Stress-induced brain responses are associated with BMI in women

Overweight and obesity are associated with altered stress reactivity and increased inflammation. However, it is not known whether stress-induced changes in brain function scale with BMI and if such associations are driven by peripheral cytokines. Here, we investigate multimodal stress responses in a large transdiagnostic sample using predictive modeling based on spatio-temporal profiles of stress-induced changes in activation and functional connectivity. BMI is associated with increased brain responses as well as greater negative affect after stress and individual response profiles are associated with BMI in females (pperm < 0.001), but not males. Although stress-induced changes reflecting BMI are associated with baseline cortisol, there is no robust association with peripheral cytokines. To conclude, alterations in body weight and energy metabolism might scale acute brain responses to stress more strongly in females compared to males, echoing observational studies. Our findings highlight sex-dependent associations of stress with differences in endocrine markers, largely independent of peripheral inflammation.


Post-hoc Analysis of regions of interest Posterior Insula/Midbrain:
To better characterize the associations of BMI and stress-induced activation, we extracted average beta values from ROIs 9 containing the significant clusters.We then performed post-hoc regression analyses on the whole sample including a Sex*BMI interaction as well as separately for males and females to explore and describe potential sex effects.The association with BMI was only significant in females (substantia nigra (SN): b = -0.05,p = .004,posterior insula R: b = -0.06p < .001,posterior insula L: b = -0.03,p = .008)but not males (SN: b = -0.001,p = .95,posterior insula L: b = -0.03,p = .069,posterior insula R: b = 0.005, p = .75,Figure 3B).

Figures
Figure S1: Detailed description of the psychosocial stress task 5 .Before the stress phase, participants were informed about being recorded in the following trials.Additional aversive verbal feedback (verbal FB) about unsatisfactory performance was given in the 2nd and 4th rest period of the Stress condition.Saliva sampling was done in all participants (N=192) and in subsample of n=73 participants blood samples were taken to assess the cortisol response with higher temporal resolution.Figure S6: Bootstrapped associations of BMI with negative affect after the task.In the analysis, data was resampled in males so that the mean (upper right) and standard deviation (upper left) approach the female distribution.For each weighting scheme (xaxis), data was resampled 1,000 times to derive average estimates and 95% confidence intervals.Associations of BMI with stress-induced negative affect in males did only change marginally after adjusting the weights.

Figure S7:
Bootstrapped associations of negative affect after the task with BMI.In the analysis, data was resampled in males so that the mean (upper left) and standard deviation (upper right) gradually approach the female distribution.For each weighting scheme (x-axis), data was resampled 1,000 times to derive average estimates and 95% confidence intervals.Associations of the observed BMI with baseline cortisol became increasingly similar between males and females if the distributions became more similar.In contrast, this was not seen for the correlation with the predicted

Figure S2 :
Figure S2: Uncorrected, partial correlations of body mass index (BMI) with all different immune markers, for the complete sample and men and women separately in a larger sample (N=198).All correlations are corrected for age, diagnosis, and current psychiatric medication.The last column shows the number of values that have been imputed.White indicates no values had to be imputed.

Figure S3 :
Figure S3: Correlation (r(140)= .42,p < .001) of the baseline morning cortisol assessment (measured from a serum sample together with the cytokines) at separate day and the first salivary cortisol sample before the stress task in n=142 participants.This sample was taken at approximately 10am and after participants had already completed a fear extinction paradigm.

Figure S4 :
Figure S4: Weights of the elastic net model retrained in the female sample only (n=120).In comparison to the combined male + female model, activation from more timepoints of the hippocampus and posterior insula contribute to a successful prediction.Additionally, there are also contributions of the dorsal anterior cingulate cortex and ventromedial prefrontal cortex.vmPFC = ventromedial prefrontal cortex, Put = Putamen, PCC = posterior cingulate cortex, Hypoth = Hypothalamus, HippP = posterior hippocampus, HippM = medial hippocampus, HippA = anterior hippocampus, dACC = dorsal anterior cingulate cortex, Cau = caudate, Amy = amygdala.

Figure S5 :
Figure S5: Activation patterns correlated with BMI in males and females are not associated.Avergage stress-induced (Stress -PreStress) activation associated with from all n=268 regions of interest in the Shen Atlas is not correlated between males and females (r=.05, p=.38)

Table S4 :
Multiple regression models predicting stress responses (subjective, cardiovascular, endocrine) by sex, BMI, and their interaction.Confidence Interval, BMI = Body mass index, In the multiple regression sex was dummycoded with 0 = females and 1 = males.All models additionally included age, diagnosis status, cortisol response to the placement of an IV (responder = 1, non-responder 0) and medication status.