Targeted proteomic analysis reveals that crocodile oil from Crocodylus siamensis may enhance hepatic energy metabolism in rats

Abstract

The liver is a key organ governing body energy metabolism. Dietary fats influence energy metabolism and mitochondrial functioning. Crocodile oil (CO) is rich in mono- and polyunsaturated fatty acids that contain natural anti-inflammatory and healing properties. Our study examined how CO affects the expressions of liver proteins involved in energy metabolism in rats. Twenty-one male Sprague Dawley rats were divided into three groups and underwent oral gavage with 3 ml/kg of sterile water (N group), CO (CO group), or palm oil (PO group) for 7 weeks. Body weight, energy intake, liver weight, liver indexes, blood lipid profiles, and liver-energy intermediates were measured. The liver proteome was analyzed using shotgun proteomics, and the functions and network interactions of several candidate proteins were predicted using the STITCH v.5.0 software. Body weights, energy intake, liver contents, and lipid profiles did not differ between the groups. However, hepatic oxaloacetate and malate levels were significantly higher in the CO group than in the PO group. Targeted proteomics reveals that 22 out of 1,790 unique proteins in the CO group were involved in energy-generating pathways, including the tricarboxylic acid cycle and oxidative phosphorylation (OXPHOS), and were correlated with the AMP-activated protein kinase signaling pathway. Cluster analysis of 59 differentially expressed proteins showed that OXPHOS-associated proteins were upregulated in the CO group and that three glycolytic metabolism-related proteins were downregulated in the CO group. CO may enhance hepatic energy metabolism by regulating the expressions of energy expenditure-related proteins.

Introduction

The liver is a major metabolic organ that methodically controls body energy metabolism and acts as a metabolic hub for various tissues, including skeletal muscle and adipose tissue. Hepatic energy metabolism is largely controlled at the genomic level by numerous transcription factors and co-regulators [1]. Metabolic hormones and dietary nutrition regulate protein expression activities, which dynamically regulate energy homeostatic processes. Dysregulation of these transcription factors, coregulators, and protein expressions in the liver contributes to defective lipid and energy metabolism and to the development of obesity [2,3,4]. A previous study showed that reduced hepatic energy-related metabolic activity was related to the progression of fatty liver disease [5], likely due to increased saturated fatty acids (SFAs). Conversely, increasing the quantities of dietary monounsaturated fatty acids (MUFAs) or polyunsaturated fatty acids (PUFAs) may reduce liver fat contents. Echeverría et al. [6] reported that high fat levels in SFA-enriched diets are associated with reduced enzymes in the tricarboxylic acid (TCA) cycle, such as citrate synthase and complexes I and II of the mitochondrial electron transport chain. Further studies reported that a reduction in essential PUFAs downregulated the activity of peroxisome proliferator-activated receptor-α (Ppar-α), a transcription factor that supports whole-body energy metabolism [7, 8]. Furthermore, PUFAs activated AMP-activated protein kinase (AMPK) [9], a central regulator of energy homeostasis, which coordinates metabolic pathways and balances nutrient supplies with energy demand.

Siamese crocodiles (Crocodylus siamensis) are native freshwater crocodiles in Southeast Asia and are commercially farmed on a large scale in Thailand. Crocodile oil (CO) is extracted from the abdominal fatty tissues of these crocodiles. CO is richer in MUFAs and PUFAs [10], including oleic and linoleic acids, than are other animal oils. A previous study reported that CO and its products from Siamese crocodiles are used as ointments for burns, scalds, and eczema in traditional Chinese medicine [11]. Another study indicated that oil extracted from crocodile fat tissue showed antimicrobial and anti-inflammatory activities [12]. One study found that fat-oxidation rates were higher after consuming a diet rich in MUFAs than after consuming high-carbohydrate and high-SFA diets [13]. Gene expression profiling and pathway analysis data in another study indicated that omega-3 PUFAs in krill oil regulated genes with diverse aspects of hepatic energy metabolism, including glucose metabolism and mitochondrial respiration [14]. Thus, the fatty-acid composition of CO may be an essential bioactive component for maintaining energy balance and preventing liver disease. However, the mechanism of action of CO remains unclear.

Quantitative studies of dietary mechanisms have been performed using mass spectrometry-based proteomics to understand the molecular physiology of CO and to discover biomarkers. In the current study, we investigated how CO affects the expression of proteins in hepatic energy metabolism in a rat model. We hypothesized that CO, which is rich in oleic acid (MUFA) and linoleic acid (PUFA), may be associated with altered expression of proteins in energy homeostasis pathways in rat liver tissues. Our findings may help reveal the mechanisms of action of CO in maintaining hepatic energy homeostasis and provide new information regarding liver responses to CO consumption.

Materials and Methods

Crocodile oil preparation

Abdominal fat samples were discarded and collected as waste products from slaughtered 3- to 5-year-old Siamese crocodiles (C. siamensis) obtained from a crocodile farm in Nakhon Pathom Province, Thailand. Crocodile oil was extracted, as reported by Santativongchai et al. [15] when the meat was trimmed and prepared. The samples were pressed through two layers of filter cloth with distilled water at a proportion of 1:1 (w/v). The solution was then left undisturbed until the mixture separated. The upper clear oil fraction was then collected, evaporated, and stored in a sealed container at room temperature.

Animals and experimental design

Twenty-one 7-week-old male Sprague Dawley rats were obtained from Nomura Siam International Co., Ltd., Samutprakan Province, Thailand. The animals were housed in individual metabolic cages at 25 ± 2°C on a 12-h light/12-h dark cycle with ad libitum access to standard chow and drinking water. Rats were randomly assigned into three groups (n=7 rats per experimental group): normal rats supplemented with 3 ml of sterile water per kilogram body weight (N group), normal rats supplemented with 3 ml of CO per kilogram body weight (CO group), or normal rats supplemented with 3 ml of palm oil (PO) per kilogram body weight (PO group). The rats were administered sterile water, CO, or PO by oral gavage once daily for 7 weeks. The fatty acid composition of the fats used is listed in Table 1. The study conducted adhered to the guidelines for the care and use under the Ethical Review Board of the office of National Research Council of Thailand (NRCT). The ethics committee of Kasetsart University Research and Development Institute, Kasetsart University, Thailand, approved this study (Approval No. ACKU61-VET-088).

Table 1. Fatty acid composition (g/100 g FA) of crocodile oil or palm oil.

Fatty acids	Crocodile oil	Palm oil
Lauric acid (C12:0)	0.11	0.35
Myristic acid (C14:0)	0.57	1.39
Pentadecylic acid (C15:0)	0.09	0.08
Palmitic acid (C16:0)	19.9	43.4
Heptadecanoic acid (C17:0)	0.16	0.13
Stearic acid (C18:0)	5.42	4.5
Arachidic acid (20:0)	0.8	0.1
Oleic acid (C18:1)	41.07	37.5
Myristoleic acid (C14:1)	0.1	0
Palmitoleic acid (C16:1)	3.83	0.54
Linoleic acid (C18:2)	21.1	10.1
Linolenic acid (C18:3)	0.96	0.4
Total SFAs	27.1	50.5
Total MUFAs	45.5	37.5
Total PUFAs	23.8	10.5

FA, fatty acid; SFAs, saturated fatty acids; MUFAs, monounsaturated fatty acids; PUFAs, polyunsaturated fatty acid.

Measurement of body weight and energy intake

Food consumption was measured by weighting the rats daily between 11:00 and 11:30 A.M. Each rat’s food intake was measured by weighing its leftover chow. Energy intake was calculated from the determined daily food consumption. The daily energy intake per rat (kcal/day) was calculated:

Daily energy intake = (daily food consumption per rat × TE of standard chow) + TE from each treatment.

TE was the total energy of each rat diet, including 3.040 kcal/g from the standard chow, 8 kcal/ml from the CO, and 8.67 kcal/ml from the PO. Meanwhile, body weight was measured weekly throughout the experiment.

Sample collection and analysis of energy metabolism-related intermediates

At the end of experiment, all animals were euthanized with 60 mg/ml pentobarbital sodium. Blood samples were collected and centrifuged at 2,200 g for 15 min at 4°C, and the serum was stored at −20°C until further analysis. Serum lipid profiles, which included triglycerides, cholesterol, high-density lipoprotein (HDL), and low-density lipoprotein (LDL), and were determined using a Hitachi 7080 analyzer (Hitachi, Tokyo, Japan).

Liver tissue was removed and weighed to determine the liver weight and indexes (liver weight/body weight). The liver tissues were excised, homogenized with ice-cold phosphate-buffered saline (20% w/v), and centrifuged at 2,000 g for 20 min at 4°C. The supernatants were stored at −80°C until further analysis. High-performance liquid chromatography (HPLC) was used to determine the energy-related intermediates according to Lillefosse et al. [16]. The frozen supernatants were mixed with methanol at a ratio of 1:4 (v/v). After centrifugation at 20,000 g for 20 min at 4°C, the supernatants were removed and evaporated using a freeze dryer at −80°C. The metabolites were then redissolved in 500 µl HPLC buffer. Each 5-µl sample was analyzed via HPLC. Chromatography was performed with an injection volume of 5 µl and a column temperature of 40°C. A 5-µm InertSustain C18 column (150 × 4.6 mm) was used to separate the mobile phase consisting of 8% 1 N sulfuric acid. Gradient elution was performed at a flow rate of 1 ml/min. A UV-Visible detector set at 205 nm was used.

Liquid chromatography-mass spectrometry (LC-MS) analysis and identification of proteins

LC-MS analysis was used to determine the hepatic protein expressions. Liver supernatants were mixed with acetone at a 2:1 (v/v) ratio and centrifuged at 10,000 g for 10 min. The pellets were suspended in lysis buffer (0.25% w/v sodium dodecyl sulfate, 50 mM Tris-HCl, pH 9.0), and the protein concentration was examined via Lowry’s method [17] using bovine serum albumin as the protein concentration reference standard. Pooled samples for each group were made by mixing equal amounts of protein from individual tissue samples.

Disulfide bonds in 5 µg of the protein samples were reduced using 5 mM dithiothreitol in 10 mM ammonium bicarbonate at 60°C for 1 h. The sulfhydryl groups were alkylated with 15 mM iodoacetamide in 10 mM ammonium bicarbonate for 45 min in the dark at room temperature. Subsequently, the protein samples were mixed with sequencing-grade trypsin (ratio of 1:20; Promega, Mannheim, Germany) and incubated at 37°C overnight. The tryptic peptides were dried, protonated with 0.1% formic acid, and then injected into an Ultimate 3000 Nano/Capillary LC system (Thermo Scientific, Loughborough, UK) coupled to an HCTultra (Bruker Daltonics, Billerica, MA, USA), with an electrospray at a flow rate of 300 nl/min into a nanocolumn (100-mm PepSwift monolithic column with a 50-mm internal diameter). A mobile phase of solvent A (0.1% formic acid) and solvent B (80% acetonitrile and 0.1% formic acid) was used to elute peptides using a linear gradient of 4–70% of solvent B for 0–20 min (the time point of retention) followed by 90% solvent B for 20–25 min to remove all peptides in the column. A final elution of 10% solvent B was performed for the final 25–40 min to remove any remaining salts. Mass spectra of the peptide fragments were acquired in a data-dependent AutoMS [2] mode with a scanning range of 300–1,500 m/z. Three averages were taken, and up to five precursor ions were selected from an MS scan range of 50–3,000 m/z.

The DeCyder MS Differential Analysis software (DeCyderMS; GE Healthcare, Chalfont St. Giles, UK) was used to quantify the proteins in individual samples, and the Mascot search engine was used to correlate the MS/MS spectra to a Macaca protein database maintained by UniProt [18, 19]. Mascot’s standard settings were used, which included a maximum of three missed cleavages, a peptide tolerance of 1.2 Da, an MS/MS tolerance of 0.6 Da, trypsin as the digesting enzyme, cysteine carbamidomethylation as the fixed modification, methionine oxidation as the variable modification, and peptide charge states. Protein levels in each sample were expressed as the log2-expression intensity.

Data analysis and statistical methods

Venn diagrams were used to counting and compare the lists of proteins in each treatment [20]. The jvenn software displays protein expression data as Venn diagrams, and two statistical charts were generated to assess the homogeneity of the list size and compare the compactness of multiple Venn diagrams. Targeted protein-related energy metabolism was sorted from candidate proteins in a group, and the proteins were classified and identified according to their appearance, which correlated the molecular junctions, biological processes, and pathways at the organismal level based on the NCBI and UniProt protein databases. The list of targeted proteins was then analyzed for associations with the predicted proteins in the energy homeostatic pathway using STITCH version 5.0. Finally, hierarchical and SOTA clustering analyses were performed using Multiexperiment Viewer (MeV) version 4.9. The SOTA analysis was utilized to identify groups of proteins that showed similar expression profiles. The protein expression heatmaps were visualized using GraphPad Prism 9 (GraphPad Software, San Diego, CA, USA).

The data are expressed as means ± SEM. Statistical analysis was performed via one-way analysis of variance, followed by Tukey’s post hoc test using the R project statistical computing package (R core team, 2019). P<0.05 was considered statistically significant.

Results

Effect of CO consumption on body weight, energy intake, liver weight, liver indexes, and lipid profiles

CO treatment did not significantly affect the change in body weight of the rats among the various groups on different days. At the end of experiment, the maximum body weight was observed in the CO group, follow by the PO group and N group, with the weights being 534.43 ± 5.98, 528.86 ± 19.51, and 522.57 ± 19.25 g, respectively. Moreover, CO also had no significant impact on energy intake during the 7-week period (Fig. 1). In addition, CO treatment did not affect the liver weights, liver indexes, or lipid profiles compared with those of the N and PO groups. Interestingly, both the CO (90.40 ± 15.14 mg/dl) and PO (83.68 ± 6.44 mg/dl) groups showed a decreasing trend in serum triglyceride levels when compared with the N (138.25 ± 30.14 mg/dl) group.

Fig. 1.

Effects of dietary crocodile oil on (A) body weight, (B) energy intake, (C) liver weight, (D) liver indexes, (E) serum cholesterol levels, (F) serum triglyceride levels, (G) serum LDL levels, and (H) serum HDL levels in rats (n=7 per group). Data are expressed as the mean ± SEM.

Effect of CO consumption on energy metabolic intermediates in the liver

The key energy-related intermediates of the rat liver among the three groups are shown in Fig. 2. The CO-treated group had significantly increased liver oxaloacetate (17.90 ± 0.53 mg/l) and malate levels (246.92 ± 6.02 mg/l) when compared with the PO-treated group (14.37 ± 0.53 mg/l and 204.41 ± 6.61 mg/l, respectively). Meanwhile, there was no difference in hepatic malate between the CO and N groups. On the other hand, CO treatment did not significantly affect hepatic lactate, pyruvate, citrate, or alpha-ketoglutarate levels.

Fig. 2.

Effects of dietary crocodile oil on liver energy metabolism-related metabolites in rats (n=7 per group): (A) lactate, (B) pyruvate, (C) oxaloacetate, (D) citrate, (E) malate, and (F) alpha-ketoglutarate levels. Data are expressed as the mean ± SEM. Different letters indicate statistically significant differences between groups (P<0.05).

Effect of CO consumption on hepatic energy-related protein expression

In total, 12,627 proteins were identified and presented in a Venn diagram (Fig. 3A) that shows the numbers of differentially expressed proteins between the groups. Of them, 3,733 proteins were expressed in all groups; 1,790 were expressed only in the CO group; 1,514 were expressed only in the N group; and 1,125 were expressed only in the PO group. The unique proteins were further analyzed for their functions by PANTHER classification system according to molecular function (Fig. 3C) and biological process (Fig. 3D). The results showed that 17.30% of the unique proteins from the N group, 16.80% of the unique proteins from the CO group, and 15.40% of the unique proteins from the PO group were involved with catalytic activity based on molecular function. Meanwhile, the functional analysis according to biological process revealed that unique proteins associated with metabolic and cellular processes were observed in the N group (13.70% and 23.30%, respectively), CO group (13.20% and 23.00%, respectively), and PO group (12.50% and 22.70%, respectively). The distribution of the unique proteins involved with energy metabolic processes in the CO group was greater than in the PO group, but the contributions of the unique proteins in the CO group were decreased compared with those in the N group.

Fig. 3.

Comparative and interaction network analyses of hepatic protein expression in sterile water (N)-treated, crocodile oil (CO)-treated, and palm oil (PO)-treated rats (n=7 per group). (A) Venn diagram of overexpressed and shared proteins in each group. (B) Chemical-protein and protein-protein interaction networks of unique proteins from CO-treated rats and predicted proteins in the energy metabolic pathway of the liver, analyzed using STITCH v. 5.0. (C) PANTHER functional classification of unique proteins among groups according to molecular function. (D) PANTHER functional classification of unique proteins among groups according to biological process.

To better determine how CO consumption affects the biomarkers and molecular mechanisms of energy metabolism, we used the UniProt database to evaluate and methodically identify the 1,790 proteins that were unique to the CO-treated group (Table 2). Only 22 proteins were associated with energy homeostasis metabolism. Their functions were related to ATP biosynthesis pathways, including glycolysis, the TCA cycle, and oxidative phosphorylation (OXPHOS). The network indicated that the unique energy-related proteins detected in the CO-treated group were strongly associated with TCA cycle metabolites and the AMPK signaling pathway in the liver (Fig. 3B).

Table 2. Overexpressed proteins in the crocodile oil-treated group based on hepatic energy metabolic pathway involvement.

Protein ID	Protein names	Gene names	Peptide	Pathway
B2RYS0	Cytochrome c oxidase subunit 7A2, mitochondrial	Cox7a2, Cox7a3, Cox7al	HFENKVPEKQK	Oxidative phosphorylation
P11951	Cytochrome c oxidase subunit 6C-2	Cox6c2, Cox6c	AYADFYR
P10888	Cytochrome c oxidase subunit 4 isoform 1, mitochondrial	Cox4i1, Cox4, Cox4a	EKADWSSLSR
Q6AY07	Fructose-bisphosphate aldolase	Aldoart2, Aldoal1 Aldoart1 rCG_62051	ALQASALKAWGGK	Glycolysis
Q5XIV1	Phosphoglycerate kinase	Pgk2 rCG_43578	ACANPGK
O88328	Triosephosphate isomerase	LOC246267 rs11	NNVEVYVAPSAIHLK
Q920L2	Succinate dehydrogenase [ubiquinone] flavoprotein subunit, mitochondrial	Sdha	ACALSIAESCRPGDKVPPIK	Tricarboxylic acid (TCA) cycle
Q7TQM4	Acetyl-CoA acetyltransferase, cytosolic	Acat2	FGDRMFYRDWWNSTSFSNYYR	Fatty acid metabolism
Q5KTC7	N-acylethanolamine-hydrolyzing acid amidase	Naaa, Asahl	NLDYPFGNALR
P51556	Diacylglycerol kinase alpha (DAG kinase alpha)	Dgka, Dagk, Dagk1	DGPEPGLRFFK
Q5XIC0	Enoyl-CoA delta isomerase 2	Eci2, Peci	AKWDAWNALGSLPK
Q5M9H2	Very long-chain specific acyl-CoA dehydrogenase, mitochondrial	Acadvl	GIVNEQFLLQR
Q2LAM0	Fatty acid 2-hydroxylase	Fa2h, Faah	RWLEQYYVGELR
Q6P7R8	Very-long-chain 3-oxoacyl-CoA reductase	Hsd17b12	AFVDFFSQCLHEEYK
A0A0G2K330	Trifunctional enzyme subunit beta, mitochondrial	Hadhb	AALSGLLYR
F8WG67	Acyl-CoA thioesterase 7, isoform CRA_a	Acot7 rCG_30783	HCNSQNGER
D3ZPF2	Malonyl-CoA-acyl carrier protein transacylase	Mcat	TLGSINIKKPLVAVHSNVSGHKYTHPQHIR
A0A0G2K330	Trifunctional enzyme subunit beta, mitochondrial	Hadhb	AALSGLLYR
D3ZPX9	Elongation of very long chain fatty acids protein 3	Elovl3, ELOVL3 Elovl3_predicted rCG_57640	MDTSMNFSRGLK
F1LX28	Acyl-CoA thioesterase 11	Acot11	HINSAFMTFVVLDKDDQPQK
Q68FZ8	Propionyl-CoA carboxylase beta chain, mitochondrial	Pccb	GFVDDIIQPSSTR	Metabolic intermediate metabolism
P97519	Hydroxymethylglutaryl-CoA lyase, mitochondrial	Hmgcl	DGLQNEKSIVPTPVK

Effects of CO consumption on up- and downregulated proteins involved in liver energy metabolism

To better understand the metabolic mechanisms of CO administration, we sorted the proteomic patterns of the differentially expression proteins into 6 clusters based on hierarchical and SOTA algorithms. Fifty-nine proteins that showed statistically significant changes in abundance were depicted in a heatmap using hierarchical clustering (Fig. 4A), and SOTA cluster analysis yielded 6 clusters based on the expression profile of individual proteins (Fig. 4B). The CO-treated group exhibited significantly different expression levels for 40 proteins (21 downregulated, 19 upregulated) compared with those of the N and PO-treated groups. We have selected 2 SOTA cluster groups (cluster 1 and cluster 4), which contained a number of proteins, to further examine the expression patterns of the opposite trends from CO treatment. Cluster 1 (C1), the upregulated cluster (Fig. 4C), contained proteins that were expressed more in the CO-treated group (Table 3) than in the N and PO-treated groups. This cluster contained mainly proteins that were strongly linked to targeted proteins and intermediates involved in OXPHOS and the TCA cycle (STITCH network, Fig. 4E). Cluster 4 (C4), the downregulated cluster (Fig. 4D), contained proteins that were expressed more in the N and PO-treated groups than in the CO-treated group. Most of the proteins in the downregulated cluster (Table 4) were involved in the TCA cycle, mitochondrial cytochrome c functions, and glucose use (Fig. 4F).

Fig. 4.

Targeted quantitative analysis of changes in the expressions of liver proteins involved in energy homeostasis in sterile water (N)-treated, crocodile oil (CO)-treated, and palm oil (PO)-treated rats (n=7 per group). (A) Hierarchical clustering and heat map of energy homeostatic proteins exhibiting differential expression. (B) Six expression cluster centroids of the 59 differentially expressed proteins based on their expression profiles using SOTA analysis. The pink line indicates the average trend for a given cluster, the experimental groups are indicated on the X-axis, and the change in expression level is indicated on the Y-axis. (C) Expression patterns of the selected upregulated CO-protein cluster (C1) corresponding to SOTA clustering. (D) Expression patterns of the selected downregulated CO-protein clusters (C4) corresponding to SOTA clustering. (E) Chemical-protein and protein-protein interaction networks of upregulated proteins in the upregulated CO-protein cluster (C1) and predicted proteins in the energy metabolic pathway following CO treatment were analyzed using STITCH v. 5.0. (F) Chemical-protein and protein-protein interaction networks of downregulated proteins in the downregulated CO-protein cluster (C4) and predicted proteins in the energy metabolic pathway following CO treatment were analyzed using STITCH v. 5.0. The green to red color spectrum represents the log2-expression intensities of the proteins between 10 and 18 (low to high expression).

Table 3. List of hyperexpressed proteins of crocodile oil-treated group involved in energy metabolic pathway according to upregulated CO-proteins cluster.

Protein ID	Protein names	Gene names	Peptides	Biological processes
G3V8M8	Poly (ADP-ribose) glycohydrolase	Parg	DAILKYNVAYSK	ATP generation from poly-ADP-D-ribose
P27008	Poly [ADP-ribose] polymerase 1 (PARP-1)	Parp1, Adprt	DPIDVNYEKLKTDIK	ATP generation from poly-ADP-D-ribose
Q9JJH5	6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase 2	Pfkfb2	DLSFIKVMNVGQRFLVNR	Glycolysis, pyruvate metabolic process
Q6P6R2	Dihydrolipoyl dehydrogenase, mitochondrial	Dld	GIEIPEVR	Alpha-ketoglutarate metabolic process
P19511	ATP synthase F(0) complex subunit B1, mitochondrial	Atp5pb Atp5f1	CIGDLKMLAK	Mitochondrial ATP synthesis coupled proton transport
M0R3X7	Hexokinase-4 (HK4)	Gck	ARGVQDTDVVNR	Glycolysis, glucose homeostasis
Q5XIH3	NADH dehydrogenase [ubiquinone] flavoprotein 1, mitochondrial	Ndufv1	AMGARAAYIYIRGEFYNEASNLQVAIR	Mitochondrial ATP synthesis coupled electron transport
P21643	Tryptophan 2,3-dioxygenase	Tdo2, Tdo	EEQMAEFRK	Tryptophan catabolic process to acetyl-CoA
B5LSW7	Cysteine sulfinic acid decarboxylase	Csad	FVNVCFWFVPPSLRGKK	Carboxylic acid metabolic process
F1LP30	Methylcrotonoyl-CoA carboxylase subunit alpha, mitochondrial	Mccc1, Mcca	EHAGKIGYPVMIKAIR	Leucine catabolic process
F1LU71	Methylglutaconyl-CoA hydratase, mitochondrial	Auh	AYGKNSLSKNLLK	Fatty acid beta-oxidation
D4ACE9	Alpha-aminoadipic semialdehyde synthase, mitochondrial	Aass	APLAPKHIKGITK	L-lysine catabolic process
P57093	Phytanoyl-CoA dioxygenase, peroxisomal	Phyh	AEFERICR	Alpha-ketoglutarate metabolic process
A0A0G2K2S1	Proline dehydrogenase	Prodh1, Prodh	DLKWCLGSRVYFK	Proline catabolic process to glutamate
B2RYG2	Phosphoenolpyruvate carboxykinase (GTP)	Pck2	EPCAHPNSRFCVPAR	Oxaloacetate metabolic process, NADH oxidation
Q5U3Z7	Serine hydroxymethyltransferase	Shmt2	DRQCRGLELIASENFCSR	Regulation of oxidative phosphorylation, L-serine catabolic process

Table 4. List of hypoexpressed proteins of crocodile oil-treated group involved in energy metabolic pathway according to downregulated CO-proteins cluster.

Protein ID	Protein names	Gene names	Peptides	Biological processes
Q7TP91	Surfeit locus protein 1	Surf1, Surf-1	FVRRTPGV	Mitochondrial cytochrome c oxidase assembly
P16617	Phosphoglycerate kinase	Pgk1	AGGFLMKKELNYFAK	Glycolytic process
Q68FY0	Cytochrome b-c1 complex subunit 1, mitochondrial	Uqcrc1	IRSGMFWLR	Mitochondrial electron transport, ubiquinol to cytochrome c
M0RAQ6	Hexokinase-1	Hk1	GAALITAVGVRLRGDPSIA	Cellular glucose homeostasis, glycolysis
Q01205	Dihydrolipoyllysine-residue succinyltransferase component of 2-oxoglutarate dehydrogenase complex, mitochondrial	Dlst	AAVEDPRVLLLDL	Alpha-ketoglutarate metabolic process
D3ZLJ6	Amine oxidase	Il4i1 LOC100360621 rCG_53598	AHACLSDRLR	Cellular amino acid catabolic process

Discussion

The liver is a key metabolic organ involved in glucose metabolism and energy homeostasis. Numerous transcription factors largely modulate hepatic energy metabolism at the genomic level. Metabolic alteration activities are regulated by the expression of proteins that dynamically regulate gluconeogenesis, beta-oxidation, the TCA cycle, and OXPHOS in the liver to meet systemic energy metabolic demands [1]. A previous study reported that hepatic fat accumulation was associated with decreased mitochondrial beta-oxidation due to mitochondrial dysfunction [21]. Here, we investigated the mechanisms underlying energy metabolic alterations in the liver by determining the changes in the liver proteome using quantitative shotgun proteomics in CO-treated rats. Our results showed that 7 weeks of CO administration did not affect body weight, energy intake, hepatic contents, or lipid profiles in the rats. However, CO treatment increased key intermediates in the hepatic TCA cycle and in various metabolic pathways of energy metabolism. We identified 22 unique proteins involved in energy metabolic pathways from among 1,790 overexpressed proteins in the CO-treated group. The chemical-protein and protein-proteins interactions of these regulated proteins were analyzed using STITCH to illustrate the associations between the obtained proteins and other proteins or ligands involved in energy homeostasis. Fifty-nine proteins expressed in all groups were related to energy metabolism. SOTA clustering revealed that 16 upregulated proteins in the CO-treated group were involved in OXPHOS regulation, and 6 downregulated proteins in the CO-treated group were associated with glucose utilization.

Dietary fat composition plays important roles in animal health and many chronic diseases. The fatty acid profiles revealed that CO had higher levels of total MUFAs and PUFAs compared with PO, while PO had higher total SFAs than CO. PO, the most commonly used vegetable oil worldwide, contains many more saturated fats than do vegetable oils [22]. In our previous research, we investigated and found that CO administration significantly lowered the total surface area of triglyceride-rich lipid droplets in hepatocytes more than PO administration [23]. Sales et al. [24] showed that PO caused lipid accumulation and altered hepatic metabolism related to calcium ion homeostasis. The intracellular calcium concentration considerably impacts mitochondrial function and directly stimulates energy metabolic activity [25, 26]. Although our results showed no differences in lipid profiles among the three groups, the effect of PO treatment on the lipid profiles contradicted the findings of many other studies. A recent systemic review and meta-analysis found that PO enriched with unsaturated fatty acids marginally increased the serum cholesterol, LDL, and HDL levels [27]. Likewise, CO caused no changes in lipid profiles in the present study and tended to lower the serum triglyceride levels compared with those of the N group. The main fatty-acid composition of CO is similar to that of PO, but CO has less palmitic acid (PA) and more linoleic acid (LA) than PO [15, 28]. LA is an important essential fatty acid required by the body. A previous study on the anti-obesity effects of conjugated LA (CLA) determined that CLA decreased body fat and energy intake, increased energy expenditure, and modulated metabolism in lipids, adipocytes, and skeletal muscle [29]. Additionally, CLA can reduce lipid accumulation by elevating energy expenditures via thermogenesis or lipid oxidation [30, 31]. Kien et al. [32] studied the effects of dietary PA on fatty-acid oxidation and energy expenditure and found that increased dietary PA decreased fatty-acid oxidation and energy expenditure and increased the risk of obesity. Thus, CO, an essential oil enriched in LA, may improve hepatic energy metabolism activity by restoring TCA cycle homeostasis and mitigating altered lipid profiles.

Our proteomic analysis revealed that there was a lower proportion of unique proteins associated with energy metabolism (catalytic and metabolic activity) in the CO group than in the normal group but a greater proportion of them than in the PO group. Consistent with our previous studies, CO improved hepatic mitochondrial morphology, the hepatic mitochondrial protein for maintaining energy metabolic activity [23], and serum energy-related proteins [33] much better than a PO diet. The current finding suggests that CO could retard energy metabolism dysfunction resulting from consumption of a high-fat diet when compared with PO administration. Furthermore, interaction networks between the isolated proteins and other proteins in the STITCH database revealed how CO affects hepatic metabolism. The overexpressed proteins in the CO-treated group (Table 2) were associated with the AMPK signaling pathway, which is involved in energy homeostasis (Fig. 3B). Cox7a2, Cox6c2, and Cox4i1, the components of cytochrome c oxidase, are the enzymes that drive OXPHOS in the mitochondrial electron transport chain. Sdha is an essential metabolic enzyme involved in the TCA cycle, and its genes encode succinate dehydrogenase complex flavoprotein subunit a. Sdh oxidizes succinate to fumarate, a key step in the TCA cycle that regulates energy production in hepatocytes associated with OXPHOS activity [34], and this corresponds to our results in Fig. 3. The TCA cycle intermediates in the CO-treated group exhibited higher levels of malate than did those in the PO-treated group. Moreover, lactate levels showed a lower trend in the CO-treated group than in the PO-treated group (Fig. 2). A previous study found that Sdh deficiency increases glycolysis, lactate production, and the pentose-phosphate pathway [35]. Further research showed that omega-3-enriched oils may affect mitochondrial function and improve TCA cycle homeostasis, preventing the development of obesity [36]. Upregulated Sdha expression likely induces malate oxidation and inhibits lactate production in the liver energy-intermediate pathway. In addition, other overexpressed proteins, including Elovl3, Acot7, and Acat2, were observed in CO treatment. Elovl3 regulates resistance to diet-induced obesity [37]. Bowman et al. [38] demonstrated that Acot7 helps prevent high-fat diet-induced metabolic dysfunction, is an important regulator of organismal responses to dietary lipids, and regulates brain-specific metabolic processes related to the whole-body response to dietary lipids. Meanwhile, Acat2 is a potential target for preventing and treating diseases involving dysregulated cholesterol metabolism, such as hypercholesterolemia and atherosclerosis [39].

The heatmap in Fig. 4A shows that the levels of differentially expressed proteins in the CO-treated group enhanced Prkaa1 expression compared with that of the PO-treated group. Prkaa1 is the catalytic α-subunit of 5′-AMPK. AMPK also maintains adequate NADPH levels by regulating fatty acid oxidation via phosphorylation-induced metabolic stresses by increasing the TCA cycle intermediates [40]. Further study revealed that a high-fat diet markedly reduced AMPK activity across several tissues and was associated with obesity and prediabetes [41]. Thus, long-term intake of dietary CO might improves energy utilization by upregulating TCA cycle, OXPHOS, and AMPK signaling pathway activities in the liver. Similarly, a previous study demonstrated that PUFA-enriched fish oil increased Ppar-γ expression in response to AMPK activation. This induced lipoprotein lipase gene expression [42] and increased Ppar-α activity [7, 8], which is associated with metabolic regulation in skeletal muscles and liver energy metabolism.

To better understand the molecular mechanisms of the metabolic impacts of CO on liver energy metabolism, we developed a clustering model of CO-protein expression among three rat groups. The upregulated CO-protein cluster contained proteins that were hyperexpressed in the CO-treated group, including ATP synthase F (0) complex subunit B1 (atp5pb), methylcrotonoyl-CoA carboxylase subunit alpha (mccc1), cysteine sulfinic acid decarboxylase (Csad), and NADH dehydrogenase (ubiquinone) flavoprotein 1 (Ndufv1). Ndufv1 is the first enzyme complex in the mitochondrial electron transport chain and catalyzes the transfer of electrons from NADH to the electron acceptor, ubiquinone. NADH dehydrogenase (complex I) is the most abundant enzyme in the electron transport chain [43] and is essential for OXPHOS in the mitochondria [44]. A previous study reported that a high-fat diet downregulates the genes necessary for OXPHOS and mitochondrial biogenesis, such as Ndufv1, Ndufs1, and Sdhh, in skeletal muscles [45]. Interestingly, García-Ruiz et al. [46] found that high-fat diet-induced nonalcoholic fatty liver disease (NAFLD) exhibited several features of hepatic metabolic syndrome, including increased gene expressions of markers of inflammation, fibrosis, and apoptosis. These researches also showed that OXPHOS complex activity and ATP contents were markedly reduced in the livers of NAFLD-induced mice compared with those of their control group. Our studies have suggested that dietary CO is an important factor that increases OXPHOS proteins. In the current study, we identified a protein-protein interaction between protein overexpression in CO-treated rats, the targeted OXPHOS-related proteins, and key TCA cycle intermediates. The Nduf family proteins (Ndufv1, Ndufv2, Ndufs1, Ndufs2, and Ndufs7) play important roles in the minimal assembly required for catalysis and are involved in OXPHOS [47, 48]. CO treatment can also enhance glycolysis and support the TCA cycle and OXPHOS in hepatocytes of CO-fed rats.

In this study, hypoexpressed proteins in CO-treated rats (downregulated CO-protein cluster) were associated with energy metabolism (surfeit locus protein 1, Surf1; phosphoglycerate kinase 1, Pgk1; Il4i1; Figs. 4D and F). However, several studies showed that mice lacking Surf1, a complex IV assembly protein, displayed an improved metabolic phenotype, including reduced adiposity, increased insulin sensitivity, and mitochondrial biogenesis despite a >50% reduction in Cox activity [49,50,51]. Phosphoglycerate kinase 1 is one of only two enzymes that can generate ATP in the glycolytic pathway. A previous study found that activated Pgk1 enhanced lactate production [52]. Lactate is an important bioenergetic metabolite formed during aerobic glycolysis and can be used by cells as an oxidative substrate [53]. Previous studies found that lactate clearance may be impaired in liver disease [54, 55], and substrates act as sources of lactate. Similarly, further studies reported that hepatic lactate concentrations were elevated in obese rodents [21, 56,57,58]. We found that CO-treated rats exhibited downward trends in hepatic lactate levels compared with PO-treated rats; this homeostatic mechanism in rats treated via CO administration may prevent stress-induced lactate production. Thus, CO-treated rats may metabolize energy to improve the production of TCA cycle-related intermediates, activate OXPHOS proteins, regulate glycolytic processes, and reduce physiological stress from lactate accumulation in the liver.

Conclusions

Our findings provide new insights into the potential effects of dietary CO on the differential expression of the liver proteome in cellular energy metabolism. CO treatment enhanced production of key metabolites (oxaloacetate) involved in hepatic energy. Our quantitative proteomic analysis revealed unique proteins associated with the maintenance of metabolic activity and the overexpression of energy metabolism-related proteins involved in the TCA cycle and OXPHOS in CO-treated rats. These overexpressed proteins influenced the AMPK signaling pathway. Furthermore, hierarchical and SOTA clustering of 59 shared differentially expressed proteins involved in energy expenditure revealed that hyperexpressed proteins (atp5pb, mccc1, Csad, and Ndufv1) in the upregulated cluster improved OXPHOS activity. Hypoexpressed proteins (Surf1, Pgk1, and Il4i1) in the downregulated CO-protein cluster managed glycolytic processes coupled with decreased lactate production. CO may play a significant role in maintaining energy homeostasis by regulating the TCA cycle and OXPHOS activity. Finally, our findings support dietary CO treatment and its effects on the expression of proteins associated with hepatic energy metabolism. CO may be a cost-effective dietary fat source for the food and nutrition industries.

Author Contributions

Conceptualization, W.F. and P.T.; methodology, K.P., K.S., P.S., S.C., and N.P.; software, K.P., S.C., and S.R.; validation, K.P., S.R., P.T., and W.F.; formal analysis, K.P.; data curation, K.P. and S.R.; writing—original draft preparation, K.P.; writing—review and editing, S.R., P.T., and W.F.; visualization, K.P.; supervision, W.F. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data Availability Statement

Not applicable.

Conflicts of Interest

The authors declare no conflict of interest.