Concerted repression of an immunoglobulin heavy - chain enhancer , 3 ' aE ( hs 1 , 2 )

The transcription factor, B-cell-specific activator protein (BSAP), represses the murine immunoglobulin heavy-chain 3' enhancer 3'aE(hsl,2) in B cells. Analysis of various 3'aE deletional constructs indicates that sequences flanking a and b BSAP-binding sites are essential for appropriate regulation of the enhancer. An octamer motif 5' of the a site and a specific G-rich motif 3' of the b site were identified by competition in electrophoretic mobility-shift assays and methylation-interference foot-printing analysis. Site-directed mutagenesis of either the octamer or G-rich sites resulted in the complete release of repression of 3'aE(hsl,2), implicating these two motifs in the repression of this enhancer in B cells. However, when both BSAP-binding sites were mutated, the octamer and G-rich motifs functioned as activators. Moreover, in plasma cells, when BSAP is not expressed, 3'aE(hsl,2) is active, and its activity depends on the presence of the other two factors. These results suggest that in B cells, 3'aE(hsl,2) is down-regulated by the concerted actions of BSAP, octamer, and G-rich DNA-binding proteins. Supporting this notion of concerted repression, a physical interaction between BSAP and octamer-binding proteins was demonstrated using glutathione S-transferase fusion proteins. Thus, concerted repression of 3'aE(hsl,2) in B cells provides a sensitive mechanism by which this enhancer, either individually or as part of a locus-controlling region, is highly responsive to any of several participating factors. Immunoglobulin genes are exquisitely regulated during B-cell development, in processes of DNA recombination, including V(D)J joining and class switch recombination, and in levels of gene expression, including alternative splicing. These events not only occur by intrinsic B-cell-specific developmental pathways but also in response to antigen through interactions with T cells and stromal cells in particular anatomic compartments (1). Heavy-chain genes are presumed to be regulated by proteins that react with cis-acting elements, such as the immunoglobulin heavy-chain (IgH) intronic enhancer, E/x (2), and four other enhancers located 3' of the Ca gene (3-7). Three of these enhancers-3'a enhancer (3'aE) (hsl,2), hs3, and hs4-constitute a locus control region (LCR) active at the plasma cell stage (6). Experiments have so far implicated a role for 3'aE(hsl,2) in class switching and in high levels of IgH expression in plasma cells (8, 9). We have shown that in B-cell lines, 3'aE(hsl,2) is repressed by the transcription factor, B-cellspecific activator protein (BSAP) (Pax 5) (10); in plasma cells, where BSAP is not expressed, 3'aE(hsl,2) is active (11, 12). In addition to its role in limiting the expression of 3'aE(hsl,2), BSAP has been shown to be essential for early B-cell development (as well as the development of the central nervous system) (13, 14). Although transcriptional studies have identified several genes that are apparently positively regulated by BSAP (15-17), including CD19 (18), those BSAP The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact. 4392 gene targets critical for B-cell differentiation have not yet been elucidated (33). In light of the general function of BSAP as a transcriptional activator, we wanted to elucidate the apparent repressive function of BSAP in the context of 3'aE(hsl,2) to gain insight into disparate mechanistic roles that BSAP may play. We have discovered concerted repression of 3'aE(hsl,2) in B cells: BSAP's role as a repressor of 3'aE(hsl,2) depends not only on a and b BSAP binding sites but also on other cis elements that are capable of protein binding, namely an octamer (Oct) motif and a stretch of G residues resembling switch sequences ("G-rich motif'). In the absence of BSAP binding, proteins binding to the Oct and the G-rich motifs function as transcriptional activators. MATERIALS AND METHODS Cell Lines. Murine B-cell lines include BASC (pro-B), 18-81 (pre-B), W231, A-20, and M12.4.1 (B cell), and S194 (plasma cell). All the cell lines were grown in suspension culture at 37°C in 5-7% CO2 in RPMI 1640 medium (BioWhittaker), as described in ref. 11. Electrophoretic Mobility-Shift Assays (EMSAs) and Methylation-Interference Foot-Printing Analysis. EMSAs were done as described (11). For competition with specific a-Oct-1 and a-Oct-2 antisera (Santa Cruz Biotechnology), 1 ,ul of undiluted antisera was added to each reaction mixture and preincubated overnight at 4°C. Methylation-interference footprinting analysis was done as described (11). RNase Protection Analysis. Plasmid vector OVEC-S containing a rabbit f3-globin reporter gene was used for RNase protection assay (19). A complementary oligonucleotide containing the wild-type b binding site of BSAP in 3'aE(hsl,2) (CCCTGGGGTGTTGAGCCACCCATCCTTGCCCATCTCCTGTCATGTCCT) was cloned into OVEC-S and used for SP6 RNA mapping (15). DNA Transfections and Chloramphenicol Acetyltransferase (CAT) Assay. 3'aE(hsl,2) was mutated at various sites using different mutant oligonucleotides, as cited in Fig. 1, and PCR, as described previously (11). Mutant enhancers were subcloned in the CAT expression vector QM 293 (11). DNA transfections used the DEAE-dextran method as described in ref. 11, except that transfection in the A-20 cell line was done as described in ref. 20. At least three to five experiments were done for each cell type. Protein-Protein Interaction Assay. For interaction of BSAP with immobilized glutathione S-transferase (GST) fusion proteins, GST and GST-POU (Oct-2), domain fusion proteins (gifts from H. Singh, University of Chicago) and GST-POU1, GST-POU1, and GST-POU3 (gifts from R. Roeder, Rockefeller University) were used as described by Luo and Roeder (21). BSAP cDNA (a gift from S. Desiderio, The Johns Abbreviations: IgH, immunoglobulin heavy chain; 3'aE, 3'a enhancer; BSAP, B-cell-specific activator protein; CAT, chloramphenicol acetyltransferase; GST, glutathione S-transferase; EMSA, electrophoretic mobility-shift assay. *To whom reprint requests should be addressed. Proc. Natl. Acad. Sci. USA 93 (1996) 4393 930 wt AAACAAACATCCATTTGCATGTGCCCTTGTGA mut AAACAAACATCCtccatacaGTGCCCTTGTGA 737 Oligoa wt CATCATCAATAGGGGTCATGGACCCCAGTCCC Oligo a: mut CATCATCAATcaaGcTtATGGACCCCAGTaCC