Beyond Clonal Selection and Network

Immunology came of age with the selective principles of antibody formation (Jerne 1955) and the clonal selection theory (Burnett 1957). The theory was clear and amenable to direct test, promoting considerable progress in cellular and molecular immunology. It posed the central question as to the genetic origin of antibody diversity, settled 20 years later when progress in molecular biology has allowed for the study of eucaryotic genes (Tonegawa 1983). It is as if everything has been solved, and good people start leaving the field. It seems to me, however, that we are just at the end of one particular period of development in the discipline. Thus, while we can today manipulate genes, engineer molecules and control cellular behaviors, we are quite disarmed when trying to manipulate the system. Surprisingly, perhaps, we continue to treat allergy as we did before IgE was known, we have no specific therapy for autoimmune diseases, we are unable to tolerize the recipient of an organ to the tissues of the donor, and we seem incompetent to derive vaccines to protect the larger part of the world population from parasite infections (Stewart & Coutinho 1989). There are at least two ways of dealing with this paradox. One, more often found, maintains that all we need is more information on details and to continue to invest in the description of genes, molecules and cells. Another, more provocatively, claims that those kinds of problems cannot be solved in the same framework which was appropriate to address other levels of organization and description. It is argued that we must move from molecular and cellular immunology to systemic immunology, for the good reason that properties such as tolerance and self-nonself distinctions are distributed, and cannot be "reduced" to isolated components. Systemic approaches would seem to have found their way with the idiotypic network theory (Jerne 1974) but we are forced to conclude, 15 years later, that the whole idea 'did not quite make it'. The expectations of many have not been