Positron emission tomography (PET) studies suggest opioidergic system dysfunction in morbid obesity, while evidence for the role of the dopaminergic system is less consistent. Whether opioid dysfunction represents a state or trait in obesity remains unresolved, but could be assessed in obese subjects undergoing weight loss. Here we measured brain μ-opioid receptor (MOR) and dopamine D 2 receptor (D 2 R) availability in 16 morbidly obese women twice—before and 6 months after bariatric surgery—using PET with [11 C] carfentanil and [11 C] raclopride. Data were compared with those from 14 lean control subjects. Receptor-binding potentials (BP ND) were compared between the groups and between the pre-and postoperative scans among the obese subjects. Brain MOR availability was initially lower among obese subjects, but weight loss (mean= 26.1 kg, sd= 7.6 kg) reversed this and resulted in~ 23% higher MOR availability in the postoperative versus preoperative scan. Changes were observed in areas implicated in reward processing, including ventral striatum, insula, amygdala and thalamus (P's< 0.005). Weight loss did not influence D 2 R availability in any brain region. Taken together, the endogenous opioid system plays an important role in the pathophysiology of human obesity. Because bariatric surgery and concomitant weight loss recover downregulated MOR availability, lowered MOR availability is associated with an obese phenotype and may mediate excessive energy uptake. Our results highlight that understanding the opioidergic contribution to overeating is critical for developing new treatments for obesity.