Duchenne muscular dystrophy (DMD) is an X-linked neuromuscular disease characterized by severe and progressive muscle weakness, due to mutations in the dystrophin gene, resulting in wheelchair dependency and death between the 2nd to 4th decade of life due to cardiorespiratory failure(1–3). Dystrophin, both full-length and smaller isoforms, is also expressed in brain tissue and its absence there is associated with behavioral and cognitive problems. The mdx mouse is the most commonly used DMD mouse model, which lacks full-length dystrophin (Dp427) and presents with memory and behavior problems(4,5). The consequences of dystrophinopathy on gross macroscopic alterations in the mouse brain are unclear. We investigated the effect of dystrophinopathy on brain structure in post-mortem mdx and wild-type control mice, as well as in a novel mouse model expressing low amounts of dystrophin, using high-resolution MRI after Gd-DOTA staining.

We used the following groups:

1. N=8 C57Bl/10ScSn-mdx/J (mdx) mice (0% dystrophin)

2. N=10 wild type C57BL/10ScSnJ (Bl10-WT) mice (100% dystrophin)

3. N=7 mdx-Xist∆hs mice: a mouse model with 1.08-12.18% dystrophin expressed in brain (measured in n=8 mdx-Xist∆hs mice of another study). Xist∆hs mice have an C57BL/6Jico background.

4. N=9 wild type C57BL10/ScSnJ-Xist∆hs (Bl10/Bl6-WT)

At the age of three months, mice were anesthetized with isoflurane and perfused for one minute with 1x phosphate buffer saline (PBS) and for four minutes with 4% paraformaldehyde (PFA). The skull was stored in 4% PFA overnight and transferred to 4% PFA+1:40 v/v Dotarem (Manufacturer: Guerbet, USA) at 4°C for three weeks. Then they were transferred to a solution of 1x PBS, 1:40 v/v Dotarem and 0.01% sodium azide and a MRI scan was acquired after 24-48 hours.

Imaging was performed on a 7 Tesla Bruker Pharmascan with ParaVision 5.1, using a 23 mm volume coil. Scan details: 3D gradient echo (FLASH) with TE=5.3 ms; TR=15ms; FA=30°; FOV=[18x13x13]mm; zero-filling=1.34; matrix size=[256x186 x186]; isotropic resolution=0.070 mm; averages=12.

Brain segmentation was performed as described in (6) using a template image with 22 manually annotated structures (based on the Waxholm, Franklin and Paxinos and the Allen Brain Atlases using v5, FEI Software, Oregon, USA) (7–9), see figure 1.

Data was analysed using independent samples T-test in SPSS (v21). Analysis was done on absolute volumes data and in this exploratory study no correction was done for multiple comparisons.

Effect of dystrophin

The whole brain volume of the mdx mice was not significantly different from Bl10-WT mice, although there was a trend for a larger brain volume, consistent with two previous studies showing a non-significant increase in whole brain volume in mdx mice (10,11). However, several brain structures showed an increased volume in mice compared to WT-Bl10, as shown in table 1. The mdx-Xist∆hs mice were compared to Bl10/Bl6-WT mice, where no differences were found between whole brain volume. However 14 out of 22 brain structures were larger in mdx-Xist∆hs mice compared to Bl10/Bl6-WT mice, as shown in table 1.

Effect of body mass and genetic background

Body mass of mdx mice was significantly larger compared to Bl10-WT, mdx-Xist∆hs and Bl10/Bl6-WT mice, due to disease-related muscle hypertrophy.

Unexpected large background effects, with Bl10/Bl6-WT background having 8.7% larger brain volume compared to mdx and Bl10-WT mice, prevented direct comparison between the strains. Therefore, mdx mice were compared to Bl10-WT mice and mdx-Xist∆hs compared to Bl10/Bl6-WT mice.

From literature it is known that dystrophin is predominantly expressed in cortex, hippocampus and cerebellum(12) as shown in figure 1. Human data suggest that DMD patients have reduced brain volume(13), though this is most prominent in patients also lacking isoform Dp140, whereas mdx mice only lack Dp427. Similar to the human data, structural changes are wide-spread, involving many structures. Unexpectedly, we found small increases in volume. Future studies will therefore look into mice lacking Dp140.

Within our study we identified two important caveats for structural analysis in general. Firstly, a number of mouse studies used brain data normalized for body weight, but for this mouse model, the muscle hypertrophy results in increased body weight, leading to the erroneous conclusion that mdx brain volume is reduced. Secondly, even though our two mouse models have closely related genetic backgrounds (Bl10 vs Bl10/Bl6J, we observed a large background effect on the mouse brain volume, indicating that the background needs to be carefully controlled. More importantly, it is likely that the background will also influence the structural and behavioral phenotype.

This study shows that mdx mice have a have non-significant increase in whole brain volume. This is in agreement with patient data, where only patients lacking Dp140 show reduced brain volume.