Background

Candida africana was isolated, for the first time, in 1995 in Madagascar, Africa and afterwards proposed as new Candida species phylogenetically closely related to C. albicans [1]. The isolates assigned to this group were originally proposed as representatives of a new species rather than a biovariant of C. albicans. And it shows remarkable differences if compared to C. albicans such as the taxonomic status. It has been reported that C. africana showed poor adhesion ability to human Hela cells, absence of biofilm formation, a notable low level of filamentation and significantly less pathogenic than C. albicans in the galleria mellonella insect systemic infection model [2,3,4]. Candida africana represents the phenotypic variation occurring in C. albicans [5]. Candida africana colonized mainly in human genitalia and cause vaginal infections [6, 7]. Its distribution appears to be worldwide [5,6,7,8,9,10,11,12]. The purpose of this study was to explore multilocus sequence typing (MLST) analysis of C. africana from vulvovaginal candidiasis and its relevance to clinical characteristics, virulence, pathogenicity, and antifungal profiles.

Methods

Fungal isolates and genotyping

The isolates were from vaginal specimens of patients with VVC and vaginal samples of pregnant women. Genomic DNA was extracted by using the E.Z.N.A™ Yeast DNA Kit (OMEGA, USA) according to the manufacturer’s instructions HWP1 gene (primers CR-f 5-GCTAC CACTTCAGAATCATCATC-3 and CR-r 5-GCAC CTTCAGTCGTAGAGACG-3) amplification was performed and the PCR products was electrophoresed in 1.2% agarose gel in TBE buffer to distinguish C. albicans, C. dubliniensis, and C. africana on the basis of the distinct size of the amplicons as previously described [9, 13] (Additional file 6: Figure S1).

The ABC genotype and mating-type was determined by using previously designed primers and experimental conditions by McCullough et al. [14] and Hull et al. [15]. Multilocus sequence typing for C. africana was performed on the basis of a previously published consensus set of seven housekeeping gene loci for C. albicans: CaAAT1a, CaACC1, CaADP1, CaMPIb, CaSYA1, CaVPS13 and CaZWF1b [16]. ALL seven loci were sequenced in both the forward and reverse directions and sequence data were assembled using ContigExpress software and checked manually for heterozygous polymorphisms. Allele numbers of each locus was determined by inputting the sequences in the C.albicans MLST database (http://calbicans.mlst.net) and diploid sequence types (DSTs) for each isolate was determined by the composite profile of all seven allele numbers. The MLST results of seven sequenced loci were concatenated into a single sequence to phylogenetic analysis by unweighted-pair group method using average linkages (UPGMA), determined by P distances using MEGA 6 software (http://www.megasoftware.net/).

Candida albicans SC5314 and Candida albicans ATCC90028 were used as reference strains.

Quantitative real-time PCR

Quantitative real-time PCR analysis of following genes: ALS1–7, ALS9, HWP1, SAP1-SAP10, PLB1–3, PLB5, CDR1, CDR2, MDR1, ERG11 and HXK1 were performed as described by previous publications [17, 18]. Each experiment was carried out with ACT1 as the housekeeping gene. Data for each target gene were calculated as fold change in comparison with the reference gene ACT1 using the ΔΔCt quantification method. The primers used for quantitative real-time PCR analysis are shown in Additional file 1: Table S1.

Sequencing of the HXK1 gene

The DNA region of the HXK1 gene carrying the mutation was amplified by or using the primer pair described by Felice et al. [19]. The amplified fragment was sequenced in both directions and sequence data were assembled using ContigExpress and compared with the corresponding wild-type C. albicans and C. africana reference strains respectively.

Extracellular enzymes assays

The activity of phospholipase, esterase and haemolysis enzyme production was evaluated by or using plate assays methodology described by Pakshir et al. [20] and Sanita et al. [21] with modifications.

Antifungal susceptibility testing

Susceptibilities of the isolates to antifungal drugs were assayed by the broth microdilution method in 96-well plates according to the proposed Clinical and Laboratory Standards Institute M27-A2 and M27-A3 [22, 23].

Biofilm production assays

The biofilm production assays was previously published by Pierce al. [24] using a rapid reproducible 96-well microtiter-based method. Each strain was performed in triplicate wells. Quantification of biofilm formation was performed at 8 h, 24 h, 48 h and 72 h respectively after inoculation.

Statistical analysis

The statistical significance of the results was determined by the t test and Pearson's chi-squared test, using SPSS version 11.5 (SPSS, Inc., Chicago, IL). The results were considered statistically significant with P values of less than 0.05.

Results

The clinical characteristics of VVC caused by C. africana were shown on Table 1. All the patients were symptomatic. Mycological cure rate of VVC caused by C. africana was 97% (33/34 cases) and 91% (31/34 cases) at day 7–14 and day 30–35 follow-up when treated with commonly used antifungal agents (Table 1).

Table 1 Demographic and clinical characteristics of 34 patients with vulvovaginal candidiasis caused by C. africana
Full size table

The MLST analysis revealed that the DST782 and DST182 were the main MLST genotypes in C. africana from vaginal specimens of patients with VVC (Fig. 1; Additional file 2: Table S2).

Fig. 1
figure 1

UPGMA dendrogram of 43 C. africana strains based on the seven loci used in the MLST analysis and DSTs assigned by C. albicans MLST database. 43 C. africana strains were divided into 1 clade in the UPGMA dendrogram. DST782 and DST182 were the main MLST types of C. africana

Full size image

Additional file 8: Figure S3 was electrophoretogram of real time PCR products of virulence genes and drug resistance genes. PCR products of C. africana were electrophoresed in agarose gel in TBE buffer and further verified the correct fragment sizes and specificities of primers used in the study. Compared with C. albicans SC5314 and C. albicans ATCC90028, the expression of virulence genes ALS1, ALS5, ALS6, SAP3 and SAP4 decreased in C. africana strains (Fig. 2; Additional file 4: Table S4).

Fig. 2
figure 2

Expression of the virulence genes in C. africana and C. albicans by quantitative real-time PCR. Compared with the control strains C. albicans ATCC90028 and SC5314, the expression of virulent genes ALS1, ALS5, ALS6, SAP3 and SAP4 decreased in C. africana

Full size image

The expression of drug resistance genes CDR1, CDR2 and MDR1 varied among strains of C. africana (Fig. 3; Additional file 3: Table S3). There were statistically significant differences in biofilm formation and phospholipase activity between C. africana and C. albicans. Assessment of biofilms at 8 h, 24 h, 48 h and 72 h at OD490 nm in microtiter plate reader showed biofilm production was significantly lower in C. africana than that in C. albicans (Fig 4).

Fig. 3
figure 3

Expression of the drug resistance genes in C. africana and C. albicans by quantitative real-time PCR. The expression of CDR1, CDR2, and MDR1 varied among C. africana

Full size image
Fig. 4
figure 4

Assessment of biofilms at 8 h, 24 h, 48 h and 72 h at OD490 nm in microtiter plate reader. Biofilm production were significantly lower in C. africana than that in C. albicans

Full size image

The sequencing of the partial HXK1-gene of C. africana strains showed that the region contained the guanine insertion (Additional file 9: Figure S4). All isolates of C. africana and control C. albicans ATCC90028 and SC5314 displayed positive phospholipase, hemolytic and esterase activities. However, C. africana showed less active in phospholipase production (Additional file 7: Figure S2). All C. africana strains examined were unable to produce chlamydospores on cornmeal agar plates supplemented with 1% Tween 80 (Qingdao Hopebio-Technology, China) and did not grow in tubes containing GlcNAc as the sole carbon source. All C. africana were the genotype A of C. albicans and heterozygous [a/α] at the mating-type-like locus.

The mycological cure rate of patients infected with C. africana was high when treated with commonly used antifungal agents, which was consistent with in vitro antifungal susceptibilities (Table 2). All C. africana isolates were susceptible to the tested antifungal agents (Tables 3 and 4).

Table 2 Therapeutic efficacy in patients with VVC caused by candida africana
Full size table
Table 3 In vitro antifungal susceptibilities of 43 clinical C. africana isolates as determined by the Clinical and Laboratory Standards Institute (CLSI) method
Full size table
Table 4 In vitro antifungal susceptibilities of 43 clinical C. africana isolates as determined by the Clinical and Laboratory Standards Institute (CLSI) method
Full size table

Discussion

MLST and sequence variability

Candida africana shows a global distribution and mainly causes genital infections [7]. Review of C. africana in literature has been summarized in Additonal file 5: Table S5 and displayed as a world map in Additonal file 10: Figure S5. By using a single pair of primers derived from HWP1 genes, Romeo and Criseo [13] described an accurate molecular assay for the discrimination among C. africana, C. albicans, and C. dubliniensis. DTS182 was the most common and geographically dispersed strain type having been isolated in Europe, South America, Africa and Asia [6,7,8]. DST782 isolates were the commonest Asian genotype [6, 7]. All these C. africana strains exhibit a phenotypic and genetic homogeneity that is different from their country of origin [7]. In current study, 5 of the 7 MLST loci sequenced showed the same sequences and without single nucleotide polymorphism (SNP) diversity. This suggests a reasonable low level of divergence in the population structure of C. africana [7, 10]. The low level of sequence change also suggests that this typing technique may not be ideal for local epidemiological studies [6, 7].

Phenotypic characterizations, virulence and pathogenicity

Compared to C. albicans, C. africana shows remarkable phenotypically differences such as the inability to produce characteristic chlamydospores and the incapacity to assimilate many carbon sources especially N-acetylglucosamine(GlcNAc), which plays an important role in cell signaling [1, 5, 25]. DNA sequence analysis of the HXK1 gene, encoding the enzyme GlcNAc-kinase, demonstrated the existence of a specific mutation [guanine insertion] that impairs GlcNAc assimilation in C. africana [19]. All of our C. africana carried a specific HXK1 gene mutation and were mating-type a/α and genotype A of C. albicans isolates.

Candida africana shows a poor adhesion to human epithelial cells as demonstrated by using both mammalian and insect infection models [2]. In our current study, there were statistically significant differences in biofilm formation, phospholipase activity between C. africana and C. albicans. All of our patients were symptomatic, which were consistent within vitro extracellular enzymes assays and the expression of virulence genes.

The clinical signs in the patients infected with C. africana were almost identical with those with C. albicans [9].

Antifungal susceptibilities and therapy outcome

Candida africana exhibits susceptibility patterns of commonly used antifungal agents similar to those of C. albicans [8, 9]. However, Al-Hedaithy and Fotedar [26] reported that chlamydospore-negative combined with germ tube negative isolates were resistant to fluconazole and itraconazole. In agreement with previous reports [6, 8, 27], our C. africana isolates tested were found to be susceptible to all antifungal agents. Mendling et al. [28] reported that C. africana could be eradicated by a single dose vaginal tablet containing 500 mg clotrimazole.

The mycological cure rate of patients infected with C. africana was high when treated with commonly used antifungal agents, which was consistent with in vitro antifungal susceptibilities.

Conclusions

Candida africana, a worldwide distribution candida, mainly cause vaginal infections and appears to be with a low level of sequence variation in MLST loci. Candida africana, a lower virulence and pathogenicity Candida, is susceptible to commonly used antifungal agents and have a high mycological cure rate.