2',3'‐Cyclic Nucleotide 3'‐Phosphodiesterase Has Characteristics of Cytoskeletal Proteins A Hypothesis for Its Function a
2',3'‐Cyclic Nucleotide 3'‐Phosphodiesterase Has Characteristics of Cytoskeletal Proteins A Hypothesis for Its Function a
复制标题
2,3-环状核苷酸 3-磷酸二酯酶具有细胞骨架蛋白的特征及其功能的假设
DOI:
10.1111/j.1749-6632.1990.tb42380.x
复制
发表时间:
1990
影响因子:
5.2
通讯作者:
L. Bernier
中科院分区:
文献类型:
--
作者:
P. E. Braun;L. Bambrick;A. Edwards;L. Bernier
The earliest appearing, myelinogenesis-associated polypeptide expressed in the CNS exclusively by myelinating cells is 2',3 '-cyclic nucleotide, 3'-phosphodiesterase (CNP). (For a review see Vogel and Thompson.') The mRNA for this protein first appears prenatally, and the protein itself can be visualized immunocytochemically in oligodendrocytes long before myelination is evident, even in putative oligodendrocyte progenitors in the subventricular region of mouse brain.* The fact that neither we nor others3 observe CNP in compact myelin, but find it concentrated in the cytoplasm of noncompacted oligodendroglial ensheathments of axons and in the paranodal loops suggests to us that CNP does not likely contribute to the architecture of the myelin sheath as do the major structural proteins, namely, proteolipid protein and myelin basic p r ~ t e i n . ~ CNP nevertheless coisolates with myelin membranes, but these can be subfractionated by density gradient centrifugation with the subsequent appearance of CNP in the denser membranes (presumably unilamellar elements originating from the paranodal loops) that contribute to the spectrum of membranes comprising the myelin fraction as routinely isolated (reviewed by Danks and matt hie^.^ Although the bulk of CNP enzymatic activity coisolates with CNS myelin, a significant fraction is also associated with nonmyelin membrane fractions.6 The incompatibility of these observations with the earlier widespread view that CNP is affiliated with the lipidrich lamellae of compact myelin is further exemplified by the behavior of CNP towards extraction. Several studiesa show that the partitioning of CNP into nonionic detergents is similar to the partitioning behavior of cytoskeletal or membranoskeletal proteins and is very different from the observed behavior of intrinsic membrane proteins. Recently, we9 and Sprinkle and HancockIo determined that CNP possesses two of the three domains believed to participate in GTP binding. These are GXXXXGKWT and DXXGIA. Another domain, YFGKRPPG, known in other proteins for its adenine recognition capacity, is also present in CNP. All of these are apparently unrelated to the putative catalytic domain responsible for the phosphodiesterase activity of CNP. In addition, we found that the C-terminal domain, CXXX, is identical to that of some of the nonrus G proteins as well as the GTP-binding family of ras proteins. In ras-related