Distinct developmental signatures of human abdominal and gluteal subcutaneous adipose tissue depots

Abstract

Context: Fat distribution differs in men and women, but in both sexes, a predominantly gluteal-femoral compared with abdominal (central) fat distribution is associated with lower metabolic risk. Differences in cellular characteristics and metabolic functions of these depots have been described, but the molecular mechanisms involved are not understood.

Objective: Our objective was to identify depot- and sex-dependent differences in gene expression in human abdominal and gluteal sc adipose tissues.

Design and methods: Abdominal and gluteal adipose tissue aspirates were obtained from 14 premenopausal women [age 27.5 ± 7.0 yr, body mass index (BMI) 27.3 ± 6.2 kg/m(2), and waist-to-hip ratio 0.82 ± 0.04] and 21 men (age 29.7±7.4 yr, BMI 27.2 ± 4.5 kg/m(2), and waist-to-hip ratio 0.91 ± 0.07) and transcriptomes were analyzed using Illumina microarrays. Expression of selected genes was determined in isolated adipocytes and stromal vascular fractions from each depot, and in in vitro cultures before and after adipogenic differentiation.

Results: A total of 284 genes were differentially expressed between the abdominal and gluteal depot, either specifically in males (n = 66) or females (n = 159) or in both sexes (n = 59). Most notably, gene ontology and pathway analysis identified homeobox genes (HOXA2, HOXA3, HOXA4, HOXA5, HOXA9, HOXB7, HOXB8, HOXC8, and IRX2) that were down-regulated in the gluteal depot in both sexes (P = 2 × 10(-10)). Conversely, HOXA10 was up-regulated in gluteal tissue and HOXC13 was detected exclusively in this depot. These differences were independent of BMI, were present in both adipocytes and stromal vascular fractions of adipose tissue, and were retained throughout in vitro differentiation.

Conclusions: We conclude that developmentally programmed differences may contribute to the distinct phenotypic characteristics of peripheral fat.

Figures

Fig. 1.

Depot differences in gene expression specific in males or in females or common in both sexes. Venn diagram depicts number of genes that were differentially expressed between gluteal and abdominal sc adipose tissue, either in a sex-specific way (only in males or only in females) or in both sexes.

Fig. 2.

Expression of HOX genes and HOX gene cofactors in abdominal (Abd) and gluteal (Glu) sc adipose tissue. A and B, Predominantly abdominal HOX genes: A, expression of HOXA3, HOXA5, HOXB8, HOXC8, and IRX2 was higher in the abdominal compared with gluteal adipose tissue in both sexes; B, expression of HOXA2, HOXA4, HOXA9, and HOXB7 was higher in the abdominal compared with gluteal adipose tissue in males only, and expression of HOXD4 was not statistically different between depots. C, Predominantly gluteal HOX genes: expression of HOXA10 and HOXC13 was higher in the gluteal compared with abdominal adipose tissue in both sexes. D, Expression of the HOX genes cofactors PBX1, PBX3, and MEIS1 in the abdominal and gluteal adipose tissue of men and women. Data are presented as mean ± sem of females (n = 14) and males (n = 21). *, P < 0.05; **, P < 0.01; ***, P < 0.001 compared with Abd of same sex; #, P < 0.05; ##, P < 0.01; ###, P < 0.001 compared with same depot of females.

Fig. 3.

Expression of selected HOX genes in adipose tissue fractions. Expression of selected HOX genes is depot specific in both isolated adipocytes and stromal-vascular fraction (SVF) (n = 11 for paired abdominal (A) and gluteal (G) tissue samples, n = 3 for isolated adipocytes and stromal-vascular fractions). Data are presented as mean ± sem. *, P < 0.05; **, P < 0.01 compared with abdominal (paired t test).

Fig. 4.

Expression of selected HOX genes during in vitro differentiation of human primary preadipocytes. Graphs depict expression of predominantly abdominal (HOXA5 and HOXB8) and predominantly gluteal (HOXA10 and HOXC13) HOX genes during in vitro differentiation of paired abdominal (ABD) and gluteal (GLUT) human primary preadipocytes (n = 4). Data are presented as mean ± sem and analyzed with repeated-measures ANOVA. AU, Arbitrary units; NS, not significant.