高血糖狀態調控間質幹細胞分化成脂肪細胞及造骨細胞之機制探討

Abstract

糖尿病是指人體內的胰臟不能製造足夠的胰島素或是胰島素失去功能，導致葡萄糖無法充分進入細胞內，使血糖濃度升高，因而形成糖尿病。雖然糖尿病是一種多樣性的症候群，但其基本表現都以高血糖為主。長期處於高血糖狀態會產生一些糖尿病慢性併發症，包括視網膜病變、神經病變、腎臟病變及糖尿病足等等。而糖尿病患者也常出現骨折後骨頭癒合不良的情況，但是糖尿病造成骨質疏鬆的原因尚未釐清。基於骨頭內的間質幹細胞可以分化成造骨細胞及脂肪細胞等等，而且在臨床上也發現隨著年齡的增加，骨頭內的造骨細胞減少，取而代之的是脂肪細胞的增加，進而造成骨質疏鬆症。因此，本篇主要是探討在高血糖狀態下影響間質幹細胞的分化及其相關的分子機制。我們使用25.5mM的高葡萄糖溶液處理間質幹細胞來模擬糖尿病患者的高血糖狀態。 首先，我們證明高葡萄糖會經由phosphoinositide 3-kinase (PI3K)/Akt而促進間質幹細胞分化成脂肪細胞。取6~8週大的FVB/N雄性小鼠的股骨及脛骨，沖出骨髓後培養七天，先單獨處理10-8 M Dexamethasone (DEX) 及 5 μg/ml胰島素，觀察間質幹細胞分化成脂肪細胞的情況。我們發現在第12天脂肪細胞數目明顯的增加，因此在10-8 M DEX 及 5 μg/ml 胰島素下合併處理25.5 mM的高葡萄糖及一個已知會促進脂肪細胞分化的內生性peroxisome proliferators-activated receptor gamma (PPARγ)受質- 15-Deoxy-△12,14-prostaglandin J2 (15d-PGJ2)。12天後以流式細胞儀分析，發現在有25.5 mM的高葡萄糖情況下，脂肪細胞分化的數目比只有處理10-8 M DEX 及 5 μg/ml胰島素的這組增加了1倍，而處理1 μM 15d-PGJ2這組增加了4倍的脂肪細胞。接著，我們探討高葡萄糖促進脂肪細胞分化的機制。在1999年，Wu等人就發現PPARγ這個轉錄因子會調控脂肪細胞的分化。因此，我們合併處理高葡萄糖溶液及15d-PGJ2 12天，發現高葡萄糖溶液及15d-PGJ2會增加PPARγ表現量，而且會被PPARγ拮抗劑- GW 9662 (20 μM) 所抑制。另外，合併處理高葡萄糖溶液及15d-PGJ2也會增加Akt的磷酸化，然而處理PI3K的抑制劑- LY 294002 (7.5 μM) 後，除了會抑制Akt的磷酸化外，也會抑制高葡萄糖溶液及15d-PGJ2促進的PPARγ表現。而且主要是抑制PPARγ的isoform-PPARγ2。由以上的實驗，我們認為高葡萄糖溶液及15d-PGJ2皆會藉由PI3K及Akt這個路徑來促進PPARγ的表現，進而促進脂肪細胞的分化。為了加以確認，我們以轉染作用給予隱性變異質體DN-p85 和 DN-Akt，皆可以減少高葡萄糖溶液及15d-PGJ2促進的PPARγ表現。 由以上實驗，我們認為PI3K/Akt路徑在調控間質幹細胞的分化可能扮演一個很重要的角色。所以，我們更進一步探討PI3K/Akt是否會調控間質幹細胞分化成造骨細胞及其相關的分子機制。我們把沖出的骨髓培養7天後，單獨處理10-8 M DEX，5 μg/ml維他命C及10 mM β-甘油磷酸觀察間質幹細胞分化成造骨細胞的情況。1990年，Owen等人就發現從大鼠的間質幹細胞分化成造骨細胞到造骨細胞死亡的過程需要28天，而在這整個時期，鹼性磷酸酶的活性是屬於早期造骨細胞分化的指標，而骨鈣素的表現及骨礦物質化是屬於晚期的造骨細胞分化指標。而我們的實驗也發現在造骨細胞的分化過程中，第12天的鹼性磷酸酶活性最高，同時合併處理1 μM 15d-PGJ2及25.5 mM 高葡萄糖溶液後，會抑制鹼性磷酸酶的活性，但處理GW 9662 (20 μM) 及LY 294002 (7.5 μM) 後會增加被15d-PGJ2所抑制的鹼性磷酸酶活性，而LY 294002也會增加被高葡萄糖溶液所抑制的鹼性磷酸酶活性。另外我們也觀察造骨細胞分化的晚期指標，發現在第22天骨鈣素的mRNA表現量最大，同時合併處理15d-PGJ2及高葡萄糖後，15d-PGJ2會抑制骨鈣素mRNA的表現，而此現象也會被GW 9662及LY 294002所阻斷，但高葡萄糖溶液則沒有作用。為了加以確認，我們以轉染作用給予隱性變異質體DN-p85 和 DN-Akt，在造骨細胞分化的第22天，發現會抑制15d-PGJ2的作用，進而增加骨礦物質化，但高葡萄糖溶液仍沒有影響。因此，第二部分，我們證明高葡萄糖會經由PI3K/Akt而抑制早期的間質幹細胞分化成造骨細胞。 最後，我們以低劑量streptozotocin (STZ)連續腹腔注射5天，誘導6~8週大的FVB/N雄性小鼠成第一型糖尿病，經過3週後，檢查其血糖是否高於400 mg/dl，沖出控制組及糖尿病組小鼠骨頭內骨髓培養，經過7天後，處理DEX、胰島素及合併高葡萄糖溶液或15d-PGJ2，發現糖尿病組小鼠的間質幹細胞分化成脂肪細胞的比例較控制組高。另外，我們也取控制組及糖尿病組小鼠的脛骨和股骨的骨幹端，大約在生長板的位置，利用液態氮磨碎後，加入溶解緩衝液，測其三酸甘油酯的含量及鹼性磷酸酶的活性，發現糖尿病組小鼠骨頭內的三酸甘油酯含量比控制組小鼠增加1倍，而糖尿病組小鼠的骨頭鹼性磷酸酶活性比控制組小鼠稍低。最後，再利用南方點墨法分析，也發現糖尿病組小鼠的骨頭內PPARγ表現量比控制組高。因此，第三部分，我們證明高血糖的糖尿病小鼠，其骨頭內脂肪組織比控制組多。 綜合以上實驗，我們認為PI3K/Akt在調控間質幹細胞分化上扮演很重要的角色，包括活化PI3K/Akt會增加PPARγ表現進而促進間質幹細胞分化成脂肪細胞，而相反地活化PI3K/Akt會減少鹼性磷酸酶、骨鈣素表現及骨礦物質化進而抑制間質幹細胞分化成造骨細胞。

Diabetes is characterized by mild to moderate hyperglycemia, glucosuria, polyphagia, hypoinsulinemia, hyperlipidemia, and weight loss. All forms of diabetes are characterized by chronic hyperglycemia and the development of many serious complications, for example, heart disease (cardiovascular disease), blindness (retinopathy), nerve damage (neuropathy), and kidney damage (nephropathy). Diabetes has also been reported with a net loss of bone. However, the effect of hyperglycemia on bone loss remains unclear. Bone loss in aging and osteoporosis are associated with a decrease in the number and activity of osteoblasts and a parallel increase in the number of adipocytes. Here we have demonstrated that hyperglycemia affected the mesenchymal stem cells (MSCs) differentiation by enhancing adipocyte differentiation (adipogenesis). We used high glucose (HG, 25.5 mM) to mimic the hyperglycemia condition. To determine the effects of hyperglycemia on adipogenesis, we cultured mouse MSCs in an adipogenic hormonal cocktail, and adipogenesis was strong enhancement by supplementation of HG and 15-deoxy-∆12,14-PGJ2 (15d-PGJ2), which has been identified as an endogenous ligand for peroxisome proliferator-activated receptor gamma (PPARγ), inducing adipogenesis in vitro. This ligand improves insulin sensitivity through the activation of the transcription factor, PPARγ. In addition to sensitizing cells to insulin, the PPARγ2 isoform appears to be critical for the regulation of osteoblast differentiation (osteoblastogenesis) and adipocyte differentiation (adipogenesis) of MSCs in bone marrow. In this HG cultures, the expression of PPARγ2 was up-regulated even prior to adipogenic induction. Moreover, treatment with PPARγ agonists, GW 9662 (20 μM) or inhibitors of phosphatidylinositol 3-kinase (PI3K), LY 294002 (7.5 μM), leads to the complete blockade of HG-enhanced adipogenesis of MSCs by inhibited the PPARγ expression. HG-activated Akt on adipogenesis of MSCs was also inhibited by LY 294002. Likewise, blocking the PI3K or Akt activity with the dominant-negative vectors DN-p85 or DN-Akt, respectively, also greatly inhibited the HG-enhanced the expression of PPARγ. These suggesting that HG enhanced adipogenesis of MSCs in the adipogenic medium may through a PI3K/Akt regulated PPARγ pathway. Another intriguing finding was that 15d-PGJ2 (1 μM) enhanced adipogenesis by increasing the PPARγ expression which was inhibited by LY 294002, DN-p85 and DN-Akt. Collectively, these data provide a new insight into the PI3K/Akt pathway on MSCs differentiation. So, next we examine the mechanisms of osteoblastogenesis of MSCs. First, we found that 15d-PGJ2 and HG decreased alkaline phosphatase activity, which was used as early differentiation markers of osteoblastogenesis from MSCs in osteoblastogenic medium, and treatment with GW 9662 or LY294002 was significantly increasing the alkaline phosphatase activity. Second, the 15d-PGJ2 inhibited the expression of the osteocalcin, which gene marked the late stages of osteoblastogenesis, and treatment with GW 9662 or LY294002 was significantly increasing the osteocalcin gene expression. But HG was no effect on osteocalcin expression. Third, mineralization was reduced by treatment of MSCs with 15d-PGJ2 in osteoblastogenic medium and dominant-negative vectors DN-p85 or DN-Akt greatly increased mineralization. HG was also no effect on mineralization. Mineralization correlated closely with osteocalcin gene expression. Taken together, these results indicate that activation of PI3K/Akt pathway in MSCs may increase adipogenesis and decrease osteoblastogenesis of MSCs. On the other hand, in in vivo study, mice were made diabetic by multiple low-dose streptozotocin (STZ) treatment, and controls were treated with vehicle alone. After 3 weeks, chose the diabetic mice which had ≧400 mg/dl blood sugar. We cultured diabetic MSCs and control MSCs in adipogenic medium combined with HG or 15d-PGJ2 and compared their adipogenesis of MSCs. We found that the level of diabetic adipogenesis of MSCs was higher than the level of control mice. Moreover, the tibial bone from the proximal metaphysis to the tibiofibular junction was snap frozen in liquid nitrogen, pulverized, and we found that the expression of PPARγ and triglyceride amounts were significantly increased in the diabetic bone whereas alkaline phosphatase activity was reduced in the diabetic bone. These findings support a reciprocal relationship between the development of bone and fat under hyperglycemia, and may prompt further exploration of the PI3K/Akt regulated MSCs differentiation as a potential target for intervention in diabetic osteoporosis.