What features of host physiology must environmental fungi overcome to cause disease in mammals?

Fungi such as Cryptococcus neoformans, Histoplasma capsulatum, Blastomyces dermatitidis, and Coccidioides spp. are important human fungal pathogens which naturally exist in the environment [1]. Consequently, these fungi must transition from their natural ecological niches to the host environment in order to cause disease in humans. The elevated body temperature of mammals is a profound physiologic bottleneck that reduces the pathogenic potential of the great majority of the world’s species of fungi [2]. In C. neoformans, for example, the ability to tolerate human body temperature is a well-established and intensively studied virulence trait that distinguishes many non-pathogenic environmental strains from those isolated from patients [3]. However, recent studies of large C. neoformans clinical isolate collections indicate that differences in temperature tolerance or other virulence traits such as polysaccharide capsule formation and cell wall melanization cannot comprehensively account for the observed variation in mammalian virulence [4]. One aspect of the mammalian host environment that is dramatically different from the external environment is carbon dioxide concentration: ambient air contains approximately 0.04% CO₂ while mammalian tissues contain 100-fold higher concentrations (~5%). We hypothesized that this dramatic difference may represent a significant stress for C. neoformans and, indeed, competitive growth assays demonstrated that clinical isolates are, generally, more tolerant of host CO₂ concentrations than strains isolated from the environment [5,6]. Furthermore, CO₂-intolerant environmental strains were less virulent than CO₂-tolerant environmental strains [5].

What aspects of C. neoformans physiology are affected by host CO₂ concentrations?

CO₂ plays an essential role in cellular physiology through its conversion to bicarbonate which, in turn, is a critical substrate for multiple reactions involved in central carbon metabolism (Fig 1A). In the low CO₂ concentrations of the external environment, C. neoformans, like other fungi, generates bicarbonate from CO₂ and water through the enzyme carbonic anhydrase [7]. In the host, the high concentrations of CO₂ drive spontaneous bicarbonate formation. As a result, C. neoformans carbonic anhydrase, which is essential in ambient air, is dispensable during mammalian infection [7]. The generation of bicarbonate releases a proton, but our work has shown that the effects of CO₂ on C. neoformans are independent of changes in pH and are not recapitulated by changes in bicarbonate concentration [5].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Carbon dioxide tolerance as a virulence trait in Cryptococcus neoformans.

A. Schematic summarizing the effects of CO₂ on C. neoformans cellular physiology and the signaling pathways that positively (green box) and negatively (red box) CO₂ tolerance. Carbonic anhydrase (CAN). B. The competitive fitness of environmental and clinical isolates from the indicated C. neoformans clades in 5% CO₂ in buffered RPMI medium at 37°C relative to the reference strain H99; data are from reference [6].

https://doi.org/10.1371/journal.ppat.1013351.g001

High CO₂ concentrations cause cellular stress in fungi and bacteria. Indeed, this stress is so significant that it has been exploited in the food industry, where high CO₂ atmospheres are used to prevent food spoilage by bacterial and/or fungal growth (modified atmosphere packaging). Studies on the effects of high CO₂ exposure indicate it alters the ratio of unsaturated/saturated fatty acids of bacterial and fungal membranes and leads to a more fluid phospholipid bilayer and causes membrane disorder [8]. Consistent with these studies, we found that the phospholipid asymmetry of C. neoformans plasma membrane is remodeled at host CO₂ concentrations [9]. Specifically, phosphatidylserine, which is normally localized to the inner leaflet of the plasma membrane, accumulates on the outer leaflet (Fig 1A). C. neoformans strains expressing high levels of an ABC transporter (PDR9) that interferes with phospholipid remodeling of the plasma membrane are less fit in CO₂.

In addition, our group has shown that the susceptibility of C. neoformans to azole drugs, which target synthesis of the fungal membrane sterol, ergosterol, is increased at host concentrations of CO₂. Although host CO₂ appears to have a profound effect on membrane homeostasis, our group has shown that CO₂ tolerance is a quantitative trait with multiple genetic loci contributing to the genetic basis for the difference between CO₂-tolerant and sensitive strains [10]. Current studies are under way to understand how other aspects of C. neoformans physiology are affected by host concentrations of CO₂ stress.

What stress response pathways affect C. neoformans CO₂ tolerance?

Our large-scale genetic screens of publicly available transcription factor and protein kinase deletion mutant collections [9,11] have identified three pathways that positively regulate C. neoformans CO₂ tolerance: (1) Target of Rapamycin (TOR); (2) Regulator of Ace2 Morphogenesis (RAM), and (3) calcineurin—that positively regulate CO₂ tolerance in C. neoformans. Conversely, the Cell Wall Integrity (CWI), Rim 101, and protein kinase A (PKA) pathways negatively regulate CO₂ tolerance (Fig 1A). The distinct roles of these important regulators of C. neoformans physiology suggest that CO₂ stress tolerance requires a balanced response from multiple stress-related pathways.

To date, we have explored the specific functions of the TOR and RAM pathways during CO₂ stress. The TOR pathway, through its target kinase Ypk1, plays a critical role in membrane homeostasis, particularly with respect to membrane rigidity/fluidity and sphingolipid biosynthesis [12]. The latter process is critical to CO₂ tolerance as demonstrated by the increased susceptibility of C. neoformans to the sphingolipid biosynthesis inhibitor myriocin in 5% CO₂ [5]. Furthermore, the TOR-Ypk1 pathway plays a critical regulatory role in remodeling phospholipid asymmetry in response to CO₂. Previous work in C. neoformans has also shown that the TOR pathway suppresses CWI pathway activation [13]. Consistent with this function, we demonstrated that host CO₂ blunts temperature-induced CWI activity, providing an explanation for the increased CO₂ fitness of deletion mutants lacking CWI pathway kinases [9].

The RAM pathway has been extensively characterized in Saccharomyces cerevisiae and regulates cell wall integrity, daughter cell-specific gene expression, mating, polarized growth, RNA turnover, and stress signaling [14]. However, the functions of the RAM pathway in ascomycetes (e.g., S. cerevisiae) are distinct from basidiomycetes (e.g., C. neoformans). For example, RAM pathway mutants in ascomycetes have reduced abilities to undergo polarized growth, whereas C. neoformans RAM mutants display a hyperpolarized morphology [15]. To determine which function(s) of the RAM pathway are related to CO₂ tolerance, we isolated two groups of mutants of strains lacking CBK1, the key RAM pathway kinase, that had regained the ability to grow at 5% CO₂. Both suppressor strains contained mutations in genes involved in RNA homeostasis: SSD1, an RNA binding protein known to be a direct substate of Cbk1 in multiple other fungal species [16], and PSC1, an uncharacterized gene that contains a Poly(A)-specific ribonuclease (PARN) domain [11]. Transcriptional profiling of C. neoformans revealed that RNA processing genes are significantly enriched in the set of genes differentially expressed in 5% CO₂ [9]. Thus, it appears that at least one RAM pathway function important for CO₂ tolerance is the regulation of RNA homeostasis.

What is the relationship between CO₂ tolerance and C. neoformans virulence?

We initially found a strong correlation between CO₂ tolerance and virulence by determining the CO₂ fitness of a set of environmental and clinical C. neoformans strains whose virulence in a mouse model of cryptococcosis had been characterized by Litvintseva and Mitchell [5,17]. We have also used bulk segregant analysis combined with fine mapping of near-isogenic progeny from crosses of CO₂-tolerant and -intolerant strains to show that virulence strongly correlated with CO₂ fitness [10]. Infection of mice with a CO₂-intolerant C. neoformans strain led to mortality in a minority of mice. Importantly, strains isolated from these mice were now CO₂-tolerant, suggesting that the strains were able to adapt to host CO₂ during infection [10]. Finally, in vitro microevolution of CO₂-sensitive strains in a high CO₂ environment generated mutants with increased CO₂ fitness [6]. Two of these evolved, CO₂-tolerant strains contained truncation mutations in AVC1, an ARID domain-containing protein thought to be involved in transcriptional regulation [16]. Deletion of AVC1 in multiple CO₂-sensitive strains led to increased CO₂ fitness and virulence in mice. Intriguingly, similar loss-of-function mutations in AVC1 have been found in C. neoformans strains isolated from patients with relapsed cryptococcosis [18,19]. Together, these data strongly support the idea that CO₂ tolerance is virulence trait and also suggest that CO₂ stress exerts a selective pressure during infection of both mice and humans.

How is CO₂ tolerance distributed among the clades of C. neoformans?

As mentioned above, we have recently characterized the CO₂-tolerance of a large set of clinical and environmental isolates [9]. This set of isolates contained strains from VNI, VNIa, VNIb, VNIc, VNBI, VNBII, and VNII (Fig 1B). Both CO₂-tolerant and -sensitive strains were found in all clades with no one clade showing a significantly increased or decreased number of CO₂-tolerant or -intolerant strains. Within a given clade, we also found that many highly related strains showed different CO₂ fitness. This distribution pattern suggests that CO₂ tolerance may have emerged many times during the evolution of C. neoformans rather than having originated in a specific lineage. Our quantitative trait loci analyses indicate that many genetic loci contribute to CO₂ tolerance. Thus, it seems likely that multiple potential genetic and molecular mechanisms could drive the emergence of CO₂ tolerance in any given strain; the exact nature of these mechanisms is an open question that we are currently investigating. It is also interesting to consider what niche or environmental conditions leads to the selection of CO₂ tolerance? What niche selects against CO₂ tolerance in environmental strains? And finally, do latent infections provide a niche under which C. neoformans adapts to the host environment and, thereby, increases its fitness as a pathogen?