Conformational transitions in the synthetic polynucleotide poly[d(G-C)]·poly[d(G-C)] double-helix

J. Mol. Biol. (1983) 168, 897-901

LETTERS TO THE EDITOR

Conformational Transitions in the Synthetic Polynucleotide Poly[d(G-C)] 9poly[d(G-C)] Double-helix Conditions are described for observing by X-ray fibre diffraction the A , B and S conformations of the poly[d(G-C)] 9poly[d(G-C)] double-helix and also a new form designated as B". For fibres with an appropriate ionic content, transitions between these conformations can be induced by varying the relative humidity of the fibre environment. With increasing relative humidity the transitions B" ---* A --* S --* B occur. However, reducing the relative humidity does not result in a simple reversal of these transitions. If the relative humidity is reduced rapidly, a B -* A transition is observed followed by an A -~ B" transition, but if it is reduced slowly, the transition is from B to S. Once the S form has been assumed, further reduction in the relative humidity does not result in a transition to the A form. The S form emerges as a particularly stable form of the poly[d(G-C)] 9poly[d(G-C)] double-helix. From the point of view of its relationship to the classical A and B forms, the S form of poly[d(G-C)]" poly[d(G-C)] is shown to exhibit similarities to the D form of poly[d(A-T)], poly[d(A-T)]. Single crystal studies by Wang et al. (1979) of the oligonucleotide dCpG(pCpG)2 and subsequent studies by Drew et al. (1980) of d(CpGpCpG) allowed the direct visualization at atomic resolution of left-handed conformations of the DNA double-helix, which were designated Z and used to account for observations on the poly[d(G-C)], poly[d(G-C)] duplex in solution (Pohl & Jovin, 1972). A r n o t t et al. (1980) reported t h a t a structure designated as S and very similar to t h a t observed in crystals of dCpG(pCpG)2 could, if extended to form a two-stranded polynucleotide, account for an X - r a y fibre diffraction p a t t e r n observed from Na poly[d(G-C)].poly[d(G-C)]. T h e y reported t h a t this form was observed after prolonged annealing of fibres of Na poly[d(G-C)].poly[d(G-C)] t h a t had previously given the B form. These fibres contained 3"6~/o retained NaC1. F o r fibres containing a minimum of retained salt, the A form was observed; this form persisted without change to the S form even after annealing for 30 months. Sasisekharan & Brahmachari (1981) reported t h a t fibres of poly[d(G-C)] 9poly[d(G-C)] drawn from 1 : 1 (v/v) water/ethanol initially gave a B pattern at 40~/o relative humidity; then, as the relative h u m i d i t y was varied, gave a modified B - t y p e p a t t e r n followed b y a mixture of B and S patterns, and ultimately an S pattern. Behe et al. (1981) also observed the S form for Na poly[d(G-C)], poly[d(G-C)] and Na poly[d(G-5MeC)] 9poly[d(G-5MeC)] and the transition within oriented fibres from B to S. Poly[d(G-C)]. poly[d(G-C)] was synthesized specially for this investigation and studied as the potassium salt following precipitation from 0.1 M-KF. Fibres were prepared using standard techniques (Fuller et al., 1967) and X - r a y fibre diffraction patterns were recorded at a range of relative humidities from 32~/o to 98~/o. 897 0022-2836/83/240897-05 $03.00/0

9 1983 Academic Press Inc. (London) Ltd.

898

A. M A H E N D R A S I N G A M

ET AL.

If a fibre was first studied at low humidity (typically between 33~/o and 66~/o), a pattern was observed (Fig. 1(a)) that was somewhat similar to the semi-crystalline B patterns observed from fibres of NaDNA at high humidity; we designate this pattern B". When the relative humidity was raised (typically from 66~/o to 75~o), a structural transition to the A form was observed (Fig. l(b)). Further increase in the relative humidity (typically from 75~/o to 92~/o) led to a transition to the S form (Fig. l(e)). On some occasions, patterns were observed that were a mixture of the A and S forms. Once the fibre had fully assumed the S form, no further transition was observed unless the fibre was exposed to a very high humidity (typically 95~/o or greater). Under these conditions, a transition occurred to a semi-crystalline B form (Fig. l(d)). Gradual reduction of the relative humidity from 98% to 920/o resulted in a transition back to the S form. Once again, this form was very stable and persisted even when the relative humidity was reduced to 33~/o. However, if the relative humidity of the fibre environment was reduced more rapidly from 98~/o to 75~/o, the S form was not observed. Instead, either an A pattern or a mixture of A and B" patterns was observed. Following further reduction in the relative humidity into the range 33% to 66~o, the B" pattern was observed. The overall intensity distribution in the B" pattern (Fig. l(a)) is similar to that observed at high humidity from the semi-crystalline B form of DNA. However, in the B" pattern, there is additional well-defined diffraction on the meridion of the 5th layer-line. This diffraction might be from ions or ordered water in the structure but it is tempting to consider the possibility that it is due to differences in the conformation of the G and C residues analogous to the variation proposed by Klug et al. (1979) in the alternating B form of poly[d(A-T)], poly[d(A-T)]. It may be that the B" pattern reported here is closely related to the modified B pattern observed by Sasisekharan & Brahmachari (1981). This is the first report in which structural transitions within a nucleic acid fibre have been shown to depend on the rate at which the relative humidity of the fibre environment is varied. It should be emphasized that these transitions represent major structural changes in the poly[d(G-C)], poly[d(G-C)] double-helix, in which the high humidity semi-crystalline B form changes rapidly to the A form or more slowly to the S form. While the preliminary analysis of the X-ray fibre diffraction pattern of the S form is not definitive, a left-handed model derived from the Z conformation observed in single crystals of dCpG(pCpG)2 does appear to account in a satisfactory way for the observed fibre diffraction data. If the S form is indeed a left-handed helix then, assuming as is generally accepted that the A and B conformations are right-handed, both the B ~ S and A -~ S transitions involve a change in helix sense. While the possibility that the A and B forms are lefthanded has been seriously canvassed (e.g. Gupta et al., 1980), models with this helix sense have not been reported that have acceptable stereochemistry and are also in as good an agreement with the observed X-ray fibre diffraction as the best right-handed models (Fuller et al., 1965; Langridge et al., 1960; Arnott & Hukins, 1972,1973). The problems of constructing a left-handed model for the A form are particularly acute (Fuller et al., 1965). While transitions between left- and right-handed helices in solution are readily

LETTERS

TO T H E E D I T O R

899

D

9

~

~'

r ~:.

i!?....

5'

(a)

(b)

(c)

(d)

FIo. 1. X-ray diffraction patterns from K poly[d(G-C)] 9poly[d(G-C)]. (a) Semi-crystalline B" type at 66% relative humidity; (b) crystalline A type at 75o/o relative humidity; (c) crystalline 8 type at 66% relative humidity; (d) semi-crystalline B type at 98% relative humidity.

visualized, there is much greater difficulty in envisaging such changes within a fibre. However, it should be recognized t h a t a t the relative humidities at which the A -* S and B - ~ S transitions occur, the fibre contains between a third and a half by volume of water so t h a t there m a y well be sufficient spatial freedom for transitions as profound as a change of helix sense to occur. I t m a y well be, however, t h a t a transition involving a change in helix sense does require rather

900

A.

MAHENDRASINGAM E T AL.

more time to occur than one in which the sense is conserved, and that this is the origin of the dependence of the particular transition observed on the rate of change in relative humidity. I t may also be that this spatial freedom is rather less when the water content of the fibre is being reduced than when it is being increased. This difference may be the reason why within approximately the same relative humidity range the A -* S transition occurs with increasing relative humidity but the S - , A transition does not occur when the relative humidity is decreasing. The extensive study by Leslie et al. (1980) has focused on the dependence of the conformation of the DNA helix on its base sequence. The two extreme compositions, i.e. exclusively A-T and exclusively G-C are of crucial interest. Synthetic DNAs with both these compositions can assume the classical A and B forms. However, whereas poly[d(A-T)], poly[d(A-T)] readily assumes the D form (Mahendrasingam et al., 1983), poly[d(G-C)].poly[d(G-C)] apparently does not (Leslie et al., 1980). Conversely, the S form is characteristic of poly[d(G-C)].poly[d(G-C)] and has been reported by Arnott et al. (1980) as occurring occasionally for poly[(dA-dC)], poly[(dG-dT)]. From the point of view of their position within the pattern of conformational transitions within fibres, the S and D forms show marked similarities. Once assumed, both are particularly stable forms of the duplex. The forms are assumed with either increasing relative humidity for a fibre that was previously in the A form or with decreasing relative humidity for a fibre previously in the B form. So far, transitions from both forms have been observed only when the relative humidity is raised to a very high level (i.e. typically 95~/o), when the S -* B and D -* B changes occur. These similarities are of particular interest since, as can be seen readily from the X-ray fibre diffraction patterns of the S and D forms, the molecular conformations of'the two structures are quite distinct. This work was supported by Science and Engineering Research Council grant GR/C/23490 (to W.F. and W.J.P.). Department of Physics University of Keele Staffordshire, ST5 5BG, England

A. MAHENDRASINGAM W . J . PIGRAM W. FULLER

Centre National de la Recherche Scientifique Institut de Biologie Moleculaire Universite Paris VII 2 Place Jussieu, Paris Ve, France

J. BRAHMS J. VERGNE

Received 18 February 1983, and in revised form 18 April 1983 REFERENCES Arnott, S. & Hukins, D. W. L. (1972). Biochem. Biophys. Res. Commun. 47, 1504-1509. Arnott, S. & Hukins, D. W. L. (1973). J. Mol. Biol. 81, 93-105. Arnott, S., Chandrasekharan, R., Birdsall, D. L., Leslie, A. G. W. & Ratliff, R. L. (1980). Nature (London), 238, 743-745.

LETTERS TO THE EDITOR.

901

Behe, M., Zimmerman, S. & Felsenfeld, G. (1981). Nature (London), 293,233-235. Drew, H., Takan, T., Tanaka, S., Itakura, K. & Dickerson, R. E. (1980). Nature (London), 286, 567-573. Fuller, W., Wilkins, M. H. F., Wilson, H. R. & Hamilton, L. D. (1965). J. Mol. Biol. 12, 6080. Fuller, W., Hutchinson, F., Spencer, M. & Wiikins, M. H. F. (1967). J. Mol. Biol. 27, 507524. Gupta, G., Bansal, M. & Sasisekharan, V. (1980). Int. J. Biol. Macromol. 2, 368-380. Klug, A., Jack, A., Viswamitra, M. A., Kennard, 0., Shakked, Z. & Steitz, T. A. (1979). J. Mol. Biol. 131,669-680. Langridge, R., Marvin, D. A., Seeds, W. E., Wilson, H. R., Hooper, C. W., Wilkins, M. H. F. & Hamilton, L. D. (1960). J. Mol. Biol. 2, 38-64. Leslie, A. G. W., Arnott, S., Chandrasekharan, R. & Ratliff, R. L. (1980). J. Mol. Biol. 143, 49-72. Mahendrasingam, A., Rhodes, N. J., Goodwin, D. C., Nave, C., Pigram, W. J., Fuller, W., Brahms, J. & Vergne, J. (1983). Nature (London), 301,535-537. Pohl, F. M. & Jovin, T. M. (1972). J. Mol. Biol. 67, 375-396. Sasisekharan, V. & Brahmachari, S. K. (1981). Curt. Sci. 50, 10-13. Wang, A. H.-J., Quigley, G. J., Kolpak, F. J., Crawford, J. L., van Boom, J. H., van der Marel, G. & Rich, A. (1979). Nature (London), 282, 680-686. Edited by A. Klug