The ultimate purpose of defining IEI causally associated with IFDs is to develop a molecular immunological framework of human immunity to mycosis; such information may even shed light on fungal susceptibility in the more complex secondary immunodeficiencies. To achieve this, stringent criteria need to be explicitly considered to attribute penetrant, genetic susceptibility to fungal disease, and such criteria are proposed herein:
The certainty of the association between a gene defect and fungal susceptibility is considered “strong” when a sufficient number of cases have been reported. Consistency in the clinical observation of susceptibility implies a persistent, inherent immunological basis. Given the rarity of IEI, and the potential ubiquitous or restricted geographic distribution of the implicated fungi with resulting variability in terms of individual exposure, a threshold of three unrelated cases is proposed and used herein. Alternatively, especially in cases where the associated inborn error of immunity or mycosis is extremely rare (e.g., with few cases or a single case reported), the demonstrable immunologic defect must be congruent with the immuno-pathophysiology for that specific mycosis previously shown in other IEI. Thus, a strong association is established when three or more unrelated cases have been reported or, in ultra-rare cases, there is pathophysiologic homogeneity for that mycosis in other IEI. Below this threshold, cases are considered “sporadic”, implying the need for additional reported cases or immuno-pathologic mechanisms to buttress the certainty of the association.
Collectively, these criteria aim to bolster the association of distinct mycoses with different IEI to more confidently identify underlying pathophysiological frameworks.
4. Thermally Dimorphic Endemic Mycoses
Thermally dimorphic endemic fungi (TDEF) are saprophytes characterized by temperature-dependent morphological variation (i.e., they exist in a mould form at ambient temperature and assume a yeast-like phase at body temperature) and geographical restriction. These fungi include
Blastomyces,
Coccidioides,
Emergomyces,
Histoplasma,
Paracoccidioides,
Sporothrix, and
Talaromyces. Updates on the microbiology of these fungi have been recently reviewed
[7][8][9][10]. In their geographic niches, these fungi are environmentally distributed, and, based on previous seroprevalence or delayed-type hypersensitivity-based investigations, exposure is both frequent and generally asymptomatic or symptomatically self-limiting. On the other hand, some of these fungi are known to cause a variety of morbid syndromes in both overtly immunocompromised and apparently immunocompetent individuals. The field of IEI provides insight into the basis for the latter.
The first report of human blastomycosis in the Americas is attributed to Thomas Gilchrist, who presented the dermatological microscopic findings of a patient that was referred to him and, subsequently, detailed a second case of a chronically progressive, waxing-and-waning form of cutaneous blastomycosis
[11]. Blastomycosis is most commonly caused by the
B. dermatitidis complex (encompassing
B. dermatitidis and
B. gilchristii). Blastomycosis is said to be endemic to specific North American states and provinces that border the St. Lawrence River, the Great Lakes, the Mississippi River, and the Ohio River, although the geographic distribution of the causative fungi (and TDEF in general) is in flux owing to climate change and the human disruption of ecological niches
[12][13][14][15]. Infection occurs via the inhalation of aerosolized conidia. The most common clinical syndromes of blastomycosis include acute pneumonia, chronic pneumonia, and extra-pulmonary (disseminated) disease, occurring in 25–40% of patients, which can involve the skin, musculoskeletal system, genitourinary tract, and/or the central nervous system
[16].
Coccidioidomycosis (also known as San Joaquin Valley Fever, Valley fever, or Desert fever) is caused by two morphologically similar but genetically distinct fungi,
C. immitis and
C. posadasii. It was first described in 1888 by Alejandro Posadas and his mentor, Robert Wernicke, after their evaluation of Domingo Ezcurra, who had suffered from an indolently progressive but ultimately fatal form of the disease, and whose head is enshrined in a museum in Argentina
[17].
Coccidioides is predominantly endemic in the Southwestern USA and Central and South America. Infection occurs following inhalation of arthroconidial spores. Coccidioidomycosis can manifest as various lung diseases (e.g., acute pneumonia, chronic progressive fibrocavitary pneumonia, pulmonary nodules, or cavities) or, in <1 to 5% of those infected, as a disseminated disease; the most common sites of involvement of the latter are the CNS, skin, musculoskeletal system, and lymph nodes.
Emergomycosis is caused by
Emergomyces sp., whose genus was newly established in 2017 to include
Emmonsia-like fungi
[18]. Although disseminated disease has been reported in patients with advanced HIV or exogenous immunosuppression, no report of such mycosis associated with IEI has been identified.
In the Americas, histoplasmosis is typically caused by
H. capsulatum (and its variants), whereas
H. duboisii (restricted to Central and West Africa) causes African histoplasmosis. Although
H. capsulatum was originally thought to be restricted to certain areas (e.g., central and eastern states surrounding the Ohio and Mississippi River Valleys, as well as the St. Lawrence River valley), it is now recognized that different species from the
Histoplasma complex can be found worldwide
[19]. Indeed, the first description of histoplasmosis by Samuel Darling in 1905 was of a disseminated, fatal form in a 27-year-old Martinique man presenting febrile vomiting and whose autopsy showed leptomeningitis with the involvement of the spleen and marrow
[20]. Infection occurs via inhalation of small hyphal fragments and microconidia. Histoplasmosis can manifest as an acute pulmonary disease (varying from a mild/pauci-symptomatic form to severe/life-threatening forms); as complications associated with acute pulmonary disease that are thought to be immunologically mediated (e.g., erythema nodosum, arthritis, pericarditis, granulomatous, or fibrosing mediastinitis); as chronic cavitary pulmonary disease; or as disseminated disease
[21].
Paracoccidioidomycosis (PCM), also known as South American blastomycosis, was first described by Adolpho Lutz in Brazil in 1908 who identified the disease in two patients with oral mucosal lesions and cervical lymphadenopathy
[22]. Although PCM was attributed to
P. brasiliensis, phylogenetic studies have identified various species within the
Paracoccidioides species complex, although no distinction in clinical manifestations have been definitively established
[23]. Infection is acquired by the inhalation of conidia. PCM is classified into distinct forms: The regressive form is a-/pauci-symptomatic, involving the lungs. The progressive disease can be acute/subacute (primarily with lymphadenopathy, hepatomegaly, and/or splenomegaly) or chronic (involving the lungs, and, usually, the skin, mucosa, and/or viscera, such as the adrenal glands or CNS)
[23].
Sporotrichosis was first reported by Benjamin Schenck in 1898 who identified the disease through multiple refractory skin ulcers along the forearm lymphatics
[24]. Unlike other TDEF, sporotrichosis more commonly occurs following the soft tissue traumatic implantation of infectious propagules from an environmental source; inhalation is a less frequent route of acquisition. Caused by members of the
S. schenckii complex, sporotrichosis patients can present skin disease (cutaneous-lymphatic or nodular lymphangitis), disseminated disease (with visceral involvement), or extracutaneous presentations (e.g., CNS, musculoskeletal, or pulmonary)
[25].
Talaromycosis (formerly Penicilliosis) is caused by
T. marneffei (previously known as
P. marneffei), the only TDEF within the
Talaromyces genus that is pathogenic to humans. The fungus is notably endemic in South and Southeast Asia. It was first described by Gabriel Segretain and colleagues, who also reported the first human infected by the disease after accidental autoinoculation
[26][27][28]. The first report of a naturally occurring infection was in the form of disseminated disease (wherein the fungus was isolated from a removed spleen) in a patient with Hodgkin’s disease
[29]. Jayanetra and colleagues reported extensive disseminated disease in five patients, three of whom had no underlying disorder, causing death in three of the patients
[30]. Infection follows inhalation of spores from the environment; although the bamboo rat is a known natural reservoir of the fungus, the exact environmental source, and how the bamboo rat contributes to transmission to humans, is unclear
[31]. Given the fact that talaromycosis is a neglected tropical disease
[32], the understanding of its clinical manifestations is limited
[33]; thus, the prevalence of asymptomatic exposure is unclear. Pulmonary talaromycosis can include mass lesions, consolidation, cavities, effusion, or mediastinal lymphadenopathy. Patients with disseminated disease can present with skin lesions (which may be progressive and disfiguring), hepato/spleno-megaly, visceral infection, and/or CNS involvement.
The general pathophysiology for these mycoses is similar to that for mycobacteria: the inhalation of the infectious propagules results in their deposition in the alveoli, where they are ingested by alveolar macrophages. Under the influence of body temperature, the propagules convert to their yeast-like form intracellularly, disseminating to regional lymph nodes. The combination of innate and adaptive immunity attempts to contain the infection, wherein the formation of granuloma occurs. T-cell-mediated immunity is critical for this containment. The failure of this process results in systematic, progressive dissemination. This similarity to the pathophysiology of mycobacteria has been central to identifying IEI associated with TDEF.
Genetic defects in the capacity to produce or respond to interferon-gamma (IFN-γ) have been the IEI most frequently reported associated with TDEF mycoses. Most of these fall in the collection of disorders referred to as Mendelian Susceptibility to Mycobacterial Disease (MSMD)
[34][35][36]. While MSMDs were initially recognized for their distinct susceptibility to mycobacteria (particularly non-tuberculous (NTM) and occasionally
M. tuberculosis) and
Salmonella (predisposing victims to extra-intestinal salmonellosis)
[37][38][39][40][41][42][43], they were also subsequently found in patients with otherwise unexplained disseminated TDEF infection, including histoplasmosis (IL12RB1
[42][44][45][46]; IFNGR1
[47]), coccidioidomycosis (IFNGR1
[48] and IL12RB1
[49]), and paracoccidioidomycosis (IL12RB1
[50]). These findings provided the framework of human immunity to TDEF, which was followed by the identification of other IEI congruent with defective IFN-γ-mediated immunity, including CD40L deficiency
[51][52][53][54][55][56][57][58][59] (discussed further in the section on “Pneumocystis”), GATA2 deficiency
[60], and STAT1 GOF
[57][58][61][62].
The recognition that defective production of/response to IFN-γ underlies at least some forms of severe TDEF mycoses led to the idea that supplementation with IFN-γ, at least for those with diminished production, may be a therapeutic strategy for the curing of infections. Indeed, IFN-γ immunotherapy, used adjunctively with antifungals, has been successful as a salvage therapy for refractory coccidioidomycosis in two IEI patients
[48]. It has also been used successfully for this mycosis in patients without identified IEI
[63][64][65], although one case presented modest improvement with IFN-γ, and the identification of a dysregulated Th2 response with the subsequent addition of dupilumab (to block IL-4- and IL-13-mediated Th2 responses) led to resolution
[65]. Whether IFN-γ, either alone or with an inhibitor of dysregulated Th2 responses, may be successful for other TDEF remains to be reported.
Given the importance of the IFN-γ-immunity framework to TDEF, it was subsequently found that this pathway can be compromised by a distinct, non-monogenic process: the inherent presence of neutralizing auto-antibodies to IFN-γ (aa-IFN-γ). Again, the discovery of this immunologic defect underlying susceptibility to an infectious disease was first found in patients with a mycobacterial disease, namely, tuberculosis
[66]. Subsequently, high-titer aa-IFN-γ concentrations were found in patients with no other immunocompromising condition in association with other severe disseminated infections, including NTM
[67][68][69][70][71], and germane to TDEF,
H. capsulatum, and T.(
P.)
marneffei [72][73][74][75][76][77][78][79][80][81][82][83][84][85][86] (as well as
Cryptococcus, see below). aa-IFN-γ is an emerging, adult-onset form of immunodeficiency that is typically recognized in the fourth to sixth decades of life. An ethnic predisposition among Southeast Asians is noteworthy: these autoantibodies have been primarily reported among those of Filipino, Thai, Vietnamese, Japanese, and Chinese ancestry, with other ethnic groups less commonly affected
[87][88]. A female predominance has been shown
[85], as has a genetic association with the HLA alleles, DRB1*16:01 and DQB1*05:02
[89][90]. aa-IFN-γ inhibit STAT1 phosphorylation and IL-12 production. Thus, in patients (particularly women of Southeast Asian descent) with TDEF disease, particularly if severe or disseminated, aa-IFN-γ should be assessed. Of note, aa-IFN-γ were originally discovered in patients living with HIV (albeit without any clear association with infection susceptibility)
[91], implying that HIV status and aa-IFN-γ are not mutually exclusive in terms of their existence and, possibly, compromising effects on immunity. The management of aa-IFN-γ has been challenging; standard broad-spectrum immunosuppressants (e.g., steroids) or immunomodulatory dosing of IVIG have been unsuccessful. Cyclophosphamide has been used with some success
[84][92]. Given that this immunodeficiency is due to the presence of rogue autoantibodies, drugs specifically depleting B cells, such as Rituximab (anti-CD20) or Daratumumab (anti-CD38), have been reported to be effective in terms of reducing aa-IFN-γ levels and clearing infection
[84][93].
5. Cryptococcosis
Cryptococcosis is principally caused by yeast members of the
C. neoformans or
C. gattii species complexes that have different geographic, microbiologically phenotypic, and genotypic characteristics
[94][95]. Cryptococci also have a distinct host-range
[96]: The
C. neoformans species complex (herein referred to as “
C. neoformans”) has traditionally been viewed as an opportunistic pathogen in various immunocompromising conditions, notably those of HIV and transplant recipients. Meanwhile, the
C. gattii species complex (herein referred to as “
C. gattii”) tends to cause disease in apparently immunocompetent people. The distinction between these species complexes is relatively recent and precludes interpretation of which cryptococci were identified in previous reports of IEI patients. Thus, the instances wherein
C. gattii has been specifically identified from such cases will be indicated.
The molecularly defined IEI associated with severe or disseminated cryptococcosis strikingly overlap with those that predispose one to infection with TDEF, namely, MSMD
[97][98][99], GATA2
[100][101], CD40L deficiency
[52][102][103][104][105][106][107][108][109], STAT1 GOF
[110][111][112], and DN-STAT3
[113][114]; these IEI have been detailed above. This overlap most likely reflects a comparable, if not homologous, immunopathophysiology. It also underlines a distinct feature of the genotype/phenotype correlations of IEI: the infection phenotype requires microbial exposure, which may be highly variable between individuals with the same lesioned gene, resulting in variable expressivity, at least of the mycoses. It should also be noted that cryptococcosis can be seen in “idiopathic CD4
+ lymphocytopenia” (ICL), an immunologic phenotype that likely reflects the convergence of heterogeneous causes; in the absence of a clear genetic etiology, ICL is not included among these IEI, although clinicians should be aware of this possibility during their investigations.
As with the TDEF, aa-IFN-γ have also been found in patients with unexplained cryptococcosis
[75][115][116]. The characteristics of aa-IFN-γ are detailed above (see “Thermally dimorphic endemic mycoses”). Compared to HIV-infected patients, these patients with aa-IFN-γ tended to be less likely to present with meningitis and were more likely to present with bone and/or joint and lung/pleura as well as skin infections
[116].
In addition to aa-IFN-γ, high-titer, neutralizing auto-antibodies to GM-CSF (aa-GM-CSF) have been found in patients who were not overtly immunocompromised but who developed cryptococcal meningo-encephalitis that, distinctly, was associated with
C. gattii [117][118]. aa-GM-CSF were first identified in patients with pulmonary alveolar proteinosis (PAP), a disorder characterized by dyspnea, alveolar infiltrates (determined radiologically), and a risk for progressive respiratory failure
[119][120]. PAP results from the unregulated accumulation of a surfactant in the airways due to defective alveolar macrophage clearance
[119]. PAP had been previously reported to be occasionally complicated by opportunistic infections, notably cryptococcosis (manifesting as the pulmonary form alone
[121] or with systemic disease
[122][123]), thus potentially linking aa-GM-CSF to cryptococcal susceptibility. GM-CSF is a pleiotropic cytokine involved in numerous functions, including the terminal differentiation of monocytes into mature alveolar macrophages, the differentiation of DC, and enhanced myeloid antimicrobial functions
[124]. aa-GM-CSF inhibit the GM-CSF-induced STAT5 phosphorylation of peripheral blood mononuclear cells in vitro, although the exact pathophysiology of how they facilitate disseminated disease remains to be deciphered. Cryptococcosis (and other opportunistic infections) can precede, occur concomitantly with, or follow the clinical development of PAP. Intriguingly, the loss of the GM-CSF pathway is associated with susceptibility to another neurotropic infection (i.e., nocardiosis
[125][126]), and the selective loss of GM-CSF response by PBMC to
C. albicans may also contribute to candidal invasion of the CNS through CARD9 deficiency
[127][128] (see “Invasive Candidiasis” above), suggesting a role for GM-CSF in systemic protection against CNS infection. The underlying basis for the development of aa-GM-CSF is unclear. In addition, while both aa-IFN-γ and aa-GM-CSF underlie susceptibility to “cryptococcosis”, it is unclear from the current published reports whether these auto-antibodies segregate with mutually exclusive susceptibility to the species complex (e.g., aa-IFN-γ to
C. neoformans vs. aa-GM-CSF to
C. gattii only) and whether these auto-antibodies can co-exist in patients.
Intriguingly, historical reports of patients with PAP complicated by histoplasmosis have also been reported
[123][129], although no reports of aa-GM-CSF underlying human cases of histoplasmosis have been identified.
6. Pneumocystis
Pneumocystis was discovered in 1909 and, at the time, was thought to be a protozoan parasite (
Trypanasoma); genomic analysis resulted in its subsequent re-classification as a fungus in 1988
[130]. Within this fungal genus are several species, each of which demonstrate strict target–host species specificity, that is, they are able to infect only one host species. Thus, humans can be infected with
Pneumocystis jirovecii and are, consequently, the only source capable of infecting other humans, while the previously more popular term “
P. carinii” is the formal name for the species that infects rats
[131].
P. jirovecii pneumonia (Pjp) was initially reported in the late 1930s in Europe, with subsequent clusters reported in the 1940s and 1950s, as an unusual pulmonary infection characterized by a prominent infiltration of mononuclear cells thought to be plasma cells (hence the term “interstitial plasma cell pneumonia”)
[132][133]. Strikingly, these epidemics affected premature and malnourished (“dystrophic”) infants, usually between 6 weeks and 4 months of age, for which mortality rates were as high as 50% and wherein autopsy series noted pneumothorax, mediastinal emphysema, and alveolar exudate filling the gray lungs. While the immunological cause of these Pjp outbreaks in infants remains unclear, Pjp was recognized as a potential cause of fatal infection in SCID through the Vetter family (with death of an older brother) and other reports
[134][135][136], and clusters of Pjp in previously healthy adults in the 1980s heralded the HIV/AIDS epidemic
[137][138][139]; this provided the first evidence that profound cell-mediated immunodeficiency was conducive to the development of disease.
Exposure to
P. jirovecii seems to be common: based on serology data of healthy US children, two-thirds had detectable
Pneumocystis antibodies by the age of 4 years
[140]. Using molecular and serologic techniques, it was found that asymptomatic exposure was frequent in Chilean children
[141]. Infection is thought to occur primarily following the person-to-person transmission of asci (cysts) via respiratory secretions. As with other fungi,
P. jirovecii is not a professional pathogen, as asymptomatic colonization has been documented at high frequencies
[142]. The adherence of trophozoites to alveolar type I pneumocytes initiates the infectious process, which progresses and may ultimately be fatal in immunocompromised patients. The exact immunopathophysiology of Pjp is a challenge to study owing to, among other things, the difficulty of culturing the organism. However, the natural history of Pjp in IEI underscores the importance of T cell immunity with respect to mitigating disease.
Combined immunodeficiency (CID) refers to a diverse group of IEI marked by defective cellular and humoral immunity. In its most severe form (Severe CID (SCID)), manifestations occur in infancy and typically include failure to thrive, recurrent infections, and opportunistic infections. Pjp can be a life-threatening infection in these IEI: prior to the routine use of prophylaxis, the overall Pjp infection rate in SCID was at least 27%, an estimated attack rate of at least 0.34 episodes per patient-year, with a true infection rate of 42%
[135]. In some cases, Pjp may even be the sentinel manifestation of SCID
[143]. SCID is fatal unless corrected early in life with hematopoietic stem cell transplantation (HSCT) or gene therapy (GT). In distinction to SCID, CID is survivable beyond infancy. While the associated defects may not be as profound as those in SCID, CID may also be associated with an increased risk of Pjp.
Pjp has been reported in individual cases of GATA2
[144], XLA
[145][146], and DOCK8
[147] but without any mechanistic evidence attributing the mutant gene to disease. If these observations consistently recur, they may provide much-needed granular data on the mechanisms of susceptibility.
7. Aspergillosis
Aspergillosis refers to a group of infections caused by molds of the genus
Aspergillus and can be divided into three broad categories
[148][149]: allergic bronchopulmonary aspergillosis (ABPA); invasive aspergillosis (IA); and chronic pulmonary aspergillosis (CPA). CPA refers to a heterogeneous group of conditions that are either indolent (>3 months) in seemingly immunocompetent individuals who have underlying structural lung disease (including ‘fungus ball’/aspergilloma, an isolated aspergillosis nodule, or chronic cavitary pulmonary aspergillosis (CCPA)) or progressive (worsening < 3 months) in moderately immunocompromised patients (i.e., subacute invasive aspergillosis (SAIA), also called chronic necrotizing pulmonary aspergillosis (CNPA)). It has been estimated that humans inhale at least several hundred
Aspergillus spores daily
[150]. Thus, the inhalation of conidia into the lower respiratory tract explains why the lungs are the primary site of disease. However, extra-pulmonary disease can occur, either as a contiguous extension or dissemination from a pulmonary portal of entry or, presumably, by inoculation that is localized and/or of a high burden (e.g., in the sinuses, gastrointestinal tract, skin, and soft tissue). Among the genetic disorders associated with these aspergillosis syndromes, ABPA is primarily seen with CFTR mutations, while IA and CPA are largely seen in chronic granulomatous disease (CGD), autosomal dominant hyper-IgE (Job’s) syndrome, GATA2 deficiency, and CARD9 deficiency.
8. Dermatophytosis
Dermatophytes are hyalinated moulds of the Arthrodermataceae family with the capacity to invade keratinized tissue (i.e., skin, hair, and nails), causing “dermatophytosis” (colloquially referred to as “ringworm” or “tinea”). Dermatophytes were traditionally classified into three genera:
Trichophyton,
Epidermophyton, and
Microsporum. Phylogenomic studies have since revised the taxonomy into seven genera:
Trichophyton,
Epidermophyton, and
Microsporum, as well as
Paraphyton,
Lophophyton,
Arthroderma, and
Nannizzia [151][152]. Infection is acquired by the deposition of spores from humans, animals, or soil, accounting for anthrophilic, zoophilic, or geophilic spread, respectively. In immunocompetent individuals, infection is restricted to the stratum corneum, the outermost layer of the epidermis that contains a thickened layer of dead keratinocytes.
Despite their broad distribution and their affinity for a ubiquitous compound (keratin), dermatophytes are not professional pathogens. Previous serial sampling data have shown that these moulds can be isolated from humans for prolonged periods of time (months to years) in the absence of symptomatic infection
[153]. Similarly, intra-familial transmission has been noted to be non-uniform. Thus, dermatophytes, like other microbes, demonstrate a spectrum of symptoms, ranging from asymptomatic infection and active disease to chronic disease. Chronic dermatophytosis has been previously defined as disease that is refractory to therapy, relapsing once treatment is stopped, and/or progressive
[154]. Based on the genetic disorders described below, one could further separate ‘chronic dermatophytosis’ into two distinct chronic syndromes: one that manifests as a relapsing or recalcitrant, superficial disease (herein termed “chronic superficial dermatophytosis” [CSD]), and the other marked by superficial disease associated with invasion into deeper, non-keratinized tissue (termed “deep dermatophytosis” (DD)).
As with CMC, genetic disorders of epithelial integrity predispose one to CSD. Although numerous cases of various ichthyosis disorders (e.g., ichthyosis vulgaris, lamellar ichthyosis, etc.) associated with dermatophytosis have been reported, dermatological diagnoses have been made clinically or, occasionally, via skin biopsy and without genetic confirmation
[155][156][157][158][159][160][161][162][163][164][165][166][167][168][169]; following the proposed criteria above, these cases are not considered as definitive in this research. On the other hand, KID syndrome (due to dominant mutations in
GJB2—which encodes Cx26—and conferring susceptibility to CMC (see above)) has also been associated with an increased risk for extensive CSD
[170][171][172][173]. Some forms of palmoplantar keratoderma (PPK; also called ‘tylosis’) were also noted to be associated with increased rates of CSD
[154][174]. PPK is a heterogeneous group of disorders characterized by the thickening (hyperkeratosis) of the palmar and plantar skin. Autosomal-dominant diffuse PPK can occur due to dominant negative mutations in keratin 9 (
KRT9)
[175], keratin 1 (
KRT1)
[176], or aquaporin 5 (
AQP5)
[177]. In previous studies from Norbotten, Sweden, the reported prevalence of associated dermatophytosis varied between 35 to 65%
[178]. Keratins are a multi-gene family of proteins that form intermediate filaments, which are the major cytoskeletal proteins in epithelial cells. Mutations in
KRT9 function through dominant negative interference with the assembly of these filaments, and their specific contributions to PPK are due to the exclusive expression of KRT9 in the epidermal keratinocytes of the palms and soles
[179]. It has been proposed that mutations in
KRT1 interfere with keratin intermediate filament interactions with other structures (e.g., desmosomes) or with the cross-linking of the cornified envelope
[176]. AQP5 is expressed in the granular layer of the skin; PPK-causing GOF mutations disrupt keratinocyte water barrier function in the epidermis of the palms and soles
[177][180]. Thus, selective defects in the barrier functions of the skin may confer susceptibility to CSD.
CSD, which primarily occurs due to
Trichophyton spp. or
Microsporon spp., has also been reported in STAT1 GOF
[110][181]. This inborn error of immunity also predisposes one to CMC (see above). Whether this reflects the convergence of host defense pathways to mitigate these two superficial mycoses, or whether STAT1 GOF has a distinct mechanism of susceptibility that implicates skin barrier function, remains to be explored.
Sentinel descriptions of chronic, progressively invasive dermatophytosis were made in the late 19th century when Majocchi described granulomatous dermal disease caused by
Trichophyton [182]. Subsequently, rare cases were reported of chronic generalized trichophytosis involving the lymph nodes, muscles, bones, brain, or other viscera, including familial cases. One of the earliest documented cases of a family with fatal dermatophytosis was reported in 1975 involving a young Algerian girl who had extensive cutaneous disease at 14 years of age that progressed over 8 years
[183]. She had two brothers, each of whom died at 11 and 13 years old, respectively, from a dermatophytic disease involving the skin, hair, and viscera that was caused by
T. schönleini, which was also identified from her skin lesions. Two other brothers were being treated for dermatophytosis (no further details were provided in this case), leading to the suspicion of a genetic immunodeficiency with susceptibility to dermatophytes. Further immunologic investigations at the time revealed intact humoral immunity but variably depressed cell-mediated immunity
[184]. Indeed, Marill observed three key characteristics of what he termed “maladie dermatophytique” (French for “dermatophytic disease”)
[185]: familial predisposition; deficiency of cell-mediated immunity; and unusual severity, with one patient developing a cerebral abscess and the other patients having died from the illness.
The breakthrough in understanding the underlying basis for this disorder came with the discovery of biallelic mutations in
CARD9 that were found in 17 patients of Algerian, Tunisian, or Moroccan descent with deep dermatophytosis (DD)
[186]. Thus, AR CARD9 deficiency not only predisposes one to CMC or extra-pulmonary aspergillosis (discussed above) but to DD as well. The vast majority of cases are caused by
Trichophyton sp. (e.g.,
T. rubrum,
T. violaceum, and
T. mentagrophytes), although
Microsporum sp.
[187] may occasionally be the culprit. DD can be quite destructive, with painful, non-healing cutaneous ulcers that often fail to resolve with antifungal therapy or risk relapsing once antifungal therapy is stopped. Importantly, DD can also manifest with lymphadenitis or visceral abscesses, from which fungal elements can be seen on biopsy and the dermatophyte is cultivable from tissue.
The mechanism(s) by which CARD9 mutations lead to seemingly disparate fungal diseases is not clear. Although CARD9 LOF mutations do impair IL-17 responses, the absence of DD in other IEI marked by IL-17 deficiency (which manifest with CMC, as presented above), other than a singular report of DD in a patient harboring a DN-STAT3 mutation
[188], implies that more specific immunologic mechanisms need to be deciphered to understand how CARD9 mediates dermatophytic disease.
9. Pheohyphomycosis (Dematiaceous Fungi)
Pheohyphomycetes (dematiaceous fungi) are environmentally distributed moulds that are darkly pigmented due to the presence of melanin in the fungal cell wall. The host–fungal interaction can result in three human syndromes: eumycotic mycetoma, chromoblastomycosis, and phaeohyphomycosis
[189]. Eumycotic mycetoma is a chronic cutaneous/subcutaneous infection characterized by the triad of ‘tumefaction’ (i.e., the indolent appearance of inflammatory nodules/mass with fibrosis); the penetration of sinuses or fistulae into deep tissue; and presence of granules in tissue or discharge
[190]. Chromoblastomycosis is a chronic cutaneous/subcutaneous infection characterized by the presence of fungal muriform cells (also called ‘sclerotic bodies’ or ‘copper pennies) in tissue. Pheohyphomycosis refers to infections caused by these dematiaceous fungi that are histologically distinct from the two other syndromes due to the presence of yeast-like cells and filamentous growth in tissue; additionally, disease is not necessarily restricted to cutaneous/subcutaneous organs but can be disseminated (including with the involvement of CNS). Importantly, the manifestation of the clinical syndrome resulting from the infection with these fungi is driven by the immune status of the patient.
The identification of dematiaceous fungi in a diagnostic microbiology laboratory traditionally involves the assessment of the sample fungus’s growth rate, its colony and microscopic morphological characteristics, and the execution of biochemical tests. Since different medium and culture conditions can lead to different morphologic forms for a given fungus, this approach has led to some confusion about nomenclature. Genomically based identification has emerged as a more accurate method of identification.
Superficial disease due to dematiaceous fungi is commonly restricted to global regions of warmer climates and lower latitudes. Unfortunately, this has hampered the understanding of underlying fungal pathophysiology. The classical paradigm has been that these fungi are widespread in the environment, including in soil and plants, and that trauma, even minor forms, account for the ensuing disease. Of course, this explanation is both simplistic and overlooks a clear knowledge gap: minor trauma must occur quite frequently in the respective socio-economic context, yet disease does not. A single case of CARD9 deficiency causing chromoblastomycosis due to
Phialophora expanda (within the
P. verrucosa complex) has been reported
[191]. Although an individual case, the authors elegantly buttressed this report with a mouse model replicating the disease and strongly suggested that CARD9 should be assessed in more cases of chromoblastomycosis.
Pheohyphomycosis has also been known for a long time but has remained enigmatic. Indeed, Marill was among the first to describe a “deep dematiaceous disease” due to
Hendersonula toruloidea (a fungus at the time known to cause disease in fruit trees, which is now called
Neoscytalidium dimidiatum) in an Algerian 30-year-old male who had, intriguingly, been diagnosed with a “deep dermatophytosis” at the age of 15 year and treated intermittently with griseofulvin
[192]. Based on his investigations at the time, Marill suspected an immunological disorder. Nearly 35 years later, the first evidence to support his hypothesis was confirmed with the identification of AR CARD9 deficiency in four patients with pheohyphomycosis due to
P. verrucosa [193]. This inborn error of immunity has since been identified in pheohyphomycosis due to
Exophiala [194][195][196][197], other
Phialophora species
[198],
Aureobasidium [199],
Alternaria [200][201], and
Exserohilum [202], as well as more esoteric dematiaceous fungi, including
Pallidocercospora crystallina,
Ochroconis musae, and
Corynespora cassiicola [195][203][204][205]. Intriguingly, disseminated infection with
C. cassiicola (with extension into the CNS) has also been reported in a patient with compound heterozygous mutations in
CLEC7A (encoding Dectin-1, which is a fungal pattern recognition receptor that signals through CARD9)
[206]. Other pheohyphomyces have also been reported with single heterozygous variants in
CLEC7A or
CARD9, although with distinct fungal etiologies than those for AR CARD9 deficiency
[206]. Whether this distinction ostensibly reflects other genetic loci capable of influencing susceptibility to infection or severity to dematiaceous fungi or reflects inherent differences in fungal pathogenicity, or both, remains to be determined.
The management of pheohyphomycosis associated with AR CARD9 deficiency is quite heterogeneous, and clinical outcomes are not consistently reported. Various antifungals, either administered alone or in combination, have been used either as a sole therapy or in conjunction with surgical resection. This fungal disease is not invariably fatal, although relapses may occur. No reports of adjunctive cytokine therapy, as in AR CARD9 deficiency with invasive candidiasis (see above section), have been identified.
10. Mucormycosis
Fungi of the Mucorales order have aseptate (or pauci-septated), ribbon-like hyphae. The classical, acquired predisposing factors for infection with this fungus include neutropenia associated with chemotherapy for hematological malignancy, diabetes mellitus (particularly with poor glucose control and ketoacidosis), trauma, and iron chelation therapy. On the other hand, Mucormycosis has not been stereotypic or pathognomonic with respect to any single IEI. In CGD, spontaneous mucormycosis is exceedingly rare
[207][208]. Most reported cases only occurred in patients receiving significant immunosuppression for several weeks
[209], following HSCT
[210][211], complicating anti-TNF-α treatment
[212], or following invasive procedures
[213][214]. However, an isolated case has been reported in the absence of immunosuppression
[215]. The aggregate of these cases would suggest that defects in the phagocyte NADPH–oxidase system are not sufficient in terms of susceptibility to these moulds. There have also been reports of mucormycosis linked to AR CARD9 deficiency, wherein every case corresponded to exclusively cutaneous disease and was caused by
Mucor irregularis (formerly
Rhizomucor variabilis var.
variabilis)
[216][217][218]. Interestingly,
M. irregularis is distinct among the Mucorales due to its tendency to cause chronic cutaneous/subcutaneous infection without angio-invasion and deeper dissemination, although the disease can be mutilating
[219]. What is equally interesting is that all three reported CARD9 deficiency cases occurred in Chinese patients, and cases of the chronic skin disease caused by
M. irregularis have exclusively been reported in Asia, particularly China
[219], perhaps reflecting some aspect of selective evolutionary pressures. Mucormycosis has also been singularly reported in association with GATA2 deficiency (although it is not clear if this case occurred during chemotherapy or HSCT)
[220] and STAT1 GOF
[221]. Clearly, more reports of native mucormycosis in IEI would significantly guide the understanding of human immunity to these moulds.