Spectacular improvement of X-ray diffraction through fast desiccation of protein crystals

Succeeding in getting a protein to crystallize is not always the final hurdle in the determination of its three-dimensional structure. A relatively frequent and particularly vexing situation is the production of macroscopically well formed crystals that exhibit no suitable diffraction pattern. In this paper, three independent cases (i.e. proteins and crystallization conditions) are reported of spectacular diffraction-pattern improvement through a simple crystal-handling procedure that was discovered serendipitously. The procedure basically consists of removing a non-diffracting frozen crystal from the X-ray beam, plunging it into a soaking solution made of the original crystallization solution supplemented with a traditional cryoprotectant and then letting it dry in the evaporating sitting drop for some time (15 min to several hours). The treated crystals are then remounted and exhibit a huge improvement in their diffraction intensity and resolution. In all three cases presented here, the crystal quality shifted from unusable to perfectly suitable for structure determination. In addition to being a 'last resort' procedure for experimentalists struggling with non-diffracting crystals, this puzzling effect constitutes one more challenging problem for theoretical protein crystallographers.