A Sweet Galactose Transfer: Metabolic Oligosaccharide Engineering as a Tool To Study Glycans in Plasmodium Infection

Abstract The introduction of chemical reporter groups into glycan structures through metabolic oligosaccharide engineering (MOE) followed by bio‐orthogonal ligation is an important tool to study glycosylation. We show the incorporation of synthetic galactose derivatives that bear terminal alkene groups in hepatic cells, with and without infection by Plasmodium berghei parasites, the causative agent of malaria. Additionally, we demonstrated the contribution of GLUT1 to the transport of these galactose derivatives, and observed a consistent increase in the uptake of these compounds going from naïve to P. berghei‐infected cells. Finally, we used MOE to study the interplay between Plasmodium parasites and their mosquito hosts, to reveal a possible transfer of galactose building blocks from the latter to the former. This strategy has the potential to provide new insights into Plasmodium glycobiology as well as for the identification and characterization of key glycan structures for further vaccine development.


Supporting Information
Quantification of Mean fluorescence intensity by confocal point-scanning microscopy after metabolic incorporation of 2b and 5 into cell membrane glycans of HepG2 cells, 9 was used as a negative control. b.2. Ratio of mean fluorescence intensity against control 9 after metabolic incorporation of 2b and 5 in HepG2 cells. Representative data from one out of three experiments, analysis of five to six picture per condition and 5-6 cells per picture, Two-tailed Mann-Whitney for non-parametric distribution, D'Agostino and Pearson omnibus normality test was performed for each data set. Figure S4: Metabolic incorporation of 2b into HepG2 cells, labeling with 6-methyl-tetrazine-PEG4-biotin and Alexa-Flour-568-streptavidin (red). Co-staining of the cell membrane with CellMask Deep Red Plasma membrane stain (green). Nuclei were stained with Hoechst 33342, 9 was used as negative control. Figure S5: Incorporation of 13 and 3b in HepG2 cells, labelling with Alexa-Fluor-568-streptavidin (red) and Hoechst (blue). I and III, as well as II and IV represent two independent experiments. Figure S6: a.1/2. Gating strategy during the analysis of metabolic incorporation of 2b in HepG2 cell membrane glycans by flow cytometry, selection of live cells, single cells and infected vs. non-infected cells. b. Gating strategy during the analysis of metabolic incorporation of 2b in HepG2 cell membrane glycans by imaging flow cytometry, selection of single cells (I), single cells in focus (II), separation of infected and non-infected cells based on the intensity of GFP (Ch02, III), Fluorescence intensity resulting of the incorporation of 2b in non-infected (IV) and infected cells (V). c.1. representative pictures acquired in Amnis ImageStreamX MarkII, HepG2 cells after metabolic incorporation of 2b, infected, non_infected and naïve. c.2. representative pictures acquired in Amnis ImageStreamX MarkII, HepG2 cells after metabolic incorporation of contol 9, infected, non-infected and naïve. Figure S7: a. Quantification of fluorescence intensity resulting from incorporated galactose derivative 2b by confocal pointscanning microscopy, Data representative from one experiment out of three, each data point represents the mean fluorescence intensity of all single cells within one picture, two-tailed Mann-Whitney test. b. Quantification of fluorescence intensity resulting from incorporated galactose derivative 2b by flow cytometry, Combined data from three independent experiments, each data point represents the median intensity of 2000-3000 gated single cells, two-tailed Mann-Whitney test. Figure S8: a. Metabolic incorporation of 2b into HepG2 cell membrane glycans, without and in presence of inhibitors 10-12.
Pentaacetyl galactose 9 was used as negative control. b. Half-offset histograms of fluorescence intensity in channel PE-CF594-A, normalized mode, after metabolic incorporation of derivative 2b in HepG2 cells, without inhibitor (blue), with increasing concentrations of WZB117 10, 10 µM (orange), 20 µM (green), 30 µM (dark green). Pentaacetyl galactose 9 was used as negative control. Figure S9: a. Shape parameters of Plasmodium berghei parasites after feeding with 3b or control 13, acquisition in Amnis ImageStreamX MarkII, two-tailed Mann-Whitney. b.1. Determination of the parasite size/area of representative pictures acquired with Amnis ImageStreamX MarkII, measurements were performed using ImageJ software package, n = 50, two-tailed Mann-Whitney. b.2. Determination of the parasite size/area of all acquired pictures with Amnis ImageStreamX MarkII, n=2000-3000, two-tailed Mann-Whitney. Figure S11: Coomassie blue staining of whole protein cell lysate and protein samples bound to streptavidin magnetic beads from HepG2 cells, cells were grown for 72h with 100 µM of 9, 2b or 5. Figure S12: HepG2 cells grown with 500 µM 9 or 2b, for a inhibition experiment both 9 and 2b were added simultaneously. b. HepG2 cells were grown with 100 µM of 9 or 2b, Benzyl-2-acetamido-2-deoxy-galactopyranse (BADG) was added with 100 µM after 24h to a culture with 2b and the cells were analysed after a total culture time of 72h. IEDDA reaction was performed with 6methyl-tetrazine-peg4-biotin, followed by staining with Alexa-Fluor-568-streptavidin (red) and Hoechst (blue).

General Procedures
The used reagents were purchased from Alfa Aesar, Carbosynth Limited, Fisher Scientific and Sigma Aldrich and were used without further purification. Purification of the compounds was performed by chromatography using Silica Gel 60 (mesh 230-400) from Material Harvest. Thin layer chromatography (TLC) was carried out on silica gel coated aluminium plates (60 F254,Merck) and the reactions were visualized with 5% sulfuric acid in ethanol and UV light (λ =254 nm).
Proton ( 1 H NMR), carbon ( 13 C NMR) nuclear magnetic resonance spectra were recorded on a Bruker 500 MHz DCM Cryoprobe or 400 MHz DPX-400 Dual spectrometer. All spectra were fully assigned using COESY, HSQC and HMBC, the chemical shifts were quoted on the δ scale in ppm and the solvent peak (CDCl3: 1 H = 7.26 ppm, 13 C = 77.16 ppm, D2O: 1 H = 4.79 ppm) was used as internal standard. Coupling constants J were reported in Hz, using the following splitting abbreviations: s = singlet, d = duplet, t = triplet, dd = duplet from duplet, m = multiplet. High resolution mass spectrometry (HRMS) were received from a Thermo Finnigan Orbitrap Classic using positive ion electronstpray ionization (ESI) for essential compounds.

Kinetic Studies
The kinetic studies were performed in aqueous PBS buffer at pH = 7.4 as solvent. The reaction progress was monitored by following the decrease in tetrazine absorption at 530 nm. The optimal tetrazine concentration was determined by an absorbance screen, determining a concentration of 0.6 mM 6-methyl-tetrzaine-amine as optimal concentration per well. A stock solution of 20 mM was prepared of every de-acetylated galactose derivative, from which further dilutions of 16 mM, 12 mM, 8 mM and 4 mM were prepared. The solutions of galactose derivatives and 6-methyl-tetrazine-amine were mixed to a final volume of 100 µL in 96-well plates and the decline in absorption at 530 nm was followed for 16 h in a microplate reader at 37 °C. The pseudo-first order rate constant kobs was calculated for each concentration using GraphPad Prism 6.01 with an exponential decay function. The corresponding second order rate constant k2 for each galactose derivative was calculated based on the concentration dependent values for kobs and the resulting linear function.

Cell Culture
All cell lines were maintained in a humidified incubator at 37°C under 5% CO2 and split before reaching confluence using TrypLE TM Express. All cell lines were grown in DMEM medium (high glucose) supplemented with 10% heat-inactivated FBS, 2 mM GlutaMAX TM , 10 mM HEPES, 1% NEAA, 100 units/mL penicillin and 100 µg/mL streptomycine, further named complete medium (cDMEM). All reagents were bought from Gibco, Life Technologies (USA).

Cell toxicity
The toxicity of the galactose derivatives was assessed using a CellTiter-Blue R Cell Viability Assay (Promega, USA). In this approach the conversion of the dye reaszurine into the fluorescent resorufin product by metabolically active cells is measured. Cells were seeded at a concentration of 10000 cells/ well (100 µl

Metabolic Labeling in Huh7 and HepG2 cells
Cells were seeded with a density of 15000 cells/well (300 µL) on glass coverslips in 24-well plates and allowed to adhere for 24 h. Following this, the cell culture medium was exchanged to complete medium supplemented with 100 µM of the corresponding galactose derivative and the cells were cultured for 72 h. After this, the medium was removed and complete medium containing 25 µg/mL streptavidine (from 1 mg/mL stock in water) was added for 40 min to block endogenous biotin. The labeling experiments were performed in the same way using acetyl protected galactose derivatives or deprotected derivatives.

Metabolic labeling of HepG2 cells in presence of inhibitors for GLUT1
HepG2 cells were seeded in 24-well plates, either on glass coverslips (15000/well) for microscopy analysis or on the plain plate (50000/well) for analysis by flow cytometry. The cells were allowed to adhere for 24h, before the medium was changed to complete medium containing 100 µM of the For microscopy analysis, the medium was removed and complete medium containing 25 µg/mL streptavidine (from 1 mg/mL stock in water) was added for 40 min to block endogenous biotin. 6.6 µg/mL Alexa-Fluor-568-streptavidin in PBS + 5% FBS (100 µL per cell pellet) was added for 20 min at RT, followed by washing the cell pellet with PBS + 5% FBS (3x100 µL per cell pellet).
The cells were fixed with 4% PFA solution (8 min, RT) and resuspended in 300 µL PBS + 5% FBS for analysis. The cells were analyzed in a BD LSRFortessa X-20 cell analyzer, using a 561 nm laser (Alexa-Fluor-568). The results were analyzed using FACSDiva, FLOWJOW and GraphPad Prism software packages.

Metabolic labeling of HepG2 cells for cell lysis and pull-down
HepG2 cells were seeded in T75 flasks (1.5*106 cells/flask) in DMEM medium (high glucose), supplemented with 10% heat-inactivated FBS, 1% GlutaMAX, 1% HEPES, 1% NEAA and 100 units/mL penicillin and streptomycine, further named complete medium (cDMEM). The cells were allowed to attache at least 12 h, before the medium was changed to cDMEM supplemented with 100 µM penta-acetyl-galactose (Ac5Gal), 1,3,4,6-tetraacetyl-2-O pentenyl-galactose (2OPent) or 1,2,3,4-tetraacetyl-6 O pentenyl-galactose (6OPent). The cells were grown for 72 h, before the media was removed and the cells were gently flushed with PBS (1x). The cells were harvested using 1 mL of lysis buffer (20 mM Tris pH = 7.6, 300 mM NaCl, 1% TritonX100, 5% glycerol, 1 mM EDTA, 1:10 protease inhibitor) and incubated on ice for 30 min. The cell lysate was obtained after centrifugation at 14000 g for 10 min at 4°C. 200 µL of whole cell lysate were treated with 50 µM of 6 methyl tetrazine peg4 bioting and the labelling reaction was performed at room temperature overnight. To each sample, 100 µl of a suspension of streptavidin magnetic beads (Pierce Streptavidin Magnetic Beads, 10 mg/mL) was added and the binding was performed at room temperature for 1 h. The supernatant was collected and the magnetic beads were washed twice with PBS (1x). The magnetic beads were suspended in 50 µL 4xSDS loading buffer with 100 µM DTT and boiled for 10 min at 90 °C. The samples were resolved in a 12% SDS gel.

Competition experiment and Inhibition of O-glycosylation
HepG2 cells were seeded on glass coverslips in 24-well plates (15000 cells/well) and were allowed to attach. On the next day, the media was changed to cDMEM supplemented with an indicated concentration of 2b, 5 or 9. Oil objective was used for acquisition and the pictures were analysed by using ImageJ 1.49v software package. Pictures were acquired from different regions on the coverslips and the fluorescence intensity resulting from incorporated galactose derivatives was quantified by assigning ROIs to single cells.

Release of cell surface glycans
Huh7 cells were seeded in 6-well plates (1x106 cells/ well) in cDMEM and were allowed to attach overnight. On the next day, the media was changed to cDMEM supplemented with 100 µM of 9 or 2b and the cells were cultured for 72 h. To analyse possible cell surface glycans, the cells were harvested in cDMEM and washed with 500 µL PBS (3x, 600 rpm, 10 min @ 4 °C). The cells were suspended in 500 µL PBS and 10 µL of Trypsin solution (20 µg trypsin in 100 µL acetic acid 50 mM) was added. The reaction was incubated at 37 °C for 15 min, before the supernatant was separated (15000 rpm, 15 min @ 4 °C). The supernatant was collected and trypsin was heatinactivated @ 98°C for 5 min.
A sample of 1 µL was diluted with 9 µL MilliQ water for analysis in an Acquity LC/MS system.
It was possible to detect a specific peak at 553nm in cells treated with compound 2b for 72 h.

Infection Studies
Sporozoites from Plasmodium berghei, expressing green fluorescent protein (GFP), were dissected in non-supplemented DMEM medium from the salivary glands of infected female A. stephensi mosquitoes, bred at Instituto de Medicina Molecular. In general, HepG2 cells were seeded one day prior to infection to adhere in the cell culture plates.

Infection of HepG2 cells and analysis by confocal microscopy
HepG2 cells were seeded on glass coverslips (15000/well, 300 µL) in 24-well plates and allowed to adhere for 24h in complete medium. On the day of infection, the medium was aspirated and The pictures were acquired with a 63x Plan-Apochromat Oil objective and processed with ImageJ 1.49v software to remove background noise. Representative images were chosen from 5 different experiments. Quantification of the fluorescence intensity resulting from the incorporated galactose derivatives and labeling with 6-methyl-tetrazine-peg4-biotin and Alexa-Fluor-568-streptavidine was done using ImageJ 1.49v software by selecting the individual cells per picture as region of interest and comparing the intensity. The values were presented as ratio to the intensity of the corresponding negative control.

Infection of HepG2 cells and analysis by flow cytometry and imaging flow cytometry
HepG2 cells were seeded (50000/well, 300 µL) in 24-well plates and allowed to adhere for 24h in complete medium. On the day of infection, the medium was aspirated and exchanged to complete medium supplemented with 100 µM of the corresponding galactose derivative (from 100 mM stock in DMSO) or the control sugar and Fungizone (1:200, from commercial stock). GFP-expressing sporozoites from Plasmodium berghei were dissected in DMEM from salivary glands of infected female A. stephensi mosquitoes and 60000 sporozoites per well were added directly to the cells.
The plate was centrifuged for 4 min at 200g to ensure simultaneous settling of the sporozoites on the cells. After 2 hpi, the medium was removed carefully and the cells were rinsed with PBS (3x 200 µL/well) to remove mosquito host debris, before adding again complete medium containing 100 µm of the galactose derivative or the control sugar and fungizone. The cells were grown in a humidified incubator at 37 °C with 5% CO2 until 48 hpi. The medium was aspirated and complete medium containing 25 µg/mL streptavidine (from 1 mg/mL stock in water) was added for 40 min at 37 °C to block endogenous biotin. The cells were rinsed with PBS (3 x 200 µL/well) and complete medium containing 200 µM 6-methyl-tetrazine-peg4-biotin was added for 5 h at 37 °C.
The medium was removed and the cells were detached with 20 mM EDTA solution (5 min, 37°C), collected by centrifugation (2000 g, 5 min) and washed with PBS + 5% FBS (100 µl per cell pellet).
The cells were fixed with 4% PFA solution (8 min, RT) and resuspended in 300 µL PBS + 5% FBS for analysis. The cells were analyzed in a BD LSRFortessa X-20 cell analyzer, using a 488 nm laser (GFP) and 561 nm laser (Alexa-Fluor-568). The results were analyzed using FACSDiva, FLOWJOW and GraphPad Prism software packages. The data shown result from a pool of at least 3 different experiments. The mean or median fluorescence intensities resulting from the incorporated unnatural galactose derivative (Alexa-Fluor-568) were presented as ratio to the corresponding negative control.
For analysis using Amnis ImageStreamX imaging flow cytometer, the cells were grown, infected and stained in the same way as described above. The final volume of the cells in PBS + 5% FBS was decreased to 100 µL. A 488nm laser (GFP) and 561 nm laser (Alexa-Fluor-568) were used for the analysis, as well as bright field light. A typical acquisition setting is represented below, starting from the gating on single cells based on their bright field aspect ratio intensity and area. These single cells were restricted to those being in the focus of the camera, followed by distinguishing between non-infected and infected cells based on the intensity of the GFP signal. Finally, the mean and median fluorescence intensity resulting from the incorporated galactose derivative (Alexa-