Current Eye Research
ISSN: 0271-3683 (Print) 1460-2202 (Online) Journal homepage: https://www.tandfonline.com/loi/icey20
Severe retinochoroidopathy: variations of humoral and cellular immunity to S-antigen in a longitudinal study Denise Jobin, Brigitte Thillaye, Yvonne de Kozak, Jean Sainte-Laudy, JeanPierre Faure & PhÙC Le Hoang To cite this article: Denise Jobin, Brigitte Thillaye, Yvonne de Kozak, Jean Sainte-Laudy, JeanPierre Faure & PhÙC Le Hoang (1990) Severe retinochoroidopathy: variations of humoral and cellular immunity to S-antigen in a longitudinal study, Current Eye Research, 9:sup1, 91-96, DOI: 10.3109/02713689008999426 To link to this article: https://doi.org/10.3109/02713689008999426
Published online: 02 Jul 2009.
Submit your article to this journal
Article views: 13
Citing articles: 16 View citing articles
Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=icey20
Current hYe Research
Volume 9 supplement 1990
Severe retinochoroidopathy: variations of humoral and cellular immunity to S-antigen in a longitudinal study Denise Jobin, Brigitte Thillaye, Yvonne de Kozak, Jean Sainte-Laudy , Jean-Pierre Faure and Phuc Le Hoang' Laboratoire d'Immunopathologie de I'Oeil, INSERM U86, Centre de Recherches BiomCdicales des Cordeliers, 75270 Paris 06 and 'HBpital de la PitiC, 75651 Paris 13, France
ABSTRACT Ten patients with birdshot retinochoroidopathy, six with isolated retinal vasculitis and eight with Behcet's disease were treated with cyclosporine for one to three years. Autoantibodies to several retinal proteins, circulating lymphocyte subsets and cellular reactivity to S-antigenwere evaluated repeatedly during this period. Autoantibody titers were similar in patients and in controls. However the serum content of antibodies to IRBP or S-antigen was lessened during inflammatory periods in some patients. In some sera, antibodies reacted with enzyme digested S-antigen preparations by immunoblot, whereas the same sera were negative for the native protein. A decrease of the CD4+ subpopulation of peripheral blood lymphocytes was associated with relapses of ocular inflammation in birdshot retinochoroidopathy. In this disease and in idiopathic retinal vasculitis, the positive lymphocyte stimulation test and basophil degranulation test with S-antigen were significantly most frequent in the period preceding a relapse of ocular inflammation. These tests could therefore be of predictive value for relapses occurring within the next few months.
INTRODUCTION Retinal vascular involvement is a very common finding in many cases of posterior, intermediate or diffuse uveitis, especially in Behcet's disease and birdshot retinochoroidopathy. Recurrent episodes of retinal vasculitis are also observed without symptoms typical of any defined entity. We present a longitudinal study of some immunological parameters in patients with severe retinochoroidopathies treated by cyclosporine for a long period. We tried to
find biological tests that could have a predictive value for recurrence and could help to adapt therapy. Five different assays were performed, and concerned: 1) T lymphocyte subpopulations in the peripheral blood; 2) circulating immune complexes (CIC); 3) lymphocyte stimulation test (LST) with bovine S-antigen; 4) basophil degranulation test (HBDT) with S-antigen; and 5) humoral response to various retinal proteins. PATIENTS AND METHODS Patients Ten patients with birdshot retinochoroidopathy ( B R C ) , attending the Pitie-Salp6tri&re Eye Departement were studied for an 11 to 33 month period. At the onset of the disease, the patients, five men and five women, ranged in age from 37 to 6 3 . Antigen HLA A29 was present in all ten patients and antigen HLA B12 in six of them. Six patients with Behcet's disease (five men and one woman) and eight patients with idiopathic retinal vasculitis ( 4 men and 4 women) were also followed up. Only one patient with idiopathic vasculitis was HLA A29 positive. Disease activity was assessed by the same ophthalmologist, using slit lamp biomicroscopy, retinal angiography and a standardized subjective grading system. Six patients with BRC and five with idiopathic vasculitis were treated with cyclosporine alone, the others with
91
Current Eye Research additional steroids (10 to 15 mg/day). All the patients with Behcet's disease were treated with both cyclosporine and corticosteroids (20 to 75 mg/day) ; one received additional colchicine (1 mg/day) . Cyclosporine doses were adapted according to the clinical course and main side effects (hypertension,creatininemia). Recurrences occurred in most patients, even under cyclosporine therapy. They were defined on the basis of the following clinical criteria: increase of vitreous haze, vasculitis and/or macular oedema and comparative angiography, rather than a decline of visual acuity that could be due to secondary cataract or macular atrophy. In two severe cases of BRC, two or three relapses occurred despite constant administration of maximal doses of cyclosporine. The improvement of the clinical stetus between the relapses was very incomplete. In other patients with BRC or idiopathic retinal vasculitis, recurrences sometimes occurred under constant doses of cyclosporine or after reduction of the dose. Inflammatory resumption was generally moderate and progressive in this case. It would be very transient j u s L after the modification of cyclosporine dosage. Cessation of cyclosporine administration was always followed by a dramatic clinical relapse, except for one patient who was cured of his macular oedema one year ago and since has only complained of floaters. The efficiency of cyclosporine on relapses was difficult to evaluate in Behcet's disease because of the added steroid therapy and a probably different physiopathology. Blood collection Heparhized blood samples were collected before the onset of cyclosporine treatment, measured using an ELISA. Plates were coated with [pg/ml] : total human retinal extract [ 5 0 ] , purified bovine S-antigen [ 5 ] , and then at regular time intervals
irrespective of the clinical stage (first, second, third, sixth, ninth, twelth month and then every six months). Lvmphocvte subDoDulations
T cells were studied using indirect immunofluorescence with monoclonal antibodies against peripheral T cells ( O K T 3 ) , inducer T cells (OKT4) and suppressor T cells (OKT8). The normal values for cryopreserved cells from 50 healthy donors were as followed : 44 L 8% inducer T cells (OKT4t1, 29 2 8% suppressor T cells (OKT8+), ratio OKT4/OKT8 : 2 ? 0.5. Lvmphocvte stimulation test with S-antiaen The test was performed on blood lymphocyte cultures i n RPMI medium, supplemented with 10% heat inactivated human AB positive serum. Cultures were set up in triplicate in microtitration plates. Each well contained 4x105 cells in 0.2 ml of culture medium. Five concentrations of purified bovine S-antigen ( 3 , 6 , 10, 20, 40 pg/well) and 3 concentrations of phytohemagglutinin (PHA) were tested for each sample. Cultures were processed for five days, followed by a fourteen hour incubation with tritiated thymidine. The results were expressed as stimulation index (S.I. = cpm in culture with S-antigen or PHA / cpm in cultures without stimulant). Indices higher than or equal to 2.0 were considered positive. BasoDhil dearanulation test with S-antiaen HBDT were performed as proposed by J. Benveniste (1) arid modified by J. Sainte-Laudy et al. ( 2 ) . Blood basophils with specific (IgE) antibody fixed on the cell surface lose their ability to be stained by toluidine blue when exposed in vitro to the corresponding antigen. Counting the basophils both with and without the antigen allows one to determine the percentage of degranulated basophils. The threshold of positivity of the test is 30% of degranulation. Autoantibodies to retinal antiaens Antibodies against retinal proteins were
interphotoreceptor retinoid binding protein (IRBP) [5], the synthetic peptide M [l], degradation products of purified bovine S-antigen (DPS) obtained by a 20 minute digestion of the protein by chymotrypsin [5] or bovine serumalbumine [51. After incubation with the diluted sera, IgG antibody binding was measured using peroxidase-labeledprotein A . Immunoblots were performed on human retinal extract, purified bovine S-antigen and UPS. Human liver extract was used as a control for specificity. Proteins were separated by SDS-polyacrylamidegel electrophoresis (gradient 10 -20%) and then electrotransferred to nitrocellulose membranes. Non specific sites of the membranes were blocked with 15% fat-free dry milk in PBS. Membrane strips were then incubated with 1/20 diluted human sera, then with a peroxidase-labeled anti-human kappa chain monoclonal antibody. 4-Chloro1-naphtolwas used to visualize the complexes. Circulatina immune comDlexes CIC determination was performed by the polyethylene glycol technique as described by Digeon et al. ( 3 ) . RESULTS AND DISCUSSION Autoantibodies The mean level of autoantibodies to retinal antigens measured by ELISA did not differ between patients and controls, as shown in other studies (4,5). However, in birdshot retinochoroidopathy, the levels of antibodies tested against degradation products of S-antigen were significantly higher in patients than in controls, whereas antibodies to native S-antigen or peptide M were similar. Antibodies to IRBP were lower in patients than in controls. In Behqets's disease, antibodies to S-antigen were also similar to controls, but the levels of anti-DPS and anti-peptideM antibodies were lower than in controls.
In some patients, variations of certain retinal autoantibody levels were observed during the course of the disease. An initial decrease in antibody titres to several antigenic preparations occurred in 7 of 10 patients with BCR during the first month of cyclosporine therapy. Then fluctuations of antibody levels seemed to become independent of cyclosporine dose, and variations were observed even with constant doses. An explanation for this independence could be the fact that two different T helper lymphocyte subpopulations, implicated either in primary or in secondary humoral immune response, have a different susceptibility to cyclosporine: cyclosporine is poorly effective on the secondary response ( 6 , 7 ) . A decrease of antibodies to IRBP was correlated with a recurrence of ocular inflammation in two patients with birdshot retinochoroidopathy (fig. la). In a patient with Behcet's disease, a progressive increase of antibodies to S-antigen was associated with a regression of the ocular inflammation (fig. lb) . These observations suggest that autoantibodies to several retinal proteins are produced naturally in normal individuals. A decrease of these antibodies in serum seems to be related to active ocular inflammation, and could be due to immune complex deposition in the ocular target. Only two patients with birdshot retinochoroidopathy presented a significant variation of CIC during the course of the disease with an elevation during the inflammatory period. For one of them, CIC levels were inversely proportional to the level of anti-retina antibodies. In Behqet's disease and idiopathic retinal vasculitis, we found no correlation between CIC levels and the clinical status. On immunoblots of soluble human retinal extract, patients' and control sera 93
Current Eye Research 3Ratio 1
Percentage of Dositive 7/20 6/20
0 *urnan retina +lRBP -S-antigen +DPS
1
.l$
---Peptide
LST 9/10 10110
HBDT
11/15 3/15
M
REMISSION No recurrence Recurrence 1-4 months after the test
NFLAMMATIO N
1-4 months after the test
Figure 2: In patients with birdshot retinochoroidopathy, positive lymphocyte stimulation tests with S-antigen (LST) and basophil degranulation tests with S-antigen (HBDT) were most frequently observed in the period preceding a recurrence of ocular inflammation. + S-antigen
* Peptide M
I d 5
I
10
15
20 Months
+ i 2.0 mglkglday
Figure 1: Variations of antibodies to retinal proteins determined by ELISA during the course of ocular disease in the sera of two patients treated with cyclosporine. (The antibody level is expressed as the ratio of the ELISA reading for each serum over the mean of ELISA readings for a series of ten controls). a) In a patient with birdshot retinochoroidopathy, the level of antibodies to IRBP or human retinal extract decreased during the relapses of ocular inflammation. (b) In a patient with Behcet's disease, a progressive increase of antibodies to S-antigen,its chymotrypsin degradation products and peptide M was observed at the time of regression of ocular inflammation.
revealed several protein bands. We did not find any pattern or band characteristic of one particular disease. The pattern in a 94
given patient was very reproducible in successive sera with variation in the intensity of staining of some proteins. Blots of purified S-antigen confirmed the presence of antibodies to this protein .in the sera of several patients. Some of them and 3 controls recognized several bands in blots of chymotrypsin D P S , whether or not they reacted with native S-antigen. Testing these sera on blots of CNBr cleavage peptides of S-antigen (gift of Dr D . S . Gregerson) showed that they recognize the C-terminal region of the protein. Further studies using degradation products of retinal autoantigens should be of interest as they can demonstrate antibodies to antigenic sites that are not accessible in the native protein. Cellular immunitv Significant variations in circulating T lymphocyte subpopulations were observed during the course of birdshot retinochoroidopathy; a decrease in the OKT4/OKT8 ratio, due to a decrease of the
Current Eye Research CD4+ population, was associated with inflammatory periods, with normalization of the ratio during remissions. These changes did not appear to be dependent on the treatment, as they occurred under constant cyclosporine dose. Is this decrease of circulating CD4+ T helper cells due to their accumulation in the ocular inflammatory focus ? Confirming previous observations ( 8 , 9 ) we frequently found positive lymphocyte stimulation tests with S-antigen. Lymphocytes were more often reactive to S-antigen during periods of acute ocular inflammation than during remissions. However, at the en6 of the remission phase, i.e. in the period preceding a relapse of inflammation, the test became positive in almost all patients with birdshot retinochoroidopathy (fig. 2 ) . This phenomenon was not observed in patients suffering from idiopathic vasculitis or Behqet's disease. This may be due to the low number of cases. Cyclosporine therapy does not seem to influence the proliferative response of circulating lymphocytes to S-antigen, as no variation of the stimulation index occurred when the treatment was stopped. This is in contradiction with the observation of Nussenblatt et al. (10) concerning the effect of cyclosporine in the experimental model of S-antigen-induceduveitis, but in agreement with the finding by Palestine et al. (11) that treatment with cyclosporine has little or no effect on the lymphocyte proliferative response of humans to tetanus toxoid or keyhole limpet hemocyanine. An explanation could be that cyclosporine is not cytotoxic and that its effect on T helper cells is easily reversed when cyclosporine is not added in the culture medium. So LST positivity does not indicate cyclosporine inefficiency. LST alone is not a valuable test to adapt cyclosporine therapy.
The basophil degranulation test allows to detect specific IgE bound to blood basophils. HBDT in the presence of S-antigen has often been found to be positive in patients with posterior uveitis ( 2 ) . In the present longitudinal study, variations in basophil reactivity to S-antigen were related to the clinical status of the eye. Even more demonstrative than for lymphocyte stimulation was the fact that HBDT with S-antigen became positive at the end of remission periods in birdshot retinochoroidopathy and idiopathic vasculitis (fig. 2 ) . The test was positive in all the patients tested during remission phase which presented a relapse of ocular inflammation in a period of 4 months following the test. This suggests that basophil or local mast cell activation by IgE specific for retinal antigens could play an initiating role in inflammation. This longitudinal study of severe retinochoroidopathies shows that some modifications of the immune status of the patient, i.e. decrease of retinal autoantibody level and positive LST and HBDT could have a predictive value for recurrence. Though these changes do not appear to depend on cyclosporine, these tests could be of interest in order to adapt drug therapy. CORRESPONDING AUTHOR Dr Denise Jobin, Laboratoire d'Immunopathologie de l'Oeil, Centre des Cordeliers, 15, rue de 1'Ecole de MBdecine, 75270 Paris 06, France. REFERENCES 1. Benveniste, J . (1981) The human basophil degranulation test as an in vitro method for the diagnosis of allergies. Clin. Allergy, 11,1-11. 2. Sainte-Laudy, J., Faure, J . P . , de Kozak,Y., Bloch-Michel,E., Le Hoang, P. and Benveniste, J . (1985) Immediate hypersensitivity in human retinal autoimmunity. In "Advances in Immunology and Immunopathology of 95
Current Eye Research
3.
4.
5.
6.
7.
8.
9.
10.
96
the Eye", (eds. 0'Connor, G.R. and Chandler, J.W. ) , Pp. 122-124, Year Book Medical Publ., Chicago. Digeon, M., Laver, M., Riza, J. and Bach, J.F. (1977) Detection of circulating immune complexes in human sera by simplified assays with polyethylene glycol. J. Immunol. Methods, 115, 165 183. Doekes, G., Van dur Gaag, R., Rothova,A.,Van Kooyk, Y., Broersma, L., Zaal, M.J.M., Dijkman, G., Fortuin, M.E., Baarsma, G.S. and Kijlstra, A . (1987) Humoral and cellular immune responsiveness to human S-antigen in uveitis. Curr. Eye Res. 6 , 909-919. Forrester, J.V., Stott, D.I. and Hercus,K.M. (1989) Naturally occurring antibodies to bovine and human retinal S antigen: a comparison between uveitis patients and healthy volunteers. Br. J. Ophthalmol. 73,155-159. Kunkl, A. and Klaus, G.G.B. (1980) Selective effects of cyclosporine A on functional B cell subsets in the mouse. J. Immunol. 125,2526-2531. Klaus, G.G.B. and Kunkl, A . (1983) Effects of cyclosporine on the immune system of the mouse. 11. Cyclosporine inhibits the effector function of primary T helper cells, but not helper cell priming. Transplantation, 36, 8084. Faure, J.P., Bloch-Michel, E., Le Hoang, P. and Vadot, E. (1988) Immunopathologie de l'Oeil, Rapport A la Socidtk Francaise d'Ophtalmologie, Pp.358-362. Paris, Masson, Nussenblatt, R.B. and Palestine, A.G. (1989) Uveitis, Fundamentals and Clinical Practice, P p . 226-227. Year Book Medical Publishers, Bethesda. Nussenblatt, R.B., Salinas-Carmona,M., Waksman, B.H. and Gery, I. Cyclosporin A: alterations of the cellular immune response in S-antigen-induced experimental autoimmune uveitis. Int. Archs Allergy Appl. Immunol. 7 0 , 289294.
ISSN: 0271-3683 (Print) 1460-2202 (Online) Journal homepage: https://www.tandfonline.com/loi/icey20
Severe retinochoroidopathy: variations of humoral and cellular immunity to S-antigen in a longitudinal study Denise Jobin, Brigitte Thillaye, Yvonne de Kozak, Jean Sainte-Laudy, JeanPierre Faure & PhÙC Le Hoang To cite this article: Denise Jobin, Brigitte Thillaye, Yvonne de Kozak, Jean Sainte-Laudy, JeanPierre Faure & PhÙC Le Hoang (1990) Severe retinochoroidopathy: variations of humoral and cellular immunity to S-antigen in a longitudinal study, Current Eye Research, 9:sup1, 91-96, DOI: 10.3109/02713689008999426 To link to this article: https://doi.org/10.3109/02713689008999426
Published online: 02 Jul 2009.
Submit your article to this journal
Article views: 13
Citing articles: 16 View citing articles
Full Terms & Conditions of access and use can be found at https://www.tandfonline.com/action/journalInformation?journalCode=icey20
Current hYe Research
Volume 9 supplement 1990
Severe retinochoroidopathy: variations of humoral and cellular immunity to S-antigen in a longitudinal study Denise Jobin, Brigitte Thillaye, Yvonne de Kozak, Jean Sainte-Laudy , Jean-Pierre Faure and Phuc Le Hoang' Laboratoire d'Immunopathologie de I'Oeil, INSERM U86, Centre de Recherches BiomCdicales des Cordeliers, 75270 Paris 06 and 'HBpital de la PitiC, 75651 Paris 13, France
ABSTRACT Ten patients with birdshot retinochoroidopathy, six with isolated retinal vasculitis and eight with Behcet's disease were treated with cyclosporine for one to three years. Autoantibodies to several retinal proteins, circulating lymphocyte subsets and cellular reactivity to S-antigenwere evaluated repeatedly during this period. Autoantibody titers were similar in patients and in controls. However the serum content of antibodies to IRBP or S-antigen was lessened during inflammatory periods in some patients. In some sera, antibodies reacted with enzyme digested S-antigen preparations by immunoblot, whereas the same sera were negative for the native protein. A decrease of the CD4+ subpopulation of peripheral blood lymphocytes was associated with relapses of ocular inflammation in birdshot retinochoroidopathy. In this disease and in idiopathic retinal vasculitis, the positive lymphocyte stimulation test and basophil degranulation test with S-antigen were significantly most frequent in the period preceding a relapse of ocular inflammation. These tests could therefore be of predictive value for relapses occurring within the next few months.
INTRODUCTION Retinal vascular involvement is a very common finding in many cases of posterior, intermediate or diffuse uveitis, especially in Behcet's disease and birdshot retinochoroidopathy. Recurrent episodes of retinal vasculitis are also observed without symptoms typical of any defined entity. We present a longitudinal study of some immunological parameters in patients with severe retinochoroidopathies treated by cyclosporine for a long period. We tried to
find biological tests that could have a predictive value for recurrence and could help to adapt therapy. Five different assays were performed, and concerned: 1) T lymphocyte subpopulations in the peripheral blood; 2) circulating immune complexes (CIC); 3) lymphocyte stimulation test (LST) with bovine S-antigen; 4) basophil degranulation test (HBDT) with S-antigen; and 5) humoral response to various retinal proteins. PATIENTS AND METHODS Patients Ten patients with birdshot retinochoroidopathy ( B R C ) , attending the Pitie-Salp6tri&re Eye Departement were studied for an 11 to 33 month period. At the onset of the disease, the patients, five men and five women, ranged in age from 37 to 6 3 . Antigen HLA A29 was present in all ten patients and antigen HLA B12 in six of them. Six patients with Behcet's disease (five men and one woman) and eight patients with idiopathic retinal vasculitis ( 4 men and 4 women) were also followed up. Only one patient with idiopathic vasculitis was HLA A29 positive. Disease activity was assessed by the same ophthalmologist, using slit lamp biomicroscopy, retinal angiography and a standardized subjective grading system. Six patients with BRC and five with idiopathic vasculitis were treated with cyclosporine alone, the others with
91
Current Eye Research additional steroids (10 to 15 mg/day). All the patients with Behcet's disease were treated with both cyclosporine and corticosteroids (20 to 75 mg/day) ; one received additional colchicine (1 mg/day) . Cyclosporine doses were adapted according to the clinical course and main side effects (hypertension,creatininemia). Recurrences occurred in most patients, even under cyclosporine therapy. They were defined on the basis of the following clinical criteria: increase of vitreous haze, vasculitis and/or macular oedema and comparative angiography, rather than a decline of visual acuity that could be due to secondary cataract or macular atrophy. In two severe cases of BRC, two or three relapses occurred despite constant administration of maximal doses of cyclosporine. The improvement of the clinical stetus between the relapses was very incomplete. In other patients with BRC or idiopathic retinal vasculitis, recurrences sometimes occurred under constant doses of cyclosporine or after reduction of the dose. Inflammatory resumption was generally moderate and progressive in this case. It would be very transient j u s L after the modification of cyclosporine dosage. Cessation of cyclosporine administration was always followed by a dramatic clinical relapse, except for one patient who was cured of his macular oedema one year ago and since has only complained of floaters. The efficiency of cyclosporine on relapses was difficult to evaluate in Behcet's disease because of the added steroid therapy and a probably different physiopathology. Blood collection Heparhized blood samples were collected before the onset of cyclosporine treatment, measured using an ELISA. Plates were coated with [pg/ml] : total human retinal extract [ 5 0 ] , purified bovine S-antigen [ 5 ] , and then at regular time intervals
irrespective of the clinical stage (first, second, third, sixth, ninth, twelth month and then every six months). Lvmphocvte subDoDulations
T cells were studied using indirect immunofluorescence with monoclonal antibodies against peripheral T cells ( O K T 3 ) , inducer T cells (OKT4) and suppressor T cells (OKT8). The normal values for cryopreserved cells from 50 healthy donors were as followed : 44 L 8% inducer T cells (OKT4t1, 29 2 8% suppressor T cells (OKT8+), ratio OKT4/OKT8 : 2 ? 0.5. Lvmphocvte stimulation test with S-antiaen The test was performed on blood lymphocyte cultures i n RPMI medium, supplemented with 10% heat inactivated human AB positive serum. Cultures were set up in triplicate in microtitration plates. Each well contained 4x105 cells in 0.2 ml of culture medium. Five concentrations of purified bovine S-antigen ( 3 , 6 , 10, 20, 40 pg/well) and 3 concentrations of phytohemagglutinin (PHA) were tested for each sample. Cultures were processed for five days, followed by a fourteen hour incubation with tritiated thymidine. The results were expressed as stimulation index (S.I. = cpm in culture with S-antigen or PHA / cpm in cultures without stimulant). Indices higher than or equal to 2.0 were considered positive. BasoDhil dearanulation test with S-antiaen HBDT were performed as proposed by J. Benveniste (1) arid modified by J. Sainte-Laudy et al. ( 2 ) . Blood basophils with specific (IgE) antibody fixed on the cell surface lose their ability to be stained by toluidine blue when exposed in vitro to the corresponding antigen. Counting the basophils both with and without the antigen allows one to determine the percentage of degranulated basophils. The threshold of positivity of the test is 30% of degranulation. Autoantibodies to retinal antiaens Antibodies against retinal proteins were
interphotoreceptor retinoid binding protein (IRBP) [5], the synthetic peptide M [l], degradation products of purified bovine S-antigen (DPS) obtained by a 20 minute digestion of the protein by chymotrypsin [5] or bovine serumalbumine [51. After incubation with the diluted sera, IgG antibody binding was measured using peroxidase-labeledprotein A . Immunoblots were performed on human retinal extract, purified bovine S-antigen and UPS. Human liver extract was used as a control for specificity. Proteins were separated by SDS-polyacrylamidegel electrophoresis (gradient 10 -20%) and then electrotransferred to nitrocellulose membranes. Non specific sites of the membranes were blocked with 15% fat-free dry milk in PBS. Membrane strips were then incubated with 1/20 diluted human sera, then with a peroxidase-labeled anti-human kappa chain monoclonal antibody. 4-Chloro1-naphtolwas used to visualize the complexes. Circulatina immune comDlexes CIC determination was performed by the polyethylene glycol technique as described by Digeon et al. ( 3 ) . RESULTS AND DISCUSSION Autoantibodies The mean level of autoantibodies to retinal antigens measured by ELISA did not differ between patients and controls, as shown in other studies (4,5). However, in birdshot retinochoroidopathy, the levels of antibodies tested against degradation products of S-antigen were significantly higher in patients than in controls, whereas antibodies to native S-antigen or peptide M were similar. Antibodies to IRBP were lower in patients than in controls. In Behqets's disease, antibodies to S-antigen were also similar to controls, but the levels of anti-DPS and anti-peptideM antibodies were lower than in controls.
In some patients, variations of certain retinal autoantibody levels were observed during the course of the disease. An initial decrease in antibody titres to several antigenic preparations occurred in 7 of 10 patients with BCR during the first month of cyclosporine therapy. Then fluctuations of antibody levels seemed to become independent of cyclosporine dose, and variations were observed even with constant doses. An explanation for this independence could be the fact that two different T helper lymphocyte subpopulations, implicated either in primary or in secondary humoral immune response, have a different susceptibility to cyclosporine: cyclosporine is poorly effective on the secondary response ( 6 , 7 ) . A decrease of antibodies to IRBP was correlated with a recurrence of ocular inflammation in two patients with birdshot retinochoroidopathy (fig. la). In a patient with Behcet's disease, a progressive increase of antibodies to S-antigen was associated with a regression of the ocular inflammation (fig. lb) . These observations suggest that autoantibodies to several retinal proteins are produced naturally in normal individuals. A decrease of these antibodies in serum seems to be related to active ocular inflammation, and could be due to immune complex deposition in the ocular target. Only two patients with birdshot retinochoroidopathy presented a significant variation of CIC during the course of the disease with an elevation during the inflammatory period. For one of them, CIC levels were inversely proportional to the level of anti-retina antibodies. In Behqet's disease and idiopathic retinal vasculitis, we found no correlation between CIC levels and the clinical status. On immunoblots of soluble human retinal extract, patients' and control sera 93
Current Eye Research 3Ratio 1
Percentage of Dositive 7/20 6/20
0 *urnan retina +lRBP -S-antigen +DPS
1
.l$
---Peptide
LST 9/10 10110
HBDT
11/15 3/15
M
REMISSION No recurrence Recurrence 1-4 months after the test
NFLAMMATIO N
1-4 months after the test
Figure 2: In patients with birdshot retinochoroidopathy, positive lymphocyte stimulation tests with S-antigen (LST) and basophil degranulation tests with S-antigen (HBDT) were most frequently observed in the period preceding a recurrence of ocular inflammation. + S-antigen
* Peptide M
I d 5
I
10
15
20 Months
+ i 2.0 mglkglday
Figure 1: Variations of antibodies to retinal proteins determined by ELISA during the course of ocular disease in the sera of two patients treated with cyclosporine. (The antibody level is expressed as the ratio of the ELISA reading for each serum over the mean of ELISA readings for a series of ten controls). a) In a patient with birdshot retinochoroidopathy, the level of antibodies to IRBP or human retinal extract decreased during the relapses of ocular inflammation. (b) In a patient with Behcet's disease, a progressive increase of antibodies to S-antigen,its chymotrypsin degradation products and peptide M was observed at the time of regression of ocular inflammation.
revealed several protein bands. We did not find any pattern or band characteristic of one particular disease. The pattern in a 94
given patient was very reproducible in successive sera with variation in the intensity of staining of some proteins. Blots of purified S-antigen confirmed the presence of antibodies to this protein .in the sera of several patients. Some of them and 3 controls recognized several bands in blots of chymotrypsin D P S , whether or not they reacted with native S-antigen. Testing these sera on blots of CNBr cleavage peptides of S-antigen (gift of Dr D . S . Gregerson) showed that they recognize the C-terminal region of the protein. Further studies using degradation products of retinal autoantigens should be of interest as they can demonstrate antibodies to antigenic sites that are not accessible in the native protein. Cellular immunitv Significant variations in circulating T lymphocyte subpopulations were observed during the course of birdshot retinochoroidopathy; a decrease in the OKT4/OKT8 ratio, due to a decrease of the
Current Eye Research CD4+ population, was associated with inflammatory periods, with normalization of the ratio during remissions. These changes did not appear to be dependent on the treatment, as they occurred under constant cyclosporine dose. Is this decrease of circulating CD4+ T helper cells due to their accumulation in the ocular inflammatory focus ? Confirming previous observations ( 8 , 9 ) we frequently found positive lymphocyte stimulation tests with S-antigen. Lymphocytes were more often reactive to S-antigen during periods of acute ocular inflammation than during remissions. However, at the en6 of the remission phase, i.e. in the period preceding a relapse of inflammation, the test became positive in almost all patients with birdshot retinochoroidopathy (fig. 2 ) . This phenomenon was not observed in patients suffering from idiopathic vasculitis or Behqet's disease. This may be due to the low number of cases. Cyclosporine therapy does not seem to influence the proliferative response of circulating lymphocytes to S-antigen, as no variation of the stimulation index occurred when the treatment was stopped. This is in contradiction with the observation of Nussenblatt et al. (10) concerning the effect of cyclosporine in the experimental model of S-antigen-induceduveitis, but in agreement with the finding by Palestine et al. (11) that treatment with cyclosporine has little or no effect on the lymphocyte proliferative response of humans to tetanus toxoid or keyhole limpet hemocyanine. An explanation could be that cyclosporine is not cytotoxic and that its effect on T helper cells is easily reversed when cyclosporine is not added in the culture medium. So LST positivity does not indicate cyclosporine inefficiency. LST alone is not a valuable test to adapt cyclosporine therapy.
The basophil degranulation test allows to detect specific IgE bound to blood basophils. HBDT in the presence of S-antigen has often been found to be positive in patients with posterior uveitis ( 2 ) . In the present longitudinal study, variations in basophil reactivity to S-antigen were related to the clinical status of the eye. Even more demonstrative than for lymphocyte stimulation was the fact that HBDT with S-antigen became positive at the end of remission periods in birdshot retinochoroidopathy and idiopathic vasculitis (fig. 2 ) . The test was positive in all the patients tested during remission phase which presented a relapse of ocular inflammation in a period of 4 months following the test. This suggests that basophil or local mast cell activation by IgE specific for retinal antigens could play an initiating role in inflammation. This longitudinal study of severe retinochoroidopathies shows that some modifications of the immune status of the patient, i.e. decrease of retinal autoantibody level and positive LST and HBDT could have a predictive value for recurrence. Though these changes do not appear to depend on cyclosporine, these tests could be of interest in order to adapt drug therapy. CORRESPONDING AUTHOR Dr Denise Jobin, Laboratoire d'Immunopathologie de l'Oeil, Centre des Cordeliers, 15, rue de 1'Ecole de MBdecine, 75270 Paris 06, France. REFERENCES 1. Benveniste, J . (1981) The human basophil degranulation test as an in vitro method for the diagnosis of allergies. Clin. Allergy, 11,1-11. 2. Sainte-Laudy, J., Faure, J . P . , de Kozak,Y., Bloch-Michel,E., Le Hoang, P. and Benveniste, J . (1985) Immediate hypersensitivity in human retinal autoimmunity. In "Advances in Immunology and Immunopathology of 95
Current Eye Research
3.
4.
5.
6.
7.
8.
9.
10.
96
the Eye", (eds. 0'Connor, G.R. and Chandler, J.W. ) , Pp. 122-124, Year Book Medical Publ., Chicago. Digeon, M., Laver, M., Riza, J. and Bach, J.F. (1977) Detection of circulating immune complexes in human sera by simplified assays with polyethylene glycol. J. Immunol. Methods, 115, 165 183. Doekes, G., Van dur Gaag, R., Rothova,A.,Van Kooyk, Y., Broersma, L., Zaal, M.J.M., Dijkman, G., Fortuin, M.E., Baarsma, G.S. and Kijlstra, A . (1987) Humoral and cellular immune responsiveness to human S-antigen in uveitis. Curr. Eye Res. 6 , 909-919. Forrester, J.V., Stott, D.I. and Hercus,K.M. (1989) Naturally occurring antibodies to bovine and human retinal S antigen: a comparison between uveitis patients and healthy volunteers. Br. J. Ophthalmol. 73,155-159. Kunkl, A. and Klaus, G.G.B. (1980) Selective effects of cyclosporine A on functional B cell subsets in the mouse. J. Immunol. 125,2526-2531. Klaus, G.G.B. and Kunkl, A . (1983) Effects of cyclosporine on the immune system of the mouse. 11. Cyclosporine inhibits the effector function of primary T helper cells, but not helper cell priming. Transplantation, 36, 8084. Faure, J.P., Bloch-Michel, E., Le Hoang, P. and Vadot, E. (1988) Immunopathologie de l'Oeil, Rapport A la Socidtk Francaise d'Ophtalmologie, Pp.358-362. Paris, Masson, Nussenblatt, R.B. and Palestine, A.G. (1989) Uveitis, Fundamentals and Clinical Practice, P p . 226-227. Year Book Medical Publishers, Bethesda. Nussenblatt, R.B., Salinas-Carmona,M., Waksman, B.H. and Gery, I. Cyclosporin A: alterations of the cellular immune response in S-antigen-induced experimental autoimmune uveitis. Int. Archs Allergy Appl. Immunol. 7 0 , 289294.
