Phytoene, a Precursor of Lycopene in Tomatoes, Protects Against Metabolic Dysfunction‐Associated Steatotic Liver Disease in Mice

ABSTRACT

The health benefits of tomato consumption may derive from bioactive compounds other than lycopene in tomatoes. 15‐Cis‐phytoene, a colorless precursor of lycopene abundant in tomatoes, has not been well studied for its biological activity. We investigated whether dietary 15‐cis‐phytoene protects against metabolic dysfunction‐associated steatotic liver disease (MASLD) and whether this protection depends on the carotenoid cleavage enzymes, beta‐carotene‐15,15’‐oxygenase 1 (BCO1) and beta‐carotene‐9’,10’‐oxygenase (BCO2). Wild‐type (WT) and BCO1^−/−/BCO2^−/− double knock‐out (DKO) mice were fed a high‐refined carbohydrate diet (HRCD) with or without 15‐cis‐phytoene supplementation for 24 weeks. 15‐Cis‐phytoene supplementation markedly increased 15‐cis and all‐trans phytoene concentrations in intestine, liver, and serum, and significantly attenuated HRCD‐induced MASLD in WT mice. The protective effects of phytoene on MASLD in WT mice were associated with increased hepatic mRNA and protein levels of SIRT1, PGC1α, and PPARα; enhanced AMPK and ACC phosphorylation; and upregulated fatty acid oxidation genes, but not with the gut microbiota composition. Despite much higher phytoene accumulation in the liver and serum, phytoene did not exhibit protection against HRCD‐induced MASLD in DKO mice. Phytoene is a biologically active compound that contributes to the benefits of tomato consumption against MASLD. Its protective effects are likely mediated by cleavage metabolites through BCO1/BCO2.

This study examined the biological activity of 15‐cis‐phytoene, a colorless lycopene precursor abundant in tomatoes, in metabolic dysfunction‐associated steatotic liver disease (MASLD). Phytoene is a dietary bioactive compound contributing to the protective effects of tomatoes against high‐refined carbohydrate diet‐induced MASLD by enhancing hepatic fatty acid oxidation through the SIRT1‐PGC1α‐PPARα axis and the AMPK/ACC pathway, in a carotenoid cleavage enzymes (BCO1 and BCO2)‐dependent manner.

Abbreviations

ACADL

long‐chain acyl‐CoA dehydrogenase

ACADM

medium‐chain acyl‐CoA dehydrogenase

ACC

acetyl‐CoA carboxylase

AMPK

adenosine 5’‐monophosphate‐activated protein kinase

ANOVA

analysis of variance

ApoB

apolipoprotein B

ApoE

apolipoprotein E

ASVs

amplicon sequence variants

ATGL

adipose triglyceride lipase

BCO1

beta‐carotene‐15,15’‐oxygenase 1

BCO2

beta‐carotene‐9’,10’‐oxygenase 2

CPT1a

carnitine palmitoyl transferase 1 a

DGAT

acyl‐coenzyme A:diacylglycerol acyltransferase

DKO

BCO1^−/−BCO2^−/− double knock‐out

DKO

double knock‐out

FAS

fatty acid synthase

HADHA

hydroxyacyl‐CoA dehydrogenase trifunctional multienzyme complex subunit alpha

HRCD

high‐refined carbohydrate diet

HSL

hormone‐sensitive lipase

ISX

intestine‐specific homeobox

LDLr

low density lipoprotein receptor

LRP

low density lipoprotein receptor‐related protein

MASLD

metabolic dysfunction‐associated steatotic liver disease

MGL

monoacylglycerol lipase

MTTP

microsomal triglyceride transfer protein

NAD⁺

nicotinamide adenine dinucleotide

PCoA

principal coordinate analysis

PERMANOVA

permutational multivariate analysis of variance

PGC1α

peroxisome proliferator‐activated receptor gamma coactivator 1 alpha

PPARα

peroxisome proliferator‐activated receptor alpha

SCD

steroyl‐coenzyme A desaturase

SIRT1

sirtuin 1

SNPs

single nucleotide polymorphisms

SR‐B1

scavenger receptor class b type 1

TG

triglyceride

WT

wild‐type

1. Introduction

Metabolic dysfunction‐associated steatotic liver disease (MASLD) has become the most common type of chronic liver disease, with a global prevalence of 30% [1]. Although the pathogenesis of MASLD is complex and multifactorial, excessive intake of low‐quality carbohydrates, such as refined grains and added sugars in foods and beverages, is a major dietary risk factor for the development of MASLD [2, 3]. In the United States, nearly half of daily caloric intake derives from low‐quality carbohydrates, which also lack the health‐promoting nutrients commonly found in fruits and vegetables [4, 5]. Given the increasing burden of MASLD and the lack of effective therapeutic options, it is essential to identify dietary bioactive compounds that can mitigate MASLD development and progression.

Epidemiological studies have linked an inverse relationship between the consumption of tomatoes and tomato products, plasma lycopene levels, and the risk of chronic diseases, including MASLD [6]. Experimental studies from our laboratory [7, 8, 9] and others [10] have shown that tomato powder, tomato extract, lycopene, and lycopene metabolites attenuate high‐fat diet‐induced MASLD, inflammation, and liver cancer. Randomized clinical trials also showed that the supplementation of semidried tomato improves the lipid profile in Japanese subjects, and tomato juice supplementation ameliorates oxidative stress and preserves a proper immune surveillance modulation in obese children with MAFLD [11, 12]. However, lycopene is found together with other carotenoids in tomatoes and tomato products, raising the question of whether these other carotenoids have been overlooked in studies linking the consumption of tomato‐derived lycopene to health benefits [13, 14]. Tomato powder often confers greater protection than purified lycopene, including prolonging survival in a prostate carcinogenesis model [13, 15], protecting against alcoholic fatty liver disease [16] and attenuating oxidative stress in vitamin E deficient rats [17], suggesting additional bioactive components or synergistic interactions in tomatoes. Moreover, tomato powder supplementation in well‐trained athletes has shown superior results to isolated lycopene in reducing exercise‐induced oxidative stress [18].

Phytoene, a colorless carotenoid, is the first 40‐carbon upstream plant precursor in the biosynthesis of carotenoid pigments, such as lycopene. It is formed by the condensation of two molecules of the 20‐carbon compound geranylgeranyl pyrophosphate, catalyzed by phytoene synthase to form 15‐cis‐phytoene, the predominant isomer in a wide variety of fruits and vegetables (e.g., tomatoes, carrots, red grapefruits, watermelon, apricot, red peppers, and some citrus) [19]. Although 15‐cis‐phytoene is present in tomatoes and tomato products at levels comparable to lycopene [20], many “white‐creamy” or “golden‐yellow‐orange” tomatoes contain significantly higher levels of phytoene than red tomatoes due to the lack of enzymes required for the conversion of phytoene into red pigments (e.g., lycopene). In addition, high quantities of phytoene can also be obtained through biotechnological production systems, including fungi (e.g., Blakeslea trispora), yeasts, algae, and plant in vitro cultures [21, 22]. Phytoene is more efficiently absorbed than lycopene in in vitro intestinal cells [23], and in animals and in humans [13, 14, 24]. However, despite clear evidence of its bioavailability, the importance of investigating the bioactive properties of phytoene has been largely overlooked [14]. Several in vitro studies, which showed that phytoene decreased intracellular reactive oxygen species levels [25], inhibited the proliferation of cancer cells [26], and induced the expression of phase II enzymes, while only a few in vivo studies demonstrated that phytoene suppressed UV‐B light‐induced skin tumor formation in mice [27], protected against nicotine‐induced oxidative stress‐mediated pancreatic islet dysfunction in rats [28], and increased the median lifespan of Caenorhabditis elegans [29]. Pipitone et al. reported that the beneficial effects were more pronounced in the group receiving golden tomatoes than in the group receiving red tomatoes in a high‐fat diet–induced MASLD rat model [30]. Since many golden tomatoes contain significantly higher levels of colorless phytoene than red tomatoes, due to the lack of enzymes required for the conversion of phytoene into red pigments (e.g., lycopene), it is interesting to consider that the greater beneficial effects of golden tomatoes may be attributable, at least in part, to their higher phytoene content. Given the dearth of in vivo studies, it is essential to investigate the biological activity of phytoene to determine whether it contributes to the health benefits associated with tomato consumption against MASLD.

Carotenoids, including lycopene, can be cleaved by beta‐carotene‐15,15’‐oxygenase 1 (BCO1) at 15,15’ double bond and/or beta‐carotene‐9’,10’‐oxygenase (BCO2) at 9,10 and 9’10’ double bonds, generating biologically active metabolites, such as apo‐10’‐carotenoids [31]. Since ablation of BCO1 results in the overexpression of BCO2 [32], we utilized a BCO1^−/−/BCO2^−/− double knock‐out (DKO) mouse model to eliminate both enzymes and better characterize the role of carotenoid cleavage enzymes in MASLD and the bioactive effects of intact carotenoids [33]. Although previous studies using ¹⁴C‐phytoene in rodents [34] and ¹³C‐phytoene in humans [24] suggested that phytoene may be metabolized in some tissues, it remains unclear whether BCO1 and BCO2 are involved in phytoene cleavage or metabolism, and whether dietary phytoene exerts protective effects on MASLD depending on the presence of these enzymes. Additionally, we have shown that tomato powder supplementation mitigates high‐fat diet‐induced MASLD development, and this is accompanied by alterations of gut microbiota composition, including increased gut microbial alpha diversity, in DKO mice [35, 36]. At the genus level, tomato powder reduced the relative abundances of Bacteroides, Mucispirillum, Clostridium, and Parabacteroides, while increasing Lactobacillus and Bifidobacterium compared with high‐fat diet‐fed controls [35]. Given that excessive intake of refined carbohydrates has been linked to microbial dysbiosis [37], it is unknown whether phytoene alone can modulate gut microbiota composition and protect against high‐refined carbohydrate diet (HRCD)‐induced MASLD.

Sirtuin 1 (SIRT1) is a highly conserved nicotinamide adenine dinucleotide (NAD⁺)‐dependent protein deacetylase that plays a central role in regulating MASLD development, lipid metabolism, oxidative stress, and insulin resistance. We previously demonstrated that the ablation of BCO1 and BCO2 induced MASLD by down‐regulation of SIRT1 and ablation of SIRT1 activity promotes MASLD [38, 39]. In contrast, supplementation with tomato powder, lycopene, and lycopene metabolites upregulated hepatic SIRT1 and attenuated high‐fat diet‐induced MAFLD [35, 36, 40]. These studies implicate that the beneficial effects of tomato carotenoids are mediated through changes in SIRT1 function, the gut microbiome, and the gut‐liver axis. However, it remains critical to understand whether the protective effects associated with tomato powder on MASLD can be recapitulated by purified phytoene in a BCO1/BCO2‐dependent manner, and whether phytoene influences both the liver and the gut microbiome by normalizing the HRCD‐dysregulated SIRT1 pathway.

In the present study, we examined whether dietary phytoene supplementation protects against MASLD development in a sex‐ and BCO1/BCO2‐dependent manner. We selected the HRCD‐induced MASLD mouse model because we previously demonstrated that an isocaloric, pair‐fed HRCD induces more hepatic steatosis and inflammation than a high‐fat diet, associated with downregulation of SIRT1 [41, 42]. This model captures key features of MASLD pathogenesis, allowing us to determine whether the protective effects of phytoene are mediated through restoration of SIRT1 signaling. Furthermore, we used BCO1^−/−/BCO2^−/− DKO mouse model to determine whether the protective effects of phytoene against MASLD are mediated by the intact parent compound or by its downstream cleavage metabolites. This model also enhances our understanding of how genetic variation in the BCO1 and BCO2 genes may influence phytoene metabolism and its biological effects. We also evaluated tissue levels and distribution of phytoene and its isomers, gut microbiota composition, and key molecular pathways to elucidate the underlying mechanisms.

2. Materials and Methods

2.1. Animals and Experimental Procedure

All experimental procedures were reviewed and approved by the Institutional Animal Care and Use Committee of Tufts University (H2025‐113 entitled “Tomato carotenoid protection via targeting the gut‐liver axis”). WT and BCO1/BCO2 DKO male and female mice on a C57BL/6J genetic background were generated as previously described [39]. The WT and BCO1/BCO2 DKO mice studies were conducted independently, but under identical experimental conditions, including diet supplementation, housing, and monitoring protocols. After a 2‐week acclimation period, 8‐week‐old WT and BCO1/BCO2 DKO mice were randomly assigned to receive either a HRCD (66.5% total calories from carbohydrates, including sucrose; #F6395; Bio‐Serv, USA) to induce MASLD, as previously described [33], or HRCD supplemented with phytoene (10 mg/kg diet) for 24 weeks. Mice were fed the experimental diets ad libitum with free access to water. Food intake was measured for first two weeks. Body weight was measured weekly throughout the experimental study. Fecal samples were collected one day before sacrifice. At the end of the dietary intervention, all mice were anesthetized with isoflurane and sacrificed for tissue collection, without overnight fasting. Liver and mesenteric adipose tissue were excised, and intestinal mucosa was scraped from the entire small intestine, then all tissues were frozen in liquid nitrogen and stored at −80°C for further analyses. The central left lobe of the liver was fixed in 10% buffered formalin at 4°C for histological analysis.

2.2. Phytoene Dose Information

15‐Cis‐phytoene [extracted from fungal cells (Blakeslea trispora) with a concentration of 3.4 mg phytoene/ml in oil and purity > 95%, Vitan LTD, Ukraine] was diluted in 1 mL ethanol and mixed into the semi‐purified HRCD diet at a concentration of 10 mg/kg diet under the red light. The diet was freshly prepared every 2 weeks and stored at 4°C. Based on the dose conversion using the standard human equivalent dose formula [43], the phytoene supplemented dose of 10 mg/kg diet is equivalent to approximately 9.6 mg phytoene per day for a 60‐kg adult. Since raw tomatoes contain variable amounts of phytoene for example, from 0.3 to 252.6 mg/100 g dry weight [44] and average is ∼1.86 mg phytoene per 100 g, and tomato paste has much higher concentrations of phytoene (8.36 mg/100 g), the amount of phytoene (9.6 mg phytoene) is achievable through a regular diet, such as by consuming ∼115 g of tomato paste [45]. Using HPLC analysis, we confirmed that 15‐cis‐phytoene was the predominant isomer in the diet (∼99%) and phytoene concentration remained stable in the diet during the feeding period.

2.3. Liver Histopathology Evaluation

Liver sections (5 µm thick) from the central left lobe, embedded in paraffin, were stained with hematoxylin and eosin. The extent of steatosis, including both macro‐ and micro‐vesicular fat accumulation, was evaluated by estimating the percentage of liver section containing lipid vacuoles as previously described [33, 39]. Grading was performed at 100 X magnification across 20 randomly selected fields, using the following criteria: grade 0, ≤5%; grade 1, 5% to 25%; grade 2, 26% to 50%; grade 3, 51% to 75%; and grade 4, 76% to 100%, as previously described [39].

2.4. HPLC Analysis

The concentration of phytoene and its isomers in serum and tissues (liver 100 mg; intestinal mucosa 60–70 mg) were analyzed by HPLC analysis using an established method from our laboratory [8]. This method allowed the determination of the geometrical isomers of the main dietary carotenoids and BCO1/BCO2 cleavage metabolites, apo‐carotenoids [46]. The HPLC system consists of a Waters 2996 photodiode array detector (Waters Corporate, USA), a Waters 2695 Alliance separation module, and a YMC C30 column (S‐3 micron, 4.6 mm × 150 mm; YMC, USA) with a flow rate of 1.00 mL/min [8]. Phytoene was quantified at 285 nm relative to the internal standard by determining peak areas, calibrated against phytoene standard. All procedures were performed under red light.

2.5. RNA Extraction and Quantitative Real‐Time PCR (qPCR)

Liver (20–30 mg) and MAT (90–100 mg) were homogenized and total RNA was extracted using RNeasy Mini columns (Qiagen, Germany) according to the manufacturer's protocol. RNA concentration and purity were determined using a Nanodrop 100 spectrophotometer. cDNA was synthesized with a High Capacity cDNA Reverse Transcription kit (Applied Biosystems, USA). qPCR was performed using FastStart Universal SYBR Green Master Mix (Roche, USA) on an Applied Biosystems 7000 machine. Primer sequences are listed in Table S1. Ppia and Actb were used as loading controls for liver and intestinal mucosa, respectively.

2.6. Western Blot Analyses

Liver tissue (60–100 mg) was homogenized to perform Western blot analyses as previously described [39]. The following antibodies were used: SIRT1 (Santa Cruz), peroxisome proliferator‐activated receptor gamma coactivator 1 alpha (PGC1α; Santa Cruz), peroxisome proliferator‐activated receptor alpha (PPARα; Abcam), adenosine 5’‐monophosphate‐activated protein kinase (AMPK), p‐AMPK (Thr172), acetyl‐CoA carboxylase (ACC), p‐ACC (Ser79; Cell Signaling), and β‐actin (Sigma‐Aldrich). Relative band intensities were quantified with ImageJ, with β‐actin as the loading control.

2.7. Hepatic Triglyceride (TG) Analysis

Hepatic TG levels were measured using a commercial TG colorimetric assay kit (Cayman, USA), according to the manufacturer's instructions.

2.8. Fecal DNA Extraction and 16S rRNA Gene Sequencing Analysis

Fecal DNA was extracted using QIAmp PowerFecal Pro DNA kits (Qiagen) according to the manufacturer's instructions. The library construction and sequencing were performed on an Illumina Miseq platform in the Tufts University Core Facility Genomics. Purified DNA was used for amplification of the V4 regions of the 16S rRNA gene using 250 bp pair‐end reads. Sequence data were processed through the DADA2 pipeline to produce amplicon sequence variants (ASVs).

Filtered ASVs were analyzed through the phyloseq package in R software. Alpha diversity and Bray–Curtis dissimilarity (beta diversity) were calculated. Principal coordinate analysis (PCoA) based on Bray–Curtis dissimilarity was performed with phyloseq, and ordination plots were visualized with ggplot2. Relative abundances were aggregated at the genus level. The top 10 most abundant genera within each group were identified, and all remaining taxa were combined into an ‘Others’ category. Bar plots were generated with ggplot2 to visualize genus level relative abundances across groups.

2.9. Statistical Analysis

All statistical analyses were performed using RStudio and GraphPad Prism version 10.0 (GraphPad Software). Data are presented as mean ± SEMs, and normality was assessed with the Shapiro–Wilk test. Two‐way analysis of variance (ANOVA) was used to evaluate the effects of phytoene, sex, and their interaction on primary outcomes followed by Tukey's post hoc test. Comparisons of phytoene levels between male and female mice within the HRCD supplemented with phytoene groups, as well as comparisons of gene and protein levels between HRCD and HRCD with phytoene groups, were performed using Student's t‐test or the Mann–Whitney U test, depending on data distribution. Differences in overall microbial community composition between groups were tested using permutational multivariate analysis of variance (PERMANOVA) with the vegan package. Differences in genus‐level relative abundance between the two groups were assessed for the top 10 most abundant genera within each group using the Wilcoxon rank‐sum test, with multiple testing correction by the Benjamini–Hochberg procedure. Statistical significance was established at a threshold of p < 0.05.

3. Results

3.1. Phytoene Inhibits HRCD‐Induced MASLD in WT Mice

Initial body weight and final body weight differed significantly between sexes, with males weighing more than females (Table 1). Phytoene supplementation did not alter daily food intake but significantly reduced final body weight and body weight gain in both males and females compared to the HRCD group (p < 0.05), without phytoene‐sex interaction. Liver weight and liver‐to‐body weight ratio were significantly lower in females compared with males (p < 0.001), but neither parameter was affected by phytoene. Hepatic triglyceride levels were significantly decreased by phytoene supplementation in both males and females (Table 1). Although modest differences in hepatic triglyceride levels were observed by sex (p < 0.001), no significant phytoene‐sex interaction was observed (Table 1), suggesting that phytoene exerted comparable protective effects against HRCD‐induced hepatic steatosis in both males and females. Phytoene supplementation significantly reduced mesenteric fat weight (p < 0.05), although this effect was not observed in mesenteric fat‐to‐body weight ratio. Histopathological analysis (Figure 1) demonstrated that HRCD feeding induced pronounced hepatic steatosis in both males and females, whereas phytoene supplementation markedly attenuated lipid droplet accumulation.

TABLE 1.

Primary outcomes of WT mice fed HRCD or HRCD supplemented with phytoene. ^a

	Male		Female		p values for two‐way ANOVA
Experimental group	HRCD	HRCD+PTE	HRCD	HRCD+PTE	PTE effect	Sex effect	PTE‐Sex interaction
Number of mice	10	13	8	10
Initial body weight (g)	24.38 ± 0.56	25.15 ± 0.60	20.41 ± 0.57	19.87 ± 0.51	0.85	<0.001	0.27
Final body weight (g)	52.94 ± 1.48	49.62 ± 2.21	47.30 ± 3.21	36.32 ± 3.52	<0.05	<0.05	0.16
Body weight gain (g)	28.55 ± 1.88	24.48 ± 2.10	26.90 ± 2.82	16.45 ± 3.19	<0.05	0.06	0.22
Liver weight (g)	4.67 ± 0.41	4.14 ± 0.39	3.03 ± 0.37	2.36 ± 0.35	0.14	<0.001	0.86
Liver weight/body weight (%)	8.72 ± 0.64	8.17 ± 0.54	6.34 ± 0.49	6.26 ± 0.39	0.57	<0.001	0.66
Hepatic triglycerides (mg/g)	0.54 ± 0.01	0.48 ± 0.01	0.48 ± 0.01	0.44 ± 0.01	<0.001	<0.001	0.36
Mesenteric adipose tissue weight (g)	0.91 ± 0.11	0.89 ± 0.07	1.02 ± 0.11	0.65 ± 0.07	<0.05	0.46	0.06
Mesenteric adipose tissue weight/body weight (%)	1.72 ± 0.20	1.78 ± 0.12	2.15 ± 0.21	1.77 ± 0.10	0.31	0.19	0.17

^a Values are presented as means ± SEM and percentages. Sample sizes were n = 8–13 per group for initial body weight, final body weight, liver weight, and mesenteric adipose tissue weight, and n = 7 for hepatic triglycerides. Two‐way ANOVA followed by Tukey's post hoc test was used to analyze the effects of PTE, sex, and PTE‐sex interactions, with results shown as p values. PTE, phytoene.

FIGURE 1.

Histopathological analysis of the livers of mice fed HRCD or HRCD supplemented with phytoene. (A) Representative images of Hematoxylin and Eosin‐stained hepatic slides at 100 X magnification. (B) Hepatic steatosis score. Hepatic steatosis was graded according to steatosis magnitude, including both macro‐ and micro‐vesicular fat accumulation. Data labeled with different letters indicate significant differences (p < 0.05). Values are means ± SEM. Statistical analysis was performed using two‐way ANOVA followed by Tukey's post hoc test. n = 8–13 animals per group. PTE, phytoene.

3.2. Phytoene Supplementation Results in Significant Accumulation of both 15‐cis‐ and All‐trans‐isomers in Intestinal Mucosa, Liver, and Serum in a Sex‐Dependent Manner

Because the HRCD contained no carotenoids, phytoene was not detected in the intestinal mucosa, liver, or serum of HRCD‐fed control mice (Table 2). In contrast, phytoene accumulated in intestinal mucosa, liver, and serum of mice fed HRCD supplemented with phytoene for 24 weeks (Table 2). Both 15‐cis‐ and all‐trans‐isomer of phytoene across tissues were detected by HPLC analysis (Figure 2B), matching the retention times of 15.6 and 16.9 min, respectively, and absorption maximum at 285 nm of the 15‐cis‐ and all‐trans‐isomer of the phytoene standard (Figure 2A). In the intestinal mucosa, phytoene concentrations were comparable between male and female phytoene‐supplemented mice (∼ 2 to 3.6 nmol/g) (Table 2). In both sexes, phytoene was predominantly present as the 15‐cis isomer, resulting in a 15‐cis:all‐trans distribution of 84:16. In the liver, total phytoene accumulation was substantially greater than in the intestinal mucosa. Females showed 3.5‐fold higher hepatic total phytoene concentrations than males (p < 0.05), largely driven by increased accumulation of the 15‐cis phytoene (Table 2). To investigate potential mechanisms underlying the higher hepatic phytoene accumulation in females, we assessed the expression of carotenoid transporters, scavenger receptor class b type 1 (SR‐B1) and CD36, in the intestinal mucosa. Within the phytoene‐supplemented groups, females exhibited significantly higher intestinal Sr‐b1 mRNA expression than males, whereas Cd36 expression did not differ between sexes. No sex differences were observed in the chylomicron assembly‐ and export‐related genes, ApoB and MTTP (Figure 3A). In the liver, expression of Apoe, a key ligand for chylomicron remnant receptors, as well as Lrp and Cd36, was significantly higher in females than in males, whereas Ldlr and Sr‐b1 expression did not differ between sexes (Figure 3B). Females maintained 15‐cis as the predominant form in the liver, whereas males showed a shift toward predominance of the all‐trans isomer (Table 2). In the serum, total phytoene concentrations were 2.2‐fold higher in females than in males (p = 0.06), and concentrations of the 15‐cis isomer were significantly higher in females than in males. The 15‐cis isomer predominated in females, whereas the all‐trans isomer predominated in males, resulting in 15‐cis/all‐trans distributions of 72:28 and 47:53, respectively.

TABLE 2.

Phytoene accumulation in tissues and serum of WT mice. ^a

		HRCD		HRCD+PTE b
Tissue	Phytoene isomers	Male	Female	Male	Female
Intestinal mucosa (nmol/g)	Total phytoene	ND	ND	3.57 ± 1.14	1.96 ± 0.59
	15‐cis‐isomer	ND	ND	3 ± 0.99	1.64 ± 0.49
	All‐trans‐isomer	ND	ND	0.57 ± 0.15	0.32 ± 0.11
	15‐cis:all‐trans ratio (%)			84:16	84:16
Liver (nmol/g)	Total phytoene	ND	ND	68.34 ± 11	238.2 ± 85 *
	15‐cis‐isomer	ND	ND	21.7 ± 3.3	170.3 ± 64.4 **
	All‐trans‐isomer	ND	ND	46.7 ± 7.9	68.0 ± 21.1
	15‐cis:all‐trans ratio (%)			32:68	71:29
Serum (nmol/L)	Total phytoene	ND	ND	241.1 ± 66.6	536.3 ± 120
	15‐cis‐isomer	ND	ND	113.5 ± 26.6	385.7 ± 94.7 *
	All‐trans‐isomer	ND	ND	127.6 ± 41.6	150.6 ± 29.3
	15‐cis:all‐trans ratio (%)			47:53	72:28

^a Values are presented as means ± SEM, n = 3–4 per group for intestinal mucosa, n = 6–10 for liver, and n = 6 for serum. ND, not detected; PTE, phytoene.

^b Comparisons of phytoene levels between male and female mice in the HRCD+PTE group were analyzed by unpaired Student's t‐test.

^* p < 0.05; significantly different from male HRCD+PTE group.

^** p < 0.01; significantly different from male HRCD+PTE group.

FIGURE 2.

HPLC chromatograms of phytoene standard and the liver sample from male mice fed HRCD supplemented with phytoene. (A) The peaks and UV spectrum (insert) of the 15‐cis‐isomer (peak 1) and all‐trans‐isomer (peak 2) of phytoene standard and (B) the liver sample measured at 285 nm. The peak 1 and peak 2 with retention times of 15.7 min and 17.0 min, respectively, and their UV spectrum (insert) were matched phytoene standard as 15‐cis and all‐trans phytoene.

FIGURE 3.

Effects of phytoene on the intestinal carotenoid transporter (SR‐B1 and CD36), chylomicron assembly and export‐related biomarkers (ApoB and MTTP), hepatic lipoprotein uptake receptor (CD36), hepatic chylomicron remnant receptors (LRP, LDLr, and SR‐B1) and their key ligand (ApoE) in mice fed HRCD supplemented with phytoene. Graphical representation of fold changes in the (A) mRNA expression of Sr‐b1, Cd36, Apob, and Mttp, and (B) mRNA expression of Apoe, Lrp, Ldlr, Cd36, and Sr‐b1 quantified via real‐time PCR in mice fed HRCD with phytoene. Data represents comparisons between male and female mice in the phytoene‐supplemented group. Values are means ± SEM. Statistical analyses were performed using unpaired Student's t‐tests (intestinal Sr‐b1 and Cd36, Apob, Mttp, Ldlr, and hepatic Sr‐b1) or Mann–Whitney U tests (hepatic Cd36, Lrp, and Apoe). *p < 0.05, **p < 0.01. n = 8 animals per group.

Because we have demonstrated that BCO2 preferentially cleaves non‐provitamin A carotenoids (e.g., cis‐lycopene and lutein) at the 9,10 or 9’,10’ double bond site to produce apo‐carotenoids [47, 48], we evaluated the potential formation of the apo‐carotenoid, geranylacetone (6,10‐dimethyl‐5,9‐undecadien‐2‐one), from phytoene via BCO2‐mediated cleavage at 9,10 or 9’,10’ double bond site of phytoene. Although geranylacetone was readily detected using an HPLC C30 column with authentic standards, it was not detected in any tissue or serum samples.

3.3. Phytoene Upregulates the Hepatic SIRT1 and Its Downstream Targets, PGC1α and PPARα, in WT Mice

In WT mice, phytoene supplementation significantly increased hepatic Sirt1 mRNA expression compared with HRCD‐fed controls (Figure 4A, left). SIRT1 protein levels were increased by 52% in phytoene‐treated mice (p < 0.05; Figure 4A, right). Phytoene significantly upregulated Pgc1α and Pparα mRNA expression (Figure 4B). Protein levels of both PGC1α and PPARα were significantly elevated in the HRCD supplemented with phytoene group, compared with HRCD‐fed controls, by 37% and 70%, respectively (Figure 4C).

FIGURE 4.

Effects of phytoene on the hepatic SIRT1 and its downstream targets, PGC1α and PPARα, in mice fed HRCD or HRCD supplemented with phytoene. Graphical representation of fold changes in the (A) mRNA expression of Sirt1 and protein expression of SIRT1, (B) mRNA expression of Pgc1α and Pparα, and (C) the protein expression of PGC1α and PPARα quantified via real‐time PCR or Western blot in liver of mice fed HRCD with or without phytoene. Values are means ± SEM. Statistical analyses were performed using unpaired Student's t‐tests (Sirt1, Pparα, PGC1α, and PPARα) or Mann–Whitney U tests (SIRT1 and Pgc1α). *p < 0.05, **p < 0.01. n = 6 – 13 animals per group. PTE, phytoene.

3.4. Phytoene Increases the Hepatic AMPK/ACC Pathway and Fatty Acid Oxidation in WT Mice

In WT mice, phytoene supplementation significantly increased AMPK phosphorylation (Thr172) compared with HRCD‐fed controls (Figure 5A). Consistent with this activation, phytoene enhanced ACC phosphorylation (Ser79) (p < 0.05; Figure 5B). Phytoene supplementation did not alter the expression of lipogenic markers, such as Fasn, Scd1, and Dgat2 (data not shown). However, fatty acid oxidation‐related genes were significantly upregulated. Specifically, phytoene supplementation significantly upregulated the expression of Cpt1a by 40% (p < 0.01), a rate liming enzyme for mitochondrial fatty acid uptake, as well as Hadha by about 20% (p < 0.05) when compared to the levels in HRCD‐fed controls (Figure 5C). Moreover, phytoene elevated the mRNA expressions of Acadm and Acadl (p < 0.05), which encode medium‐ and long‐chain acyl‐CoA dehydrogenases, respectively (Figure 5C).

FIGURE 5.

Effects of phytoene on the hepatic AMPK/ACC pathway and fatty acid oxidation in mice fed HRCD or HRCD supplemented with phytoene. Graphical representation of fold changes in the protein expression of (A) p‐AMPK (Thr172), (B) p‐ACC (Ser79), and the mRNA expression of mitochondrial fatty acid oxidation‐related genes: (C) Cpt1a, Hadha, Acadm, and Acadl quantified via real‐time PCR or Western blot in liver of mice fed HRCD with or without phytoene. Values are means ± SEM. Statistical analyses were performed using unpaired Student's t‐tests (p‐AMPK, Cpt1a, Hadha, and Acadm) or Mann‐Whitney U tests (p‐ACC and Acadl). *p < 0.05, **p < 0.01. n = 6–13 animals per group. PTE, phytoene.

3.5. No Changes in Gut Microbiome With Phytoene Supplementation

Alpha diversity, as assessed by Shannon and Simpson indices, showed no significant differences between HRCD and HRCD with phytoene supplementation groups (Figure 6A). Similarly, beta diversity analysis exhibited no clear separation of microbial communities between groups on PCoA based on Bray–Curtis dissimilarity (Figure 6B). Consistent with this observation, PERMANOVA analysis confirmed that groups explained only 9.7% of the variation in community structure and was not statistically significant (R ² = 0.097, p = 0.238). At the genus level, among top 10 genera from each group, the taxonomic composition of the gut microbiota revealed no difference between HRCD and HRCD with phytoene supplementation groups, with Lactobacillus, Faecalibaculum, Bifidobacterium, and Bacteroides as the predominant genera in both groups (Figure 6C).

FIGURE 6.

Effects of phytoene on gut microbiome in mice fed HRCD or HRCD supplemented with phytoene. (A) Alpha diversity between two groups (HRCD and HRCD supplemented with phytoene), including Shannon index and Simpson index. (B) PCoA plot on Bray–Curtis dissimilarity distances between two groups. (C) Relative abundance of the top 10 genera within each two groups. For (A), data are represented as median, interquartile range, and whiskers extending to 1.5x the interquartile range. For (B), axes 1 and 2 captured 38.5% and 26.9% in the variation between samples, respectively. Statistical analyses were performed using Wilcoxon rank‐sum test with Benjamini–Hochberg correction for multiple comparisons (A, C), and PERMANOVA for beta diversity (B). n = 7 animals per group. PTE, phytoene; ns, not significant.

3.6. Phytoene Accumulates to a Greater Extent in BCO1/BCO2 DKO Mice but Does Not Protect MASLD

As observed in WT mice, DKO males exhibited greater initial body weight, final body weight, and liver weight than females (p < 0.001; Table 3). Phytoene supplementation did not significantly affect body weight, liver weight, or adiposity parameters in either sex. Despite marked accumulation of phytoene in the liver and serum, phytoene supplementation did not reduce hepatic steatosis scores or hepatic triglyceride concentrations in DKO mice (Table 3). To determine whether the protective effects of phytoene against MASLD were dependent on BCO1 and BCO2, DKO mice were supplemented with phytoene, and tissue phytoene concentrations and isomer distributions were measured. Phytoene was not detected in tissues of DKO mice fed HRCD alone (Table 4). In contrast, phytoene supplementation resulted in significantly greater phytoene accumulation in the intestinal mucosa, liver, and serum of DKO mice than WT mice fed the same dose of phytoene. In the intestinal mucosa, total phytoene levels reached 7.34 ± 0.51 nmol/g in males and 3.50 ± 0.85 nmol/g in females (Table 4). These values represented a twofold increase in males compared with WT mice (3.57 ± 1.14 nmol/g, p < 0.05), whereas the increase in females relative to WT mice (1.96 ± 0.59 nmol/g) was not statistically significant. Consistent with WT mice (Table 2), phytoene in the intestinal mucosa was predominantly present as the 15‐cis isomer, with 15‐cis:all‐trans ratios of 81:19 in males and 82:18 in females. Hepatic total phytoene concentrations were significantly increased in both sexes, corresponding to a 2.1‐fold increase in males (p < 0.01) and a 2.6‐fold increase in females (p < 0.05) relative to WT mice. Females displayed markedly higher hepatic 15‐cis phytoene concentrations than males (p < 0.001), resulting in higher 15‐cis:all‐trans ratios in females (75:25) compared with males (34:66). Serum phytoene concentrations were also significantly greater in DKO mice than in WT mice, with 2.4‐fold and 2‐fold increases in males and females, respectively. Females exhibited higher serum 15‐cis phytoene levels than males (p < 0.01), yielding 15‐cis:all‐trans distributions of 73:27 in females and 42:58 in males.

TABLE 3.

Primary outcomes of DKO mice fed HRCD or HRCD supplemented with phytoene. ^a

	Male		Female		p values for two‐way ANOVA
Experimental Group	HRCD	HRCD+PTE	HRCD	HRCD+PTE	PTE effect	Sex effect	PTE‐Sex Interaction
Number of mice	9	10	10	10
Initial body weight (g)	25.39 ± 1.27	25.02 ± 0.88	18.88 ± 0.51	19.80 ± 0.60	0.75	<0.001	0.45
Final body weight (g)	44.97 ± 1.88	43.45 ± 1.28	37.87 ± 1.54	35.18 ± 1.67	0.20	<0.001	0.22
Body weight gain (g)	19.58 ± 1.08	18.43 ± 0.84	18.99 ± 1.08	15.38 ± 1.54	0.05	0.13	0.30
Liver weight (g)	2.80 ± 0.34	2.54 ± 0.20	1.95 ± 0.11	2.02 ± 0.12	0.65	<0.001	0.42
Liver weight/body weight (%)	6.09 ± 0.53	5.77 ± 0.32	5.20 ± 0.33	5.75 ± 0.24	0.76	0.22	0.24
Hepatic steatosis score	2.19 ± 0.19	2.13 ± 0.14	1.93 ± 0.11	1.93 ± 0.07	0.84	0.09	0.84
Hepatic triglycerides (mg/g)	0.54 ± 0.02	0.54 ± 0.01	0.53 ± 0.01	0.53 ± 0.02	0.89	0.62	0.98
Mesenteric adipose tissue weight (g)	0.70 ± 0.09	0.72 ± 0.07	0.71 ± 0.10	0.67 ± 0.04	0.93	0.79	0.68
Mesenteric adipose tissue weight/body weight (%)	1.57 ± 0.20	1.65 ± 0.13	1.85 ± 0.23	1.91 ± 0.11	0.69	0.12	0.95

^a Values are presented as means ± SEM and percentages. Sample sizes were n = 9–10 per group for initial body weight, final body weight, liver weight, mesenteric adipose tissue weight, and hepatic steatosis score, and n = 8 for hepatic triglycerides. Two‐way ANOVA followed by Tukey's post hoc test was used to analyze the effects of PTE, sex, and PTE‐sex interactions, with results shown as p values. PTE, phytoene.

TABLE 4.

Phytoene accumulation in tissues of BCO1/BCO2 DKO mice. ^a

		HRCD		HRCD+PTE b
Tissue	Phytoene isomers	Male	Female	Male	Female
Intestinal mucosa, nmol/g	Total phytoene	ND	ND	7.34 ± 0.51	3.50 ± 0.85**
	15‐cis‐isomer	ND	ND	5.96 ± 0.42	2.88 ± 0.81*
	All‐trans‐isomer	ND	ND	1.38 ± 0.24	0.62 ± 0.22
	15‐cis:all‐trans ratio (%)			81:19	82:18
Liver, nmol/g	Total phytoene	ND	ND	145 ± 22.5	610.6 ± 98.2***
	15‐cis‐isomer	ND	ND	49.7 ± 7.8	457.2 ± 92.8***
	All‐trans‐isomer	ND	ND	95.3 ± 15.0	153.4 ± 12.3*
	15‐cis:all‐trans ratio (%)			34:66	75:25
Serum, nmol/L	Total phytoene	ND	ND	588.9 ± 107.9	1087.3 ± 97.1**
	15‐cis‐isomer	ND	ND	265.8 ± 44.7	789.1 ± 109.4**
	All‐trans‐isomer	ND	ND	323.1 ± 67.4	298.3 ± 38.5
	15‐cis:all‐trans ratio (%)			42:58	73:27

abbreviations: ND, not detected; PTE, phytoene.

^a Values are presented as means ± SEM, n = 3–4 per group for intestinal mucosa, n = 3–8 for liver, and n = 6–7 for serum.

^b Comparisons of phytoene levels between male and female mice in the HRCD+PTE group were analyzed by unpaired Student's t‐test.

^* p < 0.05; significantly different from male HRCD+PTE group.

^** p < 0.01; significantly different from male HRCD+PTE group.

^*** p < 0.001; significantly different from male HRCD+PTE group.

However, despite phytoene supplementation resulting in significantly greater phytoene accumulation in the intestinal mucosa, liver, and serum of DKO mice than WT mice, phytoene supplementation did not alter hepatic Sirt1 mRNA expression and SIRT1 protein levels in DKO mice compared with HRCD‐fed controls (data not shown). Similarly, phosphorylation of AMPK or ACC was not increased by phytoene supplementation in DKO mice (data not shown), indicating the regulatory effects of phytoene on SIRT1, AMPK, and ACC pathways are mediated by cleavage metabolites generated by BCO1/BCO2 rather than by intact phytoene itself.

4. Discussion

The presence of both lycopene and phytoene in substantial amounts in tomatoes and tomato products raises the important question of whether the colorless carotenoid phytoene has been overlooked in studies linking tomato consumption with health benefits [14]. In this study, we provide strong experimental evidence that phytoene is a biologically active compound contributing to the health benefits of tomato consumption. Specifically, dietary phytoene supplemented at a physiologically achievable dose effectively inhibited HRCD‐induced MASLD in both male and female mice. We further show that phytoene protected against HRCD‐induced MASLD by modulating key hepatic molecular pathways, including regulation of SIRT1‐mediated signaling and enhancement of mitochondrial fatty acid oxidation via the PGC1α–PPARα pathway or the AMPK–ACC axis. Notably, because phytoene supplementation did not prevent MASLD in DKO mice, our findings reveal that the protective effects of phytoene are dependent on BCO1 and BCO2, suggesting that its biological activity is mediated primarily by its cleavage metabolites generated by these enzymes rather than by intact phytoene. These findings have broad implications, as phytoene is abundant in fruits and vegetables as a precursor of major carotenoids (e.g., lycopene, beta‐carotene and lutein). However, to date, epidemiological studies have not yet examined circulating phytoene concentrations or evaluated its potential role in preventing chronic disease risk [13, 14].

Although there are well‐recognized differences between mice and humans in the absorption, metabolism, and tissue accumulation of carotenoids (e.g., β‐carotene, lycopene, lutein, and β‐cryptoxanthin), phytoene exhibits high bioavailability in both species [24, 34]. This is likely due to its cis geometric conformation, which is more easily incorporated into mixed micelles in the gastrointestinal tract [24, 49, 50]. Consistent with this, dietary supplementation with cis‐phytoene at 10 mg/kg diet resulted in markedly higher serum and tissue concentrations than those previously reported in mice supplemented with purified lycopene at 100 mg/kg diet [51], further supporting the superior bioavailability of phytoene in mice. Regarding the potential translational relevance of our findings to humans, we found that phytoene supplementation (equivalent to approximately 9.6 mg/day for a 60‐kg adult) led to a serum concentration of 241 nmol/L in WT mice (Table 2), comparable to the concentration of 190 nmol/L in humans after 4 weeks of tomato juice consumption providing 5.8 mg/day of phytoene [52]. This serum level was also 3.3‐fold higher than plasma phytoene concentrations (71 nmol/L) reported in humans consuming 1 – 5 mg/day under controlled conditions [24]. Although the USDA Nutrient Database did not include phytoene [53], national dietary intake data from Luxembourg report an average daily phytoene intake of 2.0 mg [54]. Although tomatoes contain variable amounts of phytoene, depending on genetic and cultivar factors, developmental (ripening) stage, and environmental conditions, tomato paste has much higher concentrations of phytoene (8.36 mg/100 g) compared to raw tomatoes because the water is removed during production. Therefore, the amount of phytoene (9.6 mg phytoene per day for a 60‐kg adult) is achievable through a regular diet, such as by consuming ∼115 g of tomato paste [45]. Collectively, our study supports the physiological relevance and translational potential of phytoene as a dietary component for the prevention of MASLD. However, our study involves supplementation with purified phytoene rather than consumption of whole tomato products. Currently, relatively little information is available from clinical studies regarding the effects of food processing, cooking methods, or dietary fat co‐consumption on phytoene bioaccessibility and bioavailability. Therefore, further studies are needed to determine whether these factors are important considerations for whole‐food consumption rich in phytoene.

Although there are no reports regarding sex‐specific differences in phytoene levels, there is a known sex difference in circulating carotenoids in humans, with women displaying higher concentrations compared with men, independent of carotenoid intake. Our study revealed pronounced sex‐dependent differences in phytoene accumulation and isomer distribution in mice. In the intestinal mucosa of WT mice, phytoene concentrations were lower in females than males; however, the 15‐cis:all‐trans isomer distribution was comparable between sexes (∼84:16), indicating that sex did not affect phytoene isomerization in the intestine. Females showed significantly higher total phytoene concentrations in both the liver and serum than males, largely driven by greater accumulation of the 15‐cis phytoene in both WT and DKO mice. These differences may reflect sex‐dependent variation in carotenoid transport efficiency. In the intestine, we found that female mice exhibited significantly higher expression of SR‐B1, a key transporter involved in the intestinal absorption of dietary phytoene and other carotenoids, whereas Apob and Mttp expression did not differ between sexes, indicating comparable chylomicron assembly and export. It has been shown that estrogen can increase SR‐B1 expression by acting on the Sr‐b1 gene promoter through estrogen response elements [55]. Therefore, it is possible that the sex‐dependent differences in phytoene tissue accumulation are due to, at least in part, higher estrogen levels and increased intestinal SR‐B1 expression, which can enhance micellar uptake in females. In the liver, expression of Apoe, Lrp, and Cd36 was significantly higher in females than in males, suggesting enhanced hepatic uptake and clearance of chylomicron remnants. These findings suggest that females may potentially exhibit more efficient transfer of phytoene from intestine to the liver, leading to reduced intestinal retention but enhanced hepatic accumulation compared with males. Interestingly, in female mice, the 15‐cis isomer predominated across tissues, suggesting preferential systemic accumulation of cis phytoene. In contrast, male mice showed relative enrichment of the all‐trans isomer in the liver and serum despite similar intestinal isomer ratios, indicating that sex‐dependent differences in phytoene distribution likely arise primarily in the liver and are subsequently reflected in circulation. Moreover, given the comparable hepatic all‐trans isomer levels between sexes, this pattern may reflect more rapid metabolism or clearance of the 15‐cis isomer in males than in females. Interestingly, Boileau et al. showed that androgen depletion and food restriction increase hepatic lycopene accumulation in male rats and proposed that higher androgen levels may stimulate lycopene metabolism and degradation [56]. Because cis–trans isomerization can influence carotenoid bioactivity, further studies are needed to clarify whether these sex‐specific differences arise from intestinal absorption, hepatic metabolism, or systemic transport, and how they impact metabolic outcomes.

One of the important findings in the study is the strong in vivo evidence that phytoene undergoes similar enzymatic cleavage by BCO1 and BCO2, similar to lycopene. We observed that hepatic phytoene levels were elevated by 2.2‐ to 2.7‐fold, and serum levels were increased by 2.5‐ to 4.8‐fold in DKO mice relative to WT mice. These results indicate that phytoene is metabolized systemically by BCO1 and BCO2, and that whole‐body BCO1/BCO2 knockout leads to greater accumulation of phytoene in both tissues and circulation. Genetic variants in the BCO1 and BCO2 genes have been associated with alterations in carotenoid status and function in both humans and animals [57, 58]. Single nucleotide polymorphisms (SNPs) in the human BCO1 gene are common and are associated with reduced catalytic activity in the conversion of β‐carotene to vitamin A [59]. A SNP in BCO2 has been associated with an increased risk of age‐related eye disease [60] and altered regulation of the pro‐inflammatory cytokine IL‐18 concentrations [61]. Additionally, SNPs in the intestine‐specific homeobox (ISX) gene, which functions as a transcriptional repressor of BCO1 and SR‐B1, have been associated with substantial interindividual variation in the bioavailability and tissue accumulation of phytofluene in humans [62]. Our findings provide a new avenue to investigate whether genetic variants in carotenoid cleavage enzymes, including BCO1 and BCO2, as well as in the ISX and SR‐B1 genes, influence the bioavailability, tissue accumulation of phytoene, and its protective effects against MASLD.

Another key finding of this study is that, despite significantly higher tissue phytoene levels, DKO mice were not protected against HRCD‐induced MASLD, suggesting that phytoene's biological effects are mediated primarily by its cleavage metabolites rather than by intact phytoene. Carotenoid cleavage metabolites are known to exert a broad spectrum of biological functions and play important roles in mammalian health [63]. We have previously demonstrated that lycopene as a non‐provitamin A carotenoid can be preferentially cleaved by BCO2 and generates metabolites, including apo‐10’‐lycopenoids [47, 48], which may have more important biological roles than their parent compound against MASLD [64, 65, 66]. For examples, apo‐10’‐lycopenoic acid exhibits significant biological activities, including the induction of SIRT1 expression and the inhibition of MASLD in ob/ob mice [64] and high fat diet‐promoted MASLD and hepatocellular carcinoma in mice [9, 65], as well as the induction of nuclear translocation of Nrf2 (nuclear factor E2‐related factor 2), thereby inducing phase II detoxifying/antioxidant enzymes, including heme oxygenase‐1, NAD(P)H:quinone oxidoreductase 1 and glutathione S‐transferases [66]. Nrf2 plays a crucial role in maintaining mitochondrial homeostasis and defending oxidative damage in the liver caused by fatty acid accumulation during the progression of MASLD [67]. Although it remains unknown whether mammalian BCO1 and BCO2 are involved in phytoene cleavage, studies in plants have shown that tomato carotenoid cleavage dioxygenase 1 cleaves phytoene at the 9,10 and 9′,10′ double bonds to produce geranylacetone [68]. Geranylacetone is a volatile organic compound responsible for leafy, floral, and fruity aromas, and is widely used in foods, beverages, and cosmetics, with antioxidant activity [69, 70]. Therefore, we postulated that one of the potential metabolites of phytoene produced through BCO2 cleavage at the 9,10 or 9’,10’ double bonds is geranylacetone. However, we could not detect geranylacetone in the liver or serum of WT mice supplemented with phytoene, possibly due to its volatility and rapid metabolism. Further studies are needed to identify specific phytoene‐derived cleavage metabolites and elucidate their biological functions. Once the biological activity of phytoene metabolites is fully elucidated, phytoene may have potential as a therapeutic agent for established MASLD.

SIRT1 acts as a master metabolic regulator that mitigates MASLD [71, 72]. In the present study, phytoene supplementation significantly increased hepatic SIRT1 mRNA and protein expression in WT mice compared with HRCD‐fed controls, suggesting that prevention of HRCD‐induced downregulation of SIRT1 by phytoene represents a mechanism underlying its protective effects against MASLD development. Our findings may provide a potential mechanistic explanation for the previously reported increase in median lifespan in C. elegans following phytoene treatment, as overexpression of the sir‐2.1 gene (the C. elegans ortholog of mammalian SIRT1) has been shown to extend lifespan in this organism [29]. It will be interesting to investigate if the biological activity of phytoene is mediated by cleavage metabolites of phytoene in C. elegans model. Interestingly, phytoene supplementation did not alter de novo lipogenesis‐related markers, including Fasn, Scd1, and Dgat2, whereas multiple fatty acid oxidation‐related markers were significantly upregulated. Consistent with these findings, both mRNA and protein expression of PGC1α and PPARα, key downstream targets of SIRT1 that drive mitochondrial fatty acid oxidation [73], were significantly increased in phytoene‐supplemented mice compared with HRCD‐fed controls. These results suggest that phytoene stimulates fatty acid oxidation rather than suppressing lipogenesis. SIRT1 and AMPK interact to regulate lipid metabolism, and their concurrent activation has been identified as a key mechanism for alleviating MASLD [74, 75]. We previously demonstrated that ablation of SIRT1 deacetylase activity decreases the phosphorylation of AMPK in the liver [38], whereas tomato powder supplementation increases AMPK phosphorylation, which attenuates high‐fat diet‐induced hepatic steatosis [36]. In the present study, phytoene supplementation significantly increased both hepatic AMPK phosphorylation at Thr¹⁷² and ACC1 phosphorylation at Ser⁷⁹ in WT mice. Phosphorylation of AMPK at Thr¹⁷² is required for its activation, and activated AMPK subsequently phosphorylates ACC1 at Ser⁷⁹, the main regulatory site responsible for inhibiting ACC1 enzyme activity [76]. This regulatory cascade reduces malonyl‐CoA levels, promotes CPT1a expression, and enhances fatty acid oxidation [75]. Consistent with this mechanism, phytoene increased the expression of fatty acid oxidation markers without changing lipogenic gene expression, supporting a role for the AMPK/ACC pathway in mediating its hepatic benefits. However, because we did not observe the same effects of phytoene in the DKO mice, despite significantly greater phytoene accumulation in the liver than WT mice, we believe that these biological activities of phytoene are mediated by cleavage metabolites of phytoene. Additionally, AMPK activity is regulated not only by activating phosphorylation at Thr¹⁷² but also by inhibitory phosphorylation at Ser^485/491 [77]. Phosphorylation at Ser^485/491 inhibits phosphorylating the critical Thr¹⁷² activation site on AMPK and suppresses AMPK activation, which has been associated with increased lipogenesis and reduced fatty acid oxidation under conditions of nutrient excess, including HRCD. Future studies investigating whether phytoene modulates inhibitory Ser^485/491 phosphorylation, which is also mediated through insulin, IGF‐1, Akt, and mTOR‐signaling pathways, would provide a more comprehensive understanding of its effects on AMPK signaling pathway. Because fatty liver disease is primarily characterized by the accumulation of excess triglycerides within hepatocytes, which constitute approximately 60%–80% of the liver mass and serve as central regulators of lipid homeostasis, we believe that hepatocytes, which also highly express both BCO1 and BCO2, are likely the primary target of phytoene's effects. Future studies are needed to determine whether phytoene also targets other liver cell types using isolated liver cell populations. Additionally, we previously demonstrated that ablation of SIRT1 activity accelerates MASLD progression by increasing mobilization of free fatty acids from mesenteric adipose tissue into the circulation [38]. In this study, phytoene supplementation reduced mesenteric adipose tissue weight; however, expression of key lipolysis‐related genes (Atgl, Hsl, and Mgl) was unchanged, suggesting that the protective effects of phytoene are mediated predominantly within the liver. The mechanism underlying the reduction in mesenteric adiposity, particularly in females, remains to be determined.

Gut microbiome plays an important role in MASLD pathogenesis and progression [78], and the interactions of carotenoids with gut microbiota in the colon and their potential associated health benefits have been highlighted [79]. We have previously demonstrated that tomato powder feeding increased microbial richness and diversity and changed gut microbial composition in high‐fat diet‐induced MASLD model [35, 36], and recently, in HRCD‐induced MASLD model (Lee NY et al., unpublished data). In contrast, in the present study, phytoene supplementation did not significantly alter alpha or beta diversity. These differences are likely due to multiple bioactive compounds found in whole tomato compared to a single purified phytoene. Tomato powder contains diverse bioactive components, including other carotenoids, vitamins, polyphenols, and dietary fiber, which may collectively contribute to modulation of the gut microbiota. Undigested carotenoids and fermentable substrates from tomato fiber may contribute to the enhancement of microbial diversity [79]. However, in the present study, dietary supplementation with cis‐phytoene resulted in markedly higher serum and tissue concentrations. We have previously shown that hepatic accumulation of phytoene was approximately 10‐fold higher than that of lycopene in mice fed tomato extract, despite the lower concentration of phytoene than lycopene in the extract, further supporting the superior bioavailability of phytoene [8]. These findings further support the notion that phytoene is efficiently absorbed in the upper gastrointestinal tract, which may limit its exposure to the colonic microbiota. Therefore, phytoene alone is more likely to influence the regulation of hepatic metabolism rather than serve as a substrate for microbial fermentation to modulate the gut microbiome. On the other hand, our results highlight the importance of considering whole‐food approaches, such as consuming tomatoes and tomato products, which provide a complex mixture of carotenoids and other bioactive compounds that may act synergistically to modulate the gut microbiome, maintain intestinal mucosal integrity, regulate the gut‐liver axis, and attenuate the progression of MASLD to more advanced liver diseases, including metabolic dysfunction‐associated steatohepatitis and cirrhosis.

In conclusion, this study demonstrates that the colorless carotenoid phytoene functions as a dietary bioactive compound that protects against HRCD‐induced MASLD by enhancing hepatic fatty acid oxidation through activation of the SIRT1‐PGC1α‐PPARα signaling axis and the AMPK/ACC pathway. These protective effects appear to depend on enzymatic cleavage by BCO1 and BCO2, rather than by intact phytoene itself, providing novel insight into the role of phytoene metabolism and genetic variability in determining dietary phytoene efficacy. These findings may serve as a foundation for future clinical investigations evaluating the efficacy and appropriate dosing of phytoene‐rich foods in humans before population‐level dietary recommendations can be made. Given that phytoene is an upstream precursor of major carotenoids and is abundant in commonly consumed fruits and vegetables, and that many “white‐creamy” or “golden‐orange” tomatoes contain significantly higher levels of phytoene than red tomatoes, consumption of phytoene‐rich tomato cultivars may represent a promising nutritional strategy against MASLD.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting File: mnfr70596‐sup‐0001‐tableS1.xlsx.

Data Availability Statement

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