Snake venom
Naja nigricollis venom (NNV) from Tanzanian specimens was purchased in lyophilised form from Latoxan, France (cn. L1327B). Venom was reconstituted in HMI-9 media without fetal calf serum (FCS) when used in trypanosome cell culture.
Venom fractionation and proteomics characterisation
Whole venom fractionation was performed by reversed-phase high performance liquid chromatography (RP-HPLC) (Agilent 1200) on a C18 column (250 × 4.6 mm, 5 μm particle; Teknokroma), using HPLC Chromeleon software35. 2 mg/mL whole venom in phosphate-buffered saline (PBS: 137 mM NaCl, 3 mM KCl, 8 mM Na2HPO4.2H2O, 1.4 mM KH2PO4) was loaded onto the column and run through an acetonitrile/trifluoracetic acid gradient for elution. Then solvent was evaporated using a speedvac. Finally, equivalent fractions from different rounds were pooled together in HMI-9 media without FCS. Fraction were analysed by SDS-PAGE (7%) and silver stained36.
20 µg of fraction 19 were reduced with 5 mM Tris(2-carboxyethyl)phosphine (646,547, Sigma Aldrich) for 60 min at 65 °C, and afterwards alkylated with 20 mM 2-chloroacetamide (C0267, Sigma Aldrich) for 30 min at 65 °C. Proteins were digested with trypsin (V5280, Promega) overnight, with a trypsin:protein ratio of 1:50. Desalting was performed with in-house packed C18 disk tips following standard reversed-phase SPE protocols for sample cleanup37,38. LC–MS analysis was performed with 300 ng of sample injected in a Q Exactive hybrid quadrupole-Orbitrap mass spectrometer (Thermo Scientific), using an EASY-nLC 1200 System (LC140, ThermoFisher Scientific), and an EASY-Spray column (50 cm × 75 µm ID, PepMap RSLC C18, 2 µm ES803, Thermo Scientific), with a 70 min gradient length of increasing acetonitrile percentage. The mass spectrometer was operated in data dependent acquisition mode, scan range of 300–1750 m/z with MS resolution of 70,000, AGC target of 3e6 and maximum injection time of 20 ms. MS2 scans were obtained for top 10 precursors, with first mass of 120 m/z, resolution of 17,500, AGC target of 1 × 106, maximum injection time of 60 ms, 1.6 m/z isolation window, and NCE of 25.
Data analysis was conducted with Proteome Discoverer 2.4 (Thermo Scientific), with workflows adjusted for label free quantification. The FASTA database used was extracted from the reviewed snake venom protein list from the UniProt database (taxonomy: “Serpentes (snakes) [8570]” annotation: (type: “tissue specificity” venom) AND reviewed:yes) containing 2,330 venom proteins. Full digestion with trypsin was indicated in the settings, with two maximum missed cleavages allowed, and peptide length of 6–144. Precursor mass tolerance and fragment mass tolerance were set to 10 ppm and 0.02 Da, respectively. Oxidation (methionine) and deamidation (glutamine and asparagine), and acetylation (protein N-terminus) were added as variable modifications, while carbamidomethylation (cysteine) was added as static modification. Summed peptide area abundances were used to calculate master protein abundance (i.e. the protein chosen to represent the group in the cases where peptides identified match to multiple proteins), with filtering only for high confidence peptides (FDR 0.01). Data interpretation, processing, and plot generation were performed in Python 3.6 programming language.
Trypanosome cell culture
T. brucei Lister 427 VSG121 BSF cells were maintained in culture at a density of 1 × 105 to 2 × 106 cells/mL in HMI-9 supplemented with 10% (v/v) heat-inactivated FCS, penicillin (100 U/mL), and streptomycin (10 µg/mL), and kept at 37 ºC and 5% CO2. Cell counts were carried out using a hemocytometer.
Cell line engineering
A strategy using long oligonucleotide primers was used to tag endocytic pathway components39. In brief, forward primers contained the last 80 base pairs of the 5′ UTR of the target gene and the 20 base pairs of the 5′ pPOTv7 primer binding sequence; reverse primers contained the 20 base pairs of the 3′ pPOTv7 primer binding sequence and the first 80 base pairs of the target gene including the start codon, in the reverse complement39. PCR amplification from the pPOTv7 plasmid used resulted in a product carrying blasticidin resistance and mNEON green (mNG) protein. The 5′ UTR sequences were obtained from the TREU 927 reference genome for Tb927.10.12960 (Rab5A), Tb927.11.4570 (Rab5B), Tb927.8.4620 (Rab7), and Tb927.8.4330 (Rab11). PCR products were introduced into T. brucei Lister 427 expressing VSG221 by standard procedure, and transformants were selected with 10 μg/mL of blasticidin. Expected localisation of N-terminally tagged endocytic pathway proteins was confirmed by fluorescence microscopy. Following the same method, a T. brucei BSF Lister 427 cell line expressing VSG121 mNG-SBP1 was developed by C-terminally tagging SBP1 (Tb927.9.1970) with mNG. Successful tagging was confirmed by fluorescence microscopy.
The T. brucei Lister 427 cell line expressing VSG121 GFP-GPI was engineered by introducing a transgene into the VSG121 expression site using the same approach as used to insert a second VSG40 The plasmid p4716 was cut with Acc65I and SacI to release the insert, which was then used for transfections. The sequence of the insert is available in Fig. S1.
Cell morphology analysis
The morphology of all cell lines was analysed by fluorescence microscopy using Zeiss Axioimager M1. Images were processed using Zeiss AxioVision v5.6.1, analysed by NIH Image J, and made into figures in Adobe Illustrator.
For live cell microscopy, cells were incubated with 20 µg/mL of NNV, and 1 mL of cells was harvested at 3 and 6 h post-NNV addition. The sample was centrifuged for 1 min at 3000 rpm in an Eppendorf microfuge, the supernatant was removed, and the pellet was resuspended in 10 μL HMI-9. 4 μL were pipetted onto the microscope slide and covered by a 25 × 50 mm coverslip for visualisation. Untreated cell cultures were analysed alongside as negative control. For fixed cell microscopy, the procedure was identical, but 1% formaldehyde was added to the 10 µL resuspended cell pellet prior preparing the slide. Confocal microscopy was performed on live cells using the same method described above. Both the 2-dimensional images and the Z-stacks were processed by NIH Image J.
Cell cycle analysis
1 mL of cells was harvested from the trypanosome culture and recovered by centrifugation at 10,000 rpm for 1 min. The supernatant was removed, 10 µL of HMI-9 was added, and the pellet was resuspended. 4% (w/v) paraformaldehyde was added to fix the cells. Finally, 1/100,000 Hoechst 33,342 was added. Fluorescence microscopy was then performed to visualise kinetoplast and nuclei in the cells.
Cell surface analysis
Cell surface fluorescence levels due to GFP-GPI were analysed by flow cytometry using a CytoFlex Flow Cytometer (Dept. of Pathology, University of Cambridge). T. brucei Lister 427 VSG121 (wild type) and T. brucei Lister 427 VSG121 p4716 (GFP-GPI) cells were incubated with 20 µg/mL NNV, and 1 mL of the cells was taken at 3 and 6 h post-NNV addition and transferred to a capped glass tube. Cells not subjected to NNV were also analysed in parallel. Wild type cells were used as a negative control for GFP fluorescence. The cytometer was set to record 10,000 events at 60 µL/min sample intake. A blank sample (pure water) was run between samples to avoid cell carry overs. The gating strategy employed aimed at separating single live cells from those dividing, forming clumps while dying, and debris. Data analysis was done using FlowJo software, and image processing was done in Adobe Illustrator.
AF-568-transferrin uptake assay
Endocytosis fitness was studied by performing an AF-568-transferrin uptake assays on live T. brucei Lister 427 VSG121 p4716 (GFP-GPI) cells. Bovine transferrin (Thermo Scientific #11107018) was first 2× dialysed in 1× PBS, and then fluorescently labeled with AF-568 using an Alexa Fluor™ 568 Protein Labeling Kit (Thermo Scientific #A10238), following the manufacturer’s instructions. Cells were incubated for 12 h with 20 µg/mL NNV. At 0, 0.5, 1.5, and 6 h, 1 mL samples were taken, accounting for 3 × 105 cells. Each sample was spun down at 7,000 rpm for 2 min, washed in fresh HMI-9 media without serum, and resuspended in 100 µL + 1% BSA + 2 µL 2 mM FMK24 (Mu-Phe-hPhe-FMK, Sigma #M4070), a lysosomal protease inhibitor. The sample was incubated at room temperature for 5 min. Then, 80 nM AF-568-transferrin was added to the sample and incubated at 37 ºC for 1 h. Last, the sample was centrifuged at 7,000 rpm for 2 min, and the cell pellet was resuspended in 20 µL HMI-9 without serum + 4% formaldehyde + 1/10,000 Hoechst 33342. The sample was incubated at room temperature for 10 min and then visualised by fluorescence microscopy.
SDS-PAGE and immunoblotting
Cell lysates were made by harvesting BSF cell culture. After centrifugation (3000 rpm, 10 min), the pellet was resuspended in 1 mL HMI-9 without serum and transferred into a 1.5 mL tube. The sample was centrifuged (10,000 rpm, 1 min), the supernatant was pipetted out. This step was repeated twice. Finally, the pellet was resuspended in HMI-9 + 1X SDS-PAGE sample buffer to a concentration of 108 cell equivalents/mL and incubated at 95 ºC for 5 min. Then, SDS-PAGE was carried out41. For each sample, 10 μL of cell lysate (2 × 106 cell equivalents/well) was used. Gels were stained with Coomassie, silver staining36, or transferred to a membrane. Immunoblotting was carried out as previously described42. Gel analysis was performed by NIH Image J, and the image was processed by Adobe Photoshop.
In vitro trypanosome cell killing assays
All cell lines were incubated with 20 µg/mL N. nigricollis whole venom (NNV) or with the relative concentration of each individual venom fraction. Trypanosome cell concentration at the start of the experiment was 1.5 × 105 cells/mL. Growth was measured by counting cells on a hemocytometer. Negative controls were carried out for all cell lines under conditions not containing NNV. The assays were performed in triplicates and collected data were Log10 transformed.
Protein synthesis inhibition assay
T. brucei Lister 427 VSG121 (wild type) and T. brucei Lister 427 VSG121 GFP-GPI cell lines were incubated with (i) 50 µg/mL cycloheximide (CHX); (ii) 20 µg/mL NNV; (iii) 50 µg/mL CHX + 20 µg/mL NNV; or (iv) HMI-9 media only. NNV was added one hour post CHX addition. Trypanosome cell concentration at the start of the experiments was 1.5 × 105 cells/mL. Growth was measured by counting cells on a hemocytometer at 1, 4, and 7 h post CHX addition. In addition, cell surface fluorescence was analysed by flow cytometry, and total GFP within the cell was analysed by anti-GFP western blot.
