Candida auris: What Have We Learned About Its Mechanisms of Pathogenicity?

Candida auris has emerged globally as a multidrug-resistant (MDR) medical care-associated fungal pathogen. Recent reports have demonstrated that C. auris usually expresses fewer virulence factors than does Candida albicans. However, the tendency of C. auris transmission within and between healthcare facilities is unique among Candida spp. and is possibly promoted by virulence and pathogenicity factors that facilitate skin colonization and environmental persistence. To understand the ability of this yeast to cause disease, we herein discuss several virulence and pathogenicity aspects of C. auris.

Introduction

Candida auris is an emerging multidrug-resistant (MDR) yeast pathogen that causes healthcare-associated invasive infections. There is growing evidence suggesting that C. auris may persistently colonize the hospital environment and multiple body sites of patients, leading to high transmissibility, and prolonged outbreaks (Schelenz et al., 2016; Vallabhaneni et al., 2017). Indeed, we are encountering for the first time an MDR yeast pathogen with a high capacity to cause clusters of invasive infections in medical centers around the world (Kim et al., 2009; Lee et al., 2011; Chowdhary et al., 2013, 2014; Calvo et al., 2016; Prakash et al., 2016; Lockhart et al., 2017; Morales-Lopez et al., 2017). C. auris fungemia results in crude mortality rates ranging from 32 to 66%, depending on the patient’s underlying conditions, geographic region, and age as well as the clinical management of the infection (Lee et al., 2011; Sarma et al., 2013; Chowdhary et al., 2014; Calvo et al., 2016; Schelenz et al., 2016; Morales-Lopez et al., 2017; Vallabhaneni et al., 2017).The description of C. auris fungemia outbreaks worldwide illustrates the dramatically increased need to develop new strategies to prevent and control infections due to this multiresistant pathogen. Despite the large number of papers addressing C. auris infections published since 2009, few have attempted to investigate the pathogenicity and virulence of this new human pathogen. Here, we review all papers addressing biological factors that affect the pathogenicity of C. auris.

What We Have Learned From Comparative Genomics of C. auris?

C. auris was first described after it was isolated from the ear canal of a 70-year-old Japanese woman at Tokyo Metropolitan Geriatric Hospital in Japan (Satoh et al., 2009). After this report, a retrospective review of Candida strain collections revealed that the earliest known strain of C. auris dates back to 1996 in South Korea (Kim et al., 2009). However, it is unknown why C. auris has only recently emerged in a wide variety of locations. Based on whole genome sequencing (WGS), four different clades of C. auris have been described by region (East Asian, South Asian, African, and South American). Moreover, analyses of SNPs identified by WGS and multilocus sequence typing (MLST) have shown low genetic diversity between isolates within each C. auris clade. These results indicate a high degree of clonality within clades and suggest the independent, nearly simultaneous emergence of the four populations on three continents (Lockhart et al., 2017).

The complete genome of C. auris was only recently investigated (Chatterjee et al., 2015; Sharma et al., 2015, 2016), and we are far from fully understanding of the role of different genes in the pathogenicity and virulence of this emerging pathogen. The main problem is that the C. auris genome sequence contains many uncharacterized and hypothetical proteins, and it is unclear whether these proteins are involved in species-specific characteristics that promote its aggressiveness as a pathogen (Chatterjee et al., 2015). C. auris is closely related to three other rare Candida species – Candida haemulonii, Candida duobushaemulonii, and Candida pseudohaemulonii – in the family Metschnikowiaceae, along with more common Candida lusitaniae (Cendejas-Bueno et al., 2012).

Initial analyses of the C. auris genome suggest that 40% of C. auris proteins are orthologous to those of C. lusitaniae (Sharma et al., 2016). Considering the limited susceptibility of C. lusitaniae to amphotericin B and C. haemulonii to amphotericin B and azoles (Kim et al., 2009; Espinel-Ingroff et al., 2017), this orthology suggests that C. auris shares some genetic characteristics with both species that may have helped it become resistant to multiple antifungal drugs. Indeed, a significant portion of the C. auris genome encodes transporter genes and protein kinases, such as the ABC and MFS transporter families, and a number of zinc cluster transcription factor orthologs such as TAC1 (29% similarity with Candida albicans), which may facilitate acquisition of drug resistance (Sharma et al., 2016).

By comparing the C. auris genome with that of C. albicans that is well annotated and well-studied we may suggest that C. auris genome harbors genes that are well-characterized as virulence factors in other Candida species, including genes for biofilm formation, proteinases, lipases, phospholipases, adhesins, secreted aspartyl proteases and transporters belonging to the ABC and major facilitator superfamily (MDR transcription factors) which are involved with azole resistance (Chatterjee et al., 2015).

Phenotypic Traits and Implications for Pathogenicity

Phenotypic switching and morphogenesis are well-known virulence attributes of C. albicans (Polke et al., 2015). Although micromorphological studies of C. auris colonies suggest that this pathogen does not produce germ tubes, pseudohyphae or chlamydospores (Lee et al., 2011; Chowdhary et al., 2014; Borman et al., 2016; Wang et al., 2018), this yeast may present pseudohyphae-like forms under high-salt stress (Wang et al., 2018) and, occasionally, in the biofilm community (Sherry et al., 2017).Pseudohyphae-like forms are characterized by rudimentary growth, with an elongated shape and incomplete cell division (Sherry et al., 2017; Wang et al., 2018). Indeed, comparative genome analyses have confirmed that C. auris and closely related species do not have two genes, candidalysin (ECE1) and hyphal cell wall protein (HWP1), both of which are highly expressed in C. albicans and the transcription of both is strongly associated with hyphal formation (Munoz et al., 2018).

In this scenario, C. auris fails to form chlamydospores after growth on cornmeal agar when incubated for 3 days at 30°C and does not germinate when incubated with fetal bovine serum (FBS). This fact was confirmed in the Galleria mellonella model of infection of C. auris, during which the isolates did not undergo significant filamentation at 18 h or at any time post infection (Borman et al., 2016).

Growth curve analyses for C. auris and C. albicans isolates demonstrate that the two Candida species have similar growth patterns, reaching the stationary phase within approximately 20 h (Larkin et al., 2017). An interesting observation regarding the growth of C. auris is that certain isolates grow in clumps (i.e., budding occurs, but daughter cells are not released), which results in large aggregates of organisms that cannot be easily disrupted in vitro (Borman et al., 2016). These isolates are called “aggregate” strains, and certain aggregate strains are too large for G. mellonella larval inoculation, hindering accurate yeast cell counting. To overcome this limitation, homogeneous inoculum preparations have been obtained after allowing the initial suspension of fungal cells to settle for 10 min followed by removal of the supernatant containing individual yeast cells for infecting larvae (Borman et al., 2016). Based on this protocol, it was possible to compare the pathogenicity of aggregating and non-aggregating strains of C. auris with other yeast species. The results demonstrated strain-specific variability in behavior, with the aggregate-forming isolates of C. auris exhibiting significantly reduced pathogenicity compared to their non-aggregating counterparts (Borman et al., 2016). Moreover, when infected G. mellonella larvae were dissected, large numbers of individual budding yeast cells of non-aggregating strains of C. auris were found inside phagocytes. In addition, the larvae inoculated with individual yeast cells prepared from aggregate-forming strains of C. auris exhibited large aggregates of C. auris cells, with few individual yeast cells, indicating that the ability to produce large aggregates had been maintained in vivo (Borman et al., 2016). Yeast cell aggregates were also observed in the kidneys of mice infected with C. auris, suggesting that aggregation might be a mode of immune evasion and persistence in tissues, which warrants further investigation (Ben-Ami et al., 2017).

Tolerance to Thermic and Osmotic Stresses

Survival and growth at physiologic temperature are prerequisites for microbial invasion and pathogenicity. C. auris exhibits thermotolerance, growing optimally at 37°C and maintaining viability up to 42°C. In addition, this pathogen is salt tolerant, and cells aggregate into large, difficult-to-disperse clusters, which may promote persistence in the hospital environment (Satoh et al., 2009; Borman et al., 2016).

The ability of C. auris isolates to grow at 37°C and 40°C appears to be similar to that of C. albicans, and certain isolates also grow at 42°C (Ben-Ami et al., 2017). Other authors have confirmed that C. auris can grow at high temperature (40°C) and salinity (10%wt/vol) when cultured in Sabouraud (SAB) or yeast nitrogen base (YNB) broth with dulcitol or mannitol as the carbon source (Welsh et al., 2017).

Wang et al. (2018) examined the morphology of C. auris on the rich media YPD and YPD plus 10% NaCl, with a round appearance on the former medium but an elongated shape on the latter. Interestingly, a small portion of highly elongated and pseudohyphae-like forms were observed when grown on YPD plus 10% NaCl (Wang et al., 2018). These results suggest that high-salt stress may result in incomplete cell division, leading to the formation of pseudohyphae-like forms. So far, the molecular mechanism related to this phenotype of C. auris has not yet been investigated.

Importance of Lytic Enzymes for Fungal Invasiveness

The production of phospholipases and proteinases has been recognized as relevant for Candida pathogenicity in the human host, helping in the adherence to and invasion of host cells (Polke et al., 2015). Hydrolases are the largest group of enzymes (42%) found in the C. auris (strain 6684) genome, followed by transferases (25%) and oxidoreductases (19%) (Chatterjee et al., 2015). Furthermore, comparative genome analyses have revealed similar numbers of lipases in the C. auris relative to those of C. albicans and C. dubliniensis (Munoz et al., 2018). The ability to produce lytic enzymes has been demonstrated in C. auris isolates, and the production of these enzymes is strain dependent (Kumar et al., 2015; Larkin et al., 2017).

An in vitro study evaluating C. auris isolates from different geographical regions showed that 37.5% of the tested strains (6/16 isolates) exhibited phospholipase activity and that 64% (9/14 isolates) were positive for proteinase (Larkin et al., 2017). Similarly, most strains displayed hemolysin activity, conferring a high capacity for iron acquisition, growth, and invasiveness leading to widespread infection (Tsang et al., 2007; Kumar et al., 2015).

In fact, a recent study conducted by Wang et al. (2018) demonstrated that the level of aspartyl proteinase (Saps) secreted by C. auris isolate at 42°C was higher than that exhibited by C. albicans at the same temperature (Wang et al., 2018). These findings suggest that C. auris isolates are not only well adapted to temperature stress but also maintain their pathogenicity at higher temperatures.

C. auris and Its Alarming Ability to Persistently Colonize Human Host and the Environment

Studies have demonstrated the ability of C. auris to colonize and spread throughout hospital environments. One of the most alarming characteristics of C. auris is the ability of this yeast to adhere to and persist on abiotic surfaces, including dry and moist surfaces, bedding material, floors, sinks and beds, as well as human skin, ears, and nasal cavities (Schelenz et al., 2016; Piedrahita et al., 2017; Vallabhaneni et al., 2017; Welsh et al., 2017). In one study, the survival of eight C. auris isolates was compared with other Candida species on dry or moist surfaces for up to 7 days, with the recovery of C. auris being similar to that of other Candida species, vancomycin-resistant Enterococcus (VRE), methicillin-resistant Staphylococcus aureus (MRSA), and carbapenem-resistant Enterobacteriaceae (CRE) (Piedrahita et al., 2017).

Moreover, the capacity of C. auris to persist and colonize the plastic surfaces of medical devices was compared with that of Candida parapsilosis (Welsh et al., 2017) using two complementary independent methods: culture [as measured by colony-forming units (CFU)] and solid-phase cytometry (which detects viability, as measured by esterase activity) (Smith et al., 2010). For CFU determination, C. auris and C. parapsilosis suspensions of 5 × 106 cells per mL were applied to dried plastic surfaces and followed to grow for up to 28 days. C. auris and C. parapsilosis were viable for at least 14 and 28 days, respectively; however, viable C. auris was detected for longer periods than previously observed by quantitative culturing. When solid-phase cytometry was used to assess individual cells for viability for longer periods of time, the results indicated that the cells entered into a viable non-culturable state, and more viable C. auris than C. parapsilosis cells were detected at all time points (Welsh et al., 2017).Overall, the adherence of C. auris to medical devices may play a role in the development of catheter-related fungemia. In contrast to C. albicans, C. auris exhibits minimal ability to adhere to silicone elastomer (Larkin et al., 2017).

Biofilm Production and Impact on Antifungal Resistance

Biofilms are a common form of microbial growth that are crucial for the development of a broad spectrum of infections in the human host and also for defending pathogens from phagocytes and antimicrobial drugs (Fanning and Mitchell, 2012). Seven highly conserved genes (PLB3, IFF4, PGA52, PGA26, CSA1, HYR3, and PGA7) are upregulated during biofilm production across isolates representative of C. auris, C. haemulonii, C. duobushaemulonii, and C. pseudohaemulonii. The same proteins are associated with biofilm production and mechanisms of antifungal resistance in C. albicans strains (Kean et al., 2018). As documented with other Candida species (Hirakawa et al., 2015), there is a large intra-specific variation in the capacity of biofilm production by C. auris. Initial investigation with fifteen C. auris isolates cultured from ear specimens failed to detect any biofilm production (Oh et al., 2011). Later, Larkin et al. (2017) performed a comparative study to assess biofilm production by two isolates of C. auris and one isolate of C. albicans using silicone elastomer as a substrate. Quantification of the biofilms was performed using a colorimetric metabolic assay (to measure mitochondrial dehydrogenase activity-XTT) and dry weight analysis (to measure total biofilm mass). During this assay, the authors were able to detect C. auris biofilm production that was composed of yeast cells adhering to the catheter material. In contrast, the C. albicans biofilm presented a highly heterogeneous architecture composed of yeast cells and hyphae embedded within the extracellular matrix. Moreover, the C. auris biofilms, unlike the C. albicans biofilms, displayed a limited amount of extracellular matrix (Larkin et al., 2017).

Recently, a comparative study was performed to evaluate the production of biofilm by isolates of C. albicans, Candida glabrata, two non-aggregating strains of C. auris and two aggregating strains of C. auris. The authors used 96-well polystyrene microtiter plates and measured total fungal biomass using a crystal violet assay. The results showed the greatest biofilm mass for C. albicans, followed by C. auris and C. glabrata. The C. auris biofilm was predominately composed of budding yeasts and occasional pseudohyphae (Sherry et al., 2017). In the same study, the susceptibility of C. auris biofilms to drugs was assessed, with micafungin and caspofungin being ineffective against biofilms “in vitro” and requiring > 32 mg/L to inhibit sessile cells (Sherry et al., 2017).

Virulence of C. auris in Animal Models

A comparative study of virulence exhibited by C. auris and C. haemulonii isolates was performed in a mouse model of hematogenous-disseminated candidiasis after immunosuppression with cyclophosphamide (150 mg/kg intraperitoneally). The C. haemulonii isolates were found to be completely non-virulent, with 100% of the mice surviving at 12 days after inoculation and no visible signs of illness. In contrast, inoculation with C. auris resulted in rapid death, with only 20% survival at 5 days after infection (Munoz et al., 2018). Another comparative study of Candida spp. virulence using an immunocompetent murine model of disseminated infection showed high virulence of C. albicans isolates, followed by C. auris, C. glabrata, and C. haemulonii (Fakhim et al., 2018). In this study, the animals were challenged with 105 CFU/mouse injected into the lateral tail vein, and death rates were recorded up to 30 days post-infection. Mice infected with C. albicans exhibited 20% survival, with a median survival time (MST) of 13 days, where as mice infected with C. auris showed 30–40% survival until the end of the experiment, with an MST of 16–17 days post-infection (Fakhim et al., 2018).

The virulence of a single C. auris isolate obtained from a Chinese fungemic patient was recently evaluated in a mouse model of Candida systemic infection. Inoculum preparations of 1 × 106CFU/ml of C. albicans (SC5314) and 1 × 107CFU/ml C. auris were injected via the tail vein, with all animals infected with C. albicans dying by the 6th day post-infection but no animal infected with C. auris dying during the same period (Wang et al., 2018).

Due to economic and especially ethical issues, the scientific community has in the last decade limited the use of mammalian infection models to study the virulome of fungi (Krappmann, 2015). Driven by the need for validating alternative models to replace mammalian systems, invertebrate organisms, such as G. mellonella and C. elegans, have been extensively used as mini-hosts for studying the virulence attributes and host response of Candida infections (Cotter et al., 2000).

Within this context, Borman et al. (2016) developed a model of G. mellonella fungal infection using an inoculum solution of 1 × 106 yeast cells of C. auris, C. albicans and Candida tropicalis isolates, whereby it was possible to demonstrate that C. albicans and C. tropicalis exhibit significantly higher virulence in terms of the kinetics of larval death and number of larvae killed (Borman et al., 2016). In accordance with other publications, these researchers found that C. auris isolates were not able to develop significant pseudohyphae into tissues of infected larvae (Lee et al., 2011; Chowdhary et al., 2013; Kathuria et al., 2015; Borman et al., 2016). Wang et al. (2018) also confirmed that the Chinese C. auris strain exhibits reduced virulence in G. mellonella larva compared to a C. albicans isolate. Overall, compared to C. glabrata, C. auris strains are usually more virulent in animal models.

Investigation of the immune response to C. auris using the Zebrafish model of invasive candidiasis revealed a recruitment of approximately 50% less neutrophils in response to C. auris infection when compared to C. albicans (Johnson et al., 2018). Moreover, in vitro human neutrophils were co-cultured with Candida cells (2 × 107 cells) for 4 h and fungal viability was measured. Neutrophils inhibited C. albicans growth by 75%; in contrast, the burden of C. auris was not impacted replicating beyond the initial inoculums, showing very little killing of C. auris by neutrophils. Using fluorescence microscopy, they analyzed neutrophil-Candida interactions, and at 1 h, very few neutrophils (15%) were either engulfing or adherent to co-incubated C. auris cells. In contrast, 50% of neutrophils exhibited some activity against C. albicans. These findings suggest that neutrophils are more able to engage and kill C. albicans over C. auris isolates (Johnson et al., 2018). Viability of neutrophils co-cultured with C. albicans decreased by half, while upon exposure to C. auris, almost all neutrophils remained viable. The production of neutrophils extracellular traps (NETs) in response to C. auris exposition was measured by scanning electron microscopy. It was found that neutrophils engaging in phagocytosis or releasing NETs were rarely observed after exposition to C. auris. Furthermore, quantifying NET formation showed that C. auris did not trigger free DNA release, consistent with the lack of NET production observed in microscopy experiments (Johnson et al., 2018).

Taking together data provided by experimental models using mice and invertebrate animals, we may conclude that C. auris is certainly less virulent than is C. albicans. Nonetheless, C. auris is significantly more virulent than are C. glabrata and C. haemulonii (Fakhim et al., 2018) and other species of Candida also considered to be resistant to multiple drugs. Indeed, C. auris was able to induce systemic infection and mortality rates higher than 50% in 2 of 3 studies investigating its virulence in a mouse model (Ben-Ami et al., 2017; Fakhim et al., 2018; Wang et al., 2018). Despite limited virulence when compared to C. albicans, in vitro assays conducted with human neutrophils showed that C. auris is less effective than C. albicans in triggering neutrophils engulfment and NET production.

Final Comments and Future Directions

Although animal studies indicate that C. auris has reduced pathogenicity and virulence compared to C. albicans, this emerging pathogen appears to be far more able to induce systemic infection and mortality than other potential MDR yeast pathogens, such as C. glabrata and C. haemulonii (Fakhim et al., 2018). This finding is likely to be related to the tolerance of C. auris strains to osmotic and high-temperature stress as well as to its ability to produce several lytic enzymes and biofilm (Ben-Ami et al., 2017; Welsh et al., 2017).

The complete genome sequence of C. auris is available (Chatterjee et al., 2015; Sharma et al., 2015; Lockhart et al., 2017), and it is currently possible to design experiments to better characterize the molecular mechanisms responsible for the capability of this pathogen to readily become resistant to multiple antifungal drugs and to determine the presence of genes related to pathogenicity and virulence factors. An important limitation of virulence analyses based on clonal strains cultured from patients during outbreaks is that it remains unclear whether such findings may be safely extrapolated to all isolates of the species or whether they are only representative of biological properties exhibited by a few clonal strains.

An additional point that deserves more attention is the characterization of adhesins and other molecules responsible for the capability of C. auris to persistently colonize abiotic and biotic surfaces. Indeed, this pathogen is able to survive and persist under different environmental conditions, including on dry materials, bedding material, floors, sinks and beds, and it exhibits tolerance to temperature and osmotic stresses (Ben-Ami et al., 2017; Welsh et al., 2017).

In summary, our review shows that C. auris expresses several virulence traits including genes that are well-characterized in other Candida species, including genes for biofilm formation, proteinases, lipases, phospholipases, adhesins, secreted aspartyl proteases, and transporters which are involved with azole resistance. Despite exhibiting less virulence than C. albicans isolates, recent findings suggest that C. auris fails to activate the innate immune response and production of NETs by human neutrophils, what certainly may play a role in the high mortality associated to this infection.

Author Contributions

LR and AC contributed equally in the paper. Both wrote the manuscript and performed all the necessary literature searches and data compilation.

Conflict of Interest Statement

AC received educational grants from Pfizer, Gilead Sciences – United Medical (Brazil), MSD, and research grants from Astellas and Pfizer. The remaining author declares that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.