Chromosome Rearrangements and Disease
Identification of genes for X-linked intellectual disability
We aim to identify novel genes for syndromic and non-syndromic X-linked intellectual disability (XLID) and related disorders by
- mapping the breakpoints of balanced chromosome rearrangements and
- deep sequencing of all human X-chromosome specific exons.
Understanding the role of PQBP1 in the pathology of intellectual disability
Since our first description of PQBP1 mutations in families with a syndromic form of X-linked intellectual disability (XLID) PQBP1 has been linked to several XLID disorders. The main clinical features include microcephaly, short stature and other midline defects with variation in severity of the phenotypes, both within and between families. The pathogenic nature of all but one mutations in the PQBP1 gene is a premature termination codon (PTC) and we have discovered that in the patients all result in the production of a truncated PQBP1 protein (Musante et al., Hum Mutation 2009). While it is well established that PQBP1 protein localizes in nuclear speckles and is engaged in transcription, we have found that it plays a role in alternative splicing of essential splicing factor RNAs. Nothing was known about the functional role of the cytosolic pool of PQBP1. Analysis of PQBP1 complexes revealed six novel interacting proteins, namely the RNA-binding proteins KSRP, SFPQ/PSF, DDX1 and Caprin-1, and two subunits of the intracellular transport-related dynactin complex, p150(Glued) and p27. In primary neurons, PQBP1 co-localized with its interaction partners in specific cytoplasmic granules, which stained positive for RNA, suggesting that PQBP1 plays a role in cytoplasmic mRNA metabolism, which is further supported by the partial co-localization and interaction of PQBP1 with the fragile X mental retardation protein (FMRP), one of the best-studied proteins found in RNA granules. Furthermore we have discovered that oxidative stress caused relocalization of PQBP1 to cytoplasmic dynamic RNA–protein complexes, called stress granules (SGs) and that the cellular distribution of PQBP1 plays a role in. SG assembly, i.e. ectopic cytoplasmic PQBP1 resulted in a significantly lower number of cells with SGs. Of note, this reduction was much weaker in cells overexpressing disease-associated PQBP1 mutants with a truncated polyglutamine-binding DR/ER repeat region. These results suggest that cytoplasmic PQBP1 is a potential negative regulator of SGs and that its poly(Q) binding DR/ER repeat region plays a role in SG formation or disassembly. Together these data demonstrate a role for PQBP1 in the modulation of SGs and suggest its involvement in the transport of neuronal RNA granules, which are of critical importance for the development and maintenance of neuronal networks, thus illuminating a route by which PQBP1 aberrations might influence cognitive function. (Kunde et al., Hum Mol Genet 2011).
Functional studies of CDKL5 implicated in an X-linked intellectual disability syndrome with early onset epileptic encelopathy
We have shown that mutations in the X-linked gene CDKL5/STK9 are a significant cause of a severe neurodevelopmental disorder (previously called atypical Rett syndrome (RTT) or variant of RTT), which affects predominantly girls (Kalscheuer et al, Am J Hum Genet 2003; Tao et al, Am J Hum Genet 2004; Córdova-Fletes et al, Clin Genet 2010; Rademacher et al, Neurogenetics 2011). The main clinical features include early onset epileptic encephalopathy with seizures starting within the fifth month of life, severe developmental delay, deceleration of head growth, impaired communication and, often, hand stereotypies.
CDKL5 encodes a serine-threonine kinase. Its N-terminal catalytic domain shares homology with members of the cyclin-dependent kinase (CDK) family and mitogen activated proteins (MAP) kinases. We have established that in primary neurons CDKL5 is localized at excitatory synapses and contributes to correct dendritic spine structure and synapse activity. This finding has prompted us to search for novel CDKL5 binding partners within dendritic spines. Having previously found that truncation of Netrin G1 (NTNG1) by a balanced chromosome translocation caused a clinical phenotype that largely overlapped with RTT (Borg et al, Eur J Hum Genet 2005), we hypothesized that Netrin G1 and CDKL5 could act in the same signaling pathway. To test our hypothesis, we have investigated if the specific Netrin G1 receptor NGL-1, which plays a crucial role in early synapse formation and maturation, interacts with CDKL5. Interestingly, our results indicated that NGL-1 and CDKL5 directly interact and in primary neurons both proteins co-localized at the dendritic spines. Next, we investigated if NGL-1 could be a phospho-substrate of CDKL5 and we indeed could show that CDKL5 phosphorylates NGL-1 on a serine residue (S631) close to the cytoplasmic C-terminal PDZ binding domain. This phosphorylation is necessary (i) for reinforcing the interaction between CDKL5 and NGL-1 and (ii) for promoting a stable association between NGL-1 and PSD95, a major protein of the postsynaptic density which plays an important role in synaptic plasticity. Accordingly, phospho-mutant NGL-1 was not any longer able to induce synaptic contacts while its phospho-mimetic form bound PSD95 more efficiently and partially rescued the CDKL5-specific spine defects. We also investigated the phosphorylation level of NGL-1 in humans using a fibroblast cell line derived from a girl who carried a balanced chromosome translocation that truncated CDKL5 and due to inactivation of the normal X-chromosome lacked functional CDKL5. Importantly, these cells when compared to normal control fibroblasts which express endogenous CDKL5 exhibited reduced levels of phosphorylated NGL1 with respect to the total amount of NGL1 protein and this level could be increased by overexpression of CDKL5. Together, our findings suggest a critical regulatory role for CDKL5 in the formation of excitatory synapses by coupling, through NGL-1 phosphorylation, the Netrin G1-NGL-1 adhesion with the recruitment of PSD95 and thereby provide important molecular insights into the pathophysiology of this disorder (Ricciardi et al, Nat Cell Biol 2012).