Candida species are commensal organisms that are naturally part of the human microbiome. However, several factors could trigger some cells to invade mucosal surfaces and cause candidiasis [2]. Indeed, some Candida species develop effective immune evasion strategies to permit colonization of an immunologically competent host.

Interestingly, this study showed that an alternative carbon source like lactate, as well as low oxygen environment and sublethal concentrations of antifungals, trigger reduction of exposure levels of β-glucan in C. auris belonging to Clade I. Previous reports already revealed that carbon sources from the host cells affect host-pathogen interactions in C. albicans [27] but to date, this is the first study that reported lactate-induced decrease in β-glucan in C. auris.

The pathway controlling lactate-induced β-glucan masking in C. auris has not been fully elucidated. However, generally, β-glucan masking pathways differ between C. auris and C. albicans [28, 29]. Genes coding for proteins involved in β-glucan masking in C. albicans [14, 30, 31] have orthologs in C. auris based on the Candida Genome database (http://www.candidagenome.org). C. auris has orthologs of the following genes in C. albicans: GPR1 (B9J08_002066), CRZ1 (B9J08_002096), ENG1 (B9J08_002109), and XOG1 (B9J08_003251). The systematic name of the gene in C. auris is indicated in parenthesis.

The effect of oxygen level on β-glucan production on the surface of Candida sp. was first described by Pradhan et al. [13]. Hypoxia is an important condition observed in the lower gastrointestinal tract and inflamed tissues [32]. Previous studies showed that several pathogenic Candida species show varying adaptations to hypoxic environments and that most strains of C. albicans, C. krusei, and C. tropicalis showed masking of β-glucan whereas C. auris, as well as C. parapsilosis and C. glabrata, showed variable results. In contrast, most strains of C. guilliermondii do not exhibit β-glucan masking [13].

Earlier studies determined how Candida cell wall architecture is influenced by environmental pH [12, 23, 24]. To mimic the pH conditions in the gastrointestinal tract, fungal cells in this study were grown in YPD with the following pH: 4, 5.5, 7, and 8.5. These pHs represent the upper and lower stomach which is around pH 4, the small intestine which is about pH 7.0, and the colon which is nearly pH 8.5 [33]. Furthermore, relatively low ambient pH is associated with vulvovaginal niches, where certain Candida species are known to thrive. Contrary to our observation, a clinical strain of C. auris (JCH15448-1) did not show any pH- and time-dependent β-glucan unmasking [24]. We recommend checking for chitin exposure levels across a wide range of pH, as unmasking of cell wall in C. albicans may induce non-protective hyperactivation of the immune system during growth in acidic environment [23].

For the effect of antifungals, this study is the first to report the reduction of β-glucan due to fluconazole. However, a concentration of 16 µg/mL fluconazole, which is shown to trigger a reduction in β-glucan, might not be present in actual patients of candidiasis. Fluconazole levels in plasma and human tissue and body fluids are approximately below this level [34]. For the effect of echinocandins on C. auris β-glucan, our observation is supported by recent findings of a Clade 1 C. auris isolate (B11219) exhibiting reduced surface β-glucan following micafungin treatment [28].

The growth of C. auris strains used in this study is the same as the control under all environmental conditions tested except during low oxygen. MFI values of samples grown in each condition is summarized in a supplemental table (Table S1). Overall, C. auris strain used in this study (UI001) exhibits a similar β-glucan expression profile to C. albicans SC5314 [12,13,14, 23, 24]. C. auris belonging to Clade I exhibits a highly similar cell wall composition and organization to C. albicans SC5314 [16] which could partly explain the similar environment-triggered β-glucan exposures of these strains.

In line with previous reports, alteration of β-glucan levels occurs due to cell wall remodeling. To date, little to no studies regarding the dynamics between β-glucan and mannan exposures have been performed on C. auris. Mannan is found on the outermost part of the cell and could possibly affect exposure levels of β-glucan [35], thus affecting the recognition of fungal pathogens by the host immune system. It is interesting to find out if thickening of the mannan is the cause of the reduction in β-glucan in C. auris. Even more so when C. auris-induced late innate immune activation is revealed to be elicited primarily by structurally unique C. auris mannoproteins [36]. We have observed that thickening of mannan is not the cause of the reduction of β-glucan exposure in C. auris. Several studies have hinted at this [14, 31] and so, we hypothesize that although there is no direct correlation between β-glucan masking and mannan levels in C. auris, there is a possibility that the change of mannan structure may lead to less accessibility of β-glucan. To date, there are no in-depth studies that determined the cell surface expression of mannan in C. auris, specifically by concanavalin A-TRITC staining and flow cytometry analysis.

This study showed that different strains of C. auris have different cell wall responses to varying environmental conditions (Fig. S1 and Fig. S2). Other studies have also observed differences between C. auris strains [28, 36, 37]. We believe that mechanisms for environment-triggered cell wall remodelling may be different between strains regardless if they belong to the same clade. To date, there is no direct explanation about this phenomenon and thus serves as avenue for future research.

The effect of β-glucan alteration on the host immune response was determined. All Candida species display PAMPs on the cell surface but the immune reactivity of fungal cell surfaces is not correlated with the relatedness of fungal species [38]. This means that past findings on host-pathogen interaction studies mainly performed using C. albicans, may not be true for other species. So, current knowledge about C. albicans cell wall and its interaction with innate immune cells cannot be extrapolated to C. auris [16]. Indeed, an earlier study demonstrated that neutrophils preferentially target C. albicans in mixed cultures with C. auris, and C. auris evaded neutrophil capture via neutrophil extracellular trap formation [29, 39, 40]. Unlike neutrophils, macrophages can recognize and phagocytose C. auris [39] and are considered the main immune cell populations responsible for host defense against systemic candidiasis [11]. A parameter used to evaluate the host immune response is cytokine production upon infection since immunomodulation with cytokines can enhance the antifungal activity of immune cells and upregulate protective T-helper type 1 adaptive immune responses [41]. This study used human monocytic cell line THP-1 and murine macrophage cell line RAW 264.7. THP-1 monocytes differentiated into macrophages through the addition of PMA can mimic the response of human primary macrophages [42]. On the other hand, RAW 264.7 cells are a commonly used model of mouse macrophages for the study of cellular responses to microbes and their products.

In vitro infection assays were performed to determine the effect of β-glucan alteration on the rate of phagocytosis of C. auris. We focused on the effect of lactate-induced decrease in β-glucan since lactate (1%) triggers the most reduction in β-glucan among all conditions tested (Fig. 1A). To have a full picture of the effect of PAMP exposure levels on the immune response, chitin levels can also be determined since reduced cell wall chitin is shown to affect late-phase cytokine response in another species, C. albicans [43]. We have observed low production of cytokines by THP-1 macrophages infected with C. auris (Fig. 6). In a similar study, peripheral blood mononuclear cells fail to induce a potent pro-inflammatory cytokine response against C. auris [39]. Furthermore, another study showed that human M1-activated monocyte-derived macrophages stimulated less CCL3/MIP-1⍺ and less TNF-α after infection with lactate-grown C. albicans compared to glucose-only-grown cells [14]. In this study, macrophages exposed to C. auris cells exhibiting lactate-induced reduction in β-glucan produced less of CCL3/MIP-1⍺ but not TNF-α and IL-10. CCL3/MIP-1⍺ is a chemotactic cytokine produced by cells during infection or inflammation. Compared to TNF-α and IL-10, CCL3/MIP-1⍺ plays an important role in recruiting various cells such as monocytes, macrophages, lymphocytes, and eosinophils via the CCR1 or CCR5 receptor [44], thereby strengthening the immune response against a fungal infection.

Aside from an in vitro infection assay, an in vivo approach was conducted to determine the effect of lactate-induced β-glucan change in C. auris on a host. Due to some limitations in institutional facilities, an in vivo work on mice was not performed to observe the pathogenesis of C. auris and recruitment of immune cells toward the infection site. Several infection studies in which Candida sp. was injected into invertebrate hosts such as silkworm [18, 45, 46] have been carried out. Silkworms allow in vivo examinations of phagocytic cell function [46] and investigation of molecular mechanisms of infection by human fungal pathogens [18]. The larvae can thrive at mammalian physiologic temperature (37 °C), which allows for the expression of relevant temperature-regulated virulence factors [47]. Fungal cell wall components such as mannan and β-glucan can activate the immune system of B. mori. The β-1,3-glucan recognition protein (βGRP) on the surface of silkworm hemocyte binds to β-1,3-glucan leading to activation of the prophenoloxidase cascade resulting in the production of quinones and melanin, which then manifests as blackening of the injected larvae [48]. Figure 8 shows that lactate-induced change of β-glucan in C. auris leads to enhancement of virulence of cells, probably due to a reduction in recognition by hemocytes leading to less efficient clearing of the pathogen. Overgrowth of C. auris due to immune evasion may have caused the death of the larvae as the growth of fungi can largely contribute to the death of the silkworms. Our observation is in line with other studies where hypoxia-induced reduction in the β-glucan of C. albicans led to increased virulence in a C. elegans infection model [32]. Conversely, an increased level of PAMP exposure in low pH-grown C. albicans correlates with higher pathogenicity [24]. The infection experiment result complements the data presented in Figs. 5 and 7, where lactate-induced reduction in β-glucan in C. auris UI001 triggered a significant reduction in phagocytosis by THP-1 and RAW 264.7 macrophages. The phagocytic ability of these types of macrophages may not be the same as the immune cells of silkworms, but considering that silkworms can recognize β-1,3-glucan via βGRP found on the surface of silkworm hemocyte, we hypothesize that alteration of β-glucan lead to immune evasion, thus leading to increased virulence of C. auris.