Immunohistochemical Evidence for C3bi Involvement in Graves Ophthalmopathy Carl E. Rosen, MD, 1 Frank Parisi, MD, 1 Radmila B. Raikow, PhD, 2 Ronald M. Burde, MD/ John S. Kennerdell, MD2 Purpose: To determine by immunohistochemical methods if components of the complement system are present in Graves ophthalmopathy extraocular and periocular tissues compared with non-Graves ophthalmopathy ocular tissues, and, if so, whether a qualitative difference exists. Methods: Orbital muscle, periorbital muscle, and adipose tissue from 10 Graves ophthalmopathy patients were studied with in situ assays using monoclonal antibodies for C3bi and C5b-9 (the terminal attack complex) complement components. Extraocular muscle, periocular muscle, and adipose tissue from 12 patients treated for unrelated orbital disorders were used as controls. Results: All nine Graves extraocular and periocular muscle tissues exhibited C3bi positive staining in an intense, localized oval- to spindle-shaped reaction that appeared to represent cells on a diffuse staining background of the endomysia! and perimysial connective tissues with no staining of the muscle fibers themselves. Some reactivity was seen in 6 of the 12 control muscles, but this was much less intense than that of Graves ocular muscle tissue. Only two Graves muscle samples stained minimally with the monoclonal antibody for the C5b-9 terminal attack complex while none of the control muscle samples demonstrated reactivity. Orbital fat from Graves and control patients did not demonstrate any reactivity for C3bi or C5b-9. Conclusion: C3bi and not C5b-9 (the terminal attack complex) is present in Graves ophthalmopathy extraocular and periocular tissues in a qualitatively greater way than in control non-Graves ophthalmopathy ocular tissue. Consequently, C3bi may contribute to the pathophysiology of Graves ophthalmopathy. Ophthalmology 1992;99: 1325-1331
Although the literature is replete with clinical and immunologic data concerning Graves ophthalmopathy, the pathophysiology remains unclear. 1 Graves ophthalmopaOriginally received: January 16, 1992. Revision accepted: March 17, 1992. 1 Department of Ophthalmology, Montefiore Medical Center, Albert Einstein College of Medicine, Bronx. 2 Department of Ophthalmology, Allegheny General Hospital and Allegheny-Singer Research Institute, Pittsburgh. Supported in part by an unrestricted departmental grant to the Department of Ophthalmology, Albert Einstein College of Medicine, Montefiore Medical Center from Research to Prevent Blindness, Inc, New Yark, New York. Reprint requests to Carl E. Rosen, MD, Department of Ophthalmology, Montefiore Medical Center, Albert Einstein College of Medicine, Ill E 2 lOth St, Bronx, NY.
thy is usually associated with Graves hyperthyroidism, however ophthalmopathy also may occur in patients with primary hypothyroidism or Hashimoto's thyroiditis, and is sometimes found in the apparent absence of thyroid disease. 2•3 The principal pathologic feature of Graves ophthalmopathy is inflammation of the orbital soft tissues and muscles. Inflammation of the extraocular muscles seems to be restricted to the muscle bellies, with sparing oftendons as visualized by neuroimaging techniques. 4 Disease severity parallels the extent of optic nerve compression. 5 The complement system is the principal humoral effector of immunologically induced inflammation. It plays a crucial role both in immunologically induced and nonspecific resistance to infections and in the pathogenesis oftissue injury. Complement activation can be initiated by either of two pathways: the classic pathway or the al-
1325
Ophthalmology
Volume 99, Number 8, August 1992
ternative pathway. Both pathways lead to the activation of the pivotal complement component C3. The classic pathway is activated by immunoglobulin (lgG~. lgG2 , lgG 3, lgM) when bound to antigen, while the alternative pathway is activated by a variety of bacteria, yeast, parasites, insoluble immune complexes, 6 and antibodies such as lgA 1 and IgA 2 •7 The C3 component can be fragmented into C3b, which serves as a branch point for the complement cascade to either form C3bi or the terminal attack complex (C5b-9). 6 The C3bi complement fragment can bind to complement receptors present on inflammatory cells as well as connective tissues and function as an adhesive ligand. 8 The role of complement in the pathogenesis ofGraves ophthalmopathy is unknown. Werner et al 9 discovered the presence of lgE, lgM, IgG and complement components C1q and C3 in the connective tissues ofthe thyroid gland in patients with Graves disease. Wang et al 10 examined the possible role of antibody-mediated complement-dependent cytotoxicity in the pathogenesis of Graves ophthalmopathy by demonstrating that serum from patients with Graves ophthalmopathy caused a significant increase in eye muscle cell lysis when compared with control subjects. Recently, the antibody lgA 1 was discovered immunohistochemically in the endomysia! and surrounding connective tissues of Graves ophthalmopathy extraocular muscle. 11 We now report immunohistochemical evidence for the presence of the complement component C3bi in Graves extraocular and periocular muscle tissues in a focal, oval- to spindle-shaped reaction that seemed to represent cells on a diffuse staining background of the endomysia! and perimysial connective tissues with no staining of the muscle fibers themselves.
Materials and Methods Tissue Specimens Tissue was removed from 10 Graves ophthalmopathy patients to improve their vision and appearance. Twelve patients without Graves disease provided control orbital tissue, which was removed in association with surgery for strabismus, blepharoptosis, fat prolapse, or enucleation because of phthisis. Our use oftissue did not in any way impinge on diagnosis or treatment. The tissues were placed into sterile, moist gauze in the operating room and fixed within 3 hours. Tissue was fixed in Zenker's fluid, embedded in paraffin, cut into 4 Jim sections, and placed onto slides. All antibody reactions were performed at room temperature (Table I). Anti-C3bi and anti-C5b-9 monoclonal antibodies were used as primary antibodies in the 10 Graves cases and the 12 control cases, as described below. Secondary antibodies and substrates were used as recommended by their suppliers. A 30-minute block with 2% instant nonfat milk powder in phosphate-buffered saline was used to reduce nonspecific antibody adhesion to tissue. Slides were washed twice in a gently agitating bath of phosphate-buffered saline for 10 minutes in between
1326
each step. All slides were counter-stained with Meyer's hematoxylin. Aqueous mounting medium (Glycergel, DAKO Corp, Santa Barbara, CA) was used.
Primary Antibodies Mouse monoclonal antibodies, anti-human C3bi (Quidel Corp, San Diego, CA, A208) and anti-human C5b-9 (Quidel Corp, A239) were diluted to 10 JLg/ml using phosphate-buffered saline with 1% sodium azide and reacted with deparaffinized tissue sections for 2 hours. Bound primary antibody was then visualized by reacting the slides with Biogenex Super Sensitive StrAvigenKit AA000-5M (San Ramona, CA) using alkaline phosphatase, according to instructions provided by the supplier. Previous studies 11 have shown that there is no endogenous alkaline phosphatase activity in Graves or control orbital tissues after Zenker's fixation.
Case Reports Graves Ophthalmopathy Patients Case 1: Levator Muscle. A 54-year-old woman with severe Graves ophthalmopathy had a prior left lateral rectus resection and a left medial rectus recession for a large esotropia 18 years previously. The patient presented with bilateral proptosis, diplopia, and blepharoptosis on the right side with lid retraction on the left side. Computed tomography showed marked involvement of the right inferior and medial rectus muscles. Results of a tensilon test were negative. At surgery, a right inferior rectus recession, a right levator resection, and a left Mueller's muscle excision with levator recession were performed. This was case I in a previous study" in which the presence oflgA 1 was demonstrated in the connective tissues surrounding the muscle fibers. Case 2: Orbicularis Muscle. A 54-year-old woman presented with a history of thyroid disease and Graves ophthalmopathy, which was first diagnosed 2 years before she had a three-wall orbital decompression. Three years after the decompression, an excision of orbicularis muscle and lid fat was done to correct lid retraction and corneal exposure. This was case 2 in a previous study 11 in which the presence of lgA 1 was demonstrated in the connective tissues surrounding the muscle fibers. Case 3: Orbicularis Muscle. A 55-year-old woman with Graves hyperthyroidism had a 2-year history of ophthalmopathy. She had previously been treated with 113I. She presented with lid edema, tearing, and blurry vision. Prednisone did not ameliorate her symptoms, and orbital radiation was performed I year before a bilateral blepharoplasty with removal of orbital fat. This was case 3 in a previous study 11 in which the presence of lgA 1 was demonstrated in the connective tissues surrounding the muscle fibers. Case 4: Levator Muscle. A 51-year-old woman with diabetes mellitus had hypothyroid disease in the distant past. She was diagnosed with moderate Graves ophthalmopathy I year before she underwent bilateral resection of the levator muscles for blepharoptosis and edematous eyelids. This tissue was case 4 in a previous study 11 in which the presence of lgA 1 was demonstrated in the connective tissues surrounding the muscle fibers.
Rosen et al · C3bi Involvement in Graves Ophthalmopathy Case 5: Levator Muscle. A 76-year-old woman with longstanding Graves hyperthyroidism had a two-wall orbital decompression 16 years ago to treat severe Graves ophthalmopathy with bilateral compressive optic neuropathy, esotropia, and lid retraction. For cosmetic purposes, a Mueller's muscle excision with a medial and lateral marginal myotomy of levator muscle was done. This tissue was case 5 in a previous study 11 in which the presence oflgA 1 was demonstrated in the connective tissues surrounding the muscle fibers. Case 6: Mueller's Muscle. Same patient reported in case 5. This tissue was case 6 in a previous study 11 in which the presence of lgA 1 was demonstrated in the connective tissues surrounding the muscle fibers. Case 7: Medial Rectus Muscle. A 72-year-old man presented with a diagnosis of hyperthyroidism and severe Graves ophthalmopathy with prolapsed conjunctiva and lid retraction of I years' duration. Prednisone was administered on a trial basis and yielded good results. However, I year later, the patient experienced diplopia, and a medial rectus recession with an adjustable suture was performed on the left eye, which was followed
5 months later by the same procedure on the right eye. This tissue was case 7 in a previous study 11 in which the presence of lgA 1 was demonstrated in the connective tissues surrounding the muscle fibers. Case 8: Lateral Rectus Muscle. A 66-year-old woman with hypothyroidism had Graves ophthalmopathy of I year's duration. On presentation, compressive optic neuropathy was diagnosed, and a bilateral orbital wall decompression was performed. The present tissue originated from a subsequent recession of the left medial rectus muscle and a resection of the left lateral rectus muscle. This tissue was case 8 in a previous study'' in which the presence oflgA 1 was demonstrated in the connective tissues surrounding the muscle fibers. Case 9: Medial Rectus Muscle. A 51-year-old woman presented with a history of Graves disease of more than I 0 years' duration. Graves ophthalmopathy was diagnosed with bilateral proptosis, chemosis, and conjunctival injection and left lid retraction. A bilateral orbital decompression was performed at that time. Two years later, progressive hmizontal diplopia developed and a left medial rectus recession was performed.
Table 1. Summary of Tissue Reactions Case No. Age (yrs)/Sex Gravt!s 1. 54/F 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.
54/F 55/F 51/F 76/Ft 76/F 72/M 66/F 51/F 57 /F 52/F
Control 12. 29/M 13. 33/M 14. 29/M 15. 65/F 16. 1/F 17. 2/F 18. 2/F 19. 69/F 20. 10/M 21. 35/M 22. 23/F 23. 35/F 24. 42/M 25. 10/M
Tissue
lgA1
lgAz
lgM
lgG
Orbicularis/ levator Orbicularis Orbicularis Levator Levator Mueller's Medial rectus Lateral rectus Medial rectus Orbital fat Orbital fat
4*
0
0
4 2 4 2 2 3 4 2 0 0
0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0
0 0 0 0 0
Levator Medial rectus Medial rectus Lateral rectus Lateral rectus Inferior oblique Medial rectus Orbicularis Mueller's Medial rectus Medial rectus Medial rectus Orbital fat Orbital fat
0 1 2 0 0 1 1 3 0 0 1 3 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0
0
0 0 0 1 0 0 0
0 0 0 0 0
0
0 0
C3bi
C5b-9
2
0
4 2 4 4 2 4 3 2 0 0
0 0 0 1 0 0 1 0 0 0
0 2 1 3 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0
0 0 0 0 0
1 0 0 0 0 0 0
The majority of the tissue sample reactions with monoclonal antibodies anti-IgA" anti-IgA2 , anti-IgM, and anti-IgG were from a previous study 11 * Tissue staining intensity scale from 4 being most reactiv!-! to 0 being nonreactive.
t
Cases 5 and 6, 17 and 18, and 20 and 25 are tissue specimens taken from three different patients, respectively.
1327
·•
,\._
'{
~.~I} .,
. """,•'·" ~
·~
1328
...
.j • . ,
...... ,
' :·a::
'""
.
Rosen et al · C3bi Involvement in Graves Ophthalmopathy Figure 1. Antibody reactions with Graves and control ocular tissue. Shown here are representative tissue samples from reactions with antibodies to C3bi and CSb-9; no anti-immunoglobulin specimens are included (Table 1). Each column represents the same muscle type and each row represents a reaction with a specific anticomplement antibody. All reactions shown followed the described method (see Materials and Methods section). A, levator muscle from Graves patient (case 4) shown after reaction with anti-human C3bi demonstrating a localized spindle-shaped pattern which may represent cells on a diffuse perimysial and epimysia! staining background reaction. B, levator muscle from same Graves patient (case 4) with no staining evident after reaction with anti-human CSb-9. C, D, levator muscle from control patient (case 12) shows no staining after reaction with antihuman C3bi (C) or anti-human CSb-9 (D). E, F, orbicularis muscle from Graves patient (case 2) shown after reaction with anti-human C3bi (E) and anti-human CSb-9 (F). G, H, orbicularis muscle from control patient (case 19) shown after reaction with anti-human C3bi (G) and anti-human CSb9 (H). I, J, medial rectus muscle from Graves patient (case 7) shown after reaction with anti-human C3bi (I) and anti-human CSb-9 (J). K, L, medial rectus muscle from control patient (case 21) shown after reaction with anti-human C3bi (K) and anti-human CSb-9 (L). M, N, lateral rectus muscle from Graves patient (case 8) shown after reaction with anti-human C3bi (M) and anti-human CSb-9 (N). 0, P, lateral rectus muscle from control patient (case 16) shown after reaction with anti-human C3bi (0) and anti-human CSb-9 (P). Q, R, Mueller's muscle from Graves patient (case 6) shown after rection with anti-human C3bi (Q) and CSb-9 (R). S, T, Mueller's muscle from control patient (case 20) shown after reaction with antihuman C3bi (S) and anti-human CSb-9 (T). U, orbicularis muscle from Graves patient (case 2) shown after reaction with anti-human C3bi demonstrating a spindle-shaped staining pattern which may represent cells. V, levator muscle from Graves patient (case 4) shown after reaction with anti-human C3bi. W, medial rectus muscle from Graves patient (case 7) shown after reaction with anti-human C3bi (A-T, original magnification, XlOO) (U-W, original magnification, X250).
Case 10: Orbital Fat. A 57-year-old woman, who had undergone a prior thyroidectomy, presented with proptosis and diplopia. Severe Graves ophthalmopathy was diagnosed with decreased extraocular motility including right hypertropia and bilateral lid retraction. At surgery, a bilateral Mueller's excision, right levator marginal myotomy, and a left inferior rectus recession with an adjustable suture were performed. The pathology report of the Mueller's specimen indicated the presence of fat and very little muscle. This tissue was case 12 in a previous study. 11 Case 11: Orbital Fat. A 52-year-old woman presented with a 7-year history of hypothyroidism and severe Graves ophthalmopathy with bilateral proptosis. She had a Mueller's excision for lid retraction 2 years before the present surgery. The patient presented with orbital fat prolapse, which was removed. This tissue was case 13 in a previous study. 11
Control (Non-Graves) Patients Case 12: Levator Muscle. A 29-year-old man with an unremarkable medical history underwent correction for congenital blepharoptosis. This tissue was case 14 in a previous study. 11 Case 13: Medial Rectus Muscle. A 33-year-old man experienced accidental head trauma. He required recession of the lateral rectus and resection of the medial rectus muscles to correct subsequent third nerve aberrant regeneration. This was case 15 in a previous study. 11 Case 14: Medial Rectus Muscle. A 29-year-old man underwent a resection of the medial rectus muscle to correct an esotropia. The specimen was found to be fibrotic with a small amount of disrupted muscle. This was case 16 in a previous study. 11 Case 15: Lateral Rectus Muscle. A 65-year-old woman underwent enucleation of the right eye secondary to phthisis bulbi. This was case 17 in a previous study. 11 Case 16: Lateral Rectus Muscle. A 1112-year-old girl with an unremarkable medical history had a left lateral rectus resection for uncomplicated strabismus. This was case 18 in a previous study. 11 Case 17: Inferior Oblique Muscle. A 2-year-old girl with an unremarkable medical history had a right inferior oblique and a left medial rectus resection for uncomplicated strabismus. This was case 19 in a previous study. 11 Case 18: Medial Rectus Muscle. Same patient reported in case 17. This was case 20 in a previous study. 11
Case 19: Orbicularis Muscle. A 69-year-old woman with myasthenia gravis had a right orbicularis and levator resection for right upper lid blepharoptosis. Case 20: Mueller's Muscle. A 10-year-old boy had a left Mueller's resection for congenital blepharoptosis. Case 21: Medial Rectus. A 35-year-old man with a medical history of trauma 20 years previously had left medial rectus resection for left exotropia. Case 22: Medial Rectus. A 23-year-old woman with history significant for sickle cell trait disease had a right medial rectus resection for a right exotropia with anisometropic amblyopia. Case 23: Medial Rectus. A 35-year-old woman with no significant medical history had a right medial rectus resection for a right exotropia. Case 24: Orbital Fat. A 42-year-old man with a history of pseudotumor and corticosteroid treatment underwent orbital fat removal. The specimen consisted of fat with a number of cellular infiltrates. This was case 22 in a previous study. 11 Case 25: Orbital Fat. Same patient reported in case 20. This tissue sample was case 23 in a previous study. 11
Results The results obtained with the anticomplement antibodies are summarized in Table 1 and photographically arranged in Figure 1. The pattern of reactivity observed from all nine Graves extraocular and periocular muscles with antiC3bi was an intense, localized oval to spindle-shaped reaction that seemed to represent cells on a diffuse staining background of the endomysia! and perimysial connective tissues (Fig 1). Intracellular staining of internally processed C3bi molecules initially attached to a complement receptor may explain the reactivity8 of these purported cellular elements. Graves cases 5, 6, and 8 showed less focal staining and more diffuse reactivity. All positive reactions seen were localized to the endomysium and perimysium. Distinct C3bi positivity was found in Graves muscle, ranging from 2+ (moderate) to 4+ (intense) staining. Six of 14 control non-Graves tissue samples showed relatively weaker staining ranging from 1+ to 3+. Of the positive controls, three cases (cases 13, 15, and 19) showed focal staining indicative of cells while the remaining three cases
1329
Ophthalmology
Volume 99, Number 8, August 1992
showed diffuse, nonfocal reactivity. Only two of the Graves muscle samples (cases 5 and 8) reacted minimally with C5b-9 terminal attack complex in a diffuse pattern, very unlike the intense, focal pattern observed with antiC3bi. None of the control extraocular muscle reacted with the C5b-9 terminal attack complex. Orbital fat from Graves and control patients did not show reactivity with any of the antibodies tested (Table 1).
C3bi antibodies is not due to Fe or nonspecific binding. Weetman et al, 13 using immunohistochemical studies from frozen sections, showed the terminal complement complex (C5b-9) around the thyroid follicles in thyroidectomy specimens from 6 of 6 Graves patients and 2 of 2 Hashimoto's thyroiditis patients while no staining from 2 of 2 patients with normal thyroid tissue was reported. This finding when compared with the relative absence of C5b-9 from Graves ophthalmopathy ocular tissues observed in this study supports the theory that Graves ophthalmopathy may be a fundamentally distinct illness from Graves disease. As shown in Figure 2, IgA 1 can start the formation of C3bi by activating the alternative pathway C3 convertase (C3b, Bb), 7 which cleaves C3, generating C3b and C3a fragments. The C3b molecule transiently acquires a capacity to bind covalently to cell surfaces and immune complexes by a transacylation reaction involving an internal thiolester. C3bi is formed from C3b by the inactivation of C3b by factor I and requires one of several cofactor molecules (membrane cofactor protein, CRl, CR2, H, or C4-binding protein) that induce conformational changes in C3b necessary for I to exert its action. 8 Binding
Discussion The data presented indicate that C3bi, and not C5b-9, is accumulated in Graves orbital and periorbital muscles, but not to any significant degree in non-Graves ocular muscle. When we tested for the presence of four immunoglobulins; IgAh IgA2 , lgG, and IgM, only IgA 1 reactivity was found. The fixation in Zenker's fluid used in this and previous studies from this group 11 • 12 may selectively preserve some epitopes. Therefore, the absence of staining for IgA 2 , IgG, and IgM should not be taken to indicate the definite absence of these molecules. Although it does demonstrate that the reactivity with anti-IgA 1 and anti-
Classical Pathway Activation Antigen I Antibody Complexes C2 + C4
~
C 1 - - - - I.... Activated
C1
Figure 2. Schematic representation of the complement cascades showing that activation of the alternative complement pathway by IgA 1 can produce C3bi, which by deposition in the interstitial space and adherence to immune cells via complement receptors may contribute to the pathogenesis of Graves ophthalmopathy.
.
C4b2a C3 convertase)
C5 C5 convertase
l
C5a + C5b
C6-9
C5b-9 Terminal attack complex
1330
Rosen et al · C3bi Involvement in Graves Ophthalmopathy of these complement regulatory proteins to C3b leads either to amplification of the C3 convertase and initiation of the membrane attack complex (C5), or to the inactivation of C3b and the generation of C3bi. Whether amplification or inactivation occurs depends on the nature of the surface to which C3b is fixed 8 and overall favors factor B binding to C3b (allowing formation of the C3b, Bb convertase) over factor H binding. 14 We have observed distinct patterns of staining with anti-lgA 1 and anti-C3bi human monoclonal immunoglobulin in muscle from patients with Graves ophthalmopathy, with essentially no C5b-9 staining suggesting an absence of complete complement activation and possible ligand activity by C3bi. An autoimmune process whereby immunoglobulin is produced to autoantigens with concomitant complement activation secondary to trauma and phthisis bulbi may explain the presence of C3bi staining in control cases 13 and 15, respectivelyY Four different complement receptors are present on various cells: CR 1 on monocytes, eosinophils, Langerhans cells, B cells, some T cells, and erythrocytes; CR2 on B cells; CR3 on neutrophils, macrophages, monocytes, natural killer cells, and antibody-dependent cell cytotoxicity (ADCC) effector lymphocytes; CR4 on monocytes, neutrophils, macrophages, natural killer cells, and ADCC effector lymphocytes. 8 The C3bi component binds to all complement receptors with greatest specificity for the CR3 receptor. 8 CR3 is structurally different from the other complement receptors and belongs to the leukocyte integrin (LFA-1) family. CR3 has a major role in adhesion of myeloid cells to opsonized particles as well as an important role in cell adhesion functions. 16 The integrin family is composed of structurally related, cell surface heterodimers that are involved in both the attachment of cells to extracellular matrix proteins as well as in cell to cell interactions. 17 It may be that lgA 1 bound to Graves ophthalmopathy extraocular and periocular muscle perimysial and endomysia! connective tissue activates the alternative complement system and forms C3bi. Subsequently, C3bi could mediate the interaction of various immune cells via complement receptors with the extracellular matrix of the interstitial space thus contributing to the pathophysiology of Graves ophthalmopathy (Fig 2). This is the first report to demonstrate C3bi involvement in Graves ophthalmopathy. Further investigations are under way to determine if the focal pattern ofC3bi, compared with the diffuse staining pattern observed most frequently with IgA 1 represents immune cells. If this is the case, monoclonal cell markers should afford a methodology to unfold the cellular immunologic mechanisms underlying Graves ophthalmopathy.
References 1. Rosen CE, Burde RM. Pathophysiology and etiology of Graves ophthalmopathy. In: Falk SA, ed. Thyroid Disease: Endocrinology, Surgery, Nuclear Medicine, and Radiotherapy. New York: Raven Press, 1990; 255-63. 2. Solomon DH, Chopra IJ, Chopra U, Smith FJ. Identification of subgroups of euthyroid Graves' ophthalmopathy. N Eng! J Med 1977;296:181-6. 3. Salvi M, Zhang ZG, Haegart D, et al. Patients with endocrine ophthalmopathy not associated with overt thyroid disease have multiple thyroid immunological abnormalities. J Clin Endocrinol Metab 1990:89-94. 4. Trokel SL, Jakobiec FA. Correlation of CT scanning and pathologic features of ophthalmic Graves disease. Ophthalmology 1981 ;88:553-64. 5. Neigel JM, Rootman J, Belkin RI, et al. Dysthyroid optic neuropathy. The crowded orbital apex syndrome. Ophthalmology 1988;95:1515-21. 6. Fearon D, David J. The complement system. In: Rubenstein E, Federmen DD, eds. Scientific American Medicine. New York: Scientific American, Inc., 1989; Vol II, Chap 6:1-9. 7. Hiemstra PS, Biewenga J, Gorter A, et al. Activation of complement by human serum lgA, secretory lgA and lgA 1 fragments. Mol Immunol 1988;25:527-33. 8. Lambris JD. The multifunctional role ofC3, the third component of complement. Immunol Today 1988;9:387-93. 9. Werner SC, Wegelius 0, Fierer JA, Hsu KC. Immunoglobulins (E, M, G) and complement in the connective tissues of the thyroid in Graves's disease. N Eng! J Med 1972;287: 421-5. 10. Wang PW, Hiromatsu Y, Laryea E, et al. Immunologically mediated cytotoxicity against human eye muscle cells in Graves ophthalmopathy. J Clin Endocrinol Metab 1986;63: 316-22. 11. Rosen CE, Raikow RB, Burde RM, et al. Immunohistochemical evidence for lgA 1 involvement in Graves ophthalmopathy. Ophthalmology 1992;99: 146-52. 12. Raikow RB, Dalbow MH, Kennerdell JS, et al. Immunohistochemical evidence for lgE involvement in Graves' orbitopathy. Ophthalmology 1990;97:629-35. 13. Weetman AP, Cohen SB, Oleesky DA, Morgan BP. Terminal complement complexes and C1/C1 inhibitor complexes in autoimmune thyroid disease. Clin Exp Immunol 1989;77:25-30. 14. Pangburn MK, Muller-Eberhard HJ. The alternative pathway of complement. Springer Semin Immunopathol1984;7: 163-92. 15. Grisanti S, Wiedermann P, Weller M, et al. The significance of complement in proliferative vitreoretinopathy. Invest Ophthalmol Vis Sci 1991;32:2711-7. 16. Horejsi V. Surface antigens of human leukocytes. Adv Immunol1991;49:75-147. 17. Shevach EM. Accessory molecules. In: Paul WE, ed. Fundamental Immunology, 2nd ed. New York: Raven Press, 1989; 413-41.
1331
Although the literature is replete with clinical and immunologic data concerning Graves ophthalmopathy, the pathophysiology remains unclear. 1 Graves ophthalmopaOriginally received: January 16, 1992. Revision accepted: March 17, 1992. 1 Department of Ophthalmology, Montefiore Medical Center, Albert Einstein College of Medicine, Bronx. 2 Department of Ophthalmology, Allegheny General Hospital and Allegheny-Singer Research Institute, Pittsburgh. Supported in part by an unrestricted departmental grant to the Department of Ophthalmology, Albert Einstein College of Medicine, Montefiore Medical Center from Research to Prevent Blindness, Inc, New Yark, New York. Reprint requests to Carl E. Rosen, MD, Department of Ophthalmology, Montefiore Medical Center, Albert Einstein College of Medicine, Ill E 2 lOth St, Bronx, NY.
thy is usually associated with Graves hyperthyroidism, however ophthalmopathy also may occur in patients with primary hypothyroidism or Hashimoto's thyroiditis, and is sometimes found in the apparent absence of thyroid disease. 2•3 The principal pathologic feature of Graves ophthalmopathy is inflammation of the orbital soft tissues and muscles. Inflammation of the extraocular muscles seems to be restricted to the muscle bellies, with sparing oftendons as visualized by neuroimaging techniques. 4 Disease severity parallels the extent of optic nerve compression. 5 The complement system is the principal humoral effector of immunologically induced inflammation. It plays a crucial role both in immunologically induced and nonspecific resistance to infections and in the pathogenesis oftissue injury. Complement activation can be initiated by either of two pathways: the classic pathway or the al-
1325
Ophthalmology
Volume 99, Number 8, August 1992
ternative pathway. Both pathways lead to the activation of the pivotal complement component C3. The classic pathway is activated by immunoglobulin (lgG~. lgG2 , lgG 3, lgM) when bound to antigen, while the alternative pathway is activated by a variety of bacteria, yeast, parasites, insoluble immune complexes, 6 and antibodies such as lgA 1 and IgA 2 •7 The C3 component can be fragmented into C3b, which serves as a branch point for the complement cascade to either form C3bi or the terminal attack complex (C5b-9). 6 The C3bi complement fragment can bind to complement receptors present on inflammatory cells as well as connective tissues and function as an adhesive ligand. 8 The role of complement in the pathogenesis ofGraves ophthalmopathy is unknown. Werner et al 9 discovered the presence of lgE, lgM, IgG and complement components C1q and C3 in the connective tissues ofthe thyroid gland in patients with Graves disease. Wang et al 10 examined the possible role of antibody-mediated complement-dependent cytotoxicity in the pathogenesis of Graves ophthalmopathy by demonstrating that serum from patients with Graves ophthalmopathy caused a significant increase in eye muscle cell lysis when compared with control subjects. Recently, the antibody lgA 1 was discovered immunohistochemically in the endomysia! and surrounding connective tissues of Graves ophthalmopathy extraocular muscle. 11 We now report immunohistochemical evidence for the presence of the complement component C3bi in Graves extraocular and periocular muscle tissues in a focal, oval- to spindle-shaped reaction that seemed to represent cells on a diffuse staining background of the endomysia! and perimysial connective tissues with no staining of the muscle fibers themselves.
Materials and Methods Tissue Specimens Tissue was removed from 10 Graves ophthalmopathy patients to improve their vision and appearance. Twelve patients without Graves disease provided control orbital tissue, which was removed in association with surgery for strabismus, blepharoptosis, fat prolapse, or enucleation because of phthisis. Our use oftissue did not in any way impinge on diagnosis or treatment. The tissues were placed into sterile, moist gauze in the operating room and fixed within 3 hours. Tissue was fixed in Zenker's fluid, embedded in paraffin, cut into 4 Jim sections, and placed onto slides. All antibody reactions were performed at room temperature (Table I). Anti-C3bi and anti-C5b-9 monoclonal antibodies were used as primary antibodies in the 10 Graves cases and the 12 control cases, as described below. Secondary antibodies and substrates were used as recommended by their suppliers. A 30-minute block with 2% instant nonfat milk powder in phosphate-buffered saline was used to reduce nonspecific antibody adhesion to tissue. Slides were washed twice in a gently agitating bath of phosphate-buffered saline for 10 minutes in between
1326
each step. All slides were counter-stained with Meyer's hematoxylin. Aqueous mounting medium (Glycergel, DAKO Corp, Santa Barbara, CA) was used.
Primary Antibodies Mouse monoclonal antibodies, anti-human C3bi (Quidel Corp, San Diego, CA, A208) and anti-human C5b-9 (Quidel Corp, A239) were diluted to 10 JLg/ml using phosphate-buffered saline with 1% sodium azide and reacted with deparaffinized tissue sections for 2 hours. Bound primary antibody was then visualized by reacting the slides with Biogenex Super Sensitive StrAvigenKit AA000-5M (San Ramona, CA) using alkaline phosphatase, according to instructions provided by the supplier. Previous studies 11 have shown that there is no endogenous alkaline phosphatase activity in Graves or control orbital tissues after Zenker's fixation.
Case Reports Graves Ophthalmopathy Patients Case 1: Levator Muscle. A 54-year-old woman with severe Graves ophthalmopathy had a prior left lateral rectus resection and a left medial rectus recession for a large esotropia 18 years previously. The patient presented with bilateral proptosis, diplopia, and blepharoptosis on the right side with lid retraction on the left side. Computed tomography showed marked involvement of the right inferior and medial rectus muscles. Results of a tensilon test were negative. At surgery, a right inferior rectus recession, a right levator resection, and a left Mueller's muscle excision with levator recession were performed. This was case I in a previous study" in which the presence oflgA 1 was demonstrated in the connective tissues surrounding the muscle fibers. Case 2: Orbicularis Muscle. A 54-year-old woman presented with a history of thyroid disease and Graves ophthalmopathy, which was first diagnosed 2 years before she had a three-wall orbital decompression. Three years after the decompression, an excision of orbicularis muscle and lid fat was done to correct lid retraction and corneal exposure. This was case 2 in a previous study 11 in which the presence of lgA 1 was demonstrated in the connective tissues surrounding the muscle fibers. Case 3: Orbicularis Muscle. A 55-year-old woman with Graves hyperthyroidism had a 2-year history of ophthalmopathy. She had previously been treated with 113I. She presented with lid edema, tearing, and blurry vision. Prednisone did not ameliorate her symptoms, and orbital radiation was performed I year before a bilateral blepharoplasty with removal of orbital fat. This was case 3 in a previous study 11 in which the presence of lgA 1 was demonstrated in the connective tissues surrounding the muscle fibers. Case 4: Levator Muscle. A 51-year-old woman with diabetes mellitus had hypothyroid disease in the distant past. She was diagnosed with moderate Graves ophthalmopathy I year before she underwent bilateral resection of the levator muscles for blepharoptosis and edematous eyelids. This tissue was case 4 in a previous study 11 in which the presence of lgA 1 was demonstrated in the connective tissues surrounding the muscle fibers.
Rosen et al · C3bi Involvement in Graves Ophthalmopathy Case 5: Levator Muscle. A 76-year-old woman with longstanding Graves hyperthyroidism had a two-wall orbital decompression 16 years ago to treat severe Graves ophthalmopathy with bilateral compressive optic neuropathy, esotropia, and lid retraction. For cosmetic purposes, a Mueller's muscle excision with a medial and lateral marginal myotomy of levator muscle was done. This tissue was case 5 in a previous study 11 in which the presence oflgA 1 was demonstrated in the connective tissues surrounding the muscle fibers. Case 6: Mueller's Muscle. Same patient reported in case 5. This tissue was case 6 in a previous study 11 in which the presence of lgA 1 was demonstrated in the connective tissues surrounding the muscle fibers. Case 7: Medial Rectus Muscle. A 72-year-old man presented with a diagnosis of hyperthyroidism and severe Graves ophthalmopathy with prolapsed conjunctiva and lid retraction of I years' duration. Prednisone was administered on a trial basis and yielded good results. However, I year later, the patient experienced diplopia, and a medial rectus recession with an adjustable suture was performed on the left eye, which was followed
5 months later by the same procedure on the right eye. This tissue was case 7 in a previous study 11 in which the presence of lgA 1 was demonstrated in the connective tissues surrounding the muscle fibers. Case 8: Lateral Rectus Muscle. A 66-year-old woman with hypothyroidism had Graves ophthalmopathy of I year's duration. On presentation, compressive optic neuropathy was diagnosed, and a bilateral orbital wall decompression was performed. The present tissue originated from a subsequent recession of the left medial rectus muscle and a resection of the left lateral rectus muscle. This tissue was case 8 in a previous study'' in which the presence oflgA 1 was demonstrated in the connective tissues surrounding the muscle fibers. Case 9: Medial Rectus Muscle. A 51-year-old woman presented with a history of Graves disease of more than I 0 years' duration. Graves ophthalmopathy was diagnosed with bilateral proptosis, chemosis, and conjunctival injection and left lid retraction. A bilateral orbital decompression was performed at that time. Two years later, progressive hmizontal diplopia developed and a left medial rectus recession was performed.
Table 1. Summary of Tissue Reactions Case No. Age (yrs)/Sex Gravt!s 1. 54/F 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.
54/F 55/F 51/F 76/Ft 76/F 72/M 66/F 51/F 57 /F 52/F
Control 12. 29/M 13. 33/M 14. 29/M 15. 65/F 16. 1/F 17. 2/F 18. 2/F 19. 69/F 20. 10/M 21. 35/M 22. 23/F 23. 35/F 24. 42/M 25. 10/M
Tissue
lgA1
lgAz
lgM
lgG
Orbicularis/ levator Orbicularis Orbicularis Levator Levator Mueller's Medial rectus Lateral rectus Medial rectus Orbital fat Orbital fat
4*
0
0
4 2 4 2 2 3 4 2 0 0
0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0
0 0 0 0 0
Levator Medial rectus Medial rectus Lateral rectus Lateral rectus Inferior oblique Medial rectus Orbicularis Mueller's Medial rectus Medial rectus Medial rectus Orbital fat Orbital fat
0 1 2 0 0 1 1 3 0 0 1 3 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0
0
0 0 0 1 0 0 0
0 0 0 0 0
0
0 0
C3bi
C5b-9
2
0
4 2 4 4 2 4 3 2 0 0
0 0 0 1 0 0 1 0 0 0
0 2 1 3 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0
0 0 0 0 0
1 0 0 0 0 0 0
The majority of the tissue sample reactions with monoclonal antibodies anti-IgA" anti-IgA2 , anti-IgM, and anti-IgG were from a previous study 11 * Tissue staining intensity scale from 4 being most reactiv!-! to 0 being nonreactive.
t
Cases 5 and 6, 17 and 18, and 20 and 25 are tissue specimens taken from three different patients, respectively.
1327
·•
,\._
'{
~.~I} .,
. """,•'·" ~
·~
1328
...
.j • . ,
...... ,
' :·a::
'""
.
Rosen et al · C3bi Involvement in Graves Ophthalmopathy Figure 1. Antibody reactions with Graves and control ocular tissue. Shown here are representative tissue samples from reactions with antibodies to C3bi and CSb-9; no anti-immunoglobulin specimens are included (Table 1). Each column represents the same muscle type and each row represents a reaction with a specific anticomplement antibody. All reactions shown followed the described method (see Materials and Methods section). A, levator muscle from Graves patient (case 4) shown after reaction with anti-human C3bi demonstrating a localized spindle-shaped pattern which may represent cells on a diffuse perimysial and epimysia! staining background reaction. B, levator muscle from same Graves patient (case 4) with no staining evident after reaction with anti-human CSb-9. C, D, levator muscle from control patient (case 12) shows no staining after reaction with antihuman C3bi (C) or anti-human CSb-9 (D). E, F, orbicularis muscle from Graves patient (case 2) shown after reaction with anti-human C3bi (E) and anti-human CSb-9 (F). G, H, orbicularis muscle from control patient (case 19) shown after reaction with anti-human C3bi (G) and anti-human CSb9 (H). I, J, medial rectus muscle from Graves patient (case 7) shown after reaction with anti-human C3bi (I) and anti-human CSb-9 (J). K, L, medial rectus muscle from control patient (case 21) shown after reaction with anti-human C3bi (K) and anti-human CSb-9 (L). M, N, lateral rectus muscle from Graves patient (case 8) shown after reaction with anti-human C3bi (M) and anti-human CSb-9 (N). 0, P, lateral rectus muscle from control patient (case 16) shown after reaction with anti-human C3bi (0) and anti-human CSb-9 (P). Q, R, Mueller's muscle from Graves patient (case 6) shown after rection with anti-human C3bi (Q) and CSb-9 (R). S, T, Mueller's muscle from control patient (case 20) shown after reaction with antihuman C3bi (S) and anti-human CSb-9 (T). U, orbicularis muscle from Graves patient (case 2) shown after reaction with anti-human C3bi demonstrating a spindle-shaped staining pattern which may represent cells. V, levator muscle from Graves patient (case 4) shown after reaction with anti-human C3bi. W, medial rectus muscle from Graves patient (case 7) shown after reaction with anti-human C3bi (A-T, original magnification, XlOO) (U-W, original magnification, X250).
Case 10: Orbital Fat. A 57-year-old woman, who had undergone a prior thyroidectomy, presented with proptosis and diplopia. Severe Graves ophthalmopathy was diagnosed with decreased extraocular motility including right hypertropia and bilateral lid retraction. At surgery, a bilateral Mueller's excision, right levator marginal myotomy, and a left inferior rectus recession with an adjustable suture were performed. The pathology report of the Mueller's specimen indicated the presence of fat and very little muscle. This tissue was case 12 in a previous study. 11 Case 11: Orbital Fat. A 52-year-old woman presented with a 7-year history of hypothyroidism and severe Graves ophthalmopathy with bilateral proptosis. She had a Mueller's excision for lid retraction 2 years before the present surgery. The patient presented with orbital fat prolapse, which was removed. This tissue was case 13 in a previous study. 11
Control (Non-Graves) Patients Case 12: Levator Muscle. A 29-year-old man with an unremarkable medical history underwent correction for congenital blepharoptosis. This tissue was case 14 in a previous study. 11 Case 13: Medial Rectus Muscle. A 33-year-old man experienced accidental head trauma. He required recession of the lateral rectus and resection of the medial rectus muscles to correct subsequent third nerve aberrant regeneration. This was case 15 in a previous study. 11 Case 14: Medial Rectus Muscle. A 29-year-old man underwent a resection of the medial rectus muscle to correct an esotropia. The specimen was found to be fibrotic with a small amount of disrupted muscle. This was case 16 in a previous study. 11 Case 15: Lateral Rectus Muscle. A 65-year-old woman underwent enucleation of the right eye secondary to phthisis bulbi. This was case 17 in a previous study. 11 Case 16: Lateral Rectus Muscle. A 1112-year-old girl with an unremarkable medical history had a left lateral rectus resection for uncomplicated strabismus. This was case 18 in a previous study. 11 Case 17: Inferior Oblique Muscle. A 2-year-old girl with an unremarkable medical history had a right inferior oblique and a left medial rectus resection for uncomplicated strabismus. This was case 19 in a previous study. 11 Case 18: Medial Rectus Muscle. Same patient reported in case 17. This was case 20 in a previous study. 11
Case 19: Orbicularis Muscle. A 69-year-old woman with myasthenia gravis had a right orbicularis and levator resection for right upper lid blepharoptosis. Case 20: Mueller's Muscle. A 10-year-old boy had a left Mueller's resection for congenital blepharoptosis. Case 21: Medial Rectus. A 35-year-old man with a medical history of trauma 20 years previously had left medial rectus resection for left exotropia. Case 22: Medial Rectus. A 23-year-old woman with history significant for sickle cell trait disease had a right medial rectus resection for a right exotropia with anisometropic amblyopia. Case 23: Medial Rectus. A 35-year-old woman with no significant medical history had a right medial rectus resection for a right exotropia. Case 24: Orbital Fat. A 42-year-old man with a history of pseudotumor and corticosteroid treatment underwent orbital fat removal. The specimen consisted of fat with a number of cellular infiltrates. This was case 22 in a previous study. 11 Case 25: Orbital Fat. Same patient reported in case 20. This tissue sample was case 23 in a previous study. 11
Results The results obtained with the anticomplement antibodies are summarized in Table 1 and photographically arranged in Figure 1. The pattern of reactivity observed from all nine Graves extraocular and periocular muscles with antiC3bi was an intense, localized oval to spindle-shaped reaction that seemed to represent cells on a diffuse staining background of the endomysia! and perimysial connective tissues (Fig 1). Intracellular staining of internally processed C3bi molecules initially attached to a complement receptor may explain the reactivity8 of these purported cellular elements. Graves cases 5, 6, and 8 showed less focal staining and more diffuse reactivity. All positive reactions seen were localized to the endomysium and perimysium. Distinct C3bi positivity was found in Graves muscle, ranging from 2+ (moderate) to 4+ (intense) staining. Six of 14 control non-Graves tissue samples showed relatively weaker staining ranging from 1+ to 3+. Of the positive controls, three cases (cases 13, 15, and 19) showed focal staining indicative of cells while the remaining three cases
1329
Ophthalmology
Volume 99, Number 8, August 1992
showed diffuse, nonfocal reactivity. Only two of the Graves muscle samples (cases 5 and 8) reacted minimally with C5b-9 terminal attack complex in a diffuse pattern, very unlike the intense, focal pattern observed with antiC3bi. None of the control extraocular muscle reacted with the C5b-9 terminal attack complex. Orbital fat from Graves and control patients did not show reactivity with any of the antibodies tested (Table 1).
C3bi antibodies is not due to Fe or nonspecific binding. Weetman et al, 13 using immunohistochemical studies from frozen sections, showed the terminal complement complex (C5b-9) around the thyroid follicles in thyroidectomy specimens from 6 of 6 Graves patients and 2 of 2 Hashimoto's thyroiditis patients while no staining from 2 of 2 patients with normal thyroid tissue was reported. This finding when compared with the relative absence of C5b-9 from Graves ophthalmopathy ocular tissues observed in this study supports the theory that Graves ophthalmopathy may be a fundamentally distinct illness from Graves disease. As shown in Figure 2, IgA 1 can start the formation of C3bi by activating the alternative pathway C3 convertase (C3b, Bb), 7 which cleaves C3, generating C3b and C3a fragments. The C3b molecule transiently acquires a capacity to bind covalently to cell surfaces and immune complexes by a transacylation reaction involving an internal thiolester. C3bi is formed from C3b by the inactivation of C3b by factor I and requires one of several cofactor molecules (membrane cofactor protein, CRl, CR2, H, or C4-binding protein) that induce conformational changes in C3b necessary for I to exert its action. 8 Binding
Discussion The data presented indicate that C3bi, and not C5b-9, is accumulated in Graves orbital and periorbital muscles, but not to any significant degree in non-Graves ocular muscle. When we tested for the presence of four immunoglobulins; IgAh IgA2 , lgG, and IgM, only IgA 1 reactivity was found. The fixation in Zenker's fluid used in this and previous studies from this group 11 • 12 may selectively preserve some epitopes. Therefore, the absence of staining for IgA 2 , IgG, and IgM should not be taken to indicate the definite absence of these molecules. Although it does demonstrate that the reactivity with anti-IgA 1 and anti-
Classical Pathway Activation Antigen I Antibody Complexes C2 + C4
~
C 1 - - - - I.... Activated
C1
Figure 2. Schematic representation of the complement cascades showing that activation of the alternative complement pathway by IgA 1 can produce C3bi, which by deposition in the interstitial space and adherence to immune cells via complement receptors may contribute to the pathogenesis of Graves ophthalmopathy.
.
C4b2a C3 convertase)
C5 C5 convertase
l
C5a + C5b
C6-9
C5b-9 Terminal attack complex
1330
Rosen et al · C3bi Involvement in Graves Ophthalmopathy of these complement regulatory proteins to C3b leads either to amplification of the C3 convertase and initiation of the membrane attack complex (C5), or to the inactivation of C3b and the generation of C3bi. Whether amplification or inactivation occurs depends on the nature of the surface to which C3b is fixed 8 and overall favors factor B binding to C3b (allowing formation of the C3b, Bb convertase) over factor H binding. 14 We have observed distinct patterns of staining with anti-lgA 1 and anti-C3bi human monoclonal immunoglobulin in muscle from patients with Graves ophthalmopathy, with essentially no C5b-9 staining suggesting an absence of complete complement activation and possible ligand activity by C3bi. An autoimmune process whereby immunoglobulin is produced to autoantigens with concomitant complement activation secondary to trauma and phthisis bulbi may explain the presence of C3bi staining in control cases 13 and 15, respectivelyY Four different complement receptors are present on various cells: CR 1 on monocytes, eosinophils, Langerhans cells, B cells, some T cells, and erythrocytes; CR2 on B cells; CR3 on neutrophils, macrophages, monocytes, natural killer cells, and antibody-dependent cell cytotoxicity (ADCC) effector lymphocytes; CR4 on monocytes, neutrophils, macrophages, natural killer cells, and ADCC effector lymphocytes. 8 The C3bi component binds to all complement receptors with greatest specificity for the CR3 receptor. 8 CR3 is structurally different from the other complement receptors and belongs to the leukocyte integrin (LFA-1) family. CR3 has a major role in adhesion of myeloid cells to opsonized particles as well as an important role in cell adhesion functions. 16 The integrin family is composed of structurally related, cell surface heterodimers that are involved in both the attachment of cells to extracellular matrix proteins as well as in cell to cell interactions. 17 It may be that lgA 1 bound to Graves ophthalmopathy extraocular and periocular muscle perimysial and endomysia! connective tissue activates the alternative complement system and forms C3bi. Subsequently, C3bi could mediate the interaction of various immune cells via complement receptors with the extracellular matrix of the interstitial space thus contributing to the pathophysiology of Graves ophthalmopathy (Fig 2). This is the first report to demonstrate C3bi involvement in Graves ophthalmopathy. Further investigations are under way to determine if the focal pattern ofC3bi, compared with the diffuse staining pattern observed most frequently with IgA 1 represents immune cells. If this is the case, monoclonal cell markers should afford a methodology to unfold the cellular immunologic mechanisms underlying Graves ophthalmopathy.
References 1. Rosen CE, Burde RM. Pathophysiology and etiology of Graves ophthalmopathy. In: Falk SA, ed. Thyroid Disease: Endocrinology, Surgery, Nuclear Medicine, and Radiotherapy. New York: Raven Press, 1990; 255-63. 2. Solomon DH, Chopra IJ, Chopra U, Smith FJ. Identification of subgroups of euthyroid Graves' ophthalmopathy. N Eng! J Med 1977;296:181-6. 3. Salvi M, Zhang ZG, Haegart D, et al. Patients with endocrine ophthalmopathy not associated with overt thyroid disease have multiple thyroid immunological abnormalities. J Clin Endocrinol Metab 1990:89-94. 4. Trokel SL, Jakobiec FA. Correlation of CT scanning and pathologic features of ophthalmic Graves disease. Ophthalmology 1981 ;88:553-64. 5. Neigel JM, Rootman J, Belkin RI, et al. Dysthyroid optic neuropathy. The crowded orbital apex syndrome. Ophthalmology 1988;95:1515-21. 6. Fearon D, David J. The complement system. In: Rubenstein E, Federmen DD, eds. Scientific American Medicine. New York: Scientific American, Inc., 1989; Vol II, Chap 6:1-9. 7. Hiemstra PS, Biewenga J, Gorter A, et al. Activation of complement by human serum lgA, secretory lgA and lgA 1 fragments. Mol Immunol 1988;25:527-33. 8. Lambris JD. The multifunctional role ofC3, the third component of complement. Immunol Today 1988;9:387-93. 9. Werner SC, Wegelius 0, Fierer JA, Hsu KC. Immunoglobulins (E, M, G) and complement in the connective tissues of the thyroid in Graves's disease. N Eng! J Med 1972;287: 421-5. 10. Wang PW, Hiromatsu Y, Laryea E, et al. Immunologically mediated cytotoxicity against human eye muscle cells in Graves ophthalmopathy. J Clin Endocrinol Metab 1986;63: 316-22. 11. Rosen CE, Raikow RB, Burde RM, et al. Immunohistochemical evidence for lgA 1 involvement in Graves ophthalmopathy. Ophthalmology 1992;99: 146-52. 12. Raikow RB, Dalbow MH, Kennerdell JS, et al. Immunohistochemical evidence for lgE involvement in Graves' orbitopathy. Ophthalmology 1990;97:629-35. 13. Weetman AP, Cohen SB, Oleesky DA, Morgan BP. Terminal complement complexes and C1/C1 inhibitor complexes in autoimmune thyroid disease. Clin Exp Immunol 1989;77:25-30. 14. Pangburn MK, Muller-Eberhard HJ. The alternative pathway of complement. Springer Semin Immunopathol1984;7: 163-92. 15. Grisanti S, Wiedermann P, Weller M, et al. The significance of complement in proliferative vitreoretinopathy. Invest Ophthalmol Vis Sci 1991;32:2711-7. 16. Horejsi V. Surface antigens of human leukocytes. Adv Immunol1991;49:75-147. 17. Shevach EM. Accessory molecules. In: Paul WE, ed. Fundamental Immunology, 2nd ed. New York: Raven Press, 1989; 413-41.
1331
