急性與慢性葛瑞夫氏眼病變之眼窩脂肪、眼外肌與眼輪匝肌之基因表現及組織染色分析 -類固醇替代治療之新方向

Abstract

目的：葛瑞夫氏症是甲狀腺功能亢進的一種自體免疫疾病，大約有百分之二十五至五十的病人出現眼部症狀，稱為葛瑞夫氏眼病變或是甲狀腺眼疾。臨床表現包括眼瞼攣縮、突眼、軟組織水腫、牽制性肌肉病變與壓迫性視神經病變。可能原因為固定容量之眼窩內，脂肪、結締組織和眼外肌體積增加所造成。有些病人主要以突眼，眼窩脂肪組織增生表現；有些以牽制性肌肉病變如斜視為主；有些則是兩者皆有。不同組織間的表現差異可能是因為纖維母細胞不同。近期的研究顯示，眼窩脂肪組織增生、葡萄醣胺類 (其主要成份為玻尿酸) 堆積和肌肉纖維化，乃是對發炎反應的一個調控所伴隨產生的現象。之前一些葛瑞夫氏眼病變與別的疾病、組織的體內與體外實驗，發現IL-1β增加COX-2的基因表現，COX-2催化之直接產物為前列腺素H2，之後可代謝為前列腺素E2，引起發炎反應。另一方面，前列腺素H2也可代謝為15d-PGJ2，就可活化PPAR-γ，引起抗發炎反應且使脂肪組織增生。COX-2經前列腺素E2作用，產生TGF-β造成病理性纖維化，也可活化HAS，開始葡萄醣胺類的堆積。TGF-β本身就可增進玻尿酸的合成。玻尿酸與TGF-β也可刺激COX-2與前列腺素的製造。而PPAR-γ活化後可以抑制TGF-β產生纖維化的反應。這些反應和葛瑞夫氏眼病變的致病機轉可能有關。有鑑於在急性發炎時，一部份病人對大量類固醇治療並無顯著改善，且常期使用類固醇常帶來副作用。我們利用葛瑞夫氏眼病變的眼窩脂肪組織、眼外肌與眼輪匝肌，進行免疫組織化學染色，並探討其抗原、細胞激素和脂肪組織增生、合成玻尿酸酵素的基因表現是否增加，為前述之關係提出體內實驗之佐證。希望對致病機轉有更進一步瞭解後，未來能在類固醇之外，為葛瑞夫氏眼病變的病人找到其他能中止發炎反應的替代治療，如 PPAR-γ 拮抗劑或COX-2 抑制劑。 方法：臨床方面- 由病歷收集病人之資料。檢體方面- 病例組為收集葛瑞夫氏眼病變因急性期產生壓迫性視神經病變之病人，與穩定期突眼病人，接受眼窩減壓術時取得之眼窩脂肪組織與眼輪匝肌。另從穩定期牽制性肌肉病變病人接受矯正手術時取得眼外肌。對照組則由非葛瑞夫氏眼病變的病人接受眼窩手術時取得眼窩脂肪組織與眼輪匝肌 ，及病人接受斜視手術時取得眼外肌。標本取下後，作蘇木紫及伊紅之常規染色，接著做免疫組織化學染色，包括HLA-DR、CD3、CD20及CD68。另抽取訊息核醣核酸，比較 IGF-1α、IGF-1R、TSHr、IL-1β、IFN-γ、IL-6、TGF-β、COX-2、PPAR-γ、adiponectin、HAS、CXCL10、sFRP1 及CYR61基因表現量。 結果：病例組為收集十位急性期與十位穩定期葛瑞夫氏眼病變病人，接受眼窩減壓術時取得之眼窩脂肪組織與眼輪匝肌。並另多收集五位穩定期葛瑞夫氏眼病變病人之眼窩脂肪組織，分析PPAR-γ基因表現量。從五位穩定期牽制性肌肉病變病人接受矯正手術時取得眼外肌。對照組則是由十位非葛瑞夫氏眼病變的病人接受眼窩手術時取得眼窩脂肪組織與眼輪匝肌，及五位病人接受斜視手術時取得眼外肌。免疫組織化學染色方面，相較於對照組，在葛瑞夫氏眼病變的組織中，陽性染色的細胞皆較多。大部份HLA-DR陽性染色之細胞為CD20陽性染色之B淋巴球，CD3陽性染色之T淋巴球次之。只有少數是CD68陽性染色的巨噬細胞。基因表現量方面，在眼窩脂肪組織和眼輪匝肌中，大部份基因表現量皆呈現穩定期病例組大於急性期病例組大於對照組。在眼窩脂肪組織，和其他研究結果類似，除了adiponectin、IL-6、TGF-β、CXCL10、sFRP1及CYR61外，許多基因在葛瑞夫氏眼病變組表現量較高。十五位穩定期葛瑞夫氏眼病變病人的眼窩脂肪組織之PPAR-γ基因表現量比對照組高。眼輪匝肌則是 IL-6 與 PPAR-γ基因表現量有顯著差異。病例組之眼外肌，除了HAS1與CYR61無統計上顯著差異外，自體免疫抗原、細胞激素、COX-2、PPAR-γ、adiponectin與玻尿酸合成之基因表現量皆為病例組較對照組多。 結語：在眼窩脂肪組織與眼輪匝肌中，大部份基因皆呈現穩定期病例組表現量大於急性期病例組，可能是因急性期病人剛接受大劑量類固醇治療產生的現象。眼外肌幾乎所有的基因表現量在疾病組與對照組間皆有顯著差異，但眼窩脂肪組織與眼輪匝肌有差異的基因則截然不同。可能原因是不同組織間纖維母細胞不同。COX-2與HAS在病變組的眼窩脂肪組織與眼外肌表現量都增加。眼窩脂肪組織、眼輪匝肌與眼外肌之PPAR-γ表現量，在病變組都比對照組增加。TGF-β1在病變組之眼外肌表現量增加。這些基因表現量在病人檢體確實比對照組多，從旁印證了由之前與一些體外實驗產生的假設，即COX-2催化產生之15d-PGJ2可活化PPAR-γ，引起抗發炎反應，COX-2又可經前列腺素E2作用，產生TGF-β造成病理性纖維化，及TGF-β活化HAS，開始葡萄醣胺類的堆積的假說。而因此在類固醇之外的選擇，急性期眼窩脂肪組織增生與牽制性肌肉病變嚴重者，COX-2 抑制劑與PPAR-γ拮抗劑是值得考慮的新方向。至於安全性與效果如何，在臨床上是否可行，則需要更多進一步的實驗才能確定。

Purpose: Graves’ disease (GD) is an autoimmune disease with hyperthyroidism, and about 25% to 50 % patients have Graves’ ophthalmopathy (GO). The manifestations of GO, including lid retraction, proptosis, soft tissue swelling, restrictive myopathy, and compressive optic neuropathy, can be explained by an increase in the volume of the adipose, connective tissue, and extraocular muscles within the orbit. Some patients predominantly present with proptosis, which comes from profound adipogenesis. Some patients present with diplopia and strabismus as a consequence of restrictive myopathy. And others present with both types of symptoms. The different presentation of different tissues may be due to the heterogeneity of fibroblasts. Recent studies implicate adipogenesis, accumulation of glycosaminoglycans (mainly hyaluronan), and fibrosis of muscle tissue are effects accompanied with inflammation. According to previous in vitro and in vivo studies of GO and other disease, IL-1β increases the expression of COX-2. COX-2 converts arachidonic acid directly into PGH2, then PGE2, hence inducing inflammation. On the other hand, PGH2 can be further metabolized to yield 15d-PGJ2, which binds PPAR-γ and then induces anti-inflammatory activity, together with adipogenesis. Through action of PGE2, COX-2 induces TGF-β and produces fibrosis, and activates HAS, starting accumulation of glycosaminoglycan. TGF-β can increase the formation of hyaluronan. In turn, hyaluronan and TGF-β can stimulate the production of COX-2 and prostaglandin. PPAR-γ disrupts TGF-β/Smad signal transduction and blocks profibrotic responses. Some of these responses may take part in the pathogenesis of GO. Many studies have focused on finding the alternative therapy, because sometimes the symptoms of patients in acute stage of GO are not improved by high dosage of steroids and that patients often have to suffer many side effects from long-tern usage of steroids. In this study, we try to link the results of immunohistochemical stain and gene expression of autoantigen, cytokines, genes relating with adipogenesis, and hyaluronan synthesis by studying orbital adipose tissue, orbicularis oculi muscle, and extraocular muscle from acute and chronic stage of GO, and to find out if there is increased expression of these genes. This adds more information on whether the hypothesis proposed from in vitro studies is possible or not in the pathogenesis of GO. We hope that after we have more understanding of this pathogenesis, in the future we can find an alternative therapy to steroids, such as PPAR-γ antagonist or COX-2 inhibitor, for the management of GO. Methods: From chart review, we collected the clinical data. We collected orbital adipose tissue and orbicularis oculi muscle specimens from patients who received orbital decompression surgery due to compressive optic neuropathy in acute stage, and proptosis in chronic stage. Extraocular muscles were collected from patients who received muscle surgeries due to restrictive myopathy. Orbital adipose tissue, orbicularis oculi muscle and extraocular muscle specimens were collected from patients of non-GO receiving orbitotomy and surgeries for strabismus as control group. Specimens were processed for routine H&E stain and immunohistochemical stain with HLA-DR, CD3, CD20, and CD68. Messenger RNA was extracted for comparison of the expression of IGF-1α, IGF-1R, TSHr, IL-1β, IFN-γ, IL-6, TGF-β, COX-2, PPAR-γ, adiponectin, HAS, CXCL10, sFRP1, and CYR61. Results: We collected orbital adipose tissue and orbicularis oculi muscle specimens from 10 patients in acute stage and 10 patients in chronic stage of GO receiving orbital decompression surgery. Extraocular muscles were collected from 5 patients who received muscle surgeries due to restrictive myopathy. Orbital adipose tissue specimens were collected from extra five GO patients in chronic stage for analysis of PPAR-γ. Orbital adipose tissues and orbicularis oculi muscles from 10 patients with non- GO and extraocular muscles from 5 patients with strabismus were collected as control group. The immunoreactive cells were more prominent in GO than in control group. Most of the HLA-DR (+) cells were CD20 (+) B lymphocytes, and then came CD3 (+) T lymphocytes. Only a few cells were CD68 (+) macrophages. As for gene expression, the levels were highest in chronic stage group and acute stage group, higher than control group in most genes at orbital adipose tissues and orbicularis oculi muscle tissues. Like other studies, we found the expression of many genes in orbital adipose tissues of GO was higher than control group (p<0.05), except adiponectin, IL-6, TGF-β, CXCL10, sFRP1, and CYR61. The expression of PPAR-γ was higher in 15 orbital adipose tissues of chronic GO than control group. In orbicularis oculi muscles, the expression of IL-6 and PPAR-γ was higher in GO group. In extraocular muscles, almost all genes, including genes of autoantigens, cytokines, COX-2, PPAR-γ, adiponectin, and other genes of hyaluronan synthesis, were up-regulated in GO group with a statistical significant difference (p<0.05). Conclusion: In orbital adipose tissue and orbicularis oculi muscle specimens, most genes expressed more in chronic stage than acute stage. It possibly was under the anti-inflammation effect of recent and large dose of steroids used in patients in acute stage disease. Comparing GO groups with control group, almost all genes expressed higher in extraocular muscles, but the genes were totally different in orbital adipose tissues and orbicularis oculi muscles. This may be due to the heterogeneity of fibroblasts among different tissues. In diseased group, the expression of COX-2 and HAS in orbital adipose tissue and extraocular muscle was higher than control group. PPAR-γ in orbital adipose tissue, orbicularis oculi muscle, and extraocular muscle was up-regulated in GO. Also there was more expression of TGF-β1 in extraocular muscle of GO. This in vivo evidence gives us more information about the hypothesis proposed from in vitro studies. In brief, COX-2 produces 15d-PGJ2, which activates PPAR-γ and then induces anti-inflammatory activity, together with adipogenesis. Also COX-2 induces TGF-β and produces fibrosis, and activates HAS, starting accumulation of glycosaminoglycan. So besides steroid therapy, COX-2 inhibitor and PPAR-γ inhibitor are potential ways to be considered in patients of acute stage disease with severe orbital adipogenesis or restrictive myopathy. More studies will be helpful for us to get useful information on the safety and efficacy of COX-2 inhibitor or PPAR-γ antagonist in the treatment of Graves’ ophthalmopathy.