The genomic organization of the CD28 gene. Implications for the regulation of CD28 mRNA expression and heterogeneity

CD28 is a 90-kDa homodimeric glycoprotein present on the surface of a large subset of T cells that appears to play an important role in the modulation of T cell activation. Although a number of physiologic effects associated with CD28 stimulation have been defined, relatively less is known about the structure and expression of the CD28 gene itself. We now show that CD28 is expressed in both Th cells and plasma cells as a series of four distinct CD28 mRNA species: 1.3-, 1.5-, 3.5-, and 3.7-kb transcripts. The steady state expression of all four transcripts in CD28+ T cells was stimulated by PMA, suggesting that they might share a common phorbol-sensitive promoter. Consistent with this hypothesis, CD28 was found to be encoded by a single copy gene organized into four exons, each exon defining a functional domain of the predicted protein. All CD28 transcripts appear to initiate within a 61-bp palindrome. Generation of the four CD28 mRNA species from the CD28 gene involves two distinct posttranscriptional events. The longer pair of transcripts (3.5/3.7 kb) is generated by the use of an alternate nonconsensus polyadenylation signal. This results in the addition of 2167 bp beyond the first polyadenylation site utilized by the shorter (1.3/1.5 kb) pair of transcripts. The size difference between the 3.7- and 3.5-kb messages and between the 1.5- and 1.3-kb messages is generated by an internal splicing event that deletes 252 bp within exon 2, which encodes the extracellular domain. This deletion would result in the loss of 84 amino acids, including 4 of 5 extracellular cysteine residues. Although this deletion would result in significant disruption of CD28 secondary structure, it would not be expected to interfere with the ability of the resultant protein to be expressed on the cell surface. These findings suggest that variant isotypes of CD28 may be expressed on the cell surface with potentially different physiologic roles.

The genomic organization of the CD28 gene. Implications for the regulation of CD28 mRNA expression and heterogeneity

CD28 is a 90-kDa homodimeric glycoprotein present on the surface of a large subset of T cells that appears to play an important role in the modulation of T cell activation. Although a number of physiologic effects associated with CD28 stimulation have been defined, relatively less is known about the structure and expression of the CD28 gene itself. We now show that CD28 is expressed in both Th cells and plasma cells as a series of four distinct CD28 mRNA species: 1.3-, 1.5-, 3.5-, and 3.7-kb transcripts. The steady state expression of all four transcripts in CD28+ T cells was stimulated by PMA, suggesting that they might share a common phorbol-sensitive promoter. Consistent with this hypothesis, CD28 was found to be encoded by a single copy gene organized into four exons, each exon defining a functional domain of the predicted protein. All CD28 transcripts appear to initiate within a 61-bp palindrome. Generation of the four CD28 mRNA species from the CD28 gene involves two distinct posttranscriptional events. The longer pair of transcripts (3.5/3.7 kb) is generated by the use of an alternate nonconsensus polyadenylation signal. This results in the addition of 2167 bp beyond the first polyadenylation site utilized by the shorter (1.3/1.5 kb) pair of transcripts. The size difference between the 3.7- and 3.5-kb messages and between the 1.5- and 1.3-kb messages is generated by an internal splicing event that deletes 252 bp within exon 2, which encodes the extracellular domain. This deletion would result in the loss of 84 amino acids, including 4 of 5 extracellular cysteine residues. Although this deletion would result in significant disruption of CD28 secondary structure, it would not be expected to interfere with the ability of the resultant protein to be expressed on the cell surface. These findings suggest that variant isotypes of CD28 may be expressed on the cell surface with potentially different physiologic roles.

Role of the CD28 receptor in T-cell activation

Antigen-specific T-cell activation is initiated through the T-cell receptor. Recent evidence has shown that a number of additional T-cell surface receptors serve to regulate the responses of antigen-activated T cells. One such molecule, CD28, is a member of a heterophilic cell adhesion complex, and is the receptor for the B-cell-restricted B7/BB-1 antigen. As Carl June, Jeffrey Ledbetter, Peter Linsley and Craig Thompson review here, CD28 serves as the surface component of a novel signal transduction pathway that modulates T-cell lymphokine production and increases the resistance of T-cell responses to various immunosuppressive agents.

Reciprocal expression of human ETS1 and ETS2 genes during T-cell activation: regulatory role for the protooncogene ETS1

The expression of the protooncogenes ETS1 and ETS2 has been studied in purified human T cells activated either by cross-linking of the T-cell receptor-CD3 complex on their cell surface or by direct stimulation with phorbol esters and ionomycin. Our results show that resting T cells express high levels of ETS1 mRNA and protein, while expression of ETS2 is undetectable. Upon T-cell activation, ETS2 mRNA and proteins are induced, while ETS1 gene expression decreases to very low levels. Late after stimulation, ETS1 mRNA is reinduced and maintained at a high level, while ETS2 gene expression decreases to undetectable levels. Therefore, it appears that in human T cells, ETS2 gene products are associated with cellular activation and proliferation, while ETS1 gene products are preferentially expressed in a quiescent state.

CD28 ligation in T-cell activation: evidence for two signal transduction pathways

The CD28 homodimer is thought to function as a signal transducing receptor during activation of T cells. Evidence is presented that the degree of aggregation of CD28 on the cell surface regulates two distinct CD28-associated signals. Binding of bivalent CD28 monoclonal antibody (MoAb) 9.3 upregulates lymphokine production by messenger RNA (mRNA) stabilization, without direct initiation of lymphokine mRNA transcription. This signal was not dependent on inositol phospholipid production or activation of a protein tyrosine kinase (PTK). In contrast, further crosslinking of CD28 on the cell surface rapidly induced formation of large amounts of inositol trisphosphate (InsP3) and increased cytoplasmic calcium concentration [( Ca2+]i), but did not stimulate PTK. CD28 crosslinking directly activated a subset of resting T cells, since CD25 (interleukin [IL]-2 receptor alpha chain) mRNA was rapidly induced in purified T cells, and proliferation, even without addition of exogenous IL-2, was sometimes observed. CD25 expression was detected on the cell surface of approximately 20% of CD4+ T cells. The degree of CD28 aggregation required for activation was investigated by preparing soluble 9.3 x 9.3 conjugates ranging in size from approximately 300 Kd to greater than 1,000 Kd, and comparing their function in T-cell proliferation assays with phorbol-12-myristate-13-acetate (PMA), anti-CD3, or IL-2. There was a correlation between conjugate size and proliferation with IL-2, whereas costimulation with PMA or CD3 was optimized at a lower degree of CD28 aggregation. The inositol phospholipid (InsP) generation and increase in [Ca2+]i after CD28 receptor aggregation appeared to proceed through a pathway different from the CD3/T-cell receptor (TCR) pathway since it was enhanced by pretreatment with PMA, while the InsP and [Ca2+]i signal from crosslinking CD3 was suppressed by PMA. Furthermore, the proliferation response to CD28 aggregation was resistant to inhibition by CD3 modulation. Thus, CD28 aggregation appears to trigger a phospholipase C activation pathway that differs from the CD3/TCR-linked pathway.

CD28 is an inducible T cell surface antigen that transduces a proliferative signal in CD3+ mature thymocytes

The rearrangement of TCR genes during thymic ontogeny creates a repertoire of T cell specificities that is refined to ensure the deletion of autoreactive clones and the MHC restriction of T cell responses. Signals delivered via the accessory molecules CD2, CD4, and CD8 have a crucial role in this phase of T cell differentiation. Recently, CD28 has been identified as a signal transducing molecule on the surface of most mature T cells. Perturbation of the CD28 molecule stimulates a novel pathway of T cell activation regulating the production of a variety of lymphokines including IL-2. We have studied the expression and function of CD28 during thymic ontogeny, and in resting and activated PBL. A variable percentage of resting thymocytes were CD28+ (3 to 25%, n = 8), but it was found in high density only on mature CD3+(bright) CD4/CD8 cells. Both unseparated thymocytes and isolated CD3-CD28-/dull cells proliferated when stimulated with PMA plus IL-2 or PMA plus ionomycin. PMA treatment also rapidly up-regulated CD28 expression in the CD3- subset as these cells became CD3-CD28+(bright). Despite the ability of PMA to induce high density CD28 expression in CD3- cells, CD3- thymocytes did not proliferate in response to PMA plus anti-CD28 mAb, in contrast to unseparated cells. CD3+ thymocytes stimulated with immobilized anti-CD3 mAb also failed to proliferate in culture. However, the addition of either IL-2 or anti-CD28 mAb supported proliferation, suggesting that only CD3+ cells could respond to CD28 signaling. The comitogenic effect of anti-CD3 and anti-CD28 mAb was IL-2 dependent as it was abrogated by an anti-IL-2R mAb. Interestingly, the expression of CD28 on the cell surface of CD3+ cells was also inducible, as flow cytometric analysis demonstrated a 10-fold increase in cell surface CD28 by 24 to 48 h after anti-CD3 stimulation of both CD3+ thymocytes and peripheral blood T cells. This increase was accounted for by a commensurate increase in CD28 mRNA levels. Together, these results suggest that CD28 is an inducible T cell antigen in both CD3- and CD3+ cells. In addition, stimulation of the CD28 pathway can provide a second signal to support the growth of CD3+ thymocytes stimulated through the TCR/CD3 complex, and may therefore represent a mechanism for positive selection during thymic ontogeny.

CD28 is an inducible T cell surface antigen that transduces a proliferative signal in CD3+ mature thymocytes

The rearrangement of TCR genes during thymic ontogeny creates a repertoire of T cell specificities that is refined to ensure the deletion of autoreactive clones and the MHC restriction of T cell responses. Signals delivered via the accessory molecules CD2, CD4, and CD8 have a crucial role in this phase of T cell differentiation. Recently, CD28 has been identified as a signal transducing molecule on the surface of most mature T cells. Perturbation of the CD28 molecule stimulates a novel pathway of T cell activation regulating the production of a variety of lymphokines including IL-2. We have studied the expression and function of CD28 during thymic ontogeny, and in resting and activated PBL. A variable percentage of resting thymocytes were CD28+ (3 to 25%, n = 8), but it was found in high density only on mature CD3+(bright) CD4/CD8 cells. Both unseparated thymocytes and isolated CD3-CD28-/dull cells proliferated when stimulated with PMA plus IL-2 or PMA plus ionomycin. PMA treatment also rapidly up-regulated CD28 expression in the CD3- subset as these cells became CD3-CD28+(bright). Despite the ability of PMA to induce high density CD28 expression in CD3- cells, CD3- thymocytes did not proliferate in response to PMA plus anti-CD28 mAb, in contrast to unseparated cells. CD3+ thymocytes stimulated with immobilized anti-CD3 mAb also failed to proliferate in culture. However, the addition of either IL-2 or anti-CD28 mAb supported proliferation, suggesting that only CD3+ cells could respond to CD28 signaling. The comitogenic effect of anti-CD3 and anti-CD28 mAb was IL-2 dependent as it was abrogated by an anti-IL-2R mAb. Interestingly, the expression of CD28 on the cell surface of CD3+ cells was also inducible, as flow cytometric analysis demonstrated a 10-fold increase in cell surface CD28 by 24 to 48 h after anti-CD3 stimulation of both CD3+ thymocytes and peripheral blood T cells. This increase was accounted for by a commensurate increase in CD28 mRNA levels. Together, these results suggest that CD28 is an inducible T cell antigen in both CD3- and CD3+ cells. In addition, stimulation of the CD28 pathway can provide a second signal to support the growth of CD3+ thymocytes stimulated through the TCR/CD3 complex, and may therefore represent a mechanism for positive selection during thymic ontogeny.

Monovalent ligands of complement receptor 2 inhibit whereas polyvalent ligands enhance anti-Ig-induced human B cell intracytoplasmic free calcium concentration

We have performed experiments to investigate the role of ligands for complement receptor 2 (CR2) in human B cell activation. Flow microfluorimetry was used to assess changes in free intracytoplasmic calcium concentration [Ca2+] in indo-loaded B cells, immediately after exposure to anti-mu antibody and to monovalent or polyvalent CR2 ligands. As monovalent ligands we used the C3d fragment and synthetic C3 peptides (peptides P14, residues 1201-1214, and P28, residues 1187-1214). As polyvalent ligands we used i) an intact monoclonal mouse anti-CR2 antibody (HB5) and its F(ab')2 fragment, ii) tetravalent P13 [residues 1202-1214) 4-template), and iii) P28 conjugated to BSA (molar ratio 5/1). Anti-CR2 antibody HB5, tetravalent P13, and P28 conjugated to BSA, enhanced the ability of F(ab')2 fragments of the IgG fraction of goat anti-human mu antibody to increase human B cell [Ca2+]i. In contrast, the monomeric CR2 ligands C3d and P28 inhibited the anti-mu-induced increase in human B cell [Ca2+]i. Multivalent P13, P28, and the HB5, by themselves, did not affect B cell [Ca2+]i. These experiments suggest that the valence of the CR2 ligands is crucial for the nature (synergistic vs antagonistic) of the message transmitted through the CR2.

Increases in tyrosine phosphorylation are detectable before phospholipase C activation after T cell receptor stimulation

Antiphosphotyrosine immunoblots were used to characterize tyrosine phosphorylated proteins after stimulation of the human TCR. Increased tyrosine phosphorylation was evident on at least 12 substrates within 2 min after ligation of the TCR with mAb. Analysis of the time course for increased tyrosine phosphorylation revealed distinct patterns. Increased phosphorylation of 135-kDa and 100-kDa substrates was evident within 5 s, whereas increased phosphorylation of the TCR-zeta-chain required several minutes after treatment with anti-CD3 mAb. This rapid cellular tyrosine phosphorylation occurred independent of the cell cycle, as it occurred after stimulation of resting T cells, T cell blasts, and the Jurkat T cell leukemia line. When the TCR complex was cross-linked together with the CD4 receptor by heteroconjugate anti-CD3/CD4 mAb, an increased magnitude of tyrosine phosphorylation occurred, although no new substrates could be detected. The increased tyrosine phosphorylation of the 135-kDa and 100-kDa substrates was specific in that anti-HLA class I, anti-CD6, anti-CD7, and anti-CD28 antibodies did not cause increased tyrosine phosphorylation. Anti-CD4 stimulation of resting T cells did not cause increased tyrosine phosphorylation of pp100 and pp135, suggesting that the CD4-associated kinase, lck, does not account for the tyrosine phosphorylation observed after TCR stimulation. Similarly, pharmacologic treatment of cells with phorbol ester and calcium ionophore did not cause increased tyrosine phosphorylation of these substrates, indicating that activation of protein kinase C or phospholipase C does not account for these early increases in tyrosine phosphorylation. The time of onset of pp100 phosphorylation, and the magnitude of phosphorylation correlated with the magnitude of calcium mobilization when cells were stimulated with different forms of TCR stimulation. When cells were labeled with [3H]myoinositol and analyzed after stimulation by anti-CD3 mAb, increased tyrosine phosphorylation of the 135-kDa and 100-kDa substrates preceded the activation of phospholipase C, as measured by the appearance of inositol 1,4,5-trisphosphate. This occurred in both T cell blasts and in the Jurkat T cell line. Thus, these findings show that increased tyrosine phosphorylation is the earliest yet detected signal observed after ligation of the TCR complex, and furthermore suggest that tyrosine phosphorylation might link the TCR to the phosphatidylinositolbisphosphate hydrolysis signaling pathway.

Monovalent ligands of complement receptor 2 inhibit whereas polyvalent ligands enhance anti-Ig-induced human B cell intracytoplasmic free calcium concentration

We have performed experiments to investigate the role of ligands for complement receptor 2 (CR2) in human B cell activation. Flow microfluorimetry was used to assess changes in free intracytoplasmic calcium concentration [Ca2+] in indo-loaded B cells, immediately after exposure to anti-mu antibody and to monovalent or polyvalent CR2 ligands. As monovalent ligands we used the C3d fragment and synthetic C3 peptides (peptides P14, residues 1201-1214, and P28, residues 1187-1214). As polyvalent ligands we used i) an intact monoclonal mouse anti-CR2 antibody (HB5) and its F(ab')2 fragment, ii) tetravalent P13 [residues 1202-1214) 4-template), and iii) P28 conjugated to BSA (molar ratio 5/1). Anti-CR2 antibody HB5, tetravalent P13, and P28 conjugated to BSA, enhanced the ability of F(ab')2 fragments of the IgG fraction of goat anti-human mu antibody to increase human B cell [Ca2+]i. In contrast, the monomeric CR2 ligands C3d and P28 inhibited the anti-mu-induced increase in human B cell [Ca2+]i. Multivalent P13, P28, and the HB5, by themselves, did not affect B cell [Ca2+]i. These experiments suggest that the valence of the CR2 ligands is crucial for the nature (synergistic vs antagonistic) of the message transmitted through the CR2.

Increases in tyrosine phosphorylation are detectable before phospholipase C activation after T cell receptor stimulation

Antiphosphotyrosine immunoblots were used to characterize tyrosine phosphorylated proteins after stimulation of the human TCR. Increased tyrosine phosphorylation was evident on at least 12 substrates within 2 min after ligation of the TCR with mAb. Analysis of the time course for increased tyrosine phosphorylation revealed distinct patterns. Increased phosphorylation of 135-kDa and 100-kDa substrates was evident within 5 s, whereas increased phosphorylation of the TCR-zeta-chain required several minutes after treatment with anti-CD3 mAb. This rapid cellular tyrosine phosphorylation occurred independent of the cell cycle, as it occurred after stimulation of resting T cells, T cell blasts, and the Jurkat T cell leukemia line. When the TCR complex was cross-linked together with the CD4 receptor by heteroconjugate anti-CD3/CD4 mAb, an increased magnitude of tyrosine phosphorylation occurred, although no new substrates could be detected. The increased tyrosine phosphorylation of the 135-kDa and 100-kDa substrates was specific in that anti-HLA class I, anti-CD6, anti-CD7, and anti-CD28 antibodies did not cause increased tyrosine phosphorylation. Anti-CD4 stimulation of resting T cells did not cause increased tyrosine phosphorylation of pp100 and pp135, suggesting that the CD4-associated kinase, lck, does not account for the tyrosine phosphorylation observed after TCR stimulation. Similarly, pharmacologic treatment of cells with phorbol ester and calcium ionophore did not cause increased tyrosine phosphorylation of these substrates, indicating that activation of protein kinase C or phospholipase C does not account for these early increases in tyrosine phosphorylation. The time of onset of pp100 phosphorylation, and the magnitude of phosphorylation correlated with the magnitude of calcium mobilization when cells were stimulated with different forms of TCR stimulation. When cells were labeled with [3H]myoinositol and analyzed after stimulation by anti-CD3 mAb, increased tyrosine phosphorylation of the 135-kDa and 100-kDa substrates preceded the activation of phospholipase C, as measured by the appearance of inositol 1,4,5-trisphosphate. This occurred in both T cell blasts and in the Jurkat T cell line. Thus, these findings show that increased tyrosine phosphorylation is the earliest yet detected signal observed after ligation of the TCR complex, and furthermore suggest that tyrosine phosphorylation might link the TCR to the phosphatidylinositolbisphosphate hydrolysis signaling pathway.

Protein kinase C activation in B cells by indolactam inhibits anti-Ig-mediated phosphatidylinositol bisphosphate hydrolysis but not B cell proliferation

In order to examine the role of phosphatidylinositol bisphosphate (PIP2) hydrolysis in B cell activation, we studied the effect of various classes of protein kinase C (PKC) activators on anti-Ig-mediated B cell stimulation. Anti-Ig-stimulated PIP2 hydrolysis, elevations in [Ca2+]i, and induction of DNA synthesis were inhibited by PMA (a phorbol ester) as previously reported. In contrast, indolactam (an alkaloid PKC activator) inhibited PIP2 hydrolysis and elevations in [Ca2+]i, but stimulated rather than inhibited cellular proliferation. In order to examine whether the binding avidity of the PKC activators to PKC played a role in determining their activity to stimulate or inhibit B cell activation, we studied two other PKC activators, bryostatin and mezerein. Again, both inhibited anti-Ig mediated PIP2 hydrolysis and elevations in [Ca2+]i, whereas only the former inhibited induction of DNA synthesis. These data suggest that a) high levels of PIP2 hydrolysis and elevations in [Ca2+]i are not essential for anti-Ig-mediated induction of B cell DNA synthesis and b) activation of PKC may induce both stimulatory and inhibitory pathways of B cell activation, and whether stimulation or inhibition of cell activation is observed may depend on the combined intensity of these two signals.

Protein kinase C activation in B cells by indolactam inhibits anti-Ig-mediated phosphatidylinositol bisphosphate hydrolysis but not B cell proliferation

In order to examine the role of phosphatidylinositol bisphosphate (PIP2) hydrolysis in B cell activation, we studied the effect of various classes of protein kinase C (PKC) activators on anti-Ig-mediated B cell stimulation. Anti-Ig-stimulated PIP2 hydrolysis, elevations in [Ca2+]i, and induction of DNA synthesis were inhibited by PMA (a phorbol ester) as previously reported. In contrast, indolactam (an alkaloid PKC activator) inhibited PIP2 hydrolysis and elevations in [Ca2+]i, but stimulated rather than inhibited cellular proliferation. In order to examine whether the binding avidity of the PKC activators to PKC played a role in determining their activity to stimulate or inhibit B cell activation, we studied two other PKC activators, bryostatin and mezerein. Again, both inhibited anti-Ig mediated PIP2 hydrolysis and elevations in [Ca2+]i, whereas only the former inhibited induction of DNA synthesis. These data suggest that a) high levels of PIP2 hydrolysis and elevations in [Ca2+]i are not essential for anti-Ig-mediated induction of B cell DNA synthesis and b) activation of PKC may induce both stimulatory and inhibitory pathways of B cell activation, and whether stimulation or inhibition of cell activation is observed may depend on the combined intensity of these two signals.

Isolation of a new herpesvirus from human CD4+ T cells

A new human herpesvirus has been isolated from CD4+ T cells purified from peripheral blood mononuclear cells of a healthy individual (RK), following incubation of the cells under conditions promoting T-cell activation. The virus could not be recovered from nonactivated cells. Cultures of lymphocytes infected with the RK virus exhibited a cytopathic effect, and electron microscopic analyses revealed a characteristic herpesvirus structure. RK virus DNA did not hybridize with large probes derived from herpes simplex virus, Epstein-Barr virus, varicella-zoster virus, and human cytomegalovirus. The genetic relatedness of the RK virus to the recently identified T-lymphotropic human herpesvirus 6 (HHV-6) was investigated by restriction enzyme analyses using 21 different enzymes and by blot hybridization analyses using 11 probes derived from two strains of HHV-6 (Z29 and U1102). Whereas the two HHV-6 strains exhibited only limited restriction enzyme polymorphism, cleavage of the RK virus DNA yielded distinct patterns. Of the 11 HHV-6 DNA probes tested, only 6 cross-hybridized with DNA fragments derived from the RK virus. Taken together, the maximal homology amounted to 31 kilobases of the 75 kilobases tested. We conclude that the RK virus is distinct from previously characterized human herpesviruses. We propose to designate it as the prototype of a new herpesvirus, the seventh human herpesvirus identified to date.

Stimulation of CD5 enhances signal transduction by the T cell antigen receptor

After the addition of a CD3 monoclonal antibody to peripheral T cells that have been previously stimulated with phytohemagglutinin, inositol phosphates are produced at a rapid rate for 2 min and at a much slower rate thereafter. Stimulation of CD5 allows CD3-mediated production of inositol phosphates to be sustained at a brisk rate for greater than 20 min and augments the initial CD3-mediated increase in inositol trisphosphate and release of intracellular Ca2+. Thus, perturbation of CD5 by monoclonal antibody enhances the ability of the CD3-antigen receptor complex to couple to the inositol phospholipid pathway. This effect of CD5 is independent of any direct effect of the CD5 monoclonal antibody on the levels of inositol phosphates.

Regulation of interleukin 3 gene induction in normal human T cells

The regulation of IL-3 gene induction in human peripheral blood T cells was studied. IL-3 gene expression was inducible by crosslinking of the T cell receptor/CD3 complex using anti-CD3 MAb G19-4. Anti-CD3-induced IL-3 gene expression was found to be limited to the CD28+ T cell subset and could be augmented by costimulating T lymphocytes with antibodies directed against CD28. IL-3 expression could also be induced by costimulation of T cells with both phorbol ester and ionomycin, which are thought to mimic the intracellular effects of T cell receptor-antigen interaction. However, unlike other lymphokines such as IL-2 or granulocyte-macrophage colony-stimulating factor, IL-3 gene expression is not induced by stimulation of cells with phorbol myristate acetate and anti-CD28. We conclude that IL-3 gene regulation is under stringent control since IL-3 gene expression occurs only in the CD28+ subset of T cells, and since IL-3 induction obligately requires increased intracellular calcium.

Surface immunoglobulin-mediated B-cell activation in the absence of detectable elevations in intracellular ionized calcium: a model for T-cell-independent B-cell activation

We recently showed that anti-immunoglobulin conjugated to high molecular weight dextran is 1000-fold more mitogenic for B cells than unconjugated anti-immunoglobulin. This system serves as a model for T-cell-independent type 2 antigens such as haptenated Ficoll, dextran, and bacterial polysaccharides, which can also stimulate B-cell proliferation and antibody production at low concentrations. We show here that conjugated anti-immunoglobulin, at concentrations that stimulate significant increases in expression of major histocompatibility complex class II molecules and incorporation of thymidine into DNA, does not induce detectable modulation of surface immunoglobulin. These results indicate that the facilitated T-cell-independent B-cell activation by polysaccharide antigens may result from inability to modulate surface immunoglobulin, possibly resulting in persistent and/or repetitive signaling. Early large increases in Ca2+ and breakdown of inositol phospholipids presently thought to be involved in transduction of the mitogenic signal are not detectable at low concentrations of conjugated anti-immunoglobulin, raising the possibility that these biochemical events may not in fact be central to this signaling pathway.

Different patterns of inositol polyphosphate production are seen in B lymphocytes after cross-linking of sIg by anti-Ig antibody or by a multivalent anti-Ig antibody dextran conjugate

Anti-delta antibody conjugated to 2 x 10(6) m.w. dextran (dex) stimulates B lymphocyte proliferation at 10,000-fold lower concentrations than that required by the unconjugated antibody. Dex conjugated antibody also stimulates a greater and more sustained increase in intracellular ionized calcium [( Ca2+]i) than does the unconjugated anti-Ig antibody. Inasmuch as inositol phosphate metabolites have been linked to rises in [Ca2+]i, we analyzed by FPLC the relative amounts of the inositol polyphosphates (IP) in these cells. Anti-Ig-dextran induced a threefold greater increase in total IP than did the unconjugated anti-Ig. Furthermore, in cells stimulated by unconjugated anti-Ig there was a transient induction of I(1,4,5)P3 followed by a rapid accumulation of the I(1,3,4)P3 isomer with little accumulation of I(1,4)P2, whereas in anti-Ig-dex-stimulated cells there was prolonged elevation of I(1,4,5)P3 with more accumulation of I(1,4)P2. In addition, levels of I(1,3,4,5)P4 were maintained over a longer period of time in B cells stimulated by anti-Ig-dex than in those stimulated by unconjugated anti-Ig. The enhanced ratio of I(1,4,5)P3/I(1,3,4)P3 was also seen when suboptimal concentrations of anti-Ig-dex were used which stimulated a level of total inositol phosphate that was similar to that stimulated by the unconjugated anti-Ig. The possibility that the greater stimulation of increased [Ca2+] by anti-Ig-dex than by unconjugated anti-Ig was a predominant factor in influencing the metabolic pathway of I(1,4,5)P3 was excluded. These results show that 1) stimulation of increases in the various IP isomers occurs in anti-Ig stimulated normal B cells as has been shown in B cell lines and 2) that signal transduction and consequent PIP2 hydrolysis that is stimulated by Ag-mediated cross-linking of sIg is strongly influenced by the extent and type of cross-linking that is induced.

Different patterns of inositol polyphosphate production are seen in B lymphocytes after cross-linking of sIg by anti-Ig antibody or by a multivalent anti-Ig antibody dextran conjugate

Anti-delta antibody conjugated to 2 x 10(6) m.w. dextran (dex) stimulates B lymphocyte proliferation at 10,000-fold lower concentrations than that required by the unconjugated antibody. Dex conjugated antibody also stimulates a greater and more sustained increase in intracellular ionized calcium [( Ca2+]i) than does the unconjugated anti-Ig antibody. Inasmuch as inositol phosphate metabolites have been linked to rises in [Ca2+]i, we analyzed by FPLC the relative amounts of the inositol polyphosphates (IP) in these cells. Anti-Ig-dextran induced a threefold greater increase in total IP than did the unconjugated anti-Ig. Furthermore, in cells stimulated by unconjugated anti-Ig there was a transient induction of I(1,4,5)P3 followed by a rapid accumulation of the I(1,3,4)P3 isomer with little accumulation of I(1,4)P2, whereas in anti-Ig-dex-stimulated cells there was prolonged elevation of I(1,4,5)P3 with more accumulation of I(1,4)P2. In addition, levels of I(1,3,4,5)P4 were maintained over a longer period of time in B cells stimulated by anti-Ig-dex than in those stimulated by unconjugated anti-Ig. The enhanced ratio of I(1,4,5)P3/I(1,3,4)P3 was also seen when suboptimal concentrations of anti-Ig-dex were used which stimulated a level of total inositol phosphate that was similar to that stimulated by the unconjugated anti-Ig. The possibility that the greater stimulation of increased [Ca2+] by anti-Ig-dex than by unconjugated anti-Ig was a predominant factor in influencing the metabolic pathway of I(1,4,5)P3 was excluded. These results show that 1) stimulation of increases in the various IP isomers occurs in anti-Ig stimulated normal B cells as has been shown in B cell lines and 2) that signal transduction and consequent PIP2 hydrolysis that is stimulated by Ag-mediated cross-linking of sIg is strongly influenced by the extent and type of cross-linking that is induced.

IgE secretion by Epstein-Barr virus-infected purified human B lymphocytes is stimulated by interleukin 4 and suppressed by interferon gamma

The cytokine interleukin 4 (IL-4) has been shown to induce lipopolysaccharide-activated murine B cells to differentiate into IgE-secreting cells and to stimulate IgE secretion by cultured human peripheral blood lymphoid cells. It is unclear, however, whether this effect of IL-4 on human peripheral blood lymphoid cells is a direct effect on the B cell because IL-4 can stimulate T cells and monocytes as well as B cells and does not induce purified human B cells to secrete immunoglobulin. To investigate this issue we studied the ability of IL-4 to induce IgE secretion by purified human B cells (93-96% CD20+, less than 1% CD3+) that were cultured with Epstein-Barr virus (EBV). Although B cells cultured with IL-4 alone did not secrete Ig and B cells cultured with EBV alone secreted IgM, IgG, and IgA but less than 150 pg of IgE per ml, the combination of EBV and IL-4 induced an IgE response that ranged from 11.4 to 40.3 ng/ml of culture supernatant after 26 days of culture. While IL-4 also enhanced IgM, IgG, and IgA secretion, as well as proliferation by EBV-infected B cells, these effects were less pronounced, occurred earlier during culture, and required a lower concentration of IL-4 than did the stimulation of IgE secretion. Furthermore, interferon gamma at 10 units per ml was found to inhibit IL-4/EBV-induced IgE secretion without inhibiting the other stimulatory effects of IL-4. We conclude that (i) IL-4 and interferon gamma can act directly on polyclonally activated human B cells to respectively stimulate and suppress IgE secretion and (ii) IL-4, in addition to its specific effect on IgE secretion, has a general stimulatory effect on the growth and differentiation of EBV-infected human B cells.

Evidence for the involvement of three distinct signals in the induction of IL-2 gene expression in human T lymphocytes

The regulation of IL-2 gene expression during T cell activation and proliferation has been investigated in primary cultures of purified human peripheral blood T cells. Prior results indicated that stimulation of T cells by anti-CD28 mAb plus PMA could induce IL-2 expression and T cell proliferation that was entirely resistant to cyclosporine. The present studies examined whether CD28 augments IL-2 expression by a unique pathway or merely acts at a point common to CD3-induced proliferation but distal to the effects of cyclosporine. The induction of maximal IL-2 gene expression required three signals provided by phorbol ester, calcium ionophore, and anti-CD28 mAb. Stimulation of cells by optimal amounts of calcium ionophore and PMA induced IL-2 mRNA that was completely suppressed by cyclosporine. The addition of anti-CD28 to T cells stimulated with PMA plus calcium ionophore induced a 5- to 100-fold increase in IL-2 gene expression and secretion that was resistant to cyclosporine. The CD28 signal was able to increase steady state IL-2 mRNA levels even in cells treated with maximally tolerated amounts of calcium ionophore and PMA. The three-signal requirement did not reflect differential regulation of lymphokine gene expression between the CD4 and CD8 T cell subsets or differences in the kinetics of IL-2 mRNA expression. The signal provided by CD28 is distinct from that of CD3 because although anti-CD28 plus PMA-induced proliferation is resistant to cyclosporine, anti-CD3 or anti-CD3 plus PMA-induced IL-2 expression is sensitive. Thus, these studies show that three biochemically distinct signals are required for maximal IL-2 gene expression. Furthermore, these studies suggest that lymphokine production in T cells is not controlled by an "on/off" switch, but rather, that CD28 regulates a distinct intracellular pathway which modulates the level of IL-2 production on a per cell basis. The observation that CD28 stimulation results in IL-2 concentrations that exceed 1000 U/m1 in tissue culture supernatants suggests that a role in vivo for CD28 might be to amplify immune responses initiated by the CD3/T cell receptor complex. Finally, the observation that CD28 interacts with the signals provided by PMA and calcium ionophore shows that the function of CD28 is not merely to act as a scaffold to stabilize or enhance signalling through the CD3/TCR complex.

Evidence for the involvement of three distinct signals in the induction of IL-2 gene expression in human T lymphocytes

The regulation of IL-2 gene expression during T cell activation and proliferation has been investigated in primary cultures of purified human peripheral blood T cells. Prior results indicated that stimulation of T cells by anti-CD28 mAb plus PMA could induce IL-2 expression and T cell proliferation that was entirely resistant to cyclosporine. The present studies examined whether CD28 augments IL-2 expression by a unique pathway or merely acts at a point common to CD3-induced proliferation but distal to the effects of cyclosporine. The induction of maximal IL-2 gene expression required three signals provided by phorbol ester, calcium ionophore, and anti-CD28 mAb. Stimulation of cells by optimal amounts of calcium ionophore and PMA induced IL-2 mRNA that was completely suppressed by cyclosporine. The addition of anti-CD28 to T cells stimulated with PMA plus calcium ionophore induced a 5- to 100-fold increase in IL-2 gene expression and secretion that was resistant to cyclosporine. The CD28 signal was able to increase steady state IL-2 mRNA levels even in cells treated with maximally tolerated amounts of calcium ionophore and PMA. The three-signal requirement did not reflect differential regulation of lymphokine gene expression between the CD4 and CD8 T cell subsets or differences in the kinetics of IL-2 mRNA expression. The signal provided by CD28 is distinct from that of CD3 because although anti-CD28 plus PMA-induced proliferation is resistant to cyclosporine, anti-CD3 or anti-CD3 plus PMA-induced IL-2 expression is sensitive. Thus, these studies show that three biochemically distinct signals are required for maximal IL-2 gene expression. Furthermore, these studies suggest that lymphokine production in T cells is not controlled by an "on/off" switch, but rather, that CD28 regulates a distinct intracellular pathway which modulates the level of IL-2 production on a per cell basis. The observation that CD28 stimulation results in IL-2 concentrations that exceed 1000 U/m1 in tissue culture supernatants suggests that a role in vivo for CD28 might be to amplify immune responses initiated by the CD3/T cell receptor complex. Finally, the observation that CD28 interacts with the signals provided by PMA and calcium ionophore shows that the function of CD28 is not merely to act as a scaffold to stabilize or enhance signalling through the CD3/TCR complex.

Two distinct mechanisms of interleukin-2 gene expression in human T lymphocytes

Interleukin-2 (IL-2) gene regulation was investigated in primary cultures of highly purified human peripheral blood CD28+ T cells. Two discrete mechanisms for induction of T-cell proliferation could be distinguished by examining cell cycle progression and the expression of the IL-2 gene. Stimulation of cells by CD3 MoAb induced only transiently expressed, small amounts of IL-2 mRNA that was completely suppressed by cyclosporine. Costimulation of T cells with CD3 MoAb and either CD28 MoAb or PMA, but not calcium ionophore, induced a 50-100-fold increased in IL-2 gene expression and secretion. High levels of IL-2 gene expression could also be achieved by stimulation with calcium ionophore and PMA or CD28 MoAb and PMA, but not by CD28 MoAb plus calcium ionophore. IL-2 gene expression and T-cell proliferation induced by CD3 MoAb plus PMA or calcium ionophore plus PMA were completely suppressible by cyclosporine. In contrast, IL-2 gene expression and T-cell proliferation induced by CD28 MoAb plus PMA were unaffected by cyclosporine. The CD28 signal was dependent on new protein synthesis. Nuclear run-on transcription assays showed that anti-CD28 did not affect lymphokine transcription. A major effect of CD28 stimulation on mRNA stability was shown by studies using actinomycin D; CD28 stimulation substantially increased the half-life of IL-2 and TNF-alpha mRNA. The effects of anti-CD28 stimulation were specific for growth factors, and thus differ from previously described effects of cycloheximide on mRNA stability. These studies suggest the existence of two biochemical pathways for the induction of IL-2 production, one that occurs at the transcriptional level and is mediated by intracellular calcium release and protein kinase C and is cyclosporine-sensitive, and one that acts post-transcriptionally, is mediated by CD28 stimulation, and is cyclosporine-resistant.