Immunity to Cryptosporidium: Lessons from Acquired and Primary Immunodeficiencies

Protozoan parasites of the genus Cryptosporidium infect a wide range of vertebrates, and in humans, Cryptosporidium hominis and Cryptosporidium parvum are the main species that can cause disease, the latter of which can be zoonotic (1). Transmission occurs through ingestion of oocysts, which release sporozoites that infect intestinal epithelial cells (IECs) (2). The parasites then replicate asexually in an intracellular but extracytoplasmic niche present at the luminal surface of IECs, and lysis of the host cell releases merozoites that reinvade other IECs (2). Death of host cells and disruption of IEC function associated with alterations in gut physiology are likely the major underlying cause of local pathology. After three rounds of asexual replication, obligate sexual reproduction leads to generation of more oocysts, which can hatch and reinfect within the same host or be shed in feces for transmission (3). Because of this cycle of infection and reinfection, Cryptosporidium can complete its entire life cycle in a single host. Although this organism was first described in mice in the early 1900s, it was not until the 1970s that case reports emerged that associated it with human disease, most commonly in immune-suppressed patients (4). Since then, there has been recognition that Cryptosporidium frequently causes disease even in immunocompetent adults and has significant effects on long-term health of children (5). Despite this importance to public health, many questions remain about the immune basis for parasite control.

Epidemiology and experimental infections of humans have helped to understand how immunity to Cryptosporidium develops. In areas where Cryptosporidium is endemic, children younger than 2 y are most affected, where infection can lead to stunting and malnutrition that predisposes to recurring infection and severe disease (6–8). The observation that milder symptoms occur with subsequent infections provides evidence for development of protective immunity (9, 10). Experimental primary challenge in healthy human volunteers revealed the development of Cryptosporidium-specific IgA and IgG responses (9, 11). However, studies in mice have suggested that B cells are dispensable for protection to primary infection (12–14). The cytokine IFN-γ is a major mediator of resistance to numerous intracellular bacterial and parasitic infections, and the use of intestinal biopsies in human challenge studies identified increased production of IFN-γ (15). This observation was complemented by foundational studies that used immunocompromised mice to show that immunity depends on the cytokines IL-12, IL-18, and IFN-γ, as well as T cells (particularly CD4⁺) (12, 16, 17). The use of mouse-adapted strains of C. parvum or mouse-specific C. tyzzeri has provided more tractable systems to study intestinal immunity to this pathogen (14, 18, 19). Thus, we now know that in IECs, the inflammasome NLRP6 promotes processing of IL-18, which synergizes with dendritic cell (DC)-derived IL-12 to drive IFN-γ production by innate lymphoid cells (ILCs) and T cells (Fig. 1) (18–21). IFN-γ can then act on IECs to activate the transcription factor STAT1 that, in turn, promotes antimicrobial activity required to restrict the parasite (18). Although these studies emphasize the role of Th1-type responses for parasite control, there is also evidence in the mouse system for a prominent IFN-γ–independent mechanism of restriction (12, 14). A better understanding of the relative contribution of IFN-γ–dependent and –independent pathways in resistance to Cryptosporidium is needed, and understanding which immune deficiencies lead to increased risk for cryptosporidiosis may offer clues as to the host pathways that restrict Cryptosporidium that may also be relevant to other enteric pathogens.

In immunocompetent individuals, Cryptosporidium infects the ileum and can cause diarrhea and enteritis, which typically resolve without treatment (1). Severe cryptosporidiosis may warrant the use of the antiparasitic drug nitazoxanide, which can moderately reduce time to recovery in adults and children and is the only Food and Drug Administration–approved medication for cryptosporidiosis. In the immunodeficient host, the failure to clear Cryptosporidium results in chronic disease, which can manifest as biliary involvement and progress to sclerosing cholangitis associated with scarring and blockage of the bile ducts. Other extraintestinal manifestations include respiratory infection (likely a consequence of inhalation of aerosolized droplets containing oocysts), gastric disease, and pancreatitis (5, 22). Unfortunately, in immunocompromised individuals, nitazoxanide is ineffective at resolving diarrhea and oocyst shedding (23, 24). This observation suggests that the ability of this drug to inhibit parasite replication may provide a window for the immune response to clear infected cells. Alternatively, in some models of viral infection, nitazoxanide enhances the ability of cytosolic sensors of microbial presence to promote the production of type I IFNs, suggesting the drug’s mechanism of action may be partly immunomodulatory (25).

The use of immunosuppressive drugs to treat certain cancers or prevent transplant rejection provided the first examples of how acquired immune deficiencies were associated with Cryptosporidium infection. Subsequently, the AIDS pandemic emphasized the critical role of T cells in resistance to Cryptosporidium, and the relatively high incidence of this opportunistic infection in these patients underscored that exposure to this organism was perhaps more frequent than anticipated.

Many of the treatments used to manage cancer, such as steroids, whole-body radiation, and certain chemotherapies that double as immunosuppressive drugs, impair T cell function and predispose to Cryptosporidium infection (26–29). Thus, the prevalence of Cryptosporidium in chemotherapy recipients can be as high as 17%, with a heightened risk for severe disease in individuals with hematologic malignancies, especially in children (30–32). In the context of hematopoietic stem cell transplant (HSCT), in some cases, Cryptosporidium infection can be mistaken for graft-versus-host disease in the gut (33). Likewise, in solid organ transplant recipients, the standard of care to prevent allograft rejection involves treatment with immunosuppressive therapies that target T cells, which in turn predisposes to severe cryptosporidiosis (34, 35). The observation that in these patients the withdrawal of suppressive therapy usually results in parasite control highlights the close relationship between sustained immunosuppression and heightened risk for Cryptosporidium infection (35).

Untreated HIV infection causes a loss of CD4⁺ T cells and can lead to AIDS, which is characterized by life-threatening opportunistic infections and cancers. Increased risk for chronic Cryptosporidium infection occurs at a CD4⁺ T cell count of <200 cells/μl, which represents a 60–85% reduction in the normal circulating CD4⁺ T cell population (36). In this setting, cryptosporidiosis can lead to persistent severe diarrhea and AIDS cholangiopathy, defined by infection-related strictures of the biliary tract that cause obstruction (1, 37, 38). Initiation of antiretroviral therapy and restoration of CD4⁺ T cell counts are associated with clearance of the parasite (39, 40). This direct relationship between CD4⁺ T cell number and control of Cryptosporidium informs that T cells are critical in control of Cryptosporidium but does not indicate the molecular and cellular pathways that are required to generate and sustain relevant T cell responses.

Primary immunodeficiencies (PIDs; also known as inborn errors in immunity) are rare, genetic (often monogenic) disorders that impair the immune system and predispose to severe infections, autoimmunity, and some cancers. More than 450 genetic mutations have been identified that result in PIDs ranging from global defects in the development of individual immune cell populations to more specific loss- or gain-of-function mutations in immune pathways (41). Until recently, identification of individuals with PIDs was possible only after the onset of clinical symptoms, and identification of the underlying defect was a challenge. Associations between PIDs and Cryptosporidium infection emerged in early case reports, with studies that described Cryptosporidium in individuals with common variable immunodeficiency (42), hypogammaglobulinemia (43), and nonspecified SCID (44). In many instances, whether the noted immune defect was responsible for increased susceptibility was unclear. For example, one report linked Cryptosporidium with IgA deficiency (45), but specific IgA defects are a common immune deficiency but are not typically associated with susceptibility to Cryptosporidium. Likewise, Cryptosporidium was identified in an individual with a global defect in the production of IFN-γ, but the molecular basis of this immune deficiency was unclear (46). Regardless, this clinical experience indicates that chronic Cryptosporidium infection may herald the presence of a PID. Which monogenic PIDs render hosts susceptible to Cryptosporidium (described later and in Table I) reveal critical pathways involved in immune-mediated control of this organism.

There are numerous studies that implicate human CD4⁺ and CD8⁺ T cells in the control of Cryptosporidium (47–51), and several PIDs are associated with altered expression of MHC class I (MHCI) and II (MHCII) molecules required for Ag presentation to these T cells. However, it is notable that defects in MHCI, causing bare lymphocyte syndrome (BLS) type I and reported in just a handful of patients worldwide, have not yet been linked to Cryptosporidium (52, 53). In contrast, BLS type II, characterized by absence of MHCII expression, is an autosomal recessive disorder attributed to ∼200 patients worldwide that is associated with debilitating Cryptosporidium infection (53–55). MHCII deficiency is caused by defects in one of four MHCII-specific transcription factors—RFXANK, MHC2TA (MHCII transactivator, CIITA), RFX5, or RFXAP—and is associated with hypogammaglobulinemia and decreased numbers of CD4⁺ T cells (54, 56–59). In one study of 35 patients with MHCII deficiency, C. parvum was found in 34% of patients between the ages of 7 mo and 10 y, and the majority of these patients suffered progressive liver failure (55).

The importance of MHCII in control of Cryptosporidium is consistent with its role in Ag presentation to CD4⁺ T cells. Its expression by thymic epithelial cells is important for selection and generation of naive CD4⁺ T cells, whereas on APCs it is important for priming of naive CD4⁺ T cells and for effector T cell responses (60). “Atypical” APCs, including IECs, can also express MHCII, which has been implicated in graft-versus-host disease, regulation of stem cell renewal, and generation of innate-like mucosal T cells present in the small intestine (61–66). Importantly, expression of MHCII in IECs is regulated by the transcription factor CIITA that is a prominent target of IFN-γ (67). Cryptosporidium has been reported to disrupt the IFN-γ–mediated increase in CIITA expression in IECs in vitro, and in vivo IECs infected by the parasite have lower MHCII expression than bystander cells (14, 68). It is unclear what implication this has for control of Cryptosporidium, particularly given that some studies show that MHCII on IECs can activate effector CD4⁺ T cells, whereas other studies have implicated IEC expression of MHCII as promoting immune tolerance (67). Therefore, although the susceptibility of individuals with MHCII deficiency is most likely due to the absence of CD4⁺ T cell responses, it remains an open question whether IEC-expressed MHCII contributes to local activation of Cryptosporidium-specific CD4⁺ T cells.

All hyper-IgM (HIGM) disorders are characterized by the inability of B cells to undergo class-switch recombination, which results in high serum levels of IgM and low IgG (69). In humans, the most common form of HIGM is a genetic defect in the CD40L (CD40L; CD154) gene causing HIGM1 (70). Because this recessive mutation is carried on the X chromosome, HIGM1 predominantly affects males and accounts for roughly 70% of all instances of HIGM (70). CD40L is expressed by activated CD4⁺ T cells, and its ability to bind to CD40 expressed on B cells provides a critical second signal mediated through NF-κB pathways that promotes IgH class switching from IgM to IgG (69, 71). In two studies of individuals with CD40/CD40L deficiencies, 21 and 60% were affected by chronic Cryptosporidium in the United States and Europe, respectively (72, 73). Autosomal recessive defects in other genes such as activation-induced cytidine deaminase (HIGM2), CD40 (HIGM3), and uracil N-glycosylase (UNG, HIGM5) can also cause HIGM (74–77). It is informative that in individuals with pure class-switching defects (HIGM2/5) that are independent of CD40/CD40L there are no reports of increased susceptibility to Cryptosporidium (12–14).

The ability to compare the different types of HIGM disorders indicates that Cryptosporidium susceptibility in CD40/CD40L deficiencies is unlikely to be due to impaired humoral responses and is consistent with experimental studies in murine models of B cell deficiency. Together, these clinical and experimental observations indicate that the ability of the CD40/CD40L pathway to promote cell-mediated immunity is required for parasite control. Thus, CD4⁺ T cell expression of CD40L would allow them to interact with CD40 on two cell types relevant to Cryptosporidium: DCs and IECs. Engagement with CD40 on DCs promotes their expression of IL-12 and thereby drives IFN-γ production from CD4⁺ and CD8⁺ T cells (78–81). In addition, CD40L itself can be shed, and direct engagement of CD40 on epithelial cells infected with Cryptosporidium limited parasite growth in vitro, potentially by promoting host cell apoptosis or autophagic processes that contribute to parasite death (82–86). However, chimeric mouse models in which IECs lack CD40 suggest that its expression on IECs alone is not sufficient for clearance of Cryptosporidium (83). It is possible that CD40 expression by DCs and by IECs both contribute to control of Cryptosporidium, where the former is important for initiation of T cell responses and the latter for effector functions. The clinical experience with these patients gives some support to this model, and two patients with HIGM1 (lack CD40L but have CD40) with chronic Cryptosporidium infection were treated with an agonistic Ab against CD40 to determine whether this would be sufficient to restrict the parasite (87). Monocyte-derived DCs from these HIGM patients increased IL-12 production when stimulated with an agonistic anti-CD40 Ab, and PBMCs showed enhanced IFN-γ production. Toxicity limited the use of this therapeutic approach, although it was unclear whether these side effects were due to the presence of Cryptosporidium in the bile tract and/or effects of the CD40 agonist treatment. Nevertheless, for one patient, this treatment promoted suppression, but not clearance, of the infection. This important proof of principle demonstrates that immune-modulating therapy can be used to suppress chronic Cryptosporidium infection and improve clinical outcomes. It is also relevant to note that Cryptosporidium and the related parasite Toxoplasma gondii are both common opportunistic infections in acquired immune deficiencies highlighted earlier. However, in the setting of the PIDs associated with susceptibility to Cryptosporidium, it is only in HIGM1 that T. gondii is also a common opportunistic infection (72). This is an observation that emphasizes that although some immune pathways may be relevant to multiple intracellular parasites, there are also mechanisms of resistance that appear to be specific to a particular microorganism.

The NF-κB family of transcription factors is central to many innate and adaptive inflammatory processes that are downstream of TLRs, certain cytokine receptor families (CD40, TNF, IL-1), and signals mediated through the TCRs and BCRs (88–90). For the TLR and IL-1 family members, a complex series of events that involve the adapters MyD88 and IRAK4, as well as several kinases, leads to degradation of the inhibitory IκB proteins, which then allows cytosolic NF-κB to translocate to the nucleus to activate transcription. A large number of PIDs are associated with mutations in the signaling pathways that culminate in NF-κB activity (91). Because these transcription factors are expressed by many cell types and are engaged by a range of receptors, pinpointing how a specific mutation underlies increased infection risk is a challenge. It can be useful to consider the NF-κB signaling cascades as being broadly divided into canonical and noncanonical pathways (92). Defects in the canonical NF-κB pathway can be caused by deficiency in NF-κB essential modulator (NEMO), IκB kinase-β (IKKβ), or IκBα hypermorphism, which manifest in a variety of immune deficiencies (93–95). However, although deficiency in MyD88, IRAK4, NF-κB essential modulator, or IκB kinase-β is associated with Mendelian susceptibility to mycobacterial disease and Ab deficiency, it has not been linked with increased susceptibility to Cryptosporidium (96). In contrast, the noncanonical NF-κB pathway uses the NF-κB–inducing kinase (NIK; encoded by MAP3K14) to propagate signals from select TNF receptor family members (including CD40) and activates IκB kinase-α, which triggers an NF-κB signaling cascade (97). NIK/MAP3K14 deficiency has been described in just two individuals of a consanguineous family who had a single homozygous variant in the kinase domain of NIK leading to abrogation of kinase activity. This resulted in low Ig levels, decreased B cells and NK cells, and normal numbers of T cells, but impaired Ag-specific T cell proliferation and impaired memory T cells. These individuals had recurrent severe bacterial, viral, and Cryptosporidium infections (98). Because CD40L interaction with CD40 activates the noncanonical NF-κB pathway, this may help explain increased susceptibility to Cryptosporidium in these individuals.

Interestingly, deficiency in the NF-κB family member c-Rel was discovered after sequencing a young patient who presented with Cryptosporidium diarrhea (99). c-Rel is associated with the canonical NF-κB pathway and regulates IL-12 production in macrophages and DCs downstream of TLRs and CD40 (100, 101). In addition, TCR signals activate c-Rel, which has a cell-intrinsic role in the regulation of T cell activities required for resistance to infection (102, 103). One clue as to which of these roles (DC versus T cell c-Rel) is important in control of Cryptosporidium is that individuals with deficiencies in IL-12 or its receptor (both causes of Mendelian susceptibility to mycobacterial disease) do not appear to be at increased risk for Cryptosporidium infection. Similarly, mice deficient in IL-12 are acutely susceptible to Cryptosporidium but can suppress the infection long term (17). This suggests that the function of c-Rel in T cells (rather than its role in inducing IL-12 in DCs) is required for resistance to Cryptosporidium. Alternatively, the heightened susceptibility of this patient could be a consequence of defects in IL-12 combined with reduced T cell activity, which compromises two pathways associated with resistance to the parasite.

There is a group of structurally related cytokines (IL-2, IL-4, IL-7, IL-9, IL-15, and IL-21) that use the common γ-chain (CD132) as a component of their signaling receptor. However, their ability to combine with other cytokine-specific α- and β-chain subunits that then engage CD132 allows them to provide cytokine-specific signals. Deficiency in the common γ-chain, which is on the X chromosome, causes X-SCID, which underlies ∼20% of all cases of SCID in the United States and is uniformly fatal without HSCT (104). Mice that lack the common γ-chain are highly susceptible to Cryptosporidium (105), although this link has not been established for X-SCID in humans, likely because of the early lethality of the disease without HSCT. Although defects in the IL-7R (CD127) are rare, there is one case associated with disseminated Cryptosporidium (106), which would suggest a role of the broader common γ-chain pathway in resistance to Cryptosporidium. Indeed, deficiency in IL-21R (CD360) is characterized by increased susceptibility to Cryptosporidium. In these individuals, deficiency in IL-21R caused impaired B cell proliferation and Ig class-switching, defective T cell proliferation and cytokine production, and dysfunctional NK cell responses (107). Of the 13 individuals identified with IL-21R deficiency, 46% had Cryptosporidium-associated cholangitis (107–109).

Why might IL-21R deficiency associate more strongly with Cryptosporidium infection than other common γ-chain cytokine receptors? IL-21 is produced by NK T cells, T follicular helper cells, and Th17 cells and promotes CD8⁺ T cell and Th17 responses (110–114). IL-21 is highly expressed by activated CD4⁺ T cells in Peyer’s patches and the small intestine lamina propria, where it plays a key role in the development of Th17 cells important for mucosal barrier maintenance and defense against intestinal pathogens (particularly extracellular bacteria and fungi) (114–116). IL-21R can also be expressed by IECs, where it enhances expression of the chemokine CCL20 that mediates T cell and DC homing to the gut and has been reported to display direct anticryptosporidial effects (117–120). Perhaps given the role of IL-21 in inducing T cell homing and function of mucosal T cell subsets, it may be uniquely positioned among the common γ-chain cytokines to promote mucosal immunity to Cryptosporidium.

Dedicator of cytokinesis 8 (DOCK8) is a protein involved in intracellular signaling in leukocytes that is important for migration of cells through collagen-dense tissues (121–124). DOCK8 deficiency is associated with the autosomal recessive form of hyper-IgE or Job’s syndrome, which manifests as low lymphocyte counts with defects in regulatory T cell function, CD8⁺ T cell survival, NK cell function, and peripheral B cell tolerance (125–127). Individuals with DOCK8 deficiency also commonly have eosinophilia, elevated IgE, low IgM, and elevated IgG. In the gut, DOCK8 positively regulates IL-22 production by ILCs and negatively regulates type 2 immune responses (128–130). In one retrospective study of patients with DOCK8 deficiency, 6 of 28 patients were positive for Cryptosporidium, but only 1 reported chronic diarrhea, with 4 showing evidence of chronic biliary disease (131). The underlying basis for susceptibility to Cryptosporidium infection in DOCK8 deficiency could stem from its role in promoting NK and T cell production of IFN-γ (132, 133). Analysis of CD4⁺ T cells from DOCK8-deficient individuals showed a skewing toward Th2 cells and a corresponding loss of Th1 and Th17 cells. It is relevant to note that susceptibility to Cryptosporidium in DOCK8 deficiency appears to be less pronounced (in both severity and frequency) when compared with the other PIDs discussed earlier. This is an observation that may suggest that additional risk factors (microbiome, nutritional status) may be necessary to confer increased susceptibility among or within PIDs. Supporting this idea, large-scale population studies have revealed that polymorphisms in protein kinase C α (PRCKA) and deficiencies in mannose-binding lectin caused by polymorphisms in the MBL2 gene associate with increased susceptibility to Cryptosporidium infection (134–136).

The earlier sections highlight some of the real-world data from susceptible populations that have revealed pathways important in resistance to Cryptosporidium, but it is appreciated that PIDs that do not confer susceptibility to Cryptosporidium are also informative. For example, within this laboratory of natural human immunology, there are currently no obvious links of innate sensors to resistance to this parasite. Thus, MyD88 and IRAK4 deficiencies that impair TLR- and IL-1R–mediated immunity are not associated with susceptibility to Cryptosporidium (137). This may reflect the biology of Cryptosporidium infection, namely, that it is mostly restricted to the gut, where many commensal, nonpathogenic microbes exist as part of the human microbiome. In this environment, pathways involved in pattern recognition receptor signaling may be fine-tuned to avoid overactivation against nonpathogenic bacteria, which could otherwise lead to aberrant inflammation. The observation that gastrointestinal infections are rare in IRAK4 and MyD88 deficiencies suggests that these TLR-driven pathways may be less important for sensing invading microorganisms in the gut (137). Rather, TLR pathways have been shown to contribute to intestinal homeostasis by regulating barrier function and regeneration (138–140). Therefore, the lack of association with Cryptosporidium and innate sensors may reflect the role of these pathways in homeostatic rather than proinflammatory gut function.

The majority of acquired immunodeficiencies and PIDs described that confer susceptibility to Cryptosporidium center on impaired T cell function (Fig. 2). Clinical and experimental datasets have informed a paradigm where DCs present Ag and produce IL-12 to promote T cell production of IFN-γ, which in turn acts on IECs to restrict Cryptosporidium (Fig. 1). This model is relevant to many intracellular infections but is at odds with the clinical experience in which neither primary IL-12 deficiency nor pharmacologic IL-12 blockade (anti–IL-12/23 p40 therapy) is associated with heightened susceptibility to Cryptosporidium (141, 142). Similarly, despite IFN-γ being considered the major mechanism used by T cells to control the parasite, primary deficiency in IFN-γ or STAT1 is also not linked to increased susceptibility (143–147). These observations are perhaps relevant to the details of how IL-12– or IFN-γ–deficient mice respond to Cryptosporidium. These knockout mice are highly susceptible to Cryptosporidium early during infection and harbor low levels of parasite chronically, but they do have a significant long-term mechanism of resistance (in contrast with T cell–deficient mice that remain highly infected long term) (14, 17). These data highlight the existence of a T cell–dependent but IL-12– and IFN-γ–independent mechanism of control of Cryptosporidium that may be particularly prominent in humans. Two candidate effectors linked to resistance to mucosal pathogens are TNF-α and IL-17, but the widespread blockade of these cytokines in inflammatory bowel disease, rheumatologic disease, and autoimmune diseases do not appear to affect Cryptosporidium susceptibility. Perhaps the clinical datasets that emphasize the important role of IL-21 and CD40L indicate the need to better understand how these effectors might influence IFN-γ–independent pathways that contribute to parasite control. An alternative perspective is that some PIDs that lead to impaired acute resistance may be missed because infection is ultimately controlled, which would obscure the link between PIDs and susceptibility to Cryptosporidium. Indeed, Cryptosporidium infection in some of the DOCK8-deficient individuals would not have been detected without the use of PCR for diagnostics (131). Perhaps the routine use of a more sensitive screening tool across larger populations in clinical or epidemiologic settings may lead to further associations between immune components and risk for Cryptosporidium infection. Continued discovery of immune deficiencies associated with Cryptosporidium can complement laboratory models and could lead to improved treatment and vaccination strategies that are needed to control this infection.

The authors have no financial conflicts of interest.