Kinetics of Iron Uptake by the Freshwater Cyanobacterium

Kinetics of Iron Uptake by the

Freshwater Cyanobacterium

Microcystis aeruginosa

by

The Cuong Dang

A thesis submitted in fulfilment of the requirements for the degree of

Doctor of Philosophy

School of Civil and Environmental Engineering

The University of New South Wales

August, 2012

THE UNIVERSITY OF NEW SOUTH WALES Thesis/Dissertation Sheet

Surname or Family name: Dang

First name: The Cuong

Other name/s:

Abbreviation for degree as given in the University calendar: Ph.D.

School: Civil and Environmental Engineering

Faculty: Engineering

Title: Kinetics of Iron Uptake by the Freshwater Cyanobacterium Microcystis aeruginosa

Abstract

Kinetics of iron (Fe) uptake by the freshwater cyanobacterium Microcystis aeruginosa cultured under a variety of growth conditions are examined in this thesis. Visible light was observed to induce reductive dissociation of Fe(III) bound to ethylenediaminetetraacetic acid (EDTA) and to dramatically increase the short-term uptake rate of Fe by M. aeruginosa. A mathematical model based on photo-generated unchelated Fe(II) uptake by concentration gradient dependent passive diffusion of Fe(II) through outer-membrane channels adequately described the rate and extent of Fe uptake. Studies of the kinetics of Fe transport to periplasmic and cytoplasmic compartments of M. aeruginosa indicated that a Monod-type relationship exists between cytoplasmic Fe accumulation rates and steady-state concentrations of unchelated Fe in the periplasm and extracellular milieu, suggesting that translocation of Fe into the cytoplasm involves complexation of Fe by a limited number of Fe-binding sites in the periplasm followed by subsequent transport into the cytoplasm, possibly via energy-dependent plasma-membrane Fe transporters. Fe uptake kinetics were also examined in Fraquil

* medium containing a natural organic ligand, Suwannee River Fulvic Acid (SRFA). Reduced Fe

uptake rates in the presence of ferrozine and superoxide dismutase under both light and dark conditions indicated that approximately a quarter to a half of the total Fe uptake was accounted for by Fe(II) uptake likely produced via light-, SRFA- or superoxide-mediated reduction of Fe(III) bound to SRFA. To further investigate cellular characteristics under various levels of Fe stress, a chemostat system made of metal-free materials was developed and used to maintain Fe-limited cultures in nutrient-insufficient and replete Fraquil*. In the nutrient-insufficient case, Fe uptake rate was lower for cells grown under conditions of lower Fe availability, suggesting cells grown under severe Fe stress and other nutrients insufficiency are likely unable to synthesize sufficient resources required for Fe uptake. In contrast, reversion to the expected relationship between Fe uptake capacity and the degree of Fe-limitation was observed when cells were grown under nutrient-replete Fe limitation. A kinetic model describing Fe transformations and biological uptake was applied to determine the biologically available form of Fe in the continuous culture.

Declaration relating to disposition of project thesis/dissertation

I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International (this is applicable to doctoral theses only).

……………………………………………… Signature

……………………………………………… Witness

……………………………………………… Date

The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research.

FOR OFFICE USE ONLY

Date of completion of requirements for Award:

THIS SHEET IS TO BE GLUED TO THE INSIDE FRONT COVER OF THE THESIS

The Cuong Dang 28 November 2012

ii

ORIGINALITY STATEMENT

‘I hereby declare that this submission is my own work and to the best of

my knowledge it contains no materials previously published or written

by another person, or substantial proportions of material which have

been accepted for the award of any other degree or diploma at UNSW or

any other educational institution, except where due acknowledgement is

made in the thesis. Any contribution made to the research by others,

with whom I have worked at UNSW or elsewhere, is explicitly

acknowledged in the thesis. I also declare that the intellectual content of

this thesis is the product of my own work, except to the extent that

assistance from others in the project's design and conception or in style,

presentation and linguistic expression is acknowledged.’

Signed ……………………………………………...........................

Date ......………………………………………….............................

The Cuong Dang

28 November 2012

iii

COPYRIGHT STATEMENT

‘I hereby grant the University of New South Wales or its agents the right

to archive and to make available my thesis or dissertation in whole or part

in the University libraries in all forms of media, now or here after known,

subject to the provisions of the Copyright Act 1968. I retain all proprietary

rights, such as patent rights. I also retain the right to use in future works

(such as articles or books) all or part of this thesis or dissertation.

I also authorise University Microfilms to use the 350 word abstract of my

thesis in Dissertation Abstract International (this is applicable to doctoral

theses only).

I have either used no substantial portions of copyright material in my

thesis or I have obtained permission to use copyright material; where

permission has not been granted I have applied/will apply for a partial

restriction of the digital copy of my thesis or dissertation.'

Signed ……………………………………………...........................

Date ......………………………………………….............................

AUTHENTICITY STATEMENT

‘I certify that the Library deposit digital copy is a direct equivalent of the

final officially approved version of my thesis. No emendation of content

has occurred and if there are any minor variations in formatting, they are

the result of the conversion to digital format.’

Signed ……………………………………………...........................

Date ......………………………………………….............................

The Cuong Dang

28 November 2012

The Cuong Dang

28 November 2012

iv

ACKNOWLEDGEMENTS

First and foremost I wish to express my deepest gratitude to my supervisor, Professor

David Waite, for his highly valuable supervision throughout my course of study. His

scientific spirit, knowledge, engagement and cheerfulness have extremely assisted me

in the completion of this thesis. Special thanks to my co-supervisor, Assistant

Professor Manabu Fujii, who encouraged and patiently guided me through the

dissertation process. I wish to express my sincerest appreciation and thanks to him for

his invaluable suggestions, discussions and help in the research work.

I gratefully acknowledge the support of the University of New South Wales through

the award of the University International Postgraduate Award.

Thanks to Dr. Gautam Chattopadhyay and Mr. Kelvin Ong for their assistance in my

laboratory work. Thanks also given to Ms. Pattie MacLaughlin and Mr. Patrick Vuong

for their administrative support.

I would like to take this opportunity to express my sincere appreciation and special

thanks to all my research group members. I just name a few here: Dr. Mark Bligh, Dr.

Andrew Kinsela, Dr. An Ninh Pham, Dr. Shikha Garg, Dr. Adele Johns, Dr. Chris

Miller, Ms. Anna Yeung, Mr. Daniel Boland, Mr. Di He, Ms. Tian Ma, Mr. Yongjia

Xin, Ms. Lam Ho, etc. I want to thank you all for your kind help, support, interest and

valuable hints.

I am very thankful to my friends: Mr. Khoa Vo, Mr. Thao Tran, Mr. Lam Dang, Mr.

Hoang Dao, Dr. Nhat Le and Ms. Trang Trinh for their support and encouragement.

Also, I am very grateful for the love, spiritual support and encouragement of my

parents, sisters and brothers throughout my study.

Finally, I feel a deep sense of gratitude to my wife, Ms. My Van La, for standing next

to me in life, understanding and encouraging me during the completion of this thesis

regardless of the 6,300 km between us.

v

ABSTRACT

Kinetics of iron (Fe) uptake by the freshwater cyanobacterium Microcystis aeruginosa

cultured under a variety of growth conditions are examined in this thesis. Visible light

was observed to induce reductive dissociation of Fe(III) bound to

ethylenediaminetetraacetic acid (EDTA) and to dramatically increase the short-term

uptake rate of Fe by M. aeruginosa. A mathematical model based on photo-generated

unchelated Fe(II) uptake by concentration gradient dependent passive diffusion of

Fe(II) through outer-membrane channels adequately described the rate and extent of Fe

uptake. Studies of the kinetics of Fe transport to periplasmic and cytoplasmic

compartments of M. aeruginosa indicated that a Monod-type relationship exists

between cytoplasmic Fe accumulation rates and steady-state concentrations of

unchelated Fe in the periplasm and extracellular environment, suggesting that

translocation of Fe into the cytoplasm involves complexation of Fe by a limited

number of Fe-binding sites in the periplasm followed by subsequent transport into the

cytoplasm, possibly via energy-dependent plasma-membrane Fe transporters. Fe

uptake kinetics were also examined in Fraquil* medium containing a natural organic

ligand, Suwannee River Fulvic Acid (SRFA). Reduced Fe uptake rates in the presence

of ferrozine and superoxide dismutase under both light and dark conditions indicated

that approximately a quarter to a half of the total Fe uptake was accounted for by Fe(II)

uptake likely produced via light-, SRFA- or superoxide-mediated reduction of Fe(III)

bound to SRFA. To further investigate cellular characteristics under various levels of

Fe stress, a chemostat system made of metal-free materials was developed and used to

maintain Fe-limited cultures in nutrient-insufficient and replete Fraquil*. In the

nutrient-insufficient case, Fe uptake rate was lower for cells grown under conditions of

lower Fe availability, suggesting cells grown under severe Fe stress and other nutrients

insufficiency are likely unable to synthesize sufficient resources required for Fe

uptake. In contrast, reversion to the expected relationship between Fe uptake capacity

and the degree of Fe-limitation was observed when cells were grown under nutrient-

replete Fe limitation. A kinetic model describing Fe transformations and biological

uptake was applied to determine the biologically available form of Fe in the

continuous culture.

vi

PUBLICATIONS

Journal papers

FUJII, M., DANG, T. C., ROSE, A. L., OMURA, T. & WAITE, T. D. 2011. Effect of

light on iron uptake by the freshwater cyanobacterium Microcystis aeruginosa.

Environmental Science & Technology, 45, 1391-1398

DANG, T. C., FUJII, M., ROSE, A. L., BLIGH, M. & WAITE, T. D. 2012.

Characteristics of the freshwater cyanobacterium Microcystis aeruginosa grown in

iron-limited continuous culture. Applied and Environmental Microbiology, 78, 1574-

1583.

FUJII, M., DANG, T. C., ROSE, A. L. & WAITE, T. D. Kinetics of extracellular iron

transport to periplasmic and cytoplasmic compartments of the freshwater

cyanobaterium Microcystis aeruginosa (under preparation for re-submission to Applied

and Environmental Microbiology).

FUJII, M., DANG, T. C., ROSE, A. L. & WAITE, T. D. Iron uptake kinetics by the

freshwater cyanobacterium Microcystis aeruginosa in the presence of Suwannee River

fulvic acid (under preparation for submission to Environmental Science &

Technology).

DANG, T. C., FUJII, M., ROSE, A. L., BLIGH, M. & WAITE, T. D. Characteristics

of the freshwater cyanobacterium Microcystis aeruginosa grown in iron-limited

continuous culture under nutrient-replete condition (under preparation for submission

to Applied and Environmental Microbiology).

Conference papers

DANG, T.C., FUJII, M., ROSE, A.L., BLIGH, M. & WAITE, T.D. 2012. Growth and

responses to iron stress of the freshwater cyanobacterium Microcystis aeruginosa in

both nutrient-insufficient and -replete continuous cultures. 2012 ASLO Aquatic

Sciences Meeting. July 8-13th

, 2012. Lake Biwa, Shiga, Japan.

vii

TABLE OF CONTENTS

Acknowledgements ...................................................................................................... iv

Abstract .......................................................................................................................... v

Publications .................................................................................................................. vi

Table of Contents ........................................................................................................ vii

List of Figures ............................................................................................................. xiii

List of Tables .......................................................................................................... xxviii

Chapter 1 ....................................................................................................................... 1

Introduction ................................................................................................................... 1

1.1. Background to the study ...................................................................................... 2

1.1.1. Importance of Iron in Natural Waters towards Cyanobateria ....................... 2

1.1.2. Transformations of Iron in Natural Waters ................................................... 3

1.1.3. Iron Uptake Models by Phytoplankton ......................................................... 4

1.1.4. Mode of Iron Acquisition by the Freshwater Cyanobacterium M.

Aeruginosa and Knowledge Gaps ........................................................................... 7

1.2. Objectives ............................................................................................................ 9

1.3. Layout of Thesis .................................................................................................. 9

Chapter 2 ..................................................................................................................... 12

General Methodology ................................................................................................. 12

2.1. Reagents ............................................................................................................. 13

2.2. Culturing Conditions .......................................................................................... 13

viii

2.2.1. Culture Medium .......................................................................................... 13

2.2.2. Long-term Culturing Conditions ................................................................. 14

2.2.3. Continuous Culturing Apparatus ................................................................ 15

2.3. Short-term Iron Uptake Experiment .................................................................. 17

2.4. Measurement of Iron .......................................................................................... 18

2.4.1. Measurement of FeIIFZ3 with Spectrophotometer ...................................... 18

2.4.2. Measurement of Radio-labeled 55

Fe with Scintillation Counter ................. 19

2.5. Measurement of Cellular Iron Quota ................................................................. 19

2.5.1. Acid Digestion Combined with Spectrophotometry Method ..................... 19

2.5.2. Radiometry Method .................................................................................... 20

2.6. Model Fitting ..................................................................................................... 20

2.7. Analytical Quality Control ................................................................................. 20

2.7.1. Procedural Blank ......................................................................................... 20

2.7.2. Replication .................................................................................................. 20

Chapter 3 ..................................................................................................................... 21

Effect of Light on Iron Uptake by the Freshwater Cyanobacterium Microcystis

aeruginosa .................................................................................................................... 21

3.1. Introduction ........................................................................................................ 22

3.2. Materials and Methods ....................................................................................... 23

3.2.1. Materials ..................................................................................................... 23

3.2.2. Light Conditions ......................................................................................... 24

3.2.3. Photochemical Experiments ........................................................................ 28

ix

3.2.4. Short-term 55

Fe Uptake Experiments .......................................................... 28

3.3. Results and Discussion ...................................................................................... 29

3.3.1. Effect of Light on Photoreductive Dissociation and Fe Uptake .................. 29

3.3.2. Effect of Light Wavelength ........................................................................ 33

3.3.3. Fe Substrate for Uptake .............................................................................. 37

3.3.4. Kinetic Model for Fe Species ..................................................................... 39

3.3.5. Fe Uptake Machinery .................................................................................. 43

3.4. Implication of Findings ...................................................................................... 46

Chapter 4 ..................................................................................................................... 48

Kinetics of Extracellular Iron Transport to Periplasmic and Cytoplasmic

Compartments of the Freshwater Cyanobacterium Microcystis aeruginosa ......... 48

4.1. Introduction ........................................................................................................ 49


4.2.1. Reagents ...................................................................................................... 51

4.2.2. 55

Fe Accumulation Experiments ................................................................. 52

4.2.3. Determination of Periplasmic Fe(II) ........................................................... 54

4.2.4. Determination of Steady-state Concentration of Extracellular Unchelated

Fe ........................................................................................................................... 54

4.2.5. Analysis of Genome Sequences .................................................................. 55


4.3.1. Accumulation of 55

Fe in the Periplasm and Cytoplasm .............................. 55

4.3.2. Fe Species Translocated from the External Environment to the Periplasm 60

x

4.3.3. Fe Species Translocated from the Periplasm to the Cytoplasm .................. 61

4.3.4. Model for Translocation of Fe from the External Environment ................. 62

4.3.5. Fe Redox Speciation in Periplasm .............................................................. 66

4.4. Conclusions ........................................................................................................ 70

Chapter 5 ..................................................................................................................... 71

Iron Uptake Kinetics by the Freshwater Cyanobacterium Microcystis aeruginosa

in the Presence of Suwannee River Fulvic Acid ....................................................... 71

5.1. Introduction ........................................................................................................ 72


5.2.1. Reagents ...................................................................................................... 74

5.2.2. Culturing Media .......................................................................................... 75

5.2.3. Long-term Culturing Conditions ................................................................. 76

5.2.4. Light Condition ........................................................................................... 76


Fe Uptake Experiments .......................................................... 77

5.2.6. Kinetic Model for Fe Transformation and Uptake ..................................... 78


5.3.1. 55

Fe Uptake as a Function of SRFA Concentration .................................... 79

5.3.2. Effect of Chemical Treatment on 55

Fe Uptake ........................................... 82

5.3.3. Mode of Dark Fe Uptake ............................................................................ 85

5.3.4. Mode of Light-mediated Fe Uptake ............................................................ 89

5.4. Implications of Findings .................................................................................... 91

Chapter 6 ..................................................................................................................... 93

xi

Characteristics of the Freshwater Cyanobacterium Microcystis aeruginosa Grown

in Iron-limited Continuous Culture .......................................................................... 93

6.1. Introduction ........................................................................................................ 94


6.2.1. Materials ..................................................................................................... 95

6.2.2. Culturing Method ........................................................................................ 95

6.2.3. Chemostat Apparatus .................................................................................. 96

6.2.4. Cellular Fe Quota and External Fe Concentration ...................................... 96


Fe and 14

C Uptake .................................................................. 97

6.2.6. Kinetic Model for Unchelated Fe(II) Calculation ....................................... 98

6.2.7. Modified Chemostat Theory ..................................................................... 101

6.3. Results and Discussion .................................................................................... 105

6.3.1. Growth Kinetics in Batch Culture ............................................................. 105

6.3.2. Performance of Chemostat System under Fe Limitation .......................... 108

6.3.3. Cellular Fe Quota ...................................................................................... 112

6.3.4. Fe Uptake Kinetics .................................................................................... 115

6.3.5. Cellular Response to Fe Limitation in Chemostat .................................... 120

6.3.6. Characteristics of Iron-limited Cultures of M. aeruginosa Grown

Continuously in Nutrient-replete Fraquil* Medium ............................................ 123

6.4. Conclusions ...................................................................................................... 140

Chapter 7 ................................................................................................................... 142

Conclusions and Recommendations ........................................................................ 142

xii

7.1. Conclusions ...................................................................................................... 143

7.2. Implications of the findings ............................................................................. 147

7.2.1. With Regard to Knowledge of Fe Transformation and Uptake Kinetics by

Freshwater Cyanobacteria in Natural Waters ..................................................... 147

7.2.2. With Regard to Application of the Continuous Culturing System for Study

of Trace Metal Interactions with Freshwater Phytoplankton .............................. 147

7.2.3. With Regard to Knowledge of the Composition of the Growth Medium for

Freshwater Phytoplankton .................................................................................. 148

7.3. Recommendations for Future Work ................................................................. 148

References .................................................................................................................. 150

Appendix 1 ................................................................................................................. 170

Appendix 2 ................................................................................................................. 193

Appendix 3 ................................................................................................................. 201

xiii

LIST OF FIGURES

Figure 1.1. Transformations between Fe(II) and Fe(III) species in oxygenated

natural waters (Rose and Waite, 2003c).

Figure 1.2. The Fe(II)s and FeL models of Fe acquisition by phytoplankton.

The most significant difference between the two models is that the FeL model

excludes the unchelated Fe(III) in the medium as an important source of Fe(II)

for phytoplankton uptake (adapted from Morel et al., 2008).

Figure 1.3. Kinetic model for iron uptake by C. marina with Fe(III) reduction

to Fe(II) occurring by either non-reductive dissociation (NRD) or superoxide-

mediated non-dissociative reduction (NDR) or dissociative reduction (DR). In

this model superoxide plays an important role in the reduction of Fe(III) into

the more soluble form Fe(II) for uptake by marine phytoplankton (adapted from

Garg et al., 2007).

Figure 1.4. Iron uptake model for the freshwater cyanobacterium M.

aeruginosa in Fraquil* medium buffered by the model ligand EDTA in the

absence of light (Fujii et al., 2010a)

Figure 2.1. The chemostat culturing system consisting of non-metal materials.

The system was operated at four different dilution rates in triplicate.

Figure 3.1. Irradiation spectra emitted from the cool-white fluorescent tube of

the culturing incubator in the (A) absence and (B-H) presence of light filter

treatments. The spectra were measured using an Ocean Optics USB 4000

spectrophotometer equipped with an optical fiber and cosine corrector lens

(CC-3-UV) calibrated against a DH-2000 VIS-light source (hydrogen lamp).

The measurement was performed using SpectraSuite software in absolute

irradiation mode. Various light spectra were obtained by placing the cut-off

light filter between the light source and the irradiance probe. The photon flux

4

5

6

8

16

25

xiv

densities calculated for each wavelength range are shown in Table 3.1

Figure 3.2. Absorbance spectra for plastic and glass vessels. (A) blank (no

materials), (B) 1 cm quartz spectrophotometer cuvette (Starna Pty Ltd,

Australia), (C) Scintillation glass vials (20 mL, Crown Scientific), (D) 1 cm

polystyrene spectrophotometer cuvette (Starna Pty Ltd, Australia), (E)

polycarbonate container (250 mL, Nalgene), (F) high-clarity polypropylene

tube (15 mL, BD Falcon), (G) polypropylene microtube (1.5 mL, Eppendorf),

(H) high-density polyethylene bottle (125 mL, Nalgene). The absorbance

spectra were measured using a Varian Cary 50 UV-Vis spectrophotometer

(Scan mode). During measurement, the containers were filled with ultrapure

water (Milli-Q water). For large materials, the sample holder was removed

from the instrument and the materials were placed between the light source and

detector.

Figure 3.3. Effect of light on Fe(II)' formation and 55

Fe uptake by M.

aeruginosa. Time-courses of (A) FeIIFZ3 formation ([Fe]T = 0, 1 or 10 µM,

[EDTA]T = 26 µM, [FZ]T = 1 mM) and (B) 55

Fe uptake ([55

Fe]T = 200 nM and

[EDTA]T = 26 µM). Effect of light wavelength on (C) 55

FeEDTA uptake

([55

Fe]T = 200 nM, [EDTA]T = 26 µM) and (D) FeIIFZ3 formation ([Fe]T = 10

µM, [EDTA]T = 26 µM, [FZ]T = 1 mM). The incubations were performed in

modified Fraquil* (pH 8) in the light or dark at 27

oC. In the light filter

treatments (panels C and D), filters were placed between the incubated samples

and the light source to allow transmission of wavelengths longer than 400, 450,

500, 550, 600, 650 or 700 nm. In the control treatment, no light filter was

inserted in front of the sample. Incubations were performed for 4 h for the

photo-reduction experiment and 2 h for the 55

Fe uptake experiment. Asterisks

indicate that light filter treatments were significantly different from the control

at the p < 0.05 level using a single-tailed heteroscedastic t-test. Symbols and

error bars represent average data ±standard deviation from duplicate (photo-

reduction) or triplicate (55

Fe uptake) experiments. Solid lines represent linear

regression.

27

31

xv

Figure 3.4. Relationships between (A) FeIIFZ3 formation rate and

55Fe uptake

and (B) total photon flux density and 55

Fe uptake rate. At each data point, the

parameters were obtained from the incubation experiments and measurements

of irradiation spectra using the same cutoff filter. Thus, the data for 55

Fe uptake

and FeIIFZ3 formation rate are the same as those shown in parts C and D of

Figure 3.3. Details of total photon flux density are listed in Table 3.1.

Figure 3.5. UV-VIS absorbance spectra for FeIII

EDTA complex (solid line,

[Fe(III)]T = 0.5 mM and [EDTA]T = 1.3 mM) and EDTA (dotted line, [EDTA]T

= 1.3 mM) at pH 8 buffered by 15 mM NaHCO3. Enlarged absorbance spectra

in the visible light range are also shown. The average molar absorptivity of

FeIII

EDTA complex in the wavelength range from 400 nm to 500 nm was

determined to be 37 M-1

cm-1

.

Figure 3.6. Effect of (A) ferrozine (FZ) and (B) excess EDTA on 55

Fe uptake

rate in the light. The 55

Fe uptake experiment was undertaken by incubating cells

in Fraquil* containing pre-equilibrated

55Fe

IIIEDTA complex and FZ or excess

EDTA ([Fe]T = 200 nM, [EDTA]T = 26-260 µM and [FZ]T = 1 mM). While

various cell densities (2.6 × 105 – 2.7 × 10

7 cell.mL-

1) were used in the FZ

experiment, the cell density was kept constant (3.5 × 106 cell.mL-

1) in the

excess EDTA experiment. One and two asterisks indicate that treatments with

FZ or excess EDTA were significantly different from the control ([Fe]T = 200

nM and [EDTA]T = 26 µM) at p < 0.05 and p < 0.01 levels, respectively, using

a single-tailed heteroscedastic t-test. Symbols and error bars are average data

and errors represent ±standard deviation from triplicate experiments.

Figure 3.7. Fe uptake model by M. aeruginosa in the presence of light.

Unchelated Fe(II) (i.e., Fe(II)') is formed from the photoreductive dissociation

of ferric EDTA complex (FeIII

EDTA). The photoproduced Fe(II) subsequently

passes through the nonspecific outer membrane channel (porins) by diffusion.

However, cellular Fe uptake competes with Fe(II)' complexation by

extracellular Fe-binding ligands such as ferrozine (FZ) and excess EDTA if

present at appropriate concentrations. Solid arrows represent major reactions

34

35

38

39

xvi

under conditions of the short-term 55

Fe uptake experiment, whereas dotted

arrows indicate relatively minor reactions. Rate constants depicted near the

arrows correspond to those listed in Table 3.3.

Figure 3.8. Effect of competitive ligand concentrations and cellular densities

on calculated Fe uptake rate using eq 3.2.

Figure 4.1. Time course of 55

Fe accumulation in (A) periplasm and (B)

cytoplasm for M. aeruginosa strains PCC7806 (filled symbols) and PCC7005

(open symbols) grown under moderate Fe limitation. Fe uptake assays were

performed for 9 h in Fraquil* at concentrations of 0.7 µM for

55Fe and 20 µM

for citrate. Symbols and error bars represent the mean and ± standard deviation

from triplicate experiments. Solid and dashed lines represent the calculated

values for PCC7806 and PCC7005, respectively, using (A) eq. 4.7 and (B) the

integrated form of eq. 4.5 with Fe uptake parameters listed in Table 4.1.

Detailed 55

Fe accumulation data are provided in Table A2.1 of Appendix 2.

Figure 4.2. Cytoplasmic accumulation rate of 55

Fe by Fe-limited M. aeruginosa

strains PCC7806 (filled symbols) and PCC7005 (open symbols) as a function

of (A) steady-state concentration of total periplasmic 55

Fe and (B) calculated

concentration of unchelated Fe in the extracellular environment and periplasm.

Data were obtained from the assay using Fraquil* at concentrations of 0.7 µM

for 55

Fe and 5-200 µM for citrate. Solid and dashed lines in panel A were

determined for PCC7806 and PCC7005, respectively, by linear regression

analysis (p<0.05, n=15). In panel B, the solid and dashed lines represent the

calculated values using eq. 4.6 for PCC7806 and PCC7005, respectively.

Detailed 55

Fe accumulation data are provided in Table A2.2 of Appendix 2.

Calculated values of unchelated Fe concentrations are provided in Table A2.3

of Appendix 2. Symbols and error bars represent the mean and ± standard

deviation from triplicate experiments.

Figure 4.3. Cellular 55

Fe accumulation in the various culturing media at pH 8

(plasmolysis solutions, Fraquil* and 2 mM NaHCO3). Plasmolysis solutions

44

56

57

58

xvii

were 0.5 M D-sorbitol, sucrose or NaCl (buffered by 10 mM Tris-HCl, 2 mM

NaHCO3 and 1 mM for Na2EDTA) and 0.5M D-sorbitol with high EDTA

concentration (10 mM). The incubation experiment was initiated by addition of

55FeEDTA to the culture media at final concentrations of 0.7 µM for

55Fe and 2

mM for EDTA. In case of the 0.5M D-sorbitol solution containing high EDTA,

the final concentration of EDTA was adjusted to 11 mM. All incubations were

performed for 30 or 60 min in the dark at pH 8 with Fe-limited Microcystis

aeruginosa (PCC7806).

Figure 4.4. Kinetic model for Fe transport from the extracellular environment

to the intracellular environment in cyanobacteria. In the extracellular

environment, unchelated Fe (i.e., Fe′) is formed due to the (thermal or

reductive) dissociation of chelated Fe. Unchelated Fe subsequently diffuses

through non-specific outer membrane channels (such as porins). Unchelated Fe

in the periplasm is then complexed by one or more periplasmic Fe-binding

ligands (FeXperi) followed by translocation of Fe into the cytoplasm (Fecyto) by

inner membrane Fe transporters. A possible mechanism of Fe(III) and Fe(II)

transformation in the periplasm is also illustrated. Solid arrows represent major

reactions considered in the model. Rate constants depicted near the arrows

correspond to those listed in Table 4.1. MCO: multi-copper oxidase, FeoB:

ferrous iron transporter, FutA: ferric iron transporter.

Figure 4.5. Effect of ascorbate and TTM on Fe(II) accumulation in the

periplasm of M. aeruginosa PCC7806; (A) oxidation kinetics of Fe(II) in the

periplasmic extract, and (B) percentage of Fe(II) extracted from the periplasm.

PCC7806 was incubated in Fraquil* (0.7 µM Fe and 100 µM citrate) in the

presence and absence of chemical treatments (1 mM ascorbate and 1 mM

ascorbate plus 100 µM TTM). The periplasm was extracted by the cold osmotic

shock method in cold Milli-Q water followed by measurement of Fe(II) in the

extract by the luminol chemiluminescence technique. The amount of

periplasmic Fe(II) was calculated by assuming that the observed maximum

value of the chemiluminescence signal corresponds to the amount of Fe(II) in

the periplasm. Error bars represent ±standard deviation from duplicate

63

67

xviii

experiments. A single-tailed heteroscedastic t-test indicated that the treatments

with ascorbate + TTM were different from the control at a p value of 0.14.

Figure 4.6. Effect of chemical treatments on cellular 55

Fe accumulation for M.

aeruginosa PCC7806. In the control, cells were incubated for 3 hr in Fraquil*

containing 55

Fe-citrate (total concentrations for Fe and citrate were 0.7 µM and

100 µM, respectively). In the chemical treatments, cells were incubated in the

additional presence of 100 µM TTM, 1 mM ascorbate and 1 mM ascorbate plus

100 µM TTM. Error bars represent ±standard deviation from duplicate

experiments. One asterisk indicates that chemical treatments were significantly

different from the control at a p value less than 0.05 using a single-tailed

heteroscedastic t-test.

Figure 5.1. Kinetic model for Fe chemical speciation and uptake by M.

aeruginosa.

Figure 5.2. 55

Fe uptake as a function of (A) SRFA and (B) model ligand

concentrations in the absence (black symbols and bars) and presence (white

symbols and bars) of light. The 55

Fe uptake assay was performed at

concentrations of 200 nM total Fe, 1-100 mg L-1

SRFA and 26-100 µM citrate

and EDTA. Solid and dotted lines indicate model fits to the data from the Bligh

and Rose model, respectively. Effect of (C) ferrozine (FZ) and (D) superoxide

dismutase (SOD) on 55

Fe uptake. In control treatments, Fe uptake assays were

undertaken under dark (black bar) and light (white bars) at concentrations of

200 nM for Fe and 1-25 mg L-1

for SRFA. In the chemical treatments, the

identical 55

Fe uptake assay was performed except for the additional presence of

either FZ or SOD under dark (gray bar) and light (shaded bar). (E) Effect of

reducing agents on 55

Fe uptake. The control treatments were undertaken under

dark (black bar) and light (white bars) (200 nM for Fe and 5 mg L-1

for SRFA).

Chemical treatments were performed, in addition, in the presence of FZ,

ascorbate (Asc), hydroxylamine hydrochloride (HH), xanthine/xanthine oxidase

(X/XO) or their combination. All short-term Fe uptake assays were performed

in Fraquil* for 2 h at cell density of ~2 × 10

6 cell mL

-1. Symbols and error bars

69

78

80

xix

represent averaged value and ±standard deviation from triplicate experiments.

In panels C-E, asterisks indicate that 55

Fe uptake rate in the presence of a

particular chemical treatment is significantly different from control at the levels

of p < 0.01 for **

and p < 0.05 for * using a single-tailed heteroscedastic t-test.

Figure 5.3. Simulated results for unchelated Fe concentrations (gray lines for

Fe(III) and black lines for Fe(II)) as a function of SRFA ligand concentration

by using the Rose (solid lines) and Bligh (dotted lines) models.

Figure 6.1. Model for Fe uptake by M. aeruginosa in the presence of light

(Adapted from Fujii et al. (2011a))

Figure 6.2. Growth curves in batch cultures of M. aeruginosa at different total

Fe concentrations in Fraquil*. Total Fe concentrations were varied from 10 nM

to 10 µM; all other media components were constant. Symbols represent the

mean and error bars represent the standard deviation from triplicate incubations

(filled diamonds = 10 nM [Fe]T, filled squares = 20 nM [Fe]T, filled triangles =

50 nM [Fe]T, open diamonds = 100 nM [Fe]T, open squares = 1 µM [Fe]T, and

crosses = 10 µM [Fe]T).

Figure 6.3. Relationship between specific growth rate µ (d-1

) and log

concentration of unchelated Fe(II)’ (where [Fe(II)’] is in molar (M) units) in

batch culture studies of M. aeruginosa. Non-linear regression analysis yielded a

half saturation constant for growth of 'S

K = 3.6 ± 0.32 fM (with respect to

Fe(II)’) and a maximum specific growth rate µmax = 0.80 ± 0.03 d-1

. Solid and

dotted lines represent the regression line and 95% confidential interval,

respectively. Symbols indicate data for experimentally determined growth rate

under different degrees of Fe limitation.

Figure 6.4. Predicted and measured steady-state cell density and substrate

concentration in continuous cultures of M. aeruginosa as a function of dilution

rate with different total Fe concentrations in the inflowing medium (50 nM and

20 nM). Symbols represent data for steady-state cell density in Fraquil*

89

99

106

107

109

xx

medium with total Fe of 50 nM (circles) and 20 nM (triangles). Dotted lines are

the theoretical values of steady-state cell density calculated from eq. 6.16 with

growth parameters estimated from batch culture studies (µmax = 0.80 ± 0.03 d-1

,

'S

K = 3.6 ± 0.32 fM with respect to Fe(II)’, TSK = 26 ± 2.3 nM with respect to

total Fe, and Y = 8.1 ± 0.21 × 1016

cell (mol Fe)-1

), while bold lines indicate the

theoretical steady-state cell density estimated with parameters obtained from

continuous culture studies ( 'S

K = 3.4 ± 0.82 fM, TSK = 25 ± 5.0 nM and Y =

1.1 ± 0.2 × 1017

cell mol-1

), except for µmax (0.80 ± 0.03 d-1

) which was

obtained from the batch studies. Dashed and chained lines indicate predicted

steady-state unchelated Fe(II)’ concentrations estimated using parameters from

batch and continuous culture studies, respectively.

Figure 6.5. Growth of M. aeruginosa in the continuous culture system at

different dilution rates with total Fe concentrations in the inflowing Fraquil*

medium. (A) [Fe]T = 50 nM, with dilution rates of 0.07 d-1

(diamonds), 0.15 d-1

(squares), 0.30 d-1

(triangles) and 0.45 d-1

(circles). (B) [Fe]T = 20 nM, with

dilution rates of 0.09 d-1


(squares), 0.17 d-1

(triangles) and

0.25 d-1

(circles). Symbols represent the mean and error bars the standard

deviation from triplicate incubations. Dashed lines represent the 95%

confidence interval at steady-state.

Figure 6.6. Time-course of cellular Fe quotas for Fe-limited M. aeruginosa in

the chemostat with [Fe]T = 20 nM as radiolabelled 55

Fe in the inflowing

medium and dilution rates of 0.09 d-1


(squares), 0.17 d-1


(circles). Symbols represent the mean and error bars the

standard deviation from triplicate incubations.

Figure 6.7. Relationship between the cellular Fe quota (Q) and the (A) specific

uptake rate of Fe (µQ) or (B) specific growth rate (µ) for Fe-limited M.

aeruginosa under steady-state conditions in continuous cultures. The system

was operated at four different dilution rates (0.09, 0.14, 0.17 and 0.25 d-1

) and

fed with Fraquil* medium containing 20 nM radiolabeled

55Fe. In panel (A),

linear regression analysis (represented by the bold line) yielded the maximum

110

113

114

xxi

“impossible” growth rate µ’max = 0.37 ± 0.04 d-1

and minimum cellular quota

Qmin = 1.2 ± 0.2 × 103 zmol cell

-1. Symbols represent the mean and error bars

the standard deviation from triplicate incubations. In panel (B), the solid line

represents the theoretical curve calculated from the Droop equation using the

obtained estimated values of µ’max and Qmin. Symbols represent the mean from

triplicate incubations.

Figure 6.8. Time-course of 55

Fe uptake during batch short-term Fe uptake

assays using cells obtained at steady-state from the chemostat cultures grown

with [Fe]T = 20 nM in the inflowing medium and dilution rates of 0.09 d-1


(squares), 0.17 d-1


(circles). In the

short-term uptake assay, each culture was incubated in Fraquil* with either (A)

20 µM EDTA or (B) 200 µM EDTA and constant concentration of radiolabeled

55Fe (200 nM). Symbols represent the mean and error bars the standard

deviation from triplicate experiments. The continuous lines were obtained by

linear regression of data collected within 4 h (represented by closed symbols)

for each culture.

Figure 6.9. Eadie-Hofstee plots demonstrating the linear relationship between

the short-term 55

Fe uptake rate (ρFe) and the ratio ρFe/[Fe(II)’] (d-1

M-1

) for

cultures of M. aeruginosa. Linear regression analysis yielded comparable half-

saturation constants for Fe uptake (Kρ = 18 ± 1.9 fM, as Fe(II)’) but different

maximum specific uptake rates (ρmax of 270, 720, 950 and 1,010 zmol cell-1

hr-1

for cultures grown at dilution rates of 0.09 d-1


(squares),

0.17 d-1


(circles), respectively). Lines for 95%

confidential intervals were omitted for clarity.


C uptake during batch short-term uptake assays

using cells obtained at steady-state from the chemostat cultures grown with

[Fe]T = 20 nM in the inflowing medium and dilution rates of 0.09 d-1


(squares), 0.17 d-1


(circles).

Symbols represent the mean and error bars the standard deviation from

triplicate experiments.

116

118

122

xxii

Figure 6.11. Growth curves in batch cultures of M. aeruginosa at a constant

total Fe concentration in different modified Fraquil* growth media. Total Fe

concentration and its chelator EDTA were fixed at 10 and 26 µM while other

media components were varied as shown in Table 6.3. Symbols represent the

mean and error bars represent the standard deviation from duplicate incubations

(filled diamonds: control (i.e., Fraquil*); filled squares: Test 1; filled triangles:

Test 2; open diamonds: Test 3; open squares: Test 4 and open triangles: Test 5).

Figure 6.12. Growth curves in batch cultures of M. aeruginosa at various total

Fe concentrations in the nutrient-replete Fraquil* medium (i.e., Test 2 medium).

Total Fe concentration was varied from 0.05 to 10 µM while concentration of

EDTA was fixed at 26 µM. Symbols represent the mean and error bars

represent the standard deviation from duplicate incubations (filled diamonds:

[Fe]T = 0.05 µM; filled squares: [Fe]T = 0.1 µM, filled triangles: [Fe]T = 0.2

µM, filled circles: [Fe]T = 0.5 µM, open diamonds: [Fe]T = 1.0 µM, open

squares: [Fe]T = 2.0 µM, open triangles: [Fe]T = 5.0 µM, and open circles: [Fe]T

= 10 µM).


) and log

concentration of unchelated Fe(II)’ (M) in batch culture studies of M.

aeruginosa grown in nutrient-replete Fraquil* medium. Solid and dotted lines

represent the regression line and 95% confidential interval, respectively.

Symbols indicate data for experimentally determined growth rate under

different degrees of Fe limitation.

Figure 6.14. Growth of M. aeruginosa in the continuous culture system at

different dilution rates with total Fe concentrations in the inflowing nutrient-

replete Fraquil* medium [Fe]T = 100 nM, with dilution rates of 0.07 d

-1


(squares), 0.30 d-1


(circles).

Symbols represent the mean and error bars the standard deviation from

triplicate incubations. Dashed lines represent the 95% confidence interval at

steady-state.

125

128

130

132

xxiii


concentration in continuous cultures of M. aeruginosa as a function of dilution

rate with different total Fe concentrations in the two inflowing media: Fraquil*

(20 nM and 50 nM) and nutrient-replete Fraquil* (100nM). Symbols represent

data for steady-state cell density in Fraquil* medium with total Fe of 20 nM

(triangles), 50 nM (squares) and 100 nM (diamonds). Dotted lines are the

theoretical values of steady-state cell density calculated from eq. 6.16 with

growth parameters estimated from batch culture studies in Fraquil* (µmax = 0.80

± 0.03 d-1

, 'S

K = 3.6 ± 0.32 fM with respect to Fe(II)’, TSK = 26 ± 2.3 nM with

respect to total Fe, and Y = 8.1 ± 0.21 × 1016

cell (mol Fe)-1

), while bold lines

indicate the theoretical steady-state cell density estimated with parameters

obtained from batch culture studies in nutrient-replete Fraquil* (µmax = 0.89 ±

0.03 d-1

, 'S

K = 3.1 ± 0.30 fM, TSK = 23 ± 2.2 nM, and Y = 2.7 ± 0.74 × 10

17

cell (mol Fe-1

)). Chained and dashed lines indicate predicted steady-state

unchelated Fe(II)’ concentrations estimated using parameters from batch

culture studies in Fraquil* and nutrient-replete Fraquil

*, respectively.

Figure 6.16. Relationship between the cellular Fe quota (Q) and the specific

growth rate (µ) for Fe-limited M. aeruginosa under steady-state conditions in

continuous cultures. The system was operated at four different dilution rates

(0.07, 0.15, 0.30 and 0.45 d-1

) and fed with nutrient-replete Fraquil* medium

containing 100 nM Fe. The solid line represents the theoretical curve calculated

from the Droop equation using the obtained estimated values of µ’max = 0.69 ±

0.05 d-1

and Qmin = 18 ± 2.6 amol cell-1

. Symbols represent the mean from

triplicate measurements.


Fe uptake during batch short-term Fe uptake

assays using cells obtained at steady-state from the chemostat cultures grown

with [Fe]T = 100 nM in the inflowing nutrient-replete Fraquil* medium and



(squares), 0.30 d-1

(triangles) and

0.45 d-1

(circles). In the short-term uptake assay, each culture was incubated in

nutrient-replete Fraquil* with 20 µM EDTA and 200 nM radiolabeled

55Fe.

133

135

137

xxiv

Symbols represent the mean and error bars represent the standard deviation

from triplicate experiments. The continuous lines were obtained by linear

regression of data collected within 4 h (represented by closed symbols) for each

culture.

Figure 6.18. Eadie-Hofstee plots demonstrating the linear relationship between

the short-term 55

Fe uptake rate (ρFe) and the ratio ρFe/[Fe(II)’] for cultures of M.

aeruginosa. Linear regression analysis yielded comparable half-saturation

constants for Fe uptake (Kρ = 45 ± 1.9 fM, as Fe(II)’) but different maximum

specific uptake rates (ρmax of 1.0 ± 0.046, 0.89 ± 0.079, 0.67 ± 0.087 and 0.48 ±

0.035 amol cell-1

hr-1



(squares), 0.30 d-1


(circles),

respectively). Lines for 95% confidential intervals were omitted for clarity.

Figure A1.1. Time-course of FeIIFZ3 formation from Fe

IIIEDTA in (A and C)

the light and (B and D) dark. For measurement of photo-reduction rate of

FeIII

EDTA, pre-equilibrated FeIII

EDTA complex and FZ were mixed in

Fraquil* at concentrations of 1-10 µM for Fe(III), 26 µM for EDTA and 1 mM

for FZ, followed by incubation for several hours at 27oC in the presence and

absence of the light (157 µmol quanta.m-2

.s-1

). Photo-reductive dissociation rate

constants were determined by applying eq. A1-5 to the measurements with (E)

1 µM and (F) 10 µM total Fe. Symbols and error bars indicate average data and

±standard deviation from triplicate experiments. Solid lines represent linear

regression lines.

Figure A1.2. Bioavailability of pre-photolyzed 55

FeEDTA complex in the dark.

The x-axis represents the time for which 55

FeEDTA complex was exposed to

the light (157 µmol photons.m-2

.s-1

) before the commencement of the 55

Fe

uptake experiment. Immediately after irradiation, the photolyzed 55

FeEDTA

complex was added at final concentrations of 200 nM Fe and 26 µM EDTA to

the Fe and EDTA-free Fraquil* containing M. aeruginosa cells at a density of 3

× 106 cell.mL

-1. Cells were then incubated for 2 h in the dark at 27

oC. Values

shown represent the average and ±standard deviation from triplicate

139

172

174

xxv

experiments.

Figure A1.3. Kinetic data for the dissociation of FeIIEDTA complex in Fraquil

*

(pH 8); (A) time-dependent formation of FeIIFZ3 complex over a range of

[EDTA]T and (B) plots of time versus ln[Fe(II)]T/([Fe(II)]T-[FeIIFZ3]). The

value of koverall[FZ]T3 was determined as the slope of the line in the panel B.

Figure A3.1. Time-course of 55

Fe uptake under the dark (closed symbol) and

light (open symbol) conditions. 55

Fe uptake was measured by incubating cells

(at density of 1.6 × 106 cell.mL

-1) in Fraquil

* containing pre-equilibrated

55Fe

IIISRFA complex at 27

oC. Concentrations of Fe and SRFA were 200 nM

and 1 mg.L-1

, respectively. Symbols represent experimental data. Solid and

dotted lines were yielded by applying a linear regression analysis to the data

collected within 2 h under the dark and light conditions, respectively.

Figure A3.2. Comparison of measured 55

Fe uptake rate to calculated Fe(III)

uptake for M. aeruginosa PCC7806. 55

Fe uptake rates were determined in the

short-term incubational assay under the dark in modified Fraquil* containing

200 nM for Fe, 1, 5 and 25 mg L-1

for SRFA and 1 mM for FZ. In the model

calculations, steady-state concentrations for Fe(III)' were determined at the

concentration identical to those employed in the short-term assay by using rate

constants for complexation and dissociation for FeIII

SRFA complex published

by Rose (square), Jones (diamond) and Bligh (triangle). Fe(III) uptake rates

were then calculated by use of Monod-type equation with parameters listed in

Table 5.1. Solid line represents linear line with 1:1 slope.


Fe(II) uptake rate to calculated Fe(II)

uptake for M. aeruginosa PCC7806. Measured Fe(II) uptake rates in this figure

were determined by subtracting 55

Fe uptake rate in the presence of FZ from that

measured in the absence of FZ. The short-term incubational assays were

performed in the absence and presence of FZ under the dark in modified

Fraquil* containing 200 nM for Fe, 1, 5 and 25 mg L

-1 for SRFA and 1 mM for

FZ. In the model calculations, steady-state concentrations for Fe(II)' were

180

202

203

204

xxvi

determined at the concentration identical to those employed in the short-term

assay by using rate constants for complexation and dissociation for FeIISRFA

complex published by Rose (square) and Bligh (triangle). Fe(II) uptake rates

were then calculated by use of Monod-type equation with parameters listed in

Table 5.1. Solid line represents linear line with 1:1 slope. Error bar indicates

standard deviation from duplicate experiments.

Figure A3.4. Comparison of calculated steady-state concentration of Fe(II)’

under the dark and light conditions. Steady-state concentrations for Fe(II)' were

determined by using rate constants for complexation and dissociation for

FeIISRFA complex published by Rose (square) and Bligh (triangle), the

photochemical and non-photochemical reduction of FeIII

SRFA complex and

oxidation of FeIISRFA complex. Solid line represents linear line with 1:1 slope.

Figure A3.5. Effect of pH on the 55

Fe uptake rate for M. aeruginosa PCC7806

under (A) dark and (B) light. Effects of FZ (gray bar) and SOD (white bar) on

55Fe uptake were also examined compared to control treatment (black bar)

where addition of FZ or SOD was omitted. Error bar indicates standard

deviation from triplicate experiments. Asterisks indicate that 55

Fe uptake rate in

the presence of chemical treatment is significantly different from control (55

Fe

uptake rate in the absence of FZ or SOD) for each pH at the levels of p < 0.01

for ** and p < 0.05 for * using a single-tailed heteroscedastic t-test.

Figure A3.6. Primary kinetic data of FeIIFZ3 formation in the (A) light and (B)

dark conditions. The time-dependent FeIIFZ3 formation in Fraquil

* (pH 8) was

spectrophotometrically monitored for 4 h at concentrations of 1 µM for Fe(III),

1 mM for FZ, 1 mg.L-1

for SRFA, 26 µM for EDTA and 100 µM for citrate.

Symbols and error bars indicate average data and ±standard deviation from

triplicate experiments.

Figure A3.7. Determination of rate constants for photo-reduction of Fe(III)-

ligand complexes in Fraquil* (pH 8). The experimental conditions, symbols and

error bars are identical to those in Figure A3.6, except that the data measured

205

206

209

212

xxvii

under the light were only shown. The solid lines represent linear regression

lines in each ligand system.

Figure A3.8. Effect of pH on reduction of FeIII

SRFA under the light and dark.

213

xxviii

LIST OF TABLES

Table 3.1. Photon flux density at each wavelength range.

Table 3.2. Parameters used for the determination of quantum yield.

Table 3.3. Kinetic model for Fe transformation and uptake under the light.

Table 3.4 Measured and calculated Fe uptake parameters for various culture

conditions.

Table 4.1. Kinetic parameters used for the calculation of intracellular Fe

transport and extracellular Fe transformation.

Table 5.1. Kinetic model and rate constants used in this study.

Table 6.1. Kinetic model for Fe transformation and uptake in the presence of

light by M. aeruginosa (adapted from Fujii et al. (2011a) and therein).

Table 6.2. Measured and calculated Fe uptake parameters under the conditions

of short-term 55

Fe uptake experiments for four different steady-state cultures of

M. aeruginosa.

Table 6.3. Compositions of the modified nutrient-replete Fraquil* media

examined in this study.

Table 6.4. Summary of the growth constants in the batch cultures and the

behaviors of the Fe-limited chemostat cultures at different dilution rates of M.

aeruginosa grown in both Fraquil* and nutrient-replete Fraquil

*.

Table A1.1. Formation rate constant of FeIIEDTA complex (kf-EDTA) in Fraquil

*

(pH 8).

26

36

41

42

59

83

100

119

126

141

177

xxix

Table A1.2. Dissociation rate constant of FeIIEDTA complex (kd-EDTA) in

Fraquil* (pH 8).

Table A1.3. Published values of porin properties for various Gram-negative

bacteria.

Table A1.4. Range of uptake rate constant (kup) calculated using published

parameters (a, porin radius; l, channel length; Nporin, porin density; D, diffusion

coefficient of metal ions; As, surface area of Microcystis aeruginosa PCC7806).

Table A2.1. Measured and modelled values for the time course of 55

Fe

accumulation in the periplasmic and cytoplasmic fractions.

Table A2.2. Measured and modelled values for the steady-state periplasmic 55

Fe

concentration and accumulation rate of cytoplasmic 55

Fe over a range of

Fe:citrate ratios.

Table A2.3. Calculated values of unchelated Fe concentrations in the

extracellular environment and periplasm.

Table A3.1. Reduction rate constants for organically complexed Fe(III) in

Fraquil* (pH 8).

Table A3.2. Reduction rate constants for organically complexed Fe(III) in

Fraquil* (pH 6-9).

181

187

188

198

199

200

214

214

1

CHAPTER 1

INTRODUCTION

Chapter 1. Introduction

2

1.1. BACKGROUND TO THE STUDY

1.1.1. Importance of Iron in Natural Waters towards Cyanobateria

Iron (Fe) is one of the most essential micronutrients for almost all living organisms

because of its critical roles in various metabolic processes (Crichton, 2009).

Cyanobacteria in particular have a relatively high Fe requirement since Fe is needed

for the processes of photosynthetic and respiratory electron transfer and, in some cases,

nitrogen fixation (Straus, 1994). Therefore, growth of cyanobacteria is influenced

strongly by Fe availability (Wilhelm, 1995). In surface waters at circumneutral pH,

concentrations of ferrous iron (Fe[II]) and ferric iron (Fe[III]) in biologically available

unchelated inorganic forms are typically low due to rapid oxidation of Fe(II) (Rose and

Waite, 2003a) and strong complexation of Fe(III) (Kuma et al., 1996, Liu and Millero,

2002) by a range of naturally occurring ligands (Fujii et al., 2008a). When Fe is a

growth-limiting nutrient, photochemically and biologically mediated reduction of

Fe(III) to more soluble Fe(II) may become critical steps in increasing Fe availability

(Sunda, 2001, Salmon et al., 2006, Fujii et al., 2010a).

Occurrence of the bloom-forming freshwater cyanobacterium Microcystis aeruginosa

in lakes, reservoirs and slowly-flowing rivers poses serious social and ecological

concerns with excessive growth typically deteriorating water quality and jeopardizing

human and ecological health (DECC, 2005). Evidence exists that growth of this

organism can be limited by supply of the trace nutrient Fe (Nagai et al., 2006).

Additionally, Fe nutrition alters basal metabolic functions of the organism including

photosynthesis, respiration and nutrient uptake (Imai et al., 1999, Kosakowska et al.,

2007, Xing et al., 2007) as well as potentially inducing the biosynthesis of secondary

metabolites such as the potent hepatotoxin (microcystin), possibly to prevent cellular

damage from reactive oxygen species that are generated by oxidative stress (Alexova

et al., 2011). Recent studies suggested that iron limitation can induce an increase in

toxin generation of M. aeruginosa (Sevilla et al., 2008, Alexova et al., 2011). In fact,

Microcystis species can produce a wide range of secondary metabolites including

microcystins, aeruginosins, microginins, anabaenopeptins, cyanopeptolins,

microviridins and cyclamides (Welker and von Dohren, 2006) which can affect both


3

animals and human beings (Carmichael and Falconer, 1993). Microcystins, with a

molecular weight between 900-1100 daltons, are of particular concern as they are

potent hepatotoxins and are the most widespread of cyanotoxins in brackish and

freshwater environments (Sivonen and Jones, 1999, Welker and von Dohren, 2006).

In order to understand ecological functioning and adaptation of M. aeruginosa and to

provide insight into the management of aquatic environments to reduce both the

occurrence of the blooms and the toxicity potentially associated with these blooms, it

is necessary to thoroughly understand the Fe uptake mechanism of this organism.

1.1.2. Transformations of Iron in Natural Waters

In natural waters Fe occurs in five major forms including unchelated inorganic

complexes of Fe(II) and Fe(III), complexes with natural organic ligands (FeIIL and

FeIII

L) and Fe(oxy)hydroxides which may (or may not) be associated with other

(inorganic and organic) particles (Bruland and Rue, 2001). The major transformations

that occur between various iron species in natural waters are summarised in Figure 1.1

(Rose and Waite, 2003c). Ferric iron species are thermodynamically favourable in oxic

waters because of rapid oxidation of ferrous iron, particularly with dissolved oxygen at

circumneutral pH (King et al., 1995). Ferric irons tend to undergo rapid polymerisation

at low concentrations (reaction 2, Figure 1.1) to generate insoluble forms. However, in

the presence of natural organic matter (NOM) both Fe(II) and Fe(III) may additionally

form complexes with NOM, also maintaining the iron in dissolved form (reactions 4

and 6, Figure 1.1). Ferrous iron, either as inorganic ferrous species or bound to NOM,

can be returned to the ferric oxidation state by reaction with oxygen or by hydrogen

peroxide (reactions 1 and 3, Figure 1.1). In general, ferrous forms are more soluble

than Fe(III) due to the lower stability of complexes of Fe (II) with organic chelators.

Thus, the reduction of Fe(III) to Fe(II), which can be mediated by photochemical

(reaction 8, Figure 1.1) and biological processes, increases the concentration of labile

form of Fe and subsequent biological uptake.


4

Figure 1.1. Transformations between Fe(II) and Fe(III) species in oxygenated natural

waters (Rose and Waite, 2003c).

1.1.3. Iron Uptake Models by Phytoplankton

There have been various strategies to solubilise and acquire Fe by phytoplankton. The

siderophore-mediated iron uptake mechanism represents one of the most studied of

iron acquisitions. The ability of phytoplankton to synthesize siderophores during iron

limitation and actively transport them into cells has been known for some time (Estep

et al., 1975, Murphy et al., 1976, Simpson and Neilands, 1976, Wilhelm and Trick,

1994, Neilands, 1995, Martinez et al., 2001). Under iron-deplete conditions some

microorganisms have been shown to release siderophores (Greek: siderous = iron,

phorus = bearer) - low molecular weight, specific iron-binding compounds

(ferrisiderophore complex) - and to subsequently acquire iron specifically from these

strong ferric complexes. The presence of siderophores is expected to affect the cycling

of iron in the systems potentially increasing the pool of biologically available iron by

the amount of iron complexed to the siderophores.


5

Recently, numerous studies have focused on the extracellular reductive strategy of Fe

acquisition by phytoplankton. Following the Fe’ model (Hudson and Morel, 1990) in

which unchelated Fe(III) (Fe(III)’), Fe(OH)2+ , Fe(OH)3, and Fe(OH)4

- is considered as

the main substrate for uptake by marine phytoplankter, some studies reported that iron

bound in strong complexes can be taken up by some species and that both chelated

(FeIII

L) and unchelated Fe(III) (Fe(III)’) must be reduced prior to internalization

(Soriadengg and Horstmann, 1995, Maldonado and Price, 2001, Rose et al., 2005).

More recent models of Fe acquisition by marine phytoplankton have focused on the

role of Fe(II) in uptake including two well-known reduction mechanisms: the Fe(II)s

model (part A of Figure 1.2) described by Shaked et al. (2005) and the FeL model (part

B of Figure 1.2.) proposed by Salmon et al. (2006).

Figure 1.2. The Fe(II)s and FeL models of Fe acquisition by phytoplankton. The most

significant difference between the two models is that the FeL model excludes the

unchelated Fe(III) in the medium as an important source of Fe(II) for phytoplankton

uptake (adapted from Morel et al., 2008).

In the Fe(II)s model, both the chelated and unchelated Fe(III) serve as sources of

Fe(II)s at the cell surface for uptake where the parameter Fe(II)s represents both

chelated and unchelated Fe(II) which are formed by cell surface reductase at the cell

surface. Meanwhile, in the FeL model, the chelated Fe(III) serve as a sole source of

either unchelated or chelated Fe(II), for uptake. Generally, current models for Fe

acquisition suggest that unchelated Fe(II) and Fe(III) can be taken up by marine

microalgae and that reduction of both chelated and unchelated Fe(III) by either


6

photochemical and/or biological processes in the external medium or near the cell

surface prior to uptake is critical to the iron uptake process. The precise mode of

reduction may be organism specific with reduction occurring either in bulk solution

and driven by light or reduced extracellular metabolites or at the cell surface by

mediated by membrane-bound enzymes. It has also been suggested by Rose et al.

(2005) that membrane bound oxido-reductase enzymes may reduce oxygen to

superoxide which, in turn, could reduce Fe(III) to Fe(II). Garg et al. (2007)

investigated this mechanism further for the prolific superoxide producer Chattonella

marina and showed that the precise mode of Fe(III) reduction was dependent upon the

strength of any Fe(III)L complex present with both dissociative and non-dissociative

modes of superoxide-mediated Fe(III) reduction possible (Figure 1.3).

Figure 1.3. Kinetic model for iron uptake by C. marina with Fe(III) reduction to Fe(II)

occurring by either non-reductive dissociation (NRD) or superoxide-mediated non-

dissociative reduction (NDR) or dissociative reduction (DR). In this model superoxide

plays an important role in the reduction of Fe(III) into the more soluble form Fe(II) for

uptake by marine phytoplankton (adapted from Garg et al., 2007).

In conclusion, studies to date on iron uptake models for phytoplankton have shown

that not only the form of iron in the external medium but also the kinetics of

transformation between different species of iron, often mediated by the microorganism

Fe(II)'

Transporter

Fe(III)'

Transporter

kup2

kup1

kd2 kf2 kf1 kd1

FeIIIL

, kox2

, kox1

, kr1 FeIIL

FeIII FeII

−•2O

2O

2O

−•2O

Oxido-reductase

−•2O

2O

, kr2

NDR

DR

NRD


7

itself or by photochemical processes, are critical to bioavailability of iron and its rate

of acquisition by the microorganism.

1.1.4. Mode of Iron Acquisition by the Freshwater Cyanobacterium

M. aeruginosa and Knowledge Gaps

Although the modes of Fe acquisition by marine phytoplankton have been extensively

studied, little is known of the mode of Fe acquisition by freshwater phytoplankton. In

terms of the freshwater cyanobaterium M. aeruginosa, in contrast to many

cyanobacteria which can produce siderophores to facilitate Fe uptake (Murphy et al.,

1976, Simpson and Neilands, 1976, Kerry et al., 1988, Wilhelm and Trick, 1994),

excretion of siderophores to assist in acquiring Fe is not believed to be used by this

organism (Fujii et al., 2011b, Schleiff et al., 2008). A siderophore-independent iron

acquisition mechanism was also observed in Fe-limited cells of siderophore-forming

freshwater cyanobacterium Anabaena flos-aquae (Wirtz et al., 2010). Recently, a

kinetic model incorporating uptake of both unchelated Fe(II) and Fe(III) for this

organism under darkness (Figure 1.4) has been proposed (Fujii et al., 2010a). This

model suggested that in the presence of strong Fe chelators such as

ethylenediaminetetraacetate (EDTA), superoxide-mediated reductive dissociation of

organically-complexed Fe(III) with subsequent uptake of unchelated Fe(II) was a

significant pathway for Fe uptake. In contrast, assimilation of unchelated Fe(III) was

favoured in the presence of the weak Fe-binding ligand citrate (≤ 100 µM).


8

Figure 1.4. Iron uptake model for the freshwater cyanobacterium M. aeruginosa in

Fraquil* medium buffered by the model ligand EDTA in the absence of light (Fujii et

al., 2010a)

While the mode of Fe acquisition in batch cultures of M. aeruginosa in Fraquil*

medium buffered by the model ligand EDTA in the absence of light has apparently

been described satisfactorily, a variety of important questions relating to iron uptake

by M. aeruginosa remain including:

(i) Does light, which is recognized to induce a net increase in the more soluble

and bioavailable form – extracellular unchelated Fe(II) - facilitate Fe uptake

by M. aeruginosa?

(ii) How is Fe transported from the extracellular medium into the periplasmic

and cytoplasmic compartments of M. aeruginosa?

(iii) Is the Fe uptake kinetics by M. aeruginosa based on the studies using model

ligand, for example (EDTA, citrate) consistent with the mode of Fe uptake

occurring in natural waters where Fe is generally buffered by chemically

heterogeneous natural organic matter (NOM)?

Fe(III)' Fe(II)'

Fe(III)L Fe(II)L

kd1

kred1 ,

kox2 ,

kred2 ,

kf1kd2 kf2

kox1 ,

Ks

ρFeOxido-reductase

Fe-binding site

Outer membrane

; kox3 ,

O2

O2

O2 O2

•−O2•−

O2•−O2•−

O2•−O2•−

O2•−O2•−


9

(iv) What are the growth characteristics of Fe-limited continuous cultures of M.

aeruginosa and can a model of Fe uptake for this organism developed from

batch culture studies be used to describe Fe uptake by a continuous culture

of M. aeruginosa?

1.2. OBJECTIVES

The overall aim of this study is to investigate the Fe uptake kinetics of the bloom-

forming freshwater cyanobacterium M. aeruginosa under Fe limitation with particular

attention given to:

(i) Effect of light on the Fe uptake by M. aeruginosa in a chemically well-

defined culture medium (Fraquil*) in the presence of a single metal

chelator, ethylenediaminetetraacetic acid (EDTA);

(ii) Intracellular Fe transport processes of M. aeruginosa in Fraquil* medium

buffered by EDTA;

(iii) Fe uptake kinetics by M. aeruginosa in the presence of a natural organic

ligand, Suwannee River fulvic acid (SRFA); and

(iv) Characteristics of M. aeruginosa grown in iron-limited continuous culture.

1.3. LAYOUT OF THESIS

In order to achieve these objectives, the thesis is divided into seven chapters as

described below:

Chapter 1: Background information on the significance of Fe towards cyanobacteria,

literature review on iron transformations in natural waters and mechanisms of Fe

acquisition by phytoplankton as well as the objectives of the study are presented in this

chapter.

Chapter 2: Information on the experimental and computational methods used as well

as a description of analytical quality control measures used in this thesis is presented in


10

this chapter. Particular attention is given to description of the culturing conditions of

M. aeruginosa in batch and continuous incubations, the methods used to determine Fe

concentrations, the short-term radio-labeled 55

Fe uptake experiments and the methods

used to quantify cellular Fe quota.

Chapter 3: The effect of light on Fe transformation and uptake kinetics by M.

aeruginosa PCC7806 in Fraquil* (pH 8) with the free iron activity buffered by the

ligand EDTA is described in this chapter. A kinetic model for Fe acquisition by M.

aeruginosa based on photo-generation and subsequent uptake of unchelated Fe(II) is

presented.

Chapter 4: In this chapter, insight into kinetics of extracellular Fe transport to

periplasmic and cytoplasmic compartments for strains PCC7806 and 7005 of M.

aeruginosa in Fraquil* (pH 8) buffered by EDTA is obtained. Specifically, short-term

Fe accumulation in periplasmic and cytoplasmic compartments of Fe-limited M.

aeruginosa is examined using radiolabeled 55

Fe combined with the cold osmotic shock

method to quantify the amount of Fe in each compartment.

Chapter 5: Fe uptake by M. aeruginosa grown in batch culture in pH 8 Fraquil*

containing the natural organic compound Suwannee River fulvic acid (SRFA) using

short-term radiolabeled 55

Fe uptake assays is assessed in this chapter. A kinetic model

that describes extracellular Fe transformations is developed and used to elucidate the

key processes involved in cellular Fe uptake by M. aeruginosa.

Chapter 6: In this chapter, the development of a continuous culturing system

(chemostat) made of metal-free materials, used to maintain cultures of M. aeruginosa

PCC7806 in both nutrient-insufficient and nutrient-replete Fraquil* media at

nanomolar iron (Fe) concentrations (20-100 nM total Fe) in which Fe availability

limited growth is described. A modified chemostat theory for Fe-limited

phytoplankton growth is developed and applied to describe the behaviour (including

steady state cell density, Fe cell quota and Fe uptake kinetics) of M. aeruginosa strain

PCC7806 grown continuously in Fraquil* and nutrient-replete Fraquil

* media with Fe

activity buffered by the organic ligand EDTA.


11

Chapter 7: In this final chapter, the major findings of this research are summarised

and recommendations for future studies provided.

12

CHAPTER 2

GENERAL METHODOLOGY

Chapter 2. General Methodology

13

2.1. REAGENTS

All reagents were prepared in ultrapure water (18 MΩ·cm resistivity Milli-Q water,

MQ) and all solutions stored in the dark at 4oC when not in use, unless stated

otherwise. All chemicals used were of high purity (at least analytical grade). All pH

measurements were made using a pH meter (pH/ION 340i, WTW, Germany)

calibrated by pH 4.01 and pH 6.88 buffers on the free hydrogen ion activity scale.

Adjustment of pH was performed using 1 and 5 M HCl and NaOH solutions, which

were prepared from highly purified 30% w/v HCl (Fluka) and NaOH (Riedel-deHaën,

Germany). The plastic-ware used was soaked in 0.1 M HCl for at least a day, rinsed

with MQ and then dried prior to use.

2.2. CULTURING CONDITIONS

2.2.1. Culture Medium

A modified Fraquil medium (referred to as Fraquil*) designed for the study of trace

metal interactions with freshwater phytoplankton was used throughout and prepared as

previously described (Andersen, 2005). Briefly, all salt and trace metal stocks were

made up in MQ individually rather than as a mixture. Then, the stocks were mixed in

~1 L MQ, except for Fe and EDTA. The 1.0 × 10-3

M stock of ferric chloride (FeCl3,

Ajax Finechem, Australia) in 0.1 M HCl was mixed with a 1.3 × 10-2

M solution of

disodium ethylenediaminetetraacetic acid (Na2EDTA, Sigma) prior to mixing with the

other stock solutions in order to prevent precipitation of Fe(III). The pH of the medium

was then adjusted to 8, as in previous work (Fujii et al., 2010a). The medium was

sterilized using a 700 W microwave oven for 10 min in intervals of 3, 2, 3 and 2

minutes. After cooling to room temperature, filter-sterilized vitamin solution was

added. Final concentrations of the major salts, trace metals and vitamins in Fraquil*

were 2.6 × 10-4

M for CaCl2, 1.5 × 10-4

M for MgSO4, 1.0 × 10-3

M for NaNO3, 1.0 ×

10-5

M for K2HPO4, 5.0 × 10-4

M for NaHCO3, 1.0 × 10-3

M for HEPES, 1.6 × 10-7

M

for CuSO4, 5.0 × 10-8

M for CoCl2, 6.0 × 10-7

M for MnCl2, 1.2 × 10-6

M for ZnSO4,

1.0 × 10-8

M for Na2SeO3, 1.0 × 10-8

M for Na2MoO24, 3.0 × 10-7

M for thiamine HCl


14

(Vitamin B1, Sigma), 2.1 × 10-9

M for biotin (Vitamin H, Sigma) and 3.6 × 10-10

M for

cyanocobalamin (Vitamin B12, Sigma). For the Fe uptake and photochemical

experiments, culturing media were prepared with nutrient concentrations identical to

Fraquil* except that Fe and EDTA was omitted.

To avoid Fe contamination, the culture media were prepared inside a trace-metal clean

room supplied with HEPA-filtered air. Based on the Fe concentration in the reagent

grade salts using the manufacturers’ specifications, Fe contamination was calculated to

be far less than 20 nM Fe (which is the lowest concentration used in this work); thus

no additional cleaning procedures (e.g., Chelex treatment; (Andersen, 2005)) were

undertaken. In addition to this calculation, Fe contamination was measured to be below

detection limit (~1 nM) by using a colorimetric method combined with an Ocean

Optics 1 m spectrophotometry system. The apparatus and configuration of the

spectrophotometry system and detailed procedure used to measure total Fe

concentration can be found elsewhere (Fujii et al., 2008b).

2.2.2. Long-term Culturing Conditions

Toxic and non-toxic strains of M. aeruginosa (PCC7806 and PCC 7005, respectively)

were generously provided by Dr. Brett Neilan (Cyanobacteria and Astrobiology

Research Laboratory, University of New South Wales). Batch axenic cultures of M.

aeruginosa cells were maintained in Fraquil* under conditions where cellular growth

was moderately limited by Fe availability as documented earlier (Fujii et al., 2010a).

Briefly, cells were incubated at 27oC under a 14 h:10 h light:dark cycle. Light was

supplied by three cool-white fluorescent tubes (36W, 28 mm diameter, 1.2 m length,

Philips) with total radiation intensity of 157 µmol.m-2

.s-1

. In the long-term culturing

medium, total concentrations for Fe and EDTA were 0.1 µM and 26 µM, respectively.

Cells were regularly subcultured into fresh media when cultures reached stationary

growth phase (generally ~2 weeks after inoculation with initial cell density of ~104 cell

mL-1

). Cells in the cultures were enumerated using a Neubauer hemocytometer (0.1

mm depth) under an optical microscope (Nikon, Japan). All experiments were

performed using cells harvested by filtering during the light phase of the light:dark

cycle in exponential growth phase.


15

2.2.3. Continuous Culturing Apparatus

Continuous culturing systems (chemostats) have been used by several workers (Novick

and Szilard, 1950, Herbert et al., 1956, Tempest, 1969, Gerhardt and Drew, 1994,

Hoskisson and Hobbs, 2005). In this work, a chemostat system made of metal-free

materials was successfully developed and used to maintain cultures of M. aeruginosa

PCC7806 in both Fraquil* and nutrient-replete Fraquil

* media at nanomolar Fe

concentrations (20-100 nM total Fe) in which Fe availability limited growth.

Composition of the nutrient-replete Fraquil* medium will be discussed in details in

Chapter 6.

The metal-free sterile chemostat system was developed for four different flow-rates

with three replicates (Figure 2.1). The culture medium reservoirs consisted of two 2 L

polycarbonate bottles containing sterile Fraquil* medium with screw top caps vented

by a 0.22 µm air filter. Sterile fresh medium was distributed at four different flow-rates

into the 12 cultures in 250 mL polycarbonate culturing vessels using a high precision

24-channel peristaltic medium pump (Ismatec). Inline 0.22 µm-filtered air was

supplied by a 4-channel aquarium style diaphragm air pump (Aqua-one) to create the

positive pressure required in the head-space of the culture vessels. The air gap created

between the culture and waste vessels flushed out the excess culture over the elevated

weir into a 10 L polycarbonate waste vessel via an overflow vent. The volume of the

culture in each vessel was therefore maintained constant at approximately 200 mL. To

avoid sedimentation of cells, the culturing vessels were continuously gently shaken

using a Benchtop digital shaker (Thermoline Scientific, Australia) at a rotation rate of

135 ± 5 rpm. A sampling port was equipped with a sterile one-way sampling valve that

allowed sampling of the culture without bacterial or metal contamination.

The system was placed in an incubator (Thermoline Scientific, Australia) to control the

temperature at 27oC under a 14 hr:10 hr light:dark cycle with light intensity of 157

µmol photons m-2

s-1

. Prior to use, all chemostat apparatus was sterilized by

autoclaving. During this treatment, all materials were protected from bulk trace metal

contamination from metal leaching inside the autoclave by placement in a double-layer

plastic bag.


16

Figure 2.1. The chemostat culturing system consisting of non-metal materials. The

system was operated at four different dilution rates in triplicate.


17

2.3. SHORT-TERM IRON UPTAKE EXPERIMENT

The short-term uptake of radio-labelled 55

Fe by M. aeruginosa was measured by

incubating cells in Fraquil* containing

55Fe-ligand complex in the presence or absence

of light. Briefly, cells of the original batch or continuous cultures were harvested onto

a 25 mm diameter, 0.65 µm PVDF membrane (Millipore) during daytime in

exponential growth phase for the batch cultures and in steady state phase for the

continuous cultures of M. aeruginosa. To remove any adsorbed Fe from the cell

surface, washing solutions containing 50 mM Na2EDTA and 100 mM Na2oxalate at

pH 7 (hereafter referred to as “EDTA/oxalate”) and 2 mM NaHCO3 at pH 8 were used

(Tang and Morel, 2006). The filtered cells were washed gently at 1 mL.min-1

with

EDTA/oxalate solution (pH 7) and subsequently rinsed with 2 mM bicarbonate buffer

(pH 8) by continuously passing 5 mL of each solution through the harvested cells for

~10 min. Washed cells were re-suspended into Fe and EDTA-free Fraquil* medium at

densities of interest. The Fe uptake experiment was initiated by adding different pre-

equilibrated 55

FeIII

-ligand stock solutions with different Fe:ligand ratios into the

cultures to obtain expected concentrations of 55

Fe(III) and ligand. The solutions of

55Fe(III) complexed by ligand were made by mixing radiolabeled

55FeCl3 solution with

the ligand stock (i.e., EDTA, citrate or SRFA) in the bottom of a polypropylene

microtube, followed by addition of 2mM bicarbonate buffer (pH 8) to the mixture to

maintain pH 8 and let the Fe(III)-ligand stock stored for 24 h in the dark at ambient

temperature to equilibrate.

In some experiments, the culturing medium was prepared with addition of either a

single reagent such as: ascorbate (Asc), ferrozine (FZ), superoxide dismutase (SOD),

hydroxylamine hydrochloride (HH), xanthine/xanthine oxidase (X/XO), etc. or a

combination of these reagents depending on the purposes of the particular experiment

of interest. Cells were incubated at 27oC for 0-12 h in the presence and absence of light

and cut-off filters.

After incubation, cells were vacuum-filtered onto the 0.65 µm PVDF membrane filters,

and then rinsed three times with 1 mL of EDTA/oxalate solution and three times with

1 mL of 2 mM NaHCO3 solution (total rinsing time was ~10 min). The filtered cells

were then placed in glass scintillation vials with 5 mL of scintillation cocktail


18

(Beckman ReadyScint) for further measurement of radioactivity by using a Packard

TriCarb Liquid Scintillation Counter (see Section 2.4.2).

2.4. MEASUREMENT OF IRON

2.4.1. Measurement of FeIIFZ3 with Spectrophotometer

A Cary 50 Bio UV-Visible spectrophotometer (with a detection limit of ~0.1 µM) was

used to determine micromolar concentrations of the FeIIFZ3 by measurement of the

absorbance of the complex at 562 nm. Further details of the FZ method are provided

by Stookey (1970) and Viollier et al. (2000).

When determination of nanomolar concentrations of iron was required, a 1 m path-

length “waveguide” (LWCC Type II, World Precision Instruments) combined with an

Ocean Optics spectrophotometer was employed. The spectrophotometer configuration

consisted of a deuterium tungsten halogen light source (DH-2000), a UV-VIS

spectrophotometer (USB 2000 UV-VIS), two optical fibers (P400 UV-VIS) and an

Ocean Optics 1 m pathlength spectrophotometry system. In order to undertake the

analyses, Fe-free Fraquil* was initially introduced into the 1 m flow cell by pushing the

solution with a peristaltic pump and the absorbance at 562 nm was zeroed. FZ and

sample solutions were mixed and incubated under experimental conditions of interest

prior to being introduced into the 1 m cell, followed by monitoring absorbance at 562

nm for several minutes with the OOIBase 32 computer program provided by Ocean

Optics. The absorbance at 562 nm was baseline-corrected using the absorbance at 750

nm as a reference to obtain a stable absorbance.

A stock solution of Fe(II) was made by dissolving ammonium ferrous sulfate (Sigma)

in 1.0 mM HCl at a final concentration of 4 mM. Calibration of FeIIFZ3 concentration

was performed by adding known amounts of the Fe(II) stock to FZ solutions, yielding

molar absorptivity of ε562= ~28,000 M-1

.cm-1

.


19

2.4.2. Measurement of Radio-labeled 55

Fe with Scintillation Counter

Radioactivity of radio-labelled 55

Fe was measured using a Packard TriCarb Liquid

Scintillation Counter with quench correlation. Scintillation counts (counts per minute)

of the samples were converted to moles of Fe by using concurrent counts of 1-50 µL of

55Fe-ligand stock in 5 mL scintillation cocktail.

2.5. MEASUREMENT OF CELLULAR IRON QUOTA

In this study the cellular Fe quota of M. aeruginosa was quantified using either acid

digestion combined with spectrometry or the radiometry (radiolabeled 55

Fe) method as

described below.

2.5.1. Acid Digestion Combined with Spectrophotometry Method

In order to determine the intracellular Fe content of M. aeruginosa cells, 5 mL of the

culture grown in growth medium containing non-radiolabeled Fe was filtered on to a

0.65 µm PVDF membrane filter (Millipore) and gently rinsed three times with 1 mL of

EDTA/oxalate solution and three times with 1 mL of 2 mM NaHCO3 solution (total

rinsing time was ~10 min) in order to eliminate non-specifically adsorbed Fe from the

cell surface. The filter was then immersed in 2 mL of 50% HNO3 and heated at ~100

0C for 4 h. The digest solution was brought up to a final volume of 10 mL by MΩ

water and centrifuged (3000 rpm for 10 min) to separate the residue from the acid

soluble fraction. The supernatant was complexed with FZ by addition of 1 mM

hydroxylamine hydrochloride in the presence of 1 mM FZ (allowed to react overnight

to reduce all Fe to Fe(II)) followed by addition of 10 M ammonium acetate buffer (pH

9.5) to bring the pH of the final solution to a value of approximately 6 where FeIIFZ3 is

significantly formed (Tang and Morel, 2006, Fujii et al., 2010a). The total Fe

concentration was determined by quantifying the absorbance of the FeIIFZ3 complex at

562 nm using a 1 m Ocean Optics long pathlength spectrophotometry system as

described previously in Section 2.4.1.


20

2.5.2. Radiometry Method

M. aeruginosa cells were grown in growth medium containing radiolabeled 55

Fe

instead of non-radiolabeled Fe. The amount of 55

Fe incorporated within cells was

measured by filtering 1 mL of the culture on to a 25 mm diameter 0.65 µm PVDF

membrane (Millipore). The filtered cells were then gently rinsed with EDTA/oxalate

solution and 2 mM NaHCO3 solution (total rinsing time was ~10 min). Subsequently,

the washed cells were placed in glass scintillation vials filled with 5 mL of scintillation

cocktail. The activity of radioisotope 55

Fe in the washed cells was measured in a

Packard TriCarb Liquid Scintillation Counter as previously described in Section 2.4.2.

2.6. MODEL FITTING

In this thesis, the best fit of the model to the experimental data was determined by

using a least-squares method in which the mean square error between the model value

and the average of the experimental data was minimized. For the model fit, Microsoft

Excel was used.

2.7. ANALYTICAL QUALITY CONTROL

2.7.1. Procedural Blank

Procedural blanks were prepared and analyzed exactly like, and along with, the

samples. The procedural blanks provided an indication of the response of the

measurement system to a sample with zero concentration of analyte. In addition, the

procedural blanks provided an indication of analyte contamination that may occur

during sample preparation and analysis. The procedural blank responses could also be

used to estimate the detection limit of the measurement system. If the analyte

concentration of the procedural blank was less than the detection limit, no corrective

action was necessary.

2.7.2. Replication

Sample analysis was performed in triplicate, unless otherwise stated.

21

CHAPTER 3

EFFECT OF LIGHT ON IRON UPTAKE

BY THE FRESHWATER

CYANOBACTERIUM MICROCYSTIS

AERUGINOSA

Chapter 3. Effect of Light on Iron Uptake by the Freshwater Cyanobacterium Microcystis aeruginosa

22

3.1. INTRODUCTION

Iron (Fe) is an essential element for primary producers because it plays critical roles in

various metabolic processes including photosynthesis, respiration and nitrogen fixation

(Crichton, 2009), and limits phytoplankton growth in some open ocean waters because

of its low availability (e.g., Martin and Fitzwater (1988)). Although Fe is not generally

considered a major growth-limiting nutrient for freshwater phytoplankton (Hassler et

al., 2009), Fe nutrition still affects biosynthesis of primary and secondary metabolites

and cells must amend their physiological functions depending on Fe availability

(Alexova et al., 2011); for example, Fe availability may influence toxin gene

expression and toxin generation under low Fe conditions (Sevilla et al., 2008, Alexova

et al., 2011).

Fe(III) is thermodynamically favored in air-saturated surface waters because of rapid

oxidation of Fe(II) by dissolved oxygen at circumneutral pH (Pham and Waite, 2008a).

Nevertheless, substantial evidence exists that reduction of Fe(III) to Fe(II) is promoted

by photochemical and biochemical processes in surface waters (Fan (2008) and

references therein). Consequently, analytically measureable amounts of Fe(II), which

is more available than insoluble Fe(III) for biological uptake (Shaked et al., 2005, Rose

et al., 2005), are formed in surface waters (e.g., Waite et al. (1995)). Light-mediated Fe

redox reactions such as ligand-to-metal charge transfer (LMCT) and reactions with

secondarily produced organic and inorganic radicals may play critical roles in the

formation of Fe(II) during the daytime (Fan, 2008, Rose and Waite, 2006, Waite and

Morel, 1984). During LMCT, irradiation of Fe(III) complexes with aminocarboxylate,

carboxylate and catecholate ligands (among others) leads to the oxidation or

decarboxylation of the ligands at certain sites, and induces Fe(III) reduction by

electron transfer to the metal center (Barbeau et al., 2001, Sunda and Huntsman, 2003).

Kinetic modeling of LMCT Fe transformations indicates that the photoreductively

mediated dissociation of chelated Fe(III) and subsequent oxidation of unchelated Fe(II)

(Fe(II)') to unchelated Fe(III) (Fe(III)') markedly increases the concentration of

unchelated Fe (Fe') in sunlit surface waters (Sunda, 2001). Laboratory and shipboard

culturing studies have consistently shown that exposure to natural sunlight enhances

the availability of Fe when bound to various organic ligands (in situ chelators,


23

aminocarboxylates and siderophores) for uptake by in situ and laboratory strains of

marine phytoplankta by a factor of 2-15 (Barbeau et al., 2001, Maldonado et al., 2005,

Anderson and Morel, 1982) though exceptions have been reported (Hassler and Twiss,

2006).

Given the underlying concepts of classical and current Fe uptake models (Hudson and

Morel, 1990, Morel et al., 2008), where Fe' is considered the most assimilable pool of

Fe for phytoplankton, enhanced biological Fe uptake may be linked to increased Fe'

concentration in the light. However, these Fe uptake models have been essentially

developed from biological Fe assimilation assays undertaken in the absence of light.

Therefore, the mode of biological Fe uptake in the presence of light has yet to be

established despite the relatively well-defined phototransformations for certain Fe

complexes. In this work, we mechanistically examine the effect of light on Fe

transformation and uptake by a common bloom-forming freshwater cyanobacterium,

Microcystis aeruginosa, in Fraquil* (modified Fraquil) medium with the free iron

activity buffered by ethylenediaminetetraacetic acid (EDTA).

3.2. MATERIALS AND METHODS

3.2.1. Materials

Unless otherwise stated, chemicals were purchased, prepared and stored as described

in Section 2.1, Chapter 2. pH (values reported on the free hydrogen ion activity scale)

was measured using a pH/ION 340i pH meter (WTW, Germany). Only plasticware

was used for solution preparation, storage and sample incubation to prevent Fe

contamination. Plasticware was acid-cleaned by soaking in 0.1 M hydrochloric acid for

at least a day and rinsed with ultrapure Milli-Q water (MQ; Millipore, 18 MΩ.cm

resistivity) before use.

Long-term cell culturing was performed in Fraquil* medium at pH 8, prepared as

described (Andersen, 2005) excepting modified concentrations of 100 nM Fe and 26

µM EDTA (see Section 2.2.1, Chapter 2 for details). To avoid Fe contamination, the

culture medium was prepared inside a trace metal clean room supplied with HEPA-

filtered air. Fe contamination in the medium has been previously determined to be < ~1


24

nM (Fujii et al., 2010a), much less than the lowest Fe concentration used in this work.

For photochemical and Fe uptake experiments, Fraquil* media were prepared with the

omission of Fe or Fe and EDTA, but concentrations of all nutrients other than Fe and

EDTA were unchanged. Then an appropriate volume of pre-equilibrated FeIII

EDTA

stock was added shortly before the experiment. The FeIII

EDTA stock was prepared by

mixing either non-radiolabeled 1-10 mM FeCl3 (in 0.1 M HCl, Ajax Finechem,

Australia) or radiolabeled 8.3 mM 55

FeCl3 (in 0.5 M HCl, 185 MBq, Perkin-Elmer,

Australia) with an appropriate volume of 1-100 mM EDTA solution (pH 8, Sigma) in a

polypropylene microtube, followed by addition of 2-50 mM bicarbonate buffer (pH 8,

Sigma) to maintain pH 8. The FeIII

EDTA stocks were then stored for at least 2 h in the

dark at ambient temperature to equilibrate.

For use in Fe uptake and photo-reductive dissociation experiments, stock solutions of

ferrozine (FZ; 3-(2-pyridyl)-5,6-diphenyl-1,2, 4-triazine-4’,4’’-disulfonic acid sodium

salt, Sigma) were prepared in MQ at concentrations of 0.1 M. The pH of the solutions

was adjusted to 8.0 to avoid a significant pH change when added to the culture

medium.

3.2.2. Light Conditions

Most experiments and cell culturing were performed in a light- and temperature-

controlled incubator at 27oC (Thermoline Scientific). Light was vertically supplied by

three cool-white fluorescent tubes (36W, 28 mm diameter, 1.2 m length, Philips).

Samples were consistently incubated 10 cm from the fluorescent tubes, at which

distance total radiation intensity was determined to be 157 µmol quanta.m-2

.s-1

(part A

of Figure 3.1 for the emission spectrum) using an Ocean Optics USB 4000

spectrophotometer equipped with an optical fiber and cosine converter (CC-3-UV) that

was calibrated against a DH-2000 VIS light source. Samples were incubated in either

polystyrene spectrophotometer cuvettes (Starna) or polycarbonate vessels (Nalgene)

that minimally interfere with visible light transmission (see Figure 3.2 for absorbance

spectra). During dark incubations, samples were covered with aluminum foil.

Chapter 3. Effect of Light on Iron Uptake by the Freshwater Cyanobaterium Microcystis aeruginosa

Figure 3.1. Irradiation spectra emitted from the cool-white fluorescent tube of the culturing incubator in the (A) absence and (B

spectra were measured using an Ocean Optics USB 4000 spectrophotometer equipped with an optical fiber and cosine corrector le

VIS-light source (hydrogen lamp). The measurement was per

the cut-off light filter between the light source and the irradiance probe. The photon flux densities calculated for each wavelength


white fluorescent tube of the culturing incubator in the (A) absence and (B-H) presence

spectra were measured using an Ocean Optics USB 4000 spectrophotometer equipped with an optical fiber and cosine corrector lens (CC

light source (hydrogen lamp). The measurement was performed using SpectraSuite software in absolute irradiation mode. Various light spectra were obtained by placing

off light filter between the light source and the irradiance probe. The photon flux densities calculated for each wavelength range are s

25

H) presence of light filter treatments. The

ns (CC-3-UV) calibrated against a DH-2000

formed using SpectraSuite software in absolute irradiation mode. Various light spectra were obtained by placing

range are shown in Table 3.1.


26

Table 3.1. Photon flux density at each wavelength range.

Wavelength range

Photon flux density (µmol photons.m-2

.s-1

)

No filter Light filter treatments

400nm 450nm 500nm 550nm 600nm 650nm 700nm

400nm-450nm 29.1 25.4 13.5 1.4 1.5 1.0 0.5 0.4

450nm-500nm 23.9 22.0 22.1 11.1 4.2 0.8 0.5 0.4

500nm-550nm 38.1 36.3 37.4 36.6 25.5 1.0 0.8 0.6

550nm-600nm 25.4 24.2 24.9 24.7 22.1 9.1 4.8 0.4

600nm-650nm 33.8 32.2 33.1 33.0 31.8 29.8 19.1 0.6

650nm-700nm 6.2 5.7 5.6 5.7 5.4 5.2 4.7 2.2

Total 156.5 145.8 136.7 112.7 90.6 47.0 30.4 4.6

Chapter 3. Effect of Light on Iron Uptake by the Freshwater Cyanobaterium Microcystis. aeruginosa

Experiments examining the effect of

biological uptake employed visible long

Corporation), which permit the transmission of radiation at wavelengths that are

nominally longer than 400, 450, 500, 550, 60

cut-off is defined as the wavelength where ~50% of peak transmission occurs, light

transmission was typically completely blocked at a wavelength 25 nm less than this

nominal value (e.g., 475 nm for the 500 nm filter)

3.1). Experiments were conducted by placing samples in a cardboard box covered on

five sides with aluminum foil, with the front of the box (facing the light source) fitted

with a light filter to allow passage of specific wav

A control sample was incubated without placing any filters in the front of the box.

Figure 3.2. Absorbance spectra for plastic and glass vessels. (A) blank (no materials),

(B) 1 cm quartz spectrophotometer cuvette (

Scintillation glass vials (20 mL, Crown Scientific), (D) 1 cm polystyrene

spectrophotometer cuvette (Starna Pty Ltd, Australia), (E) polycarbonate container

Effect of Light on Iron Uptake by the Freshwater Cyanobaterium Microcystis. aeruginosa

Experiments examining the effect of irradiation wavelength on Fe photochemistry and

biological uptake employed visible long-wave-pass edge filters (50 mm sq., Andover


nominally longer than 400, 450, 500, 550, 600, 650 or 700 nm. Although the nominal

off is defined as the wavelength where ~50% of peak transmission occurs, light


nominal value (e.g., 475 nm for the 500 nm filter) (parts B-H Figure 3.1 and Table

Experiments were conducted by placing samples in a cardboard box covered on


with a light filter to allow passage of specific wavelengths of light as described above.


Absorbance spectra for plastic and glass vessels. (A) blank (no materials),

(B) 1 cm quartz spectrophotometer cuvette (Starna Pty Ltd, Australia), (C)



Effect of Light on Iron Uptake by the Freshwater Cyanobaterium Microcystis. aeruginosa

27

irradiation wavelength on Fe photochemistry and

pass edge filters (50 mm sq., Andover


0, 650 or 700 nm. Although the nominal

off is defined as the wavelength where ~50% of peak transmission occurs, light


H Figure 3.1 and Table

Experiments were conducted by placing samples in a cardboard box covered on


elengths of light as described above.


Absorbance spectra for plastic and glass vessels. (A) blank (no materials),

Starna Pty Ltd, Australia), (C)




28

(250 mL, Nalgene), (F) high-clarity polypropylene tube (15 mL, BD Falcon), (G)

polypropylene microtube (1.5 mL, Eppendorf), (H) high-density polyethylene bottle

(125 mL, Nalgene). The absorbance spectra were measured using a Varian Cary 50

UV-Vis spectrophotometer (Scan mode). During measurement, the containers were

filled with ultrapure water (Milli-Q water). For large materials, the sample holder was

removed from the instrument and the materials were placed between the light source

and detector.

3.2.3. Photochemical Experiments

The rate of unchelated Fe(II) formation during FeIII

EDTA photolysis was

spectrophotometrically determined using ferrozine (FZ), a strong Fe(II) complexing

agent. Experiments were initiated by spiking appropriate volumes of the pre-

equilibrated FeIII

EDTA and 0.1 M FZ (pH 8, Sigma) stock solutions into Fe and

EDTA-free Fraquil* at final concentrations of 1-10 µM for Fe, 26 µM for EDTA and 1

mM for FZ, and samples incubated under light or dark conditions in the presence and

absence of light filters. At various incubation times (0-10 h) after adding FeIII

EDTA

and FZ, the FeIIFZ3 concentration of the sample was measured by monitoring the

absorbance at 562 nm where FeIIFZ3 absorbs most strongly (Thompsen and Mottola,

1984). Absorbance was measured using an Ocean Optics spectrophotometry system

with a 1 m long path length flow cell for the 1 µM Fe system and a Varian Cary 50

UV-VIS spectrophotometer with 1 cm path length cuvette for the 10 µM Fe systems

(see Section 2.4.1. for details). The effect of Fe contamination in reagents on the Fe(II)

formation rate was examined by repeating the experiments without adding Fe(III) to

the sample.


Fe Uptake Experiments

Cells of the original cultures of M. aeruginosa PCC7806 were harvested onto a 25 mm

diameter, 0.65 µm PVDF membrane (Millipore) during daytime in exponential growth

phase at densities of 1-2 × 106

cell.mL-1

. To remove any adsorbed Fe, the filtered cells

were washed gently at 1 mL.min-1

with EDTA/oxalate solution (pH 7) and

subsequently rinsed with 2 mM bicarbonate buffer (pH 8) by continuously passing 5

mL of each solution through the harvested cells for ~10 min. Washed cells were re-


29

suspended into Fe and EDTA-free Fraquil* medium at densities of 2.6 × 10

5 to 2.7 ×

107 cell.mL

-1. The Fe uptake experiment was initiated by adding the pre-equilibrated

55Fe

IIIEDTA stock at final concentrations of 200 nM for

55Fe and 26-260 µM for

EDTA. In some experiments, the culturing medium was prepared with addition of 1

mM FZ. In all cases, cells were incubated for 2 h in the presence and absence of light

and cutoff filters. All Fe uptake experiments were performed under nonsaturating

conditions (i.e., measured uptake rate was less than the maximum uptake rate of 2.0

amol.cell-1

.hr-1

(Fujii et al., 2010a) at least by a factor of 5). Under the Fe

concentration and cell densities examined here, the amount of 55

Fe taken up by cells

during the short-term incubation were calculated to be only 3% of total 55

Fe in culture

medium at a maximum.

After incubation, cells were vacuum filtered on to 0.65 µm filters, rinsed three times

with 1 mL of EDTA/oxalate solution then three times with 1 mL of 2 mM NaHCO3

(total rinsing time was ~10 min). The filtered cells were placed in glass scintillation

vials with 5 mL of scintillation cocktail (Beckman ReadyScint) and their activity

measured in a Packard TriCarb Liquid Scintillation Counter, with scintillation counts

(counts per minute) of the samples converted to moles of Fe using concurrent counts of

1-5 µL of 55

FeEDTA stock in 5 mL scintillation cocktail. Process blanks were

determined by performing the procedure in the absence of cells.

3.3. RESULTS AND DISCUSSION

3.3.1. Effect of Light on Photoreductive Dissociation and Fe Uptake

Photoreductive dissociation of FeIII

EDTA in Fraquil* was examined by measuring

time-dependent Fe(II) formation in the presence of FZ. During irradiation (by the

fluorescent tubes in the culture cabinet), the concentration of FeIIFZ3 complex

increased linearly with time (R2 > 0.99). Time-dependent increases were negligible or

very small in the dark and in the absence of Fe. A concentration of 1 mM FZ was

chosen as Fe(II)' complexation by FZ at this concentration is rapid and should

outcompete other reactions involving Fe(II)' such as oxidation by dioxygen and

recomplexation by intact EDTA. Under these conditions therefore, assuming that the

rate of FeIIFZ3 formation is equal to the rate of Fe(II)' production would appear


30

reasonable. Because FeIIFZ3 formation followed a first-order relationship with respect

to FeIII

EDTA concentration, the rate of photochemical Fe(II)' generation in our system

can be written as:

d[FeIIFZ3]

dt=

d[Fe(II)']

dt= khv[Fe

IIIEDTA] (3.1)

where khv is a first-order rate constant for photoreductive dissociation of FeIII

EDTA

under the conditions examined. Approximating III II

T 3[Fe EDTA] [Fe] - [Fe FZ ]≈ and

[FZ] ≈ [FZ]

T (where subscript T indicates total concentration) followed by integration

of the resulting expression yields a relationship between [FeIIFZ3] and time (see part

A1.1 and Figure A1.1 of Appendix 1 for details). As shown in part A of Figure 3.3, khv

was determined to be 6.5 (± 0.25) × 10-6

s-1

for 1 µM total Fe and 6.2 (± 0.20) × 10-6

s-1

for 10 µM total Fe by linear regression analysis. Effect of total Fe concentration on khv

was statistically insignificant. In addition, the reaction proceeded in a first-order

manner with respect to total Fe concentration. These results suggest that secondary

radicals such as reactive oxygen species, which could be generated to a larger extent at

higher Fe concentrations or accumulate with time, play a minor role in Fe(II)'

formation. The minor effect of secondary radicals is probably due to either the

presence of radical scavengers in the medium (e.g., copper) or insignificant

participation of secondary radicals in the Fe(II)'-producing photoredox processes.

The effect of light on 55

Fe uptake rate was examined under conditions identical to

those used in the photochemical experiments except for the presence of M. aeruginosa

cells (exponential growth phase, ≤ 3.5 × 106

cell.mL-1 ) and the use of different

concentrations of EDTA and radiolabeled 55

Fe(III) (as a replacement for

nonradiolabeled Fe). In the absence and presence of light, 55

Fe accumulated in cells

linearly with time over the 2 h incubations (part B of Figure 3.3). At 26 µM EDTA and

200 nM Fe, the 55

Fe uptake rate (0.33 ± 0.065 amol.cell-1

.hr-1

, n=13) under the light

was two orders of magnitude greater than that under darkness (0.003 ± 0.006 amol.cell-

1.hr

-1, n=3), indicating a significant role of light in Fe uptake.

Chapter 3. Effect of Light on Iron Uptake by the Freshwater Cyanobaterium Microc

Figure 3.3. Effect of light on Fe(II)' formation and

[EDTA]T = 26 µM, [FZ]T = 1 mM) and (B) 55

Fe uptake


formation and 55

Fe uptake by M. aeruginosa. Time-courses of (A) FeIIFZ3 formation ([Fe]

Fe uptake ([55

Fe]T = 200 nM and [EDTA]T = 26 µM). Effect of light wavelength on (C)

31

formation ([Fe]T = 0, 1 or 10 µM,

. Effect of light wavelength on (C) 55

FeEDTA


32

uptake ([55

Fe]T = 200 nM, [EDTA]T = 26 µM) and (D) FeIIFZ3 formation ([Fe]T = 10 µM, [EDTA]T = 26 µM, [FZ]T = 1 mM). The incubations

were performed in modified Fraquil* (pH 8) in the light or dark at 27

oC. In the light filter treatments (panels C and D), filters were placed

between the incubated samples and the light source to allow transmission of wavelengths longer than 400, 450, 500, 550, 600, 650 or 700 nm. In

the control treatment, no light filter was inserted in front of the sample. Incubations were performed for 4 h for the photo-reduction experiment

and 2 h for the 55

Fe uptake experiment. Asterisks indicate that light filter treatments were significantly different from the control at the p < 0.05

level using a single-tailed heteroscedastic t-test. Symbols and error bars represent average data ±standard deviation from duplicate (photo-

reduction) or triplicate (55

Fe uptake) experiments. Solid lines represent linear regression.


33

3.3.2. Effect of Light Wavelength

Transport of Fe across the (cyto)plasmic membrane, is likely energy-dependent

process (Andrews et al., 2003). Thus, it is possible that the Fe uptake system in the

presence of photosynthetically active radiation (PAR) can be activated by immediate

use of the energy (e.g., ATP) produced during photosynthesis. To investigate this, we

determined which wavelengths (λ) of light particularly affected Fe uptake and

photoreduction. Use of cutoff filters to manipulate the wavelengths of light reaching

the culture indicated that only light from λ = 400-500 nm significantly contributed to

55Fe uptake by M. aeruginosa, with the uptake rate at these wavelengths comparable to

that in the control treatment (no light filter) (part C of Figure 3.3). Irradiation with

light at λ = 500-700 nm, which accounts for the majority (66%) of the total photon flux

(Table 3.1) and which has a substantial impact on cyanobaterial photosynthesis (Kirk,

1994), had almost no effect on cellular Fe uptake (part C of Figure 3.3). Similarly,

significant FeIIFZ3 formation only occurred when the sample was exposed to light in

the range λ = 400-500 nm (part D of Figure 3.3). In contrast to the total photon flux

density, the FeIIFZ3 formation rate was a linear function of

55Fe uptake rate (Figure

3.4), supporting the contention that specific light wavelength rather than total intensity

is important for the short-term 55

Fe uptake and photoreduction. These results strongly

suggest that photoinduced abiotic Fe transformations rather than biological factors are

more important for cellular Fe uptake. Such an interpretation is consistent with

previous studies (Anderson and Morel, 1982, Hudson and Morel, 1990) showing that

Fe uptake by coastal phytoplankton (the coccolithophorid Pleurochrysis carterae and

diatom Thalassiosira weissflogii) grown under Fe limitation was affected negligibly or

only slightly (by 0-37% relative to dark) by any light-mediated increase in cellular

metabolism. Although the absorbance of FeIII

EDTA rapidly decreases in the visible

light region (Figure 3.5), this complex is still capable of capturing some light at 400-

500 nm. Over this wavelength range, the average quantum yield for FeIII

EDTA (ϕ =

[FeIIFZ3 formation rate]/[number of photons absorbed]) was 0.010 (Table 3.2),

comparable to the published value at similar pH and irradiation wavelength (ϕ =

0.011, when quantum yield data ranging from ϕ = 0.005 at 500 nm to ϕ = 0.02 at 400

nm are averaged (Kari et al., 1995)).


34

Figure 3.4. Relationships between (A) FeIIFZ3 formation rate and

55Fe uptake and (B)

total photon flux density and 55

Fe uptake rate. At each data point, the parameters were

obtained from the incubation experiments and measurements of irradiation spectra

using the same cutoff filter. Thus, the data for 55

Fe uptake and FeIIFZ3 formation rate

are the same as those shown in parts C and D of Figure 3.3. Details of total photon flux

density are listed in Table 3.1.


Figure 3.5. UV-VIS absorbance spectra for Fe

= 0.5 mM and [EDTA]

8 buffered by 15 mM NaHCO

are also shown. The average molar

wavelength range from 400 nm to 500 nm was determined to be 37 M

Effect of Light on Iron Uptake by the Freshwater Cyanobaterium Microcystis aeruginosa

VIS absorbance spectra for FeIII

EDTA complex (solid line, [Fe(III)]

= 0.5 mM and [EDTA]T = 1.3 mM) and EDTA (dotted line, [EDTA]

8 buffered by 15 mM NaHCO3. Enlarged absorbance spectra in the visible light range

are also shown. The average molar absorptivity of FeIII

EDTA complex in the

om 400 nm to 500 nm was determined to be 37 M


35

EDTA complex (solid line, [Fe(III)]T

= 1.3 mM) and EDTA (dotted line, [EDTA]T = 1.3 mM) at pH

. Enlarged absorbance spectra in the visible light range

EDTA complex in the

om 400 nm to 500 nm was determined to be 37 M-1

cm-1

.


36

Table 3.2. Parameters used for the determination of quantum yield.

Parameters Symbol Value Unit

FeIII

EDTA concentration c 10 µM

Average molar absorptivity at 400-500nma)

ε 37.3 M-1

.cm-1

Length of light pathb)

l 1.0 cm

Absorbance A =εcl 2.0 × 10-4

cm-1

Incident light at 400-500 nmc)

I0 53.0 µmol.m-2

.s-1

Light absorbed Ia =I0(1-10-εcl

) 4.6 pmol.cm-3

.s-1

FeIIFZ formation rate

d) d[Fe

IIFZ3]/dt 0.046 pmol.cm

-3.s

-1

Quantum yield φ 0.010

a) The average molar absorptivity was determined from the Fe

IIIEDTA spectrum shown in Figure 3.5.

b) A 1 cm polystyrene spectrophotometer cuvette was used to incubate samples in the Fe

IIIEDTA photoreduction experiment at 10 µM Fe.

c) Values were calculated from Table 3.1.

d) Data from the photoreduction experiment at 10 µM Fe.


37

3.3.3. Fe Substrate for Uptake

A plausible explanation for the mechanism of light-facilitated uptake is that Fe

availability increased due to an increase in the concentration of photo-produced Fe(II)'

and, potentially, Fe(III)'. The presence of 1 mM FZ significantly decreased Fe uptake

by 27-70% (p < 0.05, part A of Figure 3.6), supporting the hypothesis that

photoproduced Fe(II)' aided Fe uptake since, in this event, complexation by

membrane-impermeable FZ would be expected to inhibit cellular uptake. The lesser

effect of FZ at higher cell densities suggests that M. aeruginosa cells can effectively

compete with high concentrations of FZ for Fe(II). Such a high affinity of the cell for

Fe(II) may explain the previously reported absence of effect of FZ on Fe uptake by the

macrophytic cyanobacterium Lyngbya majuscula (Rose et al., 2005). A significant

decrease in rate of Fe uptake was also observed with increasing EDTA concentration

(part B of Figure 3.6), indicating that complexation by EDTA at concentrations

examined here effectively competes with cellular uptake, as direct FeIII

EDTA

acquisition is unlikely. The limited availability of 55

Fe-EDTA for dark uptake, even

after preliminary photolysis, consistently indicates that light-facilitated Fe uptake is

tightly coupled with the availability of photo-produced Fe. This Fe is available for

uptake during illumination but readily transforms to an unavailable form once

illumination ceases, presumably via oxidation of photoproduced Fe(II) following

complexation (part A1.2 and Figure A1.2 of Appendix 1).

Chapter 3. Effect of Light on Iron Uptake by the Freshwater Cyanobaterium Microcystis ae

Figure 3.6. Effect of (A) ferrozine (FZ) and (B) excess EDTA on

the light. The 55

Fe uptake experiment was undertaken by incubating cells in Fraquil

containing pre-equilibrated

200 nM, [EDTA]T = 26

× 105 – 2.7 × 10

7 cell.mL

constant (3.5 × 106 cell.mL

indicate that treatments with FZ or excess EDTA were significantly different from the

control ([Fe]T = 200 nM and

respectively, using a single

average data and errors represent ±standard deviation from triplicate experiments.

Effect of Light on Iron Uptake by the Freshwater Cyanobaterium Microcystis ae

Effect of (A) ferrozine (FZ) and (B) excess EDTA on

Fe uptake experiment was undertaken by incubating cells in Fraquil

equilibrated 55

FeIII

EDTA complex and FZ or excess

26-260 µM and [FZ]T = 1 mM). While various cell densities (2.6

cell.mL-1) were used in the FZ experiment, the cell density was kept

cell.mL-1) in the excess EDTA experiment. One and two asterisks


200 nM and [EDTA]T = 26 µM) at p < 0.05 and

respectively, using a single-tailed heteroscedastic t-test. Symbols and error bars are

errors represent ±standard deviation from triplicate experiments.


38

Effect of (A) ferrozine (FZ) and (B) excess EDTA on 55

Fe uptake rate in

Fe uptake experiment was undertaken by incubating cells in Fraquil*

FZ or excess EDTA ([Fe]T =

. While various cell densities (2.6

) were used in the FZ experiment, the cell density was kept

excess EDTA experiment. One and two asterisks


< 0.05 and p < 0.01 levels,

test. Symbols and error bars are

errors represent ±standard deviation from triplicate experiments.

Chapter 3. Effect of Light on Iron Uptake by the Freshwater Cyanobaterium Microcystis ae

3.3.4. Kinetic Model for

Transformation of Fe species under the various conditions employed in the Fe uptake

incubations was examined using a kinetic model based on the chemical reactions

shown in Table 3.3 and presented schematically in Figure 3.7.

Figure 3.7. Fe uptake model

Fe(II) (i.e., Fe(II)') is formed from the photoreductive dissociation of ferric EDTA

complex (FeIII

EDTA). The photoproduced Fe(II) subsequently passes through the

nonspecific outer membrane channel (por

uptake competes with Fe(II)

ferrozine (FZ) and excess EDTA if present at appropriate concentrations. Solid arrows

represent major reactions under conditions o

whereas dotted arrows indicate relatively minor reactions. Rate constants depicted near

the arrows correspond to those listed in Table 3.3.

Effect of Light on Iron Uptake by the Freshwater Cyanobaterium Microcystis ae

3.3.4. Kinetic Model for Fe Species



Table 3.3 and presented schematically in Figure 3.7.

Fe uptake model by M. aeruginosa in the presence of light. Unchelated

) is formed from the photoreductive dissociation of ferric EDTA


nonspecific outer membrane channel (porins) by diffusion. However, cellular Fe

uptake competes with Fe(II)' complexation by extracellular Fe-binding ligands such as


represent major reactions under conditions of the short-term 55

Fe uptake experiment,


the arrows correspond to those listed in Table 3.3.


39



in the presence of light. Unchelated

) is formed from the photoreductive dissociation of ferric EDTA


ins) by diffusion. However, cellular Fe

binding ligands such as


Fe uptake experiment,



40

Key processes in this model are photoreductive dissociation of FeIII

EDTA to Fe(II)',

oxygenation of Fe(II)' to Fe(III)', and formation and dissociation of FeIIEDTA. This

model is similar to that used by Sunda and Huntsman (2003) in describing the

photochemical transformation of FeEDTA complexes in seawater except that rate

constants differed as different media and light intensities were used, and

recomplexation of Fe(II)' followed by oxidation of the complex was also considered

here. Because inorganic Fe complexes are nonphotoreactive at circumneutral pH (King

et al., 1993), photo-reduction of Fe(III)' was ignored.

The rate constants for photo-reductive dissociation of FeIII

EDTA determined in this

work were reasonably similar to the values reported in seawater by Sunda and

Huntsman (2003) (4.4 × 10-6

s-1

at light intensity of 500 µmol-quanta.m-2

.s-1

) and

Anderson and Morel (1982) (1.7 × 10-6

s-1

at 95 µmol-quanta.m-2

.s-1

). However, in

contrast to the Sunda and Huntsman model (Sunda and Huntsman, 2003), Fe(III)'

formation via oxidation of photoproduced Fe(II)' is negligible in the freshwater model

presented here because complexation of Fe(II)' by EDTA or FZ occurs much faster

than Fe(II)' oxygenation (by a factor of at least 104 under the experimental conditions

described here, Table 3.4). Although rates of oxidation and photoproduction of Fe(II)'

are similar in both seawater and freshwater systems, recomplexation of Fe(II)' by

EDTA was ignored in the Sunda model as complexation of Fe' by EDTA in seawater is

very slow (23 M-1

.s-1

) due to substantial precomplexation of EDTA by Ca at the high

Ca concentrations in seawater, a subject that has been extensively discussed elsewhere

(Fujii et al., 2010a).


41

Table 3.3. Kinetic model for Fe transformation and uptake under the light.

Chemical reactions / diffusion parameters Rate constants/Parameter values

(a) Chemical reactions

FeIII

EDTA + hv → Fe(II)' + EDTAox khv 6.4 × 10-6 a)

s-1

Fe(II)' + EDTA → FeIIEDTA kf-EDTA 2.1 × 10

6 b) M

-1.s

-1

Fe(II)' + 3FZ → FeIIFZ3 kf-FZ 3.1 × 10

11 c) M

-3.s

-1

FeIIEDTA→Fe(II)' + EDTA kd-EDTA 1.2 × 10

-3 b) s

-1

FeIIFZ3→Fe(II)' + 3FZ kd-FZ 4.3 × 10

-5 c) s

-1

Fe(II)' + O2 → Fe(III)' + O2- kox 8.8

d) M

-1.s

-1

FeIIEDTA + O2 → Fe

IIIEDTA + O2

- kox-EDTA 31

e) M

-1.s

-1

Fe(II)' → uptake kup 3.9 × 10-9

L.cell-1

.s-1

(b) Diffusion parameters f)

Diffusion coefficient D 0.5-1 × 10–9

m2.s

-1

Radius of porin a 0.5-1.5 × 10–9

M

Length of porin channel l 2-7.5 × 10–9

M

Density of porin Nporin 0.79-3.3 × 1016

m-2

Surface area per cell As 1.1 × 10-10

m2

a) Average value for khv determined in the 1 µM and 10 µM total Fe systems. EDTAox represents photo-oxidized EDTA formed in the ligand-to-metal charge

transfer (LMCT) process. b)

kf-EDTA and kd-EDTA were determined in this work. See part A1.3 of Appendix 1 for detailed discussion of methods and results.

c) Thompsen and Mottola (1984).

d) Pham and Waite (2008a).

e) Fujii et al. (Fujii et al., 2010a).

f) Parameters for diffusion and bacterial outer-membrane porin

properties are carefully examined in part A1.5 of Appendix 1 (See also Tables A1.3 and A1.4 of Appendix 1).

Chapter 3. Effect of Light on Iron Uptake by the Freshwater Cyanobaterium Microcystis

Table 3.4. Measured and calculated Fe uptake parameters for various culture conditions.

a) The first-order oxygenation rate constant for Fe(II)' at pH 8 was calculated by multiplying the second

saturated dissolved oxygen concentration of 0.24 mM at 27

experiments. All experiments were performed under nonsaturating conditions.

eq. 3.2. Average and ±standard deviations for all

calculate the steady-state concentrations for Fe(II)' and Fe

were performed on different dates. Although the

were slightly different possibly due to the different preconditioning of cells used.


Measured and calculated Fe uptake parameters for various culture conditions.

order oxygenation rate constant for Fe(II)' at pH 8 was calculated by multiplying the second-order rate constant of 8.8 M

saturated dissolved oxygen concentration of 0.24 mM at 27oC.

b) Averaged

55Fe uptake rates and ±standard de

experiments. All experiments were performed under nonsaturating conditions. c)

kup value for each experimental condition was determined using

verage and ±standard deviations for all kup were also calculated. d)

See part A1.4 of Appendix 1 for detailed methods employed to

state concentrations for Fe(II)' and FeIIEDTA and time-averaged concentrations for Fe

IIFZ

were performed on different dates. Although the 55

Fe:EDTA ratio and cellular density were similar in these two experiments, the uptake rates

were slightly different possibly due to the different preconditioning of cells used.

42

order rate constant of 8.8 M-1

.s-1

by a

Fe uptake rates and ±standard deviations from triplicate

value for each experimental condition was determined using

A1.4 of Appendix 1 for detailed methods employed to

FZ3. e)

The 55

Fe uptake experiments

tio and cellular density were similar in these two experiments, the uptake rates


43

3.3.5. Fe Uptake Machinery

Under nonsaturating conditions, the rate of uptake of a particular substrate is

reasonably assumed to be proportional to the substrate concentration. Provided that the

steady-state concentration of Fe(II)' ([Fe(II)']ss) is controlled by photo-reductive

dissociation of FeIII

EDTA, complexation of photo-produced Fe(II)' by EDTA and FZ

(if present), dissociation of FeIIEDTA and Fe

IIFZ (if present), oxidation of Fe(II)' and

cellular uptake, then Fe uptake rate (ρFe mol.cell-1

.s-1

) can be described by:

ρFe

= kup

[Fe(II)']ss

=k

upk

hv[FeIIIEDTA] + k

d-EDTA[FeIIEDTA] + k

d-FZ[FeIIFZ

3]( )

kf-EDTA

[EDTA] + kf-FZ

[FZ]3 + k

up[cell] + k

ox[O

2]

(3.2)

where [cell] is the cell density (cell.L-1

) and relevant reaction details are listed in Table

3.3. As explained in part A1.4 of Appendix 1, the thermal dissociation of Fe(II)

complexes with EDTA or FZ significantly influences [Fe(II)']ss while Fe(II)' oxidation

has a negligible effect on this parameter. The uptake rate constant kup was estimated by

substituting 55

Fe uptake rates measured under various conditions together with other

known kinetic parameters into eq. 3.2 (see part A1.4 of Appendix 1 for details),

yielding an average value (± one standard deviation) of 3.9 (±1.9) × 10-9

L.cell-1

.s-1

(Table 3.4). Using this value, Fe uptake rates for a range of cell densities and

competitive ligand concentrations were calculated (Figure 3.8). This model

demonstrates that higher cell densities result in a decrease in [Fe(II)']ss and thus a

decrease in 55

Fe uptake rate (Figure 3.6). This situation will arise when the rate of

cellular uptake is comparable to or higher than the rate of formation of Fe(II)

complexes with EDTA and/or FZ (i.e., k

up[cell] ≥ k

f-EDTA[EDTA] + k

f-FZ[FZ]3 ).


Figure 3.8. Effect of competitive ligand concentrations and cellular densities on calculated Fe uptake rate using eq


concentrations and cellular densities on calculated Fe uptake rate using eq

44

concentrations and cellular densities on calculated Fe uptake rate using eq. 3.2


45

The apparent high affinity of the cell for Fe(II) was further examined by considering

the possibility of passive diffusion of photo-produced Fe(II)' to the intracellular space.

In cyanobacteria, nutrients must pass through the outer-membrane into the periplasmic

space prior to active translocation across the inner-membrane to the cytosol. Some

specific forms of Fe (e.g., ferric siderophore complexes) are recognized by outer-

membrane receptors and actively transported into the intracellular compartments,

however for most small-sized hydrophilic nutrients (including metal ions) transport

from the external environment into the periplasmic space most likely occurs via

movement through non-selective transmembrane channels called porins, which are

ubiquitous in almost all Gram-negative bacteria investigated so far including

cyanobacteria (Nikaido, 2003). Given the concentration gradient across the outer-

membrane, a relationship between uptake rate and diffusional flux of a substrate in a

nonreactive cylindrical channel (Jporin mol.porin-1

.s-1

) may be formulated as follows:

ρFe

=Jporin

Nporin

AS = − Dπa2 1,000∆[Fe']

l

N

porinA

S (3.3)

where D is the diffusion coefficient (m2.s

-1), a is the radius of a porin (m), l is the

length of the channel (m), ∆[Fe'] (= in out[Fe'] - [Fe'] ) is the difference between Fe'

concentrations inside and outside the outer-membrane (M), Nporin is the number of

porins per square meter and AS is the surface area of a cell (m2). The diffusion layer

thickness of metal complexes in aqueous solution is generally on the order of tens of

micrometers (Buffle et al., 2009) such that the metal flux in proximity of the cell

surface would not be influenced by physical diffusion in the case of small

phytoplankton such as M. aeruginosa (cellular radius ~4 µm). Assuming that

unchelated Fe which enters the periplasm is rapidly captured by periplasmic Fe

transporters under the nonsaturating conditions investigated here (i.e., [Fe']in ≈ 0), then

out[Fe'] ~ -[Fe']∆ = [Fe(II)']ss, leading to an alternative expression for the uptake rate

constant kup:

k

up=

1,000Dπa2Nporin

AS

l (3.4)


46

Calculation using literature values for the diffusion coefficient of metal ions and

reported values for the properties of porins from several Gram-negative bacterial

species (Table 3.3) indicates that diffusion of Fe through such channels is faster than

an active transport process by a factor of 101-10

3. The upper limit for calculated kup

(0.06-4.1 × 10-9

L.cell-1

.s-1

, Table A1.4 of Appendix 1) is in accordance with the

measured value (i.e., 3.9 × 10-9

L.cell-1

.s-1

), regardless of the fact that all parameters

used were determined for bacterial species other than M. aeruginosa. Any

underestimation of the calculated uptake rate using this model may be because the

permeability of the outer-membrane of M. aeroginosa (determined by porin diameter,

length and density and Fe concentration gradient across the membrane) is greater than

that estimated here based on reported porin characteristics for other bacteria. The

diffusional model developed here for cyanobaterial Fe uptake contrasts with the Fe(II)s

uptake model for eukaryotic phytoplankta in the sense that the latter model considers

assimilation of extracellular Fe(II) as a process mediated by active membrane

transporters (Shaked et al., 2005).

3.4. IMPLICATION OF FINDINGS

The rate of uptake of Fe by M. aeruginosa in EDTA-buffered medium increased

substantially in the presence of light due to the photochemical transformation, at

wavelengths <500 nm, of FeIII

EDTA to a more labile form of Fe suitable for cellular

uptake. The inhibitory effect of the presence of both FZ and excess EDTA on 55

Fe

uptake rates, combined with prediction of the Fe species present using kinetic

modelling, consistently suggests that photochemically formed Fe(II)' is the major

substrate for uptake. Negligibly small uptake rates of Fe in the dark indicate that Fe

acquisition dominantly occurs during the daytime. Therefore, Fe influx during the day

is likely adequate for the biological functioning of the organism, which also occur in

the dark (e.g., respiration and dark reaction of photosynthesis). Consistent with this

notion, the steady-state Fe uptake rate (0.25-0.32 amol.cell-1

.hr-1

) calculated by

multiplying specific growth rate (µ=0.56-0.74 day-1

) by Fe quota (QFe = 7.2 amol.cell-

1) for 100 nM Fe and 26 µM EDTA culture medium (Fujii et al., 2010a) is comparable

to the short-term 55

Fe uptake rate measured at 200 nM Fe and 26 µM EDTA (0.18-0.23

amol.cell-1

.hr-1

, averaged over a 14 h:10 h light/dark cycle).


47

Although the uptake model developed using the well-characterized ligand EDTA

should be useful in ascertaining likely Fe availability to phytoplankton in laboratory

studies and natural environments, caution should be exercised before extending the

model to other microorganisms or aquatic systems. In contrast to previous observations

for Fe-stressed M. aeruginosa PCC7806 during short-term 55

Fe uptake incubations

(Fujii et al., 2010a), some cyanobacteria (e.g. Anabaena and Synechococcus) have

been shown to upregulate a siderophore-mediated Fe uptake system under Fe stress

(Ito and Butler, 2005), which presumably dominates over Fe influx by passive

diffusion. The biological response to external environment conditions and its effect on

Fe chemistry must be carefully evaluated in such cases. Additionally, the thermal- and

photolability of any Fe complexes present are important determinants of biological

uptake as are specific variables that depend on the nature of the ligand. For example,

using identical methods and conditions to those employed in this work, khv values for

the weaker ligands citrate and Suwannee River fulvic acid were determined to be

around one order of magnitude higher than that for EDTA (khv = 3.2 × 10-5

s-1

for

citrate and 1.7 × 10-5

s-1

for SRFA, unpublished data) suggesting that the rate of

bioavailable Fe formation will also be an order of magnitude greater. Under these

conditions the uptake rate reaches saturation (i.e. ~1-3 amol.cell-1

.hr-1

, unpublished

data), when ligand concentrations similar to those employed here or in natural

environments (e.g., <~10 mg.L-1

for SRFA) are used. When Fe transporters resident in

the periplasm or inner-membrane are saturated with Fe, the assumption that

out[Fe'] ~ -[Fe']∆ will not be valid and the Fe influx through the outer-membrane will

be influenced by the free Fe concentration in the periplasm. Such effects should be

investigated in future work by undertaking studies over a wider range of available Fe

concentrations. Regardless of the limitations of the model, this study provides

compelling evidence of the connection between Fe photolysis and biological uptake

and, as such, furthers understanding of Fe uptake by phytoplankton in natural waters.

48

CHAPTER 4

KINETICS OF EXTRACELLULAR IRON

TRANSPORT TO PERIPLASMIC AND

CYTOPLASMIC COMPARTMENTS OF

THE FRESHWATER CYANOBACTERIUM

MICROCYSTIS AERUGINOSA

Chapter 4. Kinetics of Extracellular Iron Transport to Periplasmic and Cytoplasmic Compartments of

the Freshwater Cyanobacterium Microcystis aeruginosa

49

4.1. INTRODUCTION

Iron (Fe) is of great importance for the growth of phytoplankton including

cyanobacteria as a cofactor of a range of proteins responsible for primary metabolic

processes including photosynthesis, chlorophyll synthesis, nitrogen fixation and

respiration as well as regulation of incidental oxidative stress (Crichton, 2009). Fe is

recognized as one of the critical nutrients limiting the growth of phytoplankton in one-

third of the world’s open ocean (Boyd et al., 2007) and some specific areas of coastal

and fresh waters where Fe availability is low (Hutchins and Bruland, 1998, Nagai et

al., 2006). Fe nutrition has also been found to be highly relevant to the biosynthesis of

secondary metabolite cyanotoxins, although the physiological functioning of these

mysterious molecules is still under debate (Alexova et al., 2011).

Over the last three decades, the kinetics and mechanisms of Fe acquisition by

phytoplankton have been studied with particular attention given to eukaryotes

(Anderson and Morel, 1982, Hudson and Morel, 1990, Shaked et al., 2005, Morel et

al., 2008) partially due to the relative abundance of diatoms in coastal and marine

ecosystems (Boyd et al., 2007). Although a number of recent studies have been

undertaken with a view to clarifying the mode of Fe uptake by cyanobacteria (Fujii et

al., 2010a, Fujii et al., 2011a, Salmon et al., 2006), existing kinetic models for Fe

uptake assume that only a single (plasma-)membrane acts as a major “permeability

barrier” in the selective transport of extracellular Fe into the cytoplasm. Cyanobacteria

are taxonomically classified as gram-negative bacteria, which possess two structurally

different phospholipid membranes surrounding the cellular body (i.e., the outer and

inner membranes) in addition to a thylakoid membrane, which is noncontiguous with

the inner plasma-membrane in the case of phototorophs (Spence et al., 2003).

Therefore, to transport nutrients from the external environment to the thylakoid or

cytoplasm where most nutrient-dependent biochemical metabolic reactions take place,

the nutrients must sequentially cross these membranes as well as the periplasmic space

between the outer and inner membranes.

The permeability of the lipid bilayer outer membrane depends on the type of transport

system involved (i.e., specific or non-specific transport) and the physicochemical



50

properties of substrates including size, polarity and charge as well as specificity to

outer membrane receptors. In some cases, uptake of specific substrates such as ferric

siderophore complexes and vitamin B12 is facilitated by specific recognition of these

substrates by outer membrane receptors (e.g., FepA and FhuA for enterobactin and

ferrichrome respectively) and subsequent active transport to intracellular

compartments via what is known as the Ton system, involving a giant protein complex

spanning from the inner to outer membrane (Faraldo-Gomez and Sansom, 2003,

Andrews et al., 2003). In contrast to such a high-affinity transport system, most small

hydrophilic nutrients (including ionic metals) are passively transported to the

periplasm via an alternative relatively lower-affinity pathway through water-filled

transmembrane channels embedded in the outer membrane, which is otherwise

impermeable to hydrophilic molecules (Hoiczyk and Hansel, 2000, Nikaido, 2003,

Jones and Niederweis, 2010). Pore-forming protein channels characterized by a β-

barrel structure (referred to as porins) are the major type of protein that is ubiquitously

found in the outer membrane of almost all gram-negative bacteria investigated so far,

including cyanobacteria (Hoiczyk and Hansel, 2000, Nikaido, 2003). Previous work

has indicated that this type of membrane channel in proteobacteria allows nonselective

permeation of hydrophilic molecules with size less than ~550-650 Da (Nikaido, 1976,

Nikaido, 1979) with the movement of nutrient in the porin most likely driven by

molecular diffusion or in some cases by slight interaction with low-affinity binding

sites (Nikaido, 2003, Pages et al., 2008). Although such information for cyanobacteria

is limited, recent studies using Fe-limited freshwater cyanobacteria Microcystis

aeruginosa PCC7806 and Anabaena flos-aquae UTEX1444 have provided indirect

evidence that only unchelated Fe is capable of permeating the outer membrane during

a short-term incubational assay with this process being independent of the siderophore-

mediated system (Fujii et al., 2010a, Wirtz et al., 2010).

Mechanisms involved in intracellular Fe transport may be quite different for the outer

membrane, periplasm and inner plasma-membrane. In the periplasm, putative

periplasmic Fe binding proteins (FutA) have been identified for cyanobacteria such as

Microcystis (Alexova et al., 2011) and Synechocystis (Katoh et al., 2001, Waldron et

al., 2007, Badarau et al., 2008), implying the selective transport or storage of Fe in this

compartment. Fe translocation across the inner membrane of proteo- and cyanobacteria

is most likely an energy-dependent process mediated by ATP-binding cassette (ABC)



51

type transporters for ferrous (FeoB) (Velayudhan et al., 2000, Andrews et al., 2003)

and ferric iron (FutB and FutC) (Katoh et al., 2001) in contrast to the diffusive

permeation of nutrients through the outer membrane.

Given the presence of an additional compartment enclosed by the outer and inner lipid

bilayers in cyanobacterial cells, the Fe uptake machinery for this type of phytoplankton

must couple the process of Fe entering the outer compartment with energy-dependent

Fe transport to the inner compartment. To our knowledge, there are no detailed studies

of Fe transport by cyanobacteria taking such intracellular compartments into account.

In this work we investigate the kinetics of transport of extracellular Fe to periplasmic

and cytoplasmic spaces in the freshwater cyanobacterium Microcystis aeruginosa by

measuring the accumulation kinetics of radio-labeled 55

Fe in chemically well-defined

media. Periplasmic and cytoplasmic 55

Fe concentrations were monitored using the cold

osmotic shock technique that has previously been used for the extraction of

periplasmic substances in a range of gram-negative bacterial species including

cyanobacteria (Neu and Heppel, 1965, Fulda et al., 1999). A recently developed

kinetic model (Fujii et al., 2010a) is extended to accommodate the Fe uptake data

obtained here.


4.2.1. Reagents

Unless otherwise stated, chemicals were purchased, prepared and stored as described

in Section 2.1, Chapter 2. For cell culturing, Fraquil* medium was prepared at pH 8

(see Section 2.2.1, Chapter 2 for the detailed preparation of Fraquil* medium) where

the transformation kinetics of extracellular Fe are well understood (Fujii et al., 2010a).

To avoid Fe contamination, the culturing medium was prepared in a trace metal clean

room supplied with HEPA-filtered air using reagent grade chemicals, resulting in Fe

contamination of less than 1 nM, as described elsewhere (Fujii et al., 2010a). Solutions

of 100 mM Na2EDTA (disodium ethylenediaminetetraacetate, Sigma) and 2.6-26 mM

Na3citrate (Sigma) were prepared at pH 8 as 55

Fe-binding ligand stock solutions.

Solutions of radiolabelled Fe complexes were then made by mixing 55

FeCl3 solution

(in 0.5 M HCl, 185 MBq, Perkin-Elmer, Australia) either with the Na2EDTA or



52

Na3citrate solutions in 1.5 mL polypropylene tubes followed by addition of 2 mM

NaHCO3 (Sigma, pH 8) to maintain circumneutral pH, providing final concentrations

of 24 µM 55

Fe, 34 mM EDTA, 0.17-6.9 mM citrate and 1-1.3 mM NaHCO3. The 55

Fe-

labelled complex stock solutions were equilibrated for 24 h prior to use.

A pH 7 solution containing 50 mM Na2EDTA and 100 mM Na2oxalate (Sigma)

(hereafter referred to as “EDTA/oxalate”) was used to remove Fe that was

nonspecifically retained on the cell surface and filter (Tang and Morel, 2006). A

solution of 2 mM NaHCO3 (pH 8) was also used to rinse the cells after the

EDTA/oxalate wash. Plasmolysis solutions at pH 8 were prepared with final

concentrations of 10 mM Tris-HCl (Sigma), 2 mM NaHCO3, 1 mM Na2EDTA and

either 0.5 M D-sorbitol, sucrose or NaCl (Sigma). All plasticware and glassware used

for cell culturing and analysis were acid cleaned using 0.1 M HCl.

4.2.2. 55

Fe Accumulation Experiments

The short-term accumulation rate of 55

Fe by M. aeruginosa strains PCC7806 and PCC

7005 was measured by incubating cells in Fraquil* containing

55Fe complexed by

either citrate or EDTA (see Section 2.3, Chapter 2 for the detailed description of the

short-term 55

Fe uptake experiment). Prior to the Fe accumulation assay, the cultured

cells were filtered on to a 25 mm diameter, 0.65 µm pore size PVDF membrane

(Millipore) and then rinsed firstly by passing 5 mL of EDTA/oxalate solution three

times and subsequently 10 mL of 2 mM NaHCO3 three times through the filter (total

rinsing time was ~10 min). The washed cells were then re-suspended into Fe- and

ligand free Fraquil* medium to provide cell densities of ~3 × 10

6 cell.mL

-1. Fe

accumulation experiments were initiated by adding a solution containing the 55

Fe-

ligand complex to the cultures at concentrations of 0.7 µM Fe, 2-11 mM EDTA and 5-

200 µM citrate. To examine the effect of Fe(II) and Fe(III) transformation on the rate

and extent of 55

Fe uptake, the assay was also undertaken in the presence of 1 mM

ascorbate and 100 µM ammonium tetrathiomolybdate (TTM). All Fe accumulation

experiments were conducted at 27oC under dark conditions to avoid complexities

associated with photochemical transformation of Fe complexes with EDTA and citrate.

After incubation, samples were harvested by vacuum-filtering on to 0.65 µm PVDF



53

membrane filters then rinsed three times with 1 mL of EDTA/oxalate solution and

once again with 1 mL of 2 mM NaHCO3 (total rinsing time was ~10 min).

To determine the amounts of 55

Fe accumulated in periplasmic and cytoplasmic spaces,

the harvested cells were subjected to cold osmotic shock in order to extract periplasmic

substances as a cold water fraction. In this process, filtered cells were incubated in 5

mL of 0.5 M D-sorbitol for 10 min. After vacuum-filtering to remove the solution, the

cells were then exposed to cold ultrapure water (5 mL, Milli-Q) for a further 10 min.

The cold water extract (containing the periplasmic 55

Fe) and the remaining cellular

mass on the filter (containing cytoplasmic 55

Fe as well as any Fe tightly bound to the

inner membrane) were each placed in glass scintillation vials. After the addition of 5

mL of scintillation cocktail (Beckman ReadyScint) to the fractionated samples, the

activity was measured in a Packard TriCarb Liquid Scintillation Counter. Scintillation

counts (counts per minute) of the samples were converted to moles of Fe by using

concurrent counts of 5-50 µL of 55

Fe-ligand stock in 5 mL of scintillation cocktail.

Process blanks were determined by performing the entire procedure in the absence of

cells.

The permeability of the outer membrane to Fe species was examined using the

plasmolysis solutions and bicarbonate buffer as the assay medium instead of Fraquil*.

Three types of plasmolysis solutions (D-sorbitol, sucrose and NaCl) were used to

examine the effect of solute molecular size on periplasmic 55

Fe accumulation. After

cells were resuspended in the assay medium, short-term 55

Fe accumulation assays were

initiated by adding 55

Fe-EDTA solution to the cultures at final concentrations of 0.7

µM 55

Fe and 2 mM EDTA. In the plasmolysis solution, the bacterial periplasmic space

can be enlarged to ~40-50% of total cell volume due to the high osmotic pressure

(Nikaido, 1979, Rose, 1982). In addition, in this assay, the presence of excess EDTA

ensured that 55

Fe was present almost exclusively in complexed form with minimal

transport of unchelated Fe into the cytoplasm. The assay was also undertaken at a

particularly high concentration of EDTA (11 mM) by using the plasmolysis solution

with D-sorbitol, which was prepared with 10 mM Na2EDTA. After incubating for 0.5

and 1 h at 27oC in the dark, cells were harvested by filtration and rinsed with the assay

medium. Total 55

Fe in the cells was then determined by collecting the filtered cells and

measuring radioactivity with the liquid scintillation counter.



54

4.2.3. Determination of Periplasmic Fe(II)

To determine the concentration of periplasmic Fe(II), the well-developed FeLume

chemiluminescence technique that has now been widely used for determination of

subnanomolar Fe(II) (King et al., 1995) was employed. In a manner identical to that

described for the 55

Fe accumulation experiment, M. aeruginosa PCC7806 cells were

harvested from the long-term incubation by filtration. The cells were then resuspended

in fresh Fraquil* which contains non-radiolabelled Fe and citrate at concentrations of

0.7 µM and 100 µM, respectively. Incubation was performed under dark condition

overnight. The incubation was also undertaken in the presence of 1 mM ascorbate and

100 µM TTM in order to examine the effect of Fe(II) and Fe(III) transformation on

Fe(II) accumulation in the periplasmic space. After the incubation, the cells were

filtered, washed with the EDTA/oxalate and bicarbonate solution, and subjected to the

cold osmotic shock protocol using the 0.5 M sorbitol solution. The cold water extract

(pH of 7.0-7.1) was immediately introduced to the flow cell of the FeLume system

where it was mixed with 0.5 mM luminol (5-amino-2,3-dihydro-1,4-phthalazinedione,

Sigma) chemiluminescence reagent prepared in 1 M ammonia (pH 10.3). The emitted

light arising from the reaction of the luminol reagent and Fe(II) present in the extract

was quantified by photomultiplier tube. The preparation of reagents, instrument

settings, measurement conditions and system calibration were all performed in a

manner identical to that described elsewhere (Fujii et al., 2010b). The total amount of

periplasmic Fe present was also determined by the procedure described above

involving the use of radiolabelled Fe.

4.2.4. Determination of Steady-state Concentration of Extracellular

Unchelated Fe

Speciation of extracellular Fe for the various conditions in the 55

Fe accumulation

experiments was calculated by use of the kinetic model described in Table 4.1A

(details of the calculation are provided in parts A2.1 and A2.2 of Appendix 2). As

discussed in Appendix A2.1, the effect of diffusional influx in the calculation of

steady-state concentration of extracellular unchelated Fe (Fe′) is negligible due to the

small concentration gradient created by extracellular and periplasmic Fe′.



55

4.2.5. Analysis of Genome Sequences

The genome sequences of M. aeruginosa strains PCC7806 and NIES843 (whose

genomes have been completely decoded (Kaneko et al., 2007, Frangeul et al., 2008)

were examined by use of the National Center for Biotechnology Information database.


4.3.1. Accumulation of 55

Fe in the Periplasm and Cytoplasm

In the Fe(III)-dominant system buffered by citrate (concentrations of Fe and citrate

were 0.7 µM and 20 µM, respectively, yielding an extracellular pFe′ of 9.3), 55

Fe

concentrations in the periplasmic and cytoplasmic fractions increased monotonically

and approached steady-state after several hours (Figure 4.1). The concentration of

periplasmic 55

Fe (< ~0.2 amol.cell-1

) was an order of magnitude less than the

cytoplasmic 55

Fe concentration (< ~5 amol.cell-1

) over the duration of the experiments

described here. Accumulation of periplasmic and cytoplasmic 55

Fe was also examined

over a range of Fe:citrate ratios yielding pFe′ from 8.6 to 10.8. The rate of cytoplasmic

55Fe accumulation was linearly correlated with the steady-state concentration of

periplasmic 55

Fe measured in 9 h incubations (R2 > 0.97 for both strains, part A of

Figure 4.2). The cytoplasmic accumulation rate also increased with increasing

extracellular Fe' concentration (part B of Figure 4.2).

In common culturing media with low osmotic pressure, the periplasmic space of gram-

negative bacteria has been determined to constitute approximately 5-40% of the total

cellular volume, depending on the type of microorganism and method of analysis

employed (Nikaido, 1979, Rose, 1982). If we assume that the concentration of 55

Fe in

the periplasm is comparable to that in the cytoplasm at saturation (e.g., after 5 h),

measured 55

Fe concentrations in each compartment under these conditions (Figure 4.1)

indicate that the periplasmic space has a volume of only 2.5-3% that of the cytoplasm

for both strains of M. aeruginosa. In the 1 h incubations,

the amount of 55

Fe

accumulated in the plasmolysed cells where the periplasmic space is enlarged (< 0.05

amol.cell-1

, pFe′ = 14.0, Figure 4.3) was even less than the periplasmic 55

Fe that was

extracted from the non-plasmolysed cells using cold osmotic shock (0.07-0.12



56

amol.cell-1

in 1h incubation with Fraquil*, part A of Figure 4.1). Addition of EDTA at

5.5-fold higher concentration (11 mM EDTA yielding pFe′ of 14.7) or use of media

with low osmotic pressure (Fraquil* and 2 mM bicarbonate buffer) yielded no

substantial changes in periplasmic 55

Fe accumulation.

Figure 4.1. Time course of 55

Fe accumulation in (A) periplasm and (B) cytoplasm for

M. aeruginosa strains PCC7806 (filled symbols) and PCC7005 (open symbols) grown

under moderate Fe limitation. Fe uptake assays were performed for 9 h in Fraquil* at

concentrations of 0.7 µM for 55

Fe and 20 µM for citrate. Symbols and error bars

represent the mean and ± standard deviation from triplicate experiments. Solid and

dashed lines represent the calculated values for PCC7806 and PCC7005, respectively,

using (A) eq. 4.7 and (B) the integrated form of eq. 4.5 with Fe uptake parameters

listed in Table 4.1. Detailed 55

Fe accumulation data are provided in Table A2.1 of

Appendix 2.

Cy

top

lasm

ic 5

5F

e

(am

ol.

cell

-1)

Pe

rip

lasm

ic 5

5F

e

(am

ol.

cell

-1)

0

1

2

3

4

5

6

7

0 2 4 6 8 10

B.

0

0.05

0.1

0.15

0.2

0.25

0 2 4 6 8 10

A.

Time (hr)



57

Figure 4.2. Cytoplasmic accumulation rate of 55

Fe by Fe-limited M. aeruginosa strains

PCC7806 (filled symbols) and PCC7005 (open symbols) as a function of (A) steady-

state concentration of total periplasmic 55

Fe and (B) calculated concentration of

unchelated Fe in the extracellular environment and periplasm. Data were obtained

from the assay using Fraquil* at concentrations of 0.7 µM for

55Fe and 5-200 µM for

citrate. Solid and dashed lines in panel A were determined for PCC7806 and

PCC7005, respectively, by linear regression analysis (p<0.05, n=15). In panel B, the

solid and dashed lines represent the calculated values using eq. 4.6 for PCC7806 and

PCC7005, respectively. Detailed 55

Fe accumulation data are provided in Table A2.2 of

Appendix 2. Calculated values of unchelated Fe concentrations are provided in Table

-20

-19.5

-19

-18.5

-18

-17.5

-17

-12 -11 -10 -9 -8

B.

Acc

um

ula

tio

n r

ate

of

cyto

pla

sm

ic 5

5F

e (

am

ol.

ce

ll-1

.hr-

1)

log [Fe'] (≈ log [Fe'peri] ) (M)

y = 10.0x - 0.22R² = 0.96

y = 6.9x - 0.14R² = 0.94

-0.5

0

0.5

1

1.5

2

2.5

3

3.5

4

0 0.1 0.2 0.3 0.4

A.

Steady-state periplasmic 55Fe (amol.cell-1)



58

A2.3 of Appendix 2. Symbols and error bars represent the mean and ± standard

deviation from triplicate experiments.

Figure 4.3. Cellular 55

Fe accumulation in the various culturing media at pH 8

(plasmolysis solutions, Fraquil* and 2 mM NaHCO3). Plasmolysis solutions were 0.5

M D-sorbitol, sucrose or NaCl (buffered by 10 mM Tris-HCl, 2 mM NaHCO3 and 1

mM for Na2EDTA) and 0.5M D-sorbitol with high EDTA concentration (10 mM). The

incubation experiment was initiated by addition of 55

FeEDTA to the culture media at

final concentrations of 0.7 µM for 55

Fe and 2 mM for EDTA. In case of the 0.5M D-

sorbitol solution containing high EDTA, the final concentration of EDTA was adjusted

to 11 mM. All incubations were performed for 30 or 60 min in the dark at pH 8 with

Fe-limited Microcystis aeruginosa (PCC7806).

0

0.01

0.02

0.03

0.04

0.05

0.06

0.07

0.08

0.09

0.1

30 min 60 min

D-sorbitol

Sucrose

Sodium Chloride

High EDTA

Fraquil*

Bicarbonate

Incubation period

Ce

llu

lar

55F

e a

ccu

mu

lati

on

(am

ol

cell

-1)



59

Table 4.1. Kinetic parameters used for the calculation of intracellular Fe transport and

extracellular Fe transformationa.

Parameters Unit

Microcystis strains

PCC7806 PCC7005

A. Rate constants for calculation of unchelated Fe concentration in extracellular

environment

kf-Cit b

M

-1.s

-1 2.1 × 10

5

kd-Cit c

s

-1 2.7× 10

-4 – 2.8 × 10

-3

kf-EDTA d

M

-1.s

-1 3.5 × 10

5

kd-EDTA e

s

-1 1.0× 10

-5

B. Parameters for calculation of Fe uptake and intracellular Fe transport

max

Fe'(peri)ρ f amol.cell

-1.h

-1 2.1 (±0.38)

*** 2.9 (±0.54)

***

Fe'(peri)K

f

pM 360 (±97)**

420 (±110)*

[Xperi]T g

amol.cell-1

0.30 0.29

k+1 h M

-1.s

-1 1.2 × 10

9 2.5 × 10

8

k-1 i s

-1 4.3 × 10

-1 1.0 × 10

-1

k2 j s

-1 1.9 × 10

-3 2.8 × 10

-3

kdif k

L.cell-1

.s-1

3.9 × 10-9

a Detailed procedure for determination of parameters was described in part A2.1 of

Appendix 2.

b kf-Cit: complexation rate constant of ferric citrate complex (Fujii et al., 2010a)

(reaction, f-Cit IIIFe(III)+Cit Fe Citk→ ).

c kd-Cit: dissociation rate constant for ferric citrate (Fujii et al., 2010a) (reaction,

d-CitIIIFe Cit Fe(III)+Citk→ ), which is a function of citrate concentration ([Cit]) due

to the formation of two mononuclear ferric citrate complexes (FeIII

Cit and FeIII

Cit2).

d kf-EDTA: complexation rate constant for ferric EDTA complex (Fujii et al., 2010a)

(reaction, f-EDTA IIIFe(III)+EDTA Fe EDTAk→ )

e kd-EDTA: dissociation rate constant for ferric EDTA complex (Fujii et al., 2010a)

(reaction, d-EDTAIIIFe EDTA Fe(III)+EDTAk→ ).

f To determine

max

Fe'(peri)ρ and Fe'(peri)K , the log-transformed eq 4.6 (eq. A2-1 of Appendix

2) was fitted to the data shown in part B of Figure 4.2 by nonlinear regression analyses



60

using R version 2.13.0 (free software for statistical computation). Asterisks represent

statistically significant levels as follows: ***, P < 0.001; **, P < 0.01; *, P < 0.05.

g [Xperi]T was determined from max

Fe'(peri)ρ = k2[Xperi]T.

h k+1 was determined from by fitting eq. 4.7 to the data shown in part A of Figure 4.1.

i k-1 was calculated from

Fe'(peri)K = (k-1+k2)/k+1.

j k2 was determined by linear regression analysis (eq. 4.5) to the data shown in part A

of Figure 4.2.

k Diffusional constant for unchelated Fe (Fujii et al., 2011a).

4.3.2. Fe Species Translocated from the External Environment to the

Periplasm

The apparent dependency of cytoplasmic Fe accumulation rates on the unchelated Fe

concentration (part B of Figure 4.2) suggests that Fe uptake is initiated by influx of

extracellular unchelated Fe into the periplasmic space. However, there is some

evidence that proteobacterial porin channels allow permeation of hydrophilic

molecules with molecular weight less than 650 Da (Nikaido, 1976, Nikaido, 1979),

potentially including chelated Fe as used in this work. Since it is difficult to identify

the Fe species resident in the periplasmic space when the periplasmic 55

Fe

concentration is substantially smaller than the total cellular 55

Fe concentration, 55

Fe

accumulation assays were performed in plasmolysis solutions where the periplasmic

space can be enlarged to approximately half of the total cell volume. As the

plasmolysis solutions contain excess EDTA, Fe transport to the cytoplasm is negligibly

slow under these conditions due to low Fe′ concentration (pFe′ > 14), and almost all

detectable 55

Fe in the medium and inside cells (if present) is expected to be present as

FeEDTA. We estimated that ~1.6-2.5 amol.cell-1

would be accumulated in the enlarged

periplasmic space if FeEDTA were to pass through the outer membrane. However,

55Fe accumulation in the plasmolysed cells was very small in 1 h incubations (< 0.05

amol.cell-1

; Figure 4.3). Therefore, the results support the notion that Fe bound to

relatively small metal-buffering organic ligands is unable to cross the outer membrane

of Microcystis, consistent with the previous finding that only unchelated Fe is capable

of crossing the outer membrane of the bacterium Micobacterium smegmai in citrate-

buffered medium (Jones and Niederweis, 2010).



61

The relatively low size-exclusion limit (less than ~340 Da) for porins in Microcystis

cells seems reasonable given the lower single channel conductance (i.e., size-

exclusion) of cyanobacterial porins (e.g., SomA) compared to typical proteobacterial

porins (e.g., OmpF and OmpC) (Hoiczyk and Hansel, 2000). It also has been reported

that while the complexation of polyamines to typical bacterial porin channels inhibits

substrate transport by modulating the channel proteins to a closed state, such treatment

does not significantly affect cyanobacterial Fe uptake (Sonier et al., 2011), suggesting

that types of porin other than OmpF and OmpC may be responsible for cyanobacterial

outer membrane nutrient transport (Hoiczyk and Hansel, 2000). This argument is

further supported by the absence of genes encoding for typical porins in proteobacteria

in the genome sequences of M. aeruginosa strains PCC7806 and NIES843.

Although one of the protein homologues involved in the siderophore-mediated system

(TonB) has been identified in cyanobacterial genomes, including those for the strains

of M. aeruginosa used in this work and in Anabaena sp. (Nicolaisen et al., 2008),

recent studies on cyanobacterial Fe uptake consistently suggest that, in contrast to

proteobacteria (Andrews et al., 2003), at least short-term Fe acquisition proceeds in a

manner independent of cellular exudates such as siderophores even under Fe-limitation

(Wirtz et al., 2010, Fujii et al., 2010a). Furthermore, the direct uptake of ferric citrate

by the specific receptor and transporters (FecA-E) used by proteobacteria (Andrews et

al., 2003) appears unlikely for Microcystis due to the dependency of 55

Fe uptake on the

unchelated Fe concentration rather than total ferric citrate concentration (part B of

Figure 4.2).

4.3.3. Fe Species Translocated from the Periplasm to the Cytoplasm

Once extracellular Fe enters the periplasm, specific forms of Fe (e.g., Fe bound to

periplasmic or transporter proteins) are eventually translocated to the cytoplasm. For

some proteobacteria and cyanobacteria such as Synechosystis PCC6803, free or

membrane-anchored periplasmic Fe-binding proteins (FutA) and membrane

transporters (FutB, FutC and FeoB) play significant roles in ferrous and ferric iron

transport into the cytoplasmic space (Velayudhan et al., 2000, Katoh et al., 2001,

Andrews et al., 2003, Waldron et al., 2007). Examination of gene sequences from M.

aeruginosa strains PCC7806 and NIES843 indicates that these strains also have a



62

homologue of the GTP-dependent plasma-membrane Fe(II) transporter (FeoB).

Although no information on the synthesis of FutA proteins was retrieved from a

similar genome analysis, a recent study on M. aeruginosa PCC7806 has suggested the

presence of FutA2 for ferric iron transport that is expressed in the periphery of cells

(Alexova et al., 2011). Since FutA2 is generally fractionated in the extract of the cold

osmotic shock method (Waldron et al., 2007) as water-soluble membrane-free

proteins, Fe bound to this protein is also expected to be found in the periplasmic Fe

fraction of M. aeruginosa. In contrast to FutA2, FutA1 (which is a water-insoluble

membrane-associated protein generally copurified with the photosystem) has been

identified recently as a homologue of an Fe deficiency-induced protein (IdiA) for the

protection of photosynthesis (Tolle et al., 2002).

4.3.4. Model for Translocation of Fe from the External Environment

Given the evidence that (i) small hydrophilic molecules such as unchelated metals are

capable of diffusing into outer membrane porin channels, and (ii) M. aeruginosa

generally possesses periplasmic Fe-binding proteins and active Fe transporters across

the inner membrane, a simple kinetic model to describe Fe transport from the

extracellular medium to the intracellular environment is proposed in Figure 4.4. The

model presented is based on the following assumptions:

(i) Unchelated Fe in the extracellular environment is capable of crossing outer

membrane channels into the periplasmic space in a diffusive manner. This

process is reversible, such that unchelated Fe in the periplasm also diffuses out to

the extracellular environment.

(ii) Unchelated Fe in the periplasm forms complexes (FeXperi) with periplasmic Fe-

binding ligands (Xperi) followed by irreversible translocation of Fe into the

thylakoid membrane or cytoplasm (Fe) by inner membrane transporters, with the

latter process occurring in a first order manner. Although we assign FeXperi as a

single major substrate for the transport of iron to the cytoplasm for simplicity,

FeXperi can be considered to represent any form of periplasmic ferric and ferrous

iron transported into the thylakoid membrane or cytoplasm.

(iii) FeXperi accounts for the majority of periplasmic Fe (Feperi) (i.e., [Feperi] =



63

[FeXperi] + [Fe′peri] ≈ [FeXperi]). Given the average cellular volume of M.

aeruginosa (~270 µm3) (Wiedner et al., 2003) and a periplasmic volume of ~3%

relative to total cell volume, the concentration of total periplasmic Fe is

calculated to be ~10-5

M, which is much higher than [Fe′] (< 10-9

M). As

described below, the steady-state concentration of Fe′ in the periplasm (Fe′peri)

was determined to be approximately equal to the concentration of extracellular

unchelated Fe (i.e., [Fe′peri] ≈ [Fe′]).

Figure 4.4. Kinetic model for Fe transport from the extracellular environment to the

intracellular environment in cyanobacteria. In the extracellular environment,

unchelated Fe (i.e., Fe′) is formed due to the (thermal or reductive) dissociation of

chelated Fe. Unchelated Fe subsequently diffuses through non-specific outer

membrane channels (such as porins). Unchelated Fe in the periplasm is then

complexed by one or more periplasmic Fe-binding ligands (FeXperi) followed by



64

translocation of Fe into the cytoplasm (Fecyto) by inner membrane Fe transporters. A

possible mechanism of Fe(III) and Fe(II) transformation in the periplasm is also

illustrated. Solid arrows represent major reactions considered in the model. Rate

constants depicted near the arrows correspond to those listed in Table 4.1. MCO:

multi-copper oxidase, FeoB: ferrous iron transporter, FutA: ferric iron transporter.

The chemical reactions that describe Fe transport to the intracellular environment can

be described as follows:

' '

periFe FeDiffusion→←

(4.1)

1

1

'

peri peri periFe + X FeX k

k

+

−

→← (4.2)

2

peri cytoFeX Fek→ (4.3)

where k+1, k-1, and k2 represent rate constants for formation and dissociation of the

periplasmic Fe complex and translocation of periplasmic Fe to the cytoplasm,

respectively. According to this model, the time-dependent changes of periplasmic and

cytoplasmic Fe concentrations can be described as follows:

'

peri peri peri

peri '

1 peri peri 1 2 peri

[Fe ] ([Fe ] [FeX ])

[FeX ][Fe ][X ] ( )[FeX ]

d d

dt dt

dk k k

dt+ −

+=

≈ = − +

(4.4)

cyto

2 peri

[Fe ][FeX ]

dk

dt= (4.5)

Making a quasi-steady-state approximation for FeXperi, the rate of Fe accumulation in

the cytoplasm ( Fe'ρ ) can be described, in accord with previous findings (Fujii et al.,

2010a), by Monod-type kinetics in terms of the periplasmic Fe′ concentration as

follows:



65

max '

Fe'(peri) peri

Fe' 2 peri SS '

Fe'(peri) peri

[Fe ][FeX ]

[Fe ]k

K

ρρ = =

+

(4.6)

where [FeXperi]SS indicates steady-state concentration of the periplasmic Fe complex,

max

Fe'(peri)ρ (= k2[Xperi]T where [Xperi]T is the total concentration of the periplasmic Fe-

binding ligand) is the maximum uptake rate and Fe'(peri)K (= (k-1+k2)/k+1) is the half

saturation constant under the conditions of the experiment. Integration of eq. 4.4 yields

the concentration of periplasmic Fe as a function of time as follows (see part A2.3 of

Appendix 2 for detailed derivation):

'

peri peri T '

peri Fe'(peri) peri 1'

Fe'(peri) peri

[Fe ][X ][FeX ] 1 exp ( +[Fe ])

+[Fe ]K k t

K+

= − − ⋅ (4.7)

At steady-state, the translocation rate of periplasmic Fe to the cytoplasm (i.e., Fe'ρ ) is

equivalent to the net rate of diffusional Fe influx that occurs across the outer

membrane (JFe′ mol.cell-1

.s-1

), yielding the following relationship between average

concentrations of Fe′ in the periplasm ([Fe′peri]) and bulk external environment ([Fe′]):

Fe' Fe'J ρ=

max '

Fe'(peri) peri'

dif '

Fe'(peri) peri

' max ' max 2 2 '

dif Fe'(peri) dif Fe'(peri) dif Fe'(peri) dif Fe'(peri) dif Fe'(peri)'

peri

dif

[Fe ][Fe ]

[Fe ]

( [Fe ] ) ( [Fe ] ) 4 [Fe ][Fe ]

2

kK

k K k k K k k K

k

ρ

ρ ρ

− ∆ =+

− − + + − + +=

(4.8)

where kdif is the diffusional constant for unchelated Fe (L.cell-1

.s-1

), and ∆[Fe′] is the

difference between unchelated Fe concentrations in extracellular bulk medium and

periplasmic space (∆[Fe′]=[Fe′peri]-[Fe′]).

By fitting eqs. 4.5-4.7 to the data in Figure 4.1 and part B of Figure 4.2 under the

assumption of [Fe′peri] ≈ [Fe′], parameters for Fe uptake and intracellular transport

were obtained as listed in part B of Table 4.1. Values for [Fe′peri] and [Fe′] were then



66

calculated using eq. 4.8 and the parameters in part B of Table 4.1, suggesting that

[Fe′peri] is comparable to [Fe′] across a range of [Fe′] concentrations from 1 fM to 10

nM ([Fe′peri]/[Fe′] > 0.999). Therefore, the approximation [Fe′peri] ≈ [Fe′] was justified

in this work (see part A2.2 of Appendix 2 for details). Overall, this model (with

relevant parameters) described the time- and concentration-dependent kinetics of Fe

transport reasonably well (Figures 4.1 and 4.2).

An intriguing finding is that the diffusional flux of Fe across the outer membrane of M.

aeruginosa (reported to be 3.9 × 10-9

L.cell-1

.s-1

; see kup in eq. 4.2 shown in reference

(Fujii et al., 2011a) is three orders of magnitude greater than k+1[Xperi]T in this work

(~10-12

L.cell-1

.s-1

). This result suggests that the diffusion of extracellular Fe′ occurs at

much faster rate than complexation by the periplasmic Fe-binding ligand and is a

particularly important factor controlling Fe′ concentration in the periplasm. Another

interesting result is that a higher transport rate of periplasmic Fe by the membrane

transporter (k2) was seen in the non-toxic strain. This finding is consistent with the

previously reported upregulation of Fe transporter proteins for the non-toxic strain

under Fe stress (Alexova et al., 2011).

4.3.5. Fe Redox Speciation in Periplasm

Although we do not consider, for simplicity, the redox state of Fe in the model

presented here, Fe redox reactions inside cells or near the cell surface are

acknowledged to be important in Fe acquisition by phytoplankton (Shaked et al., 2005,

Salmon et al., 2006). Therefore, to further investigate the effect of chemical speciation

on periplasmic and inner-membrane Fe transport, the redox state of Fe present in

periplasm was examined. For the case where Microcystis cells were incubated in the

Fe(III)-dominant system (the control in Figure 4.5), a significant amount of Fe(II)

(33% of total periplasmic Fe) was detected in the periplasmic extract. Periplasmic

Fe(II) was also detected in the Fe(II)-dominant system where ascorbate was added to

the culture (in order to reduce all extracellular and possibly some periplasmic Fe(III) to

Fe(II)). Interestingly, the amount of extracted Fe(II) present in the ascorbate treated

system was comparable to that determined in the control, suggesting that the redox

state of periplasmic Fe was negligibly influenced by the presence of extracellular

reducing agents which might also enter the periplasmic space.



67

Figure 4.5. Effect of ascorbate and TTM on Fe(II) accumulation in the periplasm of

M. aeruginosa PCC7806; (A) oxidation kinetics of Fe(II) in the periplasmic extract,

and (B) percentage of Fe(II) extracted from the periplasm. PCC7806 was incubated in

Fraquil* (0.7 µM Fe and 100 µM citrate) in the presence and absence of chemical

treatments (1 mM ascorbate and 1 mM ascorbate plus 100 µM TTM). The periplasm

was extracted by the cold osmotic shock method in cold Milli-Q water followed by

measurement of Fe(II) in the extract by the luminol chemiluminescence technique. The

amount of periplasmic Fe(II) was calculated by assuming that the observed maximum

value of the chemiluminescence signal corresponds to the amount of Fe(II) in the

periplasm. Error bars represent ±standard deviation from duplicate experiments. A

0

500

1000

1500

2000

2500

3000

3500

0 100 200 300 400

Sig

na

l

Time (s)

Control

Ascorbate

Ascorbate + TTM

A

0

20

40

60

80

100

120

control Ascorbate Ascorbate + TTM

Pro

po

rtio

n o

f p

erip

lasm

ic F

e(I

I)

rela

tive

to

tal p

erip

lasm

ic F

e (

%)

B



68

single-tailed heteroscedastic t-test indicated that the treatments with ascorbate + TTM

were different from the control at a p value of 0.14.

The comparable concentrations of periplasmic Fe(II) in both Fe(III)- and Fe(II)-

dominated systems suggests that extracellular Fe present as either Fe(III) or Fe(II) can

be reduced or oxidized in the periplasm. Thus, it is likely that the redox state of Fe is

tightly regulated in the periplasm and is likely to be substantially different from the

redox state in the external environment. Although the identity of such Fe redox

regulators in the periplasm remains largely unknown, redox reactions of extracellular

Fe with plasma-membrane reductases and multi-copper oxidases (MCO) have been

suggested to be important processes in Fe acquisition for some eukaryotic

phytoplankta (Maldonado et al., 2006). Since MCO (a periplasmic laccase) is also

identified in the genome of M. aeruginosa (PCC 7806 and NIES 843), we examined

the effect of TTM (a compound that inhibits MCO activity) on Fe(II) accumulation in

the periplasm. As shown in Figure 4.5, the presence of TTM and ascorbate in the

culture medium resulted in an increase in accumulation of periplasmic Fe(II), implying

that this enzyme plays a role in controlling the redox state of periplasmic Fe for M.

aeruginosa PCC 7806.

The effect of chemical treatments on 55

Fe accumulation in the whole of cell was also

examined (Figure 4.6). The presence of ascorbate increased the cellular 55

Fe

accumulation by 1.3-fold indicating that ascorbate-mediated reduction of Fe(III) to

Fe(II) increased unchelated Fe concentration and uptake for the Fe:Cit ratio of 0.007

(resulting in pFe′ = 10.3) where Fe(III) uptake was not saturated. The single addition

of TTM resulted in an insignificant change in the Fe accumulation compared to the

control system. In contrast, when ascorbate and TTM were simultaneously present in

the assay culture, the 55

Fe accumulation increased by 3.6-fold, which is in contrast to

the observation by Maldonado et al. (2006) that the presence of TTM in the culture

medium resulted in inhibition of Fe uptake by marine diatoms. The facilitated

periplasmic Fe(II) accumulation and cellular 55

Fe(II) uptake (but not Fe(III) uptake) in

the presence of TTM plus ascorbate for M. aeruginosa suggests that MCO-mediated

oxidation of periplasmic Fe(II) retards the transport of Fe to the cytoplasm. The

increased Fe accumulation in the absence of MCO-mediated Fe(II) oxidation suggests

that Fe(II) in the periplasm is preferably transported to inner compartments by a



69

plasma-membrane transporter (e.g., FeoB) at a faster rate than is the case for Fe(III)

(e.g., through FutA) (Figure 4.4).

Figure 4.6. Effect of chemical treatments on cellular 55

Fe accumulation for M.

aeruginosa PCC7806. In the control, cells were incubated for 3 hr in Fraquil*

containing 55

Fe-citrate (total concentrations for Fe and citrate were 0.7 µM and 100

µM, respectively). In the chemical treatments, cells were incubated in the additional

presence of 100 µM TTM, 1 mM ascorbate and 1 mM ascorbate plus 100 µM TTM.

Error bars represent ±standard deviation from duplicate experiments. One asterisk

indicates that chemical treatments were significantly different from the control at a p

value less than 0.05 using a single-tailed heteroscedastic t-test.

A negligible effect of TTM on 55

Fe accumulation in the Fe(III)-dominant system was

observed when an exogenetic reductant was absent (Figure 4.6). This lack of effect is

surprising as it was expected that the presence of TTM (which inhibits oxidation of

Fe(II) through shut-down of the MCO) would lead to a relative increase in the rate of

Fe(III) reduction. The non-observable effect of TTM in the Fe(III)-dominant system

could result from secondary effects of TTM such as inhibition of the activity of other

enzymes responsible for the reduction of periplasmic Fe(III) to Fe(II) though the

molecules responsible and detailed mechanism involved remain unclear. In addition to

oxidases and oxidoreductases such as MCO and flavin enzymes respectively,

secondary produced reactive oxygen species (ROS) are also candidates for the redox

55F

e u

pta

ke

ra

te

(re

lative

to

co

ntr

ol)

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

5

Control TTM Ascorbate Ascorbate +TTM

*

*

Cellu

lar

55F

e a

ccum

ula

tion

(rela

tive to

contr

ol)



70

regulation of periplasmic Fe, given that Fe redox cycling is potentially relevant to

generation of free radical intermediates. Superoxide, for example, can be formed in the

periplasm of Escherichia coli possibly as a result of the adventitious autooxidation of

menaquinone in the cytoplasmic membrane (Korshunov and Imlay, 2006), eventually

resulting in diffusion through the outer membrane to the extracellular environment. M.

aeruginosa also produces superoxide though the mechanism of production is unknown

(Fujii et al., 2010a). We believe that considerable scope exists for further investigation

of such processes with regard to their role in cyanobacterial Fe uptake.

4.4. CONCLUSIONS

In the present work, a previously developed model for Fe uptake by phytoplankton has

been extended to accommodate the processes involved in Fe transport from the

extracellular environment to the cytoplasm. The model incorporates the conclusion

from the current study that only Fe′ crosses the outer membrane by diffusion into the

periplasm with subsequent conversion into a form suitable for transport into the

cytoplasm. This is in contrast to the existing model in which direct internalization of

extracellular Fe′ by plasma-membrane transporters is assumed. Our model provides

consistency not only with previous kinetic studies indicating that the rate of

cyanobacterial Fe uptake follows Monod-type kinetics with respect to the

concentration of Fe′ but also with studies identifying the nature of intracellular

molecules responsible for Fe transport (Andrews et al., 2003). However, further

studies are clearly required to properly account for the effect of periplasmic redox

processes on intracellular Fe transport.

71

CHAPTER 5

IRON UPTAKE KINETICS BY THE

FRESHWATER CYANOBACTERIUM

MICROCYSTIS AERUGINOSA IN THE

PRESENCE OF SUWANNEE RIVER

FULVIC ACID

Chapter 5. Iron Uptake Kinetics by the Freshwater Cyanobacterium Microcystis aeruginosa in the

Presence of Suwannee River Fulvic Acid

72

5.1. INTRODUCTION

Iron (Fe) is one of the micronutrients essential for growth of almost all organisms. Of

the microorganisms, cyanobacteria have a particularly high requirement for Fe due to

its critical roles in metabolisms including photosynthesis, respiration, nitrogen fixation

and regulation of reactive oxygen species (Crichton, 2009). Because of the low

solubility of inorganic Fe(III) in circumneutral pH waters (e.g., 10-11

M at pH 7.5-9

(Liu and Millero, 1999)), a majority of dissolved Fe(III) in natural waters is present as

complexes with natural organic matter (NOM) including siderophores (Vraspir and

Butler, 2009), humic substances (Liu and Millero, 2002, Tipping, 2002) and possibly

polysaccharides (Hassler et al., 2011). In air-saturated surface waters at circumneutral

pH, ferrous iron (Fe[II]) is rapidly oxidized to ferric iron (Fe[III]) by dissolved oxygen

and secondarily produced organic and inorganic radicals including reactive oxygen

species with a half-life time of several minutes (Millero and Sotolongo, 1989, Rose

and Waite, 2002, Pham and Waite, 2008a). As such, Fe(III) is recognized to be the

thermodynamically favoured redox state in surface waters. As a result of extensive

investigations of Fe redox chemistry over the last decade, however, it is now evident

that Fe(II), an important substrate for Fe uptake by phytoplankton, can be generated at

appreciable rates in euphotic waters via biological, photochemical and thermal

reduction of Fe(III) species in many cases mediated by cellular membrane reductase

(Maldonado and Price, 2001, Shaked et al., 2005), superoxide (Rose and Waite, 2005,

Fan, 2008, Rose, 2012), ligand-to-metal charge transfer (LMCT) (Faust and Zepp,

1993, Waite et al., 1995) and humic substances (Pullin and Cabaniss, 2003).

The Fe uptake machinery in both freshwater and marine phytoplankton has been

investigated by many researchers in recent decades. One of the important consensuses

from the previous studies is that organically-complexed Fe, including the complexes

with metal buffering ligands used for algal culturing medium, are too large or

hydrophilic to directly permeate plasma-membrane (lipid-bilayer) of eukaryotic and

prokaryotic phytoplankta. Thus, Fe availability for uptake by phytoplankton is

described as a function of unchelated Fe concentration rather than total or chelated Fe

(Maldonado and Price, 2001, Shaked et al., 2005, Fujii et al., 2011a). Although Fe

bound to some specific molecules such as siderophores and citrate may be recognized



73

by bacterial outer-membrane receptor followed by active transport to intracellular

compartments via energy-dependent processes, recent studies have indicated that, even

under Fe-limited conditions, the siderophore-independent uptake (i.e., uptake of

unchelated Fe) is dominant for freshwater cyanobacteria such as Microcystis (Fujii et

al., 2010a) and Anabaena sp. (Wirtz et al., 2010). The lack of siderophore-associated

genes in marine phytoplankton consistently suggests that Fe uptake independent of

siderophore may be prevalent among the prokaryotic phytoplankton (Hopkinson and

Barbeau, 2012). There is also evidence that the reduction of organically complexed

Fe(III) to Fe(II) via photochemical and biological processes is a critical step in

increasing Fe bioavailability under Fe-limited environments (Maldonado et al., 2005,

Fujii et al., 2011a, Kranzler et al., 2011).

Previous studies have undoubtedly provided significant insights toward understanding

the mode of Fe uptake by phytoplankton in natural and culturing systems. However,

the underlying experimental and theoretical findings were basically provided by

incubational assays using model Fe-binding ligands such as ethylenediaminetetraacetic

acid (EDTA), desferrioxamine B (DFB) and citric acid (Maldonado and Price, 2001,

Rose et al., 2005, Shaked et al., 2005, Garg et al., 2007, Fujii et al., 2010a, Fujii et al.,

2011a), as transformation kinetics for extracellular Fe bound to these ligands are well

defined. Therefore, one of the important issues remaining to be addressed is whether

the current consensus on the Fe uptake kinetics based on the studies using model

ligands is indeed consistent with the mode of Fe uptake occurring in natural waters

where Fe is generally buffered by structurally and chemically heterogeneous NOM.

Transformation kinetics of Fe bound to NOM have been extensively studied over the

last decade, particularly for one of the most commonly used standard humic

substances: Suwannee River fulvic acid (SRFA). Using Fe complexed with this

heterogeneous molecule, chemical reactions potentially occurring in natural surface

waters including formation and dissociation of FeSRFA complexes and redox

reactions involving Fe and SRFA have been carefully examined. Consequently, a set

of published rate constants is currently available for predicting the transformation of

the FeSRFA complex under particular solution conditions (e.g., at pH~8, Table 5.1)

with these rate constants potentially useful in elucidating and/or describing the key

processes involved in Fe uptake by phytoplankton.



74

In this work, we have investigated the mechanism of Fe uptake mechanism by the

freshwater cyanobacterium Microcystis aeruginosa in the presence of SRFA with

particular attention given to examining the effect of non-photochemical and

photochemical transformations of Fe on Fe uptake. Although there are only a few

reports on the growth limitation of freshwater phytoplankton due to low Fe availability

(Nagai et al., 2006) compared to marine systems, it is recognized that the Fe nutritional

status influences the synthesis of primary and secondary metabolites including

cyanotoxins even under relatively higher Fe availability where optimal growth rates

are observed (Voelker et al., 2010, Alexova et al., 2011). Therefore, insight into the

mode of Fe uptake is considered to be of great importance in understanding the

environmental and nutritional factors influencing growth and cellular metabolism of

this organism. Proper understanding of ecological function and adaptation of

cyanobacteria is considered to be particularly important for the occurrence and

management of toxic phytoplankton blooms in water reservoirs and bodies used for

drinking water supplies.


5.2.1. Reagents

Detailed information on the grade of chemicals used, the mode of storage of stock

solutions, pH measurement and approach to cleaning of containers has been described

previously (Section 2.1, Chapter 2).

Stock solutions of Suwannee River fulvic acid (SRFA), ethylenediaminetetraacetic

acid (EDTA) and citric acid were prepared at concentrations of 1-100 g L-1

, 1-100 mM

and 1-100 mM by dissolving SRFA (International Humic Substance Society),

Na2EDTA (Sigma) and Na3citrate (Sigma) to MQ, respectively. A 0.1 M ferrozine

(FZ; 3-(2-pyridyl)-5,6-diphenyl-1,2,4-triazine-4’, 4’’-disulfonic acid sodium salt,

Sigma) solution was prepared in MQ. pH of these ligand stocks were all adjusted to

8.0 to avoid a significant pH change when added to the culture medium (Fraquil* pH

8).



75

Stock solutions of non-radiolabeled Fe(III) (FeIII

Cl3, Ajax Finechem, Australia) were

made in 0.01 M HCl at concentrations of 1-10 mM. For the Fe uptake experiments, a

23 mM stock of radiolabeled ferric chloride (55

FeIII

Cl3 in 0.5 M HCl, 185 MBq,

Perkin-Elmer, Australia) was diluted with the 1 mM non-radiolabeled FeCl3 by 24-fold

to produce a final Fe concentration of 1.9 mM. A 2 mM bicarbonate buffered solution

at pH 8 was made by dissolving sodium hydrogen carbonate (NaHCO3, Sigma) in MQ.

Solutions of organically complexed Fe(III) (FeIII

L) were then made by placing the 1.9

mM Fe(III) solution in the bottom of a 1.5 mL polypropylene container followed by

addition of an appropriate volume of SRFA, EDTA and citrate stocks. After the

bicarbonate buffer was pipetted into the mixture, the solution pH was adjusted to 8.

Before use, the solution was stored for 24 hr under dark conditions at 25oC to reach

equilibrium.

A stock solution of 6000 unit mL-1

superoxide dismutase (SOD) was prepared by

dissolution of 2519 unit mg-1

SOD (Sigma) in MQ and frozen in 100 µL aliquots at -

80oC when not in use. Denatured SOD (d-SOD) was also prepared by heating SOD

solution in boiling water for 10 min. As Fe reducing reagents, stock solutions of 100

mM ascorbate (sodium L-ascorbate, Sigma) and 100 mM hydroxylamine

hydrochloride (Sigma) were prepared in MQ followed by pH adjustment to 8.0. In the

experiments where superoxide was artificially generated, a stock solution of 12.5 mM

xanthine (Sigma) was prepared in 0.01 M NaOH solution and the pH of the solution

adjusted to 9.8 with HCl. A 1 kU.L-1

stock solution of xanthine oxidase (XO, Sigma)

was prepared in MQ water and 1 mL aliquots were individually frozen at -86 oC until

use. To remove metals non-specifically adsorbed to cell surfaces, cells were washed by

using a chelate solution containing 50 mM Na2EDTA (Sigma) and 100 mM Na2oxalate

(Sigma) (hereafter referred to as “EDTA/oxalate”) (Tang and Morel, 2006).

5.2.2. Culturing Media

The detailed preparation procedure and nutrient composition of a modified Fraquil

medium (Fraquil*) were described previously (Section 2.2.1, Chapter 2). Briefly, for

long-term incubation, Fraquil* buffered by EDTA was prepared using at least reagent

grade salts inside a trace clean room supplied with HEPA-filtered air to provide final

concentrations of 0.1 µM for Fe and 26 µM for EDTA and pH 8. For short-term 55

Fe



76

uptake studies, Fe- and ligand-free Fraquil* was prepared with the procedure identical

to those described for the long-term culturing medium, except that addition of Fe and

ligand was omitted and final pH of the medium was adjusted to 6-9.

5.2.3. Long-term Culturing Conditions

Long-term culturing conditions are described in detail in Section 2.2.2, Chapter 2.

Briefly, a batch culture of the unicellular cyanobacterium Microcystis aeruginosa

PCC7806 was incubated under sterile conditions in a temperature- and light-controlled

incubator (Thermoline Scientific) at 27oC. Cells were regularly subcultured into fresh

media when cultures reached stationary growth phase. The subculturing was

performed approximately two-weeks after the commencement of incubation with

cellular concentration of ~104 cell mL

-1. Cell numbers in the cultures were counted on

a Neubauer hemocytometer (0.1 mm depth) under an optical microscope (Nikon,

Japan). A specific growth rate in the long-term incubation was determined to be ~0.7

d-1

.

5.2.4. Light Condition

For all incubation and experiments undertaken in this work, light was vertically

supplied by three cool-white fluorescent tubes (36W, 28mm diameter, 1.2 m length,

Philips) on a 14:10 light:dark cycle. Cell culture and other abiotic samples were

consistently incubated at a distance of 10 cm from the fluorescent tubes. At this

distance, total radiation intensity was determined to be 157 µmol-quanta.m-2

.s-1

(see

part A of Figure 3.1, Chapter 3 for the emission spectrum) using an Ocean Optics USB

4000 spectrophotometer equipped with an optical fiber and cosine converter (CC-3-

UV) that was calibrated against a DH-2000 VIS light source. All incubations and

experiments were performed either in 1 cm polystyrene spectrophotometer cuvettes

(Starna Pty Ltd, Australia) or polycarbonate vessels (Nalgene) that minimally interfere

with visible light transmission between 400-800 nm (Figure 3.2, Chapter 3). For the

dark incubations, the vessels were covered with aluminium foil to prevent any light

penetration into the solution.



77


Fe Uptake Experiments

Prior to short-term Fe uptake assays, Microcystis cells in the long-term culture were

harvested onto a 25 mm diameter, 0.65 µm PVDF membrane (Millipore) during the

daytime of late-exponential growth phase (at densities ~3 × 106

cell mL-1

). The filtered

cells were washed gently with a chelate solution (50 mM Na2EDTA and 100 mM

Na2oxalate at pH 7, hereafter referred to as “EDTA/oxalate”) and subsequently rinsed

with 2 mM NaHCO3 in order to remove metals non-specifically adsorbed onto cell

surfaces. The washing treatment was achieved by passing ~30 mL of the solutions

through the filter for ~10 minutes. The washed cells were re-suspended into the Fe-

and ligand-free Fraquil* medium to provide cell densities of ~2 × 10

6 cell mL

-1. In

experiments where the effect of chemical treatment on 55

Fe uptake were examined, the

Fe- and ligand-free culture was prepared in the additional presence of either 1 mM FZ,

60 kU.L-1

superoxide dismutase (SOD), 1 mM ascorbate, 1 mM hydroxylamine

hydrochloride, 100 µM xanthine plus 1 kU.L-1

xanthine oxidase (XO) or 1 mM FZ

plus the reducing agent (ascorbate or hydroxylamine hydrochloride) by appropriately

supplementing chemical stock solutions described in Section 5.1.

A solution of Fe(III) complexed by SRFA was prepared by mixing radiolabeled ferric

chloride (55

FeCl3) stock with SRFA stock (International Humic Substance Society).

After dilution of a 23 mM radiolabeled ferric chloride stock (55

FeCl3 in 0.5 M HCl,

185 MBq, Perkin-Elmer, Australia) with 1 mM non-radiolabeled FeCl3 in 0.01 M HCl

at total Fe concentration of 1.9 mM, the 55

Fe solution was mixed with an appropriate

volume of 10 g.L-1

SRFA in the bottom of a polypropylene microtube. A 2 mM

NaHCO3 was then added to the mixture to maintain circumneutral pH followed by

acid-base titration to adjust the pH of the mixture to 6-9 depending on the purpose of

the experiments. The 55

FeIII

SRFA solution was equilibrated for 24 h under dark

conditions and ambient temperature before use.

The 55

Fe uptake assay was initiated by adding pre-equilibrated 55

FeIII

SRFA at final

concentrations of 0.2 µM in 55

Fe and 1-100 mg.L-1

SRFA to cell suspensions. In all

cases, except for the experiments where time course of 55

Fe uptake was examined,

cells were incubated at 27oC for 2 h (based on the linearity of

55Fe uptake) in the

absence and presence of light. For comparison, 55

Fe uptake experiments were also



78

undertaken in the identical manner described above except that EDTA or citrate was

used as Fe-binding ligand instead of SRFA.

After the incubation, cells were vacuum-filtered onto 0.65 µm PVDF membrane filters,

then rinsed three times with 1 mL EDTA/oxalate solution and twice with 1 mL of 2

mM NaHCO3 (total rinsing time was ~10 min). The filtered cells were then placed in

glass scintillation vial with 5 mL of scintillation cocktail (Beckman ReadyScint). The

activity was measured in a Packard TriCarb Liquid Scintillation Counter, with

scintillation counts (counts per minute) of the samples converted to moles of Fe using

concurrent counts of 5-50 µL of 55

Fe-ligand stock in 5 mL scintillation cocktail.

Process blanks were determined by performing the procedure in the absence of cells.

5.2.6. Kinetic Model for Fe Transformation and Uptake

The kinetic model used for the calculation of Fe chemical species and uptake by M.

aeruginosa is shown in Table 5.1 and illustrated in Figure 5.1.

Figure 5.1. Kinetic model for Fe chemical speciation and uptake by M. aeruginosa.

FeIIISRFA

Fe(III)'

kup

FeIIFZ3

KFe'

FeIISRFA

Fe(II)'

Outer membranePorin

Fe'

Fe(III)

fk

Fe(III)

d1k

Fe(II)

fk

Fe(III)

d2k

and Fe(II)

dk

Fe(II)

ox1k Fe(II)

ox4kto

Fe(III)

redk

IIFe L

ox1k to

IIFe L

ox4k

IIIFe L

redk ,

IIIFe L

red-darkk and

IIIFe L

red-lightk



79

In this work, we assume that only unchelated Fe (Fe') is available for Fe uptake.

Steady-state concentration of unchelated Fe ([Fe']SS) was calculated over a range of

conditions appropriate to the 55

Fe uptake assay experiments by summing steady-state

concentrations of Fe(III)' and Fe(II)' as described in Section 5.3. Monod-type

saturation theory was then used to describe the rate of Fe uptake (ρs, amol.cell-1

.hr-1

) as

follows:

[S]

[S]

S

max

SS

+=

K

ρρ (5.1)

where [S] indicates the steady-state concentration of the biologically available portion

of Fe in the extracellular environment (i.e., [Fe'] = [Fe(III)'] + [Fe(II)']). KS and max

Sρ

represent the half saturation constant and the maximum uptake rate under the

conditions examined, respectively. In order to reduce unknown complexity in

competitive uptake between Fe(II) and Fe(III), cellular affinity and uptake capacity for

the two redox states of Fe were assumed to be equal. This assumption is consistent

with the outer-membrane structure of cyanobacteria, where only small-size nutrients

are capable of passing through the outer-membrane transport channel by a

concentration-dependent diffusive process (Fujii et al., 2011a). Although the short-

term Fe uptake can be regulated by the cellular nutritional status during the

preconditioning stage, the uptake parameters for M. aeruginosa PCC7806 acclimated

under the conditions identical to those employed in this work have been previously

published (Fujii et al., 2010a, Fujii et al., 2011a) (Table 5.1).


5.3.1. 55

Fe Uptake as a Function of SRFA Concentration

55Fe uptake in the absence and presence of irradiation by fluorescent tubes (hereafter

referred to as dark and light uptake) indicate that the amount of 55

Fe accumulated in

cells increases with time over several hours (Figure A3.1 of Appendix 3). Linear

regression analysis was applied to the data collected within 2 h, yielding relatively

good linearity with correlation coefficients (R2) of 0.87 for dark uptake and 0.94 for



80

light uptake. The non-linear accumulation after a few hours’ incubation may suggest

that either (i) the concentration of Fe available for uptake varied in a time-dependent

manner due to the heterogeneity of Fe-binding strength or redox properties of SRFA,

(ii) cellular uptake became saturated during this period, and/or (iii) cellular growth or

activity varied significantly. Due to uncertainties regarding the processes responsible

for the non-linear uptake, further 55

Fe uptake assays were undertaken with only 2 hr

incubations. The results of the 2 hr 55

Fe uptake assays for a range of SRFA

concentrations (while total Fe concentration was kept constant) indicated that both

dark and light 55

Fe uptake rates decrease with increasing SRFA concentration (part A

of Figure 5.2).

Figure 5.2. 55

Fe uptake as a function of (A) SRFA and (B) model ligand

concentrations in the absence (black symbols and bars) and presence (white symbols

0

0.5

1

1.5

2

2.5

3

3.5

1 5 25

C*

****

****

0

0.5

1

1.5

2

2.5

3

3.5

1 5 25

D

**

**

SRFA concentration (mg.L-1) SRFA concentration (mg.L-1)

SRFA concentration (mg.L-1)

55F

e u

pta

ke

ra

te (

am

ol.

cell

-1.h

r-1)

0

0.2

0.4

0.6

0.8

1

1.2

1.4

5 mg/L 25 mg/L 26 μM 100 μM 26 μM 100 μM

SRFA Citrate EDTA

B

Ligand concentrations

0

0.5

1

1.5

2

2.5

3

3.5

0 20 40 60 80 100 120

A

0

0.2

0.4

0.6

0.8

1

1.2

Control FZ Asc Asc+FZ HH HH+FZ X/XO X/XO+FZ

No

t m

ea

sure

d

No

t m

ea

sure

d

E**

** ****

*

**

*

**



81

and bars) of light. The 55

Fe uptake assay was performed at concentrations of 200 nM

total Fe, 1-100 mg L-1

SRFA and 26-100 µM citrate and EDTA. Solid and dotted lines

indicate model fits to the data from the Bligh and Rose model, respectively. Effect of

(C) ferrozine (FZ) and (D) superoxide dismutase (SOD) on 55

Fe uptake. In control

treatments, Fe uptake assays were undertaken under dark (black bar) and light (white

bars) at concentrations of 200 nM for Fe and 1-25 mg L-1

for SRFA. In the chemical

treatments, the identical 55

Fe uptake assay was performed except for the additional

presence of either FZ or SOD under dark (gray bar) and light (shaded bar). (E) Effect

of reducing agents on 55

Fe uptake. The control treatments were undertaken under dark

(black bar) and light (white bars) (200 nM for Fe and 5 mg L-1

for SRFA). Chemical

treatments were performed, in addition, in the presence of FZ, ascorbate (Asc),

hydroxylamine hydrochloride (HH), xanthine/xanthine oxidase (X/XO) or their

combination. All short-term Fe uptake assays were performed in Fraquil* for 2 h at cell

density of ~2 × 106 cell mL

-1. Symbols and error bars represent averaged value and

±standard deviation from triplicate experiments. In panels C-E, asterisks indicate that

55Fe uptake rate in the presence of a particular chemical treatment is significantly

different from control at the levels of p < 0.01 for **

and p < 0.05 for * using a single-

tailed heteroscedastic t-test.

It is an intriguing result that the light 55

Fe uptake were generally similar to the dark

uptake particularly in the higher [SRFA] cases (e.g., uptake increased to only 1.1-2.2

folds under the light at 5 mg L-1

SRFA or greater) where 55

Fe uptake rates were

substantially lower than the maximum uptake rate. This result is in contrast to those

observed by using model ligands including EDTA and citrate where 55

Fe uptake in

these systems increased significantly on illumination (e.g., to 40-51-fold in EDTA

system and 1.8-6.2-fold in citrate system, part B of Figure 5.2) under similar or slightly

lower Fe(III)' availability. Since all cells used in this work were preconditioned in an

identical manner, the different effects of the light on 55

Fe uptake between the homo-

and heterogeneous Fe-binding ligands are expected to be associated with abiotic

processes rather than cellular activities (e.g., Fe uptake affinities). In addition, previous

finding indicated that biological factors that may be activated during light irradiation is

unlikely to facilitate the short-term 55

Fe uptake in the Fraquil* system (Fujii et al.,

2011a). Under identical preconditioning conditions, the light 55

Fe uptake is most likely



82

determined by abiotic factors (i.e., Fe transformation due to photochemical reduction)

rather than biological factors.

5.3.2. Effect of Chemical Treatment on 55

Fe Uptake

To examine the contribution of Fe redox transformations on net 55

Fe uptake, the 55

Fe

uptake assay was undertaken in the presence of the strong Fe(II) chelator, FZ (Rose et

al., 2005, Garg et al., 2007, Fujii et al., 2010a, Fujii et al., 2011a) (part C of Figure

5.2). The presence of this membrane non-permeable chemical resulted in the reduction

of rate of Fe uptake by 22-63% and 39-79% under dark and light conditions (at pH 8),

respectively, depending on SRFA concentration. Although there is a concern that use

of high concentrations of FZ (≥ 400 µM) might actively reduce Fe(III)' and result in

artificial inhibition of Fe uptake by simply lowering [Fe(III)'] (Shaked et al., 2004), use

of FZ at a concentration of 1 mM in previous work resulted in a negligible inhibition

of Fe(III) uptake in citrate-buffered Fraquil* (where Fe(III)' uptake dominates) (Fujii et

al., 2010a). Therefore, the reduced 55

Fe uptake in the FZ-treated system suggests that

the reduction of Fe(III) to Fe(II) is a significant process in Fe uptake under both dark

and light conditions.

It is now acknowledged that the photochemical reduction of chelated Fe(III) is

mediated by several mechanisms including ligand-to-metal charge transfer (LMCT),

superoxide-mediated Fe(III) reduction (SMIR) and Fe(III) reduction by photo-

generated organic radicals (e.g., semiquinone-type radicals). Recent studies have

proposed that while LMCT is important at acidic pHs, SMIR becomes a more

significant pathway in the photo-reduction of FeIII

SRFA at higher pH. Therefore, the

55Fe uptake assay was also undertaken at pH 8 in the presence of SOD to examine the

effect of superoxide on reductive Fe uptake in our system (part D of Figure 5.2).

Indeed, addition of SOD resulted in significant inhibition of the light-mediated 55

Fe

uptake (49-72%), while small effects were seen for the dark 55

Fe uptake (-3-4%). The

result that 55

Fe uptake decreased only in the light system is consistent with previous

findings that SMIR is a major pathway of FeIII

SRFA reduction in sunlit natural waters

at circumneutral pH (Garg et al., 2012). Under dark conditions, other mechanisms are

responsible for Fe(III) reduction as discussed below.



83

To examine if reducing entities other than light-mediated reduction influence

unsaturated 55

Fe uptake, the 55

Fe uptake assay was performed in the presence of

exegetic Fe-reducing reagents ascorbate and hydroxylamine hydrochloride (part E of

Figure 5.2). Similar to the light system, the exegetic Fe reductants, if solely present,

had negligible impact on both dark and light 55

Fe uptake. However, when FZ was

added to these systems, 55

Fe uptake was strongly inhibited. In the experiment where

superoxide was artificially generated by oxidation of xanthine with XO, 55

Fe uptake

was also substantially reduced when FZ was present. These results suggest that sole

reduction of Fe by reducing entities does not increase unsaturated 55

Fe uptake, possibly

as a result of the rapid reoxidation of FeIISRFA prior to the complex dissociation (as

discussed below), whilst sequestration of reduced Fe by FZ substantially decreases

55Fe uptake.

Table 5.1. Kinetic model and rate constants used in this study.

No. Reaction Rate constant Reference

1. Rate constants for organically complexed Fe

1 FeIII

L → Fe(II)' + L IIIFe L

red-darkk 1.3 × 10-6

s-1

This study

2 FeIII

L + hv → Fe(II)' + L

IIIFe L

red-lightk 1.6 × 10-5

s-1

This study

3 FeIII

L + O2- →Fe

IIL + O2

IIIFe L

redk 2.8 × 105 M

-1 s

-1 Rose and Waite (2005)

4 FeIIL + O2 →Fe

IIIL + O2

•-

IIFe L

ox1k 1.5 × 102 M

-1 s

-1 Miller et al. (2009)

5 FeIIL +

1O2 →Fe

IIIL + O2

•-

IIFe L

ox2k ~1010

M-1

s-1

Garg et al. (2012)

6 FeIIL + O2

- →Fe

IIIL + H2O2

IIFe L

ox3k 1.2 × 106 M

-1 s

-1 Fujii et al. (2010b)

7 FeIIL + H2O2 →Fe

IIIL + OH

-

IIFe L

ox4k 3.2 × 105 M

-1 s

-1 Miller et al. (2009)

8 Fe(III)' + L →FeIII

L Fe(III)

fk 8.7 × 10

5 –

7.9 × 106

M-1

s-1

Rose and Waite

(2003b), Jones et al.

(2009), Bligh and

Waite (2010)



84

9 FeIII

L1→ Fe(III)' + L1 Fe(III)

d1k 2.0 × 10

-6 –

3.2 × 10-4

s

-1

Rose and Waite


(2009), Bligh and

Waite (2010)

10 FeIII

L2→ Fe(III)' + L2 Fe(III)

d2k 3.2 × 10

-4 –

3.8 × 10-3

s

-1

Rose and Waite


(2009)

11 Fe(II)' + L →FeIIL

Fe(II)

fk 2.5 × 10

4 -

4.5 × 104

M-1

s-1

Rose and Waite

(2003b), Bligh and

Waite (2010)

12 FeIIL→ Fe(II)' + L

Fe(II)

dk 7.9 × 10

-4 -

4.5 × 10-1

s

-1

Rose and Waite

(2003b), Bligh and

Waite (2010)

13 % Fe bound to strong ligand

class R 61 - 100 %

Rose and Waite


(2009)

14 Fe binding capacity of SRFA CFe 260 µmol.g-1

Rose and Waite

(2003b)

2. Rate constants for unchelated Fe

15 Fe(III)' + O2- →Fe(II)' + O2

Fe(III)

redk 1.5 × 108 M

-1 s

-1

Rush and Bielski

(1985)

16 Fe(II)' + O2 →Fe(III)' + O2•-

Fe(II)

ox1k 8.8 M-1

s-1

Pham and Waite

(2008a)

17 Fe(II)' + O2- →Fe(III)' + H2O2

Fe(II)

ox2k 1.0 × 107 M

-1 s

-1

Rush and Bielski

(1985)

18 Fe(II)' + H2O2 →Fe(III)' + OH•

Fe(II)

ox3k 5.0 × 104 M

-1 s

-1

Millero and Sotolongo

(1989)

19 Fe(II)' + OH• → Fe(III)' + OH

-

Fe(II)

ox4k 5.0 × 108 M

-1 s

-1 Zuo and Hoigne (1992)

3. Rate constants for superoxide

20 O2- +

O2

-→ H2O2 + O2 kdisp 5.0 × 10

4 M

-1 s

-1 Bielski et al. (1985)



85

a A is the redox-active organic moiety which catalytically disproportionates superoxide

in the dark, but whose catalytic activity is inhibited during irradiation due to reaction

of the reduced form of A with singlet oxygen. Total concentration for A ([AT]) is 20

µmol (g-SRFA)-1

.

b Uptake parameters were determined for M. aeruginosa PCC7806 grown in Fraquil

*

under the conditions identical to those employed in this work (e.g., exponential growth

phase, light condition and nutrient compositions including Fe availability).

5.3.3. Mode of Dark Fe Uptake

To investigate the mechanism of dark Fe uptake in further detail, we examined the

chemical speciation of Fe at the Fe:SRFA ratios used in this work. The significant

cellular 55

Fe accumulation even in the presence of FZ (part C of Figure 5.2) suggests

that a certain amount of Fe was taken from the Fe(III) pool most likely via the thermal

dissociation of FeIII

SRFA. To quantitatively describe the processes involved in dark

Fe(III) uptake, the steady-state concentration of Fe(III)' present in the culture medium

([Fe(III)']SS) was calculated from the balance of FeIII

SRFA formation and dissociation

processes. The calculation of [Fe(III)']SS requires knowledge of the various Fe-binding

sites present in fulvic acid with different affinities for Fe that are classically

characterized by continuous or discrete ligand models (Tipping, 2002). While

recognised to be a simplification, this trait of fulvic acid was considered by assigning

21 Microcystis cells → O2- kprod 1.2 × 10

-18

mol cell-1

hr-1

Fujii et al. (2010a)

22 O2- +

A → A

- + O2, kSRFA

(3.3 × 10-2

)

/[AT] a

M-1

s-1

Garg et al. (2011)

4. Parameters for Fe uptake b

23

Fe' → uptake:

max

Fe'Fe'

Fe

[Fe ]

[Fe ]K

ρρ

′

′=

+ ′

KFe' 3.3 × 10-11

M Fujii et al. (2010a)

max

Fe'ρ 3.3 × 10-18

mol cell

-1

hr-1

Fujii et al. (2010a)



86

different rate constants for two representative ligand classes: i.e., strong and weak

ligand classes (L1 and L2, Table 5.1). The dissociation reactions can be described as:

( )

Fe(III)d1III

1 1Fe L Fe III ' Lk

→ + (5.2)

( )

Fe(III)d2III

2 2Fe L Fe III ' Lk

→ + (5.3)

Formation of a metal–ligand complex is generally controlled by the rate of water-loss

from the metal center of the outer-sphere complex (Margerum et al., 1978). Formation

rates of Fe complexes with each ligand class have previously been assumed to be

identical with this assumption successfully describing the kinetic data of FeIII

SRFA

complexation (Rose and Waite, 2003b). Thus, in this work, the formation reaction is

described as;

( )

Fe(III)f IIIFe III ' L Fe L

k+ → (5.4)

Due to the predominance of FeIII

SRFA relative to other Fe species, FeIII

SRFA

concentration can be approximated to be equal to the initial Fe total concentration in

the system (i.e., [FeT]≈[FeIII

L1]+[FeIII

L2]). Thus, [Fe(III)']SS was calculated using the

following equation:

Fe(III) III Fe(III) III

d1 1 d2 2SS Fe(III)

f

Fe(III) Fe(III)

d1 T d2 T

Fe(III)

f T T

[Fe L ] [Fe L ] [Fe(III)'] =

[L]

[Fe ]×R [Fe ]×(1-R)

([L ]-[Fe ])

k k

k

k k

k

+

+=

(5.5)

where R represents the proportion of Fe(III) bound to strong-binding sites. [L] and [LT]

indicate the concentrations of free Fe-binding ligand and total ligand, respectively.

[LT] is calculated from the product of SRFA concentration (mg.L-1

) and Fe binding

capacity of SRFA (CFe µmol.g-1

).

For calculation, we considered three combinations of R and rate constants for

dissociation and complexation of FeIII

SRFA published so far at pH~8 (reactions 8, 9,



87

10, 13 and 14 in Table 5.1, hereafter referred to as Rose, Bligh and Jones Fe(III)

models) (Rose and Waite, 2003b, Bligh and Waite, 2010, Jones et al., 2009). The

reported rate constants vary by up to three orders of magnitude due presumably to

different contributions of other competing reactions depending on the methods and

media employed. The competing reactions may include Fe precipitation and adjunctive

processes associated with competitive ligands that will occur at similar time-scales to

FeIII

SRFA complexation and dissociation but which are not fully considered in the

previous kinetic models, resulting in the different estimates of rate constants and Fe-

binding capacity. Comparison of calculated Fe(III) uptake (by substituting [Fe(III)']SS

into eq. 5.1) to the measured 55

Fe uptake in the presence of FZ (where a majority of

55Fe is most likely assimilated from Fe(III) pool) indicated that the measured uptake

rates fall within the range of predicted uptake rate using the three different models with

the Rose and Bligh models providing reasonable agreements (Figure A3.2 of Appendix

3). Although caution should be exercised in use of the published parameters,

particularly where unaccounted differences exist, the reasonable fit of the two models

is generally supportive of the notion that concentration of Fe(III) available for uptake

is determined solely by complexation and dissociation of FeIII

SRFA, which is

consistent with previous work using model ligands EDTA and citrate (Fujii et al.,

2010a). In this context, the decreased 55

Fe uptake with increasing concentration of

SRFA (part A of Figure 5.2) can be explained by the lower availability of unchelated

Fe at higher SRFA concentration.

With regard to Fe(II) uptake, the result that dark 55

Fe uptake was significantly reduced

in the FZ treatment (part C of Figure 5.2) suggests that cellular Fe intake is

accompanied by Fe(II) generation via non-photochemical reductive process(es). It has

been acknowledged that (i) thermal reduction by SRFA (the redox-active moieties

including hydroquinones and semiquinone-type radicals which are typically present

intrinsically in SRFA) (Pullin and Cabaniss, 2003) and (ii) SMIR (Rose, 2012)

facilitates non-photochemical Fe(II) formation. Given the negligible effect of SOD on

the dark 55

Fe uptake at pH 8 (part D of Figure 5.2), however, the latter process is

unlikely to be responsible for dark Fe(II) uptake (though the effect of SOD on 55

Fe

uptake appears to be significant at lower pH 6 as can be seen from Figure A3.5 of

Appendix 3). Thus, to examine the importance of thermal Fe(III) reduction by SRFA

at pH 8, a first-order rate constant for the SRFA-mediated Fe(III) reduction (kdark) was



88

determined in this work (which is consistent with the reported value (Pullin and

Cabaniss, 2003) as described in part A3.2 of Appendix 3). Subsequently, the steady-

state concentration of Fe(II)' ([Fe(II)']SS) was calculated by considering this effect as

well as other competing reactions for Fe(II)' (i.e., oxygenation to Fe(III)' and

recomplexation to form FeIISRFA with second order rate constants of Fe(II)

ox1k and Fe(II)

fk ,

respectively), as follows:

III IIIFe L III Fe L

red-dark red-dark TSS Fe(II) Fe(II) Fe(II) Fe(II)

ox1 2 f ox1 2 f T T

[Fe L] [Fe ][Fe(II)'] =

[O ]+ [L] [O ]+ ([L ]-[Fe ])

k k

k k k k≈ (5.6)

[Fe(II)']SS was determined by using two sets of rate constants available for FeIISRFA

complexation and dissociation (namely the Rose and Bligh Fe(II) models, reactions 11

and 12 in Table 5.1) by assuming the presence of a single ligand class (rather than the

two ligand class) with the use of this model being justified for FeIISRFA complexation

kinetics in the previous work (Rose and Waite, 2003b, Bligh and Waite, 2010). The

Fe(II) uptake rates were then calculated by substituting the two substrate

concentrations of [Fe'] and [Fe(III)'] into eq. 5.1 followed by taking a subtraction for

these two uptake rates (i.e., Fe' Fe(III)'ρ ρ− ), as experimentally determined Fe(II) uptake

represents the difference of 55

Fe uptake rates in the absence and presence of FZ

treatment. Comparison of calculated values with measurement indicate that both the

Rose and Bligh models reasonably account for the measured 55

Fe(II) uptake (Figure

A3.3 of Appendix 3) with this result supporting the conclusion that SRFA-mediated Fe

reduction facilitates Fe uptake. As found for Fe(III) uptake, the decrease in Fe(II)

uptake rate with increase in SRFA concentration is reasonably explained by the

associated decline of concentration of Fe(II)' available for uptake.

By using the Rose and Bligh models, concentrations of unchelated Fe for the two

redox states were calculated as a function of ligand concentration (Figure 5.3) resulting

in the prediction that Fe(II)' is generated at concentrations comparable to Fe(III)' at [L]

> ~10-7

M (corresponding to free SRFA concentration of 0.38 mg.L-1

). At [L] < ~10-7

M, [Fe(II)'] was calculated to be almost independent of [L] whilst [Fe(III)'] increased

with decreasing [L] over the ligand concentrations used. This prediction suggests that



89

the thermal reduction of Fe(III) by SRFA is the rate limiting step in Fe(II)' formation at

lower ligand concentrations.

Figure 5.3. Simulated results for unchelated Fe concentrations (gray lines for Fe(III)

and black lines for Fe(II)) as a function of SRFA ligand concentration by using the

Rose (solid lines) and Bligh (dotted lines) models.

5.3.4. Mode of Light-mediated Fe Uptake

The lack of a marked effect of light on 55

Fe uptake in the SRFA system suggests that

the concentration of Fe available for uptake in the light is comparable to that in the

dark. However, the photochemical reduction of Fe(III)-fulvic acid complex is

acknowledged to facilitate Fe(II) formation at circumneutral pH. Consistent with this

evidence, we also found that the visible light irradiation used in this work is capable of

reducing FeIII

SRFA to Fe(II) at a one order of magnitude greater rate than the dark

reduction by using the FZ-trapping method where Fe(II) formed in the system of

interest is determined by measuring time-course of FeIIFZ3 concentration (see part

A3.2 of Appendix 3 for details). Therefore, one might expected that the steady-state

concentration of Fe(II) species in the light is much higher than that in the dark. One of

the plausible explanations for the apparent contradiction here is that the photo-

generated Fe(II) species involve not only unchelated Fe(II) but also chelated Fe(II) that

-14

-13

-12

-11

-10

-9

-8

-10 -9 -8 -7 -6 -5 -4

Logarithm of ligand concentration (M)

Log

ari

thm

of

Fe

(III

)' o

r

Fe

(II)

' co

nce

ntr

ati

on

(M

)



90

is readily accessible to FZ (at a similar rate as that of unchelated Fe(II)) but not

available for uptake.

The comparable Fe availability under light and dark is likely accounted for by the

relatively similar availability of Fe(II) in the dark and light due to the rapid reoxidation

of photo-generated Fe(II) prior to dissociation of the Fe(II)SRFA species formed. The

dark Fe(II) oxygenation in the SRFA solution is recognized to be faster (150 M-1

.s-1

)

than the inorganic Fe(II) oxidation (8.8 M-1

.s-1

) and other synthetic ligands EDTA (31

M-1

.s-1

) and citrate (2.9 M-1

.s-1

). Furthermore, Fe(II) oxidation is facilitated during

photolysis due to the participation of photo-produced singlet oxygen (1O2) and

(possibly) other inorganic and organic radicals such as superoxide (O2-) and

semiquinones. According to Garg et al. (2012), the steady-state concentration of Fe(II)

species ([Fe(II)]SS) in photolyzed SRFA solution is well accounted for by the balance

of photo-reduction of FeIII

SRFA (IIIFe L

red-lightk ) and oxidation of FeIISRFA by dissolved

oxygen (IIFe L

ox1k ) and singlet oxygen (IIFe L

ox2k ) with the assumption that light-mediated Fe

reduction primarily generates chelated Fe(II), as follows:

III III

II II II II

Fe L III Fe L

red-light red-light T

SS Fe L Fe L 1 Fe L Fe L 1

ox1 2 ox2 2 ox1 2 ox2 2

[Fe L] [Fe ][Fe(II)] =

[O ]+ [ O ] [O ]+ [ O ]

k k

k k k k≈ (5.7)

Assuming an apparent 1O2 concentration of [

1O2]app = 3.5 pM in the photolyzed NOM

solution and a diffusion-controlled rate for the 1O2-mediated Fe(II) oxidation (

IIFe L

ox2k

~1010

M-1

.s-1

) (Garg et al., 2012), the steady-state concentration of photo-generated

Fe(II) was calculated to be 45 pM. Comparison of calculated [Fe(II)']SS under the dark

and light conditions provides relatively similar Fe(II)' concentrations when the Rose

model is used (Figure A3.4 of Appendix 3).

The photochemical FeIIFZ3 formation and

55Fe uptake were also measured in weakly

acidic to alkaline pH where large shifts in the redox potential of Fe complexed by

SRFA have been suggested (Pullin and Cabaniss, 2003) (Figure A3.5 of Appendix 3).

The time-dependent measurement of FeIIFZ3 concentration indicates that

photochemical reduction of FeIII

SRFA is substantially more significant at pH 6-7

compared to pH 8-9. In contrast, only a slight reduction of unsaturated 55

Fe uptake was



91

observed at the lower pH, suggesting that the formation of Fe(II) available for uptake

negligibly increased on light irradiation even though photochemical FeIIFZ3 formation

proceed relatively rapidly. These findings are consistent with the notion that


SRFA does not facilitate Fe uptake. The relatively

large inhibitory effect of FZ on 55

Fe uptake at the lower pH and in the presence of light

(parts C and E of Figure 5.1, and Figure A3.5 of Appendix 3) is well accounted for by

the facilitated reduction of FeIII

SRFA followed by formation of biologically

unavailable FeIIFZ3. The formation of Fe

IIFZ3 results in the decrease in concentration

of the Fe(III) pool (basically both FeIII

SRFA and Fe(III)') and, as such, reduces not

only Fe(II) but also Fe(III) availability. In the scenario discussed above, it is important

to recognise that FZ sequesters not only unchelated Fe(II) but also photo-generated

FeIISRFA directly at an appreciable rate. Although previous reports are mixed, the

ability of FZ and other strong Fe-chelator (such as DFB) to associate adjunctively with

Fe bound to weak Fe-binding ligands is now recognised (Pullin and Cabaniss, 2003,

Pham and Waite, 2008b, Ito et al., 2011).

5.4. IMPLICATIONS OF FINDINGS

Regardless of the inherent complexity of metal binding by natural organic matter, we

have attempted to elucidate the mechanism of Fe uptake by Microcystis aeruginosa in

the presence of the natural organic SRFA to the best of our knowledge with results of

Fe uptake kinetics obtained under both dark and light conditions. The negligible

impact of visible light on 55

Fe uptake by M. aeruginosa is intriguing and in contrast to

the Fe uptake in the EDTA and citrate systems. The findings of the present study are

consistent with the notion that Fe' generated by thermal and reductive dissociation of

the Fe-fulvic acid complex is a primary substrate for uptake by M. aeruginosa..

The rate of Fe uptake at fulvic acid concentrations typical of freshwaters has been

found to be relatively high at pH 8 (Figure 5.3). If ligand concentrations are

standardized by their molecular weights (~2,000 Da for SRFA) (Her et al., 2002), Fe

uptake in the presence of SRFA is comparable to that observed in the presence of

EDTA in the light but a little lower than found for uptake in the presence of citrate in

both the dark and light. Only when the fulvic acid concentration is very high (e.g., >25

mg.L-1

), is the growth rate likely to be limited by Fe availability. The 55

Fe uptake



92

measured for SRFA concentrations <25 mg.L-1

(0.14-2.3 amol.cell-1

.hr-1

) and the

cellular Fe quota under Fe-replete and limited conditions (~1-3 amol.cell-1

) (Dang et

al., 2012) suggests that Fe uptake during daytime is sufficient to sustain the optimal

rate of growth (>~0.76 day-1

).

The difference in light quality between the incubator fluorescent light used in this

work and natural sunlight is recognised. Natural sunlight for example includes

radiation in the UV range which will reduce Fe(III) bound to NOM at a much faster

rate than will visible light. In addition, the light intensity of 157 µmol m-2

s-1

used in

this work is much less than natural sunlight during the day (e.g., ~2 mmol m-2

s-1

).

Thus, under the incubational conditions used in the studies reported here, the rate and

extent of Fe(III) photoreduction is expected to be substantially lower than would be the

case in natural surface waters. Despite this, the finding that the intrinsic Fe(III)

reducing ability of NOM (at least as exemplified by SRFA in the studies described

here) is sufficient to maintain growth of M. aeruginosa at maximum rates is of

considerable significance.

93

CHAPTER 6

CHARACTERISTICS OF THE


MICROCYSTIS AERUGINOSA GROWN IN

IRON-LIMITED CONTINUOUS

CULTURE

Chapter 6. Characteristics of the Freshwater Cyanobacterium Microcystis aeruginosa Grown in Iron-

limited Continuous Culture

94

6.1. INTRODUCTION

Most previous investigations of the cellular phenotype expressed under nutrient

limitation employed batch culture incubations. However, temporal changes to physico-

chemical properties of the medium occur during incubation including pH, nutrient

concentrations and metabolic products (Hoskisson and Hobbs, 2005), such that the

batch method suffers severe limitations with regard to accurately assessing the effect

of growth conditions on cellular response. In addition, the response of cultured

microorganisms varies throughout the growth cycle, which typically consists of a lag

phase, exponential growth, stationary phase and death phase in batch cultures

(Tempest, 1969). In contrast, the growth of microorganisms in continuous culture is

maintained at steady-state throughout the incubation, with metabolic processes and

resultant growth occurring at a constant rate in a relatively stable environment (Herbert

et al., 1956).

The growth response of phytoplankton has been widely investigated in chemostats

operating under nutrient limitation by not only macro-nutrients including nitrogen

(Caperon and Meyer, 1972a, Caperon and Meyer, 1972b, Gotham and Rhee, 1981a)

and phosphorus (Fuhs, 1969, Burmaster, 1979, Gotham and Rhee, 1981b) but also

trace metals (particularly Fe) (Wilhelm and Trick, 1995, Xue et al., 1998, Weger,

1999, Weger and Espie, 2000, Collins et al., 2001, Middlemiss et al., 2001, Weger et

al., 2002, Gress et al., 2004, Weger et al., 2006, Weger et al., 2009, Sonier and Weger,

2010, Wirtz et al., 2010). However, a mathematical theory of trace-metal-limited

continuous culture is lacking. In fact, the chemostat theory used for describing cellular

growth has been developed and subsequently reviewed thoroughly by several authors

(Monod, 1950, Novick and Szilard, 1950, Herbert et al., 1956, Gerhardt and Drew,

1994, Hoskisson and Hobbs, 2005, Andersen, 2005), and shown to be applicable to

macro-nutrients studies where the concentration of limiting substrate is considered as

the total concentration of the macro-nutrient. In contrast, chemostat behavior and

theory under trace metal limited conditions is significantly different to that under

macro-nutrient limited conditions because the limiting substrate is buffered by excess

organic ligands and also because some trace metals (including Fe and Cu) are photo-

reactive. Indeed, under Fe limiting conditions, the “traditional” chemostat theory is no



95

longer suitable for description of the growth and behavior of microorganisms since,

under these conditions, Fe availability is a function of unchelated Fe concentration and

not total Fe concentration with the kinetics of both light-mediated reduction of Fe(III)

species and the oxidation of Fe(II) species critical determinants of steady state

concentration of unchelated Fe. In this work, a modified chemostat theory for Fe-

limited phytoplankton growth is developed and applied to description of the behavior

(including steady state cell density, Fe cell quota and Fe uptake kinetics) of M.

aeruginosa strain PCC7806 grown continuously in Fraquil* medium with Fe activity

buffered by the organic ligand EDTA. The modified chemostat theory developed here

and used to describe the results obtained in this study is presented in full in Section

6.2.7.


6.2.1. Materials

The grade, preparation and storage of all reagents, pH measurements, cleaning

procedure of all lab-ware are described in Section 2.1, Chapter 2.

6.2.2. Culturing Method

Cells of the toxic strain PCC7806 of M. aeruginosa were cultured in Fraquil* medium

in which the speciation of trace metals present could be precisely defined. The detailed

preparation of Fraquil* medium is described in Section 2.2.1, Chapter 2. Briefly, the

medium is buffered by single metal chelator, EDTA, and contains 0.26 mM CaCl2,

0.15 mM MgSO4, 0.5 mM NaHCO3, 0.1 mM NaNO3, 0.01 mM K2HPO4, 1 mM

HEPES, 160 nM CuSO4, 50 nM CoCl2, 600 nM MnCl2, 1.2 µM ZnSO4, 10 nM

Na2SeO3, 10 nM Na2MoO4, 300 nM thiamine HCl, 2.1 nM biotin and 0.41 nM

cyanocobalamin. In this work, the EDTA concentration was maintained constant at 26

µM, while total Fe concentration varied from 10 nM to 10 µM. All salt, trace metal and

vitamin stock solutions were made up in MQ individually rather than as a mixture.

Then, the stocks were mixed in ~1 L MQ, except for Fe and EDTA. The 1 mM stock

of ferric chloride (FeIII

Cl3, Ajax Finechem, Australia) in 0.1 M HCl was mixed with a



96

26 mM solution of EDTA (Na2EDTA, Sigma) prior to mixing with the other stock

solutions in order to prevent precipitation of Fe(III). After mixing all nutrient stocks,

the pH of the medium was adjusted to 8.0 ± 0.05 using concentrated NaOH. The

medium was then sterilized using a 700 W microwave oven for 10 minutes in intervals

of 3, 2, 3 and 2 minutes. After cooling to room temperature, the filter-sterilized vitamin

solutions were added into the medium. Fraquil* with radiolabeled Fe was also prepared

by an identical procedure except for use of radiolabeled 23 mM 55

FeIII

Cl3 (in 0.5 M

HCl, 185 MBq, PerkinElmer, Australia) instead of non-radiolabeled 1.0 mM FeIII

Cl3.

All cultures of M. aeruginosa PCC7806 in Fraquil* were grown in a temperature- and

light-controlled incubator (Thermoline Scientific, Australia) at 27oC under a 14hr:10hr

light:dark cycle with light intensity of 157 µmol photons m-2

s-1

vertically supplied by

cool-white fluorescent tubes. In the original culture used for long-term batch

cultivation, total Fe and EDTA concentrations of 100 nM and 26 µM respectively were

used. Cells were regularly sub-cultured into fresh media when cultures reached

stationary growth phase. Cell density in the culture was counted on a Neubauer

hemacytometer (0.1 mm depth) under an optical microscope (Nikon, Japan). Cellular

size was determined using a Mastersizer 2000 particle size analyzer (Malvern).

6.2.3. Chemostat Apparatus

A metal-free sterile chemostat system was developed for four different flow-rates with

three replicates (see Section 2.2.3, Chapter 2 for the detailed description of the

chemostat system used in this study).

6.2.4. Cellular Fe Quota and External Fe Concentration

In order to quantify steady-state cellular Fe quotas and extracellular Fe concentrations,

the chemostat system was operated at an inflowing 55

Fe concentration of 20 nM with

four dilution rates (0.09, 0.14, 0.17 and 0.25 d-1

). For this purpose, cells were

previously grown batch-wise in 50 nM non-radiolabeled Fe Fraquil* and harvested

during late exponential growth phase by filtration. The filtered M. aeruginosa cells

were then resuspended in 200 mL of Fraquil* medium containing 20 nM radiolabeled

55Fe for each dilution rate in triplicate at a cellular density of ~2.5 × 10

8 cell L

-1. The



97

continuous system was maintained by introducing fresh Fraquil* medium prepared

with 20 nM radiolabeled 55

Fe. The amount of 55

Fe incorporated within cells was then

monitored in triplicate every 2 d until the system approached steady-state using the

following procedure: (i) sampling 1 mL of the cultures, (ii) filtering through a 25 mm

diameter 0.65 µm PVDF membrane (Millipore), (iii) gently washing the filtered cells

at 1 mL min-1

with a solution containing 50 mM Na2EDTA (Sigma) and 100 mM

Na2oxalate (Sigma) adjusted to pH 7 (hereafter referred to as EDTA/oxalate solution)

for 15 minutes in order to eliminate non-specifically adsorbed Fe from the cell surface

(Tovar-Sanchez et al., 2004), (iv) subsequent rinsing with 2 mM sodium bicarbonate

buffer (pH 8) and (v) placing the washed cells in glass scintillation vials with 5 mL of

scintillation cocktail. When the chemostat system reached steady-state, in addition to

the cellular Fe quota, steady-state Fe concentrations were also determined by

collecting the filtrates from the filtration step and setting aside for radioactivity

measurement. The activity (counts per minute) of radioisotope 55

Fe in the washed cells

and the filtrates was measured in a Packard TriCarb Liquid Scintillation Counter and

converted to moles of Fe by performing concurrent counts of 1-5 µL of 55

FeEDTA

stock in 5 mL scintillation cocktail. Procedural blanks were measured by repeating the

identical procedure but with cells absent.


Fe and 14

C Uptake

To prepare steady-state Fe-limited cells used for the short-term uptake experiments,

the chemostat system was operated with 20 nM non-radiolabeled Fe at four different

dilution rates (0.09, 0.14, 0.17 and 0.25 d-1

). 200 mL of batch culture acclimated in

Fraquil* containing 50 nM non-radiolabeled Fe was removed in late exponential

growth phase (cellular density was ~1.5 × 109 cell L

-1) and transferred to the

continuous culture apparatus. The cell density of the cultures was then monitored

regularly every 2 d for a period of ~1 mo. When steady-state conditions were achieved,

cells were harvested onto PVDF membrane filters and rinsed with 5 mL of 2 mM

NaHCO3 for 5 min. The washed cells were then re-suspended into Fe- and EDTA-free

Fraquil* medium at cell densities of (5-7) × 10

8 cell L

-1. Pre-equilibrated

55Fe

IIIEDTA

stock solutions with different Fe:EDTA ratios were added into the cultures to obtain

concentrations of 200 nM 55

Fe and 20-200 µM EDTA. Cells were incubated at 27oC



98

for 1-12 h under light with intensity of 157 µmol photons m-2

s-1

. After the incubation,

cells were again vacuum-filtered onto PVDF membrane filters then rinsed three times

with 1 mL EDTA/oxalate solution and twice with 1 mL of 2 mM NaHCO3 (total

rinsing time was about 10 min). The filtered cells were then collected in scintillation

vials. The radioactivity was measured as described in the procedure above for

determination of cellular Fe quota. Processing steps in the experiment examining

short-term 14

C uptake by M. aeruginosa were identical to those described in the short-

term 55

Fe uptake experiments, except that cells were incubated in Fraquil* medium

([Fe]T = 20 nM and [EDTA]T = 26 µM) containing 0.5 mM 14

C prepared by

substituting the non-radiolabeled NaHCO3 stock with radiolabeled NaH14

CO3

(PerkinElmer, Australia).

6.2.6. Kinetic Model for Unchelated Fe(II) Calculation

In the presence of light, photo-produced unchelated ferrous iron (i.e., Fe(II)’) rather

than total Fe becomes the main substrate for uptake by M. aeruginosa in Fraquil*

medium, as described in detail elsewhere (Fujii et al., 2011a). In a manner similar to

that described in the previous work, the steady-state Fe(II)’ concentration ([Fe(II)’]ss)

was calculated using a kinetic model of Fe transformations that accounts for a variety

of processes including photo-reductive dissociation of FeIII

EDTA into Fe(II)’,

complexation of photo-produced Fe(II)’ by EDTA, dissociation of FeIIEDTA and

oxidation of generated Fe(II)’ to Fe(III)’ by oxygen. The [Fe(II)’]ss was calculated

from the total Fe concentration (≈ [FeIII

EDTA]) and kinetic constants using the

following expression:

[ ] [ ]

' ss

ss2

( )

III II

d Eh DTA

f E A

v

DT ox

Fe EDTA Fe Ek k

k k

DTAFe II

EDTA O

−

−

+ =

+

(6.1)

where the unknown [FeIIEDTA]ss represents the steady-state Fe

IIEDTA concentration

and can be determined from knowledge of the rate of complexation of Fe(II)’ by

EDTA and rates of dissociation and oxidation of FeIIEDTA, i.e.:



99

[ ]

[ ]

'

2

( )f EDTAII ss

ssd EDTA ox EDTA

k Fe IIEDTFe DTA

k

AE

Ok

−

− −

+= (6.2)

Figure 6.1. Model for Fe uptake by M. aeruginosa in the presence of light (Adapted

from Fujii et al. (2011a))

Rate constants reported by Fujii et al. (2011a) were assumed appropriate for use in eqs.

6.1 and 6.2 given that very similar experimental conditions were employed in both

studies (Table 6.1 and Figure 6.1). Assuming that dissolved oxygen is saturated (i.e.,

[O2]~0.25 mM at 25oC) and that [Fe

IIIEDTA] ≈ [Fe]T and [EDTA] ≈ [EDTA]T when

EDTA is in considerable excess of Fe, where the subscript T denotes total

concentration, the two unknown parameters [Fe(II)']ss and [FeIIEDTA]ss were

calculated from eqs. 6.1 and 6.2 using an iterative trial and error method (i.e., by

assuming an initial value of [Fe(II)']ss and calculating [FeIIEDTA]ss from eq. 6.2 then

substituting the calculated [FeIIEDTA]ss into eq. 6.1 to obtain a new value of [Fe(II)']ss;

repeating the process using the new estimate for [Fe(II)']ss and continuing until the

calculated [Fe(II)']ss was equal to the assumed value of [Fe(II)']ss whereupon the

solution had converged). Under the conditions examined here, the calculated steady-

state [Fe(II)’] was approximately proportional to [Fe]T (≈ [FeIII

EDTA]) (see eq. 6.1).

Chapter 6. Characteristics of the Freshwater Cyanobacterium Microcystis aeruginosa Grown in Iron-limited Continuous Culture

100

Table 6.1. Kinetic model for Fe transformation and uptake in the presence of light by M. aeruginosa (adapted from Fujii et al. (2011a) and

therein).

Reaction Rate constant/parameter Value Unit

FeIII

EDTA + hυ → Fe(II)' + EDTAox khυ 6.4 × 10-6

s-1

Fe(II)' + EDTA → FeIIEDTA kf-EDTA 2.1 × 10

-6 M

-1 s

-1

FeIIEDTA → Fe(II)' + EDTA kd-EDTA 1.2 × 10

-3 s

-1

Fe(II)' + O2 → Fe(III)' + O2- kox 8.8 M

-1 s

-1

FeIIEDTA + O2 → Fe

IIIEDTA + O2

- kox-EDTA 31 M

-1 s

-1

Fe(II)' → uptake ρmax or Kρ

'

max '

( )

( )

ssFe

ss

Fe II

Fe IIKρ

ρ ρ

= +

a mol cell

-1 hr

-1 or M

-1

a Fe uptake parameters (ρmax and Kρ) were determined as described in text.



101

6.2.7. Modified Chemostat Theory

The theory of continuous culturing was first described by Monod (1950) and Novick

and Szilard (1950) independently and reviewed subsequently by a number of authors

(Herbert et al., 1956, Gerhardt and Drew, 1994, Hoskisson and Hobbs, 2005,

Andersen, 2005). Growth kinetics in a chemostat system are characterized by key

parameters including the specific growth rate µ (d-1

), the growth rate constant µmax (d-1

)

(which is equal to the maximum value of µ at saturation levels of the limiting substrate

S (M)), the half-saturation constant Ks (M) (which represents the concentration of the

growth limiting substrate that yields a growth rate of 0.5µmax) and the yield constant Y

(cell mol-1

of limiting substrate). Such growth parameters are typically estimated from

a preliminary batch culture experiment prior to commencement of a chemostat

incubation.

The specific growth rate during the exponential growth phase in a batch system is

calculated using the following equation (Monod, 1950, Herbert et al., 1956):

dxx

dtµ= (6.3)

where x (cell L-1

) represents the population size and t (d) is the incubation period. A

straight line can be fitted to a semi-log plot of the ratio of the population size at an

arbitrary time to the initial population size (i.e.,0

lnx

x) versus time interval t. The

specific growth rate during the exponential phase µ then corresponds to the slope of

the fitted line. Both the growth rate constant and half-saturation constant can be

obtained by observing specific growth rates over a range of limiting substrate

concentrations and applying the Monod equation:

max

s

S

K Sµ µ

=

+ (6.4)



102

The yield constant Y is defined as the reciprocal of cellular nutrient quota and

quantified using growth rate and rate of utilization of the limiting nutrient as follows:

0

0

Number of organism formed

Weight of limiting substrate used S S

dxx xdtY

dSdt

−= − = =

− (6.5)

where x0 (cell L-1

) is the density of inoculated organism and S0 (M) is the initial total

concentration of limiting substrate in the medium under conditions where the total

substrate is available for cellular uptake. During stationary growth phase in a batch

culture, it is reasonable to assume that all of the limiting substrate present in the

medium has been converted to cellular material. In this situation, the yield constant in

eq. 6.5 may be estimated using the following equation:

0

0

–ssx xY

S= (6.6)

where xss (cell L-1

) is the cell density in stationary growth phase. The estimated values

of µmax, Ks and Y together with eqs. 6.3 to 6.5 provide both a quantitative description of

the growth cycle of algal cells in batch cultures and preliminary data for predicting the

growth behavior of cells in continuous cultures.

In continuous cultures operating under perfect mixing conditions, net rates of change

in organism density dx

dt

and substrate concentration dS

dt

can be expressed as

follows:

( ) max s

dx Sx D x D

dt K Sµ µ

= − = −

+ (6.7)

( ) ( )

maxI I

s

xdS x SD S S D S S

dt Y Y K S

µµ = − − = − −

+ (6.8)



103

where D (d-1

) is the dilution rate (defined as the ratio of the inflow rate of the feed

medium F (L d-1

) and the culture volume V (L)) and SI is the concentration of substrate

in the feed medium. When the continuous system reaches steady-state (i.e.,

0dx dS

dt dt= = ) with constant SI and D, the steady-state concentrations of substrate ( %S )

and cells ( %x ) are uniquely defined as follows:

%S = Ks

D

µmax

− D

(6.9)

%x = Y SI

− %S( ) = Y SI

− Ks

D

µmax

− D

(6.10)

Under steady-state conditions, D is equal to the specific growth rate

µ = µmax

%S

Ks+ %S

. D has a maximum value, generally referred to as the critical dilution rate (Dc), which

is equal to the highest value of µ obtained when S has its highest value (i.e., SI), i.e.

Ic max

s I

SD

K Sµ

=

+ . If D > Dc, any cells present in the culture vessel will be washed

out completely.

Eqs. 6.7 to 6.10 describe completely the dependency of growth of microorganisms in

continuous cultures on the total concentration of limiting substrate. However, these

equations are not applicable to continuous cultures with trace metals as the limiting

substrate since the trace metal is generally buffered by excess organic ligands in the

culturing medium, with only a small portion of the total metal present available for

uptake at any given time. While eq. 6.7 describes growth kinetics as a function of the

concentration of limiting substrate available for uptake, eq. 6.8 is derived from a total

mass balance of the substrate in the reactor. Provided the concentration of buffered

trace metal is approximately proportional to the total metal concentration, the

following relationships between total and available portions of limiting trace metals

can be introduced:



104

' T

S mS= (6.11)

' TS S

K mK= (6.12)

where ST and S’ (M) represent the total and available concentrations for the limiting

substrate in the reactor respectively, TS

K and 'S

K (M) are the half-saturation constants

obtained when ST and S’ are considered as substrate concentration, and m is a constant.

Using these relationships, eqs. 6.7 and 6.8 can be re-written as follows:

( ) m '

'

ax

S

dx Sx D x D

dt K Sµ µ

′

= − = −

+ (6.13)

( )I T

dS xD S S

dt Y

µ= − − (6.14)

At steady-state, where 0dx dS

dt dt= = and µ = D, eqs. 6.9 and 6.10 become:

%S' = K

S '

D

µmax

− D

(6.15)

%x = Y SI− %S

T( ) = Y SI

− KS

T

D

µmax

− D

(6.16)

where

%ST

= m %S' = mK

S '

D

µmax

− D

= K

ST

D

µmax

− D

When µmax, 'S

K , TS

K and Y for a given organism and growth medium are known, the

behavior of a continuous culture at steady-state under limitation of a buffered trace

metal can be completely defined by eqs. 6.15 and 6.16, where the steady-state

concentrations of substrate and cells depend solely on the values of the total substrate

concentration in the inflowing medium SI and the dilution rate D. With a fixed



105

inflowing substrate concentration, variation of dilution rate leads to variation of the

steady-state cell density and substrate concentration in the continuous system. At D =

0, for example, cell density becomes maximum and the limiting substrate

concentration approaches zero, approximately corresponding to the latter stages of a

batch culture. As the dilution rate increases towards Dc, steady-state cell density and

substrate concentration approach zero and SI, respectively. With a fixed dilution rate 0

< D < Dc, the substrate concentration in the culture must reach a level that is

independent of SI so that the specific growth rate µ is equal to the dilution rate D,

whilst cell density increases with increasing SI.


6.3.1. Growth Kinetics in Batch Culture

As discussed above (Section 6.2.7), if the values of four growth constants: maximum

specific growth rate µmax (d-1

), yield constant Y (cell mol-1

) and half-saturation

constants TS

K and 'S

K (M) obtained when total and available concentrations (ST and

S’ (M), respectively) of the limiting substrate are considered as substrate concentration

are known, the behavior of a continuous culture at steady-state under limitation of a

buffered trace metal can be completely defined by eqs. 6.15 and 6.16.

To predict the behavior of Fe-limited chemostat cultures of M. aeruginosa PCC7806, a

preliminary study of the growth kinetics of this organism in batch cultures was

conducted under incubation conditions identical to those used in the chemostat study

with regard to the culture volume and vessels, growth medium, temperature and light

intensity. Only [Fe]T was modified in order to investigate the growth kinetics under

various degrees of Fe limitation. The parent culture in exponential growth phase was

sub-cultured and grown in triplicate in Fraquil* media with [Fe]T ranging from 0.01 to

10 µM. Application of the exponential growth equation (eq. 6.3) to the initial linear

section of a semi-log plot provided specific growth rates of M. aeruginosa PCC7806 in

the batch culture ranging from 0.23 ± 0.012 to 0.82 ± 0.049 d-1

(Figure 6.2). [Fe]T > 1

µM was found to be sufficient to support optimal growth of M. aeruginosa, whilst at

[Fe]T ≤ 0.1 µM the growth rate of M. aeruginosa declined due to the depletion of Fe



106

available for uptake. These growth rates were consistent with previously reported

values for the specific growth rates of M. aeruginosa PCC7806 in batch Fraquil*

culture at somewhat higher [Fe]T of 0.1-10 µM (Fujii et al., 2010a, Fujii et al., 2011b).

Figure 6.2. Growth curves in batch cultures of M. aeruginosa at different total Fe

concentrations in Fraquil*. Total Fe concentrations were varied from 10 nM to 10 µM;

all other media components were constant. Symbols represent the mean and error bars

represent the standard deviation from triplicate incubations (filled diamonds = 10 nM

[Fe]T, filled squares = 20 nM [Fe]T, filled triangles = 50 nM [Fe]T, open diamonds =

100 nM [Fe]T, open squares = 1 µM [Fe]T, and crosses = 10 µM [Fe]T).

Maximum growth rate and half-saturation constants were estimated via non-linear

regression of the data using the Monod equation (i.e.,

max

s

S

K Sµ µ

=

+ , Figure 6.3).

The regression analysis was performed for cases where both total Fe and calculated

steady-state Fe(II)’ were treated as the appropriate substrate concentration. Since the

7.0

7.5

8.0

8.5

9.0

9.5

0 2 4 6 8 10 12 14 16

Logari

thm

of

cell

den

sity

(ce

ll L

-1)

Time (d)



107

steady-state [Fe(II)’] is essentially proportional to the total [Fe] (≈[FeIII

EDTA]), in

each case the theoretical specific growth rates as a function of Fe concentration fitted

well the measured growth rates of M. aeruginosa, yielding the same value for the

growth constant (µmax = 0.80 ± 0.03 d-1

). In contrast, a much lower value of the half-

saturation constant 'S

K = 3.6 ± 0.32 fM with respect to Fe(II)’ was deduced compared

with TS

K = 26 ± 2.3 nM for total Fe. Although response of growth rate to Fe limitation

is typically expressed in terms of total Fe, expression of 'S

K in terms of Fe(II)’ is more

appropriate given that the bioavailable form of Fe in our system is unchelated Fe(II) as

a result of the photoreductive dissociation of organically complexed Fe.


) and log concentration of

unchelated Fe(II)’ (where [Fe(II)’] is in molar (M) units) in batch culture studies of M.

aeruginosa. Non-linear regression analysis yielded a half saturation constant for

growth of 'S

K = 3.6 ± 0.32 fM (with respect to Fe(II)’) and a maximum specific

growth rate µmax = 0.80 ± 0.03 d-1

. Solid and dotted lines represent the regression line

and 95% confidential interval, respectively. Symbols indicate data for experimentally

determined growth rate under different degrees of Fe limitation.

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

-15.0 -14.5 -14.0 -13.5 -13.0 -12.5 -12.0 -11.5

Sp

eci

fic

gro

wth

ra

te µ

(d

-1)

log([Fe(II)'] (M))

R 2 = 0.9689



108

In a batch culture where growth rate is controlled solely by the concentration of a

single limiting nutrient, it would be reasonable to assume that the limiting substrate in

the growth medium has been completely consumed upon reaching stationary growth

phase. Hence, assuming that the concentration of limiting nutrient is approximately

zero at this point, the yield constant Y of M. aeruginosa under Fe limitation ([Fe]T of

0.01-0.1µM) was determined to be 8.1 ± 0.21 × 1016

cell (mol Fe)-1

by use of eq. 6.6.

6.3.2. Performance of Chemostat System under Fe Limitation

Total Fe concentrations of less than 50 nM in the Fraquil* growth medium were used

in continuous cultures in this study to ensure that cultures were maintained under Fe-

limited conditions. Using the growth parameters for M. aeruginosa PCC7806 obtained

from the Fraquil* batch culture studies, expected values of both the steady-state

concentrations of M. aeruginosa cells and unchelated photo-reductively produced

Fe(II)’ concentrations were calculated as a function of dilution rate, as illustrated in

Figure 6.4. Critical dilution rates were determined to be 0.34 d-1

for [Fe]T = 20 nM and

0.52 d-1

for [Fe]T = 50 nM. The continuous cultures were then maintained in Fraquil*

medium at dilution rates < Dc with 50 nM Fe (dilution rates of 0.07, 0.15, 0.30 and

0.45 d-1

) and 20 nM Fe (dilution rates of 0.09, 0.14, 0.17 and 0.25 d-1

) for a period of 4

wk (Figure 6.5).

In the system with [Fe]T = 50 nM (part A of Figure 6.5) there was a 4-day lag before

cells began to grow, suggesting that the cells took some time to adjust to the change in

medium conditions. At lower dilution rates (0.07, 0.15 and 0.30 d-1

), cell density

increased with time after day 4. In contrast, at the highest dilution rate (0.45 d-1

), the

cell number declined significantly from day 4 to day 20 implying that the wash-out

rate was initially higher than the net growth rate under these conditions. The

continuous cultures appeared to be at steady-state after ~20 d when the variation of the

cell density with time was less than 5% of the average cell density. In the system with

20 nM radiolabeled 55

Fe, a lag of 2 d after inoculation was also observed, which was

then followed by stable increase of cells from day 2 to day 12, when each system

achieved almost maximum cell yields. The cell concentrations then slightly decreased

to reach steady-state growth at around day 20 (part B of Figure 6.5). In both systems,

as expected, the steady state cell density declined with decreasing degree of iron



109

limitation (i.e. increasing dilution rate) which is consistent with data reported in other

Fe-limited chemostat studies (Weger, 1999, Weger et al., 2002).


concentration in continuous cultures of M. aeruginosa as a function of dilution rate

with different total Fe concentrations in the inflowing medium (50 nM and 20 nM).

Symbols represent data for steady-state cell density in Fraquil* medium with total Fe

of 50 nM (circles) and 20 nM (triangles). Dotted lines are the theoretical values of

steady-state cell density calculated from eq. 6.16 with growth parameters estimated

from batch culture studies (µmax = 0.80 ± 0.03 d-1

, 'S

K = 3.6 ± 0.32 fM with respect to

Fe(II)’, TS

K = 26 ± 2.3 nM with respect to total Fe, and Y = 8.1 ± 0.21 × 1016

cell (mol

Fe)-1

), while bold lines indicate the theoretical steady-state cell density estimated with

parameters obtained from continuous culture studies ( 'S

K = 3.4 ± 0.82 fM, TS

K = 25 ±

5.0 nM and Y = 1.1 ± 0.2 × 1017

cell mol-1), except for µmax (0.80 ± 0.03 d

-1) which was

obtained from the batch studies. Dashed and chained lines indicate predicted steady-

state unchelated Fe(II)’ concentrations estimated using parameters from batch and

continuous culture studies, respectively.

0

10

20

30

40

50

60

0 0.1 0.2 0.3 0.4 0.5 0.6

[Fe(

II)'

]ss

(fM

)

or

[Cel

l]ss

10

8(c

ell

L-1

)

Dilution rate (d-1)

Chapter 6. Characteristics of the Freshwater Cyanobacterium Microcystis aeruginosa Grown in


Figure 6.5. Growth of

dilution rates with total Fe concentrations in the inflowing Fraquil

= 50 nM, with dilution rates of 0.07 d

(triangles) and 0.45 d-


(squares), 0.17 d

represent the mean and error bars the standard deviation from triplicate incubations.

Dashed lines represent

8.8

8.9

9.0

9.1

9.2

9.3

9.4

9.5

9.6

9.7

9.8

0 5

Logari

thm

of

cell

den

sity

(ce

ll L

-1)

8.2

8.4

8.6

8.8

9.0

9.2

9.4

9.6

0 5Logari

thm

of

cell

den

sity

(ce

ll L

-1)

Characteristics of the Freshwater Cyanobacterium Microcystis aeruginosa Grown in

Growth of M. aeruginosa in the continuous culture system at different

dilution rates with total Fe concentrations in the inflowing Fraquil*

= 50 nM, with dilution rates of 0.07 d-1


-1 (circles). (B) [Fe]T = 20 nM, with dilution rates of 0.09 d

(squares), 0.17 d-1



the 95% confidence interval at steady-state.

10 15 20

Time (d)

10 15 20

Time (d)

Characteristics of the Freshwater Cyanobacterium Microcystis aeruginosa Grown in Iron-

110

in the continuous culture system at different

medium. (A) [Fe]T

(squares), 0.30 d-1

= 20 nM, with dilution rates of 0.09 d-1

(circles). Symbols


25 30

25 30



111

When the chemostat system supplied with 20 nM 55

Fe reached steady-state, cell

density and the total concentration of Fe in each reactor was determined. The steady-

state cell density () and steady-state total Fe () were subsequently used to re-

calculate the half saturation constant and the yield constant using the following

equations derived from eqs. 6.15 and 6.16 (Herbert et al., 1956):

KS

T

= %ST

µmax

− D

D

(6.17)

Y =%x

SI

− %ST

(6.18)

This produced a value of Y = 1.1 ± 0.2 × 1017

cell (mol Fe)-1

, which is comparable to

the value of 8.1 × 1016

cell (mol Fe)-1

derived in the batch culture studies. Although we

assumed that all Fe was completely consumed by cells at the stationary phase, a slight

underestimation of Y from the batch cultures suggests that during these incubations,

some portion of Fe present in the medium is transformed to a non-available form of Fe,

possibly due to precipitation of Fe (as a ferric oxyhydroxide) or loss of Fe by

adsorption to vessel surfaces. Similarly, eq. 6.17 produced a value of TS

K = 25 ± 5.0

nM in the continuous culture studies, with eq. 6.12 providing 'S

K = 3.4 ± 0.82 fM.

These values are also consistent with those determined in batch cultures.

The values for 'S

K , TS

K and Y obtained from the continuous culture studies and the

estimated value of µmax in the batch culture studies were used to re-calculate the

theoretical steady-state cell and Fe concentrations. For comparative purposes,

measured steady-state cell densities at different dilution rates are plotted on the same

graph as the theoretical data (Figure 6.4). The theoretically calculated values are in

reasonable agreement with the experimentally determined data, indicating that the cell

density decreases in accordance with the increase in dilution rates, while increasing

[Fe]T from 20 to 50 nM leads to an increase in the cell number.

Finally, to verify the hydraulic performance of the chemostat system, the water level

inside the culture vessels was monitored every 2 d and the medium inflow rates were



112

measured before and after the system had been operated continuously for 4 wk. No

significant change was observed in either parameter over this period. Thus the

chemostat system developed for the study of M. aeruginosa in this work is able to

operate continuously for a period of at least one month without any evidence of

contamination or other problems and, as such, cultures of this microorganism could be

maintained at steady-state over a range of Fe nutritional conditions.

6.3.3. Cellular Fe Quota

The amount of 55

Fe internalized by cells was measured every 2 d for a month in the

cultures supplied with 20 nM Fe (Figure 6.6). During the non-steady-state phase,

relatively large fluctuations in cellular Fe quotas were observed at each dilution rate.

After day 20, when the system approached steady-state growth, the cellular Fe quotas

at each dilution rate became constant with time. Cellular Fe quotas (Q) under steady-

state conditions increased as dilution rates (i.e. specific growth rates) increased,

consistent with Droop theory (Droop, 1973). According to this theory, specific growth

rates are predicted to hyperbolically increase with increasing steady-state Fe quota

(part A of Figure 6.7) as follows:

' minmax

Q Q

Qµ µ

−=

(6.19)

or rearranging:

' '

max max minQ Q Qµ µ µ= − (6.20)

where µ’max (d-1

) is the maximum (“impossible”) specific growth rate achieved at

infinite cellular Fe quota, Qmin (zmol cell-1

) is the minimum subsistence quota and µQ

(zmol cell-1

d-1

) represents the specific Fe uptake rate or the long-term uptake rate for

growth ρµ (µQ). The cellular Fe quota parameter µ’max in eqs. 6.19 and 6.20 (which

relates to internal substrate concentration) is unrelated to the growth rate constant µmax

(0.80 d-1

) in eq. 6.4 (which relates to external substrate concentration).



113

Figure 6.6. Time-course of cellular Fe quotas for Fe-limited M. aeruginosa in the

chemostat with [Fe]T = 20 nM as radiolabelled 55

Fe in the inflowing medium and



(squares), 0.17 d-1


(circles). Symbols represent the mean and error bars the standard deviation from

triplicate incubations.

Plotting against Q produces a linear transformation of the Droop equation with a

slope of µ’max and an intercept of µ’maxQmin, as shown in part A of Figure 6.7 for long-

term Fe uptake by M. aeruginosa. This plot yields values of Qmin= 1.2 ± 0.24 amol

cell-1

and µ’max = 0.37 ± 0.04 d-1

. Assuming that cell diameter of M. aeruginosa

PCC7806 is about 4.0 µm, i.e., cell volume of ∼33.5 µm3, this results in Qmin = 3.58 ×

10-5

mol Fe per liter-cell which was comparable to 2.1 ± 0.05 × 10-5

mol Fe per liter-

cell of Thalassiosira weissflogii (Anderson and Morel, 1982) or lower than those of

other species reported previously such as 4.8-33.3 × 10-5

mol Fe per liter-cell of

Dunaliella tertiolecta (Davies, 1970); 6.9-15.9 × 10-5

mol Fe per liter-cell of Pavlova

lutherii (Droop, 1973) and 9.3-18.6 × 10-5

mol Fe per liter-cell of T. weissflogii

(Harrison and Morel, 1986). The lower value of Fe quotas of M. aeruginosa observed

in this study relative to those of other microorganisms in previous studies may partly

due to removal of extracellular iron during washing step by EDTA/oxalate solution.

0

500

1000

1500

2000

2500

3000

3500

4000

0 5 10 15 20 25 30

Cel

lula

r F

e q

uota

(zm

ol

cell

-1)

Time (d)



114

Figure 6.7. Relationship between the cellular Fe quota (Q) and the (A) specific uptake

rate of Fe (µQ) or (B) specific growth rate (µ) for Fe-limited M. aeruginosa under

steady-state conditions in continuous cultures. The system was operated at four

different dilution rates (0.09, 0.14, 0.17 and 0.25 d-1

) and fed with Fraquil* medium

containing 20 nM radiolabeled 55

Fe. In panel (A), linear regression analysis

(represented by the bold line) yielded the maximum “impossible” growth rate µ’max =

0.37 ± 0.04 d-1

and minimum cellular quota Qmin = 1.2 ± 0.2 × 103 zmol cell

-1. Symbols

represent the mean and error bars the standard deviation from triplicate incubations. In

Ay = 0.373x - 460

R² = 0.95

0

200

400

600

800

1000

1000 1500 2000 2500 3000 3500 4000

Sp

ecif

ic u

pta

ke

rate

µQ

(zm

ol

cell

-1d

-1)

Cellular Fe quota Q (zmol cell-1)

B

0.0

0.1

0.2

0.3

0.4

0 1000 2000 3000 4000 5000 6000 7000

Sp

ecif

ic g

row

th r

ate

µ (

d-1

)

Cellular Fe quota Q (zmol cell-1)

Critical dilution rate Dc = 0.35



115

panel (B), the solid line represents the theoretical curve calculated from the Droop

equation using the obtained estimated values of µ’max and Qmin. Symbols represent the

mean from triplicate incubations.

As expected, the calculated µ’max was comparable to the critical dilution rate Dc = 0.35

d-1

for the system supplied with 20 nM Fe. The hyperbolic relationship between the

cellular Fe quotas and the specific growth rates in this work (part B of Figure 6.7) is

consistent with other observations for cyanobacteria such as Anabaena sp. and

Microcystis sp. under nitrogen or phosphorous limitation (Gotham and Rhee, 1981b,

Ahlgren, 1985, Olsen, 1989) and eukaryotic phytoplankta such as green algae

Chlamydomonas sp. and diatom Thalassiosira sp. under Fe, Mn, vitamin B12 or

phosphorus limitation (Goldman and Mccarthy, 1978, Sunda and Huntsman, 1985,

Sunda and Huntsman, 1986, Harrison and Morel, 1986).

6.3.4. Fe Uptake Kinetics

Short-term 55

Fe uptake assays were undertaken using cells collected from the

chemostat supplied with 20 nM non-radiolabelled Fe. In this assay, four batch

experiments were prepared by filtering cells from the steady-state chemostat cultures

at each of the dilution rates examined and subsequently resuspending them in Fe- and

ligand-free Fraquil*. The short-term uptake assay was then initiated by adding pre-

equilibrated 55

FeIII

EDTA to each batch experiment. Intracellular 55

Fe accumulated

during incubation in the light was then measured every 1-2 h for 12 h. Incubations

were conducted in the light based on previous findings that 55

Fe uptake rates by M.

aeruginosa in EDTA-buffered Fraquil* medium are substantially higher under light

irradiation than in the dark due to the higher concentration of bioavailable substrate

resulting from photoreductive dissociation of the FeIII

EDTA complex present in the

medium (Fujii et al., 2011a). The Fe uptake rate of each steady-state culture of M.

aeruginosa was investigated over a range of unchelated photo-produced Fe(II)’

concentrations by varying the EDTA concentrations from 20 to 200 µM, while 55

Fe

concentration was maintained constant (200 nM). Intracellular 55

Fe accumulated in a

linear manner with respect to time for at least 4 h, followed by rapid decline in uptake

rate after 6 h, indicating that uptake became saturated in the latter stages of the

experiment (Figure 6.8). A Fe mass balance in these experiments indicated that



116

depletion of extracellular Fe available for uptake was unlikely to have occurred over

the duration of the short-term assays.


Fe uptake during batch short-term Fe uptake assays using

cells obtained at steady-state from the chemostat cultures grown with [Fe]T = 20 nM in

the inflowing medium and dilution rates of 0.09 d-1


(squares),

0.17 d-1


(circles). In the short-term uptake assay, each culture

was incubated in Fraquil* with either (A) 20 µM EDTA or (B) 200 µM EDTA and

R² = 0.9979

R² = 0.9932

R² = 0.9775

R² = 0.9925

0

1000

2000

3000

4000

5000

6000

7000

0 2 4 6 8 10 12 14

Acc

um

ula

ted

5

5F

e (z

mol

cell

-1)

Time (hr)

A. [EDTA] = 20 µM

R² = 0.9986

R² = 0.9974

R² = 0.9969

R² = 0.9844

0

200

400

600

800

1000

1200

1400

1600

0 2 4 6 8 10 12 14

Acc

um

ula

ted

5

5F

e (z

mol

cell

-1)

Time (hr)

B. [EDTA] = 200 µM



117

constant concentration of radiolabeled 55

Fe (200 nM). Symbols represent the mean and

error bars the standard deviation from triplicate experiments. The continuous lines

were obtained by linear regression of data collected within 4 h (represented by closed

symbols) for each culture.

Assuming that the short-term uptake rate for M. aeruginosa follows classical

Michaelis-Menten kinetics, the Fe uptake rate can be described as follows:

'

max '

( )

( )

ssFe

ss

Fe II

Fe IIKρ

ρ ρ

= +

(6.21)

where ρmax (mol cell-1

s-1

) is the maximum uptake rate and Kρ (M) is the half-saturation

constant for short-term Fe uptake. Steady-state concentrations of Fe(II)’ were

calculated from eqs. 6.1 and 6.2 using a trial and error method (Table 6.2). To

determine the uptake parameters, an Eadie-Hofstee linear transformation was applied

to the measured 55

Fe uptake rates in each culture as shown in Figure 6.9. Linear

regression analysis gave comparable half-saturation constants (Kρ = 18 ± 2.2 fM as

Fe(II)’) but significantly different maximum uptake rates (ρmax of 0.27 ± 0.030, 0.72 ±

0.060, 0.95 ± 0.080 and 1.0 ± 0.090 amol Fe cell-1

hr-1

) for the cultures grown at

different dilution rates, suggesting that the cells in the continuous cultures adjusted to

different degrees of Fe-limitation by varying their maximum uptake rate ρmax rather

than the half-saturation constant Kρ. This ability to regulate short-term uptake has

previously been suggested in studies of algal growth under limitation by macro-

nutrients such as N, P and Si (Tilman and Kilham, 1976, Goldman and Mccarthy,

1978, Mccarthy, 1981) as well as by Fe (Harrison and Morel, 1986). In comparison to

other studies, the maximum uptake rates ρmax obtained in this study (assuming a cell

volume of M. aeruginosa of ∼33.5 µm3) ranged from 0.8 to 2.9 × 10

-5 mol Fe liter-cell

-

1 hr

-1 and was one order lower than those of Thalassiosira weissflogii reported in

Anderson and Morel (1982) (ρmax = 1.1 – 2.1 × 10-4

mol Fe liter-cell-1

hr-1

), Harrison

and Morel (1986) (ρmax = 2.4 ± 0.02 × 10-4

mol Fe liter-cell-1

hr-1

) and Morel (1987)

(ρmax = 3.2 × 10-4

mol Fe liter-cell-1

hr-1

). This is likely because lower Fe availability in

marine systems may lead to higher uptake capacity of marine microorganisms relative

to those of freshwater microorganisms.



118

Figure 6.9. Eadie-Hofstee plots demonstrating the linear relationship between the

short-term 55

Fe uptake rate (ρFe) and the ratio ρFe/[Fe(II)’] (d-1

M-1

) for cultures of M.

aeruginosa. Linear regression analysis yielded comparable half-saturation constants

for Fe uptake (Kρ = 18 ± 1.9 fM, as Fe(II)’) but different maximum specific uptake

rates (ρmax of 270, 720, 950 and 1,010 zmol cell-1

hr-1

for cultures grown at dilution

rates of 0.09 d-1


(squares), 0.17 d-1


(circles), respectively). Lines for 95% confidential intervals were omitted for clarity.

y = -15.4x + 273

R² = 0.94

y = -18.7x + 718

R² = 0.99

y = -20.7x + 951

R² = 0.99

y = -18.4x + 1014

R² = 0.95

0

100

200

300

400

500

600

700

800

0 10 20 30 40 50 60

55F

e u

pta

ke

rate

(zm

ol

cell

-1h

r-1)

55Fe uptake rate/[Fe(II)']ss (µmol cell-1 hr-1 M-1)


119

Table 6.2. Measured and calculated Fe uptake parameters under the conditions of short-term 55

Fe uptake experiments for four different steady-

state cultures of M. aeruginosa.

Dilution rate Initial concentrations Calculated steady-state concentrations Measured and calculated Fe uptake rates

[FeIII

EDTA]initial [EDTA]initial [Fe(II)']ssa

[FeII

EDTA]ssa

ρFe-measured ρFe-calculatedb

d-1

M M M M zmol cell-1

hr-1

zmol cell-1

hr-1

0.09

2.0 × 10-7

2.0 × 10-5

3.54 × 10-14

1.72 × 10-10

180 ± 22 180

2.0 × 10-7

5.0 × 10-5

1.42 × 10-14

1.72 × 10-10

118 ± 12 119

2.0 × 10-7

1.0 × 10-4

7.08 × 10-15

1.72 × 10-10

85 ± 11 76

2.0 × 10-7

2.0 × 10-4

3.54 × 10-15

1.72 × 10-10

44 ± 2 44

0.14

2.0 × 10-7

2.0 × 10-5

3.54 × 10-14

1.72 × 10-10

478 ± 24 474

2.0 × 10-7

5.0 × 10-5

1.42 × 10-14

1.72 × 10-10

301 ± 18 313

2.0 × 10-7

1.0 × 10-4

7.08 × 10-15

1.72 × 10-10

198 ± 22 201

2.0 × 10-7

2.0 × 10-4

3.54 × 10-15

1.72 × 10-10

115 ± 11 117

0.17

2.0 × 10-7

2.0 × 10-5

3.54 × 10-14

1.72 × 10-10

592 ± 33 627

2.0 × 10-7

5.0 × 10-5

1.42 × 10-14

1.72 × 10-10

395 ± 21 415

2.0 × 10-7

1.0 × 10-4

7.08 × 10-15

1.72 × 10-10

246 ± 14 265

2.0 × 10-7

2.0 × 10-4

3.54 × 10-15

1.72 × 10-10

135 ± 19 154

0.25

2.0 × 10-7

2.0 × 10-5

3.54 × 10-14

1.72 × 10-10

886 ± 52 669

2.0 × 10-7

5.0 × 10-5

1.42 × 10-14

1.72 × 10-10

540 ± 25 443

2.0 × 10-7

1.0 × 10-4

7.08 × 10-15

1.72 × 10-10

350 ± 24 283

2.0 × 10-7

2.0 × 10-4

3.54 × 10-15

1.72 × 10-10

220 ± 13 164

a The two unknown concentrations [Fe(II)’]ss and [Fe

IIEDTA]ss calculated from eqs. 6.1 and 6.2 using trial and error method where ρFe is a measured

55Fe uptake rate.

b The Fe

uptake rate estimated from eq. 6.21 using the constant half-saturation constant (Kρ =18 ± 1.9 × fM as Fe(II)’) and different maximum uptake rates (ρmax of 270, 720, 950 and

1,010 zmol cell-1

hr-1

for the cultures grown at dilution rates of 0.09 d-1

, 0.14 d-1

, 0.17 d-1

and 0.25 d-1

, respectively) where Fe(II)’ is considered as the limiting substrate.



120

6.3.5. Cellular Response to Fe Limitation in Chemostat

The half-saturation constant with respect to Fe uptake (Kρ) was determined to be

higher than that for the growth rate (KS’) by ~5-fold, suggesting that the decline in Fe

uptake due to low Fe availability does not necessarily result in a decline in capacity for

cellular growth. Rather, an optimal growth rate is maintained until Fe concentrations

are imposed that are substantially lower than those at which a decline in Fe uptake

begins to occur, after which growth rate starts to decrease. This is consistent with

previous observations that cells of the marine cyanobacterium Synechococcus exhibit

symptoms of Fe stress under low Fe availability (e.g., declining photosynthetic

activity) well before growth rate is affected, implying that cellular division is accorded

a higher priority than general metabolic functioning even in low nutrient environments

(Henley and Yin, 1998). Different preconditioning of cells in the Fe-limited chemostat

resulted in altered responses of cells with regard to Fe uptake. The Fe uptake capacity

increased as the degree of Fe-limitation decreased from the most-starved condition to

the least starved. A similar trend was also found in iron-limited chemostat cultures of

green algae Chlamydomonas reinhardtii (Weger, 1999), Chlorococcum

macrostigmatum and Stichococcus bacillaris (Weger et al., 2002) where an increase in

dilution rate (or decrease in degree of Fe limitation) led to an increase in plasma

membrane ferric chelate reductase (FC-R) activity, hence, likely an increase in Fe(II)

uptake capacity. However, this trend is the reverse of the expected relationship

between cellular Fe quota and uptake rate where cells with a higher degree of nutrient

limitation generally exhibit higher uptake rates (Gotham and Rhee, 1981a, Gotham and

Rhee, 1981b, Olsen, 1989).

Provided that Fe uptake is mediated by concentration gradient dependent passive

diffusion through non-specific trans-membrane channels (porins) as suggested by Fujii

et al. (2011a) for M. aeruginosa and by Jones and Niederweis (2010) for

Mycobacterium smegmatis, Fe uptake rate is likely proportional to cell surface area. A

slight increase (by a factor of ~1.5) in average cell volume (and hence cell surface

area) was found in P-limited continuous cultures of Monochrysis lutheri with increase

in dilution rate from 0.1 d-1

to 1.0 d-1

(Burmaster, 1979). A similar trend was found in

N-limited continuous cultures of Chlorella pyrenoidosa (Williams, 1965). Thus, an



121

increase in dilution rate may result in an increase in cell surface area and hence in cell

normalized Fe uptake rate. However, there was no substantial change in the size of M.

aeruginosa cells with increasing dilution rates in this work (mean diameters of 4.1 ±

0.07, 3.8 ± 0.01, 3.8 ± 0.05 and 3.8 ± 0.01 µm in cultures grown at dilution rates of

0.09, 0.14, 0.17 and 0.25 d-1

, respectively), consistent with the observed invariant Kρ

values among the four cultures. Therefore, change in cellular size could not account for

the ~5-fold relative increase in 55

Fe uptake rate observed in the culture grown at the

highest dilution rate versus that grown at the lowest dilution rate.

Another possible explanation is that the starved cells grown under extreme Fe stress

require time to recover Fe uptake machinery before functioning optimally, whilst less

starved cells can immediately acquire Fe at optimal rates. As a result, within the

recovery period, the less starved cells would exhibit a higher Fe uptake rate (i.e. the Fe

uptake trend observed in this work), but beyond this recovery period, more starved

cells would exhibit greater Fe uptake rates, corresponding to the expected trend

suggested by others (Morel, 1987). However, the decrease in accumulated cellular Fe

with increasing dilution rate during both the linear Fe uptake period (0-4 h) and the

non-linear Fe uptake period (4-12 h) shown in Figure 6.8 implies that such an

explanation is unlikely.

Since the transport of Fe across the (cyto)plasmic membrane of cyanobacteria is

generally mediated by ATP-binding cassette (ABC) transporters (Andrews et al.,

2003), a more plausible reason is that decline in either the efficiency of energy-

dependent processes or of energy (i.e., ATP) production during photosynthesis causes

decreasing Fe uptake under extreme Fe stress. For serverely Fe-limited cells (those

grown at lower dilution rates in this study), it is likely that the photosynthetic capacity

and subsequent ATP production from the cyclic electron transport are minimal due to

very low Fe availability, resulting in these cells being unable to drive high Fe uptake

rates. This possibility is supported by results from short-term 14

C uptake assays using

the same steady-state cultures as those used for the short-term Fe uptake experiments.

As shown in Figure 6.10, the accumulation rate of radio-labeled carbonate in cells

decreased with decreasing dilution rate during an incubation period of 12 h, similarly

to the trend observed in the short-term 55

Fe accumulation studies. Therefore, a shortage

of resources necessary for Fe uptake such as an internal transporter or ATP may



122

account for the short-term Fe uptake trend in this study. Additionally, a physiological

trade-off may occur under Fe stress between the affinity for Fe at the cell surface (i.e.

the number of surface uptake sites or Fe channels) and the maximum rate at which Fe

can be assimilated (i.e. the number of internal transporters which assimilate Fe once it

is encounted) (Smith et al., 2009). When grown under Fe stress for a long period, cells

may acclimate by maintaining the number of surface uptake sites while decreasing the

number of internal enzymes available for Fe uptake. In the presence of any pulse of Fe,

cells with fewer internal transporters will likely exhibit lower Fe uptake rates relative

to non-stressed cells. This physiological acclimation strategy has been used to explain

the observed pattern of nitrate uptake by phytoplankton in the ocean (Smith et al.,

2009).


C uptake during batch short-term uptake assays using

cells obtained at steady-state from the chemostat cultures grown with [Fe]T = 20 nM in

the inflowing medium and dilution rates of 0.09 d-1


(squares),

0.17 d-1


(circles). Symbols represent the mean and error bars

the standard deviation from triplicate experiments.

0

20

40

60

80

100

120

140

0 2 4 6 8 10 12 14

Acc

um

ula

ted

14C

(fm

ol

cell

-1)

Time (hr)



123

6.3.6. Characteristics of Iron-limited Cultures of M. aeruginosa Grown

Continuously in Nutrient-replete Fraquil* Medium

The observation that Fe uptake rates were lower in the more Fe deficient cultures and

the supposition that this effect may be related to the overall health of the organism is

highly speculative and requires further investigation. It is likely that the insufficiency

of nutrients other than Fe in the culture medium (i.e., Fraquil*) may also contribute to

the weakening strength of M. aeruginosa cells. It is worth noting that, due to the

variable composition of freshwaters compared to a remarkably constant composition of

major ions in seawater, freshwater phytoplankton are usually adapted to their ambient

chemistry. As a result, one freshwater medium may support good growth of a

microorganism but not others and even minor changes of a medium composition such

as the [Na+]/[K

+] ratio may inhibit the growth of some freshwater microorganism

species (Andersen, 2005). In fact, although the concentration of trace metal ions in

Fraquil* designed to support optimal algal growth in freshwater is comparable to those

in the synthetic seawater Aquil* medium and about an order of magnitude higher than

those in the original Fraquil medium, the concentration of most major ions and trace

components in this medium are still significantly lower than that in the growth medium

BG-11 which is widely used for the culturing of freshwater, soil, thermal, and marine

cyanobacteria (Allen and Stanier, 1968, Morel et al., 1975, Morel et al., 1979, Price et

al., 1988, Andersen, 2005). For example, the concentrations of phosphate, magnesium,

manganese, copper and cobalt in BG-11 are about 17.5, 2, 15, 2 and 3 times,

respectively, higher than in Fraquil* and, particularly, nitrate and molybdenum in BG-

11 are about 2 orders of magnitude greater than in Fraquil* (~176 and 161 times,

respectively) (see Table 6.3). Thus, in this study, the composition of Fraquil* used in

our previous chemostat study were revised to find appropriate concentrations of the

culture medium composition that likely support optimal growth of M. aeruginosa.

Following the medium selection, the chosen-modified Fraquil* medium (referred to as

nutrient-replete Fraquil*) was used to revisit the behavior (including steady state cell

density, Fe cell quota and Fe uptake kinetics) of M. aeruginosa strain PCC7806 grown

continuously in the chemostat system. All the experimental methods used in this work

were identical to those used in our previous chemostat study, except nutrient-replete

Fraquil* was employed as the culture medium instead of Fraquil

*.



124

6.3.6.1. Selection of the Nutrient-replete Fraquil*

Medium

Fraquil* medium has been widely used for examining the effects of metals on growth,

nutrient uptake, photosynthetic activity, morphology, etc. of freshwater phytoplankton

(Rueter and Ades, 1987, Rueter, 1988, Gensemer, 1990, Gensemer et al., 1993, Fujii et

al., 2010a, Fujii et al., 2011a, Alexova et al., 2011) due to its chemically well-defined

composition which can facilitate studies of trace metal interactions with freshwater

phytoplankton. Therefore, the Fraquil* recipe was chosen as the basis of the new

culture medium (i.e., nutrient-replete Fraquil*) used in this batch-wise medium

selection study as well as our further chemostat studies on behavior of Fe-limited

cultures of M. aeruginosa grown continuously under nutrient-replete conditions.

However, the concentrations of certain nutrients in the original Fraquil* was adjusted

to levels which provide improved growth of M. aeruginosa compared to that in the

original Fraquil*.

In this medium selection study, the growth of M. aeruginosa in batch incubation

cultures was examined for various nutrient concentrations of Fraquil*, with particular

attention given to the variation of nitrate and molybdenum due to the exceptionally

high concentrations of these two nutrients in BG-11 relative to Fraquil* (~176 and 161

times higher, respectively), while the concentration of bio-available Fe was maintained

at a level sufficient to prevent the adverse effect of Fe deficiency on the growth of M.

aeruginosa. The batch-wise growth of M. aeruginosa in the control medium (i.e.,

Fraquil*) was compared to those in five different nutrient-replete Fraquil

* media,

defined here as Test 1, Test 2, Test 3, Test 4 and Test 5. These nutrient-replete Fraquil*

media were prepared by simultaneously increasing the concentrations of both nitrate

and molybdenum in Fraquil* by a factor of 10 (Test 1), 20 (Test 2), 40 (Test 3), 80

(Test 4), and 176 (for nitrate) and 161 (for molybdenum) (i.e., equivalent to those in

BG-11, Test 5). Meanwhile, the concentration of the other major and trace nutrients,

except for iron, was increased to the concentration of the corresponding nutrients in

BG-11 or held at the same concentration as in Fraquil* if the concentration of that

nutrient in BG-11 was lower than that in the original Fraquil* medium. The recipes of

these culture media are presented in details in Table 6.3. The concentration of Fe and

its synthetic organic chelator EDTA in all the studied culture media was fixed at 10

and 26 µM, respectively, to ensure a sufficient level of Fe available for growth of M.



125

aeruginosa as previously reported in Section 6.3.1 and elsewhere (Fujii et al., 2010a,

Dang et al., 2012).

Figure 6.11. Growth curves in batch cultures of M. aeruginosa at a constant total Fe

concentration in different modified Fraquil* growth media. Total Fe concentration and

its chelator EDTA were fixed at 10 and 26 µM while other media components were

varied as shown in Table 6.3. Symbols represent the mean and error bars represent the

standard deviation from duplicate incubations (filled diamonds: control (i.e., Fraquil*);

filled squares: Test 1; filled triangles: Test 2; open diamonds: Test 3; open squares:

Test 4 and open triangles: Test 5).

0

2

4

6

8

10

12

14

16

18

0 2 4 6 8 10 12 14 16 18 20 22

Cel

l d

ensi

ty (

10

9×

cell

L-1

)

Time (d)


126

Table 6.3. Compositions of the modified nutrient-replete Fraquil* media examined in this study.

Composition Ratio of

[nutrient]original BG-11/[nutrient]original Fraquil*

Ratio of [nutrient]examined Fraquil*/[nutrient]original Fraquil*

Control Test 1 Test 2 Test 3 Test 4 Test 5

Salt solutions

CaCl2.2H2O 0.9 1.0 1.0 1.0 1.0 1.0 1.0

MgSO4.7H2O 2.0 1.0 2.0 2.0 2.0 2.0 2.0

NaHCO3 0.4a 1.0 1.0 1.0 1.0 1.0 1.0

NaNO3 176 1.0 10 20 40 80 176

K2HPO4 17.5 1.0 17.5 17.5 17.5 17.5 17.5

HEPES N.A.b 1.0 1.0 1.0 1.0 1.0 1.0

Trace metal solutions

CuSO4.5H2O 2.0 1.0 2.0 2.0 2.0 2.0 2.0

CoCl2.6H2O 3.4 1.0 3.4 3.4 3.4 3.4 3.4

MnCl2.4H20 15.2 1.0 15.2 15.2 15.2 15.2 15.2

ZnSO4.7H2O 0.6 1.0 1.0 1.0 1.0 1.0 1.0

Na2SeO3 N.A. 1.0 1.0 1.0 1.0 1.0 1.0

Na2MoO4.2H20 161 1.0 10 20 40 80 161

Fe(III)-ligand solutions

Na2EDTA.2H2O N.A. 1.0 1.0 1.0 1.0 1.0 1.0

FeCl3.6H2O N.A. 1.0 1.0 1.0 1.0 1.0 1.0

Vitamin solutions

Thiamine.HCl N.A. 1.0 1.0 1.0 1.0 1.0 1.0

Biotin N.A. 1.0 1.0 1.0 1.0 1.0 1.0

Cyanocobalamin N.A. 1.0 1.0 1.0 1.0 1.0 1.0

a This value represents ratio of the concentration of Na2CO3 in BG-11 and the concentration of NaHCO3 in Fraquil

*.

b N.A means “not applicable”



127

As seen in Figure 6.11, the growth of M. aeruginosa significantly increased when

additional nutrients were added. The lowest specific growth rate of M. aeruginosa

(0.81 ± 0.02 d-1

) was found in the control medium (i.e., Fraquil*) while the growth

rates in the other culture media used were comparable and averaged 0.90 ± 0.01 d-1

.

The cell density of M. aeruginosa grown in the control medium reached the maximum

value after 14 incubation days while the M. aeruginosa cell density in the modified

Fraquil* media continued to increase over the 20-d incubation period. These results

clearly show the obvious deficiency of nutrients for optimal growth of M. aeruginosa

in the original Fraquil* medium used in our previous studies. The lower cell densities

after day 10 in Test 4 and Test 5 media compared to those in the other modified

Fraquil* media (i.e., Test 1-3) suggest that the excessive concentrations of nitrate and

molybdenum in these two media adversely affected the growth of M. aeruginosa.

Overall, among the culture medium examined here, Test 1, Test 2 and Test 3 culture

media apparently provided the best growth conditions for M. aeruginosa with growth

of this organism in these culture media relatively similar. Thus, either Test 1, Test 2 or

Test 3 culture media could be chosen as the nutrient-replete medium yielding optimal

growth of M. aeruginosa. In this study, the Test 2 medium (hereafter referred to as

nutrient-replete Fraquil*) was chosen as the nutrient-replete culture medium for further

studies of M. aeruginosa growth.

6.3.6.2. Growth Kinetics in Batch Culture

In order to predict the behaviour of the Fe-limited chemostat cultures of M. aeruginosa

PCC7806 grown in nutrient-replete Fraquil* and thereby to decide on the chemosat

operating parameters (i.e., total Fe in the inflowing nutrient-replete Fraquil* medium

and the critical dilution rate, etc.) for the continuous culturing system, batch-wise

growth of this organism in nutrient-replete Fraquil* was investigated over a range of Fe

concentrations ([Fe]T = 0.05-10 µM).


128

Figure 6.12. Growth curves in batch cultures of M. aeruginosa at various total Fe concentrations in the nutrient-replete Fraquil* medium (i.e., Test 2 medium).

Total Fe concentration was varied from 0.05 to 10 µM while concentration of EDTA was fixed at 26 µM. Symbols represent the mean and error bars represent

the standard deviation from duplicate incubations (filled diamonds: [Fe]T = 0.05 µM; filled squares: [Fe]T = 0.1 µM, filled triangles: [Fe]T = 0.2 µM, filled

circles: [Fe]T = 0.5 µM, open diamonds: [Fe]T = 1.0 µM, open squares: [Fe]T = 2.0 µM, open triangles: [Fe]T = 5.0 µM, and open circles: [Fe]T = 10 µM).

0

2

4

6

8

10

12

14

0 2 4 6 8 10 12 14 16 18 20 22

Cel

l d

ensi

ty (

10

9×

cell

L-1

)

Time (d)



129

Although specific growth rate of M. aeruginosa under various Fe conditions was

relatively comparable (with an average value of 0.19 ± 0.011 d-1

) during the late

exponential growth phase (i.e., 8-14 d), it increased significantly with increasing Fe

concentration in the early exponential phase (i.e., 2-6 d), therefore, resulting in a

substantial increase in cell density in batch cultures of M. aeruginosa under high Fe

conditions (Figure 6.12). The specific growth rates in the early exponential phase for

cultures grown in the ≤ 2 µM Fe nutrient-replete Fraquil* growth media range from

0.61 ± 0.011 to 0.85 ± 0.007 d-1

and are significantly (p < 0.0002 using a single-tailed

heteroscedastic t-test) lower than the comparable growth rates of cultures grown in 5

and 10 µM Fe nutrient-replete Fraquil* (0.90 ± 0.05 and 0.92 ± 0.013 d

-1, respectively).

Thus, ≤ 2 µM of initial Fe induced growth inhibition while 5-10 µM Fe ensured

optimal growth of M. aeruginosa in nutrient-replete Fraquil*. As expected, when

grown in the same Fe concentration under identical incubation conditions, the batch

incubation cultures of M. aeruginosa in nutrient-replete Fraquil* consistently exhibited

higher specific growth rates (0.61 ± 0.011; 0.71 ± 0.022; 0.84 ± 0.011; and 0.92 ±

0.013 d-1

for [Fe]T = 0.05; 0.1; 1 and 10 µM, respectively) compared to those (0.57 ±

0.022; 0.65 ± 0.005; 0.77 ± 0.05; and 0.79 ± 0.006 d-1

for [Fe]T = 0.05; 0.1; 1 and 10

µM, respectively) in Fraquil* (Dang et al., 2012). These results are consistent with our

previous finding that the low concentration of nutrients in Fraquil* was not enough to

support optimal growth of M. aeruginosa (see Section 6.3.6.1).

The values of four growth constants of M. aeruginosa in the batch culture grown in

nutrient-replete Fraquil*: maximum specific growth rate µmax, half-saturation constants

TSK (for total Fe) and '

SK (for Fe(II)’) and yield constant Y were estimated using

identical calculation methods employed for the batch cultures of M. aeruginosa grown

in Fraquil* as discussed previously in Section 6.3.1. As such, the maximum growth

rate and the half-saturation constant were estimated via nonlinear regression of the data

using the Monod equation (eq. 6.4) for cases where both total Fe and calculated

steady-state Fe(II)’ were treated as the appropriate substrate for growth, yielding two

half-saturation constant 'S

K of 3.11 ± 0.30 fM with respect to Fe(II)’ and

TSK of 23 ±

2.2 nM for total Fe, and the same value for the growth constant (µmax = 0.89 ± 0.09 d-

1). The theoretical data of specific growth rate calculated using eq. 6.4 based on these



130

growth values fitted the observed growth rates of M. aeruginosa very well with R2 of

0.93 (Figure 6.13). Meanwhile, the average yield constant Y of M. aeruginosa under

Fe limitation condition ([Fe]T of 0.01 and 0.02 µM) was determined to be 27 ± 0.74 ×

1016

cell (mol Fe)-1

by using eq. 6.6. Compared to the growth constants of M.

aeruginsoa obtained from the batch studies in Fraquil*, the values of the half-

saturation constants 'S

K and

TSK and the growth constant µmax of this organism under

nutrient replete condiitons were comparable while the yield constant Y was about 3-

fold higher which accounts for the substantial increase in cell density when nutrient-

replete Fraquil* was used as the growth medium of M. aeruginosa instead of Fraquil

*.


) and log concentration of

unchelated Fe(II)’ (M) in batch culture studies of M. aeruginosa grown in nutrient-

replete Fraquil* medium. Solid and dotted lines represent the regression line and 95%

confidential interval, respectively. Symbols indicate data for experimentally

determined growth rate under different degrees of Fe limitation.

R2 = 0.9276

0.50

0.55

0.60

0.65

0.70

0.75

0.80

0.85

0.90

0.95

1.00

-14.5 -14 -13.5 -13 -12.5 -12 -11.5

Sp

ecif

ic g

row

th r

ate

µ (

d-1

)

log([Fe(II)'] (M))



131

6.3.6.4. Comparison between the Behavior of M. aeruginosa Grown Continuously in

Nutrient-replete Fraquil* and in Original Fraquil

*

A summary of the growth constants in the batch cultures and the behaviour of the

chemostat cultures of M. aeruginosa grown in both Fraquil* and nutrient-replete

Fraquil* is presented in Table 6.4. Detailed description of the behaviour (including

steady state cell density, Fe cell quota, cell size and Fe uptake kinetics) of M.

aeruginosa strain PCC7806 grown continuously in the nutrient-replete Fraquil*

medium with Fe activity buffered by the organic ligand EDTA is given below.

a. Performance of chemostat system under Fe limitation

Since the ≤ 2 µM Fe cultures were inhibited by depletion of Fe available for growth, a

total Fe concentration of 100 nM in the nutrient-replete Fraquil* growth medium was

used in this study to ensure that steady state chemostat cultures of M aeruginosa were

maintained under Fe limitation. At this initial Fe concentration (i.e., 100 nM Fe) in the

inflowing growth medium, the corresponding critical dilution rate for complete wash-

out (Dc) was determined to be 0.72 d-1

. Hence, the continuous cultures were

maintained in 100 nM Fe nutrient-replete Fraquil* at dilution rates of 0.07, 0.15, 0.30

and 0.45 d-1

which were less than the critical dilution rate Dc for a period of 40 d

(Figure 6.14). M. aeruginosa cells grew continuously in 100 nM nutrient-replete

Fraquil* in a similar growth pattern as seen in the 20 nM Fe Fraquil

* chemostat

cultures (part B of Figure 6.5), i.e. after 2-d lag phase, cells density of the four dilution

rates increased exponentially and appeared to reach steady state concentrations after

day 25.



132

Figure 6.14. Growth of M. aeruginosa in the continuous culture system at different

dilution rates with total Fe concentrations in the inflowing nutrient-replete Fraquil*

medium [Fe]T = 100 nM, with dilution rates of 0.07 d-1


(squares),

0.30 d-1


(circles). Symbols represent the mean and error bars

the standard deviation from triplicate incubations. Dashed lines represent the 95%

confidence interval at steady-state.

Similar to the previous chemostat studies which used Fraquil* as growth medium, the

theoretical steady state cell and Fe concentrations in the nutrient-replete chemostat

cultures as a function of dilution rate were estimated by substituting the values for the

growth constants: 'S

K , TS

K , µmax and Y into eqs 6.15 and 6.15. These data are plotted

in Figure 6.15 together with observed steady state cell densities at different dilution

rates. For comparative purposes, measured and theoretical steady state cell densities

obtained previously from the batch studies in Fraquil* are also presented in this graph.

In all cases, the observed steady state cell densities at the lowest dilution rate were

always greater than the corresponding theoretical values. This is probably because at

the lowest dilution rate (i.e., the lowest inflow rate) the evaporation rate of water inside

the culture vessel was likely comparable to the flow-rate of the inflowing medium

which leads to accumulation of nutrients in the culture vessel, hence, increase in the

8.8

9.0

9.2

9.4

9.6

9.8

10.0

10.2

10.4

10.6

10.8

0 5 10 15 20 25 30 35 40 45

Loga

rith

m o

f ce

ll d

en

sity

(ce

ll L

-1)

Time (d)



133

cell density. Overall, the theoretical steady state cell densities as function of dilution

rate fitted well the observed steady state cell densities of M. aeruginsoa, indicating that

the modified chemostat theory proposed in Section 6.2.7 (Dang et al., 2012) was

applicable to predicting the steady state cell densities in the Fe-limited chemostate

cultures of M. aeruginosa grown in both Fraquil* and nutrient-replete Fraquil

* media.


concentration in continuous cultures of M. aeruginosa as a function of dilution rate

with different total Fe concentrations in the two inflowing media: Fraquil* (20 nM and

50 nM) and nutrient-replete Fraquil* (100nM). Symbols represent data for steady-state

cell density in Fraquil* medium with total Fe of 20 nM (triangles), 50 nM (squares)

and 100 nM (diamonds). Dotted lines are the theoretical values of steady-state cell

density calculated from eq. 6.16 with growth parameters estimated from batch culture

studies in Fraquil* (µmax = 0.80 ± 0.03 d

-1, '

SK = 3.6 ± 0.32 fM with respect to Fe(II)’,

TSK = 26 ± 2.3 nM with respect to total Fe, and Y = 8.1 ± 0.21 × 10

16 cell (mol Fe)

-1),

while bold lines indicate the theoretical steady-state cell density estimated with

parameters obtained from batch culture studies in nutrient-replete Fraquil* (µmax = 0.89

± 0.03 d-1

, 'S

K = 3.1 ± 0.30 fM, TS

K = 23 ± 2.2 nM, and Y = 2.7 ± 0.74 × 1017

cell

0

50

100

150

200

250

300

350

400

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

[Fe(

II)'

]ss

(fM

)

or

[Cel

l]ss

(10

8ce

ll L

-1)

Dilution rate (d-1)



134

(mol Fe-1

)). Chained and dashed lines indicate predicted steady-state unchelated Fe(II)’

concentrations estimated using parameters from batch culture studies in Fraquil* and

nutrient-replete Fraquil*, respectively.

b. Cellular Fe quota and cell size under iron limitation

The intracellular Fe content of cells in the steady state chemostat cultures grown in 100

nM Fe nutrient-replete Fraquil* at different dilution rates was examined by using acid

digestion combined with spectrophotometry as described in Section 2.5.1. The steady-

state cellular Fe quotas at dilution rates of 0.07, 0.15, 0.30 and 0.45 d-1

were

determined to be 17.5 ± 2.1; 26.5 ± 0.7; 33.7 ± 2.4 and 48 ± 1.4 amol cell-1

,

respectively which are about one-order of magnitude higher than those observed in

Fraquil*. Similar to the data for the chemostat cultures of M. aeruginosa grown in

Fraquil*, the specific growth rates (i.e., the dilution rates) of this organism in nutrient-

replete Fraquil* increased hyperbolically with increasing steady state Fe quota (Figure

6.16) which is in accordance with the empirical Droop equation (eq. 6.19). These

results are consistent with other observations for cyanobacteria (Gotham and Rhee,

1981b, Ahlgren, 1985, Olsen, 1989) and eukaryotic phytoplankta (Goldman and

Mccarthy, 1978, Sunda and Huntsman, 1985, Sunda and Huntsman, 1986, Harrison

and Morel, 1986).

In terms of cell size, there was no significant change in the cell diameter with

increasing dilution rates observed in this study. However, M. aeruginosa cells were

slightly larger when grown continuously in nutrient-replete Fraquil* with an average

diameter of 4.01 ± 0.09 µm than in Fraquil* with an average diameter of 3.87 ± 0.15

µm. The higher nutrient levels (including Fe) in nutrient-replete Fraquil* compared to

those in Fraquil* possibly account for the increase in both the Fe quotas and the cell

size of the nutrient-replete chemostat cultures in this work.



135

Figure 6.16. Relationship between the cellular Fe quota (Q) and the specific growth

rate (µ) for Fe-limited M. aeruginosa under steady-state conditions in continuous

cultures. The system was operated at four different dilution rates (0.07, 0.15, 0.30 and

0.45 d-1

) and fed with nutrient-replete Fraquil* medium containing 100 nM Fe. The

solid line represents the theoretical curve calculated from the Droop equation using the

obtained estimated values of µ’max = 0.69 ± 0.05 d-1

and Qmin = 18 ± 2.6 amol cell-1

.

Symbols represent the mean from triplicate measurements.

c. Fe uptake kinetics and cellular response to Fe limitation in chemostat

Short-term 55

Fe uptake assays in the presence of light were undertaken using Fe-

limited cells from the steady-state chemostat cultures of M. aeruginosa grown in

100nM Fe nutrient-replete Fraquil* at dilution rates of 0.07; 0.15; 0.30 and 0.45 d

-1.

Processing steps in these short-term Fe uptake experiments were identical to those

described in Section 6.2.5 in which the examined cells were incubated in Fraquil*

medium (Dang et al., 2012). Since the photo-produced unchelated ferrous iron (i.e.,

Fe(II)’) has previously been shown to be the major form of Fe acquired by M.

aeruginosa in Fraquil* buffered by the synthetic organic ligand EDTA (Fujii et al.,

2011a, Dang et al., 2012), the Fe uptake rate of each steady-state culture of M.

aeruginosa was investigated over a range of photoproduced unchelated Fe(II)’

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

0 20 40 60 80 100

Sp

eci

fic

gro

wth

ra

te µ

(d

-1)

Cellular Fe quota (amol cell-1)

Critical dilution rate Dc = 0.72 d-1



136

concentrations by varying the EDTA concentration from 5 to 200 µM while

maintaining 55

Fe concentration at a constant value of 200 nM. As seen in Figure 6.17,

intracellular 55

Fe accumulated linearly (R2 = 0.991 ÷ 0.998) over the first 4 h during

incubation in the light followed by non-linear increase of the intracellular 55

Fe after 4-

8 h incubation, indicating that uptake reached saturated levels during the later stages of

the experiment.

Insight into Fe uptake kinetics was undertaken by applying Eadie-Hofstee linear

transformation to the measured 55

Fe uptake rates in each culture as shown in Figure

6.18, provided that the short-term uptake rates for M. aeruginosa followed classical

Michaelis-Menten kinetics (see eq. 6.21). Linear regression analysis (R2 = 0.91-0.99)

yielded comparable half-saturation constants for uptake with an average value of Kρ =

45 ± 1.9 fM but significantly different maximum uptake rates (ρmax of 1005 ± 46; 887

± 79; 684 ± 87 and 483 ± 35 zmol Fe cell-1

h-1

) for the chemostat cultures grown at

dilutions rates of 0.07; 0.15; 0.30 and 0.45 d-1

, respectively). The invariant nature of

the half-saturation constant Kρ at different dilution rates and the significant

dependence of the maximum uptake rate on the degree of Fe stress of cells are

consistent with previous classical findings where Kρ was generally observed to be

constant while ρmax was observed to increase with the degree of nutrient starvation of

cells (Gotham and Rhee, 1981a, Gotham and Rhee, 1981b, Olsen, 1989, Harrison and

Morel, 1986, Morel, 1987).



137


Fe uptake during batch short-term Fe uptake assays

using cells obtained at steady-state from the chemostat cultures grown with [Fe]T =

100 nM in the inflowing nutrient-replete Fraquil* medium and dilution rates of 0.07 d

-1


(squares), 0.30 d-1


(circles). In the short-

term uptake assay, each culture was incubated in nutrient-replete Fraquil* with 20 µM

EDTA and 200 nM radiolabeled 55

Fe. Symbols represent the mean and error bars

represent the standard deviation from triplicate experiments. The continuous lines were

obtained by linear regression of data collected within 4 h (represented by closed

symbols) for each culture.

The short-term Fe uptake data of M. aeruginsoa grown under nutrient replete

conditions were compared with those reported previously for Fe-limited chemostat

cultures of this organism grown in Fraquil* (Dang et al., 2012). As observed

previously, the half saturation-constant for uptake, Kρ, was again found to be

independent of the degree of Fe stress. However, an obvious difference between the

two studies is that the value of Kρ for the Fe-limited chemostat cultures grown under

nutrient-replete conditions (i.e., in nutrient-replete Fraquil*) was ~2.5 times higher than

that obtained in nutrient-deplete conditions (i.e., in Fraquil*), suggesting that when

R² = 0.9982

R² = 0.996

R² = 0.9971

R² = 0.9912

0

500

1000

1500

2000

2500

3000

3500

4000

0 2 4 6 8 10

Acc

um

ula

ted

55F

e (z

mol

cell

-1)

Time (hr)



138

grown under greater nutrient deficiency (including Fe), Fe-limited M. aeruginosa cells

become more competitive for photoproduced unchelated Fe(II)’. The most significant

difference in responses of cells with regard to Fe uptake between this study and the

previous chemostat study in which Fraquil* was used as growth medium was the Fe

uptake trend over the degree of Fe stress. In terms of Fe-limited cells grown in

nutrient-replete Fraquil*, Fe-starved cells exhibited much higher maximum uptake

rates as expected. In contrast, for Fe-limited cells grown in the nutrient-insufficient

growth medium (i.e., Fraquil*), the maximum short-term uptake was also dependent on

the degree of Fe stress but surprisingly decreased with increase in extent of Fe

deficiency.

Overall, when grown under nutrient-replete conditions, the kinetics of steady state

growth and short-term Fe uptake by Fe-limited cells in chemostat cultures of M.

aeruginosa strictly followed the classical hyperbolic expressions for steady state

growth (i.e., the Droop equation) and uptake (Michaelis-Menten equation), implying

that the nutrient-replete Fraquil* medium provided sufficient nutrient conditions for

optimal growth of M. aeruginosa cells. There were no abnormal responses of cells in

Fe-limited chemostat with regard to Fe uptake rate when nutrient-replete Fraquil* was

used as the inflowing growth medium, meaning that the M. aeruginosa cells and their

Fe uptake machinery functioned properly under these conditions. In comparison, under

nutrient deficient conditions (i.e., Fraquil*), although the steady state long-term uptake

kinetics of M. aeruginosa cells were not affected (i.e., the empirical Droop equation

was still followed), the short-term Fe uptake was found to surprisingly increase with

decreasing degree of Fe-limitation. It would appear that this difference in kinetics of

steady state growth and short-term Fe uptake between the cells in the Fe-limited

chemostat cultures grown in Fraquil* and those in nutrient-replete Fraquil

* stemmed

from the difference in the nutrient levels of the two culture media used.



139

Figure 6.18. Eadie-Hofstee plots demonstrating the linear relationship between the

short-term 55

Fe uptake rate (ρFe) and the ratio ρFe/[Fe(II)’] for cultures of M.

aeruginosa. Linear regression analysis yielded comparable half-saturation constants

for Fe uptake (Kρ = 45 ± 1.9 fM, as Fe(II)’) but different maximum specific uptake

rates (ρmax of 1.0 ± 0.046, 0.89 ± 0.079, 0.67 ± 0.087 and 0.48 ± 0.035 amol cell-1

hr-1



(squares), 0.30 d-1


(circles), respectively). Lines for 95% confidential intervals

were omitted for clarity.

The results from both the batch culture and chemostat culture studies in this work

suggest that the Fraquil* medium often used for study of trace metal interactions with

freshwater phytoplankton by several workers (Rueter and Ades, 1987, Rueter, 1988,

Gensemer, 1990, Gensemer et al., 1993, Fujii et al., 2010a, Fujii et al., 2011a, Alexova

et al., 2011, Dang et al., 2012) was insufficient to provide optimal growth of M.

aeruginosa. It would seem reasonable to conclude that the “unhealthy” organisms

present under nutrient deficient conditions were more prone to Fe deficiency and were

increasingly incapable of acquiring Fe as the extent of Fe deficiency increased. In

y = -42.7x + 1,005

R² = 0.99

y = -45.3x + 887

R² = 0.96

y = -47.2x + 684

R² = 0.91

y = -46.4x + 483

R² = 0.97

0

100

200

300

400

500

600

700

800

900

0 5 10 15 20 25

55Fe

up

tak

e r

ate

(zm

ol

cell

-1h

r-1)

55Fe uptake rate/[Fe(II)']ss (µmol cell-1 hr-1 M-1)



140

comparison, organisms grown with sufficient major nutrients appear able to respond to

increasing Fe deficiency by increasing their short term Fe uptake rate.

6.4. CONCLUSIONS

In this chapter we have shown that a continuous culturing system made of metal-free

material provides a valuable tool to investigate the cellular responses of Fe-limited M.

aeruginosa PCC7806 in both nutrient-insufficient and nutrient-replete Fraquil* media.

The system was successfully operated to produce steady-state cultures with different

cell densities and different cellular properties. In both nutrient-insufficient and

nutrient-replete cases, the cellular response to steady-state Fe limitation in the

chemostat system followed the Droop equation, i.e. cellular Fe quotas increased with

increasing Fe availability. Under Fe stress, cells of steady-state cultures of M.

aeruginosa regulated their short-term Fe uptake by varying their uptake capacity ρmax,

but not their affinity for Fe (i.e. the half saturation constant Kρ). Under nutrient-

insufficient case, Fe uptake data from this study show that Fe-limited M. aeruginosa

cells grown under severe Fe stress (i.e., lower dilution rates) are likely unable to

synthesize sufficient resources (such as internal transporters and/or ATP) required for

Fe uptake, and therefore exhibit lower Fe uptake ability compared to cells grown under

conditions in which Fe is more available (i.e., at higher dilution rates). In contrast, the

relationship between Fe uptake capacity and the degree of Fe-limitation reverted to

that expected with the short-term Fe uptake rate increasing with the degree of Fe stress

when Fe-limited M. aeruginosa cells are grown under nutrient-replete conditions

where M. aeruginosa cells are “healthy” and functioning normally.


141

Table 6.4. Summary of the growth constants in the batch cultures and the behaviors of the Fe-limited chemostat cultures at different dilution

rates of M. aeruginosa grown in both Fraquil* and nutrient-replete Fraquil

*

Parameters Growth medium

Unit Fraquil

* Nutrient-replete Fraquil

*

A. Batch cultures

(i) Fe limitation condition ≤ 0.1 ≤ 2.0 µM

(ii) Growth constants

- µmax 0.80 ± 0.09 0.89 ± 0.03 d-1

- TS

K 26 ± 2.3 23 ± 2.2 nM

- 'S

K 3.6 ± 0.32 3.1 ± 0.30 fM

- Y 8.1 ± 0.21 × 1016

27 ± 0.74 × 1016

cell (mol Fe)-1

B. Continuous cultures

(i) Operating parameters

- [Fe]initial 20 100 nM

- Critical dilution rate 0.34 0.72 d-1

- Operating dilution rate 0.09; 0.14; 0.17 and 0.25 0.07; 0.15; 0.30; 0.45 d-1

(ii) Cellular Fe quota Follow Droop equation (expected) Follow Droop equation (expected)

- Qmin 1.2 ± 0.24 18 ± 2.6 amol cell-1

- µ'max 0.37 ± 0.04 0.69 ± 0.05 d

-1

(iii) Cell diameter 3.87 ± 0.15 4.01 ± 0.09 µm

(v) Fe uptake kinetics Maximum uptake rate increases with

increasing dilution rates (unexpected)

Maximum uptake rate decreases with

increasing dilution rates (expected)

- ρmax 0.27 ± 0.030, 0.72 ± 0.060, 0.95 ± 0.080

and 1.0 ± 0.090 (varied)

1.0 ± 0.046; 0.89 ± 0.079; 0.67 ± 0.087

and 0.48 ± 0.035 (varied)

amol Fe cell-1

hr-1

- Kρ ~18 ± 2.2 (constant) ~45 ± 1.9 (constant) fM

142

CHAPTER 7

CONCLUSIONS AND

RECOMMENDATIONS

Chapter 7. Conclusions and Recommendations

143

7.1. CONCLUSIONS

The results of studies described in this thesis provide new insights into the Fe uptake

kinetics of the bloom-forming freshwater cyanobacterium M. aeruginosa, particularly

in regard to: (i) effect of light on iron uptake by M. aeruginosa in the presence of a

single metal-chelator, ethylenediaminetetraacetic acid (EDTA); (ii) intracellular Fe

transport processes of M. aeruginosa in a chemically well-defined culture medium

(Fraquil*) buffered by EDTA; (iii) iron uptake kinetics by M. aeruginosa in the

presence of the natural organic ligand, Suwannee River Fulvic Acid (SRFA); and (iv)

characteristics of M. aeruginosa cells grown in iron-limited continuous culture. The

main conclusions obtained from this research are summarized below.

Chapter 3. Effect of Light on Iron Uptake by the Freshwater

Cyanobacterium Microcystis aeruginosa

The aim of the studies described in this chapter was to elucidate the effect of light on the

iron uptake by M. aeruginosa in a chemically well-defined culture medium (Fraquil*)

in the presence of a single metal chelator, ethylenediaminetetraacetic acid (EDTA).

Major findings of this aspect of the study are provided below.

(i) Visible light was observed to induce reductive dissociation of organically

complexed Fe and dramatically increase the short-term uptake rate of

radiolabeled Fe by M. aeruginosa PCC7806 in Fraquil* medium at pH 8 buffered

by EDTA. Only wavelengths < 500 nm activated Fe uptake indicating that Fe

photochemistry rather than biological factors is responsible for the facilitated

uptake.

(ii) The measured rate of photochemical Fe(II) production combined with a

significant decrease in 55

Fe uptake rate in the presence of ferrozine (a strong

ferrous chelator) confirmed that photo-generated unchelated Fe(II) was the major

form of Fe acquired by M. aeruginosa under the conditions examined.

(iii) Mathematical modeling based on unchelated Fe(II) uptake by concentration


144

gradient dependent passive diffusion of Fe(II) through the non-specific

transmembrane channels (porins) could account for the magnitude of Fe uptake

and a variety of other observed effects.

(iv) Steady-state uptake rates indicated that M. aeruginosa acquires Fe

predominantly during the light cycle. This study confirms that Fe photochemistry

has a dominant impact on Fe acquisition and growth by M. aeruginosa in EDTA-

buffered culture medium.

Chapter 4. Kinetics of Extracellular Iron Transport to Periplasmic

and Cytoplasmic compartments of the Freshwater Cyanobaterium


The aim of the studies described in this chapter was to gain insight into the kinetics of

extracellular Fe transport to periplasmic and cytoplasmic compartments for strains

PCC7806 and 7005 of M. aeruginosa. Major findings of this aspect of the study are

provided below.

(i) A negligibly small amount of chelated 55

Fe accumulated in the periplasm of

plasmolysed cells, confirming that only unchelated Fe is capable of crossing the

outer membrane, most likely by a diffusive process.

(ii) The observed Monod-type relationship between cytoplasmic 55

Fe accumulation

rates and steady-state concentrations of unchelated Fe in the periplasm and

extracellular environment suggests that translocation of Fe into the cytoplasm

involves complexation of Fe by a limited number of Fe-binding sites in the

periplasm followed by subsequent transport into the cytoplasm, possibly via

energy-dependent plasma-membrane Fe transporters.

(iii) Further experimental evidences suggest that redox state of intracellular Fe is

relevant to the transport of periplasmic Fe into the cytoplasm with Fe redox state

possibly regulated by oxidoreductase enzymes such as multi-copper oxidase

resident in the periplasm.


145

Chapter 5. Iron Uptake Kinetics by a Freshwater Cyanobacterium

Microcystis aeruginosa in the Presence of Suwannee River Fulvic Acid

The aim of the studies described in this chapter was to determine the effect of the

presence of the natural organic compound, Suwannee River fulvic acid (SRFA), on

iron uptake kinetics by M. aeruginosa. Major findings of this aspect of the study are

provided below.

(i) A kinetic model was developed which satisfactorily described the results

obtained and which provided insight into the significant reactions and

processes involved in cellular Fe uptake in the presence of SRFA.

(ii) During incubations under dark and visible light, 55

Fe uptake rates similarly

decreased in an exponential manner as SRFA concentration increased, even

though photochemical reduction rate of Fe(III) bound to SRFA (FeIII

SRFA)

was one order of magnitude greater than non-photochemical reduction. The

similarity of 55

Fe uptake rate under the dark and light was accounted for by

relatively rapid oxidation of photo-generated Fe(II) prior to dissociation of

the complex forming unchelated Fe(II), resulting in the comparable Fe(II)

and Fe(III) availability in the dark and light.

(iii) Model prediction suggested that the decreased trend of 55

Fe uptake at

higher SRFA concentrations is due to limited availability of unchelated Fe

for cellular uptake.

(iv) The inhibitory effect of strong Fe(II) chelator (ferrozine) on 55

Fe uptake

indicates that approximately a half of total Fe uptake was accounted for by

uptake of unchelated Fe(II) likely produced via reductive dissociation of

FeIII

SRFA.

(v) The findings in this study suggest that Fe uptake is generally near

saturation for the fulvic acid concentrations typically encountered in natural

waters (e.g., < ~10 mg.L-1

).


146

Chapter 6. Characteristics of the Freshwater Cyanobacterium

Microcystis aeruginosa Grown in Iron-limited Continuous Culture

The aim of studies described in this chapter was to determine the characteristics of M.

aeruginosa grown in iron-limited continuous culture. Major findings of this aspect of

the study are provided below.

(i) A kinetic model describing Fe transformations and biological uptake was

developed and applied to determination of the biologically available form of Fe

(i.e., unchelated ferrous iron) that is produced by photoreductive dissociation of

the ferric EDTA complex.

(ii) Prediction by chemostat theory modified to account for the light-mediated

formation of bioavailable Fe was in good agreement with growth characteristics

of M. aeruginosa under Fe limitation.

(iii) In both nutrient-insufficient and nutrient-replete cases, cellular Fe quota

increased with increasing dilution rate in a manner consistent with Droop theory.

Short-term Fe uptake assays using cells maintained at steady-state indicated that

M. aeruginosa cells vary their maximum Fe uptake rate (ρmax) depending on the

degree of Fe stress.

(iv) Under nutrient-insufficient conditions, the rate of Fe uptake was lower for cells

grown under conditions of lower Fe availability (i.e., lower dilution rate)

suggesting that cells in the continuous cultures adjusted to Fe-limitation by

decreasing ρmax whilst maintaining a constant affinity for Fe. This result implies

that Fe-limited M. aeruginosa cells grown under severe Fe stress and other

nutrients insufficiency are likely unable to synthesize sufficient resources

required for Fe uptake.

(v) In contrast, under nutrient-replete conditions M. aeruginosa cells are “healthy”

and functioning normally. As such, the relationship between Fe uptake capacity

and the degree of Fe-limitation reverted to that expected with the short-term Fe

uptake rate increasing with the degree of Fe stress.


147

7.2. IMPLICATIONS OF THE FINDINGS

7.2.1. With Regard to Knowledge of Fe Transformation and Uptake

Kinetics by Freshwater Cyanobacteria in Natural Waters

This work provides compelling evidence of the connection between the thermal and

photochemical transformation of Fe species and biological uptake where unchelated

Fe(II)’ formed by photochemical and thermal reduction of Fe(III) species is an

important substrate for growth and represents one of the important growth factors

controlling Fe uptake by phytoplankton. Also, insights gained into the kinetics of

transport of extracellular Fe to periplasmic and cytoplasmic spaces in the freshwater

cyanobacterium Microcystis aeruginosa should be applicable to other cyanobateria

given their similarity in cell structure.

7.2.2. With Regard to Application of the Continuous Culturing

System for Study of Trace Metal Interactions with Freshwater

Phytoplankton

In this thesis, the chemostat system was successfully developed for maintaining

steady-state continuous cultures of M. aeruginosa at different dilution rates under Fe

limited conditions. The performance of the system was checked and the results

indicate that this system was effectively used to grow the Fe-limited chemosat cultures

under steady state conditions for more than one month (at least 40 d) without any

biological and trace metal contamination. In addition, the compactivity of the system

makes it possible to be incorporated into an incubator which allows experimenters to

easily control not only the chemical growth conditions such as pH, oxygen,

concentrations of nutrients, etc. but also the physical conditions such as light intensity

and temperature. These advantages of this continuous culture system make it a suitable

apparatus for investigating growth and behaviours of microorganism in steady-state

chemosat cultures under micro-nutrients limitation.

Also, the modified chemostat theory for Fe-limited phytoplankton growth proposed in

this thesis described well the observed steady state cell density of M. aeruginosa strain


148

PCC7806 grown continuously in growth medium (Fraquil* or nutrient-replete Fraquil

*)

with Fe activity buffered by the organic ligand EDTA. It can be applicable to trace

metal-limited chemostat studies where the concentration of limiting nutrient is

considered as the concentration of the bio-available forms for uptake (i.e., photo-

produced unchelated Fe(II)’ in this thesis) rather than the total concentration of

nutrient used in macronutrient-limited chemostat studies. Hence, it can be adopted as a

new model of chemostat functioning under trace metal-limited condition and used to

predict the steady-state concentrations of microorganism being cultured and limiting

substrate in the culture vessel for any value of the dilution rate and the concentration of

the inflowing limiting substrate.

7.2.3. With Regard to Knowledge of the Composition of the Growth

Medium for Freshwater Phytoplankton

The results obtained from both batch and continuous culture studies in this thesis

indicate that Fraquil* contained insufficient nutrients to support the optimal growth of

the freshwater cyanobacterium M. aeruginosa.. In comparison, the nutrient-replete

Fraquil* medium used in the studies described in this thesis was able to sustain optimal

growth of M. aeruginosa and can be recommended as a suitable medium for future

studies on the effects of metals on growth, nutrient uptakes, photosynthetic activity, or

morphology, etc. of this organism. While testing would be required to ensure that

nutrient concentrations are adequate, this modified Fraquil* medium could also be used

for the study of trace metal interactions with other freshwater phytoplankton.

7.3. RECOMMENDATIONS FOR FUTURE WORK

In order to improve our understanding of Fe uptake kinetics by M. aeruginosa in

natural waters, additional studies are required. Specific recommendations for future

work are noted below.

(i) It should be noted that all the M. aeruginosa cells in both batch and continuous

cultures used in this research was grown in the growth medium (Fraquil* or

nutrient-replete Fraquil*) with the free iron activity buffered by the well-

characterized model ligand EDTA before being harvested for further various


149

experiments of interest. In order to reflect more closely the natural aquatic

systems, study of the kinetics of Fe acquisition using Fe-limited M. aeruginosa

cells grown batch-wise or continuously in the presence of iron either complexed

to natural organic matter (NOM) rather than the model organic ligand EDTA or

present as amorphous ferric oxyhydroxide is considered necessary.

(ii) As mentioned previously in Section 5.4, Chapter 5, the nature of the light

provided by the incubator fluorescent light used in this research and natural

sunlight is acknowledged to be substantially different (e.g., light intensity of 157

µmol m-2

s-1

compared to 2,000 µmol m-2

s-1

), leading to significant differences

in photo-reduction rate of Fe(III) to Fe(II) in laboratory compared to natural

systems. As such, there is justification for further studies to investigate the Fe

uptake kinetics of M. aeruginosa using a solar simulator.

(iii) Fe redox reactions inside cells or near the cell surface are known to be important

in Fe acquisition by phytoplankton. However, for simplicity, the redox state of Fe

in the periplasmic and cytoplasmic spaces of M. aeruginosa cells has not been

considered in the model for translocation of Fe from the external medium to the

intracellular compartments developed here. Therefore, further investigation on

the effect of Fe speciation in the periplasm and cytoplasm of M. aeruginosa cells

on intracellular Fe transport is required.

References

150

REFERENCES

AHLGREN, G. 1985. Growth of Microcystis wesenbergii in batch and chemostat

cultures. Verhandlungen der Internationale Vereinigung fur Theoretische und

Angewandte Limnologie, 22, 2813 - 2820.

ALEXOVA, R., FUJII, M., BIRCH, D., CHENG, J., WAITE, T. D., FERRARI, B. C.

& NEILAN, B. A. 2011. Iron uptake and toxin synthesis in the bloom-forming

Microcystis aeruginosa under iron limitation. Environmental Microbiology, 13,

1064-1077.

ALLEN, M. M. & STANIER, R. Y. 1968. Growth and division of some unicellular

blue-green algae. Journal of General Microbiology, 51, 199-202.

ANDERSEN, R. A. E. 2005. Algal culturing techniques. Elsevier Academic Press:

Burlington, MA, 596.

ANDERSON, M. A. & MOREL, F. M. M. 1982. The influence of aqueous iron

chemistry on the uptake of iron by the coastal diatom Thalassiosiraweissflogii.

Limnology and Oceanography, 27, 789-813.

ANDREWS, S. C., ROBINSON, A. K. & RODRIGUEZ-QUINONES, F. 2003.

Bacterial iron homeostasis. FEMS Microbiol Rev, 27, 215-37.

BADARAU, A., FIRBANK, S. J., WALDRON, K. J., YANAGISAWA, S.,

ROBINSON, N. J., BANFIELD, M. J. & DENNISON, C. 2008. FutA2 is a

ferric binding protein from Synechocystis PCC 6803. Journal of Biological

Chemistry, 283, 12520-12527.

BARBEAU, K., RUE, E. L., BRULAND, K. W. & BUTLER, A. 2001. Photochemical

cycling of iron in the surface ocean mediated by microbial iron(III)-binding

ligands. Nature, 413, 409-413.

References

151

BIELSKI, B. H. J., CABELLI, D. E., ARUDI, R. L. & ROSS, A. B. 1985. Reactivity

of HO2/O2- radicals in aqueous solution. Journal of Physical and Chemical

Reference Data, 14, 1041-1100.

BLIGH, M. W. & WAITE, T. D. 2010. Role of heterogeneous precipitation in

cetermining the nature of products formed on oxidation of Fe(II) in seawater

containing natural organic matter. Environmental Science & Technology, 44,

6667-6673.

BOYD, P. W., JICKELLS, T., LAW, C. S., BLAIN, S., BOYLE, E. A., BUESSELER,

K. O., COALE, K. H., CULLEN, J. J., DE BAAR, H. J. W., FOLLOWS, M.,

HARVEY, M., LANCELOT, C., LEVASSEUR, M., OWENS, N. P. J.,

POLLARD, R., RIVKIN, R. B., SARMIENTO, J., SCHOEMANN, V.,

SMETACEK, V., TAKEDA, S., TSUDA, A., TURNER, S. & WATSON, A. J.

2007. Mesoscale iron enrichment experiments 1993-2005: Synthesis and future

directions. Science, 315, 612-617.

BRULAND, K. W. & RUE, E. L. 2001. Analytical methods for the determination of

concentrations and speciation of iron. In D. R. Turner and K. A. Hunter (eds.),

The Biogeochemistry of Iron in Seawater. J. Wiley: Chichester; New York,

255-290.

BUFFLE, J., WILKINSON, K. J. & VAN LEEUWEN, H. P. 2009. Chemodynamics

and bioavailability in natural waters. Environmental Science & Technology, 43,

7170-7174.

BURMASTER, D. E. 1979. Unsteady continuous culture of phosphate-limited

Monochrysislutheri Droop - experimental and theoreticalanalysis. Journal of

Experimental Marine Biology and Ecology, 39, 167-186.

CAPERON, J. & MEYER, J. 1972a. Nitrogen-limited growth of marine phytoplankton

.1. Changes in population characteristics with steady-state growth-rate. Deep-

Sea Research, 19, 601-618.

References

152

CAPERON, J. & MEYER, J. 1972b. Nitrogen-limited growth of marine phytoplankton

.2. Uptake kinetics and their role in nutrient limited growth of phytoplankton.

Deep-Sea Research, 19, 619-632.

CARMICHAEL, W. W. & FALCONER, I. R. 1993. Diseases related to freshwater

blue-green algal toxins, and control measures. In I. R. Falconer (ed.), Algal

Toxins in Seafood and Drinking Water. Academic Press, Ltd.; London,

England, 197-209.

COLLINS, C. M., ANDERSON, A. M. & WEGER, H. G. 2001. Iron acquisition by

the green alga Selenastrum minutum: Growth in iron-limited chemostats and

effects of chelator stability constant. Archiv Fur Hydrobiologie, 151, 283-299.

CRICHTON, R. R. 2009. Inorganic biochemistry of iron metabolism: from molecular

mechanisms to clinical consequences. 3rd ed.; John Wiley & Sons: Chichester,

West Sussex, U.K. ; Hoboken.

DANG, T. C., FUJII, M., ROSE, A. L., BLIGH, M. & WAITE, T. D. 2012.

Characteristics of the freshwater cyanobacterium Microcystis aeruginosa

grown in iron-limited continuous culture. Applied and Environmental

Microbiology, 78, 1574-1583.

DAVIES, A. G. 1970. Iron, chelation and growth of marine phytoplankton .1. Growth

kinetics and chlorophyll production in cultures of euryhaline flagellate

Dunallielatertiolecta under iron-limiting conditions. Journal of the Marine

Biological Association of the United Kingdom, 50, 65-86.

DECC 2005. Audit of the Sydney drinking water catchment. Report to the Minister for

the Environment, NSW State Government, Department of Environment and

Conservation (NSW).

DROOP, M. R. 1973. Some thoughts on nutrient limitation in algae. Journal of

Phycology, 9, 264-272.

References

153

ESTEP, M., ARMSTRONG, J. E. & VANBAALEN, C. 1975. Evidence for

occurrence of specific iron (III)-binding compounds in near-shore marine

ecosystems. Applied Microbiology, 30, 186-188.

FAN, S. M. 2008. Photochemical and biochemical controls on reactive oxygen and

iron speciation in the pelagic surface ocean. Marine Chemistry, 109, 152-164.

FARALDO-GOMEZ, J. D. & SANSOM, M. S. P. 2003. Acquisition of siderophores

in Gram-negative bacteria. Nature Reviews Molecular Cell Biology, 4, 105-

116.

FAUST, B. C. & ZEPP, R. G. 1993. Photochemistry of aqueous iron(III)

polycarboxylate complexes - Roles in the chemistry of atmospheric and surface

waters. Environmental Science & Technology, 27, 2517-2522.

FRANGEUL, L., QUILLARDET, P., CASTETS, A. M., HUMBERT, J. F.,

MATTHIJS, H. C. P., CORTEZ, D., TOLONEN, A., ZHANG, C. C.,

GRIBALDO, S., KEHR, J. C., ZILLIGES, Y., ZIEMERT, N., BECKER, S.,

TALLA, E., LATIFI, A., BILLAULT, A., LEPELLETIER, A., DITTMANN,

E., BOUCHIER, C. & DE MARSAC, N. T. 2008. Highly plastic genome of

Microcystis aeruginosa PCC 7806, a ubiquitous toxic freshwater

cyanobacterium. Bmc Genomics, 9, 274.

FUHS, G. W. 1969. Phosphorus content and rate of growth in diatoms Cyclotella nana

and Thalassiosira fluviatilis. Journal of Phycology, 5, 312-321.

FUJII, M., DANG, T. C., ROSE, A. L., OMURA, T. & WAITE, T. D. 2011a. Effect of

light on iron uptake by the freshwater cyanobacterium Microcystis aeruginosa.

Environmental Science & Technology, 45, 1391-1398.

FUJII, M., ITO, H., ROSE, A. L., WAITE, T. D. & OMURA, T. 2008a. Superoxide-

mediated Fe(II) formation from organically complexed Fe(III) in coastal

waters. Geochimica et Cosmochimica Acta, 72, 6079-6089.

References

154

FUJII, M., ITO, H., ROSE, A. L., WAITE, T. D. & OMURA, T. 2008b.

Transformation dynamics and reactivity of dissolved and colloidal iron in

coastal waters. Marine Chemistry, 110, 165-175.

FUJII, M., ROSE, A. L., OMURA, T. & WAITE, T. D. 2010a. Effect of Fe(II) and

Fe(III) transformation kinetics on iron acquisition by a toxic strain of

Microcystis aeruginosa. Environmental Science & Technology, 44, 1980-1986.

FUJII, M., ROSE, A. L. & WAITE, T. D. 2011b. Iron uptake by toxic and nontoxic

strains of Microcystis aeruginosa. Applied and Environmental Microbiology,

77, 7068-7071.

FUJII, M., ROSE, A. L., WAITE, T. D. & OMURA, T. 2010b. Oxygen and

superoxide-mediated redox kinetics of iron complexed by humic substances in

coastal seawater. Environmental Science & Technology, 44, 9337-9342.

FULDA, S., MIKKAT, S., SCHRODER, W. & HAGEMANN, M. 1999. Isolation of

salt-induced periplasmic proteins from Synechocystis sp. strain PCC 6803.

Archives of Microbiology, 171, 214-217.

GARG, S., ROSE, A. L., GODRANT, A. & WAITE, T. D. 2007. Iron uptake by the

ichthyotoxic Chattonella marina (Raphidophyceae): Impact of superoxide

generation. Journal of Phycology, 43, 978-991.

GARG, S., ROSE, A. L. & WAITE, T. D. 2011. Photochemical production of

superoxide and hydrogen peroxide from natural organic matter. Geochimica et

Cosmochimica Acta, 75, 4310-4320.

GARG, S., ROSE, A. L. & WAITE, T. D. 2012. Pathways contributing to the

formation and decay of ferrous iron in sunlit natural waters. In P. G. Tratnyek, ,

T. J. Grundl, & S. B. Haderlein (eds.), Aquatic Redox Chemistry. Washington:

American Chemical Society, 153-176

References

155

GENSEMER, R. W. 1990. Role of aluminum and growth-rate on changes in cell-size

and silica content of silica-limited populations of Asterionellaralfsii var

americana (Bacillariophyceae). Journal of Phycology, 26, 250-258.

GENSEMER, R. W., SMITH, R. E. H. & DUTHIE, H. C. 1993. Comparative effects

of pH and aluminum on silica-limited growth and nutrient-uptake in

Asterionellaralfsii var americana (Bacillariophyceae). Journal of Phycology,

29, 36-44.

GERHARDT, P. & DREW, S. W. 1994. Liquid culture. Methods for general and

molecular bacteriology. P. Gerhardt, R. G. E. Murray, W. A. Wood and N. R.

Kreig. Washington D. C., American Society for Microbiology, 224-247.

GOLDMAN, J. C. & MCCARTHY, J. J. 1978. Steady-state growth and ammonium

uptake of a fast-growing marine diatom. Limnology and Oceanography, 23,

695-703.

GOTHAM, I. J. & RHEE, G. Y. 1981a. Comparative kinetic-studies of nitrate-limited

growth and nitrate uptake in phytoplankton in continuous culture. Journal of

Phycology, 17, 309-314.

GOTHAM, I. J. & RHEE, G. Y. 1981b. Comparative kinetic-studies of phosphate-

limited growth and phosphate-uptake in phytoplankton in continuous culture.

Journal of Phycology, 17, 257-265.

GRESS, C. D., TREBLE, R. G., MATZ, C. J. & WEGER, H. G. 2004. Biological

availability of iron to the freshwater cyanobacterium Anabaena flos-aquae.


HARRISON, G. I. & MOREL, F. M. M. 1986. Response of the marine diatom

Thalassiosira weissflogii to iron stress. Limnology and Oceanography, 31, 989-

997.

HASSLER, C. S., HAVENS, S. M., BULLERJAHN, G. S., MCKAY, R. M. L. &

TWISS, M. R. 2009. An evaluation of iron bioavailability and speciation in

References

156

western Lake Superior with the use of combined physical, chemical, and

biological assessment. Limnology and Oceanography, 54, 987-1001.

HASSLER, C. S., SCHOEMANN, V., NICHOLS, C. M., BUTLER, E. C. V. &

BOYD, P. W. 2011. Saccharides enhance iron bioavailability to Southern

Ocean phytoplankton. Proceedings of the National Academy of Sciences of the

United States of America, 108, 1076-1081.

HASSLER, C. S. & TWISS, M. R. 2006. Bioavailability of iron sensed by a

phytoplanktonic Fe-bioreporter. Environmental Science & Technology, 40,

2544-51.

HENLEY, W. J. & YIN, Y. 1998. Growth and photosynthesis of marine

Synechococcus (Cyanophyceae) under iron stress. Journal of Phycology, 34,

94-103.

HER, N., AMY, G., FOSS, D. & CHO, J. W. 2002. Variations of molecular weight

estimation by HP-size exclusion chromatography with UVA versus online

DOC detection. Environmental Science & Technology, 36, 3393-3399.

HERBERT, D., ELSWORTH, R. & TELLING, R. C. 1956. The continuous culture of

bacteria - a theoretical and experimental study. Journal of General


HOICZYK, E. & HANSEL, A. 2000. Cyanobacterial cell walls: News from an

unusual prokaryotic envelope. Journal of Bacteriology, 182, 1191-1199.

HOPKINSON, B. M. & BARBEAU, K. A. 2012. Iron transporters in marine

prokaryotic genomes and metagenomes. Environmental Microbiology, 14, 114-

28.

HOSKISSON, P. A. & HOBBS, G. 2005. Continuous culture - making a comeback?


References

157

HUDSON, R. J. M. & MOREL, F. M. M. 1990. Iron transport in marine-

phytoplankton - kinetics of cellular and medium coordination reactions.


HUTCHINS, D. A. & BRULAND, K. W. 1998. Iron-limited diatom growth and Si : N

uptake ratios in a coastal upwelling regime. Nature, 393, 561-564.

IMAI, A., FUKUSHIMA, T. & MATSUSHIGE, K. 1999. Effects of iron limitation

and aquatic humic substances on the growth of Microcystis aeruginosa.

Canadian Journal of Fisheries and Aquatic Sciences, 56, 1929-1937.

ITO, H., FUJII, M., MASAGO, Y., YOSHIMURA, C., WAITE, T. D. & OMURA, T.

2011. Mechanism and kinetics of ligand exchange between ferric citrate and

desferrioxamine B. Journal of Physical Chemistry A, 115, 5371-5379.

ITO, Y. & BUTLER, A. 2005. Structure of synechobactins, new siderophores of the

marine cyanobacterium Synechococcus sp PCC 7002. Limnology and

Oceanography, 50, 1918-1923.

JONES, A. M., PHAM, A. N., COLLINS, R. N. & WAITE, T. D. 2009. Dissociation

kinetics of Fe(III)- and Al(III)-natural organic matter complexes at pH 6.0 and

8.0 and 25 degrees C. Geochimica et Cosmochimica Acta, 73, 2875-2887.

JONES, C. M. & NIEDERWEIS, M. 2010. Role of porins in iron uptake by

Mycobacterium smegmatis. Journal of Bacteriology, 192, 6411-6417.

KANEKO, T., NAKAJIMA, N., OKAMOTO, S., SUZUKI, I., TANABE, Y.,

TAMAOKI, M., NAKAMURA, Y., KASAI, F., WATANABE, A.,

KAWASHIMA, K., KISHIDA, Y., ONO, A., SHIMIZU, Y., TAKAHASHI,

C., MINAMI, C., FUJISHIRO, T., KOHARA, M., KATOH, M., NAKAZAKI,

N., NAKAYAMA, S., YAMADA, M., TABATAI, S. & WATANABE, M. M.

2007. Complete genomic structure of the bloom-forming toxic cyanobacterium

Microcystis aeruginosa NIES-843. DNA Research, 14, 247-256.

References

158

KARI, F. G., HILGER, S. & CANONICA, S. 1995. Determination of the reaction

quantum yield for the photochemical degradation of Fe(III)-EDTA -

Implications for the environmental fate of EDTA in surface waters.


KATOH, H., HAGINO, N., GROSSMAN, A. R. & OGAWA, T. 2001. Genes

essential to iron transport in the cyanobacterium Synechocystis sp strain PCC

6803. Journal of Bacteriology, 183, 2779-2784.

KERRY, A., LAUDENBACH, D. E. & TRICK, C. G. 1988. Influence of iron

limitation and nitrogen-source on growth and siderophore production by

cyanobacteria. Journal of Phycology, 24, 566-571.

KING, D. W., ALDRICH, R. A. & CHARNECKI, S. E. 1993. Photochemical redox

cycling of iron in NaCl solutions. Marine Chemistry, 44, 105-120.

KING, D. W., LOUNSBURY, H. A. & MILLERO, F. J. 1995. Rates and mechanism

of Fe(II) oxidation at nanomolar total iron concentrations. Environmental

Science & Technology, 29, 818-824.

KIRK, J. T. O. 1994. Light and photosynthesis in aquatic ecosystems. Cambridge

University Press: New York, 2nd ed.

KORSHUNOV, S. & IMLAY, J. A. 2006. Detection and quantification of superoxide

formed within the periplasm of Escherichia coli. Journal of Bacteriology, 188,

6326-6334.

KOSAKOWSKA, A., NEDZI, M. & PEMPKOWIAK, J. 2007. Responses of the toxic

cyanobacterium Microcystis aeruginosa to iron and humic substances. Plant

Physiology and Biochemistry, 45, 365-370.

KRANZLER, C., LIS, H., SHAKED, Y. & KEREN, N. 2011. The role of reduction in

iron uptake processes in a unicellular, planktonic cyanobacterium.

Environmental Microbiology, 13, 2990-2999.

References

159

KUMA, K., NISHIOKA, J. & MATSUNAGA, K. 1996. Controls on iron(III)

hydroxide solubility in seawater: The influence of pH and natural organic

chelators. Limnology and Oceanography, 41, 396-407.

LIU, X. W. & MILLERO, F. J. 1999. The solubility of iron hydroxide in sodium

chloride solutions. Geochimica Et Cosmochimica Acta, 63, 3487-3497.

LIU, X. W. & MILLERO, F. J. 2002. The solubility of iron in seawater. Marine

Chemistry, 77, 43-54.

MALDONADO, M. T., ALLEN, A. E., CHONG, J. S., LIN, K., LEUS, D.,

KARPENKO, N. & HARRIS, S. L. 2006. Copper-dependent iron transport in

coastal and oceanic diatoms. Limnology and Oceanography, 51, 1729-1743.

MALDONADO, M. T. & PRICE, N. M. 2001. Reduction and transport of organically

bound iron by Thalassiosira oceanica (Bacillariophyceae). Journal of

Phycology, 37, 298-309.

MALDONADO, M. T., STRZEPEK, R. F., SANDER, S. & BOYD, P. W. 2005.

Acquisition of iron bound to strong organic complexes, with different Fe

binding groups and photochemical reactivities, by plankton communities in Fe-

limited subantarctic waters. Global Biogeochemical Cycles, 19, GB4S23,

doi:10.1029/2005GB002481.

MARGERUM, D. W., CAYLEY, G. R., WEATHERBURN, D. C. & PAGENKOPF,

G. K. 1978. Kinetics and mechanisms of complex formation and ligand

exchange. In: MARTELL, A. (ed.) Coordination Chemistry. Washington DC:

American Chemical Society.

MARTIN, J. H. & FITZWATER, S. E. 1988. Iron-deficiency limits phytoplankton

growth in the northeast Pacific Subarctic. Nature, 331, 341-343.

MARTINEZ, J. S., HAYGOOD, M. G. & BUTLER, A. 2001. Identification of a

natural desferrioxamine siderophore produced by a marine bacterium.


References

160

MCCARTHY, J. J. 1981. The kinetics of nutrient utilization. Canadian Bulletin of

Fisheries and Aquatic Sciences, 210, 211-233.

MIDDLEMISS, J. K., ANDERSON, A. M., STRATILO, C. W. & WEGER, H. G.

2001. Oxygen consumption associated with ferric reductase activity and iron

uptake by iron-limited cells of Chlorella kessleri (Chlorophyceae). Journal of

Phycology, 37, 393-399.

MILLER, C. J., ROSE, A. L. & WAITE, T. D. 2009. Impact of natural organic matter

on H2O2-mediated oxidation of Fe(II) in a simulated freshwater system.

Geochimica Et Cosmochimica Acta, 73, 2758-2768.

MILLERO, F. J. & SOTOLONGO, S. 1989. The oxidation of Fe(II) with H2O2 in

seawater. Geochimica Et Cosmochimica Acta, 53, 1867-1873.

MONOD, J. 1950. La technique de culture continue theorie et applications. Annales

De L Institut Pasteur, 79, 390-410.

MOREL, F. M. M. 1987. Kinetics of nutrient-uptake and growth in phytoplankton.


MOREL, F. M. M., KUSTKA, A. B. & SHAKED, Y. 2008. The role of unchelated Fe

in the iron nutrition of phytoplankton. Limnology and Oceanography, 53, 400-

404.

MOREL, F. M. M., RUETER, J. G., ANDERSON, D. M. & GUILLARD, R. R. L.

1979. Aquil - Chemically defined phytoplankton culture-medium for trace-

metal studies. Journal of Phycology, 15, 135-141.

MOREL, F. M. M., WESTALL, J. C., REUTER, J. G. & CHAPLICK, J. P. 1975.

Description of the algal growth media “Aquil” and “Fraquil.” Technical Report

16. Water Quality Laboratory, Ralph Parsons Laboratory for Water Resources

and Hydrodynamics, Massachusetts Institute of Technology, Cambridge, p.p.

33.

References

161

MURPHY, T. P., LEAN, D. R. S. & NALEWAJKO, C. 1976. Blue-green-algae - their

excretion of iron-selective chelators enables them to dominate other algae.

Science, 192, 900-902.

NAGAI, T., IMAI, A., MATSUSHIGE, K. & FUKUSHIMA, T. 2006. Effect of iron

complexation with dissolved organic matter on the growth of cyanobacteria in

a eutrophic lake. Aquatic Microbial Ecology, 44, 231-239.

NEILANDS, J. B. 1995. Siderophores - Structure and function of microbial iron

transport compounds. Journal of Biological Chemistry, 270, 26723-26726.

NEU, H. C. & HEPPEL, L. A. 1965. Release of enzymes from Escherichia coli by

osmotic shock and during formation of spheroplasts. Journal of Biological

Chemistry, 240, 3685-3692.

NICOLAISEN, K., MOSLAVAC, S., SAMBORSKI, A., VALDEBENITO, M.,

HANTKE, K., MALDENER, I., MURO-PASTOR, A. M., FLORES, E. &

SCHLEIFF, E. 2008. Alr0397 is an outer membrane transporter for the

siderophore Schizokinen in Anabaena sp strain PCC 7120. Journal of

Bacteriology, 190, 7500-7507.

NIKAIDO, H. 1976. Outer membrane of Salmonella typhimurium. Transmembrane

diffusion of some hydrophobic substances. Biochimica et Biophysica Acta, 433,

118-132.

NIKAIDO, H. 1979. Permeability of the outer-membrane of bacteria. Angewandte

Chemie-International Edition in English, 18, 337-350.

NIKAIDO, H. 2003. Molecular basis of bacterial outer membrane permeability

revisited. Microbiology and Molecular Biology Reviews, 67, 593-656.

NOVICK, A. & SZILARD, L. 1950. Description of the chemostat. Science, 112, 715-

716.

References

162

OLSEN, Y. 1989. Evaluation of competitive ability of Staurastrum leutkemuellerii

(Chlorophyceae) and Microcystis aeruginosa (Cyanophyceae) under P

limitation. Journal of Phycology, 25, 486-499.

PAGES, J. M., JAMES, C. E. & WINTERHALTER, M. 2008. The porin and the

permeating antibiotic: a selective diffusion barrier in Gram-negative bacteria.

Nature Reviews Microbiology, 6, 893-903.

PHAM, A. N. & WAITE, T. D. 2008a. Oxygenation of Fe(II) in natural waters

revisited: Kinetic modeling approaches, rate constant estimation and the

importance of various reaction pathways. Geochimica et Cosmochimica Acta,

72, 3616-3630.

PHAM, A. N. & WAITE, T. D. 2008b. Oxygenation of Fe(II) in the presence of citrate

in aqueous solutions at pH 6.0-8.0 and 25 degrees C: Interpretation from an

Fe(II)/citrate speciation perspective. Journal of Physical Chemistry A, 112,

643-651.

PRICE, N. M., HARRISON, G. I., HERING, J. G., HUDSON, R. J., NIREL, P. M. V.,

PALENIK, B., & MOREL, F. M. M. :. 1988/1989. Preparation and chemistry

of the artificial algal culture medium Aquil. Biological Oceanography, 6, 443–

461.

PULLIN, M. J. & CABANISS, S. E. 2003. The effects of pH, ionic strength, and iron-

fulvic acid interactions on the kinetics of nonphotochemical iron

transformations. II. The kinetics of thermal reduction. Geochimica et


ROSE, A. H. 1982. Advances in microbial physiology, Academic Press.

ROSE, A. L. 2012. The influence of extracellular superoxide on iron redox chemistry

and bioavailability to aquatic microorganisms. Frontiers in Microbiology, 3,

124.

References

163

ROSE, A. L., SALMON, T. P., LUKONDEH, T., NEILAN, B. A. & WAITE, T. D.

2005. Use of superoxide as an electron shuttle for iron acquisition by the

marine cyanobacterium Lyngbya majuscula. Environmental Science &

Technology, 39, 3708-3715.

ROSE, A. L. & WAITE, D. 2006. Role of superoxide in the photochemical reduction

of iron in seawater. Geochimica et Cosmochimica Acta, 70, 3869-3882.

ROSE, A. L. & WAITE, T. D. 2002. Kinetic model for Fe(II) oxidation in seawater in

the absence and presence of natural organic matter. Environmental Science &

Technology, 36, 433-444.

ROSE, A. L. & WAITE, T. D. 2003a. Effect of dissolved natural organic matter on the

kinetics of ferrous iron oxygenation in seawater. Environmental Science &

Technology, 37, 4877-4886.

ROSE, A. L. & WAITE, T. D. 2003b. Kinetics of iron complexation by dissolved

natural organic matter in coastal waters. Marine Chemistry, 84, 85-103.

ROSE, A. L. & WAITE, T. D. 2003c. Predicting iron speciation in coastal waters from

the kinetics of sunlight-mediated iron redox cycling. Aquatic Sciences, 65, 375-

383.

ROSE, A. L. & WAITE, T. D. 2005. Reduction of organically complexed ferric iron

by superoxide in a simulated natural water. Environmental Science &

Technology, 39, 2645-2650.

RUETER, J. G. 1988. Iron stimulation of photosynthesis and nitrogen-fixation in

Anabaena-7120 and Trichodesmium (Cyanophyceae). Journal of Phycology,

24, 249-254.

RUETER, J. G. & ADES, D. R. 1987. The role of iron nutrition in photosynthesis and

nitrogen assimilation in Scenedesmus quadricauda (Chlorophyceae). Journal of

Phycology, 23, 452-457.

References

164

RUSH, J. D. & BIELSKI, B. H. J. 1985. Pulse radiolytic studies of the reactions of

HO2/O2- with Fe(II)/Fe(III) ions - the reactivity of HO2/O2

- with ferric ions and

its implication on the occurrence of the Haber-weiss reaction. Journal of

Physical Chemistry, 89, 5062-5066.

SALMON, T. P., ROSE, A. L., NEILAN, B. A. & WAITE, T. D. 2006. The FeL

model of iron acquisition: Nondissociative reduction of ferric complexes in the

marine environment. Limnology and Oceanography, 51, 1744-1754.

SCHLEIFF, E., NICOLAISEN, K., MOSLAVAC, S., SAMBORSKI, A.,

VALDEBENITO, M., HANTKE, K., MALDENER, I., MURO-PASTOR, A.

M. & FLORES, E. 2008. Alr0397 is an outer membrane transporter for the

siderophore schizokinen in Anabaena sp strain PCC 7120. Journal of

Bacteriology, 190, 7500-7507.

SEVILLA, E., MARTIN-LUNA, B., VELA, L., BES, M. T., FILLAT, M. F. &

PELEATO, M. L. 2008. Iron availability affects mcyD expression and

microcystin-LR synthesis in Microcystis aeruginosa PCC7806. Environ

Microbiol, 10, 2476-2483.

SHAKED, Y., KUSTKA, A. B. & MOREL, F. M. M. 2005. A general kinetic model

for iron acquisition by eukaryotic phytoplankton. Limnology and


SHAKED, Y., KUSTKA, A. B., MOREL, F. M. M. & EREL, Y. 2004. Simultaneous

determination of iron reduction and uptake by phytoplankton. Limnology and

Oceanography-Methods, 2, 137-145.

SIMPSON, F. B. & NEILANDS, J. B. 1976. Siderochromes in cyanophyceae -

Isolation and characterization of schizokinen from Anabaena sp. Journal of

Phycology, 12, 44-48.

SIVONEN, K. & JONES, G. 1999. Cyanobacterial toxins. In Toxic cyanobacteria in

water, A guide to their public health consequences, monitoring and

References

165

management. . Chorus, I., and Bartram, J. (eds). London, U.K.: Spon Press,

41-111.

SMITH, S. L., YAMANAKA, Y., PAHLOW, M. & OSCHLIES, A. 2009. Optimal

uptake kinetics: physiological acclimation explains the pattern of nitrate uptake

by phytoplankton in the ocean. Marine Ecology-Progress Series, 384, 1-12.

SONIER, M. B., CONTRERAS, D. A., TREBLE, R. G. & WEGER, H. G. 2011. Two

distinct pathways for iron acquisition by iron-limited cyanobacterial cells:

evidence from experiments using siderophores and synthetic chelators. Botany,

(submitted).

SONIER, M. B. & WEGER, H. G. 2010. Plasma membrane ferric reductase activity of

iron-limited algal cells is inhibited by ferric chelators. Biometals, 23, 1029-

1042.

SORIADENGG, S. & HORSTMANN, U. 1995. Ferrioxamine-B and Ferrioxamine-E

as iron sources for the marine diatom Phaeodactylum Tricornutum. Marine

Ecology-Progress Series, 127, 269-277.

SPENCE, E., SARCINA, M., RAY, N., MOLLER, S. G., MULLINEAUX, C. W. &

ROBINSON, C. 2003. Membrane-specific targeting of green fluorescent

protein by the Tat pathway in the cyanobacterium Synechocystis PCC6803.

Molecular Microbiology, 48, 1481-1489.

STOOKEY, L. L. 1970. Ferrozine - a new spectrophotometric reagent for iron.

Analytical Chemistry, 42, 779-781.

STRAUS, N. A. 1994. Heterocyst metabolism and development. In Bryant, D.A. (ed.),

The Molecular Biology of Cyanobacteria. Dordrecht, the Netherlands: Kluwer

Academic Publishers, 731-750.

SUNDA, W. & HUNTSMAN, S. 2003. Effect of pH, light, and temperature on Fe-

EDTA chelation and Fe hydrolysis in seawater. Marine Chemistry, 84, 35-47.

References

166

SUNDA, W. G. 2001. Bioavailability and bioaccumulation of iron in the sea. J. Wiley:

Chichester; New York, 41-84.

SUNDA, W. G. & HUNTSMAN, S. A. 1985. Regulation of cellular manganese and

manganese transport rates in the unicellular alga Chlamydomonas. Limnology

and Oceanography, 30, 71-80.

SUNDA, W. G. & HUNTSMAN, S. A. 1986. Relationships among growth-rate,

cellular manganese concentrations and manganese transport kinetics in

estuarine and oceanic species of the diatom Thalassiosira. Journal of

Phycology, 22, 259-270.

TANG, D. G. & MOREL, F. M. M. 2006. Distinguishing between cellular and Fe-

oxide-associated trace elements in phytoplankton. Marine Chemistry, 98, 18-

30.

TEMPEST, D. W. 1969. The continuous cultivation of micro-organisms. I. Theory of

the chemostat. Methods in Microbiology, 2, 259-276.

THOMPSEN, J. C. & MOTTOLA, H. A. 1984. Kinetics of the complexation of

iron(II) with ferrozine. Analytical Chemistry, 56, 755-757.

TILMAN, D. & KILHAM, S. S. 1976. Phosphate and silicate growth and uptake

kinetics of the diatoms Asterionella formosa and Cyclotella meneghiniana in

batch and semi-continuous cultures. Journal of Phycology, 12, 375-383.

TIPPING, E. 2002. Cation binding by humic substances, Cambridge University Press,

UK.

TOLLE, J., MICHEL, K. P., KRUIP, J., KAHMANN, U., PREISFELD, A. &

PISTORIUS, E. K. 2002. Localization and function of the IdiA homologue

Slr1295 in the cyanobacterium Synechocystis sp strain PCC 6803.


References

167

TOVAR-SANCHEZ, A., SANUDO-WILHELMY, S. A., GARCIA-VARGAS, M.,

WEAVER, R. S., POPELS, L. C. & HUTCHINS, D. A. 2003. A trace metal

clean reagent to remove surface-bound iron from marine phytoplankton .

Marine Chemistry, 82, 91-99.

VELAYUDHAN, J., HUGHES, N. J., MCCOLM, A. A., BAGSHAW, J., CLAYTON,

C. L., ANDREWS, S. C. & KELLY, D. J. 2000. Iron acquisition and virulence

in Helicobacter pylori: a major role for FeoB, a high-affinity ferrous iron

transporter. Molecular Microbiology, 37, 274-286.

VIOLLIER, E., INGLETT, P. W., HUNTER, K., ROYCHOUDHURY, A. N. & VAN

CAPPELLEN, P. 2000. The ferrozine method revisited: Fe(II)/Fe(III)

determination in natural waters. Applied Geochemistry, 15, 785-790.

VOELKER, B. M., VERMILYEA, A. W. & HANSARD, S. P. 2010. Dark production

of hydrogen peroxide in the Gulf of Alaska. Limnology and Oceanography, 55,

580-588.

VRASPIR, J. M. & BUTLER, A. 2009. Chemistry of marine ligands and siderophores.

Annual Review of Marine Science, 1, 43-63.

WAITE, T. D. & MOREL, F. M. M. 1984. Photoreductive dissolution of colloidal iron

oxides in natural waters. Environmental Science & Technology, 18, 860-868.

WAITE, T. D., SZYMCZAK, R., ESPEY, Q. I. & FURNAS, M. J. 1995. Diel

variations in iron speciation in northern Australian shelf waters. Marine

Chemistry, 50, 79-91.

WALDRON, K. J., TOTTEY, S., YANAGISAWA, S., DENNISON, C. &

ROBINSON, N. J. 2007. A periplasmic iron-binding protein contributes toward

inward copper supply. Journal of Biological Chemistry, 282, 3837-3846.

WEGER, H. G. 1999. Ferric and cupric reductase activities in the green alga

Chlamydomonas reinhardtii: experiments using iron-limited chemostats.

Planta, 207, 377-384.

References

168

WEGER, H. G. & ESPIE, G. S. 2000. Ferric reduction by iron-limited

Chlamydomonas cells interacts with both photosynthesis and respiration.

Planta, 210, 775-781.

WEGER, H. G., LAM, J., WIRTZ, N. L., WALKER, C. N. & TREBLE, R. G. 2009.

High stability ferric chelates result in decreased iron uptake by the green alga

Chlorella kessleri owing to decreased ferric reductase activity and chelation of

ferrous iron. Botany-Botanique, 87, 922-931.

WEGER, H. G., MATZ, C. J., MAGNUS, R. S., WALKER, C. N., FINK, M. B. &

TREBLE, R. G. 2006. Differences between two green algae in biological

availability of iron bound to strong chelators. Canadian Journal of Botany-

Revue Canadienne De Botanique, 84, 400-411.

WEGER, H. G., MIDDLEMISS, J. K. & PETTERSON, C. D. 2002. Ferric chelate

reductase activity as affected by the iron-limited growth rate in four species of

unicellular green algae (Chlorophyta). Journal of Phycology, 38, 513-519.

WELKER, M. & VON DOHREN, H. 2006. Cyanobacterial peptides - Nature's own

combinatorial biosynthesis. Fems Microbiology Reviews, 30, 530-563.

WIEDNER, C., VISSER, P. M., FASTNER, J., METCALF, J. S., CODD, G. A. &

MUR, L. R. 2003. Effects of light on the microcystin content of Microcystis

strain PCC 7806. Applied and Environmental Microbiology, 69, 1475-1481.

WILHELM, S. W. 1995. Ecology of iron-limited cyanobacteria: A review of

physiological responses and implications for aquatic systems. Aquatic

Microbial Ecology, 9, 295-303.

WILHELM, S. W. & TRICK, C. G. 1994. Iron-limited growth of cyanobacteria -

Multiple siderophore production is a common response. Limnology and


References

169

WILHELM, S. W. & TRICK, C. G. 1995. Physiological profiles of Synechococcus

(Cyanophyceae) in iron-limiting continuous cultures. Journal of Phycology, 31,

79-85.

WILLIAMS, F. M. 1965. Population growth and regulation in continuous culture algae

Doctoral dissertation, Yale University, New Haven, Connecticut, 119.

WIRTZ, N. L., TREBLE, R. G. & WEGER, H. G. 2010. Siderophore-independent iron

uptake by iron-limited cells of the cyanobacterium Anabaena flos-aquae.


XING, W., HUANG, W. M., LI, D. H. & LIU, Y. D. 2007. Effects of iron on growth,

pigment content, photosystem II efficiency, and siderophores production of

Microcystis aeruginosa and Microcystis wesenbergii. Current Microbiology,

55, 94-98.

XUE, X. P., COLLINS, C. M. & WEGER, H. G. 1998. The energetics of extracellular

Fe(III) reduction by iron-limited Chlamydomonas reinhardtii (Chlorophyta).


ZUO, Y. G. & HOIGNE, J. 1992. Formation of hydrogen peroxide and depletion of

oxalic acid in atmospheric water by photolysis of Iron(III) oxalato complexes.


170

APPENDIX 1

SUPPLEMENTAL MATERIAL FOR

CHAPTER 3 - EFFECT OF LIGHT ON

IRON UPTAKE BY THE FRESHWATER


AERUGINOSA

Appendix 1. Supplemental Material for Chapter 3

171

A1.1. FEII

FZ3 FORMATION FROM FEIII

EDTA IN THE

LIGHT AND DARK

Primary kinetic data for FeIIFZ3 formation measured in the light and dark are shown in

parts A-D of Figure A1.1. FeIIFZ3 concentration substantially increased with time in

the light, while the time-dependent increase in FeIIFZ3 concentration was small or

negligible in the dark. Although FeIIFZ3 formation was unnoticeable at low Fe

concentration (1 µM), the linear increase in FeIIFZ3 concentration with time (R

2 =

0.99) was observed at a high Fe concentration (10 µM) possibly due to the active

reduction of Fe(III) species by FZ (Shaked et al., 2004) . In the high Fe concentration

system, therefore, the photo-produced FeIIFZ3 concentration was correctly calculated

by subtracting the FeIIFZ3 concentration produced in the dark from that produced in

the light at each incubation time.

In the FeIII

EDTA system, the ligand-to-metal charge transfer (LMCT) reaction will

ultimately produce unchelated Fe(II) (Fe(II)') and photo-oxidized EDTA (EDTAox). In

the presence of a high concentration of FZ, this strong Fe(II) complexing agent would

be expected to outcompete EDTA for complexation of liberated Fe(II)' and would also

be expected to prevent the oxygenation of Fe(II)' with subsequent formation of

FeIII

EDTA. This assumption was verified by kinetic calculations using relevant rate

constants as discussed below. Thus, the system can be simply described by the

following two reactions:

FeIIIEDTA + hν

khv*

→ Fe(II)' + EDTAox

(A1-1)

3

IIFZFe 3FZ Fe(II)' FZf →+ −k (A1-2)

where k*hv and kf-FZ are rate constants for photo-reductive dissociation of Fe

IIIEDTA

and for Fe(II) complexation by FZ, respectively. The value of kf-FZ has been previously

determined to be 3.1 × 1011

M-3

s-1

in 0.1 M NaClO4 (Thompsen and Mottola, 1984)

and 2.0 × 1011

M-3

s-1

in seawater (Lin and Kester, 1992). The effect of solution pH on

kf-FZ is insignificant in the range of pH 3 to 8. Since the ionic strength was closer to


172

that of this work, the value of 3.1 × 1011

M-3

s-1

was chosen. At the FZ concentration

employed in this work (1 mM), the complexation reaction of Fe(II)' by FZ is very fast

(kf-FZ[FZ]3 = 310 s

-1) and outcompetes any other competing reactions (i.e., oxidation by

oxygen, kox[O2] = 0.0021 s-1

, and re-complexation of Fe(II)' by EDTA, kf-EDTA[EDTA]

= 52 s-1

, Table 3.4). The FeIIFZ3 complex dissociates with a first-order rate constant of

4.3 × 10-5

s-1

(Thompsen and Mottola, 1984). However, the effect of FeIIFZ3

dissociation on the photo-formation rate of FeIIFZ3 is negligible, as any Fe(II)'

produced is predominantly recaptured by FZ under the conditions of photochemical

experiment. Therefore, the FeIIFZ3 formation rate was assumed to be equal to the

photo-production rate of Fe(II)'. The rate law for the photo-reductive dissociation of

FeIII

EDTA can be written as:

d[FeIIFZ3]

dt=

d[Fe(II)']

dt= k

hv

* [hν][FeIIIEDTA] (A1-3)

where [hv] represents light intensity. Under constant illumination, k*hv[hv] is also

invariant, i.e.:

d[Fe(II)']

dt= k

hv

* [hν][FeIIIEDTA] = khv

[FeIIIEDTA] (A1-4)

where khv indicates a first-order rate constant for photo-reductive dissociation of

FeIII

EDTA under the constant irradiation intensity examined. Approximating

[FeIII

EDTA] ≈ [Fe]Total – [FeIIFZ3] (due to the low [Fe(II)'], e.g., [Fe(II)'] = 2.0×10

-

8[Fe

IIIEDTA] + 1.4×10

-7[Fe

IIFZ3]) and [FZ] ≈ [FZ]Total (due to the high [FZ]Total, e.g.,

[FZ]Total = 1 mM >> [Fe]Total = 0.01-0.001 mM) followed by the integration gives the

relationship between FeIIFZ3 concentration and time, as follows:

ln[Fe]

T

[Fe]T

− [FeIIFZ

3]

= k

hv⋅ t (A1-5)

The value of khv was determined as the slope from linear regression of plots of time

versus ln([Fe]T/([Fe]T-[FeIIFZ3])) (parts E-F of Figure A1.1).


Figure A1.1. Time-course of

and (B and D) dark. For

equilibrated FeIII

EDTA

10 µM for Fe(III), 26

several hours at 27oC in the presence and absence of the light (

1). Photo-reductive dissociati

to the measurements with (E)

indicate average data and

represent linear regression lines.

A1.2. AVAILABILITY O

IN THE DARK

To constrain the chemical

examined the availability of

purpose, a photolyzed

Material for Chapter 3

course of FeIIFZ3 formation from Fe

IIIEDTA in

and (B and D) dark. For measurement of photo-reduction rate of Fe

EDTA complex and FZ were mixed in Fraquil* at concentrations

26 µM for EDTA and 1 mM for FZ, followed by incubation for

in the presence and absence of the light (157

reductive dissociation rate constants were determined by

s with (E) 1 µM and (F) 10 µM total Fe. Symbols and error bars

indicate average data and ±standard deviation from triplicate experiments

linear regression lines.

A1.2. AVAILABILITY OF PRE-PHOTOLYZED FE

chemical form of Fe available for uptake by M.

examined the availability of pre-photolyzed 55

FeEDTA for dark uptake

ed 55

FeEDTA solution was made by exposing the

173

(A and C) the light

reduction rate of FeIII

EDTA, pre-

concentrations of 1-

for FZ, followed by incubation for

µmol quanta.m-2

.s-

on rate constants were determined by applying eq. A1-5

Symbols and error bars

plicate experiments. Solid lines

PHOTOLYZED FE-EDTA

M. aeruginosa, we

FeEDTA for dark uptake. For this

the pre-equilibrated


174

55Fe

IIIEDTA stock to light supplied from the fluorescent tube for 0.5-48 h. The treated

solution was used for the 55

Fe uptake experiment immediately after the irradiation

treatment. The 55

Fe uptake experiment was undertaken following a procedure identical

to that described in Section 3.2.4, except that (i) the pre-photolyzed 55

FeEDTA

solution instead of pre-equilibrated 55

FeIII

EDTA stock was added to the culture at

concentrations of 200 nM 55

Fe and 26 µM EDTA and (ii) the culture was incubated in

the dark for 2 h.

As shown in Figure A1.2, measured 55

Fe uptake rates were independent of the duration

of pre-exposure to the light (0.5 to 48 hrs, p > 0.05), and no significant changes in

uptake rate were seen in the experiments with and without pre-photolysis treatment.

Importantly, the observed rates were substantially smaller than the light uptake rate

measured at identical concentrations of 55

Fe and EDTA by a factor of ~102. Such a low

availability of the pre-photolyzed Fe-EDTA indicates that the Fe uptake facilitated in

the light is tightly coupled with the availability of photo-produced Fe, but Fe is not

taken up by M. aeruginosa in the dark.

Assuming that photo-oxidized EDTA (EDTAox) is produced at a rate identical to that

of Fe(II)' via LMCT, the EDTAox concentration resulting from pre-photolysis was

computed to account for only 1% relative to total EDTA even after 48 h exposure. This

calculation suggests that intact EDTA still largely exists after the irradiation treatment.

Even though Fe uptake experiments were commenced immediately (~5 min) after the

photolysis treatment, such a short period would be sufficient to form a biologically

unavailable Fe complex, as the complexation of photo-produced Fe(II)' by the intact

EDTA and subsequent oxygenation of the complex formed are rapid and significantly

occur within this time scale.

In this regard, the photochemical cycle of Fe in our EDTA-buffered Fraquil* system

appears to be different from the system of aquachelin, a suite of siderophores produced

by the marine bacterium Halomonas aquamarina strain DS40M3. Barbeau et al.

(2001) found that Fe-aquachelin C complex undergoes photolysis in natural sunlight,

resulting in oxidative cleavage of the ligand at the site of the β-hydroxyasparate

residue. Although the photo-oxidized aquachelin still retains Fe(III)-binding capacity,

its binding strength for Fe(III)' was significantly lower than that of intact aquachelin


175

(conditional stability constants log cond

Fe'FeL,K measured by CLE-ACSV were 12.5 ± 0.3 M-

1 for intact aquachelin and 11.6 ± 0.2 M

-1 for photo-oxidized aquachelin).

Consequently, the availability of the photolyzed 59

Fe-aquachelin complex for uptake

by a natural phytoplankton assemblage collected in the oligotrophic North Atlantic

Ocean (under 1-2% ambient light level) was significantly higher than that of the intact

complex.

Figure A1.2. Bioavailability of pre-photolyzed 55

FeEDTA complex in the dark. The x-

axis represents the time for which 55

FeEDTA complex was exposed to the light (157

µmol photons.m-2

.s-1

) before the commencement of the 55

Fe uptake experiment.

Immediately after irradiation, the photolyzed 55

FeEDTA complex was added at final

concentrations of 200 nM Fe and 26 µM EDTA to the Fe and EDTA-free Fraquil*

containing M. aeruginosa cells at a density of 3 × 106 cell.mL

-1. Cells were then

incubated for 2 h in the dark at 27oC. Values shown represent the average and

±standard deviation from triplicate experiments.

0

0.002

0.004

0.006

0.008

0.01

0.012

0 0.5 1 2 48

Pre-exposure time to light (hr)

55F

e u

pta

ke

rate

(am

ol cell-1

hr-1

)


176

A1.3. DETERMINATION OF FORMATION AND

DISSOCIATION RATE CONSTANTS FOR FEII

EDTA

COMPLEX

A1.3.1 Formation Rate of FeIIEDTA Complex

The rate of Fe(II) complexation by EDTA was determined using the FZ competition

method. In this method, the concentration of FeIIFZ3 complex was

spectrophotometrically determined at a wavelength of 562 nm shortly after addition of

inorganic Fe(II) solution into Fraquil* containing EDTA and FZ. At particular

concentrations of EDTA and FZ, the concentration of FeIIFZ3 complex formed is a

function of the two rate constants for the competing reactions of Fe(II) complexation

(eq. A1-8) as stated below.

A1.3.1 .1 Methods

Sample solutions were prepared by pipetting appropriate volumes of FZ and EDTA

stocks into the Fe and EDTA-free Fraquil* to create ~2 mL of solution. While the final

concentration of FZ was kept constant at 1 mM, EDTA concentration was varied from

50 to 500 µM. In order to undertake the analyses, the solution was transferred to a 1

cm polystyrene cuvette and placed in the sample holder of Cary 1E UV–Visible

spectrophotometer. The absorbance of the solution was initially zeroed. The solution in

the cuvette was then spiked with Fe(II) stock to a final concentration of 5 µM or 10

µM and mixed by shaking. Immediately, the concentration of FeIIFZ3 complex formed

was spectrophotometrically monitored. The time lag between the addition of Fe(II)

stock (t = 0) and the measurement of initial data was approximately 3-5 s. The pH

change of the solution after addition of the chemicals was previously determined to be

less than 0.1 pH units. Calibration was performed by addition of Fe(II) stock to

Fraquil* containing 1 mM FZ (in the absence of EDTA). Correlation coefficients of the

linear calibration curves of r2 > 0.99 and a molar absorptivity of ε562 = ~28,000 M

-1

cm-1

were obtained. Measurements were made at 25oC.


177

A1.3.1.2 Results and Discussion

When Fe(II) is added to a solution containing the strong complexing ligand FZ, Fe(II)'

rapidly reacts with FZ to form the FeIIFZ3 complex. If the solution additionally

contains EDTA, EDTA effectively competes with FZ for Fe(II)' complexation. As

Fe(II)' oxidation is negligible, the system can be simply described by two competing

reactions:

3

IIFZFe 3FZ Fe(II)' FZf →+ −k (A1-6)

Fe(II)' + EDTA k

f −EDTA → Fe IIEDTA (A1-7)

where kf-EDTA is a rate constant for Fe(II)' complexation by EDTA. Under the

conditions examined, Fe(II)' complexation by FZ is rapid. For example, calculation

using the rate law equation for reaction eq. A1-6 indicates that 99.9% of Fe(II) binds

with FZ in < 0.03 s, such that the competitive complexation reactions by FZ and

EDTA are completed within a few seconds after the addition of Fe(II) stock. In

contrast, dissociation of the complexes is relatively slow and the concentration of Fe

complexes on this timescale can be described by the two formation reactions only (eqs.

A1-6 and A1-7).

Approximating [FZ] = [FZ]T - [FeIIFZ3] ≈ [FZ]T, [EDTA] = [EDTA]T - [Fe

IIEDTA] ≈

[EDTA]T, [Fe(II)]T = [FeIIFZ3] + [Fe

IIEDTA] (where subscript T indicates the total

concentration) and considering that the formation reactions are complete (t > 0.03 s), a

simple first order differential equation derived from eqs. A1-6 and A1-7 can be solved

using an identical procedure to that employed by Fujii et al. (2008). This analysis

generates the following relationship between kf-EDTA and the concentration of the

FeIIFZ3 complex detected (see Electronic Annex in Fujii et al. (2008) for a complete

derivation):

kf-EDTA

= k

f-FZ[FZ]

T

3

[EDTA]T

[Fe(II)]T

[FeIIFZ

3]

− 1

(A1-8)


178

The rate constant kf-EDTA was calculated to be 2.1 (± 0.2) × 106 M

-1.s

-1 (Table A1.1) by

substituting the total concentrations of Fe(II), FZ and EDTA, the value of kf-FZ, and the

spectrophotometrically determined [FeIIFZ3] into eq. A1-8. The relatively small

variation of the determined rate constants with varying EDTA and Fe(II)

concentrations indicates that eq. A1-8 is applicable over the range of EDTA

concentrations used in this work.

Table A1.1. Formation rate constant of FeIIEDTA complex (kf-EDTA) in Fraquil

* (pH

8).

[EDTA]T [Fe(II)]T [FZ]T [FeIIFZ3]

a) kf-EDTA Average kf-EDTA

µM µM µM µM M-1

.s-1

M-1

.s-1

50 5 1000 3.85 1.9 × 106

2.1(±0.2)×106

50 10 1000 7.33 2.3 × 106

100 5 1000 3.14 1.8 × 106

100 10 1000 6.06 2.0 × 106

250 5 1000 1.86 2.1 × 106

250 10 1000 3.64 2.2 × 106

500 5 1000 1.10 2.2 × 106

500 10 1000 1.98 2.5 × 106

a) Spectrophotometrically measured data.


179

A1.3.2. Dissociation of FeIIEDTA Complex

The dissociation rate of the FeIIEDTA complex was also determined in the presence of

FZ. For this measurement, a pre-equilibrated FeIIEDTA complex was initially prepared

and thermal dissociation of the complex was measured by spectrophotometrically

monitoring the time-dependent increase in FeIIFZ3 concentration. Ascorbate was used

to prevent significant oxidation of Fe(II) during the measurement.

A1.3.2.1. Methods

Pre-equilibrated FeIIEDTA solutions were made by mixing appropriate volumes of

EDTA, Fe(II) and ascorbate stocks, followed by addition of the mixture to ~5 mL of

Fraquil*. The solution was left for 1 h to reach equilibrium, then FZ stock was added to

the solution. The final concentration of total EDTA in the solution was varied from 4

µM to 400 µM, whereas concentrations of other chemicals were kept constant at 1 mM

for FZ, 1 mM for ascorbate and 4 µM for Fe(II). The absorbance at 562 nm was zeroed

immediately after FZ stock was added. The dissociation of the FeIIEDTA complex was

then monitored by measuring absorbance at 562 nm in the dark for 30 min using a

Cary 1E UV–Visible spectrophotometer. The pH change of the solution (pH 8) after

addition of the stocks was previously determined to be less than 0.1 pH units. The

effect of ascorbate on Fe(II) complexation by FZ was also determined to be

insignificant by examining time-dependent formation of FeIIFZ3 complex in the

presence and absence of 1 mM ascorbate at pH 5. For this purpose, Fraquil* adjusted to

pH 5 using HCl was prepared. The solution pH 5 instead of pH 8 was employed, since

Fe(II) oxidation at this pH is very slow even in the absence of ascorbate and as such is

negligible over the duration of the experiment. Since pKa for ascorbate are 4.2 and

11.6, we would expect that protonation does not substantially influence to the

complexation of Fe(II) by ascorbate in the pH range from 5 to 8. Measurements were

made at 25oC.

A1.3.2.2. Results and Discussion

When the FZ stock is added to Fraquil* containing Fe

IIEDTA, the ligand-exchange

reaction will proceed as follows:


180

Fe

IIEDTA + 3FZ

koverall → Fe

IIFZ

3 (A1-9)

where, koverall indicates a fourth-order rate constant for overall ligand-exchange

reaction. Then, the rate law for FeIIFZ3 formation can be written as:

d[FeIIFZ3]

dt= k

overall[FeIIEDTA][FZ]3 (A1-10)

Approximations [FeIIEDTA] ≈ [Fe]T – [Fe

IIFZ3] (due to the relatively low [Fe(II)'])

and [FZ] = [FZ]T – [FeIIFZ3] ≈ [FZ]T (due to the relatively low [Fe

IIFZ3]) followed by

the integration give relationship between FeIIFZ3 concentration and time, as follows:

ln[Fe(II)]

T

[Fe(II)]T

− [FeIIFZ

3]

= k

overall[FZ]

T

3 ⋅ t (A1-11)

The value of koverall[FZ]T was determined from linear regression analysis of plots of

time versus ln[Fe(II)]T/([Fe(II)]T-[FeIIFZ3]) (Figure A1.3, Table A1.2).

Assuming a disjunctive mechanism, the ligand exchange reaction between FeIIEDTA

and FZ can be described by thermal dissociation of FeIIEDTA and subsequent

complexation of liberated Fe(II)′ by FZ, as follows:

FeIIEDTA

kd-EDTA

kf-EDTA

→← Fe(II)' + EDTA (A1-12)

Fe(II)' + 3FZ

kf −FZ → Fe

IIFZ

3 (A1-13)

where kd-EDTA represents the rate constant for the dissociation of FeIIEDTA.


Figure A1.3. Kinetic data for the dissociation of Fe

8); (A) time-dependent formation of Fe

plots of time versus ln

determined as the slope of

Material for Chapter 3

inetic data for the dissociation of FeIIEDTA complex

dependent formation of FeIIFZ3 complex over a range of [EDTA]

ln[Fe(II)]T/([Fe(II)]T-[FeIIFZ3]). The value of

determined as the slope of the line in the panel B.

181

EDTA complex in Fraquil* (pH

a range of [EDTA]T and (B)

The value of koverall[FZ]T3 was


182

Table A1.2. Dissociation rate constant of FeIIEDTA complex (kd-EDTA) in Fraquil

* (pH 8).

[EDTA]T [Free EDTA]Initiala)

[Fe(II)]T [FZ]T koverall[FZ]T3

kd-EDTA Average kd-EDTA

µM µM µM mM ×10-3

s-1

×10-3

s-1

×10-3

s-1

4 0 4 1 1.0 1.0

1.2 (±0.23)

5 1 4 1 1.0 1.0

6 2 4 1 1.1 1.1

8 4 4 1 1.0 1.0

14 10 4 1 1.0 1.1

44 40 4 1 0.89 1.1

104 100 4 1 0.82 1.4

100 96 4 1 0.55 0.9

200 196 4 1 0.59 1.4

200 196 4 1 0.73 1.7

200 196 4 1 0.36 0.8

400 396 4 1 0.33 1.2

400 396 4 1 0.33 1.2

a) Initial concentration of EDTA that is not bound to Fe.


183

Due to the rapid complexation of Fe(II)' by EDTA and FZ at pH 8, we assume that the

intermediate Fe(II)' is highly reactive and does not accumulate to a significant extent.

Therefore, assuming steady-state for this intermediate, the FeIIFZ3 formation rate can

be written as follows:

d[FeIIFZ3]

dt= k

f-FZ[FZ]3[Fe(II)'] = k

f-FZ[FZ]3

kd-EDTA

[FeIIEDTA]

kf-EDTA

[EDTA] + kf-FZ

[FZ]3

(A1-14)

Comparison with eq. A1-10 yields:

kd-EDTA

=k

overall(k

f-EDTA[EDTA] + k

f-FZ[FZ]3)

kf-FZ

(A1-15)

If the term kf-EDTA[EDTA] is much smaller than kf-FZ[FZ]3 due to the low concentration

of free EDTA (e.g., [EDTA]T is less than 10 µM), reformation of FeIIEDTA complex

can be ignored and kd-EDTA is simply described as:

k

d-EDTA= k

overall[FZ]3 (A1-16)

The value of kd-EDTA was determined to be 1.2 (±0.23) × 10-3

s-1

by using eqs. A1-15

and A1-16 under a range of [EDTA]T (Table A1.2). Under the condition of [EDTA]T >

10 µM where the term for Fe(II)' re-formation by EDTA in eq. A1-15 becomes

significant, the kd-EDTA values were reasonably consistent with those determined in the

low [EDTA]T system, confirming that the kf-EDTA value reported in the previous section

was also reasonable.


184

A1.4. EFFECT OF FE(II) OXIDATION AND

DISSOCIATION ON CALCULATION OF STEADY-

STATE FE(II) CONCENTRATION IN THE 55

FE UPTAKE

EXPERIMENT

A1.4.1. Fe(II)' Oxidation

First-order rate constants for Fe(II)' oxidation and complexation by ligands under the

conditions of 55

Fe uptake experiments were calculated as follows:

kox[O2] = 8.8 × 0.24 × 10-3

= 2.1 × 10-3

(s-1

) (A1-17)

kf-EDTA[EDTA] = 2.1×106 × 2.6-26 × 10

-5 = 55-550 (s

-1) (A1-18)

kf-FZ[FZ]3 = 3.1 × 10

11 × (1 × 10

-3)3 = 310 (s

-1) (A1-19)

By comparing calculated values, the Fe(II)' oxidation rate was found to be 4-5 orders

of magnitude less than the rates of Fe(II)' complexation by EDTA or FZ. Therefore,

the effect of the oxidation reaction on the steady-state Fe(II)' concentration is

negligibly small.

A1.4.2. Dissociation of FeIIFZ3 and Fe

IIEDTA Complexes

The effect of thermal dissociation of FeIIEDTA and Fe

IIFZ3 on steady-state Fe(II)'

concentration was also examined under the conditions employed in 55

Fe uptake

experiments.

The dissociation rate of FeIIFZ3 accumulated during the uptake experiment was

calculated by multiplying the dissociation rate constant (kd-FZ) by the average FeIIFZ3

concentration during the experiment. Assuming that Fe(II)' formation is equal to

FeIIFZ3 formation at high FZ concentration, then the Fe

IIFZ3 formed at any given point

in time in the 55

Fe uptake experiment is calculated as follows:


185

[FeIIFZ

3]

t= [FeIIIEDTA]

Initial1− exp −k

hvt( ) (A1-20)

At the end of the 55

Fe uptake experiment (i.e., after 2 hr, [FeIII

EDTA]Initial = 200 nM

and khv = 6.4 × 10-6

s-1

), the final FeIIFZ3 concentration is estimated to be 9.0 nM.

Since FeIIFZ3 concentration increases in a linear manner with respect to time (Figure

A1.1), the time-averaged concentration for FeIIFZ3 ([Fe

IIFZ3]ave) over the duration of

experiment is half of this value (i.e., 4.5 nM). The average dissociation rate is then

calculated as follows:

kd-FZ[FeIIFZ3]ave = 4.3 × 10

-5 × 4.5 × 10

-9 = 1.9 × 10

-13 (M.s

-1) (A1-21)

For the EDTA complex, we assume that the concentration of FeIIEDTA is determined

by formation, dissociation and oxidation rates of the FeIIEDTA complex, i.e.:

d[Fe IIEDTA]

dt= k

f-EDTA[EDTA][Fe(II)'] − k

d-EDTA[Fe

IIEDTA] − k

ox-EDTA[O

2][Fe

IIEDTA]

(A1-22)

The steady-state FeIIEDTA concentration ([Fe

IIEDTA]ss) can be calculated as follows:

[FeIIEDTA]

SS=

kf-EDTA

[EDTA][Fe(II)']SS

kd-EDTA

+ kox-EDTA

[O2]

(A1-23)

where [Fe(II)']ss represents steady-state Fe(II)' concentration. The [Fe(II)']ss is

determined by rates of cellular uptake, complexation by ligands, dissociation of

formed complexes and oxidation, i.e.:

[Fe(II)']ss

=k

hv[FeIIIEDTA]

Initial+ k

d-EDTA[FeIIEDTA]

SS+ k

d-FZ[FeIIFZ

3]

ave

kf-EDTA

[EDTA] + kf-FZ

[FZ]3 + kup

[cell] + kox

[O2]

(A1-24)

The three unknown parameters [Fe(II)']SS, [FeIIEDTA]ss and kup were then calculated

from the three equations A1-23, A1-24 and ρ

Fe= k

up[Fe(II)']

ss using the trial and error

method (where ρ

Fe is the measured

55Fe uptake rate). Calculated data are shown in


186

Table 3.4. Using the computed [FeIIEDTA]ss, the thermal dissociation rate of

FeIIEDTA is calculated as follows:

kd-EDTA[FeIIEDTA]SS = 1.2 × 10

-3 × 0.20-1.7 × 10

-10 = 0.24-2.0 × 10

-13 (M.s

-1) (A1-25)

We can also calculate the rate of photo-dissociation of FeIII

EDTA as follows:

khv[FeIII

EDTA] = 6.4 × 10-6

× 2 × 10-7

= 1.3 × 10-12

(M.s-1

) (A1-26)

Therefore, dissociation of FeIIEDTA and Fe

IIFZ3 complexes will account for up to

23% of net Fe(II)' formation.

A1.5. OUTER-MEMBRANE PERMEABILITY AND

REPORTED PARAMETERS FOR PORIN PROPERTIES

Gram-negative bacteria have two structurally different phospholipid membranes; i.e.,

outer- and inner-membranes. The space between these two membranes is referred to as

the periplasm. To assimilate nutrients, this class of microorganisms must allow

molecules or ions to pass through the outer-membrane. Permeability of the outer-

membrane depends on the type of transport system involved (e.g., specific or non-

specific transport) and physicochemical properties of substrates such as size, polarity,

charge and specificity to outer-membrane receptors. Hydrophobic molecules have

generally lower permeability, as lipopolysaccharide (LPS) anchored in outer leaflet of

the asymmetric phospholipid bilayer act as a permeability barrier (Nikaido, 2003).

Once LPS synthesis is genetically knocked out, however, the permeability of

hydrophobic solutes increases (Nikaido, 1976). In contrast, almost all small

hydrophilic nutrients (generally less than 600 Da) including metal ions can passively

diffuse into the periplasmic space in a concentration gradient manner (Nikaido and

Rosenberg, 1981). This type of transport is mediated by the water-filled

transmembrane channel known as porins. The porin is one of the most abundant outer-

membrane proteins, in contrast to other minor transporter proteins for uptake of

specific nutrients such as vitamin B12 and ferric siderophore complexes (Hall and

Silhavy, 1981). Porins have been found for almost all Gram-negative bacteria

investigated so far including cyanobactria (Nikaido, 2003). Among these, porins of E.


187

coli. may be the best studied. Three different classes of proteins designated as OmpF,

OmpC and PhoE, have been found in this organism and each protein has molecular

mass of 36,000-38,000 Da (Nikaido, 2003). Diameters of the transmembrane channels

surrounded by the 16 β-barrel structure have been reported to be 1.0 nm to 1.2 nm

based on the crystal structure and the Stokes radius of nutrients which are capable of

passing through the channels (hereafter reported parameters for porin properties and

associated references are listed in Table A1.3, unless otherwise stated). The length of

porins has been reported to be around 2.8-7.5 nm, which is equal to or less than the

thickness of the outer-membrane bilayer. The density of porins in the outer-membrane

have been reported in the range of 7.9 × 1015

to 3.3 × 1016

porins per square meters.

These parameters for E. coli porins are somewhat different from porins of other

microorganisms (Table A1.3). Although studies on cyanobacterial porins are limited,

the molecular weight of pore-forming outer-menbrane proteins for cyanobacteria such

as Synechococcus sp. (52,000 Da (Hansel and Tadros, 1998, Hansel et al., 1994) and

Anabaena variabilis (40,000-80,000 Da, (Benz and Bohme, 1985)) seems to be larger

than the other bacterial porins. A relatively large pore diameter of 1.7 nm has been

reported for Anabaena variabilis. The single channel conductance determined in 1 M

KCl ranges from 0.4 nS to 5.5 nS (Benz and Bohme, 1985, Hansel et al., 1994, Hansel

and Tadros, 1998), which is more variable than the 2-3 nS reported for E. Coli

(Nikaido, 2003), suggesting that the permeability of the outer-membrane for

cyanobacteria varies over a relatively wide range.


188

Table A1.3. Published values of porin properties for various Gram-negative bacteria.

Microorganism Class of pore- Pore diameter Channel length Porins Surface area Porin density Reference forming protein 2a (nm) l (nm) per cell of cell (m2) Nporin (m

-2)

Escherichia coli B Pore-forming protein 1.2 4–7.5a)

1.0×105 3.0×10

-12 3.3×10

16 Nikaido and Rosenberg (1981)

Escherichia coli ML308 Pore-forming protein 1.16 4–7.5 1.1×105 1.4×10

-11 7.9×10

15 West and Page (1984)

Escherichia coli K12 OmpF 1.13–1.16 6 Hancock (1987)

Escherichia coli K12 OmpC 1.02–1.13 6 Hancock (1987)

Escherichia coli K12 PhoE 1.06–1.13 6 Hancock (1987)

Escherichia coli K12 PhoE 1.0 a)

2.8–4.5 Jap et al. (1991)

Rhodopseudomonas capsulata

ATCC 23782 Outer-membrane porin 1.6 4–7.5a)

Flammann and Weckesser (1984)

Rhodobacter capsulatus 37b4 Outer-membrane porin 1.0 a)

2–4 Weiss et al. (1991)

Pseudomonas aeruginosa Protein F 1.63–2.9 6 Hancock (1987)

Anabaena variabilis

ATCC29423 Pore-forming protein 1.7 7.5 Benz and Bohme (1985)


PCC7806 1.1×10-10

Fujii et al. (2010)

a) Values were assumed in this work; pore diameter = ~1 nm and channel length = ~4–7.5 nm.


189

Table A1.4. Range of uptake rate constant (kup) calculated using published parameters (a, porin radius; l, channel length; Nporin, porin density; D,

diffusion coefficient of metal ions; As, surface area of Microcystis aeruginosa PCC7806).

Microorganism Reference πa2/l (×10

-9 m)

a) kup = 1,000Dπa2NporinAS/l (×10

-9 L.cell

-1.s

-1)b) Lower limit Average Upper limit Lower limit Average Upper limit

Escherichia coli B Nikaido and Rosenberg (1981) 0.15 0.22 0.28 0.07 0.57 1.1

Escherichia coli ML308 West and Page (1984) 0.14 0.20 0.26 0.06 0.53 1.0

Escherichia coli K12 (OmpF) Hancock (1987) 0.17 0.17 0.18 0.07 0.37 0.7

Escherichia coli K12 (OmpC) Hancock (1987) 0.14 0.15 0.17 0.06 0.35 0.6

Escherichia coli K12 (PhoE) Hancock (1987) 0.15 0.16 0.18 0.07 0.36 0.7

Escherichia coli K12 (PhoE) Jap et al. (1991) 0.17 0.23 0.28 0.08 0.57 1.1

Rhodopseudomonas capsulata

ATCC 23782 Flammann and Weckesser (1984) 0.27 0.39 0.50 0.12 1.0 1.9

Rhodobacter capsulatus 37b4 Weiss et al. (1991) 0.20 0.29 0.39 0.09 0.78 1.5

Pseudomonas aeruginosa Hancock (1987) 0.35 0.72 1.10 0.16 2.2 4.1

Anabaena variabilis ATCC29423 Benz and Bohme (1985) 0.30 0.52 a)

For πa2/l, upper and lower limits and average value were calculated using the range of published parameters for microorganisms listed. The

values used are shown in Table A1.3.


190

b) For the calculation of kup, porin radius (a), channel length (l) and porin density (Nporin) listed in Table A1.3 were used. Reported parameters for

diffusion coefficient of metal ions D = 0.5-1×10-9

m2.s

-1 (Buffle et al., 2009) and surface area of Microcystis aeruginosa PCC7806 AS = 113 µm

2

(Fujii et al., 2010) were also used. The average represents the mean value of lower and upper limits.


191

REFERENCES

BARBEAU, K., RUE, E. L., BRULAND, K. W. & BUTLER, A. 2001. Photochemical

cycling of iron in the surface ocean mediated by microbial iron(III)-binding

ligands. Nature, 413, 409-413.

BENZ, R. & BOHME, H. 1985. Pore formation by an outer-membrane protein of the

cyanobacterium Anabaena variabilis. Biochimica et Biophysica Acta, 812, 286-

292.

BUFFLE, J., WILKINSON, K. J. & VAN LEEUWEN, H. P. 2009. Chemodynamics

and bioavailability in natural waters. Environmental Science & Technology, 43,

7170-7174.

FLAMMANN, H. T. & WECKESSER, J. 1984. Porin isolated from the cell-envelope

of Rhodopseudomonas capsulata. Journal of Bacteriology, 159, 410-412.

FUJII, M., ROSE, A. L., OMURA, T. & WAITE, T. D. 2010. Effect of Fe(II) and

Fe(III) transformation kinetics on iron acquisition by a toxic strain of

Microcystis aeruginosa. Environmental Science & Technology, 44, 1980-1986.

FUJII, M., ROSE, A. L., WAITE, T. D. & OMURA, T. 2008. Effect of divalent

cations on the kinetics of Fe(III) complexation by organic ligands in natural

waters. Geochimica et Cosmochimica Acta, 72, 1335-1349.

HALL, M. N. & SILHAVY, T. J. 1981. Genetic-analysis of the major outer-membrane

proteins of Escherichia coli. Annual Review of Genetics, 15, 91-142.

HANCOCK, R. E. W. 1987. Role of porins in outer-membrane permeability. Journal

of Bacteriology, 169, 929-933.

HANSEL, A., SCHMID, A., TADROS, M. H. & JURGENS, U. J. 1994. Isolation and

characterization of porin from the outer-membrane of Synechococcus

PCC6301. Archives of Microbiology, 161, 163-167.


192

HANSEL, A. & TADROS, M. H. 1998. Characterization of two pore-forming proteins

isolated from the outer membrane of Synechococcus PCC 6301. Current


JAP, B. K., WALIAN, P. J. & GEHRING, K. 1991. Structural architecture of an outer-

membrane channel as determined by electron crystallography. Nature, 350,

167-170.

LIN, J. & KESTER, D. R. 1992. The kinetics of Fe(II) complexation by ferrozine in

seawater. Marine Chemistry, 38, 283-301.

NIKAIDO, H. 1976. Outer membrane of Salmonella typhimurium. Transmembrane

diffusion of some hydrophobic substances. Biochimica et Biophysica Acta, 433,

118-132.

NIKAIDO, H. 2003. Molecular basis of bacterial outer membrane permeability

revisited. Microbiology and Molecular Biology Reviews, 67, 593-656.

NIKAIDO, H. & ROSENBERG, E. Y. 1981. Effect of solute size on diffusion rates

through the transmembrane pores of the outer-membrane of Escherichia coli.

Journal of General Physiology, 77, 121-135.






WEISS, M. S., KREUSCH, A., SCHILTZ, E., NESTEL, U., WELTE, W.,

WECKESSER, J. & SCHULZ, G. E. 1991. The structure of porin from

Rhodobacter capsulatus at 1.8 a resolution. Febs Letters, 280, 379-382.

WEST, I. C. & PAGE, M. G. P. 1984. When is the outer-membrane of Escherichia coli rate-

limiting for uptake of galactosides. Journal of Theoretical Biology, 110, 11-19.

193

APPENDIX 2


CHAPTER 4 - KINETICS OF

EXTRACELLULAR IRON TRANSPORT

TO PERIPLASMIC AND CYTOPLASMIC

COMPARTMENTS OF THE


MICROCYSTIS AERUGINOSA


194

A2.1. DETERMINATION OF PARAMETERS RELEVANT

TO FE UPTAKE AND INTRACELLULAR TRANSPORT

Assuming that the Fe′ concentration in the periplasm ([Fe′peri]) is comparable to that in

the bulk ([Fe′]) (this assumption is justified in part A2.2 of Appendix 2), log-

transformation of the Monod-type expression for cytoplasmic 55

Fe accumulation can

be described as follows:

max ' max '

cyto Fe'(peri) peri Fe'(peri)

' '

Fe'(peri) peri Fe'(peri)

[Fe ] [Fe ] [Fe ]log log log

[Fe ] [Fe ]

d

dt K K

ρ ρ = ≈ + +

(A2-1)

The steady-state concentration of extracellular Fe′ was firstly calculated under the

particular conditions of the 55

Fe accumulation experiments undertaken in this study

from the balance of formation and dissociation of the extracellular Fe-ligand

complexes by using the kinetic model in part A of Table 4.1. The calculation of the

steady-state concentration of extracellular unchelated Fe was justified as follows.

Since concentrations of extracellular Fe′ for particular Fe:Cit ratios can be calculated

from the balance between dissociation and recomplexation rates of ferric citrate (k-d-

L[FeL] and kf-L[L][Fe′], respectively) and total diffusional flux of unchelated Fe into

the periplasm (JFe′[cell] mol.s-1

), the time-dependent change of unchelated Fe

concentration can be described as follows:

''

d-L f-L Fe'

' '

d-L f-L dif

' ' '

d-L f-L dif peri

'

d-L f-L

'

d-L T f-L T T

[Fe ][FeL] [L][Fe ] [cell]

[FeL] [L][Fe ] [Fe ][cell]

[FeL] [L][Fe ] ([Fe ] [Fe ])[cell]

[FeL] [L][Fe ]

[Fe] ([L] [Fe] )[Fe ]

dk k J

dt

k k k

k k k

k k

k k

= − +

= − + ∆

= − + −

≈ −

= − −

(A2-2)

where [cell] is cell density (cell.L-1

), and the effect of diffusional influx in the

calculation of steady-state concentration of extracellular Fe′ is negligibly small due to

the assumption that [Fe′peri] ≈ [Fe′]. In the calculation of [Fe′], [FeL] and [L] were


195

approximated to be equal to total Fe and ligand concentrations in the system ([Fe]T and

[L]T), respectively, as Fe′ concentration is substantially lower than chelated Fe

concentration under the conditions examined in this work. Thus, at steady-state, the

following expression can be used to calculate the extracellular Fe′ concentration

([Fe′]SS):

' d-L TSS

f-L T T

[Fe][Fe ]

([L] [Fe] )

k

k=

− (A2-3)

The Fe uptake parameters max

Fe'(peri)ρ and Fe'(peri)K were then determined by fitting eq. A2-

1 to the data shown in part B of Figure 4.2. In the fitting process, nonlinear regression

analysis was performed using R version 2.13.0 (free software for statistical

computation). Subsequently, the rate constant for translocation of periplasmic Fe to the

cytoplasm (k2) was determined by fitting eq. 4.5 to the data in part A of Figure 4.2.

The best fit of the model to the experimental data was obtained by Microsoft Excel

using a least-squares method in which the sum of the mean square error using the

average values of the experimental data was minimized. k+1 was determined by fitting

eq. 4.7 to the data shown in part A of Figure 4.1 then k-1 was calculated from Fe'(peri)K =

(k-1+k2)/k+1. Finally, the value for [Xperi]T was determined from max

Fe'(peri)ρ = k2[Xperi]T.

A2.2. CALCULATION OF STEADY-STATE

CONCENTRATION OF PERIPLASMIC UNCHELATED

FE

The Fe′ concentration in the periplasm was calculated by assuming that, under steady-

state conditions, the flux of periplasmic Fe transported to the cytoplasm (ρFe′) is equal

to the diffusional flux of Fe across the outer membrane (JFe′ mol.cell-1

.s-1

). This

assumption yields the following relationship between Fe′ concentrations in the

periplasm ([Fe′peri]) and bulk ([Fe′]):


196

Fe' Fe'

max '

Fe'(peri) peri'

dif '

Fe'(peri) peri

' ' ' max '

dif peri peri Fe'(peri) Fe'(peri) peri

' 2 ' ' ' '

dif peri peri Fe'(peri) peri Fe'(peri)

[Fe ][Fe ]

[Fe ]

([Fe ] [Fe ])([Fe ] ) [Fe ]

([Fe ] [Fe ] [Fe ][Fe ] [Fe ] )

J

kK

k K

k K K

ρ

ρ

ρ

ρ

=

− ∆ =+

− − + =

− + − − = max '

Fe'(peri) peri

' 2 ' max ' '

dif peri dif Fe'(peri) dif Fe'(peri) peri dif Fe'(peri)

' max ' max 2

dif Fe'(peri) dif Fe'(peri) dif Fe'(peri) dif Fe'(peri)'

peri

[Fe ]

[Fe ] ( [Fe ] )[Fe ] [Fe ] 0

( [Fe ] ) ( [Fe ] )[Fe ]

k k K k k K

k K k k K k

ρ

ρ ρ

+ − + − =

− − + + − +=

2 '

dif Fe'(peri)

dif

4 [Fe ]

2

k K

k

+

(A2-4)

where kdif is the diffusion constant for unchelated Fe (L.cell-1

.s-1

), and ∆[Fe′] is the

difference between unchelated Fe concentrations in the extracellular bulk medium and

periplasmic space. By using eq. A2-4 and relevant parameters determined in part A2.1

of Appendix 2 (Table 4.1), [Fe′peri] was calculated over a range of [Fe′] concentrations

from 1 fM to 10 nM. Calculations suggested that [Fe′peri] was similar to [Fe′]

([Fe′peri]/[Fe′] was greater than 0.999). Therefore, we assumed that [Fe′peri] ≈ [Fe′] in

our work.

A2.3. EQUATION FOR TIME-DEPENDENT CHANGE OF

PERIPLASMIC FE

According to eq. 4.4 in the text, the time-dependent change of periplasmic Fe

concentration is described as follows:

peri peri

'

1 peri peri 1 2 peri

[Fe ] [FeX ]

[Fe ][X ] ( )[FeX ]

d d

dt dt

k k k+ −

≈

= − +

(A2-5)

Substitution of the mass balance equation for Fe-binding sites (i.e., [Xperi]T = [FeXperi]

+ [Xperi]) and Fe'(peri)K = (k-1+k2)/k+1 yields the following relationship:


197

peri ' '

1 peri peri T 1 2 1 peri peri

' '1 2peri peri T Fe'(peri) peri peri

Fe'(peri)

[FeX ][Fe ][X ] ( + [Fe ])[FeX ]

[Fe ][X ] ( +[Fe ])[FeX ]

dk k k k

dt

k kK

K

+ − +

−

= − +

+= −

(A2-6)

Then, integration and rearrangement give the following:

peri ' 1 2Fe'(peri) peri'

peri peri T Fe'(peri)

peri'

Fe'(peri) peri

' '

peri peri T Fe'(peri) peri 1 2

peri'

Fe'(peri) peri Fe'(peri)

[FeX ]( +[Fe ])

[Fe ][X ][FeX ]

+[Fe ]

[Fe ][X ] ( +[Fe ])( )ln [FeX ]

+[Fe ]

d k kK dt

K

K

K k k

K K

−

−

+=

−

+− − = ⋅

∫ ∫

t C+

(A2-7)

where C is an integration constant. The condition that [FeXperi] = 0 at time zero gives:

'

peri T

'

Fe peri

[Fe ][X]ln

+[Fe ]C

K

= −

(A2-8)

By substituting C in eq. A2-7,

' '

peri peri T peri peri T'

peri Fe'(peri) peri 1' '

Fe'(peri) peri Fe'(peri) peri

[Fe ][X ] [Fe ][X ]ln [FeX ] ( +[Fe ]) ln

+[Fe ] +[Fe ]K k t

K K+

− − = ⋅ −

(A2-9)

'

peri peri T '

peri Fe'(peri) peri 1'

Fe'(peri) peri

[Fe ][X ][FeX ] 1 exp ( +[Fe ])

+[Fe ]K k t

K+

= − − ⋅ (A2-10)


198

Table A2.1. Measured and modelled values for the time course of 55

Fe accumulation in the periplasmic and cytoplasmic fractions.

No.

55

Fe concentration (amol.cell-1

) PCC7806 strain PCC7005 strain

Incubation Periplasm Cytoplasm Periplasm Cytoplasm

time (hr) experiment model experiment model experiment model experiment model

1 0 0.03 0.0 0.14 0.0 0.01 0.0 0.13 0.0

2 0 0.01 0.0 0.18 0.0 0.01 0.0 0.18 0.0

3 0 0.01 0.0 0.16 0.0 0.004 0.0 0.16 0.0

4 1 0.12 0.16 1.54 1.1 0.08 0.12 1.44 1.2

5 1 0.13 0.16 2.17 1.1 0.06 0.12 1.45 1.2

6 1 0.09 0.16 1.38 1.1 0.07 0.12 1.42 1.2

7 3 0.18 0.16 3.66 3.3 0.07 0.12 4.28 3.7

8 3 0.22 0.16 2.82 3.3 0.08 0.12 3.05 3.7

9 3 0.15 0.16 2.84 3.3 0.07 0.12 3.14 3.7

10 5 0.09 0.16 4.25 5.5 0.09 0.12 5.21 6.2

11 5 0.09 0.16 4.43 5.5 0.10 0.12 5.97 6.2

12 5 0.12 0.16 4.71 5.5 0.10 0.12 4.49 6.2

13 7 0.16 0.16 3.84 -a 0.14 0.12 4.49 -

a

14 7 0.21 0.16 3.86 -a 0.12 0.12 5.09 -

a

15 7 0.13 0.16 5.27 -a 0.15 0.12 5.67 -

a

16 9 0.20 0.16 4.11 -a 0.12 0.12 4.65 -

a

17 9 0.16 0.16 3.89 -a 0.13 0.12 4.10 -

a

18 9 0.16 0.16 4.42 -a 0.12 0.12 5.73 -

a

a Value not calculated.


199

Table A2.2. Measured and modelled values for the steady-state periplasmic 55

Fe concentration and accumulation rate of cytoplasmic 55

Fe over a

range of Fe:citrate ratiosa.

55Fe in periplasm (amol.cell

-1) Accumulation rate of cytoplasmic

55Fe (amol.cell

-1.hr

-1)

No. Sample name PCC7806 PCC7005 PCC7806 PCC7005 experiment model experiment model experiment model experiment model

1 FeCit-5µM 0.36 0.29 0.33 0.26 2.1 1.9 3.0 2.4

2 FeCit-5µM 0.34 0.29 0.30 0.26 2.7 1.9 3.4 2.4

3 FeCit-5µM 0.31 0.29 0.38 0.26 2.2 1.9 3.3 2.4

4 FeCit-20µM 0.20 0.16 0.12 0.13 0.82 0.96 0.93 1.1

5 FeCit-20µM 0.16 0.16 0.13 0.13 0.78 0.96 0.82 1.1

6 FeCit-20µM 0.16 0.16 0.12 0.13 0.88 0.96 1.1 1.1

7 FeCit-50µM 0.15 0.080 0.09 0.070 0.71 0.42 0.90 0.48

8 FeCit-50µM 0.13 0.080 0.13 0.070 0.64 0.42 0.90 0.48

9 FeCit-50µM 0.10 0.080 0.11 0.070 0.58 0.42 0.93 0.48

10 FeCit-100µM 0.053 0.053 0.061 0.048 0.29 0.23 0.31 0.26

11 FeCit-100µM 0.070 0.053 0.085 0.048 0.31 0.23 0.40 0.26

12 FeCit-100µM 0.059 0.053 0.076 0.048 0.33 0.23 0.39 0.26

13 FeCit-200µM 0.016 0.031 0.020 0.030 0.086 0.079 0.11 0.077

14 FeCit-200µM 0.013 0.031 0.025 0.030 0.17 0.079 0.14 0.077

15 FeCit-200µM 0.020 0.031 0.024 0.030 0.059 0.079 0.093 0.077

a The Fe:citrate ratios were identical to those listed in Table A2.3.


200

Table A2.3. Calculated values of unchelated Fe concentrations in the extracellular milieu and periplasm.

PCC7806 strain PCC7005 strain

No. Sample name [Fe]T [Cit]T kf-cit kd-cit [Fe'peri] [Fe'] pFe' Ratio of [Fe'peri] [Fe'] pFe' Ratio of

(nM) (µM)

(x 105 M

-1s

-

1) (x 10

-3 s

-1) (pM) (pM) (M) [Fe'peri]/[Fe'] (pM) (pM) (M) [Fe'peri]/[Fe']

1 FeCit-5µM 700 5 2.1 3.5 2691 2691 8.6 0.999949 2691 2691 8.6 0.999948

2 FeCit-20µM 700 20 2.1 2.9 507 507 9.3 0.999845 507 507 9.3 0.999840

3 FeCit-50µM 700 50 2.1 2.2 152 152 9.8 0.999768 152 152 9.8 0.999766

4 FeCit-100µM 700 100 2.1 1.6 54 54 10.3 0.999732 54 54 10.3 0.999734

5 FeCit-200µM 700 200 2.1 1.0 17 17 10.8 0.999715 17 17 10.8 0.999722

201

APPENDIX 3


CHAPTER 5 - IRON UPTAKE KINETICS

BY THE FRESHWATER


AERUGINOSA IN THE PRESENCE OF

SUWANNEE RIVER FULVIC ACID


202

A3.1. SUPPLEMENTAL FIGURES

Figure A3.1. Time-course of 55

Fe uptake under the dark (closed symbol) and light

(open symbol) conditions. 55

Fe uptake was measured by incubating cells (at density of

1.6 × 106 cell.mL

-1) in Fraquil

* containing pre-equilibrated

55Fe

IIISRFA complex at

27oC. Concentrations of Fe and SRFA were 200 nM and 1 mg.L

-1, respectively.

Symbols represent experimental data. Solid and dotted lines were yielded by applying

a linear regression analysis to the data collected within 2 h under the dark and light

conditions, respectively.

0

2

4

6

8

10

12

14

0 2 4 6 8 10 12 14

Ce

llu

lar

55Fe

(a

mo

l.ce

ll-1

)

Time (h)

y = 1.95x + 0.44

R² = 0.94y = 1.28x + 0.46

R² = 0.87


203


Fe uptake rate to calculated Fe(III) uptake for

M. aeruginosa PCC7806. 55

Fe uptake rates were determined in the short-term

incubational assay under the dark in modified Fraquil* containing 200 nM for Fe, 1, 5

and 25 mg L-1

for SRFA and 1 mM for FZ. In the model calculations, steady-state

concentrations for Fe(III)' were determined at the concentration identical to those

employed in the short-term assay by using rate constants for complexation and

dissociation for FeIII

SRFA complex published by Rose (square), Jones (diamond) and

Bligh (triangle). Fe(III) uptake rates were then calculated by use of Monod-type

equation with parameters listed in Table 5.1. Solid line represents linear line with 1:1

slope.

-20

-19.5

-19

-18.5

-18

-17.5

-17

-20 -19.5 -19 -18.5 -18 -17.5 -17

Rose model

Jones model

Bligh model

1:1

0

0.5

1

1.5

2

2.5

3

3.5

0 0.5 1 1.5 2 2.5 3 3.5

Rose model

Jones model

Bligh model

1:1

Logarithm of calculated Fe(III) uptake rate (mol.cell-1.hr-1)

Me

asu

red

55Fe

(III

) u

pta

ke r

ate

(am

ol.

cell-1

.hr-1

)

Calculated Fe(III) uptake rate (amol.cell-1.hr-1)

Loga

rith

m o

f m

ea

sure

d 5

5Fe

(III

)

up

take

rat

e (

amo

l.ce

ll-1.h

r-1)


204


Fe(II) uptake rate to calculated Fe(II) uptake

for M. aeruginosa PCC7806. Measured Fe(II) uptake rates in this figure were

determined by subtracting 55

Fe uptake rate in the presence of FZ from that measured in

the absence of FZ. The short-term incubational assays were performed in the absence

and presence of FZ under the dark in modified Fraquil* containing 200 nM for Fe, 1, 5

and 25 mg L-1

for SRFA and 1 mM for FZ. In the model calculations, steady-state

concentrations for Fe(II)' were determined at the concentration identical to those

employed in the short-term assay by using rate constants for complexation and

dissociation for FeIISRFA complex published by Rose (square) and Bligh (triangle).

Fe(II) uptake rates were then calculated by use of Monod-type equation with

0

0.2

0.4

0.6

0.8

1

0 0.2 0.4 0.6 0.8 1

Rose model

Bligh model 1:1

Me

asu

red

55F

e(I

I) u

pta

ke

ra

te

(am

ol.

cell

-1.h

r-1)

Calculated Fe(II) uptake rate (amol.cell-1.hr-1)

-20

-19.5

-19

-18.5

-18

-17.5

-20 -19.5 -19 -18.5 -18 -17.5

Rose model

Bligh model 1:1

Log

ari

thm

of

me

asu

red

55F

e(I

I)

up

tak

e r

ate

(a

mo

l.ce

ll-1

.hr-1

)

Logarithm of calculated Fe(II) uptake rate (mol.cell-1.hr-1)


205

parameters listed in Table 5.1. Solid line represents linear line with 1:1 slope. Error bar

indicates standard deviation from duplicate experiments.

Figure A3.4. Comparison of calculated steady-state concentration of Fe(II)’ under the

dark and light conditions. Steady-state concentrations for Fe(II)' were determined by

using rate constants for complexation and dissociation for FeIISRFA complex

published by Rose (square) and Bligh (triangle), the photochemical and non-


SRFA complex and oxidation of FeIISRFA complex.

Solid line represents linear line with 1:1 slope.

-14

-12

-10

-8

-6

-14 -12 -10 -8 -6

Log

ari

thm

of

calc

ula

ted

[Fe

(II)

']S

Su

nd

er

da

rk (

M)

Logarithm of calculated

[Fe(II)']SS under light (M)


206

Figure A3.5. Effect of pH on the 55

Fe uptake rate for M. aeruginosa PCC7806 under

(A) dark and (B) light. Effects of FZ (gray bar) and SOD (white bar) on 55

Fe uptake

were also examined compared to control treatment (black bar) where addition of FZ or

SOD was omitted. Error bar indicates standard deviation from triplicate experiments.

Asterisks indicate that 55

Fe uptake rate in the presence of chemical treatment is

significantly different from control (55

Fe uptake rate in the absence of FZ or SOD) for

each pH at the levels of p < 0.01 for ** and p < 0.05 for * using a single-tailed

heteroscedastic t-test.

0

0.2

0.4

0.6

0.8

1

1.2

6 7 8 9

Dark

Dark-FZ

Dark-SOD

A.

* *

*

*

0

0.2

0.4

0.6

0.8

1

1.2

6 7 8 9

Light

Light-FZ

Light-SOD

B.

*

*

*

***

****

55F

e u

pta

ke

ra

te (

am

ol.

cell

-1.h

r-1)

pH


207

A3.2. PHOTOCHEMICAL AND NON-PHOTOCHEMICAL

FE(III) REDUCTION EXPERIMENT

A3.2.1. Methods

A3.2.1.1. Fe(III) reduction experiment

Photo- and thermal reduction rates of FeIII

L were spectrophotometrically determined

by measurement of the absorbance of ferrous-ferrozine complex (FeIIFZ3) at a

wavelength of 562 nm, where the FeIIFZ3 absorbs most strongly. During the

measurement, the absorbance at 562 nm (i.e., FeIIFZ3 concentration) increased over

time after mixing the pre-equilibrated FeIII

L solution and FZ stocks in Fraquil*, as

FeIII

L is reduced to Fe(II) followed by Fe(II) complexation by FZ. All stock solutions

and vessels used in the experiments were prepared as described in Section 5.1, Chapter

5.

The solutions of FeIII

L and FZ (prepared at pH 6-9 depending on the pH of Fraquil*

interest) were spiked into an appropriate volume of the Fe- and ligand-free Fraquil*

(pH 6-9) in order to provide a total volume of 2-100 mL in an acid-washed 1 cm path

length polystyrene spectrophotometer cuvette or polycarbonate bottles (depending on

the total Fe concentration: the cuvette for 10 µM Fe and the polycarbonate bottle for 1

µM Fe) at final concentrations of 1-10 µM for Fe and 1 mM for FZ. The final

concentrations of the Fe-binding ligands were 1-250 mg L-1

for SRFA, 26 µM for

EDTA and 100 µM for citrate. The samples were then incubated either under light or

dark condition at 27oC. The photo- and thermal reduction rates were determined under

the condition identical to the Fe uptake experiment by using the same light source and

distance between the sample and the fluorescent light tube. At various times from 1

min to 36 h after mixing the FeIII

L and FZ stocks, FeIIFZ3 concentration in the sample

was measured by a Varian Cary 50 UV-VIS spectrophotometer for samples with 10

µM Fe or an Ocean Optics 1 m pathlength spectrophotometry system for samples with

1 µM Fe. In the latter measurement, the Ocean Optics spectrophotometry instruments

were used in order to determine concentrations of the FeIIFZ3 at nanomolar level (see

Section 2.4.1, Chapter 2 for the detailed description).


208

To examine the effect of Fe contamination in reagents such as the fulvic acid stock on

the FeIIFZ3 formation rate, the Fe(II) formation experiment was also performed

without addition of Fe(III) stock to the sample. The FeIIFZ3 formation so measured

was negligibly small, suggesting that Fe contamination in the sample was negligible.

Although natural organic matters such as fulvic acid significantly absorb visible light,

the absorbance of the solution at 562 nm was measured to be insignificant.

A3.2.1.2. Model fitting

The best fit of the model to the experimental data was determined by using a least-

squares method in which the mean square error between the model value and the

average of the experimental data was minimized.

A3.2.2. RESULTS AND DISCUSSION

A3.2.2.1. Rate of photochemical and non-photochemical reduction

Reduction rates for Fe(III) complexed by various ligands including SRFA, EDTA and

citrate was examined in Fraquil* by measuring Fe

IIFZ3 formed in the absence and

presence of light with primary kinetic data shown in Figure A3.6. Under both light and

dark conditions FeIIFZ3 accounting for ~10% of total Fe was rapidly formed within 1

min’s incubation after FZ was added to Fraquil* containing Fe complexed by SRFA.

Followed by the initial jump, the FeIIFZ3 concentration linearly increased with respect

to time over 4 h. The initial increase of FeIIFZ3 is likely due to rapid reaction of Fe(II)

with FZ, suggesting that Fe(II) may be formed significantly in the pre-equilibrated

FeIII

SRFA stock during the dark storage (Pullin and Cabaniss, 2003). Concentration of

FeIIFZ3 increased with time at substantially higher rate under the light compared to that

in the dark condition for the three ligands. However, only for SRFA, the dark FeIIFZ3

formation was not negligible (part B of Figure A3.6).


209

Figure A3.6. Primary kinetic data of FeIIFZ3 formation in the (A) light and (B) dark

conditions. The time-dependent FeIIFZ3 formation in Fraquil

* (pH 8) was

spectrophotometrically monitored for 4 h at concentrations of 1 µM for Fe(III), 1 mM

for FZ, 1 mg.L-1

for SRFA, 26 µM for EDTA and 100 µM for citrate. Symbols and

error bars indicate average data and ±standard deviation from triplicate experiments.

To investigate factor(s) influencing the time-dependent FeIIFZ3 formation in the dark

SRFA system, we repeated the Fe(III) reduction experiment under identical conditions

except that SRFA was replaced by other synthetic Fe chelators such as citrate and

EDTA. Since the dark FeIIFZ3 production was not discernible in such cases, it is

unlikely that light leakage into the sample or any chemicals present in Fraquil* are

0

100

200

300

400

500

0 50 100 150 200 250 300

A.

-50

0

50

100

150

200

250

300

0 50 100 150 200 250 300

Time (minutes)

EDTA

Citrate

SRFA

B.[Fe

(II)

FZ

3] (n

M)


210

responsible for the dark FeIIFZ3 formation. Although presence of FZ has been reported

to promote the Fe(III) reduction under some conditions via direct complexation of

Fe(III)' by FZ (Shaked et al., 2004), this phenomenon is unlikely to account for the

observed FeIIFZ3 formation, as negligible Fe

IIFZ3 formation was seen in the EDTA and

citrate systems which cover a range of steady-state Fe(III)' concentration from pFe' =

11.96 for EDTA to pFe' = 10.10 for citrate (part B of Figure A3.6). The dark FeIIFZ3

formation in the SRFA system, therefore, rather indicates that Fe(II) is continuously

formed by the thermal reduction of some portion of FeIII

SRFA, probably due to the

presence of redox-active moieties in fulvic acid (e.g., hydroquinones), as reported

previously (Garg et al., 2012, Pham et al., 2012).

The experiments were undertaken in excess of FZ in which any other competing

reactions relevant to Fe(II) such as complexation of Fe(II)' and re-oxygenation to

Fe(III) are likely to be ignored. For example, the SRFA-mediated Fe(II)' formation

may be followed by either complexation by FZ or recomplexation by SRFA

(represented by L in eq. A3-2). The two competing reactions can be simply described

as follows:

Fe(II)f FZ II

3Fe(II)' 3FZ Fe FZk −+ → (A3-1)

Fe(II)f-L II

Fe(II)' L Fe Lk

+ → (A3-2)

where Fe(II)

f-FZk and Fe(II)

f-Lk are bimolecular rate constants for the Fe(II) complexation by

FZ and SRFA respectively. The Fe(II)

f-FZk has been previously determined to be 3.1 × 1011

M-3

.s-1

in 0.1 M NaClO4 (Thompsen and Mottola, 1984) and 2.0 × 1011

M-3

.s-1

in

seawater (Lin and Kester, 1992). The effect of solution pH on Fe(II)

f-FZk is small in the

range of pH 3 to 8. The value of 3.1 × 1011

M-3

s-1

was used in this work as the ionic

strength was closer to that of this work (assuming that Fe(II)

f-FZk value is not influenced by

the presence of light). In the presence of 1 mM FZ, the calculation using the rate law

equation for the reaction shown in eq. A3-2 indicates that 99.9% of inorganic Fe(II)

can be complexed by FZ within < 0.03 s. In addition, the complexation rate for FeIIFZ3

( Fe(II)

f-FZk [FZ]3 = 310 s

-1) is calculated to be much larger than the complexation and


211

oxygenation rates for FeIII

SRFA ( Fe(II)

f-Lk [L] = > ~2 s-1

and kox[O2] = 0.036 s-1

) under the

condition examined. The dissociation of the FeIIFZ3 complex is relatively slow with a

first-order rate constant of 4.3 × 10-5

s-1

(Thompsen and Mottola, 1984). Given that FZ

possess such a high affinity to Fe(II), change of FeIIFZ3 concentration due to the

complex dissociation is negligible on a timescale examined in this work.

Under this condition, therefore, it is reasonable to assume that FeIIFZ3 formation rate is

equal to the Fe(II) formation rate. In addition, observed FeIIFZ3 concentration

increased in a first-order manner with respect to time regardless of the absence and

presence of light. The Fe(II) formation rates for dark and light conditions, respectively,

can be described as follows;

IIIII3

dark

[Fe FZ ] [Fe(II)][Fe L]

d dk

dt dt= = (A3-3)

IIIII3

light

[Fe FZ ] [Fe(II)][Fe L]

d dk

dt dt= = (A3-4)

where kdark and klight are first-order rate constants for the SRFA-mediated thermal

reduction of FeIII

SRFA and photochemical reduction rate of FeIII

SRFA, respectively.

Approximations [FeIII

L] = [FeT] – [FeIIFZ3] (due to the low concentration of Fe'

compared to total Fe [FeT]) and [FZ] ≈ [FZT] (due to the relatively high total FZ

concentration [FZT]) followed by the integration gives relationship between [FeIIFZ3]

and time, as follows;

Tdark/lightII

T 3

[Fe ]ln

[Fe ] [Fe FZ ]k t

= ⋅

− (A3-5)

where kdark/light represent either kdark or klight. The first-order rate constant for the thermal

reduction (kdark) for SRFA was then determined to be 1.3 (±0.0) ×10-6

s-1

from a slope

of linear regression line in plots of time versus ln([Fe]T/([Fe]T-[Fe(II)FZ3])). Similarly,

the photoreduction rate constant (klight) was determined to be 1.6 (±0.02) ×10-5

s-1

for

SRFA, 6.5 (±0.25) ×10-6

for EDTA and 3.2 (±0.03) ×10-5

for citrate (Figure A3.7) by

using the corrected FeIIFZ3 concentration by subtracting the Fe

IIFZ3 concentration


212

measured under dark. The determined rate constants were listed in Table A3.1. The

effect of pH on Fe(III) reduction was also examined for SRFA system by using the

identical experimental and analytical procedures except that pH of solutions such as

Frqauil* and Fe

IIISRFA was adjusted to 6-8 and higher total Fe (10 µM) and SRFA

concentrations (50 mg.L-1

) with Varian Cary 50UV-Vis spectrophotometer (1 cm

cuvette) were employed. As shown in Figure A3.8, the FeIIFZ3 formation rate

increased with decrease in pH in both dark and light conditions, indicating that Fe(III)

reduction is facilitated in acidic pH. The rate constants determined by using eq. A3-5

were listed in Table A3.2.

Figure A3.7. Determination of rate constants for photo-reduction of Fe(III)-ligand

complexes in Fraquil* (pH 8). The experimental conditions, symbols and error bars are

identical to those in Figure A3.6, except that the data measured under the light were

only shown. The solid lines represent linear regression lines in each ligand system.

y = 0.00039x - 0.0041

R2 = 0.99

y = 0.0019x + 0.012

R2 = 0.99

y = 0.0010x + 0.037

R2 = 0.95

0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0 50 100 150 200 250 300

Time (minutes)

EDTA

Citrate

SRFA

ln([

Fe] T

/([F

e] T

-[F

eF

Z3])


213

Figure A3.8. Effect of pH on reduction of FeIII

SRFA under the light and dark.

0

1

2

3

4

5

6

7

8

9

10

0 2 4 6 8

Fe

(II)

FZ

3fo

rma

tio

n (

μM

)

Time (hr)

(A) pH 6, [SRFA] = 50 mg/L

0

1

2

3

4

5

6

7

8

9

10

0 2 4 6 8

Fe

(II)

FZ

3fo

rma

tio

n (

μM

)

Time (hr)

(B) pH 7, [SRFA] = 50 mg/L

0

0.5

1

1.5

2

2.5

3

0 2 4 6 8

Fe

(II)

FZ

3fo

rma

tio

n (

μM

)

Time (hr)

(D) pH 9, [SRFA] = 50 mg/L

0

1

2

3

4

5

6

7

8

9

10

0 2 4 6 8

Fe

(II)

FZ

3fo

rma

tio

n (

μM

)

Time (hr)

(E) pH 6, [SRFA] = 50 mg/L

0

1

2

3

4

5

6

7

8

9

10

0 2 4 6 8

Fe

(II)

FZ

3fo

rma

tio

n (

μM

)

Time (hr)

(F) pH 7, [SRFA] = 50 mg/L

0

1

2

3

4

5

6

7

8

9

10

0 2 4 6 8

Fe

(II)

FZ

3fo

rma

tio

n (

μM

)

Time (hr)

(H) pH 9, [SRFA] = 50 mg/L

0

0.5

1

1.5

2

2.5

3

3.5

4

4.5

0 2 4 6 8 10

Fe

(II)

FZ

3fo

rma

tio

n (

μM

)

Time (hr)

(C) pH 8, [SRFA] = 50 mg/L

0

0.5

1

1.5

2

2.5

3

0 2 4 6 8 10

Fe

(II)

FZ

3fo

rma

tio

n (

μM

)

Time (hr)

(G) pH 8, [SRFA] = 50 mg/L


214

Table A3.1. Reduction rate constants for organically complexed Fe(III) in Fraquil*

(pH 8).

Ligand Thermal reduction rate constant Photoreduction rate constant

kdark (s-1

) klight (s-1

)

SRFA 1.3 (±0.0) ×10-6

1.6 (±0.02) ×10-5

EDTA N.Da)

6.5 (±0.25) ×10-6

Citrate N.Da)

3.2 (±0.03) ×10-5

a) Not determined.

Table A3.2. Reduction rate constants for organically complexed Fe(III) in Fraquil*

(pH 6-9)a)

.

Ligand pH Thermal reduction rate constant Photoreduction rate constant

kdark (s-1

) klight (s-1

)

SRFA 6 3.6 ×10-5

1.3 ×10-4

SRFA 7 3.2 ×10-5

8.3 ×10-5

SRFA 8 5.5 ×10-6

1.4 ×10-6

SRFA 9 7.8 ×10-6

2.5 ×10-5

a) The Fe reduction experiment was performed at concentrations of 10 µM for Fe, 50

mg.L-1

for SRFA and 1mM for FZ.


215

REFERENCES

GARG, S., ROSE, A. L. & WAITE, T. D. 2012. Pathways contributing to the

formation and decay of ferrous iron in sunlit natural waters. In: TRATNYEK,

P. G., GRUNDL, T. J. & HADERLEIN, S. B. (eds.) Aquatic Redox Chemistry.

Washington: American Chemical Society.

LIN, J. & KESTER, D. R. 1992. The kinetics of Fe(II) complexation by ferrozine in

seawater. Marine Chemistry, 38, 283-301.

PHAM, A. N., ROSE, A. L. & WAITE, T. D. 2012. Kinetics of Cu(II) reduction by

natural organic matter. Journal of Physical Chemistry A, 116, 6590-6599.

PULLIN, M. J. & CABANISS, S. E. 2003. The effects of pH, ionic strength, and iron-

fulvic acid interactions on the kinetics of nonphotochemical iron

transformations. II. The kinetics of thermal reduction. Geochimica et







Documents

Kinetics of Iron Uptake by the Freshwater Cyanobacterium