Abstract
Adipose tissue is central to energy homeostasis and may be affected by dietary restrictions in both the individual and its offspring. This study investigated the impact of nutritional protein restriction (NPR) in prepubertal ewes on the adipose tissue transcriptome of their offspring, which included lambs with high and low carcass fat deposition levels. RNA-Seq was performed on perirenal fat from 8 lambs born to protein-restricted mothers (MNPR group) and 8 lambs to control mothers (C group). Differential gene expression analyses were conducted between high- and low-fatness lambs within each group, followed by a meta-analysis to identify genes associated with fat deposition (Core-DEGs) and those responsive to maternal protein restriction (MNPR-responsive DEGs). Functional enrichment of the 316 Core-DEGs revealed that fatness correlated with distinct stages of adipose tissue development: Core-High DEGs were linked to vascularization, whereas Core-Low DEGs reflected ongoing adipogenesis. The 311 MNPR-responsive DEGs indicated that maternal restriction, regardless of fatness level, promotes postnatal adipogenesis and enhances the adipogenic potential of the offspring. This study provides transcriptomic evidence of the intergenerational impact of a prepubertal dietary restriction in ewes on early postnatal adipogenesis in their offspring, with translational relevance for livestock management and potential implications in other species, including humans.
Similar content being viewed by others
Introduction
Adipose tissue plays dual roles as an energy storage and endocrine organ and is critical for whole-body energy homeostasis and insulin sensitivity1. Additionally, as a metabolically active endocrine organ, adipose tissue could be used to understand the impact of stress on individuals2. Owing to its sensitivity to environmental factors, adipose tissue can be significantly affected by dietary restrictions. These effects on adipose tissue can be detected not only in individuals directly subjected to nutritional restriction but also in their offspring, both for individuals exposed during prenatal development3 and when the restriction takes place prior to conception4.
In this sense, previous studies in animal models have demonstrated that maternal diet during pregnancy influences adipocyte growth and development3,4,5. More specifically, several studies in ruminants have shown that feed restriction in pregnant females can significantly influence foetal adipose tissue development6,7,8. In addition, the effects on adipose tissue metabolism due to maternal feed restrictions may be observed not only in foetuses or newborns but also in later generations. The first evidence in this regard using sheep as an animal model and analysing more than one subsequent generation was obtained by Shasa et al.9. Their study demonstrated that maternal overnutrition/obesity (F0) through overfeeding from 60 days prior to conception until lambing can influence offspring adipose tissue and result in increased visceral adiposity, leading to greater adiposity at birth and later in life in granddaughters (F2), along with other health effects across generations (F1 and F2).
The study of intergenerational effects is quite feasible in rodents because of their short reproductive cycles and the ease of maintaining controlled experimental conditions. In ruminants, such studies are scarce, primarily because of their longer generational intervals. This limits our understanding of the effects of nutritional restrictions across generations in livestock species. However, because of their rapid growth and early fat deposition, juvenile lambs provide a unique model to study neonatal adipogenesis10. Notably, the developmental trajectory of lamb adipose tissue, including the transition from brown to white adipocytes, resembles the pattern observed in human foetal adipose tissue11 in contrast to that observed in pigs, a species that lacks brown adipose tissue12. Indeed, sheep models have been used on numerous occasions as animal models for studying the relationship between nutrition during pregnancy and obesity in future offspring13. Thus, studying the factors that influence adipose tissue development in suckling lambs enhances the fundamental biological understanding and may help address critical challenges in both human and animal health.
Currently, transcriptomic studies based on RNA sequencing (RNA-Seq) have emerged as a comprehensive and high-resolution approach for the global study of gene expression in specific tissues14,15. Using RNA-Seq on perirenal adipose tissue of suckling lambs, our research group has previously identified biological mechanisms underlying the regulation of fat deposition at early ages in sheep16,17,18. Interestingly, the transcriptomic analysis conducted by Alonso-García et al.16 revealed that the different perirenal fat deposition levels in suckling lambs are associated with different developmental stages of this fat depot. Considering the link between adipose tissue development and fat depot levels in suckling lambs, along with the relevance of intergenerational effects on this tissue, it was hypothesized that maternal dietary restrictions could influence fat deposition in offspring. In this context, our group evaluated the effects of dietary protein restriction during the prepubertal stage of ewe lambs (3–5 months of age) on fat deposition, carcass quality traits, and DNA methylation patterns in offspring lambs with different levels of carcass fatness19,20. Although no significant phenotypic differences were observed in the offspring due to maternal nutritional challenge19, the analysis of epigenetic marks in the perirenal adipose tissue in the same lambs revealed that the maternal dietary challenge altered their DNA methylation patterns and that there were epigenetic marks associated with the interaction between the fatness levels and the maternal challenge20.
Building on these findings, the objective of this study was to assess the impact of the above mentioned dietary protein restriction applied to prepubertal ewes19 on the perirenal adipose tissue transcriptome of their offspring at an early age (25–30 days), with a focus on individuals exhibiting high and low fat deposition levels. Fatness level is a major driver of transcriptomic variation in adipose tissue and can mask more subtle, biologically relevant signals of maternal dietary restriction. To address this, we conducted a meta-analysis to evaluate the potential intergenerational effects of maternal protein restriction across two fatness categories. By controlling for fatness-related variability, this approach reduces categorical noise and enhances the resolution of transcriptomic differences specifically associated with the intergenerational nutritional challenge. This approach aims to offer new insights into the intergenerational effects of early-life nutrition in ruminants, an area where multigenerational studies remain limited.
Materials and methods
Early-life nutritional protein restriction challenge
The transcriptomic study described here began with the early-life nutritional protein restriction (NPR) carried out on the dams of the lambs analysed here. This nutritional challenge was previously described by Pelayo et al.21. Briefly, 40 Assaf ewe lambs were acquired by our research group from the same flock at 2 months of age and were managed and fed under standard commercial conditions at the Instituto de Ganadería de Montaña (IGM, León, Spain) until they were 3.5 months old. At this point, at their prepubertal age, they were divided into two groups with two different diets for 64 days: a Control group (n = 20), receiving the standard diet, and an NPR group (n = 20), receiving the same diet without soybean meal, which implies a 42.3% reduction in protein in the diet (Supplementary Table 1). After this restriction period, all the ewes were managed in a single group and fed the same diet until their first lambing (15 months old). Ewes were weighed periodically throughout the experiment, with eight measurements covering the different stages of the study. Only significant differences in body weight between the Control and NPR groups were found at the end of the restriction period (see Table S1 from Pelayo et al.21). The nutritional challenge performed in the ewes was approved by the Ethics Committee of the IGM Centre (León, Spain) (Reference 100,102/2018–1) and the corresponding department of the Junta de Castilla y León regional government (Resolution 03/08/2018, Agriculture and Livestock Department, Junta de Castilla y León, Spain).
Lamb selection and phenotypic measurements
From the progeny born at the IGM Centre from the first-parity ewes previously described, a total of 34 male suckling lambs were initially selected for the present study. Half of them were born to Control ewes (referred to as C lambs; n = 17), and the other half were offspring of NPR ewes (referred to as MNPR lambs; n = 17). These lambs received colostrum from their dams, as detailed by Mateo et al.19, and subsequently, they were fed ad libitum with a milk replacer (Ovilac 60; Calfvet®) to avoid potential maternal effects. In accordance with standard commercial criteria, when the 34 lambs reached 10 to 12.5 kg in vivo weight, and at a maximum of 35 days of age19, they were sacrificed in a commercial slaughterhouse. The management, transport, and sacrifice of the animals were in accordance with Spanish and EU legislation [Spanish Laws 32/2007, 6/2013 and RD 37/2014; Council Regulation (EC) 199/2009]. On the basis of this legislation, euthanasia was not used, but the animals were subjected to mandatory prestunning before sacrifice to avoid suffering. According to the Research Ethics Committee of the University of León, formal ethical approval was not necessary for the study performed on the suckling lambs. In addition, the experimental design and the analysis performed in the current study are in accordance with ARRIVE guidelines22.
At 24 h postmortem, the 34 lamb carcasses were split into two halves (average slaughter age: 24.85 ± 5.25 days) to measure different carcass quality traits, as Mateo et al.19 reported. Because the present work focused on the perirenal fat transcriptome, measurements of the kidney knob and channel fat percentage in the half carcass (KKCF) were used to rank the animals. Hence, within each group (C and MNPR), the four animals with the highest KKCF values and the four animals with the lowest KKCF values were selected for RNA sequencing. The distribution of the studied animals according to the categories of the two factors considered (the MNPR and the KKCF groups) is detailed in Supplementary Tables 2 and summarized in Fig. 1a.
Overview of the RNA-Seq analysis pipeline. (a) Table showing the distribution of 16 samples categorized into two perirenal fatness level (KKCF, from kidney knob and channel fat) groups: High-KKCF (blue) and Low-KKCF (yellow). Each group includes offspring from mothers subjected to two prepubertal dietary conditions: Control (C, green) and Maternal Nutritional Protein Restriction (MNPR, red). (b) Bioinformatics workflow applied to RNA-Seq data, including mapping (STAR), quantification (RSEM), differential expression analysis (DEA; DESeq2) between High-KKCF and Low-KKCF independently for each maternal dietary condition (Control and MNPR), meta-analysis (metaRNASeq), and over-representation analysis (ORA) using clusterProfiler.
RNA extraction
For the 16 male lambs selected for the transcriptomic study, after the sacrifice, perirenal fat samples were collected for RNA extraction using an RNA-stabilization solution (Life Technologies’ Ambion RNAlater) and kept for 24 h at 4 °C. Then, samples were removed from the stabilization solution and were frozen at -80 °C., The RNA extraction was performed using the miRNeasy Mini Kit (Qiagen) adapted for use in adipose tissue. The RNA integrity of the samples was assessed using the Agilent 2100 Bioanalyzer equipment (Agilent Technologies), and RIN values over 7.2 were obtained. Later, cDNA libraries were constructed and sequenced at Novogene in Cambridge (UK). The 16 stranded paired-end libraries were obtained with the UltraTM RNA Library Prep Kit (NEBNext®), using a minimum of 0.8 µg of RNA per sample. Subsequently, sequencing was performed on an Illumina NovaSeq 6000 system, producing stranded paired-end reads of 150 pb to a minimum depth of 30 million paired-end reads. The raw transcriptomic datasets derived from sequencing are already available at the BioStudies repository, under accession number E-MTAB-11839 for the Control samples, and under accession number E-MTAB-15012 for the MNPR samples.
Bioinformatic analyses
The step-by-step bioinformatics workflow applied in this study to the 16 perirenal fat transcriptome samples is depicted in Fig. 1b. A detailed description of the main steps of this workflow is given below.
Alignment of transcripts and expression quantification
The raw sequences obtained were aligned to the reference sheep genome available at NCBI (ARS-UI_Ramb_v2.0) using STAR (v.2.7.6a)23 and applying the following parameters: --outFilterType BySJout for reducing the number of ”spurious” junctions, --quantMode TranscriptomeSAM to obtain an output file with alignments translated into transcript coordinates that can be used for subsequent quantification of the expression with RSEM, --outWigType bedGraph to obtain a bed Graph output format and --outWigStrand Stranded to indicate that our RNA-Seq data was stranded.
To quantify the expression of each transcript, we used RSEM (v.1.3.3)24 using the following parameters: --bam --no-bam-output to specify that input was a bam file containing reads aligned to the transcriptome, while preventing the generation of an alignment file at this stage, --estimate-rspd to estimate the read start position distribution, --calc-ci to calculate 95% credibility intervals and posterior mean estimates, --seed 12,345 to set the seed for the random number generators used in calculating posterior mean estimates and credibility intervals, and --paired-end --forward-prob 0 to indicate that our RNA-Seq data is paired-end and stranded. For each sample, the identified transcripts were quantified without normalisation (raw counts).
Differential expression analyses and meta-analysis
First, two differential expression analyses (DEAs) between the transcriptome samples of the High-KKCF and Low-KKCF animals were performed, one within the Control animals (Control-DEA) and one within the MNPR animals (MNPR-DEA). For these analyses, the raw counts obtained from RSEM for the 16 transcriptome samples were imported into the R environment using the tximport R package (v. 1.28.0)25. All the genes with a raw count below 10 in four or more samples were subsequently removed.
The DEAs were performed with DESeq2 (v.1.40.2)26 and included two variables: the perirenal fatness level group as a factor, with two categories, High-KKCF and Low-KKCF, and the age at slaughter (sAge) as a continuous variable. To be included as a continuous covariate, the sAge was standardized prior to inclusion in the DESeq2 model by centring and scaling it using the scale() function in R, with the parameters center = TRUE and scale = TRUE, to facilitate model convergence and interpretability of the continuous covariate. The classification of all the animals for the perirenal fatness group and the sAge of each animal are provided in Supplementary Table 2. sAge was included in the model because of the significant dependence observed between the KKCF variable and sAge (p value = 0.0133). DESeq2 was also used to normalize the gene counts by correcting them for library size and RNA composition bias and to assess dispersion through biological replicates.
After the DEAs, the adjusted p value (padj) corrected using the methodology proposed by Benjamini and Hochberg27 and the logarithmic fold change (log2FC) values obtained for each gene within each DEA were combined in a meta-analysis conducted using the R package metaRNASeq (v.1.0.7)28. The genes were categorized into two groups: High-KKCF and Low-KKCF, according to the sign of the log2FC value for each gene. For each comparison within the Control and MNPR groups, genes whose expression was higher in the High-KKCF group had a positive log2FC, whereas genes whose expression was higher in the Low-KKCF group had a negative log2FC. The meta-analysis approach was performed using two statistical methods—Fisher’s method and the inverse normal combination technique (INCT)—and only those genes commonly identified as significantly differentially expressed by the two methods (padj < 0.05 in Fisher’s and INCT methods) were considered differentially expressed genes (DEGs). Afterwards, we calculated the Integration-driven Discovery Rate (IDR) and the Integration-driven Revision Rate (IRR), which quantify the proportion of DEGs identified in the meta-analysis that were not detected in any individual DEAs and the proportion of DEGs identified in the individual DEA that were absent from the meta-analysis.
The DEGs identified by the meta-analyses were subsequently divided into two lists. The first list comprised genes with consistent log2FC signs (+/+ or -/-) across the two DEAs. These genes were designated “Core DEGs” because the expression pattern appeared to be independent of the maternal challenge, meaning that they were overexpressed in the same fatness level group (High- or Low-KKCF) in the two considered DEAs. Within this first list, genes were classified as Core-High or Core-Low DEGs according to the group in which they showed higher expression in both comparisons (Fig. 2a).
Schematic representation of general changes in expression levels of differentially expressed genes (DEGs) categorized as (a) Core-DEGs and (b) Maternal Nutritional Protein Restricted (MNPR)-responsive DEGs. Boxplots depict expression levels across different conditions: green boxes represent DEGs in the Control group, with darker and lighter shades indicating High- and Low-kidney knob and channel fat (KKCF) categories, respectively, while red boxes represent DEGs in the MNPR group with the same shading scheme. Connecting lines indicate expression changes: blue lines denote Core-DEGs with consistent directional changes between High- and Low-KKCF groups in both Control and MNPR conditions, whereas the purple line represents MNPR-responsive DEGs, which exhibit opposite expression trends between Control and MNPR groups.
The second gene list defined from the results of the meta-analysis included genes exhibiting opposite log2FC signs (+/- or -/+) between the two DEAs. These genes were considered to have opposite gene expression patterns between the High- and Low-KKCF groups because of the maternal challenge effect and were labelled “MNPR-responsive DEGs” (i.e., a gene with lower expression in the High-KKCF group in the Control-DEA and a higher expression in the High-KKCF group for the MNPR-DEA; see Fig. 2b). With respect to these DEGs and considering the expression pattern of DEGs in the Control-DEA group as a reference, we defined two distinct subcategories that represent different expression patterns because of the interaction of the maternal challenge effect and the developmental stage of the adipose tissue (Fig. 2b): (1) DEGs that were overexpressed in the Control-DEA group in the Low-KKCF group. In contrast, in MNPR-DEA, these genes were overexpressed in the High-KKCF group. Henceforth, these genes should be designated “Inverse-High” and (2) in the Control-DEA, DEGs that exhibited higher expression in the High-KKCF animals, and in contrast, in the MNPR-DEA, these genes were overexpressed in the Low-KKCF animals; henceforth, these DEGs will be referred to as “Inverse-Low”.
To better understand how the maternal nutritional restriction studied here could influence the biological pathways of the perirenal adipose tissue of offspring at an early stage, the clusterProfiler R package (v.4.8.3)29 was used to perform Gene Ontology (GO) overrepresentation analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway overrepresentation analysis for each of the four lists of DEGs resulting from the meta-analysis approach used here. For these analyses, the human annotation org.Hs.db (v.3.17.0)30 database in R was used. Benjamini and Hochberg multiple testing correction27 was used, considering adjusted p values less than 0.05 as significant. The terms that met this threshold were considered significantly enriched if they were supported by a minimum of five DEGs.
Results
Alignment and expression quantification
For the 16 fat transcriptomes analysed here, on average, more than 31 million (31,619,495.89 ± 982,226.62, mean ± SD) paired-end reads were generated per sample. Among them, more than 90% of the reads aligned uniquely to the ovine reference genome (ARS_UI Ramb v2.0), accounting for almost 30 million (28,947,618.22 ± 1,471,925.39, mean ± SD) paired-end reads per sample on average.
Differential expression analyses and meta-analysis
A total of 30,186 genes were considered expressed and then analysed in the two DEAs. Among them, 284 and 860 genes had padj < 0.05 between the High- and Low-KKCF groups for the Control-DEA and MNPR-DEA, respectively. The subsequent meta-analysis performed to integrate the results across the two considered contrasts provided a comprehensive overview of gene expression patterns. Exploiting the two meta-analysis methods, 915 and 641 genes were detected as DEGs according to Fisher’s method and INCT, respectively (Fig. 3a). The IDRs were 14.43 for Fisher’s method and 21.68 for INCT. The IRR values were 29.40 for Fisher’s method and 54.73 for INCT (Fig. 3a).
Summary of the meta-analysis results using Fisher’s Method and the Inverse Normal Combination Technique (INCT). (a) Table summarizing the number of differentially expressed genes (DEGs) identified by each method, along with integration-driven discoveries (IDD) – DEGs detected in the meta-analysis but not in individual studies – and DEGs lost in the meta-analysis compared to individual analyses (Loss). The table also presents the Integration-Driven Discovery Rate (IDR) and the Integration-Driven Revision Rate (IRR), representing the proportions of IDD and Loss DEGs, respectively. (b) Venn diagram illustrating the overlap of DEGs identified between High kidney knob and channel fat (KKCF) and Low-KKCF lambs across individual differential expression analyses (DEA) performed on Control (blue) and Maternal Nutritionally Protein Restriction (MNPR; purple) groups, alongside the two meta-analysis approaches: Fisher’s Method (pink) and INCT (green). DEGs identified by both meta-analysis methods are highlighted in red.
A comparison of the two meta-analysis methods revealed that 627 DEGs were common between the two procedures (highlighted in Fig. 3b by the red diamond). Among the common DEGs detected between the two methods, 316 had the same sign of the log2FC value across the two analyses and were classified as Core-DEGs (Table 1). Among these Core DEGs, 178 were more highly expressed in High-KKCF animals, whereas 138 were more highly expressed in Low-KKCF animals (Table 1).
In contrast, the meta-analysis revealed 311 MNPR-responsive DEGs, indicating alterations in expression between High- and Low-KKCF animals when the Control and MNPR-DEAs were compared. Among these genes, 124 Inverse-Low DEGs and 187 Inverse-High DEGs were identified (Table 1). These MNPR-responsive DEGs are listed in Supplementary Table 3, together with the log2FC values and the group showing the highest expression for the two DEAs under consideration.
Enrichment analysis
Enrichment analysis of the Core-High DEGs revealed 58 enriched terms. Among them, 56 belonged to the BP ontology, and the rest belonged to the CC ontology (Supplementary Table 4). Overall, the three most enriched terms were “endothelium development”, “lymph vessel development” and “endothelial cell differentiation” for BP ontology (FDR = 2.3 × 10− 2, 9, 5 and 8 DEGs, respectively) and “platelet alpha granule lumen” (FDR = 2.1 × 10− 2, 5 DEGs) for CC ontology. The most enriched terms identified for the list of Core-High DEGs were related to vasculogenesis, with the VEGFA and VEGFC genes being associated to all of them. These genes appear in the term network linking BP ontology terms related to vascular endothelial growth factor with the most enriched terms for the CC ontology, all related to the vesicle lumen (Fig. 4).
Network visualization of enriched Gene Ontology (GO) terms and their associated Core differentially expressed genes (DEGs). Blue nodes represent genes and GO terms and DEGs overexpressed in High kidney knob and channel fat (KKCF) lambs (Core-HighKKCF), while yellow nodes correspond to DEGs and GO terms overexpressed in Low-KKCF lambs (Core-LowKKCF). The size of each enriched GO term node is proportional to the number of supporting Core-DEGs.
On the other hand, 68 enriched terms were detected for the Core-Low DEGs, with 56 terms belonging to the BP category, 10 to the MF category and 2 to the CC category (Supplementary Table 4). The three most enriched terms for the Core-Low DEGs were “cellular catabolic process”, “organic acid catabolic process” and “carboxylic acid catabolic process” (FDR = 1.56 × 10− 16, 22, 22 and 20 DEGs, respectively), all belonging to the BP category. As shown in Fig. 4, the Core-Low DEGs were associated with processes related to different types of catabolic processes (“cellular molecule catabolic process”, “organic acid catabolic process”, “monocarboxylic acid catabolic process”, etc.).
Independent functional enrichment analyses were conducted for the two groups of genes comprising the 311 MNPR-responsive DEGs (see Supplementary Table 5). In the analysis of the Inverse-High DEGs, 46 significantly enriched terms were identified, the majority of which belonged to the BP category (45) with one belonging to the MF category. The most enriched term was “sterol biosynthetic process” (FDR = 5.96 × 10− 7, 10 DEGs), which belonged to the BP category. The unique enriched term for the MF ontology was “pattern recognition receptor activity” (FDR = 1.35 × 10− 2, 5 DEGs). As shown in Fig. 5, the Inverse-High DEGs were involved in cholesterol metabolism (SQLE, SCARB1, MSMO1, etc.) and were related to fatty acid metabolic processes (AACS, SCD, FASN, ELOVL6, etc.). Furthermore, some DEGs were linked to some of those terms, such as the ACLY, HMGCS1, MVD, and MVK genes (Fig. 5). With respect to the analysis of Inverse-Low DEGs, only one significantly enriched term, “response to calcium ion” (FDR = 4.11 × 10− 2, 7 DEGs), which belonged to the BP category (Supplementary Tables 5 and Fig. 5), was identified.
Network visualization of enriched Gene Ontology (GO) terms and their associated Maternal Nutrition Protein Restricted (MNPR)-responsive differentially expressed genes (DEGs). Green nodes represent GO terms and DEGs overexpressed in the Control Low-KKCF animals while in the MNPR-DEGs are overexpressed in the High-KKCF group, designated as “Inverse-HighKKCF”. The size of each enriched GO term node is proportional to the number of supporting MNPR-responsive DEGs.
In relation to the KEGG pathway overrepresentation analysis performed for the Core-DEGs, 8 and 13 significantly enriched pathways were detected for Core-High and Core-Low DEGs, respectively (Supplementary Table 6). For the Core-High DEGs, the most significantly enriched pathway was “Fluid shear stress and atherosclerosis” (FDR = 8.09 × 10− 7, 9 DEGs). Most of the significantly enriched pathways related to Core-High DEGs included “MAPK signalling pathway”, “Rap1 signalling pathway” and “Rap1 signalling pathway”. Moreover, the Core-Low DEGs were associated with “Valine, leucine and isoleucine degradation” (FDR = 1.08 × 10− 8, 9 DEGs). In terms of the Core-Low DEGs, many other pathways related to lipid metabolism (“Fatty acid degradation”) and amino acid metabolism (“Lysine degradation” and “Tryptophan metabolism”) were also detected, including pathways related to the endocrine system (“PPAR signalling pathway”).
The KEGG pathway overrepresentation analysis performed for MNPR-responsive DEGs revealed a significantly enriched pathway for each group. For Inverse-High DEGs, the pathway “Motor proteins” (FDR = 7.72 × 10− 3, 10 DEGs) was detected, whereas for Inverse-Low DEGs, the pathway “Longevity regulating pathway” (FDR = 4.64 × 10− 2, 5 DEGs) was detected.
Discussion
Understanding the factors that influence adipose tissue development is crucial because of the central role of this tissue in energy metabolism and its association with metabolic disorders such as obesity. In recent years, it has been observed that the adipose tissue of offspring is preconditioned to withstand nutritional stress comparable to that experienced by their mothers5,31,32. This is linked to the fact that the development of obesity in the next generation can be influenced by maternal status before conception, suggesting that this pathology has an intergenerational transmission33. Offspring adipose reserves can be affected by their mother’s preconception birth weight, obesity, and nutritional health34,35.
Transcriptomic differences associated with fat deposition levels in lambs – Analysis of Core-DEGs
With respect to the Core-DEGs found associated with the High- and Low-KKCF groups, the meta-analysis enabled the identification of a list of DEGs that support the distinction of the animals in two different developmental stages on the basis of the transcriptomic profile of their perirenal adipose tissue. Notably, the meta-analysis applied here (n = 16) identified 316 Core-DEGs compared with the 120 DEGs identified in the DEA analysis for the fatness factor performed in the Control animals (n = 8)16. Despite the greater complexity of this study, including the MNPR group, the meta-analysis approach allowed the exploitation of information from a larger number of samples, increasing the ability to detect DEGs associated with fatness groups and supporting the hypothesis that these groups correlate with different developmental stages of adipose tissue. Hence, among the 68 enriched terms identified for Core-Low DEGs, we highlight the “fat cell differentiation” and “fatty acid metabolic process” processes. On the one hand, “fat cell differentiation”, which represents the adipogenesis process, appears to play a pivotal role in these animals, with the involvement of genes such as UCP1, CEBPA, TRIB3, AKT2, SOD2 and SREBF1. C/EBPα, encoded by CEBPA, is a master transcription factor for adipogenesis and is among the principal regulators of this process along with PPARγ. C/EBPα is essential for the terminal differentiation of white adipocytes36 and regulates the expression of the UCP1 gene, a key mediator of nonshivering thermogenesis in brown adipose tissue (BAT)37. Moreover, TRIB3 has been identified as a key regulator in preadipocytes38, whereas the deletion of SOD2 in mouse adipocytes induces an adaptive stress response, characterized by the activation of mitochondrial biogenesis and increased mitochondrial fatty acid oxidation39. Additionally, on the basis of knockdown experiments, AKT2 appears to be essential for preadipocyte proliferation40.
With respect to the enrichment terms related to the fatty acid metabolic process, SREBF1 is the link between adipose cell differentiation and lipid metabolism. SREBF1 is closely associated with CEBPA and is expressed predominantly in two isoforms that participate at different points in the adipogenic transcription cascade. SREBF1a acts at the beginning of the cascade since its inactivation prevents the expression of transcription factors essential for later adipogenesis41. Conversely, SREBF1c inhibition results in the failure of lipid droplets to accumulate42, a process essential for the mature adipocyte phenotype. However, since our analysis evaluated gene-level expression and did not evaluate isoform-specific expression, we cannot determine which isoform was affected in this study. The detection of CEBPA and SREBF1 as Core-Low DEGs and the enriched terms detected in this group (Fig. 4) indicate that in the adipose tissue of lambs with a lower perirenal fatness level, the process of adipogenesis is ongoing. In addition, the detection of the UCP1 gene as a Core-Low DEG indicated that brown adipose tissue and mitochondrial oxidative metabolism were present.
Moreover, the detection of “Fatty acid degradation”, “Fatty acid metabolism” and “PPAR signaling pathway” KEGG pathways associated with the Core-Low DEGs (Supplementary Table 6) reinforces the idea that in the Control group, with lower levels of perirenal fat, there is very active lipid metabolism that supports ongoing adipogenesis, which is activated through PPARG signalling, despite this gene not being detected as a Core-DEG. In addition, as a result of the same KEGG pathway analysis, many metabolic pathways of different amino acids were also associated with Core-Low DEGs (Supplementary Table 6). Amino acid metabolism, particularly the breakdown of branched-chain amino acids (BCAAs), plays a significant role in adipogenesis. Specifically, the catabolism of BCAAs, including leucine, isoleucine, and valine, promotes adipocyte differentiation43. Overall, the results of this analysis reveal that the perirenal adipose tissue of these Low-KKCF animals exhibits active metabolic activity, characterized by the adipogenic processes described above.
The functional enrichment analyses reported here for the Core-High DEGs reveal the central role of vasculogenesis-related processes and the key role of VEGF-A in animals with higher perirenal fatness levels. Increased expression of VEGF-A leads to increased blood vessel density, which, in turn, enhances adipose function by improving blood perfusion, reducing adipocyte size, lowering hypoxia, and delaying adipocyte apoptosis or necrosis44. In addition, the greater statistical power of the meta-analysis presented here allowed the detection of new genes associated with vasculogenesis that had not been previously identified by Alonso-García et al.16. Among these genes, as shown in Fig. 4, we identified VEGF-C and VEGF-D, which are other angiogenic and lymphangiogenic factors that also play important roles in mediating adipose tissue inflammation associated with metabolic syndrome45.
These three genes were also linked to the MAPK, Ras, relaxin, Rap1, and TNF signalling pathways, as identified in the KEGG database (Supplementary Table 6), all of which are well known for their involvement in adipogenesis. The cAMP signalling pathway is a well-known pathway that is crucial in regulating adipocyte function, particularly in lipolysis and thermogenesis46. Other pathways that influence adipogenic gene expression include Ras signalling47,48,49, relaxin50,51 and Rap152 pathways. On the other hand, we also detected pathways that inhibit adipogenesis. MAPK activation may even antagonize adipocyte terminal differentiation53, and TNF-α is a strong negative regulator of adipogenesis through the inhibition of key transcription factors, such as PPARγ and C/EBPα54,55. Although the perirenal adipose tissue in the Control High-KKCF suckling lambs appeared fully differentiated and was primarily involved in vasculogenesis, the enrichment of these pathways may reflect residual transcriptional activity related to early adipogenesis or roles associated with vascular remodelling.
Other genes identified as Core-High DEGs were VEGF receptors. The KDR gene, also known as VEGF receptor 2, is a pivotal proangiogenic factor necessary for adipose tissue expansion and plays an essential role in regulating angiogenesis within adipose tissue56. Several studies have demonstrated that the knockdown of VEGFR1, also known as FLT1, facilitates the browning of WAT by upregulating UCP1 expression, leading to elevated thermogenesis and reducing the diameter of adipocytes57,58. The elevated expression of these two receptors in the High-KKCF animals, independent of the maternal challenge group, supports the hypothesis that the elevated vasculogenic activity in this tissue is due to the expansion of the adipose tissue in the perirenal depot.
One other gene of particular interest within the enriched gene-term network obtained for the Core-High DEGs was EGR3 (Fig. 4). EGR3 plays a pivotal role in the inhibition of adipogenesis in cultured cells59. This key role underscores that at this point, adipogenesis is not actively occurring, or it is even being suppressed, compared with animals with a lower percentage of fat in the same depot, whose adipocytes have a transcriptomic profile characteristic of adipogenesis.
Overall, the Core-High DEG support that, in the early stages of growth, adipose tissue from the animals with higher levels of perirenal fat is in a more advanced developmental stage. This is characterised by the expansion of the tissue supported by the development of a very dense network of capillaries, as was already observed by Alonso-García et al.16.
Effects of protein restriction in ewe lambs on the adipose tissue of their offspring – Analysis of MNPR-responsive DEGs.
In this study, we detected genes whose expression patterns changed from expectations based on their KKCF values because of the intergenerational effects of maternal dietary restriction. The interpretation of these genes, referred to as MNPR-responsive DEGs, remains complex.
The functional enrichment analysis for the Inverse-Low DEGs revealed that only the “response to calcium ion” GO term was significantly enriched (Fig. 5 and Supplementary Table 5). Calcium ions play a significant role in fat cell development. Changes in calcium levels, both intra- and extracellularly, can either promote or inhibit adipogenesis. Low extracellular calcium levels may promote adipogenesis and brown adipocyte differentiation, potentially by increasing the expression of brown adipocyte-specific genes such as UCP160,61. On the other hand, high extracellular calcium levels could inhibit adipogenesis through the activation of calcium-sensing receptors and the ERK signalling pathway, which can inhibit the activity of C/EBPβ, a key transcription factor in adipogenesis62,63. Moreover, intracellular calcium levels also regulate adipocyte metabolism. An increase in the intracellular calcium concentration can promote lipogenesis and adipocyte marker expression later in the differentiation process62. Some of the Inverse-Low DEGs associated with this term are also closely related to the regulation of adipogenesis. For instance, AQP3 is highly important for both adipogenic differentiation and proliferation in porcine intramuscular adipocytes in vitro64. Another transmembrane protein related to the regulation of adipocyte function is TRPM2, an ion channel that allows extracellular Ca2+ influx into the cytosol. Benzi et al.65 recently reported that this protein plays a fundamental role in brown adipose tissue activation and white adipose tissue browning. Another gene closely related to adipocyte development is THBS1, which encodes thrombospondin 1 (TSP-1). The expression of THBS1 was studied in subcutaneous and visceral adipose tissue from nondiabetic subjects, surgical patients, and patients with glucose intolerance65. This transcriptomic study revealed that TSP-1 is an adipokine that is expressed at high levels in obese and insulin-resistant subjects and is highly correlated with adipose inflammation66. Moreover, these findings were later confirmed in murine models, suggesting a role for TSP-1 in mediating obesity-associated inflammation, which contributes to weight gain, adipose growth, and the pathogenesis of metabolic dysfunction67. The remaining Inverse-Low DEGs associated with the response to calcium ions have been studied in the context of mitochondrial function (SLC25A23)68, growth and development specific to bovine skeletal muscle (CAPN3)69, regulation of neuronal excitability (KCNH1)70 and their involvement in cancer development (CPNE8)71. However, none of these genes are associated with adipogenesis.
On the other hand, the Inverse-High DEGs provided substantial support for significantly enriched terms already highlighted in Alonso-García et al.16 in the Low-KKCF group when only the Control animals were considered (“cholesterol metabolic process”, “secondary alcohol biosynthetic process”, “cholesterol biosynthetic process”, “purine nucleoside biphosphate metabolic process”, and “ribonucleoside biphosphate metabolic process”) (Supplementary Table 4, Fig. 5). Interestingly, many of the Inverse-High DEGs participate in several key points of the cholesterol biosynthesis process, such as HMGCS1, MVK, MVD, IDI1, FDPS, LSS, TM7SF2, MSMO1, and DHCR772,73,74. The observation that all stages of cholesterol synthesis were affected by MNPR in High-KKCF animals suggests that this pathway plays a pivotal role in adipose tissue metabolism and is intricately linked to the animal stress response. A very recent comparative analysis of the transcriptome of piglets’ subcutaneous adipose tissue revealed a reduction in the expression of genes involved in the early stages of the cholesterol biosynthetic pathway (ACLY, MVK, IDI1, and HMGCS1) in older piglets (21 days old) compared with younger piglets (1 day old)75. Additionally, an in vitro study comparing the preadipocyte stage (Day 0) and the differentiated adipocyte stage (Day 13) revealed that, among other genes, IDI1, MSMO1, and FDPS were downregulated on Day 1376. These results suggest that in MNPR High-KKCF lambs, adipogenesis is still active. In contrast, the expression pattern observed in the Control High-KKCF animals indicates a more advanced stage of adipose tissue development, with potential signs of adipogenesis inhibition, although it is not possible to determine whether this process has been completed.
In addition to the upregulated genes that encode enzymes that participate in cholesterol biosynthesis, the INSIG1 gene was also detected as an Inverse-High DEG (Fig. 5). INSIGs (insulin-induced genes) are considered to play pivotal roles in cholesterol metabolism77,78. In vitro studies have demonstrated that INSIG1 promotes sterol-dependent retention of SREB cleavage-activating protein, thus preventing excessive lipid accumulation and adipogenesis in adipocytes79,80. In addition, a decline in INSIG1 expression has been documented during preadipocyte differentiation in sheep81. The overexpression of INSIG1 in the studied lambs may represent an adaptive mechanism triggered by MNPR stress, programming the offspring’s adipose tissue to restrict excessive fat storage in response to a nutrient-scarce environment.
The other significantly enriched terms detected for the Inverse-High DEGs were related to acyl-CoA and thioester metabolic processes (Fig. 5 and Supplementary Table 4). These enriched terms were also reported by Alonso-García et al.16 for the Low-KKCF DEGs. The conversion of free fatty acids into fatty acyl-coenzyme A (acyl-CoAs) is required for almost all fatty acid metabolic processes, and two main types of enzymes mediate the synthesis and breakdown of acyl-CoA: acyl-CoA synthetases (ACSs) and acyl-CoA thioesterases (ACOTs). Two enzymes in the first of these groups, ACSM3 and ACSS2, were detected among the Inverse-High DEGs in this study. In goats, ACSS2 promotes fat storage and utilization on the basis of nutritional states82 and serves as a switch for PPARγ activity, increasing adipose tissue plasticity83. Moreover, the overexpression of ACSS2 increases PPARγ, CEBPα, and FABP4 expression, driving adipocyte proliferation and growth84.
Furthermore, other genes identified as Inverse-High DEGs, such as ACLY, GPAM, ELOVL6, DGAT2, and FASN, which are also associated with acyl-CoA and thioester metabolism (Fig. 5), play crucial roles in adipose tissue function and were upregulated in MNPR-High KKCF animals in this study. In contrast, in the previous study focused only on the Control animals, Alonso-García et al.16 reported that these genes were upregulated in Low-KKCF animals. All these genes are involved in lipid metabolism, as they regulate lipogenesis, the synthesis of cholesterol substrates, the synthesis of triglycerides, and lipid accumulation. These genes are also related to the development of adipose tissue by promoting the growth of adipocytes and improving adipocyte differentiation85,86,87,88,89,90,91,92.
The collective identification of all the mentioned genes as Inverse-High DEGs suggests that these animals were in a phase of metabolically active cholesterol biosynthetic pathways akin to that observed in younger animals75 or in a phase of differentiating adipocytes76. Hence, these genes appear to have dual functions and participate in both the biosynthesis of lipids and the regulation of adipocyte differentiation.
Overlap between differentially expressed genes and differentially methylated regions
The transcriptomic analyses reported here aimed to present a complementary layer of information to better understand the epigenetic signals identified in the adipose tissue of the same lambs20. That study revealed a clear interaction between MNPR and fat deposition level at the epigenomic level. By contrasting the results, we found that four DEGs identified by our meta-analysis harboured differentially methylated regions (DMRs), highlighting dual evidence of epigenetic and transcriptional regulation. Two of these three DEGs (LOC101103174 and ULK4) were classified as Core-DEGs on the basis of the meta-analysis. Even though the DMRs detected by Fonseca et al.20 for these genes were associated with the maternal challenge factor, no impact of MNPR was discerned for these two genes at the transcriptomic level in our study.
The third gene harbouring a DMR, according to Fonseca et al.20 and identified as a DEG in this study, was the PER2 gene. The DMR found in this gene was associated with the fatness level factor in the methylome analysis20, whereas in the present study, it was detected as an Inverse-Low DEG, which means that MNPR increases PER2 expression in Low-KKCF animals. PER2 has been linked to adipose tissue development and lipid metabolism, as its expression parallels that of adipogenic markers such as PPARγ and C/EBPα during 3T3-L1 cell differentiation, which suggests that this gene plays a role in lipid metabolism regulation through PPARγ93. On the basis of these results, this gene may play an important role in the response to dietary stresses and is also subjected to intergenerationally inherited regulation. In this way, PER2 emerges as a strong candidate gene for further investigation to determine whether stress experienced during the growth stage of the future mother may have long-term effects on the offspring.
The fourth gene, SYNE3, was identified as an MNPR-responsive DEG in the transcriptomic analysis. Furthermore, Fonseca et al.20 detected two DMRs within its sequence. One of the DMRs harbouring SYNE3 was associated with the interaction between maternal challenge and the fatness level factor20. The spectrin repeat-containing nuclear envelope family member 3 (SYNE3) gene is responsible for encoding nesprin-3, a crucial component of the linker of the cytoskeleton and nucleoskeleton (LINC) complex94. Yang et al.95 reported that disrupting the physiological balance of the LINC complex could affect the adipogenesis of human adipose-derived stem cells in vitro. These findings suggest a potential link between adipogenesis regulation and reorganisation of the LINC complex. In our meta-analysis, SYNE3 was identified as an Inverse-Low DEG, suggesting that maternal challenge may lead to its upregulation. This increase in expression could reflect ongoing adipogenic processes in these animals, potentially influenced by MNPR. However, despite this hypothesis, no additional evidence linking SYNE3 to adipose tissue development has been reported in the literature. Therefore, further research is needed to clarify its potential role in adipocyte differentiation and development.
Global overview of the intergenerational impact of early-life protein restriction on offspring adipose tissue
Overall, our results for MNPR-responsive genes suggest that the protein restriction suffered by ewe lambs during the prepubertal stage influenced the development of the perirenal adipose tissue of their offspring, promoting an increase in adipogenesis during the early postnatal growth stage. Moreover, this effect was especially evident in the later stage of adipose tissue development considered in this study. While no phenotypic differences in fat deposition levels were detected in the offspring of protein-restricted mothers19, our transcriptome analyses suggest that these animals may have an enhanced potential for fat accumulation. These findings are consistent with previous studies in rats, which have linked maternal protein restriction during pregnancy or lactation to increased visceral adiposity in adult male offspring31 and elevated proliferation rates of adipocyte precursors96. Additionally, some studies have reported intergenerational effects, such as increased lipogenesis in F2 offspring following a low-protein diet during the F1 lactation period97. Furthermore, by exploiting transcriptomic information, our study further improves the understanding of the mechanisms underlying the increased metabolic and cardiovascular risks reported in the offspring of mothers exposed to nutritional deficiencies, such as those observed in humans following the Dutch famine98.
Building on these observations, future research should investigate whether the observed effects in the suckling lambs studied here originate from prior alterations in their mothers’ adipose tissue. Interestingly, specific nutritional challenges, such as early-life protein restriction followed by catch-up growth, have been associated with increased adipose tissue proliferation99, which could explain the observations in this study, specifically for animals with a higher percentage of perirenal fat. The ewe lambs subjected to prepubertal protein restriction in this study, who, according to Pelayo et al.21, likely underwent compensatory growth following nutritional prepubertal restriction, may have experienced adipose tissue modifications that were subsequently transmitted to their offspring. Further research on the adipose tissue of these females is needed to validate this hypothesis and could provide a deeper understanding of the role of maternal nutrition in adipocyte development during early life, with potential implications for animal management practices and human health.
Conclusions
This study confirms that maternal protein restriction during prepubertal growth in sheep alters the transcriptomic profile of adipose tissue in offspring, enhancing the potential for fat deposition by increasing adipogenic activity and lipid metabolism across two different developmental stages. This adaptive response suggests a lasting molecular footprint that promotes fat accumulation through noncanonical pathways early in adipose tissue development. Moreover, adipocyte proliferation is hypothesised to remain uninhibited in later stages, contrary to what would be expected under normal conditions.
Notably, the intergenerational consequences of protein restriction in prepubertal ewe lambs mirror those observed when restriction occurs during pregnancy or lactation in rodent models. These findings underscore the role of adipose tissue as a key target of nutritional influences and reinforce its function in metabolic adaptation. They also suggest that this adaptive response may be conserved across mammalian species, potentially serving as a mechanism to prepare the restricted animal — or its offspring — for environmentally limited conditions.
Beyond livestock management, this research provides a model for studying the early-life origins of obesity and metabolic disorders in humans, potentially aiding in identifying risk factors for conditions such as metabolic syndrome before they manifest in adulthood.
Data availability
The raw sequencing datasets generated and analyzed during this study are publicly available in the BioStudies repository. The perirenal adipose tissue transcriptome sequences of the Control group animals can be accessed under accession number [E-MTAB-11839](https:/www.ebi.ac.uk/biostudies/ArrayExpress/studies/E-MTAB-11839?query=E-MTAB-1183) , whereas the transcriptome sequences of the MNPR group animals can be accessed under accession number [E-MTAB-15012](https:/www.ebi.ac.uk/biostudies/arrayexpress/studies/E-MTAB-15012) .
References
Ning, T. & Wang, Q. A. Adipose tissue remodeling and adipose precursors. In Reference Module in Biomedical Sciences (Elsevier, doi:https://doi.org/10.1016/B978-0-12-801238-3.65508-3. (2018).
Brydon, L. Adiposity, leptin and stress reactivity in humans. Biol. Psychol. 86, 114–120 (2011).
Lecoutre, S. & Breton, C. Maternal nutritional manipulations program adipose tissue dysfunction in offspring. Front. Physiol. 6, 1–7 (2015).
Lecoutre, S., Petrus, P., Rydén, M. & Breton, C. Transgenerational epigenetic mechanisms in adipose tissue development. Trends Endocrinol. Metabolism. 29, 675–685 (2018).
Lecoutre, S. & Breton, C. The cellularity of offspring’s adipose tissue is programmed by maternal nutritional manipulations. Adipocyte 3, 256–262 (2014).
Fernandes, A. C. et al. Metabolomics changes in meat and subcutaneous fat of male cattle submitted to fetal programming. Metabolites 14, 9 (2023).
Muhlhausler, B. S., Duffield, J. A. & McMillen, I. C. Increased maternal nutrition increases leptin expression in perirenal and subcutaneous adipose tissue in the postnatal lamb. Endocrinology 148, 6157–6163 (2007).
Rattanatray, L. et al. Impact of maternal periconceptional overnutrition on fat mass and expression of adipogenic and lipogenic genes in visceral and subcutaneous fat depots in the postnatal lamb. Endocrinology 151, 5195–5205 (2010).
Shasa, D. R. et al. Multigenerational impact of maternal overnutrition/obesity in the sheep on the neonatal leptin surge in granddaughters. Int. J. Obes. 39, 695–701 (2015).
Rodgers, A. Sferruzzi-Perri, A. N. Developmental programming of offspring adipose tissue biology and obesity risk. Int. J. Obes. 45, 1170–1192 (2021).
Gemmell, R. T. & Alexander, G. Ultrastructural development of adipose tissue in foetal sheep. Aust J. Biol. Sci. 31, 505–516 (1978).
Gómez-Prado, J. et al. Thermoregulation mechanisms and perspectives for validating thermal windows in pigs with hypothermia and hyperthermia: an overview. Front. Vet. Sci. 9, (2022).
Metges, C. C. Early nutrition and later obesity: animal models provide insights into mechanisms. in 105–112 (2009). https://doi.org/10.1007/978-1-4020-9173-5_11
Deshpande, D. et al. RNA-seq data science: from Raw data to effective interpretation. Front. Genet. 14, (2023).
Wang, Z., Gerstein, M. & Snyder, M. RNA-Seq: a revolutionary tool for transcriptomics. Nat. Rev. Genet. 10, 57–63 (2009).
Alonso-García, M. et al. Transcriptome analysis of the perirenal fat from Assaf suckling lamb carcasses showing different levels of kidney knob and channel fat. Front. Genet. https://doi.org/10.3389/fvets.2023.1150996 (2023).
Alonso-García, M. et al. A meta-analysis approach for annotation and identification of LncRNAs controlling perirenal fat deposition in suckling lambs. Anim. Biotechnol. 35, (2024).
Suárez-Vega, A. et al. Early adipose deposits in sheep: comparative analysis of the perirenal fat transcriptome of Assaf and Churra suckling lambs. Anim. Genet. 49, 605–617 (2018).
Mateo, J. et al. Relationships between growth-related traits with carcass and meat quality traits in Spanish Assaf suckling lambs. Livest. Sci. 269, 105173 (2023).
Fonseca, P. A. S. et al. Intergenerational impact of dietary protein restriction in dairy Ewes on epigenetic marks in the perirenal fat of their suckling lambs. Sci. Rep. 13, 4351 (2023).
Pelayo, R. et al. Influence of a temporary restriction of dietary protein in prepubertal Ewe lambs on first lactation milk traits and response to a mammary gland inflammatory challenge. Res. Vet. Sci. 159, 57–65 (2023).
Percie du Sert. The ARRIVE guidelines 2.0: updated guidelines for reporting animal research. PLoS Biol. 18, e3000410 (2020).
Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21 (2013).
Li, B., Dewey, C. N. & RSEM Accurate transcript quantification from RNA-Seq data with or without a reference genome. BMC Bioinform. 12, (2011).
Soneson, C., Love, M. I. & Robinson, M. D. Differential analyses for RNA-seq: transcript-level estimates improve gene-level inferences. F1000Res 4, 1521 (2015).
Love, M. I., Huber, W. & Anders, S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 15, (2014).
Benjamini, Y. & Hochberg, Y. Controlling the false discovery rate: A practical and powerful approach to multiple testing. J. Royal Stat. Soc. Ser. B (Methodological). 57, 289–300 (1995).
Rau, A., Marot, G. & Jaffrézic, F. Differential meta-analysis of RNA-seq data from multiple studies. BMC Bioinform. 15, 1–10 (2014).
Wu, T. et al. ClusterProfiler 4.0: A universal enrichment tool for interpreting omics data. Innov. 2, (2021).
Carlson M. org.Hs.eg.db: Genome wide annotation for Human. R package version 3.8.2. Preprint at (2019).
Guan, H. et al. Adipose tissue gene expression profiling reveals distinct molecular pathways that define visceral adiposity in offspring of maternal protein-restricted rats. Am. J. Physiology-Endocrinology Metabolism. 288, E663–E673 (2005).
Dumortier, O. et al. Age-Dependent control of energy homeostasis by brown adipose tissue in progeny subjected to maternal Diet–Induced fetal programming. Diabetes 66, 627–639 (2017).
Haire-Joshu, D. & Tabak, R. Preventing obesity across generations: evidence for early life intervention. Annu. Rev. Public. Health. 37, 253–271 (2016).
Power, C., Kuh, D. & Morton, S. From developmental origins of adult disease to life course research on adult disease and aging: insights from birth cohort studies. Annu. Rev. Public. Health. 34, 7–28 (2013).
Bygren, L. O. Intergenerational health responses to adverse and enriched environments. Annu. Rev. Public. Health. 34, 49–60 (2013).
Wang, Q. A. et al. Distinct regulatory mechanisms governing embryonic versus adult adipocyte maturation. Nat. Cell. Biol. 17, 1099–1111 (2015).
Manchado, C. et al. CCAAT/enhancer-binding proteins α and β in brown adipose tissue: evidence for a tissue-specific pattern of expression during development. Biochem. J. 302, 695–700 (1994).
Hernández-Quiles, M. et al. The pseudokinase TRIB3 controls adipocyte lipid homeostasis and proliferation in vitro and in vivo. Mol. Metab. 78, 101829 (2023).
Han, Y. H. et al. Adipocyte-Specific deletion of manganese superoxide dismutase protects from Diet-Induced obesity through increased mitochondrial uncoupling and biogenesis. Diabetes 65, 2639–2651 (2016).
Fischer-Posovszky, P., Tews, D., Horenburg, S., Debatin, K. M. & Wabitsch, M. Differential function of Akt1 and Akt2 in human adipocytes. Mol. Cell. Endocrinol. 358, 135–143 (2012).
Ayala-Sumuano, J. T. et al. Srebf1a is a key regulator of transcriptional control for adipogenesis. Sci. Rep. 1, 178 (2011).
Aksoy, M. O., Bilinska, A., Stachowiak, M., Flisikowska, T. & Szczerbal, I. Deciphering the role of the SREBF1 gene in the transcriptional regulation of Porcine adipogenesis using CRISPR/Cas9 editing. Int. J. Mol. Sci. 25, 12677 (2024).
Green, C. et al. Branched chain amino acid catabolism fuels adipocyte differentiation and lipogenesis. Nat. Chem. Biol. 12 (1), 15–21 (2015).
Sung, H. K. et al. Adipose vascular endothelial growth factor regulates metabolic homeostasis through angiogenesis. Cell. Metab. 17, 61–72 (2013).
Karaman, S. et al. Blockade of VEGF-C and VEGF-D modulates adipose tissue inflammation and improves metabolic parameters under high-fat diet. Mol. Metabolism. 4 (2), 93–105 (2015).
Petersen, R. K. et al. Cyclic AMP (cAMP)-Mediated stimulation of adipocyte differentiation requires the synergistic action of Epac- and cAMP-Dependent protein Kinase-Dependent processes. Mol. Cell. Biol. 28 (11), 3804–3816 (2008).
Porras, A. et al. Dissociation between activation of Raf-1 kinase and the 42-kDa mitogen-activated protein kinase/90-kDa S6 kinase (MAPK/RSK) cascade in the insulin/Ras pathway of adipocytic differentiation of 3T3 L1 cells. JBC 269 (17), 12741–12748 (1994).
Murholm, M. et al. Ras signalling regulates differentiation and UCP1 expression in models of brown adipogenesis. BBA Volume. 1800 (6), 619–627 (2010).
Tyurin-Kuzmin, P. A. et al. Angiotensin receptor subtypes regulate adipose tissue renewal and remodelling. FEBS J. 287 (6), 1076–1087 (2020).
Zhang, Q. et al. Relaxin activates the MAP kinase pathway in human endometrial stromal cells. J. Cell. Biochem. 85 (3), 536–544 (2002).
Singh, S. et al. Relaxin activates peroxisome Proliferator-activated receptor γ (PPARγ) through a pathway involving PPARγ coactivator 1α (PGC1α). J. Biol. Chem. 290 (2), 950–959 (2014).
Kannabiran, S. A. et al. Real-time monitoring of cAMP in brown adipocytes reveals differential compartmentation of β1 and β3-adrenoceptor signalling. Mol. Metabolism. 37, 100986 (2020).
Bost, F. et al. The role of MAPKs in adipocyte differentiation and obesity. Biochimie 87 (1), 51–56 (2005).
Xu, H. & Hotamisligil, G. S. Signaling pathways utilized by tumor necrosis factor receptor 1 in adipocytes to suppress differentiation. FEBS Lett. 506 (2), 97–102 (2001).
Cawthorn, W. et al. Tumour necrosis factor-α inhibits adipogenesis via a β-catenin/TCF4(TCF7L2)-dependent pathway. Cell. Death Differ. 14, 1361–1373 (2007).
Tam, J. et al. Blockade of VEGFR2 and not VEGFR1 can limit Diet-Induced fat tissue expansion: role of local versus bone Marrow-Derived endothelial cells. PLoS One 4, e4974 (2009).
Seki, T. et al. Ablation of endothelial VEGFR1 improves metabolic dysfunction by inducing adipose tissue Browning. J. Exp. Med. 215, 611–626 (2018).
Ling, M. et al. Knockdown of VEGFB/VEGFR1 signaling promotes white adipose tissue Browning and skeletal muscle development. Int. J. Mol. Sci. 23, 7524 (2022).
Wan, X. et al. Shift work promotes adipogenesis via cortisol-dependent downregulation of EGR3-HDAC6 pathway. Cell. Death Discov. 10, 129 (2024).
Zhai, M. et al. Involvement of calcium Chanel in the regulation of adipogenesis. Adipocyte 9 (1), 132–141 (2020).
Pramme-Steinwachs, I. et al. Extracellular calcium modulates brown adipocyte differentiation and identity. Sci. Rep. 7, 8888 (2017).
Thiel, G. et al. Transcriptional response to Calcium-Sensing receptor stimulation. Endocrinology 153 (10), 4716–4728 (2012).
Sée, V. et al. Calcium-dependent regulation of the cell cycle via a novel MAPK–NF-κB pathway in Swiss 3T3 cells. J. Cell. Biol. 166 (5), 661–672 (2004).
Wang, X. et al. AQP3 facilitates proliferation and adipogenic differentiation of Porcine intramuscular adipocytes. Genes 11 (4), 453 (2020).
Benzi, A. et al. The TRPM2 ion channel regulates metabolic and thermogenic adaptations in adipose tissue of cold-exposed mice. Front. Endoc. 6 (14), 1251351 (2024).
Varma, V. et al. Thrombospondin-1 is an adipokine associated with obesity, adipose inflammation, and insulin resistance. Diabetes 57 (2), 432–439 (2008).
Kong, P. et al. Thrombospondin-1 regulates adiposity and metabolic dysfunction in diet-induced obesity enhancing adipose inflammation and stimulating adipocyte proliferation. Am. J. Physiol. Endocrinol. Metab. 305 (3), E439–E450 (2013).
Hoffman, N. E. et al. SLC25A23 augments mitochondrial Ca2 + uptake, interacts with MCU, and induces oxidative stress–mediated cell death. Mol. Biol. Cell. 25 (6), 936–947 (2014).
Ge, L. et al. Bovine CAPN3 core promoter initiates expression of foreign genes in skeletal muscle cells by myod transcriptional regulation. Int. J. Biochem. Cell Biol. 127, 105837 (2020).
Stengel, R. et al. Kcnh1 Voltage-gated potassium channels are essential for early zebrafish development. J. Biol. Chem. 287 (42), 35565–35575 (2012).
Zhang, P. et al. CPNE8 promotes gastric cancer metastasis by modulating focal adhesion pathway and tumor microenvironment. Int. J. Biol. Sci. 18 (13), 4932–4949 (2022).
Duan, Y. et al. Regulation of cholesterol homeostasis in health and diseases: from mechanisms to targeted therapeutics. Signal. Transduct. Target. Ther. 7, 265 (2022).
Robertson, K. A. et al. An interferon regulated MicroRNA provides broad Cell-Intrinsic antiviral immunity through multihit Host-Directed targeting of the sterol pathway. PLoS Biol. 14, e1002364 (2016).
Mazein, A., Watterson, S., Hsieh, W. Y., Griffiths, W. J. & Ghazal, P. A comprehensive machine-readable view of the mammalian cholesterol biosynthesis pathway. Biochem. Pharmacol. 86, 56–66 (2013).
Bian, L. et al. Transcriptome analysis reveals the early development in subcutaneous adipose tissue of Laiwu piglets. Animals 14, 2955 (2024).
Xin, Y. et al. RNA-Seq analysis reveals a negative role of MSMO1 with a synergized NSDHL expression during adipogenesis of 3T3-L1. Biosci. Biotechnol. Biochem. 83, 641–652 (2019).
Yabe, D., Brown, M. S. & Goldstein, J. L. Insig-2, a second endoplasmic reticulum protein that binds SCAP and blocks export of sterol regulatory element-binding proteins. Proceedings of the National Academy of Sciences 99, 12753–12758 (2002).
Peng, Y. et al. Cloning, human chromosomal Assignment, and adipose and hepatic expression of the CL-6/INSIG1 gene. Genomics 43, 278–284 (1997).
Yang, T. et al. Crucial Step Cholesterol Homeost. Cell. 110, 489–500 (2002).
Li, J., Takaishi, K., Cook, W., McCorkle, S. K. & Unger, R. H. Insig-1 brakes lipogenesis in adipocytes and inhibits differentiation of preadipocytes. Proceedings of the National Academy of Sciences 100, 9476–9481 (2003).
Hao, Z. et al. Functional differentiation of the ovine preadipocytes —insights from gene expression profiling. Funct. Integr. Genomics. 23, 97 (2023).
Huang, Z. et al. ACSS2 promotes systemic fat storage and utilization through selective regulation of genes involved in lipid metabolism. Proceedings of the National Academy of Sciences 115, (2018).
Chen, N. et al. ACSS2 controls PPARγ activity homeostasis to potentiate adipose-tissue plasticity. Cell. Death Differ. 31, 479–496 (2024).
Kang, Z. et al. Mir-193b regulates the Differentiation, Proliferation, and apoptosis of bovine adipose cells by targeting the ACSS2/AKT axis. Animals 10, 1265 (2020).
Zhao, S. et al. ATP-Citrate lyase controls a Glucose-to-Acetate metabolic switch. Cell. Rep. 17, 1037–1052 (2016).
Yu, H. et al. Bovine lipid metabolism related gene GPAM: molecular characterization, function identification, and association analysis with fat deposition traits. Gene 609, 9–18 (2017).
Li, F. et al. miRNA-223 targets the GPAM gene and regulates the differentiation of intramuscular adipocytes. Gene 685, 106–113 (2019).
Junjvlieke, Z. et al. Transcriptional regulation of bovine elongation of very long chain fatty acids protein 6 in lipid metabolism and adipocyte proliferation. J. Cell. Biochem. 120, 13932–13943 (2019).
Junjvlieke, Z. et al. Effect of ELOVL6 on the lipid metabolism of bovine adipocytes. Genomics 112, 2282–2290 (2020).
Wang, X. et al. Elevated miR-10a-5p facilitates cell cycle and restrains adipogenic differentiation via targeting Map2k6 and Fasn, respectively. Acta Biochim. Biophys. Sin (Shanghai). 52, 1227–1235 (2020).
Yen, C. L. E., Stone, S. J., Koliwad, S., Harris, C. & Farese, R. V. Thematic review series: Glycerolipids. DGAT enzymes and triacylglycerol biosynthesis. J. Lipid Res. 49, 2283–2301 (2008).
Guo, P. P. et al. Overexpression of DGAT2 regulates the differentiation of bovine preadipocytes. Animals 13, 1195 (2023).
Grimaldi, B. et al. PER2 controls lipid metabolism by direct regulation of PPARγ. Cell. Metab. 12, 509–520 (2010).
Wilhelmsen, K. et al. Nesprin-3, a novel outer nuclear membrane protein, associates with the cytoskeletal linker protein plectin. J. Cell. Biol. 5 (171), 799–810 (2005).
Yang, Y. et al. Cross-talk between microtubules and the linker of nucleoskeleton complex plays a critical role in the adipogenesis of human adipose-derived stem cells. Stem Cell. Res. Ther. 9, 125 (2018).
Zhang, T., Guan, H., Arany, E., Hill, D. J. & Yang, K. Maternal protein restriction permanently programs adipocyte growth and development in adult male rat offspring. J. Cell. Biochem. 101, 381–388 (2007).
Vargas, R. et al. Protein restriction during lactation causes transgenerational metabolic dysfunction in adult rat offspring. Front. Nutr. 9, (2023).
Painter, R. et al. Transgenerational effects of prenatal exposure to the Dutch famine on neonatal adiposity and health in later life. BJOG 115, 1243–1249 (2008).
Bol, V. V., Delattre, A. I., Reusens, B., Raes, M. & Remacle, C. Forced catch-up growth after fetal protein restriction alters the adipose tissue gene expression program leading to obesity in adult mice. Am. J. Physiology-Regulatory Integr. Comp. Physiol. 297, R291–R299 (2009).
Acknowledgements
This work has been done in the framework of the QUALITY-LAMB project funded by the Junta de Castilla and León Regional Government [Ref. LE084P23]. MA-G is funded by a predoctoral fellowship from the European Social Fund Plus, the Operational Programme of Castilla y León, and the Junta de Castilla y León, represented by the Consejería de Educación. We thank the Instituto de Ganadería de Montaña (CSIC-ULE; Grulleros 24346, León, Spain) for providing the farm facilities for the appropriate rearing of the suckling lambs analyzed in this work. We are also grateful to the collaborating slaughterhouse.
Author information
Authors and Affiliations
Contributions
M.A.G.: formal analyses and writing - original draft preparation. A.S.V.: conceptualization, methodology, and writing – review and editing. P.A.S.F.: investigation, writing – review and editing. R.P.: investigation. G.H.: investigation. J.M.: investigation. J.J.A.: conceptualization, supervision, writing – review and editing. B.G.G.: conceptualization, methodology, supervision, and writing – review and editing. All authors read, contributed to the article and approved the submitted version.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
Below is the link to the electronic supplementary material.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License, which permits any non-commercial use, sharing, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if you modified the licensed material. You do not have permission under this licence to share adapted material derived from this article or parts of it. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by-nc-nd/4.0/.
About this article
Cite this article
Alonso-García, M., Suárez-Vega, A., Fonseca, P.A.S. et al. Intergenerational effects of early life protein restriction on adipose tissue development as revealed by sheep transcriptomic analyses. Sci Rep 15, 36491 (2025). https://doi.org/10.1038/s41598-025-20877-y
Received:
Accepted:
Published:
Version of record:
DOI: https://doi.org/10.1038/s41598-025-20877-y
