GPD scrambling using previously described radiolabeled GPD analogs3).As seen in the immunoblot (Figure 2), wild-type (WT) cells show a single intense band corresponding to fully glycosylated CPY, whereas a mutant strain lacking the GPD synthase Alg5 shows hypo-glycosylation in the form of two additional, lower molecular weight bands corresponding to -1 and -2 N-glycans. Likewise, other known glycosylation mutants lacking the glucosyltransferases Alg6 or Alg8, or the lumenal mannosyltransferases Alg3 or Alg9, show significant hypo-glycosylation, with the two latter strains displaying an additional band corresponding to -3 N-glycans (CPY glycoforms are indicated by colored dots in the alg6Δ lane; red corresponds to the fully glycosylated protein, and blue corresponds to protein lacking 3 of 4 N-glycans). In contrast, lack of the GPD scramblase candidates showed no N-glycosylation defect. We conclude that these candidates are not necessary for GPD scrambling.ResultsWe present here the results of our initial evaluation of GPD scramblase candidates. These were tested using an in vivo approach in which we compared the glycosylation phenotype of the corresponding yeast deletion mutants to a mutant (alg5Δ) with GPD synthase deficiency as well as other known glycosylation mutants. We used immunoblotting to determine the glycosylation status of the well-characterized vacuolar protease carboxypeptidase Y (CPY) which has 4 N-glycans. A deficiency in glucosylation arising from GPD synthesis or scrambling would be expected to result in hypoglycosylation. Figure 2. Evaluating GPD scramblase candidates Haploid yeast strains with disruptions of the following genes encoding GPD scramblase candidates (underlined) and control proteins: ALG5, ALG6, ALG8, ALG3, ALG9, ERC1, STE24, SUR2, SCS7, YDR051W, YBR220C, YBR338C, ALE1, ERG24, MNS1. Gene disruption was verified by colony PCR before analyzing CPY N-glycosylation. Yeast cultures were grown to exponential phase. Cells were lysed with glass beads in sample buffer and equal amounts of protein were resolved on SDS-polyacrylamide gels and taken for immunoblotting using anti-CPY antibodies.spontaneous rate of GPD flipping is extremely low, a specific transporter is needed to move GPD across the ER membrane at a physiological rate. The molecular identity of this important transport protein - herein termed GPD scramblase - is not known and its discovery remains a major challenge for molecular cell biology. The central role of glucosylation in optimal ER protein N-glycosylation was recognized more than four decades ago3), and its relevance for cellular physiology and human disease is abundantly clear as glucosylation deficiency results in Congenital Disorders of Glycosylation. Our aim is to identify GPD scramblase via a combination of bioinformatics and biochemistry. ApproachAlthough GPD is important in yeast and humans, not all N-glycosylation-competent organisms have glucose in their N-glycan precursor. Taking advantage of this fact, we implemented an innovative bioinformatics approach, termed phylogenetic profiling4,5), for assignment of protein function. Phylogenetic profiling is a powerful method to identify functionally associated, co-evolving proteins such as enzymes that catalyze sequential steps in a biochemical pathway, and as such it identifies the most significant co-evolutionary events. We therefore hypothesized that the presence or absence of GPD scramblase in a particular organism will be highly correlated with, but not necessarily identical to, the presence or absence of other proteins of the glucosylation pathway, e.g., the glucosyltransferase Alg6. Implementing this idea, we identified several polytopic ER membrane proteins as GPD scramblase candidates in yeast6). Our plan was to evaluate these candidates by in vivo approaches, for example by comparing the phenotypes of the corresponding yeast mutants to GPD synthase deficiency, as well as by using reconstitution-based assays of 85
元のページ ../index.html#85