"The three-dimensional Patterson function of ribonuclease II"

Date:
1956
Reference:
PP/CRI/I/1/9
Catalogue details

Licence: In copyright

Credit: "The three-dimensional Patterson function of ribonuclease II". Source: Wellcome Collection.

    1/5
    Reprinted from Acta Crystallographica, Vol. 9, Part 2, February 1956 The Three-Dimensional Patterson Function of Ribonuclease II* B y B eatrice S. MAGDOirf, F. H. C. C rick J and V. L tjzzati§ The Protein Structure Project, Polytechnic Institute, Brooklyn 1, N.T., U.S.A. * (Received 3 December 1954 and, in revised form 6 May 1955) All the intensities of the monoclinic form (P2 X ) of the protein ribonuclease have been measured out to 1 /d = 0-445 Â -1 , and a three-dimensional Patterson function has been computed from them. The Patterson shows that it is highly unlikely that the structure is based on polypeptide 'rods' all parallel to the o axis. However, the data show some asymmetry in the general direction indicated by the infrared dichroism. A concentration of vector fluctuations on and near the mirror planes of the Patterson has been explained on statistical grounds as a space-group effect. Introduction As a preliminary attack on the problem of the struc ture of the protein ribonuclease it was considered desirable to examine a three-dimensional Patterson function in order to obtain some idea of the simplicity, or otherwise, of the structure. For example, it has been postulated by Carlisle & Scouloudi (1951) that ribo nuclease consists of parallel a-ribbons, running in the direction of the c axis in ribonuclease II. Such a simple structure should show unmistakable signs in the three- dimensional Patterson. For such a Patterson function the smaller the num ber of molecules in the unit cell the better ; in the ideal case the space group would be PI with one molecule in the asymmetric unit. Thus the only reasonable choice among the various crystal modifications avail able (King, Magdoff, Adelman & Harker, 1956) is ribonuclease II (space group P2 1 ; cell dimensions a = 30-28, b = 38-39, c = 53-16 Á, ß = 105-83°; one molecule per asymmetric unit) and this also has the advantage of being the form most easily obtainable. It was decided, therefore, to collect the X-ray inten sities diffracted by ribonuclease II out to 20 = 40° using Cu Koc radiation and to compute a three- dimensional Patterson function from them. Carlisle and his co-workers have also calculated a similar Patterson function; at the time of writing it has not been published, though sections from it have been shown at various meetings. The authors inform us that it is now in the course of preparation for publica tion. Experimental technique The methods for growing ribonuclease crystals have been described elsewhere (King et al., 1956). Those * Contribution No. 8 from The Protein Structure Project, t Present address: Boyce Thompson Institute, Yonkers, N.Y., U.S.A. % Present address: Medical Research Council Unit, Caven dish Laboratory, Cambridge, England. § Present address: Centre de Recherches sur les Macro - molécules, Strasbourg, Prance. chosen were grown from 50% tertiary butyl alcohol. The intensities were collected from only two crystals. The first was used for 20 = 0° to 30°, and the second for 20 = 30° to 40°, with an occasional measurement between 20 = 0° and 20 = 30° to enable both sets to be put on the same scale. The experimental data were collected by two of us (B. S. M. and V. L.) with the help of all our colleagues. The intensities were measured on a General Electric XRD-3 Spectrogoniometer modified for single crystal work by the addition of an Eulerian cradle for mount ing the specimen (Furnas & Harker, 1955). This instru ment has the advantage that the intensities can be measured accurately and without giving the crystal prolonged X-ray exposures, which are known to damage the crystal (Magdoff, 1953). The total ex posures in this experiment were 20 and 25 hr. respec tively. The cell dimensions of the crystal were first measured under fine conditions, i.e. with narrow slits, etc. From these dimensions the angular coordinates of all the hkl reflections, with their values of 20, were calcu lated on IBM machines (Magdoff, unpublished). They were then in such a form that they could be used for setting the instrument to bring any required reci procal-lattice point into the reflecting position. This was done under coarse conditions (i.e. with wide slits) and for a crystal less than about mm. in size. The geometry of the apparatus is such that with the crystal stationary the counting rate is proportional to the integrated intensity. Counts were made for a fixed time of 10 sec., and the background count, from an adjacent non-lattice point in reciprocal space, was subtracted. An approximate correction, described by Magdoff & Crick (1955a), was applied for the different absorp tion of the capillary in different directions (Furnas, unpublished). This was obtained by measuring the intensity of one chosen reflection (actually 0,10,0) with various aspects of the capillary to the X-ray beam. For the first crystal the relative correction