Infra-red dichroism and chain orientation in crystalline ribonuclease By A. Elliott Research Laboratory, Courtaulds Ltd, Maidenhead, Berks Communicated by A. H. Wilson, F.R.S.—Received July 1951— ( 20 Revised 8 November 1951) Dichroism has been observed in the infra-red spectrum of a single crystal of ribonuclease. The dichroism suggests that the crystal contains folded polypeptide chains whose direction is mainly along or near that of the c axis, in agreement with conclusions arrived at by Carlisle & Scouloudi from consideration of the X-ray diffraction pattern.
Infra-red dichroism in crystalline ribonuclease 491 Rubber Manufacturers, it appears that this point was not sufficiently brought out in the paper referred to. Although it was stated that double refraction would in certain cases lower the dichroism, the way in which this effect comes about was not discussed. Dr Mann raised the question of the effect of the crystal in splitting the incident beam into two components. Since in ribonuclease the birefringence cannot be neglected, this effect is considered in the next section, where a method is described which avoids this complication.
Infra-red dichroismin crystalline ribonuclease 493 covered a range of angles 6 from +70 to —38° to the c axis. A coiisiderable number of runs was made for each crystal position, with the vector alternately parallel and perpendicular to the b axis, in order to eliminate the effect of any small shift of the crystal in its mount. Typical spectra in the overtone region (the only region in which wet protein crystals can be examined with the present technique) are shown in figure 1. The spectra are bounded on the low and high frequency sides, respectively, by absorption of liquid paraffin and of water in the crystal.
494 A. Elliott (Carlisle & Scouloudi 1951). Because of the difference in refractive index of the crystal and this immersion liquid, it is only possible to make observations with the radiation not far from normal to the large ab face. Three such observations are included in figure 2. A much greater range of angles is obtained when the crystal radiation Figure 2. Dichroic ratios from figure 1. a, N—H combination band; b, C =0 combination band. The inset shows the ac plane of the crystal, with the radiation making a positive angle 6 with the c axis: •, from partially dried crystal; x , from wet crystal.
Infra-red dichroism in crystalline ribonuclease 495 less), the errors in the density contributed by C=0 and N—H bands is con siderable, and this no doubt accounts for the spread of the points in figure 2. The main results to be deduced are: (a) The N—H combination band shows no dichroism when the radiation is in the ac plane, making an angle of about — 30° to the c axis, and the dichroism rises to a maximum in a direction at right angles to this, but also in the ac plane.
496 A. Elliott Observations have shown that the dichroism is not greatly different whether the radiation is incident normally on the ab face, or at 45° on either side of the normal, and in the ac plane. These observations are not of high accuracy, but show, wave number (cm-1) Figure 3. a, dried ribonuclease crystal, radiation incident normally on ab face. Full line, U-vector perpendicular to b axis; broken line, F-vector parallel to b axis; 6, spectrum of gliadin; c, spectrum of solid caproamide film.
498 A. Elliott Discussion It is clear from the preceding section that the amide groups in ribonuclease prevent quantitative deductions being made from the observations on dichroism. In earlier work on haemoglobin crystals (Elliott & Ambrose 1950) it was shown that the dichroism of the N—H combination band was compatible with an arrangement of a-keratin chains along the a axis, suggested by Perutz from observations of Patterson vectors. The dichroism of the N—H combination band in ribonuclease suggests that similar chains are present, their direction being at least approximately along the c axis. If the effect of amide groups is not taken into account, the results would suggest a chain direction in the ac plane, making about 30° to the c axis, since this is the direction of minimum dichroism. However, the amide groups must certainly affect the observed dichroism, and it could well be that the actual chain direction, as Carlisle has suggested, is along the c axis.