Journal of Crystal Growth, Vol.232, No.1-4, 520-535, 2001
Mosaic spread analysis of Canadian advanced protein crystallization experiment on the Russian space station, Mir
Protein crystallization experiments were performed on the Russian space station, Mir, using liquid-liquid interface diffusion. The technique was activated in orbit by the sliding together of two half-wells containing protein and precipitant fluids, respectively. Imperfections in protein crystals were analyzed from rocking curve measurements of the diffracted intensities using synchrotron radiation. Data were collected on microgravity and earth-grown crystals, and 10 different protein pairs were compared. To avoid bias, a double-blind protocol was used throughout the data analysis. Rocking curves for individual reflections were analyzed in terms of crystal domains, each fitted by a three-dimensional Gaussian profile. The results of Gaussian analysis were consistent with domain segregation corresponding to spatially different regions of the protein crystal exhibiting distinct mosaic spreads. When crystals were grown in microgravity the domain mosaic spreads were consistent with five of 10 different proteins exhibiting fewer imperfections, three other proteins showed no significant difference while a remaining two proteins displayed a greater number of apparent imperfections. Ground (earth-grown) controls were also conducted on protein samples flown to assess protein stability as a function of solution storage time prior to protein crystal growth (PCG) activation in microgravity. Protein samples were stored in ground controls at concentrations used to initiate crystallization, and aliquots were analyzed after a 30-day period by dynamic light scattering. Polydispersity estimates indicated that prolonged storage induced heterogeneity in all protein samples. Stable aggregates were present, and they were concentration independent, as shown by resistance to protein sample dilution. A PCG growth model is proposed that takes into account large scale aggregation or self-impurities present during crystal growth and predicts domain segregation. Trapping or rejection of self-impurities using this model can qualitatively explain differences in domain mosaic spreads observed as a function of gravitational environment.