氧化壓力及花生四烯酸引起細胞質及粒腺體內離子失衡和細胞死亡關係之探討

Abstract

缺血再灌流(ischemic-reperfusion)產生過量的活性氧物質(reactive oxygen species)是導致細胞死亡的主要因子之一。在動物模式中，電子顯微鏡可觀察到在同一個心肌細胞(cardiomyocytes)會同時出現細胞凋亡(apoptic)及細胞死亡(necrotic)的特徵。我們利用雷射共軛焦顯微鏡(confocal)記錄動態心肌細胞，發現給予過氧化氫(H2O2)產生的氫氧自由基(OH•; hydroxyl radical)會顯著的增加細胞質(cytosol)及粒線體(mitochondria)的鈉離子(Na+)及鈣離子(Ca2+)濃度，氫氧自由基造成鈣離子過度負荷(overload)是由於細胞內的鈉離子過度負荷而活化鈉鈣交換器(Na-Ca exchanger)的反向運轉所造成的。心肌細胞在H2O2處理40分鐘後換回不含H2O2的培養液中，等待4.5及16小時, 心肌細胞凋亡(apoptosis)的比例會從0小時的4%增加到4.5小時的55%及16小時的85%，若以無鈉溶液可完全阻斷氫氧自由基所引發細胞凋亡，若以無鈣溶液則無法阻斷。另一方面利用Na-ionophores(在無鈣溶液中)單獨提高細胞內鈉離子濃度超過30 mM而不改變鈣離子濃度即可引發caspase 3-dependent細胞凋亡，顯示細胞內鈉離子過度負荷即可引發細胞凋亡。我們也發現當粒線體而非細胞質內鈉離子濃度的增加，會造成粒線體雙層膜上之滲透轉換孔道(permeation transition pore)的開啟，接著引發cytochrome c的釋放。由我們的發現可推論H2O2所引發粒線體的鈉離子濃度過度負荷是一個導致細胞凋亡機制的重要上游訊息。 本實驗進一步去探討活性氧物質所引起的細胞腫脹性死亡(oncosis)的機制；由以上的實驗已經証實細胞凋亡是因為H2O2引起粒腺體內鈉離子過度負荷造成cytochrome c的釋放及caspase-3的活化；而細胞死亡主要是由於poly(ADP-ribose) polymerases (PARPs)的活化，造成細胞內ATP/ NAD+的排空所引起。本實驗也發現H2O2在心肌細胞所引發粒腺體內鈉離子(及鈣離子)過度負荷主要是因為開啟(melastatin-related transient receptor potential 2, TRPM2) channels其原因如下：(1) 利用免疫化學染色(immuncytochemical)顯示心肌細胞細胞膜上可被TRPM2的抗體所標示。(2) 利用細胞膜嵌制技術 (patch-clamp technique)發現在加入H2O2後可偵測到鈉離子及鈣離子電流；另外在pipette solution中加入可開啟TRPM2的化學物質(β-NAD+或ADP-ribose, ADPR)，與H2O2所引發鈉離子及鈣離子電流，在電生理的特徵是很相似的。(3) 內生性的NAD+及ADPR的量在加入H2O2後會大量增加，若降低NAD+及ADPR的量則會抑制H2O2所引起細胞內鈉離子及鈣離子的過度負荷。(4) clotrimazole是TRPM2 channel是阻斷劑，可抑制ADPR開啟TRPM2所引起的電流，也會抑制H2O2所引發鈉離子過度負荷。很重要的，同時處理clotrimazole及DHQ (PARP的阻斷劑)，可幾乎完全抑制H2O2所引發細胞凋亡及細胞死亡的路徑。 已知花生四烯酸(Arachidonic acid, AA)在生理及病理上扮演重要角色。在論文第二部份利用初級培養的大鼠神經膠質細胞(astrocyte)來探討花生四烯酸作用，發現：(1) endothelin-1或thapsigargin (Tg)可引發填充性鈣離子流(CCE)，此填充性鈣離子流會被2-aminothoxydiphenyl borane (2-APB)或La3+所阻斷。(2) 花生四烯酸(10

Overproduction of reactive oxygen species (ROS) is one of the major causes of cell death in ischemic reperfusion injury, and, in animal models, electron microscopy has shown mixed apoptotic and necrotic characteristics in the same myocyte (i.e. oncosis). Using time-lapse confocal recording of live cardiomyocytes, I have shown that H2O2 (OH•) causes a marked increase in Na+ and Ca2+ levels in both the cytosol ([Na]cyt, [Ca]cyt) and mitochondria ([Na]m, [Ca]m). The H2O2-induced intracellular Na+ ([Na]i) overload contributes to the H2O2-induced [Ca]cyt/[Ca]m overload via activation of the reverse mode of the Na-Ca exchanger. When myocytes are treated for 40 min with 100 μM H2O2 in normal medium, then returns to H2O2-free medium, the percentage of apoptotic cells increases from 4% at 0 h to 55% and 85% at 4.5 and 16 h, respectively. H2O2-induced apoptosis is completely prevented using Na-free, but not Ca-free, medium. When a Na+ ionophore cocktail in Ca-free medium is used instead of H2O2 to increase the [Na]i by more than 30 mM without any change in the [Ca]i, cytochrome c release and caspase 3-dependent apoptosis occurres, showing that [Na]i overload per se induces apoptosis. I also show that the increase in the mitochondrial, but not the cytosolic, Na+ levels results in the opening of the permeation transition pore, followed by cytochrome c release. These findings therefore suggest that H2O2-induced [Na]m overload is an important upstream signal for the apoptotic machinery. My study further identifies a novel pathway for ROS-induced oncosis in which the apoptotic features are caused by H2O2-induced mitochondrial Na+ ([Na]m) overload, resulting in cytochrome c release and caspase 3 activation, while the necrotic features are caused by PARP activation, resulting in ATP/NAD+ depletion. I also show that opening of novel TRPM2 channels in cardiomyocytes is involved in the H2O2-induced [Na+]m (and [Ca2+]m) overload, since: (i) immunocytochemical studies show that the plasmalemma is labeled by anti-TRPM2 antibody, (ii) the patch-clamp technique shows that Na+ and Ca2+ currents are activated by addition of H2O2 or by inclusion in the pipette solution of either of two specific messengers (β-NAD+ or ADPR), which open the TRPM2 channel, and that the electrophysiological properties induced are very similar, (iii) endogenous NAD+ and ADPR levels are increased by H2O2, while a decrease in their levels inhibits H2O2-induced [Na]i/[Ca]i overload, and (iv) clotrimazole, a putative TRPM2 channel inhibitor, which inhibits the ADPR/TRPM2-induced current, also inhibits the H2O2-induced Na+ overload. Importantly, co-treatment with clotrimazole and DHQ (a PARP inhibitor) nearly completely abolishes the H2O2-induced apoptotic and necrotic processes. These results therefore suggest that activation of both TRPM2 and PARP is involved in the myocyte oncosis. This could be a novel therapeutic target in I/R-induced myocyte loss. It has been shown that arachidonic acid (AA) plays important physiological or pathophysiological roles. In the second part of my study, I have found in cultured rat astrocytes that: (i) endothelin-1 or thapsigargin (Tg) induces store-depleted activated Ca2+ entry (CCE), which is inhibited by 2-aminoethoxydiphenyl borane (2-APB) or La3+; (ii) AA (10 μM) and other unsaturated fatty acids (8,11,14-eicosatrienoic acid and y-linoleic acid) have an initial inhibitory effect on the CCE, due to AA- or fatty acid-induced internal acid load; (iii) after full activation of CCE, AA induces a further Ca2+ influx, which is not inhibited by 2-APB or La3+, indicating that AA activates a second Ca2+ entry pathway, which coexists with CCE; and (iv) Tg or AA activates two independent and co-existing non-selective cation channels and the Tg-induced currents are initially inhibited by addition of AA or weak acids. A possible pathophysiological effect of the AA-induced [Ca]i overload is to cause delayed cell death in astrocytes.