Article Text


Immunogenetics: changing the face of immunodeficiency
  1. Alison M Jones1,
  2. Hubert B Gaspar2
  1. 1Department of Immunology, Great Ormond Street Hospital NHS Trust, Great Ormond Street, London WC1N 3JH, UK
  2. 2Molecular Immunology Unit, Institute of Child Health, University College London, London WC1N 1EH, UK
  1. Dr Jones email: Alison.Jones{at}

Statistics from

Primary immunodeficiency is increasingly recognised in both children and adults. Although clearly defined disorders are individually rare, collectively they form a significant group. The molecular basis of several disorders has been known for over a decade, but in the past 10 years there has been a massive increase in the definition of defects underlying many immunodeficiencies.

Management of immunodeficiency has historically consisted of supportive treatment, including antibiotics, immunoglobulin (Ig) replacement, and in some cases, immunomodulation and immunosuppression. Bone marrow transplantation has been standard treatment for the past 20 years, but only for clearly defined cases of severe combined immunodeficiency (SCID) and a few other disorders known to have a very poor outlook without bone marrow transplantation. Accumulating worldwide experience is now providing evidence for poor long term prognoses in many other situations, even with optimal medical treatment. Bone marrow transplantation is therefore being considered for increasing numbers of children affected by these disorders, and, in parallel with this, there is a pressing need to define individual disorders as accurately and as early as possible.

The recent increase in knowledge of the molecular defects underlying many immunodeficiencies has led to several improvements in diagnosis and management.

First, precise molecular diagnosis is now possible in many cases, allowing earlier decisions to be made about the most appropriate management. This is particularly applicable in children who have evidence of combined (cellular and humoral) immunodeficiency, but with “milder” clinical phenotypes than infants with classical SCID. Some of these children are found to have identical molecular defects to those causing SCID, and in these cases the long term outlook is now known to be poor enough to justify bone marrow transplantation at an early stage.

Second, accurate carrier detection and first trimester prenatal diagnosis are possible in any family where the precise mutation has been defined. In some cases prenatal diagnosis of an affected fetus may not lead to termination of pregnancy, but can allow preparation for bone marrow transplantation early in the neonatal period, or even in utero in selected cases. Earlier bone marrow transplantation is associated with a greater chance of a successful outcome.

Third, knowledge of the molecular basis of immunodeficiencies allows further research into the mechanisms of the immune defects, with the possibility of designing more rational treatment in some cases, and for developing strategies for somatic gene therapy.

Specific disorders

Those immunodeficiencies for which the underlying molecular defects have been defined are summarised in tables 1 and 2. In several of these disorders, knowledge of the molecular basis has led to the improvements in management outlined above. In other cases ongoing research is continuing to add to our understanding of the molecular mechanisms of immune function. It is not within the scope of this review to go into detail for every disorder, so several illustrative diseases will be discussed.

Table 1

X linked immunodeficiencies

Table 2

Autosomal recessive immunodeficiencies

X linked agammaglobulinaemia (XLA)

X linked agammaglobulinaemia is the archetypal humoral immunodeficiency, presenting in its classical form with susceptibility to bacterial infection, absent or low circulating mature B cells, and absence of all immunoglobulin isotypes.1 Standard treatment is with life long immunoglobulin replacement. The outlook is good provided that adequate IgG levels are maintained and infections are promptly treated. There is, however, a significant risk of enteroviral infection, and chronic meningoencephalitis can develop despite immunoglobulin replacement therapy.2 The diagnosis of XLA has been made on lack of peripheral B lymphocytes, absence of immunoglobulin production, and in some cases an X linked pedigree. However, girls with congenital agammaglobulinaemia and absent B cells3 and boys with a similar immunological profile but from an autosomal pedigree have also been reported, suggesting that the same immunophenotype can arise from a number of molecular defects.

In 1993, two groups identified the gene defective in XLA, one using a positional cloning strategy and another using a reverse genetic approach.4,5 Subsequently designated Btk (Bruton's tyrosine kinase), the gene is a non-receptor tyrosine kinase expressed in the B and myelomonocytic cell lineages but not in T cells. Btk is activated following stimulation of several cell surface receptors including most importantly the B cell receptor.6 In addition, interaction of Btk with a variety of intracellular signalling molecules including PIP3 (phosphatidylinositol-3,4,5-trisphosphate) and PLCγ1 (phospholipase Cγ1) is crucial for regulation of the sustained calcium signal in response to BCR engagement (reviewed in7). However, despite the large volume of experimental data, it is still unclear precisely how defects in Btk lead to the arrest in B cell maturation seen in XLA.

The identification of the Btk gene defect allows unambiguous assignment of a molecular diagnosis to individuals with XLA and to female carriers. This is particularly important as defects in μ heavy chain and λ5/14.1 surrogate light chain (components of the pre-B cell receptor) have also been shown to give rise to congenital agammaglobulinaemia.8,9 Furthermore, screening of individuals with atypical or less severe XLA phenotypes, including those previously labelled as common variable immunodeficiency or immunoglobulin subclass deficiency, has shown that some of these individuals may also have Btk defects.10

Mutation detection in XLA has traditionally been carried out by initial screening of the 19 exons of the Btk gene by single stranded conformation polymorphism (SSCP) analysis and subsequent direct sequencing of the affected exon. This is a time consuming technique that may take months to complete, and has a sensitivity of only 90%. More recently generation of specific anti-Btk antibodies has allowed development of more rapid methods of diagnosis. Analysis of Btk protein expression in peripheral blood mononuclear cells of XLA patients by intracellular fluorescence activated cell sorter (FACS) or western blot analysis shows that over 90% lack Btk expression.11,12 These techniques have also been useful in determining carrier status. In obligate carriers of XLA, intracellular FACS staining of monocytes shows a dual population of Btk expressing and Btk non-expressing cells since monocytes are randomly X inactivated in these women.11 Similarly, if an XLA patient expresses a Btk protein of abnormal size, western blot analysis of monocytes from carrier females shows expression of an abnormal and normal sized band, thereby confirming carrier status.

CD40 ligand deficiency

X linked hyper IgM syndrome (XHM) has been recognised since 1966,13 its characteristic immunological phenotype including very low IgG and IgA levels, normal or raised IgM, normal lymphocyte subpopulations, and apparently normal T cell function. The clinical phenotype is one of susceptibility to bacterial infection, but for many years before the responsible gene was identified it was known that boys affected by XHM were susceptible to opportunistic infection with organisms such as Pneumocystis carinii.14 More recently, particular susceptibility to Cryptosporidium parvum has also been recognised. It was therefore not surprising when the gene responsible for XHM was identified in 1993 as that coding for CD40 ligand (CD40L), a surface molecule present on activated T cells.1517 CD40L is essential for immunoglobulin isotype switching through its interaction with CD40 on the surface of B cells, but it also plays an as yet undefined role in T cell function. T cell proliferative responses to a range of antigens have been shown to be defective in CD40L deficiency.18 CD40L is also expressed on the surface of biliary epithelial cells, and CD40/CD40L interaction here is thought to be involved in control of intracellular pathogens such as C parvum.

A further problem in CD40L deficiency is that of susceptibility to liver disease and liver/gastrointestinal malignancy, with a possible link between chronic cryptosporidial infection and sclerosing cholangitis.19 The European Society for Immunodeficiency database for CD40L deficiency currently contains clinical and molecular data from 113 affected males. The actuarial survival at 25 years is only 25%, and the incidence of liver disease by the age of 20 years is 80%.20 There is so far little apparent genotype/phenotype correlation, and there is considerable intrafamilial variation in clinical phenotype, so predictions of severity or likelihood of liver disease are not possible.

Knowledge of the molecular basis of XHM allows accurate diagnosis not only in boys who have a typical phenotype or a positive family history but also in some cases of previously undefined hypogammaglobulinaemia. Several adult males diagnosed as affected by common variable immunodeficiency (CVID) have recently been found to have CD40L deficiency. Although these individuals may not develop the life threatening complications of CD40 ligand deficiency, other affected family members could have a more severe phenotype. Carrier detection and prenatal or early diagnosis is therefore important for relatives at risk. Any affected male who has a matched family donor is now recommended for bone marrow transplantation, and, where possible, unrelated donor bone marrow transplantation is offered to affected boys who have any evidence of liver disease or other complications.

Wiskott-Aldrich syndrome

Wiskott-Aldrich syndrome (WAS) is a rare X linked recessive disease characterised by immune dysregulation and microthrombocytopenia.21 Clinical manifestations of the immune disorder include susceptibility to pyogenic, viral, and opportunistic infection, and eczema. There is usually progressive loss of T lymphocytes during childhood, with defects in proliferative and delayed type hypersensitivity responses, and deficient production of antibodies to both polysaccharide and protein antigens. Conventional supportive treatment for WAS includes prophylactic antibiotics and immunoglobulin, splenectomy for thrombocytopenia, and platelet transfusion in circumstances of life threatening haemorrhage.22,23 Currently only allogeneic bone marrow transplantation can offer curative treatment for WAS.24 Like many other primary immunodeficiencies, there is considerable heterogeneity in the severity of clinical manifestations. In mildly affected individuals, diagnosis based on clinical criteria has proved difficult. However, the identification of WASP (Wiskott-Aldrich syndrome protein), the gene defective in WAS, now allows definitive molecular diagnosis. To illustrate this, the clinical syndrome of microthrombocytopenia in X linked pedigrees (X linked thrombocytopenia or XLT) was originally thought to be a distinct genetic entity, but it has now also been shown to arise from WASP defects.25 Furthermore, in parallel with the development of protein based assays for XLA, similar tests are now available for the rapid detection of WASP abnormalities using immunoblotting and FACS analysis.26,27

From a more basic perspective, the identification of the gene defect has allowed a greater understanding of the molecular pathogenesis of the WAS phenotype. The WASP gene encodes a 502 amino acid intracellular protein expressed exclusively in haematopoietic cells,28 which belongs to a recently defined family of more widely expressed proteins involved in transduction of signals from receptors on the cell surface to the actin cytoskeleton. The WASP family proteins are organised into modular domains defined by sequence homology and binding interactions with other signalling molecules (reviewed in 29). Experimental data suggest that there are intrinsic defects in cytoskeletal architecture organisation and WAS immune cell motility and trafficking that may explain the variety of immunological abnormalities. Recent studies have shown WASP to cluster physically with polymerised actin and to act as a direct effector molecule for the small GTPase protein Cdc 42, which is critical for regulation of cytoskeletal structures.30,31 Studies directly visualising WAS cells have shown that both WAS dendritic cells and macrophages have major abnormalities in the distribution of peripheral filamentous actin and fail to develop polarised filopodial extensions in response to chemotactic attractants.32,33 T cells from WAS patients show abnormalities of antigen receptor (CD3) stimulated proliferation (but not allostimulation), and both T and B lymphocytes have similar defects in the distribution of filamentous actin to those seen in dendritic cells and macrophages.34,35 These data suggest that a major component of the WAS immunophenotype may be abnormal initiation and regulation of immune responses owing to defective immune cell trafficking and motility.

X linked lymphoproliferative disease (Duncan's syndrome)

X linked lymphoproliferative disease (XLP) is a rare disorder characterised by a dysregulated immune response to Epstein-Barr virus (EBV). First recognised over 25 years ago,36,37 more than 80 kindreds have now been identified worldwide. The most usual presentation is with fulminant, often fatal acute EBV infection, but its phenotypic expression is highly variable. Other presentations include acquired hypogammaglobulinaemia, B cell lymphoma, aplastic anaemia, vasculitis, and lymphomatoid granulomatosis.38 In addition, this same study reports that 27 of 272 individuals (10%) manifested an XLP phenotype without serological and in some cases genomic evidence of EBV infection. This, together with the variability in the clinical phenotype and the lack of a reliable diagnostic test, has made firm diagnosis of XLP extremely difficult. Historically, the diagnosis was based on a typical history with a suggestive pedigree, and sometimes the finding of raised anti-EBV antibody levels in obligate female carriers. In families where XLP has been considered likely, boys have been treated with long term immunoglobulin replacement in an attempt to protect them against EBV infection. However, this was not effective in at least one boy, who developed a haemophagocytic lymphohistiocytosis-like syndrome while receiving adequate doses of immunoglobulin.

The gene responsible for XLP was localised to Xq25 in 1990 by demonstration of an Xq25 deletion in an affected family,39 but it was not until 1998 that the gene was finally identified as that coding for SAP (signalling lymphocyte activating molecule associated protein).40,41 Signalling lymphocyte activating molecule (SLAM) is a B and T cell surface marker which forms a receptor–ligand pair, triggering of which coactivates B and T lymphocyte responses. SAP is a 15 kDa protein that is primarily expressed in T cells and acts as a negative regulator of SLAM interactions. Mutations of SAP in XLP patients are postulated to dysregulate interactions between SLAM molecules on the surfaces of T and B cells, and result in uncontrolled B cell proliferation. Mutations in the SAP gene have so far been demonstrated in 13 males affected by XLP. However, there are still several families thought to carry XLP in whom no SAP mutations have been found, and the possibility remains that a second gene is involved.

Identification of the gene responsible for XLP allows accurate diagnosis by mutation detection, so that boys found to be affected can be recommended for bone marrow transplantation before the onset of progressive EBV disease or other complications. As the gene consists of only four exons, SSCP analysis (fig 1) or direct sequencing of genomic or cDNA should be relatively straightforward. It is also likely that use of an SAP antibody will the allow the development of a rapid protein based diagnostic assay. Unfortunately, lack of an identifiable mutation cannot at present exclude a diagnosis of XLP because of the uncertainty concerning a possible second genetic locus.

Figure 1

Single stranded conformational polymorphism (SSCP) analysis of the SAP gene in a family with XLP shows that the affected boy (CM) has a band shift in exon 2 in comparison with a normal control (N). Subsequent sequencing of exon 2 in CM identified a C to G change at codon 54 which resulted in a tyrosine being replaced by a stop codon (Y54X) thus confirming the diagnosis of XLP. The boy's mother (WM) shows presence of both normal and abnormal bands by SSCP and is a carrier of the Y54X mutation.

Defects in the common γ chain and JAK3 proteins

Of the different molecular defects that result in SCID (tables 1 and 2), the most common is X linked severe combined immunodeficiency (X-SCID) which arises from defects in the common γ chain (γc).42 The characteristic immunophenotype in X-SCID consists of absence of T and NK (natural killer) cell development, but normal B cell numbers, although these are dysfunctional (T-B+NK− SCID). A similar, though much rarer, clinical and immunological phenotype arises from an autosomal defect in the gene encoding the tyrosine kinase JAK3 (Janus associated kinase-3).43,44 Details of the γc/JAK3 signalling pathway are illustrated in fig 2. Data from in vitro studies and from “knockout” mice models have shown that functional interleukin (IL)-7/IL-7R and IL-15/IL-15R mediated signalling pathways are essential for normal T and NK cell development, respectively, thus explaining the lineage specific development defects present in X-SCID and JAK3 deficient patients. Abnormalities in IL-2 and IL-4 signalling may further explain the functional B cell defects.

Figure 2

γc is an essential component of the high affinity cytokine receptors for interleukin (IL)-2, -4, -7, -9 and -15. Upon activation of these receptor complexes by cytokine, the JAK3 molecule, which binds selectively to γc, is itself activated and subsequently phosphorylates STAT5 (Signal Transducer and Activators of Transcription). Phosphorylated STAT5 dimerises and then translocates to the nucleus where it binds to specific sites to initiate transcriptional events.

Identification of γc and JAK3 has had profound effects on the diagnosis and management of these conditions. Approximately two thirds of children with X-SCID have abnormal expression of γc on the surface of mononuclear cells,45 allowing confirmation of the molecular diagnosis by FACs analysis of peripheral blood mononuclear cells. In infants affected by T-B+NK− SCID who have normal γc expression, further dissection of the signalling pathway can now be undertaken. IL-2 stimulation of mononuclear cells results in tyrosine phosphorylation of JAK3 at specific tyrosine based motifs. A monoclonal antibody directed against phosphotyrosine residues can be used to demonstrate JAK3 activation, so abnormalities in this signalling pathway can be detected at a protein level before genetic analysis. The variability in clinical presentation in these forms of SCID and especially in JAK3 deficiency again underlines the need to identify the molecular defect so that earlier referral for bone marrow transplantation can be made.

The ability to make a molecular diagnosis in utero in known carriers has also changed the treatment options for these types of SCID. Tissue typing can be carried out on DNA obtained from a CVS sample and used to identify potential donors. If a genotypically identical HLA match is available, transplantation can be undertaken almost immediately after birth before the onset of any infectious complications. If no related or unrelated donors are available, parental haplo-identical transplantation in utero may be an option. This has been attempted in a small number of SCID patients worldwide, with promising results.46,47 Finally, establishment of the molecular defect opens up the possibility for treatment by somatic gene therapy. SCID caused by ADA deficiency was the first human disease to be treated in this way, although the results of initial trials were disappointing.4851 Advances in vector technology and stem cell transduction have now led to another wave of clinical gene therapy trials, and the first study of gene therapy for X-SCID is currently under way in Europe. Furthermore, correction of the immunological abnormalities by gene transfer in a JAK3 deficient mouse model52 suggests that gene therapy for JAK3 deficiency may also be possible.


Tables 1 and 2 highlight the enormous advances that have been made in the definition of the molecular defects underlying primary immunodeficiencies in the past decade. The identification of SAP as the gene defective in XLP now completes the molecular bases of all the recognised X linked syndromes. Of the autosomally inherited syndromes, only the genes for DiGeorge syndrome, hyper-IgE, and perhaps most importantly, common variable immunodeficiency remain to be elucidated.

The major clinical benefits of this information have primarily been in offering more accurate and rapid molecular diagnoses. The ability to make a molecular diagnosis also increases the options for earlier definitive treatments such as bone marrow transplantation and somatic gene therapy. Finally, as illustrated by the studies on the functions of WASP and the γc/JAK-3 pathway, identification of the gene defect is the first step to understanding the molecular pathogenesis of the immunological abnormalities.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.