Dependence of BSAP Repressor and Activator Functions on BSAP Concentration

See allHide authors and affiliations

Science  20 Mar 1998:
Vol. 279, Issue 5358, pp. 1961-1964
DOI: 10.1126/science.279.5358.1961


During a B cell immune response, the transcription factor BSAP maintains its activator functions but is relieved of its repressor functions. This selective targeting of BSAP activities was shown to be regulated by a concentration-dependent mechanism whereby activator motifs for BSAP had a 20-fold higher binding affinity than repressor motifs. An exchange of activator and repressor motifs, however, showed that the context of the motif, rather than the affinity, determined whether BSAP operated as an activator or repressor.

BSAP, the transcription factor encoded by the Pax5 gene, is a key regulator of B lymphocyte development (1). Its critical role in early B cell lymphopoiesis has been established by targeted disruption of the mouse Pax5 gene (2). B cells from the mutant mice are blocked at the pro-B stage and do not develop further. BSAP also participates in the later antigen-driven stages of B cell differentiation. Analysis of the changes in gene expression induced by antigen and cytokine signals has shown that BSAP positively regulates the expression of the CD19 component of the B cell antigen coreceptor and germ-line transcription of the ɛ gene (Iɛ), which is a prerequisite for a switch to immunoglobulin E (IgE) synthesis (3, 4). BSAP can also serve as a negative regulator during development. During an immune response, BSAP blocks are removed from the heavy chain 3′ enhancer to allow isotype switching and from the J chain gene promoter to allow synthesis and secretion of pentamer IgM antibody (5-7).

Because the positive functions of BSAP are maintained at the same time that its negative functions are relieved, we investigated how these different regulatory activities are accomplished. BSAP concentrations are high in pre-B, immature B, and mature B cells; once an antigen signal is delivered, however, the concentrations progressively decrease from the presecretor to the mature plasma cell stage, in which BSAP is almost undetectable (7, 8). The decrease in BSAP expression is induced by signals from interleukin-2 (IL-2) or IL-5, which can down-regulate BSAP RNA expression to 20 to 25% of untreated cells (7, 9). Overexpression of BSAP in plasma cell lines increases BSAP repressor activity: J-chain RNA and secreted IgA decrease (7, 10). This expression pattern suggested that BSAP activities might be selectively targeted during antigen-driven differentiation by competition among the BSAP sites for the diminishing amounts of factor (7, 11).

We obtained experimental evidence for a concentration-dependent regulation by determining the relative affinities of BSAP activator and repressor elements from gel-shift assays (Fig.1). BSAP binding to a consensus recognition sequence (12) was competed by adding increasing amounts of the BSAP recognition motifs with established functions such as the CD19, Iɛ, and J chain gene promoters and the 3′ α enhancer (13). The BSAP sites with positive regulatory activity had 20 times the affinity for the factor as those sites with repressor activity (Fig. 1, A and B). A fivefold excess of the CD19 or Iɛ oligonucleotide blocked BSAP binding to the probe by 80 to 90%, whereas a 100-fold molar excess of the J chain or 3′ α-enhancer oligonucleotides was required to achieve the same degree of binding inhibition (Fig. 1C). This difference in affinity indicated that the activator sites could out-compete the repressor sites when the supply of BSAP was limiting.

Figure 1

Affinity of BSAP binding to its recognition sites. BSAP recognition motifs were examined in gel-shift assays for their abilities to compete with the consensus recognition sequence from the sea urchin promoter, H2A2.2, for the endogenous BSAP in extracts from the mature B cell line, K46R. The competitors were added in a 5, 25, and 100 molar excess over probe. (A) Representative competition by activator elements. (B) Representative competition by repressor elements. (C) Quantitation of the competition assays by PhosphorImager measurements of the BSAP-probe complexes formed. Each point represents an average of three independent gel-shift determinations.

We obtained further evidence for a concentration-dependent mechanism by examining the effects of BSAP overexpression on both its repressor and activator functions. For the repressor experiments, representative plasma cell lines were stably transfected with BSAP expression vector or vector alone, and G418-resistant clones were selected for analysis (14). The amounts of BSAP were determined in protein immunoblots of nuclear extracts, and the amount of heavy chain and J chain in protein immunoblots of cytoplasmic extracts. Because light chain synthesis was unaffected by changes in BSAP expression, the light chain content of the cytoplasmic samples was used to normalize the densitometry data for differences in sample loading.

A concentration-dependent regulation predicts that BSAP repressor activities can be restored by increasing BSAP concentrations in plasma cells. This prediction was borne out by protein immunoblots of heavy chain isotypes regulated by the 3′ heavy chain enhancer. Of four transfected clones of a MPC11 line producing a γ2b heavy chain, the two that expressed moderate BSAP concentrations reduced γ2b content by one-half to two-thirds (Fig. 2A). In three clones of an α-producing MOPC315 cell line that expressed moderate amounts of BSAP, the α–heavy chain content was reduced by one-half to one-quarter (Fig. 2B). Similarly, J chain expression was depressed in all clones expressing transfected BSAP, and the extent of suppression roughly correlated with BSAP concentration. In S194 clones, for example, relatively low BSAP concentrations decreased J chain content by 15 to 45%, whereas high amounts decreased J chain content by 85 to 90% (Fig. 2C).

Figure 2

Effect of BSAP overexpression on its repressor functions. Protein immunoblots of the BSAP, heavy, J, and light chain contents in (A) MPC11, (B) MOPC315, and (C) S194 plasmacytomas stably transfected with a BSAP expression vector (Tf1 to Tf4) or with vector alone (Tf0). The amounts of each protein were measured by densitometry, normalized to light chain content, and expressed relative to the value obtained with the vector-transfected control, which was set to 1.0. Endogenous BSAP in the three plasma cell lines was too low to be detected by the conditions used to evaluate the concentrations of transfected BSAP. No heavy chain data are included in the S194 assays because the cells synthesize too little α chain protein for quantitative measurements.

We examined the effect of BSAP overexpression on its activator functions by stably transfecting the BSAP expression vector and vector control into a CH12LX subline that expresses IgM (15). Unlike plasma cells, CH12LX-IgM cells synthesize CD19, express low amounts of J chain but do not secrete significant amounts of pentamer IgM, and have not undergone isotype switching. The phenotype of such CH12 cells is consistent with their BSAP content, which has been reduced by the partial activation to 30% of that present in lines representative of mature, unactivated B cells (7). RNA blots were used to assay the results of the transfections (16) because they allowed the amounts of endogenous and transfected BSAP RNA to be readily distinguished: a 9- to 10-kb endogenous RNA versus a 1.2-kb transcript of the transfected BSAP cDNA (Fig.3).

Figure 3

Effect of BSAP overexpression on its activator functions. RNA blot analyses of endogenous and transfected BSAP, CD19, and J chain transcripts in CH12LX-IgM cells stably transfected with a BSAP expression vector (Tf1 to Tf5) or with vector alone (Tf0). The relative CD19 and J chain RNA values were determined from PhosphorImager measurements normalized to the tubulin RNA content. The density of the BSAP and CD19 blots was measured after a 24-hour exposure, whereas the density of the J chain blots was determined after a 4-day exposure.

A concentration-dependent control mechanism would predict that expression of the target genes under positive BSAP control would increase, or at least remain the same, with increasing BSAP concentrations. Again, the prediction was matched by the experimental data (Fig. 3). CD19 RNA was 1.6 to 3.3 times higher in the BSAP-transfected clones than in the vector control. Thus, maximum expression of CD19 is not required to maintain its coreceptor functions during the presecretor–germinal center stage of a B cell primary immune response (17). J chain RNA was also regulated in a concentration-dependent manner. Increasing BSAP in these cells increased its repressor activities so that even the low expression of the target J chain gene was reduced by 45 to 60% (Fig. 3). The μ heavy chain, however, was not affected, consistent with studies showing that the μ isotype expression is not significantly regulated by the 3′ α enhancer and thus is independent of any BSAP-induced changes (18).

We next asked whether the affinity of a BSAP site determined its activator or repressor action. The CD19 BSAP recognition sequence was substituted for the BSAP sequence (JC) in the J chain gene promoter. Nine of the 18 base pairs (bp) that make up the minimum consensus BSAP site were replaced (Fig. 4A). We evaluated the effect of the replacement by inserting the J2 (nucleotide −168 to −84) test and control fragments from the J chain promoter upstream of a minimum γ-fibrinogen promoter and the chloramphenicol acetyltransferase (CAT) reporter gene (19). These constructs were then assayed for CAT activity after transfection into J chain–expressing S194 cells (20). The BSAP element, which functioned as an activator binding site in the CD19 promoter, was converted to a repressor binding site in the J chain gene promoter (Fig. 4A). The CD19 site–containing J2 fragment induced a 90% decrease in CAT expression compared with an 84% decrease obtained with the counterpart endogenous J chain–containing J2 fragment.

Figure 4

Exchange of an activator motif for a repressor motif. (A) Effect of exchange on the function of the activator motif. Constructs of the J2 promoter fragment (nucleotide −168 to −84) containing either the endogenous JC BSAP repressor element or the CD19 BSAP activator element, and constructs of the J1 fragment (nucleotide −83 to −9) containing the positive-acting JA and PU.1 elements were assayed for CAT activity after transfection into S194 plasmacytoma cells. CAT activity is expressed relative to the value obtained with the γ-fibrinogen promoter alone, which was set to 1.0. All values are the mean of at least three independent determinations. Bars indicate the SEM. (B) Effect of exchange on the affinity of the activator motif. J2 fragments containing the CD19 and JC elements were used as competitors in gel-shift assays as described in Fig. 2.

To determine whether the change in function of the CD19 BSAP site was accompanied by a change in affinity, we used the J chain promoter fragments containing the CD19 and JC binding elements as competitors in the gel-shift assay described in Fig. 1. The patterns of competition obtained with the 85-bp fragments (Fig. 4B) duplicated those observed with the 34-bp CD19 and JC oligonucleotides (Fig. 1), indicating that the affinity of the BSAP site did not change when its context was altered. Despite this switch in affinity, the BSAP site still functioned as a repressor in the J chain promoter, specifying that affinity alone was not sufficient to dictate the repressor or activator function of a BSAP site within an individual promoter. Thus, although BSAP sites with high affinity correlate with positive regulatory function, and low-affinity sites with negative regulatory functions, the context of a site is crucial, even though the occupancy of a site is determined by the gradient of BSAP concentration.

Here, we have defined a concentration-dependent mechanism that allows a single transcription factor to alter its coordinate regulation of gene expression at different stages during B cell differentiation. A similar mechanism operates in the developing embryo of Drosophilawhere concentration gradients of transcription factors, such as the dorsal activator, are established along the dorsoventral axis (21). In this mammalian example, the gradient of BSAP concentration is instituted by signals from IL-2 or IL-5 (7), so that the mechanism allows a single external cytokine to regulate the various BSAP-controlled events in the B cell immune response. In view of these features, concentration-dependent regulation of gene expression may be a generally used mechanism in the developmental systems of vertebrates to control the coordinated activation and repression of multiple genes during development.

  • Deceased.


View Abstract

Navigate This Article