mAbs recognizing TCR V gene products have helped to define a unique group of potent T cell activators termed superantigens (1). Recognized superantigens include murine selfantigens (2, 3) and a group ofmicrobial toxins, the staphylococcal enterotoxins (SE) (1), and a soluble product ofMycoplasma arthritidis mitogen (MAM) (4). Superantigens share a number of intriguing characteristics. For example, superantigen recognition is almost entirely a function of TCR V a gene usage (1, 2).In addition, superantigens bind selectively and with high affinity to MHC class II molecules (6). In the absence of antigen processing and in an MHC-nonrestricted manner, superantigen-MHC class II complexes on the APC surface trigger the proliferation of T cells expressing the relevant TCR V gene product (5). Finally, the microbial toxins that function as superantigens are among the most potent mitogens known.The dual affinity of superantigens for MHC class II and TCR V gene products suggests that this class of antigen could promote a form of abnormal Th-B cell collaboration analogous to that observed during graft-vs .-host disease (GVHD). In GVHD, adoptive transfer of donor Th cells, specific for recipient MHC class II antigens, triggers polyclonal B cell activation . In selected strains of nonautoimmune prone mice, GVHD is characterized by an autoimmune diathesis remarkably similar to SLE (6). It has been suggested that in GVHD, the presence of both self antigens and donor-derived Th cells, reactive against recipient MHC class II antigens, provides a combination of stimuli that selectively drives autoreactive B cell differentiation (6).The experiments described below were designed to determine if a superantigen bridge between Th cells and B cells would result in an analogous form ofpolyclonal human B cell activation . MAM was selected for study in view ofthe chronic inflammatory arthritis that M. arthritidis induces in rodents (7), and scattered reports of M. arthritidis cultured from the bone marrow of patients with SLE (8).