A titanic drug resistance threat in Cryptococcus neoformans

Abstract

Increasing resistance to frontline antifungals is a growing threat to global health. In the face of high rates of relapse for patients with cryptococcal meningitis and frequent drug resistance in clinical isolates, recent insights into Cryptococcus neoformans morphogenesis and genome plasticity take on new and urgent meaning. Here we review the state of the understanding of mechanisms of drug resistance in the context of host-relevant changes in Cryptococcus morphology and cell ploidy.

The increasing magnitude of antimicrobial resistance is a cause for global concern, particularly for fungal infections with few treatment options. Antifungal drug resistance affects not only worldwide population health, but also healthcare costs and gross domestic product (GDP) [1, 2]. Development of drug resistance in the human fungal pathogen Cryptococcus neoformans, which causes an estimated 1 million symptomatic infections each year, could have a devastating impact on human health, yet only recently have studies begun to unravel the complex processes leading to antifungal drug resistance [3–6].

In 2010, a newly described morphotype, the Cryptococcus titan cell, was demonstrated to mediate pathogenesis [7–9]. Several reviews have discussed titan cell characteristics and mechanisms of formation [10, 11]. Yet a recent study shows titan cells could also play a critical role in stress adaptation and antifungal drug resistance [12]. Here we examine the available literature that links antifungal drug resistance and the titan cell morphotype and urge improved understanding of these processes in patient outcome.

How does Cryptococcus neoformans cause disease?

C. neoformans infection starts with inhalation of airborne fungal spores [13]. Within the lungs the spores can cause pulmonary infection, such as pneumonia, before disseminating to the central nervous system (CNS) to cause meningitis [14]. Cryptococcal meningitis is predominantly a disease of the immunocompromised - with organ transplant recipients, cancer chemotherapy patients, and individuals with advanced HIV at highest risk of developing cryptococcosis.

While the vast majority of C. neoformans patients present with meningitis, a number of reports have recently highlighted cryptococcal pneumonia as an under-appreciated aspect of the disease [15–19]. While reports of pulmonary disease in immunocompromised individuals may be quite rare, asymptomatic or latent C. neoformans infection in the lungs of healthy people is common [20, 21]. Serological studies show this initial pulmonary infection is acquired early in life, typically by 5 years of age [22, 23]. Most adults have a Cryptococcus granuloma in their lungs [20, 21, 24–26], and strain genotyping of C. neoformans in immigrant populations show disease in immunocompromised individuals is due to strains acquired during childhood [27–30].

The impact of these asymptomatic pulmonary C. neoformans infections is unknown. Seropositivity has been linked to increased rates of asthma [31, 32]. In addition, 20% of adult cryptococcosis patients are immunocompetent, suggesting unappreciated risk of progression to cryptococcal meningitis [11, 33]. Most intriguing, these lifelong infections could have a dramatic impact on the development of drug resistance for patients that become immunocompromised and develop cryptococcosis later in life. Antifungal drug exposure throughout an individual’s lifespan, either from environmental exposures due to agricultural antifungal use or during treatment of other fungal infections, could lead to drug resistance occurring before the individual presents with cryptococcosis. In a recent study in Uganda, Smith et al. noted increased resistance to the antifungal drug fluconazole in initial C. neoformans isolates from cryptococcal meningitis patients with advanced HIV. In the decade since antifungal drug use became widespread, over 30% of C. neoformans isolates are no longer susceptible to fluconazole [34]. A similar level of resistance was observed in Tanzanian clinical isolates, where 25% of initial cultures were resistant at time of diagnosis [35]. However, despite rising levels of resistance similar to those observed for other major fungal pathogens, there is reason to suspect that mechanisms of resistance may be distinct.

Mechanisms of antifungal drug resistance in Cryptococcus neoformans

In the absence of successful treatment, cryptococcal meningitis is lethal. The standard of care requires a short course of amphotericin B in combination with flucytosine, followed by consolidation fluconazole monotherapy. Relapse has been observed in up to 60% of meningitis patients [36–41]. In cases of relapse, fluconazole resistance is frequently observed.

Mechanisms of antifungal drug resistance in fungi can be classified in two ways: heritable and transient. Heritable fluconazole resistance in Candida albicans and Aspergillus sp. is linked to point mutations, most commonly in the drug target ERG11/CYP51 or drug transporters; ERG11 single nucleotide polymorphisms (SNPs) alter drug-target interaction or lead to an increase in Erg11 protein levels, while gene duplications lead to increased multidrug transporter gene copy number [42–49]. However, while multiple studies have looked for ERG11 point mutations in C. neoformans, these are rarely identified in patient isolates [4, 35, 50–53]. Other additional point mutations have been identified in mutator strains and other clinical isolates, but their exact contribution to drug resistance is yet to be determined [35, 51, 54–56].

In contrast, transient resistance is well documented in C. neoformans. In vitro and in vivo studies have revealed intrinsic hetero-resistance: transient resistance of a subpopulation of cells that are able to proliferate in the presence of high concentrations of fluconazole but lose their resistance once the drug stress is removed [41, 57]. Genomic analyses have shown this heteroresistance is mediated by aneuploidy of chromosomes 1, 4, 6, and 10 [12, 35, 57–59]. Interestingly, hetero-resistance can be suppressed by treating C. neoformans with fluconazole and flucytosine in combination [35].

As discussed above, during cryptococcosis, persistent antigenemia suggests that continued active infection may create a reservoir for the emergence of drug resistance during long-term therapy [22]. The emergence of this resistance was recently demonstrated in vivo in a cohort of HIV- infected Tanzanian patients [35]. Stone et al. found a fluconazole hetero-resistant subpopulation of up to 25% of colonies in certain clinical isolates that were able to grow in supra-MIC concentrations before any fluconazole treatment was given. Within 7 days of fluconazole monotherapy, 11 out 13 patients (85%) had an increase in degree of hetero-resistance leading to an increase in MIC [35]. Combinatorial treatment with fluconazole and flucytosine was seen to supress the growth of hetero-resistant cells in vitro and in vivo. Although they saw continued suppression with this combination therapy in vivo, after 24 hours the hetero-resistant colonies reappeared in vitro. Interestingly, these newly grown hetero-resistant colonies were found to have developed additional resistance to flucytosine [35].

The mechanisms driving this process remain unclear, but several possibilities have been proposed based on in vitro studies. First, Altamirano et al., demonstrated a direct impact for fluconazole on yeast cell nuclear division, consistent with findings in the ascomycete yeast Candida albicans [5, 60, 61]. The diploid C. albicans develops aneuploidy in the presence of fluconazole through production of intermediate tetraploid multimeras due to incomplete cytokinesis [61]. Similarly, C. neoformans was also found to produce multinucleate, incompletely septate cells in the presence of fluconazole [5]. Alternatively, Chang et al demonstrated that fluconazole resistance may arise from an originating cell that is uninucleate and euploid, with aneuploidy arising through mis-segregation of chromosomes [6]. Together, this suggests that fluconazole may directly (through blocking cytokinesis) or indirectly (through genome plasticity) induce drug resistance.

What is a titan cell and how does it influence antifungal drug resistance?

During the initial pulmonary infection, 20–30% of C. neoformans cells in the lungs convert into large polyploid titan cells [7, 8]. Titan cells are observed in human samples [62, 63] and in mouse models [7, 8], and recent studies have identified in vitro culture conditions [64–66]. Titan cells are defined by their high ploidy (≥4C), a cell body size larger than 10 μm in diameter, thick cell wall, and highly cross-linked capsule [7, 8, 10].

Importantly, titan cells are critical for the overall virulence of C. neoformans and dissemination to the central nervous system [9]. Detailed analyses of the role of titan cells in pathogenesis have identified at least two distinct functions. First, structural changes associated with the titan cell, such as their increased cell size and modification of the cell wall and capsule structure, alter recognition by the host immune response [7, 8, 67–73]. This altered recognition not only promotes survival of the titan cell itself through reduced phagocytosis, but also shifts the adaptive immune response to promote Cryptococcus cell survival and dissemination to the central nervous system [70].

Second, and perhaps most intriguing in the context of drug resistance, is that titan cells generate daughter cells that exhibit increased resistance to fluconazole and may be better adapted to the host environment [12]. Cryptococcus cells are typically haploid, containing a single copy of the genome. Titan cells are polyploid with 4, 8, 16, and more copies of the genome [7, 8, 12, 64, 74]. Yet titan cells undergo a genome reduction when they produce daughter cells - often generating haploid, diploid, or aneuploid daughter cells [7, 12, 64]. Gerstein et al. showed that populations of daughter cells derived from titan cells are more resistant to oxidative and nitrosative stresses similar to those used by the host immune system [12]. Gerstein et al. also showed that in vivo derived titan cells incubated in the presence of fluconazole produce populations of daughter cells that are resistant to fluconazole, and whole genome sequencing revealed these daughter cells are aneuploid [12]. Thus, the titan cell morphotype may be a mechanism for the generation of the genomic plasticity underlying hetero-resistance and aneuploidy leading to drug resistance in C. neoformans.

Other Mechanisms of Resistance

In addition to contributing to fluconazole resistance through formation of aneuploid daughter cells, titan cells exhibit unique cell surface alterations that may also play a role in the development of drug resistance. The polysaccharide capsule surrounding the C. neoformans cell, which acts both as a protective outer layer and an immune modulatory compound, shows structural differences in titan cells. While typical-sized cells have diffuse capsules, the titan cell capsule is significantly denser and more highly crosslinked [8, 72]. Additionally, titan cells have a cell wall up to 2–3 μm thick, compared to the 15–200 nm cell wall of typical-sized cells [10, 68]. Analysis of titan cell walls in comparison to typical-sized cell walls revealed a decrease in glucan, an increase in chitin, and the possibility of a novel mannan layer on the surface of the cell wall [72]. It was recently shown that the increased chitin (a long-chain polymer of N-acetylglucosamine required for fungal cell wall rigidity) in the titan cell wall stimulates a detrimental Th2-mediated immune response [70]. There have not been any studies that have directly linked titan cell surface alterations with increased drug resistance, but several studies have shown both the capsule and cell wall to be important in conferring drug resistance. A recent review emphasized the potential for in vivo physiological changes in C. neoformans, such as titan cells and capsule enlargement, to act as barriers to effective treatment due to in vitro antifungal tests overlooking these factors [75]. Elegant work from the Fries lab has shown that old C. neoformans cells are more resistant to antifungals [76, 77]. The increased drug resistance of old cells has been hypothesized to be due to the thickened cell wall of older cells, which could prevent or reduce drug penetration [76, 77]. Studies have also shown that C. neoformans strains that produce larger capsules exhibit increased resistance to amphotericin B, which may also be due to reduced drug penetration [78–80]. Interestingly, both older cells with thickened cell walls and cells with larger capsules have also been shown to be more resistant to reactive oxygen species (ROS) induced by hydrogen peroxide stress, and both fluconazole and amphotericin B induce ROS in C. neoformans [75, 77, 78, 81, 82]. Taken together, the altered cell surface structure of titan cells could facilitate survival through reduction of the inhibitory effects of ROS during treatment. However, it has yet to be determined if a thickened cell wall is important in increased ROS resistance.

Conclusion

Titan cells are an important subpopulation of C. neoformans cells that arise early in infection and undergo unique morphological changes that contribute to increased virulence. Previous studies have highlighted the critical role the cell surface alterations and enlarged size of titan cells play in evasion of the host immune system and dissemination to the brain. Additionally, it has been shown that titan cells exhibit increased ploidy levels and give rise to aneuploid daughter populations that are more stress resistant. However, the ability of titan cells to act as a source of adaptive phenotypic change to promote host adaptation and drug resistance still remains largely unexplored. Given the well-established importance of aneuploid formation in conferring fluconazole resistance in C. neoformans, the increased ploidy levels of titan cells and their ability to produce aneuploid daughter cells makes investigating the contribution of titan cells to drug resistance especially intriguing. Given that titan cells are produced in the lungs during pulmonary infection, they may play a critical role in the development of drug resistance during lifelong Cryptococcus infections. Future work examining the effects of titan cell formation on in vivo drug efficacy will help us better understand the mechanisms of drug resistance in C. neoformans and improve our current therapeutic and diagnostic approaches.

Figure 1. Titan cell characteristics likely to impact drug resistance:

(A) Cell size increase – Titan cells exhibit an increase in cell size that prevents phagocytosis and promotes survival. An increase in survival may provide additional opportunity for acquisition of drug resistance during infection. DIC image of a titan cell (left) and typical cell (right). Cells were isolated from the lungs of a mouse and stained with DAPI to visualize DNA (blue). White dashed lines outline capsules. Scale bar is 10 μm. (B) Ploidy – Titan cells have increased ploidy levels (≥4C) and produce aneuploid daughter cells (≥1C) that are more resistant to the antifungal, fluconazole. Thus the increase in gene copy number in titan cells and their unique ability to undergo genome reductions may promote survival during drug treatment. (C) Cell wall and (D) Capsule – Compared to typical cells, titan cells have a thickened cell wall that contains increased chitin and a mannan fibril layer, and they have a more highly cross-linked capsule with increased GalXM and mannan. These cell surface alterations may reduce drug efficacy through decreased drug penetrance.