Saltar al contenido
MilliporeSigma

Insight into the regulation of glycan synthesis in Drosophila chaoptin based on mass spectrometry.

PloS one (2009-05-06)
Yoshimi Kanie, Miki Yamamoto-Hino, Yayoi Karino, Hiroki Yokozawa, Shoko Nishihara, Ryu Ueda, Satoshi Goto, Osamu Kanie
RESUMEN

A variety of N-glycans attached to protein are known to involve in many important biological functions. Endoplasmic reticulum (ER) and Golgi localized enzymes are responsible to this template-independent glycan synthesis resulting glycoforms at each asparagine residues. The regulation mechanism such glycan synthesis remains largely unknown. In order to investigate the relationship between glycan structure and protein conformation, we analyzed a glycoprotein of Drosophila melanogaster, chaoptin (Chp), which is localized in photoreceptor cells and is bound to the cell membrane via a glycosylphosphatidylinositol anchor. Detailed analysis based on mass spectrometry revealed the presence of 13 N-glycosylation sites and the composition of the glycoform at each site. The synthetic pathway of glycans was speculated from the observed glycan structures and the composition at each N-glycosylation site, where the presence of novel routes were suggested. The distribution of glycoforms on a Chp polypeptide suggested that various processing enzymes act on the exterior of Chp in the Golgi apparatus, although virtually no enzyme can gain access to the interior of the horseshoe-shaped scaffold, hence explaining the presence of longer glycans within the interior. Furthermore, analysis of Chp from a mutant (RNAi against dolichyl-phosphate alpha-d-mannosyltransferase), which affects N-glycan synthesis in the ER, revealed that truncated glycan structures were processed. As a result, the distribution of glycoforms was affected for the high-mannose-type glycans only, whereas other types of glycans remained similar to those observed in the control and wild-type. These results indicate that glycan processing depends largely on the backbone structure of the parent polypeptide. The information we obtained can be applied to other members of the LRR family of proteins.