Transcription Profiles Associated with Inducible Adhesion in Candida parapsilosis

Candida parapsilosis has emerged as a frequent cause of invasive candidiasis with increasing evidence of unique biological features relative to C. albicans. As it adapts to conditions within a mammalian host, rapid changes in gene expression are necessary to facilitate colonization and persistence in this environment. Adhesion of the organism to biological surfaces is a key first step in this process and is the focus of this study. Building on previous observations showing the importance of a member of the ALS gene family in C. parapsilosis adhesion, three clinical isolates were cultured under two conditions that mimic the mammalian host and promote adhesion, incubation at 37°C in tissue culture medium 199 or in human plasma. Transcriptional profiles using RNA-seq were obtained in these adhesion-inducing conditions and compared to profiles following growth in yeast media that suppress adhesion to identify gene expression profiles associated with adhesion. Overall gene expression profiles among the three strains were similar in both adhesion inducing conditions and distinct from adhesion suppressing conditions. Pairwise analysis among the three growth conditions identified 133 genes that were differentially expressed at a cutoff of +/- 4-fold, with the most upregulated genes significantly enriched in iron acquisition and transmembrane transport, while the most downregulated genes were enriched in oxidation-reduction processes. Gene family enrichment analysis identified gene families with diverse function that may have an important role in this important step for colonization and disease. Total RNA was prepared from three clinical Candida parapsilosis isolates exhibiting adhesive phenotypes which were incubated at 37°C in YPD, M199, or pooled human plasma. All samples were prepared in triplicate. rRNA depleted RNA was sequenced with 150bp paired-end reads using the Illumina HiSeq2000 platform. Reads were aligned and gene-level expression count estimates were calculated using Salmon in mapping-based mode.