Abstract
Background
Mongolian cattle are local breeds in northern China with excellent adaptability to harsh environmental conditions. Adipose tissues play essential roles in tolerance to cold and disease, but the associated cellular and molecular mechanisms are unclear.
Methods
Single-nucleus RNA sequencing (snRNA-seq) was performed on the adipose tissues from the subcutaneous (SAT), greater omentum (OAT) and perirenal (PAT) of 3 healthy cattle. The adipogenic trajectory was analyzed, and the functional roles of gene of interest were verified in vitro.
Results
There were different cell subpopulations in adipose tissues. The lipid-deposition adipocytes identified by the PTGER3 marker exhibited outstanding characteristics in SAT. In PAT and OAT, aldosterone was expressed to provide clues for the differential brown adipocytes. Among the DEGs by comparing OAT with SAT and PAT with OAT, C3 was significantly expressed in most of the cell populations in SAT. G0S2, LIPE, LPIN1, PTGER3 and RGCC took part in the adipogenic trajectory from preadipocyte commitment to mature adipocytes. S100A4 expression affected Ca2+ signaling and the expression of UCP1 ~ 3, FABP4 and PTGER3.
Conclusion
The cell heterogeneity and genes expressed in adipose tissues of Mongolian cattle not only determine the endocrine and energy storage, but contribute to adapt to cold and disease resistance.
Supplementary Information
The online version contains supplementary material available at 10.1186/s12864-024-10913-y.
Keywords: Adipose tissue, Endocrine, Immune, Lipid, Thermogenesis, Mongolian cattle, Single-nucleus RNA sequencing (snRNA-seq)
Background
The Mongolian cattle is an important and excellent local breed in northern China; they are used for milk, meat and labor and can adapt to harsh environmental conditions because of their tolerance of rough feeding and resistance to cold temperatures and disease. Adipose tissues, including subcutaneous adipose tissue (SAT) and visceral adipose tissue (VAT), which are located under the skin and around the abdominal organs, respectively, contribute to the adaptation of mammals. SAT is an important factor affecting the taste of meat, while VAT does not have a commercial use [1]. In terms of the anatomical distribution of adipose tissue, SAT and greater omental adipose tissue (OAT) are white adipose tissue (WAT), and perirenal adipose tissue (PAT) is brown adipose tissue (BAT) [2]. In addition to playing an essential role in energy homeostasis, WAT also acts as an important endocrine organ that secretes adipokines to regulate metabolic processes [3, 4], including its role in insulin action [5]. As an endocrine organ, WAT is regulated by the nervous and immune systems [6]. Apart from perirenal location, BAT at anterior cervical, intrascapular, perivascular, supraclavicular locations were found in mice and humans [7]. The brown adipocytes contain abundant mitochondria with special expression of uncoupling protein 1 (UCP1) in the inner mitochondrial membrane to take part in burning excess energy in brown adipocytes to produce heat [8].
Adipose tissues are heterogeneous and are composed of adipose stem and progenitor cells (ASPCs), adipocytes and immune cells. Adipocytes not only store energy in lipid droplets in the cytoplasm for metabolism but also regulate physiological processes through endocrine products [9]. ASPCs constitute another major component of adipose tissue; these cells differentiate into preadipocytes and eventually into mature adipocytes [10, 11]. In humans and mice, ASPCs are classified into three subtypes: CD26+ adipogenic stem cells (ASCs), CD54+ preadipocytes (PreAs) and CD142+ adipogenesis regulators (Aregs) [12]. However, CD142+ Aregs have not been isolated from cattle fat [13]. In livestock, ASPCs are stromal vascular fraction (SVF) cells derived from adipose tissue that play roles in promoting metabolism and the production of fatty acids and lipogenesis to increase the meat composition, including fat content [13, 14]. SVF cells resemble mesenchymal stem/stromal cells (MSCs) [14]. The functions of MSCs and adipocytes are modulated by each other [15].
In adipose tissue, immune cells are the most diverse and are classified into innate immune cells (such as macrophages, eosinophils, mast cells, and neutrophils) and adaptive immune cells (including various subtypes of B cells and T cells), which take part in the nervous–endocrine–immune network with adipocytes and adipocyte progenitors to maintain and modulate the ability of adipose tissue plasticity to regulate tissue homeostasis and environmental adaptation [9, 15]. Macrophages regulate homeostasis and hypertrophy induced by inflammation in adipose tissue and the initiation of insulin resistance both locally and systemically [16]. Dendritic cells are responsible for presenting foreign antigens loaded onto major histocompatibility complexes (MHCs) to T cells [17], which may be related to adipose tissue inflammation in obese individuals [18]. Upon foreign stimulation from antigens, immune responses to defend against antigens are incited by T cells, which include various subpopulations of CD4+, CD8+ and natural killer T (NKT) cells; however, most subpopulations of T cells play roles in regulating inflammation in adipose tissue in obese individuals [19]. B cells not only produce antibodies but also present antigens and are involved in macrophage phagocytosis [18]. In addition, there are endothelial cells presenting in adipose tissues, which are essential for regulating inflammation, nutrient transport and hormonal signals [9].
To understand and reveal the metabolic and functional contributions of adipose tissue at different visceral depots to the cold and disease resistance of Mongolian cattle, in this study, the cellular composition and differential gene expression in adipose tissues were investigated at single-nucleus resolution. The major cell subsets were identified by marker genes, and profiles of the immune and metabolic features of adipose tissues were subsequently created to elucidate the cellular and molecular mechanisms of cold and disease resistance in Mongolian cattle, which could provide evidence for management strategies.
Materials and methods
Animals and treatment
Healthy Mongolian cattle (two at 6 months of age and one at 16 months of age) were selected and obtained from commercial sources (Supplementary Fig. 1). These cattle were managed on the same farm in the Alxa region of China with free access to food and water (from February to August, 2022) and were fed more local forage and a small amount of concentrate in the winter. Cattle were humanely euthanized by intramuscular injection of chlorpromazinum at a dose of 1.5 mg/kg, followed by intravenous injection of barbiturate at a dose of 90 mg/kg.
Sample collection and preparation for snRNA-seq
Cattle were humanely euthanized to collect adipose tissues from the greater omentum, the subcutaneous and perirenal locations at Aug 30, 2022 (14–28 ℃), and the tissues were immediately placed in liquid nitrogen. Approximately 100 mg of liquid nitrogen-frozen tissue was ground on ice, and 2 mL of precooled homogenization buffer was added. The mixture was incubated for 3–5 min until the cell nuclei were fully released under the microscope. The tissue homogenate was filtered through 70- and 40-µm filters and centrifuged at 500 × g at 4 °C for 5 min. After washing, the precipitate suspension was made and observed under a microscope. If the suspension background was clean, the cells were directly counted, and the cell viability was assessed under a microscope after being stained with 0.4% trypan blue. Cells with more than 80% viability were qualified and captured for library construction using a Chromium Single Cell 30 Reagent Kit v3 (10X Genomics).
Library preparation and sequencing
Briefly, the single-cell suspension was partitioned into GEMs (gel beads in emulsion), where mRNAs were reverse transcribed into cDNAs. After the cDNAs were fragmented, the ends were repaired, “A” bases were added at the 3’-end of each strand, and the cDNAs were subjected to adaptor ligation and PCR. Based on the requirements for good library quality, which were determined using the Agilent Bioanalyzer high-sensitivity chip, single-strand PCR products were obtained by denaturation and then circularized to obtain single-stranded circular DNA, which was replicated in a rolling manner to obtain a DNA nanoball (DNB) containing multiple copies of DNA. A sufficient number of DNBs were sequenced on the NovaSeq 6000 platform.
snRNA-seq computational analysis
The versions and websites to the bovine reference genome and genome annotation files used for the bioinformatics analysis were NCBI GCF_002263795.2_ARS-UCD1.3.
The raw snRNA-seq data were processed in Cell Ranger 6.1.2 (10x Genomics) to generate a raw gene expression matrix, which was aggregated using CellRanger (v5.0.1) (https://www.support.10xgenomics.com/single-cell-gene-expression/software/downloads/latest) [20]. Downstream analysis was performed in the R package Seurat (v 3.2.0) [21] (https://www.satijalab.org/seurat/), which was based on the quality control analysis of cells with fewer than 200 target genes or more than 90% of the maximum genes and the top 15% of mitochondrial reads, which were filtered out. The doublets were identified by DoubletDetection (https://www.rdrr.io/github/scfurl/m3addon/man/doubletdetection.html) and then removed. The cell cycle distribution was analyzed using the CellCycle scoring tool of the Seurat program. Principal component analysis of the normalized gene expression dataset was performed using 2000 highly variable genes, and Uniform Manifold Approximation and Projection(UMAP)was subsequently used for two-dimensional visualization of the clusters generated by marker genes, which were identified by FindClusters (with the parameter resolution = 0.5) and FindAllMarkers as implemented in the Seurat (v4.0.2) package (with the parameters logFC.threshold > 0.25, minPct > 0.1 and Padj ≤ 0.05). Clusters were subsequently assigned to a known cell type using the Cell type annotation for single-cell RNA-seq data (SCSA) method (https://www.github.com/bioinfo-ibms-pumc/SCSA) [22]. Differentially expressed genes (DEGs) were identified using FindMarkers in Seurat (with the parameters logFC.threshold > 0.25, minPct > 0.1 and Padj ≤ 0.05). Gene Ontology (GO) (UniProt: http://www.ftp.ebi.ac.uk/pub/databases/GO/goa/UNIPROT/goa_uniprot_all.gaf.gz, NCBI gene2GO: http://www.ftp.ncbi.nih.gov/gene/DATA/gene2go.gz, GO official website: http://www.ftp.pir.georgetown.edu/databases/idmapping/idmapping.tb.gz, downloaded in May 2020) and Kyoto Encyclopedia of Genes and Genomes (KEGG) (V93.0) pathway analyses were performed using phyper (a function of R, with a parameter FDR ≤ 0.05 and q value for P value adjust for significant enrichment).
CytoTRACE (https://cytotrace.stanford.edu/) was used to calculate degrees of differentiation based on cell types or clusters of 5000 cells, with higher values at lower degrees of differentiation. Red represents a lower degree of differentiation, blue represents a higher degree of differentiation (left), and UMAP was displayed according to cell type or cluster. Pseudotime analysis was performed using Monocle2 (v2.22.0) software [21]. The protein–protein interactions (PPIs) were analyzed in Cytoscape (v3.5.1) [23].
Isolation, culture and identification of preadipocytes from the adipose tissues of Mongolian cattle
A total of 12 g of adipose tissue at different depots was isolated to obtain single cells by grinding and passing through a 200-mesh cell strainer. Briefly, adipose tissues from each individual were ground in Hanks balanced salt solution (Sigma‒Aldrich, St. Louis, MO) (pH 7.4) at 37 ºC for 30 min. The mixture was passed through a cell strainer with 200 mesh and then centrifuged at 1300 × g for 10 min at room temperature. For the isolation of preadipocytes, the cells were cultured in primary preadipocyte medium (Saios, Wuhan, China). The cell smears from the isolated preadipocytes were stored in foil-wrapped packages in snap-lock bags at − 80 °C until use.
Immunofluorescence, immunohistochemistry and immunocytochemistry
Prior to staining, the cell smear slides were unwrapped to prevent condensation by incubation at room temperature for half an hour. The individual smears were outlined using a PAP pen (Japan), fixed in paraformaldehyde for 2 min and washed with Tris-buffered saline (TBS, pH 7.5) for immunostaining. Briefly, after antigen retrieval was performed by microwaving the slides in 10 mM Tris-EDTA (pH 8.0), 3% methanol-hydrogen peroxide solution was used to remove peroxidase at room temperature for 25 min, followed by incubation with 0.5% BSA for 30 min to block nonspecific antigens. The primary antibodies for the markers were subsequently added to the slides, which were then incubated at 4 °C overnight, after which the slides were incubated with rabbit horseradish peroxidase (HRP)-conjugated secondary antibody for 30 min. The slides were subsequently washed with PBS (pH 7.4) three times. CY3-TSA or FITC-TSA was incubated at room temperature, and the nuclei were counterstained with DAPI. For immunohistochemistry, tissue slides were fixed with specific buffer (Servicebio, China) and embedded in paraffin for CD4 and CD8 distribution analysis, which was similar to the immunofluorescence method involving hematoxylin-stained nuclei. The primary antibodies used were anti-FABP4 (HUABIO, China), which was conjugated to PerCP-Cy5.5 (AbD Serotec, Raleigh, NC, USA) following the manufacturer’s instructions, along with anti-KLF12 (HUABIO, China), anti-DLK1 (HUABIO, China), anti-PTGER3 (HUABIO, China), anti-CD4 (Bioss, China), anti-CD8 (Proteintech, China), and anti-S100A4 nanobodies (patent in China, No. 202311757848.4).
Oil red O staining
Frozen sections of adipose tissues with a thickness of 6 μm were cut, placed on positively charged slides and then fixed in 10% formalin for 10 min. The slides were soaked in 60% isopropanol for 20 s after being washed with water and then stained with an improved oil red O solution for 15 min in a sealed container. To remove the staining solution, the slides were gently washed in 60% isopropanol. The nuclei were stained with Mayer’s hematoxylin staining solution for 2 min and rinsed in water for 10 min. After washing gently with distilled water and absorbing excess water with filter paper, the slides were sealed with a glycerol–gelatin mixture.
Quantitative real-time PCR
Total RNA from the isolated preadipocytes was extracted using TRIzol reagent (Invitrogen, Shanghai, China) and treated with DNase I (Sigma, Shanghai, China). A total of 1 µg of RNA was reverse transcribed into cDNA for real-time PCR analysis using a PrimeScript™ RT reagent kit (TAKARA, Dalian, China). Real-time quantitative PCR was performed using SYBR®Premix Ex Taq™ II (Tli RNaseH Plus) (TAKARA, Dalian, China). The mRNA expression of S100A4, PTGER3, FABP4, KLF12 and UCP1 ~ 3 were amplified and quantified with specific primers (Supplementary Table 1), which were normalized relative to β-actin. The analysis of relative gene abundance was performed using the comparative threshold cycle (CT) method.
Plasmid construction and cell transfection
The coding sequence of S100A4 was amplified by RT-PCR using cDNA generated from isolated preadipocytes as a template with primers containing HindIII and BamHI sites. The pEGFP-N1 vector and the PCR product were digested with HindIII and BamHI and then ligated to obtain the pEGFP-N1-S100A4 construct. Three siRNAs for S100A4 were designed and synthesized by Shanghai Sangon Biotech (Supplementary Table 2). The pEGFP-N1-S100A4 construct and 3 siRNA-S100A4 were then transfected into isolated preadipocytes, respectively. Two days after transfection, preadipocytes were collected for total mRNA extraction and Ca2+ imaging.
Ca2+ imaging
Ca2+ imaging was performed in preadipocytes according to the manufacturer’s instructions. Briefly, the harvested preadipocytes were labeled with Rhod-3 (Invitrogen, USA) for 1 h at room temperature. After being washed in PBS, the cells were incubated for 1 h to de-esterify the dye. The labeled cells were subjected to live-cell imaging using a laser scanning confocal microscope (FV1000, Olympus, Japan).
Statistical analysis
The differences in the qRT‒PCR and Ca2+ signal data among the groups were analyzed using ordinary one-way ANOVA. The correction for multiple comparisons was performed using the Tukey method in GraphPad Prism software (version 8.0). The data are presented as the means ± SDs.
Results
Cellular landscape of adipose tissues identified by snRNA-seq
To identify and distinguish the cellular heterogeneity and hierarchy, cell activation states, and cell trajectory of adipose tissues at different anatomical locations in Mongolian cattle, snRNA-seq analysis was performed. In adipose tissues, a total of 50,497 cells were analyzed, with a mean gene count of 1212. The feature gene expression profiles of ten distinct types of cells in adipose tissues, including adipocytes, adipose stem cells (ASCs), vascular endothelial cells, preadipocytes, macrophages, mesothelial cells, smooth muscle-like ASCs, T cells, lymphatic nodes, and neurons, were annotated (Fig. 1A). In terms of the cellular heterogeneity of adipose tissues at different depots, differences were observed: immune cells (such as DC cells, B cells and NK cells) were present only in PAT; macrophages were present in OAT, SAT and PAT; mesenchymal stem cells (MSCs) only were present only in OAT; and preadipocytes were present in OAT and SAT. Moreover, the percentages of different cells at different depots differed, especially those of adipocytes, preadipocytes, ASCs and macrophages (Fig. 1B). The major feature gene markers for different clusters are shown in Fig. 1C and are listed in Supplementary Table 3.
Fig. 1.
Cell compositions of adipose tissues in Mongolian cattle identified by scRNA-seq. A All the cell compositions of adipose tissues. B Cellular heterogeneity in adipose tissues at the subcutaneous (SAT), greater omentum (OAT) and perirenal (PAT) sites and the percentages of different cells in adipose tissues at the different depots. C Gene marker expression profiles in different cells of adipose tissues (Top row: Gene marker expression profiles in different cells of adipose tissues. Bottom row: Gene marker expression profiles in different cells in SAT, PAT and OAT, respectively). D Examples of adipocytes in OAT were marked by KLF12 and PTGER3 expression by immunofluorescence. E Example of preadipocytes in OAT were marked by KLF12 and FABP4 by immunofluorescence
Adipocytes in adipose tissues at different depots were classified and characterized into at least 4 subpopulations based on marker gene expression. In adipocytes in SAT, adipocyte 1 (expressing marker genes, including DGAT2, GPAM, and ADIPOQ) and adipocyte 2 (LIPE, LPIN1, PNPLA2, and PTGER3) subpopulations were annotated. In adipocytes in PAT, an adipocyte subpopulation (expressing LIPE and PLIN1) was annotated. In adipocytes in OAT, adipocyte subpopulation (expressing LIPE, PLIN1, and ADIPOQ) was annotated. The macrophages were reclustered into 6 subpopulations: macrophages (F13A1, CD163) in SAT, macrophages 1 (CD163, MRC1, DAB2, MSR1, and C1QA), macrophages 2 (C1QC, FCGR3A, MS4A7, CD163, and C1QB) macrophages 3 (CD163, C1QA, CSF1R, and MRC1) in PAT, macrophages 1 (eSYK, ITGAM, and CD163), and macrophages 2 (CD163, C1QA, C1QC, SYK) in OAT.
In addition, other cells presented distribution and classification characteristics. In SAT, there were 4 and 2 subpopulations of ASCs and preadipocytes, respectively. In PAT, 2 subpopulations of ASCs and no preadipocyte population were found. In OAT, there were 4 subpopulations of preadipocytes and 1 population of ASCs (Supplementary Table 3).
In subpopulations of adipocytes and macrophages, different pathways and genes contributed to their functions (Tables 1 and 2). For example, in SAT, PLPP1, diacylglycerol O-acyltransferase 1 (DGAT1), and DGAT2 were involved in glycerolipid and fat digestion and absorption, whereas only PLPP1 was involved in glycerophospholipid metabolism. We selected 2 markers to identify cells in adipose tissues at different depots, and the results revealed that KLF12+/PTGER3+ mesothelial cells were present in OAT (Fig. 1D) and that KLF12+/FABP4+ preadipocytes were present in OAT (Fig. 1E).
Table 2.
Macrophage subpopulation in adipose tissues at different depots of Mongolian cattle
| Adipose tissues at depot | Sub- population |
Markers | Pathway and genes |
|---|---|---|---|
| Subcutaneous adipose tissue | Macrophage | F13A1,CD163 |
Complement and coagulation cascades Coronavirus disease - COVID−19 |
| Greater Omental adipose tissue | Macrophage1 | SYK, ITGAM, CD163 |
NF-kappa B signaling pathway Phospholipase D signaling pathway PI3K-Akt signaling pathway Osteoclast differentiation Platelet activation Neutrophil extracellular trap formation C-type lectin receptor signaling pathway Natural killer cell mediated cytotoxicity B cell receptor signaling pathway Fc epsilon RI signaling pathway Fc gamma R-mediated phagocytosis Tuberculosis Kaposi sarcoma-associated herpesvirus infection Herpes simplex virus 1 infection Epstein-Barr virus infection Coronavirus disease - COVID−19 Viral carcinogenesis Rap1 signaling pathway Phagosome Cell adhesion molecules Complement and coagulation cascades Hematopoietic cell lineage Leukocyte transendothelial migration Regulation of actin cytoskeleton Pertussis Legionellosis Leishmaniasis Amoebiasis Staphylococcus aureus infection Transcriptional misregulation in cancer Acute myeloid leukemia |
| Macrophage2 | CD163,C1QA, C1QC, SYK |
Alcoholic liver disease B cell receptor signaling pathway Chagas disease Complement and coagulation cascades Coronavirus disease - COVID−19 C-type lectin receptor signaling pathway Efferocytosis Epstein-Barr virus infection Fc epsilon RI signaling pathway Fc gamma R-mediated phagocytosis Herpes simplex virus 1 infection Kaposi sarcoma-associated herpesvirus infection Natural killer cell mediated cytotoxicity Neutrophil extracellular trap formation NF-kappa B signaling pathway Osteoclast differentiation Pertussis Phospholipase D signaling pathway PI3K-Akt signaling pathway Platelet activation Prion disease Staphylococcus aureus infection Systemic lupus erythematosus Tuberculosis Viral carcinogenesis |
|
| Perirenal adipose tissue | Macrophage1 | CD163,MRC1,DAB2,MSR1,C1QA |
Cell cycle - yeast Alcoholic liver disease Chagas disease Complement and coagulation cascades Coronavirus disease - COVID−19 Efferocytosis Endocytosis Pertussis Phagosome Prion disease Staphylococcus aureus infection Systemic lupus erythematosus |
| Macrophage2 | C1QC, FCGR3A, MS4A7,CD163,C1QB |
Efferocytosis Complement and coagulation cascades Alcoholic liver disease Prion disease Pertussis Chagas disease Staphylococcus aureus infection Coronavirus disease - COVID−19 Systemic lupus erythematosus |
|
| Macrophage3 | CD163,C1QA, CSF1R, MRC1 |
Efferocytosis Acute myeloid leukemia Alcoholic liver disease Cell cycle - yeast Chagas disease Complement and coagulation cascades Coronavirus disease - COVID−19 Cytokine-cytokine receptor interaction Hematopoietic cell lineage MAPK signaling pathway Osteoclast differentiation Pathways in cancer Pertussis PI3K-Akt signaling pathway Prion disease Rap1 signaling pathway Ras signaling pathway Staphylococcus aureus infection Systemic lupus erythematosus Transcriptional misregulation in cancer Viral protein interaction with cytokine and cytokine receptor |
Table 1.
Adipocyte subpopulations in adipose tissues at different depots of Mongolian cattle
| Adipose tissues at depot | Sub- population |
Markers | Pathway and genes |
|---|---|---|---|
|
Subcutaneous adipose tissue (SAT) |
Adipocyte1 |
DGAT2; GPAM; ADIPOQ |
Glycerolipid metabolism–PLPP1, DGAT1, DGAT2 Fat digestion and absorption–Cd36, PLPP1, DGAT1, DGAT2 AMPK signaling pathway–ACACA, CD36 PPAR signaling pathway–CD36 Adipocytokine signaling pathway–CD36 Longevity regulating pathway–HIF1A (Non-) Alcoholic liver disease–XBP1 Type II diabetes mellitus–ADIPO, GLUT4 |
| Adipocyte2 |
LIPE; LPIN1; PNPLA2; PTGER3 |
AMPK signaling pathway–ACACA, Insulin signaling pathway–ACACA, Apelin signaling pathway–UCP1 Regulation of lipolysis in adipocytes–ATGL Aldosterone synthesis and secretion–DAGL cAMP signaling pathway–PPARα Thermogenesis–UCP1 Alcoholic liver disease–ACACA mTOR signaling pathway–GRB10,LPIN Glycerophospholipid metabolism–PLPP1 Calcium signaling pathway–PTGER3 Neuroactive ligand-receptor interaction–PTGER3 |
|
|
Perirenal adipose tissue (PAT) |
Adipocyte1 |
LIPE; PLIN1 |
AMPK signaling pathway–CD36 Insulin signaling pathway–ACACA Apelin signaling pathway–UCP1 Regulation of lipolysis in adipocytes–ATGL Aldosterone synthesis and secretion–DAGL cAMP signaling pathway–PPARα Thermogenesis–UCP1 PPAR signaling pathway–CD36 |
|
Greater omental adipose tissue (OAT) |
Adipocyte1 |
LIPE; PLIN1; ADIPOQ |
AMPK signaling pathway–ACACA, CD36 Insulin signaling pathway–ACACA Apelin signaling pathway–UCP1 Regulation of lipolysis in adipocytes–ATGL Aldosterone synthesis and secretion–DAGL cAMP signaling pathway–PPARα Thermogenesis–UCP1 PPAR signaling pathway–CD36 Type II diabetes mellitus–ADIPO, GLUT4 (Non)-alcoholic fatty liver disease–XBP1 Adipocytokine signaling pathway–CD36 Longevity regulating pathway–HIF1A |
Differentially expressed genes (DEGs) in cell types at different adipose tissue depots
Compared with the DEGs in SAT adipocytes, 82 DEGs were identified, including 49 upregulated (including 26 ribosome proteins) and 33 downregulated DEGs, among which COL3A1 in SAT and PDK4 in OAT and PAT are involved in the diabetic cardiomyopathy pathway. In addition, GLCE, which is involved in the heparan sulfate/heparin pathway, was highly and differentially expressed in SAT, OAT and PAT; moreover, COL5A2 and COL5A3, which are involved in the protein digestion and absorption pathway, and ABACA1, which is involved in fat digestion and absorption, were differentially expressed in OAT and PAT (Table 3).
Table 3.
The DEGs with more than 1.5 Fold changes in adipocytes of SAT versus PAT and OAT
| Gene | pct.OAT | pct.PAT | pct.SAT | avg_log2 FC (OAT VS SAT) |
avg_log2 FC (PATVS SAT) |
KEGG pathway | ||
|---|---|---|---|---|---|---|---|---|
| p_val | p_val | |||||||
| LPIN1 | 0.868 | 0.998 | 0.179 | 4.629 | 1.24E-275 | 5.430 | 0 | Alcoholic liver disease; mTOR signaling pathway; Glycerolipid metabolism; Glycerophospholipid metabolism |
| TPT1 | 0.707 | 0.573 | 0.05 | 3.062 | 9.84E-182 | 2.173 | 3.50E-128 | Translationally-controlled tumor protein a |
| PDK4 | 0.652 | 0.692 | 0.176 | 2.630 | 7.40E-126 | 2.072 | 1.06E-137 | Diabetic cardiomyopathy |
| GSN | 0.522 | 0.565 | 0.004 | 2.630 | 7.19E-130 | 1.979 | 5.50E-153 | Regulation of actin cytoskeleton; Fc gamma R-mediated phagocytosis; Viral carcinogenesis |
| IL1RAPL2 | 0.145 | 0.211 | 0.011 | 2.566 | 1.51E-24 | 2.043 | 1.81E-39 | Calcium ion-dependent exocytosisa |
| TXNIP | 0.835 | 0.975 | 0.25 | 2.565 | 1.67E-200 | 2.953 | 0 | NOD-like receptor signaling pathway |
| FGF2 | 0.749 | 0.975 | 0.344 | 2.530 | 1.41E-144 | 2.997 | 0 | Calcium signaling pathway; Proteoglycans in cancer; Regulation of actin cytoskeleton; PI3K-Akt signaling pathway; Rap1 signaling pathway; EGFR tyrosine kinase inhibitor resistance; Kaposi sarcoma-associated herpesvirus infection; Ras signaling pathway; Signaling pathways regulating pluripotency of stem cells; MAPK signaling pathway; Chemical carcinogenesis - receptor activation; Gastric cancer; Melanoma; Breast cancer |
| GRB10 | 0.616 | 0.951 | 0.199 | 2.484 | 9.68E-110 | 3.159 | 0 | mTOR signaling pathway |
| ABCA1 | 0.63 | 0.811 | 0.161 | 2.426 | 3.05E-116 | 2.228 | 5.59E-199 | Cholesterol metabolism; Lipid and atherosclerosis; ABC transporters; Fat digestion and absorption; |
| TMEM236 | 0.437 | 0.491 | 0.082 | 2.391 | 6.16E-77 | 1.983 | 7.36E-96 | Transmembrane proteina |
| G0S2 | 0.636 | 0.474 | 0.195 | 2.322 | 2.31E-109 | 1.535 | 1.49E-45 | G0/G1 switch proteina |
| CEACAM1 | 0.579 | 0.64 | 0.101 | 2.188 | 3.43E-115 | 1.587 | 6.54E-134 | Carcinoembryonic antigen-related cell adhesion moleculea |
| PFKFB3 | 0.451 | 0.812 | 0.091 | 2.134 | 7.65E-78 | 2.782 | 1.44E-254 | AMPK signaling pathway; HIF-1 signaling pathway; Fructose and mannose metabolism |
| LOC101906240 | 0.624 | 0.971 | 0.233 | 2.071 | 4.39E-95 | 3.266 | 0 | Unknown |
| TP53INP1 | 0.415 | 0.727 | 0.13 | 2.022 | 5.79E-56 | 2.437 | 2.88E-190 | Tumor protein p53-inducible nuclear protein a |
| ROBO2 | 0.317 | 0.475 | 0.034 | 2.004 | 1.18E-55 | 2.258 | 6.23E-106 | Axon guidance; Axon regeneration |
| ZDHHC2 | 0.426 | 0.805 | 0.133 | 1.888 | 1.54E-57 | 2.684 | 1.56E-236 | Palmitoyltransferasea |
| PID1 | 0.249 | 0.649 | 0.001 | 1.884 | 1.31E-50 | 3.295 | 2.66E-193 | Phosphotyrosine interaction domain containinga |
| GPAT3 | 0.345 | 0.71 | 0.043 | 1.878 | 1.95E-60 | 2.636 | 1.16E-208 | Glycerolipid metabolism; Glycerophospholipid metabolism |
| RIN2 | 0.408 | 0.569 | 0.034 | 1.863 | 1.33E-77 | 1.782 | 4.38E-134 | Ras and Rab interactor 2/3a |
| HIF3A | 0.444 | 0.891 | 0.086 | 1.818 | 2.08E-73 | 2.833 | 5.09E-303 | Hypoxia-inducible factor a |
| MEG8 | 0.45 | 0.928 | 0.087 | 1.790 | 5.91E-71 | 3.651 | 0 | Unknown |
| PLXNA4 | 0.526 | 0.795 | 0.241 | 1.781 | 1.31E-55 | 2.095 | 2.22E-169 | Axon guidance |
| RNF144A | 0.394 | 0.704 | 0.079 | 1.729 | 6.57E-61 | 2.132 | 5.58E-184 | E3 ubiquitin-protein ligase RNFa |
| CEBPD | 0.232 | 0.418 | 0 | 1.721 | 3.76E-47 | 1.846 | 7.65E-101 | CCAAT/enhancer binding proteina |
| HIF1A | 0.442 | 0.625 | 0.149 | 1.702 | 1.75E-56 | 1.635 | 6.45E-119 | Autophagy - animal; Proteoglycans in cancer; Mitophagy - animal; Longevity regulating pathway - worm; Pathways in cancer; Renal cell carcinoma; Thyroid hormone signaling pathway; Kaposi sarcoma-associated herpesvirus infection; HIF-1 signaling pathway; Axon regeneration; Th17 cell differentiation; Central carbon metabolism in cancer; PD-L1 expression and PD-1 checkpoint pathway in cancer; Choline metabolism in cancer; Chemical carcinogenesis - reactive oxygen species |
| PTPRK | 0.416 | 0.876 | 0.157 | 1.670 | 5.04E-45 | 3.237 | 8.12E-283 | Receptor-type tyrosine-protein phosphatase kappaa |
| SLC24A3 | 0.319 | 0.569 | 0.074 | 1.650 | 3.33E-43 | 1.903 | 3.08E-124 | Sodium/potassium/calcium exchangera |
| FSHR | 0.349 | 0.667 | 0.122 | 1.630 | 2.80E-39 | 2.374 | 5.69E-161 | cAMP signaling pathway; Ovarian steroidogenesis; Ovarian steroidogenesis; Neuroactive ligand-receptor interaction |
| GLCE | 0.768 | 0.975 | 0.505 | 1.578 | 1.05E-88 | 2.111 | 6.89E-269 | Glycosaminoglycan biosynthesis - heparan sulfate / heparin |
| RFTN1 | 0.301 | 0.715 | 0.042 | 1.555 | 1.04E-48 | 2.463 | 7.94E-210 | Raftlin; raftlin lipid raft linkea |
| SLC25A33 | 0.428 | 0.632 | 0.18 | 1.548 | 2.06E-45 | 1.526 | 9.91E-114 | Solute carrier familya |
| ARHGAP10 | 0.465 | 0.701 | 0.169 | 1.542 | 5.87E-51 | 1.937 | 1.36E-138 | Bacterial invasion of epithelial cells |
| ZFHX4 | 0.29 | 0.428 | 0.63 | -1.522 | 4.54E-92 | -1.517 | 1.27E-74 | Zinc finger homeobox proteina |
| FHL1 | 0.294 | 0.365 | 0.638 | -1.539 | 1.41E-93 | -1.686 | 1.11E-97 | JAK-STAT signaling pathway |
| DST | 0.429 | 0.494 | 0.71 | -1.547 | 4.73E-74 | -2.053 | 1.97E-99 | Heparan sulfate N-deacetylase/N-sulfotransferasea |
| LOC112442997 | 0.02 | 0.001 | 0.296 | -1.597 | 9.10E-137 | -1.763 | 9.69E-305 | Unknown |
| COL5A2 | 0.513 | 0.74 | 0.847 | -1.629 | 2.64E-139 | -1.551 | 2.56E-134 | Protein digestion and absorption |
| ANLN | 0.053 | 0.028 | 0.359 | -1.637 | 1.66E-119 | -1.949 | 8.13E-233 | Actin-binding protein anillina |
| CCDC80 | 0.186 | 0.31 | 0.527 | -1.667 | 1.77E-86 | -1.900 | 1.32E-63 | Coiled-coil domain-containing proteina |
| LIMA1 | 0.123 | 0.082 | 0.512 | -1.686 | 1.50E-125 | -2.264 | 3.72E-239 | LIM domain and actin-binding proteia |
| MARK1 | 0.297 | 0.434 | 0.605 | -1.695 | 1.29E-82 | -1.734 | 9.79E-65 | MAP/microtubule affinity-regulating kinasea |
| DCLK1 | 0.081 | 0.04 | 0.351 | -1.725 | 9.71E-84 | -2.207 | 7.07E-186 | Doublecortin-like kinasea |
| ACER3 | 0.515 | 0.753 | 0.738 | -1.735 | 1.89E-73 | -1.625 | 3.05E-51 | Sphingolipid metabolism |
| NSMF | 0.223 | 0.153 | 0.621 | -1.764 | 1.01E-111 | -2.566 | 8.11E-225 | NMDA receptor synaptonuclear signaling and neuronal migration factora |
| DDR2 | 0.045 | 0.095 | 0.39 | -1.764 | 1.58E-146 | -1.764 | 8.25E-117 | Discoidin domain receptor familya |
| COL5A3 | 0.329 | 0.116 | 0.716 | -1.814 | 1.16E-119 | -3.167 | 0 | Protein digestion and absorption |
| B3GNT6 | 0.28 | 0.428 | 0.642 | -1.846 | 2.61E-112 | -1.740 | 1.91E-85 | Mucin type O-glycan biosynthesis |
| LAMC1 | 0.311 | 0.467 | 0.681 | -1.864 | 1.95E-119 | -1.883 | 4.08E-100 | Focal adhesion; PI3K-Akt signaling pathway; Pathways in cancer; Human papillomavirus infection; Toxoplasmosis; ECM-receptor interaction; Small cell lung cancer; Amoebiasis; Prion disease |
| CALCRL | 0.142 | 0.232 | 0.499 | -1.885 | 2.23E-108 | -1.599 | 1.44E-71 | Vascular smooth muscle contraction; Neuroactive ligand-receptor interaction |
| TBX15 | 0 | 0.029 | 0.359 | -1.916 | 3.45E-236 | -1.759 | 3.12E-224 | T-box transcription factora |
| SGIP1 | 0.049 | 0.092 | 0.378 | -1.918 | 7.36E-138 | -1.758 | 7.48E-112 | SH3-containing GRB2-like protein 3-interacting proteina |
| SMC1A | 0.1 | 0.181 | 0.495 | -1.928 | 2.52E-142 | -1.938 | 5.53E-109 | Oocyte meiosis; Cell cycle |
| FILIP1L | 0.115 | 0.18 | 0.437 | -1.932 | 2.63E-100 | -1.899 | 7.84E-77 | Filamin A-interacting proteina |
| SFXN1 | 0.117 | 0.162 | 0.538 | -1.958 | 2.97E-157 | -2.028 | 4.43E-159 | Sideroflexina |
| ECHDC1 | 0.181 | 0.369 | 0.559 | -1.998 | 8.44E-118 | -1.664 | 7.62E-59 | Propanoate metabolism |
| SCD | 0.029 | 0.008 | 0.327 | -2.046 | 4.27E-138 | -2.309 | 1.15E-291 | AMPK signaling pathway; Longevity regulating pathway - worm; Alcoholic liver disease; Fatty acid metabolism; PPAR signaling pathway; Biosynthesis of unsaturated fatty acids |
| COL3A1 | 0.526 | 0.289 | 0.938 | -2.151 | 1.31E-194 | -4.554 | 0 | Diabetic cardiomyopathy; Protein digestion and absorption; Amoebiasis; AGE-RAGE signaling pathway in diabetic complications; Platelet activation; Relaxin signaling pathway |
| COL5A1 | 0.184 | 0.186 | 0.648 | -2.158 | 1.16E-158 | -2.707 | 5.96E-221 | Protein digestion and absorption |
| MME | 0.01 | 0.002 | 0.332 | -2.160 | 4.52E-187 | -2.220 | 0 | Alzheimer disease; Protein digestion and absorption; Renin-angiotensin system; Hematopoietic cell lineage |
| COL18A1 | 0.051 | 0.023 | 0.508 | -2.278 | 5.13E-215 | -2.640 | 0 | Protein digestion and absorption; |
| ANK2 | 0.135 | 0.228 | 0.638 | -2.305 | 9.62E-196 | -2.219 | 2.47E-165 | Toll and Imd signaling pathway; Proteoglycans in cancer |
| RGS7 | 0.211 | 0.066 | 0.565 | -2.342 | 1.23E-102 | -3.625 | 0 | Regulator of G-protein signalinga |
| SYNE1 | 0.19 | 0.358 | 0.694 | -2.403 | 8.15E-197 | -2.135 | 1.24E-141 | Spectrin repeat containing nuclear envelope proteina |
| COL15A1 | 0.206 | 0.027 | 0.715 | -2.407 | 1.81E-203 | -3.556 | 0 | Protein digestion and absorption; |
| THBS1 | 0.045 | 0.091 | 0.531 | -3.176 | 2.48E-252 | -2.976 | 7.66E-240 | Malaria; p53 signaling pathway; Proteoglycans in cancer; Focal adhesion; PI3K-Akt signaling pathway; Rap1 signaling pathway; MicroRNAs in cancer; TGF-beta signaling pathway; Human papillomavirus infection; Phagosome; ECM-receptor interaction; Bladder cancer |
a meaned that the KEGG pathway from KEGG tools in website (https://www.kegg.jp/kegg/kegg2.html)
Compared with the DEGs identified in SAT preadipocytes, 90 DEGs were identified in OAT and PAT, including 50 upregulated (including 38 ribosome proteins), 35 downregulated and 5 upregulated or downregulated DEGs, among which there were fewer genes associated with metabolism. S100A4 and CAMK1D are involved in the calcium binding or signaling pathway, TRIP10 is involved in the insulin signaling pathway, and DST is involved in the heparan sulfate/heparin pathway; these genes were highly and differentially expressed in SAT, OAT and PAT. Moreover, COL5A1, which is involved in the protein digestion pathway, was differentially expressed (Table 4).
Table 4.
The DEGs with more than 1.5 Fold changes in preadipocytes of SAT versus PAT and OAT
| Gene | pct.OAT | pct.PAT | pct.SAT | avg_log2 FC (OAT VS SAT) |
avg_log2 FC (PAT VS SAT) |
KEGG pathway | ||
|---|---|---|---|---|---|---|---|---|
| p_val | p_val | |||||||
| GPM6A | 0.827 | 0.351 | 0.023 | 5.180 | 0 | 1.844 | 7.60E-82 | Glycoproteina |
| TPT1 | 0.856 | 0.893 | 0.17 | 2.849 | 0 | 2.363 | 8.68E-95 | Translationally-controlled tumor protein a |
| TXNIP | 0.754 | 0.901 | 0.142 | 2.770 | 0 | 2.943 | 2.06E-121 | NOD-like receptor signaling pathway |
| UBA52 | 0.644 | 0.718 | 0.084 | 2.696 | 0 | 1.958 | 4.49E-102 | Ribosome; Coronavirus disease - COVID-19; Shigellosis; Mitophagy - animal; Ubiquitin mediated proteolysis; Kaposi sarcoma-associated herpesvirus infection; Parkinson disease; Pathways of neurodegeneration - multiple diseases |
| ROBO2 | 0.382 | 0.611 | 0.032 | 2.590 | 1.90E-274 | 3.095 | 4.43E-175 | Axon guidance; Axon regeneration |
| ZC3H10 | 0.541 | 0.588 | 0.087 | 2.360 | 0 | 1.517 | 1.75E-64 | Zinc finger CCCH domain-containing proteina |
| FAU | 0.417 | 0.466 | 0.041 | 2.353 | 6.49E-294 | 1.515 | 1.18E-85 | Ribosome; Coronavirus disease - COVID-19 |
| WT1 | 0.229 | 0.305 | 0.001 | 2.235 | 1.74E-177 | 1.627 | 2.36E-190 | Transcriptional misregulation in cancer |
| FN1 | 0.783 | 0.031 | 0.273 | 2.152 | 0 | -2.555 | 3.37E-10 | Focal adhesion; Regulation of actin cytoskeleton; PI3K-Akt signaling pathway; Shigellosis; Bacterial invasion of epithelial cells; Rap1 signaling pathway; ECM-receptor interaction; Proteoglycans in cancer; Human papillomavirus infection; Yersinia infection; Salmonella infection; Mitophagy - animal; Pathways in cancer; Small cell lung cancer; AGE-RAGE signaling pathway in diabetic complications; Amoebiasis; NOD-like receptor signaling pathway; Parkinson disease; Amyotrophic lateral sclerosis; Pathways of neurodegeneration - multiple diseases; MAPK signaling pathway - fly |
| LOC101906240 | 0.43 | 0.824 | 0.107 | 1.972 | 5.89E-212 | 2.923 | 5.21E-128 | Unknown |
| MT-COX1 | 0.671 | 0 | 0.157 | 1.935 | 0 | -2.596 | 9.50E-07 | Unknown |
| S100A4 | 0.496 | 0.015 | 0.133 | 1.756 | 1.55E-241 | -1.731 | 6.45E-05 | Calcium binding proteina |
| SELENOP | 0.321 | 0.687 | 0.062 | 1.705 | 1.29E-163 | 2.029 | 2.39E-124 | Selenoprotein W-related proteina |
| PTPRK | 0.26 | 0.71 | 0.051 | 1.573 | 7.40E-125 | 3.036 | 7.46E-167 | Receptor-type tyrosine-protein phosphatase kappaa |
| MT-COX3 | 0.585 | 0 | 0.119 | 1.547 | 0 | -2.721 | 2.79E-05 | Unknown |
| FLRT2 | 0.349 | 0.557 | 0.1 | 1.545 | 1.82E-139 | 1.503 | 5.92E-50 | Leucine-rich repeat transmembrane proteina |
| CAMK1D | 0.121 | 0.084 | 0.319 | -1.502 | 3.67E-142 | -2.186 | 1.11E-09 | Aldosterone synthesis and secretion; Calcium signaling pathway; Glioma; Oxytocin signaling pathway |
| THBS3 | 0.016 | 0.015 | 0.13 | -1.516 | 3.43E-126 | -1.656 | 8.53E-05 | Focal adhesion; PI3K-Akt signaling pathway; ECM-receptor interaction; Human papilloma virus infection; Phagosome; Malaria |
| STXBP6 | 0.055 | 0.107 | 0.212 | -1.521 | 2.40E-128 | -1.534 | 0.000704917 | Syntaxin-binding proteina |
| SEPT5 | 0.008 | 0.015 | 0.123 | -1.536 | 2.05E-147 | -1.556 | 0.000157246 | Parkinson disease; Pathways of neurodegeneration - multiple diseases |
| RRBP1 | 0.146 | 0.26 | 0.368 | -1.635 | 3.11E-165 | -1.819 | 9.04E-06 | Protein processing in endoplasmic reticulum |
| LSP1 | 0.047 | 0.13 | 0.22 | -1.647 | 1.97E-158 | -1.527 | 0.001448676 | Tuberculosis; C-type lectin receptor signaling pathway |
| SRRM2 | 0.361 | 0.634 | 0.674 | -1.647 | 0 | -1.707 | 9.07E-16 | Serine/arginine repetitive matrix proteina |
| COL5A1 | 0.191 | 0.176 | 0.44 | -1.653 | 9.68E-198 | -2.477 | 3.50E-13 | Protein digestion and absorption |
| IFRD2 | 0.041 | 0.076 | 0.15 | -1.709 | 1.38E-84 | -1.767 | 0.007474148 | Interferon-related developmental regulatora |
| DBN1 | 0.021 | 0.038 | 0.171 | -1.744 | 8.96E-168 | -1.827 | 3.04E-05 | Drebrina |
| PDLIM7 | 0.022 | 0.031 | 0.194 | -1.831 | 3.47E-198 | -1.993 | 1.34E-06 | PDZ and LIM domain proteina |
| CRIP2 | 0.05 | 0.061 | 0.248 | -1.864 | 7.17E-190 | -2.275 | 1.74E-07 | Cysteine-rich proteina |
| EBF2 | 0.033 | 0.802 | 0.238 | -1.920 | 7.52E-226 | 1.585 | 2.45E-40 | Early B-cell factora |
| MAP1A | 0.015 | 0.038 | 0.183 | -1.937 | 4.18E-209 | -1.938 | 9.98E-06 | Microtubule-associated proteina |
| FAT4 | 0.026 | 0.046 | 0.229 | -2.016 | 4.98E-238 | -1.947 | 3.75E-07 | Hippo signaling pathway - fly; Hippo signaling pathway - multiple species |
| GLIS3 | 0.058 | 0.115 | 0.307 | -2.030 | 8.92E-258 | -1.804 | 2.14E-07 | Zinc finger protein GLIS 1/3a |
| SFRP4 | 0.001 | 0.038 | 0.163 | -2.044 | 2.61E-248 | -1.806 | 6.30E-05 | Wnt signaling pathway |
| FAP | 0.068 | 0.107 | 0.335 | -2.072 | 2.06E-274 | -2.080 | 2.11E-09 | Unknown |
| PHLDB1 | 0.055 | 0.206 | 0.309 | -2.165 | 2.88E-272 | -1.652 | 0.000113131 | Pleckstrin homology-like domain family Ba |
| WSCD2 | 0.003 | 0 | 0.199 | -2.240 | 6.43E-295 | -2.297 | 1.82E-08 | WSC domain containinga |
| DLG2 | 0.083 | 0.122 | 0.416 | -2.268 | 0 | -2.097 | 1.37E-12 | Tight junction; Human papillomavirus infection; Hippo signaling pathway |
| FGF7 | 0.005 | 0.023 | 0.211 | -2.290 | 6.62E-301 | -2.249 | 1.02E-07 | Regulation of actin cytoskeleton; PI3K-Akt signaling pathway; Rap1 signaling pathway; MAPK signaling pathway; Pathways in cancer; Melanoma; Ras signaling pathway; Calcium signaling pathway; Chemical carcinogenesis - receptor activation; Breast cancer; Gastric cancer |
| ITGBL1 | 0.009 | 0.053 | 0.228 | -2.305 | 0 | -2.082 | 6.00E-07 | Integrin beta-like proteina |
| TBX15 | 0.001 | 0.023 | 0.22 | -2.315 | 0 | -2.176 | 4.29E-08 | T-box transcription factora |
| FILIP1L | 0.102 | 0.267 | 0.446 | -2.378 | 0 | -2.068 | 1.23E-09 | Filamin A interacting proteina |
| HMCN1 | 0.024 | 0.176 | 0.279 | -2.470 | 0 | -1.643 | 0.000311027 | Hemicentina |
| DMRT2 | 0.004 | 0.053 | 0.255 | -2.525 | 0 | -2.239 | 3.38E-08 | Doublesex- and mab-3-related transcription factora |
| TRIP10 | 0.042 | 0.313 | 0.367 | -2.576 | 0 | -1.519 | 0.000391027 | Insulin signaling pathway |
| DKK2 | 0.015 | 0.137 | 0.291 | -2.618 | 0 | -1.799 | 6.12E-06 | Wnt signaling pathway; Pathways of neurodegeneration - multiple diseases; Alzheimer disease |
| TNFAIP6 | 0.002 | 0.122 | 0.259 | -2.698 | 0 | -1.618 | 4.18E-05 | Tumor necrosis factor-inducible gene 6 proteina |
| PKD1 | 0.044 | 0.13 | 0.447 | -2.888 | 0 | -2.684 | 1.29E-15 | Polycystina |
| DST | 0.27 | 0.802 | 0.831 | -2.977 | 0 | -1.721 | 1.34E-27 | Heparan sulfate N-deacetylase/N-sulfotransferasea |
| MAP1B | 0.083 | 0.069 | 0.511 | -3.057 | 0 | -3.670 | 3.42E-23 | Microtubule-associated protein 1a |
| VCAN | 0.052 | 0.145 | 0.526 | -3.229 | 0 | -3.144 | 9.22E-21 | Cell adhesion molecules |
| MMP16 | 0.085 | 0.214 | 0.601 | -3.372 | 0 | -3.015 | 2.90E-24 | MicroRNAs in cancer; Parathyroid hormone synthesis, secretion and action |
| LOC530102 | 0.006 | 0.237 | 0.545 | -4.229 | 0 | -2.797 | 2.79E-18 | Focal adhesion; PI3K-Akt signaling pathway; ECM-receptor interaction; Human papillomavirus infection; Protein digestion and absorption |
a meaned that the KEGG pathway from KEGG tools in website (https://www.kegg.jp/kegg/kegg2.html)
Compared with the DEGs in the ASCs of SAT, 72 DEGs were identified, 38 of which were upregulated (including 23 ribosome proteins), 32 of which were downregulated and 2 of which were upregulated or downregulated in OAT and PAT, among which there were fewer genes associated with metabolism. FABP4 regulates lipolysis in adipocytes on the PPAR signaling pathway, ADIPOQ is involved in the type II diabetes mellitus pathway, CACNA1C and PDE1C are involved in the taste and olfactory transduction pathways, and DST is involved in the heparan sulfate/heparin pathway. These genes were highly and differentially expressed in SAT, OAT and PAT (Table 5).
Table 5.
The DEGs with more than 1.5 Fold changes in ASC of SAT versus PAT and OAT
| Gene | pct.OAT | pct.PAT | pct.SAT | avg_log2 FC (OAT VS SAT) |
avg_log2 FC (PAT VS SAT) |
KEGG pathway | ||
|---|---|---|---|---|---|---|---|---|
| p_val | p_val | |||||||
| GPM6A | 0.879 | 0.437 | 0.026 | 5.720 | 0 | 2.479 | 0 | Glycoproteina |
| THSD4 | 0.678 | 0.419 | 0.028 | 3.600 | 0 | 1.910 | 0 | Thrombospondin type-1 domain-containing protein 4a |
| UNC5C | 0.842 | 0.009 | 0.22 | 3.187 | 0 | -1.625 | 6.98E-265 | Axon guidance |
| ROBO2 | 0.765 | 0.78 | 0.098 | 3.164 | 0 | 3.327 | 0 | Axon guidance; Axon regeneration |
| TXNIP | 0.82 | 0.875 | 0.145 | 2.979 | 0 | 3.422 | 0 | NOD-like receptor signaling pathway |
| FN1 | 0.671 | 0.073 | 0.402 | 2.815 | 1.76E-292 | -2.219 | 0 | Focal adhesion; Rap1 signaling pathway; Shigellosis; Regulation of actin cytoskeleton; Proteoglycans in cancer; Salmonella infection; Parkinson disease; PI3K-Akt signaling pathway; Bacterial invasion of epithelial cells; Pathways in cancer; Human papillomavirus infection; Pathways of neurodegeneration - multiple diseases; Yersinia infection; ECM-receptor interaction; Mitophagy - animal; AGE-RAGE signaling pathway in diabetic complications; Amyotrophic lateral sclerosis; Small cell lung cancer; Amoebiasis; NOD-like receptor signaling pathway; MAPK signaling pathway - fly |
| TPT1 | 0.842 | 0.709 | 0.105 | 2.760 | 0 | 2.178 | 0 | Translationally-controlled tumor protein a |
| SEMA5A | 0.563 | 0.394 | 0.008 | 2.635 | 0 | 1.880 | 0 | Axon guidance |
| UBA52 | 0.664 | 0.447 | 0.051 | 2.533 | 0 | 1.765 | 0 | Coronavirus disease - COVID-19;Ribosome; Shigellosis; Parkinson disease; Pathways of neurodegeneration - multiple diseases; Mitophagy - animal; Kaposi sarcoma-associated herpesvirus infection; Ubiquitin mediated proteolysis |
| IGFBP7 | 0.653 | 0.494 | 0.145 | 2.092 | 0 | 1.531 | 3.12E-251 | Insulin-like growth factor-binding proteina |
| LOC101906240 | 0.722 | 0.877 | 0.254 | 2.071 | 0 | 3.005 | 0 | Unknown |
| ADIPOQ | 0.325 | 0.331 | 0.008 | 1.944 | 4.88E-279 | 1.628 | 2.33E-288 | Type II diabetes mellitus; |
| KCNB2 | 0.303 | 0.206 | 0.022 | 1.869 | 9.41E-229 | 1.885 | 1.11E-137 | Potassium voltage-gated channel Shab-related subfamily B member 2a |
| CCSER1 | 0.539 | 0.447 | 0.108 | 1.850 | 0 | 1.562 | 1.58E-243 | Serine-rich coiled-coil domain-containing proteina |
| FABP4 | 0.212 | 0.391 | 0.004 | 1.846 | 1.01E-172 | 2.247 | 0 | Regulation of lipolysis in adipocytes; PPAR signaling pathway |
| CCND3 | 0.553 | 0.661 | 0.092 | 1.785 | 0 | 2.491 | 0 | Focal adhesion; PI3K-Akt signaling pathway; Pathways in cancer; Human papillomavirus infection; Hippo signaling pathway; Wnt signaling pathway; Cellular senescence; Chemical carcinogenesis - receptor activation; Influenza A; Human T-cell leukemia virus 1 infection; Measles; Epstein-Barr virus infection; Cell cycle; Viral carcinogenesis; p53 signaling pathway; JAK-STAT signaling pathway |
| SIPA1L2 | 0.432 | 0.57 | 0.097 | 1.593 | 1.24E-244 | 1.976 | 0 | Rap1 signaling pathway |
| ABCC9 | 0.031 | 0.004 | 0.266 | -1.506 | 1.56E-233 | -1.653 | 0 | ABC transporters; |
| IL16 | 0.167 | 0.108 | 0.455 | -1.525 | 1.89E-228 | -1.890 | 0 | Interleukina |
| CACNA1C | 0.061 | 0.07 | 0.348 | -1.527 | 3.80E-263 | -1.554 | 3.74E-273 | MAPK signaling pathway; Pathways of neurodegeneration - multiple diseases; Alzheimer disease; Prion disease; Glutamatergic synapse; Dopaminergic synapse; GABAergic synapse; Oxytocin signaling pathway; Renin secretion; Long-term potentiation; Growth hormone synthesis, secretion and action; GnRH signaling pathway; Circadian entrainment; Chemical carcinogenesis - receptor activation; cGMP-PKG signaling pathway; Hypertrophic cardiomyopathy; Cholinergic synapse; Aldosterone synthesis and secretion; Calcium signaling pathway; Arrhythmogenic right ventricular cardiomyopathy; GnRH secretion; Dilated cardiomyopathy; cAMP signaling pathway; Adrenergic signaling in cardiomyocytes; Cardiac muscle contraction; Retrograde endocannabinoid signaling; Amphetamine addiction; Cushing syndrome; Vascular smooth muscle contraction; Insulin secretion; Taste transduction |
| PDE1C | 0.067 | 0.043 | 0.232 | -1.559 | 2.08E-116 | -1.591 | 1.12E-173 | Morphine addiction; Renin secretion; Calcium signaling pathway; Purine metabolism; Olfactory transduction; Taste transduction |
| PDE7B | 0.171 | 0.235 | 0.536 | -1.580 | 6.42E-299 | -1.538 | 3.99E-246 | Morphine addiction; Purine metabolism |
| LOC112441831 | 0.005 | 0.004 | 0.154 | -1.669 | 4.33E-166 | -1.678 | 4.19E-191 | Unknown |
| FAM124A | 0.019 | 0.017 | 0.314 | -1.704 | 0 | -1.754 | 0 | Unknown |
| SLC8A1 | 0.146 | 0.058 | 0.426 | -1.732 | 3.94E-211 | -2.311 | 0 | Apelin signaling pathway; Endocrine and other factor-regulated calcium reabsorption; cGMP-PKG signaling pathway; Hypertrophic cardiomyopathy; Calcium signaling pathway; Arrhythmogenic right ventricular cardiomyopathy; Dilated cardiomyopathy; Adrenergic signaling in cardiomyocytes; Cardiac muscle contraction; Protein digestion and absorption; Mineral absorption; Olfactory transduction |
| HDAC9 | 0.121 | 0.091 | 0.309 | -1.798 | 4.82E-119 | -1.850 | 4.48E-170 | Viral carcinogenesis; Neutrophil extracellular trap formation; Alcoholism |
| PKD1 | 0.125 | 0.137 | 0.46 | -1.833 | 4.38E-304 | -1.881 | 0 | Polycystin 1a |
| SFRP4 | 0.001 | 0.027 | 0.244 | -1.900 | 1.34E-294 | -1.761 | 5.83E-241 | Wnt signaling pathway |
| LOC112448368 | 0.015 | 0.081 | 0.377 | -1.984 | 0 | -1.605 | 2.99E-291 | Unknown |
| FAT4 | 0.032 | 0.093 | 0.402 | -2.067 | 0 | -1.826 | 1.64E-303 | Hippo signaling pathway - fly; Hippo signaling pathway - multiple species |
| ELMO1 | 0.022 | 0.084 | 0.415 | -2.152 | 0 | -1.794 | 0 | Shigellosis; Salmonella infection; Bacterial invasion of epithelial cells; Yersinia infection; Chemokine signaling pathway |
| DMRT2 | 0.001 | 0.1 | 0.413 | -2.193 | 0 | -1.679 | 0 | Doublesex- and mab-3-related transcription factor 2a |
| DKK2 | 0.115 | 0.256 | 0.646 | -2.264 | 0 | -1.859 | 0 | Pathways of neurodegeneration - multiple diseases; Alzheimer disease; Wnt signaling pathway |
| FAP | 0.168 | 0.168 | 0.628 | -2.281 | 0 | -2.287 | 0 | Unknown |
| HMCN1 | 0.083 | 0.221 | 0.539 | -2.294 | 0 | -1.677 | 7.66E-265 | Hemicentina |
| GLIS3 | 0.164 | 0.233 | 0.618 | -2.324 | 0 | -1.908 | 0 | Zinc finger protein GLIS 1/3a |
| MAP1B | 0.187 | 0.163 | 0.63 | -2.341 | 0 | -2.515 | 0 | Microtubule-associated protein 1a |
| DST | 0.602 | 0.796 | 0.938 | -2.360 | 0 | -1.533 | 0 | Heparan sulfate N-deacetylase/N-sulfotransferasea |
| ITGBL1 | 0.043 | 0.077 | 0.49 | -2.427 | 0 | -2.260 | 0 | Integrin beta-like proteina |
| TBX15 | 0.004 | 0.067 | 0.489 | -2.497 | 0 | -2.100 | 0 | T-box transcription factora |
| DLG2 | 0.242 | 0.216 | 0.756 | -2.519 | 0 | -2.216 | 0 | Tight junction; Human papillomavirus infection; Hippo signaling pathway |
| WSCD2 | 0.002 | 0.001 | 0.424 | -2.546 | 0 | -2.555 | 0 | WSC domain containinga |
| FILIP1L | 0.221 | 0.43 | 0.721 | -2.603 | 0 | -1.974 | 0 | Filamin A interacting proteina |
| FGF7 | 0.009 | 0.062 | 0.483 | -2.661 | 0 | -2.376 | 0 | Rap1 signaling pathway; Regulation of actin cytoskeleton; Ras signaling pathway; MAPK signaling pathway; PI3K-Akt signaling pathway; Pathways in cancer; Melanoma; Chemical carcinogenesis - receptor activation; Calcium signaling pathway; Breast cancer; Gastric cancer |
| VCAN | 0.139 | 0.158 | 0.714 | -2.682 | 0 | -2.950 | 0 | Cell adhesion molecules |
| ABI3BP | 0.325 | 0.51 | 0.859 | -2.700 | 0 | -1.974 | 0 | Target of Nesh-SH3a |
| MMP16 | 0.312 | 0.205 | 0.817 | -3.107 | 0 | -3.451 | 0 | MicroRNAs in cancer; Parathyroid hormone synthesis, secretion and action |
| TNFAIP6 | 0.007 | 0.161 | 0.483 | -3.303 | 0 | -2.110 | 1.20E-303 | Tumor necrosis factor-inducible gene 6 proteina |
| LOC530102 | 0.012 | 0.264 | 0.626 | -3.840 | 0 | -2.495 | 0 | Focal adhesion; PI3K-Akt signaling pathway; Human papillomavirus infection; ECM-receptor interaction; Protein digestion and absorption |
a meaned that the KEGG pathway from KEGG tools in website (https://www.kegg.jp/kegg/kegg2.html)
Compared with the DEGs in macrophages of SAT, 23 DEGs were upregulated (including 2 ribosome proteins), 11 were downregulated, and 2 were upregulated or downregulated in OAT and PAT; in addition to FABP4, EBF1 is involved in the insulin resistance and insulin signaling pathways (Table 6).
Table 6.
The DEGs with more than 1.5 Fold changes in macrophages of SAT versus PAT and OAT
| Gene | pct.OAT | pct.PAT | pct.SAT | avg_log2 FC (OAT VS SAT) |
avg_log2 FC (PATVS SAT) |
KEGG pathway | ||
|---|---|---|---|---|---|---|---|---|
| p_val | p_val | |||||||
| SRGN | 0.755 | 0.58 | 0.073 | 2.908 | 0 | 2.308 | 3.89E-183 | Proteoglycan peptide core proteina |
| MT-COX1 | 0.698 | 0 | 0.137 | 2.397 | 3.27E-248 | -1.966 | 3.51E-43 | Unknown |
| FYB1 | 0.689 | 0.551 | 0.103 | 2.138 | 2.27E-269 | 1.668 | 2.45E-133 | Yersinia infection; Rap1 signaling pathway |
| BOLA-DRB3 | 0.647 | 0.49 | 0.101 | 2.023 | 4.68E-241 | 1.617 | 9.95E-111 | Phagosome; Leishmaniasis; Tuberculosis; Toxoplasmosis; Antigen processing and presentation; Influenza A; Epstein-Barr virus infection; Th17 cell differentiation; Staphylococcus aureus infection; Th1 and Th2 cell differentiation; Inflammatory bowel disease; Human T-cell leukemia virus 1 infection; Viral myocarditis; Rheumatoid arthritis; Intestinal immune network for IgA production; Allograft rejection; Asthma; Type I diabetes mellitus; Cell adhesion molecules; Hematopoietic cell lineage; Autoimmune thyroid disease; Herpes simplex virus 1 infection; Graft-versus-host disease; Systemic lupus erythematosus |
| PLXNC1 | 0.694 | 0.62 | 0.122 | 2.012 | 4.05E-256 | 1.640 | 8.97E-154 | Axon guidance |
| SELENOP | 0.639 | 0.656 | 0.126 | 1.704 | 6.79E-199 | 1.850 | 4.50E-170 | Selenoprotein W-related proteina |
| FABP4 | 0.373 | 0.372 | 0.026 | 1.689 | 1.79E-135 | 1.538 | 4.20E-120 | Regulation of lipolysis in adipocytes; PPAR signaling pathway |
| RGS1 | 0.364 | 0.472 | 0.033 | 1.609 | 1.13E-126 | 2.074 | 2.24E-163 | Regulator of G-protein signalinga |
| FKBP5 | 0.804 | 0.928 | 0.276 | 1.574 | 1.88E-218 | 2.670 | 0 | Estrogen signaling pathway |
| FN1 | 0.534 | 0.034 | 0.123 | 1.567 | 1.80E-133 | -1.509 | 1.75E-19 | NOD-like receptor signaling pathway; Salmonella infection; Shigellosis; Yersinia infection; Rap1 signaling pathway; Bacterial invasion of epithelial cells; Human papilloma virus infection; Regulation of actin cytoskeleton; Pathways in cancer; Mitophagy - animal; Proteoglycans in cancer; Amoebiasis; Parkinson disease; Focal adhesion; Pathways of neurodegeneration - multiple diseases; Small cell lung cancer; Amyotrophic lateral sclerosis; AGE-RAGE signaling pathway in diabetic complications; PI3K-Akt signaling pathway; MAPK signaling pathway - fly; ECM-receptor interaction |
| LOC112442408 | 0.216 | 0.197 | 0.373 | -1.550 | 6.92E-52 | -1.947 | 2.61E-43 | Unknown |
| DLG2 | 0.031 | 0.05 | 0.178 | -1.711 | 8.29E-68 | -1.578 | 4.46E-28 | Human papilloma virus infection; Tight junction; Hippo signaling pathway |
| AHNAK | 0.816 | 0.601 | 0.901 | -1.724 | 5.95E-262 | -2.731 | 2.70E-283 | Salmonella infection |
| VCAN | 0.046 | 0.034 | 0.202 | -1.794 | 3.08E-66 | -2.010 | 4.58E-44 | Cell adhesion molecules |
| COL5A2 | 0.065 | 0.105 | 0.231 | -1.821 | 1.84E-64 | -1.814 | 1.63E-24 | Protein digestion and absorption |
| HMCN1 | 0.01 | 0.041 | 0.192 | -1.869 | 1.04E-108 | -1.751 | 7.28E-37 | Hemicentina |
| TSHZ2 | 0.037 | 0.049 | 0.194 | -1.873 | 1.06E-69 | -1.907 | 2.02E-33 | Hippo signaling pathway - fly |
| GPC6 | 0.011 | 0.016 | 0.161 | -1.936 | 2.48E-86 | -1.889 | 2.41E-40 | Glypican 6a |
| ABI3BP | 0.047 | 0.061 | 0.198 | -1.987 | 2.24E-63 | -2.025 | 1.19E-30 | Target of Nesh-SH3a |
| ANK2 | 0.019 | 0.073 | 0.209 | -2.144 | 4.36E-105 | -1.913 | 5.17E-29 | Toll and Imd signaling pathway; Proteoglycans in cancer |
| EBF1 | 0.085 | 0.108 | 0.423 | -2.921 | 2.06E-179 | -2.145 | 3.98E-83 | Alcoholic liver disease; AMPK signaling pathway; Insulin resistance; Insulin signaling pathway; Non-alcoholic fatty liver disease |
a meaned that the KEGG pathway from KEGG tools in website (https://www.kegg.jp/kegg/kegg2.html)
Among those DEGs in adipocytes, preadipocytes and ASCs from adipose tissues at different depots, TPT1, DST, TBX15, FILIP1L and LOC101906240 were shared. Interestingly, we also found that UCP2 was expressed in SAT, OAT and PAT; however, UCP1 and UCP3 were expressed only in SAT of young Mongolian cattle.
Enrichment of KEGG pathways of DEGs in adipose tissues at different depots
Using KEGG to analyze the marker genes of adipocyte populations, the results revealed that various pathways, including the AMPK signaling pathway, insulin signaling pathway, apelin signaling pathway, aldosterone synthesis and secretion, cAMP signaling pathway, regulation of lipolysis in adipocytes, thermogenesis, and the PPAR signaling pathway, were shared by adipocytes in SAT, OAT and PAT; several pathways, such as the mTOR signaling pathway, glycerophospholipid metabolism, calcium signaling pathway, and neuroactive ligand‒receptor interaction, were expressed only in SAT; and other pathways, such as the adipocytokine signaling pathway, nonalcoholic liver disease, longevity-regulating pathway, and type II diabetes mellitus, were shared by SAT and OAT.
Compared with those in SAT, the DEGs in PAT and OAT were enriched in different pathways. The comparison of PAT versus SAT revealed that metabolism-related pathways (such as propanoate metabolism, fatty acid metabolism and degradation of valine, leucine and isoleucine), death-related metabolic pathways (such as cancer and COVID-19), cellular biology-related pathways (especially mitophagy and autophagy, bacterial invasion of epithelial cells) and general pathways (such as the longevity regulating pathway, MAPK signaling pathway and PPAR signaling pathway) were enriched. The comparison of OAT versus SAT revealed that additional pathways, including pathways related to shigellosis, the regulation of lipolysis in adipocytes, and the PI3K–Akt signaling pathway, were enriched. Metabolism-related pathways (such as propanoate metabolism, fatty acid metabolism and degradation of valine, leucine and isoleucine), disease-related metabolism pathways (such as cancer and COVID-19), cellular biology-related pathways (such as mitophagy and autophagy, bacterial invasion of epithelial cells) and general pathways (such as the longevity-regulating pathway, MAPK signaling pathway and PPAR signaling pathway) were also enriched. The top 20 pathways of the KEGG analysis are shown in Table 7.
Table 7.
Top 20 pathways enriched for differential expression genes in adipocytes by comparison between adipose tissues at different depots
| No. | PAT VS SAT | OAT VS SAT | PAT VS OAT | |||
|---|---|---|---|---|---|---|
| Pathway | P-value | Pathway | P-value | Pathway | P-value | |
| 1 | Coronavirus disease - COVID-19 | 2.26E-19 | Coronavirus disease - COVID-19 | 5.62E-25 | Coronavirus disease - COVID-19 | 8.36E-32 |
| 2 | Mitophagy - animal | 1.93E-07 | Autophagy - animal | 2.26E-09 | Focal adhesion | 2.64E-07 |
| 3 | MicroRNAs in cancer | 2.37E-07 | Proteoglycans in cancer | 9.01E-09 | Bacterial invasion of epithelial cells | 2.21E-07 |
| 4 | EGFR tyrosine kinase inhibitor resistance | 1.71E-06 | Bacterial invasion of epithelial cells | 9.39E-08 | Proteoglycans in cancer | 2.64E-07 |
| 5 | Longevity regulating pathway | 2.37E-06 | AMPK signaling pathway | 1.23E-07 | Autophagy - animal | 5.16E-07 |
| 6 | Focal adhesion | 2.10E-06 | Focal adhesion | 1.65E-07 | AMPK signaling pathway | 6.22E-07 |
| 7 | AMPK signaling pathway | 2.76E-06 | Mitophagy - animal | 5.82E-07 | Propanoate metabolism | 1.40E-06 |
| 8 | Propanoate metabolism | 4.69E-06 | Adherens junction | 9.02E-07 | PI3K-Akt signaling pathway | 1.41E-06 |
| 9 | Autophagy - animal | 1.17E-05 | Shigellosis | 1.68E-06 | MicroRNAs in cancer | 1.76E-06 |
| 10 | Adherens junction | 1.39E-05 | Protein processing in endoplasmic reticulum | 4.41E-06 | Tight junction | 2.31E-06 |
| 11 | Fatty acid metabolism | 1.84E-05 | Longevity regulating pathway | 4.78E-06 | Adherens junction | 2.78E-06 |
| 12 | PI3K-Akt signaling pathway | 2.30E-05 | Regulation of actin cytoskeleton | 4.96E-06 | Fluid shear stress and atherosclerosis | 3.51E-06 |
| 13 | Proteoglycans in cancer | 4.89E-05 | Longevity regulating pathway - worm | 9.39E-06 | Regulation of actin cytoskeleton | 6.38E-06 |
| 14 | MAPK signaling pathway | 6.64E-05 | PI3K-Akt signaling pathway | 1.31E-05 | Shigellosis | 8.75E-06 |
| 15 | Bacterial invasion of epithelial cells | 6.45E-05 | Tight junction | 2.62E-05 | Neurotrophin signaling pathway | 1.35E-05 |
| 16 | Valine, leucine and isoleucine degradation | 0.000106199 | Pathways in cancer | 3.59E-05 | Mitophagy - animal | 2.16E-05 |
| 17 | Longevity regulating pathway - worm | 0.000113677 | Longevity regulating pathway - multiple species | 6.03E-05 | Kaposi sarcoma-associated herpesvirus infection | 2.10E-05 |
| 18 | Prostate cancer | 0.000120231 | Regulation of lipolysis in adipocytes | 7.24E-05 | Fatty acid metabolism | 2.69E-05 |
| 19 | PPAR signaling pathway | 0.000134453 | Spliceosome | 0.00010141 | Insulin signaling pathway | 2.55E-05 |
| 20 | Peroxisome | 0.000134453 | Rap1 signaling pathway | 0.000101128 | Regulation of lipolysis in adipocytes | 2.69E-05 |
To further compare the KEGG pathways in different cells in SAT, PAT and PAT, the enriched pathways and DEGs in different cells were analyzed. Compared with those in SAT, the enriched pathways for DEGs with the highest gene ratios in T cells in OAT were focused on translation, amide biosynthetic processes, organonitrogen compound biosynthetic processes, peptide metabolic processes, peptide biosynthetic processes, and cellular amide metabolic processes, which were also enriched pathways of DEGs in adipocytes (Fig. 2A), and C3 was the shared DEG in most cells whose expression was significantly upregulated in OAT (Fig. 2B). Compared with those in OAT, the pathways enriched with DEGs in preadipocytes were focused only on the cellular response to organonitrogen compounds and nitrogen compounds (Fig. 2C), and C3 was the shared DEG in most cells whose expression was significantly downregulated in PAT (Fig. 2D). Compared with those in SAT, the enriched pathways for DEGs in PAT were the same as those in OAT compared with SAT (Fig. 2E), whereas none of the DEGs were shared in most cells with significant regulation, and only FABP4 was enriched in smooth muscle-like ASCs and vascular endothelial cells (Fig. 2F).
Fig. 2.
Different pathways enriched in adipose tissues at different depots. A, C and E Bubble plots of pathways enriched in cell populations in OAT vs. SAT, PAT vs. OAT and PAT vs. SAT, respectively; B, D and F Significant DEGs in cell populations in OAT vs. SAT, PAT vs. OAT and PAT vs. SAT, respectively
Interactions of genes in adipose tissues
In adipose tissues, various innate and adaptive immune cells are involved in the regulation of insulin resistance and inflammation. We found that macrophages were the predominant immune cell type in adipose tissues in SAT, OAT and PAT, featuring the expression of the shared cell marker CD163 in the subpopulations.
Using the Cytoscape tool, CD163, a macrophage receptor, was used to construct a cell‒cell interaction network. The results revealed that CDH1 formed a node connecting macrophages with preadipocytes, which was marked by S100A4; CD36 and APOE formed nodes that connected macrophages with ASCs, which were marked by FABP4; COL1A1, CAV1 and FGF2 formed nodes that connected preadipocytes with ASCs; and PECAM1, SPP1 and CD44 were shared nodes that connected macrophages, ASCs and preadipocytes (Fig. 3A).
Fig. 3.
Interactions between genes of interest and functional verification. A Interaction network showing the centers of CD163, S100A4 and FABP4 identified in Cytoscape analysis. B Violin plot showing the expression of CD163, S100A4 and FABP4. C S100A4 expression in preadipocytes by immunocytochemistry (left : negative, right : positive). D Effect of S100A4 on the Ca2+ signal in preadipocytes. (left: Preadipocytes transfected with the vector, middle: Preadipocytes transfected with the S100A4 CDS, right: Preadipocytes transfected with siRNA-S100A4). E Effects of S100A4 on the expression of FABP4, KLF12, PTGER3 and UCP 1 ~ 3 in preadipocytes, as determined by qRT‒PCR. (left: Effects of the S100A4 CDS on the expression of FABP4, KLF12, PTGER3 and UCP 1 ~ 3, right: Effects of S100A4 siRNA on the expression of FABP4, KLF12, PTGER3 and UCPs 1 ~ 3)
S100A4 expression in preadipocytes (Fig. 3B) was verified by immunocytohistology (Fig. 3C). To further verify the effects of S100A4 on preadipocytes, S100A4 was overexpressed and downregulated (basing on the effects of 3 siRNAs targeting S100A4 shown by Supplementary Fig. 2) in preadipocytes. The results revealed that S100A4 positively regulated the Ca2+ signal in the preadipocytes (Fig. 3D). Additionally, S100A4 overexpression upregulated the expression of FABP4, UCP1 and UCP2 and downregulated the expression of PTGER3 and UCP3, which were reversed by the siRNA knockdown of S100A4 (Fig. 3E).
Differentiation trajectories and gene regulation across pseudotime
To analyze the pseudotime, the cells relative to adipose tissues were clustered according to the markers LIPE/PLIN1/ADIPOQ in adipocytes, COL3A1/COL1A1/COL1A2 in preadipocytes, and PDGFRA/CD34/LAMA2 in ASCs (Fig. 4A and B).
Fig. 4.
Differentiation of different adipose-associated cells and pseudotime analysis. A UMAP display of subcell types of adipose-associated cells. B Feather plot of marker gene expression in cell types of adipose-associated cells. The markers used were as follows: LIPE/PLIN1/ADIPOQ in adipocytes, COL3A1/COL1A1/COL1A2 in preadipocytes, and PDGFRA/CD34/LAMA2 in ASCs. C Nuclei assigned to the differentiation path of the trajectory analysis and the degrees of differentiation (red represents a lower degree of differentiation, and blue represents a higher degree of differentiation). The higher the value represented the lower degree of differentiation. D Differentiation degree analysis using CytoTRACE for the calculation of 5000 cells. E Pseudotime analysis of adipose-associated cells in SAT based on cell types to display the cell differentiation trajectory. F Pseudotime analysis of adipose-associated cells in SAT based on cell clusters to display the cell differentiation trajectory. G Heatmaps generated with Monocle2 software were used to perform pseudotime analysis of adipose-associated cell types, and 5 groups with similar gene expression trends across different subtypes of cells according to pseudotime were clustered. H Heatmaps of the expression trends of the top 5 differentially expressed genes in the 5 clusters across pseudotime in SAT, with the x-axis representing the pseudotime and the y-axis representing the level of gene expression
Monocle2 software was used to analyze the pseudotime trajectory, and the results revealed that the nuclei were distributed throughout the entire adipogenic trajectory, with the adipogenic differentiation trajectories leading from ASCs to preadipocytes and then to mature adipocytes (Fig. 4C). However, the degree of differentiation of adipocytes was weak, whereas that of preadipocytes was strong (Fig. 4D).
To explore these trajectories further, DEGs of cells in SAT of Mongolian cattle were identified across the trajectory and clustered. Based on the feature gene expression patterns and the changes in gene expression in each cell across pseudotime, the trajectories of adipose-associated cells were depicted, and the dynamic programs in SAT of Mongolian cattle were revealed (Fig. 4E and F). These genes were clustered into five groups based on their temporal expression patterns (Fig. 4G). The top 5 genes highly expressed in each cluster across the trajectory (from ASC to preadipocyte and then to adipocyte) were G0S2, LIPE, LPIN1, PTGER3 and RGCC in Cluster 1, with expression trends from low to high; ANXA1, COL4A2, FABP4, pyruvate dehydrogenase kinase 4 (PDK4) and SOX5 in Cluster 2, with expression trends from high to low to high; ADIRF, HSPB1, MYL6, SDBS and VIM in Cluster 3, with expression trends from low to lower to low; COL1A1, COL1A2, COL3A1, DST and PTMS in Cluster 4, with expression trends from low to high to low; and ANTXR2, ARHGAP21, DOCK9, LRP6 and VPS13C in Cluster 5, with expression trends from low to low to lower (Fig. 4H).
Lipid accumulation and the expression of UCP1 ~ 3, CD4 and CD8 in adipose tissues
The snRNA-seq results revealed that lipid accumulation might differ in adipose tissues at different depots in Mongolian cattle and that lipid accumulation was affected by inflammatory factors. Oil red O staining can be used to monitor intracellular lipid accumulation [24]. In adipose tissues at different anatomical depots, the arrangement, size and lipid droplets of adipocytes varied. The adipocytes in PAT were larger than those in OAT and SAT. The results of gene expression analyses in adipose tissues at different depots revealed that UCP2 was expressed in all cell compositions of those adipose tissues; UCP1 and UCP3 were expressed in both adipocytes and ASCs; and UCP1 was also expressed in preadipocytes and vascular endothelial cells in SAT (Fig. 5). Here, the immunohistochemistry results revealed that UCP1 ~ 3, CD4 and CD8 were expressed in the adipose tissues of Mongolian cattle with different levels of positive signal.
Fig. 5.
Distribution of lipid droplets and UCP1 ~ 3, CD4 and CD8 in adipose tissues at different adipose depots of Mongolian cattle. First row: Lipid accumulation in adipose tissues in SAT, PAT and OAT. Second to fourth row: Distribution of UCP1 ~ 3 in adipose tissues in SAT, PAT and OAT, as determined by immunohistochemistry, respectively. Fifth to sixth row: Distribution of CD4+ and CD8+T cells in adipose tissues in SAT, PAT and OAT, as determined by immunohistochemistry, respectively. Last row: Negative control for immunohistochemistry; IgG was used as the primary antibody
Discussion
Characteristics of the cell composition of adipose tissues at different depots in Mongolian cattle
The adipose tissues in SAT, PAT and OAT of Mongolian cattle were composed of 10 types of cells. However, there were differences in the types, subtypes and numbers at different depots, which determined the function of the adipose tissues. Adipocytes constitute one of the four kinds of cell populations shared at different depots with different percentages and different subtypes. The definitions of the subpopulations of adipocytes in humans are as follows: (1) ACACA+ and Acly+ lipogenic adipocytes (LGAs), which are related to de novo fatty acid synthesis; CD36+ and Apoe+ lipid-scavenging adipocytes (LSAs), which are related to lipid uptake and transport; and (3) Hif1a+ and Rab7+ stressed LSAs (SLSAs), which are related to hypoxia and autophagy [25]. Through the identification of adipocytes using markers, 4 subtypes of adipocytes expressing those markers in humans were identified in the adipose tissues of Mongolian cattle, with an additional subtype marked by PTGER3 (EP3) in SAT, which is related to lipid deposition and diet-induced insulin resistance [26]. These results suggest that PTGER3 might act as a marker to identify lipid-deposition adipocytes and that PTGER3 has a greater ability to deposit fat in SAT than in OAT and PAT. However, the subpopulation of adipocytes in PAT seemed to lack SLSAs, which suggested that PAT might not be responsible for stress treatment.
Among the ASPCs of Mongolian cattle, the ASC and preadipocyte subtypes were identified, but there were different subpopulations in adipose tissues at different depots; in particular, no preadipocytes were identified in PAT. ASPCs play important roles in facilitating lipogenesis, fatty acid metabolism and specific fatty acids in livestock [13]. The cellular basis for determining the fat content in PAT in Mongolian cattle is lacking.
In addition to the classical description of macrophages, perivascular, inflammatory and lipid-associated macrophages are referred to as subtypes [27]. In PAT and OAT in Mongolian cattle, macrophages were identified with at least 2 subpopulations with markers of antigen presentation and inflammation-related markers, suggesting that SAT and OAT function in the immune response or defense by being recruited and adhering to sites of inflammation to clear bacteria or other residues through phagocytosis. In humans and mice, obese individuals have greater inflammatory activity in OAT than in SAT, which is correlated with insulin resistance [28]. Therefore, macrophages are widely distributed in adipose tissues at different depots, where they play individual roles or interact with adipogenic cells.
The anatomy of the kidney is specific, which is separated from the peritoneum and embedded in adipose tissue. The types of immune cells enriched in PAT in Mongolian cattle (such as NK, DC, and B cells and vascular and lymphatic endothelial cells) might provide protection in the correct microenvironment.
Differences in KEGG pathways and DEGs of interest in some adipose tissue cells at different adipose depots in Mongolian cattle
Comparison of the top 20 pathways enriched for the marker genes in adipocytes at different depots revealed that fatty acid metabolism, the PPAR signaling pathway, peroxisomes, the insulin signaling pathway, and the regulation of lipolysis in adipocytes, which are related to energy conservation and insulin action, were differentially regulated in the adipocytes of Mongolian cattle.
Interestingly, compared with SAT, one of the KEGG pathways in PAT and OAT was the longevity-regulating pathway, which is directly linked to β-cell function and type 2 diabetes (T2D) onset [29] through insulin released by β cells [30] to decrease glucose levels in the blood by stimulating the uptake of blood glucose in fat, muscle, liver, and intestinal cells [31], suggesting the crucial role of the adipocyte subpopulation in metabolic glucose homeostasis during longevity. In this pathway, aldosterone not only is responsible for regulating sodium homeostasis to control blood volume and pressure but also has a role in insulin resistance [31]. In addition to salt balance and blood pressure, aldosterone affects BAT [32] and promotes the differentiation of T37i and 3T3-L1 cells into (brown) adipocytes [33]. The dual functions of aldosterone in the longevity-regulating pathway not only help us understand the importance of PAT and OAT in Mongolian cattle but also provide clues for existing or differential brown adipocytes in PAT and OAT. The specific role of aldosterone in BAT remains to be elucidated.
Among the DEGs, the third component of complement protein (C3), which was significantly differentially regulated in SAT, was of interest. Considering the activation of the complement system, the classical pathway links the innate immune system to the adaptive immune system, and the alternative and lectin pathways are involved mainly in natural resistance to microorganisms in the absence of specific antibodies [34]. Because C3 is a central and indispensable component in various pathways and is expressed in several tissues and cells [35], it acts as a candidate gene that is naturally resistant to microorganisms [34]. Although C3 is synthesized mainly in the liver, as shown by the scRNA-seq data, C3 was expressed in most of the cells in SAT of Mongolian cattle, except in adipocytes and SMC-like ASCs, suggesting that C3 synthesized in SAT provided the ability to defend against microorganisms. Furthermore, the functional role of C3 in the complement system is important for both the innate and adaptive immune systems, as it is split into the vasoactive and chemotactic peptides C3a and C3b, which lead to the recruitment of both cellular and soluble mediators of inflammation [36], including proinflammatory and anti-inflammatory mediators. Moreover, C3 inhibits the apoptosis of inflammatory cells to cause a cascade of inflammatory reactions, resulting in the accumulation of lipids through mitochondrial damage. Overall, C3 in SAT would be an important candidate gene that would contribute to susceptibility to infections and be involved in adipose deposition.
In the inner mitochondrial membrane of adipocytes, carrier proteins such as the uncoupling protein (UCP) family control the exchange of substrates between the mitochondrial matrix and the cytoplasm [37]. UCP1, which is specifically expressed in BAT, is in charge of thermogenic activity [38] through dissociating the mitochondrial proton gradient from adenosine triphosphate (ATP) synthesis [39]. Once UCP1 meets other invalid energy cycling programs, brown adipocytes expend excess energy for thermogenesis [40]. Brown adipose tissues are typically enriched in young mammals. The scRNA-seq results revealed that UCP1 was differentially expressed in the ASCs and preadipocytes of young Mongolian cattle, which suggested that UCP1 was more essential for thermogenesis. UCP2 is an ion/anion transporter that is widely expressed in various tissues (including pancreatic islets and WAT) to regulate cellular metabolism, oxidative stress, cell proliferation and cell death [41]. UCP2 is also essential for fatty acid oxidation [42], and polymorphisms in UCP2 are correlated with diabetes and obesity because UCP2 is involved in energy expenditure and insulin secretion [41]. Additionally, significant correlations between the UCP2 and UCP3 genes and milk production traits have been identified in Jersey cows and Polish Holstein cows [43]. In the adipose tissue of Mongolian cattle, UCP2 was widely expressed in all cells. Upon verifying the expression of the UCP family in the adipose tissues of Mongolian cattle, oil red O staining and UCP1 ~ 3 expression at different depots suggested that the SAT of Mongolian cattle might be brown and that adipocytes might be sensitive to insulin.
Developmental trajectories of adipogenic cells in the adipose tissues of Mongolian cattle
Based on the scRNA-seq results, the composition and subpopulations of cells were distinguished by cell markers, which represented different cell states with similar developmental trajectories. Here, we revealed that the significantly differential expression of Cluster 1 genes (including G0S2, LIPE, LPIN1, PTGER3 and RGCC) was consistent with the cell trajectory from ASCs to preadipocytes to adipocytes in the SAT of Mongolian cattle. G0S2 (G0/G1 switch gene 2), which acts as a regulator of adipogenesis through peroxisome proliferator-activated receptor γ (PPARγ) [44] and directly interacts with ATGL and thereby inhibits its TG hydrolase activity [45], was expressed at relatively high levels in adipose tissues. G0S2 overexpression accelerated the differentiation of preadipocytes to mature adipocytes [44]. Lpin1, which is a member of the lipin family, with essential roles in lipid biosynthesis, has effects on differentiating adipocytes [46]. During cold exposure, Lpin1 is induced and contributes to thermogenic activation through the transcriptional induction of PPAR, PGC-1, and, consequently, UCP1 in BAT [24, 46] and is also required for the maintenance and function of mature brown adipocytes [24]. Cell trajectory analysis indicated that adipocytes were derived from ASCs and preadipocytes; however, the differentiation of adipocytes was weak, suggesting that the adipocytes had potential differentiation activity. The subpopulation with Lpin1 expression suggested that adipocytes in the SAT of Mongolian cattle might be plastic, allowing them to differentiate from white to brown, contributing to or resulting from adaptation to their harsh living conditions. In particular, in cold environments, Lpin1 expression would be induced, followed by UCP1 expression, which resulted in increased insulin sensitivity to glucose.
In addition, the gene expression in Cluster 2 (including ANXA1, COL4A2, FABP4, PDK4 and SOX5) might contribute to the developmental trajectories of adipogenic cells in the visceral adipose tissue of Mongolian cattle. PDK4 regulates the bioavailability of pyruvate for thiazolidinedione-activated glyceroneogenesis [47]. In enterocytes, fatty acids and monoacylglycerol hydrolyzed from triacylglycerol (TAG) are taken up and resynthesized into TAG at the ER membrane for further secretion into the blood by packaging in chylomicrons (CMs) or temporary storage within enterocytes by packaging in cytoplasmic lipid droplets (CLDs) [48]. The synthesis enzymes specific for TAG, acyl CoA: DGAT1 and DGAT2, participate in the TAG package for CM and CLD synthesis to regulate dietary fat absorption [49]. Moreover, DGAT1 is involved in retinol metabolism, glycerolipid metabolism, and fat digestion [50, 51]. Among the pathways associated with glycerolipid metabolism and fat digestion and absorption in SAT of Mongolian cattle, PLPP1, DGAT1 and DGAT2 were three shared members of both pathways. Although only adipocytes specialize in TAG storage [52], the expression of DGAT1 and DGAT2 seemed to suggest that TAG synthesis occurs only in adipocytes in SAT of Mongolian cattle. DGAT2 compensates for DGAT1 function [53]. DGAT2 and DGAT1 overexpression has been correlated with large and small lipid droplet formation, respectively [54], suggesting the important function of DGAT2 in TAG synthesis. Moreover, as a marker of adipocytes in SAT, DGAT2 interacts with autophagy-related gene 10 (ATG10) and BCL2 in adipose tissue, as indicated by the PPI analysis; retinol metabolism was not detected in SAT of Mongolian cattle. FABP4 is highly expressed in adipocytes and macrophages and can regulate the release and transport of fatty acids [55], suggesting that the homeostasis maintained by FABP4 would be essential for the developmental trajectories of adipogenic cells in the visceral adipose tissue of Mongolian cattle.
Interactions of adipogenic cells with immune cells and between molecules in the adipose tissues of Mongolian cattle
In the context of adipose tissue functions, immune cells play important roles. The roles of macrophages and their interactions with preadipocytes and adipocytes in regulating the proliferation, differentiation and lineage commitment of adipocyte progenitors could provide evidence of their contributions to the functions of adipose tissues. CD163 is a scavenger receptor present on the surface of tissue macrophages [56] and is a marker of macrophages with anti-inflammatory properties [57]. Using CD163 as a target protein, we found that FABP4 and S100A4 might likewise be targets, suggesting that there might be crosstalk between macrophages and adipocytes or preadipocytes through the expression of important genes that present antigens in adipose tissue. The expression of S100A4 in preadipocytes resulted in positive regulation of the expression of Ca2+ influx, FABP4, UCP1 and UCP2 and negative regulation of the expression of the prostaglandin E2 (PGE2) receptor 3 gene (PTGER3, also referred to as EP3) and UCP3. FABP4 takes part in lipogenesis, and excess secretion of free fatty acids and TAG causes ectopic lipid accumulation, resulting in the consequent development of metabolic syndrome [58]. UCP2 is expressed in several immune cells [59, 60] and plays roles in the immune cell activity in various diseases, including atherosclerosis and inflammatory and infectious diseases in mice and humans [41]. We found that UCP2 was expressed in SAT, OAT and PAT, with TLR4 expressed only in PAT. UCP2 controls reactive oxygen species (ROS) production and macrophage activity and is involved in Toll-like receptor 4 (TLR4)-induced ROS signaling in primary macrophages [37]. Regardless of the cold resistance of Mongolian cattle, UCP2 might play functional roles through its wide expression in adipose tissues.
In adipose tissue, CD1d presents a lipid antigen that is highly expressed in adipocytes relative to SVCs [28]. Adipocytes expressing MHC I and MHC II can mediate CD8+ and CD4+ T-cell responses, respectively. Additionally, adipocytes highly expressing CD1d can regulate the function and activation of iNKT cells in adipose tissues. Compared with those in insulin-resistant obese individuals, the proportions of naive CD4+ and CD8+ T cells were significantly greater in insulin-sensitive obese individuals [61]. We found CD4+ and CD8+ T cells in the visceral adipose tissues of Mongolian cattle, with strong signals of CD8+ T cells in PAT and CD4+ T cells in SAT and PAT, which suggested that these cells contribute to insulin sensitivity, especially in PAT. In addition to the fact that macrophage accumulation in adipose tissue is correlated with insulin resistance [62], an immune microenvironment was constructed in the adipose tissues of Mongolian cattle.
In intracellular Ca2+ dynamics, the endoplasmic reticulum (ER) and mitochondria act as Ca2+ sources and Ca2+ sinks, respectively, which play various roles in cell signal transduction [63]. Ca2+ is released from mitochondria through the activation of UCP1 by the stimulation of β3-adrenerg, which further activates plasmalemmal Ca2+ entry [64]. In the preadipocytes of Mongolian cattle, the expression of S100A4 affected Ca2+ signaling and the expression of UCP1 ~ 3, FABP4 and PTGER3. FABP4 is a multifunctional gene, and high levels of FABP4 in circulation and/or local expression in the epicardial fat tissue of obese individuals may partially result in the development of heart dysfunction [65]. In adipocytes, PGE2 inhibits lipolysis via EP3 [26, 66], whereas EP3 signaling positively regulates lipid deposition, leading to ectopic lipid distribution and subsequent insulin resistance induced by diet [26]. Therefore, S100A4 might be positively correlated with insulin resistance and metabolism through FABP4 and PTGER3. Moreover, in intact cells, UCP2/3-dependent and UCP2/3-independent mitochondrial Ca2+ sequestration meets adequate organelle Ca2+ demands from different Ca2+ sources [67]. The accumulation of Ca2+ stimulated by S100A4 occurred in a similar manner to that of UCP2 but reversed with UCP3, suggesting the existence of another pathway to regulate the Ca2+ or UCP2/UCP3 mechanism to balance Ca2+, which needs further research.
Conclusion
The heterogeneity of the cell populations, especially immune cells and adipogenic cells, and the genes expressed in adipose tissues at different depots of Mongolian cattle not only determined the endocrine and energy storage functions of adipose tissues, but contributed to adapting to cold and disease resistance. In particular, S100A4, might be target gene for repairing abnormal lipid deposition and insulin resistance.
Supplementary Information
Authors’ contributions
C.Z., F.R., M.H., designed and wrote the main manuscript. C.Z., Y.H., J.C., S.Z., W.N., performed experiments. C.Z., J.Q., R.H., processed and analyzed the data. M.B., Z.L., Q.G., A.S., W.J., A.D., fed cattle, collected samples and supplied the formation of cattle. H.J., Z.J., reviewed and edited the manuscript.
Funding
This projected is funded by the “Revelation and Leadership” project of the Department of Science and Technology of Inner Mongolian Autonomous Region (2022JBGS0023).
Data availability
Sequence data that support the findings of this study have been deposited in NCBI Gene Expression Omnibus (GEO) with the primary accession code GSE264101 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE264101). The data can be allowed to review with the token “exudwmyiltadjez”.
Declarations
Ethics approval and consent to participate
Cattle care and treatment was approved by the Animal Ethics Committee SXAU-EAW-2022MC.OA.008001171 at Shanxi Agricultural University (Taigu, Shanxi, China). All informed consent has been obtained to use these animals in the experiment.
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Contributor Information
Ruiwen Fan, Email: ruiwenfan@163.com.
Muren Herrid, Email: mherrid@gmail.com.
References
- 1.Song B, Di S, Cui S, Chen N, Wang H, Wang X, Gao Q, Tong G, Wang H, Huang X, Ding L, Gao Y, Liu J, Wang X. Distinct patterns of PPARγ promoter usage, lipid degradation activity, and Gene expression in Subcutaneous Adipose tissue of lean and obese Swine. Int J Mol Sci. 2018;19(12):3892–904. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Hausman GJ, Basu U, Wei S, Hausman DB, Dodson MV. Preadipocyte and adipose tissue differentiation in meat animals: influence of species and anatomical location. Annu Rev Anim Biosci. 2014;2(1):323–51. [DOI] [PubMed] [Google Scholar]
- 3.Bluher M. Adipokines—removing road blocks to obesity and diabetes therapy. Mol Metab. 2014;3:230–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Sadovsky R. Adipose tissue as an endocrine organ. Mol Cell Endocrinol. 2004;18(1):41–58.
- 5.Friedman JM. Leptin and the Endocrine Control of Energy Balance. Nat Metab. 2019;1(8):754–64. [DOI] [PubMed] [Google Scholar]
- 6.Larabee CM, Neely OC, Domingos AI, Obesity. A neuroimmunometabolic perspective. Nat Rev Endocrinol. 2020;16(1):30–43. [DOI] [PubMed] [Google Scholar]
- 7.Cohen P, Kajimura S. The cellular and functional complexity of thermogenic fat. Nat Rev Mol Cell Biol. 2021;22(6):393–409. [DOI] [PMC free article] [PubMed]
- 8.Mezentseva NV, Kumaratilake JS, Newman SA. The brown adipocyte differentiation pathway in birds: an evolutionary road not taken. BMC Biol. 2008;6:17. 10.1186/1741-7007-6-17. [DOI] [PMC free article] [PubMed]
- 9.Ong WK, Chakraborty S, Sugii S. Adipose Tissue: Understanding the Heterogeneity of Stem Cells for Regenerative Medicine. Biomolecules. 2021;11(7):918. 10.3390/biom11070918. [DOI] [PMC free article] [PubMed]
- 10.Liu SQ, Chen DY, Li B, Gao ZJ, Feng HF, Yu X, Liu Z, Wang Y, Li WG, Sun S, Sun SR, Wu Q. Single-cell analysis of white adipose tissue reveals the tumor-promoting adipocyte subtypes. J Transl Med. 2023;21(1):470. 10.1186/s12967-023-04256-7. [DOI] [PMC free article] [PubMed]
- 11.Chang E, Varghese M, Singer K. Gender and Sex Differences in Adipose Tissue.Curr Diab Rep. 2018;18(9):69. 10.1007/s11892-018-1031-3. [DOI] [PMC free article] [PubMed]
- 12.Ferrero R, Rainer P, Deplancke B. Toward a Consensus View of mammalian adipocyte stem and progenitor cell heterogeneity. Trends Cell Biol. 2020;30:937–50. [DOI] [PubMed]
- 13.Ishida Y, Mabuchi Y, Naraoka Y, Hisamatsu D, Akazawa C. Conservation of markers and stemness in adipose stem and progenitor cells between cattle and other species. Int J Mol Sci. 2023;24(15):11908. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sampaio RV, Chiaratti MR, Santos DCN, Bressan FF, Sangalli JR, Sá ALA, Silva TVG, Costa NN, Cordeiro MS, Santos SSD, et al. Generation of bovine (Bos indicus) and Buffalo (Bubalus bubalis) adipose tissue derived stem cells: isolation, characterization, and Multipotentiality. Genet Mol Res. 2015;14:53–62. [DOI] [PubMed] [Google Scholar]
- 15.Altun I, Yan X, Ussar S. Immune cell regulation of white adipose progenitor cell fate. Front Endocrinol. 2022;13:859044. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Ruggiero AD, Key CC, Kavanagh K. Adipose tissue macrophage polarization in healthy and unhealthy obesity. Front Nutr. 2021;8:625331. 10.3389/fnut.2021.625331. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hackstein H, Thomson AW. Dendritic cells: emerging pharmacological targets of immunosuppressive drugs. Nat Rev Immunol. 2004;4:24–34. [DOI] [PubMed] [Google Scholar]
- 18.Huh JY, Park YJ, Ham M, Kim JB. Crosstalk between adipocytes and Immune cells in adipose tissue inflammation and metabolic dysregulation. Obesity. Molecules Cells. 2014;37(5):365–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Jager A, Kuchroo VK. Effector and regulatory T-cell subsets in autoimmunity and tissue inflammation. Scand J Immunol. 2010;72:173–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Butler A, Hoffman P, Smibert P, Papalexi E, Satija R. Integrating single-cell transcriptomic data across different conditions, technologies, and species. Nat Biotechnol. 2018;36(5):411–20. [DOI] [PMC free article] [PubMed]
- 21.Hao Y, Hao S, Andersen-Nissen E, et al. Integrated analysis of multimodal single-cell data. Cell. 2021;184(13):3573–87. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Cao Y, Wang X, Peng G. SCSA: a cell type annotation tool for single-cell RNA-seq data. Front Genet. 2020;11:490. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Gulati GS, Sikandar SS, Wesche DJ, Manjunath A, Bharadwaj A, Berger MJ, Ilagan F, Kuo AH, Hsieh RW, Cai S, Zabala M, Scheeren FA, Lobo NA, Qian D, Yu FB, Dirbas FM, Clarke MF, Newman AM. Single-cell transcriptional diversity is a hallmark of developmental potential. Science. 2020;367(6476):405–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Nadra K, Medard JJ, Mul JD. Cell Autonomous Lipin 1 function is essential for development and maintenance of White and Brown Adipose Tissue. Mol Cell Biol. 2012;32(23):4794–810. [DOI] [PMC free article] [PubMed]
- 25.Maniyadath B, Zhang Q, Gupta RK, Mandrup S. Adipose tissue at single-cell resolution. Cell Metab. 2023;35(3):386–413. [DOI] [PMC free article] [PubMed]
- 26.Ceddia RP, Lee D, Maulis MF, Carboneau BA, Threadgill DW, Poffenberger G, Milne G, Boyd KL, Powers AC, McGuinness OP, Gannon M, Breyer RM. The PGE2 EP3 Receptor Regulates Diet-Induced Adiposity in Male Mice. Endocrinology. 2016;157(1):220–32. [DOI] [PMC free article] [PubMed]
- 27.Hildreth AD, Ma F, Wong YY, Sun R, Pellegrini M, O’Sullivan TE. Single-cell sequencing of Human White Adipose tissue identifies New Cell States in Health and obesity. Nat Immunol. 2021;22(5):639–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Singh R, Xiang Y, Wang Y, Baikati K, Cuervo AM, Luu YK, Tang Y, Pessin JE, Schwartz GJ, Czaja MJ. Autophagy regulates adipose mass and differentiation in mice. J Clin Invest. 2009;119:3329–39. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Liu G, Li Y, Zhang T, Li M, Li S, He Q, Liu S, Xu M, Xiao T, Shao Z, Shi W, Li W. Single-cell RNA Sequencing Reveals Sexually Dimorphic Transcriptome and Type 2 Diabetes Genes in Mouse Islet β Cells. Genomics Proteomics Bioinformatics. 2021;19(3):408–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Akintola AA, van Heemst D. Insulin, aging, and the brain: mechanisms and implications. Front Endocrinol. 2015;6:13. 10.3389/fendo.2015.00013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Bollag WB. Regulation of Aldosterone Synthesis and. Secretion. Compr Physiol. 2014;4(3):1017–55. [DOI] [PubMed] [Google Scholar]
- 32.Zennaro MC, Menuet DL, Viengchareun S, Walker F, Ricquier D, Lombè M. Hibernoma development in transgenic mice identifies brown adipose tissue as a novel target of aldosterone action. J Clin Invest. 1998;101(6):1254–60. [DOI] [PMC free article] [PubMed]
- 33.Rondinone CM, Robbard D, Baker ME. Aldosterone stimulates differentiation of mouse 3T3-L1 cells into adipocytes. Endocrinology. 1993;132:2421–6. [DOI] [PubMed] [Google Scholar]
- 34.Wimmers K, Mekchay S, Schellander K, Ponsuksili S. Molecular characterization of the pig C3 gene and its association with complement activity. Immunogenetics. 2003;54:714–24. [DOI] [PubMed] [Google Scholar]
- 35.Volanakis JE. Transcriptional regulation of complement genes. Annu Rev Immunol. 1995;13:277–305. [DOI] [PubMed] [Google Scholar]
- 36.Sacks S, Zhou W, Campbell RD, Martin J. C3 and C4 gene expression and interferon-gamma-mediated regulation in human glomerular mesangial cells. Clin Experimental Immunol. 1993;93(3):411–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Emre Y. Tobias Nübel. Uncoupling protein UCP2: when mitochondrial activity meets immunity. FEBS Lett. 2010;584(8). 10.1016/j.febslet.2010.03.014. [DOI] [PubMed]
- 38.Nicholls DG. The physiological regulation of uncoupling proteins. Biochim Biophys Acta. 2006;1757:459–66. [DOI] [PubMed] [Google Scholar]
- 39.Steinbring J, Graja A, Jank AM, Schulz TJ. Flow Cytometric Isolation and Differentiation of Adipogenic Progenitor Cells into Brown and Brite/Beige Adipocytes. Methods Mol Biol. 2017;1566:25. 10.1007/978-1-4939-6820-6_4. [DOI] [PubMed]
- 40.Nedergaard J, Cannon B. Brown adipose tissue as a heatproducing thermoeffector. Handb Clin Neurol. 2018;156:137–52. 10.1016/B978-0-444-63912-7.00009-6. [DOI] [PubMed] [Google Scholar]
- 41.Li J, Jiang R, Cong X, Zhao Y. UCP2 gene polymorphisms in obesity and diabetes, and the role of UCP2 in cancer. FEBS Lett. 2019;593(18). 10.1002/1873-3468.13546. [DOI] [PubMed]
- 42.Pecqueur C, Bui T, Gelly C, Hauchard J, Barbot C, Bouillaud F, Ricquier D, Miroux B, Thompson CB. Uncoupling protein-2 controls proliferation by promoting fatty acid oxidation and limiting glycolysis- derived pyruvate utilization. FASEB J. 2008;22:9–18. [DOI] [PubMed] [Google Scholar]
- 43.Kowalewska-Łuczak I, Głosińska JI. Czerniawska-Piatko-ws- kaE.Efect of UCP2 and UCP3 genes polymorphisms on functional traits in dairy catle. Rusian J Genet. 2018;54(7):853–7. [Google Scholar]
- 44.Choi H, Lee H, Kim TH, Kim HJ, Lee YJ, Lee SJ, Yu JH, Kim D, Kim KS, Park SW, Kim JW. G0/G1 switch gene 2 has a critical role in adipocyte differentiation. Cell Death Differ. 2014;21(7):1071–80. 10.1038/cdd.2014.26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Cerk IK, Salzburger B, Boeszoermenyi A, Heier C, Pillip C, Romauch M, Schweiger M, Cornaciu I, Lass A, Zimmermann R, Zechner R, et al. A peptide derived from G0/G1 switch gene 2 acts as noncompetitive inhibitor of adipose triglyceride lipase. J Biol Chem. 2014;289:32559–70. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Sembongi H, Miranda M, Han GS, Fakas S, Grimsey N, Vendrell J, Carman GM, Siniossoglou S. Distinct roles of the Phosphatidate Phosphatases Lipin 1 and 2 during adipogenesis and lipid Droplet Biogenesis in 3T3-L1 Cells. J Biol Chem. 2013;288(48):34502–13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Cadoudal T, Distel E, Durant S, Fouque F, Blouin JM, Collinet M, Bortoli S, Forest C, Benelli C. Pyruvate dehydrogenase kinase 4: regulation by thiazolidinediones and implication in glyceroneogenesis in adipose tissue. Diabetes. 2008;57:2272–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 48.D’ Aquila T, Hung YH, Carreiro A, Buhman KK. Recent discoveries on absorption of dietary fat: presence, synthesis, and metabolism of cytoplasmic lipid droplets within enterocytes. Biochim et biophssica acta. 2016;1861:730–47. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Hung Y-H, Carreiro AL, Buhman KK. Dgat1 and Dgat2 regulate enterocyte triacylglycerol distribution and alter proteins associated with cytoplasmic lipid droplets in response to dietary fat. Biochim Biophys Acta. 2017;1862(6):600–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50.Yen C-LE, Stone SJ, Koliwad S, Harris C, Farese RV. Jr. Thematic review series: glycerolipids. DGAT enzymes and triacylglycerol biosynthesis. J Lipid Res. 2008;49:2283–301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Kim S, Lim B, Cho J, Lee S, Dang C-G, Jeon J-H, Kim J-M, Lee J. Genome-Wide Identification of Candidate Genes for Milk Production Traits in Korean Holstein Cattle. Animals. 2021;11:1392. 10.3390/ani11051392. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Zammit VA, Buckett LK, Turnbull AV, Wure H, Proven A. Diacylglycerol acyltransferases: Potential roles as pharmacological targets. Pharmacol Ther.2008;118(3):295–302. 10.1016/j.pharmthera.2008.03.010. [DOI] [PubMed]
- 53.Balogun KA, Cheema SK. Dietary Omega-3 fatty acids prevented adipocyte hypertrophy by downregulating DGAT-2 and FABP-4 in a sex-dependent. Fashion. Lipids. 2016;51(1):25. 10.1007/s11745-015-4105-x. [DOI] [PubMed] [Google Scholar]
- 54.Stone SJ, Myers HM, Watkins SM, Brown BE, Feingold KR, Elias PM, Farese RV Jr. Lipopenia and skin bar- rier abnormalities in DGAT2-deficient mice. J Biol Chem. 2004;279:11767–76. [DOI] [PubMed] [Google Scholar]
- 55.Boord JB, Fazio S, Linton MF. Cytoplasmic fatty acid- binding proteins: emerging roles in metabolism and atheroscle- rosis. Curr Opin Lipidol. 2002;13:141–7. [DOI] [PubMed] [Google Scholar]
- 56.Kristiansen M, Graversen JH, Jacobsen C, Sonne O, Hoffman HJ, Law SKSK. Moestrup, identification of the haemoglobin scavenger receptor. Nature. 2001;409(6817):198–201. [DOI] [PubMed] [Google Scholar]
- 57.Rubio-Navarro A, Amaro Villalobos JM, Lindholt JS, Buendía I, Egido J, Blanco-Colio LM, Samaniego R, Meilhac O, Michel JB, Martín-Ventura JL, Moreno JA. Hemoglobin induces monocyte recruitment and CD163-macrophage polarization in abdominal aortic aneurysm. Int J Cardiol. 2015;201:66–78. [DOI] [PubMed]
- 58.Rasouli N, Molavi B, Elbein SC, Kern PA. Ectopic fat accumulation and metabolic syndrome. Diabetes Obes Metab. 2007;9:1–10. [DOI] [PubMed] [Google Scholar]
- 59.Rousset S, Emre Y, Join-Lambert O, Hurtaud C, Ricquier D, Cassard- Doulcier AM. The uncoupling protein 2 modulates the cytokine balance in innate immunity. Cytokine. 2006;35:135–42. [DOI] [PubMed] [Google Scholar]
- 60.Tagen M, Elorza A, Kempuraj D, Boucher W, Kepley CL, Shirihai OS, Theoharides TC. Mitochondrial uncoupling protein 2 inhibits mast cell activation and reduces histamine content. J Immunol. 2009;183:6313–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 61.Sbierski-Kind J, Goldeck D, Buchmann N, Spranger J, Volk HD, Steinhagen-Thiessen E, Pawelec G, Demuth I, Spira D. T cell phenotypes Associated with insulin resistance: results from the Berlin Aging Study II. Immun Ageing. 2020;17(1):40. 10.1186/s12979-020-00211-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 62.Klöting N, Fasshauer M, Dietrich A, Kovacs P, Schön MR, Kern M, Stumvoll M, Blüher M. Insulin-sensitive obesity. Am J Physiol Endocrinol Metab. 2010;299(3):E506–15. [DOI] [PubMed] [Google Scholar]
- 63.Carafoli E. Intracellular calcium homeostasis. Annu Rev Biochem. 1987;56:395–433. [DOI] [PubMed] [Google Scholar]
- 64.Hayato R, Higure Y, Kuba M, Nagai H, Yamashita H, Kuba K. β3-Adrenergic activation of sequential Ca2+ release from mitochondria and the endoplasmic reticulum and the subsequent Ca2+ entry in rodent brown adipocytes. Cell Calcium. 2011;49(6):400-14. SPACEadipocytes. Cell Calcium. 2011;49(6):400–14. 10.1016/j.ceca.2011.02.011. [DOI] [PubMed]
- 65.Lamounier-Zepter V, Look C, Alvarez J, Christ T, Ravens U, Schunck WH, Ehrhart-Bornstein M, Bornstein SR, Morano I. Adipocyte fatty acid-binding protein suppresses cardiomyocyte contraction: a new link between obesity and heart disease. Circulation Res. 2009;105(4):326–34. [DOI] [PubMed] [Google Scholar]
- 66.Iyer A, Lim J, Poudyal H, Reid RC, Suen JY, Webster J, Prins JB, Whitehead JP, Fairlie DP, Brown L. An inhibitor of phospholipase A2 group IIA modulates adipocyte signaling and protects against dietinduced metabolic syndrome in rats. Diabetes. 2012;61(9):2320–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 67.Waldeck-Weiermair M, Malli R, Naghdi S, Trenker M, Kahn MJ, Graier WF. The contribution of UCP2 and UCP3 to mitochondrial ca(2+) uptake is differentially determined by the source of supplied ca(2+). Cell Calcium. 2010;47(5):433–40. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Sequence data that support the findings of this study have been deposited in NCBI Gene Expression Omnibus (GEO) with the primary accession code GSE264101 (https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE264101). The data can be allowed to review with the token “exudwmyiltadjez”.