Abstract
1 International Vaccine Institute, San 4-8 Bongcheon 7 dong, Kwanak gu, Seoul, 151-818, Republic of Korea
2 School of Biological Sciences, Seoul National University, Seoul, 151-742, Republic of Korea
3 Ministry of Health, Maputo, Mozambique
4 Centro De Higiene Ambiental E Exames Medicos, Rua Correia De Brito N° 1815 Sofala Beira, Mozambique
5 International Centre for Diarrhoeal Disease Research, Bangladesh, Dhaka, Bangladesh
Serotyping and biotype analysis. The V. cholerae isolates from Mozambique were confirmed serologically by the slide agglutination test using specific antiserum to V. cholerae polyvalent O1 and serotype-specific antisera to Inaba and Ogawa serotype strains. For biotype analysis, we used chicken erythrocyte agglutination, haemolysis of sheep erythrocytes, the VogesProskauer reaction, sensitivity to polymyxin B, and Mukerjee classical phage IV and Mukerjee El Tor phage 5 tests (Basu & Mukerjee, 1968).
MLST and sequence analysis. Bacterial genomic DNA was prepared from a single colony grown overnight on nutrient agar by using the PrepMan Ultrakit (Applied Biosystems). All oligonucleotides used for the MLST analysis, PCR amplification and sequencing were purchased from Genotech (Taejon, Korea). Nine loci (dnaE, lap, rstA, gmd, recA, pgm, gyrB, cat and chi) were used for MLST analysis. Each locus was amplified by using PCR and primers as described previously (Garg et al., 2003). In the report of Garg et al. (2003), rstA was denoted as rstR, despite the fact that the amplified fragment was rstA. Here, we wish to clarify that rstA was used for MLST, and rstR was used to differentiate between the classical and El Tor biotype prophages.
The primer sets were as follows, and their locations on the CTX prophage genome are shown in Fig. 3: primer set 1 for El Tor biotype-specific rstR-rstA fragment amplification (forward, 5' TGTCTTATCAGCATACTTTC 3'; reverse, 5' GAGTGAATCGTCGTG 3'), primer set 2 for rstB-rstC (or cep)-rstR-rstA fragment amplification (forward, 5' ACGTTTGACAATGAGCCAG 3'; reverse, 5' GAGTGAATCGTCGTG 3'), primer set 3 for ctxB-rstR-rstA-rstB-psh-cep fragment amplification (forward, 5' TATGCACATGGAACACCTCAAA 3'; reverse, 5' AACCCCGAGTGAAAGCGTG 3'), primer set 4 for classical-biotype-specific rstR-rstA fragment amplification (forward, 5' AGCCTCCATCAAAATGAATA 3'; reverse, 5' GAGTGAATCGTCGTG 3') and rstC primer set (forward, 5' GGATGTTTACGATAGCCTAGAA 3'; reverse, 5' TTGAGTTGCGGATTTAGGC 3').
|
The amplified products were confirmed on agarose gel and purified with the UltraClean PCR Clean up DNA purification kit (MO BIO Laboratories). The purified PCR products were sequenced in both directions by using a Big Dye cycle sequencing kit (ABI) according to the manufacturer's instructions. Sequencing was performed on an ABI 3770 automatic sequencer. For alignment and phylogenetic analyses, we used several software programs: MultAlin [available at (Corpet, 1988)], PHYDIT (available at ) and NTSYspc version 2.11h (Exeter software). Subtyping of Mozambique isolates
All 40 Mozambican isolates were identified as V. cholerae O1 Ogawa serotype, as described previously (Ansaruzzaman et al., 2004). All isolates agglutinated with chicken cell erythrocytes, were positive for El Tor-type haemolysin by the modified tube agglutination method, yielded a positive VogesProskauer reaction, and were resistant to polymyxin B. They were sensitive to Mukerjee El Tor phage 5 but resistant to the classical phage IV (Table 1). The isolates were therefore classified as El Tor biotype.
Table 1. Biotype characterization of V. cholerae O1 isolated from Beira, Mozambique R, Resistant; S, sensitive.
Sequence type (ST) of V. cholerae isolates from Mozambique
Though a number of MLST studies have been conducted on V. cholerae, most have focused on O139 strains, and only a few O1 strains have been analysed with a limited number of loci for MLST analysis (Garg et al., 2003; Kotetishvili et al., 2003). We applied the MLST analysis method of Garg et al. (2003) to O1 El Tor and classical biotypes. Among 40 Mozambique V. cholerae O1 isolates transported to IVI, 23 were fully sequenced at nine MLST loci, and all 23 had identical DNA sequences at the loci. The remaining 17 isolates were sequenced at six to eight MLST loci, and each sequenced locus was identical to that of the other isolates. We therefore concluded that the Mozambican isolates belonged to one ST. When the sequence of each allele in our study was identical to previously reported allele sequences, the allele type number was denoted with the same number. The allele types and ST results are summarized in Table 2. We found one new allele type in rstA, recA and chi loci in all five classical biotype strains, and one new rstA allele type in the O139 strains. We also compared the DNA sequence of nine loci of the Mozambique isolates with those of the El Tor N16961 strain, for which the whole genome sequence is available (Heidelberg et al., 2000). As shown in Table 2, the ST of the Mozambique isolates was identical to that of V. cholerae O1 El Tor N16961. The allele profile of O1 El Tor N16961 and Mozambique isolates is 1, 1, 1, 2, 1, 1, 1, 1, 1 (in the order dnaE, lap, rstA, gmd, recA, pgm, gyrB, cat and chi). This allele profile does not match any of the O139 strains previously reported (Table 2; Garg et al., 2003). As this MLST method can be applied to O1 El Tor and classical biotypes, we propose further MLST analyses of O1 strains.
Table 2. Allele profiles of V. cholerae O1 classical, O1 ElTor and O139 strains, and Mozambique isolates of this study Numbers are previously reported allele profile numbers (Gargetal., 2003). N, New allele type identified in this study.
Based on allele profiles of V. cholerae, allele profile similarities were calculated and divided into four groups. The highest similarity value was found between V. cholerae Mozambique and V. cholerae O1 El Tor. After determining the V. cholerae allele profiles, we constructed a dendrogram using unweighted pair grouping with mathematical averaging (UPGMA) (Fig. 1). The Mozambique isolates were closely related to V. cholerae O1 El Tor compared to strains O139 and O1 classical biotype. When all sequences of the nine loci were used to calculate sequence similarity, the V. cholerae isolates from Mozambique and V. cholerae O1 El Tor strains were the most similar. On the basis of the allele profile, the dendrogram and the sequence similarities, we concluded that V. cholerae isolates from Mozambique are closely related to V. cholerae O1 El Tor.
|
CTX prophage of Mozambique V. cholerae isolates
A 1447 nt DNA fragment, encompassing the first 288 (of 336) nt of classical rstR, ig-2, and the first 1032 nt of rstA, was amplified only from classical strains and Mozambique isolates using primer set 4. The DNA sequence of Mozambique isolates showed 100 % homology with that of the classical biotype rstR and ig-2. With primer set 1, a 1460 nt fragment was amplified from all five O1 El Tor strains and all five O139 strains; however, there was no amplification of this DNA fragment in classical strains or in the Mozambique isolates. From these results, we concluded that only the classical rstR was present in the CTX prophage of V. cholerae isolates from Mozambique.
The DNA sequence of the 618 nt rstA internal fragment for MLST analysis (from nt 415 to 1032, out of 1083 nt of the full-length rstA) of Mozambique isolates was identical to that of the most common allele type of O139 and O1 El Tor N16961 (allele type 1). However, the rest of the rstA of the Mozambique isolates was different from rstA of El Tor N16961. When we compared the full-length sequences of rstA of Mozambique isolates to those of other V. cholerae strains, we found an overall sequence similarity of over 99 %. We identified 10 polymorphic sites (Fig. 2). The first four variable sites (27, 162, 183 and 258) of Mozambique isolates were identical to the those of the classical biotype, and the last six variable sites that belong to the internal fragment used for the MLST analysis were the same as those of the El Tor biotype. The positions of all polymorphic-site nucleotides within their codons were third positions, and all differences were synonymous changes when translated. Although the rstA of the Mozambique isolates seems to be a hybrid of classical biotype and El Tor biotype, the generation mechanisms of this mosaic structure would be quite complicated. Therefore, we propose the presence of a classical CTX phage with the same variable sites in rstA as those of the Mozambique isolates.
|
Given that the ST of the Mozambique isolates is identical to that of El Tor strain N16961 and that the CTX prophage of the Mozambique isolates is the classical type, it is likely that the Mozambique strain originated from the El Tor biotype. One possible (and the simplest) mechanism for the generation of this strain is the infection of the El Tor biotype strain by the classical CTX phage that has the Mozambique rstA. This classical CTXΦ either integrated into an O1 El Tor biotype strain that had lost the existing El Tor prophage or replaced the existing El Tor prophage. Since the classical CTX phage particle, along with the classical biotype strains of V. cholerae, is believed to have been extinct from 1994 onwards (Nair et al., 2002; Safa et al., 2005), the acquisition of the classical CTX phage genome by the El Tor biotype remains an open question. Epidemic cholera in Mozambique, caused by V. cholerae O1, Ogawa, biotype El Tor strains, was first reported in 1997 (Folgosa et al., 2001). Those strains contained two copies of CTX prophage, and were resistant to sulfonamides and trimethoprim, characteristics similar to those of the Mozambique isolates. The relationship between the isolates from the 1997 and 2004 epidemics can be established with the methods presented here. Further studies will be needed to understand the origin of the Mozambique strains, since no classical biotype strains have reached the African continent during the past 30 years of the seventh cholera pandemic. The investigation of the pathogenicity of this hybrid strain and the intensity of symptoms in patients should also be investigated, because the O1 El Tor and classical biotype strains differ in these aspects.
CTX prophage array in the genome of Mozambique V. cholerae isolates
We examined the CTX prophage array of the Mozambique isolates with respect to two aspects: 1) the presence of the RS1 element or the truncated prophage, and 2) the presence of the tandem repeat of the CTX prophage. To see the presence of the RS1 element or 3' truncated CTX' on the 5' region of the prophage, we used primer set 2, designed to amplify the fragment between rstB and rstA, as shown in Fig. 3. The same primer set produced a different size DNA fragment from the classical biotype and El Tor biotype strains. A 1962 nt fragment was amplified from the classical biotype strains C-19385, F-2427, H-18, X-19850 and Y-8661, which implies that there are truncated CTX prophages in these strains, as shown in Fig. 3. From all five El Tor strains and from three O139 strains (AR-196157, AR-18096 and AR-9954), a 2543 nt fragment was amplified, showing the presence of the RS1-CTX prophage array. No DNA fragment was amplified from the Mozambique isolates with the same primer set, indicating that no RS1 element is present upstream of the CTX prophage, or the CTX' truncated prophage. We confirmed the absence of the RS1 element in the Mozambique isolates with an rstC primer set. No DNA amplification was obtained from Mozambique isolates, but a 173 nt fragment was amplified from all O1 El Tor and O139 strains (data not shown).
We used primer set 3 to evaluate the presence of the tandem repeat of the CTX prophage on the genome of the Mozambique isolates. The PCR product encompassing ctxB-ig1-rstR-ig2-rstB-psh-cep can be amplified only from a tandem repeat of the CTX prophage. As expected, there was no amplification from classical biotype strains (data not shown). This DNA fragment was PCR-amplified from only one O1 El Tor strain (2201969) and two O139 strains (AR196157 and 2206945). The DNA sequence of the fragment from the O139 isolates showed that these strains had a similar genetic structure to that of strain O139 AS207, which contains one El Tor prophage followed by a tandem repeat of a new-type prophage, CTX Calcutta, as shown in Fig. 3 (Davis et al., 1999). A DNA fragment of similar size was amplified from most Mozambique isolates, and we analysed the DNA sequence of this fragment (GenBank accession no. DQ012295). Mozambique isolates contain an ig-1 that is homologous to that of the CTX Calcutta prophage (18 nt different out of 730 nt), followed by a classical biotype rstR instead of the Calcutta type rstR. The potential genetic structure and the array of CTX prophages of the Mozambique isolate are shown in Fig. 3. Since the infectious CTX virion particles (CTXΦ) can be produced from the tandem repeat prophage array (Davis et al., 2000), we expect that this tandem-repeat classical prophage array can yield classical CTX phage particles.
The CTX prophage integration site on the large chromosome is well defined (Heidelberg et al., 2000), and we could detect the integration site on the large chromosome of O1 El Tor, classical and O139 strains by using PCR primer pairs that straddled both ends of the prophage. However, we could not define the integration site of the CTX prophage of the Mozambique strains with the same primers (data not shown). Identification of the CTX prophage location and full sequencing of the tandem repeat prophage are planned.
This work was supported by the Bill and Melinda Gates Foundation through the Diseases of the Most Impoverished Program coordinated by the IVI, Korea. The core donors to the International Center for Diarrhoeal Disease Research, Bangladesh (ICDDR,B), supported this work. Current donors providing unrestricted support include the aid agencies of the governments of Australia, Bangladesh, Belgium, Canada, Kingdom of Saudi Arabia, The Netherlands, Sweden, Sri Lanka, Switzerland and the United States of America.The Mozambique Cholera Vaccine Demonstration Project Coordination Group includes participants from the Ministério da Saude, Maputo, Mozambique (Avertino Barreto, Arminda Mcuamule Juvenaldo Amos and Raul Vaz); IVI (Jacqueline L. Deen, Xuan-Yi Wang, Mohammad Ali and Mahesh K. Puri); Médecins Sans Frontières, Geneva, Switzerland (Claude Mahoudeau, Bruno Lab, Gérard Bedock, Valerie Perroud and Margaret McChesney); Epicentre, Paris, France (Julia Ampuero, Philippe Cavailler, Philippe J. Guerin and Dominique Legros); World Health Organization, Geneva, Switzerland (Claire-Lise Chaignat, Marie-Paule Kieny and Duncan Steele); and World Health Organization, Maputo, Mozambique (Bocar Touré and Pierre Kahozi).
References
Basu, S. & Mukerjee, S. (1968). Bacteriophage typing of Vibrio eltor. Experientia 24, 299300.[CrossRef][Medline]
Basu, A., Mukhopadhyay, A. K., Garg, P., Chakraborty, S., Ramamurthy, T., Yamasaki, S., Takeda, Y. & Nair, G. B. (2000). Diversity in the arrangement of the CTX prophages in classical strains of Vibrio cholerae O1. FEMS Microbiol Lett 182, 3540.[CrossRef][Medline]
Corpet, F. (1988). Multiple sequence alignment with hierarchical clustering. Nucleic Acids Res 16, 1088110890.
Davis, B. M. & Waldor, M. K. (2000). CTXΦ contains a hybrid genome derived from tandemly integrated elements. Proc Natl Acad Sci U S A 97, 85728577.
Davis, B. M., Kimsey, H. H., Chang, W. & Waldor, M. K. (1999). The Vibrio cholerae O139 Calcutta bacteriophage CTXΦ is infectious and encodes a novel repressor. J Bacteriol 181, 67796787.
Davis, B. M., Moyer, K. E., Boyd, E. F. & Waldor, M. K. (2000). CTX prophages in classical biotype Vibrio cholerae: functional phage genes but dysfunctional phage genomes. J Bacteriol 182, 69926998.
Davis, B. M., Kimsey, H. H., Kane, A. V. & Waldor, M. K. (2002). A satellite phage-encoded antirepressor induces repressor aggregation and cholera toxin gene transfer. EMBO J 21, 42404249.[CrossRef][Medline]
Faruque, S. M., Albert, M. J. & Mekalanos, J. J. (1998). Epidemiology, genetics, and ecology of toxigenic Vibrio cholerae. Microbiol Mol Biol Rev 62, 13011314.
Folgosa, E., Mastrandrea, S., Cappuccinelli, P., Uzzau, S., Rappelli, P., Brian, M. J. & Colombo, M. M. (2001). Molecular identification of pathogenicity genes and ERIC types in Vibrio cholerae O1 epidemic strains from Mozambique. Epidemiol Infect 127, 1725.[Medline]
Garg, P., Aydanian, A., Smith, D. J., Glenn, M. J., Nair, G. B. & Stine, O. C. (2003). Molecular epidemiology of O139 Vibrio cholerae: mutation, lateral gene transfer, and founder flush. Emerg Infect Dis 9, 810814.[Medline]
Heidelberg, J. F., Eisen, J. A., Nelson, W. C. & 29 other authors (2000). DNA sequence of both chromosomes of the cholera pathogen Vibrio cholerae. Nature 406, 477483.[CrossRef][Medline]
Kaper, J. B., Morris, J. G., Jr & Levine, M. M. (1995). Cholera. Clin Microbiol Rev 8, 4886.[Abstract]
Kimsey, H. H. & Waldor, M. K. (1998). CTXΦ immunity: application in the development of cholera vaccines. Proc Natl Acad Sci U S A 95, 70357039.
Kotetishvili, M., Stine, O. C., Chen, Y., Kreger, A., Sulakvelidze, A., Sozhamannan, S. & Morris, J. G., Jr (2003). Multilocus sequence typing has better discriminatory ability for typing Vibrio cholerae than does pulsed-field gel electrophoresis and provides a measure of phylogenetic relatedness. J Clin Microbiol 41, 21912196.
Li, M., Shimada, T., Morris, J. G., Jr, Sulakvelidze, A. & Sozhamannan, S. (2002). Evidence for the emergence of non-O1 and non-O139 Vibrio cholerae strains with pathogenic potential by exchange of O-antigen biosynthesis regions. Infect Immun 70, 24412453.
Mekalanos, J. J., Rubin, E. J. & Waldor, M. K. (1997). Cholera: molecular basis for emergence and pathogenesis. FEMS Immunol Med Microbiol 18, 241248.[CrossRef][Medline]
Nair, G. B., Faruque, S. M., Bhuiyan, N. A., Kamruzzaman, M., Siddique, A. K. & Sack, D. A. (2002). New variants of Vibrio cholerae O1 biotype El Tor with attributes of the classical biotype from hospitalized patients with acute diarrhea in Bangladesh. J Clin Microbiol 40, 32963299.
Nusrin, S., Khan, G. Y., Bhuiyan, N. A. & 9 other authors (2004). Diverse CTX phages among toxigenic Vibrio cholerae O1 and O139 strains isolated between 1994 and 2002 in an area where cholera is endemic in Bangladesh. J Clin Microbiol 42, 58545856.
Safa, A., Bhuiyan, N. A., Alam, M., Sack, D. A. & Nair, G. B. (2005). Genomic relatedness of the new Matlab variants of Vibrio cholerae O1 to the classical and El Tor biotypes as determined by pulsed-field gel electrophoresis. J Clin Microbiol 43, 14011404.
Waldor, M. K. & Mekalanos, J. J. (1996). Lysogenic conversion by a filamentous phage encoding cholera toxin. Science 272, 19101914.[Abstract]