IgG1 induction factor.

The IgG1 induction factor elevates the IgG1 response and reduces the IgG2b and IgG3 response in LPS-stimulated spleen cells. The factor is a lymphokine produced by T cells. The precursor frequency for cells secreting the IgG1 induction factor is at least tenfold lower as compared to those secreting interleukin 2. Some biochemical properties of the lymphokine are listed. The effects of gamma interferon in B cell-stimulated cultures are shown. Isolation of a cDNA clone coding for the IgG1 induction factor has been achieved and results of these studies are reviewed. Evidence is given that this lymphokine is the same as B cell stimulating factor 1, and we propose that it be renamed interleukin 4. Finally, the possible mechanisms of interleukin 4 are discussed.

IgG1 induction factor: a single molecular entity with multiple biological functions.

A cDNA clone coding for the murine IgG1 induction factor has been isolated. The translation products directed by this clone were analyzed in different biological assays. The data obtained show that the IgG1 induction factor: Is involved in the regulation of IgG responses, by increasing IgG1 and decreasing IgG3 and IgG2b secretion; Induces hyper-Ia expression on resting B lymphocytes; Synergizes with anti-Ig in inducing DNA synthesis in resting B lymphocytes; Synergizes with DxS in inducing DNA synthesis by B lymphocytes; It induces DNA synthesis by either the T cell line CTL-L or Con-A blasts. Thus, this lymphokine in addition to IgG1 inducing activity has also BSF-1, BCGF-II and TCGF like activities. The fact that a single molecule can perform all the above listed functions has implications for our view of lymphocyte activation. It indicates that considering the B cell response as an ordered series of independently controlled events, is an oversimplified view of the dynamic process through which B cells are activated and also indicate the functional interconnection of the different elements of the immune system.

The polyclonal lipopolysaccharide response is accessory-cell-dependent.

Different concentrations of spleen cells from C57BL.10 mice were activated with lipopolysaccharide (LPS), and DNA synthesis and IgM and IgG secretion were measured. The dilution curve was sigmoid, and the response was rapidly lost below a certain cell concentration. In the presence of thymocytes or spleen cells from the LPS-non-responder strain C57BL.10/ScCr the dilution curve for the LPS response of cells from C57BL.10 mice became linear, and the overall response was increased. Irradiated cells could not restore the response at suboptimal cell concentrations. In addition, enriched T cells restored the response as well as normal spleen cells, whereas enriched B cells did not. We compared the restoring capacity of different filler cells for the LPS response with that of a plasmacytoma cell line cultured at suboptimal cell concentrations. The results are compatible with the idea that the filler cells provide growth-stimulating activity to LPS-responsive cells but growth-supporting activity to the tumour cells. Furthermore, highly enriched B-cell populations respond poorly to LPS, but the response can be partly restored by filler cells. These data suggest that the LPS response is accessory-cell-dependent.

Regulation of adhesion and motility in B lymphocytes.

During the differentiation process of B lymphocytes, they go through changes in adhesion and motility. In order to investigate the molecular mechanism of such changes, in vitro culture systems are necessary. When B cells are activated by various stimuli, they form different types of homotypic aggregates. In addition, they might also spread and express microvilli and/or become polarized, the latter being a sign of motility. In this review, we summarize our own research in this area. We give evidence for involvement of different adhesion and signalling molecules, and by the end, we speculate on the in vivo significance of our findings.

Interleukin 4 induces cellular adhesion among B lymphocytes.

We here report that interleukin 4 (IL-4) alone is able to induce cellular adhesion among mouse lymphocytes, and together with lipopolysaccharide (LPS), it increases the adhesion induced by LPS. The adhesion was inhibited by antibodies against IL-4. IL-4 appears to be acting mainly on B lymphocytes, since the response caused by IL-4 alone was much less sensitive to depletion of adherent cells than the LPS response. Depletion of T cells had no effect on IL-4- or LPS-induced adhesion. IL-4 could together with Con A, but not alone, induce adhesion among T cells. Cell clusters, which were formed after 2-3 days of LPS plus IL-4 stimulation, could be completely dissociated, and when the cells were recultured in medium, they readily started to reaggregate. The adhesion molecule lymphocyte function-associated antigen 1 (LFA-1) is, at least in part, involved in LPS plus IL-4-induced adhesion. Antibodies against LFA-1 inhibited the adhesion, but antibodies against other cell surface molecules were without inhibitory effect. Adhesion induced by IL-4 alone may involve other adhesion molecules than LFA-1.

Assessment of the role of leucocyte function-associated antigen-1 in homotypic adhesion of activated B lymphocytes.

In this study we investigated how T-cell-dependent stimuli, via interleukin-4 (IL-4) or CD40 ligation, influence homotypic B-cell adhesion when compared with induction by the T-cell-independent stimulus lipopolysaccharide (LPS). Using primary murine B cells, we found that T-cell-dependent stimulation led to increased aggregation as compared to that induced by LPS. The adhesion was to a large extent dependent on the adhesion molecule, lymphocyte function-associated antigen-1 (LFA-1). We found that activation of B cells with the mitogenic stimuli induced an increased avidity of LFA-1 for its ligand, intercellular adhesion molecule-1 (ICAM-1). The increase was stable and different from that induced by phorbol esters. Although adhesion was reduced using B cells from LFA-1(-/-) mice, aggregation occurred in response to T-cell-dependent stimuli. Our data suggest that adhesion of B lymphocytes is regulated in different modes. One is induced by antigen and leads to a transient conformational change of the LFA-1 molecule. Another is induced by mitogenic stimuli and leads to stable avidity increase of LFA-1, possibly via activation of cytoskeletal anchorage. A third is LFA-1 independent, of low avidity and is induced by T-cell-dependent stimuli.

Activation of BCL1 cells by polyclonal activators and inhibition of growth and IgM secretion by anti-IgM.

Bacterial lipoprotein (LP) and lipopolysaccharide (LPS) both activated an in vitro line of the B-cell tumour BCL1 to IgM secretion, as determined by the protein A plaque assay. LPS but not LP activation was inhibited by polymyxin B. Activation with both LPS and LP resulted in a less than additive response. Several clones of BCL1 were tested, and all responded to both LPS and LP. Both LPS and LP induced broad dose-response curves in normal lymphocytes, recently cloned BCL1 cells, and cloned and synchronized (G1 phase) BCL1 cells. This suggests that the dose-response curve cannot be due to accumulation of responding cells with different threshold sensitivities for activation. We cannot exclude the possibility that the broad dose-response curve is due to a heterogeneity of the LPS or LP preparation. The results indicate that LPS and LP induce similar signals in BCL1 cells. Furthermore, binding to the cell membrane and activation of BCL1 cells by LPS or LP seem to be separate events. An anti-IgM antiserum inhibited spontaneous DNA synthesis and spontaneous and LP-induced IgM secretion of BCL1 cells. Equal inhibition was observed with F(ab')2 fragments but not with Fab fragments of the antiserum, suggesting that cross-linking of IgM bound to the cell surface membrane-induced inhibition. Supernatants from concanavalin A (Con A)-activated spleen cells induced BCL1 cells to secrete IgM. Fab anti-IgM added alone to BCL1 cells did not induce IgM secretion. Furthermore, Fab anti-IgM plus Con A supernatant did not induce a higher response than the supernatant alone. This suggests that inductive signals via the IgM receptor do not occur in BCL1 cells.

HS1,2 enhancer regulation of germline epsilon and gamma2b promoters in murine B lymphocytes: evidence for specific promoter-enhancer interactions.

During an immune response, activated B cells develop into high rate Ig-secreting plasma cells. They also switch from production of IgM to IgG, IgA, or IgE. This process requires a DNA recombination event, which is regulated at the transcriptional level by the production of isotype-specific, sterile germline (GL) transcripts. Induction of these transcripts is controlled by GL promoters and, possibly, by IgH 3' enhancers. We investigated the interaction of the GL epsilon and gamma2b promoters with the HS1,2 enhancer using transiently transfected mouse primary B cells and cell lines. The constructs used for the transfections contained a GL promoter upstream and HS1,2 downstream of a luciferase reporter gene. Both GL epsilon and gamma2b promoters synergized strongly with the HS1,2 enhancer in activated primary B cells, a mature B cell line, and a plasma cell line. We show that the major activity of HS1,2 in activated primary B cells occurs within a 310-bp fragment that includes NF-kappaB, OCT, and NF of activated B cells (Ets/AP-1) sites. By mutating the consensus sequences for various transcription factors, we have determined which sites in HS1,2 are important for synergy with the GL epsilon and gamma2b promoters. Our findings indicate that different sites in HS1,2 might selectively interact with the GL epsilon and gamma2b promoters. We also provide evidence that B cell-specific activator protein is not an absolute suppressor of HS1,2 activity.

Cholera toxin acts synergistically with IL-4 to promote IgG1 switch differentiation.

Previously, we reported that cholera toxin (CT) causes LPS-stimulated membrane (m)IgM+ B cells to undergo increased switch differentiation to IgG- and IgA-producing B cells. In this study we determined whether this effect is specific for one or several of the IgG subclasses and whether B cells exposed to CT respond differently to IL-4, a lymphokine with switching capabilities. In initial studies we found that in LPS-stimulated, mIgM+ B cell cultures, CT eightfold enhanced the formation of IgG1-producing B cells, whereas it only weakly enhanced, one- to twofold, the formation of IgG3-producing B cells. In addition, CT synergistically enhanced the induction of IgG1-producing B cells by IL-4, even at plateau concentrations of IL-4. In contrast, IgM and IgG3 responses were suppressed in the CT plus IL-4-containing cultures as compared to those containing only LPS or LPS and CT. Furthermore, CT plus IL-4 had no enhancing effect on the formation of cells producing IgA; on the contrary, the presence of IL-4 led to a reversal of the stimulatory effect of CT on the IgA response. In further studies, we found that CT affected B cell differentiation at the gene level, before final gene recombination has occurred. Thus, CT together with LPS induced faint but detectable germline gamma 1 RNA transcripts not seen with cells cultured in LPS alone. However, more strikingly, CT enhanced by several-fold expression of germline gamma 1 RNA transcripts in LPS-stimulated B cell cultures containing optimal IgG1-inducing concentrations of IL-4. In addition, despite its weakly positive effect on IgG3 production. CT inhibited expression of germline gamma 3 RNA transcripts in cultures containing LPS and caused a further decrease in such transcripts in cultures containing LPS and IL-4. Finally, we found that CT enhanced the in vivo IgG1 but not the IgG3 or IgM anti-DNP serum antibody response of mice immunized with DNP-LPS. Taken together, these studies suggest that CT more strongly promotes B cell differentiation to IgG1 than to any other IgG subclass in LPS-stimulated cultures. CT acts alone or in synergy with IL-4, early in B cell differentiation to promote IgG1 expression in LPS-stimulated B cell cultures, probably by inducing early steps in the switch to this isotype such as the production of germline gamma 1 RNA transcripts.

Induction of germ-line immunoglobulin heavy chain transcripts by mitogens and interleukins prior to switch recombination.

It has recently been postulated that immunoglobulin class switching is preceded by transcription from unrearranged heavy chain genes. In this report, we have investigated the conditions under which RNA transcribed from unrearranged C gamma 3, C gamma 1, C gamma 2b, C gamma 2a, C epsilon and C alpha genes are induced in normal spleen cells by mitogens and/or interleukin (IL) 4, IL 5 and interferon-gamma. Lipopolysaccharide (LPS) plus IL 4 induced germ-line gamma 1 and epsilon transcripts. LPS induced gamma 2b and gamma 3 transcripts and high doses of IL 4 suppressed these LPS-induced transcripts. Interferon-gamma induced low levels of germ-line gamma 2a transcripts and profoundly suppressed the gamma 1 and epsilon transcripts induced by LPS and IL 4. IL 5 alone or in combination with IL 4 and/or LPS did not induce germ-line alpha transcripts. Spleen cells of the partially immunodeficient mice CBA/N and C3H/HeJ, which do not express IgG3 could be induced, however, by polyclonal activators to express germ-line gamma 3 and gamma 2b transcripts. The data indicate that the capacity of a ligand to induce/suppress transcription of a particular unrearranged heavy chain gene is a good indicator of its capacity to induce switching to the corresponding Ig isotype. However, it is also clear that control of switching can be carried out at other levels.

STAT6 is required for the regulation of IL-4-induced cytoskeletal events in B cells.

During lymphocyte activation, changes in cell morphology are commonly observed. This reflects cell functions important for the regulation of immune responses such as cell adhesion or cell migration. Notably, IL-4 has been shown to induce adhesion and locomotion in B cells, and we have recently described that IL-4 causes dramatic changes in B cell morphology. Thus, such B cells spread with dendritic cell protrusions and produce microvilli-like structures. The molecular mechanisms by which IL-4 induces these complex changes are currently unknown. Two signal transduction pathways are well described for IL-4, i.e. one involving insulin receptor substrate (IRS)-2 and a Janus kinase (JAK)/ signal transducer and activator of transcription (STAT) pathway mediated by STAT6. In this study we therefore used B cells from STAT6-deficient mice to address the question of a possible STAT6 dependence in IL-4-induced morphology changes. By light and electron microscopy, cell spreading and polarization were found to be severely impaired and microvilli formation was reduced. In contrast, only mild impairment was observed in cell adhesion in B cells from STAT6-deficient mice. Our results show that adhesion can be induced in the absence of STAT6. However, expression of STAT6 is necessary for optimal responses in both cell adhesion and microvilli induction. STAT6 is also essential to allow an IL-4-dependent spreading or polarization response. A possible interpretation of our results is that STAT6-dependent expression of a specific gene or genes is required for IL-4 to affect changes in B cell morphology.

CD2: a functional adhesion molecule on murine B cells, involved in interleukin-4-induced aggregation.

The murine equivalent to CD2, previously known as a T cell marker, is expressed on mouse B cells. The monoclonal anti-CD2 antibody 12-15 was found to induce B cell homotypic adhesion. When treated with F(ab')2 fragments of 12-15, purified, resting B cells aggregate within 2 h of incubation and the response is optimal after 20 h. Anti-CD2-induced aggregation is a dose-related active process, dependent on temperature, metabolic energy and divalent cations. Aggregation is inhibited by two different Fab monomers of anti-CD2, implying that it is the CD2 molecule itself that functions as an adhesion molecule. We also report that interleukin 4-induced B cell homotypic adhesion involves CD2-mediated cell binding, since the antibodies specific for mouse, CD2, inhibited interleukin-4-induced cell aggregation.

T and B cell collaboration: induction of motility in small, resting B cells by interleukin 4.

In this report we investigate if IL 4 can work as a chemoattractant factor by inducing locomotion in B cells. We found that murine recombinant IL 4 (rIL 4) induced motile morphology and migration through polycarbonate micropore filters of murine, splenic B cells at an optimal concentration of 3 ng/ml. Kinetic studies revealed optimal migration at 8-16 h, although a significant response could be detected already after 1 h. Flow cytometric studies confirmed that the migrated cells were indeed B cells. We also compared the activity of small, dense B cells and large, low-density B cells, based on Percoll gradient separation. We found no difference in IL 4-induced motility among the two groups. Furthermore, we looked at B cells activated in vitro by preculture in lipopolysaccharide (LPS) or IL 4. Our data indicate that both LPS and IL 4 can increase the general capacity for motility in B cells after preculture for 24 h. T and B cell collaboration requires close cell-cell contacts in order for T cell help to be administered to the B cell. One way of enhancing such cell contacts could be through directional cell migration induced by helper factors (chemotaxis). We suggest that IL 4 can play a role as a chemoattractant factor that enhances cell contacts between T helper cells and B cells.

Characterization of CD40-dependent immunoglobulin class switching.

Previous studies have shown that signalling via CD40 together with cytokines stimulates immunoglobulin (Ig) class switching in B cells. This process includes induction of germline (GL) transcripts and switch recombination. Using an agonistic rat anti-mouse CD40 monoclonal antibody (MoAb), we investigated the role of CD40 signalling in these molecular events. We found that stimulation of murine B cells induced high steady-state levels of germline gamma1, gamma2b and low levels of epsilon transcripts. No detectable gamma2a or alphaRNA were found and the level of gamma3 transcripts was high both in stimulated and unstimulated cells. Although cells treated with anti-CD40 MoAb had high levels of GL gamma1 and gamma3 transcripts, we failed to detect switching to IgG1 or IgG3. However, anti-CD40 MoAb-stimulated cells increased expression of IgG2b. Interestingly, anti-CD40 plus interleukin (IL)-5 induced switching to IgG1. Previous work has demonstrated that CD40 signalling, but not lipopolysaccharide (LPS), induces the gamma1 promoter and that NF-kappaB motifs are important. We show here that both LPS and anti-CD40 activated NF-kappaB proteins binding to the gamma1 promoter. The bound NF-kappaB complexes were different with regard to total concentration and subunit composition. In the light of our data, the mechanism of CD40-mediated Ig class switching is discussed.

Limited life span of extensively proliferating B cells: no evidence for a continuous class or subclass switch.

The polyclonal B cell activators (PBA) dextran sulphate (DxS) and lipopolysaccharide (LPS) induce a marked synergistic response in mouse splenic B cells. Such activated B cell blasts could be maintained in culture for a total period of 15 to 20 days, provided the cells were subcultured repeatedly in fresh medium containing the PBA. After this time period activation could no longer be achieved. The B cell blasts responded at all times better to LPS plus DxS than to either PBA added alone. Furthermore, LPS-induced B cell blasts also responded better to LPS plus DxS than to LPS alone. Activated cells secreted IgM and switched to IgG3 and IgG2b secretion. At the end of the culture period, approximately one cell out of four secreted IgG. The kinetics of IgG3 and IgG2b production were similar, and the fraction of cells secreting these subclasses increased at similar rates throughout the culture period. We conclude that high-rate secreting and proliferating B cells have a limited life span. Furthermore, most activated B cells do not seem to go through a continuous subclass switch but keep on secreting the subclass they originally switched to.

A synergistic polyclonal response to dextran sulphate and lipopolysaccharide: immunoglobulin secretion and cell requirements.

The effects of the simultaneous addition of lipopolysaccharide (LPS) and dextran sulphate (DxS) were studied in a low-cell density culture system, which enables extensive cell proliferation and high immunoglobulin secretion. Using these two mitogens, a synergistic response was observed, with regard to both cell division and IgM secretion. However, only a very low IgG production could be detected. This was caused by the extensive cell proliferation, leading to suboptimal culture conditions. Thus, when the blasts were recultured at a lower density or when a lower initial cell concentration was used, a high IgG response was obtained. The synergistic response induced by LPS plus DxS was independent of T cells. Furthermore, no apparent need for phagocytic cells was found. Both the LPS- and the LPS plus DxS-induced activation led to a switch, preferentially to IgG2b and IgG3 secretion. This subclass pattern was not changed when the cultures were lacking functional T cells.

Frequencies of interleukin-5 mRNA-producing cells in healthy individuals and in immunoglobulin-deficient patients, measured by in situ hybridization.

Interleukin-5 (IL-5) has previously been demonstrated to enhance immunoglobulin synthesis, especially IgA. Thus, it could be hypothesized that a defect production of IL-5 may cause immunoglobulin deficiency. We have analysed the frequency of IL-5 mRNA-producing cells in healthy adults and in patients with common variable immunodeficiency or selective IgA deficiency. Unstimulated lymphocytes were rarely found to synthesize IL-5 as measured by in situ hybridization. However, pokeweed mitogen and several other activating ligands induced the synthesis of IL-5 mRNA in peripheral blood and spleen lymphocyte cultures. After pokeweed mitogen activation, the number of IL-5 mRNA-producing cells most often peaked on day 3 with a maximal frequency of around 1-2% of mononuclear cells. In a kinetic study we were unable to detect any peak frequency differences between healthy controls (mean 0.44%) and 20 patients (mean 0.58%). Thus, although IL-5 has been reported to be an important regulator of IgA synthesis, a defect production does not seem to be the underlying mechanism in human immunoglobulin deficiency.

Partial biochemical characterization of IgG1-inducing factor.

IgG1 induction factor elevates the IgG1 and suppresses the IgG3 and IgG2b responses in lipopolysaccharide-stimulated murine spleen cell cultures. By the use of a quantitative assay, it was found that the three activities, induction of IgG1 and reduction of IgG3 and IgG2b synthesis, were found in the same fractions after different chromatographic procedures, suggesting that the same molecule was responsible for the effects. The factor was precipitated by 60-90% saturation of ammonium sulfate and was sensitive to proteolytic cleavage and to treatment with a buffer of pH 10. It had an apparent molecular mass of 20 kDa as judged by gel filtration chromatography and was separated into two peaks after isoelectric focusing, pI 7.4-7.2 and 6.4-6.2, respectively. Finally it was weakly hydrophobic and negatively charged at pH 7.55. These characteristics indicate that the factor is different from many previously characterized lymphokines and similar or identical to the B cell stimulating factor-1 (BSF-p1). The relevance of these findings to the mechanism of the immunoglobulin class switch is discussed.