Susceptibility to caspofungin is regulated by temperature and is dependent on calcineurin in Candida albicans

ABSTRACT Candida albicans is a prevalent opportunistic human fungal pathogen. Echinocandins are first-line drugs for the treatment of invasive candidiasis. Molecular mechanisms of resistance to echinocandins have been extensively studied; however, the effect of physiological factors on antifungal efficacy of echinocandins is unexplored. Here, we found temperature-modulated susceptibility to caspofungin in medium-independent and strain background-independent manner in C. albicans. Deletions of protein kinase C (PKC) pathway and calcineurin-Crz1 signaling pathway genes conferred hypersensitivity to caspofungin, but only deletion or pharmacological inhibition of calcineurin subunits inverted the temperature effect on susceptibility to caspofungin. Furthermore, the enhanced growth at lower temperature was not due to altered expression of some established genes such as FKS, CHS, or CHT genes; PKC or calcineurin pathway genes; or HSP90. The expressions of other heat shock protein genes, including HSP12 and HSP70, were higher at lower temperature. We posit that the temperature-modulated susceptibility to caspofungin is a non-canonical mechanism that is dependent on heat shock proteins and calcineurin. IMPORTANCE Echinocandins are the newest antifungal drugs and are first-line treatment option for life-threatening systemic infections. Due to lack of consensus regarding what temperature should be used when evaluating susceptibility of yeasts to echinocandins, typically either 30°C, 35°C, or 37°C is used. However, the impact of temperature on antifungal efficacy of echinocandins is unexplored. In the current study, we demonstrated that Candida albicans laboratory strain SC5314 was more susceptible to caspofungin at 37°C than at 30°C. We also found that calcineurin was required for temperature-modulated caspofungin susceptibility. Surprisingly, the altered caspofungin susceptibility was not due to differential expression of some canonical genes such as FKS, CHS, or CHT genes. The molecular mechanism of temperature-modulated caspofungin susceptibility is undetermined and deserves further investigations.

list of health-threatening fungi.C. albicans is among the four members of the most dangerous "critical group" (4).
Currently, only four classes of antifungal drugs are available: azoles, polyenes, 5-flucytosine, and echinocandins.Echinocandins are the most optimal first-line antifungal agents for the treatment of invasive candidiasis (5).Three echinocandin drugs are approved by the United States Food and Drug Administration to treat candidemia: caspofungin (CSP), anidulafungin (ANF), and micafungin (MCF).Echinocan dins are fungicidal against Candida spp and fungistatic against Aspergillus spp but are inactive against Cryptococcus spp (6).Echinocandins act by non-competitively binding to β-(1,3)-glucan synthase (7).Inhibition of the major fungal cell wall component β- (1,3)glucan biosynthesis leads to growth inhibition or death owing to imbalance in osmotic pressure (8).
The incidence of CSP resistance is rising in recent years (9)(10)(11).Resistance to CSP is usually due to point mutations in FKS genes, which encode subunits of the β-(1,3)-glucan synthase (12).Mutation usually occurs in two highly conserved hotspot regions of FKS genes encompassing amino acids 641-649 and amino acids 1357-1364 (13).The amino acids of FKS mutation decrease the sensitivity of glucan synthase to echinocandins and cause cross-resistance to various echinocandins (9,14,15).About 80% mutation in FKS1/GSC1 of C. albicans occur in Ser645 and Phe641, which brings the most prominent resistance phenotype (16,17).The roles of FKS2/GSL1 and FKS3/GSL2 in resistance to CSP in C. albicans are largely unknown; however, deletions of FKS2 and FKS3 can upregulate expression of FKS1, increase cell wall glucan content, and decrease susceptibility to echinocandins (18).In glabrata, mutations occur more frequently in FKS2 (19).In auris, S639F mutation of FKS1 is associated with CSP resistance in clinical isolates (20).
In addition to FKS mutations, cell wall salvage mechanism also contributes to CSP resistance.Fungal cell wall consists of glucan, chitin, and glycoproteins.Inhibition of glucan biosynthesis usually results in compensatory increase of chitin synthesis (21,22).Both in vitro and in vivo, elevated chitin confers CSP resistance in C. albicans (23,24).Moreover, C. albicans clinical isolates containing FKS1 mutation generally have higher chitin levels (23).Decreased digestion of chitin, due to loss-of-function mutation in chitinase genes CHT2 and CHT3 (25), and decreased copy number of CHT2 also confer CSP resistance in C. albicans (26).
Compensatory increase of chitin in response to CSP treatment is dependent on the regulatory circuit composed of PKC pathway, calcineurin pathway, high-osmolar ity glycerol pathway, and heat shock protein 90 (Hsp90) (22,27).Deletion of PKC pathway genes, including PKC1, BCK1, MKK2, and MKC1, or genes encoding down stream transcription factors, including SWI4, SWI6 and RLM1, causes hypersensitivity to echinocandins (28,29).Pharmacological inhibition of calcineurin is synergistic with echinocandins against clinical isolates harboring FKS1 mutations (30).Genetic deletion of CMP1 and CNB1, which encodes the catalytic subunit and regulatory subunit of calcineurin, respectively, or CRZ1, which encodes the major downstream transcription factor of the calcineurin pathway, can enhance sensitivity of C. albicans to echinocan dins (29,30).Calcineurin is the client protein of Hsp90, a highly conserved molecular chaperone.Hsp90 plays a pivotal role in activation and stabilization of calcineurin (30).
Although genes and regulatory networks required for CSP resistance have been studied extensively, physiological factors affecting antifungal efficacy of CSP are largely unexplored.Recent studies indicate that temperature affects antifungal potency of azoles against C. albicans (31, 32); however, the impact of different temperatures on the antifungal effect of CSP is still unknown.There is no consensus among researchers regarding what temperature should be used in yeast studies.Usually either 30°C or 37°C is used.In the current study, we found that C. albicans was less susceptible to CSP at 30°C than at 37°C.The temperature effect was independent on medium composition, or genetic backgrounds of test strains, and was independent on the PKC pathway, albeit deletion of PKC pathway genes that caused hypersensitivity to CSP.Pharmacological inhibition or genetic knockdown of calcineurin subunits inverted temperature effect on CSP susceptibility; however, the transcription factor of the calcineurin pathway, Crz1, was not involved.Comparison of transcriptomes of cells grown at 30°C and 37°C indicated that several genes encoding chaperone proteins were more abundant at 30°C.We posit that heat shock proteins and the client protein calcineurin are required for temperaturemodulated CSP susceptibility in C. albicans.

Spot assay
Cells were suspended in distilled water and adjusted to 1 × 10 7 cells/mL.A total of 3 µL of 10-fold serial dilutions were spotted on YPD, SD, or SDC plates with or without drugs (control) at 30°C or 37°C and photographed after 2 days.Medium and temperature were indicated in the figures.

Deletions
Gene deletions were performed as described previously (31).Primers are the same as described previously (31).Plasmid pJK863 (33) was used as template for amplifying the NAT1 flipper gene deletion cassette.Approximately 500 bp regions flanking the CDS of the target gene to be deleted were amplified using the genomic DNA of YJB-T490 as the template.The upstream region of the gene was fused by PCR to the 5′ region of the cassette, and the downstream region of the gene was fused to the 3′ region of the cassette.The upstream and downstream fusion products for each gene were then simultaneously transformed in C. albicans by following the lithium acetate method (34).Transformants were selected on YPD plates supplemented with 400 µg/mL NAT.Diagnostic PCR using primers that annealed outside the flanking homologous regions of the gene was performed to confirm the replacement of the gene with the NAT1 flipper cassette.The NAT1 flipper was evicted by streaking the clones on YNB-BSA plates.

Growth curve
Approximately 1 × 10 3 cells/mL of each strain were suspended in YPD broth and 150 µL was transferred to a 96-well plate.Optical density at 595 nm (OD 595 ) was measured every 15 min for 24 h at 30°C using a Tecan plate reader (Infinite F200 PRO; Tecan, Switzerland).

RNA-Seq
RNA-seq was performed as described previously (29).SC5314 was inoculated to a starting OD 600 of 0.2 in 50 mL of YPD broth.The culture was incubated in a shaker at 30°C and 37°C until the OD 600 reached 1.0.Cultures were collected by centrifugation (comparison between 30°C and 37°C) or divided into two batches: one batch was supplemented with 100 ng/mL CSP, and the other batch was sup plemented with an equal volume of DMSO.Three hours later, the cultures were collected by centrifugation, washed, and flash frozen in liquid nitrogen.The total RNA was extracted for six independent samples, corresponding to two conditions and three biological replicates.Total RNA extraction and purification, library con struction, and sequencing were performed as described in Yang et al. (35).Raw sequence files (.fastq files) underwent quality control analysis using the FastQC tool (http://www.bioinformatics.babraham.ac.uk/projects/fastqc). Reads were mapped to the C. albicans SC5314 reference genome (http://www.candidagenome.org/download/sequence/C_albicans_SC5314/Assembly22/current/).Differential gene expression profiling was carried out using DESeq2 (36) with standard parameters.Genes with false discovery rate-adjusted P-value (<0.05) and expression fold changes of more than 1.5 or less than −1.5 were considered differentially expressed.

Temperature modulates echinocandins susceptibility
Susceptibility of the C. albicans reference strain SC5314 to CSP and MCF was measured at two different temperatures using different media.On YPD, the growth was not inhibited by 100 ng/mL CSP (Fig. 1, top panel) or 25 ng/mL MCF (Fig. 1, bottom panel) at 30°C, but the growth was completely inhibited at these drug concentrations at 37°C.On SD and SDC, the inhibition of growth by 200 ng/mL of CSP (Fig. 1, top panel) or 25 ng/mL MCF (Fig. 1, bottom panel) was more obvious at 37°C than at 30°C.Thus, independent on medium composition, SC5314 is generally less susceptible to CSP and MCF at 30°C than at 37°C.
Comparing between 30°C and 37°C, on YPD, SD and SDC, both mkk2 Δ/Δ and mkc1 Δ/Δ strains were less susceptible to CSP at 30°C than at 37°C (Fig. 2).Therefore, deletion of MKK2 or MKC1 caused hypersensitivity to CSP at both 30°C and 37°C; however, in deletion strains, lower temperature (30°C) still enables better adaptation to CSP than higher temperature (37°C).

Temperature effect on susceptibility to CSP is dependent on calcineurin
The calcineurin-Crz1 signaling pathway is conserved across multiple pathogenic fungi.Calcineurin is a major player in calcium 2+ -dependent signal transduction pathways of eukaryotes.The transcription factor Crz1 is the major effector of calcineurin (37).Calcineurin consists of a catalytic subunit and a regulatory unit, which is encoded by CMP1 and CNB1, respectively, in C. albicans genome (37).
In this study, the role of calcineurin-Crz1 signaling pathway in temperature-modula ted CSP susceptibility was evaluated in two genetic backgrounds: SC5314 and YJB-T490.Two methods were employed: genetic deletions of CMP1, CNB1, and CRZ1 and pharma cological inhibition of calcineurin.
In SC5314 background, at 30°C, the wild type was not inhibited by 90 ng/mL CSP, while cmp1Δ/Δ and cnb1Δ/Δ strains were completely inhibited by 30 ng/mL CSP.However, crz1Δ/Δ strain was not inhibited by 90 ng/mL CSP.At 37°C, the wild type was not inhibited by 50 ng/mL CSP, while cmp1Δ/Δ and cnb1Δ/Δ strains were obviously inhibited by 50 ng/mL CSP, but the crz1Δ/Δ strain was also not inhibited by 50 ng/mL.Comparing between 30°C and 37°C, interestingly, both cmp1Δ/Δ and cnb1Δ/Δ strains were less susceptible at 37°C than at 30°C; however, similar to wild type, the crz1Δ/Δ strain was still more susceptible at 37°C than at 30°C (Fig. 3A).In YJB-T490 background, at 30°C, the wild type was not inhibited by 70 ng/mL CSP, while the cmp1Δ/Δ and cnb1Δ/Δ strains were completely inhibited by 30 ng/mL and 50 ng/mL CSP, respectively.At 37°C, the wild type was not inhibited by 50 ng/mL CSP and was completely inhibited by 70 ng/mL CSP, while the cmp1Δ/Δ and cnb1Δ/Δ strains were all completely inhibited by 50 ng/mL CSP.However, crz1Δ/Δ strain had similar extent of CSP susceptibility to wild type at both 30°C and 37°C.Compared between 30°C and 37°C, both cmp1Δ/Δ and cnb1Δ/Δ strains were less susceptible to CSP at 37°C than at 30°C, while the crz1Δ/Δ strain was still more susceptible at 37°C than at 30°C (Fig. 3B).We also measured growth curves of deletion strains and the wild types in YPD broth without CSP.We found that all strains had similar growth curves (Fig. S1); thus, altered CSP susceptibility in the deletion strains was not due to change of growth caused by deletions.
Requirement of calcineurin for temperature-modulated CSP susceptibility was also evaluated by using calcineurin inhibitor cyclosporin A (CsA).In the YPD medium supplemented with 0.5 µg/mL CsA, both SC5314 and YJB-T490 were less susceptible to CSP at 37°C than at 30°C (Fig. 3C).As a control, CsA at 0.5 µg/mL was not inhibitory to the strains (Fig. S2).
Taken together, in both SC5314 and YJB-T490 backgrounds, calcineurin is required for temperature-modulated CSP susceptibility.Genetic deletion or pharmacological inhibition of calcineurin inverted extent of CSP susceptibility at 30°C and 37°C.

Canonical genes associated with CSP susceptibility are not involved in temperature-modulated CSP susceptibility
We asked why susceptibility to CSP was lower at 30°C than at 37°C.Transcriptome analysis indicated that some genes were regulated by CSP similarly 30°C and 37°C.For example, the FKS gene GSC1; the CHS genes CHS2, CHS3, CHS4, and CHS7; and the calcineurin-Crz1 signal pathway genes CMP1 and CRZ1 were upregulated by CSP at both temperatures.The CHT gene CHT3 and the HSP genes HSP21, HSP70, HSP104, and HSP12 were downregulated by CSP at both temperatures.Some genes were differentially regulated by CSP at 30°C and 37°C.For example, the PKC pathway genes BCK1 and RLM1 were upregulated by CSP at 30°C but not at 37°C, while the HSP genes HSP31 and HSP104 were downregulated by CSP at 30°C but not at 37°C (Table S2).
We compared transcriptomes of log phase cells of SC5314 incubated at 30°C and at 37°C.Genes with ratios of transcripts at 30°C vs 37°C higher than 1.5 and lower than 0.67 were considered higher and lower expressed, respectively.We found that cells did not have altered expression of FKS genes, including GSC1, GSL1, and GSL2, or of CHS genes, including CHS1, CHS2, CHS3, CHS4, CHS5, CHS6, CHS7, and CHS8.Among the four CHT genes, CHT2 and CHT3 had higher expressions (Fig. 4, top panel).Among the calcineurin-Crz1 signal pathway genes, CMP1 and CNB1 were not differentially expressed.Expression of CRZ1 was not detected.Among the PKC pathway genes, RLM1 was downregulated.The other genes were not differentially expressed, including PKC1, BCK1, MKK2, MKC1, SWI4, and SWI6 (Fig. 4, middle panel).Among the heat shock protein genes, HSP12, HSP70, HSP78, and HSP104 had higher expressions at 30°C than at 37°C, while the other heat shock protein genes were not differentially expressed, including HSP21, HSP30, HSP31, HSP60, and HSP90 (Fig. 4, bottom panel).
We also compared proteomes of log phase cells of SC5314 incubated at 30°C and at 37°C.Unexpectedly, cells did not exhibit significantly different abundance of proteins encoded by the abovementioned FKS, CHS, CHT, calcineurin-Crz1 signal pathway, or PKC pathway genes.Among the Hsp proteins, Hsp106, Hsp60, Hsp90, Hsp21, and Hsp78 were significantly more abundant in cells grown at 37°C than at 30°C (Table S3).
Taken together, lower CSP susceptibility at 30°C than at 37°C is not due to elevated expressions of FKS or CHS genes or to decreased expression of CHT genes.It is also not due to activation of PKC and calcineurin-Crz1 signaling pathways or higher expression of HSP90.It is probably due to altered expression and abundance of other heat shock protein genes such as HSP12, HSP70, HSP78, and HSP104.

DISCUSSION
Echinocandins are the newest antifungal drugs licensed for clinical use.Molecular mechanism of development of resistance to echinocandins has been extensively studied.However, little is known about the impact physiological factors on antifungal efficacy of echinocandins.The normal human body temperature ranges from 36.5°C to 37.5°C.Studies evaluating antifungal activities of echinocandins in Candida spp typically use either 30°C (38) or 37°C (39,40).To the best of our knowledge, our study is the first to investigate the effect of different temperature on the extent of CSP resistance in C. albicans.
Consistent with previous studies, deletions of PKC pathway genes MKK2 and MKC1 and calcineurin-Crz1 signaling pathway genes CMP1 and CNB1 caused hypersensitivity to CSP (29,30,41,42).Furthermore, we demonstrated that PKC pathway was not required for temperature-modulated CSP resistance.Like the wild type, both mkk2Δ/Δ and mkc1Δ/Δ strains were still more resistant to CSP at 30°C than at 37°C.We further demonstrated that both pharmacological inhibition and genetic deletion of calcineurin subunits abolished the temperature-modulated CSP resistance; however, strains with deletion of CRZ1 were still more resistant to CSP at 30°C than at 37°C, indicating that CSP resistance modulated by temperature is dependent on calcineurin but is independent on Crz1.
Unexpectedly, cells grown at 30°C and 37°C do not have altered expression of some canonical genes associated with CSP resistance, such as FKS, CHS, or CHT genes, or the PKC and calcineurin pathway genes, or HSP90.Several other genes encoding heat shock proteins were highly expressed (1.7-6.4 fold), including HSP12, HSP70, HSP78, and HSP104.Heat shock proteins exist in most organisms and have multiple broad functions.In C. albicans, there are nine kinds of heat shock proteins with varying molecular sizes: HSP12, HSP21, HSP30, HSP31, HSP60, HSP70, HSP78, HSP90, and HSP104.Calcineurin is a client protein of Hsp90.Requirement of calcineurin in resistance to azoles and echino candins is dependent on Hsp90 (30,43).Hsp70 is involved in the transfer of client proteins to Hsp90 for their subsequent activation via the Hsp90 chaperone cycle (44).Hsp12 is a small heat shock protein.Hsp12 usually binds denatured proteins with high affinity until Hsp70 reactivates them (45).In C. albicans, overexpression of HSP12 causes hypersensitivity to azoles (46).In Cryptococcus neoformans, deletion of HSP12 confers hypersensitivity to amphotericin B but not to azoles (47).In Saccharomyces cerevisiae, HSP12 is involved in response to multiple stresses including antifungal drugs (48), and deletion of HSP12 results in reduced plasticity and flexibility of cell wall (49).HSP12 and HSP70 are promising potential targets for antifungal drug development (50).The function of Hsp104 is similar to Hsp70 (51).Hsp78 is a mitochondrial heat shock protein.Hsp78 is implicated in the proteolysis required for the efficient degradation of substrate proteins in mitochondria (52).We posit differential expressions of Hsp12 and Hsp70 at 30°C and 37°C and potentiated calcineurin-dependent CSP resistance.
In conclusion, this study demonstrated a temperature-modulated, non-canonical mechanism of CSP susceptibility in C. albicans.Coordinated actions of Hsp12 and Hsp70 probably underlay the essentiality of calcineurin in this novel mechanism.

FIG 2
FIG 2 Role of PKC pathway in temperature-modulated caspofungin resistance.SC5314-derived strains with deletions of MKK2 (top panel) and MKC1 (bottom panel) were evaluated with spot assay.The media, temperatures, and drug concentrations are indicated in the figure.For each strain, the cell densities spotted on the plates were 1 × 10 7 , 1 × 10 6 , 1 × 10 5 , and 1 × 10 4 cells/mL.The plates were incubated for 48 h then photographed.

FIG 3
FIG 3 Role of calcineurin-Crz1 pathway in temperature-modulated caspofungin resistance.In SC5314 (top panel) and YJB-T490 (middle panel), CMP1, CNB1, and CRZ1 were deleted.Wild-type and deletion strains were tested with spot assay on YPD-agar medium supplemented with caspofungin (CSP).The wild-type strains were also tested with spot assay using YPD-agar supplemented with calcineurin inhibitor cyclosporin A (CsA, 0.5 µg/mL) (bottom panel).The temperatures and drug concentrations are indicated in the figure.The plates were incubated for 48 h then photographed.