Saffron as a natural modulator of reverse cholesterol transport genes in atherosclerotic rabbits, with molecular docking insights

Abstract

Atherosclerosis is a major contributor to cardiovascular disease, and one of the mechanisms that contributes to atherosclerosis is the reverse cholesterol transport (RCT) pathway, which includes SR-BI, ABCA1, and PPARγ genes. Natural compounds that modulate RCT-related genes may present promising therapeutic alternatives. Saffron (Crocus sativus L.), rich in bioactive carotenoids, exhibits both lipid-lowering and antioxidant properties. This study investigated the effects of saffron extract on hepatic expression of SR-BI, ABCA1, and PPARγ genes in the atherosclerotic rabbit model and evaluated the molecular docking of its major phytocompounds. Fifty-five male New Zealand White rabbits (NZWR) were randomly assigned to three main groups: a normal diet (ND) group, a 1% high-cholesterol diet (HCD; 4 W, 8 W) group, and intervention groups. Rabbits in the HCD and intervention groups were induced for early atherosclerosis (4 weeks) and established atherosclerosis (8 weeks). Following these induction periods, each subgroup received 8 weeks of oral treatment with saffron ethanolic extract (50 or 100 mg/kg/day), statin (2.5 mg/kg/day), or placebo while maintained on a normal chow diet. The Quantitative Real-Time Polymerase Chain Reaction (qRT-PCR) analysis showed that saffron treatment significantly upregulated hepatic SR-BI expression in early atherosclerosis (S50: 3.65-fold, p < 0.05; S100: 4.59-fold, p < 0.05) and in established atherosclerosis (S100: 8.34-fold, p < 0.01). ABCA1 and PPARγ expression levels were also increased, though not statistically significant. Molecular docking demonstrated favorable binding affinities between saffron bioactives and RCT-related targets, with crocetin (a major carotenoid compound in saffron) binding to PPARγ (–7.75 kcal/mol) and SR-BI (–7.24 kcal/mol), and quercetin binding to ABCA1 (–8.35 kcal/mol). These findings suggest that saffron may positively modulate RCT-associated gene expression, supporting its potential as a natural adjunct in atherosclerosis research and management.

Introduction

Atherosclerosis is a leading cause of cardiovascular diseases (CVDs), accounting for nearly one-third of global mortality in 2021^1,2. It is characterized by lipid accumulation, chronic inflammation, and endothelial dysfunction that progressively narrow arterial walls and lead to ischemic events such as myocardial infarction and stroke^3,4. Despite the availability of lipid-lowering therapies, effective management remains challenging due to persistent oxidative stress, inflammation, and metabolic dysregulation^5,6.

Natural products and medicinal plants have gained interest as adjuncts to conventional therapy because of their multi-targeted antioxidant and lipid-regulating mechanisms⁷. Among the key molecular pathways involved in cholesterol homeostasis is RCT, which mediates cholesterol efflux from peripheral tissues to the liver for excretion⁸. The RCT process is primarily regulated by ATP-binding cassette transporter A1 (ABCA1), scavenger receptor class B type 1 (SR-BI), and peroxisome proliferator-activated receptor gamma (PPARγ)^9–12. ABCA1 initiates cholesterol efflux to apolipoprotein A-I (ApoA-I), SR-BI facilitates hepatic uptake of HDL-cholesterol, and PPARγ indirectly modulates their expression through interaction with nuclear receptors such as LXR, linking lipid metabolism with anti-inflammatory signaling¹¹. Modulating these pathways offers an integrative approach to prevent lipid accumulation and plaque progression.

Saffron or scientifically known as Crocus sativus L., a medicinal spice rich in bioactive compounds such as crocin, crocetin, quercetin, and safranal, has demonstrated antioxidant, anti-inflammatory, and lipid-lowering effects in preclinical and clinical studies^13–16. Experimental evidence suggests saffron improves endothelial function and suppresses inflammatory markers, yet its effects on RCT-related molecular targets remain unclear. Molecular docking offers valuable predictive insights into how bioactive compounds interact with protein targets^17,18. Prior studies have shown strong binding affinities of saffron-derived compounds to angiogenic and inflammatory proteins such as VEGFR2 and COX-2^19,20, suggesting broad pharmacological potential. However, interactions with RCT-related targets (SR-BI, PPARγ, and ABCA1) are yet to be elucidated. This study investigates the effects of saffron extract on hepatic SR-BI, ABCA1, and PPARγ expression in atherosclerotic New Zealand White rabbits, integrating in vivo and in silico analyses to elucidate saffron’s potential role in cholesterol efflux and atherosclerosis mitigation.

Materials and methods

Preparation of saffron extract

Saffron Extract Preparation. Dried saffron were purchased from World Care Groups Sdn. Bhd. (Malaysia) and authenticated by the Herbarium, Universiti Kebangsaan Malaysia (UKM), under voucher number ID006/2021. Ten grams of dried saffron stigmas were ground and macerated in 500 mL of ethanol–water (80:20, v/v) with continuous stirring for 72 h at room temperature. The extract was filtered and concentrated under reduced pressure at 50 °C using a rotary evaporator (120 rpm), frozen at − 80 °C overnight, and freeze-dried to obtain the lyophilized saffron ethanolic extract (SEE). The final extract was stored at 4 °C until use. The ethanol–water (80:20) solvent system was selected to efficiently extract both polar and semi-polar bioactive compounds. The extraction procedure followed established protocols described by Ramli et al.²¹ and Rahim et al.²². Figure 1 illustrates the schematic overview of saffron ethanolic extraction.

Fig. 1.

Schematic diagram of maceration extraction process of saffron ethanolic extract.

Animals

Fifty-five male NZWR, weighing between 1.8 kg and 2.0 kg, were purchased from Chenur Sdn. Bhd., Malaysia. The rabbits were obtained from a reputable, certified laboratory animal supplier with extensive experience providing laboratory animals to public universities nationwide. The rabbits were kept in the controlled condition at an ambient temperature of 25–30 °C and relative humidity of 55–60% and 12/12 h light/dark cycle and were provided a pellet diet along with water ad libitum. All methods were performed in accordance with the relevant guidelines and regulations. All procedures of the current study were reported following the ARRIVE guidelines. All procedures were conducted in accordance with the Universiti Teknologi MARA Committee on Animal Research & Ethics (UiTM CARE) regulations and were approved by the committee (UiTM CARE: 326/2020), which ensured that the rabbits were treated throughout the experiment according to the standards for the care of experimental animals.

Study protocol

Fifty-five male New Zealand White rabbits (NZWRs), weighing 2.0–2.2 kg, were obtained from Chenur Sdn. Bhd. (Malaysia). The animals were housed individually in stainless-steel cages within an environmentally controlled clean-air facility, maintained at 22 ± 2 °C with a 12 h light/12 h dark cycle and relative humidity of 60 ± 5%. The rabbits were randomly allocated into three main groups: a normal diet (ND) group, a high-cholesterol diet (HCD) group, and an intervention group. The ND group (n = 5) received a standard chow diet and served as the control cohort. The HCD group (n = 10), also referred to as the pre-treated group, was fed a 1% cholesterol diet and further subdivided into two subgroups: 4 W (n = 5), which received HCD for 4 weeks to develop early atherosclerosis, and 8 W (n = 5), which received HCD for 8 weeks to develop established atherosclerosis. The intervention group (n = 40) consisted of rabbits induced with early or established atherosclerosis, followed by 8 weeks of specific interventions: (i) SEE at 50 mg/kg/day (S50), (ii) SEE at 100 mg/kg/day (S100), (iii) simvastatin at 2.5 mg/kg/day, and (iv) placebo (distilled water). All interventions were administered orally once daily by gavage. During the intervention period, the HCD was discontinued and replaced with a ND to minimize hepatotoxicity and mortality associated with prolonged cholesterol feeding, as documented in previous studies^21,22. This adjustment ensured animal survival and preserved physiological relevance by preventing excessive inflammation uncharacteristic of human atherosclerosis²³. The selected HCD durations for inducing early and established atherosclerotic lesions were based on previous reports demonstrating early fatty streak formation after 4 weeks (early atherosclerosis) and advanced lesions (established atherosclerosis) after 8 weeks of HCD feeding^23,24. The atherosclerotic animal model and HCD composition were adapted from established rabbit models of diet-induced atherosclerosis²⁵.

The statin-treated group served as a positive control to validate the model and provide a reference for comparison. The saffron-only group was not included since the study focused on evaluating saffron’s therapeutic effects under high-cholesterol diet–induced atherosclerosis. The SEE dosage was extrapolated from a previous rat study demonstrating efficacy at 100 and 200 mg/kg/day²⁶. The equivalent rabbit doses (50 and 100 mg/kg/day) were derived using the standard interspecies dose conversion formula²⁷. The schematic presents an overview of the study protocol illustrated in Fig. 2.

Fig. 2.

Schematic overview of the experimental study protocol. New Zealand White Rabbits (NZWR; n = 55) were assigned to a baseline group on a normal diet (ND; n = 5), a high-cholesterol diet (HCD) group fed for 4 or 8 weeks (n = 10), or an intervention group (n = 40) comprising early and established atherosclerosis models treated with saffron extract (S50: 50 mg/kg/day; S100: 100 mg/kg/day), statins, or placebo (PCB). At the end of the study, rabbits were euthanised and liver tissues were collected, snap-frozen, and processed for RNA extraction, complementary DNA (cDNA) synthesis, and quantitative real-time PCR (qRT-PCR).

Quantitative real time-PCR (qRT-PCR) analysis

At the end of the treatment period, the rabbits were euthanized via intravenous administration of 100 mg sodium pentobarbital. Following euthanasia, the rabbits were dissected, and liver tissues were collected for analysis. Approximately 30 mg of liver tissue was weighed and disrupted using a pestle and mortar in the presence of liquid nitrogen to preserve RNA integrity and prevent degradation. Total RNA was then extracted from the tissues and reverse-transcribed into complementary DNA (cDNA) to enable gene expression analysis. The qRT-PCR was performed to assess the expression of key target genes, including ABCA1, SR-BI, and PPARγ, which are critical regulators of cholesterol homeostasis and reverse cholesterol transport. Primer sequences were selected based on previously validated studies and designed according to the guidelines proposed by Bustin and Huggett²⁸ and Shipley²⁹.

The selected primer sequences and their associated annealing temperatures are presented in Table 1. Reference genes Glyceraldehyde 3-phosphate dehydrogenase (GAPDH), Hypoxanthine Phosphoribosyltransferase 1 (HPRT-1), and beta-actin (β-actin) were used for normalization to ensure reliable quantification. Each reaction was performed in technical triplicates, and mean Ct values were used for analysis. The relative changes in gene expression were calculated using the 2^-ΔΔCt (Livak) method. The Livak method is a widely accepted approach for quantifying fold changes in gene expression relative to the baseline group and fold changes with p < 0.05 were considered statistically significant.

Table 1.

Primer sequences and annealing temperatures for liver in established atherosclerosis rabbits.

Primer (ID reference)	Forward primer sequence	Reverse primer sequence	TA (°C)
GAPDH (DQ403051)	5’ ATC ACT GCC ACC CAG AAG AC 3’	5’ TGA GTT TCC CGT TCA GCT CG 3’	60
HPRT-1 (NM_001105671)	5’ CCT TGG TCA AGC AGT ATA ATC 3’	5’ GGG CAT ATC CTA CAA CAA AC 3’	50
ß-ACTIN (NM_001101683)	5’ ATC AGC AAG CAG GAG TAT GAC 3’	5’ GCC AAT CTC GTC TCG TTT CT 3’	60
PPARγ (NM_001082148)	5’ GGA GCA GAG CAA AGA AGT CG 3’	5’ CTC ACA AAG CCA GGG ATG TT 3’	56.3
ABCA1 (Li et al., 2015)	5′-GAT GGC AAT CAT GGT CAA TGG-3′	5′-AGC TGG TAT TGT AGC ATG TTC CG 3′	60
SR-BI (Li et al., 2015)	5′-CAG TGG GCA TTG TGT CCT GTC-3′	5′-GGC TCA GTG CAG GCT GAT GTC-3′	50

TA; annealing temperature, nucleotides A; adenine, G; guanine, T; thymine, C; cytosine.

ADMET prediction

The pharmacokinetic and drug-likeness properties of saffron’s key phytoconstituents were evaluated using the SwissADME webserver (ref - https://www.nature.com/articles/srep42717). Canonical SMILES retrieved from PubChem were analyzed to predict key ADME parameters, including gastrointestinal absorption, blood–brain barrier permeability, P-glycoprotein interaction, and cytochrome P450 (CYP1A2, CYP2C19, CYP2C9, CYP2D6, CYP3A4) inhibition profiles. Drug-likeness was assessed according to Lipinski’s rule of five and bioavailability scores. This analysis predicts pharmacokinetic characteristics of oral bioavailability and therapeutic potential of the saffron compounds.

Molecular docking

The X-ray crystallography structures of ABCA1 (PDB 8Y6H) and PPARγ (PDB 8BF2) were downloaded from the Protein Data Bank (https://www.rcsb.org/). As the human SR-BI experimental structure is unavailable, the structure predicted by AlphaFold (https://alphafold.ebi.ac.uk/) was utilized instead (AF Q8WTV0). The structural information of the saffron-derived compounds and simvastatin (control ligand) were retrieved from PubChem database (https://pubchem.ncbi.nlm.nih.gov/). Molecular docking was performed using the MolModa web server, a docking program that integrated Autodock Vina as the docking tool (https://pubs.acs.org/doi/abs/10.1021/acs.jcim.1c00203, https://pubmed.ncbi.nlm.nih.gov/38783339/). Prior to the docking step, MolModa was utilized to remove other co-crystallize molecules (such as ions and water) and duplicate subunits from the PDB files, add hydrogen, optimize the protein side chains, protonate the compounds and proteins at pH 7.4, and identify the binding pocket. The binding pocket location is defined based on the co-crystalized ligand in the catalytic site of the target proteins. Docking runs for each compound to a receptor protein were executed three times. The poses of compounds with the highest docking score for each receptor were selected for further analyses. The results for receptor-compound interactions were visualized using Pymol and PoseEdit (https://link.springer.com/article/10.1007/s10822-023-00522-4).

Statistical analysis

All gene expression data are presented as mean ± standard error of the mean (SEM). Statistical analyses were performed using IBM SPSS Statistics version 27 (IBM Corp., Armonk, NY, USA). The Shapiro–Wilk test was used to assess data normality prior to analysis. Differences among treatment groups were evaluated using one-way analysis of variance (ANOVA) followed by the Bonferroni post hoc test for multiple comparisons. A p-value < 0.05 was considered statistically significant for all analyses.

Results

Influence of saffron on SR-BI, PPARγ and ABCA1 gene expression

Saffron ethanolic extract modulated the hepatic expression of SR-BI, PPARγ, and ABCA1 in both early and established stages of atherosclerosis (Fig. 3). In early atherosclerosis, saffron treatment significantly upregulated SR-BI expression in a dose-dependent manner, with 4.59-fold and 3.65-fold increases observed in the S100 and S50 groups, respectively (p < 0.05). Expression remained highest in the statin-treated group with 6.21-fold increase. PPARγ expression also increased (4.99-fold in S100; 3.78-fold in S50) and ABCA1 increased (4.50-fold in S100; 3.60-fold in S50) relative to placebo, though neither reached statistical significance.

Fig. 3.

Relative fold change of hepatic (A) SR-BI, (B) PPARγ, and (C) ABCA1 gene expression in early and established atherosclerotic rabbits treated with saffron ethanolic extract. Data are presented as mean ± SEM (n = 5 per group). ND, normal diet; 4 W, 4 weeks (HCD group); 8 W, 8 weeks (HCD group); S50, 50 mg/kg/day saffron; S100, 100 mg/kg/day saffron; STATIN, statin; PCB, placebo.*p < 0.05 vs. ND; *p < 0.05 vs. STATIN; #p < 0.05 vs. PCB; †p < 0.05 vs. 4 W; §p < 0.05 vs. 8 W; ‡p < 0.05 vs. S50.

In established atherosclerosis, saffron continued to enhance SR-BI expression, with the S100 group showing the highest level (8.34-fold, p < 0.05), exceeding statin (5.52-fold) and S50 (3.34-fold) treatments. PPARγ showed moderate upregulation (2.88-fold in S100; 2.30-fold in S50) while ABCA1 expression also increased, peaking in the S50 group (4.36-fold), exceeding both S100 (2.07-fold) and statin (3.54-fold) treatments. The slightly higher response at S50 compared to S100 may reflect a non-linear dose relationship, which is discussed further in the discussion. The findings demonstrate that saffron exerts a dose- and stage-dependent modulatory effect on RCT-related genes, with significant upregulation for SR-BI, while PPARγ and ABCA1 display positive but statistically non-significant trends.

Molecular docking and binding affinity prediction

Molecular docking was performed to predict potential interactions between saffron’s bioactive compounds and key proteins involved in RCT (PPARγ, SR-BI, and ABCA1). Docking analyses revealed favourable interactions across all targets, with saffron constituents exhibiting moderate-to-strong binding affinities comparable to simvastatin (Table 2). Crocetin demonstrated the strongest predicted affinity for PPARγ (− 7.75 kcal/mol), followed by quercetin (− 7.58 kcal/mol) and dimethylcrocetin (− 7.57 kcal/mol), exceeding that of simvastatin (− 6.58 kcal/mol). For SR-BI, crocetin again showed the highest affinity (− 7.24 kcal/mol), followed by picrocrocin (− 6.79 kcal/mol) and quercetin (− 6.59 kcal/mol), all stronger than simvastatin (− 5.28 kcal/mol). In the case of ABCA1, quercetin (− 8.35 kcal/mol), crocetin (− 8.26 kcal/mol), and dimethylcrocetin (− 8.15 kcal/mol) exhibited comparable affinities to simvastatin (− 8.93 kcal/mol). Figure 4 shows docking visualization of saffron compounds with (A) ABCA1, (B) PPARγ, and (C) SR-BI, highlighting 2D interactions of the top binders. Saffron major constituents showed consistent binding potential toward lipid-regulating targets. These in silico predictions complement the in vivo gene expression findings, suggesting saffron’s multi-targeted modulation of lipid pathways. However, as molecular docking provides predictive insights, further protein–ligand and functional validation studies are required to confirm biological relevance.

Table 2.

Docking binding affinities of selected compounds from saffron and Simvastatin against RCT-related proteins (PPARγ, SR-BI, and ABCA1).

Compounds	PubChem CID	ABCA1 (PDB 8Y6H)	PPARγ (PDB 8BF2)	SR-BI (AF Q8WTV0)
	Docking binding affinities (kcal/mol)
Crocetin	5,281,232	− 8.26	− 7.75	− 7.24
Picrocrocin	130,796	− 7.18	− 6.64	− 6.79
Dimethycrocetin	5,316,132	− 8.15	− 7.57	− 6.54
Safranal	61,041	− 6.91	− 5.28	− 5.54
Quercetin	5,280,343	− 8.35	− 7.58	− 6.59
Simvastatin	54,454	− 6.58	− 5.28	− 8.93

*crocin has failed to dock with MolModa, SwissDock, SeemDock due to a large number of rotatable bonds.

Fig. 4.

Docking visualisation of the compounds from saffron to the binding pocket of (A) ABCA1, (B) PPARγ and (C) SR-BI. 2D receptor-compound interactions represent the docked compounds that have the highest binding affinity to each receptor (quercetin, crocetin, crocetin, respectively).

ADMET profiles of major phytocompounds in saffron

The pharmacokinetic and drug-likeness properties of saffron’s bioactive compounds were evaluated using the SwissADME platform. All compounds complied with Lipinski’s rule of five, indicating favorable drug-likeness with no major violations. High gastrointestinal absorption was predicted for all compounds, while most showed limited blood–brain barrier permeability. Quercetin and crocetin were predicted to inhibit CYP3A4 and CYP2C9, suggesting possible metabolic interactions, whereas safranal showed minimal enzyme inhibition potential. Predicted bioavailability scores (0.55–0.85) reflected moderate to good oral absorption. Table 2 summarizes the physicochemical and pharmacokinetic parameters of the analyzed saffron phytoconstituents.

Table 3.

ADMET analysis of saffron phytoconstituents (SwissADME predictions).

ADMET	Compounds
	Crocetin	Dimethylcrocetin	Picrocrocin	Safranal	Quercetin
Molecular weight (g/mol)	328.40	356.46	330.37	150.22	302.24
Topological polar surface area (Ų)	74.60	52.60	116.45	17.07	131.36
H-bond acceptors	4	4	7	1	7
H-bond donors	2	0	4	0	5
Molar refractivity	98.48	107.12	81.08	47.06	78.03
GI absorption	High	High	High	High	High
BBB permeant	No	Yes	Yes	Yes	No
P-gp substrate	No	Yes	Yes	No	Yes
CYP1A2 inhibitor	Yes	No	No	No	No
CYP2C19 inhibitor	Yes	Yes	No	No	No
CYP2C9 inhibitor	Yes	Yes	No	No	No
CYP2D6 inhibitor	No	No	No	No	Yes
CYP3A4 inhibitor	Yes	No	No	No	Yes
Lipinski rule violations	0	0	0	0	0
Bioavailability score	0.85	0.55	0.55	0.55	0.55
hERG inhibition	Low	Low	Low	Low	Low
Hepatotoxicity	No	No	No	No	No
Ames mutagenicity	Negative	Negative	Negative	Negative	Negative
Carcinogenicity	Non- carcinogenic	Non- carcinogenic	Non- carcinogenic	Non-carcinogenic	Non-carcinogenic
Skin sensitization	None	Minimal	Minimal	Mild	None
Overall toxicity	Safe	Safe (low CYP inhibition)	Safe (low absorption)	Safe (mild skin sensitization)	Safe (mild CYP interaction)

ADMET parameters were obtained using the SwissADME web tool (http://www.swissadme.ch) based on canonical SMILES retrieved from PubChem.GI = gastrointestinal absorption; BBB = blood–brain barrier; P-gp = P-glycoprotein. Values are computational predictions and May differ from in vivo Pharmacokinetic data.

Discussion

The present findings highlight the modulatory effects of saffron extract on key regulators of RCT (SR-B1, ABCA1, and PPAR-γ), offering insights into its potential role in atherosclerosis management. Our findings demonstrated that both statin and saffron treatments significantly upregulated hepatic SR-B1 gene expression, with expression levels varying by treatment dose and stage of atherosclerosis. SR-B1 is a critical mediator in the uptake of high-density lipoprotein (HDL)-derived cholesterol by the liver, facilitating the final step in the RCT pathway. In early-stage atherosclerosis, saffron treatment significantly upregulated SR-BI expression with 4.59-fold and 3.65-fold increases observed in the S100 and S50 groups, respectively (p < 0.05). Expression remained highest in the statin-treated group with 6.21-fold increase. In established atherosclerosis, the S100 group showed the highest SR-B1 expression (8.34-fold; p < 0.05), exceeding that of the statin group (5.52-fold). This is consistent with previous reports showing that statins enhance the expression of key cholesterol transporters, including ABCA1, ABCG1, and SR-BI, in hepatic and peripheral tissues, while also upregulating Apo-AI in hepatocytes to facilitate RCT³⁰. Saffron at 100 mg/kg (S100) also induced a significant upregulation of SR-B1(4.59-fold; p < 0.05), followed by the 50 mg/kg dose (S50), which produced a 3.91-fold increase (p < 0.05), both significantly higher than baseline and placebo. These results are supported by earlier studies indicating that crocin, a key bioactive component of saffron, upregulates SR-B1 expression, promotes HDL-C uptake, and facilitates RCT³¹. Moreover, SR-B1 has been reported to regulate macrophage activity within plaques, improving efferocytosis and clearance of apoptotic cells, thereby promoting plaque stability and reducing necrotic core formation³². Saffron’s sustained upregulatory effect may be attributed to its ability to modulate inflammatory signaling and maintain cholesterol homeostasis via nuclear receptor pathways such as LXR and PPARs³³.

PPAR-γ is a nuclear receptor involved in lipid storage, adipogenesis, insulin sensitivity, and inflammation regulation¹². In early-stage atherosclerosis, saffron treatment resulted in a dose-dependent upregulation of hepatic PPAR-γ gene expression. The S100 group demonstrated a 4.99-fold increase, while the S50 group showed a 3.78-fold increase compared to baseline. In established atherosclerosis, saffron maintained a moderate effect, with the S100 and S50 groups exhibiting 2.88-fold and 2.30-fold increases in PPAR-γ expression, respectively. The observed upregulation suggests that saffron’s bioactive constituents may influence lipid metabolism through antioxidant and anti-inflammatory mechanisms. Our findings are in line to prior research that demonstrated saffron’s ability to modulate gene expression at the epigenetic level, inducing chromatin remodeling, activating tumor suppressors, and promoting apoptosis in cancer models³⁴. Moreover, activation of PPAR-γ and PPAR-α has been shown to enhance ABCA1 expression via LXR-α activation, thereby facilitating cholesterol efflux, reinforcing the relevance of PPAR-γ upregulation in lipid regulatory pathways³⁵. This finding is consistent with past studies where crocin upregulates PPAR-γ, leading to improved lipid profiles through decreased triglyceride accumulation and increased adiponectin secretion³⁶. As a central regulator of fatty acid uptake, triglyceride synthesis, and insulin sensitivity, PPAR-γ plays a vital role in lipid and glucose homeostasis^37,38. Its activation is also associated with anti-atherosclerotic effects, including enhanced cholesterol efflux, reduced vascular inflammation, and improved endothelial function³⁹.

Statin treatment resulted in the highest PPAR-γ expression levels across both stages, with a 6.34-fold increase in early atherosclerosis and a 3.95-fold increase in established disease (both p < 0.05 vs. baseline and placebo). Previous studies have shown that simvastatin modulates PPAR-γ expression under pro-inflammatory conditions, indicating that statins may exert part of their lipid-lowering and anti-inflammatory effects through this pathway⁴⁰. Although the PPAR-γ upregulation observed with saffron did not exceed that of statins, the consistent dose-responsive elevation indicates a potential modulatory effect. These findings suggest that optimizing saffron dosage or extending the treatment duration may further enhance its therapeutic efficacy in targeting PPAR-γ–mediated pathways in atherosclerosis.

PPARs and LXRs cooperatively regulate ABCA1 expression and HDL biogenesis through an integrated signal transduction pathway. ABCA1 plays a critical role in lipid homeostasis by facilitating the efflux of cellular cholesterol and phospholipids to apolipoprotein A-I (ApoA-I), initiating the formation of nascent HDL particles. As this is the rate-limiting step in RCT, the regulation of ABCA1 has significant implications for atherosclerotic plaque development and regression⁴¹. In early-stage atherosclerosis, saffron treatment showed a moderate upregulation with a 3.60-fold increase observed in the S50 group and a 4.50-fold increase in the S100 group. In established atherosclerosis, S50 group exhibited the highest ABCA1 expression (4.36-fold), surpassing both the statin group (3.54-fold) and the S100 group (2.07-fold). The observed inverse dose–response pattern indicates that saffron’s therapeutic potential may depend on both the stage of atherosclerosis and the administered concentration. In early disease, lipid accumulation predominates, whereas later stages are characterized by heightened inflammation. This transition may enhance the relevance of saffron’s antioxidant and receptor-modulating activities, accounting for its stage-specific efficacy. These findings may be explained by the dose-dependent activation or suppression of nuclear receptors, where moderate doses may optimally activate LXR-α, while higher doses could trigger feedback inhibition or receptor desensitization, attenuating ABCA1 expression⁴². While these effects may be lower than those seen in the statin-treated group (6.15-fold), it underscores saffron’s capacity to enhance ABCA1 expression. Statins primarily act in the early stages of atherosclerosis by competitively inhibiting HMG-CoA reductase, thereby reducing hepatic cholesterol synthesis and preventing lipid accumulation^43,44. In contrast, saffron and its bioactive compounds, such as crocin and crocetin, have demonstrated antioxidant and anti-inflammatory properties that may confer greater therapeutic potential in later stages of the disease, when oxidative stress and endothelial dysfunction become predominant pathological features^13,45. Statins have been reported to indirectly upregulate ABCA1 expression, potentially involving LXR/RXR signaling, but this effect appears to be secondary to changes in intracellular cholesterol homeostasis rather than resulting from direct activation of these nuclear receptors^41,46.

Although PPARγ and ABCA1 changes did not reach statistical significance, these transcriptional patterns may reflect early or partial activation preceding measurable protein-level outcomes. Changes in cell function can happen through post-transcriptional and post-translational processes, such as mRNA stabilization, translation control, or receptor phosphorylation, even when mRNA levels stay the same⁴⁷. Previous studies have shown that crocin increases phosphorylated PPARγ protein levels, indicating receptor activation even in the absence of marked transcriptional changes⁴⁸. Similarly, longer treatment durations or purified fractions of saffron constituents have demonstrated stronger effects on lipid-regulating genes^49,50. These findings suggest that saffron modulates lipid metabolism through a multi-targeted mechanism involving PPARγ activation, LXR/RXR signaling, and SR-BI upregulation, collectively enhancing RCT efficiency.

Docking analyses revealed that saffron bioactives exhibited favorable binding affinities toward ABCA1, PPARγ, and SR-BI. Simvastatin was included as a benchmark for the interactions of saffron-derived compounds with RCT proteins. Crocetin’s affinity for PPARγ supports its role as a natural partial agonist that can modulate lipid metabolism and inflammatory signaling which is consistent with prior studies describing crocetin-mediated receptor activation through hydrogen and hydrophobic bonding^51–53. Saffron compounds showed higher predicted affinity than simvastatin for SR-BI, suggesting enhanced receptor activity. This is in line with past reports that crocin upregulates SR-BI expression and promotes plaque stabilization, potentially through activation of transcriptional regulators such as PPARγ and LXR^10,31,54. Crocin could not be successfully docked using MolModa, SwissDock, or SeemDock, likely due to its large, highly flexible structure with numerous rotatable bonds that expand the conformational search space beyond the practical limits of standard docking algorithms, a known limitation reported in molecular docking studies⁵⁵.

The ADMET analysis complements the docking findings by confirming that saffron’s bioactive compounds exhibit favorable oral drug-likeness and safety profiles. Their moderate bioavailability and limited blood-brain barrier permeability reduce systemic toxicity risks while supporting therapeutic applicability. The ADMET and docking analyses highlight saffron’s translational potential, as its major constituents, particularly crocetin and quercetin, exhibited moderate-to-strong binding affinities toward lipid-regulating targets, in some cases comparable to simvastatin. The agreement between computational and experimental findings highlights saffron’s multi-targeted modulation of cholesterol transport pathways. While these results provide valuable mechanistic insights, further in vivo pharmacokinetic and protein-binding studies are required to confirm metabolic stability, safety, and biological relevance.

While these results highlight saffron’s multi-targeted effects on cholesterol regulation, several limitations have been recognized to guide future validation and translational studies. Although this study provides molecular evidence supporting saffron’s role in modulating RCT–related pathways. However, the modest sample size may limit statistical power and generalizability. Protein-level analyses were not conducted, and future studies incorporating Western blot or proteomic validation would clarify whether the observed transcriptional changes translate to functional protein alterations. Although the molecular docking results complemented the in vivo findings by predicting plausible ligand–receptor interactions, confirmatory functional and receptor activation assays are still needed to establish causal mechanisms. Further investigations integrating pharmacokinetic, proteomic, and functional analyses will be essential to validate these findings and advance saffron’s translational potential in atherosclerosis management.

Conclusion

This study demonstrates that saffron extract modulates key regulators of reverse cholesterol transport, including SR-BI, PPARγ, and ABCA1, supported by both in vivo gene expression and in silico docking analyses. The integration of molecular and pharmacokinetic data suggests that saffron’s bioactive constituents, particularly crocetin and quercetin, contribute to its multi-targeted effects on lipid metabolism. These findings position saffron as a promising adjunct in cardiovascular disease prevention, though further studies on dose optimization, pharmacokinetics, and clinical validation are needed to confirm its therapeutic relevance.

Author contributions

NAMK contributed to conceptualization, formal analysis, methodology, project administration, supervision, validation, and writing – review & editing. YMZ contributed to data curation, formal analysis, investigation, methodology, project administration, writing – original draft, and writing – review & editing. INAR and NIH contributed to data curation, investigation, and writing – review & editing. NAAN contributed to methodology, supervision, and writing – review & editing. CPO contributed to methodology, supervision, and writing – review & editing. SAJ contributed to software, data analysis, visualization, and writing – review & editing. MMT contributed to conceptualization, supervision, funding acquisition, and resources. SAM contributed to methodology, supervision, and data analysis.

Funding

This work was supported by the Higher Institution Centre of Excellence (HICoE) research grant 600-RMC/MOHE HICoE CARE-I 5/3 (01/2025) awarded to the Cardiovascular Advancement and Research Excellence Institute (CARE Institute), Geran Inisiatif Penyeliaan Universiti Teknologi MARA (UiTM) (600-RMC/GIP 5/3 (052/2022)), Strategic Research Partnership Universiti Teknologi MARA (100-RMC 5/3/SRP (067/2021),100-RMC 5/3/SRP PRI (048/2022), Universiti Teknologi MARA.

Data availability

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

Declarations

Competing interests

The authors declare no competing interests.