Antigenic Specificity of Immunoprotective Therapeutic Vaccination for Glaucoma

Abstract

purpose. To investigate the antigenic specificity of the immune neuroprotective mechanism that can protect retinal ganglion cells (RGCs) against death caused by high intraocular pressure (IOP).

methods. A unilateral increase in IOP was induced in rats by argon laser photocoagulation of the episcleral veins and limbal plexus. Rats with high IOP were immunized with glatiramer acetate (Cop-1, a synthetic copolymer) or with myelin-derived or uveitogenic peptides. When the steroid drug methylprednisolone was used, it was administered intraperitoneally every other day for 12 days.

results. Vaccination with myelin-derived peptides that reside in the axons failed to protect RGCs from death caused by high IOP. In contrast, IOP-induced RGC loss was reduced by vaccination with R16, a peptide derived from interphotoreceptor retinoid-binding protein, an immunodominant antigen residing in the eye. The benefit of protection against IOP-induced RGC loss outweighed the cost of the monophasic experimental autoimmune uveitis (EAU) that transiently developed in a susceptible rat strain. Treatment with methylprednisolone alleviated the disease symptoms, but caused further loss of RGCs. Cop-1 vaccination was effective in both EAU-resistant and EAU-susceptible strains.

conclusions. To benefit damaged neurons, immune neuroprotection should be directed against immunodominant antigens that reside in the site of damage. In a rat model of high IOP, RGCs can benefit from vaccination with peptides derived from proteins that are immunodominant in the eye but not from myelin-associated proteins. This suggests that the site of primary degeneration in IOP-induced RGC loss is in the eye. Cop-1 vaccination apparently circumvents the site-specificity barrier and provides protection without risk of inducing autoimmune disease.

Recent studies in our laboratory, using rat and mouse models of optic nerve injury and glutamate toxicity, have led our group to postulate that autoimmunity represents the body’s mechanism of protection against harmful self-components, ^{1 2 3} such as those produced in toxic amounts as a result of injury to the central nervous system (CNS). ⁴ An autoimmune disease might be the result of a failure to control properly the injury-induced autoimmune response, the purpose of which is essentially beneficial. ^{5 6}  

While investigating the possible role of the autoimmune response as the body’s protective mechanism against destructive self-compounds induced by CNS injury, our group showed that the antigenic specificity of the relevant protective autoimmune T cells in rats or mice appears to vary with the site of the insult and the antigen(s) presented to the T cells at that site. ^{7 8 9} Further studies suggested that the one of the functions of the evoked immune response is to help the resident immune cells (microglia) clear the injured site of cell debris and other deleterious matter, such as breakdown products of degenerating nerves ^{10 11} (Shaked et al., unpublished data, 2003). In the course of studies seeking to boost the spontaneously evoked protective autoimmune response without risking an autoimmune disease, it was discovered that this goal could be achieved by vaccination with copolymer-1 (Cop-1; Copaxone; Teva Pharmaceutical Industries, Petah Tikva, Israel), ¹² which appears to operate as an antigen that cross-reacts with a wide range of self-reacting T cells. ^{10 13} Vaccination with Cop-1 effectively protects retinal ganglion cells (RGCs) against death initiated in the axons by mechanical injury or in the cell bodies by glutamate toxicity or high intraocular pressure (IOP). ^{2 12} In contrast to the effect of Cop-1, the immune response evoked by myelin-associated antigens is protective only from death induced by axonal injury, ^{14 15} and not from direct death of the RGCs caused, for example, by local injection of glutamate. ²  

Our recent study of IOP-induced RGC death in rats showed that strain differences in the ability to withstand the consequences of high IOP are immune-related. ¹⁶ In T-cell–deficient rats of a strain in which naïve rats are relatively resistant to IOP-induced death, the constitutional resistance was restored by replenishment of the rats with splenocytes from a matched donor that had the appropriate ability to harness a T cell–mediated protective activity. ¹⁶  

Accumulating evidence in our laboratory suggests that the antigenic specificity of immune intervention by therapeutic vaccination for regrowth is dependent on the site of insult rather than the type of injury. It was therefore of interest to examine which of two types of self-antigens, those residing in the nerve or those residing in the eye, would be beneficial as a therapeutic vaccine for the treatment of RGCs under stress induced by high IOP. Our results showed that vaccination with myelin-derived peptides did not lead to protection in a rat model of IOP-induced RGC death. In contrast, R16, a peptide derived from interphotoreceptor-retinoid binding protein (IRBP), an immunodominant antigen residing in the eye, conferred significant protection against the RGC death associated with high IOP. Vaccination with R16, known to induce monophasic experimental autoimmune uveitis (EAU) in susceptible rats, caused some RGC damage in immunized EAU-susceptible rats with normal IOP but if these rats had been subjected to an increase in IOP, this cost was outweighed by the vaccination’s protective benefit. Moreover, in rats that are not susceptible to EAU, no such cost was incurred. 

Materials and Methods

Animals

Inbred adult male Lewis and Sprague-Dawley (SPD) rats (8 weeks old, average weight 300 g) were supplied by the Animal Breeding Center at The Weizmann Institute of Science. The rats were maintained in a light- and temperature-controlled room and were matched for age and weight before each experiment. All animals were handled according to the regulations formulated by the International Animal Care and Use Committee (IACUC) and according to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. 

Induction of High IOP

Blockage of aqueous outflow causes an increase in IOP, which results in death of RGCs. ^{2 16 17 18} An increase in IOP was achieved in the right eyes of deeply anesthetized rats (ketamine HCl 50 mg/kg, xylazine HCl 0.5 mg/kg, injected intramuscularly) by blocking the aqueous outflow in the eye with 80 to 120 applications of blue-green argon laser. The laser beam, which was directed at three of the four episcleral veins and at 270° of the limbal plexus, was applied with a power of 1 watt for 0.2 seconds, producing a spot size of 100 μm at the episcleral veins and 50 μm at the limbal plexus. At a second laser session 1 week later, the same parameters were used, except that this time the spot size was 100 μm for all applications, and the radiation was directed toward all four episcleral veins and 360° of the limbal plexus. ²  

Measurement of IOP

Most anesthetic agents cause a reduction in IOP, ¹⁹ thus precluding reliable measurement. To obtain accurate pressure measurements while the rat was in a tranquil state, we injected the rat intraperitoneally (IP) with 10 mg/mL acepromazine and measured the pressure in both eyes 5 minutes later with a handheld tonometer (Tono-Pen XL; Automated Ophthalmics, Ellicott City, MD), after applying benoxinate HCl 0.4% to the cornea. Because of the reported effect of anesthetic drugs on IOP measured by the tonometer, ¹⁹ we always measured at the same time after acepromazine injection and recorded the average of 10 measurements made in each eye. 

Morphologic Assessment of Retinal Damage Caused by the Increase in IOP

The hydrophilic neurotracer dye dextran tetramethylrhodamine (Rhodamine Dextran; Molecular Probes, Eugene, OR) was applied directly into the intraorbital portion of the optic nerve. Only axons that survive the high IOP and remain functional, and whose cell bodies are still viable, can take up the dye and demonstrate labeled RGCs. ⁴ The rats were killed 24 hours after dye application, and their retinas were excised, wholemounted, and preserved in 4% paraformaldehyde. RGCs were counted under magnification of ×800 in a fluorescence microscope (Carl Zeiss Meditech; Oberkochen, Germany) by an observer who was blinded to the identity of the retinas. Four fields from each retina were counted, all with the same diameter (0.076 mm²) and located at the same distance from the optic disc. ^{3 20} Eyes from untreated rats were used as a control. 

Antigen Purification

The following peptides were synthesized by the Synthesis Unit at the Weizmann Institute: R16 sequence 1177-1191 of bovine IRBP (ADGSSWEGVGVVPDV); a nonencephalitogenic altered peptide, amino acids 87-99, derived from an encephalitogenic peptide of myelin basic protein (MBP) by replacing the lysine residue 91 with glycine (A91; VHFFANIVTPRTP) ^{21 22} ; and a Nogo-derived peptide (AS472; SYDSIKLEPENPPPYEEA), a myelin-associated growth inhibitory protein ^{23 24 25} recently found to be protective when used as a vaccine in rat models of spinal cord injury and optic nerve crush. ⁸  

Active Immunization with the Peptides

Rats with a laser-induced increase in IOP were immunized, immediately after the first laser session, with R16 (30 μg), A91 (75 μg), AS472 (100 μg), or Cop-1 (100 μg) or were injected with phosphate-buffered saline (PBS) as a control. All compounds were emulsified in complete Freund’s adjuvant (CFA) containing 2.5 mg/mL Mycobacterium tuberculosis (Difco, Detroit, MI). ⁸ A total volume of 100 μL was injected subcutaneously into each rat at the root of the tail. ¹²  

Assessment of Clinical Signs of Uveitis

Both eyes were examined under a microscope. The cornea, flare, and cells in the anterior chamber and iris hyperemia were evaluated according to the scale presented in Table 1 . The mean value for each parameter was calculated and graded. The examination was repeated every other day for 3 weeks. 

Steroid Administration

Methylprednisolone (MP; Solu-Medrol, 125 mg/mL, Pharmacia & Upjohn, Puurs, Belgium) was injected IP at a dosage of 30 mg/kg. ²⁶ Rats were injected on day 0 and again on days 3, 6, 9, and 12. ²⁷  

Results

Failure of Myelin-Derived Peptides to Evoke Neuroprotection

T-cell–dependent survival of RGCs was manifested in all eye models tested. Moreover, the T-cell–dependent neuroprotective mechanism was found to be amenable to boosting, eitherspecifically by antigens residing in the lesion site, or nonspecifically by Cop-1, which seems to circumvent the tissue-specificity barrier. Thus, whereas myelin-derived peptides were effective in protecting against the degenerative consequences of an insult to optic nerve axons, they could not protect against a glutamate insult imposed directly on the RGCs. As a corollary, vaccination with uveitogenic peptides protected RGCs from damage induced directly by glutamate. 

In the case of RGC loss caused by high IOP, it was not clear whether the primary site of degeneration is the optic nerve or the eye, and thus whether the antigenic specificity of the protective immune response would be directed to antigens residing in the nerve or the retina. To address this question, Lewis rats were subjected to a laser-induced increase in IOP and were vaccinated with A91 (a myelin-derived altered peptide that provides significant protection against RGC death after axonal injury) or the Nogo-derived peptide AS472 ⁸ on the day of the first laser treatment. Surviving RGCs were counted and expressed as a percentage of the number of RGCs in normal rats. As shown in Figure 1B , neither of these peptides could protect the RGCs against death caused by high IOP. In contrast, and as expected, ² significant protection was obtained when these rats were vaccinated with Cop-1 as a positive control (Fig. 1A) ; the extent of RGC survival after vaccination with Cop-1 was 87.4% ± 4.9% (n = 7) compared with 54.3% ± 8.2% in PBS-injected control subjects (n = 7; P < 0.0001). Thus, Cop-1 vaccination reduced the extent of death caused by high IOP by more than threefold. These results suggest that the death imposed by high IOP in rats was, at least during the early stages, not in the nerve but in the RGCs, and thus that the observed degeneration of the optic nerve is a late event. 

Immunization with Peptides Derived from Eye-Resident Proteins Protects against Retinal Ganglion Cell Death

To verify that the eye is the primary site of protection, we subjected Lewis rats to an increase in IOP and then immunized them with spinal cord homogenate or retinal homogenate. A beneficial effect of the vaccination was observed in the rats immunized with retinal homogenate but not in those immunized with spinal cord homogenate (Fig. 2) . These results suggest that the specificity of the immune response in the case of an IOP-induced insult should be directed against antigens that are immunodominant in the eye. On the basis of our previous observation that immunodominant antigens in a particular damaged CNS tissue are capable of eliciting an autoimmune response that is protective in that tissue, ⁹ we examined whether vaccination with an immunogenic peptide derived from the eye-abundant protein IRBP ²⁸ would protect the RGCs from a direct IOP insult. The peptide used to test this hypothesis was R16, already shown by our group to provide protection in a rat model of glutamate toxicity in the eye. ⁹ Immunization of Lewis rats with R16 resulted in significant protection of RGCs (Fig. 3A) . Substantially more RGCs survived in the group vaccinated with R16 (82.8% ± 7.8%, n = 7) than in the group injected with PBS (58.9% ± 5.3%, n = 6; P < 0.0001). Thus, the IOP-induced loss of RGCs was reduced by threefold by vaccination with R16. Similar protection from RGC death induced by high IOP was achieved in SPD rats, a strain resistant to the induction of EAU, in which the extent of RGC survival was 83.4% ± 6.7% (n = 6) after vaccination with R16, as opposed to 67.7% ± 5.4% (n = 4) after injection of PBS (P < 0.003; Fig. 3B ). 

Effect of Steroids on RGCs

Because R16 is a peptide that can cause EAU in Lewis rats, ^{29 30 31} it was important to determine whether the immunization itself could cause loss of RGCs in naïve rats with normal IOP. Vaccination of naïve Lewis rats with R16 emulsified in CFA caused EAU symptoms that were accompanied by 12% loss of RGCs compared with Lewis rats injected with PBS in CFA (n = 6; P < 0.001). A score for clinical symptoms of EAU was assigned to the R16-vaccinated Lewis rats (Fig. 4B) . According to the results, it seemed that although inflammation had some destructive effect on the healthy retina, its beneficial effect on the IOP-damaged retina outweighed the cost. In light of these findings and those in our previous work ^{15 32 33} showing that immune activity is important in promoting survival of damaged neurons, we were interested in finding out whether a high dose of steroids, often used clinically in acute CNS insults to wipe out inflammation, would be destructive to RGCs. Examination of the effect of MP on RGC survival in R16-vaccinated animals, as well as in nonvaccinated rats or control rats immunized with PBS in CFA, showed that MP prevented the R16-induced development of clinical EAU in Lewis rats (manifested by the low uveitis scores obtained by these rats compared with R16-vaccinated rats that were not treated with MP). However, in these EAU-susceptible rats, MP caused some loss of RGCs beyond the weak loss caused by R16-induced inflammation (Fig. 5) . Thus, MP did not protect RGCs from inflammation-induced death. Moreover, injection of MP into Lewis rats after they were injected with PBS in CFA showed that that the MP itself caused RGC death; the number of surviving RGCs in this group (1630 ± 285, n = 10) was even smaller (though not significantly) than that in the Lewis rats treated with MP after vaccination with R16 in CFA (1830 ± 352, n = 9; P = 0.19). In a separate set of experiments, using the same groups of rats (n = 5−10 in each group), we also compared the effect, in healthy Lewis rats, of MP injection with the effect of injection of PBS in CFA. The number of surviving RGCs per square millimeter in naïve Lewis rats injected with MP was 1815 ± 224 (n = 5) compared with 2715 ± 128 (n = 9) in naïve Lewis rats that were not injected (P < 0.01). No increase in IOP was observed in any of the Lewis rats after R16 vaccination or MP treatment (Table 2) , ruling out the possibility that the RGC death attributed to the inflammation or to the steroid treatment was actually caused by an increase in IOP. 

In further experiments, we also immunized naïve SPD rats with R16 emulsified in CFA. As expected, no clinical disease or RGC loss was observed. This finding prompted us to examine the effect of MP in rats in which EAU does not develop. Injection of MP alone in healthy SPD rats resulted in RGC loss, but the loss was reduced when these rats were also vaccinated with R16 (the RGC count in the nonvaccinated MP-treated group (n = 5) was 1341 ± 110 cells/mm², compared with 1668 ± 129 cells/mm² in the vaccinated group (n = 4), P < 0.01; Fig. 6 ). These results indicate that R16, unlike MP, does not cause RGC loss in an EAU-resistant strain. Moreover, R16 was partially capable of protecting the RGCs of SPD rats from MP-induced death. It thus appears that autoimmunity provides some protection against MP-induced death in SPD rats. 

Discussion

In this study, using a rat model for IOP-associated chronic glaucoma, we show that autoimmunity protects RGCs from death induced by high IOP. The protective autoimmune T cells were found to be specific to antigens residing in the eye, rather than to myelin-associated antigens found in the nerve fibers. 

Previous studies in our laboratory have demonstrated that T cells directed against myelin antigens protect neurons from the consequences of axonal injury. ⁹ Thus, passive transfer of T cells directed against myelin-derived peptides or active vaccination with those peptides reduced RGC loss after optic nerve injury. ^{15 34} It was further discovered that the ability to benefit from T cells directed against myelin antigens is not merely a reflection of experimental manipulation, but is the body’s natural way of coping with the injury. The physiological autoimmune response was found to be amenable to boosting. ^{3 20}  

Our group discovered, in addition, that a T-cell–mediated protective mechanism is also operative when the damage is caused by a direct injection of glutamate into the vitreous. ^{2 3} In that case however, myelin antigens failed to boost the response, whereas after an intraocular injection of glutamate, antigens residing in the eye evoked an immune response that successfully reduced the loss of RGCs. ⁹ It thus appears that for the T cells to be effective, they should be activated at the site of the stress where the relevant antigens are presented to them. ⁹  

Glaucoma belongs to a group of diseases often characterized by an increase in IOP. ^{35 36} In a previous study, using a rat model of IOP-induced RGC death, we showed that the immune system plays a key role in an animal’s ability to resist the damaging consequences of an increase in IOP. ¹⁶ Thus, rats deprived of T cells were found to be more prone to IOP-induced RGC loss than rats with normal T cell populations. In the present study we show that the ability of rats to resist IOP-induced RGC death can be improved by vaccination with antigens residing in the eye (but not in the optic nerve). The antigen found to protect RGCs against IOP-induced death in the present study is R16, a uveitis-associated peptide. ^{29 31 37 38 39}  

The present finding that protection of RGCs against IOP-induced death can be boosted by an immune response that is directed against proteins residing in the eye but not in the nerve, coupled with our previous finding that myelin proteins can elicit an immune response that protects RGCs against death induced by an axonal injury but not against death induced directly within the eye by glutamate, suggest that, at least at the early stage, IOP-induced death initiates degeneration in the eye, but not in the optic nerve head. The finding that vaccination with Cop-1 can counteract insults to RGCs imposed both by axonal injury and by toxic glutamate injection into the eye suggests that Cop-1 circumvents the tissue-specificity barrier, and is in line with the observed effect of Cop-1 in reducing IOP-induced RGC death. It is important to note that none of the therapeutic vaccinations appeared to be effective in rats deprived of mature T cells as adults (Bakalash et al., unpublished data, 2003). We have data suggesting that the beneficial effect of vaccination is manifested not only in the number of surviving RGCs, but also in functional activity, as indicated by reduction in loss of the visual evoked potential response (Ben-Shlomo et al., unpublished data, 2003). 

The finding that IOP-induced RGC death could be significantly reduced by eye-derived peptides suggests that the T cells exerting a beneficial effect are those activated by antigens presented to them in the eye. It should be emphasized that the eye-derived self-peptides that were tested in this study, and which in EAU-susceptible strains are potentially pathogenic to the eye, are not the peptides that would be selected for development of a therapy. They are useful, however, as a tool for gaining insight into the specificity of the body’s physiological immune remedy against stressful conditions associated with high IOP. According to our view, under stressful conditions the immune system is harnessed by the tissue to help remove endogenous harmful substances released as a result of the stress (Shaked et al., unpublished data, 2003). The present findings indicate that when the stress is caused by high IOP, such substances reside in the eye, at least in the early stages of the disease. 

It is important to note that the uveitogenic peptide that protects RGCs from IOP-induced death in naïve Lewis rats caused some RGC loss, which in the long run was outweighed by its benefit. In SPD rats there was no price to pay for the benefit, as this strain is not susceptible to EAU. 

Recent studies in our laboratory have shown that the phenotype of the T cells required for neuroprotection is Th1. ⁴⁰ In principle, therefore, any Th1 cells that are directed against antigens presented at the site of stress can lead to neuroprotection. Such T cells, once activated, provide a source of cytokines and possibly also of neurotrophins. ^{41 42} According to this view, it could be argued that the protection within the eye is associated with local inflammation controlled by autoimmunity. The specificity of the T-cell response—an anti-self response—has the role of amplifying and regulating the inflammation where it is needed. Once the cells become activated by antigens presented to them at the lesion site, the effect is nonspecific and is mediated by cytokines and other immune-derived factors, including neurotrophic factors. ^{34 41 43} Transient accumulation of T cells has indeed been observed in eyes of Cop-1–vaccinated rats with high IOP (Bakalash et al., unpublished data, 2003). 

One of the cytokines likely to play a role in protection is IFN-γ, a cytokine characteristic of Th1 cells. ⁴⁰ Studies in vitro have shown that IFN-γ can activate microglia, which we view as stand-by cells ready to participate in the dialog between the T cells and the neural tissue when needed and to remove deleterious matter, including glutamate toxicity, from the site of stress (Shaked et al., unpublished data, 2003). 

In Lewis rats with EAU, treatment with a high dose of the steroid drug MP attenuated the symptoms of inflammation. However, the treatment resulted in the death of more RGCs than was caused by the autoimmune disease itself. In naïve SPD rats, treatment with MP also resulted in a significant loss of RGCs, which was, however, slightly attenuated by the induction of an autoimmune response (i.e., by immunization with the self-reactive peptide R16). High-dose steroids are often used to treat patients after an acute traumatic CNS injury such as spinal cord injury. In the present study we used a high dose of steroid, not as a way to simulate therapy in an autoimmune disease like uveitis, but to highlight the paradox that autoimmunity is a defense mechanism that can be protective even against damage caused by steroids. It thus seems that autoimmunity can be viewed as a mechanism for maintenance and protection, whose absence might be more harmful than its presence. The results of this study further support our contention that epitopes of immunodominant proteins in a specific tissue are the ones selected by the tissue for maintenance, and that (in susceptible animals) development of an autoimmune disease in that tissue represents an extreme situation in which the mechanism that controls the immune response to these self-epitopes is defective. 

It was recently suggested that a correlation exists between normal-tension glaucoma and autoimmune disease. ⁴⁴ The present findings suggest that autoimmune disease, as opposed to an autoimmune response, can cause RGC loss. Accordingly, in cases in which normal-tension glaucoma is an outcome of autoimmune disease, the common practice of nonspecific immunosuppression by treatment with steroids should be critically reassessed. Moreover, after CNS injuries the potential benefit of steroid treatment to the eye is overridden by the negative effect to the RGCs, and perhaps also to other parts of the eye. ^{26 45 46}  

These findings support the development of vaccines for neuroprotection and immunomodulation designed to harness safely the body’s own repair mechanism, thereby serving the dual goal of avoiding neuronal loss due to autoimmune disease and boosting autoimmunity for protection against neuronal loss caused by nonimmune risk factors. ⁴⁷ Safe modulation can apparently be achieved by using a weak antigen, such as Cop-1, which can cross-react with self-reacting T cells without activating potentially pathogenic T cells. ¹⁰  

 Supported by Proneuron Ltd., Ness-Ziona, Israel, and in part by a grant from The Glaucoma Research Foundation (MS). MS holds the Maurice and Ilse Katz Professorial chair in Neuroimmunology.

 Submitted for publication January 26, 2003; revised March 4, 2003; accepted March 13, 2003.

 Disclosure: S. Bakalash, None; A. Kessler, None; T. Mizrahi, None; R. Nussenblatt, None; and M. Schwartz, None

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Table 1.

 

Uveitis Scores

Flare in Anterior Chamber	Cells in Anterior Chamber	Iris Hyperemia Scale	Conjunctival Congestion Grading
0, Absent	0, 0–7	0, Normal iris, no hyperemia	0, Normal, without perilimbal injection
+1, Faint flare	+1, 7–10	+1, Minimal involvement of secondary vessels	+1, Palpebral and confined perilimbal congestion
+2, Moderate flare	1–2+, 10–15	+2, Minimal involvement of both secondary and tertiary vessels	+2, Palpebral with at least 75% of perilimbal region
+3, Marked flare (details of the iris and lens are hazy)	+2, 15–20	+3, Moderate involvement of secondary and tertiary vessels with slight swelling of iris stroma	+3, Dark red congestion with petechiae
+4, Intense flare (with large amounts of fibrin)	+3, 20–50+4,+50	+4, Marked involvement of secondary and tertiary vessels with swelling of iris stroma accompanied by hyphema	+4, Confluent, dark petechiae of at least 50% of the conjunctival area

Figure 1.

 

Immunization of Lewis rats with myelin-associated antigens, unlike Cop-1 immunization, failed to rescue RGCs from death induced by high IOP. Immediately after the first increase in IOP, adult Lewis rats were immunized with Cop-1 (A) or with A91 (B) emulsified in CFA (2.5 mg/mL). The same peptides, this time emulsified in incomplete Freund’s adjuvant (IFA), were injected again 1 week later. Control rats were immunized with PBS in CFA and 1 week later with PBS in IFA. Whole-mounted retinas were excised 3 weeks after the first increase in IOP and 24 hours after injection of dye into the optic nerve. (A) The loss of RGCs in rats treated with Cop-1 was significantly smaller than in rats treated with PBS (n = 7 in each group; ***P < 0.0001). (B) No significant difference was detected between the groups injected with A91 (n = 7) and with PBS (n = 4; P = 0.3). The data in (A) are from one of three and in (B) from one of two independent experiments with similar results. Note that two PBS control experiments are shown because the experiments were performed on different days, and thus required separate controls.

Figure 2.

 

Protection of RGCs from IOP-induced death by immunization with retinal homogenate but not with spinal cord homogenate. Rats were subjected to two successive laser treatments to increase their IOP and immediately after the first treatment were immunized with spinal cord homogenate (SCH) or whole retinal homogenate (WRH) emulsified in CFA. Three weeks after immunization, the retinas were retrogradely labeled and then excised 24 hours later. Surviving RGCs were counted and expressed as a percentage of the number of RGCs in the nonimmunized control. Significantly more RGCs survived in the rats immunized with retinal homogenate than in control rats or rats immunized with spinal cord homogenate (***P < 0.001; n = 5−7 rats per group).

Figure 3.

 

Immunization of Lewis or SPD rats with the uveitogenic peptide R16, immediately after the first increase in IOP, protected RGCs from death caused by an increase in IOP. Adult (A) Lewis or (B) SPD rats were immunized with R16 emulsified in CFA. Control rats were injected with PBS in CFA. Three weeks after the first increase in IOP the retinas were stained. They were excised 24 hours later, and survival of RGCs was calculated as described in Figure 1 . In Lewis rats, significantly more RGCs survived in the group vaccinated with R16 than in the PBS-injected group (***P < 0.0001). A similar pattern was observed in the SPD rats, in which significantly more RGCs survived after vaccination with R16 (n = 6) than after injection with PBS (n = 4; **P < 0.003). As the control, normal retinas of both strains (n = 3 for each strain) were labeled at the same time, and their RGCs were counted and taken as 100%. In the SPD rats, clinical disease did not develop. The presented data are from one of four (A) and one of two (B) independent experiments with similar results.

Figure 4.

 

Experimental autoimmune uveitis (EAU) caused some death of RGCs. (A) Lewis rats were immunized with R16 emulsified in CFA, and control Lewis rats were injected with PBS in CFA (n = 6 in each group). RGC survival was measured by retrograde labeling with rhodamine dextran 3 weeks after immunization (by which time the disease had resolved itself). Immunization with R16 caused a small but significant loss of RGCs. The average number of RGCs per field in the R16-treated group was significantly lower than that in the group injected with PBS in CFA (***P < 0.001) or in the normal group (*P < 0.05). The difference between the two control groups was not significant (P = 0.87). (B) Mean clinical scores for EAU in Lewis rats injected with R16 (see grading in Table 1 ). The first signs appeared on day 10 after R16 immunization, peak symptoms were seen on day 14, and the disease resolved itself by day 21. Clinical disease did not develop in control rats injected with PBS. The presented data are from one of four independent experiments with similar results.

Figure 5.

 

Steroid treatment alleviates EAU symptoms but reduces the number of viable RGCs in Lewis rats with EAU. Adult Lewis rats were vaccinated with R16 emulsified in CFA or injected with PBS in CFA and were injected IP immediately afterward with 30 mg/kg of MP (125 mg/2 mL). The MP injection was repeated 3, 6, 9, and 12 days after the start of the experiment. After 3 weeks, the retinas were stained and 24 hours later were excised and the RGCs counted. The number of surviving RGCs in R16-vaccinated rats treated with MP (n = 9) was significantly smaller than in R16-vaccinated rats not treated with MP (n = 9; ***P < 0.0001). In addition, RGC survival in the group injected with PBS in CFA and MP (n = 10) was significantly lower than that in the group injected with PBS in CFA only (n = 4; **P < 0.01). The presented data are from one of two independent experiments with similar results.

Table 2.

 

IOPs in Lewis and SPD Rats, with and without R16 Vaccination and/or Steroids

Group (n)	Mean IOP ± SD
Normal Lewis	17.78 ± 1.08
Normal SPD	17.13 ± 1.28
Lewis+R16	19.83 ± 1.54
Lewis+steroids	18.8 ± 1.86
Lewis+R16+steroids	16.85 ± 2.25
SPD+steroids	18.04 ± 1.32
SPD+R16+steroids	17.87 ± 1.14

Figure 6.

 

R16 vaccination protected RGCs from steroid-induced death in a strain that is resistant to development of EAU. SPD rats (EAU-resistant) were vaccinated with R16 or injected with PBS and treated with MP according to the same protocol as that used for Lewis rats (described in Fig. 4 ). Three weeks later, the retinas were stained and excised, and the numbers of surviving RGCs were determined. The percentage of surviving RGCs was significantly smaller in the nonvaccinated group of rats injected every other day with MP (n = 5) than in the group vaccinated with R16 and not treated with MP (n = 12; ***P < 0.0001). These results show that steroid treatment can have a deleterious effect on neuronal survival. R16 immunization showed some protection of the RGCs from MP-induced RGC loss. Significantly fewer RGCs survived in rats treated only with MP than in rats treated with both R16 vaccination and MP (n = 4; **P < 0.01). The presented data are from one of two independent experiments with similar results.

The authors thank Shirley Smith for editing the manuscript and Avital Sharoni for animal maintenance.