Msh2 deficiency leads to dysmyelination of the corpus callosum, impaired locomotion, and altered sensory function in mice

Abstract

A feature in patients with constitutional DNA-mismatch repair deficiency is agenesis of the corpus callosum, the cause of which has not been established. Here we report a previously unrecognized consequence of deficiency in MSH2, a protein known primarily for its function in correcting nucleotide mismatches or insertions and deletions in duplex DNA caused by errors in DNA replication or recombination. We documented that Msh2 deficiency causes dysmyelination of the axonal projections in the corpus callosum. Evoked action potentials in the myelinated corpus callosum projections of Msh2-null mice were smaller than wild-type mice, whereas unmyelinated axons showed no difference. Msh2-null mice were also impaired in locomotive activity and had an abnormal response to heat. These findings reveal a novel pathogenic consequence of MSH2 deficiency, providing a new mechanistic hint to previously recognized neurological disorders in patients with inherited DNA-mismatch repair deficiency.

Figure 1. Expression of QKI is significantly lower in the absence of MSH2 expression.

(a) MSH2 was knocked down in human leukemia cells (CEM cell line), and the mRNA expression of QKI isoforms was quantified by Real-time PCR and normalized to the GAPDH expression (b,c,d). The mRNA expression of Qki isoforms was also quantified by Real-time PCR and normalized to Gapdh expression in the corpus callosum of Msh2^−/− mice (e,f,g,h). Panels (a,e) represent western blots. Error bars represent SD. Data represent triplicate experiments in the CEM cell line and n = 3 mice in each genotype. The blots have been run under the same experimental conditions.

Figure 2. Myelin ultrastructure in the corpus callosum of the Msh2^−/−mice.

Transmission electron microscopic images of cross-sections at high magnification show disorganized myelin in the Msh2^−/− mice (b,c) compared to normal, compact myelin in the wild-type (WT) mice (a). Scale bars represent 500 nm. (d) Determination of myelin sheath thickness by G-ratio quantification from 50 axons in each mouse. Error bars represent SD. N = 3 mice for each genotype.

Figure 3. Immunohistochemical labeling.

(a) Paraffin section were stained using an antibody anti GFAP. The quantification of the results showed no difference between WT and Msh2^−/− mice. Data are represented as means ± SD (n = 4 for each genotype). (b) Paraffin section were stained using an antibody anti IBA1. The quantification of the results showed no difference between WT and Msh2^−/− mice. Data are represented as means ± SD (n = 4 for each genotype).

Figure 4. MBP expression and in vivo MRI.

(a) Quantification by densitometry of all the isoforms of MBP protein levels in the corpus callosum normalized to GAPDH signal shows low expression of myelin basic protein (MBP) in the Msh2^−/− mice compared to that in WT mice. Error bars represent SD (n = 3 for each genotype). (b) Corpus callosum volumes determined by MRI and normalized to brain volume were significantly smaller in the Msh2^−/−. Error bars represent SD (n = 4 for each genotype).

Figure 5. Electrophysiological deficits in Msh2^−/− mice.

(a) Representative corpus callosal compound action potential (CAP) traces from a WT mouse showing typical N1 and N2 components (gray line) compared to a smaller amplitude N1 component in a Msh2^−/− mouse (black line). (b) Quantification of the average stimulus-response (WT, n = 12 slices, 4 mice; Msh2^−/−, n = 13 slices, 4 mice) revealed a significant decrease in the amplitude of the N1 component (myelinated axons) in the Msh2^−/− mice (P < 0.001). (c) There was no difference in the amplitude of the N2 (unmyelinated axons) component. (d) N1 and N2 conduction velocities, which were represented by the respective latencies to peak at different distances (1–2 mm), revealed no difference between the WT and the Msh2^−/− mice. Electrophysiological data are represented as the mean ± SEM.

Figure 6. Impaired locomotion and sensation in Msh2^−/− mice.

(a) Mice were placed in an open-field chamber equipped with infrared sensors. Total distance traveled (in centimeters) was measured for each group of five mice. Msh2^−/− mice walked a significantly shorter distance than did the WT mice. Error bars represent SD (n = 5 for each genotype). (b) Mice were placed in an open-field chamber equipped with infrared sensors. Anxiety index is the ratio between the time in the center and the time in the periphery (n = 5 for each genotype).

Figure 7. Altered sensation in Msh2^−/− mice.

The hot-plate test was performed at 50 °C, 52.5 °C, and 55 °C. The time(s) elapsing to the first pain response (lifting or licking or jumping the hind paws (a) or the fore paws (b)) was scored for the WT (solid black) and the Msh2^−/− (grey) mice. 3 mice for each genotype were tested. Behavioral data are represented as means ± SD.