Aromatic Constituents from the Stems of Astragalus membranaceus (Fisch.) Bge. var. Mongholicus (Bge.) Hsiao

Four new aromatic constituents, astraflavonoids A (1), B (2), C (3), and astramemoside A (4), along with sixteen known ones 5–20 were obtained from the stems of A. membranaceus (Fisch.) Bge. var. mongholicus (Bge.) Hsiao. Their structures were elucidated by chemical and spectroscopic methods. Among the known isolates, 14 was obtained from the Astragalus genus for the first time, while 7–12, 18–20 were isolated from the species for the first time. The effects of the compounds obtained from the plant on glucose consumption were analyzed in differentiated L6 myotubes in vitro, whereby compounds 1, 2, 3, 7, 8, 10, 11, 14, 15 and 18 displayed significant promoting effects on glucose consumption in L6 myotubes. Among them, the activities of 1, 2 and 7 were comparable to that of insulin, which suggested that these compounds may be involved in glucose metabolism and transport.


Introduction
Astragalus membranaceus (Fisch.) Bge. var. mongholicus (Bge.) Hsiao (AM), belongs to the Astragalus genus of the Leguminosae family. The main chemical constituents of the plant are flavonoids and terpenoids. During the course of our studies, 14 oleanane type saponins, including eight new ones, named astroolesaponins A, B, C 1 , C 2 , D, E 1 , E 2 , and F, have been obtained from its stems, and some of them showed depressing effects on triglyceride levels in sodium oleate-induced HepG2 cells [1]. During our continued research on this species, twenty aromatic constituents, including four new ones, were isolated from AM using SiO 2 gel, ODS, Sephadex LH-20 column chromatography (CC) and preparative HPLC chromatography (Prep HPLC), and their structures were clearly determined by chemical and spectroscopic methods ( 1 H-NMR, 13 C-NMR, 1 H-1 H COSY, HSQC, HMBC, UV, IR, CD, and MS). Based on the evidences of previous activity reports on the Astragalus genus [1], the glucose consumption effects of the isolates were examined.
On the basis of the activity results of kaempferol and its glycosides, we can deduce that the glucosylation of 3-hydroxyl group would increase the effect on glucose consumption in L-6 cells, and disaccharide substitution at 3-hydroxyl showed a stronger activity than monosaccharide substitution. Due to the limited number of compounds, detailed studies are in progress to evaluate more kaempferol glycosides to clarify these structure-activity relationships. Comparison of the 1 H-and 13 C-NMR spectroscopic data of 4 with those of 1 identified the sugar moiety to be β-D-apiofuranosyl(1 Ñ 2)-β-D-glucopyranose. Moreover, the 5-position of the β-D-apiofuranosyl moiety was substituted too. Meanwhile, the long-range correlations from H-2,6 to C-4, C-7; H-1 1 to C-4; H-1 11 to C-2 1 ; H 2 -5 11 to C-7 observed in the HMBC experiment further proved the correctness of the above deductions.
The effects of the compounds obtained from AM on glucose consumption were analyzed in differentiated L6 myotubes in vitro. To create the assay method, standardization of basic parameters like differentiation time, the number of cells to seed, amount of D-glucose to be used and time of incubation were determined (data not shown). As shown in Figure 7, insulin increased the glucose consumption in L6 myotubes to about 4.76˘0.33 µg/well (p < 0.001) and the percentage of the raise reached about 9.01%, which serves as a positive control for our study. Among the tested compounds, 1, 2, 3, 7, 8, 10, 11, 14, 15 and 18 possessed significant promoting effects on glucose consumption in L6 myotubes, the glucose consumption of which were 5.02˘0.29, 4.92˘0.36, 3.20˘0.58, 4.86˘0.67, 3.37˘0.62, 3.04˘0.86, 2.94˘0.60, 3.96˘1.21, 2.44˘0.59 and 3.15˘1.00 µg/well, respectively. At the concentration of 30 µmol/L, compounds 1, 2 and 7 led to 9.67%, 9.48% and 9.33% increments in glucose consumption, respectively, which was comparable to the effects of insulin. However, the other isolates showed no obvious effect on glucose consumption. These results indicated that some of the constituents in AM can stimulate glucose consumption in L6 myotubes to various degrees. The effects of the compounds obtained from AM on glucose consumption were analyzed in differentiated L6 myotubes in vitro. To create the assay method, standardization of basic parameters like differentiation time, the number of cells to seed, amount of D-glucose to be used and time of incubation were determined (data not shown). As shown in Figure 7, insulin increased the glucose consumption in L6 myotubes to about 4.76 ± 0.33 μg/well (p < 0.001) and the percentage of the raise reached about 9.01%, which serves as a positive control for our study. Among the tested compounds, 1, 2, 3, 7, 8, 10, 11, 14, 15 and 18 possessed significant promoting effects on glucose consumption in L6 myotubes, the glucose consumption of which were 5.02 ± 0.29, 4.92 ± 0.36, 3.20 ± 0.58, 4.86 ± 0.67, 3.37 ± 0.62, 3.04 ± 0.86, 2.94 ± 0.60, 3.96 ± 1.21, 2.44 ± 0.59 and 3.15 ± 1.00 μg/well, respectively. At the concentration of 30 μmol/L, compounds 1, 2 and 7 led to 9.67%, 9.48% and 9.33% increments in glucose consumption, respectively, which was comparable to the effects of insulin. However, the other isolates showed no obvious effect on glucose consumption. These results indicated that some of the constituents in AM can stimulate glucose consumption in L6 myotubes to various degrees. (1 × 10 4 cells/well) were subcultured into 48-place multiwell plates in 2% FBS/DMEM for 7 days to form myotubes. The differentiated myotubes were kept in HBS with no serum or glucose for 2 h, and then were continue incubated in HBS containing 1 mg/mL D-glucose and 2% FBS with or without insulin (Ins, 2 μmol/L) or obtained compounds (30 μmol/L) for another 4 h. Then the glucose concentrations in the supernatant were detected using glucose assay kit and the percentage of glucose consumption in each well was calculated to express the results. Each value represents the mean ± S.E.M., n = 6. *** p < 0.001, ** p < 0.01, * p < 0.05 vs. control group (Ctrl).
On the basis of the activity results of kaempferol and its glycosides, we can deduce that the glucosylation of 3-hydroxyl group would increase the effect on glucose consumption in L-6 cells, and disaccharide substitution at 3-hydroxyl showed a stronger activity than monosaccharide substitution. Due to the limited number of compounds, detailed studies are in progress to evaluate more kaempferol glycosides to clarify these structure-activity relationships.  Figure 7. Effects of compounds 1-11 and 13-20 on glucose consumption in L6 myotubes. L6 myoblasts (1ˆ10 4 cells/well) were subcultured into 48-place multiwell plates in 2% FBS/DMEM for 7 days to form myotubes. The differentiated myotubes were kept in HBS with no serum or glucose for 2 h, and then were continue incubated in HBS containing 1 mg/mL D-glucose and 2% FBS with or without insulin (Ins, 2 µmol/L) or obtained compounds (30 µmol/L) for another 4 h. Then the glucose concentrations in the supernatant were detected using glucose assay kit and the percentage of glucose consumption in each well was calculated to express the results. Each value represents the mean˘S.E.M., n = 6. *** p < 0.001, ** p < 0.01, * p < 0.05 vs. control group (Ctrl).
On the basis of the activity results of kaempferol and its glycosides, we can deduce that the glucosylation of 3-hydroxyl group would increase the effect on glucose consumption in L-6 cells, and disaccharide substitution at 3-hydroxyl showed a stronger activity than monosaccharide substitution. Due to the limited number of compounds, detailed studies are in progress to evaluate more kaempferol glycosides to clarify these structure-activity relationships.

Plant Material
The stems of Astragalus membranaceus (Fisch.) Bge. var. mongholicus (Bge.) Hsiao. were collected from Gansu Province, China, and identified by Dr. Li Tianxiang (Experiment Teaching Department, Tianjin University of Traditional Chinese Medicine, Tianjin, China). The voucher specimen was deposited at the Academy of Traditional Chinese Medicine of Tianjin University of TCM.
in the supernatant were determined using glucose assay kit (GOD-POD colorimetric method) and the ratio of glucose consumption in each well was calculated for further comparison. Results were expressed as a percentage of glucose consumption: Percentage of glucose consumption (%) = [(glucose surplus of control´glucose surplus of sample)/glucose surplus of control]ˆ100% (1)

Conclusions
In summary, four new aromatic constituents, astraflavonoids A-C (1-3), and astramemoside A (4), along with sixteen known ones 5-20 were obtained from the 70% EtOH extract of AM. Among the known isolates, 14 was isolated from the Astragalus genus for the first time, and compounds 7-12, 18-20 were isolated from the species for the first time. Their structures were elucidated by chemical and spectroscopic methods. The effects of the compounds obtained from AM on glucose consumption were analyzed in differentiated L6 myotubes in vitro. As results, compounds 1, 2, 3, 7, 8, 10, 11, 14, 15 and 18 possessed significant promotion effects on glucose consumption in L6 myotubes. Among them, the activities of 1, 2 and 7 were comparable to that of insulin, which suggested that these AM compounds may be involved in glucose metabolism and transportat. On the basis of the activity results, the structure-activity was discussed. Glucose consumption plays a role in cellular energy homeostasis. This process includes glucose uptake, translocation, glucose storage, involves many key kinase, including AMP-activated protein kinase, phosphoinositide 3-kinase, glycogen synthase kinase, and so on. Further studies will be carried out to elucidate the mechanism of action of these and other kaempferol derivatives on glucose consumption.