Existence of twisting in dislocation-free protein single crystals

The growth of high-quality protein crystals is a prerequisite for the structure analysis of proteins by X-ray diffraction. However, dislocation-free perfect crystals such as silicon and diamond have been so far limited to only two kinds of protein crystals, such as glucose isomerase and ferritin crystals. It is expected that many other high-quality or dislocation-free protein crystals still exhibit some imperfection. The clarification of the cause of imperfection is essential for the improvement of crystallinity. Here, we explore twisting as a cause of the imperfection in high-quality protein crystals of hen egg-white lysozyme crystals with polymorphisms (different crystal forms) by digital X-ray topography with synchrotron radiation. The magnitude of the observed twisting is 10−6 to 10−5°/μm which is more than two orders smaller than 10−3 to 104°/μm in other twisted crystals owing to technique limitations with optical and electron microscopy. Twisting is clearly observed in small crystals or in the initial stage of crystal growth. It is uniformly relaxed with crystal growth and becomes smaller in larger crystals. Twisting is one of main residual defects in high-quality crystals and determines the crystal perfection. Furthermore, it is presumed that the handedness of twisting can be ascribed to the anisotropic interaction of chiral protein molecules associated with asymmetric units in the crystal forms. This mechanism of twisting may correspond to the geometric frustration proposed as a primary mechanism of twisting in molecular crystals. Our finding provides insights for the understanding of growth mechanism and the growth control of high-quality crystals.