建立小鼠急性腎損傷後產生慢性腎病模式以研究新穎治療

Abstract

腎臟纖維化一直是腎病患者最難以痊癒的主因，也往往會導致器官衰竭與死亡。因此近年來有越來越多的研究投入在相關的領域當中，根據之前的研究發現，兩週後切除對側腎臟會對原本受到缺血再灌流的受損腎臟有反轉纖維化的效果。另一篇研究中主張在單側缺血再灌流之後一週摘除對側腎臟，觀察到腎功能變差，且在八週後可看到腎臟纖維化的情形，但由於缺血再灌流與摘除對側腎臟等兩個手術時間點只相隔一週，我們發現會造成小鼠高死亡率，因此在我們進一步的研究中選擇在左側腎臟進行缺血再灌流傷害四週後再摘除對側腎臟，藉此來探討腎臟摘除對缺血再灌流傷害的腎臟產生的影響。 我們使用11週大的C57BL/6公鼠來進行實驗，疾病組小鼠在利用加熱板維持小鼠體溫在37度下，使用動脈夾夾住小鼠左側腎動脈及腎靜脈25-26分鐘後再移除動脈夾，使小鼠左腎受到缺血再灌流的損害，損害四週後再將右腎摘除。左腎假手術及四週後右腎假手術的C57BL/6小鼠則做為對照組。 我們收集並檢驗同一批疾病組與對照組C57BL/6小鼠，在不同時間點的血液檢體，發現血中尿素氮和肌酸酐在對側腎臟摘除後第7天明顯升高，在往後至第126天的各時間點也持續顯著高於對照組，無法回復到對側腎臟摘除前(即第0天)的數值。以尿液白蛋白/肌酸酐比例分析發現疾病組的尿液白蛋白排洩有比對照組低的趨勢，且在摘除右側腎臟後第18天與第53天達到統計差異。疾病組的尿液中KIM-1/肌酸酐比例在摘除右側腎臟前顯著高於對照組，雖然摘除右側腎臟後第4天仍然顯著高於對照組，但之後就下降至兩組相當的程度。我們接著進行另一批對照實驗，在進行對側腎臟摘除前(第0天)與對側腎臟摘除後(第7、14、與28天)犧牲小鼠，並以第一次假手術四週後小鼠的左側腎臟作為對照組(相當於疾病組在摘除對側腎臟前的時間點)。從Periodic Acid-Schiff染色的腎臟組織切片中發現經過缺血再灌流的左腎腎絲球大小，在摘除右側腎臟前後都與對照組沒有差異，摘除右腎前後各時間點之間也無差異。以Periodic Acid-Schiff染色的腎臟組織切片評估腎小管間質傷害，發現疾病組經過缺血再灌流的左腎明顯增加腎小管間質傷害，但在右腎摘除的前後並未發現腎小管間質傷害有所改變。在以Picrosirius red染色的腎臟組織切片中發現疾病組經過缺血再灌流的左腎明顯增加纖維化，但在右腎摘除的前後並沒有看到纖維化明顯改變。於是我們比較對照組腎臟，疾病組經過缺血再灌流傷害的左側腎臟在右側腎臟摘除前，摘除後14天與摘除後28天的基因表現。包括促進腎臟纖維化的基因(Acta2, Col1a1, Col3a1, Timp2)，促進細胞外基質水解的基因(Mmp2, Mmp3, Mmp9)等都在缺血再灌流傷害的左側腎臟在右側腎臟摘除前達到表現量的最高點，右側腎臟摘除後可觀察到Col3a1, Mmp2與Mmp9的表現量顯著的下降，但Acta2, Col1a1與Mmp3的表現量下降都未達到統計上的差異。促進腎臟纖維化的基因Timp3在缺血再灌流傷害的左側腎臟則比對照組腎臟表現量低。過去文獻上發現表現於巨嗜細胞且能促進細胞外基質水解的Mmp13在缺血再灌流傷害的左側腎臟則比對照組腎臟表現量低，右側腎臟摘除後第28天可觀察到Mmp13的表現量回升至相當於對照組腎臟的水準。同樣也是促進細胞外基質水解的基因Tpa和Upa則在右側腎臟摘除後第14天表現量最低。具有促進腎小管再生與抑制發炎的基因Bmp7則在缺血再灌流傷害的左側腎臟降低表現，摘除右側腎臟並未影響Bmp7的表現。但代表細胞增生的Ki67，巨嗜細胞的Adgre1與代表腎小管傷害的Havcr1等基因則都在缺血再灌流傷害的左側腎臟持續增加表現。與纖維母細胞活化及細胞外基質蛋白折疊相關的Txndc5表現量則都沒有差異。我們延長觀察疾病組至術後6個月，在與同時進行假手術的對照組相比之下，發現缺血再灌流傷害的左側腎臟仍有顯著的纖維化與腎小管間質損傷，而且腎絲球變大達到統計差異。因此左腎經過缺血後缺血後再灌注手術的C57BL/6小鼠，在四週後摘除右腎，表現出持續升高的血中尿素氮與肌酸酐、腎臟纖維化、腎小管與間質損傷、發炎等慢性腎病特徵。 因為過去的研究發現硫氧還蛋白域蛋白5(TXNDC5)可經由促進纖維母細胞活化、細胞外基質摺疊與穩定乙型轉化生長因子受體以促進器官的纖維化，因此我們利用兩種不同的誘導型基因剔除小鼠來誘發左側腎臟缺血再灌流傷害並在四週後摘除右側腎臟以進行研究，包括可誘導全身細胞剔除Txndc5的Ubc-CreERT2Tg;Txndc5F/F小鼠與可誘導出腎臟周細胞剔除Txndc5的Gli1CreERT2/+;Txndc5F/F小鼠。我們發現將Txndc5 剔除後可減輕缺血再灌流傷害腎臟的血中尿素氮與肌酸酐上升，而且因缺血再灌流傷害升高的纖維化相關基因包含Acta2, Col1a1與Col3a1等的表現也會在Txndc5 剔除後顯著減少下來。 因此本研究成功建立一小鼠在缺血再灌流手術誘發急性腎損傷之後轉變為可評估腎功能的慢性腎病模式。在進行缺血再灌流手術之前誘導在全身細胞或血管周細胞剔除Txndc5可以減輕腎功能的惡化。未來對於本急性腎損傷產生慢性腎病的機轉應進一步釐清，尤其是右側腎臟摘除後引起在經缺血再灌流傷害後腎臟的變化，例如腎小管的再生與再分化，腎臟間質炎性細胞浸潤的變化，腎臟周細胞的活化等。剔除或抑制TXNDC5是否能促進已纖維化的腎臟進行細胞外基質的減少或腎小管細胞的再生與再分化也可利用本小鼠模式釐清。過去臨床報告指出，發生急性腎損傷的病患，即使在治療後恢復了腎功能，但其轉變為慢性腎病的機率仍比正常人高，且臨床上大多數的慢性腎病患者在開始治療前即已有顯著的腎臟纖維化。因此我們期待本研究開發的小鼠模式可作為開發新穎治療的重要工具。

Renal fibrosis, which often leads to organ failure and death, is the main challenge for patients with kidney disease. A previous study shows the rescuing effect on kidney fibrosis by removing the contralateral kidney two weeks after unilateral ischemia-reperfusion injury(uIRI). In another study, they remove the contralateral kidney one week after uIRI, which investigates the worse of kidney function and fibrosis eight weeks after uIRI. However, we found high mortality in mice undergoing contralateral nephrectomy (Nx) one week after uIRI. Therefore, we deferred Nx to four weeks after uIRI. We used a homeothermic blanket system to maintain the body temperature of 11-week-old wild type male C57BL/6 mice at 37 oC. Left renal artery and vein was clamped for 25 or 26 minutes with a nontraumatic microaneurysm clip to induce uIRI and then the right kidney was removed four weeks after uIRI surgery in the disease group. We found that plasma levels of blood urea nitrogen (BUN) and Creatinine increased and peaked at day 7 after Nx, and continued to be significantly higher than those in the control group through the whole period to day 126 after Nx. The urinary albumin/creatinine ratio was lower in the disease group after Nx. The urinary KIM-1/creatinine ratios in the disease group 5 days before Nx and 4 days after Nx were notably higher than in the control group, but they decreased to comparable levels afterward. We then collected the left kidneys at day 0, day 7, day 14 and day 28 after Nx. The volumes of glomeruli were not different between the kidneys assessed in the tissue sections. The percentage of kidney sections with tubulointerstitial injury and fibrosis increased significantly in the kidney four weeks after uIRI when compared with the control kidney. To detect any trivial change in the kidney after uIRI happening after Nx, we compared gene expression in the control kidney and the disease kidney at day 0, day 14, and day 28 after Nx. The expression of Acta2, Col1a1, Col3a1, Timp2, Mmp2, Mmp3 and Mmp9 was the highest in the uIRI kidney before Nx. The expression of Col3a1, Mmp2, and Mmp9 decreased significantly after Nx. Compared with the control kidney, the profibrotic gene Timp3 decreased in uIRI kidney. We found that Mmp13 expression decreased in the uIRI kidneys, but it increased to the levels of the control kidney at day 28 after Nx. The expression of Tpa and Upa was lowest at day 14 after Nx, but returned to the levels of the control kidney at day 28 after Nx. The expression of Bmp7 decreased in the uIRI kidney and did not change after Nx. The expression of genes involved in Ki67, Adgre1 and Havcr1 all increased in the uIRI kidneys and continued after Nx. There was no difference in the expression of Txndc5. We then extended observation to 6 months after Nx, and found persistent fibrosis and tubulointerstitial injury in the uIRI kidneys. Furthermore, increased glomerular volumes were demonstrated when compared with the control kidney. We then focused on the role of thioredoxin domain-containing 5 (TXNDC5), which has been proved to promote fibroblast activation, extracellular matrix protein folding and stabilization of TGF-β receptor signaling, in organ fibrosis. We induced Txndc5 knockout four weeks before uIRI in two mouse lines with Ubc-CreERT2Tg;Txndc5F/F and Gli1CreERT2/+;Txndc5F/F, and then performed right Nx four weeks after uIRI. We found that Txndc5 deficiency attenuated the increase of plasma BUN and Creatinine at day 2 after Nx. Besides, we also found that the profibrotic gene expression was downregulated in Txndc5 knockout mice. In conclusion, we have successfully established a murine model of acute kidney injury (AKI)- chronic kidney disease (CKD) transition. Inducing Txndc5 knockout in all cells or pericytes before uIRI could prevent renal fibrosis and attenuate renal function deterioration. In the future, we would further clarify the cellular response in the uIRI kidney after Nx, including tubular epithelial cells, inflammatory cells and pericytes. Moreover, we will use this model to study whether Txndc5 knockout promotes extracellular matrix protein degradation or tubule regeneration and re-differentiation in fibrotic kidney. Evidence has shown that AKI patients, even with functional recovery after treatment, will experience increased risk for progression to CKD. Furthermore, patients usually suffer from significant renal fibrosis before treatments. Overall, we will use this murine model as a tool to discover novel therapies for kidney fibrosis.