Development of Keap1‐Nrf2 Protein–Protein Interaction Inhibitor Activating Intracellular Nrf2 Based on the Naphthalene‐2‐acetamide Scaffold, and its Anti‐Inflammatory Effects

Abstract

Nuclear factor erythroid 2‐related factor 2 (Nrf2) and Kelch‐like ECH‐associated protein 1 (Keap1) axis is an attractive therapeutic target for various intractable diseases. Although protein–protein interaction inhibitors against Keap1‐Nrf2 have been developed over the past decade, more structural expansion is needed to improve efficacy. In this article, several candidate compounds are designed and synthesized as novel Nrf2 activators and their intracellular Nrf2‐activating effects are evaluated. Among the synthesized compounds, a novel naphthalene‐1,4‐(4‐ethoxybenzensulfonamide) bearing a tertiary acetamide side chain at the 2‐position strongly activated intracellular Nrf2. Particularly, the pyrrolidine‐type acetamide compound showed the strongest intracellular Nrf2 activation. X‐ray cocrystallography revealed that this compound can bind to the DC domain of Keap1. Additionally, the pyrrolidine‐type acetamide compound induced the mRNA expression of the representative Nrf2 target genes heme oxygenase‐1 and NAD(P)H:quinone oxidoreductase 1. Moreover, the compound exhibited anti‐inflammatory effects in a lipopolysaccharide‐stimulated macrophage cell line. Conclusively, these results suggest that the pyrrolidine‐type naphthalene‐2‐acetamide is a promising compound for the development of Nrf2 activators that can be applied to treat inflammatory diseases.

Pyrrolidine‐type naphthalene‐2‐acetamide compound 5i inhibits Kelch‐like ECH‐associated protein 1 (Keap1)‐nuclear factor erythroid 2‐related factor 2 (Nrf2) protein–protein interaction by binding to the Keap1‐DC domain; it subsequently activates intracellular Nrf2 and induces Nrf2‐target genes. Furthermore, it shows anti‐inflammatory effects in LPS‐treated macrophages.

1. Introduction

Living organisms are continuously exposed to harmful agents, such as reactive oxygen species (ROS) and environmental stressors. Nuclear factor erythroid 2‐related factor 2 (Nrf2) and Kelch‐like ECH‐associated protein 1 (Keap1) systems function as fundamental and highly conserved biological defense mechanisms against these stressors, particularly in higher organisms, including humans.^{[ 1} , ^{2 – 3 ]}

Nrf2 is a transcription factor that may be related to resistance against oxidative and electrophilic stresses, anti‐inflammatory responses, and metabolic reprograming.^{[ 4 , 5 ]} Notably, the transcriptional activity of Nrf2 is regulated by Keap1, an adaptor protein for Cullin 3‐based ubiquitin E3 ligase.^{[ 6 ]} Under basal conditions, Nrf2 is controlled through its ETGE and DLG motifs by Keap1, forming a complex that is ubiquitinated and subsequently degraded by the 26S proteasome.^{[ 7 ]} However, Keap1 detects chemical aberrations via the modification of thiol groups on specific cysteine residues under conditions of significant stress.^{[ 8 ]} This modification suppresses proteasomal recognition of the Keap1‐Nrf2 complex, leading to the accumulation of free Nrf2. Stabilized Nrf2 translocates to the nucleus, where it binds to the antioxidant response element (ARE) as a heterodimer with small musculoaponeurotic fibrosarcoma oncogene homolog proteins (sMaf).^{[ 8 ]} This complex induces the expression of various cellular defense and proliferative proteins, including heme oxygenase‐1 (HO‐1), NAD(P)H:quinone oxidoreductase‐1 (NQO1), multidrug resistance proteins, and enzymes associated with the pentose phosphate pathway.^{[ 9 ]}

Owing to its critical role in maintaining cellular homeostasis, Nrf2 is an attractive therapeutic target for various challenging disorders, such as chronic kidney disease,^{[ 10 ]} pulmonary arterial hypertension,^{[ 11 ]} and Alzheimer's disease.^{[ 12 , 13 ]} Compounds containing Michael acceptor moieties activate Nrf2 by covalently modifying cysteine residues on Keap1, thereby inhibiting its repressive function.^{[ 14 ]} For example, tert‐butylhydroquinone (tBHQ), a food additive used in several countries with antioxidant properties, is one of the most well‐characterized Nrf2 activators.^{[ 15 ]} Upon oxidation, tBHQ is readily converted into a benzoquinone‐like structure that functions as an electrophilic Michael acceptor, enabling covalent interaction with Keap1. Another example is bardoxolone methyl (CDDO‐Me), a synthetic triterpenoid with an unsaturated ketone moiety, and an orally available Nrf2 activator developed as a first‐in‐class drug for treating diabetic kidney disease and several rare diseases.^{[ 16} , ^{17 – 18 ]} Although CDDO‐Me was screened out in a phase III study, omaveloxolone (Skyclarys), a CDDO‐Me analog, has been approved as an orphan drug for Friedreich ataxia.^{[ 19 ]} Additionally, dimethyl fumarate (Tecfidera) is approved as a therapeutic agent for multiple sclerosis,^{[ 18 ]} however, it is debatable whether its mechanism of action is solely through Nrf2 activation.^{[ 18 , 20 ]}

Covalent drugs could potentially possess certain risk factors, such as off‐target effects.^{[ 21 ]} Recently, noncovalent protein–protein interaction (PPI) inhibitors have attracted attention as more selective Nrf2 activators.^{[ 21 , 22 ]} Naphthalene‐1,4‐bis(4‐methoxybenzensulfonamide) (Figure  1 ), also known as Cpd16, was identified in 2013 as a canonical noncovalent Keap1‐Nrf2 PPI inhibitor.^{[ 23 ]} It has been shown to interacts with the double glycine repeat domain and C‐terminal region (DC) domain on Keap1 and exerts its therapeutic effects by activating the transcriptional activity of Nrf2 in cultured cells.^{[ 23 , 24 ]} Additionally, naphthalene‐1,4‐bis(4‐methoxybenzensulfonamide)‐N,N′‐diacetate derivatives (e.g. CPUY192018, Figure 1) are among the most potent inhibitors of this class in vitro, and numerous derivatives have been developed.^{[ 25 , 26 ]} Moreover, several Keap1‐Nrf2 PPI inhibitors without the naphthalene skeleton have been developed.^{[ 27} , ^{28 – 29 ]} Specifically, non‐naphthalene type inhibitors, such as KI696 (Figure 1), exhibiting strong PPI inhibitory activity.^{[ 30 ]}

Figure 1.

Known Nrf2 activators.

In a previous study, we identified 2‐acetonylnaphathelene‐1,4‐bis(4‐etoxybenzenesulfonyamide), named K67 (Figure 1), as an inhibitor of PPI between Keap1 and p62/sequestosome 1 (p62).^{[ 5 ]} p62 is another protein that interacts with Keap1 and has Nrf2 activating properties. Based on the structure of K67, various 2‐substitued naphthalene derivatives, such as p62‐Keap1 and Nrf2‐Keap1 PPI inhibitors, have been developed.^{[ 31 ]} Among the synthesized compounds, naphathelene‐1,4‐bis(4‐ethoxybenzenesulfonyamide), bearing a butyrate unit named KMN003 (Figure 1), showed a half‐maximal inhibitory concentration (IC₅₀) in the submicromolar order in a fluorescence polarization (FP) assay.^{[ 31 ]} Upon structural expansion, naphthalene derivatives bearing an N,N‐dimethylacetamide structure at the C‐2 position (Figure 1) exhibited potent intracellular Nrf2‐inducing activity in reporter gene assays. Based on these results, 2‐substituted naphthalene derivatives appeared to be the preferred core structures for Nrf2 activators.^{[ 32 ]}

In this study, we designed and synthesized several candidate compounds as novel Nrf2 activators, primarily naphthalene‐2‐acetamide derivatives with cyclic structures and evaluated their intracellular Nrf2‐activating effects. Moreover, detailed analyses, such as X‐ray cocrystallography, anti‐inflammatory experiments, and an absorption, distribution, metabolism, excretion, and toxicity (ADMET) study, were performed.

2. Results and Discussion

2.1. Design and Chemistry

Several research efforts have been made to introduce side chains onto naphthalene 1,4‐bissulfonamide derivatives to increase their biological activities. In this study, the acetamide side chain was introduced to design various secondary and tertiary acetamide derivatives, using a relatively facile synthetic method (5a–5k, Scheme).

Notably, the compounds evaluated in this study were synthesized using commercially available starting materials in a six‐step process (Scheme). In the first step, 4‐nitro‐1‐naphthylamine was reduced to 1,4‐diaminonaphthalene, which was subsequently condensed with 4‐ethoxybenzenesulfonyl chloride. This intermediate (Cpd16‐OEt, Scheme) was oxidized using cerium(IV) ammonium nitrate (CAN) to yield a quinone imine‐like compound. Thereafter, dimethyl malonate was added to the oxidized intermediate and ester hydrolysis with decarboxylation occurred to generate a naphthalene‐2‐acetic acid derivative. Finally, the resulting acetate was reacted with the corresponding amine in the presence of 1‐(3‐dimethylaminopropyl)‐3‐ethylcarbodiimide hydrochloride (EDC•HCl), leading to the formation of the desired naphthalene‐1,4‐bissulfonamide derivatives bearing 2‐acetamide groups in moderate yield.

2.2. Screening of Intracellular Nrf2 Activating Compound

In this study, we used a HEK293‐ARE‐Luc reporter cell line (luciferase gene reporter assay) to screen the phenotypic intracellular Nrf2 activation (ARE stimulation) effects of the synthesized acetamide compounds. For comparison, the known Nrf2 activator tBHQ, a well‐characterized covalent Nrf2 activator, was included as a reference compound. Although the mechanism of action of tBHQ differs from that of the PPI inhibitors, it was selected as a positive control due to its robust and reproducible Nrf2‐activating potency, facilitating comparative assessment.

At a concentration of 10 μM, tBHQ increased luminescence intensity by approximately tenfold compared with that in the control group (Figure  2A). Among the naphthalene‐type inhibitors, Cpd16‐OEt, which lacked a side chain at the C‐2 position, enhanced luminescence by fivefold compared with that in the control. Additionally, some of the C‐2 substituted naphthalene derivatives demonstrated a stronger effect than Cpd16‐OEt. For example, the dimethyl acetamide compound 5c enhanced Nrf2 activity by tenfold compared with that in the control. However, all secondary amides bearing cyclic alkanes (5e–5g), except the cyclopropyl derivative 5d, failed to increase Nrf2 activity. Additionally, N‐phenylamide (5h) had no effect on Nrf2 activity. Overall, these results suggest that bulky secondary amide side chains decrease cellular activity.

Figure 2.

HEK293‐based cellular assays. A) Comparison of intracellular Nrf2 activating effect. The cell lines were treated with the test sample (10 μM) for 24 h. Nrf2 activity is expressed as fold change in luminescence intensity compared with the control group (n = 3). B) Dose‐dependency of the intracellular Nrf2 activating effect. The cell lines were treated with vary concentrations of the test sample (1, 2.5, 5, 10 μM) for 24 h (n = 3). C) Cytotoxicity of representative compounds at 50 μM (WST‐8 assay, n = 3). No Significant differences between control and each test compound groups were observed using Student's t‐test (n = 3).

In contrast, pyrrolidine (5i), piperidine (5j), and morpholine (5k) derivatives exhibited notable Nrf2 activation effects (Figure 2A). Among them, 5i was the strongest naphthalene derivative and showed a higher activity (approximately 13‐fold) than tBHQ (tenfold). Importantly, these derivatives were also effective at lower concentrations, with compound 5i, 5j, and 5k demonstrating notable activity at 5 μM (Figure 2B). These findings prove that tertiary acetamide is a useful substituent that enhances the intracellular Nrf2 activating effect of naphthalene‐1,4‐bissulfonamide derivatives, especially those which cyclic structures.

Additionally, the cytotoxicity of the representative compounds was evaluated using the same reporter cell line. All tested compounds exhibited no significant cytotoxicity at concentrations up to 50 μM (Figure 2C).

2.3. FP Assay

Among the newly synthesized compounds, three (5i, 5j, and 5k) that demonstrated significant intracellular Nrf2‐activating effects were further evaluated for their peptide‐based Keap1‐Nrf2 PPI inhibitory activities using an FP assay to confirm whether their mechanism of action was through PPI inhibition. As shown in Table  1 , all the tested compounds, including Cpd16‐OEt, inhibited the interaction between the Keap1‐DC domain and the ETGE‐motif fragment peptide of Nrf2, with IC₅₀ values in the micromolar range. Notably, the presence or absence of a substituent at the 2‐position had minimal impact on the IC₅₀ values of the tested compounds. Collectively, these results suggest that the FP assay may not fully correlate with pharmacological effects observed in a cellular context. Although the small sample size limits the generalizability of this conclusion, caution should be exercised when evaluating the potency of Nrf2 activators‐based solely on peptide fragment‐based assays, such as the FP assay.

Table 1.

PPI inhibitory activity and ClogP value.

Compound	Keap1‐Nrf2 PPI inhibitory activity (FP, IC50, μM)	ClogP valuea)
Cpd16‐OEt	3.57 ± 0.34	5.09
5i	1.07 ± 0.04	4.08
5j	2.43 ± 0.17	4.64
5k	1.38 ± 0.10	3.63

^a)Calculated using ChemDraw.

Importantly, the observed discrepancy between the FP assay results and intracellular Nrf2 activation suggests that factors other than direct Keap1‐Nrf2 PPI inhibition influence the pharmacological activities of these compounds. One possible explanation is the impact of physical or ADME properties, such as cell permeability, metabolic stability, and protein binding ratio, which could affect the intracellular concentration and bioavailability of the compounds. ClogP value (Table  2 ), which serve as an indicator of lipophilicity, did not correlate with the degree of intracellular Nrf2 activation. For instance, although the ClogP value of Cpd16‐OEt was one unit higher than that of compound 5i, its Nrf2 activation activity was less than half that of 5i. This discrepancy suggests that the acetamide side‐chain of the naphthalene scaffold is critical for intracellular Nrf2 activation. Additionally, alternative cellular regulatory pathways, such as p62‐mediated Nrf2 activation and other stress response mechanisms, may have contributed to the observed cellular effects.

Table 2.

In vitro ADME properties of compound 5i.

ADME properties	Values
Solubility in PBS (pH 7.4)	2.65 μM
Membrane permeability (PAMPA)	3.06 × 10−6 cm sec− 1
Metabolic stability in liver microsomes	546 mL/min/kg (7.4% in human) 2286 mL/min/kg (3.17% in mouse)
Protein binding ratio (unbound fraction)	0.0012(in human plasma)0.0070(in mouse plasma)0.0342(10% FBS)

2.4. Interaction of Keap1‐DC with Compound 5i

To investigate the interaction between Keap1‐DC and compound 5i, we performed cocrystallization experiments of the mouse Keap1‐DC/5i complex and determined the crystal structure at 1.8 Å resolution (Figure  3A, Table S1, Supporting Information, PDB ID: 9UO8). The asymmetric units of the crystals contained two molecules of Keap1‐DC (A and B), and both complexed with 5i (Figure S1A, Supporting Information). Compound 5i bound to the Nrf2‐binding site of Keap1‐DC (Figure 3A and Figure S1A, Supporting Information), thus competitively inhibiting the interaction between Keap1‐DC and Nrf2. Additionally, the overall conformations of Keap1‐DC were highly similar between molecules A and B (Figure S1B, Supporting Information, r.m.s.d. 0.109 Å). However, the binding mode of 5i differed between the two molecules (Figures 3A and 3B). Electron density maps revealed that the acetamide side chain of 5i adopted distinct orientations in molecules A and B (Figure S1C, Supporting Information), suggesting that 5i bound Keap1‐DC in the opposite orientation, likely due to the highly symmetric structure of the 5i.

Figure 3.

Structural analysis of compound 5i binding to Keap1‐DC. A) Binding modes of compound 5i in molecule A (left) and molecule B (right) of the Keap1‐DC/5i complex (PDB ID: 9UO8). Backbone of Keap1‐DC and side chains of the residues involved in the interactions with 5i are shown as a ribbon and sticks, respectively. Compound 5i is shown as sticks (purple for molecule A and blue for molecule B). Water and formate molecules are shown as red spheres and an orange stick, respectively. Hydrogen bonds are shown as dashed lines. B) Overlay of 5i (purple: molecule A; blue: molecule B) and K67 (green) in the structures of Keap1‐DC/5i and Keap1‐DC/K67 (PDB ID: 4ZY3), respectively. C) Binding mode of K67 (green) in Keap1‐DC complex. D) Superposition of the Keap1‐DC/5i complex (molecule A) with Keap1‐DC/Nrf2‐DLG (left, PDB ID: 3WN7) and Keap1‐DC/Nrf2‐ETGE (right, PDB ID: 1X2R) complexes. The Keap1‐DC/5i complex is shown in gray (Keap1‐DC) and purple (5i). The Keap1‐DC/Nrf2 complexes are shown in cyan (Keap1‐DC) and green (Nrf2 peptides). Nrf2 peptides are shown as semi‐transparent surfaces, with key residues of the DLG motif (residues 29–31) and the ETGE motif (residues 79–82) displayed as sticks.

In molecule A, the binding mode of 5i closely resembled that of the original ligand K67 (PDB ID: 4ZY3) (Figure 3A, left panel; Figure 3B and 3C).^{[ 5 ]} In the Keap1‐DC/5i structure (Figure 3A), the naphthalene ring of 5i is embedded in the central pore of Keap1‐DC and forms π‐cation stacking with the guanidium group of Arg415. One ethoxybenzene moiety engages in π–π stacking with Tyr525, while the other forms van der Waals interactions with surrounding hydrophobic residues Tyr334, Tyr572, and Phe577. The two sulfonamide groups form direct hydrogen bonds with Ser508 and Ser602, respectively. Water and formate molecules, derived from the crystallization solution, were clearly resolved in the electron density map. The two nitrogen atoms of the sulfonamide groups form water‐ and formate‐mediated hydrogen bonds with Asn414, Arg415, and Arg483.

In molecule A, the pyrrolidine‐based acetamide side chain of 5i is oriented toward the solvent region, similar to the acetonyl side chain of K67 (Figure 3A, left panel; Figure 3B and 3C),^{[ 5 ]} and is loosely recognized by Keap1‐DC, forming a water‐mediated hydrogen bond formed between its carbonyl oxygen and Ser555 (Figure 3A, left panel). In contrast, this hydrogen bond was not observed in molecule B, presumably because of the opposite orientation of 5i (Figure 3A, right panel). These observations suggest that the acetamide side chain of 5i makes only a minor contribution to binding affinity. This is consistent with the FP assay results, which showed that substitution at the 2‐position of the compound had minimal impact on the IC₅₀ (Table 1).

To assess potential steric interference with Nrf2, we superimposed the Keap1‐DC/5i complex with Keap1‐DC/Nrf2‐DLG (Figure 3D, left panel; PDB ID: 3WN7) and Keap1‐DC/Nrf2‐ETGE (Figure 3D, right panel; PDB ID: 1X2R).^{[ 33 , 34 ]} This comparison revealed multiple steric clashes, indicating that compound 5i acts as a competitive inhibitor by blocking the Nrf2‐binding interface of Keap1.

2.5. Nrf2 Accumulation and Target Gene Induction Activity of 5i

Among the newly synthesized compounds, 5i exhibited the most promising intracellular Nrf2 activation. Therefore, we evaluated its effects on Nrf2 accumulation and the expression of HO‐1 and NQO1, two key target genes of Nrf2, in the RAW264.7 cell line. Treatment with 5i increased Nrf2 accumulation in RAW264.7 cells in a concentration‐dependent manner (Figure  4A). Additionally, compound 5i significantly upregulated HO‐1 mRNA expression at 12.5 μM (Figure 4B), but downregulated the gene relative to the peak expression at higher concentrations (Figure 4B). Although a direct comparison is difficult due to the use of different cell lines, compound 5i did not exhibit cytotoxicity at concentrations up to 50 μM in HEK293‐ARE‐Luc cells (Figure 2C), suggesting that the decreased HO‐1 expression at higher concentrations in RAW264.7 cells is unlikely to be caused by cytotoxicity. Instead, these findings imply that compound 5i may exert concentration‐dependent biphasic effects on gene expression, and that there may be an optimal concentration range for maximal induction of specific Nrf2 target genes. Additionally, NQO1 mRNA expression was more responsive to compound 5i, with approximately 50‐fold increase at 12.5 µM and higher concentrations (Figure 4C). Overall, these results indicate that 5i induces nuclear Nrf2 and its target genes not only in HEK293 cells but also in other cell lines.

Figure 4.

Expression of Nrf2 and its target genes. Compound 5i induces A) Nrf2 accumulation and the mRNA expression of B) HO‐1 and C) NQO1 in RAW264.7 cells. The cells were treated with 5i for 8 h, followed by immunoblotting and RT‐PCR. Data are presented as mean ± SD. Statistical analyses were performed using GraphPad Prism 8.0.1. Significant differences between groups were determined using Student's t‐test, with statistical significance set at *p < 0.05 and ***p < 0.01, respectively (n = 3).

2.6. Anti‐Inflammatory Effect of 5i

Recent studies have demonstrated that Nrf2 exerts anti‐inflammatory effects that are independent of its canonical role in regulating oxidative stress responses.^{[ 4 ]} This has led to the emerging concept that Nrf2 activation may represent a distinct anti‐inflammatory signaling pathway, beyond its established function in cytoprotection. To explore this possibility, we investigated the anti‐inflammatory potential of compound 5i by assessing its ability to activate Nrf2 in a cellular model of inflammation induced by lipopolysaccharide (LPS). Treatment of LPS‐stimulated RAW264.7 cells with 5i led to a dose‐dependent reduction in the expression of key inflammatory markers, including nitric oxide (NO), inducible nitric oxide synthase (iNOS), tumor necrosis factor‐α (TNF‐α), and C–C motif chemokine ligand 2 (CCL2) (Figure  5 A–F).

Figure 5.

Anti‐inflammatory effect of 5i. A,B) RAW264.7 cells were pretreated with 5i for 8 h and stimulated with lipopolysaccharide (LPS; 1 μg mL^{− 1}) for 8 or 16 h. (A) NO levels in the cell supernatant were determined using Griess reagent. (B) iNOS mRNA expression was examined using RT‐PCR. C,D) RAW264.7 cells were pretreated with 5i for 8 h and stimulated with LPS (1 μg mL^{− 1}) for 16 h. The amounts of cytokines in supernatant were determined using ELISA kits. E,F) RAW264.7 cells were pretreated with 5i for 8 h and stimulated with LPS (1 μg mL^{− 1}) for 1.5 h. mRNA expression levels of TNF‐α (E) and CCL2 (F) were determined using RT‐PCR. Data are presented as mean ± SD. Statistical analyses were performed using GraphPad Prism 8.0.1. Significant differences between groups were determined using Student's t‐test, with statistical significance set at *p < 0.05, **p < 0.01, and ***p < 0.01, respectively (n = 3).

Among these markers, NO, a reactive nitrogen species generated by iNOS, is critically involved in sustained inflammatory responses and has been implicated in pathologies, such as chronic kidney disease and neuroinflammation. TNF‐α is a central proinflammatory cytokine whose aberrant expression contributes to the progression of various chronic diseases, including rheumatoid arthritis, Alzheimer's disease, cancer, and cellular senescence. CCL2 (MCP‐1), a chemokine that recruits monocytes to sites of inflammation, is known to exacerbate tissue damage in cardiovascular diseases and promote tumor‐associated inflammation. The simultaneous suppression of these diverse mediators by 5i not only supports its role as an Nrf2 activator but also underscores the potential of targeting Nrf2 as a multimodal anti‐inflammatory strategy. These findings raise the possibility that 5i may serve as a lead compound for therapeutic development in inflammatory and degenerative diseases driven by macrophage‐mediated immune responses.

Interestingly, increasing evidence has suggested a complex crosstalk between the Nrf2 and NF‐κB pathways, two master regulators of oxidative stress and inflammation, respectively. Nrf2 activation has been shown to inhibit NF‐κB signaling by reducing the levels of intracellular ROS, which are known to promote NF‐κB activation.^{[ 35 ]} Additionally, Nrf2 may directly suppress the transcription of NF‐κB target genes or compete for coactivators, such as cAMP response element binding protein (CREB)‐binding protein (CBP).^{[ 35 ]} Therefore, the anti‐inflammatory effects of 5i might be mediated not only by the transcriptional activation of cytoprotective genes via Nrf2, but also by indirect suppression of NF‐κB‐mediated inflammatory cascades. Further investigation into this regulatory interplay is essential to fully elucidate the mechanisms underlying the anti‐inflammatory activity of 5i.

2.7. ADME Properties of 5i

ADME properties are among the most important factors in the early stages of drug discovery and contribute to determining the subsequent structural development plan. Therefore, we evaluated the major ADME properties of compound 5i, including solubility and membrane permeability. Solubility test using phosphate‐buffered saline (PBS; pH 7.4) showed that the solubility of compound 5i in PBS was as low as 2.65 μM (Table 2). In addition, the metabolic stability of compound 5i in liver microsomes was low in both human and mouse (Table 2). To determine the passive membrane diffusion rates, a parallel artificial membrane permeability assay (PAMPA) was performed. Importantly, the membrane permeability of 5i was 3.06 × 10⁻⁶ cm sec^{− 1} (Table 2), which was superior to that of metoprolol, a high‐permeability control (1.85 × 10⁻⁶ cm sec⁻¹). Additionally, compound 5i had a high binding ratio to plasma proteins, with the unbound fraction in serum <3.42% (Table 2).

Collectively, these results strongly indicate the need to enhance the ADME properties of compound 5i and its related derivatives. A major challenge is the high ClogP value and low solubility of the compounds (Tables 1 and 2), which are particularly problematic for rigid and highly symmetrical naphthalene‐1,4‐bisulfonamide PPI inhibitors. These characteristics may be attributed to the low metabolic stability of 5i. Although comprehensive ADME profiling was not conducted for compounds other than 5i in this study, previous investigations have shown that the structurally related compound, Cpd16 (Figure 1), displayed markedly low metabolic stability.^{[ 36 ]} One potential solution to the solubility issue is the substitution of the hydrogen atoms on the two nitrogen atoms of the sulfonamide group with acetic acid moieties, which represents a canonical approach to structural expansion. However, the introduction of dicarboxylate substituents may reduce membrane permeability and impair the intracellular Nrf2 activation potential. Therefore, alternative strategies to enhance hydrophilicity, such as the incorporation of heteroatoms into the benzenesulfonamide moieties, should be explored (Scheme  1 ).

Scheme 1.

Preparation route of the naphathalene‐2‐acetamide derivatives. a) Pd/C (5%), H₂, CH₂Cl₂, 24 h, r.t., 98%; b) 4‐ethoxybenzenesulfonyl chloride, pyridine, CH₂Cl₂, 24 h, r.t., 95%; c) CAN, MeOH, r.t., 15 min, 96%; d) dimethyl malonate, Et₃N, toluene, r.t., 2 h; e) 2 M NaOH, EtOH, 100 °C, 49% over 2 steps; and f) corresponding amine, EDC•HCl, DMF, r.t., 3 h.

3. Conclusion

In this study, we identified naphthalene‐1,4‐bissulfonamide derivatives with a 2‐acetamide side chain as promising Nrf2 activators through PPI inhibition between Keap1 and Nrf2. Among the acetamide derivatives, secondary amides exhibited only weak intracellular Nrf2 activation, whereas cyclic tertiary amides 5i, 5j, and 5k demonstrated significantly stronger effects. Notably, X‐ray cocrystal structure analysis and cellular assays revealed that compound 5i formed a complex with Keap1 and exhibited anti‐inflammatory effects, suggesting its potential as a therapeutic candidate for the treatment of inflammatory diseases via Nrf2 activation.

Conclusively, the incorporation of a side chain at the 2‐position of naphthalene represents a promising strategy for designing potent PPI inhibitors targeting the Keap1‐Nrf2 interaction. Notably, the pyrrolidine‐based acetamide derivative 5i is a strong lead compound, demonstrating significant potential for the development of Nrf2 activators by disrupting PPI. Furthermore, compound 5i shows promise as a candidate for cellular Nrf2 activation with potential therapeutic applications in the treatment of inflammatory diseases. These findings highlight the efficacy of side‐chain modifications in enhancing the activity of PPI inhibitors and underscore the therapeutic potential of Nrf2 modulation in inflammatory pathologies.

4. Experimental Section

4.1.

4.1.1.

Generals

Unless otherwise stated, all commercially available reagents were purchased from FUJIFILM Wako Pure Chemical Co. (Osaka, Japan), Tokyo Chemical Industry Co. (Tokyo, Japan), Kanto Chemical Co. (Tokyo, Japan), and Merck Co. (Darmstadt, Germany), and were used as received. Proton nuclear magnetic resonance (¹H NMR) spectra (400 MHz) were recorded on a JEOL JNM‐ECZL 400S NMR spectrometer (JEOL, Tokyo, Japan), using tetramethylsilane as the internal standard (δ = 0.00). Carbon‐13 (¹³C) NMR spectra (100 MHz) were recorded using the same NMR spectrometer and the chemical shifts were compared with the signals of dimethyl sulfoxide (DMSO) (δ = 39.50). Mass spectra were recorded using an LCMS‐9030 ESI‐Q‐TOF instrument (Shimadzu, Kyoto, Japan).

Design and chemistry: Preparation of New Naphathalene‐2‐acetamide Derivatives

All the newly synthesized naphthalene‐2‐acetamide derivatives were prepared according to a previously reported method.^{[ 32 ]} After the preparation of the acetate derivative (4, Scheme), condensation between the carboxylate of 4 and the corresponding amine was performed.

Briefly, compound 4 (300 mg, 0.513 mmol) was dissolved in N,N‐dimethylformamide (DMF, 5 mL) and the corresponding amine (2.0 equiv.) and EDC•HCl (197 mg, 1.03 mmol, 2.0 equiv.) was added to the solution. After stirring for 3 h at room temperature, 2 M HCl (ca. 50 mL) was added. The water layer was extracted twice with ethyl acetate (ca. 50 mL × 2). The combined organic layers were washed twice with brine (ca. 30 mL × 2), dried over sodium sulfate, and concentrated under reduced pressure using a rotary evaporator. Thereafter, the crude compound was purified using silica gel column chromatography (eluent: n‐hexane/ethyl acetate) to obtain the desired naphthalene‐2‐acetamide derivatives. Note: In the ¹H NMR spectrum, the methylene protons of the acetamide side chain directly attached to naphthalene exhibited significant broadening in deuterated DMSO; in some cases, signals were not observed.

Design and chemistry: Preparation of new naphathalene‐2‐acetamide derivatives: 2‐(1,4‐Bis((4‐ethoxyphenyl)sulfonamido)naphthalen‐2‐yl)‐N‐cyclopropylacetamide (5d)

An off‐white solid (70 mg, 22% yield) was obtained. ¹H NMR (400 MHz, DMSO‐d ₆ ) δ 0.37–0.41 (m, 2H), 0.58–0.63 (m, 2H), 1.30 (t, J = 6.8 Hz, 3H), 1.33 (t, J = 6.8 Hz, 3H), 2.55–2.59 (oct, J = 3.6 Hz, 1H), 3.31 (brs, 2H), 4.04 (q, J = 6.8 Hz, 2H), 4.07 (q, J = 6.8 Hz, 2H), 6.94 (d, J = 8.8 Hz, 2H), 6.94 (d, J = 8.8 Hz, 2H), 6.97 (d, J = 8.8 Hz, 2H),7.12 (s, 1H), 7.18–7.22 (m, 1H), 7.29–7.33 (m, 1H), 7.40 (d, J = 8.8 Hz, 2H), 7.58 (d, J = 8.8 Hz, 2H), 7.64 (d, J = 8.8 Hz, 1H), 7.92 (d, J = 8.8 Hz, 1H), 8.11 (d, J = 4.4 Hz, 1H). 9.87 (brs, 1H), 10.10 (brs, 1H). ¹³C NMR (100 MHz, DMSO‐d ₆ ) δ 5.66, 14.39, 22.42, 38.42, 63.66, 63.70, 114.53, 114.69, 122.88, 124.36, 125.44, 125.79, 128.74, 128.92, 131.31, 131.86, 132.05, 133.23, 161.61, 161.74, 171.05. HRMS (ESI): m/z calcd for C₃₁H₃₃N₃O₇S₂–H^–: 622.1687 [M–H]^–; found: 622.1686.

Design and chemistry: Preparation of new naphathalene‐2‐acetamide derivatives: 2‐(1,4‐Bis((4‐ethoxyphenyl)sulfonamido)naphthalen‐2‐yl)‐N‐cyclobutylacetamide (5e)

An off‐white solid (130 mg, 40% yield) was obtained. ¹H NMR (400 MHz, DMSO‐d ₆ ) δ 1.31 (t, J = 7.2 Hz, 3H), 1.33 (t, J = 7.2 Hz, 3H), 1.59–1.65 (m, 2H), 1.80–1.90 (m, 2H), 2.09–2.16 (m, 2H), 3.28 (brs, 2H), 4.00–4.14 (m, 5H), 6.93–6.99 (m, 4H), 7.12 (s, 1H), 7.20–7.24 (m, 1H), 7.30–7.34 (m, 1H), 7.40 (dd, J = 6.8, 2.0 Hz, 2H), 7.58 (dd, J = 6.8, 2.0 Hz, 2H), 7.67 (d, J = 8.4 Hz, 1H), 8.29 (d, J = 8.4 Hz, 1H), 9.92 (brs, 1H), 10.09 (brs, 1H). ¹³C NMR (100 MHz, DMSO‐d ₆ ) δ 14.42, 14.43, 14.73, 30.24, 38.45, 44.04, 63.70, 63.75, 114.55, 114.74, 122.93, 124.32, 124.46, 128.79, 128.97, 131.26, 131.84, 132.10, 133.13, 161.65, 161.80, 168.84. HRMS (ESI): m/z calcd for C₃₂H₃₅N₃O₇S₂–H^–: 636.1844 [M–H]^–; found: 636.1859.

Design and chemistry: Preparation of new naphathalene‐2‐acetamide derivatives: 2‐(1,4‐Bis((4‐ethoxyphenyl)sulfonamido)naphthalen‐2‐yl)‐N‐cyclopentylacetamide (5f)

A light yellow solid (88 mg, 26% yield) was obtained. ¹H NMR (400 MHz, DMSO‐d ₆ ) δ 1.30 (t, J = 7.2 Hz, 3H), 1.35 (t, J = 7.2 Hz, 3H), 1.46–1.51 (m, 2H), 1.60–1.64 (m, 2H), 1.73–1.79 (m, 2H), 3.29 (brs, 2H), 3.92 (q, J = 7.2 Hz, 1H), 4.04 (q, J = 7.2 Hz, 2H), 4.07 (q, J = 7.2 Hz, 2H), 6.95–6.98 (m, 4H), 7.14 (s, 1H), 7.21–7.25 (m, 1H), 7.30–7.34 (m, 1H), 7.41 (d, J = 8.8 Hz, 2H), 7.58 (d, J = 8.8 Hz, 2H), 7.70 (d, J = 8.8 Hz, 1H), 7.93 (d, J = 8.8 Hz, 1H), 8.02 (d, J = 7.2 Hz, 1H), 9.96 (brs, 1H), 10.09 (brs, 1H). ¹³C NMR (100 MHz, DMSO‐d ₆ ) δ 14.37, 23.44, 32.25, 38.51, 50.42, 63.65, 63.72, 114.51, 114.72, 122.89, 124.18, 124.47, 125.46, 125.81, 128.73, 128.77, 128.81, 128.88, 131.30, 131.83, 131.88, 132.07, 133.20, 161.61, 161.77, 169.41. HRMS (ESI): m/z calcd for C₃₃H₃₇N₃O₇S₂–H^–: 650.2000 [M–H]^–; found: 650.2001.

Design and chemistry: Preparation of new naphathalene‐2‐acetamide derivatives: 2‐(1,4‐Bis((4‐ethoxyphenyl)sulfonamido)naphthalen‐2‐yl)‐N‐cyclohexylacetamide (5g)

An off‐white solid (114 mg, 33% yield) was obtained. ¹H NMR (400 MHz, DMSO‐d ₆ ) δ 1.06–1.19 (m, 6H) 1.30 (t, J = 6.8 Hz, 3H), 1.33 (t, J = 6.8 Hz, 3H), 1.53–1.56 (m, 1H), 1.64–1.70 (m, 5H), 3.30 (brs, 2H), 3.42–3.49 (m, 1H), 4.04 (q, J = 6.8 Hz, 2H), 4.07 (q, J = 6.8 Hz, 2H), 6.93–6.99 (m, 4H), 7.14 (s, 1H), 7.21–7.25 (m, 1H), 7.30–7.34 (m, 1H), 7.41 (dd, J = 7.2, 2.0 Hz, 2H), 7.57 (dd, J = 7.2, 2.0 Hz, 2H), 7.69 (d, J = 8.4 Hz, 1H), 7.92 (d, J = 8.4 Hz, 1H), 7.94 (d, J = 7.6 Hz, 1H), 9.97 (brs, 1H), 10.10 (brs, 1H). ¹³C NMR (100 MHz, DMSO‐d ₆ ) δ 14.41, 24.54, 25.21, 32.37, 38.86, 47.66, 63.66, 63.73, 114.52, 114.73, 122.91, 124.10, 124.47, 125.49, 125.85, 128.75, 128.89, 131.32, 131.84, 131.89, 132.08, 133.30, 161.62, 161.78, 168.98. HRMS (ESI): m/z calcd for C₃₄H₃₉N₃O₇S₂–H^–: 664.2157 [M–H]^–; found: 664.2169.

Design and chemistry: Preparation of new naphathalene‐2‐acetamide derivatives: 2‐(1,4‐Bis((4‐ethoxyphenyl)sulfonamido)naphthalen‐2‐yl)‐N‐phenylacetamide (5hr)

A light yellow solid (68 mg, 20% yield) was obtained. ¹H NMR (400 MHz, DMSO‐d ₆ ) δ 1.26 (t, J = 6.8 Hz, 3H), 1.30 (t, J = 6.8 Hz, 3H), 3.74 (brs, 2H), 3.91 (q, J = 6.8 Hz, 2H), 3.91 (q, J = 6.8 Hz, 2H), 4.02 (q, J = 6.8 Hz, 2H), 6.85 (dd, J = 7.2, 2.0 Hz, 2H), 6.92 (dd, J = 7.2, 2.0 Hz, 2H), 7.05–7.08 (m, 1H), 7.15–7.19 (m, 1H), 7.24 (s, 1H), 7.30–7.34 (m, 3H), 7.43 (dd, J = 7.2 Hz, 2H), 7.54–7.61 (m, 5H), 7.95 (s, J = 8.4 Hz, 1H), 9.85 (brs, 1H), 10.10 (brs, 1H), 10.12 (brs, 1H). ¹³C NMR (100 MHz, DMSO‐d ₆ ) δ 14.36, 14.37, 63.59, 63.65, 114.47, 114.66, 119.11, 122.89, 123.23, 123.94, 125.83, 128.60, 128.63, 128.74, 128.82, 128.98, 131.25, 131.83, 131.93, 131.99, 139.19, 161.61, 161.73, 168.64. HRMS (ESI): m/z calcd for C₃₄H₃₃N₃O₇S₂–H^–: 658.1687 [M–H]^–; found: 658.1689.

Design and chemistry: Preparation of new naphathalene‐2‐acetamide derivatives: N,N′‐(2‐(2‐Oxo‐2‐(pyrrolidin‐1‐yl)ethyl)naphthalene‐1,4‐diyl)bis(4‐ethoxybenzenesulfonamide) (5i)

A light‐yellow solid (179 mg, 56% yield) was obtained. ¹H NMR (400 MHz, DMSO‐d ₆ ) δ 1.30 (t, J = 7.2 Hz, 3H), 1.33 (t, J = 7.2 Hz, 3H), 1.73–1.84 (m, 2H), 3.17 (t, J = 6.8 Hz, 2H), 3.23 (t, J = 6.8 Hz, 2H), 3.45 (brs, 2H), 4.04 (q, J = 7.2 Hz, 2H), 4.07 (q, J = 7.2 Hz, 2H), 6.97 (d x 2, J = 8.8 Hz, 4H), 7.07 (s, 1H), 7.23–7.27 (m, 1H), 7.31–7.35 (m, 1H), 7.44 (d, J = 8.8 Hz, 2H), 7.58 (d, J = 8.8 Hz, 2H), 7.71 (d, J = 8.4 Hz, 1H), 7.93 (d, J = 8.4 Hz, 1H), 9.82 (brs, 1H), 10.10 (brs, 1H). ¹³C NMR (100 MHz, DMSO‐d ₆ ) δ 14.39, 24.04, 25.54, 37.45, 45.43, 46.09, 63.66, 63.72, 114.53, 114.70, 122.90, 124.41, 124.95, 125.44, 125.82, 128.73, 128.83, 128.90, 131.42, 131.74, 132.17, 132.34, 132.75, 161.57, 161.71, 167.90. HRMS (ESI): m/z calcd for C₃₂H₃₅N₃O₇S₂–H^–: 636.1844 [M–H]^–; found: 636.1848.

Design and chemistry: Preparation of new naphathalene‐2‐acetamide derivatives: N,N′‐(2‐(2‐Oxo‐2‐(piperidin‐1‐yl)ethyl)naphthalene‐1,4‐diyl)bis(4‐ethoxybenzenesulfonamide) (5j)

A light‐yellow solid (220 mg, 66% yield) was obtained. ¹H NMR (400 MHz, DMSO‐d ₆ ) δ 1.21—1.43 (m, 4H), 1.30 (t, J = 7.2 Hz, 3H), 1.33 (t, J = 7.2 Hz, 3H), 1.46–1.56 (m, 2H), 3.11 (t, J = 4.8 Hz, 2H), 3.52 (brs, 2H), 4.04 (q, J = 7.2 Hz, 2H), 4.07 (q, J = 7.2 Hz, 2H), 6.97–6.99 (m, 4H), 7.05 (s, 1H), 7.22–7.26 (m, 1H), 7.30–7.34 (m, 1H), 7.45 (d, J = 8.0 Hz, 2H), 7.58 (d, J = 8.0 Hz, 2H), 7.67 (d, J = 8.8 Hz, 1H), 7.92 (d, J = 8.8 Hz, 1H), 9.86 (brs, 1H), 10.12 (brs, 1H). ¹³C NMR (100 MHz, DMSO‐d ₆ ) δ 14.40, 24.00, 25.12, 25.75, 36.45, 42.11, 46.19, 63.68, 63,73, 114.58, 114,74, 122,95, 124.25, 124.31, 125.46, 125.89, 128.32, 128.78, 128.90, 131.42, 131.87, 132.12, 132.29, 133.07, 161.62, 161.76, 167.77. HRMS (ESI): m/z calcd for C₃₃H₃₇N₃O₇S₂–H^–: 650.2000 [M–H]^–; found: 650.2014.

Design and chemistry: Preparation of new naphathalene‐2‐acetamide derivatives: N,N′‐(2‐(2‐oxo‐2‐(morpholin‐1‐yl)ethyl)naphthalene‐1,4‐diyl)bis(4‐ethoxybenzenesulfonamide) (5k)

A light‐yellow solid (234 mg, 70% yield) was obtained. ¹H NMR (400 MHz, DMSO‐d ₆ ) δ 1.30 (t, J = 7.2 Hz, 3H), 1.32 (t, J = 7.2 Hz, 3H), 3.18 (brs, 2H), 3.41–3.42 (m, 2H), 3.52–3.54 (m, 2H), 4.04 (q, J = 7.2 Hz, 2H), 4.06 (q, J = 7.2 Hz, 2H), 6.95–7.00 (m, 4H), 7.08 (s, 1H), 7.21–7.25 (m, 1H), 7.30–7.34 (m, 1H), 7.45 (dd, J = 7.2, 2.0 Hz, 2H), 7.59 (dd, J = 7.2, 2.0 Hz, 2H), 7.65 (d, J = 8.8 Hz, 1H), 7.92 (d, J = 8.8 Hz, 1H), 9.83 (brs, 1H), 10.15 (brs, 1H). ¹³C NMR (100 MHz, DMSO‐d ₆ ) δ 14.40, 36.24, 41.67, 45.69, 63.69, 63.74, 65.80, 65.96, 114.61, 114.76, 122.93, 124.30, 125.52, 125.93, 128.39, 128.76, 128.80, 128.90, 131.39, 131.93, 132.07, 132.29, 132.84, 161.65, 161.78, 168.41.HRMS (ESI): m/z calcd for C₃₂H₃₅N₃O₈S₂–H^–: 652.1793 [M–H]^–; found: 652.1801.

Cell Culture

The NRF2/ARE luciferase reporter HEK293 stable cell line (HEK293‐ARE‐Luc, Lot No. 061821. C7) was purchased from Signosis, Inc. (San Diego, CA, USA). The cell line was cultured in Dulbecco's modified Eagle medium (DMEM, 4.5 g L^{− 1} glucose, Nacalai Tesque, Inc., Kyoto, Japan) supplemented with 10% fetal bovine serum (ThermoFisher Scientific, MA, USA) and 1% penicillin‐streptomycin (Nacalai Tesque, Inc.) at 37 °C in a humidified atmosphere with 5% CO₂.

Murine macrophage‐like RAW264.7 cells were purchased from RIKEN BRC (Ibaraki, Japan) and cultured using the same method as that used for HEK293 cells.

Intracellular Nrf2 Activation Assay

HEK293‐ARE‐Luc cells (2.0 × 10⁴ cells per well) were seeded in a white‐bottom 96 well microplate, with the medium volume adjusted to 100 μL per well. After 24 h of incubation, the test compound in DMSO (1 μL) was added and incubated for another 24 h. Thereafter, One‐Step Luciferase Assay System (BPS Bioscience Inc., CA, USA) was added to each well according to the manufacturer's protocol, followed by incubation at room temperature. After 15 min, luminescence intensity was measured using a Synergy H1 microplate reader (Agilent Technologies, Santa Clara, CA, USA). Intracellular Nrf2 activation was calculated as the ratio of luminescence intensity to that of the vehicle control group (n = 3).

Cytotoxicity (WST‐8 Assay)

HEK293‐ARE‐Luc cells (5.0 × 10⁴ cells per well) were seeded in a 96 well clear microplate, with the medium volume adjusted to 100 μL per well. After 24 h of incubation, the test compound in DMSO (1 μL) was added and incubated for another 24 hr. Thereafter, 10 μL of the Cell Counting Kit‐8 solution (Dojin Chemical Co., Kumamoto, Japan) was added and incubated for 4 h. Lastly, the absorbance at 420 nm was measured using a Synergy H1 microplate reader and cell viability was calculated (n = 3).

Fluorescent Polarization Assay

Briefly, FP assay was performed in a 384 well plate using the KEAP1‐Nrf2 Inhibitor Screening Assay Kit (BPS Bioscience, Cat. 72,020) according to the manufacturer's protocol, with slight modifications.

Expression and Purification of Keap1‐DC

Mouse Keap1‐DC expression vector was kindly provided by Professor Masayuki Yamamoto (Tohoku University, Miyagi, Japan). The gene encoding the mouse Keap1 DC domain (residues 321–609), fuzed with an N‐terminal His₆‐tag followed by a TEV protease recognition sequence, was inserted into the expression vector pET‐101 (Thermo Fisher Scientific, MA, USA). Escherichia coli BL21(DE3) cells (Agilent Technologies, CA, USA) were transformed with the expression vector and cultured in LB medium at 37 °C until they reached an OD₆₀₀ of 0.7–0.9. Thereafter, protein expression was induced by adding 0.5 mM isopropyl‐β‐D‐thiogalactopyranoside, followed by culturing at 37 °C for 4 h. Cells were harvested by centrifugation and disrupted by sonication in a buffer containing 20 mM Tris‐HCl (pH 8.0), 150 mM NaCl, 1 mM DTT, and 1 mM phenylmethylsulfonyl fluoride. The protein was purified from the clarified lysate using HIS‐Select Nickel Affinity Gel (Merck Millipore, Burlington, MA, USA), followed by His₆‐tag cleavage with TEV protease at a ratio of 1:50 (w/w) at 4 °C for 16 h. Further purification was performed using a Superdex 200 Increase 10/300 GL gel filtration column (Cytiva, Tokyo, Japan) with 20 mM Tris‐HCl (pH 8.0), 150 mM NaCl, and 1 mM DTT.

Crystallization of Keap1‐DC and Structure Determination

For crystallization, purified Keap1‐DC was concentrated to approximately 12 mg mL^{− 1} in 20 mM Tris‐HCl (pH 8.0), 150 mM NaCl, and 20 mM DTT. To prepare the Keap1‐DC/compound 5i complex, the protein solution was mixed with an approximately twofold molar excess of compound 5i from a 10 mM DMSO stock solution (final DMSO concentration: 4%). Crystallization experiments were performed using the sitting‐drop vapor‐diffusion method at 293 K. Crystallization droplets were prepared by mixing equal volumes (0.5 μL each) of the protein–compound solution and the reservoir solution (0.1 M Tris‐HCl [pH 7.5], 2.8 M Sodium formate [pH 7.0]).

Diffraction data sets were collected at a wavelength of 0.9800 Å on beamline Photon Factory BL‐17A (Ibaraki, Japan) under cryogenic condition at 100 K. Crystals were soaked in a cryoprotectant solution containing 10% glycerol and flash‐cooled by plunging into liquid nitrogen. Diffraction data were processed using X‐ray detector software.^{[ 37 ]} Initial phases of the Keap1‐DC structures were determined by molecular replacement using Molrep,^{[ 38 ]} with an unliganded mouse Keap1‐DC structure (PDB ID: 1X2J)^{[ 34 ]} as the search model. The model was further refined via stepwise cycles of manual model building using COOT^{[ 39 ]} and restrained refinement using REFMAC^{[ 40 ]} and Phenix.refine.^{[ 41 ]} The quality of the final model was validated using the PDB validation server (https://validate‐rcsb‐2.wwpdb.org/). The coordinates and structural factors were deposited in the Protein Data Bank. The crystallographic data collection and refinement statistics are summarized in Table S1, Suppporting Information. Structural representations were generated using Pymol.^{[ 42 ]}

Accession Code

The structural factors and coordinate files of the mouse Keap1‐DC domain in complex with compound 5i were deposited in the Protein Data Bank 9UO8. The authors will release the atomic coordinates upon publication of this article.

Immunoblotting

Briefly, cell lysates were prepared using lysis buffer (50 mM Tris‐HCl, pH 7.4, 10% glycerol, 50 mM NaCl, 0.5% sodium deoxycholate, 0.5% NP‐40, 20 mM NaF, and 0.2 mM Na₃VO₄) supplemented with protease inhibitors (Nacalai Tesque). Thereafter, DENATURED samples were resolved using sodium dodecyl sulfate – poly‐acrylamide gel electrophoresis, transferred onto PVDF membranes (Millipore, Billerica, MA, USA), and probed with primary antibodies, followed by incubation with horseradish peroxidase‐conjugated secondary antibodies (Cell Signaling Technology, Danvers, MA, USA). Detection was performed using the electrochemiluminescence system (GE Healthcare, Little Chalfont, UK). Anti‐Nrf2 and anti‐β‐actin antibodies were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA).

Measurement of Nitric Oxide

RAW264.7 cells (5 × 10⁵ cells/500 μL) were seeded in a 24‐well plate and treated with 5i. LPS (1 μg mL^{− 1}) was added to the wells after 8 h, followed by incubation at 37 °C for 16 h. After incubation, 100 μL of the supernatant was collected and mixed with Griess reagent, and the absorbance at 540 nm was measured. Finally, NO concentration was determined using an NaNO₂ standard solution via the internal standard method, as previously described.^{[ 43 ]}

Enzyme‐Linked Immunosorbent Assay (ELISA)

RAW264.7 cells (2 × 10⁵ cells/500 μL) were seeded in 24‐well plates and treated with 5i. After 8 h, LPS (1 μg mL^{− 1}) was added, followed by incubation at 37 °C for 16 h. Thereafter, CCL2 and TNF‐α concentrations in the cell supernatants were determined using an Immunoassay Kit (eBioscience, Inc. San Diego, CA, USA) and a TNF‐α ELISA kit (R&D Systems, Minneapolis, MN, USA), respectively.

Reverse Transcription‐Polymerase Chain Reaction (RT‐PCR)

RNA extraction, RT, and quantitative real‐time PCR were performed as previously described.^{[ 43 ]} The primer sequences used for PCR were as follows: HO‐1: 5^′‐CACGCATATACCCGCTACCT‐3^′ (forward), 5^′‐CCAGAGTGTTCATTCGAGCA‐3^′ (reverse), NQO1: 5^′‐TTCTCTGGCCGATTCAGAGT‐3^′ (forward), 5^′‐GGCTGCTTGGAGCAAAATAG‐3^′ (reverse), iNOS: 5^′‐TCTGCGCCTTTGCTCATGAC‐3^′ (forward), 5^′‐TAAAGGCTCCGGGCTCTG‐3^′ (reverse), TNF‐α: 5^′‐TACTGAACTTCGGGGTGATCGGTCC‐3^′ (forward), 5^′‐CAGCCTTGTCCCTTGAAGAGAACC‐3^′ (reverse), CCL2: 5^′‐TGGGGACACCTTTTAGCATC‐3^′ (forward), 5^′‐GCCCATCGCCAATGAGCTG‐3^′ (reverse), GAPDH: 5^′‐ACTCCACTCACGGCAAATTC‐3^′ (forward), 5^′‐CCTTCCACAATGCCAAAGTT‐3^′ (reverse).

Solubility in PBS

Solutions of the compounds were prepared by diluting 10 mM DMSO stock solution (2 μL) with 165 μL of PBS and mixing at 37 °C for 4 h via constant rotation at 1,000 rpm. Thereafter, the solution was loaded into 96‐well MultiScreen Filter Plates (product number MSHVN4510, 0.45 μm hydrophilic PVDF membrane; Millipore), followed by filtration via centrifugation. The obtained filtrates were analyzed using high‐performance liquid chromatography with UV detection at 254 nm. The solubility was determined by comparing the peak area of the filtrate with that of the 100 μM standard solution. Two technical replicates were used for each experiment.

Metabolic Stability

Briefly, metabolic stability was assessed using hepatic microsomal stability assay. Specifically, the loss of the parent compound over time was evaluated using liquid chromatography‐tandem mass spectrometry (LC–MS), with the amount of the compound at time zero used as the reference. After 5 min of preincubation, 1 mM NADPH (final concentration; the same applies to the following) was added to a mixture containing 1 μM of the compound, 0.2 mg mL^{− 1} of human or mouse liver microsomes (Sekisui XenoTech LLC, Kansas City, KS, USA), 1 mM EDTA, and 0.1 M phosphate buffer (pH 7.4). Thereafter, the mixture was incubated at 37 °C for 30 min at 60 rpm. An aliquot (50 μL) of the incubation mixture was added to 250 μL of chilled acetonitrile/methyltestosterone (IS). After centrifugation at 3,150 × g for 15 min at 4 °C, the supernatants were analyzed by LC–MS/MS. Hepatic microsomal stability (mL/min/kg, CL_int) was calculated as described in a previous report,^{[ 44 ]} using 48.8 and 45.4 mg MS protein/g liver and 25.7 and 87.5 g liver/kg body weight as scaling factors for human and mouse, respectively. Notably, the half‐life was calculated from the parent compound reduction rate in 30 min, assuming that the compound was reduced by the primary reaction, and clearance was further calculated using microsomal concentration. Two technical replicates were used for each experiment.

Membrane Permeability Assay

The PAMPA assay was performed using a Corning Gentest precoated PAMPA plate system (Corning, NY, USA). The acceptor plate was prepared by adding 200 μL of 0.1 M phosphate buffer (pH 7.4) supplemented with 5% DMSO to each well, followed by the addition of 300 μL of a solution of 1 μM of the compounds in 0.1 M phosphate buffer (pH 6.4), with 5% DMSO added to the donor wells. The acceptor plate was then placed on top of the donor plate and incubated at 37 °C for 4 h, without agitation. After incubation, the plates were separated, and the solutions from each well of both the acceptor and donor plates were transferred to 96‐well plates and mixed with acetonitrile. The final concentrations of compounds in both the donor and acceptor wells and the concentrations of the initial donor solutions were analyzed using LC‐MS/MS. The permeability of the compounds was calculated as previously described.^{[ 45 ]} Antipyrine (100 μM), metoprolol (500 μM), and sulfasalazine (500 μM) were used as reference compounds. The permeabilities of antipyrine, metoprolol, and sulfasalazine were 23.1, 1.85, and 0.077 × 10^–6 cm s^{− 1}, respectively. Two technical replicates were used for each experiment.

Protein Binding Ratio

Protein binding ratio was assessed by determining the unbound fraction of the compound in plasma and serum. Briefly, the unbound fractions of the compounds in human plasma, mouse plasma, and 10% FBS were determined using an equilibrium dialysis apparatus. An HTDialysis complete unit (HTD96b) and Dialysis Membrane Strips MWCO 12–14 kDa (HTDialysis, Galesferry, CT) were used. Plasma or 10% FBS was mixed with the test compound (1 μM), and 150 μL aliquots were loaded into the apparatus and dialyzed against 150 μL of 0.1 M phosphate buffer (pH 7.4) at 37 °C for 6 h under constant rotation at 80 rpm. A 100 μL aliquot of the buffer from the receiver side and 10 μL of blank plasma or 10% FBS were mixed. Additionally, a 10 μL aliquot of plasma or 10% FBS from the donor side and 100 μL of blank buffer were mixed. These mixtures were combined with 400 μL chilled acetonitrile/IS. After centrifugation at 3,150 × g for 15 min at 4 °C, the supernatants were analyzed using LC‐MS/MS. The unbound fraction (f_u) was calculated as the ratio of the concentration on the receiver side (Cbuffer) to that on the donor side (C_plasma): (f_u) = (C_buffer/C_plasma). Two technical replicates were used for each experiment.

LC‐MS/MS quantification Method

An LC‐MS8060 instrument equipped with a Shimadzu Nexera series LC system (Shimadzu) was used for analysis. All compounds were analyzed in the multireaction monitoring mode under electron spray ionization conditions. The analytical column employed was a CAPCELLPAK C18 MGIII (3 μm × 2.0 mm ID × 35 mm; OSAKA SODA, Osaka, Japan) maintained at 50 °C. The gradient mobile phase consisted of 0.1% formic acid in water (mobile phase A) and 0.1% formic acid in acetonitrile (mobile phase B), at a flow rate of 1 mL min^{− 1}. Initially, the mobile phase composition was 10% B, which was maintained at 0.2 min and increased linearly to 90% B over 1 min, followed by a constant hold at 0.8 min. The mobile phase was then returned to the initial condition of 10% B over 0.01 min and re‐equilibrated for 1 min. The transitions (precursor ion > product ion) of 5i and IS (methyltestosterone) were 638.2 > 197.1 and 303.1 > 109.1 (positive), respectively.

Statistical Analysis

All biological experiments were conducted independently at least thrice unless otherwise mentioned. Data are presented as mean ± standard deviation (SD). Statistical analyses were performed using GraphPad Prism 8.0.1 (GraphPad Software, San Diego, CA, USA). Significant differences between groups were determined using Student's t‐test, with statistical significance set at p < 0.05.

Conflict of Interest

The authors declare no conflict of interest.

Supporting information

Supplementary Material

Yasuda Daisuke, Toyoshima Kai, Kojima Koujin, Ishida Hanako, Kaitoh Kazuma, Imamura Riyo, Kanamitsu Kayoko, Kojima Hirotatsu, Funakoshi‐Tago Megumi, Osawa Masanori, Ohe Tomoyuki, Hirano Tomoya, ChemMedChem 2025, 20, e202500474. 10.1002/cmdc.202500474

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.