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.

    210/328 (page 188)
    BACH AND BACH AB If AB AA BB AB AB AB AO ВО OO Antigen inheritance is shown in simplified example using ABO blood groupings. In each case, child can inherit only one allele from each parent; the two matings dia¬ grammed (many more combinations are possible) could between them produce all the six possible combinations of alleles. HL-A histocompatibility system is transmitted simi¬ larly but instead of only three alleles (A, B, O) involving only two antigens, there are more than 30, each involving a different combination of many different antigens. ing them against standardized anti¬ bodies. But the latter proved singu¬ larly difficult to obtain. A good primary source of such anti¬ bodies (to define the so phenotypic products, i.e., antigens) is the serum of a multiparous woman. Such an indi¬ vidual, as is well known, forms anti¬ bodies to antigens on the fetal blood cells that were inherited from the father; Rh disease is caused by the mother's formation of such antibodies to erythrocyte Rh antigen. Our con¬ cern, however, is with the leukocyte antigens, which have been shown to be implicated in transplant rejection. Clearly, the mother will form anti¬ bodies only against such leukocyte antigens as she herself does not al¬ ready possess, just as only an Rh- negative woman will form antibodies against the Rh-positive fetus. It will be equally clear, moreover, that unless the human leukocyte antibody system is far simpler than that of the mouse - and there is no reason to suppose this — the multiparous mother's serum will be multispecific, containing not one but several antibodies, those formed against several or all the leukocyte antigens that the fetus has inherited from the father but which the mother does not possess. To obtain a presumed monospecific serum — one containing antibodies to only a single antigen — the multispe¬ cific serum can then be reacted against a panel of leukocytes from a num¬ ber of unrelated individuals. Each individual's leukocytes will react with, and remove from the serum, any anti¬ bodies keyed to their own antigens, but the remaining antibodies will still show activity against appropriate anti¬ gens in other leukocytes. If this pro¬ cedure is repeated enough times with different panel members, all the anti¬ bodies that can react with their collec¬ tive leukocytes will have been re¬ moved. What remains will be a serum that cannot react against any member of the panel but will still react with leukocytes of the original immunizer (the fetus), i.e., with an antigen (or antigens) that it, but none of the panel members, possesses. The reasoning behind this can per¬ haps be clarified somewhat by a much simplified example. Let us suppose that the fetal leukocytes contain only three antigens that the mother does not possess, called A, B, and C. The mother's serum, then, will contain only three antibodies: a, b, and c. If the first leukocytes in the panel con¬ tain, say, antigens B, D, and F, they will remove b from the serum; a sec¬ ond panel member with antigens C, F, and G will extract c. The modified serum will then react with neither member of the panel, but will react with the A antigen in the fetal leuko¬ cytes, and will therefore be defined as a monospecific serum. In practice, of course, the number of antigens would be unknown but presumably much greater than in our example. In addition, there would be no way of assigning particular anti¬ gens to particular panel members (had there been, the whole rigmarole would be unnecessary). It will be clear, likewise, that even in our simpli¬ fied example it would have been per¬ fectly possible to obtain a mono¬ specific serum that actually con¬ tained antibodies h, k, 1, and m along with a — assuming that the fetus possessed the corresponding antigens but none of the other panel members, nor the mother, did. The monospe¬ cific sera obtained by this method, in other words, are defined in a purely operational sense and, moreover, only in terms of the particular panel em¬ ployed in a given case. There are other methods by which single antigens can be defined, but these are equally oper¬ ational and even more complicated. The initial fruits of these involved operational definitions can be im¬ agined with no great difficulty: con¬ fusion. At the Second International Histocompatibility Workshop, held in Ley den in 1965, some scores of anti- sera prepared in half a dozen labora¬ tories were tested against a panel of 45 individuals. No two sera gave identical results. Happily, however, there was enough similarity in some cases to indicate that different investi¬ gators were talking about more or less the same thing. With this encourage¬ ment, and through exchange of sera among laboratories, work during sub¬ sequent years has clarified the picture to the point where, in one way or an¬ other, more than a dozen antigens can 188