Increased susceptibility to cortical spreading depression and epileptiform activity in a mouse model for FHM2

Abstract

Migraine is a highly prevalent, debilitating, episodic headache disorder affecting roughly 15% of the population. Familial hemiplegic migraine type 2 (FHM2) is a rare subtype of migraine caused by mutations in the ATP1A2 gene, encoding the α₂ isoform of the Na⁺/K⁺-ATPase, predominantly expressed in astrocytes. Differential comorbidities such as epilepsy and psychiatric disorders manifest in patients. Using a mouse model harboring the G301R disease-mutation in the α₂ isoform, we set to unravel whether α₂^+/G301R mice show an increased susceptibility for epilepsy and cortical spreading depression (CSD). We performed in vivo experiments involving cortical application of KCl in awake head-restrained male and female mice of different age groups (adult and aged). Interestingly, α₂^+/G301R mice indeed showed an increased susceptibility to both CSD and epileptiform activity, closely replicating symptoms in FHM2 patients harboring the G301R and other FHM2-causing mutations. Additionally, this epileptiform activity was superimposed on CSDs. The age-related alteration towards CSD indicates the influence of female sex hormones on migraine pathophysiology. Therefore, the FHM2, α₂^+/G301R mouse model can be utilized to broaden our understanding of generalized epilepsy and comorbidity hereof in migraine, and may be utilized toward future selection of possible treatment options for migraine.

Figure 1

Increased susceptibility to cortical spreading depression in α₂^+/G301R mice. (A) Schematic representation of the placement of ECoG electrodes and the craniotomy allowing for the application of KCl. (B) Picture of a mouse with implanted electrodes, corresponding connector and small 3D printed cylinders to allow head-restraining secured to its skull. The green area is non-toxic silicone to secure and protect the craniotomy. (C) Representative examples of 15 min ECoG traces of M1 left (pink), S1 left (black) and M1 right (green, contralateral) during occipital application of KCl in wild type mice (top), α₂^+/G301R mice (middle) and α₂^+/G301R post-menopausal (PM) females (bottom). (D) Quantification of CSD frequency based on 1 hour recordings in all groups of animals; male (light blue), female (pink), post-menopausal (PM) female (purple) and aged male (dark blue) α₂^+/G301R mice and their wild type littermates (N = 7–9). Each black dot represents an animal, error bars represent standard deviations. *p < 0.05, ***p < 0.001 (Mann Whitney U tests and Kruskal-Wallis test followed by Dunn’s Multiple Comparison tests with Bonferroni correction; see Table 1) (E) Representative example of a single CSD in a wild type (left) and α₂^+/G301R mouse (right) in the time difference in CSD onset between S1 (black) and M1 (pink) is depicted with dotted lines and arrows. (F) Quantification of propagating speed for all groups as in (D). Error bars represent standard deviations, ***p < 0.001 (Mann Whitney U tests and Kruskal-Wallis test and subsequent Dunn’s Multiple Comparison tests with Bonferroni correction; see Table 1).

Figure 2

Increased susceptibility to epileptiform activity in α₂^+/G301R mice. (A) Representative examples of the last 15 min of 1 hour ECoG recordings (M1) showing various degrees of epileptiform activity. Mice either did not develop epileptiform (green traces), developed bouts of epileptic spikes (black traces) or showed continuous epileptiform activity (pink traces). Traces were filtered using a 0.005 Hz high pass filter (top traces) or a 0.5 Hz high pass filter (middle traces). The bottom traces represent magnifications of the parts of the middle traces depicted by the dashed squares (0.5 Hz high pass filter). The red lines above the black and pink traces indicate the occurrence of epileptiform activity. (B) Quantification of the proportion of mice showing either no epileptiform activity (green), bouts of epileptiform activity (black) or continuous epileptiform activity (pink) by the end of the recording (1 hour) for all groups (male, female, aged male and post-menopausal (PM) female α₂^+/G301R mice and their wild type littermates; (N = 7–9)). (C) Quantification of onset of epileptiform activity for male, female, aged male and post-menopausal (PM) female α₂^+/G301R mice and wild types (pooled due to low number of animals developing epileptiform activity) (N = 9–10). Error bars represent standard deviations, *p < 0.05, ***p < 0.001 (Kruskal-Wallis test and subsequent Dunn’s Multiple Comparison tests with Bonferroni correction; see Table 2). (D) Representative examples of 1 hour ECoG recordings (M1; 0.005 Hz high pass filter) in α₂^+/G301R mice (bottom 2 traces) and wild type littermates (top 2 traces) showing the progression of epileptiform activity over time. Mice either did not develop epileptiform activity (green trace), developed bouts of epileptiform activity (black traces) or progressed to showing continuous epileptiform activity (pink trace). Arrows indicate onset of the epileptiform activity. Like in (A), The red lines above the black and pink traces indicate the occurrence of epileptiform activity.

Figure 3

Predictability of onset and severity of epileptiform activity by CSD frequency in α₂^+/G301R mice. (A) representative example of severe epileptiform activity (top) and the corresponding power spectrum (bottom). (B–D) Scatter plots showing linear relations between severity of epileptiform activity (as defined by the normalized power at the dominant frequency of the epileptiform activity) and CSD frequency (B), onset of epileptiform activity and CSD frequency (C) and severity and onset of epileptiform activity (D). ***p < 0.001 (linear regression, see Table 3).

Figure 4

Epileptiform activity is generalized and superimposed on CSDs. (A,B) Representative ECoG traces showing bouts of epileptiform activity (left) or late stage continuous epileptiform activity (right) that is generalized (apparent in both hemispheres) and superimposed on the CSD (note that the CSD does not spread to the contralateral hemisphere). All traces are from the same experiment but at 2 different time points; 30 min (left) or 55 min (right). Traces are shown using a 0.005 Hz (A) or 0.5 Hz (B) high pass filter and a 100 Hz low pass filter (both). The red lines above the traces indicate the occurrence of epileptiform activity. (C) Wavelets of the traces shown in (B) showing the change in frequency spectra over time. (D) Magnification of the part of the trace in (B) (right panel) depicted by the dotted square showing the timing of the epileptic spikes in all traces. (E) Average cross correlations between all combinations of the three ECoG channels (N = 33 mice). (F,G) Quantification of the maximum cross correlations (F) and timing of this maximum (G) for the three combinations of ECoG channels (M1 left vs S1 left, M1 left vs M1 right and S1 left vs M1 right). Error bars represent standard deviations, ***p < 0.001 (paired t-tests; see Table 4).