Crystal X-ray quality is usually evaluated by looking at data quality parameters such as (relative) Wilson B-factor, resolution, R-factors, signal-to-noise ratio, and others. As these parameters are correlated, most studies focus on one or two of them. As part of a study of the effects of microgravity on X-ray quality, full data sets of 35 ferritin crystals (17 PromISS-4 "space" crystals and 18 from the ground control) were collected. Sixty-three parameters commonly used as indicative of X-ray data quality taken from the output of the processing, scaling, and merging software packages were analyzed. This highly dimensional "quality parameter dataset" was reduced using a principal component analysis. About 78% of the variability in the data set could be explained with the first four principal components. A score-plot in this four-dimensional space clearly showed two tendencies, one for the crystals grown in space and one for the ground crystals. The differences between the two groups are observed irrespective of the software package. They can be attributed to the first principal component and reflect the superior quality of the space crystals.
Experimental SectionProtein. Type I horse spleen ferritin was obtained in saline solution at 85 mg/mL (cat. no. F-4503; lot: 084K7001) from Sigma-Aldrich (Steinheim, Germany). To remove higher order oligomers, 16 a gel filtration was performed using a GE Healthcare Superdex 75 HR 10/ 30 column (Freiburg, Germany) with a 0.2 M sodium acetate pH 4.5 mobile phase. Monomeric fractions were pooled and concentrated to 30 mg/mL. The protein solution was dialyzed against a buffer of 100 mM NaAc pH 5.0 with 200 mM ammonium sulfate and concentrated to 18 mg/mL. Protein concentrations were determined using the Bradford assay kit from Sigma-Aldrich.Crystallization. The crystallization experiments were performed in specially developed reactors (Figure 1a) which were designed to allow a "full geometry counter-diffusion" setup. 17 Reactors have a precipitant volume that is initially separated from the protein volume. When the release mechanism is activated, precipitant and protein are free to diffuse toward each other.Six reactors were launched on Progress 20P on December 21, 2005, installed in the PromISS instrument on board of the ISS, activated for 15 days, and returned to Earth on board of Soyuz 11S on April 8, 2006. The PromISS instrument was operated in the Microgravity Science Glovebox in the US Destiny Laboratory by Expedition 12commander Wiliam McArthur, who was also responsible for the reactor activation. A control experiment using the same conditions was performed on Earth at the same time at the ULB Microgravity Research Center in Brussels (Belgium). Care was taken to keep the same conditions (concentrations, temperature) for the space and ground