Grzeskowiak K, Yanagi K, Privé G G, Dickerson R E
Department of Chemistry and Biochemistry, University of California, Los Angeles 90024.
J Biol Chem. 1991 May 15;266(14):8861-83. doi: 10.2210/pdb1d23/pdb.
The crystal structure of the DNA decamer C-G-A-T-C-G-A-T-C-G has been solved to a resolution of 1.5 A, with a final R-factor of 16.1% for 5,107 two-sigma reflections. Crystals are orthorhombic space group P2(1)2(1)2(1), with cell dimensions a = 38.93 A, b = 39.63 A, c = 33.30 A, and 10 base pairs/asymmetric unit. The final structure contains 404 DNA atoms, 142 water molecules treated as oxygen atoms, and two Mg(H2O)6(2+) complexes. Decamers stack atop one another to simulate continuous helical columns through the crystal, as with three previously solved monoclinic decamers, but the lateral contacts between columns are quite different in the orthorhombic and monoclinic cells. Narrow and wide regions of the minor groove exhibit a single spine or two ribbons of hydration, respectively, and the minor groove is widest when BII phosphate conformations are opposed diagonally across the groove. Phosphate conformation, in turn, appears to have a base sequence dependence. Twist, rise, cup, and roll are linked as has been observed in the three monoclinic decamers and can be characterized by high or low twist profiles. In all five known decamer crystal structures and eight representative dodecamers, a high twist profile is observed with G-C and G-A steps whereas all other R-R steps are low twist profiles (R = purine). A-T and A-C steps are intermediate in character whereas C-A and C-G exhibit behavior that is strongly influenced by the profiles of the preceding and following steps. When sufficient data are in hand, sequence/structure relationships for all helix parameters probably should be considered in a 4-base pair context. At this stage of limited information the problem is compounded because there are 136 unique 4-base steps x-A-B-y in a double helix as compared with only 10 2-base steps A-B.
已解析出DNA十聚体C-G-A-T-C-G-A-T-C-G的晶体结构，分辨率为1.5埃，对于5107个两倍标准差反射，最终R因子为16.1%。晶体属于正交空间群P2(1)2(1)2(1)，晶胞参数a = 38.93埃，b = 39.63埃，c = 33.30埃，每个不对称单元含10个碱基对。最终结构包含404个DNA原子、142个视为氧原子的水分子以及两个Mg(H2O)6(2+)配合物。十聚体彼此堆叠，如同之前解析出的三个单斜晶系十聚体一样，在晶体中模拟连续的螺旋柱，但正交晶胞和单斜晶胞中柱之间的侧向接触有很大不同。小沟的窄区和宽区分别呈现单条水合脊或两条水合带，当BII磷酸构象在小沟对角相对时，小沟最宽。反过来，磷酸构象似乎依赖于碱基序列。扭转、上升、弯曲和滚动相互关联，这在三个单斜晶系十聚体中已观察到，并且可以通过高扭转或低扭转轮廓来表征。在所有五个已知的十聚体晶体结构和八个代表性的十二聚体中，G-C和G-A步呈现高扭转轮廓，而所有其他R-R步（R = 嘌呤）为低扭转轮廓。A-T和A-C步性质居中，而C-A和C-G步的行为受前后步轮廓的强烈影响。当有足够数据时，所有螺旋参数的序列/结构关系可能应在4碱基对的背景下考虑。在目前信息有限的阶段，问题更加复杂，因为双螺旋中有136个独特的4碱基步x-A-B-y，而只有10个2碱基步A-B。
