K. Landsteiner

An Agglutinable Factor in Human Blood Recognized by Immune Sera for Rhesus Blood

The capacity possessed by some rabbit immune sera produced with blood of Rhesus monkeys, of reacting with human bloods that contain the agglutinogen M has been reported previously. 1 , 2 Subsequently it has been found that another individual property of human blood (which may be designated as Rh) can be detected by certain of these sera. Upon exhaustion of such a serum with selected bloods, for instance OM, the absorbed serum still agglutinated the majority (39 out of 45) of other human bloods, independently of the group or the M, N type; moreover, reactions took place with bloods lacking the property P. An example of the reactions is given in Table I. The results are of some interest in that they suggest a way of finding individual properties in human blood, namely, with the aid of immune sera against the blood of animals. As an analogy may be cited the demonstration of differences in sheep erythrocytes with immune sera for human A blood 3 The reactions described, although of moderate intensity only, were obtained with immune sera produced at different times. Whether these observations may possibly lead to a method suitable for routine work is still under investigation.

Further Observations on Individual Differences of Human Blood.

In a previous communication 1 we described an agglutinable factor (M), independent of the blood groups and present in many but not in all human bloods. A somewhat higher incidence of M among colored than white individuals, as indicated by our first results, was confirmed by examination on a larger scale; among 902 white individuals 165 (18.3 per cent), and among 338 colored 95 (28.1 per cent) whose blood reacted negatively. A differentiation of bloods (in the same group) lacking or possessing the factor M, was possible also with dried specimens. The heredity of the property M was studied in more than 100 families. The results are in keeping with the assumption that M is inherited as a mendelian dominant. In the following table the families are arranged in 3 classes according to the presence or absence of the factor M in the parents, and its incidence among the offspring is given. It may be mentioned incidentally that the character 2 A 1 present in most bloods of group A, seems to be an inheritable quality. For, in our tests, if the property was not or slightly developed in the blood of both parents, this, as a rule, was also the case for the children. With a method similar to that employed for the detection of M, namely, suitable absorption of certain rabbit immune sera for human blood, two other agglutinable qualities were found which may be denoted as N and P. These reactions, further distinguishing individual human bloods, vary in intensity from very strong to weak or negative.

A New Agglutinable Factor Differentiating Individual Human Bloods.

By absorbing a number of anti-human blood immune sera from rabbits with the blood corpuscles of certain individuals regardless of the group, fluids were obtained from a few sera which give a sharp differentiation of individual human bloods within the common blood groups. Among 116 individuals selected from the four blood groups the distribution of the agglutinable factor (which may be designated as M) was as shown in Table II. This reaction is distinguished by its intensity from some others known to show individual differences within the groups, such as the reactions with cold agglutinins. This is a preliminary report.

Heredity of Variants of the Rh Type.

Summary Studies on 47 families with 133 children indicate that the major subtypes Rh1 and Rh2 are transmitted by means of corresponding allelic genes Rh1 and Rh2 which are both dominant over the gene rh; and, in addition, that gene Rh1 is dominant over Rh2. (A nomenclature for designating the subdivisions of the Rh positive type is proposed).

Observations on human isoagglutinins.

It has been known for a long time that there are certain exceptions to the common blood grouping. This was pointed out repeatedly by one of the writers 1 as well as by others. 2 It did not seem worth while to describe in detail some observations which we have made along this line since, at the present time, there is no practicable way of comparing such disconnected findings with similar ones of other workers and properly classifying them. In our studies we have found instances of weak agglutination of a serum type I with another blood of the same type and recorded similar reactions with specimens of types II and III. The reactions were chiefly observed at a temperature of about 15° C. or less. At higher temperatures they became gradually weaker and disappeared. Several times the clumps contained rouleaux. The observation that at lower temperature atypical agglutinin reactions may occur between human bloods and sera of different individuals has already been described by Bialosuknia and L. Hirzfeld. 3 While the conclusion of these authors that the reactions are different from isoagglutination is questionable, they certainly gave a description of the fact. Similar phenomena were described in detail by Guthrie and Pessel 4 Although it is possible that an agglutination of this kind would not occur at body temperature yet the slight difference in the constitution of the blood might be sufficient to produce undesirable effects. Regarding the practical application, the observations suggest the use of direct cross tests before transfusion (in addition to typing); furthermore, it seems advisable to make these tests at low temperatures since in that way slight reactions may appear. This deduction has previously been made by Gutlirie and Pessel. Difficulty may be experienced in distinguishing isoagglutination at low temperatures from autoagglutination since the latter is favored by decreasing temperature. 5

Skin Sensitization to a Simple Compound by Injections of Conjugates

In view of a recent paper 1 touching upon the subject, we wish to make a preliminary communication of a study under way for some time on the possibility of producing in animals skin sensitivity to drugs by immunizing with conjugates. We have in fact been able to render guinea pigs sensitive to superficial application of picryl chloride by intraperitoneal injections of a conjugate resulting from the treatment of guinea pig erythrocyte stromata with picryl chloride in alkaline solution, killed tubercle bacilli as in previous work having been injected beforehand. Since even minute quantities of the simple substance can sensitize under certain conditions and must be avoided, the chief concern in these experiments was to guard against the inclusion of unchanged picry! chloride in the injection material. This was carried out by adding an excess of glycine which removed any possible remainder of the substance through chemical combination, and by washing with aqueous alcohol. The large majority of animals treated in this way have shown upon subsequent testing with the simple substance typical reactions of the contact dermatitis type.

Note on Individual Differences in Human Blood.

1. Previously an agglutinable property of human blood designated as P was described, which could be demonstrated by the use of immune agglutinins from rabbits, 1 and it was mentioned 2 that reactions of a more or less similar specificity occur with a certain type of irregular human isoagglutinins. Inasmuch as these 2 reagents are not always easily available it seems worth mentioning that similar results can readily be obtained with various absorbed normal animal sera, e. g., those of horses, pigs and rabbits. In horse serum the agglutinins were found rather frequently. 2. An anti-dysentery immune serum from a goat was shown by Eisler 3 to contain agglutinins for human blood which can be absorbed by dysentery bacilli (Shiga). Using this serum, kindly supplied to us by Eisler, we were able to confirm fully these interesting observations. Furthermore we found that after absorbing the serum with cells of the sub-group A 1 the supernatant fluid agglutinated most cells of sub-group A 2 , O, and probably B, distinctly more strongly than those of sub-group A 1 , in a similar way as certain exceptional human sera. 4 Whether this is a singular occurrence or can be reproduced with other dysentery sera is being investigated.

On Individual Differences in Human Blood

A clear-cut differentiation of human blood, aside from the blood groups, could be made by means of special agglutinating immune sera. The observations point to the existence of several agglutinable factors for which no agglutinins are demonstrable in normal human sera. In view of the latter circumstance the results reported do not imply any change in the scheme of the four blood groups. The body of serological evidence leads to the inference of a high degree of biochemical differentiation among individuals.

Experiments on Sensitization of Guinea Pigs with Simple Chemical Compounds

The sensitization of guinea pigs with simple compounds such as primulin, 1 salvarsan, 2 and p-phenylene diamine 3 has been studied by several authors, but the development of the subject has been somewhat hindered by difficulties encountered in reproducing some of these experiments. 4 As already observed by Mayer, we found that guinea pigs rubbed repeatedly with salve containing p-phenylene diamine showed increased redness and scaling, in comparison to controls. Furthermore, increased sensitivity was obtained by the administration of a salve containing p-nitroso-dimethylaniline and by repeated intracutaneous injections of a saline solution of p-phenylene diamine, and some such effects were noticed, also, after injecting p-nitroso-dimethylaniline and p-amino-phenol, using approximately 1 mg. of the substances per injection. Following this, smaller doses of 0.02 mg. were injected on the back twice a week over a longer period, a method similar to that advocated by Kolle 5 for the production of hypersensitiveness to salvarsan, in animals. With this method definite sensitization effects were obtained, besides salvarsan, with other compounds such as p-phenylene-diamine and p-nitroso-dimethylaniline. Thus, on reinjection on the sides about 7 days after the last injection of a course consisting of up to 20 injections at one or 3 day intervals, a number of the guinea pigs showed pinkish, more or less elevated areas within 24 hours, averaging 3/4-1 cm. in diameter. Likewise, with 1:2:4 dinitro-chlorbenzene∗ distinctly increased reactions, noticeable even after less than 10 injections, were observed in most of the guinea pigs after daily intracutaneous injections of the substance. In a preliminary experiment with suberanilic-resorcine dye, 6 2 out of 6 animals treated in the manner described showed definite reactions, so that one may assume that sensitization is possible with this substance, also.

Experiments on Immunization with Haptens

Experiments were previously reported 1 confirming the findings of Gonzalez and Armangué that the antigenic properties of alcoholic organ extracts containing Forssmans substance are strikingly increased by adsorption to inorganic adsorbents such as kaolin. Additional evidence for this effect has been advanced by the above authors recently, 2 in particular by establishing that it can be observed with a variety of adsorbents. To study further the phenomenon discovered by Gonzalez and Armangué, experiments similar to those referred to were carried out with Forssman preparations which had been freed from a large portion of inactive material originally present, by methods described previously. 3 Of the two preparations used, one was easily soluble in water and the other was brought into water solution by the addition of phenol. It appeared that with these substances practically no antigenicity could be induced by adsorption to kaolin whereas on injecting rabbits with the same preparations mixed with pig serum, all of the animals reacted and the majority produced sera of good titer. These results suggest the possibility that substances other than the specifically reacting haptens have a part in the reactivation by non-antigenic adsorbents. In this connection it may be mentioned that alcoholic organ extracts (e. g., extracts of horse kidney) prepared by heating, which may be supposed to have a higher content in proteins, are by themselves more apt to induce the production of hemolysins than those prepared at ordinary temperature.∗ In experiments with adsorbed bacterial carbohydrates, also, the effect seems to vary with the degree of purity of the substances. Preparations of cholera carbohydrate giving little or no biuret reaction were inactive or only faintly antigenic upon adsorption to charcoal, while preparations giving a strong biuret reaction, though very slightly antigenic themselves (following treatment with alkali), were definitely active in inducing the formation of agglutinins and precipitins in rabbits, after adsorption to charcoal.