Abstract

   Former studies indicate that nuclear receptor subfamily 4 group A
   member 2 (Nurr1, NR4A2), a transcription factor, is regarded as a
   potential therapeutic target for central nervous system diseases, and
   many studies have focused on the development and optimization of
   agonists of Nurr1. Recent studies have shown that Nurr1 is upregulated
   in many other diseases. However, there is still a lack of effective
   inverse Nurr1 agonists as a therapeutic strategy or as pharmacological
   tools to counteract the receptor’s inherent activity. In this study, we
   screened Nurr1 ligands through a high-throughput screening system and
   identified a novel Nurr1 inverse agonist (K-strophanthoside). We
   further validated the binding site of K-strophanthoside on Nurr1 and
   investigated its effect on regulating Nurr1 function. K-strophanthoside
   directly binds to the ligand-binding domain of Nurr1 (Glu445, Glu514,
   Arg515, and His516) and mimics the function of Nurr1 knockdown by
   suppressing the intrinsic Nurr1 transcriptional activity. Our study
   contributes a valuable chemical tool for Nurr1 modulators and provides
   a potential treatment target for Nurr1-related disorders.
     __________________________________________________________________

   graphic file with name ao5c06698_0008.jpg
     __________________________________________________________________

   graphic file with name ao5c06698_0006.jpg

Introduction

   Nuclear receptor subfamily 4 group A member 2 (Nurr1, also known as
   NR4a2) is a transcription factor that plays a significant role in
   various diseases, and its dysregulation is linked to neurodegenerative
   diseases, − autoimmune diseases, , and cancers. − Our previous study
   documented a synthetic Nurr1 activator that has a strong
   antiparkinsonian effect in animal models of PD. Downregulation of Nurr1
   aggravates the pathology of AD involving neurodegeneration, Aβ
   accumulation, and neuroinflammation. Moreover, Nurr1 is increased in
   the peripheral blood of ALS patients and in the spinal cord of
   superoxide dismutase 1 (SOD1)-G93A transgenic mice in the early phase,
   which is regarded as a protective anti-inflammatory mechanism. However,
   in some diseases, increased Nurr1 is regarded as a pathogenic factor.
   ,, Previous studies have found abnormal expression of Nurr1 in multiple
   sclerosis (MS). ,, However, the relation between Nurr1 and MS is
   conflicting. Nurr1 was considered as a protective factor in MS; ,
   however, recent studies have further found that continuous upregulation
   of Nurr1 due to chronic inflammation in MS leads to neuronal death,
   indicating the critical importance of maintaining balanced Nurr1
   expression for the onset and progression of the disease. Furthermore,
   Nurr1 expression is increased in tumors, , associated with tumor growth
   and resistance. ,,,, Nurr1 mediated the ERK and AKT signaling pathways
   to facilitate tumor development in cervical cancer, and their
   inhibitors could halt tumor progression. In glioblastoma, the
   expression of Nurr1 is significantly upregulated in microglia, and both
   genetic and pharmacological inhibition of Nurr1 enhance the therapeutic
   efficacy against the tumor while suppressing its growth. , Therefore,
   Nurr1 represents a potential therapeutic target for a range of
   diseases.

   Nurr1 belongs to the nuclear receptor family of nonligand-dependent
   transcription factors. Its protein structure includes a
   nonligand-dependent transcription activation region 1 (AF1), a highly
   conserved zinc finger DNA binding domain (DBD), and a C-terminal
   ligand-binding domain (LBD) containing AF2. Previous study showed that
   unlike other nuclear receptors, the Nurr1-LBD lacks a conventional
   ligand-binding cavity and binding sites for coregulatory factors,
   leading to a lack of identifiable Nurr1 ligands. However, a series of
   small-molecule compounds were identified to modulate Nurr1
   transcriptional activities by physically binding to Nurr1-LBD,
   including synthetic (amodiaquine, AQ; chloroquine, CQ) and natural
   small-molecule compounds (docosahexaenoic acid, DHA;
   5,6-dihydroxyindole, DHI; prostaglandin E1, PGE1; and prostaglandin A1,
   PGA1). In addition to the above agonists, a number of inverse agonists
   have also been identified as chemical tools, including oxaprozin, the
   indole-based derivative, fatty acid mimetics, and nonsteroidal
   anti-inflammatory drugs. All of these results indicate the presence of
   Nurr1 ligands that bind to Nurr1-LBD and regulate its function.

   The important role of Nurr1 in related diseases indicates that the
   discovery of Nurr1 modulators remains an urgent necessity. However,
   most current research has focused on the development and validation of
   Nurr1 agonists, and there is still a lack of Nurr1 inverse agonists as
   both a potential therapeutic strategy for diseases and a valuable tool
   in elucidating Nurr1’s role in disease pathology. In this study, we
   screened a new inverse Nurr1 agonist (K-strophanthoside) from the
   Chemdiv compound library by structure-based virtual screening.
   Intriguingly, K-strophanthoside appears to modulate Nurr1 monomer
   activity more effectively than the Nurr1 homodimer and heterodimer, and
   it mimics the effects of Nurr1 silencing on various biological
   pathways. These findings warrant further exploration of the biological
   functions and potential applications of Nurr1 modulators.

Results

Compound 7 as a Potential Inverse Agonist for Nurr1

   The ChemDiv compound library, containing about 1.5 million compounds,
   was selected as the small-molecule library for virtual screening. The
   three-dimensional structure of Nurr1-LBD in the Protein Data Bank (PDB
   code: [37]6DDA) was used to dock with compounds in the ChemDiv compound
   library. The absolute values of docking scores were ranked from high to
   low, and the top 10 compounds are shown in [38]Figure S1. To explore
   the transcriptional activity of Nurr1 regulated by the top 10
   compounds, we preliminarily evaluated their toxicity using CCK-8, and
   the results showed that compounds 1 and 2 exhibited cytotoxicity, while
   others did not show cytotoxicity even at a higher concentration of 100
   μM ([39]Figure S2). Then, we constructed HEK293T cells for the
   Nurr1-mediated transcriptional activity detection system by
   transfecting a luciferase plasmid containing the Nurr1 binding site
   ([40]Figure S3). The cells cultured with 8 other compounds (0, 6.25,
   12.5, 25, 50, and 100 μM) were subjected to a luciferase reporter gene
   assay to determine the transcriptional activity of Nurr1. As shown in
   [41]Table , compounds 7 and 9 significantly inhibited the Nurr1
   transcriptional activity by up to 51% (P < 0.01) and 18% (P < 0.05),
   compounds 3 and 10 increased the activities by 1.38 and 1.36 folds
   compared with the control group at a concentration of 100 μM (P <
   0.01), respectively, and other compounds did not show significant
   changes in the Nurr1 transcriptional activity regulation ([42]Table ).
   According to the change in folds of Nurr1 transcriptional activities,
   compound 7 was selected as the primary target compound for further
   verification.

1. Summarized Transcriptional Regulation of Nurr1 by Compounds^ .

     concentration(μM)
     __________________________________________________________________

   compound 0 6.25 12.5 25 50 100
   3 1.00 ± 0.042 1.04 ± 0.074 0.91 ± 0.066 1.06 ± 0.046 1.10 ± 0.027
   1.38 ± 0.039**
   4 1.00 ± 0.043 0.97 ± 0.067 1.16 ± 0.065 1.23 ± 0.073 1.29 ± 0.088*
   1.23 ± 0.089
   5 1.00 ± 0.073 0.92 ± 0.036 0.87 ± 0.046 0.91 ± 0.010 0.88 ± 0.026 0.90
   ± 0.011
   6 1.00 ± 0.042 0.90 ± 0.019 1.06 ± 0.021 0.99 ± 0.023 1.10 ± 0.14 1.01
   ± 0.066
   7 1.00 ± 0.035 1.13 ± 0.065 0.89 ± 0.086 0.74 ± 0.13 0.76 ± 0.024 0.49
   ± 0.017**
   8 1.00 ± 0.017 0.93 ± 0.037 0.86 ± 0.029 1.19 ± 0.049 1.07 ± 0.086 1.10
   ± 0.076
   9 1.00 ± 0.048 0.97 ± 0.011 0.95 ± 0.07 0.88 ± 0.035 0.85 ± 0.045 0.82
   ± 0.024*
   10 1.00 ± 0.036 0.97 ± 0.040 1.15 ± 0.040 1.06 ± 0.070 1.23 ± 0.051*
   1.36 ± 0.036**
   [43]Open in a new tab
   ^a

   The transcriptional activity of Nurr1 regulated by compounds 3–10 in
   HEK293T cells was determined in a cellular reporter gene assay. *P <
   0.05, **P < 0.01. Data are mean ± S.E.M., n ≥ 3.

Direct Binding of K-Strophanthoside to Nurr1-LBD

   The name of compound 7 was K-strophanthoside, which was found by
   blasting the structure of compound 7 in the SciFinder database. To
   further verify the interaction between K-strophanthoside and the
   Nurr1-LBD, the computer molecular docking predicted that
   K-strophanthoside exhibited suitable spatial complementarity with the
   Nurr1-LBD by forming hydrogen bonds with Glu445, Leu509, Glu514,
   Arg515, and His516 ([44]Figure A–C). However, there was no structural
   complementarity between K-strophanthoside and other members of the NR4A
   family with known 3D structures ([45]Figure S4A,B). The results
   suggested that K-strophanthoside was potentially bound to Nurr1-LBD.
   Then, a surface plasmon resonance (SPR) experiment was performed to
   verify the direct binding between K-strophanthoside and Nurr1, with AQ
   as a positive control. The results showed that K-strophanthoside bound
   to Nurr1 in a concentration-dependent manner, similar to AQ ([46]Figure
   ). The SPR fitting results showed that the association rate constant (k
   [a]) of the complexes between K-strophanthoside and the Nurr1 protein
   was 3.352 × 10^2 M^–1 s^–1, and the dissociation rate constant (k [d])
   was 7.744 × 10^–3 s^–1. In particular, the dissociation constant (K
   [D]) of the complex for K-strophanthoside with the Nurr1 protein was
   2.310 × 10^–5 M, showing stronger affinity than AQ (K [D] = 3.695 ×
   10^–5 M, [47]Table ). Furthermore, we constructed site-directed
   mutagenesis of the Nurr1 plasmids, and the mutations did not affect
   Nurr1 activity compared to the wild-type Nurr1 ([48]Figure S4C).
   Mutations of Glu445, Glu514, Arg515, and His516 significantly improved
   the transcriptional activity of Nurr1 induced by K-strophanthoside,
   while mutation of Leu509 did not show significant changes ([49]Figure
   D). Therefore, the amino acid residues (Glu445, Glu514, Arg515, His516)
   of Nurr1 that formed hydrogen bonds with K-strophanthoside are
   essential for K-strophanthoside-induced Nurr1 transcriptional activity.
   However, these mutations did not affect the AQ-induced Nurr1 activity
   ([50]Figure S10). Collectively, these findings suggest that
   K-strophanthoside is a ligand for Nurr1.

1.

   [51]1
   [52]Open in a new tab

   Binding model of K-strophanthoside to the Nurr1 protein. (A)
   Two-dimensional binding pattern of K-strophanthoside to the Nurr1
   protein (PDB code: 6DDA). (B) Three-dimensional binding mode of
   K-strophanthoside to Nurr1-LBD. K-strophanthoside is shown as a yellow
   rod. The surrounding residues in the binding bag are green. Hydrogen
   bonds are represented by the dotted red line. (C) Electrostatic surface
   rendering of K-strophanthoside with Nurr-LBD. Blue and red surfaces
   indicate positively and negatively charged surfaces, respectively.
   K-strophanthoside is shown as a yellow rod. (D) Effects of Nurr1
   mutations in the potential K-strophanthoside binding sites on the
   transcriptional activity of Nurr1. The transcriptional activity of
   wild-type (WT) and mutant constructs (E445A, L509A, E514A, R515A, and
   H516A) was examined by luciferase reporter assay with or without
   K-strophanthoside (100 μM). *P < 0.05, **P < 0.01. Data are mean ±
   S.E.M., n ≥ 3, one-way ANOVA in comparison expression with control
   (WT-Nurr1).

2.

   [53]2
   [54]Open in a new tab

   Binding between Nurr1 and small-molecule compounds was determined by
   SPR. The Nurr1 protein was coupled to a CM5 chip and exposed to samples
   at various doses. (A) Concentration-dependent binding of AQ to Nurr1.
   AQ was injected at increasing doses (1.953, 3.906, 7.813, 15.63, 31.25,
   62.5, and 125 μM). (B) Steady-state affinity fitting diagram of AQ
   binding with Nurr1. (C) Concentration-dependent binding of
   K-strophanthoside to Nurr1. K-strophanthoside was injected at
   increasing doses (3.906, 7.813, 15.63, 31.25, 62.5, 125, and 250 μM).
   The black curves represent the fitting lines.

2. Dissociation Constant (K [D]) Values for the Binding of AQ and
K-strophanthoside with the Nurr1 Protein.

      protein        compounds        K [D](M)   R [max] (RU)
   Nurr1 protein amodiaquine (AQ)  3.695 × 10^–5 4.441
                 K-strophanthoside 2.310 × 10^–5 10
   [55]Open in a new tab

Regulation of Nurr1-Associated Gene Expression by K-Strophanthoside

   Nurr1 is an important transcriptional factor of genes (tyrosine
   hydroxylase, TH; L-aromatic amino acid decarboxylase, AADC, etc.)
   involved in the process of DA synthesis. To explore the regulation of
   K-strophanthoside on Nurr1 function, we cultured Neuro-2a cells with
   different concentrations of K-strophanthoside and screened for
   concentrations without cytotoxicity ([56]Figure S5). The results of
   Western blot showed that K-strophanthoside did not change the
   expression of Nurr1 ([57]Figure A,D), while the expression levels of TH
   and AADC were significantly downregulated to 47% and 35% by
   K-strophanthoside at a concentration of 30 μM (P < 0.05, [58]Figure
   A–C). Real-time PCR analysis showed that K-strophanthoside reduced the
   expression of TH and AADC to 75% and 62% compared with the control
   groups, respectively (P < 0.05, [59]Figure E). Notably, the expression
   levels of Nurr77, Nor1, and Nurr1 remained unchanged under the same
   conditions ([60]Figure S9). Immunofluorescence results showed that the
   levels of TH and AADC were decreased by K-strophanthoside with a
   concentration dependence ([61]Figure F,G). Moreover, we constructed
   Neuro-2a cells with Nurr1 downregulation ([62]Figure S6). Then,
   transcriptome analysis was conducted on Nurr1 knockdown cells and
   Neuro-2a cells treated with 30 μM K-strophanthoside by RNA sequencing
   to verify the role of K-strophanthoside in Nurr1-associated gene
   expression. A total of 1785 differential expression genes (DEGs),
   including 950 downregulated and 835 upregulated genes (fold change ≥
   1.5), were identified in Nurr1 knockdown cells ([63]Figure A), while
   535 downregulated and 403 upregulated genes were detected in cells
   treated with K-strophanthoside ([64]Figure B). Analysis showed that
   there were 93 genes with common expression changes between cells
   treated with K-strophanthoside and Nurr1 knockdown cells, including 40
   upregulated and 53 downregulated genes ([65]Figure C). Using the
   bioinformatics approach, we found that 73 of these genes contain
   NBRE-like binding sites within their promoter regions. These results
   are detailed in [66]Supplementary Table 2, supporting the notion that a
   substantial portion of the DEGs may be directly regulated by Nurr1
   through its canonical response elements. The clustered heatmap of
   commonly changed DEGs in Nurr1 knockdown cells and Neuro-2a cells
   treated with K-strophanthoside is shown in [67]Figure S7.

3.

   [68]3
   [69]Open in a new tab

   Biological effects of K-strophanthoside in Neuro-2a cells. (A–D) Impact
   of K-strophanthoside on Nurr1 and Nurr1-associated gene (TH and AADC)
   protein expression in Neuro-2a cells for 24 h. (E) Effect of
   K-strophanthoside on Nurr1-associated gene mRNA expression of TH and
   AADC at a concentration of 30 μM. (F, G) Fluorescence intensity of TH
   and AADC was detected by immunofluorescence after administration of
   K-strophanthoside. *P < 0.05, **P < 0.01. Data are mean ± S.E.M., n ≥
   3, one-way ANOVA in comparison expression with 0 μM compound (DMSO
   only) in Western blot analysis, t-test in comparison expression with 0
   μM compound (DMSO only) in real-time PCR analysis.

4.

   [70]4
   [71]Open in a new tab

   Integrated identification and enrichment analysis of DEGs obtained from
   Nurr1 knockdown cells and K-strophanthoside-treated cells at 30 μM. (A,
   B) Volcano plots for DEGs from Nurr1 knockdown cells (A) and
   K-strophanthoside-treated cells (B). Red and blue dots represent
   upregulated genes (q-value < 0.05 and log2FC ≥ 0.58) and downregulated
   genes (q-value < 0.05 and log2FC ≤ −0.58), respectively. (C) Venn map
   of DEGs indicates common upregulated and downregulated DEGs between
   Nurr1 knockdown cells and K-strophanthoside-treated cells. (D–F) Top 3
   GO enrichment pathways of commonly changed DEGs in three functional
   groups: Biological process (D), cellular component (E), and molecular
   function (F). (G–H) Bubble diagram of the top 3 KEGG enrichment
   pathways for common upregulated DEGs (G) and common downregulated DEGs
   (H). DEGs, differentially expressed genes.

   To further analyze the function of these DEGs, we conducted Gene
   Ontology (GO) and KEGG enrichment analyses. The GO biological process
   terms ([72]Figure D) enriched by common downregulated DEGs were the
   phosphatidylethanolamine biosynthetic process, the polyamine
   biosynthetic process, and the MAPK cascade, while those enriched by
   common upregulated DEGs were RNA polymerase II transcription,
   DNA-templated transcription, and muscle structure development. The
   commonly changed DEGs were significantly enriched in cellular compounds
   of the extracellular matrix, photoreceptor inner segment,
   MPP7-DLG1-LIN7 complex, nucleus, nucleoplasm, and protein–DNA complex
   ([73]Figure E). The molecular functions enriched by commonly changed
   DEGs contain RNA polymerase II-specific DNA-binding transcription
   factor activity, RNA polymerase II cis-regulatory region
   sequence-specific DNA binding, DNA binding, oxidoreductase, hydrolyase,
   and lyase activity ([74]Figure F).

   Moreover, the enrichment and analysis of KEGG pathways by common
   upregulated DEGs were significantly enriched in lysine degradation,
   renal cell carcinoma, and HIF-1 signaling pathway ([75]Figure G), while
   arginine and proline metabolism, beta-alanine metabolism, and valine,
   leucine, and isoleucine degradation were enriched by common
   downregulated DEGs ([76]Figure H). Altogether, these results indicated
   that K-strophanthoside acts as a Nurr1 inverse agonist, mimicking the
   function of Nurr1 knockdown.

Modulation of Nurr1 by K-Strophanthoside Depends on the DNA Response Element

   To fully understand the modulation of Nurr1 by K-strophanthoside in a
   physiological environment, it is essential to understand the effects of
   K-strophanthoside on the reporter activity via modulating the Nurr1
   protein in its monomer, homodimer, or RXR heterodimer forms. We
   constructed luciferase reporter plasmids containing NGFI-B response
   element (NBRE) for the Nurr1 monomer, Nur-response element (NurRE) for
   the Nurr1 homodimer, and direct repeat 5 response element (DR5) for the
   Nurr1-RXR heterodimer ([77]Figure A). The luciferase reporter gene
   assay results showed that AQ robustly enhanced the activities of all
   three response elements in a concentration-dependent manner ([78]Figure
   B–E): AQ (100 μM) activated a 2.72-fold increase in NBRE, a 1.56-fold
   increase in NurRE, and a 1.69-fold increase in DR5. However,
   K-strophanthoside exhibited marked inverse agonism on the NBRE response
   element (49%) and a comparatively lower inverse agonism on NurRE (22%)
   and DR5 (10%) response elements at a concentration of 100 μM
   ([79]Figure B–E). These findings suggest that K-strophanthoside has a
   selective regulatory effect on Nurr1.

5.

   [80]5
   [81]Open in a new tab

   Construction of luciferase reporter gene plasmids and luciferase
   reporter gene assays for Nurr1 modulation by small-molecule compounds.
   (A) DNA response elements for the human Nurr1 monomer (NBRE,
   AAGGTCACAAGGTCACAAGGTCACAAGGTCAC), the Nurr1 homodimer (NurRE,
   TGATATTTACCTCCAAATGCCA), or the Nurr1-RXR heterodimer (DR5,
   GGTTCACCGAAAGGTCA) were fused into pGL3-Basic vector to generate
   transgenic luciferase reporter gene plasmids. (B–D) Full-length Nurr1
   reporter gene assays with human Nurr1 response elements in HEK293T
   cells for 24 h: (B) NBRE for the Nurr1 monomer, (C) NurRE for the Nurr1
   homodimer, (D) DR5 for the Nurr1/RXRα heterodimer. (E) Transcriptional
   activity profiles of small-molecule compounds (100 μM) on Nurr1
   modulation compared with the control group. *P < 0.05. Data are mean ±
   S.E.M., n ≥ 3, one-way ANOVA in comparison expression with 0 μM
   compound.

Discussion

   Nurr1 is a transcription factor of the nuclear receptor family, whose
   expression disturbance is closely related to diseases. , The absence of
   Nurr1 exacerbates the pathology of neurodegenerative diseases, ,, such
   as neuroinflammation and the aggregation of pathological proteins,
   while Nurr1 overexpression or Nurr1 agonists could improve these
   changes. ,,− However, in some cancers − and autoimmune diseases, ,
   Nurr1 upregulation is associated with disease progression.
   Understanding the function of Nurr1 in related diseases provides a
   potential target for the effective treatment of these diseases.
   Therefore, recent studies have increasingly focused on developing Nurr1
   modulators. However, the absence of an effective intrinsic Nurr1
   activity modulator has promoted an urgent need for exploring an
   effective Nurr1 inverse agonist. In this study, we selected the Chemdiv
   compound library, containing about 1.5 million compounds, as the VS
   small-molecule library to screen Nurr1 modulators. K-strophanthoside
   was identified as a Nurr1 modulator, exhibiting a significant
   inhibitory effect on the Nurr1 transcriptional activity and regulating
   the expression of Nurr1 target genes. Our finding provides a novel
   inverse agonist for Nurr1, further indicating the presence of ligands
   for Nurr1.

   Although Nurr1 is traditionally considered to lack a classical
   ligand-binding pocket, recent years have seen the exploration of
   compounds that bind to the Nurr1-LBD and modulate Nurr1 transcriptional
   activity. ,,, Our study reveals that K-strophanthoside binds directly
   to Nurr1-LBD at a micromolar concentration, exhibiting a stronger
   affinity compared to AQ. Amino acid residues (Glu445, Glu514, Arg515,
   and His516) participate in the regulation of Nurr1 transcriptional
   function mediated by K-strophanthoside. Interestingly, the sugar
   moieties of K-strophanthoside appear to contribute to its binding with
   Nurr1. Investigating the specific interactions between these glycosyl
   groups and the Nurr1 protein will be of significant interest in future
   studies, which may provide valuable insights into the structural basis
   of ligand recognition and receptor modulation. In addition, previous
   study identified that residues Ile403, Leu409, Tyr575, and Asp580 are
   essential for the interaction of AQ and Nurr1-LBD, which differ from
   the key residues involved in K-strophanthoside binding. This suggests
   that AQ and K-strophanthoside do not directly compete for the same
   binding site within Nurr1-LBD. Moreover, further studies are needed to
   determine whether K-strophanthoside competes with AQ for binding to
   Nurr1 in order to fully elucidate the mechanism of K-strophanthoside
   action.

   Previous research has proved that both synthetic and endogenous
   ligands, such as nonsteroidal anti-inflammatory drugs and unsaturated
   fatty acids, regulate Nurr1 activity. ,, However, our understanding of
   the cellular effects of ligands on Nurr1 activity within physiological
   environments and their biological correlations is still limited. To
   address this, we further explored the regulation of the cellular
   effects of K-strophanthoside on the Nurr1 activity in the physiological
   environment. A previous study demonstrated that K-strophanthoside is
   capable of penetrating the cell membrane and localizing to the nucleus.
   Our findings further indicate that K-strophanthoside inhibits the
   expression of TH and AADC in a dose-dependent manner, two rate-limiting
   enzymes of dopamine synthesis regulated by Nurr1, , while
   K-strophanthoside does not affect the expression of the Nurr1 protein.
   These results demonstrate that K-strophanthoside functions by
   inhibiting Nurr1 activity rather than altering its expression.
   Furthermore, K-strophanthoside affected the expression of
   Nurr1-regulated genes involved in the polyamine biosynthetic process,
   release of cytochrome c from mitochondria, reactive oxygen species
   metabolic process, apoptotic process, MAPK cascade, transcription
   process, and amino acid metabolism ([82]Supplement Table 1). Similarly,
   an inverse Nurr1 agonist study based on indole found that its inverse
   agonist changed Nurr1 target gene expression in T98G cells and mimicked
   neuroinflammation caused by Nurr1 deficiency. The transcription of all
   protein-coding genes in eukaryotic genomes requires the involvement of
   RNA polymerase II. As a transcription factor, Nurr1 inhibition impacts
   RNA polymerase II and the transcription of the DNA template. In the
   pathological environment, mitochondrial dysfunction can lead to the
   increase in ROS production, a key initiator of cytochrome c release
   into the cytosol. The cytochrome c-initiated pathway, involving its
   release from mitochondria and resulting in caspase activation and
   subsequent cell apoptosis, is a significant caspase activation pathway.
   Moreover, ROS mediates mitogen-activated protein kinase (MAPK)
   cascades, including ERK1/2, JNK, p38, and the BMK1/ERK5, which are the
   major intracellular signaling pathways that regulate a variety of
   cellular processes. Previous research indicated that AQ, a Nurr1
   agonist, increased the level of phosphorylated P38 MAPK. Our study
   provides evidence that Nurr1 regulated cell apoptosis. In summary, our
   findings further demonstrate that K-strophanthoside inhibits Nurr1
   transcriptional activity in the physiological environment and provide
   insights into other potential biological effects mediated by Nurr1.

   Upon ligand interaction with Nurr1, the conformation of the Nurr1
   protein changes, and then, the activity of Nurr1 is regulated. As a
   transcription factor, Nurr1 can function as monomers, homodimers, and
   heterodimers to regulate the transcription of Nurr1 target genes via
   the specific DNA response elements. The Nurr1 monomer binds to the
   NBRE, the homodimer recognizes the NurRE, and the heterodimer formed by
   Nurr1 and RXR binds to DR5. , To comprehensively evaluate the
   regulation of Nurr1 by ligands, we investigate the impact of
   K-strophanthoside on the regulation of NBRE, NurRE, and DR5 by human
   full-length Nurr1 reporter gene experiments. However, unlike other
   inverse Nurr1 agonists, , K-strophanthoside appears to tend to modulate
   the transcriptional activity of the Nurr1 monomer in a dose-dependent
   manner rather than the homodimer or the heterodimer. Similarly, some
   nonsteroidal anti-inflammatory drugs (MFA, parecoxib, oxaprozin) have
   also been shown to act as inverse Nurr1 agonists. However, parecoxib
   and oxaprozin displayed inverse agonism on NBRE, NurRE, and DR5, while
   MFA only inhibited the activity of NurRE and DR5 but not NBRE. These
   results suggest that Nurr1 could potentially be selectively regulated
   by chemical tools. However, given the moderate potency of
   K-strophanthoside on Nurr1 inverse activation, it is necessary to
   identify the functional groups of K-strophanthoside and optimize the
   compounds to screen for the most effective small-molecule compounds in
   future studies.

   As members of the NR4A family, Nurr1, Nur77, and NOR1 share a
   significant structural homology. It is therefore essential to establish
   whether K-strophanthoside exhibits a specific binding affinity for
   Nurr1, rather than Nur77 or NOR1. The current unavailability of NOR1’s
   three-dimensional structure precludes a comprehensive investigation of
   its potential interaction with K-strophanthoside. Moreover, our study
   was limited to computational predictions, suggesting the absence of
   direct binding between K-strophanthoside and Nur77. However, this
   finding has not been experimentally validated. Consequently, future
   studies should be designed to address these limitations. In summary, we
   identified a novel inverse Nurr1 agonist, K-strophanthoside, which
   effectively inhibits the receptor’s intrinsic transcriptional activity,
   demonstrating that Nurr1 activity can be selectively regulated by its
   ligands. These results contribute to a better understanding of the
   activity of Nurr1 and promote the development of modulators aimed at
   different structural forms of Nurr1 receptors to verify the function of
   Nurr1 and assess the therapeutic effects of drugs targeting Nurr1.

Materials and Methods

Virtual Screening (VS)

   The dock module in the molecular operating environment (MOE) was used
   for structure-based VS (SBVS). Nurr1 in the PDB (PDB code: [83]6DDA)
   was defined as the receptor, and the binding site was set around the
   native ligand. The procedure of virtual screening is shown in
   [84]Figure S8. We initially conducted high-throughput rigid docking
   between Nurr1-LBD and 1.5 million compounds from the ChemDiv compound
   library, obtaining London dG scores. Subsequently, the top 15,000
   compounds, ranked according to their London dG scores, were subjected
   to flexible redocking with Nurr1 using the “induced fit” protocol. The
   top 100 poses, ranked according to the London dG scores, performed the
   force field refinement and re-evaluated using the GBVI/WSA dG scoring
   function. Following the flexible docking of the 15,000 compounds, we
   employed fingerprint-based clustering to categorize them into
   structural clusters, applying a distance threshold of 0.5 between
   clusters. From each cluster, we identified the top 100 cluster centers
   as potential hits based on the highest-ranked compound within each
   cluster. Finally, the top 10 compounds were selected for further
   functional validation. The SBVS and molecular docking were performed by
   Wecomput Technology Co., Ltd.

   To further retrospectively validate our docking procedure, we chose a
   set of known Nurr1 modulators that have been assessed for their ability
   to bind to Nurr1-LBD. The Nurr1 protein structure (PDB code: [85]6DDA)
   was prepared by the QuickPrep module in MOE. All compounds were
   energy-minimized, docked with the same protocol, and ranked according
   to their docking scores. To evaluate the predictive performance of the
   docking approach, we generated an ROC curve. The AUC was 0.778
   ([86]Figure S11), indicating good predictive accuracy of the docking
   method.

Compounds and Antibodies

   The top 10 compounds were purchased from commercial sources (ChemDiv
   Inc.), with a purity of at least 95%. AQ was purchased from TargetMol.

   The antibodies were shown as follows: anti-Nurr1 antibody
   (Proteintech), anti-GAPDH antibody (Cell Signaling Technology), anti-TH
   antibody (Millipore), anti-AADC antibody (Santa Cruz Biotechnology),
   goat anti-mouse IgG H&L (Proteintech), goat anti-rabbit IgG H&L
   (Proteintech), goat anti-rabbit Alexa 594 (Cell Signaling Technology),
   and goat anti-mouse Alexa 488 (Cell Signaling Technology).

SPR

   A Biacore T200 instrument was used for the SPR analysis to investigate
   the binding affinity between the human Nurr1 protein (Abcam) and
   compounds. The amino coupling method was adopted by the CM5 chip. The
   protein system contains Tris and was replaced with a pH 5.0 sodium
   acetate solution by an ultrafiltration tube. The protein concentration
   after ultrafiltration was about 84 μg/mL.

   The coupling conditions were as follows: a concentration of
   approximately 21 μg/mL, a system of pH 5.0 sodium acetate solution, a
   chip activation time of 420 s, and a sealing time of 420 s. The protein
   coupling amount of Nurr1 was 5950 RU. SPR binding experiments were
   performed in the same running buffer, PBS (pH 7.4) supplemented with 5%
   dimethyl sulfoxide (DMSO). Based on the solubility of the sample to be
   measured and the core surface test, a series of concentrations were
   prepared (AQ: 125, 62.5, 31.25, 15.63, 7.813, 3.906, 1.953 μM, compound
   7: 250, 125, 62.5, 31.25, 15.63, 7.813, 3.906 μM). The test conditions
   for the compounds to be measured were an injection time of 60 s, a
   dissociation time of 60 s, and no regeneration. Binding curves were
   analyzed by using Biacore-provided software.

Luciferase Reporter Assays

   To conduct transactivation assays based on full-length Nurr1, we
   constructed related reporter gene plasmids containing the human Nurr1
   response elements DR5 (pGL3-Basic-DR5; GGTTCACCGAAAGGTCA), NBRE
   (pGL3-Basic-NBRE; AAGGTCACAAGGTCACAAGGTCACAAGGTCAC), or NurRE
   (pGL3-Basic-NurRE; TGATATTTACCTCCAAATGCCA). Nurr1 and RXRα were
   overexpressed for DR5. The Renilla luciferase expression vector pGL4.74
   was obtained from Miaoling Biotechnology (Wuhan, China). The
   full-length human Nurr1 expression plasmid (pCMV-hNurr1) contains the
   full-length cDNA encoding human Nurr1 (amino acids 1–598).
   Site-directed mutagenesis was performed to generate specific Nurr1
   point mutants. The following nucleotide substitutions were introduced
   to produce alanine substitutions at the indicated positions: Glu445
   (GAA → GCA), Leu509 (CTG → GCG), Glu514 (GAG → GCG), Arg515 (AGA →
   GCA), and His516 (CAC → GCC).

   HEK293T cells were cultured and transfected as previously described.
   For assays on full-length human Nurr1, the plasmid mixtures were
   pCMV-hNurr1/pGL3-Basic-NBRE/pGL4.74 (NBRE),
   pCMV-hNurr1/pGL3-Basic-NurRE/pGL4.74 (NurRE), and
   pCMV-hNurr1/pCMV-RXRα/pGL3-Basic-DR5/pGL4.74 (DR5). The cells were
   cultured with respective concentrations of compounds and incubated for
   24 h to detect the luciferase activity by a Dual Luciferase Reporter
   Gene Assay Kit (Beyotime). The firefly luciferase data were normalized
   by dividing them with the renilla luciferase data, followed by
   calculating the fold activation of compounds at various concentrations.
   In the Nurr1 mutagenesis experiments, fold changes were calculated by
   comparing normalized values in the presence versus absence of
   K-strophanthoside for each mutant. These values were subsequently
   compared to those from wild-type Nurr1 under the same conditions to
   determine the impact of each mutation.

Cell Viability Assay

   Neuro-2a cells and HEK293T cells were seeded into 96-well plates (10^4
   cells/well), and compounds at various concentrations were added to at
   least three wells. After incubation for 24 h, CCK-8 reagent (Meilunbio)
   was added, and the mixture was incubated at 37 °C and 5% CO[2] for 1–2
   h. Absorbance at 450 nm was measured, and cell survival rates were
   calculated.

Western Blot

   Cells after culture with K-strophanthoside for 24 h were resuspended in
   RIPA buffer (Beyotime) with protease inhibitor cocktails
   (Sigma-Aldrich) and centrifuged at 12,000 rpm for 15 min. BCA assay kit
   (Takara) was used to determine the protein concentration, and sodium
   dodecyl sulfate (SDS) loading buffer (Beyotime) was added to denature
   the proteins. The protein samples were separated by SDS-polyacrylamide
   gel electrophoresis (PAGE) and transferred to polyvinylidene difluoride
   (PVDF) membranes. The membranes were subsequently blocked and incubated
   with antibodies. Chemiluminescence detection kit (Abbkine) was used to
   visualize protein bands, and ImageJ was used to analyze the results.
   The intensities of proteins were normalized to those of GAPDH.

Quantitative PCR

   Cells cultured with K-strophanthoside at 30 μM for 24 h were used for
   total RNA extraction using TRIzol reagent (Takara). After cDNA
   synthesis (TransGene Biotech), mRNA levels were analyzed by an SYBR
   Green RT-PCR kit (TransGene Biotech) and an ABI Prism 7500 assay system
   (Applied Biosystems). The results were analyzed by the 2-ΔΔCt formula.
   The primers for the real-time PCR assay were shown as follows: GAPDH
   (F: GCATTGTGGAAGGGCTCATG, R: AGGGATGATGTTCTGGGCAG); TH (F:
   CAATACAAGCAGGGTGAGCC, R: TAGCATAGAGGCCCTTCAGC); AADC (F:
   GAGCTGGAGACCGTGATGAT, R: TTAGTCCGAGCAGCCAGTAG); Nurr1 (F:
   ATTCCAGGTTCCAGGCAAAC, R: AGCAAAGCCAGGGATCTTCT).

Immunofluorescence Staining

   After administration of K-strophanthoside for 24 h, cells were
   processed as previously described: they were blocked with 2% BSA,
   incubated with antibodies, and subsequently imaged by a confocal
   microscope (Nikon).

RNA Isolation and Library Preparation

   Neuro-2a cells were cultured with K-strophanthoside at a concentration
   of 30 μM for 24 h, and the total RNA was extracted using the TRIzol
   reagent (Invitrogen). RNA purity, quantification, and integrity were
   evaluated. Then, the libraries were constructed using the VAHTS
   Universal V6 RNA-seq Library Prep Kit.

RNA Sequencing and Differentially Expressed Gene Analysis

   The transcriptome sequencing and analysis were performed by OE Biotech
   Co., Ltd. Differential expression analysis was performed using DESeq2,
   with significance thresholds set at a q-value <0.05 and
   |log2foldchange| ≥ 0.58. Hierarchical cluster analysis, along with GO
   and KEGG pathway enrichment analysis of DEGs, was carried out using R
   (v 3.2.0).

Statistical Analysis

   The experimental data obtained in this study were analyzed by the
   GraphPad Prism 7 system, and t-test or one-way ANOVA was used for
   statistical analysis. P < 0.05 was considered statistically
   significant. All data are presented as the mean ± standard error of the
   mean (S.E.M.).

Supplementary Material

   [87]ao5c06698_si_001.pdf^ (1.3MB, pdf)

Acknowledgments