Fig. 7: The C1 dimer interface provides a scaffold for the association of the ALK2/BMPR2 kinase tetramer via the N dimer interface.

a Two identical copies of the N dimer were superposed on the trajectory of an MD simulation of the C-lobe/C-lobe 1 dimer where we observed slight variations of the C-C interface. A pose that put both N dimer copies at a close distance was extracted and was subjected to more MD simulations (50 µs aggregate simulation time described in Supplementary Table 5) to generate the tetramer model where tetramer interfaces converged to a consensus pose. b The interaction between the GS domain of ALK2 and the active site of the BMPR2 kinase domain is modeled in both ALK2 monomers in the tetramer using a crystal structure of PKA (PDB: 1ATP). This interaction occurs across the N dimer interfaces and positions the GS domain for phosphorylation. The inset on the right zooms in on the interaction between the GS domain and the BMPR2 kinase active site. The phosphorylatable serine residues in the GS domain are shown in dot representation. c Zoomed in view of the N dimer interface between the ALK2 and BMPR2 kinases in the ALK2/BMPR2 tetramer model highlighting key residues within helices C of each kinase that contributes to the interface. d Effect of the N dimer interface mutants on the BMP4-mediated activation of SMAD-dependent transcription by the BMPR2/ALK2 receptor complex, measured as described in Fig. 5c. Experiments were repeated at least three times and in each experiment each condition was quadruplicated with four independent biological samples. Means and standard deviations of the means (SDs) are indicated. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001, ns, not significant by ANOVA with two-sided post hoc Tukey’s test. Western blot analysis of expression levels of ALK2 and BMPR2 receptors carrying the N dimer interface mutations expressed transiently in HEK293 cells upon cell lysis is shown in the bottom panel.