OBJECTIVE: Adipose function, not mass, underpins metabolic health. Lean and obese nonhuman primates (NHPs) naturally develop metabolic syndrome. Mitochondria-related measures in subcutaneous adipose tissue (SQ AT) and peripheral blood mononuclear cells may elucidate differences that transcend adiposity measures. METHODS: Obesity statuses ranged from very lean to severely obese (<
9%->
50%, n = 44), which were equivalent in healthy or unhealthy NHPs (metabolic syndrome score difference, p <
0.001). We evaluated SQ AT histology, electron microscopy, tissue proteins, and bioenergetics. RESULTS: Unhealthy adipocytes had mitochondria one-half the size of healthy adipocytes (p <
0.01), whereas adipocyte cell sizes were comparable. Consistent with small mitochondria, we saw deficiencies in mitochondrial fusion and quality-control proteins in SQ AT from unhealthy NHPs (all p <
0.05). Smaller mitochondria in unhealthy adipocytes were consistent with low SQ AT tissue respiration (p <
0.05). Mitochondrial size was specifically reduced with unhealthiness, as mitochondrial abundance, size, and related metrics were unrelated to adiposity. Isolated stromal vascular cells showed comparable respirometry profiles, substantiating specificity of adipocyte-related mitochondrial defects. Peripheral blood mononuclear cell bioenergetic indices were increased in unhealthy NHPs, indicative of immune cell activation, and correlated to SQ AT inflammatory cytokines. CONCLUSIONS: We conclude that targeting mitochondrial fusion processes would be a rational strategy to improve metabolic health, independent of total fat mass.