Hocart S J, Reddy V, Murphy W A, Coy D H
Department of Medicine, Tulane University School of Medicine, New Orleans, Louisiana 70112, USA.
J Med Chem. 1995 May 26;38(11):1974-89. doi: 10.1021/jm00011a017.
Somatostatin is a hypothalamic hormone that inhibits the release of growth hormone (GH). It has also been shown to inhibit the release of a broad range of hormones including insulin, glucagon, and gastrin. Presently, five different receptor subtypes of somatostatin have been characterized and cloned. Our previous work on the structure-activity relationship of somatostatin and that of many others has generated a large database of analogues with different biological activities and receptor affinities. This present work is an investigation of the growth hormone release-inhibiting potencies of somatostatin analogues by the three-dimensional quantitative structure-activity paradigm, comparative molecular field analysis (CoMFA). A total of 64 analogues were modeled in SYBYL using structural information from two NMR studies. The molecules were aligned by a root-mean-square fit of atoms and field-fit of the steric and electrostatic molecular fields and the resulting databases analyzed by partial least squares analysis with cross-validation to extract the optimum number of components. The analysis was then repeated without cross-validation to give the final QSAR models. Preliminary investigations with the CoMFA models led to the synthesis of a new somatostatin analogue. This compound together with five other newly synthesized compounds not included in the original training sets were used to test the predictive ability of the CoMFA models. Two models with good predictive powers are presented.
生长抑素是一种下丘脑激素,可抑制生长激素(GH)的释放。研究还表明,它能抑制包括胰岛素、胰高血糖素和胃泌素在内的多种激素的释放。目前,已对生长抑素的五种不同受体亚型进行了表征和克隆。我们之前关于生长抑素及许多其他物质构效关系的研究产生了一个包含不同生物活性和受体亲和力类似物的大型数据库。本研究采用三维定量构效关系范式——比较分子场分析(CoMFA),对生长抑素类似物抑制生长激素释放的效力进行了研究。利用两项核磁共振研究的结构信息,在SYBYL中对总共64种类似物进行了建模。通过原子的均方根拟合以及空间和静电分子场的场拟合对分子进行比对,并通过偏最小二乘法分析和交叉验证对所得数据库进行分析,以提取最佳成分数量。然后在不进行交叉验证的情况下重复分析,以得到最终的定量构效关系模型。对CoMFA模型的初步研究促使合成了一种新的生长抑素类似物。该化合物与其他五种未包含在原始训练集中的新合成化合物一起,用于测试CoMFA模型的预测能力。本文给出了两个具有良好预测能力的模型。