Current Eye Research
Volume 11 number 7 1992, 657-667 ~~
~
___~__
___
Induction of experimental autoimmune uveitis with rhodopsin synthetic peptides in Lewis rats Grazyna Adamus', Jacki L.Schmied', Paul A.Hargrave1,2,Anatol Arendt' and Edward J.Moticka3
Curr Eye Res Downloaded from informahealthcare.com by University of Melbourne on 11/23/14 For personal use only.
Departments of 10phthalmology, 2Biochemistryand Molecular Biology, University of Florida, Gainesville, FL and 'Department of Medical Microbiology and Immunology, Southern Illinois University School of Medicine, Springfield, IL, USA ABSTRACT Rhodopsin, a membrane protein of rod photoreceptor cells, induces an experimental autoimmune uveitis (EAU) in Lewis rats. Synthetic peptides derived from rhodopsin sequences that cover hydrophilic, exposed regions of the protein were tested for their capacity of eliciting in vitro T cell proliferation and their ability for inducing EAU in Lewis rats. Rats were injected with rhodopsin's peptides mixed in complete Freund's adjuvant containing M. tuberculosis H3zRa (5 mg/ml) three days after pretreatment with cyclophosphamide (20 mg/kg). ELISA results indicate that all peptides induce antibody responses; however antibody titers differ among sera tested. Immunization with four pephdes - the amino-terminus (2-32), loop 1-11(61-75), loop V-VI (230-251), and the carboxyl-terminus (324-348 and 331-342) induced both antibody and T cell responses. In ail cases, the proliferative responses of cells derived from peptide-injected rats were stronger against the immunizing peptide than against native protein. Three distinct uveitogenic epitopes were identified on rhodopsin's cytoplasmic surface - within the rhodopsin carboxylterminus (324-348), loop 1-11 (61-75), and loop V-VI (230-250). Histopathologically, at the immunized doses, total destruction of the photoreceptor cell layer was observed as compared to the control group. Loop V-VI caused severe inflammation of the retina while the other pathogenic peptides produced less severe destruction with few inflammatory cells present. Our study indicates that the major immunodominant T cell epitope (331-342) is also involved in EAU induction but is not the primary uveitogenic site.
inflammation of the uveal track, retina and pineal gland (1). It has been shown that specific regions of uveitogenic proteins are responsible for their pathogenicity (2-5). Several T cell epitopes for S-antigen and IRBP responsible for uveitogenic, proliferative, and adoptive transfer responses have been characterized with the use of CNBr fragments and synthetic peptides (6-11). Rhodopsin, an integral membrane glycoprotein, has been shown to induce EAU in guinea pigs, rats, and monkeys (12-19). In a recent study, we determined the specificity of T and B cell responses to bovine rhodopsin (19). T cells isolated from rhodopsin-injected rats responded
in vitro to determinants present in the carboxyl-terminus and in loops 1-11 (61-72), II-III (96-115) and IV-V (174-202)
(see topographic diagram of rhodopsin, Fig. 1). The major T cell determinant has been determined to be within the sequence 324-348. Using a series of 12 amino acid long overlapping peptides, the precise position of this epitope was localized to the sequence 331-342 (19). On the other hand, loop IV-V and the amino-terminus have been shown to be immunodominant sites for antibody binding.
To investigate the immunoregulatory mechanisms INTRODUCTION
responsible for controlling rhodopsin autoimmunoreactivity,
Experimental autoimmune uveitis is an ocular inflammatory
it is necessary to identify the immunopathogenicepitopes in
disease in susceptible animals induced by immunization with
this protein. In this report, we examined the ability of
retinal antigens such as rhodopsin, S-antigen (also known as
rhodopsin peptides to induce EAU in Lewis rats. We have
arrestin, and 48K protein), and interphotoreceptor retinoid
used rhodopsin synthetic peptides that represent hydrophilic,
binding protein (IRBP). This disease has been characterized
exposed regions of the protein to immunize animals. The
in part as a T cell-dependent disease that results in severe
immunogenicity of the peptides was correlated with
Received on January 27, 1992; accepted on June 16, 1992
657
Current Eye Research T CEL& EPlTOF'E (331-342)
Curr Eye Res Downloaded from informahealthcare.com by University of Melbourne on 11/23/14 For personal use only.
U
T CELL EPU'OPE
I
B CELL E~TTOPE .
(96-115)
(2-32)
Figure 1. A topographic model of rhodopsin in the lipid bilayer of the disc membrane of retinal rod cells (Adapted from Hargrave et. al [26]). AKOWSindicate regions of
rhodopsin that are found to represent T cell epitopes, B cell epitopes, and uveitogenic sites.
pathogenesis of the disease to determine which T-cell
days later, they were immunized in the hind footpad with a
epitopes of rhodopsin are possibly involved in the disease
single dose of peptide mixed 1:1 in complete Freund's
induction.
adjuvant with Mycobactenwn tuberculosis H37Ra (5 mg/rnl) in a total volume of 100 pl. Control animals received saline
MATERIALS AND METHODS
instead of peptide (negative control) or 100 pg rhodopsin
toimmune uveitis
(positive control) using above protocol. 25-28days after
Female Lewis rats were purchased from Charles River
immunization, the animals were killed, blood was collected
Laboratories. Groups of 3-4 rats (150-180g) were injected
and spleens were removed aseptically. All procedures
with cyclophosphamide (20 mg/kg) intrapentoneally. Three
adhered to the ARVO Resolution on the Use of Animals in
658
Current Eye Research
Curr Eye Res Downloaded from informahealthcare.com by University of Melbourne on 11/23/14 For personal use only.
Table 1. Synthetic peptides from bovine rhodopsin sequence used for immunization. RHODOPSIN REGION
RESIDUES NUMBERS
N-terminal
2-32
1-11 loop
61-75
11-111 loop
96-1 15
III-IV loop
141-153
IV-v loop
174-203
AMINO ACID SEQUENCE
Asn-Gly-Thr-Glu-Gly-Pro-Asn-Phe-Tyr-Val-Pro-Phe-Ser-Asn-Lys-Thr-Gly-V~-Val-ArgSer-Pro-Phe-Glu-Ala-Pro-Gln-Tyr-Tyr-LeuAla-
Lys-Pro-Met-Ser-Asn-Phe-Arg-Phe-Gly-Glu-Asn-His-Ala
Gly-Trp-Ser-Arg-Tyr-Ile-Pro-Glu-Gly-Met-Gln-Gly-Ile-Asp -Tyr-Tyr-Thr-Pro-His-Glu-GluThr-Asn-Asn-Glu-Ser-Phe
v-VI loop
230-252
Val-Lys-Glu-Ala-Ala-Ala-Gln-Gln-Gln-Glu-Ser-Ala-Thr-~r-Gln-Lys-~a-Glu-Lys-Glu-V~Thr-Arg
VI-VII loop
276-286
Phe-Thr-His-Gln-Gly-SerAsp-Phe-Gly-Pro-Ile
C-terminal
3 10-321
Asn-Ly s-Gln-Phe-Arg-Asn-Cys-Met-Val-Thr-Thr-Leu
C-terminal
324-348
Gly-Lys-Asn-Pro-Leu-Gly-Asp-Asp-Glu-Ala-Ser-Thr-Thr-Val-Ser-Lys-Thr-Glu-~r-Ser-GlnVal-Ala-Pro-Ala
C-terminal
33 1-342
s-Thr-Glu-Thr Asp-Glu-Ala-Ser-Thr-Thr-Val-Ser-Ly
Research. The eyes and pineal glands were removed and
Antigens
fixed in 10% buffered neutral formaldehyde solution and
Disc membrane bound rhodopsin was prepared from bovine
submitted to routine histological examination. Histological
retinas by Ficoll flotation (20). Rhodopsin was prepared
grading of EAU was from 0 (no EAU) - 5+ (most severe
from rod outer segments and purified by chromatography on
EAU) as follows: (1 +) Minimal signs of inflammation,
concanavalin A-Sepharose (21). Rhodopsin's synthetic
mild perivascular cuffing of some retinal vessels.
peptides representing exposed, hydrophilic regions (Table 1)
(2+) Moderate number of inflammatory cells infiltration of
were synthesized by the methods published previously (22).
the photoreceptor layer of the retina and infiltration of the
ELISA
vitreous cortex. (3+) Infiltration of the photoreceptor cell
The ELISA was performed in PVC microtiter plates
layer, outer nuclear layer, inner nuclear layer of retina;
(Falcon) coated overnight with Con-A purified rhodopsin,
substantial retinal detachment. (4 +) Full thickness
or synthetic peptide (1 pg/100 pl per well) in 0.1 M Tris-
involvement of the retina, complete infiltration, ujlal retinal
HC1, pH 9.0. Nonspecific binding was blocked by
detachment. (5+) Complete replacement of the retina by
incubation with 200 p1 1% BSA in PBS. 100 pl of serial
inflammatory cells and sequelae, loss of any anatomical
two-fold dilutions of heat-inactivated serum (56"C, 30 min)
structure of retina.
in PBS were incubated for 1 h at 4°C followed by
659
Curr Eye Res Downloaded from informahealthcare.com by University of Melbourne on 11/23/14 For personal use only.
Current Eye Research
660
Curr Eye Res Downloaded from informahealthcare.com by University of Melbourne on 11/23/14 For personal use only.
Current Eye Research
Figure 2. Representative micrographs from eyes of Lewis rats immunized with rhodopsin pathogenic peptides. Eyes were removed 28 days post immunization. (A) Normal retina from rats immunized with saline instead of antigen. Destruction of the photoreceptor cell layer following immunization with (€4) peptide 230-252 (loop V-VI) and (C) peptide 324-348. Note infiltration of retina with
inflammatory cells in (B) and the complete loss of the photoreceptor cell layer with diminishing inflammatory infiltrate in (C). The outer nuclear layer rests on the pigment epithelium. G - ganglion cell layer, B - bipolar cell layer, P - photoreceptor cell layer including the nuclear and outer segment layers , C - choroid.
incubation with 100 p1 of lo00 times diluted conjugate of
lo5 /200 pl were cultured with a stimulant in RPMI medium
goat anti-rat IgG (heavy and light chain) - peroxidase
containing 2 mM HEPES, 5 x
(Sigma). Color reaction was developed by adding 200 p1
mM L-glutamine, 2 % normal rat serum, 3 % CPSR-2
peroxidase substrate (0-phenylenediamine and hydrogen
(Sigma), gentamicin (50 pg/ml) and fungizone (2.5 pg/ml).
peroxide) and measured at 490 nm using Bio-Tek EIA
The cells were incubated in triplicate for 72 hr at 37°C in
Reader EL 3 10.
flat bottomed microtiter plates with different concentrations
LymDhocvte proliferation assay
of membrane-bound rhodopsin (0.125 pM) or synthetic
T cells were purified from spleen cells of immunized
peptide (2-80 pM). Proliferative responses of different doses
animals by lysis of red cells with 0.83%ammonium
of rhodopsin and peptides were assayed by measuring
chloride in 0.01 M Tris-HCI, pH 7.5, followed by passage
incorporation of 1 pCi [’HI-thymidine in 25 pl added 18 h
over a nylon wool (Polyscience) column to remove adherent
prior to cell harvesting. Results are presented as stimulation
B cells. Nylon wool non-adherent T cells at a density of 2 x
indices (SI) calculated as mean cpm of cultures with
M 2-mercapthoethanol, 2
661
Current Eye Research Table 2. Immunogenic and immunopathogenic determinants in rhodopsin ANTIGEN PEPTIDE'
STIMULATION'* PEPTIDE RHODOPSIN ~
Curr Eye Res Downloaded from informahealthcare.com by University of Melbourne on 11/23/14 For personal use only.
2-32 61-75 96-1 14 141-153 174-203 230-252 276-286 3 10-32 1 324-348
13.7f1.9 3.0k0.6 2.5f0.4 1.6f0.8 2.0f1.2 9.9f0.3 1.5f0.1 2.5f0.8 10.8f2.7
0.5f0.3 2.9k0.7 0.9f0.3 0.8k0.3 1.1 f0.5 1.4f0.6 1.0f0.1 0.8f0.5 3.4f1.6
EAU INCIDENCE ~
013 213 013
013 013 415 013 013 617
SEVERITY -
~~
0 1 k0.5 0 0 0 2.4f1.1 0 0 0.9f0.2
Data are presented as stimulation indices (SI). Stimulation was considered positive if the SI of injected rats was equal to or greater than twice the background (SI=2). Unstimulated cultures of cells from these animals incorporated 1054-2180 cpm. Rats were injected with 100 pg peptide (- 30-80 nmollrat). Peptides were tested at 20 pM and rhodopsin at 0.125 pM.
'
stimulant divided by mean cpm of the unstimulated control
carboxyl-terminal peptides and the 1-11 loop, very few
culture. Stimulation was considered positive if the SI of
inflammatory cells were present in the retina. Some areas
injected rats was equal to or greater than twice the
were infiltrated with cells, particularly near retinal vessels.
background (SI=2).
In general, the inflammatory changes in eyes of animals injected with loop 1-11 and the C-terminus injected rats
RESULTS
observed were less severe than those observed in eyes from
Induction of experimental autoimmune uveitis
animals immunized with rhodopsin (Table 2). The 12-
To determine the ability of rhodopsin peptides to induce
residue carboxyl terminal peptide, 331-342, was as active as
uveitis, Lewis rats were immunized with synthetic peptides
the 25-residue peptide, 324-348. None of the other peptides
and tested for antibody responses, T cell proliferative
tested (Table 1) induced significant inflammatory disease.
responses and pathological changes in the retina. Peptides
Inflammation of the anterior chamber was not detected. We
representing the exposed, hydrophilic regions of rhodopsin
did not observe inflammatory changes in the pineal gland
were selected for immunization (Table 1). The doses of
from animals immunized with any of rhodopsin's synthetic
injected peptides were 10 - 30 times higher than the molar
peptides.
equivalent of 100 pg rhodopsin. Only a few peptides were
Cellular resporises
capable of inducing EAU. Peptides from the C-terminal
The cellular immune responses were determined by the in
(324-348) and from loop 1-11 (61-75) were both capable of
vitro lymphocyte proliferative assay of T cells to different
inducing disease. However, the most immunopathogenic
doses of the appropriate synthetic peptides or rhodopsin in
peptide was sequence 230-25 1 (V-VI loop).
culture. Injection of two peptides from the C-terminus of
Histopathologically, at the immunized doses of 16-28
rhodopsin (324-348 and its truncated form 331-342),
nmol per animal, we observed total destruction of the
peptides representing loop sequences, and the amino-
photoreceptor cell layer when compared to the control
terminus induced different levels of T cell responses but
group that did not receive the antigen. Loop V-VI caused
only at higher doses than rhodopsin ('10-30 times higher on
severe inflammation of the retina (Fig. 2). In the case of the
a molar basis). Positive proliferative responses against the
662
Current Eye Research Table 3. Stimulation of spleen T cells from Lewis rats immunized with peptide 331-342 and its pathogenicity
324-34 S.I.'
INJECTED DOSE nmollrat
Curr Eye Res Downloaded from informahealthcare.com by University of Melbourne on 11/23/14 For personal use only.
0 21 42 84 168 336
1.8 f O . l 3.55k0.2 9.2 k2.5 7.0 f l . O 2.6 k0.6 1.8 f0 . 3
331-342 S.I.
EAU INCIDENCE
1.7 f0.2 1.45 f0.5 6.0 f0.7 5.5 f l . O 2.8 f0.5 4.6 k1.2
SEVERITY
0 0 0.9k0.2 0.9k0.2 0.4k0.35 0.4k0.4
013 012 213 213 113 113
* Data are presented as stimulation indices (SI). Stimulation was considered positive if the SI of injected rats was equal to or greater than twice the background (SI=2). Unstimulated cultures of cells from these animals incorporated 214-1817 cpm. Peptides were tested at 20 pM.
respective immunizing peptides were obtained with doses
protein, rhodopsin.
ranging from 10-20 pM.
To determine an optimal immunizing dose of
Table 2 summarizes results obtained from Lewis rats
synthetic peptide and to correlate it with pathogenicity, rats
immunized with 100 pg (-30-80 nmol/rat) of various
were injected with two peptides representing the sequences
synthetic peptides. The capacity of these peptides to elicit in
previously identified as the major immunodominant epitope.
vitro lymphocyte response is shown. Cells isolated from
The minimal immunogenic and immunopathogenicdose of
Lewis rats injected with any of the following peptides: 2-32
peptide 331-342 in Lewis rats was 42 nmole of peptide.
(N-terminus), 61-75 (loop 1-11), 230-252 (loop V-VI), or
Both carboxyl-terminal peptides (331-342 and 324-348)
324-348 (C-terminus), responded in v i m to that same
were capable of eliciting strong in vidro lymphocyte
peptide. In all cases, the proliferative response against the
proliferation responses in cells derived from rats injected
immunizing peptide was stronger than that against the intact
with peptide 331-342 (Table 3). Peptide 61-75 (loop 1-11)
Table 4. Stimulation of spleen T cells from Lewis rats immunized with peptide 61-75 and its pathogenicity INJECTED DOSE nrnollrat
6 1-75 S.I.'
EAU INCIDENCE
SEVERITY ~
0 21 42 84 168 336
0.7 f0.2 1.0 k0.4 1.8 k1.1 3.0 f0.6 1.85k 1.O 3.6 k0.3
013 113 113 313 213 113
~
~~
0 0.5k0.4 0.5*0.8
1.0k0.4 0.6 k0.25
0.51f0.8
* Data are presented as stimulation indices (SI). Stimulation was considered positive if the SI of injected rats was equal to or greater than twice the background (SI=2). Unstimulated cultures of cells from these animals incorporated 567-1895 cpm. Peptides were tested at 20 p M . 663
Current Eye Research induced overall higher antibody response in animals immunized with the whole molecule (19). Control animals
3
that received saline instead of antigen developed no antibody 0
m
*
2
response.
1
DISCUSSION
a 0
Rhodopsin has been shown to be uveitogenic in susceptible ,
0
Curr Eye Res Downloaded from informahealthcare.com by University of Melbourne on 11/23/14 For personal use only.
100
m-
1000
ANTISERUM DILUTION
strains of animals such as Lewis rats, guinea pigs and primates (12-19). Although T cell recognition sites for rhodopsin have been identified (19), until now, regions of
Figure 3. Anti-rhodopsin antibody titer in antisera of Lewis rats injected with rhodopsin synthetic peptides. Titer of specific antibody was determined by ELISA using microtiter plates coated with purified bovine rhodopsin. Each point is the mean of three separate animals. Preimmune control sera for all tested animals had an average ODdwat 1:20 dilution of 0.75 units. Animals injected with saline instead of peptide did not generated specific antibodies; level of activity was typical of those measured in preimmune sera. (0)2-32; (0)61-75; (m) 96-115; (v) 141-153; ( T ) 174203; ( 0 ) 230-252; ( + ) 276-286; (A) 310-321; (A) 324-348.
the molecule responsible for EAU induction have not been determined. Knowledge of the amino acid sequence of rhodopsin was essential for characterization of immunodominant and pathogenic sites for T cells. Synthetic peptides have been used to define several epitopes on rhodopsin that induced T cell responses in Lewis rats (19). The majority of T cells from animals primed with rhodopsin were directed against the carboxyl terminus which was determined to be a major T cell epitope. Other minor
stimulated weaker proliferative responses in cells derived
epitopes were found in the 11-111 (96-114) and IV-V
from rats injected with peptide 61-75. Representative data
(174-202) loops (see Fig. 1). The first attempt to use a
are summarized in Table 4.
rhodopsin peptide to induce EAU, injection of rhodopsin’s
Antibody responses
amino-terminal CNBr peptide into guinea pigs, did not
Antibody titers were measured against a synthetic antigen
result in disease development (12). In our studies, the major
(immunizing peptide) and native antigen (rhodopsin) in
T cell proliferative site for rhodopsin, sequence 324-348,
serum of animals injected with peptide. The ELISA results
was selected for testing pathogenic properties. Direct
indicate that all the peptides induced antibody responses.
immunization with that peptide stimulates a strong
The antibodies recognized both the appropriate synthetic
proliferative response that is crossreactive with the native
peptide and rhodopsin. However, antibody titers were
protein. Whereas this peptide seems to contain a uveitogenic
different among sera tested. Fig. 3 shows titration curves of
site, not all of rhodopsin’s peptides which elicited immune
rat antisera against rhodopsin. In general, peptides that
responses were immunopathogenic (Table 2). Immunization
represent major antibody binding sites (N-terminus and
of animals with rhodopsin peptides resulted in the display of
174-203), were better immunogens than were other
some different T cell specificities than immunization with
peptides. The highest level of specific antibodies was
the whole antigen. Moreover, immunization with peptides
observed after immunization with the amino-terminal
induces specificities that are not present after rhodopsin
peptide (sequence 2-32). Also immunization with the loop
immunization e.g., T cells specific against residues 230-251
IV-V (174-203) produced good antibody titers. Although
and 2-32. A different repertoire of T cell specificities
rats injected with some of the peptides generated good anti-
induced by peptides and whole protein has been found in the
rhodopsin and anti-peptide antibody titers, rhodopsin
immune response to other proteins (e.g. P-galactosidase (23)
664
Curr Eye Res Downloaded from informahealthcare.com by University of Melbourne on 11/23/14 For personal use only.
Current Eye Research and lysozyme (24)) and a hierarchy of dominant usage in a
vitro, and induce pathogenicity. Its truncated form, residues
multidominant protein has been suggested. Our studies show
331-342, also possessed these properties. Responses to
only two peptides that produced strong in vitro proliferative
rhodopsin’s minor determinants were not always stimulated
responses of T cells, also provoked pathology (230-252 and
by direct immunization with the peptide and were not
324-348). One of these peptides contains a major T cell
capable of provoking EAU except for one, loop 1-11
epitope (324-348). The third pathogenic site is located in the
(61-75). Peptide 230-252 was capable of inducing EAU,
region of loop 1-11 (61-75) which stimulates only mild T-cell
however only in a few cases were cells from rhodopsin-
proliferation. We do not exclude the possibility of presence
injected rats stimulated by that peptide. This determinant
of other immunopathogenic regions of rhodopsin because
could fit to a category of cryptic determinants in which only
the pathogenic doses of peptide were about 20 times higher
injection of the peptide induces a response to itself.
than the protein itself. There may be intramembrane
However, the topography studies of rhodopsin revealed that
portions of the molecule which may be immunogenic
this region of that protein is greatly exposed and accessible.
/pathogenic but which were not tested. It is interesting that
It has been shown that membrane-bound-rhodopsin is
the sequences of all three uveitogenic sites are in the
susceptible to proteolysis by a number of proteases in only a
cytoplasmic, conserved regions of the bovine rhodopsin
few selected regions of the molecule which includes loop V-
structure as compared to mouse and human. Within two
VI (30). It is therefore possible that proteolytic degradation
loop sequences, no amino acid substitutions have been
of rhodopsin which takes place in vicro, produces fragments
reported (25-27). In addition, the carboxyl-terminus
that are unable to sensitize T cells (abolish MHC binding or
(324-348) of bovine rhodopsin differs from mouse and
T cell binding). Although the range of proteases involved in
human rhodopsin by as little as a single amino acid residue.
antigen processing is unknown, it is reasonable to speculate
The uveitogenic (pathogenic) determinants for rhodopsin were defined in regions of major and minor determinants. If autoimmune disease results from a
that proteolytic attack affects the highly exposed surface loop region, sequence 230-251. Rhodopsin peptide 324-348 is not the only
breakdown in self-tolerance, there may be some T cell
immunodominant peptide that has been found to be involved
epitopes that escape the process of tolerance induction (28).
in disease induction. Studies on EAU have revealed
In the model proposed by Gammon and Sercarz (29)
immunodominance of certain fragments of other retinal
tolerance to many self-proteins will not be complete but
autoantigens such a S-antigen (31) and IRBP (7, 9) in
restricted to the major T cell-inducing determinants. The
disease induction. Similarly, in EAE (experimental
pathogenic determinants on self-antigens are not typical
autoimmune encephalomyelitis), which is often compared
major T cell-inducing determinants, but resemble minor
with EAU, immunodominant peptides have been identified
ones. The minor epitopes are available only in relatively
in myelin basic protein (MBP) involved in the disease
low amounts after in vivo processing of the entire molecule.
initiation (28).
These T cells are not normally activated and responses to poorly presented minor epitopes can be stimulated only
ACKNOWLEDGEMENTS
under special circumstances to avoid tolerance. In
Supported in part by NIH grants EY 06225 and EY 06226
rhodopsin-induced EAU however, an immunodominant site
to P.A.H., Core facilities grant EY08571, and an
of rhodopsin, the carboxyl terminal peptide, was capable of
unrestricted departmental grant from Research to Prevent
inducing the immune system to produce sensitized T cells
Blindness, Inc. to the University of Florida Department of
and antibody, stimulate good proliferative responses in
Ophthalmology.
665
Current Eye Research CORRESPONDING AUTHOR Grazyna Adamus, Department of Ophthalmology, Box 100284, University of Florida, Gainesville, F1 32610,
11.
U.S.A.
Immunopathogenic determinants are not necessarily immunodominant. Clin. Immunol. Immunopathol. 3,212-224. Vrabec, T. R., Reber, R. N., Magargal, L. E. and Donoso, L. A. (1990) S-Antigen. Identification of human T-lymphocyte proliferation sites. Arch. Ophthalmol. 1470-1473. Marak, G. E., Shichi, H., Rao, N. A. and Wacker, W. B. (1980) Pattern of experimental allergic uveitis induced by rhodopsin and retinal rod outer segments. Ophthalmic Res. J2,165-176. Meyers-Elliott, R. H., Gammon, R. A., Sumner, H. L. and Shimizu, I. (1983) Experimental retinal autoimmunity (ERA) in strain 13 guinea pig: induction of ERA-retinopathy with rhodopsin. Clin. Immunol. Immunopathol. 2, 81-95. Meyers-Elliott, R. H. and Sumner, H. L. (1982) Experimental uveitis induced by products of activated lymphocytes: intraccular effects of rhodopsin-induced lymphokines. Cell. Immunol. &, 240-253. Broekhuyse, R. M., Winkens, H. J., Kuhlmann, E. D. and VanVugt, A. H. M. (1984) Opsin-induced experimental autoimmune retinitis in rats. Curr. Eye Res. 3, 1405-1412. Schalken, J. J . , Van Vugh, A. H. M.,Winkens, H. J., Bovee-Geurts, P. H. M., De Grip, W. J. and Broekhuyse, R. M. (1988) Experimental autoimmune uveoretinitis in rats induced by rod visual pigment: rhodopsin is more pathogenic than opsin. Graefe's Arch. Clin. Exp. Ophthalmol. 226, 255-261. Schalken, J. J., Winkens, H. J., Van Vugt, A. H. M., Bovee-Geurst, P. H. M., De Grip, W. J. and Broekhuyse, R.M. (1988) Rhodopsin-induced experimental autoimmune uveoretinitis: dose-dependent clinicopathological features. Exp. Eye Res. 47, 135-145. Schalken, J. J., Winkens, H. J., Van Vugt, A. H. M., De Grip, W. J. and Broekhuyse, R. M. (1989) Rhodopsin-induced experimental autoimmune uveoretinitis in monkeys. Br. J. Ophthalmol. 168-172. Moticka, E. J. and Adamus, G. (1991) Specificity of T and B cell responses to bovine rhodopsin in Lewis rat. Cell. Immunol. U, 175-184. Smith, H. G., Stubbs, G.W. and Litman, B.J. (1975) The isolation and purification of intact discs from retinal rod outer segments. Exp. Eye Res. 211-217. Litman, B. J. (1982) Purification of rhodopsin by concanavalin A affinity chromatography. In "Methods in Enzymology" (Ed. Parker, L.) 81, Pp. 150-153 Academic Press, San Diego. Adamus, G., Zam, Z.S., Arendt, A . , Palczewski, K. McDowell, J.H. and Hargrave, P.A. (1991) Antirhodopsin monoclonal antibodies of defined specificity: Characterization and application. Vision Res. 17-31.
Curr Eye Res Downloaded from informahealthcare.com by University of Melbourne on 11/23/14 For personal use only.
m,
REFERENCES Gery, I., Mochizuki, M. and Nussenblatt, R. B. 1. (1986) Retinal specific antigens and immunopathogenic processes they provoke. Prog . Ret. Res. 5, 75-109. Donoso, L. A., Merryman, C. F., Sery, T. W., 2. Shinohara, T., Dietzschold, B., Smith, A. and Kalsow, C. M. (1987) S-Antigen: characterization of a pathogenic epitope which mediates experimental autoimmune uveitis and pinealitis in Lewis rat. Curr. Eye Res. 9, 1151-1159. Donoso, L. A . , Merryman, C. F., Sery, T. W., 3. Vrabec, T., Arbizo, V. and Fong, S.-L. (1988) Human IRBP: characterization of uveitopathogenic sites. Curr. Eye Res. 2, 1087-1095. Donoso, L. A., Yamaki, K., Merryman, C. F . , 4. Shinohara, T., Yue, S . and Sery, T. W. (1988) Human S-Antigen: characterization of uveitopathogenic sites. Curr. Eye Res. 2, 1077-1085. 5. Singh, V. K . , Donoso, L. A., Yamaki, K. and Shinohara, T. (1989) Uveitopathogenic sites in bovine S-antigen. Autoimmunity, 3, 177-187. 6. Gregerson, D. S . , Fling, S . P. and Wohlhueter, R. M. (1986) Characterization of immunologically active cyanogen bromide peptide fragments of bovine and human retinal S-antigen. Exp. Eye Res. 43, 803-818. 7. Kotake, S . , DeSmet, M. D., Wiggert, B., Redmond, T. M., Chader, G. J. and Gery, I. (1991) Analysis of the pivotal residues of the immunodominant and highly uveitogenic determinant of interphotoreceptor retinoid-binding protein. J. Immunol. 146, 2995-3001. Kotake, S . , Redmond, T. M., Wiggert, B., Vistica, 8. B., Sanui, H., Chader, G. J. and Gery, I. (1991) Unusual immunological properties of the uveitogenic interphotoreceptor retinoid-binding protein-derived protein R23. Invest. Ophthalmol. Vis. Sci. Z, 2058-2064. Kotake, S . , Wiggert, B., Redmond, T. M., Borst, 9. D. E., Nickerson, J. M., Margalit, H., Berzofsky, J. A., Chader, G. J. and Gery, I. (1990) Repeated determinants within the retinal interphotoreceptor retinoid-binding protein (IRBP): immunological properties of the repeats of an immunodominant determinant. Cell. Immunol. 2, 331-342, 10 Redmond, T. M., Sanui, H., Hu, L. H., Wiggert, B., Margalit, H., Berzofsky, J. A., Chader, G. J. and Gery, I. (1989) Immune responses to peptides derived from the retinal protein IRBP:
666
12.
13.
14.
15.
16.
17.
18.
a,
19. 20.
a,
21.
22.
a,
Current Eye Research ~~
23.
24.
25.
Shivakumar, S., Sercarz, E.E. and Krzych, U. (1989)The molecular context of determinants within the priming antigen establishes a hierarchy of T cell induction: T cell specificities induced by peptides of 0-galactoside vs. the whole antigen. Eur. J. Immunol. L 2 ,681-687. Shastri, N., Miller, A. and Serwz, E.E. (1984)The expressed T cell repertoire is hierarchical: the precise focus of lysozyme-specific T cell clones is dependent upon the structure of the immunogen. J. Mol. Cell. Immunol. 1,369-379. Baehr, W., Falk, J.D., Bugra, K., Triantafullos, J.T. and McGinnis, J.F. (1988)Isolation and analysis of the mouse opsin gene. FEBS Lett. 253-256. Nathans, J. and Hogness, D.S.(1984) Isolation and nucleotide sequence of the gene encoding human rhodopsin. Proc. Natl. Acad. Sci. USA. 81, 48514855. Hargrave, P.A., McDowell, J.H., Curtis D.R., Wang, J.K., Juszczak, E., Fong, S-L., Rao, J.K.M. and Argos, P. (1983)The structure of bovine rhodopsin. Biophys. Struct. Mech. 8, 235-244. Wraith, D.C.,McDevitt, H.O., Steinman, L. and Acha-Orbea, H. (1989)T cell recognition as the target for immune intervention in autoimmune disease. Cell, 52,709-715. Gammon, G.and Sercarz, E. (1989) How some T cells escape tolerance induction. Nature, 342, 183185. Hargrave, P.A. (1982)Rhodopsin chemistry, structure and topography. Prog. Ret. Res. 1, 1-51. Fling, S.P.,Donoso, L.A., Gregerson, D.S. (1991) In vitro Unresponsiveness to autologous sequences of the immunopathogenic autoantigen, S-antigen. J. Immunol. 147, 483-489.
Curr Eye Res Downloaded from informahealthcare.com by University of Melbourne on 11/23/14 For personal use only.
m,
26.
27.
28.
29. 30. 31.
667
Volume 11 number 7 1992, 657-667 ~~
~
___~__
___
Induction of experimental autoimmune uveitis with rhodopsin synthetic peptides in Lewis rats Grazyna Adamus', Jacki L.Schmied', Paul A.Hargrave1,2,Anatol Arendt' and Edward J.Moticka3
Curr Eye Res Downloaded from informahealthcare.com by University of Melbourne on 11/23/14 For personal use only.
Departments of 10phthalmology, 2Biochemistryand Molecular Biology, University of Florida, Gainesville, FL and 'Department of Medical Microbiology and Immunology, Southern Illinois University School of Medicine, Springfield, IL, USA ABSTRACT Rhodopsin, a membrane protein of rod photoreceptor cells, induces an experimental autoimmune uveitis (EAU) in Lewis rats. Synthetic peptides derived from rhodopsin sequences that cover hydrophilic, exposed regions of the protein were tested for their capacity of eliciting in vitro T cell proliferation and their ability for inducing EAU in Lewis rats. Rats were injected with rhodopsin's peptides mixed in complete Freund's adjuvant containing M. tuberculosis H3zRa (5 mg/ml) three days after pretreatment with cyclophosphamide (20 mg/kg). ELISA results indicate that all peptides induce antibody responses; however antibody titers differ among sera tested. Immunization with four pephdes - the amino-terminus (2-32), loop 1-11(61-75), loop V-VI (230-251), and the carboxyl-terminus (324-348 and 331-342) induced both antibody and T cell responses. In ail cases, the proliferative responses of cells derived from peptide-injected rats were stronger against the immunizing peptide than against native protein. Three distinct uveitogenic epitopes were identified on rhodopsin's cytoplasmic surface - within the rhodopsin carboxylterminus (324-348), loop 1-11 (61-75), and loop V-VI (230-250). Histopathologically, at the immunized doses, total destruction of the photoreceptor cell layer was observed as compared to the control group. Loop V-VI caused severe inflammation of the retina while the other pathogenic peptides produced less severe destruction with few inflammatory cells present. Our study indicates that the major immunodominant T cell epitope (331-342) is also involved in EAU induction but is not the primary uveitogenic site.
inflammation of the uveal track, retina and pineal gland (1). It has been shown that specific regions of uveitogenic proteins are responsible for their pathogenicity (2-5). Several T cell epitopes for S-antigen and IRBP responsible for uveitogenic, proliferative, and adoptive transfer responses have been characterized with the use of CNBr fragments and synthetic peptides (6-11). Rhodopsin, an integral membrane glycoprotein, has been shown to induce EAU in guinea pigs, rats, and monkeys (12-19). In a recent study, we determined the specificity of T and B cell responses to bovine rhodopsin (19). T cells isolated from rhodopsin-injected rats responded
in vitro to determinants present in the carboxyl-terminus and in loops 1-11 (61-72), II-III (96-115) and IV-V (174-202)
(see topographic diagram of rhodopsin, Fig. 1). The major T cell determinant has been determined to be within the sequence 324-348. Using a series of 12 amino acid long overlapping peptides, the precise position of this epitope was localized to the sequence 331-342 (19). On the other hand, loop IV-V and the amino-terminus have been shown to be immunodominant sites for antibody binding.
To investigate the immunoregulatory mechanisms INTRODUCTION
responsible for controlling rhodopsin autoimmunoreactivity,
Experimental autoimmune uveitis is an ocular inflammatory
it is necessary to identify the immunopathogenicepitopes in
disease in susceptible animals induced by immunization with
this protein. In this report, we examined the ability of
retinal antigens such as rhodopsin, S-antigen (also known as
rhodopsin peptides to induce EAU in Lewis rats. We have
arrestin, and 48K protein), and interphotoreceptor retinoid
used rhodopsin synthetic peptides that represent hydrophilic,
binding protein (IRBP). This disease has been characterized
exposed regions of the protein to immunize animals. The
in part as a T cell-dependent disease that results in severe
immunogenicity of the peptides was correlated with
Received on January 27, 1992; accepted on June 16, 1992
657
Current Eye Research T CEL& EPlTOF'E (331-342)
Curr Eye Res Downloaded from informahealthcare.com by University of Melbourne on 11/23/14 For personal use only.
U
T CELL EPU'OPE
I
B CELL E~TTOPE .
(96-115)
(2-32)
Figure 1. A topographic model of rhodopsin in the lipid bilayer of the disc membrane of retinal rod cells (Adapted from Hargrave et. al [26]). AKOWSindicate regions of
rhodopsin that are found to represent T cell epitopes, B cell epitopes, and uveitogenic sites.
pathogenesis of the disease to determine which T-cell
days later, they were immunized in the hind footpad with a
epitopes of rhodopsin are possibly involved in the disease
single dose of peptide mixed 1:1 in complete Freund's
induction.
adjuvant with Mycobactenwn tuberculosis H37Ra (5 mg/rnl) in a total volume of 100 pl. Control animals received saline
MATERIALS AND METHODS
instead of peptide (negative control) or 100 pg rhodopsin
toimmune uveitis
(positive control) using above protocol. 25-28days after
Female Lewis rats were purchased from Charles River
immunization, the animals were killed, blood was collected
Laboratories. Groups of 3-4 rats (150-180g) were injected
and spleens were removed aseptically. All procedures
with cyclophosphamide (20 mg/kg) intrapentoneally. Three
adhered to the ARVO Resolution on the Use of Animals in
658
Current Eye Research
Curr Eye Res Downloaded from informahealthcare.com by University of Melbourne on 11/23/14 For personal use only.
Table 1. Synthetic peptides from bovine rhodopsin sequence used for immunization. RHODOPSIN REGION
RESIDUES NUMBERS
N-terminal
2-32
1-11 loop
61-75
11-111 loop
96-1 15
III-IV loop
141-153
IV-v loop
174-203
AMINO ACID SEQUENCE
Asn-Gly-Thr-Glu-Gly-Pro-Asn-Phe-Tyr-Val-Pro-Phe-Ser-Asn-Lys-Thr-Gly-V~-Val-ArgSer-Pro-Phe-Glu-Ala-Pro-Gln-Tyr-Tyr-LeuAla-
Lys-Pro-Met-Ser-Asn-Phe-Arg-Phe-Gly-Glu-Asn-His-Ala
Gly-Trp-Ser-Arg-Tyr-Ile-Pro-Glu-Gly-Met-Gln-Gly-Ile-Asp -Tyr-Tyr-Thr-Pro-His-Glu-GluThr-Asn-Asn-Glu-Ser-Phe
v-VI loop
230-252
Val-Lys-Glu-Ala-Ala-Ala-Gln-Gln-Gln-Glu-Ser-Ala-Thr-~r-Gln-Lys-~a-Glu-Lys-Glu-V~Thr-Arg
VI-VII loop
276-286
Phe-Thr-His-Gln-Gly-SerAsp-Phe-Gly-Pro-Ile
C-terminal
3 10-321
Asn-Ly s-Gln-Phe-Arg-Asn-Cys-Met-Val-Thr-Thr-Leu
C-terminal
324-348
Gly-Lys-Asn-Pro-Leu-Gly-Asp-Asp-Glu-Ala-Ser-Thr-Thr-Val-Ser-Lys-Thr-Glu-~r-Ser-GlnVal-Ala-Pro-Ala
C-terminal
33 1-342
s-Thr-Glu-Thr Asp-Glu-Ala-Ser-Thr-Thr-Val-Ser-Ly
Research. The eyes and pineal glands were removed and
Antigens
fixed in 10% buffered neutral formaldehyde solution and
Disc membrane bound rhodopsin was prepared from bovine
submitted to routine histological examination. Histological
retinas by Ficoll flotation (20). Rhodopsin was prepared
grading of EAU was from 0 (no EAU) - 5+ (most severe
from rod outer segments and purified by chromatography on
EAU) as follows: (1 +) Minimal signs of inflammation,
concanavalin A-Sepharose (21). Rhodopsin's synthetic
mild perivascular cuffing of some retinal vessels.
peptides representing exposed, hydrophilic regions (Table 1)
(2+) Moderate number of inflammatory cells infiltration of
were synthesized by the methods published previously (22).
the photoreceptor layer of the retina and infiltration of the
ELISA
vitreous cortex. (3+) Infiltration of the photoreceptor cell
The ELISA was performed in PVC microtiter plates
layer, outer nuclear layer, inner nuclear layer of retina;
(Falcon) coated overnight with Con-A purified rhodopsin,
substantial retinal detachment. (4 +) Full thickness
or synthetic peptide (1 pg/100 pl per well) in 0.1 M Tris-
involvement of the retina, complete infiltration, ujlal retinal
HC1, pH 9.0. Nonspecific binding was blocked by
detachment. (5+) Complete replacement of the retina by
incubation with 200 p1 1% BSA in PBS. 100 pl of serial
inflammatory cells and sequelae, loss of any anatomical
two-fold dilutions of heat-inactivated serum (56"C, 30 min)
structure of retina.
in PBS were incubated for 1 h at 4°C followed by
659
Curr Eye Res Downloaded from informahealthcare.com by University of Melbourne on 11/23/14 For personal use only.
Current Eye Research
660
Curr Eye Res Downloaded from informahealthcare.com by University of Melbourne on 11/23/14 For personal use only.
Current Eye Research
Figure 2. Representative micrographs from eyes of Lewis rats immunized with rhodopsin pathogenic peptides. Eyes were removed 28 days post immunization. (A) Normal retina from rats immunized with saline instead of antigen. Destruction of the photoreceptor cell layer following immunization with (€4) peptide 230-252 (loop V-VI) and (C) peptide 324-348. Note infiltration of retina with
inflammatory cells in (B) and the complete loss of the photoreceptor cell layer with diminishing inflammatory infiltrate in (C). The outer nuclear layer rests on the pigment epithelium. G - ganglion cell layer, B - bipolar cell layer, P - photoreceptor cell layer including the nuclear and outer segment layers , C - choroid.
incubation with 100 p1 of lo00 times diluted conjugate of
lo5 /200 pl were cultured with a stimulant in RPMI medium
goat anti-rat IgG (heavy and light chain) - peroxidase
containing 2 mM HEPES, 5 x
(Sigma). Color reaction was developed by adding 200 p1
mM L-glutamine, 2 % normal rat serum, 3 % CPSR-2
peroxidase substrate (0-phenylenediamine and hydrogen
(Sigma), gentamicin (50 pg/ml) and fungizone (2.5 pg/ml).
peroxide) and measured at 490 nm using Bio-Tek EIA
The cells were incubated in triplicate for 72 hr at 37°C in
Reader EL 3 10.
flat bottomed microtiter plates with different concentrations
LymDhocvte proliferation assay
of membrane-bound rhodopsin (0.125 pM) or synthetic
T cells were purified from spleen cells of immunized
peptide (2-80 pM). Proliferative responses of different doses
animals by lysis of red cells with 0.83%ammonium
of rhodopsin and peptides were assayed by measuring
chloride in 0.01 M Tris-HCI, pH 7.5, followed by passage
incorporation of 1 pCi [’HI-thymidine in 25 pl added 18 h
over a nylon wool (Polyscience) column to remove adherent
prior to cell harvesting. Results are presented as stimulation
B cells. Nylon wool non-adherent T cells at a density of 2 x
indices (SI) calculated as mean cpm of cultures with
M 2-mercapthoethanol, 2
661
Current Eye Research Table 2. Immunogenic and immunopathogenic determinants in rhodopsin ANTIGEN PEPTIDE'
STIMULATION'* PEPTIDE RHODOPSIN ~
Curr Eye Res Downloaded from informahealthcare.com by University of Melbourne on 11/23/14 For personal use only.
2-32 61-75 96-1 14 141-153 174-203 230-252 276-286 3 10-32 1 324-348
13.7f1.9 3.0k0.6 2.5f0.4 1.6f0.8 2.0f1.2 9.9f0.3 1.5f0.1 2.5f0.8 10.8f2.7
0.5f0.3 2.9k0.7 0.9f0.3 0.8k0.3 1.1 f0.5 1.4f0.6 1.0f0.1 0.8f0.5 3.4f1.6
EAU INCIDENCE ~
013 213 013
013 013 415 013 013 617
SEVERITY -
~~
0 1 k0.5 0 0 0 2.4f1.1 0 0 0.9f0.2
Data are presented as stimulation indices (SI). Stimulation was considered positive if the SI of injected rats was equal to or greater than twice the background (SI=2). Unstimulated cultures of cells from these animals incorporated 1054-2180 cpm. Rats were injected with 100 pg peptide (- 30-80 nmollrat). Peptides were tested at 20 pM and rhodopsin at 0.125 pM.
'
stimulant divided by mean cpm of the unstimulated control
carboxyl-terminal peptides and the 1-11 loop, very few
culture. Stimulation was considered positive if the SI of
inflammatory cells were present in the retina. Some areas
injected rats was equal to or greater than twice the
were infiltrated with cells, particularly near retinal vessels.
background (SI=2).
In general, the inflammatory changes in eyes of animals injected with loop 1-11 and the C-terminus injected rats
RESULTS
observed were less severe than those observed in eyes from
Induction of experimental autoimmune uveitis
animals immunized with rhodopsin (Table 2). The 12-
To determine the ability of rhodopsin peptides to induce
residue carboxyl terminal peptide, 331-342, was as active as
uveitis, Lewis rats were immunized with synthetic peptides
the 25-residue peptide, 324-348. None of the other peptides
and tested for antibody responses, T cell proliferative
tested (Table 1) induced significant inflammatory disease.
responses and pathological changes in the retina. Peptides
Inflammation of the anterior chamber was not detected. We
representing the exposed, hydrophilic regions of rhodopsin
did not observe inflammatory changes in the pineal gland
were selected for immunization (Table 1). The doses of
from animals immunized with any of rhodopsin's synthetic
injected peptides were 10 - 30 times higher than the molar
peptides.
equivalent of 100 pg rhodopsin. Only a few peptides were
Cellular resporises
capable of inducing EAU. Peptides from the C-terminal
The cellular immune responses were determined by the in
(324-348) and from loop 1-11 (61-75) were both capable of
vitro lymphocyte proliferative assay of T cells to different
inducing disease. However, the most immunopathogenic
doses of the appropriate synthetic peptides or rhodopsin in
peptide was sequence 230-25 1 (V-VI loop).
culture. Injection of two peptides from the C-terminus of
Histopathologically, at the immunized doses of 16-28
rhodopsin (324-348 and its truncated form 331-342),
nmol per animal, we observed total destruction of the
peptides representing loop sequences, and the amino-
photoreceptor cell layer when compared to the control
terminus induced different levels of T cell responses but
group that did not receive the antigen. Loop V-VI caused
only at higher doses than rhodopsin ('10-30 times higher on
severe inflammation of the retina (Fig. 2). In the case of the
a molar basis). Positive proliferative responses against the
662
Current Eye Research Table 3. Stimulation of spleen T cells from Lewis rats immunized with peptide 331-342 and its pathogenicity
324-34 S.I.'
INJECTED DOSE nmollrat
Curr Eye Res Downloaded from informahealthcare.com by University of Melbourne on 11/23/14 For personal use only.
0 21 42 84 168 336
1.8 f O . l 3.55k0.2 9.2 k2.5 7.0 f l . O 2.6 k0.6 1.8 f0 . 3
331-342 S.I.
EAU INCIDENCE
1.7 f0.2 1.45 f0.5 6.0 f0.7 5.5 f l . O 2.8 f0.5 4.6 k1.2
SEVERITY
0 0 0.9k0.2 0.9k0.2 0.4k0.35 0.4k0.4
013 012 213 213 113 113
* Data are presented as stimulation indices (SI). Stimulation was considered positive if the SI of injected rats was equal to or greater than twice the background (SI=2). Unstimulated cultures of cells from these animals incorporated 214-1817 cpm. Peptides were tested at 20 pM.
respective immunizing peptides were obtained with doses
protein, rhodopsin.
ranging from 10-20 pM.
To determine an optimal immunizing dose of
Table 2 summarizes results obtained from Lewis rats
synthetic peptide and to correlate it with pathogenicity, rats
immunized with 100 pg (-30-80 nmol/rat) of various
were injected with two peptides representing the sequences
synthetic peptides. The capacity of these peptides to elicit in
previously identified as the major immunodominant epitope.
vitro lymphocyte response is shown. Cells isolated from
The minimal immunogenic and immunopathogenicdose of
Lewis rats injected with any of the following peptides: 2-32
peptide 331-342 in Lewis rats was 42 nmole of peptide.
(N-terminus), 61-75 (loop 1-11), 230-252 (loop V-VI), or
Both carboxyl-terminal peptides (331-342 and 324-348)
324-348 (C-terminus), responded in v i m to that same
were capable of eliciting strong in vidro lymphocyte
peptide. In all cases, the proliferative response against the
proliferation responses in cells derived from rats injected
immunizing peptide was stronger than that against the intact
with peptide 331-342 (Table 3). Peptide 61-75 (loop 1-11)
Table 4. Stimulation of spleen T cells from Lewis rats immunized with peptide 61-75 and its pathogenicity INJECTED DOSE nrnollrat
6 1-75 S.I.'
EAU INCIDENCE
SEVERITY ~
0 21 42 84 168 336
0.7 f0.2 1.0 k0.4 1.8 k1.1 3.0 f0.6 1.85k 1.O 3.6 k0.3
013 113 113 313 213 113
~
~~
0 0.5k0.4 0.5*0.8
1.0k0.4 0.6 k0.25
0.51f0.8
* Data are presented as stimulation indices (SI). Stimulation was considered positive if the SI of injected rats was equal to or greater than twice the background (SI=2). Unstimulated cultures of cells from these animals incorporated 567-1895 cpm. Peptides were tested at 20 p M . 663
Current Eye Research induced overall higher antibody response in animals immunized with the whole molecule (19). Control animals
3
that received saline instead of antigen developed no antibody 0
m
*
2
response.
1
DISCUSSION
a 0
Rhodopsin has been shown to be uveitogenic in susceptible ,
0
Curr Eye Res Downloaded from informahealthcare.com by University of Melbourne on 11/23/14 For personal use only.
100
m-
1000
ANTISERUM DILUTION
strains of animals such as Lewis rats, guinea pigs and primates (12-19). Although T cell recognition sites for rhodopsin have been identified (19), until now, regions of
Figure 3. Anti-rhodopsin antibody titer in antisera of Lewis rats injected with rhodopsin synthetic peptides. Titer of specific antibody was determined by ELISA using microtiter plates coated with purified bovine rhodopsin. Each point is the mean of three separate animals. Preimmune control sera for all tested animals had an average ODdwat 1:20 dilution of 0.75 units. Animals injected with saline instead of peptide did not generated specific antibodies; level of activity was typical of those measured in preimmune sera. (0)2-32; (0)61-75; (m) 96-115; (v) 141-153; ( T ) 174203; ( 0 ) 230-252; ( + ) 276-286; (A) 310-321; (A) 324-348.
the molecule responsible for EAU induction have not been determined. Knowledge of the amino acid sequence of rhodopsin was essential for characterization of immunodominant and pathogenic sites for T cells. Synthetic peptides have been used to define several epitopes on rhodopsin that induced T cell responses in Lewis rats (19). The majority of T cells from animals primed with rhodopsin were directed against the carboxyl terminus which was determined to be a major T cell epitope. Other minor
stimulated weaker proliferative responses in cells derived
epitopes were found in the 11-111 (96-114) and IV-V
from rats injected with peptide 61-75. Representative data
(174-202) loops (see Fig. 1). The first attempt to use a
are summarized in Table 4.
rhodopsin peptide to induce EAU, injection of rhodopsin’s
Antibody responses
amino-terminal CNBr peptide into guinea pigs, did not
Antibody titers were measured against a synthetic antigen
result in disease development (12). In our studies, the major
(immunizing peptide) and native antigen (rhodopsin) in
T cell proliferative site for rhodopsin, sequence 324-348,
serum of animals injected with peptide. The ELISA results
was selected for testing pathogenic properties. Direct
indicate that all the peptides induced antibody responses.
immunization with that peptide stimulates a strong
The antibodies recognized both the appropriate synthetic
proliferative response that is crossreactive with the native
peptide and rhodopsin. However, antibody titers were
protein. Whereas this peptide seems to contain a uveitogenic
different among sera tested. Fig. 3 shows titration curves of
site, not all of rhodopsin’s peptides which elicited immune
rat antisera against rhodopsin. In general, peptides that
responses were immunopathogenic (Table 2). Immunization
represent major antibody binding sites (N-terminus and
of animals with rhodopsin peptides resulted in the display of
174-203), were better immunogens than were other
some different T cell specificities than immunization with
peptides. The highest level of specific antibodies was
the whole antigen. Moreover, immunization with peptides
observed after immunization with the amino-terminal
induces specificities that are not present after rhodopsin
peptide (sequence 2-32). Also immunization with the loop
immunization e.g., T cells specific against residues 230-251
IV-V (174-203) produced good antibody titers. Although
and 2-32. A different repertoire of T cell specificities
rats injected with some of the peptides generated good anti-
induced by peptides and whole protein has been found in the
rhodopsin and anti-peptide antibody titers, rhodopsin
immune response to other proteins (e.g. P-galactosidase (23)
664
Curr Eye Res Downloaded from informahealthcare.com by University of Melbourne on 11/23/14 For personal use only.
Current Eye Research and lysozyme (24)) and a hierarchy of dominant usage in a
vitro, and induce pathogenicity. Its truncated form, residues
multidominant protein has been suggested. Our studies show
331-342, also possessed these properties. Responses to
only two peptides that produced strong in vitro proliferative
rhodopsin’s minor determinants were not always stimulated
responses of T cells, also provoked pathology (230-252 and
by direct immunization with the peptide and were not
324-348). One of these peptides contains a major T cell
capable of provoking EAU except for one, loop 1-11
epitope (324-348). The third pathogenic site is located in the
(61-75). Peptide 230-252 was capable of inducing EAU,
region of loop 1-11 (61-75) which stimulates only mild T-cell
however only in a few cases were cells from rhodopsin-
proliferation. We do not exclude the possibility of presence
injected rats stimulated by that peptide. This determinant
of other immunopathogenic regions of rhodopsin because
could fit to a category of cryptic determinants in which only
the pathogenic doses of peptide were about 20 times higher
injection of the peptide induces a response to itself.
than the protein itself. There may be intramembrane
However, the topography studies of rhodopsin revealed that
portions of the molecule which may be immunogenic
this region of that protein is greatly exposed and accessible.
/pathogenic but which were not tested. It is interesting that
It has been shown that membrane-bound-rhodopsin is
the sequences of all three uveitogenic sites are in the
susceptible to proteolysis by a number of proteases in only a
cytoplasmic, conserved regions of the bovine rhodopsin
few selected regions of the molecule which includes loop V-
structure as compared to mouse and human. Within two
VI (30). It is therefore possible that proteolytic degradation
loop sequences, no amino acid substitutions have been
of rhodopsin which takes place in vicro, produces fragments
reported (25-27). In addition, the carboxyl-terminus
that are unable to sensitize T cells (abolish MHC binding or
(324-348) of bovine rhodopsin differs from mouse and
T cell binding). Although the range of proteases involved in
human rhodopsin by as little as a single amino acid residue.
antigen processing is unknown, it is reasonable to speculate
The uveitogenic (pathogenic) determinants for rhodopsin were defined in regions of major and minor determinants. If autoimmune disease results from a
that proteolytic attack affects the highly exposed surface loop region, sequence 230-251. Rhodopsin peptide 324-348 is not the only
breakdown in self-tolerance, there may be some T cell
immunodominant peptide that has been found to be involved
epitopes that escape the process of tolerance induction (28).
in disease induction. Studies on EAU have revealed
In the model proposed by Gammon and Sercarz (29)
immunodominance of certain fragments of other retinal
tolerance to many self-proteins will not be complete but
autoantigens such a S-antigen (31) and IRBP (7, 9) in
restricted to the major T cell-inducing determinants. The
disease induction. Similarly, in EAE (experimental
pathogenic determinants on self-antigens are not typical
autoimmune encephalomyelitis), which is often compared
major T cell-inducing determinants, but resemble minor
with EAU, immunodominant peptides have been identified
ones. The minor epitopes are available only in relatively
in myelin basic protein (MBP) involved in the disease
low amounts after in vivo processing of the entire molecule.
initiation (28).
These T cells are not normally activated and responses to poorly presented minor epitopes can be stimulated only
ACKNOWLEDGEMENTS
under special circumstances to avoid tolerance. In
Supported in part by NIH grants EY 06225 and EY 06226
rhodopsin-induced EAU however, an immunodominant site
to P.A.H., Core facilities grant EY08571, and an
of rhodopsin, the carboxyl terminal peptide, was capable of
unrestricted departmental grant from Research to Prevent
inducing the immune system to produce sensitized T cells
Blindness, Inc. to the University of Florida Department of
and antibody, stimulate good proliferative responses in
Ophthalmology.
665
Current Eye Research CORRESPONDING AUTHOR Grazyna Adamus, Department of Ophthalmology, Box 100284, University of Florida, Gainesville, F1 32610,
11.
U.S.A.
Immunopathogenic determinants are not necessarily immunodominant. Clin. Immunol. Immunopathol. 3,212-224. Vrabec, T. R., Reber, R. N., Magargal, L. E. and Donoso, L. A. (1990) S-Antigen. Identification of human T-lymphocyte proliferation sites. Arch. Ophthalmol. 1470-1473. Marak, G. E., Shichi, H., Rao, N. A. and Wacker, W. B. (1980) Pattern of experimental allergic uveitis induced by rhodopsin and retinal rod outer segments. Ophthalmic Res. J2,165-176. Meyers-Elliott, R. H., Gammon, R. A., Sumner, H. L. and Shimizu, I. (1983) Experimental retinal autoimmunity (ERA) in strain 13 guinea pig: induction of ERA-retinopathy with rhodopsin. Clin. Immunol. Immunopathol. 2, 81-95. Meyers-Elliott, R. H. and Sumner, H. L. (1982) Experimental uveitis induced by products of activated lymphocytes: intraccular effects of rhodopsin-induced lymphokines. Cell. Immunol. &, 240-253. Broekhuyse, R. M., Winkens, H. J., Kuhlmann, E. D. and VanVugt, A. H. M. (1984) Opsin-induced experimental autoimmune retinitis in rats. Curr. Eye Res. 3, 1405-1412. Schalken, J. J . , Van Vugh, A. H. M.,Winkens, H. J., Bovee-Geurts, P. H. M., De Grip, W. J. and Broekhuyse, R. M. (1988) Experimental autoimmune uveoretinitis in rats induced by rod visual pigment: rhodopsin is more pathogenic than opsin. Graefe's Arch. Clin. Exp. Ophthalmol. 226, 255-261. Schalken, J. J., Winkens, H. J., Van Vugt, A. H. M., Bovee-Geurst, P. H. M., De Grip, W. J. and Broekhuyse, R.M. (1988) Rhodopsin-induced experimental autoimmune uveoretinitis: dose-dependent clinicopathological features. Exp. Eye Res. 47, 135-145. Schalken, J. J., Winkens, H. J., Van Vugt, A. H. M., De Grip, W. J. and Broekhuyse, R. M. (1989) Rhodopsin-induced experimental autoimmune uveoretinitis in monkeys. Br. J. Ophthalmol. 168-172. Moticka, E. J. and Adamus, G. (1991) Specificity of T and B cell responses to bovine rhodopsin in Lewis rat. Cell. Immunol. U, 175-184. Smith, H. G., Stubbs, G.W. and Litman, B.J. (1975) The isolation and purification of intact discs from retinal rod outer segments. Exp. Eye Res. 211-217. Litman, B. J. (1982) Purification of rhodopsin by concanavalin A affinity chromatography. In "Methods in Enzymology" (Ed. Parker, L.) 81, Pp. 150-153 Academic Press, San Diego. Adamus, G., Zam, Z.S., Arendt, A . , Palczewski, K. McDowell, J.H. and Hargrave, P.A. (1991) Antirhodopsin monoclonal antibodies of defined specificity: Characterization and application. Vision Res. 17-31.
Curr Eye Res Downloaded from informahealthcare.com by University of Melbourne on 11/23/14 For personal use only.
m,
REFERENCES Gery, I., Mochizuki, M. and Nussenblatt, R. B. 1. (1986) Retinal specific antigens and immunopathogenic processes they provoke. Prog . Ret. Res. 5, 75-109. Donoso, L. A., Merryman, C. F., Sery, T. W., 2. Shinohara, T., Dietzschold, B., Smith, A. and Kalsow, C. M. (1987) S-Antigen: characterization of a pathogenic epitope which mediates experimental autoimmune uveitis and pinealitis in Lewis rat. Curr. Eye Res. 9, 1151-1159. Donoso, L. A . , Merryman, C. F., Sery, T. W., 3. Vrabec, T., Arbizo, V. and Fong, S.-L. (1988) Human IRBP: characterization of uveitopathogenic sites. Curr. Eye Res. 2, 1087-1095. Donoso, L. A., Yamaki, K., Merryman, C. F . , 4. Shinohara, T., Yue, S . and Sery, T. W. (1988) Human S-Antigen: characterization of uveitopathogenic sites. Curr. Eye Res. 2, 1077-1085. 5. Singh, V. K . , Donoso, L. A., Yamaki, K. and Shinohara, T. (1989) Uveitopathogenic sites in bovine S-antigen. Autoimmunity, 3, 177-187. 6. Gregerson, D. S . , Fling, S . P. and Wohlhueter, R. M. (1986) Characterization of immunologically active cyanogen bromide peptide fragments of bovine and human retinal S-antigen. Exp. Eye Res. 43, 803-818. 7. Kotake, S . , DeSmet, M. D., Wiggert, B., Redmond, T. M., Chader, G. J. and Gery, I. (1991) Analysis of the pivotal residues of the immunodominant and highly uveitogenic determinant of interphotoreceptor retinoid-binding protein. J. Immunol. 146, 2995-3001. Kotake, S . , Redmond, T. M., Wiggert, B., Vistica, 8. B., Sanui, H., Chader, G. J. and Gery, I. (1991) Unusual immunological properties of the uveitogenic interphotoreceptor retinoid-binding protein-derived protein R23. Invest. Ophthalmol. Vis. Sci. Z, 2058-2064. Kotake, S . , Wiggert, B., Redmond, T. M., Borst, 9. D. E., Nickerson, J. M., Margalit, H., Berzofsky, J. A., Chader, G. J. and Gery, I. (1990) Repeated determinants within the retinal interphotoreceptor retinoid-binding protein (IRBP): immunological properties of the repeats of an immunodominant determinant. Cell. Immunol. 2, 331-342, 10 Redmond, T. M., Sanui, H., Hu, L. H., Wiggert, B., Margalit, H., Berzofsky, J. A., Chader, G. J. and Gery, I. (1989) Immune responses to peptides derived from the retinal protein IRBP:
666
12.
13.
14.
15.
16.
17.
18.
a,
19. 20.
a,
21.
22.
a,
Current Eye Research ~~
23.
24.
25.
Shivakumar, S., Sercarz, E.E. and Krzych, U. (1989)The molecular context of determinants within the priming antigen establishes a hierarchy of T cell induction: T cell specificities induced by peptides of 0-galactoside vs. the whole antigen. Eur. J. Immunol. L 2 ,681-687. Shastri, N., Miller, A. and Serwz, E.E. (1984)The expressed T cell repertoire is hierarchical: the precise focus of lysozyme-specific T cell clones is dependent upon the structure of the immunogen. J. Mol. Cell. Immunol. 1,369-379. Baehr, W., Falk, J.D., Bugra, K., Triantafullos, J.T. and McGinnis, J.F. (1988)Isolation and analysis of the mouse opsin gene. FEBS Lett. 253-256. Nathans, J. and Hogness, D.S.(1984) Isolation and nucleotide sequence of the gene encoding human rhodopsin. Proc. Natl. Acad. Sci. USA. 81, 48514855. Hargrave, P.A., McDowell, J.H., Curtis D.R., Wang, J.K., Juszczak, E., Fong, S-L., Rao, J.K.M. and Argos, P. (1983)The structure of bovine rhodopsin. Biophys. Struct. Mech. 8, 235-244. Wraith, D.C.,McDevitt, H.O., Steinman, L. and Acha-Orbea, H. (1989)T cell recognition as the target for immune intervention in autoimmune disease. Cell, 52,709-715. Gammon, G.and Sercarz, E. (1989) How some T cells escape tolerance induction. Nature, 342, 183185. Hargrave, P.A. (1982)Rhodopsin chemistry, structure and topography. Prog. Ret. Res. 1, 1-51. Fling, S.P.,Donoso, L.A., Gregerson, D.S. (1991) In vitro Unresponsiveness to autologous sequences of the immunopathogenic autoantigen, S-antigen. J. Immunol. 147, 483-489.
Curr Eye Res Downloaded from informahealthcare.com by University of Melbourne on 11/23/14 For personal use only.
m,
26.
27.
28.
29. 30. 31.
667
