CHARACTERISATION AND MOLECULAR TYPING OF CLINICAL AND ENVIRONMENTAL ISOLATES OF VIBRIO PARAHAEMOLYTICUS
by
Aklak Miah B.Sc. (Hons)
A thesis submitted to the University of Plymouth in partial fulfilment for the degree of
DOCTOR OF PHILOSOPHY
Department of Biological Sciences Faculty of Science
In collaboration with
The Centre for Environment, Fisheries and Aquaculture Science Weymouth Laboratories, Dorset, United Kingdom
2009
11 this th~si~ ls.qe(jh:atedlt(J. my'moiher, T. N: 'K:hatoon and .my:laie grandfather, A. Talukhdar
!Ill Abstract
Characterisation and Molecular Typing of Clinical and Environmental Isolates of Vibrio parahaemolyticus
Aklak Miah
Vibrio parahaemolyticus is a natural inhabitant of coastal waters worldwide and is the leading cause of seafood-borne gastroenteritis.This study reports on the use of several molecular characterisation methods to screen clinical and environmental isolates of V parahaemolyticus to assess whether such techniques can be used to distinguish pathogenic isolates reliably.
In a total of 86 isolates mainly of V parahaemolyticus but also including V cholerae, V vulnijicus and several other species, serotypes of the more virulent clonal group 03:K6 were identified, but otherwise there appeared no association with serotype and phenotype. The tdh and trh genes encoding haemolysins that are typically associated with virulent isolates were found in a significantly large number of isolates; however, poor concordance between haemolytic activity and the presence of the gene tdh was found. In an effort to establish more accurate relationships amongst clinical and environmental isolates of V parahaemolyticus, four molecular typing systems were employed; namely pulsed-field gel electrophoresis (PFGE), intergenic transcribed spacer (ITS) analysis, tDNA intergenic length polymorphisms (tDNA-ILPs) and randomly amplified polymorphic DNA (RAPD). Typing patterns and clustering analysis using these methods differentiated V parahaemolyticus from other marine species as well as at the subspecies level. PFGE with Noti was shown to be the most discriminative but suffered from not being universally applicable. Both ITS and tDNA-ILP methods were sufficiently discriminatory with discrimination indices (DI) of between 0.568 and 0.724, depending on the primers employed. The discriminatory ability of RAPD was also affected by the primers used (DI= 0.959 - 0.965) but closely matched that of PFGE (Dl = 0.976). Additionally, both RAPD methods were able to distinguish putative markers for the pandemic clonal group. Typing systems appeared largely stable in duplicate and triplicate analyses with multiple primer pairs with some obvious variability in the reproduction of faint amplicons. All methods except PFGE were simple to execute but none of the methods could distinguish V parahaemolyticus into obvious lineages based on the clinical or environmental source.
With the recent implication of a type Ill secretion system {TTSS) involved in the pathogenicity of V parahaemolyticus, a multiplex PCR system using PCR primers that spanned both TTSS l and TTSS2 regions was developed. Dot-blot analysis confirmed TTSS2 genes in at least 30% of environmental isolates. Nucleotide sequence analysis revealed lOO% sequence homology in three loci of TTSS2 putative structural genes. In comparison, a total of 34 single nucleotide polymorphisms (SNP) were identified in three TTSSI regions. In two of the regions, the SNPs were synonymous, whereas a non synonymous substitution in the structural gene vcrDJ resulted in valine replacement with isoleucine. In addition, nucleotide deletions in TTSS 1 with resultant frameshift mutations were identified. The finding that significant numbers of environmental isolates also possess TTSS2 genes is contrary to currently held opinion that TTSS2 is only present in clinical isolates. It is hypothesed that the high incidences of V parahaemolyticus infections may be related to active TTSS2 genes, whereas a high degree of polymorphisms in TTSS 1 suggest it may be inactive.
IV List of Contents
Copyright statement ...... Title...... 11 Dedication...... iii Abstract...... iv List of Contents...... v List of Figures...... x1 List of Tables...... xv1 List of Abbreviations...... xvm Acknowledgements...... xxi Author's Signed Declaration...... xxii
Chapter 1 -INTRODUCTION 1.1 Introduction...... I 1.2 Aims and Objectives...... 3
Chapter 2- LITERATURE REVIEW 2.1 The Organism...... 6 2.1.1 General characteristics...... 6 2.1.2 Genome organisation of V parahaemolyticus...... 6 2.1.3 Ecological niches...... 7 2.1.4 Viable but non-culturable state of V parahaemolyticus...... 9 2.2 Worldwide distribution of V: parahaemolyticus...... I 0 2.2.1 Europe...... 10 2.2.2 East Asia...... 11 2.2.3 Southern Asia and Africa...... 11 2.2.4 North America...... 13 2.2.S South America...... 13 2.3 Pandemic serotypes of V. parahaemolyticus...... 14 2.4 Virulence factors and pathogenicity...... IS 2.4.1 Human infection...... IS 2.4.2 Thermostable direct haemolysin (TDH)...... 16 2.4.3 Thermostable direct haemolysin related haemolysin (TRH)...... 17 2.4.4 Other haemolysins in V: parahaemolyticus...... 18 2.4.S Cytotoxicity and enterotoxicity of V: parahaemolyticus...... 18 2.4.6 Other potential pathogenicity determinants...... 19 2.4. 7 Serine protease...... 19 2.4.8 Adherence and colonisation factors...... 20 2.4.9 Siderophores...... 20 2.4.10 Insertion Sequences...... 21 2.4.11 Phage associated virulence...... 21 2.4.12 Type Ill secretion systems...... 22 2.S Conventional methods for identification...... 24 2.S.l ldentification and detection methods of V parahaemolyticus.. .. 24 2.S.2 [solation...... 24 2.S.3 Culture and growth...... 24 2.S.4 Biochemical profiling...... 2S 2.S.S API and commercial identification systems...... 26 2.S.6 Characterisation of pathogenic V parahaemolyticus...... 27 2.S.7 Immunological detection of pathogenic V: parahaemolyticus.... 28
V 2.6 Molecular detection methods...... 28 2.6.1 PCR methods for V parahaemolyticus...... 28 2.6.2 loxR gene based PCR methods...... 29 2.6.3 PCR targetted to tlh...... 30 2.6.4 Primers for gryB...... 30 2.6.5 PCR of pR72H...... 31 2.6.6 PCR methods for toxin genes...... 31 2.6. 7 Multiplex PCR...... 31 2.6.8 DNA hybridisation probes and ELISA methods...... 32 2.6.9 Quantitative real-time PCR (QRT-PCR)...... 33 2.7 Strain differentiation by molecular typing methods...... 33 2. 7 .I Epidemiological typing of clinically significant strains...... 33 2.7.2 Principles and properties of a molecular typing method...... 34 2.7.3 Pulsed field gel electrophoresis (PFGE)...... 35 2. 7.4 Restriction fragment length polymorphism (RFLP)...... 36 2.7.5 Randomly amplified polymorphic DNA (RAPD) PCR...... 37 2.7.6 Ribosomal DNA intergenic spacer (ITS) region analysis...... 39 2.7.7 Transfer DNA-PCR analysis...... 40 2.7.8 Repetitive extragenic palindromic sequence PCR (REP-PCR).... 41 2.7.9 Multi locus enzyme electrophoresis (MLEE)...... 43 2.7.10 Multi locus sequence typing (MLST)...... 43 2.7.11 Variable number tandem repeats PCR (VNTR-PCR)...... 44
Chapter 3- GENERAL MATERIALS AND METHODS 3.I Chemicals and reagents...... 4 7 3.2 Bacteriology...... 47 3.2.1 Isolates of V parahaemolyticus...... 47 3.2.2 Culture and growth of V. parahaemolyticus...... 48 3.2.3 Culture and growth of additional bacterial species...... 48 3.3 Bacterial genomic DNA extraction...... 52 3.3.1 Cell lysis and protein precipitation...... 52 3.3.2 DNA precipitation and hydration...... 53 3.4 DNA quantification...... 53 3.5 Primer design for PCR...... 53 3.5.1 Other PCR primers...... 53 3.6 Standard Polymerase Chain Reaction (PCR)...... 54 3.7 PCR product purification...... 55 3.8 Agarose Gel Electrophoresis...... 55 3.9 Restriction enzyme digests of genomic DNA...... 56 3.10 Analysis of banding profiles...... 56
Chapter 4- BIOCHEMICAL, GROWTH AND SEROLOGICAL VARIATION AMONGST VIBRIOPARAHAEMOLYTICUS 4.1 Introduction...... 58 4.2 Materials and Methods...... 60 4.2.1 Biochemical tests...... 60 4.2.2 Growth and survival characteristics...... 60 4.2.3 Thermostable Direct Haemolysin phenotype- 'Kanagawa test'.. 61 4.2.4 Determination of haemolytic activity levels...... 61 4.2.5 K and 0 antigenic variants (Serotyping)...... 62 4.3 Results...... 64 4.3.1 Biochemical tests...... 64 4.3.2 Growth rates of V parahaemolyticus by reduction ofNa+ ...... 64 4.3.3 Culturability of V parahaemolyticus isolates from -20°C...... 66
VI 4.3.4 General growth characteristics on selective and non-selective media...... 66 4.3.5 Beta-haemolysis and clinical phenotypes...... 71 4.3.6 Variation in 0 and K serotypes amongst V parahaemolyticus... 75 4.4 Discussion...... 76 4.4.1 General growth and biochemical properties...... 76 4.4.2 Growth of V parahaemolyticus in reduced salinities...... 77 4.4.3 Culturability of cells from cold storage at -20°C ...... 78 4.4.4 Beta-haemolysis attributed to thermostable direct haemolysin (TDH)...... 78 4.4.5 Urease production and TDH-related haemolysin (TRH)...... 81 4.4.6 Serotyping...... 81
Chapter 5- USE OF PULSED-FIELD GEL ELECTROPHORESIS TO SHO" CLONAL RELATIONSHIPS AMONGST VIBRIO PARAHAEMOLYTJCUS 5.1 Introduction...... 83 5.2 Materials and Methods...... 86 5.2.1 Preparation of V parahaemolyticus plugs...... 86 5.2.2 Bacterial cell lysis and de-proteination...... 86 5.2 .3 Restriction endonuclease digestion of agarose plugs...... 87 5.2.4 Agarose gels and running conditions...... 87 5.2.5 Standards and control organisms...... 88 5.2.6 Data analysis and dendrogram construction...... 88 5.3 Results...... 89 5.3.1 PFGE of V parahaemo/yticus...... 89 5.3.2 PFGE of V cholerae...... 89 5.3.3 Sfii digested restriction profiles of V parahaemolyticus...... 89 5.3.4 Noti digested restriction patterns of V parahaemo/yticus...... 90 5.3.5 Dendrogram analysis of Not I restriction PFGE patterns...... 91 5.4. Discussion...... 104 5.4.1 Clustering of clinical isolates...... I 04 5.4.2 Environmental isolates showing homology to clinical isolates... 1OS 5.4.3 Typeability with PFGE...... 106 5.4.4 Standardisation ofPFGE...... 108 5.4.5 Conclusion...... 109
Chapter 6- DIFFERENTIATION OF VIBRIO PARAHAEMOLYTJCUS ISOLATES BASED ON LENGTH POLYMORPHISMS OF THE 16S- 23S INTERNAL TRANSCRIBED SPACER (ITS) REGION 6.1 Introduction...... 111 6.2 Materials and Methods...... 113 6.2.1 Primers for amplification ofiTS regions...... 113 6.2.2 Amplification of ITS 1 and ITS2 regions...... 113 6.2.3 Data analysis and dendrogram construction...... 114 6.3 Results...... liS 6.3 .I Consensus primers for ITS I regions ...... liS 6.3.2 PCR of 23S-SS (ITS2) intergenic spacer regions...... 115 6.3.3 PCR of 16S-23S (ITSI) intergenic spacer regions of V parahaemolyticus...... 116 6.3.4 Reproducibility ofiTS 1-PCR banding profiles...... 116 6.3.5 ITS1 PCR in Vibrio species...... 117 6.3.6 Dendrogram analysis and differentiation of V parahaemolyticus isolates...... 117 6.4 Discussion...... 127
Vll 6.4.1 Primers used in the study...... 127 6.4.2 ITS1-PCR banding patterns...... 128 6.4.3 Properties of the ITS1-PCR system...... 129 6.4.4 Sequence based analysis of ITS I...... 130 6.4.5 ITS I spacer classes in V parahaemo/yticus...... 131
Chapter 7- DNA PROFILING OF VIBRIO PARAHAEMOLYTICUS USING tDNA-PCR TO GIVE INTERGENIC LENGTH POLYMORPHISMS 7.1 Introduction ...... 135 7.2 Materials and Methods ...... 137 7.2.1 tDNA primer screening with V parahaemolyticus ...... 137 7.2.2 PCR amplification oftDNA intergenic spacer regions using consensus pnmers ...... 137 7.2.3 Dendrogram analysis oftDNA-PCR ...... 137 7.2.4 Concordance between tDNA fingerprint types ...... 138 7.3 Results ...... 140 7.3.1 Consensus tDNA primer combinations ...... 140 7.3.2 tDNA-PCR subtyping based on primers T3A and T5A (tDI) .. .. 140 7.3.3 tDNA-PCR subtyping based on primers T3B and T5A (tD2) .. .. 141 7.3.4 tDNA-PCR subtyping based on primers T3A and T3B (tD3) ... . 142 7.3.5 Comparative analysis oftDNA fingerprinting with different primers ...... 142 7.3.6 Combined dendrogram analysis oftDNA-PCR...... 143 7.3.7 Reproducibility oftDNA-PCR ...... 144 7.3.8 tDNA-ILPs in other species ...... 144 7.4 Discussion ...... 157 7.4.1 Fingerprinting based on tDI, tD2 and tD3 ...... 157 7.4.2 Consensus and specific tDNA primers ...... 161 7.4.3 General considerations from tDNA-PCR typing ...... 161 7 .4.4 Typing of V parahaemolyticus based on tDNA spacer regions .. 163
Chapter 8- RANDOMLY AMPLIFIED POLYMORPHIC DNA (RAPD) METHOD FOR DIFFERENTIATING VIBRIO PARAHAEMOLYTICUS 8.1 Introduction...... 164 8.2. Materials and Methods...... 167 8.2.1 RAPD- PCR of V parahaemolyticus...... 167 8.2.2 Screening ofRAPD primers PI to P6, singly and in pairs...... 167 8.2.3 Dendrogram analysis with BioNumerics...... 168 8.2.4 Concordance between RAPD -PSI and PS2 typing...... 168 8.3Results...... 169 8.3.1 Single primer (P2) screening of V parahaemolyticus...... 169 8.3.2 Two-primer screening ofRAPD reactions...... 169 8.3.3 RAPD typing of V parahaemolyticus and related species 165 with PS 1...... 169 8.3.4 RAPD fingerprinting of V parahaemolyticus based on PS2...... 171 8.3.5 Reproducibility of RAPD-PCR...... 172 8.3.6 Comparison ofRAPD-PSI and RAPD-PS2...... 172 8.3. 7 Differentiation of clinical and environmental isolates...... 173 8.4 Discussion...... 185 8.4.1 Typing of V parahaemolyticus based on PS 1 ...... 186 8.4.2 Typing of V parahaemolyticus based on PS2...... 187 8.4.3 A marker for pandemic serovars? ...... 188
Vlll 8.4.4 Comparison of single and two-primer RAPD methods...... 188 8.4.5 Data analyses...... 190 8.4.6 General considerations of a RAPD typing system...... 190
Chapter 9- VARIATION IN THE TYPE DI SECRETION SYSTEM AMONGST CLINICAL AND ENVIRONMENTAL ISOLATES OF VIBRIO PARAHAEMOLYTICUS 9.1 Introduction...... 192 9.2 Materials and Methods...... 194 9 .2.1 PCR detection of TTSS 1 and TTSS2 genes...... 194 9.2.2 Development of Multiplex PCRs...... 195 9.2.3 Production ofTTSS 1 and TTSS2 specific DIG labelled probes.. 198 9.2.4 Preparation of dot-blots...... 198 9.2.5 DNA-DNA Hybridisation...... 199 9.2.6 Immunological detection of DNA-DNA hybrids...... 199 9.2.7 Sequencing and analysis ofTTSS genes...... 200 9.3 Results...... 200 9.3.1 PCR primers targeted to the TTSS loci...... 200 9.3 .2 PCR amplification of genes from the TTSS I region...... 20 I 9.3.3 PCR amplification of genes from the TTSS2 region...... 201 9.3.4 TTSS probes specificity in dot blots...... 202 9.3.5 Dot blot analysis ofTTSSI genes vcrDJ, vp/680 and vscCJ..... 202 9.3.6 Dot blot analysis ofTTSS2 genes vscC2, vcrD2 and vpa/321... 203 9.3.7 Conserved sequence identities of V parahaemolyticus TTSS genes...... 204 9.3.8 Correlation of phylogenetic clustering of isolates with the presence ofTTSSI and TTSS2 genes...... 204 9.3.9 V parahaemolyticus TTSS regions in other species...... 205 9.4 Discussion...... 217 9.4.1 Gene targets on the TTSS system of V parahaemolyticus...... 217 9.4.2 Detection ofTTSS regions...... 218 9.4.3 Distribution ofTTSS I genes in V parahaemolyticus...... 219 9.4.4 Distribution ofTTSS2 genes in V parahaemolyticus...... 220 9.4.5 Sequence diversity ofTTSS I regions...... 221 9.4.6 TTSS genes as epidemiological markers for virulent strains...... 223 9.4.7 Identification ofTTSSI and TTSS2 genes in other species...... 223
Chapter 10- A COMPARISON AND EVALUATION OF METHODS USED FOR DIFFERENTIATING VIBRIO PARAHAEMOLYTICUS I 0. L Introduction...... 225 I0.2 Materials and Methods...... 226 10.2.1 Phenotypic and growth characteristics of V parahaemolyticus. 226 I 0.2.2 Discrimination of molecular typing systems...... 226 I0.2.3 Concordance of molecular typing systems...... 226 I 0. 3 Results...... 227 10.3.1 Growth of V parahaemolyticus...... 227 10.3.2 Haemolytic characteristics...... 228 I 0.3 .3 Serovar diversity of V parahaemolyticus...... 228 10.3.4 Comparative analysis ofpulsed-fie1d gel electrophoresis...... 229 10.3 .5 16S-23S intergenic spacer (ITS I) based typing...... 231 10.3.6 tDNA-ILP based typing...... 232 10.3.7 Randomly amplified polymorphic DNA...... 233 I0.3.8 Evaluation of the TTSS2 in V parahaemolyticus...... 234 ix 10.4 Discussion...... 245 10.4.1 Phenotypic methods for isolate differentiation...... 245 10.4.2 Typing methods of V parahaemolyticus...... 246 10.4.3 General reliability of clustering patterns...... 249 10.4.4 Differentiation of clonally related V parahaemolyticus...... 250 I 0.4.5 Conclusion...... 25 I
Chapter 11 -GENERAL DISCUSSION AND CONCLUSION I 1.1 Introduction...... 253 11.2 Scope and feasibility of the present study...... 254 11.3 Genome plasticity of V parahaemolyticus...... 255 11.4 The differentiation of clinically significant V parahaemo/yticus...... 257 I 1.5 Conclusions...... 259
APPENDIX I- MEDIA, REAGENTS, SOLUTIONS AND SUPPLIERS 1.1 Bacteriological media...... 262 1.2 Genomic DNA isolation and purification...... 264 I.3 General molecular biology reagents...... 264 1.4 Pulsed field gel electrophoresis buffers...... 254 1.5 DNA purification kit solutions...... 266 1.6 Description of blotting solutions and buffers...... 266 1.7 DNA labelling and detection...... 267 1.8 Molecular weight markers...... 269 1.9 Suppliers...... 271
APPENDIX 11- ADDITIONAL DATA AND IMAGES Il.l Bacterial DNA quantification...... 273 II.2 Phenotypic growth characteristics of V parahaemolyticus...... 277
APPENDIX Ill -ADDITIONAL SEQUENCE DATA Ill.l Sequencing data relevant to Chapter 9...... 281
LIST OF REFERENCES...... 290
X List of Figures
Figure 2.1: Electron micrograph of a swimmer cell (A) and swarmer cell (B) types of V parahaemolyticus...... 8
Figure 2.2: Reported cases of V. parahaemolyticus food poisoning outbreaks...... 12
Figure 2.3: Unrooted trees constructed by the neighbor-joining method showing the phylogenetic interrelationships of the different Vibrio and Photobacterium species based on toxR gene nucleotide sequences (A) and 168 rRNA genes (B)...... 29
Figure 4.1: Growth rate phenotypes of clinical (C) and environmental (E) isolates of V parahaemolyticus at 0.5 and 3.0% NaCI...... 65
Figure 4.2: Representative growth of V parahaemolylicus clinical isolate 2436 (plate A) and environmental isolate VP4 (plate B) on TCBS agar after 18 hours at 37°C...... 69
Figure 4.3: Exopolysaccharide production amongst two environmental isolates of V parahaemolylicus...... 70
Figure 4.4: Variation in ~-haemolytic levels of rabbit erythrocytes on Wagatsuma agar after 18 hours at 37°C...... 72
Figure 4.5: Zones of haemolysis in clinical and environmental isolates of V parahaemolyticus...... 73
Figure 4.6: Zones of partial haemolysis attributed to TDH from thermostable ECP.. 74
Figure 4.7: Haemolytic activity of V parahaemolyticus...... 74
Figure 5.1: PFGE profiles of restriction Not! digested and undigested V. parahaemolyticus of various cell densities and treatments using BioRad method.. 93
Figure 5.2: PFGE profiles of No/land Sfi1 digested environmental isolate V parahaemolyticus VPI and V cholerae isolate VC4...... 94
Figure 5.3: PFGE profiles of Cla!, No/1 and Sfi! restriction digested V parahaemolyticus...... 95
Figure 5.4: Comparison of PFGE profiles of Noli restriction digested V parahaemolylicus clinical and environmental isolates...... 96
Figure 5.5: Comparison of PFGE profiles of Noli restriction digested V parahaemolylicus clinical and environmental isolates ofUK origin...... 97
Figure 5.6: Comparison ofPFGE profiles of Noli restriction digested V parahaemolylicus clinical and environmental isolates from UK, Spain and Japan.. 98
Figure 5.7: Comparison ofPFGE profiles of Noli restriction digested V parahaemolyticus clinical and environmental isolates from UK and Japan...... 99
XI Figure 5.8: PFGE profiles of Notl digested UK environmental and UK clinical 100 strains of V parahaemolylicus ...... 0 •••••••••••••••••••••••••••••••••••••••••••••••••••••••
Figure 5.9: PFGE profiles of Notl digested environmental isolates of V. parahaemolyticus and other Vibrio species ...... 101
Figure 5.10: Comparison of Not! digested Norwegian clinical and UK environmental isolates of V parahaemolyticus ...... 0 •••••••••••••••••••• 102
Figure 5.11: UPGMA dendrogram of clinical and environmental isolates of V parahaemolyticus ...... 103
Figure 6.1: Schematic representation of the V parahaemolyticus rRNA operon ...... 113
Figure 6.2: ITS 1-PCR banding profiles shown on agarose gel electrophoresis after PCR amplification at different temperatures ...... 120
Figure 6.3: PCR amplification ofiTS2 regions of V parahaemolyticus using primers designed for conserved 23S and SS regions ...... 0 •••••••••••••••• 121
Figure 6.4: PCR amplification of ITS I regions of clinical and environmental isolates of V parahaemolylicus and other species using primers VINTF and VINTRo 121
Figure 6.5: ITS I fingerprinting patterns of clinical and environmental isolates of V parahaemolyticus and related species using primers VINTF and VINTR ...... 0 •• 122
Figure 6.6: Comparison of ITS I fingerprinting patterns of clinical isolates of V parahaemolyticus from Norway, UK and Japan using primers VINTF and VINTR ... 122
Figure 6. 7: Comparison of ITS I fingerprinting patterns of environmental isolates of V parahaemolyticus from UK and Spain ...... 0 •••••••••••• 123
Figure 6.8: Comparison of ITS I fingerprinting patterns of environmental isolates of V parahaemolyticus, V cholerae and other species ...... 123
Figure 6.9: Comparison ofiTS-PCR to differentiate representative V parahaemolyticus from other members of Vibrio spp., and related bacteria ...... 125
Figure 6.10: Length variations of 16S-23S (ITS I) spacers found in clinical and environmental V parahaemolyticus ...... 125
Figure 6.11: ITSl-PCR banding profiles from 56 isolates of V parahaemolyticus used in the study ...... 0. 0 •• 0 0 ••••••••••••••••• 0 •••••••••••••• 126
Figure 6.12: Schematic representation of the ITSI conserved regions in V parahaemolyticus ...... 132
Figure 7.1: tDNA-PCR fingerprinting profiles with all 6 combinations oftDNA primers using a panel strain of 3 clinical and 3 environmental isolates ...... 146
Figure 7.2: tDNA-PCR fingerprints of 56 isolates of V parahaemolyticus and related species of clinical and environmental origin using primers T3A and TSA ...... 147
xii Figure 7.3: Dendrogram based on UPGMA clustering of normalised tDNA-ILPs of V parahaemolyticus obtained with primers T3A and T5A...... 148
Figure 7.4: tDNA-PCR fingerprints of 56 isolates of V parahaemolylicus and related species of clinical and environmental origin using primers T3B and T5A...... 149
Figure 7.5: Dendrogram based on UPGMA clustering of normalised tDNA-ILPs of V parahaemolyticus obtained with primers T3B and T5A...... 150
Figure 7.6: tDNA-PCR fingerprints of 56 isolates of V parahaemolyticus and related species of clinical and environmental origin using primers T3A and T3B...... 151
Figure 7.7: Dendrogram based on UPGMA clustering of normalised tDNA-ILPs of V parahaemolyticus obtained with primers T3A and T3B...... 152
Figure 7.8: Similarity plots oftDNA-PCR fingerprint types with three primer pairs...... 15 3
Figure 7.9: Combined dendrogram of composite tDI, tD2 and tD3 fingerprint data.. 154
Figure 7.10: Reproducibility of tDNA-PCR fingerprints from six isolates in repeat tests 1 and 2 with all three primer pairs...... 155
Figure 7.11: Dendrogram showing relatedness based on UPGMA clustering of tDNA-ILPs of V parahaemolyticus and 8 related species...... 156
Figure 8.1: RAPD single primer (P2) screening of V parahaemolyticus environmental isolates, VP350 and VPSC I...... 174
Figure 8.2: RAPD double primer screening of V parahaemolyticus environmental isolate VP350...... 174
Figure 8.3: RAPD analysis of a panel strain of 16 representative clinical and environmental isolates of V parahaemolyticus with primers P2 and P3 (PS 1)...... 175
8.4: RAPD banding patterns of Norwegian clinical isolates of V parahaemolyticus with PS 1...... 175
Figure 8.5: RAPD banding patterns ofUK and Spanish environmental isolates of V parahaemolyticus with PS 1...... 176
Figure 8.6: RAPD banding patterns of Japanese clinical isolates and UK environmental isolates of V parahaemolyticus with PS I and related species...... 176
Figure 8. 7: RAPD banding patterns of environmental isolates of V parahaemolyticus, V cholerae and related species with PSI...... 177
Figure 8.8: RAPD banding patterns of environmental and clinical isolates of V. parahaemolyticus and related species with PS 1...... 177
Figure 8.9: A unique RAPD marker (Band X) amongst clinical strains of V parahaemolyticus...... 178
Xlll Figure 8.10: A UPGMA dendrogram generated from curve based correlation coefficients for normalised RAPD patterns of V parahaemolyticus with PS 1...... 179
Figure 8.11: RAPD fingerprinting of 56 clinical and environmental isolates of V parahaemolyticus with PS2 ...... ~.... 180
Figure 8.12: A UPGMA dendrogram generated from curve based correlation coefficients for normalised RAPD patterns of V parahaemolyticus with PS2...... 182
Figure 8.13: Reproducibility ofRAPD fingerprinting profiles in six isolates of V parahaemolyticus...... 183
Figure 8.14: Concordance of stratification by RAPD-PS1 and RAPD-PS2 for 56 isolates of V parahaemolyticus...... 184
Figure 9.1: Genetic organisation ofTTSS's from V parahaemolyticus RIMD221 063 3...... I 97
Figure 9.2: PCR amplification at annealing temperatures of (A) 63°C, (B) 64°C and (C) 65°C for TTSS 1 and TTSS2 genes in NCIMB type strain 1902 in single and multiplex PCR reactions...... 206
Figure 9.3: PCR ofTTSS loci in clinical isolates of V parahaemolyticus from Japan, UK and Norway...... 207
Figure 9.4: PCR ofTTSS loci in 34 environmental isolates of V parahaemolyticus from UK and Spain...... 208
Figure 9.5: PCR and hybridisation signals in Vibrio spp...... 209
Figure 9.6: Hybridisation signals of 55 isolates of V parahaemolyticus with Probe A targeted to the vcrDllocus ofTTSSl...... 210
Figure 9.7: Hybridisation signals of 55 isolates of V parahaemolyticus with Probes Band C targeted to the vpl680 and vscCJ loci ofTTSS I respectively...... 211
Figure 9.8: Hybridisation signals of 55 isolates of V parahaemolyticus with Probes E and F targeted to the vscC2 and vscD2 loci of TTSS2 respectively...... 212
Figure 9.9: Hybridisation signals of 55 isolates of V parahaemolyticus with ProbeD targeted to the vpa132Jiocus downstream of the TTSS2 region of'pathogenicity island' (Vp-PA1)...... 213
Figure 9.10: RAPD-PS 1 clustering ofTTSS2 positive isolates...... 213
Figure 9.11: Multiple alignments of the vcrDllocus ofTTSS 1 in isolates of V parahaemo/yticus...... 215
Figure 9.12: Multiple alignments of the vpa 1321 locus of TTSS2 in isolates of V parahaemolyticus ...... _ 216
Figure 10.1: Summary of concordance of all molecular typing systems...... 236
XIV Figure 10.2: Concordance of stratification by PFGE and RAPD for V parahaemolyticus isolates...... 237
Figure 10.3: Concordance of stratification by PFGE and tDNA-ILP typing (A), and PFGE for ITSI typing (B) for V parahaemolyticus isolates...... 238
Figure 10.4: Concordance of stratification by ITS 1-PCR and RAPD for V parahaemolyticus isolates...... 239
Figure 10.5: Concordance of stratification by tDNA-lLP and RAPD for V parahaemolyticus isolates...... 240
Figure 10.6: Concordance of stratification by tDNA-ILP (tDl) and ITSI-PCR for V parahaemolyticus isolates...... 241
Figures located in appendices
Figure 11.1: Quantification of DNA extracted from V parahaemolyticus and other bacterial cultures...... 273
Figure 111.1: Complete nucleotide sequence of vcr D 1...... 281
Figure 111.2: Complete nucleotide sequence of vp1680...... 282
Figure 111.3: Complete nucleotide sequence of vscC1...... 283
Figure 01.4: Complete nucleotide sequence of vpa 1321...... 284
Figure 111.5: Complete nucleotide sequence ofvscC2...... 284
Figure 111.6: Complete nucleotide sequence ofvcrD2...... 285
Figure Ill. 7: Nucleotide sequences and multiple alignments of the vp1680 locus of TTSS I in isolates of V parahaemolyticus...... 286
Figure 111.8: Nucleotide sequences and multiple alignments of the vscCJlocus ofTTSSI in isolates of V parahaemolyticus...... 287
Figure 111.9: Nucleotide sequences and multiple alignments of the vscC2locus ofTTSS2 in isolates of V parahaemolyticus...... 288
Figure 111.10: Nucleotide sequences and multiple alignments of the vcrD2 locus ofTTSS2 in isolates of V parahaemolyticus...... 289
XV List of Tables
Table 1.1: List of abbreviations...... xviii
Table 2.1: Overview of V parahaemolyticus genome organisation in a clinical strain RTMD2210633...... 7
Table 3.1: Description of V parahaemolylicus and their genotypes, where known, that were used in this study...... 49
Table 3.2: Sequences of previously described PCR primers for toxR, and putative V .parahaemolyticus virulence genes tdh and trh...... 50
Table 3.3: Description of additional bacterial strains, DNA, and their genotypes, where known that were used in this study...... 51
Table 3.4: PCR primers used in this study...... 54
Table 4.1: Polyvalent and monovalent antisera containing agglutinins specific to K type...... 63
Table 4.2: Anti genic scheme of V parahaemolyticus 0 groups and K serotypes...... 63
Table 4.3: Serotypes and selected phenotypes of V parahaemolyticus isolates...... 67
Table 4.4: Distribution of 0-antigens in isolates of V parahaemolyticus...... 76
Table 6.1: Reproducibility ofiTSl-PCR bands in repeat tests...... 124
Table 6.2: Comparison of spacer classes identified in different V parahaemolylicus. 132
Table 7.1: Description of consensus tDNA primers used in this study...... 137
Table 8.1: RAPD Primers used in this study...... 167
Table 9.1: TTSS gene targets in V parahaemolyticus used in this study...... 194
Table 9.2: PCR primer sequences used to detect TTSS 1 and TTSS2 regions...... 196
Table 9.3: Vibrio species tested for TTSS genes of V parahaemolyticus...... 209
Table 9.4: Comparison of the presence ofTTSS related genes A-F, virulence characteristics and phylogenetic RAPD Typing clusters in V parahaemolyticus...... 214
Table 10.1: Discrimination of V parahaemolyticus typing methods used in this study...... 235
Table 10.2: Comparison of V parahaemolyticus TTSS2 isolates and typing groups to distinguish virulent and potentially virulent strains...... 242
Table 10.3: Comparison ofTTSS2 unconfirmed (UC) or TTSS2 confirmed negative (-) isolates of V parahaemo/ylicus to potentially distinguish avirulent strains...... 243 XVI Table 10.4: Analysis of regression in comparison of typing systems...... 244
Tables located in appendices
Table 11.1: Genomic DNA concentrations determined from agrose gels...... 276
Table 11.2: Growth characteristics of V parahaemolyticus on non-selective and selective solid media...... 277
Table 11.3: Comparison of growth rates of clinical and environmental isolates at 0.5 and 3% NaCl in TSB, pH 7.2...... 279
XVII List of Abbreviations
ADP Adenine diphosphate ATP Adenosine triphosphate bp base pair CEFAS Centre for Environment Fisheries and Aquaculture Science cfu Colony forming units CHEF contour-clamped homogenous field css cell suspension solution Da Dalton dATP deoxyadenosine triphosphate dCTP deoxycytosine triphosphate dGTP deoxyguanosine triphosphate dNTP deoxynucleotide triphosphate dTTP deoxythymidine triphosphate ddH20 distilled and deionised water DIG digoxigenin Dig-11-dUTP digoxigenin deoxyuracil triphosphate DMF dimethyl formamide DMSO dimethyl sulfoxide DNA deoxyribonucleic acid DNase deoxyribonuclease ECP extracellular products EDTA ethylenediaminetetraacetic acid ELlS A enzyme-linked immunosorbent assay Eps Exopolysaccharides ERIC enterobacterial repetitive intergenic consensus FDA Food and Drug Administration of the USA FIGE field inversion gel electrophoresis G guanine g gram h hour ITS internal transcribed spacer region ITSI 16S-23S rRNA internal transcribed spacer Region ITS2 23S-5S rRNA internal transcribed spacer region IS insertion sequence kb kilo base kg kilogram A. lambda L litre LBA Luria-Bertani agar LBB Luria-Bertani broth LPS lipopolysaccharide J.lg mtcrogram J.ll micro litre J.I.M micromolar M molar M-PCR Multiplex PCR mA miliampere Mb megabase
xviii MLEE multilocus enzyme electrophoresis mg miligram m in minute mM millimolar mRNA messenger ribonucleic acid MSA marine salts agar NCBI National Centre for Biotechnology Information Ng nanogram nt nucleotide NBT nitroblue tetrazolium Ol:Kl serotype of somatic 0 I antigen and capsular Kl antigen OD6oo optical density at 600nm wavelength ORF open reading frame PAGE polyacryamide gel electrophoresis PBS phosphate buffered saline PCR polymerase chain reaction PFGE pulsed field gel electrophoresis pmol picomolar PMSF phenylmethylsulfonylfluoride Q-PCR quantitative real time PCR RAPD randomly amplified polymorphic DNA REP repetitive extragenic palindromic RFLP restriction fragment length polymorphisms RNA ribonucleic acid RNase ribonuclease rpm revolutions per minute spin speed rRNA ribosomal RNA RT room temperature RT-PCR reverse transcription PCR s seconds SDS sodium docecyl sulphate ssc sodium salt citrate TAE tris-acetate EDT A buffer TBE tris-borate EDT A buffer TCBS Thiosulphate Citrate Bile Salt agar medium tdh thermostable direct heamolysin gene TE Tris-ethylenediaminetetraacetic acid tlh thermolabile haemolysin gene tRNA transfer RNA tDNA DNA encoding for tRNA toxR ToxR transmembrane protein gene trh thermostable direct heamolysin related gene Tris tris (hydroxymethyl) aminomethane tRNA transfer ribonucleic acid TSA tryptone soya agar TSB tryptone soya broth TTSS type lii secretion system u unit UPGMA unweighed pair group method using arithmetic averages ure urease gene cluster for urea hydrolysis uv ultra violet xix V volts v/v -vcilume:·pel' ·volume, w/.v ~eight1per volume
xx Acknowledgements
I am most indebted to Dr Martyn Gilpin who was Director of Studies, for the continual day-to-day supervision and advice throughout the course of this study. My deep appreciations extend to my supervisors Dr Colin Munn and Dr Rachel Rangdale for their thoughts on my work and reading through my thesis. Thank you Rachel for giving me the opportunity to study for a PhD, organising the funding through CEF AS throughout the full duration of my studies and the opportunity to carry out some of the serotyping and BioNumerics data analysis in Weymouth.
In addition to my supervisors, I would like to extend my thanks to Sariqa Wagley for all the collaboration and sending me most of the isolates used in the study; Dr Hilton Grayson for assistance in the design of multiplex PCR primers and Lynne Cooper for her friendship, assistance and moral support throughout the last few years.
I wish to acknowledge colleagues and friends on the School of Biological Sciences for their technical expertise, sense of humour and who have made my four years in the lab not only endurable but also enjoyable. They include: Viv Richards, Paul Waines, Kathryn Evans, Zoe Lyle and Ehab Metwali. This list is by no means an exhaustive list so apologies for those who I have missed. A warm thank you to all the technical staff in School of Biological Sciences without whom the work would not have been possible.
Finally, I would like to thank my parents and family for their enduring love and support over the last few years, in a task that inevitably proved more difficult than expected.
XXI Author's signed declaration
At no time during the registration for the degree of Doctor of Philosophy has the author been registered for any other University award without prior agreement of the Graduate Committee.
This study was financed with the aid of a part-time studentship from the Centre for Environment, Fisheries and Aquaculture Science (CEF AS) and carried out in collaboration with CEFAS, Weymouth Laboratory, Dorset UK.
A programme of advanced study was undertaken, which included postgraduate research training (MSc modules EAR51 01, ENV 5101 and IMS51 01) on research methods and skills.
Relevant scientific seminars and conferences were regularly attended at which work was often presented; external institutions were visited for consultation purposes and several papers are in preparation for publication.
Publications - In Preparation
Population diversity of clinical and environmental Vibrio parahaernolyticus based on polymorphisms of 16S - 23S and tDNA spacer regions.
Identification of Kanagawa positive V. parahaernolyticus in British coastal waters and markers for the clonally related pandemic 03:K6 subpopulation.
Type Ill secretion system two (TTSS2) of virulent Vibrio parahaernolyticus is conserved and is also found distributed amongst Kanagawa negative environmental isolates and other Vibrio spp.
Conferences
Miah, A., L. Cooper, S. Wagley, R.E. Rangdale, C.B.Munn and M. Gilpin. Randomly amplified polymorphic DNA studies to evaluate discrimination between virulent and avirulent isolates of Vibrio parahaemolyticus. Vibrio 2005, Ghent, Belgium, 6-8th November 2005.
Wagley, S., R.E. Rangdale and A. Miah. Identification of the Pandemic 03:K6 clone in cases of Vibrio parahaemolyticus related illness in the U.K. 6th International Conference ofMolluscan Shellfish Safety. Marlborough, New Zealand. 18-23rd March 2007.
Miah, A., S. Wagley, R.E. Rangdale, L. Cooper, C.B.Munn and M. Gilpin. Variation in the type III secretion system of clinical and environmental isolates of Vibrio parahaernolyticus. Vibrio 2007, Paris, France, 28th November- 1st December 2007 (abstract accepted).
xxii Oral presentations
'Lost in Letters- Bioinformatics: Genomic Sequence Analysis using Artemis and Artemis Comparison Tool' Biomolecular Scholarship Group. School of Biological Sciences, University of Plymouth. October. 2003.
'Vibrio parahaemolyticus- a menace of warm coastal waters' Ecotoxicology and Stress Biology Research Seminar, School of Biological Sciences, University of Plymouth. October. 2003.
'Pathogencity of Vibrio parahaemolyticus' CEFAS Weymouth Laboratory, October 2005.
Otber relevant conferences, courses and workshops attended
Peninsula Microbiology Forum. One day meeting sponsored by the SGM and SfAM. Marine Biological Association. March 2003
Society for General Microbiology. l52nd Meeting. Microbial Subversion of Host Cells. University of Edinburgh, 7-llth April 2003
Society for General Microbiology. l53rd Meeting. Exploiting Genomes: Bases to Megabases in 50 years. UMIST, 8-ll th September 2003
Bioinformatics Workshop- Sequence Analysis and Comparison with Artemis and the Artemis Comparison Tool (ACT). Wellcome Trust Sanger Institute & the SGM, University of Bristol, August 2003
Bioinformatics Summer School- BCS & School of Computing, University of Exeter 30th August to 2nd September 2004
BioNurnerics Workshop- The 31st International Training Workshop on the use of GelCompar 11 and BioNumerics. Applied Maths, Belgium. April llth & 12th, 2005.
External Contacts:
Dr Rachel Rangdale CEFAS- Weymouth Laboratory Barrack Road The Nothe, Weymouth, Dorset, DT4 8UB
Word count of main body of thesis: 61,762
Signed ..... ~--~------Dot< ...... 12.) 9~JP~ ......
xxiii Chapter One
INTRODUCTION
1.1 Introduction
Vibrio parahaemolyticus IS a bacterium found in the marine environment. It is phylogenetically close to V cholerae, the etiological agent of cholera and can cause human disease upon ingestion of seafood contaminated with the bacteria. Worldwide, V parahaemolyticus is a significant cause of food borne gastroenteritis, especially in warmer coastal regions. In Japan and the USA it is the leading cause of seafood related gastroenteritis (Pan et al., 1997). In the UK and Europe, V parahaemolyticus is an
increasing concern due to the rise in consumption of raw or undercooked seafood and the rise of seawater temperatures. The first UK outbreak was recorded in 1974 when 12 holiday makers became ill after eating dressed crab (Hooper et al., 1974) and since then
there have been several other reports of sporadic outbreaks of V parahaemolyticus food
poisoning in Europe (Hervio- Heath et al., 2002; Lozano-Leon et al., 2003; Mame &
Aznar, 1988; Robert-Pillot et al., 2004). Although most harvested fish and shellfish, especially those from brackish, estuarine and coastal environments contain V parahaemolyticus, only a small proportion of isolates are thought to be pathogenic (Cook
et al., 2002; DePaola et a/, 2003a; Deepanjali et al., 2005).
For many years the isolation and distinction of V parahaemolyticus from the closely
related Vibrio species that inhabit marine and estuarine habitats, as well as from
contaminated seafood, have proved to be arduous and long. The detection of virulent V
parahaemolyticus amongst a numerically greater number of avirulent V parahaemolyticus
thus poses an even greater problem. To date, there appears to be no widely accepted
method for the enumeration of virulent isolates with increasing uncertainties of using the total V parahaemolyticus count to predict the risk of infection (Zimmennan et al., 2007).
These methods of V parahaemolyticus detection usually require several days using conventional techniques (De Sousa et al., 2004). The approach is based on presumptive incubation in a non-selective broth followed by confinnatory growth on selective agar and biochemical and nutritional tests. This process together with a lack of distinct phenotype to ascertain pathogenic isolates and the inherent nature of a self-limiting gastroenteritis infection means that most cases of infection either remain unreported or the pathogen is not detected. It is usually only in severe cases requiring hospitalisation that tests for pathogenic
V parahaemolyticus can be carried out, hence, adequate surveillance methods have long been overdue for this organism.
In recent years a number of molecular methods have been developed to aid the process of detection, identification and enumeration of potentially pathogenic V parahaemolyticus using polymerase chain reaction (PCR) based approaches (Kim et al., 1999; Miwa et al.,
2003; Panicker et al., 2004). These include the use of species specific and recognised virulence gene markers, and in many cases multiple gene targets are detected in multiplex
PCR reactions (Bej et al., 1999). However, due to a lack of understanding of what constitutes a true pathogenic strain of V parahaemolyticus, there are no satisfactory detection methods to identify pathogenic isolates. Traditional typing systems such as biochemical and nutritional profiling and 0 and K antigenic serotyping, although useful in identifying certain pandemic clones, have failed to provide a reliable means for discriminating strains that are pathogenic, due to a remarkable degree of phenotypic and serological uniformity (Bhuiyan et al., 2002; Binta & Nyaga, 1982; lguchi et al., 1995;
Martinez-Urtaza et al., 2006; Rahman et al., 2006). Thus, an increasing number of single and multiple molecular characterisation systems are currently being developed and assessed, to enable reliable, reproducible and robust analysis of genetic diversity amongst virulent and avirulent strains. These include pulsed field gel electrophoresis (PFGE) 2 methods (Wong et al., 2007), various PCR methods (Ward & Bej, 2006) and the increasing use of genomic sequencing data for applications such as mapping intergenic spacer regions
(Gonzalez-Escalona et al., 2006a) and multilocus sequence typing (MLST) Gonzalez-
Escalona et al., 2008)
The use of a variety of molecular subtyping systems almost always generates divergent phylogenetic clusters as each typing system works by utilising different loci or traits.
Therefore the dependence on a few properties or just one typing system may lead to errors that may not be apparent whilst elucidating relatedness amongst isolates (Maslow &
Mulligan, 1996). For example, clonally related pathogenic strains of V parahaemolyticus may show identical PFGE profiles with respect to the restriction fragment produced but fall into two distinct serovars (Chowdhury et a/ 2000a). Hence another enzyme or second method must be employed to validate the degree of relatedness. Generally, at least two reproducible and reliable typing systems should be used, together with further characterisation of specific markers of variation in order to confirm epidemiological relatedness between isolates.
All bacterial molecular typing systems are based upon the principle that epidemiologically related isolates are derivatives of a single parent isolate (Struelens, 1998). Therefore the pathogenic isolate shares characteristics that may differ from those of non-pathogenic isolates within the same species. Thus typing systems are able to discriminate between epidemiologically related and non-related isolates, although typing methods are more frequently used to characterise organisms below the species level (Kostman, 1995).
1.2 Aims and Objectives
Despite the large progress in identification and detection of V parahaemolylicus, only two major virulence genes encoding for thermostable haemolysins have been used in detection 3 protocols having been found in 94% - 97% of clinical cases (Suzuki et al., 1997). These genes are rarely detected from environmental and food samples so the significance of finding environmental reservoirs of haemolysin genes remains unclear.
Although various serovars of V. parahaemolyticus can cause infections, to date there is no method of biochemically, serotypically or pbenotypically differentiating clinical (virulent) and environmental (mostly avirulent) isolates of this organism. Current molecular typing systems relate pathogenic isolates or outbreaks of disease to epidemiological purposes and clonal relatedness yet there are no standardised, robust methods in use and as such considerable variability in data and its interpretation seriously hamper characterising virulent isolates. This, together with inadequate means to differentiate clinical and environmental isolates further exacerbates the problem. The overall aim of the work presented in this thesis was to assess the use of molecular characterisation to differentiate clinical and environmental isolates of V. parahaemolylicus.
The specific objectives were:
i) to assess existing biochemical and molecular typing systems in the specific
characterisation and differentiation of V. parahaemolyticus using phenotypic
and genotypic characteristics including growth, haemolysis and subtyping
systems of PFGE and 16S-23S intergenic spacer analysis (chapters 4, 5 & 6)
ii) to develop previously described typing systems for use with V.
parahaemolylicus including ITSI, tRNA typing and RAPD, (chapters 6, 7 and
8)
iii) to use the recently identified type Ill secretion system (TTSS) (Makino et al.,
2003; Park et al., 2004b; Ono et al., 2006) to assess differentiation between
clinical and environmental isolates of V. parahaemolyticus (chapter 9) 4 finally; ,to ~.:mnpm:e ~ncj eyall!lltl! .alii methods described to differentiate clinlcali . ·- --- . -
•and I environmental, isolates o{thispathogen (chapters'{() tdtl ~:
A better :understanding o(the differences .and !'elatedness tbetWeei] .path:g~erli_c and tnon
p~thogt:riic V parahaemolyticus enables the de:velopmeni of more 'accurate targeting,
detection:and monltoring;of this emerging global~ pathogen.
I . Chapter Two LITERATURE REVIEW
2.1 Tbe Organism
2.1.1 General characteristics
V. parahaemolyticus is a Gram-negative, motile, rod-shaped bacterium possessing both polar and lateral flagella that enable it to swim and swarm respectively (Atsumi et al.,
1992; Kim & McCarter, 2000; McCarter, 200 I). The organism is generally oxidase positive, facultatively aerobic and non-spore forming. It is a member of the Vibrionaceae of the y-Proteobacteria family, originally termed Pasture/la parahaemolytica (Fujino,
1966). In common with other vibrios, V parahaemolyticus has a strong requirement for sodium ions for growth (Pace, 1989) and is abundant in the marine and estuarine environments (Baross & Liston, 1970; Larsen et al., 1981; Binta & Nyaga, 1982). Together with V cholerae and V. vulnificus, V. parahaemolyticus form a group of organisms that are responsible for most bacterial seafood and water-borne extra-intestinal infections (Johnson et al., 1971; Thomton et al., 2002). Recently, a number of advances in molecular methods have facilitated the study of the dynamics and pathogenicity of these organisms.
2.1.2 Genome organisation of V. paralraemolyticus
Pulsed-field gel electrophoresis (PFGE, section 2.7.3) of intact genomic DNA shows V. parahaemolyticus to possess two chromosomes (replicons) of approximately 3.2 and 1.9
Mb (Y amaichi et al., 1999) and that the two-chromosome configuration is common amongst members of the Vibrionaceae (Tagomori et al., 2002; Okada et al., 2005). The larger replicon ranged from 3.0 to 3.3 Mb in size whereas the smaller replicon varies considerably (0.8 to 2.4 Mb) amongst 34 species of vibrios, suggesting that this smaller chromosome has been more flexible in the evolution of V parahaemolyticus (Okada et al.,
2005). Genome sequencing of a clinical strain, RIMD2210633, confirmed V
6 parahaemolyticus to contain two circular chromosomes and to possess 11 copies of ribosomal operons which correlate well with its rapid growth rates. Complete genome sequencing also revealed, for the first time, that V parahaemolyticus harboured genes for type Ill secretion systems clustered together with pathogenicity islands and virulence effectors (Makino et al., 2003). These are summarised in Table 2.1.
Table 2.1: Overview of V. parahaemolyticus genome organisation in a clinical strain RIMD2210633 (Adapted from Makino et al., 2003)
Ch.l Ch.2 Length or sequence 3,288,558 bp 1,877,212 bp G+C ratio 45.4% 45.4% Open reading frames 3,080 1,752 (ORFs) Protein coding regions 86.9% 86.9% Average ORFs length 926.9 bp 931.3 bp rRNA (16S-23S-SS) 10 I tRNA 112 14 Secretion systems 0 2 Genomic Islands 5 2
2.1.3 Ecological nicbes
Coastal filter-feeding invertebrates such as oysters concentrate vibrios. However, the relationships between anthropogenic effects on coastal ecosystems (i.e. fish or shellfish farming), the natural hosts (i.e., filter feeders) and the environmental niches of vibrios are complex, dynamic and not very well understood. V parahaemolyticus has been shown to possess multiple cell types that may explain its habitat in a variety of niches (McCarter,
1999). A planktonic free-swimming state, a sessile existence within a microbial community such as the commensal relationship in shellfish, or attached to under sea surfaces (boats, seaweeds, fishes etc) and attachment to host organism in pathogenesis are just some of the adaptations of this organism (Kaneko & Col well, 1973, 1975). Extensive in-vitro studies have revealed two cell types; a "swimmer cell" and a "swarrner cell."
(McCarter, 1999). The swimmer cell consists of a single polar flagellum, powered by the
7 sodium motive force (Atsumi et al., 1992) ideally suited for the survival and proliferation
in the marine environment at a pH of approximately 8.0 (Kogure, 1998). The swarrner cell
type is believed to be more adapted for growth and colonisation of V parahaemolyticus on
surfaces or viscous environments (McCarter, 1999). Electron micrographs show a differentiated cell type with numerous lateral peritrichously arranged flagella (Fig. 2. I).
Figure 2.1: Electron micrograph of a swimmer cell (A) and swarmer cell (B) types of V. parallaemolyticus. Bar indicates approximately I mm. (Reproduced with permission; McCarter, 1999)
Densities of V parahaemolyticus in environmental and seafood samples vary considerably depending on water temperatme, salinity, location, sample types, faecal pollution and the
method used for analysis (Kelly & Stroh, 1988; DePaola et al., 1988). DePaola and
colleagues showed a significant log10 correlation between seasonal abundance of V parahaemo/yticus, salinities and temperatme in Alabama oysters (DePaola et al., 2003a).
Total V parahaemolyticus abundance was highest in May and June with sea temperatmes
over 25°C and salinities close to the optimum of 17 parts per thousand (ppt). It was
estimated that an increase of one degree Celsius resulted in a 1.2 fold increase in the
number of V parahaemolyticus colony forming units (cfu) per gram of oyster homogenate.
Similar findings in seasonal abundance have also been reported in several other studies
(Kelly & Stroh, 1988; DePaola et al., 1988, 2000).
8 Despite a trophism for warmer temperatures and tropical climates, there are an increasing number of reports of pathogenic V parahaemolyticus in colder climates such as Alaska and Southern Chile, that up until now were considered regions too cold to support the growth of this organism (McLaughlin et al., 2005; Gonzalez-Escalona et al., 2005a;
Fuenzalida et al., 2006) These recent findings suggest remarkable changes in the organism's ability to quickly adapt, survive and proliferate in different niches. The prevalence of these pathogenic V parahaemolyticus in colder climates may also be indications of global warming and thus an increasing cause for worldwide concern.
2.1.4 Viable but non-culturable state of V. paralraemolyticus
As with many human pathogens, V parahaemolyticus is also thought to exist in the viable but non-culturable (VBNC) state and this may explain the low detection rates of V. parahaemolyticus in cold temperatures. For example, Baffone et al. (2006) recently identified unculturable V parahaemolyticus in plankton fractions by PCR. Several laboratory studies have induced the VBNC state in V parahaemolyticus by exposing cells to low temperatures under starvation conditions (Jiang & Chai, 1996; Mizunoe et al., 2000;
Wong & Wang, 2004). Temperature upshifts have been reported to "resuscitate" a sub population of cells. In a recent study, VBNC V parahaemolyticus that were subjected to temperature upshifts showed a remarkably heterogeneous population of cells that were coccoid or flatted in shape and in some cases showed asymmetrical cell divisions (Coutard et al., 2007). These findings are also similar to VBNC in other Vibro spp. (Vattakavan et al., 2006; Pruzzo et al., 2003). Remarkably, despite morphological and structural differences, the VBNC cells have been shown to retain their virulent properties (Bates &
Oliver, 2004; Baffone et al., 2003) thus the possible existence of environmental VBNC V parahaemolyticus pose a significant public health concern and will hamper epidemiological investigations of this pathogen.
9 2.2 Worldwide distribution of V. paralraemolyticus
Food borne poisoning due to V. parahaemolyticus has been identified in many countries worldwide (Barker, 1974; Cabassi & Mori, 1976). Incidents are more common in countries when seawater temperatures rise over 15°C during the summer season and where seafood consumption is high (A lam et al., 2003; DePaola et al., 2003a; Lozano-Leon et al., 2003).
2.2.1 Europe
In Europe, V. parahaemolyticus has been isolated in several countries including the UK
(Hooper et al., 1974; Ayres & Barrow, 1978), Demark (Kristensen, 1974), Norway (Gjerde
& Boe, 1981; Bauer et al., 2006) and Eastern European countries of Romania and Bulgaria
(Aldova, 1989). However to date, most reports of the isolation of pathogenic V. parahaemolyticus are in southern Europe and the Mediterranean including France (Hervio
Heath et al., 2002; Robert-Pillot et al., 2004; Quilici et al., 2005), Spain (Lozano-Leon et al., 2003; Marne & Aznar, 1988; Martinez-Urtaza et al., 2004, 2005; Molero, et al., 1989),
Italy and Greece (Masini, et al., 2007).
In 1999 a large outbreak of V. parahaemolyticus related food poisoning occurred in
Galicia, northwest Spain where a total of 64 people were admitted to hospital with acute diarrhoea after consumption of raw oysters (Lozano-Leon et al., 2003). Temporary surveillance of Vibrio spp. employed at the time ensured the accurate identification of the etiological agent (Martinez-Urtaza et al., 2004). Another outbreak involved 80 cases of illnesses associated with consumption of boiled crab (Martinez-Urtaza et al., 2005). These sporadic cases in Spain may highlight possible deficiencies in the identification of V. parahaemolyticus by national reference laboratories and overall lack of surveillance and awareness of this pathogen.
10 2.2.2 East Asia
In Asia, where illness due to V parahaemolyticus food poisoning is most prevalent, both pathogenic and non-pathogenic isolates have been characterised and reported in most countries including Korea, (Chun et al., 1974) Japan (Arakawa et al., 1999), Taiwan
(Wong et al., 2005b), China (Yano et al., 2006) and Indonesia, (Marlina et al., 2007). For example, in Japan, where it was first recognised as a major cause of seafood related illness in 1950, an outbreak of 272 illnesses resulting in 20 deaths were associated with the consumption of raw sardines (Daniels et al., 2000). Since then, V parahaemolyticus has been estimated to account for over 30% of all annual reported cases of food poisoning in
Japan (Aiam et al., 2002) and the leading cause of food poisoning (171 0 incidents, 24,373 cases) between 1996 and 1998 as reported by the Infectious Disease Surveillance Centre
(IDSC) of Japan (Su & Liu, 2007).
2.2.3 Southern Asia and Africa
Pathogenic isolates in the Indian subcontinent have been reported in an increasing number of environmental samples in Bangladesh (Islam et al., 2004), Sri Lanka (Palasuntheram et al., 1977), and India where, together with V cholerae infection, fatalities have been the highest in the malnourished and immuno-compromised populations (De, 1977; Deepanjali et al., 2005; Parvathi, et al., 2006; Sen et al., 2007) (section 2.3). Gastroenteritis related to
V parahaemolyticus has also been reported in Africa, (Bockmuhl & Triemer, 1975) as well as an incident during the V cholerae El Tor cholera epidemic in West Africa
(Bockemuhl et al., 1975). A recent report also highlighted the presence of the more virulent pandemic strain (section 2.4.4) of V. parahaemolyticus in Mozambique.
(Ansaruzzarnan et al., 2005) (Fig. 2.2).
11 2.2.4 North America
Sporadic outbreaks of V parahaemolyticus food poisioning have been reported in the USA
since 1972 (Molenda et al., 1972). Between 1973 and 1998, 40 outbreaks of V parahaemolyticus food poisoning were reported to the Centers for Disease Control (CDC)
in the USA (Daniels et al., 2000). Among them, four major outbreaks involved over 700
cases of food poisoning due to raw oyster consumption, most prevalent during the warmer
months of each year (Daniels et al., 2000; DePaola et al., 2000). Perhaps more significant
for the North European community are the two recent reports of virulent V parahaemolyticus, in Washington, British Columbia and Alaska (McLaughlin et a!, 2005).
Despite the colder Alaskan and Canadian sea water temperatures, V parahaemolyticus
related food poisoning resulted in 14 and 177 cases of infection, in 2004 and 2006
respectively (Khaira & Galanis, 2007; McLaughlin et a!, 2005).
2.2.5 South America
Environmental reservoirs of V parahaemolyticus were identified in South America as
early as 1977 (Casellas et al., 1977; Franca et al., 1980). Several studies have indicated the
virulence potential of V parahaemolyticus (Magalhaes et al., 1992) with an outbreak of V
parahaemolyticus food poisoning amongst military personnel in Peru (Begue et al., 1995).
More recently, Cordova and colleagues reported two outbreaks, in 1998 and 2004 in Chile
(Cordova et al., 2002; Gonzalez-Escalona, et al., 2005a). Between November 1997 and
March 1998, over 300 cases were reported in northern Chile with another outbreak
affecting 1500 people between January and March 2004 in Puerto Montt, a region of
usually cold coastal water (Gonza1ez-Escalona et al., 2005a). A further outbreak between
January and April 2005 in the same region resulted to 3,725 cases of acute diarrhoea. As
Puerto Montt is one of the main shellfish producing areas of in Chile, contaminated
shellfish had rapidly spread to other urban regions. By the end of March 2005, a total of
I 0, 783 cases were reported, making it the largest documented occurrence of V 13 parahaemolyticus food poisoning in the world (Cabello et al., 2007). The occurrence of these large outbreaks indicates that the contamination of V. parahaemolyticus in oysters is a great safety concern to the UK shellfish industry and public health.
2.3 Pandemic serotypes of V. parallaemolyticus
Antigenic serotypes based on the somatic (0) and capsular (K) antigen of V. parahaemolyticus have been described since 1969 (Miwatani et al., 1969; Zen-Yoji et al.,
1970). To date, antiserum for 13 0 and 71 K antigens have been reported and are commercially available to type V parahaemolyticus isolates. Diverse serotypes have caused outbreaks worldwide. However, since 1996 an increasing number of cases are related to the 03:K6 serotype (Nair et al., 2007). This serovar was first identified in
Calcutta after a sudden increase in incidences during a hospital-based active surveillance program (Okuda et al., 1997b ). Between September 1996 and April 1997 this serovar accounted for 63% of all reported cases and since then this highly virulent serovar has dominated in a large number of outbreaks worldwide (Chiou et al., 2000; Chowdhury et al., 2000b). In the USA the largest outbreak of V parahaemolyticus food poisoning, with
416 cases, was attributed to the 03:K6 serovar with the same serovar later involved in another outbreak in 1999 (Daniels et a!, 2000; Su & Liu, 2007). This serovar was also responsible for the outbreaks in Chile Gonzalez-Escalona et al., 2005a), and Spain
(Martinez-Urtaza et al., 2005).
Since the year 2000, a number of reports have identified several other serovars prevalent in outbreaks. Amongst them, serovars of 01 :KUT, 01 :K25, and 04:K68, have been responsible for a large number of food poisoning cases worldwide (Chiou et al., 2000;
Chowdhury et al., 2004b; Ansaruzzarnan et al., 2005; Rizvi et al., 2006; Serichantalergs, et al., 2007). All three serotypes have been shown to possess remarkable homology in
molecular typing profiles to the 03:K6 serovar, thus it is believed that all three serovars of 14 01 :KUT, 01 :K25, and 04:K68 are derived from the parent 03:K6, despite possessing different 0 and K antigens (Chowdhury et al., 2000a; Chowdhury et al., 2004a;
Chowdhury et al., 2004b; Nair et al., 2007). Clearly, these changes in the 0 and K antigens demonstrate the dynamic nature of the evolution of virulent V parahaemolyticus by the acquisition of 0 and K antigens.
2.4 Virulence factors and pathogenicity
2.4.1 Human infection
Clinical manifestations of V parahaemolyticus infection follow an incubation period of
between 12 - 46 hours and symptoms that include a self-limiting diarrhoea, abdominal cramps, nausea, vomiting, headache, fever and chills may last up to seven days (Nolan et al., 1984). Extra-intestinal infections have also been known to occur including wound, eye and ear infections (Tacket et al., 1982; Johnson, et al., 1984; Sautter et al., 1988; Kelly &
Stroh, 1988). Between 1988 and 1997, 345 sporadic V parahaemolyticus infections were
reported to the CDC of which 45% were hospitalised (Daniels et a/, 2000). Of these, 59%
were gastroenteritis, 34% were wound infections, 5% were septicaemia, and 2% were from other exposures. In most cases of gastroenteritis, rehydration therapy is sufficient and
antibiotic treatment is not required. For more serious infections tetracycline has long been
the drug of choice, although several other antibiotics have commonly been used (Morris &
Black, 1985). However, as in most cases of food poisoning, and due to the self-limiting
symptoms of infection, mild cases of infections are not reported and consequently, actual
figures of infection remain largely unknown.
Most strains isolated from infected patients demonstrate a major virulence factor known as
thermostable direct haemolysin (TDH) that has been shown to exhibit ~-haemolysis on
Wagatsuma agar (Sakurai et al., 1973). This characteristic phenotype is also known as the
Kanagawa phenomenon (section 2.4.2). Another virulence factor, TDH related haemolysin 15 (TRH), is typically associated with Kanagawa negative or urease positive isolates of V parahaemolyticus (Eko, 1992; Kelly & Stroh, 1989; Suthienkul et al., 1995) and both TDH
and TRH have been isolated from the same organism (Xu et al., 1994). However, most
strains of V. parahaemolyticus isolated from the environment express neither TDH nor
TRH even if they harbour the tdh or trh gene sequences respectively (Nishibuchi et al.,
1989; Nishibuchi & Kaper, 1995). Furthermore, the infectious dose of pathogenic isolates
may be high or low, with the more pathogenic strains of V parahaemolyticus having been
described as having multiple copies and varying expression of TDH and TRH (Honda et
al., 1989a; Nishibuchi & Kaper, 1990; Nagayama et al., 1995).
2.4.2 Thermostable direct haemolysin (TDH)
The gene for TDH is present in 93-96 % of clinical isolates and can be found in two or
three copies either on the smaller chromosome flanking the pathogenicity island or on
plasmids (Makino et al., 2003; Nishibuchi & Kaper, 1990; Baba et al., 1991a). The cloned
tdh gene has a GC content of 35.6% and produces a 21.1 k.Da homodimer protein product
consisting of 189 amino acid residues including signal peptides (Taniguchi et al., 1985).
Although the mechanism of TDH action has not been fully elucidated, workers have
described TDH as a pore forming toxin (Honda et al., 1992; Takahashi et al., 2006)
forming a cation channel (Lang et al., 2004) and its most extensively studied effect is the
ability to induce haemolysis of erythrocytes from different mammalian species (Huntley &
Hall, 1994, 1996; Douet et al., 1996). Beta-haemolysis on specialised blood agar,
Wagatsuma agar is the most common phenotype used to ascertain TDH positive isolates of
V parahaemolyticus. However it has been shown that expression of both tdh genes are
required for an adequate ~-haemolysis phenotype (Nishibuchi & Kaper, 1990). The
workers showed that all cloned TDH genes exhibited at least 96.7% nucleotide sequence
homology but tdh specific RNA transcripts were missing in some of the clones. Therefore, 16 differences at the transcriptional control level are likely to be responsible for the variation in haemolytic phenotypes. Clearly, the variability in haemolysis amongst different isolates needs to be addressed if the haemolytic activity is to be used as a virulence marker of V parahaemolyticus (Chapter 4). Variants ofTDH have also been isolated from V cholerae non-01, V mimicus and V hollisiae (Nishibuchi et al., 1990; Baba et al., 1991a; Yoh et al., 1989; Terai, et al., 1990). The nucleotide sequences of various tdh genes are well conserved and their protein products immunologically indistinguishable (Nishibuchi &
Kaper, 1995).
TDH has also been shown to induce cytotoxic effects consisting mainly of cell degeneration, which often leads to loss of viability (Sakurai et al., 1976). Using human and rat monolayers Raimondi et al demonstrated a TDH dose dependent increase in
2 intracellular free Ca + and er secretion leading to cell death (Raimondi et al., 2000). This
work together with previous work reported by Tang et a/, showed that cytotoxity due to
2 TDH is mediated by a Ca + independent pathway (Tang et al., 1995). The haemolysin TDH
has also been described as a cardiotoxin found to stop the spontaneous beating of cultured rat and mouse myocardial cells (Goshima et al., 1976) and intravenous injection of TDH
into rats resulted in rapid death through cardiac arrest (Honda et al., 1976).
2.4.3 Thermostable direct haemolysin related haemolysin (TRH)
Thermostable direct haemolysin related haemolysin (TRH) has typically been isolated
from clinical cases of V parahaemolyticus food poisoning in almost half of Kanagawa
negative isolates (Yoh et al., 1992). More recently it has also been isolated from V
alginolyticus (Xie et al., 2005; Gonzalez-Escalona et al., 2006b). Other than a lack of~-
haemolysis on Wagatsuma agar, TRH is similar in attributes to TDH and the proteins share
up to 68% nucleotide sequence and 62% amino acid similarity (Xu et al., 1994; Nishibuchi
et al., 1989). Up to two copies of variants trh 1 and trh2 respectively have been reported in 17 a single genome (Kishishita et al., 1992) and in some cases is correlated with the presence of the urease gene cluster. Consequently, urea production may be used as a possible marker for TRH and potential virulence in V parahaemolyticus (Eko, 1992; Kaysner et al., 1994;
Suthienkul et al., 1995; Chiu et a/, 2007). lida et a/ have shown genetic linkage between the urease gene cluster, the genes ure and trh by identifying both ure and trh on the same
Notl digested PFGE fragment (Iida et al., 1997; 1998). They were able to show that both trh and ure were within 8.5 kb of each other and that the trh and tdh genes were localized within a 40 kb region. A low and aberrant CG content relative to the genome in this region, and presence of phage-like integrases subsequently led to characterisation of several genomic islands (Hurley et al., 2006).
2.4.4 Other haemolysins in V. parallaemolyticus
Taniguchi et a/ described another thermostable haemolysin (delta-VPH) that was found in all clinical and environmental V parahaemolyticus isolates (Taniguchi et al., 1990). This haemolysin had no homology to that of TDH or TRH and was also found present in V damse/a but its significance in human disease is not known. Similarly, Sakurai and colleagues demonstrated the presence of an additional haemolysin in V parahaemolyticus that unlike TDH or TRH, was deactivated upon heating at 60 for 10 mins (Sakurai et al.,
1974). The mature thermolabile haemolysin (TLH) was found to be of 45.5 kDa and its gene had a GC content similar to that of the genome (Taniguchi et al., 1986). Whilst TDH was found only in Kanagawa positive isolates, TLH was found in all strains tested (Sakurai et al., 1974). Since then, the gene for TLH has been used as a species specific marker for
V parahaemolyticus (section 2.6.3).
2.4.5 Cytotoxicity and enterotoxicity of V. paralraemolyticus
Both TDH and TRH, together with several uncharacterised factors have been shown to be responsible for the cytotoxic and enteropathogenic properties of V parahaemolyticus. In 18 some early studies, Carruthers (1975) demonstrated cytotoxicity to HeLa cells and scanning electron microscopy of cells derived from human amniotic membrane revealed a decrease in microvilli and damage to cell morphology with sub lethal amounts of TDH.
Higher dosage caused complete structural damage to the cytoplasm (Sakurai et al., 1976).
Nairn et a/ used osmotic stabilizers to show that the mode of action by TDH on erythrocytes was different from that on cultured embryonic fibroblast cells (Nairn et al.,
2001).
2.4.6 Other potential pathogenicity determinants
In one of the largest cl inica1 studies to date, Suzuki and colleagues demonstrated that 3. 7% of V. parahaemolyticus isolated from 1319 cases of traveller's diarrhoea lacked both tdh and trh genes (Suzuki et al., 1997). Indeed, when tdh deletion mutants lacking both copies of tdh were created by homologus recombination, the mutants lost their haemolytic activity yet still demonstrated toxicity to HeLa cells as well as partial fluid accumulation in rabbit intestine (Park et al., 2004a). Therefore, it is likely that additional factors are involved in the pathogenicity of V parahaemolyticus.
2.4. 7 Serine protease
Work carried out by Lee et al. (2002) demonstrated a serine protease from a clinical strain lacking both tdh and trh genes. The protease showed significant effects on the growth of
Chinese hamster ovary (CHO), HeLa, Vero and Caco-2 cells. The protease lysed erythrocytes, caused tissue haemorrhage and death in m1ce when injected both intraperitoneally and intravenously, but it was heat labile unlike TDI·I. It was shown that the serine protease was widely distributed in V. parahaemolyticus and V alginolyticus
(Ishihara et al., 2002).
19 2.4.8 Adherence and colonisation factors
An early report in 1977 demonstrated an increase in adherence by Kanagawa positive
isolates to human foetal intestinal cells, whereas Kanagawa negative isolates adhered at a
much slower rate or not at all on cultured HeLa cells (Carruthers, 1977). The work suggested a carbohydrate molecule on the surface of the Kanagawa positive bacterial cell
wall was responsible for this interaction. However, no specific receptor molecules have yet
been discovered although it has been reported that V parahaemolyticus shows an increase
in adherence under starvation conditions (Wong et al., 2005a). Nakasone et al.
demonstrated that pili play a key role in the adhesion and colonisation in the rabbit
intestine by isolating pili that were adhesive to the intestinal epithelium (Nakasone et al.,
1990, 1992). These pili were found to be indistinguishable in morphology to pili of a
pandemic 03:K6 strain, yet were antigenically distinct (Nakasone et al., 2000). Further
work is required to establish mechanisms of V parahaemolyticus adherence.
2.4.9 Siderophores
Several workers have described the potential role of siderophores in V parahaemolyticus
pathogenicity (Yamamoto et al., 1992b, 1994, 1995; Tanabe et al., 2003, 2006; Funahashi
et al., 2003). In conditions of iron limitation, Dai et al. (1992) showed an increase in
haemolytic activity of Kanagawa positive strains, together with the production of a
siderophore, an increase in adherence, and lethality of V parahaemolyticus to mice cells.
Similar findings were obtained with Kanagawa negative isolates suggesting that virulence
factors may be induced or enhanced under iron-limited conditions. The siderophore, named
vibrioferrin (Y amamoto et al., 1994) was found to sequester iron from 30% iron saturated
human transferrin for growth, suggesting that V parahaemolyticus may be able to utilise
host iron through the action of vibrioferrin for survival and proliferation (Y amamoto et al.,
1994). Under iron limiting growth conditions, clinical isolates showed higher production of
vibrioferrin than strains isolated from seafood thus suggesting a selective advantage of the 20 survival of clinical isolates in the human intestine (Y amamoto et al., 1999). Although vibrioferrin has been well characterised (Tanabe et al., 2003, 2006), and the receptor genes found widely distributed amongst V. parahaemolyticus (Funahashi et al., 2003) in both clinical and environmental isolates, its significance in the role of pathogenicity remains unclear.
2.4.10 Insertion Sequences
The possible role of insertion sequences in the spread of tdh genes was described by Terai et al. ( 1991) upon the characterisation of tdh in related species, V. cholerae non-0 I, V. mimicus and V. hollisae (Nishibuchi et al., 1990). Analysis of 7 tdh genes and their flanking regions revealed insertion sequence like elements, termed insertion sequences of
Vibrio species (ISVs) or sequences that were derived from the genetic rearrangement of the
ISVs. The possible role of ISVs in transferring V. parahaemolyticus pathogenicity determinants have also been described by several others including (Burstyn et al., 1980;
Baba et al., 1991 a; Park et al., 2000) and Okura et al., (2005) who identified a large 16-kbp insertion sequence which was present in almost all pandemic strains but not in non pandemic strains tested.
2.4.11 Phage associated virulence
A large number ofbacteriophages (vibriophages) with either a specific host or a broad host range have been described (lnoue et al., 1995; Hardies et al., 2003; Comeau et al., 2006),
including those found in the pandemic 03:K6 clone (Nakasone et al., 1999; Chang et al.,
1998). Indeed, Chang et al. (2000) showed that filamentous phage genomes of the
previously described phages Vfl2 and Vf33, integrated into genomic DNA of V. parahaemolyticus at distinctive regions and were also distributed on some plasmids and
total cellular DNAs of one V. damsela and one non-agglutinable vibrio strain tested
(Taniguchi et al., 1984; Chang et al., 1998). However, the potential of these phages as 21 tools for genetic transmission of virulence factors in V parahaemolyticus has yet to be determined.
Within the pandemic 03:K6 serovars, the open reading frame ORF8, from filamentous phage f237, has been extensively characterised and has been described as a marker for pandemic strains (Nasu et al., 2000). The phage genome was originally found in a common plasmid p03K6, in all 24 pandemic 03:K6 strains tested and genetic organisation resembled that of a previously characterised phage Vf33 (Taniguchi et al., 1984). The double stranded DNA obtained from the treatment of f237 with DNA polymerase was identical to that of p03K6 suggesting that p03K6 is a replicative form of f237. The sequence of ORF8 (orj8) has since been used to identify pandemic strains by several workers (Myers et al., 2003; Laohaprertthisan et al., 2003) although there have been reports by others that ORF8 has poor correlation with the pandemic genotype (Bhuiyan et al., 2002; Okura et al., 2003). To date, there appears to be no evidence of lysogenic phage conversion, like that of the cholera toxin in V. parahaemolyticus (Waldor & Mekalanos,
1996)
2.4.12 Type Ill secretion systems
Type III secretion enables Gram-negative bacteria to secrete and inject pathogenicity proteins into the cytosol of host cells (Hueck, 1998). The presence of genes for a type Ill secretion system (TTSS) was first described upon genomic sequencing of V parahaemolyticus (Makino et al., 2003) and since then TTSS genes have also been characterised in V. cholerae non-01 and non-0139 (Dziejman et al., 2005). The TTSS is the principle virulence factor of diarrhoea causing bacteria such as Shigella, Salmonella, and enteropathogenic Escherichia coli, which cause gastroenteritis by invading or interacting with intestinal epithelial cells (Hueck, 1998). The presence of TTSS homologues was also demonstrated in other vibrios, such as V. alginolyticus, V. harveyi, 22 and V. tubiashii, suggesting that the TTSS may have a role in protein secretion in these organisms during interaction with eukaryotic cells (Park et al., 2004b)
Park and colleagues characterised two sets of genes termed TTSS I and TTSS2. By
constructing mutant strains of TTSS 1 and TTSS2 they were able to show reduced
cytotoxic and enterotoxic activities compared with the parent strains (Park et al., 2004b).
Molecular characterisation of the TTSS I of V. parahaemolyticus most closely resembled
that of Yersinia spp., with over 30 gene homologues and a large inversion of the yscA to yscQ region together with an additional 12 hypothetical genes within the ysc region that are
not found in Yersinina spp. or other species. Work carried out by Ono et al. (2006) later
identified some of these genes to encode effector proteins found translocated into HeLa
cells and that the cytotoxicity to HeLa cells was related to one of the newly identified
secreted proteins encoded by VP1680 (Chapter 9). The second secretion system, TTSS2
was found on the smaller chromosome (section 2.1.2) in an area characterised as a
pathogenicity island. In this region only 9 genes had homologues to Yersinina spp. and
resembled the TTSS found recently on V cholerae (Dziejman et al., 2005)
Kodama et al. (2007) showed that TTSS2 induced a cytotoxic effect on human epithelial
intestinal cell lines (HCT-8 and Caco-2) partly due to a putative ADP-ribosyltransferase. A
further two TTSS2-dependent secretion proteins necessary for TTSS2-dependent
cytotoxicity and enterotoxicity were also characterised (Kodama et al., 2008). However,
despite the functional characterisation of TTSS2, the distribution of TTSS systems remains
unclear. According to Park et a/, (2004b) and Ono et a/, (2006) the TTSS2 was only found
in Kanagawa positive clinical strains tested thus suggesting that TTSS2 had a major role in
the pathogenicity of V parahaemolyticus (Chapter 9).
23 2.5 Conventional methods for identification
2.5.1 Identification and detection methods of V. parahaemolyticus
V. parahaemolyticus is one of the fastest growing organisms with doubling times as short
as 9 min in optimum growth conditions (Katoh, 1965). These include a pH of 7.8 to 8.6 and a temperature of 37°C, although it has been found to grow in a wide range of pH and
temperatures (Yeung & Boor, 2004). This, together with a toleration of moderately high
alkaline conditions up to pH 9, forms the basis of a variety of selective media used to grow
and enrich food samples and in use since 1971 (Twedt & Novelli, 1971 ).
2.5.2 Isolation
Isolation of V. parahaemolyticus from environmental and food sample usually requires
homogenisation of sample with enrichment broths such as alkaline peptone water (APW)
or glucose salt teepol broth (GSTB). Salt polymyxin broth (SPB) and salt colistin broth
(SCB) containing the appropriate antibiotics for isolation of V parahaemolyticus from
other marine vibrios have also been used. Beuchat, (1977) identified other selective
enrichment broths including 7%NaCI tryptic soy broth (TSBS), arabinose-ethyl violet
broth (HAEB) and water-blue-alizarin yellow broth (WBA Y) for the enrichment and
recovery of cold- and heat-stressed V parahaemolyticus. The use of a pre-enrichment step
before selective enrichment has also been shown to increase V. parahaemolyticus counts
(Ray et al., 1978; Ho fer & Silva, 1984; Hara-Kudo et al., 200 I).
2.5.3 Culture and growth
The most probable number method (MPN) described in the US Food and Drug
Administration Bacterial Analytical Manual (Elliot et al., 1995) is the most commonly
used method for the detection of V. parahaemolyticus. The MPN method is both labour
intensive and time-consuming (Su & Liu, 2007). Since 1963, thiosulphate citrate bile salt
sucrose agar (TCBS) has been used widely as a selective medium for many Vibrio spp. 24 including V. parahaemolyticus. This medium uses a combination of ox bile, NaCl, a high
pH and an indicator system for acid production from sucrose. However, TCBS cannot
differentiate V parahaemolyticus from other Vibrio species that may be sucrose negative
such as V mimicus or V vulnificus. Several workers have evaluated TCBS supplied by
different manufacturers to reveal inconsistencies in the recovery and quality of growth of each species (West et al., 1982) suggesting the need for better quality controls. Due to the
common presence of more than one Vibrio species in environmental and food samples, it
may be more favourable to use TCBS in clinical cases than culturing from environmental
samples.
V parahaemolyticus has also been isolated on a variety of other media including mannitol
salt agar (Carruthers, 1976), sucrose teepol tellurite agar (Chatterjee et al., 1977), and
thiosulphate chloride iodide agar (Pfeffer & Oliver, 2003) although none have been used
extensively. Trypticase soy agar containing triphenyl tetrazolium was developed to
enhance differentiation between V parahaemolyticus and V alginolyticus (Kourany, 1983)
but also does not appear to have been widely used. The performance of CHROMagar
Vibrio (CV), a medium containing a chromogenic beta-galactosidase system was also
studied (Hara-Kudo et al., 2001). In comparison to TCBS, a greater proportion of samples
were positive in CV and less interference from other vibrios as V parahaemolyticus
appears violet in colour whilst other vibrios appear pale blue. Unlike TCBS, the colour of
the colonies did not change with prolonged incubation or storage.
2.5.4 Biochemical profiling
The screening of presumptive V parahaemolyticus is necessary to exclude as many non-
target organisms as possible before further identification. These basic screening tests
include assessing cell growth, morphology, motility, swarming on growth surface, Gram
staining and several biochemical tests including arginine dihydrolase (ADH) and Voges- 25 Proskauer tests. Kaper et al. (1979) described a medium for screening of presumptive V. parahaemolyticus. This includes seven tests in one medium including the utilisation of mannitol, sucrose and lactose together with ADH, indole reactions, gas and H2S production. Although it was able to differentiate V parahaemolyticus from other Vibrio spp. some combinations of positive reactions were found to interfere with detection of other tests.
A full list of biochemical tests differentiating between closely related species have been described (Elliot et al., 1995). Of the many tests it has been noted that an increasing number of V parahaemolyticus are resistant to 150J.lg of di-isopropyl pteridine phosphate
(0/129). Several workers have also reported oxidase negative and urease positive strains that coincide with pathogenic V parahaemolyticus (Eko, 1992).
2.5.5 API and commercial identification systems
The use of API 20E and API 20NE (BioMerieux) has been shown to be the most reliable and quickest method to identify V parahaemolyticus (Overman et al., 1985, 1986). In the most recent study, Croci et al., (2007) evaluated API 20E, API 20NE and a set of 32 conventional tests described previously by Alsina and Blanch ( 1994) for the rapid identification of vibrios. Intra- and inter-laboratory test findings showed the Alsina and
Blanch scheme to be the most sensitive in correctly identifying V parahaemolyticus.
Veretta et al. compared the API20E and API20NE identification systems to a BBL Crystal enteric/nonfermenter (E/NF) identification system using Gram negative rods from clinical samples (V arettas et al., 1995). The workers found that the Crystal system was more convenient than API for routine clinical use because it required fewer repeats and was easier and less time consuming to use. O'Hara and eo-workers applied these and several other systems that included MicroScan Neg ID2, Rapid Neg ID3, Vitek GNI+ and ID
GNB to identify members of the Vibrionaceae family (O'Hara et al., 2003) with varying 26 levels of accuracy. Overall percentage accuracy ranged from 63.1% to 80.9% with the
MicroScan Neg ID2 and Crystal E/NF respectively.
In what appears to be the first study of its kind on V parahaemolyticus, Martinez-Urtaza et al. (2006) used an API 20E system to biochemically characterise differences between clinical and environmental isolates of V parahaemolyticus. By using 2% and 0.85% NaCI as diluent (as recommended by the manufacturer), they showed that environmental isolates gave more accurate identification using a 2% NaCI, whereas clinical isolates were correctly identified using 0.85% NaCI, consistent with previous findings (MacDonell et al.,
1982). Additionally, clinical and environmental isolates demonstrated significant differences in up to 5 tests. Both gelatine utilisation (GEL) and arabinose fermentation
(ARA) were positive in the environmental group and negative in the clinical isolates tested.
Environmental isolates analysed using 2% NaCl showed higher positive reactions for citrate utilisation, urease, melibiose and amygdalin than those using 0.85% NaCI thus suggesting small yet significant differences in metabolism amongst clinical and environmental isolates.
2.5.6 Characterisation of pathogenic V. parahaemolyticus
The difficulty in obtaining reliable identification of V parahaemolyticus by conventional biochemical procedures alone is widely recognised and summarised above. Once identified, there is no biochemical test or series of tests that would confirm a pathogenic isolate. However, the haemolytic properties of TDH have been well documented and correlated with human pathogenicity since 1969 (Miyamoto et al., 1969) and ~-haemolysis of human or rabbit erythrocytes on Wagatsurna agar is a commonly used indicator for TDH related virulence (section 2.4.2). The related haemolysin, TRH, is more active against calf and sheep red blood cells than human or rabbit cells but does not produce haemolysis on
27 Wagatsurna agar although the presence of TRH has been associated with the production of urease in some findings (section 2.4.3, Eko, 1992; Iida et al., 1997; Park et al., 2000).
2.5.7 Immunological detection of pathogenic V. parallaemolyticus
Honda et al. modified a BTB-Teepol agar and arabinose-ammonium sulphate-cholate agar
(MAAC) for the immunological detection of TDH producing isolates (Honda et al., 1980,
1982). Using a modified Elek test and immunohalo test with antiserum or anti haemolysin lgG, it was possible to identify the Kanagawa phenomenon on a single plate. Workers of the same group later developed several ELlS A methods for the detection of TDH and TRH
(Honda et al., 1985, 1989b, 1990) and subsequently developed and evaluated two assay kits for TDH and TRH that are commercially available (Yoh et al., 1995). The two kits, reversed passive latex agglutination (RPLA) and Bead-ELISA, although originally developed for detection of TDH, were found to detect TRH with a minimum concentration of 10 ng/ml.
2.6 Molecular detection methods
2.6.1 PCR methods for V. parallaemolyticus
A number of successful PCR-based assays have been developed to assist in the specific detection of V parahaemolyticus and its toxin genes. Most of the species specific gene targets used by workers include the genes toxR, tlh, and gyrB, the genes encoding for ToxR transmembrane protein, thermolabile haemolysin (section 2.4.4) and DNA gyrase subunit
B respectively (Kim et al., 1999; Davis et al., 2004; Venkateswaran et al., 1998). Other gene targets that have been used with varying degrees of success include the heat shock protein hsp60 gene, the species specific pR72H fragment and to some extent the 16S- 23S spacer regions (Kwok et al., 2002; Lee et al., 1995; Sparagano et al., 2002). Several of these gene targets have been used collectively as multiplex PCR systems and quantitative real-time PCR (QRT-PCR) methods. 28 2.6.2 toxR gene based PCR methods
ToxR is a trans-membrane protein and the transcriptional activator of Vibrionaceae (Miller et al. 1987; Lin et al., 1993). It has been found in many Vibrio spp. and shown to be well conserved amongst its members (Higgins et al., 1996; Li et al., 2000; Okuda et al., 2001 ).
However, some variations in sequence identity and divergence of toxR (Fig. 2.4) allow the design of PCR primers for specific Vibrio spp (Osorio & Klose, 2000).
A P. damse/ae subsp. dam se/Be B Vibrio uulnl(lcus --.;.;.;..-f P. damselaesubsp. plsc/c/da Vibrio fluvi•li•
Vibrio alglnolyticus
Vibrio parDhaemolylicus
Vibrio a/glnolytlcus VIbrio mlm/cus
Vibrio parshsemolyt/cus
P. dsmSIIIIHI subsp. d11mu/ae
P. damselao subsp. prsc/clda
..t--- Vibrio fluvia/ls "---- VIbrio ho/1/ue 1% Divergence 10% Divergence '---- VIbrio vulniflcus '----- P. pro/undum
Figure 2.3: Unrooted trees constructed by the neighbor-joining method showing the phylogenetic interrelationships of the different Vibrio and P!Jotobacterium species based on toxR gene nucleotide sequences (A) and 16S rRNA genes (B) (Osorio & Klose, 2000)
Using PCR targeted to toxR, Kim et al. ( 1999) showed that all 373 V parahaemolyticus isolates tested were positive for toxR and that the gene was present in other members of
Vibrio spp. Some of the PCR reactions exhibited amplicons of different sizes and intensities suggesting toxR sequence variation in Vibrio species that may reflect phylogenetic relationships (Fig. 2.3). This was shown to be true in a follow up study by
Osorio & Klose (2000). The toxR region has been widely used as a species specific marker in various PCR protocols including multiplex PCRs and real time PCRs (Kim et al. , 1999;
Lin et al., 1993; Alam et al., 2002; Dileep et al., 2003). In a comparison with toxR and two other loci (tlh and pR72H) Croci et al. were able to show that toxR provided the best
29 results in the identification of V parahaemolylicus with an inclusivity of 96%, no false positives, and a concordance between repeated determinations of97% (Croci et al., 2007).
2.6.3 PCR targeted to tilt
Several workers have used tlh, the gene for thermolabile haemolysin (section 2.4.4) as a species specific marker (Bej et al., 1999; Kaufman et al., 2002, Miwa et al., 2003). Work by McCarthy et al. demonstrated the use of an alkaline phosphase and digoxygenin (DIG) labelled tlh probe that worked equally as well as the tlh PCR protocol (McCarthy et al.,
1999). The tlh PCR based probe showed 97% agreement with the API 20E diagnostic strips in the identification of V parahaemolyticus (section 2.5.5) and was found to be considerably more efficient and reduce time and effort. In addition to the tlh PCR, several other targets such as a collagenase and metalloprotease gene have also been used as targets for the identification of V parahaemolyticus with varying degrees of success (Di Pinto et al., 2005; Luan et al., 2007). However, these targets have yet to be assessed and validated by others.
2.6.4 Primers for gryB
Venkateswaran and workers developed V parahaemolyticus specific DNA gyrase subunit
B (gyrB) primers. They showed that V parahaemolyticus and V alginolyticus shared a
86.8% homology in nucleotide sequences and developed a set of primers that correctly detected I 17 strains of the target strain from 267 non-target strains (V enkateswaran et al.,
1998). Using sequence data from gyrB, and other genes from multilocus sequence typing
(section 2.7.10) Chowdhury et al. were able to assess evolution of the pandemic 03:K6 strains that had arisen since 1996 (Chowdhury et al., 2004b). Other workers have developed multiplex, microarray and real-time PCR assays for the specific detection of V parahaemolylicus (Gonzalez et al., 2004; Cai et al., 2006).
30 2.6.5 PCR ofpR72H
Several workers have used the pR72H fragment obtained from a Hindlll DNA digest of V parahaemolyticus DNA cloned into pUCII9. This fragment was found in all strains of V parahaemolyticus tested and thus it was recommended as a species specific DNA probe.
(Lee et al., 1995; Croci et al., 2002, 2007) Using this fragment Hervio-Heath and colleagues identified a novel 320bp amplicon, atypical of the 387bp fragment usually found, in 24.4% of the V. parahaemolyticus isolates (Hervio-Heath et al., 2002). Workers in Chile identified two other amplicons using the same primers; a 400bp fragment and a
340bp fragment in both clinical and environmental samples. However, a third group of isolates of V. parahaemolyticus did not show a PCR band (Cordova et al., 2002) and consequently the use ofpR72H as a species specific marker remains questionable.
2.6.6 PCR methods for toxin genes
PCR for tdh and trh have widely been used both as a marker for pathogenic isolates in environmental samples as well as confirmation in clinical studies (Lee et al., 1993;
Karunasagar et al., 1996; Wong et al., 1999a; Alam et al., 2002; Robert-Pillot et al., 2004).
The most widely used sets of primers are those developed by Tada et al. ( 1992) that target the tdh2 and trhl regions. The gene tdh2 is the more representative of tdh genes and is found to be more strongly associated with haemolysis of Kanagawa positive isolates than tdh I genes (Nishibuchi & Kaper, 1990; lida et al., 1990). Although the haemolysin TDH can be assayed using immunological tests and to some extent Wagatsuma agar, to date there appears to be no conventional tests for TRH production.
2.6. 7 Multiplex PCR
The development of multiplex PCR (M-PCR) for pathogenic isolates have been reported
by several workers using both tdh, trh and several species specific genes (Bej et al., 1999;
Kong et al., 2002; DePaola et al., 2003b; Okura et al., 2003; Panicker et al., 2004; Tarr et 31 a/.,2007). Using M-PCR Lee et al. (2003) showed the detection of V. parahaemolyticus and related species from shellfish using tl (tlh) primers. The amplicons were then subjected to colorimetric CovaLink™ NH microwell plate sandwich hybridisation using phosphorylated and biotinlylated tlh probes.
Panicker et al. (2004), used tdh, trh, tlh and orj8 amplicons derived from a multiplex PCR as targets for oligonucleotide probes. The probes were deposited onto Teflon-masked slides by using a MicroGrid ll arrayer and hybridisation detected by using tyramide signal amplification with Alexa Fluor 546 fluorescent dye. This microarray approach detected 102 to 103 CFU/ml of target organism in pure cultures and 1 CFU in 1 g of oyster tissue homogenate after Sh of sample enrichment with 100% specificity. Consequently, the authors suggested the use of a microarray approach together with multiplex PCR for efficient detection of pathogenic V. parahaemo/yticus from shellfish samples.
2.6.8 DNA hybridisation probes and ELISA methods
DNA-DNA hybridisation probes have been used extensively as they can usually be applied directly onto plate grown cultures allowing identification of target genes and test all the colonies on a plate at the same time. Both digoxigenin (DIG) and alkaline phosphatase- conjugated probes have been used by several workers to target species specific and virulence genes (Yamamoto et al., 1992a; McCarthy et al., 1999; Gooch et al., 200 I;
Nordstrom et al., 2006; Raghunath et al., 2007). In comparison with API 20E identification strips, McCarthy et al. showed that specificity in identifying V. parahaemolyticus using the alkaline phosphatase probe and DIG-labelled method was 97% and 98% respectively.
However, although the probes were specific, the sensitivity of this detection method was less than expected (McCarthy et al., 2000; Gooch et al., 200 I). Lee et al. subsequently developed a CovaLink™ NH microwell plate approach using tlh primers that was reported to detect 102 cells/g of oyster tissue homogenate (Lee et al., 2003) and offered a shorter 32 hybridisation time, higher sensitivity and high throughput sample processing, although this has not been tested with pathogenic isolates or using tdh or trh probes.
2.6.9 Quantitative real-time PCR (QRT-PCR)
Increasingly, a number of workers have reported QRT-PCR methods usmg various markers including tdh, trh, t/h, toxR, gyrB and both tlurogenic and TaqMan probes (Davis et al., 2004; Iijima et al., 2004; Kaufman et al., 2004; Takahashi et al., 2005; Cai et al.,
2006). Blackstone et al. (2003) used a tlurogenic tdh probe to test 131 Alabama oyster samples that were enriched in alkaline peptone water (APW). In comparison to using a streak plate method followed by tdh probe, the 24 hour QRT-PCR method was able to detect the gene tdh in 61 samples, whereas previously the streak plate/probe method detected tdh in 15 samples, a procedure reported to take 3 days (Blackstone et al., 2003).
Similarly, Bej and coworkers multiplexed four targets tdh, trh, tlh and orj8, for a count of total and potentially pathogenic V parahaemolyticus for a QRT-PCR procedure that was sensitive for 200 pg of purified genomic DNA and I 04 CFU per ml for pure cultures (Ward et al., 2006; Rizvi et al., 2006). The high sensitivity of QRT-PCR was also demonstrated by Nordstrom et al. (2007) to detect total and potentially pathogenic V. parahaemolyticus.
2.7 Strain differentiation by molecular typing methods
2.7.1 Epidemiological typing of clinically significant strains
Due to the large number of outbreaks of V. parahaemolyticus food poisoning worldwide and an increasing number of environmentally isolated, potentially pathogenic strains, there have recently been numerous strategies developed to delineate patterns of transmission. In some cases, typing has contributed to epidemiological surveillance and evaluation of control measures by assessing prevalence and distribution of epidemic and pandemic clones. Several molecular typing techniques for V parahaemolyticus are under
33 development and more recently, there has been a growing concern for standardised typing methods for V. parahaemolyticus.
2. 7.2 Principles and properties of a molecular typing method
The accurate recognition and differentiation of subtypes within a bacterial species and the use of subtype identification methods to determine sources of infection, rates of spread of infection and mechanisms of transmission is referred to as typing. The basis for all molecular typing methods is that an outbreak of an infectious disease is the result of exposure to a pathogen that is derived from a single cell or clone and its subsequent progeny will be genetically identical or closely related to the source organisms (Streulens,
1998). The assumption is that at species level there is sufficient genetic diversity so that organisms isolated from different sources and/or at different times and locations can be classified as different isolates, subtypes or strains (Olive & Bean, 1999). The criteria for evaluating performance of typing systems may include assessments on typeability, discriminatory power, reproducibility, stability, epidemiological concordance and comparative study (Streulens, 1998).
Ideally, the techniques used in a molecular typing system should be standardised and reproducible so that they can be carried out in any laboratory and give informative and comparable data, which is applicable to all isolates. The system should be sufficiently discriminatory to distinguish a reasonable number of types amongst which isolates can be roughly equally distributed. Furthermore, a typing system should be able to differentiate unrelated strains such as those which are geographically distinct from the source organisms but at the same time demonstrate the relationships that exist between organisms isolated from individuals infected through the same source (Olive & Bean, 1999).
34 2.7.3 Pulsed field gel electrophoresis (PFGE)
Over recent years, pulsed field gel electrophoresis (PFGE) has emerged as the "gold standard" in molecular typing methodology and is the current method of choice for typing nosocomial and community acquired pathogens (Tenover et al., 1997). PFGE IS an extremely sensitive and sophisticated method that uses whole intact genomic DNA prepared by embedding bacteria in agarose plugs. Bacterial cells are lysed in situ, followed by restriction enzyme digestion of the intact genomic DNA with an infrequently cutting enzyme after removal of proteins. The large DNA fragments generated are then separated in highly purified agarose using specialist gel electrophoresis systems. Differences in the length and number of restriction fragments arise as a result of the changes in restriction enzyme sites caused by duplication, mutation, deletion, insertion or inversion (Tenover et al., 1997).
PFGE has been used widely in regions of South East Asia where gastroenteritis due to V parahaemolyticus food poisoning is endemic. Selections of different restriction enzymes including Notl, Sfil and Apal have been used with various protocols and pulse times (Bag et al., 1999; Wong et al., I999a; Lu et al., 2000; Ma1tinez-Urtaza et al., 2004; Yeung et al., 2002; Chowdhury et al., 2004a; Tanil et al., 2005). One of the first reports of an extensive use of PFGE for subtyping of V parahaemolyticus was reported for 130 clinical isolates in Taiwan in 1996 (Wong et al., 1996). Although comprehensive fingerprinting data were not presented, they reported clustering of the 130 isolates into 14 PFGE types one of which seemed to be clonally related to an isolate from Japan. Conversely, clonal diversity within the serotype 03 :K6 was demonstrated by Bag and colleagues (1999), who combined the use of PFGE with ribotyping (section 2.7.4). Although there had been genetic reassortment among the 30 clinical 03:K6 strains tested, they were still
predominantly one clone. A similar study incorporated the use of an arbitrarily primed
PCR (AP-PCR) method with PFGE to identify a second serovar, 04:K68 in Thailand and 35 India that shared 78-91% genomic similarity to that of 03:K6 strains (Chowdhury et al.,
2000a). Using PFGE, Okura et al. (2003) also suggested that the 01 :K25, 01 :K26, and
01 :K untypeable pandemic serovars were derivatives of one clone.
Since the use of PFGE in typing V parahaemolyticus in 1996, a number of methods with variations in the use of restriction enzymes (Not], Sfil and Apal), pulse switch times, equipment and other parameters have been employed with varying degrees of success
(Marshall et al., 1999). The main concern is a lack of typeabilty as DNA degradation occurs over the long preparation and processing times, not just in V parahaemolyticus but also in V cholerae and V vulnificus (Gonzalez-Escalona et al., 2007). A number of workers have recently adopted the 'One day (24-28 h) Standardised Laboratory Protocol for Molecular Subtyping of Non-typhoidal Salmonella by PFGE' (PulseNet; CDC, Atlanta,
Ga) with increased typeability. However, with an increasing number of outbreaks worldwide, and with PFGE as the preferred method of subtyping, there have been growing concerns for a standardised PFGE V parahaemolyticus protocol to facilitate inter-
laboratory strain comparisons. To this end, two standardised PFGE procedure for V parahaemolyticus have recently been described (Parsons et a/, 2007; Wong et al., 2007)
and remain to be validated by others.
2.7.4 Restriction fragment length polymorphism (RFLP) analysis
Restriction fragment length polymorphism (RFLP) analysis has been used on a variety of
occasions, usually in conjunction with other methods (Gendel et al., 200 l; Yeung et al.,
2002; Wong et al., l999c). RFLP analysis uses restriction enzymes to digest genomic
DNA and the resulting fragments are separated according to size by agarose gel
electrophoresis in much the same way as PFGE without the need for sophisticated gel
electrophoresis apparatus. This method has been used to characterise polymorphisms in
virulence genes tdh and trh (Suthienkul et al., 1996) and genes encoding for components of 36 polar flagellar, Fla locus of V parahaemolyticus (Marshall et al., 1999). However, because of the complexity of the resulting pattern due to a large number of restriction fragments in a genome, the DNA is transferred to a membrane (i.e. Southern blot) and probed by DNA hybridisation using a labelled probe. Several probes have been used to characterise V parahaemolyticus, the most common being ribosomal based probes
(ribotyping), (Wong et al., 1999b).
In what appears to be the first of two comparative studies of several molecular typing techniques, Marshall and colleagues used the rrnB rRNA operon of E. coli to ribotype 60 isolates (Marshall et al., 1999). Although full genotypic data was not shown, Bgll digested genomic ON A was reported to reveal 14 different types in comparison to 15 and 13 types as shown by PFGE and enterobacterial repetitive intergenic consensus sequence PCR
(ERIC-PCR, section 2.7.8) respectively. Using Simpson's index of diversity (DI) described by Hunter and Gaston (1988), the authors concluded that both ribotyping and ERIC PCR
(section 2.7.8) provided the best discrimination, similar to that of PFGE discrimination while Fla locus RFLP analysis provided the lowest discrimination index (Marshall et al.,
1999). Another method employing RFLP that does not appear to have been assessed in V parahaemolyticus include the use of insertion sequence (IS) probes, especially as there is evidence to suggest that tdh has been found to be spread by insertion sequence like elements (ISVs) (Baba et al., 1991b; Terai et al., 1991). IS probes such as IS200 have been used successfully to type Salmonella spp, Streptococcus pneumoniae (IS IS1167) and
Mycobacterium tuberculosis (lS6110) (Stanley & Saunders, 1996; Luna et al., 2000).
2.7.5 Randomly amplified polymorphic DNA (RAPD) PCR
Randomly amplified polymorphic DNA PCR (RAPD) is a widely used method that examines nucleotide variation throughout the entire genome and can be more discriminatory than methods that examine variation at a specific locus, particularly in 37 studies of highly conserved species. It employs the use of short oligonucleotide primers of arbitrary sequence that bind to any region in the genome bearing the complementary sequence. Banding patterns are due to mismatches or appearances of new primer sites and the length of amplified region between the primer sites. RAPD has been used by several workers to characterise diverse strains of V. parahaemolyticus with various degrees of success (Wong et al., 1999c; Islam et al., 2004; Maluping et al., 2005; Hayat et al.,2006;
Parvathi et al., 2006).
In one of the first reported studies of RAPD-PCR with V parahaemolyticus, Wong et al. screened 100 different 10-mer primers (UBC RAPD primer set 100/3, Finnzymes Oy,
Espoo, Finland) with a single clinical and environmental strain and proceeded to type 308 clinical isolates to identify 41 polymorphic RAPD patterns (Wong et al., 1999). Although quality of data was not apparent, the RAPD clustering identified 5 major clinical subtypes that accounted for 91% of all the isolates and were phylogenetically more closely related to each other indicating a largely homogeneous group amongst the clinical strains tested. In another study that involved 1500 environmental isolates of V. parahaemolyticus in
Bangladesh, the workers identified and characterised five tdh positive and clinically significant strains (Islam et al., 2004). These five isolates were found to belong to three different serotypes; two that belonged to 03:K6 however, believed to be clonally related,
showed identical RAPD bands. Similar results were also published by Hayat et al. (2006)
who showed 0.3% of environn1ental isolates to be toxigenic and to possess identical RAPD
bands to that of pandemic reference strains. Similar findings were reported by others on the
application of RAPD using arbitrary primers to distinguish between V. parahaemolyticus
strains (Matsumoto et al. 2000; Chowdhury et al., 2000b; Okura et al., 2003). These
methods employed one or more of the primers provided with the Ready-To-Go RAPD
Analysis Kit (Amersham Biosciences). However, a low number of amplicons and the use
of different control strains do not facilitate meaningful comparison of data. 38 2.7.6 Ribosomal DNA intergeoic spacer (ITS) region analysis
The genes encoding subunits of bacterial rRNA are typically arranged in a characteristic formation, 16S, 23S and SS. Regions between the 16S and 23S rRNA genes and between
23S and SS rRNA genes are known to be more variable in both size and nucleotide sequence than the flanking genes and are termed ITS 1 and ITS2 respectively. The acronyms, IGS (intergenic spacer) and ISR (intergenic spacer region) have also been frequently used. Within a species, the 16S rRNA is often highly conserved with variable regions often used for phylogentic and taxonomic purposes. The intergenic spacer region shows greater variability and can be used to distinguish between closely related bacterial species and, sometimes, between strains and isolates. A number of copies of the rRNA operon are known to exist throughout the genome in different species and this may provide adequate variation for epidemiological typing purposes (Gurtler & Stanisich, 1996).
Variation in the ITS I and ITS2 regions have been the basis for the development of a number of subtyping systems for a wide number of species although sequence variation in the ITS 1 region has been more widely used due to its larger size and consequently higher probability of nucleotide polymorphisms. In the recent genome sequencing project of V parahaemolyticus, 11 rRNA operons (rrn) that fell into 6 classes were identified (Makino et al., 2003) with identical 16S rRNA genes (rrs) (Gonzalez-Escalona et al., 200Sb).
Although this large number of operons may produce complex ITS 1 banding patterns, the relative frequency and distribution of rRNA genes throughout the genome amongst strains of V parahaemolyticus may provide a significant tool for intraspecies subtyping of this organism. For example, a strain of V parahaemolyticus that had been isolated in 19S1 only contained two rrn sequence types, one of which was found in seven of the 11 rrn operons and the other in four (Maeda el al., 2000).
39 The ITS I region was first analysed in V cholerae by Coelho and colleagues (1995) who used primers that flanked 16S and 23S rDNAs of E. coli to show that ITS I PCR patterns from 01 classical, 01 El Tor and 0139 strains were different and therefore provide a potential tool for studying the epidemiology of V cholerae. In a follow up study by Chun et al., (1999), V cholerae and V mimicus ITS I regions were studied using more specific primers based on sequences from E. coli, Pseudomonas aeruginosa, V cholerae and V vulnificus (Chun et al., 1999). Perhaps not surprisingly due to a high specificity of primers and the use of only I 0 isolates in the study, only four ITS I variants were identified in both
V cholerae and V mimicus where up to I 0 rrn operons had previously been described
(Nandi et al., 1997).
2.7.7 Transfer DNA-PCR analysis
Conserved genes encoding tRNAs can be found widely distributed throughout a genome, usually clustered in multiple tandem-repeat units (Green & Void, 1993). DNA that encodes for RNA is termed tDNA and the intergenic space between these tRNA genes can vary considerably between species and sub species (Welsh & McCelland, 199la). tDNA gene regions appear to be hot-spots for the integration of prokaryotic genetic elements (Reiter et al., 1989; section 2.4.1 0, insertion sequences) and has been shown in E. coli to confer specific traits such as enhanced virulence, antibiotic resistance and fitness traits necessary to adapt to certain niches (Dobrindt et al., 2004; Hayashi et al., 2001; Welch et a/ 2002).
Low stringency PCR products based on primed tDNA regions can amplify comparable inter-length polymorphisms (tDNA-ILPs) between different strains and species and can be used to assess genome wide variation (Welsh & McCelland, 1991a). tDNA-ILPs have been used by several workers to rapidly subtype a wide variety of bacterial species including
Acinetobacter spp., Staphylococcus spp., Renibacterium salmoninarum, Enterobacter cloacae and Lactobacillus spp., (Ehrenstein et al., 1996, Maes et al., 1997; Alexander et al., 200 I; Clementino et al., 200 I, Baele et a/ ., 2002). 40 Welsh and McCielland developed a set of 'consensus' tRNA gene primers from multiple alignments of tRNA genes from a wide variety of bacterial species including E. coli. The primers were designed to face outwards from the end of the tRNA genes to amplify a tDNA-ILP fingerprint determined by the arrangement of tRNA genes on the bacterial genome. Due to the conserved order and arrangement of tRNA genes on the genome, the banding profiles that are generated using consensus primers are often characteristic of a particular species and can be identical or nearly identical within every species (Welsh &
McCelland, 1991a, 1992). Considering the relatively large number of tRNA genes that have been found in the Vibrionaecae, (Maeda et al., 2000), and in particular, 126 in V parahaemolyticus, (Makino et al., 2003) that coincide with multiple rrn operons, tDNA-
ILPs show significant potential as a method of sub typing V parahaemolyticus.
2.7.8 Repetitive extragenic palindromic sequence PCR (REP-PCR)
Repetitive extragenic palindromic (REP) sequences are repetitive elements that also appear to be distributed throughout bacterial genomes and consist of a conserved 33 bp inverted repeat region (Stem et al., 1984). The function of the REP sequence in the prokaryotic genome remains largely unknown although the sequences appear to stabilise upstream mRNA regions and influence gene expression (Newbury et al., 1987). Although a high discrimination index was shown by Wong et al. (2001) in subtyping V parahaemolyticus,
Maluping et al. demonstrated less reproducibility even though the same primers and PCR conditions were used (Maluping et al., 2005)
Another system that employs the use of conserved regions that have widely been used for bacterial typing includes ERIC- PCR. These sequences are a family of repetitive elements that were first described in the genomes of E. coli and Salmonella Typhimurium and also found in other species including V cholerae (Hulton et al., 1991). The sequence is 126 bp long and has been shown to be highly conserved and restricted to transcribed regions of the 41 genome, either in the intergenic regions of polycistronic operons or in the untranslated regions that lie upstream or downstream of open reading frames (Oebruijn, et al., 1992).
Although ERIC sequences are highly conserved, their chromosomal locations are known to vary between different bacterial species. ERIC-PCR has been widely used to characterise
V cholerae since 1991 and more recently used for V parahaemolyticus with varying degrees of success (Khan et al., 2002; Maluping et al., 2005; Tanil et al., 2005).
As well as ERIC and REP sequences, repetitive BOX sequences that appear dispersed throughout the genome have also been used to subtype microorganisms. BOX elements have been shown to consist of three subunits termed boxA, boxB, and boxC. Sequence analysis relevealed that boxA was highly conserved in bacteria, whilst boxB and boxC were highly variable and found only within S. pneumoniae (Versalovic et al., 1995) and since then primers for the boxA region have been used widely in several species and more recently in Aeromonas spp. (Tacao et al., 2005) and B. pseudomallei (Currie et al., 2007).
In a comparative study of molecular typing systems, Wong & Lin (2001) compared three
PCR methods for subspecies typing V parahaemolyticus that included ERIC- PCR, REP
PCR and ITS-PCR. Forty strains representing a wide range of PFGE profiles were grouped into 27, 27 and 15 patterns with discrimination indexes of 0.98, 0.97 and 0.91 by ERIC
PCR, REP-PCR and ITS-PCR respectively (Wong & Lin, 2001). In another study, PFGE was shown to be a more discriminatory method than BOX-PCR although both methods were able to distinguish between closely related strains of E. coli. However, neither of the methods used were effective in clustering E. coli strains according to the source of the organism (Cesaris et al., 2007).
42 2.7.9 Multi locus enzyme electrophoresis (MLEE)
Multi locus enzyme electrophoresis (MLEE) is a technique that examines the electrophoretic mobility of a selection of metabolic enzymes. The variation in mobility is visualised by staining of gels to reveal specific enzyme activity and is attributed to isozymes or allozymes. MLEE is commonly used to study long-term epidemiology on a global scale where minor differences between related strains are not required.
Farfan et al. (2000) used MLEE to study genetic relationships between a diverse group of
107 clinical and environmental isolates of V cholerae to determine allelic variation in IS housekeeping enzyme loci. They identified 99 electrophoretic types with a genotypic diversity (G) of 0.9872 for all samples. All IS enzyme loci studied were polymorphic, the number of alleles ranging from 2 to 7. The authors concluded that the 0139 and 01 El Tor isolates were genetically more related to each other than all the other sub-populations studied.
Another study using 62 strains of V vu/nificus strains, placed the population into two divisions that were comparable to the two divisions obtained using RAPD (section 2.7.4)
(Gutacker et al., 2003). The method did not show any significant clustering or association between distinct clones and as such its use is inappropriate for subtyping isolates from local outbreaks of infection.
2.7.10 Multi locus sequence typing (MLST)
Multi-locus sequence typing (MLST) is a nucleotide sequence based approach that was proposed in 1998, borrowing from the concepts of MLEE (Maiden et al., 1998). Instead of using the variation in electrophoretic mobility of a selection of enzymes to assess allelic genotypes, the variation of the gene sequences, is bioinformatically assessed for relatedness using software and database approaches. 43 In a recent finding using three housekeeping gene loci and three virulence gene loci,
Kotetishvili et al., (2003) demonstrated a better discriminatory ability of MLST than PFGE and since then several workers have used MLST for subtyping of V cholerae (Lee et al.,
2006; Danin-Poleg et al., 2007). Using MLST sequence data from six housekeeping genes of V vulnificus, Cohen et al. (2007) were again able to subtype 94% of the isolates into the two distinct lineages commonly associated in this species. To date, ten housekeeping genes have been proposed for V vulnificus (Bisharat et al, 2007a; 2007b). Of these ten, five of these loci have also been adopted for the MLST typing of V parahaemolyticus with an additional two housekeeping loci, recA, dnaE and the variable gene ompK (Gonzalez
Escalona et al., 2008). Using MLST of 100 isolates of V parahaemolyticus, three major clonal complexes and 62 sequence types were identified. Based on these initial studies it was concluded that V parahaemolyticus is genetically diverse exhibiting a semi-clonal population structure that is driven primarily by frequent recombination rather than mutation events (Gonzalez-Escalona et al., 2008). However, as with all MLST methods, it is considerably more resource intensive, expensive and impractical for typing a large number of V parahaemolyticus isolates from shellfish samples.
2.7.11 Variable number tandem repeats PCR (VNTR-PCR)
Short DNA sequence motifs that occur in specific regions of a genome where there are variations in the number of repeat units can provide a source of high genomic polymorphism are often termed variable number tandem repeats (VNTR), single sequence repeats (SSR) or 'multi locus variable number of tandem repeat analysis' (ML V A) (van
Belkum, 1999; 2007). The number of repeat units that occur at a locus can vary between strains (and within strains) due to many reasons including slipped strand mispairing, deletions and insertions. VNTR analysis makes use of PCR primers complementary to the stable flanking sequences and polymorphisms in the numbers of repeat units are
44 detennined by agarose gel electrophoresis (van Belkum et al., 1998). In ML V A, the approach can be supported or extended by locus-specific DNA sequencing.
VNTR has been used for a variety of applications including human genetics and bacterial epidemiological studies most notably in subtyping Mycobacterium spp. (Roring et al.,
2004). More recently, it has been applied to members of the Vibrioneacae; V. cholerae and
V. vulnificus. Danin-Poleg et al. (2007) screened the V. cholerae genome sequence to reveal thousands of SSR tracts with an average frequency of one SSR every I 52 bp. Using
32 strains of V. cholerae that represented both clinical and environmental isolates, the workers tested for polymorphism in SSR loci with two methods including polymorphism of SSR tracts longer than 12 bp and mononucleotide repeats as assessed by sequencing.
Phylogenetic analysis of combined data was able to discriminate between the clinical 01 and 0139 serogroups and environmental isolates. This ability to discriminate between all but three clinical and environmental isolates from a total of 45 strains V. cholerae was also demonstrated (Danin-Poleg et al., 2007).
Unlike the various other PCR methods previously described, VNTR regions are sequence specific for each bacterial species and may occur at only a single locus. Whilst VNTRs
have been widely used on variety of applications, until recently, there had been no reports of the technique applied to V parahaemolyticus. Kimura et al. (2008) recently showed
MLVA to have high resolution and reproducibility for typing of V parahaemolyticus
03:K6 that were clonally related. In 28 pandemic V. parahaemolyticus 03:K6 isolates
from human cases, ML V A produced 28 distinct VNTR patterns at eight loci (Kimura et al.
2008) and differentiated a further six environmental 03:K6 isolates that apparently showed
the same PFGE pattern. However, the parameters that were used to obtain the PFGE clusters were unclear, and three clinical isolates were untypeable by PFGE. As with
45 ' ' .
M!!,ST;. analysis! of sequence data for a. large number· of envirortrhehtal is9lates is'
;Impractical ,andi wo_tJ!dii:Je cliffic~IHQ implement for routine :use.
46 Chapter Three
GENERAL MATERIALS AND METHODS
3.1 Chemicals and reagents
Unless otherwise stated all general laboratory chemicals were purchased from Amersham
Biosciences, Bio-Rad Laboratories, Invitrogen, Them10-Fisher Scientific, or Sigma
Aidrich, and, if possible, were of molecular biology grade. All bacteriological culture media was obtained from Oxoid and PCR related reagents and restriction enzymes were obtained from Roche Biosciences. Recipes for all of the buffers that were used and the suppliers' details are outlined in Appendix I. Details of specific reagents can be found in the Materials and Methods section of individual chapters.
3.2 Bacteriology
3.2.1 Isolates of V. parallaemolyticus
The designation codes of V parahaemolylicus obtained from various locations between
2003 and 2006 and related organisms used in this study together with their countries of origin and source of isolation where known, and PCR data for tdh, trh and toxR where available are listed in Table 3.1. A total of 86 different isolates of V parahaemolyticus and closely related members of the Vibrionaecae were used for various typing procedures.
Unless otherwise stated, the isolates were held as 50% glycerol stocks ( -70°) or on 2%
Marine Salts Agar (MSA) in the University of Plymouth culture collection. A selection of biochemical tests and growth characteristics on several bacteriological media were perfonned regularly to confinn identity and purity of the cultures. Together with PCR and hybridisation analysis, molecular typing was performed on isolates as appropriate for which further details are provided in each chapter. These techniques included the analysis of one or more of the following: pulsed field gel electrophoresis (Chapter 5), l6S-23S and
47 23S-5S rRNA intergenic spacer analysis (Chapter 6), tDNA intergenic length polymorphism analysis (chapter 7) and randomly amplified polymorphic DNA (chapter 8).
The characterisation of V parahaemolyticus was the first of its kind in the UK and sourcing of clinical isolates from UK bacterial culture collections proved to be very difficult. Where possible, clinical isolates were sought from other countries and this study represents a preliminary study to assess the feasibility of molecular typing techniques to determine clinical and environmental isolates of V parahaemolyticus.
3.2.2 Culture and growth of V. parallaemolyticus
All V parahaemolyticus isolates were recovered from 50% glycerol stocks or grown from
MSA slopes stored at room temperature in the dark. Isolates were cultured for 8 to 18 hours (depending on use) at 37°C in either 15 ml volumes of tryptone soya broth (TSB) with a final NaCI concentration of 3%, or MSA with a fmal salt concentration of 2%, shaking in a water bath (Ciifton NE5-28D) at 70 rpm.
3.2.3 Culture and growth of additional bacterial species
Bacterial isolates other than V parahaemolyticus that were used in this study are listed in
Table 2.3 together with growth medium. All halophilic Vibrio spp. were grown in TSB and
MSA supplemented with the appropriate amount of NaCI. Escherichia coli was grown on
Luria-Bertani (LB) medium overnight at 37°C.
48 Table 3.1: Description of V. parallaemolylicus and their genotypes, where known, that were used in this study. The Strain ID is used throughout the study.
Strain CEFAS Environmental/Clinical/ PCR (2) Ure ID Rderence tnslitute(l) Location Reference strain toxR tdlr trh + 1902 V05/00I CEFAS PHLS, UK NCIMB Reference Strain + + VPI0884 V05/002 CEFAS PHLS, UK NCTC Reference Strain + VPI0885 VPI0885 NCTC Reference Strain + 219 VOS/009 NSVS Oslo, Norway Clinical + 220 V05/0IO NSVS Oslo, Norway Clinical + + + + 221 VOS/011 NSVS Oslo, Norway Clinical + + (+) 222 VOS/012 NSVS Oslo, Norway Clinical + + 223 VOS/013 NSVS Oslo, Norway Clinical + + + + 224 VOS/014 NSVS Oslo, Norway Clinical + + 225 VOS/015 NSVS Oslo, Norway Clinical + + 226 VOS/016 NSVS Oslo, Norway Clinical + + 227 VOS/017 NSVS Oslo, Norway Clinical + (+) + 228 V05/018 NSVS Oslo, Norway Clinical + 1022/4 V05/020 CEFAS Galicia, Spain Environmental + 1024/5 V05/021 CEFAS Galicia, Spain Environmental + 1021/5 VP 102115 CEFAS Galicia, Spain Environmental + 1022/10 VPI022/IO CEFAS Galicia, Spain Environmental + 1023/1 VP 102311 CEFAS Golicia, Spain Environmental + 1023/2 VP 102312 CEFAS Galicia, Spain Environmental + 1023/3 VP 1023/3 CEFAS Galicia, Spain Environmental + 1103/9 VP 1103/9 CEFAS Galicia, Spain Environmental + + 2053 VP 2053 KUMS Kyoto, Japan Clinical + + + + 2436 VP 2346 KUMS Kyoto, Japan Clinical + + 5421 VP 5421 KUMS Kyoto, Japan Clinical + + 6134 VP 6134 KUMS Kyoto, Japan Clinical + + 6316(3) VP 6136 KUMS Kyoto, Japan Clinical + Vibl Vib I Southampton, PHL Environmental + Vib2 Vib2 Southampton, PHL Environmental + Vib3 Vib3 Southan1pton, PHL Environmental + Vib5 Vib 5 Southan1pton, PHL Environmental + Vib6 Vib6 Southampton, PHL Environmen!al + + Vib7 Vib 7 Southampton, PHL Environmental + Vib8 Vib 8 Southampton, PHL Environmental + + Vib9 Vib9 Southampton, PHL Environmental + + VibiO Vib 10 Southampton, PHL Environmental + VPI VP I Southan1pton, PHL Environmental + VP2 V05/022 CEFAS Southampton, UK Environmental + + VP3 VOS/023 CEFAS Southampton, UK Environmental VP4 VP4 CEFAS Southan1pton, UK Environmental VPS VP 5 CEFAS Southampton, UK Environmental + VP6 VP6 CEFAS Southampton, UK Environrnenlal + + + VP7 V05/024 CEFAS Southampton, UK Environmental + + + VP8 V05/025 CEFAS Southampton, UK Environmental + + + VP9 V05/026 CEFAS Southampton, UK Environmental + + + VPIO V05/027 CEFAS Southampton, UK Environmental
49 Table 3.1: COIII.
Strain CEFAS En•ironmentai/Ciiniral/ PCR(2) Ure ID Reference lnslitutejl) Location Reference strain toxR tdll trh + + E154482 VP E154482 PHLS Unknown Clinical + + El55855 VP El55855 PHLS Unknown Clinical + + El68143 VP E168143 PHLS Unknown Clinical + + + + VP F3305 Clinical + VP331 VP 02/331 Holes Bay En•ironmental (G) DOM Atlantic + VP345 VOS/028 CEFAS Shellfish, UK En•ironmental (G) + VP346 VP 021346 Dom Fota Farm En•ironmental (G) + + VP349 VOS/029 CEFAS Holes Bay, UK En•ironmental (G) + + + VP350 VOS/030 CEFAS Arne, UK En•ironmental (G) VP361 VP 02/361 DOM Fota Farm En•ironmental (G) DOM Atlantic + + VP377 VP 02/377 Shellfish Environmental (G) + VP378 VP02/378 DOM Fota Fann Environmental (G) + + VP416 VP 01/416 Holes Bay Environmental (G) + VP491 VOS/031 CEFAS Weymouth, UK Environmental (M) + VP939 VP 02/939 Coedcanlas spit Environmental (E) + + VPIECI VOS/032 CEFAS Portland, UK Environmental (P) VPSCI VP 02/SCI Portland Area Environmental (S)
(+) Very faint amplicon in PCR (I) Isolates obtained from the following institutions: Centre for Environment, Fisheries and Aquaculture Science (CEF AS) Health Protection Agency (HPA), formerly Public Health Laboratory Service, (PHLS) Kyoto University Medical School (KUMS) Norwegian School of Veterinary Science (NSVS)
(2) PCR for V. parahaemo/yticus species specific toxR, and virulence tdh and trh genes (Table 3.2) Ure: Urease test. Data provided by S. Wagley, CEFAS Weymouth Laboratories.
(3) Identity of isolate 6316 as V. parahaemolyticus uncertain: Chapters 6 and 7.
Source of environmental isolates (where known) from following species: G: Crassostrea gigas (Pacific oyster), M: Pecten maximus (Great scallop), S: Maja squinado (Common spider crab), P: Cancer pagurus (edible crab}, E: Mytilus edu/is (blue mussel)
Table 3.2: Sequences of previously described PCR primers for toxR, and putative V .parallaemolyticus virulence genes tdlr and tr/r
Target gene PCR primers jS' to l') Reference F: GTCITCTGACGCAATCGITG Kim et al.. (1999) lo.TR R:ATACGAGTGGITGCTGTCATG F:CCACTACCACTCTCATATGC Tada et al., (1992) tdh R:GGTACTAAATGGCTGACATC F: GGCTCAAAATGGITAAGCG Tadaetal., (1992) I rh R:CATITCCGCTCTCATATGC
50 Table 3.3: Description of additional bacterial strains, DNA, and their genotypes, where known, that were used in this study
Environmental/Clinical/ PCR (2) Ure StraioiD Institute !I! Location S2ecies I DNA Rererence strain toxR tdlr trlr BL21 Amersham £.coli DNA LyophilisedControl DNA Cl a Amersham £.coli DNA LyophilisedControl DNA ST5 CEFAS Salmonella tyhimurium Environments] + PM5 CEFAS Proteus mirabilis Environmental PAl CEFAS P. aeruginosa Environmental + + KA5 CEFAS Klebsiella aerogenes Environmental + LD365 CEFAS Lis/one/la damsela Environmental 2046 CEFAS PHLS, UK V. vulnificus NCTC Reference 11327 CEFAS PHLS, UK V. j/uvia/is NCTC Reference VOS/005 CEFAS PHLS, UK V. mimicus NCTC Reference + VOS/006 CEFAS PHLS, UK V. lrollisae NCTC Reference VOS/007 CEFAS PHLS, UK V. algino/yticus NCTC Reference 8042 CEFAS PHLS, UK V. clrolerae NCTC Reference CMI04 UoP V. anguil/arum Environmental VCI Southampton V. clrolerae Environmental + VC2 Southampton V. c/rolerae Enviromnental + VC3 Southampton V. cholerae Environmental VC4 Southampton V. cho/erae Environmental VC5 Southampton V. cho/erae Environmental + Vib4(3) Southampton Unspeciated Environmental + 2341 Kyoto, Japan V. alginolyricrrs Clinical VA430 River Exe V. alginolyticus Environmental + VF430 V. jluvialis Environmental + VM499 DOM Fota V. mimicus Environmental + VCBADOI V. cholerae Environmental + VVII067 CEFAS V. vulnificus Environmental V. vulnificus Environmental + KIDBBVV CEFAS
(I) Isolates obtained from the following institutions: Centre for Environment, Fisheries and Aquaculture Science (CEF AS) Health Protection Agency, formerly Public Health Laboratory Service, (PI-lLS) Kyoto University Medical School (KUMS) University of Plymouth Culture Collection (UoP) Amersham Biosciences: Lyophilised DNA
(2) PCR for V. parahaemolyticus species specific toxR, and virulence tdh and trh genes (Table 3.2) Ure: Urease test. PCR data provided by S. Wagley, CEFAS Weymouth Laboratories.
(3) Isolate Vib4, together with Vib I - Vib I 0 (Table 3 .I) originally obtained as a collection of unspeciated isolates. Whereas Vibl- Vib3 and Vib5 to VibiO were later confirmed to be V. parahaemolyticus, Vib4 remained unspeciated
51 3.3 Bacterial genomic DNA extraction
3.3.1 CeU lysis and protein precipitation
Genomic DNA was isolated from V parahaemolyticus and other bacterial species using the 'Puregene D-6000 DNA isolation Kit' according to the manufacturer's instructions for
Gram-negative bacteria (Gentra Systems Inc., 'DNA Isolation Kit Instruction Manual', pages 54-57). A I Oml overnight culture of the organism was placed into a 15ml Falcon tube, centrifuged (1000 x g, 10min, 4°C) and the cell pellet resuspended in 3ml cell lysis solution with gentle pipetting. Samples were incubated at 80°C for 5 min to lyse cells.
Rnase A solution (15!11) was then added to the cell lysate before incubating the samples at
37°C for 1 hour. Following this, samples were cooled to room temperature (20°C) and 1 ml of protein precipitation solution added to the cell lysate and tubes vortexed vigorously at high speed for 20 seconds to mix the protein precipitation solution evenly with the cell lysate. The samples were kept on ice for 5 min before being centrifuged (2000 x g, 10 min,
4oC).
3.3.2 DNA precipitation and hydration
The supematant containing the DNA was poured into a clean 15 ml centrifuge tube containing a volume of isopropanol (3 ml) and mixed by inverting the tubes several times to precipitate the DNA followed by centrifugation (2000 x g, 3 min, RT). The supematant was discarded and the tubes left to drain over clean absorbent paper. Ethanol (70%, 3 ml) was used to wash the pellet of DNA before re-centrifugation (2000 x g, lmin, RT). Again, the supematant was discarded, tubes drained and allowed to air dry for an hour. An appropriate volume (50 - I 00 111) of DNA hydration solution (containing TE buffer) was added, depending on the amount of DNA visible upon precipitation, to dissolve the DNA.
Samples were centrifuged briefly, transferred to 1.5ml microfuge tubes and incubated for 1 hour at 65° to rehydrate the DNA. Samples were mixed and briefly centrifuged again before storage at -20°C. 52 3.4 DNA quantification
The quality and quantity of genomic DNA was established by agarose gel electrophoresis
(Section 3.8). Images of each agarose gel were captured with a gel documentation system
(UViTec, UK) and the DNA concentration was estimated for each isolate by comparison with a range of the following known amounts of 'A DNA standards: 0.552 11g, 0.276 11g,
0.138 11g and 0.069 11g using UV photo software (UViTec, UK). Genomic DNA was also quantified with a UV spectrophotomer (NanoDrop 100).
3.5 Primer design for PCR
Unless stated otherwise, primers for PCR reactions were designed using the aid of DNAsis for Windows software (Hitachi) and Amplify based on a Mac platform (Engels, 1993).
Primers consisting of24 bases were generated on the basis of a 50% G+C ratio and relative proximity to the desired target sequence. The presence of 2-3 guanine or cytosine bases at the 3' end of the primer, a lack of secondary structure and primer-dimer formation was assessed using AmplifY. Primers were synthesised by Sigma-Genosys.
3.5.1 Other PCR primers
Primers used for detection of V parahaemolyticus virulence genes, tdh, trh and toxR have been described previously, (Tada et al., 1992; Kim et al., 1999) (Table 3.2). Other primers used in the current work are described in detail in individual chapter and summarised in
Table 3.4.
53 Table 3.4: PCR primers used in this study. Specific details are provided in each chapter.
Target regions/ Designation Sequence (5' to 3') Reference Method ITS I VINTF TGGGGTGAAGTCGTAACAAGG Maeda et al., 2000 VINTR TCCTTCATCGCCTCTGACTG Chapter 6
ITS2 S2F CTAATTGCCCGTGAGGCTTA This st11dy, S2R GAGTCAGGTGGGTCCAAAAC Chapter 6
IONA spacer TJA GGGGGTTCGAATTCCCGCCGGCCCCA Welsh& TJB AGGTCGCGGGTTCGAATCC McClelland, 199Ia; TSA AGTCCGGTGCTCTAACCAACTGAG 1992 TSB AATGCTCTACCAACTGAACT Chapter 7 RAPD PI GGTGCGGGAA P2 GTTTCGCTCC Amersham P3 GTAGACCCGT Biosciences PS AACGCGCAAC Chapter 8
------TTSSI VPI662F TCACCGCTGGTATTATCGTTACGC VPI662R GAGATACAACGCGCGTCTTACTCT VPI680F GCCGAAGCGTATCATCATCAACTC VPI680R CACAGAGCTTACACCAAACGTACC VPI696F ATTGCTTCGGTATTGAGCCGTGTG Makino et al., 2003 VPI696R AACTCTGACCACTGCCCAATGTGA This st11dy Chapter 9 TTSS2 VPAI321F GGTTAGTGAATCCAACCAAACCGC VPAI321R TTGCCGTGCATGTCATACAACCAG VPAI339F GACACTCGCTGTTGTTCTCAGGTA VPAI339R GTAAGCGCGTGATGTTAGCTCTTC VPAI355F GGCATGTGGTGTCTATTTGACACG VPAI355R TACGACAACTGCAGGTAGTCCAAG
3.6 Standard Polymerase Chain Reaction (PCR)
All PCR reactions were prepared within a PCR cabinet and performed with either a MWG
Primus 96 Plus (MWG Biotech) or a Perkin Elmer (TCI) thermal cycler and based on a
standard reaction mixture as follows. Each reaction mixture contained 1 unit (U) of Taq
polymerase (Roche), reaction buffer containing 1.5mM MgCb (Roche), 24 pmol of each
primer, each deoxynucleoside triphosphate (dNTP) at a concentration of 0.2mM, between
2 - l 0 ng of bacterial DNA and the reaction volume was made up to 25 J.ll or 50 fll using
molecular biology grade water (Sigma). Negative controls were set up in exactly the same
way except the addition of DNA template was replaced with an equivalent volume of
water. Reactions that took place in the Perkin-Elmer thermal cycler were overlaid with
54 mineral oil, due to the absence of a heated lid. Various cycling conditions were used as described elsewhere. Specific details are provided in each individual chapter.
3. 7 PCR product purification
PCR products were purified using the 'High Pure PCR Product Purification Kit' (Roche
Applied Science) according to the manufacturer's instructions. Details of the individual solutions included in this kit can be found in Appendix I. After the PCR reaction was complete the final volume of the reaction mixture was adjusted with TE buffer to I 00 J.d.
This was transferred to a clean 1.5ml microcentrifuge tube containing 500 ~-tl Binding
Buffer (containing guanidine thiocyanate) and mixed thoroughly. The mixture was centrifuged (13,000 x g, 60s, RT), on a column (containing the manufacturer's filter) and flow through solution removed. The column filter tube containing PCR DNA was washed firstly with 500 ~-tl of Wash Buffer by centrifugation (13,000 x g, 60 s, RT) and secondly with 200 ~-tl of Wash Buffer (13,000 x g, 60s, RT). The filter tube was reconnected to a fresh 1.5ml microcentrifuge tube and an appropriate volume (50 ~-tl - I 00 ~-tO of Elution
Buffer added to the column before centrifugation (13,000 x g, 60s, RT) to elute the DNA.
This DNA was quantified using the procedures in sections 3.4 and 3.8.
3.8 Agarose Gel Electrophoresis
Depending upon the application, agarose gel electrophoresis was performed using 0.8%- 2
% agarose made with I x TBE buffer. Three sizes of agarose gel were used, 30 ml, 60 ml or 250 ml volume (equating to approximately 50 x 100 mm, 80 x 100 mm and 200 x 200 mm agarose slabs). Agarose gels of 250 ml were electrophoresed at 90V for 6 h, while 60 ml gels were run for 90 min at 120V. Gels of 30 ml were run at 75V for between 20 - 60 min depending on the expected fragment sizes. A suitable volume of 6x gel loading solution (Bromophenol Blue I Glycerol, Appendix 1) was added to all san1ples before
55 loading the gels. The sample volumes and concentrations varied according to each analysis and are stated in the Material and Methods section of each chapter. The sizes of the DNA bands that were resolved on agarose gels were estimated by comparison with either 100 bp,
I kb standards (0.3 J.tg) or f... ladders and these are detailed in Appendix I. Following electrophoresis, the DNA bands were stained by immersing the gel for 20 min in a solution containing 1J.tg of ethidium bromide (EtBr) per m! of TE buffer followed by 10 min de stain in distilled water to remove any unbound EtBr. The stained bands were visualised with UV light (309 nm) using a trans-illuminator and gels were recorded as digital TIFF images using a gel documentation system (UVI-Tech)
3.9 Restriction enzyme digests of genomic DNA
Various restriction enzymes purchased from Roche Applied Science were used for several applications. Usually, between 0.2 - 2.0 J,tg DNA was incubated with 10-20 units of enzyme, using the buffer supplied by the manufacturer. Digests were incubated for various periods up to overnight, at 37°C or 50°C, depending on the enzyme to ensure appropriate digestion.
3.10 Analysis of banding profiles
Banding patterns were analysed with the aid of BioNumerics (Applied Maths, Ghent,
Belgium). Generally, the bands from each lane were subjected to curve calculations and analysis to exclude background noise from PCR reactions. The tracks were then normalised using two external reference positions that consisted of 1OObp molecular weight markers on either side of the gel, and an internal reference; a common band that appeared in several lanes, as identified by visual inspection. Dendrograms calculated by the unweighted pair group method by arithmetic averaging (UPGMA) algorithm using a
Pearson product-moment correlation coefficient or Dice coefficient, as appropriate to
56 "' --J
1obtain the ·most reliable •Clustering patterns~ BioNumerics. po"sition_ toleran9e, s~t}in~s; generally. Wet(;! (h)%, optimisation,, ar position ·tolerance of 5% of get. andi i~oring· faint bands. Clusters ·were rdetermined ibased on visual inspection of deild!'ogr@i• ahalysi!),
Spedfic detail~ .a:re giv.en in ~acil chapter. Chapter 4
BIOCHEMICAL, GROWTH AND SEROLOGICAL VARIATION AMONGST VIBRIOPARAHAEMOLYTICUS
4.1 Introduction
Bacteria have traditionally been characterised using various phenotypic procedures that include biochemical properties, antimicrobial sensitivity tests, growth characteristics on selective media, and motility. Biochemical profiling (or biotyping) has widely been used to identify V parahaemolyticus at the species level. In particular, API 20E is a standardised system, used for identifying members of the Vibrionaeceae. Up to 21 differential characteristics have been described which show considerable variation amongst Vibrio spp., and in many cases resistance to antibiotics and di-isopropyl pteridine phosphate
(0/129) has been used to identify V parahaemolyticus from other members of this genus
(Baumann, & Schubert, 1984; Elliot et al., 1995).
Using the API 20E system Martinez-Utarza and colleagues (2006) showed that clinical isolates of V parahaemolyticus were correctly identified using a 0.85% NaCl diluent as specified by the manufacturer. However, environmental isolates were more correctly identified with a 2% NaCl diluent than the standard concentration. Clinical isolates grown on a 2% NaCl diluent showed non-specific profiles. Furthermore, differences in citrate
(CIT), amygdalin (AMY), gelatine (GEL) and arabinose (ARA) utilisation between the two groups of strains suggest wide biochemical diversity amongst V. parahaemolyticus.
Environmental isolates grown in vitro have been shown to exhibit characteristics different
from clinical and established laboratory strains, that include less growth, induction of a
viable but non-culturable (VBNC) state in certain conditions and excessive polysaccharide production. There have been few published reports into the study of growth characteristics
58 amongst clinical and environmental isolates. Considering that clinical and environmental isolates show enhanced biochemical fitness at different salinities of 0.85% and 2% respectively, little comparison has been made of these two heterogenic groups of isolates.
To assess phenotypic traits amongst members of clinical and environmental isolates of V parahaemolyticus, standard identification tests including Gram staining and oxidase tests were performed throughout the course of study. Of these, test results for cytochrome oxidase have been shown to vary amongst V parahaemolyticus (Twedt et al., 1969) and the absolute requirement for Na+ prompted some NaCl limiting growth tests. Early work also suggested a loss of viability of some cells stored in glycerol and hence viability of storage at -20°C was periodically monitored. These simple tests allowed monitoring of purity, viability, and growth characteristics on the most popular media used to cultivate V parahaemolyticus. Growth was also assessed on Wagatsurna agar (WA), used principally to detect haemolysis attributed to TDH production. Finally, serotype variation amongst clinical and environmental isolates was investigated in collaboration with CEF AS
Laboratories, Weymouth, UK.
59 4.2 Materials and Methods
4.2.1 Biochemical tests
Bacteria used in this study were Gram-stained according to standard procedures. The oxidase test was performed with oxidase strips (Fiuka). Reference strains were used as appropriate. A full set of API 20E biochemical characterisation was performed in association with colleagues at CEFAS, Weymouth, UK.
4.2.2 Growth and survival characteristics
Growth of V parahaemolyticus was assessed using several solid and broth media including
Marine Broth (Difco Laboratories), tryptic soy broth (TSB), Wagatsuma agar (WA) and thiosulphate citrate bile salts (TCBS) agar (Oxoid). Colony morphology was examined and recorded and growth tests carried out. Briefly, growth of 20 clinical and 20 environmental isolates was monitored in broth at 0.5% and 3.0% NaCI. A 0.5% NaCI is the standard concentration of NaCI in TSB (Oxoid) whereas a 3% NaCI is widely employed for the routine culturing of V parahaemolyticus. A 50 1-11 volume of an overnight culture grown in
TSB (3% NaCI) was used to inoculate 20 ml of fresh TSB (3% NaCI) in a shaking water bath (30°C) and growth rates were determined by monitoring absorbance at 600nm. The average growth rates of clinical and environmental isolates were determined and means tested for statistical significance with a two sample /-test (NCSS software).
Assessment of viability from cold storage was made by inoculating 50!-11 of 50% glycerol stock stored at -20°C into I Oml of TSB (3% NaCI), after I, 2 and 3 month intervals since initial preparation of stock culture. Cultures were grown overnight and 201-11 directly streak plated onto TCBS. Viability was monitored during the first three months of setting up glycerol stocks.
60 4.2.3 Thermostable Direct Haemolysin (TDH) phenotype- 'Kanagawa test'
Wagatsuma agar (W A) was prepared using fresh rabbit blood in heparin using the methods described in the FDA guidelines (Appendix 1). Briefly, fresh (within 24 h of drawing) rabbit blood was mixed with equal volume of physiological saline and centrifuged at 4000 x gat 4°C for 15 min. Supernatant was removed and the blood cells washed a further two times. After the third wash, supernatant was poured off and cells resuspended to original volume with fresh saline. Blood was used at 5% v/v with remaining agar mixture that had been steam sterilised and cooled to 50°C before use. Dried (30°C, 30 min) Wagatsuma agar plates were streak or spot plated using 5, 10 and 20 ~-tl of overnight grown cultures of
V. parahaemolyticus. To determine whether haemolysis was solely a result of bacterial growth on W A or heat stable haemolysin present in broth, the supernatant of the overnight grown cultures was heated to l00°C for 10 m in and spot plated in 5, 10 and 20 ~-tl W A overnight at 37°C.
4.2.4 Determination of haemolytic activity levels
Quantifying relative amounts of haemolytic activity between several strains was attempted using a modification to a method by Yoh et at, ( 1986). This method was used to potentially assess the uncertain P-haemolytic phenotypes on Wagatsuma agar. Briefly, one ml supernatant volumes of overnight cultures of test bacteria containing crude TDH were heated to l00°C for 10 min. A 50~-tl volume of each of the supernatants were transferred separately to a round bottomed microtitre plate followed by the addition of 50 ~-tl PBS containing 20 mM CaCh and rabbit erythrocytes at a final concentration of2% v/v. After 3 hours at 37°C the assay was performed using a microtitre plate reader (Optimax Tunable) set to As4o using no-lysis and full lysis as relative controls.
61 4.2.5 K and 0 antigenic variants (Serotyping)
V parahaemolyticus was serotyped for somatic (0) and capsular (K) antigens using a commercially available serotyping kit from Denka, Seiken Corp, Japan. using a polyvalent and monovalent anti-sera combination according to the scheme outlined in Table 4.1 and
4.2. This work was carried out in collaboration with colleagues at CEFAS Weymouth laboratory.
The K serotype was determined from the polyvalent antiserum I to IX (Table 4. 1) using some modifications from the manufacturer's guidelines. Test bacteria were grown overnight on TSA supplemented with 3% NaCI and fresh colonies were densely suspended in 500 Jll 3% saline. A double-welled slide, each well containing 20 Jll of the polyvalent serum was used; to one well 20 Jll of test bacteria was added and the other PBS (negative control). The mixture was stirred with the pipette tip and the slide tilted around to facilitate agglutination which was seen as the formation of a cloudy, granular precipitate. Only strong agglutination that occurred within one minute was considered a positive reaction.
Once a polyvalent antiserum had been ascertained, the procedure was repeated with monovalent sera until a positive agglutination was achieved. Where no agglutinations took place, a new type of K antigen was presumed and was indicated by K -untypable or KUT.
Overnight cultured test bacteria as described above were suspended in 3% saline at an approximate concentration of 10 mg/ml. Glycerine at 5% w/v was used in the mixture to prevent spontaneous agglutination of V parahaemolyticus. Cells were heat-lysed by autoclaving at 121 oc for 15 minutes and centrifuged for 10 mins. The supernatant was discarded and the sediment used for the agglutination test. When the K type had previously been identified, the antigenic scheme in Table 4.2 was used to test for serotype. If the 0 antigen corresponding to the K type as shown in Table 4.2 did not agglutinate, the remaining 0 serum was tested.
62 Table 4.1: Polyvalent and monovalent antisera containing agglutinins specific to K type. Each polyvalent antisera contained a mixture of 7 agglutinins. A positive agglutination in the polyvalent sera served to narrow down and test for the monovalent K antigen (Denka Seiken Corp.)
Pol~valent K antiserum K agglutinins (lresent 3 4 5 6 7 8 11 9 10 11 12 13 15 17 Ill 18 19 20 21 22 23 24 IV 25 26 28 29 30 31 32 V 33 34 36 37 38 39 40 VI 41 42 42 44 45 46 47 VII 48 49 50 51 52 53 54 VIII 55 56 57 58 59 60 61 IX 63 64 65 66 67 68 69 70 71
Table 4.2: Antigenic scheme of V. parallaemolyticus 0 groups and K serotypes. K types that have been shown to correspond to 0 antigens to facilitate serotyping (Denka Seiken Corp.)
0-antigen Grou(l K ~(le 1 25 26 32 38 41 56 58 64 69 2 J 28 3 4 5 6 7 29 30 31 33 37 43 45 48 54 57 58 59 65 4 4 8 9 10 11 12 13 34 42 49 53 55 63 67 68 15 17 30 47 60 61 68 6 18 46 7 19 52 8 20 21 22 39 70 9 23 44 10 19 24 52 66 71 11 36 40 50 51 61
1 Note: The antigenic scheme by Denka Seiken is different to the first established by Sakazaki et al., ( 1963) and later revised versions (Hugh & Feeley, 1972; Kaysner & DePaola, 2004)
63 4.3 Results
4.3.1 Biochemical tests
Results obtained from 63 isolates of V parahaemolyticus and various reference strains show Gram negative and cytochrome oxidase positive strains although 4 isolates (VPS,
VP349, VP377 and VP378) showed weak blue/purple colouration of oxidase test strips even after repeat testing. Whilst all V alginolyticus and V cholerae isolates were positive for acid production upon sucrose utilisation as identified on TCBS agar, all V parahaemolyticus isolates were negative (section 4.3.4). Isolates were confirmed as V parahaemolyticus by colleagues in CEF AS based on API galleries (BioMerieux) and if they met the following criteria: positive for oxidase, negative for Voges Proskauer and
Ortho-nitrophenyi-~-D-galactopyranoside (ONPG).
4.3.2 Growth rates of V. para haemolytic us by reduction of Na+
Doubling times generated from growth curves of a selection of clinical and environmental isolates are summarised in Figure 4.1. The average doubling time (n = 20, including two references strains) for clinical isolates at a NaCI concentration of 3.0% and 0.5% were 21.1 min and 27.2 min, respectively representing a 29% increase in doubling time at the lower
NaCI concentration. The average doubling time for environmental isolates at the same
NaCl concentrations were 22.2 min and 31.1 min respectively, the latter this time representing a 40% increase in doubling time upon decreasing NaCI concentration (Fig.
4.1 ). A comparison between clinical and environmental isolates showed that at a lower Na+ level, environmental isolates showed more variability in growth times with less growth, as represented by a mean doubling time significantly higher (p < 0.00 I) than that of clinical isolates (Fig. 4.1) in t-test analysis (Appendix II).
64 Doubling times (min) Clinical Environmental NaCI (0.5%): 27.25 ± 0.62 3 1.1 ± 1.14 NaCI (3 .0%): 2 1.10 ± 0.7 1 22.2± 0.81
E 40.0 c
--.. E c ._..E 30.0 c -- Cl) Q) E +=i 0> ·-c ..c :::J 20.0 0 0
10.0 0.5 3.0 NaCI (%)
Figure 4.1: Growth rate phenotypes of clinical (C) and environmental (E) isolates of V. parallaemolyticus at 0.5 and 3.0% NaCI. The average times and error values are indicated above. Each box-and-whisker plot summarises 20 isolates grown at 30°C. Maximum and minimum doubling times outside of the inter-quartile range are indicated by error bars. Average growth rates of clinical isolates were significantly higher (p < 0.001 in !-test) than the selected environmental isolates at 0.5 % NaCl (NCSS software).
65 4.3.3 Culturability of V. parahaemolyticus isolates from -20°C
Viability of 50% glycerol stock cultures was monitored for a total of 59 isolates and this showed 28 (47%) of V. parahaemolyticus cultures were not culturable after a month of storage at -20°C. After another month, a further 10 (17%) isolates could not be cultured using the overnight method adopted, followed by another 5 (8%) on the third month. Out of 59 cultures, 73% of the cultures were not culturable by the third month. This included
18 clinical and 25 environmental isolates representing 100% of total clinical and 61% of total environmental isolates respectively (Table 4.3).
4.3.4 General growth characteristics on selective and non-selective media
Isolates of V. parahaemolyticus and related organisms (section 3.2.3) were grown on four bacteriological media and growth levels scored and recorded (Appendix II). Most organisms grew on all media at varying intensities over a period of 18 h at 37°C. V. parahaemolyticus grew well on both MSA and 3% NaCl TSA with more pronounced growth and larger sized colonies on MSA. Colonies appeared circular on TSA whereas growth on MSA was less raised and generally opaque, although this varied amongst members of the genus. V. cholerae colonies were generally more translucent on MSA and
TSA. Growth on a more selective media, TCBS, usually used for isolation and confirmation purposes, generally showed less growth than MSA or TSA, yet distinct colonies of various green intensities were observed (Fig. 4.2). Unusually, amongst the V. parahaemolyticus cultures one clinical isolate, 6316, grew remarkably small colonies on all four types of media tested. V. cholerae isolates showed distinct yellow colonies upon acid production with sucrose fermentation as did V. alginolyticus isolates used in the study.
The production of acid alters the pH and is indicated by the dyes bromothymol blue and thymol blue. The colour reverted to green within a few days.
66 Table 4.3: Serotypes and selected phenotypes of V. paralraemolytlcus isolates.
Phenotypes Genotypes' Serotype' Culturability' Environmental/Clinical/ From -2o•c Beta- Urease tdh+ trh+ Isolate Reference strain haemollsls' OI:KI + 1902 NCIMB Reference 03:K4 +++ + VPI0884 NCTC Reference 02: K3 ND ND ND VPI0885 NCTC Reference 06:KUT 0 219 Clinical OJ:K6 ++ + 220 Clinical Oi:KUT 0 + (+) 221 Clinical 03:K6 0 ++ + 222 Clinical 03:K6 2 ++ + + 223 Clinical 04:K9 0 + + + (+) 224 Clinical 03:KUT 0 +++ + 225 Clinical 03:KUT 0 ++ + 226 Clinical 05:K68 +++ + 227 Clinical OII:KUT + (+) 228 Clinical 03:K37 2 1022/4 Environmental OJ:KUT 0 + I 024/5 Environmental 02:KUT 0 I 02115 Environmental Oii:KUT + 1022110 Environmental OI:K32 0 + 102311 Environmenlal ND 0 ND 102312 Environmental ND + 1023/3 Environmental OI:KI + 1103/9 Environmental 03:K6 0 + + 2053 Clinical 06:KI8 0 ++ + + + 2436 Clinical 03:K6 0 + 5421 Clinical 04:KI2 +++ + 6134 Clinical 03:K6 ++ + 6316 Clinical OI:K32 0 Vib I Environmental OI:K32 0 + Vib2 Environmental Oi:K32 2 + Vib3 Environmental OI:K32 3 Vib4 Environmental OI:K32 0 + Vib 5 Environmental ND 3 + Vib6 Environmental ND 3 + Vib 7 Environmental ND 3 Vib8 Environmental ND 3 + Vib9 Environmental ND 3 + + Vib 10 Environmental OI:K33 0 + VPI Environmental ND 0 + VP2 Environmental ND 0 + + VP3 Environmental OI:K33 3 VP4 Environmental Oi:K33 3 ND VP 5 Environmental 02:K28 3 + VP6 Environmental ND 0 + + + VP7 Environmental ND 0 + + + VP8 Environmental ND 3 + + + VP9 Environmental ND + + + VPIO Environmental
67 Table 4.3: cont.
Phenotypes Genotypes Serotype' Culturability1 Environmental/Clinical/ From -20°C Beta- Urease tdh+ /rh+ Isolate Reference strain haemoll:sis3 OI:KI 2 + + E154482 Clinical 03:K6 0 +++ + El55855 Clinical 03:K6 0 + + El68143 Clinical OI:K69 ND ND + + + VP F3305 Clinical ND 3 VP331 Environmental ND 3 + + VP345 Environmental OI:KUT VP346 Environmental ND 2 + + VP349 Environmental ND 0 + + VP350 Environmental OI:KUT 0 + VP361 Environmental 04:KUT 3 + VP377 Environmental 06:Kl8 3 VP378 Environmental 04:KUT 0 +++ + VP416 Environmental ND 0 + VP491 Environmental 08:KUT 0 + VP939 Environmental ND 3 + VPECI Environmental 08:KUT 3 VPSCI Environmental
NO: Not detennined
1 Serotyping perfonned in collaboration with colleagues at CEF AS. KUT: Untypeable K antigen
2 Culturability scores are: 0 and I for non-viability and viability on the first month. 2 indicates viability in the second month respectively. Viability on month 3 is recorded as 3
3 Levels of P-haemolysis was monitored at I 8 hours and scored as,+++ High,++ Medium, +Low/uncertain,- No visible haemolysis. P-haemolysis is resultant phenotype of gene tdh expression.
4 Potentially virulent genes: tdh and trh encoding for thermostable direct haemolysin and related haemolysin respectively.
(+) Weak amplification bands
68 A
Figure 4.1: Representative growth of V. ptut~hllemo/ydcus clinical oolate 1436 (plate A) and environmental isolate VP4 (plate B) on TCBS agar after 18 hours at 37°C. There were no obvious growth characteristics differentiating isolates of V. parahaemolyticus on TCBS agar.
69 c
Figure 4.3: Esopolysaccbaride production amongst two environmental isolates or V. partlhaemolyticus, VPECl (A), VPSCI (B) and swarming phenotype or V. a/glnolyticus, VA430 (C). Isolates were grown on Wagatsuma agar for 18 hours at 37°C.
70 Growth on Wagatsuma agar (WA), principally used to test for Kanagawa haemolysin
TDH, showed much variation as well as an unusual amount of exopolysaccharide
production for two environmental isolates, VPSC I and VP EC I, (Fig. 4.3). Of significance
is that six isolates showed poor growth including NClMB type strain 1902, clinical strain
6316, and four environmental isolates 1022/4, Vib4, VP350 and VP361. With the exception of these isolates, there appeared no obvious or distinct growth characteristics on selective or non-selective media.
4.3.5 Beta-haemolysis and clinical phenotypes
Varying degrees of haemolysis were demonstrated by isolates of V. parahaemolylicus and related species as indicated on Table 4.3 and in Fig. 4.4. In many cases, it was not possible to ascertain whether haemolysis had actually occurred as the zone around the colony was not characteristic of a clear P-haemolysis phenotype. Of the distinct haemolytic phenotype, all of the isolates except clinical strain 6316 were positive for tdh. In some cases of P
haemolysis, there appeared a secondary zone of haemolysis (Fig. 4.5) but without a green pigmentation (characteristic of a-haemolysis). This secondary zone appeared similar to the zone induced by heat treated ECPs (Fig. 4.6) and was only present in isolates that appeared to be the most haemolytic (Table 4.3).
Out of 27 tdh positive V. parahaemolyticus isolates, 12 (44%) showed distinct P
haemolysis, I 0 (37%) isolates showed some signs of haemolysis, but remained
unconfirmed, and a further five (19%) isolates did not exhibit any visible haemolytic
activity (Table 4.3). Of these five strains, one was from a clinical isolate (5421) and four
(Vib7, Vib9, VP350 and VPECl) were environmental. The use of a spectrophotometer to
assay haemolytic levels for uncertain P-haemolytic phenotypes proved to be inconclusive
with the method described (Fig. 4.7).
71 Figure 4.4: Variation in P-haemolytit levels of rabbit erythrocytes on Wagatsuma agar after 18 boun at 37°C. Kanagawa positive strains are isolates of V. parahaemo/yticus that express haemolysin TDH and exhibit various degrees ofp-haemolysis (plates B-D, scored+, ++,+++ respectively) from the control plate A(-). Plate E shows clinical tdh+ isolate with an uncertain haemolytic phenotype.
72 A
Primary zone: !}-haemolysis (Jmm) --+--
Secondary zone: incomplete haemolysis (7mm)
8
Figure 4.5: Zones of haemolysis in cllnkal and environmental isolates of Jl. parahaemo/ytlcus. A primary zone off}-haemolysis is followed by a larger zone of secondary incomplete (alpha) haemolysis in NCTC clinical reference strain VP I 0884 (A). Inoculum spot volumes of 5, I 0 and 25~tl of overnight TSB cultures of V. parahaemolyticus supplemented with 3% NaCI. Image B shows (}-haemolysis attributed to an environmental strain, VP416, isolated from Crassostrea gigas in comparison with a non haemolytic environmental isolate.
73 Figure 4.6: Zones of partial haemolysis attribated to TDH from thermostable ECP. Supematant of overnight V. parahaemo/yticus ,(isolate 6134) culture was heat-treated and spot inoculated with 5 (Q4), 10 (Q3) and 20 ~I (Q2). Arrows indicate diameter of partially haemolysed zone. First quadrant (Ql), shows growth and both primary and secondary zones of haemolysis from unheated supematant.
Haemolysin + 50% 25% 15% 5% 0% Isolate
6134
VP10884
VP2
PBS
Figure 4.7: Haemolytic activity of V. parahaemolyticMs. Lane 1, Control (equal volume H20 and RBC; Lanes 2-5; decreasing concentration ofhaemolysin (heat treated ECP); Lane 6, negative control (PBS & RBC). Varying haemolysin concentration had no apparent significant haemolytic effect on rabbit erythrocytes at 3 hours.
74 In addition to V. parahaemolyticus, there were three isolates of different species: V. cholerae (V CS), V. alginolyticus (V A430) and V. mimicus (VM499) that possessed gene
tdh but exhibited no haemolysis (data not shown). Data available on the presence of the trh
gene (Table 4.3) showed no association with haemolytic activity on Wagatsuma agar.
Based on the current data, there did not appear to be a link between urease activity and the
presence of the trh gene in V. parahaemolyticus, as both urease positive and negative
isolates were present with and without the gene trh (Table 4.3).
4.3.6 Variation in 0 and K serotypes amongst V. parallaemolyticus
Of the 44 isolates of V. parahaemolyticus serotyped, a total of 20 different O:K serotypes
were found amongst the 19 clinical, 22 environmental and three reference strains studied
(Table 4.3). Amongst these isolates, six had previously been characterised as 03:K6 and
were effectively used as controls and included three Norwegian and three Japanese
isolates. In total, there were eight different 0-types, 13 different K-types and 22 of the
environmental isolates were K untypeable (KUT). Serotype 01 appeared the most common
(36%) amongst all the isolates tested and was found mainly amongst the environmental
group. Similarly, serotype 03 was the most common 0-antigen in the clinical group (Table
4.3). Other than these two serotypes, and a larger proportion of environmental untypeable
K types, there were no apparent association of 0- or K- serotypes with clinical or
environmental isolates. Of significance, was the identification of two UK clinical isolates
(E155855 & E168143) that were of the same 03:K6 serotype associated with the pandemic
genotypes.
75 Table 4.4: Distribution of 0-antigens in Isolates of V. parallaemolyticus. Reference strains not included in the table. A total of eight out of 13 0-types were identified.
No of isolates 0- Serogroup Clinical Environmental 01 3 12 02 0 2 03 10 2 04 2 2 05 I 0 06 2 I 08 0 2 011 I I
4.4 Discussion
4.4.1 General growth and biochemical properties
Several methods were used to study the biochemical and physical growth properties of V parahaemolyticus with the aim of assessing any differences between clinical and environmental isolates. Despite the relatively small number of isolates that were available, the strains used in the study show variation in physical growth, haemolytic and serological
properties. Initial confirmatory biochemical test results were consistent with the typical characteristics of V parahaemolyticus (Gram negative, oxidase positive and motile).
Isolates received were routinely tested for purity and assessed using these methods.
Growth on several media routinely used in the isolation and characterisation of V parahaemolyticus were adequate for the purposes used, although some strains (e.g., 6316
and 6134) did not grow well and others (e.g., VPECl and VPSCl) showed atypical growth
on selected media. However, these characteristics were present in only a limited number of
isolates. Based on these growth observations, there were no obvious differences between
clinical and environmental isolates that would allow differentiation.
76 4.4.2 Growth of V. parahaemolyticus in reduced salinities
Salt tolerance has been used by several workers to differentiate Vibrio spp. (Beuchat, 1973;
Palasuntheram, 1981) and to assess properties within V parahaemolyticus (Twedt et al.,
1969). To the author's knowledge, there did not appear any published studies that compared growth rates between clinical and environmental isolates of V parahaemolyticus. As isolates were cultured throughout the course of this present study with 3% NaCl, it was of interest to assess if a reduction in NaCl would affect growth of clinical and environmental isolates. On the basis of this present limited study, it was shown that environmental isolates grew at more variable growth rates at both 0.5 and 3.0% NaCl than clinical isolates. However, at the reduced Na+ level, a significantly larger proportion of environmental isolates grew slower than clinical isolates and may indicate environmental isolates are less suited for growth at reduced salinities. It would thus appear that clinical and environmental isolates of V parahaemolyticus do not differentially exploit excess Na + (e.g. 3% NaCl) for growth when it is freely available. For the purposes of experiments that involved culturing of V parahaemolyticus isolates (e.g. tests for
Kanagawa haemolysin), it would seem culturing at the higher NaCl concentration is more appropriate as this did not favour the growth of clinical or environmental isolates.
Recent work by Martinez-Urtaza and colleagues described how clinical and environmental isolates were better identified according to the API 20E profile using two different NaCl concentrations (Martinez-Urtaza et al., 2006). In addition to environmental and clinical isolates showing significant differences in biochemical tests, clinical isolates were correctly identified according to the API 20E profile using 0.85% NaCl diluent, but they presented nonspecific profiles with 2% NaCl diluent. In contrast, the use of a 2% NaCl diluent facilitated correct identification of the environmental isolates. The findings from the present study supports the work by Martinez-Urtaza and colleagues by providing evidence of better growth of clinical isolates at a lower NaCl concentration and thus a
77 possible explanation of more accurate API 20E profiles. Further analysis of growth rates at different NaCI concentrations would be necessary to validate these findings.
4.4.3 Culturability of cells from cold storage at -20°C
Cold temperatures have been known to induce a viable but non-culturable VBNC state in
V. parahaemolyticus and several other species (Baffone et al., 2003). Stock glycerol cultures were prepared using a standard method (Sambrook et al., 1989) and cells monitored for survival over three months. Of some significance is that within the first month, 50% of clinical and 41% of clinical strains became non-culturable. A somewhat similar finding was made by Jiang and Chai ( 1996) who showed that an environmental,
Kanagawa negative strain took longer to achieve non-culturability than a Kanagawa positive one. Upon observing decline of culturable cells, it was decided to keep some remaining samples of stock culture at -20°C and monitored for culturability or viability to assess any difference amongst the V. parahaemolyticus strains. Although tests specifically for the VBNC state of cells were not carried out here, work by Johnston and Brown (2002) showed V. parahaemolyticus was not inactivated at -20°C and therefore it is speculated that the V. parahaemolyticus strains may have entered a VBNC state. Others have shown various degrees of resuscitation of VBNC cells by treatment with catalase or sodium pyruvate (Mizunoe et al., 2000) and that resuscitation was dependent on the age of the culture and strain dependent (Wong & Wang, 2004) although this approach was not taken here.
4.4.4 Beta-haemolysis attributed to thermostable direct haemolysin (TDH)
Haemolysis on Wagatsuma agar (WA) was demonstrated by various strains of V. parahaemolyticus. This ~-haemolysis phenotype of TDH is widely described as a marker for virulent strains of V. parahaemolyticus although haemolysis levels varied considerably amongst tdh gene positive strains tested. Of immediate interest is an environmental isolate,
78 VP416, recovered from oyster (C. gigas), off the Dorset coast which demonstrated remarkably strong ~-haemolytic activity (Fig 3.3), appearing more haemolytic than clinically acquired isolates expressing the haemolysin TDH. As clinical isolates must originate from environmental samples, it questions whether V parahaemolylicus TDH is expressed in C. gigas. To date, there appears little information on oyster physiology and immune status upon levels of potentially pathogenic V parahaemolyticus.
Of the clinical isolates used in the study, approximately 55% exhibited ~-haemolysis; a significantly larger proportion than that previously described amongst clinical isolates
(Kelly & Stroh 1989). However, four tdh+ clinical strains (PCR data: Table 3.I) showed an uncertain ~-haemolytic phenotype and such findings have also been described by others
(Nair et al., I985). Therefore, in an attempt to quantify levels of haemolysis, three strains that gave distinct phenotypes (Fig. 4.4) were assessed to see whether it was possible to assay for haemolytic activity using a simple spot plating and microtitre assay method.
Previous reports have shown that TDH is able to retain 70% of its haemolytic properties after treatment at I 00°C for I 0 m in (Nishibuchi & Kaper, 1995) and thus it was thought it may be possible to obtain relative haemolysis levels that may be used to characterise those isolates that visually showed uncertain haemolytics levels. However, both the spot plating method and the microtitre method were unable to generate discernable figures of haemolytic levels. It would appear that without concentrating TDH, a significant haemolytic reaction would not be able to be determined with a microtitre assay. Spot plating a higher volume of ECP containing TDH on Wagatsuma agar did not show any significant haemolysis either. ELISA methods have been used to quantify TDH levels
(Honda et al., I985; Yoh et al., 1995) but the difficulties and costs associated in obtaining purified TDH products mean it is not a feasbilbe option at present for rapid quantification ofTDH amongst isolates.
79 The large variation m TDH expressed amongst tdh+ isolates (Kelly & Stroh 1989;
Nishibuchi & Kaper, 1990) adds to the uncertainty of the significance of finding tdh+ in environmental isolates. For example, several clinical isolates did not exhibit clear haemolysis, one being distinctly negative (5421 ), and a further two isolates that were haemolytic, showed poor growth on WA. Additionally, false positive haemolytic reactions can occur due to pH changes around the colonies, fragility of erythrocytes or haemolysins other than TDH (Nair et al., 1985). Such inconsistencies in the use of W A suggest it may not be a reliable method for screening environmental isolates of V parahaemolyticus for
TDH. Despite the present findings, some workers have employed the Kanagawa test for confirmatory purposes. Cabrera-Garcia et al., (2004) identified four tdh+ isolates amongst
46 environmental samples of V parahaemolyticus and all four isolates were shown to be
Kanagawa positive. However, Kanagawa tests are not routinely performed and the use of
W A may be hindered by the availability of fresh human or rabbit blood as recommended in the FDA guidelines (Kaysner & Depaola, 2004).
In order to study actual expression of tdh genes, the use of reverse transcriptase RT-PCR can address the issue of estimating TDH production by monitoring mRNA levels.
However, putative virulence genes tdh and trh have been reported not to be expressed as well as housekeeping genes under certain conditions (Coutard et al., 2005; 2007). Others have used a transcription-reverse transcription concerted (TRC) method to quantify tdh and trh mRNA transcripts (Masuda et al., 2004; Nakaguchi et al., 2004), but it appears few workers who employ tdh and trh gene detection protocols in V parahaemolyticus actually studied the expression of these genes. Future developments on these methods and gene expression studies are necessary, and may provide more reliable indications as to whether the tdh or trh genes in environmental isolates are active and the degrees to which they are expressed.
80 4.4.5 Urease production and TDH-related haemolysin (TRH)
The correlation of urease production with pathogenic isolates and more importantly lrh+ has been reported in several studies (Suthienkul et al., 1995; Okuda et al., 1997b; Iida et al., 1997). Data obtained from previous studies on the strains tested show only five strains that are trh+ and urease positive. The trh primers used were based on consensus regions of
I rh I and trh2 and are therefore reliable indicators of /rh (Tada et al., 1992). A further eight strains, seven of which were environmental, were shown to be positive for trh yet did not show urease activity. Similarly, three isolates that expressed the urease phenotype lacked the trh genes. Consequently, there is insufficient evidence in this study to support published data that urease production is a reliable indicator for pathogenic strains.
4.4.6 Serotyping
Serotyping is important in distinguishing the small number of strains that actually cause disease in a large number of species including E. coli, V cholerae and Salmonella (reviews by Stenutz, et al., 2006; Nair, 1996; Kingsley & Baumler, 2000). The finding of a heterogeneous population of 0 and K antigens whilst serotyping V parahaemolyticus demonstrates the wide diversity of isolates present in the study. A large number of isolates that were not K-typeable (KUT) accounted equally for approximately a fifth of the clinical and environmental isolates. The total number of different serovars ( 13 and 16 in clinical and environmental isolates respectively) have only two common serotypes 01 :K1 and
06:K18, and other than the 03:K6 isolates amongst clinical isolates, there appear no correlation amongst the two groups of isolates. Additional environmental V parahaemolyticus isolates that were serotyped do not show a predominant serotype amongst UK acquired environmental strains (Pers. Comm. S. Wagley, 2007).
The difficulty of being able to differentiate V parahaemolyticus by this method is hampered because the manufacturers of the test sera (Denka Seiken, Japan) have stated that
81 the determination of K antigens can be affected by the presence of certain 0 antigens and similarly the agglutination ofO type can be inhibited by K antigens.
Of all methods studied, the growth levels in different NaCI concentrations have been of most interest. The difficulty in assessing haemolytic levels of tdh expression need to be addressed if the haemolytic phenotype is to be used successfully. Although biochemical and physical growth characteristics have been explored, the relationships and ability to differentiate between clinical and environmental isolates using these approaches can be long and often difficult to interpret. Larger sample numbers and greater number of methods are required to help assess feasibility of biochemical, growth and serological characterisation. The next few chapters explore the use of molecular methods in characterising V parahaemolyticus.
82 Chapter 5
USE OF PULSED-FIELD GEL ELECTROPHORESIS TO SHOW CLONAL RELATIONSHIPS AMONGST VIBRIO PARAHAEMOLYTICUS ISOLATES
5.1 Introduction
Pulsed-field gel electrophoresis (PFGE) is based on the observation that during continuous field electrophoresis, DNA above 30-50 kb migrates with the same mobility regardless of size. If the DNA is forced to change direction during electrophoresis, different sized fragments begin to separate from each other. PFGE thus allows separation of larger pieces of DNA than conventional agarose gel electrophoresis, and in recent years it has been used extensively to sub-type bacterial species, and is widely regarded as the method of choice for typing nosocomial and community acquired pathogens (Tenover et al., 1995). PFGE has been shown to be more discriminating than other molecular methods of typing for many species of bacteria and with V parahaemolyticus it has been reported to have the highest discrimination index of a single typing method when compared with several other methods including ERIC-PCR, and an RFLP method based on the Fla locus (Marshall et al., 1999).
Intact genomic DNA is digested with an infrequently cutting restriction enzyme and the
resulting DNA fragments are separated in agarose gels using highly specialised equipment
to switch the direction of the current for different time periods. This allows the separation
of DNA fragments sized between 50kb and l 0 M b. In practice, whole bacterial cells are
suspended in molten agarose in small plug moulds. The solidified plug containing a known
number of bacterial cells is then treated to lyse the cells in situ and remove interfering
constituent proteins. The resultant linearised DNA fragments are then separated by PFGE.
83 Differences in the number and sizes of fragments after separation are due to variation in the positions and frequency of restriction sites on the genome (Tenover et al., 1995).
PFGE has also been used for macro restriction analysis to study entire bacterial genomes.
The ability to separate large DNA fragments has enabled the identification of multiple and linear chromosomes as well as genomic reorganisation caused by duplication, mutation, insertion or deletion. Indeed, this technique has been used to unravel the complex genomic structure of several bacterial genomes and produce detailed physical maps, including that of Vibrio parahaemolyticus (Fonstein & Haselkom, 1995; Kolsto, 1997; Tagomori et al.,
2002; Y amaichi et al., 1999;)
The use ofPFGE for subtyping V parahaemolyticus was first reported in 1996 by Wong et al., ( 1996) and since then, several workers have developed various protocols for the application of PFGE for genomic typing of V parahaemolyticus based on the initial procedures of Cantor et al., (Bag et al., 1999; Chowdhury et al., 2000a; 2004a; Lu et al.,
2000; Cantor et al., 1988; Wong et al., 1996). However, variations in the use of restriction enzymes (Notl, Sfil and Apal), pulse switch times, equipment and other parameters mean that PFGE patterns produced by different laboratories are not readily comparable.
Moreover, in many cases variation in the degree of DNA degradation has hampered the use of PFGE sub-typing methods (Marshall et al., 1999).
The current study was initially based on the procedures described in the BioRad PFGE Kit
(2001) using the electrophoresis and pulse settings of Wong et al., (2000) for V parahaemolyticus. Upon the observation of large degrees of DNA degradation, several modifications were made. This work was carried out to:
84 (i) compare V parahaemo/yticus UK isolates with each other and with isolates
obtained from other countries,
(ii) to assess existing methods for PFGE of V parahaemo/yticus for the ability to
differentiate clinical (virulent) and environmental (mostly avirulent) V
parahaemolyticus, and
(iii) to establish a benchmark method using PFGE (described as the "gold standard"
by Streulens, 1998) to which other molecular typing systems could be
compared.
85 5.2 Materials and Methods
5.2.1 Preparation of V. paralzaemolyticus plugs
V parahaemolyticus cells grown in Marine Broth were harvested at OD600 of between 0.8
7 8 and 1.0 corresponding to a cell density of between l 0 - 10 cells/m I. Where possible, all
reagents used were of PFGE quality from BioRad. Details of the buffers used in this section are in outlined in Appendix I. V parahaemolyticus cells were embedded into agarose plugs using the method as described in the "CHEF Genomic DNA Plug Kits
Instruction Manual" (CHEF Genomic DNA Pulsed Field Electrophoresis Systems -
Instruction Manual and Applications guide, BioRad). V parahaemolyticus cell density was established using a haemocytometer; contrast obtained by staining cells with Gram Crystal
Violet. A volume corresponding to 5 x 107 bacteria was centrifuged (13, 000 x g, for 3 min, RT) and the supematant discarded. Cells were washed twice with 500J.ll of cell
suspension buffer (CSB), and then re-suspended in 250J.ll of CSB and equilibrated to 50°C.
Alternatively, 1.25ml from an overnight culture grown in TSB (3% NaCI) was harvested
by centrifugation (without cell counting) and resuspended in 250J.ll of buffer containing
lOmM Tris, lOOmM EDTA and lmM NaCI (pH 8.0) (Wong et al., 1996). Next a 250J.ll volume of molten solution of2% (w/v) 'Pulsed-field CleanCut agarose' (50°C) was added and mixed gently. The mixture was transferred into 1OOJ.ll disposable plug moulds and left
to solidify at 4°C for 15 mins. Plugs were also prepared with the same method for selected
isolates of V cholerae.
5.2.2 Bacterial cell lysis and de-proteination
Solidified plugs were removed from plug moulds and transferred to sterile plastic
disposable bottles containing 50J.ll Lysozyme stock and 1.25ml of Lysozyme Buffer for
each 500J.ll of agarose plugs. Plugs were incubated for 2 hours at 37°C, after which the
lysozyme solution was discarded and the plugs were washed briefly with 1Oml sterile
86 water. Plugs were then incubated for 24 h with a proteolytic solution containing 1.25ml of
Proteinase K Reaction Buffer and 501J.I Proteinase K stock per 5001J.I agarose plugs at 50°C without agitation. Afterwards, proteinase K solution was removed and the plugs washed four times in lx Wash Buffer, 1 hour each wash at room temperature with gentle agitation.
For each plug 1ml of 1x Wash Buffer was used. During the third wash a final concentration of 1mM PMSF was used to inactivate residual Proteinase K. If undigested
DNA plugs were required, then no further procedures were performed and the plugs were stored at 4°C.
5.2.3 Restriction endonuclease digestion of agarose plugs
Prior to restriction digestion, plugs were cut (4 x 9 x 1.2 mm approx) and washed for I hour in 1ml O.lx Wash Buffer per 100 111 plug in a 1.5 ml microfuge tube. This was followed by a second wash (30 m in) using 300 Ill of 0.1 x Wash Buffer. Wash Buffer solution was removed and I 001J.I of the appropriate restriction enzyme buffer added and left at 4°C for 30 min. Restriction enzyme buffer was removed and replaced with fresh restriction enzyme buffer and 20U of the enzyme. The contents were gently mixed and incubated overnight at the appropriate temperature.
5.2.4 Agarose gels and running conditions
Agarose gels were cast using pulsed-field certified agarose in BioRad's gel casting kit and run at 14°C using a CHEF DR-II machine, incorporating a cooling system and buffer pump. Agarose gel concentration, switch times, running conditions and other parameters were initially based on work by Wong et al., (2000) and modified as appropriate and are included in the legends of each figure. Following electrophoresis, all PFGE gels were immersed for 30 min in 250 ml of ethidium bromide solution at 1 jlg/ml to facilitate visualisation of DNA under UV illumination.
87 5.2.5 Standards and control organisms
A commercially available A standard was used (Bio-Rad Laboratories) to size fragments throughout the experiments. These 48.5 kb concatemers were preparations of A cl85sam7 ranging between approximately 0.05 and 1 mb in size. Saccharomyces cerevisiae and
Schizosaccharomyces pombe standards supplied by BioRad were used as controls and size markers. V parahaemolyticus NCIMB reference strain 1902 was used to optimise conditions. Internal controls (DNA of same isolates) were placed at various gel lanes to assist data analysis.
5.2.6 Data analysis and dendrogram construction
DNA bands were initially compared by visual inspection and interpreted according to
Tenover et al., ( 1995). Based on the number of DNA fragments that exhibited different pulse patterns, the isolates were considered to be indistinguishable (identical pulse patterns), closely related (two to three fragments with different patterns), or possibly related (four to six fragments with different patterns). If more than six DNA fragments exhibited a different migration pattern, the isolates were considered to be unrelated. A dendrogram showing relationships amongst the clinical and environmental isolates of V parahaemolyticus that appeared typeable were generated using BioNumerics Software with modules for Fingerprints and Cluster Analysis (version 5.1, Applied Maths, Belgium) as described in section 3.1 0, with data normalised using internal (same isolates placed at different gel lanes) and external controls (A ladders). Dendrogram was calculated by the unweighted pair group method by arithmetic averaging (UPGMA) algorithm using the
Dice coefficient. BioNumerics position tolerance settings were 5% optimisation, a position tolerance of 2% with all visible bands scored. Pulse Types (PTs) were assigned at a cut off point of 88% similarity value that was based on visual inspection of identical and closely related isolates as described above.
88 5.3 Results
5.3.1 PFGE of V. parallaemolyticus
PFGE was perfonned on a total of 61 isolates of V parahaemo/yticus and 11 related species using restriction enzymes Noti and to a lesser extent with Sjil. The data reflect the wide genetic heterogeneity amongst the isolates used in the study (Figs. 5.1 to 5.8) and many isolates were untypeable due to excessive amount of DNA degradation. In total, 13 of the environmental isolates and one clinical isolate were not typeable using the methods described. Using the BioRad method, visible PFGE bands were present at cell densities greater than 108 cells and various degrees of DNA degradation, including that of controls,
S. cerevisiae and S. pombe visible throughout the gels (Figs. 5.1 & 5.2).
5.3.2 PFGE of V. cllolerae
V cho/erae plugs prepared using the same methods, were digested with both Sjil and Notl.
Using the same electrophoretic parameters as those used with V. parahaemolyticus. V cholerae showed distinct banding profiles (Fig. 5.2). Only two strains of V cholerae, VC1 and VC BAD01 could be typed and subsequently used for comparative purposes. Noti digested plugs of V. cholerae (VC4 & VC1) showed up to five common bands in a total of approximately 15 bands with V parahaemolyticus (Figs. 5.2 & 5.8).
5.3.3 Sfil digested restriction profiles of V. para haemolyticus
Up to 14 bands were visible with NCIMB type strain 1902 that ranged in size between 50 kb and 430 kb. A larger, unresolved, faint band (>600kb) and several smaller bands
(<48.5kb), although visible, were very faint amongst other widely distributed sizes of restriction fragments (Figs. 5.2 & 5.3). The appearance of these faint bands with Sjii digestion in comparison with Notl restriction bands initially favoured the use of a Notl restriction profiling and therefore Notl was used on the entire collection of V
89 parahaemolyticus isolates available. Isolates that were not typeable usmg Noli also appeared untypeable using Sjil.
5.3.4 Notl digested restriction patterns of V. parahaemolyticus
Noli digested PFGE profiles showed between 11 - 18 bands that ranged in size between 30 kb to greater than 600 kb and was therefore selected for use on the entire collection of isolates available. Of the 47 typeable isolates, i.e. those that showed visible bands even within DNA smearing, 19 were of clinical origin (including two reference strains) and 28 were environmentally acquired strains. These represent 95% and 68% typeability of clinical and environmental isolates respectively using the methods described.
Environmental isolates used in the study showed a wide variation in banding profiles and included many unique restriction patterns. For example, at least seven environmental isolates were single members of their own Pulse Types (PT) and could not be grouped based on common bands (Figs. 5.5 - 5.1 0). The environmental isolates (n=28) were identified as 17 unique PFGE Types found distributed throughout dendrogram (Fig. 5.11).
Seven of these groups consisted of between two to four isolates per Pulse Type that was
usually from the same source of origin. For example, Vibl, Vib2 and Vib3, all of serotype
OI:K32 from Southampton appeared to be clonally related as did Vib7 to VibiO (Figs. 5.8
& 5.9). Similarly, the homology ofVPEC1 and VPSCI (Fig. 5. 11) could be attributed to the same source of isolation (Portland, Dorset). However, between all the environmental groups, although several bands were common, there was no band that was universally
present amongst all the strains, thus reflecting a true heterogeneous population of V parahaemolylicus in this study (Fig. 5.11 ).
90 PFGE profiles generated by clinical strains of Pulse Types 18 to 22 exhibited the most similar banding patterns with at least 8 common bands out of approximately 1S bands
(>80% similarity) (Figs. S.4- S.8 & S.10). This group of 10 isolates (Fig. S.1l) fell into one hierarchical cluster and included two strains obtained from the UK that had almost identical PFGE profiles to strains from Japan and Norway. These clinical isolates were of serovars 03:K6, 03:KUT and OS:K68 and were all positive for tdh and did not contain trh genes. Clinical isolates that were in Pulse Types IS and 22, however, contained trh only strains. Additionally, an environmentally acquired tdh positive strain (VP416) showed a near identical match to a clinical Japanese strain, 6134. Norwegian isolate 221 showed the least number of bands amongst the clinical isolates with 11 fragments that were widely dispersed in fragment sizes and was thus used as a size marker for dendrogram analysis.
5.3.5 Dendrogram analysis of Notl restriction PFGE patterns
Cluster analysis based on UPGMA and a cut off point at 88% banding similarity, as generated using BioNumerics, the isolates used in the study were grouped into a total of 30
Not I restriction PTs (Fig. S.I1 ). Of these types, only one type, (PT 09) had a mixture of
both an environmental and clinical strain in the group at 88% similarity. Isolate VP416, a
UK shellfish acquired, Kanagawa positive strain matched the profile of a Japanese clinically sourced Kanagawa positive strain. Other environmental isolates were also grouped with clinical isolates but appeared to be more distant (<88% similarity) on the dendrogram. Seven clinical isolates, (PTs 02, 06, 1S, 19, 24, 2S & 27) consisted of unique
PFGE types and included the UK reference strain, 1902 (Fig. S.11). Other than these
strains, the clinical isolates used in the study generally clustered into one group of the dendrogram (PTs 18 - 22) with distinct clonality in PT 18. The environmental isolates exhibiting clonality were clustered in PTs 03 (Vibl - Vib3), 04 (Vib7-Vib10), OS (VPEC1
& VPSC 1), 10 (VP1, VP2 & VibS), 12 (VP4 & VPS) and 30 (VP6 & VP7) although
91 • .. ·
I, isolates in IPUlse Types 28 -~ JO app~:aredi more <;IoseJy rel[lted than thllt indic~t_e_dl by th-e dendrograrn. (Fig s. U) .. I
92 +PMSF -PMSF r.--_ __,.A...._----..,
f-291 kb
f-194 kb
f-97 kb
f-48.5 kb
Figure 5.1: PFGE proftles of restriction Notldigested and undigested V. parahaemolyticus of various cell densities and treatments using BioRad method. Lanes 1- 12 are NCIMB type strain 1902 treated with (Lanes l-6) and without (Lanes 7-12) PMSF. Lanes I, 2, 7 & 8 correspond to 108 cells; lanes 3, 4, 9 & 10 correspond to 107 cells; lanes 5, 6, 11 & 12 correspond to 106 cells; lane 13: S. cerevisiae; lane 14: S. pombe; lanes 15 & 16: undigested 1902; lanes 17 & 18: environmental isolate Vib 3; M: A. ladders size indicated by arrows. Gel prepared with l% agarose, 0.5xTBE and run under the following conditions: 20 min pulse, at 65 volts (1 .9v/cm), for 18 h.
93 VPI VC4 VP1 VC4 ABC ABCABCAB C BB BB M 1 2 3 4 5 6 7 8 9 lO 11 12 13 14 15 16 17 18 M
~29 1 kb
~194 kb
~97kb
~48.5 kb
N s N s N S N S
Figure 5.2: PFGE profiles of Notlaod Sfil digested environmental isolate V. parahaemolyticus VPl and V. cholerae isolate VC4 with cell densities of 10 9(A), 5x 108 (B) and 108 cells (C). Lanes 1 -7, 14 & 15: isolate VP1 ; lanes 8 -13, 17 & 18: isolate VC4. Not1 digested are indicated by Nand Sfil digested indicated by S above. Lanes 14 to 18 represent plugs prepared as replicates. M: A. ladders, sizes indicated by arrows. Gel prepared with 1% agarosc, 0.5xTBE and run under the following conditions: 20 rnin pulse at 65 volts (1.9v/cm), for 20 h.
94 M 1 2 3 4 5 6 7 8 9 10 1112 13 14 15 16 17 18 M
~388 kb
~291 kb
~194 kb
~97 kb
~48 .5 kb
Cla I Notl Sfil Not I
Figure 5.3: PFGE profiles of Clal, Notl and Sftl restriction digested V. parahaemolyticus NCIMB type strain1902 with and without PMSF treatment and modification of method based on Wong et al. (2000). Lanes l - 6: Clal digests with (lanes 3 & 4) and without (lanes l & 2) PMSF. Lanes 5 & 6 correspond to doubling restriction enzyme concentration (30u). Lanes 7 - 10: Notl digests with (lanes 9 & 10) and without (lanes 7 & 8) PMSF treatment. Lanes 11 - 14: S.ftl digests with (lanes 13 & 14) and without (lanes 11 & 12) PMSF. S. cerevisiae digested with Not I treated with (lane 16) and without (lane 17) PMSF. Lane 18: undigested VP1902. M: A. ladders indicated by arrows. Gel prepared with 1% agarose, 0.5xTBE and run under the following conditions: 20 min pulse, at 65 volts (1.9v/cm), for 18 h.
95 Env Clin Environmental ,;..-.. ,;..-.. M I 2 3 4 5 6 7 8 9 10 M
f-388 kb
f-291 kb
f-194 kb
f-97 kb
f-48.5 kb
Figure 5.4: Comparison of PFGE proft.les of Notl restriction digested V. parahaemolyticus clinical and environmental isolates. Lanes 1-6 correspond to the following isolates: VP2a, VP2b, El54482a, El54482b, Vib 6a and Vib6b respectively. Letters a & b indicate OD600 of0.4 and 0.5 respectively after overnight growth in TSB. Lanes 8 - I 0 correspond to isolates VP2, VP7 and Vib 6 respectively, prepared with plugs that were 7 days old. M: /...ladders indicated by arrows. Gel prepared with I% agarose, 0 .5xTBE and run under the following conditions: 20 min pulse, at 65 volts (1 .9v/cm), for 20 h.
96 Environmental Clinical ~
~485kb
~388kb
~29lkb
~194kb
~97kb
~48.5kb
Figure 5.5: Comparison of PFGE proftles of Notl restriction digested V. parahaemolyticus clinical and environmental isolates of UK origin. Lanes 1-l 0 correspond to the following isolates: VP3a, VP3b, VP6a, VP6b, VP7a, VP7b, El55855a, El68143a, El68143b and El55855b respectively. Letters a and b indicate OD60o of0.4 and 0.5 respectively after overnight growth in TSB. M : /. ladders indicated by arrows. Gel prepared with 1% agarose, 0.5xTBE and run under the following conditions: 20 min pulse, at 65 volts (1.9v/cm), for 20 h
97 Environmental Clinical
~485 kb
f-388 kb
~291 kb
f-194 kb
~97kb
f-48.5 kb
Figure 5.6: Comparison of PFGE profiles of Not1 restriction digested V. parahaemolyticus clinical and environmental isolates from UK, Spain and Japan. Lanes 1-18 correspond to the following isolate designations: VP1 , VP2, VP3, VP4, VP7b, Vib3, 1022/4, 2053x, 2053, 2436, 5421, 6134, 63 16, E154482, El55855a, El68143a, El68143b and VP10884 respectively. Letters a and b (after isolate numbers) indicate 00600 of0.4 and 0.5 respectively after overnight growth in TSB. M: A. ladders with selected sizes indicated by arrows. Gel prepared with 1% agarose, 0.5xTBE and run under the following conditions: 20 min pulse, at 65 volts (1 .9v/cm), for 20 h
98 Clinical Environmental Clinical ,A, M 1 2 3 4 5 6 7 8 9 10 ll 12 13 14 15 16 17 18 M
f-485 kb
f-388 kb
f-291 kb
f-194 kb
f-97 kb
f-48.5kb
Figure 5. 7: Comparison of PFGE proftles of Notl restriction digested V. parahaemolyticus clinical and environmental isolates from UK and Japan. Lanes 1-17 correspond to the following isolate designations: Vibl, VP10884, 6316, VPl, VP2, VP3, VP4, VP7, Vib3, 1022/4, 2053, 2436, 5421,6134, 6316 and El54482 respectively. Lane 18: NCTC type strain VP10884. M: I.. ladders with selected sizes indicated by arrows. Gel prepared with l% agarose, 0.5xTBE and run under the following conditions: 20 rnin pulse, at 65 volts (1.9v/cm), for 20 h
99 Environmental Clinical ~ M 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 M
Figure 5.8: PFGE profiles of Notl digested UK environmental and UK clinical strains of V. parahaemolyticus. Lanes 1-18 correspond to the following isolate designations: VPl, VP2, VP3, VP4, VP6, VP7, VP8, VP9, VPlO, Vib3, Vib5, Vibl, Vib2, 1103/9, E154482, El55855, El68143 and VCl respectively. Lane 14 is a Spanish isolate of V parahaemolyticus and lane 18 corresponds to V cho/erae. M: I. ladders with selected sizes indicated by arrows. Gel prepared with I% agarose, 0.5xTBE and run under the following conditions: 20 min pulse, at 65 volts (1.9v/cm), for 20 h
100 M 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 M
~485kb
~388kb
~291kb
~194kb
~97kb
Figure 5.9: PFGE profiles of Notl digested environmental isolates of V. parahaemolyticus and other Vibrio species. Lanes l-18 correspond to the following isolate designations: Vib2, Vib3, Vib5, Vib6, Vib7, Vib8, Vib9, Vib10, 1022/4, 1022/10, VP350, VP5, VP378, VP416, 219, 220, VCBAD01 and VM499 respectively. Lanes 15 & 16 correspond to Norwegian clinical isolates of V parahaemolyticus. Lanes 17 & 18 correspond to V cholerae and V mimicus respectively. M: A. Ladders with selected sizes indicated by arrows. Gel prepared with 1% agarose, 0.5xTBE and run under the following conditions: 20 min pulse, at 65 volts (1 .9v/cm) , for 20 h
101 Environmental Clinical
Figure 5.10: Comparison of Notl digested Norwegian clinical and UK environmental isolates of V. parahaemolyticus. Lanes 1-18 correspond to the following isolate designations: 219, 219*, 221 , 221 *, 222, 222*, 223, 224, 225,226,227, 228, VP350, VP350, VP416, VP416, VPl and VPl (undigested) respectively (duplicates are indicated with *). Lanes 15 & 16 correspond to Norwegian clinical isolates of V parahaemolyticus. M: A ladders with selected sizes indicated by arrows. Gel prepared with l% agarose, 0.5xTBE and run under the following conditions: 20 min pulse, at 65 volts ( l .9v/cm) , for 20 h
102 % Similarity Isolate Location Source Pulse Type 0.53 0.24 0.05 (PT)
~ 1022/10 Spain Environ menta I 01 228 Norway Oinical 02 Vib1 Sou1hampton, UK Environmental 03 Vib2 Sou1hampton, UK Environmental 03 Vib3 Sou1hampton, UK Environmental 03 Vib8 Sou1hampton, UK Environmental 04 Vib9 Sou1hampton, UK Environmental 04 Vib7 Sou1hampton, UK Environmental 04 Vib10 Sou1hampton, UK Environmental 04 VPEC1 Dorset, UK Environmental 05 VPSC1 Dor.;el UK Environmental 05 2436 Japan Oinical 06 VP1 0884 PI-I.S, UK NCTC Reference Strain 07 -~- _ ~~~15.. ___ -~vi_ro_!!"l_en_ta.!_ ___ -~ _ 6134 Japan Oinical 09 I 1 VP416 Dorset, UK Environmental 09 ------Vib5 Sou1hampton, UK Environmental------10 1 VP2 Sou1hampton, UK Environmental 10 VP1 Sou1hampton, UK Environmental 10 VP3 Sou1hampton, UK Environmental 11 VP4 Sou1hampton, UK Environmental 12 VP5 Sou1hampton, UK Environmental 12 110319 Spain Environmental 13 VP378 Dorset, UK Environmental 14 ------,E154482 PI-I.S, UK Oinical 15 1 __~~ __ ~~a~~!!, !,!K__ ~~ro.!'~~taJ. ____ 12 _1 6316 Japan O inical 17 ------~ E155855 PHLS, UK Oimcal 18 1 E168143 PHLS, UK Oinical 18 I 2053 Japan Oinical 18 _ ,!2~ - --~ '!"'! __ - -- _?~ i~l_ - -- - __ !B_ ! 226 Norway Oinical 19 227 Norway Oinical 20 5421 Japan Oinical 20 221 Norway O inical 20 220 Norway Oinical 21 222 Norway Oinical 22 1022/4 Spain Environmental 23 VP939 Dorset, UK En111ronmental 23 224 Norway Oinical 24 1902 PI-I.S, UK NCIMB Reference Strain 25 VP10 Sou1hampton, UK Environmental 26 219 Norway Oinical 27 VP8 Sou1hampton, UK Environmental 28 VP9 Sou1hampton, UK Environmental 29 VP6 Sou1hampton, UK Environmental 30 VP7 Sou1hampton, UK Environmental 30
Figure 5.11: UPGMA dendrogram of clinical and environmental isolates of V. parahaemolyticus. A total of 47 isolates were typeable by PFGE. Clusters (Pulse Types) created are based on a 88% similarity value indicated by the blue line. Red dotted area indicate two isolates from UK with clonal relatedoess with Japanese and Norwegian 03 :K6 isolates. Environmental isolates sharing similar proftles to clinical isolates are highlighted in green and blue. Scale above banding profiles indicates approximate sizes (Mb) . BioNumerics, Applied Maths, Belgium.
103 5.4. Discussion
This study looked at the variation in banding patterns of Notl digested PFGE profiles amongst clinical and environmental isolates of V parahaemolyticus. The technique of
PFGE has gained increasing popularity and is currently the most widely employed typing method for V parahaemolyticus with recent reports of standardised procedures (Wong et al., 2007; Parsons et al., 2007; Kam et al., 2008). The present work described, was conducted in 2004 to establish initial relationships between the isolates used using modifications to the BioRad PFGE method and to assess whether PFGE could distinguish pathogenic from non- pathogenic isolates. It was hoped that PFGE would serve as a benchmark subtyping method to compare clusters obtained with additional typing systems.
5.4.1 Clustering of clinical isolates
PFGE analysis showed a large genomic variation amongst clinical and environmental isolates with Pulse Types 18 - 22 showing the highest similarities of all the strains. Out of
I 0 clinical isolates that exhibited similar banding profiles, seven were of 03:K6 serotype, one of 05:K68 serotype and two were K untypeable (KUT). Within the homogeneous
03:K6 group of isolates, PFGE demonstrated the sensitivity to discriminate with the presence or absence of between two to three bands. A Norwegian isolate of serotype
Ol:KUT, believed to be one of the pandemic group, showed a similar banding profile
(>88% similarity) to 03:K6 strains (Figure 5.11), consistent with previous work demonstrating clonal relationships of pandemic serotypes (Chowdhury et al., 2000b;
Matsumoto et al., 2000; Okura et a/, 2003). The finding of UK acquired isolates to possess almost identical banding patterns to pandemic 03:K6 isolates from Japan and Norway demonstrated for the first time the presence of clonally related, more virulent pandemic isolates in the UK. These isolates were obtained from travellers arriving from India and
Thailand, so it appears that infection of V parahaemolyticus due to 03:K6 due to
104 consumption of indigenous shellfish in the UK have not yet been identified or tested for.
This highlights inadequate surveillance measures of pathogenic V parahaemolyticus in the
UK that may partially be due to low numbers of sporadic V parahaemolyticus infections in the UK, most of which are attributed to travel related cases, (Anon, 2006) and the self limiting nature of the infection.
Although some clinical isolates were grouped regardless of geographical origins, others such as El54482 and the NCIMB reference strain, 1902, with Pulse Types 22 and 23 respectively, showed less relatedness, even though they were of the same 01 :KI serotype.
An additional strain, environmental Spanish isolate, 1103/9 with the same 01 :K I serotype, showed yet a different Pulse Type to types 22 and 23 thereby PFGE, in this instance, was clearly able to discriminate isolates within the same serotype.
Typically, molecular differentiation is used to characterise strains that are phenotypically, biochemically or serotypically similar as described. In this study however, it was also found that within the same Pulse Field type, different serotypes were present, such as the presence of a 05: K68 strain amongst clinical strains of 03:K6 and also the presence of two serotypes in an environmental cluster, Pulse Type 05. This phenomenon was also described by Martinez-Urtaza et al. who showed different serotypes from the same area in
Spain were included in the same PFGE cluster (Martinez-Urtaza et al., 2004), thereby reinforcing the need for careful interpretation of data and the use of additional molecular
characterisation to determine accurate relationships between isolates.
5.4.2 Environmental isolates showing homology to clinical isolates
PFGE was able to group two environmentally acquired strains with known clinical strains
as demonstrated by Pulse Types 09, 15 and 16. An environmental isolate, VP416 that was
105 previously identified as a Kanagawa positive haemolytic strain, showed remarkably similar banding profiles (>94%) to a Japanese clinical isolate (6134) with serotypes 04:KUT and
04:K12 respectively. As isolate VP416 was obtained from an oyster sample from the south coast of UK in 2001, this finding may provide some evidence of the inter-continental spread of virulent strains of V parahaemolyticus amongst coastal waters, and thus highlights the need for increased surveillance of shellfish samples in the UK.
In addition to isolate 6134, an environmental isolate Vib6, from Southampton, UK, also exhibited some homology to a trh positive clinical isolate, El54482. With a total of 17 bands, the isolates shared 14 common bands with >86% dendrogram similarity value (Fig.
5.11 ). As E 154482 was isolated from a travel related gastroenteritis incident arriving from
Thailand, the similarities of PFGE profiles to that of Vib6 may suggest the presence of UK environmental reservoirs of V parahaemolyticus that may have originated from South East
Asia. PFGE data with a different restriction enzyme would provide more conclusive information regarding the relatedness of these two environmental isolates used in the study. Additional sub-typing methods (chapters 6 to 8) explore these intra-species relationships further.
5.4.3 Typeability with PFGE
In this study, a large degree of DNA degradation was evident throughout the experiments.
It was found that the use of both turbidity measurements and cell density counts to estimate
DNA concentration, and the use of chloramphenicol to inhibit replication, as used by the
BioRad method, was not deemed necessary. This reduced preparation time and DNA degradation. The method by Wong et al., (1996) that simply used a volume of an overnight culture of V parahaemo/yticus was implemented; whereby the harvested cells were washed in buffer containing l OmM Tris, I OOmM EDT A and 1mM NaCI (pH 8.0). This
106 buffer replaced the Cell Suspension Buffer (CSB) as supplied in the BioRad PFGE kit, the content of which remains publicaly unknown. Interestingly, the same isolates that were not typeable with Notl restriction were also not typeable with Sjil restriction. Untypeability of environmental V parahaemolyticus strains has widely been documented by several others including Marshall et al. (1999) and Yeung et al. (2002). Marshal! et al. reported 77% typeability amongst clinical isolates whereas Y eung et al. found only a single isolate untypeable amongst clonally related 03:K6 isolates.
The lengthy PFGE process described in the present work required numerous procedures that required several days to complete. Preparation of plugs and their treatment inevitably introduced DNase in addition to residual nuclease activity which may not have been adequately deactivated using the methods described. Once plugs were prepared it was found that they only remained stable for a few days therefore suggesting that meaningful comparisons could only be made using freshly prepared plugs.
Recently, a number of workers have employed the 'One Day Standardised Laboratory
Protocol for Molecular Subtyping of Non-typhoidal Salmonella by PFGE' (Pulse-Net;
Centers for Disease Control, Atlanta, Ga) procedures for PFGE typing of V parahaemolyticus (Martinez-Urtaza et al., 2004). This method differs mainly in that an overnight culture grown on solid plate media is directly transferred to cell suspension buffer and OD adjusted to 0.48 to 0.52 before cell lysis. Immediate cell lysis with proteinase K is followed by addition of molten agarose, premixed with SOS, to prepare the plugs. In a parallel study by colleagues elsewhere, this method correctly grouped the two
UK 03:K6 strains (E155855 and E168143) with clonally derived pandemic isolates from
Japan, Spain, Korea and Laos (Wagley et al., 2008). However, whereas the authors showed clinical isolates 2053 and 5421 in the same cluster, the BioRad method used in this study
107 identified them in two different pulse types (18 and 20 for isolates 2053 and 5421 respectively) at an approximately 84% similarity level. Clearly, there is an essential requirement for standardised procedures for data acquisition and analysis for the PFGE of
V parahaemolyticus to enable the meaningful exchange of inter-laboratory results.
5.4.4 Standardisation of PFGE
Despite a large degree of untypeability in the current study, PFGE has gained increasing popularity over recent years and these methods have so far yielded an increase in typeability. However, there do not appear to be many reports of this method being used extensively on environmental strains of V parahaemolyticus, as it has been shown that these are more likely to be untypeable (this work; Marshall et al., 1999). More recently, the
CDC PulseNet have put forward a standardised protocol for the PFGE of V parahaemolyticus (Parsons et al., 2007). In a multicenter evaluation of this method, that incorporated 36 well characterised strains of various serotypes from different geographical regions, a high discriminatory power (34 Pulse Types in 36 strains) and typeability was shown. The study involved clinical and environmental strains. However, there was no mention of methods of data analysis and an increasing number of bands (markers) with the addition of new strains can induce errors that are compounded exponentially thereby reducing the robustness of PFGE (Call et al., 2008). It remains to be seen whether this method, or others (Wong et al., 2007; Kam et al., 2008) that have recently been described, will gain widespread acceptance and increased typeability of environmental strains of V parahaemolyticus. Ideally, a protocol that would employ the same methods and pulse conditions to type other related organisms found in the same environment would be more beneficial. As the CDC PulseNet protocol was based on the procedures of V cholerae and
E. coli 0157:H7 typing, it would seem likely that V cholerae could also be typed.
108 5.4.5 Conclusion
To date, there has been very limited published data on the detection and prevalence of V parahaemolyticus in the UK. This chapter presents the first known study of the characterisation of UK isolates, using modifications to a protocol initially based on commercially available PFGE kits (BioRad Laboratories). Due to the nature of the clinical symptoms of V parahaemolyticus infection, together with the difficulty in correctly identifying the organism, it is probable that clinical cases are currently under reported or misdiagnosed with other self-limiting gastroenteritis related illnesses.
This chapter described the use of PFGE in assessing the relationships between clinically and environmentally acquired isolates of V parahaemolyticus. It was found that two UK isolates from gastroenteritis cases were closely related to those of Japan and Norway as demonstrated by PFGE profiles as well as belonging to the same 03:K6 serotype. The
03:K6 serotype is widely considered a clonally derived, pandemic strain responsible for disease worldwide (Hara-Kudo et al., 2003; Matsumoto et al., 2000; Martinez-Urtaza et al., 2005). This together with the finding of a non pandemic, 04:KUT isolate amongst C. gigas in British coastal waters that showed almost identical PFGE profile to a Japanese clinical isolate and presents additional evidence of clinically significant V parahaemolyticus in the UK and the importance of reliable typing systems for pathogenic isolates. Using PFGE, only 47 isolates could be characterised representing 77% of the total number of isolates. This, together with the long time required to process samples, as only a limited number of samples can be processed at any one time, adds to the difficulty in the routine use of PFGE for surveillance and epidemiological purposes. Clearly, more appropriate rapid, robust methods are required to screen large numbers of environmental and shellfish samples to differentiate potentially pathogenic isolates, that can then be
109 subjected ,to J~FGE analysis' or ;other, more di~criminaiory,. ·nu_cl~otige sequence base~ fypin_g sy~tents.
1:10 Chapter 6
DIFFERENTIATION OF VIBRIO PARAHAEMOLYTICUS ISOLATES BASED ON LENGTH POL YMORPHISMS OF THE 16S- 23S INTERNAL TRANSCRIBED SPACER (ITS) REGION
6.1 Introduction
Ribosomal RNA (rRNA) operons can be found in all species of bacteria with the genes usually arranged in a defined order (l6S-23S-SS), often with several copies distributed throughout the genome. These highly conserved regions make them ideal for the classification, identification and typing of isolates (Abed et al., 199S; Tilsala-Timisjarvi et al., 2001; Dingman, 2004). Both the 16S rRNA gene and SS rRNA gene have been used to establish phylogenetic relationships between a wide variety of organisms (Gazumyan et al., 1994; Yoon et al., 1997; Maiwald et al., 2000). The internal transcribed spacer (ITS!) region between 16S and 23S genes shows the most variation, both in length and sequence
(Gurtler & Stanisich 1996; Dennis et al., 1998; Perez-Luz et al., 2002; Wojciech et al.,
2004; Dillon et al., 200S) and this variation can be used for distinguishing between closely related bacterial species and, sometimes, between strains (Gurtler & Stanisich 1996;
Leblond-Bourget et al., 1996; Daffonchio et al., 2006). In addition to this ITS1 region, the spacer region between the 23S and SS (ITS2) may be useful for designing species-specific probes and establishing relationships between closely related species (Ji et al., 1994; Nour et al., 1998).
Sequence information of a number of V parahaemo/yticus isolates has revealed that they possess up to 11 copies of the rRNA operon (rrn) each having identical 16S rRNA genes
(rrs) but with 6 spacer classes differing in size with either I, 2 or 3 copies of each class,
111 (Makino et al., 2003; Gonzalez-Escalona et al., 2005b). However, a type strain isolated in
1951, varied in two of the spacer classes; one was found to be 45 bp larger and the other 15 bp larger containing genes for tRNA Ala and tRNAGlu (Maeda et al., 2000). A preliminary analysis of the spacers, demonstrated on a few clinical isolates by Gonzalez-Escalona and colleagues, differentiated clinical isolates of V. parahaemolyticus ITS l spacers into four groups based on two spacer classes (Gonzalez-Escalona et al., 2006a). By sequencing two clinical isolates, VpAQ and Vpl, it was further found that ITSl regions of 40bp and
208bp, close to the l6S and 23S genes respectively, were conserved in three clinical isolates of V. parahaemo/yticus. Despite these structural similarities and variation in spacer classes, little is known about intra-species relationships between clinical and environmental isolates based on ITS l regions. There are currently no published data assessing the feasibility of typing V. parahaemolyticus based on variation of the ITS 1 and
ITS2 regions or addressing its significance as an epidemiological typing tool with this organism.
This chapter applies the use of consensus primer pairs for amplifying ITSl regions. ITS2 regions were also amplified with primers developed using existing sequence information of
V parahaemolyticus (Mak.ino et al., 2003) described in section 3.5. This chapter describes the:
i) molecular typing of V parahaemolyticus based upon PCR amplification of ITS
l and 2 regions
ii) the usefulness of ITS typing as a molecular typing tool for V parahaemolyticus
isolates
iii) evaluation of the ability to discriminate clinical and environmental isolates of V.
parahaemolyticus based on ITS I typing.
112 6.2 Materials and Methods
6.2.1 Primers for amplification of ITS regions
Primers VINTF (5'-TGGGGTGAAGTCGTAACAAGG-3') from 16S rRNA corresponding to E coli l6S positions 1485 to 1505 and VINTR (5'-
TCCTTCATCGCCTCTGACTG-3') from E. coli 23S positions 37 to 56 were used to amplify ITS l regions (Maeda et al., 2000). E. coli 16S is notably larger than V. parahaemolyticus l6S: !54! bp compared to 147lbp (Brosius et al., 1978; Makino et al.,
2003) and primer VINTF is complementary to a conserved region (section 6.4.4;
Gonzalez-Escalona et al., 2006a) outside of V. parahaemolyticus 16S (Fig. 6.1 ). Primers for ITS2 were designed based on conserved 23S and 5S regions that flanked the ITS2 region of V. parahaemolyticus (Makino et al., 2003) as described in section 3.5 (Fig. 6.1).
ITS2 primer sequences were designated S2F, 5'-CTAATTGCCCGTGAGGCTTA-3' and
S2R, 5'-GAGTCAGGTGGGTCCAAAAC-3 for forward and reverse primers respectively.
ITS2 SS V!NTF.. 8! ~obpJ (1471bp)16S 2 3 ITS;------~~======(=28= + 9__ls,-b_p=)======~rc+ VlNTR S2R
Figure 6.1: Schematic representation of the V. paralraemolyticus rRNA operon (rmA), ITS regions and primers used in the study. Sizes are based on clinical isolate RIMD221 0633 (Makino el al., 2003). Primer VINTF is complementary to a conserved region in V. parahaemolylicus ITS I.
6.2.2 Amplification of ITS I and ITS2 regions
PCR amplifications were performed as described in section 3.6 using an MWG thermo cycler (Primus 96 Plus). Conditions were optimised with the following modifications: reactions were performed in 55 J..ll volumes and contained 20 ng of genomic DNA and 25 pmol of each forward and reverse primer. Each reaction mixture contained I U of Taq 113 polymerase (Roche), 1 x reaction buffer containing l.SmM MgCh (Roche) and each deoxynucleoside triphosphate (dNTP) at a final concentration of 0.2mM ..
The final cycling conditions for the ITS-PCR were as follows: an initial denaturation step at 94°C for 2 minutes followed by 30 cycles of94°C for 60 sec, 62°C for 45 sec, and 72°C for 90 sec and a final extension at 72°C for 5 min. Cycling conditions were optimised by performing PCR reactions at different annealing temperatures and with genomic DNA of three strains; type strain NCIMB 1902, Japanese clinical isolate 2053, and Spanish environmental isolate 1023/1, and PCR performed on a total of 56 isolates of V parahaemolyticus including a number of replicates. Additional species that were examined included, V cholerae, V vulnificus, V mimicus and several members of non-Vibrio spp.
All PCR products were analysed and compared by electrophoresis of 12 J.ll of each reaction mixture in 2% agarose gels as described in section 3.8.
6.2.3 Data analysis and dendrogram construction
Amplicons generated with primers VINTR and VINTF were scored both empirically by means of visual inspection and by software analysis. Isolates were grouped according to
fingerprint pattern types and also by using a curve based correlation coefficient and a band
based Dice coefficent with BioNumerics as described in section 3. The parameters that
were varied in BioNumerics during the identification of bands were "minimum profiling"
(used to identify bands based on the elevation of densitometric peaks relative to the highest
peak on the pattern), "minimum area" (used to identify bands on the area of densitometric
features relative to the total area of the pattern) and "shoulder sensitivity" an option that
allows identification of doublet bands where the curve appears as a "shoulder" to another
band. BioNumerics optimisation, position tolerance settings and percentage similarity cut
off points were altered in small increments until clusters resembling similar banding
114 profiles were achieved. A final position of 93% similarity was used for the dendrogram and ITS I groupings.
6.3 Results
6.3.1 Consensus primers for ITSl regions
Primers described by Maeda et al. (2000) were tested at PCR annealing temperatures of 62
- 68°C on two isolates of V. parahaemolyticus, NCIMB type strain 1902 and environmental isolate 1023/1 (Fig. 6.2). Bands of varying intensities were identified on the fingerprint pattern with more bright bands indicative of multiple ITS I regions of similar sizes present in V. parahaemolyticus (Section 6.4.5). It was found that varying the temperatures within these ranges did not significantly alter the banding profiles with higher annealing temperatures resulting in decreased yield of PCR products (Fig. 6.2). Increasing the annealing temperature reduced backgrow1d PCR noise but also reduced the intensity of amplicons generated (Fig. 6.2). However, increasing template and primer concentrations, from 10 to 20 ng and I 0 to 50 pmols respectively, was shown to specifically increase the amplification of the 390 and 635 bp spacer regions (Fig. 6.2, Plate A, lanes 6-8) and was used to help deduce optimised reaction conditions (section 6.2.2). However, increasing primers and templates did not increase the intensity of the two larger amplicons of approximately 790 and 810 bps. Faint bands above 790 bps were ignored from
BioNumerics analysis.
6.3.2 PCR of 238-SS (ITS2) intergenic spacer regions
Primers that were designed based on published 23S and SS sequences of V. parahaemolyticus amplified a single unambiguous band of approximately 150bp in UK
type strain 1902, Spanish environmental isolate 102311 and clinical isolate 2053 (Fig. 6.3).
Amplicons from all three isolates appeared consistent in size reflecting an ITS2 region of
approximately 90 bp (without primer regions). 115 6.3.3 PCR of 16S-23S (ITSl) intergenic spacer regions of V. paraltaemolyticus
PCR with primers VINTF and VINTR flanking the 16S-23S rRNA regions was carried out on 56 strains of V parahaemolyticus (Figs. 6.4 to 6.8). Between four and eight bands were amplified with two distinct markers of approximately 390 and 635 bps that distinguished all but one isolate (6316) of V parahaemolyticus from all the other species (Fig. 6.9). With the exception of isolate 6316, these two amplicons appeared of similar size in all of the remaining isolates tested (section 6.3 .4; Fig 6.11) with some apparent variability in size of the 390 bp amplicon amongst the environmental isolates (Figs 6.4 & 6.8). Isolate 6316 was characterised by a unique profile and did not match any of the other Vibrio spp. tested by
ITS 1-PCR. In addition to the 390 and 635 bps bands, a further three bands of approximately 640, 790 and 81 0 bps were present in most of the isolates, with the larger
sizes of bands generally appearing more faint. A further four bands, (between 390 and 640
bps, Fig. 6.1 0) some faint, were found distributed in some of the isolates. Based on general
banding profiles, there appeared no obvious correlation of banding patterns with the source
or the location of the isolates.
6.3.4 Reproducibility ofiTSl-PCR banding profiles
A comparison of repeated ITS 1 fingerprint profiles (Figs. 6.4 - 6.8, isolates underlined)
demonstrated the degree of reproducibility of the ITS-PCR technique. A total of 17
arbitrarily chosen isolates, that included three non Vibrio spp. were subjected to repeat
ITS 1-PCR and gel runs. Five UK clinical isolates in Figure 6.6 (Lanes 16 - 20) amplified
the same bands as Figure 6.5 (Lanes 12 - 14, 16 & 18). However, banding patterns, in
common with the rest of the isolates on the gel, appeared more intense with more
background PCR noise in Figure 6.6. Similarly, eight environmental isolates (Figs. 6.5, 6. 7
& 6.8, underlined) showed reproducibility of the two main amplicons. However,
identification of faint bands appeared problematic in some isolates and in some of the
repeats performed faint bands could not be identified at all. For example, initial 116 experiments with NCIMB reference strain 1902, (Fig 6.2, Plate A, 62°C) identified two faint bands between 500 and 600 bps, both bands were absent in Figure 6.4 (Lane2) and only one present in Figure 6.6 (Lane 20). A full comparison of the bands produced in repeat analysis is shown in Table 6.1. Of the total 17 repeats, a number of the faint bands could not be ascertained in at least five isolates including 1902, VP491, VP939, VPEC1 and VPSC1 (Table 6.1).
6.3.5 ITSl PCR in Vibrio species
The 16S-23S rRNA spacers in Vibrio species varied amongst members of the genera tested
(Figs. 6.3, 6.4, 6.5 & 6.7). Using the same consensus primers, up to seven bands of various intensities were present. Amongst the six known Vibrio spp. tested, V. cholerae shared a common band of approximately 635 bp with V. parahaemolyticus that was present in all six isolates including V. cholerae NCTC type strain 8042 (Fig. 6.8). An unspeciated Vibrio sp., Vib4 (Table 3.3), showed a profile matching that of Listonella damsela. LD365 (Fig.
6.5) Based on initial ITS 1-PCR, L. damsela appeared more closely related to V. vulnificus than V. cholerae or V. parahaemolyticus, although a larger number of isolates and sequence analysis would be required to verify ITS 1-PCR homologies between these species. Based on a limited number of isolates, the PCR developed was able to distinguish between all six Vibrio spp. and five non-vibrios (Fig. 6.9).
6.3.6 Dendrogram analysis and differentiation of V. parallaemolyticus isolates
Dendrogram analysis employing a curve based correlation coefficient was used to cluster
V. parahaemolyticus into thirteen groups (A - M) at a 93% similarity value (Fig. 6.11 ).
These groups appeared to be based on the positions and intensities of the two more prominent bands (approximately 390 and 635 bps) with faint bands largely unaccounted for (Fig. 6.11 ).
117 ITS types A and B consisted of a single clinical isolate, clustered with two environmental ones and a reference strain, NCIMB 1902. Isolates of these two groups appeared similar in banding profiles to those of Group C. This group, ITS-C appeared the largest with a mixture of both 32 clinical and environmental isolates and appeared closely related with
VP346 (ITS Type D); the isolate differentiated based on the presence of two distinct bands between 390 and 500 bps (Fig 6. 7). Most of the clinical isolates were found clustered in one of two groups amongst ITS-C with two bands around 640bp appearing as 'doublets' in comparison with environmental isolates of the same group (Figs. 6.6 & 6.11). Furthermore, two faint bands between 390 and 635 bp appeared variable amongst the 13 ITS Types. For example the 540 bp amplicon (Fig 6.1 0) was present in 14 out of 17 of clinical isolates whereas this amplicon could be identified anywhere in between 13 to 20 of 36 environmental isolates and thus highlighting variability in banding intensities (Section
6.3.4).
Other than isolate 6316 and reference strain VP I 0884, the remaining ITS Types, E-L, did not contain a mixture of both clinical and environmental isolates in any of the groups (Fig.
6.11 ). The most significant variation was found in the amplification of multiple ITS 1 spacers of various lengths that were not identified in any of the clinical ITS types. In addition to the 390 bp amp! icon, four environmental isolates, VP346, VP939, VPEC 1 and
VPSC I, amplified two other spacer regions of approximately 425 and 450 bps (Figs. 6.4 &
6.7). In all four isolates the 390 bp fragment was present, but was of reduced intensity.
Several other isolates also exhibited a 415 amplicon, but appeared faint and unaccounted for in the dendrogram. Environmental isolates also differed by the variable size of the main spacers found in V parahaemolyticus. These were clustered together in ITS groups E, G and H (Fig. 6.11 ). In addition to the new spacer types and intensity of some of the amplicons, length polymorphisms in the '390 bp' amplicon were observed. For example, isolates of E, G and H groups showed smaller amp! icons than typically found in the other 118 groups {Fig, :6.1111} Similar iiength ipdlyrnmphisrns. in 'spacers were also; identified in
environmental I isolates of IV iparahaemolyil'cus. 8linical :isolates~ in genenil, appeared ,more. homogeneous in: bilfidiiig profiles~
1119 M 2 3 4 5 6 7 8 9 M
Figure 6.3: PCR amplification of ITS2 regions of V. parahaemolyticus using primers designed for conserved 23S and 5S regions. Lanes l - 3 correspond to NClMB type slrain 1902, lanes 4- 6; Spanish isolate 1023/1, lanes 7- 9; Japanese isolate 2053 with decreasing template DNA concentrations of 20ng, lOng and 5ng respectively. Lane M: 1OObp ladders, sized of selected markers indicated by arrows on the right.
c E r.-----A~-...... , M l 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 M
Figure 6.4: PCR amplification of ITS I regions from clinical (C) and environmental (E) isolates of V. parahaemolyticus and other species (unmarked) using primers VINTF and VINTR. Lanes I - 20 correspond to the following isolates:- 1: E. coli Cla; 2: 1902 Reference strain; 3: 2053; 4: 2436; 5: 5421 ; 6: 6 134; 7: 6316; 8: V alginolyticus VA234 L; 9: Vib3; 10: Vib5; 11 : VP416; 12: VP491 ; 13 : VP939; 14: VPECl; 15: VPSCl; 16: V anguillarum CM104; 17: S. typhimurium ST5; 18: K. aerogenes KA5; 19: P. aen1ginosa PA 1; 20: E. coli BL2l. Lane M: 1OObp ladders, size of selected markers indicated by arrows on the right
121 E E E c r___ __..;A.______" ~ M 1 2 3 4 5 6 7 8 9 10 ll 12 13 14 15 16 17 18 19 20 M
Figure 6.5: ITSl fmgerprinting patterns of cUnical (C) and environmental (E) isolates of V. parahaenwlyticus and related species (unmarked) using primers VINTF and VINTR. Lanes I - 20 correspond to the following isolates:- 1: Negative control; 2: VP10; 3: Vjluvialis 11327; 4: Vibl; 5: Vib2; 6: Vib3; 7: Vib4 (unspeciated Vibrio sp.); 8: Vib5; 9: Vib6; 10: Vib7; 11: Vib8; 12: Vib10; 13 : V vulnificus KiDbW; 14: V vulnificus Wll067; 15: Listonella damsela LD365; 16: E154482; 17: El55855; 18: El68143; 19: 2436; 20: 6134. Isolates underlined are repeat PCRs. Lane M: lOObp ladders, size of selected markers indicated by arrows on the right
Norway Japan R
Figure 6.6: Comparison of ITSl fmgerprinting patterns of clinical isolates of V. parahaemolyticus from Norway, UK and Japan using primers VINTF and VINTR. Lanes l - 20 correspond to the following isolates:-I: Negative control; 2: 219; 3: 220; 4: 221 ; 5: 222; 6: 223; 7: 224; 8: 225; 9: 226; 10: 227; 11 : 228; 12: E 154482;13: E155855; 14: E168143; 15: 2053; 16: 2436; 17: 5421; 18: 6134; 19: 6316; 20:1902 Reference strain (R). Isolates underlined are repeat PCRs. Lane M: lOObp ladders, size of selected markers indicated by arrows on the right.
122 UK Spain UK ~ M 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 M
Figure 6.7: Comparison of ITSl fingerprinting patterns of environmental isolates of V. parahaemolyticus from UK and Spain. Lanes 1 - 20 correspond to the following isolates:- l: Negative control; 2: VP1 ; 3: VP2; 4: VP3; 5: VP4; 6: VP6; 7: VP7; 8: VP8; 9: VP9; 10: VP10; 11 : 1021/5; 12: 1022/10; 13: 1023/l; 14: 1023/3; 15: 1024/5; 16: 1103/9;17: VP331; 18: VP345; 19: VP346; 20: VP349. Isolate underlined represents a repeat PCR Lane M: lOObp ladders, size of selected markers indicated by arrows on the right
V. parahaemolyticus V. cholerae
M 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 M
Figure 6.8: Comparison of ITSl fmgerprinting patterns of environmental isolates of V. parahaemolyticus, V. cholerae and other species. Lanes 1 -20 correspond to the following isolates:- 1: VP350; 2: VP36L; 3: VP377; 4: VP378; 5: VP416; 6: VP491; 7: VP939; 8: VPECI; 9: VPSC1 ; 10: VP10884 (Type Strain); Ll: V. cholerae Reference Strain 8042; 12: VCI; 13: VC2; 14: VC3; 15: VC4; 16: VC5; L7 : Proteus mirabilis PM5; 18: S typhimurium ST5; 19: K aerogenes KA5; 20: P. aeruginosa PAl. Isolates underlined are repeat PCRs. Lane M: 100bp ladders, size of selected markers indicated by arrows on the right.
123 Table 6.1: Reproducibility of ITSI-PCR bands in repeat tests. Tests I and 2 refer to independent ITS PCR and gel runs. Uncertain band numbers are hyphenated. Corresponding band numbers in two repeats are indicated in bold. The 390 and 635 bp bands were present in all 14 isolates of V. parahaemolylicus in both tests. Discrepancies in the number of bands were also the result of inconsistencies in gel running conditions.
Isolate No. of bands References Test I Test 2 Figures: Lanes V. f!.Orahaemol}!,ticus: 1902 (Type Strain) 2-3 5 6.4: 2 & 6.6: 20 Vib3 4-5 5 6.4: 9 & 6.5: 6 Vib5 4-5 5 6.4: 10 & 6.5: 8 V PlO 3 3 6.5: 2 & 6.7: 10 VP416 3 4 6.4: 11 & 6.8: 5 VP491 2-3 4 6.4: 12 & 6.8: 6 VP939 3-5 7 6.4: 13 & 6.8: 7 VPECI 4-5 6-8 6.4: 14 & 6.8: 8 VPSCI 4-5 6-8 6.4: 15 & 6.8: 9 El54482 5 5 6.5: 16 & 6.6: 12 El55855 5 5 6.5: 17 & 6.6: 13 El68143 5 5-6 6.5: 18 & 6.6: 14 2436 5 5-6 6.5: 19 & 6.6: 16 6134 4 3-4 6.5: 20 & 6.6:18 Other species S. lypimurium, ST5 2 2 6.4: 17 & 6.8: 18 K. aerogenes, KAS 4 4 6.4: 18 & 6.8: 19 P. aeruginosa, PAl 2-3 2-3 6.4: 19 & 6.8: 20
124 % Similarity Species 810bp 635bp 390bp Isolate 2053 V. parahaemolyticus 6134 V. parahaemo/yticus E154482 V. parahaemo/yticus 1023/1 V. parahaemo/yticus VP7 V. parahaemo/yticus VP331 V. parahaemolyticus VP346 V. parahaemo/yticus 1902 V. parahaemo/yticus 219 V. parahaemo/yticus VP10884 V. parahaemo/yticus VP416 V. parahaemo/yticus 8042 V. cholerae 11327 V. fluvia/is VanCM104 V. anguillarum KA5 Klebsiella aerogenes LD365 Ustonella damsela W11067 V. vulnificus BL21 E. coli PA1 P. aeruginosa ST5 S. typhimurium
Figure 6.9: Comparison of ITS-PCR to differentiate representative V. parahaemolyticus from other members of Vibrio spp., and related bacteria. Arrow indicates a 390 bp spacer unique to V parahaemolyticus
2 3 4 5 6 7 8 9 10
f- 810 bp f- 640 bp f- 590 bp
f- 450 bp f- 390 bp
635 790 590 450 630 390 640 540 425 380
Figure 6.10: Length variations of 16S-23S (ITS I) spacers found in clinical and environmental V. parahaemolyticus. Lanes 1 & 3: bands found only in V parahaemolyticus with bands of Lane 1 as the predominant markers for the species. Lanes 3, 4, 5, 6 & 9: Bands found in some isolates of V parahaemolyticus. Bands above 790 bp were ignored. Data normalised in BioNumerics with internal and external reference positions with approximate sizes of bands (bp) indicated by arrow and below each lane.
125 % Similarity Isolate Location Source ITS 81Obp 635bp 390bp Types ~ ~ ~ !' VP1 Salllllfi'CliCil. UK Emlra1rn Vl'6 Salihanl>~Cil , UK ErMronmenlll c Vl'8 SalihanlJICil. UK ErMronmenlal c VP2 Sal!MirPCil. UK ErMronmenlll c VP3 SalihanlJICil, UK ~ronmenlll c 6134 J- Clillc:al c VP331 Dorset. UK Em4rorvne'llal c VP345 Oorsel. UK EnWcrvn Figure 6.11: ITSl-PCR banding proftles from 56 isolates of V. parahaemolyticus used in tbe study. The dendrogram was generated using BioNumerics and a densitometric curve based analysis with groups determined as shown with a dotted line at 93% similarity. Groupings (A - M) resulted from bias of predominant (darker) bands over faint bands. Based on this study, the identity of isolate 63 16 is questioned. 126 6.4 Discussion Both the length and the sequence of the ITS region separating the 16S and 23S rRNA genes are increasingly used to assess relationships in bacterial population studies (Daffonchio et al., 2006; Stewart & Cavanaugh, 2007). This chapter assessed the feasibility of ITS length polymorphisms as a simple method for sub-typing V parahaemolyticus based on consensus 16S and 23S primers. The 56 isolates used in the study were assigned to 13 ITS Types, of which all but two clinical isolates fell into a single ITS Type that included an isolate (6316) initially identified as V parahaemo/yticus, the identity of which became questionable after ITS 1-PCR analysis. Environmental isolates appeared more variable and were found included in all but two ITS Types thereby providing a simple and efficient means of differentiating isolates of V parahaemolyticus. ITS 1-PCR was thus shown to be a quick and reliable way of inter and intra-species differentiation of V parahaemolyticus. The following discussion addresses some of the issues with the ITS 1-PCR technique and highlights some further work in this area of study. 6.4.1 Primers used in the study The consensus primers used in the present study (Maeda et al., 2000) were shown to be of identical complementary match to V parahaemo/yticus regions based on BLAST comparisons at all 11 operon sites of the V parahaemolyticus genome based on strain RIMD 2210633 (Makino et al., 2003). Expected amplicon sizes for the sequenced strain are shown in Table 6.2. Based on these primers VINTF and VINTR successfully amplified up to eight bands despite the various degrees of background PCR noise that was present. Several approaches were taken to optimise reaction conditions and minimise the background PCR noise that was expected from the use of consensus primers in amplifying a number of products in a single PCR reaction. Bands of various intensities were produced implying either a favoured amplification of certain regions due to base mismatch in others or, as in this case, the presence of multiple amplicons of the same size (section 6.4.4). 127 6.4.2 ITSl-PCR banding patterns V. parahaemolyticus isolates used in the study were categorised into 13 groups based on a similarity score of 93% in dendrogram analysis. This point was selected as it differentiated isolate VP346 (ITS-D) from ITS-C, with three distinct bands less than 450bp that clearly distinguished this isolate from others. Despite this prominence, there was only a marginal difference in BioNumerics cluster analysis of less than 1% between VP346 and ITS-C. However, by using a 93% dendrogram similarity, other isolates that appeared visually similar were differentiated into their own types. For example, isolates in ITS groups A and B appeared similar to those of group C but it is likely that gel intensities and backgrowtd subtraction used in BioNumerics would have had some impact on reducing actual band intensities and thereby affecting calculations in the BioNumerics curve based analysis method. It is noteworthy that all four profiles of groups A and B, were from the furthermost lanes of gels where the UV illumination and the lightening system of gel photography may not have been optimal and hence background processing methods in BioNumerics may have incorrectly grouped similar banding isolates in separate clusters. This error is likely to be more apparent in techniques such as ITS 1-PCR as the analysis method is reliant on the discrimination based only on a few prominent bands. RAPD analysis (Chapter 8) explores whether a more complex banding pattern can minimise any such errors. Faint bands appeared largely unaccounted for in dendrogram analysis. Isolates in ITS-C could be visually identified as at least 4 banding types based on the presence of a 'doublet', and the presence or absence of faint bands between the 390 and 63Sbps. However, verification of these doublets would require more reliable size determination by use of acrylamide gels, or more accurately with the use of a DNA sequencer. The presence of these bands, were variable in some of the repeat tests and identification remained subjective. Despite this, and as isolates of ITS-C were from two gels (Figs. 6.6 & 6.7), the 128 grouping illustrates some degree of comparability between two independently run ITS 1- PCR experiments. However, environmental isolates of ITS-I, 1 and K that showed similar banding profiles to clinical isolates of ITS-C were distinguished into their own groups. This highlights a further factor for consideration when comparing different gels: bands that appeared apparently the same by visual inspection were grouped differently on the dendrogram. To what extent this could be attributed to BioNumerics' inaccuracy in normalisation of gels or to the ITS 1-PCR technique itself remains uncertain. For example longer gel runs may have been able to better resolve small differences in amplicon sizes. Inherently, the ITS-PCR method, as with most other PCR typing methods, contains variables such as temperature control and buffer quality that are difficult to control, some of which are discussed in section 6.4.3. To enable more accurate comparisons in BioNumerics, isolates should be randomly distributed amongst the gels together with an internal reference for gel normalisation to verify ITS 1-PCR as a useful typing method. 6.4.3 Properties of the ITSl-PCR system ITS1-PCR was shown to be a simple method of typing V parahaemolyticus throughout the study where all isolates of V parahaemolyticus appeared typeable. In some cases, individual agarose gels appeared unresolved (e.g. Figs. 6.4 & 6.8) and these are likely to be due to agarose gel running conditions and may reflect the age of buffers employed. A concentration of 2% ( w/v) agarose gels was used to achieve higher resolution of bands and thus increased gel running times and higher buffer temperatures may be partially responsible for these effects. To overcome such issues and to resolve bands which essentially all lay within a 400 bp region, other workers have employed polyacrylamide gels and have targeted more variable regions (section 6.4.4) within the ITS 1 domain to obtain smaller amplicons (Maeda et a/, 2000; Gonzalez-Escalona et al., 2006). 129 Perhaps the most significant criticism related to the ITS 1-PCR system is also the attribute that would most favour its use. Multiple ITS I regions of the same length, especially likely in strains possessing a large number of rRNA operons, may be difficult to distinguish and thus be unaccounted for, leading to bias in interpretation and phylogenetic groupings (Boyer et al., 2001). For example, the dendrogram presented in this present work clearly shows a bias in favour of the position of the more intense bands with weaker bands largely ignored (Fig. 6.11 ). Although this may be useful in its own right, as it highlights the significance of two, three or more copies of the ITS I spacer, it may not necessarily reflect true diversity of single copy spacers that may exhibit greater variability and thus more useful for a dendrogram based analysis. To overcome the bias generated by BioNumerics, several parameters (section 6.2.3) were altered for dendrogram construction to obtain reliable groups that corresponded with similar banding profiles. A 'band based' Dice coefficient was also employed but failed to adequately group similar isolates together, most likely due to the close proximity of the bands. A recent study that assessed BioNumerics parameters for optimal correspondence of E. coli strains showed that a curve based method was found to be more robust and reliable than the 'band based' banding profiles in REP-PCR (Goldberg et al., 2006). Clearly, further work optimising parameters would be necessary for more accurate software based analysis. The findings from these experiments highlight the need for careful empirical assessment of groupings that are automatically generated from software based analysis. 6.4.4 Sequence based analysis of ITSl Due to the relative large copy number of the rRNA operons present in V parahaemolyticus and the presence of multiple ITS I of the same size, sequence analysis of spacer regions was not pursued in the present study. Clearly, the source of the ITS! variation present in some of the isolates have not been established. The 11 copies of the 16S genes of the 130 sequenced strain, RIMD 2210633, have been shown to possess identical nucleotide sequences (Makino et al., 2003). The design of operon specific primers would therefore involve initial 'Long PCR' followed by primers targeted to the conserved sequences, flanking ITS 1 regions. Gonzalez-Escalona et al. (2006a) demonstrated that conserved regions of 40 bp and 209 bp, flanking the 16S and 23S regions respectively, were present in the ITS1 of V parahaemolyticus (Fig. 6.12). Therefore, cloning and sequence analysis of a smaller region may help to establish further relationships between clinical and environmental isolates of V parahaemolyticus. The findings in the present study show that the 540 and 590 bp amplicons exhibit the greatest variability in size, and thus one or both of these ITS 1 regions may provide a source for genetic heterogeneity. 6.4.5 ITSl spacer classes in V. paralraemolyticus The presence of multiple rRNA operons in the same genome have been widely documented in a large number of species including E. coli, Salmonella, V cholerae and V vulnificus (Boyer et al., 200 I; Chun et al., 2002; Gonzalez-Escalona et al., 2006a). Up to 14 rRNA operons in psychrotolerant strains of Bacillus cereus have recently been reported (Candelon et al., 2004) with operon numbers of seven commonly found in bacteria. However, approximately 40% of bacterial species are believed to have just one or two copies of the rRNA operons (Acinas et al., 2004) and in such instances it is unlikely that length polymorphisms identified with ITS 1-PCR would serve as a useful tool for sub typing. 131 I Conserved region 1 6S r DN ~~ Co!'5"rcg1on~ved I bp I 1 2JSrDN~ 100 209 hp 40 bp ------~-.;~~;;;;~;;;------r.s:······ Figure 6.12: Schematic representation of the ITS I conserved regions in V. parahaemolyticus. Up to six spacer classes between 277 and 706 bps identified possessing genes for up to four tRNA (Table 6.2; Maeda et al. , 2000; Gonzalez-Escalona et al., 2006a). Table 6.2: Comparison of spacer classes identified in different V. paralwemolyticus, Variations in spacer sizes (a- c) based on four sequenced clinical strains that may be used to discriminate isolates are underlined. Column d shows approximate sizes based on current study. ITSI Spacer tRNA genes found in ITSI region ITS I S(!acer sizes {bf!) C lass a b c d 0 None 277 278,278 278 268-278 E Glu (UUC) 475 476, 461 lli 428-478 lA Ile(GAU), Ala(UGC) 523 524, 524 524 523 AE Ala(GGC), Glu(UUC) 54 1 542,542 527 528 EKV Glu(UUC), Lys(UUU), Vai(UAC) 668 669, 669 669 678 EKAV Glu(UUC), Lys(UUU), Ala(UGC), Vai(UAC) 705 706, 706 706 698 a Based on Maeda et al., 2000 b Based on two isolates, type strain ATCC 17802(Aq and RJMD 22 10856 (Gonzalez-Escalona, et a/, 2006) c Based on clinical strain RM1D 22 10633 (Makino et a/, 2003) dEstimated sizes aud li kely class assigned based on this study (band sizes minus 11 2 bp) 132 Based on the present limited study, it is most likely that the intensity of some of the bands (390 and 635 bps) are due to multiple copies of the same ITS spacer found in V parahaemolyticus (Maeda et al., 2000; Makino et al., 2003; Gonzalez-Escalona et al., 2005b, 2006a) although sequence analysis of representative strains would be required to confirm these findings. Studies on the structural variation of the 16S-23S spacer regions of V parahaemolyticus have identified six spacer classes based on different tRNA gene compositions (Table 6.2). These classes were based on four clinical strains two of which were of the pandemic 03:K6 serovar that are widely regarded to be clonally related (Matsumoto et al, 2000; Chowdhury et al., 2000b). Furthermore, ITS I lengths between all four strains were in close agreement with each other (Table 6.2) with the lengths of spacers supporting the current study with 55 isolates of V parahaemolyticus. Present work has demonstrated that clinical isolates do not exhibit significant ITS 1 length heterogeneity in repeated PCR analysis and thus the varying banding intensities offer further evidence to support the case for multiple amplicons. Additionally, pandemic 03:K6 isolates that were used in the current study show identical banding profiles to other clinical and environmental isolates. Based on these findings, ITS1 lengths (at least of the clinical isolates used in the study) correspond favourably to those of sequenced strains. These lengths take into account a subtraction of 112 bps corresponding to the primer positions near the 16S and on the 23S regions (Table 6.2). Based on these initial assumptions, it seems likely that the ITS I Spacer Classes 0 11 1 (possessing no tRNA genes) and lA (possessing genes tRNA e and tRNAA ") present in three and two copies respectively, in the sequenced strains are commonly found in V parahaemolyticus. However the significance of having multiple copies of certain classes of operons remains unclear. Based on band intensities, it is probable that some environmental isolates may possess variations in operon copy number, a phenomenon suggested by others in B. cereus (Candelon et al., 2004, Clostridium difficile (Sadeghifard et al., 2006) and 133 Acinetobacter spp. (Maslunka, et al., 2006). The present data also suggest that Spacer 01 Class E, possessing a single tRNA " gene, was the most variable of regions supporting 01 previous work showing that indels near the tRNA " gene was responsible for ITS 1 spacer variation (Gonzalez-Escalona et al., 2006). The presence of one, two or both bands (540 and 590 bps) in some of the isolates in the current work suggest that some isolates may possess two copies of this spacer, rather than just the one identified in previous studies (Makino et a/, 2003, Gonzalez-Escalona et al., 2006a). The presence of these two bands may also correlate with the two rrs classes identified in V parahaemolyticus (Moreno et al., 2002; Harth et al., 2007). Clearly, further work and sequence analysis would be required to verify any of these claims and establish the significance of these two bands and spacer classes to the virulence of V parahaemolyticus. This work has shown that multiple rRNA operons present in the V parahaemolyticus genome can be exploited for the sub-typing of this organism using a simple PCR reaction. The primers used in the study identified additional ITS 1 spacers in environmental isolates that previously had not been reported. The main ITS 1 spacers (i.e. the more intense bands) consisting of multiple ITS 1 regions offered a quick method to differentiate V parahaemolyticus from other members of the aquatic environment but clinical isolates in general appeared indistinguishable from environmental strains with ITSl-PCR. Nonetheless, a previous study has shown that ITS spacers can be resolved directly from bacteriallysates without the need for further DNA purification (Gonzalez et al., 2003) thus favouring ITS-PCR for V parahaemolyticus. The large numbers of genes for tRNA molecules present within the ITS I spacer regions also suggest that a similar method may be employed to identify length polymorphisms in V parahaemolyticus and is explored further in the next chapter. 134 Chapter 7 DNA PROFILING OF VIBRIO PARAHAEMOLYTICUS ISOLATES USING tDNA-PCR TO GIVE INTERGENIC LENGTH POLYMORPHISMS (ILPs) 7.1 Introduction The genes for tRNA can be found arranged in clusters distributed throughout the bacterial genome often within multiple tandem repeat units. The order and arrangement of these genes is highly conserved, whereas the intergenic spacer regions between them can vary considerably, being prone to the insertion of mobile genetic elements such as insertion sequences (IS), transposons, bacteriophages and genomic islands (Gurtler & Mayall, 2001). Length polymorphisms between adjacently primed tDNA sequences have been used to assess the degree of relatedness in a wide range of bacterial species and sub species. The PCR amplification of tDNA intergenic length polymorphisms (tDNA-ILPs) can be achieved by employing consensus primers, or specifically designed primers that are annealed at low stringency. Welsh and McClelland developed a set of consensus tRNA gene primers designed to face outwards from the end of the tRNA genes that have been shown to amplify a tDNA-PCR fingerprint that is determined by the arrangement oftRNA genes on the bacterial genome (Welsh & McClelland, 199la; McClelland et al, 1992). This technique has been successfully applied to distinguish at the species and subspecies level in a wide group of bacteria including Renibacterium salmoninarum (Alexander et al., 2001 ), Streptococcus (McCielland et a/, 1992; Zschock et al., 2005), Staphylococcus (Welsh & McCielland, 1992; Maes et al., 1997), Acinetobacter (Ehrenstein et al., 1996) and more recently in Klebsiella pneumoniae (Lopes et al., 2007) 135 Genome sequence information for V parahaemolyticus has shown the organism to possess up to 126 tRNA genes distributed between the two chromosomes in comparison to 11 rRNA operons (Makino et al., 2003; Gonzalez-Escalona et al., 200Sb). Thus, the successful application of consensus primers targeted to the tRNA regions of the V parahaemolyticus genome has a greater banding pattern potential than 16S-23S ITS 1-PCR (Chapter 6) to show intra-genomic variation, and hence provide a simple and rapid means of typing of this organism. The primers developed by Welsh and McClelland were based on I) at least a perfect match in five bases between the 3' end of the primer and many tRNA genes and 2) extensive homology in the rest of the primer with a number of different tRNA genes from a wide variety of sources (Welsh & McCielland, 1991a). The recent use of these consensus primers with Klebsiella spp. revealed different tRNA arrangements and highlighted intraspecies heterogeneity (Lopes et al., 2007). Considering the ease of use and simplicity, this method could show high efficiency for the typing of V parahaemolyticus. This chapter describes the use of consensus primers developed by Welsh and McClelland and the development of a PCR method to assess the differentiation of V parahaemolyticus. Thus the following were investigated: i) molecular typing of V parahaemolyticus based upon PCR amplification of tRNA spacer regions ii) comparative analysis of different sets of consensus primers described by Welsh and McClelland (1991) with V parahaemolyticus iii) the usefulness of tRNA-PCR ILPs typing as a molecular typing tool in V parahaemolyticus and iv) an evaluation of the ability to discriminate clinical and environmental (that may also be pathogenic) isolates of V parahaemolyticus based on tDNA-PCR ILPs typing. 136 7.2 Materials and Methods 7.2.1 tDNA primer screening with V. para/raemolyticus Primers T3A, T3B, T5A and T5B (Table 7.1) were initially screened in all 6 pair wise combinations on a panel of strains that included NCTC reference strain VPI 0884 and clinical and environmental isolates, in accordance with the published protocols (Welsh & McClelland 1991, 1992; Clementino et a/, 2001) and reaction conditions optimised for use with V parahaemolyticus genomic DNA. Table 7.1: Description or consensus tDNA primers used in this study Primers Sequence Reference T3A 5'-GGGGGTTCGAA TTCCCGCCGGCCCCA-3' } T5A 5'-AGTCCGGTGCTCTAACCAACTGAG-3 Welsh & McCielland, 1991a T58 5'-AA TGCTCTACCAACTGAACT-3' McCielland et al., 1992 T38 5'-AGGTCGCGGGTTCGAA TCC-3' 7.2.2 PCR amplification of tDNA intergenic spacer regions using consensus primers The tDNA-PCR reactions were performed with DNA extracted from 56 isolates of V parahaemo/yticus and a further ten species that included other members of Vibrionaceae, Listonella spp., Klebsiella spp., and E. coli. Of the six primer sets, three were used to deduce fingerprinting profiles for all 69 isolates studied, viz. primer set I (T3A/T5A), primer set 3 (T3B/T5A) and primer set 5 (T3A/T3B); fingerprints were designated tDI, tD2 and tD3 respectively. Reactions were performed using a 50 111 mixture containing 1 U of Taq polymerase and reaction buffer with 1.5 mM MgCh (Roche ), 20 pmol of each tRNA gene primer (Sigma), 0.2 mM deoxynucleoside triphosphates, and 20 ng of bacterial DNA. PCR amplification was facilitated in a DNA thermal cycler (Model: Primus 96, MWG Biotech). The reaction 137 mixture was incubated at 94 oc for 2 m in, and then subjected to 40 cycles of 94 oc for 30 s, 50°C for 30 s, and 72°C for 2 min and a final cycle of 94°C for 30 s, 50°C for 30 s, and 72°C for I 0 min. The amplification products were visualised after electrophoresis and banding profiles compared on 2% agarose gels. Banding profiles were also compared for reproducibility from two independent experiments. 7.2.3 Dendrogram analysis of tDNA-PCR Dendrograms showing the relationships between isolates of V parahaemolyticus were generated using BioNumerics Software with modules for Fingerprints and Cluster Analysis (version 5.1, Applied Maths, Ghent, Belgium). All banding patterns for the three consensus primer sets were used for the cluster analysis of 56 isolates of V parahaemolyticus. The bands from each lane were subjected to curve calculations and analysis to exclude background noise from PCR reactions. The tracks were then normalised using two external reference positions that consisted of 50 bp molecular weight markers on either side of the gel and a common band that appeared in several lanes as an internal reference. Up to nine bands per lane, many of which were also common between isolates, were scored according to their presence or absence within the I 00 to 800 bp range only. Dendrograms calculated by the unweighted pair group method by arithmetic averaging (UPGMA) algorithm using both band matching coefficients (Dice) and curve matching correlation coefficients (Pearson product-moment correlation) were used as appropriate to obtain the most reliable clustering patterns. A dendrogram that combined the data for all three fingerprint sets was also constructed. BioNumerics position tolerance settings were 0.5% optimisation, a position tolerance of 5% of gel and ignoring faint bands. 7.2.4 Concordance between tDNA fingerprint types The concordance between tO I, tD2 and tD3 fingerprint types were determined using BioNurnerics by comparing the values from the similarity matrices of the typing methods 138 ,, that \\fete piotted: IOWX allo y axes. iEach dot on a scatterplotrepresents corresponding ,. •simil!llity values.· for ;two isolates by· different ;primer :pairs given :on :the x and y axes ·andi I ' thus :gives :an: ·indict~tioll ofthe degree 911, concordance betWeen. two ·techitiqi.ies, The . . ' 'Pearson·, product~moment ·correlation. :coefficient was used ,to· calculate the ·.congruence ,, 'l r: I : I I [' I I I I 139 7.3 Results 7.3.1 Consensus tDNA primer combinations All six combinations of the four primers described by Welsh and McClelland were used on a test panel consisting of three clinical and three environmental isolates of V parahaemolyticus. These primers were designed to amplify the intergenic spacer region from the 5' and 3' ends of tRNA genes and were selected based on the successful amplification of the most number of bands that were widely spread in fragment size (Fig 7.1). Primer pairs T3B/TSB and T3A/T5B showed a similar banding profile to T3B/T5A and T3A/T5A respectively, but fewer bands. Similarly, primers T5AIT5B showed a maximum of three bands amongst the six panel strains and therefore the remaining three primer pairs T3AIT5A, T3B/T5A and T3A/T3B were selected to investigate their ability to produce reproducible IONA fingerprinting patterns and assess the ability to subtype clinical and environmental isolates of V parahaemolyticus. 7.3.2 tDNA-PCR subtyping based on primers TJA and TSA (tDl) The use of primers T3A/T5A (tO I) produced fingerprints that included four major bands of approximately 205, 355, 410 and 740 bp in most of the isolates of V parahaemolyticus. Additional bands were present but appeared faint and were excluded from the analysis. Most of the differences in fingerprinting profiles were based on small but significant differences in the relative sizes of fragments that were amplified (Figs. 7.2 and Fig. 7.3). Using UPGMA cluster analyses on V parahaemolyticus is was possible to group the 56 isolates of V parahaemolyticus into seven clusters at a 90% similarity value, labelled tDl- A to tDl-0 (Fig. 7.3). Cluster tDI-A was the largest (25 isolates in fingerprint type tDl) but unable to differentiate three clinical UK isolates from the remaining 22 environmental isolates. Similarly, two Japanese clinical strains were grouped with 7 environmental isolates in cluster tD 1-C. However, cluster tO 1-D consisted entirely of 13 clinical isolates 140 that included all I 0 isolates from Norway, two from Japan and a UK clinical reference strain. Although these isolates were from three different locations, all 13 isolates were grouped into a single cluster that included isolates of diverse serotypes and Pulse Types (section 5.3.5). Cluster tD 1-F contained five isolates of environmental origin and from the same geographical region. Isolates Vib8 and 6316 could not be grouped into any clusters; therefore, in this study they are the single representatives of two other groups tD 1-E and tDI-G respectively (Fig. 7.3). As with Chapter 6, together with fingerprint types tD2 and tD3, the identity of isolate 6316 as V parahaemolyticus remained questionable. 7.3.3 tDNA-PCR subtyping based on primers T38 and TSA (tD2) Fingerprints produced by the primers T3B/TSA (tD2) showed a maJor band of approximately 370 bp that was present in all but one isolate (6316) of V parahaemolyticus (Fig. 7.4 ). A further nine bands of less intensity ranging in size between I 00 and 500 bp were also present in most of the isolates and could be used to distinguish V parahaemolyticus into 8 types (Fig. 7.4: E). However, as with primers T3AITSA, faint bands were excluded in the BioNumerics cluster analysis (Fig. 7.5). Cluster analysis at a similarity value of 95% enabled the 56 isolates of V parahaemolyticus to be grouped into 7 clusters (tD2-A to tD2-G). Cluster tD2-A was the largest accounting for 64% of the isolates but was unable to differentiate between a total of 36 clinical and environmental isolates at 95% similarity (Fig. 7.5). Group tD2-B was the only group to consist of clinical isolates, both of which were from Norway. Clusters tD2-C and tD2-F both consisted of a mixture of clinical and environmental isolates, whereas tD2- D and tD2-E contained environmental isolates only. 141 7.3.4 tDNA-PCR subtyping based on primers T3A and T3B (tD3) Primers T3A/T3B (tD3) amplified a maximum of nine visible amplicons with three major bands of approximately 240, 270 and 500 bp that were present in all but one isolate of V parahaemolyticus (6316). A distinct feature of tDNA-ILPs with these primers was that the sizes of all bands fell within a relatively short range of240-550 bp (Fig. 7.6). At a 95% similarity value, a UPGMA cluster analysis grouped the 56 isolates into 9 clusters (tD3-A to tD3-I). Cluster tD3-B appeared the largest consisting of 26 clinical and environmental isolates (Fig. 7.7). Two other clusters (tD3-A and tD3-C) also consisted of mixed clinical and environmental isolates. Group tD3-D contained three clinical strains and NCTC UK reference strain, VP I 0884 (Fig. 7. 7). In addition, two clinical strains (219 and E155855) were clustered with an otherwise predominantly environmental group of isolates in clusters B and C respectively. Three Norwegian clinical isolates that were previously grouped with other clinical isolates (based on fingerprint types tD l and tD2) from the same geographical region fell into a distinct cluster (tD3-D) of its own (Fig. 7.7). Furthermore, using primers T3A and T3B isolates VPSCI, 2053 6316 and 1021/5 represented unique groups of their own. 7.3.5 Comparative analysis of tDNA fingerprinting with different primers Using a curve based correlation coefficient to generate dendrograms for all three primer sets, it was possible to group isolates at relatively high similarity values of >90%. With dendrogram analysis the 56 isolates of V parahaemolyticus were grouped into seven, seven and nine clusters for the three fingerprint types, tDl, tD2 and tD3 respectively, as described above. With the exclusion of isolate 6316 that was distinguished into its own type with all three primer sets, the remaining 55 isolates showed similarities greater than 66%, 93% and 83% in each of the fingerprint types tDI, tD2 and tD3 respectively (Figs. 7.3, 7.5 & 7.7). Of the groups obtained, there appeared some concordance in groupings 142 amongst the three fingerprint types. For example, four Spanish isolates (1022110, 1023/1, 1023/3 and 11 03/9) grouped in the first cluster of tD 1 and tD2 were also sub-grouped in t03-C but a further isolate (1021/S) was not identified as concordant. Likewise, whereas two isolates (VP349 & 20S3) from tD 1-B were found together, they were identified in separate groups in tD2 and tD3. Similar qualitative measures of concordance could be identified throughout the three fingerprint types. A comparison of the similarity matrices obtained from the three pnmer pairs was performed (Fig. 7.8). Three scatterplots representing all three comparisons demonstrated data skew towards high percentage similarity values whereas plots at the lower percentages were representative of distance relatedness of isolate 6316, the identity of which remained uncertain from typing experiments. A measure of concordance calculated with the Pearson product-moment correlation coefficient yielded congruence values of 69.9, 47.2 and 74.6 percent for comparisons tD I /t02, tO llt03 and t02/t03 respectively (Fig. 7 .8). Linear regression models showed good data fit of 7S, 66 and 82% for the respective comparisons (Fig. 7.8) Therefore, fingerprint types tD2 and t03 were identified most concordant and tDl and t03 as the least concordant of the three comparisons. 7.3.6 Combined dendrogram analysis of tDNA-PCR Fingerprints tO 1, tD2 and tD3 were combined as a composite data set (tD-combined) to assess whether better discrimination between V parahaemolytiocus isolates could be achieved (Fig. 7.9). At 9S% similarity, isolates were grouped into 12 clusters (tO- combined A to L) of which five isolates (1 024/S, VP939, 2436, 20S3 & 6316) formed single members of their own clusters. With the identity of isolate 6316 questioned, the remaining SS isolates appeared >84.S% similar in combined dendrogram analysis. Cluster tO-combined -C grouped three clinical strains with environmental isolates from Southampton whereas Clusters 0, E, J and K consisted entirely of environmental isolates. 143 Group F clustered 12 clinical strains with NCTC reference strain V P 10884. The composite tD-combined data did not discriminate more than single tDNA primer pairs alone. tDI formed a larger number of groups at 95% similarity but groupings appeared to correspond largely to the isolates from the same gels. The combined data is evaluated for concordance with other typing systems in Chapter I 0. 7.3.7 Reproducibility of tDNA-PCR A side-by-side comparison of six isolates of V. parahaemolyticus was performed to assess the reproducibility of the tDNA-PCR technique (Fig. 7.1 0) with all three sets of primers. A total of 18 comparisons of two repeats were analysed. Isolates from tDNA-PCR described in the reproducibility test were obtained from Figures 7.1 (20 pmols of primers), 7.2, 7.4 and 7.6. Fingerprints appeared largely reproducible with regards to presence or absence of bright bands, but faint bands showed more variability (Fig. 7 .I 0). For example, in some of the repeats, faint bands could not always be ascertained, appeared absent or present in the repeat. Additionally, in some tests the brighter bands did not appear consistent between the two repeats (tD 1: I 021/5; tD3: 1021/5 & 2053). However, of the 18 comparisons, the main amplicons (4 in tDI, I in tD2 and 3 in tD3) could be identified in all but one comparison (Primer set tDI, isolate VP10884) (Fig. 7.1 0). 7.3.8 tDNA-ILPs in other species Distinct fingerprint profiles were obtained for isolates of V. anguillarum, V. mimicus, V. fluvialis, V. cholerae, V. vulnificus, V. alginolyticus and L. damsela with PCR conditions optimised for V. parahaemolyticus in all fingerprint types (Figs. 7.2, 7.4 & 7.6). In all three fingerprint types isolate 6316, which had previously shown a unique fingerprint pattern, matched that of V. mimicus (Figs. 7.2, 7.4 & 7.6). Additionally, the profile of isolate Vib4 144 an unspeciated Vibrio sp., matched that of L. damsela (LD365). Furthermore, common amplicons were present in several Vibrio spp. For example, with primers T3B and TSA, V cholerae and V parahaemolyticus shared two common sized fragments of approximately 200 and 470 bp (Fig 7.3). Similarly, V mimicus and V vulnificus shared common bands with V parahaemolyticus. Dendrogram analysis revealed the close tDNA-ILP similarities of V alginolyticus (2341) and V vulnificus (KIDBVV) with that of V parahaemolyticus (Fig. 7.11). The data showed a high degree of consistency of the tDNA-PCR technique in differentiating V parahaemolyticus at the species level. 145 Gell: Primer sets T3Aff5A and T3Aff5B M I 2 3 45 6 78 910 11121314151617181920M M21222324252627282930 - M 700bp 350bp IOObp Gel 2: Primer sets T3Bff5A and T3Bff5B M I 2 3 45 6 78 910 1112 1314151617181920M M21222324252627282930M Gel3: Primer sets T3Aff3B and T5Aff5B ~700bp ~350bp Figure 7.1: tDNA-PCR fingerprinting profiles with all 6 combinations of tDNA primers using a panel strain of 3 clinical and 3 environmental isolates. Lanes corresponding to the following isolates: 1-5: 102115; 6-10: 2053; 11-1 5: Vibl ; 16-20: El55855; 2 1-25: VP350; 26-30: NCTC reference strain VPI0884. In each set of 5 lanes, the first two lanes correspond to I 0 pmol of primers, and the second two lanes correspond to 20 pmol of primers. Each lane alternates with the first and second primer set in Gels I, 2 and 3 above. The fifth lane for each isolate is DNA only. Negative controls are marked with - and M corresponds to 50 bp molecular size markers: selected sizes are indicated on the right of each gel. 146 ~700bp ~350bp ~200bp ~IOObp ~700bp ~350bp ~200bp ~IOObp - - .. - - . _: : - ~ - - :. - : - ·; - -- - ..: 1 ~700bp - -- ~: ~~J· ; - 111.. . l 1i'11· .f __;u ~!!!!j ~350bp - - -!"!it~~ t~~w,•!""""lft!! ll~ . - . - __ .. ___ ---·-- - - ~200bp ~IOObp Figure 7.2: tONA-PCR fingerprints of 56 isolates of V. paraltaemolyticus and related spedes of clinical and environmental origin using primers T3A and TSA. Lane marked(-) is no DNA control and each lane is marked with the corresponding isolate at the top. NCIMB and NCTC reference strains for V. parahaemolyticus are 1902 and VP I 0884 respectively. Strains 11327 and 8042 are NCTC reference strains for V. vulnificus and V. cho/erae respectively. In each gel, 50 bp ladders were used, sizes of selected markers are indicated by arrows on the right 147 Isolate Location So11rct tDl clusttr % Similarity 730bp 4ll))p JS5bp 205bp I t I I IIP3n 0cnot. UK EnWonmentll IIP378 Ocnat, UK En-tll VP350 0cnot. UK En-Ill VP416 Ocnat, UK En-tll 111>3 ~•. EnWonmentll Vl>5 ~ .. EnWonmentll VP4 SoulllaTc*Jn,. En-tll VP7 ~ .. En-111 VP2 ~en .. EnW'clnnw>tll IIP3 ~en •. EnWcnmonllf VP10 ~ .. enwon...... VPI ~ .. EnWmmlnllf VP8 ~ •. E.-111 A VP!I ~en .. Em4ronnwrt:ol VP8 SoulllaTc*Jn.. EnWcnmonllf 1022/10 Spoin Em4ronnwrt:ol VP491 Ocnat, UK E,....,._lll 102311 Spoin E~ 102313 Spoin EnW'clnnw>llf 110319 Spoin e.-111 E154482 PHLS, UK c.1lcll E15585S PHLS, UK c.1lcll \11)1 ~ - EnWonmentll \11)2 ~• . EnWonmonllf El68143 PHLS, UK c.1lcll 102115 Spoin &Wmmontll }- B 11102 PHLS. UK NCIIIIB R-..:e S1roin IIP346 Ocnat, UK ~ IIP349 Ocnat, UK EnWon.,.,tll 102415 Spoin EnWonmonlll Vl>5 ~ .. EnW<>nmenllf \11)7 ~•. EnWonmontll c Vl>10 ~ . - ~ 2053 J- Clinicol 5421 c.1lcll VP939 J-Ocnat, UK EnWonmonllf 226 ~ Clinicll 227 ~ c.1lcll 228 ~ c.1lcll VP101l84 PHLS, UK NCTC Rolorwnce StlWn 221 ~ c.1lcll 223 ~ c.1lcll 220 D 224 225 222 8134 219 2438 Clinicll \11)8 J-~ •. EnWonnw!tll IIP345 0cnot. UK En...... ,_tll VP361 0cnat. UK EnWonnwltll IIP331 Ocnat, UK EnW<>nmentll VPEC1 Ocnat, UK EnW<>nmenllf I/PSC1 Ocnat, UK EnWonnwllll 6316 J- c.1lcll Figure 7.3: Dendrogram based on UPGMA clustering or normalised tDNA-ILPs or V. parohaemolyticus obtained with primers T3A and TSA. At least 4 bands were present in all isolates (except 6316), varying in size and intensity. Clusters tDI-A to tDI-F are based on a 90% similarity score indicated by the dotted line. Scale above banding profiles indicates approximate sizes of amp Iicons . 148 E ~700bp ~350bp ~200bp 1 2 3 4 ~IOObp ..... 00 ~ ~ ~700bp ~350bp ~200bp ~IOObp 5 6 7 8 ~700bp ~350bp ~200bp ~IOObp Figure 7.4: tDNA-PCR fingerprints of 56 isolates of V. parahaemolyticus and related species of clinical and environmental origin using primers TJB and T5A. Lane marked(-) is no DNA control and each lane is marked with the corresponding isolate at the top. NCIMB and NCTC reference strains for V. parahaemolyticus are 1902 and VP I0884 respectively. Strains 11327 and 8042 are NCTC reference strains for V. vu/nificus and V. cho/erae respectively. ln each gel 50 bp ladders were used, sizes of selected markers are indicated by arrows on the right. Insert (E) shows the eight tD2 banding types identified using primers T3Bff5A. 149 liOiate Loattion Sotlrce 102 duster % Similarity 4JOI>p J70bp JSSbp 20Sbp I I I I VP8 Soullla'npm, UK Envi'onmonlol VPil ~ . UK EnWor'llW1Ial VP10 Soullla'npm, UK Envi'onmonlol VP8 Soullla'npm. UK Envi'onmonlol VP7 Soullla'npm, UK Envi'onmonlol VP1 SOUINr'f1llan, UK Envi'onmonlol VP4 Soullla'npm. UK Envi'onmonlol VP2 Soullla'npm, UK ~ VP3 Soullla'npm, UK Envi'onmonlol E154482 PHLS, UK Clnital 102311 Splin Envi'onmonlol \IP34& OcrHI, UK Envi'onmonlol 102313 Splin Envi'onmonlol 102o4/5 Splin Envi'onmonlol 110319 Splin ~ 1022/10 Splin Envi'onmonlol 225 ~ ~ 22e N.._ Clnital 227 ~ Clinic* A 5421 J- Clnital 220 ~ Clilicll 222 ~ Clnital 223 ~ Cllnicol 221 ~ Clinic* 8134 J- Clnital 228 ~ Clnital 2438 J- Clnital VP1oe&l PHLS, UK NCTC R...... S,.., Vl>1 ~ . UK~ Vl>10 Soullla'npm, UK E~ Vl>3 Soullla'npm, UK ~ VI>& Soullla'npm, UK EnWormontal Vll6 ~ . UK Envi'onmonlol Vl>7 ~ UK EnWonmon1o1 VI>& ~ . UK~ Vl>2 ~. UK Emtruvnontol 219 ~ Clnital 224 ~ Clinic* }-B E1551155 PHLS, UK E168143 PHLS, UK 102115 Splin EnW1wnontaf c 1902 PHLS, UK =NCIMB Rlforw1ce- } VP331 OcrHI, UK VP345 OcrHI, UK En- VP381 OcrHI, UK -~ }n VPEC1 OcrHI, UK VP491 Dcnol, UK E...... ,_ VPil39 Dcnol, UK - VP349 Oone1, UK EnW1wnontaf VP350 OcrHI, UK - VP3n OcrHI, UK ErMronmonlol }E VP418 OcrHI, UK - VP378 OcrHI, UK EnYnwvnon111 2053 -Clnital J"*' }-F VPSCf OcrHI, UK 8318 J"*' -Clinic* -G Figure 7.5: Dendrogram based on UPGMA clustering of normalised tDNA-lLPs of V. parahaemolyticus obtained with primers T3B and T5A. Up to eight bands were present in most isolates (except 6316), varying in size and intensity. Clusters A to G are based on a 95% similarity score indicated by the dotted line. Scale above banding profiles indicates approximate sizes of amplicons. 150 ~700bp ~350bp ~200bp ~IOObp ~700bp ~350bp ~200bp ~IOObp .. - ...... · .- . ~.. : . ~700bp ·I_. ,t_. a·_. • " _ . _. ... ---~Jlg·• _a·_f_· · · ••it.,e& ·iJ ~..- -- ~= --- _. ~- ..1 __· . D 2 -~ .._ : . - :: ~350bp -..... ---~-. =- ..... ------·.. --- - ~200bp ~IOObp ~700bp ~350bp ~200bp ~IOObp Figure 7.6: tDNA-PCR fingerprints of 56 isolates of V. parahaemolytlcus and related species of clinical and environmental origin using primers TJA and TJB. Lane marked(-) is no DNA control and each lane is marked with the corresponding isolate at the top. NCIMB and NCTC reference strains for V. parahaemolyticus are 1902 and VP10884 respectively. Strain 11327 and 8042 are NCTC reference strains for V. vulniflcus and V. cholerae respectively. In each gel 50 bp ladders were used, sizes of selected markers are indicated by arrows on the right. 151 Isolate Loutlon Souce tDl cluter % Similarity 7llllp 41111p 24Sbp ' ' ' f ~ IIP1 Soi.Chon1*>n. lJ( ~ IIP7 Soi.Chon1*>n. lJ( Elwlr- E1544. PHtS, Ul( c.,_ }A 2QO ~ ~ E1881. PHtS, Ul( Clnk:ol IIPEC1 o.r-. Ul( ~ 5421 ~ Clnk:ol 8134 .-.. Clnk:ol m ~ Clnk:ol 223 Nolwo\' c.- 225 ~ Clnk:ol 220 ~ Clnk:ol 221 ...._ Clnk:ol 22oC ~ Clnk:ol IIP2 Soi.Chon1*>n' lJ( E- IIP3 ~ . lJ( \/PI Soi.Chon1*>n. lJ( ~ B IIP8 -...... lJ( - IIP8 -...... UI( ~ IIP4 -...... lJ( -E- 21V Nolwo\' Clnk:ol 1111>10 -...... lJ( f- WJ5 -.-. lJ( EJMrortmnol Wlll -....,, UK ~ ~ -....,, lJ( 111>7 --. lJ( \/PlO Soi.Chon1*>n. lJ( - Vlb1 Soi.Chon1*>n. lJ( -~ V1b2 -....,, lJ( -~ V1b3 Soi.Chon1*>n. lJ( 102311 Si>oln - 102313 Si>oln - o.r-. Ul( fiMr- £1558. PHtS, lJ( -Clnk:ol -IIP34V D 102115 Spolft ~ -~-~ -8318 Clnk:ol .-.. - I Figure 7.7: Dendrogram based on UPGMA clustering of normalised tDNA-ILPs of V. parahaemolyticus obtained with primen T3A and T38. Up to 10 bands were present in all isolates (except 6316), varying in size and intensity. Clusters tD3-A to tD3-I are based on a 95% similarity score indicated by the dotted line. Scale above banding profiles indicates approximate sizes of amplicons. 152 r = 75% eo 70 so 9 0 100 tD2 typing(% similarity) r = 66% . ; ..~: ~.:. .· :~:. --:.~r/ . eo 90 tD3 typing (% similarity) r = 82% ·~·g 6l "i3 .:::; V) ~ .....00 . :::; I< p. .G· 6 0 70 80 90 100 tD3 typing (% similarity) Figure 7.8: Similarity plots of tDNA-PCR fin gerprint types with three primer pairs. Congruence of fingerprint types tDI/tD2, tD I/tD3 and tD2/tD3 were 69.9%, 47.2% and 74.6% respectively. The linear regression correlation, r, is also calculated and expressed as a percentage fit (BioNumerics, Applied Maths). 153 Isolate Location So urce tO-combined cluster % Similarity \! l! ¥ ' ' ' ' 102415 Spain Environmental A VP939 Dorset, UK Environmental 8 E154482 PHLS. UK Cfinical E155855 PHLS, UK Cfinical VP2 Soulhamplon, UK Environmental VP3 Soulh81l1J!on, UK Environmental VP4 Soulhamplon. UK Environmental VP6 Soulhamplon, UK Environmental VPB Soulhamplon. UK Environmenlal VP9 Soulhamplon, UK Environmenlal VP10 Soulhamplon, UK Envirorvnental c VP1 Soulhamplon, UK Envlronmenlal VP7 Soulh81l1J!on. UK Enllironmental Vib3 Soulhamplon, UK Environmental Vob5 Soulh8f1'lllon, UK Environmenlal Vob1 Soulhampton, UK Environmental Vib2 Soulhamplon, UK Environmental E168143 PHLS. UK Cinical VP346 Dorset. UK Environmental VP349 Dorset. UK Environmenlal VP350 Dorset, UK Environmental 102311 Spain Environmental 102313 Spain Environmental 110319 Spain Environmental D 102V10 Spain Environmental VP491 Dorset, UK Environmental VP377 DorseL UK Environmenlal VP378 Dorset. UK Environmental VP416 DorseL UK Environmental Vib6 Soulhamplon. UK Environmenlal Vib7 Soulhampton. UK Environmental Vib10 Soulh8f1'lllon, UK Environmental }E Vib8 Soulh8f1'lllon, UK Environmenlal 222 Norway Clinical 223 NO/Way Clinical 220 NOIWay Clinical 221 Norway Clinical 224 Norway Clinical 6134 Japan Clinical 219 Norway Clinical F 226 Norway Clinical 227 Norway Clinical 225 Norway Clinical 228 Norway Clinical 5421 Japan Clinical VP10884 PHLS. UK NCTC Reference Strain 2436 Japan Clinical G 102115 Spain Environmental } H 1902 PHLS, UK NCIMB Reference Strain 2053 Japan Clinical VPEC1 Dorset. UK Environmenlal VPSC1 Dorset, UK Environmental } VP331 Dorset, UK Environmental VP345 Dorset, UK Environmental } K VP361 DorseL UK Environmenlal 6316 Japan Clinical L Figure 7.9: Combined dendrogram of composite tDI, tD2 a nd tD3 fingerprint data. 12 clusters tO combined -A to L are based on a 95% similarity score indicated by the dotted line. 154 Primers T3AffSA (tDl): 1 2 2 2 1 2 2 1 2 E1S58SS VPJSO VPI0844 Primers TJBffSA (tDl): 2 1 2 2 2 2 2 VPI0844 Primers T3Aff38 (tD3): 2 2 2 1 2 2 1 2 1021/S EISS8SS VP3SO VPI0844 Figure 7.10: Reproducibility of tDNA-PCR fingerprints from six isolates in repeat tests J and 2 with all three primer pairs. Arrows indicate common amplicon of approximately 355, 370 and 270 base pairs in fingerprint types tDI , tD2 and tD3 respectively and indicate relative migration ofbands on each gel. 155 Isolate Species Location Source 730bp 410bp JSSbp 20Sbp I I I I 11 228 V . .,...-.,olyticut Notway Clilic* 5421 V . .,..._olyticut Jopen Clilic* 2053 V. peno-..otj4icla Jopen Clilic* E155855 V. ~olyticuo PHLS, UK Clinlcel \11)1 V. ~olyticut sout\lmpton, UK Enwonmental VP1 v. .,.,.-..olyticuo sout\lmpeon, UK EnW'Onmenlel 1802 V. ~olyticut PHLS, UK NCIMB R-..oo S~ 2438 V. .,_.-.,olyticuo Jopen Clilic* VP101!84 V. ~olyticut PHLS, UK NCTC Reterwnoo s...n 102~ V. l*ai\Mmolyticuo Spaln EnWonn1-l VP345 V. ~olyticuo Do KJOBVV V. wlnillalt ~UK EIMronmonlll VanCM1 04 v. •nguillerum Plymouth ErMronmonl81 EA5 E. MrOgonel L.D365 L domMIM DorMt, UKUK EIMronmonlll \IC4 V. chol..... Sou!hampton, UK ErMn>nmon181 VC5 V choleree sout\lmpton, UK EIMronm-1 VC3 V. chol.,.. sout\lmpton, UK EnWonn1onl81 11042 V chol- PHLS, UK NCTC Relll.- Snin VC1 V. chol..... sout\lmpeon, UK ErMronmonlal VC2 V chol.,_ sout\lmp!On, UK EnWnnmonlal EC5 E.coli I/M4119 V minlc&J• Oo Figure 7.11: Dendrogram showing relatedness based on UPGMA clustering of tDNA-ILPs of Jl. parahllemolyticus and 8 related species. tDNA consensus primers T3A and T5A were used and data analysed with a curve based Pearson product-moment correlation coefficient. 156 7.4 Discussion To the author's knowledge, this chapter is the fust study that assesses tDNA-ILP typing for V parahaemolyticus by comparing and evaluating the performance of three pairs of tDNA primers. The use of these primers provided a simple and largely reproducible PCR based approach to subtype clinical and environmental isolates of V parahaemolyticus. This created divisions between isolates that had previously been shown to be biochemically and serologically indistinguishable and where clinical and environmental isolates could not be grouped by other means (Chapters 4 and 6). The method developed was applied to a total of 36 environmental and 20 clinical isolates sourced from the UK, Spain, Norway and Japan as well as isolates from other species. Based on the present study, the identity of one clinical V parahaemolyticus isolate, 6316, remained questionable. The tRNA genes occur in multiple copies dispersed throughout the bacterial genome in most species. The shared sequence motifs of tRNA genes mean that the use of primers that contain consensus tRNA sequences in the PCR are likely to result in a number of characteristic PCR products (Welsh & McClelland, 199la). In the current study, the use of consensus primers enabled the subgrouping of V. parahaemolyticus isolates into distinct clusters based on a similarly score of greater than 90%. These groups, although not identical, fared well in concordance studies with correlation greater than 66%. 7.4.1 Fingerprinting based on tDl, tD2 and tD3 Up to four bright bands and a further five faint bands were common in most isolates with tDl and these are representative of the multiple copies of tRNA genes found within clusters, which in turn are arranged within tandem repeat units. These bands can be generated from both within and between these multiple repeat regions (Honeycutt et al, 1995). Upon closer inspecti on, it can also be seen that variation lies mainly in the relative sizes of the fragments and intensity of some of the bands rather than in the presence or 157 absence of bands, except for one clinical isolate, 222. Here, two fajnt bands were absent altogether, suggesting the absence or sequence variation of primer binding sites present in all the other strains. The absence of such bands also highlights a major drawback in the use of consensus tDNA primers in producing a robust test. Whereas mismatches between primer and target sequence can give ri se to variation evident by PCR, the same mismatches compromise the stability and hence the reproducibility of faint bands evident in some of the repeat tests. However, a trade-off between reasonably reproducible results and ease of use formed the basis of the present study. With the large degree of similarities in banding profiles (similarity value of 75% accounting for 89% of all isolates), it was decided to cluster tD 1 groups at 90% similarity values, resulting in large clusters and only seven divisions but interestingly formed an entirely new clinical division that had previously not been clustered using PFGE. Of this group, four isolates were of the pandemic serotypes, 03 :K6 (isolates 220, 222 and 223) and 01 :KUT (isolate 221) . Isolates of these serovars have repeatedly shown a high degree of clonality in previous studies (Okuda et al., 1997b; Chowdhury et al, 2000b; Matsumoto et al., 2000; Laohaprertthisan et al., 2003). Based on tDNA-PCR it appears that these clonally related groups fell within a group of 13 clinical isolates of diverse serovars forming a unique cluster. However, as all 13 isolates were from the same gel (Fig. 7.2: Gel B), careful interpretation is necessary to determine whether this grouping could be attributed to close banding similarities or due to the inability of BioNumerics to differentiate isolates on the same gel, especially at such hi gh similarity values. A further criticism of the method developed is the use of an internal marker based on common bands in a gel in the BioNumerics normalisation process. Additional external markers (i.e. molecular size markers within the gel) would have been preferable, but was omitted from the experimental design as the author was unaware at the time that data would be analysed in BioNumerics. These errors in experimental design need to be addressed in future studies 158 to validate such groupings. Nonetheless, a further two clonally related 0 3 :K6 isolates (5421, 2053, Table 4.3) from cluster tD 1-C showed a similar degree of relatedness to each other and were on the same gel, but failed to be clustered in tO 1-D. Group specific PCR (GS-PCR) markers for the post 1995 clonal pandemic strains (Bhuiyan et al. , 2002)., have been shown to be positive for eight out of the 13 isolates incJuding the two clinical isolates oftDl-B (Pers comm: S. Wagley). Environmental isolates of tO 1-F all ori ginating from nearby oyster (Crassostrea gigas) samples showed similar profil es. However, two isolates VP491 and VP939, isolated from the same region but different sources: scallop (Pecten maximus) depuration and mussel (Mytilus edulis) samples respecti vely, were mapped differently in the dendrogram. The observation that isolate VP49 1 clustered with environmental isolates from Spain is particularly interesting since environmental strains of V parahaemolyticus have been known to exist as genetically heterogeneous populations in comparison to clinical isolates (Chapter 5; Wong et al., 1999a; Martinez-Urtaza et al., 2004). Despite isolates being obtained from different geographical regions, intergenic regions, at least with primers T3A and T5A, between tRNA gene clusters appeared the same. Whereas dendrogram analysis seemed to have differentiated tO 1 isolates based on small differences in fragment sizes and banding intensity of four amplicons, primers fo r tD2 produced eight unique fingerprint patterns but only a single intense band. As mentioned earlier, faint bands did not always appear consistent in repeat comparisons of isolates in tD2 and therefore it is presently not clear whether such banding patterns could be used as reference types. It is also worthy of note that only two repeats were compared to assess the reproducibility of tDNA-PCR. Clearly, a further series of repeats, together with more accurate band sizing techniques would be necessary to substantiate the eight t02 types described. Additionally, by combining three or more repeats in BioNumerics to produce an 159 averaged curve based banding pattern, it may have been possible to reduce inconsistencies of band intensities and to overcome this drawback of the use of consensus primers. With large clusters based on high percentage similarities, empirical assessment of concordance in clusters between tD 1 and tD2 became difficult. It became apparent that although some isolates from the groups of two fingerprint types were concordant (as expected with such large clusters), others were not. A scatterplot comparing the values of percentage similarity between two fmgerprint types demonstrated the skew of data towards high matrix similarity figures. A congruence value approaching 70% between tO I and t02 (regression = 75.0%) suggested a high degree of concordance between tDI and t02 fmgerprint types; however, it is not known to what extent this could be attributed to isolates being placed on the same gel in each of the fingerprint types and fm1her emphasises the need for random distribution oftDNA-PCR products between agarose gels. It is of particular interest that the Norwegian isolates that had previously been grouped together, largely in one main group (clusters D and A for fingerprint types tO I and tD2 respectively), was found split in two distinct groups (clusters B and D) suggesting some discrimination of clonally related isolates in t03. Concordance studies of t03 with fingerprint types tDI and t02 provided some confirmatory results. Whereas tDI and tD3 appeared the least congruent, fmgerprint types t02 and tD3 exhibited a greater degree of concordance. These results may reflect that greater variation in tD 1 and least variation in tD2 (55 V parahaemolyticus isolates showed > 66% similar in tDI , >92% similarity in t02 and >88% similarity in t03) could be partially responsible for the congruence values in the three fingerprint types. A combined dendrogran1 of composite data, as expected, produced a greater number of clusters at 95% similarity than tD2 or tD3, but failed to give more discrimination than tDl. Interestingly, in congruence tests (Chapter IO, Fig. I0.8) it appears that the composite data, tO-combined, is more congruent with tO I than t02 or 160 tD3 and thereby confirming primers T3A and T5A for better discrimination of V parahaemolyticus. 7.4.2 Consensus and specific tDNA primers The primers designed by Welsh and McClelland were based on consensus tRNA sequences from a number of bacterial species (Welsh & McClelland, l991a, 1992). Preliminary work investigated the potential development and use of specific V parahaemolyticus tRNA primers. Based on availability of a single genome sequence, it was found that tDNA sequences between tRNAs for different amino acids varied considerably. Copies of the san1e tRNA genes were not found clustered adequately to yield an appropriate number and size of intragenomic length polymorphisms. BLAST analysis of consensus primer sequences designed by Welsh and McClelland identified significant homologies to the tRNA genes of V parahaemolyticus. Of the three primers (T3A, T3B and T5A) that were investigated by BLAST searches, three, six and ten tDNA sites were found to be homologous to the primers T3A, T3B and T5A respectively, with 12 to 17 bp homology and only a single bp mismatch. It is likely that those were the priming sites and it is therefore unlikely that the use of specific tDNA primers would have been able to further differentiate V parahaemolyticus. 7.4.3 General considerations from tDNA-PCR typing This chapter has demonstrated that tDNA-PCR can serve as a reasonably discriminatory epidemiological typing method for V parahaemolyticus that is easy to execute and relatively cost effective. As the method is less time consuming than others such as ribotyping, tDNA-PCR could be implemented to screen a larger number of environmental samples. However, several important considerations need to be made. Firstly, as with other typing systems, the stability of the tDNA-PCR needs to be assessed. The stability of a typing system is defined as the ability to express constant markers over time and 161 generations, and the method should be stable for at least the time fran1e of the investigation (Struelens, 1998). Length polymorphisms of tRNA spacer regions may not necessarily be coding regions and hence their stability could be questioned (Burr et al., 1998). Other than identifying isolates in distinct clusters in independent fingerprinting experiments, this study has not fully assessed the stability of the tDNA-PCR technique over a period of time, and this should be explored iftDNA-PCR is to be implemented for routine monitoring. Reproducibility is a second factor for consideration. In fingerprint types tD 1, tD2 and t03, all but one isolate (6316) was clustered together at similarity coefficients of 67%, 92% and 89% respectively. The presence of intense bands common to most isolates in repeat tests demonstrated a fairly good degree of reproducibility of this technique. An inter-laboratory study that incorporated capillary electrophoresis in the identification of Streptococcus spp also concluded reproducibility in tDNA-PCR (Baele et al., 200 l ). These findings are consistent with those of several other workers (Baele et al., 2000; De Gheldre et al., 2001; Clementino et al, 2001, Lopes et al., 2007). In addition to stability and reproducibility, the discriminatory powers of tRNA primer sets and typeability are of major importance. Clearly, the three sets of primers grouped isolates at different similarity coefficients and gave different groupings. Primer set tO 1 was most discriminatory and tD2 least discriminatory based on a few amplification products. Ease of data interpretation is another factor that is applicable with tDNA-PCR. Almost all typing experiments require the use of data analysis software. The large numbers of groups obtained are the result of sophisticated data manipulation, normalizations and curve calculations. Clearly, without the aid of such fingerprint analysis software, such groupings would not have been possible. 162 7.4.4 Typing of V. parahaemolyticus based on tDNA spacer regions As with 16S - 23S ITS-PCR (Chapter 6), tDNA-PCR discriminates between isolates by measuring molecular variation within defined regions, i.e. the tRNA gene spacer regions both between and within tRNA gene dusters. Up to nine an1plicons were generated using the various prin1er pairs, and in each case, variation was found mainly in the size and intensity of the bands produced, although the number of amplicons was the same in each fmgerprint type in most isolates. This could suggest that numbers of tRNA genes remain the same in most isolates. A total of 126 tRNA genes in V parahaemolyticus (in strain RIMD) is relatively high compared to that found in E. coli K 12, E. coli 0157 and S. typhimurium (86, 99 and 85 for the three strains respectively) and this may correlate with the comparatively high ribosomal operon number, reflecting V parahaemolyticus fast growth properties (Makino et al., 2003; Withers et al, 2006). However, from tDNA-PCR fmgerprint data alone, it was not possible to deduce possible variations in copy numbers of tRNA genes an1ongst V parahaemolyticus. As both methods of spacer regions analysis (tDNA-PCR and ITS-PCR) were looking at specific genomic region, it was logical to examine the isolates further, through a method which gives access to the entire genome. Although PFGE studies aimed to establish this, a lack of comprehensive typing data resulted in the development of a sensitive RAPD method described in Chapter 8. 163 Chapter 8 RANDOMLY AMPLIFIED POLYMORPHIC DNA (RAPD) METHOD FOR DIFFERENTIATING VIBRIO PARAHAEMOLYTICUS 8.1 Introduction Randomly amplified polymorphic DNA (RAPD) is a PCR based technique where short primers (usually 10-mers) of arbitrary sequence are used to generate an1plification products characteristic of the template DNA (Williams et al., 1990). The method is unique in that it requires no previous knowledge of the target genome sequence, requires only small quantities of template DNA, can be analysed with standard electrophoresis methods and be used for any organism. The simplicity of this arbitrarily primed PCR method (also known as AP-PCR) has enabled this technique to be adopted in a wide variety of applications such as determining parent-sibling connecti ons (Hadrys et al, 1993), identification of genetic alterations in tumours (Dil-Afroze et al. , 1998) and the detection of UV induced genotoxicity (Atienzar et al. , 2000) . In bacteriology, RAPD analysis has been successfully used to differentiate between strains of a wide variety of species including Renibacterium salmoninarum (Grayson et al., 1999, 2000), Burkholderia pseudomal/ei (Leelayuwat et al., 2000) and Vibrio vulnificus (Ryang et al., 1997). More recently, RAPD was used to differentiate clonally related V cholerae that could not be discriminated by ribotyping or PFGE (Pourshafie et al, 2007). Wong and colleagues (1999c) described the use of RAPD to establish relationships among 308 clinical isolates of V parahaemolyticus from food poisoning outbreaks in Taiwan. A total of 41 polymorphic RAPD patterns were identified based on the presence of up to 10 bands that were grouped into 16 types (Wong et al., 1999c). Earlier work examined the genotypes of urease positive clinical strains of V parahaemolyticus (Okuda et al., 1997a), 164 demonstrating that RAPD was able to differentiate between 04:K12 serotypes and non- 04:K12 serotypes in a sample of eight clinical isolates. In a further report, the clonal relationship of 03 :K6 isolates from Calcutta and Japan was demonstrated with RAPD (Okuda et al., 1997b). Both of the studies by Okuda and colleagues used single primers PI and P2 and, as expected, clonally derived isolates yielded similar comparative results. However, both single primers produced a mere five major amplicons and several faint bands that were hardly discernable from background PCR noise (Okuda el al., 1997a, 1997b). These primers, together with primers P3, P4, PS, P6 (Amersham Biosciences) were investigated by Maluping et al., (2005) who were only able to show a maximum of between 2 to 4 bands with primers P4 and P6 in environmentally acquired isolates and thus it was difficult to obtain sensitive discrimination of data. To date, there has been no attempt to assess the ability of a RAPD method to differentiate pathogenic isolates of V parahaemolyticus from environmentally acquired samples. Previous work that employed consensus ITS and tRNA gene primers showed up to eight distinct amplicons (Chapters 6 and 7 respectively) and enabled the grouping of isolates. However to further assess relatedness between isolates and to allow a greater probability for a possible discrimination in relationships, it was necessary to produce a more complex banding pattern. This chapter describes the use of combined RAPD primers that have been used in both bacteria (Mycoplasma spp., and E.coli), and plants (Brassica napus, and Hordeum spp.). Combined primers have been shown to demonstrate greater polymorphisms in banding than single primers alone (Hu et al., 1995; Reddy & Soliman, 1997; Charlton et al., 1999; Hopkins & Hilton, 200 I). The following are investigated in this chapter: i) screening of commercially available RAPD pnmer patrs for complex and reliable banding patterns 165 ii) analY.sis of RA:PJil fingerprinting· in V parahaemolyticus based:on tw:o1 di'ffererit iprill11!t setf)'for co111parison and ,reliability ofCilit!t (J~oCillcedJ - . - --· ·i}O: an. evaluation [of,the: ability ;to· discriminate clinlcal.and :environmental isolates of V •parahaemolyticus based ion RAPD fingerprinting, 1•66! 8.2. Materials and Methods 8.2.1 RAPD - PCR of V. parallaemolyticus RAPD analysis was performed on 56 isolates of V. parahaemolyticus usmg RAPD Analysis Beads (Amersham) containing 6 distinct random 10-mer primers as shown in Table 8.1. Each bead, when resuspended in a 25 ~-tl volume contained 1 u Amplitaq DNA polymerase and Stoffel fragment, 0.4 mM of each dNTP, 2.5 11g BSA and buffer [3 mM MgC12, 30 mM KCI and 10 mM Tris, (pH 8.3)]. To each reaction bead 25 pmol (5 ~-tO of the appropriate primer and DNA template of appropriate volume equivalent to 10 ng, were added to generate optimum banding pattems upon agarose gel electrophoresis. The PCR reaction conditions were as follows: 1 cycle at 95°C fo r 5 min, fo llowed by 45 cycles at 95°C for 1 min, 36°C for 1 min and 72°C for 2 min. All RAPD products were analysed by electrophoresis of 15 ~-t l of each reaction mixture in 2% agarose gels as described in section 3.6. Table 8.1 : RAPD Primers used in this study (Amersham Biosciences). RAPD Primers Primer Sequence %GC PI 5' -d[GGTGCGGGAA]-3' 70 P2 5'-d[GTTTCGCTCC]-3' 60 P3 5'-d[GTAGACCCGT ]-3' 60 P4 5'-d[AAGAGCCCGT]-3 ' 60 PS 5'-d[AACGCGCAAC]-3' 60 P6 5'-d[CCCGTCAGCA]-3' 70 8.2.2 Screening of RAPD primers Pl to P6, singly and in pairs Genomic DNA (section 3.3) from a panel of six isolates of V parahaemolyticus was screened with single primer, and in all possible paired combinations (15 reactions) with primers PI to P6. In order to determine the best discriminating banding patterns. Subsequently, primer pair P2 and P3 (designated primer set 1, PSI) was used on a larger panel of 18 isolates, before they were applied to all 56 isolates used in the study. Primer pair Pl and PS (designated PS2) were also used as a duplicate set of experimental results to assess the reliability ofRAPD clustering used in this method. 167 8.2.3 Dendrogram analysis with BioNumerics Dendrograms showing relationships amongst clinical and environmental isolates of V parahaemolyticus were generated using BioNumerics Software with modules for Fingerprints and Cluster Analysis (version 5.1, Applied Maths, Belgium) as explained in section 6.2.3. All banding patterns for PSI and PS2 were used for the cluster analysis of 56 isolates of V parahaemolyticus and to make comparisons between the two sets of data. The bands from each lane were subjected to curve calculations and analysis to exclude background noise from PCR reactions. The tracks were then normalised using two external reference positions that consisted of 100 bp molecular weight markers on either side of the gel, and a common band that appeared in several lanes as an internal reference. Up to 13 bands per lane, many of which were also common between isolates, were scored according to their presence or absence within the 100 - 2072 bp range only. Dendrograms were calculated by the unweighted pair group method by arithmetic averaging (UPGMA) algorithm using a Pearson product-moment correlation coefficient. BioNumerics position tolerance settings were I% optimisation, a position tolerance of I% of gel and faint bands were ignored. Clusters were based on 90 % similarity value in both fingerprint types, PS I and PS2. 8.2.4 Concordance between RAPD -PSl and PS2 typing The degree of concordance between RAPD-PS I and RAPD-PS2 was determined using BioNumerics software with modules for Fingerprints and Cluster Analysis (version 5.1, Applied Maths, Belgium). Values from the similarity matrices of the PS I and PS2 typing methods were plotted by BioNumerics on an x-and y- graph. Each dot represents corresponding similarity values for two isolates by the typing methods given on the x and y axes and thereby gives an indication of concordance between PS I and PS2 typing. The congruence between PSI and PS2 was calculated using the Pearson product-moment correlation coefficient in BioNumerics. 168 8.3 Results 8.3.1 Single primer (P2) screening of V. parallaemolyticus Two environmental isolates, VP350 and VPSCI, were screened with primer P2 that had previously been shown to amplify at least five prominent bands in clinical strains of V parahaemolyticus (Okuda et al., 1997b). Between one and four bands were observed; these varied considerably based on DNA template concentration and primer concentration used (Fig. 8.1 ). Clearly, the use of a single primer failed to give the complexity of banding patterns sought. 8.3.2 Two-primer screening of RAPD reactions Between four and 10 bands were observed when two primers were combined (Fig. 8.2). Out of the 15 different primer combinations, primers P3/P6, P3/P5, and P4/P4 showed the lowest number of bands, whilst most exhibited 5 - I 0 amplicons. A greater number of bands were exhibited with primers P2/P3 whereas primers PI/PS showed higher intensities of the two main amp Iicons generated. Based on these findings, primer combinations P2/P3 (designated PS 1) and PI/PS (designated PS2) were used to produce fingerprinting patterns of a panel strain of 18 clinical and environmental isolates of V parahaemolyticus (Fig 8.3). 8.3.3 RAPD typing of V. parallaemolyticus and related species with PSI A total of 56 isolates of V parahaemolyticus, plus 14 isolates of other Vibrio spp., and nine non-vibrio species were typed using PSI (Figs 8.4 to 8.8). There was 100% typeability, where all 79 isolates showed comparable banding profiles that largely fell within the range of the 100 bp size markers from 100 - 2072 bp (Fig 8.3). A comparison of RAPD-PCR reproducibility in a selection of isolates is described in section 8.3.5. Approximately nine to 20 bands of varying intensity and size were observed with V parahaemolyticus isolates with at least nine common arnplicons of various sizes. Distinct bands of around 180, 200, 280, 370, 590, 610, 860, 1000, and 1350 bp as well as several higher molecular weight 169 bands were amplified (Fig. 8.3). Significant differences in banding patterns in other Vibrio spp. were also identified. Six isolates of V cholerae showed distinct banding patterns (Fig 8. 7). Likewise, V anguillarum, V alginolyticus, V mimicus, V jluvialis, L. damsela and non-vibrios could also be differentiated by unique banding profiles (Figs. 8.5 to 8.8). A common band (Band X) of approximately 910 bp was present in 10 out of the 20 clinical isolates studied. Other than the possibility of this band present in isolate 1103/9 (Figs. 8.5 & 8.9), no other environmental isolate seemed to possess this band. However, unlike the clinical isolates, Band X appeared very faint in 1103/9. A similar sized amplicon was also identified amongst some members of RAPD Types 22 and 23, but appeared characteristically smaller (Fig. 8.6) or faint (Fig. 8.8). The 10 clinical isolates were of the 03:K6, 03:KUT or 05: K68 serovars accounting for seven, two, and one isolate respectively (Fig. 8.9). Both 03:K6 isolates from the UK (El55855 & El68143) have been shown to be related to the pandemic clones (Wagley et al., 2008). At a 90% similarity value based on dendrogram analysis, V parahaemolyticus isolates were differentiated into 30 RAPD types (Fig. 8.1 0) with clinical isolates in RAPD types (RT) I 0, 11, 12, 15, 21, 24, 26 and 27. Of these, four types only had single members (219, 224, 2436 and 6316) with the identity of isolate 6316 questioned from earlier chapters. An environmental isolate VP416, isolated from the common oyster, was clustered closely with a Japanese clinical isolate 6134. Clinical isolates from UK, Japan and Norway were clustered into distinct groups. Of the 20 environmental types, 13 isolates fell into their own types and could not be grouped with other isolates at a 90% similarity level. However, PS 1 clustered environmental isolates VP1 to VP4 together and VP6 to VP9 together into two groups (RT 20 and 16 respectively). Similarly four isolates, Vibl- Vib3 and Vib5 (RT 23) were found 170 to be more closely related than Vib6- ViblO (RT 22) from the Southampton environmental isolates (Fig 8.1 0). As with the environmental isolates from Spain, V parahaemolyticus isolated from environmental samples in the UK fell generally into a heterogenic population. 8.3.4 RAPD fingerprinting of V. parallaemolyticus based on PS2 All 79 isolates that included members of other Vibrio spp and non-vibrios were also typed using a second set of primers Pl and PS (PS). To ease analysis and comparison of data, only V parahaemolyticus isolates are shown (Fig. 8.11 ). As with PS I, there was I 00% typeability with approximately nine to 18 bands of varying intensity and size observed for V parahaemolyticus. Most isolates showed at least six common amplicons of various sizes with bands predominantly around 130, 250, 310, 590, 730 and 1130 bp (Fig. 8.11 ). At a 90% similarity value based on UPGMA dendrogram analysis, V parahaemolyticus isolates were differentiated into 24 RAPD types (Fig. 8.12) with clinical isolates present in seven RAPD types; 02, 08, I 5, 16, 18, 22 and 23. Of these, only one isolate (2436) constituted its own group (RT 16). RAPD Type 15 clustered two UK 03:K6 isolates with two Japanese 03:K6 isolates, whereas RT 22 and 23 grouped all the Norwegian isolates together. As with the previous primer set, the banding pattern of an environmental isolate, VP416, closely resembled Japanese clinical isolate, 6134. Furthermore, a common bandY of approximately 610 bp was present in 11 out of the 20 clinical isolates studied (Fig. 8.1 0). This band seemed to be present in some environmental isolates as well but as with Band X, appeared distinctly weaker. The 11 clincial isolates included those I 0 previously described with PSI that were of03:K6, 03:KUT or OS:K68 serovars, with the addition of isolate 221 of the 0 I :KUT serotype (Fig. 8.12). 171 Of the IS environmental types, six isolates were sole representatives of their own types and could not be grouped with other isolates at a 90% similarity level. As previously found with PS l, PS2 clustered environmental isolates VP 1-VP4 together and VP6-VP9 together into two groups (RT 12 and 13 respectively) with VPIO in a distinct RT of its own. As with PS l, isolates Vib 1-Vib3 and Vib5 (RT 17) grouped separately to Vib6 - Vib 10 (RT 09) from the Southampton environmental isolates (Fig 8.12). Environmental isolates from Spain fell into four types (RT 04, 05, 07 & l 0) and the remaining environmental isolates from Dorset, where clustered, fell into RAPD Types 0 I, 03 and 06 (Fig. 8.12). 8.3.5 Reproducibility of RAPD-PCR A comparison of repeat RAPD-PCR with six isolates is shown in Figure 8.13. Most of the bands were present in repeat PCRs (A - F) with variability found in both the presence or absence of weak bands and the intensity of what appeared to be the same amplicon (Fig 8.13). For example, the 610 bp amplicon was present in all six repeats of type strain 1902 but intensities were variable (Fig. 8.13: Lanes A & B; isolate 1902). At least six faint bands were absent in Lane B (reflecting Fig. 8.4) in comparison with the remaining five tests where most of these bands could be identified (Fig. 8.13: isolate 1902). Similarly, repeat tests with isolate 2436, VP345, Vib2 and 1021/5 identified most of the prominent bands with the presence or absence of 2 or 3 very faint bands differentiating amongst the repeats (Fig. 8.13). A comparison of repeats with VP2 also demonstrated the degree of reproducibility of RAPD-PCR bands. Whereas Lane B (Fig. 8.13: VP2) showed the absence of minor bands, certain bands such as the 610 bp amplicon, in contrast, appeared brighter. Minor bands were ignored in BioNurnerics cluster analyses. 8.3.6 Comparison of RAPD-PSI and RAPD-PS2 RAPD-PS I appeared more discriminating than RAPD-PS2. PS l was able to differentiate two clinical isolates (219 and 224) that existed in the same group in PS2. Similarly, PSI 172 differentiated between three Spanish environmental isolates (1022/10, 1023/1 and 1023/3) and three UK environmental isolates (VP349, VP350 and VP361) that were grouped together in Pulse Types 03 and 06 with PS2. Additionally, PS2 was able to differentiate isolates that were grouped together in RAPD-PS I. For example, clinical isolate E154482 was differentiated from the other two clinical isolates in RT 21 of PS I. A comparison of the two typing methods where the values from the similarity matrices were plotted on x and y axes showed a congruence of 56.6% with Pearson's correlation showing some concordance between the two primer pairs (Fig. 8.14). 8. 3. 7 Differentiation of clinical and environmental isolates It was possible to differentiate clinical isolates from environmental isolates based on the RAPD patterns shown (Figs. 8.10 and 8.12). The two markers amongst clinical isolates, Band X and Band Y, account for 50% of the clinical isolates studied. Band X was approximately 910 bp in size and BandY was approximately 605 bp length (Figs. 8.9, 8.10 and 8.11 ). Isolates were considered positive for bands X and Y if they appeared distinct and intense. However, with a large number of RAPD types, there was no single unique banding pattern or the presence or absence of a band that could clearly distinguish pathogenic V parahaemolyticus. 173 1 2 3 4 5 6 7 8 9 10 M t- 1000 bp t- 600 bp t- 200 bp Figure 8.1: RAPD single primer (Pl) screening of V. p11raltaemolytiCNs environmental Isolates, VP350 and VPSCI. Lanes correspond to the following: I and 10; Control DNA (E. coli: BL21 and Cl a respectively). Lanes 2, 3, 4 and 5; VP350 {12 ng DNA) with 0, 10, 20 and 30 pmol primer respectively. Lanes 6, 7, 8 and 9; VPSCI (25 pmol primer) with 0, 15, 30, 45 ng of template respectively. Lanes M, 100 bp Ladders (Gibco). Arrows indicate non-specific PCR products typical ofRAPD reactions. t• Primer: M Cl C2 PS PS P4 P4 P3 P3 P3 P3 P2 P2 P2 P2 P2 PI PI PI PI PI M 2od Primer: P6 P6 P6 PS P6 P6 PS P4 P6 P6 PS P4 P3 P6 PS P4 P3 P2 600bp7 IOOOb 600bp 200 bp PSI PS2 Figure 8.l: RAPD double primer screening of V. paraltaenwlytlciiS environmental isolate, VPJSO. Primers PI to P6 were used in pairs for RAPD PCR. Lanes M, 100 bp Ladders (Gibco). Lane Cl and C2 are no DNA template and no primer controls respectively. Primer PS I and PS2 were used for further studies. 174 •M I I 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 M2 - c•)IC·~ .". ·. ;"7·:- ,_.;.. I!._· .. . : -.. . - - - e; - - . . - ~ . f.;. .._ - . ... t -1=-- I : '' Ill " ...... ' •.~~ -=: -. .... 1-. ::.. =. ~~-.--.,Ji-• .. ---- - ~ 600 bp ~ aill... ~ · "- !". - • ~-• ~1- ·1' · .:. _ ,'·· •~· •~ .. s-~. - .:._ - -- -· ...... ·-- . . ~ 200bp Figure 8.3: RAPD analysis or a panel strain or 16 representative clinical and environmental isolates or V. parallamwlydcus with primers Pl and P3 (PSI). Lanes I and 2: no template and no primer controls respectively. Lanes 3 to 20 correspond to the following isolates:- 3: E coli BL21 (control DNA); 4: V. anguil/arum CMI04; 5: VPSCI; 6:VPECI; 7: VP939; 8: VP377; 9: VP350; 10: VP345; 11: VP7; 12: VP4; 13:VP3; 14: VP2; 15: VPI; 16: 5421; 17: 2436;18: 6316; 19: 1021/5; 20: NCIMB type strain 1902. Lanes M1 and M2, IKb and 100 bp ladders respectively, with the sizes of selected markers indicated by the arrows. M I 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 M ~ 1000 bp ~ 600 bp ~ 200bp Figure 8.4: RAPD banding patterns or Norwegian cUnicallsolates or V. parahaemo/yticus with PSl. Lane 1; no template control. Lanes 2 to 20 correspond to the following isolates: 2; NCIMB reference strain 1902, 3; 219, 4; 220, 5; 221, 6; 222. 7; 223, 8; 224, 9; 225, 10; 226, 11; 227, 12; 228, 13; NCTC reference strain VPI0884, 14; 2436, 15; E155855, 16; 2053, 17; VP345, 18; 1021/5, 19; Vib2, 20; VP2. Lanes M, 100 bp ladders with the sizes of selected markers indicated by the arrows. 175 M I 2 3 4 5 6 7 8 9 10 I I 12 13 14 15 16 17 18 19 20 M ~ 1000 bp ~ 600bp ~ 200 bp Figure 8.S: RAPD banding pattems ofUK and Spanish environmental isolates of V. parlllftlemolydcus with PSI. Lane 1: no template control. Lanes 2 to 20 correspond to the following isolates:- 2: NCJMB reference strain 1902 (internal control); 3: 1021/5; 4: 1022/ 10; 5: 1023/ 1; 6: 1023/3; 7: 1024/5; 8: 1103/9; 9: VP378; 10: VP33l; 11 : VP345; 12: VP346; 13: VP349; 14: VP350; 15: VP361; 16: VP377; 17: Proteus mirabilis; 18: Salmonella typimurium; 19: E. coli EC5; 20: E.coli Cl a. Lanes M, 100 bp ladders with the sizes of selected markers indicated by the arrows. M 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 M ~ 1000 bp Band X ~ 600bp ~ 200 bp Figure 8.6: RAPD banding pattems of Japanese clinical isolates and UK environmental isolates of V. portllftlemolyde~~s with PSI and related species. Lane I: no template control. Lanes 2 to 20 correspond to the following isolates:- 2: NCJMB reference strain 1902(intemal control) ; 3: 2053; 4: 2436; 5: 542 1; 6: 6134; 7: 6316; 8: V. a/ginolyticus2341 ; 9: Yib3; 10: Vib5; 11 : VP416; 12: VP49 1; 13: VP939; 14: VPECI; 15: VPSCI; 16: V. anguillarum CM104; 17: Escherichia aerogenes EA5; 18: Klebsiella aerogenes KA5; 19: Pseudomonas aeruginosa PA I; 20: E. coli BL21 . Lanes M: l OObp ladders with the sizes of selected markers indicated by the arrows. 176 M I 2 3 4 5 6 7 8 9 I 0 11 12 13 14 15 16 17 18 19 20 M ~ 1000 bp ~ 600bp ~ 200bp Figure 8.7: RAPD banding patterns of environmental isolates of V. [HiralraelffO/ytlcus, V. clrolerae and related spedes with PSI. Lane 1: no template control. Lanes 2 to 20 correspond to the following isolates:- 2: NCIMB reference strain 1902 (internal control); 3: 8042; 4: VC 1; 5: VC2; 6: VC3; 7: VC4; 8: VC5; 9: V. alginolyticusVA430; IO: VPI ; ll : VP2; 12: VP3 ; t3: VP4; 14: VP6; 15: VP7; 16: VP8; 17: VP9; 18: VPIO; 19: NCTC reference strain VP10884; 20: V. mimicus 499. Lanes M: IOObp ladders with the sizes of selected markers indicated by the arrows. M I 2 3 4 5 6 7 8 9 I0 I 1 12 13 14 15 16 17 18 19 20 M ~ 1200 bp Band X ~ 600 bp ~ 200 bp Figure 8.8: RAPD banding patterns of environmental and clinical isolates of V. paralranno/ydcus and related species with PSI. Lane I: no template control. Lanes 2 to 20 correspond to the following isolates:- 2: NCIMB reference strain 1902 (internal control); 3: V. jluvialis NCTC type strain 11327; 4: Vib 1; 5: Vib2; 6: Vib3; 7: Vib4; 8: Vib5; 9: Vib6; 10: Vib7; 11 : Vib8; 12: ViblO; 13: V. vulnificus KIDBVV; 14: V. vulnificus 11067; 15: Listonella damselea LD365; 16: El54482; 17: El55855; 18: E l68143;19: 2436; 20: 61 34. Lanes M: tOObp ladders with the sizes of selected markers indicated by the arrows. 177 M I 2 3 4 5 6 7 8 9 10 M OType KType 1000 bp Band X 800 bp 600 bp Figure 8.9: A unique RAPD marker (Band X) amongst clinical strains of V. paraht1nnolydcus. Clinical isolates correspond to the following lanes:- I: NCIMB reference strain 1902; 2: 220; 3: 221 ; 4: 222; 5: 223; 6: 225; 7: 226; 8: 227; 9: E155855; 10: 2053. Amplicon also present in isolates 5421 and E168143 (Fig. 8.6: Lane 5 and Fig. 8.8: Lane 18 respectively). 0 and K serotypes are indicated in each lane. Lanes M: 100 bp ladders with the sizes of selected markers indicated by the arrows. 178 M - 2 3 4 5 6 7 8 9 10 11 M 1000 bp BandY 600bp 200 bp M 12 13 14 15 16 17 18 19 20 21 22 23 24 25 M 1000 bp 600bp 200bp Figure 8.11: RAPD fingerprinting of 56 clinical and environmental isolates of V. partllrtlemolydcus wltb PSl. Lanes correspond to the following:- I: NCIMB reference strain 1902; 2: 219;3: 220; 4: 221; 5: 222; 6:223; 7: 224; 8: 225; 9: 226; 10: 227; 11: 228; 12: 1021 /5; 13: 1022/10; 14: 102311; 15: 1023/3; 16: 1024/5; 17: 11 03/9; 18: VP378; 19: VP331; 20: VP345; 21 : VP346; 22: VP349; 23: VP350; 24: VP361 : 25: VP377. Lane marked - is no template control. Arrow indicates Band Y found mainly in clinical isolates of V. parahaemolyticus. Lanes M, I OObp ladders with the sizes of selected markers indicated by the arrows. 180 M 26 27 28 29 30 31 32 33 34 3S M M 36 37 38 39 40 41 42 43 44 4S M BandY 600 bp 200 bp M 46 47 48 49 50 SI 52 S3 S4 SS 56 M ~IOOObp ~ 600 bp ~ 200bp Figure 8.11: Continued. Lanes correspond to the following:- 26: 2053; 27: 2436; 28: 5421; 29: 6134; 30: 6316; 31:VP416; 32: VP491 ; 33: VP939; 34: VPECI ; 35: VPSCI ; 36: VPI ; 37: VP2; 38: VP3; 39: VP4; 40: VP6; 41: VP7; 42: VP8; 43: VP9; 44: VPIO; 45: NCTC reference strain VPI0884; 46: Vibl; 47: Vib2; 48: Vib3 ; 49: Vib5; 50: Vib6; 51 : Vib7; 52: Vib8; 53: VibiO; 54: El54482; 55: El55855; 56: El68143. Arrow indicates BandY found mainly in clinical isolates of V. parahaemolyticus. Lanes M: lOObp ladders with the sizes of selected markers indicated by the arrows. 181 A B c D E F A B D F A B E A B c B F B C 102115 Figure 8.13: Reproducibility of RAPD fingerprinting profiles in sill isolates of V. parah11emo/ydcus. Lanes A to F are comparisons of independent RAPD-PCR and gel runs from Figures 8.3 to 8.8 respectively. Reference arrow on the left shows the position of a 610 bp amplicon that appeared in isolates 1902,2436, VP2, Vib2 and 1021/5 indicating relative migration of bands on each gel. Isolate VP345 is marked at a 500 bp amplicon All band sizes are approximate. 183 0 0 ., ··. ·. :;·.. ... r = 68.8% ..· .... ...... ··' ·.··. . . · .. ··. :' ·.. ·. ...... :: .. ... 0 N 40 60 80 100 RAPD -PS2 typing(% Similarity) Figure 8.14: Concordance of stratification by RAPD-PSI and RAPD-PS2 for 56 isolates of V. parahaemolyticus. Each dot represents corresponding similarity va lues for two isolates obtained by both primer sets given on the x and y axes. The congruence between the two primer pairs as calculated with the Pearson coefficient was 56.5%. Data showed a linear regression fit of 68.8%. 184 8.4 Discussion This chapter examined the feasibility of the RAPD technique to differentiate clinical and environmental isolates of V parahaemolyticus, as well as assess the reproducibility and reliability of the method used. Using pairs of RAPD primers produced a more complex banding pattern than single primer alone, as this increases the chances that two priming sites are recognised and are sufficiently close to result in a PCR amplification product (Welsh & McCielland, 1990). The method was easy to execute and the resulting large number of amplicons allowed intra-species differentiation in isolates that were previously grouped together using 16S-23S ITS and tDNA-ILPs spacer region analyses. The method that was employed also allowed Vibrio species to be clearly differentiated from one another with the exception of one isolate, 6316, that could not be grouped with other members of V parahaemolyticus. This isolate has repeatedly shown unique banding profiles in Chapters 5, 6 and 7 with a fingerprinting profile that is characteristic of V mimicus. Further tests would be required to accurately speciate this isolate. RAPD is a widely used method that is known to be sensitive in detecting small changes in the genome between isolates and is used to determine genetic relatedness both amongst species and sub-species. The use of a wide range of RAPD primers have been reported to type V parahaemolyticus, but often typing experiments are compromised by the relatively small number of bands that are used to discriminate within a typing experiment. This was shown by Maluping and colleagues and further supported by this work (Section 8.3.1 and Fig. 8.1) (Maluping et al., 2005). This study concentrated on four primers PI, P2, P3 and PS that are widely used in typing experiments, but used in pairs to develop and optimise a RAPD method that appeared reproducible and that could cluster distinct clinical and environmental groups of isolates. 185 8.4.1 Typing of V. parallaenrolyticus based on PSI Initial primer screening experiments suggested that pnmers P2 and P3 may produce banding patterns of suitable complexity. Up to nine common bands were amplified continually in most isolates of V parahaemolyticus, and this allowed confirmation of relatedness. Several of these bands were also common in related Vibrio species suggesting that this RAPD method may I) be a helpful tool to differentiate V parahaemolyticus from other vibrios and 2) be able to differentiate between members of other species such as clinically significant and clonally related V cholerae. Of the 56 isolates of V parahaemolyticus studied, PS I differentiated the isolates into 30 RAPD Types (RTs) of which I9 were environmental only, eight were clinical only, a single RAPD Type consisting of both on an environmental and clinical isolate, and two RAPD types accounted for the two reference type strains. This data suggests that both clinical and environmental isolates were discriminated almost equally with PS 1, and that RAPD can differentiate clinical isolates that have previously shown to be a more homogeneous group (Chapters 4, 6 and 7). The sensitivity of the RAPD technique has been widely documented (reviews by Atienzar & Jha, 2006; Yakubu et al., 1999). Clinical isolates of219, 224, 2436 (as well as 6316) that fell into their own RAPD types and the clustering of a Kanagawa positive (haemolytic) environmental isolate with a Japanese clinical isolate is in agreement with PFGE findings (Chapter 5). Whilst the RAPD method is sensitive, there are several criticisms of the technique. For example, whilst a complex banding profile with two primers was favoured in this study, it is likely this contributed to the presence of additional faint bands and possibly reducing the intensity of others (Fig 8.13). Weak bands have been shown to be irreproducible in most cases as identified in repeat tests. Whilst BioNumerics settings allowed the option of omitting very faint bands in cluster analysis it did not account for varying band intensities I86 inherent in the RAPD method that were used in the curve calculations. This may provide at least one explanation why the same method with two different sets of random primers failed to give a high concordance (i.e. a congruence of greater than 70%) even though isolates in both dendrograms initially appeared grouped similar from visual inspection. A further explanation for discordance using the same method but a different set of primers could be attributed to the GC content of the primers. Whereas primers P2, P3 and PS were all 60% GC, primer PI had a 70% GC ratio (Table 8.1). Invariably, such differences in% GC are likely to result in considerable changes to primer and template DNA interactions. A complex gel banding pattern also meant that agarose gels required adequate running times to obtain good resolution. In practice, consistency was difficult to achieve between different gel runs even with the use of a tracking dye and thus further reliance on BioNumerics normalisation procedures. The use of additional markers within the gel would therefore facilitate more accurate normalisation of bands and to adjust for possible gel smiling. 8.4.2 Typing of V. parallaemolyticus based on PS2 It was necessary to validate groupings and clusters determined with PS I and therefore the second primer set PS2 utilised Primers I and 5. Although a smaller number of bands were produced, a fair degree of concordance was shown between the two typing experiments. Isolates could be characterised within a cluster, forming smaller groups or identified as its own type. For example, the lower number of bands produced in PS2 meant that the ability to discriminate was less than that of PS I. However, although PS I discriminated 10 isolates that were grouped in PS2, PS2 also discriminated three isolates that were grouped in PS 1; therefore, the results of the two sets of RAPD primers were complementary rather than contradictory. Chapter 10 addresses the evaluation of RAPD typing with other molecular typing systems employed. 187 8.4.3 A marker for pandemic serovars? One of the most interesting findings of typing V parahaemolyticus with two primers was the identification of a band in 10 of the clinical isolates (Fig. 8.9) that was present using both primer sets, PS 1 and PS2. These distinct bands were found in isolates of four serovars (01 :KUT, 03:K6, 03:KUT and 05:K68) from three different geographical regions, but appeared faint in many of the environmental isolates. PFGE grouped them all under one cluster at 76% similarity that consisted of Pulse Types 18, 19, 20 and 21 (Fig 5.11) but did not manifest as a singularly identifiable band. RAPD dendrogram analysis was also able to differentiate these isolates into RAPD types 10, 21 and 24 with PS 1, and RAPD types 15, 22 and 23 with PS2. In order to confirm the identity of these bands, and whether they are truly different to those from environmental isolates, it would be necessary for further repeats, as band intensities mentioned earlier, are variable m the RAPD method. Furthermore, more reliable band sizing techniques, such as polyacrylamide gel electrophoresis, would be necessary for more accurate analysis. This should be followed by cloning and sequencing of bands to investigate the 910 bp and 610 bp RAPD markers identified in clinical isolates. It has also been proposed that the arbitrary bands produced in RAPD do not reflect real and relevant population diversity and as such it would be difficult to monitor the stability of the typing system during the time frame of the investigations (Burr et a/, 1998). 8.4.4 Comparison of single and two-primer RAPD methods RAPD typically employs the use of short I 0-mer primers. Several workers have used the I 0-mer primers PI, P2 and P4 singly to demonstrate clonal relationships in 03:K6 isolates of V. parahaemolyticus (Matsumoto et al., 2000; Okuda et al., 1997a; 1997b). Of all these primers used, P I showed the most bands and the presence or absence of one or two bands was used as criteria to differentiate the old and new 03:K6 isolates; those that appeared before and after 1996 respectively - the latter referring to the more virulent 03 :K6 types. 188 Additionally, the use of Primer 2 showed even fewer total bands and, again, the presence or absence of one or two bands differentiated the two groups in a sample of eight 'old' 03:K6 and nine 'new'03:K6 isolates (Matsumoto et al., 2000). Thus the use of a two primer method described here that generated up to 20 bands may be better able to differentiate clonally related isolates. It is unfortunate that the study did not contain a significant number of comparative old and new 03:K6 types to test this claim. PCR product s1ze with single pnmer PI has been shown to range in size from approximately 0.6 - 6.5 kbp (Okuda et al., 1997b) whereas this work has demonstrated that most of the PCR products could be resolved within 0.1 and 2.0 kbp. This is to be expected as the distance between primed sites is likely to be shorter when two different arbitrary primers are available than a single primer alone (Welsh & McClelland 1991 b). Shorter amplicons may also amplify more efficiently because of the shorter extension times required and the increased likelihood that DNA polymerase will complete strand synthesis (Hopkins & Hilton, 2001). Therefore the use of two primers would further increase efficiency to subtype V parahaemolyticus, by minimising times needed for agarose gel electrophoresis. The use of a third primer might be advantageous, but in one reaction this has been shown to be inappropriate for typing purposes as PCR product sizes were significantly smaller than one- or two-primer PCR reactions and required the use of polyacrylamide gels to resolve smaller fragments (Hopkins & Hilton, 2001). This may not be practical for routine typing purposes to process a large number of samples. Additionally, the presence of too many bands in a typing system increases the complexity of the visual interpretation leading to reliance on more sophisticated band analysis software. 189 8.4.5 Data analyses As with previous chapters, the quantitative assessment of fingerprinting profiles was necessary throughout the analysis of RAPD fingerprints. One of two methods usually employed are either a densitometric curve based approach (as used in this chapter and chapters 6 and 7) or a band based approach as used in Chapter 5. Whereas the binary nature of distance matrices in the band based approach scores bands as either present or absent, a densitometric curve analysis method takes into account the intensities of different bands produced by RAPD. Despite, variation in band intensities shown in the present study, this method has been shown to be more reliable in producing RAPD clustering than the band based Dice method (Duarte et al 2005). Additionally, the availability of fingerprint analysis software facilitated the process of determination of groupings. Nonetheless, much caution is still needed in the interpretation of RAPD data. For example, the eo-migration of bands may be a result of amplicons of the same size, yet generated from different regions of the genome. If the presence of these bands is used to infer similarity between isolates, then clearly two identical RAPD patterns may not necessarily imply they are the same strain. Therefore, when analysing large amounts of data that are used to construct a dendrogram that infers phylogenetic relationships, careful interpretation of banding patterns is required. Thus, the use of a second primer set PS2 in this chapter was necessary to confirm groupings obtained, and the larger number of amplicons in a more complex banding pattern may serve to reduce errors from eo-migration of bands. 8.4.6 General considerations of a RAPD typing system The main advantages of the RAPD method are its relative simplicity to set up and use. However, several problems may potentially hamper its use and these have been the subject of criticism of the technique. As mentioned, these include the lack of reproducibility between laboratories due to sensitivity to changes in reagents, protocols and machines, the non-universal use of primers and RAPD markers as well as the aforementioned method of 190 data analysis used (Chang et al., 2007). However, when the same method was used with different thermocyclers, comparable results for V parahaemolyticus have been achieved (Matsumoto et al., 2000). For example, RAPD profiles generated with a "Hybaid HB TRIL" and "Zymoreactor II - AB-1820" thermocyclers were used to reliably compare strains from Bangladesh with those isolates from many other countries to study the spread of pandemic 03:K6 clones (Matsumoto et al., 2000). In this chapter, internal controls using the DNA of several isolates as well as NCIMB reference strain 1902 were used. These results have shown some reproducibility and consistency in banding profiles in independent PCR reactions (Figs. 8.4 to 8.8), however more thorough inter-laboratory comparisons are required to assess the reproducibility of this two-primer method. Despite the criticisms, the initial concerns about the reproducibility of RAPD subtyping have been partially addressed by the development of more robust RAPD protocols, and in particular, standardised RAPD reagents and 'analysis beads'. Therefore, a comparison of RAPD with and without the use of analysis beads would be interesting. The RAPD method developed in this chapter allowed discrimination equal to that of PFGE, yet requiring a fraction of the time to execute. It is estimated that a total of 5 hours of DNA/PCR preparation, thermocycling and gel running times gave data comparable to the sensitivity ofPFGE with no concerns ofuntypeabilty. This would be more cost and time efficient than the recently described PulseNet USA standardised pulsed-field gel electrophoresis protocol for subtyping of Vparahaemolyticus (Parsons et al., 2007) Analysis on a larger number of samples using primers P2 and P3 would confirm the general usefulness of this method and, in particular, to determine if the markers can be used to distinguish clinical isolates. 191 Chapter 9 VARIATION IN THE TYPE Ill SECRETION SYSTEM AMONGST CLINICAL AND ENVIRONMENTAL ISOLATES OF VIBRIO PARAHAEMOLYTICUS 9.1 Introduction The type Ill secretion system (TTSS) is an apparatus used by several Gram-negative bacteria to secrete and translocate virulence effectors across the bacterial inner and outer membranes directly into the cytosol of host eukaryotic cells (Comelis & Van Gijesegem, 2000; Hueck, 1998; Ghosh, 2004). The TTSS is found in both animal and plant pathogens and is composed of more than 20 proteins that form a macromolecular assembly (TTSS I, Fig. 9.1 ). The proteins of the TTSS apparatus are highly conserved among bacteria that possess the system (Tampakaki, et al., 2004). In contrast, the effector proteins secreted via the TTSS are not conserved among those bacteria, and hence, the biological effects vary widely in target host cells (Hueck, 1998; He et al., 2004). Genes encoding the TTSS apparatus are generally found on pathogenicity islands (PAis) or virulence plasmids (Hueck, 1998; Winstanley & Hart, 2001 ). The TTSS has been shown to be especially important in the pathogenicity of Yersinia spp., Burkholderia spp., Salmonella spp., Shigellajlexneri, and enteropathogenic Escherichia coli (Hueck, 1998; Winstanley & Hart, 2001). Genome sequencing of the clinical V parahaemolyticus isolate RIMD221 0633 revealed that the bacterium possesses two sets of genes for a TTSS, termed TTSS I and TTSS2, located on chromosomes I and 2 respectively (section 2.1.2) (Makino et al., 2003). The gene organization of TTSS I was found to be very similar to that of the TTSSs reported for Yersinia species. In contrast, TTSS2 is more similar to that found in non-0 I and non-0 139 192 V. cholerae pathogenic strains and seems to contain at least essential genes for protein secretion (Dziejman et al, 2005). The distribution of those genes among various V. parahaemolyticus strains and the GC content of the DNA regions containing TTSS I or TTSS2 suggested that TTSS I is intrinsic in the species while TTSS2 has a feature of laterally transferred DNA (Makino et al., 2003). Other bacterial species also possess two or more TTSSs. For example, multiple TTSSs can be found in S. typhimurium and Burkholderia spp. (Hueck, 1998; Winstanley & Hart, 200 I). Enterohemorrhagic E. coli 0157:H7 has been known to possess two TTSSs termed LEE and ETT2 (Perna et a!, 200 I). LEE is involved in the formation of attaching and effacing lesions in epithelial cells whereas the function of ETT2 remains unknown (ldeses et al., 2005). Except for the case of the TTSSs in Salmonella spp., the roles of multiple TTSSs in a single bacterium have not yet been elucidated. In V. parahaemolyticus, TTSS I was found in all of the strains tested while TTSS2 was present only in the KP-positive strains (Makino et al., 2003). The present study was conducted to: (i) develop multiplex PCRs to assess the distribution of both TTSS systems in clinical and environmental isolates that possess the gene tdh and shown to be K+ or K- (Chapters 3 and 4) (ii) examine variation in both TTSS I and TTSS2 in V. parahaemolyticus isolates by the development of probes specific to genes ofTTSSI and TTSS2 and DNA hybridisation assays (iii) assess the relationships between clinical and environmental isolates possessing TTSS I and TTSS2 systems with phylogenetic clusters obtained from typing experiments with a view to discrimination of potentially pathogenic isolates. 193 9.2 Materials and Methods 9.2.1 PCR detection of TTSSI and TTSS2 genes Using sequence information from isolate RIMD221 0633 (Makino et al., 2003), three pairs of PCR primers spanning each of the TTSSs were designed with the aid of DNA-sis (Hitachi Software) and Amplify software (Engels, 1993). The primers were targeted to three genes on TTSS I and two genes on TTSS2 as well as a region flanking the TTSS2 genes representing the V. parahaemolyticus pathogenicity island (Vp-PAI) (Izutsu et a/, 2008) (Table 9.1, Fig. 9.1). Primers consisting of24 bases were designed on the basis of a 50% GC ratio, the presence of 2-3 guanine or cytosine bases at the 3' end of the primer where possible, a lack of secondary structure and primer dimer formation, and their proximity in relation to the desired target sequence (Table 9.2). Table 9.1: TTSS gene targets in V. parahaemolyticus used in this study. Genes vcrDI & vscCJ are components of the TISSI. Gene vcrD/ has 74% identity to Yersina sp. yscD (encoding the largest protein). Gene vscC/ has 48% identity to Yersinia sp. yscC. An absence of these genes is likely to result in defective protein secretion via the TISS l. Region Gene number Designation Function Reference of robe lTSSI VPI662 Probe A Low calcium response protein. (vcrDI) lTSS structural protein Park et al., 2004b VPI680 Probe B 'Cytotoxic effector' protein Ono e/ al., 2006 (vp/680) VPI696 Probe C lTSS structural protein Park e/ al., 2004b (vscC/) TISS2 VPAI321 ProbeD Cytotoxic necrotizing factor Makino e/ al., 2003 (vpa/32 /) V. parahaemolyticus PA I VPAI339 Probe E Putative lTSS structural protein Dziejman et al., 2005 (vscC2) VPAI355 Probe F Putative TTSS structural protein Dziejman et al., 2005 (vcrD2) The reaction conditions for PCR amplification of V. parahaemolyticus TTSS loci in separate reactions are described in section 3.6 except that 25 J.ll rather than 50 J.ll reaction volumes were used and consequently half the stated volume of each reagent was added. Details of the primers used are provided in Table 9.2. Optimised PCR cycling conditions were 2 min at 94°C followed by 30 cycles of I min at 94°C, 45 sec at 65 oc and 90 sec at 194 72°C followed by a final extension time of 5 min at 72°C. PCR products were visualised by gel electrophoresis of 6- 12 Ill sample volumes on 1.5% agarose gels, as described in section 3.8. 9.2.2 Development of Multiplex PCRs Two multiplex PCRs (M-PCR) were developed using the 6 primer pairs described in Table 9.2. A maximum of three loci per reaction were amplified. The PCR reactions were performed in a 50 Jll volume as described in section 3 .6; primers for the TTS 1 regions used an equimolar 20 pmol concentration with the same PCR cycling conditions as in section 9.2.1. Amplification ofTISS2 regions used an equimolar 30 pmol concentration of each of the primers. The M-PCR products were visualised by electrophoresis of 10 Jll volumes of the reactions in a 1.5 % agarose gel as described in section 3.8. 195 Table 9.2: PCR primer sequences used to detect TTSSl and TTSS2 regions1 Region Gene %GC Amplicon Primer designation & sequence (5'-3') number size (bp) TTSS1 VP1662 53.9 458 VP 1662+ 764: TCACCGCTGGTA TTATCGTT ACGC (2127bp) VP 1662-1221: GAGATACAACGCGCGTCTTACTCT VP1680 53.3 183 VP 1680+613: GCCGAAGCGTATCATCATCAACTC (1479bp) VP 1680-795: CACAGAGCTTACACCAAACGT ACC VP1696 52.0 415 VP 1696+655: ATTGCTTCGGTATTGAGCCGTGTG (1899bp) VP 1696-1069: AACTCTGACCACTGCCCAATGTGA TTSS2 VPA1321 36.9 485 VPA1321+195: GGTTAGTGAATCCAACCAAACCGC (1107bp) VPA1321-679: TTGCCGTGCATGTCATACAACCAG VPA1339 41.6 272 VPA1339+ 398: GACACTCGCTGTTGTTCTCAGGTA (1464bp) VP A 1339-669: GTAAGCGCGTGATGTT AGCTCTTC VPA1355 40.6 388 VPA1355+747: GGCATGTGGTGTCTATTTGACACG (1884bp) VPA1355-1134: TACGACAACTGCAGGTAGTCCAAG 1Primers based on published sequence of clinical strain RIMD221 0633 (Makino et al., 2003) 196 TTSSl gene cluster ,vp •"Cr ·~J"cN'"'~'~<« (p~((~ (1 2kb approx Probe• A Probe• B Probe• C TTSS2 gene cluster ProbeD Probe E Probe F t t t ======~~--~---... ~~~ -- ~·~~~.. ~-- ~----~.. ~~ .. -- ~- ~~-=~ lpal321 fOpP IOXR spaJJ esc.U w:rDl upD tprK VPA I370 Figure 9.1: Genetic organisation of TTSSs from V. parahaemolyticus RIMD2210633 showing targets for PCR and DIG labelled probes. Probes A to Probe F targeted to approximate regions of the TTSS regions. ProbeD lies outside of the TTSS2 regions identified as a V parahaemolyticus pathogenicity island (VpPAl), indicated by flanking black/orange lines (Izutsu et al, 2008). Grey arrows show annotated TTSS genes where known. Unshaded arrows indicate hypothetical genes (adapted from Ono et al, 2006 & Dziejman et al., 2005). 197 9.2.3 Production ofTTSSl and TTSS2 specific DIG labelled probes PCR products corresponding to six of the TTSS regions from the genome of NCIMB UK type strain 1902 was used to develop digoxigenin (DIG) labelled probes and were designated Probes A to Probes F, according to PCR amplification regions (Fig 9.1 ). Details of probes labelled are shown in Tables 9.1 and 9.2. Additionally, three PCR products corresponding to the TTSS2 regions of an 03:K6 clinical isolate (220) that gave a strong PCR signal was also used to develop probes. The random primed method of labelling of DNA probes was performed according to the protocol provided in the Roche instruction manual 'DIG DNA Labeling and Detection Kit'. Details of all reagents referred to in this section can be found in Appendix I. PCR products were purified using the 'High Pure PCR Product Purification Kit' (Roche) and DNA quantified using the methods described in section 3.4. The reactions were performed in 0.5 ml Eppendorf tubes, each containing 0.4 - I Jlg of purified PCR product, 2 Jll of hexanucleotide mixture (stock concentration I .56 mg/ml), 2 Jll dNTP labelling mixture; the mixture was made up to 19 Jll with the addition of molecular grade water and then 1 Ill of Klenow enzyme (2 U/Jll), was added. The tubes were vortexed, centrifuged briefly at room temperature (13000-16000 x g, 5 s) and incubated at 37°C for at least 1 h, after which 2 Ill of0.2M EDTA solution was added to each tube to stop the reactions. 9.2.4 Preparation of dot-blots Genomic DNA was transferred onto positively charged nylon membrane by using a Minifold I 96-well dot-blot system (Schleicher & Schuell, Germany) using the manufacturer's instructions. A volume corresponding to I 00 ng DNA was diluted to 100 Ill with molecular grade water, boiled for 10 m in and kept on ice. An equal volume of freshly prepared NaOH (1M) was added to the samples and incubated at room temperature 198 for 20 min before transferring onto the nylon membrane. After drying for I 0 m in, the membrane was removed from the apparatus and DNA UV cross-linked for a further 5 min (Chromato-Vue cabinet, Module: CC-60 UVP Inc), followed by incubation in neutralising solution for 30 min. Membrane was rinsed twice in 2 x SSC and air dried and hybridised. 9.2.5 DNA-DNA Hybridisation DIG labelled probes were hybridised in DIG Easy Hyb solution (Roche) according to manufacturer's instructions at temperatures of 65 - 68°C. Briefly, 20 m! of DIG Easy Hyb solution was preheated to hybridisation temperature in a roller bottle and the blot was added and left for 20 min for a prehybridisation wash. DIG labelled DNA probe (prepared as in Section 9.2.3) was denatured and added to a preheated 5 m! of Easy Hyb solution at a final concentration of 10 ng/ml. Prehybridisation solution was discarded and replaced with the probe/DIG Easy Hyb mixture. Probe DNA was hybridised overnight in the roller bottles at the appropriate hybridisation temperature. Post hybridisation stringency washes were performed twice in 20 m! of 2x SSC containing 0.1% SDS for 5 m in at 20°C, with two further washes in 10 m! of 0.5x SSC containing 0.1% SOS at 68°C for 15 m in in roller bottles. 9.2.6 Immunological detection of DNA-DNA bybrids Details of all reagents and buffers referred to in this section can be found in Appendix I. Immunological detection of DNA hybrids was performed according to the protocol provided in the Roche instruction manual 'DIG DNA Labelling and Detection Kit'. Blots prepared as previously described were submerged for I min in 200 ml of buffer number I with constant gentle agitation, and then placed in 200 m! buffer number 2 for 30 min. The membranes were rinsed with buffer 1 then placed in a solution containing anti-digoxygenin antibody conjugated to alkaline phosphatase diluted I: 10,000 in buffer I for 2 h at room temperature. Any unbound antibody was subsequently removed by washing the 199 membranes twice in buffer I for 15 min. The membranes were equilibrated for 2 min in 100 ml buffer 3 and then immersed in 70 ml of developing solution. The development reaction was allowed to proceed in the dark until a colour precipitate formed. The reaction was terminated by washing the membrane several times with TE buffer. Images of the developed membranes were captured onto a digital camera (Sony Cybershot DSC-I 0) with the macro settings enabled and the intensities of hybridisation signals were scored 0 to 3. 9.2.7 Sequencing and analysis ofTTSS genes Both strands of PCR products corresponding to all six TTSS loci derived from NCIMB UK type strain 1902 were sequenced by MWG Biotech AG. For comparative purposes, amplicons derived from two additional isolates were also sequenced (03:K6 clinical isolate 220 and environmental isolate VP416) These sequences, together with the published genome sequence of V parahaemolyticus strain RIMD221 0633, were aligned using the online programs MultAiin, Artemis, ACT and Blastn 2. (Carver et al., 2005; Corpet, 1988; Makino et al., 2003; Tatiana et al., 1999). Sequences were also compared with those from other organisms on the GenBank database using the gapped BLAST program at http://blast.ncbi.nlm.nih.gov (Aitschul et al., 1990). 9.3 Results 9.3.1 PCR primers targeted to the TTSS loci PCR primers based on the published genome sequence of V parahaemolyticus RIMD221 0633 successfully an1plified regions of both sets ofTTSS genes in most of the 55 isolates of V parahaemolyticus tested, as well as in some isolates of V cholerae, V vulni.ficus and V alginolyticus (Figs. 9.2, 9.3 & 9.4). Most of the TTSS amplicons from V parahaemolyticus appeared consistent in size, with various degrees of intensity. Sequencing of the PCR products of three isolates (1902, 220 and VP416) followed by 200 BLAST searches confirmed their identity as TTSS regions of the V parahaemolyticus genome (section 9.3.7). An annealing temperature of 64°C proved to be the most reliable in the amplification of both TTSS1 and TTSS2 genes using the same cycling conditions (Fig 9.2). Two multiplex PCRs targeting the TTSS 1 and TTSS2 regions respectively both amplified three loci of correct sizes (Table 9.2) in most cases. The bands of the TTSS2 regions appeared weaker in the multiplex PCR reactions and an increase of primer concentration by 5 pmol provided more consistent results (Fig 9.2). 9.3.2 PCR amplification of genes from the TTSSl region Of the 55 isolates of V parahaemolyticus tested, PCR was positive for all three loci of TTSS I (representing the entire TTSS 1 system) in 48 (87%) of the isolates. Of the remaining seven isolates, three clinical isolates (2436, VPI0884 and 219) and two environmental isolates (VP331 and Vib1) PCR only amplified two out of the three genes (Figs. 9.3 & 9.4). Of the five isolates, only the amplicon corresponding to vpl680 showed a similar sized PCR product to the other 48 isolates (Figs. 9.3 & 9.4). In isolates 2436, VP331 and VP10884 PCR did not amplify vcrDJ whereas in isolates 219 and Vib1 vscCJ did not amplify. Repeated PCR on these isolates provided the same results. A further two environmental isolates VP345 and VP361 did not amplify any TTSS 1 loci. The absence of the TTSS 1 region was confirmed by DNA hybridisation (Table 9.4, section 9.3.5) 9.3.3 PCR amplification of genes from the TTSS2 region The TTSS2 regions in both clinical and environmental isolates of V parahaemolyticus showed much more variability in intensity than the TTSS 1 amplicons (Figs. 9.3 & 9.4). Genes near the TTSS2 region corresponding to vpa132 1 were not amplified in two clinical isolates (2436 and 228) and 20 environmental isolates, accounting for 11% and 59% of the 201 two groups respectively. Clinical isolates that showed weak TTSS2 signals included 219, 221 and 224. Of the 15 environmental isolates that showed positive PCR for the TTSS2 region only two isolates (VPI 0 and Vib3) did not give an amplicon corresponding to vscC2 (Fig 9.4). In most cases apart from isolate VP416, PCR signals for the amplification of TTSS2 regions in environmental strains were weak (Fig 9.4). Again, the presence of the TTSS2 region was confirmed by DNA hybridisation (Table 9.4, section 9.3.5). 9.3.4 TTSS probes specificity in dot blots DIG labelled probes A to E, developed from the specific PCR products of UK type strain 1902, hybridised successfully to most of the DNA samples from both clinical and environmental isolates (Figs 9.8 and 9.9). The hybridisation data compared well with PCR data and in most cases confirmed the presence of TTSS 1 or TTSS2 regions in both clinical and environmental isolates of V parahaemolyticus (Table 9.4). However, a weak hybridisation signal was obtained with some isolates that did not show the presence of TTSS regions by PCR. Additionally, a positive 'halo effect' in some of the samples was observed (Figs. 9.8 & 9.9). In such instances, the hybridisation signals were scored by the relative intensity of the halo as compared to the positive control. Only scores of 2 or 3,for the relative intensity of hybridisation signal, together with the presence of PCR signals were counted as positive for the presence of TTSS regions. There was no evidence to suggest that TTSS regions existed in multiple copies in any of the 56 isolates tested. Results are summarised in Table 9.4. 9.3.5 Dot blot analysis ofTTSSl genes vcrDJ, vpl680 and vscCJ In support of the PCR results, dot blot hybridisation confirmed the presence or absence of TTSS I regions in most of the clinical and environmental isolates of V parahaemolyticus. Hybridisation signals with probe A (vcrDJ) were present in all clinical isolates including isolate 2436 that was negative by PCR. However, five environmental isolates (VP6, VP7, 202 VibiO, 1024/5 and I023/3) that had shown to be positive for the vcrD1 locus by PCR appeared negative or showed very weak signals in dot blot analysis. Furthermore, dot blot reliably confirmed the absence of vcrD1 in isolates VP331, VP345, VP36I, VPECI and VPSCI (Fig 9.6, Rows C & D). Signals with probe B, representing the presence of the vp 1680 locus, were present in all of the clinical isolates although signals were distinctly weaker in nine out of the I7 isolates (Fig 9.7, Rows A and B). However, three environmental isolates, (I022/10, 1024/5 and VP7) that were shown to be positive for the vp1680 locus by PCR appeared negative in dot blot analysis. Additionally, isolate VP345 that had shown to be PCR negative for vp1680 also showed a weak hybridisation signal. Hybridisation signals with probe C, representing the presence of the vscC I locus were present in all the clinical isolates and hybridisation signals were stronger than with Probe B (Fig 9.8, Probe C, Rows A and B). However, five environmental isolates (VP7, VP346, VP4I9, I 023/3 and I 024/5) that had shown to be positive for the vscC1 locus by PCR appeared negative in dot blot analysis. Furthermore, dot blot reliably confirmed the absence ofvscC1 in isolates, VP345, VP36I, VPECI and VPSCI (Fig 9.8, Probe C, Rows C&D). 9.3.6 Dot blot analysis of TTSS2 genes vscC2, vcrD2 and vpa1321 Analysis of the hybridisation of probes specific to vscC2, vcrD2 and vpa1321 regions all gave similar results and in most cases confirmed PCR findings (Figs. 9.8 & 9.9). Clinical isolates that gave weak (219, 221 and 224) or no (2436 and 228) PCR signals for all three loci also gave weak dot blot hybridisation signals, confirming the presence ofTTSS2 genes in all clinical isolates of V parahaemolyticus examined. However, II environmental isolates that were negative by PCR for TTSS2 regions demonstrated weak to moderate 203 hybridisation signals in comparison to the negative control isolates. Additionally, three environmental isolates (VP331, VP346 and Vib 10) that appeared positive in PCR did not hybridise to probes D, E or F. All results are summarised in Table 9.4. 9.3. 7 Conserved sequence identities of V. parallaemolyticus TTSS genes Multiple sequence alignments from three isolates together with RMID at all six loci used in this investigation revealed I 00% congruence in nucleotide sequence in all three loci of the TTSS2 regions (Fig. 9.12). However, the sequences of the three TTSS1 regions showed some variation amongst the three isolates studied. Sequencing of PCR amplicons showed up to nine single nucleotide polymorphisms (SNPs) present within a fragment size of 458 bp for vcrDJ (Fig 9.11 ). Additionally, vpl680 and vscCJ showed 11 and 7 nucleotide substitutions respectively. These together with the TTSS2 sequence homologies account for between 97 and 100% similarity between all three isolates in the TTTSS 1 and TTSS2 regions. The SNPs do not translate into a different amino acid in vcrDJ, except for one codon where isoleucine is encoded in type strain 1902 instead of valine in isolates RlMD, 220 and VP416 (position 119; Fig. 9.11). In addition to the SNPs, vpl680 also exhibited nucleotide deletions at position 30 (in isolates 220, 1902 and VP416) and further deletions at positions 40, 50 and 147 in isolate VP416 with a resultant frameshift mutation in vpl680 (Appendix Ill; Fig. III.7). 9.3.8 Correlation of phylogenetic clustering of isolates with the presence of TTSSl and TTSS2 genes Four environmental isolates of V parahaemolyticus that were negative for TTSS 1 region (VP345, VP361, VPEC1 and VPSC1) were clustered together and can be found in RAPD types 28, 29 and 30 (Chapter 8 & Fig. 9.10) with an additional isolate, VP331. Environmental isolates positive for the TTSS2 region generally fell into one group and two 204 sub clusters at a 76% similarity value (Fig 9.1 0). Of these clusters (indicated by red and blue in Fig 9.1 0), one group of 8 consisted entirely of environmental isolates and the other group consisted of a mixture of three environmental and four clinical isolates. A further three environmental isolates (VP2, Vib3 and Vib7) positive for the TTSS2 region were found scattered in various clusters (Table 9.4). 9.3.9 V. parahaemolyticus TTSS regions in other species The PCR primers and probes developed from template of V parahaemolyticus type strain 1902 were found to amplify PCR products in five other species tested including Vibrio mimicus (VM499), V vulnificus (VV11067 and KIDBVV), V cholerae (type strain 8042, VC3, VC4 and VC5), V alginolyticus (2341) and V anguillarum (Figs. 8.3, 8.4, 8.5, Table 8.3) Probes gave no hybridisation signals with Proteus mirabilis (PM5), Salmonella typhimurium (ST5), Escherichia aerogenes (EA5), Klebsiella aerogenes (KA5) and Pseudomonas aeruginosa (Fig 8.5) 205 M 2 3 4 5 6 7 8 9 M - - '" •• - t • - - ••• -...... -... -..... - tl t t M 2 3 4 5 6 7 8 9 M - - tll • t ..- .,., - ••• • t - -~ .. .. - - tl - •• .. - •• M 2 3 4 5 6 7 8 9 M Figure 9.2: PCR amplification at annealing temperatures of (A) 6JOC, (B) 64°C and (C) 6~°C for TISSJ and TISS2 genes In NCfMB type strain 1902 in single and multiplex PCR reactions. Lanes correspond to amplicons from the following, with approximate sizes: I: M-PCR for all three ITSS I products; 2: VPI662 (458 bp); 3: VPI680 (183 bp); 4: VPI696 (415 bp); 5: M-PCR for all three ITSS2 products; 6: VPAI321 (485 bp); 7: VPAI339 (272 bp); 8: VPA1355 (388 bp); 9: no template control; M: tOO bp markers with selected sizes indicated by arrow. 206 TTSSl TTSS2 vcrDI vpa/321 vscC2 vscCJ/ vcrD2 Vp/680- M 2053 2436 5421 6134 6316 E154482 E155855 E168143 VP10884 LD365 M M 219 220 221 222 223 224 225 226 227 228 M _, , I ,~ 'I) ' ' :I) ' ' ' ~ '4 '" '(, 'S ~~~ ~ I '' ' '' ~ -l ' ... ~" ' ~s 411 I - ... • • - - - = - --- = - = - .....------11 •• Figure 9.3: PCR of 1TSS lod in clinical isolates of V. JHlralraeMO/ytiCIIS from Japan, UK and Norway. (A): amplicons of isolate 542 I derived from primers based on sequences of strain RJMD2210633. (B) and (C): For each isolate (as indicated above the lanes) a pair of lanes corresponds to TTSS I and TTSS2 regions respectively. Lanes I- 10 Japan isolates with amplicons representative of all six loci; lanes 11- 16 UK PHLS isolates; lanes 17 - 18 NCfMB reference strain VP I 0884; lanes 19 - 20 LD365(L. damsela) ; lanes 21 • 40 Norway isolates. Lane M corresponds to I 00 bp ladders with selected markers indicated on the right. 207 M VPI VP2 VP3 VP4 VP6 VP7 VP8 VP9 VPIO 361 M bp bp M 331 34S 346 349 3SO 377 378 416 491 939 M M Vibl Vib2 VibJ VibS Vib6 Vib7 Vib8 VibiO bp M 1902 102115 1022/10 102311 102313 1024/S 1103/9 VCI 2341 11327 M Figure 9.4: PCR of'ITSS loci in 34 environmental isolates of V. parllltaemolyticlls from UK and Spain. For each isolate (as indicated above the lanes) a pair of lanes corresponds to ITSSI and ITSS2 regions respectively. Lanes I - 18 Southampton isolates; lanes 19 - 40 isolates from Dorset coast; lanes 31 - 56 Southampton isolates; lanes 57 - 58 unspeciated Vibrio sp. Vib 4; lanes 59 - 60 V. mimicus; lanes 61 - 62 NCIMB reference (clinical) strain; lanes 63 - 74 Spanish isolates; lanes 75 - 76 V. cho/erae; lanes77-78 V. alginolyticus; lanes 79-80 V. jluvialis. Lane M corresponds to I 00 bp ladders 208 M 8042 VCJ VC4 VVII067 A B C D E F A• a• c• o• P F• M A bp bp B VVII067 KID VA430 VM499 LDJ65 Vanl04 8042 VCI VC2 VCJ VC4 VCS PMS Figure 9.5: PCR and hybridisation signals in VIbrio spp. (A) Single primer and multiplex PCR signals of TISS loci in Vibrio species with primers based on V. parahaemo/yticus RMD2210633. Lanes l-2: V. cho/erae type strain 8042; lanes 3-6: V. cholerae environmental isolates; lanes 7-8: V vulnijicus; lanes 9-14: V. vulnificus VVII067 single primer PCR with all six primer sets; lanes 15-20: V. vulnificus environmental isolate KIDBVV single primer PCR with all six primer sets respectively. Lane M corresponds to 100 bp markers. (B) Dot blot hybridisation signals (probe E) with isolates labelled above the dots. Isolates EC5, PM5, STS, EA5, KA5 and PAl are non vibrios. Table 9.3: VIbrio species tested for TTSS genes of V. paralulemolytlcus. TISSI TISSl Isolate A 8 c 0 [ F DesipaliOil Species Lotation Solll"tle 1902 V. porahaemo/yt1CM!J PHLS, UK NCTMB Reference Strain + + + + + + 8042 V. cho/erae PHLS, UK Ncrc Reference Sttain + + + 11327 V. fluvial is Oslo, N01W8y NcrC Reference sttain 4 2341 V. a/gmo/yticus Kyoto, Japan Clinical ·~ ... VCI V. cho/erae Southampton Environmental VC2 V. cho/erae Southampton Environmental .;; VC4 V. cholerae Soutlwnpton Environmental + + V. Environmental VC5 cholerae Soutbampton + + V. mim1cus VM499 Dorset Environmnetal ... V. vulnificus "' $ VVII067 Dorset Environmnetal + * V. vulnificw ~· ;;. KIDBVV Dorset Environmental + LD365 Llstonel/a damsellae Dorset Environmental Vib4 Vibrio spp. Southampton Environmental BL21 E. coli (control) Amersham Unlrnown A- E correspond to all six loci described in Fig. 9.1 and Section 9.2 +/- Presence I absence ofTISS genes based on bolh PCR and dot blot analysis ilt Identified with dot blot but not confirmed with PCR • Identified with PCR but not confirmed with dot blot 209 Probe A 2 3 4 5 6 7 8 9 10 11 12 1902 2053 2436 5421 6134 (6316) (2341) EU4-182 EI$S85S FI6814J A VPI0884 219 220 221 222 223 224 225 226 227 228 B -ve 331 3-15 346 349 ]50 361 377 378 491 416 c 939 ECI SCI 1021/5 1022110 102311 102313 102415 1103/9 D VPI VP2 VP3 VP4 VP6 VP7 VPS VP9 VPIO E V1hl Vib2 Vib3 (Vib4) Vib5 Vib6 Vib7 Vib8 VihiO F Figure 9.6: Hybridisation signals of 55 Isolates of V. parallll~molydcus witb Probe A targeted to tbe vcrD/Iocus ofTTSSI. The isolate corresponding to each dot is indicated. Rows correspond to the following: A and B clinical isolates; C and D environmental isolates from Dorset coast and Spain; E and F environmental isolates from Southampton, UK. White regions represent artefacts from photography of wet membrane. 210 Probe C 2 3 4 5 6 7 8 9 10 11 12 1902 2053 2436 542 1 6134 (63 16) (23-H) El54482 El55855 E168 A VP10884 219 220 221 222 223 224 225 226 227 228 B -ve 331 345 346 349 350 361 377 378 491 416 c 939 ECI SC I 102115 1022/10 102311 I 023/3 I 024/5 1103~ D VPI VP2 VP3 VP4 VP6 VP7 VP8 VP9 VPIO E Vibl Vib2 Vib3 (Vib4) Vib5 Vll>6 Vib7 Vib8 VibJO F Figure 9.7: Hybridisation signals of ~5 isolates of V. parahaemolydcus witb Probes 8 and C targeted to tbe vp1680 and vscCJ loc:i ofTfSSl respectively. The isolate corresponding to each dot is indicated. Rows correspond to the following: A and B clinical isolates; C and D environmental isolates from Dorset coast and Spain; E and F environmental isolates from Southampton, UK. 211 3 4 5 6 7 8 9 10 ll 12 2053 2436 5421 6134 (6316) (2341) A 219 220 221 222 223 224 225 226 B c 331 345 346 349 350 31\1 177 1711 939 ECI SCI 1021/5 102211 0 1023/1 1023/3 D VPI VP2 VP3 VP4 VP6 VP7 VP8 E V1bl Vib2 Vib3 (Vib4) VibS Vib6 Vib7 Vib8 F Figure 9.9: Hybridisation signals of 55 Isolates of V. parahaemolyticus wltb Probe D targeted to the vpa/321 locus downstream of tbe lTSSl region of 'pathogenicity island' (Vp-PAl). The isolate corresponding to each dot is indicated. Rows correspond to the following: A and B clinical isolates; C and D environmental isolates from Dorset coast and Spain; E and F; environmental isolates from Southampton, UK. TTSS I l ! ! r 1022/10 Spain Envronmertal + 110319 Spain Envronmertai + uc 102415 Spain Envronmertal uc 1021/5 Spain Envronmertal + + 1023/1 Spain Envronmertal + uc VP377 Donillt, U< Envronmertal + + VP349 Donillt, U< Envronmertal + + VP350 Dorset. U< Envronmertal + + VP346 Dorset, U< Envronmertal + uc Spain Envronmertal uc 1102313 VP378 Dorset. U< Envronmertal + + 2053 Japan Clini:al + + 5421 Japan Clini:al + + 2436 Jap..-t Cllni:al + + 6134 Jap.-. Cllni:al + + VP416 D01'8111, U< Envronmertal + + VP939 Dolll!l, U< Envronmertal + + VP491 Donlllt, U< Envronmertal uc Figure 9. 10: RAPD-PSJ clustering oflTSSl positive isolates. Three clusters are shown in green, red and blue based on a 76% similarity value. Both TTSS 1 and TTS2 correspond to all three loci described. Isolates indicated with + were conf11tned by PCR and dot blot whereas isolates marked with uc are unconfirmed by either PCR or dot blot. All isolates marked - are negative for both PCR and dot blot. 213 Table 9.4: Comparison of the presence ofTTSS related genes A-F, virulence characteristics and phylogenetic RAPD Typing clusters in V. parai/Uemolyticus. TISSI TI'SS2 RAPD Virulence ABCDEF PSI Isolate Source Choraderi!tics1 1902 NCIMB Reference K-, tdh-. trh ++++++ 20 + + + + + 18 10884 NCfC Reference K+, tdh ...... trh ++ilt+++ 27 219 Clinical K-,tdh-. trh ++++++ 25 220 Clinical K+,tcDr+, trh ++++++ 25 221 Clinical K-, tdh-, trh+ ++++++ 222 Clinical K+,tdlr+,lrlr+ 25 ++++++ 25 223 Clinical K+,tdlr+, trlr ++++++ 26 224 Clinical K-, tdh ..... trh+ ++++++ 25 225 Clinical K+,tdh+, trh + + + + + + 25 226 Clinical K+, tdh+. trlr + + + + + + 25 227 Clinical K+,tdh t, trlr + + + + + 228 Clinical K-, ufh-, trh+ 25 03 1024/5 Environmental K + + + + + + 04 1021/5 Environmental K-,tdh-.trlr + + 1022/10 Environmemnl K-, tdh-, trh + 01 +++lil!V 102311 Environmenull K-, tdh-, trh 04 10231) Environmental K-, tdh-, trh 09 1103/9 Environmema I K-. tdJr-. trh + + + * -€-- .;- 02 ++++++ 11 2053 Clinical K-, tdh 1, trh i$1++~t~ilte 12 2436 Clinical K+,tdJr+. trh+ + + + + + + 5421 Clinical K-, tdh+, trh 11 ++++++ 6134 Clinical K+,tdh+, trh 13 ++trtese Vib I Environmental K-, tdh·. trh 24 .. + 24 Vib 2 Environmental K-, tdh-, trh ++++$+ Vib 3 Environmemnl K-, tdh-. trh 24 + + + 24 Vib5 Environmental K-, tdh-. trh· +++S!!IS 23 Vib6 Environmental K-, tdh-. trh ++++++ 23 Vib7 Environmental K-, uih +, trh + + + $ ;{! ill Vib8 Environmental K-, 1dh-. trh 23 +-1>++-1·+ Vib 10 Environmental K-, tdlt+, trh 23 +++-ll!ttiltl VPI Environmental K-, tdJr-. lrh 21 ++++++ 21 VP2 Environmental K-,tdh-. trh-+ VP3 Environmental K-, tdh-. trh' +++E!G~ 21 ++·t-ee.q:. VP4 Enviromnental K-,tdh-. trh 21 + + + 17 VP6 Environmenral K-, tdh-, trh + ... 17 VP7 Environmental K-,tdh+. trhr + +++GaS 17 VP8 Environmental K-,tdh+, trh + +++l!.'i!i$ 17 VP9 Environmental K-,ldh+. trlr+ VPIO Environmental K-,tdh+,trh+ ++++$+ 19 ++++++ El54482 Clinical K-, tdh-, trh+ 22 ++++++ 22 E155855 Clinical K+,tclh+. trh ++++++ El68143 Clinical K-,Idh-+, lrh 22 + + + + VPJJI Environmental K-, tdh-. trh + 28 VP345 Environmental K-, tcUr+. trlr 28 ++++-!-+" " 08 VP346 Environmental K-, tdh-, trh 1Virulcnce characteristics ++++++ 06 VP349 Environmental K-, tdh+, trh K: Kanagawa haemolysis and ++++++ 07 VP350 Enviromnental K-,tdh+, rrlr virulence genes (tdh. trh) 29 VP361 Environmental K-, tdh-, trh +I· Presence /absence ofTTSS ++++++ 05 VP377 Environmental K-,tdh·,trh genes based on both PCR and dot VP378 Environmental K-, tdh-, lrh ++++++ 10 blot analysis + + + + + + 13 VP416 Environmental K+,tdh+,trh + + 4 15 l!> Identified wilh dol blot bul VP491 Environmental K-. tdh-. trh ++++++ 14 not confirmed with PCR VP939 Environmental K-, tdlr-. trh 30 VPECI Environmental K-, tdh .... _trh ~ Identified with PCR bui nol VPSC! Environmental K- tdh-, rrh- 30 confirmed wilh dot blot 214 1 - ~ I I I I I I :-1------1o______2o______30 ______4o______so __: so 10 eo 90 100 uo 21o : 130 t I I I I I I 1 t I I I I I ~ i I RIJI)221~ TCRCC6CT66TRTTRTC6TTIICGC6T6TTTCCCACGRRGRCTC66CCGRCTTQ66CTCC60CATT66C666CAfliTCACG6CTtfiGCCTiliR6CRCT6TT6fiTT66T66TGTGCTGCTm~~GTTC6C 2209 TRGG>---tTCTIIRGRCTC66CiliR-TT$iGCTCCGfiCRTT6GCGGGOIRGTCRC66CTCftGCCTC6R&CRCTGTT6fiTTGGT6GTGTGCT6CT GTTCGC VP4161( TR6ft6T ---cTR6HAGfiiiTC66Ct:&ft-TTQ6GCTCC6fiCRTT6GC666CRA61 CRCG6CTCIIGCCTiliR6CIICTGTT6fiTTGGT1iliTGTGCTGCT OrTGTTC6C 19029 TGIIGT---t TIIGfiR6flfiTC66CCGfl-TTQiGCTCC611CRTT6GTG66CflfiGTCACG6CTCRGCCTtGTGCIICTGTT6fiTT66TG6C6TGCT6CT&RTC'miTTCGC tonsensu.: •••••••••••••••••••••• 6aGT. •••• . Ct.aeMGJ~armrcGR. TTq&GCTCC&IICRTTGGcGGGCRR6TCACGGCTCIIGCCTC6a6CIICTGTT6fiTTGGT6Gt.GTGCTGCTO.eJCfTGTTCGC ~------m ~ ~ ------~ ~ ~ · ~ ~ ~ ~ ~ ~ ~ ~ I I RIJI)2210633 TCTGRTCCCTGGCTTTCCGAAAATTRCRTTCCTR6T6CTG6CTTT6GTGGTCG6T6GCG6C6GTTTTTRCCTCTTCTRTCftRCRGRfiRRfiGGARTC8GflftAGCII6CGTCT6CCtiRGCTTT6TC 220ft TCT6fiTCCCTG6CTTTCCGflflfiiTTRCRTTCCTA6TGCTGGtTTT66T66TCG6TGGCGGCG6TTTTTIICCTCTTCTRTCRIICR6flllfiiiCR6fiRR6CRiitGfTCTGCC6R6tTTT6TC VP4168 TCT6fiTCCCTG6CTTTCCGRfiRRTTRCRTTCCTRGT6CTGGCTTT6GTG6TC66T66C6GC6GTTTTTRCCTCTTCTRTCRRCA6ARRRR&CfiRfi:CGRRTCR6ARR6CRGCGRTCT6CCGR6CTTT6Tt 1902ft TCTTRTCCCT6GCTTTCCGfiRARTTRCRTTCCTR6T6CTG6CTTT6GTG6TCG6T66C66C6GTTTTTRCCTCTTCTRTCRfiCAGAfllflfiGAGTCT6RAII6CRGCGATCTGCCGRGCTTT6TC Consensus TCT efiTCCCTGGCTTTCC6fiiiRTTRCRTTCCTR6T6CT66CTTTGGT66TC66T6Gt66C66TTTTTRCCTCTTCTRTCRRC11611111111GCAIIICC6Ra TCaGRflfiGCRGCGfiTCT6CCGfi6CTTT6TC ~ m ~ ~ ~ ~ ~ ~ ~ ~ ~ m ~ ~ I I RIJI)2210633 GCGCR6G6R6CG66TTCTCCTiiCC6CRRIIICCCRRTfiiiiCCRRCGCC6TCTCGT66TIIGCRRR66CRRRTTG6GCGft6tfiR6RH6R6TTT6CGRTGRCIIGTGCCRCTGCTTRTiliRTCT6GRTTCRfi6CT 220ft GCGCRGGGA6CGGGTTCTCCTGCC6CAAAACCCIIATRAIICCARCGCCGTCTCGTG6TIIGCAARGGCAfiRTTGG6CGfiGCIIAGARTTT6CGfiTGRCftGT6CCRCTGCTTRTCGRTCT6GRTTCRRGCT VP416R 6C&CR66GRGC6GGTTCTCCTGCC1itfiARRCCCRRTfiRfiCCRRCGCC6TCTC6TG6TRGCIIARGGCAfiRTTGGGCGfi6CRRGATTT6CGRTGRCRGT6CCRCT6CTTRTC6RTCT6GRCTCRRGCT 1902R GC6C8666AGCGGGTTCTCCTGCCGCRRfiACCCRRTIIRfiCCRfiCGCCGTCTCGTG6TIIGCRRRGGCRRRTTG6GCGRGCflfiGRRGC6CGRT6RCR6T6CCRCTGCTTATCGRTCT6GRTTCRR6CT Conaensua GC6CR6GGRGCGGGTTCTCCTGCtGCRIIRRCCCRRTRRRCCRRCGCCGTCTC6TGGTIIGCRAIIGGCflfiHTTGGGCGRGCflfiGIIAGRGTTt.6CGfiT6RCR6T6CCRCTGCTTATCGATCT6GRt.TCRAGCT 391 400 410 420 430 : - 440'"" - - - --<15lT-- - '158 - i I I I RIJI)2210633 TRCAA6RfiAGCCTRGAAGCC6TCGCRCT~!~TRRGRCGCGCGTT6TRTCTC : 220ft TRtRRGHRA6CCTR6RRGCC6TCGCRCT6fl TRR6RCGC6-GTT VP416R TRCRAGfiARGCCTRGfiiGCC&TCGCRCTGRIICGfiCGARCTCGCCft&RGTRR6RCGCG-GTT 1902R TfiCRAGRfiRGCCTRGARGCC6TCGCRCTGfiRCGRCGfiACT~TRRiiRCGCGG Consenaua TRCARGARR6CCTIIGRfliCC1iTC6CRCT6RfiCGfiC6fiiiCTCGCC~RRGOCGC6 . att ••••••• ------I J Figure 9.11: Multiple alignments of the vcrDJiocus of TTSSl in isolates of V. parahaemolytkus. RIMD2210633 is a sequenced clinical strain. Sequences 220A and VP416A are from clinical and environmental isolates respectively and 1902A is NCTC reference strain from which probes were developed (Corpet, 1988; Makino et al., 2003). Consensus regions are shown in red; blue and black show substitutions or deletions. Codon marked in green is valine (GTC) or isoleucine (ATC). Regions marked in dotted line represent quality control clipped regions. SNPs in TISSI are also present in vp1680 and vscCJ (Appendix rrn 215 r------~ I I lL 10 20 ~ 40 50 60 70 80 90 100 110 120 130 11 I RDIJ2210633 :r&eTTCTIIIITTT6fiiiRIITRRTRGTR6tfmRRTCGRRRTGRTRTCRATGTIIRRfiRRR6CA6RTGCRRRRTCRRTRACCGOiCT6RfiRR6T66T6RTTTGCRCRTRCTTRRIIGGCIII666RRTTRTR66T 2200 I GRGTTCCRTRGTGATRRRTRGRfiGOmRRTCGRRftTGRTRTCRRTGTIIIIIIARRGCRGRTGCRRRRTCRRTIIICCGCGCTGRARRGTGGTGRTTTGCRCRTRCTTRIIIGGCJIAG6GRRTTRTRGGT YP4160 : GRGTTCCRTRGT6RTRRRTRGRf161tmRRTCGRRRTGRTRTCfiiT6TIIflllflfiRRGCRGRTGCRRfiiTCRRTRRCCGCGCTGRRIIRGT66T6RTTTGCRCRTRCTTIIRfi66CIIfi666TTRTRGGT Consensus ~••• eae TtccaTa~RtaRRTRGaRGOTTTRRTCGRRRTGRTRTCIIITGTRRRRRRRGCR6RT6CRRRRTCRRTRRCCGCGCT6RRRRGT66T6RTTTGCRCRTRCTTRRRG6CRRG6GRRTTRTRG6T ~------1131 140 150 160 170 180 190 200 210 220 230 240 250 260 I I RDID2210633 RTGRR6666GGRGRTRRTIII6CTCCCGTTCAART6TRCTRTT6TRRRT6RTGRTRRRRRTG6T6CGCRTCTTRGTCRR666RCTIIITTTRGCRRCCRRT66TRTTRRGR6TRTGGCTG666RTTC86TTC 2200 RTGRRGGG6GGRGRTIIATRRGCTCCCGTTCRRRmACTRTT6TRRAT1iRTGRTRRfiRRTG6TGCGCRTCTTfiGTCRAGGGRCTIIITTTAGCRACCRAT6GTRTTRRGRGTRT6GCTGGG6RTTCRGTTC YP4160 RTGRA6G6GGGRGRTRRTRRGCTCCCGTTCARRT6TRCTRTT6TRIIIT6RT6RTRRRRRT66TGCGCRTCTTR6TCRRGGGRCTRRTTTRGCRRCCRRTG6TRTTRRGRGTRT6GCTGGGGRTTCR6TTC ~ RT6RRGGG6GGAGRTRRTflfiGCTCCCGTTCRART6TRCTRTT6TRRRTGRTGRTRARIIIT6GTGCGCRTCTTR6TCIIR6GGRCTRRmiiGCARCCRRT6GTRTTRRGRGTRT6GCTGGGGRTTCRGTTt 261 270 280 290 300 310 320 330 340 350 360 370 380 390 I I RDIJ2210G33 6T6CCGCTORRCTRRTCCCT66TRCGCCTCTTGGRCR6TTTTRTRRTRGCGCGCCTCTRGRT6RTR6CTTTRRTGT6GTRCRTTTRCCTRRTGGTCRRRGR66T6TRRRTGGRCTTRRRRTRCCRCT6Tt 2200 GTGCCGCTCRRCTRRTCCCTGGTRCGCCTCTTGGRCRGTTTTRTRRTRGCGCGCCTCTRGRTGRTRGCmRRTGTGGTRCRmRCCTRRTGGTCRRRGRGGTGTRRRTGGRCTTRRRATRttRCTGTt YP4160 GT6CC6CTORRCTRRTtttTGGTRCGCCTCTTGGRCR6TTTTRTRRTRGCGCGCCTCTRGRTGRTRGCmiiiiT6T66TRCRmRCCTRRTGGTCARRGRGGTGTRRRT66RCTTRRRRTRCCRCTGTt Consensus GTGCCGCTCRRCTRRTCCCTGGTRCGCCTCTTGGRCRGTTTTRTRRTRGC6C6CCTCTRGRT6RTRGCmRRTGT66TRCRTTTRCCTRRT66TCRRRGRG6T6TRRRTGGRCTTRRRRTRCCRCTGTC·------, J91 400 410 420 430 I 44() 45152 I I I I : RDI02210633 TGRGTTTTRTTCTGRRRRRRARTTmRTTCTCRRRTGGTGt~tTGTt 1 2200 T6R6TTTTRTTCTGRRRRRRRRTTmATTCTCARRT6GTGCOCT6TCTGGTTGTRTGRCT6 : YP4160 TGRGTmRTTtTGRRARRRARTTmRTTtTCAARTGGT~TGTtTGGTT&TRT&ACT& 1 Consensus TGRGTTTTRTTCT6RRRARRRRTTmRTTCTCRRRTGGTGCOCT6Ttteattetataacte : I I ~------~ Figure 9.12: Multiple alignments of the vpal321 locus of TTSS2 in isolates of V. parahaemolytkus. RIMD2210633 is a sequenced clinical strain. Sequences 220D and VP416D are from clinical and environmental isolates respectively. (Corpet, 1988; Makino et al., 2003). Consensus regions are shown in red, blue and black show substitutions or deletions. Regions marked in dotted line represent clipped regions. Nucleotide sequences of vscC2 and vcrD2 of TTSS2 are also conserved (Appendix ill) 216 9.4 Discussion Genome sequencing of V parahaemolyticus (RIMD2210633) revealed TTSS regions, one with extensive homology to that described for Yersinia spp, and the other that had previously not been characterised before but was later found to show similarities to the TTSS2 found in non-01 and non-0139 V. choleae (Dziejman et al., 2005). Both TTSSI and TTSS2 were shown to be functional and the fact that TTSS2 was only reported in Kanagawa positive (~-haemolytic) strains, suggest that it may be employed in the pathogenicity of V parahaemolyticus (Ono et al., 2006). The haemolytic reaction of thermostable direct haemolysin (TDH) is considered a marker of pathogenic strains (Sakurai et al., 1973), although not alltdh bearing isolates exhibit this phenotype (Chapter 4; Nishibuchi et al., 1989). Thus, the significance of finding tdh alone in environmental isolates remains unknown. Additionally, a small proportion of clinical isolates apparently lack both tdh and /rh and the exact pathogenesis of V parahaemolyticus has yet to be elucidated. The current work was carried out to assess the distribution of TTSS I and TTSS2 regions in V parahaemolyticus and may help to differentiate virulent and avirulent isolates of V parahaemolyticus. 9.4.1 Gene targets on the TTSS system of V. paraltaemolyticus Three loci in TTSS 1, two in TTSS2 and an additional locus recently identified just outside of the TTSS2 border comprising of the V. parahaemolyticus pathogenicity island (Vp- PAl) were targeted by PCR (Izutsu, et al., 2008; Nishioka et al., 2008). Two genes, vcrDJ and vcsCJ targeted on TTSS 1 correspond to structural proteins with homologues found in P. aeruginosa, S. typimurium, enteropathogenic E. coli (EPEC), B. pseudomallei as well as several other species (Hueck, 1998). Additionally, VPI680 described as a TTSS effector protein, to date only described in V parahaemolyticus, was a third target for PCR in this study (Ono, Park et al., 2006). The three loci were used to represent the entire TTSS 1 region of approximately 40 kbp (Park et al., 2004b). Similarly, the three genes vpa1321, 217 vscC2 and vcrD2 were selected to represent the entire TTSS2 region. The TTSS I and TTSS2 are two distinct secretion systems. Comelis and Van Gijesegem (2000) classified the TTSS of bacteria into three major groups: the Ysc-plus-Psc system, Salmonella pathogenicity-island 'S-PAI-plus-Mxi/Spa' system and a 'S-PAI-plus-EPEC-EHEC' system that constituted a system of Salmonella, enteropathogenic and enterohaemorrhagic E. coli genes. Whereas the TTSSI system falls into the Ysc-plus-Psc system, the TTSS2 of V parahaemolyticus is unlike any of the three described (Ono et al., 2006). Sequence alignments between vscCJ and vscC2 and between vcrDJ andvcrD2 ofTTSSI and TTSS2 in three experimental isolates and two sequenced strains revealed no significant homologies for the cross hybridisation of probes or PCR primers, thus were used as PCR and probe targets. 9.4.2 Detection of TTSS regions Single and multiplex PCR targeted to the six loci described were generally found to be specific in amplifying the correct sized fragments in V parahaemolyticus. Upon closer inspection, it seems unlikely that isolate Vib I, (red arrow, Fig 9.4) showed an ambiguous sized band, as first thought. In this isolate only two TTSS I sections were amplified with primers for vcrDJ and vscCJ, showing what appeared to be an ambiguous sized fragment between 415 and 458 bp. However, dot blot analysis confirmed that all three loci were present and nucleotide sequence analysis would confirm the true identity of these bands. Length polymorphisms in both virulence and housekeeping genes are widely reported in bacterial species. For example, PCR of the gene encoding toxin A (tcdA) in Clostridium difficile has shown length polymorphisms that enable some strains to be distinguished from others and a simple PCR to amplify capsule genes of Haemophilus influenzae can be used to discriminate between isolates (Rupnik et al., 1998; Falla., et al. 1994). However, in this study, it was necessary to confirm isolates lacking a PCR product by hybridisation 218 analysis. A lack of a PCR product may be due to nucleotide mismatches, the lack of primer target sites in some isolates or even simply due to a reagent missing in the PCR reaction mix. Conversely, the presence of priming sites does not necessarily mean the correct region is being amplified in a strain especially if amplicon sizes vary from the expected. The finding of isolates that were positive for hybridisation and negative for PCR (and vice versa) illustrates the need for careful interpretation ofPCR and hybridisation data. To overcome the magnitude of large scale Southern blotting and to save time, a dot blot hybridisation system was employed. Six DIG-labelled probes were developed based on the PCR amplification products that generated two different signals. The type strain 1902 showed weak PCR signals for the TTSS2 regions (Fig. 9.4) and hence three additional probes targeting vpal321, vscC2 and vcrD2 were also developed from an 03:K6 serotype isolate 220. Despite isolate 1902 showing low PCR signals for TTSS regions, the probes developed from 1902 showed almost identical hybridisation profiles to those of probes developed from the template of isolate 220. However, the limitations of the dot blot analysis technique meant that a more detailed comparison of TTSS fragments between isolates was not possible. The finding of weakly positive signals together with the 'halo effect' present on some of the dots (most likely as a result of impurities in the DNA mixture deterring DNA cross-linking to membrane) indicate some of the drawbacks of this technique. 9.4.3 Distribution ofTTSSl genes in V. parallaemolyticus Genes for TTSS I were found in all 17 clinical isolates of V parahaemolyticus in this study, including the two type strains; this is in agreement with previously published work (Makino et al., 2003; Ono et al., 2006; Park et al., 2004b). However, the presence of these genes in environmental isolates is much more variable. Out of37 environmental isolates of V parahaemolyticus used in the study, PCR and dot blot analysis revealed that four 219 isolates (VP345, VP361, VPECl and VPSCl) lacked all three ofthe TTSSl loci examined, suggesting that the TTSS 1 may not be an essential component for the survival of V parahaemolyticus and may not be intrinsic to this species as suggested previously (Makino et al., 2003; Ono et al., 2006). An additional seven isolates (1024/5, 1023/3, ViblO, VP6, VP7, VP331 and VP491) were identified as possessing TTSS1 by dot blot hybridisation but not by PCR amplification, suggesting a lack of primer target sites. These findings support existing studies that have identified TTSS 1 in all isolates of V parahaemolyticus, both clinical and environmental (Makino et al., 2003; Ono et al., 2006; Park et al., 2004b). A more extensive study with these environmental isolates using Southern hybridisation techniques and sequencing the entire TTSS region may help decipher structural variations in the TTSS I region. 9.4.4 Distribution of TTSS2 genes in V. parahaemolyticus Genes for TTSS2 were found both in clinical and environmental isolates of V parahaemolyticus. These include isolates that showed presence or absence of the virulence genes tdh and trh regardless of whether they expressed the Kanagawa W-haemolytic) phenotype or not (Table 9.4). This is contrary to earlier published information, suggesting that TTSS2 is only present in Kanagawa positive isolates (Makino et al., 2003; Ono et al., 2006; Park et al., 2004b). However, the findings in this chapter are consistent with the latest report that clinical isolates that are not related to the pandemic clone and are not Kanagawa positive also exhibit TTSS2 (Meador et al., 2007). Furthermore, as vpal 321 lies approximately 10 kbp outside of the recently described boundary of the TTSS2 region within the V parahaemolyticus pathogenicity island (Vp-PAI), it can also be proposed that all V parahaemolyticus carrying a TTSS2 system reside on the Vp-PAI (Probe D, Fig. 9.9), which is typical of several pathogenic species with TTSSs (Hueck, 1998; Winstanley & Hart, 2001 ). This is further supported by the finding of the eo-presence of the virulence gene tdh and TTSS2 in V parahaemolyticus (Table 9.4) in most of the isolates. A copy of 220 the tdh gene (TDH-A) has been shown to lie close to the TTSS2 region approximately 4.3 kbp downstream of vpal321 and, as expected, all but two isolates (VP7 and Vib9) that were positive for the vpa 1321 locus in this study were also positive for tdh (Makino et al., 2003). The presence of TTSS2 together with the Vp-PAI and tdh all pertain to potentially pathogenic V parahaemolyticus and thus routine analysis of all tlu·ee regions rather than the tdh alone would better enable the monitoring of these strains. Although TTSS systems are present in many Gram-negative species, to date only four other species have been characterised to possess two or more different TTSS in the same bacterium: S. typhimurium, Enterohaemorraghic E. coli 0157: H7, Yersinia enterocolitica and B. psueudomallei (Hueck, 1998; Foultier et a/, 2002; Stevens et al., 2002). In S. en/erica, the first TTSS cluster (SPI-I) mediates enterocyte invasion while the second (SPI-2) is responsible for the intracellular survival within macrophages (Hensel et al., 1995). However, whilst the first TTSS of enterohaemorrhagic E. coli (LEE) is responsible for the attachment and formation of lesions in epithelial cells, the precise function of the second TTSS (LEE2) remains unknown (Ideses et al., 2005). Similarly, the presence of both TTSS gene clusters in all clinical isolates may suggest a requirement of both systems for the pathogenicity of V. parahaemolyticus. TTSS I has been shown to secrete cytotoxic effector proteins from both the TTSS I and TTSS2 regions, but the role of TTSS2 remains unclear (Ono et a/, 2006). The finding of TTSS2 in at least I 0 environmental isolates with both the presence and absence of tdh and trh genes have demonstrated that TTSS2 is not exclusive to clinical isolates as previously thought and these genes are present in other members of the Vibrio spp. (Table 9.3, Fig. 9.5). 9.4.5 Sequence diversity of TTSSl regions Sequence analysis of three isolates within the same six loci, together with the comparison with a further two sequenced strains has revealed that the TTSS2 regions, where present, 22I are more conserved than TTSS I. Single nucleotide polymorphisms (SNPs) were widely found in all three TTSS I regions examined. These polymorphisms usually appeared on the third base of the codon and often did not alter the translated amino acid. In bacterial pathogens, a variety of SNPs have been identified that may provide selective advantage during a course of an infection, epidemic spread or long term evolution of virulence (Sokurenko et al., 1999). For example, Boddicker and colleagues observed that two strains of S. typhimurium with comparable phenotypes demonstrated an eightfold difference in adhesion to human laryngeal epithelial (HEp2) monolayers. Two amino acid polymorphisms were identified in the high- bindingfimH locus of the type I fimbriae and both changes were identified to two single nucleotide substitutions (Boddicker et a/, 2002). Although some SNPs may be favourable and enhance pathogenicity, most cases ofSNPs in both eukaryotic and prokaryotic systems are thought to result in gene knockouts (Weissman et al., 2003). In the present study, valine was replaced with isoleucine in position 119 ofvcrDJ ofTTSS1 in three pandemic 03:K6 strains (RIMD, 220 and VP416) upon comparison with the type strain. TTSS 1 of isolate RIMD221 0633 has been shown to be functional, and isolates 220 and VP416 exhibit strong ~-haemolysis (Chapter 4). However, the large number of base substitutions identified in all three isolates raises the issue of whether the TTSSl regions, more specifically the vcrDJ, vpl680 and vscCJ genes, are functional and whether the presence of valine is responsible for a more virulent ~ haemolytic phenotype. The finding of nucleotide deletions with resultant frameshift mutations in the vpl680 locus in all three isolates sequenced raises additional support to question the functionality of this gene. A further comparison of these loci in a larger number of isolates may provide more conclusive information. 222 9.4.6 TTSS genes as epidemiological markers for virulent strains To assess whether isolates possessing TTSS fell into distinct epidemiological clusters, RAPD-PSI groupings were compared with TTSS positive isolates (Table 9.4). The finding that most of the environmental isolates possessing TTSS2 clustered in a hierarchical group that included four clinical isolates was of interest. Of the I 0 environmental isolates that were confirmed for TTSS2 using both PCR and dot blot analysis, seven isolates fell into this unique cluster that was further divided into two sub-clusters (represented in red and blue, Fig. 9.1 0). Three isolates exhibited the presence of tdh; however, four isolates (1021/5, VP377, VP378 and VP939) in this group, as well as Vib3 from RAPD Type 24, were all negative for haemolysin genes tdh and trh. The close hierarchical clustering of these isolates negative for haemolysin genes and positive for TTSS2 warrants further investigation to determine their virulence status and to assess the feasibility of using TTSS2 as an additional virulence marker to tdh and trh. The finding of an environmental isolate (VP416) that is Kangawa positive (considered a strong indicator for virulent isolates) within this cluster may further suggest that other isolates may also be virulent in this group (Miyamoto et al., 1969). 9.4. 7 Identification of TTSSl and TTSS2 genes in other species Since the identification of TTSS in V. parahaemolyticus, genome sequencing of a number of related vibrios have shown the presence of TTSS I genes, including V. harveyi, V. alginolyticus, and V. tubiashii (Henke & Bassler, 2004; Park et a/, 2004b). This work has additionally demonstrated these TTSS genes in V. vulnificus (VV 11 067), V. anguil/arum (VanCMI 04) as well as confirming its presence in V. cho/erae and V. alginolyticus (Dziejman, et al., 2005). Only the genes in non-01, non-0 139 V. cholerae have so far been shown to be homologous to the TTSS2 regions of V. parahaemolyticus (Dziejman, et al., 2005). Despite the finding of TTSS genes in other species, little is known about their activity. The ubiquitous nature of vibrios mean that a large array of virulence genes, 223 including some identified in clinical isolates of other species, may confer upon this organism a greater degree of virulence. This chapter has described work to determine the presence of TTSS I and TTSS2 genes amongst both clinical and environmental isolates, something which had previously been thought to be a feature of Kanagawa positive (beta-haemolytic) strains only. Although the presence of the TTSSI has been shown to be responsible for translocating virulence determinants across the cytosol in V parahaemolyticus, its role in pathogenicity with regards to the haemolysins TDH and TRH has not yet been established (Ono et al., 2006). It is likely that the presence of both the TTSS systems together with the variously described effector molecules result in pathogenicity. The finding ofTTSS2 in a significant number of environmental isolates suggests that there may be more potentially pathogenic isolates in environmental reservoirs than currently thought. 224 Chapter 10 A COMPARISON AND EVALUATION OF METHODS USED FOR DIFFERENTIATING VIBRIO PARAHAEMOLYTICUS 10.1 Introduction In this present study, several phenotypic, serological and molecular techniques were applied in an attempt to assess and develop methods that may help differentiate pathogenic and non-pathogenic isolates of V. parahaemolyticus. Isolates were acquired from environmental, shellfish and clinical samples from several locations. Although etiological conclusions were hampered by the lack of information available on the isolates investigated, several techniques were successfully developed and applied. Phenotypic methods such as assessments of growth characteristics on general and selective media, P haemolysis, 0 and K serotyping provided more discrimination amongst the isolates. The ITSl-PCR and tDNA-ILP methods measured localised variation within PCR amplifiable 16S-23S and tRNA spacer regions, whilst PFGE and RAPD provided a measure of interspecies variation throughout the genome. PCR methods, in general, were easy to execute and demonstrated varying degrees of diversity. Previous chapters have discussed issues of typeability, reproducibility and discriminatory abilities of the individual typing system employed. This chapter aims as far as possible, to summarise, compare and contrast the properties and groupings obtained between different phenotypic and molecular typing systems by assessing measures of discrimination and concordance as applied to V. parahaemolyticus isolates throughout the study. 225 10.2 Materials and Methods 10.2.1 Phenotypic and growth characteristics of V. parahaemolyticus Assessments of growth and haemolytic characteristics were summarised together with findings of 0- and K- serotyping from Chapter 4. These methods are compared and evaluated. 10.2.2 Discrimination of molecular typing systems The discriminatory power of the typing methods was estimated by the use of Simpson's index of diversity, as defined by Hunter and Gaston (1988). Discrimination index (DI): 1 s n-(n -1) D = 1 - """.£.J J J N(N-1) j=l Where N is the total number of strains used in the sample population, s is the total number of types described and n1 is the number of strains belonging to the jth type (Hunter & Gaston, 1988). For example, PFGE typing with the use of restriction enzyme Notl (Chapter 5) shows 30 Pulse Types (n1 to n30) where n1 = I, n2 = 1, n3 = 3, 114 = 4, etc and N = 47. Calculations employing all 30 types give a discrimination index of 0.976. An index close to 0 indicates that little diversity is shown by the typing method whereas a DI approaching 1.0 reflects increasing diversity identified by the typing system. Discrimination indices were calculated for all five typing systems and fingerprint types 10.2.3 Concordance of molecular typing systems The concordance between the four typing systems was determined as previously using BioNumerics software (Version 5.1). Values from the similarity matrices of the two typing 226 systems under comparison were plotted on x and y axes. Each dot on the scatterplot represents corresponding similarity values for two isolates by any two of the typing methods. The graph therefore gives an indication of concordance between two typing techniques. The congruence, expressed as a percentage, was calculated in BioNurnerics using the Pearson product-moment correlation coefficient (PMCC). The regression (r), based on the least square fitting of the regression line for the data points, was also calculated in BioNumerics and gives an indication of the reliability of the regression model. A first degree (linear) model with monotonous fit was employed. Second (r2) and third degree (r3) models were also compared for correlation of x and y data. A dendrogram showing the relationships of all the typing systems, including fingerprints from combined tDl, tD2 and tD3 data (ID-combined), and well as combined RAPD-PS I and RAPD-PS2 (RAPD-combined) in the typing of V. parahaemolyticus was also produced. 10.3 Results 10.3.1 Growth of V. parallaemolyticus Except for two environmental and one clinical isolate, growth on a selection of culture media (Chapter 4) in general, did not show any obvious signs of variability. Isolates VPECI and VPSCI produced large amounts of exopolysaccharide on Wagatsuma agar and appeared clustered together in dendrogram analysis in all four of the molecular typing systems employed. The third isolate (6316) produced remarkably small colonies on TCBS agar, similar to V mimicus, with unique fingerprint profiles throughout the course of the study. Based on Gram-staining, oxidase and basic morphological identification analysis as outlined in Chapter 4, all isolates (except 6316) appeared biochemically and morphologically indistinguishable with no apparent differences in growth characteristics. However, growth kinetics in reduced Na+ levels from 3.0% to 0.5% NaCI, significantly increased doubling times by 30% and supports recent evidence of increased fitness of environmental isolates at higher salinities (Martinez-Utarza et al., 2006). Environmental 227 isolates also appeared more tolerant to cold temperatures demonstrating increased culturabilty from cold storage. 10.3.2 Haemolytic characteristics Out of 27 tdh positive isolates used in the study, only 12 (44%) isolates exhibited ~- haemolysis; an attribute considered significant amongst clinical strains (Sakurai et al., 1973; Nishibuchi et al., 1989; Nishibuchi & Kaper, 1995). Of the 12 environmental isolates possessing the gene tdh, only a single isolate, VP416, was haemolytic even though isolates were confirmed as TTSS2 positive in at least three other environmental isolates. The large proportion (29%) of environmental isolates in the study possessing the tdh gene is unusually high compared to typical levels of the tdh gene (I - 6 %) found in environmental samples (Nishibuchi & Kaper, 1995) and is clearly the result of sampling bias and the acquisition of clinically significant isolates from culture collections. The suitability of the Kanagawa test as a means of differentiating tdh positive strains is further questioned as three clinical isolates (2053, 5421 and El68l43) possessing tdh did not exhibit any haemolysis on Wagatsuma agar. This data suggests poor concordance between the gene tdh and haemolysis. The difficulty in the availability of fresh rabbit (or human) blood, limits the use of KP reaction for the routine surveillance of TDH. Difficulty also arose in assessing the degrees of haemolytic activity as it was apparent that varying levels of haemolysis were present (section 4.3.4). An attempt to quantify haemolysis levels attributed to TDH using a spectrophotometer showed little success. 10.3.3 Serovar diversity of V. parahaemolyticus Out of a total of 63 isolates of V parahaemolyticus, 44 were serotyped into one of 20 O:K types (Chapter 4). Serotyping had a good discriminatory power (0.937) where all 0 antigens were typeable with one of the 13 0-antisera commercially available. However, not all K-antigens were typeable and were designated KUT (Tablel0.2 & 10.3). Of the 71 228 K-antigens currently available for V parahaemolyticus, the isolates tested fell into one of 14 K-types that included KUT. Nine clinical isolates in the study were of the O:K6 serotype. A further two 03 serotypes (isolates 225 & 226) were K-untypeable but were positive for all four ITS! and RAPD markers that are found present in the pandemic clones (Table I 0.2). Within clinical isolates, an additional six serovars were present reflecting a diverse O:K population of disease causing types. A number of studies have shown a wide variety of serovars responsible for human disease (Nair et al., 1985; 2007). Serotypes amongst environmental isolates varied widely, with the predominant 01-type, present in SS% of environmental isolates. Other than the presence of 03:K6 isolates amongst clinical isolates, there appeared to be no other apparent serovar association in clinical and environmental isolates or their respective markers. As well as serotyping being a discriminative method, it also forms an easily reproducible technique due to the standardised reagents available in the serotyping kit. However, serotyping proved to be labour intensive and expensive. The availability of antisera for V parahaemolyticus 0 and K antigens is difficult, with only a few distributors worldwide (Kaysner & DePaola, 2004). With the rise of pandemic clones of serotypes 03:K6, 01 :KUT, Ol:K25 and 04:K68 as the dominant serovars, the present limited study highlights the investigation of the RAPD markers (Bands X and Y) for the potential identification of pandemic isolates. 10.3.4 Comparative analysis of pulsed-field gel electrophoresis In chapter five, Noti was used to generate restriction profiles of V parahaemolyticus with PFGE. Of the 59 isolates investigated using the methods described, a total of 47 isolates were used for dendrogram construction which included isolates that exhibited various degrees of DNA degradation. PFGE analysis, together with serotype variation, confirmed a heterogeneous population of V parahaemolyticus under investigation with 30 Pulse Types 229 and with the largest clusters (PT 04 & PT 18) containing only 4 isolates each. Despite the untypeability of approximately 20% of the isolates, PFGE showed the highest discrimination index (0.976) of all typing systems (Table 10.1). Apart from environmental isolates from Southampton (Vibl to Vib3 & Vib7 to Vib!O) that appear to be clonally related or even the same, isolates from the Dorset coast generally appeared untypeable; those that were typeable could not be assigned into clusters based on the location or the host source. In general, PFGE typing showed low correspondence with all the other typing systems (Figs. 10.1- 10.3) with congruence values ranging from 10.5% (with tD3) to 21.6% (with RAPD-combined). A comparison of PFGE with RAPD-PS1 and RAPD-PS2 showed that RAPD-PS2 was more congruent to PFGE than RAPD-PS I, with the composite RAPD PS 1 & PS2 data appearing more congruent (Figs. 10.1 & 10.2). Isolates that appeared grouped in pairs in PFGE such as VPSC1 and VPSCI or 6134 and VP416 were also identified grouped together in RAPD-PSI and RAPD-PS2 (Chapters 5 & 8: Figs. 5.11, 8.10 & 8.12. Clonally related 03:K6 isolates were also generally grouped together in all three fingerprint types to further support the reliability of these groupings. However, despite some similar groupings, there appeared general discordance. Firstly, isolate El54482 appeared closely related to two clinical isolates from the UK, E155855 and E168482 in RAPD-PS 1 whereas PFGE and RAPD-PS2 clearly differentiated isolate E154482 from the two pandemic 03:K6 isolates. Furthermore, a few groups were attributed to different lineages by dendrogram analysis of different fingerprint types. For example, isolates Vibl - Vib3 and Vib5- ViblO that were found grouped together in most of the fingerprint types, appeared in the same hierarchical branch as pandemic isolates in RAPD-PS 1. However, the same isolates appeared to branch off into different groups m PFGE. Such differences may explain such low congruence values. 230 10.3.5 16S-23S intergenic spacer (ITSl) based typing Independent experiments showed that the ITS 1-PCR typing method developed was reproducible with respect to the two main amplicons of approximately 390 and 635 bps. Other than isolate 6316, these two bands distinguished V parahaemolyticus from other species (Chapter 6). Up to a total of eight bands within 390 to 810 bps were present with considerable variability of intensity. At 93% similarity, 13 groups were identified that included a single large group accounting for 56% of the isolates. An overall discrimination index of 0.665 and 100% typeability amongst the 56 isolates (Table 10.1) was shown with ITS 1-PCR. This method was shown to be a simple technique to assess intra-species relationships but a thorough investigation with accurate band sizing techniques is required to determine whether clinical or environmental isolates possess unique banding patterns. Due to the unavailability of a complete set of PFGE data, a visual comparison of groups obtained by RAPD-PS I and ITS 1-PCR was made. Clonal groups (Table I 0.2) in ITS 1- PCR could not be as easily distinguished as RAPD-PSI. Based on ITS1-PCR, these clinical isolates appeared indistinguishable from several environmental isolates from Spain and Southampton, although at sufficiently high similarly values (>95%), clonal isolates appeared to be clustered together (Chapter 6: Fig. 6.11 ). Several other groups such as VPEC1 and VPSC1 and Vib1 to Vib!O were also found clustered together in both RAPD- PS 1 and ITS 1-PCR.In addition to these concordant groups, a larger and significant number of discordant types were also identified. A visual representation of the data by means of a scatterplot demonstrated low concordance both with RAPD-PS I and RAPD-PS2 with values of 25% and 32% respectively (Fig. 10.4). A comparison with PFGE, even though a significant number of isolates were untypeable, similarly showed low correspondence (Fig. I 0.2: B). It appears that these comparisons do not follow a linear model and using a quadratic or higher model (r2 or r3 respectively) demonstrated higher regression values (Table 10.3). However, a comparison of ITSI-PCR with tDNA-ILP showed a higher 231 degree of concordance of between 35.5 and 51.8 %. Of the three tDNA-PCR typing systems (tDl, tD2 &tD3), tD3 showed at least 16% less concordance of ITSI-PCR than tDl or tD2 (Fig. 10.1 ). In general, ITS 1-PCR did not differentiate V. parahaemolyticus isolates as well as PFGE or RAPD, but provided general information on the isolates available. Almost all of the clonally related (Table 10.2) types were found clustered together in the largest group (ITS 1-C) and thereby these isolates were not differentiated by ITSI-PCR. Vibl to Vib3 and Vib5 to Vib8 and ViblO that clustered together based on ITS! polymorphisms, could not be grouped together with any of the tDNA-ILP fingerprint types. 10.3.6 tDNA-ILP based typing The use of three sets of consensus tRNA gene primers provided a set of data representative oftDNA ILP typing of V. parahaemolyticus. Discrimination indices were 0.724, 0.568 and 0. 703 (for fingerprint types tDl, tD2 and tD3 respectively) with I 00% typeability in all three fingerprint types. Up to eight amplicons were identified in PCR analyses with a good degree of reproducibility of the main amplicons in repeat tests. Only a single isolate (VP10884) failed to amplify one ofthe four main amplicon of approximately 740 bp found amongst the 6 isolates tested. As with ITS 1-PCR, variation in reproducibility lay mainly in the amplification of weak bands in tDNA-PCR. A comparison of fingerprint types tDI, tD2 and tD3 with the combined data of all three tDNA-PCR experiments showed that tDI was the most differentiating of the three sets of primers and therefore used for comparative purposes (Fig. 10.1 ). A lower discrimination index was obtained for tD2 even though isolates in tD2 were grouped at a higher similarity value of 95% instead of 90% as with tDI further suggesting primers for tD2 was the least discriminative of the three sets of tDNA primers. 232 Fingerprint type tDNA-ILP (tDI) was compared to RAPD-PSI. A fair degree of concordance was demonstrated between the two fingerprint types (Congruence = 47%, r = 56.7%, Fig. 10.5:A). For example, isolates VPI to VPIO that formed two groups in RAPD-PSI was also identified within a subset of the largest group oftDI (tDI cluster A). Similarly, most isolates of the pandemic clonal group (Table 10.2) appeared clustered together in the comparison (Fig. 7.6, cluster tDl-C). Two further, smaller, groups consisting of isolates VP331, VP345, VP361, VPECI and VPSCl) were also grouped in both fingerprint types. It is not known to what extent this could be attributed to isolates placed on the same gel and therefore this needs to be addressed in further studies. Most other environmental isolates appearing heterogeneous in more sensitive methods such RAPD and PFGE were found clustered together and sometimes indistinguishable based on tDNA-ILPs. Likewise. a comparison oftDl with RAPD-PS2 demonstrated some degree of concordance (Congruence= 56.6%, r = 62.1 %). 10.3.7 Randomly amplified polymorphic DNA RAPD was shown to be the second most discriminative method after PFGE with discrimination indices of 0.965 and 0.959 for the two RAPD fingerprint types, PS I and PS2 respectively. The two-primer RAPD methods developed were almost as discriminative as PFGE (DI = 0.976) but unlike PFGE, did not suffer from issues of untypeability and were thus used as a method to which other typing systems were compared. RAPD identified 30 and 24 RAPD types at ~ 90% similarity with PS I and PS2 respectively, with good correspondence between the two RAPD types (Section 8.3.6 & Fig. 10.6). However, despite this discriminatory ability and congruence between two sets of RAPD primers, concordance with PFGE was low (Fig I 0.1 ). For example, Vib6 that appeared similar to clinical isolate El54482 by PFGE analysis was identified in different groups by both RAPD methods. Examination of ITSI-PCR and tDNA-ILPs found little correspondence between isolates Vib6 and E I 54482. Thus in this case, similarities observed by PFGE 233 could not be verified by other typing clusters. In addition, the RAPD method appeared to be the least reproducible in comparison with ITS 1-PCR and tDNA-ILP. Nonetheless, the RAPD method identified two possible markers that corresponded well with the pandemic group specific (GS) PCR (Table I 0.2). Further investigation is required to confirm whether such bands are unique to pathogenic isolates only. 10.3.8 Evaluation of the TTSS2 in V. paralraemolyticus Based on the data presented in Chapter 9, isolates confirmed as possessing the second type three secretion system (TTSS2) were those that gave positive signals for both PCR and DNA probe hybridisation for at least two of the three loci examined. The groupings and comparisons of TTSS2 positive isolates are shown in Table I 0.2. For comparative purposes, genotypes and characteristics of TTSS2 positive and TTSS2 negative isolates are presented in Tables I 0.2 and 10.3 respectively. It became immediately apparent that all clinical isolates and a significant number of environmental isolates possessed a TTSS2. 234 Table 10.1: Discrimination of V. parallaemolyticus typing methods used in this study Method Fingerprint No. of Size(%) of oe Similaritl % 2 {Ch&(!ter~ T~(!e T~(!eS largest l~l!e of t~l!es T~l!eabilit~ + Serotyping1 (4) n/a 20 19 0.937 n/a 70 PFGE (5) Not I 30 4 0.976 ~88% 80 ITSI-PCR(6) ITS I 13 57 0.665 ~93% 100 tDNA-ILP (7): tDI 7 45 0.725 ~90% 100 tD2 7 64 0.568 ~95% 100 tD3 9 46 0.703 ~95% 100 RAPD (8): PSI 30 13 0.965 ~90% 100 PS2 24 11 0.959 >90% 100 1 Serotyping is character type data; Chapter 4: Table 4.3. Typeability refers to untypeable K-antigen only. All 0-antigen typing, where performed, were typeable. 2 Includes types assigned at ~ 88% similarity. 3 Discrimination Index based on Hunter & Gaston ( 1988); a value closer to 1.0 suggests a more discriminate typing system. 4 Based on dendrogram + Typeability was scored as a percentage of the typing system to assign a type to each isolate (Struelens, 1996) 235 '2 c i ,.. :i5 .E U) E (I) ,Q 0 "' E 0,. '/ 0,. 0 c c c w ... 0.. 0.. (.!) y N U) ... c( LL Q e e ""9 t: ~ ~ ~ a. tD1 tO-combined tD2 tD3 ITS1 RAPD-PS1 RAPD-combined 28.8 22.4 34.0 RAPD-PS2 17.4 32.3 32.8 PFGE Figure 10.1: Summary of concordance for aD molecular typing systems. Matrix shows percentage congruence values as calculated with the PMCC in BioNumerics. Composite data for all three tDNA-PCR and both RAPD experiments are marked as tO-combined and RAPD-combined respectively. Dendrogram shows a measure of tl1e relatedness of the molecular typing systems based for the typing of V parahaemolyticus. 236 0 0 ~ A . ' ..: . . • : · :.· ! -:_ .' ·,~. ·. • .. .. ~· .: :~:: ·:.: :.):· :t C= 13.2% ,I ·'• •; :;, .· ,':·· : :: :·:· · .._: :; ':: r=34.4% ,.... ·.·: . . · : ·.· .=: ~: .. : . . . . : . # : (/) 0.. . ' 0 : .: 0.. <( 0:: 60 70 80 100 PFGE typing (% Similarity) B .. : : .· .: · : ! • '•:. :. I .. ·, • ~ . : : .. . . . ·. :· ,. . : . : . !': ·...... ; . ' I • ' I' • '• • ~ ;... : : .· ' .. C = 17.7% . : t:;'• .=.. ~ ·::- : : . : . ~ .. :.:: : . ·.. ·. r = 38.7% . ~- . I . , • • . :' .' ; ; . · .· : PFGE typing (% Similarity) Figure 10.2: Concordance of stratification by PFGE and RAPD for V. parallaemolyticus isolates. Each dot represents corresponding similarity values for two isolates obtained by PFGE and RAPD-PS I typing (A) and PFGE and RAPD-PS2 typing (B). Concordance as calculated with the PMCC is shown as C and regression by r. 237 A >: .t,_ .E..!!! i:i5 "::!?. ~ 0> C= 16.2% ·a..c c r = 18.9% 0:: () 0... __...- I ___. <( z 0... -- PFGE typing (% Similarity) 0 0 B ·: ...; :. ; · ~ .. ~ ~ t ~~-~ ::· -~ ' I • ..• ·.: ·.. . : ... , • I • ' .. ·.. .. .=:. · .. ·:.. : : :: ; . :~ :; :: ~ . · :. 0 CO s·c ·. · .. : . ~ E ii5 ~ 0 ... ·. C= 13.8% ~ <0 0> c r = 17.4% ~ ...- (/) t: 0 ' 0 N PFGE typing (% Similarity) Figure 10.3: Concordance of stratification by PFGE and tDNA-ILP typing (A), and PFGE and ITS I typing (8) for V. paraltaemolyticus isolates. Each dot represents corresponding similarity values for two isolates obtained by PFGE, tDNA-PCR (tD I) and ITS 1-PCR. Concordance as calculated with the PMCC is shown as C and regression by r. 238 0 0 A I I c = 25.0% r = 40.3% ..... (/) a_ ..·.·· ..,: · ... 0 0 .. :·· :. . :":",', a_ .., ;.t:, <( 0:: 0 N 20 40 60 eo 100 ITS1 typing (% Similarity) 0 0 ~ B . • .• t. ~.. .. ·. ... : .:~·::·: ;~:i: .·_...... \ .· .··. ~g ·c ..!!! : ~ en ~ 0) C =32.6% c ·a. 0 ?;-eo r = 46.7% N en ; ...... ~-~-. ': . :· -~ .. : :~: .. a_ . . ·.. a_0 .· .... -~? :=·.·:· ... ·-F ..· .. ~ 0 V 20 40 60 80 100 ITS1 typing (% Similarity) Figure 10.4: Concordance of stratification by ITSI-PCR and RAPD for V. parallaemolyticus isolates. Each dot represents corresponding similarity values for two isolates obtained by ITS 1-PCR and RAPD-PS I typing (A), and ITSI-PCR and RAPD-PS2 typing (B). Concordance as calculated with the PMCC is shown as C and regression by r. 239 0 0 A ~ I I .I c = 47.0% ·.• . . '• . ·:·. ·! .. ::·' ..... r = 56.7% a..(/) 0a.. 0..,. <{ Q:: ... ,. 0 N / 60 tDNA·PCR typing (% Similarity) 0 0 B >: 1:: 0 .!!! <0 E Uj ~ 0) C=56.6% c: ·c. . . · ·.·: z, r =62.1% N g . . (/) ,., .: ·~ . ' . .. ·.' ··... · .: :~ ... a.. . ··. "·: ·· .. ··.., .: 0 ·: / a.. . : ... ~ ' . . . ~ ·. :-· ..,.0 I tDNA-PCR typing (% Similarity) Figure 10.5: Concordance of stratification by tDNA-ILP and RAPD for V. paralwemolyticus isolates. Each dot represents corresponding similarity values for two iso lates obtained by tDNA-PCR and RAPD-PS I typing (A), and tDNA-PCR (tO I) and RAPD-PS2 typing (B). Concordance as calculated with the PMCC is shown as C and regression by r. 240 C = 46.5% r = 62.9% 0 0 0 CO Cl) 1- 0 N 0 20 40 60 80 100 tDNA-PCR typing (% Similarity) Figure 10.6: Concordance of stratification by tDNA-ILP (tDI) and ITSI-PCR for V. paraltaemolyticus isolates. Each dot represents corresponding similarity values for two isolates obtained by tDNA-PCR and ITS 1-PCR typing. Concordance as calculated with the PMCC is shown as C and regression by r. 241 Table 10.2: Comparison of V. parahaemolyticus TTSS2 isolates and typing groups to distinguish vir ulent and potentially virulent strains. All isolates are confirmed TTSS2 positive both by PCR and dot blot hybridisation. PCR Typing groups Isolate Serotype KP2 tdh ,,, GS PF ITS I tDt tD2 tDJ RAPD RAPD PSt PS2 Clinica l 220• 03:K6 + + + 2 1 c D A 8 24' 23* 222• 03:K6 + + + + 22 c D A 8 24* 23* 223• 03:K6 + + + 18 c D A 8 24* 23* 227• 03:K6 + + + 20 c D A D 24* 23* El 55855e 03:K6 + + + 18 c A c c 21' 15* El68143e 03:K6 + + 18 c A c 8 21' 15* 2053• 03:K6 + + 18 c c F F 10* 15' 5421• 03:K6 + + 20 c c A 8 10* 15* 22 1• OI:KUT + 20 c D A 8 25 22* 225• 03:KUT + + UT c D A 8 24* 23* 226• 03:KUT + + 19 c D A D 24' 23* 219 06:KUT 27 8 D 8 8 27 22 224 03:K6 + + 24 c D 8 8 26 22 228 OII:KUT + 02 c D A D 24 22 2436 06:Kl8 + + + UT c D A A 11 16 6134 04:Kl2 + + 09 c D A 8 12 08 El54482 Ol:Kl + 15 c A A ;\ 21 18 Type Strains 1902 O I:Kl 25 A B c G 19 11 VPI0884 03:K4 + + 07 1-1 D A D 17 20 Environmental 102 1/5 02:KUT UT c 8 c 1-1 04 10 Vib3 Ol:K32 03 K A A B 23 17 Vib7 NO + 04 c A B 22 09 VP2 NO + 10 c ;\ A B 20 12 VPIO NO + + 26 c A A B 18 14 VP33 1 NO UT c p D c 28 03 VP349 NO + UT A c E c 06 06 VP350 NO + 08 L ;\ E c 06 06 VP377 04:KUT UT E A E c 05 06 VP378 06:KI8 14 E A E c 09 07 VP416 04:KUT + + 09 E A E c 12 08 VP939 04:KI8 23 ~I c E G 13 19 1 Wagley et al., (2008) tdh: thermostable direct haemolysin (TOH) gene trh: TOH related haemolysin gene GS: Group Specific PCR for pandemic types (Matsumoto et a/, 2000) 2 Kanagawa phenomenon - P-haemolytic isolates 3 Groups based on ~ 88 % similarity. PF: Pulsed-field gel electrophoresis with Notl, ITS: ITS 1-PCR, tO l-t03: tONA-PCR with three primer pairs, RA PO-PS I and RAPO-PS2 types indicated with * correspond to distinct amp! icons (Band X and Band Y) as possible markers of pandemic strains in RA PO-PS I and RAPO-PS2 respectively. NO: Not determined UT: Untypeable Clonal C• ) and clonally related isolates (•) based on GS-PCR, PFGE, RAPO, ITS and IONA PCR. 242 Table 10.3: Comparison ofTTSS2 unconfirmed (UC) or TTSS2 confirmed negative (-) isolates of V. parallaemolyticus to potentially distinguish avirulent strains. All isolates are environmental. PCR1 Typing groups Isolate Serotype KP2 tdll !rh GS PF ITS I tDI 102 IDJ RAPD RAPD PSI PS2 ITSS2 (UC) 1023/ 1 OI:K32 UT c A A c 04 04 1103/9 OI : KI 13 c A A c 02 07 Vibl OI :K32 03 K A A B 23 17 Vib3 Ol:K32 03 K A A B 23 17 Vib6 ND 16 J c A B 22 09 Vib8 ND 04 E A B 22 09 Vib!O ND + 04 c A B 22 09 VPI OI :K33 10 A A A A 20 12 VP3 ND + 11 c A A B 20 12 VP4 OI:K33 12 c A A B 20 12 VP8 ND + + 28 c A A B 16 13 VP9 ND + + 29 c A A B 16 13 VP345 ND + UT c F D c 28 03 VP346 OI:KUT UT D c A c 07 21 VPECI ND + 05 H F D B 30 01 TTSS2 (-) I 024/5 03:KUT ND ND UT c c A c 03 05 1022/10 OJI:KUT 01 c A A c 01 04 1023/3 ND UT c A A c 08 04 Vib2 OI:K32 03 A A B 13 17 VibS Ol :K32 10 K A A B 23 17 Vib9 ND + 0~ ND ND ND ND ND ND VP6 02:K28 30 c A A B 16 13 VP7 ND + + 30 c A A A 16 13 VP361 O I:KUT UT F F D c 29 03 VP491 ND UT G A E c 14 24 VPSCI 08:KUT 05 H I' F E 30 0 1 1 Wagley et al., (2008) tdh: thermostable direct haemolys in (TDH) gene trh: TDH related haemo lysin gene GS: Group Specific PCR for pandemic types (Matsumoto et a/, 2000) 2 Kanagawa phenomenon - P-haemolytic isolates 3 Groups based on 2:88% similarity in typing systems. PF: Pulsed-field gel electrophoresis with Not!, ITS: ITS I-PCR, tDI-tD3 corresponds to tDNA-PCR with three primer pairs. ND: Not determined UT: Untypeable 243 Table 10.4: Analysis of regression in comparison of typing systems. Regression models r" r2 and r3 correspond to first, second and third degree regression models respectively. Values are as calculated from BioNumerics software using the method of least squares. Figures l 0.2 to l 0.6 show a r 1 linear regression line whereas in some instances r2 and r3 curves provide better data fit. Increasing regression values were obtained with higher regression models in comparisons where localised variation (i.e. typing systems tDNA-PCR or ITSI-PCR) was compared with high discriminatory genome-wide variation (i.e. PFGE or RAPD). T:yuin2; comuarisons Regression PFGE ITS I-PCR tONA-ILP (tO I} model PSI PS2 tO I ITS I PSI PS2 PS I PS2 ITS I rl 34.4 38.7 18.9 17.4 40.3 46.7 56.7 62.1 62.9 rz 40.8 51.1 64.4 57.1 58.3 62.7 74.2 78.5 63.3 rJ 40.9 51.1 79.6 73.8 65.4 67.8 77.2 80.9 63.4 244 10.4 Discussion Subtyping or isolate classification can be accomplished by using a number of different techniques. These methods must meet several criteria in order to be useful. A comparison of different typing methods used in this study has outlined various degrees of typeability, reproducibility and discriminatory powers, together with an evaluation of the ease of use of a typing system. This chapter aimed to compare and evaluate the groupings obtained to help establish relationships and assess the reliability of the methods and the relatedness infened by different characterisation and typing systems. 10.4.1 Phenotypic methods for isolate differentiation The TDH phenotype could not be attributed to a significantly large proportion of tdh gene possessing isolates. The haemolysin TDH is known to be a pore forming toxin that causes 2 cytotoxicity in a Ca + independent pathway (Honda et al., 1992; Tang et al., 1995; Takahashi et al. , 2006). However, the precise mechanisms of its mode of action still remain unknown. The present data show a greater proportion of environmental isolates possessing the gene tdh than clinical isolates not exhibiting haemolysis. It is not known whether the gene tdh is switched off and/or present in an inactive fonn amongst environmental isolates. The conserved nature of tdh (Nishibuchi & Kaper, 1995) may suggest, however, that the copy is functional. The significance of finding tdh in environmental isolates therefore remains unclear. Of interest would be to assess the actual numbers of environmental isolates harbouring tdh genes to exhibit haemolysis. Non haemolytic clinical isolates possessing the gene tdh indicate that passage in the human host may not be responsible for the activation of tdh genes. As far as the author is aware, antisera for V parahaemolyticus 0 and K antigens is only produced by Denka-Seiken in Japan with a few distributors worldwide (Kaysner & DePaola, 2004 ). The costs and labour associated with serotyping, together with the 245 untypeabilty as shown in the present study, and by others (Nair et al., 1985) suggest that serotyping of a large number of environmental isolates may not a feasible option. 10.4.2 Typing methods of V. parallaemolyticus PFGE was performed so that it could be used as a "benchmark" typing method to which additional typing systems, which were simpler to perform, could be compared. Despite the recent trend towards nucleotide sequencing typing methods, to date, PFGE remains the most widely adopted method for typing nosocomial, community acquired and foodborne pathogens (Tenover et al., 1997; Hyytia-Trees et al., 2007). Due to the large degree of untypeability of V parahaemolyticus by PFGE (and thus the unavailability of a complete set of PFGE data) for all 56 isolates of V parahaemoly ticus, an initial comparison of PFGE with both RAPD-PS 1 and RAPD-PS2 was undertaken. Struelens (1996) suggested that a sufficiently discriminate typing system should have a discrimination index of at least 0.950 and therefore both PSI and PS2 have shown to be adequate for subtyping of V parahaemolyticus. In a previous comparative study, using a single primer RAPD method, Wong et al., (2001) identified several typing systems with RAPD as the lowest discrimination index (0.840). The workers found ERIC (DI = 0.980) and REP-PCR (DI = 0.97) to be more discriminatory than PFGE (DI = 0.960) based on 40 strains typed (Wong et al., 2001). Whilst the present PFGE data (DI = 0.976) support these findings, the two-primer RAPD method (DI = 0.959- 0.961) has been shown to be a more discriminate method and presents an enhanced typing system achievable without significant addition to costs or labour times. However, based on the present study, low concordance of PFGE was shown with all the typing systems. PFGE suffered from high degrees of DNA degradation where a significant proportion of isolates were not typeable. Of those that appeared typeable, a further subset (such as isolate 226 and VP6 - VP9) showed distinctly weak bands. It is likely that some bands could not be identified amongst 246 the DNA smears. Furthermore, based on scatterplots and regression analysis it seems likely that some of the comparisons could better be associated with quadratic or higher models (i.e. r2 or r3) rather than a linear relationship. A comparison between first degree (linear) and non-linear models showed better data fit with non-linear models in some instances. This seemed more apparent when comparisons were made between high and low discriminatory typing systems. BioNwnerics employed a Pearson product-moment correlation coefficient to assess the linear relationships between two comparative typing methods. As some models don't appear linear, it is possible that small changes apparent in ITS 1 and tDNA spacer regions identified from typing experiments could be indicative of greater or lesser genome variation as measured by RAPD or PFGE. The use of RAPD for typing V parahaemolyticus has been suggested by others, above all on the grounds of the ease of use before more labour intensive approaches such as MLST and VNTR analyses are performed (Giske et al. , 2006). However, inter-laboratory reproducibility of this method is a widespread concern and has been discussed earlier (section 8.4.6). Despite this, the data presented in this thesis show some degree of reproducibility in independent experiments and tills may be partially due to the use of standardised reagents such as RAPD analysis beads (Amersham Biosciences). However, in general, levels of reproducibility did not appear to be adequate and further work is required. Based on the preliminary studies described in this report, the use of ITS 1-PCR and tDNA-ILP offer less discrin1inative alternatives to RAPD. The work was caJ.Tied out with standard, easily available laboratory reagents and may prove to be more robust than RAPD in inter-laboratory assessments. For example, changing annealing temperatures did not significant! y affect banding patterns in ITS 1-PCR in the present study and therefore variation arising as a result of PCR amplification at rugh annealing temperatures could be more attributed to mutations in the primer binding sites. This is in contrast to the low annealing temperatures for RAPD that would result in more non-specific primer 247 hybridisation to DNA template (Bart-Delabesse, et al., 2001). Therefore, ITSl-PCR and tDNA-ILP warrant further analysis on larger sample numbers. The use of more sensitive typing methods such as PFGE and RAPD may not always be appropriate. For example, environmental populations of V parahaemolyticus, as shown in this study, and by others, are genotypically diverse (Nair et al., 1985). Both clinical and environmental isolates were identified as single members in 18 and 20 of their own types in RAPD-PSI and PFGE respectively. Therefore, methods such as ITSl-PCR or tDNA- ILP may be satisfactorily used as a first step towards strain confirmation and differentiation. The present work has collectively demonstrated that the less discriminate methods grouped distantly related V parahaemolyticus isolates and provides useful low resolution typing information. Combining data from multiple experiments, when high similarities are obtained (i.e. tD-combined) can provide greater differentiation than single primer pairs alone but in the present case, combined data appears to group together isolates from the same gels more than single gels alone thereby exemplifying the need for random isolate distribution on a gel and careful data analysis. As described in section 6.4.5, a number of tRNA genes can be found within ITS 1 regions of V parahaemolyticus (Makino et al., 2003). Whereas ITS 1-PCR provided a measure of spatial heterogeneity between conserved ribosomal genes, the genome wide distribution of tRNA genes provided a method in which more of the genome was subject to analysis. This may help to explain the discrepancy in concordance of 46.5% (linear regression = 62.6% fit) between ITS 1 and tDNA (tD l) ILP typing. It is not known if the tRNA gene primers used were specific for particular tDNA regions encoding for specific tRNA classes or whether such inferences could be made. However, based on the amplicon sizes it would appear that there were limitations posed by PCR amplification conditions. For example, the largest amplicon from ITS 1-PCR (81 0 bp) was expected in three copies ((Makino et al., 248 2003; Gonzalez-Escalona et al., 2005b; 2006a) and hence to result in an intense band. Clearly this was not the case observed as arnplicons above 700 bp appeared faint (Chapter 6). Similarly, tDNA-ILP bands (Chapter 7) above 600 bp appeared faint. In both cases, there were significant amounts of background PCR noise that would need to be addressed in further studies employing these methods. 10.4.3 General reliability of clustering patterns It was apparent that no two typing systems used m the present study gave a high concordance and there remain questions on the reliability of the data obtained. The discordance of PFGE with other typing systems highlights the need for the use of an additional restriction enzyme for PFGE analysis. As both RAPD-PS 1 and RAPD-PS2 appear congruent to each other and to some degree with ITS 1-PCR and tDNA-PCR typing, it is likely that the current PFGE data needs further reassessment. For example, similar banding isolates such as VP6 - VP9 appeared distant on the PFGE dendrogram that may suggest sub-optimal BioNumerics band tolerance settings. As mentioned earlier, this could also arise if faint bands were not selected in the analysis. A similar explanation of isolates that appeared to share the same apparent fingerprint pattern yet were clustered differently in ITS 1, tON A and RAPD dendrograms can be provided. It is also worthy of note that analysis of PFGE was performed using the Dice, band-based method, whereas other typing systems employed a curve based method. Analysis of the concordance tests were limited by the reliability of the data originally obtained and may further explain low congruence values in the comparisons. Discordant results between RAPD and PFGE have been reported before. In a recent external quality assessment collaborative study with a network of laboratories, RAPD was shown to be 47% concordant to PFGE with Staphylococcus aureus isolates (Deplano et al., 2006). In another study, RAPD demonstrated only 26.5% concordance with other typing techniques such as RFLP (Bart-Delabesse, et al., 2001). Despite such discrepancies, RAPD continues to be widely employed in typing systems. 249 It was found with PFGE that isolate 6316 showed more similarity to Vib6 and El54482 than any of the other isolates, whereas both lTSl-PCR and tDNA-lLP typing clearly discriminated 6316. With a large number of bands (> 15) and a few common bands, this isolate did not show as much variability by PFGE for it to be classified as such. As mentioned earlier, isolate 63 16 showed unique growth characteristics similar to that of V mimicus. The placement of this isolate in the PFGE dendrograrn thus seems inappropriate and can be used as a measure to further illustrate the genetic diversity of V parahaemolyticus isolates under study. Therefore ITS 1-PCR or tDNA-ILP typing provided further evidence that isolate 6316 may not be closely related to V parahaemolyticus or may even be a different species altogether. Inconsistencies in groupings obtained in the typing systems may also pru1ially be attributed to the method used for dendrograrn analysis. In the 'curve based' analysis methods used, it was found that the presence of much darker bands gave unreliable clusters where faint bands apperu·ed ignored. Thi s has been identified by others as an inherent problem of curve based analysis methods (Werner et al., 2003; Rasschaert et al. , 2005) and based on present data seems more of a concern when there are few bands generated in the typing system. 10.4.4 Differentiation of clonally related V. parahaemolyticus There are an increasing number of reports of the prevalence and dominance of clonally related V parahaemolyticus in disease outbreaks worldwide (Chiou et al., 2000; Chowdhury et al., 2004b; Ansaruzzarnan et al., 2005; Rizvi et al., 2006; Serichantalergs, et al., 2007). To date, these isolates generally have been shown to fall into one of four serotypes: 03:K6, Ol:KUT, 01 :K25, and 04:K68 (Chowdhury et al., 2004a), and appear indistinguishable even with PFGE methods, contrary to the belief that isolates with highly related PFGE patterns shru·ed the same phenotype (Hall et al., 1996). Only a small number 250 of isolates appeared clonal in the present study (four isolates of PFGE type 18; Fig. 5. 11) . The remaining seven isolates were categorised as clonally related (Table 10. 9) due to incomplete isolate histories. Based on this data alone and the limited number of isolates, it is difficult to postulate the extent to which PFGE has differentiated clonally related isolates in this study. Recent developments in MLST for V parahaemolyticus provide more appropriate means to assess diversity amongst isolates indistinguishable by PFGE or other means (Chowdhury et al., 2004b; Gonzalez-Escalona et al. , 2008). Amongst a total of 37 clinical and 63 environmental isolates, three major clonal complexes and 62 sequence types were identified amongst seven house-keeping genes (Gonzalez-Escalona et al., 2008). The stud y showed the same MLST sequence type (ST) in fo ur different serotypes and thus further supports the non-correspondence between serotype and molecular based typing systems. 10.4.5 Conclusion To date, there appear no published studies of the typing of V parahaemolyticus by ITS 1- PCR or by tDNA-ILPs. This study is the first to compare these methods wi th RAPD and PFGE typing. The initial findings in this work suggest both ITS 1-PCR and tDNA-ILP methods are applicable for routine analysis wi th a two-primer RAPD method far easier to execute than PFGE, but gave almost equivalent discrimination. The different molecular typing systems generated different groupings and lineages, and the results, in general showed various degrees of concordance. This was expected with different typing systems as each typing method targets different loci and in some cases not enough loci may be targeted to result in a true reflection of genome diversity. For example, it is unlikely that ITS1-PCR typing with species with a low ribosomal RNA copy number would give sufficient variation of length polymorphisms to be able to distinguish between isolates. Serotype data and the PCR identification of virulence genes in shellfish and seafood 251 samples, although useful adjuncts to characterise pathogenic strains, do not necessarily provide useful information about the virulent status of the organism. The identification of TTSS2 in all clinical isolates and a significant proportion of environmental ones, may be a more useful a marker for virulent V parahaemolyticus. 252 Chapter 11 GENERAL DISCUSSION AND CONCLUSIONS 11.1 Introduction Members of the genus Vibrio are ubiquitous in the marine and estuarine environment. Accurate and reliable methods for the identification and differentiation of pathogenic vibrios are thus essential if epidemiological studies are to be successful. This study employed several simple methods that could easily be adopted to assess diversity amongst populations of Vibrio spp. commonly found in shellfish and other seafood samples. Additionally, as conventional approaches to the identification and confirmation of V parahaemolyticus are lengthy and labour intensive, some of the methods described in this work may help to identify V parahaemolyticus and ease the epidemiological assessment of large samples. Methods to rapidly and easily distinguish pathogenic isolates of V parahaemolyticus from environmental reservoirs of mixed vibrio populations are necessary to (i) address seafood safety concerns; (ii) monitor the spread of vim lent strains; and (iii) to study and understand the evolution of V parahaemolyticus. Knowledge of pathogenicity traits and specifically the spread of virulent forms, may further enable the development of models that may be applicable to other pathogenic bacteria from the marine ecosystem. Following on from the methods comparison in the previous chapter, this discussion attempts to address the use and applications of the adopted strain differentiation methods of V parahaemolyticus in a wider context in light of the findings presented in this thesis and recent literature. 253 11.2 Scope and feasibility of the present study Although traditional biochemical tests and growth experiments proved to be labour intensive, they may still be useful because of some recent evidence to suggest biochemical heterogeneity amongst clinical and environmental isolates (Chapter 4; Martinez-Urtaza et al., 2006). This study aimed to apply existing molecular typing methods and to develop new ones to assess whether pathogenic V parahaemolyticus could be differentiated from non-pathogenic isolates. It was envisaged that by targeting different loci with several typing systems, the study would (i) establish a measure of the diversity present in V parahaemo/yticus genomes; (ii) allow the identification of V parahaemolyticus based on source; i.e., environmental or clinical, animal host or geographical location; and (iii) determine whether a proportion of environmental isolates may possess characteristic attributes similar to established vimlent strains as a means for classifying potentially pathogenic isolates. However, the scope of the present work did not include establishing the pathogenicity status of any of the environmentally acquired isolates. This study was conducted with a limited number of isolates but acquired from several sources (section 3.2.1) and therefore forms essentially a preliminary study of general molecular typing of V parahaemolyticus. The methods described in this thesis should be validated on a larger population of isolates to assess the in-vitro stability of the ITS 1-PCR and tDNA-ILP methods. The isolates used were acquired and typed over a period ofthree years (2003 to 2006). Based on a single laboratory user, and two PCR machines (section 3.6), the methods described seem stable, reproducible and applicable to other members of the Vibrio genus. The ease of use and reduced time required in comparison to techniques such as PFGE facilitated the generation of multiple fingerprint types with different primer pairs. Perhaps the most significant advantages of both ITS 1-PCR and tDNA-ILP methods are their practicality, as the generation of a small number of amplification bands enables easier visual assessment without the need for complicated software analysis. If these 254 methods are to be further explored, a larger proportion of clinical isolates with more complete histories and from more diverse geographical sources would be recommended. Based on the percentage of tdh and trh genotypes present in the isolates tested, it seems unlikely that the environmental isolates used in the study represent a true reflection of tdh or trh genes present amongst environmental reservoirs of V parahaemolyticus (section 10.2.2) and therefore, a bias towards potentially pathogenic strains. Such sample bias in favour of tdh + isolates may initially have suggested greater relationships amongst those environmental and established clinical isolates. Although tdh + isolates were found in several clusters, much the same as TTSS2 positive isolates (Table 9.5; Fig. 9.10) there appeared little evidence of environmental and clinical tdh + isolates being grouped together. This would suggest that such haemolysin genes have been incorporated into the genome more recently. However, despite these differences there was some evidence to suggest that isolates obtained from the same shellfish species could be clustered using the methods described. For example, the majority of isolates sourced from oysters were found in separate groups to those isolates from other shellfish species. Isolates VP491, VP939 (obtained from scallop and crab respectively) were clearly differentiated from each other and the majority of those obtained from oysters. 11.3 Genome plasticity of V. parahaemolyticus The isolates were found to be from diverse 0 and K antigenic serotypes and this heterogeneity amongst V parahaemolyticus was clearly confirmed by PFGE analysis with Notl. Amongst this heterogeneity, and as expected, environmental isolates accounted for the majority of the serotypes (70%). In both clinical and environmental cases, isolates of the san1e serotype showed distinct PFGE genotypes; conversely, isolates of similar PFGE patterns were of different serotypes. This phenomenon has been demonstrated widely in V parahaemolyticus, as up to 21 different serovariants have been found to be clonally related 255 to the parent 03:K6 clone (Martinez-Urtaza et al., 2004; Nair et al., 2007). It has been suggested that the acquisition of additional serotypes may be a selected response to host immunological pressure (Chowdhury et al., 2004b). Such changes in the 0- and K antigens thereby clearly illustrate the fluidity of V parahaemolyticus. Based on the variations in length polymorphisms identified in both 16S-23S (ITSl) and between tRNA genes, it seems highly likely that some of these regions are undergoing large degrees of change in V parahaemolyticus populations. Based on banding profiles alone, it appears that some of the tDNA regions varied more than others amongst environmental isolates. The largest amplicon with primers T3A/T5A (tDl) showed the greatest variation, whereas this amplicon appeared uniform in most of the clinical isolates (Fig. 7.2). By using another pair of discriminatory primers T3B/T5A (t02) it was possible to confirm and establish the differences in relationships. Thus, the use of a secondary or even tertiary fingerprint pattern (i.e. t03) would be essential to establish true diversity and hence more accurate relationships. The presence or absence of bands in the tDNA-ILPs suggests there may be rearrangements in the genome that need to be examined to assess for possible insertions of mobile genetic elements. Of the typing systems employed, PFGE revealed the most variation in isolates and very recently, standardised methods that show increased reliability for typing, have been put forward (Wong et al., 2007; Kam et al., 2008). Variations in PFGE and RAPD profiles in bacteria are due to discrete insertions or deletions that contribute to restriction or primer site changes between strains (Kudva et al., 2002). In E. coli, it is thought that bacteriophage variation is a major factor in generating genomic diversity that underlies genomic variability detected by PFGE (Ohnishi et al., 2002). With the identification of an increasing number of non-host specific vibriophages (Matsuzaki et al., 2000; Millar et al., 256 2003; Comeau et al., 2006) in the marine environment, it is likely that V parahaemolyticus will continue to evolve at a relatively fast rate. 11.4 The differentiation of clinically significant V. parahaemolyticus This work was based on the founding molecular typing principle that epidemiologicall y related isolates are derivatives of a single parent isolate (Streulens, 1998). Therefore, pathogenic isolates share characteristics that may differ from those of non-pathogenic isolates within the same species. The typing experiments carried out did not give a single distinct cluster of pathogenic V parahaemolyticus. Clinical isolates were identified in several different clusters and in some cases clinical isolates formed single members (219, 224,228 and E l 54482) of their own groups. PFGE analysis identified fo ur members of the clonal pandemic 03:K6 group, which belonged to a larger cluster (>76% similarity) of 10 entirely clinical isolates. Another cluster (>76% similarity) contained five isolates, four of which showed P-haemolysis (an environmental and clinical isolate appearing closely related) and clustered with an environmental isolate, VP350 (Fig. 5.11). Similar fmdings were also observed in RAPD analysis. Whilst these methods clearly generate more diverse patterns than ITS 1-PCR or tDNA-PCR, there appears evidence that more discriminate methods such as PFGE or ribotyping appear to correspond with strain source even less than lesser discriminate methods (Tamplin et al. , 1996; Hoi et al., 1997; Chatzidaki-Livanis et al., 2006; Gonzalez-Escalona et al., 2006a). The inability to identify isolates of single pathogenic lineages, even by less discriminate methods such as ITSl-PCR and tDNA-ILP, may suggest that acquisition of pathogenicity by V parahaemolyticus is a more recent event, occurring by lateral transfer. These findings would support current opinions on the acquisition of the ITSS2 by V parahaemolyticus (Park et al., 2004b; Ono et al., 2006; Izutsu et al., 2008) and the hypothesis of a rapidly evolving pathogen. 257 The rise and dominance of clonally related 03:K6 V. parahaemolyticus (Chiou et al. , 2000; Chowdhury. et al., 2004b; Ansaruzzaman et al., 2005 ; Rizvi et al., 2006; Serichantalergs, et al. , 2007) amongst infections reported worldwide has prompted the development of sequenced based methods (Gonzalex-Escalona et al. , 2008). Presently, it is not clear why this particular clone and its derivatives have gained prevalence, although in- vitro assessments have clearly shown enhanced virulence traits in comparative studies between 03 :K6 related and non-03 :K6 related strains and is reviewed by Nair et al. (2007). Until the molecular basis of pathogenicity of V. parahaemolyticus is elucidated, it seems unlikely that there will be answers to such questions as, 'what makes the 03:K6 related strains more dominant amongst clinical strains?' This 'pandemic' isolate is thought to have been derived from the acquisition of the gene tdh by a non-pathogenic isolate (Okura et al., 2003), and this fact, together with the regional dominance of several other serovars both in Asia and the USA, suggests that a particular clone could gain dominance at any time. Other than the increased hospitalisations due to the 03:K6 clonally related serovariants (Nair et al. , 2007), to date there appear no repo11s on the severity of the illnesses associated with V. parahaemolylicus infections. This could be attributed to the lack of reporting due to the self-limiting nature of the infection and inadequate surveillance due to the present inability to differentiate clonally related V. parahaemolyticus. In clonal populations of E. coli 0 l57:H7 it was found that a subpopulation belonging to clade 8 was responsible for contributing to more severe disease (Manning et al., 2008). This was identified based upon the analysis of 96 SNP loci and the identification of 39 SNP polymorphisms differing at 20% of the SNP loci. This present work has identified at least 34 polymorphic sites on the three structural genes of the TTSS l region and hence indicates the potential of SNP typing of clonal populations to monitor the severity of disease and better surveillance of V parahaemolyticus. Despite the large increase in infections resulting from clonally derived 258 03 :K6 strains, a recent review suggested that these clones do not seem to have the propensity for elevating hospital admissions to the same degree as pandemic isolates that were initially characterised in 1996. This led Nair et al., (2007) to speculate that that there may be some degree of decay in the epidemic process. Thus, what was essentially considered a dominant serovar amongst infections seems to be on the decrease in hospital acquired infections in Calcutta, India. This would further support the concept that any one of the clones could gain dominance at any time, and highlights the need for increased vigilance. 11.5 Conclusions Cleary, the scope of the work presented in this thesis is limited both in isolate numbers, and analytical methods employed. This study focussed on three techniques that up until now, had not been assessed in V. parahaemolyticus before: a two-primer RAPD method, ITSl-PCR and tDNA-ILP typing. The work formed the first of its kind in the UK in collaboration with the Community Reference Laboratory for monitoring bacteriological and viral contamination of bivalve molluscs (CEF AS). It is hoped that at least some of the fmdings would be explored further; possibly a blind inter-laboratory test which examines the usefulness of the RAPD markers to distinguish 03 :K6 related isolates and the reproducibility of the typing methods in the hands of multiple users. Additional suggestions for further work have been described in previous chapters. V. parahaemolyticus is genotypically a diverse organism. It appears pathogenic isolates possess diverse genotypes that stem from different lineages with evidence of recent lateral transfer of virulence determinants. Although most efforts over the last decade have centred on the epidemiology of the 03 :K6 clonally related isolates, clinical disease can be caused by a wide variety of types. The ability for V. parahaemolyticus to survive and proliferate in diverse environmental conditions and exist as various cell types is clearly a result of its 259 genome fluidity. With its remarkably fast growth properties, not only is this a model organism to study bacterial growth, but also to study the evolution of marine bacteria. The development of simple and more robust typing systems are essential for the routine monitoring of this emerging pathogen. The present study has shed some light on the genome plasticity of V parahaemolyticus yet many more questions have arisen and remain unanswered. 260 Appendix I MEDIA, REAGENTS, SOLUTIONS AND SUPPLIERS 261 1.1 Bacteriological media Tryptone soya broth (TSB) TSB powder was purchased as pre-made media from Oxoid. 30g of media was dissolved in 1L of ddH20, the pH adjusted to 7.5 and then autoclaved. The formula of the resulting media is as follows: Tryptone (casein digest USP) 17.0g Soya peptone 3.0g Sodium chloride 5.0g Dibasic potassium phosphate 2.5g Dextrose 2.5g Tryptone soy agar (TSA) TSA powder was purchased as pre-made media from Oxoid. 37g of media was dissolved in 1L of ddH20, the pH adjusted to 7.5 and then autoclaved. TSA contained the same components as stated in the TSB media although TSA with the addition of 1.5% (w/v) agar. Luria-Bertani medium (LB) Per litre of ddH20: Tryptone I Og Yeast extract 5g Sodium Chloride 1Og pH 7.5 Sterilised by autoclaving LB agar was prepared as stated above but with the addition of 1.5% (w/v) agar Marine salts broth (MSB) Marine Broth 2216 powder was purchased as pre-made media from Becton Dickinson, distributors for Difco Labs. 37.4g of media was dissolved in I L of ddH20, boiled for I nun to completely dissolve the powder and autoclaved. Approximate formula per litre (pH 7.6 ± 0.2 @ 25°C) Peptone 5.00 g Yeast Extract 1.00 g Ferric Citrate 0.10 g Sodium Chloride 19.45 g Magnesium Chloride 5.90 g Magnesium Sulphate 3.24 g Calcium Chloride 1.80 g Potassium Chloride 0.55 g Sodium Bicarbonate 0.16 g Potassium Bromide 0.08 g Strontium Chloride 34.0 mg Boric Acid 22.0 mg Sodium Silicate 4.0mg Sodium Fluoride 2.4 mg Ammonium Nitrate 1.6 mg Disodium Phosphate 8.0 mg 262 Marine salts agar (MSA) Marine Broth 2216 power (3 7 .4g) of media was dissolved in 1L of ddH20 with the addition of 1.5% w/v agar TCBS agar TCBS Cholera Mediwn (CM0333) powder was purchased as pre-made media from Oxoid. 88g of media was dissolved in 1L of ddH20. The formula of the resulting media is as follows: Yeast extract 5.00g Bacteriological peptone 10.00g Sodiwn thiosulphate 10.00g Sodiwn citrate 1O.OOg Ox Bile 8.00g Sucrose 20.00g Sodium chloride 1O .OOg Ferric citrate 1.00g Bromothymol blue 0.04g Thymol blue 0.04g Agar 14.00g pH 8.6 ± 0.2 @ 25°C Wagatsuma agar (WA)- M178 (Bacteriological Analytical Manual, 8th Edition, Revision A, 1998) Yeast extract 3g Peptone 10 g NaCl 70 g K2HP04 5g Mannitol 10 g Crystal violet 0.001 g Agar 15 g Distilled water 1 litre Human or rabbit red blood cells, fresh (24 h), with anticoagulant 50 ml Fresh rabbit blood was washed with same volwne of physiological saline three times (Centrifuge cells at about 4000 x g at 4 °C for 15 m in each time). All above ingredients, except blood, was suspended in distilled water and boiled to dissolve agar. The pH was adjusted to 8.0 and steam sterilised (30 min). A volume of 50 ml of washed red blood cells was added to the cooled (45°C) medium. 263 1.2 Genomic DNA isolation and purification Bacterial genomic DNA was extracted using the ' Puregene D-6000 DNA Isolation Kit' manufactured by Gentra Systems Inc, Minneapolis, MN5544l, USA. Details ofthe contents of each solution are as follows. Concentration of individual solution components is unknown. Cell Suspension Solution Tris [hydroxymethyl] aminomethane EDTA Sorbitol Cell Lysis Solution Tris [hydroxymethyl] an1inomethane EDTA SDS Protein Precipitation Solution Ammonium acetate DNA Hydration Solution Tris EDTA Lytic enzyme solution and RNase A stock solution were stored at 4°C. All other Puregene reagents were stored at room temperature (21 °C). 1.3 General molecular biology reagents TBE buffer TBE was made up at 5x strength according to the following formula and was then diluted to lx: TRIS-base 54.0g Boric acid 27.5g EDTA 0.5M [pH 8.0] 20ml TE buffer TE buffer was prepared either by diluting x 100 TE supplied by Sigma in sterile ' Analar' water or made up as a 1 x solution according to the following: Tris-HCL IOmM EDTA !mM pH 8 Autoclave 264 Gel loading solutions (1) General gel loading (x 6) solution (Sigma): Bromophenol blue 0.25% (w/v) Xylene cyanole FF 0.25% (w/v) Sucrose 40% (w/v) (2) Tracking buffer (x 6) for RAPD (Amersham Biosciences) 30% Glycerol. 0.125% bromophenol blue 20 mM Tris-HCl (pH 8.0) Roche Taq polymerase 10 x reaction buffer Tris-HCl 1OOmM MgCh 15mM KCl 500nlM pH 8.3 1.4 Pulsed field gel electrophoresis buffers 'CHEF Bacterial Genomic DNA Plug Kit' (BioRad) contained: Cell Suspension Buffer (CSB) 600 U/ml Proteinase K, Proteinase K Reaction Buffer 2% CleanCut Agarose I Ox Wash Buffer 25 mg/ml Lysozyme, Lysozyme Buffer Contents and concentration of CSB and Wash Buffer are unknown. Tris acetate EDTA 50 x stock: Tris base 242g Glacial acetic acid 57. L ml 0.5M EDTA lOOm! pH 8 This solution was diluted immediately to a 1 x working concentration. Tris-borate EDT A (TBE) 100 x stock: Tris base 108g Boric acid 55g 0.5M EDTA 40ml pH 8 This solution was diluted immediately to a 0.5 x working concentration. 265 1.5 DNA purification kit solutions DNA was purified using the 'High Pure PCR Product Purification Kit' supplied by Roche. Details of the content of each solution are as follows: Binding Buffer 3 M guanidine-thiocyanate, 10 mM Tris- HCl, 5% ethanol (v/v), pH 6.6 (25° C) Washing Buffer (final concentrations after addition of 40ml ethanol) 20 mMNaCl, 2 mM Tris-HCl, pH 7.5 (25°C) Elution Buffer 10 mM Tris-HCI, pH 8.5 (25°C) 1.6 Description of blotting solutions and buffers (Chapter 9) Depurination solution Per litre of ddH20: HCI 0.2M Denaturing solution Per litre of ddH20: NaCl 0.5M Tris-HCI 1.5M pH 8 Neutralisation solution Per litre of ddH20: NaOH 1.5M NaCl 0.5M 20 X SSC Per litre of ddH20: Sodium chloride 175.32g Sodium citrate 88.23g pH7 The above solution was diluted to 10 x, 5 x, 2 x, or 0.1 x strength as required with sterile ' Analar' water. 266 1.7 DNA Labelling and Detection (Chapter 9) DNA was labelled using the ' DIG DNA Labeling and Detection Kit' supplied by Roche. Details of the content of each solution are as follows: Unlabelled control DNA 1 20 111 pBR328, 100 j.tg/mJ pBR328 is digested separately with Barn HI, Bg ll and Hinfl Unlabelled control DNA 2 20 111 pBR328, 200 j.tg/ml Linearised with Barn HI Labeled Control DNA 50 111 pBR328 DNA (5 j.tg/ml) Linearised with Barn HI DNA Dilution Buffer 50 j.tg/ml herring sperm DNA in 10 mM Tris-HCl, 1 mMEDTA pH 8.0 at 20°C Hexanucleotide Mix (1 0 x concentrated hexanucleotide reaction mix) Random hexanucleotides 62.5 A26o units/ml (1.56mg/ml) Tris-HCl 500mM MgCh 100mM Dithioerythritol (DTE) l mM BSA 2mg/ml pH 7.2 dNTP labelling mixture 10 x concentrated dNTP labeling mix containing: 1 mM dATP 1 mMdGTP 1 mM dCTP 0.65 mM dTTP 0.35 mM DIG-11-dUTP pH 7.5 (20° C) Klenow enzyme, labelling grade Klenow enzyme (2 U/ j.t) Anti-Digoxigenin-AP Conjugate Polyclonal sheep anti-digoxigenin Fah-fragments, conjugated to alkaline phosphatase 750 U/ml NBTIBCIP Concentrated stock solution (exact concentration unknown) 267 Buffer 1 - Washing Buffer 0.1 M Maleic acid 0.15MNaCl pH 7.5 (20° C) 0.3% (v/v) Tween 20 Buffer 2 - Blocking reagent (content unknown) 1Ox stock and 1x working solutions were prepared with maleic acid buffer Buffer 3 - Alkaline phosphatase detection buffer M Tris-HCI, 0.1 M NaCI, pH 9.5 (20° C) Maleic acid buffer - for dilution of blocking reagent 0.1 M Maleic acid 0.15MNaCl Adjusted with NaOH (solid) to pH 7.5 (20° C) TE-buffer 10 mM Tris-HCI 1 mM EDTA pH 8.0 Stripping and reprobing of DNA blots Dimethylformamide (DMF) 0.2 N NaOH, 0.1% SOS (w/v), 2 x SSC 268 1.8 Molecular weight markers Tracklt™ 50 bp DNA Ladder Gibco (Invitrogen) 0.5 ~lane bp -2652 The 50 bp ladders range from 50-800 bp in -800 50 bp increments, plus a 2652 bp fragment. The 350 bp band is ~ 2 times brighter than the others. -500 Tracking dyes, Xylene Cyanol FF (XCFF) -350 and Orange G, allows tracking of DNA migration during electrophoresis and indicate when maximum resolution is achieved. -150 -50 Orange 100 bp Ladders 100 - 2072bp Gibco (Invitrogen) 0. 5 llgllane) The 100 bp ladders range from 100-1500 bp in 100 bp increments, plus a 2072 bp fragment. The 600 bp band is 2-3 times brighter than the others. During 2% agarose gel electrophoresis with -600 tris-acetate (pH 7.6) as the running buffer, bromophenol blue migrates near the I 00-bp fragment. -100 269 lkb Ladder 75- 12,216bp Gibco (lnvitrogen) 0. 5 flg/lane The 1bkb ladders range from 75bp to 12,216 10,160 9162 bp in various increments. 7126 6144 Hinf I fragments of the vector A: Lambda Ladder 48.5- 873 kb (BioRad) B: Saccharomyces cerevisiae Ladders 0.2 - 2.2 MB (BioRad) A B :2Jtoo kb A: Lambda ladders (concatemers of 1 .600 A.cl857sam7) range from 48.5kb - 873 kb in 48.5kb increments purchased pre-prepared. B: Saccharomyces cerevisiae size standards range from 0.225 - 2.20 MB. 270 1.9 Suppliers Becton Dickinson The Danby Building Edmund Halley Road Oxoid Limited Oxford Science Park Wade Road Oxford Basingstoke OX44DQ Hampshire http://www.bd.com RG24 8PW hnp://www.oxoid.com Bio-Rad Laboratories Ltd. Bio-Rad House Qiagen Ltd. Maxted Road Qiagen House Hemel Hempstead Fleming Way Hertfordshire HP2 7DX Crawley Tel: 0800 181134 West Sussex www.bio-rad.com RHlO 9NQ Tel: 01 293-422-9 11 Fisher Scientific UK Ltd www1.qiagen.com Bishop Meadow Road, Loughborough, Roche Molecular Biochemicals Leicestershire Bell Lane LE11 5RG Lewes http://www.fi sher.co.uk East Sussex BN71LG Gentra S~stems (now Qiagen Ltd.) www.biochem.roche.com 13355 101 Avenue North Suite 120 Sigma-Aldrich Company Ltd Minneapolis Fancy Rd MN55441 Poole USA Dorset BH1 2 4QH lnvitrogen I Gibco www.sigmaaldrich.com Invitrogen Ltd 3 Fountain Drive Thermo Fisher Scientific lnchinnan Business Park ABgene House Paisley Blenheirn Road PA4 9RF Epsom www.invitrogen.com Surrey KT19 9AP MWG Biotech www.thermofisher.com Eurofins MWG Operon 318 Worple Road Raynes Park London SW20 8QU http://www.mwg-biotech.com 271 Appendix II ADDITIONAL DATA AND IMAGES 272 ILl Bacterial DNA quantification The concentration of extracted genomic DNA was established by agarose gel electrophoresis (Section 3.8). The DNA concentration was estimated for each isolate by comparison with a range of the following known amounts of A. DNA standards indicated in Figure II.l using UV photo software (UViTec, UK) (Fig. II.l ). For confirmation and quantification of DNA of additional isolates, genomic DNA was quantified with a UV spectrophotomer (NanoDrop 100) . A 2 3 4 5 6 7 8 9 10 B 2 3 4 5 6 7 8 9 10 Figure ll.l: Quantification of DNA extracted from V. parahaemolyticus and other bacterial cultures. Panel A: lanes 1-10 correspond to the following, l; 280ng of!... standard DNA, 2; 140ng ofA. DNA standard, 3; 70ng of!... standard DNA, 4; 35ng of!... standard DNA, 5; DNA from isolate I 103/9, 6; DNA from isolate VC2, 7; DNA from isolate Vib3, 8; DNA from isolate VP6, 9; DNA from isolate Vib6, 10: DNA from isolate VC5. Panel B: lanes 1-10 correspond to tlte following, 1; 280ng of A. standard DNA, 2; 140ng ofA. DNA standard, 3; 70ng of A. standard DNA, 4; 35ng of A. standard DNA, 5; DNA from isolate 6 134, 6; DNA from isolate VP10884, 7; DNA from isolate Vib5, 8; DNA from isolate VP8, 9; DNA from isolate VCl, 10; DNA from isolate 1022/10. 273 c 2 3 4 5 6 7 8 9 10 11 D 2 3 4 5 6 7 8 9 10 11 E 2 3 4 5 6 7 8 9 10 11 Figure 11.1: Cont. Panel C: lanes 1-10 correspond to the following, 1; 280ng of/... standard DNA, 2; 140ng on DNA standard, 3; 70ng of!.. standard DNA, 4; 35ng on standard DNA, 5; DNA from isolate Vibl, 6; DNA from isolate 1023/1, 7; DNA from isolate VP361, 8; DNA from isolate Vib2, 9; DNA from isolate 11327, 10; DNA from isolate VP491, ll; DNA from isolate VP9. Panel D: lanes l -10 correspond to the following, 1; DNA from isolate VM499, 2; DNA from isolate VP33 1, 3; DNA [rom isolate VP349, 4; DNA from isolate Vib7, 5; DNA from isolate E154482, 6; DNA from isolate VA430, 7; DNA from isolate 8042, 8; DNA from isolate VibJO, 9; DNA from isolate VP346, 10; DNA from isolate E l 68143, 11 ; l40ng on standard DNA Panel E : lanes 1-lO correspond to the following, J; 280ng orA. standard DNA, 2; 140ng of A. DNA standard, 3; 70ng on standard DNA, 4; DNA from isolate PM5, 5; DNA from isolate KA5, 6; DNA from isolate LD365, 7; DNA from isolate EC5, 8; DNA from isolate Vib4, 9; DNA from isolate ST5, 10; DNA from isolate EA5. 11 ; DNA from isolate E l55855. 274 F 2 3 4 5 6 7 8 9 10 ll G 1 2 3 4 5 6 7 8 9 10 H 2 3 4 5 6 7 8 9 10 Figure ILl: Cont. Panel F: lanes l-10 correspond to the following, 1; 280ng oD. standard DNA, 2; 140ng on DNA standard, 3; 70ng on standard DNA, 4; DNA from isolate 2053, 5; DNA from isolate 2341 , 6; DNA from isolate KIDBVV, 7; DNA from isolate Vib3, 8; DNA from isolate 5421 , 9; DNA from isolate VP416, 10; DNA from isolate VP378, 11; DNA from isolate Vib5. Panel G : lanes l-10 correspond to the following, I; 280ng oO.. standard DNA, 2; 140ng on DNA standard, 3; 70ng of A. standard DNA, 4; 35ng of A. standard DNA, 5; 17 ng of A. standard DNA, 6; DNA from isolate 219, 7; DNA from isolate 220, 8; DNA from isolate 221 , 9; DNA from isolate 222, 10; DNA from isolate 223. Panel H: lanes 1-10 correspond to the following, 1; 280ng of A. standard DNA, 2; l40ng of A. DNA standard, 3; 70ng on standard DNA, 4; 35ng of A. standard DNA, 5; 17 ng on standard DNA, 6; DNA from isolate 224, 7; DNA from isolate 225, 8; DNA from isolate 226, 9; DNA from isolate 227, 10; DNA from isolate 228. Note: Plates C - F are scanned digitised images 275 Table 11.1 Genomic DNA concentrations determined from agrose gels. Isolates not in Fig. 11.1 were quantified with a UV spectrophotomer. DNA from other species were purchased lyophilised and of known concentrations. Isolate DNA cone. (ng/!ll) Isolate DNA (ng/111) 1902 20 £154482 60 8042 50 £155 855 50 11327 100 £ 168143 70 1021 /5 50 VP331 70 1022/10 280 VP345 40 1023/1 30 VP346 70 1023/3 150 VP349 70 1024/5 250 VP350 60 1103/9 30 VP361 100 2053 50 VP377 20 2436 50 VP378 100 5421 50 VP4 16 150 6134 90 VP49 1 100 6316 50 VP939 50 2341 200 VPEC 1 100 VCl 200 VPSC I 100 VC2 70 VA430 70 VC3 260 VM499 30 VC4 150 KIDBVV 250 VC5 140 LD365 350 Vibl 50 PM5 200 Vi b2 70 ST5 300 Vib3 100 EC5 200 Vib4 350 EA5 350 Vib5 50 KA5 200 Vib6 40 PAl 60 Vib7 30 Van CM104 20 Vib8 40 VPl 0884 90 Vib10 70 VV11067 100 VPl 10 219 50 VP2 40 220 80 VP3 30 22 1 150 VP4 20 222 40 VP6 140 223 50 VP7 200 224 100 VP8 40 225 100 VP9 30 226 280 VP10 30 227 100 BL21 50 228 80 Cl a 10 276 ll.2 Phenotypic growth characteristics of V. parahaemolyticus Table 11.2: Growth characteristics of V. parahaemolyticus on non-selective and selective solid media (Chapter 4) 1 Environmental/CiinicaV ---;-;;;::-:------;:;:::-:-=G:.:r..::o~w..:.lh::...:.:m:::e.:::.d::;ia~.,------Isolate Reference strain 1\JSA TSA W A TCBS 1902 NC IMB Reference Strain +++ +++ + +++ VPI 0884 NCTC Reference Strain +++ +++ ++ +++ VPI0885 Clinical +++ +++ ++ +++ 219 Clinical +++ +++ +++ +++ 220 Clinical +++ +++ ++ +++ 221 Cl ini cal +++ +++ +++ +++ 222 Clinical +++ ++ ++ +++ 223 Clinical +++ ++ ++ +++ 224 Clinical +++ +++ ++ +++ 225 Clinical +++ +++ ++ +++ 226 Clinical +++ +++ ++ +++ 227 Clinical +++ +++ +++ +++ 228 Clinical +++ +++ + +++ 1022/4 Environmental +++ ++ +++ ++ 1024/5 Environmental +++ +++ +++ +++ 102 1/5 Environmental +++ +++ +++ +++ 1022/10 Environmental +++ +++ +++ ++ 1023/1 Environmental +++ +++ +++ +++ 1023/2 Environmental +++ +++ ++ +++ 102313 Environmental +++ +++ ++ +++ 2053 Environmental +++ +++ ++ +++ 2346 Clinical +++ +++ ++ +++ 542 1 Clinical +++ +++ ++ +++ 6134 Clinical ++ + + ++ 6316 Clinical ++ ++ + + Vibl Clinical +++ +++ +++ +++ Vib2 Environmental +++ +++ ++ +++ Vib3 Environmental +++ +++ +++ +++ Vib4 Environmental +++ +++ + ++ Vib5 Environmental +++ +++ +++ +++ Vib6 Environmental +++ +++ +++ +++ Vib7 Environmental +++ +++ +++ +++ Vib8 Environmental +++ +++ +++ +++ Vib9 Environmental +++ +++ +++ +++ VibiO Environmental +++ +++ +++ +++ VPI Environmental +++ +++ ++ +++ VPI Environmental +++ +++ +++ +++ VP2 Environmental +++ +++ +++ +++ VP4 Environmental +++ ++ ++ +++ VP5 Environmental ++ +++ +++ +++ VP6 Environmental ++ ++ +++ +++ VP6 Environmental +++ +++ +++ +++ VP6 Environmental +++ +++ +++ +++ VP6 Environmental +++ +++ +++ +++ VP6 Environmental +++ +++ +++ +++ 277 Table 11.2: Cont. Environmental/Clinical/ Growth media' Jsolote Reference strain MSA TSA WA TCBS E154482 Clinical +++ +++ +++ +++ E 155855 Clinical +++ +++ +++ +++ El68143 Clinical +++ +++ +++ +++ VP331 Environmental +++ +++ +++ +++ VP345 Environmental +++ +++ +++ +++ VP346 Environmental +++ +++ +++ +++ VP349 Environmental +++ +++ +++ +++ VP350 Environmental +++ +++ + ++ VP361 Environmental +++ +++ + +++ VP377 Environmental +++ +++ +++ +++ VP378 Environmental +++ +++ ++ +++ VP416 Environmental +++ +++ +++ +++ VP491 Environmental +++ +++ +++ +++ VP939 Environmental +++ +++ +++ +++ VPECI Environmentol +++ +++ P+++ +++ VPSCJ Environmental +++ +++ P+++ +++ 1 Growth on media: MSA: Marine salts agar (Marine Agar) TSA: Tryptic soy agar (3% NaCI) WA: Wagatsuma agar TCBS: Thiosulphate citrate bile salts agar Growth was monitored at 18 hours and scored as,+++ Very good,++ Good,+ Poor, -No growth Visible exopolysaccharide production in Wagatsuma agar present only in isolates VPECI and VPSCI 278 Table 11.3: Comparison of growth rates of clinical and environmental isolates at 0.5 and 3% NaCI in TSB, pH 7.2. Doubling times were obtained by monitoring absorbance at 600nm and graphical plots of log absorbance {y) and time (x). Statistical significance testing (/-lest) was carried out on the means and variance of data with NCSS statistical software analysis. Box plots (section 4.3.2) show variability of data. Clinical 3%NaCI 0.5% NaCI Environmental 3% NaCI 0.5% NaCI 1902 (Type) 17 28 1022/10 23 32 VPI0884 (Type) 16 26 1023/2 25 28 2053 18 23 Vibl 30 38 2436 16 23 Vib2 22 29 5421 25 26 Vib3 23 33 6134 22 24 Vib7 26 28 6136 22 28 VibiO 23 34 El54482 21 29 VPI 23 31 El55855 24 27 VP2 25 36 El68143 26 34 VP3 22 36 219 21 29 VP7 21 35 220 24 28 V PlO 19 31 221 17 25 VP331 24 38 222 19 27 VP346 24 37 223 25 32 VP350 27 35 224 22 29 VP377 24 26 225 19 28 VP416 19 23 226 21 26 VP939 15 23 227 23 25 VPECI 14 24 228 24 28 VPSCI 16 25 279 ------. ' SEQUENCE DA'F A 28,0; 111.1: SEQUENCING DATA RELEVENT TO CHAPTER 9 Vibrio parahaemolyticus strain RIMD221 0633 sequence information retrieved from: http://genome.naist.jplbacteria/vpara/index.html and also available with GenBank accession numbers BA00003l and BA000032. Primer binding sites (underlined) for ITSS l and ITSS2 genes are shown below (Figs. Ill.! - 111.6) >VP1662 GTGAATCTCATGAATAAGTTGATCGATATTCTAAACAAGGTAGGGCAACGCAAAGACATC ATGCTCGCAGTGATGTTGCTGGCGATCGTCTTTATGATGATTTTGCCTTTGCCAACTGCA CTAGTGGATGTGCTGATCGGCGCAAACATGAGTATTGCCGTCGTGTTGCTGATGTTGGCG ATTTACATCACCACGCCACTGGAGTTTTCGGCGTTTCCGGCAGTATTGTTGATAACGACG CTATTTCGTTTGTCCCTGTCTATCACCACGACACGTTTGATCCTGCTTCAAGGCGATGCG GGTCAAATCGTTTACACCTTTGGTAACTTTGTTGTTGGCGGCAACCTTGTCGTTGGTATC GTGGTCTTCCTGATCATAACTATTGTACAGTTTATGGTGATCACCAAAGGTTCGGAGCGT GTTGCTGAAGTTAGTGCTCGATTCTCTCTTGATGCCATGCCGGGCAAACAGATGAGTATC GACGGTGACATGCGCGCTGGAGTTATCGATGTTCACGAGGCGCGTCATCGCCGTTCTCTG ATCGAGAAAGAGAGTCAAATGTACGGCTCGATGGACGGCGCAATGAAGTTCGTAAAAGGC GACTCCATTGCGGGCTTGGTGATCATCATCGTGAACATCTTGGGTGGCGTCACCATTGGC GTCACGCAAAAAGGCATGAGCGCGTCAGAGGCGCTTGAGCTGTTTGCGATTTTGACCGTA GGGGATGGCTTGGTATCGCAGATTCCTGCGCTATTTATCGCGATCACCGCTGGTATTATC GTTACGCGTGTTTCCCACGAAGACTCGGCCGACTTGGGCTCCGACATTGGCGGGCAAGTC ACGGCTCAGCCTCGAGCACTGTTGATTGGTGGTGTGCTGCTGGTCTTGTTCGCTCTGATC CCTGGCTTTCCGAAAATTACATTCCTAGTGCTGGCTTTGGTGGTCGGTGGCGGCGGTTTT TACCTCTTCTATCAACAGAAAAAGCAAACCGAATCAGAAAGCAGCGATCTGCCGAGCTTT GTCGCGCAGGGAGCGGGTTCTCCTGCCGCAAAACCCAATAAACCAACGCCGTCTCGTGGT AGCAAAGGCAAATTGGGCGAGCAAGAAGAGTTTGCGATGACAGTGCCACTGCTTATCGAT CTGGATTCAAGCTTACAAGAAAGCCTAGAAGCCGTCGCACTGAACGACGAACTCGCCAGA GTAAGACGCGCGTTGTATCTCGATCTTGGTGTACCGTTTCCGGGAATTCACCTGCGTTTT AACGACGGGATGAAAAACGGCGAATACTTGATTCAGTTGCAAGAAGTGCCCGTTGCACGT GGTCGTATTGAGAAAGACAAATTGCTGGTCACAGAAGGAAGCGACCAGATTGAGTTGCTT GGTGTGCCGTTTGAGCAAGATGACGACTTCCTCCCGGGAGTCTCGAGTCTTTGGGTCGCA CAAAGCTATCAAGAGAAGTTGACTGCGAGCCATGTCGGCTTCCTTACACCAGATCGTATT TTGACGTTCCATCTATCACACGTATTGAAAGAGTACGCGCAAGATTTTATTGGGATTCAA GAAACTCGCTACCTACTTGAGCAAATGGAAGGCAGTTATTCTGAATTGGTGAAAGAGGCG CAGCGCATCGTCCCGCTACAAAAAATGACGGAGATTTTGCAACGTTTGGTATCTGAGGAC ATTTCGATTCGTAACCTTCGCGTCATTTTGGAGGCCATGGTCGAGTGGGGACAAAAGGAA AAAGATGTCGTTCAGTTGACGGAATACATTCGTTCAAGCTTAAAGCGTTACATCTGTTAC AAGTACGCGAGCGGTCAGAACATGCTGCCGGCGTACCTTCTCGATCAGAGCCTCGAAGAC ACGATTCGCAGCGGTATTCGACAAACCTCGGCGGGCAGCTACTTAGCGCTTGACCCTTCG GTTACGCAGCAATTTGTTAGCGACGTAAAACAGACCGTAGGCGATTTAAGCCGTATGCCG AACAAGCCTGTTCTGGTGGTTTCTATGGATGTGCGTCGTTATGTTCGAAAGCTGATTGAA TCCGAATATTACGATTTGCCCGTTCTATCGTTTCAAGAACTCACTCAGCAGATCAACATT CAGCCACTTGGACGGGTAGGGATGTAA Figure 111.1: Complete nucleotide sequence of vcrDI. Low calcium response TTSS I structural protein. PCR product size: 458 bp. 281 >VP1680 ATGGTGAATACAACGCAAAAAATCAGTCAAAGCCCAGTACCCGATTTGGAACAGTTTCGT GCCATTGCCGCGCAAAAAGATGACCGAGTAATCAGCAAAAGAGGCGAAGTCAAAGAGCCA TCTACGTTCCATAAAGGCCACAAGTTTGCTTCGGTTAGCGAAGGCGTATTAAGAAAAAAA TACACCAAGTTCTTCCAAGAAAACATCAAAACGCATCTCGACTTGAAGCAGGCACTGCTT AAGGAAGAAAAACCAGAAACCGCACTTTTGGCTTATTCATTAGTCAGTCCGTCAGGCTAC CGAGGAGAACCATTAACCGAGCGAAAAATTTTGGAAGTGGTGAGCCTACTTGATGAAGTA AAAGTCGAAGGCGACACCTATCAGCAGCTCAAAAATACGTTTGATTCGATCAGCAAAGAT CCGCGTATGCAGGTCAGTTTGGAAAACCAGTATCCGGGGAAAATGGACGGTTTCGGCGCG CAGTTGTTGGAAATGGGAAAAGAGAAGCTGAAAGGCTCGGGTGTTAACGCTGCCATCAAT TTGGCGTTACCCGGTGTAGGCTTGCTCGTCGCCACAGGGCGTGAACTTCATAAAGCGTCC GTCAATGGCGATGCCGAAGCGTATCATCATCAACTCGAACAAATCTCACAATTGCCGGGA AGAGATCAGCGCTTATCGATGCCAATGCAACAAACGTTAGCCATTGGTCACGCGATGTTG TCGGCTGAGGGTGCAGTTGGTGCCACACTTGGTATGGCGACAGGCGGTTTGGGTACGTTT GGTGTAAGCTCTGTGGCAACGGCAGGAGTTACGCCGATAGCAAAAGAAGCCATTGGAACT GCTTTAGCTACTGGCATTATCAGCGGCGGGGGTTTTGTTGCTGGCCAAGCTGGGGCGTAT GGCTTAAATAACGAAGTGCAAGACCAACTCAAACAAGGGCCAATGTCTGGTGTGCTGCCA CGTTTGGAAATTTCAAACGTGAAGGGCGATTTCACTTTTAGCATGCAGGAGCCGGCGGCA GTTCGCGCACTGATGGCGTATCTTGGTCCTAAAGAAGATACATCAATGAGCTCACCTCAA GCGCCAAAAGAAGCGCAAGAGATGGAAGCCGCACGCTTAACACTCAAGCAAATGCTTGGC TCAAGCCCGAATGAACACCTCGTTCCAGATGTCGATTCACTGCTTAAACTGAGCGATGAA GACATGCCAAGTCAGACGGAATCTACGGCGAACGGTGCATTTAAAAAGCTGTTAAGTGAA GATTGGGATTGGCTGATGCCTGCTGTCCGAGCAATGGATAAAGGTGAAGCGAACAAAATC AACGAGAAGTTGACGTACAAGCTGCCTTTGGACGCGGCAAACGGACGTGTGTATCTCGAT AAGAGCCCGAACCTAAGTGGCGCTCAGCTTGACGCATTAGATAAACTCGGCTCACCAAGC CAGTTACGACTAATGTACTTAGCCGAAGGCTGGATTTAA Figure 111.2: Complete nucleotide sequence of vp/680. 'Cytotoxic effector' protein. PCR product size: 183 bp. 282 >VP1696 GTGGTTACCGTAATGCGAACTCTCATGCCAAAAATAGGGCGCATTGCCGCCAAAATGACA TTGTGCGCACTGTGTGTCGCTCCGATGTTTTCTGTACAAGCGACAGAGCTCAATTGGCCA GAACAACCGTTTCGTTACTATGCCGATAACGATTCGCTGAAAGATCTGCTTAACAACTTT GGTGCCAACTATCGAGTATCCGTGTCTGTGAGCGACAAAGTGAATGACCGAGTGAGTGGT CGATTCACGCCAGAAGATCCAGCGGAATTTCTCGACTACCTAGCGCAAGTTTACAACTTG ATGTGGTACTTCGATGGCGCAGTACTGCACGTTTACAAAGCGACAGAAACGCGCTCCCGA TTGCTGCAATTGGAACTGCTTACCGCGCGCGAGCTTCGCAGTACCTTGATTTCAACCGGT GTTTGGGATGCTCGTTATGGTTGGCGCGCTGCCGAAAATAAAGGGCTGGTATATCTAGCT GGGCCGCCGCGTTACGTTGAGTTGGTGGTGCAAACCGCAGAAGCGTTAGAGAGCCGATTG CTGCAAAAGTCCAACTCGACGGATGAGCTGTTTGTGGAACTCATTCCGCTTAAATATGCC TCGGCGACGGACAGAAGTATCTCGTATCGCGACCAAAGCATTACCGTTCCGGGGATTGCT TCGGTATTGAGCCGTGTGGTTGGCGGTGTTCAAACTCAAATTACGGATTCTGCGTCTGTA CAAACATCGTCTGTAAATGGATTACCCGCAGAAGCAGCGAAACCTCGTGGCAAAACTGCG TCTGTGCACGGCGGTGCAACCGTAGAAGCGGAGCCGGGGCTGAATGCGATCATCGTTCGC GATACCCAAGCAAGACTGCCGTTGTATCGCAAGCTGGTCGCTCAACTCGATCAGCCACAA TCTCGTATCGAGGTTGCGTTATCTATTGTGGACATCAGCGCAAACGATTTACGTCAGCTC GGTGTGGATTGGCGTGCAGGTGTATCGGTTGGAAACAACCGCATTGTCGACATCAAAACA ACCGGCGACGTTGATAACGGTGATGTCACATTGGGCAGTGGTCAGAGTTTCAAGTCCTTG TTGGATTCGACCAACTTGAACTACTTGTTGGCGCAAATTCGTTTATTGGAATCGAAAGGC AGTGCGCAAGTCGTTTCCAGACCAACGCTGCTCACGCAAGAAAACGTCGAAGCGGTGCTA AACAACAGCAGCACGTTTTACGTCAAATTGGTTGGCAAAGAGACGGCGGCTTTGGAAGAA GTGACGTATGGCACCTTGTTGCGTATTGTGCCTCGTATCGTCGGCGACAGATTCGCGACG CGACCAGAGATCAACTTGAGCCTTCACTTAGAAGATGGCGCAAAGATCCCAGATGGTGGT GTGGATGATCTTCCATCAGTACGTAAAACCGAAATCAGTACGTTAGCGACGGTAAAACAA GGGCAAAGCTTACTCATTGGTGGCGTATACCGAGACGAAGTTAGCCATCAGCTACGTAAA GTCCCGCTGTTGGGCGACATTCCGTACCTAGGTGCTTTGTTCCGTAGTAATACCAACACA ACGCGCCGAACAGTTCGCATGTTCATTATCGAGCCGAGAATCGTGGTCGATGGGATTGGC GACAGCGTGTTGATAGGTAATGAGCATGATTTGCGCCCGAGTATTGGCCAGCTCAACAAC ATTTCGAACAACAGCGCAGAATTTAAATCGGTTGTTGAAGTATTCAGCTGTACGTCAAAA ACGCAGGCAGAACGTTACCAACAAGACTTACTGTCACAGCAAAAGTCTTCGCTGCTAACT CAATGTCAATTGCCGTCAGGGCAAGTAGGCTGGCGAGTCAAAGTGGCGGAATGTGACCTT TCACAGGCCGAATGTGTTCGCCCTTCAGAGGAGCCATAA Figure 111.3: Complete nucleotide sequence of vscCJ. TTSS I structural protein. PCR product size: 415 bp. 283 >VPA1321 ATGCCAATATTAAATATTAGTAAATTTAGTAATACAGAGTATGCCTTTGTTAATAATCGT AAATTAAAAGAGCCATGTATAAAGGTGAAATCTGTTAGAACTAGTAGAGAAGGAGGCGAA ATATTATTCATTAATATGCCTAGTAAGTCAAAATATAAAGACTTGAGAGCTATGGTTAAA AGCAGCGTGGTGGTTAGTGAATCCAACCAAACCGCTGCTTCTAAATTTGAAAATAATAGT AGGTTTAATCGAAATGATATCAATGTAAAAAAAGCAGATGCAAAATCAATAACCGCGCTG AAAAGTGGTGATTTGCACATACTTAAAGGCAAGGGAATTATAGGTATGAAGGGGGGAGAT AATAAGCTCCCGTTCAAATGTACTATTGTAAATGATGATAAAAATGGTGCGCATCTTAGT CAAGGGACTAATTTAGCAACCAATGGTATTAAGAGTATGGCTGGGGATTCAGTTCGTGCC GCTCAACTAATCCCTGGTACGCCTCTTGGACAGTTTTATAATAGCGCGCCTCTAGATGAT AGCTTTAATGTGGTACATTTACCTAATGGTCAAAGAGGTGTAAATGGACTTAAAATACCA CTGTCTGAGTTTTATTCTGAAAAAAAATTTTTATTCTCAAATGGTGCGCTGTCTGGTTGT ATGACATGCACGGCAATAGATAAAAATAACCTTTATATCTTTCATGTAGGTAAAGATGGA AATGACACATCACCTTGGAAGACAAATGTGGATGGTTCTAGTTTAATACAAAAAAACATG AAAATGTTACTTGGCCAAAATTCTGATTCGTTGAATAATGGTATTCAGGGTTTGATAGAC TATTGTTCTAAAAACTTTGATAAAGCTATTATTCAATATTGTGGTCACGGAGAACAATAT AGTGGTCGAAAAAACATACATTTATTTGATTATAATACTCCTCAAAAAAACAATCCTTTG CGTGTAGGAAATAATTTGACATTGATAAGCCATTCTGATAATGGCTCATTGAGTATTTCA ACATTGTGTGATGATATGATTATTAATTCAAAAACCTGCGAAACCAACTCTGTTAATAGC AAATTAGTGCTATTGAAAAATGGTTAA Figure 111.4: Complete nucleotide sequence ofvpa/321. Cytotoxic necrotizing factor. PCR product size: 485 bp. >VPA1339 ATGAGGTTTAGTCAACTTTTGCTTGCGCTATGCTGTCTGACCTTTGGCGCGCAAGCCGAT GTTTTTATCAAACAAGAAGATATGGATTTGAAGCAGGCGTTGGAAGCCATTGCGAAAGAT ATGCAGCTCAAACTGGTGTCTGACATAGATAGCAAAACTAGTAGCCAACCGATAACTCAA ACATTATCTGGGTCGGGGCCTATGTTATTGTCAAAGTTATCTTCTGTTTTTGATTTCGAC TGGTATACCTTTGGTGGCACGCTGACTATCCAGTCAGGTCAGCAATACGCTAACTATACC TATAAGCCTAAAAACATTTCTCCTAGTCAATTACTATCGGAGATGAAGGGGGCGTTTAAA ACTAATAGTACAACCAAAATTAAGTTGGTTGAGCGTGGACACTCGCTGTTGTTCTCAGGT ACGCGCAAGTTTGTTAACGATGCTATGGGGTATGCCTCTATGGTTGATAGAAACCAGTTT TTAGAAAAAGGCAACAACCTTGAAGTGGCTCGTATCGAGTTTAACTATGTTTCGGTTATG GATCGAAATATTGCGACTTACGATGGTCAGGTGAACTTTCCTGGCGCTCAATCATTAATT GCAGCAGCCATTACGAACATTGGGCAATTCACTAATCTCAATGATGAAGAGCTAACATCA CGCGCTTACAGAGTTAAGCTTAGCCAAGGTGAAAAGCAAGCACTCGAAGAGGATGAAAAA TCCACGAAGGTTCAGGCGTTGCCGGGAACAAATGCGCTGTTGATTCGCGGAACTCAAGAA GAAGTAAAATTAGCTAAACGTATTGCTGCTCTAATTGATGTAAAACGTCGTCAGTTACTG TTTTCGCTTAAAGTGTATGATATTTCGGCTGAGCGAAAAGAAAATCTTGGTATTGATAGT GGGTGGCTAAATGGTAACCGAGGTATTTACGATATTGTTGTGCCACCTTTTACCGAAACA AAGGATTTTGTGAAAAACTTTCAAGCACTTTATTCTAATGGTATTGCTCGTAGTGTCTAT GAGACCAACCTAGTGGCATTGGAAAATCAACAAAGTCAGTTTGGTAAAAAGCAAACTGCA ACATTGACGTTGATCTCGGACAAGCAAGTTAAAACAGAAAAGATTGAAGCCGATAATAGT CTGTATGTGACAGGTCGTTTATTGCCATCAGGTCAAGTGCAAGCTCGTATTCAGTATGTT GAGGAGTCTCTAGATGAAGATGATAATGAGGATAGCAATAGACCGCCGCGAGTCAGTTCT CAGTCACTCAACTCGGAGGTTTATATTCAACCAAATCAGACTGTAATTTTAGGCGGTTTT GATAATACTGTCACGCAACAACAAGAGAGTGGTGTACCAATTATATCGAGTATTCCTTTG CTTGGTGAGTTGTTTAAATATACTTCCGAGAGCAAGCATAAATACAAACGTTACGTATCG ATTTCATTTCAGGTGATTGAATAA Figure 111.5: Complete nucleotide sequence of vscC2. Putative TTSS2 structural protein. PCR product size: 272 bp. 284 >VPA1355 ATGATGGCGAAGTTACGTAAATCATCATTATTGTTTAATCCCAGCAGTGTGATGATTGCG TTTTTATTACCGACAATCATTCTAGTGACGCTGCCGGGGATTGTGATCGATTTCTTTTTG ATGATCAGTTTTATCAGTTCTGCACTGTTATTGGTCATTATGTTAGAAAATGATGAGCCG TTGAAAGTGACGTTTCTACCTACAATGGTTCTCTTACTAACAACGTTTCGTTTGTTGCTT TCTATTGCGACAACGCGGAATATCATTGCTAATGAAGATGTTGGTCAAGTTATTGAAACG GTTGGTCAGTTTGTTATGGGGGGGAACTTACTCTCTGGGCTGCTGATTTTCATCATCATT ACTATTGTTCAATTTTTAGTAGTAACCAAAGGTGGTGAACGCGTTGCTGAAGTTGGTGCT CGCTTCTCTCTGGATGCGTTGCCTGGTAAACAAATGACCATTGATGGGGATTTAAAAAGC GGCTTAATCACCGGGGAGCAAGCTCAAAAATTACGTGCTGATTTAGGGACAGAAAACAGA CTTTTTGGTTCTCTTGATGGGGCAATGAAGTTTGTTAAAGGTGATGCGATCGCGGGTATT TTGATCAGTTTAGTTAACTTATTCGGCGGTATTTATGTTGGTATTAACCAGTTTGACCTC TCTTTAGGAGATAGCGTTAGCCGCTTTTCAGTCTTGACAGTAGGGGATGGGTTAGTATCT CAAATACCAAGTTTATTGCTATCTATGGCATGTGGTGTCTATTTGACACGTATTAAAGGT TCTGATGATGAGTCTAGCTCATTTATGTCTCAACTAATGTTGCAAATTCGAACCTTTTGG AAAAGTTTGTTTGTCATTGGTGGCATCATTATTGTCTTGGGTATTGCTAATCCGAATTTG TTGTATGTGTGCTTGCCTCTCGCAGGGGTTTGTGCCGCGTTAGCGATTTGGCTACAAAAA AAAGACGGCAGTAAAAATCCAACGCTCAGTTTAGAACCGCAAGGACATGGCTTATTTCAG TCTTTAAAGTTTGTTTTCCGAGAAGAGCAATATCAGCACTTGATTCAACGACATGTGGAA GAGAAAGAAAAGTCGTTATGGGGACAAGCCCTTGGACTACCTGCAGTTGTCGTAGACCCA ACGCTTGATTGCGATATCAAGGTGTACATTTCCGATATCGAAATGTACAGCTATAAAAAT ACTGAACCAGGTGTAGTGATTGAAATTCTTAATGATGAGCTTTTGATGTCAAGTCATGAT GTGGTGAGTAATATTAATCATATTATAAGCTACTTTATTCATAATCGTTTTGTTGAAAAA TATAACCTGCAATACAGCAGCAATATTGTCACAGGTTTAGAACAACAAAGTGAAATTATG AAAAGCGAGATTGATTCTGCTGTCGGATTAAACCGAGTGCATGATGTACTGAAGCAGATG ATTCGCTCACCTGAGTTTTATTTGGATCGTGTTAGTTTTTTCGAGGCTCTTATCTATTGG TCACGCGTAGAGAGTGACCCTCGTAACTGGTTAACACGCATGCGTAGTCAAGCCCGTTAT GACATTACAAGCAAATTGTTAAATAGCGAAGGTAAGTTTAATGTCATATTACTGACTCCC GAACTTACAGAAGAAATTTCCGGATTTGTTGCTGGTGAGTTTGACGATGCTGAGAGATTG ATTTCGGTACAAACTACACTGCGTAGCGAAATAAAACGTTTGCTTTTTCAATACGGTTCG AAACCTGTCGTCGTGGTCAATGAAAATGAACTGGATTATATAAAATCATTTTTCCAACAA GTTATGAGTAGAATATTTGTATGTAGTCATGCTGATATTGTTAACTCAAACTGTATTGAA AATACAGATACAATTCATATTTAA Figure 111.6: Complete nucleotide sequence of vcrD:Z. Putative TTSS2 structural protein. PCR product size: 388 bp. 285 iT------~o------"2tr-~ 30 4o so so 10 oo 90 100 110 120 130 :•------+------+-+------·------·------·------·------·------+------·------·------+------·------1 RDfD2210633 : cGRRCRRRTCTCRCRRTTGCC~RGRGRTCRGCGCTTRTCGRTGCCARTGCRRCRRRCGTTAGOCRTTGGTCRCGCGRTGTTGTCGGCTGRGGGTGCR&TTGGTGCCRCRCTTGGTRTGGCGRCRGG C RQ3810B 1 RCRRTTGCCSGGftRGRGRTCRGCGCTTRTCGRTGCCRRTGCRRCRRRCGTTRGCCRTTGGTCRCGC&RTGTTGTCGGCTGRGGGTGCRGTTGGTGCCRCRCTTGGTRTGGCGRCRGGC 2208 : RGTRRGTCR---TTGCCGGGRRGRG- TCRGCGCTTRTCGRTGCCRRTGCRACRRRCGTTRGCCRTTGGTCRCGCGRTGTTGTCGGCTGRGGGTGCRGTTGGTGCCRCRCTTGGTRTGGCGRCRGGC 19028 : GRCTGTCGCRGTTGCl:GGGRRGRG- TCRGCGCTTRTCGRTGCCRRTGCARCRRACGTTRGCCRTTGGTCRCGCGRTGTTGTCGGCTGRGGGCGCRGTTGGTGCCRCACTTGGCRTGGCGACRGGT VP4168 : TGGGRCTRCTCGRRTGGCC~CGRR-RCAGCGCTT-TCGATGCCA-TGCRHCRRRCGTTRGCCRTTGGTCRCGCGRTGCTGTCCGCTGAGGGCGCA&TTGGTGCCRCACTTGGCATGGCGRCAGGT Consensus : ••••• ea . t . tcacaaTtGCC~aGRg . tCRGCGCTTa TCGATGCCAaTGDRACRRRCGTTRGCCRTTGGTCHCGCGRTGtTGTCeGCTGAGGGtGCRGTTGGTGOCACRCTTGGtATGGCGRCAGGc I I -~----l~o------f5b-- ~s~-~s~- RD1D2210633 GGTTTGGGTACGTTTG• ------+------•------~-·---1GTGTRR6CTCt.6TGGCRA RQ38108 6GTTTGGGTRCGTTTGGTGTRRGCTC',GT6GC 2208 GGTTT666TRC6TTTGGTGTRRGCTCT6TGR 19028 GGTTTGGGTROGTTTGGTGTRRGCTCToTGR VP4168 GGTTTGGGTRCGTTTG- TGTRRGCTCf,GTGR Consensus GGTTTGGGTRCGTTTGgTGTRRGCTCl.GTGa ••• I------ Figure m 7: Nucleotide sequences and multiple alignments of the vpl680 locus of TTSSl in isolates of V. parahaenwlyticus. RIMD2210633 is a sequenced clinical strain. Sequences AQ3810B and 220B are from clinical strains and VP416B from an environmental isolate. 1902B is NCIMB reference strain from which probe was developed. Consensus regions are shown in red; blue and black show substitutions or deletions. Isolate VP416 shows Regions marked in dotted line represent quality control clipped regions. SNPs in TISSI are also present in vcrDJ and vscCJ . 286 :--1------io______io------~o 4o so so 1o so 9o 1oo uo 120 13o RinD22106a: GTTGGC1------·------·------.+------·---GGTGTTCRRRCTCRRRTTRC66ATTCTGCGTCTGTRCRRRCRTCG------+------·------·------·------+----TCT6TRRRTGGRTTRCCCGCRGRR6CRGCGRARCCTCGTGGCRRRRCTGCGTCTGTGCACG6CGGTGCRRCCGTRGRR6- ·------+------·-----l 220 6RGTTTCRGTTTRGTTCRTTRC66ijTTCTGCGTCTGTRCRRRCRTCGTCT6TRRRT66ATTRCCCGCRGRRGCAGCGRRRCCTC6TGGCRRRRCTGCGTCTGTGCACG6CGGTGCRRCCGTRGRR6 VP41G RRGGRRCTCRTGRTGTCT TGTRG~TTCTGOGTCTGTRCRRRCRTCGCCTGTRRRCGGRTTRCCCGCAGRRGCRGCRRRRCCTCGTGGCARRRCTGCGTCTGTGCACGGCGGTGCRRCCGTRGARG 1902~ GRGTTRCGGTCTRTTCRGTCGGATTCTGCGTCTGTRCRRRCRTCGCCTGTRRRCGGRTTRCCCGCAGRRGCRGCRRRRCCTCGCGGCRRRRCTGCGTCTGTGCACGGCGGTGCRRCCGTRGRRG RQ381Q RCGGATTCTGOGTCTGTRCRRRCRTCGCCTGTARRCGGRTTRCCCGCRGRRGCRGCGRRRCCTCGCGGCRRRRCTGCGTCTGTGCRCGGCGGTGCRRCCGTRGRRG Consensu~ •••••• g ••••• c •••• t . t •• t . acGGijTTCTGCGTCTGTRCRRRCRTCGcCTGTRRRcGGRTTRCCCGCR6RRGCRGCgHRRCCTCGt6GCRRRRCTGCGTCTGT6CRCGGCGGTGCRRCCGTRGRRG '------~ ~ ~ - -~~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ RIHD2210633 CGGRGCCGGGGCTGRRTGCGRTCRTCGTTCGCGATRCCCRRGCRRGRCTGCCGTTl------·------·------·------·------·------·------·------·-GTRTCGCRRGCTGGTCGCTCRRCTCGRTCRGCCRCRRTCTCGTRTCGRGGTTGCGTTRTCTRTTGTGGRCRTCRG------·------·------+------·------1 220C CGGRGCCGGGGCTGRRTGCGRTCRTCGTTCGCGRTRCCCRRGCRRGRCTGCCGTTGTRTCGCRAGCTGGTCGCTCRRCTCGRTCRGCCRCRRTCTCGTRTCGRGGTTGCGTTRTCTRTTGTGGACRTCRG VP416C CGGR6CCGGG6CT6RRTGCGRTCRTCGTTCGCGRTRCCCAR6CRRGRCTGCCGTTRTRTCGCAAGCTGGTCGCTCRRCTCGRTCRGCCRCRRTCTCGTRTCGR66TTGCGTTRTCTRTTGT66RCRTCRG 1902C CGGRGCCGGGGCTGRRTGC6RTCRTCGTTCGCGRTRCCCRR6CRRGRCTGCCGTTRTRTC6CRRGCT66TCGCTCRRCTCGRTCRGCCRCRRTCTC6TRTCGRG6TTGCGTTRTCTRTTGT66RCRTCR6 RQ3810 CGGAGCCGGGGCTGARTGCGRTCRTCGTTCGCGATRCCCRRGCRRGRCTGCCGTTGTRTCGCRAGCTGGTCGCTCRRCTCGRTCRGCCRCRRTCTCGTRTCGRGGTTGCGTTRTCTRTTGTGGRCRTCAG Consensus CGGRGCCGGGGCTGAATGCGRTCATCGTTCGCGATRCCCRRGCRAGACTGCCGTTgTRTCGCARGCTGGTCGCTCARCTCGRTCRGCCACARTCTCGTRTCGAGGTTGCGTTRTCTRTTGTGGRCRTCRG 261 270 280 290 300 310 320 330 340 350 360 370 i--3'1J6'------'39{1 ~ RIHD2210633 CGCRRACGRTTTRCGTCRGCTCGGTGTGGRTTGGCGl------·------·------·------T-GC--·R-GGTGTRTCGGTTGGRRRCRRCCGCATT------·------+------GTCGACATCRRRRCRRCCGGCGRCGTTGRTRRCGG+------+------+------+---TGRTGTCRCATT------+----~GGG----·------1 :I 220C CGCRRRCGRTTTRCGTCAGCTCGGTGTGGRTTGGCGTGCRGGTGTRTCGGTTGGRRRCRRCCGCRTTGTCGRCRTCRRRRCRRCCGGCGRCGTTGRTRRCGGTGRTGTCRCRTT~CRGTGGTTCRGRG : VP416C CGCRRRCGRTTTRCGTCAGCTCGGTGTGGRTTGGCGTGCGGGTGTRTCGGTTGGRRACRRCCGCRTTGTCGRCRTCRRRRCRRCCGGCGRCGTTGRTRRCGGTGRTGTCACRTTt GGCRTG : 1902C CGCRRRCGRTTTRCGTCRGCTCGGTGTGGRTTGGCGCGCGGGTGTRTCGGTTGGRRRCRRCCGCHTTGTCGRCRTCRRRRCRRCCGGCGRCGTTGRTRRCGGTGRTGTCRCRTTGGGCRGGG--TC 1 RQ3810 CGCRRRCGRTTTRCGTCRGCTCGGTGTGGRTTGGCGTGCGGGTGTRTCGGTTGGRRRCRRCCGCRTTRTCGRCRTCRRRRORRCCGGCGRCGTTGRTRRCGGCGRTGTCRCRTTGGGCR6TGG- TCRGR6 : Consensus CGCRRRCGRTTTRCGTCRGCTCGGTGTGGRTTGGOG~GCeGGTGTRTCGGTTGGRRRCRRCCGCRTTgTCGRCRTCRRRRCRRCCGGCGRCGTTGRTRRC6Gt6RTGTCRCRTT~GGcae . g •• tc •••• : ~ ------~ Figure m.s: Nucleotide sequences and multiple alignments of the vscCJlocus of TTSSl in isolates of V. parahaemolyticus. R.IMD2210633 is a sequenced clinical strain. Sequences AQ3810C and 220C are from clinical strains and VP416C from an environmental isolate. 1902C is NCIMB reference strain from which probe was developed. Consensus regions are shown in red; blue and black show substitutions or deletions. Regions marked in dotted line represent quality control clipped regions. SNPs in TISSI are also present in vcrDJ and vpl680. 287 ~------, :1 10 20 30 : 40 50 60 70 80 90 100 110 120 130 1 RD102210633 :c1 G------·------·-----CGCRRGTTTGTTRHCGRTGCTRTGG66TA----·-~------·------·------·------·------·------·------·------+------·------16CCTCTRTGGTTGRTRGRRACCRGTTTTTRGRRRRRGGCHRCRRCCTTGRRGTGGCTCGTRTCGRGTTTRRCTRTGTTTCGGTTRTGGRTCGRRRTRTl 220E : GRGTCGCGCTGT--RCGRTGCTRTGGGGTR, GCCTCTRTGGTT6RTR6RRACCRGTTTTTR6RRRRRGGCHRCRRCCTTGRR6TGGCTCGTRTC6RGTTTRRCTRT6TTTCGGTTRTGGRTC6RRRTRT VP416E : TTGTTCRRTGTTTCCTRTGCTC-6GGGTRl GCCTCTRTGGTTGRTRGRRACCRGTTTTTRGRRRRRGGCRRCRRCCTTGRR6TGGCTCGTRTCGRGTTTRRCTRTGTTTC6GTTRTGGRTCGRRRTRT Consensus :. g. g• •• c •• TGT t . aCgRTGCTat6GGGTA~6CCTCTRTGGTTGRTRGHRRCCR6TTTTTHGRRRHR6GCRRCRACCTTGRR6T66CTCGTRTCGRGTTTRRC!~!~~~~~~~!~lCGRRRTRT .. ------1 Ir .,I 131 140 150 160 170 180 190 200 210 220 230 : 240 249 RD102210633 TGCGRCTTRCGRTGGTCRGGTGRRCTTTCCTGGCGCTCARTCATTRRTTGCRGCRGCCRTTRCGRACRTTGGGCRRTTCRCTRHTCTCARTGRTGRRGRGCTRl------·------·------·------·------·------·------·------·------·------·---·-----+------OCRTCRCGC -1 220E TGCGRCTTRC6RTGGTCRGGTGRRCTTTCCTG6CGCTCHRTCRTTRRTTGCRGCR6CCRTTRCGRRCRTT6GGCRRTTCRCTRRTCTCRRTGRTGRRGRGCTR~CRTCRCGC6CTTRCR VP416E TGCGRCTTACGRTGGTCRGGTGRRCTTTCCTGGCGCTCRRTCRTTRRTTGCRGCRGCCRTTRCGRRCRTTGGGCRRTTCRCTRRTCTCARTGRTGRRGRGCTR~CRCRCCG TGCGRCTTACGRTGGTCRGGTGRRCTTTCCTGGCGCTCRRTCRTTRRTTGCRGCRGCCHTTRCGRHCRTT6GGCRRTTCRCTRRTCTCRRTGRT6RRGRGCTRACAtcaC6c...... 1 Consensus I_------_I Figure ID.9: Nucleotide sequences and multiple alignments of the vscC2 locus of TTSS2 in isolates of V. p arahaemolyticus. RIMD221 063 3 is a sequenced clinical strain. Nucleotide sequence 220E is from a clinical strain and VP416B from an environmental isolate. Consensus regions are shown in red; blue and black show substitutions or deletions. Regions marked in dotted line represent quality control clipped regions. Regions within clipped areas were homologous in all three TISS2 loci. 288 r------j :1 10 20 : 30 40 50 60 70 80 90 100 110 120 130 RDt02210633 :l:rR------·------·------·-·------+------+------TTRRRGGTTCTGRTGRTGRGTCTRG[ TCRTTTRTGTCTCRRCTRRTGTTGCRRRTTCGRRCCTTTTGGRRRR6TTT&TTTGTCRTTG6T66CRTCRTTRTTGTCTT666TRTT6CTRATCCGRRTTT6-+------+------+------+------·------+------+------1 2208 : CTGGCRG--TCTGRTGRTGRG-CTRGt TCRTTTRTGTCTCRRCTRRTGTTGCRRRTTCGRRCCIIJJGGARRR6TTTGTTTGTCRTTGGTGGCRTCRTTRTTGTCTTGGGTRTTGCTRRTCCGRRTTTG VP4168 • RRRGTCGCT TGGTG6RGCTRG(TCRTTTRTGTCTCRRCTRRTGTTGCRRRTTCGRROCII11GGRRRRGTTTGTTTGTCRTTGGT66CRTCRTTRTTGTCTTGGGTRTTGCTRRTCCGRRTTT6 Consensus : •• ~ . aaRG •• ~CTGaT6aT6ai . CTRG~TCRTTTRTGTCTCRRCTRRTGTT6CRRRTTCGRRCCTTTTG6RRRR6TTTGTTTGTCRTTGGTGGCRTCRTTRTTGTCTT666TRTTGCTRRTCCGRRTTT6 L------J 131 140 150 160 170 180 190 200 210 220 230 240 250 260 RDt02210633 TTGTRTGTGTGCTTGCCTCTCGCRGGGGTTTGTGCCGCGTTRGCGRTTTGGCTRCRRRRRRRRGRC66CRGTRRRRRTCCARC6CTCRGTTTRGRRCCGCRRGGRCRT66CTTRTTTCRGTCTTTRRRGTl------+------·------·------·------·------+------·------·------·------·------+------·------1 2208 TTGTATGTGTGCTTGCCTCTCGCRGGGGTTTGTGCCGOGTTRGCGATTTGGCTRCRRRRRRRRGRCGGCRGTRRARRTCCARCGCTCRGTTTRGRRCCGCRRGGRCATGGCTTATTTCRGTCTTTRRRGT VP4168 TTGTRTGTGTGCTTGCCTCTCGCRGGGGTTTGTGOCGCGTTRGOGRTTTGGCTRCRRRRRRRRGRC66CRGTRRARRTCCARC6CTCAGTTTRGRRCC6CRRGGRCATG6CTTRTTTCRGTCTTTRARGT Consensus TTGTRTGTGTGCTTGCCTCTCGCRGGGGTTTGT6CCGOGTTRGCGRTTTGGCTRCRRRRRRRRGRC6GCRGTRRRRRTCCARC6CTCRGTTTRGARCCGCRRGGRCRT66CTTRTTTCRGTCTTTRRRGT.. ------261 270 280 290 300 310 320 330 340 350 :' 360 365 RDt02210633 TTGTTTTCCGRGARGRGCRRTRTCRGCRCTTGRTTCRRCGRCRTGTGGRRGRGRRRGRARRGTCGTTRTGGGGRCRAGCCCTTGGRCTRC1------·------·------·------+------+------+------·------+------•-t------·--UTGCRGTTGTCG · 2208 TTGTTTTCCGAGRRGRGCRATATCRGCACTTGRTTCRACGACRTGTGGARGRGRRRGRRRRGTCGTTRTGGGGRCAAGCCCTTGGRCTRCGTGCRGTTGTCGTRR VP4168 TTGTTTTCCGRGRRGRGCRRTRTCRGCRCTTGRTTCRRCGRCRTGTGGRRGRGRRRGRRRRGTCGTTRTGGGGRCRRGCCCTTGGRCTRCCTG Consensus TTGTTTTCCGRGRRGRGCRRTRTCRGCRCTTGRTTCRRCGRCRTGTGGRRGRGRRRGRRRRGTCGTTRTGGGGRCRRGCCCTT6GRCTR~TGcai~tg~c2 ••• Figure 01.10: Nucleotide sequences and multiple alignments of the vcrD2 locus of TTSS2 in isolates of V. parahaenwlyticus. RIMD2210633 is a sequenced clinical strain. Nucleotide sequence 220E is from a clinical strain and VP416B from an environmental isolate. Consensus regions are shown in red; blue and black show substitutions or deletions. Regions marked in dotted line represent quality control clipped regions. Regions within clipped areas were homologous in all three TISS2 loci. 289 List of References Abed, Y., C. Bollet, and P. De Micco. 1995. Demonstration of Mycobacterium kansasii species heterogeneity by the amplification of the 16S-23S spacer region. J Med Microbiol. 43:156-158. Acinas, S.G., L. A. Marcelino, V. Klepac-Ceraj and M. F. Polz. 2004. Divergence and redundancy of 16S rRNA sequences in genomes with multiple rm operons. J Bacteriol. 186:2629-2635. Alam, M. J., S. Miyoshi, and S. Shinoda. 2003. Studies on pathogenic Vibrio parahaemolyticus during a warm weather season in the Seto Inland Sea, Japan. Environ Microbiol. 5:706-710. Alam, M. J., K. I. Tomochika, S. I. Miyoshi, and S. Shinoda. 2002. Environmental investigation of potentially pathogenic Vibrio parahaemolyticus in the Seto-Inland Sea, Japan. FEMS Microbiol Lett. 208:83-87. Aldova, E. 1989. Serovars of Vibrio parahaemolyticus. J Hyg Epidemiol Microbiol Immunol. 33:219-228. Alexander, S. M., T. H. Grayson, E. M. Chambers, L. F. Cooper, G. A. Barker, and M. L. Gilpin. 2001. Variation in the spacer regions separating tRNA genes in Renibacterium salmoninarum distinguishes recent clinical isolates from the same location. J Clin Microbiol. 39:119-128. Alsina, M. and A. R. Blanch. 1994. A set of keys for biochemical identification of environmental Vibrio species. J Appl Bacteriol. 76:79-85. Altschul, S. F., W. Gish, W. Miller, E. W. Myers and D. J. Lipman. 1990. Basic local alignment search tool. J Mol Bioi. 215:403-410. Anonymous. 2006. Travel Health Infection Report. Communicable Disease Report Weekly. 16:40 http://www.hpa.org.uk/CDR/pages/travel.htrn (accessed online 09/01/08). Ansaruzzaman, M., M. Lucas, J. L. Deen, N. A. Bhuiyan, X. Y. Wang, A. Safa, M. Sultana, A. Chowdhury, G. B. Nair, D. A. Sack, L. von Seidlein, M. K. Puri, M. Ali, C. L. Chaignat, J. D. Clemens, and A. Barreto. 2005. Pandemic serovars (03:K6 and 04:K68) of Vibrio parahaemolyticus associated with diarrhea in Mozambique: spread of the pandemic into the African continent. J Clin Microbiol. 43:2559-2562. Arakawa, E., T. Murase, T. Shimada, T. Okitsu, S. Yamai, and H. Watanabe. 1999. Emergence and prevalence of a novel Vibrio parahaemolyticus 03:K6 clone in Japan. Jpn J Infect Dis. 52:246-247. Atienzar, F. A., B. Cordi, M. E. Donkin, A. J. Evenden, A. N. Jha and M. H. Depledge. 2000. Comparison of ultraviolet-induced genotoxicity detected by random amplified polymorphic DNA with chlorophyll fluorescence and growth in a marine macroalgae, Palmaria palmata. Aquat Toxicol. SO: 1-12. 290 Atienzar, F. A. and A. N. Jha. 2006. The random amplified polymorphic DNA (RAPD) assay and related techniques applied to genotoxicity and carcinogenesis studies: a critical review. Mutat Res. 613:76-102. Atsumi, T., L. McCarter, and Y. Imae. 1992. Polar and lateral flagellar motors of marine vibrio are driven by different ion-motive forces. Nature 355:182-184. Ayres, P.A. and G. I. Barrow. 1978. The distribution of Vibrio parahaemolyticus in British coastal waters: report of a collaborative study 1975--6. J Hyg (Lond) 80:281-294. Baba, K., H. Shirai, A. Terai, K. Kumagai, Y. Takeda, and M. Nisbibuchi. 1991a. Similarity of the tdh gene-bearing plasmids of Vibrio cholerae non-01 and Vibrio parahaemolyticus. Microb Pathog. 10:61-70. Baba, K., H. Shirai, A. Terai, Y. Takeda, and M. Nishibuchi. 1991 b. Analysis of the tdh gene cloned from a tdh gene- and trh gene-positive strain of Vibrio parahaemolyticus. Microbiol Immunol. 35:253-258. Baele, M., P. Baele, M. Vaneecboutte, V. Storms, P. Butaye, L. A. Devriese, G. Verschraegen, M. Gillis and F. Haesebrouck. 2000. Application oftRNA intergenic spacer PCR for identification of Enterococcus species. J Clin Microbiol. 38:4201-4207. Baele, M., V. Storms, F. Haesebrouck, L. A. Devriese, M. Gillis, G. Verschraegen, T. de Baere and M. Vaneechoutte. 2001. Application and evaluation of the interlaboratory reproducibility oftRNA intergenic length polymorphism analysis (tDNA-PCR) for identification of Streptococcus species. J Cl in Microbiol. 39: 1436-1442. Baele, M., M. Vaneecbouttc, R.Verhelst, M. Vancanneyt, L.A. Devriese and F. Haesebrouck F. 2002. Identification of Lactobacillus species using tDNA-PCR. J Microbiol Methods. 50:263-271. Baffone, W., B. Citterio, E. Vittoria, A. Casaroli, R. Campana, L. Falzano, and G. Donelli. 2003. Retention of virulence in viable but non-culturable halophilic Vibrio spp. Int J Food Microbiol. 89:31-39. Baffone, W., R. Tarsi, L. Pane, R. Campana, B. Repetto, G. L. Mariottini, and C. Pruzzo. 2006. Detection of free-living and plankton-bound vibrios in coastal waters of the Adriatic Sea (Italy) and study of their pathogenicity-associated properties. Environ Micro bioi. 8:1299-1305. Bag, P. K., S. Nandi, R. K. Bhadra, T. Ramamurthy, S. K. Bhattacharya, M. Nishibuchi, T. Hamabata, S. Yamasaki, Y. Takeda, and G. B. Nair. 1999. Clonal diversity among recently emerged strains of Vibrio parahaemolyticus 03:K6 associated with pandemic spread. J Clin Microbiol. 37:2354-2357. Barker, W. H., Jr. 1974. Vibrio parahaemolyticus outbreaks in the United States. Lancet 1:551-554. Baross, J. and J. Liston. 1970. Occurrence of Vibrio parahaemolyticus and related hemolytic vibrios in marine environments of Washington State. Appl Microbiol. 20: 179- 186. Bart-Delabesse, E., J. Sarfati, J.P. Debeaupuis, W. van Leeuwen, A. van Belkum, S. Bretagne and J.P Latge. 200 I. Comparison of restriction fragment length polymorphism, 291 microsatellite length polymorphism, and random amplification of polymorphic DNA analyses for fingerprinting Aspergillusfumigatus isolates. J Clin Microbiol. 39:2683-2686. Bates, T. C. and J. D. Oliver. 2004. The viable but nonculturable state ofKanagawa positive and negative strains of Vibrio parahaemolyticus. J Microbiol. 42:74-79. Bauer, A., 0. Ostensvik, M. Florvag, 0. Ormen, and L. M. Rorvik. 2006. Occurrence of Vibrio parahaemolyticus, V cholerae, and V vulnificus in Norwegian Blue Mussels (Mytilus edulis). Appl Environ Microbiol. 72:3058-3061. Baumann, P. and R. H. W. Schubert. 1984. Family Jl Vibrionaceae Veron 1965, 5245AL. In: Krieg NR (Ed) Bergey's Manual of Systematic Bacteriology. 1:516-517. The Williams & Wilkins Co., Baltimore. Begue, R. E., R. Meza, G. Castellares, C. Cabezas, B. Vasquez, A. Ballardo, J. Cam, and J. L. Sanchez. 1995. Outbreak of diarrhea due to Vibrio parahaemolyticus among military personnel in Lima, Peru. Cl in Infect Dis. 21:1513-1514. Bej, A. K., D. P. Patterson, C. W. Brasher, M. C. Vickery, D. D. Jones, and C. A. Kaysner. 1999. Detection of total and hemolysin-producing Vibrio parahaemolyticus in shellfish using multiplex PCR amplification of tl, tdh and trh. J Microbiol Methods. 36:215-225. Beuchat, L.R.\973. Interacting effects of pH, temperature, and salt concentration on growth and survival of Vibrio parahaemolyticus. Appl Microbiol. 25:844-846. Beuchat, L. R. 1977. Suitability of some enrichment broths and diluents for enumerating cold- and heat-stressed Vibrio parahaemolyticus. Can J Microbiol. 23:630-633. Bhuiyan, N. A., M. Ansaruzzaman, M. Kamruzzaman, K. Alam, N. R. Chowdhury, M. Nishibuchi, S. M. Faruque, D. A. Sack, Y. Takeda, and G. B. Nair. 2002. Prevalence of the pandemic genotype of Vibrio parahaemolyticus in Dhaka, Bangladesh, and significance of its distribution across different serotypes. J Cl in Microbiol. 40:284-286. Binta, M. G. and P. N. Nyaga. 1982. The distribution of Vibrio parahaemolyticus serotypes in Kenyan seafish, shellfish, marine water and sediment. Trans R Soc Trop Med Hyg. 76:497-499. Binta, M. G., P. N. Nyaga, and T. B. Tjaberg. 1982. Vibrio parahaemolyticus: occurrence in marine seafish, shellfish, water and sediment in Kenya. East Afr Med J. 59:442-445. Bisharat, N., C. Amaro, B. Fouz, A. Llorens, and D. I. Cohen. 2007a. Serological and molecular characteristics of Vibrio vulnificus biotype 3: evidence for high clonality. Microbiology .153:84 7-856. · Bisharat, N., D. I. Cohen, M. C. Maiden, D. W. Crook, T. Peto and R. M. Harding. 2007b. The evolution of genetic structure in the marine pathogen, Vibrio vulnificus. Infect Genet Evol. 7:685-693 Blackstone, G. M., J. L. Nordstrom, M. C. Vickery, M. D. Bowen, R. F. Meyer, and A. Depaola. 2003. Detection of pathogenic Vibrio parahaemolyticus in oyster enrichments by real time PCR. J Microbiol Methods 53:149-155. 292 Bockemuhl, J., A. Amedome and A. Triemer. 1975. Vibrio parahaemolyticus gastroenteritis during, the El Tor cholera epidemic in Togo (West Africa). Am J Trap Med Hyg. 24:101-104. Bockmuhl, J. and A. Triemer. 1975. Serotypes of Vibrio parahaemolyticus from clinical and environmental sources in Togo (West Africa). Jpn J Med Sci Bioi. 28:215-221. Boddicker, J. D., N. A. Ledeboer, J. Jagnow, B. D. Jones and S. Clegg. 2002. Differential binding to and biofilm formation on, HEp-2 cells by Salmonella enterica serovar Typhimurium is dependent upon allelic variation in the fimH gene of the fim gene cluster. Mol Microbiol. 45:1255-1265. Boy er, S. L., V. R. Flechtner and J. R, Johansen. 2001. Is the 16S-23S rRNA internal transcribed spacer region a good tool for use in molecular systematics and population genetics? A case study in cyanobacteria. Mol Bioi Evol. 18:1057-1069. Brosius, J., M. L. Palmer, P. J. Kennedy, and H. F. Noller. 1978. Complete nucleotide sequence of a 16S ribosomal RNA gene from Escherichia coli. Proc Natl Acad Sci USA. 75: 4801-4805. Bubb, H. D. 1975. Vibrio parahaemolyticus- a marine pathogen detected in South African coastal waters. S Afr Med J. 49: 1514-1 516. Burr, M. D., K. L. Josephson, and I. L. Pepper. 1998. An evaluation of ERIC PCR and AP PCR fingerprinting for discriminating Salmonella serotypes. Lett Appl Microbiol. 27:24-30. Burstyn, D. G., L. A. McNicol, and M. J. V oil. 1980. Isolation and characterization of spontaneously arising auxotrophic and Kanagawa phenomenon-negative mutants of Vibrio parahaemolyticus. Infect lmmun. 27:889-896. Cabanillas-Beltran, H., E. LLausas-Magana, R. Romero, A. Espinoza, A. Garcia Gasca, M. Nishibucbi, M. Ishibashi, and B. Gomez-Gil. 2006. Outbreak of gastroenteritis caused by the pandemic Vibrio parahaemolyticus 03:K6 in Mexico. FEMS Microbiol Lett. 265:76-80. Cabassi, E. and L. Mori. 1976. Vibrio parahaemolyticus: aetiological agent of food poisoning. Folia Vet Lat. 6:335-354. Cabello, F. C., R. T. Espejo, M. C. Hernandez, M. L. Rioseco, J. Ulloa, and J. A. Vergara. 2007. Vibrio parahaemolyticus 03:K6 epidemic diarrhea, Chile, 2005. Emerg Infect Dis. 13:655-656. Cabrera-Garcia, M. E., C. Vazquez-Salinas and E. I. Quifiones-Ramirez. 2004. Serologic and molecular characterization of Vibrio parahaemolyticus strains isolated from seawater and fish products of the Gulf of Mexico. Appl Environ Microbiol. 70:6401-6406. Cai, T., L. Jiang, C. Yang, and K. Huang. 2006. Application of real-time PCR for quantitative detection of Vibrio parahaemolyticus from seafood in eastern China. FEMS Imrnunol Med Microbiol. 46:180-186. Call, D. R., L. Orfe, M. A. Davis, S. Lafrentz and M. S Kang. 2008. Impact of compounding error on strategies for subtyping pathogenic bacteria. Foodborne Pathog Dis. 5:505-516. 293 Candelon, B., K. Guilloux, S. D. Ehrlich, A. Sorokin. 2004. Two distinct types ofrRNA operons in the Bacillus cereus group. Microbiology. 150:601-11. Cantor, C. R., C. L. Smith and M. K. Mathew. 1988. Pulsed-field gel electrophoresis of very large DNA molecules. Annu Rev Biophys Chem. 17:287-304. Carruthers, M. M. 1975. Cytotoxicity of Vibrio parahaemolyticus in HeLa cell culture. J Infect Dis. 132:555-560. Carruthers, M. M. and W. J. Kabat. 1976. Isolation of Vibrio parahaemolyticus from fecal specimens on mannitol salt agar. J Cl in Microbiol. 4:175-179. Carruthers, M. M. 1977. In vitro adherence of Kanagawa-positive Vibrio parahaemolyticus to epithelial cells. J Infect Dis. 136:588-592. Carver, T. J., K. M. Rutherford, M. Berriman, M. A. Rajandream, B. G. Barrell and J. Parkhill. 2005. Artemic Comparison Tool. Bioinformatics. 21:3422-3423. Casellas, J. M., M. A. Caria, and M. E. Gerghi. 1977. Isolation of Vibrio parahemolyticus from mussel varieties in Argentina. Rev Asoc Argent Microbiol. 9:41-53. Cesaris, L., B. E. Gillespie, V. Srinivasan, R. A. Almeida, A. Zecconi, and S. P. Oliver. 2007. Discriminating between strains of Escherichia coli using pulsed-field gel electrophoresis and BOX-PCR. Foodborne Pathog Dis. 4:473-480. Chang, A., W. C. Liew, A. Chuah, Z. Lim, Q. Lin and L. Orban. 2007. FluoMEP: a new genotyping method combining the advantages of randomly amplified polymorphic DNA and amplified fragment length polymorphism. Electrophoresis. 28:525-534 Chang, B., H. Taniguchi, H. Miyamoto, and S. Yoshida. 1998. Filamentous bacteriophages of Vibrio parahaemolyticus as a possible clue to genetic transmission. J Bacteriol. 180:5094-510 I. Chang, B., S. Yoshida, H. Miyamoto, M. Ogawa, K. Horikawa, K. Ogata, M. Nishibuchi, and H. Taniguchi. 2000. A unique and common restriction fragment pattern of the nucleotide sequences homologous to the genome ofvf33, a filamentous bacteriophage, in pandemic strains of Vibrio parahaemolyticus 03:K6 04:K68, and 01 :K untypeable. FEMS Microbiol Lett. 192:231-236. Charlton, B. R., A. A. Bickford, R. L. Walker and R. Yamamoto. 1999. Complementary randomly amplified polymorphic DNA (RAPD) analysis patterns and primer sets to differentiate Mycoplasma gallisepticum strains. J Vet Diagn 1nvest.11: 158- 161. Chatterjee, B. D., P. K. De, and T. Sen. 1977. Sucrose teepol tellurite agar: a new selective indicator medium for isolation of Vibrio species. J Infect Dis. 135:654-658. Chatterjee, B. D., P. K. De, A. K. Hati, and N. Tandon. 1978. Fly borne Vibrio parahaemolyticus in Calcutta. Trop Geogr Med. 30:499-503. Chatzidaki-Livanis, M., M. A. Hubbard, K. Gordon, V. J. Harwood and A.C. Wright. 2006. Genetic distinctions among clinical and environmental strains of Vibrio vulnificus. Appl Environ Microbiol. 72:6136-6141. 294 Chiou, C. S., S. Y. Hsu, S. I. Chiu, T. K. Wang, and C. S. Chao. 2000. Vibrio parahaemolyticus serovar 03:K6 as cause of unusually high incidence of food-borne disease outbreaks in Taiwan from 1996 to 1999. J C1in Microbiol. 38:4621-4625. Chiu, T. H., J. Duan, and Y. C. Su. 2007. Characteristics of virulent Vibrio parahaemolyticus isolated from Oregon and Washington. J Food Prot. 70:1011-1016. Chowdhury, A., M. Ishibashi, V. D. Thiem, D. T. Tuyet, T.V. Tung, B. T. Chien, L. L. Seidlein, D. G. Canh, J. Clemens, D. D. Trach, and M. Nishibuchi. 2004a. Emergence and serovar transition of Vibrio parahaemolyticus pandemic strains isolated during a diarrhea outbreak in Vietnam between 1997 and 1999. Microbiol Immunol. 48:319-327. Chowdhury, N. R., 0. C. Stine, J. G. Morris, and G. B. Nair. 2004b. Assessment of evolution of pandemic Vibrio parahaemolyticus by multilocus sequence typing. J Clin Microbiol. 42: 1280-1282. Chowdhury, N. R., S. Chakraborty, B. Eampokalap, W. Chaicumpa, M. Chongsa Nguan, P. Moolasart, R. Mitra, T. Ramamurthy, S. K. Bhattacharya, M. Nishibuchi, Y. Takeda, and G. B. Nair. 2000a. Clonal dissemination of Vibrio parahaemolyticus displaying similar DNA fingerprint but belonging to two different serovars (03:K6 and 04:K68) in Thailand and India. Epidemiol Infect. 125:17-25. Chowdhury, N. R., S. Chakraborty, T. Ramamurthy, M. Nishibuchi, S. Yamasaki, Y. Takeda, and G. B. Nair. 2000b. Molecular evidence of clonal Vibrio parahaemolyticus pandemic strains. Emerg Infect Dis. 6:631-636. Chun, D., J. K. Chung, S. Y. Scol, and R. Tak. 1974. Vibrio parahaemolyticus in the Republic of Korea. Am J Trop Med Hyg. 23:1125-1130. Chun, J., A. Huq and R. R. Colwell. 1999. Analysis of 16S-23S rRNA intergenic spacer regions of Vibrio cholerae and Vibrio mimicus. Appl Environ Microbiol. 65:2202-2208. Chun, J., I. N. Rivera and R. R. Colwcll. 2002. Analysis of 16S-23S rRNA intergenic spacer of Vibrio cho/erae and Vibrio mimicus for detection of these species. Methods Mol Bioi. 179:171-178. Clementino, M. M., I. De Fillipis, C. R. Nascimento, R. Branquinho, C. L. Rocha, and 0. B. Martins. 2001. PCR analyses oftRNA intergenic spacer, 16S-23S internal transcribed spacer, and randomly amplified polymorphic DNA reveal inter- and intraspecific relationships of Enterobacter c/oacae strains. J Clin Microbiol. 39:3865-3870. Coelho, A., A. C. Vicente, M. A. Baptista, H. Momen, F. A. Santos, and C. A. Salles. 1995. The distinction of pathogenic Vibrio cholerae groups using arbitrarily primed PCR fingerprints. Res Microbiol. 146:671-683. Cohen, A. L., J. D. Oliver, A. DePaola, E. J. Fcil and E. F. Boy d. 2007. Emergence of a virulent clade of Vibrio vulnificus and correlation with the presence of a 33-kilobase genomic island. Appl Environ Microbiol. 73:5553-5565. Comeau, A. M., A. M. Chan, and C. A. Suttle. 2006. Genetic richness of vibriophages isolated in a coastal environment. Environ Micro bioi. 8:1164-1176. 295 Cook, D. W., J. C. Bowers, and A. Depaola. 2002. Density of total and pathogenic (tdh+) Vibrio parahaemolyticus in Atlantic and Gulf coast molluscan shellfish at harvest. J Food Prot. 65:1873-1880. Cordova, J. L., J. Astorga, W. Silva, and C. Riquelme. 2002. Characterization by PCR of Vibrio parahaemolyticus isolates collected during the 1997-1998 Chilean outbreak. Bioi Res. 35:433-440. Comelis and Van Gijesegem. 2000. Assembly and function of type Ill secretion systems. Ann Review Microbiology 54: 735-774 Corpet, F. 1988. Multiple sequence alignment with hierarchical clustering. Nucl Acids Res.16:10881-10890 Coutard, F., M. Pommepuy, S. Loaec, D. J. Hervio-Heath. 2005. mRNA detection by reverse transcription-PeR for monitoring viability and potential virulence in a pathogenic strain of Vibrio parahaemolyticus in viable but nonculturable state. Appl Microbial. 98:951-961. Coutard, F., S. Lozach, M. Pommepuy and D. Hervio-Heath. 2007. Real-time reverse transcription-PeR for transcriptional expression analysis of virulence and housekeeping genes in viable but noncu1turable Vibrio parahaemolyticus after recovery of culturability. Appl Environ Microbial. 73:5183-5189. Croci, L., E. Suffredini, L. Cozzi, and L. Toti. 2002. Effects of depuration of molluscs experimentally contaminated with Escherichia coli, Vibrio cholerae 01 and Vibrio parahaemolyticus. J Appl Microbial. 92:460-465. Croci, L., E. Suffredini, L. Cozzi, L. Toti, D. Ottaviani, C. Pruzzo, P. Serratore, R. Fischetti, E. Goffredo, G. Loffredo, and R. Mioni. 2007. Comparison of different biochemical and molecular methods for the identification of Vibrio parahaemolyticus. J Appl Microbial. 102:229-237. Currie, B. J., D. Gal, M. Mayo, L. Ward, D. Godoy, B. G. Spratt and J. J. LiPuma. 2007. Using BOX-PCR to exclude a clonal outbreak of melioidosis. BMC Infect Dis. 7:68. Daffonchio, D., N. Raddadi, M. Merabishvili, A. Cherif, L. Carmagnola, L. Brusetti, A. Rizzi, N. Chanishvili, P. Visca, R. Sharp and S. Borin. 2006. Strategy for identification of Bacillus cereus and Bacillus thuringiensis strains closely related to Bacillus anthracis. Appl Environ Microbial. 72: 1295-30 I. Dai, J. H., Y. S. Lee, and H. C. Won g. 1992. Effects of iron limitation on production of a siderophore, outer membrane proteins, and hemolysin and on hydrophobicity, cell adherence, and lethality for mice of Vibrio parahaemolyticus. Infect Immun. 60:2952- 2956. Daniels, N. A., L. MacKinnon, R. Bishop, S. Altekruse, B. Ray, R. M. Hammond, S. Thompson, S. Wilson, N. H. Bean, P. M. Griffin, and L. Slutsker. 2000. Vibrio parahaemolyticus infections in the United States, I 973-1998. J Infect Dis. 181:1661-1666. Danin-Poieg, Y ., L. A. Cohen, H. Gancz, Y. Y. Broza, H. Goldshmidt, E. Malul, L. Valinsky, L. Lerner, M. Broza and Y. Kashi. 2007. Vibrio cho/erae strain typing and phylogeny study based on simple sequence repeats. J Clin Microbial. 45:736-746. 296 Davis, C. R., L. C. Helier, K. K. Peak, D. L. Wingfield, C. L. Goldstein-Hart, D. W. Bodager, A. C. Cannons, P. T. Amuso, and J. Cattanii. 2004. Real-time PCR detection of the thermostable direct hemolysin and thermolabile hemolysin genes in a Vibrio parahaemolyticus cultured from mussels and mussel homogenate associated with a foodbome outbreak. J Food Prot. 67:1005-1008. De, S. P., M. Banerjee, B. C. Deb, P. G. Sengupta, J. Sil, B. K. Sircar, D. Sen, A. Ghosh, and S. C. Pal. 1977. Distribution ofvibrios in Calcutta environment with particular reference to V. parahaemolyticus. Indian J Med Res. 65:21-28. De Bruijn, F. J. 1992. Use of repetitive (repetitive extragenic palindromic and enterobacterial repetitive intergeneric consensus) sequences and the polymerase chain reaction to fingerprint the genomes of Rhizobium meliloti isolates and other soil bacteria. Appl Environ Microbiol. 58:2180-2187. De Gheldre, Y., N. Maes, F. L. Presti, J. Etienne and M. Struelens. 2001. Rapid identification of clinically relevant Legionella spp. by analysis of transfer DNA intergenic spacer length polymorphism. J Clin Microbiol. 39: 162-169 De Sousa, 0. V., R. H. Vieira, F. G. de Menezes, C. M. dos Reis, and E. Hofer. 2004. Detection of Vibrio parahaemolyticus and Vibrio cholerae in oyster, Crassostrea rhizophorae, collected from a natural nursery in the Coco river estuary, Fortaleza, Ceara, Brazil. Rev Inst Med Trop Sao Paulo 46:59-62. Deepanjali, A., H. S. Kumar, I. Karunasagar, and I. Karunasagar. 2005. Seasonal variation in abundance of total and pathogenic Vibrio parahaemolyticus bacteria in oysters along the southwest coast of India. Appl Environ Microbiol. 71:3575-3580. Dennis, P. P., S. Ziesche, and S. Mylvaganam. 1998. Transcription analysis of two disparate rRNA operons in the halophilic archaeon Haloarcula marismortui. J Bacteriol. 180:4804-4813. DePaola, A., L. H. Hopkins and R. M. McPhearson. 1988. Evaluation of four methods for enumeration of Vibrio parahaemolyticus. Appl Environ Microbiol. 54:617-618 DePaola, A., C. A. Kaysner, J. Bowers, D. W. Cook. 2000. Environmental investigations of Vibrio parahaemolyticus in oysters after outbreaks in Washington, Texas, and New York (1997 and 1998). Appl Environ Microbiol. 66:4649-4654. DePaola, A., J. L. Nordstrom, J. C. Bowers, J. G. Wells, and D. W. Cook. 2003a. Seasonal abundance of total and pathogenic Vibrio parahaemolyticus in Alabama oysters. Appl Environ Microbiol. 69: 1521-1526. DePaola, A., J. Ulaszek, C. A. Kaysner, B. J. Tenge, J. L. Nordstrom, J. Wells, N. Puhr, and S. M. Gendel. 2003b. Molecular, serological, and virulence characteristics of Vibrio parahaemolyticus isolated from environmental, food, and clinical sources in North America and Asia. Appl Environ Microbiol. 69:3999-4005. Deplano, A., R. De Mendon~a, R. De Ryck, and M. J. Struelens. 2006. External Quality Assessment of Molecular Typing of Staphylococcus aureus Isolates by a Network of Laboratories. J Clin Microbiol. 44: 3236-3244. 297 Di Pinto, A., G. Ciccarese, G. Tantillo, D. Catalano, and V. T. Forte. 2005. A collagenase-targeted multiplex PCR assay for identification of Vibrio alginolyticus, Vibrio cholerae, and Vibrio parahaemolyticus. J Food Prot. 68: 150-153. Dil-Afroze., A. Misra, I. M. Sulaiman, S. Sinha, C. Sarkar, A. K. Mahapatra and S. E. Hasnain. 1998. Genetic alterations in brain tumors identified by RAPD analysis. Gene. 206:45-8 Dileep, V., H. S. Kumar, Y. Kumar, M. Nishibuchi, I. Karunasagar, and I. Karunasagar. 2003. Application of polymerase chain reaction for detection of Vibrio parahaemolyticus associated with tropical seafoods and coastal environment. Lett Appl Microbiol. 36:423-427. Dillon, B., J. Iredell, E. B. Breitschwerdt, and R. G. Maggi. 2005. Potential limitations of the 16S-23S rRNA intergenic region for molecular detection of Bartonella species. J Clin Microbiol. 43:4921-4922. Dingman, D. W. 2004. Characterization of Paenibacillus popilliae rRNA operons. Can J Microbiol. 50:779-791. Dobrindt, U., B. Hochhut, U. Hentschel and J. Hacker. 2004. Genomic islands in pathogenic and environmental microorganisms.Nat Rev Microbiol. 5:414-424. Douet, J. P., M. Castroviejo, A. Dodin, and C. Be bear. 1996. Study of the haemolytic process and receptors of thermostable direct haemolysin from Vibrio parahaemolyticus. Res Microbiol. 147:687-696. Duarte, R. S., R. R. Barros, R. R. Facklam and L. M. Teixeira.2005. Phenotypic and genotypic characteristics of Streptococcus porcinus isolated from human sources. J Cl in Microbiol. 43:4592-4601. Dziejman, M., D. Serruto, V. C. Tam, D. Sturtevant, P. Diraphat, S. M. Faruque, M. H. Rahman, J. F. Heidelberg, J. Decker, L. Li, K. T. Montgomery, G. Grills, R. Kucherlapati, and J. J. Mekalanos. 2005. Genomic characterization of non-0 I, non- O 139 Vibrio cholerae reveals genes for a type Ill secretion system. Proc Natl Acad Sci USA. 102:3465-3470. Ehrenstein, B., A. T. Bernards, L. Dijkshoorn, P. Gerner-Smidt, K. J. Towner, P. J. Bouvet, F. D. Daschner, and H. Grundmann. 1996. Acinetobacter species identification by using tRNA spacer fingerprinting. J Clin Microbiol. 34:2414-2420. Eko, F. 0. 1992. Urease production in Vibrio parahaemolyticus: a potential marker for virulence. Eur J Epidemiol. 8:627-628. Elliot, E. L., C. A. Kaysner, L. Jackson, and M. L. Tamplin. 1995. Vibrio cholerae, V parahaemolyticus, V vulnificus and other Vibrio spp, p. 9.01-9.27. In FDA Bacteriological Analytical Manual, 8th ed. AOAC International, Gaithersburg, MD. Engels, W. R. 1993. Contributing software to the internet: the Amplify program. Trends Biochem Sci. 18:448-450. Falla, T. J., D. W. Crook, L. N. Brophy, D. Maskell, J. S. Kroll and E. R. Moxon. 1994. PCR for capsular typing of Haemophilus influenzae. J Clin Microbiol. 32:2382- 2386. 298 Farfan, M., D. Miiiana, M. C. Fuste and J. G. Loren. 2000. Genetic relationships between clinical and environmental Vibrio cholerae isolates based on multilocus enzyme electrophoresis. Microbiology. 146:2613-2626. Ferrer, C., F. Colom, S. Frases, E. Mulet, J. L. Abad, and J. L. Alio. 2001. Detection and identification of fungal pathogens by PCR and by ITS2 and 5.8S ribosomal DNA typing in ocular infections. J Clin Microbiol. 39:2873-2879. Fonstein, M. and R. Haselkorn. 1995. Physical mapping of bacterial genomes. J Bacteriol. 177:3361-3369. Foultier, B., P. Troisfontaines, S. Miiller, F. R. Opperdoes and G. R. Cornelis. 2002. Characterization of the ysa pathogenicity locus in the chromosome of Yersinia enterocolitica and phylogeny analysis of type Ill secretion systems. J Mol Evol. 55:37-51. Franca, S. M., D. L. Gibbs, P. Samuels, and W. D. Johnson, Jr. 1980. Vibrio parahaemolyticus in Brazilian coastal waters. JAMA 244:587-588. Fuenzalida, L., C. Hernandez, J. Toro, M. L. Rioseco, J. Romero and R. T. Espejo. 2006. Vibrio parahaemolyticus in shellfish and clinical samples during two large epidemics of diarrhoea in southern Chile. Environ Microbiol. 4:675-683. Fujino, K. 1966. Discovery of Vibrio parahaemolyticus. Nippon Saikingaku Zasshi 21:373-388. Fujino, T., T. Miwatani, Y. Takeda, and A. Tomaru. 1969. A thermolabile direct hemolysin of Vibrio parahaemolyticus. Biken J. 12: 145-148. Funahashi, T., T. Tanabe, H. Aso, H. Nakao, Y. Fujii, K. Okamoto, S. Narimatsu, and S. Yamamoto. 2003. An iron-regulated gene required for utilization ofaerobactin as an exogenous siderophore in Vibrio parahaemo/yticus. Microbiology 149: 1217-1225. Gazumyan, A., J. J. Schwartz, D. Liveris, and I. Schwartz. 1994. Sequence analysis of the ribosomal RNA operon of the Lyme disease spirochete, Borrelia burgdorferi. Gene 146:57-65. Gendel, S. M., J. Ulaszek, M. Nishibuchi, and A. Depaola. 200 I. Automated ribotyping differentiates Vibrio parahaemolyticus 03:K6 strains associated with a Texas outbreak from other clinical strains. J Food Prot. 64:1617-1620. Georghiou, P. R., R. J. Hamill, C. E. Wright, J. Versalovic, T. Koeuth, D. A. Watson, and J. R. Lupski. 1995. Molecular epidemiology of infections due to Enterobacter aerogenes: identification of hospital outbreak-associated strains by molecular techniques. Clin Infect Dis. 20:84-94. Ghosh, M. M., B. K. Sarkar, B. C. Deb, S. P. De, D. Sen, S. C. Pal, and B. N. Ghosh. 1986. Role of common house flies in transmission of Vibrio parahaemolyticus infection in Calcutta slum in relation to carrier rate of the said pathogen. Indian J Public Health 30:145- 149. Ghosh, P. 2004. Process of protein transport by the type III secretion system. Microbiol Mol Bioi Rev. 68:771-795. Giske, C.G., B. Libisch, C. Colinon, E. Scoulica, L. Pagani, M. Fiizi,G. Kronvall G, and G.M. Rossolini. 2006. Establishing clonal relationships between VIM-1-Iike metallo- 299 beta-lactamase-producing Pseudomonas aeruginosa strains from four European countries by multilocus sequence typing. J Clin Microbial. 44:4309-4315 Gjerde, J. and B. Boe. 1981. Isolation and characterization of Vibrio alginolyticus and Vibrio parahaemolyticus from the Norwegian coastal environment. Acta Vet Scand. 22:331-343. Goldberg, T. L., T. R. Gillespie, R. S. Singer. 2006. Optimization of analytical parameters for inferring relationships among Escherichia coli isolates from repetitive element PCR by maximizing correspondence with multi locus sequence typing data. Appl Environ Microbial. 72:6049-6052. Gonzalez, N., J. Romero, and R. T. Espejo. 2003. Comprehensive detection ofbacteria1 populations by PCR amplification of the l6S-23S rRNA spacer region. J Microbial Methods. 55:91-97. Gonzalez-Esca1ona, N., G. M. Blackstone, and A. Depaola. 2006b. Characterization of a Vibrio alginolyticus strain, isolated from Alaskan oysters, carrying a hemolysin gene similar to the thermostable direct hemolysin-related hemolysin gene (trh) of Vibrio parahaemo/yticus. Appl Environ Microbial. 72:7925-7929. Gonzalez-Escalona, N., V. Cachicas, C. Acevedo, M. L. Rioseco, J. A. Vergara, F. Cabello, J. Romero, and R. T. Espejo. 2005a. Vibrio parahaemolyticus diarrhea, Chile, 1998 and 2004. Emerg Infect Dis. 11:129-131. Gonzalez-Escalona, N., J. Romero, and R. T. Espejo. 2005b. Polymorphism and gene conversion of the 16S rRNA genes in the multiple rRNA operons of Vibrio parahaemolyticus. FEMS Microbial Lett. 246:213-219. Gonzalez-Escalona, N., J. Romero, C. A. Guzman, and R. T. Espejo. 2006a. Variation in the 16S-23S rRNA intergenic spacer regions in Vibrio parahaemolyticus strains are due to indels nearby their tRNAGlu. FEMS Microbial Lett. 256:38-43. Gonzalez-Escalona, N., J. Martinez-Urtaza, J. Romero, R. T. Espejo, L. A. Jaykus and A. DePaola. 2008. Determination of molecular phylogenetics of Vibrio parahaemolyticus strains by multilocus sequence typing J Bacterial. 190:2831-2840 Gonzalez, S. F., M. J. Krug, M. E. Nielsen, Y. Santos, and D. R. Call. 2004. Simultaneous detection of marine fish pathogens by using multiplex PCR and a DNA microarray. J Clin Microbial. 42:1414-1419. Gooch, J. A., A. Depaola, C. A. Kaysner, and D. L. Mars hall. 200 I. Evaluation of two direct plating methods using nonradioactive probes for enumeration of Vibrio parahaemolyticus in oysters. Appl Environ Microbial. 67:721-724. Goshima, K., T. Honda, Y. Takeda, and T. Miwatani. 1976. Stopping of spontaneous beating of cultured mouse and rat myocardial cells by a toxin (thermostable direct hemolysin) from Vibrio parahaemolyticus. Recent Adv Stud Cardiac Struct Metab 11:615- 620. Grayson, T. H., F. A. Atienzar, S. M. Alexander, L. F. Cooper, and M. L. Gilpin. 2000. Molecular diversity of Renibacterium salmoninarum isolates determined by randomly amplified polymorphic DNA analysis. Appl Environ Microbial. 66:435-438. 300 Grayson, T. H., L. F. Cooper, F. A. Atienzar, M. R. Knowles, and M. L. Gilpin. 1999. Molecular differentiation of Renibacterium salmoninarum isolates from worldwide locations. Appl Environ Microbiol. 65:961-968. Green, C. J. and B. S. Void. 1993. Staphylococcus aureus has clustered tRNA genes. J Bacteriol. 175:5091-5096. Gurtler, V. and B. C. Mayall. 2001. Genomic approaches to typing, taxonomy and evolution of bacterial isolates. lnt J Syst Evol Microbiol. 51:3-16. Gurtler, V. and V. A. Stanisich. 1996. New approaches to typing and identification of bacteria using the 16S-23S rDNA spacer region. Microbiology. 142:3-16. Gutacker, M., N. Conza, C. Benagli, A. Pedroli, M. V. Bernasconi, L. Permin, R. Aznar and J. C. Piffaretti. 2003. Population genetics of Vibrio vulnificus: identification of two divisions and a distinct eel-pathogenic clone. Appl Environ Microbiol. 69:3203- 3212. Hadrys, H., B. Scbierwater, S. L. Dellaporta, R. DeSalle and L. W. Buss. 1993. Determination of paternity in dragonflies by Random Amplified Polymorphic DNA fingerprinting. Mol Ecol. 2:79-87 Hagbigbi, L. and N. S. Waleh. 1978. Isolation of Vibrio parahaemolyticus from Persian Gulf. J Trop Med Hyg. 81:255-257. Hall, L. M., R. A. Whiley, B. Duke, R. C. George, and A. Efstratiou. 1996. Genetic relatedness within and between serotypes of Streptococcus pneumoniae from the United Kingdom: analysis of multi locus enzyme electrophoresis, pulsed-field gel electrophoresis, and antimicrobial resistance patterns. J Clin Microbiol. 34:853-859. Hara-Kudo, Y., T. Nishina, H. Nakagawa, H. Konuma, J. Hasegawa, and S. Kumagai. 2001. Improved method for detection of Vibrio parahaemolyticus in seafood. Appl Environ Microbiol. 67:5819-5823. Hara-Kudo, Y., K. Sugiyama, M. Nishibuchi, A. Chowdhury, J. Yatsuyanagi, Y. Obtomo, A. Saito, H. Nagano, T. Nishina, H. Nakagawa, H. Konuma, M. Miyahara and S. Kumagai. 2003. Prevalence of pandemic thermostable direct hemolysin-producing Vibrio parahaemolyticus 03:K6 in seafood and the coastal environment in Japan. Appl Environ Microbiol. 69:3883-3891. Hardies, S. C., A. M. Comeau, P. Serwer, and C. A. Suttle. 2003. The complete sequence of marine bacteriophage VpV262 infecting Vibrio parahaemolyticus indicates that an ancestral component of a T7 viral supergroup is widespread in the marine environment. Virology. 310:359-371. Harth, E., J. Romero, R. Torres, R. T. Espcjo. 2007. lntragenomic heterogeneity and intergenomic recombination among Vibrio parahaemolyticus 16S rRNA genes. Microbiology. 153:2640-2647 Hayashi, T., K. Makino, M. Ohnishi, K. Kurokawa, K. Ishii, K. Yokoyama, C. G. Han, E. Ohtsubo, K. Nakayama, T. Murata, M. Tanaka, T. Tobe, T. Iida, H. Takami, T. Honda, C. Sasakawa, N. Ogasawara, T. Yasunaga, S. Kuhara, T. Shiba, M. Hattori, H. Shinagawa. 2001. Complete genome sequence of enterohemorrhagic 301 Escherichia coli 0157:H7 and genomic comparison with a laboratory strain K-12. DNA Res. 8:11-22. Hayat, M. Z., A. Kassu, A. Mohammad, M. Yamato, N. A. Bhuiyan, N. G. Balakrish, and F. Ota. 2006. Isolation and molecular characterization of toxigenic Vibrio parahaemolyticus from the Kii Channel, Japan. Microbiol Res. 161:25-37. He, S. Y., K. Nomura and T. S. Whittam. 2004. Type Ill protein secretion mechanism in mammalian and plant pathogens. Biochim Biophys Acta. 11: 181-206. Henke, J. M. and B. L. Bassler. 2004. Quorum sensing regulates type III secretion in Vibrio harveyi and Vibrio parahaemolyticus. J Bacteriol. 186:3794-3805 Hensel, M., J. E. Shea, C. Gleeson, M. D. Jones, E. Dalton, D. W. Holden. 1995. Simultaneous identification of bacterial virulence genes by negative selection. Science. 269:400-403. Hervio-Heath, D., R. R. Colwell, A. Derrien, A. Robert-Pillot, J. M. Fournier, and M. Pommepuy. 2002. Occurrence of pathogenic vibrios in coastal areas of France. J Appl Microbiol. 92:1123-1135. Higgins, D. E. and V. J. DiRita. 1996. Genetic analysis ofthe interaction between Vibrio cholerae transcription activator ToxR and toxT promoter DNA. J Bacteriol. 178: I 080- 1087. Hofer, E. and C. H. Silva. 1984. An evaluation of the efficiency of enrichment media in the isolation process for Vibrio parahaemolyticus. Zentralbl Bakteriol Mikrobiol Hyg. 256:456-465. Hoi, L., A. Dalsgaard, J. L. Larsen, J. M. Warner and J. D. Oliver. 1997. Comparison ofribotyping and randomly amplified polymorphic DNA PCR for characterization of Vibrio vulnificus. Appl Environ Microbiol. 63:1674-1678. Honda, T., S. Chearskul, T. Takeda and T. Miwatani. 1980. Immunological methods for detection ofKanagawa phenomenon of Vibrio parahaemolyticus. J Clin Microbiol. 11:600-603. Honda, T., K. Goshima, Y. Takeda, and T. Miwatani. 1976. A bacterial cardiotoxin: thermostable direct hemolysin produced by Vibrio parahaemolyticus. Recent Adv Stud Cardiac Struct Metab 11:609-614. Honda, T., S. Honda, Y. X. Ni, and T. Miwatani. 1990. Development and application of enzyme-linked immunosorbent assay using monoclonal antibody against a hemolysin (Vp TRH) of Vibrio paraheamolyticus--evidence that Vp-TRH producing-Kanagawa phenomenon-negative V parahaemolyticus is a human pathogen. Kansenshogaku Zasshi 64:767-773. Honda, T., Y. Ni, and T. Miwatani. 1989a. Purification of a TDH-related hemolysin produced by a Kanagawa phenomenon-negative clinical isolate of Vibrio parahaemolyticus 06: K46. FEMS Microbiol Lett. 48:241-245. Honda, T., Y. Ni, T. Miwatani, T. Adachi, and J. Kim. 1992. The thermostable direct hemolysin of Vibrio parahaemolyticus is a pore-forming toxin. Can J Microbiol. 38:1175- 1180. 302 Honda, T., Y. Ni, M. Yoh, and T. Miwatani. 1989b. Production ofmonoclonal antibodies against thermostable direct hemolysin of Vibrio parahaemolyticus and application of the antibodies for enzyme-linked immunosorbent assay. Med Microbiol lmmunol. 178:245-253 Honda, T., C. Sornchai, Y. Takeda, and T. Miwatani. 1982. Immunological detection of the Kanagawa phenomenon of Vibrio parahaemolyticus on modified selective media. J Clin Microbiol. 16:734-736. Honda, T., M. Yoh, U. Kongmuang, and T. Miwatani. 1985. Enzyme-linked immunosorbent assays for detection of thermostable direct hemolysin of Vibrio parahaemolyticus. J Clin Microbiol. 22:383-386. Honeycutt, R. J., B. W. Sobral and M. McCielland. 1995. tRNA intergenic spacers reveal polymorphisms diagnostic for Xanthomonas albilineans. Microbiology. 141:3229- 3239. Hooper, W. L., G. I. Barrow, and D. J. McNab. 1974. Vibrio parahaemolyticus food poisoning in Britain. Lancet 1: 1100-1102. Hopk.ins, K. L. and A. C. Hilton. 2001. Use of multiple primers in RAPD analysis of clonal organisms provides limited improvement in discrimination. Biotechniques.30: 1262- 4, 1266-7. Hu, J., J. van Eysden and C. F. Quiros. 1995. Generation of ON A-based Markers in Specific Genome Regions by Two-primer RAPD reactions. PCR Methods Appl. 4:346-351 Hueck, C. J. 1998. Type III protein secretion systems in bacterial pathogens of animals and plants. Microbiol Mol Bioi Rev. 62:379-433. Hugh, R. and J. C. Feeley. 1972. Report (1966-1970) of the subcommittee on taxonomy of vibrios to the International Committee on Nomenclature of Bacteria. lnt J Syst Bacteriol. 2:123-126. Hulton, C. S., C. F. Higgins and P. M. Sharp.199l. ERIC sequences: a novel family of repetitive elements in the genomes of Escherichia coli, Salmonella typhimurium and other enterobacteria. Mol Microbiol. 5:825-834. Hunter, P. R. and M. A. Gaston. 1988. Numerical index of the discriminatory ability of typing systems: an application of Simpson's index of diversity. J Clin Microbiol. 26:2465- 2466. Huntley, J. S. and A. C. Hall. 1994. Aspects of the haemolytic reaction induced by Kanagawa haemolysin of Vibrio parahaemolyticus. Toxicon 32:1397-1412. Huntley, J. S. and A. C. Hall. 1996. Nature of the cation leak induced in erythrocyte membranes by Kanagawa haemolysin of Vibrio parahaemolyticus. Biochim Biophys Acta 1281:220-226. Hurley, C. C., A. Quirke, F. J. Reen, and E. F. Boyd. 2006. Four genomic islands that mark post-1995 pandemic Vibrio parahaemolyticus isolates. BMC Genomics 7: I 04. Hyashi, T., K. Mak.ino, M. Ohnishi, K. Kurokawa, K. Ishii, K. Yokoyama, C. G. Han, E. Ohtsubo, K. Nakayama & other authors. 2001. Complete genome sequence of 303 enterohemorrhagic Escherichia coli 0157:H7 and genomic comparison with a laboratory strain K-12. DNA Res. 8: 11-22. Hyytia-Trees, E.K., K. Cooper, E. M. Ribot and P. Gerner-Smidt. 2007. Recent developments and future prospects in subtyping of food borne bacterial pathogens. Future Microbial. 2:175-178 ldeses, D., U. Gopbna, Y. Paitan, R. R. Cbaudburi, M. J. Pallen and E. Z. Ron. 2005. A degenerate type Ill secretion system from septicemic Escherichia coli contributes to pathogenesis. J Bacterial. 187:8164-8171. Igucbi, T., S. Kondo, and K. Hisatsune. 1995. Vibrio parahaemolyticus 0 serotypes from 01 to 013 all produceR-type lipopolysaccharide: SOS-PAGE and compositional sugar analysis. FEMS Microbial Lett. 130:287-292. Iida, T., 0. Sutbienkul, K. S. Park, G. Q. Tang, R. K. Yamamoto, M. Isbibashi, K. Yamamoto, and T. Honda. 1997. Evidence for genetic linkage between the ure and trh genes in Vibrio parahaemolyticus. J Med Microbial. 46:639-645. Iida, T., K. S. Park, 0. Suthienkul, J. Kozawa, Y. Yamaichi, K. Yamamoto, and T. Honda. 1998. Close proximity of the tdh, trh and ure genes on the chromosome of Vibrio parahaemolyticus. Microbiology 144:2517-2523. Iida, T. and K. Yamamoto. 1990. Cloning and expression of two genes encoding highly homologous hemolysins from a Kanagawa phenomenon-positive Vibrio parahaemolyticus T4750 strain. Gene 93:9-15. Iijima, Y., N. T. Asako, M. Aihara, and K. Hayashi. 2004. Improvement in the detection rate of diarrhoeagenic bacteria in human stool specimens by a rapid real-time PCR assay. J Med Microbial. 53:617-622. Inoue, T., S. Matsuzaki, and S. Tanaka. 1995. Cloning and sequence analysis of Vibrio parahaemolyticus ompK gene encoding a 26-kDa outer membrane protein, OmpK, that serves as receptor for a broad-host-range vibriophage, KVP40. FEMS Microbial Lett. 134:245-249. Ishihara, M., A. Kawanishi, H. Watanabe, K. Tomocbika, S. Miyosbi, and S. Shinoda. 2002. Purification of a serine protease of Vibrio parahaemolyticus and its characterization. Microbial Immunol. 46:298-303. Islam, M. S., R. Tasmin, S. I. Khan, H. B. Bakht, Z. H. Mahmood, M. Z. Rahman, N. A. Bhuiyan, M. Nishibucbi, G. B. Nair, R. B. Sack, A. Huq, R. R. Colwell, and D. A. Sack. 2004. Pandemic strains of03:K6 Vibrio parahaemolyticus in the aquatic environment of Bangladesh. Can J Microbial. 50:827-834. Izutsu, K., K. Kurokawa, K. Tashiro, S. Kuhara, T. Hayashi, T. Honda and T. Iida. 2008. Comparative genomic analysis using microarray demonstrates a strong correlation between the presence of the 80-kilobase pathogenicity island and pathogenicity in Kanagawa phenomenon-positive Vibrio parahaemolyticus strains. Infect Immun. 76: I 016- 1023. Ji, Y. E., K. E. Kempsell, M. J. Colston, and R. A. Cox. 1994. Nucleotide sequences of the spacer-1, spacer-2 and trailer regions of the rrn operons and secondary structures of 304 precursor 23S rRNAs and precursor SS rRNAs of slow-growing mycobacteria. Microbiology 140:1763-1773. Jiang, X. and T. J. Chai. 1996. Survival of Vibrio parahaemolyticus at low temperatures under starvation conditions and subsequent resuscitation of viable, nonculturable cells. Appl Environ Micro bioi. 62:1300-1305 Johnson, D. E., L. Weinberg, J. Ciarkowski, P. West, and R. R. Colwell. 1984. Wound infection caused by Kanagawa-negative Vibrio parahaemolyticus. J Clin Microbiol. 20:811-812. Johnson, H. C., J. A. Baross, and J. Liston. 1971. Vibrio parahaemolyticus and its importance in seafood hygiene. JAm Vet Med Assoc. 159:1470-1473. Johnston, M. D. and M. H. Brown. 2002. An investigation into the changed physiological state of Vibrio bacteria as a survival mechanism in response to cold temperatures and studies on their sensitivity to heating and freezing. J Appl Microbiol. 92:1066-1077 0 Kam, K. M., C. K. Luey, M. B. Parsons, K. L. Cooper, G. B. Nair, M. Alam, M. A. Islam, D. T. Cheung, Y. W. Chu, T. Ramamurthy, G. P. Pazhani, S. K. Bhattacharya, H. Watanabe, J. Terajima, E. Arakawa, 0. A. Ratchtrachenchai, S. Huttayananont, E. M. Ribot, P. Gerner-Smidt, B. Swaminathan; Vibrio paraltaemolyticus PulseNet PFGE Protocol Working Group. 2008. Evaluation and validation of a PulseNet standardized pulsed-field gel electrophoresis protocol for subtyping Vibrio parahaemolyticus: an international multicenter collaborative study. J Clin Microbiol. 46:2766-2773. Kaneko, T. and R. R. Colwell. 1973. Ecology of Vibrio parahaemolyticus in Chesapeake Bay. J Bacteriol. 113:24-32. Kaneko, T. and R. R. Colwell. 1975. Incidence of Vibrio parahaemolyticus in Chesapeake Bay. Appl Microbiol. 30:251-257. Kaper, J., R. J. Seidler, H. Lockman and R. R. Colwell. 1979. Medium for the presumptive identification of Aeromonas hydrophila and Enterobacteriaceae. Appl Environ Microbiol. 38: 1023-1026. Karunasagar, I., G. Sugumar, I. Karunasagar, and P. J. Reilly. 1996. Rapid polymerase chain reaction method for detection ofKanagawa positive Vibrio parahaemolyticus in seafoods. lnt J Food Microbiol. 31:317-323. Katoh, H. 1965. Studies on the growth rate of various food bacteria. The generation time of Vibrio parahaemolyticus. Nippon Saikingaku Zasshi 20:94-99. Kaufman, G. E., M. L. Myers, C. L. Pass, A. K. Bej, and C. A. Kaysner. 2002. Molecular analysis of Vibrio parahaemolyticus isolated from human patients and shellfish during US Pacific north-west outbreaks. Lett Appl Microbiol. 34:155-161. Kaufman, G. E., G. M. Blackstone, M. C. Vickery, A. K. Bej, J. Bowers, M. D. Bowen, R. F. Meyer, and A. Depaola. 2004. Real-time PCR quantification of Vibrio parahaemolyticus in oysters using an alternative matrix. J Food Prot. 67:2424-2429. Kaysner, C. A., C. Abeyta, Jr., P.A. Trost, J. H. Wetherington, K. C. Jinneman, W. E. Hill, and M. M. Wekell. 1994. Urea hydrolysis can predict the potential pathogenicity 305 of Vibrio parahaemolylicus strains isolated in the Pacific Northwest. Appl Environ Microbiol. 60:3020-3022. Kaysner C. C. and A. DePaola. 2004. Vibrio. In: Bacteriological Analytical Manual Online (revised). Chapter 9. US Food and Drug Admisistration. Available at: http://www.cfsan.fda.gov/-ebam/bam-9.html Kelly, M. T. and E. M. Stroh. 1988. Temporal relationship of Vibrio parahaemolyticus in patients and the environment. J Clin Microbiol. 26:1754-1756. Kelly, M. T. and E. M. Stroh. 1989. Urease-positive, Kanagawa-negative Vibrio parahaemolyticus from patients and the environment in the Pacific Northwest. J Clin Microbiol. 27:2820-2822. Khaira, G. and E. Galanis. 2007. Descriptive epidemiology of Vibrio parahaemolyticus and other Vibrio species infections in British Columbia: 2001-2006. Can Commun Dis Rep. 33:12-22. Khan, A. A., S. McCarthy, R. F. Wang, and C. E. Ceroiglia. 2002. Characterization of United States outbreak isolates of Vibrio parahaemolyticus using enterobacterial repetitive intergenic consensus (ERIC) PCR and development of a rapid PCR method for detection of03:K6 isolates. FEMS Microbiol Lett. 206:209-214. Kim, Y. B., J. Okuda, C. Matsumoto, N. Takahashi, S. Hashimoto, and M. Nishibuchi. 1999. Identification of Vibrio parahaemolyticus strains at the species level by PCR targeted to the toxR gene. J Cl in Microbiol. 37:1173-1177. Kim, Y. K. and L. L. McCarter. 2000. Analysis of the polar flagellar gene system of Vibrio parahaemolyticus. J Bacteriol. 182:3693-3704. Kimura, B., Y. Sekine, H. Takahashi, Y. Tanaka, H. Obata, A. Kai, S. Morozumi and T. Fujii. 2008. Multiple-locus variable-number of tandem-repeats analysis distinguishes Vibrio parahaemolyticus pandemic 03:K6 strains. J Microbiol Methods. 72:313-320. Kingsley, R. A. and A. J. Baumler. 2000. Host adaptation and the emergence of infectious disease: the Salmonella paradigm. Mol Micro bioi. 36: I 006-1014. Kishishita, M., N. Matsuoka, K. Kumagai, S. Yamasaki, Y. Takeda, and M. Nishibuchi. 1992. Sequence variation in the thermostable direct hemolysin-related hemolysin (trh) gene of Vibrio parahaemolyticus. Appl Environ Microbiol. 58:2449-2457. Klontz, K. C. 1990. Fatalities associated with Vibrio parahaemolyticus and Vibrio cholerae non-01 infections in Florida (1981 to 1988). South Med J. 83:500-502. Kodama, T., H. Hiyoshi, K. Gotoh, Y. Akeda, S. Matsuda, K. S. Park, V. V. Cantarelli, T. Iida and T. Honda. 2008. Identification of two translocon proteins of Vibrio parahaemolyticus type Ill secretion system 2. Infect Immun. 76:4282-4289. Kodama, T., M. Rokuda, K. S. Park, V. V. Cantarelli, S. Matsuda, T. lida and T. Honda. 2007. Identification and characterization of V opT, a novel ADP-ribosyltransferase effector protein secreted via the Vibrio parahaemolyticus type Ill secretion system 2. Cell Microbiol. 9:2598-2609 Kolsto, A. B. 1997. Dynamic bacterial genome organization. Mol Microbiol. 24:241-248 306 Kogure, K. 1998. Bioenergetics of marine bacteria. Curr Opin Biotechnol. 9:278-282. Kong, R. Y., S. K. Lee, T. W. Law, S. H. Law, and R. S. Wu. 2002. Rapid detection of six types of bacterial pathogens in marine waters by multiplex PCR. Water Res. 36:2802- 2812. Kostman, J. R., M. B. Alden, M. Mair, T. D. Edlind, J. J. LiPuma, and T. L. Stull. 1995. A universal approach to bacterial molecular epidemiology by polymerase chain reaction ribotyping. J Infect Dis. 171:204-208. Kotetishvili, M., 0. C. Stine, Y. Chen, A. Kreger, A. Sulakvelidze, S. Sozhamannan and J. G. Morris Jr. 2003. Multilocus sequence typing has better discriminatory ability for typing Vibrio cholerae than does pulsed-field gel electrophoresis and provides a measure of phylogenetic relatedness. J Cl in Microbiol. 41:2191-2196. Kourany, M. 1983. Medium for isolation and differentiation of Vibrio parahaemolyticus and Vibrio a/ginolyticus. Appl Environ Microbiol. 45:310-312. Kourany, M., R. J. Kinney, and M. A. Vasquez. 1974. Vibrio parahaemolyticus in seawater off the Pacific coast of Panama. Am J Trop Med Hyg. 23:714-715. Kristensen, K. K. 1974. The occurrence of Vibrio parahaemolyticus and Vibrio alginolyticus in the Sound. Nord Vet Med. 26:188-196. Kudva, I. T., P. S. Evans, N. T. Peroa, T. J. Barrett, F. M. Ausubel, F. R. Blattner and S. M. Calderwood. 2002. Strains of Escherichia coli 0157:H7 differ primarily by insertions or deletions, not single-nucleotide polymorph isms. J Bacteriol.184: 1873-1879. Kwok, A. Y., J. T. Wilson, M. Coulthart, L. K. Ng, L. Mutharia, and A. W. Chow. 2002. Phylogenetic study and identification of human pathogenic Vibrio species based on partial hsp60 gene sequences. Can J Microbiol. 48:903-91 0. Lang, P.A., S. Kaiser, S. Myssina, C. Birka, C. Weinstock, H. Northoff, T. Wieder, F. Lang, and S. M. Huber. 2004. Effect of Vibrio parahaemolyticus haemolysin on human erythrocytes. Cell Microbiol. 6:391-400. Laohaprertthisan, V., A. Chowdhury, U. Kongmuang, S. Kalnauwakul, M. Ishibashi, C. Matsumoto, and M. Nishibuchi. 2003. Prevalence and serodiversity of the pandemic clone among the clinical strains of Vibrio parahaemolyticus isolated in southern Thailand. Epidemiol Infect. 130:395-406. Larsen, J. L., A. F. Farid, and I. Dalsgaard. 1981. Occurrence of Vibrio parahaemolyticus and Vibrio alginolyticus in marine and estuarine bathing areas in Danish coast. Zentralbl Bakteriol Mikrobiol Hyg. 173:338-345. Leblond-Bourget, N., H. Philippe, I. Mangin, and B. Decaris. 1996. 16S rRNA and 16S to 23S internal transcribed spacer sequence analyses reveal inter- and intraspecific Bifidobacterium phylogeny. lnt J Syst Bacteriol. 46: I 02-111. Lee, C. and S. F. Pan. 1993. Rapid and specific detection ofthe thermostable direct haemolysin gene in Vibrio parahaemolyticus by the polymerase chain reaction. J Gen Microbiol. 139:3225-3231. 307 Lee, C. Y., S. F. Pan, and C. H. Chen. 1995. Sequence of a cloned pR72H fragment and its use for detection of Vibrio parahaemolyticus in shellfish with the PCR. Appl Environ Microbial. 61: 1311-1317. Lee, C. Y., G. Panicker, and A. K. Bej. 2003. Detection of pathogenic bacteria in shellfish using multiplex PCR followed by CovaLink NH microwell plate sandwich hybridization. J Microbial Methods 53:199-209. Lee, J. H., K. H. Han, S. Y. Choi, M. E. Lucas, C. Mondlane, M. Ansaruzzaman, G. B. Nair, D. A. Sack, L. von Seidlein, J. D. Clemens, M. Song, J. Chun and D. W. Kim; Mozambique Cholera Vaccine Demonstration Project Coordination Group. 2006. Multilocus sequence typing (MLST) analysis of Vibrio cholerae Ol El Tor isolates from Mozambique that harbour the classical CTX prophage. J Med Microbial. 55:165-70 Leelayuwat, C., A. Romphruk, A. Lulitanond, S. Trakulsomboon and V. Thamlikitkul. 2000. Genotype analysis of Burkho/deria pseudomallei using randomly amplified polymorphic DNA (RAPD): indicative of genetic differences amongst environmental and clinical isolates. Acta Trop. 77:229-237 Li, C. C., J. A. Crawford, V. J. DiRita and J. B. Kaper. 2000. Molecular cloning and transcriptional regulation of ompT, a ToxR-repressed gene in Vibrio cholerae. Mol Microbial. 35:189-203. Lin, Z., K. Kumagai, K. Baba, J. J. Mekalanos, and M. Nishibuchi. 1993. Vibrio parahaemolyticus has a homolog of the Vibrio cholerae toxRS operon that mediates environmentally induced regulation of the thermostable direct hemolysin gene. J Bacterial. 175:3844-3855. Lopes, A. C., J. F. Rodrigues, M. B. Clementino, C. A. Miranda, A. P. Nascimento, and M. A. Morais Jr. 2007. Application ofPCR ribotyping and tDNA-PCR for Klebsiella pneumoniae identification. Mem Inst Oswaldo Cruz 102:827-832. Lozano-Leon, A., J. Torres, C. R. Osorio, and .J. Martinez-Urtaza. 2003. Identification of tdh-positive Vibrio parahaemolyticus from an outbreak associated with raw oyster consumption in Spain. FEMS Microbial Lett. 226:281-284. Lu, P. L., S. C. Chang, H. J. Pan, M. L. Chen, and K. T. Luh. 2000. Application of pulsed-field gel electrophoresis to the investigation of a nosocomial outbreak of Vibrio parahaemolyticus. J Microbial Imnumol Infect. 33:29-33. Loan, X. Y., J. X. Chen, X. H. Zhang, J. T. Jia, F. R. Sun, and Y. Li. 2007. Comparison of different primers for rapid detection of Vibrio parahaemolyticus using the polymerase chain reaction. Lett Appl Microbial. 44:242-247. Luna, A., J. F. Julian, A. Mariscal, N. Sopena, J. Fernandez-Liamazares and M. Broggi. 2000. Breast tuberculosis in a man. Breast. 9:58-59. MacDonell, M. T., F. L. Singleton, and M. A. Hood. 1982. Diluent composition for use of API 20E in characterizing marine and estuarine bacteria. Appl Environ Microbial. 44:423-427. Maeda, T., N. Takada, M. Furushita, and T. Shiba. 2000. Structural variation in the 16S-23S rRNA intergenic spacers of Vibrio parahaemolyticus. FEMS Microbial Lett. 192:73-77. 308 Maes, N., Y. De Gheldre, R. De Ryck, M. Vaneechoutte, H. Meugnier, J. Etienne, and M. J. Struelens. 1997. Rapid and accurate identification of Staphylococcus species by tRNA intergenic spacer length polymorphism analysis. J C1in Microbial. 35:24 77-2481. Magalhaes, M., Y. Takeda, V. Magalhaes, and S. Tateno. 1992. Brazilian urease positive strains of Vibrio parahaemolyticus carry genetic potential to produce the TDH related hemolysin. Mem Inst Oswaldo Cruz 87:167-168. Maiden, M. C., J. A. Bygraves, E. Feil, G. Morelli, J. E. Russell, R. Urwin, Q. Zhang, J. Zbou, K. Zurtb, D. A. Caugant, I. M. Feavers, M. Achtman, and B. G. Spratt. 1998. Multilocus sequence typing: a portable approach to the identification of clones within popu1ations of pathogenic microorganisms. Proc Natl Acad Sci USA. 95:3140-3145. Maiwald, M., A. von Herbay, P. W. Lepp, and D. A. Relman. 2000. Organization, structure, and variability of the rRNA operon of the Whipple's disease bacterium (Tropheryma whippelii). J Bacterial. 182:3292-3297. Makino, K., K. Oshima, K. Kurokawa, K. Yokoyama, T. Uda, K. Tagomori, Y. lijima, M. Najima, M. Nakano, A. Yamashita, Y. Kubota, S. Kimura, T. Yasunaga, T. Honda, H. Shinagawa, M. Hattori, and T. Iida. 2003. Genome sequence of Vibrio parahaemolyticus: a pathogenic mechanism distinct from that of V cholerae. Lancet 361:743-749. Maluping, R. P., C. Ravelo, C. R. Lavilla-Pitogo, K. Krovacek, and J. L. Romalde. 2005. Molecular typing of Vibrio parahaemolyticus strains isolated from the Philippines by PCR-based methods. J Appl Microbial. 99:383-391. Manning, S. D., A. S. Motiwala, A. C. Springman, W. Qi, D. W. Lacher, L. M. Ouellette, J. M. Mladonicky, P. Somsel, J. T. Rudrik, S. E. Dietrich, W. Zhang, B. Swaminatban, D. Alland and T. S. Whittam. 2008. Variation in virulence among clades of Escherichia coli 0157:H7 associated with disease outbreaks. Proc Natl Acad Sci US A.105:4868-4873. Marlina, S. Radu, C. Y. Kqueen, S. Napis, Z. Zakaria, S. A. Mutalib, and M. Nisbibuchi. 2007. Detection of tdh and trh genes in Vibrio parahaemolyticus isolated from Corbicula moltkiana prime in West Sumatera, Indonesia. Southeast Asian J Trop Med Public Health 38:349-355. Marne, C. and R. Aznar. 1988. Vibrio parahemolyticus-associated diarrhea in Spain. Eur J Clin Microbial Infect Dis. 7:439-440. Marshall, S., C. G. Clark, G. Wang, M. Mulvey, M. T. Kelly, and W. M. Jobnson. 1999. Comparison of molecular methods for typing Vibrio parahaemolyticus. J Clin Microbial. 37:2473-2478. Martinez-Urtaza, J., A. Lozano-Leon, A. Depaola, M. Ishibashi, K. Sbimada, M. Nishibuchi, and E. Liebana. 2004. Characterization of pathogenic Vibrio parahaemolyticus isolates from clinical sources in Spain and comparison with Asian and North American pandemic isolates. J Clin Microbial. 42:4672-4678. Martinez-Urtaza, J., A. Lozano-Leon, A. Vina-Feas, J. de Novoa, and 0. Garcia Martin. 2006. Differences in the API 20E biochemical patterns of clinical and environmental Vibrio parahaemolyticus isolates. FEMS Microbial Lett. 255:75-81. 309 Martinez-Urtaza, J., L. Simental, D. Velasco, A. Depaola, M. lshibashi, Y. Nakaguchi, M. Nishibuchi, D. Carrera-Fiores, C. Rey-Aivarez, and A. Pousa. 2005. Pandemic Vibrio parahaemolyticus 03 :K6, Europe. Emerg Infect Dis. 11:1319-1320. Masini, L., G. De Grandis, F. Principi, C. Mengarelli, and D. Ottaviani. 2007. Research and characterization of pathogenic vibrios from bathing water along the Conero Riviera (Central Italy). Water Res. 41:4031-4040 Maslow, J. and M. E. Mulligan. 1996. Epidemiologic typing systems. Infect Control Hosp Epidemiol. 17:595-604. Maslunka. C., E. Carr, V. Giirtler, P. Kiimpfer and R. Seviour. 2006. Estimation of ribosomal RNA operon (rrn) copy number in Acinetobacter isolates and potential of patterns of rrn operon-containing fragments for typing strains of members of this genus. Syst Appl Microbiol. 29:216-228. Masuda, N., K. Yasukawa, Y. Isawa, R. Horie, J. Saitoh, T. lshiguro, Y. Nakaguchi, M. Nishibuchi, T. Hayashi. 2004. Rapid detection of tdh and trh mRNAs of Vibrio parahaemolyticus by the transcription-reverse transcription concerted (TRC) method. J Biosci Bioeng. 98:236-243. Matsumoto, C., J. Okuda, M. Ishibashi, M. lwanaga, P. Garg, T. Rammamurthy, H. C. Wong, A. Depaola, Y. B. Kim, M. J. Albert, and M. Nishibuchi. 2000. Pandemic spread of an 03 :K6 clone of Vibrio parahaemolyticus and emergence of related strains evidenced by arbitrarily primed PCR and toxRS sequence analyses. J Cl in Microbiol. 38:578-585. Matsuzaki,. S, T. Inoue, S. Tanaka, T. Koga, M. Kuroda, S. Kimura and S. lmai. 2000, Characterization of a novel Vibrio parahaemo/yticus phage, KVP241, and its relatives frequently isolated from seawater. Microbiollmmunol. 44:953-956. McCarter, L. 1999. The multiple identities of Vibrio parahaemolyticus. J Mol Microbiol Biotechnol. 1:51-57 McCarter, L. L. 2001. Polar flagellar motility of the Vibrionaceae. Microbiol Mol Bioi Rev. 65:445-462. McCarthy, S. A., A. Depaola, D. W. Cook, C. A. Kaysner, and W. E. Hill. 1999. Evaluation of alkaline phosphatase- and digoxigenin-labelled probes for detection of the thermolabile hemolysin (tlh) gene of Vibrio parahaemolyticus. Lett Appl Microbiol. 28:66- 70. McCarthy, S. A., A. Depaola, C. A. Kaysner, W. E. Hill, and D. W. Cook. 2000. Evaluation of nonisotopic DNA hybridization methods for detection of the tdh gene of Vibrio parahaemolyticus. J Food Prot. 63:1660-1664. McClelland, M., C. Petersen, and J. Welsh. 1992. Length polymorphisms in tRNA intergenic spacers detected by using the polymerase chain reaction can distinguish streptococcal strains and species. J Clin Microbiol. 30:1499-1504. McLaughlin, J. B., A. Depaola, C. A. Bopp, K. A. Martinek, N. P. Napolilli, C. G. Allison, S. L. Murray, E. C. Thompson, M. M. Bird, and J. P. Middaugh. 2005. Outbreak of Vibrio parahaemolyticus gastroenteritis associated with Alaskan oysters. N Engl J Med. 353:1463-1470. 310 Meador, C. E., M. M. Parsons, C. A. Bopp, P. Gerner-Smidt, J. A. Painter and C. J. V ora. 2007. Virulence gene- and pandemic group-specific marker profiling of clinical Vibrio parahaemolyticus isolates. J Clin Microbiol. 45:1133-1139 Miller, E. S., J. F. Heidelberg, J. A. Eisen, W. C. Nelson, A. S. Durkin, A. Ciecko, T. V. Feldblyum, 0. White, I. T. Paulsen, W. C. Nierman, J. Lee, B. Szczypinski and C. M. Fraser. 2003. Complete genome sequence of the broad-host-range vibriophage KVP40: comparative genomics of a T4-related bacteriophage. J Bacteriol. 18:5220-5233. Miller, V. L., R. K. Taylor and J. J. Mekalanos. 1987. Cholera toxin transcriptional activator toxR is a transmembrane DNA binding protein. Cell. 48:271-279 Miwa, N., T. Nishio, Y. Arita, F. Kawamori, T. Masuda, and M. Akiyama. 2003. Evaluation of MPN method combined with PCR procedure for detection and enumeration of Vibrio parahaemolyticus in seafood. Shokuhin Eiseigaku Zasshi 44:289-293. Miwatani, T., S. Shinoda, T. Tamura, H. Nishimune, A. Tomaru, A. Yoshihara, and T. Fujino. 1969. Antigens of Vibrio parahaemolyticus. Preparation of specific antisera to somatic (0) antigen and their application in antigen analysis of Vibrio parahaemolyticus. Biken J. 12:9-15. Miyamoto, Y., T. Kato, Y. Obara, S. Akiyama, K. Takizawa, and S. Yamai. 1969. In vitro hemolytic characteristic of Vibrio parahaemolyticus: its close correlation with human pathogenicity. J Bacteriol. 100:1147-1149. Mizunoe, Y., S. N. Wai, T. Ishikawa, A. Takade and S. Yoshida. 2000. Resuscitation of viable but nonculturable cells of Vibrio parahaemolyticus induced at low temperature under starvation. FEMS Microbiol Lett.186: 115-120. Molenda, J. R., W. G. Johnson, M. Fishbein, B. Wentz, I. J. Mehlman and T. A. Dadisman Jr. 1972. Vibrio parahaemolyticus gastroenteritis in Maryland: laboratory aspects. Appl Microbiol. 24:444-448. Molero, X., R. M. Bartolome, T. Vinuesa, L. Guarner, A. Accarino, F. Casellas, and R. Garcia. 1989. Acute gastroenteritis due to Vibrio parahaemolyticus in Spain. Presentation of 8 cases. Med Cl in. (Bare.) 92:1-4. Moreno, C., J. Romero and R. T. Espejo. 2002. Polymorphism in repeated 16S rRNA genes is a common property of type strains and environmental isolates of the genus Vibrio. Microbiology. 148:1233-1239. Morris, J. G., Jr. and R. E. Black. 1985. Cholera and other vibrioses in the United States. N Engl J Med. 312:343-350. Myers, M. L., G. Panicker, and A. K. Bej. 2003. PCR detection of a newly emerged pandemic Vibrio parahaemolyticus 03:K6 pathogen in pure cultures and seeded waters from the Gulf of Mexico. Appl Environ Microbiol. 69:2194-2200. Nagayama, K., K. Yamamoto, T. Mitawani, and T. Honda. 1995. Characterisation of a haemolysin related to Vp-TDH produced by a Kanagawa phenomenon-negative clinical isolate of Vibrio parahaemo/yticus. J Med Microbiol. 42:83-90. Nairn, R., T. Iida, A. Takahashi, and T. Honda. 2001. Monodansylcadaverine inhibits cytotoxicity of Vibrio parahaemolyticus thermostable direct hemolysin on cultured rat embryonic fibroblast cells. FEMS Microbiol Lett. 196:99-1 OS. 311 Nair, G. B. 1996. Molecular insights into the evolution & epidemiology of Vibrio cholerae. Indian J Med Res. 104:5-13. Nair, G. B., M. Abrabam, and R. Natarajan. 1980. Distribution of Vibrio parahaemolyticus in finfish harvested from Porto Novo (S. India) environs: a seasonal study. Can J Microbial. 26:1264-1269. Nair, G. B., B. L. Sarkar, M. Abrabam, and S. C. Pal. 1985. Serotypes of Vibrio parahaemolyticus isolates from hydrobiologically dissimilar aquatic environments. Appl Environ Microbial. 50: 724-726 Nair, G. B., T. Ramamurtby, S. K. Bbattacharya, B. Dutta, Y. Takeda, and D. A. Sack. 2007. Global dissemination of Vibrio parahaemolyticus serotype 03:K6 and its serovariants. Clin Microbial Rev. 20:39-48. Nakagucbi, Y., T. lsbizuka, S. Obnaka, T. Hayasbi, K. Yasukawa, T. Isbiguro and M. J. Nishibuchi. 2004. Rapid and specific detection of tdh, trhl, and trh2 mRNA of Vibrio parahaemolyticus by transcription-reverse transcription concerted reaction with an automated system. Clin Microbial. 42:4284-4292. Nakasone, N., M. Ikema, N. Higa, T. Yamasbiro, and M. lwanaga. 1999. A filamentous phage of Vibrio parahaemolyticus 03:K6 isolated in Laos. Microbial Immunol. 43:385-388. Nakasone, N., S. Insisengmay and M. Iwanaga. 2000. Characterization of the pili isolated from Vibrio parahaemolyticus 03:K6. Southeast Asian J Trop Med Public Health. 31:360-365. Nakasone, N. and M. Iwanaga. 1990. Pili of a Vibrio parahaemolyticus strain as a possible colonization factor. Infect Immun. 58:61-69. Nakasone, N. and M. Iwanaga. 1992. The role of pili in colonization of the rabbit intestine by Vibrio parahaemolyticus Na2. Microbiollmmunol. 36:123-130. Nandi, S., G. Kbetawat, S. Sengupta, R. Majumder, S. Kar, R.K. Bbadra, S. Roychoudbury and J. Das. 1997. Rearrangements in the genomes of Vibrio cholerae strains belonging to different serovars and biovars. Int J Syst Bacterial. 47:858-862. Nasu, H., T. lida, T. Sugahara, Y. Yamaichi, K. S. Park, K. Yokoyama, K. Makino, H. Sbinagawa, and T. Honda. 2000. A filamentous phage associated with recent pandemic Vibrio parahaemolyticus 03:K6 strains. J Clin Microbial. 38:2156-2161. Newbury, S. F., N. H. Smith, E. C. Robinson, I. D. Riles and C. F. Higgins. 1987. Stabilization oftranslationally active mRNA by prokaryotic REP sequences. Cell. 48:297- 310. Nishibucbi, M. and J. B. Kaper. 1990. Duplication and variation of the thermostable direct haemolysin (tdh) gene in Vibrio parahaemolyticus. Mol Microbial. 4:87-99. Nishibuchi, M. and J. B. Kaper. 1995. Thermostable direct hemolysin gene of Vibrio parahaemolyticus: a virulence gene acquired by a marine bacterium. Infect Immun. 63:2093-2099. Nisbibuchi, M., V. Khaeomanee-iam, T. Honda, J. B. Kaper, and T. Miwatani. 1990. Comparative analysis of the hemolysin genes of Vibrio cholerae non-01, V. mimicus, and 312 V. hollisae that are similar to the tdh gene of V. parahaemolyticus. FEMS Microbiol Lett. 55:251-256. Nishibuchi, M., T. Taniguchi, T. Misawa, V. Khaeomanee-iam, T. Honda, and T. Miwatani. 1989. Cloning and nucleotide sequence of the gene (trh) encoding the hemolysin related to the thermostable direct hemolysin of Vibrio parahaemolyticus. Infect Immun. 57:2691-2697. Nishioka, T., M. Kamruzzaman, M. Nishibuchi and Y. Satta. 2008. On the origin and function of an insertion element VPal-1 specific to post-1995 pandemic Vibrio parahaemolyticus strains. Genes Genet Syst. 83:101-110. Nolan, C. M., J. Ballard, C. A. Kaysner, J. L. Lilja, L. P. Williams Jr. and F. C. Tenover. 1984. Vibrio parahaemolyticus gastroenteritis. An outbreak associated with raw oysters in the Pacific northwest. Diagn Microbiollnfect Dis. 2: 119-128. Nordstrom, J. L., R. Rangdale, M. C. Vickery, A. M. Phillips, S. L. Murray, S. Wagley, and A. Depaola. 2006. Evaluation of an alkaline phosphatase-labeled oligonucleotide probe for the detection and enumeration of the thermostable-related hemolysin (trh) gene of Vibrio parahaemolyticus. J Food Prot. 69:2770-2772. Nordstrom, J. L., M. C. Vickery, G. M. Blackstone, S. L. Murray, A. DePaola. 2007. Development of a multiplex real-time PCR assay with an internal amplification control for the detection of total and pathogenic Vibrio parahaemolyticus bacteria in oysters. Appl Environ Microbiol. 73:5840-5847. Nour, M. 1998. 16S-23S and 23S-5S intergenic spacer regions of lactobacilli: nucleotide sequence, secondary structure and comparative analysis. Res Microbiol. 149:433-448. O'Hara, C. M., E. G. Sowers, C. A. Bopp, S. B. Duda, and N. A. Strockbine. 2003. Accuracy of six commercially available systems for identification of members of the family Vibrionaceae. J Clin Microbial. 41:5654-5659. Ohnishi, M., J. Terajima, K. Kurokawa, K. Nakayama, T. Murata, K. Tamura, Y. Ogura, H. Watanabe and T. Hayashi. 2002. Genomic diversity of enterohemorrhagic Escherichia coli 0157 revealed by whole genome PCR scanning.Proc Natl Acad Sci US A. 99:17043-17048 Okada, K., T.lida, K. Kita-Tsukamoto and T. Honda. 2005 Vibrios commonly possess two chromosomes. J Bacteriol. 187:752-757 Okuda, J., M. Ishibashi, S. L. Abbott, J. M. Janda, and M. Nishibuchi. 1997a. Analysis of the thermostable direct hemolysin (tdh) gene and the tdh-related hemolysin (trh) genes in urease-positive strains of Vibrio parahaemolyticus isolated on the West Coast ofthe United States. J Clin Microbiol. 35:1965-1971. Okuda, J., M. Ishibashi, E. Hayakawa, T. Nishino, Y. Takeda, A. K. Mukhopadhyay, S. Garg, S. K. Bhattacharya, G. B. Nair, and M. Nishibuchi. 1997b. Emergence of a unique 03 :K6 clone of Vibrio parahaemolyticus in Calcutta, India, and isolation of strains from the same clonal group from Southeast Asian travelers arriving in Japan. J Clin Microbial. 35:3150-3155. Okuda, J., T. Nakai, P. S. Chang, T. Oh, T. Nishino, T. Koitabashi, and M. Nishibuchi. 2001. The toxR gene of Vibrio (Listonella) anguillarum controls expression of 313 the major outer membrane proteins but not virulence in a natural host model. Infect Immun. 69:6091-6101. Okura, M., R. Osawa, E. Arakawa, J. Terajima, and H. Watanabe. 2005. Identification of Vibrio parahaemolyticus pandemic group-specific DNA sequence by genomic subtraction. J Clin Microbiol. 43:3533-3536. Okura, M., R. Osawa, A. Iguchi, E. Arakawa, J. Terajima, and H. Watanabe. 2003. Genotypic analyses of Vibrio parahaemolyticus and development of a pandemic group specific multiplex PCR assay. J Cl in Microbiol. 41:4676-4682. Olive, D. M. and P. Bean. 1999. Principles and applications of methods for DNA-based typing of microbial organisms. J Cl in Microbiol. 37:1661-1669. Ono, T., K. S. Park, M. Ueta, T. Iida, and T. Honda. 2006. Identification of proteins secreted via Vibrio parahaemolyticus type Ill secretion system 1. Infect Immun. 74:1032- 1042. Osorio, C. R. and K. E. Klose. 2000. A region of the transmembrane regulatory protein ToxR that tethers the transcriptional activation domain to the cytoplasmic membrane displays wide divergence among Vibrio species. J Bacteriol. 182:526-528. Ottem, K. F., A. Nylund, E. Karlsbakk, A. Friis-Moller, and B. Krossoy. 2007. Characterization of Francisella sp., GM2212, the first Francisella isolate from marine fish, Atlantic cod (Gadus morhua). Arch Microbiol. 187:343-350. Overman, T. L., J. F. Kess1er, and J. P. Seabolt. 1985. Comparison of API 20E, API rapid E, and API rapid NFT for identification of members of the family Vibrionaceae. J Clin Microbiol. 22:778-781. Overman, T. L. and J. K. Overlcy. 1986. Feasibility of same-day identification of members of the family Vibrionaceae by the API 20E system. J Clin Microbiol. 23:715- 717. Pace, J. and T. J. Chai. 1989. Comparison of Vibrio parahaemolyticus grown in estuarine water and rich medium. Appl Environ Microbiol. 55:1877-1887. Palasuntheram, C. 1981. The halophilic properties of Vibrio parahaemolyticus. J Gen Microbiol. 127:427-428. Palasuntheram, C., N. Withana, and D. G. Amaratunga. 1977. First isolation in Sri Lanka of Vibrio parahaemolyticus from a case of gastro-enteritis. Ceylon Med J. 22:93-96. Pan, T. M., T. K. Wang, C. L. Lee, S. W. Chien, and C. B. Horng. 1997. Food-borne disease outbreaks due to bacteria in Taiwan, 1986 to 1995. J Cl in Micro bioi. 35:1260- 1262. Panicker, G., D. R. Call, M. J. Krug, and A. K. Bej. 2004. Detection of pathogenic Vibrio spp. in shellfish by using multiplex PCR and DNA microarrays. Appl Environ Microbial. 70:7436-7444. Park, K. S., T. Iida, Y. Yamaichi, T. Oyagi, K. Yamamoto, and T. Honda. 2000. Genetic characterization of DNA region containing the trh and ure genes of Vibrio parahaemolyticus. Infect Immun. 68:5742-5748. 314 Park, K. S., T. Ono, M. Rokuda, M. H. Jang, T. lida, and T. Honda. 2004a. Cytotoxicity and enterotoxicity of the them10stable direct hemolysin-deletion mutants of Vibrio parahaemolyticus. Microbiol Immunol. 48:313-318. Park, K. S., T. Ono, M. Rokuda, M. H. Jang, K. Okada, T. Iida, and T. Honda. 2004b. Functional characterization of two type Ill secretion systems of Vibrio parahaemolyticus. Infect Immun. 72:6659-6665. Parsons, M.B., K. L. Cooper, K. A. Kubota, N. Puhr, S. Simington, P. S. Calimlim, D. Schoonmaker-Bopp, C. Bopp, B. Swaminathan, P. Gerner-Smidt and E. M. Ribot. 2007. PulseNet USA standardized pulsed-field gel electrophoresis protocol for subtyping of Vibrio parahaemolyticus. Food borne Pathog Dis. 4:285-92 Parvathi, A., H. S. Kumar, A. Bhanumathi, M. Ishibashi, M. Nishibuchi, I. Karunasagar, and I. Karunasagar. 2006. Molecular characterization of thermostable direct haemolysin-related haemolysin (TRH)-positive Vibrio parahaemolyticus from oysters in Mangalore, India. Environ Microbiol. 8:997-1004. Pattison, J. A., S. K. Samuelian, and C. A. Weber. 2007. Inheritance ofPhytophthora root rot resistance in red raspberry determined by generation means and molecular linkage analysis. Theor Appl Genet. 115:225-236. Perez-Luz, S., J. Fernandez, F. Rodriguez-Valera, L. Pascual, C. Moreno, A. Amo, D. Apraiz, and V. Catalan. 2002. Sequence diversity of the internal transcribed spacer (ITS) region of the rRNA operons among different serogroups of Legionella pneumophila isolates. Syst Appl Microbiol. 25:212-219. Perna, N. T., G. Plunkett 3rd, V. Burland, B. Mau, J. D. Glasner, D. J. Rose, G. F. Mayhew, P. S. Evans, J. Gregor, H. A. Kirkpatrick, G. Posfai, J. Hackett, S. Klink, A. Boutin, Y. Shao, L. Miller, E. J. Grotbeck, N. W. Davis, A. Lim, E. T. Dimalanta, K. D. Potamousis, J. Apodaca, T. S. Anantharaman, J. Lin, G. Yen, D. C. Schwartz, R. A. Welch and F. R. Blattner. 2001. Genome sequence of enterohaemorrhagic Escherichia coli 0157:H7. Nature. 409:529-533. Pfeffer, C. and J. D. Oliver. 2003. A comparison of thiosulphate-citrate-bile salts-sucrose (TCBS) agar and thiosulphate-chloride-iodide (TCI) agar for the isolation of Vibrio species from estuarine environments. Lett Appl Microbiol. 36: 150-151. Pourshafie, M. R., B. Bakhshi, R. Ranjbar, M. Sedaghat, N. Sadeghifard, J. Zaemi Yazdi, M. Parzadeh and J. Raesi. 2007 Dissemination of a single Vibrio cholerae clone in cholera outbreaks during 2005 in Iran. J Med Microbiol. 56: 1615-1619. Pruzzo, C., R. Tarsi, M. M. Lleo, C. Signoretto, M. Zampini, L. Pane, R. R. Colwell and P. Canepari. 2003. Persistence of adhesive properties in Vibrio cholerae after long term exposure to sea water. Environ Microbiol. 5:850-858 Quilici, M. L., A. Robert-Pillot, J. Picart, and J. M. Fournier. 2005. Pandemic Vibrio parahaemolyticus 03:K6 spread, France. Emerg Infect Dis. 11:1148-1149. Raghunath, P., B. Pradeep, I. Karunasagar, and I. Karunasagar. 2007. Rapid detection and enumeration of /rh-carrying Vibrio parahaemolyticus with the alkaline phosphatase-labelled oligonucleotide probe. Environ Microbiol. 9:266-270. 315 Rahman, M., N. A. Bhuiyan, I. Kuhn, T. Ramamurthy, M. Rahman, R. Mollby, and G. B. Nair. 2006. Biochemical fingerprinting of Vibrio parahaemolyticus by the PhenePlate system: comparison between pandemic and non-pandemic serotypes. Epidemiol Infect. 134:985-989. Raimondi, F., J. P. Kao, C. Fiorentini, A. Fabbri, G. Donelli, N. Gasparini, A. Rubino, and A. Fasano. 2000. Enterotoxicity and cytotoxicity of Vibrio parahaemolyticus thermostable direct hemolysin in in vitro systems. Infect Immun. 68:3180-3185. Rampersad, J. N., J. D. Watkins, M. S. Samlal, R. Deonanan, S. Ramsubeik, and D. R. Ammons. 2005. A nested-PCR with an Internal Amplification Control for the detection and differentiation of Bartonella henselae and B. clarridgeiae: an examination of cats in Trinidad. BMC Infect Dis. 5:63. Rasschaert. G., K. Houf, H. lmberechts, K. Grijspeerdt, L. De Zutter and M. Heyndrickx. 2005.Comparison of five repetitive-sequence-based PCR typing methods for molecular discrimination of Salmonella enterica isolates. J Clin Microbiol. 43:3615-3623. Ray, B., S. M. Hawkins, and C. R. Hackney. 1978. Method for the detection of injured Vibrio parahaemolyticus in seafoods. Appl Environ Microbiol. 35:1121-1127. Reddy, P. V. and K. M. Soliman. 1997. Identification of wild and cultivated Hordeum species using two-primer RAPD fragments. Biologia Plantarum. 39:543-552 Reiter, W. D., P. Palm and S Yeats. 1989. Transfer RNA genes frequently serve as integration sites for prokaryotic genetic elements. Nucleic Acids Res. 17:1907-1914. Rizvi, A. V., G. Panicker, M. L. Myers, and A. K. Bej. 2006. Detection of pandemic Vibrio parahaemolyticus 03:K6 serovar in Gulf of Mexico water and shellfish using real time PCR with Taqman fluorescent probes. FEMS Microbiol Lett. 262:185-\92. Robert-Pillot, A., A. Guenole, J. Lesne, R. Delesmont, J. M. Fournier and M. L. Quilici. 2004. Occurrence of the tdh and trh genes in Vibrio parahaemolyticus isolates from waters and raw shellfish collected in two French coastal areas and from seafood imported into France. Int J Food Micro bioi. 91:319-325. Romesburg, H. C. 1984. Cluster analysis for researchers. Lifetime Learning Publications, Belmont, CA. Roring, S., A. N. Scott, H. R. Glyn, S. D. Neill, and R. A. Skuce. 2004. Evaluation of variable number tandem repeat (VNTR) loci in molecular typing of Mycobacterium bovis isolates from Ireland. Vet Microbiol. 101:65-73. Rupnik, M., V. Avesani, M. Janc, C.von Eichel-Streiber, and M. Delmee. 1998. A novel toxinotyping scheme and correlation of toxinotypes with serogroups of Clostridium difficile isolates. J Clin Microbiol. 36: 2240-2247. Ryang, D. W., S. W. Cho, M. G. Shin, J. H. Shin, and S. P. Suh. 1997. Molecular typing of Vibrio vulnificus isolates by random amplified polymorphic DNA (RAPD) analysis. Jpn J Med Sci Bioi. 50: 113-121. Sadeghifard, N., V. Giirtler, M. Beer and R.J. Seviour. 2006. The mosaic nature of intergenic 16S-23S rRNA spacer regions suggests rRNA operon copy number variation in Clostridium difficile strains. Appl Environ Microbiol. 72:7311-7323. 316 Sakazaki, R., S. lwanami and H. Fukumi. 1963. Studies on the enteropathogenic, facultatively halophilic bacteria, Vibrio parahaemolyticus. Morphological, cultural and biochemical properties and its taxonomical position. Jpn J Med Sci Bioi. 16:161-88 Sakurai, J., T. Honda, Y. Jinguji, M. Arita, and T. Miwatani. 1976. Cytotoxic effect of the thermostable direct hemolysin produced by Vibrio parahaemolyticus on FL cells. Infect Inunun. 13:876-883. Sakurai, J., A. Matsuzaki, and T. Miwatani. 1973. Purification and characterization of thermostable direct hemolysin of Vibrio parahaemolyticus. Infect Irnmun. 8:775-780. Sakurai, J., A. Matsuzaki, Y. Takeda, and T. Miwatani. 1974. Existence of two distinct hemolysins in Vibrio parahaemolyticus. Infect Immun. 9:777-780. Sambrook, J., E. F. Fritsch and T. Maniatis. 1989. Molecular cloning: A laboratory manual. 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY Samore, M. H., K. M. Bettin, P.C. DeGirolami, C. R. Clabots, D. N. Gerding, and A. W. Karchmer. 1994. Wide diversity of Clostridium difficile types at a tertiary referral hospital. J Infect Dis. 170:615-621. Sautter, R. L., J. S. Taylor, J. D. Oliver and C. O'Donnell. 1988. Vibrio parahaemolyticus (Kanagawa-negative) wound infection in a hospital dietary employee. Diagn Microbiol Infect.Dis. 9:41-45. Schoeffield, A. J. and H. N. Williams. 1990. Efficiencies of Recovery of Bdellovibrios from Brackish- Water Environments by Using Various Bacterial Species as Prey. Appl Environ Microbial. 56:230-236. Sen, B., B. Dutta, S. Chatterjee, M. K. Bhattacbarya, R. K. Nandy, A. K. Mukhopadbyay, D. N. Gangopadbyay, S. K. Bhattacharya and T. Ramamurtby. 2007. The first outbreak of acute diarrhea due to a pandemic strain of Vibrio parahaemolyticus 03:K6 in Kolkata, India. Int J Infect Dis.11:185-187. Serichantalergs, 0., N. A. Bhuiyan, G. B. Nair, 0. Chivaratanond, A. Srijan, L. Bodbidatta, S. Anuras, and C. J. Mason. 2007. The dominance of pandemic serovars of Vibrio parahaemolyticus in expatriates and sporadic cases of diarrhoea in Thailand, and a new emergent serovar (03 : K46) with pandemic traits. J Med Microbial. 56:608-613. Shaaly. A., M. G. Tellevik, N. Langeland, E. A. Hoiby and R. Jureen. 2005. Comparison of serotyping, pulsed field gel electrophoresis and amplified fragment length polymorphism for typing of Streptococcus pneumoniae. J Med Microbial. 54:467-472 Shinagawa, M. Hattori, and T. Iida. 2003. Genome sequence of Vibrio parahaemolyticus: a pathogenic mechanism distinct from that of V cholerae. Lancet 361:743-749. Sokurenko, E. V., D. L. Hasty and D.E. Dykhuizen. 1999. Pathoadaptive mutations: gene loss and variation in bacterial pathogens. Trends Microbial. 7:191-195. Sparagano, 0. A., P. A. Robertson, I. Purdom, J. Mclnnes, Y. Li, D. H. Yu, Z. J. Du, H. S. Xu, and B. Austin. 2002. PCR and molecular detection for differentiating Vibrio species. Ann N Y Acad Sci. 969:60-65. 317 Stanley, J. and N. Saunders. 1996. DNA insertion sequences and the molecular epidemiology of Salmonella and Mycobacterium. J Med Microbial. 45:236-251 Stenutz. R., A. Weintraub and G. Widmalm. 2006. The structures of Escherichia coli a polysaccharide antigens. FEMS Microbial Rev. 30:382-403 Stern, M. J., G. F. Ames, N. H. Smith, E. C. Robinson and C. F. Higgins. 1984. Repetitive extragenic palindromic sequences: a major component of the bacterial genome. Cell. 37:1015-1026. Stevens, M. P., M. W. Wood, L. A. Taylor, P. Monaghan, P. Hawes, P. W. Jones, T. S. Wallis and E. E. Galyov. 2002. An Inv!Mxi-Spa-like type III protein secretion system in Burkholderia pseudomallei modulates intracellular behaviour of the pathogen. Mol Microbial. 46:649-659 Stewart, F. J. and C. M. Cavanaugh. 2007. lntragenomic variation and evolution of the internal transcribed spacer of the rRNA operon in bacteria. J Mol Evol. 65:44-67 Struelens, M. J. 1996. Consensus guidelines for appropriate use and evaluation of microbial epidemiologic typing systems. Clin Microbial Infect. 2:2-11 Struelens, M. J. 1998. Molecular epidemiologic typing systems of bacterial pathogens: current issues and perspectives. Mem Inst Oswaldo Cruz. 93:581-585 Su, Y. and C. Liu. 2007. Vibrio parahaemolyticus: A concern for seafood safety. Food Microbiology. 24. 549-558 Suthienkul, 0., T. Iida, K. S. Park, M. Ishibashi, S. Supavej, K. Yamamoto, and T. Honda. 1996. Restriction fragment length polymorphism of the tdh and trh genes in clinical Vibrio parahaemolyticus strains. J Cl in Micro bioi. 34:1293-1295. Suthienkul, 0., M. Ishibashi, T. Iida, N. Nettip, S. Supavej, B. Eampokalap, M. Makino, and T. Honda. 1995. Urease production correlates with possession of the trh gene in Vibrio parahaemolyticus strains isolated in Thailand. J Infect Dis. 172:1405-1408. Suzuki, N., Y. Ueda, T. Furukawa, Y. Takegaki, K. Miyagi, K. Noda, H. Hirose, S. Hashimoto, S. Yano, M. Ishibashi, and T. Honda. 1997. Incidence ofKanagawa phenomenon-positive and -negative Vibrio parahaemolyticus strains isolated from traveller's diarrhea and their relation to tdh and I rh genes. Kansenshogaku Zasshi 71:417- 420. Tacao, M., A. Alves, M. J. Saavedra and A. Correia. 2005. BOX-PCR is an adequate tool for typingAeromonas spp year. Antonie Van Leeuwenhoek.88: 173-179. Tacket, C. 0., T. J. Barrett, G. E. Sanders, and P.A. Blake. 1982. Panophthalmitis caused by Vibrio parahaemolyticus. J Cl in Microbial. 16:195-196. Tada, J., T. Ohashi, N. Nishimura, Y. Shirasaki, H. Ozaki, S. Fukushima, J. Takano, M. Nishibuchi, and Y. Takeda. 1992. Detection of the thermostable direct hemolysin gene (tdh) and the thermostable direct hemolysin-related hemolysin gene (trh) of Vibrio parahaemolyticus by polymerase chain reaction. Mol Cell Probes 6:477-487. Tagomori, K., T. lida and T. Honda. 2002 Comparison of genome structures ofvibrios, bacteria possessing two chromosomes. J Bacterial. 184:4351-4358 318 Takabashi, A., C. Yamamoto, T. Kodama, K. Yamashita, N. Harada, M. Nakano, T. Honda, and Y. Nakaya. 2006. Pore fonnation ofthennostable direct hemolysin secreted from Vibrio parahaemolyticus in lipid hi layers. Int J Toxicol. 25:409-418. Takabashi, H., Y. Iwade, H. Konuma, and Y. Hara-Kudo. 2005. Development of a quantitative real-time PCR method for estimation of the total number of Vibrio parahaemolyticus in contaminated shellfish and sea water. J Food Prot. 68: I 083-1088. Tampakaki, A. P., V. E. Fadouloglou, A. D. Gazi, N. J. Panopoulos and M. Kokkinidis. 2004. Conserved features of type Ill secretion. Cell Microbiol. 6:805-816. Review. Tamplin, M. L., J. K. Jackson, C. Bucbrieser, R. L. Murphree, K. M Portier, V. Gangar, L. G. Miller and C. W. Kaspar. 1996. Pulsed-field gel electrophoresis and ribotype profiles of clinical and environmental Vibrio vulnificus isolates. Appl Environ Microbiol. 62:3572-3580. Tanabe, T., T. Funahashi, H. Nakao, S. Miyoshi, S. Shinoda, and S. Yamamoto. 2003. Identification and characterization of genes required for biosynthesis and transport of the siderophore vibrioferrin in Vibrio parahaemolyticus. J Bacterial. 185:6938-6949. Tanabe, T., H. Nakao, T. Kuroda, T. Tsuchiya, and S. Yamamoto. 2006.lnvolvement of the Vibrio parahaemolyticus pvsC gene in export of the siderophore vibrioferrin. Microbial Immunol. 50:871-876. Tang, G. Q., T. lida, K. Yamamoto, and T. Honda. 1995. Ca(2+)-independent cytotoxicity of Vibrio parahaemolyticus thermostable direct hemolysin (TDH) on Intestine 407, a cell line derived from human embryonic intestine. FEMS Microbiol Lett. 134:233- 238. Tanigucbi, H., H. Hirano, S. Kubomura, K. Higasbi, Y. Mizuguchi. 1986 Comparison of the nucleotide sequences of the genes for the thennostable direct hemolysin and the thennolabile hemolysin from Vibrio parahaemolyticus. Microb Pathog. 1:425-432. Taniguchi, H., S. Kubomura, H. Hirano, K. Mizue, M. Ogawa and Y. Mizuguchi. 1990. Cloning and characterization of a gene encoding a new thennostable hemolysin from Vibrio parahaemolyticus. FEMS Microbial Lett. 55:339-345. Taniguchi, H., H. Ohta, M. Ogawa, and Y. Mizuguchi. 1985. Cloning and expression in Escherichia coli of Vibrio parahaemolyticus thermostable direct hemolysin and thennolabile hemolysin genes. J Bacterial. 162:5 I 0-515. Tanigucbi, H., K. Sato, M. Ogawa, T. Udou, and Y. Mizuguchi. 1984. Isolation and characterization of a filamentous phage, Vf33, specific for Vibrio parahaemolyticus. Microbiol Immunol. 28:327-337. Tanil, G. B., S. Radu, M. Nisbibuchi, R. A. Rahim, S. Napis, L. Maurice, and J. W. Gunsalam. 2005. Characterization of Vibrio parahaemolyticus isolated from coastal seawater in peninsular Malaysia. Southeast Asian J Trop Med Public Health 36:940-945. Tarr, C. L., J. S. Patel, N. D. Puhr, E. G. Sowers, C. A. Bopp, and N. A. Strockbine. 2007. Identification of Vibrio isolates by a multiplex PCR assay and rpoB sequence determination. J Clin Micro bioi. 45: 134-140. 319 Tatusova, A.T. and T. L. Madden. 1999. Blast 2 sequences- a new tool for comparing protein and nucleotide sequences. FEMS Microbial Lett. 174:247-250 Tenover, F. C., R. D. Arbeit, and R. V. Goering. 1997. How to select and interpret molecular strain typing methods for epidemiological studies of bacterial infections: a review for healthcare epidemiologists. Molecular Typing Working Group of the Society for Healthcare Epidemiology of America. Infect Control Hasp Epidemiol. 18:426-439. Tenover, F. C., R. D. Arbeit, R. V. Goering, P.A. Mickelsen, B. E. Murray, D. H. Persing and B. Swaminathan. 1995. Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J Clin Microbial. 33:2233-2239. Terai, A., K. Baba, H. Sbirai, 0. Yoshida, Y. Takeda, and M. Nisbibuchi. 1991. Evidence for insertion sequence-mediated spread of the thermostable direct hemolysin gene among Vibrio species. J Bacterial. 173:5036-5046 Terai, A., H. Sbirai, 0. Yoshida, Y. Takeda, and M. Nisbibuchi. 1990. Nucleotide sequence of the thermostable direct hemolysin gene (tdh gene) of Vibrio mimicus and its evolutionary relationship with the tdh genes of Vibrio parahaemolyticus. FEMS Microbial Lett. 59:319-323. Tbomton, V., W. Hazell, and G. Simmons. 2002. Acute gastroenteritis associated with seafood privately imported from the Pacific Islands. NZ Med J. 115:234-236. Tilsala-Timisjarvi, A. and T. Alatossava. 2001. Characterization of the 16S-23S and 23S-5S rRNA intergenic spacer regions of dairy propionibacteria and their identification with species-specific primers by PCR. lnt J Food Microbial. 68:45-52. Twedt, R. M. and R. M. Novelli. 1971. Modified selective and differential isolation medium for Vibrio parahaemolyticus. Appl Microbiol. 22:593-599. Twedt, R.M., P.L. Spaulding and H.E. Hall. 1969. Morphological, cultural, biochemical, and serological comparison of Japanese strains of Vibrio parahemolyticus with related cultures isolated in the United States. J Bacterial. 98:511-518. Van Belkum, A. 1999. Short sequence repeats in microbial pathogenesis and evolution. Cell Mol Life Sci. 56:729-734 Van Belkum, A. 2007. Tracing isolates of bacterial species by multilocus variable number of tandem repeat analysis (MLV A). FEMS Immunol Med Micro bioi. 49:22-27. Van Belkum, A., S. Scherer, L. van Alphen and H. Verbrugh. 1998. Short-sequence DNA repeats in prokaryotic genomes.Microbiol Mol Bioi Rev. 62:275-293 Van Leeuwen, W., H. Verbrugh, J. van der Velden, N. van Leeuwen, M. Heck and A. van Belkum. 1999. Validation of binary typing for Staphylococcus aureus strains. J Cl in Microbial. 37:664-674. Varettas, K., C. Mukerjee and M. Schmidt. 1995. A comparative study of the BBL crystal enteric/nonfermenter identification system and the biomerieux API20E and API20NE identification systems after overnight incubation. Pathology. 27:358-61. 320 Vattakaven, T., P. Bond, G. Bradley and C. B. Munn. 2006. Differential effects of temperature and starvation on induction of the viable-but-nonculturable state in the coral pathogens Vibrio shiloi and Vibrio tasmaniensis. Appl Environ Microbio1.72:6508-6513. Venkateswaran, K., N. Dohmoto, and S. Harayama. 1998. Cloning and nucleotide sequence of the gyrB gene of Vibrio parahaemolyticus and its application in detection of this pathogen in shrimp. Appl EnvironMicrobiol. 64:681-687. Versalovic, J., V. Kapur, T. Koeuth, G.H. Mazurek, T.S.Whittam, J.M. Musser and J.R. Lupski.l995. DNA fingerprinting of pathogenic bacteria by fluorophore-enhanced repetitive sequence-based polymerase chain reaction. Arch Pathol Lab Med. 119:23-9 Wagley, S., K. Koofhethile, J. B. Wing and R. Rangdale. 2008.Comparison of V. parahaemolyticus isolated from seafoods and cases of gastrointestinal disease in the UK. Int J Environ Health Res. 18:283-93. Waldor, M. K., and J.J. Mekalanos. 1996. Lysogenic conversion by a filamentous phage encoding cholera toxin. Science. 272: 1910-1914 Ward, L. N. and A. K. Bej. 2006. Detection of Vibrio parahaemolyticus in shellfish by use of multiplexed real-time PCR with TaqMan fluorescent probes. Appl Environ Microbiol. 72:2031-2042. Weissman, S. J., S. L. Moseley, D. E. Dykhuizen and E. V. Sokurenko. 2003. Enterobacterial adhesins and the case for studying SNPs in bacteria. Trends in Microbiology. 11:115-117 Welch, R. A., V. Burland, G. Plunkett. 3•d, P. Redford, P. Roesch, D. Rasko, E. L. Buckles, S. R. Liou, A. Boutin & other authors. 2002. Extensive mosaic structure revealed by the complete genome sequence of uropathogenic Escherichia coli. Proc Natl Acad Sci U S A. 99: 17020-17024 Welsh, J. and M. McClelland. 1990. Fingerprinting genomes using PCR with arbitrary primers. Nucleic Acids Res. 18:7213-7218. Welsh, J. and M. McClelland. 1991 a. Genomic fingerprints produced by PCR with consensus tRNA gene primers. Nucleic Acids Res. 19:861-866. Welsh, J. and M. McClelland. 1991 b. Genomic fingerprinting using arbitrarily primed PCR and a matrix of pairwise combinations of primers. Nucleic Acids Res. 19:5275-5279. Welsh, J. and M. McClelland. 1992. PCR-amplified length polymorphisms in tRNA intergenic spacers for categorizing staphylococci. Mol Microbiol. 6:1673-1680. Weng, J. H., H. C. Wong, C. C. Liu, S. Y. Lee, C. M. Yu, Y. S. Lee, H. C. Tseng, and K. Y. Lu. 1995. Characterization of the manganese-resistant mutants derived from Vibrio parahaemolyticus. Zhonghua Min GuoWei Sheng Wu Ji MianYi Xue Za Zhi. 28:256-269. Werner, G., R. J. Willems, B. Hildebrandt, I. Klare and W. Witte. 2003. Influence of transferable genetic determinants on the outcome of typing methods commonly used for Enterococcusfaecium. J Clin Microbiol. 41:1499-1506. West, P.A., E. Russek, P. R. Brayton, and R. R. Colwell. 1982. Statistical evaluation of a quality control method for isolation of pathogenic Vibrio species on selected thiosulfate citrate-bile salts-sucrose agars. J Cl in Microbiol. 16: Ill 0-1116. 321 Williams, J. G., A. R. Kubelik, K. J. Livak, J. A. Rafalski and S. V. Tingey. 1990 DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Res.18:6531-6535. Winstanley, C. and C. A. Hart. 200 I. Type Ill secretion systems and pathogenicity islands. J Med Microbiol. 50:116-126 Withers, M., L. Wernisch and M. dos Reis. 2006. Archaeology and evolution of transfer RNA genes in the Escherichia coli genome. RNA. 12:933-942. Wojciech, L., Z. Staroniewicz, A. Jakubczak, and M. Ugorski. 2004. Typing of Yersinia Enterocolitica Isolates by ITS profiling, REP- and ERIC-PCR. J Vet Med B Infect Dis Vet Public Health 51:238-244. Wong, H. C. and C. N. Chang. 2005a. Hydrophobicity, cell adherence, cytotoxicity, and enterotoxigenicity of starved Vibrio parahaemolyticus. J Food Prot. 68:154-156. Wong, H. C., C. H. Chen, Y. J. Chung, S. H. Liu, T. K. Wang, C. L. Lee, C. S. Chiou, M. Nishibuchi, and B. K. Lee. 2005b. Characterization of new 03:K6 strains and phylogenetically related strains of Vibrio parahaemolyticus isolated in Taiwan and other countries. J Appl Microbiol. 98:572-580. Wong, H. C., M. C. Chen, S. H. Liu, and D. P. Liu. 1999a. Incidence of highly genetically diversified Vibrio parahaemolyticus in seafood imported from Asian countries. Int J Food Microbiol. 52:181-188. Wong, H. C., C. Y. Ho, L. P. Kuo, T. K. Wang, C. L. Lee, and D. Y. Sbih. 1999b. Ribotyping of Vibrio parahaemolyticus isolates obtained from food poisoning outbreaks in Taiwan. Microbiol Immunol. 43:631-636. Wong, H. C. and C. H. Lin. 2001. Evaluation of typing of Vibrio parahaemolyticus by three PCR methods using specific primers. J Clin Micro bioi. 39:4233-4240. Wong, H. C., S. H. Liu, C. S. Chiou, M. Nishibucbi, B. K. Lee, 0. Suthienkul, G. B. Nair, C. A. Kaysner, and H. Taniguchi. 2007. A pulsed-field gel electrophoresis typing scheme for Vibrio parahaemolyticus isolates from fifteen countries. Int J Food Microbiol. 114:280-287. Wong, H. C., C. C. Liu, T. M. Pan, T. K. Wang, C. L. Lee, and D. Y. Shih. 1999c. Molecular typing of Vibrio parahaemolyticus isolates, obtained from patients involved in food poisoning outbreaks in Taiwan, by random amplified polymorphic DNA analysis. J Cl in Micro bioi. 37:1809-1812. Wong, H. C., S. H. Liu, L. W. Ku, I. Y. Lee, T. K. Wang, Y. S. Lee, C. L. Lee, L. P. Kuo, and D. Y. Shih. 2000. Characterization of Vibrio parahaemolyticus isolates obtained from foodbome illness outbreaks during 1992 through 1995 in Taiwan. J Food Prot. 63:900-906. Wong, H. C., K. T. Lu, T. M. Pan, C. L. Lee, and D. Y. Shih. 1996. Subspecies typing of Vibrio parahaemolyticus by pulsed-field gel electrophoresis. J Cl in Microbiol. 34:1535- 1539. 322 Wong, H.C., P. Wang. 2004. Induction of viable but nonculturable state in Vibrio parahaemo/yticus and its susceptibility to environmental stresses. J Appl Microbiol. 96:359-366 Xie, Z. Y., C. Q. Hu, C. Chen, L. P. Zhang, and C. H. Ren. 2005.1nvestigation of seven Vibrio virulence genes among Vibrio alginolyticus and Vibrio parahaemolyticus strains from the coastal mariculture systems in Guangdong, China. Lett Appl Microbiol. 41:202- 207. Xu, M., T. lida, K. Yamamoto, Y. Takarada, T. Miwatani, and T. Honda. 1994. Demonstration and characterization of simultaneous production of a thermostable direct hemolysin (TDH/1) and a TDH-related hemolysin (TRHx) by a clinically isolated Vibrio parahaemolyticus strain, TH3766. Infect Immun. 62:166-171. Yakubu, D.E., F. J.Abadi and T. H. Pennington. 1999. Molecular typing methods for Neisseria meningitidis. J Med Micro bioi. 48: I 055-1064. Yamaicbi, Y., T. Iida, K. S. Park, K. Yamamoto and T. Honda. 1999. Physical and genetic map of the genome of Vibrio parahaemolyticus: presence of two chromosomes in Vibrio species. Mol Microbiol. 31:1513-1521. Yamamoto, K., T. Honda, T. Miwatani, S. Tamatsukuri, and S. Sbibata. 1992a. Enzyme-labeled oligonucleotide probes for detection of the genes for thermostable direct hemolysin {TDH) and TDH-related hemolysin (TRH) of Vibrio parahaemolyticus. Can J Microbiol. 38:410-416. Yamamoto, S., T. Akiyama, N. Okujo, S. Matsu-ura, and S. Shinoda. 1995. Demonstration of a ferric vibrioferrin-binding protein in the outer membrane of Vibrio parahaemolyticus. Microbiollmmunol. 39:759-766. Yamamoto, S., Y. Fujita, N. Okujo, C. Minami, S. Matsuura, and S. Sbinoda. 1992b. Isolation and partial characterization of a compound with siderophore activity from Vibrio parahaemolyticus. FEMS Microbiol Lett. 73:181-186. Yamamoto, S., N. Okujo, S. Matsuura, I. Fujiwara, Y. Fujita, and S. Shinoda. 1994. Siderophore-mediated utilization of iron bound to transferrin by Vibrio parahaemolyticus. Microbiol Immunol. 38:687-693. Yamamoto, S., N. Okujo, S. Miyosbi, S. Sbinoda, and S. Narimatsu. 1999. Siderophore production of Vibrio parahaemolyticus strains from different sources. Microbiol Immunol. 43:909-912. Yano, Y., M. Kaneniwa, M. Satomi, H. Oikawa, and S. S. Chen. 2006. Occurrence and density of Vibrio parahaemo/yticus in live edible crustaceans from markets in China. J Food Prot. 69:2742-2746. Yeung, P.S. and B.J. Boor. 2004. Epidemiology, pathogenesis, and prevention of foodborne Vibrio parahaemolyticus infections. Foodborne Pathog Dis. 1:74-88 Yeung, P. S., M. C. Hayes, A. Depaola, C. A. Kaysner, L. Kornstein, and K. J. Boor. 2002. Comparative phenotypic, molecular, and virulence characterization of Vibrio parahaemolyticus 03:K6 isolates. Appl Environ Microbiol. 68:2901-2909. 323 Yob, M., T. Honda, T. Miwatani. 1986. Purification and partial characterization of a non- 01 Vibrio cholerae hemolysin that cross-reacts with thermostable direct hemolysin of Vibrio parahaemolyticus. Infect Immun. 52:319-322. Yoh, M., T. Honda, and T. Miwatani. 1989. Homogeneity of a hemo1ysin(Vh-rTDH) produced by clinical and environmental isolates of Vibrio hollisae. FEMS Microbiol Let!. 52:171-175. Yoh, M., N. Kawakami, Y. Funakoshi, K. Okada, and T. Honda. 1995. Evaluation of two assay kits for thermostable direct hemolysin (TDH) as an indicator ofTDH-related hemolysin (TRH) produced by Vibrio parahaemolyticus. Microbiol Immunol. 39:157-159. Yoh, M., T. Miwatani, and T. Honda. 1992. Comparison of Vibrio parahaemolyticus hemolysin (Vp-TRH) produced by environmental and clinical isolates. FEMS Microbiol Lett. 71: 157-161. Yoon, J. H., S. T. Lee, S. B. Kim, M. Goodfellow, and Y. H. Park. 1997. Inter- and intraspecific genetic analysis of the genus Saccharomonospora with 16S to 23S ribosomal DNA (rDNA) and 23S to SS rDNA internally transcribed spacer sequences. lnt J Syst Bacteriol. 47:661-669. Zen-Yoji, H., S. Sakai, Y. Kudoh, T.ltoh, and T. Terayama. 1970. Antigenic schema and epidemiology of Vibrio parahaemolyticus. Health Lab Sci. 7:100-108. Zimmerman, A. M., A. DePaola, J. C. Bowers, J. A. Krantz, J. L. Nordstrom, C. N. Johnson and D. J. Grimes. 2007. Variability of total and pathogenic Vibrio parahaemolyticus densities in northern Gulf of Mexico water and oysters. Appl Environ Microbiol. 73:7589-7596. Zscbock, M., S. Manhold-Maurer, A. Wescher, K. Merl, I. Khan and C. Liimmler. 2005. Evaluation oftRNA intergenic spacer length polymorphism analysis as a molecular method for species identification of streptococcal isolates from bovine mastitis. J Dairy Res. 72:333-337. 324 This copy ofthe thesis has 'been< supplied on condiiion thai anyone ,who! consults 'it is understood: to ~recognise thai iis copyright 'res is with iis, author and that no:quotdtion from the thesis and no·.in/ormation.derived from' it: may be publishedwithouf, the author's,prior ,li~J;itten'consent. . I I I f I, r ,J,,'