Overlooked Candida glabrata petites are echinocandin tolerant, induce host inflammatory responses, and display poor in vivo fitness

Abstract

Candida glabrata is a major fungal pathogen, which is able to lose mitochondria and form small and slow-growing colonies, called "petite." This attenuated growth rate has created controversies and questioned the clinical importance of petiteness. Herein, we have employed multiple omics technologies and in vivo mouse models to critically assess the clinical importance of petite phenotype. Our WGS identifies multiple genes potentially underpinning petite phenotype. Interestingly, petite C. glabrata cells engulfed by macrophages are dormant and, therefore, are not killed by the frontline antifungal drugs. Interestingly, macrophages infected with petite cells mount distinct transcriptomic responses. Consistent with our ex vivo observations, mitochondrial-proficient parental strains outcompete petites during systemic and gut colonization. Retrospective examination of C. glabrata isolates identified petite prevalence a rare entity, which can significantly vary from country to country. Collectively, our study overcomes the existing controversies and provides novel insights regarding the clinical relevance of petite C. glabrata isolates.

Keywords: Candida glabrata; drug resistance; petite; toelrance.

Fig 1

Characteristics of petite C. glabrata isolates. All petite isolates included in this study were fluconazole resistant (A) and did not carry PDR1 gain-of-function mutations, as confirmed by PDR1 sequencing. Petite isolates have a higher basal level of PDR1 and efflux pumps under its control (CDR1, CDR2, and SNQ2) once measured by real-time PCR (three biological replicates, ***<0.01, ****<0.001, two-tailed t-test) (B). Petites had a lower ATP level (five biological replicates, ****<0.001, two-tailed t-test) (C) and mitochondrial membrane potential (three biological replicates, ***<0.01 and **=0.01, two-tailed t-test) (D) than their respiratory proficient counterparts. ATP, adenosine triphosphate.

Fig 2

MIP1 sequencing of diverse clinical isolates revealed that polymorphisms occurring in isolates better reflect the sequence type rather than petite phenotype (A). The interaction of petite and their respective parental isolates with THP1 macrophages. Macrophages infected with respective isolates and intracellular replication were measured 3, 6, 24, and 48 h after infection, and the data were normalized against the initial inoculum used to treat the macrophages. Petite isolates did not show intracellular replication, unlike their respiratory proficient isolates (four to eight biological replicates, ***<0.001, two-tailed t-test) (B), whereas petite isolates had a significantly higher rate of phagocytosis rate (four biological replicates, **<0.01, two-tailed t-test) (C). Macrophages were infected with fluorescein isothiocyanate (FITC)-stained BYP40 and NYP41 and counterstained with AF647 after macrophage lysis at each timepoints, and the single and double positive events were measured by fluorescent-activated cell sorting (FACS). FITC staining and AF647 counterstaining of intracellular BYP40 and BYP41 revealed that petites have an extremely limited growth as all cells were double-stained, while BYP40 showed a significant intracellular growth as evidenced by a high proportion of single-stained cells (three biological replicates) (D). Genomic interaction of green fluorescent protein (GFP) and red fluorescent protein (RFP) into CBS138 did not impact the intracellular growth and both transformants showed equally replicated inside the macrophages (four biological replicates) (E). RFP-expressing petite isolates were effectively phagocytosed by THP1 macrophages (four biological replicates, ****<0.00001, two-tailed t-test) (F), whereas they were outcompeted by their parental strains (nine biological replicates, ***<0.01 and ****<0.00001, two-tailed t-test) (G). Petite mutants enter a near dormant state after internalization by macrophages. The ATP level of BYP40 and BYP41 was normalized against colony-forming unit (CFU) incubated in RPMI or macrophages at different timepoints. Intracellular BYP41 had a significantly lower ATP level at early hours, whereas its ATP level was significantly higher than BYP40 48 h (six biological replicates) (H). Unlike their parental isolates, petite isolates imposed the least cytotoxicity 24 h post-infection as measured by lactate dehydrogenase (LDH) (eight biological replicates, ****<0.00001, two-tailed t-test) (I). FACS, fluorescent-activated cell sorting.

Fig 3

Principal component analysis (PCA) plot of all studied C. glabrata samples across studied conditions. The plot is based on vst-transformed read count data generated by DESeq2. Labels on the data points correspond to timepoints of the experiments. Percentages on PC1 and PC2 axes indicate the total amount of variance described by each axis (A). GO term enrichment analysis (category “Biological Process”) of up-regulated genes of C. glabrata at a given comparison shown on the X axis. The numbers underneath the comparisons correspond to the “counts” of clusterProfiler (i.e., total number of genes assigned to GO categories). GeneRatio corresponds to the ratio between the number of input genes assigned to a given GO category and “counts.” Only significant (P _adj < 0.05) enrichments are shown. Adjustment of P-values is done by Benjamini-Hochberg procedure (B). Petitness-specific GO terms such as tRNA and rRNA related processes, biosynthesis of several amino acids as arginine and lysine, among others. When compared to laboratory-derived petites, clinical petites showed down-regulation of carbohydrate biosynthesis and fungal cell-wall-related processes. Finally, the comparisons of non-clinical strains of normal size showed down-regulation of various processes in the clinical stains, such as nucleosome and mitotic spindle assembly, methionine, and acetate metabolism (C). Principal component analysis plot of all studied macrophage samples. The plot is based on vst-transformed read count data generated by DESeq2. Labels on the data points correspond to internal sample identifiers. Percentages on PC1 and PC2 axes indicate the total amount of variance described by each axis (D). GO term enrichment analysis (category “Biological Process”) of up-regulated genes of macrophages infected with C. glabrata strains (as depicted on X axis) compared to unchallenged macrophages (E). The numbers underneath the comparisons correspond to the “counts” of clusterProfiler (i.e., total number of genes assigned to GO categories). GeneRatio corresponds to the ratio between the number of input genes assigned to a given GO category and “counts.” Only significant (P _adj < 0.05) enrichments are shown. Adjustment of P-values is done by Benjamini-Hochberg procedure. C. glabrata petite strains induce a pro-inflammatory transcriptional program in human THP-1 macrophage cells. Summary data of differentially expressed transcripts in THP-1 macrophages at 24 h post-challenge; comparisons are between THP-1 transcriptomes challenged with petite vs non-petite C. glabrata laboratory or clinical strains (F). Gene set enrichment analysis (GSEA) indicating significantly enriched “Hallmark” pathways of the Molecular Signatures Database, based on the RNA-seq data from THP-1 cells at 24 h post-fungal challenge. The pathways are displayed based on the normalized enrichment score (NES) and the false discovery rate (FDR). The dotted line marks an FDR value of 0.25, while a select top enriched pathways are indicated Blue (enriched in non-petite) and Red (enriched in Petite) (G). GSEA enrichment plots depicting enrichment of the M1 macrophage transcriptional module (29) comparing transcriptomes THP-1 cells challenged with the petite vs non-petite C. glabrata laboratory or clinical strains at 24 h post-challenge (H). The P value reported here is the nominal P-value, while NES is the normalized enrichment score.

Fig 4

Petite phenotype is advantageous under certain stresses and echinocandin treatment. Petite and parental strains were resuspended in RPMI containing tunicamycin (endoplasmic reticulum stress; 10 µg/mL), SDS (membrane stress; 0.02%), Congo Red (cell wall integrity; 10 µg/mL), and menadione (0.5 mM); survival was assessed at designated timepoints; and the survival data were normalized to untreated control. Petites had a higher tolerance to endoplasmic reticulum (four biological replicates, ***<0.01 and ****<0.00001, two-tailed t-test) and membrane stresses (four biological replicates, **=0.01, two-tailed t-test), whereas petite isolates showed similar tolerance to oxidative and cell wall stresses (all experiments were carried out in four biological replicates, **=0.01, ***<0.01 and ****<0.00001, two-tailed t-test) (A). The survival assessment of planktonic BYP40 and BYP41 under micafungin (0.125 µg/mL) and caspofungin (0.25 µg/mL) revealed that BYP41 was more tolerant and showed monophasic and slow-killing dynamic reminiscent of tolerance phenotype defined in bacteriology (eight biological replicates) (B). Intracellular BYP41 was not responsive to either micafungin or caspofungin, whereas intracellular parental strains showed 1,000- to 10,000-fold killing compared to petite strains (four biological replicates) (C). Intracellular BYP40 and CBS138 were outcompeted by BYP41 and D5, respectively, under micafungin treatment (three biological replicates, ****<0.00001, two-tailed t-test) (D and E). BYP40 and BYP41 were equally killed by AMB in either planktonic (eight biological replicates) or intracellular conditions (four biological replicates) (F).

Fig 5

In-vivo competition of petite and their parental strains in gut colonization and systemic infection mice models. The fecal samples of CBS138-GFP and non-labeled CBS138 were collected at days 1, 3, 5, and 7, plated on agar YPD plates, and the number of GFP- and non-fluorescent colonies were enumerated. Our gut colonization model showed a slight fitness cost of genomic GFP integration in the context of gut colonization (5 mice per each group, each dot represents one mouse) (A). Both BYP41 (B) and D5 (C) were outcompeted by their parental strains in the context of gut colonization models. Genomic GFP integration carried a significant fitness cost at day 7 in the kidney (D), whereas it did not impact fitness in the spleen (four mice per each timepoint; each dot represents one mouse) (E). Similar to gut colonization, BYP41 was outcompeted by BYP40 in both immunocompromised (F and G) and immunocompetent mice (H and I).

Fig 6

BYP41 shows an improved survival in systemic infection mice treated with humanized dose of caspfungin (5 mg/kg) administered 2 h prior (A and B) or 4 h post infection (C and D).