By combining results from atomistic MD simulations, DFT calculations and NMR experiments, we will study the adsorption of phosphoserine molecules at the surface of calcium hydroxyapatite (HA), which is the mother structure of bone mineral. Phosphoserine (Pser) is an ester of serine and phosphoric acid, and is abundant in many proteins believed to initiate and regulate bone growth . Our studies of Pser/inorganic phosphate interactions intimately connect two research areas—biomaterials and biomineralization:
1. Pser-bearing calcium phosphate cements (CPCs) for the next-generation bone implants: After a mixture of calcium phosphates is injected in a bone/tooth void, it hardens to a HA-like cement, which strongly interfaces with the surrounding bone tissue. Including Pser in the cement improves its mechanical properties and also stimulates further bone growth around the implant. However, the reasons behind these favorable properties are largely unknown. As one part of the SSF-funded project “Active Calcium Phosphate Cements”, we aim at a better understanding of the Pser/CPC interface by its selective experimental probing and by MD/DFT modeling.
2. Biomineralization: A long-standing and highly debated question without a clear consensus concerns the mechanism behind bone formation and if/how this process is initiated and controlled by non-collagenous proteins (NCPs). NCPs involved in bone mineralization in vertebrates share the feature of a high abundance of phosphorylated serine residues , thereby giving the proteins a high affinity for binding at inorganic phosphate surfaces, such as bone mineral, HA, and CPCs.
The modeling will mainly involve atomistic MD simulations incorporating meta-dynamics of a system with HA, water, and Pser components, where we will explicitly account for the dependence of the HA surface speciation on the pH of the solution ; this is rarely considered in literature reports on similar calculations on related systems. The computations will explore the relative affinities/binding energies of Pser molecules at different crystallographic surfaces, locate the preferred Pser orientation at the surface, and include the dependence of these parameters/properties on the pH and Pser concentration. The MD simulations will be benchmarked against Car-Parrinello DFT calculations.
The modeled results will be compared with data from advanced solid-state NMR experiments on CPC-Pser composites and model systems of nano-crystalline HA grown from solutions comprising Pser. Central to our research work is a synergistic combination of in-group-derived atomic-scale information from MD simulations, DFT calculations, and solid-state NMR experiments (e.g., see [3, 4]), as utilized within our previous SNIC project .
 A George, A Veis, Chem. Rev. 108, p4670 (2008)
 T-J Lin, H Heinz, J. Phys. Chem C, 120, p4975 (2016)
 A Jaworski, T Charpentier, B Stevensson, M Edén, J. Phys. Chem C, 121, p18815 (2017)
 Y Yu, B Stevensson, M Edén, J. Phys. Chem B, 121, p9737 (2017)