Structure and thermodynamics of nonalternating C.G base pairs in Z-DNA: the 1.3-A crystal structure of the asymmetric hexanucleotide d(m5CGGGm5CG).d(m5CGCCm5CG).
Schroth, G.P., Kagawa, T.F., Ho, P.S.(1993) Biochemistry 32: 13381-13392
- PubMed: 8257675 
- DOI: https://doi.org/10.1021/bi00212a002
- Primary Citation of Related Structures:  
145D - PubMed Abstract: 
We have solved the single-crystal X-ray structure of the complementary hexanucleotides d(m5-CGGGm5CG) and d(m5CGCCm5CG). The hexamer duplex was crystallized as Z-DNA, but contains a single C.G base pair that does not follow the alternating pyrimidine/purine rule for Z-DNA formation. This is the first crystal structure which serves to illustrate the structural consequences of placing a cytosine in the sterically disfavored syn conformation. In addition, since these sequences are not self-complementary, the individual strands of this asymmetric hexamer are unique in sequence and therefore distinguishable in the crystal lattice. Nevertheless, the crystal of this duplex is isomorphous with other Z-DNA hexamer structures. The asymmetry of this hexamer sequence required that the structure be solved using two unique models, which are distinguished by the orientation of hexanucleotides in the crystal lattice. In one model (the GG model) the cytosine in the syn conformation is packed against the terminal guanine base of a symmetry-related hexamer, while in the alternative model (the CC model) this cytosine sits exposed in a solvent channel of the lattice. We find that neither model alone can completely account for the observed electron densities. The two models ultimately were refined together. A composite structure consisting of 65% GG model and 35% CC model refined to an R-factor of 19.3%, which was significantly lower than refinements using either model alone. A detailed analysis of these two structures shows that, in spite of the out-of-alternation C.G base pair, the features characteristic of Z-DNA have been maintained. Both models, however, show significant local structural adjustments to accommodate the single cytosine base which is forced to adopt the syn conformation in each hexamer. In general, it appears that in order to relieve the energetically unfavorable steric contacts between the cytosine base in the syn conformation and the deoxyribose sugar, the base is forced into a highly buckled conformation, and that this large buckle in turn alters the conformation of neighboring residues. This unusual conformation also significantly weakens base-stacking interactions between the cytosine in syn and the adjacent residues in the helix and affects the exposure of the bases to solvent. We conclude that this crystal structure provides a molecular rationale for why nonalternating bases are energetically disfavored in Z-DNA.
Organizational Affiliation: 
Department of Biochemistry and Biophysics, Oregon State University, Corvallis 97331.