Microglia-specific recombination to remove Tgfbr2 was performed using the inducible Cre-LoxP system. Generation of microglia-specific Tgfbr2-knockout mice has been described previously (16). In brief, Tgfbr2flox/flox lines were crossed with Cx3cr1CreERT2 mice. These animals were further crossed with B6.129×1-Gt(ROSA)26Sortm1(EYFP)Cos/J mice which carry a stop codon flanked by LoxP before yellow fluorescent protein (YFP) to generate Cx3cr1CreERT2:R26-YFP: Tgfbr2flox/flox mice. The mice were maintained at a temperature of 22 ± 2 °C, following a 12-h light/dark cycle, with unlimited access to chow and water. All animal experiments were performed in strict adherence with the German Federal and local ethical guidelines (2024-248-Grundantrag). To induce the recombination, adult male and female mutant mice were fed with chow containing tamoxifen citrate (TAM chow) (500mg/kg body weight, ssniff, Soest, Germany) and/or control chow (ssniff, Soest, Germany) for 4 weeks. The mice were then euthanized and brains were promptly removed. Microglia were then isolated using the Adult Brain Dissociation Kit (130-107-677, Miltenyi Biotec, Bergisch Gladbach, Germany) and MACS technology as per the manufacturer´s instructions. Next, magnetic labelling and flow cytometric analysis of microglia were performed. Briefly, microglia were incubated with CD11b MicroBeads (130-126-725, Miltenyi Biotec, Bergisch Gladbach, Germany) and subsequently separated using MS Columns (130-042-201, Miltenyi Biotec, Bergisch Gladbach, Germany) yielding CD11b enriched cell population. Flow cytometry was performed using using MACSQuant 10 and MACSQuantify Software V2.13.3. Cells were stained with anti-mouse CD11b-Vioblue (130-113-810, Miltenyi Biotec, Bergisch Gladbach, Germany) at a concentration of 1:50 in flow cytometry buffer for 10 min. Afterwards, cells were washed and applied to the flow cytometer. First, cells were gated as single events and then for YFP⁺ events as the cells that underwent recombination express YFP due to the deletion of preceding stop codon. The separated cells were then subjected to RNA isolation and qPCR analysis as described in the RNA isolation and quantitative RT-PCR sections.

Primary microglia (pMG) were prepared and cultured as previously described (17). Vessels and meninges were removed from the brains of P0/P1 C57BL/6JRj mice (Janvier), and brains were washed and collected in ice-cold Hank´s Buffered Salt Solution (HBSS, Gibco, Germany). After enzymatic dissociation with Trypsin-EDTA (Gibco, Germany) for 15 minutes at 37 °C, an equal volume of fetal calf serum (FCS, Gibco, Germany) and DNase (Roche, Mannheim, Germany) at a final concentration of 0.05 mg/ml was added. Cells were dissociated using wide- and narrow-bored polished Pasteur pipettes and further centrifuged and resuspended in DMEM/F12 medium (Gibco, Germany) containing 10% FCS and 1% penicillin/streptomycin (Invitrogen). Dissociated cells from 2–3 brains were plated on poly-D-lysine-coated (Sigma-Aldrich, Schnelldorf, Germany) 75 cm² culture flasks. Cells were kept in a 5% CO₂/95% humidified atmosphere at 37 °C. After 10–14 days in culture, microglia were shaken off (250–300 rpm for 1 h) from adherent astrocytes and plated according to the experimental designs. Treatment with recombinant human TGFβ1 (Peprotech, Hamburg, Germany) was performed at a concentration of 5 ng/mL while LPS (L8274, Sigma-Aldrich, Schnelldorf, Germany) was added at a concentration of 1 µg/mL. For the activation of the inflammasome, Nigericin (InvivoGen, Toulouse, France) was used at a concentration of 1.34 mM. For in vitro TGFβ signaling inhibition, primary microglia were treated with SB431542 (Selleckchem, 5µM) for 24h or left untreated. All the treatments were performed under serum-free conditions.

BV2 cells were cultured as described by Zhou et al. (8). Cells were kept in DMEM/F12 culture medium (ThermoFisher Scientific) supplemented with 10% fetal calf serum (FCS) and 1% penicillin/streptomycin (Invitrogen) and maintained at 37 °C in a 5% CO₂/95% humidified atmosphere. Treatments with 1 µg/ml LPS and/or 5 ng/ml TGFβ1 (Peprotech, Hamburg, Germany) for qPCR and western blotting experiments were always performed under serum-free conditions.

RNA was isolated from pMG using TRIzol reagent (Invitrogen, Karlsruhe, Germany) according to manufacturer´s instructions. RNA concentration and quality were determined using the NanoDrop 2000 (Thermo Scientific, Germany). 1 µg total RNA from each sample was reverse transcribed to cDNA using Protoscript^® II First Strand cDNA Synthesis Kit (#E6560S, New England Biolabs, Frankfurt, Germany) according to the manufacturer’s instructions.

Quantitative RT-PCR was performed using the CFX Connect™ System (Bio-Rad, München, Germany) in combination with the SYBR Green GoTaq^® qPCR Kit (A6002, Promega, Madison, WI, USA). 5 µl of cDNA template was used in 20 µl reaction mixture. Results were analyzed using the CFX Connect™ System (Bio-Rad, München, Germany) Software and the comparative CT method. All data are expressed as 2^-ΔΔCT for the gene of interest normalized to the housekeeping gene Gapdh and presented as fold change relative to controls. The following primers have been used throughout this study: Nlrp3for 5´-CACCTTGTGGAGTACATGGAAC-3´, Nlrp3rev 5´- CTACCTCCCTTTCAAGACGGT-3´ [NM_133859.2], Casp1for 5´-CACCTTGTGGAGTACATGGAAC-3´, Casp1rev 5´- CTACCTCCCTTTCAAGACGGT-3´ [NM_133859.2], Il18for 5´-CACCTTGTGGAGTACATGGAAC-3´, Il18rev 5´- CTACCTCCCTTTCAAGACGGT-3´ [NM_133859.2], Il1bfor 5´-CACCTTGTGGAGTACATGGAAC-3´, Il1brev 5´- CTACCTCCCTTTCAAGACGGT-3´ [NM_133859.2], Gapdhfor 5´-GGCATTGCTCTCAATGACAA-3´, Gapdhrev 5´- ATGTAGGCCATGAGGTCCAC-3´ [NM_001289726].

Total proteins were extracted from primary microglia with RIPA Buffer (#89900, Thermo Fisher Scientific, Germany) according to the manufacturer’s instructions. Proteins were also isolated from conditioned cell culture medium. Briefly, acetone was added to culture medium at a ratio of 3:1 and incubated at -20 °C for 30min. Then the mixtures were centrifuged at maximum speed for 20min. The supernatant was discarded and the resultant pellet was resuspended in 1X PBS. Protein concentrations for both cell lysates and supernatants were determined using Pierce™ BCA Protein Assay Kit (#23225, Thermo Fischer Scientific, Germany) as per the manufacturer’s instructions. 10 mg of protein per lane was loaded into 4–20% Mini-PROTEAN^® TGX™ Precast Gels (Bio-Rad, Munich, Germany). Electrophoresis was performed at 120 volts for 90 minutes. Then, the proteins were transferred using the Trans-Blot^®Turbo™ RTA Midi PVDF Transfer Kit (#1704275, Bio-Rad, Munich, Germany) for the Trans-Blot^®Turbo™ Transfer System (Bio-Rad, Munich, Germany) onto a PVDF membrane. The membranes were then briefly rinsed with 1X Tris-buffered saline with 0.1% Tween^® 20(1X-TBST) and blocked with 5% bovine serum albumin (Carl Roth, Karlsruhe, Germany) in 1X-TBST for 90 mins at room temperature. They were later incubated with primary antibodies against Il1β (AF-401-NA, R&D Systems, 1:500), Nlrp3 (AG-20B-0014, Adipogen, 1:500), Caspase-1 (AG-20B-0042, Adipogen, 1:500) and β-Actin (#4967, Cell Signaling,1:2000) overnight at 4 °C. After the primary antibody incubation, membranes were washed with 1X-TBST three times for 10 minutes at room temperature. Then, the membranes were incubated with horseradish peroxidase (HRP)-conjugated anti-goat (sc-2354, Santa Cruz Biotechnology, 1:2000), anti-rabbit (#7074, Cell Signaling, 1:2000) and anti-mouse (#7076, Cell Signaling, 1:2000) in 1X-TBST at room temperature for 90 minutes. Then, the proteins were detected using SignalFire™ Elite ECL Reagent (#12757, Cell Signaling) and visualized using the ChemiDoc MP imaging system (Bio-Rad, Munich, Germany). Densitometric analysis of protein bands was carried out using ImageJ software (National Institutes of Health, Bethesda, MD, United States).

ILβ1 was detected in in medium from pMG cultures treated for 6 h followed by additional 2 h of nigericin treatment using a mouse IL-1 beta/IL-1F2 ELISA Kit (#MLB00C, R&D Systems, Wiesbaden-Nordenstedt, Germany) according to the manufacturer´s instructions. Absorbances were detected using a Multiskan FC plate reader (Thermo Fischer) and concentrations of ILβ1 were calculated from standard curves using GraphPad Prism10 software (GraphPad Software Inc., La Jolla, CA, USA).

The micro-array data for pMG treated with TGFβ1 used in the study is available on the Gene Expression Omnibus database (GEO) under the number GSE115652. Heatmap was made using SRplot (18). Pathway enrichment analysis was performed using Metascape (19).

Data are given as means ± standard error of the mean (SEM). Statistical differences between two groups were determined using Student´s t-test. Multiple-group analysis was performed using one-way ANOVA followed by Bonferroni´s multiple comparison post-test. P-values ≤ 0.05 were considered as being statistically significant. All statistical analyses were performed using the GraphPad Prism10 software (GraphPad Software Inc., La Jolla, CA, USA).

Inflammasome-relevant genes were screened from a previously published study (20). To confirm the pathways represented by the gene set from Hytti et al., 2021, pathway enrichment analysis was performed for GO Biological Processes, GO Cellular Components, GO Molecular Functions, and KEGG Pathway using Metascape. All genes in the genome have been used as the enrichment background. This analysis has revealed that pathways such as the NOD-like receptor signaling pathway, NF-κB signaling pathway, canonical inflammasome assembly, and cytokine-producing pathways are highly enriched (Figures 1C, D). In order to analyze the effect of TGFβ1 signaling on inflammasome-related genes, a previously published transcriptomic analysis (16) in which pMG cultures were treated with 5 ng/ml TGFβ1 for 24 h was used (Figure 1A). Results from this analysis have shown that TGFβ1 treatment reduces the expression of inflammasome-related genes such as Nlrp3, Nlrlp1a, Casp1, Casp4, and Aim1 (Figures 1B, E). In addition to this, genes involved in several other inflammatory pathways were also downregulated. Toll-like receptor 2 (TLR2) pathway genes such as Myd88 and Irak4, Nuclear factor-κB (NF-κB) signaling genes like Nfkbia, Nfkbib, and Bcl2l1, interferon regulatory factor (IRF) family genes Irf1 and Irf2, and Il1b and Il18 belonging to interleukin (IL) receptor signaling were among the genes downregulated upon TGFβ1 treatment (Figures 1B, E). Overall, these findings indicate that TGFβ1 has a significant regulatory effect on the genes involved in the inflammasome and other inflammatory pathways by significantly reducing priming of primary microglia.

In the next step, the microglia cell line BV2 was used to evaluate the effects of TGFβ1 on the expression of NLRP3 inflammasome-related genes. As shown in Figure 2, treatment of BV2 cells with LPS at a concentration of 1 µg/ml for 6 h resulted in significant upregulation of Nlrp3 (Figure 2A), Casp1 (Figure 2C) as well as the cytokines Il18 (Figure 2E) and Il1b (Figure 2G). Whereas the treatment with TGFβ1 (5 ng/ml) had no significant effects on the expression of Nlrp3, Casp1, Il18, and Il1b after 6 h, co-treatment with LPS and TGFβ1 resulted in significantly reduced expression of Nlrp3, Casp1, Il18, and Il1b as compared to the treatment with LPS alone (Figures 2A, C, E, G). Similar results were obtained after treatment of BV2 cells for 12 h. Increased expression of Nlrp3, Casp1, and Il1b were detected after treatment with LPS alone (Figures 2B, D, H). Although LPS treatment resulted in higher Il18 expression, the increase was found not to be significant (Figure 2F). The incubation of BV2 cells with a combination of LPS and TGFβ1 resulted in decreased expression of Nlrp3, Casp1, Il18, and Il1b after 12h when compared to LPS treatments alone (Figures 2B, D, F, H). However, TGFβ1 only significantly inhibited LPS-induced upregulation of Nlrp3 after treatment for 12 h (Figure 2B). Taken together, these data indicate that TGFβ1 can abrogate LPS-induced transcriptional upregulation of the inflammasome-related genes Nlrp3, Casp1, Il18, and Il1b in BV2 cells. We next analyzed the protein levels of NLRP3, CASP1, and IL1β after treatment of BV2 cells for 6 h and 12 h using western blots. As shown in Figure 3A, LPS treatment alone resulted in a significant increase of NLRP3 after 6h and 12 h. Quantification of NLRP3 confirmed this observation (Figure 3B). TGFβ1 was able to significantly inhibit the LPS-induced increase of NLRP3 at both time points (Figures 3A, B). Analysis of CASP1 protein levels revealed a slight increase after LPS treatment for 6 h and a significant increase after 12 h (Figures 3C, D). Again, TGFβ1 significantly blocked the LPS-induced increase of CASP1 levels after treatment for 12 h (Figures 3C, D). Finally, intracellular levels of IL1β were highly elevated after treatment of BV2 cells with LPS for 6 h as well as 12 h (Figures 3E, F). At both time points analyzed, TGFβ1 was able to significantly inhibit the LPS-triggered increase of IL1β. In accordance with the qPCR data, TGFβ1 inhibited the LPS-mediated increases of NLRP3, CASP1, and IL1β in the microglia cell line BV2.

Since it has been reported that BV2 display a distinct and different gene expression profile compared to pMG, we aimed to confirm the abovementioned results using primary mouse microglia cultures. Due to the limited availability of pMG, we decided to focus on treatments for 6 h. Moreover, this timepoint has been used in several studies addressing the expression of inflammasome-related genes as well as inflammasome assembly in pMG (21, 22). As demonstrated in Figure 4, treatment of pMG with LPS (1 µg/ml) resulted in robust transcriptional upregulation of Nlrp3 (Figure 4A), Casp1 (Figure 4B), Il18 (Figure 4C), as well as Il1b (Figure 4D). Noteworthy, pMG reacted with stronger increase of all 4 genes analyzed when compared to BV2 cells. Similar to the observed effects of TGFβ1 in BV2 cells, LPS-induced upregulations of Nlrp3, Casp1, Il18, and Il1b were significantly abrogated after co-treatment with TGFβ1 (Figures 4A–D). Analysis of NLRP3, CASP1, and IL1β protein levels revealed significant increases after treatment with LPS for 6 h (Figures 4E–H). In accordance with the transcriptional data, TGFβ1 significantly inhibited LPS-induced effects on NLRP3, CASP1 as well as IL1β in pMG (Figures 4E–H). It should be mentioned that the abundancy of pro-IL1β (31 kDa) after LPS treatment was much stronger than in BV2 cells indicating a more robust reaction of pMG to LPS treatment. Moreover, we were also able to detect low amounts of mature IL1β at a molecular weight of 17 kDa after LPS treatment. Together, these data confirm our result from BV2 cells and further demonstrate that TGFβ1 can inhibit LPS-induced priming as evidenced by abrogated upregulation of inflammasome-related genes in pMG.

In the next step, we analyzed inflammasome assembly and subsequent secretion of mature IL1β from pMG using Nigericin, a well-established inflammasome activator. Therefore, pMG were treated with LPS, TGFβ1 as well as the combination of both factors for 6 h. Cells were further treated for 2 h with or without Nigericin (1.34 mM). Whole cell lysates (WCL) and supernatants were used for western blots as well as IL1β ELISA. As shown in Figure 5A, Nigericin treatment resulted in robust release of pro-IL1β and mature IL1β from pMG. In the presence of TGFβ, Nigericin-triggered secretion of IL1β was significantly reduced (Figures 5A, B). Using an ELISA, concentrations of secreted IL1β after different treatments with and without Nigericin-induced inflammasome activation were analyzed in microglia supernatants. As shown in Figure 5C, very low levels of IL1β were detectable in the absence of Nigericin. However, Nigericin-triggered inflammasome activation resulted in significant secretion of IL1β (138.9 ± 23.44 pg/ml) in LPS-treated pMG. Co-treatment with TGFβ1 significantly reduced the levels of secreted IL1β (43.6 ± 0.94 pg/ml). These TGFβ1 effects are comparable to the total protein analyses shown in Figure 4, indicating that TGFβ1 is able to inhibit LPS-induced upregulation of inflammasome-related genes but seems to have a minor influence on Nigericin-induced inflammasome assembly and activation. In summary, our data demonstrate that TGFβ1 negatively regulates LPS-induced upregulation of the inflammasome genes Nlrp3, Casp1, and Il1b resulting in reduced secretion of mature IL1β from pMG after inflammasome activation.

In order to check whether inhibition of microglial TGFβ signalling affects the transcription of the inflammasome genes Nlrp3, Casp1, Il18, and Il1b, pMG were treated with SB431542, a potent inhibitor TGFβ type I receptor ALK5, for 24 h or left untreated (Figure 6A). As shown in Figures 6B–D, qPCR analysis revealed that inhibition of TGFβ signaling resulted in significantly increased expression of Nlrp3, Casp1, and Il18 in primary microglia. However, expression levels of Il1b were only slightly increased without reaching significancy (Figure 6E). These findings indicate that TGFβ signalling in microglia controls the expression levels of inflammasome genes. Since serum-free cultures of primary microglia cannot be treated for long time periods and hardly resemble the situation in vivo, we further used microglia-specific Tgfbr2 mutant mice to address whether lack of TGFβ signaling affects expression of inflammasome genes. We have previously shown that targeting Tgfbr2 in microglia leads to impaired homeostasis and render them immunologically primed (16). In order to test if this priming comprises of inflammasome related gene activation, Cx3cr1CreERT2:R26-YFP: Tgfbr2flox/flox mice were fed with control chow or chow containing tamoxifen (TAM) for 4 weeks were analysed in this study (Figure 6F). Initially, flow cytometry analysis showed increased YFP expression in TAM chow mice when compared to mice that received control chow, thereby confirming efficient recombination and deletion of Tgfbr2 in microglia (Figures 6G, H). Subsequent qPCR analysis of RNA isolated from microglia after CD11b-based immunomagnetic cell separation revealed that deletion of microglial Tgfbr2 results in significantly increased expression of Casp1, Il18, and Il1b (Figures 6J–L), while Nlrp3 transcriptional levels were comparable between both groups (Figure 6I). Together, these results indicate that silencing of microglial TGFβ signalling leads to increased expression of inflammasome genes, a change that could be an important feature of immunologically primed Tgfbr2-deficient microglia.