IMPROVED PARAMETERS FOR PREDICTION OF RNA STRUCTURE
IMPROVED PARAMETERS FOR PREDICTION OF RNA STRUCTURE
复制标题
DOI：
10.1101/sqb.1987.052.01.017
复制
发表时间：
1987-01-01
期刊：
影响因子：
--
通讯作者：
KIERZEK, R
中科院分区：
文献类型：
--
作者：
TURNER, DH;SUGIMOTO, N;KIERZEK, R
要完全理解酶催化作用的分子基础，需要了解酶的三维结构。确定酶的序列比确定结构容易得多。因此，结构测定是理解结构-功能关系的限制性步骤。这种限制在RNA酶的情况下特别严重，因为它是非常困难的结晶大RNA分子。原则上，加速我们理解RNA酶的一种方法是开发从序列可靠地预测结构的方法。对于RNA来说，开发这样的方法应该比蛋白质更容易，因为决定RNA结构的局部相互作用往往比决定蛋白质结构的局部相互作用更强。例如，短的、分离的RNA螺旋比它们的蛋白质对应物更稳定（Freier et al. 1986 b; Shoemaker et al. 1987）。Tinoco等人（1971）开发了一种预测RNA二级结构的方法，该方法基于最小化与最近邻相互作用相关的自由能变化。然而，这种方法的应用受到了阻碍，缺乏实验数据提供参数的最近邻相互作用。合成寡核糖核苷酸方法的最新改进（Neilson等，1980; Beckett和Uhlenbeck 1984; Markiewicz等，1984; Kierzek等，1986）使得有可能极大地扩展提供最近邻相互作用的数据库。我们报告改进的参数来自短RNA双链体的稳定性的热力学研究。研究结果还提供了对基本相互作用（如堆叠和氢键）强度的深入了解。与酵母苯丙氨酸tRNA的三维结构的比较表明，这些见解也可能是有用的预测RNA的三维结构。
Complete understanding of the molecular basis of catalysis by an enzyme requires knowledge of the threedimensional structure of the enzyme. Determining sequences of enzymes is much easier than determining structures. Thus, structure determination is a limiting step in understanding structure-function relationships. This limitation is particularly severe in the case of RNA enzymes because it is extremely difficult to crystallize large RNA molecules. In principle, one way to accelerate our understanding of RNA enzymes is to develop methods for reliably predicting structure from sequence. It should be easier to develop such methods for RNA than for proteins because the local interactions determining RNA structures tend to be stronger than those determining protein structure. For example, short, isolated RNA helixes are more stable than their protein counterparts (Freier et al. 1986b; Shoemaker et al. 1987).Prediction of secondary structure is one step toward prediction of three-dimensional structure. Tinoco et al.(1971) developed a method for predicting the secondary structure of RNA that is based on minimizing the free-energy change associated with nearest-neighbor interactions. Applications of this method, however, have been hindered by lack of experimental data providing parameters for the nearest-neighbor interactions. Recent improvements in methods for synthesizing oligoribonucleotides (Neilson et al. 1980; Beckett and Uhlenbeck 1984; Markiewicz et al. 1984; Kierzek et al. 1986) make it possible to greatly expand the database for providing nearest-neighbor interactions. We report improved parameters derived from thermodynamic studies of the stability of short RNA duplexes. The results also provide insight into the strengths of fundamental interactions such as stacking and hydrogen bonding. Comparisons with the threedimensional structure of yeast phenylalanine tRNA suggest that these insights may also be useful for predicting the three-dimensional structure of RNA.