Edited by: Moo-Seung Lee, Korea Research Institute of Bioscience and Biotechnology (KRIBB), South Korea
Reviewed by: Cheleste Thorpe, Tufts University School of Medicine, United States; Phillip I. Tarr, Washington University School of Medicine in St. Louis, United States
This article was submitted to Molecular Bacterial Pathogenesis, a section of the journal Frontiers in Cellular and Infection Microbiology
This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
Shiga toxin is the main virulence factor of non-invasive enterohemorrhagic
Shiga toxin 2 (Stx2) is an AB5 toxin that consists of a pentameric B-subunit, which mediates binding, and an enzymatically active A-subunit (Endo et al.,
Stx2 is the main virulence factor of enterohemorrhagic
Stx2-positive microvesicles were taken up in murine glomerular endothelium in the EHEC infection model (Ståhl et al.,
To this end we investigated the effect of Stx2 delivered within microvesicles on Gb3-positive and Gb3-negative cells. We used Chinese hamster ovary (CHO) cells that are inherently Gb3-negative and generated Gb3-positive transfected CHO cells. We decreased Gb3 synthesis in HeLa cells using a glycosylceramide synthase inhibitor and also used DLD-1 human intestinal cells, naturally lacking Gb3. Cells were incubated with Gb3-positive Stx2-positive microvesicles. The intracellular transport route of Stx2 delivered via microvesicles was investigated. The specific goal was to determine if the presence of Gb3 in recipient cells was essential for cytotoxicity of Stx2 delivered within microvesicles.
Stx2a was purchased from Phoenix Lab (Tufts Medical Center, Boston, MA). Lipoplysaccharide (LPS) contamination was measured using the Limulus Amebocyte Lysate method (Thermo Fisher Scientific, Rockford, IL) detecting minute amounts (183.4 ng/mg toxin). For certain experiments Stx2 was labeled with Alexa Fluor 488 or Alexa Fluor 555 using the Microscale Protein Labeling Kit (both from Thermo Fisher Scientific) according to the manufacturer's instructions. The toxic activity of Stx2 was retained after labeling with fluorescent dyes, as determined by the cell metabolism assay described below.
Human whole blood was drawn from healthy volunteers (
The Stx2 content of the microvesicle suspension and the supernatant that was removed from the last wash from the microvesicle supernatant was determined by ELISA. White 96-well Maxisorp plates (Nunc, Roskilde, Denmark) were coated with mouse anti-Stx2 (1 μg/mL, 11E10, Hycult Biotech, Uden, Netherlands) suspended in 0.1 M carbonate buffer with pH 9.6 (Merck, Darmstadt) overnight at 4°C. Wells were washed thrice with PBS Tween 0.05% (PBS-T, Medicago, Uppsala, Sweden) and blocked with bovine serum albumin (1%, BSA, Sigma-Aldrich) in PBS for 1 h at rt. Microvesicle suspensions and supernatants were incubated with saponin. Saponin (0.5%, Sigma-Aldrich) was used to permeabilize the microvesicles to enable antibody detection of intravesicular Stx2. Microvesicle samples and a standard curve of Stx2 (range 500–1.6 ng/mL), all diluted in BSA, were incubated overnight at rt. Wells were washed with PBS-T thrice and incubated with rabbit anti-Stx2 (1:1,000, BEI Resources, Manassas, VA) diluted in BSA for 1 h at rt. Wells were washed with PBS-T as above and incubated with goat anti-rabbit HRP (1:1,000, Dako, Glostrup, Denmark), in BSA. Wells were washed and developed using SuperSignal ELISA Pico (Thermo Fisher Scientific) according to the manufacturer's instructions. Luminescence was measured using a Glomax Discover System (Promega, Madison, WI) at 1 s integration time per well. The Stx2 concentration of the saponin-treated microvesicle suspension was 30.3 ng/mL. No toxin was detected in the supernatant from the microvesicle suspension. The percentage of the recovered toxin in the microvesicle suspension was 0.13% of the initial toxin dose after adjusting for volumes.
Microvesicle samples were analyzed for size distribution using nanoparticle tracking analysis. Briefly, samples were diluted 1:10 in PBS filtered through 0.2 μm pore-size filters and loaded into a syringe pump. The microvesicle suspensions were recorded under flow using a NanoSight LM10 instrument equipped with a 405 nm laser (NanoSight, Amesbury, UK). Three consecutive recordings were analyzed with NTA software 3.2 (NanoSight) giving a peak at 124.3 nm (range 56.5–672.5 nm) for Stx2-positive microvesicles and 133.7 nm (range 41.5–808.5) for Stx2-negative microvesicles. These data suggest that most of the vesicles were shed microvesicles (100–1,000 diameter) but the presence of some exosomes (30–100 nm) could not be ruled out.
Chinese hamster ovary (CHO-native) epithelial cells (ATCC, Manassas, VA), DLD-1 (colonic epithelial) cells (ATCC), or HeLa cells (cervical epithelial, a kind gift from L. Johannes, Institute Curie, Paris) were cultured in DMEM supplemented with 10% fetal calf serum and 1% penicillin-streptomycin (both from Gibco) in 5% CO2 at 37°C.
CHO-native cells were transfected with the pEF1α-IRES-ZsGreen1 plasmid (CloneTech Laboratories, Mountain View, CA) containing A4GALT cDNA or the corresponding control vector (a kind gift from Martin L. Olsson, Transfusion Medicine, Lund University), lacking the cDNA insert, using Lipofectamine 3000 (Thermo Fisher Scientific). Two days after transfection G418 (400 μg/mL, Sigma-Aldrich) was added to the cells to select for transfected clones. After 7 days of selection cells were seeded out at a density of 1 cell per well in a 96-well plate (Corning, Corning, NY). Clones were picked based on the presence of ZsGreen and Stx2:Alexa555 binding in A4GALT-positive CHO cells (CHO-Gb3) or a lack of Stx2:Alexa555 binding in the vector control CHO cells (CHO-control).
HeLa cells were transfected with the CD77 synthase shRNA plasmid or with the control shRNA Plasmid A using Plasmid Transfection Reagent (all from Santa Cruz Biotechnology, Dallas, TX). Two days after transfection puromycin (2 μg/mL, Sigma-Aldrich) was added to the cells to select for transfected clones. Two clones from the cells transfected with the CD77 synthase shRNA plasmid and two clones from the cells transfected with the control plasmid were isolated and further cultured. The clones were analyzed for the presence of Gb3 by glycosphingolipid extraction and separation using thin layer chromatography followed by visualization by orcinol staining, as described below. Results showed that reduction in Gb3 had not been obtained and therefore inhibition of Gb3 synthesis was pursued as described below.
HeLa cells were treated with the glycosylceramide synthase inhibitor D-threo-1-Phenyl-2-hexadecanoylamino-3-morpho lino-1-propanol (PPMP 5 μM, Abcam, Cambridge, UK) for at least 10 days to inhibit synthesis of Gb3.
CHO-native, CHO-Gb3 (expressing Gb3), CHO-control (the vector control), HeLa cells, PPMP-treated HeLa cells and DLD-1 cells were grown in two T75 cell culture flasks (Thermo Fisher Scientific) each. Once confluent cells were treated with trypsin and washed twice in PBS, by centrifugation at 500 × g for 10 min.
Microvesicles were purified, as described above, from 12 mL human whole blood drawn from healthy volunteers into citrated blood collection tubes (Becton Dickinson), diluted 1:1 with DMEM and incubated with calcium ionophore A23187 (10 μM, Sigma-Aldrich), to stimulate microvesicle release, for 40 min at 37°C.
Lipid extraction and thin layer chromatography (TLC) of cells and microvesicles was performed as described previously (Hedlund et al.,
Cells (15,000 CHO-Gb3 or CHO-control or 10,000 DLD-1 cells/well) were seeded out in black 96-well plates (Corning) with clear bottoms 24 h before the start of the experiments. Cells were incubated with blood cell-derived Stx2-positive microvesicles (final Stx2 concentration/well: 2 ng/mL as determined by ELISA), Stx2-negative microvesicles (CHO-Gb3 or CHO-control), free Stx2 (2 ng/mL CHO-Gb3, CHO-control or DLD-1 cells), or PBS alone (CHO-Gb3, CHO-control or DLD-1), all diluted in serum-free DMEM (Gibco). The amount of microvesicles in Stx2-positive microvesicles and Stx2-negative microvesicles was equalized based on protein content, as measured by light absorption at 280 nm (NanoDrop, Thermo Fisher Scientific). After 24 h incubation cells were washed twice with PBS and Alamar Blue (Thermo Fisher Scientific) diluted 1:10 in serum-free DMEM was added to the cells. The time-point of 24 h was chosen as all cells detached after longer incubations. When a clear shift in the dye was visible the plates were read in a Glomax Discover System (Promega), using 520 nm excitation light and 580–640 nm emission filters. Using this assay we could demonstrate that the CHO-Gb3 cells were sensitive to free Stx2 whereas the CHO-control (vector control) cells were not (
Using the same method a dose response assay was designed to determine the effect of Stx2-positive microvesicles vs. free Stx2 on cell metabolism. For these experiments Stx2-positive microvesicles were added at concentrations between 0.015 and 3.75 ng/mL and free Stx2 between 0.007 and 3.75 ng/mL. The IC50 value of Stx2-positive microvesicles was 0.088 ng/mL and the IC50 value of free Stx2 was 0.11 ng/mL (
CHO-Gb3 or CHO-control cells (15,000 cells/well) or HeLa, HeLa-PPMP, or DLD-1 cells (10,000 cells/well) were seeded out and cultured in black 96-well plates with clear bottoms (Corning) 24 h before incubation with blood cell-derived Stx2-positive microvesicles, Stx2-negative microvesicles, free Stx2 (2 or 3 ng/mL), or PBS alone, as described above. In certain experiments CHO-Gb3 cells were pre-treated with Retro-2.1. Retro-2.1 is an early-endosome-to-Golgi-transport inhibitor that has been shown to protect cells from Stx2-induced protein synthesis inhibition (Gupta et al.,
CHO-Gb3 or CHO-control cells (15,000 cells/well) were seeded out and cultured in 96-well plates 24 h before incubation with blood cell-derived microvesicles. After 2 or 24 h the microvesicle-suspensions that had been incubated with the cells and microvesicle suspensions that had not been in contact with cells, were collected and analyzed by a CD41/CD61 ELISA (Novus Biologicals, Centennial, CO), according to the manufacturer's instructions. Absorbance was measured at 450 nm using the Glomax Discover System.
Whole blood was drawn as described above and centrifuged at 200 × g for 15 min. The platelet-rich supernatant was collected and further centrifuged at 2,000 × g for 10 min to obtain a pellet. The supernatant was discarded, and platelets were suspended and stained in 5 μM CellTrace CFSE (Thermo Fisher Scientific) diluted in washing buffer (NaCl 140 mM, EDTA 9 mM, Na2HPO4 26 mM, adjusted to pH 7) for 30 min in the dark. Platelets were washed thrice in washing buffer and resuspended in cold Hank's Balanced Salt Solution with Ca2+/Mg2+ (HBSS), containing Stx2 labeled with Alexa 555 (200 ng/mL, final concentration) and GPRP (1 mM) for 1 h in 4°C and washed twice in cold washing buffer. Platelets were resuspended in cold HBSS and stimulated with calcium ionophore A23187 (10 μM final concentration) for 40 min at 37°C followed by centrifugation at 10,000 × g for 10 min. The microvesicle-containing supernatant was collected and further washed as described above. These microvesicles were used for detection of microvesicle uptake by CHO-native cells described below.
HeLa cells were seeded out at a density 150,000 cells/mL in Fluorobrite DMEM media (Gibco) in T75 culture flasks (Thermo Fisher Scientific). One day later the cells were incubated with TNF-α (20 ng/mL, Sigma-Aldrich) at 37°C for 24 h. Cells were washed twice in PBS and incubated in ice-cold FluoroBrite DMEM media with Stx2:Alexa 488 (200 ng/mL, final concentration) for 1 h in 4°C. Cells were washed as above and incubated with A23187 10 μM for 40 min to induce microvesicle release. The cell media was collected and centrifuged at 2,500 × g for 5 min and the supernatant, containing microvesicles, was collected and washed as described above. These microvesicles were used for detection and quantification of microvesicle uptake by DLD-1 cells.
Cells (CHO-native (45,000) or DLD-1 (30,000) cells/well) were seeded out in an Ibidi 8-well chamber slide (Ibidi, Gräfelfing, Germany) 1 day before the start of the experiment. CHO-native cells were washed twice with PBS and incubated with the platelet-derived microvesicle suspension derived from a total of 1.2 ml of whole blood or Stx2:Alexa555 200 ng/mL, corresponding to the concentration used to generate the Stx2-positive platelet microvesicles.
DLD-1 cells were incubated with the microvesicle suspension derived from 350,000 HeLa cells, or Stx2:Alexa 488 at 31 ng/mL, which gave the equivalent fluorescence as the microvesicle suspension when measured in the Glomax Discover System.
Microvesicles and labeled Stx2 were diluted in Opti-MEM (Invitrogen, Carlsbad, CA) and incubated with the cells for 4 h. After the incubation, cells were fixed in 4% paraformaldehyde (Histolabs, Västra Frölunda, Sweden), for 20 min. CHO-native cells were stained with NucBlue Live Cell Stain (Thermo Fisher Scientific) and DLD-1 cells were stained with Cellmask Deep Red plasma membrane stain (1.6 μg/mL, Thermo Fisher Scientific) and visualized in an Axio Observer.A1 inverted fluorescence microscope (Zeiss) or in a Ti-E inverted fluorescence microscope equipped with a Nikon structured illumination microscopy module (Nikon Instruments, Tokyo, Japan).
Images captured using fluorescence microscopy of DLD-1 cells that had been incubated with HeLa cell-derived Stx2:Alexa488-positive microvesicles or free toxin were quantified using ImageJ (ImageJ version 1.52n, Bethesda, MA). Each image stack was converted to a max intensity image. Individual cells were outlined and a minimum threshold for green fluorescence was set to 2100 pixel intensity. The area of the green pixels in each cell was quantified and plotted (
Gb3 (200 μg, Matreya LLC, State College, PA), phosphatidylethanolamine (Larodan, Solna, Sweden), and phosphatidylserine (Sigma-Aldrich) diluted in methanol:chloroform (16:13) were mixed and dried under a stream of nitrogen. The lipids were resuspended in PBS 400 μL, vortexed (4 × 30 s, setting 10, Vortex Genie 2, Scientific Industries, Bohemia, NY) and sonicated in a water-bath (Metler Electronics, Anaheim, CA) for 30 min to form liposomes. The Gb3-containing liposomes were immediately added to CHO-native or DLD-1 cells at a concentration of 3.75 μg/100,000 cells. To verify that the liposomes were able to bind Stx2, some of the liposome suspension was added to a chamber slide (Ibidi) in the absence of cells together with Stx2:Alexa 488 200 ng/mL or PBS and imaged in an Axio Observer A1 inverted fluorescence microscope (
For comparison between two groups the two-tailed Mann-Whitney
We transfected CHO-native cells that thereby express Gb3 and inhibited synthesis of Gb3 from HeLa cells using PPMP. CHO-Gb3 and CHO-control cells were incubated with blood cell-derived Stx2-positive and Stx2-negative microvesicles or free toxin, at the same concentration as in Stx2-positive microvesicles, for 24 h. CHO-Gb3 cells incubated with Stx2-positive microvesicles exhibited a significantly lower cell metabolism compared to cells incubated with Stx2-negative microvesicles (
Gb3-positive but not Gb3-negative CHO cells are susceptible to microvesicle-delivered Stx2. CHO-Gb3 and CHO-control cells were incubated with Shiga toxin 2-positive microvesicles (Stx2-pos MVs), Stx2-negative microvesicles (Stx2-neg MVs), free Stx2 or PBS and cell metabolism was measured.
Cell metabolism and protein synthesis in cells exposed to Stx2-positive microvesicles and free toxin.
Stx2-pos MVs | Cell metabolism | UC |
↓↓ |
NA | NA | NA | NA |
Protein synthesis | UC |
↓↓ |
↓ |
↓↓ |
UC |
UC |
|
Free Stx2 | Cell metabolism | UC |
↓↓ |
NA | NA | NA | UC |
Protein synthesis | UC |
↓↓ |
NA | ↓↓ |
UC |
UC |
Stx2-positive microvesicles are toxic to Gb3-positive but not Gb3-negative cells. CHO-Gb3, CHO-control and CHO native cells, HeLa and HeLa-PPMP and DLD-1 cells were incubated with Shiga toxin 2 positive microvesicles (Stx2-pos MVs), Stx2-negative microvesicles (Stx2-neg MVs), Stx2 or PBS and protein synthesis was measured.
To investigate the intracellular transport route of Stx2 delivered via microvesicles, CHO-Gb3 cells were treated with the retrograde transport inhibitor Retro-2.1 or PBS vehicle for 30 min before addition of Stx2-positive microvesicles. Results showed that the protein synthesis associated with microvesicle-delivered Stx2 was lower in the cells that had been treated with PBS compared to the Retro-2.1-treated cells (
Shiga toxin 2 delivered by microvesicles utilizes the intracellular retrograde pathway. The intracellular pathway of Shiga toxin 2-positive microvesicles (Stx2-pos MVs) was determined using Retro-2.1. CHO-Gb3 cells were treated with Retro-2.1 or PBS before addition of Stx2-positive microvesicles (Stx2-pos MVs) (
To investigate if Stx2, delivered within microvesicles, is taken up by Gb3-negative cells, fluorescently labeled microvesicles, containing Stx2:Alexa555, were isolated from stimulated platelets and incubated with CHO-native cells. Some of the labeled microvesicles were associated with the CHO-native cells within 4 h and a portion of the microvesicles were positive for Stx2:Alexa555 (
Shiga toxin 2 can be taken up by Gb3-negative cells within microvesicles.
To determine if Gb3-positive and Gb3-negative cells take up equivalent amounts of microvesicles, CHO-Gb3 and CHO-control cells were incubated with blood cell-derived microvesicles, followed by analysis of CD41/CD61, platelet markers, in the microvesicle-containing supernatants. The median values of CD41/CD61 in the microvesicle suspensions after 2 h incubation were 84.1 ng/mL (range 69.9–100.1,
To test if exogenously administrated Gb3 could introduce Stx2-sensitivity, Gb3-negative DLD-1 or CHO-native cells were treated with exogenous Gb3-liposomes followed by incubation with Stx2 (range 25–200 ng/mL) for 24 or 48 h. Exogenous Gb3 did not introduce sensitivity to Stx2 in DLD-1 cells (
Exogenous administration of Gb3 did not introduce Stx2 susceptibility in DLD-1 cells. DLD-1 cells were incubated with Gb3-containing liposomes or PBS control followed by incubation with Shiga toxin 2 (Stx2) for 24 h. No significant difference in cell metabolism could be seen in PBS-treated or Gb3-liposome treated cells at any of the Stx2 concentrations. Median cell metabolism is denoted by the bar.
Shiga toxin can be taken up by cells after binding to its receptor, Gb3, on the cell surface, or by cellular uptake of bacterial outer membrane vesicles (Bielaszewska et al.,
Studies from our group have shown that Stx2 circulates bound to blood cell-derived microvesicles and thereby reaches the kidney (Ståhl et al.,
The microvesicles used herein were shed from blood cells that were stimulated with Stx2 and contained Stx2 within the vesicle. The toxin binds to Gb3 on the blood cells, is taken up and shed within microvesicles (Karpman et al.,
Cell viability was assessed using the redox Alamar blue assay and staining was indicative of the number of living cells. An interesting incidental observation was that cells exposed to microvesicles that did not contain Stx2 exhibited a higher cellular metabolism. CHO-Gb3 cells, incubated with microvesicles lacking Stx2, and CHO-control cells, incubated with microvesicles with and without Stx2, had a significantly higher cellular metabolism compared to untreated cells (
Stx that is bound to Gb3 not present in lipid rafts will be directed to lysosomes for destruction (Falguieres et al.,
In this study we chose to work with the CHO cell line as these Gb3-negative cells have previously been shown to become sensitive to Stx when transiently transfected with the Gb3-synthase A4GALT (Keusch et al.,
Gb3 is essential for a response to Stx in a murine
In summary, our findings suggest that uptake of Gb3- and Stx2-positive microvesicles is not sufficient to induce cell death and that the recipient cell must possess Gb3, and possibly a certain glycosphingolipid composition within its intracellular lipid microdomains, for Stx2 delivered within microvesicles to exert its toxic effect. Uptake of Stx2-positive microvesicles to Gb3-negative cells will not affect cell viability but may enable passage of the vesicles through the cells in an inflammatory microenvironment.
All data in this study are available from the corresponding author upon request.
The studies involving human participants were reviewed and approved by Regional Ethics Review Board of Lund University. The participants provided their written informed consent to participate in this study.
KJ, AW, A-CK, AT, and AS designed and conducted experiments. DG contributed material. KJ, AW, A-CK, AT, AS, and DK analyzed data. KJ and DK wrote the manuscript. All authors read and approved the final manuscript.
The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
The authors thank Martin L. Olsson (Dept. of Transfusion Medicine, Lund University) for supplying the pEF1α-IRES-ZsGreen1 plasmid containing A4GALT cDNA and Ludger Johannes (Institut Curie, Paris, France) for providing the HeLa cell line. A preliminary version of this manuscript appeared in the PhD thesis of KJ.
The Supplementary Material for this article can be found online at: