Inhibitory Effects of Gymnema (Gymnema sylvestre) Leaves on Tumour Promotion in Two-Stage Mouse Skin Carcinogenesis

Ethanol extracts of gymnema (Gymnema sylvestre) leaves exhibited marked antitumour-promoting activity in an in vivo two-stage carcinogenesis test in mice using 7,12-dimethylbenz[a]anthracene as an initiator and 12-O-tetradecanoylphorbol-13-acetate (TPA) as a promoter. From the active fraction of the ethanol extract of the gymnema leaves, three triterpenoids were isolated and identified. These compounds were evaluated for their inhibitory effects on TPA-induced inflammation (1 µg/ear) in mice. The tested compounds showed marked anti-inflammatory effects, with a 50% inhibitory dose of 50–555 nmol/ear.


Introduction
The chemoprevention of cancer is an urgent priority in the field of public health. A method of prevention that acts at the promotion stage of carcinogenesis is most desirable, as such a method could be applied even after exposure to tumour-promotion agents, which in many cases is unavoidable in daily life. Many tumour promoters have inflammatory activity [1]. A correlation between the inhibitory effects against 12-O-tetradecanoylphorbol-13-acetate (TPA)induced ear inflammation and inhibition of TPA-induced tumour promotion in a two-stage carcinogenesis experiment was observed in mice [2]. To screen for new inhibitors as chemopreventive agents, we intentionally selected natural compounds.

TPA-Induced Inflammation Assay in Mice.
TPA (1 g) dissolved in acetone (20 L) was applied to the right ear of ICR mice by means of a micropipette. A volume of 10 L was delivered to both the inner and outer surfaces of the ear. The sample (0.02-1.0 mg/ear) or vehicle, methanolchloroform-water (2 : 1 : 1; 20 L) or methanol-chloroform (1 : 1; 20 L), as a control, was applied topically about 30 min before TPA treatment. For ear thickness determination, a pocket thickness gauge (Mitsutoyo Co. Ltd., Tokyo, Japan) with a range of 0-9 mm (graduated at 0.01-mm intervals and modified so that the contact surface area was increased, thus reducing tension) was applied to the tip of the ear.
Ear thickness was determined before TPA treatment ( ). Oedema was measured at 6 h after TPA treatment ( : TPA with vehicle; : TPA with sample). The following values were then calculated: Oedema A: oedema induced by TPA with vehicle ( − ).

Inhibitory ratio (%) = [(oedema A − oedema B)/oedema A] × 100.
Each value was the mean of individual determinations from four mice. acetone and applied to the shaved area in a volume of 100 L using a micropipette. The back of each animal was shaved once a week to remove hair. The number and diameter of skin tumours were measured every week, and the experiment was continued for 20 weeks. Experimental and control groups each consisted of 15 mice.

Statistical
Analysis. The 50% inhibitory dose (ID 50 ) values and their 95% confidence intervals (95% CI) were obtained by nonlinear regression using the GraphPad program 5.0 (Intuitive Software for Science, San Diego, CA). Differences between experimental groups were compared by Student's -test and Mann-Whitney exact test.

Results and Discussion
As can be seen in Table 1, extracts from gymnema leaves inhibited TPA-induced inflammation in mice. The inhibitory effects of the ethanol extract of gymnema leaves in a twostage carcinogenesis test on mouse skin using DMBA as an initiator and TPA as a tumour promoter were then investigated. Figure 1(a) illustrates the time course of skin tumour formation in the groups treated with DMBA plus TPA, with or without the ethanol extract of gymnema leaves. The first tumour appeared at week 5 in the group treated with DMBA plus TPA and all 15 mice had tumours at week 13. In the group treated with DMBA plus TPA and ethanol extract of gymnema leaves, the first tumour appeared at week 9. The percentage of tumour-bearing mice treated with DMBA plus TPA and ethanol extract of gymnema leaves was 27% at week 20. Figure 1(b) shows the average number of tumours per mouse. The group treated with DMBA plus TPA produced 7.7 tumours per mouse at week 20; the group treated with DMBA plus TPA and ethanol extract of gymnema leaves had 1.8 tumours per mouse. Treatment with ethanol extract of gymnema leaves caused a 77% reduction in the average number of tumours per mouse at week 20. Active components were then isolated from the ethanol extract of gymnema leaves. The isolated compounds showed inhibitory activity against TPA-induced ear inflammatory oedema. As can be seen in Table 2, the ID 50 values of 2-4 against TPA-induced inflammation were 49.7-555 nmol/ear, respectively.
This is the first report to find that ethanol extracts of gymnema leaves inhibit tumour promotion by TPA following initiation with DMBA in ICR mouse skin. Furthermore, the active constituents were identified as three triterpenes from ethanol extracts of gymnema leaves. This is the first report of phenethyl glucoside in the genus Gymnema and the first report of 28-acetyl-21-tigyolgymnemagenin in nature. These results demonstrate the efficacy to two triterpenes and triterpene glycosides in the components of gymnema leaves. In our study, we reported that numerous triterpenes and their glycosides are effective for preventing cancer [27,28]. Therefore, we inferred that triterpenes and their glycosides in gymnema leaves are the active components.