Oct-2, a transcription factor expressed in the B lymphocyte lineage and in the developing CNS, functions through of a number of discrete protein domains. These include a DNA-binding POU homeodomain flanked by two transcriptional activation domains. In vitro studies have shown that the C-terminal activation domain, a serine-, threonine- and proline-rich sequence, possesses unique qualities, including the ability to activate transcription from a distance in a B cell-specific manner. In this study, we describe mice in which the endogenous oct-2 gene has been modified through gene targeting to create a mutated allele, oct-2ΔC, which encodes Oct-2 protein isoforms that lack all sequence C-terminal to the DNA-binding domain. Surprisingly, despite the retention of the DNA-binding domain and the glutamine-rich N-terminal activation domain, the truncated protein(s) encoded by the oct-2ΔC allele are unable to rescue any of the previously described defects exhibited by oct-2 null mice. Homozygous oct-2ΔC/ΔC mice die shortly after birth, and B cell maturation, B-1 cell self renewal, serum Ig levels, and B lymphocyte responses to in vitro stimulation are all reduced or absent, to a degree equivalent to that seen in oct-2 null mice. We conclude that the C-terminal activation domain of Oct-2 is required to mediate the unique and indispensable functions of the Oct-2 transcription factor in vivo.

The B cell-restricted transcription factor Oct-2 was discovered through its ability to bind with high affinity to the conserved octamer DNA motifs found in Ig gene promoters and enhancers (1, 2). A strong activator in B lymphocytes, Oct-2 was presumed to act as the primary mediator of Ig gene transcription in B cells, but its role on these potential target genes, and its mechanism of action have not been fully elucidated. Although Oct-2 is described as a B cell-specific protein because of its high level expression in B lineage cells, it is also expressed outside of this cell lineage. Activated T lymphocytes express Oct-2, as do some myeloid cell lines (3, 4). In addition, a yet to be defined, critical, nonhemopoietic site of expression underlies the fact that Oct-2 is required for postnatal survival. Despite these observations, hemopoietic reconstitution of immunodeficient (SCID or recombination-activating gene (Rag)3-1-deficient) mice with control or mutant fetal liver stem cells revealed that only the B cell lineage was impacted significantly by the loss of Oct-2 (5, 6). T cell development and function appeared normal, and hemopoietic colony assays, performed using fetal liver progenitors, indicated that Oct-2 is dispensable for differentiation of myeloid progenitors (L. M. Corcoran and D. Metcalf, unpublished observations). Early B cell development in the bone marrow was normal for oct-2−/− cells, but in the periphery, the mature conventional B lymphocyte and the peritoneal B-1 cell populations were both largely absent (5). In vitro, B cell responses to a number of mitogens were abnormal and in vivo, serum Ig levels were low in both naive and immunized mice (6). The latter was due, at least in part, to the low number of Ig-secreting cells generated in vivo in mutant mice and in vitro upon stimulation (7).

These studies indicated that Oct-2 regulates the transcription of genes that play very significant roles in a number of B cells responses. Surprisingly, the low levels of serum Ig in oct-2−/− reconstituted mice were not due to a direct effect of Oct-2 loss on the expression of Ig genes, because Ig gene expression seems not to require Oct-2, either in primary splenic B lymphocytes or in a B cell line (7, 8). However, a small number of genes have been shown to rely directly on Oct-2 for their expression (9). One of these encodes the cell surface scavenger receptor CD36 (10). In B cells, Oct-2 directly activates the CD36 gene through a promoter octamer motif that cannot recruit the B cell transcriptional coactivator OBF-1/OCA-B/Bob.1 (11). The functional deficiencies of oct-2 null B cells are not the consequence of CD36 dysregulation, however, as CD36−/− mice have a normal B cell compartment (12). The critical genes that Oct-2 regulates in peripheral B cells remain to be identified. Their unequivocal identification is complicated by the overlapping influences of Oct-2 and Oct-1, a ubiquitously expressed factor that binds to DNA through an almost identical DNA-binding domain (13), and by the additional influences of B cell specific coactivators such as OBF-1 (14, 15).

Like most transcription factors, Oct-2 is comprised of a number of functional domains, generally represented as the N-terminal glutamine-rich activation domain, the central bipartite POU homeo DNA-binding domain, and the C-terminal serine-, threonine-, and proline-rich activation domain (16, 17). Although the structures and macromolecular interactions mediated by the transactivation domains have not been determined, the POU homeodomain has been extensively studied. Its multihelical structure embraces the DNA binding site (18, 19), and residues in both the POU-specific and POU homeodomains specifically interact with transcriptional coactivators and components of the basal transcription machinery, to enable assembly of the preinitiation complex and transcription initiation (19, 20, 21, 22).

There is strong evidence from in vitro experiments to suggest that the Oct-2 C-terminal transcriptional activation domain has unique properties, including an ability to activate transcription from a distal position (16, 23, 24, 25). This property is not observed for the N-terminal activation domain of Oct-2, or for any domain of the related protein Oct-1. Furthermore, this special property requires a B cell-specific activity, perhaps a modification or a cofactor. These insights into the importance of this domain came from assays performed in nonlymphoid cells and B cell lines, using ectopic expression of oct-2 and its derivatives, and synthetic reporters to measure transcriptional activity. The studies described here were undertaken to ascertain the requirements for this unique functional domain of the Oct-2 protein in vivo. The unexpected finding is that the phenotype of oct-2ΔC/ΔC mice is identical with that of oct-2 null mice in all assays performed. Thus, unique and indispensable features of the Oct-2 transcription factor lay C-terminal to its DNA-binding domain, in a specialized transcriptional activation domain.

A targeting strategy was chosen that would preserve the N-terminal activation and DNA-binding POU domain of Oct-2 in the targeted locus. In addition, all sequences C-terminal to the DNA-binding domain would be removed in any protein translated from the targeted allele, regardless of oct-2 splicing pattern. Finally, potential influences on oct-2 regulation were avoided by removal of the selection cassette by Cre-mediated deletion.

Three intermediate plasmids were generated to construct the pOct-2ΔC targeting vector. They contained the 5′ oct-2 homology region, the lox P-flanked selection cassette and the 3′ oct-2 homology region, respectively. The first, pBK (5′Oct+XbaSTOP), was obtained by inserting an XbaI “nonsense” linker (5′-GCATCGAGACGAATGTCCGCTTCG-3′; New England Biolabs, Beverly, MA) into the genomic SmaI site (exon 10 as defined by Wirth et al. (26)) of a plasmid containing a 2.6-kb BamHI/KpnI fragment of oct-2 genomic sequence upstream of the KpnI site within intron 10. To generate the loxP-flanked neomycin-resistant/thymidine kinase selection cassette plasmid, ploxPNeoNTRtk, the XhoI/NsiI fragment of the selection vector pPGKneoNTRtkpA (a kind gift of Dr. R. Jaenisch, Whitehead Institute for Biomedical Research, Cambridge, MA, as described in Ref. 27) was substituted for the SalI/XbaI neomycin resistance gene fragment of ploxPneo-1 plasmid (a kind gift from Dr. J. Rossant, University of Toronto, Toronto, Ontario, Canada). The plasmid pBlueOct-3′ (Xba) was obtained by the subcloning a 3.6-kb KpnI/XhoI fragment of oct-2 genomic DNA located 3′ of the DNA-binding domain (exon 10) into a pBlueScript II vector (Stratagene, La Jolla, CA) lacking the polylinker XbaI site. Three fragments were assembled as follows to give rise to pOct-2ΔC (NotI). A BamHI/KpnI fragment from pBK (5′Oct+XbaSTOP), and a KpnI/XhoI fragment from pBlueOct-3′ (Xba) were joined in a three-way ligation to SalI/BamHI linearized pGem4. To introduce a unique NotI site into the vector, a KpnI internal fragment was first removed to eliminate one of two BamHI sites within pOct-2ΔC (NotI), and a NotI linker was put into the remaining BamHI site. Later restoration of the KpnI fragment then gave rise to the final targeting vector, pOct-2ΔC. pOct-2ΔC was linearized with NotI before electroporation into embryonic stem (ES) cells. Both loxP sites and the nonsense mutation within the vector were sequenced to ensure their integrity.

W9.5 ES cells were propagated on embryonic fibroblasts as described in Köntgen and Stewart (28). ES cells (2 × 107) were electroporated (500 μF, 250 V) in the presence of 25 μg of NotI linearized oct-2ΔC targeting vector and plated on irradiated (3000 rad) neomycin-resistant embryonic fibroblasts. G418 selection (200 μg/ml active substance) was initiated 24 h later and individual clones picked after 9–10 days selection, expanded, frozen and screened by Southern blot analysis. Clones that harbored the targeted mutation were expanded for Cre transfection. Targeted ES cells (2 × 107) were electroporated in the presence of 30 μg of pMC-cre (a kind gift of K. Rajewsky as described in Ref. 29). 1-(2-deoxy-2-fluoro-b-d-arabinofuranosyl)-5-iodouracil (FIAU) selection (0.1 μg/ml) was initiated 24 h later and individual clones picked after 9–10 days of selection. The cells were expanded and DNA prepared for Southern blot and PCR analysis (30).

A 1200-bp StuI/BamHI oct-2 genomic fragment was used for Southern blot analyses. ES cell DNA was digested with XbaI and hybridized with this oct-2 probe, which contained sequences 5′ to the oct-2 homology region of the targeting vector (position of restriction sites and probe are shown in Fig. 1,A). Clones bearing the desired mutation should contain a 3.9-kb band due to the presence of the XbaI-nonsense linker within the targeting vector, in addition to the wild-type allele (4.6-kb XbaI band; see Fig. 1 B, left panel).

FIGURE 1.

Targeting and screening strategies, and confirmation of Oct-2 protein alteration. A, A map of the Oct-2 locus surrounding the region to be mutated is shown, with exons shown as boxes (top); those encoding the DNA-binding domains are labeled. Arrows indicate the locations and orientations of PCR primers used for screening and brackets show the location of a fragment used as a probe in Southern blot analyses. The XbaI linker containing the nonsense mutation (Xba/STOP) was introduced into the SmaI site in the POU homeodomain. Also shown are the targeting vector with its double selection cassette (see Materials and Methods), and the predicted product of Cre-mediated deletion of this cassette, leaving the targeted allele shown (bottom). The wild-type Oct-2 protein (bottom) and the truncated Oct-2ΔC mutant are depicted, approximately to scale, with the glutamine-rich (Q), DNA binding (POU/hom; ▦), and serine-, threonine-, and proline-rich CTD (S/T/P) highlighted. B, Genomic Southern blot (left) of XbaI digested ES cell DNA probed with the fragment previously described, showing two targeted clones with the predicted 3.9-kb fragment corresponding to the mutated allele. A PCR assay (right) using the primers previously described on Cre-expressing transfectants of targeted clone 209B. The smaller product corresponds to the wild-type allele, whereas the larger product corresponds to the targeted allele, which includes a loxP site. C, EMSA on nuclear extracts of peripheral blood cells (left) and splenocytes (center) using an octamer DNA probe and anti-Oct-2 Abs. Retarded bands, indicating octamer-binding proteins (Oct-1, Oct-2, and Oct-2ΔC) in extracts from cells of the indicated genotypes are compared with an extract from a B lymphoma cell line (I29B). The position of excess, unbound probe is indicated. Western blot (right) of protein extract from purified primary splenic B cells of the genotypes shown, showing endogenous expression levels of wild-type and oct-2ΔC alleles, and the actin control. D, Western blot on phoenix cells (57 ) transiently transfected with the pMX-puro vector (58 ), or with corresponding clones encoding wild-type Oct-2 and the truncated Oct-2ΔC protein, probed with anti-Oct-2 and anti-actin. E, Relative luminescence (RLU) of firefly/renilla resulting from the Oct-2 transactivation of an octamer-containing reporter plasmid in phoenix cells. Horizontal triangles represent increasing amounts of transactivator DNA in each assay (see Materials and Methods).

FIGURE 1.

Targeting and screening strategies, and confirmation of Oct-2 protein alteration. A, A map of the Oct-2 locus surrounding the region to be mutated is shown, with exons shown as boxes (top); those encoding the DNA-binding domains are labeled. Arrows indicate the locations and orientations of PCR primers used for screening and brackets show the location of a fragment used as a probe in Southern blot analyses. The XbaI linker containing the nonsense mutation (Xba/STOP) was introduced into the SmaI site in the POU homeodomain. Also shown are the targeting vector with its double selection cassette (see Materials and Methods), and the predicted product of Cre-mediated deletion of this cassette, leaving the targeted allele shown (bottom). The wild-type Oct-2 protein (bottom) and the truncated Oct-2ΔC mutant are depicted, approximately to scale, with the glutamine-rich (Q), DNA binding (POU/hom; ▦), and serine-, threonine-, and proline-rich CTD (S/T/P) highlighted. B, Genomic Southern blot (left) of XbaI digested ES cell DNA probed with the fragment previously described, showing two targeted clones with the predicted 3.9-kb fragment corresponding to the mutated allele. A PCR assay (right) using the primers previously described on Cre-expressing transfectants of targeted clone 209B. The smaller product corresponds to the wild-type allele, whereas the larger product corresponds to the targeted allele, which includes a loxP site. C, EMSA on nuclear extracts of peripheral blood cells (left) and splenocytes (center) using an octamer DNA probe and anti-Oct-2 Abs. Retarded bands, indicating octamer-binding proteins (Oct-1, Oct-2, and Oct-2ΔC) in extracts from cells of the indicated genotypes are compared with an extract from a B lymphoma cell line (I29B). The position of excess, unbound probe is indicated. Western blot (right) of protein extract from purified primary splenic B cells of the genotypes shown, showing endogenous expression levels of wild-type and oct-2ΔC alleles, and the actin control. D, Western blot on phoenix cells (57 ) transiently transfected with the pMX-puro vector (58 ), or with corresponding clones encoding wild-type Oct-2 and the truncated Oct-2ΔC protein, probed with anti-Oct-2 and anti-actin. E, Relative luminescence (RLU) of firefly/renilla resulting from the Oct-2 transactivation of an octamer-containing reporter plasmid in phoenix cells. Horizontal triangles represent increasing amounts of transactivator DNA in each assay (see Materials and Methods).

Close modal

For PCR screening, ∼105 ES cells or 1 μg ES cell DNA in PCR buffer (10 mM Tris HCl, pH 9.0, 2.0 mM MgCl2, 50 mM KCl, 200 μg/ml gelatin, 60 μg/ml proteinase K, 0.45% Nonidet P-40, and 0.45% Tween 20) were incubated for 30 min at 55°C, boiled for 10 min and 2 μl was used in each PCR. PCR assays used the following conditions (primer pairs): lox5′, 5′-AGACCTGTGAGGCACCTGCAGC-3′; lox3′, 5′-AAGCCTTTGTGCACGCTGGACC-3′; 35 cycles at 94°C for 60 s, at 70°C for 90 s. An aliquot of the products was analyzed on an agarose gel (see Fig. 1,B, right panel). The primers are anchored within the oct-2 gene adjacent to the insertion site of the selection cassette. The wild-type allele is detected as a 250 bp band (see Fig. 1,B, (WT)), whereas all Cre-transfected ES cell lines examined contained the rearranged band (see Fig. 1 B, (ΔloxP)). Sequencing of the ΔloxP products confirmed the nature of the junction, i.e., loss of the neomycin-resistant/thymidine kinase cassette and the presence of a single loxP site. For sequencing, the amplified bands were purified using the Qiaquick gel extraction kit (Qiagen, Valencia, CA) and sequenced using the PRISM Ready Reaction DyeDeoxy Terminator Cycle sequencing kit (Applied Biosystems, Foster City, CA). The reactions were analyzed on an Applied Biosystems 373A DNA sequencer.

Rag-1-deficient mice were reconstituted with oct-2+/+, oct-2−/−, or oct-2ΔC/ΔC fetal liver stem cells as described (6). Splenocytes from fully reconstituted mice (possessing abundant B and T lymphocytes in peripheral blood at least 2 mo after reconstitution) were cultured at 106 cells/ml in 200 μl either alone, with LPS (10 μg/ml) or with anti-CD40 (5 μg/ml purified clone FGK45.5; a kind gift from Dr. A. Rolink, Basel Institute for Immunology, Basel, Switzerland) for 72 h in triplicate wells. Cells were then pulsed with [3H]thymidine (1 μCi/well) for 6 h, as previously described (6), to measure proliferative responses to these mitogens.

For Fig. 3, sorted B cells were cultured in 96-well plates at 106 cells/ml in 100 μl as described (5). Cells were either cultured alone, or were stimulated with anti-μ (F(ab′)2 goat anti-mouse IgM (Jackson ImmunoResearch Laboratories, West Grove, PA) at 10 μg/ml. After 3 days, replicate wells were analyzed for cell survival and for cell proliferation. Cell survival was assessed by flow cytometry, by determining the proportion of total cells that excluded the vital dye propidium iodide. Proliferation was measured using the incorporation of 3[H]thymidine, as previously described.

FIGURE 3.

Assessment of B cell maturation by phenotype and function. A, Histogram of HSA staining on B220-positive cells from spleens of reconstituted animals. The vertical line indicates the HSA level found on the majority of mature, splenic B cells in control animals. B, Flow cytometry plots for splenocytes stained for B220 and CD23. The sorting gates enrich for immature (CD23-negative) and more mature (CD23-positive) cells. C, Survival of sorted B cells in vitro, after 72 h of culture in normal medium (▧), or in the presence of anti-μ Abs (▪). D, Proliferation of sorted B cells in response to surface IgM cross-linking. The stimulation index is calculated as the ratio of the incorporation of 3[H]thymidine in stimulated cultures to the incorporation measured in matched, unstimulated cultures. The values shown are the means of triplicate wells ± the SD for individual mice.

FIGURE 3.

Assessment of B cell maturation by phenotype and function. A, Histogram of HSA staining on B220-positive cells from spleens of reconstituted animals. The vertical line indicates the HSA level found on the majority of mature, splenic B cells in control animals. B, Flow cytometry plots for splenocytes stained for B220 and CD23. The sorting gates enrich for immature (CD23-negative) and more mature (CD23-positive) cells. C, Survival of sorted B cells in vitro, after 72 h of culture in normal medium (▧), or in the presence of anti-μ Abs (▪). D, Proliferation of sorted B cells in response to surface IgM cross-linking. The stimulation index is calculated as the ratio of the incorporation of 3[H]thymidine in stimulated cultures to the incorporation measured in matched, unstimulated cultures. The values shown are the means of triplicate wells ± the SD for individual mice.

Close modal

EMSA was performed on nuclear extracts prepared as described (31) from splenocytes or from peripheral blood cells after red cell depletion, using an octamer-containing DNA fragment from the Ig VH17.2.25 promoter as a probe. Retarded bands containing Oct-2 were identified using a mAb raised against the 44 N-terminal amino acids common to all Oct-2 isoforms (L. M. Corcoran, unpublished observations), which caused a supershift of the protein-DNA complex.

Phoenix cells were transfected with the pMX-puro expression vector, or with this vector bearing a full-length oct-2.1 cDNA, or an Oct-2ΔC-encoding construct containing a stop codon in the same position as in the oct-2ΔC targeted allele (see Fig. 1 A). The reporter plasmid was the pGL-3 Enhancer vector (Promega, Madison, WI), into which a DNA fragment containing a single octamer was introduced in the promoter position. The octamer-containing DNA fragment was from the Ig VH17.2.25 promoter, from −59 to +57 with respect to the start of transcription. The artificial promoter driving firefly luciferase expression therefore resembles an endogenous Ig gene promoter. Phoenix cells were transfected with 250, 500, and 1000 ng of each expression plasmid, using the FuGENE 6 reagent and manufacturer’s protocol (Roche, Indianapolis, IN), along with 200 ng of the reporter construct. As a control for transfection efficiency, all cells were simultaneously transfected with 100 ng of p-RL, which encodes renilla luciferase (Promega). Reporter activity was measured using the Dual-Luciferase Reporter Assay System (Promega). Luminescence was measured using a LUMIstar Galaxy luminometer (BMG Labtechnologies, Offenburg, Germany) and relative luciferase activity was calculated by measuring the ratio of firefly to renilla luciferase activities.

Abs to cell surface markers on red cell depleted peripheral blood cells used to distinguish B and T lymphocytes were, respectively, PE-coupled anti-B220 (clone RA3-6B2) and FITC-coupled anti-thy-1.2 (clone 30-H12; both from BD PharMingen, San Diego, CA). Preparation, staining and analysis of splenocytes and peritoneal cells were performed as previously described (5). In all cases, acquisition gates were set to include all nucleated, live (propidium iodide-negative) cells.

For cell sorting experiments, splenocytes were stained for B220 and CD23 (using FITC-coupled anti-CD23, clone B3B4; BD PharMingen). Live, B220+ cells expressing different levels of CD23 were collected using a MoFlo cell sorter (Cytomation, Fort Collins, CO).

Sera from naive mice housed in a clean, conventional animal facility were analyzed for levels of IgM and IgG1 by ELISA, as previously described (6). Mid-log serial dilutions starting from an initial 1/100 dilution of serum were used.

B cells (B220+, IgM+) were sorted from spleens of C57BL/6 control mice, or from mice reconstituted with oct-2−/− or oct-2ΔC/ΔC lymphoid cells. Total RNA was prepared using the RNeasy Mini Kit (Qiagen), and first strand cDNA was synthesized using the SuperScript First-Strand Synthesis System (Invitrogen, San Diego, CA), all according to manufacturers’ instructions. cDNAs corresponding to β-actin and CD36 mRNA were amplified and quantified as described (11).

Splenic B cells were purified using anti-B220 magnetic beads (Miltenyi Biotec, Bergisch Gladbach, Germany). Transfected phoenix cells were harvested at 48 h posttransfection by scraping. In all cases, cell lysates were prepared using Triton X-100 lysis buffer (1% v/v in 20 mM Tris, pH 8.1, 150 mM NaCl containing 1 mM PMSF, 10 mg/ml leupeptin, 10 mM NaF, 10 mM Na2 H2P2O7, 5 mM EDTA, and 1 mM Na3VO4). Proteins were separated on a 10% SDS-polyacrylamide gel and transferred to a nitrocellulose membrane for Western blotting analysis. Oct-2 was detected using the Oct-2-specific mAb previously described and an HRP-labeled anti-rat IgG (DAKO, Glostrup, Denmark). Goat anti-actin antiserum and rabbit anti-goat HRP (both from Santa Cruz Biotechnology, Santa Cruz, CA) were used to assess protein loading. Ab binding was assessed using ECL Western blotting reagent (Amersham Pharmacia Biotech, Piscataway, NJ).

A specific deletion of the C-terminal transactivation domain of Oct-2 was made via homologous recombination in ES cells. The Oct-2ΔC targeting vector (Fig. 1 A) took into account the potential for alternative splicing of the targeted allele, by inserting a translational stop codon at the end of the POU homeodomain encoding exon, which is included in all oct-2 mRNA splice forms described to date (26, 32, 33). The mutant allele is designed to encode Oct-2 protein isoforms with normal N termini, but which are uniformly truncated immediately C-terminal to the DNA-binding domain.

The nonsense linker used contained multiple STOP codons and introduced an XbaI site, which was used for screening. The mutation was generated by conventional gene targeting in ES cells (see Materials and Methods), using a selection cassette (27) containing an internal ribosomal entry site to allow the expression of a bicistronic mRNA containing both the neomycin resistance gene (for positive selection) and the HSV thymidine kinase gene (for negative selection). From a total of 336 neomycin-resistant clones, five independent targeted clones were shown to contain a single integration of the targeting vector at the oct-2 locus (e.g., 209B; Fig. 1,B). The selection cassette was flanked by loxP sites (Fig. 1,A), and when Cre recombinase was introduced transiently into clone 209B in vitro, this cassette was deleted correctly (Fig. 1 B).

Chimeric mice transmitting the targeted allele (which still contained the selection cassette) were crossed with EIIa-Cre transgenic mice (34) on a C57BL/6 background, and their progeny screened for deletion of the selection cassette by the PCR strategy shown in Fig. 1,B. Mice for which the selection cassette had been correctly deleted were intercrossed. Their progeny were tested for inheritance of the EIIa-Cre transgene by PCR (34), and positive mice were not used further. Nuclear extracts of peripheral blood cells from mice heterozygous for the oct-2ΔC allele revealed a structurally altered Oct-2 protein in a gel shift assay. The truncated protein bound the octamer motif, was expressed at a level equivalent to the wild-type allele and was specifically supershifted by an anti-Oct-2 anti-serum (Fig. 1,C, left and center panels). Western blots of B cell extracts (Fig. 1,C, right panel) and of phoenix cells transiently expressing the wild-type and truncated Oct-2 proteins showed that the Oct-2ΔC protein was stable, and it was expressed at an appreciable level. In B cells, the level of the Oct-2ΔC protein was similar to the Oct-2B isoform, and slightly less than the Oct-2A isoform, whereas in phoenix cells, both proteins were expressed at similar levels, arguing that the truncated protein is not inherently less stable than full-length Oct-2. Finally, both proteins were equally capable of activating transcription from an octamer-containing promoter (Fig. 1 E). This conclusion agrees with the results of structure/function analysis of human Oct-2, in which an analogous C-terminal truncation mutant was found to be stable and transcriptionally active in vitro (16). The activity in each case was modest, as the reporter bore only a single octamer motif in the context of a genomic IgH promoter fragment. All of these properties indicate that the oct-2ΔC allele encodes the expected Oct-2ΔC protein, and that the protein is stable and active.

Mice homozygous for a null allele of the oct-2 gene are live born, but die shortly after birth for unknown reasons (7). We reasoned that the truncated Oct-2 protein might rescue the mice, as both the DNA-binding domain and a strong activation domain in the N terminus of the protein, identified biochemically (16, 17), were intact in Oct-2ΔC. Surprisingly however, mice homozygous for the oct-2ΔC allele also failed to survive. No homozygous oct-2ΔC/ΔC mice have been generated in over six generations, despite their presence in a normal Mendelian frequency during late stages of gestation (data not shown). Heterozygous oct-2+/ΔC mice are phenotypically normal and fertile. Therefore the survival role played by the Oct-2 transcription factor requires sequences in the C terminus for its fulfillment. Consequently, as with the oct-2 null mutants (6) all experiments described in this study used adult Rag-1-deficient mice in which immune systems had been reconstituted with fetal liver stem cells from wild-type or oct-2ΔC homozygous littermates, and are designated “control mice” and “oct-2ΔC/ΔC mice” in this study. They have been compared in some cases to mice reconstituted with oct-2 null stem cells, designated in this study as “oct-2−/−” mice for brevity.

The gene modification described used a 129/Sv-derived ES cell line. We have shown that this strain background can significantly influence B cell behavior (35). We therefore backcrossed the mice for at least six generations onto a C57BL/6 background before performing the studies described in this study.

Mice with an Oct-2-deficient immune system show a number of aberrations, all of which reflect defects in peripheral B cell maturation and function. Developmental processes in the bone marrow appear to be unaffected (7).

In the circulation of oct-2−/− mice, the proportion of B cells is significantly reduced compared with controls, with a corresponding increase in the proportion of T cells (Fig. 2 A). oct-2ΔC/ΔC mice also show a significantly lower proportion of B cells in the circulation (the proportions of B and T cells in heterozygous oct-2+/ΔC mice were normal; data not shown). The reduction is due at least in part to a maturation block in both of the oct-2 mutants (see below).

FIGURE 2.

Features of the humoral immune systems of naive, reconstituted mice. A, Proportions of nucleated cells in peripheral blood that represent B (▪) or T (▧) lymphocytes, as determined by flow cytometry using anti-B220 or anti-thy-1 Abs, respectively. Values shown are the mean ± SD for four (oct-2−/− and oct-2ΔC/ΔC) or eight (oct-2+/+) reconstituted mice. B, ELISA on titrations of serum from naive mice. Absorbance values are means of four mice reconstituted with cells of each genotype (▪, +/+; ○, −/−; ▵, ΔC/ΔC). C, B cell populations in a peritoneal wash of mice reconstituted with Oct-2+/+ or Oct-2ΔC/ΔC fetal liver. The stains used reveal conventional (B-2) and B-1 lymphocytes in the control, but not the mutant sample. B-1 cells (insets) have a characteristic surface phenotype (B220-dull and Mac-1-positive, or CD23-negative).

FIGURE 2.

Features of the humoral immune systems of naive, reconstituted mice. A, Proportions of nucleated cells in peripheral blood that represent B (▪) or T (▧) lymphocytes, as determined by flow cytometry using anti-B220 or anti-thy-1 Abs, respectively. Values shown are the mean ± SD for four (oct-2−/− and oct-2ΔC/ΔC) or eight (oct-2+/+) reconstituted mice. B, ELISA on titrations of serum from naive mice. Absorbance values are means of four mice reconstituted with cells of each genotype (▪, +/+; ○, −/−; ▵, ΔC/ΔC). C, B cell populations in a peritoneal wash of mice reconstituted with Oct-2+/+ or Oct-2ΔC/ΔC fetal liver. The stains used reveal conventional (B-2) and B-1 lymphocytes in the control, but not the mutant sample. B-1 cells (insets) have a characteristic surface phenotype (B220-dull and Mac-1-positive, or CD23-negative).

Close modal

Naive oct-2−/− mice have less Ig in the serum compared with control animals. An 8–10-fold reduction is consistently observed for most isotypes, including IgM and IgG (6). oct-2ΔC/ΔC Mice show a similar reduction in IgM and IgG1 titers to oct-2−/− mice (Fig. 2 B). Therefore, Oct-2 mediates optimal Ig production in vivo, and the C-terminal domain (CTD) of the protein is required for this function.

The B-1 lymphocyte population of the peritoneal cavity is maintained through self-renewal via a signal through the B cell receptor (BCR) and associated signaling molecules (36) with support from cytokines such as IL-5 (37). Oct-2 is absolutely required for the development or maintenance of the B-1 cell compartment (5). The data in Fig. 2 C indicate that the Oct-2 C terminus is essential in this capacity, as oct-2ΔC/ΔC mice lack peritoneal B-1 cells.

B cell differentiation in the bone marrow yields surface IgM-positive, naive B cells in which differentiation continues after emigration to the periphery. These immature cells can be distinguished on the basis of surface markers, such as the heat-stable Ag (HSA; Refs. 38 ,39) or a combination of markers (IgM/IgD/CD21/CD23) that distinguish two transitional differentiation states (T1 and T2) through which cells pass before entering the mature, long-lived circulating B cell pool (40). In addition, immature, transitional B cells are labile cells that exhibit a high turnover rate, and are killed rather than activated by a BCR signal (38, 39, 41, 42).

Peripheral Oct-2-deficient B cells exhibit an immature phenotype, expressing the high HSA levels and turnover rates of immature cells (5). oct-2ΔC/ΔC splenic B cells also have high HSA levels (Fig. 3,A), implying immaturity. When the markers used by Loder et al. (40) were used, both oct-2−/− and the oct-2ΔC/ΔC mutants appeared to be fully mature, in having normal numbers of cells with the mature B220+, CD23+ phenotype (Fig. 3,B and data not shown). However, when these cells were assessed for their behavior in response to BCR signaling, they were clearly immature. Immature (B220+, CD23) and mature (B220+, CD23+) cells from spleens of control, oct-2−/− and oct-2ΔC/ΔC mice (Fig. 3,B), were sorted and cultured for 3 days in the presence of cross-linking anti-μ Abs. Immature cells from all three mice behaved similarly; they did not divide, but were killed by anti-μ treatment (Fig. 3, C and D). Mature cells from control animals were refractory to this negative signal, and were activated by surface IgM cross-linking to proliferate. Strikingly, the phenotypically mature oct-2 mutant cells behaved as immature cells; they failed to proliferate in response to BCR cross-linking, and instead were killed by the treatment (Fig. 3, C and D). Therefore, the C terminus of Oct-2 is critical for the peripheral maturation of B cells, and their acquisition of the ability to respond positively to a BCR signal, as oct-2ΔC/ΔC cells behaved exactly as Oct-2-deficient cells in these assays.

In addition to their aberrant response to BCR signaling, Oct-2-deficient B cells are hyporesponsive to LPS in a proliferation assay (6). In normal mice, this response is not affected by the maturity of the cells (38). oct-2ΔC/C cells have the same poor response as Oct-2-deficient B cells, demonstrating that the response to LPS requires the Oct-2 CTD (Fig. 4,A). Whereas Oct-2 is expressed at a significant basal level in B lineage cells, the gene is further induced in normal B cells activated by LPS (43). The up-regulation of Oct-2 in response to LPS fails to occur in oct-2ΔC/ΔC cells (Fig. 4,B), consistent with the inability of these cells to respond to LPS activation. Signaling through CD40 is not compromised by loss or mutation of Oct-2 (Fig. 4 A and in Ref. 6), indicating that the cells are capable of proliferation when given the appropriate signal.

FIGURE 4.

Responses to other mitogens in vitro and Oct-2 target gene expression. A, Each graph shows the response of oct-2+/+ (▪), oct-2−/− (□), or oct-2ΔC/ΔC (▧) splenocytes to different mitogenic stimuli, represented as stimulation index (mean of triplicates ± SD) for individual mice. B, Western blots of cytoplasmic (C) and nuclear (N) extracts from purified wild-type, Oct-2 null, or oct-2ΔC/ΔC B cells, using an anti-Oct-2 mAb that binds to the N terminus of Oct-2. C, Semiquantitative RT-PCR on titrated cDNA templates made from purified, resting B cells, to measure the expression levels of the Oct-2-dependent CD36 gene, relative to a β-actin control.

FIGURE 4.

Responses to other mitogens in vitro and Oct-2 target gene expression. A, Each graph shows the response of oct-2+/+ (▪), oct-2−/− (□), or oct-2ΔC/ΔC (▧) splenocytes to different mitogenic stimuli, represented as stimulation index (mean of triplicates ± SD) for individual mice. B, Western blots of cytoplasmic (C) and nuclear (N) extracts from purified wild-type, Oct-2 null, or oct-2ΔC/ΔC B cells, using an anti-Oct-2 mAb that binds to the N terminus of Oct-2. C, Semiquantitative RT-PCR on titrated cDNA templates made from purified, resting B cells, to measure the expression levels of the Oct-2-dependent CD36 gene, relative to a β-actin control.

Close modal

Of the number of genes proposed as Oct-2 target genes, by virtue of octamer motifs in promoter regions, only a small number has been shown to rely directly on Oct-2 for transcription (9). One of these is the gene for the scavenger receptor molecule CD36 (10). The CD36 gene has an octamer motif in its promoter whose sequence prohibits the recruitment of the cofactor OBF-1 (11). Using RT-PCR analysis of cDNA from purified splenic B cells from control, OBF-1−/−, Oct-2-deficient, and oct-2ΔC/ΔC mice, we found that CD36 expression is indeed Oct-2-dependent and OBF-1 independent, and that the C terminus of Oct-2 is essential for normal CD36 transcription (Fig. 4 C). Collectively, these data show that the Oct-2 C terminus is required for a number of biological properties of B lymphocytes, and imply that this protein domain is critical for the activity of Oct-2 in vivo.

We have demonstrated that mice and B cells expressing a C terminally truncated Oct-2 protein from the endogenous oct-2 locus behave aberrantly, and in all respects mirror the phenotypes of Oct-2-deficient mice and B cells. In agreement with early in vitro structure/function assays (16, 23), we found no evidence that the truncated protein acts in a dominant-negative fashion. In all assays, heterozygous oct-2+/ΔC B cells appeared and behaved like wild-type cells (data not shown).

Based on in vitro studies, the Oct-2 CTD has been ascribed two unique properties in B cells. These are abilities to activate transcription from a distal position and to interact with a B cell-specific activity to mediate tissue-specific expression (23, 24, 44). In the plasmacytoma crossed to T lymphoma hybrids of Sharif et al. (25), the Oct-2 C terminus was critical for the preservation of the plasma cell gene expression program in hybrid cells in vitro. The data presented in this study demonstrate that functions mediated by the CTD of Oct-2 are critically important in vivo, both for B cell development and functional capacity, and for neonatal survival, a process that is not immunologically mediated. Indeed, without this domain, the Oct-2ΔC protein behaves as a null in vivo, in all assessments that we have conducted.

The nature of the B cell-restricted activity cooperating with Oct-2 is unknown, but could be a posttranslational modification, such as phosphorylation, or a physical interaction with a novel coactivator. All currently known Oct-1 and/or Oct-2 coactivators, such as viral transcription regulators or the B cell-restricted coactivator OBF-1, interact via the POU-specific or POU homeodomains (15, 45, 46). However, the data presented in this study imply that protein-protein interactions mediated by the N terminus and DNA-binding POU domains are not sufficient to ensure that Oct-2 functions in vivo in B cells, as both of these domains are intact in the Oct-2ΔC protein (Fig. 1), and yet this protein is largely nonfunctional. Also, as previously mentioned, OBF-1 is not required for transcription of the CD36 gene, an example of a gene-dependent on the Oct-2 CTD. Interestingly, the cofactor HMG I(Y), which cooperates with Oct-2 to activate the human HLA-DRA gene, also interacts with the Oct-2 POU domain, but does require the C-terminal activation domain to mediate its affect on class II expression (47). HMG I(Y) is therefore an example of the sort of coactivator that might enhance tissue-specific transcription mediated by the Oct-2 C terminus.

Alternatively, the B cell-restricted activity may be an enzymatic modifier of Oct-2, such as a kinase. Several investigators have described the phosphorylation of Oct-2, the putative kinases and the possible functional consequences (48, 49, 50). The latter include affects on DNA binding and transcriptional activation, but all result from phosphorylation sites in the POU domain. The Oct-2 C terminus contains a number of putative structural elements that are perfectly conserved in human Oct-2, but whose functional significance has not been established (Fig. 5). There is a predicted phosphorylation site for the cell cycle regulated kinase p34cdc2, and there are overlapping consensus sites for casein kinase I and glycogen synthetase kinase 3 (51, 52). The latter are clustered in a potential amphipathic leucine zipper motif, and so phosphorylation there might influence the interaction of Oct-2 with other proteins. We noted that in the Western blot analysis (Fig. 4), wild-type B cells had at least two bands that reacted with the anti-Oct-2 Ab. These might reflect differentially spliced isoforms of Oct-2, which are known to exist (26, 32). Alternatively, they may represent differentially phosphorylated versions of Oct-2, as have been observed by others (17, 24). LPS activated homozygous oct-2ΔC/ΔC B cells displayed only one band in the Western blot (Fig. 4 B), and the underlying basis for this difference, whether splicing or phosphorylation, is under investigation.

FIGURE 5.

Amino acid sequence of the C terminus shared by the major Oct-2 isoforms. This sequence is that which is lost in the truncated Oct-2ΔC protein. The putative leucine zipper (Prosite prediction tool) is underlined, and consensus sites for predicted protein phosphorylation are boxed. The gray box highlights a consensus p34cdc2 phosphorylation site, and the open boxes show sites that conform to overlapping casein kinase I and glycogen synthetase kinase 3 phosphorylation sites. These sites were identified using the NetPhos 2.0 and PhoshoBase prediction tools. The C-terminal nine amino acids shown to be critical for function in the in vitro assays of Annweiler et al. (24 ) are underscored with a dotted line.

FIGURE 5.

Amino acid sequence of the C terminus shared by the major Oct-2 isoforms. This sequence is that which is lost in the truncated Oct-2ΔC protein. The putative leucine zipper (Prosite prediction tool) is underlined, and consensus sites for predicted protein phosphorylation are boxed. The gray box highlights a consensus p34cdc2 phosphorylation site, and the open boxes show sites that conform to overlapping casein kinase I and glycogen synthetase kinase 3 phosphorylation sites. These sites were identified using the NetPhos 2.0 and PhoshoBase prediction tools. The C-terminal nine amino acids shown to be critical for function in the in vitro assays of Annweiler et al. (24 ) are underscored with a dotted line.

Close modal

Many in vitro studies have contributed to the notion that the C-terminal transactivation domain of Oct-2 has special properties, particularly when compared with the N-terminal, glutamine-rich transactivation domain of Oct-2 or to any domain of the closely related Oct-1. Indeed, in such a comparison, Mead et al. (53) showed that only the Oct-2 CTD can activate transcription in yeast. In an earlier study, Seipel et al. (54) attempted to classify transcriptional activation domains functionally, and found that those of the glutamine-rich class acted as “proximal” activation domains, whereas serine-, threonine-, or proline-rich domains could act from both proximal and distal binding enhancer sites. Oct-2 has two activation domains, one from each of these functional classes. Therefore, Oct-2 has the potential to positively influence expression of target by separate mechanisms, in a variety of contexts. Indeed, in the model recently proposed by Bertolino and Singh (22), the possession of both a POU domain, which directly facilitates recruitment of TATA-binding protein to a promoter (55), and transactivation domains that can act from a both proximal and remote positions confers on Oct-2 the potential to act as a self-sufficient factor, simultaneously directing the promoter/enhancer cooperatively that underlies strong transcriptional activation. Oct-1 lacks the capacity to mediate this cooperativity, most likely because it lacks a domain analogous to the Oct-2 CTD (22). Finally, the functional versatility of Oct-2 is enhanced through differential splicing of the oct-2 gene, which is encoded by a relatively large number of small exons (14 exons; Refs. 26 ,32). Inclusion of particular small exons in a given oct-2 mRNA has been shown to either augment or diminish the potency of the resulting protein in transactivation assays (23, 24). In B cells, the multiple isoforms of Oct-2 might be regulated in their expression to enable activation from a distance in some circumstances (for example, when the IgH 3′ or intronic enhancers become active late in B cell differentiation). In the CNS, neuronal splicing patterns convert Oct-2 from an activator into a transcriptional repressor (56). We have shown in this study that the differential inclusion or exclusion of the CTD in the Oct-2 protein would have significant consequences for cell development and function, and for the survival of the animal.

We thank Drs. Andreas Strasser and Antonius Rolink for mAbs, Drs. Rudolph Jaenisch, Janet Rossant, and Klaus Rajewsky for plasmids, and Stephen Nutt for comments on the manuscript.

1

This work was supported by the Cancer Research Institute, the National Health and Research Council of Australia, and The Arthur and Mary Osborn Estate.

3

Abbreviations used in this paper: RAG, recombination-activating gene; ES, embryonic stem; CTD, C-terminal domain; BCR, B cell receptor; HSA, heat-stable Ag.

1
Staudt, L. M., R. G. Clerc, H. Singh, J. H. LeBowitz, P. A. Sharp, D. Baltimore.
1988
. Cloning of a lymphoid-specific cDNA encoding a protein binding the regulatory octamer DNA motif.
Science
241
:
577
.
2
Clerc, R. G., L. M. Corcoran, J. H. LeBowitz, D. Baltimore, P. A. Sharp.
1988
. The B-cell-specific Oct-2 protein contains POU box- and homeo box-type domains.
Genes Dev.
2
:
1570
.
3
Kang, S. M., W. Tsang, S. Doll, P. Scherle, H. S. Ko, A. C. Tran, M. J. Lenardo, L. M. Staudt.
1992
. Induction of the POU domain transcription factor Oct-2 during T-cell activation by cognate antigen.
Mol. Cell. Biol.
12
:
3149
.
4
Cockerill, P. N., S. P. Klinken.
1990
. Octamer-binding proteins in diverse hemopoietic cells.
Mol. Cell. Biol.
10
:
1293
.
5
Humbert, P. O., L. M. Corcoran.
1997
. Oct-2 gene disruption eliminates the peritoneal B-1 lymphocyte lineage and attenuates B-2 cell maturation and function.
J. Immunol.
159
:
5273
.
6
Corcoran, L. M., M. Karvelas.
1994
. Oct-2 is required early in T cell-independent B cell activation for G1 progression and for proliferation.
Immunity
1
:
635
.
7
Corcoran, L. M., M. Karvelas, G. J. Nossal, Z. S. Ye, T. Jacks, D. Baltimore.
1993
. Oct-2, although not required for early B-cell development, is critical for later B-cell maturation and for postnatal survival.
Genes Dev.
7
:
570
.
8
Feldhaus, A. L., C. A. Klug, K. L. Arvin, H. Singh.
1993
. Targeted disruption of the Oct-2 locus in a B cell provides genetic evidence for two distinct cell type-specific pathways of octamer element-mediated gene activation.
EMBO J.
12
:
2763
.
9
Pfisterer, P., J. Hess, T. Wirth.
1997
. Identification of target genes of the lymphoid-specific transcription factor Oct2.
Immunobiology
198
:
217
.
10
Konig, H., P. Pfisterer, L. M. Corcoran, T. Wirth.
1995
. Identification of CD36 as the first gene dependent on the B-cell differentiation factor Oct-2.
Genes Dev.
9
:
1598
.
11
Shore, P., W. Dietrich, L. M. Corcoran.
2002
. Oct-2 regulates CD36 gene expression via a consensus octamer, which excludes the co-activator OBF-1.
Nucleic Acids Res.
30
:
1767
.
12
Corcoran, L. M., D. Vremec, M. Febbraio, T. Baldwin, E. Handman.
2002
. Differential regulation of CD36 expression in antigen-presenting cells: Oct-2 dependence in B lymphocytes but not dendritic cells or macrophages.
Int. Immunol.
14
:
1099
.
13
Herr, W., R. A. Sturm, R. G. Clerc, L. M. Corcoran, D. Baltimore, P. A. Sharp, H. A. Ingraham, M. G. Rosenfeld, M. Finney, G. Ruvkun, et al
1988
. The POU domain: a large conserved region in the mammalian pit-1, oct-1, oct-2, and Caenorhabditis elegans unc-86 gene products.
Genes Dev.
2
:
1513
.
14
Strubin, M., J. W. Newell, P. Matthias.
1995
. OBF-1, a novel B cell-specific co-activator that stimulates immunoglobulin promoter activity through association with octamer-binding proteins.
Cell
80
:
497
.
15
Luo, Y., R. G. Roeder.
1995
. Cloning, functional characterization, and mechanism of action of the B-cell-specific transcriptional co-activator OCA-B.
Mol. Cell. Biol.
15
:
4115
.
16
Muller-Immergluck, M. M., W. Schaffner, P. Matthias.
1990
. Transcription factor Oct-2A contains functionally redundant activating domains and works selectively from a promoter but not from a remote enhancer position in non-lymphoid (HeLa) cells.
EMBO J.
9
:
1625
.
17
Tanaka, M., W. Herr.
1990
. Differential transcriptional activation by Oct-1 and Oct-2: interdependent activation domains induce Oct-2 phosphorylation.
Cell
60
:
375
.
18
Sivaraja, M., M. C. Botfield, M. Mueller, A. Jancso, M. A. Weiss.
1994
. Solution structure of a POU-specific homeodomain: 3D-NMR studies of human B-cell transcription factor Oct-2.
Biochemistry
33
:
9845
.
19
Klemm, J. D., M. A. Rould, R. Aurora, W. Herr, C. O. Pabo.
1994
. Crystal structure of the Oct-1 POU domain bound to an octamer site: DNA recognition with tethered DNA-binding modules.
Cell
77
:
21
.
20
Inamoto, S., N. Segil, Z. Q. Pan, M. Kimura, R. G. Roeder.
1997
. The cyclin-dependent kinase-activating kinase (CAK) assembly factor, MAT1, targets and enhances CAK activity on the POU domains of octamer transcription factors.
J. Biol. Chem.
272
:
29852
.
21
Arnosti, D. N., A. Merino, D. Reinberg, W. Schaffner.
1993
. Oct-2 facilitates functional preinitiation complex assembly and is continuously required at the promoter for multiple rounds of transcription.
EMBO J.
12
:
157
.
22
Bertolino, E., H. Singh.
2002
. POU/TBP cooperativity: a mechanism for enhancer action from a distance.
Mol Cell
10
:
397
.
23
Friedl, E. M., P. Matthias.
1995
. Transcriptional activation and repression, two properties of the lymphoid-specific transcription factor Oct-2a.
Eur. J. Biochem.
234
:
308
.
24
Annweiler, A., S. Zwilling, T. Wirth.
1994
. Functional differences between the Oct2 transactivation domains determine the transactivation potential of individual Oct2 isoforms.
Nucleic Acids Res.
22
:
4250
.
25
Sharif, M. N., H. S. Radomska, D. M. Miller, L. A. Eckhardt.
2001
. Unique function for carboxyl-terminal domain of Oct-2 in Ig-secreting cells.
J. Immunol.
167
:
4421
.
26
Wirth, T., A. Priess, A. Annweiler, S. Zwilling, B. Oeler.
1991
. Multiple Oct2 isoforms are generated by alternative splicing.
Nucleic Acids Res.
19
:
43
.
27
Wu, H., X. Liu, R. Jaenisch.
1994
. Double replacement: strategy for efficient introduction of subtle mutations into the murine Col1a-1 gene by homologous recombination in embryonic stem cells.
Proc. Natl. Acad. Sci. USA
91
:
2819
.
28
Köntgen, F., C. L. Stewart.
1993
. Simple screening procedure to detect gene targeting events in embryonic stem cells.
Methods Enzymol.
225
:
878
.
29
Gu, H., Y. R. Zou, K. Rajewsky.
1993
. Independent control of immunoglobulin switch recombination at individual switch regions evidenced through Cre-loxP-mediated gene targeting.
Cell
73
:
1155
.
30
Laird, P. W., A. Zijderveld, K. Linders, M. A. Rudnicki, R. Jaenisch, A. Berns.
1991
. Simplified mammalian DNA isolation procedure.
Nucleic Acids Res.
19
:
4293
.
31
Schreiber, E., P. Matthias, M. M. Muller, W. Schaffner.
1989
. Rapid detection of octamer binding proteins with ‘mini-extracts’, prepared from a small number of cells.
Nucleic Acids Res.
17
:
6419
.
32
Hatzopoulos, A. K., A. S. Stoykova, J. R. Erselius, M. Goulding, T. Neuman, P. Gruss.
1990
. Structure and expression of the mouse Oct2a and Oct2b, two differentially spliced products of the same gene.
Development
109
:
349
.
33
Stoykova, A. S., S. Sterrer, J. R. Erselius, A. K. Hatzopoulos, P. Gruss.
1992
. Mini-Oct and Oct-2c: two novel, functionally diverse murine Oct-2 gene products are differentially expressed in the CNS.
Neuron
8
:
541
.
34
Lakso, M., J. G. Pichel, J. R. Gorman, B. Sauer, Y. Okamoto, E. Lee, F. W. Alt, H. Westphal.
1996
. Efficient in vivo manipulation of mouse genomic sequences at the zygote stage.
Proc. Natl. Acad. Sci. USA
93
:
5860
.
35
Corcoran, L. M., D. Metcalf.
1999
. IL-5 and Rp105 signaling defects in B cells from commonly used 129 mouse substrains.
J. Immunol.
163
:
5836
.
36
Hardy, R. R., Y. S. Li, D. Allman, M. Asano, M. Gui, K. Hayakawa.
2000
. B-cell commitment, development and selection.
Immunol. Rev.
175
:
23
.
37
Kopf, M., F. Brombacher, P. D. Hodgkin, A. J. Ramsay, E. A. Milbourne, W. J. Dai, K. S. Ovington, C. A. Behm, G. Kohler, I. G. Young, K. I. Matthaei.
1996
. IL-5-deficient mice have a developmental defect in CD5+ B-1 cells and lack eosinophilia but have normal antibody and cytotoxic T cell responses.
Immunity
4
:
15
.
38
Allman, D. M., S. E. Ferguson, M. P. Cancro.
1992
. Peripheral B cell maturation. I. Immature peripheral B cells in adults are heat-stable antigen(hi) and exhibit unique signaling characteristics.
J. Immunol.
149
:
2533
.
39
Allman, D. M., S. E. Ferguson, V. M. Lentz, M. P. Cancro.
1993
. Peripheral B cell maturation. II. Heat-stable antigen(hi) splenic B cells are an immature developmental intermediate in the production of long-lived marrow-derived B cells.
J. Immunol.
151
:
4431
.
40
Loder, F., B. Mutschler, R. J. Ray, C. J. Paige, P. Sideras, R. Torres, M. C. Lamers, R. Carsetti.
1999
. B cell development in the spleen takes place in discrete steps and is determined by the quality of B cell receptor-derived signals.
J. Exp. Med.
190
:
75
.
41
Carsetti, R., G. Kohler, M. C. Lamers.
1995
. Transitional B cells are the target of negative selection in the B cell compartment.
J. Exp. Med.
181
:
2129
.
42
Norvell, A., L. Mandik, J. G. Monroe.
1995
. Engagement of the antigen-receptor on immature murine B lymphocytes results in death by apoptosis.
J. Immunol.
154
:
4404
.
43
Bendall, H. H., D. C. Scherer, C. R. Edson, D. W. Ballard, E. M. Oltz.
1997
. Transcription factor NF-κB regulates inducible Oct-2 gene expression in precursor B lymphocytes.
J. Biol. Chem.
272
:
28826
.
44
Pfisterer, P., H. Konig, J. Hess, G. Lipowsky, B. Haendler, W. D. Schleuning, T. Wirth.
1996
. CRISP-3, a protein with homology to plant defense proteins, is expressed in mouse B cells under the control of Oct2.
Mol. Cell. Biol.
16
:
6160
.
45
Kristie, T. M., P. A. Sharp.
1990
. Interactions of the Oct-1 POU subdomains with specific DNA sequences and with the HSV α-trans-activator protein.
Genes Dev.
4
:
2383
.
46
Stern, S., W. Herr.
1991
. The herpes simplex virus trans-activator VP16 recognizes the Oct-1 homeo domain: evidence for a homeo domain recognition subdomain.
Genes Dev.
5
:
2555
.
47
Abdulkadir, S. A., V. Casolaro, A. K. Tai, D. Thanos, S. J. Ono.
1998
. High mobility group I/Y protein functions as a specific cofactor for Oct-2A: mapping of interaction domains.
J. Leukocyte Biol.
64
:
681
.
48
Tang, H., P. A. Sharp.
1999
. Transcriptional regulation of the murine 3′ IgH enhancer by OCT-2.
Immunity
11
:
517
.
49
Grenfell, S. J., D. S. Latchman, N. S. Thomas.
1996
. Oct-1 [corrected] and Oct-2 DNA-binding site specificity is regulated in vitro by different kinases.
Biochem. J.
315
:(Pt. 3):
889
.
50
Pevzner, V., R. Kraft, S. Kostka, M. Lipp.
2000
. Phosphorylation of Oct-2 at sites located in the POU domain induces differential down-regulation of Oct-2 DNA-binding ability.
Biochem. J.
347
:(Pt. 1):
29
.
51
Kennelly, P. J., E. G. Krebs.
1991
. Consensus sequences as substrate specificity determinants for protein kinases and protein phosphatases.
J. Biol. Chem.
266
:
15555
.
52
Pinna, L. A., M. Ruzzene.
1996
. How do protein kinases recognize their substrates?.
Biochim. Biophys. Acta
1314
:
191
.
53
Mead, J., K. Elliston, D. F. Mark, S. W. Ludmerer.
1994
. The C-terminus of the B cell activator Oct-2 functions as an activation domain in yeast.
Biochem. Biophys. Res. Commun.
205
:
1998
.
54
Seipel, K., O. Georgiev, W. Schaffner.
1992
. Different activation domains stimulate transcription from remote (’enhancer’) and proximal (’promoter’) positions.
EMBO J.
11
:
4961
.
55
Zwilling, S., A. Annweiler, T. Wirth.
1994
. The POU domains of the Oct1 and Oct2 transcription factors mediate specific interaction with TBP.
Nucleic Acids Res.
22
:
1655
.
56
Dent, C. L., K. A. Lillycrop, J. K. Estridge, N. S. Thomas, D. S. Latchman.
1991
. The B-cell and neuronal forms of the octamer-binding protein Oct-2 differ in DNA-binding specificity and functional activity.
Mol. Cell. Biol.
11
:
3925
.
57
Grignani, F., T. Kinsella, A. Mencarelli, M. Valtieri, D. Riganelli, F. Grignani, L. Lanfrancone, C. Peschle, G. P. Nolan, P. G. Pelicci.
1998
. High-efficiency gene transfer and selection of human hematopoietic progenitor cells with a hybrid EBV/retroviral vector expressing the green fluorescence protein.
Cancer Res.
58
:
14
.
58
Krebs, D. L., Y. Yang, M. Dang, J. Haussmann, M. R. Gold.
1999
. Rapid and efficient retrovirus-mediated gene transfer into B cell lines.
Methods Cell Sci.
21
:
57
.