Virulence is a multi-faceted feature of microorganisms that includes the ability to thrive within host tissue, withstand rapid elimination by the immune system and the capability to cause tissue and/or organ damage. Keeping this in mind, it is not surprising that genes involved in various processes, from basic cell biology to specific interactions with host cells, have been demonstrated to affect the virulence potential of C. albicans. Here, we will focus on factors that influence the interaction of the fungus with host cells, nutrient acquisition and will briefly discuss the role of the cell wall and morphogenesis for candidiasis.
Interaction of Candida albicans with Epithelial Cells
As a commensal C. albicans lives in close contact with– or depending on the presence of a mucus layer, in proximity to– epithelial cells without causing overt damage. This changes during infection characterized by increased fungal load, invasion into epithelial barriers (and underlying tissue), induction of an immune response and tissue damage. The factors that trigger the transition from commensalism to infection are only incompletely understood, but the disturbance of the microbiota clearly plays an important role.
The first step toward infection is attachment of the yeasts to host cells, in the case of mucosal membranes epithelial cells (Fig. 1). This complex process is likely initiated by passive forces, e.g., hydrophobic cell–cell interactions, followed by specific interactions between fungal and host cell surface molecules. While some adhesins also occur on yeast cells, hyphae express adhesins, like Als3 and Hwp1, in larger number and more efficiently. Targeted host structures include cadherins, integrins, and components of the extracellular matrix. Of note, adhesion to host cells is a strong trigger of filamentation in vitro and hypha formation is important for the second step during infection: invasion (Fig. 1). Invasion into host cells occurs by two mechanisms, induced endocytosis and active penetration. Active penetration is a fungal-driven process that requires directed filamentous growth. The current knowledge suggests that in addition to physical forces provided by the turgor pressure, hydrolytic enzymes, especially secreted aspartic proteases (Saps) support this process. Inhibition of aspartic proteases by pepstatin A significantly reduces invasion; the contribution of individual Saps to this process is however still controversially discussed. In contrast to active penetration, induced endocytosis requires the activity of host cells and occurs even with inactivated fungal cells. It is initiated by the recognition of distinct fungal surface proteins (“invasins”) by host cell receptors, which triggers rearrangement of the cytoskeleton leading to uptake of the fungal cell. The first C. albicans invasin identified was Als3 that binds to cadherins on host cells. Another known C. albicans invasin is Ssa1 and further host receptors include the EGF receptor and the human epidermal growth factor receptor 2. Importantly, as Als3 is a hypha-specific protein, both active penetration and induced endocytosis depend largely on hypha formation. Therefore, hyphae are considered to be the invasive morphology in C. albicans.
Fig1. Adhesion, invasion, and damage by C. albicans in the human gut. Major steps of the C. albicans life-cycle during the switch from commensal to pathogen. Hyphae-associated adhesins (blue anchors) mediate the attachment to enterocytes. Growing hyphae and subsequent release of candidalysin and proteases (orange scissors) and lipases penetrates enterocyte barrier and destroys underlying cell layers. Illustration by M.J. Niemiec and M. Polke.
Invasion is a prerequisite for damage but not necessarily linked to damage of host cells: As the host membrane remains intact following initial invasion, forming a glove-like structure surrounding growing hyphae (Wächtler et al., 2012; Zakikhany et al., 2007), the host cell is not immediately lysed. Indeed, specific C. albicans mutants have been identified that show unaltered adhesion and invasion but significantly reduced damage in the interaction with epithelial cells. One of these mutants is C. albicans eed1Δ/Δ, lacking EED1, a factor that is essential for maintenance of hyphal growth and thus hyphal elongation. Following invasion, this mutant reverts back to yeast growth intracellularly, thereby reducing the strain on the host cell membrane afflicted by growing hyphae of wild type C. albicans. Other examples are mutants lacking either DUR31 or HSP21, that likewise show reduced hyphae formation following invasion. Most important, however, was the discovery that an ECE1 deletion mutant is highly attenuated in epithelial damage, despite normal filamentation. ECE1 was already identified in the 1990s as one of the genes most highly expressed on hyphae compared to yeast cells. Yet again, its function was only elucidated in 2016: ECE1 codes a protein that is cleaved into eight short peptides by Kex2. Peptide 3 generated by this cleavage adopts an a-helical structure and integrates into host cell membranes, leading to permeabilization and lysis. Due to this effect, the peptide was named candidalysin. Thus, candidalysin links hyphal morphology and damage beyond physical rupture by continuous filamentous growth and provides further explanation for the significant role of filamentation in pathogenesis.
Interaction of Candida albicans with the Immune System
Immune cells are rapidly recruited during candidiasis and especially innate phagocytes are efficient and essential effector cells during fungal infections (Fig. 2). In fact, impairment of the immune system represents a risk factor for candidiasis, but not all patients suffering from disseminated C. albicans infections are obviously immunocompromised. To be a successful pathogen, C. albicans needs to be able to either prevent recognition or survive antimicrobial mechanisms employed by immune cells to a certain extend. As the immune response to C. albicans is addressed in detail in another module of this series, we will focus here on some of the mechanisms by which the fungus withstands elimination of the immune system.
Fig2. Killing of C. albicans by professional human phagocytes. Killing of C. albicans by neutrophils and macrophages occurs intra- and extracellularly (IC/EC). IC: Internalization of Candida via phagocytosis leads to inactivation of fungus by ROS (yellow lightning) and host effector proteins. EC: Chemokines are released (purple glow) by activated phagocytes to recruit more phagocytes. Antifungal effector proteins are released into cell environment during degranulation. Neutrophils release extracellular traps (NETs) to entangle and kill Candida. Notably, C. albicans can outgrow macrophages, not neutrophils. Illustration by M.J. Niemiec.
The immune system can be roughly divided into cellular and humoral effector mechanisms. Humoral factors include antibodies, the complement system and antimicrobial peptides (AMPs). Antibodies are readily produced in humans and mice following exposure to C. albicans but naturally induced antibodies are considered to have only a limited protective effect (recently reviewed in Richardson and Moyes. The complement system and AMPs are evolutionary old and comprise two of the first lines of defense against invading microbes. Furthermore, AMPs are produced by mucosal epithelial cells even in the absence of a pathogenic threat and thereby control the commensal microbiota. Several AMPs, b-defensins, the cathelicidin LL-37 and histatin 5, have been shown to kill C. albicans in vitro. It has furthermore been proposed that the reduced levels of histatin 5 observed in the saliva of HIVþ patients contribute to the increased risk of these patients for oropharyngeal candidiasis. Although this and the hypothesis that AMPs control fungal growth on mucosal surfaces in vitro appear plausible, only few studies have been conducted to address this experimentally. Nonetheless, C. albicans also seems to have evolved AMP resistance mechanisms, like inactivation through proteolytic cleavage of AMPsbySap9 andSap10and inactivation by a secreted glycodomain of Msb2. Tolerance to AMPs can furthermore be induced by induction of compensatory mechanisms, for example, Pbs2 via the Hog1 stress response pathway, or stabilization of mitochondrial integrity via Ssd1 and Bcr1. Similarly, C. albicans interferes with complement activation by various means, including proteolytic cleavage of active complement components by Saps and binding of host factors that regulate complement activation. To which extend these immune evasion mechanisms contribute to pathogenesis is not fully clear; however, complement defects render mice more susceptible to systemic candidiasis. Furthermore, opsonization by complement enhances phagocytosis and fungal killing, thereby linking complement evasion to other survival strategies of C. albicans.
Professional phagocytes, especially neutrophils and monocytes/macrophages, constitute cellular components of the first line of defense against C. albicans. Following recognition of the fungus via interaction of pattern recognition receptors (PRRs) with fungal surface components acting as pathogen-associated molecular patterns (PAMPs) , immune cells become activated and either phagocytose the pathogen or, in the case of neutrophils, can undergo degranulation or form neutrophil extracellular traps (NETs) via programmed cell death to release antifungal effector molecules (Fig. 2). The cell wall composition and, more importantly, exposure of distinct components on the surface significantly influences the efficacy of recognition and the subsequent response induced; thus, cell wall modifications are one way by which C. albicans can influence recognition by phagocytes and, additionally, also phagosome maturation. These mechanisms are discussed in detail below.
Phagocytosis requires physical uptake of the pathogen and is therefore influenced by size and shape. C. albicans hyphae can grow to a length that prevents complete phagocytosis. Therefore, the ability of C. albicans to form filaments affects the interaction with phagocytes. However, yeast cells can readily be phagocytosed and phagolysosome maturation creates a hostile environment characterized by reduced pH, a high concentration of AMPs, reactive oxygen and nitrogen species (ROS and RNS, respectively), hydrolytic enzymes and nutrient starvation C. albicans is able to escape from macrophages by either triggering pyroptosis, a programmed cell-death pathway, or by affecting phagolysosome maturation, development of long hypha, and physical piercing of macrophages. Induction of pyroptosis by C. albicans is associated with hypha induction but does not require morphogenesis, but rather remodeling of the cell wall associated with increased 1,3 b-glucan exposure, and ergosterol. Furthermore, the fungal peptide toxin Candidalysin can trigger apoptosis.
Phagocytes typically produce significant amounts of ROS and RNS upon encountering C. albicans and oxidative mechanisms are important for killing of C. albicans . The fungus responds by activating a general stress response that involves the protein kinase Hog1 and the more specific oxidative and nitrosative stress response pathway involving the transcription factors Cap1 and Cta4, respectively. The activated stress response results in increased expression of detoxification systems, that include superoxide dismutases (Sods), catalase 1, a glutathione-based system including glutathione peroxidases and reductases, and flavodoxin-like proteins, that are protective against oxidative stress. Resistance against nitrosative stress depends on the flavohemoglobin NO scavenger Yhb1. Upon phagocytosis, C. albicans furthermore upregulates certain metabolic pathways, including gluconeogenesis, enzymes for the utilization of alternative carbon sources, b-oxidation of fatty acids and the glyoxylate cycle, whereas glycolysis and protein synthesis are downregulated. As the phagosome is generally considered to be poor in nutrients, this metabolic adaptation contributes to survival of the fungus in this specificenvironment. Finally, the upregulation of uptake systems for trace metals (discussed in detail below, see Fig. 3)byphagocytosedC. albicans cells likely contributes to overcome restriction of these micronutrients within host cells.
Fig3. Micronutrient acquisition by C. albicans. C. albicans imports, stores, and exports trace elements during its life-cycle. In the mammalian host, the availability of micronutrients is limited. Depicted are the transport systems for zinc and iron. Illustration by M. Kapitan.
The Cell Wall as a Dynamic Factor in Candida-Host Interactions
As the outer layer of the cell, the cell wall and associated proteins present the structures that mediate contact of C. albicans with the environment and host cells. While on the one hand providing the major outer physical structure for the cell and mediating resistance to adverse environmental conditions, the cell wall is also presenting pathogen-associated molecular patterns (PAMPs) that are recognized by PRRs on professional and associate immune cells. Thus, the cell wall is critical for host-pathogen interactions of C. albicans. The fungal cell wall consists of different b-glucans, chitin and chitosan, which are located in an inner layer, and mannans attached to proteins in an outer layer. Most of the cell wall molecules are carbohydrates, which all contribute to immune recognition. The receptors involved in recognition and the response of immune cells differs depending on the cell wall moiety that is recognized: b-1,3 glucan interacts with the C-type lectin Dectin-1, which triggers a strong inflammatory response. Mannans can be recognized by Toll like receptors (TLRs), the Mannose Receptor DC-SIGN, Mincle and Galectin-3, likewise inducing a proinflammatory response which is, in comparison to b-1,3 glucan, less pronounced. Chitin is sensed by NOD2, TLR9 and the mannose receptor, but in contrast to the other components can reduce inflammatory responses. Thus, the relative exposure of the different carbohydrates determines the quality and quantity of subsequent immune responses. This is exemplified in studies of C. albicans mutants with defects in mannosylation which lead to increased exposure of inner cell wall components, resulting in altered interaction with immune cells.
Importantly, the cell wall composition and relative exposure of the different carbohydrate components is highly dynamic and influenced by the morphological state, environmental conditions and the interaction with immune cells. Yeast and hypha differ in the types of glucan and the amount of chitin, with chitin being more abundant in hypha. Furthermore, the distribution of chitin differs; while it is found along the cell wall of hypha, it is concentrated at the bud scar in yeast. Furthermore, some mannoproteins are differentially expressed depending on the morphology, e.g., hypha-specific proteins like Hwp1 and Als3 and the mannan composition in yeast and hypha is different. As a consequence of these differences, immune cells respond differently to yeast and hypha .
The environment influences the cell wall composition and structure. Mannan structures and the levels of glucan and chitin, for example, vary depending on the culture conditions. As mentioned above, both pH and hypoxia affect the exposure of b-glucan. Lactate induces active b-glucan masking while cell wall stress induced by the antifungal echinocandins leads to increased chitin and enhanced b-glucan exposure both in vitro and during in vivo treatment. Importantly, in the absence of antifungals, unmasking of b-glucan also occurs during systemic candidiasis in mice as a result of active cell wall remodeling in response to damage by neutrophil extracellular traps. Chitin can also be upregulated during infection, affecting recognition by Dectin-1, a phenomenon that varies between different C. albicans strains. Increased chitin in C. albicans is furthermore associated with reduced virulence in a systemic mouse model, highlighting the importance of the cell wall and its dynamic modifications for the interactions C. albicans with its host.
The Influence of Macronutrients on Candida-Host Interaction
As both a commensal and a pathogen, C. albicans can thrive at different anatomical sites. These differ not only in pH, oxygen levels and host defense systems but also regarding the quality and quantity of available carbon and nitrogen sources. Its success at the different niches suggests that C. albicans is metabolically versatile and has evolved means to assimilate a range of nutrient sources. Furthermore, the dissemination into various tissues during systemic infection requires fast adaptation to local nutrient availability. Indeed, C. albicans can utilize a vast variety of different sugars and amino acids and a range of secreted hydrolytic enzymes contributes to the liberation of nutrients from complex molecules. Complex regulatory networks mediate the response of C. albicans to nutrient starvation or alterations in the available carbon or nitrogen sources, leading to altered expression of hydrolases, transporters and metabolic pathways like gluconeogenesis, glycolysis or fatty acid b-oxidation .In comparison to the related baker’s yeast Saccharomyces cerevisiae, substantial transcriptional and post-translational rewiring has occurred in C. albicans, contributing to the metabolic flexibility of C. albicans .
In addition to providing the necessary energy for fungal survival, growth, and production of effector molecules, metabolic adaptation is directly linked via transcriptional networks to the expression of different fitness and virulence attributes such as morphogenesis, stress resistance, adhesions, and hydrolytic enzymes. Cell wall composition is also influenced by nutrients, especially carbon metabolism, and thereby metabolism indirectly influences the interaction with immune cells. One well studied example is the growth of C. albicans on lactate in comparison to glucose and the changes caused by this. While most growth media used in the laboratory are rich in glucose, glucose availability is limited within the host. Thus, C. albicans has to use alternative carbon sources such as lactate. Growth on lactate in vitro led to alterations in cell wall architecture and secretome, with consequences for adhesion, stress resistance, and interaction with immune cells: Lactate grown cells were phagocytosed less by macrophages while fungal escape was enhanced; at the same time lactate-grownC. albicans cells induced more IL-10 but less IL-17, thereby leading to a reduced proinflammatory response. As a consequence, fitness of lactate-grown C. albicans cells in vivo was enhanced.
Due to the link between metabolism and fitness/virulence traits, the adaptation of C. albicans to different environments and nutrient availability within the host simultaneously affects expression of virulence factors. This linkage of metabolic responses to environmental clues and virulence can be interpreted as “adaptive prediction”, by which the availability of distinct nutrients serves as a clue for further subsequent environmental changes, e.g., attack by immune cells following translocation into the blood stream, leading to oxidative stress due to ROS production. From an evolutionary perspective, this linkage is likely to be highly advantageous to the fungus and likely contributes to its success in diverse anatomical niches both as a commensal and a pathogen.
Acquisition of Micronutrients by C. albicans
Micronutrients are as essential for microbial growth as carbon- and nitrogen sources. Trace metals such as iron, zinc, copper, and manganese are essential co-factors in enzymes and stabilize three-dimensional protein structures. However, their free availability in the host is limited by sequestration and binding to host molecules. Host mediated nutrient depletion from invading pathogens is also termed ‘nutritional immunity,’ a mechanism that has gained increased attention in recent years, and is therefore described in some detail below.
Every aspect of nutritional immunity, from the host or pathogen perspective, reflects the co-evolutionary nature of these interactions. Best studied is the competition for iron, but most concepts apply to all transition metals in eukaryotes. In the host, iron is mainly kept intracellular– free iron is practically unavailable in blood or other extracellular liquids, with exception of the gut. Iron is bound to transferrin in plasma and lactoferrin in external secretions. However, in order to grow within the host, microbial pathogens cope with these limitations. In case of direct physical contact between the host iron source and the pathogen, host iron-retaining proteins might be manipulated to release their iron load into the extracellular environment for uptake. Secondly, many bacteria and fungi secrete siderophores: Small molecules with exceptionally high iron affinity that free iron from host proteins followed by resorption into the pathogen cell. In response, mammals produce lipocalin-2, to intercept siderophores. This can in turn be counteracted by many pathogens which disguise their siderophores by glycosylation. Iron is a common co-factor in enzymes and one of the most important direct functions of iron during host-pathogen interactions is likely the detoxification of H2O2 by the iron-dependent heme-enzyme catalase. Iron uptake in C. albicans occurs by a reductive and the non-reductive mechanism (Fig. 3). In a nutshell, the high affinity reductive uptake system functions via the interplay of three enzymes. A reductase (Fre) converts ferric Fe3+ to ferrous Fe2+. The actual uptake is mediated by a permease (Ftr) and a multicopper ferroxidase (Fet). C. albicans relies on this pathway to acquire iron from ferritin and transferrin. Probably due to functional redundancy due to gene expansion of the reductases and multicopper oxidases, only the permease Ftr1 has been shown to be crucial in blood stream infections by C. albicans. Non reductive Iron uptake is based on the import of iron-containing proteins, namely, hemoglobin/heme and siderophores. C. albicans has receptors for heme (e.g., Rbt5p), but also for ferritin (reductive uptake). The ferritin receptor and adhesin Als3 is expressed only by hyphae. Siderophores are not produced by C. albicans, but they can be utilized as an iron source even if produced by other microbes.
In contrast to other sites in the human body, the gut is probably the only site where microbes encounter high iron levels. Accordingly, as a colonizer of this niche, C. albicans is able to adapt not only to iron limitation but also to relatively high iron concentrations. To do so, an intricate system of three transcriptional regulators has evolved that controls iron uptake in C. albicans. The transcriptional repressor Sfu1 and the virulence regulator Hap43 dampen iron uptake and Sfu1 is required for persistence of C. albicans in the gut. In contrast, during blood stream infections, when iron is limited, iron uptake is upregulated by Sef1, a process essential for virulence in the systemic mouse model. Sfu1 represses Sef1, thereby interconnecting the response to low and high iron and adaptation to distinct niches in the host.
Uptake of iron and copper are very tightly connected. The ferroxidase (Fet) is actually a copper-dependent enzyme, and the iron reductases (Fre) are indeed ferric/cupric reductases that reduce Cu2+ to Cu1+ through the oxidation of Fe2+ to Fe3+, and vice-a versa . Reduced Cu1+ is taken up by Cu transporters (Ctr). Ctr1 and Ctr3 are plasma membrane-localized, Ctr2 is vacuolar– and all three transport towards the cytosol. Interestingly, the number of CTR copies varies between fungal species and might affect virulence due to an effect on, e.g., hyphal growth. During infection, pathogens face copper starvation as well as copper intoxication. For example, excess copper has been described as a mode of action against intraphagosomal Mycobacteria in macrophages. Therefore, copper export and copper detoxifying metallothioneins are important to maintain copper homeostasis. Metallothioneins are conserved in prokaryotes and eukaryotes. A potential role in virulence was demonstrated in Cryptococcus neoformans, a facultative intracellular fungal pathogen, but not yet in C. albicans. However, C. albicans is known to export excess copper, indicating that fungi are primed to resist potential copper poisoning in vivo. At the same time, copper is essential for pathogenic survival. Most fungal SODs, crucial to withstand superoxide stress, have copper in their active site. In C. albicans, Sod1, Sod4, and Sod6 are copper/zinc-dependent, Sod5 was recently described to be a Cu-only SOD. C. albicans extracellular Sod4 and Sod5 were demonstrated to detoxify superoxides produced during interactions with macrophages.
Of all metal-containing enzymes, 9% use zinc– so zinc is the second most abundant metal cofactor after manganese. Further, zinc stabilizes zinc-finger domains, a function that is especially important in eukaryotic organisms. Several studies have demonstrated the importance of zinc in pathogenic and non-pathogenic fungi. C. albicans is very sensitive to zinc starvation. Zinc can actually be the limiting factor if all other nutrients are available. Zinc acquisition in fungal pathogens is mediated by a small number of proteins and is conserved in C. albicans (Fig. 3): Zinc enters the cytoplasm via the Zrt plasma membrane transporters (CaZrt1, CaZrt2). From there, zinc can be transported further into the vacuole through ZnT transporters (in C. albicans: Zrc1). In return, zinc can be mobilized from the vacuolar pool by Zrt exporters (CaZrt3). Interestingly, no cellular zinc exporter has been described in fungi, yet. The homeostasis of zinc is regulated on two levels: the expression of Zrt importers, controlled by Zap1, and zinc import and export from subcellular organelles, resulting in biphasic zinc compartmentalization which is essential for virulence. Vacuolar storage might be especially important during infection, since it allows proliferation even under zinc-restricting conditions. In addition, C. albicans uses a zincophore system– a secreted protein that sequesters zinc from host cells and re-associates with CaZrt1 to deliver zinc to the fungal cell: Pra1. The expression of Pra1 is repressed under acidic conditions, but induced under zinc starvation or at neutral/alkaline pH. Remarkably, Pra1 is not only beneficial for the fungus, it is actually immunogenic and immune-modulatory. Amongst many functions, Pra1 can serve as a ligand to neutrophil integrin receptor aMb2 and promote neutrophil killing. Probably due to this “side effect,” Pra1 is in not conserved in all pathogenic fungi. Zinc is furthermore important for biofilm formation, and involved in tolerance to intrinsic oxidative stress by the cytoplasmic Sod1. Sod1 is a Cu/zinc superoxide dismutase (SOD). Notably, the cell surface superoxide dismutases Sod4p and Sod6p that are crucial to survive host-derived ROS, have Cu/zinc as their active site.
In non-plant eukaryotes, manganese function is restricted to mitochondrial enzymes, including Sod2 and Sod3. The human host restricts access to manganese by releasing calprotectin that is capable of binding manganese and zinc. Furthermore, manganese (and iron) are actively depleted from the macrophage phagosome. Manganese restriction is important during bacterial infections, but its role during fungal pathogenesis requires more investigation.
