Roeder – the coactivator OCA-B, the first cell-specific coactivator, discovered by Roeder in 1992, is unique to immune system B cells

Larry H Bernstein, MD, FCAP, Curator

Leaders in Pharmaceutical Intelligence
The B-cell-specific transcription coactivator OCA-B/OBF-1/Bob-1 is essential for normal production of immunoglobulin isotypes

Unkyu Kim*, Xiao-Feng Qin†, Shiaoching Gong†, Sean Stevens*, Yan Luo*, Michel Nussenzweig† & Robert G. Roeder*
Nature 383, 542 – 547 (10 October 1996);  http://dx.doi.org:/10.1038/383542a0
^* Laboratory of Biochemistry and Molecular Biology, and ^† Laboratory of Molecular Immunology, Howard Hughes Medical Institute, The Rockefeller University, New York, New York 10021, USA

OCA-B was initially identified as a B-cell-restricted coactivator that functions with octamer binding transcription factors (Oct-1 and Oct-2) to mediate efficient cell type-specific transcription of immunoglobulin promoters in vitro ^1–3. Subsequent cloning studies led to identification of the coactivator as a single poly-peptide, designated either as OCA-B (ref. 3), OBF-1 (ref. 4) or Bob-1 (ref. 5). OCA-B itself does not bind to DNA directly, but interacts with either Oct-1 or Oct-2 to potentiate transcriptional activation^1–5. To determine the biological role of OCA-B, we generated OCA-B-deficient mice by gene targeting. Mice lacking OCA-B undergo normal antigen-independent, B-cell differentiation, including appropriate expression of both immunoglobulin genes and other early B-cell-restricted genes. However, antigen-dependent maturation of B cells is greatly affected. The pro- liferative response to surface IgM crosslinking is impaired, and there is a severe deficiency in the production of secondary immunoglobulin isotypes including IgGl, IgG2a, IgG2b, IgG3, IgA and IgE in OCA-B-deficient B cells. This defect is not due to a failure of the isotype switching process, but rather to reduced levels of transcription from normally switched immunoglobulin heavy-chain loci. In accord with the defective isotype production, germinal centre formation is absent in these mutant mice.

References

Cloning, Functional Characterization, and Mechanism of Action of the B-Cell-Specific Transcriptional Coactivator

Oca-B Yan Luo & Robert G. Roeder*
Laboratory of Biochemistry and Molecular Biology, The Rockefeller University, New York, New York 10021
Molecular And Cellular Biology, Aug. 1995;  15(8):4115–4124 0270-7306/95/
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC230650/pdf/154115.pdf

Biochemical purification and cognate cDNA cloning studies have revealed that the previously described transcriptional coactivator OCA-B consists of a 34- or 35-kDa polypeptide with sequence relationships to known coactivators that function by protein-protein interactions. Studies with a recombinant protein have proved that a single OCA-B polypeptide is the main determinant for B-cell-specific activation of immunoglobulin (Ig) promoters and provided additional insights into its mechanism of action. Recombinant OCA-B can function equally well with Oct-1 or Oct-2 on an Ig promoter, but while corresponding POU domains are sufficient for OCA-B interaction, and for octamer-mediated transcription of a histone H2B promoter, an additional Oct-1 or Oct-2 activation domain(s) is necessary for functional synergy with OCA-B. Further studies show that Ig promoter activation by Oct-1 and OCA-B requires still other general (USA-derived) cofactors and also provide indirect evidence that distinct Oct-interacting cofactors regulate H2B transcription.

PNAS  http://www.pnas.org/content/100/15/8868.full.pdf

The tissue-specific transcriptional coactivator OCA-B is required for antigen-dependent B cell differentiation events, including germinal center formation. However, the identity of OCA-B target genes involved in this process is unknown. This study has used large-scale cDNA arrays to monitor changes in gene expression patterns that accompany mature B cell differentiation. B cell receptor ligation alone induces many genes involved in B cell expansion, whereas B cell receptor and helper T cell costimulation induce genes associated with B cell effector function. OCA-B expression is induced by both B cell receptor ligation alone and helper T cell costimulation, suggesting that OCA-B is involved in B cell expansion as well as B cell function. Accordingly, several genes involved in cell proliferation and signaling, such as Lck, Kcnn4, Cdc37, cyclin D3, B4galt1, and Ms4a11, have been identified as OCA-B-dependent genes. Further studies on the roles played by these genes in B cells will contribute to an understanding of B cell differentiation.

Identification and Characterization of a Novel OCA-B Isoform: Implications for a Role in B Cell Signaling Pathways

Xin Yu, Lu Wang†, Yan Luo, Robert G. Roeder
Immunity Feb 2001; 14(2): 157–167   http://dx.doi.org:/10.1016/S1074-7613(01)00099-1

The B cell–restricted function of immunoglobulin (Ig) promoters is mediated mainly by an octamer element (5′-ATGCAAAT-3′) that is conserved in virtually all Ig heavy (H) and light (L) chain gene promoters, as well as in some Ig enhancers (reviewed by Staudt and Lenardo, 1990). However, this same element is also a key central element for transcription of differentially regulated genes that include ubiquitously expressed small nuclear RNA genes (snRNA) and cell cycle-regulated histone H2B genes (reviewed inLuo et al., 1992). The regulatory functions of octamer elements, therefore, are likely dependent on transcription factors that bind this DNA sequence. The well-characterized octamer binding transcription factors include the ubiquitous Oct-1 and the B cell–enriched Oct-2, both of which belong to the POU family and share a conserved DNA binding structure called the POU domain (reviewed by Herr et al. 1988 and Wegner et al. 1993). It was originally thought that Oct-2 would account for the tissue-specific activity of Ig promoters, whereas Oct-1 would facilitate transcription of the ubiquitously expressed genes regulated through octamer elements (e.g., snRNA and histone H2B genes) Staudt et al. 1986, Cockerill and Klinken 1990 and Murphy et al. 1992. However, subsequent biochemical Pierani et al. 1990 and Luo et al. 1992 and genetic (Corcoran et al., 1993)analyses clearly demonstrated that this was not the case. Instead, the promoter specificity was shown to be due to an Oct-1 interacting factor called OCA-B (Luo et al., 1992), and the purification of related p35 and p34 isoforms with apparently equivalent activity in vitro (Luo and Roeder, 1995) set the stage for further studies of the structure and function of OCA-B.

Subsequent to the biochemical identification of OCA-B and its mechanism of action, cognate cDNAs were cloned using both biochemical (Luo and Roeder, 1995) and genetic screening Gstaiger et al. 1995 and Strubin et al. 1995 methods. Analyses of recombinant OCA-B (p34) function in cell-free systems and in transfection assays confirmed both physical and functional interactions with Oct-1 and Oct-2 (via their POU domains) on Ig promoters Gstaiger et al. 1995, Luo and Roeder 1995 and Strubin et al. 1995 and led to the definition of an N-terminal OCA-B domain that interacts with the Oct POU domainCepek et al. 1996, Gstaiger et al. 1996, Babb et al. 1997 and Chaseman et al. 1999 and a C-terminal activation domain that acts synergistically with Oct activation domains to recruit additional coactivators (Luo et al., 1998).

The physiological roles of OCA-B were further investigated by genetic disruption of OCA-B expression in mice Kim et al. 1996, Nielson et al. 1996 and Schubart et al. 1996. These studies showed that, although not required for early B cell development, OCA-B functions are essential both for germinal center formation and for efficient secondary Ig isotype production (including IgGs, IgA, and IgE). In accordance with the biochemical function of OCA-B in activating Ig promoter transcription, it has been found that the decrease of secondary antibody production in OCA-B-deficient mice is largely due to reduced levels of transcription from normally switched IgH chain loci, rather than a reduced capacity for class switching events per se Kim et al. 1996 and Schubart et al. 1996. Recent results further demonstrated that OCA-B plays an essential role in efficient transcription from switched IgH loci by directly regulating 3′ IgH enhancer function in conjunction with Oct-1 or Oct-2 Tang and Sharp 1999 and Stevens et al. 2000b. On the other hand, the lack of germinal center formation in OCA-B-deficient mice cannot be explained by reduced Ig isotype production, since these are two independent events in B cell development (Vajdy et al., 1995). Therefore, OCA-B may regulate germinal center formation by activating the expression of other target genes or by mediating signal pathways that in turn trigger a specific genetic program. At least two lines of evidence support this idea: (1) B cells lacking OCA-B are defective in the proliferative response to surface IgM cross-linking (Kim et al., 1996); (2) OCA-B expression, which is very low in early B cells but high in activated B cells in vivo, can be dramatically and synergistically induced in naive B cells by B cell stimuli (CD40L, Ig cross-linking, and IL4) that are required for germinal center formation (Qin et al., 1998).

Our findings in this report raise the possibility that OCA-B may be directly involved in B cell signaling pathways through novel mechanisms. We report the presence of a novel isoform of OCA-B (p40) that results from utilization of an upstream alternative translation initiation codon and that serves as a precursor to the p35 isoform of OCA-B. Relative to the conventional p34 OCA-B isoform, p35 shows distinct protein modification, subcellular localization, and transcriptional coactivator properties. The unique features of p35 suggest a novel function for this molecule in signal transduction.

Synergism with the Coactivator OBF-1 (OCA-B, BOB-1) Is Mediated by a Specific POU Dimer Configuration

Alexey Tomilin1, 2, #, Attila Reményi2, 4, #, Katharina Lins1, Hanne Bak2, Sebastian Leidel1, Gerrit Vriend3, Matthias Wilmanns4, Hans R Schöler1, 2
Cell   Dec 2000; 103(6):853–864  doi:10.1016/S0092-8674(00)00189-6

Development of multicellular organisms is characterized by an intricate series of genetic and epigenetic events that generate the complex adult body from the unicellular zygote. A refined and sophisticated regulatory network that is established during embryogenesis reflects the complexity of organisms. Although embryonic development is a multistep process characterized by the sequential activation and repression of many genes, only a relatively small number of transcription factors are responsible for regulating the expression of developmental genes. This diversity in transcriptional control by a limited array of transcription factors is achieved through a complex network of interactions between these proteins and specific DNA sequences found in promoters and enhancers of developmental genes. The primary structure of these DNA elements defines the composition and architecture of the transcriptional activation complexes that ultimately control gene expression in the appropriate temporo-spatial context of the developing organism. For example, nonsteroid members of the nuclear receptor superfamily that possess a zinc-finger DNA binding domain operate by binding to the hormone response elements (HREs). HREs consist of two minimal core hexad sequences, AGGTCA, which can be configured into various functional motifs. The orientation and spacing between these two hexamers as well as subtle differences in their sequence dictate the identity and the mode (monomer, hetero-, or homodimer) of nuclear receptor binding that results in diverse effects on transcription (Mangelsdorf and Evans 1995).

The operation of members of the POU domain family of transcription factors is also highly dependent on the nature of cognate DNA elements. The 160 amino-acid-long DNA binding domain of these proteins is composed of two structurally independent subdomains: the POU-type homeodomain (POU-homeo or POU_H), and the POU-specific domain (POU_S) that are connected by a flexible linker region (27 and 36). POU domain proteins demonstrate impressive versatility in how they regulate transcription. This is due to several, often interdependent, factors: (1) flexible amino acid–base interaction, (2) variable orientation, spacing, and positioning of DNA-tethered POU subdomains relative to each other, (3) posttranslational modification, and (4) interaction with heterologous proteins (Herr and Cleary 1995).

POU domain proteins are able to bind to DNA cooperatively, thus conferring additional functional variability. The homo- and heterodimerization of Oct-1 and Oct-2 on immunoglobulin (Ig) heavy chain promoters (V_H) provided evidence of cooperativity, with a yet unknown dimer arrangement (13, 16 and 23). The cis-elements are considered to consist of low-affinity heptamer and high-affinity octamer sites separated by two nucleotides (AT).

The pituitary-specific POU domain protein Pit-1 binds to DNA either as a homodimer or as a heterodimer with Oct-1 (Voss et al. 1991). Crystallographic studies determined the structure of a Pit-1 homodimer assembled on the synthetic motif ATGTATATACAT (referred to here as PitD) that had been derived from the natural Pit-1 cognate element within the prolactin gene promoter (ATATATATTCAT) (Jacobson et al. 1997). The structure of the Pit-1 POU_S and POU_H domains, and their docking onto DNA, are very similar to that observed in the cocrystal of the Oct-1 POU domain monomer with the octamer site (ATGCAAAT, Klemm et al. 1994). The Oct-1 POU_S domain recognizes the ATGC subsite whereas the Pit-1 POU_S domain binds to the sequence ATAC. However, the latter subsite lies on the opposite strand and, as a consequence, the orientation of POU_S relative to the POU_H domain is inverted (Jacobson et al. 1997).

Another mechanism outlining cooperative DNA binding by POU proteins was recently determined during the course of an Oct-4 target gene characterization (Botquin et al. 1998). The P alindromic O ct factor R ecognition E lement (PORE), ATTTGAAATGCAAAT (15 bp), of the Osteopontin (OPN) enhancer interacts with an Oct-4 dimer, thereby mediating strong transcriptional activation in preimplantation mouse embryos. Homo- and heterodimerization of other Oct factors like Oct-1 and Oct-6 on the PORE has also been demonstrated.

The aforementioned examples provide evidence of the various ways in which POU domain proteins are able to cooperatively bind to substrate DNA. The particular mode of binding employed is primarily defined by the DNA sequence. To address the question of whether diversity in cooperative binding is reflected in transcriptional regulation, we have assessed and compared the ability of two different types of POU dimers to interact with the coactivator OBF-1 (OCA-B, Bob-1). This coactivator synergistically interacts with Oct-1 and Oct-2 monomers bound to the octamer motif (18, 9, 17 and 33). We have investigated one type of POU dimer that is formed on the PORE and another that is formed on another palindromic DNA motif called MORE (M ore P ORE), ATGCATATGCAT. The data presented in this study provide an example of how POU domain molecules that bind to DNA in the same stoichiometry but in different configurations can differentially recruit a transcriptional coactivator to the promoter resulting in differential transcriptional activation.

B-cell-specific Coactivator OCA-B: Biochemical Aspects, Role in B-Cell Development and Beyond

Cold Spring Harb Symp Quant Biol 1999 64: 119-132;
http://dx.doi.org:/10.1101/sqb.1999.64.119

Laboratory of Biochemistry and Molecular Biology

Selected PublicationsClick here for a printable version (pdf) of this list Reviews General Initiation Factors and Mechanisms for Transcription by RNA Polymerase II Transcriptional Activation by Nuclear Receptors in Homeostasis and Cell Differentiation Transcriptional Regulation During Cell Growth, Proliferation and Differentiation Transcriptional Regulation During B-Cell Differentiation and Immune Responses Transcriptional Activation via Chromatin Structural Modifications