(ACAID) in Aged Retinal Degeneration - Semantic Scholar

Loss of Anterior Chamber-Associated Immune Deviation (ACAID) in Aged Retinal Degeneration (rd) Mice Ulrich Welge–Lu ¨ ßen,1 Caren Wilsch,1 Thomas Neuhardt,1 J. Wayne Streilein,2 and Elke Lu ¨ tjen–Drecoll1 PURPOSE. To determine whether the capacity to induce ACAID by antigen injection into the anterior chamber is altered in animals with genetically determined retinal degeneration and increased age. METHODS. Anterior chamber–associated immune deviation (ACAID) induced by injection of ovalbumin into the anterior chamber of the eye was studied in three rodent strains with different forms of hereditary retinal degeneration (Royal College of Surgeon [RCS] rats, retinal degeneration [rd] mice, and Norrie–Disease [ND] mice) and in different age groups (age range, 1–23 months). The data were compared with those of age-matched controls. Aqueous humors of rd mice, RCS rats, and age-matched congenic controls were investigated for concentrations of transforming growth factor-b2 (TGF-b2) using enzyme-linked immunosorbent assay. RESULTS. ACAID was readily induced in RCS rats and ND mice irrespective of amount of retinal degeneration or aging. In rd mice ACAID could be induced in young animals but not in animals more than 12 months of age. In old rd mice, loss of ACAID was accompanied by a marked reduction in total TGF-b2 levels in aqueous humor. CONCLUSIONS. Rd mice more than 1 year of age lose the capacity of the anterior chamber to support the induction of ACAID by intracameral injection of soluble protein antigen. Because loss of ACAID correlated with a decrease in TGF-b2 concentration in aqueous humor, it is proposed that eyes of rd mice are unable to maintain an immunosuppressive microenvironment necessary for ACAID. (Invest Ophthalmol Vis Sci. 1999;40:3209 –3214)

I

mmune privilege in the eye can be defined as the prolonged survival of allografts placed in ocular compartments or tissues, such as the anterior chamber, the vitreous cavity, the subretinal space, and the corneal stroma. The stereotypical systemic immune response evoked by intracamerally injected soluble antigens is termed “anterior chamber–associated immune deviation” (ACAID) and is characterized by a deficiency of antigen-specific delayed-type hypersensitivity (DTH)–mediating T cells and of complement-fixing antibodies.1 It is well documented that the establishment of the ACAID phenomenon is dependent on the immunosuppressive microenvironment of the aqueous humor. A number of immunoregulatory cytokines are normally present in the aqueous humor, including transforming growth factor-b (TGF-b), a-melanocyte–stimulating hormone, calcitonin gene–related peptide, free cortisol, and vasoactive intestinal peptide.1 In vitro TGF-b has been demonstrated to play a specific role for the induction of ACAID.2 Recent studies have also shown that the ocular immune response can be modulated by external factors. Dark-

rearing of mice abolished suppression of DTH and elicitation of ACAID. In these mice a different pattern in the concentration of neuropeptides has been observed and discussed as being responsible for the suppression of ACAID.3 In previous studies, we found that early-onset retinal degeneration correlates with alterations in components of eye growth and development of structures in the anterior eye segment. We hypothesized that these changes are due to qualitative or quantitative changes in factors released from the degenerating retina.4 Changes in factors associated with hereditary retinal degeneration might also influence the immune privilege of the anterior chamber. Therefore, in the present study we investigated ACAID in different animal models with hereditary retinal degeneration.

MATERIALS

AND

METHODS

Experimental Animals

From the 1Department of Anatomy II, Universita¨tsstrabe 19, 91054 Erlangen, Germany; and 2Schepens Eye Research Institute, Harvard Medical School, Boston, Massachusetts. Supported by IZKF Erlangen (ELD); by a grant from the Academy of Science, Mainz (ELD); and by BIOMED BMH4-CT96-1593 (ELD). Submitted for publication August 10, 1998; revised January 21 and May 18, 1999; accepted June 30, 1999. Commercial relationships policy: N. Corresponding author: Elke Lu ¨ tjen–Drecoll Department of Anatomy II, University of Erlangen–Nu ¨ rnberg, Universita¨tsstr. 19, 91054 Erlangen, Germany. E-mail: anat2.gl@ anatomie.uni-erlangen.de

Three rodent strains with different forms of inherited retinal degeneration (rd) were tested. C57BL/6 mice were purchased from Charles River (Wilmington, MA). Rd mice (C57BL/6J rd/ rd, fast degeneration) were generously provided by Joe Hollyfield (The Cleveland Clinic Foundation, Ophthalmic Research, Cleveland, OH), Norrie–Disease (ND) mice by Wolfgang Berger (Department of Human Genetics, University of Nijmegen, the Netherlands), and Royal College of Surgeon (RCS) rats and their genetic controls (RCS-control rats) by Matthew La Vail (Beckman Vision Center, University of California, San Francisco). All animals were treated according to the ARVO Statement for the Use of Animals in Ophthalmic and Vision Research. All

Investigative Ophthalmology & Visual Science, December 1999, Vol. 40, No. 13 Copyright © Association for Research in Vision and Ophthalmology

3209

3210

Welge–Lu ¨ ßen et al.

treatments were conducted under anesthesia with an intramuscular injection of a 5% ketamine hydrochloride/xylazine solution (rats, 12–20 IU; mice, 4 –10 IU). Experimental groups consisted of 4 to 6 male animals. All experiments were performed at least twice for each animal strain and age group. Dystrophic RCS Rats. Dystrophic RCS rats develop retinal degeneration due to a specific phagocytotic defect of shed rod outer segments (ROS) in the retinal pigment epithelium (RPE). The increase of ROS debris in the subretinal space leads to a complete loss of photoreceptors, followed by reduction of the retinal cell layers. Reduction in RPE and adjacent choriocapillaris occurs almost exclusively in the upper temporal quadrant.5 In dystrophic RCS rats, ACAID experiments were performed in young (6-month-old) and old (23 months old) males. In addition, young (6-month-old) and old (23-month-old) male, age-matched congenic controls were tested. ND Mice. The ND mouse is a mouse model on the genetic background of the C57BL/6 strain for ND in humans, an x-chromosomal recessive disorder, characterized by blindness and sensorineuronal deficit. In these mice the murine gene orthologue for ND is replaced by an inactive gene copy. The gene product has similarities to mucins and TGF-b.6 The eyes of the animals exhibit degenerative changes of the neuronal retina and malformation of retinal blood vessels. The morphology of the RPE and choroid appeared to be normal.7 ACAID experiments were performed in young (5-monthold) and old (17-month-old) ND mice. C57BL/6J rd/rd (rd Mice). In C57BL/6J rd/rd (rd mice) the retina degenerates because of an autosomal recessive gene defect of the rod photoreceptor cGMP–phosphodiesterase b-subunit mapped on chromosome 5. These animals show reduction and loss of cells in the outer and inner nuclear layers and gliosis in the ganglion cell layer.8 In the central retinal area of older animals, we found pronounced degeneration of RPE cells and atrophy of the adjacent choriocapillaris.9 ACAID experiments were performed in different age groups (3-, 6-, 7-, 11-, 12-, 16-, and 20-month-old). C57BL/6 Mice (Control). C57BL/6 mice (pigmented) served as controls for all mouse strains. Animals were tested at 3, 12, and 21 months of age.

Assay for Induction of ACAID On day 0, animals received 50 mg chicken egg albumin (OVA; Sigma Chemical, St. Louis, MO), per 3 ml volume of either Hanks’ balanced salt solution (HBSS; Life Technologies, Grand Island, NY) or mock aqueous humor (MAH), into the anterior chamber (AC) of their right eyes, as described previously.10 On day 7 all animals received an immunizing subcutaneous (SC) dose of OVA (100 mg in HBSS or MAH) emulsified 1:1 in complete Freund’s adjuvant. A total volume of 100 ml was injected into the nape of the neck of each animal. Seven days after immunization animals received an intradermal injection of OVA (400 mg in 20 ml) into the right ear pinnae. Before injection, as well as 24 and 48 hours after injection, the ear-swelling response, as a measure of DTH, was assessed using a micrometer (Mitutuyo; MIT, Paramus, NJ). Animals that had not received AC injection but that did receive immunization and ear challenge served as positive controls. As negative controls for correct intradermal ear injection, unmanipulated mice received ear injection and ear thickness measurement at the time intervals indicated.

IOVS, December 1999, Vol. 40, No. 13

Statistical Evaluation Statistical treatment of ear-swelling measurements was accomplished using a two-tailed Student’s t-test.

Analysis of TGF-b2 by Enzyme-Linked Immunosorbent Assay TGF-b2 concentrations were determined in aqueous humor obtained from 24 rd mice (2, 3, 7, 12, 13, and 16 months old), 20 control mice (C57BL/6; 2, 3, 11, 13, and 15 months old) as well as from 12 RCS rats (3, 11.5, and 12 months old) and the same number of RCS-control rats (3, 5, and 13 months old). The aqueous humor from 8 eyes of each strain and age group was pooled in one tube, immediately stored at 280°C and thawed just before use. Unfortunately, TGF-b2 levels in aqueous humor of ND mice could not be determined because not enough mice of different age groups were available. For the determination of TGF-b2 concentrations in aqueous humor we used a double-antibody “sandwich” enzymelinked immunosorbent assay (ELISA; R&D Systems, Wiesbaden, Germany). This assay only detects the activated form of TGF-b2 and does not recognize the latent form. To activate latent TGF-b2 to the immunoreactive form, we acidified aqueous humor (16.2 ml) by the addition of 1 N HCl. This mixture was incubated for 10 minutes at room temperature and neutralized with 1.2 N NaOH/0.5 M HEPES. After adding 100 ml assay diluent to each well, we pipetted the probe or the provided standard to the microtiter ELISA plate. After 2 hours of incubation at room temperature, each well was washed three times with 400 ml washing buffer, and then 200 ml of polyclonal antibody against TGF-b2 conjugated to horseradish peroxidase was added. The incubation time was 2 hours. After washing 3 times with washing buffer the probe was incubated for 20 minutes at room temperature in 200 ml of substrate solution (a mixture of H2O2 and tetramethylbenzidine). The color development was stopped by adding 50 ml stop-solution to each well. The optical density was determined using ELISA reader (kinetic microplate reader; Molecular Devices, Ebersberg, Germany) set to 450 nm. For wavelength correction, readings at 540 nm were substracted from the readings at 450 nm. For the measurements of TGF-b2 concentrations by ELISA, each aqueous humor sample (derived from 8 eyes per age group) was used only once, because the entire sample volume was needed for one determination. The determinations were performed on 3 different days with the probe derived from 1 young (up to 11 months) and 1 old (older than 11 months) animal group each.

RESULTS Induction of ACAID rd Mice. In rd mice the ability to establish ACAID was dependent on age. The results of these experiments are displayed in Figures 1A and 1B. In 26 of 27 animals tested until the age of 7 months, we were able to elicit ACAID. DTH response, measured as mean ear swelling 6 SE, showed that animals at the ages of 3 and 6 months showed significantly lower mean response than positive controls. Only 1 of 9 7-month-old animals and two of eleven 11-month-old animals developed intense DTH even though they had received antigen by the AC route before sensitization. With progressing age, starting at the age of 12 months (6/8 animals), older experimental groups

IOVS, December 1999, Vol. 40, No. 13

Loss of ACAID in Old rd Mice

3211

FIGURE 1. ACAID was inducible in young rd mice at the age of 3 and 6 months. ACAID was not inducible in the great majority of rd mice aged between 11 and 20 months. DTH was measured in rd mice at the age of 3 to 20 months (A). Mice received an AC injection of OVA (ACI) in HBSS or MAH on day 0 and were immunized SC with OVA in HBSS or MAH and complete Freund’s adjuvant on day 7. Ear challenge was performed on day 14. DTH was expressed as ear swelling (in micrometers) 48 hours after ear challenge. The age of animals in each experimental group is indicated at each box. Bar represents 6 SEM. As positive controls, animals were immunized SC and ear challenged but did not receive ACI. Naive animals that underwent ear challenge only served as negative controls. rd mice aged 3, 6, 7, and 11 months showed significantly a lower mean response than positive control animals (P , 0.05). rd mice aged 12, 16, and 20 months showed a significantly higher mean response than negative control animals (P , 0.05). Single values of ear assessment are presented in (B).

3212

Welge–Lu ¨ ßen et al.

IOVS, December 1999, Vol. 40, No. 13

FIGURE 2. ACAID was inducible in control mice and control rats and in ND mice and RCS rats with hereditary retinal degeneration in all tested age groups. DTH was measured in ND mice (5 and 17 months) RCS rats (6 and 23 months; A), C57BL/6 mice (3, 12, and 21 months), RCS-control rats (6 and 23 months; B). Animals received an AC injection of OVA (ACI) in HBSS or MAH on day 0 and were immunized SC with OVA in HBSS or MAH and complete Freund’s adjuvant on day 7. Ear challenge was performed on day 14. DTH was expressed as ear swelling (in micrometers) 48 hours after ear challenge. The age of animals in each experimental group is indicated at each box. Bar represents 6 SEM. As positive controls, animals were immunized SC and ear challenged but did not receive ACI. Naive animals that underwent ear challenge only served as negative controls. All groups that received ACI showed a significantly lower mean response than positive control animals (P , 0.05).

failed the ACAID experiment (Fig. 1B). In old animals without significant suppression of DTH, the ear-swelling values increased approximately 10-fold compared with younger animals with positive ACAID responses and with age-matched controls (Figs. 1A and 1B).

RCS Rats and ND Mice with Hereditary Retinal Degeneration In dystrophic RCS rats (6- and 23-month-old) and ND mice (5and 17-month-old) the mean DTH response, measured as mean

Loss of ACAID in Old rd Mice

IOVS, December 1999, Vol. 40, No. 13

TABLE 1. Quantitative Measurement of Total TGF-b2 in Pooled Aqueous Humor of 48 rd-Mouse Eyes, 40 Control Mouse Eyes, and 24 RCS and 24 RCS-Control Rat Eyes Strain

Age, mo

Total TGF-b2 pg/ml

C57BL/6 C57BL/6 C57BL/6 C57BL/6 C57BL/6 rd rd rd rd rd rd RCS control RCS control RCS control RCS RCS RCS

2 3 11 13 15 2 3 7 12 13 16 3 5 13 3 11.5 12

950 776 753 1172 700 480 560 960 316 210 110 1056 1389 1106 640 906 980

Aqueous humor from 8 eyes was pooled for each age group and animal strain.

ear swelling 6 SE, was not significantly different (P . 0.05) from the corresponding negative controls (Fig. 2A).

Young and Old Control Animals In young (3-month-old) and old (12- and 21-month-old) C57BL/6 mice as well as in 6- and 23-month-old RCS-control rats mean DTH response, measured as mean ear swelling 6 SE, was significantly lower than in positive control animals (P , 0.05; Fig. 2B).

ELISA for TGF-b2 in rd and Control Mice After acid activation, the total amount of TGF-b2 was detected in all aqueous humor samples from rd and control mice of different age groups. In control mice, no age-related reduction in total TGF-b2 concentrations in the aqueous humor was observed, as the amount of total TGF-b2 in 3-month-old animals was 776 pg/ml, and 700 pg/ml in 15-month-old animals. Total TGF-b2 values in control mice ranged between 1172 pg/ml (in 13-month-old mice) and 700 pg/ml (in 15-month-old animals; Table 1). In rd mice TGF-b2 concentrations in the aqueous humor of all age groups were reduced compared with concentrations in age-matched controls, and there was an age-related reduction of total TGF-b2 levels in the aqueous humor between different age groups. In the 2- to 7-month-old rd mice the concentration of total TGF-b2 ranged between 480 and 960 pg/ml (Table 1), whereas in old rd mice (16-month-old) the amount of total TGF-b2 decreased to 110 pg/ml, which reflects approximately 13% of the mean total TGF-b2 concentration in control mice. In the group of 12- and 13-month-old rd mice, which usually already lack ACAID, total TGF-b2 levels were 316 and 210 pg/ml (Table 1).

ELISA for TGF-b2 in RCS and RCS Control Rats After acid activation, the total amount of TGF-b2 was determined in all aqueous humor samples from RCS and RCS-control rats of different age groups (Table 1).

3213

In RCS controls total TGF-b2 concentrations of all age groups (3-, 5-, and 13-month-old) were within the range of 1056 pg/ml (3-month-old) to 1389 pg/ml (5-month-old). Total TGF-b2 values obtained from RCS rats of all age groups (mean, 842 pg/ml) were slightly reduced compared with values of total TGF-b2 concentrations in RCS-control rats (mean, 1184 pg/ml). There was, however, no age-related decrease in total TGF-b2 levels in the aqueous humor of the different age groups (3, 11.5, and 12 months), as in young (3-month-old) RCS rats 640 pg/ml and in older (12-month-old) RCS rats 980 pg/ml were measured (Table 1).

DISCUSSION In the present study we demonstrated that rd mice older than 12 months lose the capacity of the aqueous humor to support ACAID by intracameral injection of soluble protein antigen. Loss of the ability to support ACAID might be expected to reflect a process of retinal degeneration or aging itself. The experimental finding that the capacity to establish ACAID was not lost in aged C57BL/6, RCS-control rats, or RCS rats and ND mice that suffer from retinal degeneration and consequently from blindness, strongly argues against retinal degeneration and age as the causal mechanism for the abolishment of ACAID in old rd mice. Immune privilege in the eye is a multifactorial process involving anatomic, physiologic, and immunologic components. In vitro studies indicate that TGF-b is one of the most important factors for the generation of ACAID-inducing properties.2 In the eyes of normal control mice we measured concentrations of total TGF-b2 within the range of 700 to 1170 pg/ml. In the experiments reported here, antigen was injected into the AC of eyes of aged rd mice, in which the concentration of total TGF-b2 in aqueous humor was markedly reduced. Especially when rd mice reached the age of 12 months and older, total TGF-b2 concentrations in the aqueous humor were reduced to 316 to 110 pg/ml and antigen injection in the AC failed to induce ACAID. Decreased levels of total TGF-b2 concentrations were not seen in old C57Bl/6 mice and RCS rats, or in their age-matched controls, indicating that age and retinal degeneration are not solely responsible for total TGF-b2 decline in rd mice. There appears to be a link between the capacity of an eye to support ACAID induction and a threshold concentration of total TGF-b2 in the aqueous humor. Unfortunately, an exact threshold could not be determined because due to the sensitivity of ELISA and the small volume of aqueous humor that can be obtained from one mouse eye (3–5 ml), we needed to pool aqueous humor. As the total TGF-b2 levels in aqueous humor from animals aged 12, 13 and 16 months were 316 to 110 pg/ml, 480 pg/ml from 11-month-old animals, we assume that the critical concentration of total TGF-b2 is within the range of 316 to 480 pg/ml. To the best of our knowledge this is the first demonstration that a deficit of total TGF-b2 in aqueous humor in vivo correlates with loss of ACAID. The mechanism by which the observed decline in total TGF-b2 concentrations in the aqueous humor of old rd mice is caused is not known. In a previous study it has been shown that old rd mice in contrast to RCS rats and ND mice develop a profound age-related loss of RPE cells in the central posterior pole, accompanied by a loss of adjacent choriocapillaris.9 It has been demonstrated that RPE cells con-

3214

Welge–Lu ¨ ßen et al.

stitutively produce TGF-b,11 but it is not yet known whether the RPE changes in rd mice are directly or indirectly involved in the reduction of total TGF-b2 in the aqueous humor. It is well known that several cells in the anterior eye segment are able to secrete TGF-b12 and that, even if there are no major morphologic changes in the anterior segment of old rd mice, formation of TGF-b by these cells may be reduced. Although our studies have sought a correlation between the reduction of intraocular concentrations of total TGF-b2 and the loss of ACAID, we are aware that changes in other factors in the aqueous humor or in the expression of FAS or FASligand13 may be equally important in the loss of immune privilege in the AC of aged rd mice. The rd mouse may provide a good model for further investigations of factors that are necessary in the induction of ACAID.

References 1. Streilein JW. Immunological non-responsiveness and acquisition of tolerance in relation to immune privilege in the eye. Eye. 1995;9: 236 –240. 2. Takeuchi M, Kosiewicz M, Alard P, Streilein JW. On the mechanisms by which transforming growth factor-beta 2 alters antigenpresenting abilities of macrophages on T cell activation. Eur J Immunol. 1997;27:1648 –1656. 3. Ferguson TA, Fletcher S, Herndon J, Griffith TS. Neuropeptides modulate immune deviation induced via the anterior chamber of the eye. J Immunol. 1995;155:1746 –1756.

IOVS, December 1999, Vol. 40, No. 13 4. Schreckenberger M, Eichhorn M, Gottanka J, Do ¨ big C, Lu ¨ tjen– Drecoll E. Altered proportions of RCS-rat eyes. Exp Eye Res. 1994; 59:409 – 416. 5. May CA, Horneber M, Lu ¨ tjen–Drecoll E. Quantitative and morphological changes of the choroid vasculature in RCS rats and their congenic controls. Exp Eye Res. 1996;63:75– 84. 6. Berger W, Meindl A, van de Pol TJR, et al. Isolation of a candidate gene for Norrie disease by positional cloning. Nat Genet. 1992;1:199 –203. ¨tjen– 7. Richter M, Gottanka J, May CA, Welge–Lu¨ben U, Berger W, Lu Drecoll E. Morphological changes in the retinal vasculature of Norrie Disease-mice. Invest Ophthalmol Vis Sci. 1998;39:2450 –2457. 8. La Vail MM. The retinal pigment epithelium in mice and rats with inherited retinal degeneration. In: Zinn KM, Marmor MF, eds. The Retinal Pigment Epithelium. Cambridge, UK: Harvard University Press; 1979:357–380. 9. Neuhardt T, May CA, Eichhorn M, Wilsch C, Lu ¨ tjen–Drecoll E. Morphological changes of retinal pigment epithelium and choriocapillaris in rd-mice. Exp Eye Res. 1999;68:75– 83. 10. Kosiewicz MM, Okamoto S, Miki S, Ksander BR, Shimizu T, Streilein JW. Imposing deviant immunity on the presensitized state. J Immunol. 1994;153:2962–2973. 11. Tanihara H, Yoshida M, Matsumoto M, Yoshimura N. Identification of transforming growth factor-b expressed in cultured human retinal pigment epithelial cells. Invest Ophthalmol Vis Sci. 1993; 34:413– 419. 12. Peress N, Perillo E. TGF and TGF-b3 immunoreactivity within the ciliary epithelium. Invest Ophthalmol Vis Sci. 1994;35:454 – 476. 13. Griffith TS, Brunner T, Fletcher SM, Green DR, Ferguson TA. Fas ligand-induced apoptosis as a mechanism of immune privilege. Science. 1995;270:1189 –1192.