Increased susceptibility to cortical spreading depression in the mouse model of familial hemiplegic migraine type 2

Abstract

Familial hemiplegic migraine type 2 (FHM2) is an autosomal dominant form of migraine with aura that is caused by mutations of the α2-subunit of the Na,K-ATPase, an isoform almost exclusively expressed in astrocytes in the adult brain. We generated the first FHM2 knock-in mouse model carrying the human W887R mutation in the Atp1a2 orthologous gene. Homozygous Atp1a2(R887/R887) mutants died just after birth, while heterozygous Atp1a2(+/R887) mice showed no apparent clinical phenotype. The mutant α2 Na,K-ATPase protein was barely detectable in the brain of homozygous mutants and strongly reduced in the brain of heterozygous mutants, likely as a consequence of endoplasmic reticulum retention and subsequent proteasomal degradation, as we demonstrate in transfected cells. In vivo analysis of cortical spreading depression (CSD), the phenomenon underlying migraine aura, revealed a decreased induction threshold and an increased velocity of propagation in the heterozygous FHM2 mouse. Since several lines of evidence involve a specific role of the glial α2 Na,K pump in active reuptake of glutamate from the synaptic cleft, we hypothesize that CSD facilitation in the FHM2 mouse model is sustained by inefficient glutamate clearance by astrocytes and consequent increased cortical excitatory neurotransmission. The demonstration that FHM2 and FHM1 mutations share the ability to facilitate induction and propagation of CSD in mouse models further support the role of CSD as a key migraine trigger.

Figure 1. Knock-in construct strategy.

A. Genomic structure of the targeting vector and wt Atp1a2 allele. B. Predicted structure after homologous recombination (Atp1a2^R887-neo allele), and after Flp-mediated deletion of the neo-cassette (Atp1a2^R887 allele). FRT sites are indicated by green circles. Grey boxes indicate respective exons, and yellow star the R887 mutation in exon19. C. Southern blot of F1 Atp1a2^{+/R887 -neo} mutant mice. EcoRV and BamHI-digested genomic DNA from two genotypes for wild-type and Atp1a2^+/R887neo mutant strains probed 5′ and 3′ probe.

Figure 2. In vivo expression of mutant Atp1a2.

Total RNA and protein samples were isolated from brain of wild type (+/+), Atp1a2^+/R887 (+/R887) and homozygous Atp1a2^R887/R887 (R887/R887) mice at E19.5. A. Left panel. Semi quantitative Atp1a2 RT-PCR (254 bp fragment) on brain cDNA. The b-actin fragment (610 bp) is included as in-tube normalizer. Right panel, same Atp1a2 RT-PCR fragments digested by MspI. B. Protein blot of microsomal fraction probed with anti-α2 Na,K-ATPase antibody and anti-neogenin as loading control; the α2 Na,K-ATPase and neogenin bands appear at the expected size of 110 kDa and 52 kDa, respectively. C. Total brain lysates from adult wild type and Atp1a2^+/R887 mice probed with anti-α1, α2, and α3 Na,K-ATPase antibodies; anti- tubulin as loading control. Densitometric quantization shows a 50% reduction of the heterozygous mutant α2 level compared to wild type. Error bars represent ± SD; Student's t test p<0.05, n = 3; α1, α2, and α3 Na,K-ATPases migrate as single bands according to the expected size of 110 kDa. D. Region specific immunoblot from cortex, cerebellum and total brain of adult wild type and Atp1a2^+/R887 mice probed with anti-α2 and anti- glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as loading control (Figure 2D). Densitometry evaluation shows significant reduction in α2 level (p<0.05) of 50%, 35% and 40% in cortex, cerebellum and total brain, respectively. Error bars represent ± SD; Student's t test p<0.05, n = 3.

Figure 3. In vitro expression and localization of α2-ATPase.

A. Immunoblot of HeLa cells transfected with pA2-wt-Myc (wt) and pA2-R887-Myc (R887) probed with anti α2-ATPase antibody. B. Calnexin (red) and α2 ATPase (probed with anti-myc; green) immunofluorescence staining of wt or R887 transfected Hela cells. R887 α2 ATPase mostly colocalises with the endoplasmic reticulum. In the right graphs, the scatter plot of red (ch1) vs. green (ch2) colocalization intensities. (wt parameters: Rr 0.369, R 0.6, Ch1:Ch2 0.999; R887 parameters: Rr 0.558, R 0.755, Ch1:Ch2 0.997). C. HeLa cells expressing wt or R887 variant of α2 ATPase are treated with the reversible proteasome inhibitor MG132 (10 µM for 4, 6 and 8 hrs). Blots are probed with anti- α2 isoform antibody, anti-ubiquitin antibody and anti-tubulin as loading control.

Figure 4. In vivo CSD recording.

A. On the left, a sketch of the positioning of the stimulating (stim) and recording electrodes (1, 2) is outlined. Example of traces recorded in wild type (+/+) and Atp1a2 ^+/R887 mutants (+/R887) at the two electrodes with dotted line indicating the wave recorded at the first electrode and full line indicating the wave recorded at the second electrode in both wild type and mutated animals. B. CSD threshold, velocity and duration of wild type and mutant are reported in box plots reporting bottom-up the 5^th percentile, 25^th percentile, median, 75^th percentile and 95^th percentile. The single dot represents the mean value described for each group. Atp1a2 ^+/R887 heterozygous animals are more sensitive to CSD induction (+/R887, 13.00±1.7 µC, n = 20; +/+, 19.9±1.9 µC, n = 18; p<0.01; left graph). CSD velocity of propagation is increased in Atp1a2 ^+/R887 mutants (wild type, 3.85±0.35 mm/min n = 18); mutant, 5.41±0.41 mm/min, n = 20; p<0.01; middle graph). No significant difference was observed in CSD duration (wild type 41.1±3.5 sec, n = 18; Atp1a2 ^+/R887 40.1±3.19 sec, n = 20; p = 0.83; right graph).