Chronic Consumption of Ghee (Clarified Butter) and Coconut Oil, but Not Palmolein, Induces Microvesicular Steatosis, Insulin Resistance, and Inflammation in Mice Fed a High-Fat Diet

1. Introduction

Although metabolic dysfunction-associated steatotic liver disease (MASLD) is multifactorial, nutritional factors, including excess calorie intake contribute to its development [1]. Excessive intrahepatic triglycerides , a hallmark of MASLD, is closely associated with obesity, insulin resistance, and abnormal lipid metabolism [2]. Dietary fat, both the amount and composition, may play a pivotal role in the development of MASLD since the quality of dietary fat appears to influence intrahepatic triglyceride (TG) accumulation [3]. Evidence suggests that a higher saturated fat intake leads to greater visceral and hepatic fat accumulation than unsaturated fats [4]. Increased SFA intake was found to be more metabolically harmful for the human liver than simple sugar or unsaturated fat due to its greater induction of intrahepatic TG content, insulin resistance, and harmful ceramides [5].

Dietary SFAs showed a greater stimulatory effect on weight gain and hepatic lipid accumulation than unsaturated fats due to the overflow of fat to the distal intestine which induced changes in the gut microbiota and host metabolism [6]. Evaluation of twelve commonly consumed dietary fats in mice showed that fat source differentially influenced body weight, circulating triglycerides, and gut microbiota composition, indicating that different dietary fats elicit distinct metabolic effects, in part through their differential impact on the composition and functional activity of the host gut microbiota [7]. Moreover, dietary fatty acid composition interacts with the gut microbiota to influence hepatic steatosis and host metabolism [8]. Furthermore, studies comparing different high-fat diets have reported variable effects on metabolic outcomes. Mice fed high-fat diets containing lard, soybean oil, and milk fat for 12 weeks exhibited comparable weight gain and hepatic steatosis. In contrast, milk fat increased serum cholesterol and advanced glycation end products more than lard or soybean oil [9]. Similarly, feeding HFD containing lard, palmolein, or interesterified palmolein resulted in increased body weight gain, fasting glucose, HOMA-IR, and hepatic steatosis in all high-fat diet groups. Still, the magnitude of these effects differed by fat type [10].

Different types of oils with distinct fatty acid profiles might have varied metabolic effects when consumed over a longer period. Palmolein, ghee, and coconut oil are commonly used dietary fat sources that differ markedly in their SFA content and fatty acid composition. Short-term studies have reported the metabolic effects of peanut [11], palm oil [12,13], ghee [7,14,15], and coconut oil [16,17] on hepatic lipid deposition and associated factors. In human subjects, consuming 30 grams of ghee per day, which makes up approximately 15% of total daily energy intake over four weeks, resulted in increased levels of apolipoprotein B and non-HDL cholesterol compared to olive oil [15]. These changes in the lipid profile, underscoring the rationale of investigating the impact of ghee intake on hepatic fat homeostasis and the development of MASLD.

Ghee (clarified butter), typically an anhydrous milk fat, when consumed as a sole source of fat at 10% of the diet, altered blood lipid profiles in rats [18]. Dietary ghee supplementation (2.5 or 5.0 wt.% ) for 8 weeks lowered serum cholesterol in rats primarily by increasing biliary excretion of cholesterol and related bile constituents, rather than by inhibiting hepatic cholesterol synthesis [14]. Feeding a 10% ghee diet for four weeks did not affect serum total cholesterol levels but increased serum triglyceride levels in Fischer rats [19]. A systematic review of animal studies suggests that the effects of ghee on the lipid profile are largely influenced by both the level of dietary intake (5%, 10%, and 20% of total energy intake) and the duration (28, 60, 90, 120, and 270 days) of intervention [20]. A diet containing coconut oil (7%) for 12 weeks produced hepatic steatosis in wild and PPARα KO mice by lowering insulin sensitivity, gut microbial diversity, and mitochondrial function and fatty acid oxidation [17]. Palm oil in HFD (60%) in mice induced obesity, and disrupted glycemic control after 19 weeks [21]. A long-term palmolein oil-based diet (20%) induced obesity and altered hypothalamic insulin sensitivity and redox activity in Wistar rats [22]. However, the effects of chronic consumption of such diets on the development of fatty liver are limited. Palmolein-based diet for eight weeks in mice fed a high-fat diet (45% kcal from fat) promoted hepatic triglyceride accumulation [23]. However, the long-term impact of such a moderate high-fat dietary regimen of palmolein diet on hepatic steatosis are limited.

Despite available data, the long-term relative metabolic effects of commonly consumed SFA-enriched dietary fats/oils, which mimic daily intake, remain limited. Moreover, comparative effects of palmolein, ghee, and coconut oil on hepatic steatosis relative to other fats are not reported. Our previous data suggest that substitution of fatty acid type (linoleic acid with α-linolenic acid) could prevent MASLD in rats [24]. Moreover, in our recent study, supplementation with bioactive compounds from dietary rice bran oil protected mice against diet-induced MASLD [25]. Since fatty acid turnover is slow, longer-term comparative measurements of increasing levels of SFA-enriched dietary oils and fats can provide early metabolic insights into their contribution to the development of MASLD. Therefore, the study evaluated the effects of dietary fats such as peanut, palmolein, ghee, and coconut oil, each incorporated at 25% of a high-fat diet for 6 months, on hepatic fat deposition and relevant metabolic parameters in C57BL/6 mice. The combination of fat/oil chosen in this study mimics the dietary practice of commonly used sources. The specific oils studied were the sole source of fat in the diet, and total SFA was the only variable across the test diets. Here is the first report of a long-term in vivo metabolic trial assessing the effects of ghee intake on MASLD risk.

2. Materials and Methods

2.1. Diet

Composition of the chow diet was prepared as per AIN-93M recommendations and presented in Table 1.

Peanut oil, buffalo ghee, and palmolein oil were procured from the local market, while cold-pressed coconut oil was obtained from local manufacturers. Fatty acid composition of the dietary fat sources is presented in Table S1. HF diets were prepared by thoroughly mixing the respective oils/fat with chow diet (75% of the chow diet without oil and 25% respective oils). The chow diet provided 303 kcal/100 g diet, with 25% of calories derived from fat, whereas HF diets provided 428 kcal/100 g diet, with 58.4% of calories derived from oil/fat and all these high fat diets were isocaloric. The contents of total SFA, MUFA, and PUFA in the dietary oils and fat were analyzed by gas chromatography as described later. Fatty acid composition of the experimental diets is presented in Table 2.

2.2. Animal Experiment

Three-month-old male C57BL/6 mice (n= 40) were obtained from the animal house facility of ICMR-NIN. Mice were acclimatized on a chow diet (AIN-93M) for 7 days and randomly divided in to five groups (n=8 per group). Mice fed with chow diet was used as a control group (5% peanut oil) and HF diet (all 25% w/w) enriched with peanut, palmolein, ghee, and coconut oil. Mice were housed individually in wire mesh bottomed polypropylene cages under controlled environmental conditions (22 ± 2 °C, 12-h light/12-h dark cycle) with free access to food and water. Daily food intake was recorded, and body weights were measured once every ten days. All animals were sacrificed at the end of the six-month.

2.3. Blood Collection and Biochemical Assays in Plasma

At the end of the metabolic trial, mice were fasted for 6 h prior to blood collection. Blood was collected from retro orbital plexus, and stored in EDTA-coated vials. Plasma was separated and stored (-80 °C) for subsequent analysis. Plasma glucose (Biosystems, Barcelona, Spain) and alanine aminotransferase (ALT) activity (Span Diagnostics Ltd., India) were estimated on the same day using commercial kit methods as per their instructions. Plasma insulin levels were measured using an ELISA kit (Mercodia AB, Sweden). Insulin resistance was assessed using the homeostasis model assessment of insulin resistance (HOMA-IR) index, calculated as: HOMA-IR = Fasting glucose (mmol/L) × Fasting insulin (mU/L) / 22.5. Plasma leptin levels were measured using an ELISA kit (Elabscience, USA). Total cholesterol (TC), triglycerides (TG), and high-density lipoprotein cholesterol (HDL-C) were estimated in plasma using commercial kits (Biosystems, Spain).

2.4. Termination, Organ, and Tissue Collection

After blood collection, mice were euthanized by CO₂ asphyxiation. Liver, retroperitoneal (RP) fat, and epididymal (EP) fat depots were excised, rinsed with saline, weighed, snap-frozen in liquid nitrogen, and stored at -80 °C for further analyses. A portion of liver tissue was preserved in RNA later (Sigma-Aldrich, India) for transcriptome analysis, while another portion was fixed in 10% neutral buffered formalin for histological examination. The adiposity index was calculated as sum of epididymal, retroperitoneal and mesenteric adipose tissues (g)/body weight (g)*100 [26].

2.5. Determination of Triglycerides and Total Cholesterol in the Liver

Briefly, 50-100 mg of liver tissue was homogenized in 5% aqueous Nonidet P-40 (NP-40; Sigma, India) solution. The homogenate was gradually heated to 80 °C in a water bath until the solution became cloudy, cooled to room temperature, and reheated once more to ensure complete lipid solubilization, and centrifuged at 3000 rpm, 20 min as described before [24]. The triglycerides and total cholesterol were measured in the supernatant using commercial assay kits as per the supplier’s instruction (Biosystems, Barcelona, Spain).

2.6. Fatty Acid Compositional Analysis of the Diets

Fatty acid extraction and analysis was done as described previously [27]. In brief, dietary lipid extraction was performed by adding 250 µg of heptadecanoic acid (C17; mg/mL) as an internal standard and 20 mL of chloroform: methanol (2:1, v/v) containing butylated hydroxytoluene (BHT; 10 mg/L) to 200 mg of diet sample. Samples were mixed thoroughly, flushed with nitrogen, and incubated overnight. After filtration, 3 mL of water was added to the filtrate, vortexed, and allowed to stand for 30 min. The lower chloroform layer was collected. The aqueous phase was reextracted twice with chloroform: methanol (2:1, v/v), and the pooled chloroform extracts were evaporated to dryness under nitrogen. The dried lipid extract was saponified followed by acidification and methylation. The final extract was stored at -20 °C until fatty acid analysis by gas chromatography (PerkinElmer Clarus 680).

2.7. Estimation of Hepatic Lipids by Fatty Acid Composition

Total lipids from liver tissue were extracted using the Folch method. A 10% liver homogenate (500 µL) was mixed with 5 mL of chloroform: methanol (2:1, v/v), vortexed vigorously, filtered, and reextracted with an additional 3 mL of solvent. The pooled extract was washed with one fifth volume of 0.22% KCl and centrifuged at 2000 rpm for 2 min. The lower organic phase was collected and evaporated under nitrogen. The dried extract was spiked with a known quantity of heptadecanoic acid (C17) internal standard and methylated using 2% sulfuric acid in methanol containing BHT (10 mg/L) at 70 °C for 4 h. After cooling, fatty acid methyl esters were extracted with petroleum ether, washed with 0.2 M sodium bicarbonate, dried over anhydrous sodium sulfate, and stored at -20 °C until analysis. Fatty acid composition was analyzed using gas chromatography (PerkinElmer Clarus 680) equipped with a flame ionization detector and an SP-2300 fused silica capillary column (30 m × 0.25 mm × 0.2 μm; Supelco, USA). Fatty acids were identified by comparison with authentic standards (Nuchek, USA) based on retention times.

2.8. Histology and MASLD Activity Score

Formalin fixed liver tissues were embedded in paraffin blocks, sectioned at 5 µm thickness, and mounted on glass slides. Sections were stained with hematoxylin and eosin (H&E) and examined under a light microscope for hepatic steatosis and lobular inflammation. The severity of MASLD was assessed using the MASLD activity score (MAS) as described before [28]. MAS was calculated based on steatosis (0–3), lobular inflammation (0–3), and hepatocyte ballooning (0–2) in a blinded manner. A MAS score > 4 was considered indicative of NASH.

2.9. Lipid Peroxidation and Antioxidant Enzymes in the Liver

Lipid peroxidation levels in liver were estimated by measuring malondialdehyde (MDA) using the thiobarbituric acid reactive substances (TBARS) assay [29]. Catalase, glutathione peroxidase, reduced glutathione activity in the liver supernatant was measured using assays as mentioned before [24].

2.10. Gene Expression Analysis by Real-Time qPCR

Total RNA was isolated from liver tissue using TRI reagent (Cat#T9424, Sigma) according to the manufacturer’s protocol. Briefly, 50 mg of liver tissue was homogenized in TRI reagent (Cat#T9424, Sigma) and homogenized using 2-mm zirconia beads in a bead beater (Mini Bead Beater, BioSpec). Phase separation was achieved by the addition of chloroform, and RNA was precipitated with isopropanol, followed by dissolution in nuclease-free water as described previously [30]. Total RNA was purified using DNase 1 (Cat#AMPD1, Sigma) to remove any traces of contaminating DNA. The quality and quantity of total RNA were assessed by a spectrophotometer (Multiskan Skyhigh, Thermo Scientific). cDNA was synthesized using a Bio-Rad iScript cDNA synthesis kit (Cat#1708891, Bio-Rad). Real time-qPCR was performed in Bio-Rad CFX 96 system (Bio-Rad, USA) using KiCqStart SYBR green primers (Table S2) and TB green pre-mix Ex-Taq II (Cat#RR208, Takara). The relative mRNA expression was quantified by the ddCt method, and β-actin was used as an endogenous reference gene. Values are obtained from repeated independent experiments from n=6 per experimental group.

2.11. Immunoblotting

The protein expression of hepatic pro-inflammatory marker interleukin-6 (IL-6) was analyzed by Western blotting. Liver tissue (100 mg) was homogenized in RIPA buffer (pH 7.5) containing protease inhibitors. The tissue was finely homogenized in liquid nitrogen, centrifuged at 12000g, 4ºC for 20 min as described previously. In brief, the blot membrane was blocked and incubated overnight at 4ºC with primary antibodies of anti-IL-6 (#P620, Thermo Fischer) with a dilution of 1:2500, and incubated with a HRP conjugated secondary antibody. The immunoblots was detected by ECL substrate (#1705061, Bio-Rad), and captured their signal in gel imaging system (G-Box, Syngene). The band intensity signal was calculated using Image J 1.50i (NIH, USA) and analyzed statistically as mentioned in legend.

2.12. Statistical Analysis

All data are expressed as mean ± standard error of the mean (SEM). A statistical difference between groups were analyzed using one-way analysis of variance (ANOVA) with least significant difference (LSD) post-hoc test using SPSS v21. A p-value <0.05 was considered statistically significant.

3. Results

3.1. Fatty Acid Composition of Experimental Diets

As shown in Table 2, the fatty acid composition of the experimental diets varied in the range of SFA content across groups. Total SFA content was lowest in the chow diet (1.5%), with incremental increases in peanut (5.8%), palmolein (10.82%), ghee (15.84%), and coconut (21.59%) enriched HF diets. Total monounsaturated fatty acid (MUFA) content also followed an incremental trend across the groups, except that ghee and coconut showed closely similar MUFA content. Total polyunsaturated fatty acid (PUFA) content was lowest in the ghee and coconut enrich HF diet. The n-6/n-3 ratio was lowest in the ghee group compared to other groups, whereas peanut showed the highest ratio (Table 2).

3.2. Food Intake, Body Weight, Organ Weights and Feed Efficiency

Food intake of mice was captured for 16 consecutive weeks after feeding diets enriched with different edible oils and fats. The food intake values are expressed in calories (Figure 1A- B).

Food intake observed each week differs across the groups and the number of time points (weeks) with significantly increased food intake also differed across groups compared with the chow diet. For example, food intake was significantly increased (p<0.05) at 8-, 7-, 5-, and 4-time points for the peanut, palmolein, coconut, and ghee groups, respectively, compared to the chow diet. Ghee fed mice are the only group that showed a significant reduction (p<0.05) in food intake at 2- time points than chow diet (Figure 1B). Again, compared to the ghee diet, calorie intake was significantly (p<0.05) lower at 5 time points, indicating that chow-fed mice have the lowest calorie intake among these experimental diets (Figure 1B). At the end of 16-week feeding, average calorie intake was significantly increased (p<0.05) in peanut, palmolein, and coconut diets as compared to the chow diet, while it remained insignificant (p>0.05) with ghee-fed mice (Figure 1B). These data indicate preferential food intake in the mice in the order of peanut > palmolein > coconut> ghee as compared to chow diet where ghee fed mice showed a satiety effects compared to the rest.

Concomitant with food intake, the body weight of mice was also monitored every 10 days and considered as individual data points (Figure 2A-B). The measured body weight of the mice differs across the edible oil-fed diets at each data point. For example, mice body weight was significantly increased (p<0.05) at 9-, 7-, 5-, 4 - time points for ghee, coconut, peanut, and palmolein groups respectively, as compared to chow diet (Figure 2B).

Although initial body weight remained insignificant across the groups (Table 3), at the end of the 6 months feeding with different dietary oils/fats, both final body weight and weight gain were significantly increased (p<0.05) in ghee and coconut group as compared to chow fed mice, while the bodyweight changes remained insignificant in peanut group. Although final body weight did not differ significantly, a significant weight gain (p<0.05) was also observed in the palmolein group as compared to chow-fed mice. Feed efficiency ratio (FER) was consistently elevated (p<0.05) in all groups as compared to chow-fed mice. However, the FER of ghee-fed mice was the highest (Table 3) among all groups, indicating that these mice convert food into body mass more efficiently than the other groups, which may be due to increased body and organ weights. For example, mice fed ghee exhibited a significant and exclusive increase in weight (p<0.05) of metabolically sensitive organs, such as retroperitoneal and epididymal fat, kidney, and adiposity index, compared to the other dietary groups. Normalized liver weight was significantly elevated (p<0.05) in all dietary groups compared with chow-fed mice, with the highest value in the ghee group (Table 3). In addition to the liver, the kidney weight was marginally increased in coconut-fed mice compared with other groups.

3.3. Long-Term Intake of Saturated Fatty Acids-Enriched Diets and Their Effects on Metabolic Parameters Detected in Plasma

Except for peanut, fasting plasma glucose was significantly elevated (p<0.05) in all the groups after 6 months of feeding with different SFA-enriched dietary oils, as compared to chow-fed mice (Table 4). Ghee-fed mice showed insulin resistance after 6 months, as indicated by a significant insulin spike and elevated HOMA-IR values (p<0.05), whereas peanut- and palmolein-fed mice remained insulin sensitive. Although insulin levels remained insignificant, HOMA-IR was also significantly increased in coconut oil-fed mice compared to the chow group. Plasma cholesterol was significantly elevated across the HF-fed groups compared with chow-fed mice (p<0.05). However, triglyceride and leptin levels were significantly increased (p<0.05) only in ghee-fed mice among the groups (Table 4), indicating that most altered metabolic parameters in plasma were detected in ghee-fed mice.

3.4. Hepatic Fatty Acid Composition Was Differentially Altered After Feeding Edible Oil-Enriched Diets for Six Months

After feeding HF diets enriched with different edible oils for 6 months, liver fatty acid composition was altered differently across groups compared with the chow diet (Table 5). For example, mice fed a peanut diet exhibited the lowest levels of SFAs in the liver (p<0.05) compared with the other groups. In contrast, hepatic SFA levels were significantly higher (p<0.05) in the ghee and coconut groups compared with the others. Except for the coconut group, total MUFA levels were significantly elevated (p<0.05) in all groups compared with the chow diet. Among all, hepatic n-6 PUFA levels were highest (p<0.05), and n-3 PUFA levels were lowest (p<0.05) in the peanut group. Moreover, the liver of peanut-fed mice exhibited the highest n-6/n-3 ratio of 15:1, while ghee showed the least n-6/n-3 ratio (Table 5).

3.5. Dietary Intake of Ghee and Coconut Oil Significantly Increased Liver Triglyceride Levels, Fatty Liver Score and IL-6 Expression

The morphology of the livers appeared regular in color with an apparent smooth texture in all the groups (Figure 3A). However, hepatic triglyceride content was significantly elevated in ghee and coconut oil-fed mice (Figure 3B), although hepatic total cholesterol did not differ significantly among the groups (Figure 3C). Histological examination of liver sections stained with H & E revealed a regular hepatic architecture in chow, peanut, and palmolein-fed mice (Figure 3D i, ii, iii). In contrast, mice fed ghee (Figure 3D iv) and coconut oil (Figure 3D v) exhibited hepatocytes with microvacuoles, focal collections of lymphocytes in the parenchyma, and a moderate number of hepatocytes ballooning, indicating mild microvesicular steatosis in ghee- and coconut-fed mice. MASLD activity score (MAS) assessment showed a significant increase of approximately 1.3, indicating the occurrence of simple steatosis in ghee- and coconut oil-fed mice, but did not develop MASH (Figure 3E). Furthermore, IL6 expression was significantly upregulated (p<0.05) in the mice livers fed with ghee and coconut oil (Figure 3F), while it remained insignificant in the peanut and palmolein groups, indicating the absence of liver inflammation in the latter groups. Again, hepatic malondialdehyde (MDA) level was significantly increased (p<0.05) only in ghee group compared to others [Chow vs peanut, palmolein, ghee, coconut (nmoles /mg protein): 1.5 ± 0.14 vs 1.13 ± 0.04, 1.26 ± 0.08, 2.3 ± 0.15, 1.57 ± 0.17 (n=8 per group)] indicating oxidative stress in the ghee-fed liver, while no significant changes were observed in glutathione level, catalase and glutathione peroxidase activity among these groups (data not shown).

3.6. Expression of Fatty Acid Oxidative and Lipolytic Genes Was Downregulated in Almost All Edible Oils-Enriched Diets After Feeding for Six Months

Fatty acid oxidation genes such as CPT-1α, PGC-1α and LPL mRNA expression were significantly decreased (p<0.05) in the mice fed with diets containing palmolein, ghee, and coconut oils (Table 6). Peanut-fed mice also showed a downregulated expression of CPT-1α and LPL mRNA (p<0.05), indicating downregulated beta-oxidation in the liver after six months of feeding dietary oils as compared to the chow diet. Mice fed with a diet containing palmolein showed decreased mRNA expression of PPAR-α (p<0.05), whereas in other experimental groups, PPAR-α mRNA remained unaltered. The expression of PPAR-γ remained insignificant among the experimental groups. The expression of antioxidant genes, such as catalase (CAT) mRNA levels, was significantly downregulated (p<0.05) in the liver of mice fed with all the dietary oils as compared to the chow diet. Glutathione peroxidase 4 (GPX4) and glutathione reductase (GSR) mRNA levels were significantly downregulated in the liver of mice fed with palmolein and ghee-fed mice compared to those of others.

4. Discussion

The present study is the first report to demonstrate that a ghee-enriched diet (25% w/w) for six months can develop mild steatosis and inflammation in the liver by triggering visceral adiposity, insulin resistance, and dyslipidemia in adult mice. Unlike ghee-fed mice, coconut oil-fed mice exhibited hepatic steatosis without increased visceral adiposity, suggesting that insulin resistance may have developed through mechanisms independent of adipose tissue expansion. In a similar setup, both peanut- and palmolein-fed mice were protected from developing steatosis, as reflected by lower TG accumulation, lower MAS score, and reduced IL6 expression in the liver, compared with the ghee and coconut oil diet. Although the palmolein diet contains a balanced proportion of saturated and unsaturated fatty acids, when supplied as a 25% fat source, it showed a mild increase in fasting glucose and body weight gain compared to the peanut oil diet. The peanut oil diet exhibited a neutral effect, with no significant changes in fasting glucose, insulin sensitivity, body weight gain, adiposity, or TG levels, even though the liver showed a higher n-6/n-3 fat ratio (15:1) compared to the others. Collectively, these findings indicate that fat type, its composition, dietary level, and their metabolic fate can contribute to the development of hepatic steatosis and inflammation in the liver.

In summary, a preferential food intake was observed after six months of feeding with different oils/fats, in the order of peanut > palmolein > coconut > ghee diets, compared with the chow diet. Only ghee-fed mice showed the highest leptin levels and reduced food intake at two time points, indicating a satiety effect. In contrast, body weight gain was increased in the order of ghee > coconut > palmolein, while it remained insignificant in peanut oil-fed mice compared with chow-fed mice, indicating a mismatch between food intake and body weight gain. Feed efficiency ratio (FER) increased in the order of ghee > coconut > palmolein> peanut, indicating that ghee-fed mice absorbed and utilized the diet more efficiently to build organ mass and body mass, as observed by the highest normalized liver weight, adipose fat, kidney, and body weight gain compared to the rest. Except for peanut oil, plasma glucose was elevated in all the groups compared to chow-fed mice. Ghee-fed mice exhibited elevated TG and insulin resistance, as indicated by a spike in insulin and HOMA-IR values, whereas peanut- and palmolein-fed mice remained insulin-insensitive. A decrease in CPT-1α, PGC-1α, and LPL mRNA expression in the palmolein, ghee-, and coconut-fed livers suggests a potential reduction in fatty acid oxidation and impaired lipid utilization in these mice.

FER was highest in ghee compared with other test groups, indicating that these mice convert food into body mass more efficiently than other groups. A previous study also showed that mice fed milk fat exhibited higher feed efficiency, body weight gain, and adipose tissue content than those fed other animal fat sources [31]. A higher FER may indicate more of the consumed calories/nutrients are absorbed and converted into tissue or energy storage, improved digestive/absorptive efficiency as the gut may extract nutrients more effectively or may be burning fewer calories for maintenance, greater fat accumulation, reduced metabolic rate, insulin-related metabolic alterations, or an “energy-saving” phenotype.

Despite the lowest food intake observed in ghee-fed mice, body weight gain was the highest among the SFA-fed test groups, indicating that body weight gain is influenced more by the source of dietary fat than by the proportion of SFA in the diet, as shown previously [7]. Our findings are consistent with a previous report demonstrating that, despite reduced or unchanged food intake, an increase in body weight depends on the fat source [7]. Consistent with our observations, a previous study demonstrated that C57BL/6 mice fed with milk fat (15% w/w) for five weeks exhibited higher feed efficiency, body weight gain, and adipose tissue expansion compared with mice receiving cow tallow, buffalo tallow, mutton tallow, or chicken fat. These results suggest that milk fat is utilized more efficiently by the body than the other fat sources [31].

Ghee, a typically anhydrous milk fat, contains triglycerides with palmitic acid partly at the sn-2 position, and during digestion pancreatic lipase selectively hydrolyzes the sn-1 and sn-3 fatty acids, producing sn-2 monoacylglycerol that is efficiently absorbed and re-esterified [32]. However, the impact of sn-2 palmitate-derived absorption on lipid transport and hepatic triglyceride accumulation in this study remains unclear. The overall fatty acid profile of the ghee diet (Table 2) shows a higher amount of total SFAs (15.84%) than the peanut and palmolein diet, with a similar proportion of total MUFAs (24.93%) with coconut oil, and the lowest n-6/n-3 ratio (~10 :1) among the diets. Although we did not measure it, data suggest that bovine milk fat like ghee contains a relatively high proportion of palmitic acid at the sn-2 position [33,34] because of the way milk fat is biosynthesized in the mammary gland of ruminants. Acyltransferase involved in milk fat synthesis preferentially incorporates short and medium chain fatty acids at sn-3, oleic acid at sn-1, and palmitic acid (C16:0) is frequently esterified at the sn-2 position of the glycerol moiety. This positional distribution critically determines the metabolic fate since fatty acids at sn-2 position are predominantly retained as sn-2 monoacylglycerol (MAG) during digestion whereas pancreatic lipase hydrolyses sn-1 and sn-3 fatty acids. Evidence showed that MAG carrying palmitate in the sn-2 position is absorbed efficiently in the intestine and transported faster than native FFAs [35], which could be the reason for the highest FER in ghee-fed mice (Table 3) among all groups in this study. The excess sn-2 MAG in the ghee diet might have been directly incorporated into chylomicron triglycerides, thereby increasing lipid accumulation in the liver. The overload of intermediary mediators (free fatty acids) might have served as precursors of reactive oxygen species for lipid peroxidation, as evidenced by elevated MDA levels, which might have triggered inflammation in the liver, as reflected by increased IL-6 expression in this study. Thus, compared to eucaloric peanut- and palmolein-fed mice, ghee-fed mice might have received a different quality of fats that contain a greater proportion of SFAs acylated at the sn-2 position than the vegetable oils used in this study, which predominantly contain unsaturated fatty acids in the same position.

Despite no increase in the mean caloric intake of ghee-fed mice (Figure 1), they were not protected from the accumulation of serum and hepatic TGs, which accompanied steatosis in the liver. Similar to a previous study in Fischer inbred rats fed with a 10% ghee diet for four weeks [19], the current study with a 25% ghee diet also reported increase in serum TGs. A previous study showed that a higher-saturated-fat diet is more obesogenic than diets with lower saturated fat content [36]. However, in this study, a higher adiposity index was observed in lower SFA-enriched ghee-fed mice than in coconut (Table 2), which could be due to differences in fat sources (milk fat vs vegetable oil), species (SD rats vs C57 mice), duration (50 days vs 180 days), and composition of the diet between the study designs.

The metabolic inflammation and insulin resistance in ghee-fed mice could be due to both adipose tissue inflammation (AT) and hepatic inflammation, resulting from a dysfunctional adiposity index (DAI) and a steatosis phenotype [37]. Although this study did not measure AT inflammation, the DAI observed could be independently linked to steatosis [38]. Excessive calorie intake is an independent risk factor for developing fatty liver [39]. However, this study does not support the notion that MASLD was caused by changes in calorie intake, but rather by changes in the quality of fats consumed.

The greater weight gain and excess hepatic lipid accumulation observed with SFA-rich diets in this study may also be associated with increased availability of endogenous fatty acids and their metabolites to the gut microbiota, thereby altering microbial metabolism and host lipid homeostasis [40]. Dietary SFAs promote greater weight gain and hepatic lipid accumulation than unsaturated fats, potentially because their lower absorption efficiency in the proximal intestine allows more fat to reach the distal intestine, where it alters the gut microbiota in ways that promote inflammation, metabolic dysfunction, and increased lipid deposition in adipose tissue and the liver [6]. Previous studies have demonstrated that HFD containing different fat sources, including corn oil, peanut oil, soybean oil, sunflower oil, canola oil, and lard, elicit distinct alterations in gut microbiota composition [7]. In particular, diets richer in SFAs increased the abundance of Firmicutes and reduced the abundance of Bacteroidetes; for example, canola oil (lower SFA content) produced the smallest microbiota shifts, whereas lard (higher SFA content) induced the greatest changes. These findings suggest that the elevated hepatic triglyceride levels and lipid accumulation observed in SFA-enriched ghee- and coconut oil-fed mice may be mediated, at least in part, by metabolites produced by the altered gut microbiota.

Unlike ghee-diet, coconut oil diet promoted insulin resistance and lipid accumulation in the liver independent of visceral adiposity. Like ghee-fed mice, a HOMA-IR together with elevated glucose suggests insulin resistance, even if fasting insulin itself was not different. Since HOMA-IR incorporates both fasting glucose and insulin, the observed increase in HOMA-IR appears to be driven primarily by higher fasting glucose levels and reflects altered insulin sensitivity that was not evident from fasting insulin measurements alone. The development of insulin resistance in coconut oil-fed mice may be attributable to excess hepatic lipid accumulation. Because the sn-2 position of coconut oil triacylglycerols is enriched with lauric (C 12:0), and myristic acid (C 14:0) as reported [34], these SFAs accumulated to a greater extent in the liver of coconut oil-fed mice (Table 5). Unlike shorter medium-chain fatty acids (C8:0 and C10:0), a considerable proportion of lauric acid is re-esterified into triglycerides within enterocytes and incorporated into chylomicrons, allowing transport through the lymphatic circulation and delivery to the liver. Excess hepatic uptake of lauric acid might therefore have promoted triglyceride accumulation in the liver, leading to hepatic insulin resistance even in the absence of visceral fat expansion in this study. SFAs present in coconut oil, particularly lauric acid and other medium-chain saturated fatty acids, have been reported to induce inflammatory signaling, endoplasmic reticulum stress, and defects in insulin signaling pathways in metabolic tissues. Consequently, insulin resistance may develop before or independently of detectable visceral fat accumulation [16]. Thus, the observed increase in HOMA-IR despite unchanged visceral adiposity may reflect liver-specific metabolic dysfunction rather than obesity-driven insulin resistance.

In this study, palmolein-fed mice exhibited a modest increase in weight gain without insulin resistance, in contrast to several reports showing palm oil-induced insulin resistance and glucose intolerance at different energy levels [10,41,42]. Consistent with a study at a comparable fat level (20%) with oleic acid-rich palmolein in Wistar rats, our study also observed a mild increase in fasting glucose after 24 weeks. However, fasting insulin remained unchanged, in contrast to the earlier finding of hyperinsulinemia [22]. These findings suggest that while long term palmolein- enriched diet (25%) can promote weight gain and mild disturbances in glucose homeostasis, its effects on insulin regulation may vary depending on specific fatty acid composition of the oil used (44.3% oleic acid and 39.5% palmitic acid vs. 44.9% oleic acid and 35.87% palmitic acid), duration of exposure (21 vs 24 weeks), species (rat vs mice). Unlike the previous study, where palm oil in high-fat diets (41% fat) fed for 8 weeks in C57Bl/6J mice disrupted blood lipid profile, accumulated liver fats and hepatocyte ballooning [43], a six-month palmolein oil-enriched diet (25% fat) in this study did not induce hepatic steatosis and liver inflammation in similar mice, possibly due to the difference in the fat intake. Unlike this work, a palm oil-based diet with comparable fat (23.8%) showed hepatic steatosis in C57BL/6 mice after 8 weeks [10], which could be due to differences in the fatty acid composition of diets (palm oil vs palmolein), and the adapted metabolic response over time (8 vs 24 weeks).

A two-month crossover trial comparing palmolein and olive oil (48 g per day) showed no differences in body weight, blood lipids, glucose, or insulin in healthy adults [44]. Again, in a 16-week randomized controlled trial, diets enriched with palmolein, cocoa butter, or soybean oil had no differential effect on liver fat concentration when total dietary fat was maintained at 33% of energy [13], indicating inconsistency in the effect of palmolein intake on hepatic lipid accumulation. Irrespective of the composition of dietary oils, the metabolic impact of oils on liver fats might depend on the absolute energy of the diets derived from fat sources. For example, compared to the control diet (11% energy from fat), both high-fat diets rich in palm and olive oils (56% energy from fat) intake showed hepatic lipid accumulation in male Wistar rats after 12 weeks [42], indicating that olive oil in excess could have an equal adverse impact on liver fat metabolism as palm oil.

The protective effects of palmolein on hepatic steatosis may be partly attributed to its tocotrienol content, as tocotrienols are known to exhibit hypolipidemic [45], antioxidant, and anti-inflammatory properties [46]. However, the extent to which these effects contribute to the benefits of palmolein remains to be fully established, particularly in human studies. The protection observed in peanut-fed mice compared with coconut- and ghee-fed mice may also reflect the higher MUFA content of peanut oil, with preferential incorporation of MUFAs into hepatic triglycerides (Table 5). The liver might actively channel fatty acids into membrane phospholipids and triglycerides to preserve membrane fluidity, consistent with evidence that MUFAs are preferentially esterified into triglycerides and phospholipids and are less lipotoxic than SFAs [47]. Moreover, lipid-synthesizing enzymes, such as monoacyl- and diacylglycerol acyltransferases, prefer unsaturated acyl chains over saturated ones during triglyceride synthesis. The increase in MUFA levels in peanut- and palmolein-fed liver (Table 5) might have prevented the accumulation of hepatic triglyceride by stabilizing triglyceride storage. Since triglycerides containing MUFAs remain more fluid and are easier to package into lipid droplets [48], this may drive such physiological alterations.

In this study, plasma TC was elevated in all HF-fed groups (peanut, palmolein, ghee, and coconut), whereas TG was increased only in the ghee group, possibly due to distinct metabolic pathways regulating TC and TG. Ghee’s effects on cholesterol metabolism could not be attributed solely to cholesterol absorption from the intestine but may reflect differences that occur after the release of chylomicrons/VLDL from the intestine [15]. Plasma TG concentrations are determined largely by the balance between hepatic VLDL and intestinal chylomicron production and their subsequent clearance, particularly through LPL-mediated lipolysis and hepatic remnant uptake, rather than by TC [49]. All HF diets in this study may have altered hepatic lipid and lipoprotein metabolism, thereby increasing circulating cholesterol. At the same time, the specific fatty acid composition of each fat group might have determined whether triglyceride production or clearance was also substantially affected [50]. Ghee has a unique composition, with a high proportion of short- and medium-chain SFAs, as well as longer-chain SFAs like palmitic and stearic acids. This compositional difference can also increase TG specifically in the ghee group due to differences in fatty acid availability and clearance compared with other groups.

Again, liver TG levels were elevated in the ghee and coconut group, as supported by histological evidence of microvesicular steatosis, inflammatory mediators, and increased MAS scores. Liver fatty acid composition further demonstrated increased SFA content in these groups, supporting the role of SFA-rich diets in hepatic fat accumulation. Consistent with our findings, previous studies have shown increased liver TG and steatosis in ghee- and coconut oil-fed rodents [17,31]. In contrast, the palmolein diet did not increase hepatic lipid content in our study, aligning with reports indicating minimal hepatic fat accumulation at similar intake levels [42]. However, palmolein intake in HFD setup (41 to 45% kcal from fat) has been associated with hepatic steatosis and hepatocyte ballooning [23,43]. Ghee-fed mice also exhibited elevated plasma leptin levels and reduced food intake, supporting leptin’s role in appetite regulation [51]. In addition to increased adiposity, insulin resistance, and higher MASLD activity score (MAS) in the ghee and coconut groups, elevated malondialdehyde (MDA) levels in the liver indicate enhanced oxidative stress in these groups. Additionally, a relative increase in IL-6 expression further confirms that the liver was inflamed in addition to accumulating fat.

The present dietary trial confirms long-term changes in the metabolically sensitive organ, such as the liver, after 6 months of feeding with different SFA-enriched diets, as reflected by the metabolic trace of fatty acids (Table 2) and their proportionate incorporation into the liver (Table 5). For example, SFA-enriched coconut- and ghee-fed mice result in a higher incorporation of total SFAs in the liver as compared to other diets. Similarly, a relative proportion of MUFAs in the peanut and palmolein diet was also incorporated proportionately in the livers compared to the chow diet. Again, total n-6 PUFA levels were elevated in the peanut group and decreased in the palmolein-fed mice. Such dietary incorporation of fatty acids into the liver could be due to differences in dietary lipid content, preferential incorporation of unsaturated fatty acids, or increased desaturation of SFAs in the liver. Overall, the present dietary trial established a defined model for assessing the metabolic changes induced by dietary modification.

Although both phenotypic and biochemical indices clearly established metabolic disruption of liver homeostasis, the study is limited in explaining the mechanisms underlying coconut- and ghee-induced hepatic disruption at the molecular level. The study did not include other dietary oils, such as sunflower, rice bran, and soybean, which have SFA contents similar to those of the peanut and palmolein oils used in this study.

5. Conclusions

Despite the low food intake, the ghee diet (25% w/w) increased body weight gain, visceral adiposity, and insulin resistance, making it more susceptible to developing fatty liver than the other diets used in this study. Although the relative body weight gain was increased in all high-SFA-fed mice, the adiposity index was only increased in ghee-fed mice, indicating that increased visceral fat may predispose to the development of fatty liver. Among the sets of SFA containing oils chosen in this dietary intervention trial, ghee-, and coconut-oil enriched diets selectively induced mild microvesicular steatosis and inflammation in mice liver compared with the peanut and palmolein diet. However, further work is required to confirm whether recommended intake of fats (20-30% of total calories) for a similar duration yields similar effects in the liver. Collectively, these findings indicate that the source of dietary fat, degree of saturation, and fatty acid composition of the diets may play a crucial role in the development of metabolic risk factors such as obesity, impaired glucose metabolism, dyslipidemia, and hepatic steatosis.