Tissue Antigens (1978), 11,475-478 Published by Munksgaard, Copenhagen, Denmark No part may be reproduced by any process without written permission from the author(s)
Possible Associations between HLA Antigens and the Immune Responsiveness to Attenuated Rubella Vaccine Shunichi Kato, * Mikio Kimura,* Iwao Takakura,* Tatsuro Sakakibara,* Hiroo Inouye,* * and Kimiyoshi Tsuji,* * *Department of Pediatrics, Tokai University School of Medicine, Isehara, Kanagawa, and **Transplantation Immunology Center, Tokai University School of Medicine, Isehara, Kanagawa, Japan
We investigated possible associations between HLA antigens and the antibody response patterns during a clinical trial with a live attenuated rubella vaccine in 172 Japanese schoolgirls. Those vaccine recipients were divided into three groups, 42 low responders, 102 intermediate responders and 28 high responders, according to convalescent-phase hemagglutination-inhibiting antibody titers. High frequencies of HLA-A11 and HLA-B15 in high responders and HLA-Aw24 and HLA-BS in low responders were found. Among HLA-A antigens HLA-A11 had the highest geometric mean HI antibody titer and HLA-Aw24 had the lowest. Among HLA-B antigens HLA-B15 had the highest and HLA-B5 had the lowest. Strong linkage disequilibrium was found between HLA-A11 and HLA-B15 in high responders and between HLA-Aw24 and HLA-B5 in low responders. These results suggest that there may be a relationship between HLA antigens and specific immune response genes in man as well as in other vertebrates. Received forpublication 21 October, revised, accepted 18 November 1977
I t has been demonstrated in several animal species, including mice (McDevitt & Tyan 1968), rats (Armerding & Rajewsky 1970), guinea pigs (Ellman et al. 1970) and rhesus monkeys (Dorf e t al. 1974), that the recognition of specific antigens as immunogens is governed by the products of immune response genes (Ir genes) linked to the major histocompatibility complex. Although reports of Ir genes in man are scarce and controversial because of the
complexity of genetic background and the difficulty of conducting experimental models in man, several HLA associations with certain immune responses have been reported (Jersild et al. 1973, Marsh et al. 1973, Greenberg et al. 1975, Spencer et al. 1976, Ilonen et al. 1977). Recently we had the opportunity to study possible associations of HLA types and antibody response patterns during a clinical trial with a live attenuated rubella
476
KATO ET AL.
vaccine. In order to minimize factors influencing immune response other than HLA types, our materials were limited t o the same race, sex and age, immunization was done on the same day, tissue typing was performed using the same antisera and rubella antibody was measured at the same time, by the same technicians, using the same reagents on all sera obtained. Materials and Methods 182 seronegative, unrelated, normal, healthy, Japanese schoolgirls, aged 1 3 and 14 years, were immunized with a live attenuated rubella vaccine (‘‘TO3 36”strain) supplied by Takeda Pharmaceutical Company. In this age group 55% of the schoolgirls were found t o be seronegative for rubella. Blood specimens for serologic study and HLA typing were obtained before vaccination and 6 weeks later t o follow IgG response. During this period no known exposure to rubella cases was observed nor were other vaccinations given t o this study group. Ten girls were omitted from the study due t o intercurrent minor infections, although their distribution of antibody response did not differ from that in the study group. HLA typing was done by the microdroplet lymphocyte cytotoxicity test (Ray et al. 1973), using antisera provided from the Seventh International Histocompatibility Testing Workshop and local antisera of 26 defined HLA-A and -B specificities. Rubella hemagglutination-inhibition (HI) tests were performed by the microtiter technique (Sever 1962). Results and Discussion Geometric mean HI antibody titers to rubella in 172 recipients of “T0336”-
Table 1 Convalescent-phase Geometric Mean Rubella Hemagglutination-Inhibiting Antibody Titers in 172 Recipients of “T0336”-Strain Live, Attenuated Vaccine, Analyzed According to HLA types HLA types
No. of Recipients
Geometric Mean HI titers
0 81 1 94 32 28 30
-
Locus A HLA-A1 HLA-A2 HLA-A3 HLA-Aw24 HLA-A26 HLA-A11 HLA-Aw31
76 128 72 87 105 75
Locus B HLA-BS HLA-B7 HLA-B8 HLA-B12 HLA-B13 HLA-B14 HLA-B15 HLA-Bwl6 HLA-B17 HLA-B 1 8 HLA- BwZ 1 HLA-Bw22 HLA-B27 HLA-Bw35 HLA-Bw37 HLA- B40 HLA-Bw54
61 14 0 30
65 74 -
77
0
-
0
-
34 21 2 0 0
108 81 128
-
-
30
67
0
-
26 0 50 23
88
85 70
strain vaccine are presented according to HLA types in Table 1. Among HLA-A antigens HLA-A11 had the highest titer and HLA-Aw24 had the lowest; and the difference was significant ( P < 0.05 by Student’s t-test). Among HLA-B antigens HLA-B1S had the highest titer and HLA-BS had the lowest; the difference again proving significant ( P < 0.01 by Student’s t-test). However, if these P values are corrected by the number of pairwise combinations, the differences become insignificant. HLA-A3 and HLA-B17
ASSOCIATIONS BETWEEN HLA-A AND ANTIBODY RESPONSE
477
Table 2 HJ-A Phenotype Frequencies in Recipients of “TO336 ”-Strain Rubella Vaccine According t o Hemagglutination-Inhibiting Antibody Response Categories
HLA antigens
(1) Low Responders HI titers < 32 (N = 42)
(11) Intermediate Responders 64 < HI titers < 1 2 8 (N = 102)
(111) High Responders HI titers 3 256 (N = 28)
50.0 0.0 71.4’ 11.9 4.g2 14.3
47.1 1.o 49.0 21.6 19.6 18.6
42.9
17.9 21.4 17.9
57.1 7.1 23.8 4.8 9.s4 11.9
28.4 8.8 12.7 3.9 20.6 11.8 1.o 17.6 19.6 33.3 14.7
28.6 7.1 25.0 7.1 32.1 14.3 3.6 14.3 10.7 25.0 10.7
(IV) Total (N = 172)
Locus A HLA-A2 HLA-A3 HLA-Aw24 HLA-A26 HLA-A11 HLA-Aw3 1
0.0 50.0
47.1 0.6 54.7 18.6 16.3
17.4
Locus B HLA-BS HLA-B7 HLA-B12 HLA-B13 HLA-B15 HLA-Bwl6 HLA-B17 HLA-Bw22 HLA-Bw35 HLA-B40 HLA-Bw54
0.0
19.0 7.1 21.4 11.9
35.5 8.1 17.4 4.7 19.8 12.2 1.2 17.4 15.1 29.1 13.4
Comparisons between group I and I11 (Fisher’s exact test) I P = 5.91 X lo-’ P = 4.00 X P = 1.69 X lo-’ P = 1.99 X
were omitted from the comparison because of the small number of subjects. The HLA phenotype frequencies in vaccine recipients are presented according to HI antibody response categories in Table 2. 42 (24%) were considered “low responders” (HI titer < 3 2), 28 (16%) were “high responders” (HI titer Z 256), whereas the majority, 102 (59%) were intermediate (64 < HI titer < 128) in response. Frequencies of HLA-All and HLA-B15 were higher in high responders than in low responders (P = 4.00 x
and P = 1.99 x by Fisher’s exact test), whereas frequencies of HLA-Aw24 were higher in low and HLA-BS responders than in high responders ( P = 5.91 x and P = 1.69 x lo-* by Fisher’s exact test). But these differences become insignificant after correction by multiplying by the number of antigens studied. Strong linkage disequilibrium was found between HLA-Aw24 and HLA-B5 in low responders and between HLA-A11 and HLA-B15 in high responders. Delta-values
478
KATO ET AL.
between HLA-Aw24 and HLA-B5 were 8.7% in 4 2 low responders, 5.7% in 28 high responders, 3.0% in 102 intermediate responders and 3.8% in 609 random unrelated Japanese. Those between HLA-A11 and HLA-€315 were 2.6% in 28 high responders, -0.2% in 42 low responders, - 0.1% in 102 intermediate responders and -0.0% in 310 random unrelated Japanese. These results suggest that there might be a relationship between HLA antigens and specific immune response genes in man as well as in other vertebrates. The fact that HLA-B antigens showed higher associations than HLA-A antigens is compatible with, but does not confirm the idea that Ir genes may be located near HLA-D locus. Although the significance of these associations is not clarified in natural rubella infection, a similar mechanism must be involved in its immune response. Further studies should be conducted along this line, A ck nowledgments This work was supported in part by a Research Grant from the Ministry of Health and Welfare, Japan.
References Armerding, D. & Rajewsky, K. (1970) The genetic control of immunological responsiveness to lactic dehydrogenase (LDH) isoenzymes. In: Protides of the Biological Fluids, pp. 185-187. ed. Peeters, H. Proceedings of the 17th Colloquium, Bruges, 1969.
Dorf, M. E., Balner, H., de Groot. M. L. & Benacerraf, B. (1974) Histocompatiblitylinked immune response genes in the Rhesus monkey. Transplant. Proc. 6, 119-123.
Ellman, L., Green, I:, Martin, W.J. & Benacerraf, B. (1970) Linkage between the poly-L-lysine gene and the locus controlling the major histocompatibility antigens in strain 2 guinea pigs. Proc. nat. Acad. Sci. 66, 322-328. Greenberg, L. J., Gray, E. D. & Yunis, E. J. (1975) Association of HL-AS and immune responsiveness in vitro to streptococcal antigens.]. exp. Med. 141,935-943. Ilonen, J., Herva, E., Reunanen, M., Panelius, M., Meurman, O., Arstila, P. & Tilikainen, A. (1977) HLA antigens and antibody responses to measles and rubella viruses in multiple sclerosis. Acta Neuroi. Scand. 55, 299-309. Jersild, C., Ammitsbdll,, T., Clausen, J . & Fog, T. (1973) Association between HL-A antigens and measles antibody in multiple sclerosis. Lancet 1, 151-152. Marsh, D. C., Bias, W. B., Hsu, S. H. and Goodfriend, L. (1973) Association of the HL-A7 cross-reacting group with a specific reaginic antibody response in allergic man. Science 179,691-693.
McDevitt, H. 0. & Tyan, M. L. (1968) Genetic control of the antibody response in inbred mice. J. exp. Med. 128, 1-11. Ray, J. G. Jr., Hare, D. B. & Kayhoe, D. E. (1973) In: Mannual of Tissue Typing Technique, pp. 20-22. National Institute of Health, Maryland. Sever, J. L. (1962) Application of a microtechnique to viral serological investigations. J. lmmunol. 88,320-329. Spencer, M. J., Cherry, J. D. & Terasaki, P. I. (1976) HL-A antigens and antibody response after influenza A vaccination. New Engl. J . Med. 294,13-16.
Address: Shunichi Kato, M.D. Department of Pediatrics Tokai University School of Medicine Bohseidai Isehara Kanagawa Japan 259-11.
Possible Associations between HLA Antigens and the Immune Responsiveness to Attenuated Rubella Vaccine Shunichi Kato, * Mikio Kimura,* Iwao Takakura,* Tatsuro Sakakibara,* Hiroo Inouye,* * and Kimiyoshi Tsuji,* * *Department of Pediatrics, Tokai University School of Medicine, Isehara, Kanagawa, and **Transplantation Immunology Center, Tokai University School of Medicine, Isehara, Kanagawa, Japan
We investigated possible associations between HLA antigens and the antibody response patterns during a clinical trial with a live attenuated rubella vaccine in 172 Japanese schoolgirls. Those vaccine recipients were divided into three groups, 42 low responders, 102 intermediate responders and 28 high responders, according to convalescent-phase hemagglutination-inhibiting antibody titers. High frequencies of HLA-A11 and HLA-B15 in high responders and HLA-Aw24 and HLA-BS in low responders were found. Among HLA-A antigens HLA-A11 had the highest geometric mean HI antibody titer and HLA-Aw24 had the lowest. Among HLA-B antigens HLA-B15 had the highest and HLA-B5 had the lowest. Strong linkage disequilibrium was found between HLA-A11 and HLA-B15 in high responders and between HLA-Aw24 and HLA-B5 in low responders. These results suggest that there may be a relationship between HLA antigens and specific immune response genes in man as well as in other vertebrates. Received forpublication 21 October, revised, accepted 18 November 1977
I t has been demonstrated in several animal species, including mice (McDevitt & Tyan 1968), rats (Armerding & Rajewsky 1970), guinea pigs (Ellman et al. 1970) and rhesus monkeys (Dorf e t al. 1974), that the recognition of specific antigens as immunogens is governed by the products of immune response genes (Ir genes) linked to the major histocompatibility complex. Although reports of Ir genes in man are scarce and controversial because of the
complexity of genetic background and the difficulty of conducting experimental models in man, several HLA associations with certain immune responses have been reported (Jersild et al. 1973, Marsh et al. 1973, Greenberg et al. 1975, Spencer et al. 1976, Ilonen et al. 1977). Recently we had the opportunity to study possible associations of HLA types and antibody response patterns during a clinical trial with a live attenuated rubella
476
KATO ET AL.
vaccine. In order to minimize factors influencing immune response other than HLA types, our materials were limited t o the same race, sex and age, immunization was done on the same day, tissue typing was performed using the same antisera and rubella antibody was measured at the same time, by the same technicians, using the same reagents on all sera obtained. Materials and Methods 182 seronegative, unrelated, normal, healthy, Japanese schoolgirls, aged 1 3 and 14 years, were immunized with a live attenuated rubella vaccine (‘‘TO3 36”strain) supplied by Takeda Pharmaceutical Company. In this age group 55% of the schoolgirls were found t o be seronegative for rubella. Blood specimens for serologic study and HLA typing were obtained before vaccination and 6 weeks later t o follow IgG response. During this period no known exposure to rubella cases was observed nor were other vaccinations given t o this study group. Ten girls were omitted from the study due t o intercurrent minor infections, although their distribution of antibody response did not differ from that in the study group. HLA typing was done by the microdroplet lymphocyte cytotoxicity test (Ray et al. 1973), using antisera provided from the Seventh International Histocompatibility Testing Workshop and local antisera of 26 defined HLA-A and -B specificities. Rubella hemagglutination-inhibition (HI) tests were performed by the microtiter technique (Sever 1962). Results and Discussion Geometric mean HI antibody titers to rubella in 172 recipients of “T0336”-
Table 1 Convalescent-phase Geometric Mean Rubella Hemagglutination-Inhibiting Antibody Titers in 172 Recipients of “T0336”-Strain Live, Attenuated Vaccine, Analyzed According to HLA types HLA types
No. of Recipients
Geometric Mean HI titers
0 81 1 94 32 28 30
-
Locus A HLA-A1 HLA-A2 HLA-A3 HLA-Aw24 HLA-A26 HLA-A11 HLA-Aw31
76 128 72 87 105 75
Locus B HLA-BS HLA-B7 HLA-B8 HLA-B12 HLA-B13 HLA-B14 HLA-B15 HLA-Bwl6 HLA-B17 HLA-B 1 8 HLA- BwZ 1 HLA-Bw22 HLA-B27 HLA-Bw35 HLA-Bw37 HLA- B40 HLA-Bw54
61 14 0 30
65 74 -
77
0
-
0
-
34 21 2 0 0
108 81 128
-
-
30
67
0
-
26 0 50 23
88
85 70
strain vaccine are presented according to HLA types in Table 1. Among HLA-A antigens HLA-A11 had the highest titer and HLA-Aw24 had the lowest; and the difference was significant ( P < 0.05 by Student’s t-test). Among HLA-B antigens HLA-B1S had the highest titer and HLA-BS had the lowest; the difference again proving significant ( P < 0.01 by Student’s t-test). However, if these P values are corrected by the number of pairwise combinations, the differences become insignificant. HLA-A3 and HLA-B17
ASSOCIATIONS BETWEEN HLA-A AND ANTIBODY RESPONSE
477
Table 2 HJ-A Phenotype Frequencies in Recipients of “TO336 ”-Strain Rubella Vaccine According t o Hemagglutination-Inhibiting Antibody Response Categories
HLA antigens
(1) Low Responders HI titers < 32 (N = 42)
(11) Intermediate Responders 64 < HI titers < 1 2 8 (N = 102)
(111) High Responders HI titers 3 256 (N = 28)
50.0 0.0 71.4’ 11.9 4.g2 14.3
47.1 1.o 49.0 21.6 19.6 18.6
42.9
17.9 21.4 17.9
57.1 7.1 23.8 4.8 9.s4 11.9
28.4 8.8 12.7 3.9 20.6 11.8 1.o 17.6 19.6 33.3 14.7
28.6 7.1 25.0 7.1 32.1 14.3 3.6 14.3 10.7 25.0 10.7
(IV) Total (N = 172)
Locus A HLA-A2 HLA-A3 HLA-Aw24 HLA-A26 HLA-A11 HLA-Aw3 1
0.0 50.0
47.1 0.6 54.7 18.6 16.3
17.4
Locus B HLA-BS HLA-B7 HLA-B12 HLA-B13 HLA-B15 HLA-Bwl6 HLA-B17 HLA-Bw22 HLA-Bw35 HLA-B40 HLA-Bw54
0.0
19.0 7.1 21.4 11.9
35.5 8.1 17.4 4.7 19.8 12.2 1.2 17.4 15.1 29.1 13.4
Comparisons between group I and I11 (Fisher’s exact test) I P = 5.91 X lo-’ P = 4.00 X P = 1.69 X lo-’ P = 1.99 X
were omitted from the comparison because of the small number of subjects. The HLA phenotype frequencies in vaccine recipients are presented according to HI antibody response categories in Table 2. 42 (24%) were considered “low responders” (HI titer < 3 2), 28 (16%) were “high responders” (HI titer Z 256), whereas the majority, 102 (59%) were intermediate (64 < HI titer < 128) in response. Frequencies of HLA-All and HLA-B15 were higher in high responders than in low responders (P = 4.00 x
and P = 1.99 x by Fisher’s exact test), whereas frequencies of HLA-Aw24 were higher in low and HLA-BS responders than in high responders ( P = 5.91 x and P = 1.69 x lo-* by Fisher’s exact test). But these differences become insignificant after correction by multiplying by the number of antigens studied. Strong linkage disequilibrium was found between HLA-Aw24 and HLA-B5 in low responders and between HLA-A11 and HLA-B15 in high responders. Delta-values
478
KATO ET AL.
between HLA-Aw24 and HLA-B5 were 8.7% in 4 2 low responders, 5.7% in 28 high responders, 3.0% in 102 intermediate responders and 3.8% in 609 random unrelated Japanese. Those between HLA-A11 and HLA-€315 were 2.6% in 28 high responders, -0.2% in 42 low responders, - 0.1% in 102 intermediate responders and -0.0% in 310 random unrelated Japanese. These results suggest that there might be a relationship between HLA antigens and specific immune response genes in man as well as in other vertebrates. The fact that HLA-B antigens showed higher associations than HLA-A antigens is compatible with, but does not confirm the idea that Ir genes may be located near HLA-D locus. Although the significance of these associations is not clarified in natural rubella infection, a similar mechanism must be involved in its immune response. Further studies should be conducted along this line, A ck nowledgments This work was supported in part by a Research Grant from the Ministry of Health and Welfare, Japan.
References Armerding, D. & Rajewsky, K. (1970) The genetic control of immunological responsiveness to lactic dehydrogenase (LDH) isoenzymes. In: Protides of the Biological Fluids, pp. 185-187. ed. Peeters, H. Proceedings of the 17th Colloquium, Bruges, 1969.
Dorf, M. E., Balner, H., de Groot. M. L. & Benacerraf, B. (1974) Histocompatiblitylinked immune response genes in the Rhesus monkey. Transplant. Proc. 6, 119-123.
Ellman, L., Green, I:, Martin, W.J. & Benacerraf, B. (1970) Linkage between the poly-L-lysine gene and the locus controlling the major histocompatibility antigens in strain 2 guinea pigs. Proc. nat. Acad. Sci. 66, 322-328. Greenberg, L. J., Gray, E. D. & Yunis, E. J. (1975) Association of HL-AS and immune responsiveness in vitro to streptococcal antigens.]. exp. Med. 141,935-943. Ilonen, J., Herva, E., Reunanen, M., Panelius, M., Meurman, O., Arstila, P. & Tilikainen, A. (1977) HLA antigens and antibody responses to measles and rubella viruses in multiple sclerosis. Acta Neuroi. Scand. 55, 299-309. Jersild, C., Ammitsbdll,, T., Clausen, J . & Fog, T. (1973) Association between HL-A antigens and measles antibody in multiple sclerosis. Lancet 1, 151-152. Marsh, D. C., Bias, W. B., Hsu, S. H. and Goodfriend, L. (1973) Association of the HL-A7 cross-reacting group with a specific reaginic antibody response in allergic man. Science 179,691-693.
McDevitt, H. 0. & Tyan, M. L. (1968) Genetic control of the antibody response in inbred mice. J. exp. Med. 128, 1-11. Ray, J. G. Jr., Hare, D. B. & Kayhoe, D. E. (1973) In: Mannual of Tissue Typing Technique, pp. 20-22. National Institute of Health, Maryland. Sever, J. L. (1962) Application of a microtechnique to viral serological investigations. J. lmmunol. 88,320-329. Spencer, M. J., Cherry, J. D. & Terasaki, P. I. (1976) HL-A antigens and antibody response after influenza A vaccination. New Engl. J . Med. 294,13-16.
Address: Shunichi Kato, M.D. Department of Pediatrics Tokai University School of Medicine Bohseidai Isehara Kanagawa Japan 259-11.
