A great deal of experimental evidence supports the phenomenon of immunological suppression. The molecular mechanisms to explain the phenomenology have, however, remained controversial. In this review, the data are reinterpreted in light of the recent advances in the understanding of T-cell subsets, the crossregulatory properties of lymphokines and the differential presentation capacities of different antigen-presenting cell types.

No problem is of greater fundamental importance to immunology than tolerance, yet no realm of immunology has less credibility than that of suppressor T (Ts) cells. Reasons are neither hard to discern nor, in fact, wholly unjustified. Merely contemplating one of the hypothetical conceptualizations of the suppressor cascade of the mouse (Fig. 1) is, to say the least, a daunting exercise. Four major difficulties are encountered. First, one is confronted by a multifarious array of T-cell players, that include both CD4+ and CD8+ T cells, and a plethora of soluble antigen-specific and nonspecific factors that comprise a functionally unique network. The cascade involves antigen-specific, IJ-restricted, CD4+ suppressor-inducers (T s 1), CD8+ antiidiotype-specific cells (Ts2), followed by CD8+ antigenspecific, effector cells (Ts3) whose suppression is not restricted by the major histocompatibility complex (MHC) L, 2. Some of these cells are reported to be uniquely capable of binding directly to immobilized antigen in the absence of MHC molecules. Connectivity in this cellular cascade is mediated by a series of soluble factors (TsFs): TsF1 is idiotypic, antigen-specific, and, curiously, immunoglobulin heavy chain variable region (Ig VH) restricted; TsF2 is anti-idiotypic and requires delivery by an armed macrophage; and TsF3 acts totally nonspecifically. TsF1 and TsF2 require antigen-presenting cells (APCs), but TsF3 does not. The second problem is the mystery of IJ, which in congenic mouse strains represents a genetic restriction element for suppression, mapping between E (x and EIB. L From DNA sequence data of that region of the MHC, no sufficiently large uncharted portion of chromosome 17 is available to encode a unique 1-J polypeptide 4. Nor, after prodigious effort, has there been any molecular characterization of an IJ molecule that discriminates between the appropriate congenic strains. Third, and perhaps even more devastating, are molecular studies of antigen specific, murine TT suppressor hybridomas. These studies revealed that half had deleted T-cell receptor (TCR) 6 genes and the remainder had unrearranged TCR (x or 13 genes, and hence would be incapable of producing a functional receptor protein 5. Fourth, of the soluble TsFs described as having extraordinary properties, for example specificity for antigen and IJ or Ig Vn, until recently no two had similar properties, and over a period of 20 years not one had been convincingly characterized at a molecular level.