Post-transcriptional regulation of adult CNS axonal regeneration by Cpeb1

Adult mammalian CNS neurons are unable to regenerate following axonal injury, leading to permanent functional impairments. Yet, the reasons underlying this regeneration failure are not fully understood. Here, we study the transcriptome and proteome shortly after spinal cord injury. Profiling of the total and ribosome-bound RNA in injured and naïve spinal cords identify a substantial post-transcriptional regulation of gene expression. In particular, transcripts associated with nervous system development were downregulated in the total RNA-fraction while remaining stably loaded onto ribosomes. Interestingly, motif association analysis of post-transcriptionally regulated transcripts identified the cytoplasmic polyadenylation element (CPE) as enriched in a subset of these transcripts that was more resistant to injury-induced reduction at transcriptome level. Modulation of these transcripts by overexpression of the CPE binding protein, Cpeb1, in mouse and Drosophila CNS neurons promoted axonal regeneration following injury. Our study uncovers a global conserved post-transcriptional mechanism enhancing regeneration of injured CNS axons.

functional impairments. Yet, the reasons underlying this regeneration failure are not fully understood. 23 Here, we study the transcriptome and proteome shortly after spinal cord injury. Profiling of the total 24 and ribosome-bound RNA in injured and naïve spinal cords identify a substantial post-transcriptional 25 regulation of gene expression. In particular, transcripts associated with nervous system development 26 were downregulated in the total RNA-fraction while remaining stably loaded onto ribosomes. 27 Interestingly, motif association analysis of post-transcriptionally regulated transcripts identified the 28 cytoplasmic polyadenylation element (CPE) as enriched in a subset of these transcripts that was more 29 resistant to injury-induced reduction at transcriptome level. Modulation of these transcripts by 30 overexpression of the CPE binding protein, Cpeb1, in mouse and Drosophila CNS neurons promoted 31 axonal regeneration following injury. Our study uncovers a global conserved post-transcriptional 32 mechanism enhancing regeneration of injured CNS axons. Early in 1913 Santiago Ramón y Cajal already described that besides the weak and sterile end of 44 axotomized axons set to degenerate, there are active axonal ends, capable of sprouting, for which he 45 termed "bud" or "club of growth", due to their analogy to the growth cones of embryonic axons (Cajal 46 et al., 1991). Interestingly, formation of sprouts at the axonal tip of axotomized dorsal root ganglion 47 (DRG) neurons is accompanied by increased expression of regeneration-associated genes (Ylera et al.,48 2009). Successful regrowth of central DRG axons as induced by a preconditioning peripheral lesion 49 requires the assembly of these regenerating terminal bulbs that are observed 5-7 hours following 50 injury (Ylera et al., 2009). Processes like membrane sealing, regulation of proteolytic processes, RNA 51 stability and local translation are major determinants of successful assembly of a regenerating axonal 52 terminal. Therefore, injured CNS axons do attempt to regrow in the early post-injury phases but 53 ultimately fail to do so. Consistent with this it has been shown that intrinsic pro-regenerative response 54 has to be stimulated before the onset of overt inflammatory response and scar formation (Bradke et   used for RNA profiling to gain insights into the processes taking place in the proximal and distal axonal 111 ends affected by the injury. Polysome-bound transcripts were isolated via sucrose gradient 112 fractionation, and fractions containing RNA bound to two or more ribosomes were collected. Samples 113 were subsequently hybridized onto Affymetrix microarrays. Total and polysome-bound RNAs were 114 normalized separately, as polysome-bound RNA is a subset of total RNA and standard microarray 115 normalization methods that assume equality of distributions of total intensity between arrays could 116 not be used. Notably, correlation plots between arrays shows low variability between replicates, and 117 assessment of expression changes by quantitative real-time PCR (qPCR) largely validated the 118 expression changes obtained from microarray analysis ( Fig. S1A-B). 119 120 Since the chosen tissue not only contains the injured neuronal processes, but also other cellular 121 subtypes, we first analyzed whether the injury would have a major impact on cellular tissue 122 composition at this early time point. To this end we compared the intensities of probesets mapped to 123 marker genes for motor neurons and other local neurons, oligodendrocytes, microglia, precursor cells 124 present in the central canal, and blood-borne cells (see Table S2). The analysis revealed a high 125 correlation between expression patterns of naïve and injured spinal cord (0.97 and 0.99 Pearson 126 correlation for total and polysome fraction, respectively), indicating the absence of major changes in 127 tissue composition upon injury (Fig. S1C). By contrast, a high proportion of probesets in both total and 128 polysome-bound fractions exhibited significant changes upon injury (Fig. 1B). Importantly, for many 129 differentially expressed probesets, the changes in total and polysome-bound RNA do not correlate 130 (Fig. 1B-D, Table S1). A large proportion of changes occur only in the total RNA fraction with no 131 corresponding changes in the polysome-bound RNA. This agrees with previous observations in which 132 stress conditions trigger a general shut down of translation to maximize cell survival (Park et al., 2008;133 Yamasaki and Anderson, 2008). Differentially regulated genes in the total RNA fraction are similarly 134 distributed between up-and down-regulation (1B-C). The polysome-bound RNA fraction showed 135 fewer differentially regulated genes than the total RNA fraction, with most of those being down-136 regulated ( Fig. 1B-C). The difference in numbers of differentially regulated genes between the total 137 and polysome-bound fractions indicates that the translational response to injury is highly uncoupled 138 from RNA availability. In addition, many genes displayed opposing directions of regulation, suggesting 139 considerable influence of post-transcriptional regulation (Fig. 1D).

141
Uncoupled genes are functionally clustered and regulate neuronal regeneration 142 143 To assess the functional role of the observed uncoupling effect, Gene Ontology (GO) (Ashburner et al., 144 2011) enrichment analysis was performed and visualized using Cytoscape (Shannon et al., 2003).

145
Enrichment of up-and down-regulated genes was represented as red and blue nodes respectively. In 146 the total-RNA fractions, transcript availability of genes related to translation, RNA processing, protein 147 catabolic processes and protein transport was increased upon injury, but decreased for genes related 148 to CNS development ( Fig. 2A and Table S3). Excitingly, in the polysomal-bound RNA fraction, injury 149 increased ribosome-loading of genes related to regulation of CNS development, as well as cell death, 150 transcription, RNA processing and immune response ( Fig. 2B and Table S3). Notably, there were no 151 significantly under-represented categories in the polysome fraction, indicating that the decrease in 152 translation after injury is a general effect and neither directed nor functionally clustered.

154
Different trends of enrichment were observed for many GO categories between the total and 155 polysome-bound fractions, suggesting that uncoupling serves specific functional purposes. Of 156 particular note, categories related to CNS development, which are decreased in the total RNA fraction, 157 remain stable or enriched in ribosomal-loaded transcripts. This might explain the temporary 158 regeneration observed following injury, which is absent at later stages of the injury response, as 159 existing transcripts from regenerative genes continue to be translated at first, but are not replaced 160 upon eventual degradation. To investigate if transcripts exhibiting this highly uncoupled behavior 161 affect axonal growth, we turned to Drosophila. Using the UAS-Gal 4 (Brand and Perrimon, 1993), we 162 expressed each of a total of 38 candidate uncoupled genes -for which a fly homologue exists and a 163 UAS line was available -in a fly CNS neuronal population called the small ventral lateral neurons 164 (sLNvs). Most of the genes tested have reduced transcript availability but stable ribosomal loading in 165 both naïve and injured spinal cord (Table S1). We find that 19 (50%) of tested candidates influenced 166 the developmental growth of the sLNv axonal projection: 12 increased outgrowth, while 7 resulted in 167 shorter sLNv projections ( investigate the association of 3'UTR motifs in a given transcript with its expression upon injury. Since 179 this has to be performed on the transcript level, only probe sets mapping to a unique transcript were 180 used. We investigated the presence of cytoplasmic polyadenylation element (CPE), Pumilio binding 181 element (PBE), Musashi binding element (MBE), Hex (hexanucleotide involved in polyadenylation) and 182 AU-rich elements (AREs) (Table S4). To investigate differences in expression changes upon injury 183 relative to motif-free transcripts, we plotted the density curves showing the probability of a data point 184 to have a given log2-fold change, thus reflecting the pattern of distribution of expression of the set of 185 transcripts of interest (Fig. 3).

187
This analysis revealed that transcripts possessing CPE are associated with resistance to injury-induced 188 down-regulation, as compared to transcripts devoid of CPE (Fig. 3A). Notably, this association was also 189 seen for genes related to axon and CNS development GO categories (Fig. 3B). In contrast, presence or 190 absence of CPE did not influence injury-induced changes on the level of ribosomal-loading ( Fig. 3A and  191 S2A). Likewise, RNA transcripts possessing PBE, MBE, Hex and AREs were also associated with 192 resistance to injury-induced down-regulation in the total but not in the polysomal RNA fractions ( Fig.  193 S2B-C, S3A-B). CPE and AREs were also found to co-occur in the mouse transcriptome, and this is 194 associated with resistance to injury-induced down-regulation, suggesting that the two motifs might 195 function in a synergistic manner ( Fig. S3C-D). To ensure that the observed associations are genuine, 196 the same analysis was performed with the motifs on the 5'UTR or with random motifs. As expected, 197 this control experiment does not show any significant association (Fig. S4). Taken together, the data 198 suggest that CPE confers transcript stability against the global decrease induced by spinal cord injury 199 by increasing RNA stability in conjunction with ARE motifs.

201
Although the 3'UTR-tested motifs are associated with higher resistance to injury-induced down-  (Table S1). Together with the fact that Cpeb1 overexpression in Drosophila 207 promoted robust axonal outgrowth of developing sLNvs (Table 1), we chose Cpeb1 for further detailed 208 investigation.

210
To elucidate whether CPE has a general functional role, GO enrichment studies were performed on the 211 prevalence of CPE among all protein coding genes of the mouse and fly genomes. Many nervous 212 system development categories were enriched among CPE containing genes in the mouse genome, 213 including neuron projection morphogenesis, axonogenesis and axon guidance ( Fig. 3C and Table S5).

214
There were no categories found with an under-representation of CPE. Interestingly, almost all 215 categories in the mouse genome enriched in CPE-containing genes are also enriched in Drosophila, 216 suggesting a high level of conservation of CPE function between the two species.

218
Cpeb1 promotes regeneration following neuronal injury 219 220 Our data suggest that CPE-enriched transcripts are temporarily protected from degradation after 221 injury. Surprisingly, neither Cpeb1 mRNA nor protein levels were changed significantly upon SCI 222 (although a trend to down-regulation could be observed (Fig. S5). Whether Cpeb1 is specifically down-223 regulated in neurons following SCI could not be assessed due to the lack of a working anti-Cpeb1 224 antibody for immunohistochemistry. To address the neuronal specific role of CPEB1 in axonal 225 regeneration, and specifically whether the failure to upregulate Cpeb1 might in part explain the 226 transient and abortive nature of the regenerative response to injury, we overexpressed the fly 227 homologue of Cpeb1, Orb, exclusively in the sLNvs. Fly brains were dissected and kept in culture as To investigate whether this effect is conserved in mammals too, we turned to a mouse model of optic 235 crush injury that allows overexpression of Cpeb1 in mouse retinal ganglion cells (RGCs) via infection 236 with adeno-associated viral (AAV) vectors. Thereafter, regeneration of RGC-axons was assessed 237 following a crush injury of the optic nerve. Importantly, as in Drosophila, over-expression of Cpeb1 238 enhanced axon regeneration in the mouse optic nerve, as both the number and length of regenerated 239 axons were higher when measured 2 weeks after injury as compared to AAV-GFP infected control 240 RGCs ( Fig. 5E-F). The number of RGCs in the retina remained constant, indicating that the regenerative 241 effect is not due to reduced cell death after injury ( Fig. 5G-H). To test whether knockout of Cpeb1 242 produces an opposite effect, we knocked out Cpeb1 in primary cultures of mouse cortical neurons. 243 Efficient knockout is triggered via AAV-Cre mediated deletion of exon 4 (Fig. S6A), which causes a 244 frameshift that affects the activation and RNA recognition domains of Cpeb1. Neurons were cultured 245 in a transwell chamber which specifically allows neurites to grow on the underside of the chamber. 246 Scraping the lower side of the transwell mimicked a transection-injury. Thereafter, regenerating 247 neurite on the underside could be examined. Notably, knockout of Cpeb1 was found to reduce the 248 number and length of regenerated neurites 24 hours after injury ( Fig. S6B-C). Together, these data 249 support the notion that Cpeb1 is an enhancer of regeneration, and that this function is conserved 250 between mice and Drosophila. These findings agree well with the increasing number of studies that report the role of axonal Animals were subjected to laminectomy and then an 80% transectional spinal cord injury by cutting 339 the spinal cord with irridectomy scissors. Naïve mice were subjected only to laminectomy. Nine hours 340 after the operation, 2.5 cm of spinal cord tissue centering on the lesion site was extracted. Procedures 341 were conducted in accordance with the DKFZ guidelines and approved by the Regierungpräsidium 342 Karlsruhe.

344
Translation state array analysis (TSAA) 345 346 RNA was isolated from the chosen tissue and segregated into a total RNA fraction and a polysome-347 bound RNA fraction. Polysome-bound RNA was isolated via fractionation in sucrose gradient as 348 previously described (Lou et al., 2014), with fractions containing RNA bound to two or more 349 ribosomes collected.

351
The samples were profiled with Affymetrix arrays (model Mouse430_a2) with RNA input amounts of 5 352 µg and 3µg for total and polysome-bound RNA respectively. Array data is accessible from GEO 353 (GSE92657). Array data corresponding to each fraction were normalized separately, as polysome-354 bound RNA is a subset of total RNA and standard microarray normalization methods that assume 355 equality of distributions of total intensity between arrays could not be used.

357
Normalization was performed using the vsn method as implemented in the vsnrma function from the 358 R/bioconductor package vsn (Huber et al., 2002). lts.quantile=0.5 was used to allow for robust 359 normalization when many genes are differentially expressed. Differential expression was calculated 360 using limma (Smyth, 2004) from R/bioconductor at the level of probesets, the parameter "trend" was 361 set to TRUE for the empirical moderation of standard errors. Probesets with Benjamini-Hochberg 362 false-discovery rates (FDR) <0.05 were considered as differentially expressed. Probesets were then 363 translated to Ensembl gene IDs (Ensembl v72; www.ensembl.org), and those mapping to multiple IDs 364 were excluded from subsequent analysis. For cell marker analysis normalised microarray intensities 365 were averaged over replicates. The expression values for 103 measured marker genes were used to 366 show the composition of samples.

368
Gene Ontology enrichment 369 370 Enrichment analysis of GO biological process categories between up-and down-regulated genes were 371 performed separately for each of the RNA fractions. To this end, up-regulated genes were compared 372 against a background of all differentially expressed genes with the hypergeometric distribution, using 373 GOstats (Falcon and Gentleman, 2007) and annotation from the org.Mm.eg.db v2.14.0 package, both 374 from R/bioconductor. Under-representation of up-regulated genes in a category is equivalent to 375 having an enrichment of down-regulated genes, and is displayed as such.

377
For the enrichment study of CPE in the mouse and fly genomes, GO annotations from annotation 378 packages org.Mm.eg.db v2.14.0 and org.Dm.eg.db v2.14.0 with experimental evidence code (i.e. EXP, 379 IDA, IPI, IMP, IGI, IEP) were used. The selectiveness is because GO annotations are based on different 380 kinds of evidence including homology, and circularity would occur when comparing results from two 381 different genomes with homology included. To prevent artificially inflating the number of motifs when 382 genes have more than one transcript with common 3'UTR, we performed the enrichment analysis at 383 the level of genes. Transcripts with annotation of transcript biotype as protein coding were translated 384 to Ensembl gene ID and then to Entrez. Genes with only CPE-containing transcripts were compared 385 against all genes (excluding those with both CPE-containing and CPE-free transcripts) with the 386 hypergeometric distribution using GOstats.

389
Results were represented on the GO subnetworks comprising of categories significant in at least one 390 of the comparisons (Benjamini-Hochberg FDR <1e-4 for Fig. 2  Optic nerve crush injury 408 409 Experimental procedures were performed in compliance with animal protocols approved by the 410 Animal and Plant Care Facility at the Hong Kong University of Science and Technology. C57BL/6 mice of 411 5-6 weeks of age were anesthetized with ketamine (80 mg/kg) and xylazine (10 mg/kg) and received 412 Meloxicam (1 mg/kg) as analgesia after the surgery. AAV2 vectors expressing either Cpeb1 or Gfp 413 under the neuron-specific human synapsin promoter were injected into the vitreous bodies with a 414 Hamilton microsyringe. Five weeks after vector injection, the optic nerve was gently exposed 415 intraorbitally and crushed with jeweler's forceps (Dumont #5; Fine Science Tools) around 1 mm behind 416 the optic disk. Mice were kept for 2 weeks after injury before tracing. To visualize RGC axons in the 417 optic nerve, 1.5 μL cholera toxin β subunit conjugated with Alexa555 (CTB555, 2 μg/μl, Invitrogen) 418 were injected into the vitreous bodies. Two days after the CTB injection, animals were sacrificed by 419 transcardial perfusion for histology examination. In each mouse, the completeness of optic nerve 420 crush was verified by showing that anterograde tracing did not reach the superior colliculi.  Gal4 line was obtained from P. Taghert. All flies were dissected 2-10 days after eclosion. 477

Drosophila outgrowth and injury assays 478
To measure axonal outgrowth during development, flies were reared at 25°C and were dissected in 2012)and dishes were kept in a humidified incubator at 25°C. Four days later, cultured brains were 491 fixed and immunohistochemical staining was performed as for freshly dissected samples. 492

Imaging, morphometric measurements and statistics 493
For the outgrowth experiments, brains were visualized under a fluorescent microscope equipped with 494 a GFP filter and classified as having "increased outgrowth", "reduced outgrowth" or "no observable 495 effect" according to comparison of sLNv length with that of controls.
For the injury experiments, de novo growth was assessed four days after injury by measuring injured 497 sLNv projection that has formed at least one new axonal sprout of a minimum length of 12µm. The 498 exact injury location was accessed by comparison with axonal projection length at 5 hours (where no 499 de novo growth has occurred). Imaging was performed on a Zeiss 500 or 700 confocal microscope and 500 analyzed with Image J. Regenerated length was defined as the de novo axon lengths using the manual 501 tracing tool. Projected distance was defined as the displacement of the axon sprouts from the lesion 502 site, measured in a straight line. All images were analyzed in a blind manner. Statistical comparisons 503 were performed with two-tailed student's t-test. 504

506
We thank Thomas Hielscher, Axel Benner, Tim Holland-Letz and Simone Braun from the DKFZ 507 Biostatistics Division for statistical advice; Ha Nati from the University of Heidelberg for help with 508 Cytoscape; and Damir Krunic from the DKFZ Light Microscope Core Facility for providing the macro for 509 neuronal process tracing. This study was supported by the German Ministry of Education and 510 Research ( stabilization applied to microarray data calibration and to the quantification of differential expression. 578 Bioinformatics 18, S96-S104. Differentially regulated probesets upon SCI      Figure S1: Correlation and validation of microarray. A) Correlation plot of normalised arrays (log2[normalised intensities]) for total and polysome-bound RNA fractions. B) Comparison of expression changes of selected genes derived from microarray or qPCR. C) Expression patterns of cell-type-specific genes are similar in both, the condition 'naive' and 'injured'. Colours represent normalised intensities of microarray probes mapped to marker genes for motor neurones, other neurones, oligodendrocytes, microglia, precursors, and blood cells. The expression patterns between the conditions 'naive' and 'injured' of these markers show a Pearsons' correlation of 0.97 for total RNA and 0.99 for RNA bound to polysomes.    Figure S4: Control analysis for motif analysis. Substituting the motif analysis with A) the same motifs but in the 5' UTR and B) random motifs shows no association with expression changes.