Enhancement of ABCA1 and ABCG1 Expression and Cholesterol Efflux by a Metabolite of Tipelukast: A Potential Therapeutic Strategy for Atherosclerosis

Abstract

Aims: MN-001 (tipelukast), a compound with lipid-modulating and anti-inflammatory properties, and its active metabolite MN-002, have been suggested to influence cholesterol metabolism. This study aimed to investigate whether MN-001 and MN-002 enhance cholesterol efflux via ABCA1 and ABCG1, thereby reducing foam cell formation. We also evaluated cholesterol efflux capacity in patients with diabetes before and after MN-001 administration.

Methods: Cholesterol efflux was assessed in THP-1 macrophages treated with MN-001 and MN-002 in the presence of ApoA-I or HDL. ABCA1 and ABCG1 expression were evaluated using western blot and qPCR analyses. A 12-week observational study in patients with diabetes evaluated the cholesterol efflux capacity using ApoB-depleted serum and radiolabeled J774.1 macrophages. Molecular docking simulations were conducted to explore MN-002 binding affinities, aiming to identify potential target proteins and elucidate the molecular mechanisms underlying their effects on cholesterol metabolism.

Results: MN-002 enhanced ABCA1-mediated cholesterol efflux and upregulated ABCA1 expression independently of PKA. It also increased ABCG1 expression; however, neither MN-001 nor MN-002 influenced HDL-mediated efflux. MN-001 showed no significant improvement in cholesterol efflux capacity (p = 0.6507) in patients with diabetes. Molecular docking simulations indicated that MN-002 may bind to PPAR-alpha, suggesting a potential mechanism for its effects.

Conclusion: MN-002 offers a novel therapeutic approach for atherosclerosis by upregulating ABCA1 and ABCG1 expression and enhancing ApoA-I-mediated cholesterol efflux. Further studies are required to clarify the underlying mechanisms and assess their clinical potential in atherosclerosis and metabolic disorders.

See editorial vol. 33: 26-28

Introduction

Atherosclerotic cardiovascular diseases (ASCVD) remain a leading cause of mortality in developed countries, primarily driven by dyslipidemia and chronic inflammation ^{1 - 3)} . Low-density lipoprotein cholesterol (LDL-C) is a well-established major risk factor for ASCVD, and LDL-C-lowering therapies, such as statins, have significantly reduced the incidence and mortality associated with ASCVD ^{1 , 2)} . However, even with optimal LDL-C reduction, a substantial residual risk persists. This residual risk is driven by factors such as elevated triglycerides (TG), low high-density lipoprotein cholesterol (HDL-C) levels, and chronic inflammation, which contribute to the progression of atherosclerosis ^{4 - 6)} .

Addressing this residual risk requires therapeutic strategies beyond lowering the LDL-C levels. Targeting the alternative pathways involved in lipid metabolism and inflammation is essential for advancing ASCVD management ^{7 , 8)} . Therapies targeting LDL-C, such as statins and PCSK9 inhibitors, have been proven effective in lowering LDL-C levels and reducing cardiovascular events ⁹⁾ . However, even with optimal LDL-C reduction, significant residual risk persists owing to factors such as high triglycerides, low HDL-C, and unresolved inflammation ^{4 , 10)} . This underscores the need for additional therapies addressing these non-LDL-related contributors ^{11 , 12)} .

Cholesterol efflux from macrophages in arterial walls is crucial for preventing foam cell formation and mitigating plaque development ¹³⁾ . This process is regulated by ATP-dependent transporters ABCA1 and ABCG1, which facilitate cholesterol removal from cells and prevent intracellular cholesterol accumulation ^{14 , 15)} . These transporters are controlled by nuclear receptors such as liver X receptor (LXR)-alpha and peroxisome proliferator-activated receptors (PPARs). LXR-alpha is activated by intracellular cholesterol buildup, promoting the transcription of ABCA1 and ABCG1, enhancing cholesterol efflux, and reducing inflammation ¹⁶⁾ . Similarly, PPAR-alpha and PPAR-gamma regulate lipid and glucose metabolism and promote HDL-cholesterol levels, further supporting cholesterol homeostasis ¹⁷⁾ . The protein kinase A (PKA) signaling pathway also influences cholesterol uptake and efflux by integrating multiple metabolic and inflammatory signals ¹⁸⁾ .

MN-001 (tipelukast) is an orally administrable small-molecule compound with multiple mechanisms of action, including leukotriene (LT) receptor antagonism, phosphodiesterase (PDE) 3/4 inhibition, 5-lipoxygenase (5-LO) inhibition, phospholipase C inhibition, and thromboxane A2 inhibition ^{19 , 20)} . Following its discovery, this compound was used in the developed of treatments for asthma and interstitial cystitis. MN-001 had been evaluated in more than 600 subjects in six Phase I studies and five Phase II studies with favorable safety and tolerability profiles (MN-001-CL-001, MN-001-CL-002, MN-001-CL-003/NCT 00295854, and MN-001-CL-004). Notably, previous clinical trials unexpectedly revealed that serum TG levels decreased in MN-001-treated groups, raising expectations about its potential impact on lipid metabolism.

It has been shown that 5-LO expression and LT formation were increased in livers in animals with induced cirrhosis ^{21 , 22)} , suggesting that 5-LO plays an important role in Kupffer cell survival and in the pathogenesis of liver inflammation and fibrosis. Thus, MN-001 is thought to exert an inhibitory effect on 5- LO and the 5-LO/LT pathway. Therefore, it was considered to have potential utility for treating fatty liver disease.

MN-001 treatment significantly suppressed hepatic lipid accumulation and fibrosis in preclinical studies with non-alcoholic steatohepatitis (NASH) and advanced NASH mouse models ¹⁹⁾ . These results suggested that MN-001 may have therapeutic potential for patients with non-alcoholic fatty liver disease (NAFLD) or NASH accompanied by dyslipidemia.

In this study, we focused on MN-002, the major metabolite of MN-001. MN-002 is structurally similar to MN-001 and is believed to have lipid-modulating and anti-inflammatory effects, although its precise mechanism of action is not yet fully understood. Based on its chemical similarity to MN-001 and preliminary observations, MN-002 is hypothesized to act as a lipid-regulating agent via nuclear receptor signaling pathways, potentially offering therapeutic benefits in atherosclerosis and other metabolic disorders. Thus, the aim of this study was to investigate whether MN-002 exerts anti-atherosclerotic effects by enhancing cholesterol efflux through the increased expression of ABCA1 and ABCG1 mediated through the PPAR-alpha and LXR-alpha pathways.

Aim

We hypothesized that MN-001 and MN-002 enhance cholesterol efflux by upregulating ABCA1 and ABCG1 expression. The aim was to assess their therapeutic potential by evaluating their effects on cholesterol efflux and ABCA1 and ABCG1 expression.

Methods

Materials

MN-001 (tipelukast, MediciNova, San Diego, CA, USA), MN-002 (a metabolite of MN-001, MediciNova, CA, USA), T0901317 (T090, Sigma-Aldrich, St. Louis, MO, USA), and the PKA inhibitor H89 2HCl (Selleckchem, Houston, TX, USA) were dissolved in dimethyl sulfoxide (DMSO; Sigma-Aldrich). The chemical structures of MN-001 and MN-002 are illustrated in Fig.1A and Fig.1B , respectively. Human apolipoprotein A-I (ApoA-I) was obtained from Sigma-Aldrich. [1,2-^3H(N)]-cholesterol (437 MBq/mL, 1.0 mCi/mL) was purchased from Perkin-Elmer Analytical Sciences (Waltham, MA, USA).

Fig.1. Chemical structures of MN-001 and MN-002.

(A) Chemical structure of MN-001.

(B) Chemical structure of MN-002.

Cell Culture

THP-1 cells (Riken Cell Bank, Tokyo, Japan) were cultured in RPMI-1640 medium (FUJIFILM Wako Pure Chemical Corporation,, Tokyo, Japan) supplemented with 10% fetal bovine serum (FBS; Sigma-Aldrich). Differentiation of THP-1 monocytes into macrophages was induced by incubating the cells with 100 nM phorbol 12-myristate 13-acetate (PMA) for 72 hours. After differentiation, the cells were washed with phosphate-buffered saline (PBS) and treated with MN-001, MN-002, T090, or DMSO. The cells were then cultured in serum-free DMEM (FUJIFILM Wako Pure Chemical Corporation,) containing 1% fatty acid-free bovine serum albumin (BSA, Sigma-Aldrich) for 24–48 hours.

Real-time Quantitative RT-PCR

RNA transcript expression was quantified using real-time quantitative PCR (qPCR). Total RNA was extracted from the cells using an RNeasy Plus Mini Kit (Qiagen, Germantown, MD, USA) at specified time points following treatment with MN-001, MN-002, or T090. RNA concentrations were determined by measuring absorbance at 260 and 280 nm. First-strand cDNA was synthesized from total RNA using the SuperScript IV VILO Master Mix with ezDNase Enzyme (Invitrogen, Carlsbad, CA, USA). Quantitative PCR was performed using the QuantStudio 3 system (Thermo Fisher Scientific, Waltham, MA, USA) with Fast SYBR Green Master Mix (Thermo Fisher Scientific) to determine mRNA levels. PCR primers specific for human ABCA1, ABCG1, PPAR-α, PPAR-γ, LXR-α, and β-actin were synthesized by Integrated DNA Technologies (IDT Inc., Coralville, IA, USA). Primer sequences used for each gene are listed below.

• ABCA1:

- forward,

5ʹ-CTGCTGCCGCATCTCACTG-3ʹ

- reverse,

5ʹ-TCCCTTCTGCCTTCATCCTTC-3ʹ

• ABCG1:

- forward,

5ʹ-GGAAGAGACTGCTAATTGCCAGACGG-3ʹ

- reverse,

5ʹ-GCTGACAAATGTGTACTGTTCGTTGTACATC-3ʹ

• PPAR-α:

- forward,

5ʹ-CCAGTATTTAGGAAGCTGTCCTG-3ʹ

- reverse,

5ʹ-CGTTGTGTGACATCCCGACAG-3ʹ

• PPAR-γ:

- forward,

5ʹ-TGGAATTAGATGACAGCGACTTGG-3ʹ

- reverse,

5ʹ-CTGGAGCAGCTTGGCAAACA-3ʹ

• LXR-α:

- forward,

5ʹ-TGGACACCTACATGCGTCGCAA-3ʹ

- reverse,

5ʹ-CAAGGATGTGGCATGAGCCTGT-3ʹ

• β-actin:

- forward,

5ʹ-CATGTACGTTGCTATCCAGGC-3ʹ

- reverse,

5ʹ-CTCCTTAATGTCACGCACGAT-3ʹ.

Melting curve analyses were conducted for all qPCR products to confirm the specificity of the amplified products and the absence of primer-dimer artifacts. Relative expression levels were calculated by normalizing the expression of each target gene to that of the β-actin gene.

Molecular Docking Simulation Verification

Molecular docking simulations were performed to verify the binding affinity of the compounds for the target protein. The Q07869/ PPAR-α-HUMAN target was selected for molecular docking experiments. The 3D structure of the compound was retrieved from PubChem, and multiple conformations were generated using the “conformational search” feature in Molecular Operating Environment (MOE) software (Chemical Computing Group, Montreal, Canada). Eight conformations were produced based on specific parameters. PPAR-α-HUMAN, comprising 468 amino acids, was modeled using the crystal structure with PDB ID: 6LXA, which features a distinct small-molecule binding site (site 1) for docking. During the docking process, the receptor was protonated using the amber10 force field, and five docking regions were defined. The induced-fit docking protocol was used by applying the triangle-matching algorithm to generate docking modes. The London δG was used as the scoring function to assess binding energy for each pose, and the top 100 poses were retained. These poses were further optimized using the induced-fit algorithm with the GBVI/WAS δG scoring function, allowing the calculation of binding affinity for the optimized poses. The top five docking positions of the compounds were selected based on the docking regions, and the conformation with the highest docking score was selected for analysis. In the binding model, the compound is depicted in green and the protein is represented as orange cartoon structures ( Supplementary Fig.1 ) .

Supplementary Fig.1. Molecular docking for MN-002 with PPAR-alpha and LXR-alpha.

(A) 3D structures of PPAR-alpha_HUMAN.

(B) Binding model. The green spherical models represent small molecules, and the orange cartoons represent proteins.

(C) Binding model. The orange cartoon represents the structure of the PPAR-alpha protein. The green spheres represent the MN-002 small molecule, and the blue ones represent the EPA molecule.

(D) Binding model. The orange cartoon represents the structure of the PPAR-alpha protein. The green sticks represent the MN-002 small molecule, and the blue ones represent the EPA (eicosapentaenoic acid) molecule.

(E) Binding model. The orange cartoon represents the structure of the PPAR-alpha protein. The green spheres represent the MN-002 small molecule, and the red ones represent the pemafibrate molecule.

(F) Binding model. The orange cartoon represents the structure of the PPAR-alpha protein. The green sticks represent the MN-002 small molecule, and the red ones represent the pemafibrate molecule.

(G) Binding model. The orange cartoon represents the structure of the PPAR-alpha protein. The green spheres represent the MN-002 small molecule, and the yellow ones represent the fenofibrate molecule.

(H) Binding model. The orange cartoon represents the structure of the PPAR-alpha protein. The green sticks represent the MN-002 small molecule, and the yellow ones represent the fenofibrate molecule.

(I) Binding model. The orange cartoon represents the structure of the PPAR-alpha protein. The green spheres represent the MN-002 small molecule, and the magenta ones represent the bezafibrate molecule.

(J) Binding model. The orange cartoon represents the structure of the PPAR-alpha protein. The green sticks represent the MN-002 small molecule, and the magenta ones represent the bezafibrate molecule.

(K) Binding model. The silver cartoon represents the structure of the LXR-alpha protein. The green sticks represent the MN-002 small molecule.

(L) Binding model. The silver cartoon represents the structure of the LXR-alpha protein. The green sticks represent the T0901317 small molecule.

Western Blot Analysis

Cells were harvested and lysed using Cell Lysis Buffer (Cell Signaling Technology, Danvers, MA, USA) to extract total proteins, which were subsequently quantified using the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific). For SDS-PAGE, equal amounts of protein (1–5 µg per lane) were loaded onto a 10% Mini-PROTEAN TGX gel (Bio-Rad, Hercules, CA, USA) and separated by electrophoresis. Following electrophoresis, the proteins were transferred onto polyvinylidene difluoride (PVDF) membranes. The membranes were then probed with the following primary antibodies: mouse anti-ABCA1 (ab18180, Abcam, Cambridge, UK), rabbit anti-ABCG1 (NB400-132, Novus Biologicals, Centennial, CO, USA), and mouse anti-β-actin (A5316, Sigma-Aldrich). The membranes were subsequently incubated with horseradish peroxidase (HRP)-conjugated secondary antibodies: anti-mouse IgG (Cell Signaling Technology, 7076S) or anti-rabbit IgG (Cell Signaling Technology, 7074S). Protein bands were detected using enhanced chemiluminescence (ECL) (Bio-Rad, Hercules, CA, USA) and visualized with the ImageQuant LAS 4000 system (Fujifilm Life Science, Tokyo, Japan). The intensity of the protein bands was quantified using the ImageJ software (NIH, Bethesda, MD, USA). The expression levels of ABCA1 and ABCG1 were normalized to β-actin, which was used as an internal control.

Cholesterol Efflux Assay

THP-1 cells were differentiated into macrophages by treatment with phorbol 12-myristate 13-acetate (PMA) for 72 hours. After differentiation, the cells were radiolabeled by incubating overnight with [^3H]-cholesterol (1.0 µCi/mL) in DMEM supplemented with 1% fatty acid-free BSA. The following day, the medium was removed, and the cells were washed twice with PBS. Subsequently, the cells were treated with the test compounds MN-001, MN-002, or T0901317 (positive control), with or without ApoA-I (10 µg/mL), and incubated for 24 hours. This time frame was selected based on preliminary studies showing that shorter ApoA-I exposure (2–16 hours) failed to induce consistent efflux and that prolonged preincubation with MN-002 (48 hours) reduced efficacy. The 24-hour co-incubation enabled both ABCA1 upregulation and reproducible cholesterol efflux under low ApoA-I conditions. Alternatively, the cells were treated with MN-001, MN-002, or T090 for 24 hours, followed by incubation with or without HDL (25 µg/mL) for an additional 4 hours. The efflux medium was carefully collected after the incubation period, and the cells were lysed with 0.1 N NaOH to obtain the cell-associated radioactivity. Both medium and cell lysates were analyzed for radioactivity using a liquid scintillation counter. The cholesterol efflux capacity was determined by calculating the percentage of radioactivity in the medium relative to the total radioactivity (medium plus cells) according to the following formula:

{3 H-cholesterol in medium/(3 H-cholesterol in medium + 3 H-cholesterol in cells)} × 100 – the percentage of passive diffusion in the case of no acceptor (ApoA-I).

Study Participants

This cross-sectional study included 19 patients from the Southern California Research Center, Scripps Clinic, and the University of Washington. The participants, aged 27–67 years, included both males and females. Most patients were Caucasian, with one participant identified as a Native Hawaiian or Pacific Islander. In terms of ethnicity, the majority were Hispanic or Latino. All patients had prior medical histories with common diagnoses including hypertension, fatty liver, type 2 diabetes, hypertriglyceridemia, and obstructive sleep apnea. The procedures and potential risks associated with the study were thoroughly explained to all patients, and written informed consent was obtained. This study adhered to the ethical standards and was conducted in accordance with the Declaration of Helsinki, with approval from the ethics committees of each participating institution.

The Clinical Study of MN-001

The basic characteristics of the study participants are summarized in Supplementary Table 1 . This study was designed as a multicenter, non-randomized, open-label Phase II clinical trial sponsored by MediciNova, Inc. The study was conducted at multiple academic institutions, including Scripps Clinic (Torrey Pines, San Diego, California), where the first Institutional Review Board (IRB) review was performed. The study protocol was approved by the Scripps Institutional Review Board under Protocol No. IRB-16-6736 on January 26, 2016 ²³⁾ . Other participating institutions included the Southern California Liver Center and several additional clinical sites that used a central IRB, the Western Institutional Review Board (WIRB), which approved the study under Protocol No. MN-001-NATG-201 on February 10, 2017 ²⁴⁾ . The study population consisted of adults aged 21 to 65 years who were diagnosed with NASH via liver biopsy or confirmed to have NAFLD by ultrasound with hypertriglyceridemia (fasting serum triglyceride level ≥ 150 mg/dL). The dosing regimen consisted of an initial administration of MN-001 at 250 mg/day for 4 weeks, followed by 500 mg/day (250 mg twice daily) for 8 weeks (MN-001-NATG-201, ClinicalTrials.gov Identifier: NCT02681055) ²⁵⁾ . Outcome assessments were performed at different time points depending on the parameter. Serum lipid profiles, including triglyceride levels, were evaluated at week 8 to capture early treatment effects, while hepatic fat content was assessed after 12 weeks of administration. Subgroup analyses based on the presence or absence of type 2 diabetes mellitus (T2DM) were conducted post hoc and were not pre-specified in the original study protocol.

Supplementary Table 1. Changes in cholesterol efflux capacity before and 12 weeks after administration of MN-001.

Subject	All (n = 19)	With T2DM (n = 10)	Without T2DM (n = 9)
Mean Age (y)	54.6	56.7	52.2
Gender
Male/Female	8 / 11	4 / 6	4 / 5
Race Caucasian	18	10	8
Pacific Islander	1	0	1
Mean Weight (kg)	91.3	92.3	90.2

The table shows individual cholesterol efflux capacity values measured before and 12 weeks after administration of MN-001 in patients. Cholesterol efflux capacity was assessed using radiolabeled cholesterol in serum samples collected from patients, and the values are expressed as relative ratios. Each row represents data from an individual patient, identified by their unique patient number.

Assessment of Cholesterol Efflux Capacity

The cholesterol efflux capacity of the patients was quantified using a slightly modified method ²⁶⁾ . THP-Blood was collected from patients with diabetes and controls, and after centrifugation, the serum was frozen at −70℃ until use. Apolipoprotein B (ApoB)-depleted serum was obtained by precipitating ApoB-containing lipoproteins from whole serum using a polyethylene glycol solution ²⁷⁾ . J774.1 cells (National Institute of Biomedical Innovation, Osaka, Japan), derived from a murine macrophage cell line, were radiolabeled with 0.33 µCi of 3H-cholesterol (Perkin-Elmer Analytical Sciences) per milliliter for 24 hours. The cells were then washed with PBS and incubated for 4 hours in an efflux medium containing 2.8% ApoB-depleted serum. All steps were performed in the presence of the acyl–coenzyme A: cholesterol acyltransferase inhibitor Sandoz 58-035 (Santa Cruz Biotech, Dallas, TX, USA, 2 µg per milliliter). Liquid scintillation counting (Perkin-Elmer Analytical Sciences) was used to quantify the efflux of radioactive cholesterol from the cells. The amount of radioactive cholesterol incorporated into cellular lipids in control wells not exposed to patient serum was calculated using hexane and isopropanol extractions. The efflux percentage was calculated using the following formula:

[(3H-cholesterol in media containing 2.8% ApoB-depleted serum (µCi) – 3H-cholesterol in serum-free media (µCi)) / 3H-cholesterol in cells extracted before the efflux step (µCi)] × 100.

All assays were performed in duplicates. Pooled serum from 11 healthy volunteers was included in each plate, and the values of serum samples from the patients were normalized to the pooled sample value in subsequent analyses to correct for inter-assay variations across plates.

Statistical Analysis

All reported p-values are two-tailed, with statistical significance set at p＜0.05. All results are presented as the mean±standard deviation (SD), unless otherwise specified. All experiments were conducted in triplicate unless stated otherwise in the figure legends. Statistical differences between three or more groups were assessed using Dunnett’s test or Student’s t-test. Analyses were performed using JMP software, version 12.0 or JMP Pro 17 (SAS Institute Inc., Cary, NC, USA).

Results

1. MN-002 Enhances Cholesterol Efflux in THP-1 Macrophages

We investigated the effects of MN-001 and MN-002 on ApoA-I- and HDL-mediated cholesterol efflux in THP-1-derived foam cells using a 24-hour co-incubation protocol ²⁸⁾ . Under these conditions, MN-002 significantly increased ApoA-I-mediated cholesterol efflux (T0901317 (positive control) vs. control: 3.09, p＜0.0001; MN-002 vs. control: 1.53, p = 0.0377; Fig.2A ), while MN-001 had no significant effect (MN-001 vs. control: 1.17, p = 0.6318; Fig.2A ). In contrast, neither compound affected HDL-mediated efflux (T0901317 (positive control) vs. control: 1.59, p＜0.0001; MN-001 vs. control: 1.04, p = 0.9331; MN-002 vs. control: 1.07, p = 0.7652; Fig.2B ). To further evaluate the time-dependent effects of MN-001 and MN-002 on cholesterol efflux, we conducted additional experiments using shorter incubation durations. Under the 4-hour ApoA-I exposure condition, cholesterol efflux was highly variable and showed no significant increase in the MN-001 or MN-002 group compared to the control. Similarly, HDL-mediated cholesterol efflux following 16-hour exposure did not show statistically significant enhancement by either MN-001 or MN-002 ( Supplementary Fig.2 ) . These findings support the selected conditions in our main experiments, which yielded more consistent and robust results for MN-002.

Fig.2. MN-002 promotes ApoA-I-specific cholesterol efflux in THP-1 macrophages.

THP-1 macrophages were labeled with [³H]-cholesterol, followed by treatment with MN-002 (10 µM), T0901317 (10 µM, used as a positive control), or the solvent vehicle (DMSO). The cells were then treated with MN-001, MN-002, or T090 in the presence or absence of ApoA-I (10 µg/mL) and incubated for 24 hours (Fig. 2A). Alternatively, the cells were treated with MN-001, MN-002, or T090 for 24 hours, followed by an additional incubation with or without HDL (high-density lipoprotein) (25 µg/mL) for 4 hours (Fig. 2B). Cholesterol efflux was quantified by calculating the percentage of radioactivity released into the media relative to the total radioactivity (media plus cell-associated radioactivity). Data are presented as the mean±standard deviation of three independent experiments. DMSO, dimethyl sulfoxide. p＜ 0.05 vs. control (Dunnett’s test). n.s.: not significant (p＞0.05), compared to the control unless otherwise specified.

Supplementary Fig.2. Cholesterol efflux under modified incubation durations.

(A) ApoA-I–mediated cholesterol efflux after shortened (4-hour) exposure

(B) HDL-mediated cholesterol efflux after prolonged (16-hour) exposure

Cholesterol efflux was quantified by calculating the percentage of radioactivity released into the media relative to the total radioactivity (media plus cell-associated radioactivity). Data are presented as the mean±standard deviation of three independent experiments. DMSO, dimethyl sulfoxide. p＜0.05 vs. control (Dunnett’s test). n.s.: not significant (＞0.05), compared to the control unless otherwise specified.

2. Upregulation of ABCA1 and ABCG1 Protein Levels by MN-002

We evaluated whether MN-002 treatment leads to increased expression of ABCA1 and ABCG1 at the protein level. As shown in Fig.3A , MN-002 treatment significantly increased the protein levels of both ABCA1 and ABCG1. Specifically, Fig.3B depicts the relative protein expression levels of ABCA1 and ABCG1, normalized to β-actin. MN-002, but not MN-001, significantly increased the protein expression of ABCA1 (5 µM: 1.83, p = 0.0134; 10 µM: 1.99, p = 0.0424) and ABCG1 (10 µM: 2.82, p = 0.0083) following 48 hours of treatment.

Fig.3. Upregulation of ABCA1 and ABCG1 protein levels by MN-002.

(A) Protein levels of ABCA1 (ATP-binding cassette transporter A1) and ABCG1 (ATP-binding cassette transporter G1) following MN-002 treatment. THP-1 cells were differentiated and subsequently treated with MN-001 (10 µM) or MN-002 (2.5, 5, and 10 µM) for 24 hours. The protein expression levels of ABCA1 and ABCG1 were assessed using western blot analysis. The solvent vehicle (DMSO) and T0901317 (0.1 µM) were used as negative and positive controls, respectively.

(B, C) Expression data are presented as the mean±standard deviation from three independent experiments. ^＊p＜0.05 vs. control (Dunnett’s test). DMSO, dimethyl sulfoxide

3. Induction of ABCA1 and ABCG1 gene Expression by MN-002

As pivotal regulators of reverse cholesterol transport (RCT), ABCA1 and ABCG1 are essential for maintaining cellular cholesterol homeostasis and play critical roles in preventing lipid accumulation within macrophages—a key event in the pathogenesis of atherosclerosis. In this study, we comprehensively analyzed the temporal effects of the pharmacological agents MN-001 and MN-002 on the expression of ABCA1 and ABCG1 in THP-1-derived macrophages, using quantitative RT-PCR to measure mRNA levels. The results are shown in Fig.4A . Fig.4B demonstrates that the LXR agonist T090, which was used as a positive control, induced rapid and robust upregulation of both ABCA1 and ABCG1 mRNA levels within 6 hours of treatment. In contrast, MN-002 exhibited a more gradual yet sustained increase in ABCA1 and ABCG1 mRNA expression, indicating a delayed but persistent activation of these genes. Specifically, MN-002 significantly upregulated ABCA1 mRNA expression by 1.95-fold (p = 0.0026) at 5 µM and by 3.01-fold (p＜0.0001) at 10 µM after 24 hours of exposure. Similarly, ABCG1 mRNA levels were increased by 2.88-fold (p = 0.0276) at 5 µM and by 5.39-fold (p＜0.0001) at 10 µM, demonstrating a clear dose-dependent response.

Fig.4. MN-002 promotes ABCA1, ABCG1 LXR-alpha, PPAR-alpha, and PPAR-gamma gene expression.

Relative mRNA expression levels of ABCA1, ABCG1, LXR-alpha, PPAR-alpha, and PPAR-gamma following treatment with MN-001 and MN-002. THP-1 cells were differentiated with phorbol-12-myristate-13-acetate (PMA, phorbol 12-myristate 13-acetate,100 nM) for 72 hours. Subsequently, cells were treated with MN-001 (10 µM), MN-002 (2.5, 5, 10 µM), T0901317 (0.1 µM, used as a positive control), or the solvent vehicle (dimethyl sulfoxide, DMSO) for various time points (6 and 24 hours). The mRNA levels of ABCA1 (A), ABCG1 (B), LXR-alpha (C), PPAR-alpha (D), and PPAR-gamma (E) were quantified using real-time quantitative PCR. Data are presented as mean±standard deviation from three independent experiments. ^＊p＜0.05 vs. control (Dunnett’s test).

Notably, MN-001 did not induce any significant changes in the expression of these genes, highlighting the unique efficacy of MN-002 in modulating cholesterol efflux pathways.

4. MN-002 Promotes LXR-alpha, PPAR-alpha, and PPAR-gamma gene Expression

We focused on the LXR signaling pathway, a critical transcriptional regulator of these genes to elucidate the molecular mechanisms by which MN-002 upregulates ABCA1 and ABCG1 at both the gene and protein levels. We performed a detailed analysis of the expression of LXR-alpha and its downstream targets, considering their pivotal roles in lipid metabolism and reverse cholesterol transport. Our data revealed that MN-002 significantly enhanced the expression of LXRα (5 µM: 1.77-fold, p = 0.0142; 10 µM: 2.44-fold, p＜0.0001; Fig.4C ) after 24 hours of treatment. These expression patterns mirrored those of ABCA1 and ABCG1, suggesting a coordinated regulatory mechanism. Furthermore, MN-002 significantly enhanced the expression of PPARs, specifically PPARα (5 µM: 1.20-fold, p = 0.0264; 10 µM: 1.22-fold, p = 0.0112; Fig.4D ) and PPARγ (5 µM: 1.49-fold, p = 0.004; 10 µM: 1.67-fold, p＜0.0001; Fig.4E ) after 24 hours of treatment. Notably, PPARγ expression was also upregulated at an earlier time point—after 6 hours of treatment with MN-002 (10 µM: 1.41-fold, p = 0.0413; Fig.4E )—indicating a potential sequential activation mechanism that may contribute to regulating cholesterol efflux genes. We conducted molecular docking simulations using the human PPARα ligand-binding domain (PDB ID: 6LXA), comprising 468 amino acids ( Supplementary Fig.1A ) , to verify the interaction of MN-002 with PPARα. The docking was performed at the primary small-molecule binding site of PPARα. Among the generated conformations, the one with the highest docking score was selected for further analysis. MN-002 showed a docking score of –10.92 kcal/mol, suggesting a strong binding affinity to PPARα ( Supplementary Fig.1B ) . Since docking scores are expressed in negative values (kcal/mol), a larger absolute value indicates a stronger binding affinity between the ligand and the receptor. We conducted comparative docking analyses with known PPARα ligands to further evaluate the specificity and strength of this interaction. Eicosapentaenoic acid (EPA), which is a recognized PPARα ligand ²⁹⁾ , showed a similar binding region to that of MN-002 ( Supplementary Fig.1C, D ) . Pemafibrate, a selective PPARα modulator, showed a docking score of –10.65 kcal/mol ( Supplementary Fig.1E, F ) . In contrast, fenofibrate and bezafibrate demonstrated weaker binding, with docking scores of –8.46 kcal/mol and –8.90 kcal/mol, respectively ( Supplementary Fig.1G–J ) . These results indicated that MN-002 has a PPARα binding affinity comparable to EPA and pemafibrate and higher than that of fenofibrate and bezafibrate. These findings support the hypothesis that MN-002 may affect ABCA1 and ABCG1 expression through PPARα activation, possibly involving downstream signaling via the LXRα pathway. Detailed information on the specific amino acid residues interacting with each compound in the ligand-binding domain of PPARα is provided in Supplementary Fig.1B, C, E, G, and I , corresponding to MN-002, EPA, pemafibrate, fenofibrate, and bezafibrate, respectively. In addition to PPARα, we conducted molecular docking analyses for PPARγ and LXRα to assess the binding specificity of MN-002. The docking score for PPARγ was –8.57 kcal/mol, indicating moderate affinity. Compared to these, MN-002 showed the strongest binding affinity for PPARα (–10.92 kcal/mol), suggesting that PPARα is the primary molecular target among the tested nuclear receptors. To assess whether MN-002 interacts with LXRα, we conducted a molecular docking analysis. The docking score of MN-002 for LXRα was –0.56 kcal/mol, suggesting very weak or negligible binding affinity. In contrast, the known LXRα agonist T0901317 showed a strong binding affinity with a docking score of –8.34 kcal/mol ( Supplementary Fig.1K, L ) . These results further support the specificity of MN-002 for PPARα rather than LXRα.

5. MN-002 Enhances ABCA1-Mediated Cholesterol Efflux Independently of PKA Signaling

We examined its effects under PKA-inhibiting conditions using the specifc PKA inhibitor H89. The results demonstrated that MN-002 significantly increased cholesterol efflux, whereas H89 alone had no effect (T0901317 (positive control) vs. DMSO: p＜0.0001; MN-002 vs. DMSO: p = 0.0146; H89 vs. DMSO: p = 0.4087; Fig.5A ). Furthermore, the cholesterol efflux induced by MN-002 was not significantly affected by co-treatment with H89 (MN-002 vs. MN-002 + H89, p = 0.3711; Fig.5A ). These findings suggested that MN-002 promotes cholesterol efflux independently of the PKA signaling pathway. We also assessed whether the enhanced cholesterol efflux was accompanied by increased ABCA1 protein expression. Western blot analysis was performed after 24 hours of treatment with MN-002 and/or H89, and protein levels were normalized to β-actin ( Fig.5B ) . The results showed that MN-002 significantly increased ABCA1 protein expression (T0901317 (positive control) vs. DMSO: p = 0.0002; MN-002 vs. DMSO: p = 0.0233; Fig.5C ), whereas H89 alone had no effect (H89 vs. DMSO, p = 0.8617; Fig.5C ). Importantly, co-treatment with H89 did not alter MN-002-induced ABCA1 expression (MN-002 vs. MN-002 + H89, p = 0.5312; Fig.5C ). These results indicated that the enhancement of ABCA1-mediated cholesterol efflux by MN-002 occurs independently of PKA signaling, suggesting a potential mechanism by which MN-002 modulates cholesterol homeostasis in macrophages.

Fig.5. MN-002 enhances ABCA1-mediated cholesterol efflux independently of PKA signaling.

(A) THP-1 macrophages were labeled with [³H]-cholesterol and treated with MN-002 (10 µM), H89 (10 µM), MN-002 (10 µM) + H89 (10 µM), T0901317 (10 µM, positive control), or solvent vehicle (DMSO). The cells were then incubated with 10 µg/mL of human ApoA-I for 24 hours. Cholesterol efflux was assessed by calculating the percentage of radioactivity released into the media relative to the total radioactivity (sum of media and cell-associated). Data represent the mean±standard deviation (SD) of three independent experiments. ^＊p＜0.05 vs. control (Student’s test).

(B) ABCA1 protein levels following treatment with MN-002, the PKA inhibitor H89, or their combination. Differentiated THP-1 cells were treated with MN-002 (10 µM), H89 (10 µM), or both for 24 hours. ABCA1 and β-actin protein levels were assessed using western blot analysis. The solvent vehicle (DMSO) and T0901317 (0.1 µM) were used as negative and positive controls, respectively.

(C) Expression data are presented as the mean±SD from three independent experiments. DMSO, dimethyl sulfoxide. p＜0.05 vs. control (Dunnett’s test). n.s.: not significant (p＞0.05), compared to the control unless otherwise specified.

6. Lipid Profile Changes after MN-001 Administration

With the clinical trial targeting NASH and NAFLD patients with hypertriglyceridemia (MN-001-NATG-201 NCT 02681055), treatment with MN-001 demonstrated significant improvements in serum lipid profiles after 8 weeks of MN-001 administration. As described in the Methods section, lipid parameters were assessed at week 8 to reflect early pharmacodynamic effects of MN-001. The 8-week time point was selected based on prior data indicating early improvements in serum TG. Serum TG levels decreased from an average of 345.7 mg/dL to 206.9 mg/dL, representing a 40.2% reduction (n = 19; Supplementary Table 2 ). Notably, serum TG levels decreased from 444.7 mg/dL to 218.7 mg/dL in the subgroup of patients with a medical history of type 2 diabetes (n = 10), showing a 50.8% reduction (p = 0.098). In contrast, the non-diabetic subgroup (n = 9) exhibited a reduction from 235.7 mg/dL to 193.8 mg/dL, a 17.8% decrease ³⁰⁾ . Although the reduction in TG levels did not reach statistical significance (p = 0.098), the observed trend indicates a clinically meaningful improvement, particularly in the diabetic subgroup. In addition to the reduction in serum TG levels, the lipid profile analysis also demonstrated significant improvements in serum HDL-C levels ( Supplementary Table 3 ) . Serum HDL-C levels increased from an average of 38.7 mg/dL to 41.9 mg/dL after 8 weeks of MN-001 administration, representing an 8.3% increase (n = 19). HDL-C levels increased from 36.0 mg/dL to 41.7 mg/dL in the subgroup of patients with type 2 diabetes (n = 10), indicating a 15.8% increase (p＜0.0002). These p-values represent within-group comparisons of mean change from baseline at week 8 in each subgroup (with or without T2DM), as clarified in Supplementary Tables 2 and 3 . In contrast, the non-diabetic subgroup (n = 9) showed a slight increase from 41.8 mg/dL to 42.2 mg/dL, accounting for a 0.96% increase. The increase in HDL-C levels was particularly pronounced in the type 2 diabetic subgroup, suggesting that MN-001 may have a more substantial impact on HDL-C modulation in patients with type 2 diabetes. LDL-C levels also showed a downward trend after the 8-week intervention ( Supplementary Table 4 ) . The average LDL-C decreased from 118.1 mg/dL to 104.4 mg/dL, corresponding to an 11.6% reduction (n = 19). In the type 2 diabetes subgroup (n = 10), LDL-C levels decreased from 126.9 mg/dL to 107.4 mg/dL, showing a 15.4% reduction, while the non-diabetic subgroup (n = 9) exhibited a decrease from 108.3 mg/dL to 101 mg/dL, corresponding to a 6.7% reduction. Although the reduction in LDL-C levels did not reach statistical significance, the trend suggested a potential benefit in lipid profile improvement.

Supplementary Table 2. Mean TG before and 8 weeks after MN-001 administration.

Mean TG (mg/dL)	Baseline	Week 8	Change	p-value
All Subjects (N=19)	345.7	206.9	-40.2 %	–
With T2DM (n = 10)	444.7	218.7	-50.8%	p = 0.098
Without T2 DM (n = 9)	235.7	193.8	-17.8%	–

This table presents individual serum triglyceride (TG) levels measured before and 8 weeks after MN-001 administration. Patients are stratified by type 2 diabetes mellitus (T2DM) status (with and without T2DM). The values are expressed in milligrams per deciliter (mg/dL). Each row corresponds to an individual patient. Mean, percentage change, and p-values are included to assess changes in TG levels over the treatment period. Note: All lipid profile data were measured at week 8. p-values indicate within-group comparisons between baseline at week 8 in each subgroup (with or without T2DM), and the tables summarize the mean changes in TG levels before and after treatment in both groups.

Supplementary Table 3. Mean serum HDL before and 8 weeks after MN-001 administration.

Mean serum HDL (mg/dL)	Baseline	Week 8	Change	p-value
All Subjects (N = 19)	38.7	41.9	+8.26 %	–
With T2 DM (n = 10)	36	41.7	+15.8 %	p＜0.0002
Without T2 DM (n = 9)	41.8	42.2	+0.96 %	–

This table presents individual high-density lipoprotein cholesterol (HDL-C) levels measured before and 8 weeks after MN-001 administration. Patients are divided into subgroups based on type 2 diabetes mellitus (T2DM) status. Values are expressed in milligrams per deciliter (mg/dL). Each row represents one patient. Mean changes and statistical comparisons (including p-values) are provided for each subgroup.

Note: All lipid profile data were measured at week 8. p-values indicate within-group comparisons between baseline at week 8 in each subgroup (with or without T2DM), and the tables summarize the mean changes in HDL-C levels before and after treatment in both groups.

Supplementary Table 4. Mean serum LDL before and 8 weeks after MN-001 administration.

Mean serum LDL (mg/dL)	Baseline	Week 8	Change
All Subjects (N = 19)	118.1	104.4	-11.6 %
With T2 DM (n = 10)	126.9	107.4	-15.4 %
Without T2 DM (n = 9)	108.3	101	-6.7 %

This table shows individual low-density lipoprotein cholesterol (LDL-C) levels measured before and 8 weeks after MN-001 administration. Patients are stratified by type 2 diabetes mellitus (T2DM) status. The data are shown in milligrams per deciliter (mg/dL), and each row represents a patient. Average changes and statistical comparisons (including p-values) are included to evaluate the effect of treatment.

7. Limited Impact of MN-001 on Cholesterol Efflux Capacity in Diabetic Patients: a 12-Week Observational Study

The effect of MN-001 on cholesterol efflux capacity in patients with diabetes was assessed over a 12-week treatment period. Pre- and post-administration comparisons revealed that, although a subset of patients exhibited slight increases, the majority showed no improvement in cholesterol efflux capacity. No statistically significant changes were observed across the cohorts (p = 0.6507; Supplementary Fig.3 ; Supplementary Table 5 ).

Supplementary Fig.3. Cholesterol efflux capacity before and after MN-001 administration.

Expression data are presented as the mean±standard deviation from three independent experiments. ^＊p＜0.05 vs. control (Dunnett’s test).

Supplementary Table 5. Cholesterol efflux capacity before and 12 weeks after MN-001 administration.

	Before administration	12 weeks after administration
00031	1.246290452	1.141589268
00034	0.971795949	0.993630957
00058	0.964276534	0.83289327
05013	1.105736233	1.115891039
05015	1.12187032	1.043501986
05017	0.901246305	0.731251765
05019	1.125398125	0.958769167
05021	1.032861991	1.030276076
05024	0.850727341	0.750611838
05025	0.854982541	1.013391168
05029	0.879029361	1.026877801
05030	0.961126419	1.038563107
05031	1.034662718	0.984327647
05032	0.860952197	0.945123728
05037	0.955887344	0.949249591
05033	1.099483978	0.987507446
05036	0.95695515	1.063417951
05035	1.083323006	1.106777495
05034	1.145674442	1.196481603

This table shows the mean and standard deviation of lipid profile changes, comparing patients with and without type 2 diabetes mellitus. Data include total cholesterol, LDL-C, HDL-C, and triglycerides measured before and 12 weeks after MN-001 administration. p-values indicate the significance of changes between baseline and 12 weeks within each group.

Discussion

This study demonstrated that MN-002 significantly enhanced ApoA-I-mediated cholesterol efflux in THP-1-derived macrophages by activating nuclear receptors, including PPAR-alpha, PPAR-gamma, and LXR-alpha. Notably, MN-002 upregulated the expression of ABCA1 and ABCG1, which play critical roles in reducing foam cell formation and potentially mitigating atherosclerosis progression of atherosclerosis ^{12 , 31 , 32)} . In contrast, MN-002 and MN-001 had no significant effect on HDL-cholesterol efflux ( Fig.2B ) .

This discrepancy highlights the selective action of MN-002 on ApoA-I-mediated pathways. While Scavenger Receptor Class B Type I (SR-BI) is a key receptor involved in HDL-mediated cholesterol efflux, our experiments were inconclusive in detecting SR-BI protein expression because of uncertainties in identifying the correct band (data not shown). This suggested that SR-BI does not significantly contribute to the lack of HDL-mediated efflux enhancement observed with MN-002.

Notably, MN-002 increased ABCG1 expression without a corresponding increase in HDL-cholesterol efflux, a phenomenon previously reported in the literature. Studies have shown that elevated intracellular sphingomyelin (SM) levels can impair ABCG1-mediated efflux by altering lipid raft domains and reducing cholesterol availability for HDL-mediated transport ³³⁾ . Additionally, ABCG1 redistributes cholesterol to specific cellular domains accessible to HDL; however, its functionality can be modulated by factors such as post-translational modifications or changes in lipid dynamics, potentially limiting its capacity to enhance efflux ^{34 , 35)} . These findings suggested that ABCG1 expression alone may not be sufficient to facilitate HDL-mediated efflux under certain cellular conditions.

PPAR-alpha is a key regulator of lipid metabolism, driving the expression of ABCA1 and ABCG1 to enhance cholesterol efflux and suppress foam cell formation. Similarly, LXR-alpha is activated in response to intracellular cholesterol accumulation and synergistically induces ABCA1 and ABCG1 expression, thereby promoting reverse cholesterol transport. Together, these nuclear receptors facilitate a multifaceted approach to improving lipid metabolism and reducing inflammation ³⁶⁾ . Understanding why MN-002 selectively enhances ApoA-I-mediated efflux while leaving HDL-mediated pathways unaffected will provide valuable insights into optimizing its therapeutic potential in atherosclerosis. Molecular docking simulations were conducted to examine its interaction with PPAR-alpha and further elucidate the role of MN-002 in lipid metabolism. The results of the molecular docking simulations strongly supported the role of PPAR-alpha in the mechanism of action of MN-002. Using the crystal structure of PPAR-alpha_HUMAN (PDB ID: 6LXA), MN-002 demonstrated a high binding affinity with a docking score of -10.92 kcal/mol ( Supplementary Fig.1 ) .

Our findings suggested that MN-002 exerts a delayed yet sustained transcriptional activation of ABCA1 and ABCG1, likely mediated through PPARα-induced upregulation of LXRα. These findings are summarized in Fig.6 , which schematically illustrates the proposed mechanism by which MN-002 activates the PPARα–LXRα axis, leading to increased expression of ABCA1 and ABCG1 at the whole-cell level. This figure illustrates our Western blot results, emphasizing that the upregulation involves total cellular expression rather than membrane-specific localization. Unlike T0901317, which directly activates LXRα and rapidly induces target gene expression, MN-002 may regulate LXRα expression indirectly via PPARα activation, resulting in a slower response. This pattern implied that MN-002’s effects are mediated through a multi-step signaling cascade rather than direct LXRα stimulation. We conducted molecular docking simulations to further investigate this possibility. The results indicated that MN-002 binds to the ligand-binding domain of PPARα with high affinity, showing a docking score of –10.92 kcal/mol. Comparative analysis revealed that MN-002 exhibits a similar binding conformation to known PPARα ligands such as EPA (–10.92 kcal/mol), pemafibrate (–10.65 kcal/mol), and greater binding affinity than fenofibrate (–8.46 kcal/mol) and bezafibrate (–8.90 kcal/mol). These results indicated that MN-002 may act as a PPARα agonist with potency comparable to or exceeding that of some clinically used fibrates. Given that pemafibrate has been classified as a selective PPARα modulator (SPPARMα) and shown to robustly upregulate ABCA1/ABCG1 and promote cholesterol efflux, the distinct binding profile and functional characteristics of MN-002 suggested it may belong to a different subclass of PPARα modulators. Furthermore, the delayed transcriptional effects observed with MN-002 resemble those reported for EPA, which exerts lipid-modulating effects via both PPARα-dependent and independent pathways. Thus, MN-002 may represent a mechanistically unique agent with a sustained regulatory impact on lipid metabolism, potentially offering therapeutic benefits in chronic lipid disorders. These observations highlighted the potential of MN-002 as a novel therapeutic agent that engages PPARα signaling with distinct temporal dynamics and molecular interactions compared to traditional fibrates and LXR agonists. Further investigation to fully elucidate its mechanism of action and therapeutic implications is warranted. Our findings indicated that MN-002 activates the PPARα–LXRα–ABCA1 axis primarily through direct binding to PPARα. Although MN-002 showed moderate affinity for PPARγ, the observed upregulation of LXRα and ABCA1 is likely a downstream event following PPARα activation. This interpretation aligns with previously reported crosstalk between PPARα and LXRα in regulating lipid metabolism and cholesterol efflux.

Fig.6. Hypothetical model of MN-002 enhancing ABCA1-mediated cholesterol efflux independently of PKA signaling.

Our study demonstrated that MN-002 significantly enhances ABCA1-mediated cholesterol efflux in THP-1-derived macrophages through a PKA-independent mechanism. This effect was corroborated by the upregulation of ABCA1 protein expression, which was not influenced by PKA inhibition. These findings suggested that MN-002 may have potential as a therapeutic agent with anti-inflammatory effects and the ability to regulate lipid metabolism, which could contribute to atherosclerosis prevention.

Moreover, the PKA-independent mechanism of MN-002 represents a novel therapeutic approach, particularly for patients who are unresponsive to PKA-dependent lipid-modulating therapies. While many established lipid-regulating agents exert their effects through PKA-dependent signaling, various PKA-mediated pathways have been reported to promote ABCA1 expression and cholesterol efflux. These include the cAMP–PKA–CREB1 transcriptional pathway ³⁷⁾ , adenylyl cyclase 1-mediated cAMP production ³⁸⁾ , A2A adenosine receptor-induced cAMP–PKA signaling ³⁹⁾ , and intermedin-mediated activation of PKA ¹⁸⁾ . In addition, PKA-dependent phosphorylation stabilizes ABCA1 protein ⁴⁰⁾ , and apolipoprotein A-I binding protein (AIBP), a PKA-phosphorylated cofactor, has been shown to enhance ABCA1-mediated cholesterol efflux ⁴¹⁾ . In contrast, MN-002 appears to activate the ABCA1-mediated efflux pathway via PPARα–LXRα signaling without requiring PKA activation. Our data demonstrated that MN-002 significantly upregulates ABCA1 and ABCG1 mRNA expression and binds with high affinity to PPARα, suggesting that PPARα is the primary upstream regulator. Nevertheless, the possibility of indirect crosstalk between PPARα and PKA signaling remains. PKA modulates ABCA1 activity through phosphorylation ^{32 , 42)} , raising the question of whether MN-002-induced PPARα activation may synergize with or bypass PKA-dependent regulation. Furthermore, other kinase pathways such as AMP-activated protein kinase (AMPK) regulate lipid metabolism through mechanisms independent of PKA ^{43 , 44)} , and may potentially intersect with MN-002’s mode of action. While MN-002 likely exerts its effects predominantly through a PPARα–LXRα–ABCA1 axis independently of PKA, further studies using pharmacological PKA inhibitors or siRNA knockdown models are warranted to conclusively determine whether MN-002 operates entirely outside the PKA regulatory framework.

MN-001/002 was developed to prevent fibrosis and inflammation, as many patients exhibit atherogenic profiles and chronic inflammatory states. In addition to its antifibrotic and anti-inflammatory effects, MN-001/002 may contribute to the prevention of atherosclerosis by promoting cholesterol efflux ¹¹⁾ . Further studies are required to confirm the efficacy of MN-001/002 for these therapeutic applications. Preclinical models and clinical trials are crucial to validate these findings and facilitate the translational application of MN-002 ⁴⁵⁾ .

The effect of MN-001 on the cholesterol efflux capacity in patients with diabetes was evaluated in a 12-week observational study. Although some patients exhibited slight increases in cholesterol efflux capacity, the majority showed no improvement, and no statistically significant changes were observed across the cohort (p = 0.6507; Supplementary Fig.3 ; Supplementary Table 1 ). These findings suggested the limited efficacy of MN-001 in enhancing the cholesterol efflux capacity. This variability in patient responses may be attributed to baseline lipid profiles, inflammatory status, or individual metabolic differences. Although some patients demonstrated measurable improvements in HDL function, these effects were inconsistent and insufficient to achieve cohort-level statistical significance.

This study had several limitations. First, the observational design introduces variability due to uncontrolled factors such as patient adherence and lifestyle differences. Second, the relatively small sample size reduced statistical power, limiting the ability to detect subtle or subgroup-specific effects. Third, while cholesterol efflux capacity was assessed, other relevant biomarkers, such as HDL functionality and inflammatory markers were not evaluated, which could have provided deeper insights into MN-001’s mechanisms of action ^{13 , 46)} . Finally, the pharmacokinetics and dose-response relationships of MN-001 in patients with diabetes were not investigated, potentially affecting the observed efficacy. Additionally, glycemic parameters such as HbA1c and fasting blood glucose were not collected in this trial, limiting our ability to evaluate the relationship between MN-001 treatment and glycemic control. Future studies should incorporate these biomarkers to better assess the therapeutic potential of MN-001 in diabetic populations.

Future controlled clinical trials should involve larger and more diverse populations to address these limitations. Expanding the range of biomarkers, including HDL functionality and systemic inflammation, will provide a more comprehensive understanding of MN-001’s effects ²¹⁾ . Additionally, pharmacokinetic and dose-response studies will help optimize its therapeutic potential. Identifying specific patient subgroups that are more likely to benefit from MN-001 is critical for developing personalized treatment strategies ^{13 , 26 , 46)} .

Despite these insights, further validation using PPARα knockdown models is warranted to definitively confirm the role of PPARα in MN-002-mediated cholesterol efflux. Due to the discontinuation of our collaborative research laboratory following the retirement of the principal investigator, we are currently unable to conduct additional experiments such as PPARα knockdown or siRNA-based studies. Nonetheless, the consistent upregulation of ABCA1 and ABCG1 expression, in conjunction with the strong binding affinity of MN-002 to PPARα, provides robust indirect evidence supporting our mechanistic hypothesis.

Conclusion

This study demonstrated that MN-001/MN-002, a therapeutic agent for diseases associated with fibrosis and inflammation, may contribute to the prevention of atherosclerosis by enhancing ApoA-I-dependent cholesterol efflux through the upregulation of ABCA1 expression. Molecular docking simulations suggested that MN-002 acts as a PPAR-alpha agonist, indicating its potential impact on hepatic lipid metabolism and macrophages. Future studies are needed to investigate the role of MN-001/MN-002 in lipid metabolism, explore their effects on fibrosis and inflammation, and evaluate their long-term safety and efficacy. Additionally, the potential of combination therapy with lipid-modulating agents, such as statins, warrants further investigation.

In contrast, MN-001 showed limited efficacy in improving the cholesterol efflux capacity in patients with diabetes, with notable variability in individual responses. Further research is required to confirm these findings and clarify the potential of MN-001/MN-002 as therapeutic agents.

Acknowledgements

We express our sincere gratitude to the following individuals for their invaluable contributions as Principal Investigators in this study.

Paul Pockros, MD (SC Liver Research Consortium): For his significant role in patient recruitment, trial coordination, and data analysis.

Julio Gutierrez, MD, MS (Scripps Green Hospital): For his significant role in patient recruitment, trial coordination, and data analysis.

Charles S. Landis, MD, PhD (University of Washington): For his significant role in patient recruitment, trial coordination, and data analysis.

Their expertise and dedication were instrumental in the successful execution of this study.

The authors are grateful to the TargetMol group for their technical aid in the CADD job described in this work.

Conflict of Interest

Huicheng Qi and Kazuko Matsuda are employees of MediciNova, Inc. Masatsune Ogura has received honoraria from Amgen, Kowa, Ultragenyx Japan, and research funding from MediciNova, Inc. Takashi Miida has received research funding from Roche Diagnostics Japan and MediciNova, Inc.