Clonal anergy of B cells: a flexible, reversible, and quantitative concept.

t the conclusion of the 1986 Immunology Congress in Toronto, the late Georges K6hler presented a sum- mary lecture which was, unfortunately, poorly attended. It dealt with the impact of the new genetics on immunology, and, among other wise things, he said that transgenic tech- nology was set to revolutionize the way we studied immu- nologic tolerance (1). The greatest barrier to uncovering the details of what happened to anti-self cells in the repertoire was the heterogeneity oflymphocytes. Cells potentially re- active with a given self antigen were so rare that most at- tempts to study their fate retied on inferential rather than direct methods. Successful attempts to create mice trans- genic for both heavy (H) and tight (L) immunoglobutin (Ig) chains, or for the oe and [3 chains of the recently discovered T cell receptor (TCR) in theory, at least, had the potential to create monoclonal mice, i.e., animals in which all or at least the greater part of the B or T cell population consisted only of cells bearing the transgenically imposed specificity. Anti- idiotypic reagents could identify such cells, or, even more easily, simple enumeration of B or T cell numbers under various experimental circumstances could yield information about the fate of anti-self cells. How tight K6hler was; but it took some time to prove it. Burnet first articulated the concept of deletion as the key mechanism of tolerance, postulating that if an en- counter between a self antigen and a cell reactive with it occurred in early life, while the immune system was imma- ture, deletion rather than selection would be the end result (2). Lederberg refined this notion, placing the transition from paralyzability to inducibility at the level of each differentiating immunocyte, regardless of the age of the animal (3). When it first became possible to enumerate antigen-binding B cells by autoradiographic techniques, early results indeed favored deletion as a tolerance mechanism (4). For T cells, it was still necessary to rely on functional techniques to support repertoire purging as a tolerance mecha- nism as not even enumeration of antigen-specific precur- sors of effector T cells by in vitro cloning techniques could distinguish deletion from some form of effective non- lethal silencing (5). However, the latter concept had articu- late supporters on theoretical grounds (6). Careful work on antigen-specific B cells, involving both their enumeration and purification, showed that high doses of antigen could cause deletion, but tolerance could be achieved with much lower doses, this non-deletional phenomenon being termed anergy (7). Support for anergy among T lymphocytes also gradually emerged (8, 9). Enter transgenic mouse technology. As far as B cell tol- erance was concerned, the first results were disappointing. Antigen-transgenic mice were either non-tolerant or vari- ably tolerant (10, 11). However, antigen-transgenic animals were not the main game, as they left the investigator with the task of studying the minority of (unidentifiable) reactive lymphocytes. Things really took off when B (12, 13) or T (14) cell receptor transgenic mice were used. From the viewpoint of the present story (15), the critical model was an ingeniously devised double-transgenic strategy (12; re- viewed in 16, 17). Mice were rendered transgenic for mono- B cell IgM and IgD receptors with high affinity for hen egg lysozyme (HEL) and such mice were mated with mice transgenic for HEL itself. Further features of the model were a capacity to distinguish transgenic from endogenous antibody via an allotype marker; the use of various founder lines constitutively producing various amounts of soluble HEL (s-HEL); a capacity to raise low HEL levels to high ones through zinc feeding in mice where the metallothio- nein promoter formed part of the HEL transgenic con- struct; and comparison of the effects of s-HEL with a membrane-anchored form of the same antigen (m-HEL). Some key findings of this model were as follows, s-HEL at very low concentrations caused T cell tolerance but with no discernible effects on the B cell compartment. Higher concentrations also resulted in T cell tolerance but in addi- tion caused a non-deletional, anergy type of B cell tolerance, readily observed as the great majority of B cells carried the transgenic Ig. This B cell tolerance was accom- panied by a down-modulation of surface IgM and a marked shortening of the B cell's life-span (18). The transgenic B cells clustered chiefly in follicular areas of the spleen, espe- cially the mantle zone of the follicles. Only when the strongly cross-linking signal of transgenic m-HEL was in place did a deletional type of tolerance, representing elimination shortly after Ig expression (while the immature B cell was still in the bone marrow) become manifest. Still, the delayed deletion of the anergic cells at 3-4 d does raise the question of whether anergy in B cells is entirely discrete from clonal abortion (19). The plot thickens as we move to studies that seek to track the migration of transgenic B cells after transfer into various kinds of host mice (20). Anti-HEL B cells trans- 1953 j. Exp. Med. 9 The Rockefeller University Press 9 0022-1007/96/05/1953/04 $2.00 Volume 183 May 1996 1953-1956