Cerebral μ-opioid and CB1 receptor systems have distinct roles in human feeding behavior

Eating behavior varies greatly between individuals, but the neurobiological basis of these trait-like differences in feeding remains poorly understood. Central μ-opioid receptors (MOR) and cannabinoid CB1 receptors (CB1R) regulate energy balance via multiple neural pathways, promoting food intake and reward. Because obesity and eating disorders have been associated with alterations in the brain’s opioid and endocannabinoid signaling, the variation in MOR and CB1R system function could potentially underlie distinct eating behavior phenotypes. In this retrospective positron emission tomography (PET) study, we analyzed [11C]carfentanil PET scans of MORs from 92 healthy subjects (70 males and 22 females), and [18F]FMPEP-d2 scans of CB1Rs from 35 subjects (all males, all also included in the [11C]carfentanil sample). Eating styles were measured with the Dutch Eating Behavior Questionnaire (DEBQ). We found that lower cerebral MOR availability was associated with increased external eating—individuals with low MORs reported being more likely to eat in response to environment’s palatable food cues. CB1R availability was associated with multiple eating behavior traits. We conclude that although MORs and CB1Rs overlap anatomically in brain regions regulating food reward, they have distinct roles in mediating individual feeding patterns. Central MOR system might provide a pharmacological target for reducing individual’s excessive cue-reactive eating behavior.


Supplementary Material
i.
PET scanners and smoking status (Supplementary Table 1) ii.
References for the original studies (Supplementary Table 2) iii. vii.
Descriptive correlations of the sample (Supplementary Figure 1) viii. Visualization of regional associations in three representative regions of interest (Supplementary Figure 2) ix.
Regional correlations between MOR and CB 1 R availabilities (Supplementary Figure 3) x.
External eating and µ-opioid receptor availability in males (Supplementary Figure 4)
CB 1 R availability was quantified as [ 18 F]FMPEP-d 2 volume of distribution (V T ) using graphical analysis 1 (Logan). The frames starting at 36 minutes and later since injection were used in the model fitting, since Logan plots became linear after 36 minutes 1 . Plasma activities were corrected for plasma metabolites as described previously 2 . We were unable to obtain metabolite data from two subjects due to laboratory measurement issues, so we used the mean metabolite fraction of the group for these two subjects.

[ 11 C]carfentanil scans
A cannula was inserted before the scan in an antecubital vein for [ 11 C]carfentanil injection. 223-279 MBq of [ 11 C]carfentanil was administered intravenously and the brain's radioactivity was followed 51 minutes (13 frames). The modeling of the [ 11 C]carfentanil data and the acquisition of binding potentials (BP ND ) has been previously described 3 .

Supplementary Text 2. Summary of the complementary [ 11 C]carfentanil analyses.
Smoking status did not influence [ 11 C]carfentanil results. When adjusting simultaneously for smoking, sex and BMI (in addition to age and scanner), the association with [ 11 C]carfentanil binding potential (BP ND ) and External eating score remained essentially the same (significant at cluster forming threshold p < 0.02, FWE corrected).

The male subsample
In the subsample of 70 males, the associations between External eating score and µ-opioid receptor availability were mostly similar than with the full sample, associations in caudatus being slightly less prominent (significant at cluster forming threshold p < 0.05, FWE corrected).

The female subsample
In the subsample of 22 females, there were no statistically significant associations between External eating score and central µ-opioid receptor availability. This is potentially due to the relatively low number of female subjects. Since Restrained eating score was higher in females than in males and females were also older than males (Table 1), we examined the correlations of age and restrained eating also in both female and male subgroups. Restrained eating correlated positively with age in both groups, but more strongly in the female sample (females r = 0.33, p = 0.01; males r = 0.25, p = 0.03).  Shown are clusters where p < 0.05, FWE corrected at cluster level.