Immunology of Neurologic and Endocrine Diseases that Affect the Eye
SHARON K. SRA, KARAN K. SRA, MITCHELL FRIEDLAENDER and STEFAN D. TROCME
Table Of Contents
|Immunologic mechanisms in neurologic and endocrine diseases have become a focus of interest and investigation. In multiple sclerosis (MS), an infectious agent may trigger an immune response that affects the tissues of the nervous system. In other disorders, autoantibodies or immune cells may interact with and destroy host tissues. Although the complete pathogenesis of these diseases has not been fully understood, innovative research has led to the development of new methods of diagnosis and treatment.|
|Multiple sclerosis is an inflammatory demyelinating disease of the central
nervous system. It is the most common neurologic disease affecting
young adults (20–40 years of age) and occurs more commonly
in whites, especially those of European descent. Women are also
more affected than men. The victims of this progressive, devastating disease
have debilitating symptoms that last for many years, with periods
of remission and exacerbation. The symptoms depend on the area affected
by demyelination. Patients may exhibit motor difficulties or disturbances
in sensation, and/or ocular disturbances. Ophthalmic findings, such
as optic neuritis and ocular motility disturbances, are often
among the early signs of the disease. There may also be an association
between MS and uveitis and/or retinal phlebitis.1,2|
MS is closely associated with geographic factors, with its prevalence increasing as the victim's distance from the equator increases, north or south. The risk of MS seems to be determined by some event that takes place in early childhood. If an adult moves from a high-risk to a low-risk area, the chance of MS developing is that of the original environment. However, if the move is made before the age of 15, the risk is that of the new environment. An environmental factor of infectious viral nature has been implicated from such observations.
Recently, much interest has focused on viruses as the possible cause of multiple sclerosis. The generation of a predominately cytotoxic cellular immune response suggests a viral role in inducing demyelination. Several theories have been hypothesized regarding the mechanism of how viruses may cause MS. Possible mechanisms include direct viral damage of oligodendroglia with resultant myelin degeneration and/or damage to the central nervous system (CNS) by mediators and cytokines released from inflammatory cells.3,4
The human herpesvirus type 6 (HHV-6) has recently received great attention as a possible cause of MS. HHV-6 is a neurotropic and latent lymphotropic virus affecting the CNS. It can remain latent or persistant and can be reactivated by stress or infection with other organisms.5 HHV-6 DNA has been found to be significantly more common in MS plaques compared to normal-appearing white matter, thus implicating a pathogenic role of HHV-6 in the development of MS.6 It has been hyposethized that T cells infected with HHV-6 have elevated proinflammatory gene expression, which would indirectly induce oligodendrocyte death.7 Other findings associating HHV-6 with MS include increased titers of antibody to HHV-6 in the serum and cerebrospinal fluid (CSF) of patients with MS and the demonstration of HHV-6 DNA in serum of MS patients.8,9 HHV-6 reactivation has also been correlated with MS exacerbation via modulation of IL-12 synthesis. Patients with active MS have been shown to exhibit significantly higher levels of serum IL-12 concentrations than patients with latent MS.10
Epstein-Barr virus (EBV), also a member of the herpes family, has also been associated with an increased risk of MS. It is a highly prevalent virus occurring worldwide. It is believed that EBV infection precedes the occurrence of the development of MS. Investigators have found that elevated serum levels of IgG antibodies to EBV viral capsid antigen and EBV nuclear antigens are the strongest predictors of the development of MS.11 A study has also suggested that EBV nuclear antigen-1 (EBNA-1) may be a target of oligoclonal bands. The investigators observed a distinctive oligoclonal antigen-specific banding pattern for EBNA-1 in patients with MS.12
Other viruses have also been suggested as the possible cause of MS, including Chlamydia pneumoniae and the measles virus. There is considerable evidence that C. pneumoniae and measles-specific antibody titers are elevated in CSF of patients with MS.13–15
Although the diagnosis of MS is based primarily on clinical presentation, magnetic resonance imaging (MRI) has become an invaluable diagnostic aid in the investigation of MS. MRI was recently recognized by the International Panel on MS as the most sensitive paraclinical test used in the diagnosis of MS.16 The standard of lesion detection during the course of MS is the focal enhancement of lesions in a MRI with FLAIR. MS plaques commonly can be observed in the periventricular and/or juxtacortical white matter; however, they can be seen anywhere in the CNS.17 Changes in cerebral perfusion before the formation of plaques has also been detected by newer MRI techniques,18 perhaps aiding in early diagnosis of the disease. Spinal cord MRI has also been advocated; however, the correlation between MRI findings and clinical findings remains weak.
One of the most common immunologic abnormalities of MS is an elevated synthesis of intrathecal immunoglobulins. The specificity of these antibodies, although speculated to be of viral origin, remains obscure. IgG oligoclonal bands, the most prevalent CSF abnormality, are detectable in 95% of patients with MS.19,20 The synthesis of IgG has been found in CSF lymphocytes obtained from patients with MS, and plaque lesions contain abnormally high levels of immunoglobulins.21,22 Although elevated levels of kappa and lambda free light chains correlate with MS, the prevalence of free kappa light chains may be more specific to support a clinical diagnosis of MS.23,24 All of these antibody studies suggest that there is a local synthesis of antibodies to an unknown antigen within the CNS.
Much of the evidence that the immune system participates in MS comes from the study of an experimental model of the disease known as experimental allergic encephalitis (EAE). This disease was first recognized approximately 40 years ago among humans who were given rabies vaccine containing rabbit brain cells. EAE has been produced in a variety of laboratory animals by injecting them with CNS tissues in complete Freund's adjuvant. Within 2 weeks of the injection, the animals begin to lose weight, develop paralysis, and ultimately die.
EAE is mediated by T lymphocytes and can be transferred passively with lymph node cells containing sensitized T cells. The protein responsible for sensitization is known as myelin basic protein. Some investigators have been able to identify leukocytes sensitized to myelin basic protein in patients with MS. Interestingly, myelin basic protein not only initiates EAE but also can prevent its initiation and can alleviate or even stop its symptoms after they have begun. Myelin basic protein has even been used to treat a few carefully selected cases of MS, but whether EAE is in fact a good model of MS is controversial.
Genetic factors may play an important role in MS. Although the exact gene remains unknown, it is known that the human leukocyte antigen (HLA) phenotype Dw2 is associated with increased risk for MS.25,26 Tumor necrosis factor-alpha (TNF-α) has also been implicated in MS by causing oligodendrocyte cell death. Recent studies have suggested that levels of p53 increase after stimulation with TNF-α resulting in apoptosis of oligodendrocytes.27
Immunomodulators, specifically interferon beta (IFN-β) and glatiramer, have been demonstrated to be effective in the treatment of MS. Glatiramer, a synthetic copolymer with an amino acid composition based on the structure of myelin basic protein, has been shown to significantly reduce enhancing lesions as measured by MRI.28,29 IFN-β has been established to significantly reduce the exacerbation rate in patients with relapsing-remitting MS and inhibit cognitive deteriation.30,31 However, prolonged treatment with IFN-β may induce neutralizing antibodies against INF-β, thus reducing its effectiveness.32,33
|Acute optic neuritis is more frequently associated with MS than with any
other disease entity, although the reported incidence of the association
varies widely.34 However, the probability that patients with optic neuritis will develop
MS appears to increase with time—the longer they are followed-up, the
more MS is found. In a study by the Optic Neuritis Study
Group, the 10-year risk for the development of MS after an initial
episode of acute optic neuritis was 38%, with the risk increasing
to 56% in patients who had at least one lesion on MRI. A
higher number of lesions was not associated with an increased risk.35|
Patients with optic neuritis usually present with sudden onset of monocular visual loss, pain with extraocular movement, and an afferent papillary defect unless the optic neuropathy is bilateral. The retrobulbar form of optic neuritis occurs more commonly and is associated with a normal optic disc appearance. Causes of optic neuritis other than MS include viral, vasculitic, or granulomatous processes.
Oligoclonal IgG can be found in the CSF of 67% of patients with optic neuritis.36 Patients may have elevated intrathecal titers of viral antibody to measles, varicella zoster, rubella, and mumps.37 Many patients with optic neuritis also demonstrate free kappa and lambda oligoclonal bands in their CSF.38
The histocompatibility antigens HLA-A3, HLA-A7, and HLA-LD-a have been reported to be increased in optic neuritis and MS.39 In other studies, no significant differences were found in HLA distribution between patients with optic neuritis and controls.40,41 Such discrepancies are probably caused by the different criteria that have been used by different observers to establish the diagnosis of optic neuritis.
Optic neuritis has also been reported in association with the Guillain-Barré syndrome.42,43 In vitro cellular immunity to central and peripheral nervous tissue myelin has been demonstrated by the macrophage migration inhibition test.
Corticosteroid therapy in the treatment of optic neuritis has been evaluated in Optic Neuritis Treatment Trial (ONTT). Patients treated with standard-dose oral prednisone did not improve vision. Further, patients treated with oral prednisone were found to have an increased rate of new attacks of optic neuritis. Patients treated with intravenous (IV) corticosteroids followed by oral prednisone recovered vision more rapidly in the first 2 weeks, but there was no long-term benefit to vision. Visual recovery can occur without any treatment. The recovery is rapid initially and visual improvement continues for up to 1 year.44
|Myasthenia gravis (MG) is a chronic disease characterized by
abnormal fatigability of striated muscle. The disease may last for many
years and includes several remissions. It may be generalized or limited
to a single muscle group. Severe wasting and weakening may eventually
develop, and death may occur if the respiratory muscles are affected. Ptosis, the
most common sign of MG, and extraocular muscle abnormalities
are seen in MG. Recent investigations have suggested that autoimmune
factors are important in the pathogenesis of MG. Patients develop
antibodies directed toward acetylcholine receptors at the neuromuscular
junction of skeletal muscles.|
In 1960, Simpson45 suggested an autoimmune basis for MG because of its association with other autoimmune disorders such as systemic lupus erythematosus, rheumatoid arthritis, pemphigus, pernicious anemia, and myxedema. The coexistence of MG and thyroid autoimmune diseases has also been well recognized. Up to 10% of patients with MG develop Graves disease, whereas only up to 0.2% of patients with Graves disease develop MG.46
Antibodies to acetylcholine receptors (AChR) are present in as many as 85% of patients with MG.47 The injection of sera from myasthenic patients into experimental animals produce electrophysiologic changes characteristic of MG, including reduction of successive muscle action potentials and postjunctional sensitivity to acetylcholine.48 The active fraction in the serum has been identified as IgG, and its effect can be increased by the third component of complement. This and other experiments suggest that MG is the result of an antibody-mediated autoimmune attack on acetylcholine receptors at the neuromuscular junction.
The neuromuscular defect in MG is a result of a reduced number of functional AChR occurring by three different mechanisms: (1) accelerated endocytosis and degradation of AchR; (2) antibody blockage of acetylcholine receptor sites on the AchR; and (3) complement-mediated damage to the postsynaptic membrane. The cross-linking of the Fab fragment of the antibody with the AChR stimulates the internalization process of endocytosis and degradation by lysosomal enzymes. This reduces the half-life of AChR. In MG, the antibodies are directed against the a subunit of the AChR. By blocking the binding of the acetycholine to AChR, the antibody also inhibits the function of the AChR. The Ab-AChR complex initiates a complement-mediated destruction of postsynaptic membrane, resulting in a reduction of junctional folds.49
Abnormalities of the humoral immune system are probably the most important abnormalities in MG, but the role of cellular immunity has also been undergoing investigation and is becoming increasingly more apparent. Peripheral lymphocytes from myasthenic patients are stimulated when exposed to purified acetylcholine receptor in vitro.50 Delayed hypersensitivity to muscle and thymic antigens often occurs in patients with MG, and approximately half of all myasthenic patients show cellular immunity to myelin basic protein.51,52 Lymphocytes from affected patients are cytotoxic to fetal muscle tissue when stimulated with phytohemagglutinin.53 All of these investigations show the possible importance of cellular immunity to patients with MG, and the possible importance of an autoimmune mechanism is suggested by the fact that immune complexes containing IgG and C3 have been localized to the motor end plates of such patients.54
Thymic abnormalities are reported in 80% of myasthenic patients: 70% have thymic hyperplasia with increased numbers of germinal centers, and 10% have true thymomas. These hyperplastic thymuses contain high levels of B lymphocytes. Two-thirds of the patients who undergo thymectomy have complete or partial remission of MG, suggesting that the thymus is a site for the production of a neuromuscular blocking agent. Soluble fractions of thymus gland show immunologic cross-reactivity with acetylcholine-receptor protein.
HLA typing of patients with MG shows that they have HLA-B3 more often than do normal subjects.55 This association is particularly prevalent in females in whom the disease begins early and correlates with the presence of thymic follicular lymphoid hyperplasia. HLA-B3 is increased in males in whom the disease begins late and is common among myasthenics with thymomas.
Plasmapheresis has been used with considerable success in severe cases of MG. Presumably, this technique removes acetylcholine receptor antibodies or immune complexes that contribute to the patient's symptoms.
|Some forms of diabetes mellitus may be associated with autoimmunity but
just how closely is not known. This and other immunologic abnormalities
in diabetes may be related to the presence of antibodies to insulin
and to a variety of autoantibodies.|
Patients with diabetes show a greater than normal frequency of autoimmune disease of the thyroid gland, the adrenal glands, and the gastric mucosa (pernicious anemia). Thyroid microsomal antibody and gastric parietal cell antibody have been found, respectively, in 20% and 16% of patients with diabetes mellitus.56 These antibodies, and antibodies to pancreatic islet cells, are particularly common in patients with juvenile diabetes mellitus of recent onset. The islet cell antibodies may precede the onset of diabetes by several years. Whether these antibodies and other autoantibodies are concerned in the genesis of diabetes or are a cause of diabetogenic islet cell damage is presently uncertain. Antibodies to pancreatic islet cells are IgG and are found by indirect immunofluorescence in patients with insulin-dependent diabetes.57 In vitrectomy specimens from humans, there may be deposition of immunoglobulins and complement components.58 However, there does not seem to be any unique pattern or one that differs from nondiabetic proliferative retinopathies. Unusual deposition patterns for complement components have been noted in the diabetic cornea.59
Delayed hypersensitivity to crude pancreatic islet antigen in diabetic patients can be shown by the macrophage migration inhibition test.60 Transient impairment of glucose tolerance occurs when mice are immunized with mouse pancreatic extract. Immunized animals show delayed hypersensitivity skin test reactions and inhibition of macrophage migration when tested with pancreatic islet antigen.
The histocompatibility antigen HLA-B8 is found with greater frequency in patients with insulin-dependent diabetes, Graves disease, and Addison disease than in normal subjects, and a high frequency of HLA-Bwl5 has been found in insulin-dependent diabetic subjects.61,62
Because insulin is a foreign protein, it can evoke an antibody response, and repeated injections of insulin into humans can give rise to antibody formation.63 (Beef insulin is apparently more antigenic for humans than pork insulin.) The immunologic response to insulin may take the form of an allergic reaction or of increasing resistance to the effects of insulin. The allergic reaction is usually a generalized urticaria, presumably mediated by IgE. Chronic insulin resistance seems to be caused by insulin-binding IgG antibodies.63 Insulin resistance and allergy rarely occur in the same patient. Insulin antibody levels tend to be lower in pregnant patients than in others.64
The vasoproliferative effects of insulin have been studied by Shabo et al.,65,66 who suggest that vascular proliferation in diabetes may be caused by the immunogenic nature of insulin as well as to the intrinsic vascular abnormality of diabetics. Insulin injected intravitreally into sensitized monkeys causes proliferation of preretinal vessels and connective tissue. It also causes rubeosis iridis, vascular hemorrhages, vessel tortuosity, and the beading of vessels. In the development of this experimental neovascularization, inflammation may be a significant factor.
|Although Graves disease was described 150 years ago, it is only in the
past 15 years that we have begun to understand its pathogenesis. Abnormalities
in both the humoral and cellular immune systems have now been
recognized in both Graves disease and Hashimoto thyroiditis. In addition
to immunologic phenomena, hormonal and genetic factors contribute
to the exophthalmos of Graves disease. Autoimmune responses directed against
orbital tissue antigens are probably responsible for many of the
typical ocular features.67|
Many lines of evidence point to the importance of autoimmunity in the pathogenesis of Graves disease. Thyroid antibodies are present in virtually all patients and in 50% of their relatives.68,69 Other autoantibodies and certain autoimmune disorders may be found as well. Various immunoglobulins occur within the stroma of the thyroid gland, and lymphocytes and plasma cells infiltrate the thyroid gland and retro-orbital tissues.
Early studies of autoimmune thyroiditis suggested that thyroglobulin was a secluded antigen during fetal life; hence, no tolerance to it was developed. If the thyroid gland were damaged by trauma or infection, however, cytotoxic thyroid antibodies could be formed. This theory is no longer tenable, because although circulating thyroglobulin has been found in 60% of normal individuals, thyroglobulin antibodies are rarely found unless there is thyroid disease.70 Experimental thyroiditis can be produced by immunizing animals with thyroid tissue emulsified in Freund's adjuvant, and it can then be transferred passively with immune serum or lymphocytes.71 Transfer with lymphocytes suggests that cell-mediated hypersensitivity may play a role in the development of the experimental disease.
The most compelling evidence that hyperthyroidism has an immunologic basis comes from studies of long-acting thyroid stimulator (LATS). When serum from patients with diffuse thyroid hyperplasia is injected into mice or guinea pigs, a prolonged stimulation of the animals' thyroid glands occurs. LATS, the substance responsible for this phenomenon, can be isolated in the gammaglobulin fraction of human serum. It is precipitated and neutralized by anti-IgG and is probably either an immunoglobulin or a soluble complex of thyrotropin and its antibody. LATS is found in 50% to 80% of patients with Graves disease.72
Other immunoglobulin-containing substances have also been identified in the sera of patients with Graves disease. One such substance, known as LATS protector (LATS-p), can block the binding of LATS to a human thyroid protein fraction.73 Human thyroid-stimulating substance stimulates intracellular colloid droplet formation in human thyroid slices and is found in virtually all LATS-negative patients with Graves disease.74 Complement-fixing antibodies to thyroid microsomes are frequently found in hyperthyroidism and may increase the function of thyroid cells. As mentioned earlier, antibodies to thyroglobulin, predominantly of the IgG and IgA classes, may also be found.
Antibodies to eye muscle are found in 70% of patients with Graves disease,75 and there can be a cross-reactivity between eye muscle antibodies and thyroid antigen.76 IgE-positive material has been found in association with the majority of leukocytes and with muscle fibers found in orbital tissue.77
The role of cell-mediated immunity may also be important in thyroid disorders. Lymphocytes from patients with Graves disease undergo blast transformation and produce migration inhibition factor (MIF) in response to human thyroid antigen (except during remissions of the disease when MIF is not produced).78,79 MIF is also produced in response to liver mitochondrial antigens in patients with either Graves disease or Hashimoto thyroiditis, perhaps caused by cross-reactivity between the thyroid and the liver or by the presence of more than one population of sensitized T lymphocytes.
MIF production to retro-orbital muscle antigen can be demonstrated in patients with exophthalmos who have no evidence of either thyroid disease or cellular immunity to thyroid antigens.80 It would appear, therefore, that antigens concerned with the production of the exophthalmos are separate from those responsible for the thyroid abnormality. Exophthalmos may be a separate autoimmune process that overlaps the process concerned with the hyperthyroidism of Graves disease.81 The cause of the exophthalmos in Graves disease is unknown, but immune factors may participate in its pathogenesis. Exophthalmos-producing factor (EPF) is derived from thyrotropin and binds to membrane receptors of retro-orbital tissues. This binding affinity is enhanced by serum immunoglobulins from patients with severe degrees of Graves ophthalmopathy.82 Both thyroglobulin and immune complexes bind to human extraocular muscle membrane.83 Lymphatic connections between the thyroid gland and the orbits may exist so that exophthalmos in Graves disease could be the result of stimulation by thyroid glandular antigens or antibodies and their complexes. These elements, interacting with orbital muscle receptors, could then produce muscle injury and inflammation.83
Unlike normal lymphocytes, the lymphocytes from patients with Graves disease can be stimulated by phytohemagglutinin (PHA) to produce thyroid-stimulating immunoglobulins.81 Because PHA stimulates only T lymphocytes, and because T lymphocytes cannot produce immunoglobulins, PHA must turn-on helper T lymphocytes, which in turn stimulate B lymphocytes to produce these immunoglobulins. It has been suggested that in Graves disease there is an inherited defect of immunologic control over a specific forbidden clone of helper T lymphocytes. If this clone appeared by normal, random mutation, it might not be destroyed. It would then proceed to interact with thyroid antigen, stimulate the replication of the forbidden T lymphocytes, and cooperate with B lymphocytes in the production of thyroid-stimulating immunoglobulins. Thus both cellular and humoral immune mechanisms would have roles to play in this hypothetical model.
Certain genetic factors may also participate in the pathogenesis of Graves disease. In whites, more patients with the disease have the histocompatibility antigen HLA-B8 than do control subjects, and in Japanese, more persons with the disease have HLA-Bw40 than do controls.84 The concordance rate in monozygotic twins is high. The prevalence of Graves disease is much greater in women than in men, with the X chromosome or estrogens possibly playing a role in its development.
9. Tejada-Simon MV, Zang YC, Hong J, Rivera VM, Killian JM, Zhang JZ: Detection of viral DNA and immune responses to the human herpesvirus 6 101-kilodalton virion protein in patients with multiple sclerosis and in controls. J Virol 76(12):6147–6154, 2002
13. Krametter D, Niederwieser G, Berghold A, et al: Chlamydia pneumoniae in multiple sclerosis: humoral immune responses in serum and cerebrospinal fluid and correlation with disease activity marker. Mult Scler 7(1):13–18, 2001
16. McDonald WI CA, Edan G, Goodkin D, Hartung HP, Lublin FD, McFarland HF, Paty DW, Polman CH, Reingold SC, Sandberg-Wollheim M, Sibley W, Thompson A, van den Noort S, Weinshenker BY, Wolinsky JS: Recommended diagnostic criteria for multiple sclerosis: guidelines from the International Panel on the diagnosis of multiple sclerosis. Ann Neurol 50(1):121–127, 2001
18. Wuerfel J B-SJ, Brunecker P, Aktas O, McFarland H, Villringer A, Zipp F: Changes in cerebral perfusion precede plaque formation in multiple sclerosis: a longitudinal perfusion MRI study. Brain 1–9, 2003
23. Fagnart OC SC, Laterre C: Free kappa and lambda light chain levels in the cerebrospinal fluid of patients with multiple sclerosis and other neurological diseases. J Neuroimmunol 19(1–2):119–132, 1988
28. Comi GFM, Wolinsky JS: European/Canadian multicenter, double-blind, randomized, placebo-controlled study of the effects of glatiramer acetate on magnetic resonance imaging–measured disease activity and burden in patients with relapsing multiple sclerosis. European/Canadian Glatiramer Acetate Study Group. Ann Neurol 49:290–297, 2001
32. Bellomi FSC, Tomassini V, Gasperini C, Paolillo A, Pozzilli C, Antonelli G: Fate of neutralizing and binding antibodies to IFN beta in MS patients treated with IFN beta for 6 years. Neurol Sci 215(1–2):3–8, 2003
33. Sorensen PSRC, Clemmesen KM, Bendtzen K, Frederiksen JL, Jensen K, Kristensen O, Petersen T, Rasmussen S, Ravnborg M, Stenager E, Koch-Henriksen N: Danish Multiple Sclerosis Study Group. Clinical importance of neutralising antibodies against interferon beta in patients with relapsing-remitting multiple sclerosis. Lancet 362(9391):1184–1191, 2003.
35. Beck RW, Trobe JD, Moke PS, et al: High- and low-risk profiles for the development of multiple sclerosis within 10 years after optic neuritis: experience of the optic neuritis treatment trial. Arch Ophthalmol 121(7):944–949, 2003
37. Frederiksen JL, Sindic CJ: Intrathecal synthesis of virus-specific oligoclonal IgG, and of free kappa and free lambda oligoclonal in acute monosymptomatic optic neuritis. Comparison with brain MRI. Mult Scler 4(1):22–26, 1998
44. Beck RW, Cleary PA, Trobe JD, et al: The effect of corticosteroids for acute optic neuritis on the subsequent development of multiple sclerosis. The Optic Neuritis Study Group. N Engl J Med 329(24):1764–1769, 1993
54. Engel AG, Lambert EH, Howard FM: Immune complexes (IgG and C3) at the motor end-plate in myasthenia gravis: ultrastructural and light microscopic localization and electrophysiologic correlations. Mayo Clin Proc 52(5):267–280, 1977
68. Bastenie PA, Ermans AM: Thyroiditis and thyroid function: Clinical, morphological, and physiopathological studies. In International Series of Monographs in Pure and Applied Biology, Modern Trends in Physiological Sciences. Vol 36. Oxford: Pergamon Press, 1972
71. Nakamura RM, Weigle WO: Passive transfer of experimental autoimmune thyroiditis from donor rabbits injected with soluble thyroglobulin without adjuvant. Int Arch Allergy Appl Immunol 32(5):506–520, 1967
74. Onaya T, Kotani M, Yamada T, Ochi Y: New in vitro tests to detect the thyroid stimulator in sera from hyperthyroid patients by measuring colloid droplet formation and cyclic AMP in human thyroid slices. J Clin Enocrinol Metab 35(5):859–866, 1973
75. Kahaly G, Moncayo R, Stover C, Beyer J: Relationship of eye muscle antibodies with HLA phenotypes and thyroid-stimulating immunoglobulins in endocrine orbitopathy. Res Exp Med 191(2):137–144, 1991
80. Munro RE, Lamki L, Row VV, Volpe R: Cell-mediated immunity in the exophthalmos of Graves' disease as demonstrated by the migration inhibition factor (MIF) test. J Clin Enocrinol Metab 37(2):286–292, 1973