Comparative evaluation of the effect of resveratrol and carnitine on the full transcriptomic profile of liver tissue in mice with different sensitivity to the development of alimentary obesity

Specialized food products and biologically active food supplements enriched with minor biologically active substances are considered as a useful supplement in the treatment of obesity and other nutrition-dependent diseases. Biologically active substances of food can have a complex effect on the expression of a large number of genes, which can affect the results of a therapy. The aim of the study was to analyze the nutrigenomic mechanisms of the effect of biologically active substances - l-carnitine and resveratrol on the expression of liver genes of DBA/2J mice and DBCB tetrahybrid, differing in genotype and sensitivity to the development of diet-induced obesity, using the method of full transcriptomic profiling of liver tissue. We carried out the experiment on male DBA/2J mice and the hybrid of the 2nd generation DBCB, obtained by crossing 4 lines of mice (DBA/2J, BALB/c, CBA/ lac and С57Black/6J). Mice for the experiment were obtained from Stolbovaya nursery, Federal State Budgetory Scientific Institution Scientific Center of Biomedical Technologies of the Federal Medical-Biological Agency (Moscow region, Russia). We worked with animals in accordance with international recommendations (Directive 2010/63/EU on the protection of animals used for scientific purposes adopted on September 22, 2010; Guide for the care and use of laboratory animals. Eighth Edition / Committee for the Update of the Guide for the Care and Use of Laboratory Animals; Institute for Laboratory Animal Research (ILAR); Division on Earth and Life Studies (DELS); National Research Council of the National Academies. Washington: The National Academies Press. 2011).The mice were divided into four groups with an equal number of 8 individuals. During 65 days, animals of the 1st (control) groups received a balanced semi-synthetic diet and purified drinking water, the 2nd groups received a high-carbohydrate and high-fat diet with a high fat content (30% by of dry matter of the diet) and replacing drinking water by 20% fructose solution, 3rd groups - high-carbohydrate and high-fat diet with the addition of resveratrol at a dose of 25 mg/kg body weight, 4th groups - high-carbohydrate and high-fat diet with the addition of l-carnitine at a dose of 300 mg/kg body weight. Full transcriptome analysis was performed using the Gene Expression Hybridization Kit (Agilent Technologies, USA) on SurePrint G3 Mouse GE 8×60K Microarray Kit microarrays. Differential gene expression was expressed as base 2 logarithm of increasing or decreasing fluorescence (log2 FC) compared to control groups, separately for DBA/2J and DBCB mice. Chip scan data and calculation of differential expression values were exported to the “R” IDE and bioinformatics analysis was performed with quantile normalization and further analysis in the limma package. The packages AnnotationDbi, org.Rn.eg.db, pathview, gage, gageData were used to identify metabolic pathways among the genes, metabolic pathways and functions of biological systems presented in the international database Kyoto Encyclopedia of Genes and Genomes and to visualize them. To visualize the results at all stages, the standard “R” graphics and additional packages ggplot2, ggrepel, and gplots were used. Liver morphology was studied by light microscopy after staining with hematoxyline-eosine (See Fig. 1). We revealed differential expression for at least one of the intergroup comparisons in the amount of │log2FC│≥0.5 (towards both enhancement and attenuation) and at a p-value ≤ 0.05 for 415 transcripts, of which 311 were identified with proteins or RNA with a known function (See Tables 1-3). Consumption of a high-carbohydrate and high-fat diet was reflected in differential expression of 62 genes in DBA/2J mice and 97 in DBCB mice. In DBA/2J mice fed on a high-carbohydrate and high-fat diet, supplementation with resveratrol and l-carnitine caused a differential expression of 26 genes each. At the same time, only 2 genes (Pklr, Tkfc) responded to resveratrol and l-carnitine in mice of this strain. In DBCB tetrahybrid mice, resveratrol consumption corresponded to differential expression of 147 genes, and l-carnitine consumption corresponded to 221 genes. 61 genes from DBCB mice responded to both supplements, and the number of genes simultaneously targeted by high-carbohydrate and high-fat diets, resveratrol and l-carnitine was 10 (See Fig. 2). The gene expression profiles in DBA/2J and DBCB mice formed two separate clusters, the differences within which, determined by the composition of the diets, were less significant than the interstrain differences (See Fig. 3). Differential expression values in DBCB and DBA mice responding to HFCD and both supplements correlated negatively (See Fig. 4). The consumption of a high-carbohydrate and high-fat diet in DBA/2J mice resulted in significant changes in 4 metabolic pathways, and in DBCB mice, in addition, in 5 more metabolic pathways. Resveratrol consumption did not cause significant changes in DBA/2J mice, and in tetrahybrid mice it affected mmu04512 ECM-receptor interaction. L-carnitine supplementation caused significant changes in mmu00830 Retinol metabolism only in DBCB mice (See Table 4). Consumption of a high-carbohydrate and high-fat diet produced similar changes in the mmu00830 Retinol metabolism pathway in both mice (See Fig. 5). In metabolic pathway mmu03320 PPAR signaling pathway DBA/2J and DBCB mice showed positive differential expression of the PPARγ gene and negative Scd1. At the same time, only DBCB mice in this metabolic pathway are characterized by activation of the RXR gene expression and suppression of FABP, and the direction in changing Cyp4a1 in both mice is opposite (See Fig. 6). Changes in the metabolic pathway mmu00590 Arachidonic acid metabolism characterized by the imbalance in the expression of Cyp4a and Cyp2 isoforms, which are responsible for the synthesis of various hydroxy and epoxy derivatives of arachidonic acid, is characteristic only of DBCB mice (See Fig. 7). Thus, the experiments performed revealed both a certain similarity and differences in the response of the transcriptome of DBA/2J and DBCB mice to the consumption of a high-carbohydrate and high-fat diet, resveratrol and l-carnitine. The mechanisms that determine the direction of changes induced in the transcriptome of mice (and in coupled phenotypic changes) are, apparently, in the intervention of the studied dietary factors in key metabolic pathways, such as the PPAR signaling pathway, the metabolism of retinoids and eicosanoids. The data obtained indicate the importance of an adequate choice of a in vivo model of obesity and metabolic syndrome in preclinical studies of biologically active substances, in diet therapy and the enrichment of specialized food products with them.