Cutting Edge: BASH, A Novel Signaling Molecule Preferentially Expressed in B Cells of the Bursa of Fabricius

Antibodies

The anti-chicken IgM Ab, M4, was provided from Dr. C.-L. H. Chen (University of Alabama at Birmingham, AL). The anti-porcine Syk polyclonal antiserum, which was shown to cross-react with chicken Syk (9), was a gift from Dr. T. Kurosaki (Kansai Medical University, Osaka, Japan). The anti-T7 epitope Ab, anti-phosphotyrosine (pY) Ab (RC20), and polyclonal anti-Shc Ab directed against the SH2 domain of human Shc were purchased from Novagen (Madison, WI), Signal Transduction Laboratories (Lexington, KY), and Upstate Biotechnology (Lake Placid, NY), respectively.

Isolation of chicken BASH cDNA and construction of an expression plasmid

The cDNA from chicken bursal B cells was subtracted by the cDNA from spleen lymphocytes using the PCR-select cDNA subtraction kit (Clontech, Palo Alto, CA), and the subtracted PCR products were cloned into pCR2.1 vector (Invitrogen, San Diego, CA). To further screen differentially expressed products, the subtracted fragments were run on an agarose gel and immobilized on duplicate nylon membrane filters, and then hybridized to cDNA probes derived from either a DT40 B cell line or a 132B T cell line poly(A)⁺ RNA. One of the subtracted cDNA fragments hybridized with the DT40 cell cDNA probe but not to the 132B cell cDNA probe was used to screen a DT40 cDNA library constructed in λZAP-XR (Stratagene, La Jolla, CA). The positive clones were subjected to sequencing by the dideoxy-chain termination method with an automatic DNA sequencer (Applied Biosystems, Foster City, CA). A full-length cDNA encoding the short-form of BASH was ligated in-frame with N-terminal double T7 epitope tag in pAT7neo expression vector, which was driven by the chicken β-actin promoter (pAT7neo-BASH). Details of restriction maps and cloning sites of the construct are available upon request.

Northern blot and in situ hybridization analysis

Total RNA samples (10 μg) extracted from tissues and cell lines were electrophoresed in 1.2 M formaldehyde/1.2% agarose gel and transferred to nylon membranes. The blots were hybridized with a ³²P-labeled 900-bp chicken BASH cDNA fragment initially isolated by the subtraction. Whole-mount in situ hybridization was performed essentially as described (10). Digoxigenin-labeled sense and anti-sense 900-bp BASH riboprobes were produced with the Digoxigenin RNA Labeling Kit (Boehringer Mannheim, Mannheim, Germany) according to the manufacturer’s instruction. After whole-mount in situ hybridization, tissues were embedded in Tissue-Tek (Miles, Elkhart, IN), frozen in liquid nitrogen, and sectioned at 10 μm on a Leiz cryostat. Sections were mounted with Mowiol, and photographed on a Olympass BX30 Nomarwsky microscope (Tokyo, Japan).

Immunoprecipitations and immunoblot analysis

DT40 cells were electroporated with pAT7neo-BASH and then stable clones expressing T7-tagged BASH protein were selected by G418 (2 mg/ml). These clones were stimulated with the anti-chicken IgM Ab (15 μg/ml) at 40°C for the indicated period of time and then lysed at 2 × 10⁷ cells/ml in the lysis buffer containing 1% NP40, 10 mM Tris, pH 7.8, 150 mM NaCl, 2 mM EDTA, and protease and phosphatase inhibitors. The lysates were immunoprecipitated with the indicated Abs and then analyzed by Western blotting with the indicated Abs and secondary Abs conjugated with horseradish peroxidase. Immunoreactive proteins were detected by ECL kit (Amersham, Arlington Heights, IL). COS7 cells (5 × 10⁵) were cotransfected with 1 μg of pAT7neo-BASH with either 1 μg of pME-Lyn, pME-Syk (gifts from Dr. H. Nishizumi, Institute of Medical Science, University of Tokyo, Tokyo, Japan), or both plasmids using a TransIT-LT1 transfection reagent (Pan Vera, Masison, WI). After the incubation for 48 h, cell lysates were prepared and analyzed for tyrosine phosphorylation as described above.

Luciferase assay

DT40 cells were cotransfected with 10 μg of a luciferase reporter plasmid driven by seven tandem copies of the NF of activated T cells (NF-AT) response element from the mouse IL-2 gene promoter (NF-AT-Luc; a gift from Dr. K. Arai, Institute of Medical Science, University of Tokyo), together with 15 μg of either empty pAT7neo or pAT7neo-BASH in serum-free RPMI 1640 at a density of 10⁷ cells/400 μl per cuvette with a gene pulser (Bio-Rad Laboratories, Richmond, CA) set at 250 V and 975 μF. After electroporation, the cells were transferred to complete RPMI 1640 and incubated at 40°C for 48 h. Triplicates of 5 × 10⁵ viable cells were then stimulated with anti-IgM Ab and subsequently assayed for luciferase activity, as described previously (11). Light emission was measured in a Lumat LB9501 luminometer (Berthold, Wildbad, Germany).

Isolation of a cDNA fragment whose mRNA is preferentially expressed in B cells of the bursa of Fabricius

We sought to clarify the molecular basis of the unique bursal differentiation process by identifying genes selectively expressed in bursal B cells through PCR-based subtraction combined with differential hybridization. One positive cDNA clone, designated B-1-2, hybridized to a 2.5-kb mRNA transcript that was abundantly expressed in the bursa, not in thymus, bone marrow, or other tissues, except for a low level of expression in spleen and ovary (Fig. 1⇓A). All chicken B cell lines, but no T cell lines expressed this mRNA transcript, although the level of the expression was highly variable among B cell lines. High levels of transcripts were detectable in immature B cell lines such as DT40 and CL18, in which gene conversion persists (12), and TLT-1, whose Ig light chain gene is in germline configuration. In contrast, very low levels of transcripts were detected in relatively mature B cell lines, such as 293B9 and 249L4 (Fig. 1⇓B). In situ hybridization analysis revealed a strong positive signal in B cell follicles in the bursa (Fig. 1⇓C), especially in medullar B cells surrounded by the follicular epithelium (Fig. 1⇓D).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1.

The expression pattern of BASH mRNA. A, Northern blot analysis of total RNA from chicken tissues. The blots were hybridized with the B-1-2 cDNA probe (top) and reprobed with the chicken β-actin cDNA (bottom). B, Northern blot analysis of total RNA from chicken cell lines. B cell lines are CL-18, DT40, TLT1, 293 B9, and 249L4; T cell lines are RP1, JP2, and MSB1. C, Whole-mount in situ hybridization analysis of the bursa of Fabricius, showing that B-1-2 transcripts are localized at B cell follicles. D, Transverse section of the sample in C.

BASH is a novel member of the family of signaling proteins containing SH2 domains

Using a B-1-2 cDNA fragment as a probe, we isolated two full-length cDNA clones with different lengths (GenBank/EMBL/DDBJ accession number AB015289). The long-form cDNA contains the longest open reading frame encoding a 553-amino acid protein, with a calculated molecular mass of 62 kDa (Fig. 2⇓). The shorter form of cDNA lacks 57 nucleotides corresponding to the 19 amino acids located at the N terminus of the long-form protein. These two forms of mRNA transcripts were detectable in the bursa as well as DT40 cells by PCR using a pair of primers flanking the deleted portion (data not shown), although we could not assess the corresponding protein species expressed in B cells due to the lack of a discriminating Ab. The two predicted proteins contain N-terminal acidic domains with seven potential phosphotyrosine-based binding motifs for SH2 domains (13, 14) and a putative SH2 domain in the C-terminal region. The intervening region between the two domains is rich in proline residues and may represent binding sites for proteins containing an SH3 domain (15). Based on these molecular characteristics, this protein was termed BASH (B cell-specific adaptor containing an SH2 domain).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 2.

Deduced amino acid sequences of chicken BASH (B), aligned with mouse SLP-76 (S). Identical and similar amino acids are indicated by black and gray shading, respectively. Dashes denote gaps introduced to optimize similarity. The numbers in the left column indicate the position of amino acid residues. The amino acid residues missing in the short-form of BASH are indicated by a dashed line above the sequence. Tyrosine residues in the potential SH2 domain-binding motifs are shown by closed circles above the sequence. The SH2 domain is underlined.

A database search revealed substantial sequence similarity between BASH and mouse SLP-76, a tyrosine-phosphorylated hematopoietic cell-specific protein (16). The SH2 domain of BASH shows high homology to that of mouse SLP-76 (42% identity). Among seven potential phosphotyrosine-based binding motifs for SH2 domains, three motifs (positions of tyrosine residues: 91, 103, and 115 in the long-form) are conserved between BASH and SLP-76 (Fig. 2⇑).

BASH is tyrosine-phosphorylated after cross-linking of surface-IgM on B cells

Because SLP-76 is known to be tyrosine-phosphorylated upon TCR-cross-linking (16), we examined whether BASH is phosphorylated upon BCR stimulation. The T7-epitope tagged BASH cDNA was transfected into DT40 cells, and then the BASH protein was analyzed by immunoprecipitation with anti-tag Ab. The BASH protein was found to migrate on SDS-PAGE with an apparent molecular mass of 85 kDa, which is larger than the predicted molecular mass of 60 kDa plus the additional 4 kDa of the T7-epitope tag. The anomalous migration may result from posttranslational modification or the abundance of charged amino acids present in the protein. Maximal BCR-induced tyrosine phosphorylation of BASH was detectable after 1 min and was sustained for at least 15 min after BCR stimulation (Fig. 3⇓A). In addition, we found that lysates of the stimulated DT40 cells contained several additional tyrosine-phosphorylated proteins (∼70 kDa, 52 kDa, and 46 kDa), which were specifically coprecipitated with BASH.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 3.

BASH is tyrosine-phosphorylated after BCR stimulation and associates with Shc and Syk. A, Tyrosine phosphorylation of BASH. DT40 cells stably transfected with a T7-tagged BASH cDNA (DT40/BASH) were stimulated with anti-IgM Ab for the indicated periods of time. Whole-cell lysates were immunoprecipitated with an anti-T7 Ab, immunoblotted with an anti-phosphotyrosine Ab (top) and reprobed with an anti-T7 Ab (bottom). B, Coimmunoprecipitation of Shc with BASH. Lysates of DT40/BASH cells, either unstimulated or stimulated with anti-IgM mAb for 1 min, were immunoprecipitated with an anti-T7 Ab and then probed with anti-phosphotyrosine Ab (top) or anti-Shc Ab (bottom). C, Coimmunoprecipitation of Syk with BASH. Immunoprecipitates of unstimulated and stimulated (anti-IgM, 1 min) DT40/BASH cells with anti-T7 (bottom) or anti-Syk (top) Abs were reciprocally probed with the same Abs. D, Lysates from COS7 cells transfected with expression vectors for T7-tagged BASH, Lyn, or Syk in the indicated combinations were immunoprecipitated with anti-T7 Ab and immunoblotted with an anti-phosphotyrosine Ab (top) or an anti-T7 Ab (bottom).

Association of BASH with Shc and Syk

By blotting the BASH immunoprecipitate with Abs against known tyrosine-phosphorylated proteins, we found that the 52-kDa BASH-associated phosphoprotein reacted with an anti-human Shc Ab, suggesting an association of BASH with the chicken Shc homologue. Although this association was seen before stimulation, it was significantly enhanced after BCR cross-linking (Fig. 3⇑B). The 70-kDa phosphoprotein bound to BASH was identified as Syk; BASH was detectable in the anti-Syk immunoprecipitate, and Syk was coimmunoprecipitated with BASH. The BASH-Syk association was enhanced by BCR-stimulation (Fig. 3⇑C).

Both Syk and Lyn are required for the maximal phosphorylation of BASH

To identify the tyrosine kinases responsible for the phosphorylation of BASH following BCR cross-linking, COS7 cells were cotransfected with a BASH plasmid and plasmids encoding Syk and Lyn. Anti-phosphotyrosine immunoblots of the BASH-immunoprecipitate from these cell lysates revealed that coexpression of either Syk or Lyn induced only a very weak tyrosine phosphorylation of BASH (Fig. 3⇑D). In contrast, coexpression of Syk and Lyn dramatically increased the tyrosine phosphorylation of BASH, indicating that both kinases are required for maximal activity.

Overexpression of BASH interferes with NF-AT activation upon BCR-stimulation

Because SLP-76, a potential T-cell counterpart of BASH, has been demonstrated to augment TCR signals leading to IL-2 promoter and NF-AT activation (17, 18), we examined whether overexpression of BASH influences the activation of NF-AT, which is also involved in BCR-mediated transcriptional events (19). In contrast to the positive effect of SLP-76 on NF-AT activation in T cells, overexpression of BASH resulted in significant suppression of NF-AT activation when compared with transfection of the control vector (Fig. 4⇓). This finding suggests that BASH may play an inhibitory role in a BCR-mediated signaling pathway leading to the activation of NF-AT.

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 4.

Overexpession of BASH inhibits BCR-mediated NF-AT activation. DT40 cells were cotransfected with NF-AT-Luc plasmid and either the empty vector or a BASH expression vector. After 48 h, cells were either left unstimulated or stimulated with anti-IgM Ab for 6 h and subsequently assayed for luciferase activity. The results are shown as the fold induction of luciferase activity as compared with the activity in unstimulated cells transfected with empty plasmid. Luciferase activity was determined in triplicate in the experiment.

The presence of SH2 and proline-rich domains as well as several potential SH2-binding sites in BASH, in addition to its BCR-mediated tyrosine phosphorylation and association with Syk and presumably with Shc, suggests that BASH functions as an adaptor protein interacting with multiple signaling molecules in BCR-mediated signaling. The abundant expression of BASH in B cells of the bursa of Fabricius, together with its functional involvement in BCR-mediated signal transduction, suggests a critical role of BASH in the regulation of B cell development in the bursa.

Very recently, novel mammalian B cell-specific adaptor proteins, BLNK (20) and SLP-65 (21), have been described. These proteins are very homologous to BASH, particularly in the N-terminal and SH2 domains, although the length and homology of the intervening region have diverged considerably between BLNK/SLP-65 and BASH. In contrast to BASH, BLNK does not associate with Shc and augments NF-AT activation following BCR-stimulation (20), suggesting that BASH may not simply be a homologue of BLNK/SLP-65. Moreover, the expression patterns of the proteins are quite different. BASH is found primarily in immature B cell lines and is expressed only weakly in mature B cells, while the converse is true of BLNK/SLP-65. Therefore, BASH may have evolved to fulfill unique functions in developing B cells in the avian bursa.