Sample preparation for routine and advanced structural biology, including serial data collection, microED, and cryoEM

Serial data collection and microED techniques usually require “slurries” of tiny, well-ordered crystals.  Neutron diffraction requires very large single crystals.  Microseeding effectively generates such samples since the seed stock can be concentrated or diluted as necessary.  During the 16 years since the random microseed matrix-screening (rMMS) method was published, understanding of the theoretical advantages of the method has increased, and several practical variations of the technique have emerged.  Moreover, seeding can be carried out in a microbatch-under-oil setup, which has two important advantages: (1) easily interpret phase diagrams that can be constructed in a few minutes; (2) batch experiments are easy to scale up.  We present case studies using these approaches to increase control and crystal quality for routine and advanced data collection.

Protein structure determination by cryoEM requires expensive equipment that has low throughput.  It is, therefore, wasteful to examine samples that can be shown in advance to be aggregated since such samples are unlikely to be suitable.  We used a high-throughput screening approach with dynamic light scattering to explore 96 chemical conditions, with as little as 10 µL of protein solution in total, to identify conditions with reduced aggregation.