Discovery of Novel NLRP3 Inflammasome Inhibitors Composed of an Oxazole Scaffold Bearing an Acylsulfamide

Abstract

The NLRP3 inflammasome plays an important role in the defense mechanism of the innate immune system and has recently attracted much attention as a drug target for various inflammatory disorders. Among the strategies for generating the novel chemotype in current drug discovery, scaffold hopping and bioisosteric replacement are known to be attractive approaches. As the results of our medicinal chemistry campaign, which involved exploration of core motifs using a ring closing approach, a five-membered oxazole-based scaffold was identified, and subsequent implementation of bioisosteric replacement led to discovery of a novel chemical class of NLRP3 inflammasome inhibitor bearing the acylsulfamide group. Further optimization of aniline and sulfamide moieties to improve potency in human whole blood assay led to the identification of the orally bioactive compound 32 in the LPS challenge model. Furthermore, compound 32 attenuated kidney injury in adriamycin-induced glomerulonephritis in mice. These investigations indicated that the NLRP3 inhibitor could be a potential therapeutic agent for glomerulonephritis.

Pathogens and tissue damages are mainly sensed by tissue resident cells, including innate immune cells as well as nonimmune cells such as tubular epithelial cells. The NOD-like receptor, leucine-rich repeat and pyrin domain-containing protein 3 (NLRP3) is an intracellular sensor that detect a broad range of pathogens, environmental irritants, and tissue damage-derived factors.¹ Upon activation, NLRP3 engages pro-caspase-1 through apoptosis-associated specklike protein containing a carboxyterminal CARD (ASC) to form NLRP3 inflammasome. This results in activation of caspase-1, which cleaves the proinflammatory cytokines pro-IL-1β and pro-IL-18 to their active forms.¹ Activation of the NLRP3 inflammasome has been implicated in various inflammatory disease including gout, ulcerative colitis, rheumatoid arthritis, systemic lupus erythematosus, and glomerulonephritis.²⁻⁴ Therefore, the NLRP3 inflammasome has recently attracted much attention as a drug target molecule for the development of novel therapeutic agents to treat the aforementioned conditions.^5,6

In 2001, a Pfizer group first reported that diarylsulfonyl urea derivatives potently suppressed IL-1β production in lipopolysaccharide (LPS)-activated monocytes and macrophages.⁷

Subsequent detailed research regarding the mechanism of action revealed that CP-456,773 (CRID3, MCC950), composed of diarylsulfonyl urea, inhibited activation of the NLRP3 inflammasome followed by reduced IL-1β in pharmacological in vitro experiments and in vivo model studies in mice.^8,9 After those pioneering research efforts were reported, development of various NLRP3 inflammasome inhibitors have been enthusiastically investigated.¹⁰⁻¹⁵ In Figure 1, the chemical structures of representative compounds that act as NLRP3 inhibitors are illustrated. In this report, we describe our medicinal chemistry efforts to identify novel NLRP3 inflammasome inhibitors and their therapeutic potential for glomerulonephritis.

Figure 1.

Structures of representative compounds as NLRP3 inflammasome inhibitors.

In the current drug discovery program for reducing the high attrition rates,¹⁶ explorations of novel chemical classes are effective for improving bioactivity and pharmacokinetic properties and to mitigate unexpected outcomes such as a critical toxicity derived from the specific chemical structures. Scaffold hopping is a powerful approach to discover distinct chemotypes.¹⁷⁻¹⁹ To pursue novel chemical classes from the reported NLRP3 inhibitors, we focused on compound 1, which had a flexible chemical structure composed of a urea core structure accompanying an amino acid ester unit. Among the categories classified in scaffold hopping,¹⁷ the ring closing approach was applied because it seemed to be suitable for the investigation of core structure that can be effectively placed the accompanying substituents for suitable interaction with the NLRP3 proteins. Therefore, we started our investigation on compound 4, an analogue of compound 1, which served as the reference compound (shown in Table 1) for the discovery of a novel NLRP3 inhibitor.

Table 1. SAR Exploration of the Core Motif.

^aThe values of IC₅₀ are mean values determined from four replicates. CP-456,773 was used as a positive control in each assay and gave IC₅₀ = 0.060 ± 0.049 (mean ± SD) μM, n = 163.

We conducted a SAR exploration of the core motif using a ring closing approach starting from compound 4. We evaluated the NLRP3 inflammasome inhibitory activity of the test compounds by measuring the IL-1β production in LPS/nigericin treated THP-1 cells. Several types of cyclization at the α-position carbon of the ester adjacent to the urea unit and cyclization in the urea unit were performed (Table 1). First, compounds 5 and 6, which were closed at the α-carbon of the ester adjacent to urea’s nitrogen to form a five-membered ring (type a), showed less potency relative to reference compound 4. Compound 7, bearing the imidazolidine-dione in the core motif (type b) also exhibited no IL-1β inhibitory effect. Subsequently, five-membered heteroazole derivatives were synthesized by cyclization at the carbon in the α-position relative to the carbonyl group of urea (type c). We found that oxazole analogues (9 and 10) exhibited IL-1β inhibitory activities at acceptable levels compared to reference compound 4. On the other hands, the other 5- and 6-membered heterocyclic analogues showed insufficient or substantially decreased IL-1β inhibitory activities (8, 11–13).

At this point, with the oxazole-based compounds (9 and 10) in hand, we considered that the chemical stability of these compounds was insufficient due to the ease of hydrolysis of the ester group. On the other hand, compound 14, the carboxylic acid form, also exhibited IL-1β inhibitory activity, suggesting that the ester group functions as a prodrug of the carboxylic acid, transforming to the active form in the target cell.¹⁰ Accordingly, bioisosteric replacement²⁰ of the carboxylic acid was performed to explore the more potent NLRP3 inhibitory activity (Table 2). Although classical replacements of carboxylic acid to amide groups (15 and 16) led to reduced IL-1β inhibitory activities, introduction of the acylsulfonamide (17) afforded moderate potency. As a result of further modification, compound 18 incorporating the acylsulfamide group provided significant enhancement of IL-1β inhibitory activity. Because a decrease of potency was observed in compound 19 in which the NH bond of the sulfamide group was blocked, we considered that the hydrogen bond acceptor (HBA) of the sulfonyl substituent contributed to NLRP3 inhibitory activity. The application of a sulfamide group to the oxazole isomer afforded compound 20 ,which showed potent IL-1β potency comparable with compound 18. Based on bioisosteric replacement, we have thus successfully obtained a novel chemotype for NLRP3 inhibitors composed of oxazole-based scaffolds bearing an acylsulfamide group (18 and 20). Before proceeding to further optimization, physicochemical properties such as solubility and Caco-2 membrane permeability between the two compounds (16 and 18) were compared. While the solubilities of both compounds were comparably high, compound 20 gave better results with respect to Caco-2 membrane permeability due to increased lipophilicity by the internal hydrogen bond interaction between the nitrogen atom in the oxazole and the NH bond in the acylsulfamide group.²¹ Accordingly, compound 20 with favorable physicochemical properties was selected for further optimization to improve the IL-1β inhibitory activity in THP-1 cells and human whole blood.

Table 2. SAR Exploration of the Carboxylic Acid Group.

^aThe values of IC₅₀ are mean values determined from four replicates. CP-456,773 was used as a positive control in each assay and gave IC₅₀ = 0.060 ± 0.049 (mean ± SD) μM, n = 163.

^bFaSSIF

^cCaco-2 membrane permeability (10^–6cm/s).

^dNot tested.

As a first attempt, we undertook an SAR exploration of the highly lipophilic aniline group in compound 20 (Table 3). Compounds 21 and 22, which were modified with methyl groups for the purpose of reduction of lipophilicity showed slight decreases in IL-1β inhibitory activities. Introduction of an i-Pr substituent at the ortho-position did not lead to an improvement of IL-1β inhibitory activity (23). Therefore, we next investigated the effect of substituents in the meta-position with respect to compound 20, which exhibited moderate potency along with low lipophilicity. We observed enhanced IL-1β inhibition by the installation of alkyl substituents, and compound 25 showed potent IL-1β inhibitory activity. However, the IC₅₀ value of compound 25 in the whole blood assay was not improved. In further modification, compound 27 bearing a trifluoromethyl group exhibited enhancement of potency in the whole blood assay. This improvement of potency in whole blood assay was presumably considered to be reduction of human protein binding (hPB) (20: >99.9%; 25: 99.9%; 27: 99.6%). On the other hand, the conversion of benzene to pyridine for the purpose of reduction of hPB did not improve the potency in whole blood assay (28).

Table 3. SAR Exploration of the Aniline Group.

^aThe values of IC₅₀ are mean values determined from four replicates. CP-456,773 was used as a positive control in each assay and gave IC₅₀ = 0.060 ± 0.049 (mean ± SD) μM, n = 163.

^bThe values of IC₅₀ are mean values determined from two donors. CP-456,773 was used as a positive control in each assay and gave IC₅₀ = 1.3 ± 0.4 (mean ± SD) μM, n = 192.

Since we found that the aniline substituent reduced lipophilicity in our chemotype, we conducted an exploration of the ability of amines in the acylsulfamide group to improve potency in whole blood assay along with reducing hPB (Table 4). The compounds 29 and 30, which incorporated branched and cyclic amines, showed potent IL-1β inhibitory activity. The hPB of those compounds were increased due to slightly increased LogD, and as expected, the potency of those compounds in the whole blood assay was decreased substantially. In an attempt to reduce the level of hPB, the incorporation of polar cyclic amines was implemented. Compound 31 possessing a morpholine unit showed potent IL-1β inhibitory activity with lower LogD, but the expected reduction of hPB was not observed. On the other hand, the compounds bearing 7-membered polar cyclic amines such as a 1,4-oxazepane and the substituted 1,4-diazepane unit (32–34) exhibited significant improvement in whole blood assay activities. It was suggested that installation of those polar cyclic amines was effective in reducing the human protein binding ratio, resulting in an increase in the free fraction of compound with potent IL-1β inhibitory activity. Among those compounds, compound 32, which showed the most potent inhibitory activity in whole blood assay, was successfully identified in our present medicinal chemistry campaign.

Table 4. SAR Exploration of Amines in the Acylsulfamide Group.

^aThe values of IC₅₀ are mean values determined from four replicates. CP-456,773 was used as a positive control in each assay and gave IC₅₀ = 0.060 ± 0.049 (mean ± SD) μM, n = 163.

^bThe values of IC₅₀ are mean values determined from two donors. CP-456,773 was used as a positive control in each assay and gave IC₅₀ = 1.3 ± 0.4 (mean ± SD) μM, n = 192.

^cHuman protein binding.

To confirm target engagement of the novel discovered NLRP3 inhibitor 32, X-ray cocrystal analysis was implemented, which revealed that compound 32 directly binds to the NACHT domain of human NLRP3 (Figure 2). According to the structure, the NH bond of the aniline group and the nitrogen atom in the oxazole scaffold formed a strong hydrogen bond with Ala228 (2.8 Å) in the NLRP3 protein, and the acylsulfamide group was found to form hydrogen bonds at two locations, Arg351 (2.6 Å) and Val353 (3.0 Å). Those interactions in this chemotype are uniquely distinct from the interaction of the sulfonyl urea derivatives including CP-456,773 with Arg578 in NLRP3 NACHT domain.²² The lipophilic substituents in the aniline moiety showed van der Waals interactions in the lipophilic region at the binding pocket, and the 1,4-oxazepane unit was exposed to the solvent region. It was presumed that the potent NLRP3 inhibitory activity of compound 32 was attributed to those multiple unique interactions.

Figure 2.

X-ray structure of compound 32 in the human NLRP3 NACHT domain. (PDB 8WSM) Hydrogen bonds are depicted as dashed lines (yellow).

The profiles of the representative compound 32 are shown in Table 5. The results showed that compound 32 suppressed both NLRP3-dependent IL-1β production, and inflammatory cell death known as pyroptosis in THP-1 cells with a similar potency. Furthermore, compound 32 exhibited high selectivity against NLRC4 which was a member of the NLRP family. Physicochemical properties such as solubility and Caco-2 membrane permeability were generally favorable. Regarding the pharmacokinetic profiles, moderate metabolic stability was observed in mouse, whereas compound 32 demonstrated highly metabolic stabilities in human, rat, and dog. Those preferable characteristics reflected the favorable PK profiles so that low total clearance with high oral bioavailability were confirmed in rat. Furthermore, evaluation of CYP inhibition indicated that such an off-target risk of compound 32 seemed to be low.

Table 5. In Vitro and in Vivo PK Profiles of Compound 32.

IC50 (μM)/THP-1			solubility	Caco-2	protein binding
NLRP3			(μM)	Papp	(%)
IL-1β	pyroptosis	NLRC4	FaSSIF	(cm−6/s)	human	rat	mouse
0.021	0.028	>30	444	27.8	99.1	96.6	93.8

metabolic stability				rat PK						CYPc
(% of remaining 60 min)				iv (0.3 mg/kg)a			po (1 mg/kg)b			3A4
human	rat	mouse	dog	MRTd (h)	CLtote (L/h/kg)	Vdssf (L/kg)	Cmax (μM)	MRTd (h)	Fg (%)	IC50 (μM)
71.2	90.3	45.5	90.0	4.7	0.06	0.28	3.8	6.3	121	>10

^aIV administration at corresponding dose in DMSO solution.

^bPO administration at corresponding dose in propylene glycol solution.

^cCYP inhibition.

^dMean residence time.

^eTotal body clearance.

^fVolume of distribution at steady state.

^gBioavailability.

Next, compound 32 was advanced to in vivo animal testing. A robust increase of plasma IL-1β was observed in LPS challenge mice model, and oral treatment with compound 32 resulted in suppression of plasma IL-1β in a dose related manner (Figure 3a). To evaluate the therapeutic potential of compound 32 on glomerulonephritis, compound 32 was administered orally for 2 weeks in mice with adriamycin-induced glomerulonephritis.^23,24 We found that treatment of the mice model with compound 32 improved the urine albumin-to-creatinine ratio (UACR) in a dose related manner (Figure 3b). These data confirmed that NLRP3 inhibition was effective against glomerulonephritis.

Figure 3.

(a) Effect of compound 32 on the plasma IL-1β levels in the LPS challenge in murine (n = 4). (b) Effect of compound 32 on UACR in adriamycin-induced glomerulonephritis mice (n = 5–10). Data are presented as mean ± SD *p < 0.05 **p < 0.01 (Dunnett’s test or Steel test) versus vehicle group.

The synthetic pathway toward the NLRP3 inhibitor 32 was shown in Scheme 1 (Synthetic procedures of other compounds investigated in this report are described in Supporting Information). 1,4-Oxazepane HCl 35 was converted to sulfonylurea 36, followed by deprotection of the Boc group under acidic conditions to furnish sulfonamide intermediate 37. Regarding the oxazole scaffold, commercially available 2-bromo oxazole 4-ethyl ester 38 and 2-methyl-5-trifluoroaniline were coupled by a palladium catalyst to obtain 39. After Boc protection of the NH group to afford 40, the corresponding carboxylic acid 41 was prepared by hydrolysis of the ester group under basic condition. After coupling of 37 and 41 by CDI to obtain 42, finally deprotection of the Boc group was facilitated by TFA to give compound 32.

Scheme 1. Synthesis of Compound 32.

Reagents and conditions: (a) chlorosulfonyl isocyanate, t-BuOH, Et₃N, CHCl₃, 0 °C to rt, 79%; (b) 4 N HCl in EtOAc, rt, 97%; (c) 2-methyl-5-(trifluoromethyl)aniline, t-BuBrettPhos-Pd-G3, K₃PO₄, t-BuOH, 100 °C, 56%; (d) (Boc)₂O, DMAP, THF, rt, 99%; (e) 2 N NaOH aq, MeOH, rt, 86%; (f) 37, CDI, DBU, THF, 60 °C, 43%; (g) TFA, rt, 80%.

In summary, we report herein our medicinal chemistry efforts to discover a potent and orally bioavailable oxazole based NLRP3 inflammasome inhibitor bearing an acylsulfamide group. To explore a distinct chemical class, we demonstrated a ring closing approach and the effects of bioisosteric replacement. Subsequent optimization of substituents bearing aniline and introduction of polar cyclic amines into the acylsulfamide group to reduce protein binding resulted in a significant improvement in the potency in the whole blood assay. Finally, we identified compound 32, which was an orally bioavailable, potent, and selective NLRP3 inhibitor. Furthermore, compound 32 improved kidney injury in adriamycin-induced glomerulonephritis in mice. These investigations, obtained by profiling compound 32, suggest that the use of small molecules to inhibit the NLRP3 inflammasome may serve as a potential pathway for the treatment of glomerulonephritis as well as other inflammatory diseases involving NLRP3 inflammasomes.

Glossary

Abbreviations

CDI

1,1-carbonyldiimidazol

CYP

cytochrome P450

DBU

1,8-diazabicyclo[5.4.0]undec-7-ene

DMAP

4-dimethylaminopyridine

FaSSIF

fasted state simulated intestinal fluid

IL-1β

interleukin 1 beta

IL-18

interleukin 18

LPS

lipopolysaccharide

NLRP3NOD-like receptor

NOD-like receptorLeucine-rich repeat and pyrin domain containing protein 3

NLRC4

NLR family CARD domain containing 4

PK

pharmacokinetics

SAR

structure–activity relationship

TFA

trifluoroacetic acid

THF

tetrahydrofuran

UACR

urine albumin-to-creatinine ratio.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsmedchemlett.3c00433.

• Synthetic schemes, procedures, experimental data, assay procedures (PDF)

The authors declare no competing financial interest.