Adipocyte mitochondrial function is reduced in human obesity independent of fat cell size

Abstract

Context: It has been suggested that mitochondrial dysfunction in adipocytes contributes to obesity-related metabolic complications. However, obesity results in adipocyte hypertrophy, and large and small adipocytes from the same depot have different characteristics, raising the possibility that obesity-related mitochondrial defects are an inherent function of large adipocytes.

Objective: Our goal was to examine whether obesity, independent of fat cell size and fat depot, is associated with mitochondria dysfunction.

Design: We compared adipocyte mitochondrial function using a cross-sectional comparison study design.

Setting: The studies were performed at Mayo Clinic, an academic medical center.

Patients or other participants: Omental and/or abdominal subcutaneous adipose samples were collected from 20 age-matched obese and nonobese nondiabetic men and women undergoing either elective abdominal surgery or research needle biopsy.

Intervention: Interventions were not conducted as part of these studies.

Main outcome measures: We measured mitochondrial DNA abundance, oxygen consumption rates, and citrate synthase activity from populations of large and small adipocytes (separated with differential floatation).

Results: For both omental and subcutaneous adipocytes, at the cell and organelle level, oxygen consumption rates and citrate synthase activity were significantly reduced in cells from obese compared with nonobese volunteers, even when matched for cell size by comparing large adipocytes from nonobese and small adipocytes from obese. Adipocyte mitochondrial content was not significantly different between obese and nonobese volunteers. Mitochondrial function and content parameters were not different between small and large cells, omental, and subcutaneous adipocytes from the same person.

Conclusion: Adipocyte mitochondrial oxidative capacity is reduced in obese compared with nonobese adults and this difference is not due to cell size differences. Adipocyte mitochondrial dysfunction in obesity is therefore related to overall adiposity rather than adipocyte hypertrophy.

Figures

Figure 1.

Correlations between BMI and mtDNA. mtDNA content values are plotted vs BMI. Correlation coefficients were r = −0.33, P = .18 in omental large cells; r = −0.37, P = .128 in omental small cells; r = −0.51, P = .028 in subcutaneous large cells; r = −.26, P = .39 in subcutaneous small cell group.

Figure 2.

Correlation between BMI and oxidative capacity. OCR values were normalized by mitochondrial protein concentration (A) and cell numbers (B), respectively. OCR values were plotted against BMI. Correlation coefficients were r = −0.50, P = .03 in the omental large cell group; r = −0.47, P = .04 in the omental small group; r = −0.57, P = .007 in the subcutaneous large cell group; r = −0.53; P = .013 in the subcutaneous small cell group for OCR per milligram mitochondrial protein (A). Correlation coefficients were r = −0.83, P = .001 in the omental large cell group; r = −0.86, P = .0011 in the omental small cell group; r = −0.72, P = .001 in the subcutaneous large cell group; r = −0.51; P = .032 in the subcutaneous small cell group for OCR per cell (B). We were unable to calculate the cell number for all individuals, including two with BMI >50; thus, the x-axis for B is different than A.

Figure 3.

Correlations between BMI and CS-specific activity of adipocytes. CS activities were normalized by mitochondrial protein (A) and cell number (B), respectively. Correlation coefficients were r = −0.57 P = .01 in the omental large cell group; r = −0.19 P = .43 in the omental small cell group; r = −0.58 P = .005 in the subcutaneous large cell group; r = −0.52; P = .016 in the subcutaneous small cell group for CS activity per milligram mitochondrial protein (A). Correlation coefficients were r = −0.80, P = .003 in the omental large cell group; r = −0.69, P = .027 in the omental small cell group; r = −0.81, P = .0001in the subcutaneous large cell group; r = −0.35; P = .15 in the subcutaneous small cell group for CS activity per cell (B). We were unable to calculate the cell number for all individuals, including two with BMI >50; thus, the x-axis for B is different than A.