| Diabetes is a kind of endocrine metabolic disease caused by hypoinsulinism or Insulin Resistance, featured in hyperglyceamia and melitura. It now ranks next to cardiovascular disease and cancers as the third most serious disease, while the non-insulin-dependent diabetes mellitus (NIDDM) takes up 80 % of the cases.Nowadays, there are several kinds of antidiabetic medicine in clinical application. Mainly they are Insuline,Insuline analog,sulfonylurea,biguanides,thiazolidine (TZD), etc. however most of them have such insufficiencies as inconvenient for use,insufficient curative effect,toxicity, while the Chinese antidiabetic medicines are composite, the effective component is not sure, which all make the applications for patents very difficult. Therefore, to develop a new antidiabetic medicine with greater effect and lower side effect becomes the hot spot among the international new drug study area.In this experiment we established three screening models based respectively on cell,molecular and living animal. We try to use those models to find some small compound with anti-diabetic activity by novel mechanism or novel structure. [1] In vitro glucose uptake model. was established to exam the stimulation effect of 230 compounds on glucose uptake. The 3T3-L1 cells were differentiated into adipocyte by hormones in the microtitre plate, then the mature adipocyte cells were treated respectively with insulin at 10 mg·L-1 as positive control or 230 compounds at 50,5,0.5 mg·L-1for 48 h, and then the glucose concentration of cells culture supernatants was detected. 63 compounds showed different rate of glucose uptake effects with NO.33,51,96,200,202,211,212,213 at concentration of 50 mg·L-1, and NO.170 at concentration of 5 mg·L-1 ; [2] PPARγsystem screening model was established to find compounds with PPARγ, activation within those 63. A recombinant reporter vector contained PPRE and a PPARγ, expression vector were transiently co-transfected into cells. The expression of reporter gene was examed to determine the PPARγactivation. 7 compounds were 'hit', the induction fold of report gene in 293T cells by NO.60 at dose of 1mg·L-1 was above 2 times, the effect close to rosiglitazone at dose of 10mol·L-1 (p>0.05); [3]The ALX diabetic mice model was established to exam the in vivo hypoglycemic activity of 4 compound of the 7. we give the ALX diabetic mice MET at dose of 250 mg·Kg-1·d-1 as positive drug, NO.51 at doses of 12.5,25,50 mg·Kg-1·d-1, NO.60,192,203 at doses of 25,50,100 mg·Kg-1·d-1for 7 consecutive days, then to exam the blood sugar at 4 and 7 days respectively, the mice blood sugar of NO.51 sample at dose of 25 mg·Kg-1·d-1, of NO.60 sample at dose of 100mg·Kg-1·d-1, of NO.203 sample at 100 mg·Kg-1·d-1 were significantly decreased (p<0.01 or p<0.05 compared with model control group); NO.192 sample didn't show any hypoglycemic effect at dose from 25~100 mg·Kg-1·d-1 (p>0.05 compared with model control group).In this experiment two vitro models and one animal model were developed successfully. The glucose uptake model was characterized as low cost, less sample quantity demanded, the higher screening efficiency etc, except that poor stability and no specificity; The recombinant reporter drug screening model was characterized as single targeted, higher specialized and stable; Intravenous injection of ALX resulted in significant increase of blood sugar in 80% mice which can remain stable for 7 days, and the death rate is low. 63 of 230 compounds have different level of stimulating glucose uptake effect. 7 of which can activate PPARγ, signal pathway. NO.51,NO.60 and NO.203 was confirmed to have in vivo hypoglycemic effect, which is close to MET. |