Medical genetics / edited by Victor A. McKusick and Robert Claiborne.

Date:
[1973]
Catalogue details

Licence: Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)

Credit: Medical genetics / edited by Victor A. McKusick and Robert Claiborne. Source: Wellcome Collection.

    211/328 (page 189)
    THE GENETICS OF HISTOCOMPATIBILITY now be detected and identified by sera from a sufficient number of labora¬ tories. These antigens have been given official identifying numbers (hl-a1, H l-a2, etc.). The identity, however, is still a matter of operational definition. We emphasize this because it is altogether too easy to fall into the trap of equat¬ ing an antigen vñth some specific sub¬ stance, e.g., a particular protein mole¬ cule. To some degree this is probably true, but how true remains to be seen. It is a fact that proteins, e.g., bovine serum albumin, possess not one but several antigenic determinants. It is also easy to imagine two different pro¬ teins, each capable in theory of gener¬ ating a distinct antibody, that for one reason or another invariably occur to¬ gether in leukocytes (or at least in all the leukocytes that have thus far been typed). With present methods, there is no way of producing an anti¬ serum to either of such a pair singly. and therefore no way of distinguish¬ ing them; operationally, they consti¬ tute a single antigen. There are other problems. For all the progress that has been made in defining leukocyte antigens, it has now become obvious that we could not have defined them all; to begin with we could only test for the pres¬ ence of known antigens, and therefore had to be missing the unknown ones. Add to that our new-found knowl¬ edge that in addition to the serologi¬ cally defined antigens, there is an¬ other whole category of antigens that cannot be detected by currently used serologic methods, the lymphocyte-de¬ fined antigens, which are probably important in graft rejection. (Inci¬ dentally it should be noted that in using the term antigens for pheno- typic products that do not elicit a detectable antibody response, we may offend immunologic purists. Neverthe¬ less, it is our feeling that on a func¬ tional basis, the nomenclature is justi¬ fied.) Nor, for that matter, can we be sure that all leukocyte antigens ( either s d or L D ) play a role in transplant rejec¬ tions, or, if they do, whether they play equally important roles. These prob-* lems have stimulated further re¬ searches, some of which will be re¬ counted later on. Because of the present limitations of leukocyte typ¬ ing, it has proved of unequivocal prac¬ tical value largely in matching trans¬ plant donors and recipients among siblings. The evidence bearing on typ¬ ing in relationship to matching be¬ tween unrelated donor-recipient pairs is less definitive. More important are the genetic findings. Analysis of antigen patterns in families has established a strong presumption that all the leukocyte antigens are controlled by a single genetic system comparable to the M H с in the mouse. Confirmation of One Locus Two Loci Three Loci 100 80 ^ 60 ■M <V I- 40 I 30 I 20 in ^ 10 Ш 8 О fi so с О) 00 nJ 1.0 0.9 0.6 0.4 0.3 0.2 0.1 10 15 20 25 30 35 О 1 2 Number of Alleles 3 4 The number of hoi and alleles involved in control of reactions in MLC tests can be predicted by calculating the expected fre¬ quency of nonstimulation for siblings (color curves) and unre¬ lated persons (black curves), using different assumed numbers of loci and alleles. Horizontal bars show observed frequencies of nonreaction in MLb tests at 95% confidence levels (darkened areas show where expected and observed frequencies coincide). For each assumed number of loci, the observed data define a possible range in number of alleles but only in the case of a single locus do the ranges for siblings and unrelated persons overlap. 189