A small-molecule TrkB ligand improves dendritic spine phenotypes and atypical behaviors in female Rett syndrome mice

Abstract

Rett syndrome (RTT) is a neurodevelopmental disorder caused by mutations in methyl-CpG-binding protein-2 (MECP2), encoding a transcriptional regulator of many genes, including brain-derived neurotrophic factor (Bdnf). BDNF mRNA and protein levels are lower in RTT autopsy brains and in multiple brain regions of Mecp2-deficient mice, and experimentally increasing BDNF levels improve atypical phenotypes in Mecp2 mutant mice. Due to the low blood-brain barrier permeability of BDNF itself, we tested the effects of a brain penetrant, small molecule ligand of its TrkB receptors. Applied in vitro, LM22A-4 increased dendritic spine density in pyramidal neurons in cultured hippocampal slices from postnatal day (P) 7 male Mecp2 knockout (KO) mice as much as recombinant BDNF, and both effects were prevented by the TrkB receptor inhibitors K-252a and ANA-12. Consistent with its partial agonist activity, LM22A-4 did not affect spine density in CA1 pyramidal neurons in slice cultures from male wildtype (WT) mice, where typical BDNF levels outcompete its binding to TrkB. To identify neurons of known genotypes in the mosaic brain of female Mecp2 heterozygous (HET) mice, we treated 4-6-month-old female MeCP2-GFP WT and HET mice with peripheral injections of LM22A-4 for 4 weeks. Surprisingly, mutant neurons lacking MeCP2-GFP showed dendritic spine volumes comparable to that in WT controls, while MeCP2-GFP-expressing neurons showed larger spines, similar to the phenotype we described in symptomatic male Mecp2 KO mice where all neurons lack MeCP2. Consistent with this non-cell-autonomous mechanism, a 4-week systemic treatment with LM22A-4 had an effect only in MeCP2-GFP-expressing neurons in female Mecp2 HET mice, bringing dendritic spine volumes down to WT control levels, and without affecting spines of MeCP2-GFP-lacking neurons. At the behavioral level, we found that female Mecp2 HET mice engaged in aggressive behaviors significantly more than WT controls, which were reduced to WT levels by a 4-week systemic treatment with LM22A-4. Altogether, these data revealed differences in dendritic spine size and altered behaviors in Mecp2 HET mice, in addition to provide support to the potential usefulness of BDNF-related therapeutic approaches such as the partial TrkB agonist LM22A-4.