Mutations of the Igβ gene cause agammaglobulinemia in man

As part of an ongoing effort to extensively genotype agammaglobulinemic patients for known and candidate genes causing agammaglobulinemia (23), we found for the first time a homozygous mutation in the Igβ-encoding gene in a patient clinically diagnosed with agammaglobulinemia. The patient, a 20-yr-old Italian male, is the first-born child of healthy parents without known consanguinity. He has two healthy sisters, and no positive family history for primary immunodeficiencies is present in the pedigree. The pregnancy and delivery were uneventful. At the age of 8 mo, the patient was hospitalized because of pneumonia of the left lobe and Salmonella-caused enteritis. Immunological work up showed a marked hypogammaglobulinemia (IgG, 100 mg/d; IgA and IgM, undetectable) and absence of peripheral CD19-positive cells (<1%), in the presence of normal numbers of T and NK cells; T cell function, assessed by mitogen-induced proliferation, was normal. Intravenous immunoglobulin replacement therapy was initiated at the age of 8 mo. During follow up, despite appropriate immunoglobulin substitution therapy, the patient suffered several episodes of bacterial conjunctivitis, acute otitis media, sinusitis, and bronchitis. A computed tomography scan of the nasal sinuses showed a pattern compatible with chronic sinusitis, whereas no bronchiectasis was observed. BTK gene sequence analysis was normal.

Genomic DNA from this patient was analyzed by PCR amplification and direct sequencing of exons and exon–intron junctions of genes encoding for the preBCR components. We found a homozygous C>T nucleotide substitution in exon 3 of the B29 gene, encoding for the Igβ protein (Fig. 1), at position c.238 (GenBank accession no. M89957) corresponding to amino acid 80. The mutation causes the replacement of Gln80 with a stop codon within the extracellular immunoglobulin domain of Igβ, preventing the expression of the functional transmembrane protein and possibly interfering with the assembly of the preBCR on cell surface.

Both parents of the patient were heterozygous for this mutation. The Gln80X mutation was not detected in DNA samples obtained from 90 healthy controls; in addition, sequence analysis of the whole B29 gene reveals the presence of a common haplotype represented by a T>C silent change at codon 122 in cis with a 3′-untranslated region T>C transition, which is present at a high frequency (f_CC = 0.76) in the 90 controls. Our patient is homozygous for the less common 122/T and 3′-untranslated region /T haplotype, suggesting the possibility of a founder effect for the mutated allele. Although not reported, consanguinity of patient's parents cannot be excluded.

It has been shown that efficient transport of IgM to the cell surface requires coexpression of Igβ furthermore, IgM and Igβ alone are sufficient to reconstitute antigen-specific signal transduction by Igβ (24). The effect of the Gln80X Igβ mutation on BCR assembly was assessed by a reconstitution experiment in Drosophila melanogaster S2 cells.

The human Igβ cDNA was cloned in an appropriate expression vector, and D. melanogaster S2 cells were cotransfected with mouse λ1LC, μHC, FLAG-tagged Igα, and either human wild-type Igβ or mutant Gln80X Igβ. After induction of expression of the recombinant proteins, cells were harvested and reconstitution of the IgM BCR was detected by staining with a phycoerythrin (PE)-conjugated anti-FLAG antibody that binds to Igα and a biotinylated anti-IgM antibody. PE- and Cy5-conjugated streptavidin double-positive cells indicate the expression of the BCR on the cell surface.

Fig. 2 shows that the wild-type human Igβ chain is as efficient as the mouse Igβ chain at promoting assembly of the IgM BCR complex on the cell membrane (compare the bottom left and top right graphs). On the contrary, insertion of the Gln80X mutation in Igβ prevents reconstitution of the IgM BCR on the cell surface, which is similar to expression of Igα only (compare the bottom right and top left graphs).

S2 cells cotransfected with FLAG-tagged Igα and wild-type Igβ and stained with the PE-conjugated anti-FLAG antibody showed that the human Igβ alone was not able to carry Igα to the cell surface, proving that, in the aforementioned experiments, any Igα chain detected at the S2 cell surface is, indeed, part of the assembled BCR complex (unpublished data).

PBMCs from the patient carrying the Igβ mutation, three patients with Btk deficiency, three patients with μHC deficiency, and three healthy controls were compared using standard flow cytometry markers. The dramatic reduction of circulating B cells in the Igβ-deficient patient was confirmed using the typical B cell marker CD19 (unpublished data). As expected, PBMCs from the patient, as well as PBMCs from patients with mutations in Btk and μHC, did not express mature IgM on the cell surface (unpublished data).

The analysis of the alternative pan–B cell marker CD22 in the Igβ-deficient patient, as well as in Btk- and μHC-deficient patients, revealed the presence of a cell population expressing CD22 at lower levels compared with healthy controls (unpublished data). The CD22^lo subset was poorly represented (<1%) in 25 healthy controls, whereas in patients CD22^lo cells range from 2 to 4.5% of the gated lymphocytes.

The CD22^lo cells did not express either the pan–B cell marker CD19 or the early B cell marker CD10; because the CD22 molecule was previously found expressed in human basophiles (25), we tested if the CD19−/CD22^lo population in agammaglobulinemic patients was accounted for by basophiles. This turned out to be the case, as demonstrated by Toluidine blue staining of the sorted CD22^lo cells, showing pure basophiles with typical large, dark granules. Furthermore, the CD22^lo cells were positive for CD13, which is a myelomonocytic lineage marker (unpublished data). Therefore, the CD22^lo cell population from the Igβ-deficient patient did not belong to the B lineage.

To determine at which stage B cell development is blocked, bone marrow cells from the patient and an age-matched control were evaluated by cytofluorimetric analysis. CD19+ B cells were markedly reduced in the patient (∼1%) when compared with the healthy control (>10%) (Fig. 3). Two-color immunofluorescence was used to define the percentages of pro-B cells (CD34+CD19−) and preB/immature B cells (CD34−CD19+). The coexpression of CD19 and CD34 showed that >85% of the patient's B cells belong to the pro-B stage of differentiation, whereas <5% of the healthy control's B cells presented the pro-B phenotype. These findings indicate a block of B cell development at the transition from the pro-B to preB stage.

To further characterize the point of arrest of B cell development, expression of TdT and surface Ig were evaluated. Remarkably, the Igβ-deficient patient presented an accumulation of TdT-positive cells that did express CD19. Furthermore, no surface immunoglobulin expression could be detected on the patient's B cells. These results further confirm that the arrest of B cell development in the Igβ-deficient patient occur at the pro-B to preB transition; a similar block at the pro-B stage has been described in patients affected by mutations of other members of the preBCR complex, i.e., μHC deficiency (23) and Igα deficiency (19). Unfortunately, we were unable to perform further biochemical and functional analyses, such as assessment of the extent of V(D)J recombination, because of the low number of bone marrow lymphocyte precursors obtained from the patient biopsy.

In principle, the occurrence of a stop codon in the N-terminal portion of Igβ could result in a truncated protein lacking the transmembrane domain because of premature termination; however, it is also possible that the occurence of a premature stop codon causes nonsense-mediated RNA decay. In any event, the mutation in Igβ prevents the assembly of the IgM BCR on the cell surface, as demonstrated by the reconstitution experiment shown in Fig. 2.

The preBCR checkpoint is essential for B cell development. So far, animal models deficient for different preBCR components were shown to have an arrest at the pro-B to preB cell stage of development, although some models presented a leaky phenotype (e.g., the presence of peripheral B cells in λ5-deficient mice [reference 26]). Human mutations in preBCR components further elucidated the role of the preBCR complex in B cell development, and underlined that animal models are not always concordant with the phenotype of patients carrying mutations in the same gene; the aforementioned human λ5 deficiency presents a severe defect in B cells, with absence of peripheral B cells. On the other hand, human Igβ deficiency closely resembles the phenotype observed in knockout mice in that these mice have no mature peripheral B cells and B cell development is blocked at the pro-B cell/preBI cell stage (22).

In conclusion, this is the first report of a patient with agammaglobulinemia caused by a homozygous Gln80X mutation in Igβ, resulting in a stop codon that abrogates the expression of the preBCR complex on the cell surface. Bone marrow B cell analysis of the Igβ-deficient patient indicated an arrest at the pro-B–preB transition, a phenotype resembling the one observed in other known forms of agammaglobulinemia.

Genomic DNA was extracted from peripheral blood leukocytes using standard techniques. Exons and exon–intron junctions were amplified by PCR using FastStart Taq DNA Polymerase (Roche) under standard conditions. Purified PCR products were sequenced on both strands using a Big Dye Terminator v1.1 Cycle Sequencing kit (Applied Biosystems) and were run on an ABI 3730 Genetic Analyzer (Applied Biosystems). The primers used in this study are available upon request. The M89957 GenBank entry was used as reference sequence.

For the genetic diagnosis, the informed consent was signed by the patient.

For inducible expression of human Igβ in S2 cells, wild-type and mutant cDNAs were cloned into the S2 plasmid pRmHa-3 containing a copper-inducible metallothionein promoter. Total RNA extracted from PBMC was amplified by RT-PCR using the following oligonucleotides: Igβ_cDNA-F 5′-GGTCGGGGACAGAGCAG-3′ and Igβ_cDNA-R 5′-CGACCTGGCTCTCACTCCT-3′, and the corresponding PCR fragment was subcloned into the pRmHa-3 plasmid. The C>T single-nucleotide substitution was introduced into the Igβ cDNA by site-directed mutagenesis.

D. melanogaster S2 Schneider cells were grown in Schneider's Drosophila medium (Life Technologies, Inc.) supplemented with 5–10% FCS at 27°C with atmospheric CO₂ levels. Transient transfection of S2 cells and induction of expression (5 h) was performed as previously described in Wossning et al. (27). For FACS analysis, cells were stained with anti-FLAG PE antibody and biotinylated anti-IgM antibody, followed by Cy5-conjugated streptavidin. To normalize the transfection efficiencies, S2 cells were cotransfected with the vector pDEGFP containing the enhanced GFP (EGFP) coding sequence.

PBMCs from fresh blood samples and bone marrow MCs from patients and healthy controls were isolated by Ficoll-Hypaque gradient centrifugation and stained with the following conjugated anti–human monoclonal antibodies: FITC-labeled anti–human CD19, PE-labeled anti–human CD22, PE-labeled anti–human CD34, PE-labeled anti–human surface Ig, and PE-labeled anti–human TdT. Cells were resuspended in PBS and 0.1% BSA at the concentration of 5 × 10⁵–1 × 10⁶ cells/ml, and then stained with and the anti–human antibodies using standard protocols. Stained cells were washed twice with PBS/BSA and analyzed with a FACSCalibur flow cytometer (Becton Dickinson).

The human experiments have been approved by the hospital Ethical Committee of Spedali Civili of Brescia.

We acknowledge the patient and his family for their cooperation and the members of the Italian Primary Immunodeficiencies Network.

This work was supported by a grant from Fondazione Telethon to S. Ferrari. We thank the Fondazione Golgi, the Associazione Immunodeficienze Primitive and the Centro di Studio Immunodeficienze Primitive Mazio Di Martino from Brescia for their support.

The authors declare there are no conflicts of interest.