Source of microorganisms.
The human gut derived strains Methanobrevibacter smithii ALI (DSM 2375), and Methanosphaera stadtmanae (DSM 3091, type strain) were obtained from the German Collection of Microorganisms and Cell Cultures (DSMZ) GmbH, Braunschweig, Germany. Candidatus M. intestini WWM1085 (DSM 116060) was obtained from the Department of Microbiology, University of Illinois, USA, where it was isolated from a stool sample of a healthy woman24. In the following, we will use the abbreviation “M. intestini” instead of Candidatus M. intestini.
M. smithii GRAZ-2 (DSM 116045) was isolated in 2018 at the Medical University of Graz, Graz, Austria, from a stool sample of a healthy woman24. Instead of opting for the Methanobrevibacter smithii type strain (PS, DSM 861), our choice was M. smithii ALI, as it sourced from a human fecal sample and not from sewage water. Enterotoxigenic Escherichia coli (ETEC) H10407 and Bacteroides fragilis ATCC® 25285 have been reported previously25.
Growth media and cultivation.
For the cultivation of all methanogens standard methanogenium medium (MS) with some modifications as previously described 24. For vesicle production, aliquots of 250 ml media in 1000 ml infusion bottles were sealed, pressurized with H2/CO2 (4:1) and autoclaved. Before inoculation and incubation at 37°C, sodium acetate (0.001g/ml, anoxic, sterile) and yeast extract (0.001g/ml, anoxic, sterile, YE) were added to the media. Vesicles of ETEC and B. fragilis were retrieved from stocks prepared earlier 25.
Electron microscopy
Electron microscopy (EM) was undertaken at the Core Facility Ultrastructure Analysis, Medical University of Graz, Graz, Austria and at the Core Science Resources Quadram Institute Bioscience, Norwich, United Kingdom. For ultrastructural analyses of cells, isolates were cultivated in 20 ml aliquots in 100 ml serum bottles for 7 days under anaerobic conditions at 37°C in an incubation shaker (shaking speed: 80 rpm). Followed by the centrifugation of 2 ml of medium containing each strain at 4000 g, 4°C, for 10 min. Cell pellets were then directly handed over to the Core Facility Ultrastructures, Medical University Graz, Graz, Austria for further preparation. AEVs (1x1011/ml) were directly handed over to the Core Science Resources Quadram Institute Bioscience, Norwich, United Kingdom.
Transmission electron microscopy: thin sections and tomography
Cells were fixed in 2.5% (w/v) glutaraldehyde and 2% (w/v) paraformaldehyde in 0.1 M cacodylate buffer, pH 7.4, for 1 h, postfixed in 1% (w/v) osmium tetroxide for 2 h at room temperature, dehydrated in graded series of ethanol and embedded in TAAB (Agar Scientific, Essex, GB) epoxy resin. Ultrathin sections (70 nm thick) were cut with a UC 7 Ultramicrotome (Leica Microsystems, Vienna, Austria) and stained with lead citrate for 5 min and with platinum blue for 15 min. Images were taken using a Tecnai G2 20 transmission electron microscope (Thermo Fisher) with a Gatan ultrascan 1000 charge coupled device (CCD) camera (temperature − 20°C; acquisition software Digital Micrograph; Gatan, Munich, Germany). The acceleration voltage was 120 kV. The tilt series was reconstructed using FLARA, a joint alignment and reconstruction algorithm for electron tomography. This iterative algorithm allows for acquisitions without fiducial gold markers, since an effective shift computation can be obtained by using a global alignment technique based on a linearized approximation of the disruptive shifts in each iteration26. For negative staining cell suspensions were placed on glow discharged carbon coated copper grids for 1 min. The solution was removed after incubation by filter paper stripes. A drop of 1% aqueous uranyl acetate solution was placed afterwards for 1 min, dried with filter paper and later on air dried at room temperature. Specimens were examined with an FEI Tecnai G 2 (FEI, Eindhoven, Netherlands) equipped with a Gatan ultrascan 1000 charge coupled device (CCD) camera (-20°C, acquisition software Digital Micrograph, Gatan, Munich, Germany).
AEV suspensions were visualized using negative staining with TEM. Briefly, 4 µL AEV suspension was adsorbed to plasma-pretreated carbon-coated copper EM grids (EM Solutions) for 1 min before wicking off with filter paper and negatively staining with 1% Uranyl Acetate solution (BDH 10288) for 1 min. Grids were air-dried before analysis using a FEI Talos F200C electron microscope at 36,000×-92,×000 magnification with a Gatan OneView digital camera.
Scanning electron microscopy
For scanning electron microscopy, cells were affixed to coverslips and treated with a fixing solution consisting of 2% paraformaldehyde and 2.5% glutaraldehyde in 0.1 M phosphate buffered saline (pH 7.4). Subsequently, a graded ethanol series was used for dehydration. Post-fixation involved 1% Osmium tetroxide for 1 hour at room temperature, followed by additional dehydration in an ethanol series (ranging from 30–100% EtOH). Hexamethyldisilazane (HMDS) was applied, and coverslips were positioned on stubs using conductive double-coated carbon tape. Imaging was performed with a Sigma 500VP FE-SEM equipped with a SEM Detector (Zeiss Oberkochen) operating at an acceleration voltage of 5 kV.
AEV Isolation
To obtain a sufficient amount of biomass for the isolation of AEVs, 250 ml of MS medium was aliquoted into 1000 ml infusion bottles (VWR) and further handled the same way as described above. These cultures were then cultivated for 10 days under anaerobic conditions at 37°C in an incubation shaker (shaking speed: 80 rpm). When the pressure of cultivation bottles dropped due to growth, they were re-gassed with H2/CO2. Growth was surveyed by optical density photometry at 600 nm. On day ten, the cell suspensions were centrifuged at 14,000 x g, 4°C, 20 min (Thermo Scientific™ Sorvall™ LYNX™ 6000). To remove cell debris and remaining cells, the supernatant was filtered with 0.22 µm PES bottle-top filters (Fisherbrand™ Disposable PES Bottle Top Filters). If not immediately processed, the supernatant containing the vesicles was stored at 4°C.
Isolation of vesicles was done according to Stentz et al.27 (Workflow see Supplementary Fig S1). In brief, a filtration cassette (Vivaflow 50R, 100,000 MWCO, Hydrostat, model VF05H4, Sartorius or Vivaflow 200 100,00 MWCO, PES, model VF20P4) was used to concentrate 1 L of sample down to approx. 5 ml. Then, 500 ml PBS buffer (pH 7.4) was added for washing purposes, and the liquid was concentrated to 1–4 ml. The sample was then centrifuged for 20 min at 10,000 g, 4°C to remove protein and lipid aggregates. Next, the sample was transferred to Pierce™ Protein Concentrators (PES, 100,000 MWCO, Thermo Scientific) and centrifuged at 3,000 g until the samples were concentrated down to 1 ml. Residual contaminants and proteins were further eliminated through size exclusion chromatography (SEC) using an IZON qEV1 column (pore size 35 mm) according to the manufacturer’s instructions. The vesicles were eluted in the 2.8 ml fraction containing the purified extracellular vesicles underwent a final filter sterilization using a 0.22 µm syringe filter (ROTILABO® PES, 0,22 µm), and were subsequently stored at 4°C until further use.
To ensure that the final AEV suspension does not contain any yeast vesicles or other residues, the YE was sterile-filtered previous to medium preparation.
For the metabolomics analyses, 1 L of blank MS medium underwent the same procedure to serve as a control.
BEV Isolation
BEVs for the HT-29 experiment were isolated as described previously with minor modifications25,28. Briefly, overnight cultures were either grown with aeration (180 rpm, Infor shaker) in case of ETEC or anaerobically (GasPak™ EZ Systems, BD) in case of B. fragilis to ensure sufficient growth. The respective cultures were diluted (1:100) in BHI medium and grown at 37°C either with aeration for 8 h or overnight anaerobically (GasPak™ EZ Systems, BD). The cells were then removed from the supernatant by centrifugation (9,000 x g, 15 min) and subsequent sterile filtration (0.22 µm). The BEVs present in the supernatant were pelleted through subsequent ultracentrifugation (150,000 x g, 4°C, 4 h), resuspended in appropriate volumes of PBS to generate a BEV suspension 1000-fold more concentrated than in the original culture supernatant. Quantification and size distribution of BEVs were investigated by nanoparticle tracking analysis (NTA) using a Nanosight NS300 (see below).
AEV characterization
Nanoparticle tracking analysis (NTA)
Quantification and size distribution of AEVs were investigated by nanoparticle tracking analysis (NTA) using ZetaView and Nanosight NS300. ZetaView was used by following established protocols27,29. In brief, particles were quantified using the ZetaView instrument (Particle Metrix, Germany) with ZetaView (version 8.05.12 SP1) software running a 2 cycle 11 position high frame rate analysis at 25°C. Samples were diluted with ultrapure water allowing the optimal detection range. Camera control settings: 80 Sensitivity; 30 Frame Rate; 100 Shutter. Post-acquisition parameters: 20 Min Brightness; 2000 Max Area; 5 Min Area; 30 Trace Length; 5 nm/Class; 64 Classes/Decade.
For NanoSight NS300 (Malvern Instruments, UK) samples were diluted in 1x PBS according to the manufacturer’s guidelines (final concentration between 107 − 109 particles per ml), and a 405 nm laser was used. Between samples, the instrument was flushed with 10% Ethanol and Aqua.dest. Reads of 1-minute duration were performed in five replicates for each sample with the following capture settings: cell temperature: 25°C, syringe load/flow rate: 30, camera: sCMOS. For capture settings, camera level was adjusted so that all particles were distinctly visible (Camera level 12–15). The ideal detection threshold was set including as many particles as possible and debris (blue cross count) with a maximum of five (detection threshold 5). Data output was acquired using NanoSight NTA software version 3.3 (Malvern Instruments). For each sample, the mean particle number in the Experiment Summary output was adjusted by the dilution factor.
Protein, DNA, and RNA content
As previously described30–34, quantification of vesicle content, including protein, DNA, and RNA, was conducted using the Qubit® Protein Assay, Qubit® dsDNA high sensitivity assay, and RNA high sensitivity assay kits, respectively (Thermo Fisher Scientific). Protein, DNA, and RNA measurements were performed using a Qubit® 4 or Qubit® 3 Fluorometer. Instructions of the manufacturer were followed.
Lipid content
The quantification of lipid content in AEVs was conducted using the FM4-64 lipophilic fluorescent dye and a linoleic acid standard, a method previously employed for bacterial extracellular vesicle (BEV) lipid quantification35. The modified procedure for quantifying vesicles released in culture was previously described in Juodeikis et al.29 and includes the following steps: In duplicate, 20 µL of 30 µg/ml FM4-64 (Thermo Fisher Scientific) was combined with 180 µL of filtered culture supernatant or a linoleic acid standard in water (100, 75, 50, 20, 10, 5, 1, 0 µg/ml, prepared from a 1 mg/ml stock) in black 96-well plates. Following a 10-minute incubation at 37°C, endpoint fluorescence was analyzed using the FLUOStar Omega microplate reader with pre-set FM 4–64 settings (Excitation: 515 − 15; Dichroic: auto 616.2; Emission 720 − 20), employing an enhanced dynamic range. Linear standard curves from the linoleic acid samples were established for lipid quantification.
Proteomics
Protein profiles of whole cell lysates (WCL) and AEVs were analyzed. Therefore, 20 mg of cell biomass (3 replicates per species) were subjected to extensive ultrasonication with 400 µl of PBS. Cell debris was removed with centrifugation at 800 g at 4°C, for 5 min. The supernatants were collected for proteomic analysis. The protein content of the whole cell lysate was determined by Pierce BCA protein assay according to the manufacturer's protocol (Thermo, USA). Protein concentration of AEVs was measured by Qubit® Protein Assay (Thermo Fisher Scientific), as described above.
Mass spectrometry analysis
For LC-MS/MS analysis, 2 (for AEVs) or 5 µg (for WCLs) of protein were reduced and alkylated for 10 min at 95°C with final 10 mM TCEP (tris(2-carboxyethyl)phosphine) and 40 mM CAA (2-Chloroacetamide). The sample was processed according to the SP3 protocol36
and digested overnight with trypsin (Promega, enzyme/protein 1:50). Peptides were desalted using SBD-RPS tips as previously described37. 400 ng per sample (re-dissolved in 2% acetonitrile/0.1% formic acid in water) was subjected to LC-MS/MS analysis. Protein digests were separated by nano-HPLC (Dionex Ultimate 3000, Thermo Fisher Scientific(Dionex Ultimate 3000) equipped with a C18, 5 µm, 100 Å, 100 µm x 2 cm enrichment column and an Acclaim PepMap RSLC nanocolumn (C18, 2 µm, 100 Å, 500 x 0.075 mm) (all Thermo Fisher Scientific, Vienna, Austria). Samples were concentrated on the enrichment column for 5 min at a flow rate of 15 µl/min with 0.1 %formic acid as isocratic solvent. Separation was carried out on the nanocolumn at a flow rate of 300 nl/min at 60°C using the following gradient, where solvent A is 0.1 %formic acid in water and solvent B is acetonitrile containing 0.1 %formic acid: 0–5 min: 2 %B; 5-123 min: 2–35 %B; 123–124 min: 35–95 %B, 124–134 min: 95 %B; 134–135 min: 2 %B; 135–150 min: 2% B. The maXis II ETD mass spectrometer (Bruker Daltonics, Germany) was operated with the captive source in positive mode with the following settings: mass range: 200–2000 m/z, 2 Hz, capillary 1,600 V, dry gas flow 3 L/min with 150°C, nanoBooster 0.2 bar, precursor acquisition control top 20 (collision induced dissociation (CID). The mass spectrometry proteomics data were deposited to the ProteomeXchange Consortium38 via the partner repository with the dataset identifier PXD053245 (Reviewer access details: Log in to the PRIDE website using the following details: PDX accession: PXD053245;Username: [email protected]; Password: MFqECDz7Uyv6)38.
The LC-MS/MS data were analyzed by MSFragger39,40 by searching the public Methanobrevibacter protein databases (UP000232133; UP000003489; UP000004028; UP000018189; UP000001992), the archaeal protein catalogue described in Chibani et al.22 and a list of common contaminants41. Additional information on proteins found in all vesicles was retrieved via MaGe42 and the implemented functions SignalP (version 4.1)43, MHMM (version 2.0c)44,45 and InterProScan46,47, as well as from the InterPro Database47 (Supplementary Table 5).
Carbamidomethylation of cysteine and oxidation on methionine were set as a fixed and as a variable modification, respectively. Detailed search criteria were used as follows: trypsin, max. missed cleavage sites: 2; search mode: MS/MS ion search with decoy database search included; precursor mass tolerance ± 20 ppm; product mass tolerance ± 15 ppm; acceptance parameters for identification: 1% protein FDR48.
Data from EV and whole cell lysates were processed with Perseus software version 1.6.15.0. Data was filtered for decoy hits and contaminants. After log2 transformation, and subtracting the median from the column proteins were filtered for containing at least 2 valid values in at least one group.
Mass spectrometry derived AEV metabolomics
Biological triplicates of the vesicle preparations were used for the LC-MS analysis, and a technical duplicate of a non-cultured medium that had passed through the pipeline for vesicle isolation was used as a medium blank. All samples were stored at -70°C until processing at the Vienna BioCenter Metabolomics Core Facility.
The samples were diluted with 50 µL ACN and subjected to analysis with liquid chromatography-mass spectrometry (LC-MS). 11 µL of each sample was pooled and used as a quality control (QC) sample. Samples were randomly injected on an iHILIC®-(P) Classic HPLC column (HILICON AB, 100 x 2.1 mm; 5 µm; 200 Å, Sweden) with a flow rate of 100 µL/min delivered through an Ultimate 3000 HPLC system (Thermo Fisher Scientific, Germany). The stepwise gradient has a total run time of 35 min, starts at 90% A (ACN), and takes 21 min to 60% B (25 mM ammonium bicarbonate) followed by 5 min hold at 80% B and a subsequent equilibration phase at 90%. The LC was coupled to a high-resolution tandem MS instrument (Q-Exactive Focus, Thermo Fisher Scientific, Germany). The ionization potential was set to + 3.5/-3.0 kV, the sheet gas flow to 20, and an auxiliary gas flow of 5 was used. Samples were flanked by a blank and a QC sample for background labeling and data normalization, respectively.
The obtained data set was processed by “Compound Discoverer 3.3 SP2” (Thermo Fisher Scientific). Annotation of the compounds was done through searching against our internal mass list database generated with authentic standard solutions (highest confidence level). Additionally, the mzCloud database was searched for fragment matching and ChemSpider hits were obtained using BioCyc, Human Metabolome Database, E. coli Metabolome Database, and KEGG databases. Only metabolites identified with highest confirmation (match with internal database) were examined in more detail; additional ones are provided in Supplementary Table 7).
The log2 fold changes, as well as p-values, were calculated by the Compound Discoverer software (Tukey HSD test (posthoc), after an analysis of variance (ANOVA) test).
Co-incubation experiments with cell lines
Cytotoxicity tests of AEVs and BEVs
3-(4,5-Dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT) cell viability assays were routinely performed at the end of the HT-29 cell culture assays49, but no significant reduction in metabolic activity could be observed for any condition used in this study (data not shown).
Additionally, CellTiter-Glo® 2.0 Cell Viability Assay (Promega) was used to investigate the cytotoxicity of AEVs on THP1-Blue cells, but no reduction in the viable cells could be detected (data not shown).
Confocal Microscopy
M. smithii ALI, M. intestini, M. smithii GRAZ-2, and M. stadtmanae-derived AEVs (1x1011/ml) were labeled with 5% DiO at 37°C for 30 minutes. Labeled DiO - AEVs (1x1011/well [10 µl]) were added to THP1-b cell monolayers cultured on collagen solution (Merck) coated 12-well chamber slides (IBIDI) overnight (16 hrs). THP1-b monocytes were previously induced to differentiate into macrophages using 150 nM PMA (Phorbol 12-myristate 13-acetate; Sigma, P8139). Samples were fixed using Pierce 4% PFA (ThermoFisher), permeabilized with 0.25% Triton X1000 (Sigma), and blocked with 10% goat serum in PBS. For nuclear visualization, cells were incubated with Hoechst 33342 (ThermoFisher), Alexa 647-Phalloidin to visualize intracellular membranes. As a second approach, AEVs were incubated with Archaea specific primary antibodies (Davids Biotechnologie GmbH, affinity purified, specific for Methanobrevibacter and Methanosphaera) and AF647 as the secondary antibody, and cells were labeled with Hoechst 33342 (ThermoFisher). Images were taken using a Zeiss LSM880 confocal microscope equipped with a 63x/1.40 oil objective. Fluorescence was recorded at 405 (blue, nucleus), 488 (green, AEVs), and 594 nm (red, intracellular membranes or AEVs). The red channel was adjusted using the ZEISS ZEN 3.9 (ZEN lite) software by the best-fit function.
HT-29 cytokine release
The HT-29 cytokine release assay was performed at the Institute of Molecular Biosciences, University of Graz. HT-29 (intestinal epithelial cells) were grown in T-175 tissue culture flask, containing Dulbecco's Modified Eagle's medium/ Nutrient F-12 (DMEM-F12) medium (Gibco, USA) supplemented with 10% fetal bovine serum (FBS), Penicillin-Streptomycin (100 µg/ml streptomycin and 100 Units/ml penicillin) and L-Glutamine (2 mM) at 37oC in a CO2 incubator. To investigate the pro-inflammatory potency of AEVs and BEVs, HT-29 cells were seeded in a 24 well tissue culture plates at a concentration of 6 x 105 cells/well and cultivated for 24 h in DMEM-F12 medium supplemented with 10% fetal bovine serum (FBS), Penicillin-Streptomycin and L-Glutamine. Then, intestinal epithelial cells were washed once with PBS and the medium was replaced with AEVs or BEVs (108 particles/ ml) resuspended in DMEM-F12 medium without FBS. After incubation for 20 h the cell culture supernatant was harvested, centrifuged for 2500 rpm at 4°C for 10 min to remove the cell debris and stored at -20°C for subsequent Interleukin 8 (IL-8) quantification by ELISA, which was performed as previously described according to the manufacturer’s protocol25.
Statistics and data visualization
Vesicle properties (Concentration, size, nucleic acids, and protein content) and metabolites were plotted as boxplots in R (R-Core-Team, 2024) using the ggplot2 Package (v3.5.1)50.
Creation of Venn diagrams was performed by using the online tool interactiVenn51. PCA was created with Perseus software (v1.6.15.0)52.
The overview of proteins identified in archaeal vesicles and whole cell lysates, as well as proteins annotated as adhesins, were displayed in heatmaps using ggplot2 (v3.5.1) 50, with data transformation performed using the reshape253 package (v1.4.4; Wickham, 2007). Barchart of mean intensities of protein categories was plotted with ggplot2 (v3.5.1) 50, and dplyr (v1.1.4)54 was used for the calculation of mean and standard deviation. IL-8 excretion in the HT-29 cell line was visualized as a bar chart using ggplot2 (v3.5.1) 50, with data transformation by reshape253 (v1.4.4; Wickham, 2007), FSA (v0.9.5)55, and ggsignif (v0.6.4)56. For IL-8 excretion Kruskal-Wallis test followed by Dunn’s Multiple comparison where all EV samples were compared to the NTC (no treatment control).