Atherosclerosis is a pathological condition characterized by the development of lesions in the arterial walls. These lesions result from inflammation and lipid accumulation on the vessel wall, leading to the formation of atheromatous plaques [1]. As accumulation continues, these atherosclerotic plaques narrow the arterial lumen. Although often asymptomatic, this condition can lead to significant complications, including coronary heart disease, stroke, cerebral ischemia, organ dysfunction, and mortality.

Atherosclerosis is the leading cause of mortality and disability in the developed countries [2]. Despite its importance, the precise etiology of atherosclerosis remains elusive, with multiple factors contributing. Elevated plasma low-density lipoprotein (LDL) cholesterol (CHO) is a major risk factor. LDL accumulates in the vascular wall and triggers an inflammatory response [3,4]. This response involves interactions between the extracellular matrix, migrating inflammatory cells, and platelets, leading to plaque formation [5]. These plaques impede blood flow and reduce arterial flexibility [6]. Activated platelets and P-selectin from mononuclear cells, endothelial cells, and smooth muscle cells contribute to plaque formation and exacerbate inflammation and thrombosis [7,8].

All-trans retinoic acid (ATRA), a derivative of vitamin A, functions as a signaling molecule involved in various physiological processes, including embryonic development, immune regulation, and lipid metabolism [9]. ATRA also plays a role in regulating inflammatory processes. Numerous studies have demonstrated its anti-inflammatory effects in various inflammatory diseases [10]. In atherosclerotic models, ATRA has shown potential in regulating vascular injury and repair by affecting CHO homeostasis and platelet activation [11-13]. Nevertheless, high doses of ATRA may result in significant adverse effects, necessitating precise dosing and targeted delivery [14].

9-cis retinoic-acid (9cRA), another vitamin A metabolite, activates the retinoic acid receptor and the retinoid X receptor (RXR) [15,16]. RXR, which has lower binding affinity for ATRA, regulates plasma lipid and apolipoprotein concentrations, cell migration, and inflammation—all of which contribute to atherosclerosis [17]. RXR agonists have demonstrated potential in ameliorating atherosclerosis in both in vitro and in vivo models [18,19].

It is therefore proposed that the combination of ATRA and 9cRA may improve the regulation of vascular injury and repair in atherosclerosis by exploiting the effects of RXR. It is anticipated that the synergistic effect will also result in a reduction in the toxicity of ATRA. This study aims to evaluate whether potassium retinoate (PA9RA), a synthetic combination of 9cRA and ATRA, provides a superior effect on atherosclerosis in ApoE^-/- mice compared to traditional therapies such as clopidogrel (Plavix) or atorvastatin (Lipitor).

Animal model

Male B6.129P2-Apoetm1Unc/J (apolipoprotein knock-out, ApoE^-/-) mice at approximately 8 weeks of age were supplied by Orient Bio Co. After 1–2 weeks of acclimation, animals were randomly assigned to 6 groups (each with n = 8). The groups were 1) ApoE^-/- + vehicle (double distilled water [DDW]), 2) ApoE^-/- + clopidogrel (Plavix) 25 mg/kg/day, 3) ApoE^-/- + atorvastatin (Lipitor) 10 mg/kg/day, 4) ApoE^-/- + PA9RA 5 mg/kg/day, 5) ApoE^-/- + PA9RA 15 mg/kg/day, 6) ApoE^-/- + PA9RA 45 mg/kg/day. All animals were fed a Western-type diet for 10 weeks starting at 12 weeks of age. The animals were housed in separate standard plastic cages in a specific pathogen-free facility. They had controlled light and temperature conditions with free access to food and water ad libitum. The Institutional Animal Care and Use Committee (IACUC) approved all animal care procedures under protocol No. 2021-02-290 (Asan Institute for Life Science, Seoul, Korea).

Test article preparation and administration

The test article (PA9RA) were obtained from ArcaEir Inc. PA9RA was dissolved in DDW before treatment. Animals were treated with PA9RA daily by oral gavage for at least 10 weeks.

In-life procedures, observations, and measurements

All animals were observed daily for mortality and clinical signs. Individual body weights were measured daily before treatment using an electronic scale (CIAA233; CAS).

Necropsy

After 10 weeks, the animals were weighed and sacrificed under isoflurane anesthesia, followed by exsanguination. All animals were subjected to a complete necropsy. The carcass and musculoskeletal system, all external surfaces and orifices, the cranial cavity and external surfaces of the brain, and the thoracic, abdominal, and pelvic cavities with their associated organs and tissues were grossly examined.

After gross examination, the following organs were weighed before 10% neutral buffered formalin fixation: the heart, inguinal fat tissue, kidney, liver, spleen, and thymus. Paired organs were weighed together.

Serum biochemistry

Blood samples of 0.8 ml were collected from the caudal vena cava at the time of sacrifice. The serum concentrations of alanine aminotransferase (ALT), alkaline phosphatase (ALP), aspartate aminotransferase (AST), CHO, LDL, high-density lipoprotein (HDL), total protein, glucose (GLU), apolipoprotein A1, apolipoprotein A2, calcium and triglyceride were determined using the blood biochemistry analyzer (HITACHI 7180; Hitachi Co., Ltd.).

Histopathological examinations and morphometric analysis

Representative samples of the following tissues were collected and preserved in 10% neutral buffered formalin: cecum, colon, duodenum, heart, ileum, inguinal fat, jejunum, kidney, liver, mesenteric lymph nodes, rectum, skin, spleen, stomach, and thymus. Tissues were embedded in paraffin, sectioned, mounted on glass slides, and stained with hematoxylin and eosin. The pathologist evaluated all tissues histopathologically.

For liver injury, liver centrilobular microvesicular steatosis were semi-quantitatively scored from 0–5, depending on the severity (0: no evidence of lipid vacuoles; 5: very prominent lipid vacuoles). Lobular inflammation was semiquantitatively scored from 0–3 depending on the severity (0: no foci; 1: < 2 foci / 200x; 2: 2–4 foci / 200x; 3: > 4 foci / 200x) [20].

The heart was preserved in 10% neutral buffered formalin for 24 h and then substituted with a 30% sucrose solution. The heart then was embedded in the O.C.T compound (Tissue-Tek), making a frozen block, sectioned at 10 µm. Add 0.625 g of Oil Red O (ORO, O0625; Sigma) in 100 ml of isopropanol. Stir the saturated solution in dark bottle for 1 h at room temperature. Prepare working solution by mixing 75 ml of saturated ORO solution with 50 ml of deionized water. Add 0.5 ml of ORO working solution to cover the sections. Incubate the sections for 5 min at room temperature. Counter stain the section with Mayer’s hematoxylin. Rinse the sections under running tap water for 30 min and mount the slides. All slide were photographed. The area of atherosclerotic lesions on the aortic root was measured and analyzed using Image J software (Fiji).

Statistical analysis

All results were expressed as mean ± standard error of the mean (SEM). Statistical significance was analyzed using IBM SPSS Statistics (v27) software. Levene’s test was used to assess homogeneity of group variances, and a one-way analysis of variance (ANOVA) was employed to assess significant differences. Scheffé's multiple comparison test was performed as a post-hoc test when both homogeneity of variance and significance of the difference were demonstrated. In the absence of homogeneity of variance, the Dunnett's T3 test was performed. The significance threshold was p ≤ 0.05.

Mortality and clinical signs

One animal was euthanized on study day 34. Notably, animals receiving PA9RA at a dose of 45 mg/kg/day exhibited keratinization and hair loss. However, no other observations were found to be related to the administration of PA9RA.

Body weight

Fig. 1 illustrates the variation in body weight over the study period. Notably, there was a statistically significant reduction in body weight in the PA9RA-treated group (p ≤ 0.05), with a more than 10% reduction in the PA9RA 45 mg/kg/day group.

Figure 1. Mean body weight is decreased in PA9RA treated mouse. ApoE^–/– mice were fed a Western-type diet for 10 weeks and orally administered with DDW, clopidogrel, atorvastatin, PA9RA (5 mg/kg/day, 15 mg/kg/day, and 45 mg/kg/day). Body weight changes during study period is presented as mean ± SEM (n = 8, n = 7 for PA9RA 45 mg/kg/day from study day 34). PA9RA, potassium retinoate; DDW, double distilled water.

Organ weight

Table 1 presents that mice treated with PA9RA at doses of 15 mg/kg/day and 45 mg/kg/day showed significant reductions in inguinal fat weights. In contrast, the group treated with PA9RA at 45 mg/kg/day showed an increase in spleen weight.

Table 1 . Comparison of mean organ weights between negative control, clopidogrel, atorvastatin, PA9RA 5 mg/kg/day, 15 mg/kg/day, and 45 mg/kg/day treated groups.

Variable	Negative control	Clopidogrel, 25 mg/kg/day	Atorvastatin, 10 mg/kg/day	PA9RA, 5 mg/kg/day	PA9RA, 15 mg/kg/day	PA9RA, 45 mg/kg/day
Number of animals per group	8	8	8	8	8	7
Terminal body weight
Absolute value (g)	36.76	–0.91	–1.25	–4.78	–6.05*	–13.08**
Heart
Absolute value (g)	0.1567	–0.0007	0.0016	–0.0120	–0.0176	–0.0248
% of body weight	0.4262	0.0113	0.0253	0.0291	0.0274	0.1366*
Inguinal fat
Absolute value (g)	1.6164	0.0324	–0.0055	–0.2136	–0.5933*	–1.1768**
% of body weight	4.3870	0.1003	0.0252	–0.1283	–1.0754	–2.5958**
Kidney
Absolute value (g)	0.3476	0.0037	0.0139	–0.0122	0.0081	–0.0186
% of body weight	0.9445	0.0413	0.0896	0.1105	0.2153**	0.4591**
Liver
Absolute value (g)	1.7496	–0.0715	–0.1663	–0.3803*	–0.3855	–0.4575
% of body weight	4.7530	–0.1113	–0.3436	–0.4973	–0.3233	0.6461*
Spleen
Absolute value (g)	0.1031	0.0055	0.0073	–0.0083	–0.0101	0.0848*
% of body weight	0.2799	0.0243	0.0345	0.0164	0.0259	0.5305**
Thymus
Absolute value (g)	0.0503	–0.0139	–0.0109	–0.0078	–0.0126	–0.0286*
% of body weight	0.1368	–0.0357	–0.0260	–0.0030	–0.0120	–0.0482*

PA9RA, potassium retinoate. *p ≤ 0.05, **p ≤ 0.01.

Serum biochemistry

In the study, reductions in LDL level were observed in the atorvastatin and PA9RA-treated groups in serum biochemistry. In contrast, HDL levels were increased in PA9RA-treated groups. Additionally, GLU levels were reduced in the PA9RA treated groups. However, ALP levels were increased in the PA9RA treated groups. Table 2 provides a summary of the results.

Table 2 . Serum biochemistry analysis results in negative control, clopidogrel, atorvastatin, PA9RA 5 mg/kg/day, 15 mg/kg/day, and 45 mg/kg/day treated groups.

Variable	Negative control	Clopidogrel, 25 mg/kg/day	Atorvastatin, 10 mg/kg/day	PA9RA, 5 mg/kg/day	PA9RA, 15 mg/kg/day	PA9RA, 45 mg/kg/day
Number of animals per group	8	8	8	8	8	7
Alkaline phosphatase (U/l)	200.99 ± 8.27	191.75 ± 8.21	221.35 ± 12.45	316.60 ± 9.51**	339.81 ± 21.55*	223.09 ± 30.50
Alanine aminotransferase (U/l)	62.06 ± 8.73	56.34 ± 10.31	81.48 ± 24.57	68.66 ± 16.05	40.09 ± 10.19	27.07 ± 3.80
Apolipoprotein A1 (mg/dl)	0.26 ± 0.02	0.23 ± 0.04	0.13 ± 0.03	0.25 ± 0.04	0.19 ± 0.08	0.31 ± 0.05
Aspartate aminotransferase (U/l)	82.53 ± 6.32	84.96 ± 8.36	115.26 ± 23.85	84.89 ± 11.14	65.60 ± 6.34	68.37 ± 5.89
Calcium (mg/dl)	10.02 ± 0.15	9.93 ± 0.10	10.19 ± 0.15	9.59 ± 0.07	9.56 ± 0.08	10.06 ± 0.11
Cholesterol (mg/dl)	1,432.94 ± 36.10	1,407.95 ± 105.36	1,250.88 ± 58.91	1,151.03 ± 125.54	1,223.89 ± 110.59	981.66 ± 163.77
Glucose (mg/dl)	185.84 ± 10.72	153.94 ± 11.78	166.90 ± 8.89	123.31 ± 10.85*	135.88 ± 13.12	117.20 ± 14.00*
High-density lipoprotein (mg/dl)	12.51 ± 1.03	11.78 ± 0.99	17.84 ± 1.93	21.96 ± 2.70	26.63 ± 2.71**	25.60 ± 2.88**
Low-density lipoprotein (mg/dl)	338.39 ± 8.73	333.18 ± 26.45	265.06 ± 12.43**	247.60 ± 24.01	249.90 ± 26.33	164.20 ± 32.55*
Triglyceride (mg/dl)	82.91 ± 8.28	77.95 ± 7.94	102.85 ± 9.78	106.63 ± 12.86	102.39 ± 9.68	110.97 ± 21.80
Total protein (g/dl)	5.32 ± 0.07	5.06 ± 0.09	5.06 ± 0.07	5.11 ± 0.08	4.97 ± 0.07	5.62 ± 0.21

Values are presented as mean ± SEM. PA9RA, potassium retinoate. *p ≤ 0.05, **p ≤ 0.01.

Histopathologic evaluation and morphometric analysis

In the heart, foam cells in were found the intima of the aorta from all groups except the PA9RA-treated group. For the skin, foamy histiocytes in the deep dermis were found in the clopidogrel-treated group. Epidermal hyperplasia was observed in the PA9RA-treated groups.

In the forestomach, epithelium hyperplasia and submucosal inflammation were observed in the PA9RA-treated groups. Similar lesions were found in the glandular stomach of PA9RA-treated animals with globular leukocytes. The limiting ridge was also hyperplastic with inflammation. The incidence and severity of microscopic findings are summarized in Table 3.

Table 3 . Microscopic findings in negative control, clopidogrel, atorvastatin, PA9RA 5 mg/kg/day, 15 mg/kg/day, and 45 mg/kg/day treated groups.

Variable	Negative control	Clopidogrel, 25 mg/kg/day	Atorvastatin, 10 mg/kg/day	PA9RA, 5 mg/kg/day	PA9RA, 15 mg/kg/day	PA9RA, 45 mg/kg/day
Number of animals per group	8	8	8	8	8	7
Heart
Examined	2	2	2	2	2	2
No visible lesions	0	0	0	0	0	2
Foam cells, aorta/intima
Minimal	0	0	1	1	2	0
Slight	1	2	1	1	0	0
Jejunum
Examined	5	5	6	6	6	6
No visible lesions	5	5	6	4	4	5
Zymogen granule (Paneth-like cell), cytoplasm, mucosa
Minimal	-	-	-	2	2	1
Skin
Examined	6	6	6	6	6	6
No visible lesions	6	4	6	5	5	0
Foamy histocyte, deep dermis
Minimal	0	1	0	0	0	0
Slight	0	1	0	0	0-	0
Hyperplasia, epidermis
Minimal	0	0	0	1	1	1
Spleen
Examined	6	6	6	6	6	6
No visible lesions	4	3	3	5	4	0
Extramedullary hematopoiesis
Minimal	1	0	0	1	1	1
Slight	1	2	2	0	0	0
Moderate	0	1	1	0	0	0
Stomach
Examined	6	6	6	6	6	6
No visible lesions	6	5	5	3	2	0
Hyperplasia, forestomach, epithelium
Minimal	0	0	0	0	0	1
Slight	0	0	0	0	1	1
Marked	0	0	0	0	0	3
Inflammation, forestomach, submucosa
Slight	0	0	0	0	0	3
Moderate	0	0	0	0	0	1
Eosinophilic crystalline inclusion, glandular stomach
Minimal	0	0	0	1	0	1
Globule leukocyte, glandular stomach, foveolar epithelium
Minimal	0	0	0	0	1	0
Slight	0	0	0	0	0	3
Moderate	0	0	0	0	0	1
Hyperplasia, glandular stomach, foveolar epithelium
Minimal	0	0	0	0	1	2
Slight	0	0	0	0	0	4
Inflammation, glandular stomach, submucosa
Minimal	0	0	0	0	1	1
Slight	0	1	0	0	1	1
Moderate	0	0	0	0	0	1
Hyperplasia, limiting ridge, epithelium
Minimal	0	0	0	2	2	2
Slight	0	0	0	0	1	1
Marked	0	0	0	0	0	3
Inflammation, limiting ridge, submucosa
Minimal	0	0	1	1	1	2
Slight	0	0	0	0	1	3
Moderate	0	0	0	0	0	1

_{PA9RA, potassium retinoate.}.

In the liver, centrilobular microvesicular steatosis was significantly ameliorated in the PA9RA 45 mg/kg/day treated groups (Fig. 2). The arteriosclerotic lesion within the aorta showed a decreased area in the PA9RA 45 mg/kg/day treated group compared to the other groups (Fig. 3).

Figure 2. PA9RA alleviates liver lesion in ApoE^–/– mouse. Centrilobular microvesicular steatosis and lobular inflammation in the liver were analyzed for liver injury using H&E. (A) Semi-quantitative analysis of liver injury. Data is presented as mean ± SEM (n = 6). **p ≤ 0.001 vs. negative control group. (B) Representative image of the liver from negative control, clopidogrel, atorvastatin, PA9RA 5 mg/kg/day, 15 mg/kg/day, and 45 mg/kg/day treated groups. Centrilobular steatosis is decreased in PA9RA 45 mg/kg/day treated group. Scale bar: 200 µm. PA9RA, potassium retinoate.

Figure 3. PA9RA alleviates atherosclerotic lesion in ApoE^–/– mouse. Atherosclerotic plaques in the heart aortic root were stained with Oil-red O (ORO) and plaque area was measured. (A) Quantitative analysis of atherosclerotic lesion area (percent of ORO positive area in the total staining area). Data is presented as mean ± SEM (n = 6). *p ≤ 0.05 vs. negative control group. (B) Representative image of sections of heart aortic root from negative control, clopidogrel, atorvastatin, PA9RA 5 mg/kg/day, 15 mg/kg/day, and 45 mg/kg/day treated groups. The atherosclerotic lesion area is decreased in PA9RA 45 mg/kg/day treated group. Scale bar: 100 µm. PA9RA, potassium retinoate.

This study demonstrates that PA9RA has better protective effects in the complex disease processes of dyslipidemia and atherogenesis in male ApoE^-/- mice. Specifically, PA9RA is more effective than clopidogrel and atorvastatin in ameliorating fatty liver formation and reducing the size of atherosclerotic lesions. These findings highlight the potential of PA9RA as a promising therapeutic agent for the treatment of dyslipidemia and atherosclerosis.

Adipose tissue is a potential target for ATRA [21]. Our study showed a significant reduction in body weight and inguinal adipose tissue weight in the PA9RA-treated groups, which is consistent with previous research showing that ATRA reduces fat accumulation in mice by promoting fat mobilization and catabolism [22]. ATRA is known to decrease adipogenesis, enhance lipolysis, and reduce triacylglycerol levels in mature adipocytes [23]. However, it is important to acknowledge that other studies have reported an ATRA-induced increase in body weight [24].

9cRA contributes to the activation of RXRs. RXRs are known to be downregulated in the macrophage of the aged population, which may partly explain the increased risk of atherosclerosis [25]. RXR also contributes to the regulation of ApoE expression [26]. 9cRA also increases CHO efflux in macrophages, thereby blocking atherosclerotic lesion formation [27].

This discrepancy underscores the complex interplay between ATRA and 9cRA in adipose tissue, emphasizing the need for further research to clarify their mechanisms of action across different experimental contexts.

Atherosclerosis is a chronic condition associated with lipid deposition, endothelial damage, and inflammation. The progression of atherosclerosis is influenced by endothelial cells, inflammatory cells, and intimal smooth muscle cells [28]. ATRA and 9cRA may potentially influence the development of atherosclerosis by regulating immune cell proliferation, migration, and transformation, as well as blood GLU and lipid metabolism [29]. Our study found a decrease in serum GLU levels in the PA9RA-treated groups, which supports this theory. In addition, ATRA has been shown to attenuate procoagulant properties, promote fibrinolysis, and reduce monocyte induction and tissue factor expression, thereby inhibiting arterial thrombus formation [30].

Clopidogrel is the first-line treatment for atherosclerosis, and has multiple effects on inhibiting intimal injury, inflammation and, endothelial dysfunction [31,32]. Clopidogrel also attenuates cellular infiltration and decreases adhesion molecule cytokine expression , and inhibits platelet aggregation [33]. Atorvastatin delays the development of atherosclerosis by reducing CHO production and increasing the number of LDL receptors [34]. Our findings suggest that the combination of ATRA and 9cRA, PA9RA may exert enhanced protective effects against atherosclerosis through multiple mechanisms.

Multiple studies have demonstrated that ATRA can inhibit atherogenesis in ApoE^-/- mice fed a high-fat diet [35]. The protective effect of ATRA against atherosclerosis may involve multiple mechanisms. ATRA has been demonstrated to promote the transformation of both M1 and M2 macrophages in various disease contexts [36]. M1 macrophages exacerbate plaque formation, causing systemic inflammation and leading to plaque rupture. In contrast, M2 macrophages are abundant in stable plaques [37]. In addition, ATRA regulates smooth muscle cells, whose expression is decreased in atherosclerosis and contributes to foam cell formation [38]. Foam cells are generated by the unregulated uptake of modified forms of LDL [39]. In our study, no foam cells were observed in the intima of the cardiac aorta in the group treated with PA9RA 45 mg/kg/day. ATRA also modulates inflammatory cells, such as neutrophils and T cells. However, further research is needed to elucidate the mechanisms by which ATRA and 9cRA ameliorate atherosclerosis.

Since atherosclerosis results from a systemic chronic inflammatory state characterized by elevated lipid levels, metabolic dysfunction-associated steatotic liver disease (MASLD) can develop in the liver [40]. MASLD is characterized by lipid steatosis, macrophage accumulation, and dysregulated immune responses [41]. Several studies have shown that ATRA can alleviate MASLD by reducing serum CHO, LDL, AST, and ALT levels, along with histologic improvements in liver steatosis [42]. In our study, we observed a significant decrease in centrilobular microvesicular steatosis scores in the PA9RA-treated groups, suggesting that PA9RA may also be a promising potential treatment for MASLD.

Toxicologically, the epidermal hyperplasia of the skin observed in the PA9RA 45 mg/kg/day group was consistent with clinical observations. Retinoic acid and retinoids have known effects on skin growth and differentiation [43]. Specifically, ATRA induces re-epithelialization by increasing keratinocyte proliferation and cell migration in wounded skin [44]. ATRA is thought to induce the expression of heparin-binding EGF-like growth factor in keratinocytes, thereby mediating epidermal hyperplasia [45]. ATRA also promotes the proliferation of basal keratinocytes and accelerates epidermal cell turnover, resulting in epidermal thickening [46]. However, these effects have primarily been demonstrated with topical administration of ATRA. The epidermal hyperplasia observed in the PA9RA-treated groups in this study may be related to the test article and the pharmacological effects of PA9RA, but further studies are needed to elucidate the underlying mechanism [47].

The hyperplastic changes observed in the stomach (including the forestomach, limiting ridge, and glandular stomach) appear to be a response to irritation, as evidenced by inflammation and the presence of eosinophilic globules [48]. In previous studies, the mucosal surfaces of organs lined by squamous epithelium were found to be hyperplastic and hyperkeratotic in global knockout mice lacking the enzyme responsible for ATRA clearance [49]. In addition, hypereosinophilic Paneth cells were observed in these knockout mice, which were associated with poor nutritional status. The presence of Paneth-like cells in the jejunum in our study may be related to this condition. Further studies using more standardized sectioning techniques, such as Swiss rolls, is necessary to clarify this hypothesis.

The changes in heart, kidney, spleen, and thymus weights observed in the PA9RA group did not correlate with any histologic findings. It is recognized that changes in thymus and spleen weights do not consistently correlate with histological findings, and histopathology is generally considered to be a more sensitive indicator of PA9RA effects [50]. In addition, decreased heart weight without microscopic correlation is often associated with a treatment-related decreases in body weight. Furthermore, the changes in kidney weight observed were minimal, less than 10%. Therefore, these findings are not considered to be adverse.

The limitation of this study is that we evaluated the protective effect of PA9RA in male mice. The focus on male subjects was motivated by the observation that males generally exhibit a greater plaque and inflammatory burden, which increases the risk of ischemia [51]. Moreover, the results of studies on females are often challenging to interpret accurately due to the inherent variability associated with the female sexual cycle. It would therefore be beneficial to investigate whether there are any differences in the efficacy and toxicity of PA9RA between the sexes.

In conclusion, our study demonstrates that the combination of 9cRA and ATRA significantly ameliorates both atherosclerosis and fatty liver. To the best of our knowledge, this study is the first to investigate the combinatorial effect of 9cRA and ATRA in atherosclerosis models. For future research, we hypothesize that PA9RA activates reverse CHO transport through CHO efflux from macrophages. This could potentially lead to plaque-stabilizing changes and intimal remodeling by reducing macrophage-derived foam cells and increasing smooth muscle cells. Further studies are warranted to investigate this hypothesis and to elucidate the underlying mechanisms of action of the combination of ATRA and 9cRA in the treatment of atherosclerosis and related diseases.

None.

The authors declare no conflicts of interest.

This research was financially supported by the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare of the Korean government under grant number: RS-2023-00265860.