(a) Phylogeny of I. orientalis strains selected for this study. Six tolerant strains (purple) and six susceptible strains (orange) were selected. The heatmap shows the relative growth of each strain under low pH conditions compared to the control pH (~5.5). (b) Optical density (OD) curves representing growth profiles of 12 selected strains at pH = 1.5. Purple and orange curves represent tolerant and susceptible strains, respectively. (c) Pairwise Spearman correlation between gene expression profiles [log2(CPM) of all genes, after 1 h of treatment at pH 1.5] of all replicates of all strains. (d-f) Low-dimensional representations of transcriptomic profiles of all replicates of all strains (log2(CPM)) in the control condition (pH ~ 5.5) (d) or pH = 1.5 treatment after 1 h (e). Panel f shows the low-dimensional representations of changes in [log2(CPM)] transcriptomic profiles between 0 (control) and 1 h (treatment) samples, for all strains. Orange and purple points represent susceptible and tolerant strains, respectively. For dimensionality reduction, we first used principal component analysis (PCA) to obtain PCs, which were reduced to two-dimensional representations using UMAP. (g) UMAP plot of log2(CPM) counts for all replicates of tolerant strains subjected to varying pH levels. Colors represent the pH scale. The gray arrow indicates pH change toward more acidic media.Because of its natural stress tolerance to low pH, Issatchenkia orientalis (a.k.a. Pichia kudriavzevii) is a promising non-model yeast for bio-based production of organic acids. Yet, this organism is relatively unstudied, and specific mechanisms of its tolerance to low pH are poorly understood, limiting commercial use. In this study, we selected 12 I. orientalis strains with varying acid stress tolerance (six tolerant and six susceptible) and profiled their transcriptomes in different pH conditions to study potential mechanisms of pH tolerance in this species. We identified hundreds of genes whose expression response is shared by tolerant strains but not by susceptible strains, or vice versa, as well as genes whose responses are reversed between tolerant and susceptible strains. We mapped regulatory mechanisms of transcriptomic responses via motif analysis as well as differential network reconstruction, identifying several transcription factors, including Stb5, Mac1, and Rtg1/Rtg3, some of which are known for their roles in acid response in Saccharomyces cerevisiae. Functional genomics analysis of short-listed genes and transcription factors suggested significant roles for energy metabolism and translation-related processes, as well as the cell wall integrity pathway and RTG-dependent retrograde signaling pathway. Finally, we conducted additional experiments for two organic acids, 3-hydroxypropionate and citramalate, to eliminate acid-specific effects and found potential roles for glycolysis and trehalose biosynthesis specifically for response to low pH. In summary, our approach of comparative transcriptomics and phenotypic contrasting, along with a multi-pronged bioinformatics analysis, suggests specific mechanisms of tolerance to low pH in I. orientalis that merit further validation through experimental perturbation and engineering.
