透過您的圖書館登入
IP:18.217.200.151
  • 學位論文

骨蝕作用、骨形成、以及免疫耐受性對於台灣人僵直性脊椎炎發生之效應

Effects of bone resorption, bone formation, and immune tolerance on the development of Taiwanese ankylosing spondylitis

指導教授 : 翁瑞宏

摘要


僵直性脊椎炎 (ankylosing spondylitis;AS) 是慢性發炎疾病,而發炎已經被報告可以促進蝕骨細胞 (osteoclast) 的分化與成熟,並且骨蝕作用 (bone resorption) 也會影響成骨細胞 (osteoblast) 的高度表現,導致骨形成 (bone formation) 與脊椎韌帶骨贅 (syndesmophyte) 的現象。此外,自體免疫耐受性 (autoimmune tolerance) 之失衡與自體免疫疾病之發生具有相關性,CD4+ T細胞以及CD8+ T細胞也被發現在僵直性脊椎炎患者中相較於健康對照有較高的表現,如此的表現可能與活化T細胞的抑制訊息失衡具有關連。 細胞核因子kB接受器活化因子配位基 (receptor activator for nuclear factor kappa B ligand;RANKL) 結合於細胞核因子kB接受器活化因子 (receptor activator of nuclear factor kappa B;RANK),可導致骨蝕作用的活化,而蝕骨細胞抑制因子 (osteoprotegerin;OPG) 也與RANK競爭來與RANKL結合以及抑制骨吸收作用。因此,我們設計一項病例對照研究來評估僵直性脊椎炎的發生及其臨床表徵與RANK、RANKL以及OPG基因多形性之相關。總計,有330名僵直性脊椎炎病患與330名以性別及年齡配對的健康對照被納入。結果顯示,OPG GG基因型攜帶者相較於GC與CC基因型者具有增加的僵直性脊椎炎發生危險 (odds ratio [OR] 1.74;95% confidence interval [CI] 1.26-2.40)。此外,初始症狀發病年齡與周邊關節炎之比例在不同的OPG G1181C基因型之僵直性脊椎炎病患中具有顯著差異,攜帶OPG C對偶基因之HLA-B27陽性病患具有最早的初始症狀發病年齡。因此,OPG G1181C基因多形性可能是相關於僵直性脊椎炎之發展與臨床表徵。 微型核醣核酸 (microRNA;miR)-146a結合在TNF接受器相關因子6 (TNF-receptor-associated-factor-6;TRAF-6) 和IL-1接受器相關激酶1 (interleukin-1 receptor-associated kinase 1;IRAK1) 來抑制細胞核因子κB (nuclear factor κB;NF-κB) 活性。我們探討是否miR-146a rs2910164 G/C、IRAK1 rs3027898 A/C和rs1059703 T/C基因多形性與僵直性脊椎炎的發展和臨床徵狀具有相關。總計,450名台灣僵直性脊椎炎病患與438名健康對照被納入於我們的研究。在miR-146a rs2910164/IRAK1 rs3027898對偶基因成對分析 (pairwise analysis) 中顯示,G/A (OR 2.84;95% CI 1.34-6.03)、G/C (OR 1.71;95% CI 1.27-2.30)、與C/A (OR 1.53;95% CI 1.09-2.16) 相較於C/C對偶基因具有顯著較高的僵直性脊椎炎危險;如此的結果在男性中被發現,但不在女性中。此外,攜帶miR-146a rs2910164/IRAK1 rs3027898 G/A成對對偶基因的僵直性脊椎炎病患相較於攜帶miR-146a rs2910164/IRAK1 rs3027898 C/C成對對偶基因的病患具有一個顯著較高的虹彩炎 (uveitis) 危險。因此,miR-146a rs2910164與IRAK1 rs3027898多形性可能是相關於僵直性脊椎炎的發生以及其臨床徵狀。 腫瘤壞死因子 (tumor necrosis factor;TNF)-α可能誘導發炎反應和骨蝕生成作用。此外,miR-125b表現抑制TNF-α表現,並且較高的miR-155表現藉由結合細胞激素訊息抑制者-1 (suppressor of cytokine signaling-1;SOCS-1) 以負向誘導TNF-α表現。我們比較TNF-α mRNA、miR-125b、miR-155、與SOCS-1 mRNA表現在僵直性脊椎炎病患與健康對照之間的差異,風濕藥物在僵直性脊椎炎病患的這些分子表現之影響也被評估。以反轉錄定量聚合酶鏈鎖反應 (RT quantitative PCR;RT-qPCR) 評估119名僵直性脊椎炎病患與120名健康對照的全血TNF-α mRNA、miR-125b、miR-155、與SOCS-1 mRNA表現。在無服用非類固醇抗發炎藥物 (nonsteroidal antiinflammatory drugs;NSAIDs) 或疾病調節抗風濕藥物 (disease-modifying antirheumatic drugs;DMARDs) 的僵直性脊椎炎病患中,TNF-α mRNA表現分別負向相關於miR-125b表現 (r = -0.63,p = 0.03);正向相關於巴斯僵直性脊椎炎功能性指數 (Bath Ankylosing Spondylitis Functional Index;r = 0.82,p < 0.01)、紅血球沉降速率 (erythrocyte sedimentation rate;r = 0.70,p = 0.02)、以及C-反應蛋白 (C-reactive protein;r = 0.79,p < 0.01)。然而,TNF-α mRNA表現並無相關於僵直性脊椎炎病患的骨質代謝。 MiR-21與計畫性細胞死亡-4 (programmed cell death 4;PDCD4) 的結合可能抑制PDCD4的表現,並且進而導致蝕骨細胞的活化。我們比較全血miR-21表現在僵直性脊椎炎病患與健康對照之間的差異,並且評估僵直性脊椎炎病患的全血miR-21、PDCD4 mRNA、與骨蝕作用之相關;風濕藥物在僵直性脊椎炎病患的miR-21與PDCD4 mRNA表現的影響也被評估。以RT-qPCR偵測122名僵直性脊椎炎病患與122名健康對照的全血miR-21與PDCD4 mRNA表現量,並且以酵素免疫分析法 (Enzyme-linked immunosorbent assay;ELISA) 來偵測CTX的血清濃度。相較於健康對照,僵直性脊椎炎病患具有顯著較高的miR-21、PDCD4 mRNA表現、與CTX濃度。在沒有服用非類固醇抗發炎藥物或疾病調節抗風濕藥物的病患中,miR-21表現是負相關於PDCD4 mRNA表現。此外,僵直性脊椎炎病患之miR-21與CTX的表現具有正相關存在 (r = 0.35,p < 0.001);特別是在具有疾病持續時間< 10年 (r = 0.49,p < 0.001) 與活性期疾病 (r = 0.51,p < 0.001) 的病患中。 骨形成蛋白 (Bone morphogenic proteins;BMPs) 是由成骨細胞 (osteoblasts) 所分泌,可影響僵直性脊椎炎的新骨形成 (bone formation)。miR-29抑制無翼的 (Wingless;Wnt) 蛋白抑制者Dickkopf 1 (DKK1) 與分泌型捲曲相關蛋白 (secreted frizzled related protein;sFRP) 2,miR-27也抑制Wnt抑制者sFRP1、DKK2、與adenomatous polyposis coli (APC) 以調節Wnt訊息路徑。當Wnt訊息路徑被活化時,可能進一步地誘發BMP-2、BMP-4、與BMP-7表現。我們比較BMP-2、BMP-4、與BMP-7、miR-29、miR-27、以及其標的基因表現在僵直性脊椎炎病患和健康對照之差異。風濕藥物在僵直性脊椎炎病患的這些分子表現之影響也被評估。結果觀察到,73名脊椎黏合的僵直性脊椎炎病患相較於62名健康對照和79名無脊椎黏合的病患具有較高的血清BMP-4與BMP-7濃度 (ps < 0.01)。脊椎黏合與無脊椎黏合的僵直性脊椎炎病患也分別相較於健康對照,具有顯著較高的miR-27b、miR-29a、miR-29b、以及較低的sFRP1與sFRP2 mRNA表現。相較於未接受藥物治療的僵直性脊椎炎病患,接受藥物治療的病患具有較低的miR-27b表現。因此,BMPs、miR-27b與Wnt抑制者蛋白可能相關於僵直性脊椎炎的新骨形成。 第四類細胞毒殺性T淋巴抗原 (cytotoxic T lymphocyte antigen-4;CTLA-4) 與蛋白酪氨酸磷酸酶N22 (protein tyrosine phosphatase, nonreceptor 22;PTPN22) 基因可能在自體免疫耐受性的初期扮演著抑制過度活化T細胞的角色,而計畫性細胞死亡-1 (programmed cell death-1;PD-1) 與配位基-1 (PD-L1) 和配位基-2 (PD-L2) 鍵結所引起的抑制訊息可能在免疫耐受性的晚期維持著平衡。因此,我們評估CTLA-4與PTPN22基因型以及PD-1、PD-L1和PD-L2基因型在僵直性脊椎炎發生之效應。結果顯示,同時攜帶PTPN22 CC/CTLA-4 AA或是PTPN22 GC/CTLA-4 AA基因型者相較於其他PTPN22與CTLA-4基因型合併組合者具有2.10倍 (95% CI 1.07-4.13) 的僵直性脊椎炎增加危險。此外,PD-1 GG + GA/PD-L1 CC/PD-L2 CC合併基因型攜帶者、PD-1 AA/PD-L1 CC/PD-L2 CC合併基因型攜帶者、與PD-1 GG + GA/PD-L1 AC + AA/PD-L2 CC合併基因型攜帶者分別相較於其它PD-1/PD-L1/PD-L2合併基因型攜帶者具有6.63倍 (95% CI 1.28-34.46)、4.33倍 (95% CI 0.36-52.64) 和3.05倍 (95% CI 2.16-4.32) 的僵直性脊椎炎增加危險。我們的結果建議著,PTPN22 -1123 G/C與CTLA-4 +49 A/G基因型和PD-1 G-536A、PD-L1 A8923C、以及PD-L2 C47103T基因型在僵直性脊椎炎的發生具有合併效應。 總體而言,骨蝕作用、骨形成、與免疫耐受性失衡相關基因和miRNA以及其標的基因可能相關於僵直性脊椎炎發生和臨床表徵。

並列摘要


Ankylosing spondylitis (AS) is a chronic inflammatiory disorder. Inflammation has been reported to promote the differentiation and maturation of the osteoclast, and bone resorption also affects the overexpression of the osteoblast resulting in the bone formation and syndesmophyte. In addition, imbalance of autoimmune tolerance is associated with the occurrence of autoimmune diseases. The higher expression of CD4+ and CD8+ T cells has also been found in AS patients than in healthy controls, and such expression may relate to the imbalance in the negative signal of activated T cells. Receptor activator of nuclear factor kappa B ligand (RANKL) binds to receptor activator of nuclear factor kappa B (RANK) could cause the activation of bone resorption. Osteoprotegerin (OPG) also competes with RANK by binding to RANKL and inhibiting bone absorptions. Therefore, we designed a case-control study to evaluate the association between occurance and clinical features of AS and RANK, RANKL and OPG genetic polymorphisms. A total of 330 AS patients and 330 age- and gender-matched healthy controls were recruited. Results revealed OPG GG genotype carriers had an elevated risk of AS compared with those with the GC and CC genotypes (odds ratio [OR] 1.74, 95% confidence interval [CI] 1.26-2.40). In addition, age of symptom onset and frequency of peripheral arthritis significantly differed among AS patients by different OPG G1181C genotypes. HLA-B27+ patients with the OPG C allele had the earliest age of symptom onset. Therefore, OPG G1181C polymorphism may be associated with AS development and clinical manifestations. MicroRNA (miR)-146a targets TNF-receptor-associated-factor-6 (TRAF-6) and interleukin-1 receptor-associated kinase 1 (IRAK1) to suppress nuclear factor kB (NF-kB) activity. We investigated whether miR-146a rs2910164 G/C, IRAK1 rs3027898 A/C and rs1059703 T/C genetic polymorphisms associated with AS development and clinical characteristics. A total of 450 Taiwanese AS patients and 438 healthy controls were recruited in our study. Pairwise analysis of the miR-146a rs2910164/IRAK1 rs3027898 alleles showed G/A (OR 2.84, 95% CI 1.34-6.03), G/C (OR 1.71, 95% CI 1.27-2.30), and C/A (OR 1.53, 95% CI 1.09-2.16) had a significantly greater risk of AS than the C/C alleles. Such results were found in males, but not in females. In addition, AS patients with miR-146a rs2910164/IRAK1 rs3027898 G/A pairwise alleles had a significantly higher risk of uveitis development than patients with miR-146a rs2910164/IRAK1 rs3027898 C/C pairwise alleles. Therefore, miR-146a rs2910164 and IRAK1 rs3027898 polymorphisms might be associated with the development of AS, as well as its clinical manifestations. Tumor necrosis factor (TNF)-α might induce the inflammation and osteoclastogenesis. In addition, miR-125b expression inhibits the TNF-α expression, and higher miR-155 expression inversely induces TNF-α expression by targeting suppressor of cytokine signaling-1 (SOCS-1). We compared the differences of TNF-α mRNA, miR-125b, miR-155, and SOCS-1 mRNA expressions between AS patients and healthy controls. The influence of rheumatologic drugs on the expressions of these molecules in AS patients was also assessed. Expressions of whole blood TNF-α mRNA, miR-125b, miR-155, and SOCS-1 mRNA were assessed by RT quantitative PCR (RT-qPCR) among 119 AS patients and 120 healthy controls. TNF-α mRNA expression was negatively correlated with miR-125b expression (r = -0.63, p = 0.03); positively correlated with Bath Ankylosing Spondylitis Functional Index (r = 0.82, p < 0.01), erythrocyte sedimentation rate (r = 0.70, p = 0.02), and C-reactive protein (r = 0.79, p < 0.01) in AS patients without taking nonsteroidal antiinflammatory drugs (NSAIDs) or disease-modifying antirheumatic drugs (DMARDs), respectively. But TNF-α mRNA expression was not correlated with bone metabolism in AS patients. The binding of miR-21 and programmed cell death 4 (PDCD4) might inhibit the expression of PDCD4, and further induce the activation of osteoclasts. We compared the difference of miR-21 expression between AS patients and healthy controls, and evaluated the relationships of miR-21, PDCD4 mRNA, and bone erosion in AS patients. The influence of rheumatologic drugs on the expressions of miR-21 and PDCD4 mRNA in AS patients was also assessed. The expressions of serum miR-21 and PDCD4 mRNA were assessed by RT-qPCR among 122 AS patients and 122 healthy controls, and serum CTX was detected using enzyme-linked immunosorbent assay (ELISA). Compared to healthy controls, AS patients had significantly higher levels of miR-21, PDCD4 mRNA, and CTX. In AS patients who were taking neither NSAIDs nor DMARDs, miR-21 expression was negatively correlated with PDCD4 mRNA expression. In addition, positive correlation of miR-21 and CTX level was observed in AS patients (r = 0.35, p < 0.001), especially in those with disease duration < 10 years (r = 0.49, p < 0.001) and active disease (r = 0.51, p < 0.001). Bone morphogenic proteins (BMPs), is secreted by osteoblasts, might influence the bone formation in AS. MiR-29 suppresses Dickkopf 1 (DKK1) and secreted frizzled related protein 2 (sFRP2) of the Wingless (Wnt) inhibitors, and miR-27 also suppresses sFRP1, DKK2, and adenomatous polyposis coli (APC) of the Wnt inhibitors to regulate the Wnt signal pathway. When Wnt signal pathway was activated, it might further induced BMPs expression. We compared the differences of miR-29, miR-27, and their target genes expressions between AS patients and healthy controls. The influence of rheumatologic drugs on the expressions of these molecules in AS patients was also assessed. Results observed that 73 patients with spinal fusion had higher serum levels of BMP-4 and BMP-7 than either 62 healthy controls or 79 patients without spinal fusion (ps < 0.01), respectively. AS patients with spinal fusion and those without spinal fusion also had the significantly increased miR-27b, miR-29a, miR-29b, and lower sFRP1 mRNA and sFRP2 mRNA expressions than healthy controls; respectively. Compared to AS patients without receiving drug treatment, those with receiving drug treatment had a lower miR-27b expression Therefore, BMPs, miR-27, and Wnt inhibitor proteins might relate to the bone formation in AS. Cytotoxic T lymphocyte antigen-4 (CTLA-4) and protein tyrosine phosphatase, nonreceptor 22 (PTPN22) might play the roles in the inhibition of activated T cells in the initial period of autoimmune tolerance. Programmed cell death-1(PD-1) binds programmed cell death-1 ligand 1(PD-L1) and ligand 2 (PD-L2) to induce negative signals that might maintain the balance of immune tolerance during late period. Therefore, we evaluated the effects of CTLA-4, PTPN22, PD-1, PD-L1 and PD-L2 genptypes in AS occurrences. Results revealed subjects with both of PTPN22 CC/CTLA-4 AA or PTPN22 GC/CTLA-4 AA genotypes had a 2.10 fold (95% CI 1.07-4.13) greater risk of AS development than those with other combinations of PTPN22 and CTLA-4 genotypes. In addition, subjects carrying combinations of PD-1 GG + GA/PD-L1 CC/PD-L2 CC genotypes, those carrying combinations of PD-1 AA/PD-L1 CC/PD-L2 CC genotypes, and those carrying combinations of PD-1 GG + GA/PD-L1 AC + AA/PD-L2 CC genotypes had 6.63 (95% CI 1.28-34.46), 4.33 (95% CI 0.36-52.64), and 3.05 (95% CI 2.16-4.32) greater risks of AS development than those with other combinations of PD-1/PD-L1/PD-L2 genotypes; respectively. Our results suggest that PTPN22 G-1123C, CTLA-4 A49G, PD-1 G-536A, PD-L1 A8923C, and PD-L2 C47103T genotypes have the combined effects on AS development. In conclusion, bone resorption, bone formation, and imbalance of immune tolerance-related gene and miRNA, and their target genes might associate with the development of AS and clinical syndromes.

參考文獻


30. Shih HC, Liu SC, Chang CP, Tschen JS, Chiu HY, Liu HC, Chang JG. Positive association of ankylosing spondylitis with homozygous HLA-B2704, but protection with B2705 in Taiwan Chinese. Kaohsiung J Med Sci. 17(10):509-16, 2001.
2. Brewerton DA, Hart FD, Nicholls A, Caffrey M, James DC, Sturrock RD. Ankylosing spondylitis and HL-A 27. Lancet. 1(7809):904-7, 1973.
3. Feltkamp TE, Mardjuadi A, Huang F, Chou CT. Spondyloarthropathies in eastern Asia. Curr Opin Rheumatol. 13(4):285-90, 2001.
4. Sieper J, Braun J, Rudwaleit M, Boonen A, Zink A. Ankylosing spondylitis: an overview. Ann Rheum Dis. 61 Suppl 3:iii8-18, 2002.
5. Khan MA. Update on spondyloarthropathies. Ann Intern Med. 136(12):896-907, 2002.

延伸閱讀