Abstract:
The purpose of this study was to elucidate the role of epigenetic mechanisms for the regulation of genes associated with axonal regeneration following peripheral sciatic nerve axotomy (SNA) as compared to central dorsal column axotomy (DCA). Axon regeneration is usually only successful in the peripheral system. According to the hypothesis, injury type dependent differential changes of the gene expression pattern were expected to be associated with corresponding changes of the epigenetic code. Specifically, gene promoter DNA hypermethylation would correlate with downregulation of regeneration-associated genes (RAGs) following injury, compared to sham. Inversely, promoter hypomethylation would correlate with an upregulation. Furthermore, specific histone modifications might be associated with injury-induced RAG expression.
Mouse dorsal root ganglia (DRG) were used as a suitable in vivo model for axon regeneration that allows the investigation of differential responses and effects to both types of nerve lesion within the same neurons. One, three, or seven days following either SNA or DCA, injury-induced changes of promoter-bound epigenetic marks were assayed and correlated to gene expression changes associated with axon regeneration. First, methylated DNA immunoprecipitation was combined with a special microarray for methylated DNA (MeDIP-chip) in the frame of a genome-wide promoter and CpG island (CGI) DNA methylation analysis. Altogether, 179 hyper- or hypomethylated genes were identified for both injury conditions. A subset of these genes was differentially methylated (DM) exhibiting injury-induced changes of methylation levels only upon SNA or DCA. Many of these genes were associated with functions in chromatin remodeling, transcription regulation, or neural development or differentiation. For a subset of the DM genes, the promoter methylation status correlated with gene expression changes upon injury according to the hypothesis. Gene expression of known major RAGs such as Gap43, Sprr1a, and Bdnf was verified to be upregulated solely upon SNA. However, most of these genes were not significantly methylated. Consequently, a promoter CGI analysis was performed for RAGs and DM genes to identify predictable correlations between promoter methylation status and CpG dinucleotide distribution. Almost all investigated genes, especially RAGs, have one or more CGIs within the proximal promoter region or close to the transcription start site (TSS), except Sprr1a. DM genes exhibited a higher normalized CpG density around the TSS than analyzed RAGs. Thereby, differentially hypermethylated genes had higher normalized CpG values than hypomethylated genes, and moderately induced RAGs had higher values compared to highly induced RAGs.
Second, increased expression of major RAGs was shown to be associated with increased promoter histone-3 lysine-9 acetylation (H3-K9ac), and decreased dimethylation (H3-K9me2) following peripheral injury. Overexpression of PCAF (KAT2B), a histone acetyltransferase for H3-K9, in dissociated mouse DRG neuron cultures, or in mouse cerebellar granule neurons enhanced neurite outgrowth, even on inhibitive CNS myelin. This was associated with the prevention of decreasing promoter H3-K9ac occupancy, and with increased RAG expression on myelin compared to a permissive substrate.
Altogether, promoter DNA methylation seems not to be majorly involved in regeneration-associated gene expression regulation although it seemed to be important for specific induced genes. However, H3-K9 acetylation specifically plays a role for RAG expression upon SNA and for promoting neurite outgrowth. Enhancing local promoter histone acetylation by overexpression of PCAF can even overpower the outgrowth inhibiting effects of CNS myelin in vitro inducing a certain degree of regeneration in cultured DRG neurons or CGN. This study succeeded to show that specific epigenetic mechanisms are involved in gene regulation in the frame of axonal regeneration, although the whole picture is still far from being complete. Further studies are mandatory and might help developing combinational clinical therapies for treatment of spinal cord injury.