Evaluation of the Antihyperuricemic Activity of Phytochemicals from Davallia formosana by Enzyme Assay and Hyperuricemic Mice Model

Abnormal serum urate levels are recognized as a critical factor in the progression of several chronic diseases. To evaluate the antihyperuricemic effect of Davallia formosana, the inhibitory activities of 15 isolated phytochemicals, including five novel compounds of 6,8-dihydroxychromone-7-C-β-d-glucopyranoside (1), 6,8,3′,4′-tetrahydroxyflavanone-7-C-β-d-glucopyranoside (2), 6,8,4′-trihydroxyflavanone-7-C-β-d-glucopyranoside (3), 8-(2-pyrrolidinone-5-yl)-catechin-3-O-β-d-allopyranoside (4), and epiphyllocoumarin-3-O-β-d-allopyranoside (5), were examined against xanthine oxidase (XOD) and in a potassium oxonate-(PTO-) induced acute hyperuricemic mice model. The results indicated that compounds 3 and 5 significantly inhibited XOD activity in vitro and reduced serum uric acid levels in vivo. This is the first report providing new insights into the antihyperuricemic activities of flavonoid glycosides which can possibly be developed into potential hypouricemic agents.


Introduction
Hyperuricemia means high levels of uric acid in the blood, a condition considered to be closely associated with increased risks for developing gout, cardiovascular diseases, hypertension, and metabolic syndrome [1,2]. Xanthine oxidase (XOD) is an important enzyme responsible for the catabolism of purines in humans; it oxidizes hypoxanthine into xanthine and then further forms uric acid [3,4]. Allopurinol is currently the most effective XOD inhibitor, which is used for treating hyperuricemia and gout by reducing circulating levels of uric acid and vascular oxidative stress [5]. However, serious side effects include skin rashes and allergic reactions that occur in some clinical patients [6,7].
Recently, several naturally occurring compounds were reported to inhibit XOD activity [8][9][10]. In particular, plant phenolic compounds, such as phenolic acids and flavonoids, exhibit strong antioxidant activities via scavenging free radicals. Moreover, many studies also indicated that both types of compounds obviously inhibited XOD activity [11][12][13]. Davallia formosana is a popular herbal medicine used to treat osteoporosis [14]. Several flavan-3-ols, triterpenoids, proanthocyanidins, and mericprocyanidins were isolated from the rhizome of D. formosana [15,16]. Our preliminary studies revealed that the crude extract of D. formosana rhizomes could inhibit XOD activity. Therefore, in this study, we investigated the constituents of D. formosana and their antihyperuricemic effects. An in vitro XOD-inhibitory assay and in vivo potassium oxonate-(PTO-) induced acute hyperuricemic mouse model were used to evaluate the uric acidlowering effects of compounds isolated from D. formosana.

Determination of XOD-Inhibitory Activity.
The inhibitory effect on XOD was determined spectrophotometrically [17]. The reaction mixture consisted of 100 L of 50 mM potassium phosphate buffer (pH 7.5), 50 L of 1.5 mM xanthine, 10 L of sample solution dissolved in dimethyl sulfoxide (DMSO), and 25 L XOD (0.05 U). The absorption increments at a UV absorbance of 295 nm indicated the formation of uric acid. All determinations were performed in triplicate. Pure compounds and allopurinol for the XO inhibitory activity assays were examined at concentrations of 0, 25, 50, and 100 M, respectively. The inhibitory activity of XOD was assessed as the inhibitory percent (%) = (1 − b/a) × 100, where "a" is the change in absorbance per minute without the sample, and "b" is the change in absorbance per minute with the sample.

Hypouricemic Effects Examined in Mice with PTO-Induced Hyperuricemia.
Six-week-old male ICR mice with body weights of about 30.0 g were purchased from BioLASCO (A Charles River Licensee Corp., Yilan, Taiwan). Before the experiments, mice were raised for 1 week to allow them to acclimate to the environment and diet. All mice were given a standard laboratory diet (no. 5001; PMI Nutrition International, Brentwood, MO) and distilled water ad libitum and kept on a 12 h light/dark cycle at 22 ± 2 ∘ C. This study was conducted according to institutional guidelines and was approved by the Institutional Animal Care and Utilization Committee (IACUC) of National Taiwan Sport University, Taoyuan, Taiwan. This study was approved by the IACUC ethics committee under protocol IACUC-10004.

Xanthine Oxidase Molecular
Docking. Models of compound 5 in complex with xanthine oxidase were generated through docking compound 5 to the active site of the X-ray crystal structure of bovine xanthine oxidase (PDB id: 3B9J) ( Figure 5). In order to predict the position of compound 5 in the active site, we implemented the docking program (GOLD Genetic Optimization for Ligand Docking) (Cambridge Crystallographic Data Center (CCDC), version 3.2) with the Goldscore scoring function. Before docking, the substrate 2-hydroxy-6-methylpurine and all water molecules were removed. The 3D structure of compound 5 was generated and optimized by energy minimization using Discovery Studio v.3.5 (Accelrys Software Inc., USA). GOLD was used to dock compound 5 into the proteins with the flexible docking option turned on. Initially, 500 independent genetic algorithm cycles of computation were carried out with ligand torsion angles varying between −180 ∘ and 180 ∘ . The search efficiency was set at 200% to ensure the most exhaustive search for the docking conformational space. All other parameters were kept the same as the default settings. Finally, from the 500 docking conformations of compound 5, the top one with the highest GOLD fitness score was chosen to explore the "inhibitor-bond" conformations in the xanthine oxidase active site using Goldscore within the GOLD program. The molecular models of compound 5 were displayed using the PyMOL software (http://www.pymol.org).

Statistical
Analysis. All data are expressed as the mean ± standard error of the mean (SEM). A one-way analysis of variance (ANOVA) was performed with Duncan's post hoc test for multisample testing.
< 0.05 was considered statistically significant. The HMBC correlation between glucopyranose H-1 and aglycone C-7 suggested that glucose was substituted at C-7 of the aglycone. The molecular weight of 3 lost 16 units compared to that of 2, which supported compound 3 having lost a hydroxyl group at position C-5 . Accordingly, the structure of compound 3 was assigned as 6,8,4 -trihydroxyflavanone-7-C--d-glucopyranoside.

Results and Discussion
Compound 4 was isolated as a yellow amorphous solid. The molecular formula was deduced to be C 25  Additionally, the HMBC correlation between H-3 and C-1 demonstrated that the allopyranose residue was linked to C-3 of the aglycone. These results indicated that compound 4 was similar to davallioside A and B [20]. Therefore, compound 4 was a catechin-3-O--d-allopyranoside with a -lactam substitution at C-8, and it was determined by 8-(2pyrrolidinone-5-yl)-catechin-3-O--d-allopyranoside.
Compound 5 was a white amorphous powder, and the molecular formula was determined to be C 24  activity in dose-dependent manners, compared to allopurinol ( Figure 3). The IC 50 of compounds 3 and 5 was 57.4 and 124.0 M, respectively, whereas there was no detectable effect for compound 2.

In Vivo Hypouricemic Effect Determined in Mice with PTO-Induced Hyperuricemia.
To further confirm the capabilities of compounds 2, 3, and 5 to reduce the uric acid level in vivo, a PTO-induced hyperuricemia mice model was investigated. After 3 h of PTO treatment, the level of serum uric acid had increased to 12 ± 0.14 mg/dL. As shown in Figure 4, PTO-induced serum uric acid levels were reduced by the three test compounds, as well as the reference (allopurinol). At the same concentration (100 mmol/kg), compounds 2, 3, 5, and allopurinol significantly reduced the level of serum uric acid by 33.9%, 41.7%, 46.0%, and 58.1%, respectively, compared to the PTO group ( < 0.005). XOD is an important purine metabolic enzyme, which is a significant target for developing antihyperuricemic drugs. The in vitro and in vivo results suggested that inhibition of XOD activity played an important role in the antihyperuricemic effects of compounds 3 and 5.

Computational Docking Studies of Compound 5.
In addition to its uric acid-lowering activity in vivo, compound 5 also inhibits xanthine oxidase activity in vitro. We were interested in visualizing the effects of compound 5 on XOD in order to gain insights into the observed activities. Many studies have shown that flavonoids can inhibit XOD activity via hydrogen bonding and hydrophobic interaction with key amino acid residues on XOD-catalyzed sites such as Arg880, Phe914, Phe1009, and Thr1010 [32]. To determine the preferred positions of binding sites on XOD for compound 5, the 3D model of interaction was analyzed by docking using bovine milk XOD (PDB id: 3B9J) [33].
As shown in Figure 5, the carbonyl group on the benzopyranone forms hydrogen bonds with the active sites, including Arg880 and Thr1010. Furthermore, the 5-hydroxyl group of compound 5 forms hydrogen bonds with Glu802. The docking results show again that the A-ring of compound 5 is sandwiched between Phe914 and Phe1009 and participates in the formation of aromatic interactions ( − effects) with the two phenylalanines. In addition, three hydrophobic interactions were observed involving the methylene and Pro1076, the 3 ,4 -dihydroxyphenyl moiety and Leu1014, and Val1011. It can be seen that binding residues are the same among [34], suggesting that Arg880 and Thr1010 might be of significance for the selective inhibition of XOD by compound 5.
This study obtained five new compounds from D. formosana, and two of them exhibited potent antihyperuricemic activity. Accordingly, our results can provide the scientific basis for development of antihyperuricemic drugs.