HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Western blots Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via CrossRef Citing Articles via Google Scholar Google Scholar Articles by Miyazato, T. Articles by Iwanaga, M. Search for Related Content PubMed PubMed Citation Articles by Miyazato, T. Articles by Iwanaga, M. Agricola Articles by Miyazato, T. Articles by Iwanaga, M.

Molecular analysis of VcfQ protein involved in Vibrio cholerae type IV pilus biogenesis

¹Department of Bacteriology, Faculty of Medicine, University of the Ryukyus, Uehara 207, Nishihara, Okinawa 903-0215, Japan ²Faculty of Health and Welfare Science, Okayama Prefectural University, Soja, Okayama 719-1197, Japan

Correspondence Tomoko Miyazato k008765{at}med.u-ryukyu.ac.jp

Received 9 May 2002 Accepted 23 December 2002

The nucleotide sequence of an ORF (vcfQ) within the type IV pilus gene cluster of Vibrio cholerae O34 strain NAGV14 was determined, thereby completing the sequence analysis of the structural operon. The vcfQ gene showed homology to the mshQ gene of the mannose-sensitive haemagglutinin pilus gene cluster. The vcfQ was 651 bp larger than mshQ, and the G+C content of the extra 651 bp portion (35.6 mol%) was lower than that of the overall vcfQ gene (42.5 mol%). Except for the first 270 aa residues, the deduced amino acid sequence of VcfQ showed high homology to the MshQ protein. There was immunological cross-reaction between VcfQ and MshQ by Western blotting. Cell fractionation studies showed that VcfQ is located in both the inner and the outer membranes. Mutational analysis showed that vcfQ-deficient mutant expressed detectable levels of major pilin (VcfA), but failed to assemble them into pili, indicating that VcfQ is essential for pilus assembly. Colony-blotting analyses showed that the N-terminal region of vcfQ is variable in V. cholerae strains.

The GenBank/EMBL/DDBJ accession number for the vcfQ sequence of V. cholerae strain NAGV14 is AB064660.

Figures showing the results of Western blotting analyses are available as supplementary data in JMM Online (http://jmm.sgmjournals.org).

Vibrio cholerae is known to be the causative agent of the acute gastrointestinal disease cholera. Strains of just two of the approximately 200 currently known O-antigen serogroups, O1 and O139, produce cholera toxin and cause cholera disease symptoms (Faruque et al., 1998; Yamai et al., 1997). However, sporadic cases of gastroenteritis including cholera-like diarrhoea due to organisms with serogroups that are non-O1/non-O139 have been reported (Bhattacharya et al., 1998; Dalsgaard et al., 1999; Sharma et al., 1998). In the process of infection, the initial events are attachment of V. cholerae to the intestinal epithelium and colonization of the intestine by the organisms. Therefore, in studies of pathogenic mechanisms and in the efforts to develop vaccines, much attention has been focused on these events.

Type IV pili are flexible appendages which are found in a number of important pathogenic bacteria that affect humans and animals. Expression of type IV pili is associated with colonization of epithelial cells, twitching motility of the organisms and biofilm formation on a variety of surfaces (Strom & Lory, 1993). Several kinds of type IV pili of V. cholerae have been isolated, purified and characterized. The toxin-coregulated pilus (TCP) (Taylor et al., 1987) has been recently identified as a coat protein of a filamentous bacteriophage (Karaolis et al., 1999). Expression of a mannose-sensitive haemagglutinin (MSHA) pilus that promotes adherence to zooplankton and biofilm formation appears to be important for survival in an aquatic environment (Chiavelli et al., 2001; Watnick & Kolter, 1999). The MSHA pilus is encoded in a locus that contains 16 genes organized in secretory and structural operons (Marsh & Taylor, 1999). In the structural operon, five genes encode prepilin-like proteins and two encode putative outer-membrane (OM) proteins (Marsh & Taylor, 1999).

In our previous studies, several kinds of pili of V. cholerae were purified and characterized (Iwanaga et al., 1989; Nakasone & Iwanaga, 1990; Yamashiro et al., 1993, 1996). Among them, only the pili isolated from strain NAGV14 adhered to rabbit intestinal epithelium (Yamashiro et al., 1996). The major pilin structural gene of the NAGV14 pilus (vcfA) was highly homologous to the MSHA pilin gene in the N-terminal region, but there was no homology in the C-terminal region (Kuroki et al., 2001). A subsequent study demonstrated that NAGV14 pilus gene cluster contained five ORFs that encoded prepilin-like proteins and corresponded to mshB, A, C, D, O (Toma et al., 2002). An ORF that encoded a putative OM protein and corresponded to mshP (Marsh & Taylor, 1999; Toma et al., 2002) was also demonstrated. Considering this similarity of the both gene cluster organization in the two organisms, an orthologous protein (same functional protein) (Alm et al., 2000) of MshQ might exist in the NAGV14 pilus structural operon. In the present study, the vcfQ gene encoding the putative MshQ-orthologous protein was investigated to clarify the gene function and to complete the nucleotide sequence of the NAGV14 pilus structural operon. The distribution of the gene in V. cholerae isolates was also studied.

METHODS

Bacterial strains, vector and media.

The strains and plasmids used in this study are listed in Table 1. Other strains stocked in our laboratory were also used in colony blotting. The organisms were cultured in Luria–Bertani (LB) broth [1 % tryptone (Difco), 0.5 % yeast extract (Difco), 0.5 % NaCl (nacalai tesque)] or on LB agar plates that were supplemented, as necessary, with 100 µg ampicillin ml^–1, 5 µg kanamycin ml^–1 and/or 5 or 25 µg chloramphenicol ml^–1.

DNA manipulation.

Chromosomal DNA was isolated by using QIAGEN Genomic-tip (Qiagen). Plasmid DNA was isolated according to the method described by Birnboim & Doly (1979) or by using Qiagen resin columns. Restriction enzyme digestion, ligation, gel electrophoresis, and transformation of DNA were carried out as described by Bagdasarian & Bagdasarian (1994) or electroporation in a 0.1 cm cuvette, using the Gene Pulser II Electroporation System (Bio-Rad) set at 0.75 kV. PCR products were purified on a GFX column (Amersham).

Sequencing and cloning of vcfQ.

For sequencing and cloning of vcfQ, construction of the primers and PCR were performed as shown in Fig. 1. First, primers mshQ3 and mshQ4 were designed on the basis of the mshQ sequence previously reported by Marsh & Taylor (1999), and the PCR product was named fragment-1. The nucleotide sequence of this fragment was determined, and new primers, mshQ10 and LA3 were constructed. For amplification of fragment-2, fragment-3 and fragment-4, primers DSV11, mreB, DSV13 and DSV8 were designed on the basis of previously reported sequences (Marsh & Taylor, 1999; Toma et al., 2001). Fragment-5, which contains the complete vcfQ gene, was amplified by using primers DSV13 and mshQ6 with KOD plus DNA polymerase (TOYOBO). Fragment-5 was cloned into pCR 2.1 vector to obtain plasmid pCTQ20 after addition of 3' A-overhangs which were necessary for TA cloning with Taq DNA polymerase (TOYOBO). All primers used in this study are presented in Table 2. Nucleotide sequencing was performed on both strands with Big Dye Terminator Cycle Sequencing FS Ready Reaction kits (Applied Biosystems) and analysed with an ABI PRISM 310 Genetic Analyzer (Perkin-Elmer). The program BLAST 1.4.9 was used to search for homologous sequences in the database.

View larger version (15K): [in this window] [in a new window]   Fig. 1. Sequencing strategy for vcfQ gene (black box). Arrows and open boxes below the ORF represent the primers and the fragments (F) amplified by PCR, respectively. Horizontally striped box shows the 3' portion of ORF5 (Toma et al., 2002) and the vertically striped box shows the 5' portion of mreB. The two probes used for colony blotting in this study are indicated.

Cloning of mshQ.

The same protocol used for cloning of vcfQ was used to clone the mshQ gene from V. cholerae O1 strain 86B3. The complete mshQ gene was amplified by using primers QCE1 and mshQ6. The PCR product was cloned into pCR 2.1 vector and the resulting plasmid (pQ28-4) was used to transform Escherichia coli XL-1 Blue. E. coli XL-1 Blue harbouring pQ28-4 was used for immunological detection of MshQ.

Cloning the C-terminal region of vcfQ into the expression vector.

The C-terminal region (VcfQT) of vcfQ was amplified by PCR using primers nagQ15 and nagQ18. The PCR product was digested with BamHI and SmaI and cloned into the corresponding sites of pQE30 vector. The recombinant plasmid (pTM2) was transformed into E. coli M15.

Expression and purification of recombinant His-tagged protein.

VcfQT was purified from E. coli M15 harbouring plasmid pTM2 as a (His)₆-tagged recombinant protein. Overnight culture of the recombinant organism was diluted 1 : 50 in LB medium (containing ampicillin and kanamycin) and induced with 1 mM IPTG. The harvested cells were dissolved in start buffer (8 M urea, 0.1 M NaH₂PO₄, 0.01 M Tris, 10 mM imidazole, pH 8.0). The soluble fraction was collected and purified using a His Trap nickel column (Amersham) using 500 mM imidazole phosphate buffer as the elution buffer.

Antiserum preparation.

A rabbit was immunized with 87.5 µg purified VcfQT protein in Freund's complete adjuvant. The animal was boosted three times with the same antigen in Freund's incomplete adjuvant at 2-week intervals.

SDS-PAGE and Western blotting analysis.

SDS-PAGE and Western blotting were carried out according to the methods of Laemmli (1970) and Towbin et al. (1979), respectively. Pre-stained molecular markers (New England Biolabs) were used as size standards.

Cellular fractionation.

Cellular fractionation was accomplished by a modification of the method of Ramer et al. (2002). V. cholerae 88UDT119 was cultured in 500 ml LB broth at 37 °C overnight with shaking. The harvested cells were suspended in 8 ml 1 M sucrose in 30 mM Tris/HCl (pH 8.0), to which 80 µl 0.5 M EDTA and 80 µl of a 20 mg ml^–1 lysozyme solution were added. The suspension was then incubated on ice for 40 min. MgCl₂ was added to a final concentration of 75 mM. The cells were centrifuged at 15 000 g for 30 min and the resulting supernatant was collected as the periplasmic fraction. The cell pellet was resuspended in 8 ml 1 M sucrose-30 mM Tris/HCl (pH 8.0)-75 mM MgCl₂. The resulting suspension was sonicated four times (1 min each) and then subjected to two rounds of freezing and thawing. The unbroken cells were removed by centrifugation at 5000 g for 10 min and supernatant was subjected to further centrifugation at 113 000 g for 1 h. The resulting supernatant was termed the cytoplasmic fraction. The pellet was resuspended in 1 ml 10 mM PBS, and then 100 µl 10 % Sarkosyl was added. The suspension was treated with a glass homogenizer, and after centrifugation at 113 000 g for 1 h, the Sarkosyl-soluble fraction was termed the inner-membrane (IM) fraction. The pellet resuspended in 500 µl PBS was termed the OM fraction.

Construction of vcfQ insertion mutant.

To confirm that VcfQ is required for type IV pilus biogenesis, the vcfQ gene was disrupted by insertion of a suicide plasmid. A 1.5 kb central fragment amplified by PCR with primer mshQ3 and mshQ4 (fragment-1, Fig. 1) was cloned in the suicide plasmid pKY719 (Whayeb et al., 1996) to obtain plasmid pQM14. Bacterial conjugation was accomplished by mating E. coli SM10pir bearing pQM14 as a donor and V. cholerae 88UDT119 as recipient. The vcfQ exconjugants were selected on TCBS agar containing 5 µg chloramphenicol ml^–1. The vcfQ plasmid insertion mutant, WM4, was identified by PCR using primer pUCF, which is specific for the multiple cloning site of pKY719, and primer nagQ7, which is specific for sequences upstream of primer mshQ3. To complement the vcfQ mutation in WM4, pCTQ20, containing vcfQ, was electroporated into WM4 to obtain strain 50-2 (Table 1).

Crude type IV pili preparation.

Crude NAGV14 pili were prepared as described previously (Nakasone & Iwanaga, 1990).

Colony blotting.

A single colony was picked and placed onto a nylon membrane Hybond-N⁺ (Amersham). Alkaline lysis of the cells was performed and then probed with peroxidase-labelled PCR-amplified DNA fragments (Fig. 1). Hybridization was performed overnight at 42 °C, and high-stringency washes were performed before detection with the ECL direct nucleic acid labelling and detection system (Amersham).

RESULTS AND DISCUSSION

Nucleotide sequence analysis

The nucleotide sequences of fragments-1, 2, 3 and 4 (Fig. 1) were determined. To confirm the sequence, fragment-5 (Fig. 1) was cloned and sequenced. The sequence of fragment-5 contained a complete ORF (vcfQ) of 4410 bp. The deduced amino acid sequence revealed that vcfQ encoded a 1470 aa residue with a calculated molecular mass of 157.9 kDa and an isoelectric point of 4.77. VcfQ is predicted by PSORT analysis (Nakai & Kanehisa, 1991) to be an OM protein with a potential signal sequence of 19 aa.

Homology searches comparing the nucleotide sequence of vcfQ with other sequences indicated that portions of vcfQ showed homology to those of mshQ; however, vcfQ was 651 bp larger than mshQ. The size difference between vcfQ and mshQ might be explained by the difference of G+C content of the extra 651 bp region. The G+C content of vcfQ (42.5 mol%) and mshQ (45 mol%) are similar; however, the G+C content was lower (35.6 mol%) in the extra region. This may indicate that the region was introduced into the NAGV14 strain from some other organism since different G+C content is a reliable indicator of horizontal transfer (Lawrence & Ochman, 1998). As in V. cholerae O1, a portion of the mreB gene (Marsh & Taylor, 1999) was also found downstream of vcfQ.

BLASTP analysis of the deduced amino acid sequence revealed that the first 270 residues of VcfQ had no similarity with any known protein. However, high homology was detected between the VcfQ (residues 271–1470) and MshQ (residues 43–1253) (61 % identity, 70 % similarity) (Fig. 2). This homology was supported by the immunological cross-reaction found in Western blotting using VcfQ specific antiserum (anti-VcfQ antiserum) and whole-cell lysates of E. coli harbouring pCTQ20 or pQ28-4 (see gel in the supplementary data system in JMM Online; http://jmm.sgmjournals.org). By determining the vcfQ gene nucleotide sequence, we completed the genetic sequence analysis of the NAGV14 pilus structural operon. The vcfQ gene was identified as the last gene of the pilus operon.

View larger version (7K): [in this window] [in a new window]   Fig. 2. Comparison of the deduced amino acid sequences of VcfQ and MshQ. Black boxes represent homologous regions of VcfQ and MshQ. White boxes represent variable regions. Numbers show amino acid position relative to the methionine of VcfQ and MshQ taken as 1. Asterisks show signal sequence. The location of recombinant protein VcfQT is indicated.

Localization of VcfQ

Many Gram-negative bacteria express type IV pili on their surface; however, the mechanism of type IV pili biogenesis is poorly understood. The genes required for type IV pili biogenesis encode major pilin, minor pilin(s) and IM or OM proteins (Marsh & Taylor, 1999). The predicted amino acid sequence of vcfQ showed that VcfQ might be an OM protein. To investigate this possibility, cell fractions prepared from V. cholerae 88UDT119 were analysed by Western blotting with anti-VcfQ antiserum. As shown in the gel in the supplementary data system in JMM Online (http://jmm.sgmjournals.org), anti-VcfQT antiserum recognized a 157 kDa VcfQ protein in OM fraction of 88UDT119. The result shows that this protein is OM-associated. However, a detectable level of VcfQ was also found in the IM. Therefore, VcfQ might coalesce into an assembly complex by anchoring to IM and thus may articulate between the proteins in the IM and OM components. The presence of VcfQ in the IM and OM could be explained by disruption of this complex during cell fractionation. In E. coli type IV pilus (bundle-forming pilus, BFP), the assembly complex has been reported to consist of IM, OM and periplasmic components. One of these components, BfpL, was also shown to be associated with the IM and OM fractions and essential for BFP biogenesis (Ramer et al., 2002).

Phenotype of vcfQ-deficient mutant

To assess the role of vcfQ in type IV pilus biogenesis, we constructed a V. cholerae 88UDT119 strain containing insertion mutation in the vcfQ gene (WM4). Western blotting analysis using whole-cell lysates of WM4 with an anti-VcfA antiserum (Toma et al., 2002) revealed that the WM4 did produce normal levels of vcfA (see gel in the supplementary data system in JMM Online; http://jmm.sgmjournals.org). Multiple bands that appeared at 20, 21 and 22 kDa were likely due to the post-translational modification of pili as previously described by Kuroki et al. (2001). The type IV pilus of Pseudomonas aeruginosa and Neisseria meningitidis have been reported to be post-translationally modified by glycans (Castric, 1995; Power et al., 2000). Despite the presence of pilin, no pilus was detected in the crude pili fraction of WM4. Therefore, WM4 failed to assemble VcfA into pili. The complemented strain, 50-2, produced normal levels of pilin and extracellular pili. These results indicate that VcfQ is essential for pilus assembly but not for expression of VcfA.

Colony-blotting analysis

To investigate the conservation of vcfQ and mshQ in various strains, 93 strains of V. cholerae (Table 3) were tested by colony blotting using probe 1 and probe 2 (Table 2, Fig. 1). The results presented in Table 3 show that only one of the 25 V. cholerae O1 strains tested and none of the 18 V. cholerae O139 strains tested gave a positive signal with probe 1, which anneals to the N-terminal region of vcfQ. For V. cholerae non-O1/non-O139, 20 % (10 of 50) of the strains tested gave positive signals with probe 1. In contrast, 92 % (23 of 25) of V. cholerae O1 strains, 100 % (18 of 18) of V. cholerae O139 strains and 98 % (49 of 50) of V. cholerae non-O1/non-O139 strains tested produced strong signals with probe 2, which anneals to the C-terminal region of vcfQ and mshQ. We previously demonstrated that the NAGV14 pilus and MSHA pilus gene clusters are different in the C-terminal region of the gene (vcfA) encoding the major pilin subunit (Kuroki et al., 2001). In this study we elucidated that the N-terminal region of the gene (vcfQ) encoding the OM-associated protein is also variable.

One possible way for V. cholerae to cause epidemic disease is adhesion to the zooplankton using type IV pili and survival in the environment (Chiavelli et al., 2001). V. cholerae has to assemble the pili under many environmental conditions. Thus, the molecular variation detected here in the gene encoding an OM protein included in the type IV pilus gene cluster might be required for type IV pilus biogenesis.

ACKNOWLEDGEMENTS

ACKNOWLEDGEMENTS

This work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology, Japan.

REFERENCES