The application of in situ AFM to the study of molecular and macromolecular crystallization
The crystallization of molecules from solution encompasses a number of key areas in science and technology, ranging from the purification and separation of industrial chemicals through to the arrangement of fragile biomacromolecules in ordered arrays suitable for structural analysis. However, as there are only a small number of techniques suitable for the study of such assemblies, many fundamental aspects governing the crystallization of molecules from solution are still poorly understood. In the studies presented here we have attempted to improve the understanding of this subject by investigating the crystallization of a series of molecules of both pharmaceutical and biological importance, using in situ Atomic Force Microscopy (AFM). A particular aim of the PhD was to develop the experimental protocols necessary to investigate macromolecular crystals known to exhibit poor diffraction properties, and subsequently to relate AFM data to these properties. The first study carried out concerned the crystallization and habit modification of a pharmaceutical excipient molecule, adipic acid. By using AFM we were able, for the first time, to directly observe the behavior of the dominant (100) face in both an air and liquid environment. A number of important observations were made including solute reorganization in air, etch pit formation and growth inhibition by the structurally related habit modifier, octanoic acid. We subsequently investigated various aspects of the crystallization of the model protein, lysozyme. The rate and mechanisms of growth of the (110) surface of the tetragonal crystal were observed using in situ AFM at a range of supersaturations, a study that uncovered a previously unreported mechanistic event. The (110) and (101) faces were then both investigated at higher resolution, revealing molecular resolution features corresponding directly to basic crystallographic data. The polymorphic characteristics displayed by many macromolecular crystals were then investigated in a short study concerning the growth and structure of the monoclinic form of the lysozyme crystal. The dominant (101) face of the crystal was investigated at both high and low protein/precipitant concentrations, allowing us to unambiguously distinguish between two crystalline forms of the same macromolecule. Finally, by utilizing the experimental techniques developed throughout the previous studies, we investigated a poorly diffracting crystal constructed from a protein found in Streptococcus pneumoniae, Response Regulator 02 receiver domain (RR02rec). By studying the surface of crystalline RR02rec with in situ AFM, we were able to uncover various features of the crystal lattice that may have contributed to the poor diffraction properties displayed by the crystal during previous X-ray studies. Besides revealing a range of new molecular scale details concerning the structure and growth of each of these crystal systems, these studies culminated in a successful attempt to relate direct microscopical observations of growth dynamics of a protein crystal system (RR02rec) to the limited results obtained from previous crystallographic studies performed on this protein. As an approach this offers considerable promise in identifying problems with certain crystals and, in conjunction with future advances in AFM technology, may offer information that could lead to the acceleration and enhancement of X-ray diffraction analyses.