Comparison of global transcription responses allows identification of Vibrio cholerae genes differentially expressed following infection

FEMS Microbiology Letters 190 (2000) 87^91

www.fems-microbiology.org

Comparison of global transcription responses allows identi¢cation of Vibrio cholerae genes di¡erentially expressed following infection Soumita Das a , Amit Chakrabortty a;1 , Rajat Banerjee a , Susanta Roychoudhury b , Keya Chaudhuri a; * a

Biophysics Division, Indian Institute of Chemical Biology, 4, Raja S.C. Mullick Road, Jadavpur, Calcutta 700 032, India b Human Genetics Department, Indian Institute of Chemical Biology, Jadavpur, Calcutta 700 032, India Received 18 April 2000 ; received in revised form 18 June 2000; accepted 1 July 2000

Abstract Comparison of global transcription profiles of Vibrio cholerae grown in vitro and in vivo revealed that 20% of the genome was repressed and about 5% was induced under in vivo conditions. Hybridization with the cloned genes revealed that the virulence genes ctx, toxR, toxT and tcpA were induced under in vivo conditions. Dissection of two in vivo induced cosmids identified another set of three genes homologous to cheY1 involved in motility and chemotaxis, pnuC encoding the major component of the nicotinamide mononucleotide transport system and icmF belonging to a cassette involved in multiplication inside host cells. These results demonstrate that the global transcription profile approach might be a powerful method for identification of differentially expressed transcripts under in vivo conditions. ß 2000 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Global transcription pro¢le ; In vivo expressed gene; Vibrio cholerae

1. Introduction Cholera still remains a major public health problem in developing countries as well as in some developed countries. The causative organism Vibrio cholerae, is a Gramnegative curved rod and the life threatening watery diarrhea is largely due to the action of the secreted cholera toxin (CT) on the epithelium of the small intestine. The so far identi¢ed virulence factors of this pathogen include other toxins like Ace, Zot and colonizing factors like TCP [1]. The exact regulatory mechanism of the major virulence genes including ctx and tcp is yet to be conclusively established. The current model of the regulation of virulence factor expression in V. cholerae postulates that ToxR and TcpP become activated in response to particular environmental signals and act cooperatively to induce expression of ToxT [2]. ToxT in turn directly activates transcription of biosyn* Corresponding author. Tel. : +91 (33) 473-0350; Fax: +91 (33) 473-5197/0284; E-mail : [email protected] 1 Present address: Department of Biochemistry and Molecular Biology (MC536) College of Medicine, A-312 College of Medicine West, 1819 West Polk Street, Chicago, IL 60612-7334, USA.

thetic genes for CT, TCP [3]. In the intestinal lumen, where the pathogen colonizes during infection, conditions are such, that the expression of ToxR controlled virulence factors is repressed [4,5]. Thus to induce ToxR-controlled virulence genes in vivo, V. cholerae may recognize other unknown external signals in the host environment [6]. Recently using tnpR operon fusions and signature tagged mutagenesis (STM) to conduct a screen for random insertion mutations, V. cholerae genes induced within the host and critical for colonization were isolated [7,8]. Only one gene in the tcp operon was isolated suggesting that one third of all the genes responsible could be identi¢ed by the STM method. By these methods, involving transposon mutagenesis, only genes which are induced in vivo are identi¢ed. However, the genes that are repressed may hold some major information regarding the mechanism of pathogenesis of the organism. We have used the global transcription response technique to identify regions of the V. cholerae genome that are di¡erentially expressed following infection. By this technique, detection, cloning, and mapping of a responding gene or co-regulated gene in response to environmental signals can be achieved. This has been used to identify and map transcriptionally active regions of the Haloferax volcanii genome [9] and to study mRNA levels in an over-

0378-1097 / 00 / $20.00 ß 2000 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 1 0 9 7 ( 0 0 ) 0 0 3 2 6 - 8

FEMSLE 9550 10-8-00

88

S. Das et al. / FEMS Microbiology Letters 190 (2000) 87^91

lapping set of clones in V vectors of the Escherichia coli K12 under various environmental stimuli [10]. In this report we describe the identi¢cation of some of the functions that are di¡erentially expressed following infection. 2. Materials and methods 2.1. Bacterial strains and plasmids The V. cholerae 569B and E. coli DH5K strains used in this study were maintained at 370³C in Luria^Bertani (LB) medium containing 20% (v/v) glycerol. E. coli cells were grown in LB medium and V. cholerae cells were grown in nutrient Broth or LB medium. All cloning experiments were done in pBluescript KS+. The plasmids were maintained and ampli¢ed in E. coli DH5K and XL1 Blue strains. Ampicillin was used at 100 Wg ml31 for E. coli and 50 Wg ml31 for V. cholerae. The 92 sets of cosmid clones in Lorist M vector [11] were grown in TB medium containing phosphate bu¡er. Ampicillin (100 Wg ml31 ) and kanamycin (30 Wg ml31 ) were used where appropriate. 2.2. Ligated rabbit ileal loop model In vivo growth of V. cholerae strains were carried out in the rabbit ileal loop model essentially as described previously [12]. Brie£y, 48-h fasted rabbits were anesthetized and the small intestine was tied into consecutive 6-cm and 2-cm segments proximally to the mesoappendix. An inoculum of 0.5 ml of V. cholerae strains containing about 5U106 CFU was introduced into each 6-cm segment, while one loop was inoculated with saline as negative control. After 16^18 h the animals were killed and the small intestine removed. Fluid accumulated in each loop was separately collected and measured. V. cholerae count in the £uid was measured by plating on thiosulfate- citratebile salt-sucrose (TCBS) agar plates. Although no £uid accumulated in the loops inoculated with saline, these loops were also washed and scraped and CFU were assayed in the washes to determine whether the intestine contained any inhabitant. 2.3. Isolation of in vivo bacteria Fluid accumulated from each loop was separately collected and spun at 1000 rpm for 5 min to remove the debris. The supernatant was then centrifuged at 5000 rpm for 10 min to pellet the bacteria. 2.4. RNA isolation and analysis For isolation of RNA cells were grown to the exponential phase in LB pH 7.2 or overnight in rabbit intestine. Total RNA was extracted and puri¢ed by using guanidi-

um isothiocyanate as described earlier [12]. RNA samples (15^25 Wg well31 ) were electrophoresed in duplicate in 1% agarose^2.1 M formaldehyde-morpholinepropanesulfonic acid gels, and one part was stained with ethidium bromide and visualized with UV light to con¢rm equal loading of all samples. The other part of the gel was blotted onto nylon membranes by using 20USSC and hybridized with labeled probes prepared by the random priming method. The desired DNA fragment was labeled using [K-32 P]dCTP (speci¢c activity 600 Ci nmol31 , Amersham, UK). In 100 ng of heat-denatured DNA, 1 Wl random hexamer, 1.2 Wl of 0.5 mM dNTP (except dCTP), 4 Wl of 50 mM DTT, 5 Wl of [K-32 P]dCTP and 4 Wl 10UKlenow bu¡er was added. DNase free water was added to make the volume 40 Wl. 1 Wl (0.5 U) DNA polI Klenow fragment was added and the reaction mixture was incubated at 37³C for 1 h. The reaction was stopped by EDTA to a concentration of 0.5 mM. Labeled DNA was puri¢ed from unincorporated [K-32 P]dCTP by passing through a Sephadex G-50 column equilibrated with STE bu¡er. 2.5. Preparation of K-32 P-labeled total cDNA The random hexamer priming method [10] was used to prepare labeled cDNA. The reaction mixture contains 15 Wg of total RNA, 5 Wg of pd(N)6 (random hexamer from NEB), 0.5 mM dATP, 0.5 mM dGTP, and 0.5 mM dTTP, 5 WM dCTP, 100 WCi of [K-32 P]dCTP (3000 Ci/mmol), 50 mM Tris^HCl (pH 8.3), 6 mM MgCl2 , 40 mM KCl, 0.5 U of RNase block (Stratagene) per Wl and 50 U of superscript II RNase H3 reverse transcriptase (Life Technologies) in a total volume of 50 ml. The reaction was allowed to continue overnight at room temperature. EDTA and NaOH were then added to ¢nal concentrations of 50 mM and 0.25 M, respectively, and the mixture was incubated at 65³C for 30 min to degrade the RNA templates. The cDNA was then ready to use after neutralization by adding HCl and Tris bu¡er. Free nucleotides were not removed. Usually more than half of the isotope was incorporated. 2.6. Dot blot hybridization A set of 92 cosmid DNAs were denatured by boiling at 100³C for 5 min and with subsequent chilling at 4³C. Individual samples were spotted on a membrane ¢lter and the membrane was dried and baked in a vacuum desiccator at 70³C for 2^3 h. For hybridization with appropriate probes, the blots were prehybridized for 6 h at 60³C in the prehybridization solution and hybridized overnight with K-32 P-labeled DNA fragment at 60³C in 6USSC and 0.5% SDS. The blots were washed successively (15 min each at room temperature) in 2USSC, 0.5% SDS and in 2USSC, 0.1% SDS with a ¢nal stringency wash in 1USSC, 0.5% SDS at 60³C for 2 h. The blots were dried and exposed to Kodak XR-5 ¢lm using an intensifying

FEMSLE 9550 10-8-00

S. Das et al. / FEMS Microbiology Letters 190 (2000) 87^91

89

screen or to a phosphorimager (Bio-Rad, USA) for analysis. 2.7. Subcloning and DNA sequencing Selected cosmids were digested with BamHI and HindIII and cloned into BamHI^HindII-digested pBluescript. The nucleotide sequence was determined with double stranded plasmid DNA as the template by cycle sequencing in a ABI prism 377 DNA sequencer (Applied Biosystems, Perkin-Elmer) using company supplied kit reagents and protocols. 2.8. Nucleotide sequence and accession number The nucleotide and deduced amino acid sequences of the gene fragments appear in EMBL, GenBank, DDBJ databases under the accession numbers AF171076, AF239736 and AF239738. Fig. 2. A: Northern blot analysis of toxR (A), toxT (B), tcpA (C), ctxAB(D). B: Northern blot analysis of icmF(E), pnuC(F) and cheY1(G). L and R represent in vitro and in vivo RNA.

3. Results and discussion High molecular mass RNA was isolated from V. cholerae 569B cells grown overnight in rabbit ileal loop (in vivo) and in rich media (in vitro) conditions. K32 P-Labeled cDNA was prepared by reverse transcription primed with mixed hexanucleotides and hybridized with dot blots of a minimal set of 92 cosmid clones representing about 90% of the V. cholerae genome. The minimal set of cosmid clones were taken from the cosmid library constructed in the Lorist M vector by Chatterjee et. al. [11] in our laboratory. The control and experimental hybridizations were performed at least thrice by reusing the same blot. The result-

Fig. 1. Transcription pro¢le of a minimal set of 92 cosmid clones showing regions of the V. cholerae 569B genome that are di¡erentially expressed under in vivo conditions. The left panels represent the hybridization patterns under in vitro (top) and in vivo (bottom) conditions. The right panels show the base line transcription pro¢le under in vitro conditions (top) and induction and repression ratios under in vivo conditions (bottom).

ing data were analyzed in a phosphorimager (Bio-Rad, USA) instrument. The percentage of total counts for each spot was obtained directly from the machine after subtracting the background level, which was taken from the blank area of each blot (Fig. 1). Each cosmid clone contained an insert of V. cholerae genomic DNA between 20 and 40 kb. The restriction map of each clone was veri¢ed with the original data [11], to check for deletions. The clones containing rrn operons were excluded as they gave saturated values and caused errors during calculations. From the hybridization patterns (Fig. 1), ratios of in vivo to in vitro counts were calculated and plotted in Microsoft Excel (Fig. 1). We regarded induction or repression ratios of s 2 and 6 0.5 as signi¢cant. The same blot was used with stripping to reduce errors. The experiments were repeated with at least ¢ve di¡erent batches of RNA. In comparison to baseline expression, some clones showed di¡erential expression when the bacteria were grown in the animal host. About 20% of the clones showed repressed expression under in vivo conditions, and about 5% were induced. It is not surprising that most of the clones showed a repression in vivo, as in the stressful environment within the host it is possible that very few proteins were allowed to be synthesized at a high rate. This helps the pathogen to make proper utilization of the limited resources in vivo and synthesize proteins essential for virulence and survival within the host. However, the cosmid clones containing 20^40 kb of V. cholerae DNA need to be further dissected for ¢ne tuning of the expression pro¢le. As expected the clones containing the virulence cassettes ctxAB, zot, ace, orfU, cep showed an induction in the level of expression in vivo. So did the toxR and tcpAB, toxT

FEMSLE 9550 10-8-00

90

S. Das et al. / FEMS Microbiology Letters 190 (2000) 87^91

Table 1 Clones from the V. cholerae genome induced in vivo Clone#/gene(s)

Homologous to

Tentative function

Reference

Sb1 Sb2 Sb3 toxR toxT ctxAB tcpA

PnuC protein of E. coli IcmF protein of L. pneumophila CheY1 protein of E.coli ToxR of V. cholerae ToxT of V. cholerae CtxAB of V. cholerae TcpA of V. cholerae

NMN transport Involved in intracellular multiplication Chemotaxis Virulence regulator Virulence regulator Enterotoxin Pilus required for colonization

Present study Present study Present study [20] [21] [22] [21]

containing clones. Since toxR tcpA, ctxAB, and toxT constitute the virulence cassette of the V. cholerae genome, we decided to investigate this further. Northern blot experiments with the cloned genes (laboratory collection) revealed that under in vivo conditions ctx, toxR, toxT and tcpA were induced, the expression of tcpA was found to be always higher compared to the other three (Fig. 2). Very recently using RIVET technology it has been suggested that tcpA induction is biphasic and is required both at the initial and later stages of infection [13] which might be the cause of higher tcpA expression in our experiments. The clones showing an overall induced level of expression were digested with BamHI and HindIII and then were further cloned in pBScript KS+ to identify the genes responsible for the increased signal in vivo. Northern blot experiments were used to con¢rm the induction of these genes. Three in vivo induced segments identi¢ed using this strategy were sequenced and the database was searched for homology using the BLASTX [14] programme (Table 1). Interestingly, one of the in vivo induced segments identi¢ed in this study is homologous to cheY1 that relates to motility and chemotaxis in several Gram-negative enteric bacteria. Induction of such a gene is quite expected as it has been found that V. cholerae isolated from rabbit intestine is 350% more motile than those grown under normal laboratory conditions [12]. Some evidence has also suggested that motility and chemotaxis are prerequisites for full virulence of V. cholerae [15^17]. Since V. cholerae adheres to the mucous coat in the small intestine, these motility and chemotaxis genes may play an important role in penetration of the mucous coat. The presence of the icmF gene in V. cholerae and its induction within the host is a new mystery in the pathogenesis of this organism. In Legionella pneumophila, icmF belongs to the icm locus shown to be essential for virulence in guinea pigs and for intracellular multiplication of Legionella within human macrophages [18]. Although intracellular multiplication has not been documented in non invasive V. cholerae 569B, it is still exciting as we have recently shown that V. cholerae 569B grown under in vivo conditions becomes resistant to human serum [12] in which icmF might play an important role. The increased expression of pnuC, the major component of the nicotinamide mononucleotide (NMN) transport system under in vivo conditions in V. cholerae might be re-

quired for survival of V. cholerae in the anaerobic environment prevailing inside rabbit intestine. Under anaerobic conditions the NAD pool goes down [19] and increased NMN transport will help the organism to synthesize more NAD. Including this report, in the last 5 years there has been at least ¢ve di¡erent approaches to identify di¡erential expression and regulation of V. cholerae genes during growth in vivo [7,8,12,13]. Our results suggest that with the sequencing of V. cholerae genome nearing completion, a genomic approach towards identi¢cation of in vivo expressed genes in this organism using DNA microarrays would possibly yield more detailed information necessary to understand its complex mechanism of pathogenesis. Acknowledgements This work was supported by the research Grant SP/SO/ D-56/96 from the Department of Science and Technology, Government of India. A.C is grateful to the Council of Scienti¢c and Industrial Research for a research fellowship. We thank S.N. Dey and I. Guhathakurta for excellent technical support.

References [1] Kaper, J.B., Morris Jr., J.G. and Levine, M.N. (1995) Cholera Clin. Microbiol. Rev. 8, 48^86. [2] Hase, C.C. and Mekalanos, J.J. (1998) TcpP protein is a positive regulator of virulence gene expression in Vibrio cholerae. Proc. Natl. Acad. Sci. USA 95, 730^734. [3] DiRita, V.J., Parsot, C., Jander, G. and Mekalanos, J.J. (1991) Regulatory cascade controls virulence in Vibrio cholerae. Proc. Natl. Acad. Sci. USA 88, 5403^5407. [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] Miller, V.L. and Mekalanos, J.J. (1988) A novel suicide vector and its use in construction of insertion mutations: osmoregulation of outer membrane proteins and virulence determinants in Vibrio cholerae requires toxR. J. Bacteriol. 170, 2575^2583. [6] Parsot, C. and Mekalanos, J.J. (1990) Expression of ToxR, the transcriptional activator of virulence factors of Vibrio cholerae, is modulated by the heat shock response. Proc. Natl. Acad. Sci. USA 87, 9898^9902. [7] Camilli, A. and Mekalanos, J.J. (1995) Use of recombinase gene

FEMSLE 9550 10-8-00

S. Das et al. / FEMS Microbiology Letters 190 (2000) 87^91

[8]

[9]

[10]

[11]

[12]

[13]

[14]

fusions to identify Vibrio cholerae genes induced during infection. Mol. Microbiol. 18, 671^683. Chiang, S.L. and Mekalanos, J.J. (1998) Use of signature-tagged transposon mutagenesis to identify Vibrio cholerae genes critical for colonization. Mol. Microbiol. 27, 797^805. Trieselmann, B.A. and Charlebois, R.L. (1992) Transcriptionally active regions in the genome of the archaebacterium Haloferax volcanii. J. Bacteriol. 174, 30^34. Chuang, S.E., Daniels, D.L. and Blattner, F.R. (1993) Global regulation of gene expression in Escherichia coli. J. Bactriol. 175, 2026^ 2036. Chatterjee, S., Mandal, A.K., Begum, N.A., Roychoudhury, S. and Das, J. (1998) Ordered cloned DNA map of the genome of V. cholerae strain 569B and localization of genetic markers. J. Bacteriol. 180, 901^908. Chakrabortty, A., Das, S., Majumdar, S., Mukhopadhyay, K., Roychoudhury, S. and Chaudhuri, K. (2000) Use of RAP-PCR ¢ngerprinting to identify Vibrio cholerae genes di¡erentially expressed in host following infection. Infect. Immun., in press. Lee, S.M., Hava, D.L., Waldor, M.K. and Camilli, A. (1999) Regulation and temporal expression patterns of Vibrio cholerae virulence genes during infection. Cell 99, 625^634. Altschul, S.F., Madden, T.L., Scha«¡er, A.A., Zhang, J., Zhang, Z., Miller, W. and Lipman, D.J. (1997) Gapped BLAST and PSI-BLAST : a new generation of protein database search programs. Nucleic Acids Res. 25, 3389^3402.

91

[15] Freter, R., O'Brien, P.C.M. and Macsai, M.M.S. (1981) Role of chemotaxis in the association of motile bacteria with intestinal mocosa in vivo studies. Infect. Immun. 34, 234^240. [16] Richardson, K. (1991) Roles of motility and £agellar structure in pathogenecity of Vibrio cholerae : analysis of motility mutants in three animal models. Infect. Immun. 59, 2727^2736. [17] Gardel, C.L. and Mekalanos, J.J. (1996) Alterations in Vibrio cholerae motility phenotypes correlate with changes in virulence factor expression. Infect. Immun. 64, 2246^2255. [18] Segal, G., Purcell, M. and Shuman, H.A. (1998) Host cell killing and bacterial conjugation require overlapping sets of genes within a 22 kb region of the Legionella pneumophila genome. Proc. Natl. Acad. Sci. USA 95, 1669^1674. [19] Leonardo, M.R., Dailly, Y. and Clark, D.P. (1996) Role of NAD in regulating the adhE gene of Escherichia coli. J. Bacteriol. 178, 6013^ 6018. [20] Waldor, M. and Mekalanos, J.J. (1994) ToxR regulates virulence gene expression in non -O1 Vibrio cholerae that cause epidemic cholera. Infect Immun. 62, 72^78. [21] Keasler, S.P. and Hall, R.H. (1993) Detecting and biotyping Vibrio cholerae O1 with multiplex polymerase chain reaction. Lancet 341, 1661. [22] Kaper, J.B., Lockman, H., Baldini, M.M. and Levine, M.M. (1984) A recombinant live oral cholera vaccine. BioTechnology 2, 345^ 349.

FEMSLE 9550 10-8-00