Analysis of T hybridomas prepared from a T cell clone with three specificities. Recognition of self + X and allo-H-2 determinants segregates from recognition of Mlsa determinants.

To see k information on T cell recognition of Mlsa determinants, hybridomas were prepared from a well-characterized F23.2+ (V beta 8.2+) T cell clone specific for three different ligands, i.e., 1) Mlsa determinants, 2) fowl gamma-globulin (F gamma G) plus self-H-2 (H-2d), and 3) allo-H-2, e.g., H-2p, determinants. Fusion of the clone to the BW5147 thymoma line produced a triple-reactive T hybridoma which generated two types of spontaneous variants. One type of variant (type I) lost Mlsa reactivity but retained reactivity to both F gamma G/H-2d and allo-H-2p. These variants, which were generated at high frequency, stained strongly with a mAb, A1.57, with idiotypic specificity for the TCR molecules of the parental clone. The second type of variant (type II) reacted to Mlsa determinants but showed no reactivity to F gamma G/H-2d or to allo-H-2p. These variants failed to express the A1.57 idiotypic determinants of the parent clone, but were F23.2+ (V beta 8.2+); nonequilibrium pH gradient electrophoresis analysis suggested that these hybrids expressed a mixed TCR heterodimer composed of the parental clone beta-chain and the BW5147 alpha-chain. Three aspects of the data are very difficult to accommodate with the view that Mlsa, F gamma G, and allo-H-2 determinants are all recognized via a common TCR molecule: 1) the independent (and frequent) segregation of Mlsa reactivity from F gamma G/H-2d and allo-H-2p reactivity, 2) the retention of Mlsa reactivity by the type II variants despite loss of the parental clone alpha-chain, and 3) the loss of Mlsa reactivity by the type I variants despite high expression of the A1.57+ TcR molecules derived from the parental clone. The data support a model in which Mlsa determinants are recognized by a separate T cell structure, which we envisage as a monomorphic accessory molecule unrelated to the TCR. Since the type II hybridoma variants invariably retained quantitatively normal TcR expression, the triggering phase of anti-Mlsa responses appears to be TCR dependent. The model we favor is that anti-Mlsa/Mlsa interaction increases TCR binding with Ia epitopes to above the threshold required for cell triggering. A key feature of this model is that Mlsa and Ia determinants are recognized as separate structures rather than as a complex.