- Top of page
- Patients and methods
X-linked agammaglobulinaemia (XLA) is a primary immunodeficiency caused by mutations in the gene coding for Bruton's tyrosine kinase (Btk) and is characterized by an arrest of B-cell development. We analysed Btk protein expression in platelets using flow cytometry and found that normal platelets express large amounts of Btk. Assessment of affected males from 45 unrelated XLA families revealed that platelets of the majority of the patients (37 out of 45 families) had decreased or absent Btk expression, and that platelets from carrier females of these families had both normal and mutated Btk expression, indicating that megakaryocytes in XLA carriers undergo random X-chromosome inactivation. These observations demonstrate that Btk is not crucial for maturation of megakaryocytes and the production of platelets. No correlation between Btk expression in platelets and clinical phenotype was observed in this study. Flow cytometric evaluation using platelets is a simple and rapid method to test Btk expression. It may be used as a screening test for XLA and for carrier detection, followed, if necessary, by more expensive mutation analyses.
X-linked agammaglobulinaemia (XLA), first described by Bruton (1952), remains the prototypic example of a primary immunodeficiency disease in which the defect is limited to B lymphocytes. As a result, affected males have profoundly abnormal antibody production and present with recurrent bacterial infections (Ochs & Smith, 1996). XLA is caused by mutations of Bruton's tyrosine kinase (Btk) which is required for B-cell development (Vetrie et al, 1993; Tsukada et al, 1993). Btk is present in pre-B cells and B lymphocytes, but not in plasma cells and T lymphocytes. Btk is a member of the cytoplasmic Tec family of tyrosine kinases and consists of a pleckstrin homology (PH) domain, a Tec homology (TH) domain, a Src homology 3 (SH3) domain, a SH2 domain and a SH1 (kinase) domain. Mutations of Btk, which include missense and nonsense mutations, deletions, insertions and splice site mutations (Vetrie et al, 1993), are distributed throughout the gene and may result in complete absence of the protein, or the presence of truncated or full-length non-functional proteins.
Because of this variability in the clinical phenotype, the diagnosis of XLA may be difficult in some cases (Ochs & Smith, 1996). The most decisive approach for confirming the diagnosis of XLA is mutation analysis of Btk. Once the mutation in a given family has been determined, carrier females from that family can be readily identified. However, sequence analysis of the Btk gene requires a specialized laboratory, is time consuming and may be difficult. A simple screening test that allows easy identification of XLA patients and carrier females has recently been designed, based on the fact that monocytes express Btk and, in contrast to B cells, undergo random X-chromosome inactivation (Futatani et al, 1998). To further simplify the screening procedure, we have studied Btk in platelets and found that normal platelets have large amounts of Btk, that platelets of the majority of XLA patients have decreased or absent Btk and that platelets of carrier females have both normal and mutated Btk, suggesting that X-chromosome inactivation is random in megakaryocytes.
- Top of page
- Patients and methods
The clinical diagnosis of XLA is based on a history of recurrent bacterial infections starting in infancy, low serum immunoglobulin levels and a decreased number of circulating B cells, with or without a positive family history. The diagnosis is confirmed by demonstrating a mutation within the Btk gene. Mutation analysis, however, requires a specialized laboratory, is time consuming and technically difficult. In some instances, patients with classic XLA were found to have normal sequence analysis of the coding region of Btk, but complete absence or reduced levels of protein (Hashimoto et al, 1996; Gaspar et al, 1998; Holinski-Feder et al, 1998). Alternative methods have been used successfully to establish a molecular diagnosis of XLA in patients with hypogammaglobulinaemia and low B-cell numbers, including in vitro kinase assays (Gaspar et al, 1998), linkage analysis (Kwan et al, 1994) and X-chromosome inactivation studies of obligate XLA carriers (Fearon et al, 1987; Conley & Puck, 1988; Allen et al, 1994). A novel strategy to establish a diagnosis has been introduced by Futatani et al (1998), who demonstrated that XLA patients have normal numbers of peripheral blood monocytes that lack Btk and that carrier females have circulating monocytes that show random X-chromosome inactivation. In contrast, circulating B cells of carriers are known to consistently undergo non-random X-chromosome inactivation (Fearon et al, 1987; Conley & Puck, 1988; Allen et al, 1994). To analyse peripheral blood monocytes for the presence of Btk, the cell membranes were stained with a monocyte-specific (anti-CD14) mAb and the cytoplasm stained with the Btk-specific mAb 48–2H (Futatani et al, 1998).
In this study, we used flow cytometric analysis to evaluate platelets from normal controls and from XLA patients with known mutations for the presence of Btk and to test their female relatives for possible carrier status. The presence of large amounts of Btk in circulating platelets made this analysis possible. As XLA patients have normal platelet counts, a blood sample of less than 5 ml is sufficient to harvest enough platelets for Btk analysis and mononuclear cells for subset analysis, and to isolate mRNA and genomic DNA for sequence analysis. The observation that nearly 100% of normal control platelets express Btk demonstrates that Btk is stable in platelets throughout their lifespan. Although platelets are known to actively synthesize mRNA and protein (Kieffer et al, 1987), very little if any of the platelet-associated Btk is synthesized de novo as eukaryotic mRNA has an average half-life of between 15 min and 17 h (Rajagopalan & Malter, 1997). Thus, most of the platelet-associated Btk must be generated prior to the formation of platelets, which have a lifespan of 8–10 d (Leeksma & Cohen, 1956; Schmidt et al, 1985). The role of Btk for the function of platelets is unknown. Interestingly, XLA patients who lack functional Btk protein owing to mutations of the gene have normal numbers of platelets. Furthermore, there is no disadvantage in the production and survival of Btk-negative platelets as our study has shown that megakaryocytes of carrier females have random X-chromosome inactivation. It has been reported recently that, during platelet activation, Btk is tyrosine phosphorylated. This occurs in platelets that are stimulated with thrombin through engagement of the αIIb/β3 integrin and activation of phosphoinositide 3-kinase (Laffargue et al, 1999), or with collagen through the collagen receptor (Oda et al, 2000). These observations indicate the involvement of Btk in signalling pathways that are downstream of the adhesive receptors of platelets. Although XLA patients do not suffer from increased haemorrhage, a recent report found diminished responses in platelet aggregation, dense granular secretion and calcium mobilization in XLA platelets, following stimulation through the collagen receptor glycoprotein VI (GPVI) (Quek et al, 1998). The fact that Wiskott–Aldrich syndrome protein (WASP) (which acts downstream of Btk) can be phosphorylated in the absence of Btk further suggests the existence of alternative signalling pathways, e.g. tyrosine kinases such as Tec kinase that bypass Btk in platelets (Oda et al, 2000).
The patients and carrier females included in this study were derived from 45 consecutively seen families with one or more affected males and with known mutations of Btk, and their female relatives. More than half the mutations resulted in single amino acid substitutions. Nonsense mutations, genomic insertions and deletions, and splice site mutations were represented at proportions similar to those observed in the European XLA registry (Vihinen, 1996; Lappalainen et al, 1997; Vihinen et al, 1997). The staining of platelets with mAb 48–2H followed by flow cytometric analysis confirmed the diagnosis of XLA in 37 of the 45 families (78%) with known BtK mutations. Simultaneous staining of circulating monocytes with the same mAb showed identical results, demonstrating that platelets and monocytes are equally useful to screen for Btk mutations. Using the same reagents, a study of Btk expression by monocytes performed in Japan led to the identification of XLA patients in 98% of the families studied (Futatani et al, 1998). This discrepancy is due to a difference in the target population. Of the Japanese group of 41 families, only seven had missense mutations and only one of those had normal expression of Btk (type III). In contrast, in the present study of XLA patients residing in North America, about 50% (22 out of 45) had missense mutations, and affected males in eight of those families showed normal expression of Btk in both platelets and monocytes.
Three patterns of Btk expression were observed in our patient population. A markedly reduced amount or complete absence of Btk (type I) was found predominantly in patients with nonsense mutations, with insertions and deletions resulting in frame shift and premature termination, and in most patients with splice site mutations. Reduced but substantial expression of Btk that allowed differentiation from normal controls (type II) was present in 9 out of 22 families with missense mutations, one out of seven families with nonsense mutations, two out of six families with splice site mutations and in a single family with a mutation affecting the promoter region. Normal or slightly reduced Btk expression (type III) that did not allow differentiation of patients from normal controls or the identification of carrier females using flow cytometry was observed only in families with missense mutations (eight families).
Reduced Btk mRNA or protein expression by PBMCs has been reported in patients with mutations of Btk that resulted in early termination of transcription (Hashimoto et al, 1996; Vorechovsky et al, 1997; Gaspar et al, 1998), possibly owing to the rapid decay of mRNA containing premature stop codons (Peltz et al, 1993; Pulak & Anderson, 1993; Muhlrad & Parker, 1999). The reduced amount of Btk found in 14 of our 22 families with missense mutations is probably as a result of unstable Btk that degrades faster than wild type (Saffran et al, 1994). Gaspar et al (1998) have also observed defective Btk expression in PBMCs from patients with missense mutations by Western blot using polyclonal anti-Btk antibody. Some missense mutations or mutations resulting in in frame deletions or insertions are expected to code for mutated Btk with a reduced affinity for some anti-Btk mAbs but not for others. Such an example is family B-30, which has a mutation within the SH3 domain that causes an in frame deletion of 21 amino acids, resulting in a truncated protein that can be detected by a polyclonal anti-Btk antibody (Zhu et al, 1994), but not by the mAb 48–2H used in this study. Similar findings of discrepancies in results when polyclonal or monoclonal antibodies were used have been observed in X-linked hyper-IgM syndrome (Seyama et al, 1998). As reported by others (Gaspar et al, 1998), there was no recognizable correlation between clinical phenotype and the presence or absence of mutated Btk. The exception was one patient who presented with a mutation in the promoter region of Btk that presumably interferes with the binding of the nuclear protein, PU.1. This patient's platelets and monocytes contained reduced but substantial amounts (type II) of wild-type Btk, which may explain his mild clinical phenotype. De Weers et al (1997) have observed another XLA patient with a mutation within the promoter region and a mild phenotype.
In female carriers of some X-linked conditions such as XLA or X-linked severe combined immunodeficiency (SCID) (Puck et al, 1987; Puck & Willard, 1998), the expected random X inactivation fails to occur in the lymphocyte population targeted by the gene defect. If the gene defect results in a selective disadvantage in the proliferation and survival of the targeted cell lineage, a ‘marked skewing’ of the activation in favour of the X chromosome with the wild-type gene is observed. Such non-random X inactivation has been demonstrated in B lymphocytes of females that are carriers for XLA (Fearon et al, 1987; Conley & Puck, 1988; Allen et al, 1994; Puck & Willard, 1998), but not in their monocytes (Futatani et al, 1998). In this study, we have demonstrated that platelets of female carriers for XLA undergo random X-chromosome inactivation, suggesting that megakaryocytes can develop, mature and generate platelets equally well with or without functional Btk. Consistent with this observation is the finding that XLA carrier females have two populations of platelets, one expressing normal Btk, the other expressing the mutated Btk. The ratio of the two populations varied greatly among female carriers but no tendency of one-directional skewing was observed, further supporting the postulate that absence of Btk does not affect platelet production by megakaryocytes or the survival of circulating platelets.
The expression of two populations of Btk (wild and mutated) during flow cytometry accurately identified carrier females. However, the demonstration of normal Btk expression by a female at risk belonging to an XLA family with a type I or type II pattern does not exclude carrier status. We have identified two females from two unrelated families in which the affected males showed a type I pattern of Btk expression, who both carried the expected mutation (a nonsense mutation and a one base pair deletion respectively) but showed normal Btk expression by their platelets as well as monocytes. This ‘false negative’ result was due to extreme skewed X chromosome inactivation resulting predominantly in the use of the normal X chromosome. Such skewing of X-chromosome inactivation in normal females has been well recognized (Busque et al, 1996; Tonon et al, 1998). A skewing of X-chromosome inactivation of over 90% has been observed in 7% of normal females (Racchi et al, 1998). Thus, it is to be expected that there are female carriers (< 10% in our study) with extremely skewed X-chromosome inactivation in favour of the normal X chromosome, making it impossible to detect the carrier status based on flow cytometric analysis of Btk in peripheral blood platelets or monocytes. For genetic counselling, it is imperative to proceed to mutation analysis in those females at risk of carrier status whose flow cytometric analysis of Btk in platelets shows a normal pattern.
These results demonstrate the usefulness but also the limitation of flow cytometry to analyse Btk expression in platelets as a rapid screening test to identify patients with Btk mutations and carrier females. The simplicity of the test allows the screening of any male with a history of recurrent infections, low serum immunoglobulin levels and decreased numbers of circulating B lymphocytes. If the binding of mAb 48–2H is reduced or absent, the diagnosis of XLA is highly probable, and mutation analysis should be considered as an option to confirm the diagnosis. Because a normal Btk expression does not rule out XLA, as demonstrated in patients with selected missense mutations, sequence analysis is recommended as the final tool. In those families where the affected males express Btk abnormally, carrier females can be identified successfully. The amount of blood required for this analysis is small enough to study infants and, if combined with mutation analysis, can effectively identify affected males and carrier females.