Sox10 Activity and the Timing of Schwann Cell Differentiation Are Controlled by a Tle4-Dependent Negative Feedback Loop

Abstract

The HMG-domain containing transcription factor Sox10 plays a crucial role in regulating Schwann cell survival and differentiation and is expressed throughout the entire Schwann cell lineage. While its importance in peripheral myelination is well established, little is known about its role in the early stages of Schwann cell development. In a search for direct target genes of Sox10 in Schwann cell precursors, the transcriptional co-repressor Tle4 was identified. At least two regions upstream of the Tle4 gene appear involved in mediating the Sox10-dependent activation. Once induced, Tle4 works in tandem with the bHLH transcriptional repressor Hes1 and exerts a dual inhibitory effect on Sox10 by preventing the Sox10 protein from transcriptionally activating maturation genes and by suppressing Sox10 expression through known enhancers of the gene. This mechanism establishes a regulatory barrier that prevents premature activation of factors involved in differentiation and myelin formation by Sox10 in immature Schwann cells. The identification of Tle4 as a critical downstream target of Sox10 sheds light on the gene regulatory network in the early phases of Schwann cell development. It unravels an elaborate regulatory circuitry that fine-tunes the timing and extent of Schwann cell differentiation and myelin gene expression.

Keywords: HMG; Schwann cells; Sox10; glia; transcriptional regulation.

Figure 1

Stage-specific expression of Tle4 in undifferentiated Schwann cells. (a) Relative Tle4 transcript expression in primary rat Schwann cell cultures in proliferative conditions and after cAMP treatment to induce terminal differentiation for 7 days (7 d diff, modified from [17]). (b,c) Relative expression levels of Tle4, Egr2, and Mpz transcripts (b) or proteins (c) in primary rat Schwann cell cultures in proliferative conditions (prol) or after two days of in vitro differentiation (2 d diff). (b) Transcript levels were normalized to Gapdh and Rpl8 levels and the proliferative conditions were arbitrarily set to 1. Data represent mean values ± SEM. Statistical analyses were performed with Student’s two-tailed t-tests (** p ≤ 0.01, *** p ≤ 0.001). (c) Western Blot was performed on whole cell extracts of primary rat Schwann cells with antibodies or antisera directed against Tle4, Egr2, and Gapdh (Gapdh blots refer to Tle4 or Egr2 blots depicted above). Numbers on the right indicate the detected molecular weight in kDa. (d–o,s,t) Representative immunohistochemical stainings of transverse spinal nerve sections on forelimb level from wildtype mice at 13.5 days postcoitum (dpc), 16.5 dpc, and 18.5 dpc using antibodies or antisera directed against Tle4 (green), Sox10 (magenta), Sox2 (cyan (f,j,n); magenta in (s)), and Oct6 (red). Arrowheads mark Sox2/Tle4-positive, Oct6-negative cells, asterisks mark Sox2/Oct6/Tle4-positive cells, and arrows mark Oct6-positive, Tle4-negative cells (s,t). Scale bar: 50 µm. (p–r) Quantifications of immunohistochemical stainings represented in (d–o). Percentages of Sox2/Sox10-positive undifferentiated (undiff.) Schwann cells (SCs) of all Sox10-positive SCs (p), Tle4-positive cells among all Sox10-positive SCs (q) or all Sox2-positive undiff. SCs (r) are depicted as mean ± SEM. Statistical analyses were performed using Student’s two-tailed t-tests (ns p > 0.05, * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001).

Figure 2

Sox10 directly binds to and activates two newly identified Tle4 upstream enhancers. (a) Localization of the evolutionary conserved regions (ECRs, yellow boxes) and the conserved Sox binding site upstream of the mouse Tle4 gene (exons depicted as blue boxes). (b) Schematic representation of the luciferase reporter constructs used in (c,d). The Tle4 ECRs (yellow box) and a minimal promoter (TATA, white box) control the expression of a firefly luciferase gene (green box). (c,d) Transient transfections of the Tle4 ECR luciferase reporter constructs were performed in Sox10-negative Neuro2a neuroblastoma cells. The reporters were co-transfected with either a control plasmid (−) or increasing amounts of Sox10 (+, ++, +++). Reporter gene expression for the Tle4 ECR-19 kb (c) and the wildtype (Sox-wt) or mutated (Sox-mut) Tle4 ECR-78 kb constructs (d) was measured, and effector-dependent activation rates are presented as relative transactivation ± SEM with relative light units of control plasmid-transfected cells arbitrarily set to 1. Statistical analyses were performed using Student’s two-tailed t-tests (* p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001). (e) EMSAs were performed with radiolabeled probes for the wildtype Sox consensus site (Sox-wt) in ECR-78kb and a mutated (Sox-mut) version without cell extracts (−) or cell lysates from HEK293T cells transfected with empty pCMV5 vector (ctrl) or pCMV5-Sox10-MIC (Sox10) vector. Specific bands for Sox10 are marked by an arrowhead.

Figure 3

Tle4 and Hes1 bind to and repress Sox10-dependent gene regulatory regions of Schwann cell differentiation genes and peripheral myelin genes. (a–e) Luciferase assays in Neuro2a cells transiently transfected with reporter constructs containing the regulatory regions Cx32 promoter (a,d,e), Egr2 MSE (b), and Pmp22 intronic enhancer (c) upstream of the firefly luciferase open reading frame. Co-transfections with empty expression vectors (−) or varying amounts of expression vectors (+, ++, +++) for Sox10, Tle4, and Hes1 or a combination were performed as indicated. In (d) either the wildtype Sox10 (wt) or a non-sumoylatable Sox10 variant (K55/246/357R) was used. Reporter gene expression was measured and transfections with empty expression vectors were arbitrarily set to 1, except in (d) where Sox10 activation alone was set to 100% for a better comparison of the specific effects in Tle4 repression (d). Data indicate mean ± SEM and statistical analysis was performed with a one-way ANOVA with Tukey’s multiple comparisons tests (ns p > 0.05, * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001). (f,g) Co-immunoprecipitations were performed with HEK293T lysates overexpressing either myc-Tle4, T7-Hes1 or a combination. Numbers on the right indicate the molecular weight in kDa. (f) Precipitations were performed with either the anti-T7 antibody (f) or anti-myc antibody (g). T7-Hes1 was detected with an anti-T7 antibody and myc-Tle4 was detected with an anti-myc antibody. (h) EMSAs were performed with radiolabeled probes containing the Nbox consensus sites Nbox1 and Nbox2 present in the analyzed Cx32 promoter fragment without cell extracts (−) or with cell lysates from HEK293T cells transfected with the pCMV5 empty vector (ctrl) or pCMV5-T7-Hes1 vector (Hes1). Specific bands for Hes1 are marked by an arrowhead and unspecific bands by an asterisk. (i) Luciferase assays were performed in Neuro2a cells transiently transfected with luciferase reporter constructs containing the regulatory region Cx32 prom in a wildtype variant (wt) or with mutations in the Nbox1 (Nbox1mut), Nbox2 (Nbox2mut) or both Nbox consensus sites (Nbox1/2mut). Co-transfections with empty pCMV5 expression vector (−), expression vectors for Sox10 and Hes1 or a combination were performed as indicated. Reporter gene expression was measured and mean relative transactivation rates ± SEM are depicted. Transfections with Sox10 vector only were arbitrarily set to 1. Statistical analysis was performed with a two-way ANOVA with Tukey’s multiple comparisons tests between meaningful groups (ns p > 0.05, ** p ≤ 0.01).

Figure 4

Forced expression of Tle4 and Hes1 in primary rat Schwann cells represses their differentiation. (a–i) Primary rat Schwann cells were transduced with an EGFP and tdTomato only virus (ctrl, black bar), a Tle4-EGFP and tdTomato only virus (Tle4, dark grey bar), a Hes1-tdTomato and EGFP only virus (Hes1, light grey bar) or a Tle4-EGFP and Hes1-tdTomato virus (Tle4 + Hes1, white bar). (a–h) Representative images of transduced primary rat Schwann cells differentiated for 4 days and stained for Mbp (cyan). Fluorescence of retrovirally expressed EGFP and tdTomato is shown in green and magenta. Scale bar: 100 µm. Arrowheads mark double-transduced cells. (i) Percentage of Mbp-positive cells of all double-transduced Schwann cells. Data represent mean values ± SEM. Statistical analysis was performed with a one-way ANOVA with Tukey’s multiple comparisons tests (ns p > 0.05, ** p ≤ 0.01, *** p ≤ 0.001). (j,k) Co-immunoprecipitations were performed with lysates of HEK293T cells overexpressing myc-Tle4 and Sox10 with or without additional T7-Hes1 (j), or with lysates of primary rat Schwann cell cultures in proliferative conditions (k). Precipitations were performed with anti-T7 antibody (j) or anti-Tle4 antibody (k). Subsequently, immunoblotting was performed to detect Hes1 (anti-T7 antibody), Tle4 (anti-myc antibody) or Sox10 (anti-Sox10 antiserum). The control lane (ctrl) contains lysates of HEK293T cells overexpressing myc-Tle4 and Sox10 for easy identification of endogenous protein bands (k). Numbers on the right indicate the detected molecular weight in kDa. (l) S16 cells were cultured in proliferative (white bars) and in differentiating conditions for 3 days (black bars), comparable to primary Schwann cells. The transcript levels of Schwann cell lineage markers specific to the immature state and the differentiated state were quantified in whole RNA isolations using qRT PCR. Data represent mean values ± SEM. Statistical analysis was performed with Student’s two-tailed t-tests (ns p > 0.05, * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001). (m) S16 cells were transfected with a pCMV5 empty vector (ctrl, black bar), a pCAG-myc-Tle4-IRES-EGFP vector (Tle4, dark grey bar), a pCMV5-T7-Hes1 vector (Hes1, light grey bar), or both expression vectors (Tle4 + Hes1, white bar) and differentiated for 3 days. Immunocytochemistry was performed for Sox10, T7-Hes1, and EGFP as a proxy for Tle4, and Sox10 intensity was measured by flow cytometry in EGFP and T7 double positive cells. Data represent median values ± SEM. Statistical analysis was performed with a one-way ANOVA with Tukey’s multiple comparisons tests (ns p > 0.05, * p ≤ 0.05, ** p ≤ 0.01). (n–w) Representative immunocytochemical stainings, differential interference contrast (DIC), and phase contrast images of S16 cells kept for 1 day in proliferative conditions (1 d prol) or for 3 days in differentiating conditions (3 d diff) using antisera directed against Sox10 (cyan), Oct6 (magenta), and Mpz (green). Scale bar: 100 µm.

Figure 5

Tle4 and Hes1 repress activity of Sox10 enhancers active during Schwann cell development. (a–f) Luciferase assays were performed in S16 cells (a–d) or Neuro2a cells (e,f) transiently transfected with luciferase reporter constructs containing the Sox10 enhancers U1, U2, U3, or D6 and a β-globin minimal promoter (TATA) upstream of the firefly luciferase open reading frame. Co-transfections with empty pCMV5 expression vectors (−), varying amounts of expression vectors (+, ++, +++) for Sox10, Tle4, and Hes1 or a combination of them were performed as indicated. Reporter gene expression was measured and the mean relative transactivation rates ± SEM are depicted. Transfections with the pCMV5 vector only were arbitrarily set to 1. Statistical analysis was performed with a one-way ANOVA with Tukey’s multiple comparisons tests between meaningful groups (ns p > 0.05, * p ≤ 0.05, ** p ≤ 0.01, *** p ≤ 0.001).