Regulated production and molecular diversity of human liver and activation-regulated chemokine/macrophage inflammatory protein-3 alpha from normal and transformed cells.

Liver and activation-regulated chemokine (LARC), also designated macrophage inflammatory protein-3alpha (MIP-3alpha), Exodus, or CCL20, is a C-C chemokine that attracts immature dendritic cells and memory T lymphocytes, both expressing CCR6. Depending on the cell type, this chemokine was found to be inducible by cytokines (IL-1beta) and by bacterial, viral, or plant products (including LPS, dsRNA, and PMA) as measured by a specific ELISA. Although coinduced with monocyte chemotactic protein-1 (MCP-1) and IL-8 by dsRNA, measles virus, and IL-1beta in diploid fibroblasts, leukocytes produced LARC/MIP-3alpha only in response to LPS. However, in myelomonocytic THP-1 cells LARC/MIP-3alpha was better induced by phorbol ester, whereas in HEp-2 epidermal carcinoma cells IL-1beta was the superior inducer. The production levels of LARC/MIP-3alpha (1-10 ng/ml) were, on the average, 10- to 100-fold lower than those of IL-8 and MCP-1, but were comparable to those of other less abundantly secreted chemokines. Natural LARC/MIP-3alpha protein isolated from stimulated leukocytes or tumor cell lines showed molecular diversity, in that NH(2)- and COOH-terminally truncated forms were purified and identified by amino acid sequence analysis and mass spectrometry. In contrast to other chemokines, including MCP-1 and IL-8, the natural processing did not affect the calcium-mobilizing capacity of LARC/MIP-3alpha through its receptor CCR6. Furthermore, truncated natural LARC/MIP-3alpha isoforms were equally chemotactic for lymphocytes as intact rLARC/MIP-3alpha. It is concluded that in addition to its role in homeostatic trafficking of leukocytes, LARC/MIP-3alpha can function as an inflammatory chemokine during host defense.

Isolation of the CXC chemokines ENA-78, GRO alpha and GRO gamma from tumor cells and leukocytes reveals NH2-terminal heterogeneity. Functional comparison of different natural isoforms.

Chemokines are a family of chemotactic peptides affecting leukocyte migration during the inflammatory response. Post-translational modification of chemokines has been shown to affect their biological potency. Here, the isolation and identification of natural isoforms of the neutrophil chemoattractants GRO alpha and GRO gamma and the epithelial-cell-derived neutrophil attractant-78 (ENA-78), is reported. Cultured tumor cells produced predominantly intact chemokine forms, whereas peripheral blood monocytes secreted mainly NH2-terminally truncated forms. The order of neutrophil chemotactic potency of these CXC chemokines was GRO alpha > GRO gamma > ENA-78 both for intact and truncated forms. However, truncated GRO alpha (4,5,6-73), GRO gamma (5-73) and ENA-78(8,9-78) were 30-fold, fivefold and threefold more active than the corresponding intact chemokine. As a consequence, truncated GRO alpha (4,5,6-73) was 300-fold more potent than intact ENA-78 indicating that both the type of chemokine and its mode of processing determine the chemotactic potency. Similar observations were made when intact and truncated GRO alpha, GRO gamma and ENA-78 were compared for their capacity to induce an increase in the intracellular calcium concentration in neutrophilic granulocytes, and to desensitize the calcium response towards the CXC chemokine granulocyte chemotactic protein-2 (GCP-2). It must be concluded that physiological proteolytic cleavage of CXC chemokines in general enhances the inflammatory response, whereas for CC chemokines NH2-terminal processing mostly results in reduced chemotactic potency.

Differential induction of monocyte chemotactic protein-3 in mononuclear leukocytes and fibroblasts by interferon-alpha/beta and interferon-gamma reveals MCP-3 heterogeneity.

Monocyte chemotactic protein-3 (MCP-3) is a pluripotent CC chemokine, attracting most leukocytic cell types. With the use of a sensitive and specific ELISA, MCP-3 was found to be inducible in fibroblasts and peripheral blood mononuclear cells (PBMC) by cytokines and cytokine inducers. MCP-3 production levels (1-10 ng/ml) were tenfold lower compared to those of MCP-1. In diploid fibroblasts, synergistic induction of MCP-3, but not of MCP-1, mRNA and protein was observed by combined treatment with IL-1beta and IFN-gamma. In PBMC, IFN-alpha and IFN-beta (but not IFN-gamma), as well as measles virus and double-stranded RNA, were potent inducers of MCP-3, which suggests a role for this chemokine in an early stage of viral infections. In contrast, endotoxin failed to induce MCP-3 production in fibroblasts and PBMC. Purification of MCP-3 from PBMC revealed biochemical heterogeneity. In monocyte chemotaxis and calcium mobilization assays, pure 11-kDa MCP-3 from PBMC showed similar potencies as MCP-3 from tumor cells. It was concluded that the induction of MCP-3 by IFN is regulated differently in fibroblasts and PBMC. In view of the multiple target cells for MCP-3, local and strictly regulated chemokine production might be important to conduct selectively the immune response in infection or inflammation.

Posttranslational modifications affect the activity of the human monocyte chemotactic proteins MCP-1 and MCP-2: identification of MCP-2(6-76) as a natural chemokine inhibitor.

Chemokines are important mediators in infection and inflammation. The monocyte chemotactic proteins (MCPs) form a subclass of structurally related C-C chemokines. MCPs select specific target cells due to binding to a distinct set of chemokine receptors. Recombinant and synthetic MCP-1 variants have been shown to function as chemokine antagonists. In this study, posttranslationally modified immunoreactive MCP-1 and MCP-2 were isolated from mononuclear cells. Natural forms of MCP-1 and MCP-2 were biochemically identified by Edman degradation and mass spectrometry and functionally characterized in chemotaxis and Ca2+-mobilization assays. Glycosylated MCP-1 (12 and 13.5 kDa) was found to be two- to threefold less chemotactic for monocytes and THP-1 cells than nonglycosylated MCP-1 (10 kDa). Natural, NH2-terminally truncated MCP-1(5-76) and MCP-1(6-76) were practically devoid of bioactivity, whereas COOH-terminally processed MCP-1(1-69) fully retained its chemotactic and Ca2+-inducing capacity. The capability of naturally modified MCP-1 forms to desensitize the Ca2+ response induced by intact MCP-1 in THP-1 cells correlated with their agonistic potency. In contrast, naturally modified MCP-2(6-76) was devoid of activity, but could completely block the chemotactic effect of intact MCP-2 as well as that of MCP-1, MCP-3, and RANTES. Carboxyl-terminally processed MCP-2(1-74) did retain its chemotactic potency. Although comparable as a chemoattractant, natural intact MCP-2 was found to be 10-fold less potent than MCP-1 in inducing an intracellular Ca2+ increase. It can be concluded that under physiologic or pathologic conditions, posttranslational modification affects chemokine potency and that natural MCP-2(6-76) is a functional C-C chemokine inhibitor that might be useful as an inhibitor of inflammation.

Purification and Identification of Natural Chemokines

A novel family of chemotactic cytokines or chemokines, essential for the directed migration of leukocytes to sites of inflammation, has been identified during the past decade. To obtain microgram amounts of natural chemokines, normal (e.g., freshly isolated leukocytes, connective tissue cell cultures) or malignant cell lines have to be selectively induced with endogenous (cytokines) or exogenous (bacterial, viral, or plant) products. We have developed a four-step procedure that allows for the complete purification of active C-C (MCP-1, MCP-2, MCP-3, RANTES, MIP-1alpha and MIP-1beta) and C-X-C (IL-8, GRO-alpha, GRO-beta, GRO-gamma, GCP-2, ENA-78, IP-10, PF-4, and CTAPIII/betaTG/NAP-2) chemokines from bulk volumes of culture supernatant. This method is applicable for the isolation of recombinant chemokines. Conditioned medium was first concentrated and partially purified on silicic acid or controlled pore glass beads. Further purification to homogeneity was achieved using heparin-Sepharose or antibody affinity chromatography, cation exchange FPLC, and reverse-phase HPLC. Purification of chemokines was monitored by testing column fractions for biological (chemotaxis) or immuno (RIA, ELISA) activity and protein content (SDS-PAGE). Homogeneous proteins were identified by amino-terminal or internal protein sequence analysis.

Identification of mouse granulocyte chemotactic protein-2 from fibroblasts and epithelial cells. Functional comparison with natural KC and macrophage inflammatory protein-2.

Neutrophil and monocyte chemotactic factors were isolated from conditioned media of mouse fibroblasts and epithelial cells. Neutrophil chemotactic activities were purified to homogeneity using a four-step chromatographic procedure, and the corresponding proteins were identified by amino acid sequence analysis. Natural forms of the murine chemokines KC and macrophage inflammatory protein-2 were isolated from virus-infected fibroblasts. However, the major neutrophil chemotactic activity from fibroblasts stimulated with endotoxin plus double-stranded RNA and from PMA-treated epithelial cells resided in other 7- and 8-kDa proteins. Amino acid sequence analysis revealed a novel Cys-Xaa-Cys chemokine structure, characterized by the conservation of four cysteines and the Glu-Leu-Arg motif. Based on the completely identified primary structure of this natural protein, this chemokine must be considered to be the murine homologue of human and bovine granulocyte chemotactic protein-2 (GCP-2; 61 and 64% identical residues, respectively). Due to NH2-terminal cleavage, 11 different forms of mouse GCP-2 were discovered. In contrast to human and bovine GCP-2, functional comparison of long and short NH2-terminal forms of mouse GCP-2 demonstrated that truncated mouse GCP-2 (short form) has a higher specific activity in neutrophil activation (gelatinase B release) and chemotaxis assays. Furthermore, mouse GCP-2 was more potent than human GCP-2 on human neutrophils, and more active than murine KC and macrophage inflammatory protein-2 on mouse neutrophils. In view of the absence of a murine homologue for IL-8, NH2-terminally processed GCP-2 can be considered a major neutrophil chemoattractant in the mouse during the inflammatory response.

Induction of monocyte chemotactic proteins MCP-1 and MCP-2 in human fibroblasts and leukocytes by cytokines and cytokine inducers. Chemical synthesis of MCP-2 and development of a specific RIA.

Monocyte chemotactic proteins (MCP) belong to a group of structurally and functionally related factors, called chemokines. To facilitate additional characterization of the recently identified MCP-2, the 76-residue protein was chemically synthesized. The synthetic 7-kDa monomeric protein was chemotactic for monocytes at 1 nM and was biochemically similar to natural MCP-2. Sensitive radioimmunoassays for both MCP-1 and MCP-2 were developed. These RIAs were specific in that no cross-reactivity could be observed, and other chemokines or cytokines were not detected. Induction of MCP-1 and MCP-2 in human diploid fibroblasts and peripheral blood leukocytes as well as osteosarcoma, epidermal carcinoma, and melanoma cells by the cytokines IL-1 beta, IFN-beta, and IFN-gamma and cytokine inducers such as dsRNA, virus, endotoxin, mitogen, and phorbol ester was studied. In connective tissue cells, IL-1 beta was the best inducer of MCP-1, but IFN-gamma was a superior inducer of MCP-2. Mononuclear cells also proved to be a source of MCP-1 and MCP-2 when stimulated by most of the inducers tested. Granulocytes, however, were inefficient producers. Measles virus induced MCP-1 and MCP-2 in most cell types. In general, the yields of MCP-2 were at least 10-fold lower than those of MCP-1. It is concluded that, although MCP-2 is often coproduced with MCP-1, regulation of expression of the two chemokines is not identical. It remains to be studied under which pathological conditions MCP-2 is released in vivo and whether MCP-1 and MCP-2 can activate different target cells.

Human and bovine granulocyte chemotactic protein-2: complete amino acid sequence and functional characterization as chemokines.

Tumor cells are capable of simultaneously producing a number of related inflammatory peptides, now classified as chemokines. We have isolated a new human granulocyte chemotactic protein (GCP-2), coproduced with interleukin-8 (GCP-1/IL-8) by osteosarcoma cells. Furthermore, the bovine homologue of human GCP-2 was purified from kidney tumor cells using the same isolation procedure. Both chemokines occur in at least four NH2-terminally truncated forms. These 5-6 kDa proteins do not differ in potency and efficacy as granulocyte chemotactic factors using a standard in vitro migration assay. The complete primary structures of human and bovine GCP-2 were disclosed by sequencing peptide fragments derived from the natural proteins. On the basis of the conservation of four cysteine residues, the two molecules are to be classified within the C-X-C chemokine family, including IL-8. Human and bovine GCP-2 are 67% similar at the amino acid level. Their sequences show only weak similarity with that of IL-8, and human GCP-2 does not cross-react in a radioimmunoassay for IL-8. Human and bovine GCP-2 are specific granulocyte chemotactic factors in that they do not attract human monocytes. Bovine GCP-2 is not species specific since it is at least as active as human GCP-2 on human granulocytes. Both chemokines can also activate postreceptor mechanisms leading to release of gelatinase B by granulocytes. This is indicative for a possible role in inflammation and tumor cell invasion.

Identification of a novel granulocyte chemotactic protein (GCP-2) from human tumor cells. In vitro and in vivo comparison with natural forms of GRO, IP-10, and IL-8.

Stimulated human osteosarcoma cells (MG-63) were used as a source of granulocyte chemotactic protein (GCP). In addition to the previously isolated GCP-1/IL-8, natural forms of GRO alpha, GRO gamma, and IP-10 were purified and identified by amino acid sequence analysis. Further, a novel GCP, GCP-2, was isolated in its natural form (6 kDa) and was found to be structurally related to the other members of the IL-8 family. GRO alpha, IP-10, and GCP-2 showed heterogeneity, in that several forms of each protein were recovered. These differed in truncation at the amino terminus. Reverse phase HPLC allowed us to separate four such different forms of GCP-2. These tumor-derived factors were compared in granulocyte activation and chemotaxis assays. IL-8 induced neutrophil gelatinase B release at 2 nM, but GRO alpha and GCP-2 showed a 5- to 10-fold lower specific activity. When the migration of granulocytes through polycarbonate micropore membranes was measured, GCP-2 and GRO alpha had a maximal chemotactic index comparable to that of IL-8. The minimal effective dose for GCP-2 and GRO alpha was 3 to 10 nM, whereas the specific activity of IL-8 was at least 10-fold higher. IP-10 was not active in this assay at doses up to 100 nM. Finally, in vivo chemotaxis was measured by using granulocyte recruitment in the rabbit skin model. After intradermal injection of 200 ng/site, GCP-2 provoked a significant granulocyte infiltration, albeit to a lesser extent than did IL-8 and GRO alpha. GCP-2 did not attract monocytes in vivo nor did it induce the cells in vitro to migrate or to produce enzyme. In conclusion, this study reveals a new member of the IL-8 family and shows that these related inflammatory mediators possess different potencies and efficacies towards granulocytes.

Production and identification of natural monocyte chemotactic protein from virally infected murine fibroblasts. Relationship with the product of the mouse competence (JE) gene.

Primary cultures of mouse embryonic fibroblasts and confluent monolayers of mouse fibroblastoid cells (L929) were found to secrete a chemotactic factor specific for monocytes. It biological activity was deduced from both the migration distance under agarose and the number of migrated monocytes in the micropore filter method. The monocyte chemotactic protein (MCP) was inducible in these cells by double-stranded RNA and by infection with virus. In embryonic fibroblasts MCP was also produced in response to the cytokines interleukin-1 (IL-1) and tumor necrosis factor (TNF). Under all conditions for induction of MCP tested no production of chemotactic activity for granulocytes could be detected. MCP activity from virally infected L929 cells was concentrated and purified by sequential adsorption to controlled pore glass, heparin-Sepharose chromatography, ion-exchange FPLC and reversed-phase HPLC. Pure MCP was found to occur mainly as a 7-8-kDa protein. Although the mature protein possessed a blocked NH2-terminus, it was identified by enzymatic cleavage and sequence analysis of an internal fragment. The sequence obtained corresponded to a part of the cDNA-derived protein sequence of the murine 'competence' (JE) gene, inducible in fibroblasts by cytokines and virus. In all probability the 7-8-kDa MCP form represents the natural product of the mouse gene JE. Murine MCP can thus be classified in the novel family of small inducible inflammatory proteins.

The neutrophil-activating proteins interleukin 8 and beta-thromboglobulin: in vitro and in vivo comparison of NH2-terminally processed forms.

Isolation of the human neutrophil activating protein (NAP) interleukin 8 (IL8) from leukocytes has revealed that it is structurally related to beta-thromboglobulin (beta TG) from platelets. Both these proteins occur as natural mixtures of multiple forms, differing from each other by unequal truncation at the NH2 terminus. In this study we have compared IL8 and beta TG forms for in vitro and in vivo neutrophil activation. In contrast to IL8, none of the beta TG forms were found to exert granulocyte chemotactic activity in vitro, as measured in the agarose assay. However, fractions rich in the most extensively processed forms of beta TG (e.g. NAP-2) as well as pure NAP-2 did induce lactoferrin release from granulocytes, whereas fractions containing only the longer forms (e.g. connective tissue-activating peptide III) were inactive. In order to observe this in vitro effect, about 10-fold less IL8 (10 nM) than NAP-2 was required. In the presence of a vasodilator substance low doses (2-20 pmol) of IL8 and the shorter forms of beta TG caused granulocyte accumulation and plasma leakage in rabbit skin whereas the longer forms of beta TG again failed to do so. Finally, granulocytosis induction following i.v. injection was found to occur with NAP-2. At the maximal dose tested (250 pmol), this in vivo effect of NAP-2 was less pronounced than that of IL8. In the case of IL8, NH2-terminal processing did not seem to affect granulocyte stimulatory activity. It should be noted, however, that the extent of processing of IL8 is less than that occurring with beta TG. It can be concluded that the platelet factor beta TG, structurally related to the monokine IL8, can also play a role in neutrophil activation during inflammatory reactions.

Identification of the monocyte chemotactic protein from human osteosarcoma cells and monocytes: detection of a novel N-terminally processed form.

The chemotactic activity for monocytes in culture supernatants from double-stranded RNA-stimulated human MG-63 osteosarcoma cells and from LPS-stimulated human monocytes was purified to homogeneity and characterized by amino acid sequence analysis. The chemotactic protein derived from the fibroblastoid osteosarcoma cells had a blocked N-terminus but sequencing of tryptic fragments showed that it was identical with a recently identified monocyte chemoattractant designated MCP-1 or MCAF isolated from glioma or myelomonocytic cells, respectively. Preparations of monocyte -derived chemotactic activity appeared to contain not only the blocked protein, but also a novel N-terminally processed form of this molecule, lacking 5 amino acid residues.

Characterization of granulocyte chemotactic activity from human cytokine-stimulated chondrocytes as interleukin 8.

Human articular chondrocytes, when stimulated with interleukin 1 beta (IL 1 beta), tumor necrosis factor-alpha (TNF-alpha), or with the double stranded RNA poly (rI).poly (rC), produce a chemotactic activity for granulocytes. The induction with IL 1 beta could be abolished by an antibody to IL 1 beta but not by an antibody to interleukin 6 (IL 6), indicating that the latter is not a mediator for the production of chemotactic activity. The inducers had no direct chemotactic effect on granulocytes. The granulocyte chemotactic factor from chondrocytes was characterized with a specific antibody against leukocyte-derived interleukin 8 (IL 8). The specificity of this antibody was demonstrated by immunochemical and biological criteria such that it could immunoprecipitate only the 6-7 kDa IL 8 protein from fibroblasts, and that it did not neutralize a structurally related monocyte chemotactic protein. This antibody against IL 8 completely neutralized the granulocyte chemotactic activity from stimulated chondrocytes. This demonstrates the identity of chondrocyte IL 8 with leukocyte- and fibroblast-derived IL 8. Our data show that leukocyte chemotaxis into the inflamed joint can be mediated by IL 8, induced in both synovial fibroblasts and chondrocytes by the inflammatory cytokines IL 1 and TNF-alpha.

Identification by sequence analysis of chemotactic factors for monocytes produced by normal and transformed cells stimulated with virus, double-stranded RNA or cytokine.

A monocyte chemotactic activity was found to be released by various types of cultured human cells after appropriate stimulation: normal diploid fibroblasts, peripheral blood mononuclear cells or monocytes isolated therefrom, and a number of tumor cell lines, including osteosarcoma (MG-63) and hepatoma (Malavu) but not melanoma (Bowes) cells. Cultures of diploid human fibroblasts and these tumor cells stimulated with interleukin (IL) 1 or double-stranded RNA [poly(rI).poly(rC)], or infected with viruses (measles or rubella viruses) were found to produce chemotactic activity for both monocytes and granulocytes. Media collected from fibroblasts treated with E. coli or IL 6 did not contain such activity. Granulocyte and monocyte chemotactic activities were serologically distinct, and could be separated by successive chromatographical procedures. While the granulocyte chemotactic activity of both fibroblasts and MG-63 cells had previously been identified as granulocyte chemotactic protein/IL 8, the monocyte chemotactic activity from MG-63 cells was identified by amino acid sequence analysis as a different protein recently described to be released by human glioma and myelomonocytic cell lines. In view of the similarity in their chromatographical behavior, monocyte chemotactic activities from fibroblasts, MG-63 cells and fresh monocytes can probably be assigned to identical molecules. Cultures of unfractionated peripheral blood cells, however, were found to release an additional monocyte chemotactic protein, identifiable by amino acid sequence analysis as platelet factor 4.

The chemotactic activity for granulocytes produced by virally infected fibroblasts is identical to monocyte-derived interleukin 8.

So far, the role of fibroblasts in inflammatory processes has been underestimated. We have previously shown that stimulation of fibroblasts with viruses or bacteria results in a simultaneous production of several cytokines, including interferon-beta, interleukin (IL) 6 and colony-stimulating factors. We here report that virally infected fibroblasts produce also a chemotactic factor for granulocytes. The activity is inducible not only by measles virus but also by IL 1 beta and the double-stranded RNA poly(rI).poly(rC). This factor, when purified to homogeneity, occurs as a 6-7-kDa protein doublet upon sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The pure protein is serologically related to a fully characterized granulocyte chemotactic peptide (GCP) from monocytes, designated IL8. Furthermore, the chemotactic factor from fibroblasts has an NH2-terminal sequence identical to that of GCP/IL8, small differences in NH2-terminal processing being observed. Finally, in addition to diploid fibroblasts, the osteosarcoma MG-63 cell line is also a producer of GCP/IL8. It can thus be concluded that GCP/IL8 can be produced by several cell types in response to infection and that fibroblasts can contribute to chemotaxis in inflammation.

Purification of granulocyte chemotactic peptide/interleukin-8 reveals N-terminal sequence heterogeneity similar to that of beta-thromboglobulin.

Stimulated human peripheral blood leukocytes produce a chemotactic factor for granulocytes (granulocyte chemotactic peptide/interleukin-8; GCP/IL-8), which is structurally related to platelet-derived beta-thromboglobulin. Analytically pure CGP/IL-8 and beta-thromboglobulin could be obtained after three purification steps, comprising adsorption to silicic acid, heparin-Sepharose chromatography and ion-exchange chromatography. Although GCP/IL-8 and beta-thromboglobulin had a similar affinity for heparin, they could be separated on a cation-exchange column. Both molecules were heterogeneous in that 6-7-kDa protein doublets were detected upon SDS/PAGE. N-terminal amino acid sequence analysis revealed the presence of six immunologically related but differently truncated polypeptides of beta-thromboglobulin, of which only two corresponded to previously described forms. Similarly, apart from a major polypeptide, five minor species of GCP/IL-8 were detected that also differed by N-terminal truncation. The most processed forms of beta-thromboglobulin and GCP/IL-8 were found to have their N-terminus in that region of the primary structure where a significant similarity between the two molecules starts. GCP/IL-8 was found to be chemotactic for granulocytes with a specific activity of 10(5) units/mg, whereas none of the beta-thromboglobulin species possessed detectable chemotactic activity.

Purification and characterization of human fibroblast-derived hybridoma growth factor identical to T-cell-derived B-cell stimulatory factor-2 (interleukin-6).

Human fibroblast cultures, when stimulated with interleukin-1 (IL-1) produce a growth factor for B-cell hybridoma and plasmocytoma cell lines. The availability of both a fast-growing and high-producer cell line (MG-63 osteosarcoma cells) and of a highly sensitive and specific assay system for this hybridoma growth factor (HGF) allowed us to obtain analytically pure preparations. Crude HGF from MG-63 cells was processed through a five-step concentration and purification schedule. Sequential adsorption to controlled pore glass (CPG) beads, antibody affinity chromatography and gel filtration resulted in a 10,000-fold purification to a specific activity of 10(9) units/mg HGF. Electrophoretically pure HGF was obtained after additional purification by cation-exchange chromatography and reversed-phase HPLC. The purification procedure revealed two distinct biologically active HGF components. The amino-terminal sequence of one of the two components was determined and found to correspond to that already predicted from cDNA clones of a protein alternatively called 26-kDa protein, interferon-beta 2 (IFN-beta 2) or B-cell stimulating factor-2 (BSF-2). The first two designations (26-kDa protein and IFN-beta 2) refer to a postulated fibroblast secretory protein with so far no unambiguously defined function; the latter designation (BSF-2) refers to a T-cell product possessing differentiation stimulatory effect on B-cell lines. The reported results firmly establish that the protein is secreted by fibroblasts and reveal that it possesses B-cell growth stimulatory activity. The new designation interleukin-6 (IL-6) is proposed to resolve prescribing nomenclature confusion.