Interaction between serotonin transporter and dopamine D2/D3 receptor radioligand measures is associated with harm avoidant symptoms in anorexia and bulimia nervosa

Abstract

Individuals with anorexia nervosa (AN) and bulimia nervosa (BN) have alterations of measures of serotonin (5-HT) and dopamine (DA) function, which persist after long-term recovery and are associated with elevated harm avoidance (HA), a measure of anxiety and behavioral inhibition. Based on theories that 5-HT is an aversive motivational system that may oppose a DA-related appetitive system, we explored interactions of positron emission tomography (PET) radioligand measures that reflect portions of these systems. Twenty-seven individuals recovered (REC) from eating disorders (EDs) (7 AN-BN, 11 AN, 9 BN) and nine control women (CW) were analyzed for correlations between [(11)C]McN5652 and [(11)C]raclopride binding. There was a significant positive correlation between [(11)C]McN5652 binding potential (BP(non displaceable(ND))) and [(11)C]Raclopride BP(ND) for the dorsal caudate, antero-ventral striatum (AVS), middle caudate, and ventral and dorsal putamen. No significant correlations were found in CW. [(11)C]Raclopride BP(ND), but not [(11)C]McN5652 BP(ND), was significantly related to HA in REC EDs. A linear regression analysis showed that the interaction between [(11)C]McN5652 BP(ND) and [(11)C]raclopride BP(ND) in the dorsal putamen significantly predicted HA. This is the first study using PET and the radioligands [(11)C]McN5652 and [(11)C]raclopride to show a direct relationship between 5-HT transporter and striatal DA D2/D3 receptor binding in humans, supporting the possibility that 5-HT and DA interactions contribute to HA behaviors in EDs.

Figure 1

Comparison of [¹¹C]raclopride BP_ND in the antero-ventral striatum between groups; CW, control women; REC, recovered; AN, anorexia nervosa, restricting type AN; AN-BN, bulimic type AN; BN, bulimia nervosa

Figure 2

Correlation of [¹¹C]McN5652 BP_ND and [¹¹C]raclopride BP_ND in the dorsal caudate (Panel A) and antero-ventral striatum (AVS) (Panel B) in the REC ED sample

Figure 3

REC ED showed positive correlations between [¹¹C]raclopride BP_ND and Harm Avoidance in the dorsal caudate (r(27) = .48; p = .01) (Panel A) and dorsal putamen (r(27) = .42; p = .03) (Panel B)

Figure 4

For each region, scatterplots of [¹¹C]McN5652 BP_ND (y-axis) versus Harm Avoidance (x-axis), with each individual coded with a symbol based on a median split of [¹¹C]raclopride BP_ND, depict the relationship between these three continuous variables. Panel A. In the dorsal putamen, women with [¹¹C]raclopride BP_ND values below the median show a negative relationship between [¹¹C]McN5652 BP_ND and Harm Avoidance, while those with [¹¹C]raclopride BP_ND values above the median show a positive relationship between [¹¹C]McN5652 BP_ND and Harm Avoidance. Panel B. In the dorsal caudate, women with [¹¹C]raclopride BP_ND values below the median show a slight positive relationship between [¹¹C]McN5652 BP_ND and Harm Avoidance, while those with [¹¹C]raclopride BP_ND values above the median show a negative relationship between [¹¹C]McN5652 BP_ND and Harm Avoidance.

Figure 4

For each region, scatterplots of [¹¹C]McN5652 BP_ND (y-axis) versus Harm Avoidance (x-axis), with each individual coded with a symbol based on a median split of [¹¹C]raclopride BP_ND, depict the relationship between these three continuous variables. Panel A. In the dorsal putamen, women with [¹¹C]raclopride BP_ND values below the median show a negative relationship between [¹¹C]McN5652 BP_ND and Harm Avoidance, while those with [¹¹C]raclopride BP_ND values above the median show a positive relationship between [¹¹C]McN5652 BP_ND and Harm Avoidance. Panel B. In the dorsal caudate, women with [¹¹C]raclopride BP_ND values below the median show a slight positive relationship between [¹¹C]McN5652 BP_ND and Harm Avoidance, while those with [¹¹C]raclopride BP_ND values above the median show a negative relationship between [¹¹C]McN5652 BP_ND and Harm Avoidance.