Pathogenesis of the yeast Candida albicans

Virulence of Candida albicans: The goal of our research is to understand the mechanisms that control the pathogenicity of the yeast C. albicans. C. albicans is the most common and arguably the most important causative agent of human fungal infections. It is a major opportunistic pathogen of immunocompromised hosts, including AIDS patients and those undergoing chemotherapy, tissue transplants or with central venous catheters. Studies indicate that up to 90% of AIDS patients suffer from oropharyngeal and esophageal candidiasis, in which C. albicans is the major causative agent. C. albicans can exist in different morphological states including budded or yeast-like cells and hyphal forms (Fig. 1), and there is strong evidence that the ability to switch between these morphological states is essential for its virulence. Understanding the input signals and output responses that lead to the budded-to-hyphal morphological transition will undoubtedly lead to tremendous insight into virulence mechanisms and may ultimately lead to new anti-fungal therapeutic targets and drugs. This point is especially critical since there are serious side effects due to renal and liver dysfunction associated with the polyenes (i.e., amphotericin B, nystatin) that are usually used to treat C. albicans infections. In addition, a significant increase in resistance to the less toxic azole drugs (i.e., fluconoazole) has occurred within the patient population, especially HIV-positive patients.

C. albicans pathogenicity can be attributed to its ability to survive and thrive in multiple microenvironments within the host, including multiple organs, the mucosa, and the bloodstream, and to virulence factors that aid in the adherence and invasion of multiple cell types. C. albicans exists within immunocompetent individuals in a commensal relationship on mucosal linings of the oral cavity, esophagus, vagina, and gastrointestinal tract. Oropharyngeal, esophageal, vulvovaginal, and cutaneous candidiasis leads to significant morbidity while lethality is usually associated with systemic infections in immunocompromised patients. Systemic infections arise from colonization of mucosal surfaces by adherence to epithelial cells, followed by the penetration of epithelial and endothelial cell barriers and dissemination throughout the body. Therefore, the vascular endothelium has a key role in the dissemination of C. albicans which leads to establishment of systemic candidiasis. The endothelial cells are sensitive to invasion and damage by C. albicans, which requires the attachment of germinated C. albicans to the endothelial cells, their phagocytosis, and induction of hyphal-specific genes. Both budded and hyphal forms of C. albicans have been identified in infections and are important for virulence. The budded form is found predominantly attached to epithelial cells and the hyphal form is found disseminated, presumably due to its enhanced ability to invade cells through the production and secretion of proteinases and hydrolases. The budded-to-hyphal transition is critical for systemic infections, a premise that was reinforced in non-virulent C. albicans mutants defective in hyphal formation.

C. albicans exist in different morphological states (Fig. 1; adapted from Calderone, 2002, Candida and Candidiasis, ASM Press) including round to ovoid budded or yeast-like cells (A) and filamentous forms including pseudohyphae (B, chains of ellipsoidal cells with constricted septa between mother and daughter cells) and hyphae (C, long filaments with no constriction at septa). Hyphal cells are morphologically distinct from budded and pseudohyphal cells, with differences in septation patterns, actin dynamics, and growth properties. Hyphal growth is the result of alterations in polarized growth patterns and in transit through the cell cycle. The budded-to-hyphal transition occurs in response to environmental stress signals such as temperature above 35oC, pH above 6.5, nitrogen and/or carbon starvation, low oxygen concentration, and changes in cell density, growth in serum or other chemicals such as N-acetylglucosamine, proline and other amino acids, or alcohols, and the presence of host macrophages. In response to these hyphal-inducing signals, a variety of hypha-specific genes are increased in expression, including those that encode cell surface adhesins and secreted proteinases. Induced genes ALS3, ALS8, and HWP1 (adhesins), HYR1 (cell surface glycoprotein), SAP4, 5, 6 (secreted aspartyl proteinases), and ECE1 (unknown function) are important reporters for examining the effects of different signaling pathway on the budded-to-hyphal transition.

Signaling pathways regulating C. albicans morphogenesis: Given that C. albicans cells respond to a variety of different growth and environmental signals to induce the budded-to-hyphal transition, it is not surprising that multiple signaling pathways function in hyphal development (Fig. 2). Mutational analyses indicate that the Tup1p, Cph1p, and Efg1p transcriptional regulatory proteins function independently in promoting (Cph1p, and Efg1p ) or inhibiting (Tup1p) hyphal development. The Cst20p-Hst7p-Cek1p MAP kinase pathway, which is homologous to the Ste20p-Ste7p-Kss1p pathway in S. cerevisiae, signals to the Cph1p (S. cerevisiae Ste12p) transcription factor to induce hypha-specific genes. Two Cdc42p-interacting proteins, the p21-activated kinase (PAK) CaCla4p and a phosphatidylinositol 3-kinase CaVps34p, are also involved in hyphal development. In a parallel pathway, the Efg1p transcription factor functions downstream of Tpk2p, the catalytic component of protein kinase A, and the Ras GTPase to integrate a cAMP signal into hyphal development. Recently, the Cos1p/Nik1p two-component histidine kinase, a Hog1p MAP kinase, and the Crk1p cyclin-dependent kinase have also been shown to be involved in hyphal development. It is clear that these different signaling pathways are activated by separate but overlapping environmental signals, eventually leading to rearrangements of the actin- and tubulin-based cytoskeleton. For instance, the Efg1p pathway can be triggered by pH, N-acetylglucosamine, proline, and nitrogen and/or carbon starvation while the Cph1p pathway is stimulated by nitrogen starvation but not in response to N-acetylglucosamine, amino acids, or serum. In addition, the hypha-specific genes ALS8, ECE1, HWP1, and HYR1 are regulated primarily by the Efg1p pathway with the Cph1p pathway being less important. However, the molecular mechanisms that underlie these different signaling pathways and their potential cross-talk are still unclear.

The budded-to-hyphal transition in C. albicans is analogous in some respects to the yeast-to-pseudohyphae transition in S. cerevisiae. Diploid S. cerevisiae cells respond to nitrogen starvation by altering their cell cycles, budding patterns, polarized growth patterns, cell shape, and cell separation patterns, resulting in polarized elongated budded cells that resemble fungal hyphae. Haploid S. cerevisiae cells can also be induced to filamentous growth, which is manifested as invasive growth into agar plates. Mutational analyses indicate that the actin cytoskeleton plays a critical role in various aspects of pseudohyphal growth. One route leading to pseudohyphal growth encompasses the Ras2p GTPase signaling to the Cdc42p GTPase and to several components of the pheromone response MAP kinase cascade, including Ste20p, Ste11p, Ste7p, and Kss1p. This signaling cascade leads to the activation of the Ste12p transcription factor that together with the Tec1 transcription factor induces the expression of genes necessary for filamentous growth. The role of Cdc42p in this pathway was deduced from the observations that expression of the dominant negative Cdc42D118A mutant protein inhibited Ras2-dependent filamentous growth, and that expression of the activated Cdc42G12V mutant protein induced filamentous growth. The mechanisms by which Cdc42p responds to cell cycle and nutritional signals to regulate the switching between different morphogenetic patterns in S. cerevisiae is still unclear, but certainly Cdc42p interactions with the actin cytoskeleton and MAP kinase pathways do play a critical role in this regulation.

Cdc42p signaling pathways: Cdc42p and other Rho-type GTPases regulate the signal-transduction pathways that mediate cellular asymmetry or cell polarity in many, if not all, eukaryotic cell types. GTPase signaling modules consist of regulators of the guanine-nucleotide bound state, i.e., guanine-nucleotide exchange factors (GEFs), guanine-nucleotide dissociation inhibitors (GDIs), and GTPase-activating proteins (GAPs), as well as downstream effectors. In S. cerevisiae, Cdc24p is the sole Cdc42p-GEF and Bem3p, Rga1p, and Rga2p are three Cdc42p-GAPs. C. albicans Cdc42p is 88% identical to S. cerevisiae Cdc42p (69), and a preliminary characterization of its functions in C. albicans morphogenesis indicated that CaCdc42p was essential for viability and both budded and hyphal growth. Based on amino acid similarity, we have identified a single C. albicans Cdc24p homolog that displays 35% amino acid identity to S. cerevisiae Cdc24p, which approximates the level of identity usually seen between Cdc42p-GEFs from different species. This relatively low level of sequence identity suggests that C. albicans Cdc24p may have unique (i.e., C. albicans-specific) sequences not present in other mammalian GEFs, making it a good candidate for a therapeutic target. In addition, C. albicans homologs of other components of Cdc42-dependent signaling pathways, including Bem1p, Bem3p, Bni1p, Bud6p, Cla4p, Iqg1p, Pfy1p, Ras2p, Rga1p, Rga2p, Rsr1p, and Spa2p (Fig. 3), have been identified through the C. albicans Genome Project, raising the possibility that these signaling pathways are conserved.

A myriad of downstream effectors (see S. cerevisiae research page) interact with the activated (GTP-bound) form of Cdc42p through an ~25 amino acid effector domain. Recent data from our lab and others indicate that there are differential interactions between downstream effectors and S. cerevisiae Cdc42p effector-domain residues. These differential interactions likely affect the formation of different multi-protein complexes and hence different morphological processes. For instance, the S. cerevisiae cdc42V44A and cdc42D38E mutations both affect the Swe1p-dependent morphogenetic checkpoint and G2/M transition through a decrease in interaction with the Cla4p PAK-like kinase, but the cdc42D38E mutation also affects the maintenance of polarized growth post-bud emergence through altered interactions with the Bem3p GAP. Recently, S. cerevisiae cdc42 mutations were isolated that abolished its role in pseudohyphal growth without affecting its essential mitotic function. These mutations affected interactions with a small subset of downstream effectors and would be predicted, when inserted into C. albicans Cdc42p, to lead to defects in the budded-to-hyphal transition and decreased virulence. We have incorporated several of these mutations into C. albicans CDC42 and showed that the budded-to-hyphal transition was affected, as were several signal transduction pathways (vandenBerg et al., 2004). Hence, disruption of specific Cdc42p-effector interactions may also be a potential therapeutic goal.

Use of small molecules to study cell biological process: There is no question that the use of small organic molecules in deciphering complex biological processes has been tremendously fruitful. For example, much of what is known about actin-based processes within the cell comes from studies using specific inhibitors of actin structure or function such as the cytochalasins, latrunculins, and phalloidins. In addition, molecules such as nocodozole and hydroxyurea have been used extensively as sychronization tools in the study of the cell division cycle in eukaryotes. These types of small molecules have proven invaluable because they interact with their target proteins as agonists or antagonists in a highly specific manner, which allows definitive conclusions to be drawn regarding a protein¹s function in a particular process. This type of approach is analogous to a reverse genetic approach in that molecules are used to study known protein targets that are presumed to be involved in a cellular process. Small molecules can also be used in a forward genetic approach in which no presumptions are made about what proteins are involved in the process. This occurs through the screening of large, diverse sets of individual molecules for a phenotype (or reversal of a phenotype) associated with a cellular process. This approach has been successful in studying a wide variety of biological processes. For example, a new inhibitor of the mitotic kinesin Eg5, termed monastrol, was identified from a high-throughput phenotype-based screen and has proved very useful in the study of mitotic spindles. It should be noted that this type of phenotype-based screening is particularly useful in situations where classical forward genetic approaches are not possible (i.e., isolation of recessive mutations in a diploid species), and is another reason why we have decided to use this approach to study the budded-to-hyphal transition in the constitutive diploid C. albicans. We have identified 5 novel molecules that can inhibit the budded-to-hyphal transition without affecting budded growth and shown that these molecules differentially affect several cellular processes including the induction of hyphal-specific genes and endocytosis (Toenjes et al., 2005). We have also identified 16 molecules with known cellular targets that inhibit the budded-to-hyphal transition and these molecules affect multiple signaling pathways in the cell.

It is clear that Cdc42-dependent signal-transduction pathways are highly conserved among eukaryotes. The role of these pathways in the pathogenicity of C. albicans is unclear and given the unique characteristics of C. albicans versus S. cerevisiae as a human pathogen, it is likely that these pathways will be more complex and respond to different signals in C. albicans. Thererfore, our extensive knowledge of S. cerevisiae pathways and protein functions is used primarily as a guide to the study of C. albicans. Our studies bring to bear an arsenal of genetic, molecular genetic, biochemical, and cell biological approaches, designed to test several hypotheses centered around the tenet that Cdc24p-dependent activation of Cdc42p signal transduction pathways is required for morphological changes and subsequent virulence of C. albicans. Specifically, we are elucidating the mechanisms by which environmental signals that induce hyphal growth and virulence regulate Cdc42p signaling pathways and localization of signaling proteins. In addition, we are circumventing the inherent complexities associated with classical genetic screens with the diploid C. albicans by using a new high-throughput small molecule screen to identify inhibitors and enhancers of the morphological changes. The bioactive molecules identified in this screen are being used to identify and characterize components of the pathways regulating the budded-to-hyphal transition and the molecular mechanisms by which these components function. In addition, the effects of these bioactive molecules, or their derivatives, on C. albicans virulence is being examined, with the ultimate goal to identify new anti-fungal molecules or targets for anti-fungal therapeutics. Taken together, these approaches will illuminate the role of Cdc42p-dependent signal transduction pathways in inducing C. albicans virulence and may provide valuable insight into new and novel therapeutic targets and drugs.