磷酸鈣鹽水泥腔室作為免疫隔離裝置在人工胰臟之運用

Abstract

胰島移植 (islet transplantation) 對於第一型糖尿病 (Type 1 diabetes) 為一良好之治療方式。然而，免疫排斥與捐贈者不足限制了胰島移植的運用。因此免疫隔離 (immunoisolation) 的概念被提出，使得我們在不使用抗免疫排斥藥物下，可利用異種胰島細胞作為移植來源，克服捐贈者不足的問題。大部分相關的研究皆利用高分子作為免疫隔離材料來包覆胰島細胞做為人工胰臟 (bioartificial pancreas, BAP)，並移植於腹腔中(peritoneal cavity)。但卻很容易因為纖維組織包覆以及氧氣供給不充足，造成人工胰臟失效。因此，本實驗利用磷酸鈣鹽水泥 (calcium phosphate cement) 來取代高分子材料作為免疫隔離裝置，並將人工胰臟移植入骨髓腔中 (intramedullary cavity)，我們希望新生骨組織可以長到磷酸鈣鹽水泥腔室上，來避免纖維組織包覆；同時藉由骨髓腔中高氧分壓，來克服貧氧 (hypoxia) 的問題，以增長人工胰臟之效期。本研究的目的為探討磷酸鈣鹽水泥作為免疫材料在人工胰臟運用之可能性，以及評估骨髓腔作為人工胰臟移植處之可行性。 首先製備四鈣磷酸鹽 (tetracalcium phosphate)，並以等莫爾比 (equimolar) 混合雙鈣磷酸鹽 (dicalcium phosphate) 來建造磷酸鈣鹽水泥腔室 (chamber)，並利用X光繞射儀 (X-ray diffraction)、掃描式電子顯微鏡 (scanning electron microscope) 以及水銀孔洞分析儀 (mercury intrusion porosimetry) 來分析磷酸鹽水泥，其分子穿透性質 (molecular weight cut-off) 以及材料毒性 (cytotoxicity) 也進行評估。結果顯示磷酸鈣鹽水泥具有良好生物相容性，其微結構符合免疫隔離裝置之要求，並可阻斷分子量大於12.4千道爾頓 (kDa) 分子之穿透。 接著，胰島瘤細胞株 (insulinoma) 被選擇作為細胞來源，包覆在瓊脂 (agarose) 中成為微包埋球 (microenspheres)。含有胰島細胞之微包埋球被填入磷酸鈣鹽水泥腔室中，創造出一新型人工胰臟。此人工胰臟在體外 (in vitro) 以胰島素分泌、細胞活性、細胞存活以及細胞激素 (cytokine) 毒性試驗來評估，並植入糖尿病大鼠體內作初步的體內 (in vivo) 試驗。非禁食血糖以及血清胰島素濃度則在術前、術後測定，並且在實驗後將植入之人工胰臟從大鼠體中取出進行分析。結果顯示，即使培養在含有細胞激素之培養液，人工胰臟內之胰島瘤細胞仍保有正常活性、存活率以及胰島素分泌。糖尿病大鼠在腹腔內移植人工胰臟四天後，非禁食血糖由460±50下降至132±43 mg/dl，並維持在正常血糖達22天。其血清胰島素濃度在術後由0.34±0.11上升至1.43±0.30 μg/dl。組織切片顯示，纖維組織包覆以及免疫相關細胞競爭氧氣可能為造成人工胰臟失效之主因。 接著，本人工胰臟移植入糖尿病犬之骨髓腔中，並定期測量術前、術後血糖值。血液樣本則收集以利分析C-peptide濃度，犬隻生理狀況同時觀察記錄，在實驗結束後並取出人工胰臟作分析。另一隻自發性糖尿病貓也接受本人工胰臟之植入。結果顯示，糖尿病犬其血糖值在術後一天由424±23下降至243±33 mg/dl，並維持在219-349 mg/dl達12週。術後血清C-peptide濃度則由5.3±2.8上升至105.7±19.4 pmol/l。犬隻餐後血糖下降速率也增快。組織切片則顯示，骨組織與植入之人工胰臟表面建結，中間並無纖維組織產生。免疫染色顯示，人工胰臟內的胰島瘤細胞株持續分泌胰島素。關於糖尿病貓，兩小時葡萄糖曲線之峰點 (peak point of two hours glucose curve) 在術後由398下降至165-290 mg/dl。外源性胰島素之效率也由2小時延長到10-14小時，糖尿病貓之體重以及生理狀況皆改善。 本實驗證明了磷酸鈣鹽水泥腔室作為免疫隔離材料之可能性，同時證明了骨髓腔做為人工胰臟移植處之可行性，並論證了此一策略可避免纖維組織包覆以及貧氧導致的人工胰臟失效。將來，利用正常胰島來取代胰島瘤細胞株，並增加植入之人工胰臟數量，應可達到胰島素獨立 (insulin independent)，並可提供第一型糖尿病病患一嶄新的治療方式。

Immune rejection and insufficient donor supply are the restrictions of islet transplantation for type 1 diabetes (T1D). A strategy so called as immunoisolation was purposed to facilitate the use of xenogeneic cell sources to resolve the issue of an insufficient donor supply in the absence of immunosuppression. Most studies utilized polymeric materials as an immunoisolative material to encapsulate islets as a bioartificial pancreas (BAP) and implanted in the peritoneal cavity. However, BAPs would be dysfunctional because of the fibrous tissue overgrowth and an insufficient oxygen supply. Accordingly, a calcium phosphate cement (CPC) chamber was utilized to replace the polymeric material as an immunoisolative device. BAPs were implanted in the intramedullary cavity instead of the peritoneal cavity. We hope the newborn bone tissues can grow on the CPC chamber to avoid the development of fibrous tissue. In the meantime, the sufficient partial oxygen pressure in the intramedullary cavity can overcome the issue of hypoxia. The purpose of this study was as to evaluate the feasibility of the use of a CPC chamber as an immunoisolative device, and to demonstrate the possibility of intramedullary cavity as an implanted site for BAPs. Tetracalcium phosphate was first prepared and mixed with dicalcium phosphate in equimolar to fabricate a CPC chamber. The CPC chamber was analyzed by X-ray diffraction, scanning electron microscope and mercury intrusion porosimetry. The molecular weight cut-off (MWCO) and material-mediated cytotoxicity of CPC chamber were also evaluated. Results showed that the CPC chamber was non-cytotoxicity to insulinoma cells. The microstructure of CPC chamber satisfied the requirements of an immunoisolative device with the MWCO of 12.4 kDa, which can totally cut-off the diffusion of immune stimulatory molecules. Next, an insulinoma cell line was applied as a cell source and encapsulated in agarose microspheres. Insulinoma/agarose microspheres were enclosed in a preformed CPC chamber to create a BAP. BAPs were evaluated in vitro by insulin secretion, cell viability, live/dead cell ratio, cytokine-mediated cytotoxicity assay, and implanted in the peritoneal cavity of diabetic rats as preliminary in vivo study. Non-fasting blood glucose level (NFBG) and serum insulin level were analyzed perioperatively; BAPs were also retrieved for histological examination. Results showed insulinoma cells enclosed in the CPC chamber had normal viability, survival and insulin secretion even when cultured in media with cytokines. The NFBG of rats was decreased from 460±50 to 132±43 mg/dl four day post-operation and maintained in euglycemia for 22 days; serum insulin level was increased from 0.34±0.11 to 1.43±0.30 μg/dl post-operation. Histological examination revealed the fibrous tissues overgrowth and immune related cells competed for oxygen resulting in hypoxia could be attributed to the dysfunction of BAPs. BAPs were then implanted in the femoral intramedullary cavity of diabetic canines. Pre- and postprandial blood glucose level was monitored perioperatively. Blood samples were collected for the analysis of C-peptide level and physiological conditions were observed at pre-determined intervals. BAPs were retrieved 12 weeks post-operation for histological examinations. A spontaneous diabetic feline was also received the BAPs implantation. Results showed the preprandial blood glucose level of diabetic canines was decreased from 424±23 to 243±33 mg/dl one day post-operation and maintained in the range of 219-349 mg/dl for 12 weeks. Serum C-peptide level was increased from 5.3±2.8 to 105.7±19.4 pmol/l. The decreasing rate of postprandial blood glucose was accelerated. Histological examinations revealed recipients’ bone tissues binding to the surfaces of BAPs directly; there was no fibrous tissue development. Immunohistochemical stain showed positive insulin staining for the enclosed insulinoma cells. For the diabetic feline, the peak point of two hours glucose curve was decreased from 398 to 165-290 mg/dl post-operation; the effectiveness of exogenous insulin was extended from 2 to 10-14 hrs and the physiological conditions were improved. This study proves the feasibility of using a CPC chamber as an immunoisolative device. The possibility of intramedullary cavity as an implanted site for BAPs had also been proved. BAPs implanted into the intramedullary cavity can avoid the problems of fibrous tissue outgrowth and hypoxia. The use of native islets to replace the insulinoma cells and to increase the number of implanted BAPs or the islet cells density within BAPs should be considered to achieve insulin independent, and may provide a new treatment for T1D in the future.