The Spectrum of Fungi That Infects Humans

Abstract

Few among the millions of fungal species fulfill four basic conditions necessary to infect humans: high temperature tolerance, ability to invade the human host, lysis and absorption of human tissue, and resistance to the human immune system. In previously healthy individuals, invasive fungal disease is rare because animals’ sophisticated immune systems evolved in constant response to fungal challenges. In contrast, fungal diseases occur frequently in immunocompromised patients. Paradoxically, successes of modern medicine have put increasing numbers of patients at risk for invasive fungal infections. Uncontrolled HIV infection additionally makes millions vulnerable to lethal fungal diseases. A concerted scientific and social effort is needed to meet these challenges.

Of the millions of fungal species, only a few can naturally infect humans. But medical advances have rendered more people—especially the immunocompromised—susceptible to infections with fungi not previously considered pathogenic.

Fungal infections today are among the most difficult diseases to manage in humans. Some fungi cause disease in healthy people, but most fungal infections occur in individuals already experiencing serious illness, and frequently jeopardize the success of the newest medical advances in cancer care, solid organ and hematopoietic stem cell transplantation, neonatal medicine, autoimmune disease therapies, trauma and intensive care, and sophisticated surgery. In fact, these medical advances themselves often make their beneficiaries vulnerable to fungal disease. Treatment can involve breaching normal immune functions, or susceptible patients, such as extremely premature newborns who survive long enough to become infected by a fungus.

The following discussion intends to touch on highlights of the evolutionary developments by which living humans became substrates for fungi. Given the tremendous wealth of recent findings on fungal evolution, phylogenetics, genomics, development, and pathogenesis, this overview will necessarily omit much work critical to our understanding of fungi, which the other articles in this collection will focus on in detail.

Among the estimated 1.5–5.0 million fungal species on planet Earth (O’Brien et al. 2005), only several hundred cause disease in humans, and very few are able to affect healthy people. Animals coevolved with fungi, and the sophisticated and potent human immune system arose from the constant challenge posed by the microbial world. Fungal pathogens likely adapted their pathogenic repertoire to other animal prey—mammals, insects, and even unicellular amoebae—before encountering and attacking humans. But unlike plants, insects, and ectothermic vertebrates (animals whose body temperature fluctuates with their surroundings), mammals are highly resistant to invasive fungal diseases, and evolution of endothermy and homeothermy enhanced antifungal immunity (Robert and Casadevall 2009; Bergman and Casadevall 2010). The remarkable resistance of mammals to fungal pathogens has been hypothesized to be responsible for emergence of mammals as the dominant land species, when proliferation of fungi at the end of the Cretaceous era created a “fungal filter” that selected for this animal group (Casadevall 2005, 2012).

For a fungus, parasitizing a human is a demanding strategy. Four criteria must be fulfilled. (1) It must be able to grow at a high temperature, at or above 37°C. (2) It must be able to reach the tissues it will parasitize, by penetrating host tissue barriers, or by circumventing them through small airborne cells that enter air-filled spaces of lungs and sinuses directly. (3) It must be able to digest and absorb components of human tissues. (4) Finally, it must be able to withstand the human immune system.

Virulence factors can be divided into aspects of physiology that allow a fungus to fulfill these four criteria. Growth at high temperature is a stringent criterion, because land-colonizing fungi likely evolved in association with plants (Humphreys et al. 2010), whose nighttime temperature must be cool enough to minimize the ratio of carbon-expending respiration to carbon-assimilating photosynthesis. Casadevall and colleagues have discussed how the stable elevated temperature of endothermic (warm-blooded) animals may have been one of the most potent developments in antifungal immunity (Robert and Casadevall 2009; Bergman and Casadevall 2010; Garcia-Solache and Casadevall 2010). The energy cost of the human fever response to infection, which is tightly regulated by cytokines like TNF-α, represents another testimony to the evolutionary importance of temperature as a major host defense against fungi.

The role of fungal heat intolerance is illustrated by the different clinical scenarios of disseminated Fusarium infection versus disseminated candidiasis. Fusarium spp. are important plant pathogens that can infect neutropenic patients. As filamentous Fusarium spp. produce cells resembling yeast in the human host, which readily float within the bloodstream, a hallmark of their dissemination is development of numerous metastatic foci in the skin, presumably because the skin is the coolest organ. The development of metastatic lesions in warm organs distant from the primary focus, like spleen and kidneys, is much more unusual in fusariosis. In contrast, invasive infections with human temperature-adapted Candida spp., which also use yeast to disseminate through the bloodstream, frequently lead to innumerable metastatic foci in deep organs.

Morphogenesis is an important virulence factor related to fungal locomotion. Although fungal hyphae can pierce tissue barriers owing to the turgor pressure at their tips, yeast can readily disseminate to distant sites in a large animal. Fungi that infect healthy humans do so almost exclusively in their yeast form. But for most fungi, the ability to assume various shapes is critical for infecting humans, because many enter the body in the form of small, round airborne dispersal propagules, sporangiospores or conidia, which are produced from hyphal cells. In the oceans, fungi originated as unicellular, oval cells propelled by flagella, so growth as a round or oval cell was a fundamental fungal trait. Lucking et al. (2009), in their comparison of fossil and molecular dating systems of fungal evolution, state that “all dating estimates show that the evolution of filamentous fungi occurred much later than the origin of the fungal lineage itself, suggesting that for a long time after their origin, fungi were heterotrophic, unicellular, flagellate, aquatic organisms.” An option for the hyphal form, whose adaptive utility is shown by convergent evolution of hyphae in fungi and oomycete water molds (Money et al. 2004), may have evolved when fungi colonized land as plant symbionts (Redecker et al. 2000; Heckman et al. 2001; Lucking et al. 2009). Morphogenesis also contributes to protection against amoeboid cell predation. Filamentous forms of Cryptococcus neoformans are resistant to amoeba (Neilson et al. 1978), and incubation of Histoplasma capsulatum and Blastomyces dermatitidis yeast with amoeba at 37°C triggers transition to filamentous growth (Steenbergen et al. 2004).

Tremendous variability of cell forms and the ability to switch between them continues in extant fungal phyla, and few lineages have completely lost a round or a filamentous cell form. A highly successful human-host-associated fungus, Candida albicans, is notable for the facility and frequency with which it switches between a broad spectrum of growth forms in the host, between round yeast, elongated pseudohyphae, and filamentous hyphae (Sudbery et al. 2004). Adhesion molecules are virulence factors related to fungal locomotion strategies because a hypha must anchor itself to host tissue to exert tissue-penetrating pressure at its tip, and because once a fungal cell has reached a favorable location, it needs to be able to stay at the site.

Secretion of digestive enzymes appropriate to dissolve host tissue is the first step for a fungus to use the host as a nutritious substrate. The next step is to absorb the small molecules released by digestion of host tissues; fungal cells have transporters of nitrogen and carbon sources as well as metal ions and other micronutrients. Iron acquisition plays a special role because of its importance in critical enzymatic processes including respiration, and because the human host actively sequesters iron during infection to withhold it from the pathogen (Boelaert et al. 1993; Ibrahim et al. 2006; Kronstad et al. 2013; Noble 2013). The importance of iron sequestration in the host to defend against fungal growth was dramatically illustrated by disseminated Mucorales infections in patients receiving the iron chelator, deferoxamine (Windus et al. 1987).

Most fungi, even with these capabilities, cannot withstand the phagocytes unleashed against them by an immunocompetent human, which have help from opsonins such as complement and specific antibody, and from activating T lymphocytes. Fungi that infect healthy humans devote a large portion of their physiology to withstanding or evading the immune system. To this end they use an amazingly diverse array of strategies, which sometimes differ even between lineages of a single species. In contrast, many species that grow at 37°C can readily infect a severely immunocompromised patient.

Last, coordinating the processes to fulfill these four criteria of pathogenicity demands a complex network of sensing and signaling systems, which inform the fungal cell of external conditions and set in motion its appropriate responses. Loss of virulence was shown many times in many fungal pathogens, when just a single element of a sensing and signaling network was genetically or pharmacologically disturbed, such as mechanosensor, calcineurin, protein kinase A, MAP kinase, Tor, and high osmolarity signaling pathways (Rohde and Cardenas 2004; Cramer et al. 2008; Kumamoto 2008; Bastidas and Heitman 2009; Argimon et al. 2011; Ernst and Pla 2011; Hogan and Muhlschlegel 2011; Shapiro et al. 2012; Chen et al. 2013).

FUNGI THAT INFECT HEALTHY HUMANS

The capability to infect an immunocompetent human has arisen independently multiple times among three major fungal phyla: the Entomophthoromycota (Kwon-Chung 2012), the Ascomycota, and the Basidiomycota.

Entomophthoromycota

Entomophthoromycota contains effective pathogens of insects. They occur worldwide, but have been found as agents of invasive human infection only in subtropical and tropical regions. The species of genera pathogenic for humans, Basidiobolus and Conidiobolus, can be isolated from plant debris and soil especially during rainy months (Bittencourt 1988). As emphasized recently by Kwon-Chung (2012), these genera are evolutionarily distant from each other as well as from the Mucorales, with whom they traditionally were classified in Zygomycota.

Conidiobolus spp. cause submucosal disease of the nose, sinuses, and central face. In most cases, gradual progression causes swelling of submucosal and subcutaneous tissue, disfigurement, breathing difficulties, and chronic bacterial sinusitis caused by the blockage of ostia. Unlike the Mucorales causing sinus disease, Conidiobolus does not usually invade blood vessels nor penetrate into the central nervous system (CNS) (Prabhu and Patel 2004).

Two species, Conidiobolus coronatus and Conidiobolus incongruus, have been isolated from central facial disease. Presumably, the route of infection is inhalation of large ballistospores forcibly launched from the sporangiophores on which they are produced singly to distances up to 30 cm (Isa-Isa et al. 2012). Owing to their large size of 30–38 µm (Isa-Isa et al. 2012), spores presumably land on the mucosa of nasal air passages and fail to reach the distal airways (Bittencourt et al. 2006).

Conidiobolus spp. infect arthropods including mites, spiders, and insects (Isa-Isa et al. 2012). They elaborate lytic enzymes including elastases, collagenases, and lipases, likely to parasitize and kill arthropods (Comerio et al. 2008). These enzymes enable Conidiobolus to digest human tissue, growing by extension from nasal submucosa through the facial soft and bony tissue (Gugnani 1992). Metabolites toxic to insects have also been identified but because the fungus is highly lethal to arthropods and only of moderate virulence in mammals, it remains to be seen whether these metabolites play any role in human infection (Wieloch et al. 2011).

The host sends neutrophils, eosinophils, and histiocytes to contain the fungus, and granulomas (structured assemblies of histiocytes) develop. Histopathology can show dense eosinophilic material surrounding the hyphae, the Splendore-Hoeppli phenomenon, which has been suggested to consist of antigen–antibody precipitate (Isa-Isa et al. 2012).

Basidiobolus ranarum causes subcutaneous disease primarily in children of tropical and subtropical Africa, Asia, and the Americas. Nodules appear at sites of inoculation through insect bites, scratches, or small wounds (Ribes et al. 2000). In addition, Basidiobolus has caused gastrointestinal infections, presumably after ingestion of large inocula, and symptoms resemble inflammatory bowel disease (Zavasky et al. 1999; Vikram et al. 2012). When insects infected by Basidiobolus are eaten by amphibians and reptiles, the fungus is shed in the feces (Ribes et al. 2000) and attaches to plant matter with which it is accidentally inoculated under the skin, for instance, by thorns, leaves used for cleansing, etc. As an insect pathogen, it also elaborates proteases and lipases (Echetebu and Ononogbu 1982; Okafor et al. 1987; Okafor and Gugnani 1990; Okafor 1994). Its inability to invade deep organs is likely related to limited thermotolerance; 37°C may be its maximal growth temperature (Ribes et al. 2000).

Ascomycota

Pathogenic Onygenales

Several soil-inhabiting members of the ascomycete order Onygenales have evolved to parasitize mammals and cause systemic infection. They have been classified in the family Ajellomycetaceae (Untereiner et al. 2004; Bagagli et al. 2008), which include B. dermatitidis (teleomorph, Ajellomyces dermatitidis), H. capsulatum (teleomorph, Ajellomyces capsulatus), Paracoccidioides brasiliensis, Paracoccidioides lutzii, Lacazia loboi, Coccidioides immitis, Coccidioides posadasii, and related fungi with similar properties. The intracellular pathogenic strategy of some of these organisms is similar in protozoa and macrophages (Steenbergen et al. 2001, 2003), raising the possibility that their capacity for virulence arose accidentally as a result of environmental pressures (Casadevall and Pirofski 2007).

Disease caused by these fungi begins asymptomatically in the lungs and progresses to an influenza-like illness or frank pneumonia. They can disseminate to distant sites, can persist and reactivate, with different organ predilections among different genera: Paracoccidioides in oral and respiratory mucous membranes (Queiroz-Telles and Escuissato 2011; Marques 2012), Blastomyces in bones, joints, and skin (Bradsher et al. 2003), and Histoplasma in multiple organs including the gastrointestinal tract and adrenals, as well as bones and skin (Kauffman 2007).

Their primary lifestyle is saprobic. Most genera occur in defined geographic areas, which follow particular soil consistencies, for instance, dry and alkaline for the Coccidioides spp. and acidic for Histoplasma. It has not been ruled out that their geographic distribution follows that of specific host mammals. In the soil, they grow as a branched network of hyphal filaments, a mycelium, and saprobically use organic matter. They produce airborne dispersal cells: Conidia produced from specialized hyphae, as Histoplasma and Blastomyces do, or arthrospores from regulated fragmentation of hyphal compartments, as Coccidioides and Paracoccidioides do. Paracoccidioides produces both arthroconidia and aleuriconidia (conidia directly produced along the hyphae). Once conidia or arthrospores are taken up by a mammal, usually by inhalation, they convert into parasitic yeast (or spherules, in the case of Coccidioides spp.) and initiate an infection. The principal signal for their conversion to the yeast form is the elevated mammalian body temperature, so that this switch can be induced in vitro when they are cultured at 37°C or above. For this reason they are grouped as thermal dimorphs. These fungi infect, persist, and cause disease in healthy hosts. In immunocompromised hosts, the infections are more clinically apparent, more severe, and more likely to disseminate to multiple organs.

Adaptive cellular immunity is required to control the infection with these fungi and can be detected by skin tests with purified antigens. For this reason, clinically apparent infections with some members of this group, such as H. capsulatum, became much more common with the rising prevalence of AIDS. These Onygenales (Ajellomycetaceae) are the only phylogenetically related group whose members all cause systemic disease in otherwise healthy people.

Other thermally dimorphic fungi are Sporothrix schenckii (order Ophiostomatales), which is primarily introduced into the host through skin injuries, and Talaromyces (Penicillium) marneffei (order Eurotiales), which does not infect healthy hosts but because of the AIDS epidemic has become a prevalent and important invasive pathogen in Southeast Asia. These fungi coevolved with higher vertebrates, as it is the adaptive immune system by which they are controlled. The fact that these evolutionarily distant fungi convert from saprobic hyphal to pathogenic yeast growth at the temperatures of endothermic animals suggests that an ability to access large nitrogen and other nutrient stores of an animal host is an advantageous lifestyle option for a soil dweller, and a result of convergent evolution in several phyla.

H. capsulatum will be discussed at greater length because it is well studied and also because it exemplifies important principles for other thermally dimorphic Onygenales. This fungus inhabits soil and other organic matter such as decaying wood enriched with bat and bird droppings. Its hyphae produce macro- and microconidia, and the latter are small enough, <5 µm, to penetrate into the alveoli of mammalian lungs. There they are phagocytized by alveolar macrophages and converted to yeast that is able to proliferate in the hostile environment of the phagolysosome.

Aspects of histoplasmosis resemble infections with a well-adapted bacterial intracellular pathogen, Mycobacterium tuberculosis. Depending on inoculum size, infection is most often inapparent or leads to a self-limited influenza-like illness. The pathogen can persist indefinitely until immune decline allows it to reactivate. Lung disease resembling TB can ensue immediately after infection, and dissemination to all organs occurs depending on host immune status and age. The response of an immunocompetent host is granuloma formation.

Histoplasma occurs on all continents, although it is rare in Eurasia and most common in North and South America. Two studies of H. capsulatum showed that its growth and clinical phenotypes, which had traditionally been the basis for its division into varieties (capsulatum, duboisii, and farciminosum), did not align with the divisions of its seven phylogenetic clades (Kasuga et al. 1999, 2003), and that the different clinical disease states attributed to these (now obsolete) varieties were caused by the same clades.

Phylogenetic analyses suggest that H. capsulatum arose in South America 3.2–13.0 million years ago and its present distribution reflects a period of rapid spread to all continents except Eurasia (where it was later introduced), followed by restriction to the warmest areas during a period of intense cold 1.8 million years ago (Kasuga et al. 1999, 2003). From there, it spread to temperate regions as the earth warmed again. This idea is consistent with its low genetic diversity in temperate regions and high diversity in tropical areas (Kasuga et al. 1999), mirroring the diversity of plant species in areas of temperate and tropical forest, which are its geographic areas of distribution.

The two North American clades are not closely related and are thought to have been genetically isolated for the past ∼20 million years (Kasuga et al. 1999), despite overlapping territories. These two clades differ in virulence, with North American clade 1 having been isolated predominantly from AIDS patients, and clade 2 infecting immunocompetent hosts (Kasuga et al. 1999). A clade 2 isolate and a representative of the Panama clade have been used in genetic studies of virulence factors, and surprising differences were found, as discussed below.

The habitat of Histoplasma is soil enriched with bird and bat droppings, in specific geographic locations including the Ohio and Mississippi valleys of North America and moist regions of Central and South America. Whereas birds are very rarely infected with H. capsulatum (Quist et al. 2011), bats are frequently infected but respond with minimal inflammation (Taylor et al. 1999), suggesting active repression of the host immune response by the fungus. It was suggested that, “in at least some bat species natural exposure may result in a chronic, controlled infection, which allows the bat to excrete viable fungi over a long period” (Hoff and Bigler 1981). Experimentally infected bats also had a minimal inflammatory response (McMurray and Greer 1979).

A broad variety of other infected mammals such as cats (Aulakh et al. 2012), dogs (Brömel and Sykes 2005), sea otters (Morita et al. 2001), and badgers (Bauder et al. 2000) do respond with granulomatous inflammation to the presence of Histoplasma yeast. Perhaps H. capsulatum has evolved to infect bats, in which it avoids activation of host responses, by mechanisms only partially adapted to other immunocompetent mammals. If this is the case, the fungus has established an efficient amplification cycle by which its mycelial form grows in droppings under bat roosting places, for instance, in caves, and its microconidia are released into the air to be inhaled by the bats that fertilize its soil substrate (Taylor et al. 1999).

Several mechanisms by which H. capsulatum yeast successfully interacts with mammalian hosts have been identified. Surprisingly, different clades use different strategies to achieve an intracellular lifestyle in macrophages, and to suppress macrophage activation. For example, macrophages recognize the important pathogen associated molecular pattern (PAMP) β(1,3)-d-glucan with the receptor dectin-1 (Rappleye et al. 2007), and ligand binding triggers responses that enhance phagocyte activation responses (Underhill et al. 2005). β(1,3)-d-glucan is the main structural polymer of ascomycete cell walls, so Histoplasma cannot forgo its use to avoid activating their hosts’ macrophages. Instead, Histoplasma yeast of most clades hide it by coating their cell surface with α(1,3)-d-glucan. Perturbation of genes whose products are required for α(1,3)-d-glucan synthesis renders cells of the Panama clade avirulent (Rappleye et al. 2004). Notably, this α(1,3)-d-glucan coat is lacking in a North American clade 2 strain (Edwards et al. 2011), and an α(1,3)-d-glucan synthase mutant in this strain background is fully virulent (Edwards et al. 2011).

Another cell surface protein, Yps3, is expressed to a significant level only in the more virulent North American clade 2 (Bohse and Woods 2007). This protein, a homolog of B. dermatitidis BAD1, contributes to dissemination of H. capsulatum to extrapulmonary foci (Holbrook and Rappleye 2008). Its gene is present in most strains but substantial transcription is known to occur only in North American clade 2, the clade lacking α(1,3)-d-glucan. Yps3, like BAD1, may directly suppress TNF-α production, circumventing the need for this clade to use α(1,3)-d-glucan to hide the stimulatory PAMP β(1,3)-d-glucan (Holbrook and Rappleye 2008). The known variations in their pathogenic repertoires among clades of Histoplasma are based in differences of regulatory rather than coding regions between the sequenced strains (Edwards et al. 2013).

H. capsulatum yeast have evolved a repertoire of manipulative molecules to turn the host compartment specifically evolved to kill pathogens, the phagolysosome, into a habitat in which they thrive (Sebghati et al. 2000; Youseff et al. 2009). Remarkably, within the teleomorph genus Ajellomyces, originally comprising the anamorph sister species H. capsulatum and B. dermatitidis, one species H. capsulatum chose this route of intracellular life, whereas the other, B. dermatitidis, is a successful extracellular pathogen.

Distinct H. capsulatum clades differ in specific details of the pathogenic program of their yeast, but they have in common that their suite of virulence factors is turned on simultaneously with the switch from hyphal to yeast growth. Mammalian body temperature of 37°C or above is the main signal to trigger conversion from the mycelium to yeast. This conversion entails differential transcription of ∼20% of the genome (Nguyen and Sil 2008), and is controlled by a network of transcriptional regulators apparently responding to a histidine kinase signal (Nemecek et al. 2006; Webster and Sil 2008; Cain et al. 2012; Beyhan et al. 2013). Mutants in these regulators are incapable of the transition to yeast growth, and are avirulent (Nemecek et al. 2006; Webster and Sil 2008; Cain et al. 2012; Beyhan et al. 2013).

Other members of the Ajellomycetaceae, all of which associate with mammals, have equally distinctive ecology and pathobiology. Mycelia of the Coccidioides spp., growing near rodent burrows in arid regions of the Americas (Nguyen et al. 2013), produce arthroconidia. When arthroconidia are inhaled by a mammal, they give rise to round cells—spherules—in which growth and mitosis occur to eventually fill the enlarging mother cell with hundreds of endospores. On maturity, the spherule ruptures and releases the endospores into host tissue. Newly released endospores repeat the growth cycle locally or disseminate hematogenously. Phylogenetic analysis shows that the fungus evolved in North America before that continent’s geologic connection with South America 2.5–3.5 million years ago (Fisher et al. 2001), and that the genetically homogeneous South American strains, derived from populations in present-day Texas, reached their current distribution 8940–134,000 years ago, possibly with the migration of humans from North into South America (Fisher et al. 2001). Another unique member of the Ajellomycetaceae, L. loboi, shares an important feature with a distant member of its phylum Ascomycota, Pneumocystis jirovecii. They are the only human fungal pathogens not grown in culture (Vilela et al. 2009). Lacazia, a sister genus to Paracoccidioides, has been found only in subcutaneous infections of previously healthy humans in the Amazon basin and in dolphins of the Atlantic ocean and of the Amazon (Vilela et al. 2009; Theodoro et al. 2012), and (analogous to Mycobacterium leprae) can be amplified only in a mouse model. In contrast, Paracoccidioides spp., prevalent fungal pathogens that cause pneumonia, systemic disease of the monocyte–macrophage system, and destructive lesions of skin and oral mucous membranes (Fortes et al. 2011; Queiroz-Telles and Escuissato 2011; Marques 2012, 2013), occur in Central and South America mainly outside the Amazon region (Theodoro et al. 2012). Paracoccidioides causes clinical disease in many more men than women, in a ratio of 13:1, possibly because estrogen blocks the conversion of inhaled arthroconidia to the tissue-invasive yeast form (Shankar et al. 2011). This fungus frequently infects armadillos, and its virulence factors such as the immunodominant adhesin glycoprotein 43 (Puccia et al. 1999, 2008, 2011; Fortes et al. 2011) may have evolved in its coevolution with these ancient mammals (Bagagli et al. 2006; Richini-Pereira et al. 2009).

Some members of the Onygenales infect healthy humans, but very rarely cause invasive disease. Dermatophytes of the genus Arthrodermataceae specialize in degrading keratins, important structural proteins in vertebrate skin whose extensive cross-linking by disulfide bonds makes them proteinase resistant (Gradisar et al. 2005). With the evolution of animals with substantial keratin appendages such as scales, feathers, nails, and hair, specializing in use of these nutrient sources must have been a worthwhile niche for the presumed ancestor of the dermatophytes (Weitzman and Summerbell 1995).

Reptilian amniotes had evolved keratin, so that the basis for this specialization was available as a fungal substrate 300 million years ago (Eckhart et al. 2008). Recent analysis of mitochondrial genomes of six dermatophytes comprising the three morphologically defined anamorphic genera, Trichophyton, Microsporum, and Epidermophyton, together with 29 other members of Ascomycota, confirmed that the dermatophytes are descended from a common ancestor (are monophyletic) and separated from other fungi only 32–50 million years ago (Wu et al. 2009). This study also found that classification of species by molecular characters frequently contradicted the traditional classifications based on morphologic traits (Wu et al. 2009), emphasizing the fact that fungal morphology is enormously flexible and convergent evolution is common, so that only limited predictions of their physiology can be made from the appearance of fungal structures. In contrast, molecular phylogenetic studies have confirmed the grouping of dermatophytes by their ecologic and clinical features, including the distinct sites of the human body they tend to infect (Cafarchia et al. 2013).

Anthropophilic dermatophytes travel over the world with their human hosts. Although many species are endemic to specific geographic regions, some species occur worldwide. Their prevalence varies with the lifestyle and socioeconomic conditions of their human hosts, and is undergoing continuous epidemiologic changes (Ameen 2010). Unlike their thermal dimorphic Onygenales relatives, they can reach a new host by person-to-person transmission, and their access to new human substrate is made even easier by the ability of their arthroconidia to persist for years in fomites (Weitzman and Summerbell 1995). Anthropophilic dermatophytes, relying on person-to-person access to new hosts, can forgo long-distance travel via airborne conidia, which their geophilic relatives produce in larger profusion to reach distant deposits of animal keratin (Weitzman and Summerbell 1995).

Coevolution with their animal hosts presumably allowed dermatophytes to evade or resist the evolutionarily more ancient innate immune mechanisms, so that today their control requires adaptive cellular immunity (Dahl 1993). By restricting their usual habitat to the most superficial keratinized layer of the skin and its appendages, dermatophytes reduce their contact with immune cells. Human-specific dermatophytes (anthropophiles) are able to down-regulate host inflammation to establish a chronic infection (Blake et al. 1991; Shiraki et al. 2006). In contrast, soil-dwelling (geophilic) and animal-specific (zoophilic) dermatophytes are unable to manipulate host immunity and are eliminated by a vigorous inflammatory response, illustrating the principle that a well-adapted parasite carefully calibrates its virulence.

Basidiomycota

Cryptococci

Basidiomycete yeast with a worldwide distribution, cryptococci, in the past century, infected humans only rarely (Molez 1998). In the 1950s, increasing numbers of cryptococcal meningoencephalitis were reported from central Africa, viewed in retrospect as sentinels of the emergence of AIDS around the Congo River (Molez 1998). AIDS is still the setting in which the vast majority of cryptococcosis occurs (Mitchell and Perfect 1995; Pukkila-Worley and Mylonakis 2008); for 2006, 957,900 cases of cryptococcal meningitis associated with AIDS were estimated, resulting in 624,700 deaths (Park et al. 2009). Furthermore, recent outbreaks of Cryptococcus gattii infections with significant mortality on Vancouver Island and the northwestern United States raise the concern that the fungus may be evolving to become virulent for healthy humans (Fraser et al. 2005; Byrnes et al. 2010; Springer et al. 2012).

The pathogenic cryptococci, C. neoformans and its sister species C. gattii, enter a human by inhalation of infectious cells: dried yeast or possibly basidiospores, the products of meiosis after mating. When these small airborne cells, <5 µm in size, are inhaled, pneumonia can ensue in susceptible hosts, and on gaining access to the bloodstream, the yeast can disseminate to all organs. Cryptococcus has a special predilection for the CNS and causes subacute meningoencephalitis, in which high intracranial pressure plays an especially deleterious role, which, if left untreated, is lethal. The fungus can persist for years in the lung or in sites of previous dissemination and reactivate only on weakened immune surveillance.

C. gattii causes invasive disease and death in wild and domestic land and ocean mammals and birds and pneumonia and meningoencephalitis in previously healthy people (Krockenberger et al. 2002; Miller et al. 2002; Raso et al. 2004; Santos et al. 2008; McGill et al. 2009). A wealth of analyses of pathogenic cryptococci over the past several decades have shown that C. gattii in many settings is a more common primary pathogen than C. neoformans (Speed and Dunt 1995; Springer et al. 2012), but new reports continue to blur the boundaries between primary and opportunistic pathogens among the Cryptococcus species (Chen et al. 2008; Chau et al. 2010; Choi et al. 2010), illustrating the limited utility of these terms (Casadevall and Pirofski 2001). A recent report has indicated that the presence of anti-GM-CSF autoantibody is a risk factor for C. gattii CNS infection in otherwise healthy individuals (Saijo et al. 2014). It suggests that patients with cryptococcal CNS infection considered “immunocompetent” may carry immune defects that can yet not be identified by routine immunological screening. An important condition for a clearer understanding of ecology, epidemiology, and evolution of cryptococci is that more resources for microbiologic analysis become available in poor countries. Cultures of blood and cerebrospinal fluid currently cannot be performed in medical settings in which most patients with cryptococcosis will seek help.

The separation of the sister species C. gattii and C. neoformans is thought to have occurred 45 million years ago (Simwami et al. 2011). Molecular typing recapitulates traditional serologic classification of varieties into C. neoformans var. grubii (serotype A, VNI, VNII, VNB) and var. neoformans (serotype D, VNIV) and their hybrids (serotype AD, VNIII), as well as C. gattii (serotypes B and C) (Mitchell and Perfect 1995; Meyer et al. 2003). C. gattii has been subdivided into molecular varieties or cryptic species VGI through VGIV, among which further subdivisions correlate with geographic location and virulence (Kidd et al. 2004; Fraser et al. 2005; Byrnes et al. 2010).

In the environment, C. gattii has been isolated from numerous tree species in tropical and more recently in temperate regions, particularly from sites of wood decay and insect consumption of vegetable matter (Ellis and Pfeiffer 1990; Fortes et al. 2001; Kidd et al. 2003); although it was first found on Eucalyptus trees and believed to follow their worldwide distribution (Ellis and Pfeiffer 1990), its isolation from the midst of pristine Amazon rainforest revises the interpretation of its origin (Fortes et al. 2001). Environmental sampling on Vancouver Island to define the reservoirs of the outbreak that began in 1999 found more than 10 species of trees to yield C. gattii, which comprised ∼10% of the trees sampled (Kidd et al. 2007), and it survives for at least 1 yr in fresh- and seawater, suggesting it could spread with ocean currents (Kidd et al. 2007).

Supporting the idea of an origin of pathogenic C. gattii in South America is the uniquely symmetric distribution of mating types in haploid isolates from that continent. Although clinical and environmental isolates of the α mating type vastly predominate in other parts of the world, and isolates of the a type are almost never encountered, the South American distribution is 0.8a–1.0α (Hagen et al. 2013).

C. neoformans var. grubii, which is responsible for ∼95% of cryptococcal infections worldwide and 98% of infections among AIDS patients (Simwami et al. 2011), is genotypically most diverse in southern Africa; and a collection of strains from Botswana contained 12% isolates of the a mating type, which in other sites is extremely rare (Litvintseva and Mitchell 2012). In that study, distinct genotypes were thought to have diverged 5000 years ago; the investigators speculate that pigeons contributed to the worldwide spread of one specific strain of C. neoformans var. grubii (Litvintseva and Mitchell 2012) because C. neoformans is enriched in bird droppings and has been shown to mate on pigeon droppings (Nielsen et al. 2007). A study from Thailand supports the concept of an origin of C. neoformans var. grubii in Africa (Simwami et al. 2011).

Like C. neoformans, with which it shares virulence traits, and unlike nonpathogenic cryptococcal species (Findley et al. 2009; Araujo Gde et al. 2012), C. gattii possesses a thick polysaccharide capsule composed of glucurono- and galactoxylomannan, which, analogous to encapsulated bacteria like Streptococcus pneumoniae, blocks phagocytosis of the organism unless it is opsonized (Casadevall et al. 1998). Presumably, the dense capsule also provides some protection against hydrolytic enzymes exocytosed by neutrophils. Capsular glucuronoxylomannan and other macromolecules, as well as virulence-associated proteins, are carried through the cell wall in an amazing process involving vesicles that originated from the late endosome (multivesicular body) (Rodrigues et al. 2007, 2008). This process, first discovered in Cryptococcus, subsequently was found in other pathogenic fungi like Paracoccidioides (Vallejo et al. 2011, 2012), and in the model yeast Saccharomyces cerevisiae (Oliveira et al. 2010). Like other fungi, cryptococci can diversify morphologically in the host, producing a population of Titan cells, giant cells of up to 50 µm in diameter (Okagaki et al. 2010; Zaragoza et al. 2010) created by DNA replication and growth not followed by mitosis (Okagaki et al. 2010). Other antiphagocytic mechanisms, independent of the capsule, are controlled by two GATA transcription factors and are likely to involve coordinated regulation of numerous physiologic processes because hundreds of genes are differentially expressed in their mutants (Liu et al. 2008; Chun et al. 2011).

Once ingested by alveolar macrophages not activated by T lymphocytes, the fungus can survive and proliferate, and it is thought that some cryptococci may enter the brain within macrophages crossing the blood–brain barrier (a Trojan horse-like mechanism), whereas free yeast in the bloodstream cross the blood–brain barrier by transcytosis (Casadevall 2010; Shi et al. 2010). To survive in phagocytes, it buffers oxidative and nitrosative stress with melanin deposited in its cell wall (Wang et al. 1995), and with neutralizing enzymes like superoxide dismutase, glutathione reductase, and thioredoxins (Kronstad et al. 2011). Both in macrophages and in the free-living amoeba Acanthamoeba castellanii, it disrupts phagolysosome membranes to prevent acidification of this compartment and to dilute the lytic enzymes it contains; it also sheds large amounts of capsular polysaccharide to fill the phagocytic cells’ cytoplasm (Steenbergen et al. 2001; Tucker and Casadevall 2002). Virulent strains not only resist killing but proliferate in the amoebae (Steenbergen et al. 2001) so that the predator finds itself prey.

How did some cryptococci evolve from survivors of amoeba attacks to occasional parasites of humans? A recent study examined evolutionary relationships and virulence traits of C. gattii and C. neoformans strains, and related fungi belonging to the order Tremellales, which do not infect mammals (Findley et al. 2009). Other Tremellales parasitize fungi by attaching to the host fungus with a specialized hypha and accessing its cytoplasm through a newly created pore (Zugmaier et al. 1994; Millanes et al. 2014). An evolutionary trajectory from parasitizing another live fungus to parasitizing an animal host may have proceeded through association with and parasitism of insects, and through selection in soils by amoebae for traits that can also function in mammalian virulence. The human-pathogenic Cryptococcus species killed larvae of the wax moth Galleria mellonella in one study, whereas their Tremellales relatives partially or completely lacked virulence in this insect model (Findley et al. 2009). Of note, insect cellular immune defenses consist of amoeboid phagocytes (Williams 2007; Browne et al. 2013), so that resistance to free-living amoebae may transfer well to resistance to insect plasmatocytes. However, in nature, cryptococci so far have only been found in association with insect frass (the excreta of plant-eating insects), not as insect parasites (unlike, e.g., the entomophthorales and the ascomycetous Cordyceps spp.). Further ecologic research will have to show whether insect parasitic cryptococcal relatives can be identified, perhaps in their ancestral tropical forest habitats. The role of its versatile mating system in virulence evolution of Cryptococcus will be discussed in the dedicated articles in this collection.

FUNGI THAT INFECT IMMUNOCOMPROMISED HUMANS

A sufficiently immunocompromised host can be infected by hundreds of environmental fungal species that grow at human core temperatures. However, a predictable set of actors is known to cause the most common invasive infections in immunocompromised individuals, and they will be discussed next according to their phylogenetic affiliations, which predict important parameters of their physiology.

Ascomycota

Candida

Candida species are a polyphyletic group of the order Saccharomycotina, which live as commensals on mammalian mucous membranes, particularly of the gastrointestinal tract (Wrobel et al. 2008). They have rarely been found in the soil (Marples and Di Menna 1952; Skinner and Fletcher 1960). A limited number of species are commonly associated with humans as colonizers and opportunistic pathogens: C. albicans, Candida glabrata, Candida parapsilosis, Candida tropicalis, Candida lusitaniae, and Candida krusei. Of these, C. albicans is the most commonly isolated human commensal and pathogen (Odds 1988; Kam and Xu 2002; Krcmery and Barnes 2002). The frequency of colonization with nonalbicans Candida species shifts according to age of the host, with C. parapsilosis being more prevalent among children and C. glabrata among older adults (Soll 2002).

Invasive candidiasis takes many forms, depending on the setting in which a host became susceptible to this opportunist, and depletion of normally competing bacterial flora by antibiotics often plays a role. For example, in cancer patients receiving chemotherapy, candidiasis is often caused by fungal transmigration through disrupted intestinal epithelium, and the organs with the heaviest fungal burden are those connected by the portal circulation—liver and spleen—but not the brain. In contrast, premature newborns with candidiasis often have abscesses in the brain, because their blood–brain barrier may be immature and because their brain receives such a large fraction of cardiac blood output.

C. albicans is the human companion fungus par excellence. Many healthy humans carry C. albicans, which harmlessly colonizes mucous membranes to high numbers at different anatomic sites of a single individual (Odds 1984, 1988; Soll et al. 1991). It has evolved to flourish in a wide range of environmental conditions: high pH in the intestine versus low pH in the vagina, feast and famine nutritional conditions in the gastrointestinal tract according to its host’s mealtimes versus steady glycogen supply in the vagina, and aerobic conditions on oral surfaces versus anaerobic conditions in the intestine. Unlike the dermatophytes, in which different clades inhabit different anatomic sites, C. albicans biotypes are generalists and each can adapt to all sites (Odds 1984).

Not only can C. albicans survive and thrive in highly disparate microenvironments of the human host, but as part of the normal flora it has also honed its ability to avoid triggering human immune defenses, for instance, by covering its main structural cell wall component, β(1,3)-d-glucan, with glycoproteins to avoid engaging the macrophage dectin-1 receptor (Wheeler and Fink 2006).

Analysis of its populations, which comprise at least five clades, suggests that C. albicans migrated throughout the world with its human hosts (Lott et al. 2005). It is unknown whether the fungus accompanied the mammalian lineages that led to primates throughout evolution, or whether it made the jump to mammalian commensal at a more recent time before emergence of modern humans (Lott et al. 2005).

Perhaps in its long cohabitation with the human, which as to host numbers proved to be a winning strategy, C. albicans became so closely hostadapted that major genomic rearrangements, as facilitated by sex and meiotic recombination, became less advantageous (Goddard et al. 2005). In a stable and benign environment, such as the equilibrium between thriving hosts and commensals, the expenditure of time and energy required for sexual reproduction may cost more than it benefits. C. albicans appears to consist of stable clonal diploid populations (Bougnoux et al. 2008), and its mating and parasexual cycle seem to be such rare events that their discoveries were sensational paradigm shifts in the community of Candida researchers (Hull and Johnson 1999; Magee and Magee 2000; Miller and Johnson 2002; Alby et al. 2009; Hickman et al. 2013). Although the parasexual cycle occurs infrequently, it can generate progeny with extensive genetic diversity by shuffled combinations of eight chromosomes or by recombination between homologous chromosomes mediated by the conserved Spo11 protein integral to meiotic recombination (Forche et al. 2008). More discoveries are sure to follow on the role of mating and various ploidy states in Candida ecology today, and virulence associated with the mating loci (Lockhart et al. 2005; Wu et al. 2007) may become more important as Candida habitat changes with current changes in human epidemiology.

In the millions of years of coevolution, human predecessors and humans survived only briefly with diseases during which C. albicans could significantly invade the host. Invasive candidiasis was very rare (Browne 1954; Zimmerman 1955). C. albicans is an efficient invasive pathogen, causing mucous membrane infections in individuals with ineffective adaptive cellular immunity and fatal disseminated infections in patients lacking functional innate immune cells, neutrophils. For a microorganism that relies on person-to-person transmission and does not have a significant soil reservoir (or mode of locomotion back from the soil to its primary host), the benefit of multiplying to the point of host death seems mysterious. One idea might be that the Candida pathogenic repertoire evolved to take advantage of a constant large number of temporarily susceptible humans, infants, who develop oral candidiasis because of immature adaptive immunity. Human infants’ propensity to widely distribute their oral secretions (to slobber) may allow infecting Candida to launch its increased numbers from infected mucous membranes of these hosts and to colonize more of their contacts. If this is the case, Candida virulence factors evolved to efficiently infect mucous membranes, and invasive disease is for it an accidental dead end.

The pathogenic repertoire of Candida comprises all features needed for a human fungal pathogen. It is able to grow well at human febrile temperatures of 39°C–40°C. It can penetrate host tissues with hyphal cells and it has multiple adhesion molecules to facilitate the drilling action of the hyphal tip (Staab et al. 1999; Sundstrom 2002; Hoyer et al. 2008; Liu and Filler 2011). Its hyphae constitutively produce yeast, constantly diversifying the population with more mobile cells (Shen et al. 2008). An array of lytic enzymes suitable for digesting human macromolecules is induced during tissue invasion (Ghannoum 2000; Staib et al. 2000; Calderone and Fonzi 2001; Schaller et al. 2005; Albrecht et al. 2006; Trofa et al. 2011), and transporters are regulated coordinately to absorb the released monomers into the fungal cell, as reviewed by (Morschhäuser 2011), and to mobilize micronutrients from the host (Weissman et al. 2008; Citiulo et al. 2012; Noble 2013). Unbiased genetic screening yielded novel virulence factors like glucosylceramide synthesis (Noble et al. 2010) and is expected to reveal new aspects of Candida invasive lifestyle (Holland and Summers 2008).

Resistance of Candida to the human immune system is limited. Features increasing its virulence were likely selected against in its long coevolution with us as a commensal. Yet today we are saving, and improving the quality of, human lives that are vulnerable to this opportunist, and any line of investigation that might result in better control of invasive candidiasis has the potential worldwide to save hundreds of thousands of lives every year (Pfaller and Diekema 2007).

Aspergillus

Aspergillus fumigatus, the species responsible for ∼90% of invasive aspergillosis (Schmitt et al. 1990), causes relentless pneumonia, sinusitis progressing through tissue planes, and brain abscesses in neutropenic patients and those with phagocyte defects like chronic granulomatous disease. It is feared in immunosuppressed individuals because its susceptibility to antifungals is limited. This organism also causes disease at another extreme of immune function: allergic reactions.

Normal human innate immunity controls A. fumigatus, a versatile thermophilic plant saprobe. It is well adapted to the high temperatures that occur during bacterial decomposition of dead plants and tolerates thermophilic bacteria’s optimum of around 55°C (Latgé 1999). At human febrile temperatures of 39°C–40°C, it grows well. Its secretion of a very broad and redundant range of hydrolases (Kothary et al. 1984; Kwon-Chung and Sugui 2013), evolved in the tumultuous competition for nutrients among the microbiome of decaying plants, allows it to easily access human tissues as sources of nutrition (Abad et al. 2010).

The mycelia of A. fumigatus give rise to conidiophores, specialized hyphae that produce small (∼3 µm) airborne dispersal cells, conidia. Their small size allows them to remain airborne for long periods and, incidentally, to enter human alveoli. On average, a person is estimated to inhale several hundred A. fumigatus conidia each day (Hospenthal et al. 1998); with exposure to grass cuttings, leaf litter, or compost, this number may be orders of magnitude higher (Mullins et al. 1976; Poole and Wong 2013). Conidia are coated with hydrophobic proteins (Thau et al. 1994; Paris et al. 2003) and with the chemoprotectant melanin (Pihet et al. 2009) to withstand harsh environmental stressors like freezing, sunlight, and desiccation (Kwon-Chung and Sugui 2013). They are recognized by innate immune cells of a human host only when they begin to germinate to produce a growing hypha (Levitz and Diamond 1985; Aimanianda et al. 2009) after landing in an air-filled space like a paranasal sinus or lung. Aspergillus conidia can give rise to allergic lung disease, and the fungus can colonize the bronchiectasis of cystic fibrosis or chronic obstructive lung disease; but lethal invasive infection occurs when there is a dearth of functional neutrophils and macrophages to control germinating conidia. Hyphae proliferate in lung parenchyma or sinuses, and hyphal fragments occasionally are carried to highly perfused organs like the brain in which they initiate foci of metastatic infection. On encountering blood vessels, Aspergillus hyphae tend to enter and follow their course, clogging the vessel and causing infarction of downstream tissue. This angioinvasive behavior may be attributable to thigmotropism, the ability to sense and follow contours, which aspergilli have in common with fungi of diverse phyla (Perera et al. 1997; Bowen et al. 2007).

Pathogenicity studies of A. fumigatus highlight general themes of innate immunity (Morton et al. 2012). For example, specialized alveolar epithelial cells, type II pneumocytes, phagocytose and kill conidia of A. fumigatus in their lysosomes, unless the conidia manage to germinate before their death (Wasylnka and Moore 2003). Neutrophils are attracted by the chemokines secreted by resident alveolar macrophages encountering A. fumigatus during their patrols of the alveolar space. Whereas small hyphae are engulfed and killed by neutrophils, larger hyphae are killed by their extracellular release of reactive oxygen and nitrogen species and antimicrobial peptides, in dependence on the pathogen recognition receptors (PRRs) TLR2 and 4, and dectin-1 (Netea et al. 2003; Kennedy et al. 2007; Werner et al. 2009). Neutrophil extracellular traps (NETs), critical to control polymicrobial infection in appendicitis (Brinkmann et al. 2004), nutritionally constrain A. fumigatus hyphae (McCormick et al. 2010) and down-regulate excessive inflammation elicited by their presence (Rohm et al. 2014).

Non-Fumigatus Aspergilli, Fusarium, Pseudoallescheria, and Other Opportunistic Ascomycetous Fungal Pathogens

Profound immunosuppression, for instance, during prolonged neutropenia, graft-versus-host disease, or severe rejection episodes of transplanted solid organs, permits invasive disease of many other types of environmental filamentous ascomycetes. Among them are the non-fumigatus species of Aspergillus (Torres et al. 2003), Fusarium solanum, Fusarium oxysporum, and other Fusarium spp., and their teleomorph Nectria spp. (Nucci and Anaissie 2007), Pseudoallescheria boydii, and its Scedosporium anamorphs (Quan and Spellberg 2010). Infections with these fungi are often lethal because the hosts they usually infect are incapable of mounting an effective immune response, and because they tend to be more resistant to currently available antifungals.

Fusarium spp. have evolved to infect plants, and the genomes of certain lineages contain one or more entire chromosomes encoding plant pathogenicity factors, which may be horizontally transferred within the genus (Ma et al. 2010). Encoded on pathogenicity chromosomes, and other genomic clusters reminiscent of bacterial pathogenicity islands, are secreted hydrolytic enzymes and signaling molecules expressed during early plant infection (Ma et al. 2010; Rep and Kistler 2010). Whether possession of a pathogenicity chromosome corresponds to virulence for humans, for instance, in the F. oxysporum lineage identified to have caused 70% of invasive fusariosis in a San Antonio hospital (O’Donnell et al. 2004), is not yet known. In Fusarium, morphogenesis contributes to virulence for humans, because yeast-like cells are produced from hyphae in the host, and typically spread widely through the bloodstream to cause numerous foci of infection in the skin.

Pigmented filamentous ascomycetes, also called dematiaceous fungi, infect immunocompromised and, rarely, immunocompetent individuals to cause phaeohyphomycosis. Their pathobiology is diverse including brain abscesses (Cladophialphora bantianum, Ramichloridium spp., and Dactylaria gallopava), keratitis, sinus or soft tissue infections, ulcers, and cysts (Exophiala jeanselmei, Exophiala dermatitidis, Curvularia spp., Bipolaris, or Alternaria spp.). When directly inoculated into tissue, these fungi can cause devastating disease, as emphasized in a tragic outbreak caused by contaminated injectable steroids (Smith et al. 2013).

Another epidemiologically, biologically, and profoundly important ascomycete, P. jirovecii, is so distinctive in its biology, that only a reference to the related article in this collection can be made (Gigliotti et al. 2014).

Basidiomycota

In addition to the Cryptococcus spp. discussed above, other Basidiomycota growing predominantly as yeast in the host, Malassezia furfur, Trichosporon asahii, and other members of human skin flora, are opportunists in patients with venous catheters and in immunosuppressed patients.

Mucorales

Profoundly immunocompromised patients at risk for the environmental fungi described above are also at risk for Mucorales infections. Additionally, Mucorales cause severe disease in diabetic patients especially at times of uncontrolled blood glucose, and patients with elevated serum iron (e.g., attributed to hemosiderosis), as described in the excellent series by Roden et al. (2005). Like the aspergilli, these fungi cause disease where airborne dispersal cells, sporangiospores, enter air-filled spaces (i.e., in sinuses and lungs). Infections with Mucorales progress rapidly as the fungus grows fast and relentlessly through tissue planes and bone, penetrating the eye and the brain when originating in sinuses, and causing widespread infarction because of its angioinvasive behavior. Depending on the underlying condition, such infections are often fatal (Roden et al. 2005), especially because these fungi tend toward resistance to current antifungals.

The Mucorales are an ubiquitous, ancient group mostly of saprotrophs, which, unlike the Ascomycota and Basidiomycota (the Dikarya), do not manifest sophisticated adaptations to diverse substrates, but rapidly use up available sugars and then move on via their sporangiospores (Hoffmann et al. 2013). Their evolutionary distance from the phyla of other human parasitic fungi is apparent in the structure of their hyphae, which are fragile, thin walled, and lack septa, and in the composition of their cell walls in which chitin and chitosan play the structural roles that glucans fulfill in the Dikarya (Dijksterhuis and Samson 2006). Some species like Mucor circillenoides are dimorphic and form yeast in the host (Dizbay et al. 2009; Khan et al. 2009; Lee et al. 2013), contributing to dissemination.

Among the Rhizopus spp., Rhizopus oryzae is responsible for ∼70% of human disease (Spellberg et al. 2005). Several of its virulence factors are areas of active investigation: its iron scavenging from the host (Fu et al. 2004; Ibrahim et al. 2008, 2010; Spellberg et al. 2009; Ibrahim 2011) and its ability to bind GRP78/BiP, a Kar2 homolog, on the surface of endothelial cells with its spore coat protein homolog CotH3 (Liu et al. 2010; Gebremariam et al. 2014). Like a broad phylogenetic variety of fungi, Mucorales are angioinvasive. Thigmotropism, the proclivity to sense and follow curvatures of a surface, has been shown for one species, Mucor mucedo (Perera et al. 1997), and is likely to be common to all. Much remains to be learned about the biology of these evolutionarily old fungi and historically new human pathogens.

CONCLUSION

Immunologically intact humans manifest robust defenses against fungal diseases. Recently, human social evolution produced scientific medicine, whose progress has rendered a large population susceptible to infections with fungi not considered human pathogens as recently as a hundred years ago. Human social evolution may have reached a stage where prioritization of the resources needed for understanding fungal biology and for successful development of multiple new classes of antifungals is possible, as the example of AIDS shows. Effective antiretroviral therapy was developed with a massive research effort less than two decades ago, and if human society evolves to a point of valuing all human lives equally, near elimination of AIDS is as feasible as the quasi elimination of HIV mother-to-child transmission has been in wealthy countries (Siegfried et al. 2011; Cohen et al. 2013; Nicol et al. 2013). Unlike with HIV, which has no reservoir in nature, humans will always have to cope with fungal infections, because potential fungal pathogens are part of our normal flora and of soil, water, and air. But building on the tremendous efforts and findings of the past decades, new low-toxicity therapies and novel preventative measures of fungal infections can be within our reach.