In our lectures we have discussed that reglucosylation of the protein-linked oligosaccharide mannose9-N-acetylglucosamide2 (M9N2, M = mannose, N = N-acetylglucosamine) by UDP- glucose glycoprotein glucosyltransferase (UGGT) using UDP-glucose as glucose donor can trigger a second round of interaction of a folding protein with the lectin chaperones calnexin and calreticulin. We have also discussed that comparatively slow demannosylation of protein- linked M9N2 to successively demannosylated species such M8N2, M7N2, M6N2, and M5N2, are superimposed over the deglucosylation-reglucosylation cycle and ultimately targets slowly folding glycoproteins to the ER-associated protein degradation machinery.
In this summative assessment you will analyse and interpret some of the experimental evidence for this model of the calnexin/calreticulin cycle.

ANSWER ALL PARTS (QUESTIONS 1 TO 4) OF THIS SUMMATIVE ASSESSMENT.

1. Figure 1 shows the profile of protein-linked oligosaccharides that were radioactively labelled in vivo by incubation of rat liver cells (Figure 1A), common bean seedlings (Figure 1B), or yeast cells (Figure 1C) for ~20 min with [U-14C]-D-glucose (U stands for ‘universal’, meaning that the radioactive isotope 14C of carbon has been incorporated equally into all six positions of the carbon atom in the glucose molecule).
After lysis of cells, protein-linked oligosaccharides were released from proteins by digestion with endoglycosidase H (Endo H). Release of oligosaccharides from proteins with Endo H leaves the terminal N-acetylglucosamine on the protein. For example, for the protein-linked oligosaccharide G3M9N2, Endo H will release the oligosaccharide G3M9N, for the protein-linked oligosaccharide M9N2 Endo H will release M9N, and for protein- linked M8N2 Endo H will release M8N.
The released oligosaccharides were then separated by a chromatographic technique, and their carbohydrate (sugar) composition identified by comparing their migration during the chromatography to standards of known carbohydrate composition and structure (Figure 1).
a) Based on the data shown in Figure 1, which cells show incorporation of [U-14C]- glucose into the protein-linked oligosaccharide glucose1-mannose9-N- acetylglucosamine2 (G1M9N2, G = glucose, M = mannose, N = N- acetylglucosamine) (10 % of the marks)?
b) For yeast cells (Figure 1C), how may you explain the formation of radioactively labelled M9N and M8N when the cells are labelled with [U-14C]-glucose (10 % of the marks)?

2. Further investigation shows that all of the radioactivity in G1M8N and G1M7N, where synthesised (Figure 1) resides in the glucose residue; no radioactivity is found in any of the mannose residues or the N-acetylglucosamine residue. Two explanations for formation of G1M8N2 and G1M7N2 are proposed:

Figure 1. Chromatographic separation of protein-linked oligosaccharides in vivo in (A) rat liver cells, (B) common bean seedlings, and (C) yeast cells. Arrows marked G1M9N, G1M8N, G1M7N, M9N, M8N, and M7N show the migration positions of the corresponding marker oligosaccharides. Abbreviation: cpm – counts per minute.

i. Protein-linked G1M8N2 and G1M7N2 are formed by glucosylation of protein-linked M8N2 and protein-linked M7N2, respectively.
ii. Protein-linked G1M8N2 is formed by demannosylation of G1M9N2 to G1M8N2. Similarly, G1M7N2 is formed by demannosylation of G1M8N2.

Figure 2. Effect of inhibition of 1,2-mannosidases on the chromatographic profiles of protein-linked oligosaccharides. (A) No inhibition of 1,2-mannosidases, (B) ~40% inhibition of 1,2-mannosidases, and (C) ~70% inhibition of 1,2-mannosidases. Arrows marked G1M9N, G1M8N, and G1M7N show the migration positions of the corresponding marker oligosaccharides. Numbers represent the 14C radioactivity in the fraction (in cpm, counts per minute). Numbers shown in black correspond to the G1M9N peak, numbers shown in red to the G1M8N peak, and numbers shown in blue to the G1M7N peak.

To distinguish between these two possible explanations, rat liver microsomes [Microsomes are vesicles isolated from both the endoplasmic reticulum and the Golgi complex. They contain all the enzymes for protein glycosylation] were labelled with UDP-[U-14C]-glucose under conditions that fully inhibit transfer of oligosaccharides from the lipid-linked precursor G3M9N2-P-P-dolichol (P = phosphate group) onto proteins (i.e. in the absence of detergents). Protein-linked oligosaccharides were then released by digestion with Endo H and chromatographically separated (Figure 2A). The experiment was repeated in the presence of increasing concentrations of oligosaccharides of mannose, which act as inhibitors of 1,2-mannosidases (Figures 2B, C).
Discuss how the data shown in Figure 2 support either mechanism i., mechanism ii., both mechanisms, or none of the two mechanisms (25 % of the marks).

3. Figure 1 shows evidence that the protein-linked oligosaccharide G1M9N2 exists. In the calnexin/calreticulin cycle this protein-linked oligosaccharide (G1M9N2) is deglucosylated to terminate the interaction with calnexin/calreticulin.
Discuss how the data shown in Figure 3 either support the hypothesis that protein-linked G1M9N2 can be deglucosylated to M9N2, do not support this hypothesis (i.e. support the hypothesis that G1M9N2, once formed, is stable, and cannot be deglucosylated), or are inconclusive with regard to distinguishing between these two hypotheses. How might the red curve look if it would support the alternative hypothesis (i.e. the hypothesis that you decided that has been ruled out by the data in Figure 3)? (25 % of the marks).

4. Degradation of the protein-linked oligosaccharide M8N2 to species with a lower mannose content such as M7N2, M6N2, and M5N2, is thought to terminate the cycling of a glycoprotein in the calnexin/calreticulin cycle and to target the affected glycoprotein to the ER-associated protein degradation (ERAD) machinery. It is now very worthwhile to

Figure 3. Labelling of protein-linked oligosaccharides in rat liver microsomes with UDP- [14C]-glucose. At the indicated times (●, black circles on the black line) samples of equal volumes were withdrawn, and the radioactivity incorporated into protein determined. After 5 min, the reaction was split into two equal parts, in one part the labelling reaction was allowed to continue as before; the second part (○, open red circles on the red line) was added to a tube containing a large excess on non-radioactive [UDP]-glucose. After 10 and 15 min equal volumes are withdrawn from both reactions and the radioactivity incorporated into protein determined.

compare the information from question 3, as well as some information from the formative assessment, with the stability of protein-linked M8N2.
To do this we determine the half-life of protein-linked M8N2 by labelling cells with radioactive [2-3H]-mannose for 10 min, and then incubating them in a vast excess of non- radioactive mannose for another 10, 20, or 40 min. Protein-linked oligosaccharides are then released with Endo H, separated by gel filtration, and the radioactivity of individual fractions is measured and plotted against the fraction number (Figure 4).
Are the half-lives of protein-linked M8N2 and protein-linked G3M9N2 (determined in the formative assessment) consistent with the model of the calnexin/calreticulin cycle or not? Explain your answer (30% of the marks).

Figure 4. Demannosylation of the protein-linked oligosaccharide M8N2 (represented by M8N in the gel filtration profiles). Gel filtration profiles of radioactively labelled oligosaccharides isolated from cells that were (A) labelled for 10 min with [2-3H]mannose and then chased with an excess of non-radioactive mannose for (B) 10 min, (C) 20 min, and (D) 40 min. Arrows marked M8N show the elution positions of the corresponding marker oligosaccharides. Numbers show the radioactivity (in dpm/0.1 ml, where dpm = disintegrations per minute) in each data point of the M8N peak.