Introduction

Nucleus accumbens (NAc) is a key structure that contributes to the etiology of mood and anxiety disorders^1,2. Proton magnetic resonance spectroscopy (¹H-MRS) is a very promising tool to study neurochemistry in vivo and has a true potential to reveal metabolic alterations associated with psychiatric disease³. Current spectroscopy studies on human lack however the power to describe the alterations of molecular and cellular mechanisms underlying the visible metabolic changes. Animal models are thus of great importance to provide a link between visible biomarkers and their biological meaning. The aim of this study was to investigate metabolic differences associated with basal anxiety in the NAc of naive mice and to compare it with medial prefrontal cortex (mPFC), a region known to be very sensitive to stress.

Methods

All animal studies were performed with the approval of local animal care and use committee. We first segregated naïve C57bl/6J male mice (n=28) of around 8 weeks for their trait-anxiety using a standardized Light Dark Test (LDT)⁴. The time spent in the lit compartment of the LD box was used to assess anxiety-like behaviour and mice were separated into high susceptible (HS) and low susceptible (LS) based on the median⁵. After a resting period of a week, ¹H-MRS scan was conducted and metabolites concentrations were measured in both the nuclei accumbens and mPFC of anesthetized mice at 14.1T ⁶. Animals under isoflurane anesthesia (1-2%) were fixed in holder with a bite piece and two ear bars and anatomical MR images were obtained using T₂-weighted FSE images (15×0.4mm slices, TE_eff /TR=43/2000ms, averages=2). This set of images was used to localize the volumes of interests (VOIs), i.e. mPFC (1.7x1.4x1.2 mm³) and NAc (1.4x4.1x1 mm³). After adjusting field inhomogeneity to reach a water linewidth under 24Hz on the target VOI, localized ¹H MRS was applied using SPECIAL sequence (TE/TR=2.8/4000ms) ⁷. Full width at half maximum (FWHM) measured on the water signal was 15±1 Hz for mPFC and 18±2 Hz for NAc. Scan time was around 20 minutes for NAc and 25 minutes for mPFC, in order to reach satisfactory Signal to noise ratio (SNR) i.e. >10. The spectral data were frequency corrected and summed for quantification using LCMODEL referencing the endogenous water (80% in brain tissue) from the identical VOI. Cramér-Rao lower bounds (CRLB) > 50% were considered not reliable. Statistical comparison was done between these two groups (Student t test) as well as correlations between anxiety and metabolite levels.

Results

Based on T₂ images, bilateral NAc were clearly identified (Fig 1). SNR of in vivo spectra was 15±3 for PFC and 17±2 for NAc with time of acquisition of 25 and 20 minutes respectively. The resulting FWHM were 0.018 ppm (11Hz) for mPFC and 0.020 ppm (12Hz) for NAc, both of which were close to the difference between the 3.9 ppm resonance peaks of phosphocreatine (PCr) and creatine (Cr) (i.e. 0.017ppm) allowing to differentiate them. Together with the ability to visually see these two peaks, the obtained resolution indicate a robust and reliable PCr/Cr measurement. Results indicate strong regional differences in the metabolic profile, mainly in the levels of taurine and glutamate/GABA. Trait-anxiety showed a Gaussian distribution in the sample with a median of 133s regarding the time spent in lit compartment. Anxiety-related behaviour negatively correlated with the PCr/Cr ratio (p=0.01, R=0.51) and was significantly increased in the HS as compared to LS individuals (Unpaired t-test, two tailed t=2.46, p=0.02*). Finally, we also found that glutamate, glutamine (as well as the sum Glx) and taurine concentrations tend to be higher in HA than LA mice specifically in the NAc.

Discussion

The high levels of GABA in NAc can be explained by the abundance of GABAergic neurons (~95% of the neurons), which is a particularity of this structure. This study showed that natural variation in trait-anxiety of a group of naïve mice was associated with metabolic changes that are visible in ¹H-MRS in the NAc but not the mPFC. Increased phosphocreatine levels might reflect a higher mitochondrial activity as would be expected in a region with high energy demands. This observation was in line with previous findings suggesting a critical role of mitochondria in the NAc related to anxiety in rats⁷. In our knowledge, this is the first study showing a direct link between individual differences in basal anxiety and metabolic activity in the NAc of mice in vivo. Finally, this work has the potential of discovering important targets to treat behavioural dysfunctions related to neuroenergetics and will help further unravel and understand the metabolic changes associated to psychiatric conditions.

Acknowledgements

This work was supported by the CIBM of the UNIL, UNIGE, HUG; CHUV, EPFL and Leenaards and Louis-Jeantet Foundations.

References

¹Volkow, et al. Cell (2015); ²Shirayama, et al. Curr Neuropharmacol (2006); ³Maddock and Buonocore, Curr Topics Behav Neurosci (2012) ⁴Campos, et al. Rev Brea Psiquiatr (2013); ⁵Nasca et al, Mol Psy (2015); ⁶Duarte et al, Neuroimage (2012); ⁷Mlynarik, et al. Res Med (2006); ⁸Hollis et al. PNAS (2015);