O
-phospho-
l
-serine (Pser) and its Ca
salt, Ca[
O
-phospho-
l
-serine]·H
2
O (CaPser), play important roles for bone mineralization and
were recently also proposed to account for the markedly improved bone-adhesive
properties of Pser-doped calcium phosphate-based cements for biomedical
implants. However, the hitherto few proposed structural models of
Pser and CaPser were obtained by X-ray diffraction, thereby leaving
the proton positions poorly defined. Herein, we refine the Pser and
CaPser structures by density functional theory (DFT) calculations
and contrast them with direct interatomic-distance constraints from
two-dimensional (2D) nuclear magnetic resonance (NMR) correlation
experimentation at fast magic-angle spinning (MAS), encompassing double-quantum–single-quantum
(2Q–1Q)
1
H NMR along with heteronuclear
13
C{
1
H} and
31
P{
1
H} correlation NMR
experiments. The Pser and CaPser structures before and after refinements
by DFT were validated against sets of NMR-derived effective
1
H–
1
H,
1
H–
31
P, and
1
H–
13
C distances, which confirmed the improved
accuracy of the refined structures. Each distance set was derived
from one sole 2D NMR experiment applied to a powder without isotopic
enrichment. The distances were extracted without invoking numerical
spin-dynamics simulations or approximate phenomenological models.
We highlight the advantages and limitations of the new distance-extraction
procedure. Isotropic
1
H,
13
C, and
31
P chemical shifts obtained by DFT calculations using the gauge including
projector augmented wave (GIPAW) method agreed very well with the
experimental results. We discuss the isotropic and anisotropic
13
C and
31
P chemical-shift parameters in relation
to the previous literature, where most data on CaPser are reported
herein for the first time.