Vibrio parahaemolyticus isolates derived from an outbreak of gastroenteritisin the Republic of Maldives did not have the genetic potential to produce the ...
INFECTION AND IMMUNITY, Sept. 1989, p. 2691-2697
Vol. 57, No. 9
0019-9567/89/092691-07$02.00/0 Copyright C 1989, American Society for Microbiology
Cloning and Nucleotide Sequence of the Gene (trh) Encoding the Hemolysin Related to the Thermostable Direct Hemolysin of Vibrio parahaemolyticus MITSUAKI NISHIBUCHI,t* TOORU TANIGUCHI, TADASHI MISAWA, VANNA KHAEOMANEE-IAM, TAKESHI HONDA, AND TOSHIO MIWATANI Research Institute for Microbial Diseases, Osaka University, 3-1 Yamadaoka, Suita, Osaka 565, Japan Received 18 August 1988/Accepted 15 May 1989
Vibrio parahaemolyticus isolates derived from an outbreak of gastroenteritis in the Republic of Maldives did not have the genetic potential to produce the thermostable direct hemolysin, but one such isolate produced a hemolysin immunologically related to the thermostable direct hemolysin (T. Honda, Y. Ni, and T. Miwatani, Infect. Immun. 56:61-965, 1988). The Maldives isolates hybridized with the DNA probe for the gene encoding the thermostable direct hemolysin (the tdh gene) under reduced stringencies. A DNA fragment containing the probe-reactive nucleotide sequence was isolated from a selected strain and cloned into pBR322 in Escherichia coli. A clone producing the thermostable direct hemolysin-related hemolysin was obtained by screening with hemolysis assays and by an immunological assay. Nucleotide sequence analysis of the cloned DNA fragment revealed that the gene encoding the thermostable direct hemolysin-related hemolysin (the trh gene), like the tdh gene, encoded the hemolysin subunit composed of 189 amino acid residues. The trh gene had significant nucleotide sequence homology with the tdh gene (68.4% with the tdhl gene copy and 68.6% with the tdh2 gene copy). The amino acid sequences of the hemolysin subunits deduced from the nucleotide sequences of the trh gene and tdh gene were homologous (61.9% homology with the tdhl-encoded subunit and 63.0% homology with the tdh2-encoded subunit) and contained the two cysteine residues to form an intrachain bond at the same positions, and their possible conformations appeared to be similar as determined by hydrophobicityhydrophilicity analysis and a secondary structure prediction. The trh and tdh genes may have had a common ancestor and may have evolved by single-base changes so that they maintained the fundamental architecture of the molecules. We recently studied the cases of traveller's diarrhea originating in an outbreak of gastroenteritis in the Republic of Maldives and found that the outbreak was caused by another type of KP- V. parahaemolyticus (5). The isolates did not carry the tdh gene, but it was subsequently found that one of such isolates produced the TDH-related hemolysin (TRH). TRH was immunologically similar to TDH, but the two hemolysins had significantly different physicochemical characteristics and lytic activities for various erythrocytes (6). In this study, we cloned the gene encoding TRH (named the trh gene) and compared TDH and TRH on the basis of the nucleotide sequences of the genes. Our results suggested that the two hemolysins have diverged significantly from a common ancestor but that they maintain a fundamental molecular structure.
Vibrio parahaemolyticus, a marine bacterium, may cause gastroenteritis when introduced into humans through seafoods (1). Past epidemiological studies revealed that almost all clinical isolates induced Kanagawa phenomenon (KP+), i.e., a zone of p-type hemolysis around the colony on a special blood agar (Wagatsuma agar), whereas only 1 to 2% of environmental isolates were KP+ (14, 22). Production of Kanagawa hemolysin or the thermostable direct hemolysin (TDH) is responsible for the Kanagawa phenomenon; thus, TDH has been considered a major virulence factor (26). TDH is cytotoxic to various cultured cells and is a potent cardiotoxin, but its enteropathogenicity has not been clearly demonstrated in experimental animals (26). In addition, isolation of the strains which do not show the Kanagawa phenomenon (KP-) from clinical sources, although frequency has been very low, apparently disproves a pathogenic role for TDH. We analyzed the gene encoding TDH (the tdh gene) and made the DNA probe and oligonucleotide probes specific to the tdh gene (17-19). The studies using the probes resulted in two findings which can explain the above-mentioned contradiction. Some of the KP- clinical strains had the tdh genes, but the genes were not efficiently expressed under in vitro conditions (17, 18; M. Nishibuchi and J. B. Kaper, submitted for publication). Another finding was that a KP+ isolate spontaneously lost the tdh gene and became KP- (18).
MATERIALS AND METHODS Bacterial strains, media, and plasmids. V. parahaemolyticus WP1 and S162-71 were used as a positive and a negative control, respectively, in the DNA colony hybridization test with the tdh gene. The sources of the two strains were described previously (19). Eleven KP- strains of V. parahaemolyticus were isolated from travelers with diarrhea returning from the Republic of Maldives at the Osaka International Airport Quarantine Station during the year 1985 (5). Other Vibrio species used as negative controls in the DNA colony hybridization test had been maintained as stock cultures in our laboratory. Escherichia coli HB101 (3) and JM103 (12) were used as hosts for propagating plasmids and bacteriophage M13 clones, respectively. To grow E. coli HB101 and its derivatives, LB medium (13) was used as the
* Corresponding author. t Present address: Department of Microbiology, Faculty of Medicine, Kyoto University, Konoe-cho, Yoshida, Sakyo-ku, Kyoto 606, Japan.
2691
2692
INFECT. IMMUN.
NISHIBUCHI ET AL.
basal medium. When necessary, the medium was supplemented with ampicillin (200 ,ug/ml) or calf erythrocytes (10% [vol/vol]) or both. YT medium (13) was used for strain JM103 to propagate phages. Plasmid vector pBR322 (bla+ tet+) (2) was used for cloning DNA fragments carrying the trh gene. The recombinant plasmids constructed in this study were pBIK8 (a 2.3-kilobase-pair [kb] AvaI fragment carrying the trh gene and another 2.3-kb AvaI fragment, both derived from V. parahaemolyticus AQ4037 DNA, cloned in tandem into the AvaI site of pBR322 [bla+ tet+ trh+]) and pBIK9 (a 3.1-kb AvaI-Sall fragment of pBIK8 carrying the trh gene cloned into AvaI- and Sall-cleaved pBR322 [bla+ trh+]). DNA colony hybridization test. The DNA colony hybridization test with the 415-base-pair (bp) DNA probe derived from the internal portion of the tdh gene was carried out as described previously (16, 18) with minor modifications. The method was modified so that the hybridization was performed under different stringencies. The hybridization stringency was controlled by changing the concentration of formamide in the hybridization solution and the temperature at which the colony blot was washed after hybridization (15). Under high stringency (stringency A), the hybridization solution containing 50% formamide was used and the blots were washed at 65°C. Reduced stringencies (designated stringencies B, C, D, and E) were set up by using the combinations of formamide concentration and washing temperature of 40% and 65°C, 35% and 61.5°C, 25% and 54.5°C, and 20% and 51°C, respectively. Plasmid assay. Plasmid contents of V. parahaemolyticus isolates which originated in the Maldives outbreak were assayed by the method of Kado and Liu (7). Southern blot hybridization. The DNA digested with a restriction endonuclease was electrophoresed in a 0.8% agarose gel, blotted onto a nitrocellulose sheet by the method of Southern (25), hybridized with the tdh gene probe (18), and autoradiographed. The hybridization procedure was the same as that used for the DNA colony hybridization test described above, and the formamide concentration and washing temperature selected were 35% and 61.5°C, respectively (corresponding to stringency C [see Table 1]). General cloning techniques. Total DNA was extracted by a modification of the method of Marmur (10) as described by Seidler and Mandel (24). Digestion of DNA with restriction enzymes, gel electrophoresis, isolation of DNA fragments from the gel, ligation, and transformation were performed as previously described (19). Detection of intracellular TRH in E. coli clones. Intracellular TRH produced in E. coli clones was detected by examining hemolysis in a blood agar plate and reactivity with anti-TDH polyclonal serum in the Ouchterlony immunodiffusion plate. The assays were performed by the methods previously described for detection of intracellular TDH in E. coli clones (19), with a modification. The blood agar plate was modified to contain calf erythrocytes in place of rabbit erythrocytes because TRH had much higher lytic activity for calf erythrocytes than for rabbit erythrocytes (6). Nucleotide sequence analysis. Various DNA fragments generated by restriction endonuclease digestion were cloned into appropriate restriction sites of vectors M13, mpl8, and mpl9 (20, 30) and pBR322 (2). The nucleotide sequences of the cloned fragments were determined by the dideoxy-chain termination method of Sanger et al. (23) as described by Messing (11) or by Wallace et al. (28). The amino acid sequence deduced from the nucleotide sequence of the gene was analyzed for hydropathic character by the method of Kyte and Doolittle (8) by using a span of 5 amino acid
TABLE 1. Results of the DNA colony hybridization test with the tdh gene probe Hybridization reactiona under
stringencyb:
Organism A
V. parahaemolyticus WP1 (tdh+) S162-71 (tdh) 11 Maldives isolates (KP-)
01 V. cholerae 569B Non-O1 V. cholerae 34 V. damsela ATCC 33539 V. furnisii ATCC 35061 V. mimicus ATCC 33653
B
C
D
E
+ +W +
+ +W + -
+
+
+
-
-
-
-
+W
+
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
+W
-
-
-
+W
+, Positive; +W, weakly positive; -, negative. b Conditions were set up by the method of Moseley and Falkow (15). Formamide concentrations in the hybridization solution and the temperatures at which the colony blot was washed after hybridization were 50% and 65°C, 40% and 65°C, 35% and 61.5°C, 25% and 54.5°C, and 20% and 51°C for stringencies A, B, C, D, and E, respectively. a
residues as a test segment and for possible secondary structure by the method of Chou and Fasman (4). RESULTS DNA colony hybridization test. V. parahaemolyticus isolates derived from the outbreaks in the Maldives and other control strains were examined with the tdh gene probe under different hybridization conditions by the DNA colony hybridization method, and the results are summarized in Table 1. Hybridization results under a high stringency (stringency A in Table 1) indicated that 11 Maldives isolates did not carry the tdh gene. They did, however, exhibit positive reactions under a reduced stringency, while other control strains (except WP1) were still negative under the same conditions (stringency C in Table 1). The results indicated that Maldives isolates carry nucleotide sequences homologous to but not identical to the tdh gene. The homology was estimated to be more than 72.5% but less than 80% on the basis of the hybridization conditions (15). The homologous sequence was assumed to be the trh gene. Cloning the trh gene. Eleven Maldives isolates were assayed for their plasmid contents. Only three of the isolates contained small plasmids (data not shown), suggesting that the trh gene was present on the chromosome. Total DNA was extracted from one selected strain, AQ4037, which was shown to produce TRH (6). The DNA was digested with one of the following restriction endonucleases: AvaI, BamHI, BglII, EcoRI, HindIII, PstI, or Sall. The digested DNA was subjected to Southern blot hybridization analysis with the tdh gene probe under the reduced stringency (stringency C). AvaI and HindIll digests contained relatively small DNA fragments (2.3 and 3.0 kb, respectively) which hybridized with the tdh gene probe (data not shown). The AvaI digest was size fractionated by agarose gel electrophoresis, and DNA fragments with an approximate size of 2.3 kb were isolated and ligated with Aval-cleaved and bacterial alkaline phosphatase-treated pBR322. The ligation mixture was transformed into E. coli HB101. The transformants were screened on LB agar containing ampicillin and calf erythrocytes for hemolysin production. After overnight incubation, one clone was found to show a very small zone of hemolysis around the colony. The clone harbored a recombinant plasmid (pBIK8) containing two nonidentical 2.3-kb AvaI frag-
tdh-RELATED HEMOLYSIN GENE OF V. PARAHAEMOLYTICUS
VOL. 57, 1989
2693
2.3 kb Aval
TMgAenSI
A
H
(S H
A)
(
pWIK8
--
*.4 kb
1.0 kb
A
r th
D
A
H
AVal Sail H
A S Aval
H (SXA)
~a
trt
pBRIK9
WAS)
H
(A)
FIG. 1. Cloning of the DNA fragments carrying the trh gene of V. parahaemciyticus into plasmid pBR322. Liii, DNA derived from V. parahaemoiyticus AQ4037. ~, DNA of pBR322 origin. Restriction sites are abbreviated as follows: A, Aval; H, Hindill; S, Sall; (A), nucleotide sequence originally cleaved with Aval (the AvaI site was not regenerated after ligation); (S), nucleotide sequence originally cleaved with Sall (the Sall site was not regenerated after ligation). The locations and transcriptional directions of the genes coding for ampicillin resistance (bla) and tetracycline resistance (tet) are indicated. The location and transcriptional direction of the trh gene (trh) also shown were determined later by analyzing the sequences (see text).
ments inserted in tandem (Fig. 1). Southern blot hybridization analysis of various restriction fragments of pBIK8 indicated that, of the two cloned AvaI fragments, the fragment containing a Hindlll site carried the presumed trh gene (data not shown). Since AvaI-AvaI ligation did not regenerate the AvaI site at one end of the above-mentioned AvaI fragment, probably because of ligation between nonidentical Aval-end sequences, the 3.1-kb AvaI-SalI fragment of pBIK8 carrying the presumed trh gene was isolated and ligated with AvaI- and Sall-cleaved pBR322 (Fig. 1) and transformed into E. coli HB101. Transformants were screened for hemolysin production on LB agar containing ampicillin and calf erythrocytes. One hemolytic clone harbored a recombinant plasmid, pBIK9 (Fig. 1). Restriction mapping revealed that pBIK9 was not generated by simple Sall-Sall and AvaI-AvaI sticky-end ligation. It contained the 3.1-kb AvaI-SalI fragment cloned in the reverse orientation. Possible reasons for the ligation event are discussed later. The hemolysis zone around the colony of E. coli HB101 (pBIK9) was larger than that of E. coli HB101(pBIK8). Intracellular hemolysin production in E. coli HB101(pBIK9) was also examined. The cell lysate induced clear lysis of calf erythrocytes in the blood agar plate. The cell lysate also
FIG. 2. Ouchterlony immunodiffusion analysis of the cell lysate of E. coli HB101 (pBIK9). 1, Cell lysate of E. coli HB101; 2, cell lysate of E. coli HB101 (pBIK9); 3, purified TDH; anti-TDH, anti-TDH rabbit serum.
formed a precipitation line with anti-TDH serum in an Ouchterlony immunodiffusion plate, as shown in Fig. 2, indicating that the cell lysate contained a molecule which shared an antigenic determinant(s) with TDH. These characteristics agreed with those previously described for purified TRH (6); thus, the presence of the trh gene in pBIK9 and the expression of the gene in E. coli were confirmed. Comparison of the trh gene and the tdh gene. The restriction map of the 2.3-kb insert of pBIK9 carrying the trh gene was made, and Southern blot hybridization analysis of pBIK9 digested with various restriction enzymes was carried out. The results (not shown) predicted that the trh gene existed within the 1.0-kb AvaI-HindIII fragment shown in Fig. 3. Comparison of the restriction maps of the DNA fragments containing the trh gene and tdh gene copies (all KP+ V.
H SHCC NV I I
tdhl tdh2
L
I
I
EI Al
trh
1m1
EHHC I
HI IC
lee
H ICC Nf I -
A i
I KB FIG. 3. Comparison of the restriction maps of the DNA fragments carrying the tdh gene copies and the trh gene. KP+ V. parahaemolyticus carries two tdh gene copies named tdhl and tdh2 (Nishibuchi and Kaper, submitted). Restriction sites are abbreviated as follows: A, AvaI; B, BamHI; C, Cial; E, EcoRI; H, Hindlll; I, HpaI; P, PstI; S, Sall. _, Area within which the gene is located. Locations of tdhl and tdh2 were determined based on the nucleotide sequences of the gene. Location of the trh gene was predicted by
Southern blot hybridization analysis of pBIK9 digested with various restriction enzymes.
2694
NISHIBUCHI ET AL. 10
trF
20
INFECT. IMMUN. an
40
CCACATAAX8_ATTAATTAT
50 60 GAAACTAAAA CTCTACTTTG TAT-TTGCAA TAT-TTGCAA
TCiAGCA&TC IJGJCAAAA dh----ACC-t1dh ---GTACCG--
10
20
30
40
50
60
MKLKLYFAFS LLLASIFSVS KSFAIDLPSI PFPSPGSDEL LFVVRNTTIK TESPVNAIVN Itr tdhi --HQYFAKK- F-FI-MLAAF -TS-FE---V ---A----- I-----D--FN -QA---VK-S tdh2 --YRYFAKK- F-FI-MLAAF -T--FE---V ---A----- I-----D--FN -NA---VE-S
70 s0 90 100 110 120 CTTTCAGTTT GCTATTGGCT TCGATATTTT CAGTATCTAA ATCATTCGCG ATTGACCTAC AAAAATCA-- TT----TATA --C--G--GG -T-C--TC-- -A---CT--T T----G--TAAAAATCA-- TT----TATA --C--G--GG -T-C--TC-- -A----T--C T----G--T-
70 80 90 100 110 120 DYWTNRNIKR KPYKSVHGQS IFTTSGSKWL SAYITVNING NNYTMAALSG YKDGLSTVFT -F-----V-- ---ED-Y--- V-----T--- TS-M-----D D-----V-- --S-H-A--V -F-----V-- ----D-Y--- V-----T--- TS-M-----D D-----V-- --H-li-A--V
130 140 150 160 170 180 CATCCATACC TTTTCCTTCT CCAGGTTCGG ATGAGCTACT ATTTGTCGTT AGAAATACAA ----TG-C-- -------G-C --C-----T- -----A--T- G-----T--- C--a---------TG-C-- -------G-C --C-----T- -----A--T- G-----T--- C--G------
130 140 150 170 160 180 KSEKTSLNQN YSSVSDFVGE NEESLPSVTY LDETPEYFVN VEAYESGNGH MFVMCISNKS --GQVQ-QHS -N--AN---- D-G-I--KM- ---------- -------S-N IL-------E --DQVQ-QHS -D--AN---- D-D-I--KM- ---------- -------S-N IL-------E
190 200 210 220 230 240 CAATAAAAAC TGAATCACCA GTTAACGCAA TCGTTAATGA CTACTGGACA AACCGAAACA -TT-T--T-- CC--G-T--G --C--T-T-- AG--CTC--- --TT------ -----T--TG -TT-T--T-- CA-TG----G --C--T-T-G AG--CTC--- --TT------ -----T--TG 250 260 270 280 290 300 TAAAACGAAA ACCATATAAA AGCGTTCACG GTCAATCTAT TTTCACGACT TCAGGCTCAA -A---- ---G--CG-- aAT---T-T- -------AG- A-----A--G -----TA-TA---- ---G--C--- GAT---T-T- -------AG- A-----A--G -----TA-T-
310 320 330 340 350 360 AATGGTTAAG CGCCTATATA ACGGTAAATA TTAATGGAAA TAACTACACA ATGGCTGCTC -------G-C AT----C--G --T--G--C- ------AT-- AGa----T --A--GG -----C-G-C AT----C--G --T--G--C- ------ATr-- AG----T--- -----A--GG 370 380 390 400 410 420 TTTCTGGCTA TAAAGATGGC CTTTCAACGG TCTTCACAAA ATCAGAAAAA ACAAGCCTAA -G-------- ---GAGC--T -A---TG-T- -G---GT--- -----GTC-- GT-CAA--TC -G-------- ---GC-C--T -A---TG-T- -G---GT--- ------TC-- GT-CAG--TC 430 440 450 460 470 480 ATCAGAACTA TTCTTCTGTT AGTGATTTCG TTGGTGAAAA TGAAGAATCA TTGCCAAGTG -A--TTC--- -AA------A GC-A-C--T- -------- a-----GT--T A-T------A -A--TTC--- -GA------A GC-A-C--T- --------G- ------T--T A-T------A 490 500 510 520 530 540 TAACGTATTT GGATGAAACG CCAGAATATT TCGTCAATGT CGAAGCATAT GAGAGCGGAA A--T------ ---------T ---------- -T--T----- A--------- -----T--TA--T------ ---------T ---------- -T--T----- A--------- -----T--T-
550 560 570 580 590 600 ATGGGCATAT GTTTGTTATG TGTATTTCCA ATAAATCATC ATTTGATGAA 'I'GTATGTCAC G---TA---- A--G--A--- -----AA--AG---Tr- --ACATG---TA---- A--G--A--- -----A---- -C---GA--- ----rTT---- ----AACAT610 640 620 630 650 AAAATTAAAC AATCATAAAT CACCACTATT AGTGGTGATI' TATATTTCTA --C-A----C-A---
660
TCTrTrACATAT
FIG. 4. Nucleotide sequence of the trh gene. Numerals indicate the positions of the bases starting at the 5' end. The sequence TCGAG (positions 1 through 5) corresponds to the AvaI site present at the left end of the restriction map for the trh gene in Fig. 3. The nucleotide sequence of the coding region (positions 38 through 608) is compared with those of tdhl and tdh2 gene copies (20; Nishibuchi and Kaper, submitted). -, Base identical to that of the trh gene. The sequence showing homology with the consensus sequence of the E. coli Shine-Dalgarno sequence (9) is indicated (E Eaa) above the underlined sequence. < , Inverted repeat sequence.
parahaemolyticus strains carry two similar but not identical tdh gene copies named tdhl and tdh2 [Nishibuchi and Kaper, submitted]) indicated that the nucleotide sequences surrounding the trh gene and those surrounding the tdh gene copies were not similar (Fig. 3). The results suggest that the trh gene was not generated by simple mutation of the tdh gene or vice versa. The fine restriction map of the trh gene-bearing AvaIHindlIl fragment was made (data not shown), and smaller subfragments containing the portions of the AvaI-HindIII fragment were cloned into M13 vectors or pBR322 and their nucleotide sequences were determined and analyzed. A 567-bp nucleotide sequence encoding an open reading frame started at the nucleotide 38 bp away from the AvaI site. The sequence was homologous to the nucleotide sequences of the coding region of the tdhl and tdh2 gene copies (5; Nishibuchi and Kaper, submitted) and thus was judged to be the trh gene (Fig. 4). A sequence homologous to the consensus sequence of the E. coli Shine-Dalgarno sequence (9) was present 4 bp upstream of the coding region (positions 29 through 34 in Fig. 4). Inverted repeat sequences existed upstream (positions 3 through 16) and downstream (positions 615 through 642) of the coding sequence. The latter could possibly be involved in transcriptional termination (21). The coding sequence contained the HpaI site (GTTAAC, positions 201 through 206) that existed in the restriction map
190 SFDECMSQN* --F--KH-Q- -F--KH-Q-
200
210
220
230
240
FIG. 5. Comparison of the amino acid sequences of the TRH subunit and of TDH subunits deduced from the nucleotide sequences of the trh gene and the tdh gene copies (tdhl and tdh2). Sequences of the tdhl- and tdh2-encoded TDHs were taken from other studies (19; Nishibuchi and Kaper, submitted). The first 24-amino-acid sequences constitute putative signal peptides.-, Same amino acid residue as that of the TRH; *, termination code.
shown in Fig. 3, and transcription was directed toward the HindIII site (698 bp downstream of the HpaI site, not shown in Fig. 4). Therefore, the trh gene was presumably transcribed from the promoter of the truncated tet gene in pBIK9 (Fig. 1). When the sequences outside the coding regions were compared, sequence homology between the trh gene and tdh gene copies was not detected (not shown in Fig. 4). The coding region of the trh gene had 68.4 and 68.6% homology with the tdhl and tdh2 gene copies, respectively. The difference in the nucleotide sequences among the trh, tdhl, and tdh2 gene copies can be accounted for by various types of base changes but they had similar G-plus-C contents (35.1, 35.6, and 35.6%, respectively). Like the tdh gene copies, the trh gene encoded a polypeptide composed of 189 amino acid residues (Fig. 5). The first 24 amino acid residues encoded by the tdh gene constituted a putative signal peptide (19). Assuming that this is the case with the trh gene, the calculated molecular weight of the preprotein is 21,082, and that of the mature protein is 17,959. Like TDH, TRH was supposed to be composed of two identical subunits, and the molecular weight of the TRH subunit was estimated to be 23,000 by sodium dodecyl sulfate-polyacrylamide gel electrophoresis of the purified protein (6). Such a discrepancy between the nucleotide sequence-based calculation and protein gel electrophoresis-based estimation was also observed in calculating the molecular weight of the TDH subunit (19). The nucleotide sequence homology was reflected in the deduced amino acid sequence homology. The trh-coded polypeptide had 61.9 and 63.0% homology with the tdhl- and tdh2-encoded polypeptides, respectively. The putative signal peptide sequence encoded by the trh gene (positions 1 through 24 in Fig. 5) was significantly different from those encoded by the tdh gene copies (29.2 and 33.3% homology with the tdhl- and tdh2-encoded putative signal peptide sequences, respectively). Two cysteine residues formed an intrachain bond in the subunit of TDH (27). The two cysteine residues were also conserved in the TRH subunit encoded by the trh gene (positions 175 and 185 in Fig. 5). The hydrophobic-hydrophilic properties of the TRH subunit and the TDH subunit were compared by using the hydropathic index of Kyte and Doolittle (8). Since the hydropathic profiles of the tdhl- and tdh2-encoded TDH subunits were extremely similar (tdhl and tdh2 gene copies had 97.2% nucleotide sequence homology), and the tdh2 but not the tdhl gene copy seemed to be expressed in a KP+ strain (Nishibuchi and Kaper, submitted), a comparison was made between the TRH subunit encoded by the trh gene and
tdh-RELATED HEMOLYSIN GENE OF V. PARAHAEMOLYTICUS
VOL. 57, 1989
2695
3.0 x LU
0*.0 a0
IOL
101.
-3.0 3.0
0.
201
-3.
RESIDUE POSITION FIG. 6. Comparison of the hydropathic indices of the TRH subunit encoded by the trh gene and the TDH subunit encoded by the tdh2 gene copy. Broken and solid curves indicate, respectively, the TRH subunit encoded by the trh gene and the TDH subunit encoded by the tdh-2 gene copy. Positive and negative values on the ordinate denote, respectively, the hydrophobic and hydrophilic indices (8). Numerals shown along the axis indicate the positions of amino acid residues starting at the N terminus. Putative signal peptides correspond to positions 1 through 24.
the TDH subunit encoded by the tdh2 gene copy (Fig. 6). There was no significant differerice between them except that the putative signal peptide of TRH was more hydrophobic than that of TDH. The TRH subunit coded by the trh gene and the TDH subunit coded b)y the tdh2 gene copy were also compared for possible secondlary structure predicted by the method of Chou and Fasman ((4). There are some minor differences between the two prediicted structures in each of the regions where the amino acid sequences could possibly nucleate (x-helices or n-sheets cwr form ,-turns (Fig. 7). However, the region where the c,onformation could drastically differ over a large stretch w;as not noted.
This made us suspect that the trh gene may have significant nucleotide sequence homology with the tdh gene. Results of the DNA colony hybridization test with the tdh gene probe under various stringencies supported this hypothesis (Table 1). The homology between the tdh gene and the trh gene was estimated to be more than 72.5% but less than 80% by the hybridization analysis. The 415-bp tdh gene probe used in the hybridization test was derived from an internal portion of the tdhl gene copy (18). The homology between the two genes within this region was actually 72.6% (nucleotide positions 171 through 576 in Fig. 4) and thus shows that the hybridization conditions used for cloning the trh gene were
appropriate. In cloning the trh gene, unusual ligation events confused us. This can be explained by the lack of the promoter sequence of the trh gene on the 2.3-kb AvaI fragment. It was found that the promoter-proximal coding region of the trh gene started at the nucleotide 38 bp away from the AvaI site. The 38-bp stretch contained a putative Shine-Dalgarno sequence but was too short to accommodate a functional __promoter sequence. Therefore, the cloned trh gene needed to be transcribed from an exogenous promoter. In an attempt to clone the trh gene-bearing AvaI-SalI fragment of pBIK8 into the AvaI-SalI sites of pBR322, a hemolytic clone -- containing the recombinant plasmid with the insert in the __ - right orientation was not obtained. A very unusual recombi_ nant plasmid (pBIK9) with the AvaI-SalI fragment inserted in the opposite orientation, probably formed by rare in vivo ligation, was obtained instead (Fig. 1). Most likely, the trh - - - -gene was transcribed by the promoter of the truncated tet - - - -gene in pBIK9. There is no known promoter transcribing in the opposite direction around the AvaI site of pBR322. This
DISCUSS1ION TRH shared a common antigIenc cdeterminant(s) with TDH and had a subunit structure similar to that of TDH (6).
HELIX
- -
trh
tdh2
__ __ -__ __
tdh2
-- - - - - -
trh
TURN
tdh2
I
So
100
_
lS0
200
RESIOUJE POSITION1 FIG. 7. Comparison of the possible secondary structures of TRH encoded by the trh gene and of TDH ericoded by the tdh2 gene copy. The possible secondary structures wetre predicted by the method of Chou and Fasman (4). The regions whiere the amino acid sequences may possibly nucleate ot-helices (HELAX) or P-sheets (SHEET) or form ,-turns (TURN) are indicated bzy the straight lines. Numerals on the axis denote the positions of the- amino acid residues starting at the N terminus. Putative signal pep tides correspond to positions 1 through 24.
is probably the reason why the recombinant plasmid with the
in the right orientation was not detected AvaI-SalI fragment In the transformants with the
hemolysis assay. by screening pBIK8, the trh gene was probably transcribed from a weak promoter existing on the AvaI fragment of V. parahaemolyticus origin, which was ligated to the trh gene-bearing AvaI fragment. A prominent difference in the amino acid sequences deduced from the nucleotide sequences of the TRH and TDH subunits was found in the putative signal peptide region (Fig. 5). The putative signal peptide encoded by the
2696
NISHIBUCHI ET AL.
trh gene was more hydrophobic than that of the tdh gene (Fig. 6). It is tempting to speculate that the hydrophobic signal peptide coded by the trh gene is functional to some degree in E. coli. It can be a possible explanation for the observation that screening of the trh gene-bearing E. coli clones on the blood agar medium was possible, while the tdh gene-bearing E. coli clones had to be screened by examining intracellular hemolysin production (19). Another possibility is a difference in signal peptidase processing. Both the trh and tdh genes had similar G-plus-C contents, coded for polypeptides of identical lengths, and had significant nucleotide sequence homology. Despite 37% mismatches of the amino acid sequences deduced from the nucleotide sequences, the possible conformations of TRH and TDH appear to be similar; the hydrophobic-hydrophilic properties and the predicted secondary structures of the TRH and TDH subunits were not significantly different, and the two cysteine residues needed to form an intrachain bond were conserved in both hemolysin molecules. It seems, therefore, that both genes have a common ancestor and evolved by single-base changes so that they maintained the fundamental architecture of the molecules. This is reminiscent of the relationship between the cholera toxin of V. cholerae and the heat-labile enterotoxin of enterotoxigenic E. coli, both of which cause diarrhea through activation of adenylate cyclase of the target cells (15, 29). Some of the physicochemical characteristics of TRH and TDH and the susceptibilities of various erythrocytes of the two hemolysins were dissimilar (6). These can be explained by the differences in composition and conformation of the two hemolysin molecules caused by the mismatches in the amino acid sequences. An epidemiological study with tdh-specific and trh-specific gene probes revealed that strains potentially capable of producing TDH or TRH are associated with gastroenteritis cases (manuscript in preparation). If TRH and TDH play similar roles in inducing gastroenteritis, they should share a similar functional domain(s). Such a functional domain(s) may be predicted by amino acid sequence homology. We are currently examining the genes encoding TDH-like hemolysins of other Vibrio species isolated from gastroenteritis cases. Comparative analysis of these genes together with the present results may help to predict the amino acid sequences constituting the functional domain(s), which will lead to elucidation of the role of TDH, TRH, and related hemolysins in the pathogenesis of V. parahaemolyticus. Our future studies will be mutagenesis of the tdh and trh genes to assess the structure-function relationship of the hemolysin molecules and animal experiments with isogenic mutants to assess the pathogenic role of the hemolysins. It is interesting to find that two hemolysin genes which have 68.4 to 68.6% nucleotide sequence homology exist in one bacterial species. We previously hypothesized that the tdh gene may have moved as a small genetic unit because the nucleotide sequences surrounding the structural genes of representative gene copies were heterologous, whereas the structural genes had very homologous sequences (Nishibuchi and Kaper, submitted). A similar observation was made when the trh gene and the tdh gene and their surrounding sequences were compared by restriction mapping (Fig. 3) and direct sequencing. The results support the hypothesis that the trh gene and the tdh gene may have been derived from a common ancestor and moved into different genetic backgrounds, and then the nucleotide sequences were modified therein. Hemolytic activities of TDH and TRH were completely or partially neutralized with antisera raised to TDH and to TRH (6). It seems, therefore, that the divergent
INFECT. IMMUN.
hemolysin gene sequences are due to random drift and clonal separation. ACKNOWLEDGMENTS The study was supported by grants 62770289 and 63570197 provided by the Ministry of Education, Science, and Culture of the Japanese government and by a grant from the U.S.-Japan Cooperative Science Program Cholera Panel.
LITERATURE CITED 1. Blake, P. A., R. E. Weaver, and D. G. Hollis. 1980. Diseases of humans (other than cholera) caused by vibrios. Annu. Rev. Microbiol. 34:341-367. 2. Bolivar, F., R. L. Rodriguez, P. J. Greene, M. C. Betlach, H. L. Heyneker, H. W. Boyer, J. H. Crosa, and S. Falkow. 1977. Construction and characterization of new cloning vehicles. II. A multipurpose cloning system. Gene 2:95-113. 3. Boyer, H. W., and D. Roulland-Dussoix. 1969. A complementation analysis of the restriction and modification of DNA in Escherichia coli. J. Mol. Biol. 41:459-472. 4. Chou, P. Y., and G. D. Fasman. 1978. Empirical predictions of protein conformation. Annu. Rev. Biochem. 47:251-276. 5. Honda, S., I. Goto, I. Minematsu, N. Ikeda, N. Asano, M. Ishibashi, Y. Kinoshita, M. Nishibuchi, T. Honda, and T. Miwatani. 1987. Gastroenteritis due to Kanagawa negative Vibrio parahaemolyticuis. Lancet i:331-332. 6. Honda, T., Y. Ni, and T. Miwatani. 1988. Purification and characterization of a hemolysin produced by a clinical isolate of Kanagawa phenomenon-negative Vibrio parahaemolyticus and related to the thermostable direct hemolysin. Infect. Immun. 56:961-965. 7. Kado, C. I., and S. T. Liu. 1981. Rapid procedure for detection and isolation of large and small plasmids. J. Bacteriol. 145: 1365-1373. 8. Kyte, J., and R. F. Doolittle. 1982. A simple method for displaying the hydropathic character of a protein. J. Mol. Biol. 157:105-132. 9. Lewin, B. 1987. Genes. John Wiley & Sons, Inc., New York. 10. Marmur, J. 1961. A procedure for the isolation of deoxyribonucleic acid from microorganisms. J. Mol. Biol. 3:208-218. 11. Messing, J. 1983. New M13 vectors for cloning. Methods Enzymol. 101:20-78. 12. Messing, J., R. Crea, and P. H. Seeburg. 1981. A system for shotgun DNA sequencing. Nucleic Acids Res. 9:309-321. 13. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 14. Miyamoto, Y., T. Kato, Y. Obara, S. Akiyama, K. Takizawa, and S. Yamai. 1969. In vitro hemolytic characteristics of Vibrio parahaemnolyticus; its close correlation with human pathogenicity. J. Bacteriol. 100:1147-1149. 15. Moseley, S. L., and S. Falkow. 1980. Nucleotide seqeunce homology between the heat-labile enterotoxin gene of Escherichia coli and Vibrio cholerce deoxyribonucleic acid. J. Bacteriol. 144:444 446. 16. Moseley, S. L., I. Huq, A. R. M. A. Alim, M. So, M. SamadpourMotalebi, and S. Falkow. 1980. Detection of enterotoxigenic Escherichia coli by DNA colony hybridization. J. Infect. Dis. 142:892-898. 17. Nishibuchi, M., W. E. Hill, G. Zon, W. L. Payne, and J. B. Kaper. 1986. Synthetic oligodeoxyribonucleotide probes to detect Kanagawa phenomenon-positive Vibrio parahaemolyticus. J. Clin. Microbiol. 23:1091-1095. 18. Nishibuchi, M., M. Ishibashi, Y. Takeda, and J. B. Kaper. 1985. Detection of the thermostable direct hemolysin gene and related DNA sequences by the DNA colony hybridization test. Infect. Immun. 49:481-486. 19. Nishibuchi, M., and J. B. Kaper. 1985. Nucleotide sequence of the thermostable direct hemolysin gene of Vibrio parahaemolyticus. J. Bacteriol. 162:558-564. 20. Norrander, J., T. Kempre, and J. Messing. 1983. Improved M13 vectors using oligonucleotide-directed mutagenesis. Gene 26: 101-106.
VOL. 57, 1989
tdh-RELATED HEMOLYSIN GENE OF V. PARAHAEMOLYTICUS
21. Rosenberg, M., and D. Court. 1979. Regulatory sequences involved in the promotion and termination of RNA transcription. Annu. Rev. Genet. 13:319-353. 22. Sakazaki, R., K. Tamura, T. Kato, Y. Obara, S. Yamai, and K. Hobo. 1968. Studies on the enteropathogenic, facultatively halophilic bacteria, Vibrio parahaemolyticus. III. Enteropathogenicity. Jpn. J. Med. Sci. Biol. 21:325-331. 23. Sanger, F., S. Nicklen, and A. R. Coulson. 1977. DNA sequencing with chain-terminating inhibitors. Proc. Natl. Acad. Sci. USA 74:5463-5467. 24. Seidler, R. J., and M. Mandel. 1971. Quantitative aspects of deoxyribonucleic acid renaturation: base composition, state of chromosome replication, and polynucleotide homologies. J. Bacteriol. 106:608-614. 25. Southern, E. 1979. Gel electrophoresis of restriction fragments. Methods Enzymol. 68:152-176. 26. Takeda, Y. 1983. Thermostable direct hemolysin of Vibrio
2697
parahaemolyticus. Pharmacol. Ther. 19:123-146. 27. Tsunasawa, S., A. Sugihara, T. Masaki, F. Sakiyama, Y. Takeda, T. Miwatani, and K. Narita. 1987. Amino acid sequence of thermostable direct hemolysin produced by Vibrio parahaemolyticus. J. Biochem. 101:111-121. 28. Wallace, R. B., M. J. Johnson, S. V. Suggs, K. Miyoshi, R. Bhatt, and K. Itakura. 1981. A set of synthetic oligodeoxyribonucleotide primers for DNA sequencing in the plasmid vector pBR322. Gene 16:21-26. 29. Yamamoto, T., T. Nakazawa, T. Miyata, A. Kaji, and T. Yokota. 1984. Evolution and structure of two ADP-ribosylation enterotoxins, Escherichia coli heat-labile toxin and cholera toxin. FEBS Lett. 169:241-246. 30. Yanisch-Perron, C., J. Vieira, and J. Messing. 1985. Improved M13 phage cloning vectors and host strains: nucleotide sequences of the M13mpl8 and pUC19 vectors. Gene 33:103-119.
