Target Cell Activation of a Structurally Novel NOD-Like Receptor Pyrin Domain-Containing Protein 3 Inhibitor NT-0796 Enhances Potency S

The NOD-like receptor pyrin domain-containing protein 3 (NLRP3) inflammasome is a central regulator of innate immunity, essential for processing and release of interleukin-1 b and pyroptotic cell death. As endogenous NLRP3 activating triggers are hallmarks of many human chronic inflammatory diseases, inhibition of NLRP3 has emerged as a therapeutic target. Here we identify NDT-19795 as a novel carboxylic acid-containing NLRP3 activation inhibitor in both human and mouse monocytes and macrophages. Remark-ably, conversion of the carboxylate to an isopropyl-ester (NT-0796) greatly enhances NLRP3 inhibitory potency in human monocytes. This increase is attributed to the ester-containing pharmacophore being more cell-penetrant than the acid species and, once internalized, the ester being metabolized to NDT-19795 by carboxylesterase-1 (CES-1). Mouse macrophages do not express CES-1


Introduction
Inhibition of the NOD-like receptor pyrin domain-containing protein 3 (NLRP3) inflammasome represents an attractive therapeutic target for the treatment of multiple human disorders (Guo et al., 2015;Mangan et al., 2018).As a key innate sensor responsive to both endogenous and exogenous danger signals, NLRP3 serves a critical role in regulating monocyte and macrophage function.When a monocyte or macrophage encounters an appropriate activating stimulus, individual NLRP3 This work received no external funding.P.S., C.D., H.W., J.R.D., K.S., M.G.B., D.H., A.P.W., and C.A.G. are employees of NodThera.B.H.K. received a grant from NodThera, Ltd.B.H.K. and M.N. declare they have no financial interest in NodThera.
The authors declare that they have no conflicts of interest.dx.doi.org/10.1124/jpet.123.001941.S This article has supplemental material available at jpet.aspetjournals.org.
subunits are recruited along with the adaptor protein ASC and procaspase-1 to form a supra-molecular structure within the cytosol known as an inflammasome (Martinon et al., 2002;He et al., 2016;Chan and Schroder, 2019;Swanson et al., 2019;Fu and Wu, 2023).This ensemble facilitates caspase-1 activation which in turn catalyzes post-translational processing of inflammatory cytokines pro-interleukin (IL)-1b and pro-interleukin (IL)-18.The primary translation products of these two cytokines lack leader sequences required for entry into the rough endoplasmic reticulum (Auron et al., 1984;March et al., 1985;Ghayur et al., 1997).As a result, the pro-cytokines accumulate intracellularly (Singer et al., 1988;Sitia and Rubartelli, 2020).However, following NLRP3 inflammasome-mediated caspase-1 activation both pro-IL-1b and pro-IL-18 are processed to their mature, biologically active forms (Man and Kanneganti, 2016).Caspase-1 cleaves an additional intracellular protein, gasdermin D, yielding an amino-terminal polypeptide which spontaneously multimerizes and forms pores in membranes resulting in loss of ionic homeostasis (He et al., 2015;Shi et al., 2015).Mature IL-1b and IL-18 are released extracellularly following pyroptosis where they can bind to their cognate receptors on target cells (Fink and Cookson, 2006).Inhibition of NLRP3 activation thus is envisioned as a treatment of disorders where endogenous danger signals (e.g., monosodium urate, amyloid-b aggregates, and cholesterol crystals) promote production of IL-1b and/or IL-18 which contribute to disease pathology.
Prior to discovery of NLRP3 Pfizer scientists identified a compound 773) which inhibited IL-1b output from monocytes activated sequentially with lipopolysaccharide (LPS) to promote pro-IL-1b synthesis and ATP to activate posttranslational processing and release of the mature cytokine (Perregaux et al., 2001;Laliberte et al., 2003).Coll and colleagues subsequently identified CP-456,773 as a selective inhibitor of NLRP3 activation (Coll et al., 2015).Using CP-456,773 as a tool, a large number of animal disease models have provided evidence that NLRP3 inhibition offers therapeutic benefit (see review by Corcoran et al., 2021).In addition, a variety of additional compounds, including natural products and drug substances, have been reported to impair NLRP3 activation in the context of rodent disease models (Abderrazak et al., 2016;Li et al., 2016;Ye et al., 2022).More recent directed chemistry efforts have identified additional NLRP3 activation inhibitors (Li et al., 2023;Povero et al., 2023; and see reviews Baldwin et al., 2016;Mangan et al., 2018;Zahid et al., 2019) and several compounds have completed initial clinical evaluations (Hissaria et al., 2023;Tang et al., 2023).As part of our own medicinal chemistry effort, a series of carboxylic acid analogs were identified possessing pharmacological profiles similar to CP-456,773.Remarkably, when the carboxylate moiety was esterified, potency was dramatically enhanced with respect to inhibition of IL-1b output from LPS/ATP-activated human monocytes (Harrison et al., 2020).Esterification of carboxyl groups is frequently employed during drug optimization to generate prodrugs (Rautio et al., 2008(Rautio et al., , 2018;;Huttunen et al., 2011).Neutralization of negative charge can facilitate passage of small molecules across plasma membranes and endothelial and epithelial barriers as well as reduce transport via various membrane transporters.Following in vivo administration, esterases metabolize the prodrug moiety yielding the active carboxylate species (Laizure et al., 2013;Nishimuta et al., 2014).
In this study, we present cellular pharmacology of a novel ester-containing NLRP3 inhibitor NT-0796 and its carboxylic acid metabolite NDT-19795.Moreover, we describe generation and characterization of a genetically engineered mouse line (hCES-1) which provides a humanized model for studying ester-containing NLRP3 inhibitors both in vitro and in vivo.We show that blood and macrophages isolated from hCES-1 mice are highly sensitive to NT-0796, mimicking potent pharmacological responses seen with human cells.With the aid of the hCES-1 mouse, we demonstrate that orally administered NT-0796 possesses greater pharmacodynamic impact than comparable doses of NDT-19795 and reveal mononuclear phagocyte CES-1 expression is key for the observed in vivo activity.Pharmacological profiling of NT-0796 in hCES-1 mice thus provides translational insight into how this clinical development candidate is expected to behave in human studies.
Generation of a Syntenic Mouse Line in which the Endogenous Ces-1c Locus is Replaced with Human CES-1.Mice dif- fer markedly from humans with respect to CES biology in that they: 1) do not express functional CES in monocytes (Needham et al., 2011) and 2) possess high plasma CES activity which stems from liver expression and subsequent release of Ces1c (Duysen et al., 2011).To mitigate both differences, a displacer vector (Supplemental Methods section) was introduced by electroporation into murine embryonic stem cells (Phnx43); this vector was designed to replace the endogenous Ces1c gene with the human CES-1 gene.To ensure wide-spread expression of human CES-1 in cells of monocyte/macrophage lineage the endogenous human CES-1 promoter was replaced with the mouse colony stimulating factor 1 receptor promoter as this promoter was Characterization of a Novel NLRP3 Inhibitor NT-0796 799 at ASPET Journals on March 12, 2024 jpet.aspetjournals.org

Downloaded from
successfully employed previously to drive transgene expression throughout the mouse mononuclear phagocyte system (Sasmono et al., 2003;Hawley et al., 2018).Cloned ES cells expressing a single copy of the human CES-1 locus were injected into blastocysts and the modified embryos were implanted into pseudo pregnant foster mice.Resulting offspring were crossed for six generations to C57BL/6 mice, after which heterozygous animals were intercrossed to generate cohorts of mice (referred to as hCES-1) homozygous for the humanized locus.Expression of the human CES-1 did not alter fertility, nor could mice homozygous for the locus be distinguished from C57BL/6 mice based on development size or health.
C57BL/6 mice lacking functional Ces1c (B6.Cg-Ces1ctm1.1Loc/J;stock #014096) were obtained from The Jackson Laboratory.Two male and six female Ces1c À/À mice were initially obtained and mated upon arrival at UNC.These mice were used to establish a breeding colony and to produce the animals for the experiments described.This ensured that the Ces1c À/À mice were born and reared in the same vivarium as the C57BL/6 wild-type and the hCES-1 mice.All mice were exposed to a similar environment with identical food, water, noise levels, light cycles, and vibrations, thereby limiting variation between groups due to different rearing environments.
Cytokine Output from Human Peripheral Blood Mononuclear Cells (PBMCs).Heparin-stabilized blood from healthy volun- teers was obtained from Bloodworks NW (Seattle); blood from both male and female donors was employed with no noticeable differences in assay performance.During the course of these studies the donor pool employed consisted of 15 individuals, of which 20% were males and 80% females.PBMCs were prepared using Ficoll-Paque Plus centrifugation.Total cell numbers were determined using a hemacytometer and adjusted to 2.6 x 10 6 cells/ml in RPMI Glutamax medium containing 1% FBS, 1% penicillin/streptomycin, and 20 mM HEPES, pH 7.3.To assess IL-1b output, 0.1 ml (2.6 × 10 5 cells) of this cell suspension was added to each well of 96-well plates.After a 2-hour incubation at 37 C in a 5% CO 2 incubator to allow adherence of monocytes, media and non-adherent cells were aspirated.0.1 ml of RPMI Glutamax medium containing 5% FBS, 1% penicillin/streptomycin, 20 mM HEPES, pH 7.3 (Base Medium) was added to all wells, and the plates were incubated overnight at 37 C in a 5% CO 2 incubator.To wells designated to receive LPS, 0.04 ml of Base Medium containing 350 ng/ml LPS was introduced; wells not receiving LPS received 0.04 ml of Base Medium.Plates were incubated at 37 C for 2 hours to allow transcription/translation of proIL-1b.At this point, media were removed by aspiration and 0.143 ml of fresh RPMI Gluta-MAX medium containing 1% FBS and 1% penicillin/streptomycin were added; the presence of LPS was maintained in wells previously exposed to this stimulus.Test compound at various concentrations (or 0.2% DMSO) was added to designated wells and plates were placed at 37 C in a 5% CO 2 incubator for 60 minutes.To designated wells, 0.0075 ml of a 100 mM ATP solution (final concentration 5 5 mM) was then introduced to promote NLRP3 activation, and the plates returned to a 37 C/5% CO 2 incubator.After a 60-minute stimulation, the plates were subjected to centrifugation after which the media supernatants were harvested for assessment of IL-1b levels by ELISA.
To assess impact on IL-6 and TNFa output, 0.14 ml of a freshly isolated PBMC suspension (2.8 × 10 5 cells) were added to each well of 96-well plates.To wells designated to receive test compound, 0.05 ml of Base Medium containing 4X the desired final test agent concentration were added; control wells received 0.05 ml of Base Medium without compound.Plates were placed at 37 C in a 5% CO 2 incubator for 30 minutes.To wells designated to receive LPS, 0.01 ml of Base Medium containing 2 mg/ml LPS were added to achieve a final LPS concentration of 100 ng/ml and the plates were returned to a 37 C/5% CO 2 incubator.After 4 hours of LPS stimulation the plates were subjected to centrifugation and media supernatants were recovered and assessed for cytokine output using TNFa-and IL-6-specific ELISA kits.
NLRP3-Mediated ASC Fluorescent Speck Formation.A suspension of THP1-ASC-GFP cells (Invivogen) was incubated in culture medium containing 10% FBS with or without 1 mg/ml LPS for 2 hours at 37 C. Cells were collected by centrifugation, resuspended in assay medium containing 1% FBS, and dispensed into a 96-well plate (0.1 ml containing 2 × 10 5 cells/well).Diluted compounds or equivalent dilutions of vehicle (DMSO) were then added to the cells, and the cells were incubated for 30 minutes at 37 C; the final DMSO concentration in all wells was 0.4%.At this point the caspase inhibitor zVAD-FMK was added to all wells to yield a final concentration of 20 mM, and the cells were incubated for 5 minutes at 37 C; z-VAD-FMK was included to prevent pyroptosis and preserve speck detection.Next, 1 ml of a 500 mM nigericin solution (in 10% EtOH) was added to yield a final concentration of 5 mM.Control wells not treated with nigericin received an equivalent dilution of ethanol.After a 1-hour incubation at 37 C, cells were fixed with a final concentration of 1% paraformaldehyde and then incubated with Hoechst 33342 trihydrochloride to fluorescently label DNA.Images of the cells were acquired using an ImageXpress Micro Confocal High Content Imaging System (Molecular Devices) and analyzed using CellProfiler software.
LPS/ATP-Induced IL-1b Output in Blood.Heparin-stabilized blood from healthy volunteers was subjected to a 2-step activation protocol as previously described (Perregaux et al., 2000;Primiano et al., 2016).Blood samples were diluted with RPMI GlutaMax medium containing 20 mM HEPES, pH 7.3 (two parts blood to one part medium) after which 0.14 ml of the diluted samples were placed in individual wells of a 96-well plate designated not to receive LPS.To the remaining diluted blood sample, an appropriate volume of 1 mg/ml LPS was added to achieve a final concentration of 100 ng/ml.0.14 ml of the LPS-containing diluted blood sample were then placed into wells of the 96-well plate designated to receive LPS.The plates were incubated for 3 hours at 37 C in a 5% CO 2 incubator to allow activation of proIL-1b transcription/translation.After the LPS activation phase, 0.05 ml of RPMI GutaMax media containing 1% FBS, 1% penicillin/streptomycin, 20 mM HEPES, pH 7.3, and test agent (at a 4-fold higher concentration than final) were added to designated wells.Wells designated not to receive test agent received 0.05 ml of RPMI GlutaMAX media containing 1% FBS, 1% penicillin/streptomycin, 20 mM HEPES, pH 7.3, and 1.6% DMSO.The plates were returned to the 37 C/5% CO 2 incubator for an additional 30 minutes.At this point, 0.01 ml of 100 mM ATP and 20 mM HEPES, pH 7.3, were added to designated wells to promote NLRP3 activation.The plates were incubated for an additional 60 minutes at 37 C/5% CO 2 after which they were subjected to centrifugation, and plasma supernatants were harvested for assessment of IL-1b levels by ELISA.A similar protocol was employed to assess IL-1b output from heparin-stabilized mouse blood with the following changes: blood samples from multiple mice were pooled, the LPS concentration was increased to 1 mg/ml, and cytokine levels in plasma supernatants were assessed using an ELISA specific for murine IL-1b.
IL-1b Output from Mouse Peritoneal Macrophages.Cells within the peritoneal cavity were harvested by lavage following euthanasia.Cells were collected by centrifugation and resuspended in RPMI-1640 medium containing 5% heat-inactivated FBS, 2 mM Glu-taMAX, 1% penicillin/streptomycin, and 25 mM HEPES, pH 7.3 (Culture Medium). 5 × 10 4 cells (in 0.1 ml of Culture Medium) were seeded per well of 96-well tissue culture plates and the plates were incubated overnight at 37 C in a 5% CO 2 incubator.The following day, 1 mg/ml LPS was introduced to designated wells and the cultures incubated for an additional 2 hours.Culture supernatants then were aspirated and fresh RPMI-1640 medium containing 1% heat-inactivated FBS, 2 mM GlutaMAX, 1% penicillin/streptomycin, and 25 mM HEPES, pH 7.3 (Compound Medium) was added; where indicated, a test agent was present in the Compound Medium.The plates were returned to the 37 C incubator for 60 minutes, after which ATP was introduced to achieve a final concentration of 5 mM.The plates were incubated for an additional 60 minutes at 37 C prior to being subjected to centrifugation (5 minutes at 700 × g).Supernatants were collected and assessed for IL-1b content by ELISA.

Downloaded from
Expression Analysis by qPCR.Total RNA was isolated from indicated cells or tissues using RNA STAT-60 (AMSBIO, Cambridge, MA) according to the instruction manual.Reverse transcription of total RNA to cDNA for quantitative real-time PCR was performed by using a High-Capacity cDNA Reverse Transcription kit (Applied Biosystems, Foster City, CA) according to the manufacturer's instructions.All reactions were performed with TaqMan PCR Universal Master Mix (Applied Biosystems) using the Applied Biosystems QuanStudio 6 Flex Real-Time PCR System.Reactions were carried out in duplicate in a 20 ml final volume reaction.PCR conditions for all reactions were as follows: 2 minutes at 50 C and 10 minutes at 95 C, followed by 40 cycles at 95 C for 15 seconds and 60 C for 1 minute.Expression levels of genes of interest were normalized to 18S.Data were analyzed using the comparative C T method as described by Applied Biosystems.
Plasma carboxylesterase Detection.EDTA-stabilized plasma was prepared from wild type (WT) and hCES-1 mice.Total plasma-associated esterase activity was assessed using a non-selective substrate 4-NPB (Williams et al., 2011).Reaction mixtures contained 200 mM 4-NPB in 50 mM HEPES, pH 7. Plasma samples were added (resulting in a 600-fold dilution) and the mixtures were incubated at 37 C. Production of 4-NP was monitored spectrophotometrically over time by assessing OD 405 and converted to nmoles by comparison with a 4-NP standard curve.
WT and hCES-1 plasma samples were also treated with a serine hydrolase activity-based probe (TAMRA-FP) to visualize targeted polypeptides (Simon and Cravatt, 2010).Prior to labeling, plasma samples were depleted of serum albumin and IgG by incubation with resinbound antibodies following the supplier's protocol.The resulting depleted fractions were concentrated using 3 kD mol.wt.cut-off centrifugation filtration units.Protein concentrations were determined using the Pierce BCA total protein reagent.The depleted plasma samples were adjusted to 1 mg/ml of total protein and treated with 4 mM TAMRA-FP for 45 minutes at room temperature.The mixtures then were quenched by addition of an equal volume of 2X SDS sample buffer containing 50 mM dithiothreitol and heated at 70 C for 10 minutes.Forty microliters of each sample was subjected to SDS gel electrophoresis and resulting gels were washed and imaged using GE Typhoon Trio instrumentation with 532 nm and 720 nm lasers.
Western Blotting for hCES-1.Peritoneal macrophages were isolated from WT B6 female and homozygous hCES-1 female mice.An equal number of cells (27.5 × 10 6 ) were pelleted and suspended into 0.5 ml of an extraction buffer containing 1% Triton X-100 and protease inhibitors.The cellular extracts were clarified by centrifugation and resulting supernatants were recovered and disaggregated with SDS sample buffer and subjected to SDS gel electrophoresis.Separated polypeptides were blotted onto nitrocellulose membranes and probed for human CES-1 and the housekeeper protein GAPDH.
Stability of NT-0796 in Plasma as Assessed by Liquid Chromatography and Tandem Mass Spectrometry (LC-MS/ MS) Analysis.NT-0796 was added to EDTA-stabilized plasma sam- ples from WT and hCES-1 mice which were diluted 2-fold with 50 mM HEPES, pH 7.4, 1% FBS.As a positive control, NT-0796 also was incubated with Supersomes expressing CES-1b (the major isoform of CES-1 expressed by human liver).Reaction mixtures were incubated at 37 C for 60 minutes, after which 1 mM bis-p-nitrophenyl-phosphate was introduced to stop further hydrolysis.Standard curves were generated for both NT-0796 and NDT-19795 by preparing a common serial dilution series titration from 10 to 0.1 mM in 50 mM HEPES, pH 7.4, 1% FBS.Samples were stored at À80 C until transferred to the University of Washington School of Pharmacy for LC-MS/MS analysis.Briefly, 10 ml of each sample was combined with 20 ml of an internal standard, glyburide (50 ng/ml in water), and vortexed.0.1 ml of acetonitrile was then added to each sample, after which the samples were vortexed and clarified by centrifugation.Fifty microliters of each supernatant was collected and combined with 50 ml of water in a limited volume HPLC vial.Five microliters of these samples were injected into a Waters Xevo-TQ-XS.Recoveries of NT-0796 and NDT-19795 in the reaction mixtures were calculated based on comparison of peak areas of each analyte to standard curves generated with each compound.
CNS Expression of Human CES-1.Right hemispheres of brains isolated from WT and hCES-1 mice were homogenized in a lysis buffer composed of 10 mM Tris, pH 8, 150 mM NaCl, 1 mM EDTA, and 1% NP-40 and the lysates clarified by centrifugation.Recovered supernatant fractions (1 mg of total protein) were incubated with 0.06 ml of streptavidin-agarose, after which the beads were removed by centrifugation.To the precleared supernatants, FP-biotin was added (2 mM) and the mixtures were incubated at 25 C for 45 minutes.Reaction mixtures were quenched by addition of an equal volume of 10 mM urea in lysis buffer.The mixtures were then re-treated with streptavidin-agarose to capture biotinylated polypeptides.Beads were collected by centrifugation and washed 3X by repeated centrifugation with 5 M urea in lysis buffer.Final bead pellets were resuspended in 0.04 ml of SDS sample buffer containing 50 mM dithiothreitol and boiled for 5 minutes.Aliquots of the soluble extracts were subjected to SDS gel electrophoresis, and the resulting gels blotted onto nitrocellulose membranes.These blots were probed with Alexa Fluor-680 streptavidin (diluted 1 to 4000) and re-probed with rabbit anti-human CES-1 (diluted 1 to 5000) followed by a DyLight 880-labeled goat anti-rabbit secondary antibody (diluted 1 to 10,000).
Enriched preparations of microglial cells from WT and hCES-1 mice were isolated as described previously (Lee and Tansey, 2013).Briefly, each perfused brain was diced into small pieces and transferred to a 15-ml tube containing 3 ml of dissociation medium (Dulbecco's Modified Eagles Medium/F12, 1 mg/mL papain, 1.2 U/ml dispase II and 20 U/mL DNase1) for 20 minutes at 37 C with gentle inversion every 5 minutes.At the end of the incubation period, 5 ml of neutralization media (Dulbecco's Modified Eagles Medium/F12, 10% FBS and 4.5 mg/ml glucose) was added to the tube to neutralize the enzymes, followed by 5 minutes of centrifugation at 250 × g.Pelleted tissues were resuspended in 6 ml of Dulbecco's Modified Eagles Medium/F12 media and dissociated by pipetting up and down using a Pasteur pipette.The cell suspension then was passed through a 40 mm cell strainer and the filtrate was centrifuged at 250 × g for 4 minutes.The cell pellet was resuspended in 4 ml of 37% isotonic Percoll and 4 ml of 70% isotonic Percoll was gently underlaid.Then 4 ml of 30% isotonic Percoll was overlaid on top of the 37% layer, followed by 2 ml of phosphate-buffered saline.The gradient was centrifuged at 300 × g for 40 minutes at 18 C with no brake.Microglia were collected at the 37%-70% interface; based on enrichment of Tmem119 expression, the isolated cells were estimated to be 30% of microglial origin.
Immunohistochemical Assessments.Liver, lungs, and brains from WT and hCES-1 mice were harvested and subjected to formalin fixation and paraffin embedding.All staining procedures were conducted using a Leica Bond Automated Immunostainer.Briefly, sections were deparaffinized then processed for immunolocalization of human CES-1 using a primary rabbit polyclonal antibody against CES-1.Liver sections also were probed with a polyclonal antibody against endogenous mouse Ces-1.After staining with an appropriate goat anti-rabbit secondary antibody, slides were treated with an antigoat antibody conjugated to horseradish peroxidase and antibody complexes then were visualized using 3,3 0 -diaminobenzidine as substrate.Tissues were counterstained with hematoxylin.Slides were scanned in brightfield with a 20X objective using a NoanZoomer Digital Pathology System (Hamamatsu City, Japan).The digital images were then imported into Visiopharm software (version 2023.01.1.13563;Hoersholm, Denmark) for analysis.Images were quantified as a percentage of positive stain to tissue area by a blinded analyst using the same Visiopharm software.
In Vivo Studies.To compare pharmacokinetics of NT-0796 in WT (C57BL/6J) and hCES-1 mice, animals were intravenously administered 3 mg/kg NT-0796; the dosing formulation was composed of 1 mg/ ml NT-0796 in vehicle consisting of 40% polyethylene glycol 400/10% Cremophor RH-40/50% water.Three male mice of each genotype were dosed with this formulation after which blood samples were harvested at various times by saphenous vein collection.Twenty-microliter Characterization of a Novel NLRP3 Inhibitor NT-0796 801 at ASPET Journals on March 12, 2024 jpet.aspetjournals.orgaliquots of the harvested blood samples were collected into prechilled polypropylene tubes containing 2 ml of 0.5 M potassium EDTA (anticoagulant) and 3 ml of a quench solution containing 950 mM sodium fluoride, 722 mM potassium oxalate, and 5 mM phenylmethylsulfonyl fluoride (to neutralize esterase activity).Two hundred microliters of 75% methanol/25% acetonitrile (spiked with internal calibration standards) were added to each stabilized blood sample, the samples vortexed, and subsequently subjected to a 10-minute centrifugation at 12,000 × g at 4 C. Resulting supernatants were analyzed by LC-MS/ MS analysis.Plasma T 1/2 estimates were generated using Phoenix Winnonlin 6.3 software.PK studies were conducted under contract by WuXi AppTech in Shanghai, China.
Pharmacodynamic impact of NDT-19795 and NT-0796 was evaluated in an acute LPS/ATP peritonitis model (Primiano et al., 2016).Mice for these studies were bred and maintained in specific pathogenfree housing in ventilated cages.Studies were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals; these studies were also approved by the Institutional Animal Care and Use Committee (IACUC) at the University of North Carolina, Chapel Hill.Additional details relating to animal care are provided in the Supplemental Methods section.Cohorts of male hCES-1 mice were dosed by oral gavage with NDT-19795 or NT-0796 suspended in 0.5% methylcellulose, 2% Tween 80; doses of 10, 30, and 100 mg/kg were employed.NT-0796 doses were selected based on IC 50 values observed in the context of the hCES-1 mouse blood assay and knowledge of compound bioavailability.One-hour post-dosing, the animals received an intraperitoneal injection of 1 mg of LPS followed 2 hours later by intraperitoneal injection of 0.5 ml of 30 mM ATP. Thirty minutes later, the animals were euthanized (inhaled isoflurane anesthesia followed by physical euthanasia) and 3 ml of phosphatebuffered saline containing 10% FBS, 25 units/ml heparin and protease inhibitors was injected to facilitate peritoneal lavage.Levels of IL-1b and IL-6 recovered in the isolated lavage fluids subsequently were determined by ELISA.Similarly, age-and sex-matched (female) cohorts of hCES-1 (backcrossed to a C57BL/6 background), Ces1c À/À , and WT C57BL/6 mice were dosed orally with NT-0796 and then subjected to the LPS/ATP challenge model.
Statistical Analysis.Data presented were graphed and analyzed using GraphPad Prism software (version 10 sourced from Graphpad Software, Boston, MA).Individual points corresponding to compound titration curves represent the mean and standard deviation of triplicate determinations.Concentration titration curves were fit and IC 50 values determined using a four-parameter logistic, variable slope model.Statistical analyses were conducted using ordinary one-way ANOVA with Dunnett's multiple comparisons test or Welch's one-way ANOVA and Dunnett's T3 multiple comparisons test.

Results
Pharmacological Profile of NT-0796 and NDT-19795.The prototype NLRP3 activation inhibitor CP-456,773 contains a sulfonylurea core scaffold; at neutral pH CP-456,773 carries a negative charge resulting from loss of an acidic H 1 (Laliberte et al., 2003).In seeking an alternative to the sulfonylurea scaffold, a carboxylate series was discovered which inhibited NLRP3 activation in human monocytes as exemplified by NDT-19795 (Fig. 1A).Like CP-456,773, NDT-19795 is predominantly an anionic species at neutral pH.As shown in Fig. 1B (inset), a high level of IL-1b output was achieved when human PBMCs were stimulated sequentially with LPS and ATP; little IL-1b output was observed with either stimulus alone.When NDT-19795 was added to LPS-activated human PBMCs prior to introduction of ATP, IL-1b output was inhibited in a concentration-dependent manner (Fig. 1B).Like CP-456,773, NDT-19795 completely suppressed release of IL-1b to the medium; IC 50 values for CP-456,773 and NDT-19795 were 0.027 and 0.047 mM, respectively.A selective NLRP3 activation inhibitor is not expected to inhibit LPS-induced output of IL-6 and TNFa as these cytokines are exported from cells via the traditional secretory apparatus.Indeed, robust production of IL-6 was observed from LPS-stimulated human monocytes, and CP-456,773 did not affect its output (Fig. 1C, inset).In contrast, a p38 kinase inhibitor, BIRB796, effectively inhibited IL-6 production, as would be expected based on this kinase being involved in mediating signaling via the TLR4 receptor (Regan et al., 2003).NDT-19795 likewise did not inhibit IL-6 output when tested at concentrations #10 mM (Fig. 1C).Similar findings were observed with respect to LPS-induced TNFa output (not shown).When assessed by fluorescence microscopy, NLRP3-mediated inflammasome assembly results in formation of a single "speck" structure within the cell; formation of these fluorescent specks is inhibited by CP-456,773 (Coll et al., 2015).THP-1 cells engineered to express an LPS-inducible GFP-tagged ASC construct produced speck-like structures following activation of NLRP3 using the potassium ionophore nigericin.NDT-19795 and CP-456,773 inhibited speck formation concentration-dependently yielding IC 50 values of 0.46 and 0.069 mM, respectively (Fig. 1D).Thus, NDT-19795 possesses a pharmacological profile expected of a selective NLRP3 activation inhibitor.
NT-0796 is an isopropyl-ester derivative of NDT-19795 (Fig. 1A).Like NDT-19795, NT-0796 inhibits IL-1b output from human PBMCs stimulated with LPS and ATP.However, the potency of NT-0796 is 200-fold greater than observed with NDT-19795 (Fig. 1B); IC 50 values for NDT-19795 and NT-0796 were 0.047 and 0.0002 mM, respectively.Despite the enhanced potency against IL-1b output, NT-0796 remained a selective cytokine inhibitor; the highest concentration tested (2 mM) did not affect output of IL-6 from LPS-activated human PBMCs (Fig. 1C).In the THP-1 cell inflammasome assembly assay NT-0796 inhibited formation of the GFP-ASC speck with an IC 50 value of 0.002 mM, which is 230-fold more potent than NDT-19795 (Fig. 1D).Thus, esterification enhanced the potency of NDT-19795 as an inhibitor of NLRP3 activation.Further evidence of the beneficial impact of esterification is seen in the context of a human blood assay.Freshly isolated heparin-stabilized blood sequentially treated with LPS and ATP resulted in release of IL-1b (Fig. 1E, inset).As was observed with PBMC cultures, optimal IL-1b output from blood was dependent on addition of both LPS and ATP (Fig. 1E; Perregaux et al., 2000).The prototype NLRP3 inhibitor CP-456,773 blocked IL-1b output in the blood matrix with an IC 50 of 2 mM.The shift in the potency of CP-456,773 in moving from a low serum environment employed in the PBMC assay to the blood assay is consistent with this agent binding extensively to serum proteins, as expected for a lipophilic anion (Primiano et al., 2016).NDT-19795 and NT-0796 also inhibited IL-1b output in the context of human blood yielding IC 50 values of 3.4 mM and 0.009 mM, respectively (Fig. 1E).Although not shown, NDT-19795 and NT-0796 also inhibited IL-18 output from LPS/ATP-treated PBMCs with average (n 5 2) IC 50 values of 0.057 mM and 0.00018 mM, respectively.
Like human monocytes, mouse macrophages produce extracellular mature IL-1b in response to sequential LPS and ATP activation; stimulation with LPS or ATP individually is insufficient to produce extracellular IL-1b (Fig. 2A, inset).Both CP-456,773 and NDT-19795 inhibited IL-1b output from LPS/ ATP-treated mouse macrophages (Fig. 2A); observed IC 50 values versus WT mouse macrophage IL-1b output were 0.11 mM and 0.43 mM, respectively, for CP-456,773 and NDT-19795 (low serum environment).In sharp contrast, ester-containing NT-0796 did not inhibit IL-1b output from mouse macrophages stimulated sequentially with LPS and ATP (Fig. 2A).
Lack of pharmacological impact of NT-0796 in the context of isolated WT mouse macrophages is consistent with NDT-19795 being the active species and, due to lack of intracellular CES activity, conversion of NT-0796 to NDT-19795 not occurring following cellular uptake.

Downloaded from
In the context of a WT mouse blood assay format, stimulation with both LPS and ATP again was required to promote externalization of IL-1b (Fig. 2B, inset).Output of IL-1b from LPS/ATP-treated WT mouse blood was inhibited by CP-456,773 and NDT-19795 with IC 50 values of 0.7 and 4.5 mM, respectively.In contrast to its ineffectiveness as an inhibitor of IL-1b output from LPS/ATP-treated WT mouse macrophages, however, NT-0796 concentration-dependently inhibited output in the context of the mouse blood assay format (Fig. 2B).The IC 50 value observed (3 mM), however, is comparable to that for NDT-19795.This correspondence is explained by the presence of Ces1c in WT mouse plasma which is expected to metabolize NT-0796 to NDT-19795 following addition to the blood assay matrix (see below).
Generation of a Syntenic Mouse Line in which the Endogenous Ces-1c Locus is Replaced with Human CES-1.Lack of sensitivity of WT mouse macrophages to ester-based NLRP3 inhibitors prompted development of a mouse line that would enable a translationally-relevant assessment of vivo pharmacological properties of these compounds.As noted earlier, mice differ from humans not only in the lack of CES expression in monocytes and macrophages but also in expression of an additional CES-1 ortholog (Ces1c), which is released by the liver to the plasma compartment.To mitigate both differences, a genetic engineering strategy was undertaken in which the endogenous Ces1c locus was replaced by a displacer vector (Supplemental Figs. 1 and 2) encoding the human CES-1 gene.
hCES-1 Mice Lack Ces1c.To establish that the expected recombination event designed to remove the entire Ces1c locus from the mouse genome had taken place, liver RNA from heterozygous and homozygous animals was isolated and analyzed by qPCR using primers specific for Ces1c.The level of Ces1c expression was reduced in animals containing a single copy of the human gene relative to WT levels and undetectable in mice possessing two copies of the human gene (Fig. 3A).
To demonstrate that Ces1c levels were reduced in a functional context, plasma fractions initially were assessed for esterase activity using 4-NPB as substrate; while not specific for Ces1c, 4-NPB offers a facile analytical method for assessing esterase catalytic activity.WT plasma promoted time-dependent formation of the product 4-NP (Supplemental Fig. 3).Plasma from hCES-1 mice, however, demonstrated a much-reduced rate of substrate hydrolysis (Supplemental Fig. 3), as would be expected by loss of Ces1c.
Further demonstration of Ces1c depletion from hCES-1 mouse plasma was accomplished using a serine hydrolase activity-based TAMRA-labeled FP probe (Patricelli et al., 2001).Following treatment with TAMRA-FP and SDS gel electrophoresis, WT plasma contained a prominent fluorescently-labeled band which migrated with an apparent mobility of 64 kDa, the expected molecular weight of Ces1c (Fig. 3B).This species migrated slightly slower on the gel than the fluorescently labeled polypeptide species detected in CES-1 Supersomes (Fig. 3B); this difference in mobility may stem from a difference in Nlinked glycosylation as the amino acid sequence of Ces1c encodes 5 potential N-linked glycosylation sites whereas that of human CES-1 contains only a single potential site.The TAMRA-FP-labeled polypeptide was not detected in heat-inactivated CES-1 Supersomes, indicating labeling is dependent on an active conformation.Importantly, plasma derived from hCES-1 mice was devoid of the 64 kDa fluorescent species found in WT plasma; at both a 2.5 and 10 mg total protein load, no 64 kDa species was detected (Fig. 3B).Densitometric analysis of the gel confirmed the TAMRA-labeled species was greatly reduced in hCES-1 plasma (Fig. 3B).Following staining of the same gel with Coomassie blue, the banding patterns and loads were comparable between the WT and hCES-1 albumindepleted plasma samples (Fig. 3B, lower image).Thus, based on TAMRA-FP labeling, Ces1c represents an abundant serine hydrolase present in WT mouse plasma but absent in plasma isolated from hCES-1 mice.
As expected, when NT-0796 was incubated with EDTAstabilized WT mouse plasma it was efficiently metabolized to NDT-19795 as detected by LC-MS/MS analysis (Table 1).Conversion of NT-0796 to NDT-19795 did not occur in buffer alone incubated on ice or at 37 C. CES-1 Supersomes also effectively catalyzed conversion of NT-0796 to NDT-19795 (Table 1).
In stark contrast, NT-0796 remained intact when incubated with EDTA-stabilized plasma isolated from hCES-1 mice (Table 1).This stability suggests Ces1c is the dominant activity in WT mouse plasma responsible for conversion of NT-0796 to NDT-19795, and absence of Ces1c in hCES-1 mouse plasma results in markedly enhanced stability of NT-0796.hCES-1 mice Express Human CES-1 in Monocytes and Macrophages.Resident peritoneal macrophages were isolated from WT and hCES-1 mice and analyzed by Western blot analysis for the presence of human CES-1.As shown in Fig. 4A, WT macrophages possessed a single antigenic species migrating between the 49 and 62 kDa molecular weight standards.This species migrated faster than recombinant standards of human CES-1 (rhCES1) and murine CES-1 (rmCES1) and is assumed to represent an endogenous mouse polypeptide which cross reacts with the anti-CES-1 antibody.In contrast, hCES-1 peritoneal macrophages possessed the same faster migrating species as WT macrophages but also a doublet migrating with a mobility comparable to the recombinant CES-1 standards.This doublet is likely caused by differential glycosylation of human CES-1.Similarly, human THP-1 cells included as a positive control yielded a cross-reactive species whose mobility matched that of recombinant CES-1; these cells lacked the faster migrating species seen in mouse macrophages.Densitometric analysis of the CES-1 staining region of the blot indicated that hCES-1 macrophages and THP-1 cells possessed comparable levels while levels of the antigenic species in WT macrophages were minimal (Fig. 4A).The control marker, GAPDH, indicated similar total cellular protein loads were applied to relevant lanes of the gel (Fig. 4A).Thus, peritoneal macrophages from hCES-1 mice and THP-1 cells, but not macrophages from WT mice, express CES-1 as assessed by Western blotting.
Isolated hCES-1 peritoneal macrophages were subjected to sequential LPS/ATP activation to promote IL-1b output.In terms of requirements for release of IL-1b to the medium, hCES-1 macrophages retained the need for a two-step activation.No extracellular cytokine was observed in the absence of both LPS and ATP (-LPS/-ATP) or after stimulation with ATP-only (-LPS/1ATP).Likewise, cells stimulated with LPSonly (1LPS/-ATP) released minimal cytokine to the medium (Fig. 4B, inset).However, following sequential treatment with  LPS and ATP (1LPS/1ATP), large quantities of IL-1b were released to the medium (Fig. 4B, inset).Output of IL-1b from hCES-1 macrophages in response to LPS/ATP activation was inhibited by CP-456,773 and NDT-19795 concentrationdependently (Fig. 4B); IC 50 values of 0.09 and 0.23 mM were observed for CP-456,773 and NDT-19795, respectively, and are comparable to IC 50 values observed using WT mouse macrophages (Fig. 2).Notably, ester-containing NT-0796 effectively inhibited LPS/ATP-induced IL-1b output from hCES-1-derived macrophages, yielding an IC 50 value of 0.002 mM (Fig. 4B).Thus, in contrast to WT mouse macrophages, NLRP3-dependent IL-1b output from peritoneal macrophages isolated from hCES-1 mice is highly sensitive to the ester-containing inhibitor.
To probe whether the genetic engineering also resulted in monocyte human CES-1 expression, blood from hCES-1 mice was subjected to sequential LPS/ATP activation in the presence of NLRP3 activation inhibitors and IL-1b output assessed by ELISA.As with WT mouse blood, output of IL-1b was dependent on sequential stimulation with LPS and ATP (Fig. 4C, inset).Both CP-456,773 and NDT-19795 inhibited LPS/ATP-induced IL-1b output yielding IC 50 values of 0.5 and 2.1 mM, respectively (Fig. 4C).Importantly, NT-0796 inhibited IL-1b output from hCES-1 mouse blood with an IC 50 value of 0.035 mM (Fig. 4C).In contrast to WT blood where NT-0796 yielded a potency equivalent to NDT-19795, in the hCES-1 blood matrix the IC 50 for NT-0796 was reduced 90-fold relative to NDT-19795.This enhanced potency of NT-0796 in the context of blood isolated from hCES-1 mice is consistent with absence of plasma Ces1c-mediated metabolism and presence of monocyte-expressed human CES-1.
Immunohistochemical staining using a monoclonal anti-CES-1 antibody specific for human CES-1 was employed to further probe for expression.As shown in Fig. 5A, the anti-human CES-1 antibody did not detect any antigen in liver sections prepared from WT mice.On the other hand, liver sections from hCES-1 mice contained a subset of cells which stained positive for human CES-1 as detected by the presence of the brown diaminobenzidine reaction product (Fig. 5B).Quantification of the images indicated that 2.0% of the hCES-1 liver section stained positive relative to the total tissue area, versus only 0.12% of the WT liver section (Fig. 5D).The positivestaining cells may represent Kupffer cells and/or other  4. Characterization of NT-0796-mediated NLRP3 inhibition in hCES-1-derived macrophages and blood monocytes.(A) Western blot of detergent extracts of peritoneal macrophages isolated from WT and hCES-1 mice probed with anti-CES-1 antibody.THP-1 cells were included as a control, as were recombinant standards of both mouse (rmCES1) and human (rhCES1) CES-1.The blot was counterstained with anti-GAPDH as a loading control.Both WT and hCES-1 macrophages possessed a polypeptide which migrated faster than CES-1 and cross-reacted with the antiserum; the identity of this species is not known.Densitometry analysis of anti-CES-1 positive staining bands (co-migrating with the recombinant CES-1 standards) in the hCES-1 macrophage and THP-1 cell lysates as well as the corresponding area of the WT macrophage lane was performed with ImageJ software.Levels of fluorescence are indicated in the graph on the right.A larger image of this blot is presented in Supplemental Fig. 7.This western blot comparison was conducted once.(B) Peritoneal macrophages isolated from hCES-1 mice were stimulated with LPS and/or ATP and the amount of IL-1b released extracellularly was determined by ELISA.The inset shows levels of IL-1b output as a function of the indicated stimulus while the plot shows levels output of IL-1b from LPS/ATP-treated peritoneal macrophages as a function of the concentration of NDT-19795, NT-0796, or CP-456,773.Each data point of a titration curve is the mean and S.D. of three technical replicate wells.In this cell system, NDT-19795 yielded an average IC 50 value of 0.16 mM (n 5 2 biologic replicates), while NT-0796 yielded mean and S.E.M. IC 50 values of 0.007 ± 0.0055 mM (n 5 3 biologic replicates).(C) Pooled heparin-stabilized blood from hCES-1 mice (all female, 4 months of age) was stimulated with LPS and/or ATP and the amount of IL-1b released extracellularly was determined by ELISA.The inset shows levels of IL-1b output as a function of the indicated stimulus, while the plot shows levels of IL-1b output from LPS/ATP-treated samples as a function of the concentration of NDT-19795, NT-0796, or CP-456,773.Each data point of a titration curve is the mean and S.D. of three technical replicate wells.In this assay context, NT-0796 yielded an average IC 50 value of 0.065 mM (n 5 2 biologic replicates).Where denoted, IC 50 values (mM) were calculated using Graphpad Prism software and are shown adjacent to curve fits.resident macrophages found in liver.This assignment is further supported by the human CES-1 positive-staining liver cells co-staining with anti-F4/80, a macrophage marker (Supplemental Fig. 4).Staining of lung sections from hCES-1 mice with the anti-human CES-1 antibody also revealed the presence of a positive-staining subpopulation of cells (Supplemental Fig. 4).
The engineering strategy employed to generate the hCES-1 mouse line was expected to remove Ces1c while maintaining expression of other endogenous Ces-1 isoforms.Mouse hepatocytes are considered the primary source of secreted Ces1c, but these cells also express forms of Ces-1 that remain cell-associated, including Ces1d (Jones et al., 2013;Lian, et al., 2019).Sections of hCES-1 mouse liver show abundant positive staining when probed with a primary antibody recognizing mouse Ces-1 (Fig. 5C), confirming that hCES-1 mice retain endogenous liver Ces-1 expression.Mouse antigen staining accounted for 86% of the total tissue area (Fig. 5D), as expected based on endogenous Ces-1 being expressed by hepatocytes.
Brains of hCES-1 mice also were interrogated for human CES-1 expression.By immunohistochemical analysis, hCES-1 mouse brain sections contained a subset of CES-1-positive cells which were not present in WT brains (Supplemental Fig. 4).Brain/microglial cell human CES-1 expression was further established biochemically.First, using a biotinylated-FP probe enrichment strategy, human CES-1 was detected in biotinylated protein isolates of hCES-1 but not WT brain homogenates by Western blotting (Supplemental Fig. 5A).Second, qPCR analysis of mRNA from a microglial cell population isolated from hCES-1 mice showed high enrichment of both a microglial cell marker (Tmem119) as well as human CES-1 relative to whole brain and WT mouse peritoneal macrophages (Supplemental Fig. 5B).Finally, enriched microglial cell populations isolated from hCES-1 but not WT mice contained an immunoreactive protein species corresponding to human CES-1 when probed by Western blotting (Supplemental Fig. 5C).
Demonstration of the Utility of hCES-1 Mice for Studying In Vivo Pharmacology of NT-0796.WT and hCES-1 mice were dosed with NT-0796 (3 mg/kg; i.v.), after which plasma samples were collected and analyzed for NT-0796 and NDT-19795.In WT mice, levels of NT-0796 rapidly declined such that no NT-0796 was detected beyond 1 hour post-dosing (Fig. 6).Loss of NT-0796 was accompanied by the appearance of NDT-19795 which remained detectable for up to 8 hours.In hCES-1 mice, on the other hand, NT-0796 disappeared more slowly, being detectable out to 4 hours.Loss of NT-0796 from hCES-1 mice was accompanied by appearance of NDT-19795, but at lower levels than in WT mice (Fig. 6).T 1/2 estimates for NT-0796 in WT and hCES-1 mice were 0.18 Fig. 6.Comparison of NT-0796 in vivo pharmacokinetics in WT and hCES-1 mice.WT and hCES-1 mice received NT-0796 (3 mg/kg) by intravenous administration.At the indicated times, blood samples were collected into tubes containing a cocktail of esterase inhibitors to stabilize NT-0796 and plasma fractions isolated.These subsequently were analyzed for the presence of NT-0796 and NDT-19795 by LC-MS/MS.Each value is the average concentration seen in three separate mice.This study was conducted once.
Characterization of a Novel NLRP3 Inhibitor NT-0796 and 0.87 hours, respectively.Robust metabolism of NT-0796 in hCES-1 mice despite absence of Ces1c is expected based on the abundant expression of endogenous Ces in hCES-1 livers (Fig. 5C).
To demonstrate NT-0796 possesses greater pharmacodynamic potential than its carboxylic acid metabolite NDT-19795, an acute in vivo model was conducted in which mice were subjected to sequential intraperitoneal challenge with LPS and ATP (Primiano et al., 2016).As indicated in the schematic shown in Fig. 7A, cohorts of hCES-1 mice were dosed orally with NT-0796 or NDT-19795 1 hour prior to receiving an intraperitoneal injection of LPS and subsequent injection of ATP to promote NLRP3 activation.Relative to the level of IL-1b in peritoneal lavage fluid recovered from hCES-1 mice dosed with vehicle, NDT-19795-treated mice required a dose of 100 mg/kg to achieve a 50% reduction in cytokine output (Fig. 7D).The reduction in IL-1b output at 100 mg/kg was statistically significant, but lower doses of 10 and 30 mg/kg NDT-19795 did not result in significant inhibition.NT-0796, on the other hand, was a much more effective inhibitor following oral dosing yielding significant inhibition at all three doses (Fig. 7B).At a dose of 100 mg/kg NT-0796, IL-1b output was near baseline while 10 and 30 mg/kg doses of NT-0796 provided > 75% inhibition of IL-1b output.Importantly, IL-6 output into the peritoneal lavage fluid was not affected by either NDT-19795 or NT-0796 confirming their pharmacological impact is limited to cytokines dependent on NLRP3 activation (Fig. 7C and 7E).
Both absence of plasma Ces1c and expression of human CES-1 by monocytes/macrophages are expected to enhance NT-0796's therapeutic impact.To provide more clarity as to the relative importance of these two traits, a comparison of NT-0796 as an inhibitor of IL-1b output in response to the LPS/ATP peritoneal challenge was conducted in mice with three different genotypes.C57BL/6 WT and Ces1c-deficient (Ces1c À/À ; Duysen et al., 2011) mice were profiled alongside hCES-1 mice in the peritoneal challenge model post-dosing with NT-0796.Cohorts of female age-matched mice were dosed with NT-0796 or vehicle 1-hour prior to intraperitoneal injection of LPS followed 2-hour later by administration of ATP to promote NLRP3 activation.All three genotypes showed robust output of IL-1b when dosed with vehicle (Fig. 8A).When dosed at 3 mg/kg, NT-0796 did not reduce IL-1b output.A dose of 30 mg/kg NT-0796 modestly reduced levels of IL-1b produced by WT but not Ces1c À/À mice relative to vehicle-treated counterparts.Likewise, a dose of 100 mg/kg resulted in a modest reduction in IL-1b levels generated by WT and Ces1c À/À mice.In contrast, levels of IL-1b recovered from hCES-1 mice treated with 30 and 100 mg/kg NT-0796 were reduced to near baseline relative to levels recovered from vehicle-treated counterparts (Fig. 8A).Plasma samples collected at termination and analyzed for drug levels yielded expected outcomes.Both the Ces1c À/À and hCES-1 plasma samples contained similar levels of NT-0796 and NDT-19795, while plasma samples from WT mice lacked NT-0796 and possessed higher NDT-19795 levels than the other genotypes (Fig. 8B).Plasma samples from mice of each of the three genotypes were independently assessed for NT-0796 metabolism.As expected, NT-0796 was rapidly metabolized by WT plasma but stable in plasma from both Ces1c À/À and hCES-1 mice (Supplemental Fig. 6).

Discussion
Esterification of carboxylic acid moieties is a well-described methodology for improving bioavailability and/or distribution of small molecule therapeutics (Rautio et al., 2008(Rautio et al., , 2018;;Huttunen et al., 2011).In the vast majority of these cases, hydrolysis of the resulting prodrug following administration to humans is thought to occur as a result of metabolism by CES-1 or CES-2 which are highly expressed in the liver and gastrointestinal tract (Wang et al., 2018).In these cases, the bioactive acid drug species is then released to the systemic circulation, where it can exert its intended pharmacology.Although monocytes are not typically thought of as a site of drug metabolism, the presence of CES-1 in these cells (Uphoff and Drexler, 2000;Satoh et al., 1999), as well as in alveolar macrophages (Munger et al., 1991) and tissue resident macrophages (Elfiky et al., 2022), provides an opportunity for ester-based prodrugs to be employed as a means to selectively deliver an active carboxylate drug species to these cells.Appreciation of this attribute was previously employed to selectively target histone deacetylase inhibitors to cells of myeloid linage (Needham et al., 2011;Zabkiewicz et al., 2016).In view of mononuclear phagocyte lineage cells being prime expressors of NLRP3 and producers of IL-1b, an ester-based prodrug strategy affords an opportunity to selectively deliver active NLRP3 inhibitors precisely to the chief target cells of interest.
The power of this approach is apparent when comparing the potency of NDT-19795 and NT-0796 as inhibitors of NLRP3 activation in human monocytes and THP-1 cells; these two cell types both express CES-1.The free acid species NDT-19795 is an effective and selective inhibitor of LPS/ATPinduced IL-1b output from human PBMCs and of ASC-speck formation in LPS/nigericin-treated THP-1 cells.However, the corresponding ester-containing NT-0796 is >100-fold more potent than NDT-19795 at inhibiting these same cellular responses.This difference in potency is attributed to neutral NT-0796 being taken up by the cells more readily than negatively-charged NDT-19795.Once internalized, NT-0796 then gains access to CES-1 allowing formation of the active pharmacophore NDT-19795.Evidence to support this mechanistic understanding is based on comparison of the effectiveness of NDT-19795 and NT-0796 as inhibitors of IL-1b output from WT mouse peritoneal macrophages.As with human PBMCs, LPS/ATP treatment of peritoneal macrophages elicits IL-1b production and release to the medium.Output of IL-1b is inhibited by NDT-19795 and the prototype NLRP3 inhibitor CP-456,773, and the potency of each is comparable to that seen using human PBMCs.In sharp contrast, NT-0796 is an ineffective inhibitor of LPS/ATP-induced IL-1b output from WT peritoneal macrophages.However, LPS/ATP-induced IL-1b output from peritoneal macrophages isolated from hCES-1 mice, which express human CES-1, is highly sensitive to NT-0796 inhibition.Thus, while both WT and hCES-1 peritoneal macrophages are sensitive to NDT-19795, only the human CES-1-expressing macrophages are sensitive to the ester species NT-0796.
Another key difference in CES biology between mice and humans is the abundance of Ces1c in the plasma compartment (Rudakova et al., 2011;Nishimuta et al., 2014); CES levels in mouse and human plasma are estimated to be 80 and 0 mg/ml, respectively (Li et al., 2005).Ces1c lacks a C-terminal amino acid motif (KDEL) which is responsible for retention of other CES isoforms within the endoplasmic reticulum (Medda and Proia, 1992).As a result, Ces1c is released into the circulation where it can contribute to metabolism of ester-bearing drug substances.Recognition that high levels of plasma Ces activity in mice complicate toxicology studies of organophosphorus compounds led to the engineering of a mouse line lacking serum esterase activity (Duysen et al., 2011).However, while this mouse line is depleted of plasma Ces1c, it does not overcome the lack of CES expression by mouse monocytes/macrophages.A separate transgenic mouse line was engineered in which macrophages were programmed to express human CES-1 via use of a CD68 promoter.These mice subsequently were crossbred with a naturally plasma esterase-low Es1elo mouse to generate a line with CES-1 expression in macrophages and lower serum Ces1c levels (Luque-Martin et al., 2019).The advantage of the hCES-1 mouse line developed in our study is that endogenous Ces1c is completely eliminated via the targeting strategy employed.Absence of Ces1c was demonstrated through the use of an activity-dependent fluorescent tag; whereas plasma from WT mice contained a TAMRA-FP tagged polypeptide possessing the expected mol.wt. of Ces1c, this species was absent in hCES-1 plasma.Moreover, hCES-1 plasma demonstrated markedly reduced 4-NPB hydrolase activity.Further confirmation of Ces1c absence was achieved by comparing metabolism of NT-0796 in WT and hCES-1 EDTA-stabilized plasma.Whereas NT-0796 was readily metabolized to NDT-19795 by WT plasma, no significant conversion took place with hCES-1 mouse plasma.
The colony stimulating factor 1 receptor promoter used to drive expression of human CES-1 in the current study has previously been employed to drive transgene expression in a wide array of mononuclear phagocytes (Sasmono et al., 2003;Hawley et al., 2018).Assuming similar expression control was achieved in this study, human CES-1 would be expected to reside in a wide array of mononuclear phagocytes.By comparison of the sensitivity to NT-0796 inhibition of LPS/ATP-induced IL-1b output from blood of WT and hCES-1 mice, we can infer that human CES-1 is expressed by blood borne-monocytes.By immunohistochemistry, evidence of human CES-1 presence is observed in a subset of cells residing in lung and liver tissues of hCES-1 mice.These positive staining cells may represent resident tissue macrophage populations.Finally, human CES-1 is detected in brains of hCES-1 mice by immunohistochemistry and several independent biochemical assessments.Thus, the colony stimulating factor 1 receptor promoter effectively led to human CES-1 expression in monocytes and macrophages within the periphery as well as microglial cells within the CNS.
The genetic engineering resulted in a mouse line more representative of human CES-1 biology than WT mice.To demonstrate that hCES-1 mice can facilitate modeling of an estercontaining NLRP3 inhibitor in humans, hCES-1 mice were subjected to an acute LPS/ATP peritoneal challenge following administration of NDT-19795 or NT-0796.In this model, resident peritoneal macrophages are assumed to be the primary producers of IL-1b in response to sequential challenge with LPS and ATP.NDT-19795 produced a pharmacodynamic response in the model when administered orally at 100 mg/kg; this is expected as the carboxylate species is considered the active pharmacophore.However, doses of NT-0796 #100 mg/kg administered to hCES-1 mice yielded greater inhibition of IL-1b output than achieved with NDT-19795.This enhanced response may result from better tissue distribution of the ester species and from greater sensitivity of peritoneal macrophages to NT-0796 (versus NDT-19795) resulting from their expression of human CES-1.Evidence that the latter trait is important was achieved by comparing the impact of NT-0796 in hCES-1 and Ces1c À/À mice following oral administration of NT-0796.Both lines lack Ces1c and thus are expected to possess similarly improved NT-0796 pharmacokinetics.Based on plasma drug levels at termination, NT-0796 pharmacokinetics were enhanced in both hCES-1 and Ces1c À/À compared with WT mice.However, the pharmacodynamic response was significantly greater in hCES-1 mice, supporting the belief that CES-1 presence within NLRP3-expressing cells enhances therapeutic utility of the ester species.In view of the apparent widespread expression of CES-1 in monocytes and macrophages, hCES-1 mice should afford a useful host for studying impact of ester-containing NLRP3 inhibitors, such as NT-0796 in the context of various disease models.The current findings demonstrate NT-0796 is a very potent inhibitor of NLRP3 activation both in vitro and in vivo and highlight the potential of this novel agent as a human therapeutic.

Supplemental Methods
Design and Construction of hCES-1 Displacer Vector The vector, herein referred to as the hCES-1 Displacer Vector, was designed such that homologous recombination of the hCES-1 Displacer vector with the mouse genome results in replacement of a segment of endogenous genomic DNA encoding ces1c with a segment of DNA carried by the vector.However, instead of simply disrupting an endogenous gene by replacing one or several exons with a selectable marker cassette, the hCES-1 Displacer Vector replaces the entire mouse gene with a segment of DNA carrying an orthologous human gene along with a selectable marker cassette flanked by mutant loxP sites.The selectable marker gene can be removed by transient expression of Cre recombinase, leaving only a nonfunctional loxP site in its place.The result is a locus in which a chosen mouse gene has been replaced with its human ortholog, with the mouse-human and human-mouse junctions defined to basepair resolution.
As shown in Supplemental Figure 1, the hCES-1 Displacer vector comprises two homology arms (one long and one short), a portion of the hCES-1 gene, a portion of the Csf1r promoter and a selectable marker cassette flanked by mutant loxP sites.
The hCES-1 Displacer Vector was constructed by recombineering using segments of four separate bacterial artificial chromosomes (BACs) and a floxed neomycin resistance cassette.The homology arms of the construct were derived from two 129 mouse BACs.The shorter arm corresponded to chr8: 93,141,145,371 (coordinates based on Mus musculus (house mouse) genome assembly GRCm38/mm10) and was derived from the bMQ248j13 BAC.The longer arm corresponded to chr8: 93,027,227 -93, 096,304 and was derived from the bMQ445g02 BAC.
The segment of human genomic DNA in the Displacer construct was derived from the RP11-449J19 BAC.This segment of DNA corresponds to chr16: 55,798,833,124 Verified, individual clones of the modified ES cells were injected into mouse blastocyts and the modified embryos were subsequently implanted into pseudo pregnant foster mice.The resulting chimeric mice were bred to achieve germline transmission and the resulting pups were genotyped.Heterozygous hCES-1 mice were crossbred to yield homozygous hCES-1 mice.

Animal environmental enrichment
Water: Prepared by Reverse Osmosis (RO), via a filtration process that uses a semipermeable membrane to remove ions, molecules, and larger particles from drinking water.It is delivered to cages by an auto-water system.
Cages: Employed cages were made by Tecniplast and are part of the Green Line GM500 series, manufactured in Italy.
Enrichment: Each cage was equipped with a "Mouse House" from Tecniplast, designed to enhance the animals' positive behavior.It provides a sense of protection to its occupants, as it is crafted from a special plastic that only allows red wavelength light to pass through.Additionally, Nestlets™ from Ancare are added to each cage.While originally used exclusively in cages with litters, it was discovered that nulliparous adult female and male mice also showed interest in this enrichment.The activity of shredding the Nestlets™ provides the mice with natural activity and exercise.
In vivo study considerations Both male and female mice were employed in studies detailed in this report.We have not identified sex differences in terms of response to the acute LPS/ATP peritonitis model or in terms of macrophage responsiveness.All mice were maintained throughout, in ventilated cages within an SPF barrier vivarium.Females were housed up to 6 mice per cage.Males were housed as littermates, with no more than 5 mice per cage.Male hCES-1 mice (a total of 59) were employed for the study shown in Figure 7.All mice in this experiment were of an identical genotype.The only variable in this experiment was their exposure to NT-0796 or NDT-19795.The use of male mice minimizes variation potentially caused by cage-to-cage differences in the estrus cycle.The objective of the experiment shown in Figure 8 was distinct.Here, impact of genetic differences in CES (Ces1c and human CES-1) on the effectiveness of NT-0796 was compared.The primary variable of interest, therefore, is dependent on the genetics of the mice.In this study female mice were employed as this allowed co-housing of mice from different litters (litters from wild type, Ces1c -/-and hCES-1 breeding cages).Males from different litters, and thus different genotypes, are challenging to co-house due to the development of hierarchies and displays of aggressive behavior which isn't observed with female mice.The use of female mice permitted co-housing of Ces1c -/-, hCES-1, and wild type mice in the same cage, which limits the differential response attributed to the "cage effect."This effect is especially significant when mice are housed on ventilated racks.Co-housing ensures that the microbiome is shared among all mice, regardless of their genotype.A total of 24 mice of each of the 3 genotypes was employed in the study shown in Figure 8.
Power calculations were used to estimate groups size based on our experience with the in vivo LPS/ATP assay, including expected standard deviation and desire to observe differences of ~25% between treatment groups or various mouse lines.The calculator employed is accessed at: https://clincalc.com/stats/samplesize.aspxtreatment with TAMRA-FP.Migration of molecular weight standards (kDa) are shown on the left and right of the images.Top image shows fluorescence and the bottom image Coomassie staining of the same gel.B) Larger image of the Western blot shown in Figure 4A.Detergent extracts of peritoneal macrophages isolated from WT and hCES-1 mice were separated by SDS gel electrophoresis, blotted onto nitrocellulose, and the blot probed with anti-CES-1 antibody (green).THP-1 cells were included as a control, as were recombinant standards of both mouse (rmCES1) and human (rhCES1) CES-1.The blot was counterstained with anti-GAPDH (red) as a loading control.Migration of molecular weight standards (kDa) are shown on the left of the image.

Fig. 1 .
Fig. 1.Pharmacological impact assessed in human cell systems.(A) Structures of the prototype NLRP3 activation inhibitor CP-456,773 and two novel inhibitorsNDT-19795, a carboxylic acid, and the corresponding isopropyl-ester NT-0796.(B) Human PBMC IL-1b output assessment.The inset shows levels of IL-1b output as a function of treatment, while the plot shows levels of IL-1b output from LPS/ATP-treated cells as a function of test agent concentration.Each data point of a titration curve is the mean and S.D. of three technical replicate wells.Independent determinations with different PBMC preparations yielded mean and S.E.M. IC 50 values of 0.066 ± 0.011 mM for NDT-19795 (n 5 6 biologic replicates) and of 0.00022 ± 0.00006 mM for NT-0796 (n 5 14 biologic replicates).(C) Output of IL-6 from LPS-treated PBMCs is not affected by NDT-19795 or NT-0796.Each data point of a titration curve is the mean and SD of three technical replicate wells, and this example is representative of two biologic replicates.Inclusion of 1 mM CP-456,773 throughout the 3-hour LPS stimulation did not inhibit IL-6 output, whereas 0.2 mM BIRB796, a p38 kinase inhibitor, did (inset).(D) Inhibition in formation of fluorescent "specks" as a function of test agent concentration following sequential LPS and nigericin stimulation of ASC-GFP-expressing THP-1 cells.Each data point is the sum derived from imaging four sites within a single well.Data are representative of three independent biologic replicates in which NDT-19795 and NT-0796 were profiled in a side-by-side comparison.(E) Human blood IL-1b output assessment.The inset shows levels of IL-1b output as a function of treatment while the plot shows levels of IL-1b produced following sequential stimulation with LPS and ATP in the presence of the indicated test agent concentration.Each data point of a titration curve is the mean and S.D. of three technical replicate wells.Independent determinations with different blood preparations yielded mean and S.E.M. IC 50 values of 4.7 ± 0.41 mM for NDT-19795 (n 5 5 biologic replicates) and of 0.0084 ± 0.00084 mM for NT-0796 (n 5 9 biologic replicates).Where denoted, IC 50 values (mM) were calculated using Graphpad Prism software and are shown adjacent to curve fits.Control panel insets shown in Figs.1, 2, and 4 depict IL-1b output from cells subjected to no stimulation (-LPS/-ATP), ATP alone (-LPS/1ATP), LPS alone (1LPS/-ATP), and sequential LPS and ATP (1LPS/1ATP).Each of the four conditions comprised 8 wells of technical replicates (with the exception of only four wells being employed for 1LPS/-ATP and 1LPS/1ATP conditions in Fig.2B).In the case of the control panel inset shown in Fig.1C, the number of technical replicate wells were: no stimulation (n 5 3), LPS (n 5 6), LPS1CP-456,773 (n 5 4), and LPS1BIRB796 (n 5 4).Horizontal lines shown in the inset dot plots mark the median value.Samples which readout below the lower limit of quantitation of the ELISA were plotted as 0.

Fig. 2 .
Fig.2.Sensitivity of IL-1b output from WT mouse peritoneal macrophages and blood.(A) Peritoneal macrophages were stimulated with LPS and/or ATP and the amount of IL-1b released extracellularly was determined by ELISA.The inset shows levels of IL-1b output as a function of the indicated stimulus, while the plot shows levels of output of IL-1b from LPS/ATP-treated peritoneal macrophages as a function of the concentration ofNDT-19795, NT-0796, or CP-456,773.Each data point of a titration curve is the mean and S.D. of three technical replicate wells.In independent determinations, mean and SEM IC 50 values were 0.29 ± 0.05 mM for NDT-19795 (n 5 4 biologic replicates) while the IC 50 for NT-0796 was consistently > 2 mM (n 5 3 biologic replicates).(B) Pooled heparin-stabilized blood from a cohort of WT C57BL/6 mice (all male, 4 month of age) was subjected to LPS/ATP and IL-1b output was assessed by ELISA.The inset shows levels of IL-1b output as a function of stimulus while the plot shows levels of output of IL-1b from LPS/ATP-treated blood samples as a function of the concentration ofNDT-19795, NT-0796, or CP-456,773.Each data point of a titration curve is the mean and S.D. of three technical replicate wells.Where denoted, IC 50 values (mM) were calculated using Graphpad Prism software and are shown adjacent to curve fits.

Fig. 3 .
Fig.3.Assessing Ces1c expression.(A) Total mRNA was isolated from liver homogenates of WT (Ces1c/Ces1c), hCES-1 heterozygous (Ces1c// hCES-1) and homozygous (hCES-1/hCES-1) mice and assessed by reverse transcription qPCR for Ces1c expression.The level of expression observed in WT mice was set to 1; ND 5 not detected.Four livers from each genotype were analyzed; this assessment was conducted once.(B) Albumin/IgG-depleted plasma samples from WT and hCES-1 mice were treated with TAMRA-FP to covalently label serine hydrolases after which the reaction mixtures were separated by SDS gel electrophoresis.CES-1 Supersomes were treated with TAMRA-FP to provide an internal control.The top panel is the fluorescent image and the bottom panel is an image of the same gel after Coomassie staining.Lane assignments: 1 & 2, 10 and 2.5 mg, respectively, of WT plasma; 3 & 4, 10 and 2.5 mg of hCES-1 plasma; 5 & 8, blank; 6, 3 mg of CES-1 Supersomes incubated with TAMRA-FP; 7, 3 mg of CES-1 Supersomes which were heat-inactivated prior to treatment with TAMRA-FP.Migration of molecular weight standards (kDa) are shown on the left of the images.Densitometry analysis of the bands corresponding to TAMRA-labeled Ces-1c in lanes 1 and 2 and the corresponding areas in lanes 3 and 4 was performed using ImageJ software (version 1.54f; NIH, Bethesda, MD).Levels of TAMRA fluorescence as a function of plasma protein loading are indicated in the graph on the right.This experiment is representative of two conducted.Larger images of these blots are presented in Supplemental Fig.7.
Fig. 4. Characterization of NT-0796-mediated NLRP3 inhibition in hCES-1-derived macrophages and blood monocytes.(A) Western blot of detergent extracts of peritoneal macrophages isolated from WT and hCES-1 mice probed with anti-CES-1 antibody.THP-1 cells were included as a control, as were recombinant standards of both mouse (rmCES1) and human (rhCES1) CES-1.The blot was counterstained with anti-GAPDH as a loading control.Both WT and hCES-1 macrophages possessed a polypeptide which migrated faster than CES-1 and cross-reacted with the antiserum; the identity of this species is not known.Densitometry analysis of anti-CES-1 positive staining bands (co-migrating with the recombinant CES-1 standards) in the hCES-1 macrophage and THP-1 cell lysates as well as the corresponding area of the WT macrophage lane was performed with ImageJ software.Levels of fluorescence are indicated in the graph on the right.A larger image of this blot is presented in Supplemental Fig.7.This western blot comparison was conducted once.(B) Peritoneal macrophages isolated

Fig. 5 .
Fig. 5. Immunohistochemical detection of CES in hCES-1 livers.Livers from an individual WT and hCES-1 mouse were fixed and processed and multiple sections were prepared.Sections from WT (A) and hCES-1 (B, C) mice were probed with an antibody specific for human CES-1 (A, B) or an antibody recognizing endogenous mouse Ces-1 (C).Human CES-1 is readily detected in liver sections of hCES-1 mice (B) but absent from WT livers (A).Endogenous mouse isoforms of Ces-1 remain abundant in the livers of hCES-1 mice (C).Quantification of the area represented by diaminobenzidine-positive staining as a percentage of total tissue area for the three images is indicated (D).Scale bars 5 100 mm.0 2 4 6 8 1 0 1

Fig. 7 .
Fig. 7. Comparison of NT-0796 and NDT-19795 pharmacodynamic impact when administered orally to hCES-1 mice.As indicated in the schematic (A), NT-0796 and NDT-19795 were administered to cohorts of hCES-1 mice (male, 8 weeks of age) at 10, 30, or 100 mg/kg (vehicle 0.5% methylcellulose, 2% Tween 80) prior to being subjected to an acute LPS/ ATP challenge.Recovered peritoneal lavage fluids were analyzed for IL-1b and IL-6 by ELISA.Individual mouse cytokine levels (pg/ml) are shown as a function of treatment with NT-0796 [IL-1b (B) and IL-6 (C)] orNDT-19795 [IL-1b (D)  and IL-6 (E)].Group sizes in the NT-0796 arm of the study were n 5 8 in the vehicle and n 5 7 in the 10, 30, and 100 mg/kg cohorts.Group sizes in the NDT-19795 arm of the study were n 5 9 in the vehicle and n 5 7 in the 10, 30, and 100 mg/kg cohorts.Statistical analysis of mean cytokine (both IL-1b and IL-6) levels recovered in peritoneal lavage fluids was conducted using ordinary one-way ANOVA with Dunnett's multiple comparisons test.A similar study was conducted in which hCES-1 mice were dosed with 1, 3, or 10 mg/kg NT-0796; inhibition of IL-1b output was seen at 10 mg/kg, but not at the lower doses.

Fig. 8 .
Fig.8.Comparison of NT-0796's pharmacodynamic impact in hCES-1 and Ces1c À/À mice.Cohorts of female mice (12-22 weeks of age) corresponding to each of the indicated genotypes (WT, Ces1c À/À , and hCES-1) were dosed orally with NT-0796 (3, 30, or 100 mg/kg) or vehicle 1 hour prior to being subjected to the acute peritoneal LPS/ATP challenge.(A) Levels of IL-1b recovered in peritoneal lavage fluids as a function of dose and genotype.Values for each individual mouse are presented with the cohort mean indicated by a horizontal bar.(B) Terminal plasma levels of NDT-19795 and NT-0796 as determined by LC-MS/MS are indicated as a function of genotype and dose.Levels of NT-0796 in WT mouse plasma samples all were below the lower level of quantification (3 nM) except for one mouse in the 100 mg/kg cohort.Group sizes were n 5 6 for all cohorts.Statistical analysis of mean IL-1b levels between vehicle and NT-0796 treated means for each genotype were conducted using Welch's one-way ANOVA and Dunnett's T3 multiple comparisons test.A separate pilot study was conducted in which cohorts of the three genotypes were dosed with a single 30 mg/kg dose of NT-0796; IL-1b output and metabolism to NDT-19795 showed the same genotype dependent differences as depicted.

TABLE 1
fold with buffer (50 mM HEPES, pH 7, 150 mM NaCl, 1% FBS).CES-1 Supersomes were employed as a positive control to facilitate metabolic conversion.NT-0796 (input concentration of 3 mM and 0.4% DMSO) was added to each indicated matrix and the mixtures were incubated at 37 C for 60 minutes (except for a buffer control which remained on ice) after which they were quenched with an equal volume of 0.3 M phosphoric acid and frozen at À80 C pending LC-MS/MS analysis.Recoveries of NT-0796 and NDT-19795 are indicated; values are the average of duplicate determinations.Standard curves generated with each compound established a lower limit of quantification of 0.04 and 0.014 mM, respectively, for NT-0796 andNDT-19795.