Candida albicans exhibits heterogeneous and adaptive cytoprotective responses to antifungal compounds.

Candida albicans, an opportunistic human pathogen, poses a significant threat to human health and is associated with significant socio-economic burden. Current antifungal treatments fail, at least in part, because C. albicans can initiate a strong drug tolerance response that allows some cells to grow at drug concentrations above their minimal inhibitory concentration. To better characterize this cytoprotective tolerance program at the molecular single-cell level, we used a nanoliter droplet-based transcriptomics platform to profile thousands of individual fungal cells and establish their subpopulation characteristics in the absence and presence of antifungal drugs. Profiles of untreated cells exhibit heterogeneous expression that correlates with cell cycle stage with distinct metabolic and stress responses. At 2 days post-fluconazole exposure (a time when tolerance is measurable), surviving cells bifurcate into two major subpopulations: one characterized by the upregulation of genes encoding ribosomal proteins, rRNA processing machinery, and mitochondrial cellular respiration capacity, termed the Ribo-dominant (Rd) state; and the other enriched for genes encoding stress responses and related processes, termed the Stress-dominant (Sd) state. This bifurcation persists at 3 and 6 days post-treatment. We provide evidence that the ribosome assembly stress response (RASTR) is activated in these subpopulations and may facilitate cell survival.

Many drugs currently used to treat fungal diseases are becoming less effective. This is partly due to the rise of antifungal resistance, where certain fungal cells acquire mutations that enable them to thrive and proliferate despite the medication. Antifungal tolerance also contributes to this problem, wherein certain cells can continue to grow and multiply, while other ­ genetically identical ones ­ cannot. This variability is partly due to differences in gene expression within the cells. The specific nature of these differences has remained elusive, mainly because their study requires the use of expensive and challenging single-cell technologies. To address this challenge, Dumeaux et al. adapted an existing technique to perform single-cell transcriptomics in the pathogenic yeast Candida albicans. Their approach was cost effective and made it possible to examine the gene expression in thousands of individual cells within a population that had either been treated with antifungal drugs or were left untreated. After two to three days following exposure to the antifungal treatment, C. albicans cells commonly exhibited one of two states: one subgroup, the 'Ribo-dominant' cells, predominantly expressed genes for ribosomal proteins, while the other group, the 'Stress-dominant' cells, upregulated their expression of stress-response genes. This suggests that drug tolerance may be related to different gene expression patterns in growing cell subpopulations compared with non-growing subpopulations. The findings also indicate that the so-called 'ribosome assembly stress response' known to help baker's yeast cells to survive, might also aid C. albicans in surviving exposure to antifungal treatments. The innovative use of single-cell transcriptomics in this study could be applied to other species of fungi to study differences in cell communication under diverse growth conditions. Moreover, the unique gene expression patterns in C. albicans identified by Dumeaux et al. may help to design new antifungal treatments that target pathways linked to drug resistance.

A Deep Learning Approach to Capture the Essence of Candida albicans Morphologies.

We present deep learning-based approaches for exploring the complex array of morphologies exhibited by the opportunistic human pathogen Candida albicans. Our system, entitled Candescence, automatically detects C. albicans cells from differential image contrast microscopy and labels each detected cell with one of nine morphologies. This ranges from yeast white and opaque forms to hyphal and pseudohyphal filamentous morphologies. The software is based upon a fully convolutional one-stage (FCOS) object detector, a deep learning technique that uses an extensive set of images that we manually annotated with the location and morphology of each cell. We developed a novel cumulative curriculum-based learning strategy that stratifies our images by difficulty from simple yeast forms to complex filamentous architectures. Candescence achieves very good performance (~85% recall; 81% precision) on this difficult learning set, where some images contain hundreds of cells with substantial intermixing between the predicted classes. To capture the essence of each C. albicans morphology and how they intermix, we used a second technique from deep learning entitled generative adversarial networks. The resultant models allow us to identify and explore technical variables, developmental trajectories, and morphological switches. Importantly, the model allows us to quantitatively capture morphological plasticity observed with genetically modified strains or strains grown in different media and environments. We envision Candescence as a community meeting point for quantitative explorations of C. albicans morphology. IMPORTANCE The fungus Candida albicans can "shape shift" between 12 morphologies in response to environmental variables. The cytoprotective capacity provided by this polymorphism makes C. albicans a formidable pathogen to treat clinically. Microscopy images of C. albicans colonies can contain hundreds of cells in different morphological states. Manual annotation of images can be difficult, especially as a result of densely packed and filamentous colonies and of technical artifacts from the microscopy itself. Manual annotation is inherently subjective, depending on the experience and opinion of annotators. Here, we built a deep learning approach entitled Candescence to parse images in an automated, quantitative, and objective fashion: each cell in an image is located and labeled with its morphology. Candescence effectively replaces simple rules based on visual phenotypes (size, shape, and shading) with neural circuitry capable of capturing subtle but salient features in images that may be too complex for human annotators.