The H-2 chart (Tables 1 and 2) provides information on the reaction pattern of allo-immune sera against gene products of the H-2K and H-2D regions of the major histocompatibility system (MHS) of the mouse.
The extent and arrangement of the H-2 chart is subject to continual change as new information is obtained from the study of inbred mouse strains having new H-2 haplotypes, and as new advances are made in the understanding of the genetic composition of the H-2 system. In this sense, the intrinsic supplement of the H-2 chart is the list of allo-immune sera (with the donor-recipient combinations and absorptions) used for the definition of H-2 antigens.
H-2K and H-2D antigens are operationally subdivided into private and public antigens. The public antigens are shared by two or more H-2 alleles. Public antigens illustrate the cross-reactivity pattern of the respective anti-H-2 sera. The molecular basis of the extensive cross-reactivity pattern of the MHS has not yet been sufficiently elucidated. Two possible factors that may account for cross-reactivity appear not to be mutually exclusive. In the first instance, an allo-immune serum is composed of a heterogeneous family of antibodies directed against components of a complex allo-antigen. Second, in addition to recognizing the specific target, a homogeneous antibody population can also recognize specificities that are different but closely related to those of the original immunogen. We use the term "cross-reactions" or "cross-reactivity pattern" of an anti-H-2 serum produced against one K- or D-region difference for both these possibilities.
The private H-2 antigens are restricted to one H-2K or H-2D allele product. They are defined by an anti-donor strain antibodies produced in a recipient or F1 hybrid and/or appropriately absorbed to reveal only anti-donor activity. Serotyping of wild mice has shown, however ( 7, 8), that some "monospecific" anti-private sera still exhibit cross-reactions with products of other H-2 alleles, if the panel of available targets is extended beyond those defining the present boundaries of the H-2 chart. Absorption experiments with cells from positively reacting wild mice have disclosed the cross-reactivity pattern of "monospecific" anti-private sera.
Most public antigens (coded by one K or D allele) are "parts" of one complex H-2 molecule. Therefore, it might be supposed that "privateness" is a reflection of the sum of a number of public specificities. Caution should, however, be exercised when attempts at generalization are made, and the need for this has been illustrated by the discovery of a separate locus for certain H-2D region "public" specificities ( 9, 10). More details and alternatives have been discussed by Snell et. al. ( 3), Démant ( 5), Shreffler and David ( 11), and Klein ( 6, 12, 13). The number of thoroughly analyzed H-2 alleles is about 10. Studies on wild mice ( 7, 12, 14, 15, 16, 17) have shown the polymorphism of H-2 loci to be much more extensive.
The impressive amount of H-2 serological work carried out during histocompatibility and immunogenetic studies has been inevitably and unavoidably performed by using serum pools. These serum pools were obtained by immunization of groups of recipients of one inbred strain (or F1 hybrids) with cells of the donor strain. It is probably tacitly assumed, because all serum producers are members of one highly inbred strain, that no substantial differences in the content of individual sera might be expected. The use, as a routine procedure in H-2 laboratories, of pools of sera from several recipients is usually dictated by the need to obtain a sufficient amount of a serum. Although considerable variations must have occurred in the different laboratories in the course of preparation of the individual serum pools, in general terms the procedures from different centers are roughly comparable. These can be characterized as being repeated intraperitoneal injections of different amounts of lymphoid cells, administered over differing time schedules, with the subsequent bleedings and pooling of positive sera to produce one batch of anti-H-2 serum. There are data about differences in the content of specificities from early and late pools during prolonged immunizations (e.g., the NIH catalogue of anti-H-2 sera), but to the best of our knowledge, there are no reports about differences in serum samples obtained from individual mice. We do not know whether this is because such difference were not found, or whether they had been noticed and then considered as trivial, or whether the possibility of divergence was not tested at all. The number of animals in one group used for the preparation of an anti-H-2 serum pool probably differs greatly from experiment to experiment, but groups of more than 20 recipients are frequently used.
Our first look at the diversity of the antibody content of individual H-2 sera (i.e., in sera from individual mice of one inbred strain) came from experiments in which these sera were tested on human cells. H-2 sera exert a strong cytotoxic effect on human cells, possibly due to cross-reactions with HLA antigens. The reaction pattern of H-2 sera on human cells depends on the mouse donor-recipient combination and the HLA phenotype of the human cells ( 18, 19, 20, 21). Thus, the wide variety of HLA alleles, presented by a panel of human lymphocytes, served as a first indicator that antibody variability was present in individual H-2 sera ( 22).
A short summary of this work will be given here. Anti-H-2 sera were produced by immunization of B10.A(4R) (H-2h4; kkbbbbb) recipients with lymphoid cells of H-2 congenicB10.A (H-2a; kkkdddd) donors. In this combination anti-Dd antibodies are primarily produced. The donor (Dd)-recipient (Db) H-2 antigen difference, according to the H-2 chart, enables the generation of anti-H-2.4 (private) and anti-H-2.3,13,35,36,41,42,43,44,49 (public) antibodies. Instead of one serum pool from several immunized recipients, the serum samples of each individual mouse were tested on a panel of about 100 HLA-typed human peripheral blood lymphocytes. It was found (for summary see Table 1) that individual mouse sera differed both in strength and specificity. A number of sera were negative, whilst others reacted in various titers on human lymphocytes. The reaction pattern of the positively reacting individual mouse sera was correlated with the presence of certain HLA antigens. Significant correlations were found between the high cytotoxicity scoring grade (CSG) and the presence of HLA-B12, B17, B27, and B37. (all these antigens are included in the Bw4/4a supertypic specificity.) Positive correlations were also found with some HLA-A locus antigens, and this was most pronounced for A11 and Aw31. However, as in previous studies, none of the individual H-2 sera behaved as "monospecific" with regard to individual HLA antigens. It is presumed that this reaction pattern of the individual mouse sera is attributable to the presence of certain common structures shared by all these antigens; individual mice of one inbred strain might recognize, or produce, antibodies against different parts of the complex allo-antigenic difference.
During this study, it became obvious that a similar approach should also be applied to the mouse H-2 system, and preliminary results are given in this work.
Anti-H-2Dd sera from individual mice of one highly inbred strain were tested on a number of H-2 haplotypes of inbred strains which were further supplemented with a small number of wild mice. The wild mice were included in the investigation to extend the number of H-2 targets as an expedient in studying variation in the sera of individual mice. These results on mouse cells showed an analogous extent of variation in titer and specificity similar to that already observed on human cell targets.
These findings, if confirmed on further combinations, might have implications for the understanding of 1) the generation of antibody diversity during MHS allo-immunization, and 2) the target H-2 specificities that are responsible for the cytotoxic effect of H-2 sera. It has been hypothesized ( 23) and widely discussed that the MHS itself is involved in the generation of antibody diversity.
Mice. The mouse strains were maintained at our Institute. The breeding nuclei were obtained from Drs. J. Colombani (Paris), J. Klein (Dallas), P. Démant (Amsterdam), and J.P. Levy (Paris), and from The Jackson Laboratory (Bar Harbor).
Serology. The cytotoxic action of H-2 sera on mouse lymph node cells ( 24) was tested by the microlymphocytotoxicity test ( 25), which was regularly performed in serial geometrical dilutions (for details, see reference 15). Absorbed rabbit serum was used as a source of complement. Serum dilutions and cell suspensions were prepared in RPMI + 10% FCS medium. The reaction in each well was recorded as -, +, ++, +++, ++++ according to the finding of not more than 15, 30, 50, 80 and more than 80% dead cells, respectively.
Immunizations. Donor spleen + thymus + lymph node cells (in doses of one donor per ten recipients) were injected intraperitoneally into female mice 4 (+/- 1) months old. In spite of possible individual variations (mostly due to differences in spleen weight) in the total number of lymphoid cells received from individual donors, no effort was made to equalize the number of cells injected each week. However, each mouse received the same aliquot from one common pool of the prepared lymphoid cell suspension. The immunizations were performed at weekly intervals with a one-week intermission after every fourth injection. Serum samples from individual mice were collected every week 3-4 days after injection. Samples obtained after the first injection were labeled A, after the second injection B, and so on. Most work was performed with serum samples of the immunization week T (i.e., after a series of 19 injections). When it was decided to terminate a given donor-recipient immunization combination, the recipients were bled 3, 7, and 9 days after the last immunization, and the three samples of sera (from one individual mouse) were pooled and designated as being the last sample of the immunization (with a letter of the last immunization week).
Wild mice. Wild mice nos. 1, 2, 9, and 10 were captured at different localities in or near Amsterdam, mice nos. 3-8 were captured in South Holland in one building complex.
The anti-H-2 sera were raised by immunization of B10.A (4R) (H-2h4; kkbbbbb) recipients with B10.A (H-2a; kkkdddd) donor cells (serum no. CLS-3). Serum samples of bleeds S or T from 38 individual recipients were tested for their cytotoxicity on lymph node cells of inbred mouse strains representative of the well-defined H-2 haplotypes. The reaction of all sera was negative on B10.A(2R) (kkkdd.b) cells. This shows that only anti-H-2Dd antibodies were present. The data are summarized in Table 1.
For general information, the reaction strength and major HLA antigen associations of the individual sera are given in the top part of Table 1 (for details, see reference 22).
The following conclusions and questions emerge from the reaction pattern observed on mouse cells.
1. One serum was negative and three gave only very weak reactions with the cells of the donor strain. All other sera reacted with the donor strain cells in titers higher than 1:64 up to titers of 1:2000-4000. Eight sera, however, only reacted with the donor strain. Such sera represent monospecific anti-private (anti-H-2.4) sera. As will later be shown, this designation is only operational because some of the "monospecific" sera cross-reacted with certain wild mice.
2. Further sera exhibited a variable extent of cross-reactivity with other H-2 haplotype products. At the right hand end of Table 1 appear those sera that cross-reacted with all H-2 haplotypes having H-2 public specificities shared with the cells of the donor strain. The only exception to the expected cross-reactions were in the negative reaction of all sera with the H-2k haplotype. It is not clear why anti-Dd sera produced in Dd recipients did not react with the H-2k haplotype. Both Dd and Dk contain H-2.3. This H-2 public antigen was originally defined (by hemagglutination) with an anti-H-2d recipients ( 26). H-2.3 was later described as actually being a "family" of 3-like specificities ( 27). The family consists of H-2.3, 13, 35, 36, 41, 42, 43, 44 specificities. Only H-2.3 is present on H-2k. Demant et al. ( 27) noticed that the presence of H-2b in the recipient can block the formation of cytotoxic anti-H-2.3 antibodies. It was presumed that the presence of H-2.35 and H-2.36 of H-2b(Kb) are responsible for this situation. However, in our combination, only Db (H-2.35, 36 negative) was present in the recipients. If we admit that the sera do not contain cytotoxic anti-3 antibodies, it would become difficult to explain the reactions with H-2r and Kd alleles. We therefore propose the following explanation. H-2.3 of the Dd allele is a D-end antigen with a counterpart of a similar structure at the K-end of H-2d. H-2r has the H-2Dd-like characteristics of H-2.3. However, H-2.3 is primarily a K-end specificity on the H-2k haplotype. The formation of anti-K-end antibodies in our sera was blocked by the presence of identical Kk alleles in the donor and recipient. In other terms, we presume that the absence of anti-H-2.3 in our sera (blocked by the presence of Kk in the recipient) is responsible for the negative reactions with H-2k.
3. Between the two extremes (negative sera versus "strong" cross-reactive sera) individual sera exerted a variable reaction pattern when tested on the panel of H-2 haplotypes (q, s, r, r). Serum no. 46 only reacted strongly with Dd and H-2s. Sera nos. 6, 36, and 37 only reacted strongly with Dd and H-2q (although weaker reactions also occurred with other haplotypes). Sera nos. 24 and 17 gave stronger reactions on H-2p than on H-2q and H-2s targets. Many more sera cross-reacted with H-2p when compared with H-2r. No sera reacted strongly with H-2p and/or H-2r, which did not have concurrent reactivity against H-2q and H-2s.
There are clear-cut differences in the frequency of occurrence of different cross-reacting patterns. (See anti-q and anti-s versus anti-p and anti-r). Further experiments might elucidate the biological meaning of these differences.
Two interpretations emerge from these findings: 1) certain individual mice produce anti-H-2.4 (the anti-"private" antibody) with a variable cross-reactivity pattern for other specificities: and 2) individual sera represent different combinations of distinct antibody populations with variable predominance for some specificities. These predominant components (or cross-reactivity patterns) are characteristic for some individual sera, and correspond with public H-2 antigens H-2.13 (for H-2q), H-2.36 (for H-2s), H-2.35 (for H-2p) and H-2.? (for H-2r).
4. Cross-reaction of anti-Dd antibodies with K region products. a) Anti-H-2.35 and 36 antibodies can be responsible for cross-reactions of anti-Dd antibodies with H-2Kb products of the H-2b haplotype. A rather low number of sera exerted a strong cross-reaction of this type (21%). In one serum (no. 9), anti-Kb activity was the strongest. There was good accord for the requirement that anti-H-2s and anti-H-2p cross-reactivity be present in sera showing anti-H-2b positivity. However, not all sera that reacted strongly with H-2p and/or H-2r reacted with H-2b, Thus some as yet unknown requirement(s) must be fulfilled to achieve anti-K-end (Kb) cross-reactivity in individual anti-D-end sera.
b) An unexpected finding was the strong reactions of three sera with HTG cells (H-2g; ddddd.b). These reactions point to shared specificities of Dd with Kd allele products. According to the H-2 chart, H-2.3 is the only specificity that might be responsible for these reactions. However, as mentioned under point 2, all sera were negative with H-2k. Furthermore, the reactivity of at least two of the anti-Kd-positive sera was either negative or very weak with cells of other strains bearing H-2.3. Thus sera 3x-9 and 3-19 either detect unknown Kd allele products or they visualize the complex nature of H-2.3. The latter possibility seems to be more valid, because both DBA/1 (H-2q) and A.SW (H-2s) cells absorbed anti-HTG (Kd) activity from serum 3x-9.
Anti-H-2.3 sera have been reported as exerting very weak or inconsistent reactions on HTG cells (H-2g) ( 27). This observation might be due to the use of serum pools in which the content of anti-H-2g (anti-Kd) might have been diluted out by the high number of negative sera.
As mentioned before for sera cross-reacting with Kb, it seems in a similar sense that some unknown requirement(s) must be fulfilled for the achievement of anti-K-end (Kd) cross-reactivity in individual anti-D-end sera. Only three out of 38 sera of individual mice from one inbred strain had this capacity.
5. Although both the donor and recipient strains have the "whole" H-2.28 complex (H-2.27, 28, 29), we have tested for the presence of anti-H-2L locus antibodies ( 9, 10). This may occur if the donor and recipient strains differ in subtypes of the H-2.28 complex. Serotyping of BALB/c-H-2d and BALB/cH-2db cells, and absorption experiments, revealed no difference between these two cell types although it was concluded that the H-2.28 public specificity was absent from cells of the H-2db mutant ( 28).
These experiments were aimed at extending the available panel of H-2 haplotypes and thus to allow for a better resolution of the extent of diversity of the individual H-2 sera.
The same sera as those tested on inbred mouse strains and human cells were tested for cytotoxicity on lymph node cells of 10 wild mice. The data are summarized in Table 2 (the sera remain arranged in the same order as for Table 1). The following conclusions and questions emerged from the reaction pattern observed with wild mice cells. 1. None of the wild mice tested can be classified as H-2.4 or Dd positive. This is because none of them reacted positively with all sera containing anti-donor antibodies. 2. The extent of cross-reactivity of the sera on wild mice cells (strength and number of positive reactions) was similar to that found on cells from mice of an inbred strain. This is certainly true for sera grouped to the left ("narrow" sera) and for most of the sera to the right ("broad" sera) of Table 2. Sera nos. 25, 9 and 33, which reacted with all wild mice cells, were obviously noted as being the most polyvalent sera on cells of inbred strains. However, there are examples that do not exactly follow this tendency.
3. Some sera that behaved as monospecific anti-H-2.4 (sera nos. 3x-37 and 31) on inbred strains reacted with cells of certain wild mice. This shows that the monospecific nature of these sera was operational and the "hidden" polyvalency or cross-reactivity pattern only emerged when sera were tested on wild mice cells. This finding parallels previous data obtained with "monospecific" (pooled) anti-H-2-private antigen sera ( 7, 8).
4. The interesting finding obtained from these preliminary tests on wild mice cells was the actual extent of cross-reactivity (broadness) of some sera. While some sera were negative or only reactive with a restricted number of individuals (inbred or wild mice) a few sera reacted with a very high number of unrelated individuals with different H-2 haplotypes (sera nos. 25, 9 and 33 as extreme examples). Sera of this kind show that not only the extent of polymorphism, but the extent of cross-reactivity among allelic products is unique for the MHS. Individuals of one inbred strain differ up to the two extreme ends of this very great individual variability; some form cytotoxic antibodies reactive only with the donor strain and possibly a few other haplotypes, whereas others produce antibodies that react with a very high proportion of the polymorphic variants present in the given species. Although illustrated only in a preliminary approach, knowledge about the extent of these variations may be important for further studies on the mechanisms involved in the regulation of this kind of individual diversity in anti-MHS antibody production.
We conclude that the observed variations among individual sera were surprising as they have not yet been reported despite the great amount of H-2 serological work. (We cannot exclude similar notions in some work that we were unable to review.) In several instances where physiological factors involved in the regulation of anti-H-2 antibody formation were studied and in studies on genetic differences in anti-H-2 antibody production among inbred strain, the data were obtained by testing serum pools. In some studies, mean values from individual sera are shown (with or without values for S.D.). Furthermore, the respective sera were mostly tested on the donor strain only. As also shown by our data, in such a situation individual variations could be considered as being "exceptions" and thus escaped interest for immunobiological evaluation ( 29, 30, 31, 32).
Experiments are in progress to test further combinations by a similar approach. We have seen a similar kind of individual variations when B10.A(2R) recipients were immunized by B10.A donor cells. In this combination, the donor and the recipient differ only in the D region.
We should like to stress that the important point, inherent in the approach that we follow, is that a high number of individual sera should encounter a far larger variety of MHS polymorphic targets. It seems clear that the higher the number of possible targets, the greater the extent of the observed diversity. This could be put forward for consideration as a general rule when we attempt to characterize the extent of individual variations in antibody formation in the frame of MHS. The same sera which were used in this work were also tested for cytotoxicity activity on human cells which provide a much more extensive panel (variety) of target structures than the limited panel of available H-2 haplotypes. The individual mouse sera differed in strength, reaction frequency, and specificity (related to HLA antigens) ( 22). There was a good correlation between the broadness of individual sera on both mouse and human cells ( Figure 1). This is a further parameter that illustrates the extent of the observed differences of individual sera. While some sera were rather restricted when tested on the mouse panel and exerted a low reaction frequency on human cells, other sera reacted with the majority of mouse haplotype products and also with about 80% of human cells from unrelated individuals (more than 100 cells tested).
The pronounced differences in the reaction pattern of individual mouse sera are of interest for further studies of the antibody diversity that occurs during MHS allo-immunization. Probably, we are faced with an extensive, but hitherto not experienced, variability in the number of clones that became or remained specifically effective in the production of antibody to components of the H-2D region. Several studies show different aspects of individual variations in antibody response among individual mice of one inbred strain when these were immunized with heterologous antigens (bacterial, erythrocyte, hapten, etc.) ( 33, 34, 35, 36, 37, 38, 39). For estimates of the extent of antibody diversity in inbred mouse strains and the proportion of this potential that is being used by an individual mouse, see Kreth and Williamson ( 36) and Pink and Askonas ( 37). Some of the most restricted antisera can be predicted to be analogous to mono- or oligo-clonal antibody populations generated by techniques in vitro ( 40). These latter antisera are produced to unravel the complex nature of the immunological response to products of the MHS region. Experiments can now be designed to test the possible homogeneity of our restricted (with regard to target specificity) antibody populations, and to attempt to elucidate the role of biological factors ( 41) that contribute to the regulation in vivo of the observed antibody diversity during the course of MHS allo-antigenic immunization.
Finally, for operational and practical purposes, elucidation of the type of response occurring in an individual experimental animal might be useful before the setting up in vitro of a series of continuous lymphoid cell cultures (hybridomas) from any given MHS immunized individual.
We express our gratitude to Peter van Mourik for his excellent technical assistance and to Sietse Weide and William Molenaar for their help in the preparation of the anti-H-2 sera. We are grateful to Anja Maas and Jetty Gerritsen for their secretarial assistance.
1. Snell, G.D., and Stimpfling, J.H. (1966). In Biology of the Laboratory Mouse, 2nd ed. (E.L. Green, ed.), p. 457. McGraw-Hill Book Co., New York.
See also
MGI.
2. Snell, G.D., Hoecke, G., Amos, D.B., and Stimpfling, J.H. (1964). Transplantation 12: 777.
See also
PubMed.
3. Snell, G.D., Cherry, M., and Demant, P. (1973). Transplant. Rev. 15: 3.
See also
PubMed.
4. Shreffler, D.C., and Snell, G.D. (1969). Transplantation 8: 435.
5. Démant, P. (1973). Transplant. Rev. 15: 162.
6. Klein, J. (1975). Biology of the Mouse Histocompatibility-2 Complex. Springer Verlag, New York.
See also
MGI.
7. Micková, M., and Iványi, P. (1971). In Proceedings of the Symposium on Immunogenetics of the H-2 System (A. Lengerová and M. Vojtisková, eds.), p. 20. S. Karger AG, Basel.
8. Iványi, P., and Micková, M. (1972). Transplantation 14: 802.
See also
PubMed.
9. Démant, P., Snell, G.D., Hess, M., Lemonnier, F., Neauport-Sautes, C., and Kourilsky, F. (1975). J. Immunogenetics 2: 263.
10. Démant, P., Neauport-Sautes, C., and Joshowtiz, M. (1977). Tissue Antigens 10: 252.
11. Shreffler, D.C., and David, C.S. (1975). Adv. Immunol. 20: 125.
See also
MGI.
12. Klein, J. (1971). Nature 229: 635.
See also
PubMed.
13. Klein, J. (1973). Transplant. Proc. 5: 11.
See also
PubMed.
14. Iványi, P., Démant, P., Vojtisková, M., and Iványi, D. (1969). Transplant. Proc. 1: 365.
See also
PubMed.
15. Micková, M., and Iványi, P. (1976). Folia Biol. (Praha) 22: 169.
See also
PubMed.
16. Klein, J. (1970). Science 168: 1362.
See also
PubMed.
17. Zaleska-Rutczynska, Z., and Klein, J. (1977). J. Immunol. 119: 1903.
See also
PubMed.
18. Ivasková, E., Dausset, J., and Iványi, P. (1972). Folia Biol. (Praha) 18: 194.
See also
PubMed.
19. Ivasková, E., Iványi, P., Micková, M., and Démant, P. (1975). Tissue Antigens 6: 286.
20. Iványi, P., Pavljuková , H., and Ivasková, E. (1976). Transplantation 22: 612.
21. Iványi, P. (1978). Proc. Roy. Soc., @02: 117.
See also
PubMed.
22. Iványi, P., van den Berg-Loonen, E.M., and de Greeve, P. (1978). Tissue Antigens 12: 32.
See also
PubMed.
23. Jerne, N.K. (1971). Eur. J. Immunol. 1: 1.
See also
PubMed.
24. Gorer, P.A., and O'Gorman, P. (1966). Transpl. Bull. 3: 142.
25. Terasaki, P.I., and McClelland, J.D. (1964). Nature 204: 998.
See also
PubMed.
26. Hoecker, G., Counce, S., and Smith, P. (1954). Proc. Natl. Acad. Sci. USA 40: 1040.
27. Démant, P., Snell, G.D., and Cherry, M. (1971). Transplantation 11: 242.
See also
PubMed.
28. McKenzie, I.F.C, Morgan, G.M., Melvold, R.G., and Kohn, H.I. (1977). J. Immunogenetics 4: 333.
29. Stimpfling, J.H., and Durham, T. (1972). J. Immunol. 108: 947.
See also
PubMed.
30. Lilly, F., Graham, H., and Coley, R. (1973). Transplant. Proc. 5: 193.
See also
MGI.
31. Peters, K., Schreiber, B., Gleichmann, H., and Gleichman, E. (1974). J. Immunogenetics 1: 141.
32. Shinohara, N., Ricks, Hansen, T.D., and Sachs, D.H. (1977). J. Immunol. 119: 1732.
See also
PubMed.
33. Gershon, R.K., and Kondo, K. (1972). Immunology 23: 321.
See also
PubMed.
34. Silver, D.M.. McKenzie, I.F.C, and Winn, H.J. (1972). J. Exp. Med. 136: 1063.
See also
PubMed.
35. Briles, D.E., and Krause, R.M. (1972). J. Immunol. 109: 1311.
See also
PubMed.
36. Kreth, H.W., and Williamson, A.R. (1973). Eur. J. Immunol. 3: 141.
37. Pink, J.R.L., and Askonas, A. (1974). Eur. J. Immunol. 4: 426.
38. Eichmann, K. (1974). Eur. J. Immunol. 4: 296.
See also
PubMed.
39. Eichmann, K. (1975). J. Immunogenetics 2: 491.
40. Lemke, H., Hammerling, G.J., Hohmann, Ch., and Rajewsky, K. (1978). Nature 271: 249.
See also
PubMed.
41. Pilarski, L.M., and Cunningham, A.J. (1975). Eur. J. Immunol. 5: 10.
See also
PubMed.
42. Iványi, P. (1970). Curr. Top. Microbiol. Immunol. 53: 1.
See also
PubMed.