The structure of the (110) face of tetragonal lysozyme crystals can be predicted employing Periodic Bond Chain theory.  As shown in figure 1, this will consist of growth layers bimolecular in height.  The growth units will correspond to the 4₃ helix, with the minimum size being a tetramer.  The goal of this project was to use the AFM linescans technique we had developed to measure the size of the growth units being added to this face in the two growth directions.

Figure 1: The predicted structure of the (110) face and its growth steps and growth units.  The molecular packing on the surface corresponds to complete 43 helices as shown, with growth steps at least bimolecular in height and corresponding to this helix.  In order to fit the bimolecular step and retain the 4-fold screw symmetry of the crystal, the minimum growth unit for this step must be a tetramer corresponding to a single turn of the 43 helix as shown here.

The growth step heights on the (110) face were known to be at least bimolecular (5.6 nm) from earlier studies.  The results of many AFM linescans done in the two growth directions are shown in figures 2 and 3.  In figure 2 the results show that the growth proceeds by complete 4₃ helices.  The growth units are of even-molecular width, mostly bimolecular or four-molecular.  Growth units of unimolecular or other odd-molecular dimensions were not observed.  In figure 3 the growth unit dimensions mostly correspond to one, two and three turns of the 4₃ helix.  These results suggest that growth of the (110) face proceeds by the addition of many growth units, all corresponding to the 4₃ helix.  The minimum size is a tetramer corresponding to a single turn of this helix.
 

Figure 2: Distribution of growth unit dimensions obtained from 100 linescan experiments on the (110) face.  The scans here were done in the direction as shown in the figure on top left.  The measured dimension was the width of the growth unit as shown on the figure in top right.  The dimensions corresponding to the width of single (5.6 nm) and two (11.2 nm) 43 helices are shown for comparison purposes (see figure 1).

 

Figure 3: Distribution of growth unit dimensions obtained from 65 linescan experiments done in the direction shown in the figure on top left.  The measured dimension was the height of the growth unit, corresponding to the number of turns of the 43 helix, as shown on the figure in top right.  The dimensions corresponding to single (3.8 nm), two (7.6 nm) and three (11.4 nm) turns of the 43 helix are shown for comparison purposes (see figure 1).

The results presented above suggest the advantages of molecular packing analyses and AFM measurements to determine the molecular growth mechanisms of protein crystals.  The linescan technique developed here demonstrates the use of AFM to follow molecular processes in real time.  It also shows that resolutions of one nanometer or less can be achieved with this technique.  For more details see H. Li, A. Nadarajah & M.L. Pusey, Acta Crystallographica, D55, 1036-1045 (1999).

Return to the homepages of:
 

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