In the early 1950s, studies on the immune system were still dominated by immunochemistry and immunological theories focused primarily on the problem of Ab formation. Although the hypothesis that Ags could function as a template for the acquisition of a complementary Ab configuration (direct template theory) was experimentally unsupported, it had remained since 1930 a quite popular theory for describing Ab specificity (1).
In 1955 Niels Jerne published “The Natural-Selection Theory of Antibody Formation,” which proposed that the Ag entering the host selects the best fits among a vast array of pre-existing Abs and triggers their enhanced production (2). The theory insisted on the notion of natural Ab diversity and revived the selective concept present in Paul Ehrlich’s side-chain theory of Ab formation (3).
As early as 1941, F. M. Burnet was skeptical about instructive theories of Ab formation and was trying to formulate an alternate hypothesis that could accommodate immunological features not covered by the direct template theory, such as the “secondary type immunological response.” Although he expressed reservations about the possibility that Abs could stimulate their own replication after the capture of Ag-Ab complexes by phagocytic cells, Burnet became highly interested in Jerne’s point of view because “most immunological phenomena can be well described in terms of the theory” (4). In 1957, D. W. Talmage and F. M. Burnet simultaneously embraced the “natural-selection” feature of Jerne’s theory with the major modification that the unit of selection must be the cell bearing the Ig rather than the Ig itself (4, 5). Both suggested that the Ag selects cells with the capacity to express the adequate Ab and that only those cells expand and differentiate (with particular emphasis on the notion of cell clones in the case of Burnet’s formulation). The so-called cell/clonal selection theory of acquired immunity was born (6, 7).
The 1958 report by G. Nossal and J. Lederberg (8), reprinted in this issue of The Journal of Immunology, indeed represents nothing less than an attempt to challenge the clonal selection theory. That Ab-forming cells could individually produce Ig of a unique specificity was a major prediction of the new theory, and at the time there was no observation to suggest that it could be valid. The question was this: could one identify plasma cells with two specificities? If this were the case, the core of the 1957 clonal selection theory would have to be re-examined. The idea of the study was to immunize an animal with two distinct Ags and then determine at the single cell level whether an Ab-producing cell did or did not produce Abs against both Ags.
Nossal and Lederberg chose two Salmonella strains (Salmonella adelaide and Salmonella typhi) serologically distinguishable by their H flagellar determinant without detectable cross-reactivity. Rats were immunized subcutaneously with a 1:1 mixture of fixed S. adelaide and S. typhi and cell suspensions were prepared from draining lymph nodes. By taking advantage of micromanipulation techniques developed by microbiologists, single cells were isolated, carefully washed, and maintained in microdroplets under mineral oil for several hours to allow for a substantial Ab concentration to be reached. Motile S. adelaide or S. typhi were then introduced into microdroplets containing plasma cells, and the preparations were monitored by dark-field microscopy. The readout was the capacity of single cell microcultures to abrogate the motility of every single bacterium present in the sample due to the presence of Abs directed to the flagellar Ag. Among the 456 single cells examined in the motility assay, Nossal and Lederberg observed that 33 mounted reactivity against S. adelaide whereas 29 were reactive to S. typhi. Most remarkably, none of the single cell-containing microdroplets that immobilized S. adelaide could immobilize S. typhi and vice versa. In other words, at the single cell level none of the Ab-producing cells examined revealed signs of dual reactivity. Hence, the technical tour de force by Nossal and Lederberg yielded results that, far from challenging the clonal selection theory, in fact supported it.
An independent study that analyzed lymphoid organ sections from animals immunized with two different fluorescent Ags also failed to detect cells with double reactivity (9). In follow-up studies of Salmonella-based systems with up to four different Ags and distinct immunization approaches, Nossal and Makela analyzed up to 3,628 single cells and detected 41 cells (1.13%) with dual reactivity (10, 11). Along the same lines, it was found that lymph node cells from animals immunized with hapten-protein conjugates could make Abs against either the hapten or the carrier. But again, no cells with dual reactivity could be observed (12).
In contrast, after hyperimmunization of rabbits with two serologically unrelated bacteriophages, it was found that up to 15% of the single cells analyzed could display dual reactivity (13). The reasons that could account for such marked differences have not been fully clarified. Retrospectively, it appears plausible that the experimental procedure used by Attardi et al. might well have favored the detection of cells with incomplete allelic exclusion (13). It is remarkable that this possibility was actually discussed (in contemporary terms) by Lederberg himself in 1959 in his elaboration on the original clonal selection theory (14).
Thus, for the most part the data available in 1967 supported the hypothesis that plasma cells were essentially cells producing Abs of a single specificity. There is no doubt that this notion contributed greatly to the general acceptance of the clonal selection theory at the Cold Spring Harbor Symposium on Quantitative Biology that was held the same year. There is also no doubt that the pioneering study by Nossal and Lederberg was instrumental in this process, as it constituted the first step toward the formulation of what is now known as the “one cell-one antibody” rule (8). That Ag-reactive B cells were making Abs of a single type was further supported by the observation that the cognate Ag was capable of aggregating all surface Igs on individual cells (15). Soon afterward, Nossal and Pike showed that the specificity of a given Ab and that of the Ag receptor expressed on the surface of the cell secreting it are the same (16). Finally, in their 1975 publication Köhler and Milstein described a revolutionary methodology allowing for the generation of continuous cell lines producing mAbs (17). The latter achievement, already celebrated as a classic in this series (18), reasserted the proof of the clonal selection theory proclaimed in this month’s Pillars of Immunology article.
The authors have no financial conflict of interest.
↵1 Address correspondence and reprint requests to Dr. Christophe Viret, Institut National de la Santé et de la Recherche Médicale Unité 563, Département d’Immunologie, Centre Hospitalier Universitaire Purpan, BP 3028, 31024 Toulouse Cedex 3, France. E-mail address:
- Copyright © 2009 by The American Association of Immunologists, Inc.