Inhibition of myogenesis by Notch: evidence for multiple pathways

Abstract

Notch signaling is critical for skeletal muscle development and regeneration, permitting the expansion of progenitor cells by preventing premature differentiation. We have interrogated the pathways through which ligand-mediated signaling inhibits myogenesis by identifying Notch target genes and assessing their impact on differentiation in vitro. Notch activation led to the robust induction of the transcriptional repressors Hey1 and HeyL in myoblasts, but only constitutive expression of Hey1 blocked myogenesis. siRNA-mediated knockdown of Hey1 had no effect on Notch's ability to inhibit differentiation, suggesting the existence of additional, possibly redundant pathways. We identified 82 genes whose expression was activated when C2C12 myoblasts were cultured in the presence of the Notch ligand Dll4. One of these, MyoR, is a novel Notch-responsive gene, whose protein product is known to repress myogenesis in vitro. siRNA-mediated knockdown of MyoR alone, or in combination with Hey1, was also ineffective at rescuing differentiation in the presence of Dll4. Our data support a model in which Notch signaling inhibits myogenesis through multiple pathways, two of which are defined by the Notch target genes Hey1 and MyoR.

Figure 1

Ligand-induced Notch signaling blocks myogenesis. 6-well plates were coated with 4.5 ml of ligand-containing supernatant per well. C2C12 cells were grown on Fc-Dll4-coated or Fc-control-coated plates and switched from growth medium (GM) to differentiation medium (DM) as indicated. Cells were analyzed after 24 hours for A) cleaved Notch1 and Myogenin proteins (Western); B) MEF2A, MEF2D, MEF2C and Myogenin RNAs (RT-PCR); C) MyoD and D) Myf-5 RNAs (quantitative RT-PCR). MyoD and Myf-5 levels shown in (C–D) are normalized to the Fc-control-GM condition (defined as 1) and plotted as the average +/− standard deviation of two replicate samples. The upper bands of the MEF2A and MEF2D RNA doublets are the differentiation-induced splice variants. β-tubulin protein and HPRT or 18S RNA were used as loading controls.

Figure 2

Ligand-induced Notch signaling significantly induces Hey1 and HeyL expression, but only constitutively expressed Hey1 blocks myogenesis. A) 6-well plates were coated with 3 ml of ligand-containing supernatant per well. C2C12 cells were plated on Fc-Dll4 or Fc-control ligand and propagated for 48 hours in growth medium (GM). Hey1, Hey2 and HeyL RNA levels were determined by quantitative RT-PCR using 18S as a loading control. Expression levels (x) for individual genes were computed from ΔCt values (relative to 18S) according to the formula (x) * 2^ΔCt = (c), where c is an arbitrary constant, and plotted as the average +/− standard deviation of three replicate samples. B) C2C12 cells were stably transduced with parental retrovirus (pBABE-puro) or retroviruses expressing FLAG-tagged Hey1 or HeyL and analyzed for expression of the indicated cDNAs by RT-PCR. RT, reverse transcriptase. C) Transduced cells were propagated in GM, shifted to DM and analyzed for expression of Myogenin, MHC or FLAG-tagged proteins after 1, 2, or 4 days by Western immunoblotting using β-tubulin as a loading control. D) Human myoblasts were grown on Fc-Dll4-coated or Fc-control-coated plates and switched from GM to DM for 24 hours as indicated. Myogenin protein was assessed by Western immunoblotting using β-tubulin as a loading control, and Hey1 RNA was determined by RT-PCR using GAPDH as a loading control.

Figure 3

Ligand-induced Notch signaling effectively blocks myogenesis in cells expressing reduced levels of Hey1. Individual tissue culture wells were treated with Fc-Dll4 or Fc-control as indicated in the Methods. C2C12 cells transfected with either non-silencing (NS) control siRNA oligonucleotides or Hey1-directed siRNAs were propagated on the coated plates and then shifted to DM for 24 hours. Expression of Hey1 (left panel) and Myogenin (right panel) RNA was assessed by quantitative RT-PCR. Q-PCR values are presented as the average +/− standard deviation of three replicate samples. p values were computed by a standard unpaired t-test.

Figure 4

Validation of microarray targets by RT-PCR. 10 cm dishes were coated with 2.5 ml of ligand-containing supernatant. C2C12 myoblasts were plated on either Fc-Dll4 or Fc-control ligand and propagated in growth medium (GM) for six hours. RNA expression of selected genes was determined by (A) RT-PCR or (B) quantitative RT-PCR using HPRT or 18S as a loading control, respectively. Quantitative RT-PCR values for individual genes are normalized to the Fc-control condition (defined as 1) and plotted as the average +/− standard deviation of three replicate samples. (C) C2C12 cells were plated on Fc-control or Fc-Dll4 ligand, propagated in GM, and switched to DM for 24 hours. RNA expression of indicated targets was analyzed by RT-PCR, using HPRT as a loading control.

Figure 5

Constitutive expression of Nrarp or Trib2 does not impair myogenesis, while IL-6 inhibits at high doses. (A) C2C12 cells were stably transduced with parental retrovirus (pBABE-puro) or retroviruses expressing Nrarp or Trib2. Lines were propagated in growth medium (GM) and then shifted to differentiation medium (DM) for 24 hours and analyzed for expression of Myogenin or the indicated cDNAs by RT-PCR using HPRT as a loading control. RT, reverse transcriptase. (B) C2C12 cells were stably transduced with parental retrovirus (pBABE-puro) or retroviruses expressing FLAG-tagged Nrarp or Trib2. Lines were propagated in GM and analyzed for expression of FLAG-tagged proteins by Western immunoblotting. (*) indicates the position of FLAG-Nrarp, and (**) indicates the position of FLAG-Trib2. (C) Transduced cells were propagated in GM, shifted to DM and analyzed for expression of Myogenin, MHC or FLAG-tagged proteins after 1, 2, or 4 days by Western immunoblotting using β-tubulin as a loading control. (D) Fusion of myoblasts into myotubes was examined in the indicated cell lines after three days in DM. (E) C2C12 cells were maintained in GM and then switched to DM for 24 hours in the absence or presence of increasing concentrations of IL-6 (2.8, 10 and 100ng/mL). Myogenin RNA levels were assessed by RT-PCR using HPRT as a loading control.

Figure 6

Constitutive expression of MyoR inhibits myogenesis. C2C12 cells were stably transduced with parental retrovirus (pBABE-puro) or a MyoR-expressing retrovirus (pBABE-MyoR), propagated in growth medium (GM) and then shifted to differentiation medium (DM). Expression levels of the indicated differentiation markers and MyoR were determined by (A) RT-PCR, after 24 hours in DM or (B) Western immunoblot, after 1, 2, or 4 days in DM. (C) Myoblast fusion was examined in pBABE control cells or MyoR-expressing cells after three days in DM. D) ABF-1 (human MyoR) RNA was assessed by quantitative RT-PCR in human myoblasts plated on Fc-Dll4 or Fc-control and propagated in GM. The induction level represents an average +/− standard deviation obtained from two independent myoblast isolations.

Figure 7

(A) Ligand-induced Notch signaling effectively blocks myogenesis in cells expressing reduced levels of MyoR. Individual tissue culture wells were treated with Fc-Dll4 or Fc-control as indicated in the Methods. C2C12 cells transfected with either non-silencing (NS) control siRNA oligonucleotides or MyoR-directed siRNAs were propagated on the coated plates and then shifted to DM for 24 hours. Expression of MyoR (left panel) and Myogenin (right panel) RNA was assessed by quantitative RT-PCR. Q-PCR values are presented as the average +/− standard deviation of three replicate samples. (B) Ligand-induced Notch signaling effectively blocks myogenesis in cells expressing reduced levels of Hey1 and MyoR. Individual tissue culture wells were treated with Fc-Dll4 or Fc-control as indicated in the Methods. C2C12 cells transfected with either non-silencing (NS) control siRNA oligonucleotides or a mixture of Hey1-directed and MyoR-directed siRNAs were propagated on the coated plates and then shifted to DM for 24 hours. Expression of Hey1 (left panel), MyoR (middle panel), and Myogenin (right panel) RNA was assessed by quantitative RT-PCR. Q-PCR values are presented as the average +/− standard deviation of three replicate samples.