Discovery of multi-target anti-gout agents from Eurycoma longifolia Jack through phenotypic screening and structural optimization

Abstract

Developing anti-gout medications that simultaneously reduce uric acid and exert anti-inflammatory effects represents a critical breakthrough for managing gout progression. Natural products with polypharmacological properties offer promising leads for drug discovery. In this study, β-carboline-1-propionic acid, a bioactive constituent of Eurycoma longifolia Jack, served as the starting point for drug design. Guided by a dual-target pharmacophore model, we design and synthesize 64 derivatives. Through systematic screening, 32 emerges as a drug candidate, demonstrating potent uric acid-lowering activity in male hyperuricemia mouse models (efficacy comparable to febuxostat and superior to lesinurad and benzbromarone) by inhibiting key urate transporters. In a male rat model of acute gouty arthritis, 32 mitigates NOD-like receptor protein 3 inflammasome-mediated inflammation. Notably, 32 exhibits enhanced safety compared to control drugs. This study exemplifies a natural product-inspired, dual-mechanism drug discovery approach, showcasing the potential of a rational polypharmacology and thus offering therapeutic opportunities for gout management.

Developing anti-gout medications that simultaneously reduce uric acid and exert anti-inflammatory effects is of interest for managing gout progression. Here, the authors employ a dual-target pharmacophore model to design derivatives of β-carboline-1-propionic acid and identify a drug candidate demonstrating potent uric acid-lowering activity in male hyperuricemia mouse models and mitigating NLRP3-mediated inflammation.

Introduction

Gout is a chronic disease characterized by the deposition of monosodium urate (MSU) crystals in the joints or surrounding tissues, which triggers an immune response mediated by IL-1β via the NOD-like receptor protein 3 (NLRP3) inflammasome¹. Hyperuricemia (HUA), a fundamental pathological basis for gout, primarily arises from purine metabolic disorders or inadequate serum uric acid (SUA) excretion, making it the fourth most prevalent basal metabolic disease following hypertension, hyperlipidemia and hyperglycemia^2,3. Numerous studies have demonstrated an association between gout, HUA, and various cardiovascular and cerebrovascular diseases, metabolic syndrome, fatty liver, chronic kidney disease, and diabetes^4,5. Therefore, the pathogenesis and selection of therapeutic drugs have emerged as the central focus of medical research due to the increasing significance of HUA and gout in posing threats to human health.

The current approach to anti-gout involves the utilization of pharmaceutical interventions to effectively lower SUA levels and control inflammatory response⁶. Among these interventions, long-term urate-lowering therapy leads to the dissolution of MSU crystals, ultimately resulting in the prevention of gout flares or tophi and in improving quality of life⁷. The SUA-lowering small molecule medications can be classified into two main groups: xanthine oxidase inhibitors (XOIs) that inhibit uric acid formation and urate transporter 1 (URAT1) inhibitors that promote uric acid excretion^{8, 9, 10–11}. Nevertheless, clinically utilized XOIs, including febuxostat (a) and allopurinol (b) (Fig. 1a), have been associated with significant adverse effects, while demonstrating limited efficacy in the treatment of acute gouty arthritis^12,13. And currently approved URAT1 inhibitors also have limitations that hinder their widespread use: benzbromarone (c) can cause fatal fulminant hepatitis, while lesinurad (d) has a black-box warning of severe nephrotoxicity and is no longer commercially available (Fig. 1a)^{14, 15–16}. On the other hand, the anti-gout drugs used to control the inflammatory response in clinic practice are mainly traditional anti-inflammatory drugs, such as colchicine (e) and indomethacin (f). Although they can rapidly control joint inflammation and alleviate pain; however, the relevance of their targets is relatively weak and prolonged usage may lead to tolerance. With the continuous investigation into the pathogenesis of gout, research has demonstrated a significant correlation between the initiation of primary gout and polymorphism in the NLRP3 gene during its treatment. The increased expression of NLRP3 in peripheral blood mononuclear cells obtained from individuals with acute gouty arthritis indicates the potential role of the NLRP3 inflammasome in the development of this condition^17,18. Furthermore, it is noteworthy that patients with acute gouty arthritis frequently exhibit concurrent HUA. Therefore, targeting NLRP3 directly can effectively inhibit the upstream of the inflammatory cascade, leading to a significant reduction in the production of inflammatory mediators and effective control over both acute onset and chronic progression of gout. However, currently there is no commercially available NLRP3 inhibitor on the market, and clinical trials of MCC950 (g), an NLRP3 inhibitor with a well-defined mechanism of action and excellent in vitro and in vivo anti-inflammatory activity, have been discontinued due to hepatotoxicity¹⁹. In summary, the limitations of simple uric acid-lowering or anti-inflammatory therapy for the treatment of gout are evident and easily discernible. Moreover, the concomitant use of existing urate-lowering drugs and anti-inflammatory drugs often leads to complex pharmacokinetic and pharmacodynamic interactions, potentially resulting in cumulative toxicity and adverse effects. To address this challenge, we propose two potential strategies: (1) developing improved derivatives of current drugs with better safety profiles through structural optimization; (2) designing single-molecule dual-functional agents‌ that concurrently lower uric acid and suppress inflammation. Natural products with multi-target activities may be particularly suitable for this purpose, as they often exhibit inherent polypharmacological properties that could be optimized for dual therapeutic effects²⁰. Consequently, exploring the skeleton of natural products exhibiting uric acid-lowering activity and utilizing them as lead compounds for rational structural modifications would be a significant approach towards the discovery of anti-gout medications.

Fig. 1 The design of compounds. [Images not available. See PDF.]

a Structures of representative drugs used in the treatment of gout (a–f) and representative NOD-like receptor protein 3 (NLRP3) inflammasome inhibitor (g); b Pharmacophore models of β-carboline-1-propionic acid and lesinurad (The figures were generated by Schrödinger suite 2023-1); c The pharmacophore model of urate transporter 1 (URAT1) inhibitors (This model was generated by Molecular Operating Environment (MOE, 2022.02; Chemical Computing Group ULC, 910-1010 Sherbrooke St. W., Montreal, QC H3A 2R7, 2022)); d The pharmacophore model of NLRP3 inhibitors (This model was generated by MOE); e The dual-target pharmacophore model of URAT1 and NLRP3 inhibitors (This model was generated by MOE); f Pharmacophore model-guided design of target compounds.

Eurycoma longifolia Jack (EL), an indigenous tropical plant to Southeast Asian countries such as Vietnam, Malaysia, and Indonesia, is renowned for its diverse bioactivities, including anti-inflammatory, anti-gout, anti-cancer, hypoglycemic effects, and enhancement of sexual functions^{21, 22, 23–24}. The primary chemical constituents of EL are triterpenoids with the quassinoid skeleton, β-carboline alkaloids, coumarins, squalenes, and monophenyl rings, etc²⁵. Recently, a study revealed that the administration of EL significantly decreased blood uric acid levels and prevented kidney pathological changes in an animal model of HUA induced by potassium oxonate, thereby improving renal urate transport. As research into the uric acid-lowering mechanisms of EL has progressed, it has been revealed that a variety of its components can inhibit the activity of xanthine oxidase (XOD), thereby reducing uric acid levels²⁶. Additionally, eurycomanol isolated from EL has been reported to inhibit uric acid uptake in HEK293T cells expressing the human uric acid transporter 1 (hURAT1)²⁷. However, it is unknown whether β-carboline alkaloids, which rank second only to triterpenoids in EL, possess uric acid-lowering activity.

In this work, we utilize the phenotypic screening to assess the uric acid-lowering effects of β-carboline alkaloids in the acute HUA mouse model²⁸. The findings demonstrate that β-carboline-1-propionic acid (h, Fig. 1a), present in the natural product EL, exhibits a specific uric acid-lowering activity (decrease ratio, DR = 41.70%), surpassing the efficacy of the marketed drug lesinurad (DR = 32.48%). In order to further improve the uric acid-lowering effects, incorporate anti-inflammatory activities, and mitigate toxicity, we employ the pharmacophore models of lesinurad, β-carboline-1-propionic acid, URAT1 inhibitors and NLRP3 inhibitors as a guide to multi-round optimization of the structure of β-carboline-1-propionic acid²⁹. Subsequently, we successfully design and synthesize a total of eight carboline compounds and fifty-six indole compounds. All newly synthesize compounds underwent phenotypic screening to assess their uric acid-lowering activity in mice. Compound 32 represents a promising anti-gout drug candidate following comprehensive target identification and rigorous safety and drug-likeness assessments. This compound exhibits both uric acid-lowering and anti-inflammatory activities, while demonstrating high efficacy along with low toxicity.

Results

Structural optimization of β-carboline-1-propionic acid

The β-carboline-1-propionic acid composed of a β-carboline ring and a side-chain propionic acid, is classified as a β-carboline alkaloid with potential anti-gout activity in EL³⁰. The phenotypic screening results indicate that β-carboline-1-propionic acid (DR = 41.70%) exhibits a certain SUA-lowering activity in our well-established acute HUA mouse model, surpassing the effectiveness of the marketed agent lesinurad (DR = 32.48%). Regrettably, β-carboline-1-propionic acid exhibited no inhibitory effect on XOD and URAT1 (Table 1), the two primary targets for reducing uric acid levels. Upon examining the structure of β-carboline-1-propionic acid, we observed a resemblance to the structure of lesinurad, prompting us to construct respective pharmacophore models for both lesinurad and β-carboline-1-propionic acid. The pharmacophore model of lesinurad, which encompasses four essential features including an anionic group, a hydrogen-bonded acceptor, an aromatic ring and a hydrophobic group, aligns consistently with the three structural components (the core ring A, the hydrophobic domain B, and the side chain C containing an anionic group) elucidated in our previous work on lesinurad and its derivatives³¹. Interestingly, the distance between the anionic group and the hydrogen bond acceptor of β-carboline-1-propionic acid and lesinurad is nearly identical (lesinurad is 5.40 Å and β-carboline-1-propionic acid is 5.74 Å). Subsequently, we integrated the two pharmacophore models and observed that in comparison to the lesinurad pharmacophore, the β-carboline ring of β-carboline-1-propionic acid can be identified as the core ring A while its side chain propionic acid can be regarded as the anionic side chain C, lacking solely the hydrophobic domain B (Fig. 1b).

Table 1. Re-evaluation of uric acid-lowering activity in vivo and assessment of target activity in vitro for representative compounds

Compds.	SUA (μM)	c DR %	d URAT1 IC50 (μM)	d,e XOD IR (%)
1	241.33 ± 109.70	78.82	3.02 ± 0.82	17.79
5	336.33 ± 107.55	72.89	13.80 ± 4.43	16.39
14	332.33 ± 140.97	73.28	16.09 ± 2.94	NA
17	270.00 ± 49.15	82.53	12.01 ± 2.39	20.31
25	246.00 ± 109.80	86.10	NA	17.21
27	188.20 ± 14.70	94.68	5.85 ± 2.86	5.31
32	196.67 ± 26.50	93.42	3.81 ± 0.99	23.56
34	317.75 ± 77.51	75.44	9.76 ± 2.33	1.21
41	180.00 ± 4.00	95.89	4.09 ± 0.79	2.49
45	334.67 ± 188.60	72.93	14.00 ± 2.08	NA
47	251.00 ± 77.43	85.35	NA	NA
58	326.75 ± 64.88	74.11	NA	NA
63	327.25 ± 101.13	74.03	5.80 ± 0.74	NA
64	287.33 ± 98.74	79.96	5.56 ± 0.90	NA
h	484.25 ± 86.28	50.73	NA	NA
d	549.00 ± 130.93	41.12	6.88 ± 0.81	NA
a	f178.00 ± 0.00	96.16	-	100.24
c	283.00 ± 154.08	80.60	-	-
aModel	826.00 ± 127.14	-	-	-
bVehicle	152.33 ± 36.36	-	-	-

^aModel group: male HUA mice induced with hypoxanthine (p.o.) and potassium oxonate (s.c.), and not treated with test compound (p.o., untreated HUA mice); ^bVehicle group: male normal mice untreated with hypoxanthine (p.o.) and potassium oxonate (s.c.), or test compound (p.o., untreated normal mice); ^cDR = (model SUA – compound SUA) / (model SUA – vehicle SUA); ^dNA: No activity; ^eXOD IR: the XOD inhibitory ratio of these compounds at a relatively high single concentration (10 μM); ^fWhen using a uric acid analyzer with a lower limit of quantification (LLOQ) of 178 μM, all “LO” flags output by the instrument (indicating measurements below the LLOQ) are uniformly recorded as 178 μM following the principle of conservative assignment.

First, to improve the SUA-lowering activity of β-carboline-1-propionic acid, we utilized it as the lead compound and employed a molecular hybridization strategy guided by the pharmacophore model of lesinurad. While preserving the core ring structure of β-carboline-1-propionic acid, the bromine- or cyclopropyl-substituted benzene rings and naphthalene rings were incorporated into the core ring via molecular hybridization in hydrophobic domain B, affording a total of eight carboline compounds (series I, compounds 1–8). In the subsequent round of modification, while preserving hydrophobic domain B, we employed a scaffold hopping strategy to introduce flexibility into the rigid β-carboline ring by opening it and to ensure preservation of hydrogen-bonded acceptor features by substituting pyridine N with carbonyl oxygen. Meanwhile, the introduction of hydrophobic groups (such as methyl and dimethyl) in close proximity to the carboxylic acid, or by appropriately elongating the carbon atom chain to investigate its impact on uric acid-lowering activity, resulted in a total of eight indole amide compounds (series II, compounds 9–16).

Subsequently, to enhance the NLRP3 inflammasome inhibitory activity while maintaining the SUA-lowering activity, we analyzed the binding pocket of NLRP3 inhibitors that can be categorized into three distinct zones: the hydrophobic domain located within the pocket, the electropositivity channel situated in its central domain, and the solvent-exposed domain positioned externally and constructed a dual-target pharmacophore model of URAT1 and NLRP3 inhibitors which comprises three hydrophobic groups (S1, S2, and S3) and an anionic center, with respective spacings of 3.94, 6.20, and 3.58 Å, as well as angles of 86.6° and 90.9° ^{32, 33–34} (Fig. 1c–e). Based on this pharmacophore model, the indole ring was retained as S1 while the hydrophobic naphthalene ring remained as S2; additionally, we introduced a sulfonamide group as the anionic center through molecular hybridization. Finally, 48 indole acylsulfonamide compounds were designed by connecting different aryl and alkyl groups in S3 (series III, Fig. 1f).

Effects of newly synthesized compounds on reducing SUA

The reduction of uric acid levels represents a promising strategy for the prevention of gout flares and the achievement of normal uric acid levels. These structural compounds were initially evaluated for their SUA-lowering activity in acute HUA mouse models at the oral dose of 2 mg/kg. Febuxostat (a), benzbromarone (c), lesinurad (d), and β-carboline-1-propionic acid (h) were treated as positive controls. The summarized results can be found in Supplementary Table 1. The structure-activity relationship (SAR) analysis showed that the introduction of the benzene and naphthalene rings in series I significantly enhanced the in vivo SUA-lowering activity. Among them, compound 1 (82.86%) and 5 (74.54%) exhibited more than a twofold increase in potency compared to the lead compound β-carboline-1-propionic acid (42.11%). Further analysis revealed that incorporation of a larger naphthalene ring in zone B was more favorable for enhancing activity compared to a benzene ring, while bromine-substituted compounds demonstrated superior potency compared to cyclopropyl-substituted compounds. In series II, when the carboline skeleton of the core ring (zone A) was replaced with the indole ring, most compounds still demonstrated SUA-lowering activity in mice. Among them, 14 (72.19%) exhibited the highest potency. Unlike series I, the potency of the cyclopropyl-substituted naphthalene ring (zone B) was significantly greater than that of the bromine-substituted naphthalene ring, and the introduction of a methyl group in the side chain proved beneficial for enhancing activity. In series III, the overall in vivo activity of a total of 36 compounds surpassed that of β-carboline-1-propionic acid (47.9%, 50.33%) (Supplementary Table 2). Among them, 25 (80.7%), 27 (84.1%), 32 (83.5%), 41 (85.97%) and 47 (84.18%) demonstrated significant SUA-lowering effects, with DR > 80%. Similar to series II, the cyclopropyl-substituted naphthalene ring in zone B had superior activity than the bromine-substituted naphthalene ring. The subsequent SAR analysis revealed that the activity was significantly affected by various substituted sulfonamide groups (Fig. 2).

Fig. 2 Structure-activity relationship of compounds in series III. [Images not available. See PDF.]

The color black indicates no activity; orange indicates DR > 30%; blue indicates DR > 50% and red indicates DR > 70%.

The representative compounds in series I ~ III (DR > 70%) underwent a secondary evaluation to confirm the activity of the compounds, and the corresponding results are presented in Table 1. These compounds demonstrated significant efficacy in reducing uric acid levels. Specifically, six compounds (17,25,27,32,41, and 47) exhibited higher activity than benzbromarone (80.60%), with approximately twice the potency of lesinurad (41.12%), and a 1.6-fold increase compared to the lead compound β-carboline-1-propionic acid (50.73%). The most potent compounds among them were 27 (94.68%), 32 (93.42%), and 41 (95.89%), showing comparable activity to febuxostat (96.16%). Subsequently, these compounds were subjected to experiments investigating mechanisms for lowering SUA levels and anti-inflammation.

Evaluation of representative compounds on URAT1 in vitro and molecular simulation

In order to preliminarily investigate the targets of representative compounds (DR > 70%) with uric acid-lowering activity, the URAT1-inhibitory activities of those compounds were evaluated in HEK293-URAT1-overexpressing cells with ¹⁴C labeled uric acid as the substrate³⁵. The results are presented in Table 1. Among them, 1 and 32 displayed the most potent inhibition on URAT1 with IC₅₀ values of 3.02 ± 0.82 μM and 3.81 ± 0.99 μM, which were about 2.3-fold and 1.8-fold higher than that of lesinurad (IC₅₀ = 6.88 ± 0.81 μM), respectively. 27 (IC₅₀ = 5.85 ± 2.86 μM), 41 (IC₅₀ = 4.09 ± 0.79 μM), 63 (IC₅₀ = 5.80 ± 0.74 μM) and 64 (IC₅₀ = 5.56 ± 0.90 μM) exhibited slightly superior URAT1 activity compared to lesinurad in vitro. This indicates that compounds 1,27,32,41,63 and 64 exhibit significant potential in reducing the reabsorption of uric acid in renal tubules and promoting its excretion by inhibiting URAT1.

Afterward, we proceeded to investigate the binding mode of 1 and 32 with URAT1 (Fig. 3a) through molecular docking, aiming to gain insights into the underlying mechanism behind the enhanced activities³⁶. As shown in Fig. 3b, the residues PHE395 and PHE241 exhibited potential π-π stacking interactions with compound 1. The carboxylic acid of compound 1 also displayed hydrogen bonds with residues ARG477 and GLN473. Compared with 1, the anionic group of 32 reduces the interaction with GLN473, while the indole ring of 32 forms an additional π-π stacking interaction with PHE364 (Fig. 3c). The overall findings suggest that compounds 1 and 32 exhibit enhanced additional interactions in comparison to lesinurad (Fig. 3d), thereby contributing to their superior inhibitory activity against URAT1.

Fig. 3 Results of the predicted compound 32 binding conformation to URAT1. [Images not available. See PDF.]

a An overview of binding pocket of compounds in urate transporter 1 (URAT1) (PDB code: 9JDZ); b–d Predicted binding modes of compound 1, compound 32 and lesinurad with URAT1 (The figures were generated by PyMOL (Version 2.5.2 Schrödinger). Yellow dashed lines are hydrogen bonds, blue dashed lines are π-π stackings); e RMSD analysis for predicted URAT1 complexes with compounds 1,32 and lesinurad.

Molecular dynamics simulations were performed for 100 ns to study the stability of the 1-predicted URAT1, 32-predicted URAT1 and lesinurad-predicted URAT1 docking complexes. In order to clarify the structural characteristics of URAT1, the root mean square deviation (RMSD) of 1-predicted URAT1 complex, 32-predicted URAT1 complex and lesinurad-predicted URAT1 complex was analyzed throughout the simulation time (Fig. 3e). The 1-predicted URAT1 complex and 32-predicted URAT1 complex showed a stable trend from 0 to 100 ns, with only slight fluctuations observed at the 80 ns mark. However, the lesinurad-predicted URAT1 complex demonstrated a persistent upward trend throughout the simulation process. The 1-predicted URAT1 complex and 32-predicted URAT1 complex exhibited a more stable RMSD value (less than 4.00 Å) compared to the lesinurad-predicted URAT1 complex.

Study on anti-Inflammatory activity of representative compounds in vitro and in vivo

The NLRP3 inflammasome is a key component of the innate immune system and plays a crucial role in the pathogenesis of various inflammatory diseases, including gout. Activation of the NLRP3 inflammasome leads to the cleavage of pro-IL-1β and pro-IL-18 into their active forms, which subsequently trigger a cascade of inflammatory responses³². In this study, we investigated the potential of representative compounds in inhibiting NLRP3 inflammasome activation by evaluating their effects on IL-1β release in lipopolysaccharide (LPS) and adenosine triphosphate (ATP) stimulated mouse bone marrow-derived macrophages (BMDMs). The results showed that all other compounds could exhibit IL-1β inhibitory at the concentration of 3 μM with the exception of compounds 27, 63, and 64 (Supplementary Table 3). Among them, the most potent compounds 25, 32, and 45 were further evaluated for their IC₅₀ values in inhibiting IL-1β release in BMDMs stimulated by LPS and ATP (Fig. 4a). The results showed that 32 (IC₅₀ = 2.61 μM) and 45 (IC₅₀ = 2.61 μM) exhibited greater inhibitory activity compared to 25 (IC₅₀ = 7.28 μM).

Fig. 4 Anti-inflammatory evaluation of the representative compounds and the effects of 32 on NLRP3. [Images not available. See PDF.]

a The half-inhibitory concentration of compounds on IL-1β secreted by BMDMs (n = 3, expressed as mean ± SEM); b, c32 does not affect the secretion of TNF-α and IL-6 (n = 3, expressed as mean ± SEM, the Tukey’ multiple comparions test was used to the multiple comparison analysis); d Alterations in rat ankle circumference (n = 6, expressed as mean ± SEM); e Illustrative images depicting the progression of rat ankle joints over various time intervals; f The surface plasmon resonance (SPR) of 32 and NOD-like receptor protein 3 (NLRP3); g Cellular Thermal Shift Assay (CETSA) determines the binding of NLRP3 to 32. This experiment was performed once, with technical consistency observed across the temperature gradient (48–78 °C) in the assay; h The interaction between 32 with NLRP3. (The figures were generated by PyMOL. Yellow dashed lines are hydrogen bonds.); i Dynamics simulation trajectory analysis; j Timeline representation of interactions and contacts (H-bonds, hydrophobic, water bridges) in 32-predicted NLRP3 complex during 100 ns molecular dynamics simulation. (Compared with the control group, ns, P > 0.05.).

Subsequently, considering the uric acid-lowering activity, 32 was selected for further investigation into its anti-inflammatory mechanisms. Upon treatment of BMDMs with 32, no significant changes in TNF-α and IL-6 production were observed (Fig. 4b, c), indicating that 32 does not interfere with the initiation phase of NLRP3 inflammasome activation. Thereafter, the in vivo efficacy of 32 was assessed in a rat model of gout arthritis at the injection dose of 2 mg/kg. Acute inflammation was induced by intra-articular injection of MSU crystals into the ankle joint, and the extent of ankle joint swelling was monitored at 2, 4, 6, 8, 12 h following administration while assessing the release of IL-1β. The administration of 32 demonstrated significant efficacy in reducing ankle swelling in rats with acute gouty arthritis (Fig. 4d, e), while also exerting potent anti-inflammatory effects (Supplementary Fig. 1), which was comparable to MCC950. It is worth noting that the concentration of 32 in the ankle joint (83,733 ng/g) was significantly higher than that of MCC950 (9709 ng/g) at 2 h, suggesting that its anti-inflammatory effects may be enhanced through local tissue accumulation (Supplementary Table 4). Although 32 has relatively weak NLRP3 inhibitory activity, the high exposure level may partially compensate for the activity deficiency through a concentration-dependent effect, thereby effectively regulating the NLRP3.

To determine whether compound 32 can bind to NLRP3 protein, the affinity of the inhibitor for NLRP3 was analyzed using surface plasmon resonance (SPR)³⁷. The results in Fig. 4f showed that 32 could bind to NLRP3 protein (K_D = 27.8 μM). Simultaneously, we employed the cell thermal shift assay (CETSA) for additional validation (Fig. 4g)³⁸. The results were consistent with those obtained from the SPR experiment, indicating that 32 enhances the thermal stability of NLRP3, which suggests a potential binding interaction between the 32 and NLRP3. After determining that 32 exerts a direct effect on NLRP3, we proceeded to investigate the binding mode of 32 with NLRP3. The docking results showed that 32 can be anchored within the electropositivity channel of the NLRP3 central pocket, and form hydrophobic interactions with multiple amino acid residues (ARG578, ALA228) and additional π-π stacking interactions with PHE575 and TYR632 (Fig. 4h). 32 was further studied for molecular dynamics simulation. The results are shown in Fig. 4i, the RMSD analysis demonstrated stable maintenance of the NLRP3 protein backbone within 2-3 Å, while the ligand exhibited fluctuations around 4 Å, indicating a solid binding of 32 to the pocket. The contacts established between NLRP3 protein and 32 throughout the trajectory are depicted in Fig. 4j, with the top panel illustrating interactions at different frames. Notably, multiple protein residues exhibited specific contacts with 32, indicated by a darker shade of orange color. This observation suggests the stability of the 32-predicted NLRP3 complex.

In vitro evaluation of inhibitory effect of representative compounds on other targets associated with gout

The majority of natural products have been reported to be effective urate-lowering drugs by inhibiting XOD, while a few studies have demonstrated the extract from EL to possess XOD activity. The XOD inhibitory potency of these compounds was initially assessed at a relatively high single concentration of inhibitor (10 μM) in order to identify potent XOD inhibitors. Unfortunately, the results demonstrate that the inhibitory effect of these compounds on XOD was found to be less than 30% at a concentration of 10 μM, indicating their negligible inhibitory activity (Table 1). In the same experimental system, febuxostat achieved complete inhibition (100%), while lesinurad exhibited no inhibitory effect.

Additionally, we have also systematically investigated the inhibitory effects of 32 on key urate reabsorption transporters, such as the organic anion transporter 4 (OAT4) and glucose transporter 9 (GLUT9)^{39, 40, 41–42}. OAT4, localized in the apical membrane of human renal proximal tubule epithelial cells, has been demonstrated to effectively promote urate excretion upon functional inhibition (as exemplified by lesinurad). Using HEK293-OAT4 overexpressing cell models and 6-carboxyfluorescein (6-CFL) as a characteristic substrate for uptake assays, compound 32 (IC₅₀ = 6.08 ± 2.58 μM) exhibited comparable inhibition against OAT4 as lesinurad (IC₅₀ = 4.35 ± 1.48 μM, Fig. 5a). Notably, GLUT9 serves as a high-capacity urate transporter that synergistically mediates renal urate reabsorption alongside OAT4 and URAT1, yet currently lacks clinically available inhibitors. The relative current was elicited in a HEK293 cell model overexpressing GLUT9 by incubating with 1 mM uric acid, and compound 32 exhibited potent inhibitory activity against GLUT9 (IC₅₀ = 11.02 ± 1.33 μM, Fig. 5b). The SPR experiments further demonstrated that the compound 32 exhibits strong affinities with OAT4 (K_D = 0.71 μM) and GLUT9 (K_D = 6.37 μM, Fig. 5c, d).

Fig. 5 Results of molecular dynamics simulations of the predicted compound 32 binding conformation to URAT1, OAT4 and GLUT9. [Images not available. See PDF.]

a Dose-dependent effects of benzbromarone and compound 32 on organic anion transporter 4 (OAT4) cells (n = 4, biological replicates, expressed as mean ± SEM); b Dose-response relationship of glucose transporter 9 (GLUT9) currents by 32 (n = 5, biological replicates, expressed as mean ± SEM); c The binding affinity of 32 to OAT4 was evaluated by surface plasmon resonance (SPR); d The binding affinity of 32 to GLUT9 was evaluated by SPR; e Predicted binding modes of 32 with OAT4 (PDB code: 8WJH) (The figures were generated by PyMOL. Yellow dashed lines are hydrogen bonds, green dashed lines are π-π stackings); f Predicted binding modes of 32 with GLUT9 (PDB code: 8Y65) (The figures were generated by PyMOL and Schrödinger. Yellow dashed lines are hydrogen bonds, green dashed lines are π-π stackings); g Root mean square deviation (RMSD) analysis for predicted OAT4 complexes with 32; h RMSD analysis for predicted GLUT9 complexes with 32; i The binding affinity of 32 to urate transporter 1 (URAT1) was evaluated by SPR; j–l RMSD analysis for predicted OAT4 complexes with 32.

Given the unexpected identification of OAT4 and GLUT9 during target activity screening, the specific binding sites of these proteins remain unclear. Therefore, we employed Chai-1 to predict the complex structure and utilized molecular dynamics simulations to verify conformational stability⁴³. As shown in Fig. 5e, sulfonamide group of 32 has stable polar interactions with ARG389 and TYR361 of OAT4, and its aromatic ring system has π-π stacking with PHE211. Moreover, its aromatic domain is embedded within hydrophobic pocket of GLUT9, forming multiple hydrophobic interactions with PHE477, TYR369 and TRP378, while its sulfonamide group forms a hydrogen-bonding network with the side chain of ASN500 (Fig. 5f). Subsequently, the RMSD of compound 32 with transmembrane of OAT4 and GLUT9 has been analyzed, focusing on the RMSD of the backbone of the transmembrane domain protein and the heavy atom of the small molecule. The results of RMSD showed that 32 can bind to OAT4 and GLUT9 in a relatively stable manner (Fig. 5g, h). Furthermore, we conducted systematic validation of the interaction between compound 32 and URAT1 using SPR and RMSD analysis. The SPR results revealed moderate-affinity specific binding between them (K_D = 2.84 μM, Fig. 5i). During the 100 ns molecular dynamics simulation, the RMSD values of the backbone atoms in the URAT1 transmembrane domain remained stable at 3.0 ± 1.0 Å, indicating maintained protein conformational stability. Meanwhile, the heavy atoms of compound 32 showed minimal RMSD fluctuations, suggesting its stable positioning within the binding pocket. Notably, the distance between the key binding residue PHE241 and the aromatic ring center of compound 32 consistently remained at 5.0 ± 0.5 Å, forming stable π-π stacking interactions (Fig. 5j, l).

Preliminary drug-likeness evaluation of compound 32

In order to further validate the efficacy, we displayed a discussion on the minimum effective dose of compound 32 in the acute HUA mouse model. Lesinurad, febuxostat and benzbromarone were selected as positive controls. The results are presented in Supplementary Table 5. Compound 32 exhibited notable SUA-lowering activity at the dosage of 1 mg/kg, with a DR value of 90.6%, which is comparable to that of febuxostat (93.1%). However, the activity of lesinurad and benzbromarone at that dose was limited, with DR values of 10.8% and 17.3%, respectively. Taken together, these findings provide further confirmation of the improved in vivo activity and effective dose of 32 compared to those of lesinurad and benzbromarone. Furthermore, we conducted a comprehensive investigation into the relationship between the exposure and pharmacological effects of the compound 32. In the acute HUA mouse model, increasing the oral dose from 1 mg/kg to 2 mg/kg resulted in pharmacokinetic changes. The absorption phase displayed saturation kinetics, evidenced by marginal 10.8% plasma concentration elevation at 15 min (from 223 to 247 nM) despite 100% dose increment. This was followed by disproportionate exposure increases: C_max rose 26.5% (from 223 to 282 nM) while at 1 h concentration surged 115.8% (from 63.1 to 135.9 nM), indicative of altered distribution or elimination. Crucially, despite a 71.5% increase in AUC_0-∞, the urate-lowering efficacy improved by merely 2.8%. This nonlinear exposure-response relationship indicates proximity to an efficacy plateau, implying limited therapeutic gain from further dose escalation. The IC₅₀ value (3.81 μM) of compound 32 against URAT1 is much higher than the maximum concentration (223 nM) at the effective in vivo oral dose (2 mg/kg). This disparity can potentially be attributed to multi-target synergy, as the compound not only inhibits URAT1 but also demonstrates inhibitory effects on OAT4 and GLUT9.

In light of gout being a persistent ailment demanding an extended therapeutic intervention, the therapeutic potential and safety profile of compound 32 in addressing chronic HUA were further explored utilizing a 30-day model (Fig. 6a). According to the findings in Fig. 6b, it can be observed that the chronic HUA model in this experiment may have an impact on mouse growth, while febuxostat, lesinurad and 32 demonstrate potential for promoting normal growth in HUA mice. As shown in Fig. 6c-g, the values of SUA, serum creatinine (SCr), blood urea nitrogen (BUN), alanine aminotransferase (ALT) and aspartate aminotransferase (AST) in the model group were significantly elevated compared to the vehicle group, indicating successful establishment of a chronic HUA model induced by potassium oxonate and hypoxanthine, but this model resulted in liver and kidney function impairment. Compared to the model group, the compound 32 group exhibited varying degrees of reduction in all aforementioned indicators, whereas the lesinurad and febuxostat groups showed a slight increase in AST levels. These findings suggest that intragastric administration of compound 32 exhibit therapeutic effects on mice with chronic HUA, potentially providing hepatorenal protection and ameliorating liver and kidney damage induced by chronic HUA. However, it should be noted that while lesinurad also demonstrates efficacy in treating chronic HUA, it may exacerbate liver and kidney damage. The liver and kidney specimens were sectioned, stained with hematoxylin and eosin (H&E) (Supplementary Fig. 2), and the results were generally consistent with the aforementioned observations. Compared to the model group, compound 32 group exhibited less severe renal and hepatic damage, while the lesinurad group demonstrated increased damage that could potentially lead to expansion of the renal capsule.

Fig. 6 In vivo efficacy evaluation and safety data of compound 32. [Images not available. See PDF.]

a Flow chart of the establishment of chronic hyperuricemia (HUA) model in mice; b Effects of 32 on the body weight of chronic HUA mice during the treatment (n = 6, expressed as mean ± SEM); c–g Effects of 32 on the levels of serum uric acid (SUA), serum creatinine (SCr), blood urea nitrogen (BUN), alanine aminotransferase (ALT), aspartate aminotransferase (AST) in chronic HUA mice (n = 6, expressed as mean ± SEM, one-way analysis of variance (ANOVA) with Tukey’s multiple comparison test (two-sided)); h IC₅₀ of 32 and positive control Cisapride on hERG tail current (n = 3, biological replicates, expressed as mean ± SEM); i, j The weight change of mice was assessed after oral administration of test compounds at a dose of 1000 mg/kg one week (n = 5, expressed as mean ± SEM); k The body weight of mice before and after oral administration of test compounds at a dose of 50 mg/kg for one month (n = 5, biological replicates, expressed as mean ± SEM, one-way analysis of variance (ANOVA) with Tukey’s multiple comparison test (two-sided)); l Appearance of mouse organs (n = 5); m Effects of organ ratio in mice (n = 5, expressed as mean ± SEM, one-way analysis of variance (ANOVA) with Tukey’s multiple comparison test (two-sided)); n Histopathological study of subacute toxicity in mice. Heart, liver, spleen, lung and kidney were sectioned and stained with H&E (scale bar = 100 μm). Six mice were included per experimental group, with three randomly selected animals per group undergoing detailed pathological examination. This experiment was performed in a single independent replicate, with no inter-individual variability in tissue pathology observed across the analyzed mice; o Plasma concentration-time profiles of 32 in rats following i.v. or p.o. administration (n = 3, biological replicates, expressed as mean ± SD). (Compared with the control group, ***P < 0.001; **P < 0.01; *P < 0.05; ns, P > 0.05).

For the safety evaluation, we initially conducted an in vitro hERG channel blockade assay using a manual patch-clamp method, employing cisapride monohydrate as a reference compound. The results indicated that compound 32 is unlikely to induce cardiotoxicity (IC₅₀ > 10 µM) (Fig. 6h).

Then, the acute and subacute toxicity were assessed on healthy Kunming (KM) mice. The administration of 1000 mg/kg of compound 32 via oral gavage (p.o.) during the acute toxicity study did not result in any mortality or significant changes in body weight (Fig. 6i, j), while the febuxostat group showed a mortality rate of 50% (with sex-specific rates of 40% in males and 60% in females), whereas the lesinurad group exhibited 100% mortality under the same dosage conditions (Supplementary Table 6).

In the subacute toxicity experiment, all mice tolerated oral administration of 32, febuxostat and lesinurad at the dose of 50 mg/kg twice daily for 30 days without any deaths, abnormal behaviors or statistically significant differences in body weight throughout the entire study period (Fig. 6k, weekly body weight variations in Supplementary Fig. 3). The blood and organs including heart, liver, spleen, lung and kidney were collected for observation on the final day. The blood routine examination showed that the administration of compound 32 had minimal impact on total white blood cell count, red blood cell count, hemoglobin level, lymphocyte count and platelet count in mice when compared to the control group (Supplementary Table 7). As shown in the Fig. 6l, the impact of compound 32 on the morphological characteristics of these organs was minimal compared to febuxostat and lesinurad. Additionally, compound 32 exhibited negligible effects on the organ ratios of these organs and did not demonstrate any observable pathological alterations when compared to the control group (Fig. 6m, n). However, in the group treated with febuxostat, occasional observations of cardiomyocyte edema, cell swelling, cytoplasm loosening and lightly stained can be noted. And our preceding investigations have revealed that lesinurad possesses the potential to induce hepatotoxicity and nephrotoxicity, such as extensive ballooning degeneration and condensation of hepatocytes in the liver, hydropic degeneration of hepatocytes, and enlargement of glomeruli⁴⁴. Overall, 32 showed enhanced preclinical safety profiles relative to lesinurad and febuxostat.

To assess the drug-likeness of compound 32, we conducted an evaluation of its DMPK properties. Firstly, the inhibitory activity of 32 against human cytochrome P450 (CYP) enzymes was measured. In this assay, 32 did not show obvious inhibitory activity against CYP1A2, 2D6, and 3A4 (IC₅₀ > 30 µM) (Supplementary Table 8). However, it exhibited moderate inhibitory activity against CYP2C9 (IC₅₀ = 2.77 μM) and CYP2C19 (IC₅₀ = 2.03 μM); therefore, further evaluation is necessary to assess the specific risk of drug-drug interaction (DDI) mediated by enzyme inhibition. Next, the pharmacokinetic profiles of 32 were evaluated in Wistar rats, and the blood concentration-time profiles and main pharmacokinetic parameters are depicted in Fig. 6o and Supplementary Table 9. The oral bioavailability of 32 was 53.2%, which meets the requirements for a potential candidate drug.

Discussion

The global prevalence of gout and HUA is significantly elevated. Currently available therapeutic drugs primarily consist of rapidly acting anti-inflammatory agents and life-long uric acid-lowering agents, but their therapeutic outcomes remain unsatisfactory. Moreover, these drugs exhibit a single-target inhibitory mechanism, which give rise to significant adverse effects and potential long-term safety concerns associated with prolonged medication. Therefore, there is an urgent imperative to develop anti-gout medications that exhibit enhanced efficacy, diminished toxicity, and are suitable for prolonged administration. According to statistics, impaired renal urate excretion is observed in ~90% of patients with primary HUA⁴⁵. During the process of renal excretion of uric acid, URAT1 and GLUT9 act as primary reabsorption transporters, thereby serving as major determinants for plasma uric acid levels and gout development⁴⁶. This suggests that targeting URAT1 and GLUT9 may be an effective approach for treating HUA. On the other hand, the activation of NLRP3 inflammasome by MSU crystals holds significant relevance in both the initiation and progression of gout arthritis, thereby presenting a promising prospective therapeutic target⁴⁷. Consequently, the pharmacological inhibition of urate reabsorption transporters and modulation of NLRP3 inflammasome assembly and activation present a promising avenue for therapeutic intervention in gout.

Natural products and active ingredients derived from traditional Chinese medicine possess distinct advantages in the treatment of multifactorial chronic diseases owing to their diverse therapeutic effects and minimal adverse reactions, rendering them a pivotal source for innovative drugs targeting major chronic ailments. In view of this, we employed a mouse model of acute HUA for phenotypic screening and discovered that β-carboline-1-propionic acid, a naturally occurring compound found in EL, demonstrated notable efficacy in reducing uric acid levels in mice surpassing that of the marketed drug lesinurad (DR = 41.70% VS 32.48%). Then, the pharmacophore model was utilized to conduct three series of structural optimization for β-carboline-1-propionic acid. Subsequently, indole acyl sulfonamide compounds with triple-target and dual-mechanism anti-gout effects were discovered through in vitro and in vivo activity screening.

Notably, the drug candidate 32 exhibited twofold inhibitory activity compared to lesinurad in the URAT1 target activity test conducted in vitro (IC₅₀ = 3.81 μM VS 6.88 μM). Moreover, its acute uric acid-lowering efficacy in mice (DR = 93.42%) was significantly superior to that of lesinurad (DR = 41.12%) and comparable to febuxostat (DR = 96.16%). The 30-day murine model for assessing chronic HUA demonstrated that compound 32 exhibited a sustained reduction in SUA levels, while indicators, such as SCr, BUN, ALT, and AST showed no significant signs of hepatorenal damage. Concurrently, the liver XOD activity assay results indicated that compound 32 displayed no inhibitory effect on XOD activity, suggesting its mechanism of action in reducing uric acid levels may involve promoting uric acid excretion. Comprehensive pharmacological evaluation combining target activity profiling and SPR analysis demonstrated that compound 32 exerts its uric acid-lowering effect by simultaneously inhibiting three key urate transporters OAT4, GLUT9, and URAT1. Additionally, compound 32 displayed a concentration-dependent impact on the release of IL-1β induced by ATP in BMDMs, demonstrating potent anti-inflammatory effects through the inhibition of NLRP3 inflammasome. The DMPK evaluation results indicate that compound 32 exhibits a rat oral bioavailability of 53.2%, meeting the requirements for oral administration. The safety evaluation revealed the absence of hERG toxicity for compound 32, with a maximum tolerated dose exceeding 1000 mg/kg in mice, and no evident organ damage following long-term high-dose (50 mg/kg) administration. Thus, compound 32 exhibits superior safety compared to existing drugs.

Overall, this study establishes 32 as a natural product-derived anti-gout agent with a unique polypharmacology profile, targeting four key proteins (URAT1/GLUT9/OAT4/NLRP3) to achieve dual uric acid-lowering and anti-inflammatory effects. By avoiding excessive single-target inhibition, its multi-mechanism design mitigates the severe adverse effects of current lifelong uric acid-lowering drugs. Beyond gout, this work exemplifies a broadly applicable drug discovery paradigm: leveraging natural product scaffolds and pharmacophore-guided polypharmacology to address complex diseases. While clinical validation is pending, 32 represents a promising candidate for advancing gout therapy toward safer, more effective, and disease-modifying outcomes.

Methods

Compounds and cells

The compounds (1–64) shown in Supplementary Table 1-2 were synthesized in our internal facilities. And the detailed synthesis methods and characterization data of these compounds are listed in the supplementary materials (Supplementary Figs. 4-67). HEK293 (cat.no. CL-0003) and HEK293T (cat.no. CL-0005) and THP-1 (cat.no. CL-0233) cells were obtained from Pricella Life Science & Technology Co., Ltd.

Ethics statements

Animal care and handling were conducted in compliance with the Guide for the Care and Use of Laboratory Animals (U.S. Department of Health and Human Services, NIH Publication no. 93-23, revised 1985), as well as the Animal Care and Use Committee of Shandong University (ECSBMSSDU2021-2-156) and the Animal Management Rules of the Ministry of Health of the People’s Republic of China.

In vivo HUA mouse model

In the process of acclimation and study, animals were raised in groups. We carefully controlled the environment of the animal room (target conditions: temperature 20 °C–25 °C, relative humidity between 40% and 70%, with half-day light exposure and half-day darkness). The 3-week-old healthy KM mice were divided into four groups: a vehicle group, a model group, an active-control group, and an experimental group. HUA was induced in the model group using potassium oxonate (s.c.) and hypoxanthine (p.o.). Detailed information regarding dosages and administration protocols can be found in the supplementary materials. Blood samples were collected after 4 h of acute modeling and 30 days of chronic modeling.

In vitro targets inhibitory potency

A HEK293 cell-based transient expression system for human URAT1 was established to assess ¹⁴C-uric acid transport (detailed protocol in Supplementary Materials).

The in vitro inhibitory effects of compounds on bovine XOD were evaluated spectrophotometrically. Briefly, XOD enzyme powder was reconstituted in PBS containing 0.1 M sodium pyrophosphate (pH 7.5) to a final concentration of 0.5 U/mL. The enzyme (50 μL) was pre-incubated with test compounds (50 μL, 10 μM) for 15 min at 37 °C. The reaction was initiated by adding 100 μL of 0.3 mM xanthine, and absorbance changes were immediately monitored at 295 nm using a microplate reader. Febuxostat and lesinurad served as positive and negative controls, respectively.

The inhibitory effects on GLUT9 were assessed via whole-cell patch-clamp recordings. Currents were measured using a MultiClamp 700B amplifier/Digidata 1550B digitizer and analyzed with pClamp 10 software (Molecular Devices, CA, USA). Detailed procedures are provided in the Supplementary Materials.

OAT4 inhibitory activity was determined using 6-CFL, a classical OAT4 substrate. HEK293 cells (70% confluent in 96-well plates) were transfected with 100 ng/well pcDNA3.1( + )-OAT4 plasmid. After 24 h, cells were pre-treated with test compounds (or vehicle) in HBSS buffer for 30 min, followed by incubation with 100 μM 6-CFL for 15 min. Cells were washed three times with ice-cold DPBS, lysed in 0.1 M NaOH, and fluorescence intensity was measured (excitation/emission: 485/535 nm).

Molecular docking and dynamics simulation

Protein and ligand preparations, as well as docking studies, were performed utilizing the Schrödinger platform suite 2023-1. Chai-1 was utilized to predict the possible binding mode between the protein and ligand. Exported 3D files were visualized in PyMOL (Version 2.5.2 Schrödinger, LLC). All simulations were conducted by using Schrödinger (Schrödinger, LLC. Desmond Molecular Dynamics System, version 2023-1. New York, NY, USA: Schrödinger, LLC, 2023) and employed the Desmond module⁴⁷. The resulting trajectory was then analyzed, and the RMSD of the complex was calculated. Molecular dynamics trajectory data (initial and final configurations) are provided in Supplementary Data 1.

Inflammasome stimulation

The effect of the compounds over IL-1β release was evaluated in BMDMs of mice. Macrophages were obtained from the tibias and femurs of 6-week old C57BL/6 J WT male mice and were cultured overnight in 12-well plates. Subsequently, they were primed with 1 μg/mL LPS for 4 h, followed by the addition of fresh medium containing the different compounds at specified concentrations for 15 min, with MCC950 serving as a positive control. After that, cells were stimulated with 5 mM ATP for 30 min to fully activate the inflammasome. Supernatants were collected and stored at −80 °C for IL-1β, TNF-α and IL-6 detection. The determination of the IC₅₀ values was performed by using GraphPad Prism 9.5.1.

In vivo anti-gout arthritis activity

A total of 42 male Sprague-Dawley (SD) rats (7-8 weeks old, 180–220 g) were acclimated under standard conditions for at least 3 days. They were then randomly divided into the following groups: control group, MSU group, MCC950 group (2 mg/kg), compound 32 groups (2 mg/kg). The weights of the rats were recorded, and each group received the corresponding drug or placebo solution once daily. After continuous administration for 3 days, the rats’ thighs were shaved to expose the ankle joint, and subsequently, the Coderre method was employed to establish the model. The right ankle joint cavity was subsequently injected with a sterile syringe containing an MSU suspension at a concentration of 50 mg/mL to complete the model establishment. After one hour since the establishment of the model, intragastric administration was resumed. The swelling and gait changes in rat toes were closely observed and documented. Blood was collected before modeling, 2, 4, 6, 8, 10 and 12 hours after molding, and the supernatant was obtained by centrifugation for the determination of IL-1β level. The ankle joint tissue exposure and plasma exposure of the compounds at 2, 6, and 24 h were detected.

Surface plasmon resonance

The protein was diluted to 100 ng/μL in sodium acetate buffer at the appropriate pH and immobilized on a CM5 sensor chip via amine coupling. PBS+ buffer was used as the running buffer. Serial dilutions of compound 32 (containing 5% DMSO) were prepared in PBS + . Different concentrations of 32 were injected sequentially over the chip surface. Interaction between 32 and the target protein produced characteristic association and dissociation sensorgrams. No response signal was observed in the absence of binding. Surface regeneration was achieved by washing with buffer between injections. Sensorgram data were globally fitted to a 1:1 Langmuir binding model using BIAcore 1 K Evaluation software (Cytiva, Marlborough, MA, USA) to determine the association rate constant and dissociation rate constant. Processed data were exported to Origin 7 software (Version 7.0552, OriginLab) for graphical representation.

Cellular thermal shift assay

THP-1 cells were lysed on ice and centrifuged. The supernatant was collected, aliquoted, and incubated with either DMSO or compound 32 (10 μM) at room temperature for 10 min. Following incubation, each aliquot was further divided and thermally denatured at different temperatures for 5 min. After centrifugation, supernatants were harvested and subjected to Western blot analysis to detect NLRP3 protein (Cell Signaling Technology, cat. no. 15101).

Assay of hERG activity

HEK293 cells stably transfected with hERG cDNA were used to test the inhibitory effect of compound 32 on the hERG potassium channel. Whole-cell recordings were conducted with a commercial patch clamp amplifier. Percent inhibition was calculated according to the following equation % inhibition = (1-(peak tail current compound)/(peak tail current vehicle)) ×100. The detailed procedures can be found in the supplementary materials.

In vivo toxicity study

Male and female KM mice (3 weeks, 18–22 g) were used to assess the in vivo acute and subacute toxicity of compound 32. Compound 32 were dissolved in 75:20:5 0.5% CMC-Na/water/DMSO. The specific information is showed in supplementary materials.

H&E staining

Animal specimens were fixed in 10% neutral buffered paraformaldehyde, dehydrated using a semiautomatic tissue processor (Leica, Germany), and embedded in paraffin. Sections were stained with Gill’s hematoxylin and eosin Y, and visualized under an inverted light microscope.

Pharmacokinetic properties

The information of compounds, doses and administration scheme for PK studies are showed in supplementary materials. Briefly, male SD rats (n = 3) were treated by intravenous (i.v.) or gavage (p.o.). The vehicle was consisted with 5% N-methyl-2-pyrrolidone (NMP), 10% cremophor EL, 30% PEG400 and 55% water. After administration, the blood samples were collected at indicated time points and analyzed by liquid chromatography tandem mass spectrometry (LC-MS/MS).

Assay of CYP-inhibitory activity

Compound 32 was prepared and incubated with human liver microsomes and cofactor NADPH in the presence of the corresponding probe substrates. Select α-naphthoflavone, sulfaphenazole, ( + )-N-3-benzylnirvanol, quinidine, and ketoconazole were as positive controls. Cold acetonitrile solution containing tolbutamide and labetalol was added to terminate the reaction. 200 μL supernatant was taken and dediluted with 100 μL HPLC water after centrifuged. Finally, the mixtures were subjected to LC-MS/MS analysis.

Statistics/bioinformatics

The data analysis was performed using GraphPad Prism 9.5.1. Results are presented as mean ± SD/SEM deviation of n = 3 independent experiments unless stated otherwise.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Acknowledgements

We gratefully acknowledge financial support from the National Natural Science Foundation of China (No. T2321004, F.Y.), the National Natural Science Foundation of China (No. 82473769, P.Z.; No. 82373921, J.P.) and the Shandong Laboratory Program (SYS202205, P.Z. and X.L.). In December 2020, during a conversation with Peng Zhan, Dengke Gong (Founder of Shandong Medical Alliance), an undergraduate classmate of Peng Zhan, mentioned that Eurycoma longifolia is used as a natural medicine for anti-gout treatment in Malaysia. This information served as the initial inspiration for the project. The authors would like to express their sincere gratitude to Mr. Dengke Gong for providing this valuable information, which laid an important foundation for the initiation of this research.

Author contributions

P.Z., F.Y., J.P., and X.L. conceived the project and designed the experiments. Z.Z., X.S., R.L., and Q.Y. synthesized the compounds and performed biological experiments. Z.G., S.W., and M.W. carried out computational experiments. Y.W., Z.W., M.Y., D.Q., and Z.W. completed the SUA-lowing activity evaluation in vivo. H.L., F.Z., Z.Z, and Y.C. completed the in vitro activity experiment of URAT1, GLUT9, and OAT4. J.E. completed the in vitro activity experiment of NLRP3 inflammasome. T.W., W.Y., completed the in vivo activity experiment of anti-gout. T.Z. assisted in confirming the structure of the compounds. Z.Z and X.S. prepared the manuscript with input from all authors. P.Z. revised the manuscript of the article throughout. S.X. and Z.H. put forward valuable opinions on the writing of the article. All authors have read and approved the article.

Peer review

Peer review information

Nature Communications thanks the anonymous reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Data availability

All data generated or analyzed during this study are included in this published article and its supplementary information files (Supplementary Information file and Source Data file), and are available from the corresponding authors (P.Z., J.P., F.Y., or X.L.).  are provided with this paper.

Competing interests

The authors declare no competing interests.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1038/s41467-025-62645-6.

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

1. Smith, E; Diaz-Torne, C; Perez-Ruiz, F; March, L. Epidemiology of gout: an update. Best. Pract. Res. Clin. Rheumatol.; 2010; 24, pp. 811-827. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/21665128]

2. Lin, S et al. Ganoderma lucidum polysaccharide peptide alleviates hyperuricemia by regulating adenosine deaminase and urate transporters. Food Funct.; 2022; 13, pp. 12619-12631. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36385640]

3. Yanai, H; Adachi, H; Hakoshima, M; Katsuyama, H. Molecular biological and clinical understanding of the pathophysiology and treatments of hyperuricemia and its association with metabolic syndrome, cardiovascular diseases and chronic kidney disease. Int. J. Mol. Sci.; 2021; 22, 9221. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34502127][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8431537]

4. Kuwabara, M et al. Update in uric acid, hypertension, and cardiovascular diseases. Hypertens. Res.; 2023; 46, pp. 1714-1726. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37072573]

5. Kuwabara, M et al. Exploring the multifaceted nexus of uric acid and health: a review of recent studies on diverse diseases. Biomolecules; 2023; 13, 1519. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37892201][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10604821]

6. Li, X; Liu, J; Ma, L; Fu, P. Pharmacological urate-lowering approaches in chronic kidney disease. Eur. J. Med. Chem.; 2019; 166, pp. 186-196.2019pso.book...L [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30769179]

7. Peng, X et al. Gout therapeutics and drug delivery. J. Control Release; 2023; 362, pp. 728-754. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37690697]

8. Kumar, N; Kaur, K; Kaur, N; Singh, E; Bedi, P. Pathology, target discovery, and the evolution of XO inhibitors from the first discovery to recent advances (2020-2023). Bioorg. Chem.; 2024; 143, 107042. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38118298]

9. Shin, HJ et al. Interactions of urate transporter URAT1 in human kidney with uricosuric drugs. Nephrology; 2011; 16, pp. 156-162. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/21272127]

10. Dai, Y; Lee, CH. Transport mechanism and structural pharmacology of human urate transporter URAT1. Cell Res; 2024; 34, pp. 776-787. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/39245778][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11528023]

11. Tasic, V et al. Clinical and functional characterization of URAT1 variants. PloS one; 2011; 6, e28641.2011PLoSO..628641T [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/22194875][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3241677]

12. Waller, A; Jordan, KM. Use of febuxostat in the management of gout in the United Kingdom. Ther. Adv. Musculoskelet. Dis.; 2017; 9, pp. 55-64. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28255339]

13. Seth, R; Kydd, AS; Buchbinder, R; Bombardier, C; Edwards, CJ. Allopurinol for chronic gout. Cochrane Database Syst. Rev.; 2014; 2014, CD006077. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25314636][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8915170]

14. Lee, M-HH; Graham, GG; Williams, KM; Day, RO. A benefit-risk assessment of benzbromarone in the treatment of gout: was its withdrawal from the market in the best interest of patients?. Drug Saf.; 2008; 31, pp. 643-665. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/18636784]

15. Miner, JN et al. Lesinurad, a novel, oral compound for gout, acts to decrease serum uric acid through inhibition of urate transporters in the kidney. Arthritis Res. Ther.; 2016; 18, pp. 1-10.

16. Deeks, ED. Lesinurad: a review in hyperuricaemia of gout. Drugs Aging; 2017; 34, pp. 401-410. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28425024]

17. Liu, YR; Wang, JQ; Li, J. Role of NLRP3 in the pathogenesis and treatment of gout arthritis. Front. Immunol.; 2013; 14, 1137822.

18. Tian, Y et al. Recent advances in the treatment of gout with NLRP3 inflammasome inhibitors. Bioorg. Med. Chem.; 2024; 112, 117874. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/39167977]

19. Coll, RC et al. MCC950 directly targets the NLRP3 ATP-hydrolysis motif for inflammasome inhibition. Nat. Chem. Biol.; 2019; 15, pp. 556-559. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31086327]

20. Luo, Z; Yin, F; Wang, X; Kong, L. Progress in approved drugs from natural product resources. Chin. J. Nat. Med.; 2024; 22, pp. 195-211. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38553188]

21. Zhang, Y et al. Anti-inflammatory canthin-6-one alkaloids from the roots of Thailand Eurycoma longifolia Jack. J. Nat. Med.; 2020; 74, pp. 804-810. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32638295]

22. Ye, G et al. A quassinoid diterpenoid Eurycomanone from Eurycoma longifolia Jack exerts anti-cancer effect through autophagy inhibition. Molecules; 2022; 27, 4398. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35889271][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9324291]

23. Tsai, CH et al. The powdered root of Eurycoma longifolia Jack improves beta-cell number and pancreatic islet performance through Pdx1 induction and shows antihyperglycemic activity in db/db mice. Nutrients; 2020; 12, 2111. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32708678][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7400842]

24. Rehman, SU; Choe, K; Yoo, HH. Review on a traditional herbal medicine, Eurycoma longifolia Jack (Tongkat Ali): Its traditional uses, chemistry, evidence-based pharmacology and toxicology. Molecules; 2016; 21, 331. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/26978330][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6274257]

25. Peng, L et al. Experimental study on the anti-hyperuricemia effect of different extracts of Tongkat Ali. China Pharm.; 2018; 21, pp. 224-227.

26. Bao, R et al. Effect of eurycoma longifolia stem extract on uric acid excretion in hyperuricemia mice. Front. Pharmacol.; 2019; 10, 1464. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31920654][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6914847]

27. Bao, R et al. Eurycomanol alleviates hyperuricemia by promoting uric acid excretion and reducing purine synthesis. Phytomedicine; 2022; 96, [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34785103]153850.

28. Moffat, JG; Vincent, F; Lee, JA; Eder, J; Prunotto, M. Opportunities and challenges in phenotypic drug discovery: an industry perspective. Nat. Rev. Drug Discov.; 2017; 16, pp. 531-543. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/28685762]

29. Wang, F et al. A novel SIK2 inhibitor SIC-19 exhibits synthetic lethality with PARP inhibitors in ovarian cancer. Drug Resist Updat.; 2024; 74, 101077. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/38518726]

30. Liu, Y; Lai, L; Ju, Y; Liu, C; Meng, D. Chemical constituents and synergistic anti-gout studies on Eurycoma longifolia and potential mechanisms evaluation based on systemic analysis approach. Bioorg. Chem.; 2019; 92, 103302. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/31634989]

31. Zhao, T et al. Novel human urate transporter 1 inhibitors as hypouricemic drug candidates with favorable druggability. J. Med. Chem.; 2020; 63, pp. 10829-10854. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/32897699]

32. Hochheiser, IV et al. Structure of the NLRP3 decamer bound to the cytokine release inhibitor CRID3. Nature; 2022; 604, pp. 184-189.2022Natur.604.184H [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35114687]

33. Mangan, MS et al. Targeting the NLRP3 inflammasome in inflammatory diseases. Nat. Rev. Drug Discov.; 2018; 17, pp. 588-606. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/30026524]

34. Dekker, C et al. Crystal structure of NLRP3 NACHT domain with an inhibitor defines mechanism of inflammasome inhibition. J. Mol. Biol.; 2021; 433, [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/34687713]167309.

35. Shi, X et al. Novel urate transporter 1 (URAT1) inhibitors: a review of recent patent literature (2020–present). Expert Opin. Ther. Pat.; 2022; 32, pp. 1175-1184. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36625031]

36. Wu, C et al. Molecular mechanisms of urate transport by the native human URAT1 and its inhibition by anti-gout drugs. Cell Discov.; 2025; 11, 33. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/40169562][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11962085]

37. Olaru, A; Bala, C; Jaffrezic-Renault, N; Aboul-Enein, HY. Surface plasmon resonance (SPR) biosensors in pharmaceutical analysis. Crit. Rev. Anal. Chem.; 2015; 45, pp. 97-105. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25558771]

38. Shi, C et al. Discovery of NLRP3 inhibitors using machine learning: Identification of a hit compound to treat NLRP3 activation-driven diseases. Eur. J. Med. Chem.; 2023; 260, 115784. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/37672931]

39. Bibert, S et al. Mouse GLUT9: evidences for a urate uniporter. Am. J. Physiol. Ren. Physiol.; 2009; 297, pp. F612-F619.

40. Le, MT; Shafiu, M; Mu, W; Johnson, RJ. SLC2A9—a fructose transporter identified as a novel uric acid transporter. Nephrol. Dial. Transplant.; 2008; 23, pp. 2746-2749. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/18606621][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2734172]

41. Yang, B et al. New insight into the management of renal excretion and hyperuricemia: Potential therapeutic strategies with natural bioactive compounds. Front. Pharmacol.; 2022; 13, 1026246. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/36483739][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9723165]

42. Nigam, SK et al. The organic anion transporter (OAT) family: a systems biology perspective. Physiol. Rev.; 2015; 95, pp. 83-123. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/25540139][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4281586]

43. Chai Discovery team, et al. Chai-1: decoding the molecular interactions of life. BioRxiv. (2024).

44. Zhao, T et al. Discovery of novel bicyclic imidazolopyridine-containing human urate transporter 1 inhibitors as hypouricemic drug candidates with improved efficacy and favorable druggability. J. Med. Chem.; 2022; 65, pp. 4218-4237. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/35084182]

45. Li, F et al. Clinical classification of hyperuricemia in patients with chronic kidney disease. Int Urol. Nephrol.; 2021; 53, pp. 1665-1674. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/33387225]

46. So, A; Thorens, B. Uric acid transport and disease. J. Clin. Invest.; 2010; 120, pp. 1791-1799. [PubMed: https://www.ncbi.nlm.nih.gov/pubmed/20516647][PubMedCentral: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2877959]

47. Bowers, K. J. et al. Scalable algorithms for molecular dynamics simulations on commodity clusters. In Proc. ACM/IEEE Conference on Supercomputing (SC06), 11-17 (IEEE, 2006).