Prevalence of Closely Related Candida albicans Species among Patients with Vulvovaginal Candidiasis in Southern Poland Based on the hwp1 Gene Amplification

Despite advances in the treatment of fungal infections, vulvovaginal candidiasis (VVC) remains a common problem among women of childbearing age worldwide (Sustr et al. 2020). This disease affects women to varying degrees, regardless of ethnic origin, and social or living conditions. This ailment can be induced by various factors ranging from host factors such as age, estrogen levels, sexual activity, exposure to stress, allergic diseases, or medications (Farr et al. 2021). However, most cases do not have an identifiable trigger. It is estimated that 70–75% of women will be diagnosed with vulvovaginal candidiasis at least once in their lifetime (Hedayati et al. 2015), and 5–8% of adult women have recurrent vulvovaginal candidiasis (RVVC), which is defined as four or more episodes per year (Sobel 1992). VVC affects approximately 138 million women annually (in the range of 103 to 172 million), with a global annual prevalence of 3,871 per 100,000 women; 372 million women are affected by RVVC during their lifetime (Denning et al. 2018). More than 40 species of the Candida genus are capable of causing infection (Pfaller et al. 2014), but Candida albicans remain a significant cause of VVC (Sobel and Sobel 2018). The incidence of C. albicans vulvovaginitis is 70% to 90%.

The taxonomy of some Candida species has undergone significant changes due to the development of identification techniques, primarily molecular biology. Due to difficulties in identification, the newly described related species were grouped into complexes and named cryptic yeasts. The C. albicans complex includes three closely related species C. albicans sensu stricto, Candida dubliniensis, and Candida africana (Sullivan et al. 1995; Tietz et al. 2001). C. dubliniensis and C. albicans have many phenotypic resemblances, such as the capacity to form pseudohyphae, chlamydospores on cornmeal agar, and germ tubes in serum (Moran et al. 2012). Phylogenetic studies indicate that C. dubliniensis is most closely related to C. albicans (McManus et al. 2008), and it is often difficult to discriminate between these two species in clinical samples. The second species from the C. albicans complex, C. africana, had previously been considered an atypical C. albicans strain. However, in 2001 it was proposed to isolate a separate species of C. africana that produces pseudohyphae but not chlamydospores (Tietz et al. 2001). Several other features, such as the inability to grow in 42°C, the incapacity to assimilate the aminosugars N-acetyl glucosamine and glucosamine, as well as the disaccharide trehalose, and the organic acid DL-lactate, could be helpful in proper identification, but during routine diagnostics misidentification often occurs in mycological laboratories (Sullivan et al. 2004). To avoid these drawbacks of classical techniques, molecular analyses have been applied in multiple studies to identify and distinguish between Candida species (Byadarahally Raju and Rajappa 2011; Neppelenbroek et al. 2014). A singleplex PCR assay using a single primer pair targeting the hyphal wall protein 1 (hwp1) gene encoding a mannoprotein specific for the filamentous form of C. albicans, has been proposed to distinguish C. albicans sensu stricto, C. dubliniensis, and C. africana according to the distinct size of the amplicon (Romeo and Criseo 2008; 2009).

The available clinical and epidemiological data show that C. africana and C. dubliniensis are isolated from patients all over the world; however, there are scant data on the frequency of occurrence of these species in Poland, identified as the cause of VVC. Due to limited knowledge concerning its prevalence, this study aims to confirm the phenotypic identification of C. albicans and its closely related species isolated from the genital tract by amplifying the hwp1 (hyphal wall protein 1) gene in the PCR assay.

Results of phenotypic analysis. According to a phenotypic analysis based on culture on chromogenic media, all 326 Candida strains included in the study were obtained from patients diagnosed with vulvovaginal candidiasis. In addition, the presence of aerobic bacteria in vaginal swabs was confirmed, such as: Gram-positive cocci (Staphylococcus spp.) in one patient, streptococci (group B Streptococcus – GBS) in 55 patients, Gram-negative bacilli (Escherichia coli in 13, and Klebsiella spp. in one patient) and colonization by anaerobic bacteria (G. vaginalis) in 73 patients (Table I). All women with co-occurrence of Candida spp. with bacteria presented reduced Lactobacillus spp.

Results of molecular analysis. Based on the PCR assay targeting the hwp1 gene, preceded by MALDI TOF analysis, we confirmed the initial phenotypic identification in 307 (94.17%) samples as C. albicans, while 17 (5.22%) strains belonged to the C. dubliniensis species (Table II). Moreover, molecular identification demonstrated two co-infections (0.61%) by C. albicans and C. dubliniensis. No strain of C. africana was detected (Fig. 2).

Fig. 2.

Comparison of C. albicans complex identification results from VITEK^®2, MALDI-TOF MS, and PCR assays. Phenotypic analysis based on chromogenic medium (chromID^® Candida; bioMérieux, France) identified all clinical isolates as C. albicans strains. Despite slight differences in the appearance and colors of the yeast colonies (Fig. 1B), no distinction between C. albicans sensu stricto, C. dubliniensis, and C. africana species was possible according to the manufacturer’s recommendations.

The reference method for this study found as “a gold standard”, was molecular identification based on the hwp1 gene amplification, which confirmed that all reference strains comprised the C. albicans complex: C. albicans sensu stricto (ATCC^® 90028^™), C. dubliniensis (NCPF 3949), and C. africana MYA 2669^™.

The VITEK^®2 system (bioMérieux, France) showed agreement with the molecular assay in 97.85% of clinical strains. Most C. albicans sensu stricto were identified at an excellent and very good discrimination level. All the incorrect identification results (2.15%) referred to C. dubliniensis strains (Table III, IV), which were finally discriminated by PCR assay. The VITEK^®2 system did not allow identification of C. africana. The reference strain of C. africana MYA 2669^™ was identified as C. dubliniensis/C. parapsilosis at the low discrimination level (Table IV).

The protein analysis by MALDI-TOF MS (VITEK^®2 MS Biotyper; bioMérieux, France) did not require lengthy biochemical reactions. It was more accurate method than the biochemical analysis with the VITEK^®2 system. Based on MALDI-TOF MS, 99.08% of clinical isolates were correctly identified as C. albicans or C. dubliniensis. Only 0.92% of the identification results differed from the results obtained with PCR (Table III). However, MALDI-TOF MS failed to detect the C. africana reference strain, and neither did the VITEK^®2 system (Table IV).

Results of antifungal susceptibility testing. Antifungal susceptibility testing covered susceptibility to azoles, which are recommended for VVC. The susceptibility to amphotericin B and 5-flucytosine was found clinically unsuitable for VVC treatment (Workowski et al. 2021).

In total, 100% of C. albicans isolates were susceptible to fluconazole and ketoconazole, while 99.35% (305/307) of them were susceptible to miconazole, and 96.74% (297/307) to itraconazole. The ratio of C. albicans intermediate to itraconazole (referring to “susceptible, increased exposure” according to the EUCAST recommendations (EUCAST 2020)) amounted to 3,26% (10/307). Only 0.65% (2/307) were intermediate resistant to miconazole (Fig. 3). All C. dubliniensis strains were susceptible to all antifungals tested.

Fig. 3.

Vulvovaginal candidiasis (VVC) remains a threatening problem among women (Sustr et al. 2020). All 326 Candida strains included in the study were obtained from women with confirmed VVC. Furthermore, in 128 cases (~ 39%), the coincidence of yeast and aerobic and/or anaerobic bacteria was revealed by phenotypic methods. We are in line with the other observations that the occurrence of yeasts in the genital tract disrupted the natural bacterial microbiota, especially the number of Lactobacillus spp., which is generally considered to constitute a major part of the normal vaginal microbiota and are natural competitor of Candida spp. (Zangl et al. 2019).

C. albicans is the most common isolated species among women with VVC; however, there are still problems with discriminating closely related species among C. albicans complex; thus, their prevalence and clinical importance remains misestimated. Despite the phenotypic and biochemical differences (e.g., production chlamydospores and ability to assimilate N-acetylglucosamine and glucosamine), the identification of C. africana remains challenging with the use of currently existing methods in routine laboratory practice (Tietz et al. 2001; Criseo et al. 2015).

One of the most commonly utilized, rapid, and cost-effective tests are chromogenic media, which are frequently used to detect and identify yeasts of the Candida genus. However, culture-based methods may still need to be improved as they are often influenced by various culture conditions (Tietz et al. 2001; Criseo et al. 2015; Scharmann et al. 2020). Despite subtle differences in the C. albicans sensu stricto, C. dubliniensis, and C. africana colonies morphology on the ChromID^® medium used in this study, no distinction of these species is possible according to the manufacturer’s recommendations (Fig. 1B).

The other method commonly used in routine laboratory practice for yeast identification is the VITEK^®2 system. This automated system is based on kinetic analysis detecting metabolic changes and continuous colorimetric-based monitoring of reactions which provides reliable identification of 27 medically relevant species of Candida genus, including both C. albicans and C. dubliniensis. To this day, there is no possibility of C. africana identification by VITEK^®2, ID YST Card. Identifying newly described species, species of modified systematic names, and rare ones is problematic for most commercial systems because the databases do not include a representative set of profiles, resulting in misidentifying closely related microorganisms.

Due to the limited ability of phenotypic approaches, such as chromogenic media and the VITEK^®2 system, to identify rare species and to confirm the identification of C. africana isolates, phenotypic tests have to be supplemented with molecular methods.

According to the performed molecular assay based on the hwp1 gene amplification, we revealed that the diagnosis of 326 strains initially identified by phenotypic methods as C. albicans was confirmed in 94.17% (307/326), while 5.21% (17/326) strains were identified as C. dubliniensis. Moreover, molecular identification allowed the detecting of two cases of co-infection with C. albicans and C. dubliniensis. Although C. africana is isolated from vaginal specimens worldwide, we could not isolate any species of this organism from the presumptive vaginal isolates of C. albicans.

The meta-analysis provided by Ghareghbolagh et al. (2020), showed that the distribution of C. albicans closely related species varies among the world’s regions, and the global prevalence of C. africana among C. albicans complex reached 1.67%. Our results align with those from Turkey, Malaysia, and China, where 195, 98, and 87 Candida isolates from vulvovaginitis cases were detected (Gumral et al. 2011; Yazdanpanah and Khaithir 2014; Hu et al. 2015). Furthermore, a single study from Iran has confirmed the lack of C. africana in vaginal specimens, even though the overall prevalence of C. africana in this country was one of the highest (~ 3%) (Pakshir et al. 2017). It confirms that variations in the prevalence of this species could be observed among and across continents and even within one country.

Epidemiological data indicated that the incidence of C. dubliniensis in vaginal samples has varied within the tested population from 0 to even 29.52% (Jamilian et al. 2007; Shan et al. 2014). Several studies demonstrated the very low prevalence of C. dubliniensis in VVC/RVVC patients (Theill et al. 2016; Hazirolan et al. 2017; Mucci et al. 2017). Our results based on molecular discrimination confirmed that 5.22% of the VVC cases were caused by C. dublinienis and approximated those reported by Borman et al. (2013) (6%). Furthermore, we revealed two patients (0.61%) who presented with concurrent co-infection by C. albicans and C. dubliniensis. Although C. dubliniensis has been frequently associated with oral infections in HIV-positive patients, it has also demonstrated significant levels of adhesion to vaginal epithelial cells (average adherence 85.6%), suggesting that this species is well adapted, in terms of adhesion capability, to the vaginal environment (Vidotto et al. 2003).

Multiple global studies have shown that C. albicans and the related species isolated from VVC patients are usually susceptible to antifungal agents (Theill et al. 2016). The results of the Fungitest™ assay used in this study confirmed these findings. Whereas only single strains of C. albicans have shown reduced susceptibility to itraconazole and miconazole, all C. dubliniensis strains were susceptible to all antifungals tested. However, C. dubliniensis is known to develop rapid resistance to fluconazole, and consequently, patients do not respond to standard doses of fluconazole (Kolekar et al. 2019). Furthermore, global overuse of azole antifungal agents without prescription for the treatment of Candida vaginal infection and repeated exposure to antifungal agents can lead to the induction of azole resistance in strains (Naeimi et al. 2018; Mohammadi et al. 2021). In our study, we did not observe any strains resistant to fluconazole. At the same time, resistance to this antifungal was frequently reported among C. albicans isolates from VVC patients, with a resistance rate ranging from 14.7% to even 53.7% (Mohammadi et al. 2021). The microbroth dilution method is the reference method for antifungal susceptibility testing of glucose-fermenting yeasts (EUCAST 2020; CLSI 2022); however, the commercial kits (e.g., Fungitest™; BioRad, France) are still in use as a screening test for the detection of antifungal efficacy (Cuenca-Estrella et al. 2005). The microdilution method recommended by EUCAST is demanding and rarely used in routine diagnostics of C. albicans strains isolated from women with VVC. Methods such as E-test strips or automatic VITEK^®2 YST08 methods are expensive and are not applicable in the routine diagnosis of genital tract infections. In addition, the VITEK^®2 system and the available anti-biogram charts provide data for fluconazole, only one of the drugs used in VVC treatment. Moreover, the interpretation criteria according to EUCAST for drugs used in the treatment of VVC apply only to fluconazole and itraconazole. There are currently no other recommendations regarding topical antifungal drugs (e.g., miconazole, econazole, clotrimazole).

According to the Centers for Disease Control and Prevention recommendations (CDC), treatment of uncomplicated VVC (which constitutes the majority of VVC) is based on empirical therapy. Culture and antifungals susceptibility testing should be considered only for patients who remain symptomatic after failure of the empirical treatment (Workowski et al. 2021). However, according to the demands reported by the physicians, susceptibility tests are performed. Currently, there are few commercial tests available that can be used to assess the susceptibility testing of drugs applicable in the treatment of VVC. Fungitest^™ (BioRad, France) was the only method used to test antifungal susceptibility in this study and is still routinely used in many microbiological laboratories in Poland. However, the concentrations of drugs available in this kit do not meet the current European (EUCAST 2020) and American criteria (CLSI 2022). According to the manufacturer’s instructions, the interpretation of drug susceptibility is based only on scientific reports. Moreover, this semi-quantitative method has many advantages: it does not require complex handling and is cost-effective. Furthermore, the agreement with the European Committee on Antibiotic Susceptibility Testing (EUCAST) data for Fungitest^™ was high, with correlation indices of 0.95 (p < 0.01) (Cuenca-Estrella et al. 2005).

C. albicans sensu stricto remains the major factor of vulvovaginal candidiasis, while C. dublinienis, and C. africana are a minor species within the C. albicans complex (Gharehbolagh et al. 2020). Our results based on molecular discrimination confirmed that the most common species causing vulvovaginal candidiasis in Southern Poland is C. albicans sensu stricto (94.17%). However, we argue that the reported incidence of C. dubliniensis and C. africana suffers from certain limitations, such as a low number of studies from each country and the unavailability of data for the majority of countries. While it has been estimated that C. albicans is responsible for 78.3% of genital tract candidiasis in Poland (Wójkowska-Mach et al. 2021), the regional occurrence of closely related C. albicans species such as C. dubliniensis and C. africana is unknown. As far as we know, this is the first study concerning the prevalence of C. albicans and its closely related species isolated from the genital tract in our country.

The global prevalence of C. albicans may be overestimated due to a lack of routinely used tests aimed at the discrimination of closely-related species; therefore, C. dubliniensis and C. africana isolates continue to be misidentified as C. albicans (Shokoohi et al. 2021). Due to minor phenotypic differences between closely-related species belonging to the C. albicans complex, it is not easy to recognize them properly by routine diagnostic methods (e.g., biochemical tests, chromogenic media).

A PCR assay targeting the hwp1 gene was reliable in identifying species among the C. albicans complex. Its simplicity, rapidity, and selectivity make this method helpful in investigating C. albicans and its closely related species epidemiology.