While X-ray crystallography provides the detailed molecular structure and the internal molecular packing arrangements of protein crystals, their surface packing must be determined by other means.  The surface packing influences the growth mechanism of the crystal and hence its importance.  We have pioneered the use of high resolution AFM to precisely determine the surface packing arrangements of protein crystals and their relationship to the internal packing ones determined by X-ray crystallography.  This work is done in collaboration with Dr. John H. Konnert at the Naval Research Laboratory, Washington, DC.

The packing arrangements are determined by first constructing a theoretical AFM image for a given packing arrangement on a crystal face.  If the actual AFM images matches this it confirms the arrangement, if not another packing arrangement must be tried. Periodic Bond Chain theory can be employed to obtain initial predictions for the packing arrangements.  This approach is illustrated below.
 

Figure 1: Image obtained from a high resolution AFM scan of the (110) face of tetragonal lysozyme crystals. The original experimental image was averaged over the repeating units to obtain the image shown.  The c axis runs vertically in this image. Individual molecules forming five 43 molecular helices are clearly visible.

	Figure 2: Theoretical image constructed by convoluting the tip shape with the calculated surface morphology for a given surface packing arrangement.  In this case the packing arrangement was for complete 43 helices on the (110) face of tetragonal lysozyme crystals.  The correlation between this image and the experimental image in figure 1 is 62%.
	Figure 3: Theoretical image constructed by convoluting the tip shape with the calculated surface morphology for a given surface packing arrangement.  In this case the packing arrangement was for complete 21 helices on the (110) face of tetragonal lysozyme crystals.  The correlation between this image and the experimental image in figure 1 is 25%.

Figures 2 & 3 correspond to the two possible packing arrangements on the (110) face.  The poor agreement between figure 1 & 3 suggests that the packing on this face correspond to that shown in figure 2.  However, the correspondence between figures 1 & 2 is not perfect, although it is adequate to select the correct packing arrangement.  This lack of perfect agreement between figures 1 & 2 is because of surface reconstruction on the crystal faces.  For more details see H. Li, M.A. Perozzo, J.H. Konnert, A. Nadarajah & M.L. Pusey, Acta Crystallographica, D55, 1023-1035 (1999).

Return to the homepages of:
 

AFM in Protein Crystallization	Macromolecular Crystallization Laboratory	Chemical & Environmental Engineering	UT College of Engineering	The University of Toledo