Clonal Selection

Burnet's response to receiving Talmage's paper was to write a "preliminary communication" titled "A Modification of Jerne's Theory of Antibody Production Using the Concept of Clonal Selection" (Burnet 1957). This paper cites Talmage, pays homage to his classification of previous theories, and explicitly acknowledges Talmage's priority in the idea that cells were the replicating elements responsible for the exponential expansion of antibody production. It claims priority for Burnet in viewing the processes involved from a clonal perspective. Burnet's reworking of Talmage's model can be presented in an abridged form (Burnet 1957):

Among [antibodies] are molecules that can correspond probably with varying degrees of precision to all, or virtually all, the antigenic determinants that occur in biological material other than that characteristic of the body itself. Each type of pattern is a specific product of a clone of [lymphocytes] and it is the essence ofthe hypothesis that each cell automatically has available on its surface representative reactive sites equivalent to those of the globulin they produce... It is assumed that when an antigen enters the blood or tissue fluids it will attach to the surface of any lymphocyte carrying reactive sites which correspond to one of its antigenic determinants... It is postulated that when antigen-[antibody] contact takes place on the surface of a lymphocyte the cell is activated to settle in an appropriate tissue... and there undergo proliferation to produce a variety of descendents. In this way, preferential proliferation will be initiated of all those clones whose reactive sites correspond to the antigenic determinants on the antigen used. The descendents will [be] capable of active liberation ofsoluble antibody and lymphocytes which can fulfil the same functions as the parental forms.

Burnet's preliminary communication was very brief - only two manuscript pages - but it rings with the excitement of a man who has suddenly gained a great insight. While Talmage presented the model as a logically sound solution to a much-pondered puzzle, Burnets text is alive to the explanatory power and corollaries of such a hypothesis. In particular, he could see how the theory might explain the operation of neonatal tolerance, simply and elegantly (Burnet 1957):

Its chief advantage over [the Burnet-Fenner theory] is its relevance to the nature of normal antibodies including the red cell isoagglutanins (i.e. "natural antibodies," such as those involved in ABO blood group antigen reactions) and the simpler interpretation of tolerance to potential antigens experienced in embryonic life... Despite the speculative character of much of the detail of this modification ofJerne's theory - which might be called the "clonal selection hypothesis" - it has so many implications calling for experimental enquiry that it has been thought justifiable to submit this preliminary account for publication.

Burnet explored the full implications of the clonal selection theory to self/non-self discrimination the following year in a series of lectures presented at Vander-bilt University in Tennessee; these were subsequently published by Cambridge University Press (Burnet 1959) as The Clonal Selection Theory of Acquired Immunity. This book, like those he published in 1941, 1946, and 1949, provided a con cise but comprehensive summary of current findings pertaining to the character and regulation of immunological responses. As in the past, he reviewed theories of antibody production, but in a broad outline borrowed from Talmage, he described the contributions by Ehrlich, Pauling, Jerne, and himself and Fenner. In his description of the clonal selection theory, he attributed contributions by Jerne, Ehlich, and Talmage (Burnet 1959):

The great contribution of ¬°erne's theory was that it drew attention to the theoretical possibility that the recognition of self from not-self could be achieved in another fashion than by the recognition of "self markers." As Talmage points out, Ehrlich's side-chain theory was in many ways the logical equivalent ofJerne's concept... [At the time] it seemed, and to most immunologists still seems, inconceivable that all types of antibody could be pre-existent in the normal complement of gamma globulins. Nevertheless, if Jerne is correct... this would be an effective and much more elegant way ofaccount-ingfor the differentiation of self from not-self.

The outstanding difficulty in acceptingJerne's theory is the claim that when a given type of natural antibody is brought to a cell by antigen, the cell then proceeds to make more natural antibody molecules of the same type... Talmage pointed out that it would be more satisfactory ifthe replicating elements essential to any such theory were cellular in character...

Burnet then presented his own perspective (Burnet 1959, underlined):

Our own view is that any tenable form ofJerne's theory must involve the existence of multiple clones of globulin-producing cells, each responsible for one genetically determined type of antibody globulin... Clearly it would simplify matters a great deal ifthe antigen were in a position to react with natural antibody or a pattern equivalent thereto on the surface ofthe cell which produced it. This is the crux ofthe clonal selection hypothesis... Self-not-self recognition means simply that all those clones which would recognize (that is, produce antibodies against) a self component have been eliminated in embryonic life. All the rest are retained.

Burnet (1959) described in detail how the clonal selection model could explain a broad range of immunological phenomena, including immunological memory (expanded clone), anamnestic responses ("original antigenic sin" - lower-affinity binding of expanded numbers of cells dominating a response to a new antigen), the effects of adjuvants (prolonged release of antigen from a depot allowing per sistent expansion of a clone), mucosal immunity (preferential settling of clones in sites of stimulation), natural antibodies (all clones spontaneously make small amounts of antibody), and autoimmunity (somatic mutation of clones occasionally leading to acquisition of self-reactivity).

Corollaries of the Clonal Selection Theory

Perhaps of greater importance than the explanatory power of the clonal selection theory were the corollaries of the theory that related to the ontogeny of the immune system (including the "one cell-one antibody" hypothesis), affinity maturation, and the existence of peripheral tolerance mechanisms.

The one cell-one antibody hypothesis was an axiom of the theory. If a single lymphocyte, or a clone of lymphocytes, were able to make a variety of antibodies, then the specificity of antibody produced could not be used to select cells for expansion or destruction. Burnet regarded this axiom as so important that Joshua Lederberg, who was visiting the laboratory at the time, and his student Gustav Nossal were immediately set to address the question by assessing the specificity of antibody produced by individual plasma cells isolated from rats immunized with both Salmonella typhi and Salmonella adelaide (Nossal and Lederberg 1958).

A second axiom was that a vast array of specificities must be represented on the surfaces of various lymphocytes prior to antigen exposure. Since such a variety of specificities seemed unlikely to be encoded in the germ line, Burnet postulated that they be generated by an intense period of somatic mutation of the antibody-encoding genes within lymphocyte precursors early in ontogeny. This "phase of differentiation or randomization" had to be relatively short-lived, for in order for the development of tolerance by clonal deletion to be effective, the specificity of the developing lymphocytes had to be relatively stable at the time it occurred. The period immediately following tolerance induction, which was around the time of birth in mice, was proposed by Burnet to be the "critical point of Medawar". At this time, antigen exposure no longer caused tolerance, but did not yet result in immunization. This point was followed by maturation, with the liberation of natural antibodies and the potential of antigen-induced antibody production.

The concept of affinity maturation arose from Burnets belief that all dividing cells are subject to replication errors. In the context of the clonal selection theory, this could result in a neutral, a beneficial, or a degenerative effect on the affinity of antibody binding. Those clones that by chance mutate to higher affinity should compete better for antigen and replication signals, eventually dominating the response (Burnet 1959):

The combination of frequent minor mutation and a highly effective selective process would rapidly improve the accuracy of the complementary relationship [ofantibody] to new antigenic determinant.

If the right mutation occurs, the cell in question is, in the presence ofthe antigenic determinant, given a major advantage in the struggle to produce a larger clone ofdescendants than its congeners. If our general hypothesis is correct, there is no escape from the picture of(lymphocytes) as an evolving population as subject to mutation and selective survival as any large animal population in nature.

The postulated existence of peripheral tolerance arose from a similar train of thought. Since replication errors were bound to occur at least occasionally in rapidly dividing clones of lymphocytes, a clone might mutate to a "forbidden" or deleted specificity. It was therefore necessary that a mechanism exist to inactivate such clones.

It should not escape the notice of the reader that each of these corollaries is testable. The general form of the model was easily adapted to incorporate the subsequent discovery of T cells (Miller 1961), MHC restriction (McDevitt 1968; Zinkernagel and Doherty 1974), the summing of activation signals (Forsdyke 1968; reviewed in Baxter and Hodgkin 2002), and the concept of active immunoregulation (Gershon et al. 1972). It seems that testing the relevance of Burnets corollaries has occupied a significant proportion of experimental immunol-ogists for much of the last 50 years.

Self Post-Burnet

In the 45 years since the clonal selection theory was proposed, almost every experimental finding in immunology either has been consistent with it or has confirmed it. There have been quibbles, of course. Silverstein and Rose (1997) point out that it is possible to induce tolerance in adults (although it is much easier to induce it prenatally) and that "there are no fundamental differences in mechanism between the acquisition of tolerance to autologous and heterologous antigens." It should be painfully obvious that Burnet was aware of this. It was this very fact that raised his awareness that a tolerance induction mechanism existed at all. But Silverstein and Rose (1997) go further. The main theme of their thesis is that "the immune system does not and cannot discriminate between intrinsically harmful and intrinsically harmless substances." This is an odd point of view. If we define "the immune system" as those organs, cells, and cell products involved in defense against infectionx), these components and their functions must be subject to evolutionary pressure - indeed, one would imagine that they are subject to considerable evolutionary pressure. As natural

1) Many published definitions of "immune system" define it in terms of self/non-self discrimination (e.g., Stedman's Medical Dictionary; Merriam-Webster Medical Dictionary; Langman and Cohn 1997). Such definitions were eschewed, as in this particular context the arguments by all participants would have become circular.

selection (Darwin's sort, not Burnet's) operates at the level of reproductive fitness and (from an evolutionary point of view) "harm" is something that reduces reproductive fitness, the ability of the immune system to differentiate "harm" from "harmless" would emerge sooner or later, even if it did not exist in the first place. In his criticism of the paper, Brent (1997) argued along similar lines.

Silverstein and Rose had a point to their piece, of course. They were provoked by the publication of three manuscripts (for an immediate reaction, see Silverstein 1996) announcing the possibility of inducing immune responses, rather than tolerance, by immunizing neonates (Ridge et al. 1996; Sarzotti et al. 1996; Forsthuber et al. 1996). It was not so much these papers themselves that wound them up, but a bizarrely gushing accompanying "news" piece by Elizabeth Pen-nisi, published in the same issue of the journal, which claimed the authors were "trying to topple one of immunology's seemingly most solid pillars" (Pen-nisi 1996). This interesting example of tabloid journalism prompted a spate of letters (and responses to the letters) as well as commentaries in other leading journals on the subject. All were critical, but all seem to have missed a critical point: Medawar did not publish a model of neonatal tolerance; he published a model of prenatal tolerance. His own findings support the possibility of immune competence of neonatal mice, depending on strain, antigen, and mode of administration. Burnet, who did not originally have the benefit of Medawafs data, covered his bets by claiming tolerance would be induced in the "embryonic period or immediately post-embryonic stages" (Burnet and Fenner 1949), but subsequently settled for "during the last stage of embryonic life" (Burnet 1956).

As Langman and Cohn (1997) pointed out, there is a period in embryonic life when self-antigens are present and antigen-specific lymphocytes are absent. As the first potential immune effectors develop, the effects of the induction of tolerance to antigens present in the primary lymphoid organs can be easily experimentally measured. Although, as with much of hematology, the processes clearly observed in ontogeny operate throughout productive life, Medawar's "critical point" remains critical because of the relative ease with which tolerance can be induced at an age when the periphery has not yet been seeded with potential effectors (Brent 1997).

0 0

Post a comment