Inhibition of p21-activated kinase rescues disrupted phenotype in a mouse model of CDKL5 deficiency disorder

Introduction: CDKL5 Deficiency Disorder (CDD) is a rare neurodevelopmental condition characterized by severe behavioral and physiological symptoms. About 90 % of patients diagnosed with CDD are females. CDD is caused by mutations in the X-linked cyclin-dependent kinase-like 5 (CDKL5) gene. This gene encodes for the homonym serine/threonine kinase, a protein involved in several neurodevelopmental processes, including regulation of spine morphogenesis and synapse formation. To date, no cure is available for this disorder. Mice carrying mutations in the CDKL5 gene have been generated, and represent good models for this disorder as they recapitulate both the behavioral and morphological phenotype disrupted in CDD patients [1], [2]. We recently demonstrated that a CDD mouse model present brain overactivation of group I p21-activated kinases (PAKs), a family of serine/threonine kinases that are critically involved in the regulation of ultrastructural neuronal morphology and activity-dependent actin dynamics [3]. Prolonged activation of group I PAKs affects cognitive function and social behavior in mice and has been associated with several neurologic disorders [4]. Furthermore, recent studies demonstrate that pharmacological inhibition of group I PAKs rescues neurobehavioral defects in preclinical models of Fragile X and schizophrenia [5]. Aim: The present study evaluated whether overactivation of group I PAKs might contribute to the behavioral phenotype observed in CDD, and whether pharmacological inhibition of these kinases might rescue CDD-related behavioral alterations in female mice, the gender most frequently affected by CDD. To this aim, we used Frax486 (supplied by DBA Italia), a brain penetrant PAK inhibitor with an excellent selectivity over group I PAKs. Methods: Cdkl5 heterozygous (Het) female mice were injected daily with FRAX486 (20 mg/kg, s.c. injections, for 5 days) or saline at an advanced stage of the disease (1 year of age). Wild-type (wt) littermates treated with saline were used as controls. A battery of behavioral tests was carried out to evaluate genotype differences and treatment efficacy. Data were analyzed with one way ANOVA. Post-hoc comparisons were performed using the Tukey's HSD. When relevant, repeated measures ANOVA was applied. Results: A 5-day long treatment with FRAX486 at an advanced stage of the disease normalized the general health status (p<0.001 after post hoc comparison on the experimental group: F(1,36)=6.414; p=0.0001) and the hyperactive profile of Cdkl5-Het mice (p<0.01 after post hoc comparison on the experimental group: F(1,35)= 6.414; p= 0.004). It also increased the defective preference for the novel social stimulus of CDD mice in the Three-chamber social test, thus restoring wt-like levels of social novelty (p<0.05 after post hoc comparison on the experimental group: F(1,34)= 3.796; p= 0.032). Conclusion: Present results provide evidence that a 5-day long treatment with FRAX486 rescues relevant behavioral alterations in a CDD female mouse model, suggesting that group I PAK inhibitors may represent an innovative therapeutic strategy for this disorder.