Toxocara canis of canids is a parasitic nematode (ascaridoid) that infects humans and other hosts, causing different forms of toxocariasis. This species of Toxocara appears to be the most important cause of human disease, likely followed by Toxocara cati from felids. Although some studies from Malaysia and China have shown that cats can harbor another congener, T. malaysiensis, no information is available about this parasite for other countries. Moreover, the zoonotic potential of this parasite is unknown at this point. In the present study, we conducted the first investigation of domestic dogs and cats for Toxocara in Vietnam using molecular tools. Toxocara malaysiensis was identified as a common ascaridoid of domestic cats (in the absence of T. cati), and T. canis was commonly found in dogs. Together with findings from previous studies, the present results emphasize the need to explore the significance and zoonotic potential of T. malaysiensis in Vietnam and other countries where this parasite is endemic and prevalent in cats.
Babesia gibsoni is a tick-borne hemoprotozoan parasite of dogs that often causes fever and hemolytic illness. Detection of B. gibsoni has been predominantly reported in Asian countries, including Japan, Korea, Taiwan, Malaysia, Bangladesh and India. The present study shows the first molecular characterization of B. gibsoni detected from dogs in Bangladesh. Blood samples were collected on FTA® Elute cards from 50 stray dogs in Mymensingh District in Bangladesh. DNA eluted from the cards was subjected to nested PCR for the 18S rRNA gene of Babesia species. Approximately 800bp PCR products were detected in 15 of 50 dogs (30%). Based on restriction fragment length polymorphism (RFLP) and direct sequencing of the PCR products, all parasite isolates were identified as B. gibsoni. Furthermore, the BgTRAP (B. gibsoni thrombospondin-related adhesive protein) gene fragments were detected in 13 of 15 18S rRNA gene PCR positive blood samples. Phylogenetic analysis of the BgTRAP gene revealed that B. gibsoni parasites in Bangladesh formed a cluster, which was genetically different from other Asian B. gibsoni isolates. In addition, tandem repeat analysis of the BgTRAP gene clearly showed considerable genetic variation among Bangladeshi isolates. These results suggested that B. gibsoni parasites in a different genetic clade are endemic in dogs in Bangladesh. Further studies are required to elucidate the origin, distribution, vector and pathogenesis of B. gibsoni parasites circulating in dogs in Bangladesh.
A novel member of the Pteropine Orthoreovirus species has been isolated and sequenced for the whole genome from flying foxes (Pteropus vampyrus) imported to Italy from Indonesia. The new isolate named Indonesia/2010 is genetically similar to Melaka virus which has been the first virus of this species to be shown to be responsible for human respiratory disease. Our findings highlight the importance of flying foxes as vectors of potentially zoonotic viruses and the biological hazard that lies in the import of animals from geographical areas that are ecologically diverse from Europe.
The knowledge of the diversity of haemosporidian parasites is of primary importance as their representatives include agents of bird malaria. We investigated the occurrence of Haemoproteus spp. and Plasmodium spp. in bird populations from a single locality in the State of Selangor, Peninsular Malaysia, and report on the parasite prevalence of the two genera. A combination of methods (molecular and morphological) was used for detecting these parasites. Seventy-nine bird individuals were caught using mist-nets in July and August 2010 at Gombak Field Station of the University of Malaya, Kuala Lumpur. In total, 23 birds were identified as positive for Haemoproteus or Plasmodium infection and one individual was recognized as carrying mixed infection. The total prevalence of haemosporidians in the collected samples was 30.3%. Infections with parasites of the genus Haemoproteus were predominant compared to those of the genus Plasmodium. In total, 10 new cyt b lineages of Haemoproteus spp. and 3 new cyt b lineages of Plasmodium spp. were recorded in this study. From all recorded haemosporidian lineages (16 in total), 3 were known from previous studies - hCOLL2, hYWT2 and pNILSUN1. Two of them are linked with their corresponding morphospecies - Haemoproteus pallidus (COLL2) and Haemoproteus motacillae (YWT2). The morphological analysis in the present study confirmed the results obtained by the PCR method relative to prevalence, with 25.3% total prevalence of Haemoproteus and Plasmodium parasites. The intensities of infection varied between 0.01% and 19%. Most infections were light, with intensities below 0.1%. The present study is the first molecular survey of the protozoan blood parasites of the order Haemosporida recorded in Malaysia.
Virulent Newcastle disease virus (NDV) isolates from new sub-genotypes within genotype VII are rapidly spreading through Asia and the Middle East causing outbreaks of Newcastle disease (ND) characterized by significant illness and mortality in poultry, suggesting the existence of a fifth panzootic. These viruses, which belong to the new sub-genotypes VIIh and VIIi, have epizootic characteristics and do not appear to have originated directly from other genotype VII NDV isolates that are currently circulating elsewhere, but are related to the present and past Indonesian NDV viruses isolated from wild birds since the 80s. Viruses from sub-genotype VIIh were isolated in Indonesia (2009-2010), Malaysia (2011), China (2011), and Cambodia (2011-2012) and are closely related to the Indonesian NDV isolated in 2007, APMV1/Chicken/Karangasem, Indonesia (Bali-01)/2007. Since 2011 and during 2012 highly related NDV isolates from sub-genotype VIIi have been isolated from poultry production facilities and occasionally from pet birds, throughout Indonesia, Pakistan and Israel. In Pakistan, the viruses of sub-genotype VIIi have replaced NDV isolates of genotype XIII, which were commonly isolated in 2009-2011, and they have become the predominant sub-genotype causing ND outbreaks since 2012. In a similar fashion, the numbers of viruses of sub-genotype VIIi isolated in Israel increased in 2012, and isolates from this sub-genotype are now found more frequently than viruses from the previously predominant sub-genotypes VIId and VIIb, from 2009 to 2012. All NDV isolates of sub-genotype VIIi are approximately 99% identical to each other and are more closely related to Indonesian viruses isolated from 1983 through 1990 than to those of genotype VII, still circulating in the region. Similarly, in addition to the Pakistani NDV isolates of the original genotype XIII (now called sub-genotype XIIIa), there is an additional sub-genotype (XIIIb) that was initially detected in India and Iran. This sub-genotype also appears to have as an ancestor a NDV strain from an Indian cockatoo isolated in 1982. These data suggest the existence of a new panzootic composed of viruses of subgenotype VIIi and support our previous findings of co-evolution of multiple virulent NDV genotypes in unknown reservoirs, e.g. as recorded with the virulent NDV identified in Dominican Republic in 2008. The co-evolution of at least three different sub-genotypes reported here and the apparent close relationship of some of those genotypes from ND viruses isolated from wild birds, suggests that identifying wild life reservoirs may help predict new panzootics.
In the Arabian Peninsula malaria control is progressing steadily, backed by adequate logistic and political support. As a result, transmission has been interrupted throughout the region, with exception of limited sites in Yemen and Saudi Arabia. Here we examined Plasmodium falciparum parasites in these sites to assess if the above success has limited diversity and gene flow.
Streptococcus pneumoniae is an epidemiologically important bacterial pathogen. Recently, we reported the antibiotic susceptibility patterns of a limited collection of pneumococcal isolates in Malaysia with a high prevalence of erythromycin resistant strains. In the present study, 55 of the pneumococcal isolates of serotype 19F were further analysed by pulsed field gel electrophoresis (PFGE) and multilocus sequence typing (MLST). The generated genotypic patterns were then correlated with the antibiograms previously reported. Forty-seven different PFGE profiles (PTs) were obtained, showing that the isolates were genetically diverse. MLST identified 16 sequence types (STs) with ST-236 being predominant (58.2%), followed by ST-81 (10.3%). Among the ST-236 isolates, 22 were erythromycin resistant S. pneumoniae (ERSP) and 15 were trimethoprim/sulfamethoxazole (TMP/SMX) resistant, while among ST-81, four isolates were ERSP and two were TMP/SMX resistant. The high prevalence of erythromycin resistant serotype 19F isolates of ST-236 in this study has also been reported in other North and South East Asian countries.
Rapidly growing, non-tuberculous mycobacteria (NTM) in the Mycobacterium abscessus (MAB) species are emerging pathogens that cause various diseases including skin and respiratory infections. The species has undergone recent taxonomic nomenclature refinement, and is currently recognized as two subspecies, M. abscessus subsp. abscessus (MAB-A) and M. abscessus subsp. bolletii (MAB-B). The recently reported outbreaks of MAB-B in surgical patients in Brazil from 2004 to 2009 and in cystic fibrosis patients in the United Kingdom (UK) in 2006 to 2012 underscore the need to investigate the genetic diversity of clinical MAB strains. To this end, we sequenced the genomes of two Brazilian MAB-B epidemic isolates (CRM-0019 and CRM-0020) derived from an outbreak of skin infections in Rio de Janeiro, two unrelated MAB strains from patients with pulmonary infections in the United States (US) (NJH8 and NJH11) and one type MAB-B strain (CCUG 48898) and compared them to 25 publically available genomes of globally diverse MAB strains. Genome-wide analyses of 27,598 core genome single nucleotide polymorphisms (SNPs) revealed that the two Brazilian derived CRM strains are nearly indistinguishable from one another and are more closely related to UK outbreak isolates infecting CF patients than to strains from the US, Malaysia or France. Comparative genomic analyses of six closely related outbreak strains revealed geographic-specific large-scale insertion/deletion variation that corresponds to bacteriophage insertions and recombination hotspots. Our study integrates new genome sequence data with existing genomic information to explore the global diversity of infectious M. abscessus isolates and to compare clinically relevant outbreak strains from different continents.
Staphylococcus aureus biofilm associated infections remains a major clinical concern in patients with indwelling devices. Quantitative real-time PCR (qPCR) can be used to investigate the pathogenic role of such biofilms. We describe qPCRs for 12 adhesion and biofilm-related genes of four S. aureus isolates which were applied during in vitro biofilm development. An endogenous control (16S rRNA) was used for signal normalization. We compared the qPCR results with structural analysis using scanning electron microscopy (SEM). The SEM studies showed different cellular products surrounding the aggregated cells at different times of biofilm formation. Using qPCR, we found that expression levels of the gene encoding fibronectin binding protein A and B and clumping factor B (fnbA/B and clfB), which involves in primary adherence of S. aureus, were significantly increased at 24h and decreased slightly and variably at 48 h when all 4 isolates were considered. The elastin binding protein (ebps) RNA expression level was significantly enhanced more than 6-fold at 24 and 48 h compared to 12h. Similar results were obtained for the intercellular adhesion biofilm required genes type C (icaC). In addition, qPCR revealed a fluctuation in expression levels at different time points of biofilm growth of other genes, indicating that different parameter modes of growth processes are operating at different times.
Giardia duodenalis is considered the most common intestinal parasite in humans worldwide. In Malaysia, many studies have been conducted on the epidemiology of giardiasis. However, there is a scarcity of information on the genetic diversity and the dynamics of transmission of G. duodenalis. The present study was conducted to identify G. duodenalis assemblages and sub-assemblages based on multilocus analysis of the glutamate dehydrogenase (gdh), beta-giardin (bg) and triose phosphate isomerase (tpi) genes. Faecal specimens were collected from 484 Orang Asli children with a mean age of 7 years and examined using light microscopy. Specimens positive for Giardia were subjected to PCR analysis of the three genes and subsequent sequencing in both directions. Sequences were edited and analysed by phylogenetic analysis. G. duodenalis was detected in 17% (84 of 484) of the examined specimens. Among them, 71 were successfully sequenced using at least one locus. Genotyping results showed that 30 (42%) of the isolates belonged to assemblage A, 32 (45%) belonged to assemblage B, while discordant genotype results were observed in 9 specimens. Mixed infections were detected in 43 specimens using a tpi-based assemblage specific protocol. At the sub-assemblages level, isolates belonged to assemblage A were AII. High nucleotide variation found in isolates of assemblage B made subtyping difficult to achieve. The finding of assemblage B and the anthroponotic genotype AII implicates human-to-human transmission as the most possible mode of transmission among Malaysian aborigines. The high polymorphism found in isolates of assemblage B warrants a more defining tool to discriminate assemblage B at the sub-assemblage level.
Dengue viruses are mosquito-borne viruses that cause dengue fever and dengue hemorrhagic fever, both of which are globally important diseases. These viruses have evolved in a transmission cycle between human hosts and mosquito vectors in various tropical and subtropical environments. We previously isolated three strains of dengue type 1 virus (DENV1) and 14 strains of dengue type 3 virus (DENV3) during an outbreak of dengue fever and dengue hemorrhagic fever in Jakarta, Indonesia in 1988. Here, we compared the nucleotide sequences of the entire envelope protein-coding region among these strains. The isolates were 97.6-100% identical for DENV1 and 98.8-100% identical for DENV3. All DENV1 isolates were included in two different clades of genotype IV and all DENV3 isolates were included in a single clade of genotype I. For DENV1, three Yap Island strains isolated in 2004 were the only strains closely related to the present isolates; the recently circulated Indonesian strains were in different clades. Molecular clock analyses estimated that ancestors of the genotype IV strains of DENV1 have been indigenous in Indonesia since 1948. We predict that they diverged frequently around 1967 and that their offspring distributed to Southeast Asia, the Western Pacific, and Africa. For DENV3, the clade containing all the present isolates also contained strains isolated from other Indonesian regions and other countries including Malaysia, Singapore, China, and East Timor from 1985-2010. Molecular clock analyses estimated that the common ancestor of the genotype I strains of DENV3 emerged in Indonesia around 1967 and diverged frequently until 1980, and that their offspring distributed mainly in Southeast Asia. The first dengue outbreak in 1968 and subsequent outbreaks in Indonesia might have influenced the divergence and distribution of the DENV1 genotype IV strains and the DENV3 genotype I strains in many countries.
This paper describes the development and application of multilocus sequencing typing (MLST) and multi-virulence locus sequencing typing (MVLST) methods in determining the genetic variation and relatedness of 43 Vibrio cholerae strains of different serogroups isolated from various sources in Malaysia. The MLST assay used six housekeeping genes (dnaE, lap, recA, gyrB, cat and gmd), while the MVLST assay incorporated three virulence genes (ctxAB, tcpA and tcpI) and three virulence-associated genes (hlyA, toxR and rtxA). Our data showed that the dnaE and rtxA genes were the most conserved genes in V. cholerae O1 strains. Among the 12 studied genes, transitional substitutions that led to silent mutations were observed in all, except for gmd and hlyA, while non-synonymous substitutions occurred more frequently in virulence and virulence-associated genes. Five V. cholerae O1 strains were found to be the El Tor variant O1 strains because they harboured the classical ctxB gene. In addition, the classical ctxB gene was also observed in O139 V. cholerae. A total of 29 MLST types were observed, and this assay could differentiate V. cholerae within the non-O1/non-O139 serogroups. A total of 27 MVLST types were obtained. MVLST appeared to be more discriminatory than MLST because it could differentiate V. cholerae strains from two different outbreaks and could separate the toxigenic from the non-toxigenic subtypes. Although the O1 V. cholerae strains were closely related, the combined MLST and MVLST analyses differentiated the strains isolated from different localities. In conclusion, sequence-based analysis in this study provided a better understanding of mutation points and the type of mutations in V. cholerae. The MVLST assay is useful to characterise O1 V. cholerae strains, while combined analysis may improve the discriminatory power and is suitable for the local epidemiological study of V. cholerae.
Given the HIV epidemic in Malaysia, genetic information on opportunistic pathogens, such as Cryptosporidium and Giardia, in HIV/AIDS patients is pivotal to enhance our understanding of epidemiology, patient care, management and disease surveillance. In the present study, 122 faecal samples from HIV/AIDS patients were examined for the presence of Cryptosporidium oocysts and Giardia cysts using a conventional coproscopic approach. Such oocysts and cysts were detected in 22.1% and 5.7% of the 122 faecal samples, respectively. Genomic DNAs from selected samples were tested in a nested-PCR, targeting regions of the small subunit (SSU) of nuclear ribosomal RNA and the 60kDa glycoprotein (gp60) genes (for Cryptosporidium), and the triose-phosphate isomerase (tpi) gene (for Giardia), followed by direct sequencing. The sequencing of amplicons derived from SSU revealed that Cryptosporidium parvum was the most frequently detected species (64% of 25 samples tested), followed by C. hominis (24%), C. meleagridis (8%) and C. felis (4%). Sequencing of a region of gp60 identified C. parvum subgenotype IIdA15G2R1 and C. hominis subgenotypes IaA14R1, IbA10G2R2, IdA15R2, IeA11G2T3R1 and IfA11G1R2. Sequencing of amplicons derived from tpi revealed G. duodenalis assemblage A, which is of zoonotic importance. This is the first report of C. hominis, C. meleagridis and C. felis from Malaysian HIV/AIDS patients. Future work should focus on an extensive analysis of Cryptosporidium and Giardia in such patients as well as in domestic and wild animals, in order to improve the understanding of transmission patterns and dynamics in Malaysia. It would also be particularly interesting to establish the relationship among clinical manifestation, CD4 cell counts and genotypes/subgenotypes of Cryptosporidium and Giardia in HIV/AIDS patients. Such insights would assist in a better management of clinical disease in immuno-deficient patients as well as improved preventive and control strategies.
Dengue virus type 4 (DENV-4) circulates in tropical and subtropical countries from Asia and the Americas. Despite the importance of dengue virus distribution, little is known about the worldwide viral spread. Following a Bayesian phylogenetic approach we inferred the evolutionary history of 310 isolates sampled from 37 countries during the time period 1956-2008 and the spreading dynamics for genotypes I and II. The region (tropical rainforest biome) comprised by Malaysia-Thailand was the most likely ancestral area from which the serotype has originated and spread. Interestingly, cross-correlation analysis on demographic time series with the Asian sequences showed a statistically significant negative correlation that could be suggestive of competition among genotypes within the same serotype.
Foot-and-mouth disease (FMD) is endemic in the countries of mainland Southeast Asia where it represents a major obstacle to the development of productive animal industries. The aim of this study was to use genetic data to determine the distribution of FMD virus (FMDV) lineages in the Southeast Asia region, and in particular identify possible sources of FMDV causing outbreaks in Malaysia. Complete VP1 sequences, obtained from 214 samples collected between 2000 and 2009, from FMD outbreaks in six Southeast Asian countries, were compared with sequences previously reported. Phylogenetic analysis of these sequences showed that there were two patterns of FMDV distribution in Malaysia. Firstly, for some lineages (O/SEA/Mya98 and serotype A), outbreaks occurred every year in the country and did not appear to persist, suggesting that these incursions were quickly eradicated. Furthermore, for these lineages FMD viruses in Malaysia were closely related to those from neighbouring countries, demonstrating the close epidemiological links between countries in the region. In contrast, for O/ME-SA/PanAsia lineage, viruses were introduced and remained to cause outbreaks in subsequent years. In particular, the recent incursion and maintenance of the PanAsia-2 sublineage into Malaysia appears to be unique and independent from other outbreaks in the region. This study is the first characterisation of FMDV in Malaysia and provides evidence for different epidemiological sources of virus introduction into the country.
Japanese encephalitis virus (JEV) is the most important cause of epidemic encephalitis worldwide but its origin is unknown. Epidemics of encephalitis suggestive of Japanese encephalitis (JE) were described in Japan from the 1870s onwards. Four genotypes of JEV have been characterised and representatives of each genotype have been fully sequenced. Based on limited information, a single isolate from Malaysia is thought to represent a putative fifth genotype. We have determined the complete nucleotide and amino acid sequence of Muar strain and compared it with other fully sequenced JEV genomes. Muar was the least similar, with nucleotide divergence ranging from 20.2 to 21.2% and amino acid divergence ranging from 8.5 to 9.9%. Phylogenetic analysis of Muar strain revealed that it does represent a distinct fifth genotype of JEV. We elucidated Muar signature amino acids in the envelope (E) protein, including E327 Glu on the exposed lateral surface of the putative receptor binding domain which distinguishes Muar strain from the other four genotypes. Evolutionary analysis of full-length JEV genomes revealed that the mean evolutionary rate is 4.35 × 10(-4) (3.4906 × 10(-4) to 5.303 × 10(-4)) nucleotides substitutions per site per year and suggests JEV originated from its ancestral virus in the mid 1500s in the Indonesia-Malaysia region and evolved there into different genotypes, which then spread across Asia. No strong evidence for positive selection was found between JEV strains of the five genotypes and the E gene has generally been subjected to strong purifying selection.
Human enterovirus 71 (EV-71) is genotyped for molecular epidemiological investigation mainly using the two structural genes, VP1 and VP4. Based on these, EV-71 is divided into three genotypes, A, B and C, and within the genotypes B and C, there are further subgenotypes, B1-B5 and C1-C5. Classification using these genes is useful but gives incomplete phylogenetic information. In the present study, the phylogenetic relationships amongst all the known EV-71 and human enterovirus A (HEV-A) isolates with complete genome sequences were examined. A different tree topology involving EV-71 isolates of subgenotypes, C4 and B5 was obtained in comparison to that drawn using VP1. The nucleotide sequence divergence of the C4 isolates was 18.11% (17-20%) when compared to other isolates of subgenotype C. However, this positions the C4 isolates within the cut-off divergence value of 17-22% used to designate the virus genotypes. Hence, it is proposed here that C4 should be designated as a new genotype D. In addition, the subgenotype B5 isolates had an average nucleotide divergence of only 6.14% (4-8%) when compared to other subgenotype B4 isolates. This places the B5 isolates within the subgenotype B4. It is proposed here that the B5 isolates to be redesignated as B4. With the newly proposed genotype D and inclusion of subgenotype B5 within B4, the average nucleotide divergence between genotypes was 18.99% (17-22%). Inter- and intra-subgenotype average divergences were 12.02% (10-14%) and 3.92% (1-10%), respectively. A phylogenetic tree built using the full genome sequences is robust as it takes into consideration changes in the sequences of both the structural and non-structural genes. Similar nucleotide similarities, however, were obtained if only VP1 and 3D RNA polymerase genes were used. Furthermore, addition of 3D RNA polymerase sequences will also show recombination events. Hence, in the absence of full genome sequences, it is proposed here that a combination of VP1 and 3D RNA polymerase gene sequences be used for initial genotyping of EV-71 isolates.
There have been three major rabies epidemics in China since the 1950s. To gain more insights into the molecular epidemiology of rabies viruses (RVs) for the third (the current) epidemic, we isolated RV from dogs and humans in major endemic areas, and characterized these isolates genetically by sequencing the entire glycoprotein (G) gene and the G-L non-coding region. These sequences were also compared phylogenetically with RVs isolated in China during previous epidemics and those around the world. Comparison of the entire G genes among the Chinese isolates revealed up to 21.8% divergence at the nucleotide level and 17.8% at the amino acid level. The available Chinese isolates could be divided into two distinct clades, each of which could be further divided into six lineages. Viruses in clade I include most of the Chinese viruses as well as viruses from southeast Asian countries including Indonesia, Malaysia, the Philippines, Thailand, and Vietnam. The viruses in the other clade were found infrequently in China, but are closely related to viruses distributed worldwide among terrestrial animals. Interestingly, most of the viruses isolated during the past 10 years belong to lineage A viruses within clade I whereas most of the viruses isolated before 1996 belong to other lineages within clades I and II. Our results indicated that lineages A viruses have been predominant during the past 10 years and thus are largely responsible for the third and the current epidemic in China. Our results also suggested that the Chinese RV isolates in clade I share a common recent ancestor with those circulating in southeast Asia.
Anopheles sundaicus s.l. is a malaria vector in coastal areas of Southeast Asia. Previous studies showed at least four distinct species within the complex. The present study investigated the phylogeography and the status of A. sundaicus s.l. populations from Cambodia, Thailand, Malaysia and Indonesia with regard to A. sundaicus s.s. from Sarawak, Malaysian Borneo and A. epiroticus in Vietnam and Thailand. Three lineages recovered by analyses of Cyt-b and COI (mtDNA) confirmed the presence of A. sundaicus s.s. in Malaysian Borneo, the distribution of A. epiroticus from southern Vietnam to peninsular Malaysia, and recognised a distinct form in Indonesia that is named A. sundaicus E. The phylogenetic and demographic analyses suggest that the three species were separated during the Early Pleistocene (1.8-0.78 Myr) and experienced bottlenecks followed by a genetic expansion in more recent times. Based on the results and knowledge of the biogeography of the area, we hypothesise that the combination of cyclical island and refugium creation was the cause of lineage isolation and bottleneck events during the Pleistocene.
Recurrent vulvovaginal candidiasis affects women worldwide and the resistance to azole drugs may be an important factor. The extent of strain-to-strain variation within a species and its relationship to the ability of the organism to colonize the vulvovaginal mucosa is not well established. The aims of this study were to compare: (i) the genotypes of Candida strains in sequential infections in patients with recurrent vaginitis, (ii) the genotypes of strains in patients with only one episode of infection in a period of 1 year and (iii) determine the in vitro antifungal susceptibilities of strains that cause recurrent vaginitis. Fifty-one cultured specimens from six distinct Candida species were genotyped via random amplified polymorphic DNA-polymerase chain reaction (RAPD-PCR) method using the ERIC1 and ERIC2 primers (ERIC, enterobacterial repetitive intergenic consensus). Statistical analyses allowed three different scenarios to be discerned for recurrent cases: (i) strain maintenance without genetic variation, (ii) strain maintenance with minor genetic variation and (iii) outright strain replacement. The genetic relatedness between strains from patients with recurrent vaginitis and patients with single episode of vaginitis were demonstrated by the dendogramme and the mean pairwise similarity coefficient S(AB) for the intergroup comparison was 0.223. However, intragroup genetic relatedness was slightly higher than intergroup comparison, with mean S(AB) of 0.261 and 0.331 for Groups I and II, respectively. A high proportion of Group I isolates (87.5%) causing recurrent infections were resistant to ketoconazole, whereas 41.7% of these isolates were cross-resistant to both clotrimazole and ketoconazole as shown by the in vitro antifungal susceptibility test, especially for C. glabrata isolates. Pregnancy status of patients displayed a highly significant association with C. albicans species whereas non-albicans species had a markedly higher prevalence in non-pregnant patients (p<0.001). These results may have a profound impact on the management of vaginal candidiasis, especially in recurrent cases.