Fabrega Sylvie, Durand Patrick, Mornon Jean-Paul, Lehn Pierre
INSERM U458, Höpital Robert Debré, 48, Bd Sérurier, 75019 Paris.
J Soc Biol. 2002;196(2):151-60.
Gaucher disease is a lysosomal storage disorder caused by a deficiency in glucocerebrosidase which cleaves the beta-glucosidic linkage of glucosylceramide, a normal intermediate in glycolipid metabolism. Glucocerebrosidase belongs to the clan GH-A of glycoside hydrolases, a large group of enzymes which function with retention of the anomeric configuration at the hydrolysis site. Accurate three-dimensional (3D) structure data for glucocerebrosidase should help to better understand the molecular bases of Gaucher disease. As such 3D structure data were not available, we used the two-dimensional hydrophobic cluster analysis (HCA) method to make structure predictions for the catalytic domains of clan GH-A glycoside hydrolases. We found that all the enzymes of clan GH-A may share a similar catalytic domain consisting of an (alpha/beta)8 barrel with the critical acid/base and nucleophile residues located at the C-terminal ends of strands beta 4 and beta 7, respectively. In the case of glucocerebrosidase, Glu 235 was predicted to be the putative acid/base catalyst whereas the nucleophile was located at Glu 340. Next, in order to obtain experimental evidence supporting these HCA-based predictions, we used retroviral vectors to express, in murine null cells, E235A and E340A mutant proteins, in which alanine residues unable to participate in the enzymatic reaction replace the presumed critical glutamic acid residues. Both mutants were found to be catalytically inactive although they were correctly folded/processed and sorted to the lysosome. Thus, Glu 235 and Glu 340 do indeed play key roles in the active site of human glucocerebrosidase as predicted by the HCA analysis. In a broader perspective, our work points out that bioinformatics approaches may be highly useful for generating structure-function predictions based on sequence-structure interrelationships, especially in the context of a rapid increase in protein sequence information through genome sequencing.
戈谢病是一种溶酶体贮积症，由葡糖脑苷脂酶缺乏引起，该酶可裂解糖脂代谢的正常中间产物葡糖神经酰胺的β-糖苷键。葡糖脑苷脂酶属于糖苷水解酶的GH-A家族，这是一大类在水解位点以异头构型保留的方式发挥作用的酶。葡糖脑苷脂酶准确的三维（3D）结构数据应有助于更好地理解戈谢病的分子基础。由于尚无此类3D结构数据，我们使用二维疏水簇分析（HCA）方法对GH-A家族糖苷水解酶的催化结构域进行结构预测。我们发现，GH-A家族的所有酶可能共享一个相似的催化结构域，该结构域由一个（α/β）8桶组成，关键的酸/碱和亲核残基分别位于β4和β7链的C末端。就葡糖脑苷脂酶而言，预测Glu 235为假定的酸/碱催化剂，而亲核试剂位于Glu 340处。接下来，为了获得支持这些基于HCA预测的实验证据，我们使用逆转录病毒载体在小鼠缺失细胞中表达E235A和E340A突变蛋白，其中无法参与酶促反应的丙氨酸残基取代了假定的关键谷氨酸残基。尽管这两种突变体都正确折叠/加工并分选到溶酶体中，但发现它们均无催化活性。因此，正如HCA分析所预测的，Glu 235和Glu 340确实在人葡糖脑苷脂酶的活性位点中起关键作用。从更广泛的角度来看，我们的工作指出，生物信息学方法对于基于序列-结构相互关系生成结构-功能预测可能非常有用，特别是在通过基因组测序蛋白质序列信息迅速增加的背景下。
