Macromolecular Crystallography for Synthetic Abiological Molecules: Combining xMDFF and PHENIX for Structure Determination of Cyanostar Macrocycles

Crystal
structure determination has long provided insight into
structure and bonding of small molecules. When those same small molecules
are designed to come together in multimolecular assemblies, such as
in coordination cages, supramolecular architectures and organic-based
frameworks, their crystallographic characteristics closely resemble
biological macromolecules. This resemblance suggests that biomacromolecular
refinement approaches be used for structure determination of abiological
molecular complexes that arise in an aggregate state. Following this
suggestion we investigated the crystal structure of a pentagonal macrocycle,
cyanostar, by means of biological structure analysis methods and compared
results to traditional small molecule methods. Cyanostar presents
difficulties seen in supramolecular crystallography including whole
molecule disorder and highly flexible solvent molecules sitting in
macrocyclic and intermolecule void spaces. We used the force-field
assisted refinement method, molecular dynamics flexible fitting algorithm
for X-ray crystallography (xMDFF), along with tools from the macromolecular
structure determination suite PHENIX. We found that a standard implementation
of PHENIX, namely one without xMDFF, either fails to produce a solution
by molecular replacement alone or produces an inaccurate structure
when using generic geometry restraints, even at a very high diffraction
data resolution of 0.84 Å. The problems disappear when taking
advantage of xMDFF, which applies an optimized force field to realign
molecular models during phasing by providing accurate restraints.
The structure determination for this model system shows excellent
agreement with the small-molecule methods. Therefore, the joint xMDFF-PHENIX
refinement protocol provides a new strategy that uses macromolecule
methods for structure determination of small molecules and their assemblies.