Conservation of cholera toxin gene in a strain of cholera toxin non-producing Vibrio cholerae O1

FEMS Microbiology Letters 154 (1997) 111^116

Conservation of cholera toxin gene in a strain of cholera toxin non-producing Vibrio cholerae O1 Yasuko Honma a b

a

*, Masaaki Iwanaga

a ;b

Department of Bacteriology, Faculty of Medicine, University of the Ryukyus, Nishihara, Okinawa 903-01, Japan

Research Institute of Comprehensive Medicine, Faculty of Medicine, University of the Ryukyus, Nishihara, Okinawa 903-01, Japan

Received 14 April 1997; revised 14 June 1997 ; accepted 28 June 1997

Abstract

BT23, a Vibrio cholerae O1 El Tor isolate, possesses the cholera toxin (CT) gene as determined by PCR. However, CT was not detected in the culture medium by the reversed passive latex agglutination test, nor in the whole cell lysate as examined by Western blotting. The toxin-coregulated pilus (TCP) was not detected by Western blotting. This suggests the presence of defects in the regulatory cascade. toxR, toxS and toxT, members of the regulatory cascade, were examined by PCR. toxR and toxS were conserved but toxT was not. CT and TCP production was complemented by transformation of toxT. The lack of toxT was suspected to be the cause of the undetectable production of CT in strain BT23. Keywords : Vibrio cholerae

O1; Cholera toxin productivity; Polymerase chain reaction

1. Introduction

O1 [1]. The production of CT agreed with conservation of the ctx in 48 of the 49 strains. Only one strain, BT23, was positive in PCR and negative in RPLA in spite of culture in an AKI-SW culture condition, which has been shown to optimize cholera toxin production by El Tor strains [2]. Naturally occurring isolates of V. cholerae that did not produce CT have been reported to lack ctx and the rest of the core region [3]. Also non-toxinogenic strains of V. cholerae have been reported to lack the entire ctx genetic element and all V. cholerae, including nontoxinogenic strains, had an intact regulatory system for ctx genes [4]. These observations suggest the presence of a mutation in the ctx of strain BT23, producing an altered CT which did not react with anti-CT antiserum, or defects in the ctx regulating system. Therefore, we examined the DNA sequences of ctx and performed combined reversed transcripcholerae

Cholera is caused by the enterotoxin producing . When V. cholerae O1 is isolated, it is important to determine the production of cholera toxin (CT) by the isolate. To identify CT-producing strains, the isolates are routinely examined for the presence or absence of the CT-encoded gene (ctx) or production of CT in the culture medium. However, in general the existence of the structural gene does not always mean the expression of the gene products. We have reported the use of reversed passive latex agglutination (RPLA) and polymerase chain reaction (PCR) tests for identi¢cation of CT-producing V. Vibrio cholerae

* Corresponding author. Tel.: +81 (98) 895-3331; Fax: +81 (98) 895-2951; E-mail: [email protected]

0378-1097 / 97 / $17.00 ß 1997 Federation of European Microbiological Societies. Published by Elsevier Science B.V. PII S 0 3 7 8 - 1 0 9 7 ( 9 7 ) 0 0 3 1 0 - 8

FEMSLE 7738 25-10-97

112

Y. Honma, M. Iwanaga / FEMS Microbiology Letters 154 (1997) 111^116

tion-PCR (RT-PCR) for transcription into the mRNA of ctx. toxR, toxT and toxS, members of the toxR regulon [5,6], were also examined. 2. Materials and methods

2.1. Bacterial strains and plasmids

BT23, a strain with V. cholerae O1 biotype El Tor characteristics, was isolated from a cholera patient in Bangladesh. V. cholerae O1 El Tor strain 82P7, isolated from a cholera patient producing a high level of CT, was used as a control. For general molecular cloning, Escherichia coli XL1-Blue (Stratagene) and plasmid vector pBluescriptII KS‡ (Stratagene) were used to clone PCR products. 2.2. Media and culture conditions

The bacterial strains were cultured on nutrient agar plates overnight at 37³C to prepare PCR templates. For detection of CT, the organisms were cultured in AKI-SW [2] at 30³C in YEP medium (1.5% bactopeptone, 0.4% yeast extract, 0.5% NaCl). To collect the mRNA for RT-PCR, the organisms were cultured in AKI-SW with 4 h stationary culture and subsequent 4 h shaking. 2.3. Detection of CT in culture £uid

To detect CT in the culture £uid, the £uid was centrifuged at 14 000Ug for 10 min. The amount of CT in the supernatant was assayed by RPLA [2] and Chinese hamster ovary (CHO) cell assay [7]. 2.4. Electrophoresis and Western blotting

To detect proteins, SDS-PAGE and immunoblotting were carried out by the methods of Laemmli [8] and Towbin et al. [9], respectively. The PCR and RT-PCR products were detected by agarose gel electrophoresis. 10 Wl of each sample was applied on 1.0% agarose gel. 2.5. Antiserum

The rabbit anti-CT and anti-toxin coregulated pi-

lus (TCP) serum to detect the antigens was obtained from Japanese White rabbits immunized by multicutaneous injections of 100 Wg puri¢ed CT or TCP mixed with the same volume of Freund's complete adjuvant and several booster injections with Freund's incomplete adjuvant at 2 week intervals. The puri¢ed CT and TCP were previously prepared and the N-terminal amino acid sequences were determined to be identical to those already reported. The antiserum was highly speci¢c for CT in culture medium or TCP examined by Western blotting or electron microscopic immunogold labelling. 2.6. PCR primers

The primers used for PCR to detect or clone the structural genes are listed in Table 1. The set of primers AX2 and AX3 [10] (amplifying 564 bp) was used for detection of ctx. For the DNA sequencing of ctx, primers I, J, K and M were used additionally. The set of TX3 and TX4 amplifying 488 bp was used for detection of toxR, and primers TX5 and TX6 for amplifying the structural gene. The set of toxS1 and toxS3 amplifying 292 bp was used for detection of toxS, and the set of toxS1 and toxS4 for amplifying the structural gene. The set of toxT1 and toxT3 amplifying 472 bp and another set of toxT2 and toxT4 amplifying 410 bp were used for detection of toxT. T7 and T3 primers were used to determine the DNA sequence of PCR products cloned in the plasmid pBluescriptII KS‡ . 2.7. PCR conditions and sequencing of the PCR products

For detection of ctx, the conditions used were as previously reported [10]. For toxR, toxS, and toxT, the template DNAs were denatured at 94³C for 1 min, annealed with the primers at 52³C for 1 min, and extended at 72³C for 1 min for a total of 25 cycles. To prepare templates, the bacterial colonies were suspended in sterile distilled water and boiled for 10 min. The DNA sequences of the PCR products were determined using the Applied Biosystems Dye Deoxy1 Terminator Cycle Sequencing Kit and ABI 373 DNA Sequencer (ABI).

FEMSLE 7738 25-10-97

Y. Honma, M. Iwanaga / FEMS Microbiology Letters 154 (1997) 111^116

2.8. Preparation of total RNA and RT-PCR conditions

113

and TCP in the whole cell lysate was detected by RPLA and Western blotting, respectively.

The total RNA for RT-PCR was puri¢ed with

1

RNAzol

(Biotecx). Reversed transcription was car-

3. Results

ried out at 42³C for 15 min using 10 U of reverse

3.1. CT production and its gene (ctx)

transcriptase, Superscript (Stratagene), in 10 mM of Tris-HCl (pH 8.3) containing 2 mM MgCl2 , 50 mM KCl, 30

Wg total RNA, 15 WM each of four

Strain BT23 did not produce CT as examined by

dNTPs, and 100 pM of the primer set AX2 and

RPLA, CHO cell assay [1] and Western blotting with

AX3. Subsequently, cDNA was ampli¢ed for 30

anti-CT rabbit serum (Fig. 1). The concentrated

cycles using 2 U of Tth DNA polymerase (Toyobo)

supernatant

in the same bu¡er as above. Each cycle consisted of

with antibody, but the pattern was di¡erent from

96³C for 1 min, 42³C for 1 min, and 54³C for 1 min,

that of control strain 82P7 CT (Fig. 1). The whole

and the reaction mixture was incubated at 60³C for

cell lysate of strain BT23 did not react with anti-CT

10 min.

antibody in Western blotting (data not shown). The

ctx

2.9. Complementation of toxT

revealed some cross-reacting proteins

of BT23 was detected by PCR [1]. The DNA

sequence revealed two base replacements in

ctx

re-

sulting in one amino acid replacement at the 96th The structural gene of

toxT

from 82P7 was ampli-

amino acid position in the A subunit (Asn to Lys)

¢ed by PCR using the primers toxT1 and toxT4. The

compared with the reported sequence [11] (accession

product was recombined into the vector and the

number D84337). The sequence of

cloned

toxT

was transformed into BT23 by electro-

ctxB was identical

to that of group 3 strains, the group of El Tor strains

poration. The transformant (BT23/pTXT1) was cul-

distributed worldwide [12].

tured in AKI-SW, and CT in the culture medium

Table 1 DNA sequences of oligonucleotide primers used for PCR

C

Gene

Primer

DNA sequence (5P

ctx-F ctx-R ctx-F ctx-R ctx-F ctx-R

AX2

cgggcagattctagacctcctg

3P)

Position 73 to 94

[11]

AX3

cgatgatcttggagcattcccac

636 to 614

I

taaacaaagggagcattat

3

[11]

J

tctcatcatcgaaccac

1197 to 1181

19 to

3

Reference

1

[11] [11]

K

acacctcaaaatattactga

837 to 856

[11]

M

tgcggcaatcgcatgaggcg

1130 to 1111

[11]

toxR-F toxR-R toxR-F toxR-R

TX3

ccgaattctttggctgctggcccaacgtcc

156 to 185

[14]

TX4

cgccatcgacaaccgttagggg

643 to 622

cagggagatactgggaca

3

[14]

TX5

5

[14]

TX6

cccatggcgatgtgtcta

920 to 903

[14]

toxS-F toxS-R toxS-R

toxS1

ggatcttgctatgcaaa

3

[15]

toxS3

aatgacgctttctgcac

282 to 266

[15]

toxS4

ttaagaattactgaac

522 to 507

[15]

toxT-F toxT-F toxT-R toxT-R

toxT1

gtagaacgcaatgattg

3

10 to 7

[16]

toxT2

cagatgagttcctaaaa

419 to 435

[16]

toxT3

tttcgagaagaaccctg

462 to 446

[16]

toxT4

tttttctgcaactcctg

828 to 812

[16]

F, forward (sense strand) ; R, reverse (antisense strand).

FEMSLE 7738 25-10-97

22 to

3

10 to 7

Y. Honma, M. Iwanaga / FEMS Microbiology Letters 154 (1997) 111^116

114

Fig. 1. Western blotting analysis for detection of CT antigen in AKI-SW culture medium. Fifty times concentrated supernatant samples did not react with anti-CT serum. Two thousand times concentrated sample reacted with the anti-CT rabbit serum, but the pattern was not the same as for strain 82P7 CT(82P7). M, molecular (

U

100) ;

size

marker.

second

goat IgG (

U

First

antibody,

antibody,

peroxidase

anti-CT

labeled

rabbit

serum

anti-rabbit

IgG

Fig. 2. Western blotting analysis of TCP antigen in

V. cholerae

O1 El Tor strains BT23, 82P7 and BT23/pTXT1. In the whole cell lysate of strains 82P7 and BT23/pTXT1, TCP antigen was

1000).

detected as a 20.5 kDa protein, but it was not detected in strain BT23. Strain BT23/pTXT1 was cultured until the late exponential phase, and incubated for 1 h more with 2 mM IPTG. First antibody, anti-TCP rabbit serum (

3.2. Analysis of TCP production

U

100) ; second antibody, peroxi-

dase labeled anti-rabbit IgG goat IgG (

U

1000).

The organisms were cultured in AKI-SW to increase the production of TCP [13]. Whole cell lysates

strain 82P7 by PCR (data not shown). The DNA

of BT23 and 82P7 were applied to Western blotting

sequence of

with the anti-TCP rabbit antibody. TCP antigen was

nucleotides as compared with the sequence reported

toxR

showed 14 replacements of mono-

detected in strain 82P7 but not in strain BT23 (Fig. 2).

Table 2 Replacements of mononucleotides accompanied with amino acid

3.3. RT-PCR analysis of ctx translation

replacement in BT23

toxR

Amino acid number

The transcription of

ctx

to mRNA was examined

by RT-PCR (Fig. 3). The mRNA of

ctx

3.4. Analysis of the regulatory genes toxR, toxS and toxT, and complementation

b BT23 a

Ser

agc

569B

Ala

gcg

b BT23 a

Thr

acg

184

569B

Phe

ttt

b BT23 a

Leu

ctt

204

569B

Ser

tcc

Thr

acc

BT23

a b

toxR

was detected in strain BT23 as well as in

Nucleotide aac

172

of strain 82P7. We concluded that there was a regulatory failure in BT23.

Amino acid Asn

569B

was positive

in a trace amount in strain BT23 compared with that

a

134

b

Miller and Mekalanos [4]. The nucleotide sequences have been deposited in the DDBJ data

base and are available under accession number D84191.

FEMSLE 7738 25-10-97

Y. Honma, M. Iwanaga / FEMS Microbiology Letters 154 (1997) 111^116

115

previously [14], resulting in four amino acid replacements (Table 2). The promoter area of

toxR

was

identical to the reported sequence [14], except for one nucleotide deletion.

toxS

was detected in strain BT23 as well as in

toxS

strain 82P7 (Fig. 4). The DNA sequence of

from strain BT23 was identical to the reported sequence [15].

toxT was detected in strain 82P7 but not

in strain BT23 with two sets of primers (Fig. 4) even when the annealing temperature was reduced to 48³C (data not shown). The DNA sequence of the PCR product of

toxT

from strain 82P7 was identical to the reported sequence [16] except for one nucleotide replacement. The

transformant

toxT-recombined

of

strain

BT23

containing

the

plasmid (BT23/pTXT1) produced

Fig. 4. PCR analysis for detection of

ae

toxS

and

toxT

in

V. choler-

O1 El Tor strains BT23 and 82P7. Lanes 1, 2, the PCR prod-

uct of

toxS.

Lanes 3, 4, the PCR product of

and toxT3) ; lanes 5, 6, the PCR product of

toxT toxT

(primer toxT1 (primer toxT2

and toxT4). Lanes 1, 3, 5, BT23 ; lanes 2, 4, 6, 82P7. M, DNA size marker.

CT in culture medium as examined by RPLA (100 1 ng ml ), and TCP in whole cell lysate after induc-

3

tion with 2 mM IPTG (Fig. 2).

4. Discussion

V. cholerae O1 El Tor strain, appeared ctx by PCR but CT was not detected. examined by RT-PCR, the ctx was slightly tran-

BT23, a

positive for As

scribed into mRNA (Fig. 3). ToxS stimulates the activity of ToxR [15]. ToxR regulates virulence gene expression by activating the expression of

toxT,

and ToxT directs the coordinate

expression of CT and TCP [15]. Since neither TCP nor CT was detected in strain BT23, a defect was considered [4,14]. However,

toxR toxR

regulon was de-

tected in strain BT23 by PCR as well as in strain 82P7, and sequencing of Fig. 3. RT-PCR analysis to detect mRNA of

ctx.

Cells in late

to

cDNA

and

ampli¢ed

with

ctx

PCR.

was reverse transcribed

Lane

M,

molecular

size

marker ; lane 1, strain BT23 ; lane 2, strain 82P7 ; lane 3, negative control (RT-PCR for

ctxA

of 82P7 without reverse transcrip-

tase). In strain BT23, the amount of than in strain 82P7.

ctx

mRNA was smaller

revealed only some

Examined by PCR, strain BT23 lacked the

exponential phase with AKI-SW were collected and the total mRNAs were puri¢ed. The mRNA of

toxR

replacements of mononucleotides (Table 2).

toxT

gene (Fig. 4). Examined by the colony hybridization test, strain BT23 did not react with the

toxT

probe

(pTXT1) either (data not shown). Complementation of

toxT suggested that the lack of toxT was the cause

of the undetectable production of CT in strain BT23.

FEMSLE 7738 25-10-97

Y. Honma, M. Iwanaga / FEMS Microbiology Letters 154 (1997) 111^116

116

Strain BT23 was isolated from a cholera patient. However, the data in this report suggest that strain BT23

was

hypotoxinogenic.

As

expected,

strain

[6] Miller, V.L., DiRita, V.J. and Mekalanos, J.J. (1989) Identi¢cation of

toxS,

a regulatory gene whose product enhances

ToxR-mediated activation of the cholera toxin promoter. J. Bacteriol. 171, 1288^1293.

BT23 did not cause £uid accumulation in the rabbit

[7] Honda, T., Shimizu, M., Takeda, Y. and Miwatani, T. (1976)

intestinal loop (data not shown). The reason why the

Isolation of a factor causing morphological changes of chinese

hypotoxinogenic strain BT23 was isolated from the

hamster ovary cells from the culture ¢ltrate of

patient is not clear. It might have been isolated from other toxinogenic strains by chance. Of course, it cannot be excluded that

toxT

was lost during sub-

culturing.

haemolyticus.

Vibrio para-

Infect. Immun. 14, 1028^1033.

[8] Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227, 680^685. [9] Towbin, H., Staehelin, T. and Gordon, J. (1979) Electropho-

V. cholerae have been reported to lack ctx, and all V. cholerae, including non-toxinogenic strains, have toxR [3,4]. An intact regulatory system for ctx has been present Most CT non-producing isolates of

retic transfer of proteins from polyacrylamide gels to nitrocellulose sheets :

procedure and source applications. Proc.

Natl. Acad. Sci. USA 76, 4350^4354. [10] Fields, P.I., Popovic, T., Wachsmuth, I.K. and Olsvic, O. (1992) Use of polymerase chain reaction for detection of toxi-

Vibrio cholerae

in non-toxinogenic strains as well as in toxinogenic

genic

strains [4]. This is the ¢rst report of a naturally oc-

epidemic. J. Clin. Microbiol. 30, 2118^2121.

curring hypotoxinogenic strain with defects in the regulatory system for

ctx.

strains from the Latin American cholera

[11] Mekalanos, J.J., Swarts, D.J., Pearson, G.D., Harford, N., Groyne, F. and Wilde, M. (1983) Cholera toxin genes : nucleotide sequence, deletion analysis and vaccine development. Nature 306, 551^557. [12] Olsvik, Ò., Wahlberg, J., Petterson, B., Uhlen, M., Popovic,

References

T., Wachsmuth, I.K. and Fields, P.I. (1993) Use of automated sequencing of polymerase chain reaction-generated amplicons

[1] Honma, Y., Higa, N., Tsuji, T. and Iwanaga, M. (1995) Comparison of a reversed passive latex agglutination and a polymerase chain reaction for identi¢cation of cholera toxin producing

Vibrio cholerae

O1. Microbiol. Immunol. 39, 59^61.

[2] Iwanaga, M., Yamamoto, K., Higa, N., Ichinose, Y., Nakasone, N. and Tanabe, M. (1986) Culture condition for stimulating cholera toxin production by

Vibrio cholerae O1 El Tor.

Microbiol. Immunol. 30, 1075^1083. [3] Kaper, J.B., Moseley, S.L. and Falkow, S. (1981) Molecular characterization of environmental and nontoxigenic strains of

V. cholerae.

Infect. Immun. 32, 661^667.

[4] Miller, V.L. and Mekalanos, J.J. (1984) Synthesis of cholera toxin is positively regulated at the transcriptional level by

toxR.

Proc. Natl. Acad. Sci. USA 81, 3471^3475.

[5] DiRita, V.J., Parsot, C., Jander, G. and Mekalanos, J.J. (1991) Regulatory cascade controls virulence in

ae.

Vibrio choler-

to identi¢ed three types of cholera toxin subunit B in

cholerae

Vibrio

O1 strains. J. Clin. Microbiol. 31, 22^25.

[13] Waldor, M.K. and Mekalanos, J.J. (1994) ToxR regulates virulence gene expression in non-O1 strains of

Vibrio cholerae

that cause epidemic cholera. Infect. Immun. 62, 72^78. [14] Miller, V.L., Tayler, R.K. and Mekalanos, J.J. (1987) Cholera toxin

transcriptional

activator

ToxR

is

a

transmembrane

DNA binding protein. Cell 48, 271^279. [15] DiRita, V.J. and Mekalanos, J.J. (1991) Periplasmic interaction between two membrane regulatory proteins, ToxR and ToxS, results in signal transduction and transcriptional activation. Cell 64, 29^37. [16] Higgins, D.E., Nazareno, E. and DiRita, V.J. (1992) The virulence gene activator toxT from

Vibrio cholerae

is a member

of the araC family of transcriptional activators. J. Bacteriol. 174, 6974^6980.

Proc. Natl. Acad. Sci. USA 88, 5403^5407.

FEMSLE 7738 25-10-97