Immunodeficiency With Hyper-Igm, Type 1
A number sign (#) is used with this entry because X-linked immunodeficiency with hyper-IgM type 1 (HIGM1) is caused by mutation in the CD40LG gene (300386) on chromosome Xq26.
DescriptionHIGM is a rare immunodeficiency characterized by normal or elevated serum IgM levels associated with markedly decreased IgG, IgA, and IgE, resulting in a profound susceptibility to bacterial infections and an increased susceptibility to opportunistic infections. Patients with X-linked HIGM also tend to have neutropenia, as well as a high rate of gastrointestinal and central nervous system infections, often resulting in severe liver disease and/or neurodegeneration (summary by Levy et al., 1997).
Genetic Heterogeneity of Immunodeficiency with Hyper-IgM
Other forms of HIGM include HIGM2 (605258), which results from mutation in the AICDA gene (605257), HIGM3 (606843), which results from mutation in the CD40 gene (109535), and HIGM5 (608106), which results from mutation in the UNG gene (191525). See also HIGM4 (608184).
Clinical FeaturesThe clinical course of X-linked hyper-IgM syndrome is similar to that of X-linked Bruton-type agammaglobulinemia (300755) except for a greater frequency of 'autoimmune' hematologic disorders (neutropenia, hemolytic anemia, thrombocytopenia). Neutropenia may be accompanied by gingivitis, ulcerative stomatitis, fever, and weight loss (Levy et al., 1997).
Jamieson and Kerr (1962) reported a pedigree in which 4 boys were affected. Levitt et al. (1983) reported 4 male patients with recurrent infections. Two of them had agranulocytosis or neutropenia. One had an uncle (presumably maternal) who died in infancy after developing agranulocytosis and Candida sepsis and who showed atrophic lymphoid tissue at autopsy.
Pathologically, lymphoid tissue shows disorganization of the follicular architecture and PAS-positive plasmacytoid cells containing IgM. Lymph nodes lack germinal centers (Ramesh et al., 1999). Tonsillar hypertrophy due to infiltration with these cells may occur. (The tonsils and other lymphoid tissues are atrophic in Bruton agammaglobulinemia.)
Levy et al. (1997) estimated that only 20% of patients will reach the third decade of life and that 75% of these patients will have liver complications. Hayward et al. (1997) described various gastrointestinal cancers, including cholangiocarcinoma, hepatocellular carcinoma, and adenocarcinoma in a cohort of boys with the hyper-IgM syndrome 1 and cholangiopathy. In that study, 70% of the boys who were systematically screened for infection had Cryptosporidium parvum infection (protozoan that causes bowel infection, usually in the setting of immunosuppression or immunodeficiency) and all had clinically significant chronic liver disease.
Cunningham et al. (1999) reported 3 patients with X-linked hyper-IgM syndrome from 2 families who developed enteroviral encephalitis at ages 30 months, 21 months, and 30 months. All presented with central nervous system abnormalities and the 2 surviving patients showed developmental delay. The authors stressed the importance of CSF PCR testing in similar instances.
Aschermann et al. (2007) reported a 19-year-old male patient with X-linked hyper-IgM syndrome, confirmed by genetic analysis, who developed progressive multifocal leukoencephalopathy due to opportunistic infection with the JC virus. He had decreased serum IgA, slightly increased IgM, and normal IgG due to monthly infusions. Despite combined antiviral treatment, he died after 6 weeks. The report indicated that, in addition to immunoglobulin deficiency, patients with this disorder have impaired cellular immune responses due to decreased T cell activation.
Hasegawa et al. (2014) reported a 21-year-old Japanese man, born of unrelated parents, with HIGM1 confirmed by genetic analysis. He presented in infancy with failure to thrive and recurrent otitis media and was treated with immunoglobulin. He showed clumsiness in childhood, and by age 20 years he had developed involuntary movements of the extremities, dysarthria, and hyperactive reflexes. He also had significant cognitive impairment (IQ of 58). Laboratory studies showed low serum IgG and increased serum IgM. No pathogens were detected in the cerebrospinal fluid. Brain imaging showed atrophy of the cerebral cortex and striatum, and EEG showed abnormalities in the absence of clinical seizures. Within 6 months, he was unable to walk due to severe choreoathetosis. Whole-exome sequencing detected a truncating mutation in the CD40LG gene. He also carried an in-frame deletion in the POLG gene (174763) that was not thought not to contribute to the phenotype. The patient was part of a cohort of 9 individuals with neurodegenerative features and hypogammaglobulinemia who underwent whole-exome sequencing. Hasegawa et al. (2014) noted that patients with CD40LG deficiency are susceptible to central nervous system infections, but also suggested that CD40LG may play a role in neuronal function. The report illustrated that whole-exome sequencing can lead to unpredictable molecular diagnoses and unexpected clinical features.
InheritanceHIGM1 is inherited as an X-linked recessive trait. Female carriers manifest normal IgG and IgA production (Hendriks et al., 1990).
Other inheritance patterns have been suggested. Kyong et al. (1978) reported 2 cases in male and female patients and suggested autosomal recessive inheritance. They referred to a case reported by Gleich et al. (1965) in which a female infant had reduced levels of IgG and IgA, elevated IgM, recurrent otitis media, pneumonia, and cervical abscesses. Brahmi et al. (1983) reported father and 2 daughters with the hyper-IgM syndrome. They concluded that the genetics of the hyper-IgM syndrome is 'still unresolved.' Probable autosomal dominant inheritance of one form was suggested. In a review paper, Notarangelo et al. (1992) stated that hyper-IgM syndrome had been shown to be X-linked, autosomal recessive, and autosomal dominant.
DiagnosisLin et al. (1996) pointed to PCR-SSCP screening of genomic DNA as a reliable way to establish a diagnosis of hyper-IgM syndrome 1 unequivocally and to identify carriers. Patients with the X-linked form of the disease have the onset of infections in the first few years of life and are more likely to have opportunistic infections and/or neutropenia than are patients with autosomal recessive or multifactorial disease. However, these features are not sufficiently specific to permit a definitive diagnosis of X-linked hyper-IgM syndrome.
Clinical ManagementDunn et al. (1982) found that large doses of fresh plasma corrected the neutropenia. Notarangelo et al. (1992) stated that treatment is mainly based upon regular administration of intravenous immunoglobulins, and that, in addition, steroids may be used in the treatment of neutropenia and severe autoimmune manifestations.
Thomas et al. (1995) performed successful allogeneic bone marrow transplantation in a boy with hyper-IgM syndrome 1 using his carrier sister as the donor. Full engraftment was shown by several means, including changes in red cell antigens, the results of fluorescence in situ hybridization for X and Y chromosomes, polymorphism of the CD40LG gene, and expression of the CD40 ligand by activated T cells. Transplantation was considered indicated because the patient had had P. carinii pneumonitis and came from a family in which 2 maternal uncles had died of protracted diarrhea at the ages of 6 months and 2 years, respectively. A first cousin had the same disorder with persistent diarrhea caused by cryptosporidium and with cholangitis associated with liver cirrhosis.
Hadzic et al. (2000) performed a cadaveric orthotopic liver transplantation together with nonmyeloablative bone marrow transplantation from a matched, unrelated donor in an 18-year-old man with end-stage chronic liver disease associated with the X-linked hyper-IgM syndrome. The removed liver was severely cirrhotic with alternating areas of macronodular hypertrophy and collapse. Fourteen months after liver transplantation and 13 months after bone marrow transplantation, the patient was in excellent health, with satisfactory function of both grafts.
Gennery et al. (2000) reported successful bone marrow transplant in a patient with X-linked hyper-IgM syndrome with a 6/6 antigen matched unrelated donor.
PathogenesisIt was first thought that the defect in this disorder was within the B cells themselves (see HISTORY section). Levitt et al. (1983) demonstrated that this disorder has a primary dysfunction of B-lymphocyte heavy chain isotype switching from IgM to IgG and IgA. Clinically, however, the recurrence of opportunistic infections (Pneumocystis carinii, toxoplasmosis) suggested anomalies of T-cell function. Moreover, isotype switch obtained in HIGM1 B cells after cocultivation with Sezary syndrome T cells, as well as a random pattern of X-chromosome inactivation in obligatory carriers of HIGM1, argued against a primary B-cell defect (Mayer et al., 1986).
Fuleihan et al. (1993) evaluated isotype switch recombination in 3 affected males by examining interleukin 4-driven IgE synthesis. T-cell-dependent IgE synthesis was completely absent in the B lymphocytes of the patients. CD40 mAb plus interleukin-4 induced the patients' B cells to synthesize IgE and to undergo deletional switch recombination. In contrast, T cells from the patients failed to induce IgE synthesis in interleukin-4-treated B cells and were unable to express the CD40 ligand on their surface. These results suggested that defective expression of the CD40 ligand underlies the failure of isotype switching in HIGM1.
Aruffo et al. (1993) found that patients with HIGM1 express functional CD40 but their T cells do not have functional CD40 ligand (which Aruffo et al. (1993) called gp39) as measured by T-cell binding of CD40-Ig. The patients expressed normal levels of gp39 mRNA, but these RNAs encoded defective gp39 proteins owing to mutations in the extracellular domain of gp39. Soluble recombinant forms of gp39 containing these mutations were unable to bind CD40 and drive normal B-cell proliferation.
Bossaller et al. (2006) found that CD40L-deficient patients, like ICOS (604558)-deficient patients, had abrogated germinal center formation and a severe reduction of CXCR5 (BLR1; 601613)-positive T cells.
Using flow cytometric analysis, van Zelm et al. (2014) found reduced numbers of all memory B-cell subsets except CD27 (TNFRSF7; 186711)-negative/IgA-positive B cells in both CD19 (107265)-deficient patients and CD40L-deficient patients compared with controls. Analysis of transcripts after class switching demonstrated that patient transcripts had fewer somatic mutations and reduced usage of IgG2 and IgA2 subclasses. There was also a deficiency in selection strength of mutations for antigen binding in patients compared with controls, whereas selection to maintain superantigen binding was normal. Selection against the autoreactive properties of immunoglobulins was impaired in patients. Somatic hypermutation analysis revealed decreased AICDA and UNG activity in CD40L deficiency, but increased UNG activity and decreased mismatch repair in CD19 deficiency. Van Zelm et al. (2014) concluded that both the B-cell antigen receptor and CD40 signaling pathways are required for selection of immunoglobulin reactivity, but that they differentially mediate DNA repair pathways during somatic hypermutation and thereby together shape the mature B-cell repertoire.
X-Inactivation Studies
If the defect in the switch mechanism is intrinsic to the B cells, a skewed X chromosome inactivation pattern would be observed in IgG- and IgA-expressing B lymphocytes of female carriers. Hendriks et al. (1990) studied lymphoblastoid B cells from 2 female carriers (see HISTORY section). Hendriks et al. (1990) concluded that the HIGM1 gene encodes a class switch inducer that is transferred to B lymphocytes from a cell of synthesis, possibly T lymphocytes.
Contrary to the findings of Hendriks et al. (1990) and those of Conley et al. (1988), Notarangelo et al. (1991) found nonrandom X-chromosome inactivation in T cells, B cells, and neutrophils, but not in fibroblasts, of obligate carriers, suggesting that several different hematopoietic cell lineages are primarily involved in HIGM1. Preferential inactivation of the paternally derived X chromosome was demonstrated by analysis of segregation of the alleles defined by 2 DNA probes. Notarangelo et al. (1991) suggested that the HIGM1 mutation may confer an advantage in differentiation and/or proliferation to hematopoietic precursors carrying the mutant allele on the active X chromosome.
In studies of X-chromosome inactivation in carriers of HIGM1, Hollenbaugh et al. (1994) found that the CD40L gene is not selectively inactivated. Furthermore, even when there was extremely skewed inactivation, normal levels of serum immunoglobulins were found. Unlike some other X-linked defects in which extreme lyonization may lead to disease, a small population of cells expressing the wildtype protein was sufficient to maintain normal humoral immunity and prevent the clinical symptoms of the disorder. Kipps (1994) commented that the findings have encouraging implications for patients with the disorder, since it seems that only a relatively small fraction of the T cells need express a functional CD40-ligand for effective immunity. Even a partial reconstitution with precursor T cells capable of expressing a functional ligand might suffice.
MappingMensink et al. (1987) concluded that the locus for immunodeficiency with increased IgM (symbolized XHM by them) is linked to the DXS42 RFLP locus, which maps to Xq24-q27. Recombination between XHM and DXS17 was observed, whereas no recombination between XLA and DXS17 has been found; thus, XHM and XLA are apparently determined by separate gene loci. Padayachee et al. (1992, 1993) narrowed the location to Xq26 by multipoint linkage studies demonstrating that it is close to HPRT (308000), a gene that forms part of an extensive YAC contig mapping to Xq26; a maximum lod score of 4.89 was obtained. The existence of an easily scorable VNTR of 5 alleles within the HPRT gene means that other families with X-linked hyper-IgM syndrome are likely to be informative for this polymorphism.
Aruffo et al. (1993) mapped the GP39 gene to Xq26 by PCR analysis of a regional mapping panel, followed up by fluorescence in situ hybridization for precise localization. By YAC analysis, Pilia et al. (1994) mapped the CD40L locus between DXS144E and DXS300 in Xq26 and determined its transcription to be from 5-prime centromeric to 3-prime telomeric. This corresponded to the site where the clinical phenotype of the hyper-IgM syndrome had been mapped.
Allen et al. (1993) mapped the CD40LG gene (300386) to the proximal region of the mouse X chromosome, linked to Hprt. Hprt maps to the Xq26-q27.2 region, which suggested that the human CD40LG gene would also map to this region. This was confirmed by fluorescence in situ hybridization studies of CD40LG by Graf et al. (1992) and Allen et al. (1993).
Molecular GeneticsAllen et al. (1993) presented conclusive evidence that the defect in X-linked hyper-IgM syndrome resides in the gene for the CD40 ligand (300386). Because CD40LG induces B-cell proliferation in the absence of any costimulus and because the hyper-IgM phenotype and the CD40LG gene map to the same location, CD40LG was suggested as the site of the mutation in HIGM1. Allen et al. (1993) demonstrated this to be case by the finding of point mutations in 3 of 4 patients with the syndrome (300386.0003-300386.0005). Similarly, Aruffo et al. (1993) identified mutations in the CD40LG gene in patients with the syndrome (300386.0001-300386.0002).
Animal ModelRosen (1975) stated that 'a similar syndrome of X-linked immunodeficiency with increased IgM has been found in mice.' Xu et al. (1994) generated CD40LG-deficient mice by targeted disruption.
Using intravital microscopy and histologic examination of arterioles in mice lacking CD40L, Andre et al. (2002) observed frequent rupture and embolization of thrombi. The thrombi showed lower platelet density compared to those of wildtype mice, which contributed to platelet fragility. Administration of recombinant soluble CD40L (rsCD40L), but not rsCD40L with a mutation changing the KGD motif sequence to KGE, restored thrombus stability even in CD40 -/- mice, indicating that CD40L is not acting through CD40 ligation. Evaluation of hemostasis suggested that the thrombus instability in CD40L -/- mice is not due a lack of fibrin formation but rather a defect in platelet-platelet interaction which could be stabilized by the wildtype, but not by the mutant, rsCD40L. Flow cytometric analysis demonstrated that rsCD40L binds to activated platelets of wildtype as well as of CD40 -/- mice, but that this binding can be inhibited by a peptide interfering with ITGA2B (607759)/ITGB3 (173470) binding. Plate-binding analysis indicated specific saturable binding of rsCD40L to ITGA2B/ITGB3. Fluorescence microscopy showed that human platelets spread on but did not adhere to an rsCD40L-coated glass surface only in the absence of an inhibitor of ITGA2B/ITGB3 binding. Andre et al. (2002) concluded that CD40L is an ITGA2B/ITGB3 ligand, a platelet agonist, and necessary for the stability of arterial thrombi. They also noted that these findings suggest the careful evaluation of clinical trials with anti-CD40L therapy.
HistoryRamesh et al. (1999) reviewed the history as well as other aspects of the hyper-IgM syndrome, which they abbreviated HIM. In the WHO classification of immunodeficiencies, an entity termed X-linked immunodeficiency with increased IgM was listed (Fudenberg et al., 1970). A definition of the disorder was an outcome of an international workshop (Cooper et al., 1974). It was originally hypothesized that B lymphocytes from patients with HIGM1 have an intrinsic inability to undergo immunoglobulin isotype switch.
Levitt et al. (1983) suggested that this disorder has a primary dysfunction of B-lymphocyte isotype switching. In 4 male patients with hyper-IgM immunodeficiency, the number, proportion, and proliferation of T lymphocytes were shown to be normal. IgG and IgA B lymphocytes were completely absent. In vitro stimulation of patients' B cells with both T cell-dependent and T cell-independent activators failed to induce any IgG or IgA-producing B cells. They concluded that individuals with this disorder possess an intrinsic B cell dysfunction that is not related to abnormal T cell regulation.
The observation that patients with HIM and particularly those with the X-linked form of the disorder (XHIM) were prone to opportunistic infections suggested a T-cell defect in spite of laboratory evidence for a humoral immune deficiency. The hypothesis of a primary T-cell defect was elegantly supported by the studies of Mayer et al. (1986), who demonstrated that B cells from XHIM patients could be driven to secrete immunoglobulins of various isotypes in the presence of pokeweed mitogen when cocultivated with 'helper T lymphoblasts' from a patient with a Sezary-like syndrome, a neoplastic disorder of T cells.
Hendriks et al. (1990) studied lymphoblastoid B cells from 2 female carriers who had IgG- and IgA-expressing B cells, in order to determine if the defect in the switch mechanism is intrinsic to the B cells. In an analysis of differential methylation of the polymorphic DXS255 locus, random X chromosome inactivation patterns were found in populations of T lymphocytes, in IgM-expressing B lymphocytes, and in IgG- or IgA-expressing B lymphocytes. Similar extent of heterogeneity was found for Ig H chain rearrangements and the Ig light chain usage in the IgA- or IgG-expressing B cell that had inactivated the X chromosome which carried the intact gene and in clones with the mutant-bearing X chromosome inactivated. The results indicated that the intrinsic Ig H chain class switch mechanism in this disorder is fully intact in B lymphocytes. When B lymphocytes with the X chromosome bearing the mutant IMD3 gene on the active X chromosome are placed in an in vivo environment that contains the intact gene product, i.e., in the female heterozygotes, the switch from IgM to IgG or IgA production occurs at normal frequencies. These findings are consistent with the observation that switch of hyper-IgM B cells is induced by Sezary-like T lymphoblasts (Mayer et al., 1986).