Current Eye Research

Volume I I supplement 1992, 81 -86

Genetic factors in susceptibility and resistance to experimental autoimmune uveoretinitis R.R.Caspi, C.-C.Chan, Y .Fujino, S.Oddo, F.Najafian, S.Bahmanyar, H.Heremans*, R.L.Wilderl and B .W iggert

Curr Eye Res Downloaded from informahealthcare.com by Mcgill University on 12/10/14 For personal use only.

National Eye Institute, 'National Institute of Arthritis and Musculoskeletal and Skin Diseases, NIH, Bethesda, USA and *Laboratory of Immunobiology, Rega Instituut, Katholieke Universiteit Leuwen, Belgium ABSTRACT Experimental autoimmune uveoretinitis (EAU) can be induced in susceptible strains of rats and mice by immunization with purified retinal antigens, and serves as a model for human uveitis. Because strong HLA associations have been noted in a number of human uveitic diseases, we investigated the role of major histocompatibility complex (MHC) vs. non-MHC genes in the control of susceptibility to ocular autoimmunity, using the mouse and the rat EAU models. It was shown that EAU expression in mice requires both a susceptible MHC haplotype and a "permissive" genetic background. MHC control of susceptibility was tentatively mapped to the I-A subregion in H-2k. I-Ek expression appeared to have an ameliorating effect on disease. Susceptible H-2 haplotypes exhibited highest disease scores on the B10 background, and disease was reduced, or even absent, on some other (nonpermissive) backgrounds. Factors which may determine "permissiveness" or "nonpermissiveness" of a particular genetic background, as studied in mice and rats, may include diverse genetic mechanisms spanning regulation of cytokines, hormones, vascular effects and the T cell repertoire. Taken together, the data suggest that, in individuals susceptible to uveitis by virtue of their MHC, the final expression of disease will be determined by the genetic background. INTRODUCTION Experimental autoimmune uveoretinitis (EAU), can be induced in a variety of rodent species as well as in primates, by immunization with purified antigens from the retina (for reviews see (1-3). The mammalian retina includes a number of uveitogenic proteins. The best characterized are interphotoreceptor retinoid-binding protein (IRBP), the soluble retinal antigen (SAg) and opsin (2), but additional, as yet uncharacterized uveitogens may exist. The uveitogenicity of these proteins varies in different species, eg. IRBP is a more potent uveitogen than SAg for mice, while the reverse is true in guinea pigs, and both proteins are strongly uveitogenic in the Lewis rat . The diseases induced by all the uveitogenic proteins share essential characteristics, such as similar pathology and cellular mechanisms (2,3). EAU in animals serves as a model for

human noninfectious posterior uveitic diseases of unknown origin, which are considered to have an immune-mediated or autoimmune etiology. The putative antigens which may be involved in human uveitis have not yet been characterized. Some uveitis patients have elevated lymphocyte responses to SAg (4), but it is not clear to what extent these represent primary or secondary responses, or how they are related to immunopathogenesis. However, the fact that different retinal proteins induce essentially identical disease manifestations in a number of animal species, suggests that the findings in animals can be generalized to the human. Susceptibility to uveitic disease appears to be a genetically controlled trait. Strain-dependence of susceptibility to EAU is seen in all animal models (3,5-7). and strong HLA associations have been observed in a number of human uveitic diseases (8). Table 1 depicts the HLA associations seen in human posterior uveitis. Immunogenetic studies of uveitis in humans have been difficult, due to the complex genetics and outbred nature of the human species, and because of the overall low incidence of posterior uveitic disease in the population (as compared, eg. to arthritis or diabetes). We have therefore turned to the EAU model in inbred rodents, primarily the mouse, to try to understand the genetic control of ocular autoimmunity. Our results indicate a dual regulation of EAU by both MHC and non-MHC genes (6). A susceptible MHC is required for the uveitic response to occur, however, in individuals having a susceptible haplotype the final expression of disease is strongly modulated by the genetic background. MATERIALS AND METHODS

Animals Mice were obtained from sources previously described (6,7). Rats were purchased from Charles River, Indianapolis, IN. The animals were treated in accordance

Received on December 30. 1991: accepted on March 4, 1992

0 Oxford University Press

81

Curr Eye Res Downloaded from informahealthcare.com by Mcgill University on 12/10/14 For personal use only.

Current Eye Research with the NIH Guidelines and the ARVO Resolution for Use of Animals in Research. Antigens and adjuvants IRBP and SAg were prepared from bovine retinas as reported previously (9, 10). Complete Freund's adjuvant (CFA) was supplemented with M. tuberculosis swain 37RA to 2.5 mg/ml (Difco). B. pertussis toxin was from Sigma, St. Louis, MO. Immunization Mice were immunized with 100 pg of IRBP and 1 pg of pertussis toxin by the split-dose protocol, essentially as described (1 l), and killed at 4 to 5 weeks after immunization. Rats were immunized with 30 pg of SAg or the R16 peptide, representing the major pathogenic epitope of IRBP (aa 1177-1191) (12), and killed 2 to 3 weeks after immunization. Evaluation of disease and presentation of results Enucleated eyes were processed for histopathology as described (7). Incidence and severity of EAU were scored by an independent observer on a scale of 0 to 4, using a semiquantitative grading system (7, 13). Mean EAU scores were calculated as an average of all eyes from all animals in the group, and are shown f standard error of the mean (SE). Incidence is presented as the number of positive out of total animals in the group. RESULTS AND DISCUSSION Although EAU has been most widely studied in rats, the mouse is the obvious species of choice in the study of immunogenetics, because of the availability of hundreds of inbred, genetically defined strains. The MHC of the mouse (H-2) has been well characterized, and known H-2 haplotypes have been bred onto distinct genetic backgrounds. For this reason, most of the work reviewed here has been done in mice, using H-2 congenic strains (same MHC - different background, or same background different MHC) and intra-H-2 recombinant strains (differing from each other by known genes within the MHC). Expression of EAU requires a susceptible H-2 haplotvpe MHC dependence of susceptibility to EAU in mice is seen in selected strains having the same genetic background, but different H-2 haplotypes. Among the "independent" H-2 haplotypes immunized with IRBP, the only ones to develop EAU were b, d, k and r (6) (Table 2). Highest disease scores were seen in mice having the k and the r haplotypes (high

82

Table 1: HLA associations in posterior uveitic diseases Disease

Antigen*

Relative Risk

Sarcoidosis

HLA-DRw52 (0)

2-3

Behcet's disease

HLA-B51 (0)

4-6

Birdshot retinochoroidopathy

HLA-A29 (C)

49

Sympathetic ophthalmia

HLA-DR4 (0)

82.5

Vogt-Koyanagi-Harada disease

HLA-DRw53 (0)

74.5

*C - Caucasian, 0 - Oriental. The data were compiled partly from reference (8), and partly from S. Ohno, Yokohama City Univ., Sch. of Med., Japan, personal communication.

Table 2: Susceptibility of independent H-2 haplotypes to IRBP-induced EAU Strain B10 (C57bV10) B10.D2 BIO.M B1O.BR Bl0.Q B1O.RIII BIO.S B1O.PL B1O.SM

H-2

EAU score*

incidence**

b d f k q r

1.3 f 0.4 0.5 k 0.2 0 2.7 f 0.2 0 2.7 f 0.2 0 0 0

10113 9/33 0118 23/24 0117 21/21 0126 0116 0117

S U V

* Average of all eyes in the group f SE. ** Positive out of total animals responders), intermediate scores in b, and lowest scores in the d haplotype (low responder). This disparity in the level of responses might be due to differences in epitope recognition by the various H-2 types: the IRBP molecule is a complex antigen, with a number of pathogenic sites that differ in potency, which have been characterized in the rat system (12, 14-16). For this reason, it is also important to keep in mind that susceptibility could be a quantitative, rather than an absolute phenomenon, i.e., haplotypes seen here as nonsusceptible might display some ocular pathology if immunized with a sufficiently high dose of antigen.

Current Eye Research Table 3: Susceptibility is dependent on the I-A subregion of H-2 Haplotype STRAIN

(K A ED)*

EAU incidence

---ffff ssss

011 8 0126 718 518 214 718

Table 5: Effect of the genetic background on EAU expression strain

H-2

background

incidence

BIO.A A/J

a* a

B 10 A

26/27 13/23

2.6 f 0.2 1.2 k 0.3

B10.D2 Ba1b/c

d d

B 10 Balb

9/33 0123

0.5 f 0.2 0

BlO.BR AWJ

k k

B 10 AKR

23/24 0117

2.7 f 0.2 0

B1O.RIII LP.RIII

r r

B10 LP

2 1121 11/26

2.7 f 0.2 0.6 f 0.2

~~

Curr Eye Res Downloaded from informahealthcare.com by Mcgill University on 12/10/14 For personal use only.

B1O.M BIO.S BlO.BR BlO.TBR6 BlO.S(8R) BlO.TBFR8

kkkk Skkb kk-S sk-f

* Underlines signify nonsusceptible alleles. Dashes signify null alleles.

EAU score

~~

* H-2a is I-Ak (genotype Kk,Ak,Ek,Dd) Table 4: EAU scores are lower in I-Ek-expressing mice Strain

Haplotype (KAED)

I-E expr.

incidence

BlO.TBFR8 BlO.TBR6 BIO.A

sk-f skkb kkkd

-

18/19 13/16 10113

2.4f0.3 1.5 f0.4* 1.7 f 0 . 4

A.TBR2 A.TBR16 A.TBFR1 A

sk-b skkb skkb kkkd

-

16/16 13/16 418 6/12

1.7f0.3 0.9f0.2* 0.9 f 0.4 0.6 f0.2*

+ +

EAU score

.................................................................................................

+ + +

* Significantly different from the respective I-Erecombinant (p10.05), as analyzed by Snedcor and Cochran’s z-test for linear trend in proportions.

MHC control of susceptibility is dependent on class II genes of the I-A subregion MHC class I1 molecules restrict recognition of antigenic epitopes by preferentially binding and presenting selected fragments of processed antigens to T lymphocytes. If a susceptible H-2 is a function of epitope recognition, MHC control of susceptibility should map to class I1 genes. This has been confirmed using intra-H-2 recombinants based on the susceptible H-2k haplotype, in which individual genes within the H-2 were combined with alleles derived from nonsusceptible strains (6) (Table 3). The presence of a susceptible I-A subregion (functional equivalent of the human HLA-DR) was both necessary and sufficient to permit EAU expression. Although these results were obtained with recombinants based on H-2k, it is reasonable

to assume that MHC control of susceptibility in other H-2 types is similar. Expression of the I-& molecule may have an ameliorating effect on EAU Although expression of the I-Ek gene product (functional equivalent of the human HLA-DQ is not necessary for EAU induction (Table 3, strains BlO.S(8R) and BlO.TBFRS), I-E may have a modulating effect on disease. In a series of H-2 recombinants on two different backgrounds, strains expressing I-Ek had reduced EAU scores in comparison to strains having the same I-A, but which lack expression of the I-E gene product (6) (Table 4). This trend was more pronounced on the A background, where the strain of origin displays moderate EAU susceptibility. Two reasons, not mutually exclusive, can be proposed to explain this observation. 1) The I-Ek molecule may preferentially present epitopes recognized by T suppressor lymphocytes (17); 2) I-E-related deletion of certain T cell populations may be involved. Expression of the I-E molecule causes thymic deletion of lymphocytes whose T cell receptor (TCR) has affinity for I-E (18), namely Vp5, Vpll, Vp12 and Vpl7a. If IRBP-specific uveitogenic T lymphocytes make substantial use of these TCRs, it could explain the reduced susceptibility to EAU in I-Ekexpressing strains. Disease exmession in susceDtible haDlotvD - es is modulated by the genetic background The effect of non-MHC genes on the expression of uveitis was tested in a series of strains sharing a susceptible H-2 haplotype in the context of different genetic

83

Curr Eye Res Downloaded from informahealthcare.com by Mcgill University on 12/10/14 For personal use only.

Current Eye Research backgrounds (6). A major influence of background genes on susceptibility is indicated by the finding that substituting a different background for B10, resulted in a partial to complete reduction of EAU scores in all the tested haplotypes (Table 5). Thus, B10 can be regarded as a "permissive" background. Backgrounds such as LP or AKR are "nonpermissive", in that, despite the presence of a susceptible I-A , there is little or no EAU expression. Similar results were obtained in three strains of rats sharing the same class I1 subregion (RTIB1) but having different backgrounds, in EAU induced either by SAg or by the IRBP-derived R16 peptide. The Lewis and CAR backgrounds are permissive, while the F344 background is nonpermissive (Table 6). . . the "perrmsslrenessll The identification of non-MHC genes which might make a genetic background "permissive" or "nonpermissive" is obviously a complex task, because many parts of the genome outside of the MHC may be involved. However, a number of mechanisms can be proposed even at this early stage.

A known difference between the Lewis and F344 backgrounds is that the Lewis, but not the F344, has a defective hypothalamic-pituitary-adrenal (HPA) axis. This defect results in the inability of the Lewis strain to effectively upregulate serum corticosterone levels in response to the physiological stress of immunization (19).

Table 8: Systemic neutralization of IFNy upregulates EAU and can change the pattern of response in low susceptibility strains. Strain

Treatment

incidence

EAU score

NJ

Cont.* F-3**

10112 12112

0.7 f0.2 1.7 k 0.5

AKR

Cont. F- 3

0113

6/12

0.0 f 0.0 0.4 f 0.2

Cont. F- 3

8/13 13/13

0.6 f 0.2 3.4 f:0.4

DBN1

* Cont.: nonspecific control ascites.

** F-3: monoclonal anti-IFNy (ascites fluid).

Table 9: Kinetics of EAU in mast cell-deficient mice Table 6: Effect of the genetic background on EAU in rats SAg

Weeks*

incidence

EAU score

incidence

EAU score

~~

Lewis CAR F344

10111 loll0 9/10

2.7 f 0 . 4 3.8 fO.l 0.9f0.2

84

(W

1.5f0.3 (1m13)

0.9f0.4 (4W

2.0f0.4 (N)

ND

ND

0.7 fO.l (lalo)

1.7 f 0 . 4

3

5 0.1 f 0.04

~~

~

~

(W

(W

~~

* Weeks after immunization with IRBP.

Table 7: Treatment of Rldimmunized Lewis rats with physiological stress doses of dexamethasone downregulates EAU

none 10 30 100

1.4f0.4

1.3 f 0.2

1.2 f 0.2

* Not done

DEX (ps/kg)

w/+ 1.6f0.3

(W

w/wv 0.5 f 0.2 (3W

(1W

~

21/22 ND* 4/23

SV+

SVSld 0.5f0.4

R16 peptide 2

Strain

EAU score (incidence)

incidence

EAU score

22/24 20124 18/24 3/24

0.8 f 0.1 0.6 k 0.1 0.4 L 0.1 0.1 f 0.0

Table 10: The Mls allotype shapes the T cell repertoire in mice* Genotype

Background

TCRs deleted

mlsa mlsb

AKR, DBA B10, B6

Vp6, Vp8.1, Vp9 None

mlsc mlsd('a+c)

A CBNJ

VP3 Vp3, Vp6, Vp8.1, Vp9

* Data compiled from published literature

Curr Eye Res Downloaded from informahealthcare.com by Mcgill University on 12/10/14 For personal use only.

Current Eye Research Because corticosteroids are potent regulators of the immune response, the hypothesis is that the HPA axis defect underlies the susceptibility of the Lewis strain to EAU. This hypothesis is supported by the observation that treatment of R16-immunized Lewis rats with physiological stress doses of the synthetic corticosteroid dexamethasone downregulates EAU in a dose-dependent fashion (20) (Table 7). HPA axis regulation in other EAU-susceptible strains, such as the CAR, has not been studied. Consequently, it is not known whether the "permissiveness" of the CAR background might have a basis similar to, or different from, that of the Lewis. Lvmphokine regulation . We have recently reported that EAU in mice is exacerbated by systemic treatment with a monoclonal antibody to interferon-gamma (IFNy) (21). In the case of one low responder strain, this treatment resulted in a conversion to a high responder pattern, suggesting that in some genetic backgrounds IFNy-related mechanisms play a major role in conferring resistance to EAU (Table 8). Vascular mechanisms. EAU susceptibility of different rat strains was shown by Mochizuki et a1.(22) to correlate with their numbers of choroidal mast cells. This study was done in strains having different backgrounds, which differ in many other genetic characteristics. We therefore decided to reexamine this issue in mice. The W I W and SllSld mutant strains are profoundly deficient in tissue mast cells due to a bone marrow defect, but are otherwise genetically identical to the wild type. Both these strains had reduced susceptibility to IRBP-induced EAU (23), confirming a direct correlation between susceptibility to uveitis and the number of mast cells present in the eye (Table 9). T cell repertoire. Mice having different backgrounds vary in the expression of minor lymphocyte stimulating (MIS) antigens. Mls molecules are superantigens, and (as in the case of I-E) their expression causes thymic deletion of lymphocytes bearing TCRs with affinity for Mls, leaving detectable "holes" in the repertoire (18). The TCR elements deleted by different mouse strains depend on their MIS allotype (Table 10). It is interesting to note the correlation between the MIS type and the susceptibility to EAU of the different backgrounds. The B10, which does not delete any known Vp elements, is the most permissive background for EAU expression. The A, which deletes only Vp3, is moderately permissive, while the AKR, which deletes a number of TCRs, is nonpermissive (Table 5 and ref. 6).

This correlation provides suggestive evidence that MIS genes may be involved in the control of susceptibility to EAU. CONCLUSIONS The studies described above demonstrate a dual regulation of susceptibility to uveitis by both MHC class II and background genes. The finding that the genetic background can largely control whether or not a susceptible haplotype will develop EAU, might help to explain the incomplete penetrance of uveitis in susceptible human HLA genotypes, where only a minority of the individuals having a susceptible HLA type will in fact develop uveitic disease. Further research into the complex genetic mechanisms that govern the induction and expression of EAU in animal models may provide insights into the genetic factors that predispose to sight-threatening uveitic diseases in humans. CORRESPONDING AUTHOR Rachel R. Caspi, Ph.D., Laboratory of Immunology, National Eye Institute, NIH Building 10, Room 10N222, Bethesda, MD 20892. REFERENCES 1. Caspi, R.R. (1989) Basic mechanisms in immunemediated uveitic disease. In "Immunology of Eye Disease", (Ed. Lightman, S.L.). pp. 61-86. Kluwer Academic Publishers, Lancaster, UK. 2. Gery, I., Mochizuki, M. and Nussenblatt, R.B. (19 8 6) Retinal specific antigens and immunopathogenic processes they provoke. Prog. Retinal Res. 5,75-109. 3. Faure, J.P. (1980) Autoimmunity and the retina. Curr. Topics. Eye Res. 2,215-301. 4. Nussenblatt, R.B., Mittal, K.K., Ryan, S., Green, W.R. and Maumenee, A.E. (1982) Birdshot retinochoroidopathy associated with HLA-A29 antigen and immune responsiveness to retinal S-antigen. Am. J. Ophthalmol. 94,147-158. 5. Gery, I., Robinson, W.G., Jr., Shichi, H., El-Saied, M., Mochizuki, M., Nussenblatt, R.B. and Williams., R.M. (1984) Differences in susceptibility to experimental autoimmune uveitis among rats of various strains. In "Proceedings of the Third International Symposium on Immunology and Immunopathology of the Eye", (Eds. Chandler, J.W. and O'Conner, G.R.). pp. 242-245. Masson Publishing, NY. 6. Caspi, R.R., Grubbs, B.G., Chan, C.C., Chader, G.J. and Wiggert, B. Genetic control of susceptibility to experimental autoimmune uveoretinitis in the mouse model: Concomitant regulation by MHC and nonMHC genes. J. Immunol. (in press). 7. Caspi, R.R., Roberge, F.G., Chan, C.C., Wiggert, B., Chader, G.J., Rozenszajn, L.A., Lando, Z. and Nussenblatt, R.B. (1988) A new model of autoimmune disease. Experimental autoimmune uveoretinitis induced in mice with two different retinal antigens. J. Immunol. 140,1490-1495.

85

Curr Eye Res Downloaded from informahealthcare.com by Mcgill University on 12/10/14 For personal use only.

Current Eye Research 8. Nussenblatt, R.B. and Palestine, A.G. (1989) "Uveitis: Fundamentals and Clinical Practice". Year Book Medical Publishers, Inc. 9. Redmond, T.M., Wiggert, B., Robey, F.A., Nguyen, N.Y.,Lewis, M.S., Lee, L. and Chader, G.J. (1985) Isolation and characterization of monkey interphotoreceptor retinoid-binding protein, a unique extracellular matrix component of the retina. Biochemistry 24,787-793. 10. Dorey, C., Cozette, J. and Faure, J.P. (1982) A simple and rapid method for isolation of retinal S antigen. Ophthalmic Res. 14,249-255. 11. Caspi, R.R., Chan, C.C., Leake, W.C., Higuchi, M., Wiggert, B. and Chader, G.J. (1990) Experimental autoimmune uveoretinitis in mice. Induction by a single eliciting event and dependence on quantitative parameters of immunization. J. Autoimmun. 9.237246. 12. Sanui, H., Redmond, T.M., Kotake, S., Wiggert, B., Hu, L.H., Margalit, H., Berzofsky, J.A., Chader, G.J. and Gery, I. (1989) Identification of an immunodominant and highly immunopathogenic determinant in the retinal interphotoreceptor retinoid-binding protein (IRBP). J. EXP.Med. 169, 1947-1960. 13. Roberge. F.G., Lorberboum, G.H., Le Hoang, P., de Smet, M., Chan, C.C., Fitzgerald, D. and Pastan, I. (1 989) Selective immunosuppression of activated T cells with the chimeric toxin IL-2-PE40. Inhibition of experimental autoimmune uveoretinitis. J. Immunol. 3498-3502. 14. Donoso, L.A., Menyman, C.F., Sery, T., Sanders, R., Vrabec, T. and Fong, S.L. (1989) Human interstitial retinoid binding protein. A potent uveitopathogenic agent for the induction of experimental autoimmune uveitis. J. Immunol. M,79-83. 15. Kotake, S., Wiggert, B., Redmond, T.M., Borst, D.E., Nickerson, J.M., Margalit, H., Berzofsky, J.A., Chader, G.J. and Gery, I. (1990) Repeated determinants within the retinal interphotoreceptor retinoid-binding protein (IRBP): immunological properties of the repeats of an immunodominant determinant. Cell. Immunol. 126. 331-342. 16. Kotake, S., Redmond, T.M., Wiggert, B., Vistica, B., Sanui, H., Chader, G.J. and Gery, I. (1991) Unusual immunologic properties of the uveitogenic interphotoreceptor retinoid-binding protein-derived peptide R23. Invest. Ophthalmol. Vis. Sci. 2 , 2 0 5 8 2064. 17, Oliveira, D.B. and Mitchison, N.A. (1989) Immune suppression genes. Clin. Exp. Immunol. E. 167-177. 18. Pullen, A.M., Kappler, J.W. and Marrack, P. (1989) Tolerance to self antigens shapes the T-cell repertoire. 125-139. Immunol. Rev. 19. Sternberg, E.M., Hill, J.M., Chrousos, G.P., Kamilaris, T., Listwak, S.J., Gold, P.W. and Wilder, R.L. (1989) Inflammatory mediator-induced hypothalamicpituitary-adrenal axis activation is defective in streptococcal cell wall arthritis-susceptible Lewis rats. Proc Natl Acad Sci U S A 86,2374-2378. 20. Fujino, Y., Oddo, S., Najafian, F., Wilder, R.L., Chrousos, G.P., Sternberg, E., Chan, C.C., Nussenblatt, R.B. and Caspi, R.R. (1991) The role of steroid hormone in the regulation of EAU. Invest. hhthalmol. Vis. Sci. 32 (suppx), 232. 21, Caspi, R.R., C. Parsa, Chan, C.C., Grubbs, B.G., Bahmanyar, S., Heremans, H., A. Billiau and Wiggert,

m,

m,

86

B. (1991) Neutralization of endogenous Interferon-y exacerbates experimental autoimmune uveoretinitis in the mouse model. Invest. Ophthalmol. Vis. Sci. 22 (suppl.), 790. 22. Mochizuki, M., Kuwabara, T., Chan, C.C., Nussenblatt, R.B., Metcalfe, D.D. and Gery, I. (1984) An association between susceptibility to experimental autoimmune uveitis and choroidal mast cell numbers. J. Immunol. 1699-1701. 23. Bahmanyar, S., Grubbs, B.G., Chan, C.C., Li, Q., Wiggert, B., Nussenblatt, R.B. and Caspi, R.R. (1991) The role of mast cells in experimental autoimmune uveoretinitis. Invest. Ophthalmol. Vis. Sci. 32 (sup&), 934.

m,

Genetic factors in susceptibility and resistance to experimental autoimmune uveoretinitis.

Experimental autoimmune uveoretinitis (EAU) can be induced in susceptible stains of rats and mice by immunization with purified retinal antigens, and ...
506KB Sizes 0 Downloads 0 Views