Introduction

Chronic unpredictable mild stress (CUMS) is used to model depression-like symptoms in rodents, representing currently the most reliable model for the disease. However, 97% of CUMS studies focus on male animals¹, while depression has a twice as high prevalence in females. Therefore, our comprehensive study explored the impact of chronic stress on both behavior and brain function with multimodal neuroimaging techniques particularly in female rats.

Methods

The study involved n=30 Wistar rats, which were exposed to a six-week CUMS procedure to induce a depression-like phenotype through repeated microstressors, and a stress-free control group of n=8 rats. After the CUMS procedure, both groups underwent five behavioral tests to assess the severity of depression-like symptoms. MRI experiments were conducted on a 9.4T MR-scanner with a cryogenically cooled coil. Anesthesia combined 0.5% isoflurane and medetomidine (0.06 mg/kg/h). Respiratory and cardiac signals were monitored to correct functional data for physiological noise. The MRI acquisition protocol included a high-resolution 3D-scan (T2-weighted RARE, (0.15)²×0.288mm³ resolution), MR-spectroscopy (PRESS, 12µl-voxel in the prelimbic-cingulate cortex, TR/TE=4000/10ms, frequency-shift -2ppm) and resting-state fMRI (rsfMRI) (T2*-weighted EPI-FID, TR/TE 1500/17.5ms, (0.365mm)²×0.7mm, 350 acquisitions, 32 slices). An additional field map was acquired to correct for geometrical distortions. Analyses were performed using SPM12, in-house MATLAB scripts, FSL and LCModel.
VBM: Preprocessing steps included brain extraction, bias correction, coregistration, segmentation and nonlinear normalization using DARTEL. Normalized and modulated gray matter images were analyzed in a second-level GLM using total intracranial volume and age as covariates.
MRS: Individual frequency and phase correction was done over coils and averages as well as eddy current correction using an unsuppressed water-signal. Spectra were quantified by LCModel using a calibrated phantom basis dataset. Metabolite concentrations were referenced to the unsuppressed water signal, assuming mean water concentration of 46.106mol/L, and corrected for relaxation effects (T1/T2met=1500/300ms, T2water=45ms).
rsfMRI: Preprocessing included correction for field inhomogeneities and movement, regression of physiological noise², slice-timing correction and spatial normalization to a template using DARTEL-based flow fields from the 3D images, regression of movement parameters and CSF signal, DVARS scrubbing³ and band-pass filtering (0.01–0.1Hz). Regional time courses were extracted using a bilateral anatomical atlas. Preprocessed rsfMRI data was analyzed with NBS⁴, and seed-based approaches.

Results

CUMS rats displayed a depression-like phenotype
CUMS-exposed rats demonstrated a reduction of locomotor activity in their home cages (Fig. 2A) and an extended duration of immobility in the forced swim test (Fig. 2B), an indicator of hopelessness. Although the total sucrose consumption was diminished in CUMS-exposed rats (Fig. 2C), no differences in sucrose preference could be detected (Fig. 2D). CUMS-exposed rats exhibited a tendency to spend less time in the open arms of the elevated plus maze (Fig. 2E) and in the central area of the open field (Fig. 2F), implying elevated anxiety.
Chronic stress leads to gray matter volume loss in prefrontal, striatal and somatosensory regions
CUMS-exposed rats displayed a wide-spread reduction in gray-matter within prefronto-cortical (PFC) and striatal regions (Fig. 3A). CUMS-induced volume reduction predominantly affected the PFC, encompassing prelimbic (PL) and primary cingulate cortex (Cing1), the left striatum (Str-L), the left and right somatosensory cortex (S1), as well as the primary motor areas (M1) (Fig 3A-B).
MR spectroscopy shows decreased glutamate levels in the PFC after chronic stress exposure
CUMS-exposed rats showed significantly lower levels of the neurotransmitter Glu in the PFC (Fig. 4B). In addition, levels of Tau, Cho, and Cr (Fig. 4B) were also decreased after CUMS exposure. We could not find differences for GABA, Gln, and NAA (Fig. 4C).
Rats exposed to chronic stress show bidirectional alterations of functional connectivity (FC) in th PFC and limbic system
On a whole-brain level, CUMS rats demonstrated increased FC within the striato-hippocampal and the somatosensory-parietal network. Our seed-based FC assessment of the nucleus accumbens (Acb) confirmed higher FC to dorsal striatal and hippocampal regions in CUMS animals (Fig. 5B). In contrast, PL connections to the Acb were decreased (Fig. 5C). The FC between PL and the hippocampus was modulated by the level of glutamate in the PFC (Fig. 5D).

Summary/Discussion

In summary, our results suggest the PFC as a central hub region for stress-induced brain changes in female rats. With that, our data contributes to the current evidence for the PFC describing it as specifically sensible to chronic stress, and hypothesizing a chronic stress-induced enhancement of glutamatergic transmission in the PFC as a crucial factor to changes within the prefrontal GABAergic system. Reported evidence of increased glutamatergic transmission onto prefrontal parvalbumin neurons, particularly in female rodents, is clearly in line with the results of our study⁵.

Acknowledgements

We thank Felix Hörner for his skillfully conducted MR-measurements.