A study of the association between virulence factors and antifungal susceptibility profile of Candida species recovered from cases of vulvovaginal candidiasis

ABSTRACT

Background:

The incidence of mycotic infections, especially of Candida, has gradually increased over the past few years. In clinical practice, azoles are the most frequently used antifungal agents and the growing incidence of systemic candidiasis and resistance to antifungals have become a matter of concern worldwide. Virulence factors in Candida spp. may be critical for predicting the response of antifungal drugs.

Objectives:

This study aimed to identify the relationship between virulence factors and the antifungal susceptibility of Candida.

Methodology:

This cross-sectional study was conducted on a sample of 55 Candida strains isolated from vulvovaginal samples of patients in the reproductive age group, presenting with signs and symptoms of vulvovaginitis in a large tertiary care hospital in central India.

Results:

A majority of the Candida were sensitive to three tested drugs (89% to amphotericin B, 76.4% to fluconazole, and 89.1% to voriconazole). Resistance to fluconazole was highest at 16.4%. No significant relationships were identified between antifungal sensitivity of the three azoles with biofilm formation, phospholipase, or proteinase synthesis.

Conclusions:

High level of antifungal resistance to the three antifungals, especially to voriconazole, is worrisome; however, none of the virulence markers have a significant association with antifungal sensitivity of Candida species. This finding rules out the effect of the virulence of the pathogen on drug response.

Introduction

Fungal infections in humans are relatively common and range from common or mild superficial infections to life-threatening invasive infections. The incidence of mycotic infections has gradually increased over the past few years. Recent reports cite more than 1.6 million people worldwide to have suffered from serious fungal diseases that can even be fatal.[1]

Among the fungal pathogens, Candida spp. are most common.[2] Though they form a part of the normal flora, Candida spp. may cause diseases ranging from superficial candidiasis to life-threatening disseminated infections in immunocompromised hosts.[3–6] A majority of the invasive infections are caused by five species of Candida, namely, C. albicans, C. parapsilosis, C. glabrata, C. tropicalis, and C. krusei.[7] Although C. albicans is the predominant species involved in both superficial and disseminated infections, there has been a significant increase in the number of infections caused by non-albicans Candida (NAC) species.[8] An increase in the incidence of invasive fungal infections is attributed to the widespread use of broad-spectrum antibiotics, hormones, immunosuppressants, chemotherapy, and central venous catheters.[9]

Candida may colonize several organs including the skin and the respiratory, gastrointestinal, and genitourinary tracts.[10,11] It has many virulence factors that determine its invasion of host tissues. Virulence factors of Candida include biofilm production, production of tissue-damaging hydrolytic extracellular enzymes like proteinases, phospholipases, and hemolysin, besides adherence to host tissues as well as medical devices and formation of pseudohyphae.[12–14] In clinical practice, antifungal agents of various classes are used for treating fungal infections,[15] of which azoles are the most frequently used and are studied widely for their pharmacological properties, mode of action, and resistance by microorganisms.[16] Growing incidence of systemic candidiasis and resistance to antifungals have become a matter of concern worldwide. Accurate identification of virulence factors in Candida spp. is therefore critical for predicting the response of antifungal drugs and detecting the emergence of strains with greater resistance.

In light of the above-mentioned observations, that is, a widening spectrum of disease-causing Candida species coupled with increasing antifungal resistance, it is crucial to identify the relationship between virulence factors and antifungal susceptibility of Candida. For this study, our null hypothesis stated that the response of Candida to antifungal agents has no relation with the presence of virulence factors. Although there is a broad range of literature on virulence and antifungal resistance of Candida species separately, only a small portion of studies have examined the effect of resistance on virulence.[17] Family physicians and primary care physicians, besides the gynecological specialists, often have to deal with cases of Candida infection. Consequently, it is of high importance for them to be at pace with changing epidemiology and antifungal resistance of Candida species. In this paper, we describe the results from an analysis of 55 Candida strains isolated from vulvovaginal samples from patients in the reproductive age group, presenting with signs and symptoms of vaginitis in a large tertiary care hospital in central India. It is believed that the information generated from this study will be particularly useful for clinicians and help to improve patient care.

Material and Methods

This cross-sectional study included Candida specimen isolated from 55 women who tested positive for vulvovaginal candidiasis in the obstetrics and gynecology department of a teaching hospital of central India. For detection of Candida, three high vaginal swabs per patient were collected using sterile cotton-tipped swabs and transported within 30 minutes to the lab in sterile tubes without use of any transport media. The ethical approval was obtained from the institutional ethics committee vide number No.Comm.Med. 2016/779.

Processing of specimen: One swab was used for wet mount and second for Gram staining. Round-to-oval budding cells with or without pseudohyphae were considered as positive for Candida. Third swab was used for culture on Sabouraud dextrose agar (SDA), one plain and another with chloramphenicol (0.05 mg/ml) and incubated at 25°C for three to four days. Identification of culture growth was done based on colony characteristics, Gram staining, and germ tube test. Morphological characters on cornmeal agar with Tween 80 including the pattern of growth and presence or absence of chlamydospores were observed. Final identification using fermentation and assimilation tests was done.[18]

i. Detection of Virulence Factors:

a) In vitro biofilm formation/Slime Production: A loopful of Candida colony from the surface of SDA plate were inoculated into polystyrene tubes containing 10 ml of SDA broth supplemented with glucose (final conc. 8%) and incubated at 35°C for 48 hours. The broth was aspirated out, tubes washed with distilled water twice and its walls stained with 2% safranine for 10 minutes. They were then examined for the presence of an adherent layer. The biofilm production was scored as follows:

Status of Biofilm Formation	Interpretation
No formation	Negative
Formation at the bottom of the tube	1+ (Weak positive)
Formation both at the bottom and on the internal wall of the tube	2+ (Moderate positive)
Formation at the bottom, on the internal wall, and at the top of the tube (ring)	3+ (Strong positive)

Positive control: Staphylococcus epidermidis (ATCC35984) was used.[19]

b) Proteinase detection: Candida proteinase was detected by the slightly modified Staib method. Proteinase activity was measured in terms of bovine serum albumin (BSA) degradation.

BSA medium (dextrose 2%, KH₂PO₄ 0.1%, MgSO₄ 0.05%, agar 2% mixed after cooling to 50°C with 1% BSA solution) was used.

Inoculating 10 μl aliquots of the yeast suspension (approximately 10⁸ yeast cells/ml) into the wells punched onto the surface of the medium were incubated at 37°C for two days, fixed with 20% trichloracetic acid and stained with 1.25% amido black. Decolorization was performed with 15% acetic acid. Opaqueness of the agar, corresponding to a zone of proteolysis around the wells that is not stained with amido black, indicated degradation of the protein. The proteinase index (Pz) was measured in terms of ratio of growth, to the diameter of proteolytic unstained zone.

Pz ≥1--no proteinase activity detected in the strain.

Pz <1 --positive for proteinase production.

Control: Candida albicans (ATCC10231).[20]

c) Phospholipase detection: Egg yolk medium consisting of 13.0 g SDA, 11.7 g NaCl, 0.111 g CaCl₂, and 10% sterile egg yolk was used (after centrifuging at 500 g for 10 min at room temperature, and 20 ml of the supernatant added to the sterilized medium). Inoculating 10 μl of aliquots of the yeast suspension (approximately 10⁸ yeast cells/ml) into the wells punched onto the surface of the egg yolk medium. The diameter of the precipitation zone around the well was measured after incubation at 37°C for 48 h.

Interpretation- Pz ≥1 will indicate no phospholipase activity.

Pz <1 will indicate positive phospholipase activity.

Positive control: C. albicans (ATCC 10231).[20]

ii. MIC determination by VITEK 2: Minimum inhibitory concentration (MIC) is the lowest drug concentration that prevents visible microorganism growth after overnight incubation. The VITEK 2 system incorporates the AST-YS01 card, which is designed for susceptibility testing of amphotericin B (AMB), fluconazole (FLC) and voriconazole (VRC). MIC of antifungals was determined as described in the company’s manual.[21]

Drug	MIC (Depends on species as per VITEK 2 manual: FDA/CLSI)
	Sensitive	Intermediate	Resistant
Amphotericin-B	≤1	2	≥4
Fluconazole	≤2	4-32	≥64
Voriconazole	≤0.125 (C. albicans & C. parapsilosis)	0.25-0.5	≥1
	≤0.5 (C. krusei)	1	≥2
	≤1 (Candida spp.)	2	≥4

Statistical analysis was performed using the Statistical Package fot the Social Sciences (SPSS) version 23. All categorical variables were presented in frequency and percentage. To measure the association between virulence factors and antifungal susceptibility, Chi-squared test or Fisher’s exact test was used as applicable. Level of significance was considered at P value less than 0.05.

Results

This study included 55 specimens of Candida species isolated from vulvovaginal specimens. The findings regarding antifungal sensitivity, virulence factors, and the association between the two are described in the following tables and accompanying text.

Table 1 depicts details of antifungal sensitivity of Candida species isolated from vulvovaginal specimen of 55 women. A majority of the Candida were sensitive to the three tested drugs (89% to amphotericin B, 76.4% to fluconazole, and 89.1% to voriconazole). Resistance to fluconazole was highest at 16.4% [Figure 1].

Table 1.

Descriptive statistics of antifungal sensitivity testing by VITEK 2 in patients with vulvovaginal candidiasis (n=55)

Antifungal	Mean±Std. Deviation
Amphotericin B	1.46±2.910
Fluconazole	5.64±10.655
Voriconazole	0.2567±79074

Figure 1.

Percentage distribution of antifungal sensitivity by VITEK 2

Table 2 presents the descriptive information of virulence markers of Candida species isolated from vulvovaginal specimen of the study group. Proteinase was present in 70.9% of the specimens, phospholipase in 83.6%, and biofilm was produced in 74.5% of specimens, and it was strong in 3.6% of specimens. Figures 2 and 3 depict these details.

Table 2.

Descriptive statistics of virulence factors in patients with vulvovaginal candidiasis (n=55)

Virulence factor	Mean±S.D
Proteinase (Pr)	0.77±0.189
Phospholipase (Pz)	0.76±0.155

Figure 2.

Percentage distribution of virulence factors

Figure 3.

Proportion of mild, moderate, and strong positive biofilm formation (n = 41)

Table 3 presents the results of the Chi-squared test that was performed to study the association between biofilm formation and antifungal sensitivity. Of the Candida specimens that produced a moderately strong biofilm, 87.5% exhibited sensitivity to amphotericin B while 75% were sensitive to fluconazole and 81.3% to voriconazole. None of the antifungal drugs studied exhibited any statistically significant relationship with biofilm formation.

Table 3.

Relationship between biofilm production and antifungal susceptibility

Antifungal	Biofilm				Total	Chi-Squared Test	P
	−	+	++	+++
Amphotericin B
Sensitive	12 (85.7)	7 (100)	28 (87.5)	2 (100)	49 (89.0)
Intermediate	1 (7.1)	0 (0.0)	2 (6.3)	0 (0)	3 (5.5)	1.350	0.969
Resistant	1 (7.1)	0 (0.0)	2 (6.3)	0 (0)	3 (5.5)
Fluconazole
Sensitive	10 (71.4)	6 (85.7)	24 (75)	2 (100)	42 (76.4)
Intermediate	3 (21.4)	0 (0.0)	1 (3.1)	0 (0)	4 (7.3)	7.215	0.301
Resistant	1 (7.1)	1 (14.3)	7 (21.9)	0 (0)	9 (16.4)
Voriconazole
Sensitive	14 (100)	7 (100)	26 (81.3)	2 (100)	49 (89.1)
Resistant	0 (0.0)	0 (0.0)	6 (18.8)	0 (0.0)	6 (10.9)	4.841	0.184

Table 4 presents the results of the Chi-squared test that was performed to study the relationship between proteinase synthesis and antifungal sensitivity. Of the Candida specimen that exhibited a positive proteinase test, 89.7% were sensitive to amphotericin B, 76.9% to fluconazole, and 92.3% to voriconazole. None of the antifungal drugs studied exhibited any statistically significant relationship with proteinase synthesis.

Table 4.

Relationship between proteinase production and antifungal susceptibility

Antifungal Sensitivity	Proteinase n (%)		Total	Chi-Squared Test	P
	Negative (≥1)	Positive (<1)
Amphotericin B
S	14 (87.5)	35 (89.7)	49 (89.1)
I	1 (6.3)	2 (5.1)	3 (5.5)	0.059	0.971
R	1 (6.3)	2 (5.1)	3 (5.5)
Fluconazole
S	12 (75)	30 (76.9)	42 (76.4)
I	1 (6.3)	3 (7.7)	4 (7.3)	0.116	0.943
R	3 (18.8)	6 (15.4)	9 (16.4)
Voriconazole
S	13 (81.3)	36 (92.3)	49 (89.1)	1.427	0.232
R	3 (18.8)	3 (7.7)	6 (10.9)

S=Sensitive, I=Intermediate, R=Resistant

Table 5 presents the results of the Chi-squared test that was performed to study the relationship between phospholipase synthesis and antifungal sensitivity. A majority (89.1%) of the Candida specimens that exhibited a positive phospholipase test were sensitive to amphotericin B, 76.1% to fluconazole, and 89.1% to voriconazole. None of the antifungal drugs studied exhibited any statistically significant relationship with phospholipase synthesis.

Table 5.

Relationship between phospholipase production and antifungal susceptibility

Antifungal Sensitivity	Phospholipase		Total	Chi-Squared Test	P
	Negative (≥1)	Positive (<1)
Amphotericin B
S	8 (88.9)	41 (89.1)	49 (89.1)
I	1 (11.1)	2 (4.3)	3 (5.5)	1.218	0.544
R	0 (0.0)	3 (6.5)	3 (5.5)
Fluconazole
S	7 (77.8)	35 (76.1)	42 (76.4)
I	0 (0.0)	4 (8.7)	4 (7.3)	1.011	0.603
R	2 (22.2)	7 (15.2)	9 (16.4)
Voriconazole
S	8 (88.9)	41 (89.1)	49 (89.1)
R	1 (11.1)	5 (10.9)	6 (10.9)	0.000	0.983

S=Sensitive, I=Intermediate, R=Resistant

Table 6 presents results of the Pearson correlation analysis. No significant correlation was found between antifungal sensitivity of Candida and virulence factors.

Table 6.

Correlation between virulence factors and minimum inhibitory concentrations of antifungals in Candida isolated from patients with vulvovaginitis

Antifungal	Proteinase r (P)	Phospholipase r (P)
Amphotericin B	−0.139 (0.519)	0.040 (0.854)
Fluconazole	0.310 (0.140)	−0.029 (0.893)
Voriconazole	0.317 (0.131)	0.028 (0.898)

r=Pearson correlation

Discussion

Candida species are considered opportunistic pathogens. For disease to occur, there must be a breakdown in host defence. Numerous virulence factors exist and may play different roles in different sites and stages of a given infection. In Candida spp., the transition from commensal to potential pathogen is determined by virulence attributes of the infecting species such as adherence to host tissue and medical devices, biofilm formation, and secretion of extracellular enzymes like phospholipases and proteinases.[22] In-hospital use of devices such as stents, shunts, prostheses, implants, endotracheal tubes, pacemakers, and catheters supports colonization and biofilm formation by Candida which may lead to nosocomial septicemia. Biofilm lifestyle of Candida leads to dramatically increased levels of resistance to the most commonly used antifungal agents. However, newer antifungal agents, such as the echinocandins and liposomal formulations of amphotericin B, show increased activity against Candida biofilms.[20–22] Aspartyl proteinases are secreted by pathogenic species of Candida in vivo during infection. Extracellular proteolytic activity of C. albicans is due to secreted aspartyl proteinases. It can break down a number of host substrates, including epithelial keratin, dermal collagen, albumin, hemoglobin, and immunoglobulin A (IgA), and enhances the ability of the organism to colonize and penetrate host tissues and to evade the host’s immune system. Following adhesion, secreted aspartyl proteinases production is exhibited primarily by pathogenic Candida species. Extracellular hydrolytic enzymes appear to play an important role in adherence, tissue penetration, invasion, and the destruction of host tissues.[20,22]

Recent emergence of antifungal resistance has caused major concern. Resistance to azole group of antifungal agents, which are safe and effective, is of concern because azoles like fluconazole are among the most commonly used antifungal agents for the treatment of candidiasis.[23,24] Antifungal resistance toward all three antifungal drug classes frequently occurs in clinical settings and may be related to various factors. A recent review reported that resistance to polyenes and echinocandins resulted in significant decrease in virulence in different Candida species. On the other hand, resistance to azole-type antifungals was also shown to result in increased virulence depending on the species and isolates.[17] Previous studies have postulated that accurate identification of putative virulence factors in Candida spp. is critical for predicting the response of antifungal drugs and detecting the emergence of strains with greater resistance.[25,26] These findings underline the importance of studies aiming to dissect the connections of virulence and resistance in Candida species.

In India, few studies have discussed the prevalence and antifungal susceptibility of Candida and its virulence markers,[21,27] and the lack of research on association of the two is absolute. The current study aimed to expand the knowledge concerning antifungal sensitivity and virulence factors of pathogenic Candida isolated from vulvovaginal specimen of pregnant and non-pregnant women in the reproductive age group, presenting to a tertiary care hospital in central India. The study included 55 Candida isolates. These isolates were identified to species level using fermentation and assimilation tests which demonstrated similar results, denoting that they represent valuable methods for identification of Candida species, and use of VITEK 2 for MIC values. The findings were presented in a previous publication.[27] The focus of the current paper is on the relationship of antifungal sensitivity of isolated Candida species with the presence of virulence factors, namely, biofilm formation, proteinase synthesis and phospholipase synthesis.

Concerning the azole susceptibility profile of our study, findings revealed that a majority of the Candida specimens were sensitive to the three tested antifungal drugs: amphotericin B, fluconazole, and voriconazole. The highest resistance was towards fluconazole. This is an important observation as fluconazole has remained the drug of choice for treating candidemia for over several years now because of its efficacy against clinical Candida infection.[24] However, there have been reports of increased resistance and treatment failure associated with the use of fluconazole.[23,25] When we reviewed the literature to compare our findings with previous research, we found varying results. Some studies from Egypt reported no resistance against fluconazole,[28–30] whereas another study reported that 25% of isolates were resistant to fluconazole, 12% were resistant to voriconazole, and one isolate (6%) was resistant to amphotericin B.[29] China, Taiwan, and Kenya reported low resistance.[31–33] A large study published in 2017 by Shawky et al.[30] indicated high azole resistance percentages at 24.6% and 17.5% to fluconazole and voriconazole, respectively. Discrepant results regarding azole susceptibility have also been reported from the Middle East.[34]

These differences are explained by the difference in sample sizes, methods used for determination of antifungal susceptibility, and breakpoints used for interpretations, in addition to different uses of azoles in prophylaxis therapy among countries or institutions.[35]

With regard to the descriptive information of virulence factors of Candida species isolated from the study group, proteinase was produced by 70.9% of the specimens, phospholipase by 83.6%, and biofilm was produced in 74.5% of the specimens. Our study showed a high proteinase and phospholipase activity. There are discordant results in literature regarding these virulence factors.[36–41] For example, lower percentages of isolates show proteinase[9,37] and phospholipase enzymatic activity in some studies[37,39] whereas high phospholipase activity in others.[38,40] No activity at all was also reported in few studies.[36,41] The inconsistencies observed regarding proteinase and phospholipase activity could be attributed to several factors, such as method of media preparation, and difference in incubation temperature and duration. Discrepancies could also be attributed to the plate method used, as it may not detect the activity in low-phospholipase-producing strains.

Biofilm-associated infections are difficult to treat, representing a source of reinfections.[42,43] Biofilm production represents the most important virulence factor of Candida species.[44] The mortality rates in patients infected by biofilm-forming isolates are greater than those infected by non-biofilm-forming isolates.[45] Regarding biofilm formation, results similar to our study have been reported by a few studies.[9,46] However, variations among results concerning biofilm formation could be due to physiological differences between strains according to the origin of the isolates.[47]

We discuss below the relationship of antifungal sensitivity of the specimen with each of the virulence factors. We performed the Chi-squared test to study the relationship between biofilm formation and antifungal sensitivity. Although we found fluconazole resistance in 25% of specimens that exhibited a moderate strong biofilm formation, this association was not statistically significant. In fact, none of the antifungal drugs studied exhibited any statistically significant relationship with biofilm formation. This finding is in agreement with previous reports.[9,48] Nevertheless, contradictions are reported as positive correlation between biofilm production and resistance to fluconazole by other researchers.[26]

To test our hypothesis that increased virulence can affect drug sensitivity, we analyzed statistically the relationship between proteinase synthesis and antifungal sensitivity. Of the Candida specimens that exhibited a positive proteinase test, 89.7% were sensitive to amphotericin B, 76.9% were sensitive to fluconazole, and 92.3% were sensitive to voriconazole. None of the antifungal drugs studied exhibited any statistically significant relationship with proteinase synthesis. This established that proteinase activity, though marking the virulence of Candida spp., did not have a significant association with drug resistance. Phospholipase production is another virulence factor that we hypothesized could affect the antifungal sensitivity; hence we employed statistical tests to study this association. We found that the majority of Candida specimens that exhibited a positive phospholipase test were sensitive to amphotericin B, fluconazole, and voriconazole. None of the antifungal drugs studied exhibited any statistically significant relationship with phospholipase synthesis. Contradictory findings have been reported in literature.[17] These differences can be attributed to differences in the characteristics of different strains according to the origin of the isolates and the population differences.[47]

Strengths and limitations

To the best of our knowledge, this is the first study in the region to explore the association of antifungal sensitivity with virulence factors of Candida. However, there are some limitations. The sample size is small, and the study was carried out in a single center; hence, the results cannot be generalized. Nevertheless, the study provides valuable information for microbiologists and clinicians to improve patient care. It is recommended that larger, multi-center studies focusing on species and organ-specific Candida infection should be undertaken for a better understanding and generalizability of findings.

Conclusion

There is a growing body of evidence that supports the need for improved management of Candida species, due to its increasing antifungal susceptibility.[49,50] In this direction, our study findings have important implications for primary care physicians and mycologists in the region. The observed antifungal resistance to the two azoles, especially toward voriconazole, is worrisome. However, it is relieving that none of the virulence factors have a significant association with antifungal sensitivity of the Candida species affecting the cases. This finding is comforting for the clinician as well as the microbiologists as it rules out the effect of virulence of the pathogen on drug response.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.