Evaluation of Adherence, Hydrophobicity, Aggregation, and Biofilm Development of Flavobacterium johnsoniae-Like Isolates

Evaluation of Adherence, Hydrophobicity, Aggregation, and Biofilm Development ofFlavobacterium johnsoniae-Like IsolatesAuthor(s): A. Basson, L. A. Flemming and H. Y. CheniaSource: Microbial Ecology, Vol. 55, No. 1 (Jan., 2008), pp. 1-14Published by: SpringerStable URL: http://www.jstor.org/stable/25153434 .

Accessed: 08/08/2013 08:11

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

Springer is collaborating with JSTOR to digitize, preserve and extend access to Microbial Ecology.

http://www.jstor.org

This content downloaded from 146.230.128.27 on Thu, 8 Aug 2013 08:11:51 AMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=springer

http://www.jstor.org/stable/25153434?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


Microbial

I_Ecology

Evaluation of Adherence, Hydrophobicity, Aggregation, and Biofilm Development of Flavobacterium Johnsoniae-L\ke Isolates

A. Basson, L. A. Flemming and H. Y. Chenia

Department of Microbiology, University of Stellenbosch, Private Bag XI, Matieland, 7602, South Africa

Received: 18 January 20071 Accepted: 6 March 20071 Online publication: 31 March 2007

Abstract

Flavobacterium spp. isolates have been identified in diverse biofilm structures, but the mechanism of adher ence has not been elucidated. The absence of conven

tional biofilm-associated structures such as fimbriae, pili, and flagella suggest that surface hydrophobicity, and/or

autoaggregation and coaggregation may play an impor tant role in adherence and biofilm formation. The

biofilm-forming capacity of 29 Flavobacterium johnso niae-like isolates obtained from South African aquacul ture systems was assessed using microtiter plate assays. The role of hydrophobicity [salting aggregation test

(SAT) and bacterial adherence to hydrocarbons (BATH)

assays], autoaggregation, and coaggregation on biofilm formation by Flavobacterium spp. was also investigated, while biofilm structure was examined using flow cells and microscopy. All isolates displayed a hydrophilic nature, but showed varying levels of adherence in microtiter assays. Significant negative correlations were

observed between adherence and biofilm-forming capac ity in nutrient-poor medium at 26?C and BATH

hydrophobicity and motility, respectively. Isolates dis

played strain-to-strain variation in their autoaggregation indices and their abilities to coaggregate with various

Gram-negative and Gram-positive organisms. Micro

colony and/or biofilm development were observed

microscopically, and flavobacterial isolates displayed stronger biofilm structures and interaction with a Vibrio

spp. isolate than with an Aeromonas hydrophila isolate. The role of extracellular polysaccharides and specific outer membrane proteins will have to be examined to

reveal mechanisms of adherence and coaggregation employed by biofilm-forming F. johnsoniae-like strains.

Correspondence to: H. Y. Chenia; E-mail: [email protected]

Introduction

Aquatic fish pathogens such as Vibrio, Yersinia, and Aeromonas spp., have been shown to form biofilm structures in aquaculture environments, and survival of the aquatic bacterium outside the fish host may be

dependent on biofilm formation [10, 16, 43]. Flavobac terium columnare, Flavobacterium psychrophilum, Flavo bacterium branchiophilum, Flavobacterium aquatile, Flavobacterium johnsoniae, Flavobacterium hydatis, and Flavobacterium succinicans have been associated with fish disease and have also been detected in surrounding water

during disease outbreaks [4, 5]. Various species of fish serve as hosts of bacterial infections caused by these

pathogens, and great losses of fish in aquaculture farming worldwide has led to an increasing interest in the

pathogenicity of Flavobacterium species [23].

Apart from causing fish disease, various Flavobacterium

spp. isolates have also been identified in biofilms from domestic drain conduits [26] and potable dental lines [35].

Oppong et al. [32] reported the presence, within biofilm structures associated with paper mill slimes, of Flavobacte rium spp. showing 92% 16S rDNA sequence identity to fish

pathogenic Flavobacterium isolates. Biofilm formation begins with the initial attachment

of the bacterium to a surface followed by the formation of microcolonies and leading to the maturation of the microcolonies into three-dimensional structures, enclosed and stabilized by extrapolymeric substances (EPS). Adher ence is a complex multistep process that can be subdivided into the stages of attraction, adhesion, and aggregation [30]. Traditionally, analysis of biofilms have indicated that cell surface structures such as pili, flagella, fimbriae, and outer membrane proteins (OMPs) and EPS play an integral role in enabling microorganisms to initially adhere to a surface and to form biofilms. Many of the cell surface structures are important components of bacterial motility,

DOI: 10.1007/s00248-007-9245-y Volume 55, 1-14 (2008) ? Springer Science + Business Media, LLC 2007 1



2 A. Basson et al.: Adherence and biofilm formation of F. johnsoniae-uke isolates

and nonmotile mutants have been shown to lack biofilm

forming ability, compared to wild-type cells [14, 30]. Properties, which have been found to enhance bacterial

attachment, include increased whole cell hydrophobicity and the ability to coaggregate and autoaggregate [13, 35]. Surface hydrophobicity is generally associated with bacte rial adhesiveness and varies from organism to organism, from strain to strain and is influenced by the growth medium, bacterial age, and bacterial surface structures

[38, 42]. Coaggregation is a process by which genetically distinct bacteria become attached to one another via

specific molecules. Coaggregation between pairs of bacte ria is highly specific and is typically mediated by a protein "adhesin" on one cell type and a complementary saccharide "receptor" on the other [34]. A larger proportion of biofilm strains, are able to coaggregate and autoaggregate compared to their planktonic counter

parts [35]. Cell surface properties, along with factors such as the influence of surface-active compounds secreted by the microorganisms, the hydrodynamic surroundings, surface roughness, availability of nutrients, differences in shear rate [46], profoundly affect the final structure of the biofilms and biofilm-associated microbial diversity and

may play an important role in the potential of opportu nistic pathogens to integrate into freshwater biofilms [35].

Aquaculture systems provide the ideal habitat for

biofilm-forming pathogenic bacteria, for not only is there a rich flow of nutrients and close proximity to the hosts, but a variety of surfaces amenable for bacterial coloni zation are also provided. Biofilm formation is not only an important stage in the pathogenicity of organisms [3], but biofilm establishment on host tissue or inanimate surfaces inhibits effectiveness of antimicrobial therapy, protects against host defense mechanisms and facilitates bacterial communication leading to the expression of virulence determinants [21]. This then provides an ideal environment and niche for microorganisms to instigate outbreaks of disease or the recurrence of infection in

aquaculture environments [10, 19]. Flavobacterial isolates have been reported to adhere

to the gill epithelium tissue and to the scales and fins of the fish [12]. Flavobacterium spp. isolates do not possess the conventional motility structures, namely, flagella or

typical type IV pili and are motile over solid surfaces

using gliding motility [27]. Non-spreading mutants of F.

johnsoniae were reported to possess a cell surface which is less hydrophobic than that of the wild-type parent strain,

suggesting that cell-surface hydrophobicity is involved in the gliding motility of this organism [17] and adherence for colonization. Adhesion of F. columnore may be related to the quantity of polysaccharide constituents

present in its capsule, while a lectin-like carbohydrate binding substance was shown to be responsible for the adhesion of the bacterium to fish tissue [12, 13]. It is

likely that factors such as the presence of adhesins and a

capsule and surface hydrophobicity play an integral role in enabling the bacterium to attach to a biotic or abiotic surface and to form biofilms.

Although Flavobacterium spp. are important patho gens in the aquaculture setting [4] and have been isolated from industrial, domestic, and medical environment biofilms [26, 32, 35], the manner in which they form biofilms has not been elucidated. Given that aquaculture surfaces are easily colonized [10, 19, 43], persisting Flavobacterium spp. inhabiting biofilms might serve as a source of infection or reinfection. The present study thus

investigated the ability of F. johnsoniae-like isolates, isolated from diseased fish and aquaculture tank bio

films, to form biofilms using microtiter-plate adherence

assays and flow cells. In addition, the roles of hydropho bicity, autoaggregation, and/or coaggregation in Flavo bacterium spp. biofilm formation were examined.

Methods

Isolation and Maintenance of Study Isolates. Twenty nine F. johnsoniae-like isolates obtained from a variety of diseased fish species, including rainbow trout, koi, and

longfin eel from various aquaculture farms situated

throughout South Africa during the period 2003-2005, and previously characterized with respect to their

phenotypic and molecular diversity [15] were selected for study.

Biofilm-forming ability. All F. johnsoniae-like isolates were cultured overnight in enriched Anacker and Ordal's broth (EAOB) and their microtiter plate adherence determined [39]. Cells were washed and

resuspended in phosphate-buffered saline (PBS, pH 7.2) to a turbidity equivalent to a 0.5-M McFarland standard.

Wells of sterile 96-well U-bottomed microtiter plates (Deltalabs S.L, Barcelona, Spain) were each filled with 90 ul EAOB/tryptic soy broth (TSB; Merck Chemicals,

Gauteng, RSA) and 10 ul of each cell suspension, in

triplicate [39]. Negative control wells contained only broth and a Vibrio sp. isolate was used as a positive control. Plates were placed on a CI platform shaker (New Brunswick

Scientific, Edison, NJ, USA) and/or the benchtop to simulate dynamic and static conditions, respectively, and incubated aerobically at room temperature (21 ?2?C) and/ or 26?C for 24 h, in either EAOB/TSB. Contents of each well were aspirated, washed three times with 250 ul sterile PBS and the remaining cells were fixed with 200 ul of methanol for 15 min. After air-drying, wells were stained with 150 ul of 2% Hucker's crystal violet for 5 min. Dye bound to

adherent cells was resolubilized with 150 \il of 33% (v/v)

glacial acetic acid, and the optical density (OD) of each well was obtained at 595 nm using an automated

microtiter-plate reader (Microplate Reader model 680, BioRad Laboratories Inc., Hercules, California). Tests were



A. Basson et al.: Adherence and biofilm formation of F. johnsoniae-uke isolates 3

done in triplicate on three separate occasions and the results averaged [39]. The cutoff OD (ODc) for the

microtiter plate test was defined as three standard deviations above the mean OD of the negative control. Isolates were classified as follows: OD<ODc

= non

adherent, ODc < OD < (2 x ODc) = weakly adherent;

(2 x ODc) < OD < (4 x ODc) = moderately adherent and

(4 x ODc) < OD = strongly adherent [39].

A second method of assessing the biofilm-forming capacity for each Flavobacterium spp. isolate was

expressed relative to the average value of all isolates as follows:

Relative biofilm capacity

/" 29

J2(An-A0)/29 ,

where Ax = absorbance at 595 nm for isolate x, A0

= ab

sorbance for uninoculated growth medium [40].

Bacterial Hydrophobicity. Surface hydrophobicity was assessed using the bacterial adherence to

hydrocarbons (BATH) and modified salting aggregation test (SAT) assays and Congo Red accumulation.

Bacterial Adherence to Hydrocarbons. Bacteria

grown in EAOB were harvested during the exponential growth phase (18-h-old cultures), washed three times and resuspended in sterile 0.1 M phosphate buffer (pH 7) to an OD of 0.8 at a wavelength of 550 nm (A0 of 108 CFU/

ml), using a DU 640 spectrophotometer (Beckman Coulter, Fullerton, California, USA). Samples (3 ml) of bacterial suspensions were placed in glass tubes with 400 ul of the hydrocarbon, xylene (BDH, VWR International, Leicestershire, UK) equilibrated in a water bath at 25?C for 10 min and vortexed [36]. After a 15-min phase separation, the lower aqueous phase was removed and its

OD550 determined (Ax). Values were then expressed as the percentage of bacteria adhering to hydrocarbon (A) compared with the control suspension as follows:

A=[(A0-A1)/A0] x 100. Each value represents the mean of experiments done in triplicate and on two

separate occasions [9, 36]. Strains were considered

strongly hydrophobic when values were >50%,

moderately hydrophobic when values were in the range of 20-50%, and hydrophilic when values were <20%

[25]. PBS was used as a negative control, and a Vibrio

spp. isolate was used as a positive control.

Salting Aggregation Test. Overnight EAOB cultures were harvested, washed twice, and resuspended in PBS (pH 7.2). Isolates were "salted out" (aggregated)

by combining 25 ul volumes, containing 2 x 109 bacteria, with 25 ul volumes of a series of ammonium sulfate

[(NH4)2S04] concentrations (0.2, 0.5, 1, 1.5, 2, 2.5, 3, and 4 M) on microscope slides [38]. Addition of 400 ul methylene blue solution to 10 ml volumes of

(NH4)2S04 solution [37] facilitated better visualization of aggregation. After 2 and 4 min of mixing, respectively, on a rocking shaker at ambient temperature, slides were

visually examined and scored against a white background. The lowest final concentration of (NH4)2S04 causing aggregation was recorded as the SAT value. Experiment were done in triplicate on two separate occasions, and

respective (NH4)2S04 concentrations were used as nega tive controls [38]. Classification proceeded as follows: <0.1

M = highly hydrophobic, 0.1-1.0 M =

hydrophobic, and >1.0 M =

hydrophilic [29].

Congo Red Accumulation. F. johnsoniae-like isolates were streaked out onto enriched Anacker and

Ordal's agar (EAOA) plates containing 0.003 g/1 Congo Red (Merck Chemicals, Gauteng, RSA) and incubated at 26?C. Congo red accumulation was recorded over a

period of 2-5 days [11].

Capsule Staining. Capsule presence was tested

using the Anthony Capsule Staining procedure [18]. All slides were examined microscopically (x 1,000

magnification, Zeiss light microscope) under oil immersion. A laboratory strain of Klebsiella pneumoniae

was used as a positive control.

Autoaggregation and Coaggregation Assays. Twelve F. johnsoniae-like isolates (Y012, Y015, Y019, Y034, Y045, Y051, Y053, Y059, YO60, Y063, Y064, and

Y067) were examined for their ability to autoaggregate and coaggregate with the following bacterial partner strains, i.e., Aeromonas hydrophila, Aeromonas sobria, Aeromonas

salmonicida, Aeromonas media, Edwardsiella tarda, Salmonella enterica serovar Arizonae, Chryseobacterium spp., Myroides spp., Acinetobacter spp., Staphylococcus aureus ATCC 25923, Enterococcus faecalis ATCC 2912, Listeria monocytogenes, Listeria innocua LMG 13568, Pseudomonas aeruginosa, and Micrococcus spp. Bacteria

were grown in 20 ml EAOB or TSB, harvested after 36 h, washed and resuspended in sterile distilled H20 to an OD of 0.3 at a wavelength of 660 nm [24].

The degree of autoaggregation of all isolates was determined using the equation:

OD0 -

OD60 % Autoaggregation =-?-x 100,

OD0 refers to the initial OD of the organism, while

OD60 was obtained after 60 min at room temperature following centrifugation at 2,000 rpm for 2 min [24].



4 A. Basson et al.: Adherence and biofilm formation of F. johnsoniae-uke isolates

The degree of coaggregation was determined for

paired isolate suspensions (500 ul of each isolate) [24],

using the equation:

ODTot -

ODs %Coaggregation

=-??-x 100, ODTot

ODTot value refers to the initial OD immediately after pairing of isolates; and ODs refers to the OD of

supernatant after a period of 60 min at room tempera ture [24]. Experiments were carried out in triplicate on two separate occasions.

Characterization of Biofilm Formation Using Flow Cell

Systems. Biofilm formation by three non-adherent

(Y012, Y051, and YO60) and two adherent (Y019 and

Y064) isolates was investigated using continuous culture

once-through eight channel flow-cell systems [44]. A Vibrio spp. strain was used as a positive control for biofilm formation as well as being used with an A. hydrophila isolate obtained from diseased trout, to examine mixed bacterial biofilm growth, in combination with flavobac terial isolates, Y019 and Y051, respectively.

The eight-channel perspex flow cell (channel size 30 x 4.5 x 3 mm), the glass cover-slip covering (no. 1

thickness, 75 mm x 50 mm), and attached silicone

tubing (1 x 1.6 mm x 3 mm x 5 m tubing; The Silicone

Tube, RSA) was assembled as described by Wolfaardt et

al. [44]. Silicone tubing was connected to a reservoir

containing 2 1 of EAOB, and the flow cell was filled with

EAOB, with a flow rate of 0.3 ml/min being maintained

using a multichannel peristaltic pump (Model 205S, Watson-Marlow, UK) located upstream of the flow cell. One millimeter volumes of EAOB overnight cultures of selected flavobacterial isolates, the Vibrio spp. and A.

hydrophila isolates, respectively, were inoculated into each channel, below the clamps sealing silicone tubes

upstream of each channel, using sterile syringes. One millimeter mixed pure culture inoculations, consisted of 0.5 ml combinations of either isolate Y019 or Y051 with 0.5 ml of either the Vibrio spp. or A. hydrophila isolate,

respectively. Stagnant conditions were maintained for the first hour to allow attachment, before inoculated chan nels were exposed to flowing EAOB at a constant flow rate of 0.3 and 0.6 ml/min (isolate Y019 only), respec

tively. Flow cell systems were kept at room temperature (21?C?2) throughout the experiments. Each flow cell channel was investigated by transmitted light using a

Nikon Eclipse E400 (Nikon, Japan) microscope at 600

fold magnification and after 24 and 48 h, respectively, to

visualize bacterial attachment to a glass surface and

biofilm development. Images were documented with a

CHU high-performance charge-coupled camera device

(model 4912-5010/000).

Statistical Analyses. Statistical significance of dif ferences (p<0.05) due to altered variables (temperature,

medium, agitation) in the microtiter adherence assays were determined using repeated measures analysis of variance (ANOVA); the means were separated by a

Bonferroni least significant difference test using Statistica 7 (Statsoft, Tulsa, USA). Relationships between microtiter

adherence, hydrophobicity (BATH and SAT assays, respectively), autoaggregation, motility, and/or proteolytic activity [15] were determined by 2D scatterplot analysis (p value s < 0.05 were considered significant), using Statistica.

Results

Microtiter adherence assays. All 29 isolates were

screened for adherence to polystyrene microtiter plate wells following incubation for 24 h at 26?C or room

temperature (21?C?2), under static or dynamic conditions in nutrient-rich (TSB) or nutrient-poor (EAOB) media (Table 1; Fig. 1A and B), respectively.

Variation in biofilm formation (Table 2) was observed

among isolates cultured with EAOB, with values ranging from 0.08 (isolate Y049) to 1.06 (isolate Y053), and

among the isolates cultured with TSB, with values

ranging from 0.07 (isolate YO50) to 1.10 (isolate Y053), respectively (p<0.05). Biofilm formation and relative biofilm-forming capacity <0.1, respectively, were

considered a negative result; thus, 16/29 (55.2%) of isolates were considered to be non-adherent in the

presence of either EAOB or TSB. Of the 44.8% (13/29) of isolates displaying adherence, 69.2% (9/13) were able to adhere with both nutrient-poor and nutrient-rich

medium, while 23.1% (3/13) and 7.7% (1/13, isolate

Y053) of isolates adhered in EAOB or TSB alone,

respectively (Fig. 1A and B). Significant differences

(p<0.05) in adherence were observed when isolates were assayed under static/dynamic conditions and at room temperature/26?C only when nutrient-rich and

nutrient-poor media were used (Table 1). Agitation and temperature variation did not significantly affect biofilm formation if the medium was unchanged (p>0.05). Majority of the isolates had a relative biofilm

forming capacity that was significantly (p<0.05) lower than that demonstrated by the Vibrio spp. isolate (Table 2; Fig. 1A and B). While the Vibrio spp. and flavobacterial isolates preferred EAOB to TSB for

adherence, 23.1% (3/13) of biofilm-forming isolates were most strongly adherent in both TSB and EAOB,

respectively (Table 2; Fig. 1A and B). Significant negative correlations were observed between adherence and/or

biofilm-forming capacity at both room temperature and 26?C (static and dynamic) and motility (EAOB 26?C, r = -0.43 to -0.51, p

= 0.01; EAOB room

temperature, r = -0.39 to -0.48, p =

0.01-0.04).




d I I I I l|m(N(N(N(NM(N(N| ? H Qoooooooog JS ^,^006060660

? sf +1 +1 +1 +1 +1 +1 +1 +1 +1 -?

*g -^lOoooooooog

.S* ^0 O O OS vo o -o * S sk-S?C ? O r? h5o>OSVOmON00VOVOr-j

fe Q00000000.2 n3 & ?o o 6 6 6 6 6 6 6 ro

fl Cf +| +1 +1 +1 +1 +1 +1 +1 +1 g

cj ^lO 60606606 So o H

2 ^ .y

ti t>o ^ K [>.' 6 6 6 6 6 ? L?,

g co I 2ir-imcorocovomm^ ^ LIT

S I K-v^^^ fO Tf ^ .5 O Q ? ? ? 00 ?a 8 & P0 ?' ?* ?' ?* ?' ^l?

Z ̂ +1 +1 +1 +1 +1+1 +1 G

^ M CS CN CN CS CS h

y ^dlOoooooooohe

+j I I I I &o

fl ?\ Le o ^ ^

ts S _^ U-^VOOs" O^f^ OS O d >H fc ^ cs "^ CO (N ^ +1

fl ? > ~ ~~ X .5 -<i I <'-HOOCNOOSVOO<Nt? *> I M *Cj I t?I ,_IT-Hl_I,_l.l.l.lL^ ? QOOOOOOOOg ?d ^h^ 66666666 ,rt

?g ^ +1 +1 +1 +1 +1 +1 +1 +1 +1 "3 ^-v ^/-vOCNOr-HfSIOr-H^HO

CN ÎvJOOOOOOOOW 11 Lb

3 ^ ^ \ 9 'o ?2 v>^^ô^vqr^co^.? ?s *rt .. ôoinriin ocjirioscskj

i vh I jy ? ? N S. .J2 . _ <u

?"? Q ?? ? ? ? & Is &r? dd ?" ?' ? rs O ^ Sf +| +1+1 +1 +1 +1 -

g w >Q 00 o o o 2i 3 2 8 V ^ I O 1 1 6 6 1 6 6 6 hS ? W l? .8 o u s~^

L^ w +* d

^fl^sl? ^ ^?^ ^ ^ ^ - ^ S ^ 2 Kl ^ ^^

wÛ g s <i I oq I <i I < 1 I fo^ 1 -h ro co L3

is ^ >

S ^ I o u I o w S y '? y c ^ ̂ (3 .H S cs .y *S o

eH a +3 n .. a +j q n .h

.2 .y pQpQ'dtflpqpq'd1?g PQV UOOcQfflOOcQW L^^

? ? ? UUUUUUUU -8^ ^+i v-ooooooooPQO-i Hfll Î(N(NMM(N(NM(N| >i

Bacterial Adherence to Hydrocarbons. Overall, the BATH values for isolates ranged from 7.8-27.1%,

with 69.0% (20/29) being classified as hydrophilic (7.8 19.6) and 31.0% as moderately hydrophobic (20.6-27.1),

respectively (Table 2). The 13 biofilm-forming isolates

(Y019, YO20, Y021, Y026, Y034, Y035, Y038, Y052, Y053, Y056, Y059, Y063, and Y064) had BATH values

ranging from 9.8-20.6% (Table 2). Significant negative correlations were observed between flavobacterial ad herence to polystyrene and biofilm-forming capacity at

26?C in EAOB (static and dynamic) and BATH

hydrophobicity (EAOB, r = -0.35 to -0.37, p = 0.05).

Significant positive correlations were observed between BATH hydrophobicity and motility (r = 0.47, p

= 0.01) and proteolytic activity (r=0.39, p

= 0.04), respectively [15].

Salt Aggregation Test. Isolates were classified as

hydrophilic, with SAT values ranging from 2-4 M,

although 20.7% (6/29) of isolates did not show any

aggregation (Table 2). Using both the BATH and SAT

tests, study isolates thus displayed a hydrophilic nature. Of the biofilm-forming isolates, 46.2% (6/13) displayed no aggregation with (NH4)2S04, and 53.8% had SAT values of 2-4 M (Table 2). Although, no relationship was

observed between microtiter adherence and (NH4)2S04

aggregation (EAOB, r = -0.02, p = 0.92; TSB, r = 0.20,

p = 0.29), a significant negative correlation was observed

between SAT hydrophobicity and motility (r = -0.41,

p = 0.03).

Congo Red Accumulation. Visual examination revealed that isolates displayed varying levels of Congo Red accumulation, and this could not be correlated with BATH and/or SAT profiles.

Capsule Staining. Capsule stains revealed the

presence of capsule, although no noticeable difference in capsule thickness could be observed between flavo bacterial isolates.

Autoaggregation and Coaggregation. Autoaggregation appeared to be strain-specific and indices ranged from 12.5 43.8%. Autoaggregation did not correlate with a specific biofilm formation phenotype (EAOB, r=-0.40, p

= 0.20; TSB, r = -0.35, p

= 0.26). Autoaggregative ability was posi tively correlated with surface hydrophobicity by the BATH

assay (r=0.59, p = 0.05), however, no such relationship was

observed with the SAT assay (r=0.04, p = 0.90).

Coaggregation occurred with all 15 partnered strains to varying degrees (Table 3). Flavobacterial isolates showed overall high rates of coaggregation with S. enterica serovar Arizonae, 5. aureus ATCC 25923, E.

faecalis ATCC 2912, and to a lesser extent to, L.

monocytogenes, L. innocua, and Micrococcus spp. Flavo



6 A. Basson et al.: Adherence and biofilm formation of F. johnsoniae-like isolates

A

2.1 T-1 12.5 -i--?-1 2.0 T T ?*' 19 T 11"S T

18- 11j0' ^10.5 I

'

P"'1*7' l?10.0 I 1-6- ? 9.5 u> 1-5- . 'ff ?0

'

8 I-4' 3 *6"

1.2- ! I ? t' i fi "

' ' ' f T I III f

1 ?*9' r I ?' T

O ?-8" "" T I "I 5.(H III c ?-71

' :

' ' -

'

"5 4-5 1 L J

0.6- "

. . . . r K .. |

4.0' I

F^^^^^^^pt?^ lllllilfllllipillllllillllll ?tetv$octe^ Flm/obad&nim spp. isolates '

B 2.6-,-??-

- -?, 11*1-:-1

2.5- 10.5

H'\ T IM- T 2A'\

' 95 i 2-2" K ^ 2.1- :' '"

?> 9'?1 t

| 210- l 1 * T

m4

'

\ ml j i 8:

T " \ \ 9 7'?* I

8 1,,S ? eel . -

S 1-1- t ? 50 t

f ?8 ' '

T T ' s ?T ii

'

? "Si "'

' " _S 3.5-

' i '

***T ' "

I r 1 3-0' i "

*:** 5 2-5" '

'"''' I I T '

[' ' 1 1- '

IHIIII?ltlfll!ll?H?BI?ii? -1 ||t I Ilff||i Igf fg f Ifff If gff If f f

F7avo6actem/m spp. isolates Fl&vobacterkm spp. \s<meB

Figure 1. Biofilm formation and relative biofilm-forming capacity of Flavobacterium johnsoniae-like isolates following growth at (A) 26?C

with agitation and (B) room temperature (21?C?2) with agitation in nutrient-poor (EAOB? black bars) and nutrient-rich (TSB?grey bars) media, respectively. Biofilm formation and relative biofilm-forming capacity <0.1, respectively, were considered a negative result.

Bars represent means ? SD for three independent replicate experiments.

bacterial isolate Y051 appeared to be the weakest overall

partner in coaggregation assays (Table 3). Both strong (isolates Y019, Y034, and YO60) and weak- (isolates

Y012, and Y015) biofilm-forming isolates displayed strong coaggregation indices with a variety of different

partners (Table 3). The coaggregation assay did not,

however, identify isolates capable of forming moderate to strong- biofilm structures as being more efficient at

coaggregation than non-biofilm-forming isolates.

Flow Cell Study. Microscopy enabled the direct observation of biofilm development from initial attach ment (Fig. 2A) to the appearance of clusters on glass or

Perspex surfaces. F. johnsoniae-l\ke isolates Y012, Y019, Y051, YO60, and Y064 revealed microcolony or biofilm structures on the glass slides within 48 h (Fig. 2).

Isolate Y012 formed microcolonies with pillar-shaped filaments extending into the flowing medium 48 h postin oculation (Fig. 2B). However, no biofilm structures were

observed. Biofilm formation was not observed for isolate YO60 24 h postinoculation. Although a single micro

colony was observed at 24 h, no further microcolony formation was observed after 48 h. Biofilm formation by isolate Y051 was only observed 48 h postinoculation. Thick, multilayered biofilm production was observed (Fig. 2C) and was more prominent along the edges of the glass




Table 2. Biofilm formation, biofilm-forming capacity, and surface hydrophobicity determinations of individual Flavobacterium

johnsoniae-like isolates

Biofilm formation (OD 595 nm)a Relative biofilm-forming capacity13 Surface hydrophobicity0

Isolate Fish host EAOB TSB EAOB TSB BATH (%) SAT (NH4)2S04

Vibrio spp. - 1.9?0.05 0.18?0.01 7.66 + 0.27 0.93 + 0.14

Y012 Rainbow trout 0.13 + 0.02 0.10 + 0.01 0.24 + 0.10 0.11+0.12 23.6 + 0.03 2 + 0 Y015 Rainbow trout 0.12 + 0.01 0.10 + 0.01 0.17 + 0.02 0.11+0.01 15.4 + 0.02 2 + 0 Y019 Rainbow trout 0.37 + 0.03 0.19 + 0.06 1.23 + 0.15 1.11+0.67 9.8 + 0.01 NAd YO20 Koi 0.39 + 0.06 0.10 + 0.00 1.29 + 0.25 0.12 + 0.06 18.7 + 0.01 NA Y021 Rainbow trout 0.33 + 0.02 0.12 + 0.01 1.03 + 0.07 0.34 + 0.16 16.3 + 0.01 4 + 0 Y026 Rainbow trout 0.47 + 0.06 0.12 + 0.01 1.62 + 0.19 0.33 + 0.11 12.3 + 0.02 NA Y034 Rainbow trout 0.40 + 0.03 0.13 + 0.01 1.36 + 0.11 0.40 + 0.09 14.4 + 0.03 NA Y035 Rainbow trout 0.34 + 0.01 0.12 + 0.01 1.10 + 0.08 0.29 + 0.12 19.4 + 0.01 NA Y038 Rainbow trout 0.25 + 0.04 0.12 + 0.00 0.73 + 0.16 0.33 + 0.03 20.6 + 0.01 NA Y045 Rainbow trout 0.11+0.01 0.08 + 0.02 0.15 + 0.04 0.04 + 0.08 22.6 + 0.02 2.5 + 0.27 Y049 Rainbow trout 0.08 + 0.01 0.07 + 0.00 0.01 + 0.01 0.00 + 0.00 14.4 + 0.02 3 + 0 YO50 Rainbow trout 0.09 + 0.02 0.07 + 0.01 0.05 + 0.06 0.00 + 0.00 20.6 + 0.02 2.5 + 0 Y051 Rainbow trout 0.12 + 0.02 0.11+0.04 0.19 + 0.06 0.23 + 0.39 13.0 + 0.02 2 + 0 Y052 Koi 0.33 + 0.06 0.10 + 0.00 1.08 + 0.21 0.13 + 0.07 17.8 + 0.02 4 + 0 Y053 Koi 0.09 + 0.01 0.22 + 0.02 0.05 + 0.02 1.32 + 0.19 19.6 + 0.02 2 + 0 Y054 Koi 0.10 + 0.01 0.10 + 0.01 0.08 + 0.05 0.09 + 0.08 16.8 + 0.02 2 + 0 Y055 Koi 0.08 + 0.01 0.11+0.01 0.03 + 0.04 0.16 + 0.05 11.8 + 0.02 4 + 0 Y056 Koi 0.35 + 0.07 0.10 + 0.01 1.13 + 0.25 0.13 + 0.13 19.9 + 0.01 4 + 0 Y057 Longfin eel 0.09 + 0.00 0.10 + 0.00 0.04 + 0.01 0.05 + 0.05 24.9 + 0.01 2.5 + 0 Y058 Longfin eel 0.09 + 0.01 0.10 + 0.01 0.06 + 0.04 0.08 + 0.08 20.9 + 0.01 2.5 + 0 Y059 Longfin eel 1.06 + 0.08 1.10 + 0.07 4.07 + 0.34 10.66 + 0.86 10.9 + 0.02 4 + 0 YO60 Longfin eel 0.09 + 0.01 0.10 + 0.00 0.04 + 0.02 0.05 + 0.06 20.9 + 0.02 3 + 0 Y061 Longfin eel 0.08 + 0.01 0.09 + 0.00 0.03 + 0.01 0.04 + 0.06 21.7 + 0.02 3 + 0.27 Y062 Longfin eel 0.09 + 0.01 0.10 + 0.00 0.07 + 0.05 0.08 + 0.08 27.1 + 0.02 2 + 0 Y063 Longfin eel 0.77 + 0.02 0.58 + 0.02 2.89 + 0.05 5.23 + 0.50 17.3 + 0.02 2.5 + 0.27 Y064 Longfin eel 0.91+0.23 0.79 + 0.07 3.44 + 0.85 7.35 + 0.34 14.5 + 0.02 2.5 + 0.27 Y065 Biofilm 0.08 + 0.00 0.10 + 0.00 0.02 + 0.03 0.07 + 0.08 19.0 + 0.02 2 + 0 YO66 Biofilm 0.09 + 0.01 0.10 + 0.00 0.05 + 0.05 0.12 + 0.09 7.8 + 0.02 3 + 0 Y067 Biofilm 0.11+0.01 0.10 + 0.01 0.12 + 0.05 0.11+0.03 16.5 + 0.03 2.5 + 0

flBionlm formation assay data is the mean of three independent experiments carried out in triplicate + SD following growth in minimal (EAOB) and rich

(TSB) media at 26?C under dynamic conditions, respectively [39].

Biofilm-forming capacity was determined using equation described by Van Houdt et al. [40]. cCell surface hydrophobicity was calculated by BATH and SAT assays. Bacterial adherence to the hydrocarbon xylene values (%) represent the mean of two

independent replicate experiments + SD as described by Rosenberg et al. [36]; SAT values are expressed as molar concentration of (NH4)2S04 necessary to

cause agglutination of the bacterial cells. Values represent the mean of three replicate experiments + SD.

NA refers to no aggregation as isolates did not agglutinate at any of the (NH4)2S04 concentrations assayed.

slide, where the perspex and silicone met with the glass slide.

Cells of isolate Y064 displayed gliding motility and

polar attachment to the glass surface within 24 h postin oculation. Biofilm formation was observed after 48 h as mono- and multilayered growth with abundant polar attached cells (Fig. 2D).

Isolate Y019 cells attached to the glass surface and formed biofilm structures most rapidly. Monolayered biofilms produced were observed 24 h postinoculation (Fig. 2E). After 48 h, thick layers of biofilm cells were

observed, covering most of the glass slide surface (Fig. 2F). Biofilm surfaces, in contact with the flowing media,

appeared irregular with pillar-shaped extensions pro truding into the flowing media (Fig. 2G). Microcolonies, similar to those of isolate Y012, were observed within the first 24 h by increasing the flow rate from 0.3 to 0.6 ml/min

(Fig. 2H). Although microcolonies increased in size after

48 h, no layered biofilm formation was observed at this faster flow rate. Long filaments were also present among the microcolonies, usually extending into the flowing

media (Fig. 21). Both the A. hydrophila (Fig. 3A) and Vibrio spp. (Fig.

3C) isolates attached to the glass surface and formed biofilm structures within the first 24 h postinoculation. The A. hydrophila and Vibrio spp. isolates displayed a more rapid rate of biofilm formation in comparison to that of F. johnsoniae-like isolates. Differences were observed with respect to the biofilm formation character istics and architectures for both A. hydrophila (Fig. 3A) and Vibrio spp. (Fig. 3C) isolates and those of F.

johnsoniae-like isolates (Fig. 2C, D, F). When co-inoculated, both flavobacterial isolate

Y019 and A. hydrophila cells appeared to randomly attach to the glass slide surface within the first 24 h postinocu lation. Biofilm formation by isolate Y019 was observed



00

Table 3. Coaggregation indices of selected biofilm- and non-biofilm-forming Flavobacterium johnsoniae-like isolates with 15 bacterial partners

Coaggregation indices (%)a

Flavobacterium johnsoniae-like isolates_ yQU yQ15 yQ19 yQ34 yQ45 yQ51 yQ53 yQ59 yQ6Q yQ63 yQ64 yQ67

Biofilm phenotypeb_ + + _ _ + +++ _ ++ ++

Autoaggregation index (%)c_ 339 435 16J 22g 2?9 139 2g3 16J 2?5 252 2QJ lg5

Partner strains Range (%)

Aeromonas salmonicida (41.8%) 19.1-37.8 21.6 30.9 24.9 34.6 31.3 19.1 34.7 30.7 22.5 37.8 22.9 32.3

Aeromonas sobria (27.5%) 10.8-28.4 23.3 17.3 12.9 21.7 19.1 10.8 26.8 23.1 23.7 28.4 11.1 18.1

Aeromonas hydrophila (28.3%) 13.2-32.7 19.5 19.2 14.9 23.4 22.4 13.2 27.4 20.8 22.3 32.7 13.9 21.2 >

Aeromonas media (20.3%) 10.4-26.9 23.4 24.9 14.8 16.7 16.7 10.3 26.9 16.7 21.3 25.4 16.5 14.9 w Myroides spp. (12.0%) 10.5-25.5 22.6 13.6 18.9 25.5 20.9 10.5 20.8 15.3 23.1 17.3 18.7 19.3 %

Chryseobacterium spp. (20.3%) 10.7-26.5 24.1 29.5 15.4 26.5 24.8 10.7 23.1 21.0 21.7 19.2 15.9 14.8 ? Edwardsiella tarda (25.6%) 16.1-38.2 38.2 28.1 20.7 20.3 20.9 16.1 27.0 21.0 26.1 20.1 21.2 15.8 3

Salmonella enterica serovar Arizonae (71.9%) 24.1-78.1 78.8 41.1 66.5 57.8 38.3 70.3 26.3 24.1 62.4 25.8 53.9 58.3 r

Acinetobacter spp. (25.4%) 20.9-34.5 30.2 24.1 26.8 31.9 20.9 25.5 23.8 27.3 34.5 20.2 26.3 25.6 >

Staphylococcus aureus ATCC 25923 (76.1%) 41.1-67.5 67.5 48.2 58.5 55.5 48.9 54.2 36.3 41.1 59.1 30.8 52.7 59.3 | Enterococcus faecalis ATCC 2912 (45.0%) 24.5-49.2 47.8 29.9 49.2 42.4 29.9 42.3 26.1 41.3 46.4 24.5 42.7 41.7 |

Listeria monocytogenes (28.9%) 22.6-67.2 42.3 37.5 67.2 25.6 27.3 22.6 32.1 23.6 36.0 50.1 30.2 24.6 ? Listeria innocua LMG 13568 (56.2%) 31.1-45.0 43.9 45.0 34.6 31.1 35.7 32.1 36.4 35.6 41.5 35.5 35.9 32.7 g

Pseudomonas aeruginosa (24.3%) 16.2-33.9 33.9 25.6 18.9 20.6 23.9 16.2 22.8 26.2 26.0 19.5 20.2 18.6 ?

Micrococcus spp. (51.1%) 28.2-41.9 40.3 40.6 32.3 32.9 37.1 28.2 36.9 34.8 36.1 41.9 31.8 28.8 |

-.-.-.-.-.- g

"Coaggregation indices represent the means of two independent replicate experiments as described by Malik et al. [24]. ^

^Biofilm phenotypes were determined by microtiter assays [39, 40]. ~, +, ++, +++ refer to no biofilm, weak, moderate and strong biofilm formation, respectively. 2 cAutoaggregation indices were determined according to assay of Malik et al. [24]. For partner strains, autoaggregation indices are indicated within parentheses. >

o o I I o > H W




Figure 2. Light microscope images of Flavobacterium johnsoniae-]ike biofilm formation and architecture under 0.3 ml/min EAOB medium

continuous flow, (x600 magnification). A Loosely attached F. johnsoniae-]ike cells at 24 h. B Microcolony formation by isolate Y012 at

48 h postinoculation. C Thick multilayered biofilm growth by isolate Y051 at the glass-perspex interface at 48 h postinoculation. D Mono

and multilayered biofilm-formation by isolate Y064 48 h postinoculation. E Monolayered biofilm formation by isolate Y019 24 h post inoculation. F Multilayered biofilm formation by isolate Y019 48 h postinoculation. G Irregular surface of biofilm by isolate Y019 with

pillar-like structures (black arrow) protruding into the flowing medium, 48 h postinoculation. H Microcolony formation by isolate Y019 at an

increased flow rate of 0.6 ml/min, 24 h postinoculation. I Enlarged microcolony formation by isolate Y019 at a flow rate of 0.6 ml/min,

48 h postinoculation with long filaments (black arrow) extending into flowing medium.

after 48 h in the form of scattered, multilayered, pillar shaped microcolonies among monolayers of A. hydrophila biofilm growth. A. hydrophila cells were interspersed among flavobacterial cells within the microcolonies and also attached to the surface of the microcolonies to form

complex heterogeneous microcosms (Fig. 3B). When flavobacterial isolate Y051 was co-inoculated

with A. hydrophila, the A. hydrophila cells attached within 24 h. Isolate Y051 associated with the glass slide surface and revealed rapid gliding motility among attached A.

hydrophila cells. Biofilm formation after 48 h by A.

hydrophila cells was observed in the form of monolayers partially covering the glass slide surface. Although gliding cells of isolate Y051 were present among the scattered, attached A. hydrophila cells on the glass slide, no

attachment or biofilm formation was observed at any stage within 48 h postinoculation.

The partnering of isolate Y019 and the Vibrio spp. isolate, revealed distinct scattered microcolonies formed

by the Vibrio spp. isolate after 24 h, with randomly

attached flavobacterial cells in between. Isolate Y019 formed multilayered biofilms, resembling thick carpets after 48 h in certain areas (Fig. 3D). Microcolonies com

prised of Vibrio spp. cells attached to the outer surface of isolate Y019 biofilm growth and extended toward the

glass slide surface, forming a thick layer of complex growth containing channels and pillar-like structures

protruding toward the flowing medium. Biofilms formed

by both organisms seemed homogenous in appearance. When JF. johnsoniae-like isolate Y051 and the Vibrio

spp. isolate were combined, the Vibrio spp. cells

displayed the same characteristic scattered microcolonies, as described previously, 24 h postinoculation (Fig. 3E).

Rapidly gliding cells of isolate Y051 were observed

among the microcolonies formed by the Vibrio spp. isolate. Similar complex biofilm structures were observed as in the mixed biofilm growth of isolate Y019 and Vibrio spp. isolate after 48 h (Fig. 3D). The base layers were formed by thick, tightly packed cells of isolate Y051 with microcolonies of Vibrio spp. on the outer surface,




Figure 3. Light microscope images of flow cell (0.3 ml/min EAOB medium continuous flow) biofilm formation by Flavobacterium

johnsoniae-like isolates in the presence of Aeromonas hydrophila or a Vibrio spp. isolate, (x 600 magnification). A Multilayered biofilm

formation by A hydrophila cells 48 h postinoculation. B Cells of F. johnsoniae-Yke isolate Y019 and A hydrophila forming complex heterogeneous microcosms 48 h postinoculation. C Multilayered biofilm formation by Vibrio spp. cells 48 h postinoculation. D Multilayered biofilm formation by F.

johnsoniae-Uke isolate Y019 {black arrow) and microcolonies of the Vibrio spp. isolate (dashed arrow) on the outer surface 48 h postinoculation. E Multi

layered biofilm formation by F. johnsoniae-like isolate Y051 (black arrow) and microcolonies of the Vibrio spp. isolate (dashed arrow) on the outer

surface 48 h postinoculation.

extending to the surrounding glass slide surface to form

complex structures containing channels and pillar-like structures (Fig. 3E).

Discussion

It is now recognized that most bacteria in aquatic environments are associated with surfaces rather than in a planktonic state [10, 45]. Bacteria belonging to

genera that live in aquatic environments (Pseudomonas, Flavobacterium, and Vibrio) have been identified in

aquaculture recirculating systems, where they formed

large biofilms, which released about 104 CFU/ml of free bacteria in the rearing medium [22]. The potential for infection and reinfection is thus great.

The ability of F. johnsoniae-like isolates to form biofilms appeared to be strain-dependent. Although biofilm-forming isolates, with a single exception (isolate Y053), shared phenotypic similarities (growth in both EAOB and TSB, smooth colony morphology, nonmotile, and weakly proteolytic), they appeared to be genetically heterogeneous [15], and no specific genotype could be associated with biofilm formation. The exception, isolate Y053 grew in EAOB, displayed hazy colony morphology,

was strongly motile and displayed strong protease activity. Biofilm-forming isolates did, however, display a

specific whole cell protein profile, with the exception of isolate Y053 [15]. Isolates showed a preference for biofilm formation at 26?C, in nutrient-poor EAOB

medium, under dynamic flow conditions, although they were able to form biofilm structures to a lesser extent in nutrient-rich TSB. Isolate Y053 was significantly differ

ent, as it was the only isolate to display strong adherence in TSB and no adherence with EAOB.

The presence of gliding bacteria in aquatic environ ments and their ability to adhere to various substrata has

suggested that gliding bacteria are likely to be members of microbial biofilms [7], as temporary adhesion is a

requisite for function of their motility machinery. Many Flavobacterium spp., pathogens and opportunistic patho gens included, have been identified in biofilm material collected from diverse environments [26, 32, 35]. Alvarez et al. [1] have suggested that in F. psychrophilum gliding

motility-defective mutants, gliding motility and biofilm formation are antagonistic properties which appear to be affected by TlpB, a putative periplasmic thiobdisulfide oxidoreductase. Their mutant displaying induced biofilm formation in half-strength nutrient broth was not only



A. Basson et al.: Adherence and biofilm formation of F. johnsoniae-like isolates 11

defective in gliding but showed defective extracellular

protease activity, decreased virulence, cytotoxicity, and

growth rate disruption [1], characteristics similar to those

displayed by biofilm-forming isolates in the present study [15]. Alvarez et al. [1] have also reported the inability of

wild-type F. psychrophilum, F. columnare, and F. johnso niae cultures to form biofilms in undiluted nutrient broth.

However, we report here the ability of F. johnsoniae-like isolates, from a variety of aquaculture systems and different fish hosts, to form strong biofilm structures, in a strain-dependent manner.

Whole cell hydrophobicity, autoaggregation, and coag

gregation are important for colonization and biofilm

development in flowing environments [35]. Although the

general rule has been that adhesiveness increases with

increasing and decreases with decreasing hydrophobicity [41], a number of studies have shown contradictory results

where no relationship was found between the bacterial strain's surface hydrophobicity and the extent of initial

binding to either a hydrophilic or hydrophobic substrate

[9, 25]. Isolates in the present study ranged from

moderately hydrophobic (27.1%) to very hydrophilic (7.8%) by the BATH test, while Vatsos et al. [42] described

F. psychrophilum isolates as hydrophilic with BATH values of 16.6-17.6. Sorongon et al. [38], however, reported that adhesion-defective mutants of a Cytophaga (Flavobacte rium) strain U67 and a Cytophaga (Flavobacterium)

johnsoniae strain displaying poor adherence were hydro philic by the BATH assay. In the present study, a significant negative correlation was observed for adherence and

biofilm-forming capacity of isolates in EAOB at 26?C and BATH hydrophilicity. Although Moller et al. [29] described a relatively high (0.1-0.9 M) to very high surface (<0.1 M)

hydrophobicity for F. psychrophilum isolates by the SAT

assay, the F. johnsoniae-like isolates in the present study displayed hydrophilic SAT values. SAT phenotypes could not be correlated with biofilm formation following growth in either EAOB or TSB; however, a significant negative correlation was observed with motility. BATH and SAT

hydrophobicity tests sometimes fail to correlate [2, 31], and this has also been observed in the present study (BATH vs. SAT r =- 0.0006, p

= 0.99). This might be explained by the SAT assay measuring the hydrophobicity of the outer surface as a whole, while the BATH assay measured it in terms of adhesion [25]. In addition, hydrophobicity and surface charge of bacteria may differ between species, serotypes, or strains, change with variation in growth conditions, physiological state of cells, and composition of

suspension media, or might involve variable expression of surface-associated proteins between strains [25, 38]. Col

lectively, this might account for the diversity of BATH and SAT indices obtained for the F. johnsoniae-like isolates.

Capsule components have been shown to form part of the virulence mechanism of certain pathogenic Flavobacterium spp. [13] and also play an important role

in the adhesion and biofilm formation of certain bacterial species [3, 14]. In the present study, no

relationship could be established between the presence of capsule and biofilm phenotypes nor was a relationship to virulence established [15].

Autoaggregation and/or coaggregation play impor tant roles in the development of several different

multispecies biofilms into integrated biological struc

tures, by mediating the juxtapositioning of species next to favorable partner species within taxonomically diverse biofilms [20]. Autoaggregation is a "selfish" mechanism

whereby a strain within the biofilm will express polymers to enhance the integration of genetically identical strains

[34] and may thus enhance the development of freshwa ter biofilms. Autoaggregation interactions are enhanced

by increased hydrophobicity and tend to be stronger than

coaggregation [35]. Study isolates displayed considerable

variability in their autoaggregating abilities suggesting differences among Flavobocterium spp. and strains. No correlation was observed between biofilm-forming ca

pacity and autoaggregation for study isolates. However, a

significant relationship was observed between surface

hydrophobicity using the BATH assay only and autoag gregative ability of JP. johnsoniae-like isolates.

Autoaggregation of aquatic biofilm-forming bacteria, is more profound at high shear flow rates, than

coaggregation [35], resulting in decreased biofilm diver

sity, and altered biofilm formation and biofilm structures.

Although a few microcolonies were observed, no biofilm formation was observed for isolate Y019 when the flow cell flow rate was increased (Fig. 2H-I), which may be related to its low autoaggregation index. However, isolates Y012 and Y015 with the highest autoaggregation indices were unable to form biofilm structures, suggesting that characteristics, other than autoaggregation and

hydrophobicity, determine biofilm-forming capacity. Coaggregation often occurs between bacteria that are

taxonomically distant (intergeneric coaggregation) and

occasionally between strains belonging to the same

species (intraspecies coaggregation) [8]. Rickard et al.

[33, 35] found that Flavobocterium spp. were either unable to autoaggregate or showed low aggregation indices but displayed varying levels of coaggregation

with diverse aquatic bacteria. Isolates in the present study displayed both autoaggregation and coaggregation. Of concern is the ability of F. johnsoniae-like isolates to

partner strongly with bacteria important from a food

microbiology and public health perspective, i.e., S.

enterica, S. aureus, E. foecalis, and L. monocytogenes. The role of Flavobocterium spp. in biofilms, i.e.,

primary colonizers or secondary colonizers, will have to be clarified with more detailed flow cell experiments.

Micrococcus luteus is an important bridging organism in the development of aquatic biofilms, as it is able to

coaggregate with many aquatic heterotrophs [6], and F.




johnsoniae-like isolates appeared to attach fairly well to these bacteria. Preliminary flow cell data with Vibrio spp. (Fig. 3D-E) and A. hydrophila (Fig. 3B) suggests that the role of flavobacterial cells would be dependent on the

coaggregating partner, growth rate and a variety of other

fluctuating conditions including shear forces, nutrient

availability, or physiological conditions. Coaggregation between freshwater bacteria has been shown to be mediated by growth-phase-dependent lectin-saccharide

interactions, which are optimal but modulated in the

stationary phase cultures [29, 34]. Determining the

coaggregation partners of Flavobacterium spp. isolates

may thus indirectly reveal the cell surface properties of this bacterium and factors involved in establishing biofilm

relationships with neighboring organisms. In mixed culture flow cell experiments involving A. hydrophila, biofilm formation by isolate Y019, involved small

complex heterogeneous microcolonies consisting of both flavobacterial and A. hydrophila cells. The rapid and even

spread of Aeromonas cells across the glass-slide surface

may have prevented the initial attachment and subsequent biofilm formation by isolates, or it may have been due to the release of inhibitory bacteriocin-like molecules by the

A. hydrophila strain [28]. In contrast, biofilm formation by flavobacterial isolates Y019 and Y051 appeared enhanced in Vibrio spp. mixed-culture flow cell experiments. Initial formation of scattered microcolonies by the Vibrio spp. isolate may have facilitated attachment of flavobacterial isolates to the glass slide surface before intense biofilm formation by Vibrio spp. cells. The different interactions in the mixed culture biofilms might involve differences in bacterial coaggregation (Table 3), as Aeromonas spp. isolates did not show strong coaggregation partnerships

with flavobacterial isolates. Variations with respect to

lectin or carbohydrate moieties might explain the strong interaction between the Vibrio spp. isolate and flavobacte

rial isolates and lack of interaction displayed by the A.

hydrophila isolate and flavobacterial isolates, respectively. Microscopic analysis of biofilm formation using flow

cells revealed important qualitative characteristics and differences displayed by these isolates. Biofilms formed

by each of the five Flavobacterium spp. isolates differed in terms of structure and morphology, from small micro colonies formed by isolate Y012 (Fig. 3B) to even,

multilayered biofilms formed by isolates Y019 and Y064

(Fig. 3E-G and D). It is interesting to note that isolate Y051 only formed biofilms at the glass-perspex/silicone interface (Fig. 3C). This could be due to weak association

with the glass surface by isolate Y051 or reduced flow

along the edges of channels, enhancing the growth of isolate Y051 in these areas. Biofilm formation on the

perspex walls of the flow cell channels, which was

superior to biofilm growth on the glass slide surface,

could be attributed to difference in texture, such as

roughness, and difference in charge or surface hydro phobicity, which enhanced bacterial attachment [9, 10, 35]. Although study isolates were very hydrophilic, they showed high affinity to hydrophobic substances, such as

perspex. This has been reported for hydrophilic Yersinia ruckeri cells which showed affinity to both hydrophilic and hydrophobic surfaces [9]. The ability of fish

pathogens such as Y. ruckeri and Vibrio harveyi to attach to materials commonly found on fish farms in tanks, fish

transport tubs, water pipes, and wedges, viz., concrete,

fiberglass, polyvinylchloride (PVC), and wood has been

investigated [9, 19]. Similarly, some of the flavobacterial isolates in the present study had been isolated from PVC-attached biofilms from rainbow trout tanks [15]. Regardless of cell hydrophobicity, bacteria showed not

only high levels of adhesion to PVC, but surface

roughness seemed to play a major role in bacterial adherence. These materials are not unique to the

aquaculture environment and are found in medical,

farming, food, and industrial environments and are thus

constantly available for bacterial adherence, colonization, and biofilm-formation.

We have thus shown strain-to-strain variation in the

ability of F. johnsoniae-like isolates obtained from fish, to form biofilms. Although the hydrophobicity character istics of these organisms were negatively correlated with the capacity to form biofilms, motility, and proteolytic activity, their hydrophilic nature is significant for their

ability to autoaggregate. Biofilm formation by F. johnso niae-like isolates, as for Y. ruckeri [10], may well be an

adaptive advantage for survival in the detrimental

oligotrophic conditions that prevail in the aquatic environment. Investigation of F. johnsoniae-like biofilms is important due to their pathogenicity, especially in fish.

Revealing the specific mechanisms by which Flavobocte rium spp. adhere to solid supports or in vivo to fish

tissue, may aid in treating outbreaks of the pathogen. Preventing primary adhesion would prevent biofilm formation and may lead to better and faster eradication of the fish pathogen. Eliminating the coaggregating bacterial partners may also weaken biofilm structures.

Acknowledgements

The authors wish to thank Gideon Wolfaardt for his technical assistance and stimulating discussions on

biofilms and AquaStel (Pty) Ltd. for access to aquacul ture fish and facilities. This work was funded by a

Researchers' in Training?Thuthuka Program grant to H. Y. Chenia from the National Research Foundation of South Africa (TTK2003032000142).



A. Basson et al.: Adherence and biofilm formation of F. johnsoniae-like isolates 13

References

1. Alvarez, B, Secades, P, Prieto, M, McBride, MJ, Guijarro, JA (2006) A mutation in Flavobacterium psychrophilum tlpB inhibits gliding motility and induces biofilm formation. Appl Environ Microbiol 72: 4044-4053

2. Balebona, MC, Morinigo, MA, Borrego, J J (2001) Hydrophobicity and adhesion to fish cells and mucus of Vibrio strains isolated

from infected fish. Int Microbiol 4: 21-26

3. Bell, M (2001) Biofilms: a clinical perspective. Curr Infect Dis Rep 3:

483-486

4. Bernardet, J-F, Bowman, JP (2005) The genus Flavobacterium. In:

Dworkin, M (Ed.) The Prokaryotes, Springer, New York

5. Bernardet, J-F, Segers, P, Vancanneyt, M, Berthe, F, Kersters, K,

Vandamme, P (1996) Cutting a Gordian knot: emended classifi

cation and description of the genus Flavobacterium emended

description of the family Flavobacteriaceae and proposal of

Flavobacterium hydatis nom. nov. (Basonym Cytophaga aquatilis Strohl and Tait 1978). Int J Syst Bacteriol 46: 128-148

6. Bos, R, van der Mei, HC, Busscher, HJ (1999) Physico-chemistry of initial microbial adhesive interactions?its mechanisms and

methods for study. FEMS Microbiol Rev 23: 179-230 7. Burchard, RP, Sorongon, ML (1998) A gliding bacterium strain

inhibits adhesion and motility of another gliding bacterium strain in a marine biofilm. Appl Environ Microbiol 64: 4079-4083

8. Buswell, CM, Herlihy, PD, Marsh, PD, Keevil, CW, Leach, SA

(1997) Coaggregation amongst aquatic biofilm bacteria. J Appl Microbiol 83: 477-484

9. Coquet L, Cosette P, Junter G-A, Beucher, E, Saiter, J-M, Jouenne, T (2002) Adhesion of Yersinia ruckeri to fish farm materials:

influence of cell and material surface properties. Colloids Surfaces B Biointerfaces 26: 373-378

10. Coquet, L, Cosette, P, Quillet, L, Petit, F, Junter, G-A, Jouenne, T

(2002) Occurrence and phenotypic characterization of Yersinia ruckeri strains with biofilm-forming capacity in a rainbow trout farm. Appl Environ Microbiol 68: 470-475

11. Crump, EM, Perry, MB, Clouthier, SC, Kay, WW (2001) Antigenic characterization of the fish pathogen Flavobacterium psychrophi lum. Appl Environ Microbiol 67: 750-759

12. Decostere, A, Haesebrouck, F, Charlier, G, Ducatelle, R (1999) The association of Flavobacterium columnare strains of high and low virulence with gill tissue of black mollies (Poecilia sphenops). Vet

Microbiol 67: 287-298

13. Decostere, A, Haesebrouck, F, Van Driessche, E, Charlier, G, Ducatelle, R (1999) Characterization of the adhesion of Flavobacterium columnare

(Flexibacter columnaris) to gill tissue. J Fish Dis 22: 465-474 14. Donlan, RM (2002) Biofilms: Microbial life on surfaces. Emerg

Infect Dis 8: 881-890 15. Flemming, L, Rawlings, DE, Chenia, HY (2007) Phenotypic and

molecular characterization of fish-borne Flavobacterium johnso niae-like isolates from aquaculture systems in South Africa. Res

Microbiol 158: 18-30 16. Gavin, R, Merino, S, Altarriba, M, Canals, R, Shaw, JG, Tomas, JM

(2003) Lateral flagella are required for increased cell adherence, invasion and biofilm formation in Aeromonas spp. FEMS Micro biol Lett 224: 77-83

17. Gorski, L, Godchaux, III, W, Leadbetter, ER, Wagner, RR (1992)

Diversity in surface features of Cytophaga johnsoniae motility mutants. J Gen Microbiol 138: 1767-1772

18. Harley, JP, Prescott, LM (1996) Laboratory exercises in Microbi

ology. Wm. C. Brown Publishers. Dubuque, pp 35-36 19. Karunasagar, I, Otta, SK, Karunasagar, I (1996) Biofilm formation

by Vibrio harveyi on surfaces. Aquacult 140: 241-245

20. Kolenbrander, PE (2000) Oral microbial communities: biofilms

interactions and genetic systems. Annu Rev Microbiol 54: 413-437

21. Lavender, HF, Jagnow, JR, Clegg, S (2004) Biofilm formation in

vitro and virulence in vivo of mutants of Klebsiella pneumoniae. Infect Immun 72: 4888-4890

22. Leonard, N, Blancheton, JP, Guiraud, JP (2000) Populations of

heterotrophic bacteria in an experimental recirculating aquacul ture system. Aquae Eng 22: 109-120

23. Madsen, L, Dalsgaard, I (1998) Characterization of Flavobacterium

psychrophilum: comparison of proteolytic activity and virulence of

strains isolated from rainbow trout (Oncorhynchus mykiss). In:

Barnes, AC, Davidson, GA, Hiney, MP, Mcintosh, D (Eds.),

Methodology in Fish Diseases Research, Fisheries Research

Services, Aberdeen, pp 45-52 24. Malik, A, Sakamoto, M, Hanazaki, S, Osawa, M, Suzuki, T,

Tochigi, M, Kakii, K (2003) Coaggregation among nonflocculating bacteria isolated from activated sludge. Appl Environ Microbiol

69: 6056-6063 25. Mattos-Guaraldi, AL, Formiga, LCD, Andrade, AFB (1999) Cell

surface hydrophobicity of sucrose fermenting and nonfermenting Corynebacterium diphtheriae strains evaluated by different meth

ods. Curr Microbiol 38: 37-42

26. McBain, AJ, Bartolo, RG, Catrenich, CE, Charbonneau, D, Ledder, RG, Rickard, AH, Symmons, SA, Gilbert, P (2003) Microbial

characterization of biofilms in domestic drains and the establishment of stable biofilm microcosms. Appl Environ Microbiol 69: 177-185

27. McBride, MJ (2004) Cytophaga-Flavobacterium gliding motility. J Mol Microbiol Biotechnol 7: 63-71

28. Messi, P, Guerrieri, E, Bondi, M (2003) Bacteriocin-like substance

(BLS) production in Aeromonas hydrophila water isolates. FEMS

Microbiol Lett 220: 121-125

29. Moller, JD, Larsen, JL, Madsen, L, Dalsgaard, I (2003) Involvement of a sialic acid-binding lectin with hemagglutination and hydro phobicity of Flavobacterium psychrophilum. Appl Environ Micro biol 69: 5275-5280

30. O'Toole, G, Kaplan, HB, Kolter, R (2000) Biofilm formation as

microbial development. Annu Rev Microbiol 54: 49-79 31. Ofek, I, Hasty, DL, Doyle, RJ (2003) Bacterial Adhesion to Animal

Cells and Tissues. ASM Press. Washington DC 32. Oppong, D, King, VM, Bowen, JA (2003) Isolation and character

ization of filamentous bacteria from paper mill slimes. Int Biodeterior Biodegrad 52: 53-62

33. Rickard, AH, Buswell, CM, Leach, SA, High, NJ, Handley, PS

(2002) Phylogenetic relationships and coaggregation ability of

freshwater biofilm bacteria. Appl Environ Microbiol 68: 3644-3650 34. Rickard, AH, Gilbert, P, High, NJ, Kolenbrander, PE, Handley, PS

(2003) Bacterial coaggregation: an integral process in the develop ment of multi-species biofilms. Trends Microbiol 11: 94-100

35. Rickard, AH, McBain, AJ, Stead, AT, Gilbert, P (2004) Shear rate moderates community diversity in freshwater biofilms. Appl Environ Microbiol 70: 7426-7435

36. Rosenberg, M, Gutnick, D, Rosenberg, E (1980) Adherence of bacteria to hydrocarbons: a simple method for measuring cell surface hydrophobicity. FEMS Microbiol Lett 9: 29-33

37. Rozgonyi, F, Szitha, KR, Ljungh, A, Baloda, SB, Hjerten, S, Wadstrom, T (1985) Improvement of the salt aggregation test to

study bacterial cell-surface hydrophobicity. FEMS Microbiol Lett 30: 131-138

38. Sorongon, ML, Bloodgood, RA, Burchard, RP (1991) Hydropho bicity adhesion and surface-exposed proteins of gliding bacteria.

Appl Environ Microbiol 57: 3193-3199 39. Stepanovic, S, Vukovic, D, Davie, I, Savic, B, Svabic-Vlahovic, M

(2000) A modified microtiter-plate test for quantification of

staphylococcal biofilm formation. J Microbiol Methods 40: 175-179




40. Van Houdt, A, Aertsen, A, Jansen, AL, Quintana, Michiels, CW

(2004) Biofilm formation and cell-to-cell signaling in Gram

negative bacteria isolated from a food processing environment. J

Appl Microbiol 96: 177-184 41. Van Loosdrecht, MCM, Lyklema, J, Norde, W, Scraa, G, Zehnder,

AJB (1987) The role of bacterial cell wall hydrophobicity in

adhesion. Appl Environ Microbiol 53: 1893-1897

42. Vatsos, IN, Thompson, KD, Adams, A (2001) Adhesion of the

pathogen Flavobacterium psychrophilum to unfertilized eggs of

rainbow trout (Oncorhynchus mykiss) and ?-hexadecane. Lett Appl Microbiol 33: 178-182

43. Wang, S-Y, Lauritz, J, Jass, J, Milton, DL (2003) Role for the major outer-membrane protein from Vibrio anguillarum in bile resis tance and biofilm formation. Microbiol 149: 1061-1071

44. Wolfaardt, GM, Lawrence, JR, Robarts, RD, Caldwell, SJ, Caldwell, DE (1994) Multicellular organization in a degradative biofilm

community. Appl Environ Microbiol 60: 434-446

45. Wong, H-C, Chung, Y-C, Yu, J-A (2002) Attachment and

inactivation of Vibrio parahaemolyticus on stainless steel and glass surface. Food Microbiol 19: 341-350

46. Yuehuei, HA, Friedman, RJ (2000) Handbook of Bacterial

Adhesion. Humana Press. Totowa NJ, pp 30-58



Documents

Evaluation of Adherence, Hydrophobicity, Aggregation, and Biofilm Development of Flavobacterium johnsoniae-Like Isolates