Expression of IRGs is increased in whole blood obtained from patients with type I interferonopathies

As type I IFN has been shown to up-regulate inflammasome activation in human monocytes in vitro from healthy donors and in SLE patients characterised by high levels of type I IFN (Liu et al, 2017), we set to analyse the expression of inflammasome genes in patients with monogenic type I interferonopathies. For this, we used the RNAseq data set from our previously published project (Duncan et al, 2019). This was obtained from whole blood samples from age-matched healthy donors (n = 5) and patients with mutations in different genes that cause monogenic type I interferonopathies (ACP5, ADAR1, DNASE2, IFIH1, RNASEH2A, RNASEH2B, RNASEH2C, RNASET2, SAMHD1, STAT2, TMEM173, TREX1, and PEPD [n = 42 overall]). The expression of the following ISGs was analysed: STAT1, STAT2, STAT3, IRF9, PKR, CGAS, STING, OAS1, OAS2, OAS3, IFI16, RSAD2, CXCL9, CXCL10, and ISG15. We observed an increased expression of ISGs in patients compared with healthy controls (Fig 1A and B), confirming the previously described type I IFN signature in these patients. In addition, PKR, RSAD2, CGAS, and STING genes were found to be up-regulated in patients compared with the control group (Figs 1A and S1). The analysis of inflammasome gene expression showed an up-regulation of the expression of inflammasome-associated genes (Fig 1C and D) NLRP3 (P = 0.0309), ASC (P = 0.0403), CASP1 (P = 0.0376), IL18 (P = 0.0458), and GSDMD (P = 0.0008) in donors with interferonopathies compared with healthy controls. In addition, other IRGs were analysed and there was a general trend towards increased expression compared with controls (Fig S1). Thus, our data suggest an association of NLRP3 inflammasome gene transcription with type I IFN in these patients.

Type I IFN induces the gene expression of CASP1 and GSDMD in human macrophages in vitro

After the observation of an increased expression of inflammasome component genes in patients with monogenic interferonopathies, we undertook an in vitro assay to determine the direct effect of type I IFN on the expression of inflammasome genes in healthy macrophages eliminating other confounding factors present in these patients. We treated human blood MDMs from healthy donors, in the monocytic human cell line THP-1 (PMA-differentiated) frequently used as a model of human macrophages for 4 h with different concentrations of IFN-α and IFN-β, and then performed a quantitative polymerase chain reaction (qRT-PCR) to determine expression levels of ISGs and IRGs. To determine whether the results obtained were only applicable to macrophages, we also investigated gene expression in CD14⁺ human blood monocytes from healthy donors. We found a dose-dependent induced expression of the ISGs IRF9, ISG15, CXCL9, CXCL10, and USP18 in all cell types treated with type I IFN compared with untreated cells (UT) (Figs 2A and S2A and C). This shows that these cells are responsive to type I IFN.

Next, we analysed the expression of IRGs NLRP3, CASP1, GSDMD, IL18, and IL1B. We detected an up-regulation of CASP1 expression in MDMs, THP-1 cells, and human monocytes treated with high concentrations of IFN-α and IFN-β (100 and 1,000 U/ml) (Figs 2B and S2B and D). In addition, in MDMs we observed a trend towards an increased expression of GSDMD, which was significant in IFN-treated THP-1 cells (at 100 U/ml) (Fig 2B and D). NLRP3 expression decreased in MDMs when they were treated with IFN-β at high concentrations, but no effect on THP-1 cells or human monocytes was observed. In addition, IL18 expression tended to be reduced after treatment with type I IFN in both cell types. No effects on IL1B expression were detected in any of the cell types after IFN treatment. These data show that the gene expression of only CASP1 and GSDMD is up-regulated by type I IFN after short-term treatment of 4 h, and that NLRP3 expression is down-regulated by IFN-α in MDMs.

LPS priming does not alter IFN-mediated up-regulation of CASP1 and GSDMD gene expression in human macrophages

After confirming that MDMs and PMA-differentiated THP-1 cells can respond to type I IFN and that this leads to an increase in CASP1 and GSDMD gene expression, we decided to investigate whether this resulted in an altered activation of the NLRP3 inflammasome. As the priming step plays a key role in NLRP3 inflammasome activation, we first tested whether the observed effects of type I IFN on CASP1 and GSDMD expression were altered by macrophage priming with LPS. To assess this, we used IFN-α because we did not see major differences in gene expression between IFN-α or IFN-β treatments (Fig 2) and it had been used in previous studies that showed inflammasome regulation by this cytokine (Liu et al, 2017). We treated human MDMs from healthy donors with IFN-α (1,000 U/ml) for 6 h (short term, ST) or overnight (ON, 18 h). MDMs were co-treated with IFN-α in the presence or absence of LPS (10 ng/ml) for the last 6 h of each different treatment (ST and ON) (Fig 3A). Samples were collected at this point, and qRT-PCR was performed to measure the expression of ISGs (Fig 3B) and IRGs (Fig 3C). We found that the different ISGs tested (IRF9, ISG15, CXCL9, CXCL10, USP18, and STAT1) were up-regulated after IFN-α treatment alone, as we found before (Fig 2), and that this increase was time-dependent. The induced expression of ISGs by IFN-α alone was not altered by LPS treatment (Fig 3B). For the tested IRGs (NLRP3, CASP1, GSDMD, IL1B, ASC, and IL18), we found that NLRP3 expression was significantly decreased in MDMs after treatment with IFN-α alone for 6 h (Fig 3C), similar to the effect observed at 4 h (Fig 2B), and then significantly increased when treated with IFN-α-ON compared with untreated cells and IFN-α-ST treatment (Fig 3C). LPS treatment for 6 and 4 h alone also led to a decrease in NLRP3 mRNA that was reversed by IFN-α treatment overnight (Figs 3C and S3A). CASP1 and GSDMD increased their expression with IFN-α treatment alone at both time points tested, and this was not altered by the presence of LPS. IL1B gene expression was reduced by IFN-α treatments alone at both time points. We observed that LPS alone induced an initial increase in IL1B gene expression at 4-h treatment (Fig S3B), followed by a reduction at 6-h treatment (Fig 3C) (although protein levels were still induced at this latest time point [Fig S3C]), and that these gene expression levels at 6 h were not changed by the treatment with IFN-α. IFN-α treatment did not alter ASC and IL18 expression levels, even in the presence of LPS.

Having established the effect of LPS priming on the type I IFN–induced gene expression of NLRP3 inflammasome components, we next investigated whether the priming step alone, in the presence of type I IFN, led to inflammasome activation. To assess this, we treated human MDMs and CD14⁺ monocytes from healthy donors with IFN-α (1,000 U/ml) for 6 h (ST) or overnight (ON, 18 h). MDMs and monocytes were co-treated with IFN-α and LPS (10 ng/ml) or left untreated for the last 6 h of IFN treatments as in Fig 3A. Then, supernatants were collected to determine lactate dehydrogenase (LDH) enzyme release as a measure of cell death, caspase-1 activity, and the release of the pro-inflammatory cytokines IL-1β and IL-18 (Figs 4A–D and S4A–D). In MDMs, we found that none of the treatments induced cell death (Fig 4A), whereas we observed an increase in caspase-1 activity in cell supernatants (Fig 4B), as well as IL-1β release in MDMs treated with IFN-α-ON and LPS compared with the group treated with LPS alone (Fig 4C). In contrast, IL-18 release was not affected (Fig 4D). In human monocytes, no changes were observed in cell death or the release of IL-1β and IL-18, whereas caspase-1 activity was decreased when cells were pre-treated with IFN-α-ON compared with untreated monocytes or those treated with IFN-α-ST (Fig S4B).

Type I IFN treatment alters IL-1β and IL-18 secretion after the activation of the NLRP3 inflammasome in human macrophages

We next wanted to see whether the regulation of the expression of the IRGs was reflected by an altered inflammasome activation response. Human MDMs and monocytes from healthy donors were treated with IFN-α-ON or IFN-α-ST and then in the last 6 h co-treated with LPS (10 ng/ml) to induce priming of the inflammasome (as in Fig 3), or left untreated, and then activated with ATP 5 mM for 2 h in the presence of LPS (Figs 4E and S4E). ATP is a known ligand of the P2X7 receptor and leads to the activation of the inflammasome in a less saturating way than the nigericin toxin (Herman & Pasinetti, 2018), which is useful to monitor the effects of type I IFN without being masked by a rapid cell death that could be caused by a more potent activator. In MDMs, treatment of LPS-primed cells with ATP was successful in activating the inflammasome as it increased cell death, caspase-1 activity, and the release of the cytokines IL-1β and IL-18 (Fig 4F–I) when compared to the control. This was also confirmed by Western blot of IL-1β (Fig 4J and K) in supernatants. We observed that IFN treatment led to a slight decrease in LDH release after ATP activation (Fig 4F), as well as a reduction in secretion of IL-1β and IL-18 in response to ATP (Fig 4H and I). IFN treatment, however, did not alter caspase-1 activity induced by ATP, but increased caspase-1 activity induced by LPS alone (Fig 4G). Meanwhile, in human monocytes, a decrease in cell death, as well as a decrease in caspase-1 activity, was observed in monocytes treated with IFN-α-ON in the absence of LPS priming (Fig S4F and G). Despite this, no changes were observed in the release of IL-1β (Fig S4H) and IL-18 (Fig S4I) triggered by LPS and ATP.

To determine whether the absence of regulation of inflammasome activation by IFN-α was just limited to NLRP3 activation by P2X7R, we next tested the effect of IFN treatment on inflammasome activation by the more potent NLRP3 trigger, nigericin (Herman & Pasinetti, 2018). Here, MDMs from healthy donors were treated for 4 h or ON with IFN-α (1,000 U/ml) and LPS was added for the last 2 h of the treatment followed by nigericin treatment (10 μM) for 45 min (Fig 5A). We observed that the inflammasome was activated in response to LPS priming and nigericin as expected and as evidenced by increased cell death, caspase-1 activity, and the release of IL-1β and IL-18 (Fig 5B–I). Pre-treatment with IFN-α at different time points had no effect on the levels of inflammasome activation mediated by nigericin (Fig 5B–I). These experiments were also carried out in PMA-differentiated THP-1 cells (Fig S5A–I). Although the inflammasome was activated after LPS and nigericin treatment, no type I IFN–mediated changes were observed in this cell type either.

Differentiation of CD14⁺ cells into macrophages in the presence of type I IFN does not affect MDM response to inflammasome activation

Interferonopathies are characterised by the presence of persistently high levels of IFN in the body (d’Angelo et al, 2021; Crow & Stetson, 2022). Hence, we wanted to determine the effect that differentiation of human CD14⁺ monocytes into MDMs in the presence of IFN has on the subsequent inflammasome response. Thus, human monocytes from healthy donors were treated with M-CSF (0.5 ng/ml) as usual (Sierra-Filardi et al, 2014) in the presence or absence of IFN-α at 1 and 5 U/ml. After 3 d, half of the medium was removed and supplemented again with M-CSF and type I IFN as before. After 7 d, no change in cell death (Fig 6A and B) or basal IL-1β (Fig 6C) or IL-18 (Fig 6D) release was observed compared with MDMs differentiated in the absence of IFN-α. The expression of ISGs IRF9, ISG15, CXCL9, CXCL10, and USP18 was analysed via qRT-PCR to confirm the effect of IFN-α treatment on these cells. A significant increase in the expression of IRF9 and ISG15 was observed after treatment with 1 or 5 U/ml of IFN-α, whereas CXCL9, CXCL10, and USP18 only showed a tendency to increase, but this was not significant (Fig 6E). Subsequently, MDMs were treated with LPS (10 ng/ml) for 2 h to induce priming followed by treatment with nigericin (10 μM) for 45 min to activate the inflammasome. We observed the expected effects of nigericin on inflammasome activation through increased cell death, caspase-1 activity, and the release of IL-1β and IL-18 (Fig 6F and G) in all different treatments; however, there were no differences in the levels of inflammasome activation between cells differentiated in the presence of IFN-α at 1 U/ml (Fig 6F), IFN-α at 5 U/ml (Fig 6G), or those differentiated in the absence of type I IFN. This suggests that in vitro differentiation of monocytes into MDMs in the presence of type I IFN does not alter NLRP3 inflammasome activation in this experimental context.

Reagents and antibodies

LPS (Escherichia coli 026:B6), adenosine 5′-triphosphate disodium salt hydrate (A2383; ATP), nigericin (N7143), and penicillin–streptomycin (Pen/Strep, P4333) were purchased from Sigma-Aldrich. Human IFN Beta (Beta 1a, Mammalian Expressed) (11415-1) and Human IFN Alpha Hybrid (Universal Type I IFN) (11200-1) were obtained from PBL Assay Science. Phosphatase inhibitor cocktail (78420) was sourced from Thermo Fisher Scientific, and FBS was obtained from Gibco.

Primary antibodies for Western blot analysis and their final concentrations were as follows: anti-human IL-1β (0.1 μg/ml, goat polyclonal, AF-201-NA; R&D Systems) and anti-β-actin HRP (0.2 μg/ml, mouse monoclonal, A3854; Sigma-Aldrich).

Cell cultures and treatments

THP-1 cells were purchased from ATCC and were cultured in RPMI with 10% FBS and Pen/Strep (100 U/ml) and plated at a concentration of 1 × 10⁶ cells/ml. PBMCs were obtained from the National Blood Transfusion Service with full ethical approval from the Proportionate Review Sub-committee of the London—South East Research Ethics Committee (REC reference: 21/PR/0526). PBMCs were isolated from leucocyte cones from healthy donors by density centrifugation using a Ficoll gradient. The PBMC layer was separated and washed to remove platelets. Monocytes were harvested by incubation of the PBMCs with magnetic CD14 MicroBeads (130-050-201; Miltenyi) for 15 min at 4°C. Monocytes were positively selected for using an LS column (130-042-401; Miltenyi). For the experiments where only monocytes were used, they were cultured (at a concentration of 1 × 10⁶ cells per millilitre) in RPMI supplemented with 10% FBS, Pen/Strep (100 U/ml). To differentiate MDMs, monocytes were cultured for 6 d (at a concentration of 5 × 10⁵ cells per millilitre) in RPMI supplemented with 10% FBS, Pen/Strep (100 U/ml) and treated with 0.5 ng/ml M-CSF (300-25; PeproTech). On day 3, half of the media was removed and replaced with fresh media.

Cells were untreated or incubated for 4 h with IFN-α (1,000 or 100 or 10 U/ml) or IFN-β (1,000 or 100 or 10 U/ml). Total RNA was isolated and used for real-time qRT-PCR.

In ATP activation experiments, human monocytes or MDMs were untreated or pre-incubated with IFN-α (1,000 U/ml) for 6 h (short term, ST) or overnight (18 h) (ON). In the last 6 h of incubation, cells were left untreated or primed for 2 h with LPS (10 ng/ml) in ET buffer (147 mM NaCl, 10 mM Hepes, 13 mM D-glucose, 2 mM KCl, 2 mM CaCl₂, and 1 mM MgCl₂). Cells were treated with ATP (5 mM) in the last 2 h of the experiment.

In nigericin activation experiments, THP-1 cells and MDMs were untreated or pre-incubated with IFN-α (1,000 U/ml) for 4 h or overnight (18 h) (ON). In the last 2 h of incubation, cells were untreated or primed for 2 h with LPS (10 ng/ml). The priming stimulus was then removed and replaced with ET buffer. Cells were treated with nigericin (10 μM, 45 min).

RNA isolation and real-time qPCR

Total RNA was isolated from THP-1 cells, human monocytes, and MDMs using RNeasy Mini Kit (74104; QIAGEN) according to the manufacturer’s instructions. 500 ng of total RNA was reverse-transcribed into cDNA using High-Capacity RNA-to-cDNA Kit (4387406; Applied Biosystems). Real-time qPCR was performed in triplicate using Fast SYBR Green Master Mix (4385612; Thermo Fisher Scientific). The sequence of the primers used in this study is described in Table 1. Data were normalised to expression levels of the housekeeping gene HPRT1 (QT00059066) across each treatment (Ishii et al, 2006). Representative melt curve plots for each of the primers used can be found in Fig S6. Gene expression was calculated by the comparative threshold cycle (Ct) method, and fold change compared with basal RNA levels of untreated cells was expressed as 2^−ΔΔCt.

RNA sequencing

Whole blood samples from three patients with mutations in ACP5, six with mutations in ADAR1, three with mutations in DNASE2, two with mutations in IFIH1, three with mutations in RNASEH2A, eight with mutations in RNASEH2B, six with mutations in RNASEH2C, one with mutation in RNASET2, eight with mutations in SAMHD1, four with mutations in STAT2, three with mutations in TMEM173, eight with mutations in TREX1, three with mutations in PEPD, and five age-matched controls were analysed. RNA integrity was analysed with Agilent 2100 Bioanalyzer (Agilent Technologies). mRNA purification and fragmentation, complementary DNA (cDNA) synthesis, and target amplification were performed using Illumina TruSeq RNA Sample Preparation Kit (Illumina). Pooled cDNA libraries were sequenced using the HiSeq 4000 Illumina platform (Illumina). For this, we used the RNAseq data set from our previously published project (Duncan et al, 2019). Written informed consent for these studies was provided, and ethical/institutional approval was granted by the South Central-Hampshire A (ref: 17/SC/0026) and Leeds (East) (ref: 07/Q1206/7).

Cell death assay

The supernatant was centrifuged for 5 min at 500g at 4°C to remove any cells. Cell death was quantified using a colorimetric assay for the release of LDH into cell supernatants using CytoTox 96 Nonradioactive Cytotoxicity Assay (G1780; Promega), according to the manufacturer’s instructions. Absorbance values were measured at 490 nm, and the results were expressed as a percentage normalised to total cell lysis or as a fold change relative to untreated cells.

Caspase-1 inflammasome assay

Caspase-1 release was measured using a quantitative assay of cell supernatants. Caspase-Glo 1 Inflammasome Assay (G9951; Promega) was used according to the manufacturer’s instructions. The luminescence of the assay plates was read. The results were expressed as a fold change relative to untreated cells.

ELISA

Levels of human IL-1β (DY201) and IL-18 (DY318) were measured in the cell supernatants using ELISA kits from R&D Systems. ELISAs were performed following the manufacturer’s instructions.

Western blot analysis

Cells were lysed on ice using a RIPA lysis buffer (50 mM Tris–HCl, pH 8, 150 mM NaCl, 1% NP-40, 0.5% sodium deoxycholate, and 0.1% SDS), supplemented with a protease inhibitor cocktail (P8340, 1:100; Sigma-Aldrich). Lysates were then centrifuged at 21,000g for 10 min to remove the insoluble fraction. Protein concentrations of each sample were measured using the BCA assay (23225; Thermo Fisher Scientific Pierce), following the manufacturer’s guidelines, to standardise the amount of protein in each sample. Cell supernatants were centrifuged at 500g for 5 min to remove cells and concentrated using 10-kD MW cut-off filters (Amicon, Merck Millipore), as described by the manufacturer. Supernatants and lysates were diluted in 1× reducing Laemmli buffer containing 1% β-mercaptoethanol. Samples were heated at 95°C for 5 min and separated by Tris-glycine SDS–PAGE. Proteins were transferred onto nitrocellulose membranes (0.2 μm), followed by blocking with PBS–Tween (0.1%) containing 5% skimmed milk for 1 h at RT. Membranes were then incubated overnight with the specific primary antibody in blocking buffer at 4°C. The next day, membranes were labelled with an HRP-conjugated secondary antibody for 1 h at RT. After washing, membranes were imaged using Clarity Western ECL Blotting Substrate (1705061; Bio-Rad) in ChemiDoc MP Imager (Bio-Rad). Densitometry analysis was performed using ImageJ (rsb.info.nih.gov) to measure the intensity of mIL-1β bands. Processed IL-1β release was compared with control treatment and expressed as fold change.

Statistical analysis

GraphPad Prism 10 software was used to carry out all statistical analysis. Data were first tested for normality using the Shapiro–Wilk test. Differences between 3+ groups were analysed using one-way ANOVA with the post hoc Dunnett test or two-way ANOVA with the post hoc Tukey test for multiple comparisons. Data were shown as the mean ± SEM. Two-way ANOVA used a multiple comparison. Accepted levels of significance were *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. Where no significance levels are indicated, no significance difference was detected. In Figs 4 and S4, significance levels are only shown within the unprimed and LPS-primed treatments, but not between them. In addition, significance was indicated between the UT, IFN-α-ST, and IFN-α-ON groups.