Stafford and colleagues report in this issue that IgG anti-ribosomal P protein antibodies (anti-P) in systemic lupus erythematosus (SLE) bind to a human lymphocyte surface membrane-associated 38-kD molecule apparently identical to the P 0 ribosomal protein [1]. Their data extend those of Koren et al. [2], who demonstrated cell surface reactivity of anti-P with human hepatoma, neuroblastoma, and skin fibroblast cell lines. Although these are not novel observations-sporadic descriptions of cell surface binding by autoantibodies to various nuclear and intracytoplasmic antigens have been in the literature for 20 years (reviewed in [3])-special care was taken by Stafford to minimize the ambiguity inherent in experiments of this type. IgG Fc/Fc receptor interactions were ruled out. Specificity of binding was established by antigen inhibition, use of affinity-purified anti-P, and immunoblotting. Reactivity of anti-P with viable cells and ribosome-free plasma membrane preparations was demonstrated. And experiments with cell surface proteins purified by biotinylation revealed reactivity only with a 38-kD molecule, not the small P 1 and P 2 ribosomal proteins that would be expected if ribosomes released from dead cells were becoming attached to living cells. While it would have been nice to also have immunoprecipitation and complementdependent cytotoxicity data, and begging the question of how a ribosomal protein gets to the outer surface of the cell in the first place, Stafford's observations clearly support her conclusions that 'anti-ribosomal antibodies are a subset of anti-lymphocyte autoantibodies'.If anti-P antibodies are a subset of anti-lymphocyte antibodies (ALAB), they certainly differ from the traditional version, which are described below and in Table 1.Originally described as 'cold-lymphocytotoxins' because they kill lymphocytes in the presence of complement better at 15ЊC than at 37ЊC [4], classic IgM ALAB in SLE react broadly with autologous lymphocytes and with lymphocytes from unrelated donors (reviewed in [5]). Early studies established a relative specificity for T cells, especially thymocytes, and detected interesting specificities for functionally significant T cell subsets. IgM antibodies to B cells were described, as well, and are at least partially distinct from autoantibodies to T cells [6]. More recent data have substantially clarified the nature and potential significance of autoantibodies in this system. Thus, reactivity of T cells to IgM ALAB was shown to be dependent primarily on the presence of antigen(s) that are shared by CD4 + and CD8 + subsets, with a superimposed specificity for an additional antigen(s) expressed on CD4 + T cells [7]. Activated T cells, which exhibit a higher density of many surface membrane antigens than do resting cells and also express neoantigens, were found to be special targets of such autoantibodies [8] Most patients with SLE express IgM ALAB, which vary in titre with disease activity status in the same fashion as anti-dsDNA autoantibodies [6,20]. This association...