Between the fall of 1986 and the spring of 1987, a number of groups involved in the identification of apparently unrelated factors came to the unsettling conclusion that they had cloned or purified the same protein. As a result of this hectic start, the molecule that we now call interleukin-6 (IL-6) was originally referred to by about ten different names, each reflecting a different characteristic of the protein. To clarify this rather confusing nomenclature, I start this review with a brief historical note. IL-6 is probably unique among cytokines because it was cloned almost inadvertently long before the discovery of its major biological activities. In the course of their efforts to clone interferon-fJ (IFN-fJ) cDNA from human fibroblasts, Weissenbach et al (1) isolated two cDNA clones derived from a 1.3-kb mRNA inducible by poly (rJ). poly (rC) and cycloheximide. The corresponding 26-kd protein was termed" IFN-fJz" because Xenopus laevis oocytes injected with this mRNA were found to produce an antiviral activity that could be inhibited by antisera to IFN-fJ. Following a similar approach, Content et al cloned the same 1.3-kb mRNA species (2) but concluded that the protein, which they termed 26K factor, had no antiviral activity and was not serologically related to IFN-fJ. The sequences pub lished several years later by the two groups (3, 4) demonstrated the lack of structural homology with IFN-fJ, but failed to resolve the controversy surrounding the interferon activity.

A second, more conventional, road to IL-6 started from the observation that activated T cells produce a late-acting B cell-differentiation factor capable of inducing immunoglobulin (I g) production in activated B cells or in Epstein-Barr virus (EBV)-immortalized B-celliines (5, 6). This mol ecule, termed B-cell stimulatory factor 2 (BSF-2), was purified to homo-