Coordinated Activation of Toll-Like Receptor8 (TLR8) and NLRP3 by the TLR8 Agonist, VTX-2337, Ignites Tumoricidal Natural Killer Cell Activity

Reagents

Fluorochrome-conjugated mAbs against CD3, CD16, CD56, IFNγ, and CD107a were purchased from eBiosciences (San Diego, CA). RPMI culture media, phosphate-buffered saline (PBS), penicillin-streptomycin, and L-glutamine were purchased from Invitrogen Life Technologies (Grand Island, NY). VTX-2337 (VentiRx Pharmaceuticals, Seattle, WA) is a synthetic, small molecule TLR8 agonist based on a 2-aminobenzazepine core structure, previously described and characterized [3]. Acetone was purchased from Fisher Scientific (Houston, TX). The neutralizing anti-IL-18 mAb was purchased from Invitrogen-Biosource (Camarillo, CA).

Analysis of IL-1β secretion and caspase-1 activation in THP-1 cells

Both wild type THP-1 and NLRP3 defective (NLRP3^def) THP-1 cells were purchased from InvivoGen (San Diego, CA, Catalog numbers: thp-null and thp-dnlp). THP-1 cells or NLRP3^def THP-1 cells (180,000 cells/well) were seeded in a 96-well plate, differentiated with 12-O-tetradecanoylphorbol-13-acetate (TPA), and treated with VTX-2337 (3 nM to 10 μM) overnight to assess IL-1β secretion. Levels of IL-1β in culture supernatant were analyzed using an ELISA kit from eBiosciences. In some studies, HEK-Blue^™ IL-1β cells (InvivoGen), were used to assess levels of IL-1β secreted into the media by THP-1 cells. The production of secreted embryonic alkaline phosphatase (SEAP) by HEK-Blue^™ IL-1β cells in response to IL-1 was quantified using the QUANTI-Blue detection media and reported as OD650.

Western blot analysis of IL-1 β, pro-IL-1β, and caspase-1

Protein levels of IL-1β, pro-IL-1β, and caspase-1 in both culture supernatant and whole cell lysate were evaluated using Western blot. After washing, the cells were re-suspended in serum-free RPMI with or without VTX-2337 (3 or 10 μM). Lysates of THP-1 cells were prepared by using the RIPA lysis buffer (0.15M NaCl, 1% NP-40, 0.1% SDS, 50 mM Tris pH8.0) with protease inhibitors. Proteins in culture supernatant from control or VTX-2337 treated THP-1 cells were precipitated using acetone. Proteins were separated on a 12% SDS-PAGE gel (Invitrogen) and transferred to nitrocellulose membrane. The membrane was first probed with anti-IL-1β mAb (clone 3ZD, from NCI-FCRDC), anti-caspase-1 mAb (clone D57A2, from Cell Signaling Technology, Boston, MA), or anti-β-actin (clone 13E5, from Cell Signaling Technology). After washing, the membrane was incubated with horseradish peroxidase-labeled secondary antibodies allowing the signal for IL-1β and caspase-1 to be detected by chemiluminescence (SuperSignal West Pico, Thermo Scientific, Rockford, IL). For β-actin, the membrane was incubated with DyLight-labeled secondary antibody and the signal was detected using LI-COR Odyssey.

In vitro studies utilizing human PBMC

Astarte Biologics (Redmond, WA) was contracted by VentiRx Pharmaceuticals to provide blood samples from healthy donors. Therefore, the protocol for the collection of blood by Astarte Biologics was approved by their Institutional Review Board (IRB), Schulman and Associates. The blood collection was performed following written informed consent. PBMC were isolated from the blood by Ficoll gradient centrifugation, suspended in RPMI, placed in 96-well plates (200,000 cells/well), treated with serial dilutions of VTX-2337 (0.003 μM-10 μM), and incubated overnight. The culture supernatant was analyzed for levels of IL-1β and IL-18 by ELISA. To demonstrate that caspase-1 activation is required for VTX-2337 induced IL-1β and IL-18 production, the caspase-1 inhibitor z-VAD-FMK (10 μg/ml, Invitrogen) was added to the culture medium 30 minutes before the addition of VTX-2337 (1 μM). The PBMC were incubated overnight, and the levels of IL-1β, IL-18, IFNγ and TNFα were analyzed by ELISA using kits from eBiosciences.

Real-time RT-PCR analysis of pro-IL-1β and pro-IL-18 mRNA expression

Total RNA was extracted from THP-1 cells or PBMC isolated from blood obtained from healthy donors by Astarte Biologics, as previously described, was treated with VTX-2337 (3 or 10 μM, 24 h) using an RNAqueous4PCR kit (Ambion, Austin, TX). The cDNA was synthesized using the Superscript III reverse transcription kit (Invitrogen). Real-time Taqman PCR was run on an ABI 7900HT instrument using primer and probes from Applied Biosystems (Foster City, CA). The level of target gene expression was normalized to β-actin using the delta Ct method as previously described [14].

TruCulture^® blood collection and whole blood culture system

Whole blood from healthy volunteers was collected using the TruCulture^® system (Myriad RBM, Inc., Austin, TX) as previously described [15]. As previously described, Astarte Biologics was contracted to VentiRx Pharmaceuticals to perform the collection. The blood was collected following written informed consent under a protocol approved by the IRB used by Astarte. The TruCulture^® syringe tubes used in this study were preloaded with culture medium only, or VTX-2337 at a final concentration of 0.3 or 1 μM and stored at -20°C, then thawed the day before blood collection. Blood was drawn into the tubes and incubated in a dry heat block incubator for 24 h. Following the incubation, cells were separated from the supernatant using the supplier plunger, then frozen at -20°C until the time of cytokine quantification using the human MAP v.1.6 inflammation panel (Myriad RBM).

Flow cytometric analysis of CD107a and IFNγ expression by NK cells

To evaluate NK cell activation, PBMC were cultured overnight in RPMI + 10% Human AB serum with or without VTX-2337 (0.50 μM). Some samples were further stimulated by exposing the PBMCs to either K562 tumor cells at a 5:1 ratio or to plate-bound anti-CD16 mAb for the last 5 h of incubation. Brefeldin A was included during the last 4 h of incubation to block the secretion of IFNγ, while CD107a-PE was added for 5 h. Following the culture activation, cells were stained with fluorophore-conjugated antibodies to surface markers including anti-CD3-Alexa 488, anti-CD16-PerCP-Cy5.5, and anti-CD56-APC. After subsequent fixation and permeabilization, the cells were stained with anti-IFNγ-eFluor450. Samples were analyzed using a FACS Canto II instrument, and data collected in list mode were analyzed using Flow Jo software (Ashland, OR).

In vivo administration of VTX-2337 in cynomolgus monkeys

Studies in cynomolgus monkeys were conducted at Charles River Laboratories (CRL), Preclinical Services, (Shrewsbury MA) in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the National Institutes of Health. The study was reviewed and approved by the CRL Institutional Animal Care and Use Committee, under submission number DPKW-101. Study animals were colony animals that were returned to the colony on completion of the study. The male monkeys (2.9–4.9 kg) were housed individually (cage dimensions of 0.76 m wide x 0.74 m deep x 0.81 m in height), but commingled periodically as part of the environmental enrichment program. The animals were also given fruit, vegetable, or additional supplements as a form of environmental enrichment, as well as given various cage enrichment devices. Animals were given Certified Primate Diet #2055C (Harlan Teklad), two times daily and water ad libitum. Environmental controls for the housing were set to maintain 18–26°C, a relative humidity of 30–70%, a minimum of 10 room air changes/h and a 12-h light/12 h dark cycle. While doses of VTX-2337 were well tolerated, provisions including use of anti-inflammatory agents to moderate the immune response were considered in the study design.

VTX-2337 was administered as a bolus subcutaneous (SC) injection in the intrascapular area at doses of 1 and 10 mg/kg. Blood samples were collected at baseline (pre-dose), and 6, 12, 24, and 96 h post injection to monitor levels of IL-1β and IL-18 in the plasma using the human MAP v.1.6 inflammation panel (Myriad RBM). Due to the routine, non-invasive procedures for dosing and blood collection, anesthetics were not considered necessary for the study.

Administration of VTX-2337 to patients with head and neck cancer and immune monitoring of NK cell responses in treated patients

The safety and tolerability of cetuximab in combination with VTX-2337 was evaluated in a Phase 1 clinical trial in adult patients with advanced recurrent squamous cell carcinomas of the head and neck (SCCHN) (Study A103; ClinicalTrials.gov NCT01334177). The study was conducted at a single study center (University of Washington, Seattle Cancer Care Alliance, Seattle, WA, USA) from June 2011 to June 2014 and was performed in accordance with good clinical practice guidelines and the ethical principles outlined in the Declaration of Helsinki. Approval for study procedures was obtained from the institutional review board of the study site, and all subjects provided written informed consent before study enrollment. Patients who were eligible for this study were adults with advanced or recurrent SCCHN that was no longer amenable to treatment by surgery or radiation therapy or patients with distant or metastatic disease.

The primary objective this study was to determine the safety, tolerability and to assess the principal toxicities of VTX-2337 when given in conjunction with cetuximab. The secondary objective was to determine the pharmacodynamic response of VTX-2337 in combination with cetuximab. The primary endpoint was to determine the maximum tolerated dose (MTD)/recommended Phase 2 dose (RP2D) and to define the toxicities of VTX-2337 in combination with cetuximab. Secondary endpoints included the analysis of biologic correlative assays. The sample size was depended upon the observed safety profile, which determined the number of patients per dose level and the number of dose escalations.

Study medications (cetuximab and VTX-2337) were administered in the clinic by appropriately qualified and trained personnel. This was an open-label study with no blinding. Each patient in this dose-escalation study was assigned to a dose level of VTX-2337 at the time of study enrollment. For each cohort, cetuximab was administered using a loading dose (400 mg/m² IV), followed by a weekly maintenance dose (250 mg/m², IV). Each cetuximab dose was administered as an IV infusion: the initial dose was infused over 2 h, subsequent doses were administered over 1 h. VTX-2337 was administered by the SC route on days 1, 8 and 15 of a 28-day treatment cycle. The first cohort received a 2.5 mg/m² dose of VTX-2337 following cetuximab administration; this dose was escalated in subsequent cohorts using a 3+3 design to 3.0 mg/m² and finally 3.5 mg/m². After successful completion of Cycle 1, patients were eligible to receive subsequent treatment cycles until the criteria for study discontinuation or withdrawal were met, including disease progression, intolerable toxicity, or death. A Consort Flow Diagram for the Clinical study evaluating VTX-2337 in adults with advanced or recurrent SCCHN is provided as Fig 1. The Study Protocol, VTX-2337 Phase 1 Trial in SCCHN Protocol A103 is available as supporting information, S1 Protocol. A TREND Statement Checklist for the study is provided as supporting information, S1 Checklist.

Heparinized blood was collected pre-VTX-2337 dose and at 24 h post-dose to assess NK cell function. Freshly isolated PBMC were cultured in medium alone, stimulated with K562 cells that express transmembrane IL-15 and the 4-1-BB ligand (K562-15mb-41BBL) (16) at a 5:1 ratio, or stimulated with plate-bound anti-CD16 mAb (3G8, BD Biosciences, San Jose, CA). Flow cytometric measurement of surface CD107a expression in CD3^-CD56⁺ NK cells was performed similarly to the method described above, using anti-CD3-AF488, anti-CD56-APC, and anti-CD107a-PE (all, eBioscience, San Diego, CA).

Statistical analysis

Statistical analysis was performed using GraphPad Prism, version 6 for Windows software (San Diego, CA). Differences between the treatment groups were analyzed using the two-tailed unpaired Student’s t test or the Wilcoxon signed-rank test for matched pair analysis when normal distribution could not be assumed. A value of p<0.05 was considered statistically significant. Missing data was not imputed.

The datasets from experiments and studies presented in Figs 2–6 are provided as supporting information: S1 File.

VTX-2337 induces NLRP3 inflammasome activation and production of IL-1β and IL-18 by THP-1 cells and PBMC

Treatment of THP-1 cells with VTX-2337 induced a highly significant (p<0.001, unpaired t test), concentration-dependent increase in IL-1β and IL-18 mRNA, consistent with TLR-mediated activation of NF-ĸB (Fig 2A and 2B). Western blot analysis showed induction of both pro-IL-1β (35 kDa) in THP-1 cell lysate and the release of mature IL-1β (17 kDa) into the culture supernatant (Fig 2C). The active form of caspase-1 (p20, MW of 20kD), which is responsible for the cleavage of pro-IL-1β, was also induced in these cells by VTX-2337. Inhibition of caspase-1 with z-VAD-FMK resulted in a marked reduction of IL-1β release from VTX-2337-treated THP-1 cells (Fig 2D). To demonstrate that activation of the NLRP3 inflammasome is required for the generation of mature IL-1β and IL-18 by THP-1 cells, the VTX-2337 response was evaluated in NLRP3-deficient cells. As shown in Fig 2E and 2F, the concentration-related release of mature IL-1β and IL-18 in response to VTX-2337 (0.03–25 μM), did not occur in NLRP3 deficient THP-1 cells.

To look for comparable activity in primary human cells activated with VTX-2337, the induction of IL-1β and IL-18 was assessed using PBMC from healthy donors. In the absence of known activators of NLRP3, VTX-2337 induced the secretion of both IL-1β and IL-18 in a concentration-dependent manner (Fig 3A and 3B). As additional confirmation of this response, whole blood collected from multiple healthy donors (n = 11) was activated using multiple VTX-2337 concentrations, and production of IL-1β and IL-18 was assessed using the MAP1.6 inflammation panel to complement the ELISA results described above. As shown in Fig 3C and 3D), IL-1β levels increased from a mean of 2.6 ± 0.3 ng/mL in untreated controls to 729 ± 424 ng/mL (p < 0.001, Wilcoxon test) for cells activated with 1 μM VTX-2337. There was also a significant, (p<0.01 Wilcoxon test), concentration-dependent increase in IL-18 release in response to VTX-2337 activation.

Caspase-1 activation and IL-18 production drive NK cell activation by VTX-2337

Caspase-1 activation is an essential step in the processing and secretion of the mature forms of IL-1β and IL-18 [1]. To demonstrate that caspase-1 is activated in response to VTX-2337, PBMC from multiple healthy donors were pretreated with the caspase-1 inhibitor, z-VAD-FMK (10 μg/ml), prior to activation with VTX-2337 (1 μM, 24 h). Pretreatment with z-VAD-FMK reduced IL-1β and IL-18 release from VTX-2337-activated PBMCs by 85–90% (Fig 4A and 4B). Interestingly, z-VAD-FMK also blocked the production of IFNγ, but had little effect on the TNFα response induced by VTX-2337 (Fig 4C and 4D). We have previously reported that NK cells are a major source of IFNγ in VTX-2337-stimulated PBMC [3]. The reduction in IFNγ seen with caspase-1 inhibition is consistent with VTX-2337 driving mDC and monocytes to produce IL-18 that subsequently contributes to NK cell activation. To further evaluate the activity of IL-18 on NK cells, expression of intracellular IFNγ and CD107a, markers of activation and degranulation respectively [17], were monitored by flow cytometry. The number of IFNγ⁺ NK cells increased from 0.30 ± 0.24 in untreated cultures to 1.17 ± 0.91 (p<0.01, Wilcoxon test) with VTX-2337 treatment (Fig 4E). The addition of a neutralizing anti-IL-18 mAb (20 μg/ml) resulted in a partial, but significant reduction in the percentage of NK cells activated by VTX-2337 (p<0.05, Wilcoxon test). Expression of the degranulation marker CD107a by NK cells showed a similar pattern, where VTX-2337 induced a significant increase in expression that was partially blocked by the anti-IL-18 mAb (Fig 4F). These results are consistent with our previous report demonstrating that NK cells respond directly to TLR8 agonists and IL-18 release from activated accessory cells can augment their activation [3].

VTX-2337 enhances NK cell responses to K562 target cells and FcγRIII (CD16) stimulation in vitro

Previous studies have shown that IL-18 provides a stimulatory signal that works cooperatively with the engagement of activating surface receptors to enhance NK cell activation. Based on the release of both IL-12 and IL-18 from TLR8-activated mDC/monocytes, we hypothesized that VTX-2337 should augment NK cell activation in response to K562 target cells that express NKG2D ligands and signaling through FcγRIII cross-linking with anti-CD16 mAbs. To test this hypothesis, PBMC were initially pretreated with VTX-2337 (0.5 μM) for 24 h. NK cells were subsequently stimulated with either K562 tumor cells to activate the NKG2D pathway or plate-bound anti-CD16 mAb to signal through FcγRIII. In the absence of any stimuli, the percentage of IFNγ⁺, CD107a⁺ NK cells was < 0.1% (Fig 5A). As single activation agents, VTX-2337, K562 cells, and anti-CD16 each produced a modest degree of activation, with the percentage of IFNγ⁺, CD107a⁺ NK cells increasing to 7.33%, 4.41% and 0.47%, respectively (Fig 5A). When NK cells were pre-treated with VTX-2337, then exposed to either K562 target cells or immobilized anti-CD16 mAb, there was a much greater effect than with any single stimuli (Fig 5A). Sequential activation with VTX-2337 followed by co-culture with K562 cells increased the population of IFNγ+.CD107a + NK cells to 46.4%, while activation by VTX-2337 followed by FcγRIII stimulation with immobilized anti-CD16 mAb increased this population to 61.4%.

NK cell activation, where VTX-2337 pretreatment was followed by exposure to K562 cells or immobilized anti-CD16 was expanded to assess the response in 11 healthy volunteers. As shown in Fig 5B and 5C, VTX-2337 alone resulted in significant increases in IFNγ+ (p<0.01 Wilcoxon test) and CD107a+ (p<0.001 Wilcoxon test) NK cell populations. The percentage of activated NK cells was markedly augmented when VTX-2337 pretreatment was followed by co-culture with K562 cells or immobilized anti-CD16 treatment. The VTX-2337-mediated activation of NK cell subsets, CD56^bright and CD56^dim cells, was also assessed (S1 Fig). Both NK cell subsets were activated by VTX-2337 as demonstrated by increased IFNγ and CD107a expression, and the response was enhanced by stimulation of NKG2D by K562 target cells or FcγRIII using immobilized anti-CD16 mAb.

In vivo administration of VTX-2337 in cynomolgus monkeys induces circulating IL-1β and IL-18

To demonstrate that VTX-2337 drives the production of both pro-IL1β and pro-IL18, and activates the NLRP3 inflammasome in the absence of any other activating signals, cynomolgus monkeys were treated with the compound and plasma was monitored for these mediators. Cynomolgus monkey were chosen as they are highly responsive to VTX-2337 and predictive of the human TLR8-mediated response. This is in contrast to rodent species, where agonists optimized for activity on human TLR8 have limited activity due to sequence differences in the molecules ecodomain [18].

Monkeys received a subcutaneous injection of VTX-2337 (1 or 10 mg/kg), and plasma was collected predose, 6, 12, 24, and 96 h post-injection. For the 10 mg/kg dose, mean plasma levels of IL-1β increased from baseline levels of 0.5 pg/mL, up to 9.12 ± 2.7 ng/mL (p<0.05, t-test) at 6 h post-administration of VTX-2337 (10 mg/kg, Fig 6A). Circulating levels of IL-18 also increased from a baseline of ~ 1 pg/mL to 68.7 ± 4.4 pg/mL (p < 0.05, t-test) at 6 h in response to the VTX-2337 treatment (10 mg/kg, Fig 6B). Levels of IL-6 were monitored (Fig 6C), as this mediator is induced in response to TLR8 activation, but the release is independent of NLRP3 inflammasome activation. In addition, plasma levels of IFNγ were assessed as a measure of NK cell activation in response to VTX-2337 treatment (Fig 6D). As expected, this biomarker was undetectable in plasma prior to treatment, and increased to 11.1 ± 5.4 pg/mL at 6 h in monkeys administered the 10 mg/kg dose of VTX-2337. Overall, these results demonstrate that the coordinated activation of TLR8 and NLRP3 by VTX-2337 observed in vitro studies also occurs in vivo. Additionally, the relatively low levels of plasma IL-1β and IL-18 relative to IL-6 is consistent with the hypothesis that production of IL-1 family cytokines is more tightly regulated to minimize collateral damage to the host [19].

Treatment of SCCHN patients with VTX-2337 enhances NK cell activation

Thirteen patients with recurrent or metastatic SCCHN were enrolled on Study A103. Cetuximab was administered in combination with 3 escalating dose levels of VTX-2337: 2.5 mg/m² (n = 3), 3.0 mg/m² (n = 6), and 3.5 mg/m² (n = 4). The median age was 62 years (range, 51 to 78) and the majority of patients (10 of 13; 77%) were male. The study population was generally representative of patients with recurrent or metastatic SCCHN. A Consort Flow Diagram for the study is shown in Fig 1.

NK cell activation was assessed in two SCCHN patients treated with 3.0 mg/m² VTX-2337. Blood samples were collected prior to VTX-2337 treatment and again 24 h following VTX-2337 administration. Isolated PBMC from the pre- and post-VTX-2337 treatment blood samples were placed into culture for 4 h without additional stimulation or with ex vivo stimulation using either K562-15mb-41BBL target cells [16], or immobilized anti-CD16 mAb, as previously described. Activation of the NK cell population was monitored by increased expression of CD107a.

In blood samples collected prior to VTX-2337 administration, the prevalence of CD107a-expressing CD56+ NK cells in the control cultures was low (3.8–4.1%) for both patients, as shown in Fig 7. For Patient 1, the ex vivo stimulation with either the K562 cells or immobilized anti-CD16 mAb, produced only a small increase in the percent of CD107a+ CD56+ NK cells. In contrast, ex vivo stimulation of PBMC from Patient 2 with the K562 cells increased the percent of CD107a+ CD56+ NK cells to 18.8%, while the immobilized anti-CD16 mAb increase the percentage to 35.2%.

Following VTX-2337 treatment, the percent of CD107a-expressing CD56+ NK cells in the unstimulated control cultures remained low for both patients. However, the introduction of a second activation signal by ex vivo stimulation with either K562 cells or immobilized anti-CD16 produced a more robust response than seen in pre-VTX-2337 blood samples. NK cells from Patient 1, which were initially refractory prior to VTX-2337 administration, became responsive to both NKG2D and FcγRIII signaling following VTX-2337 treatment. For Patient 2 where the NK cells from the pre-VTX-2337 blood sample were initially responsive to NKG2D and FcγRIII activating signals; the response was further augmented by VTX-2337 treatment.