Poly(ADP-ribosylation) is present in murine sciatic nerve fibers and is altered in a Charcot-Marie-Tooth-1E neurodegenerative model

BACKGROUND Poly-ADP-ribose (PAR) is a polymer synthesized by poly-ADP-ribose polymerases (PARPs) as a postranslational protein modification and catabolized mainly by poly-ADP-ribose glycohydrolase (PARG). In spite of the existence of cytoplasmic PARPs and PARG, research has been focused on nuclear PARPs and PAR, demonstrating roles in the maintenance of chromatin architecture and the participation in DNA damage responses and transcriptional regulation. We have recently detected non-nuclear PAR structurally and functionally associated to the E-cadherin rich zonula adherens and the actin cytoskeleton of VERO epithelial cells. Myelinating Schwann cells (SC) are stabilized by E-cadherin rich autotypic adherens junctions (AJ). We wondered whether PAR would map to these regions. Besides, we have demonstrated an altered microfilament pattern in peripheral nerves of Trembler-J (Tr-J) model of CMT1-E. We hypothesized that cytoplasmic PAR would accompany such modified F-actin pattern. METHODS Wild-type (WT) and Tr-J mice sciatic nerves cryosections were subjected to immunohistofluorescence with anti-PAR antibodies (including antibody validation), F-actin detection with a phalloidin probe and DAPI/DNA counterstaining. Confocal image stacks were subjected to a colocalization highlighter and to semi-quantitative image analysis. RESULTS We have shown for the first time the presence of PAR in sciatic nerves. Cytoplasmic PAR colocalized with F-actin at non-compact myelin regions in WT nerves. Moreover, in Tr-J, cytoplasmic PAR was augmented in close correlation with actin. In addition, nuclear PAR was detected in WT SC and was moderately increased in Tr-J SC. DISCUSSION The presence of PAR associated to non-compact myelin regions (which constitute E-cadherin rich autotypic AJ/actin anchorage regions) and the co-alterations experienced by PAR and the actin cytoskeleton in epithelium and nerves, suggest that PAR may be a constitutive component of AJ/actin anchorage regions. Is PAR stabilizing the AJ-actin complexes? This question has strong implications in structural cell biology and cell signaling networks. Moreover, if PAR played a stabilizing role, such stabilization could participate in the physiological control of axonal branching. PARP and PAR alterations exist in several neurodegenerative pathologies including Alzheimer's, Parkinson's and Hungtington's diseases. Conversely, PARP inhibition decreases PAR and promotes neurite outgrowth in cortical neurons in vitro. Coherently, the PARP inhibitor XAV939 improves myelination in vitro, ex vivo and in vivo. Until now such results have been interpreted in terms of nuclear PARP activity. Our results indicate for the first time the presence of PARylation in peripheral nerve fibers, in a healthy environment. Besides, we have evidenced a PARylation increase in Tr-J, suggesting that the involvement of cytoplasmic PARPs and PARylation in normal and neurodegenerative conditions should be re-evaluated.