Journal of Autoimmunity (1990) 3,247-255
Pathology of Experimental Autoimmune Uveoretinitis in Mice
Chi-Chao Ghan,* Rachel R. Caspi,* Ming Ni,* William C. Leake,* Barbara Wiggert,? Gerald J. Chader’f and Robert B. Nussenblatt* *Laboratory National
of Immunology,
TLaboratory
Eye Institute, National
of Retinal Cell and Molecular
Institutes of Health,
Bethesda,
MD,
Biology, USA
The histopathology and immunopathology of murine experimental autoimmune uveoretinitis (EAU) following active immunization with the interphotoreceptor retinoid-binding protein (IRBP) were studied. The methods used included conventional light microscopy and immunoperoxidase staining. Lesions were located mainly in the uvea and the retina and were characteristically focal. The prominent histopathologic findings in the retina were vasculitis, granuloma, retinal fold, focal serous detachment, and loss of photoreceptors. Granulomas, formation of Dalen-Fuchs nodules, inflammatory cellular infiltration and increase in the thickness of the choroid and ciliary body were frequent findings. Subretinal neovascularization occurred in 10% of the experimental animals. Mild to moderate inflammation was also noted in the vitreous. The predominant infiltrating cells in the retinal and uveal granuloma and the Dalen-Fuchs nodules were macrophages. In contrast, the predominant infiltrating cell types in the vitreous were T helper/inducer lymphocytes. T suppressor/cytotoxic cells were rarely seen. Expression of Ia antigens on the ocular cells was confined to the immediate area of the inflammatory sites. The kinetics of histopathology showed two peaks at the 5th and 10th week after immunization, suggesting a relapsing course of the disease.
Introduction
Experimental autoimmune uveoretinitis (EAU) is an organ-specific, T-cell mediated autoimmune disease that has been used as a model for human intra-ocular inflammatory disorders (uveoretinitis) of unknown etiology [l-3]. Recently, it has been reported that EAU can be induced in mice by the retinal proteins, S-antigens or IRBP, an interphotoreceptor retinoid-binding protein [4,5]. Since the mouse is Correspondence to: Chi-Chao Chan, MD, National Eye Institute, Bldg. 10, Rm. lON206, Bethesda, MD 20892, USA. 247 0896-8411/90/030247
+ 09 $03.00/O
0 1990 Academic Press Limited
248 C.-C. Chan et al. an immunologically and genetically well-defined species, a murine EAU model provides definite advantages for understanding ocular autoimmune diseases with respect to basic immunological mechanisms and immunogenetics of the uveitogenic response. Previous results showed that the course of the disease was prolonged and that the ocular histopathologic findings were mainly located in the posterior pole with focal retinal and choroidal lesions, characterized by vasculitis and granuloma formation [4]. Histopathology in murine EAU was noted to be different from that in rats, but similar to that in some cases of human disease. In the present study, we followed the kinetics of EAU induced by interphotoreceptor retinoid binding protein (IRBP) in the BIO.A mouse strain, with respect to the dynamic changes in histopathology, inflammatory infiltrate and major histocompatibility complex (MHC) Class II-determinant expression on ocular resident cells. Materials and methods Female BIO.A mice (Jackson Laboratory, Bar Harbor, ME) at 2-5 months of age were housed in standard (non-specific pathogen-free) conditions. IRBP was prepared from bovine retinas by Con-A-Sepharose affinity chromatography and fast performance liquid chromatography as described previously [6]. For induction of EAU, the same protocol was used as reported previously [4]. Briefly, mice were pretreated 2 days before immunization with 20 mg/kg cyclophosphamide and were immunized with 50 pg of IRBP in a 1:l (V/V) emulsion with complete Freund’s adjuvent (CFA) containing 2.5 mg/ml of Mycobacterium tuberculosis strains H37RA, in a total volume of 0.1 ml divided between both hind footpads. Bordetella pertussis vaccine, lot 94 (Michigan Department of Public Health, Lansing, MI), containing pertussis organisms (10”) in a volume of 0.2 ml, was injected intraperitoneally at the same time. The animals were given a second dose of B. pertussis 3 days after primary immunization, followed by a secondary intramuscular immunization with 50 ng of IRBP in CFA one week after primary immunization. On several occasions, control groups of mice were immunized either with CFA/pertussis alone, without additional antigens other than the mycobacteria contained in CFA, or using bovine serum albumin (BSA) as irrelevant antigen. Different groups of animals were sacrificed at weekly intervals for 10 weeks after primary immunization (Figure 1). Freshly enucleated eyes were snap frozen in O.C.T. (Miles Scientific, Naperville, IL). Serial 4-6 pm frozen sections were prepared for immunohistochemical study by the avidin-biotin-peroxidase complex (ABC) method [7]. Adjacent frozen tissue sections were stained by standard hematoxylin and eosin. The intensity of EAU was scored in a masked fashion from 0 to 4 according to the histopathologic grading method developed previously for murine EAU [4]. Focal non-granulomatous, monocytic infiltration in the choroid, ciliary body and retina were scored as 0.5. Retinal perivascular infiltration and monocytic infiltration in the vitreous were scored as 1. Granuloma formation in the uvea and retina, the presence of occluded retinal vasculitis, along with photoreceptor folds, serous detachment and loss of photoreceptor were scored as 2. In addition, the formation of Dalen-Fuchs nodules (granuloma at the level of the retinal pigmented epithelium) and the development of subretinal neovascularization were scored as 3 and 4 according to the number and the size of the lesions.
Autoimmune
2
3
4
5
6
Weeks after
Figure 1. Histopathologic demonstrated a chronic,
7
8
uveoretinitis in mice
9
249
IO
immunization
kinetics of EAU induced by active immunization with IRBP in BIO.A mice possibly relapsing course. -O-, score; RI incidence.
Using the ABC method of tissue staining, biotin-conjugate anti-mouse IaK, H,Kk, Thyl,2, Lytl, Lyt2, and L,T, (Becton Dickinson, Mountain View, CA), Ml /70.15 (macrophage) and LR62B6B6.C9 (B cell) (Sera Lab., Westbury, NY) were applied to the tissue. After incubation, avidin-biotin and peroxidase complexes (Vector Laboratory, Burlingame, CA) were applied. The tissue sections subsequently were developed in diaminobenzidine. For quantitation of the labeled cells in each tissue section, the positive cells were counted and recorded. The results from each group of animals were averaged, compared and analyzed. Result
Only mice immunized with IRBP and not the control groups developed EAU. Kinetics Histopathological disease was observed as early as 2 weeks after immunization, with an incidence of 28.6% and an average grade of 1.3. Both the incidence and the average grade of histopathology varied through the lo-week examination period (Figure 1). By week 5 after immunization, the EAU incidence rose to 85.7% and the average grade was 2.71. Interestingly, both the EAU incidence and histopathological grade dropped to 0 in the 7th week after immunization. Thereafter, both EAU incidence and average grade slowly rose between the 8th and 10th week postimmunization. An incidence of 83.3% and an average grade of 2.5 were recorded at week 10 postimmunization. Histopathology The histopathologic features were essentially identical to those noted in earlier experiments [4]. The main pathologic features were located in the ocular posterior
250
C.-C.
Chan
et aZ.
Figure 2. Routine histopathology of the posterior segment. Left: A focal retinal granuloma (open arrow), retinal fold (arrowhead) and choroidal infiltration. Center: marked retinal vasculitis (arrow) and choroidal infiltration. Right: Focal infiltration between retinal pigment epithelium and choroid (Dalen-Fuchs nodule, asterisk) and choroidal infiltration (R = retina; C = choroid; hematoxylin and eosin, x 60).
Figure 3. Frozen sections of the posterior segment. T helper/inducer cells (arrows) were present in the vitreous (V) and retina(R). (L,T,, avidin-biotin-immunoperoxidase, x 150).
pole, as cited in our original report [4]. Minimum anterior involvement and mild vitritis were noted. The main pathologic lesions included inflammatory cell infiltration in the choroid and retina, and retinal damage. Most lesions were focal and discrete (Figure 2). On average, about 10% of the mice were found to have subretinal neovascularization, usually coinciding with the presence of a more severe choroidal inflammation (Figure 2). Immunopathology The inflammatory infiltrate consisted mainly of cells positive for IaK + , HZKk+, Thyl,2+, Lytl+, LsT,+ an d macrophage. The cells that bore surface markers for T helper lymphocytes (Thy1,2+, Lytl +, L3T4+) were located in the vitreous, uvea and
Autoimmune uveoretinitis in mice
Figure 4. Frozen retinal granuloma.
sections of the posterior R = retina, V =vitreous.
segment. Macrophages (arrow) were the main component (M1/70.15, avidin-biotin-immunoperoxidase, x 150).
251
in
perivascular regions of the retina (Figure 3). The cells that bore surface markers for macrophages (IaK+, HZKk+, macrophage) were mainly composed of granulomas and located in the uvea and the retina (Figure 4). Cells with the surface marker for T suppressor/cytotoxic lymphocytes (Lyt2+) or B lymphocytes were rarely seen. The composition of the inflammatory infiltrate in respect to the relative numbers of cells bearing the different markers did not change appreciably throughout the course of the disease. The ratios between T lymphocytes (Thy1,2+) and macrophages, and between T helper (L3T4) and T suppressor/cytotoxic (Lyt2+) lymphocytes were found to be 15 and 2O:l respectively, both in the early active (3rd to 5th weeks) and late, possible relapsing (9th to 10th weeks) stages. However, the ratios between T and B lymphocytes changed from 1O:l in the early stage to 6: 1 in the late stage. No influx of T suppressor/cytotoxic cells (Lyt2+) was observed in the intermediate (i.e., apparent remission) stage (6th to 8th weeks). The majority of inflammatory cells stained positively for Ia antigens. The ocular resident cells (the vascular endothelium and pigment epithelium) stained positively for Ia antigens (IaK+) only at and around the inflammatory sites. Resident cells where granuloma was present expressed MHC Class I and II antigens most strongly (Figure 5). No obvious enhancement of the staining intensity on ocular resident cells was noted as EAU scores became higher or the course of the disease progressed. Ia antigens were expressed only on the vascular endothelia, pericytes and cells in the perivascular infiltrates; detectable Ia expression coincided with the appearance of inflammatory cells. Expression of Ia antigens on ocular resident cells persisted during the active stages of disease, decreased in number of foci in parallel with the decrease in inflammatory lesions, and became undetectable in remission. Discussion
This study demonstrates for the first time the kinetics of EAU pathology in mice induced by immunization with the retinal antigen. Two weeks after immunization,
252 C.-C. Chan et al.
Figure 5. Frozen sections of the posterior segment. Most inflammatory cells (arrows) in the vitreous (V) and retina (R) stained positively for MHC Class I antigen (x 150). Inset: The retinal endothelia at the inflammatory sites stained positively for Ia antigen (x 240). [H,Kk, Iak (inset), avidin-biotinimmunoperoxidase].
histopathologic lesions began to develop with a chronic, focal, granulomatous inflammation in the retina and uvea. As the disease progressed, the number and size of these lesions increased. Both the incidence and severity scores were highest. By the 5th week the ocular inflammation was accompanied by photoreceptor degeneration and subretinal neovascularization. These two pathologic features are known to be dependent on the severity of inflammation [l, 21. By the 6th and 7th weeks after immunization, the disease appeared to go into remission, showing a decrease in histopathologic involvement and a drop in incidence. Begining at 8 weeks after primary immunization, the disease appeared to relapse as evidenced by a rise in the incidence as well as the number of lesions in the 9th and 10th weeks. Pathologically, the late-appearing lesions were similar to those of early EAU. The presence of a relapse, shown in this experiment by histopathology using different groups of mice, has subsequently been confirmed in a 3 month longitudinal follow-up of the same animals by weekly fundoscopic examination. The clinical data showed recurrent chorioretinal lesions in the same mouse (unpublished results). In a previous study [4] we observed slightly different kinetics, where disease appeared more severe with respect to histopathological grading and incidence. Animals were still positive at 7 weeks, although disease intensity was already subsiding. This difference is most likely due to the different pertussis vaccine lots used, since disease course and severity is directly related to the dosage of the pertussis adjuvant (Caspi et al., submitted). The animals were not followed beyond 7 weeks; however, we would expect the relapse to have been delayed in this case. Although EAU in the rat model is a non-relapsing disease, it is common to find focal recurrent lesions in chronic uveitis in humans. [8]. EAU in the Lewis rat is an acute inflammatory disease that occurs approximately 14 days after immunization, characterized by panuveioretinitis and lasting about 2 weeks. Complete photoreceptor destruction and retinal gliosis are found postinflammation [ 1,2]. EAU in
Autoimmune uveoretinitis in mice
253
the PVG rat [9] and Lister black hooded rat [lo] may present as milder, chronic disease with focal lesions. However, these alternative r&t models have not yet gained wide acceptance. When guinea-pigs are immunized with a low dose of uveitogenic antigen, a granulomatous infiltration in the uvea and destruction of photoreceptors lasting a few months have been described [ 1,2]. Repeated episodes of ocular inflammation by follow-up examination of fundus and fluorescein angiogram also have been seen in monkey EAU (Le Hoang, personal communication). In chronic murine experimental allergic encephalomyelitis (EAE), a relapse pattern was documented by clinical and histopathologic evaluations [ 11, 121. It was also noted that relapse in murine EAE is not a factor for increasing the degree of demyelination. In our study, both macrophages and T helper/inducer lymphocytes were found to comprise the majority of infiltrating inflammatory cells in the early and the putatively relapsing stages of murine EAU. Macrophages are the main component of the granulomatous infiltrate, and T helper/inducer cells are the main component of the non-granulomatous infiltrates inside the eye. Only occasional T suppressor/ cytotoxic cells are observed. Such cellular kinetics are quite different from those in Lewis rat EAU, where the inflammatory cellular population inside the eye changes markedly dur.ing a relatively short course (approximately 4 weeks). In the Lewis rat, T lymphocytes and not macrophages constitute the majority of the cells in the ocular infiltrates throughout the course of disease, with early predominant T helper/ inducers and late predominant T suppressor/cytotoxic cells [13,14]. Macrophages and polymorphonuclear leukocytes are seen at the beginning of ocular infiltration before the arrival of lymphoid cells in the early stage. It is conceivable that, in the mouse model, the persistent infiltrating macrophages and T helper/inducer cells may be conducive to the formation of ocular granulomas, while the paucity of T suppressor/cytotoxic cells may contribute to prolonging and possibly the relapsing course of murine EAU. It has been suggested that T suppressor lymphocytes may be needed for downregulation of the inflammatory response and modulation of the disease, both in rat EAU [ 13,141 and the end stage of some uveitides in humans [ 151. The finding that ocular resident cells located at or surrounding the inflammatory sites inside the eye bear Ia antigens is of particular interest, since similar findings have been reported in human uveitis [ 15,161 When the disease goes into remission, no inflammatory foci are found, and no Ia expression on ocular resident cells is seen. This observation is similar to that in murine EAE [17], where Ia antigens are induced on the vessels and astrocytes located at the inflammatory sites during the acute and relapsing stages. It has been suggested that local antigen presentation can induce damage to the blood-brain barrier [ 181. Therefore, at the site of Ia antigen expression, more inflammatory cells might penetrate and perpetuate the formation of a local ocular inflammation. The limited Ia antigen expression in murine EAU contrasts with widespread expression of MHC Class II antigens in rat EAU [ 19,201, and is probably related to the focal nature of the inflammatory cellular infiltrates. In conclusion, the murine EAU model offers differences in pathology and kinetics from EAU models in the Lewis rat, which today constitutes the best studied and most widely used rodent model of EAU. These differences include chronicity, focal nature, presence of Dalen-Fuchs nodules and subretinal neovascularization, relatively steady inflammatory cellular population, localized expression of Ia antigens, as well as a relapsing nature. In these respects murine EAU could be useful for the study
254 C.-C. Chan et al.
of mechanisms involved in some human uveitides such as idiopathic chronic uveitis and ocular autoimmune disorders for which no satisfactory rodent model has existed previously. Acknowledgement The authors thank Ms. Janet Edds for excellent secretarial assistance. References 1. Faure, J. I’. 1980. Autoimmunity and the retina. Curr. Top. Eye Res. 2: 215-302 2. Gery, I., M. Mochizuki, and R. B. Nussenblatt. 1986. Retinal specific antigens and immunopathogenic processes they provoke. Prog. Retinal Res. 5: 75-109 3. Nussenblatt, R. B., T. K. Kuwabara, F. de Monasterio, and W. B. Wacker. 1981. S-antigen uveitis in primates: A new model for human disease. Arch. Ophthalmol. 99: 1090-1092 4. Caspi, R. R., F. G. Roberge, C. C. Chan, B. Wiggert, G. J. Chader, L. A. Rozenszajn, Z. Lando, and R. B. Nussenblatt. 1988. A new model of autoimmune disease. Experimental autoimmune uveoretinitis induced in mice with two different retinal antigens.J. Zmmunol. 140: 1490-1495 5. Iwase, K., Y. Fujui, I. Nakashima, N. Kato, Y. Fujino, H. Kawashima, and M. Mochizuki. 1988. A new model for experimental autoimmune uveoretinitis (EAU) in the mouse. Invest. Ophthalmol. Vis. Sci. (suppl) 29: 39 6. Redmond, T., B. Wiggert, N. Nguyen, M. Lewis, L. Lee, and G. J. Chader. 1985. Isolation and characterization of monkey interphotoreceptor retinoid-binding protein, a unique extracellular matrix component of the retina. Biochemistry 24: 787-793 7. Hsu, S. M., L. Raine, and H. Franglor. 1981. The use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase technique. J. Histochem. Cytochem. 29: 577-580. and Clinical 8. Nussenblatt, R. B. and A. G. Palestine. 1988. In Uveitis: Fundamentals Practice. R. B. Nussenblatt and A. G. Palestine (eds). Year Book Medical Publishers, New York. 9. deKozak, Y., J. Sakai, B. Thillaye, and J. P. Faure. 1981. S antigen-induced experimental autoimmune uveo-retinitis in rats. Curr. Eye Res. 1: 327-340 10. Stanford, M. R., E. C. Brown, E. Kasp, E. M. Graham, M. D. Sanders, and D. C. Dumonde. 1987. Experimental posterior uveitis. I: A clinical, angiographic, and pathological study. Br. J. Ophthalmol. 71,585-592 11. Brown, A. M. and D. McFarlin. 1981. Relapsing experimental allergic encephalomyelitis in the SJL/ J mouse. Lab. Invest. 45: 278-284. 12. Tabiza, T. and K. Sakai. 1987. Demyleination induced by T cell lines and clones specific for myelin basic protein in mice. Lab. Invest. 56: 518-525 13. Chan, C. C., M. Mochizuki, R. B. Nussenblatt, A. G. Palestine, C. McAllister, I. Gery, and D. BenEzra. 1985. T lymphocyte subsets in experimental autoimmune uveitis. Clin. Zmmunol. Zmmunopathol.
35: 103-110.
14. Brown, E. C., E. Kasp, and D. C. Dumonde. 1989. Morphometric analysis of T lymphocyte compartmentation in experimental autoimmune uveoretinitis. Clin. Exp. Immunol. 77: 422-427 15. Chan, C. C., R. B. Nussenblatt, L. S. Fujikawa, A. G. Palestine, G. Stevens, L. M. Parver, M. W. Luckenback, and T. Kuwabara. 1986. Sympathetic ophthalmia: Immunopathologic findings. Ophthalmology 93: 690-695 16. Chan, C. C., R. P. Wetzig, A. G. Palestine, T. Kuwabara and R. B. Nussenblatt. 1987. Immunohistopathology of ocular sarcoidosis. Report of a case and discussion of immunopathogenesis. Arch. Ophthalmol. 105: 1398-1402 17. Sakai, K., T. Tabira, M. Endoh, and L. Steinman. 1986. Ia expression in chronic relapsing experimental allergic encephalomyelitis induced by long-term cultured T cell line in mice. Lab. Invest. 54: 345-352
Autoimmune uveoretinitis in mice
255
18. Traugott, U., C. S. Raine, and D. E. McFarline. 1985. Active experimental allergic encephalomyelitis in the mouse: immunopathology of the developing lesion. Cell. Immunol. 91: 240-254 19. Chan, C. C., J. J. Hooks, R. B. Nussenblatt, and B. Detrick. 1986. Expression of Ia antigen on retinal pigment epithelium in experimental autoimmune uveoretinitis. Curr. Eye Res. 5: 325-330 20. Fujikawa, L. S., C. C. Chan, C. McAllister, I. Gery, J. J. Hooks, B. Detrick and R. B. Nussenblatt. 1987. Retinal vascular endothelium expression fibronectin and class II histocompatibility complex antigens in experimental autoimmune uveitis. Cell. Zmmunol. 98: 139-150.
Pathology of Experimental Autoimmune Uveoretinitis in Mice
Chi-Chao Ghan,* Rachel R. Caspi,* Ming Ni,* William C. Leake,* Barbara Wiggert,? Gerald J. Chader’f and Robert B. Nussenblatt* *Laboratory National
of Immunology,
TLaboratory
Eye Institute, National
of Retinal Cell and Molecular
Institutes of Health,
Bethesda,
MD,
Biology, USA
The histopathology and immunopathology of murine experimental autoimmune uveoretinitis (EAU) following active immunization with the interphotoreceptor retinoid-binding protein (IRBP) were studied. The methods used included conventional light microscopy and immunoperoxidase staining. Lesions were located mainly in the uvea and the retina and were characteristically focal. The prominent histopathologic findings in the retina were vasculitis, granuloma, retinal fold, focal serous detachment, and loss of photoreceptors. Granulomas, formation of Dalen-Fuchs nodules, inflammatory cellular infiltration and increase in the thickness of the choroid and ciliary body were frequent findings. Subretinal neovascularization occurred in 10% of the experimental animals. Mild to moderate inflammation was also noted in the vitreous. The predominant infiltrating cells in the retinal and uveal granuloma and the Dalen-Fuchs nodules were macrophages. In contrast, the predominant infiltrating cell types in the vitreous were T helper/inducer lymphocytes. T suppressor/cytotoxic cells were rarely seen. Expression of Ia antigens on the ocular cells was confined to the immediate area of the inflammatory sites. The kinetics of histopathology showed two peaks at the 5th and 10th week after immunization, suggesting a relapsing course of the disease.
Introduction
Experimental autoimmune uveoretinitis (EAU) is an organ-specific, T-cell mediated autoimmune disease that has been used as a model for human intra-ocular inflammatory disorders (uveoretinitis) of unknown etiology [l-3]. Recently, it has been reported that EAU can be induced in mice by the retinal proteins, S-antigens or IRBP, an interphotoreceptor retinoid-binding protein [4,5]. Since the mouse is Correspondence to: Chi-Chao Chan, MD, National Eye Institute, Bldg. 10, Rm. lON206, Bethesda, MD 20892, USA. 247 0896-8411/90/030247
+ 09 $03.00/O
0 1990 Academic Press Limited
248 C.-C. Chan et al. an immunologically and genetically well-defined species, a murine EAU model provides definite advantages for understanding ocular autoimmune diseases with respect to basic immunological mechanisms and immunogenetics of the uveitogenic response. Previous results showed that the course of the disease was prolonged and that the ocular histopathologic findings were mainly located in the posterior pole with focal retinal and choroidal lesions, characterized by vasculitis and granuloma formation [4]. Histopathology in murine EAU was noted to be different from that in rats, but similar to that in some cases of human disease. In the present study, we followed the kinetics of EAU induced by interphotoreceptor retinoid binding protein (IRBP) in the BIO.A mouse strain, with respect to the dynamic changes in histopathology, inflammatory infiltrate and major histocompatibility complex (MHC) Class II-determinant expression on ocular resident cells. Materials and methods Female BIO.A mice (Jackson Laboratory, Bar Harbor, ME) at 2-5 months of age were housed in standard (non-specific pathogen-free) conditions. IRBP was prepared from bovine retinas by Con-A-Sepharose affinity chromatography and fast performance liquid chromatography as described previously [6]. For induction of EAU, the same protocol was used as reported previously [4]. Briefly, mice were pretreated 2 days before immunization with 20 mg/kg cyclophosphamide and were immunized with 50 pg of IRBP in a 1:l (V/V) emulsion with complete Freund’s adjuvent (CFA) containing 2.5 mg/ml of Mycobacterium tuberculosis strains H37RA, in a total volume of 0.1 ml divided between both hind footpads. Bordetella pertussis vaccine, lot 94 (Michigan Department of Public Health, Lansing, MI), containing pertussis organisms (10”) in a volume of 0.2 ml, was injected intraperitoneally at the same time. The animals were given a second dose of B. pertussis 3 days after primary immunization, followed by a secondary intramuscular immunization with 50 ng of IRBP in CFA one week after primary immunization. On several occasions, control groups of mice were immunized either with CFA/pertussis alone, without additional antigens other than the mycobacteria contained in CFA, or using bovine serum albumin (BSA) as irrelevant antigen. Different groups of animals were sacrificed at weekly intervals for 10 weeks after primary immunization (Figure 1). Freshly enucleated eyes were snap frozen in O.C.T. (Miles Scientific, Naperville, IL). Serial 4-6 pm frozen sections were prepared for immunohistochemical study by the avidin-biotin-peroxidase complex (ABC) method [7]. Adjacent frozen tissue sections were stained by standard hematoxylin and eosin. The intensity of EAU was scored in a masked fashion from 0 to 4 according to the histopathologic grading method developed previously for murine EAU [4]. Focal non-granulomatous, monocytic infiltration in the choroid, ciliary body and retina were scored as 0.5. Retinal perivascular infiltration and monocytic infiltration in the vitreous were scored as 1. Granuloma formation in the uvea and retina, the presence of occluded retinal vasculitis, along with photoreceptor folds, serous detachment and loss of photoreceptor were scored as 2. In addition, the formation of Dalen-Fuchs nodules (granuloma at the level of the retinal pigmented epithelium) and the development of subretinal neovascularization were scored as 3 and 4 according to the number and the size of the lesions.
Autoimmune
2
3
4
5
6
Weeks after
Figure 1. Histopathologic demonstrated a chronic,
7
8
uveoretinitis in mice
9
249
IO
immunization
kinetics of EAU induced by active immunization with IRBP in BIO.A mice possibly relapsing course. -O-, score; RI incidence.
Using the ABC method of tissue staining, biotin-conjugate anti-mouse IaK, H,Kk, Thyl,2, Lytl, Lyt2, and L,T, (Becton Dickinson, Mountain View, CA), Ml /70.15 (macrophage) and LR62B6B6.C9 (B cell) (Sera Lab., Westbury, NY) were applied to the tissue. After incubation, avidin-biotin and peroxidase complexes (Vector Laboratory, Burlingame, CA) were applied. The tissue sections subsequently were developed in diaminobenzidine. For quantitation of the labeled cells in each tissue section, the positive cells were counted and recorded. The results from each group of animals were averaged, compared and analyzed. Result
Only mice immunized with IRBP and not the control groups developed EAU. Kinetics Histopathological disease was observed as early as 2 weeks after immunization, with an incidence of 28.6% and an average grade of 1.3. Both the incidence and the average grade of histopathology varied through the lo-week examination period (Figure 1). By week 5 after immunization, the EAU incidence rose to 85.7% and the average grade was 2.71. Interestingly, both the EAU incidence and histopathological grade dropped to 0 in the 7th week after immunization. Thereafter, both EAU incidence and average grade slowly rose between the 8th and 10th week postimmunization. An incidence of 83.3% and an average grade of 2.5 were recorded at week 10 postimmunization. Histopathology The histopathologic features were essentially identical to those noted in earlier experiments [4]. The main pathologic features were located in the ocular posterior
250
C.-C.
Chan
et aZ.
Figure 2. Routine histopathology of the posterior segment. Left: A focal retinal granuloma (open arrow), retinal fold (arrowhead) and choroidal infiltration. Center: marked retinal vasculitis (arrow) and choroidal infiltration. Right: Focal infiltration between retinal pigment epithelium and choroid (Dalen-Fuchs nodule, asterisk) and choroidal infiltration (R = retina; C = choroid; hematoxylin and eosin, x 60).
Figure 3. Frozen sections of the posterior segment. T helper/inducer cells (arrows) were present in the vitreous (V) and retina(R). (L,T,, avidin-biotin-immunoperoxidase, x 150).
pole, as cited in our original report [4]. Minimum anterior involvement and mild vitritis were noted. The main pathologic lesions included inflammatory cell infiltration in the choroid and retina, and retinal damage. Most lesions were focal and discrete (Figure 2). On average, about 10% of the mice were found to have subretinal neovascularization, usually coinciding with the presence of a more severe choroidal inflammation (Figure 2). Immunopathology The inflammatory infiltrate consisted mainly of cells positive for IaK + , HZKk+, Thyl,2+, Lytl+, LsT,+ an d macrophage. The cells that bore surface markers for T helper lymphocytes (Thy1,2+, Lytl +, L3T4+) were located in the vitreous, uvea and
Autoimmune uveoretinitis in mice
Figure 4. Frozen retinal granuloma.
sections of the posterior R = retina, V =vitreous.
segment. Macrophages (arrow) were the main component (M1/70.15, avidin-biotin-immunoperoxidase, x 150).
251
in
perivascular regions of the retina (Figure 3). The cells that bore surface markers for macrophages (IaK+, HZKk+, macrophage) were mainly composed of granulomas and located in the uvea and the retina (Figure 4). Cells with the surface marker for T suppressor/cytotoxic lymphocytes (Lyt2+) or B lymphocytes were rarely seen. The composition of the inflammatory infiltrate in respect to the relative numbers of cells bearing the different markers did not change appreciably throughout the course of the disease. The ratios between T lymphocytes (Thy1,2+) and macrophages, and between T helper (L3T4) and T suppressor/cytotoxic (Lyt2+) lymphocytes were found to be 15 and 2O:l respectively, both in the early active (3rd to 5th weeks) and late, possible relapsing (9th to 10th weeks) stages. However, the ratios between T and B lymphocytes changed from 1O:l in the early stage to 6: 1 in the late stage. No influx of T suppressor/cytotoxic cells (Lyt2+) was observed in the intermediate (i.e., apparent remission) stage (6th to 8th weeks). The majority of inflammatory cells stained positively for Ia antigens. The ocular resident cells (the vascular endothelium and pigment epithelium) stained positively for Ia antigens (IaK+) only at and around the inflammatory sites. Resident cells where granuloma was present expressed MHC Class I and II antigens most strongly (Figure 5). No obvious enhancement of the staining intensity on ocular resident cells was noted as EAU scores became higher or the course of the disease progressed. Ia antigens were expressed only on the vascular endothelia, pericytes and cells in the perivascular infiltrates; detectable Ia expression coincided with the appearance of inflammatory cells. Expression of Ia antigens on ocular resident cells persisted during the active stages of disease, decreased in number of foci in parallel with the decrease in inflammatory lesions, and became undetectable in remission. Discussion
This study demonstrates for the first time the kinetics of EAU pathology in mice induced by immunization with the retinal antigen. Two weeks after immunization,
252 C.-C. Chan et al.
Figure 5. Frozen sections of the posterior segment. Most inflammatory cells (arrows) in the vitreous (V) and retina (R) stained positively for MHC Class I antigen (x 150). Inset: The retinal endothelia at the inflammatory sites stained positively for Ia antigen (x 240). [H,Kk, Iak (inset), avidin-biotinimmunoperoxidase].
histopathologic lesions began to develop with a chronic, focal, granulomatous inflammation in the retina and uvea. As the disease progressed, the number and size of these lesions increased. Both the incidence and severity scores were highest. By the 5th week the ocular inflammation was accompanied by photoreceptor degeneration and subretinal neovascularization. These two pathologic features are known to be dependent on the severity of inflammation [l, 21. By the 6th and 7th weeks after immunization, the disease appeared to go into remission, showing a decrease in histopathologic involvement and a drop in incidence. Begining at 8 weeks after primary immunization, the disease appeared to relapse as evidenced by a rise in the incidence as well as the number of lesions in the 9th and 10th weeks. Pathologically, the late-appearing lesions were similar to those of early EAU. The presence of a relapse, shown in this experiment by histopathology using different groups of mice, has subsequently been confirmed in a 3 month longitudinal follow-up of the same animals by weekly fundoscopic examination. The clinical data showed recurrent chorioretinal lesions in the same mouse (unpublished results). In a previous study [4] we observed slightly different kinetics, where disease appeared more severe with respect to histopathological grading and incidence. Animals were still positive at 7 weeks, although disease intensity was already subsiding. This difference is most likely due to the different pertussis vaccine lots used, since disease course and severity is directly related to the dosage of the pertussis adjuvant (Caspi et al., submitted). The animals were not followed beyond 7 weeks; however, we would expect the relapse to have been delayed in this case. Although EAU in the rat model is a non-relapsing disease, it is common to find focal recurrent lesions in chronic uveitis in humans. [8]. EAU in the Lewis rat is an acute inflammatory disease that occurs approximately 14 days after immunization, characterized by panuveioretinitis and lasting about 2 weeks. Complete photoreceptor destruction and retinal gliosis are found postinflammation [ 1,2]. EAU in
Autoimmune uveoretinitis in mice
253
the PVG rat [9] and Lister black hooded rat [lo] may present as milder, chronic disease with focal lesions. However, these alternative r&t models have not yet gained wide acceptance. When guinea-pigs are immunized with a low dose of uveitogenic antigen, a granulomatous infiltration in the uvea and destruction of photoreceptors lasting a few months have been described [ 1,2]. Repeated episodes of ocular inflammation by follow-up examination of fundus and fluorescein angiogram also have been seen in monkey EAU (Le Hoang, personal communication). In chronic murine experimental allergic encephalomyelitis (EAE), a relapse pattern was documented by clinical and histopathologic evaluations [ 11, 121. It was also noted that relapse in murine EAE is not a factor for increasing the degree of demyelination. In our study, both macrophages and T helper/inducer lymphocytes were found to comprise the majority of infiltrating inflammatory cells in the early and the putatively relapsing stages of murine EAU. Macrophages are the main component of the granulomatous infiltrate, and T helper/inducer cells are the main component of the non-granulomatous infiltrates inside the eye. Only occasional T suppressor/ cytotoxic cells are observed. Such cellular kinetics are quite different from those in Lewis rat EAU, where the inflammatory cellular population inside the eye changes markedly dur.ing a relatively short course (approximately 4 weeks). In the Lewis rat, T lymphocytes and not macrophages constitute the majority of the cells in the ocular infiltrates throughout the course of disease, with early predominant T helper/ inducers and late predominant T suppressor/cytotoxic cells [13,14]. Macrophages and polymorphonuclear leukocytes are seen at the beginning of ocular infiltration before the arrival of lymphoid cells in the early stage. It is conceivable that, in the mouse model, the persistent infiltrating macrophages and T helper/inducer cells may be conducive to the formation of ocular granulomas, while the paucity of T suppressor/cytotoxic cells may contribute to prolonging and possibly the relapsing course of murine EAU. It has been suggested that T suppressor lymphocytes may be needed for downregulation of the inflammatory response and modulation of the disease, both in rat EAU [ 13,141 and the end stage of some uveitides in humans [ 151. The finding that ocular resident cells located at or surrounding the inflammatory sites inside the eye bear Ia antigens is of particular interest, since similar findings have been reported in human uveitis [ 15,161 When the disease goes into remission, no inflammatory foci are found, and no Ia expression on ocular resident cells is seen. This observation is similar to that in murine EAE [17], where Ia antigens are induced on the vessels and astrocytes located at the inflammatory sites during the acute and relapsing stages. It has been suggested that local antigen presentation can induce damage to the blood-brain barrier [ 181. Therefore, at the site of Ia antigen expression, more inflammatory cells might penetrate and perpetuate the formation of a local ocular inflammation. The limited Ia antigen expression in murine EAU contrasts with widespread expression of MHC Class II antigens in rat EAU [ 19,201, and is probably related to the focal nature of the inflammatory cellular infiltrates. In conclusion, the murine EAU model offers differences in pathology and kinetics from EAU models in the Lewis rat, which today constitutes the best studied and most widely used rodent model of EAU. These differences include chronicity, focal nature, presence of Dalen-Fuchs nodules and subretinal neovascularization, relatively steady inflammatory cellular population, localized expression of Ia antigens, as well as a relapsing nature. In these respects murine EAU could be useful for the study
254 C.-C. Chan et al.
of mechanisms involved in some human uveitides such as idiopathic chronic uveitis and ocular autoimmune disorders for which no satisfactory rodent model has existed previously. Acknowledgement The authors thank Ms. Janet Edds for excellent secretarial assistance. References 1. Faure, J. I’. 1980. Autoimmunity and the retina. Curr. Top. Eye Res. 2: 215-302 2. Gery, I., M. Mochizuki, and R. B. Nussenblatt. 1986. Retinal specific antigens and immunopathogenic processes they provoke. Prog. Retinal Res. 5: 75-109 3. Nussenblatt, R. B., T. K. Kuwabara, F. de Monasterio, and W. B. Wacker. 1981. S-antigen uveitis in primates: A new model for human disease. Arch. Ophthalmol. 99: 1090-1092 4. Caspi, R. R., F. G. Roberge, C. C. Chan, B. Wiggert, G. J. Chader, L. A. Rozenszajn, Z. Lando, and R. B. Nussenblatt. 1988. A new model of autoimmune disease. Experimental autoimmune uveoretinitis induced in mice with two different retinal antigens.J. Zmmunol. 140: 1490-1495 5. Iwase, K., Y. Fujui, I. Nakashima, N. Kato, Y. Fujino, H. Kawashima, and M. Mochizuki. 1988. A new model for experimental autoimmune uveoretinitis (EAU) in the mouse. Invest. Ophthalmol. Vis. Sci. (suppl) 29: 39 6. Redmond, T., B. Wiggert, N. Nguyen, M. Lewis, L. Lee, and G. J. Chader. 1985. Isolation and characterization of monkey interphotoreceptor retinoid-binding protein, a unique extracellular matrix component of the retina. Biochemistry 24: 787-793 7. Hsu, S. M., L. Raine, and H. Franglor. 1981. The use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase technique. J. Histochem. Cytochem. 29: 577-580. and Clinical 8. Nussenblatt, R. B. and A. G. Palestine. 1988. In Uveitis: Fundamentals Practice. R. B. Nussenblatt and A. G. Palestine (eds). Year Book Medical Publishers, New York. 9. deKozak, Y., J. Sakai, B. Thillaye, and J. P. Faure. 1981. S antigen-induced experimental autoimmune uveo-retinitis in rats. Curr. Eye Res. 1: 327-340 10. Stanford, M. R., E. C. Brown, E. Kasp, E. M. Graham, M. D. Sanders, and D. C. Dumonde. 1987. Experimental posterior uveitis. I: A clinical, angiographic, and pathological study. Br. J. Ophthalmol. 71,585-592 11. Brown, A. M. and D. McFarlin. 1981. Relapsing experimental allergic encephalomyelitis in the SJL/ J mouse. Lab. Invest. 45: 278-284. 12. Tabiza, T. and K. Sakai. 1987. Demyleination induced by T cell lines and clones specific for myelin basic protein in mice. Lab. Invest. 56: 518-525 13. Chan, C. C., M. Mochizuki, R. B. Nussenblatt, A. G. Palestine, C. McAllister, I. Gery, and D. BenEzra. 1985. T lymphocyte subsets in experimental autoimmune uveitis. Clin. Zmmunol. Zmmunopathol.
35: 103-110.
14. Brown, E. C., E. Kasp, and D. C. Dumonde. 1989. Morphometric analysis of T lymphocyte compartmentation in experimental autoimmune uveoretinitis. Clin. Exp. Immunol. 77: 422-427 15. Chan, C. C., R. B. Nussenblatt, L. S. Fujikawa, A. G. Palestine, G. Stevens, L. M. Parver, M. W. Luckenback, and T. Kuwabara. 1986. Sympathetic ophthalmia: Immunopathologic findings. Ophthalmology 93: 690-695 16. Chan, C. C., R. P. Wetzig, A. G. Palestine, T. Kuwabara and R. B. Nussenblatt. 1987. Immunohistopathology of ocular sarcoidosis. Report of a case and discussion of immunopathogenesis. Arch. Ophthalmol. 105: 1398-1402 17. Sakai, K., T. Tabira, M. Endoh, and L. Steinman. 1986. Ia expression in chronic relapsing experimental allergic encephalomyelitis induced by long-term cultured T cell line in mice. Lab. Invest. 54: 345-352
Autoimmune uveoretinitis in mice
255
18. Traugott, U., C. S. Raine, and D. E. McFarline. 1985. Active experimental allergic encephalomyelitis in the mouse: immunopathology of the developing lesion. Cell. Immunol. 91: 240-254 19. Chan, C. C., J. J. Hooks, R. B. Nussenblatt, and B. Detrick. 1986. Expression of Ia antigen on retinal pigment epithelium in experimental autoimmune uveoretinitis. Curr. Eye Res. 5: 325-330 20. Fujikawa, L. S., C. C. Chan, C. McAllister, I. Gery, J. J. Hooks, B. Detrick and R. B. Nussenblatt. 1987. Retinal vascular endothelium expression fibronectin and class II histocompatibility complex antigens in experimental autoimmune uveitis. Cell. Zmmunol. 98: 139-150.
