Advances in structures required of polyphenols for xanthine oxidase inhibition

Abstract

Polyphenols have been used as natural medicaments for the management of hyperuricemia for a long history. They have been attracted many interests because of the little side effects in curing hyperuricemia, which is an important advantage over the antihyperuricemic drugs. In this review, the structure–activity relationships for polyphenols as xanthine oxidase (XO) inhibitors were discussed. It is concluded that the presence of hydroxyl groups, which influences the inhibitory effects, is closely related to whether the substitutions increase the steric hindrance or disturb the interaction of flavonoid with the catalytic site of XO, and the increased size of the molecule after glycosylation may increase the steric hindrance between flavonoid and XO, and consequently reducing the competitive inhibition behaviors. However, there is no obtained simple general rule that can comprehensively describe the effects of structural alteration on the inhibition activity because the results are varied among different subclasses of polyphenols. In addition, the inhibition mechanisms are mainly assumed as polyphenol binding to the active site of XO and hindering the entrance of xanthine or the discharge of uric acid and diffusion of O²⁻ radical.

INTRODUCTION

Gout is caused by the deposition of monosodium urate crystals in and around the joints that come from stacked excessive uric acid due to purine metabolism disorder (Gliozzi, Malara, Muscoli, & Mollace, 2016). The prevalence of gout has been recently reported to be nearly 10%, and the incidence is almost six cases/1,000 person/year (Kuo, Grainge, Zhang, & Doherty, 2015). Because the uric acids in the body are mainly products of purine catalyzed by xanthine oxidase (XO), reducing the uric acid level by inhibition of XO activity has become a mainstream treatment of gout (Sharma & Ashraf, 2018). Common medicines for gout include allopurinol, febuxostat, and topiroxostat. Allopurinol is a purine-analogue inhibitor of XO, which can competitively react with XO to reduce the amount of purines being catalyzed to produce uric acids. However, allopurinol sometimes can cause adverse effects such as looseness, hepatitis, and interstitial nephritis, which extremely limits its use (Vargas-Santos, Peloquin, Zhang, & Neogi, 2018). Febuxostat also has a 2% chance to cause elevated liver enzyme levels, rashes, joint pain, and even higher cardiovascular risks for elder patients (Choi, Neogi, Stamp, Dalbeth, & Terkeltaub, 2018). Topiroxostat is generally regarded as the safest commercial gout medicine, but a recent study declared its majority of adverse effects were mild to moderate (Hosoya, Sasaki, & Ohashi, 2017). Hence, searching new XO inhibitors is still an appealing work for pharmaceutical companies and related researchers (El-Tantawy, 2019).

Natural products have been taken as an ideal source of bioactive compounds with specific pharmacological activities. Many enzyme inhibitors have been identified and isolated from the extracts of vegetables and herbs (Liu et al., 2018, 2019; Zhang et al., 2011). Various bioactive compounds, including polyphenols, saponins, terpenoids, phenylethanoid glycosides, and alkaloids, have been reported to be effective XO inhibitors (Amessis-Ouchemoukh et al., 2017; Mehmood et al., 2019; Shi, Chen, Bao, Zeng, & Cai, 2019; Zhu et al., 2017). Among them, polyphenols have attracted most of the attentions from researchers because of their huge amounts and diverse biological activities such as antioxidation, antitumor, anti-inflammatory, antiviral, antihypertensive, hypoglycemic, and enzyme inhibitory effects (Cao et al., 2019; Xie, Yang, Chen, & Xiao, 2014; Xie, Yang, Tang, Chen, & Ren, 2015; Zhao et al., 2018).

As a group of important secondary metabolites, polyphenols play an indispensable role in helping plants resist ultraviolet radiation and pathogen attack. They are ubiquitously found in vegetables and well known as health benefit components in human diet (Brglez Mojzer, Knez Hrnčič, Škerget, Knez, & Bren, 2016; Cao et al., 2019; Cao, Liu, Ulrih, Sengupta, & Xiao, 2019; Jinju et al., 2020). Dietary polyphenols are even used as a natural medicament for the management of hyperuricemia, because they are not reported to have any side effects in curing hyperuricemia, which is unlike available antihyperuricemic drugs (Mehmood et al., 2019). Polyphenol molecules usually have aromatic rings with hydroxyl groups or other substituents. They can be divided into different classes according to their chemical structures such as flavonoids, stilbenes, lignans, and so on (Xie & Chen, 2013). Among them, flavonoids are the most abundant ones that can be further classified according to the level of oxidation and pattern of substitution of the C-ring, as well as the linking way of B-ring to C-ring (as shown in Figure 1).

FIGURE 1. Skeletons of flavonoids

Because the properties of polyphenols are decided by their structures, the relationships between their structures and bioactivities have attracted great interests among researchers. Lin, Zhang, Liao, Pan, and Gong (2015) investigated the structure–activity relationships (SAR) of dietary flavonoids as XO inhibitors and indicated that the hydrophobic interaction played an important role in binding between flavonoids and XO, and the affinities were partially proportional to the XO inhibitory abilities for flavones and flavonols. However, another recent study showed that methylation and hydroxylation of flavonoids in the A-ring were going to cause opposite changes for their XO affinities and XO inhibitions (Yuan et al., 2019). Mathew et al. (2015) reviewed literatures about the XO inhibition of flavonoids and concluded a SAR overview of flavonoids toward XO inhibition. And it is worth mentioning that the conclusions of the review are very different from those of the previous mentioned studies. Hence, it is still a challenge and research hotspots to confirm the exact structural requirements of polyphenols for XO inhibition, and more importantly, to find the underlying mechanisms. The present review mainly cited and analyzed the related literatures of recent 5 years to tentatively summarize the structures required of polyphenols for XO inhibition and the advances on the inhibition mechanisms.

EXTRACTS RICH IN POLYPHENOLS

As so called “Diseases of King,” gout has long been treated with herbs in the medical practice of ancient times (Zychowicz, 2011). In modern pharmacy, searching potential XO inhibitors in plant has been naturally a rational way to develop potential drugs for gout treatment (Rogowska et al., 2017; Song et al., 2017). Ragoo et al. investigated the in vitro bioactivities of 12 endemic plants and correlated their XO inhibitory activities with the amounts of polyphenols (Ramhit, Ragoo, Bahorun, & Neergheen-Bhujun, 2018). Hou et al. verified the XO inhibitory activity of extract from the leaves of Perilla frutescens both in vitro and in vivo, and identified five potent inhibitors using a bioassay-guided isolation method, including caffeic acid, vinyl caffeate, rosmarinic acid, methyl rosmarinate, and apigenin (Huo et al., 2015). Apigenin also has been confirmed as the mainly responsible compound for the XO inhibitory effect of Selaginella moellendorffii Hieron extracts (Zhao, Chen, Zhang, Deng, & Li, 2017). Chen et al. also have found that green tea polyphenols markedly reduced XO activity in serum and liver of hyperuricemic mice, and consequently decreasing the uric acid production and increasing uric acid excretion (Chen, Tan, Li, Leung, & Ko, 2015). In our previous study, four compounds including corilagin, geraniin, ellagic acid, and myricetrin were identified as XO inhibitors that were screened from Geranium wilfordii, a commonly used traditional medicine (He et al., 2019). In addition, the in vitro XO inhibitory activities of the extracts of walnut (Wang et al., 2015), faba bean (Choudhary & Mishra, 2019), cocoa leaves (Irondi, Olanrewaju, Oboh, Olasupo, & Boligon, 2017), sardinian honeys (Di Petrillo et al., 2018), kelulut honey (Majid, Fadzelly Abu Bakar, & Mian, 2019), bee pollen (Wang et al., 2018), butterfly pea flower (Mehmood et al., 2019), and onion (Ouyang, Hou, Peng, Liu, & Deng, 2018) have been confirmed and mostly attributed to the abundant polyphenols in them. It must be admitted that few new active structures have been found in most of the recent investigated extracts. Nevertheless, it is still a reliable and useful route to search XO inhibitors in plant extracts.

FLAVONOIDS Flavone and flavonol

Flavones are the most common flavonoids being contacted and investigated, such as luteolin, apigenin, and baicalein. The flavone amounts have been related to the XO inhibitory activities of some herb extracts for treating gout (Pereira, Catarino, Afonso, Silva, & Cardoso, 2018; Song et al., 2018). When the site 3 on C-ring of flavone is substituted by hydroxyl group, it turns into flavonol. Quercetin, myricetin, and galangin are the representative flavonols consumed daily by human via plant-derived food and beverage (Cao, Jia, Shi, Xiao, & Chen, 2016). Both Liu et al. and Lin et al. investigated the effects of hydroxylation of flavonoids on their XO inhibitory activities. It has been concluded that hydroxyls at 5 and 7 sites of A-ring are favorable for XO inhibition, whereas hydroxylation on B-ring may increase or decrease the inhibition activities depending on the intrinsic substitution patterns (Lin et al., 2015; Yuan et al., 2019). How the presence of hydroxyl groups influences the inhibitory effects is closely related to whether the substitutions increase the steric hindrance or disturb the interaction of flavonoid with the catalytic site of XO (Mathew et al., 2015; Van Hoorn et al., 2002).

It has been reported that luteolin and baicalin inhibited XO in a concentration-dependent manner with the IC₅₀ values of 47.11 ± 4.80 and 55.58 ± 1.71 μM, respectively, compared with 27.21 ± 0.82 μM for allopurinol (Wang, Li, Yu, & Qi, 2017). Besides, quercetin has been reported to exhibit certain XO inhibitory activity in vitro (Huang, Wang, Zhu, Chen, & Zhu, 2011). But, its derivative isoquercitrin, with the 3-OH being substituted by glucoside, has shown a remarkable XO inhibition activity much better than allopurinol (IC₅₀: 1.60 vs. 50.51 μM) (Liu, Wang, Yang, & Meng, 2017). Similar changes also have been found for luteolin to luteolin-4′-O-glucoside and kaempferol-4′-glucoside to camaroside (dos Santos et al., 2018; Zhang, Zhang, Pan, & Gong, 2016)[28]. However, all of the 3-OH glycosylation products of myricetin have shown distinct decreases of the XO inhibitory activity (Nguyen et al., 2016). Besides, the galactoside of quercetin (hyperin) showed a lower XO inhibition ability than quercetin, and the inhibitions on XO of rutin and baicalin were much lower than their aglycones (Lin et al., 2015). The increased size of the molecule after glycosylation may increase the steric hindrance between flavonoid and XO, and consequently reducing the competitive inhibition behaviors (Minh et al., 2019). It has been indicated that flavones and flavonols inhibit XO mainly by hydrogen bonding interaction and occupying the active site of the enzyme (Ren et al., 2019).

Liu et al. (2018) isolated two compounds from extract of Clerodendranthus spicatus using XO-immobilized silica-coated Fe₃O₄ nanoparticles and identified them as gardenin B and eupatorin (seen in Figure 2). Both of them possessed multiple methoxyl groups and showed potent XO inhibitory effects in vitro with IC₅₀ values of 1.488 and 11.197 μg/mL, respectively. As shown in Table 1, methoxylation seems to be favorable to XO inhibition of flavonoids, but methylation of hydroxyls sometimes may decrease the inhibitory activity. The enhancement effect of methoxyl group was attributed to the enhanced binding affinity toward XO, whereas the decline was probably related to the block of the active site of XO (Minh et al., 2019).

FIGURE 2. Potent flavonoid XO inhibitors

TABLE 1 Effects of structural alterations on the inhibitory activity of flavones and flavonols against XO

Type	Position	Example	Effect	Reference
Glycosylation	3	Myricetin → 3-O-glucoside	↓↓	Nguyen et al., 2016
	3	Myricetin → myricetin-3-O-rhamnoside	↓↓	Nguyen et al., 2016
	3	Myricetin → myricetin-3-O-glucuronide	↓↓	Nguyen et al., 2016
	3	Myricetin → myricetin-3-O-(6″-galloyl) galactoside	↓↓	Nguyen et al., 2016
	6,8	Apigenin → apigenin 6-C-glucoside-8-C-xylpyranoside	↓↓	Zhao et al., 2017
	6,8	Apigenin → apigenin 6-C-glucoside-8-C-arabinoside	↑	Materska, 2015
	3	Quercetin → isoquercitrin	↑↑	Liu et al., 2017
	4′	Quercetin → quercetin-4′-O-monoglucoside	↓	Nile, Nile, Keum, & Sharma, 2017
	3,4′	Quercetin → quercetin-3,4′-O-diglucoside	↓	Nile et al., 2017
	4′	Luteolin → luteolin -4′-O-glucoside	↑	Zhang et al., 2016
	6	Luteolin → luteolin -6-C-glucoside	↓	Materska, 2015
	6,8	Luteolin → luteolin -6-C-glucoside	↓↓	Materska, 2015
	3	Quercetin → hyperin	↓↓	Lin et al., 2015
	3	Quercetin → rutin	↓↓	Lin et al., 2015
	7	Bacalein → baicalin	↓↓	Lin et al., 2015
	3,7	Kaempferol → kaempferitrin	↓	Yuan et al., 2019
Methylation	3	Galangin → 3-methoxyl-galangin	↓↓	Santos et al., 2018
	4′	Kaempferol → kaempferide	↓	Yuan et al., 2019
	4′,7	Quercetin → ombuin	↑	Ranjana et al., 2019
	7	Prudomestin → tambulin	↑	Ranjana et al., 2019
	6	Apigenin → hispidulin	↓	Yuan et al., 2019
Methoxylation	4′	Galangin → kaempferide	↑	Santos et al., 2018
	6	Genistein → tectorigenin	↑	Yuan et al., 2019
	8	Chrysin → wogonin	↑	Yuan et al., 2019
	3′	Tangeretin → nobiletin	↑	Yuan et al., 2019
	4′	Galangin → kaempferide	↓	Yuan et al., 2019

Note. ↑/↑↑: enhanced after change; ↓/↓↓: declined after change.

In addition to the substitutions of groups containing carbon or oxygen atom, flavonoids with substituents containing nitrogen or sulfur have depicted potent XO inhibitory activities. A luteolin derivative (Figure 2) with unique 1,4-thiazine ring unit on the B-ring was prepared by an oxidative coupling reaction with a cysteine ester, which showed 4.5 times more strongly XO inhibition ability than luteolin (Masuda, Nojima, Miura, Honda, & Masuda, 2015). Besides, the hydrazine- and aniline-substituted derivatives of rutin, especially Rua₃ (Figure 2), also showed remarkable inhibitions on XO even much stronger than allopurinol did (IC₅₀ value is only half of that of allopurinol). The two aniline derivatives of rutin also showed XO inhibitory effects as potent as allopurinol did (Malik, Dhiman, & Khatkar, 2019).

Flavanone and flavanonol

Li et al. screened and identified pinocembrin, a flavanone with 5,7-dihyrxoyls, from the roots of Lindera reflexa Hemsl. It showed a XO inhibitory effect almost as potent as febuxostat (IC₅₀: 1.093 vs. 0.93 μM) (Fu et al., 2019). When the 7-OH of pinocembrin was substituted by 7-O-β-D-glucopyranoside, the inhibition ability decreased dramatically. In addition, the 3-OH substitution gave pinobanksin rather stronger inhibition activity than pinocembrin (Dong et al., 2016). In addition, just like rutin, the synthesized aniline-substituted derivatives of hesperidin also showed much stronger inhibitory activities against XO than allopurinol did (IC₅₀ value is only 1/10 of that of allopurinol).

Flavan-3-ols

It has been suggested that a planar structure is a requirement in flavonoid compounds that present XO inhibition activity (Silva et al., 2004). However, tea extracts rich in flavanols exhibited uric acid lowering effect, which was attributed to the modulation of both XO and urate excretion (Peluso, Teichner, Manafikhi, & Palmery, 2017). It has been suggested that the pu-erh tea extracts can attenuate hyperuricemia through XO and renal urate transporters in hyperuricemic mice (Zhao, Chen, & Wu, 2017). The green tea catechins also decreased XO activity in washed cream, and (-)-epigallocatechin gallate (EGCG) was the most potent one among the tea catechins (Rashidinejad, Birch, & Everett, 2016).

Isoflavone

Isoflavones are ubiquitously found in soy products and soybeans and have been reported to be XO inhibitors (Pyo, Hwang, & Seong, 2018). The structures of some isoflavone XO inhibitors are shown in Figure 3. Liu et al. screened four XO inhibitors from Puerariae flos extract with centrifugal ultrafiltration coupled with HPLC-MS, which were identified as tectoridin, daidzin, ononin, and biochanin A (Liu et al., 2018). Another three isoflavonoids, 2′-hydroxygenistein, 3′-methoxygenistein, and lupinalbin A, were isolated from the ethyl acetate fraction of Apios americana on the catalytic reaction mediated by XO (Kim & Jin, 2019). They exhibited potential inhibitory activity with the IC₅₀ value of 21.8 ± 0.7, 31.6 ± 1.1, and 38.8 ± 3.5 μg/mL, respectively, which indicated that the catechol groups in the B-ring (2′-hydroxygenistein) were favorable for XO inhibition. In addition, it has been suggested that both the methoxylation of 6-H (daidzein → tectorigenin) and the methylation of 4′-OH (daidzein → formononetin; genistein → biochanin A) decreased the inhibition activities (Yuan et al., 2019). The inhibition mechanism has been attributed to the insertion of isoflavone into the active site of XO and occupancy of catalytic center, then consequently reducing the landing and oxidation of substrate (Lin, Zhang, Pan, & Gong, 2015).

FIGURE 3. Isoflavone derivatives inhibitors

Chalcone

Chalcones are the intermediates in flavonoid biosynthesis without appreciable accumulations in most plants, which have been reported to possess many heath beneficial biological effects (Rozmer & Perjési, 2016). Both natural and synthetic chalcones have been reported to be potential XO inhibitors (Gomes et al., 2017; Hofmann et al., 2016). As shown in Table 2, the hydroxylation on both A-ring and B-ring is favorable to the XO inhibition, especially the catechol groups on the 3,4 site and the meta hydroxyl groups at 2′,4′ site. The methoxylation of hydroxyl groups seems to decrease the activity when it happens on the A-ring, but to increase the activity when it happens on the B-ring. It is worth noting that an additional caffeoyl substitution at 3′ remarkedly increases the XO inhibition activity of the corresponding chalcone derivative (7.68 to 0.96). In addition, chalcones can form heterodimers with each other or with other flavonoids. Yang et al. isolated a chalcone–flavonone heterodimer, termipaniculatone A from Terminthia paniculata, and proved its antihyperuricemic effects through decreasing serum uric acid levels and inhibiting XO activity in both serum and liver (Yang et al., 2019). Bermaglu et al. synthesized eight bis-chalcone derivatives by Claisen–Schmidt condensation reaction. Among them, the ones with fluoro group at the 2 or 2,5-position of B-ring were found to be potent XO inhibitors with sevenfold higher inhibitory effects than allopurinol (Burmaoglu et al., 2019).

TABLE 2 XO inhibitory activities of some chalcones

	Substitutions
[IMAGE OMITTED. SEE PDF.]	OH	OCH3	Others	XO inhibition ability indexa	Reference
Butein	2′,4′,3,4			2.66	Hou et al., 2019
2′,4′-Dimethoxy-4,5′,6′-trihydroxychalcone	4,5′,6′	2′,4′		0.10	Liu et al., 2019
2,2′,4′-Trihydroxychalcone	2,2′,4′			2.90	Xie et al., 2017
2′,4′-Dihydroxychalcone	2′,4′			5.46	Xie et al., 2017
4,2′,4′-Trihydroxychalcone	4,2′,4′			2.41	Xie et al., 2017
4,4′-Trihydroxychalcone	4,4′			5.97	Xie et al., 2017
2,2′-Dihydroxychalcone	2,2′			4.46	Xie et al., 2017
2′,4-Dihydroxychalcone	2′,4			4.05	Xie et al., 2017
2′,4′,3,4-Tetrahydroxychalcone	2′,4′,3,4			1.72	Bui, Nguyen, Dang, Nguyen, & Nguyen, 2016
4′-Dimethoxy-3,4,2′-trihydroxychalcone	2′,3,4,	4′		7.68	Bui et al., 2016
3ʹ-Caffeoyl-3,4,2ʹ-trihydroxy-4ʹ-methoxychalcone	3,4,2ʹ	4′	3ʹ-Caffeoyl	0.96	Bui et al., 2016
4-Dimethoxy-3,4′,2′-trihydroxychalcone	3,4′,2′	4		14.68	Bui et al., 2016
2′-Dimethoxy-3,4,4′-trihydroxychalcone	4′,3,4	2′		1.56	Bui et al., 2016
2′,4′-Dimethoxy-3,4-dihydroxychalcone	3,4	2′,4′		1.00	Bui et al., 2016
2′,4′,2,4-Tetrahydroxychalcone	2′,4′,2,4			6.52	Bui et al., 2016
2′,4′,2-Trihydroxy-4-methoxychalcone	2′,4′,2	4		8.72	Bui et al., 2016
3,4-Dihydroxychalcone	3,4			16.36	Bui et al., 2016
3-Hydroxy,4-methoxychalcone	3	4		Weak	Bui et al., 2016
3,4-Dimethoxychalcone		3,4		Weak	Bui et al., 2016

a

The value equals to IC₅₀ of chalcone/IC₅₀ of allopurinol (μM/μM in the same test, smaller value means stronger activity).

Aurone and anthraquinone

Liu et al. isolated an Z type aurone derivative (Z)-4,6-dimethoxy-7,4′-dihydroxyaurone (as shown in Figure 4) from the leaves of Perilla frutescens, but found no apparent inhibitory activity against XO (Liu et al., 2019). However, the carboxylated aurone derivatives have been shown to be potent XO inhibitors. The carboxylic acid group at the 4′-position of B-ring gave aurones good XO inhibitory activities with IC₅₀ values lower than sulfuretin but little higher than febuxostat, whereas the A-ring-modified aurones with carboxymethoxy group at the 6-position showed much weaker inhibitory activities (Muzychka, Kobzar, Popova, Frasinyuk, & Vovk, 2017). Minh et al. isolated two emodin derivatives from the ethyl acetate extract of the root of Rumex crispus L. The two compounds, chrysophanol and physcion, showed strong inhibition against the activity of XO with IC₅₀ values of 36.4 and 45.0 μg/mL, respectively (Minh et al., 2019). In another study, however, physcion, the emodin 3-methyl ether, was found inactive for XO inhibition, whereas emodin showed limited inhibitory effect on XO (Molee et al., 2018). The difference between the two assays may be attributed to the different materials and specific procedures they employed.

FIGURE 4. Aurone and anthraquinone inhibitors

Xanthone and coumarin

The XO inhibitory abilities of xanthones have been previously studied and the presence of a cyano group at the para position of benzyl moiety was preferred substitution pattern (Hu et al., 2011). Recently, a natural xanthone, norathyriol (Figure 5), has been found to have dual actions as hypouricemic agent, including inhibiting XO in a uncompetitive manner and accelerating the excretion of uric acid (Lin et al., 2019). It has also been suggested that hydroxylation at the 1-, 3-, and 6-positions is important for the XO inhibition activity of norathyriol (Niu et al., 2016). Thongchai et al. have investigated the influence of allylic substitutions on the XO inhibition activities of xanthone. It has been revealed that 2,4-diallyl-1,3-dihydroxythioxanthone showed the most potent inhibitory activity on XO with IC₅₀ value of 0.69 μM (positive control allopurinol, 0.12 μM), and the hydroxyl groups at C-1 and C-3 positions seemed crucial for the inhibition because of the comparably high activity of 2,4-diallyl-1,3-dihydroxyxanthone (0.76 μM) (Khammee, Jongsu, Kuno, & Suksanrarn, 2018). Fais et al. have evaluated 28 coumarin derivatives for their XO inhibitory activities, and found that 5,7-dihydroxy-3-(3′-hydroxyphenyl) coumarin had sevenfold higher inhibition effect on XO than that of allopurinol. In addition, the compound seemed to bind with the residues different from the catalytic site of enzyme, indicating an uncompetitive inhibition mode (Fais et al., 2018).

FIGURE 5. Xanthone and coumarin inhibitors

Anthocyanin and proanthocyanin

It has been reported that the extracts of Aronia melanocarpa enriched in procyanidins showed moderate XO inhibitory activities. And cyanidin-3-arabinoside (Figure 6) exhibited the most potent XO inhibitory effects among the isolated cyanidins (Bräunlich et al., 2013). Kirakosyan et al. investigated the XO inhibitory potential of the compounds from the montmorency tart cherry, and found that cyanidin 3-rutinoside showed a slightly higher XO inhibition activity than cyanidin 3-glucoside did (Kirakosyan, Gutierrez, Ramos Solano, Seymour, & Bolling, 2018). Besides, cyanidin-3-sambubioside and cyanidin-3-glucoside have shown a little weaker XO inhibition activities than their aglycone cyanidin (Wang et al., 2017). Huang et al. investigated the antioxidant functions of the main anthocyanins of blueberries in endothelial cells and found that both malvidin-3-glcoside and malvidin-3-galactoside exhibited greater inhibitory effects than malvidin, and malvidin-3-glcoside showed stronger effect than malvidin-3-galactoside (Wu-Yang, Ya-Mei, Jian, Xing-Na, & Chun-Yang, 2014). In addition, the procyanidin B1 in seed extracts from Washingtonia filifera palm fruit has been proved to have a good XO inhibitory ability with an IC₅₀ value of 53.5 mg/mL (Floris et al., 2019). Moreover, the proanthocyanin A2 isolated from longan flowers showed a significant inhibition on XO activity in vitro (Sheu et al., 2016).

FIGURE 6. Anthocyanin and proanthocyanin inhibitors

STILBENE

Stilbenes are a class of polyphenols that have a structural core of two benzene rings linked by a vinyl group (as shown in Figure 7). In our previous study, resveratrol and its three derivatives have been screened out, and XO inhibitory effects have been identified and evaluated. According to the results, methoxylation at 3-position decreased the inhibition ability of resveratrol, whereas glycosylation of the 3-hydroxyl group markedly increased the inhibition activity (Tang, Tang, Ma, & Liu, 2019). Methylation of the 3-hydroxyl group also has been suggested to decrease the inhibition activity of pinosylvin (Fu et al., 2019).

FIGURE 7. Stilbene inhibitors

PHENOLIC ACIDS

The herb extracts rich in phenolic acids such as rosmarinic, chlorogenic, caffeic, and ferulic acid have recently been proved to have potent inhibitory activities against XO (Choi, Park, Song, Yoon, & Cho, 2018; Gavarić et al., 2015; Liu et al., 2017). Dziki et al. have confirmed that the XO inhibitory activities of green coffee bean and whole meal wheat flour were attributed to chlorogenic acids and ferulic acids, respectively. And a synergistic effect was found when the two extracts were used in combination (Gawlik-Dziki, Dziki, Świeca, & Nowak, 2017). The hydroxyl groups of phenolic acids are often naturally substituted by other groups, which can markedly influence the XO inhibition ability. As shown in Table 3, methoxylation of the hydroxyl group of carboxyls in rosmarinic acid increased the XO inhibition activity by threefold. The vinyl substitution increased the XO inhibition activity of caffeic acid by about fourfold, and the benzene ring substituent also dramatically increases the activity of caffeic acid by about 20-fold. In addition, the alkylation of the hydroxyl group of carboxyls in caffeic acid can dramatically increase the XO inhibition activity, and the enhanced degree increases with the increasing alkyl length. However, gallic acid shows much lower XO inhibition activity than pyrogallol dose, and the additional methyls on the benzene ring will dramatically decrease the activity. But, an additional conjugate seven-membered oxy heterocycle(purpurogallin) has been found to enhance the XO inhibition activity of pyrogallol by eightfold (Honda, Fukuyama, Nishiwaki, Masuda, & Masuda, 2017).

TABLE 3 Effects of structural alterations on the inhibitory activity of phenolic acids against XO

Compounds	Structural alteration(site)	XO inhibition	Reference
Caffeic acid → vinyl caffeate	H → [IMAGE OMITTED. SEE PDF.] (carboxyl)	≈ 4-fold ↑	Huo et al., 2015
Caffeic acid → methyl caffeate	H → CH3 (carboxyl)	≈ 2-fold ↑	Masuoka & Kubo, 2018
Caffeic acid → hexyl caffeate	H → n-C6H13 (carboxyl)	≈ 10-fold ↑	Masuoka & Kubo, 2018
Caffeic acid → decyl caffeate	H → n-C10H21 (carboxyl)	≈ 20-fold ↑	Masuoka & Kubo, 2018
Caffeic acid → (E)-phenyl 3-(3,4-dihydroxyphenyl)acrylate	H → [IMAGE OMITTED. SEE PDF.] (carboxyl)	≈ 20-fold ↑	Choi et al., 2019
Rosmarinic acid → methyl rosmarinate	OH → OCH3 (carboxyl)	≈ 3-fold ↑	Huo et al., 2015
Pyrogallol → gallic acid	H → COOH (5)	≈ 6-fold ↓	Honda & Masuda, 2016; Ranjana et al., 2019
Pyrogallol → 4-methylpyrogallol	H → CH3 (4)	≈ 20-fold ↓	Honda et al., 2017
Pyrogallol → 5-methylpyrogallol	H → CH3 (5)	> 125-fold↓	Honda et al., 2017

The mono phenolic acids can be linked by sugar units to form dimers called phenylethanoid glycosides. Verbascoside (seen in Figure 8), a typical phenylethanoid glycoside, has been demonstrated to have potent XO inhibitory activity in vitro, which was attributed to the destroyed hydrogen-bond network of XO and changed conformation induced by verbascoside binding to molybdopterin domain of XO (Wan et al., 2016). Liu et al. (2017) isolated and identified three phenolic glycosides, and evaluated their XO inhibition effects in vitro. It was found that brachyanin F and another phenolic glycosides showed potent inhibitory activities on XO, whereas brachyanin C with an additional tetrahydrofuran ring showed no XO inhibition activity, indicating that the oxidation of phenolic structure would increase the XO inhibitory effect.

FIGURE 8. Phenolic acid inhibitors

BIOTRANSFORMATION PRODUCTS OF POLYPHENOLS

The biotransformation of polyphenols usually took place not only in the procedure of preparing food but also in the digestion process (Cao, Chen, Jassbi, & Xiao, 2015; Cianciosi et al., 2020). Hence, evaluating the bioactivities of the biotransformation products of polyphenols is important for understanding the practical bioactivities of polyphenols consumed by human being. For the polyphenols in cooking, the most important factor is temperature, which may induce the thermal reactions. During roasting coffee beans, phenylindanes were produced from the thermal reaction of chlorogenic and caffeic acids. The phenylindanes have been revealed to show potent XO inhibitory activities that were attributed to the XO inhibition-related functions of roasted coffee beverages (Fukuyama, Hidaka, Masuda, & Masuda, 2018). It has also been found that different products were produced when the coffee was roasted in different temperatures, and the thermal reaction products of caffeic acid being produced via the intermediate 4-vinylcatechol at 140–170°C have showed the most strong XO inhibitory activities among the other types of oligomers produced at other temperatures (Masuda et al., 2019). In addition, Mohos et al. have investigated the products of quercetin metabolized by human enzymes and the colonic microflora, and found that the sulfate and methyl conjugates inhibited XO activity with an approximately 10-fold stronger ability of allopurinol, which was also higher than quercetin did (Mohos et al., 2019).

QUANTITATIVE STRUCTURAL–ACTIVITY RELATIONSHIP

Quantitative structural–activity relationship (QSAR) is a powerful method to study the relationships between structure of chemical substances like polyphenols and their enzymic inhibitions to obtain a reliable statistical model for prediction of the activities of new chemical entities or structures (Mahfoudi, Djeridane, Benarous, Gaydou, & Yousfi, 2017; Verma, Khedkar, & Coutinho, 2010). It has been applied for decades to explore the structural requirements of polyphenols for their bioactivities using numerous readily computable descriptors in combination with observation techniques (Hamzeh-Mivehroud, Rahmani, Rashidi, & Dastmalchi, 2016; Hamzeh-Mivehroud, Rahmani, Rashidi, Hosseinpour Feizi, & Dastmalchi, 2013). Muthuswamy et al. (2011) have contributed both in silico docking studies and in vitro XO inhibitory activity of commercially available 10 flavonoids and polyphenolic compounds. They found that all the 10 compounds exhibit lower binding energy (–8.08 to –4.48 kcal/mol) than allopurinol (–4.47 kcal/mol). And the XO inhibitory activity of flavonoids was greater than polyphenolic compounds, which might be attributed to the presence of benzopyran ring in the flavonoids. Choudhary et al. have made a molecular docking study on catechin, epicatechin, gallic acid, and ellagic acid in faba bean. It revealed that it binds effectively with XO by binding energy of –7.78, –6.11, –6.39, and –5.78 kcal/mol, respectively, compared to allopurinol drug having binding energy of –4.94 kcal/mol. The four polyphenols and allopurinol bind other than catalytic residues (Glu-1261) of XO (Choudhary & Mishra, 2019). After the isolation, Yang et al. (2019) also conducted a molecular modeling study of termipaniculatone A. It revealed that termipaniculatone A was well located into the active site of XO by interacting with Glu802, Arg880, Thr1010, and Val1011 residues. Accordingly, QSAR modeling analysis would help to understand how chemical structure relates to the XO inhibition with some credible predictions.

ADVANCES IN INHIBITION MECHANISMS

The mechanism study usually depends on the kinetic and thermodynamics of the inhibition process, characterization of the conformation change of XO, and molecular docking study for simulating the interaction between the enzyme and substrate (Xiao, Kai, Yamamoto, & Chen, 2013; Xiao, Ni, Kai, & Chen, 2013; Xiao, Ni, Kai, & Chen, 2015). Generally, the polyphenols has been indicated to inhibit XO in a competitive manner through binding to the active site of the enzyme, of which the inhibiting activities are directly proportional to their concentrations (de Araújo et al., 2017; Ren et al., 2019). Besides, it has also been revealed that some polyphenols inhibit XO in a noncompetitive or mixed way, which means that the molecular interactions may occur both in the active site of XO and the region around the molybdopterin cofactor (Honda et al., 2017). As shown in Table 4, most of the flavonoids but butein have been reported to be reversible XO inhibitor whose inhibition activity increases with the increasing dosage (Lin, Zhang, Liao, & Pan, 2015). The flavonoids usually spontaneously bind to the active site and induce the conformational change of XO, consequently blocking the entrance of xanthine or the diffusion of O²⁻ radical out of the catalytic pocket. Besides, some phenolic acids have been suggested to be noncompetitive XO inhibitors, of which the detail mechanisms need further investigations (Gawlik-Dziki et al., 2017).

TABLE 4 The mechanisms of flavonoids inhibiting XO activity

Comp.	Activity	Inhibition type	Binding forces	Binding sites	Reference
Chrysin	0.43	Competitive manner; conformational change; avoid the entrance of xanthine	Hydrogen bonds; van der Waals forces	MO-Pt domain (the active cavity of XO);	Lin et al., 2015
Langin	1.26	Competitive manner; conformational change; inhibition of O2− generation	Hydrogen bond; hydrophobic interaction	Molybdenum atomic domain	Zhang et al., 2016
Baicalein	2.98	Mixed type manner; conformational change; block the diffusion of O2− radical	Hydrogen bond (mainly)	Flavin adenine dinucleotides (FAD) center	Zeng et al., 2018
Baicalin	0.49	Mixed type manner; conformational change; block the diffusion of O2− radical	Hydrogen bond; hydrophobic interaction	FAD center	Zeng et al., 2018
Quercetin	1.02	Mixed-type manner; block the diffusion of O2− radical	Hydrogen bonds; van der Waals forces	FAD center	Zhang, Wang, Zhang, & Gong, 2018
Fisetin	1.55	Mixed-type manner; avoid the entrance of xanthine	Hydrogen bond; hydrophobic interaction	FAD center	Zeng, Zhang, Hu, Pan, & Gong, 2019
Genistein	0.49	Competitive manner; conformational change; avoid the entrance of xanthine	Hydrogen bond; hydrophobic interaction	Mo center	Lin et al., 2015
Butein	2.66	Irreversible manner; conformational change; avoid the entrance of xanthine	Hydrogen bond; hydrophobic interaction	Molybdenum atomic (Mo) domain	Hou et al., 2019

Note. Activity equals to the value of dividing IC₅₀ of the compound by IC₅₀ of allopurinol and smaller value means stronger inhibition activity.

CONCLUSIONS

Within this review, the SARs for polyphenols as XO inhibitors were discussed. It is concluded that the presence of hydroxyl groups, which influences the inhibitory effects, is closely related to whether the substitutions increase the steric hindrance or disturb the interaction of flavonoid with the catalytic site of XO, and the increased size of the molecule after glycosylation may increase the steric hindrance between flavonoid and XO, and consequently reducing the competitive inhibition behaviors. However, there is no obtained simple general rule that can comprehensively describe the effects of structural alteration on the inhibition activity because the results are varied among different subclasses of polyphenols. In addition, the inhibition mechanisms are mainly assumed as polyphenol binding to the active site of XO and hindering the entrance of xanthine or the discharge of uric acid and diffusion of O²⁻ radical. Although the current knowledge on the inhibitory activity of polyphenols on XO is still not clear, the application potentials of polyphenols for attenuating the gout symptoms are widely accepted. However, it is noted that most of the assays were tested in vitro, and only few studies have investigated the in vivo activity. Further studies should pay more attentions on the relationships between bioavailability and inhibition activity. The possibility of adverse effects of polyphenols should also be evaluated in advance. Nevertheless, polyphenols have surely become a “Sword of victory” for conquering the “King of Diseases” (Singh, 2016).

ACKNOWLEDGMENTS

The authors thank for the financial support from the National Science Foundation of China (No. 31701613) and the Central Public-interest Scientific Institution Basal Research Fund (No. Y2019PT22-02).

CONFLICT OF INTEREST

The authors declare no conflict of interest.