Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Microbiology (2000), 146, 999–1009 Printed in Great Britain Molecular analysis of genetic differences between virulent and avirulent strains of Aeromonas hydrophila isolated from diseased fish Y. L. Zhang, C. T. Ong and K. Y. Leung Author for correspondence : K. Y. Leung. Tel : j65 8747835. Fax : j65 7792486. e-mail : dbslky!nus.edu.sg Department of Biological Sciences, Faculty of Science, National University of Singapore, 10 Kent Ridge Crescent, Singapore 119260 Aeromonas hydrophila, a normal inhabitant of aquatic environments, is an opportunistic pathogen of a variety of aquatic and terrestrial animals, including humans. A. hydrophila PPD134/91 is defined as virulent whereas PPD35/85 is defined as avirulent on the basis of their different LD 50 values in fish. Suppression subtractive hybridization (SSH) was used to identify genetic differences between these two strains. Sixty-nine genomic regions of differences were absent in PPD35/85, and the DNA sequences of these regions were determined. Sixteen ORFs encoded by 23 fragments showed high homology to known proteins of other bacteria. ORFs encoded by the remaining 46 fragments were identified as new proteins of A. hydrophila, showing no significant homology to any known proteins. Among these PPD134/91-specific genes, 22 DNA fragments (21 ORFs) were present in most of the eight virulent strains studied but mostly absent in the seven avirulent strains, suggesting that they are universal virulence genes in A. hydrophila. The PPD134/91-specific genes included five known virulence factors of A. hydrophila : haemolysin (hlyA), protease (oligopeptidase A), outer-membrane protein (Omp), multidrug-resistance protein and histone-like protein (HU-2). Another 47 DNA fragments (44 ORFs) were mainly present in PPD134/91, indicating the heterogeneity among motile aeromonads. Some of these fragments encoded virulence determinants. These included genes for the synthesis of O-antigen and type II restriction/modification system. The results indicated that SSH is successful in identifying genetic differences and virulence genes among different strains of A. hydrophila. Keywords : Aeromonas hydrophila, genomic subtraction, virulence genes INTRODUCTION Aeromonas hydrophila is the causative agent of motile aeromonad septicaemia, which occurs in a wide variety of freshwater fish species (Thune et al., 1993 ; Austin & Adams, 1996). Outbreaks usually occur only when the fish are immunocompromised by stresses such as overcrowding or concurrent disease (Stevenson, 1988). The pathogenesis of A. hydrophila is multifactorial. A. hydrophila produces several toxins including haemolysins and enterotoxin (Chakraborty et al., 1984 ; Howard ................................................................................................................................................. Abbreviations : R/M, restriction/modification ; SSH, suppression subtractive hybridization. The GenBank accession numbers for the sequences determined in this work are given in Table 5. et al., 1996), and a repertoire of enzymes which digest cellular components, such as proteases, amylases and lipases (Leung & Stevenson, 1988a ; Pemberton et al., 1997). Other virulence factors such as the S layer (Dooley & Trust, 1988), ability to internalize (Tan et al., 1998), serum resistance (Mittal et al., 1980 ; Janda et al., 1984) and resistance to phagocyte-mediated killing (Leung et al., 1995a) are also implicated in the resistance of A. hydrophila to the host’s non-specific immune defences. A. hydrophila strains PPD134\91 and PPD35\85 were isolated from our local environment and used in our previous studies. A. hydrophila PPD134\91 was defined as virulent whereas PPD35\85 was defined as avirulent on the basis of their differences in LD values in tilapia &! PPD134\91 can and blue gourami (Leung et al., 1995b). survive and proliferate in tilapia serum and phagocytes 0002-3723 # 2000 SGM 999 Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Wed, 09 Aug 2017 20:51:14 Y. L. Z H A N G, C. T. O N G a n d K. Y. L E U N G (Leung et al., 1995a, b), and it can internalize and induce severe morphological changes in EPC (epithelioma papillosum of carp) cells (Tan et al., 1998). On the other hand, A. hydrophila PPD35\85 is serum-sensitive, unable to invade or produce cytotoxic effects in EPC cells, and is incapable of surviving in phagocytes after opsonization. It is important to determine whether there are any special virulence genes present in A. hydrophila PPD134\91 but not in PPD35\85. Subtractive hybridization has been used to identify sequences that are present in one genome but absent in another (Straus & Ausubel, 1990 ; Mahairas et al., 1996). The analysis of the differences between two complex genomes holds promise for the discovery of unknown virulenceassociated factors and probes useful for genetic studies (Lisitsyn et al., 1993 ; Quinn et al., 1997). Traditional subtractive hybridization methods involved several rounds of hybridization and physical separation of single-stranded and double-stranded DNA. Recently a new technique called suppression subtractive hybridization (SSH) overcame some of these limitations. The step of suppression PCR can prevent undesirable amplification while enrichment of target molecules proceeds (Diatchenko et al., 1996 ; Gurskaya et al., 1996). The objective of this study was to identify genetic differences between virulent and avirulent strains of A. hydrophila. Sixty-nine subtracted genomic regions unique to PPD134\91 were identified. Among them, 22 fragments (21 ORFs) were present in most of the virulent strains. These presumptive virulence genes were identified, sequenced, and their biological functions analysed. This will form a foundation for further studies in elucidating how A. hydrophila causes disease in humans and fish. METHODS Bacterial strains, plasmids, and growth conditions. The bacterial strains and plasmids used in this study are listed in Table 1. The characteristics and virulence of A. hydrophila strains studied were determined previously (Mittal et al., 1980 ; Leung & Stevenson, 1988b ; Leung et al., 1995b). Virulent strains were defined as having a lower LD value in blue gourami or rainbow trout ( 10'n&) than the&!avirulent strains ( 10(n&). A. hydrophila strains were maintained on tryptic soy agar (TSA) (Difco) or in tryptic soy broth (TSB) (Difco) at 25 mC. Escherichia coli strains were maintained on L agar (Difco) or in Luria broth (LB) (Difco) at 37 mC. As necessary, media were supplemented with antibiotics (Sigma) at the following concentrations : ampicillin at 50 µg ml−" and kanamycin at 50 µg ml−". Bacteria were stored as frozen cultures at k80 mC in either TSB or LB containing 25 % (v\v) glycerol. DNA manipulations. Bacterial genomic DNA was extracted according to the manual of the QIAGEN Genomic DNA Kit and the Wizard Genomic DNA Purification Kit (Promega). Plasmid DNA was prepared by using Qiagen and Promega columns. Restriction endonuclease digestion was accomplished by standard methods (Sambrook et al., 1989). subtractive hybridization (SSH). Bacterial genome subtraction was performed following the user manual of the PCR-Select Bacterial Genome Subtraction Kit (Clontech). Briefly, the tester (PPD134\91) and driver (PPD35\85) genomic DNAs were digested with RsaI. The tester DNA was then subdivided into two portions, each of which was ligated with a different adaptor provided by the subtraction kit. Two hybridizations were performed. In the first, an excess of driver was added to each adaptor-ligated tester sample. The samples were then heat-denatured and allowed to anneal. In the second hybridization, the two primary hybridization samples were mixed together without denaturing. The entire population of molecules was then subjected to PCR to amplify the tester-specific sequences. The PCR amplification product was cloned into pT-Adv and transformed into E. coli TOP10Fh competent cells according to the manual of the Advantage PCR Cloning Kit (Clontech). Positive clones were screened on LB medium supplemented with X-Gal (Sigma), IPTG (Sigma) and ampicillin. Southern hybridization. Southern blots were performed to identify the subtractive clones that contained PPD134\91 unique fragments. At the same time, similar probes were used to screen genomic DNA of several virulent and avirulent strains of A. hydrophila. Nick-translated RsaI fragments of recombinant plasmids were used as probes to hybridize with RsaI digests of PPD134\91, PPD35\85, and other bacterial genomic DNA. Transfer of DNA to nylon membrane (GeneScreen, NEM Research Products) and hybridization conditions were in accordance with standard methods (Sambrook et al., 1989). Probe DNAs were labelled by nicktranslation with biotin-14-dATP (BioNick Labelling System, Gibco-BRL) and visualized with strepavidin–alkaline phosphate conjugate (BluGene Nonradioactive Nucleic Acid Detection System, Gibco-BRL). DNA sequencing. DNA sequencing was carried out on an Applied Biosystems PRISM 377 automated DNA sequencer by the dye termination method. The ABI PRISM dRhodamine Terminator Cycle Sequencing Ready Reaction Kit was used (Applied Biosystems). The sequences were edited by using the manufacturer’s software. Sequence assembly and further editing were done with DNA analysis software. and sequence homology analyses were performed by using the National Centre for Biotechnology Information network service. Genome walking. Unknown genomic DNA sequences adjacent to the subtracted DNA fragments (F3, 33, 44, 46, 52, 101, 106, 108, 109 and 121) were identified following the user manual of the Universal GenomeWalker Kit (Clontech). GenomeWalker libraries using five restriction enzymes (DraI, EcoRV, PvuII, ScaI and StuI) were constructed and two PCR amplifications were performed for the DNA walking. Specific primers were synthesized by Gibco-BRL for primary (P) and nested (N) PCR as well as for right (R) and left (L) sides of each subtracted fragment. PCR was performed using an Advantage Tth Polymerase Mix purchased from Clontech and following two-step cycle parameters : 7 cycles of 25 s at 94 mC, 3 min at 72 mC, 32 cycles of 25 s at 94 mC and 3 min at 67 mC. The oligonucleotides for genome walking were as follows : F3LP, 5h-GCATCCCCGTTTCGCATTATC-3h, F3-LN, 5h-CCGTTTCGCATTATCTGAAC-3h, F3-RP, 5h-GATAATGCGAAACGGGGATGC-3h, F3-RN, 5h-CGGGGATGCCACGGCATCC-3h ; F33-LP, 5h-ATTTGATGCGCTTTTGTCCC-3h, F33-LN, 5h-CGCTTTTGTCCCATTGACAG-3h, F33-RP, 5hTCGTTGCTTTTGGGTTACCAAG-3h, F33-RN, 5h-GGGTTACCAAGATACTACGTTC-3h ; F44-LP, 5h-ACACTCCCACGTCGTTTTAC-3h, F44-LN, 5h-GCTGTTTTACTCASuppression 1000 Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Wed, 09 Aug 2017 20:51:14 Screening virulence genes of A. hydrophila Table 1. Bacterial strains and vectors used in this study Strain or plasmid Genotype and/or clone description* Source† A. hydrophila ATCC 7966 L15 L31 L36 L37 PPD35\85 PPD11\90 PPD64\90 PPD88\90 PPD45\91 PPD70\91 PPD122\91 PPD134\91 TF7 Xs91\4\1 Wild-type, virulent, type strain Wild-type, avirulent Wild-type, virulent Wild-type, avirulent Wild-type, avirulent Wild-type, avirulent Wild-type, virulent Wild-type, avirulent Wild-type, avirulent Wild-type, avirulent Wild-type, virulent Wild-type, virulent Wild-type, virulent Wild-type, virulent Wild-type, virulent ATCC BAU, Indonesia BAU, Indonesia BAU, Indonesia BAU, Indonesia PPD, Singapore PPD, Singapore PPD, Singapore PPD, Singapore PPD, Singapore PPD, Singapore PPD, Singapore PPD, Singapore UG, Canada IHB, China E. coli TOP10Fh Amps Kans Clontech Subtracted fragments cloned into the RsaI sites of the pT-Adv vector Ampr Kanr LacZ This study Plasmids pSC1–129‡ pT-Adv Vector Clontech * Virulent strains were defined as having a lower LD value in blue gourami or rainbow trout ( 10'n&) &! than the avirulent strains (10(n&). † ATCC, American Type Culture Collection ; BAU, Bogor Agricultural University of Indonesia ; IHB, Institute of Hydrobiology, China ; PPD, Primary Production Department of Singapore ; UG, University of Guelph, Canada. ‡ The 14 empty clones or false positives were as follows : pSC4, 5, 19, 24, 27, 41, 67, 113, 115, 116, 118, 122, 125 and 127. TAGCTAC-3h, F44-RP, 5h-GCTTGCATCGTTATGCGTCTGT-3h, F44-RN, 5h-GCGTCTGTAGCTATCAATGTG-3h ; F46-RP, 5h-GCAGGTGGGGAAATCGATGAAC-3h, F46RN, 5h-TCGATGAACTCAAGGCGGT-3h ; F52-LP, 5h-TGGAGTGATGTCGCGTGCGG-3h, F52-RP, 5h-GTAATCCCCAAAACCCG-3h ; F101-LP, 5h-CATCATCGTCAGAAAATGCG-3h, F101-LN, 5h-AGAAAATGCGTTATTTCTAC-3h, F101-RP, 5h-CTAAGATTGTGTCTGCGAGTG-3h, F101RN, 5h-GTCTGCGAGTGATAAACAAAAG-3h ; F106-LP, 5hTGC TCT CAT TGC TGG GGG GC-3h, F106-LN, 5hGCTGGGGGGCACCTTGTCCAC-3h, F106-RP, 5h-ACATCTAGCGCACGAGATAAAT-3h, F106-RN, 5h-CACGAGATAAATCAGGCCCAAGC-3h ; F108-LP, 5h-AGACAAGCAGAATAACGCCCCG-3h, F108-LN, 5h-AATAACGCCCCGAAATATAACCG-3h, F108-RP, 5h-CAGCGGATTGGCGAAGGTATT-3h, F108-RN, 5h-TTGGCGAAGGTATTTATGTTG-3h ; F109-LP, 5h-TCGTCCTTATTTCGGGTAGGGATCAAGCGG-3h, F109-RP, 5h-CTCCTTGGAAGGTAGACCCCGAACTCTACT-3h ; F121-LP, 5h-GCCCATAGCATCCACATCGG-3h, F121-LN, 5h-TCCACATCGGCCGTATATTC-3h, F121-RP, 5h-GATTACCAGAGGTTTGGCCAAT-3h, F121-RN, 5h-GAGGTTTGGCCAATATTGCCC-3h. Nucleotide sequence accession numbers. Sixteen ORFs, which derived from 23 subtracted fragments, had homologues with known sequences in GenBank and were assigned with 16 accession numbers (Table 5). Seven of these (F3, 33, 46, 52, 101, 106 and 121) represented complete ORFs via genome walking while nine of them represented partial ORFs. RESULTS Genomic subtraction between A. hydrophila PPD134/91 and PPD35/85 SSH was carried out between the RsaI-digested genomic DNA of virulent strain PPD134\91 (tester) and avirulent strain PPD35\85 (driver) with the aim of isolating PPD134\91-specific genes (virulence genes). Subtracted clones [E. coli TOP10Fh(pSCT)] and the DNA fragments (FT) were given a number (represented here by T) for identification (Table 1). One hundred and twenty-nine clones were obtained from one round of subtraction and 14 of these were found to have no inserts (Table 1). Therefore, 115 were confirmed as positive clones (Table 2). Southern analysis of subtracted clones Plasmid DNA was extracted from these subtracted clones and was digested by RsaI. The molecular masses of subtracted fragments were measured and divided into 1001 Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Wed, 09 Aug 2017 20:51:14 Y. L. Z H A N G, C. T. O N G a n d K. Y. L E U N G Table 2. Summary of subtraction and the use of subtracted fragments in Southern hybridization analysis between A. hydrophila PPD134/91 (tester) and PPD35/85 (driver) Size of subtracted genomic fragments (kb) No. of fragments 0n18–0n68 0n70–1n4 Total Total Tester only* Testerdriver† False positive‡ 90 25 52 17 3 1 35 7 115 69 (60 %) 4 (3n5 %) 42 (36n5 %) * Subtracted fragments that hybridized only to A. hydrophila PPD134\91. Refer to Tables 3 and 4 for fragment (F) numbers in this group. When using F51, 60 and 111 as probes for hybridization, two bands were found on tester genomic DNA. † Subtracted fragments that hybridized to both A. hydrophila PPD134\91 and PPD35\85. More bands were found in tester than driver. Fragment numbers : F14, 76, 95 and 105. ‡ Subtracted fragments that hybridized to both A. hydrophila PPD134\91 and PPD35\85. One band was found in both tester and driver. Fragment numbers : F6, 7, 9, 10, 12, 13, 15, 22, 23, 26, 28, 29, 37, 39, 40, 42, 43, 45, 47, 48, 53, 54, 56, 57, 62, 63, 64, 65, 68, 69, 71, 74, 78, 96, 98, 102, 103, 117, 120, 124, 128 and 129. 33 T 66 D T 51 D T 60 D T 95 D T 105 D T 10 D T 40 D T D ................................................................................................................................................................................................................................................................................................................. Fig. 1. Southern hybridization analysis of subtracted fragments on RsaI genomic digest of A. hydrophila PPD134/91 and PPD35/85. A. hydrophila PPD134/91 (tester strain) (T) and PPD35/85 (driver strain) (D) were hybridized with biotin-labelled subtracted fragments. Fragment designations are shown at the top. two groups : less than 0n69 kb and more than 0n69 kb in size (Table 2). Their RsaI digests were biotinylated and hybridized to RsaI-digested tester and driver. These 115 clones could be divided into three groups according to the results of our Southern analysis (Table 2). In the first group (tester only), subtracted fragments only hybridized to the tester genome digest. There were 69 clones (60 % of the total population studied) in this group. Examples of this group included F33, 51, 60 and 66 (Fig. 1). In F51, 60 (Fig. 1) and 111 (data not shown), double bands were detected on the tester genomic DNA. Fragments in the second group (tester driver) could hybridize to both tester and driver. However, multiple bands were observed during hybridization with the tester whereas only one or two bands were found when hybridized with the driver (Fig. 1). Examples of this group included F14, 76, 95 and 105 and they represented 3n5 % of the total population. These different hybridization patterns may reflect multiple copies of a unique gene present in PPD134\91 or restriction fragment polymorphism. The bands formed on both tester and driver genomic DNA could be of either the same size (F105, Fig. 1) or different sizes (F95, Fig. 1). Inserts in the last group (false positive) hybridized to both tester and driver, and had the same hybridization patterns (F10 and 40 ; Fig. 1). These clones were false 1002 Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Wed, 09 Aug 2017 20:51:14 Screening virulence genes of A. hydrophila A B C 109 D E I J K L M N F G H O A B C 101 D E F G 121 I J K L M N O H 32 A B C D E F I J K L M N G H O A B C I J K D E F G L M N O H ................................................................................................................................................................................................................................................................................................................. Fig. 2. Southern hybridization analysis of subtracted fragments on RsaI genomic digest of virulent and avirulent strains of A. hydrophila. Virulent strains PPD122/91, PPD70/91, PPD11/90, PPD134/91, L31, Xs91/4/1, TF7 and ATCC 7966 (lanes A to H), and avirulent strains PPD35/85, PPD45/91, PPD64/90, PPD88/90, L15, L36 and L37 (lanes I to O) were hybridized with biotin-labelled subtracted fragments F109, F101, F121 and F32, as indicated at the top. Table 3. Distribution of group 1 PPD134/91 (tester)-specific DNA fragments among virulent and avirulent strains of A. hydrophila ................................................................................................................................................................................................................................................................................................................. Hybridization results were scored as : j, strains hybridized with PPD134\91-specific DNA fragments ; k, strains not hybridized with PPD134\91-specific DNA fragments. Fragment no. … 61 Virulent strains ATCC 7966 L31 PPD11\90 PPD70\91 PPD134\91 PPD122\91 TF7 Xs91-4-1 Homology\total† Avirulent strains L15 L36 L37 PPD35\85 PPD45\91 PPD64\90 PPD88\90 Homology\total* 58/72 88/89/109 93 34 52/108 8 99/106 20/97 92 32 85 87 11 2/3 j j j j j j j j 8\8 j j j j j j j j 8\8 j j j j j j j j 8\8 j j j j j j j j 8\8 j j j j j j k j 7\8 j j j j j j k j 7\8 j j j k j j k j 6\8 j j j k j j k k 5\8 k j j k j j k k 4\8 k j j k j k k j 4\8 k j j k j j k k 4\8 k k k j j j k j 4\8 j j k j j k k k 4\8 j k j k j k k j 4\8 k j j j k k k j 4\8 k j k k k j k 2\7 k j k k k j j 3\7 k j k k k j j 3\7 k j k k j j j 4\7 k k k k k j k 1\7 k j k k k j j 3\7 j k k k k j k 2\7 k j k k k k k 1\7 k k k k k k k 0\7 k k k k k k k 0\7 k k k k k j k 1\7 k k k k k k j 1\7 k j k k k k k 1\7 k j k k k j k 2\7 k k j k k j j 3\7 * No. of hybridized strains\total no. of strains studied. positive, which is a result of the inefficiency of subtraction hybridization. Our Southern analysis indicated that we achieved a high level of enrichment of PPD134\ 91-specific DNA whereby 60 % of our subtracted clones were present only in the tester but not in the driver (Table 2). In the tester-only group, 75 % of the subtracted fragments were less than 0n69 kb. Identification of common virulence genes in A. hydrophila Southern hybridization was carried out to survey the distribution of these 69 PPD134\91-specific fragments among eight virulent and seven avirulent strains of A. hydrophila. Fig. 2 and Tables 3 and 4 show the 1003 Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Wed, 09 Aug 2017 20:51:14 Y. L. Z H A N G, C. T. O N G a n d K. Y. L E U N G Table 4. Distribution of group 2 PPD134/91 (tester)-specific DNA fragments among virulent and avirulent strains of A. hydrophila ................................................................................................................................................................................................................................................................................................................. Hybridization results were scored as : j, strains hybridized with PPD134\91-specific DNA fragments ; k, strains not hybridized with PPD134\91-specific DNA fragments. Fragment no. … Virulent strains ATCC 7966 L31 PPD11\90 PPD70\91 PPD134\91 PPD122\91 TF7 Xs91\4\1 Homology\total† Avirulent strains L15 L36 L37 PPD35\85 PPD45\91 PPD64\90 PPD88\90 Homology\total† A* (24) B* (10) 90 38 66 C* (3) 16 94/104 18 73 110 55 k k k k j k k k 1\8 k k k k j k k k 1\8 k k k k j k k k 1\8 k k k k j k k k 1\8 k k k k j k k k 1\8 k k k k j k k j 2\8 k k k k j j k k 2\8 k j k k j k k k 2\8 k k k k j k k j 2\8 k k j k j k k k 2\8 k k k k j k k j 2\8 k j j k j k k k 3\8 k k k k k k k 0\7 k k k k k j k 1\7 k j k k k k j 1\7 k k k k k j k 2\7 k k k k k j j 2\7 k k k k k k k 0\7 k k k k k k k 0\7 k k k k k k k 0\7 k j k k k k k 1\7 k k k k k j k 1\7 k k k k k j k 1\7 k k k k k k k 0\7 * PPDI34\91-specific DNA fragments that showed a similar hybridization pattern were grouped as follows : group A, F1, 21, 30, 31, 33, 35, 46, 50, 59, 60, 70, 77, 79, 80, 81, 82, 83, 91, 100, 101, 107, 112, 114 and 119 ; group B, F25, 36, 44, 51, 75, 86, 111, 121, 123 and 126 ; group C, F17, 49 and 84. † No. of hybridized strains\total no. of strains studied. hybridization results. The fraction of hybridization was calculated for both virulent and avirulent strains (x\8 for virulent and y\7 for avirulent strains, where x and y are the number of hybridized strains). Two groups were derived based on the hybridization patterns. Group 1 (which represents common virulence genes, x 4 and y 4, n l 22) consisted of subtracted fragments that were highly enriched in other virulent strains of A. hydrophila (Table 3). As seen in F109 and F32, more bands were found on the genomic digest of virulent than avirulent strains (Fig. 2). All eight virulent strains hybridized to F109 while only three of the seven avirulent strains hybridized to it. Subtracted fragments that were tester-specific (i.e. hybridized only to PPD134\91 plus at most two other strains) were classified as group 2 (Table 4, x 3 and y 2, n l 47). In F101 and 121, among all the virulent and avirulent strains, only PPD134\91 and PPD64\90 (for F121 only) hybridized (Fig. 2). Sequence analysis and gene walking The 69 subtracted fragments were sequenced and their DNA sequences were subjected to homology search in PIR\GenBank. Of the 69 subtracted fragments, 46 (66n7 %) demonstrated no significant matches with entries in the databases and potentially represented new and novel virulence determinants in A. hydrophila (data not shown). In the remaining 33n3 % of the clones, 23 DNA fragments representing 16 ORFs showed high homology to known proteins of other bacteria and four identical pairs were found (Table 5). From the 16 ORFs that had homologues, five belong to group 1 (F2\3, 52, 106, 108 and 88\109 ; Table 3), 11 to group 2 (30, 33, 36\121, 44\86, 46, 50, 51, 60\ 91\101\119, 82, 110 and 126 ; Table 4). Full sequences were determined by genome walking for F3, 33, 44, 46, 52, 101, 106 and 121 for detailed analysis (Table 5). Table 5 shows subtracted and full-length fragments that showed high homology to known sequences in GenBank. Five ORFs in group 1 were found to correspond to known bacterial virulence genes of A. hydrophila. These include haemolysin (hlyA), protease (oligopeptidase), histone-like protein (HU-2), outer-membrane protein (Omp) and multidrug-resistance gene. In group 2, five groups of genes were found to represent the heterogeneity of motile aeromonads. These include genes for the synthesis of O-antigen of LPS (phosphomannomutase, rhamnosyl transferase, O-acetyltransferase, O-antigen methyltransferase, GDP-mannose 4,6dehydratase and mannosyltransferase B), type II restriction\modification system (modification methyltransferase of PstI and BsuBI), CII of bacteriophage P4, 1004 Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Wed, 09 Aug 2017 20:51:14 Screening virulence genes of A. hydrophila Table 5. Summary of sequence analysis of PPD134/91-specific DNA fragments Fragment no. ORF size (bp) Start–stop (bp) Predicted protein (aa) 2, 3 1423 227–1249 341 AF146597 30 454 3–452 150 AF146606 33 1858 566–1600 345 AF146603 36, 121 1691 258–1388 377 Accession no. AF146601 44, 86 1452 436–1419 328 AF148126 46 1079 88–972 295 AF146602 50 800 325–798 158 AF146604 51 965 3–398 132 AF146596 52 489 58–351 98 AF146598 60, 91, 101, 119 82 88, 109 1597 895 1791 346–1590 411–893 3–1328 415 AF146609 161 AF146605 442 AF146608 106 1672 267–1493 409 AF146029 108 518 294–518 75 AF146599 110 126 942 934 1–942 262–933 314 224 AF146607 AF146595 Homologies to predicted encoded protein E value† Homologue accession no. Omp of A. salmonicida (OmpAI and OmpAII) (Costello et al., 1996) CII protein of bacteriophage P4 (Ghisotti et al., 1990 ; Halling et al., 1990) Mannosyltransferase B (rfb) of Synechocystis sp. (Kaneko et al., 1996) Mannosyltransferase B (rfb) of E. coli (Sugiyama et al., 1994) Mannosyltransferase B (rfb) of Salmonella choleraesuis (Brown et al., 1992) GDP-mannose 4,6-dehydratase of V. cholerae GDP-mannose 4,6-dehydratase of E. coli Phosphomannomutase (PMM) (rfbK) of Salmonella choleraesuis Rhamnosyltransferase of Streptococcus pneumoniae (Morona et al., 1997) Regulatory protein (repA) in E. coli (Scholz et al., 1989) Regulatory protein (repA) in E. coli (Dorrington & Rawlings, 1990) O-specific LPS biosynthesis of E. coli (O-acetyltransferase, rfb) (Yao & Valvano, 1994) HU alpha gene (NS2 or HU-2) of Serratia marcescens HU alpha gene (NS2 or HU-2) of A. proteolytica (Giladi et al., 1992) Modification methyltransferase of PstI and BsuBI of B. subtilis (Xu et al., 1992) Modification methyltransferase of PstI and BsuBI of R. leguminosarum (Rochepeau et al., 1997) Sensory histidine protein kinase in Calothrix viguieri Oligopeptidase A (opdA) of Salmonella typhimurium (Conlin & Miller, 1992) Oligopeptidase A (opdA) of E. coli (Conlin et al., 1992) Oligopeptidase A (opdA) of H. influenzae (Fleischmann et al., 1995) Multidrug-resistance protein 2 of B. subtilis (Ahmed et al., 1995) Haemolysin (hlyA) of A. hydrophila (Wong et al., 1998) β-Haemolysin of A. salmonicida (Hirono et al., 1992) Glycosyltransferase of V. cholerae (Zhang et al., 1996) Glycosyltransferase of Yersinia enterocolitica (Zhang et al., 1996) Putative O-antigen methyltransferase of Burkholderia pseudomallei (DeShazer et al., 1998) 1ek147 5ek97 2ek06 CAA63036 5ek31 D90901 6ek24 I76776 2ek22 S22619 1ek169 Y07786 1ek166 P32054 1ek110 Q00330 3ek14 2209213 2ek32 P20356 7ek20 A36134 1ek25 P37750 7ek36 P52680 2ek35 P28080 3ek99 P33563 9ek17 X99520 7ek23 Y09899 1ek179 P27237 1ek176 451277 1ek173 P44573 1ek07 P39843 8ek26 U81555 5ek06 X65049 4ek42 AB012956 1ek22 U46859 0n002 P13059 AF064070 * The E value indicates the probability of the match. A match with an E value of 1ek5 and below was taken to be significant (i.e. the match is not due to chance). An E value of 0n002 (for F126) was also taken into consideration. 1005 Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Wed, 09 Aug 2017 20:51:14 Y. L. Z H A N G, C. T. O N G a n d K. Y. L E U N G histidine protein kinase and DNA regulatory protein (repA). DISCUSSION A prerequisite to comprehending the pathogenicity mechanisms of an organism is the identification and examination of all its virulence genes (Strauss & Falkow, 1997). There are many general procedures to detect virulence genes such as genomic subtraction (Diatchenko et al., 1996 ; Mahairas et al., 1996), transposon tagging (Berg et al., 1994) and in vivo expression technology (Mahan et al., 1993). By using SSH in this study, 60 % of the clones obtained were tester-specific and 69 PPD134\91-specific fragments were derived from one round of subtraction. Only 36n5 % of the subtracted fragments were false positive. The studies described here show that the SSH technique is simple and efficient for identifying genetic differences between virulent and avirulent strains. This method offers a possible alternative to identify presumptive virulence genes of pathogenic bacteria. Surveying the distribution of virulence genes in A. hydrophila Two strains were chosen for the SSH protocol : A. hydrophila PPD134\91 as the tester strain and PPD35\85 as the driver strain. In order to confirm that subtracted fragments were unique not only in the tester strain but also among other virulent strains, Southern hybridization analysis was used to probe eight virulent and seven avirulent strains of A. hydrophila (Tables 3 and 4). Two groups of clones were derived based on the hybridization patterns. Presumptive universal virulence genes (21 ORFs including five predicted proteins) were classified in group 1. Genes encoding heterogeneity or strain variations (44 ORFs including 11 predicted proteins) are in group 2 and these also include some known virulence genes. Genes in group 2 may have undergone extensive modifications such that they are not even conserved among different strains of the same species. Many of these genes are described for the first time in motile aeromonads. Studying the distribution of these gene fragments provides insight to their relative importance. Group 1 (common virulence genes) Haemolysin (F108). F108 is highly homologous to the hlyA gene of an A. hydrophila strain, A6 (Wong et al., 1998), and a β-haemolysin gene of A. salmonicida (Hirono & Aoki, 1993). Haemolysin is one of the virulence factors produced by motile aeromonads and has been cloned from different strains of aeromonads (Howard et al., 1996 ; Wong et al., 1998). A. hydrophila secretes at least two types of haemolysins. One of them is aerolysin (AerA) which can oligomerize on erythrocyte cell membranes, form channels, and lead to cell lysis (Wilmsen et al., 1990). The second is a non-channelforming haemolysin (HlyA) and is proposed to be a Vibrio cholerae-HlyA-like haemolysin (Wong et al., 1998). These two types of haemolytic toxins are low in homology and believed to be distinct. It was suggested that these two haemolytic toxins contribute to virulence in A. hydrophila and are widespread within virulent strains of motile aeromonads (Vadivelu et al., 1995 ; Hirono et al., 1992). Therefore, it is not a surprise to detect the absence of the hlyA gene in four of our avirulent strains including PPD35\85. In addition, our hybridization results confirm that hlyA is present in most of our virulent strains (7\8). Histone-like protein (F52). F52 is homologous to HU-2 genes of Aeromonas proteolytica (Giladi et al., 1992), Salmonella typhimurium (Higgins et al., 1988) and E. coli (Kano et al., 1987). These are histone-like proteins of prokaryotes (such as HU proteins and integration host factors) that are small, basic, heat-stable and bind to single- and double-stranded DNA (Drlica & Rouviere-Yaniv, 1987). Although the biological functions of histone-like proteins are not fully understood, they are known to alter DNA recognition by changing DNA dynamic flexibility and accessibility (Flashner & Gralla, 1988). The resulting alterations in DNA structure and topology affect several cellular processes, including initiation of DNA replication, DNA partitioning and cell division, and transposition of bacteriophage Mu (Huisman et al., 1989 ; Jaffe et al., 1997). In addition to the physiological functions, secreted histone-like protein has been demonstrated to have a potential role in the pathogenesis of Streptococcus-induced tissue inflammation (Stinson et al., 1998). The biological function of the HU-2 gene in A. hydrophila will be elucidated in future experiments. Protease (oligopeptidase A) (F88, 109). A number of extracellular proteases of A. hydrophila (metallopeptidases and serine peptidases) have been described (Howard et al., 1996) and correlations have been made between the production of proteases and virulence (Leung & Stevenson, 1988b). Extracellular proteases may aid the organism in overcoming initial host defences such as resistance to serum killing, and provide amino acids for cell proliferation (Leung & Stevenson, 1988b). Furthermore, proteases are needed for the maturation of exotoxins such as aerolysin (Howard & Buckley, 1985). For F88 and 109, the predicted protein showed similarity to the OpdA sequence of S. typhimurium, Haemophilus influenzae and E. coli (Conlin & Miller, 1992 ; Conlin et al., 1992 ; Fleischmann et al., 1995). Oligopeptidase A is the major soluble enzyme in E. coli which is able to hydrolyse free lipoprotein signal peptide in vitro (Novak et al., 1986). Fragments F88 and F109, encoding the putative OpdA protein, were present in all the virulent strains of A. hydrophila we tested and may be involved in peptide processing of virulence factors which in turn affect pathogenicity. Omp (F2 and 3). A homologue of the outer-membrane protein OmpA was identified in our tester strain. The OmpA protein is one of the major outer-membrane proteins of a wide range of Gram-negative bacteria such as A. salmonicida (Costello et al., 1996), Shigella dysenteriae (Braun & Cole, 1982) and E. coli (Beck & 1006 Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Wed, 09 Aug 2017 20:51:14 Screening virulence genes of A. hydrophila Bremer, 1980). Major physiological functions include maintenance of the structural integrity and morphology of the cells and porin activity, as well as a role in conjugation and bacteriophage binding. A role in bacterial virulence has been implicated in increased serum resistance as in E. coli (Weiser & Gotschlich, 1991) and Neisseria gonorrhoeae (Rice et al., 1986). Drug-resistance gene (F106). Bacteria have developed a number of mechanisms to protect them from environmental toxins and antibiotics, including degradation and inactivation of drugs by enzymic modifications, alteration of the drug target, and the production of multidrug transporters (Lewis, 1994 ; Nikaido, 1994). The predicted protein of F106 showed similarity to multidrug-resistance protein 2 of Bacillus subtilis (P39843, E l 1ek07) (Ahmed et al., 1995). It is possible that this multidrug transporter in A. hydrophila PPD134\91 and other virulent strains is used for the transport of other antibiotics or toxic substances. This presumptive virulence factor will be studied in future experiments. Group 2 (PPD134/91-specific sequences) O-antigen/polysaccharide (F33, 36/121, 44/86, 46, 51, 110 and 126). LPS is a major component of the outer membrane of Gram-negative bacteria. It consists of three regions : the lipid A, the core oligosaccharide and the O-antigen (Schnaitman & Klena, 1993). Serotypes are distinguished on the basis of O-antigens, and heterogeneity is known to exist among motile aeromonads (MacInnes et al., 1979 ; Leblanc et al., 1981). The unique set of Oantigen modification enzymes (such as rfb) carried by various strains of bacteria contributes to the diversity of O-antigen. A. hydrophila PPD134\91 and PPD35\85 belong to different serogroups (Leung et al., 1995b). This is probably the reason for the absence of some O-antigen modification enzymes in A. hydrophila PPD35\85. These include mannosyltransferase B (F33), GDP-mannose 4,6-dehydratase (F36, 121), phosophomannomutase (F44, 86), rhamnosyltransferase (F46), O-acetyltransferase (F51), glycosyltransferase (F110) and O-antigen methyltransferase (F126). The O-antigen is thought to be important in the pathogenesis of many bacteria. It is involved in serum resistance and protecting bacteria from phagocytosis (Stinavage et al., 1989). The O-antigen of A. hydrophila is also an important adhesin (Merino et al., 1996). When aeromonads were devoid of O-antigen, their LD value increased by 100-fold (Merino et al., 1991). &! Further studies will be conducted to investigate the differences between the LPSs of these two strains and their roles in pathogenicity. The ORF encoded by F60, 91, 101 and 119 showed high homology match to site-specific DNA-methyltransferase BsuBI of B. subtilis (Xu et al., 1992) and Rhizobium leguminosarum (Rochepeau et al., 1997). These enzymes belong to a type II restriction\modification (R\M) Restriction/modification system (F60, 91, 101 and 119). system that functions to destroy foreign DNA (Xu et al., 1992). Thus, it is reasonable to postulate the existence of a similar type II R\M system in A. hydrophila PPD134\91. This is the first report of a R\M system in A. hydrophila. Its presence may be the reason why previous attempts to introduce plasmids and transposons into the tester strain (PPD134\91) failed (data not shown). Understanding and crippling this R\M system will allow easier genetic manipulation of the bacteria. Conclusions Elucidating the genetic differences between virulent and avirulent strains of A. hydrophila provides insight into the types of virulence genes. In our previous studies, many virulence factors present in PPD134\91 but not in PPD35\85 were characterized (Leung et al., 1995a, b ; Tan et al., 1998). Attempts are made to correlate some of these phenotypes with the presumptive virulence genes we obtained in this study. Features such as resistance to serum and to phagocyte-mediated killing in PPD134\91 can be explained by the detection of Omp and Oantigen-modification enzymes. The previous failure to introduce transposons into PPD134\91 has also been explained by the detection of a type II R\M system. Furthermore, the presence of haemolysin, histone-like protein, multidrug-resistance protein and other proteins is also important in virulence. Elucidation of other virulence characteristics such as adhesion and invasion awaits further study of the novel genes and examination of all the subtracted clones. Although the subtraction was successful, the library isolated is by no means complete. This was shown by the presence of only four pairs of identical clones (2\3, 44\86, 60\101, 91\119). New rounds of SSH will be carried out to complete the mapping of genetic differences between virulent and avirulent strains. Strains PPD64\90 and L36 drew particular attention. Although avirulent, they contained many tester-specific fragments (Tables 3 and 4). There might exist some regulatory mechanism or other key virulence factors that silenced their expression. It is hoped that the identification of critical genetic differences between virulent and avirulent A. hydrophila can provide insights into the pathogenic mechanisms, thus supplying the groundwork for the development of new therapies for A. hydrophila infections in both fish and humans. On the other hand, the presence of HlyA and a bacteriophage protein CII among the subtracted fragments may signify the presence of pathogenicity island(s) in A. hydrophila (Hacker et al., 1997). Experiments are currently under way to address this possibility. ACKNOWLEDGEMENTS The authors are grateful to the National University of Singapore for providing a research grant for this work. They would like to thank Drs T. T. Ngiam and H. Loh (the Primary Production Department of Singapore) for providing the A. hydrophila isolates. They also wish to thank Dr P. Tang and Ms S. H. M. Ling for helpful constructive criticism. 1007 Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Wed, 09 Aug 2017 20:51:14 Y. L. Z H A N G, C. T. O N G a n d K. Y. L E U N G REFERENCES Ahmed, M., Lyass, L., Markham, P. N., Taylor, S. S., VazquezLaslop, N. & Neyfakh, A. A. (1995). Two highly similar multidrug transporters of Bacillus subtilis whose expression is differentially regulated. J Bacteriol 177, 3904–3910. Austin, B. & Adams, C. (1996). Fish pathogens. In The Genus Aeromonas, pp. 197–229. Edited by B. Austin, M. Altwegg, P. J. Gosling & S. W. Joseph. Singapore : Wiley. Beck, E. & Bremer, E. (1980). Nucleotide sequence of the gene ompA coding the outer membrane protein II of Escherichia coli K-12. Nucleic Acids Res 8, 3011–3027. Berg, C. M., Berg, D. E. & Groisman, E. A. (1994). Transposable elements and the genetic engineering of bacteria. In Mobile DNA, pp. 880–925. Edited by E. D. Berg and M. M. Howe. Washington, DC : American Society for Microbiology. Braun, G. & Cole, S. T. (1982). The nucleotide sequence coding for major outer membrane protein ompA of Shigella dysenteriae. Nucleic Acids Res 10, 2367–2378. Brown, P. K., Romana, L. K. & Reeves, P. R. (1992). Molecular analysis of the rfb gene cluster of Salmonella serovar Muenchen (strain M67) : genetic basis of the polymorphism between groups C2 and B. Mol Microbiol 6, 1385–1394. Chakraborty, T., Montenegro, M. A., Sanyal, S. C., Helmuth, R., Bulling, E. & Timmis, K. N. (1984). Cloning of enterotoxin gene from Aeromonas hydrophila provides conclusive evidence of production of a cytotoxic enterotoxin. Infect Immun 46, 435–441. Conlin, C. A. & Miller, C. G. (1992). Cloning and nucleotide sequence of opdA, the gene encoding oligopeptidase A in Salmonella typhimurium. J Bacteriol 174, 1631–1640. Conlin, C. A., Trun, N. J., Silhavy, T. J. & Miller, C. G. (1992). Escherichia coli prlC encodes an endopeptidase and is homologous to the Salmonella typhimurium opdA gene. J Bacteriol 174, 5881–5887. Costello, G. M., Vipond, R. & MacIntyre, S. (1996). Aeromonas salmonicida possesses two genes encoding homologs of the major outer membrane protein, OmpA. J Bacteriol 178, 1623–1630. DeShazer, D., Brett, P. J. & Woods, D. E. (1998). The type II Oantigenic polysaccharide moiety of Burkholderia pseudomallei lipopolysaccharide is required for serum resistance and virulence. Mol Microbiol 30, 1081–1100. Diatchenko, L., Lau, Y. F. C., Campell, A. P. & 8 other authors (1996). Suppression subtractive hybridization : a method for generating differentially regulated or tissue-specific cDNA probes and libraries. Proc Natl Acad Sci USA 93, 6025–6030. Dooley, J. S. G. & Trust, T. J. (1988). Surface protein composition of Aeromonas hydrophila strains virulent for fish : identification of a surface array protein. J Bacteriol 170, 499–506. Dorrington, R. A. & Rawlings, D. E. (1990). Characterization of the minimum replicon of the broad-host-range plasmid pTF-FC2 and similarity between pTF-FC2 and the IncQ plasmids. J Bacteriol 172, 5697–5705. Drlica, K. & Rouviere-Yaniv, J. (1987). Histonelike proteins of bacteria. Microbiol Rev 51, 301–319. Flashner, Y. & Gralla, J. D. (1988). DNA dynamic flexibility and protein recognition : differential stimulation by bacterial histonelike protein HU. Cell 54, 713–721. Fleischmann, R. D., Adams, M. D, White, O. & 37 other authors (1995). Whole-genome random sequencing and assembly of Haemophilus influenzae Rd. Science 269, 496–512. Ghisotti, D., Finkel, S., Halling, C., Deho, G., Sironi, G. & Calendar, R. (1990). Nonessential region of bacteriophage P4 : DNA sequence, transcription, gene products, and functions. J Virol 64, 24–36. Giladi, H., Wang, W. X. & Oppenheim, A. B. (1992). Isolation and characterization of the hupA gene coding for HU of Aeromonas proteolytica. Nucleic Acids Res 20, 4092. Gurskaya, N. G., Diatchenko, L., Chenchik, A. & 7 other authors (1996). Equalizing cDNA subtraction based on selective sup- pression of polymerase chain reaction : cloning of Jurkat cell transcripts induced by phytohemagglutinin and phorbol 12myristate 13-acetate. Anal Biochem 240, 90–97. Hacker, J., Blum-Oehler, G., Muhldorfer, I. & Tschape, H. (1997). Pathogenicity islands of virulent bacteria : structure, function and impact on microbial evolution. Mol Microbiol 23, 1089–1097. Halling, C., Calendar, R., Christie, G. R. & 13 other authors (1990). DNA sequence of satellite bacteriophage P4. Nucleic Acids Res 18, 1649. Higgins, N. P. & Hillyard, D. (1988). Primary structure and mapping of the hupA gene of Salmonella typhimurium. J Bacteriol 170, 5751–5758. Hirono, I. & Aoki, T. (1993). Cloning and characterization of three hemolysin genes from Aeromonas salmonicida. Microb Pathog 15, 269–282. Hirono, I., Aoki, T., Asao, T. & Kozaki, S. (1992). Nucleotide sequences and characterization of haemolysin genes from Aeromonas hydrophila and Aeromonas sobria. Microb Pathog 13, 433–446. Howard, S. P. & Buckley, J. T. (1985). Activation of the holeforming toxin aerolysin by extracellular processing. J Bacteriol 163, 336–340. Howard, S. P., Macintyre, S. & Buckley, J. T. (1996). Toxin. In The Genus Aeromonas, pp. 267–286. Edited by B. Austin, M. Altwegg, P. J. Gosling & S. Joseph. Singapore : Wiley. Huisman, O., Faelen, M., Girard, D., Jaffe, A., Toussaintand, A. & Rouviere-Yaniv, J. (1989). Multiple defects in Escherichia coli mutants lacking HU protein. J Bacteriol 171, 3704–3712. Jaffe, A., Vinella, D. & D’Ari, R. (1997). The Escherichia coli histone-like protein HU affects DNA initiation, chromosome partitioning via MukB, and cell division via MinCDE. J Bacteriol 179, 3494–3499. Janda, J. M., Brenden, R. & Bottone, E. J. (1984). Differential susceptibility to human serum by Aeromonas spp. Curr Microbiol 11, 325–328. Kaneko, T., Sato, S., Kotani, H. & 21 other authors (1996). Sequence analysis of the genome of the unicellular cyanobacterium Synechocystis sp. strain PCC6803. II. Sequence determination of the entire genome and assignment of potential protein-coding regions. DNA Res 3, 109–136. Kano, Y., Osato, K., Wada, M. & Imamoto, F. (1987). Cloning and sequencing of the HU-2 gene of Escherichia coli. Mol Gen Genet 209, 408–410. Leblanc, D., Mittal, K. R., Olivier, G. & Lallier, R. (1981). Serogrouping of motile Aeromonas species isolated from healthy and moribund fish. Appl Environ Microbiol 42, 56–60. Leung, K. Y. & Stevenson, R. M. W. (1988a). Characteristics and distribution of extracellular protease from Aeromonas hydrophila. J Gen Microbiol 134, 151–160. Leung, K. Y. & Stevenson, R. M. W. (1988b). Tn5-induced protease-deficient strains of Aeromonas hydrophila with reduced virulence for fish. Infect Immun 56, 2639–2644. Leung, K. Y., Low, K. W., Lam, T. J. & Sin, Y. M. (1995a). Interaction of the fish pathogen Aeromonas hydrophila with 1008 Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Wed, 09 Aug 2017 20:51:14 Screening virulence genes of A. hydrophila tilapia, Oreochromis aureus (Steindachner), phagocytes. J Fish Dis 18, 435–447. Leung, K. Y., Yeap, I. V., Lam, T. J. & Sin, Y. M. (1995b). Serum resistance as a good indicator for virulence in Aeromonas hydrophila strains isolated from diseased fish in South-east Asia. J Fish Dis 18, 511–518. Lewis, K. (1994). Multidrug resistance pumps in bacteria : variations on a theme. Trends Biochem Sci 19, 119–123. Lisitsyn, N., Lisitsyn, N. & Wigler, M. (1993). Cloning the differences between two complex genomes. Science 259, 946–951. MacInnes, J. I., Trust, T. J. & Crosa, J. H. (1979). Deoxyribonucleic acid relationships among member of the genus Aeromonas. Can J Microbiol 25, 579–586. Mahairas, G. G., Sabo, P. J., Hickey, M. J., Singh, D. C. & Stover, C. K. (1996). Molecular analysis of genetic differences between Mycobacterium bovis BCG and virulent M. bovis. J Bacteriol 178, 1274–1282. Mahan, M. J., Slauch, J. M & Mekalanos, J. J. (1993). Selection of bacterial virulence genes that are specifically induced in host tissue. Science 259, 686–688. Merino, S., Camprubi, S. & Tomas, J. M. (1991). Characterization of O-antigen bacteriophage from Aeromonas hydrophila. Can J Microbiol 38, 235–240. Merino, S., Rubires, X., Aguilar, A. & Tomas, J. M. (1996). The O : 34-antigen lipopolysaccharide as an adhesin in Aeromonas hydrophila. FEMS Microbiol Lett 139, 97–101. Mittal, K. R., Lalonde, G., LeBlanc, D., Olivier, G. & Lallier, R. (1980). Aeromonas hydrophila in rainbow trout : relation between virulence and surface characteristics. Can J Microbiol 26, 1501–1503. Morona, J. K., Morona, R. & Paton, J. C. (1997). Molecular and genetic characterization of the capsule biosynthesis locus of Streptococcus pneumoniae type 19B. J Bacteriol 179, 4953–4958. Nikaido, H. (1994). Prevention of drug access to bacterial targets : permeability barriers and active efflux. Science 264, 382–388. Novak, P., Ray, P. H. & Dev, I. K. (1986). Localization and purification of two enzymes from Escherichia coli capable of hydrolyzing a signal peptide. J Biol Chem 261, 420–427. Pemberton, J. M., Kidd, S. P. & Schmidt, R. (1997). Secreted enzymes of Aeromonas. FEMS Microbiol Lett 152, 1–10. Quinn, D. D., Newman, G. W. & King, C. H. (1997). In search of virulence factors of human bacterial disease. Trends Microbiol 5, 20–26. Rice, P. A., Vayo, E. H., Tam, M. R. & Blake, M. S. (1986). Immunoglobulin G antibodies directed against protein III block killing of serum-resistant Neisseria gonorrhoeae by immune serum. J Exp Med 164, 1735–1748. Rochepeau, P., Selinger, L. B. & Hynes, M. F. (1997). Transposonlike structure of a new plasmid-encoded restriction-modification system in Rhizobium leguminosarum VF39SM. Mol Gen Genet 256, 387–396. Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning : a Laboratory Manual. Cold Spring Harbor, NY : Cold Spring Harbor Laboratory. Schnaitman, C. A. & Klena, J. D. (1993). Genetics of lipopolysaccharide biosynthesis in enteric bacteria. Microbiol Rev 57, 655–682. Scholz, P., Haring, V., Wittmann-Liebold, B., Ashman, K., Bagdasarian, M. & Scherzinger, E. (1989). Complete nucleotide sequence and gene organization of the broad-host-range plasmid RSF1010. Gene 75, 271–288. Stevenson, R. M. W. (1988). Vaccination against Aeromonas hydrophila. In Fish Vaccination, pp. 112–123. Edited by A. E. Ellis. New York : Academic Press. Stinavage, P., Martin, L. E. & Spitznagel, J. K. (1989). O antigen and lipid A phosphoryl groups in resistance of Salmonella typhimurium LT-2 to nonoxidative killing in human polymorphonuclear neutrophils. Infect Immun 57, 3894–3900. Stinson, M. W., McLaughlin, R., Choi, S. H., Juarez, Z. E. & Barnard, J. (1998). Streptococcal histone-like protein : primary structure of hlpA and protein binding to lipoteichoic acid and epithelial cells. Infect Immun 66, 259–265. Straus, D. & Ausubel, F. M. (1990). Genomic subtraction for cloning DNA corresponding to deletion mutations. Proc Natl Acad Sci USA 87, 1889–1893. Strauss, E. J. & Falkow, S. (1997). Microbial pathogenesis : genomic and beyond. Science 276, 707–712. Sugiyama, T., Kido, N., Komatsu, T., Ohta, M., Jann, K., Jann, B., Saeki, A. & Kato, N. (1994). Genetic analysis of Escherichia coli O9 rfb : identification and DNA sequence of phosphomannomutase and GDP-mannose pyrophosphorylase genes. Microbiology 140, 59–71. Tan, E., Low, K. W., Wong, W. S. F. & Leung, K. Y. (1998). Internalization of Aeromonas hydrophila by fish epithelial cells can be inhibited with a tyrosine kinase inhibitor. Microbiology 144, 299–307. Thune, R. L., Stanley, L. A. & Cooper, K. (1993). Pathogenesis of Gram-negative bacterial infections in warm water fish. Annu Rev Fish Dis 3, 37–68. Vadivelu, J., Puthucheary, S. D., Phipps, M. & Che, Y. W. (1995). Possible virulence factors involved in bacteremia caused by Aeromonas hydrophila. J Med Microbiol 42, 171–174. Weiser, J. N. & Gotschlich, E. C. (1991). Outer membrane protein A (OmpA) contributes to serum resistance and pathogenicity of Escherichia coli K-1. Infect Immun 59, 2252–2258. Wilmsen, H. U., Pattus, F. & Buckley, J. T. (1990). Aerolysin, a hemolysin from Aeromonas hydrophila, forms voltage-gated channels in planar lipid bilayers. J Membr Biol 115, 71–81. Wong, C. Y. F., Heuzenroeder, M. W. & Flower, R. L. P. (1998). Inactivation of two haemolytic toxin genes in Aeromonas hydrophila attenuates virulence in a suckling mouse model. Microbiology 144, 291–298. Xu, G. L., Kapfer, W., Walter, J. & Trautner, T. A. (1992). BsuBI – an isospecific restriction and modification system of PstI : characterization of the BsuBI genes and enzymes. Nucleic Acids Res 20, 6517–6523. Yao, Z. & Valvano, M. A. (1994). Genetic analysis of the O-specific lipopolysaccharide biosynthesis region (rfb) of Escherichia coli K12 W3110 : identification of genes that confer group 6 specificity to Shigella flexneri serotypes Y and 4a. J Bacteriol 176, 4133–4143. Zhang, L. J., Toivanen, P. & Skurnik, M. (1996). The gene cluster directing O-antigen biosynthesis in Yersinia enterocolitica serotype O : 8 : identification of the genes for mannose and galactose biosynthesis and the gene for the O-antigen polymerase. Microbiology 142, 277–288. ................................................................................................................................................. Received 31 August 1999 ; revised 2 December 1999 ; accepted 11 December 1999. 1009 Downloaded from www.microbiologyresearch.org by IP: 88.99.165.207 On: Wed, 09 Aug 2017 20:51:14