Fig. 4.

Virtual screening of Smurf1 small molecular inhibitors and in vitro effects of the inhibitors. a Homology modeling of Smurf1 WW1-WW2 domains. Left beta sheet, WW1 domain; right beta sheet, WW2 domain. b Effects of the top-ranked 15 small molecules (5.0 μM) on Smurf1 activity (Smad1 bound to Smurf1) in osteoblasts from BMP-2ⁿ/Smurf1^e subgroup of 15-month-old osteoporotic mice (OVX at 6 months old). A01, a previously reported Smurf1 inhibitor. c Effects of the 15 small molecules (5.0 μM) on osteocalcin mRNA expression in the above osteoblasts with presence of 100 ng ml^–1 rhBMP-2 or vehicle (PBS). d Structural formula of the chalcone derivative (2-(4-cinnamoylphenoxy)acetic acid). e Relative level of Smad1 bound to Smurf1 in the above osteoblasts incubated with the chalcone derivative (2.5, 5.0, and 10.0 μM). f Relative levels of p-Smad1 and osteocalcin mRNA in the above osteoblasts incubated with the chalcone derivative with presence of 100 ng ml^–1 rhBMP-2 or vehicle (PBS). g Alp activity and mineralized nodule formation in the above osteoblasts incubated with the chalcone derivative with presence of 100 ng ml^–1 rhBMP-2 or vehicle (PBS). h In vitro cell viability of the chalcone derivative at a series of concentrations (1.0, 2.5, 5.0, 10.0, 20.0, 40.0, 60.0, 80.0, 100.0, and 120.0 μM). The levels of Smad1 bound to Smurf1, p-Smad1 and osteocalcin mRNA were normalized to the mean values of osteoblasts treated with vehicle (PBS) or DMSO. n = 6 per group. Data are mean ± s.d. followed by one-way ANOVA with a post-hoc test