YAP and TAZ Regulate Cc2d1b and Purβ in Schwann Cells

Schwann cells (SCs) are exquisitely sensitive to the elasticity of their environment and their differentiation and capacity to myelinate depend on the transduction of mechanical stimuli by YAP and TAZ. YAP/TAZ, in concert with other transcription factors, regulate several pathways including lipid and sterol biosynthesis as well as extracellular matrix receptor expressions such as integrins and G-proteins. Yet, the characterization of the signaling downstream YAP/TAZ in SCs is incomplete. Myelin sheath production by SC coincides with rapid up-regulation of numerous transcription factors. Here, we show that ablation of YAP/TAZ alters the expression of transcription regulators known to regulate SC myelin gene transcription and differentiation. Furthermore, we link YAP/TAZ to two DNA binding proteins, Cc2d1b and Purβ, which have no described roles in myelinating glial cells. We demonstrate that silencing of either Cc2d1b or Purβ limits the formation of myelin segments. These data provide a deeper insight into the myelin gene transcriptional network and the role of YAP/TAZ in myelinating glial cells.


INTRODUCTION
The function of the nervous system relies on the ability of peripheral nerve fibers to transmit information to and from the target tissues. The speed of propagation of action potentials in these fibers is regulated by myelin, a multilamellar structure produced by Schwann cells (SCs; Monk et al., 2015). Damage to SC or peripheral myelin can be caused by numerous factors, including genetic mutations, toxic agents, inflammation, viral infections, metabolic alterations, hypoxia or physical trauma and results in severe peripheral neuropathies. SC integrate biochemical signaling pathways and mechanical stimuli coming from the extracellular matrix or from the axon (Michailov et al., 2004;Feltri and Wrabetz, 2005;Taveggia et al., 2005;Belin et al., 2017). These signals regulate an intricate network of transcription factors that control differentiation of SCs and myelination (e.g., EGR2, YY1, ZEB2, Topilko et al., 1994;Nagarajan et al., 2001;He et al., 2010;Weng et al., 2012;Quintes et al., 2016;Wu et al., 2016).
The identification and characterization of the complete repertoire of transcription factors that modulate myelination is still incomplete (Svaren and Meijer, 2008;Fulton et al., 2011).
The identification of transcription factors responding to a specific signal is one of the first steps in dissecting the underlying regulatory networks. We showed that in Yap fl/+; Taz fl/fl; Mpz-Cre (Yap cHet; Taz cKO) sciatic nerves, SCs lacking YAP/TAZ are unable to myelinate and experience a global dysregulation of transcription . YAP/TAZ are two transcriptional activators of the HIPPO pathway, and play important roles in controlling organ growth, cell differentiation, proliferation and survival (Dupont, 2016). Mechanical stimulation can regulate YAP/TAZ through signals involving FAK, Src, PI3K and JNK pathways (Codelia et al., 2014;Mohseni et al., 2014;Kim and Gumbiner, 2015;Elbediwy et al., 2016), or the formation of actomyosin filaments and accumulation of F-actin (Dupont et al., 2011;Aragona et al., 2013). In addition, YAP/TAZ in SCs can be activated through Crb/Amolt proteins and laminin/G-protein signaling (Colciago et al., 2015;Fernando et al., 2016;Poitelon et al., 2016;Deng et al., 2017). YAP/TAZ regulate gene expression by binding to other DNA-binding transcription factors, especially TEAD transcription factors, but also p73, ERBB4, EGR-1 SMADs RUNXs and TBX5 (Kim et al., 2018). TEADs role in myelination is unknown Lopez-Anido et al., 2015), but TEAD1 binding to transcriptional enhancers is induced during myelination . Furthermore, genes encoding for essential myelin proteins (i.e., Mpz, Pmp22, Mbp and Mag) harbor TEAD elements and are downregulated in Yap cHet; Taz cKO sciatic nerves . Finally, it was suggested that YAP/TAZ and TEAD1 regulate myelin wrapping in cooperation with master myelin regulators EGR2 and SOX10 Poitelon et al., 2016).
To identify novel regulators that are essential for myelination, we integrated RNA-seq and bioinformatics analyses and looked for DNA binding proteins dysregulated in Yap cHet; Taz cKO sciatic nerves at 3 days of age. We identified two highly expressed proteins, i.e., Cc2d1b and Purβ, downregulated in Yap cHet; Taz cKO sciatic nerves. We found that silencing of Cc2d1b or Purβ in vitro significantly decreased the number of myelin segments and silencing of Purβ also significantly decreased the length of myelin segments, independently for effect on SC number, proliferation or apoptosis. These data demonstrate that CC2D1B and PURβ are necessary for myelination in vitro.

Cell Culture
Primary rat SCs were produced as described (Poitelon and Feltri, 2018) and grown with DMEM supplemented with 4 g/l glucose, 2 mM L-glutamine, 5% bovine growth serum, 2 µM forskolin, 50 ng/ml nerve growth factor, penicillin and streptomycin. SCs were not used beyond the fourth passage. Rat dorsal root ganglia (DRG) neurons from Sprague-Dawley rats embryos were isolated at embryonic day 14.5 embryos. DRG were dissociated by treatment with 0.25% trypsin and mechanical trituration and 1.5 DRGs were seeded on collagencoated glass coverslips as described (Poitelon and Feltri, 2018). DRGs cultures were then cycled with fluoroxidine (FUDR, Sigma-Aldrich) to eliminate all non-neuronal cells. Once all non-neuronal cell remove, rat SCs were added (200,000 cells per coverslip) to establish myelinating cocultures of DRG neurons, and myelination was initiated by supplementing the medium with 50 µg/ml ascorbic acid (Sigma-Aldrich). For verteporfin (Sigma SML0534) treatment, verteporfin was solubilized in DMSO at 20 mM, then SCs were treated with either 0.5% of DMSO; 2 or 10 µM of verteporfin for 24 h. mRNA was extracted and cDNA was analyzed by RT-qPCR, as described in Poitelon et al. (2016). This study was carried out in accordance with the principles of the Basel Declaration and recommendations of ARRIVE guidelines issued by the NC3Rs and approved by the Albany Medical College Institutional Animal Care and Use Committee (no. 17-08002).

Bioinformatics
RNAseq data were obtained from NCBI GEO: GSE79115 . Genes encoding for DNA-binding protein genes were predicted thanks to transcriptionfactor.org database. The expression data for Cc2d1b and Purβ were obtained from mousebrain.org/ and gtexportal.org on 09/2018.

Statistical Analyses
Experiments were not randomized, but data collection and analysis were performed blind to the conditions of the experiments. Data excluded are reported in the legend of the figures. Data are presented as mean ± standard error of the mean (SEM) or SD. No statistical methods were used to predetermine sample sizes, but our sample sizes are similar to those generally employed in the field. Two-tailed Student's t-test, One-way analysis of variance (ANOVA) and Two-way ANOVA were used for statistical analysis of the differences between multiple groups according to the number of sample groups. Values of P ≤ 0.05 were considered to represent a significant difference.

Identification of Novel Myelin Regulators in Schwann Cells
Cc2d1b, also named Freud-2, encodes for Coiled-coil and c2 domain containing 1B protein and is highly expressed in peripheral nerves and myelinating oligodendrocytes 2,3 . Purβ encodes for the Purine Rich elem//ent binding protein B. Purβ binding elements have already been characterized in numerous genes, including Mbp and Plp1 (Tretiakova et al., 1999;Dobretsova et al., 2008), yet it is unclear if Purβ is necessary for their expression.
We first confirmed our RNAseq data by qRT-PCR and showed that Cc2d1b and Purβ are downregulated in Yap cHet; Taz cKO sciatic nerves (Figure 2A). Because dysregulation of gene expression in sciatic nerves can be due to alterations of mRNA level in SCs, axons, perineurial or endothelial cells or fibroblasts, we confirmed that Cc2d1b and Purβ are expressed by primary SCs (Figures 2B,E). Finally, we showed that treatment of primary rat SCs with verteporfin, a drug that inhibits YAP/TAZ regulation of transcription by disrupting its interactions with TEAD transcription factors, reduces expression of Cc2d1b and Purβ (Figures 2C,D). Altogether, these data indicate that CC2D1B and PURβ are expressed by SCs and regulated by YAP/TAZ/TEAD.

Cc2d1b and Purβ Regulate Myelination in vitro
To determine the function of CC2D1B and PURβ in SCs, we asked if silencing the expression of Cc2d1b and Purβ in SCs would affect the capability of SC to myelinate axons. SCs were infected with viruses expressing different shRNAs for either Cc2d1b or Purβ. All shRNAs reduced expression of Cc2d1b or Purβ, as shown by quantitative RT-PCR and western blot  . On the right, genes encoding for a DNA-binding protein were selected for the presence of a DNA binding domain (http://www.transcriptionfactor.org) expression levels of all DNA-binding proteins expressed in sciatic nerves are represented. Among these, 64 out of 1,445 are dysregulated. Average logCPM were calculated as log 2 (average CPM + 0.5). (B) Heatmap for the significantly dysregulated DNA-binding proteins in Yap cHet; Taz cKO sciatic nerves at 3 days of age. The differential expression of genes encoding for a DNA binding domain protein was tested in wildtype and Yap cHet; Taz cKO sciatic nerves. Of the 1,445 genes tested, 64 showed statistical significance. Colors in this heatmap correspond to expression levels in Yap cHet; Taz cKO vs. control sciatic nerves, on a scale of yellow for lowest values to magenta for highest values with cyan for moderate values. Genes are categorized based on their expression levels in wildtype sciatic nerves (black denotes high expression levels, whereas white depicts low expression levels). Heatmap data are calculated using Z-score, where z = x-mean in the samples/standard deviation in the samples. (C) The chart categorized DNA-binding proteins repressed Yap cHet; Taz cKO sciatic nerves according to their expression levels in wildtype sciatic nerves (segment width indicates expression level). Among the 10 most expressed DNA-binding proteins (Egr2, Nr2f1, Srebf2, Purβ, Zeb2, Cc2d1b, Klf6, Hif1a, Nfe2l2, Cers4), eight have a role to play in myelin formation (blue). The role of Purβ and Cc2d1b in myelination is unknown (brown).
( Figures 3A,B). In addition, we show that silencing Cc2d1b or Purβ does not affect the protein level of their homolog CC2D1A and PURα ( Figure 3B). When SCs silenced for Cc2d1b and Purβ were seeded on DRG neurons and cocultured in myelinating conditions by adding ascorbic acid to the cultures, myelination of axons was impaired (Figures 3C,D). Because a defect of myelination can be caused by a reduced number of SCs attached to axons, we asked whether silencing Cc2d1b or Purβ caused changes in apoptosis or proliferation. However, silencing of Cc2d1b and Purβ did not affect SC proliferation or apoptosis (Figure 4). Finally, during in vivo and in vitro myelination, Mbp and Mpz gene expression peak 3 days after birth and 40 days after addition of ascorbic acid, respectively (Notterpek et al., 1999). We showed that Purβ expression was not significantly altered during SC development ( Figure 3E) but appears, in vitro, to spike after 3 days in culture, before the start of myelination ( Figure 3F). In contrast, Cc2d1b was highly expressed in vivo between postnatal day 5 (P5) and P20 (Figure 3E), and in vitro after 5 days in culture (Figure 3F), when SCs myelinate axons. Taken together, these data show that CC2D1B or PURβ is required for SC myelination in vitro. Error bars indicate SD. A logarithmic scale was used for the y-axis of (A,C), and the origin was set to 1. Two-way ANOVA with Bonferroni post hoc test. F (2,12) = 817.9, P < 0.0001, Cc2d1b P < 0.0001, Purβ P < 0.0001. (E) Immunolocalization of PURβ in cross section of sciatic nerve at P30 show localization of PURβ in SC cytoplasm and nuclei, although at P30 PURβ staining appears to be more intense in SC cytosol than in the nuclei. Scale bar = 10 µm. Insert display a magnified SCs stained for PURβ. Insert scale bar = 2 µm. Immunoabs displays a validation for PURβ antibody for which PURβ antibody was pre-absorbed with 30-fold excess of the PURβ protein before immunohistochemistry. * * P < 0.01, * * * * P < 0.0001.

DISCUSSION
In this study, we identify novel regulators essential for myelination. Yap cHet; Taz cKO sciatic nerves present an arrest SC early development and an abolition of subsequent SC myelination. We first hypothesized that genes regulated by YAP/TAZ include novel regulators of myelin formation. Thus, we analyzed gene expression by RNA-Seq analysis in Yap cHet; Taz cKO sciatic nerves. In contrast to classical analyses based on gene dysregulation, which would highlight genes highly regulated by YAP/TAZ , we used our dataset to look at DNA binding proteins highly expressed in Yap cHet; Taz cKO sciatic nerves, with the secondary assumption that their level of expression would be correlated to their importance in myelin formation. In this report, we validate our hypotheses and found that global gene expression stratification allows for the identification of genes essential for myelination. We were able to identify that most of the genes known to be either activators or inhibitors of myelination are FIGURE 3 | Cc2d1b and Purβ regulate myelination in vitro. (A) Quantitative RT-PCR analysis showed the silencing efficiency of shRNAs targeting Cc2d1b and Purβ in primary rat SCs. n = 3 experiments. One-way ANOVA with Bonferroni post hoc test. Cc2d1b F (3,16) = 10.83, P = 0.0004, shCc2d1b #1 (−43%) P < 0.0001 shCc2d1b #2 (−36%) P < 0.0001, Purβ F (3,16) = 35.09, P < 0.0001, shPurβ #1 (−64%) P < 0.0001, shPurβ #2 (−75%) P < 0.0001. Error bars indicate SD. A logarithmic scale was used for the y-axis and the origin was set to 1. (B) Western analysis showed that silencing Cc2d1b and Purβ reduce CC2D1B (−41%) and PURβ (−80%) protein levels, but do not after their respective homologs (CC2D1A and PURα). (C,D) Immunolocalization of myelin protein revealed a defect in myelin production in SCs silenced for either Cc2d1b or Purβ. SCs (200,000 cells) infected with shRNA were seeded on dorsal root ganglia (DRG) neurons and allowed to myelinate for 10 days. Cultures were stained for Myelin Basic Protein (MBP, green), neurofilament H (NFH, red) and DAPI (blue). The number and length of myelin segments were quantified. A lower number of segments were observed in both shCc2d1b and shPurβ and shortened myelin segments were observed in shPurβ. n = 4 coverslips for three independent experiments. One-way ANOVA with Bonferroni post hoc test. #myelin segments F (3,44) = 24.31, P < 0.0001, shCc2d1b P < 0.0001, shPurβ P < 0.0001. myelin segments length F (3,44) = 28.03, P < 0.0001, shPurβ P < 0.0001. Error bars indicate SD. (E) Quantitative RT-PCR analysis demonstrated a transient upregulation of Cc2d1b mRNA (dotted line, yellow circles) in sciatic nerves during in vivo myelination, from 10 to 20 days of age. Purβ expression (dashed line, blue triangles) also tend to be increased at 10 days of age. n = 3 animals. One-way ANOVA with Bonferroni post hoc test. F (4,20) = 7.51, P = 0.0007, Cc2d1b 10 days P = 0.001, Cc2d1b 20 days P = 0.017, Purβ 10 days P = 0.09. Error bars indicate SEM. (F) Quantitative RT-PCR analysis demonstrated an upregulation of Cc2d1b (dotted line, yellow circles) in SC-neuron cocultures during in vitro myelination, from 5 days after addition of ascorbic acid (AA). Purβ (dashed line, blue triangles) is also transiently upregulated, 3 days after addition of AA. n = 6 coverslips. One-way ANOVA with Bonferroni post hoc test. F (3,40) = 17.26, P < 0.0001, Cc2d1b 5 days P < 0.0001, Cc2d1b 10 days P < 0.0001, Purβ 3 days P < 0.0001, Purβ 5 days P = 0.05. Error bars indicate SD. * * * * P < 0.0001, * * * P < 0.001, * * P < 0.01, * P < 0.05.
highly expressed in sciatic nerves and are dysregulated in Yap cHet; Taz cKO. Following our reasoning, we identify two novel DNA binding proteins CC2D1B and PURβ, with previously unsuspected role in myelinating glial cells. CC2D1B and PURβ are highly expressed in sciatic nerves and in SCs and downregulated in Yap cHet; Taz cKO. We demonstrate that CC2D1B and PURβ are required for normal myelination. Ablation of either CC2D1B or PURβ impairs myelination in vitro, independently from effects on SC proliferation or apoptosis. Altogether our data demonstrate that CC2D1B and PURβ are both involved in myelin formation in vitro. CC2D1B and PURβ are both DNA-binding proteins, but their role as a regulator in myelinating glial cells or other cells remains undefined. CC2D1B protein structure is close to its homolog CC2D1A.
Cc2d1a is highly expressed in neurons and has been implicated in intellectual disability and autism spectrum disorder (Basel-Vanagaite et al., 2006;Manzini et al., 2014), Cc2d1b is expressed in myelinating glial cells and peripheral nerves 4,5 , which indicates that its role and function might not be fully redundant with Cc2d1a.
Few studies have suggested a redundant role between both proteins for the regulation of serotonin receptors (Hadjighassem et al., 2009(Hadjighassem et al., , 2011). Yet, the transcriptional role of CC2D1 remains controversial, as other studies showed that both CC2D1 are confined to the cytoplasm and perinuclear endosomes (Drusenheimer et al., 2015). Thus, it remains unclear whether CC2D1B can directly control transcription in vivo and whether it translocates from the cytoplasm to the nucleus. Interestingly, other functions have been proposed for CC2D1 proteins independently from their DNA-binding domain. CC2D1 proteins belong to the evolutionary conserved Lgd protein family which was shown to be involved in the regulation of signaling receptor degradation via the endosomal pathway (Jaekel and Klein, 2006). Loss of Lgd function results in an ectopic and ligand-independent activation of the Notch pathway (Childress et al., 2006;Gallagher and Knoblich, 2006;Jaekel and Klein, 2006). Notch signaling promotes the early stage of SC development but inhibits myelination (Woodhoo 4 www.mousebrain.org 5 www.gtexportal.org et al., 2009). Thus, it is possible that CC2D1B modulates myelination through the recruitment of specific signaling complexes. Finally, Cc2d1b KO mice have been reported and present delayed memory acquisition and retention (Zamarbide et al., 2018). There is growing evidence, both from animal studies and human neuroimaging that myelin plays a role in learning (McKenzie et al., 2014;Sampaio-Baptista and Johansen-Berg, 2017) and it might be worthwhile to also consider the role of CC2D1B in central nervous system myelination.
PURβ belongs to the purine-rich element binding (PUR) protein family, which includes of PURα, PURβ and PURγ. There is substantial evidence for PUR role in DNA binding (Rumora et al., 2013;Ferris and Kelm, 2019). Among them, PURα was studied the most, for its implication in fragile × syndrome and PURA syndrome, a disorder characterized by intellectual disability and delayed development of speech and motor skills, such as walking Hunt et al., 2014;Lalani et al., 2014). Interestingly, Purα knockout mice present with reduced myelin production and pathologic development of glial cells (Khalili et al., 2003). However, no knockouts of the other PUR family genes have been reported. PURβ is known to play a role in cell differentiation and modulates transcriptional regulation of gene expression of the α-and β-myosin heavy chain and actin α-2 (Gupta et al., 2003;Ramsey and Kelm, 2009). Notably, TEAD-1, the main transcriptional partner of YAP/TAZ, was reported to be upstream of Purβ (Tsika et al., 2010). TEAD motif-harboring enhancers (GGAAT) can be found in numerous genes dysregulated in Yap cHet; Taz cKO, including CC2D1B and PURβ. Yet, no binding motifs for EGR2 or SOX10 were found in the PURβ enhancer region (Heinz et al., 2010). Thus, it is possible that PURβ is a promyelinating regulator directly downstream of YAP/TAZ/TEAD1. Finally, all PUR isoforms have been associated with neoplasia .
Although the Cc2d1b and Purβ expression appear to be downstream YAP/TAZ/TEADs, the signals contributing to the regulation of their expression are essentially unknown. Elucidation of the upstream pathways and signals that induce CC2D1B and PURβ will be important both for understanding the molecular control of the myelination program, but also potentially for identifying strategies to promote remyelination in demyelinating disease. Indeed, numerous transcription factor involved in developmental myelination is also involved in remyelination following injury. Thus, it will be critical to characterize the role of CC2D1B and PURB in peripheral nerve repair.
Finally, our study extends regulatory mechanisms directing SC myelinogenesis Boerboom et al., 2017;Quintes and Brinkmann, 2017) and supports the transition from a gene-centric to a network-systems view of the myelin formation. Further characterization of the transcription factor network controlling myelin gene expression should help refine our understanding of SC development as well as suggest novel therapeutic strategies to potentiate their regenerative capacity.