Distribution of the type of N-glycan occupancy related to the isoelectric point (pI) of the acceptor site peptide. The predicted pI values for each sequon ±5 amino acids were binned in unit intervals from 3 to 13. For each interval, bars are shown to indicate the observed proportions of sites modified with exclusively EndoH-resistant glycans (black), EndoH-sensitive glycans (white), both EndoH-resistant and EndoH-sensitive glycans (grey).

Animal infectivity studies with STT3ARNAi and STT3BRNAi cell lines. (A) T. brucei TbSTT3A,B,C^+/−-STT3ARNAi cells or (B) TbSTT3A,B,C^+/−-STT3BRNAi cells present in the blood of mice dosed with (white bar) and without doxycycline (grey bar) for 2 days before and throughout the 3 day infection. (C) Purified sVSG221 of wild-type cells grown in culture (lanes 1–3, as a control), and TbSTT3A,B,C^+/−-STT3ARNAi plus doxycycline (lanes 4–6) and TbSTT3A,B,C^+/−-STT3ARNAi minus doxycycline (lanes 7–9) cells isolated from mouse blood, digested or not, as indicated, with EndoH or PNGaseF were subjected SDS–PAGE and western blot with anti-VSG221 antibody. (D) Purified sVSG221 of wild-type cells grown in culture (lanes 1–3, as a control), and TbSTT3A,B,C^+/−-STT3BRNAi plus doxycycline (lanes 4–6), and TbSTT3A,B,C^+/−-STT3BRNAi minus doxycycline (lanes 7–9) cells isolated from mouse blood, digested or not, as indicated, with EndoH or PNGaseF were subjected SDS–PAGE and western blot with anti-VSG221 antibody.

Correlations of N-glycan type in T. brucei and N-glycosylation efficiency in Δstt3 yeast with acceptor site pI. (A) Box plot (Tukey, 1977) of predicted pI distributions for the sequons ±5 amino acids occupied by endoH-resistant (R), mixed (M) or endoH-sensitive (S) N-glycans. (B, C) Box plots of predicted pI distributions for the sequons ± 5 amino acids exhibiting efficient (E), moderate (M) and poor (P) N-glycosylation in Δstt3 +TbSTT3C and TbSTT3B yeast cells, respectively. Note: Box plots are concise ways of illustrating a distribution of values. The black line shows the median, the box encompasses 50% of the data, the whiskers, 75% of the data and the outliers are shown as discrete points.

Mass spectrometric analyses of intact mature sVSG221 glycoforms before and after selective knockdown of TbSTT3A and TbSTT3B expression. Samples of sVSG221 from (A) wild type cells, (B) TbSTT3A,B,C^+/−, (C) TbSTT3A,B,C^+/−-STT3ARNAi minus Tet, (D) TbSTT3A,B,C^+/−-STT3ARNAi plus Tet, (E) TbSTT3A,B,C^+/−-STT3BRNAi minus Tet and (F) TbSTT3A,B,C^+/−-STT3BRNAi plus Tet were analysed by ES-MS. The spectra show the masses of the various glycoforms of the intact mature sVSG221 molecules. The inset cartoons represent our interpretation of those glycoform masses in terms of the ranges of N-glycans present at each of the two N-glycosylation sites. These assignments combine additional data from the ES-MS and ES-MS/MS analyses of Pronase glycopeptides from the same sVSG221 preparations (Supplementary Figure S3; Table SIV). In the inset cartoons, endo-H-resistant N-glycans are in white and endo-H-sensitive N-glycans are in black. Circles and squares (filled and open) represent mannose and N-acetylglucosmaine residues, respectively, and grey circles represent galactose residues.

Complementation of yeast Δstt3 by T. brucei OTases. (A) Yeast cells with (STT3) or without (Δstt3) genomic STT3, with the yeast STT3 gene encoded on a URA3 plasmid (pSTT3), and with various TbSTT3 genes (pTbSTT3A; pTbSTT3B; pTbSTT3C) or with empty vector (p) on a LEU2 plasmid were grown in serial dilution on minimal media (MM) or media containing 5-FOA for selection of ura-cells and (B) after selection were grown at different temperatures in serial dilution on MM. (C) Glycosylation occupancy of all sites measured for each strain, mapped from 100% occupancy, black, to 0% occupancy, white. Values are averages of three independent measurements. Data are shown in Tables SI and SII. (D) Proportion of threonine in the +2 position of efficiently and poorly glycosylated sequons. (E) Cross correlation of site-specific glycosylation efficiency by TbSTT3B and TbSTT3C. (F) Glycosylation of TOS1_417 and CCW14_87 sites by T. brucei STT3B and STT3C that are not used by yeast OTase. Error bars show range.

The effects of RNAi silencing of STT3B on protein glycosylation in T. brucei. (A) Northern blot of total RNA extracted from the following cell lines: wild type (lane 1), TbSTT3A,B,C^+/− (lane 2), TbSTT3A,B,C^+/−-STT3ARNAi minus Tet (lane 3), TbSTT3A,B,C^+/−-STT3ARNAi plus Tet (lane 4), TbSTT3A,B,C^+/−-STT3BRNAi minus Tet (lane 5) and TbSTT3A,B,C^+/−-STT3BRNAi plus Tet (lane 6) were probed with the STT3B probe (upper panel) and stained with ethidium bromide for rRNA as loading control (lower panel). (B) Growth of the following T. brucei cells at 37°C: wild type (closed circles), TbSTT3A,B,C^+/−-STT3BRNAi minus Tet (open squares) and TbSTT3A,B,C^+/−-STT3BRNAi plus Tet (closed triangles). (C) Cell ghosts containing ER-associated VSG221 and (D) cell surface-derived sVSG221 from the following cells: TbSTT3A,B,C^+/− (lanes 1–3), TbSTT3A,B,C^+/−-STT3BRNAi plus Tet (lanes 4–6) and TbSTT3A,B,C^+/−-STT3BRNAi minus Tet (lanes 7–9) were digested or not, as indicated, with EndoH or PNGaseF were subjected SDS–PAGE and either western blot with anti-VSG221 antibody (C) or staining with Coomassie blue (D). The cartoons above the gels in panels (C) and (D) summarise the types of VSG221 present with respect to EndoH-resistant biantennary paucimannose and complex N-glycans (grey) and EndoH-sensitive triantennary oligomannose N-glycans (black). (E) Blots of total glycoprotein extracts of the following cells: wild type (lane 1), TbSTT3A,B,C^+/− (lane 2), TbSTT3A,B,C^+/−-STT3BRNAi plus Tet (lane 3) and TbSTT3A,B,C^+/−-STT3BRNAi minus Tet (lane 4) were incubated with ConA without (upper panel) and with carbohydrate inhibitors (specificity control, middle panel) or stained with Pounceau red (loading control, lower panel).

The effects of RNAi silencing of STT3A on protein glycosylation in T. brucei. (A) Northern blot of total RNA extracted from the following cell lines: wild type (lane 1), TbSTT3A,B,C^+/− (lane 2), TbSTT3A,B,C^+/−-STT3ARNAi minus Tet (lane 3), TbSTT3A,B,C^+/−-STT3ARNAi plus Tet (lane 4), TbSTT3A,B,C^+/−-STT3BRNAi minus Tet (lane 5) and TbSTT3A,B,C^+/−-STT3BRNAi plus Tet (lane 6) were probed with the STT3A probe (upper panel) and stained with ethidium bromide for rRNA as loading control (lower panel). (B) Growth of the following T. brucei cells at 37°C: wild type (closed circles), TbSTT3A,B,C^+/−-STT3ARNAi minus Tet (open squares) and TbSTT3A,B,C^+/−-STT3ARNAi plus Tet (closed triangles). (C) Cell ghosts containing ER-associated VSG221 and (D) cell surface-derived sVSG221 from the following cells: TbSTT3A,B,C^+/− (lanes 1–3), TbSTT3A,B,C^+/−-STT3ARNAi plus Tet (lanes 4–6) and TbSTT3A,B,C^+/−-STT3ARNAi minus Tet (lanes 7–9) were digested or not, as indicated, with EndoH or PNGaseF were subjected SDS–PAGE and either western blot with anti-VSG221 antibody (C) or staining with Coomassie blue (D). The cartoons above the gels in panels (C) and (D) summarise the types of VSG221 present with respect to EndoH-resistant biantennary paucimannose and complex N-glycans (grey) and EndoH-sensitive triantennary oligomannose N-glycans (black). (E) Blots of total glycoprotein extracts of the following cells: wild type (lane 1), TbSTT3A,B,C^+/− (lane 2), TbSTT3A,B,C^+/−-STT3ARNAi minus Tet (lane 3) and TbSTT3A,B,C^+/−-STT3ARNAi plus Tet (lane 4) were incubated with ricin without (upper panel) and with carbohydrate inhibitors (specificity control, middle panel) or stained with Pounceau red (loading control, lower panel).