Molecular Epidemiology, Antifungal Susceptibility, and Virulence Evaluation of Candida Isolates Causing Invasive Infection in a Tertiary Care Teaching Hospital

Abstract

Background: The incidence of invasive candidiasis is increasing worldwide. However, the epidemiology, antifungal susceptibility, and virulence of Candida spp. in most hospitals remain unclear. This study aimed to evaluate invasive candidiasis in a tertiary care hospital in Nanchang City, China.

Methods: MALDI-TOF MS and 18S rDNA ITS sequencing were used to identify Candida strains. Randomly amplified polymorphic DNA analysis was used for molecular typing; biofilm production, caseinase, and hemolysin activities were used to evaluate virulence. The Sensititre™ YeastOne YO10 panel was used to examine antifungal susceptibility. Mutations in ERG11 and the hotspot regions of FKS1 of drug-resistant strains were sequenced to evaluate the possible mechanisms of antifungal resistance.

Results: We obtained 110 Candida strains, which included 40 Candida albicans (36.36%), 37 C. parapsilosis (33.64%), 21 C. tropicalis (19.09%), 9 C. glabrata (8.18%), 2 C. rugose (1.82%), and 1 C. haemulonii (0.91%) isolates. At a limiting point of 0.80, C. albicans isolates could be grouped into five clusters, C. parapsilosis and C. tropicalis isolates into seven clusters, and C. glabrata isolates into only one cluster comprising six strains by RAPD typing. Antifungal susceptibility testing revealed that the isolates showed the greatest overall resistance against fluconazole (6.36%), followed by voriconazole (4.55%). All C. albicans and C. parapsilosis isolates exhibited 100% susceptibility to echinocandins (i.e., anidulafungin, caspofungin, and micafungin), whereas one C. glabrata strain was resistant to echinocandins. The most common amino acid substitutions noted in our study was 132aa (Y132H, Y132F) in the azole-resistant strains. No missense mutation was identified in the hotpot regions of FKS1. Comparison of the selected virulence factors detectable in a laboratory environment, such as biofilm, caseinase, and hemolysin production, revealed that most Candida isolates were caseinase and hemolysin producers with a strong activity (Pz < 0.69). Furthermore, C. parapsilosis had greater total biofilm biomass (average Abs₆₂₀ = 0.712) than C. albicans (average Abs₆₂₀ = 0.214, p < 0.01) or C. tropicalis (average Abs₆₂₀ = 0.450, p < 0.05), although all C. glabrata strains were either low- or no-biofilm producers. The virulence level of the isolates from different specimen sources or clusters showed no obvious correlation. Interesting, 75% of the C. albicans from cluster F demonstrated azole resistance, whereas two azole-resistant C. tropicalis strains belonged to the cluster Y.

Conclusion: This study provides vital information regarding the epidemiology, pathogenicity, and antifungal susceptibility of Candida spp. in patients admitted to Nanchang City Hospital.

Keywords: Candida species; RAPD; antifungal susceptibility; biofilm; caseinase; hemolysin.

Figure 1

Species distribution of 110 clinical isolates. The 110 Candida strains analyzed included 40 C. albicans (36.36%), 37 C. parapsilosis (33.64%), 21 C. tropicalis (19.09%), 9 C. glabrata (8.18%), 2 C. rugose (1.82%), and 1 C. haemulonii (0.91%) isolate.

Figure 2

Dendrogram presenting the genetic relatedness of 107 Candida spp. based on the random amplified polymorphic DNA (RAPD) data. The virulence factors (i.e., caseinase, hemolysin, and biofilm), drug resistance, and specimen sources are indicated in the figure. The vertical line divides the strains based on the level of genetic similarity into related and unrelated strains. The red boxes indicate azole resistance (fluconazole and voriconazole), whereas the blue boxes indicate azole-susceptible (S) dose dependence (SDD). Red ellipses represent resistance to echinocandins, while the blue ellipses represent intermediate resistance to micafungin. (A) Dendrogram of 40 C. albicans isolates. (B) Dendrogram of 37 C. parapsilosis isolates. (C) Dendrogram of 21 C. tropicalis isolates. (D) Dendrogram of 9 C. glabrata isolates.

Figure 3

Virulence level (biofilm biomass, caseinase activity, and hemolysin activity) of the 107 Candida species isolates from different specimen sources. Y-axis (far left panel): biofilm biomass (positively correlated with virulence); Y-axis (far right panel): the Pz values of caseinase and hemolysin activities (negatively correlated with virulence). No significant differences were noted in these factors. (A) The virulence level of 40 C. albicans isolates from different specimen sources. (B) The virulence level of 37 C. parapsilosis isolates from different specimen sources. (C) The virulence level of 21 C. tropicalis isolates from different specimen sources. (D) The virulence level of 9 C. glabrata isolates from different specimen sources.

Figure 4

Comparison of the virulence levels (biofilm biomass, casein activity and hemolysin activity) of four Candida species. * indicates statistically significant (p < 0.05).