Structure and binding specificities of the VWC domains of Chordin and Chordin-like 2
Zusammenfassung der Projektergebnisse
The focus of the project was the analysis of interaction with and inhibition of BMPs by VWC domains of Chordin and Chordin-like 2. Our first attempt was to determine the structures of different VWC domains of both modulator proteins either in their unbound conformation or bound to BMP-2. As the VWC1 domain of Chordin alone exhibited poor solubility properties, only structure analysis bound to BMP-2 seemed possible. However, crystals of this complex were only obtained under very unusual conditions with very low ionic strength making reproducibility and optimization very difficult. Thus the VWC3 domain of CHL2 became more the focus. Preparation and crystallization of the VWC3:BMP-2 complex unfortunately did not yield crystals diffracting to high resolution despite huge efforts, e.g. use of different constructs, different expression systems and various purification strategies. One possible explanation for the failure might be the lower affinity of the VWC3 domain for BMP-2 compared to the VWC1 domain of CV2. Particularly the much faster dissociation rate of CHL2 VWC3 from the complex with BMP-2 causes instability of the protein-protein complex. The low solubility of BMP-2 under these crystallization conditions then leads to “removal” of BMP-2, the resulting heterogeneity then yields only crystals of poor diffractive quality. To overcome this problem the full-length CHL2, which has a greater affinity for BMP-2, was used for preparing the CHL2:BMP-2 complex, which also yielded single crystals. Gelfiltration shows indeed a higher complex stability, however, the glycosylation heterogeneity and likely flexible termini/loops might be the reason that these crystals did not diffract Xrays. NMR structure determination of CHL2 VWC3 was thus the only available route to obtain structure data on this interesting VWC domain, which is the only known VWC domain that binds solely to the type I receptor epitope (wrist epitope) of BMPs. Different constructs and schemes were tested as yields for minimal medium were far too low to allow isotope labeling for NMR structure determina-tion. Optimization of the expression now provided a VWC3 construct, which is currently used to acquire NMR data. The unusual binding of CV2 VWC1 to BMP-2 raised the question whether and how much of the binding site is preformed, as the epitopes are very small for generation of such a high binding affinity in the low nanomolar range. The high-resolution NMR structure of CV2 VWC1 in its unbound form now shows that the Clip segment, which according to mutagenesis is crucial for high-affinity binding is indeed preformed to a large extent thereby explaining the highly cooperative binding mode. Additional functional mutants are currently analyzed to support our folding hypothesis. A large effort in the project was then put onto the functional and structural analysis of Twisted gastrulation (Tsg) as it seems the common factor interacting with and binding directly to all BMP modulators of the Chordin family. It is the likely molecular switch of all Chordin-like modulator’s functionality between the pro- and anti-BMP activities, which is a unique feature of this modulator family. We succeeded in establishing a pro- and eukaryotic expression system for Tsg enabling us to study its structure/function relationship. Currently crystallization trials for Tsg alone as well as in complex bound to BMP-2 have been setup. As Tsg also strongly enhances the affinity of CHL2 (full-length CHL2 and possibly also for CHL2 VWC3) a ternary complex of Tsg and CHL2 (or CHL2 VWC3) bound to BMP-2 might be a better target for obtaining structural data on the BMP-CHL2 interaction. To get an insight into the functional interaction between Tsg and BMP-2 we have started a mutagenesis study. As no structural data or model is available for Tsg an efficient expression/analysis system allowing for “high-throughput” production of Tsg variants was developed and mutants of Tsg are currently analyzed. The latter results will also be incorporated into a new grant application.
Projektbezogene Publikationen (Auswahl)
- Crystal structure analysis reveals how the Chordin family member crossveinless 2 blocks BMP-2 receptor binding. Developmental cell 14 (2008) 739-750
J.L. Zhang, L.Y. Qiu, A. Kotzsch, S. Weidauer, L. Patterson, M. Hammerschmidt, W. Sebald, T.D. Mueller
- Crystallization and preliminary X- ray analysis of the complex of the first von Willebrand type C domain bound to bone morphogenetic protein 2. Acta crystallographica Section F, Structural biology and crystallization communications 64 (2008) 307-312
L.Y. Qiu, J.L. Zhang, A. Kotzsch, W. Sebald, T.D. Mueller
- Intricacies of BMP receptor assembly. Cytokine Growth Factor Rev 20 (2009) 367-377
J. Nickel, W. Sebald, J.C. Groppe, T.D. Mueller
- Glycosylation of Twisted Gastrulation is Required for BMP Binding and Activity during Craniofacial Development. Frontiers in physiology 2 (2011) 59
C.J. Billington, Jr., J.E. Fiebig, C.L. Forsman, L. Pham, N. Burbach, M. Sun, T. Jaskoll, K. Mansky, R. Gopalakrishnan, M.B. O'Connor, T.D. Mueller, A. Petryk
- Promiscuity and specificity in BMP receptor activation. FEBS letters 586 (2012) 1846-1859
T.D. Mueller, J. Nickel
