The aim of this study was to elucidate the cellular receptors and MAMPs that are involved in sensing the methanogenic archaeon—M. stadtmanae (15, 26). Based on our hypothesis that intracellular PRRs and nucleic acids might be involved, recognition of M. stadtmanae should lead to an antiviral-type cellular immune response. Thus, we first analyzed the time-dependent mRNA expression of type-I and type-III IFNs in moDCs (Figure 1A) and PBMCs (Figure 1B). Expression of the genes encoding IFN-α14, IFN-β, and IFN-λ1 (IL-29) was upregulated in PBMCs and moDCs upon exposure to M. stadtmanae. As NF-κB, IRF1, and IRF5 are reported to be involved in the response against pathogens and particularly IRF5 in the induction of type-I and type-III IFN responses (27), we next examined the subcellular localization and expression of these transcription factors following cellular activation by M. stadtmanae. Using confocal microscopy, we observed that 4 h after stimulation, all three transcription factors were translocated from the cytoplasm into the nucleus (Figure 1C). These findings indicated that these transcription factors were active, and provided further evidence that exposure to M. stadtmanae induces an antiviral type-I/III IFN response.

As such immune responses are often induced in response to viral or bacterial nucleic acids, either detected in the cytosol or endocytic compartments, and given that the response to M. stadtmanae is also dependent on phagocytosis (19), we hypothesized that the nucleic acids from M. stadtmanae might similarly act as MAMPs and activate immune cells. To determine the immunogenicity of archaeal nucleic acids, we transfected purified DNA or RNA into PBMCs using the liposomal transfection reagent DOTAP, and analyzed the secretion of two pro-inflammatory cytokines, TNF-α and IL-1β. Transfection of archaeal DNA did not induce secretion of TNF-α or IL-1β in human PBMCs (Figure S1 in Supplementary Material). In contrast, RNA from M. stadtmanae induced TNF-α and IL-1β release from moDCs (Figure 1D), as well as PBMCs (Figure 1E), which was absent upon addition of RNase A to the transfection mix. To analyze if the RNA-dependent activity is restricted to certain RNA species, we purified single ribosomal 5S, 16S, and 23S RNA and examined their capability to activate moDCs (Figure 1F) and PBMCs (Figure 1G). In this experimental setup, no significant changes in cytokine secretion by the different rRNA fractions were observed.

Having demonstrated that RNA from M. stadtmanae is a potent activator of human immune cells, our next step was to identify the cognate PRR involved in its recognition by human monocytes. Since knockdown approaches in primary human monocytes and dendritic cells were unable to achieve a complete inhibition of gene expression (data not shown), we decided to instead use the BLaER1 cell line (28), in which genetic modifications and stable knock-out via CRISPR/Cas9 are possible (29). This line is a B-ALL-derived cell line that can be transdifferentiated into a monocyte-like cell type, so we first confirmed whether these BLaER1 cells, upon differentiation, respond to M. stadtmanae in a similar manner to PBMCs and moDCs, and thus whether they would provide a suitable model in which to unpick the molecular mechanisms involved in recognition of this archaeon. Exposure of BLaER1 monocytes to M. stadtmanae led to an increase in TNF-α, IL-6, and IL-1β secretion 18 h after stimulation, in a similar manner to that observed with moDCs. Using Cytochalasin D, an inhibitor of phagocytosis, and Bafilomycin A1, an inhibitor of endosomal acidification, we confirmed that the recognition of M. stadtmanae in BLaER1 monocytes is also dependent on both mechanisms (Figure S2A in Supplementary Material), similar to our findings in human moDCs (19). The phagocytosis of M. stadtmanae by BLaER1 monocytes was verified by confocal microscopy (Figure S2B in Supplementary Material). These data confirmed the use of BLaER1 monocytes as a suitable model system for this study.

Our data suggest that M. stadtmanae RNA is detected by intracellular PRRs in human monocytes. All human nucleic acid-specific TLRs (TLR3 and TLR7-9) are located in the endosomal compartment, and these receptors rely on the protein UNC93B1, which is mandatory for their trafficking from the endoplasmic reticulum (ER) to the endosomes (30). To confirm our assumption that M. stadtmanae is recognized by endosomal TLRs, we generated UNC93B1^−/− cells (see Figures S3A,B in Supplementary Material for further information regarding all CRISP/Cas9-generated BLaER1-KO cells in this study) and determined whether these cells retained an ability to respond upon exposure to archaeal cells. UNC93B1^−/− BLaER1 monocytes failed to secrete any TNF-α and IL-1β after stimulation with M. stadtmanae (Figure 2A), confirming the involvement of either TLR3, or TLR7-9 in the induction of cytokine responses.

We next generated KO cells lacking MyD88 or TRIF, which are the central adapter molecules used by different TLRs for signal transduction (31). Upon exposure to M. stadtmanae, TRIF^−/− cells showed a slightly decreased but still high secretion of TNF-α and no reduction in the levels of IL-1β and RANTES (CCL5), compared with that induced in stimulated WT cells (Figure 2B). In contrast, this response was completely absent in MyD88^−/− cells exposed to this archaeal species (Figure 2B). Our results showed that M. stadtmanae is recognized through one or more MyD88-dependent TLRs by human monocytes.

As TLR9 recognizes DNA, TLR3 signals via TRIF, and human TLR7 and TLR8 are the only RNA-specific receptors that signal via MyD88 and are dependent on UNC93B1, we assumed that either or both of TLR7 and TLR8 were involved in the detection of M. stadtmanae. To identify which of these receptors recognizes M. stadtmanae, we generated single-gene knockout cell lines for each these receptors, as well as a double knockout (TLR7^−/− × TLR8^−/−) cell lines. Using these cells, we demonstrated that the secretion of TNF-α in response to M. stadtmanae can be induced via TLR7 as well as TLR8, with reduced levels of cytokine secretion detected compared with that produced by stimulated WT cells in both knockout cell lines. TLR8^−/− cells did, however, show a greater decrease in TNF-α secretion compared with that of the TLR7^−/− cells (Figure 2C). Unexpectedly, the secretion of IL-1β was entirely dependent on TLR8, as no secretion of this cytokine was detected in TLR8^−/− cells after stimulation with M. stadtmanae. The TLR7/8 agonist R848, the TLR7 agonist CL264, and the TLR8 agonist TL8–506 were used as controls to confirm the specificity of the clones used in this study (Figure 2C). Additionally, we transfected purified RNA from M. stadtmanae into TLR7^−/−, TLR8^−/−, and TLR7^−/− × TLR8^−/− BLaER1 monocytes using DOTAP and measured the secretion of TNF-α and IL-1β after 18 h. Similar to whole archaea, RNA from M. stadtmanae is recognized by TLR7 and TLR8, but the knockout of TLR8 revealed that the latter receptor has a stronger effect (Figure 2D).

Recently, TREML4 was shown to be involved in the recruitment of MyD88 to TLR7, thereby amplifying its signaling (32). However, TREML4^−/− BLaER1 monocytes showed no alterations in secretion of IL-6 or TNF-α after stimulation with M. stadtmanae or different TLR7/8 agonists (Figure S4 in Supplementary Material) indicating that this protein, at least in human monocytes, is not essential for TLR7/8 activation. To further characterize the individual roles of TLR7 and TLR8 in the response to M. stadtmanae, we used quantitative reverse-transcription polymerase chain reaction (qRT-PCR) to analyze the expression of type-I and type-III IFN mRNAs in the different TLR-KO cell lines after 8 h stimulation with M. stadtmanae. In TLR7^−/− cells, only IFN-β showed a slight reduction in expression compared with the levels in WT stimulated monocytes (Figure 2E), whereas expression of genes encoding all three IFNs, IFN-α14, IFN-β, and IFN-λ1, were significantly reduced from WT levels in TLR8^−/− cells, and an even stronger effect was detected in double-KO cells (Figure 2E). These data demonstrated that not only pro-inflammatory cytokines such as TNF-α and IL-1β but also type-I/III IFNs are induced by activation of TLR8 (and to a minor degree TLR7) with M. stadtmanae.

As mice and humans differ in their repertoire of RNA-specific TLRs (33), we also analyzed the cytokine response of bone marrow-derived dendritic cells (BMDCs) from different knockout mice after stimulation with M. stadtmanae (Figure S5 in Supplementary Material). Using these cells, we demonstrated that in mice, the recognition of this archaeon is completely dependent on mTLR7, whereas the absence of mTLR8 or mTLR13 had no effect on the secretion of IL-6.

Maturation and release of IL-1β along with that of the pro-inflammatory cytokine IL-18, occurs upon activation of caspase-1, following activation of the inflammasome and recruitment of apoptosis-associated speck-like protein containing a CARD (ASC) in response to PRR signaling. Inflammasomes are cytosolic multiprotein complexes that differ in their utilization of receptor proteins such as NLRP3 or AIM2 (20). To clarify the process of IL-1β release and inflammasome activation induced by M. stadtmanae, we first analyzed the release of active caspase-1 p20 and mature IL-1β p17 in the supernatant of human monocytes using immunoblotting. After stimulation for 18 h with M. stadtmanae, caspase-1 p20 and IL-1β p17 were detected in the supernatant of both primary human monocytes (Figure 3A) and BLaER1 monocytes (Figure 3B), indicative of inflammasome activation in both these cell types. To identify the type of inflammasome that is activated by M. stadtmanae, we first determined the effects of specific inhibitors for caspase-1 (Ac-YVAD-cmk) and NLRP3 (MCC950) (34) upon the induction of pro-inflammatory cytokine responses. Both inhibitors almost completely abolished the release of IL-1β observed in untreated primary human monocytes (Figure 3C) and BLaER1 monocytes (Figure 3D) after stimulation with M. stadtmanae, although release of IL–6 remained unaffected by either inhibitor confirming their specificity. These findings demonstrated the dependency of M. stadtmanae-induced inflammasome activation on caspase-1 and NLRP3. An equivalent observation was made for primary moDCs and PBMCs (Figures S6A,B in Supplementary Material).

To confirm activation of the NLRP3 inflammasome by M. stadtmanae, we generated additional BLaER1 cell lines in which NLRP3, ASC, or caspase-1 were knocked out, and stimulated these cells with M. stadtmanae (Figure 3E). The results were consistent with those obtained using inhibitors, showing that all three molecules were required for the secretion of IL-1β. However, NLRP3^−/− and caspase-1^−/−, but not ASC^−/−, cells also showed a decrease in IL-6 secretion after stimulation with M. stadtmanae. In addition, we tested the involvement of NEK7, a molecule that was recently shown to be required for NLRP3 inflammasome activation (35). BLaER1 NEK7^−/− cells also displayed a significant decrease in IL-1β secretion (Figure 3E), but the effect was not as strong as that in the NLRP3, ASC, and caspase-1 knockouts.

Pyroptosis, an inflammatory form of cell death, has been reported to follow NLRP3 inflammasome activation, and to be involved in the release of pro-inflammatory cytokines (20). To test whether pyroptosis was a consequence of M. stadtmanae-induced NLRP3 activation, cell death was measured by detecting the release of lactate dehydrogenase (LDH) from stimulated and unstimulated BLaER1 WT and KO monocytes. BLaER1 WT cells showed a release of approximately 20% of the amount of LDH released by the cell lysis control after stimulation with M. stadtmanae. By contrast, LDH release, and thus cell death, was undetectable in cells lacking the inflammasome components NLRP3, ASC, caspase-1, or NEK7 (Figure 3F), and was also absent from TLR8^−/− and TLR7^−/− × TLR8^−/− cells (Figure 3G) indicating that activation of TLR8 by M. stadtmanae hence leads to induction of pyroptosis.

Several pathways for inflammasome activation have previously been described, including canonical, non-canonical, and alternative activation. They differ in terms of the stimuli responsible, and the formation of ASC specks: in particular, alternative activation in response to LPS stimulation and TLR4 signaling was recently shown to be independent of ASC speck formation (29) otherwise characteristic of both canonical and non-canonical inflammasome activation. To distinguish which of these activation pathways was induced upon stimulation of human monocytes by M. stadtmanae, we examined whether ASC specks could be observed in these cells. As expected, ASC speck formation was detectable in control BLaER1 cells treated with LPS and nigericin for 4 h to induce canonical inflammasome activation (Figure 4A). Conversely, in cells treated with LPS alone, which induces the alternative activation pathway, ASC specks were absent both 4 and 18 h after treatment (Figure 4A). Although we were able to show TLR8-dependent induction of pyroptosis in BLaER1 monocytes, usually a hallmark of canonical activation, cells stimulated with M. stadtmanae showed no ASC speck formation, a phenotype more comparable to that of cells stimulated with LPS alone and alternative activation. Secretion of IL-1β in BLaER1 monocytes after 4 h with M. stadtmanae or LPS alone was relatively weak (Figure 4B), however, by 18 h, the level of IL-1β secreted by these cells was much higher and comparable to that induced by canonical inflammasome activation, despite the lack of formation of ASC specks. Taken together, these results demonstrated that M. stadtmanae triggers a TLR8-dependent NLRP3 inflammasome activation pathway in human monocytes that shares characteristics of canonical as well as alternative inflammasome activation.

Approval for these studies was obtained from the Institutional Ethics Committee at the University of Lübeck (Lübeck, Germany; Protocol no. Az. 12-202A) according to the Declaration of Helsinki. All donors gave written informed consent.

Methanosphaera stadtmanae (DSM 3091) was grown as previously described (55). Immune cell stimulation experiments were carried out with exponentially growing M. stadtmanae cells that were centrifuged at 3,200× g for 30 min, washed, and suspended in aerobic 50 mM Tris–HCl (pH 7.0).

For the isolation of nucleic acids, M. stadtmanae cells were grown as previously described (55), harvested at 4°C (3,200× g for 30 min), and lysed in liquid nitrogen using a Mikro-Dismembrator S laboratory ball mill (Sartorius) for 3 min at 1,600 bpm. RNA was isolated by TRIzol extraction followed by DNase I treatment. DNA was isolated using the Wizard Genomic DNA Purification Kit (Promega) according to the manufacturer’s protocol. Concentration and purity were determined using the DS-11 spectrophotometer (DeNovix). RNA with an A₂₆₀/A₂₈₀ ratio ≥ 2.0 was considered as pure RNA; DNA with an A₂₆₀/A₂₈₀ ratio of approximately 1.8 was considered as pure DNA.

The separation of ribosomal RNA fractions (5S, 16S, and 23S rRNA) was carried out by excision of the respective fractions after agarose gel electrophoresis and subsequent purification using NucleoSpin Gel and PCR Clean-up Kit according to the manufacturer’s protocol (Macherey-Nagel).

Peripheral blood mononuclear cells were prepared from the heparinized blood of donors by gradient centrifugation (56) using Biocoll (Merck). Subsequently, monocytes were isolated by counter-flow elutriation centrifugation (57). Monocyte-derived dendritic cells were then generated from monocytes as previously described (58). All cells were cultured in RPMI 1640 medium with stable glutamine supplemented with 10% FCS and antibiotics [100 U/mL penicillin and 100 µg/mL streptomycin (P/S; all from Merck)] from now on referred to as complete medium. Cells were grown and incubated in a humidified atmosphere of 5% carbon dioxide at 37°C.

BLaER1 cells were a gift from Thomas Graf (Center for Genomic Regulation, Barcelona, Spain). The cells were cultured under the same conditions as the moDCs and maintained at a cell density between 1 × 10⁵ and 2 × 10⁶ cells/mL. Transdifferentiation was induced by cultivating 3 × 10⁵ cells/mL for 6–7 days in complete medium with 10 ng/mL IL-3, 10 ng/mL M-CSF (both Peprotech), and 100-nM β–estradiol (Sigma), as previously described (28). For stimulation, cells were replated in complete medium without IL-3, M-CSF, or β-estradiol. The process of transdifferentiation was regularly checked using standard flow cytometry analysis with CD19-BV421, CD11b-APC, and CD14-PE antibodies (all from BioLegend) on a MACSQuant Analyzer 10 (Miltenyi Biotec).

Peripheral blood mononuclear cell and moDC were stimulated for 18 h at a cell density of 1 × 10⁵ cells/well in a 96-well flat-bottom plate with a total volume of 200 µL. Monocytes and BLaER1 cells were stimulated at a cell density of 0.25 × 10⁵ and 0.5 × 10⁵ cells/well, respectively. Whole cells of M. stadtmanae were applied at 10⁷ cells per well, unless otherwise indicated. LPS (from Escherichia coli O111, a gift from Otto Holst, Research Center Borstel) was added at a final concentration of 50 ng/mL. R848, CL264, and TL8–506 (all from InvivoGen) were used at 5 µg/mL. Archaeal RNA was complexed with the liposomal transfection reagent DOTAP (Carl Roth) prior to stimulation at a ratio of 6.5 µL DOTAP per 1 µg RNA in 50 µL of pure RPMI 1640. Total RNA was added at a concentration of 5 µg/mL, and purified rRNA was added at 2.5 µg/mL. Three hours after the addition of RNA, 50 µL of complete RPMI 1640 was added to the cells. When indicated, RNA was digested using RNase A, DNase I or Proteinase K (Thermo Fisher) prior to complexing with DOTAP as recommended by the manufacturer.

For the inhibition experiments, 7.5 µM MCC950, 50 µM Ac-YVAD-cmk (both Sigma-Aldrich), 2 µM Cytochalasin D (Sigma), or 10 nM Bafilomycin A1 (Merck) was added to cells 1 h prior to stimulation.

Gene-specific gRNA sequences were designed using Benchling¹ online software, in which off-target scores were kept as low as possible. The gRNA sequences that were used in this study are listed in Table S1 in Supplementary Material. The BbsI restriction site was used to clone ODNs (Thermo Fisher) encoding gRNA sequences into pU6-(BbsI)-CBh-Cas9-T2A-BFP [a kind gift from Ralf Kuehn (59); Addgene plasmid #64323] as previously described (60). Using the Human B Cell Nucleofector Kit and Nucleofector I device (program U-15; both Lonza), 1 × 10⁶ cells were transfected with 2 µg plasmid DNA. Forty-eight hours after transfection, GFP⁺/BFP⁺ cells were sorted into 96-well plates (1 cell/well) by the Fluorescence Cytometry core unit at the Research Center Borstel using a FACSAria IIu (BD Biosciences). Two to three weeks after sorting, DNA from the clones was isolated using QuickExtract DNA Extraction Solution (Epicenter). Specific gene fragments were amplified using standard PCR methods and sequenced at Eurofins Genomics. The occurrence of InDel mutations was analyzed using the Tracking of Indels by Decomposition (TIDE)² online software (61). Only clones with frameshift mutations on all target alleles were used for further experiments. Clones were characterized using western blot and/or ELISA depending on the availability of specific antibodies. The results of the immunoblot and TIDE analysis are shown in Figures S3A,B in Supplementary Material, respectively.

The concentrations of released cytokines in the supernatants were determined after 16 h using commercial ELISA kits specific for hIL-1β, hIL-6, mIL-6, hTNF-α (Thermo Fisher), and hRANTES (CCL5) (R&D Systems). LDH release was measured using the Pierce LDH Cytotoxicity Assay Kit (Thermo Fisher) according to the manufacturer’s instructions. LDH release (%) was calculated as follows: (stimulated sample − unstimulated sample)/(lysis control − unstimulated sample) × 100.

For expression analysis via qRT-PCR, 0.5 × 10⁶ or 1 × 10⁶ PBMCs, moDCs, or BLaER1 monocytes were stimulated for the indicated time periods (3, 6, and 12 h for PBMCs and moDCs; 8 h for BLaER1 monocytes) with M. stadtmanae cells (100:1) in a total volume of 0.5 or 1 mL at 37°C. After stimulation, cells were harvested, and RNA was isolated using the NucleoSpin RNA Isolation Kit (Macherey-Nagel) according to the manufacturer’s instructions. The concentration and purity of RNA were analyzed using a DS-11 spectrophotometer (DeNovix). Reverse Transcription of RNA into cDNA was achieved using SuperScript III, RNase Out, dNTPs, and oligo dT primer (all from Thermo Fisher) according to the manufacturer’s instructions. Quantitative PCR was performed using SYBR Green Master Mix on a LightCycler 480 II (both Roche). All primers were purchased from Thermo Fisher, and their sequences are listed in Table S2 in Supplementary Material. The following cycling conditions were applied: 95°C for 10 min (95°C for 10 s; 63 −58°C for 10 s with −0.5°C/cycle and 72°C for 6 s) × 45 cycles. The data were analyzed using LightCycler 480 software (v. 1.5.1) by calculating the ratio of the target gene expression to expression of the reference gene HPRT. Data were presented either as log2 ratio or as n-fold induction compared with unstimulated cells.

Human monocytes or BLaER1 cells were stimulated with M. stadtmanae at a ratio of 1:100 in RPMI 1640 with 1% P/S (both Merck) for 18 h. LPS at 50 ng/mL was used as a control, and 13.4 µM nigericin (Invivogen) was added for the last 2 h. Supernatants were precipitated with trichloroacetic acid (1:10, Sigma), and the pellet was washed three times with acetone and suspended in LDS buffer (Thermo Fisher) with 5% β-mercaptoethanol (Carl Roth). Cell lysates were generated by direct lysis of the cells in LDS buffer with 5% beta-mercaptoethanol. To analyze KO clones, cell lysates from BLaER1 KO monocytes were generated in the same manner. Gel electrophoresis was performed using NuPage Novex 12% Bis-Tris protein gels (Thermo Fisher) for 45 min at 200 V, and proteins were subsequently blotted onto PVDF membranes (Carl Roth) for 1 h at 30 V. Membranes were blocked using Roti-Block (Carl Roth), and primary antibodies were added overnight at 4°C. Caspase-1 p20 (Bally-1, 1:1,000, Adipogen, used for detection of p45 and p20), IL–1β (D3U3E, 1:1,000, used for detection of p31 and p17), β-actin (8H10D10, 1:5,000, both Cell Signaling), ASC (B-3, 1:1,000, Santa Cruz Biotechnology), NLRP3 (Cryo-2, 1:1,000, Adipogen), and NEK7 (EPR4900, 1:5,000, Abcam) were used as primary antibodies. Goat-anti-mouse-Alexa 680 (Thermo Fisher) and goat-anti-rabbit-IRDye 800CW (LI-COR Biosciences) were used as secondary antibodies at 1:10,000 dilution. The blots were visualized with the LI-COR Odyssey system (LI-COR Biosciences).

For confocal laser scanning microscopy, 10⁵ moDC were incubated at 37°C for 2 h on 6 channel μ-Slide (Ibidi) and subsequently stimulated for 4 h with 10⁷ M. stadtmanae cells. After fixation in 3% paraformaldehyde, primary antibodies were added at 1:100 in PBS with 3% BSA and 0.1% saponin and incubated overnight at 4°C. Cells were incubated with anti-NF-κB p65 (1:100, F-6), anti-IRF1 (1:100, B-1), or anti-IRF5 (1:100, H-56, all from Santa Cruz Biotechnology) overnight at 4°C, followed by staining with an Alexa Fluor 488-conjugated goat anti-mouse IgG (H + L) secondary antibody (1:300, Invitrogen). The nuclei were counterstained using Hoechst 33342 dye (1:3,000, Life Technologies). For detection of ASC specks, BLaER1 cells were stimulated for 4 and 18 h, and staining was performed as previously described. ASC (1:100, B-3, Santa Cruz Biotechnology) was used as the primary antibody, and Alexa Fluor 546-conjugated goat anti-mouse IgG (H + L) was used as the secondary antibody (1:300, Invitrogen). Images were captured using the TCS SP5 confocal microscope and LAS AF software (both from Leica).

The data were analyzed for statistical significance using a one-way or two-way ANOVA with a Bonferroni, Dunnett’s or Tukey post hoc test using Graph Pad Prism 7.02 software. The values of P ≤ 0.05 were considered to be statistically significant.