Decreased echinocandin susceptibility in Candida parapsilosis causing candidemia and emergence of a pan-echinocandin resistant case in China

Abstract

Candida parapsilosis is becoming a predominant non-albicans cause of invasive candidiasis (IC). Echinocandins are the preferred choice for IC treatment and prophylaxis. Resistance to echinocandins in C. parapsilosis has emerged in several countries, but little is known about the susceptibility profile in China or about mechanisms of resistance. Here, we investigated the echinocandin susceptibilities of 2523 C. parapsilosis isolates collected from China and further explored the resistance mechanism among echinocandin-resistant isolates. Anidulafungin exhibited the highest MICs (MIC_50/90, 1 and 2 µg/mL; GM, 0.948 µg/mL), while caspofungin showed better activity (0.5 and 1 µg/mL; 0.498 µg/mL). Significantly higher echinocandin MICs were observed among blood-derived isolates compared to others, especially for caspofungin (GM, 1.348 µg/mL vs 0.478 µg/mL). Isolates from ICU and surgical wards also showed higher MICs. Twenty isolates showed intermediate phenotypes for at least one echinocandin. One was resistant to all three echinocandins, fluconazole and voriconazole, which caused breakthrough IC during long-term exposure to micafungin. WGS revealed this isolate carried a mutation S656P in hotspot1 region of Fks1. Bioinformatics analyses suggested that this mutation might lead to an altered protein conformation. CRISPR Cas9-mediated introduction of this mutation into a susceptible reference C. parapsilosis strain increased MICs of all echinocandins 64-fold, with similar results found in the subspecies, C. orthopsilosis and C. metapsilosis. This is the first report of a multi-azole resistant and pan-echinocandin resistant C. parapsilosis isolate, and the identification of a FKS1^S656P conferring pan-echinocandin resistance. Our study underscores the necessity of rigorous management of antifungal use and of monitoring for antifungal susceptibility.

Keywords: Candida parapsilosis; FKS1; S656P; breakthrough candidemia; echinocandin resistance.

Figure 1.

Profile of C. parapsilosis susceptibility to echinocandins. (A) Distribution of susceptibility to three echinocandins of 2523 clinical isolates from China collected by CHIF-NET 2015–2019 programme. (B and C) Susceptibility differences of isolates regarding different samples (B), and hospital departments (C). Geometric mean of the MIC (GM) values are shown in the boxes; differences between each two groups were analyzed, while only significant differences compared to the “others” group are shown in the figure. AND, anidulafungin; CAS, caspofungin; MF, micafungin. *, P < 0.05; ****, P < 0.0001.

Figure 2.

Strain characteristics of the pan-echinocandin resistant C. parapsilosis clinical isolate. (A) In vitro susceptibility to nine common antifungal drugs. The dashed red line indicates the breakpoints for defining drug resistance or cut-off values for non-wild type. (B) Transmembrane helix predictions for Fks1 of C. parapsilosis. The location of the amino acid 656 is labelled.

Figure 3.

Structural and functional effect of the S656P substitution in Fks1. (A) The predicted structural model for the variants of the Fks1 protein (amino acids 400∼900), the S656P substitution would result in the helix-to-coil transition, disrupting conformation of this ɑ-helix. (B) Echinocandin MICs for susceptible C. parapsilosis strain ATCC 22019 and its mutant harbouring Fks1 S656P, and the pan-echinocandin resistant clinical isolate TJ1197 and its mutant with WT of Fks1. (C) Changes in expression of the FKS1 and CHS3 genes in response to micafungin at the sub-MICs. (D) Echinocandin MICs for susceptible C. orthopsilosis ATCC 96139, C. metapsilosis ATCC 96143, and their mutants carrying homologous modifications, S649P and S656P, respectively. Dashed red lines indicate the breakpoints for defining drug resistance or cut-off values for non-wild type. AND, anidulafungin; CAS, caspofungin; MF, micafungin. **, P < 0.01; ***, P < 0.001.