Immunologic tolerance: self-nonself discrimination versus costimulatory factors and second signals

An attempt is made to assess the total number of the theoretically possible antigenic epitopes, the number of lymphocyte clones and the number of epitopes on cell surface antibody of a clone may be able to recognize. We suggest that the structures of recognizing antibody sites of a human are far from random and derived from a limited number of structures adapted to the recognition of pathogenic agents. The probability of recognition for a random epitope-like structure is likely to be very small. Discriminative power of BCR and TCR binding sites, together with T-helper control, should be sufficient for prevention of peripheral autoimmune response, but second-signal-type controls should not be neglected as supplementary mechanisms to prevent such response.