Dolichol-linked oligosaccharide selection by the oligosaccharyltransferase in protist and fungal organisms

Abstract

The dolichol-linked oligosaccharide Glc3Man9GlcNAc2-PP-Dol is the in vivo donor substrate synthesized by most eukaryotes for asparagine-linked glycosylation. However, many protist organisms assemble dolichol-linked oligosaccharides that lack glucose residues. We have compared donor substrate utilization by the oligosaccharyltransferase (OST) from Trypanosoma cruzi, Entamoeba histolytica, Trichomonas vaginalis, Cryptococcus neoformans, and Saccharomyces cerevisiae using structurally homogeneous dolichol-linked oligosaccharides as well as a heterogeneous dolichol-linked oligosaccharide library. Our results demonstrate that the OST from diverse organisms utilizes the in vivo oligo saccharide donor in preference to certain larger and/or smaller oligosaccharide donors. Steady-state enzyme kinetic experiments reveal that the binding affinity of the tripeptide acceptor for the protist OST complex is influenced by the structure of the oligosaccharide donor. This rudimentary donor substrate selection mechanism has been refined in fungi and vertebrate organisms by the addition of a second, regulatory dolichol-linked oligosaccharide binding site, the presence of which correlates with acquisition of the SWP1/ribophorin II subunit of the OST complex.

Figure 1.

OS-PP-Dol donors and predicted subunit compositions of the OST from selected eukaryotes. The left portion of A–D shows the oligosaccharide structure of the in vivo donor for N-linked glycosylation. N-acetylglucosamine residues are designated by squares, mannose residues are shown as circles, and glucose residues are shown as triangles. Red saccharides are transferred by cytoplasmically oriented ALG gene products, and blue residues are transferred by luminally oriented ALG gene products. The right section of each panel shows the predicted (A, T. cruzi; B, C. neoformans; or C, T. vaginalis and E. histolytica) or experimentally determined (D, S. cerevisiae) composition of the OST complex. The color code of the subunits (red, green, and blue) designates subcomplexes detected in higher eukaryotes (Karaoglu et al., 1997; Spirig et al., 1997). The yellow bar designates the ER membrane.

Figure 2.

Donor substrate selection from an OS-PP-Dol library. The purified yeast OST (A) or detergent extracts prepared from T. cruzi (B), T. vaginalis (C), C. neoformans (D), or E. histolytica (not depicted) membranes were assayed for OST activity in the presence of glucosidase and mannosidase inhibitors using 1.2 μM OS-PP-Dol and 10 μM tripeptide acceptor (Nα-Ac-N-[¹²⁵I]-Y-T NH₂). Glycopeptide products ranging in size between M₃GN₂-NYT (M3) and G₃M₉GN₂-NYT (G3) were resolved by HPLC and identified by migration relative to authentic standards (M5, M9, and G3). (E and F) The normalized initial transfer rate (OS-NYT/OS-PP-Dol) for the eight most abundant OS-PP-Dol donors was calculated by dividing the composition of the glycopeptide products by the composition of the OS-PP-Dol donor library. The OST from all organisms was assayed twice to determine the composition of the product pool. The composition of the donor pool was determined by duplicate assays (Fig. S1, available at http://www.jcb.org/cgi/content/full/jcb.200611079/DC1). Note the difference in ordinate scales for E and F.

Figure 3.

Reduced transfer of biosynthetic M₅GN₂-PP-Dol by the T. cruzi OST. (A) Biosynthetic M₅GN₂-PP-Dol (b) and M₅GN₂-PP-Dol isomers (c–i) produced by mannosidase digestion of M₉GN₂-PP-Dol (a). GlcNAc residues are indicated by squares, α-1,2–linked mannose residues are indicated by red circles, and α-1,3– and α-1,6–linked mannose residues are indicated by open circles. (B) Glycopeptide products obtained in an OST endpoint assay (>95% conversion of OS-PP-Dol to OS-NYT) were resolved by preparative HPLC to isolate the M₅GN₂-NYT glycopeptide (left). HPLC resolution of the α-1,2 mannosidase digestion products derived from M₅GN₂-NYT (right). The M₃GN₂-NYT (M3) peak is derived from isomer b, the M₄GN₂-NYT (M4) peak is derived from isomers c–h, and the M₅GN₂-NYT (M5) peak corresponds to isomer i. (C) HPLC profiles of α-1,2 mannosidase digestion products derived from M₅GN₂-NYT synthesized by the E. histolytica and T. cruzi OST. Redigestion of the M4 peak with a-1,2 mannosidase did not yield smaller products (not depicted); hence, the initial digestion had gone to completion. (D) The distribution of the three isomer classes (2, 1, or 0 α-1,2–linked mannose residues) was calculated for the total M₅GN₂-PP-Dol pool (OS) and for M₅GN₂-NYT synthesized by the S. cerevisiae (Sc), T. cruzi (Tc), E. histolytica (Eh), and T. vaginalis (Tv) OST. Values for the OS, Sc, and Tc are means of two independent experiments; error bars designate one of two independent data points. The OS values are derived from two replicates of B.

Figure 4.

Oligosaccharide donor competition assays. OST activity was assayed using a constant concentration of the acceptor tripeptide (5 μM in A, B, and F and 10 μM in C–E). (A) Glycopeptide products from assays of the T. vaginalis (a and c) or S. cerevisiae (b) OST using 1 μM G₃M₉GN₂-PP-Dol (a) or 0.6 μM G₃M₉GN₂-PP-Dol plus 1.5 μM M₅GN₂-PP-Dol (b and c) were resolved by HPLC. For clarity, column profiles have been offset on the vertical axis. (B) The T. vaginalis (squares) or S. cerevisiae (circles) OST were assayed using 1.5 μM M₅GN₂-PP-Dol and increasing concentrations of G₃M₉GN₂-PP-Dol. Glycopeptide products were resolved by HPLC to determine the percentage of M₅GN₂-NYT. The dashed line indicates the composition (in percentage of M₅GN₂-PP-Dol) of the donor substrate mixtures. (C–E) Purified S. cerevisiae (Sc) or detergent extracts of C. neoformans (Cn), E. histolytica (Eh), T. vaginalis (Tv), or T. cruzi (Tc) membranes were assayed using the following donor substrate mixtures: 1 μM G₃M₉GN₂-PP-Dol + 1 μM M₅GN₂-PP-Dol (C), 1 μM G₃M₉GN₂-PP-Dol + 1 μM M₉GN₂-PP-Dol (D), and 1 μM M₉GN₂-PP-Dol + 1 μM M₅GN₂-PP-Dol (E). Glycopeptides were resolved by HPLC to determine product composition. (F) The S. cerevisiae and T. vaginalis OST were assayed using the following mixture: (G₂M₉GN₂-PP-Dol/G₁M₉GN₂-PP-Dol/M₉GN₂-PP-Dol/M₅GN₂-PP-Dol, 35:6:2:57). Glycopeptides were resolved by HPLC (top). The M₅GN₂-NYT (M5), G₁M₉GN₂-NYT (G1), and G₂M₉GN₂-NYT (G2) peaks are labeled. The T. vaginalis OST (closed bars), but not the yeast OST (open bars), shows reduced utilization of G₂M₉GN₂-PP-Dol and G₁M₉GN₂-PP-Dol relative to M₅GN₂-PP-Dol (bottom).

Figure 5.

Kinetic parameters for the oligosaccharide donor substrate. OST activity was assayed using a constant concentration of the acceptor tripeptide substrate (10 μM in A, B, and D and 15 μM in C) and variable concentrations of M₅GN₂-PP-Dol (A–C) or M₉GN₂-PP-Dol (D). (A, C, and D) Lineweaver-Burk plots (1/OST activity versus 1/[OS-PP-Dol]) for the T. vaginalis (A; K_m = 0.22 μM), E. histolytica (C; K_m = 0.72 μM), or T. cruzi (D; K_m = 0.49 μM) OST were linear. (B) An Eadie-Hofstee plot (OST activity vs. OST activity/[OS-PP-Dol]) for the T. vaginalis (K_m = 0.19 μM) OST was linear. The inset shows an Eadie-Hofstee plot for the S. cerevisiae OST using M₅GN₂-PP-Dol as the donor substrate.

Figure 6.

Kinetic parameters for the tripeptide acceptor substrate. OST activity was assayed using a constant concentration of the OS-PP-Dol donor (1 μM M₅GN₂-PP-Dol or G₃M₉GN₂-PP-Dol in A and 0.8 μM M₅GN₂-PP-Dol or M₉GN₂-PP-Dol in B) and increasing concentrations of acceptor tripeptide substrate. (A) Lineweaver-Burk plots for the T. vaginalis OST yielded apparent K_m values of 17 μM (M₅GN₂-PP-Dol donor) and 53 μM (G₃M₉GN₂-PP-Dol) for the acceptor tripeptide. (B) Lineweaver-Burk plots for the T. cruzi OST yielded apparent K_m values of 2.3 μM (M₉GN₂-PP-Dol donor) and 6.9 μM (M₅GN₂-PP-Dol) for the acceptor tripeptide.