| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Enteric Diseases Reference Unit, National Institute for Communicable Diseases/University of the Witwatersrand, Johannesburg, South Africa
2 Department of Medical Microbiology, Nelson R Mandela Medical School, University of KwaZulu-Natal, South Africa
Correspondence
Karen H. Keddy
karenk{at}nicd.ac.za
Received 14 February 2007
Accepted 28 August 2007
 |
INTRODUCTION |
|---|
 TOP INTRODUCTION METHODSRESULTSDISCUSSIONREFERENCES |
|---|
A previous cholera epidemic in South Africa occurred in the 1980s, with over 20 000 culture-confirmed cases being documented. Again, KwaZulu-Natal was the worst-affected province, although cases were described in Limpopo and Mpumalanga. This epidemic was primarily due to V. cholerae O1 Inaba (Küstner & Du Plessis, 1991). This followed an outbreak of cholera in Maputo in Mozambique between 1980 and 1981 (Swerdlow & Isaäcson, 1994).
Two serogroups, O1 and O139, of V. cholerae have been associated with epidemic disease, although V. cholerae O139 has to date not been identified in epidemics in Africa. Serogroups are defined by the antigenicity of the O-antigen portion of the lipopolysaccharides (Manning et al., 1994). Serotype switching in V. cholerae O1 has been described previously. It is due to a change in the genetic make-up of the wbeT (previously known as rfbT) gene (Reeves et al., 1996), resulting in premature truncation of the gene product (Manning et al., 1994; Stroeher et al., 1992). This genetic change may be due to either a point mutation or a deletion, both resulting in the transcription and translation of a premature stop codon (Stroeher et al., 1992; Ito et al., 1993). The serotype of V. cholerae O1 depends on which of the genes encoding somatic antigens A, B and C are expressed. Serotype Ogawa carries antigens A and B; serotype Inaba carries antigens A and C. The third serotype, V. cholerae O1 Hikojima, expresses antigens A, B and C. Expression of antigens A and B in V. cholerae O1 Ogawa would thus change to expression of antigens A and C in V. cholerae O1 Inaba (Manning et al., 1994).
V. cholerae has been recognized to exist in the environment in a viable but non-culturable form from which it can re-emerge (Colwell & Spira, 1992). The purpose of this study was to identify whether South Africa was affected by a single clone of V. cholerae in the recent epidemic, which underwent a serotype switch from Ogawa to Inaba, or whether the epidemic was due to two different clones, resulting from the introduction of V. cholerae O1 Ogawa and the re-emergence of the V. cholerae O1 Inaba from the 1980s epidemic. To differentiate between the two possibilities, we performed PFGE as well as mutation analysis of the wbeT gene of isolates from the recent epidemic and of stored isolates from the previous epidemic.
|
METHODS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
).
|
Serotyping. Serotyping was performed in accordance with guidelines prescribed by the Clinical and Laboratory Standards Institute, USA, using commercially available antisera (Murex Biotech).
DNA extraction. Total genomic DNA was extracted from overnight growth of isolates on 10 % blood agar (Diagnostic Media Products). Suspensions containing 105–106 organisms ml–1 in 0.3 ml sterile water were boiled for 10 min at 99 °C. This was followed by centrifugation for 2 min at 12 000 r.p.m. and the DNA-containing supernatant was stored at 4 °C.
PFGE. PFGE was performed according to the protocol of the National Molecular Subtyping Network for Foodborne Disease (Pulse-Net) as set out by the Centers for Disease Control and Prevention, Atlanta, USA, on a CHEF-DR III System (Bio-Rad Laboratories) (Centers for Disease Control and Prevention; http://www.cdc.gov/pulsenet/protocols/vibrio_May2006.pdf). Total genomic DNA was restricted with NotI endonuclease (Roche Diagnostics). A strain of Salmonella enterica serotype Braenderup digested with XbaI was used as reference size marker. The digested DNA was separated on a 1 % agarose gel (Seakem Gold agarose; BioWhittaker Molecular Applications) at 14 °C in 0.5x Tris-borate-EDTA (TBE). The following electrophoresis conditions were used: block 1 , 6 V cm–2 for 12 h with 2 and 10 s (initial and final switch times) pulses; block 2 , 6 V cm–2 for 15 h with 10 and 15 s (initial and final switch times) pulses. The gel was stained with ethidium bromide (1 µg ml–1) (BDH Laboratory Supplies), destained in nuclease-free water and viewed by UV illumination. The image was photographed under UV light using a DS34 Polaroid Direct Screen Instant Camera.
Amplification of the ctxA and wbeT genes. To establish the toxicity of the isolates, PCR amplification of the cholera toxin gene ctxA was carried out using the CTX2/CTX3 primer pair, which produces an amplification product of 563 bp (Mekalanos et al., 1983). A second set of primers was designed to amplify the wbeT gene. This included primers SN3 (5'-AGGTCCTCAAACCTGCATCTG-3') and SN2 (5'-CCCATTAATATCCTGCGTTGC-3') targeting positions bp 17 113–17 133 and bp 18 003–18 023 of the gene, respectively. These primers amplify a region of 910 bp within the wbeT gene. This primer set was constructed from the nucleotide sequence of a strain of V. cholerae serotype Ogawa, accession number X59554, made available by the Entrez Nucleotide Sequence database (NCBI, http://www.ncbi.nlm.nih.gov/entrez/viewer.fcgi?db=nucleotide&val=48381; accessed 15 July 2003). They were produced by Inqaba Biotechnical Industries, Pretoria, South Africa.
For each of the reactions, the 50 µl mix consisted of 2 µl DNA template, 1.5 mM MgCl2 (Promega), 1.25 µM deoxynucleotide triphosphates (Promega), 50 mM KCl, 10 mM Tris/HCl (pH 9.0), 1 % Triton X-100, 1.0 µM of each primer (Inqaba) and 1 U Taq DNA polymerase (Promega). The PCR assay included initial incubation at 95 °C for 2 min, followed by 25 cycles of 95 °C for 1 min, 60 °C for 1 min and 72 °C for 1 min. A 2 min extension at 72 °C was included at the end of the final cycle.
All PCR amplifications were carried out in an iCycler thermal cycler (Bio-Rad) and DNA amplicons were visualized by ethidium bromide (BDH) staining and UV illumination.
Sequencing and protein translation. The wbeT gene amplicons were sequenced using an automated AB310 sequencer (Applied Biosystems). PCR products obtained with the SN3/SN2 primer set were excised from the agarose gel and purified using the Gene Clean protocol (Bio 101). Cycle sequencing was performed on the purified DNA fragments using the ABI PRISM BigDye Terminator v3.0 Cycle Sequencing Ready Reaction kit with AmpliTaq DNA Polymerase, FS (Applied Biosystems) using both reverse and forward primers. Sequences generated by SN2 were reversed in the direction of reading frame and the complementary sequence was then created using the tools made available at http://www.cbio.psu.edu/sms/rev_comp.html (accessed 11 November 2003–10 February 2004; website no longer available). Nucleotide sequences were translated to proteins using the tools made available at http://au.expasy.org/tools/dna.html (accessed 11 November 2003–10 February 2004) and compared with one another by manual alignments. The region of the nucleotide sequence of V. cholerae serotype Ogawa, accession number X59554, encoding the wbeT gene was selected for the sequence alignment and comparison.
|
RESULTS |
|---|
|
TOPINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
.
PFGE
Irrespective of the geographical origin, the PFGE patterns obtained from serotype Inaba isolates from the 2001/2002 epidemic differed from those isolated in 1980 (Fig. 2). Of the 112 isolates from the 2001/2002 epidemic that were available for genotyping, 90 (80.4 %) were indistinguishable and formed the major clone. An additional 11 (9.8 %) showed a pattern closely associated with that of the major clone. Of these, 5 showed an additional band and 6 had a band deletion. This family of strains comprised isolates from KwaZulu-Natal as well as from other regions of RSA. Restriction patterns completely different from the most common patterns observed were exhibited by 11 isolates, all of which were Inaba isolates from the 1980 outbreak. The PFGE patterns obtained were analysed using Gel Compare II (www.applied-maths.com) (results not shown). Thirty-one representative strains from each PFGE cluster, based on origin, 100 % identity within each clone and year of isolation, were selected for sequencing analysis by primer set SN3 and SN2.
|
).
|
Comparison of nucleotide sequences of wbeT genes
The wbeT gene was detected in all 31 isolates tested. The nucleotide sequences of the amplification product of isolates belonging to the Inaba and the Ogawa serotypes were determined and compared with one another. Two isolates demonstrated identical sequences of the amplicons using either the forward or the reverse primer, which indicated the presence of tandemly repeated positions of primer homology. The remaining 29 isolates could be sequenced by the forward and reverse primers yielding separate products.
Analysis of the peptide sequences was done by comparison with the peptide sequence translated from the nucleotide sequence of the selected reference strain V. cholerae serotype Ogawa, accession number X59554 (Fig. 4). Relative to the amino acid arrangement of the wbeT gene in the reference strain, a distinct difference at position 46 was noted in each of the 31 RSA isolates analysed. The result of this change is the replacement of the leucine indicated in the reference strain by tryptophan in the RSA isolates. Additional amino acid changes were identified in four of the Inaba isolates, of which three showed the same pattern. All 12 Inaba isolates (two from 1980 and 10 from 2001/2002) exhibited the presence of a stop codon sequence at the same position (bp 17 618) that was not present in the sequence of the selected reference strain of V. cholerae O1 serotype Ogawa (Stroeher et al., 1992) or in any of the other RSA Ogawa isolates. This termination sequence was generated in the Inaba strain by a one-base difference at nucleotide 17 610, namely the deletion of a single adenine residue, which resulted in premature termination of translation (IQT – stop instead of NTDI). No other mutations observed resulted in premature stop codons.
|
|
DISCUSSION |
|---|
|
TOPINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
While the V. cholerae O1 isolates from the epidemic in the 1980s belonged to serotype Inaba, the recent epidemic started with serotype Ogawa. However, fairly early in the epidemic, serotype Inaba emerged. Serotype switches do occur in response to selective pressure due to the development of population or individual immunity (Sheehy et al., 1966; Sack & Miller, 1969), and are therefore expected later in an epidemic. We therefore considered that the V. cholerae O1 Inaba could be the cause of a concomitant epidemic due to the re-emergence of the strain from the 1980s. Such a re-emergence by a strain dormant in the environment has been reported before (Colwell & Spira, 1992). In order to differentiate between the strains from the current and previous epidemics, we performed PFGE to establish the clonality of the different serotypes and to identify whether the 2001/2002 V. cholerae O1 Inaba was related to the strain from the previous epidemic in the 1980s or to the V. cholerae O1 Ogawa from the recent epidemic.
PFGE clearly showed a distinct clustering of V. cholerae O1 Inaba isolates from 1980 with a pattern that differed significantly from that of the 2001/2002 isolates (Fig. 2, Fig. 3, full data not shown). Although two serotype Ogawa strains from 2000/2001 (U15104 and U15823) appeared to have some similarity with serotype Inaba strains from 1980, they were still more closely related to the current 2000/2001 strains, approaching 80 % similarity on analysis of banding patterns. Sequencing results of the wbeT genes for these two strains did not show any significant differences from the other strains analysed in that portion of the genome and the mutations that resulted in the PFGE differences must lie elsewhere in the genome. The remaining strains from the recent epidemic were highly clonal, irrespective of serotype, with only minor banding differences on PFGE in some of the strains (Fig. 3). This suggests the occurrence of a serotype switch in the strains from 2001/2002 from serotype Ogawa to serotype Inaba. This was confirmed by sequence analysis of the wbeT gene and was due to a deletional mutation at nucleotide 17 610, resulting in a premature stop codon at position 17 618. Other investigators have also observed that serotype switching may be the result of either deletional mutations or point mutations resulting in the truncation of the outer-membrane protein during translation (Stroeher et al., 1992; Ito et al., 1993; Manning et al., 1994). The reversal of this switch, Inaba to Ogawa, as occurred in the cholera epidemic in South America in the 1980s and India in the 1990s (Wachsmuth et al., 1993; Garg et al., 2000) would then require a second mutation, removing the premature stop codon from the gene. This has been calculated to occur at a lower frequency of 104 compared with an Ogawa to Inaba switch (Manning et al., 1994).
It is interesting to note that certain amino acids in the strains of V. cholerae O1 from both the 1980s as well as 2001 were consistent, but differed from those of the published reference strain, irrespective of serotype (Fig. 4) (Stroeher et al., 1992). Also, despite the temporal difference in the years of isolation, the specific mutation resulting in the premature stop codon in the Inaba strains was conserved (Fig. 4). This may reflect that these strains of V. cholerae O1 are prone to mutation at a specific site on the gene, as the PFGE patterns do not support a relationship between the 1980s Inaba and 2001 Inaba isolates.
Significant amino acid changes occurred in two isolates but this did not affect their serotyping. As the remaining sequences of the wbeT genes in these two strains were comparable with the other wbeT gene sequences, and both of these isolates had restriction patterns similar to the other strains from the current epidemic, these mutations were thus unlikely to reflect an alternative origin of the two isolates and suggested that they were part of the current epidemic. One of these isolates, strain U15098/02 from the Eastern Cape (sequence 29), which was serotype Ogawa appeared to have certain nucleotide sequences more in common with the Inaba strains than with the Ogawa strains from the current epidemic (full data not shown). As the first cases of cholera in that area were found later in the epidemic, in 2002, this may reflect a rare reverse mutation from serotype Inaba to Ogawa (Manning et al., 1994). The second strain, K15854/02, exhibited differences between amino acids 128 and 135, i.e. HSLSSS instead of PLVEFHH.
Other strains also showed point mutations within the wbeT gene that did not affect serotyping (Ito et al., 1993). Three further strains, two isolated from Mpumalanga (S69145/80 and S69175/80) and one from KwaZulu-Natal (K1290/01) (neighbouring provinces in South Africa), showed the same amino acid changes at position 157–160 (Fig. 4). Review of the PFGE gels and dendrogram (Fig. 3) confirmed that these isolates were less closely related and that this mutation does not reflect an epidemiological linkage.
Earlier work on the isolates from the most recent epidemic as well as antimicrobial susceptibility testing on the isolates from 2001 to 2002 has shown them to be multidrug-resistant and to carry a transmissible SXT element, an aadA2 gene cassette and the tetA gene (Dalsgaard et al., 2001). Little is known about the epidemiology of the origin of the 2001/2002 epidemic. Some isolates were from cholera cases imported from Mozambique, suggesting that, as in the 1980s, migration from that country could have happened. Ribotyping of South African isolates from the recent epidemic showed these to be related to the Mozambican isolates identified collected in 1997 (Dalsgaard et al., 2001).
This work provides evidence that the emergence of V. cholerae O1 Inaba during the 2001/2002 epidemic was due to a serotype switch, rather than to the re-emergence of a previously dormant strain. It has been suggested that serotype switches are the result of selective pressure on the organism due to the acquisition of immunity. This was shown in previous work by Stroeher et al. (1992), and has been borne out by others (Ito et al., 1993; Wachsmuth et al., 1993; Garg et al., 2000).
The early appearance of this switch may have resulted from the strain circulating in the population for a considerable period of time without being recognized, as suggested by Dalsgaard et al. (2001). This could have resulted in a build-up of sufficient population immunity to allow for preferential transmission of the Inaba mutant. Little is known regarding the epidemiology of cholera in a population with a high HIV prevalence, and this was greater than 30 % in the antenatal seroprevalence survey during the outbreak period (Department of Health; http://www.doh.gov.za/docs/reports/2000/hivreport.html) and may have impacted on the serotype switch. No studies were done to show that the Inaba strain may have acquired increased colonization capacity, resulting in its early appearance.
Cholera in Mozambique is considered to be an endemic disease, with seasonal variation (Lucas et al., 2005), but in South Africa, this does not appear to be true. The 2001/2002 cholera epidemic was due to the introduction of a new strain of V. cholerae O1 that appeared in South Africa in the late 1990s. The serotype switch from Ogawa to Inaba arose de novo in KwaZulu-Natal in 2001 and the isolates are not clonally related to the V. cholerae O1 serotype Inaba isolates from the 1980s.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. M. Smith, A. Sooka, H. Ismail, S. Nadan, N. Crisp, E. Weenink, K. H. Keddy, and for the Group for Enteric, Respiratory and Meninge Analysis of Vibrio cholerae isolates from the Northern Cape province of South Africa J. Med. Microbiol., January 1, 2009; 58(1): 151 - 154. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | J MED MICROBIOL | MICROBIOLOGY | J GEN VIROL | ALL SGM JOURNALS |
