New insights into aging-associated characteristics of female subcutaneous adipose tissue through integrative analysis of multi-omics data

Abstract

Aging could be critical in limiting the application of subcutaneous adipose tissue (SAT) in tissue repair and reconstruction. However, no systematic study on the characteristics of SAT aging has been conducted. In this study, a scanning electronic microscope was used to detect the structural and compositional changes of SAT collected from nine females in three age groups. Multi-omics data of SAT from 37 females were obtained from Gene Expression Omnibus database, and 1860 genes, 56 miRNAs, and 332 methylated genes were identified as being differentially expressed during aging among non-obese females. Using Weighted Correlation Network Analysis (WGCNA), 1754 DEGs were defined as aging-associated genes for non-obese females, distributed among ten co-expression modules. Through Gene Ontology enrichment analysis and Gene Set enrichment analysis on those aging-associated DEGs, SAT aging was observed to be characterized by variations in immune and inflammatory states, mitochondria, lipid and carbohydrate metabolism, and regulation of vascular development. SUPV3L1, OGT, and ARPC1B were identified as conserved and core SAT-aging-related genes, as verified by RT-qPCR among 18 samples in different age groups. Multi-omics regulatory networks of core aging-associated biological processes of SAT were also constructed. Based on WGCNA, we performed differential co-expression analysis to unveil the differences in aging-related co-expression patterns between obese and non-obese females and determined that obesity could be an important accelerating factor in aging processes. Our work provides a landscape of SAT aging, which could be helpful for further research in fields such as repair and reconstruction as well as aging.

Keywords: Subcutaneous adipose tissue; aging; multi-omics profiling; obesity; tissue repair and reconstruction.

Figure 1.

Morphology and quantification of adipocytes and collagen of SAT during aging. (a) SEM images of SAT in different age groups. Scale bar = 100 μm or 50 μm. (b) Numbers of the mature adipocytes, percentage of small adipocytes and diameter of mature adipocytes in different age groups in the same-sized fields.

Figure 2.

Differentially expressed mRNAs, miRNAs, and methylated genes in aging processes for obese and non-obese females.

Figure 3.

The co-expression patterns of genes and visualization of core GOBP pathways of SAT in aging process. (a) Visualization of tree diagram and co-expression patterns of genes based on the adjacency-based dissimilarity of the hierarchical clustering genes. (b) 10 aging-related gene co-expression modules with different enrichment status in each group. Enrichment Score indicated the ratio of DEGs in the specific module to all genes involved in the WGCNA network. (c) GOBP enrichment analysis for selected DEGs. (d) Gene set enrichment analysis of genes based on age groups. Normalized enrichment score (NES) was presented in chart. P Value < 0.05 was considered significant.

Figure 4.

The regulatory networks between DEGs and DEMs and RT-qPCR. (a) The regulatory networks of DEGs-DEMs involved in 7 core functional subnetwork in SAT agingbased on WGCNA. (b,c) Heatmaps of DEGs and DEMs involved in the 7 regulatory networks. DEGs and DEMs were distributed in different co-expression modules with different colors. (d) Quantitative analysis of the RT-qPCR assays for SAT.

Figure 5.

DiffCoEx analysis based on aging related genes of SAT between obese and non-obese female. (a) The comparative correlation heatmap showed the differentially co-expressed modules of SAT between obese and non-obese female based on aging related genes. The upper diagonal of the main matrix showed a correlation between pairs of genes among the non-obese female. The lower diagonal showed a correlation between the same gene pairs among the obese female. Color bars represented the differentially co-expressed modules. (b) 7 DiffCoEx modules with different enrichment status of SAT in each group. Enrichment Score indicated the ratio of DEGs in a specific module to all genes involved in the DiffCoEx network. (c) Differential co-expression network of SAT between obese and non-obese female. Colors represented the DiffCoEx modules. Red lines showed stronger correlation in obese female but blue lines in non-obese female. The thickness represented the correlation strength. (d) Bubble plots foraging related GOBP enrichment pathways. (e) The regulatory networks of DEGs-DEMs involved in the aging associated biological processes based on DiffCoEx analysis of aging process between obese and non-obese female. ‘*’ represented the potentially core molecules in the regulatory networks of obese female according to DiffCoEx analysis. P Value < 0.05 was considered significant.