Dorsomorphin and LDN-193189 inhibit BMP-mediated Smad, p38 and Akt signalling in C2C12 cells

Abstract

Bone morphogenetic proteins (BMPs) are key regulators of cell fate decisions during embryogenesis and tissue homeostasis. BMPs signal through a coordinated assembly of two types of transmembrane serine/threonine kinase receptors to induce Smad1/5/8 plus non-Smad pathways, such as MAPK and Akt. The recent discovery of BMP receptor inhibitors opened new avenues to study specific BMP signalling and to delineate this effect from TGF-β and Activin signalling. Here we present comprehensive and quantitative analyses on both canonical and non-Smad mediated BMP signalling under Dorsomorphin (DM) and LDN-193189 (LDN) treatment conditions. We demonstrate for the first time, that both compounds affect not only the Smad but also the non-Smad signalling pathways induced by either BMP2, BMP6 or GDF5. The activation of p38, ERK1/2 and Akt in C2C12 cells was inhibited by DM and LDN. In addition "off-target" effects on all branches of BMP non-Smad signalling are presented. From this we conclude that the inhibition of BMP receptors by DM and more efficiently by LDN-193189 affects all known BMP induced signalling cascades.

Fig. 1

Activation of Smad1/5/8, p38 and Akt by different BMP family members is inhibited by Dorsomorphin (DM) and LDN-193189 (LDN) in a dose dependent manner. Serum starved C2C12 cells were treated with DM or LDN for 30 min prior to ligand addition. Samples were incubated for 1 h with 1 nM BMP6 (A), 5 nM BMP2 (B) or 10 nM GDF5 (C). Different ligand concentrations were chosen to consider their individual activity on C2C12 cells. Final DMSO concentration was adjusted to 0.05% (excepting w/o DMSO controls). After cell lysis, samples were immunobloted against p-Smad1/5/8, p-Akt, p-p38 as well as for GAPDH or tubulin as loading control.

Fig. 2

High concentrations of Dorsomorphin (DM) and LDN-193189 (LDN) not only inhibit BMP2-induced activation of Smad1/5/8, p38 and Akt, but also affect their activity in non stimulated cells. Serum starved C2C12 cells were treated with DM (A) and LDN (B) for 30 min prior to ligand addition. Samples were incubated with 5 nM BMP2 for 60 min. Final DMSO concentration was adjusted to 0.2% in DM treated and 0.1% in LDN treated samples respectively (excepting w/o DMSO controls). After cell lysis, samples were immunobloted against p-Smad1/5/8, p-Akt, p-p38 as well as for tubulin as loading control.

Fig. 3

Dorsomorphin (DM) and LDN-193189 (LDN) inhibit BMP2 mediated activation of Smad and non-Smad signalling cascades. Serum starved C2C12 cells were treated with 5 µM DM (A) and 0.5 µM LDN (B) for 30 min prior to addition of 5 nM BMP2. Final DMSO concentration was adjusted to 0.05% in DM treated and 0.005% in LDN treated samples. Cells were lysed at indicated timepoints and subsequently probed for western blotting using different antibodies against p-Smad1/5/8 and the non-Smad signalling targets p-p38, p-Akt, p-ERK1/2, p-ATF2 and p-CREB. Each membrane was also probed against tubulin or GAPDH as loading control.

Fig. 4

BMP dependent activation of ATF2 occurs downstream of ERK1/2, while CREB is activated via p38. (A) Serum starved C2C12 cells were treated with different inhibitors against BMP type I receptors 5 µM Dorsomorphin (DM) or 0.5 µM LDN-193189 (LDN), p38 (PD169316, 10 µM), ERK1/2 (U0126, 10 µM) or against JNK (SP600125, 10 µM) for 30 min prior to ligand addition. After inhibitor pretreatment cells were stimulated by adding 5 nMBMP2 for 30 min. Final DMSO concentration was adjusted to 0.2% in all samples. Cells were lysed at indicated timepoints and subsequently probed for western blotting using different antibodies against p-Smad1/5/8 or the non-Smad signalling targets p-p38, p-AKT, p-ERK1/2, p-ATF2 and p-CREB. Each membrane was also probed against tubulin as loading control. (B) Schematic presentation of BMP-mediated Smad and non-Smad pathways; the point of interference of all used inhibitors is indicated.