Activation of the Interleukin-18 Signaling Pathway via Direct Receptor Dimerization in the Absence of Interleukin-18

Materials

Plasmids containing DmrA/C sequences and the A/C Heterodimerizer ligand were obtained from Takara (Cat No. 635067). pBEF1A _IRES _PGK-Hygromycin plasmids were made in-house and used to make dual receptor expression cell lines. IL-18R1 and IL-18RAP expression fragments fused with DmrC or DmrA, respectively, were cloned into the dual expression pBEF1A backbone and were verified by sequencing. The plasmids used in this study are listed in Table 1.

Cell culture and stable cell line generation

HEK-Blue™ Null-1 cells (Cat No. hkb-null1) and Jurkat-Dual™ cells (Cat No. Jktd-isnf) were obtained from InvivoGen. HEK and Jurkat cells were maintained in DMEM or RPMI medium with 10% fetal bovine serum, 2 mM l-glutamine and 1% penicillin–streptomycin (P/S). NK-92 cells were obtained from ATCC (CRL-2407) and were cultured in MEM Alpha Medium (Life Technologies, Grand Island, NY) supplemented with 12% horse serum (Life Technologies), 12% FCS, 0.2 mM i-inositol, 20 mM folic acid, 40 mM 2-ME, 2 mM l-glutamine, antibiotics, and 100 IU/mL human rIL-2 (Chiron, Emeryville, CA).

For generating stable cell lines expressing chimeric IL-18 receptors, 1.5 × 10⁶ Null-1 Reporter HEK cells or 1 × 10⁶ Jurkat/NK-92 cells were either transfected (JetPrime reagents from Polyplus) or electroporated (Neon transfection system) with a total of 2 μg of DNA consisting of each protein expressing plasmid (Table 1) plus piggyback transposase (pBO vector) in a ratio of 3:1 following manufacturer's protocol. After 72 h, the transfected cells were subcultured in fresh media containing 400 μg/mL hygromycin for selection and expansion.

Antibodies and flow cytometry

Monoclonal antibodies for human IL-18R1 (Clone H44), human IL-18RAP (Cat No. BAF118) and mouse IL-18R1 (Cat No. MAB1216) were obtained from Biolegend, Novus biologicals and R&D systems, respectively. Rabbit polyclonal antibodies were raised in-house against mouse IL-18RAP.

For flow cytometry analyses, 1 × 10⁶ individual cells were stained with anti-human or anti-mouse primary antibodies against IL-18 receptors IL18R1 and IL-18RAP for 1 h followed by 30 min of staining with phycoerythrin-secondary antibody at 4°C. Afterward, the cells were washed, resuspended in FACS buffer, and analyzed using a FACS Attune Nxt 14 Color. The expression analyses were carried out using FlowJo_v10.8.1 software.

Reporter assay

5 × 10⁴ IL-18 chimeric receptor expressing HEK reporter cells or 1 × 10⁵ Jurkat-Dual NF-κB/interferon regulatory factor (IRF) cells were seeded into 96-well plates. The cells were stimulated with increasing concentrations of the A/C Heterodimerizer ligand (Cat No. 635055). IFNa obtained from Peprotech was used to treat the Jurkat Hu-N cell line at a concentration of 5 × 10⁴ U/mL. After overnight incubation, the reporter activities were detected through Quanti-Blue assay for HEK-blue cells and Quanti-Luc/Blue assay for Jurkat cells following manufacturer's protocol (InvivoGen).

Homogeneous time-resolved fluorescence assay

IL-18 pathway activation in NK-92 cells were measured by detecting the release of IFNγ in the supernatant. Before measuring IFNγ, detection antibodies IFNγ Eu cyprate antibody and IFNγ XL antibody from the Human IFNγ homogeneous time-resolved fluorescence (HTRF) Kit (PerkinElmer) were diluted and mixed together in detection buffer following the manufacturer's instruction (Cat No. 62HIFNGPEG). Four microliters of the antibody mixture was combined with 16 μL of NK-92 cell culture supernatant in a white 384-well low volume plate (Cat No. 784075; Greiner Bio-One).

The plate was sealed and incubated overnight at room temperature. HTRF signals were recorded using a PHERAstar microplate reader. The ratio of the acceptor and donor emission signals was calculated for each well using the following equation: ( r a t i o = 6 6 5 n m 6 2 0 n m × 1 0 4 ). The ratio was then converted to pg/mL based off the IFNγ standard curve. GraphPad prism was used to plot the data points.

Construction of stable cell lines expressing the chimeric IL-18 receptors

There has been significant interest in exploring cytokine-independent receptor dimerization for pathway activation. Different approaches were explored, including the use of diabodies (Moraga et al., 2015), leucine zipper domains (Nakase et al., 2012) and fluorescent proteins (Engelowski et al., 2018). In this study, artificial IL-18 receptor dimerization was induced by fusing the dimerization domains (DmrA and DmrC) to the receptors and supplementing the dimerization inducer, the A/C Heterodimerizer ligand.

The plasmids expressing both the IL-18R1/DmrC and IL-18RAP/DmrA chimeras (N- or C-terminus dimerization domain fusion) were transfected into HEK Null-1 NF-κB reporter cells to make stable cell lines (Fig. 1a). The activation of the IL-18 pathway can be detected using the Quanti-Blue assay (InvivoGen), which measures the activity of the secreted embryonic alkaline phosphatase (SEAP) in an absorbance assay.

OPEN IN VIEWER

FIG. 1.

Construction of reporter cell lines expressing IL-18 chimeric receptors and functional assessment of IL-18 pathway activation in HEK cells. (a) Schematic diagram of the strategies for cell line construction. (b–e) Evaluation of IL-18 chimeric receptor expression in flow cytometry and shown here are the representative histography as well as geometric mean values, the specific antibodies used were human IL-18R1 (b), human IL-18RAP (c), mouse IL-18R1 (c), and mouse IL-18RAP (d), respectively. (f) NF-κB reporter cells expressing indicative chimeric receptors were dose titrated with the A/C Heterodimerizer ligand and assessed using Quanti-Blue assay. (g) NF-κB reporter cells expressing human or mouse IL-18 receptors were dose titrated with their syngeneic IL-18 molecule and assessed using Quanti-Blue assay. Shown here are representing results of at least 3 independent experiments. All data are shown as mean ± s.d. of 3 replicates (f, g). IL, interleukin.

Cell lines with either human or murine IL-18 receptor chimeras were made and the N or C terminal fusion were designated as Hu/Ms-N or Hu/Ms-C, respectively. The expression of these receptors was verified using flow cytometry (Fig. 1b–e). Dimerization domains fused at the N-terminus showed better protein expression in both species (Fig. 1b–e). For human IL-18RAP, we can hardly detect its expression when its C-terminus was fused to DmrA (Fig. 1c). These data demonstrated that the receptor proteins did not tolerate C-terminal fusions very well.

Functional assays to test the IL-18 signaling activation

IL-18 receptors are commonly activated upon binding of the IL-18 cytokine. This process starts by the binding with the low-affinity receptor IL-18R1 and is followed by the recruitment of IL-18RAP (Yasuda et al., 2019). This resulting ternary complex then activates downstream signaling, including NF-κB (Yasuda et al., 2019). To test whether we could activate the IL-18 pathway by bridging the IL-18 receptors through chemically induced domain heterodimerization, the culture medium of the stable cell lines was supplemented with an increasing dose of the A/C Heterodimerizer.

As measured by the NF-κB reporter activation assay, a robust dose–response was detected in stimulated Hu-N cell line (Fig. 1f). Similar but weaker activation was detected in cell lines expressing the murine chimera. For cell lines expressing Hu-C, we barely detected any activity, which might be due to the failure of the expression of the chimeric IL-18RAP (Fig. 1f). To explore why significantly higher activity was observed for the cell line expressing Hu-N compared with the one expressing Ms-N, we compared the response of cell lines expressing the wild-type IL-18 receptor.

We found that the cell line expressing the human IL-18 receptors was 10-fold more sensitive to syngeneic IL-18 stimulation than the one expressing the murine receptors (EC₅₀ 5 pM versus EC₅₀ 55 pM, for the human and mouse system respectively), suggesting such difference might be an intrinsic variance between the two species or due to the compromised activity of murine receptors when expressed in a human cell line (Fig. 1g).

At concentrations above 8 μM, a hook effect was detected. This was expected as oversaturating the chimeric receptors with the A/C Heterodimerizer ligand would result in the compound binding to the receptor chains individually instead of serving as a domain dimerization bridge.

To test whether we could detect IL-18 independent pathway activation in physiologically relevant cells, Jurkat and NK-92, human T and NK cell lines, respectively, were selected as parent cell lines to make similar chimeric receptor expressing stable cell lines. Based on the results obtained in HEK cells (Fig. 1f), we only constructed the cell lines expressing Hu-N. The expression of the IL-18 chimeric receptors was validated by flow cytometry (Fig. 2a, b).

OPEN IN VIEWER

FIG. 2.

Construction of IL-18 chimeric receptor cell lines and functional assessment of IL-18 pathway activation in Jurkat and NK-92 cells. (a, b) Human N-terminal IL-18 receptor chimeras were stably expressed in Jurkat reporter (a) and NK-92 (b) cell lines. The expression of IL-18R1, IL-18RAP in these cells were evaluated in flow cytometry, histograms of chimeric cell lines stained with the isotype control or the specific antibody are shown in red and blue, respectively. Histograms of the parent Jurkat (a) or NK-92 (b) cell lines stained with the same specific antibodies are shown in orange for comparison. Shown here are representing graphs of 3 independent experiments. (c, e) Jurkat and NK-92 cells expressing the chimeric IL-18 receptors were stimulated with the A/C Heterodimerizer ligand in a dose–response experiment, and the downstream IL-18 pathway activation was measured by Quanti-Luc assay (c) and HTRF assay (f), respectively. (d) Jurkat Hu-N cells were stimulated with IFNa or the A/C Heterodimerizer ligand in a dose–response experiment, the IRF reporter activity was measured by Quanti-Blue assay. All data are presented as mean ± s.d. of 3 (c, d) or 2 replicates (e). HTRF, homogeneous time-resolved fluorescence; IRF, interferon regulatory factor.

Both the IL-18R1 and the IL-18RAP receptors were expressed in these stable cell lines and their expression levels were significantly higher than the endogenous IL-18 receptors in the parent cell lines, shown as the blue and orange histograms, respectively (Fig. 2a, b). The Jurkat cell line contains dual reporters, which enable us to detect the activation of the NF-κB pathway by the inducible NF-κB responsive Lucia luciferase in the Quanti-Luc assay and the interferon pathway by the IRF-inducible SEAP in the Quanti-Blue assays (InvivoGen).

For NK-92 cells, IFNγ was detected from the supernatant of the treated cells in HTRF assays. As shown in Fig. 2c and e, we detected dose-dependent activation by adding the A/C Heterodimerizer in these engineered cell lines by measuring NF-κB reporter activity or IFNγ secretion directly. To further confirm the DmrA/C system does not activate other receptors nonspecifically, we tested whether the A/C Heterodimerizer could activate the interferon pathway in the Jurkat Hu-N cell line. In Fig. 2d, we could detect robust interferon pathway activation upon IFNa treatment but not by supplementing A/C Heterodimerizer ligand, demonstrating the DmrA/C system has no effect on the interferon receptors.

Compared with the HEK cells, we noticed that the Jurkat and NK cells were much more sensitive as we started to detect activity at low picomolar concentrations of the compound. Similarly, we detected the hook effect at higher compound concentrations. Overall, these results demonstrated that the IL-18 signaling pathway can be directly activated by receptor dimerization, even in the absence of IL-18.