Hoedemaeker F J, Davidson A R, Rose D R
Ontario Cancer Institute and Department of Medical Biophysics, University of Toronto, Canada.
Proteins. 1998 Feb 15;30(3):275-86.
ABC transporters are a large superfamily of integral membrane proteins involved inATP-dependent transport across biological membranes. Members of this superfamily play roles in a number of phenomena of biomedical interest, including cystic fibrosis (CFTR) and multidrug resistance (P-glycoprotein, MRP). Most ABC transporters are predicted to consist of four domains, two membrane-spanning domains and two cytoplasmic domains. The latter contain conserved nucleotide-binding motifs. Attempts to determine the structure of ABC transporters and of their separate domains are in progress but have not yet been successful. To aid structure determination and possibly learn more about the domain boundaries, we set out to model nucleotide-binding domains (NBDs) of ABC transporters based on a known structure. Previous attempts to predict the 3D structure of NBDs were based solely on sequence similarity with known nucleotide-binding folds. We have analyzed the sequences of a number of nucleotide-binding domains with the algorithm THREADER, developed by D.T. Jones, and a possible fold was found in the structure of aspartate aminotransferase. We present a model for the N-terminal NBD of CFTR, based on the large domain of the A chain of aspartate aminotransferase. The model is refined using multiple sequence alignment, secondary structure prediction, and 3D-1D profiles. Our model seems to be in good agreement with known properties of nucleotide-binding domains and has some appealing characteristics compared with the previous models.
ABC转运蛋白是一类庞大的整合膜蛋白超家族，参与依赖ATP的跨生物膜转运。该超家族的成员在许多具有生物医学意义的现象中发挥作用，包括囊性纤维化（CFTR）和多药耐药性（P-糖蛋白、多药耐药相关蛋白）。大多数ABC转运蛋白预计由四个结构域组成，即两个跨膜结构域和两个胞质结构域。后者包含保守的核苷酸结合基序。确定ABC转运蛋白及其单个结构域结构的工作正在进行中，但尚未成功。为了辅助结构确定并可能更多地了解结构域边界，我们着手基于已知结构对ABC转运蛋白的核苷酸结合结构域（NBD）进行建模。先前预测NBD三维结构的尝试仅基于与已知核苷酸结合折叠的序列相似性。我们使用由D.T.琼斯开发的THREADER算法分析了多个核苷酸结合结构域的序列，并在天冬氨酸转氨酶的结构中发现了一种可能的折叠。我们基于天冬氨酸转氨酶A链的大结构域提出了CFTR N端NBD的模型。该模型通过多序列比对、二级结构预测和3D-1D图谱进行了优化。我们的模型似乎与核苷酸结合结构域的已知特性高度一致，并且与先前的模型相比具有一些吸引人的特征。
