Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2014 Nov 7;289(45):31150-9.
doi: 10.1074/jbc.M114.579771. Epub 2014 Sep 18.

Regulation of bone morphogenetic protein 9 (BMP9) by redox-dependent proteolysis

Zhenquan Wei  1 , Richard M Salmon  1 , Paul D Upton  1 , Nicholas W Morrell  2 , Wei Li  3
Affiliations

Regulation of bone morphogenetic protein 9 (BMP9) by redox-dependent proteolysis

Zhenquan Wei et al. J Biol Chem. .

Abstract

BMP9, a member of the TGFβ superfamily, is a homodimer that forms a signaling complex with two type I and two type II receptors. Signaling through high-affinity activin receptor-like kinase 1 (ALK1) in endothelial cells, circulating BMP9 acts as a vascular quiescence factor, maintaining endothelial homeostasis. BMP9 is also the most potent BMP for inducing osteogenic signaling in mesenchymal stem cells in vitro and promoting bone formation in vivo. This activity requires ALK1, the lower affinity type I receptor ALK2, and higher concentrations of BMP9. In adults, BMP9 is constitutively expressed in hepatocytes and secreted into the circulation. Optimum concentrations of BMP9 are essential to maintain the highly specific endothelial-protective function. Factors regulating BMP9 stability and activity remain unknown. Here, we showed by chromatography and a 1.9 Å crystal structure that stable BMP9 dimers could form either with (D-form) or without (M-form) an intermolecular disulfide bond. Although both forms of BMP9 were capable of binding to the prodomain and ALK1, the M-form demonstrated less sustained induction of Smad1/5/8 phosphorylation. The two forms could be converted into each other by changing the redox potential, and this redox switch caused a major alteration in BMP9 stability. The M-form displayed greater susceptibility to redox-dependent cleavage by proteases present in serum. This study provides a mechanism for the regulation of circulating BMP9 concentrations and may provide new rationales for approaches to modify BMP9 levels for therapeutic purposes.

Keywords: Bone Morphogenetic Protein (BMP); Cell Signaling; Crystal Structure; Disulfide; Endothelial Cell; Redox Regulation.

PubMed Disclaimer

Figures

FIGURE 1.
FIGURE 1.
Expression, activity, and purification of recombinant BMP9. A, schematic diagram of BMP9 production and processing. B, schematic diagram of the generation of HBMP9. A His tag (filled star) is introduced at the beginning of the mature BMP9. C, conditioned medium from HEK-EBNA cells transfected with prepro-BMP9 (lane 1) or prepro-HBMP9 (lane 2) were separated on a non-reducing SDS-PAGE and probed with anti-BMP9 antibody (MAB3209). The diagram on the right shows the schematic drawing of the BMP9 molecules. D denotes dimer on non-reducing SDS-PAGE, and M denotes monomer on SDS-PAGE. D- and M-forms migrate slower in the HBMP9 due to the addition of the His6 tag at the N terminus of the mature ligand as depicted in B. D, HEK cell-produced BMP9 and HBMP9 have comparable activity with BMP9 from R&D Systems. Conditioned media containing pBMP9 (because it is very likely to be present as a prodomain bound complex in the conditioned media) or pHBMP9 were quantified by ELISA using R&D BMP9 as a standard and subjected to signaling assay using C2C12 cells transfected with ALK1. Plasmid containing Renilla was co-transfected with reporter plasmid containing BMP response element-luciferase, and the luciferase activity induced by BMP9 signaling was read using the Promega Dual-Luciferase system. E, purified pBMP9 and HBMP9 were fractionated on a 12% non-reducing SDS-PAGE and stained by Coomassie Blue. F, identical samples of R&D Systems BMP9, pBMP9, and HBMP9 were run in parallel on three 12% non-reducing SDS-PAGE, then blotted separately against the following: anti-BMP9 antibody (left), anti-BMP9 prodomain antibody (middle), and anti-His tag antibody (right). A single asterisk indicates a minor band that could not be seen on the SDS-PAGE in E but reacted very strongly with anti-BMP9 antibody. This may be a species of partially processed BMP9. Double asterisks indicate nonspecific carrier protein from R&D Systems BMP9. Pro, prodomain.
FIGURE 2.
FIGURE 2.
M-form BMP9 is a non-covalently linked dimer. A, HBMP9 was loaded onto a Superdex 75 10/30 gel filtration column pre-equilibrated in 50 mm Tris·HCl, pH 7.4, containing 150 mm NaCl. Non-reducing SDS-PAGE of the peak fractions (B3 to B5) revealed the D- and M-forms of HBMP9 co-elute under the same peak. The doublet in the M-form on SDS-PAGE was probably due to a partial reduction of intramolecular disulfide bonds. B, a representative of the HBMP9 crystal (left) and two examples of washed single crystals ran on a non-reducing SDS-PAGE (right) demonstrated that each single crystal contains a mixture of D- and M-forms of HBMP9. C, crystal structure of HBMP9 (left), colored according to the B-factors (spectrum, blue to white to red, from 20 to 100) with Cys-73 in two conformations shown in sticks. Electron density (2Fo − Fc map at 1σ) clearly shows two conformations of Cys-73 (right). D, HBMP9 was overlaid with the published BMP9 structures (Protein Data Bank codes 1ZKZ and 4FAO, all colored as described in C). In the semi-transparent schematic, type I receptor ALK1 is shown in yellow, and activin receptor type 2B is shown in cyan as in BMP9·ALK1·activin receptor type 2B complex (Protein Data Bank code 4FAO).
FIGURE 3.
FIGURE 3.
M-form BMP9 can bind to ALK1 and prodomain as the D-form. A, M- and D- forms of HBMP9 co-migrate on native PAGE, and both can form complexes with ALK1 ECD. B, M- and D-forms of BMP9 co-migrate, both as free form and as prodomain-bound form. In A and B, bands 1 to 6 from native PAGE were cut out, boiled in 2× SDS-loading buffer for 20 min before loading onto a non-reducing SDS-PAGE to confirm the identities. For SDS-PAGE in A, two parts of the same gel are shown.
FIGURE 4.
FIGURE 4.
Intermolecular disulfide bond is not required for BMP9 signaling activity. A, conditioned media (6 μl) from HEK-EBNA cells transfected with transfection reagent alone (lane 1), empty vector (lane 2), pro-BMP9 (lane 3), pro-HBMP9 (lane 4), or pro-HBMP9 C73S (lane 5) were fractionated on a 12% SDS-PAGE and probed with anti-BMP9 prodomain antibody. The intensities of bands were quantified using ImageJ, and the ratio of HBMP9 C73S to HBMP9 obtained. HBMP9 C73S concentration in the conditioned medium was normalized to HBMP9 using the above ratio. B, left: hPAECs were serum-restricted in EGM-2, 0.1% FBS overnight, and stimulated with HBMP9 or HBMP9 C73S (both 0.25–3 ng/ml) for 1 h. Cells were harvested, and total cell protein was immunoblotted with anti-pSmad1/5/8 antibody. Band intensities of pSmad1/5/8 blots were analyzed using ImageJ, corrected by ratios obtained from the α-tubulin blot, and normalized to a 0.25 ng/ml wild type sample. Data of the mean ± S.E. from three repeats are shown on the right. C, after quiescence overnight in EGM-2, 0.1% FBS, hPAECs were treated with 1 ng/ml of HBMP9 or HBMP9 C73S. Samples were harvested at 1, 2, 6, and 24 h, and immunoblotting was carried out as described in B. Band intensity of pSmad1/5/8 blots were analyzed using ImageJ, corrected by ratios obtained from the α-tubulin blot and normalized to a wild type 1-h treatment sample. Data of the mean ± S.E. from three repeats are shown on the right. NS, not significant.
FIGURE 5.
FIGURE 5.
BMP9 is regulated by redox potential. Purified HBMP9 (A), pBMP9 (B), or BMP6 (C) were incubated at room temperature overnight with PBS alone or redox buffer containing 0.1 mm GSSG and 0–20 mm GSH. Proteins were then run on a non-reducing SDS-PAGE and detected by Coomassie Blue. M* is the fully reduced form of BMP9.
FIGURE 6.
FIGURE 6.
M-form BMP9 is more susceptible to redox-dependent proteolysis. HBMP9 (A) or pBMP9 (B) was incubated in PBS or redox buffer containing 0.1/1 mm GSSG/GSH (0.1/1) or 0.1/4 mm GSSG/GSH (0.1/4) at room temperature overnight. The following day, an aliquot of each treatment was subjected to limited trypsin digestion at 37 °C, 3 h for HBMP9 or overnight for pBMP9. Reactions were stopped by addition of SDS-loading buffer and boiling at 100 °C for 10 min. Cleavage was monitored by fractionating samples on a 12% non-reducing SDS-PAGE and Coomassie Blue staining.
FIGURE 7.
FIGURE 7.
M-form BMP9 is preferentially cleaved in serum and model of BMP9 regulation by redox-dependent proteolysis. A, pBMP9 (0.3 μg) was reconstituted into 10 μl of serum (10 healthy human controls), and equal aliquots of samples were taken at 0 h (lane 1) and after overnight incubation at 37 °C (lane 2). Samples were then fractionated on a 12% SDS-PAGE under non-reducing conditions and detected by anti-BMP9 antibody (AF3209). The D- and M-forms of BMP9 at 0 h and overnight were quantified using ImageJ, and the % BMP9 remaining after overnight incubation was calculated and plotted using GraphPad Prism. Data are mean ± S.E. (n = 10). B, model of the regulation of BMP9 concentration by redox-dependent proteolysis.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. David L., Mallet C., Keramidas M., Lamandé N., Gasc J. M., Dupuis-Girod S., Plauchu H., Feige J. J., Bailly S. (2008) Bone morphogenetic protein-9 is a circulating vascular quiescence factor. Circ. Res. 102, 914–922 - PMC - PubMed
    1. David L., Mallet C., Mazerbourg S., Feige J. J., Bailly S. (2007) Identification of BMP9 and BMP10 as functional activators of the orphan activin receptor-like kinase 1 (ALK1) in endothelial cells. Blood 109, 1953–1961 - PubMed
    1. Johnson D. W., Berg J. N., Baldwin M. A., Gallione C. J., Marondel I., Yoon S. J., Stenzel T. T., Speer M., Pericak-Vance M. A., Diamond A., Guttmacher A. E., Jackson C. E., Attisano L., Kucherlapati R., Porteous M. E., Marchuk D. A. (1996) Mutations in the activin receptor-like kinase 1 gene in hereditary haemorrhagic telangiectasia type 2. Nat. Genet. 13, 189–195 - PubMed
    1. Liu F., Ventura F., Doody J., Massagué J. (1995) Human type II receptor for bone morphogenic proteins (BMPs): extension of the two-kinase receptor model to the BMPs. Mol. Cell Biol. 15, 3479–3486 - PMC - PubMed
    1. Oh S. P., Seki T., Goss K. A., Imamura T., Yi Y., Donahoe P. K., Li L., Miyazono K., ten Dijke P., Kim S., Li E. (2000) Activin receptor-like kinase 1 modulates transforming growth factor-β 1 signaling in the regulation of angiogenesis. Proc. Natl. Acad. Sci. U.S.A. 97, 2626–2631 - PMC - PubMed

Publication types

Associated data