DNA-ligand interactions studied through the chiral nature of DNA

DNA配体对DNA碱基序列的识别是所有生命形式中最基本、最重要的问题之一。为了在分子水平上了解DNA与配体的相互作用，我们一直在研究低分子金属络合物，特别是非手性金属卟啉络合物，因为它们在Soret带区显示出大量的诱导CD。我们之前已经揭示了诱导CD光谱反映了DNA的主槽结合、小槽结合和碱基对之间的嵌入等多种结合模式，为了获得关于结合模式的定量信息，我们采用了两种独立的方法。一种是基于遗传算法和最速下降技术相结合的方法，另一种是基于更经验性的方法，即用[DNA]对不同r(r=[卟啉]/[DNA])的诱导圆二色谱进行归一化。前者涉及大量使用C语言的计算机编程。不采用通常的洛伦兹或高斯曲线，用t分布的一般形式对光能进行了模拟。组件的数量由误差减少的程度决定。总的来说，模拟是成功的，我们目前正在编写一个新的程序，根据模拟的CD曲线计算位点大小和结合常数。另一种方法最多只涉及三种成分。为了将这一过程扩展到其他系统，特别是DNA结合蛋白，我们合成了与多肽或氨基酸残基结合的偶联物。

DAN base-sequence recognition by DNA ligands is one of the most fundamental and important issues in every life forms. To understand DNA-ligand interactions at the molecular level, we have been working on low molecular-weight metal complexes, especially achiral metal porphyrin complexes as they exhibit substantial induced CD in the Soret band region. We have previously revealed that induced CD spectra reflect various binding modes such as major groove binding, minor groove binding and intercalation between base-pairs of DNA.To obtain quantitative information on the binding modes, we have adopted two independent methods. One is based on the genetic algorithm combined with the steepest descent technique, and the other is based on more empirical approach by normalizing induced CD spectra recorded at different r (r=[porphyrin]/[DNA]) values by [DNA]. The former involved substantial computer programming using c-language. Without adopting the usual Lorentzian or Gaussian curves, the spectra were simulated with general forms with t-distribution against light energy. The number of components was determined by the degree of reduction of errors. Generally speaking, simulation was successful and we are currently writing a new program which calculates site sizes and binding constants based on the simulated CD curves. The other method involved only three components at maximum. Despite the small numbers of components, the simulation was good.To extend the process to other systems especially to DNA-binding proteins, we have synthesized conjugated with a peptide or amino acid residue.