Proteomic and metabolomic profiling of extracellular vesicles produced by human gut archaea

Abstract

Gastrointestinal bacteria interact with the host and each other through various mechanisms, including the production of extracellular vesicles (EVs). However, the composition and potential roles of EVs released by gut archaea are poorly understood. Here, we study EVs produced by four strains of human gut-derived methanogenic archaea: Methanobrevibacter smithii ALI, M. smithii GRAZ-2, M. intestini, and Methanosphaera stadtmanae. The size (~130 nm) and morphology of these EVs are comparable to those of bacterial EVs. Proteomic and metabolomic analyses reveal that the archaeal EVs are enriched in putative adhesins or adhesin-like proteins, free glutamic and aspartic acid, and choline glycerophosphate. The archaeal EVs are taken up by macrophages in vitro and elicit species-specific responses in immune and epithelial cell lines, including production of chemokines such as CXCL9, CXCL11, and CX3CL1. The EVs produced by M. intestini strongly induce pro-inflammatory cytokine IL-8 in epithelial cells. Future work should examine whether archaeal EVs play roles in the interactions of archaea with other gut microbes and with the host.

Fig. 1. Ultrastructure of archaeal cells and vesicle-like structures (M. smithii ALI, M. intestini, M. smithii GRAZ-2, M. stadtmanae).

a, e, i, l Scanning electron micrographs of whole cells showing potential vesicle development on their cell surface. b, f, g, j, k Ultra-thin transmission electron micrographs of whole cells, highlighting vesicles located inside or attached to the cells. c, d, h, m Transmission electron micrographs of isolated vesicles. Arrows indicate the presence of archaeal extracellular vesicles (AEVs). The experiments were repeated two times independently with similar results. Source data are provided as a Source Data file.

Fig. 2. Vesicle properties of M. smithii ALI, M. intestini, M. smithii GRAZ-2, and M. stadtmanae.

a Vesicle size [nm]: A one-way ANOVA followed by Tukey’s HSD post hoc test revealed significant differences between M. smithii ALI and M. intestini (p = 0.046), as well as M. intestini and M. smithii GRAZ-2 (p = 0.007). No significant differences were observed for other comparisons. b Vesicle concentration [particles/ml]: No statistically significant differences were found between strains via one-way ANOVA and Tukey’s HSD post hoc test (p > 0.05). Protein (c) was normalized to [µg/10¹⁰ particles], and DNA (d) and RNA (e) content to [ng/10¹⁰ particles] (Supplementary Data 3): A Kruskal-Wallis test was applied to DNA content (d), while one-way ANOVA and Tukey’s HSD post hoc tests were performed for protein (c) and RNA (e) content. No significant differences were observed between strains for c, d or e. One outlier was removed for d and e. Dots represent the number of biological/technical replicates (Supplementary Data 2): M. smithii ALI, n = 10/52; M. intestini = 12/52; M. smithii GRAZ-2 = 5/26; M. stadtmanae = 6/17. The line inside the boxplot indicates the median, while the box spans the first to third quartile; whiskers represent the smallest and largest values within 1.5× the interquartile range. One asterisk represents p < 0.05, ** represents p < 0.01. Source data are provided as a Source Data file.

Fig. 3. Mass spectrometry-based profiling of AEV proteomes.

a Principal component analysis (PCA) plot illustrating the protein profiles of AEVs and WCLs of M. smithii ALI and M. intestini, including only proteins detected in all three replicates per group (AEV M. smithii ALI, AEV M. intestini, WCL M. smithii ALI, WCL M. intestini). b Overlap of 229 proteins identified in the AEVs of M. smithii ALI (left, n = 3 biological replicates) and M. intestini (right, n = 3 biological replicates), visualized and organized by intensities/relative abundance (circle size) and functional categorization (see Supplementary Data 6–8 for details). Data were visualized using RawGraphs and InkScape. c Bar chart displaying mean intensities/relative abundances of protein categories in AEVs and WCLs, based on proteins detected in all three biological replicates of both M. smithii ALI and M. intestini (n = 229) (Supplementary Data 6–8). d Heatmap showing enrichment of 46 proteins annotated as adhesin/adhesion/IG-like present in all six AEV extracts (three biological replicates each of AEV M. smithii ALI and AEV M. intestini) compared to the whole cell lysates based on relative abundances. WCL whole cell lysate, AEV archaeal extracellular vesicles. Source data are provided as a Source Data file.

Fig. 4. Metabolites detected in archaeal vesicles.

Metabolite levels in AEVs from three biological replicates per species, compared to a non-cultured control medium processed through the vesicle isolation pipeline (technical duplicates) (Supplementary Data 10). The Y-axis represents the normalized peak area (LC-MS). Significantly changed compounds are marked with an asterisk (aspartic acid: p = 0.0010 (***); glutamic acid: p = 0.0276 (*); Tukey HSC post hoc test after ANOVA, which applies a correction for multiple comparisons. All tests were two-sided). Data points represent individual measurements; black diamonds indicate the group means, and error bars show standard deviation. Source data are provided as a Source Data file.

Fig. 5. Immunofluorescence microscopy of DiO-labeled AEV uptake by human macrophages.

Human macrophages incubated for 25 h with DiO-labeled (green) AEVs derived from a M. smithii ALI, b M. intestini, c M. smithii GRAZ-2, and d M. stadtmanae. Macrophage monolayers were stained with antibodies to visualize cytoskeleton (Alexa 647-Phalloidin, red), nuclei (Hoechst 33342, blue). e Representative z-stack of M. stadtmanae AEVs internalized by a macrophage. AEVs were additionally labeled with specific anti-archaea antibodies (red) and macrophages were stained with Hoechst 33342 (blue) for visualization of the nuclei. Orthogonal views (XY, YZ and X-Z; left panel) show the plane of view (yellow lines) with example z-slice images at different z-depths (right panel). Images were acquired using a Zeiss LSM880 confocal microscope equipped with a 63 × /1.40 oil objective. Scale bar 20 µm. Experiments were performed twice for each strain. Source data are provided as a Source Data file.

Fig. 6. Heat map showing the induction of cytokine release by macrophages (differentiated THP-1 cells), and intestinal epithelial cells (HT-29).

Cell lines were exposed to archaeal (M. smithii ALI, M. intestini, M. smithii GRAZ-2, and M. stadtmanae) and bacterial (ETEC, B. fragilis) EVs. Cytokine levels were measured by Luminex analyses from supernatants of HT-29 and THP-1 cells exposed to the different EVs for 24 h. Individual cytokines are indicated on the left. EV dose is indicated (A) 10⁸ particles/ml cell culture medium and B 10⁹ particles/ml cell culture medium on the bottom, as well as the vesicle origins. Saline (no EVs) for HT-29 and DMSO for THP-1 cells served as no treatment controls (NTC) to determine non-stimulated secretion levels of cytokines in the respective cell line, which were subtracted as blank from samples. Display of values is log10 transformed. Blue boxes show high levels, and yellow/beige boxes show negative levels of cytokine release upon EV exposure (see scale). White boxes indicate no measurable induction of cytokine release (Supplementary Data 12). Source data are provided as a Source Data file.