c-Myb binds to a sequence in the proximal region of the RAG-2 promoter and is essential for promoter activity in T-lineage cells

Abstract

The RAG-2 gene encodes a component of the V(D)J recombinase which is essential for the assembly of antigen receptor genes in B and T lymphocytes. Previously, we reported that the transcription factor BSAP (PAX-5) regulates the murine RAG-2 promoter in B-cell lines. A partially overlapping but distinct region of the proximal RAG-2 promoter was also identified as an important element for promoter activity in T cells; however, the responsible factor was unknown. In this report, we present data demonstrating that c-Myb binds to a Myb consensus site within the proximal promoter and is critical for its activity in T-lineage cells. We show that c-Myb can transactivate a RAG-2 promoter-reporter construct in cotransfection assays and that this transactivation depends on the proximal promoter Myb consensus site. By using a chromatin immunoprecipitation (ChIP) strategy, fractionation of chromatin with anti-c-Myb antibody specifically enriched endogenous RAG-2 promoter DNA sequences. DNase I genomic footprinting revealed that the c-Myb site is occupied in a tissue-specific fashion in vivo. Furthermore, an integrated RAG-2 promoter construct with mutations at the c-Myb site was not enriched in the ChIP assay, while a wild-type integrated promoter construct was enriched. Finally, this lack of binding of c-Myb to a chromosomally integrated mutant RAG-2 promoter construct in vivo was associated with a striking decrease in promoter activity. We conclude that c-Myb regulates the RAG-2 promoter in T cells by binding to this consensus c-Myb binding site.

FIG. 1

c-Myb transactivates the RAG-2 promoter through a proximal-region binding site. (A) Critical sequences in the murine RAG-2 promoter. The major transcription start site is numbered +1. The boxed regions represent sequences important for promoter activity in transiently transfected Jurkat T cells (32). The human and murine proximal-region DNA sequences are shown, with the BSAP binding site in boldface and a potential c-Myb site indicated by the thin horizontal bar above the sequence. The thick bar denotes a 20-bp region of complete identity between murine and human sequences. The sites and DNA sequences of four substitution mutations (labeled 1, 2, 3, and 4) are shown in boxes below the sequence. (B) Diagram of the transient-transfection assay vector for RAG-2 promoter activity. Each substitution mutation was analyzed in the context of the full −279 to +123 promoter. The arrow indicates the transcription start site. (C) Luciferase assay analysis of transient transfections into 2017 murine pre-T cells. The cells were cotransfected with the RAG-2 promoter-reporter construct and either empty expression vector (−) or c-Myb expression vector (+). Two 5′ promoter deletion mutants (−71/+123 and −45/+123) and four 5-nucleotide substitution mutations (labeled 1 to 4) were also tested. The results are shown as the fold increase in luciferase activity with c-Myb cotransfection compared to the activity with empty vector (the actual luminometer unit readings were 2,759 ± 434 for the wild-type promoter cotransfected with an empty expression vector and 194,886 ± 228 for the wild-type promoter cotransfected with the c-Myb expression vector). The data shown are the average (+ standard deviation) of two experiments, each performed in duplicate and adjusted for transfection efficiency using β-galactosidase. (D) Luciferase assay analysis of transient transfections into 293T human embryonic kidney cells as described for panel C. The actual luminometer unit readings were 1,723 ± 96 for the wild-type promoter cotransfected with an empty expression vector and 16,573 ± 67 for the wild-type promoter cotransfected with the c-Myb expression vector.

FIG. 2

Jurkat T cells express a readily detectable amount of RAG-2 mRNA. Total RNA was purified from Jurkat, HeLa, and 293 cells and converted to cDNA by random-primed reverse transcription. The cDNAs were analyzed by multiplex PCR for β-actin and RAG-2 transcripts. On top is a digital image of an ethidium bromide (EtBr)-stained gel, and on the bottom is a phosphorimage of the same gel blotted to a membrane and probed with an internal RAG-2-specific radiolabeled oligonucleotide.

FIG. 3

Anti-c-myb antibody specifically enriches RAG-2 promoter DNA sequences in a (ChIP) assay. Chromatin proteins were cross-linked to DNA in intact Jurkat T cells (A) and in freshly purified mouse thymocytes (B) by formaldehyde treatment, and purified nucleoprotein complexes were fractionated using either anti-Myb antibody (lanes 1), nonspecific IgG (lanes 2), or no antibody (lanes 3). The precipitated DNA fractions were analyzed by PCR for the presence of the proximal RAG-2 promoter region, RAG-2 major exon DNA, or a region of the Ig kappa locus. In each case, the input DNA was used as a positive control (lanes 4). Amplification products were analyzed on a 2% agarose gel and visualized by ethidium bromide staining. IP, immunoprecipitate.

FIG. 4

The RAG-2 promoter proximal region is occupied in the vicinity of the c-Myb site in vivo. Purified nuclei from Jurkat (lanes 5 and 6) and HeLa (lanes 1 and 2) cells or purified genomic DNA from Jurkat cells (lanes 3 and 4) were treated with increasing concentrations of DNase I. DNase-treated DNAs were then repurified and subjected to linker ligation and PCR amplification. Radiolabeled extension products were electrophoresed on a denaturing polyacrylamide gel alongside a DNA sequencing ladder, and the dried gel was analyzed by phosphorimaging. This ligation-mediated PCR analysis reveals DNase-sensitive sites on the top strand of the promoter through the use of a radiolabeled bottom-strand primer. The numbers indicate positions with respect to the transcription start site. The arrows point to in vivo DNase I-hypersensitive sites in Jurkat cells, and the position of the c-Myb binding site is indicated (box). The DNA sequencing ladder shows the bottom-strand sequence.

FIG. 5

The c-Myb binding site is essential for RAG-2 promoter activity in an integrated reporter construct in Jurkat T cells. (A) Reporter construct used in the stable-transfection assay. Linearized constructs contain a GFP cDNA and the wild-type or mutant 3 RAG-2 promoter (−279 to +123) in the presence or absence of the TCR β locus enhancer (Eβ). Ins, the 1.2-kb chicken β-globin insulator sequence (10), included to prevent the neomycin resistance cassette (PGK-neo) from influencing GFP activity. (B and C) Reporter constructs containing the wild-type or mutant 3 RAG-2 promoter in the presence or absence of Eβ were transfected into Jurkat cells by electroporation. Transfected cells were selected as a pool in G418 and then analyzed by flow cytometry for GFP expression. The fluorescence-activated cell sorter histograms of untransfected (dashed line) and wild-type promoter-transfected cells (bold solid line) are coincident. The thin solid line represents transfectants containing the wild-type promoter and Eβ, and the dotted line represents the mutant 3 promoter and Eβ.

FIG. 6

The c-Myb site in the proximal region is essential for in vivo recruitment of c-Myb to the RAG-2 promoter. Pools of G418-resistant cells containing the wild-type (A) or mutant 3 (B) RAG-2 promoter analyzed in Fig. 5 were subjected to the ChIP assay using anti-c-Myb antibody (Ab) (lanes 1) or nonspecific IgG (lanes 2). Mock (lanes 3) indicates samples processed in the absence of antibody, and input (lanes 4) is DNA prior to immunoprecipitation. Primers specific for either the transfected or endogenous RAG-2 promoter were used in PCR analysis of the various precipitated chromatin fractions. Images of ethidium bromide-stained gels are shown.