可溶性晚期糖基化終末產物接受器在急性腦中風的功能性角色

Abstract

背景及目的 腦中風是造成全球死亡的第三大原因及永久傷殘的首要原因。一旦發生急性缺血性腦中風，經由靜脈注射血栓溶解劑是唯一被證實有效的治療。然而考慮治療的時間限制以及可能的出血併發症，目前臨床上只有很小比例的急性腦中風病患能夠接受此項治療。因此，研究急性缺血性腦中風的病生理機制，並發展可能的治療方式是相當重要及迫切的課題。當腦血管阻塞引起急性缺血性腦中風時，處於缺血中心的腦組織會立即梗塞壞死，周圍受到較輕微缺血影響的腦細胞則進入缺血連鎖反應，並可能因而凋零死亡。在複雜的缺血連鎖反應中，腦神經系統的發炎反應包括上游的接受器及下游的訊息產物扮演相當關鍵的角色。 晚期糖基化終末產物接受器（receptor for advanced glycation end-products, RAGE）是一種橫跨細胞膜，在細胞表面表現的細胞激素接受器。RAGE及RAGE配合子[如AGE、high mobility group box 1(HMGB1)及S100等]調控的發炎反應被認為與心腦血管的慢性動脈硬化及急性缺血性傷害的病程密切相關。但除了細胞上的RAGE，還有循環的可溶性RAGE異構體(soluble RAGE,sRAGE)。sRAGE可以藉由RAGE的mRNA產生選擇性剪貼（alternative splicing）或者在完整的RAGE蛋白上，藉由蛋白溶解脢(protease)的作用，將細胞外RAGE切除而游離至細胞外。 循環的sRAGE具有與細胞上的RAGE競爭配合子的特性，藉由與RAGE配合子產生連結，抑制細胞內RAGE下游的相關訊息傳遞活化。相關的研究顯示，血液中的sRAGE可以作為臨床上各種慢性發炎，免疫疾病及心血管動脈硬化疾病的生物指標，且重組的sRAGE在上述疾病也可能具有治療的潛力。但sRAGE是否在急性腦中風具有類似的重要功能仍屬未知。本研究從臨床與基礎的角度，同時探討sRAGE在急性缺血性腦中風的重要生理意義與可能的治療潛力。 研究對象與方法 （1）臨床研究： 本研究為多中心臨床研究設計。急性腦中風病患的收案條件為缺血性腦中風發病後24小時內入院並在48小時內接受第一次抽血。急性腦中風病患共接受三次抽血，抽血的時間點分別為腦中風發生後小於48小時，2到3天（48-72小時），以及5到7天，對照組則接受一次抽血。血漿中的sRAGE及HMGB1濃度以酵素免疫分析法，（Enzyme-linked immunosorbent assay，ELISA）的方式測量，並與臨床的腦中風嚴重程度及三個月的功能性預後做相關分析。 （2）基礎研究： 小鼠接受中大腦動脈缺血-再灌流之損傷模式後，同時收集缺血側的腦組織及周邊血液來分析(s)RAGE及HMGB1表現。並在缺血再灌流傷害後3小時靜脈注射重組的sRAGE、HMGB1或生理食鹽水，之後評估24，48及72小時的行為功能，腦梗塞範圍及死亡率。我們也使用初級皮質神經細胞培養，研究缺氧缺糖模式傷害下細胞的RAGE表現量，並外加重組的sRAGE或HMGB1以觀察神經細胞的存活率。同時評估重組sRAGE對於腦組織及神經細胞在缺血傷害下的NFκB活性，磷酸化JNK及斷裂性caspase3蛋白的影響。 結果 本研究從2008年七月到2010七月，總共有72位急性缺血性腦中風個案及同樣個案數的對照組符合收案條件。腦中風病患於急性腦中風發生後，小於48小時的血中sRAGE及HMGB1的濃度皆顯著提高(p<0.01)，sRAGE之後呈現逐漸下降的趨勢，並且5-7天的血中濃度與第一個抽血時間點（小於48小時）相比顯著較低(p<0.05)。此外，腦中風後小於48小時的sRAGE濃度不但與NIHSS的數值呈顯著正相關(γ=0.36, p=0.002)，並且濃度較高與三個月的功能性預後不好顯著相關（p=0.037, OR=18.22）。 動物實驗顯示，小鼠血漿中的sRAGE及HMGB1濃度在發生缺血性腦傷害後，在sRAGE的部分可以見到急性缺血性傷害發生後明顯增加，並在較延遲的時間點轉而顯著降低的變化。而HMGB1則持續維持在相較對照組為顯著較高的濃度。免疫沈澱法的結果進一步顯示，不論是小鼠或者臨床病患，血中HMGB1與sRAGE結合的量在腦中風發生後，明顯隨著時間而增加。此外，外加重組的sRAGE可以減少缺血傷害後的腦組織周圍的免疫細胞浸潤，改善小鼠中風後的預後，減少缺氧缺糖培養環境的神經細胞死亡，以及逆轉外加重組HMGB1對小鼠及神經細胞在腦中風模式下的不良反應。 討論 我們的臨床研究部分得到了幾個重要的結果。第一個是急性缺血性腦中風的發生會引發血中的sRAGE以及HMGB1濃度明顯增高。第二個是缺血性腦中風發生後小於48小時的血中sRAGE的數值與腦中風的嚴重程度成正相關，且可以作為三個月後的功能性預後指標（數值越高，預後越差）。第三個是缺血性腦中風發生後，血中sRAGE的數值一開始顯著增高，但在往後的時間點卻逐漸下降。這樣的臨床研究結果，尤其是血中sRAGE的數值在腦中風發生後的快速動態性變化，讓我們進一步推論sRAGE不僅僅是能夠做為急性缺血性腦中風後有意義的生物指標（反應腦中風嚴重程度與預後），可能也參與了急性缺血性腦中風的重要病生理機制。基礎實驗的結果除了驗證了臨床研究所觀察到的，血中的sRAGE以及HMGB1在發生急性缺血性腦中風後的變化趨勢，也提供了HMGB1在急性腦中風後與sRAGE結合增加的直接證據。除此之外，外加重組的sRAGE能夠改善小鼠缺血性腦傷害模式的梗塞範圍及預後，可能的機制包括直接的神經保護作用，以及減少受傷腦組織的發炎細胞浸潤。 結論 我們的研究顯示循環性sRAGE在臨床上可以作為急性缺血性腦中風病患嚴重程度及預後的生物指標，並且外加重組sRAGE可能成為急性缺血性腦中風的治療方式。我們的研究結果支持後續進行更多的相關基礎與臨床實驗來研究(s)RAGE及RAGE配合子在急性腦中風的病生理機制中的重要角色，及驗證靜脈注射sRAGE在急性缺血性腦中風的治療果效。

Background and purpose: Stroke is the 3rd major cause of death worldwide. Nowadays, only intravenous administration of recombinant tissue plasminogen activator has been shown effective in management of acute ischemic stroke (IS). However, its clinical use remains limited mainly due to the short therapeutic window and an increased incidence of hemorrhagic stroke. There is an urgent need for novel therapies in acute IS patients. Theoretically, acute occlusion of the cerebral arteries results in immediate loss of oxygen and glucose to the core region of the affected brain tissue. Hence, a complex of ischemic cascades involving a series of biochemical reactions is rapidly activated, and delayed inflammatory mechanisms are among the critical mediators causing further cell death and functional deficits. The receptor for advanced glycation end products (RAGE) is located on the plasma membrane of many types of cells and ligands [such as AGE, high mobility group box 1 (HMGB1), and S100 protein] binding to RAGE leads to the activation of several intracellular inflammatory pathways, such as nuclear factor-κB. There are soluble isoforms of RAGE (sRAGE), which form either by alternative splicing of RAGE mRNA or by proteolytic cleavage of full-length RAGE protein. Biological studies have shown that circulating sRAGE can function as a decoy to compete with the mRAGE for ligand binding and consequently blocks RAGE-associated intracellular signaling. There is growing evidence implicating the participation of RAGE-ligand interactions in the development and progression of various immune-mediated disorders, including vascular diseases. However, little is known about the involvement of the soluble form of RAGE (sRAGE) in acute IS. Here, we aim to identify the role of plasma sRAGE and HMGB1 in IS patients, as well as mice subjected to focal ischemic injury. Subjects and Methods: In clinical part, seventy-two acute IS patients and the same number of controls were recruited and plasma samples were collected at <48 hours (H), 2-3 days (D) and 5-7 D after stroke. The levels of sRAGE and HMGB1 were measured via enzyme-linked immuno sorbent assay. The data were correlated with the clinical parameters including 3 months functional outcome after stroke. In basic part, expression levels of (s)RAGE and HMGB1 in plasma and ischemic brain tissues were investigated in mice model of focal ischemic injury. Behavior score, infarct volume and mortality in mice receiving recombinant sRAGE、HMGB1 or saline 3H after ischemic reperfusion injury (I/R) were evaluated at 24, 48 and 72H post stroke. Primary cortical neurons subjected to oxygen and glucose deprivation (OGD) were used to evaluate the cellular RAGE expression and the changes of cell viability with recombinant sRAGE and HMGB1. The impacts of recombinant sRAGE on NFκB activity, and expressions of phosphor JNK and cleaved caspase 3 in brain tissue and cultured neuron receiving ischemic injury were also evaluated. Results: In clinical part, levels of sRAGE and HMGB1 were both significantly increased at <48 H after IS, but sRAGE then decreased significantly at 5-7 D (p<0.05). sRAGE <48 H correlated well with baseline NIH stroke scale(γ=0.36, p=0.002) and was an independent predictor of poor functional outcome（p=0.037, OR=18.22）. In basic part, mice I/R injury markedly increased levels of RAGE and HMGB1 in the affected brain tissues as early as 3H and even higher at 24 and 72H after I/R compared with the controls. Importantly, the dynamic expression patterns of plasma sRAGE and HMGB1 in I/R injury were similar to our clinical data from IS patients. Immuno-precipitation assays revealed that the binding of plasma HMGB1 to sRAGE increased progressively after IS both in patients and mice. Administration of recombinant sRAGE significantly reduced infiltrating immune cells, improved the outcome of injury in mice, protected cultured neurons against OGD-induced cell death and ameliorated the detrimental effect of recombinant HMGB1. Discussion: Our study shows that the plasma levels of sRAGE and HMGB1 both were increased within 48 h after stroke onset, and the level of sRAGE was positively correlated with the stroke severity and as an independent outcome predictor. Considering the data from 3 time points of acute IS, there was a dynamic pattern of plasma sRAGE, from initial high to low within a short period of time after stroke. These characteristics indicate that plasma sRAGE is not only a surrogate maker for acute IS, but may also play some functional roles in the pathogenesis of acute IS. Our basic study not only confirmed the dynamic expression patterns of plasma sRAGE and HMGB1 in I/S patients, but also provided the direct evince of increasing sRAGE and HMGB1 complexes after I/S. In addition, possible mechanisms from recombinant sRAGE in protecting mice I/R regarding to the infarct volume and outcome, include direct neuron protection and reducing infiltration of inflammatory cells in brain tissue status post ischemic injury. Conclusion: Early post-stroke plasma sRAGE may play a protective role in IS by capturing HMGB1. Hence, recombinant sRAGE is a potential therapeutic agent in acute IS. Our results supported further studies from both basic and clinical aspects to investigate the role of (s)RAGE and RAGE ligands in the pathophysiological mechanism of I/S and the therapeutic effect of intravenous injection of recombinant sRAGE in acute I/S.