Fig 1.
Summary of the similarities and differences between general eukaryote protein N-glycosylation and processing systems (A) and those of bloodstream-form T. brucei (B).
Abbreviations: LLO, lipid-linked oligosaccharide; OST, oligosaccharyltransferase; pNAL, poly-N-acetyllactosamine; UGGT, UDP-Glc: glycoprotein glucosyltransferase. Adapted from [45], created with BioRender.com.
Fig 2.
Generation of a bloodstream form TbGT8 conditional null mutant in a TbGT10 null mutant background.
A. Gene replacement strategy using a previously described TbGT10 null cell line (TbGT10-/-) [17] which constitutively expresses diCre recombinase from the ribosomal small subunit (SSU). This was used to generate a TbGT8 conditional knockout cell line lacking TbGT10 (referred to in the text as TbGT10-/-/TbGT8Flox/- conditional null). The first allele of TbGT8 was replaced using a blasticidin resistance gene (BSDr) to generate TbGT10-/-/TbGT8+/- mutants. To make a conditional null mutant of TbGT8 the transgene was introduced at the second allele using a dicistronic construct containing a hygromycin resistance (HYG) Thymidine kinase (TK) fusion flanked by loxP (black arrows) sites (TbGT8Flox). B. PCR was performed using gDNA harvested 72 h after rapamycin treatment (+ Rap) and oligonucleotide primers SMD357 and 358 (open arrows) that anneal outside of the homologous recombination site. Expected amplicon sizes are underlined. Resolution of PCR products by agarose gel electrophoresis confirms the replacement of endogenous TbGT8 by TbGT8Flox and the excision of TbGT8Flox upon rapamycin treatment. Two TbGT10-/-/TbGT8Flox/- clones (1.1 and 1.2) are shown. Wild-type (WT) and TbGT10-/-/TbGT8+/- mutants were included as controls. C. Growth of TbGT10-/-/TbGT8Flox/- conditional null mutant cells cultured with (+Rap) or without (−Rap) 100 nM rapamycin for 3 days. Cells were seeded in the presence or absence of hygromycin (HYG) to assess floxed gene loss by hygromycin sensitivity. Data are means ± SD (n = 4 clones with 3 technical replicates per n) D. Growth kinetics of TbGT10-/-/TbGT8Flox/- conditional null mutant cells grown with or without (control) 100 nM rapamycin for 7 days. Cells were seeded at 2 × 103 cells/ml and diluted to 2 × 104 cells/ml every 2 days from day 3. Cell density was determined by counting at 24 h intervals. Data are means ± SD (n = 4 clones with 3 technical replicates each).
Fig 3.
A stable TbGT10/TbGT8 double KO clone is viable in mice.
A. PCR was performed using gDNA extracted from wild-type (WT), TbGT10-/-/TbGT8Flox/- conditional null mutant and TbGT10-/-/TbGT8-/- double null mutant (clone 1.1) cells and oligonucleotide primers SMD357 and 358 (open arrows) that amplify TbGT8. Expected amplicon sizes are underlined. PCR amplification of TbGT2b and TbGT10 coding sequences was carried out in parallel as controls. Resolution of PCR products by agarose gel electrophoresis confirms the loss of both TbGT10 and TbGT8 from the TbGT10-/-/TbGT8-/- double null mutant. B. Balb/c mice were infected with 2 x105 WT and TbGT10-/-/TbGT8-/- double null mutant cells and viable cells were counted 3 days post infection. No significant difference was observed between WT and TbGT10-/-/TbGT8-/- double null mutant cells. Data are means ± SD (n = 4–5 mice with 3 technical replicates per n).
Fig 4.
Reactivity of TbGT mutant cell lysates with RCA lectin.
A. Schematic shows the predicted outcome of double gene deletion TbGT10 and TbGT8 on BSF complex N-glycan elaboration based on previous analysis of their glycosyltransferase activity [8,13]. GlcNAc: βGal transferase activities due to TbGT8 and TbGT10 deletion prevents glycolinkages 16 and 17, respectively, resulting in deficiencies in the synthesis of 4GlcNAcβ1-3(-4GlcNAcβ1–6)Galβ1-branch points in N-linked glycans. Thus, loss of either glycosyltransferase activity similarly prevents branch point formation, and loss of both is predicted to inhibit the synthesis of linear pNAL, leading to the synthesis of N-glycans terminating in RCA-reactive β4-Galactose. B. Lysates of WT, TbGT10-/-/TbGT8Flox/- conditional null mutant and TbGT10-/-/TbGT8-/- double null mutants were subjected to SDS-PAGE and transferred to nitrocellulose membrane in duplicate. Upper panel, membranes were incubated with biotinylated RCA without (RCA—sugar) or with pre-incubation with 30 mg/ml galactose and lactose (RCA + sugar) as a binding specificity control. Lower panels, equal loading and transfer are demonstrated by Ponceau S staining. The molecular weight markers are indicated on the left. Increased detection of a discrete ~60 kDa product by RCA was observed in TbGT10-/-/TbGT8-/- double null mutant lysates (marked by *). C. Lysates of WT, TbGT10-/-/TbGT8Flox/- conditional null and TbGT10-/-/TbGT8-/- double null mutants were subjected to SDS-PAGE and transferred to nitrocellulose membrane in duplicate. Upper panel, membranes were incubated with biotinylated RCA without (RCA—sugar) or with pre-incubation with 30 mg/ml galactose and lactose (RCA + sugar) as a binding specificity control. Lower panels, equal loading and transfer are demonstrated by Ponceau S staining. The molecular weight markers are indicated on the left. An enhanced signal for a distinct ~60 kDa species (marked by *) in TbGT10-/-/TbGT8-/- double null mutant cell lysates was again observed.
Table 1.
List of identified proteins from the RCA reactive band near 60 kDa.
emPAI values scores are indicated for WT, TbGT10-/-/TbGT8Flox/- conditional null and TbGT10-/-/TbGT8-/- double null mutant cells. Non-detected (nd) proteins are indicated. Prediction of a Signal Peptide (SP) by SignalP analysis indicated. N-glycosylation site prediction was performed [27]. Expanded sequence analysis of ESAG2 and CBP1B (bold) are shown in S4 Fig.
Fig 5.
TbGT knockout mutants are deficient in TL binding and uptake.
A. Whole cell lysates (107 cell equivalents per condition) from WT, TbGT10-/-/TbGT8Flox/- conditional null mutant and TbGT10-/-/TbGT8-/- double null mutant cell lines were subjected to SDS-PAGE, transferred to PVDF membranes, and probed with TL::Biotin only (TL, top left) or with TL::Biotin pre-incubated with chitin hydrolysate at 1:10 dilution (TL/chitin, top right) as competitive inhibitor. Membranes were counter-stained with Ponceau S as loading control (bottom panels). B. Binding and internalisation of TL::Dylight488, acting as a surrogate for receptor-mediated endocytic cargoes, were measured by flow cytometry. Cells were incubated with TL::Dylight488 at 4°C for 5 min and then transferred to 14 or 37°C for 10 min to activate endocytosis. Histogram profiles show fluorescent intensities distributions of 10,000 WT, TbGT10-/-/TbGT8Flox/- conditional null mutant and TbGT10-/-/TbGT8-/- double null mutant cells. C. Bar chart shows median fluorescence intensities (MFI, arbitrary units) for each population of cells, presented as means ± SD, n = 3 biological replicates. Unpaired t-test with Welch’s correction indicated significant differences (*P < 0.0339, **P < 0.008) between WT and each of the mutant cell lines at the specified temperatures. Chitin hydrolysate at 1:10, 1:100 and 1:000 dilution (inhibitor) was used as specificity control for each temperature and the MFI values were < 22 arbitrary units.
Fig 6.
Methylation linkage analysis of glycopeptides from BSF wild type (WT) and TbGT8-/-/TbGT10-/-double null (dKO) mutants confirm reduced linear pNAL and 3,6-GlcNAc branch synthesis.
Methylation linkage analysis of glycopeptides from BSF wild type (WT, black bars) and TbGT10-/-/TbGT8-/-double null (dKO, grey bars) mutants confirms reduced linear pNAL and 3,6-GlcNAc branch synthesis. The bar graph shows the percentage of partially methylated alditol acetate (PMAA) derivatives, compared to wild type PMAAs. All the PMAAs counts were first normalised to the derivative for non-reducing terminal-mannose (t-Man). The PMAAs were analysed using selected ion monitoring (SIM) on GC-MS, where characteristic PMAA fragment ions were used to collect and extract the data. (For WT, n = 6, 3 biological replicates with 2 technical replicates and for dKO, n = 4, 2 biological replicates with 2 technical replicates). The derivatives measured and the respective characteristic ions used for analysis are mentioned in Table 2.
Table 2.
GC-MS methylation linkage analysis of glycopeptides from BSF wild type (WT) and TbGT8-/-/TbGT10-/-double null (dKO) mutants.
The PMAAs derivatives measured and the respective characteristic ions used for analysis in Fig 6.
Fig 7.
Affinity purification of TfR, Endo H digestion, and TL blotting.
TfR was immunoprecipitated from whole cell lysates (107 cell equivalents/lane) of WT, TbGT10-/-/TbGT8Flox/- conditional null mutant, and TbGT10-/-/TbGT8-/- double null mutant cell lines. The affinity-purified precipitates were treated with (+) or without (-) Endo H to digest oligomannose glycoconjugates, and matched samples were subjected to blotting with TL::biotin only (top panel, TL 10 blots) or TL::Biotin pre-incubated with competing chitin hydrolysate at 1:10 dilution (bottom panel, TL/chitin, 10 blots). The primary blots were re-probed with anti-TfR (aTfR, 20) without stripping to ensure efficient pulldowns and Endo H digestion. Mobilities of undigested ESAG6, ESAG7 (Endo H -) and Endo H de-glycosylated ESAG6, ESAG7 (dESAG6 and dESAG7, Endo H +) are shown on the right of each blot. Positions of molecular weight markers are shown on the left. Asterisks indicate the position of Endo H protein, while the arrowhead indicates mobility of a non-specific cross-reactive band. Data presented are representative of three independent biological replicates. Panels were digitally separated after image processing for clarity of presentation.
Fig 8.
Effect of TbGT knockout on TfR processing, endocytosis, and localisation.
A. Trypanosome cell extracts were prepared from WT, TbGT10-/-/TbGT8Flox/- conditional null mutant and TbGT10-/-/TbGT8-/- double null mutant cell lines. 107 cell equivalents per condition were affinity-precipitated with either anti-TfR (aTfR) or holo-transferrin (Tf beads), separated by SDS-PAGE, transferred to PVDF membranes, and blotted with aTfR antibodies. The mobilities of fully glycosylated mature ESAG6 (mESAG6), unprocessed immature ESAG6 (iESAG6), and ESAG7 are shown on the left. Molecular weight markers are shown on the right. B. Live cells from WT and KO mutants were incubated with Alexa488-conjugated holotransferrin (Tf::488) at 4, 14, and 37°C for 30 min. Unbound Tf:488 was washed, the cells were formaldehyde fixed, and analysed by flow cytometry. Bar chart shows the median fluorescence intensity from 10,000 cells per condition. The difference in Tf::488 binding at 4°C or uptake at 14 or 37°C between WT and TbGT10-/-/TbGT8Flox/- conditional null mutant or TbGT10-/-/TbGT8-/- double null mutant cells was not statistically significant (n.s.) as determined by unpaired t-tests. Error bars represent means ± SD, n = 3 biological replicates. C. To determine the kinetics of endocytosis, live cells were incubated with Tf::488 at indicated time intervals (x-axis), washed and analysed flow cytometry without fixing, to measure the rate of receptor-mediated endocytosis of Tf. Error bars represent means ± SD, n = 3 biological replicates. D. Binding of FITC-holotransferrin was measured in WT, TbGT10 KO, TbGT11 (TbGNTI) KO, and TbGT15 (TbGNTII) KO mutants (sites/cell). Data are means ± SD (n = 3 biological replicates). E. Localisation of TfR in WT, TbGT10-/-/TbGT8Flox/- conditional null mutant and TbGT10-/-/TbGT8-/- double null mutant cells. Microscopy was performed on formaldehyde fixed, permeabilised cells stained with DAPI (magenta) and anti-TfR (αTfR, green). Two representative cells are presented for each cell line. Arrowhead indicates flagellar pocket localisation of TfR. Scale-bar: 5 μM. Additional images of TfR localisation are presented in supplementary figure (S9 Fig).
Fig 9.
Effect of chitin hydrolysate on TL, Tf and Dextran uptake.
Live, wild-type bloodstream form cells were incubated with (A) Alexa647-conjugated tomato lectin (TL::647), (B) Alexa647-conjugated transferrin (Tf::647), or with (C) Alexa488::Dextran at 37°C for 10 min to allow uptake. This was done in the absence (- CH, unstained, grey) or presence of varying dilutions (1:10 to 1:1000, black) of Chitin hydrolysate (+) as indicated in the plots (insets). The actual concentration of CH used is unknown, as this information is proprietary to the manufacturer of the product (Vector Laboratories, 2BScientific, UK). Histogram profiles show fluorescent intensities distributions of 10,000 cells per condition analysed by flow cytometry. Inhibition of TL and Tf uptake by chitin hydrolysate at a 1:10 dilution was consistent across three biological replicates in Figs 5 and 7, respectively. The Dextran experiment was performed once as control for fluid phase endocytosis.