Metabolome analysis of the yeast Candida albicans during morphogenesis

Abstract

Candida albicans is a human commensal and opportunistic pathogenic fungus that causes a high mortality rate in severely immunocompromised patients. The ability to change from yeast to filamentous growth and vice-versa, in response to various environmental factors, is considered to be a critical virulence factor of this fungus. Despite the fact that many studies have elucidated the effect of signal transduction pathways and quorum sensing molecules on fungal morphogenesis, the downstream metabolic mechanisms that respond to such signalling molecules and trigger the morphological change at a system level, however, remain unclear. Therefore, the principal aim of this research was to investigate the metabolic reprogramming of C. albicans morphogenesis in vitro using an emerging post-genomics approach, metabolomics. The specific objectives of this research were: (i) to determine changes in the metabolic pathways of C. albicans during the yeast-to-filamentous transition using metabolomics; (ii) to determine the metabolic response of C. albicans to quorum sensing molecules involved in morphogenesis; and (iii) geneknockout mutagenesis of selected metabolic pathways from the central carbon metabolism likely to be associated with morphogenesis. The chapters of this thesis are a compilation of the work that has been published or submitted for publication during my PhD studies. Chapter one reviews the role of primary metabolism and quorum sensing molecules of C. albicans in morphogenesis. The central carbon metabolism and sterol biosynthesis pathways of C. albicans were reconstructed in silico based on its genome sequence, emphasising the metabolic pathways known to be related to the morphological transition and virulence along with pathways engaged in the biosynthesis of quorum sensing molecules. This initial investigation laid out an important metabolic framework that guided the interpretation of subsequent metabolomics data and was published as a review article in the journal Fungal Genetics & Biology. To understand the metabolic mechanisms behind C. albicans’ morphogenesis, I compared the metabolite profiles of C. albicans cells grown under hyphae-inducing conditions to the metabolite profiles of growing yeast cells. This work is described in chapter two and was published in the journal Metabolomics. The results revealed a global downregulation of cellular metabolism during the yeast-to-hypha transition. Specifically, seventeen metabolic pathways involved in the central carbon metabolism were significantly downregulated under all three hyphae-inducing conditions. These included metabolic pathways associated with metabolism of amino acids, C5-branched dibasic acid, glutathione, nicotinate/nicotinamide, nitrogen, purine, pyruvate, and acetyl-CoA biosynthesis. These results indicate that downregulation of these central carbon metabolic pathways is likely to be intrinsically involved in the C. albicans morphogenetic process. In addition, I demonstrated that filamentous cells contained significant lower concentrations of ATP compared to cells growing in the yeast form. Therefore, these data corroborated the metabolomics results and provided strong evidence that a global downregulation of central carbon metabolism is taking place during the filamentous growth of C. albicans. In order to validate my observations concerning C. albicans’ metabolic response to hyphae-inducing conditions, I investigated the metabolic changes of C. albicans when hyphal induction was repressed by quorum sensing molecules such as farnesol (chapter 3) and phenylethyl alcohol (chapter 4). These signalling molecules are known to suppress the germ tube formation (initial phase of filamentous growth) of C. albicans. Confirming my observations during hyphae-inducing conditions, I demonstrated a general upregulation of metabolic pathways from the central carbon metabolism when filamentous growth was inhibited by either quorum sensing molecule. By integrating the metabolic profiles from farnesol (chapter 3) and phenylethyl alcohol (chapter 4) experiments with the earlier metabolomics data (chapter 2); I was able to identify seven metabolic pathways that were affected in a consistent fashion in all these studies. Therefore, these seven metabolic pathways are likely to be closely related with the morphogenetic process of C. albicans. I also observed a metabolic reprogramming, especially the upregulation of lipid metabolism in response specifically to farnesol and phenylethyl alcohol. This indicates that C. albicans may modify its membrane composition in order to minimise the antimicrobial effects of these quorum sensing molecules when they are present at high concentrations. Thus, these results present important new understandings of the metabolic role of quorum sensing in C. albicans metabolism. In order to determine whether or not the quorum sensing molecules were taken up and metabolised by C. albicans, I used isotope labelling experiment to trace the metabolic fate of phenylethyl alcohol under hyphae-inducing conditions (chapter 4). The isotope labelling patterns showed that phenylethyl alcohol was taken up by C. albicans cells and broken-down intracellularly. Its labelled carbons ended up in the majority of amino acids as well as in lactate and glyoxylate. However, the highest level of carbon labelling was in the pyridine ring of NAD+/NADH and NADP+/NADPH molecules, indicating that these nucleotides were the major products of phenylethyl alcohol catabolism. Coincidently, two metabolic pathways where these nucleotides play important roles, nitrogen metabolism and nicotinate/nicotinamide metabolism, were among the seven pathways short-listed from the earlier metabolomics results as metabolic pathways likely to be closely involved in C. albicans morphogenesis. Therefore, the final experimental work of this PhD project (chapter 5) was the disruption of a metabolic reaction important to nitrogen and nicotinate/nicotinamide metabolism through gene-knockout mutagenesis and subsequent evaluation of the morphogenetic ability of the knockout mutants. Thus, I selected two important reactions involved in both the nitrogen metabolism and redox balance of C. albicans to be individually deleted using the SAT1 gene disruption method. The selected reactions were those catalysed by NAD+-dependent glutamate dehydrogenase (encoded by GDH2) and by NADPH-dependent glutamate dehydrogenase (encoded by GDH3). The ability of gdh2/gdh2 and gdh3/gdh3 mutant strains to undergo morphogenesis was investigated under various hyphaeinducing conditions. The mutants displayed altered morphogenesis and growth rates only when grown on arginine or proline as the sole carbon and nitrogen sources. Both arginine and proline belong to the glutamate family of amino acids. For them to be used as carbon and nitrogen sources they must be first catabolised into glutamate through a series of biochemical reactions, followed by a transamination reaction to form α-ketoglutarate and ammonia, which is catalysed by glutamate dehydrogenase dependent on either NAD+ (GDH2) or NADPH (GDH3). Therefore, the mutants were likely experiencing a change in their redox balance when growing on media supplied with only arginine or proline as carbon and nitrogen sources. The ghd2/gdh2 mutant strain showed reduced filamentous growth on both proline and arginine, whilst the ghd3/ghd3 mutant strain showed reduced filamentous growth only when growing on proline. Thus, I demonstrated that some amino acids such as arginine and proline induce morphogenesis in C. albicans through nitrogen metabolism, most likely altering the redox balance of the cell. However, to better understand the mechanism of the reduced filamentous growth of these mutants would require a metabolomic study of the mutant strains in comparison to the wild type. Unfortunately these experiments were beyond the scope of this project and could be further pursued in future studies. In conclusion, I have demonstrated that a metabolomics approach is an effective bottom-up approach to understand the metabolic changes involved with the morphogenetic process of C. albicans. My results not only contribute to the current understanding of fungal morphogenesis and quorum sensing mechanisms, but they also provided the first attempt to connect signal transduction regulation of cellular process with downstream central carbon metabolism (chapter 6).