Sox10 cooperates with the mediator subunit 12 during terminal differentiation of myelinating glia

Abstract

Several transcription factors are essential for terminal differentiation of myelinating glia, among them the high-mobility-group-domain-containing protein Sox10. To better understand how these factors exert their effects and shape glial expression programs, we identified and characterized a physical and functional link between Sox10 and the Med12 subunit of the Mediator complex that serves as a conserved multiprotein interphase between transcription factors and the general transcription machinery. We found that Sox10 bound with two of its conserved domains to the C-terminal region of Med12 and its close relative, Med12-like. In contrast to Med12-like, substantial amounts of Med12 were detected in both Schwann cells and oligodendrocytes. Its conditional glia-specific deletion in mice led to terminal differentiation defects that were highly reminiscent of those obtained after Sox10 deletion. In support of a functional cooperation, both proteins were jointly required for Krox20 induction and were physically associated with the critical regulatory region of the Krox20 gene in myelinating Schwann cells. We conclude that Sox10 functions during terminal differentiation of myelinating glia, at least in part by Med12-dependent recruitment of the Mediator complex.

Figure 1.

Sox10 and Med12l interact with each other. A, In addition to full-length Med12, Med12l, and Sox10, fragments used in interaction studies are schematically represented. Numbers represent amino acid positions. Conserved regions include the PQL and OPA domains in Med12 and Med12l and the Dim/HMG, K2, and TA domain of Sox10. B, Pull-down assays were performed with Sox10 fragments immobilized as GST fusions on Sepharose beads and Med12l CT produced in 293 cell extracts. Detection of Med12l CT was by Western blot using an antibody directed against the T7 epitope tag. C, D, Med12l fragments NPQL, CPQL, and OPA were fused to GST and used in pull-down assays to precipitate full-length Sox10 (C) or the conserved Dim/HMG, K2, and TA domains of Sox10 (D). Full-length Sox10 was generated in 293 cells, its conserved domains as T7-tagged versions in bacteria. Antibodies directed against Sox10 or the T7 epitope tag were used to detect the precipitated proteins by Western blot. Input corresponds to one-tenth of the amount of the protein used in the assay. Control pull-down experiments were performed with GST only.

Figure 2.

The interaction between Sox10 and Med12l is mediated by multiple conserved regions in both proteins. A, Pull-down assays were performed with the Sox10 transactivation domain immobilized as GST fusion on glutathione Sepharose beads and the T7-tagged Dim/HMG domain of Sox10 in the absence (top) or presence (bottom) of CPQL. B, Microdeletions del1–del6 were introduced into CPQL as outlined. Numbers define the deleted regions. C, Pull-down assays with GST fusions carrying CPQL and its deletion variants served to map the contact points for the Sox10 TA domain in Med12l. D, The interaction interface between the Sox10 Dim/HMG domain and Med12l was investigated in pull-down assays with GST fusions carrying the CPQL subfragments 1–3 (Fig. 1A). The proteins to be pulled down are listed on the left of each panel and were all supplied as bacterially produced and purified versions. Detection was by Western blot using an antibody directed against the T7 epitope present in all proteins. Generally, input corresponds to one-tenth of the amount of the protein used in the assay. Control pull-down experiments were performed with GST only.

Figure 3.

Sox10 and Med12 interact and are coexpressed in glia. A–C, Full-length, T7-tagged Med12 was coexpressed with Sox10 (A), Sox10ΔK2 (B), or Oct6 (C) in 293 cells and analyzed for its ability to interact with these proteins in coimmunoprecipitation (IP) assays. Antisera directed against Sox10 (αSox10) or Oct6 (αOct6) and preimmune serum (PI) were used. Top, Western blot (WB) detection of Sox10 (A,B) or Oct6 (C). Bottom, Western blot detection of Med12 using specific antibodies. Input corresponds to one-tenth of the amount of the protein used in the assay. D, Conserved Sox10 regions Dim/HMG, K2, and TA were fused to GST and used in pull-down assays to precipitate T7-tagged Med12. Med12 in the precipitate was detected by Western blot using an antibody directed against the T7 epitope. E–H, Expression levels of Med12 (black bars) and Med12l (white bars) were determined by qRT-PCR in RNA prepared from sciatic nerve (E) during postnatal weeks 1, 2, 3, and 4 (w1, w2, w3, w4) and in the adult (ad), from embryonic (emb) and postnatal (w1) spinal cord (F), from primary Schwann cells (SC) and the S16 Schwann cell line (G), and from primary oligodendrocytes kept under proliferating (pOL) or differentiating (dOL) conditions, as well as CG4 and OLN93 cell lines (H). Transcript levels were normalized to β-actin levels in the respective samples. For each RNA source, RT-PCRs were performed on at least three independent samples.

Figure 4.

Sox10 interacts with Med12 and the Mediator complex under physiological conditions. A, B, Coimmunoprecipitation (IP) of endogenous Med12 and other mediator subunits with anti-Sox10 antiserum (αSox10) or preimmune serum (PI) from S16 (A) and OLN93 (B) cell extracts. Top, Western blot (WB) detection of Sox10. Bottom panels probe the presence of Med12, Med13, Cdk8, Med1, and Med15 in the precipitate using antibodies specifically directed against the respective protein. Input corresponds to one-tenth of the amount of the protein used in the assay.

Figure 5.

Myelin expression is absent from mouse PNS and CNS in the absence of Med12. A–H, In situ hybridizations were performed on wild-type embryos (A,C,E,G) and Med12^ΔCNP1 littermates (B,D,F,H) at embryonic day 18.5 with probes directed against Mbp (A,B,E,F), Mpz (C,D), and Plp (G,H). Shown are spinal nerve (A–D) and spinal cord (E–H) in transverse section. Size bars correspond to 200 μm with the bar in D being valid for A–D and the one in H for E–H. I, J, Quantification of Mbp- and Plp-positive cells in the wild-type and mutant spinal cord at embryonic day 18.5. Numbers in the wild-type spinal cord were set to 100%. At least five separate 14 μm sections from the forelimb region of two independent embryos were counted for each genotype. Data are presented as mean ± SEM. Differences from the wild-type were statistically significant as determined by the Student's t test (p ≤ 0.001).

Figure 6.

In the absence of Med12, Schwann cells fail to upregulate Krox20 expression and do not induce the myelination program. A–H, Immunohistochemistry was performed on spinal nerves of wild-type embryos (A,C,E,G) and Med12^ΔCNP1 littermates (B,D,F,H) at embryonic day 18.5 with antibodies directed against Sox10 (A,B), Sox2 (C,D), Oct6 (E,F), and Krox20 (G,H). Scale bar, 50 μm. I, J, Quantification of Oct6- and Krox20-positive cells as the percentage of total Sox10-positive Schwann cells in wild-type and mutant spinal nerves at embryonic day 18.5. At least 10 separate nerve sections from two independent embryos were counted for each genotype. Data are presented as mean ± SEM. Differences from the wild-type were statistically significant only for the number of Krox20-positive cells as determined by the Student's t test (p ≤ 0.001). K, Expression levels of the Schwann cell regulators Krox20, Yy1, Nfatc4, and Gpr126 were determined by qRT-PCR in RNA prepared from sciatic nerve of wild-type embryos (black bars) and Med12^ΔCNP1 littermates (white bars) at embryonic day 18.5. Transcript levels were normalized to β-actin levels in the respective samples. For each RNA source, RT-PCRs were performed on three independent samples, each in triplicate.

Figure 7.

In the absence of Med12 oligodendrocytes fail to induce the myelination program. Immunohistochemistry was performed on transverse spinal cord sections of wild-type embryos (A,C,E,G,I,K,M,O,Q,S) and Med12^ΔCNP1 littermates (B,D,F,H,J,L,N,P,R,T) at embryonic day 18.5 (A–R) or immediately after birth (S,T) with antibodies directed against Olig2 (A–D), Pdgfra (E–H), Sox10 (I–L), Nkx2.2 (M,N), Gpr17 (O, P), Mrf (Q,R), and Mbp (S,T). Photographs were taken both from the mantle zone (A,B,E,F,I,J) and the marginal zone (C,D,G,H,K-T) of the ventral spinal cord as schematically indicated in U. Scale bar, 50 μm.

Figure 8.

Sox10 and Med12 cooperate functionally in Schwann cells. A, Detection of Krox20 transcripts in S16 cells by RT-PCR and consecutive gel electrophoresis of PCR products. − indicates a water control. B–D, Chromatin immunoprecipitation was performed on chromatin prepared from S16 Schwann cells with antibodies directed against Med12 and control IgGs. qRT-PCRs were performed on immunoprecipitated chromatin to determine the relative enrichment of several regions from the Krox20 gene locus in the Med12 precipitate over the control. Experiments were performed at least three times with each PCR in triplicate. The regions probed by qRT-PCR (−2, −0.6, TS, +36, MSE) and their relative location to the Krox20 gene are depicted in B. The enrichment obtained for each region relative to the IgG control is summarized in C, and D shows a representative chromatin immunoprecipitation experiment performed in triplicate for the Krox20 MSE with input chromatin and chromatin precipitated by control IgG or anti-Med12 antibodies. − indicates a water control. E, Determination of shRNA efficiencies by Western blot with anti-Med12 antibodies using extracts from 293 cells transfected with mouse Med12 in the presence of pSuper-based expression plasmids. In addition to the empty expression plasmid, versions were used that contained Med12-specific shRNAs (shM1 and shM2) or an shRNA with scrambled sequence (scr). Acetylated tubulin (acTub) served as a loading control. F, G, Determination of shRNA efficiencies by qRT-PCR on RNA prepared from S16 cells transfected with expression plasmids for Med12-specific (shM1 and shM2) or srambled (scr) shRNAs. After normalization to β-actin, Med12 (F) and Krox20 (G) expression levels in S16 cells transfected with scrambled shRNA were set to 1, and expression levels in the presence of Med12-specific shRNAs were expressed relative to it. Error bars represent the differences between two biological replicates with PCR on each sample performed in triplicate. H–M, Transient transfections were performed in S16 cells with luciferase reporters under the control of the Krox20 MSE (H–K), the HR1a (L), or the HR2b (M) element of the Oct6 SCE. As indicated below the lanes, expression plasmids were added to some transfections. These included expression plasmids for Sox10 (H,J,K,L,M), a Sox10VP16 fusion protein (I), and Oct6 (J), as well as for shRNAs directed against Med12 (shM1 and shM2), Cdk8 (sh8), Med1 (sh1), Med4 (sh4), or containing a scrambled version (scr). Luciferase activities in extracts from transfected cells were determined in three experiments each performed in duplicate. The luciferase activity obtained in the absence of cotransfected transcription factor plasmid, but in the presence of shRNA, was arbitrarily set to 1 and the fold inductions were calculated for all other samples relative to it. Data are presented as mean ± SEM. Differences between activation rates were statistically significant for Med12-specific shRNA-containing transfections in H and J, and for Cdk8-, Med1- and Med4-specific shRNA-containing transfections in K according to the Student's t test (p ≤ 0.001).