Identifying Toxigenic Algae Using

RNA-Based Molecular Technologies

Dr. Dianne I. Greenfield Environmental Sciences Initiative, CUNY Advanced Science Research Center

School of Earth and Environmental Sciences, Queens College

Advantages of Quantitative

Molecular Technologies

• Microscopy is time-consuming and many species look alike

• Molecular approaches often faster, enable species or gene-

specific ID and quantification, ‘early warnings’

• Examples: DNA - polymerase chain reaction (PCR), rRNA -

sandwich hybridization assay (SHA) for species, protein-

based (ELISAs and others) for toxin

Why rRNA?

(Winnebeck et al. 2010)

• High numbers in cell

• Species-specific sequences

• Characterizes live

organisms

• Transcribed as single operon

Sandwich Hybridization Assay (SHA)

Target

rRNA

Capture probe

Signal probe

DIG

HRP Substrate

A B C

H G F E D

Advantages Rapid (~1 hr); multiplex (up to

12 rxns); species or group-

specific IDs; cost-effective

Scholin et al. 1999; Goffredi et al. 2005; Greenfield et al. 2006, 2008; Doll et al. 2014, Main et al. 2014, 2018

>1 Species together

(PN, Alex, etc.)

Harmful Algae Invertebrate Larvae

Examples of Target Organisms

Balanus glandula

(Acorn barnacle)

Pseudo-nitzschia spp.

Heterosigma akashiwo

Marine Microbes

Carcinus maenus sp.

(Green crab)

Roseobacter

Cytophaga

SAR86

Pelagibacter

Picophytoplankton

Marine Group I/II Archaea

Marine Delta

OM60/KTC1119

S-oxidizing symbionts

Osedax

Karenia spp.

Mytilus sp.

(Shore mussels)

Polychaete

Alexandrium catenella

Cochlodinium polykrikoides

Sciaenops ocellatus

Hilton

Head

Charleston

Myrtle

Beach

Ideal for Detecting Multiple HABs

• Example: Coastal SC

• 1,300+ events since 2001

− ~430 FKs, 1 in 4 HAB-

related

− Raphidophytes &

cyanobacteria are most HABs

• Primarily urban regions

Maximum C. subsalsa:

~40,000 cells/ml

0

5

10

15

20

25

# E

ven

ts

Species

1 Year: summer 2014-2015 • Most common bloom and fish

kill species = Chattonella

subsalsa (~30%)

• Raphidophytes ~41% combined

bloom + FK

• Next = cyanobacteria (~55%

blooms)

• Remainder = Pseudo-nitzschia,

dinoflagellates, euglenas, others

Multiple Causative HABs

SHA applications developed for many of those species

New SHA for Microcystis spp.

Cyanobacteria: largest #HABs worldwide; Microcystis is the

most common genus. Enhances early warnings for blooms to

safeguard public health, prediction, and management

Capture Probe Design

• 16s DNA GenBank® sequences, ≥1,000 bp length

• Within 250 bp of signal probe

• GC content at least 40%

Microcystis

Outgroups

Considerable genetic similarity among Microcystis strains

No cross-reactivity with non-target species

Field Sampling • Southeast among the most rapidly

growing regions

• >21,000 stormwater ponds

• Shallow, high residence times,

stagnate, accumulate nutrients

Numerous HABs and fish kills,

high likelihood of public contact.

55% of these HABs are cyanos!

Field Sampling

0

1

2

3

4

5

6

7

8

9

29-A

pr

2-M

ay

9-M

ay

11-M

ay

13-M

ay

16-M

ay

15-J

un

17-J

un

22-J

un

24-J

un

28-J

un

6-J

ul

11-J

ul

14-J

ul

20-J

ul

26-J

ul

2-A

ug

Lo

g1

0 c

hl a

(μ

g L

-1)

an

d c

elll

s m

l-1

Date (2016) Anabaena Microcystis Anabaenopsis <1 ppb 1 ppb 2 ppb

Highest toxin, Microcystis.

SHA detected cells on all

+Microcystis dates

Elevated DON,

DOP

Upscaling Temporal Resolution:

Environmental Sample Processor (ESP)

• Enables near-real time in situ

SHA and protein microarrays,

sample archival, qPCR

• Surface, mid- and 4K depth

configurations, AUV

• Partner with other sensors

control

Alexandrium tamarense/catenella

Pseudo-nitzschia multiseries

P. pseudodelicatissima/ multiseries

P. australis

Heterosigma akashiwo

May 17 May 25

June 4

ESP

Field

Deployment

Monterey Bay, CA May 17-June 11, 2007

In situ Detection

of Harmful Algae

May 21 May 29

1L Sample Volume

Greenfield et al. 2006, 2008 L&O: Methods

?

Adaptable for non-HAB taxa

Sciaenid Spawning in SC Rivers

and Estuaries Month

J F M A M J J A S O N D

Red drum

Atlantic Croaker

Spot croaker

Black drum

Silver perch

Banded drum

Weakfish

Southern kingfish

Spotted sea trout

Star drum

Silver sea trout

Gulf kingfish

Similarities:

• Concentrating a sample

• Lysing cell membranes

• Using DNA probes to identify sequences

• Quantification of genetic material

SHA and qPCR

Differences:

SHA qPCR

NA Extraction? No Yes

Detection mode Direct Amplified product

Genetic target Large subunit rRNA DNA

Quantification Absorbance Fluorescence emission

MERHAB: Methods ‘Bake Off’

• Sedgewick rafter as

“gold standard”

(Godhe et al. 2007)

• 9 counts per sample

collection

• Multiple filters

with specified

cell number

• Flash-frozen (N2)

• Add lysis buffer

• Heat, combine

lysate, filter

Same

homogenate:

qPCR and

SHA (96-well

plate)

Photo: Tom Murphy

www.ifremer.fr

• Globally-distributed euryhaline HAB: causes

fish kills and declining water quality

• Validated SHA and qPCR methods

• Low global diversity in non-chloroplast

genome

Study organism:

Heterosigma akashiwo

0.0

0.5

1.0

1.5

2.0

OD

(450 n

m)

Control Room (23 °C) 4 °C

*

*

0.0

1.0

2.0

3.0

4.0

5.0

0 h 1 h 24 h 7 d 30 d 90 d 120 d

∆∆

CT

Time After Preservation

*

*

*

Calibration and Preservation

Doll et al. 2014, L&O Methods

Calibration • Geographically distinct H.

akashiwo strains exhibited

variability, but it was minor

• SHA and qPCR were nearly

identical; SHA had higher pre-

bloom sensitivity, qPCR had

wider range (pre-dilution)

Preservation • T and assay type influenced

quantification

Bloom Assessment and Prediction

• Delaware Inland Bays

(DIBs), May-Aug 2015

• Max: 3 x 106 cells/L, 18-

May; 1 x 106 cells/L 8-June;

both Rehoboth Bay (RB64)

Main et al. 2018, J. Applied Phycology

Great agreement at bloom concentrations – but…below and

unlike lab findings – qPCR overestimated H. akashiwo. WHY?

qPCR vs. SHA

R^2 = 0.845 at >10,000 cells/L (bloom)

Main et al. 2018, J. Applied Phycology

Bloom Assessment and Prediction

Nutrients and co-occurring phytoplankton H. akashiwo abundances had no real pattern associated with N-form but positively

and significantly (p < 0.01-0.001) correlated with Si and P

qPCR:SHA elevated at high Karlodinium veneficum (red) lower at high Prorocentrum

minimum (orange); overall concentrations did not correlate with H. akashiwo.

Main et al. 2018, J. Applied Phycology

T and phytoplankton biomass

• Strong agreement between

methods <25 oC

• Most H. akashiwo blooms

occur in this T-range,

suggesting thermal stress

• Greater agreement >30 mg/L

Chl a, consistent with Handy

et al. (2005) showing greater

qPCR accuracy with mixed

communities

• Outliers >30 mg/L were

during late blooms – cell

senescence?

Main et al. 2018, J. Applied Phycology

Regional Applications

• Multiple regional HAB species and shellfish toxins

• Recent Pseudo-nitzschia blooms

• Active toxin surveillance New England and NY areas

• Several SHA protocols (Alexandrium, Margalefidinium [ Cochlodinium], etc

Acknowledgments

SC and NY Labs: Lara Brock, Nicole Dearth, Cameron Doll, Jessica

Espinosa, Suzanne Kacenas, Chuck Keppler, Krystyn Kibbler, Dominique

Maldonado, Bec Mortensen, Reima Ramsamooj, Michelle Reed, Kimberly

Sitta, Shawn Stormer,

Funding: SC Sea Grant Consortium, CDC, USC and CUNY Research

Foundations, State-appropriated funds, NOAA, EPA, Gordon and Betty Moore

Foundation, Private contracts

ASRC link: http://environment.asrc.cuny.edu/people/dianne-greenfield/