Final Report Summary - PIMT AND SIGNALING (Interaction between the protein repair enzyme L-isoaspartyl methyltransferase and insulin/IGF-1 signaling in mice and worms)
The original overall objective of this project was to progress in our understanding of the link between PIMT and insulin-like signalling. The specific aims were the following:
1) Consolidate the interaction between PIMT and insulin-like signalling in worms
2) Analyse the effect of PIMT deficiency on the phosphoproteome of mouse brains
3) Elucidate the mechanism by which PIMT deficiency leads to the observed changes in protein phosphorylation.
The first aim was successfully addressed during the first reporting period using C. elegans strains deficient in or overexpressing the gene encoding the isoaspartyl methyltransferase (pcm-1) in these worms (Khare S, Linster CL, Clarke SG. The interplay between protein L-isoaspartyl methyltransferase activity and insulin-like signaling to extend lifespan in Caenorhabditis elegans. PLoS One. 2011; 6(6):e20850). In this work, we have found that the lifespan extension observed in PCM-1 overexpressing worms is reduced when DAF-16 is knocked down by RNA interference. As DAF-16 is the downstream transcriptional effector of the insulin-like signalling pathway in worms, this and other observations made during the first reporting period strongly support that in worms, as in mice, the isoaspartyl methyltransferase interacts with the insulin signalling pathway and that therefore C. elegans can be used as a model organism to further explore the mechanism underlying this interaction. Our studies in C. elegans also indicate that the signalling role of the isoaspartyl methyltransferase may be independent from its function in overall protein repair.
Western blot analyses in brain extracts of PIMT knockout mice also allowed to better characterize which signalling proteins are hyperphosphorylated under PIMT deficiency. To investigate the cross-talk between PIMT and growth factor signalling in a more controlled system, we have devoted time and efforts towards establishing a mammalian cell model deficient in PIMT protein and activity. Our first attempts to do so using lentiviral transduction with shRNAs in HEK293 and glioblastoma U87 were only partially successful. Despite an up to 70% reduction of PIMT activity in the HEK293 cells, only a very small increase in damaged protein levels was detected between HEK knockdown versus control cells. We therefore turned to the Crispr/Cas9 system to achieve complete gene deletion. We also changed the cell line, as we wanted to get closer to the cell type which seems to be most affected in terms of growth signalling in the PIMT KO mice, namely hippocampal cells. We managed to create a homozygous PIMT knockout cell line, using the mouse hippocampal neuronal cell line HT22. This PIMT KO cell line had no detectable isoaspartyl methyltransferase activity and accumulated 60% higher levels of damaged proteins than control cells. In this cell line, we also found evidence for dysregulation of growth factor signalling, as ERK phosphorylation reached up to 5-fold higher levels after EGF stimulation in the KO cells compared to control cells. In our current experimental conditions we did not detect changes in the activation status of the insulin/IGF1 signaling pathway, but we are currently testing various alternative cultivation and growth factor stimulation protocols to mimic, in the cell model, the effects of PIMT deficiency on insulin/IGF1 signalling observed in the brain of mice. The HT22 PIMT KO cell line created within this project represents a promising model to further investigate the mechanism underlying the cross-talk between PIMT and growth factor signalling under controlled conditions.
We have also started to investigate PIMT deficiency in zebrafish. This is a new objective added within the ‘PIMT and signalling’ project. This task was slightly complicated by the fact that the zebrafish genome encodes two homologs of the mammalian PIMT gene (designated Pcmt and Pcmtl). Producing purified recombinant forms of the zebrafish Pcmt and Pcmtl proteins, we have found that they both act as isoaspartyl methyltransferases with kinetic properties that are highly similar to human PIMT. Zebrafish has emerged over the past years as a model organism to study epileptic disorders. PIMT knockout mice die prematurely of massive epileptic seizures, but the link between the function of PIMT and epilepsy is not understood. We therefore joined forces with Alexander Crawford’s group at LCSB to test whether PIMT deficiency also leads to the development of an epilepsy phenotype in zebrafish, to subsequently exploit this model to pursue our more general aim: better understand how PIMT supports normal brain activity. Using the morpholino methodology, we managed to efficiently knock down Pcmt and Pcmtl, individually or together. We observed dysmorphologies in the Pcmt and Pcmtl morphants, including absence of swim bladder development. When knocking down both the Pcmt and Pcmtl genes, we observed a twitching movement of the embryos tails which is reminiscent of other zebrafish models of epilepsy. These preliminary results need to be confirmed, notably by rescue experiments, but indicate that zebrafish is an attractive model to further understand the precise role of PIMT in brain function.
The C. elegans work was carried out in collaboration with Steven Clarke’s research team at the University of California, Los Angeles. This is the laboratory where I did my postdoctoral studies, before returning to Europe to first integrate the laboratory of Prof. Emile Van Schaftingen at the de Duve Institute (Brussels) in April 2010 as a Research Associate and then move to the University of Luxembourg in January 2012. Here I joined Prof. Rudi Balling’s group as a Research Associate at the Luxembourg Centre for Systems Biomedicine and became a Principle Investigator leading my own research group since January 2013. A PhD student in my group (Remon Soliman) is also working on the ‘PIMT and signalling’ project since March 2013.