Research Article: A generic protocol for protein crystal dehydration using the HC1b humidity controller

Date Published: May 01, 2016

Publisher: International Union of Crystallography

Author(s): Carina M. C. Lobley, James Sandy, Juan Sanchez-Weatherby, Marco Mazzorana, Tobias Krojer, Radosław P. Nowak, Thomas L. Sorensen.


A generic protocol for investigating crystal dehydration is presented and tested with a set of protein crystal systems using the HC1b high-precision crystal humidifier/dehumidifier.

Partial Text

The process of obtaining X-ray diffraction-quality protein crystals can be fraught with difficulties. Large quantities of pure protein must be obtained and crystallization conditions must be determined through an empirical process (see McPherson, 1999 ▸ and references therein). Typically, any crystal obtained will then be removed from the mother liquor in which it was grown, exposed to a cryoprotectant and flash-cooled in liquid nitrogen (Garman & Schneider, 1997 ▸). These challenges navigated, the protein crystal is exposed to X-rays, usually using the intense X-ray beams available at a synchrotron source. Ideally, this process will result in the collection of useful diffraction data and the protein structure can be solved. More often the crystals will diffract X-rays weakly, with high mosaicity or anisotropy, or worse, not at all. In these cases it is not possible to collect a suitable X-ray diffraction data set.

Nine crystal systems have been investigated using a generic dehydration protocol to assess the changes induced by crystal dehydration. The initial crystal characterization, which is the first step in the generic protocol, was an efficient analysis of the crystal. Data were collected at 295 K, with crystals cooled to 100 K using standard cryoprotection, with naked crystals cooled to 100 K and with crystals following a one-step dehydration to a RH of 91% and cooled to 100 K. The data presented above (§3.2) provide an evidence base for testing crystals in a number of ways prior to discarding them in favour of finding a new crystal type. Firstly, they support the collection of 295 K data from all crystals. In seven of the eight indexable examples presented, the mosaicity of the crystals was lower at 295 K than after manipulation and cryocooling. Secondly, they provide a test set of samples for which the flash-cooling of a naked crystal without dehydration was sufficient to prevent the formation of crystalline ice in the sample and was therefore a suitable alternative to chemical cryoprotection, as observed in Pellegrini et al. (2011 ▸). Thirdly, we demonstrate that cryocooling the crystal in the presence of an appropriate cryoprotectant causes a thermal contraction of the crystal in seven of the eight indexable systems. Additionally, in every case presented flash-cooling the naked crystal in the absence of dehydration caused a contraction of the unit cell.