Phenolics from Glycyrrhiza glabra and G. uralensis Roots and Their PPAR-γ Ligand-Binding Activity: Possible Application for Amelioration of Type 2 Diabetes Phenolics from Glycyrrhiza glabra and G. uralensis Roots and Their PPAR- γ Ligand-Binding Activity: Possible Application for Amelioration of Type 2 Diabetes

The EtOH extract of Glycyrrhiza glabra roots and the EtOAc extract of Glycyrrhiza uralensis roots exhibited considerable PPAR-γ ligand-binding activity. Bioassay-guided fraction - ation of these extracts resulted in the isolation of 52 phenolics, including 11 novel ones. The PPAR-γ ligand-binding activity of more than 10 isolated phenolics at 10 μg/mL was approximately three times greater than that of 0.5 μM triglitazone. Glycyrin ( 44 ), isolated from the EtOAc extract of G. uralensis roots as a PPAR-γ ligand, reduced the blood glu cose levels of genetically diabetic KK-A y mice through its PPAR-γ ligand-binding activity. absorption bands at 3374 cm −1 (hydroxy groups), 1673 cm −1 (a carbonyl group), and 1608, 1502, and 1463 cm −1 (aromatic rings) in its IR spectrum. The 1 H NMR spectrum of 7 showed signals that we assigned to a prenyl group at δ H 5.39 (1H, m), 3.38 (2H, d, J = 7.3 Hz), and 1.74 and 1.72 (each 3H, br s), and two methines bearing an oxygen function at δ H 5.03 and 4.59 (each d, J = 11.9 Hz). Furthermore, two 1,3,4-trisubstituted aromatic rings were identi fied from six aromatic protons comprising two ABX-type spin-coupling systems at δ H 7.74 (d, J = 8.6 Hz), 6.64 (dd, J = 8.6, 2.2 Hz), and 6.41 (d, J = 2.2 Hz) and δ H 7.35 (d, J = 2.0 Hz), 7.27 (dd, J = 8.2, 2.0 Hz), and 6.90 (d, J = 8.2 Hz). The above data that 7 was a dihydroxy -flavan-3-ol derivative with a prenyl unit. The HMBC correlations between H-5 (δ H and C-4 (δ C 192.7)/C-9 (δ C 164.1), H-8 (δ H 6.41) and C-7 (δ C 165.4)/C-9, H-6 (δ H 6.64) and C-7, H-2′ (δ H 7.35) and C-2 (δ C 84.6)/C-4′ (δ C 155.8), H-6′ (δ H 7.27) and C-2/C-4′, and H-1′′ (δ H 3.38) and C-2′ (δ C 130.0)/C-3′ (δ C 128.1)/C-4′ hydroxy prenyl


Introduction
Peroxisome proliferator-activated receptor (PPAR)-γ is the primary molecular target for insulin-sensitizing thiazolidinedione drugs. These drugs activate PPAR-γ, increasing the number of small adipocytes that differentiate from preadipocytes and inducing apoptosis in large adipocytes. Because small adipocytes function normally, whereas large adipocytes hyperproduce and hypersecrete adipocytokines, an increased ratio of small adipocytes to large adipocytes improves insulin resistance. Therefore, compounds with PPAR-γ ligand-binding activity may be useful for the prevention and improvement of type 2 diabetes, a representative insulin resistance syndrome. We found that the EtOH extract of Glycyrrhiza glabra roots and the EtOAc extract of G. uralensis roots exhibited higher activity than did the other materials tested. Bioassay-guided fractionation of these extracts resulted in the isolation of 52 phenolics, including 11 novel ones [1,2].
In this chapter, we describe the results of the bioassay-guided fractionation of G. glabra and G. uralensis roots using a GAL-4-PPAR-γ chimera assay method.

PPAR-γ ligand-binding activity
PPAR-γ ligand-binding activity was assessed using a GAL-4-PPAR-γ chimera assay system (Figure 1) [3]. CV-1 monkey kidney cells from the American Type Culture Collection (ATCC) were suspended in Dulbecco's Modified Eagle medium (DMEM) containing 10% fetal bovine serum (FBS), 50 IU/mL Penicillin G sodium salt, 50 μg/mL streptomycin sulfate, and 37 mg/L ascorbic acid. The cells were then inoculated into a 96-well culture plate at 6 × 10 3 cells/well and incubated in 5% CO 2 /air at 37°C for 24 h. Cells were washed with OPTI-minimum essential medium (MEM) and pM-hPPAR-γ and p4 × UASg-tk-luc were transfected into cells using LipofectAMINE PLUS (Gibco). pM and p4 × UASg-tk-luc were transfected into CV-1 cells as a mock control. Twenty-four hours after transfection, the medium was changed to DMEM containing 10% charcoal-treated FBS [4] and the cells were further cultured for 24 h. The cells were then washed with phosphate-buffered saline containing Ca 2+ and Mg 2+ , and luciferase activity was measured using LucLite (Perkin-Elmer). Luminescence intensity was measured using a TopCount Microplate scintillation/luminescence counter. PPAR-γ ligand-binding activity was expressed as the relative luminescence intensity (test group/control group) determined for each sample.
In the same way, the structures of 2-4, 6, 9, and 10 were established as shown in Figure 3. Compounds 3, 4, and 9 showed neither specific rotation nor Cotton effects in their CD spectra, indicating that these compounds were racemates.

PPAR-γ ligand-binding activity of compounds 40-52 isolated from G. uralensis
Of the isolated compounds, the new compound 40 and known compounds 41-45 exhibited significant PPAR-γ ligand-binding activity (Figure 6). The activity of 40 at 5.0 μg/mL (=13.6 μM) was stronger than that of 2.0 μM TRG (relative luminescence intensity of 3.7). The coumestan derivative 46, which was less active than 40, was structurally similar to the active compound 43, and the only detected difference between 43 and 46 was the formation of a five-membered ether ring between C-4 and C-2′ in 46. This suggested that the presence of a hydroxy group at C-2′ in the isoflavan, isoflavene, or arylcoumarin skeleton is necessary for PPAR-γ ligand-binding activity. Furthermore, the isoflavones, 48 and 49, which have a hydroxy group at C-2′ and no isoprenyl group at C-6, did not exhibit activity, suggesting that the isoprenyl group at C-6 was also involved in PPAR-γ ligand-binding activity. In conclusion, the isoprenyl group at C-6 and the C-2′ hydroxy group in the aromatic C ring of the isoflavan, isoflavene, or arylcoumarin skeleton were structural requirements for PPAR-γ ligand-binding activity (Figure 9).

Ameliorative effects on diabetic KK-A y mice
The ameliorative effects of glycyrin (44) in KK-A y mice, an animal model of genetic type 2 diabetes, were studied using pioglitazone as a positive control. There was no difference in the food intake or body weight of mice between the treated groups and the control group. Test compound intake, calculated from the food intake and body weight of the mice, was approximately 100 mg/(kg day) in the glycyrin and glycyrol (46) groups and 23 mg/(kg day) in the pioglitazone group. Blood glucose levels significantly decreased after 4 days of feeding in both the glycyrin-and pioglitazone-treated groups compared to that in the control group, whereas the blood glucose levels of the glycyrol-treated group were comparable to those of the control group ( Table 1). Pioglitazone, a potent PPAR-γ agonist that activates PPAR-γ, resulted in the improvement of insulin resistance and type 2 diabetes mellitus. Glycyrin exhibited significant PPAR-γ ligandbinding activity and appeared to reduce the blood glucose levels of KK-A y mice by the same biological mechanism as pioglitazone. This finding was supported by the observation that glycyrol, structurally related to glycyrin but lacking PPAR-γ ligand-binding activity, failed to improve the hyperglycemia of KK-A y mice.

Conclusion
Fractionation of the EtOH extract of G. glabra roots and the EtOAc extract of G. uralensis roots, guided by a GAL-4-PPAR-γ chimera assay method, resulted in the isolation of 52 phenolics, including 11 new compounds. The structures of the new compounds were determined by spectroscopic analysis. Of the isolated compounds, more than 10 phenolics exhibited significant PPAR-γ ligand-binding activity and the prenylflavone derivative, licoflavanone A (31), exhibited the most potent ligand-binding activity. The activity of these compounds at a sample concentration of 10 μg/mL was approximately three times greater than that of 0.5 μM TRG. Six phenolics were isolated from the EtOAc extract of G. uralensis roots as PPAR-γ ligands and one, glycyrin (44), reduced the blood glucose levels of genetically diabetic KK-A y mice through its PPAR-γ ligand-binding activity. We have therefore discovered a possible new application of G. glabra and G. uralensis roots and their constituents for the amelioration of type 2 diabetes, a representative insulin resistance syndrome that is becoming a serious worldwide public health problem. Body weights and blood glucose levels are expressed as means ± SE of five mice.
b Calculated as (total food intake) (number of mice day).
c Calculated as (average food intake/average body weight of mice).
Statistical significance is indicated as ** (P < 0.01) as determined by Dunnett's multiple comparison test.