Comparison of the calcium requirement for the induction and maintenance of B cell class II molecule expression and for B cell proliferation stimulated by mitogens and purified growth factors

As an important intracellular second-messenger, the concentration of calcium in the cytosol [Ca+2]i influences diverse cellular activities. To investigate the calcium requirement for distinct phases of B cell activation, we studied the effect of altering the quantity of extracellular calcium on the induction of increased B cell MHC class II molecule (Ia) expression and DNA synthesis by different B cell mitogens. During short-term cultures (less than 24 hr), the induction of class II molecules by anti-immunoglobulin (anti-Ig) and calcium ionophore were dependent on the presence of extracellular calcium, whereas activation induced by B cell stimulation factor-1 (BSF-1) was minimally dependent on extracellular calcium, and that induced by LPS was independent of it. During longer-term cultures (i.e., greater than 24 hr), the heightened class II molecule expression that was induced by all of the B cell mitogens used was significantly compromised by depletion of extracellular free calcium. Although the anti-Ig-stimulated increase in expression of Ia could be restored by the addition of calcium to the medium at 12 hr, it could not be restored when the addition of calcium was delayed to 24 hr after the onset of culture. This was in marked contrast to the finding that BSF-1-stimulated B cell responses which were suppressed after 24 hr of culture in the presence of EGTA could be restored by the addition of calcium. Activation of B cells along the pathway leading to DNA synthesis demonstrated a requirement for extracellular calcium which was greater than that required for induction of MHC class II molecule expression. Thus, LPS-stimulated size increases of B cells after 12 hr of culture was dependent on extracellular calcium while its induction of MHC class II molecule expression was independent of extracellular calcium at this early time point. These observations indicate that the extracellular calcium requirement for B cell activation is dependent both on the activation pathway utilized by the mitogenic signal and on the duration of cell activation. Furthermore, they demonstrate that B cell stimuli that can initiate B cell activation in the relative absence of extracellular calcium may require extracellular calcium for maintenance of this activational state.     

BSF-1 action on resting B cells does not require elevation of inositol phospholipid metabolism or increased [Ca2+]i

B cell stimulatory factor-1 (BSF-1) acts on resting B cells to increase expression of class II major histocompatibility complex (MHC) molecules and to prepare for more prompt entry into S phase in response to anti-IgM and lipopolysaccharide. It also acts as a costimulant, with low concentrations of anti-IgM, to cause resting B cells to synthesize DNA. Unlike anti-IgM, BSF-1 does not cause elevation in inositol phospholipid metabolism or in concentration of intracellular free calcium, nor does it enhance such biochemical responses to anti-IgM. Furthermore, increased expression of class II MHC molecules to BSF-1 is observed when essentially all extracellular calcium is chelated by EGTA, whereas lower concentrations of EGTA completely inhibit increases in class II molecules in response to anti-IgM. These results indicate that BSF-1 effects on resting B cells are not mediated by the inositol phospholipid metabolic pathway.     

Induction of class I and class II MHC antigen expression on murine bone marrow-derived macrophages by IL-4 (B cell stimulatory factor 1)

We have studied the effects of IL-4 (B cell stimulatory factor 1) on the expression of MHC gene products in normal bone marrow-derived macrophages, peritoneal macrophages, and the myelomonocytic cell line WEHI-3. Using both IL-4-containing T cell supernatant and rIL-4, we have observed significant induction of both class I and class II MHC surface expression (about 1.5- to 4-fold increase) in 2-, 3-, and 4-day cultures of bone marrow-derived macrophages. This induction was also apparent at the mRNA level as assessed by Northern blot analysis using A beta, E alpha, and class I probes. Kinetic analysis revealed that induction of class II mRNA by IL-4 was slower than induction by IFN-gamma, requiring 48 h before a significant increase was noted. The magnitude of MHC induction by IL-4 was not as great as that seen with IFN-gamma, which was found to increase surface expression of MHC antigens two- to eightfold. IL-4 also differs from IFN-gamma in the repertoire of macrophages responsive to it. IL-4 was unable to induce class I or class II expression in either thioglycolate-elicited peritoneal macrophages or WEHI-3 cells whereas IFN-gamma induced MHC antigen expression on both cell types under the same conditions. These data demonstrate that IL-4 is capable of inducing both class I and class II MHC gene products in some, but not all, macrophages.     

IL-6-mediated MHC class II induction on RIN-5AH insulinoma cells by IFN-gamma occurs via the G-protein pathway

Major histocompatibility complex (MHC) class II antigen expression has been implicated in the pathogenesis of autoimmune type 1 diabetes. In this study we examined the role of various cytoldnes that may induce MHC class II surface antigen expression, using the rat insulinoma line RIN-5AH as a pertinent model system. As in another study, the ability of IFN-gamma to amplify MHC class II antigen expression 4-fold is demonstrated. At the same time we noted a 5-fold increase of these histocompatibility antigens by IL-6. Signal transduction analysis reveals that IL-6-induced MHC class II expression is specifically mediated by the G-protein system (activation of p21(ras) by IL-6) since mevalonic acid lactone (a Gprotein inhibitor) abolishes the action of IL-6. In contrast, IFN-gamma, which does not activate p21(ras), is not inhibited by protein kinase C (PKC) inhibitors but by those of the G-protein pathway. This finding raises the possibility that IFN-gamma induces RIN cells to secrete IL-6 (as shown previously, as well as in this paper) which, in turn, increases class II antigen expression via the G-protein pathway. This action may be unique to IL-6 or in synergy with IFN-gamma. Other cytokines such as IL-1alpha and beta, and TNF-alpha induce a smaller increase in MHC class II antigens on RIN cells, and appear to activate both the G-protein and the PKC signal transduction pathways to varying degrees. Therefore, injury of pancreatic beta-cells and possible induction of autoimmune type 1 diabetes via various cytokines may be caused by IL-6 or IFN-gamma, or by their ability to induce MHC class II antigen upregulation.     

Leukotriene B4 potentiates the expression and release of Fc epsilon RII/CD23, and proliferation and differentiation of human B lymphocytes induced by IL-4

This study documents the influence of leukotriene (LT) B4 on human B lymphocyte responses. Incubation of freshly isolated B lymphocytes with LTB4, but not LTC4, induced a slight but significant, time- and dose-dependent increase in the surface expression of Fc epsilon RII/CD23 and class II MHC Ag and in the release of soluble CD23. These changes were maximal at 10 nM LTB4 after an incubation period of 48 h. When B lymphocytes were preactivated in vitro with Staphylococcus aureus Cowan strain I (SAC), neither LTB4 nor LTC4 was able to promote proliferation and/or IgG and IgM secretion. In contrast, when resting B lymphocytes were stimulated with a suboptimal concentration (3 U/ml) of IL-4, LTB4, but not LTC4, potentiated both the Fc epsilon RII/CD23 and the class II MHC antigen expression, and the release of soluble CD23 in a dose-dependent manner, without affecting the kinetics of these responses. Furthermore, LTB4, but not LTC4, amplified both the proliferative response and the IgG and IgM secretion induced by addition of a suboptimal dose of IL-4 (3 U/ml) to SAC-preactivated B lymphocytes. Again, LTB4 did not modify the kinetics of the proliferative response promoted by IL-4. Although LTB4 potentiated IL-4-induced IgG and IgM secretion from SAC-activated B lymphocytes, no production of IgE was observed. These data indicate that LTB4 could play a regulatory role in the modulation of IL-4-induced signaling in human B lymphocytes.     

Requirement for BSF-1 in the induction of antigen-specific B cell proliferation by a thymus-dependent antigen and carrier-reactive T cell line

We report here a role of B cell stimulatory factor 1 (BSF-1) in the induction of antigen-specific proliferation of affinity-purified small B lymphocytes by a thymus-dependent antigen and a carrier-reactive T cell line. By using an ovalbumin-reactive T cell line (designated Hen-1), which does not produce BSF-1 following activation, it was possible to demonstrate that the antigen-specific proliferative response of trinitrophenyl (TNP)-binding B cells to TNP-ovalbumin required exogenous BSF-1 in addition to direct interaction with irradiated Hen-1 T cells. The activation obtained under these conditions was highly efficient, being sensitive to antigen doses as low as 0.001 microgram/ml. The addition of saturating amounts of BSF-1 did not alter the antigen-specificity or the requirements for hapten-carrier linkage or major histocompatibility complex-restricted T-B interaction in this system. The involvement of BSF-1 was confirmed by the ability of 11B11 anti-BSF-1 antibody to specifically suppress the response of TNP-binding B cells to TNP-ovalbumin, BSF-1, and irradiated Hen-1 T cells. Finally, this response was augmented by addition of the monokine interleukin 1. These data indicate that the proliferative response of small B cells to the thymus-dependent antigen and carrier-reactive T cell line used in our experiments can be regulated by the same factors that govern B cell proliferation induced by thymus-independent type 2 antigens or anti-IgM antibodies.     

Capacity of B cells to function as stimulators of a primary mixed leukocyte reaction

The capacity of B cells to serve as stimulator cells for a primary mixed leukocyte reaction (MLR) was evaluated. Percoll-fractionated B cells were stimulated with lipopolysaccharide and dextran sulfate (L/D) or a B cell stimulatory factor (BSF-1)-containing culture supernatant, and then were fixed before being used as stimulator cells to more precisely define the state of activation associated with MLR stimulatory capacity. It was found that unstimulated B cells or B cells stimulated for 1 day with L/D or BSF-1 were incapable of initiating a primary MLR, whereas B cells incubated for 3 days in L/D were potent stimulators. The differential activity of 1 day L/D- and BSF-1-activated B cells compared with 3 day L/D-activated B cells was not related to the amount of the relevant MHC class I or class II alloantigens on these cell populations, because all three groups had large increments in MHC class II expression in the following order: BSF-1 greater than 3 day L/D greater than 1 day L/D, and had little difference in MHC class I expression. Also, all three populations were capable of stimulating both MHC class I- and class II-specific T cell hybrids. It was concluded that the capacity of 3 day L/D-activated cells to stimulate a primary MLR was due to the elaboration of necessary co-stimulator molecules. We evaluated whether interleukin 1 (IL 1) was the co-stimulator involved. That this was not the case was indicated by two findings. First, 3 day-activated L/D cells failed to express IL 1 activity as measured by a highly sensitive IL 1 assay that utilizes the T cell line D10.G4.1. Second, recombinant IL 1 added to MLR cultures containing 1 day L/D- or BSF-1 activated B cells failed to function as a co-stimulator. In contrast, the phorbol ester PMA was a potent co-stimulator in this system. We conclude from these experiments that appropriately activated B cells can function as stimulators of a primary MLR, and that they elaborate critical co-stimulator molecules, distinct from IL 1, that enable them to function in this regard.     

Production of BSF-1 during an in vivo, T-dependent immune response

BSF-1, a cytokine produced by some T lymphocyte tumors, has been shown to act with anti-Ig antibodies to stimulate B lymphocyte proliferation, to independently induce resting B lymphocytes to increase their expression of surface Ia antigen, and to induce some activated B lymphocytes to differentiate into IgG1- or IgE-secreting cells. To determine whether BSF-1 might be secreted by normal lymphoid cells in the course of a physiologic immune response, BALB/c mice were injected with an affinity-purified goat antibody to mouse IgD (GaM delta), which induces the generation of a large, polyclonal T-dependent IgG1 response; 4-hr culture supernatants of spleen cells from these mice were prepared, and these supernatants were assayed for BSF-1 activity by analyzing their ability to induce BALB/c nu/nu spleen cells to increase their expression of cell surface Ia in vitro. Culture supernatants of unfractionated spleen cells removed from mice 4 to 8 days after GaM delta antibody injection induced substantial increases in B lymphocyte surface Ia expression; these increases were blocked by a monoclonal anti-BSF-1 antibody. Culture supernatants of spleen cells from untreated BALB/c mice or from untreated or GaM delta antibody-treated BALB/c nu/nu mice induced small to moderate increases in B cell surface Ia expression, and GaM delta antibody itself induced large increases in B cell surface Ia expression; however, these increases were not significantly blocked by a monoclonal anti-BSF-1 antibody. A culture supernatant of T cell-enriched spleen cells from untreated mice induced small increases in B cell surface Ia expression that were inhibited by anti-BSF-1 antibody, as was the larger increase in B cell Ia expression induced by a culture supernatant of T cell-enriched spleen cells from mice sacrificed 3 days after GaM delta injection. On the other hand, T cell-depleted spleen cells from BALB/c mice injected with GaM delta antibody 7 days before sacrifice failed to generate culture supernatants with BSF-1 activity. Supernatants prepared from spleen cells taken from untreated mice or mice treated with GaM delta antibody 1 to 3 days before sacrifice did not block the ability of purified BSF-1 to induce an increase in B cell surface Ia expression, and thus did not contain inhibitors of BSF-1 activity.(ABSTRACT TRUNCATED AT 400 WORDS)     

Major histocompatibility complex class II antigen expression during potentiation of line-10 tumor immunity after intralesional administration of bacillus Calmette-Guérin

Summary Intralesional injection of BCG into an established line-10 hepatocellular carcinoma in the strain-2 guinea pig causes regression of the tumor and induction of line-10 immunity. We found that the animals were already protected for a second challenge with line-10 tumor cells 7 days after BCG treatment. We studied whether this early induction of immunity was correlated with the expression of MHC class II antigens on line-10 tumor cells and was correlated with an increased expression of MHC class II antigens on leukocytes in the primary tumor and in the regional lymph node (Ln. axillaris accessorius). The MHC class II antigens and the leukocyte subpopulations were measured with monoclonal antibodies and flow cytofluorometry. In the draining lymph node the number of nucleated cells increased about 10-fold during the first 5 days after intralesional injection of BCG. At this time the MHC class II antigen expression of these cells was increased from 21%–32% in the naive controls to 39%–53% in animals with BCG-treated tumors. This implies that the number of MHC-class-II-positive cells increased about 20-fold in the draining lymph node. Surprisingly, the increase in percentage of MHC-class-II-antigen-positive cells was mainly due to an increase of IgM-positive B cells from 8%–11% to 22%–41% and an increase of IgG-positive B cells from 7%–27% to 25%–44%. In the tumor, BCG treatment induced a small increase of MHC-class-II-antigen-positive cells from 11%–12% to 15%–20%. Probably this increase came not from tumor cells but mainly from a BCG-induced infiltration of mononuclear cells, as an increase of T cells from 14% to 20%, an increase of macrophages from 8% to 18%, and an increase of B cells from 0 to 6% was observed. We conclude that the potentiation of anti-(line-10 tumor cell) immunity correlated with a 20-fold increase of MHC-class-II-antigen-positive cells in the lymph nodes and a small increase in the number of MHC-class-II-antigen-positive tumor-infiltrating cells.     

B cell stimulatory factor-1 (interleukin 4) activates macrophages for increased tumoricidal activity and expression of Ia antigens

Macrophages are activated by lymphokines (LK) to kill tumor cell and microbial targets. Interferon-gamma (IFN) is the major LK activity in conventional, antigen or mitogen-stimulated spleen cell culture fluids for induction of these macrophage effector functions. In view of the recent demonstration that murine macrophage-like cell lines have receptors for B cell stimulatory factor-1/interleukin 4 (BSF-1), a possible role for BSF-1 in regulation of macrophage function was considered. In this communication, thioglycollate-elicited murine peritoneal macrophages were shown to express about 2300 high affinity (Ka approximately 2 X 10(10) M-1) BSF-1 receptors/cell. Peritoneal macrophages treated with purified, T cell-derived BSF-1 developed potent tumoricidal activity against fibrosarcoma target cells. The concentration of BSF-1 that induced 50% of maximal tumor cytotoxicity was 38 +/- 4 U/ml for seven experiments; similar dose-responses were observed with recombinant BSF-1. That BSF-1 dose-responses for induction of macrophage-mediated tumor cytotoxicity were not affected by 5 micrograms/ml polymyxin B suggested that contaminant endotoxins played little or no role in cytotoxic activity. BSF-1 alone (less than or equal to 500 U/ml) was not directly toxic to tumor cells or macrophages. Macrophage tumoricidal activity induced by BSF-1 but not by IFN was inhibited greater than or equal to 90% with monoclonal anti-BSF-1 antibody. BSF-1 induced Ia antigen expression on peritoneal macrophages and increased (twofold to threefold) FcR(II)-dependent binding of murine IgG immune complexes to bone marrow-derived macrophages (greater than 98% macrophages). Based on these findings, it was concluded that BSF-1 is a potent macrophage activation factor.     

Modulation of interferon-gamma-induced major histocompatibility (MHC) and CD14 antigen changes by lipophilic muramyltripeptide MTP-PE in human monocytes

The lipophilic muramylpeptide derivative muramyltripeptide-phosphatidylethanolamine (MTP-PE, 0.05 to 5 micrograms/ml) and human recombinant interferon-gamma (rIFN-gamma, 1 to 100 U/ml) were applied singly or in combination to fresh human mononuclear blood leucocytes in vitro. After 15 to 72 hr incubation, culture- and drug-induced changes in beta 2-microglobulin (MHC class I associated), HLA-DR (MHC class II), and Leu-M3 (CD14) antigen expression were investigated by flow cytometry; changes in monocyte morphology (forward light scatter and side scatter) were assessed by scatter analysis. It was found that (1) rIFN-gamma caused a simultaneous down-regulation of the CD14 antigen and an up-regulation of MHC class I and class II molecules on the surface of cultured monocytes; (2) MTP-PE, which by itself failed to influence the expression of these antigens, synergized with rIFN-gamma in increasing MHC antigens and reducing CD14; (3) at high concentrations rIFN-gamma reduced monocyte viability to a small but significant extent and this effect was further potentiated by MTP-PE; and (4) untreated monocytes in culture showed an apparently MTP-PE-insensitive increase in size, density, and beta 2-microglobulin, HLA-DR, and CD14 antigen expression. The influence of MTP-PE on rIFN-gamma-induced surface marker changes may contribute to its immunoadjuvant activity in vivo.     

LPS-activated CBA/N mouse B cells respond to anti-Ig and a BSF-1-like factor

Highly purified small splenic CBA/N B cells show little or no proliferative response to LPS, soluble anti-Ig, LPS plus anti-Ig, or anti-Ig plus the B cell stimulatory factor BSF-1. An excellent proliferative response is obtained, however, if CBA/N B cells are cultured concurrently with LPS, anti-Ig, and a supernatant rich in T cell-derived lymphokines. The pertinent T cell-derived CBA/N B cell co-stimulating factor has the same m.w., isoelectric point range, and hydrophobicity as BSF-1, and co-migrates with BSF-1 throughout a two-step biochemical scheme developed for BSF-1 purification. These data therefore suggest that CBA/N B cells respond to a BSF-1-like lymphokine under appropriate activation conditions. In support of this conclusion, separate experiments demonstrated that unstimulated small CBA/N B cells respond to HPLC-purified BSF-1 by increased expression of membrane-bound class II major histocompatibility antigens. Taken together, these findings indicate that small CBA/N B cells express the receptor for a factor resembling BSF-1, and acquire the capacity to proliferate in response to anti-Ig and this BSF-1-like factor when co-stimulated with LPS.     

The effects of cytokines and adherent cells on the interleukin 4-mediated induction of Ia antigens on resting B cells

In this report we have extended our previous studies on interleukin 4 (IL-4) [previously termed B-cell stimulatory factor-1 (BSF-1)]. Our results demonstrate that 8 hr of exposure to IL-4 is sufficient to induce maximal expression of Ia antigens. This increase in expression of Ia antigens on resting B cells is due to the direct action of IL-4 on the B cells since adding or removing adherent cells or utilizing low density cultures of B cells at 50-100/culture had no effect on the IL-4-mediated increase in Ia. Monoclonal anti-IL-4 antibody completely abrogated the Ia-inducing activity of IL-4. A variety of other purified lymphokines including interleukin 2 (IL-2), interleukin 1 (IL-1), and a source of either B-cell differentiation factor for IgM (BCDF mu), or B-cell growth factor II (BCGF II), did not alter the expression of Ia antigens on resting B cells. However, interferon-gamma can partially inhibit the IL-4-mediated induction of Ia.     

Analysis by flow cytometry of the subpopulations of lymphocytes from the peripheral blood of procain or diethylaminoethanol treated rabbits

The authors worked on 3 groups of 8 male rabbits, New Zealand race: 1) controls; 2) procain injected i.m., 15 mg/kg body weight, daily, for 30 days; 3) i.m. injected with diethylaminoethanol (DEAE), 15 mg/kg body weight, daily, for 35 days. The expression of the MHC I, MHC II, CD43, CD4 and IgM antigenic markers on the plasmatic membrane of the lymphocytes was studied using flow cytometry and monoclonal antibodies. Procain or DEAE treatment reduced the percentage of lymphocytes expressing I MHC, from 99.06 in the control group, to 94.51 in procain group and to 96.91 in the DEAE group. The intensity of expression of MHC complexes of class II decreases from 160.94 in the control group, to 107.21 in the procain group and to 104.05 in the DEAE group. No significant differences were noticed between the three groups of rabbits concerning the rate of lymphocytes that have on their surface expressed markers for CD43 (lymphocytes T), CD4 (Th), or IgM (lymphocytes B). Lymphocytosis induced in rabbits as a result of the DEAE treatment took place without a change in the proportions of lymphocyte subpopulations. The authors consider that owing to their capacity to reduce the expression of antigens MHC of class I and class II on the membrane of lymphocytes, procain and DEAE can have benefic effects in some autoimmune, autoaggression and inflammatory diseases.     

Major histocompatibility complex class II antigen expression on leucocyte subpopulations in the draining lymph node and tumour in the early phase of bacillus-Calmette-Guérin-induced tumour regression

Summary We investigated the cellular composition and the major histocompatibility complex (MHC) class II antigen expression in the draining lymph node and the tumour during potentiation of the immune response by intralesional bacillus Calmette-Guérin (BCG) administration in the line 10 hepatocellular carcinoma in the strain 2 guinea-pig. Five days after its injection BCG induced a ninefold increase in the number of draining lymph node cells and an increased MHC class II expression. This increased MHC class II expression was mostly due to the selective increase of B cells in the lymph nodes, and to a lesser extent to the increase of T cells expressing MHC class II antigens. Taking into account this nine-fold increase, intralesional treatment of BCG increased considerably the number of T helper/inducer (anti-CT7) and T suppressor/cytotoxic (anti-CT6) lymph node cells expressing MHC class II antigen. The percentage of tumourinfiltrating T cells expressing MHC class II antigen in the tumour was higher than the percentage of T cells in the regional draining lymph node of non-treated guinea-pigs, indicating the presence of activated T cells in the tumour. After treatment with BCG no further increase in MHC class II expression was measured in the tumour, nor was any phenotypical change of the tumour-infiltrating T cells found. In conclusion, with the use of two-colour flow cytofluorometry we have shown that the potentiation of the already existing immune response to line 10 is accompanied by a considerable increase in T helper/inducer, T suppressor/cytotoxic cells and MHC class II antigen in the regional lymph node. Whether this is essential for the potentiation of the immune response causing tumour regression and long-lasting immunity is a subject for further study.     

Regulation of class II gene expression in a murine pre-B cell line

Class II major histocompatibility complex (MHC) gene expression has been studied in an Abelson virus-transformed pre-B cell line R8, and its Ia-negative variant R8205. These variant cells contained barely detectable levels of RNA specific for all class II genes, including the nonpolymorphic invariant chain gene (Ii), and did not express cell surface Ia. Fusion of this murine Ia-negative cell line to the human Ia-positive Raji cell produced an interspecies hybridoma that expressed the murine Ia. These data are further evidence for the existence of trans-acting factors that can regulate class II gene expression. Furthermore, the T cell-derived lymphokine B cell stimulatory factor 1 (BSF-1) induced expression of class II genes in the R8205 cells. Exposure of R8205 cells to an antibody that has been shown to mimic BSF-1 activity on normal B cells also resulted in expression of class II genes. These data demonstrate that three distinct signals--a lymphokine, an alloantibody binding to membrane structures, and an interspecies trans-acting factor--can induce expression of class II genes.     

Differential induction of stromelysin mRNA by bovine articular chondrocytes treated with interferon-γ and interleukin-1α

Articular chondrocytes from rheumatoid joints have been shown to express class II major histocompatibility (MHC) antigens that were correlated with the presence of interferon-gamma (IFN-γ) in the inflamed joint. Chondrocytes expressing MHC antigens function as antigens function as antigen presenting cells and thus stimulate lymphocyte proliferation. These responses suggest a powerful role for the IFN-γ stimulation of chondrocytes. The present studies were designed to examine the functional role of chondrocytes exposed to IFN-γ during cartilage degradation that occurs in synovial disease. Destruction of cartilage in arthritis is partially attributable to metalloproteinases released by the chondrocytes in response to interleukin-1 (IL-1). Bovine articular chondrocytes treated with interleukin-1 alpha (IL-1α) produced enhanced levels of stromelysin mRNA, however, Northern blots could not determine the percentage of cells responding. Exposure of bovine articular chondrocytes to IFN-γ induced the expression of bovine HLA-DR (boHLA-DR) antigen in 50% of the cells. Using a modified cell sorting technique, chondrocytes that expressed class II MHC antigens produced two fold greater stromelysin mRNA than chondrocytes that did not express this antigen. In contrast, collagen type II mRNA levels were similar in chondrocytes, regardless of the expression of class II MHC antigens. In situ hybridization studies showed that less than half of all cartilage chondrocytes were induced to synthesize stromelysin mRNA. These observations suggest that IFN-γ stimulates specific subpopulations of chondrocytes to be functionally active in inflammation-induced metalloprotease secretion. © 1993 Wiley-Liss, Inc.