Molecular and genetic approaches to identify the role of CDKL5 in the nervous system.

Rett syndrome (RS) is a progressive neurodevelopmental disorder with a large impact on society due to its high incidence in the female population. After a period of apparently normal development, RS girls lose acquired skills and start manifesting stereotypic hand movements, epilepsy, mental retardation, respiratory disturbances, motor deterioration and cardiac troubles. In 1999, mutations in the X-linked gene coding for methyl-CpG binding protein 2 (MeCP2) were identified as the molecular cause of RS. MeCP2 is a transcriptional repressor that, by binding methylated promoters, down-regulates transcription of its target genes. According to the fact the RS is an exclusively neurological disorder, MeCP2 regulates the transcription of specific neuronal genes such as Bdnf. In particular, Bdnf expression is modulated by an activity dependent phosphorylation of MeCP2. In addition to classical RS, which is strongly associated with MECP2 mutations, RS variants have been described presenting some general features of the classic form but with different onset and severity. Less than half of the patients with these variants display MECP2 mutations, indicating that other genetic loci are involved opening the way to unravel the relationships between MeCP2 and other genes causing RS variants. In recent years, mutations in the X-linked cyclin-dependent kinase like 5 (CDKL5) were identified in the RS Hanefeld variant, which is characterized by the absence of a normal period of development and the early onset of intractable seizures. CDKL5 is a serine/threonine kinase containing a catalytic domain in the N-terminal region and a large COOH-terminal tail. CDKL5 is a rather uncharacterized protein, however, its involvement in RS pathogenesis can be explained by the fact that the kinase interacts with MeCP2 in vivo mediating its phosphorylation in vitro. The aim of this PhD thesis was to elucidate the role of CDKL5 in the nervous system both in physiological and in pathological conditions. The expression profile of the kinase in embryonic and postnatal brains from normal mice was compared to that of MeCP2 in order to better understand where and when the two proteins might enter in contact. Using an immunohistochemistry approach, we found that, despite an expression profile generally overlapping that of MeCP2, CDKL5 peaks later in development and, in some cases, its levels appear to be modulated in an opposite manner to that of MeCP2. A similar distribution was observed in adult human brain tissues reinforcing the relevance of the kinase in the nervous system. By a series of immunofluoresence analyses on cultured primary hippocampal neurons we found that CDKL5 is a neuronal protein detectable in glutamatergic as well in GABAergic neurons but not in astrocytes. In mature neurons, the kinase is distributed both in the nucleus, with a particular dotted organization, and in the cytoplasm where it localizes in the cell soma and in the dendritic branches. Regarding the modulation of the subcellular distribution of CDKL5 in different stages of in vitro neuronal differentiation and in cultured non-neuronal cells, we demonstrated that an active nuclear export mechanism is involved in localizing the protein to the cytoplasm. Taken together, our results indicate the importance of CDKL5 for postnatal brain functioning. Eventually, we identified CDKL5 also in non-neuronal murine tissues, including kidney and liver. However, the neurological phenotype observed in patients with CDKL5 mutations leads to hypothesize a crucial role of the kinase in the nervous system where it is expressed at higher levels. Since animal model represents an optimal tool to better understand the biological role of a protein and the mechanisms underlying a pathogenesis, we decided to generate a mouse possibly manifesting some pathological features observed in patients with CDKL5 defects. We therefore planned to conditionally inactivate the endogenous Cdkl5 gene in mice by a Cre-recombinase strategy. An ES-cell clone containing the correctly targeted Cdkl5 locus was obtained and injected into blastocysts of pseudopregnant females and the resultant chimeric animals were breeded for germline transmission. Unfortunately, no mice have so far been obtained carrying the targeted Cdkl5 locus in germline. As an important by-product of this project, however, an ES-cell line devoid of CDKL5 that can be differentiated into the neuronal lineage has been generated. This provides an important tool for studying neuronal differentiation and maturation in the absence of CDKL5 thereby helping answering some questions regarding the onset of RS and related variants in patients mutated in CDKL5.