1

Vibrio harveyi adheres to and penetrates tissues of the European 1

abalone Haliotis tuberculata within the first hours of contact 2

3

Marion Cardinaud1, Annaïck Barbou, Carole Capitaine, Adeline Bidault, Antoine Dujon, 4

Dario Moraga, Christine Paillard1. 5

6

UMR 6539-LEMAR (Laboratoire des Sciences de l'Environnement Marin), IUEM (Institut 7

Universitaire Européen de la Mer), Université de Bretagne Occidentale (UBO), CNRS, IRD, 8

Ifremer, Technopôle Brest Iroise, 29280 Plouzané, France 9

Phone: (33) 2 98 49 86 44 - Fax: (33) 2 98 49 86 45 10

11

1: corresponding authors marion.cardinaud@gmail.com ; christine.paillard@univ-brest.fr 12

13

Keywords: Vibrio harveyi, abalone, infection, quantitative PCR, gills 14

AEM Accepts, published online ahead of print on 8 August 2014Appl. Environ. Microbiol. doi:10.1128/AEM.01036-14Copyright © 2014, American Society for Microbiology. All Rights Reserved.

2

Abstract 15

Vibrio harveyi is a marine bacterial pathogen responsible for episodic epidemics generally 16

associated with massive mortalities in many marine organisms including the European 17

abalone Haliotis tuberculata. The aim of this study was to identify the portal of entry and the 18

dynamics of infection of V. harveyi in the European abalone. 19

The results indicate that the duration of contact between V. harveyi and the European abalone 20

influenced the mortality rate and precocity. Immediately after contact, the epithelial and 21

mucosal area situated between gills and the hypobranchial gland was colonized by V. harveyi. 22

Real time PCR analyses and culture quantification of a GFP-tagged strain of V. harveyi in 23

abalone tissues revealed a high density of bacteria adhering to and then penetrating the whole 24

gill-hypobranchial gland tissue after one hour of contact. V. harveyi was also detected in the 25

hemolymph of a significant number of European abalone after 3h of contact. 26

In conclusion, this article has shown that a Taqman Real-time PCR assay is a powerful and 27

useful technique for the detection of a marine pathogen such as V. harveyi, in mollusk tissue, 28

and for the study of its infection dynamics. Thus, we have revealed that the adhesion and then 29

penetration of V. harveyi in European abalone organs begin in the first hours of contact. We 30

also hypothesize that the portal of entry of V. harveyi in the European abalone is the area 31

situated between the gills and the hypobranchial gland. 32

3

1. Introduction 33

The halophilic Gram-negative bacterium Vibrio harveyi is a common pathogen of many 34

marine vertebrate and invertebrate species (1). V. harveyi is known to induce gastro-enteritis, 35

inflammation of the circulatory system or eye and skin lesions in various species of fishes, 36

crustaceans and mollusks (2). In the European abalone Haliotis tuberculata, V. harveyi is 37

responsible for massive mortalities that occur during the summer spawning period both in 38

natural populations and in farmed stocks (3, 4) . Vibriosis of the European abalone is 39

characterized by the presence of white spots on the foot and inflammation of the pericardial 40

tissue, resulting in impaired mobility and septicemia (5). 41

Schematically, the bacterial infection process can be summarized in three stages: 1/ 42

colonization, adhesion and initial multiplication of the pathogen to the host surface, and then 43

its penetration inside the body through one or several portals, 2/ invasion of the pathogen 44

inside the host organs and/or the circulatory system concurrently with the expression of 45

virulence factors, and 3/ exit of the pathogen and transmission of the disease (6, 7). Invasion 46

and multiplication of V. harveyi inside the circulatory system – hemolymph – seem to take 47

place after 24h of contact with the European abalone (5), suggesting that adhesion and 48

penetration of V. harveyi must occur in the first 24h of contact. The portal of entry and the 49

precise timing of the earlier stages of V. harveyi infection in the European abalone remain 50

uncharacterized. 51

Many studies which have focused on infection dynamics of fish pathogens have revealed a 52

wide variety of modes of infection among Vibrio species. For example, V. anguillarum seems 53

to penetrate preferentially through the gastro-intestinal tract in the turbot Scophthalmus 54

maximus (8), while both skin and gills are respectively the portal of entry in the Atlantic 55

Salmon Salmo salar (9) and in the rainbow trout Salmo gairdneri (10). In marine mollusks, 56

4

the organic matrix of the shell may be a putative portal of entry of Vibrio pathogens, as 57

observed with V. tapetis which is known to invade Ruditapes phillipinarum clams after 58

adhesion on the periostracal lamina shell and the extrapalleal fluid (11, 12). Considering the 59

different modes of infection in pathogenic Vibrio species, the objectives of this study are to 60

localize and verify the existence of a portal of entry of V. harveyi in the European abalone and 61

to determine the infection dynamics occurring during the early stages of vibriosis. 62

63

2. Material and methods 64

2.1. Abalone and bacterial strains 65

Sexually mature abalone were transported in containers filled with macroalgae Palmaria 66

palmata from France Haliotis hatchery, Plouguerneau, France (francehaliotis.com) to our 67

laboratory at the European Institute for Marine Studies, Brest, France in June 2011. The 68

abalone were used in the experiments if they did not present any tissue injuries or pedal 69

muscle pustules, and were capable of moving and adhering to the tank surface. The abalone 70

were placed in 5 L tanks of seawater which was replaced daily with pumped seawater from 71

the Bay of Brest (dissolved oxygen concentration > 7.5 mg. L-1, salinity 35%) and filtered at 1 72

µm. Seawater temperature was increased by one degree Celsius per day until it reached 19°C 73

and the animals were then maintained at this temperature for one week before beginning the 74

experiments. 75

Two virulent strains of V. harveyi were used in this study. The first ORM4 strain had been 76

isolated from diseased abalone H. tuberculata in Normandy, France, during an episode of 77

massive mortalities in 1999 and affiliated to V. carchariae, synonym species of V. harveyi, 78

after 16S gene sequencing and phylogenetic analysis (3, 13). The second strain is a green 79

fluorescent protein-tagged (GFP) and kanamycin resistant bacterium constructed from the 80

5

ORM4 strain by Travers et al. (5). The initial ORM4 strain and the GFP-ORM4 strain have 81

similar growth rate and virulence, and the GFP-labelled plasmid is stable in the GFP-ORM4 82

strain during 8 days (5). Bacteria were grown in Luria-Bertani Broth (LB, Sigma) 83

supplemented with salt (20 g. L-1, f.c) – LBS – in a temperature controlled shaker, at 28°C 84

for 16 hours before challenge. Prior to use in experiments, bacteria cultures were washed and 85

re-suspended in sterile seawater. 86

87

2.2. Experiment 1: effect of contact duration on abalone mortality 88

Five triplicate groups of 15 sexually mature abalone (n=270, 2.5 years old, 38.8 mm ± 2.1 ) 89

were challenged with a suspension of ORM4 strain, estimated by spectrophotometric method 90

(5) at a final concentration of 105 CFU.mL-1 per tank, concentration usually used to reproduce 91

European abalone vibriosis in controlled conditions (4), during one hour, 3h, 6h, 9h and 24h. 92

A control group, not exposed to bacteria, was also included. Dead abalone were counted and 93

removed daily until mortality rate stabilization. The mortality of an abalone was attributed to 94

vibriosis if the ORM4 strain was detected in its hemolymph. Comparison of mortality curves 95

between different challenge durations was performed with a Kaplan-Meier analysis followed 96

by a log-rank test (14). 97

98

2.3. Experiment 2: microscopic observations 99

Two non-infected abalone were removed from their shell and visualized using a LUMAR 100

stereomicroscope (Carl Zeiss Group, Oberkochen, Germany) to identify autofluorescent 101

organs prior to adding a 100 µL GFP-V. harveyi ORM4 strain suspension, at a concentration 102

6

of 105 bacteria/mL, at 19°C. Microscopic observation was immediately done to detect a 103

putative preferential tropism of V. harveyi for one or several organs. 104

105

2.4. Experiment 3: quantification of V. harveyi in European abalone organs 106

during the first hours after contact 107

2.4.1. Bacterial challenge and animal sampling 108

Two triplicate groups of 10 sexually mature abalone (n= 60, 2 years old, 34.6 mm ± 1.9) were 109

placed in tanks: a first group contained animals challenged with the GFP strain at a final 110

concentration of 106 CFU.mL-1, and a second group contained animals immersed in filtered 111

sea water (control group). A high bacterial concentration was used in this experiment to avoid 112

inter-individual variability in response to the pathogen virulence. Ten animals were removed 113

from the tanks prior to infection and were used as t0 samples. Five control and five challenged 114

individuals were randomly sampled from the tanks after one hour, 3h, 6h, 9h and 24h post-115

infection. 116

117

2.4.2. Hemolymph and tissue sampling 118

The hemolymph of each abalone was collected from the pedal sinus using a 2.5 mL sterile 119

syringe mounted with a 23-gauge needle. One milliliter of hemolymph was reserved for 120

bacterial quantification by plating on LBS agar dishes and then keeping on ice until use, and 121

one milliliter was saved for Real-time quantitative PCR (qPCR) analysis and stored at -20°C 122

until DNA extraction. After hemolymph sampling, all animals were dissected under sterile 123

conditions. The following tissues were dissected for further analysis: (1) the two gills and 124

attached hypobranchial gland, (2) the central portion of the foot-muscle, (3) a section of the 125

7

gonad-digestive gland, (4) mouth and (5) epipodia. All organs were cut into two equal 126

portions. The first one was immediately washed in an ethanol 70% bath to remove external 127

bacteria and then rinsed with sterile seawater. The second portion was briefly plunged in a 128

sterile seawater bath to detect both adhered and penetrated bacteria and was considered as 129

untreated tissue. Each organ piece was weighed and homogenized in 1 mL sterile seawater 130

using a mechanical disruptor. Two hundred microliters of homogenate were reserved for V. 131

harveyi quantification by plating and kept on ice until use, and 800 µL were stored at -20°C 132

until DNA extraction. 133

134

2.4.3. Bacterial quantification by plating 135

The bacterial quantification of abalone samples collected after 1h, 3h and 6h of contact with 136

V. harveyi was done by plating tissue homogenates. Two series of ten-fold dilutions (10-1 and 137

10-2) of tissue homogenates and hemolymph samples were made with sterile seawater. One 138

hundred microliters of initial and diluted homogenates were plated onto LBS agar dishes 139

containing 100 µg.mL-1 kanamycin, to ensure GFP bacteria selection. Dishes were incubated 140

overnight at 28°C and the number of 'colonies fluorescing under UV light were counted. The 141

number of colonies was reported per 100 mg of tissue (density) or one mL of hemolymph 142

(concentration). 143

144

2.4.4. Bacterial quantification by Real-time quantitative PCR 145

2.4.4.1. DNA extraction procedure 146

DNA extraction was done on hemolymph and untreated tissues. Five hundred microliters of 147

hemolymph or homogenates of untreated tissue were diluted in 500 µL of extract salted buffer 148

8

(100 mM TrisHCl, 100 mM NaCl and 50 mM EDTA pH8). Total DNA was extracted using a 149

traditional phenol/chloroform/isoamyl alcohol method (15), after treatment with RNase A (10 150

mg/mL), lysozyme (10 mg/mL), 10% SDS (v/v), 10% Sarkosyl (v/v), and Proteinase K (25 151

mg/mL). Finally, DNA was suspended in 50 µL of DNase free water. Concentration and 152

quality of the nucleic acids were determined using a spectrometer Nanodrop ND-1000 153

(Thermo Fisher scientific, USA). 154

The standard samples used for qPCR were obtained after DNA extraction from 500 µL of 155

hemolymph or 100 mg organ homogenates, dissected from non-infected abalone and 156

supplemented with 10-fold serial dilutions of V. harveyi, from 108 CFU to 0. Viable bacteria 157

were counted by plating and optical microscopy. 158

159

2.4.4.2. Real-time quantitative PCR conditions and analysis 160

The concentration of V. harveyi in hemolymph was measured by qPCR using a 7500 Fast 161

Real-Time PCR System instrument and the TaqMan® Universal Master Mix (Applied 162

biosystems, Life Technologies Corporation, Carlsbad, CA, USA). Amplification was done 163

with V. harveyi tox-R gene specific primers: Forward CCA-CTG-CTG-AGA-CAA-AAG-CA 164

and Reverse GTG-ATT-CTG-CAG-GGT-TGG-TT. Fluorescent visualization of amplification 165

was done using a tox-R probe dually labeled with a 5’ reporter dye Texas Red and a 166

downstream 3’ quencher dye BHQ-2: CAG-CCG-TCG-AAC-AAG-CAC-CG (16). After 167

optimization of primer and TaqMan® probe concentrations, each reaction was run in 168

duplicates, for a final volume of 20 µL with 4 µL of DNA sample, 1X concentration of 169

TaqMan master mix, 300 nm of each primer and 200 nm of probe. Reactions were initiated 170

after activation of the Thermo-Start DNA polymerase during 10 minutes at 95°C followed by 171

45 amplification cycles (denaturation at 95°C for 15 s, annealing at 60°C for 1 min). 172

9

Threshold cycle (Ct) is defined as the first PCR cycle in which probe reporter fluorescence is 173

detectable above a baseline signal, and is inversely proportional to the logarithm of the initial 174

bacteria density or concentration. Thus, Ct values, obtained for each sample using 7500 175

software v2.0.3, were compared with the standard curves to determine the initial V. harveyi 176

density or concentration. 177

The sensitivity of the amplification was estimated for each abalone organ from 3 separate 178

PCR assays. The primer efficiency was determined by the slope of the standard curves using 179

the method described by Yuan et al.: E = 10[-1 / slope] (17). 180

The influence of time on bacterial density or concentration was assessed for the different 181

organs using a non-parametric Wilcoxon test. Differences of bacterial concentrations or 182

densities between two infection intervals were detected by a Wilcoxon signed-rank test (18). 183

Statistic tests were done using JMP 10.0.0 software (SAS Institute Inc.). 184

185

3. Results 186

3.1. Effect of contact duration on abalone mortality 187

European abalone mortality due to vibriosis was assessed during a 12 day period after 188

different durations of contact with V. harveyi (Fig. 1, table 1). The cause of mortality was 189

assessed, each day, by detection of V. harveyi in abalone hemolymph. The cumulated 190

mortality rate of infected abalone increased while the lethal time period (LT50), where 50% 191

of abalone are dead, decreased throughout the different contact durations (Table 1). Kaplan 192

Meier analyses inferred that mortality curves are different between 1h and 6h (p = 0.0114) 193

and between 6h and 24h of contact (p = 0.042). A contact of one hour with V. harveyi was 194

sufficient to induce 80% and 77% of mortality in 2 tanks whereas only one abalone died in the 195

10

last one, after 11 days of experiment. Abalone mortality reached respectively 72% and 80% 196

when abalone were in contact during 6h or 24h to V. harveyi, after 8 days of experiment. No 197

mortality was reported in control groups. 198

199

3.2. Portal of entry 200

3.2.1. Microscopic observations 201

Fluorescent bacteria were observed on the surface of gill filaments immediately after contact 202

with the GFP-V. harveyi strain (data not shown). After 6h of contact, fluorescent bacteria 203

were aggregated on the epithelial and mucosal area between gills and the hypobranchial gland 204

(data not shown). 205

206

3.2.2. Portal of entry and targeted organs 207

Quantification by plating revealed that V. harveyi densities increased exponentially in all 208

organs during the first few hours after contact and throughout the different contact times (Fig. 209

2). The density of adhered GFP bacteria was highest on the surface of the gills, being 5-fold 210

greater than in the digestive gland and 12-fold greater than on the surface of foot-muscle after 211

6h of contact (Fig. 2A). Penetrated bacteria were also 10-fold more numerous in gills than in 212

the digestive gland after 6h of contact (Fig. 2B). 213

214

3.3. Efficiency and qPCR sensitivity for V. harveyi quantification 215

Sensitivity and efficiency of V. harveyi qPCR quantification via ToxR gene amplification 216

were assessed in gills, digestive gland, foot-muscle and hemolymph (Fig. S1). The sensitivity 217

11

threshold was estimated at 102 bacteria/100 mg of tissue for gills, 103 bacteria/100 mg for 218

digestive gland, 101 bacteria/100 mg for foot-muscle and 102 bacteria/mL for hemolymph. 219

The equation of the regression line was then used to calculate bacteria concentration or 220

density from obtained qPCR data. The average amplification efficiency ranged between 95% 221

in gills, 84% in digestive gland, and 105% in muscle and hemolymph. 222

223

3.4. Infection kinetics 224

V. harveyi concentration depended on contact duration in all abalone compartments (p < 0.01 225

for the Wilcoxon test). In hemolymph, bacterial concentration increased with contact 226

duration, except for samples measured by qPCR quantification method between 3h or 6h and 227

9h of contact (Fig. 3, table 2). In gills, the density of adhered and penetrated vibrios also 228

increased significantly with duration until 6h of contact, for both of quantification methods 229

used (Table 2). Moreover, penetrated bacteria in gills surpassed 103 CFU/ 100 mg after the 230

first hour of contact, while penetrated bacteria in the digestive gland and muscle did not reach 231

the 102 CFU/ 100 mg threshold until 6h of contact (Fig. 3). For the digestive gland, 232

quantification by plating revealed a significant increase of density either for adhered or 233

penetrated bacteria during infection. However, qPCR measures showed a stabilization of V. 234

harveyi densities in the digestive gland after one hour, except the difference detected between 235

3h and 6h of contact (Fig. 3, table 2). In foot-muscle, V. harveyi density measured by plating 236

increased significantly during the first hour and after 6h of contact (Fig. 3, table 2). 237

Meanwhile, qPCR analysis in foot-muscle showed significant differences between controls 238

and infected animals until 9h of contact (Table 2). 239

12

For the majority of the measures, bacterial count by qPCR analysis was slightly higher than 240

plating count. No bacteria were counted either in t0 samples or in controls, by plating or by 241

qPCR analysis. 242

13

4. Discussion 243

Duration of contact with the pathogen V. harveyi influenced the mortality rate and infection 244

kinetics in European abalone (Fig. 1). Immediately after contact, a large number of bacteria 245

could be observed in the gills and hypobranchial gland. Moreover, after 6h of contact, the 246

density of V. harveyi was at least 5-fold greater in gills and in the hypobranchial gland than in 247

other tissues (Fig. 2). Finally, V. harveyi was detected in hemolymph after 3h of contact (Fig. 248

3). 249

Our results show that the duration of contact between V. harveyi and the European abalone 250

was correlated with mortality rate. Moreover, one hour of exposure was sufficient to induce 251

mortality. This observation suggests that V. harveyi is able to adhere and then penetrate 252

abalone tissues during the first hour of contact. The relative rapidity of adhesion and 253

penetration in the host has been described for many Vibrio species and in V. harveyi, and may 254

contribute to their symbiotic process. For example, in the association between the luminous 255

bacterium V. fischeri and the sepiolid squid Euprymna scolopes, the symbiont is able to 256

adhere to the ciliated epithelium of juvenile squid light organs after only 3h (19). In the 257

Kuruma prawn Penaeus japonicus, the study of Vibrio sp infection dynamics after oral 258

inoculation showed that the pathogen can be detected in stomach and hemolymph after 3h, 259

and in all other sampled organs after 6h (20). In the giant tiger prawn Penaeus monodon, a 260

modulation of immune gene expression and of the synthesis of protein involved in immune 261

function was observed in hemocytes after 6h of immersion with a pathogenic strain of V. 262

harveyi, suggesting that its earlier capacities of adhesion, penetration and invasion (21, 22). 263

The quantification of adhered or penetrated V. harveyi by plating showed a high density of V. 264

harveyi on the surface then within gills filaments and the hypobranchial gland after one hour 265

of contact. Microscopic observation confirmed a tropism of V. harveyi for the epithelial and 266

14

mucosal region situated between gills and the hypobranchial gland in the European abalone. 267

Previous studies showed that gills seem to be an ideal place of adhesion, hence a relevant 268

entrance of Vibrio pathogens in marine mollusks. In the case of the challenge with a GFP 269

virulent strain of V. parahaemolyticus, the density of bacteria was maintained in gills of 270

oysters Tiostrea chilensis for 48h after contact (23). Sawabe et al. also revealed a tropism of a 271

GFP virulent strain of V. harveyi for gills in the abalone H. discus hannai (24). Our results 272

add another degree of clarification and functional complexity regarding the observations done 273

by Travers et al., who visualized an accumulation of V. harveyi on the surface of gills in 274

infected animals (5). This region may constitute the portal of entry of V. harveyi in the 275

European abalone. The proximity of this targeted region with the heart, and the anterior aorta 276

may allow the pathogen to quickly invade the circulatory system of the European abalone (25, 277

26). This hypothesis of an entry route of V. harveyi via the epithelial region between gills and 278

the hypobranchial gland could be confirmed using some additional advanced electron 279

microscopy techniques, which may also specify the entry strategy of V. harveyi in gill tissue 280

via intra- or extracellular way. 281

In the current study, we have used and tested a Taqman® probe developed by Schikorski et 282

al. (2013) for qPCR assays of V. harveyi (14). Measures done with Taqman qPCRs following 283

a serial dilution scheme of pure culture of V. harveyi in samples of hemolymph or tissues 284

showed an excellent linear correlation and have proven to be a powerful and sensitive 285

technique for V. harveyi detection and quantification in European abalone. TaqMan qPCRs 286

have been previously useful for the detection of pathogenic strains of V. vulnificus, V. 287

parahemolyticus and V. aestuarianus in marine mollusks (27, 28, 29, 30, 31). Nevertheless, 288

an extrapolation of V. harveyi count by qPCR was observed in our measures compared to the 289

plating method. This observation may be due to the quantification of dead or uncultured 290

bacteria by qPCR. 291

15

Quantification of adhered and penetrated V. harveyi in abalone organs (i.e., gills-292

hypobranchial gland, the digestive gland and the foot-muscle), and in hemolymph, have 293

elucidated infection dynamics. After one hour of contact, V. harveyi was already detected by 294

plating and by qPCR on the surface of all the abalone tissue analyzed, hence inferring that V. 295

harveyi is able to adhere to abalone during the first hour of contact. The concentration of V. 296

harveyi in the hemolymph increased progressively between 3h and 24h after contact. This 297

observation suggests that the concentration of V. harveyi in hemolymph may be a proxy to the 298

development of infection in the European abalone, and that the stage of multiplication and 299

invasion may occur during the first hours after contact. The relative rapidity of Vibrio 300

invasion was also observed in the oyster Crassostrea gigas in which virulent strains of V. 301

spendidus and V. aestuarianus were detected by qPCR in hemolymph after 2h of contact, and 302

concentrations of both pathogens were maximal after 6h (32). On observation of all the organs 303

analyzed here, the density of V. harveyi increased significantly with the contact duration in 304

gills in which the density was maximal after 6h of contact. This increase in pathogen density 305

may be due to a local multiplication of V. harveyi on the surface of gills and the 306

hypobranchial gland, which supports the hypothesis of a putative portal of entry of V. harveyi 307

in this region. 308

In conclusion, V. harveyi seems to have the capacity to adhere and penetrate the European 309

abalone during the first hours of contact. We hypothesize that the portal of entry of V. harveyi 310

is the epithelial region situated between the gills and the hypobranchial gland. Further 311

observations need to be conducted to verify the putative portal of entry of V. harveyi in the 312

European abalone and localize it precisely. Invasion of V. harveyi in the abalone organs is fast 313

and progressive during the first hours of contact and may be characterized by quantification in 314

the hemolymph. In this study, we have also successfully tested a Taqman qPCR assay for V. 315

harveyi quantification in European abalone tissue which may be useful for pathogen detection 316

16

in the natural environment and in farmed stocks. This article demonstrates the importance of 317

pathogen concentration measures during the first hours of contact with the host, improving 318

our knowledge and understanding of the infection dynamics of Vibrio pathogens in marine 319

mollusks. 320

17

Acknowledgments 321

This study was supported by Pole Mer Bretagne and the Université de Bretagne Occidentale. 322

We are also grateful to Flavia Nunes from the Laboratoire des sciences de l’Environnement 323

MARin – LEMAR and Sarah Culloty from the University College Cork for language support, 324

and to Isabelle Calves for her experimental help. 325

18

References 326

1. Austin B, Zhang XH. 2006. Vibrio harveyi: a significant pathogen of marine vertebrates 327

and invertebrates. Lett. Appl. Microbiology. 43: 119-124. 328

2. Austin B. 2010. Vibrios as causal agents of zoonoses. Vet. Microbiol. 140: 310-317. 329

3. Nicolas JL, Basuyaux O, Mazurie J, Thebault A. 2002. Vibrio carchariae, a pathogen of 330

the abalone Haliotis tuberculata. Dis. Aquat. Organ. 50: 35-43. 331

4. Travers MA, Basuyaux O, Le Goic N, Huchette S, Nicolas JL, Koken M, Paillard C. 332

2009. Influence of temperature and spawning effort on Haliotis tuberculata mortalities 333

caused by Vibrio harveyi: an example of emerging vibriosis linked to global warming. 334

Glob. Chang. Biol. 15: 1365-1376. 335

5. Travers MA, Barbou A, Le Goic N, Huchette S, Paillard C, Koken M. 2008. 336

Construction of a stable GFP-tagged Vibrio harveyi strain for bacterial dynamics 337

analysis of abalone infection. FEMS Microbiol. Lett. 289: 34-40. 338

6. Mims C, Nash A, Stephen J, Fitzgerald R. 2002. Mims' Pathogenesis of Infectious 339

Disease. In Academic Press (ed.), London, UK. 340

7. Finlay BB , Falkow S. 1997. Common themes in microbial pathogenicity revisited. 341

Microbiol. Mol. Biol. Rev. 61: 136–169 342

8. Olsson JC, Joborn A, Westerdahl A, Blomberg L, Kjelleberg S, Conway PL. 1996. Is 343

the turbot, Scophthalmus maximus. L., intestine a portal of entry for the fish pathogen 344

Vibrio anguillarum? J. Fish. Dis. 19: 225-234. 345

9. Svendsen YS, Bogwald J. 1997. Influence of artificial wound and non-intact mucus 346

layer on mortality of Atlantic salmon. Salmo salar L. following a bath challenge with 347

Vibrio anguillarum and Aeromonas salmonicida. Fish Shellfish Immunol. 7: 317-325. 348

19

10. Laurencin FB, Germon E. 1987. Experimental infection of Rainbow trout, Salmo 349

Gairdneri, by dipping in suspensions of Vibrio anguillarum - Ways of bacterial 350

penetration - Influence of temperature and salinity. Aquaculture. 67: 203-205. 351

11. Paillard C, Maes P. 1995. The brown Ring Disease in the Manila clam, Ruditapes 352

phillipinarum. J. Invertebr. Pathol. 65: 91-100. 353

12. Allam B, Paillard C, Ford SE. 2002. Pathogenicity of Vibrio tapetis, the etiological 354

agent of brown ring disease in clams. Dis. Aquat. Organ. 48: 221-231. 355

13. Gauger EJ, Gómez-Chiarri M. 2002. 16S ribosomal DNA sequencing confirms the 356

synonymy of Vibrio harveyi and V. carchariae. Dis Aquat Organ. 52(1):39-46. 357

14. Kaplan EL, Meier P. 1958. Nonparametric-estimation from incomplete observations. J. 358

Am. Stat. Assoc. 53: 457-481. 359

15. Sambrook J, Russell DW. 2001. Molecular Cloning: A Laboratory Manual. J.F. 360

Sambrook and D.W. Russell, eds., Cold Spring Harbor Laboratory Press, New York, 3rd 361

edition. 362

16. Schikorski D, Renault T, Paillard C, Bidault-Toffin A, Tourbiez D, Saulnier D. 2013. 363

Development of Taqman Real-Time PCR assay targeting both the marine bacteria 364

Vibrio harveyi, a pathogen of European abalone Haliotis tuberculata, and plasmidic 365

DNA harbored by virulent strains. Aquaculture. 392-395: 106-112. 366

17. Yuan JS, Reed A, Chen F, Stewart CN. 2006. Statistical analysis of real-time PCR data. 367

Bmc Bioinformatics. 7. 368

18. Wilcoxon F. 1945. Individual comparisons by ranking methods. Biometrics Bulletin 1: 369

80-83. 370

19. Altura MA, Heath-Heckman EAC, Gillette A, Kremer N , Krachler AM. 2013. The first 371

engagement of partners in the Euprymna scolopes-Vibrio fischeri symbiosis is a two-372

20

step process initiated by a few environmental symbiont cells. Environ. Microbiol. 373

15(11): 2937–2950. 374

20. Delapena LD, Nakai T, Muroga K. 1995. Dynamics of Vibrio sp in organs of orally 375

infected Kuruma prawn Penaeus japonicus. Fish Pathology. 30:39-45. 376

21. Somboonwiwat K, Supungul P, Rimphanitchayakit V, Aoki T, Hirono I , 377

Tassanakajon, A. 2006. Differentially expressed genes in hemocytes of Vibrio harveyi-378

challenged shrimp Penaeus monodon. J. Biochem. Mol. Biol. 39(1):26-36. 379

22. Somboonwiwat K, Chaikeeratisak V, Wang HC, Lo CF, Tassanakajon A. 2010. 380

Proteomic analysis of differentially expressed proteins in Penaeus monodon hemocytes 381

after Vibrio harveyi infection. Proteome Sci. 8: 39 382

23. Cabello AE, Espejo RT, Romero J. 2005. Tracing Vibrio parahaemolyticus in oysters. 383

Tiostrea chilensis. using a Green Fluorescent Protein tag. J. Exp. Mar. Biol. Ecol. 327: 384

157-166. 385

24. Sawabe T, Inoue S, Fukui Y, Yoshie K, Nishihara Y, Miura H. 2007. Mass mortality of 386

Japanese abalone Haliotis discus hannai caused by Vibrio harveyi infection. Microbes 387

Environ. 22: 300-308. 388

25. Crofts DR. 1929. Haliotis. Memoirs on Typical British Marine Plants and Animals. 29. 389

University Press of Liverpool, Liverpool, UK. 390

26. Manganaro M, Laura R, Guerrera MC, Lanteri G, Zaccone D, Marino F. 2012. The 391

morphology of gills of Haliotis tuberculata. Linnaeus, 1758. Acta Zoologica 93: 436-392

443. 393

27. Panicker G, Bej AK. 2005. Real-time PCR detection of Vibrio vulnificus in oysters: 394

Comparison of oligonucleotide primers and probes targeting vvhA. Appl. Environ. 395

Microbiol. 71: 5702-5709. 396

21

28. Gordon KV, Vickery MC, DePaola A, Staley C, Harwood VJ. 2008. Real-time PCR 397

assays for quantification and differentiation of Vibrio vulnificus strains in oysters and 398

water. Appl. Environ. Microbiol. 74: 1704-1709. 399

29. Blanco-Abad V, Ansede-Bermejo J, Rodriguez-Castro A, Martinez-Urtaza J. 2009. 400

Evaluation of different procedures for the optimized detection of Vibrio 401

parahaemolyticus in mussels and environmental samples. Int. J. Food. Microbiol. 129: 402

229-236. 403

30. Saulnier D, De Decker S, Haffner P. 2009. Real-time PCR assay for rapid detection and 404

quantification of Vibrio aestuarianus in oyster and seawater: A useful tool for 405

epidemiologic studies. J. Microbiol. Methods. 77: 191-197. 406

31. Baker-Austin C, Gore A, Oliver JD, Rangdale R, McArthur JV, Lees DN. 2010. Rapid 407

in situ detection of virulent Vibrio vulnificus strains in raw oyster matrices using real-408

time PCR. Environ. Microbiol. Rep. 2: 76-80. 409

32. De Decker S, Saulnier D. 2011. Vibriosis induced by experimental cohabitation in 410

Crassostrea gigas: Evidence of early infection and down-expression of immune-related 411

genes. Fish Shellfish Immunol. 30: 691-699. 412

22

Tables 413

Table 1: Percentages of cumulated mortalities in European abalone after 12 days of infection, 414

and lethal time during which 50% of abalone stock were dead (LT50), reported for each 415

contact duration with V. harveyi. Results correspond to the means of 3 experiment replicates 416

of 15 animals. 417

Contact duration % of mortalities after

12 days of infection

Standard

deviation LT50 *

1h 57.8 44.4 6.8

6h 84.4 10.2 5.3

9h 88.9 3.8 4.5

24h 93.3 6.7 3.7

* lethal time in days. 418

23

Table 2: Statistical significance of V. harveyi concentrations or densities in European abalone 419

organs after the first hours of contact. Differences were detected by Wilcoxon signed-rank test 420

(Wilcoxon, 1945). An asterisk (*) indicates a statistically significant difference at p < 0.1, two 421

asterisks (**) indicate a statistically significant difference at p < 0.05. 422

Quantification method Plating without ethanol treatment

qPCR Plating with ethanol treatment

Organ Timea 1 3 6 1 3 6 9 24 1 3 6

HLb

0 ** ** ** Ndc ** ** ** **

1 ** ** ** ** ** **

3 ** ** ns **

6 ns **

9 **

Gills

0 ** ** ** ** ** ** ** ** ** ** **

1 ** ** ** ** ** ** ns **

3 ** ** ** ** **

6 * ns

9 ns

Digestive gland

0 ** ** ** ** ** ** ** ** ** ** **

1 ns ** ns ns ns ns ** **

3 ** * ns ns ** 6 ns ns

9 ns

Muscle

0 ** ** ** ** ** ** ** ns ** ** **

1 ns ns ns ns ns ns ns **

3 ** ns ns ns *

6 ns ns

9 ns a duration of contact between the European abalone and V. harveyi, in hour. 423

b hemolymph. 424

C not detected. 425

24

Figure legends 426

Figure 1: Influence of contact duration with Vibrio harveyi on mortality kinetics in the 427

European abalone. 428

Animals were exposed to a 105 bacteria/mL challenge by immersion during one hour, 6h, 9h 429

and 24h at 19°C. Results correspond to the mean from 3 experimental replicates of 15 abalone 430

per tank. Bars indicate standard error rates. Control animals were only immersed in filtered 431

seawater. Comparison of mortality curves was performed with a Kaplan-Meier analysis 432

(Kaplan and Meier, 1958). 433

434

Figure 2: Vibrio harveyi densities in European abalone organs during the first hours of 435

contact. 436

Densities were expressed in number of bacteria per 100 mg of tissue. Quantification of GFP-437

tagged and kanamycin resistant V. harveyi ORM4 strain was done after tissue homogenate 438

plating on kanamycin-LBS agar dishes after overnight incubation at 28°C. 439

A: quantification in organs that were not treated with ethanol – adhered and penetrated 440

bacteria; B: quantification in organs that were washed in an ethanol bath – only penetrated 441

bacteria. 442

25

Figure 3: infection dynamics of Vibrio harveyi in European abalone tissues. 443

Concentrations or densities (bacteria count per mL of hemolymph or per 100 mg of tissue) 444

were estimated in hemolymph (A), gills (B), digestive gland (C) and foot-muscle (D), during 445

the first hours of contact. Results are given in averaged concentrations or densities of five 446

biological replicates, estimated by plating on kanamycin-LBS agar dishes from organs treated 447

or untreated with ethanol, and by qPCR using specific primers and a Taqman probe. Bars 448

indicate standard errors. No bacterium was detected in t0 samples, or in control individuals 449

immersed in filtered-seawater without V. harveyi suspension. 450

Cu

mula

tive

mo

rta

litie

s (

%)

40

20

0

100

80

60

0 1 2 3 9 4 5 11 10 6 12 7 8

Challenge time (in days)

control

1 hour

6 hours

9 hours

24 hours

control

1 hour

6 hours

9 hours

24 hours

Control one hour 6 hour 9 hour 24 hour

Fig. 1

100125 G

Di

0255075

100

ncen

tratio

n a

/ 100

mg

FoMEp

A0

10152025

Bac

teria

lcon

*103

bact

eria

Con0

0

510B

B

Con

FiFi

illsigestive glandoot-muscleouthpipodia

ntact duration1h 3h 6h

ntact duration (in hours)

i 2ig. 2

plating quantification froQ-PCR quantificationplating quantification fro

10 6de

nsiti

es

A

plating quantification frowith ethanol

10 6

10 2

10 4

entra

tion

or d

B10 6

10 2

10 4

cter

ialc

once

C10 6

10 2

10 4

Bac

D10 2

10 4

Contact 0 1h

FigFig

om untreated tissue

om tissue treatedom tissue treated

duration (in hours)3h 6h 9h 24h

g 3g. 3