A Novel Small-Molecule Enantiomeric Analogue of Traditional (−)-Morphinans Has Specific TLR9 Antagonist Properties and Reduces Sterile Inflammation-Induced Organ Damage

Rafaz Hoque, Ahmad Farooq, Ahsan Malik, Bobby N. Trawick, David W. Berberich, Joseph P. McClurg, Karen P. Galen and Wajahat Mehal
J Immunol April 15, 2013, 190 (8) 4297-4304; DOI: https://doi.org/10.4049/jimmunol.1202184

Abstract

TLR9 is a key determinant of the innate immune responses in both infectious and sterile injury. Specific antagonism of TLR9 is of great clinical interest to reduce tissue damage in a wide range of pathologies, and has been approached by modification of nucleic acids, the recognized ligand for TLR9. Such oligonucleotide-derived pharmacotherapeutics have limitations in specificity for nucleic acid receptors, significant potential for immunologic recognition with generation of innate and adaptive immune responses, and limited bioavailability. We have identified enantiomeric analogues of traditional (−)-morphinans as having TLR9 antagonist properties on reporter cell lines. One of these analogues (COV08-0064) is demonstrated to be a novel small-molecule antagonist of TLR9 with greater specificity for TLR9 than oligo-based antagonists. COV08-0064 has wide bioavailability, including the s.c. and oral routes. It specifically inhibits the action of TLR9 antagonists on reporter cells lines and the production of cytokines by TLR9 agonists from primary cells. It also has efficacy in limiting TLR9-mediated sterile inflammation in in vivo models of acute liver injury and acute pancreatitis. The identification of a morphinan-based novel small-molecule structure with TLR9 antagonism is a significant step in expanding therapeutic strategies in the field of sterile inflammatory injury.

Introduction

Toll-like receptor 9 has been identified as a key determinant of innate immune responses to both infectious and sterile inflammatory injury (1–3). TLR9 is an endosomal receptor for dsDNA in the extracellular compartment and has greatest affinity for unmethylated and CpG-rich DNA (4, 5). Unmethylated and CpG DNA is less prevalent in mammalian genomic DNA, yet the ability of mammalian DNA to induce TLR9-mediated inflammation has been demonstrated in multiple pathologies (5).

Receptor–ligand interaction has been investigated through homology modeling, mutational analysis, and use of chemically modified ligands (6, 7). These investigations have led to antagonism of TLR9 function through competitive inhibition at ligand binding sites using nucleic acids known as inhibitory oligonucleotides (ODNs) or immunoregulatory DNA sequences with chemical modification of the oligophosphate backbone to prevent degradation by nucleases (8–11). These nucleic acid–based antagonists of TLR9 also function as antagonists of TLR7, an endosomal receptor for single-stranded RNA of mammalian and prokaryotic origin (10, 12). Although TLR7 activation parallels TLR9 immune response in many experimental models, the interaction of TLR7 and TLR9 activation and suppression is considerably more complex (13, 14). Inhibition of TLR9 signaling has also been achieved through use of nucleic acid binding polymers and other nucleic acid binding agents, including antimalarials such as chloroquine and imidazoquinolines (15). These agents do not have TLR9-specific effects, either binding and masking other nucleic acid TLR ligands or, in the case of imidazoquinolines, stimulating TLR7 and TLR8 directly. Recently, an oral inhibitor of nucleic acid–sensing endosomal TLRs, E6446, has shown efficacy in mitigating inflammatory injury in a murine model of cerebral malaria (16).

Beyond lack of specificity for TLR9, nucleic acid–based antagonists have immunoregulatory functionality independent of TLR9 (17). Similarly, DNA vaccines do not require the presence of host TLR9 for generation of potent adaptive immune responses (18). In addition, TLR9-independent sensors of cytosolic DNA have been identified, specifically absent in melanoma 2 and IFI16 (19–21). The bioavailability of nucleic acid–derived TLR9 antagonists is well characterized in vivo in parenteral administration. However, significant bioavailability after oral administration is less clear and appears unlikely to date.

We have identified a new small-molecule TLR9 antagonist COV08-0064 with a novel structure based on the (+)-morphinan scaffold. The (+)-morphinan class of compounds refers to the enantiomeric analogues of traditional (−)-morphinans such as (−)-naltrexone and (-)-naloxone. Compound COV08-0064 emerged as a potential lead candidate from a broad TLR9 antagonist screen of a library of such functionalized (+)-morphinan derivatives. The pharmacokinetics and TLR specificity of COV08-0064 for a range of TLRs is determined. In addition, in vivo efficacy for TLR9 antagonist function is established in two models of TLR9-dependent sterile inflammation, acetaminophen (APAP) hepatotoxicity and cerulein hyperstimulation acute pancreatitis.

Materials and Methods

TLR9 antagonist screen and other TLR reporter screens

TLR9 antagonist screen was performed in a commercially available HEK-Blue TLR9 cell line expressing a TLR9-responsive, NF-κB–inducible, secreted placental alkaline phosphatase (SEAP). Specifically, small-molecule candidates were screened for inhibition of SEAP activity induced by the TLR9 ligand ODN2006 as determined by QUANTI-Blue colorimetric assay (Invivogen, San Diego, CA). Candidates with strong TLR9 antagonism in this screen were subsequently screened for TLR agonist and antagonist activity using HEK-Blue TLR reporter cell lines for TLR2, TLR3, TLR4, TLR5, TLR7, TLR8, TLR9, and the NF-κB transgene alone treated with, respectively, heat-killed Listeria monocytogenes, polyinosinic:polycytidylic acid, Escherichia coli K12 LPS, flagellin, CL097, CL075, ODN2006, and TNF-α (Invivogen).

Pharmacokinetics of COV08-0064 in rats and mice

Conscious Sprague–Dawley rats were administered COV08-0064 by i.v. tail vein injection, s.c. injection in the dorsum of the neck, i.p. injection, or oral gavage in nonfasted animals as single doses. i.v., s.c., and i.p. injections were at 10 mg/kg body weight and per oral administration was at 100 mg/kg body weight. Serum concentrations of COV08-0064 were assessed by serial sampling from the jugular vein, tail vein, and at time of euthanasia by cardiac puncture at pretreatment and at 5, 15, 30, 60, 90, 120, 180, 240, and 360 min after administration.

Isolation and treatment of peritoneal elicited cells

Thioglycollate elicited peritoneal macrophages were elicited by i.p. injection of C57BL/6 male mice with sterile 4% thioglycollate broth (Sigma, St. Louis, MO) and harvest of peritoneal elicited cells (PECs) after 72 h. PECs were plated on 24-well plates at 106 cells/well in 24-well dishes, nonadherent cells removed at 1 h, and the cells recovered for 2 additional hours before treatment with ODN1826, ODN1585, and Imiquimod (R837; Invivogen) with or without COV08-0064 for 3 h. All compounds were reconstituted in sterile PBS.

Quantitative PCR assessment of target genes

RNA was extracted from PECs using an RNeasy Plus Mini Kit from Qiagen (Valencia, CA). Reverse transcription was performed with Transcriptor reverse transcriptase from Roche Applied Science (Indianapolis, IN). Quantitative real-time PCR was performed for mRNA expression for pro–IL-1β, Nlrp3, and Gapdh using commercial primer probe sets from Applied Biosystems (Foster City, CA) and the Applied Biosystems 7500 real-time PCR System. Expression of Gapdh was used to standardize the samples. Results were expressed as a ratio to Gapdh expression for pro–IL-1β and Nlrp3 expression, respectively, and normalized to the ratio in untreated PECs.

Caspase-1 activity assays

Pancreatic and liver tissues were flash frozen in liquid nitrogen and then homogenized with a rotor stator homogenizer in cell lysis buffer (Biovision, Milpitas, CA). To 50 μl of sample, 50 μl of 2× reaction buffer (Biovision) was then added. Caspase-1 substrate Ac-YVAD-MCA (Peptides International, Osaka, Japan) was resuspended in DMSO at 20 mM and then added to a final concentration of 100 μM to each sample in opaque 96-well dishes preheated to 37°C. The samples were then incubated for 5 min in a Synergy II fluorescence plate reader with excitation and emission wavelengths of 380 and 460 nm (Biotek, Winooski, VT). Mean velocity of emission fluorescence was calculated from 10 readings over 5 min per sample. For liver tissue sample, this was normalized against total protein content in the sample as assessed using a Coomassie Protein Assay Kit (Pierce Thermo Scientific, Rockland, IL). For pancreatic tissue samples, this was normalized to total amylase content as determined using a Phadebas Kit (Amersham Pharmacia, Rochester, NY).

IL-6 and IL-12p40 ELISA assays

Supernatant was collected from PECs cotreated with ODN1826 and COV08-0064 for 8 h. Pancreatic and liver tissues were flash frozen in liquid nitrogen and then homogenized with a rotor stator homogenizer in cell lysis buffer (Biovision). IL-6 and IL-12p40 was performed using commercially available assays (eBioscience, San Diego, CA). For pancreatic tissue, values were normalized to total protein content. For liver tissue, values were normalized to total protein content.

Animals

C57BL/6 male mice 5–8 wk of age were purchased from the National Cancer Institute. All experiments and animal handling were performed under approved protocols at the Yale University Institutional Animal Care and Use Committees. Sprague–Dawley rat experiments were performed on-site by Covidien (Mansfield, MA).

APAP-induced hepatotoxicity, COV08-0064 treatment, and COV08-0055 treatment

APAP (Sigma-Aldrich) solution was prepared as 20 mg/ml in sterile Dulbecco’s PBS (DPBS). COV08-0055 and COV08-0064 were prepared as 10 mg/ml in DPBS and filter sterilized. Mice were fasted for 15 h and then treated with COV08-0055 or COV08-0064 at 60 mg/kg body weight or DPBS vehicle by i.p. injection. One hour later, mice were administered APAP at 500 mg/kg body weight by i.p. injection. Animals were euthanized by isoflurane at 12 h for collection of serum and liver tissue.

Cerulein hyperstimulation pancreatitis and COV08-0064 treatment

C57BL/6 male mice were administered cerulein sulfate (Sigma-Aldrich) at 50 μg/kg body weight by i.p. injection in sterile normal saline for 10 sequential hourly injections. COV08-0064 was prepared as 10 mg/ml in DPBS and filter sterilized. Animals were administered COV08-0064 at 30 mg/kg body weight or DPBS vehicle by s.c. injection concurrent with the first, fourth, and seventh doses of cerulein. Animals were euthanized by isoflurane at 1 h after the last cerulein injection for collection of serum and pancreatic tissue.

Pancreatic trypsin activity and serum amylase and alanine aminotransferase assays

Pancreatic tissue was flash frozen in liquid nitrogen and then homogenized with a rotor stator homogenizer in 5 mM MOPS, 250 mM sucrose, 1 mM MgSO4, pH 7.00. To 50 μl of pancreatic homogenate was added 175 μl of 50 mM Tris HCl, 150 mM NaCl, 1.3 mM CaCl2, pH 8.10 and 25 μl of 10 mM fluorescent trypsin substrate Boc-QAR-MCA (Peptides International). The samples were then incubated for 5 min in a Synergy II fluorescence plate reader with excitation and emission wavelengths of 380 and 460 nm (Biotek). Mean velocity of emission fluorescence was calculated from 10 readings over 5 min per sample. This was normalized to total amylase content as determined using a Phadebas Kit (Amersham Pharmacia). Heparinized mouse blood was centrifuged at 10,000 × g at 4°C and assessed for serum amylase activity using this assay kit as well. Serum alanine aminotransferase (ALT) levels were determined in the Yale Clinical Chemistry Laboratory (New Haven, CT).

Quantitation of liver- and pancreas-infiltrating neutrophils

Neutrophil quantitation was performed in paraffin-embedded liver sections after immunolabeling with Ly-6B.2 mAb (AbD Serotec, Raleigh, NC) by scoring for positive cells in five high-power fields (×40).

Statistics

Pharmacokinetic data were analyzed by noncompartmental analysis using Pharsight WinNonLin software. Statistical analyses were performed using Microsoft Excel 2007. Unpaired two-tailed Student t test was used to compare groups. A p value <0.05 was considered significant. Percent bioavailability for s.c. and i.p. doses for COV08-0064 were calculated as the ratio relative to IV dosing of area under the curve for the respective route divided by dose administered.

Results

COV08-0064 is a specific small-molecule antagonist of TLR9 signaling

A library of small molecules structurally related to traditional opioids, but having opposite stereochemistry and novel functional group substitutions, were screened for antagonism of TLR9 using the HEK-Blue TLR9 cell line. Several compounds were thereby identified as TLR9 antagonists by antagonism to dose-dependent inhibition of ODN2006 TLR9 ligand stimulation of an NF-κB–inducible transgene for SEAP (Fig. 1A). In particular, the two most efficacious compounds, COV08-0064 and COV08-0093, were noted to have similar five-membered saturated heterocyclic rings containing oxygen (Fig. 1B). COV08-0064 was identified as the more efficacious antagonist of TLR9 for further study with an IC50 of ∼100 nM.

FIGURE 1.
FIGURE 1.

COV08-0064 is a small-molecule antagonist of TLR9 signaling. Small molecules based on the (+)-morphinan structures were assessed for dose-dependent antagonism of TLR9 using a HEK-Blue TLR9 cell line expressing a TLR9-responsive, NF-κB–inducible SEAP. (A) NF-κB–induced SEAP was quantitated in culture supernatant using a colorimetric assay as described in Materials and Methods. TLR9 agonist ODN2006 was used at 1 μg/ml (closed boxes) and 100 ng/ml (closed triangles). Each data point represents the mean of three independent experiments with SDs shown in the brackets. Asterisks denote significant differences relative to groups with no compound added (p < 0.01), and two asterisks denote significant differences relative to 100 nM compound added using an unpaired two-tailed Student t test. (B) Chemical structures of four of the small molecules tested, specifically COV08-001, which was (+)-naloxone, as well as COV08-0037, COV08-0064, and COV08-0093.

COV08-0064 did not demonstrate antagonist activity at TLR2, TLR3, TLR4, TLR5, TLR7, TLR8, or the Null-1 NF-κB reporter cell line (Fig. 2A). COV08-0064 had no noted functionality as an agonist of TLR2, TLR3, TLR4, TLR5, TLR7, TLR8, TLR9, or the Null-1 NF-κB reporter cell line using standard agonists as positive controls (Fig. 2B).

FIGURE 2.
FIGURE 2.

COV08-0064 is a specific antagonist of TLR9 with no detected antagonism or agonism at other TLRs assayed, specifically TLR2, TLR3, TLR4, TLR5, TLR7, TLR8, and the parent cell line Null-1 HEK293, which is endogenously responsive to TNF-α induction of NF-κB. (A) NF-κB–induced SEAP was quantitated in culture supernatant using a colorimetric assay as described in Materials and Methods. COV08-0064 was coincubated at 1 nM, 100 nM, and 10 μM with 10 μM of each TLR agonist heat-killed L. monocytogenes (closed box), polyinosinic:polycytidylic acid (closed rhombus), LPS (closed triangle), CLO97 (open box), CLO75 (open rhombus), ODN2006 (open triangle), or TNF-α (gray box) in the respective HEK-Blue TLR reporter cell lines. Each data point represents the mean of three independent experiments with SDs shown in the brackets. (B) COV08-0064 (closed box) or the respective TLR ligand (closed rhombus) were separately incubated with of each of HEK-Blue TLR reporter cell lines noted. SEAP activity was quantitated. Each data point represents the mean of three independent experiments with SDs shown in the brackets. Asterisks denote significant differences between bracketed groups (p < 0.01) using an unpaired two-tailed Student t test.

COV08-0064 was further screened for alterations of binding of radioligand agonist and antagonists of many receptor families, as well as inhibition or activation of enzymes in cell culture at Cerep (Redmond, WA; data not shown). COV08-0064 modestly inhibited yohimbine binding to α-2B–adrenergic receptors and protriptyline binding to norepinephrine transporters. In addition, COV08-0064 was assessed for aqueous solubility, cell culture cytotoxicity, as well as CYP metabolism and interaction at Cerep, and was not cytotoxic (data not shown).

COV08-0064 is widely bioavailable with a long serum half life

COV08-0064 dosing by i.v. and s.c. dosing resulted in similar peak serum concentrations (Fig. 3A, 3B). Peak serum concentrations were significantly and markedly reduced by i.p. administration by comparison (Fig. 3C). In addition, per oral administration at 10 times the i.v. dose resulted in similar peak serum concentrations (Fig. 3D). The t1/2 of i.v., i.p., and s.c. administration were similar at 116–122 min. Per oral administration resulted in longer serum half life at 149 ± 9 min (Fig. 3E). In addition, s.c. and oral bioavailability were high at 92.5 ± 6.6% and 55.1 ± 7.1% of the i.v. dose, respectively.

FIGURE 3.
FIGURE 3.

COV08-0064 is widely bioavailable with a long serum half life. Sprague–Dawley rats were administered COV08-0064 at the doses noted (n = 3). Serum levels of COV08-0064 were sampled by jugular venous cannulation at the time pints indicated for (A) i.v. tail vein injection, (B) s.c. injection, (C) i.p. injection, and (D) per oral administration. Each data point represents one rat. (E) Serum half life and percent bioavailability in comparison with i.v. administration were determined as noted in Materials and Methods. Percent bioavailability for s.c. and i.p. doses for COV08-0064 were calculated as the ratio relative to i.v. dosing of area under the curve for the respective route divided by dose administered. Data are presented as mean and SD from three animals.

COV08-00064 dose dependently inhibits TLR9 but not TLR7 effects in PECs

COV08-0064 significantly decreases TLR9 type B ligand ODN1826-stimulated induction of pro–IL-1β and Nlrp3 expression in PECs in a dose-dependent manner (Fig. 4A). COV08-0064 at the same doses does not inhibit imiquimod (TLR7 ligand)–stimulated induction of pro–IL-1β and Nlrp3 expression in PECs (Fig. 4B). COV08-0064 also significantly inhibits TLR9 type B ligand ODN1826-stimulated induction of IL-12p40 in PECs from 342 ± 6 to 108 ± 27 pg/ml (Fig. 4C).

FIGURE 4.
FIGURE 4.

COV08-0064 dose dependently suppresses TLR9- but not TLR7-mediated proinflammatory gene expression and inflammasome priming in murine PECs. (A) Transcript for pro–IL-1β was quantitated by quantitative PCR in PECs treated with increasing doses of COV08-0064 as detailed in Materials and Methods. (B) Transcript for pro–IL-1β was quantitated by quantitative PCR in PECs coincubated with imiquimod at 5 μM and increasing doses of COV08-0064. (A) Transcript for pro–IL-1β and Nlrp3 were quantitated by quantitative PCR in PEC-treated ODN1826 at 1 μM with increasing doses of COV08-0064 as detailed in Materials and Methods. (B) Transcript for pro–IL-1β and Nlrp3 were quantitated by quantitative PCR in PECs coincubated with imiquimod at 5 μM and increasing doses of COV08-0064. (C) Supernatant protein levels of IL-12p40 were quantitated in PECs coincubated with ODN182 at 1 μM and COV08-0064 at 10 μM. Each data point represents the mean of three independent experiments with SDs shown in the brackets. Asterisks denote significant differences between bracketed groups (*p < 0.01). NS, No significant differences using an unpaired two-tailed Student t test.

COV08-0064 protects from APAP-induced acute liver injury and inflammation

COV08-0064 pretreatment at 60 mg/kg body weight protects wild-type mice from APAP-induced acute liver injury and inflammation. Liver necrosis was significantly reduced with COV08-0064 pretreatment compared with saline pretreatment as noted on H&E-stained liver tissue histology and on blinded histology scoring with necrosis scores of 1.2 ± 0.4 versus 2.1 ± 0.4, respectively (Fig. 5A, 5B). Liver hemorrhage was also significantly reduced with COV08-0064 pretreatment compared with saline pretreatment with hemorrhage scores of 1.1 ± 0.6 versus 2.2 ± 0.4, respectively (Fig. 5A, 5B). Neutrophil infiltration into the liver parenchyma was significantly reduced with COV08-0064 pretreatment compared with saline pretreatment with neutrophil counts per ×40 magnified high-powered field of 15 ± 2 versus 36 ± 3, respectively (Fig. 5C, 5D). Serum ALT levels were significantly reduced with COV08-0064 pretreatment from 10,375 ± 1990 to 3010 ± 549 U/ml (Fig. 5E). Liver caspase-1 activity was significantly reduced with COV08-0064 pretreatment from 93 ± 32 to 14 ± 21 activity U/g liver tissue (Fig. 5F). Liver IL-12p40 levels were significantly reduced with COV08-0064 pretreatment from 125 ± 13 to 76 ± 17 pg/g liver tissue (Fig. 5G). Liver IL-6 levels were significantly reduced with COV08-0064 pretreatment from 292 ± 17 to 238 ± 18 pg/g liver tissue (Fig. 5H).

FIGURE 5.
FIGURE 5.

COV08-0064 pretreatment decreases liver inflammation and injury in APAP hepatotoxicity in mice. C57BL/6 mice were administered COV08-0064 at 60 mg/kg body weight or saline by i.p. injection 1 h before administration of APAP at 500 mg/kg body weight or PBS. Mice were sacrificed at 12 h after APAP treatment for assessment detailed in this article. (A) Representative histology of H&E-stained, formalin-fixed liver sections at ×10 original magnification. (B) Blinded histology scoring of liver H&E-stained sections for hemorrhage and necrosis as described in Materials and Methods. (C) Representative immunostaining for Ly-6B.2 in formalin-fixed liver sections at ×10 magnification as described in Materials and Methods. (D) Blinded scoring for neutrophils per 40× field in liver sections immunostained for Ly-6B.2. (E) Serum ALT values. (F) Liver caspase-1 activity normalized to total protein level. (G) Liver IL-12p40 levels normalized to total protein level. (H) Liver IL-6 levels normalized to total protein level. Saline and PBS group (n = 2; white bars), PBS and APAP group (n = 8; black bars), COV08-0064 and PBS group (n = 2; dark gray bars), and COV08-0064 and APAP (n = 8; light gray bars). Each data point represents the mean with SDs shown in the brackets of these respective groups. Asterisks denote significant differences between the PBS and APAP group and the COV08-0064 and APAP group (p < 0.01) using an unpaired two-tailed Student t test.

COV08-0055, a morphinan structurally related to COV08-0064, has no TLR9 antagonist activity and does not protect from APAP-induced acute liver injury when administered at the same dose, timing, and route of administration as COV08-0064 (Supplemental Fig. 1).

COV08-0064 protects from cerulein hyperstimulation-induced acute pancreatic injury and inflammation

COV08-0064 treatment at 30 mg/kg body weight administered for three s.c. injections concurrent with cerulein treatment protects wild-type mice from acute pancreatic-induced acute liver injury and inflammation. Pancreatic edema is markedly reduced with COV08-0064 cotreatment, as noted in tissue sections (Fig. 6A). Neutrophil infiltration into the pancreas was significantly reduced with COV08-0064 cotreatment compared with saline cotreatment with neutrophil counts per ×40 magnified high-powered field of 10 ± 1 versus 24 ± 4, respectively (Fig. 6A, 6C). Serum amylase levels were significantly reduced with COV08-0064 cotreatment from 0.8 ± 0.1 to 0.5 ± 0.1 U/ml (Fig. 6B). Pancreatic trypsin activity was significantly reduced with COV08-0064 cotreatment from 13.1 ± 4.5 to 2.4 ± 0.5 normalized units (Fig. 6D). Pancreatic caspase-1 activity was significantly reduced with COV08-0064 cotreatment from 5.8 ± 0.9 to 2.3 ± 0.4 normalized units (Fig. 6E). Pancreatic IL-12p40 levels were significantly reduced with COV08-0064 cotreatment from 101 ± 17 to 40 ± 11 pg/g pancreatic tissue (Fig. 6F). Finally, pancreatic IL-6 levels were also significantly reduced with COV08-0064 cotreatment from 43 ± 7 to 25 ± 3 pg/g pancreatic tissue (Fig. 6G).

FIGURE 6.
FIGURE 6.

COV08-0064 cotreatment decreases pancreatic inflammation and injury in cerulein hyperstimulation induced acute pancreatitis in mice. C57BL/6 mice were administered COV08-0064 at 30 mg/kg body weight or saline by s.c. injection concurrent with doses 1, 4, and 7 of 10 sequential hourly doses of cerulein administered at 50 μg/kg body weight by i.p. injection. Mice were sacrificed at 1 h after the last dose of cerulein. (A) Representative immunostaining for Ly-6B.2 in formalin-fixed pancreatic sections at ×10 and ×40 magnification as described in Materials and Methods. (B) Serum amylase values determined by commercial colorimetric assay, Phadebas kit. (C) Blinded scoring for neutrophils per ×40 field in pancreatic sections immunostained for Ly-6B.2. (D) Pancreatic trypsin activation as determined by cleavage of fluorescent trypsin substrate Boc-QAR-MCA normalized to total amylase content and trypsin activity in untreated pancreas. (E) Pancreatic caspase-1 activation as determined by cleavage of fluorescent caspase-1 substrate Ac-YVAD-MCA normalized to total amylase content. (F) Pancreatic IL-12p40 levels normalized to total protein content. (G) Pancreatic IL-6 levels normalized to total protein content. For all experiments, untreated animals (n = 2; white bars), saline and cerulein group (n = 7; black bars), and COV08-0064 and cerulein group (n = 7; dark gray bars). Each data point represents the mean with SDs shown in the brackets of these respective groups. Asterisks denote significant differences between the saline plus cerulein group and COV08-0064 plus cerulein group (p < 0.01) using an unpaired two-tailed Student t test.

Discussion

Pattern-conserved molecules in bacteria and viruses were initially identified as ligands for TLRs. Activation of TLRs in vivo results in a range of biological responses with initiation of innate and adaptive immunity being dominant. Subsequently, it was identified that a wide range of self-molecules can also activate TLRs and result in sterile inflammatory responses that are responsible for a wide range of pathologies from acute liver failure, pancreatitis, to end-organ damage after crush injuries (22). This has identified specific TLRs as candidates for antagonism, with TLR9 being a required receptor for the development of many of these pathologies (2, 3, 10, 23–27).

Nuclear and mitochondrial DNA are known to be endogenous ligands for TLR9, and synthetic antagonists have the structure of nucleic acids and inhibitory ODNs or immunoregulatory DNA sequences, with chemical modification of the oligophosphate backbone to prevent degradation (11, 28). These are a rational choice but have a number of limitations. First, they have antagonist activity against TLR7 that is activated by single-stranded RNA molecules. This may be of significance as TLR7 and TLR9 have complex interactions that at times are opposing (13, 14). In addition to a lack of specificity, nucleic acid–based antagonists have immunoregulatory consequences independent of TLR9 that may become an issue with repeated administrations (17, 18).

To identify novel molecules with TLR9 antagonism, we screened >200 compounds using a TLR9 reporter cell line and identified 4 with some antagonist activity. Of these, COV08-0064 had the best antagonist profile, with almost complete inhibition of TLR9 response at a concentration of 1 μM (Fig. 1A). These compounds are unique TLR9 antagonists as their structure is based on the (+)-morphinan scaffold. The (+)-morphinan class of compounds refers to the enantiomeric analogues of traditional (-)-morphinans such as (-)-naltrexone and (-)-naloxone. These compounds are predicted to have no activity on any of the morphine receptors, and this was the case when tested by us (data not shown).

In contrast with DNA-based TLR9 antagonists, COV08-0064 had a high degree of specificity for TLR9, with no antagonist activity against TLR2, TLR3, TLR4, TLR7, or TLR8 at a concentration 10 times greater than required to inhibit most TLR9 responses (Fig. 2A). These results from the TLR reporter cell lines are confirmed by in vitro antagonism of primary peritoneal macrophages where COV08-0064 inhibited responses from a TLR9 but not a TLR7 agonist (Fig. 4A, 4B). Tissue macrophages including Kupffer cells have been implicated as major innate immune effector cells of sterile inflammation in the liver (29) and, more recently, in the pancreas (30, 31). Our laboratory has found that TLR9 contributes to tissue injury in APAP hepatotoxicity (2) and multiple mouse models of acute pancreatitis (3). We have also established that TLR9 induces proinflammatory signals in Kupffer cells (2), is expressed in the immune cell compartment of the pancreas, which is predominantly tissue macrophages, and is required for full inflammatory signaling in this compartment in the pancreas (3). Because our investigation is on early acute inflammation and tissue injury, the TLR9 contribution to innate immune responses through tissue macrophages was our focus for in vitro investigation.

In addition to the therapeutic implications of TLR9 selective antagonism, the fact that TLR7 and TLR9 share many aspects of their biology including compartmentalization and downstream signaling suggests that COV08-0064 is functioning very proximally. To further obtain information on unpredicted off-target effects, we performed a large screen of the effect of COV08-0064 on receptor binding and enzyme activity in many pathways. This screen demonstrated that COV08-0064 did not have a significant effect on major pathways (data not shown). COV08-0064 modestly inhibited yohimbine binding to α-2B–adrenergic receptors, protriptyline binding to norepinephrine transporters, and inhibition of COX-1 by diclofenac. In addition, COV08-0064 was assessed for aqueous solubility, cell culture cytotoxicity, as well as CYP metabolism, all of which were negative. Collectively, this entirely novel structure of molecules has potent TLR9 antagonist activity with a high degree of selectivity. In addition to selectivity, DNA-based antagonists have poor oral bioavailability. COV08-0064 had 55% oral bioavailability and a half life of 149 min. i.v. and s.c. routes had slightly shorter half lives \16 ± 15 and 122 ± 20), but still significantly longer than the half lives of DNA-based antagonists, which have half lives in the range of 20–25 min.

Because TLRs were initially identified as sensors of molecules on infectious agents, the use of TLR antagonists raises questions regarding increased susceptibility to pathogens (11). This is, however, unlikely to be the case because there is now a significant amount of data on the infectious consequences of the loss of these pathways in mice and in humans. Mice deficient in individual TLRs or in common signaling pathways are viable and healthy without special care. Humans with loss of analogous pathways or individual TLRs also have a phenotype limited to childhood (32). Direct interventional information is available from a phase 1 trial of a combined TLR9 and seven antagonists (DV1179) that did not produce any adverse effects (33).

The ability of this novel structure of molecules to inhibit TLR9 activation is unexpected, and many of the earlier characteristics of specificity, delivery, and pharmacokinetics make them attractive for therapeutic antagonism. The range of noninfectious pathologies in which TLR9 has been demonstrated to play a role has been rapidly expanding. Currently, these include acute liver failure, pancreatitis, liver and lung fibrosis, ischemia-reperfusion injury, graft-versus-host disease after bone marrow transplantation, as well as the systemic inflammatory response after a crush injury (2, 3, 23, 24, 34). We tested the ability of COV08-0064 in two models, one of acute liver injury by APAP and one of acute pancreatitis. In both conditions, COV08-0064 significantly reduced serum markers of tissue injury, and reduced tissue inflammation and damage as assessed by histology, neutrophil migration, caspase-1 activity, and TLR9-induced cytokine production (Figs. 5, 6). These data have a number of important implications. The first is that COV08-0064, in addition to having the desired TLR9 antagonist activities in in vitro tests, is also able to reduce a full in vivo disease response. The additional point is that COV08-0064 was protective in disease models of two different organs, in which injury was initiated by very different mechanisms. To confirm that off-target effects were not responsible for COV08-0064–mediated protection, we assessed COV08-0055, a structurally similar analogue with no TLR9 antagonist function, and found the latter to be without hepatoprotective effects in vivo (Supplemental Fig. 1). The liver injury was initiated by APAP, which is responsible for the majority of cases of acute liver failure in the United States. Toxic metabolites of APAP induce a necrotic hepatocyte death that results in the release of damage-associated molecular patterns including nuclear and mitochondrial DNA further amplifying the initial injury (2, 35). Cerulein-induced acute pancreatitis is not due to the production of toxic metabolites of the drug, but by hyperstimulation of the exocrine pancreas that results in the excess production of digestive enzymes. This is a mechanism that is known to be active in many forms of acute pancreatitis. The ability of COV08-0064 to protect in two models demonstrates the importance of downstream activation of TLR9 after initial tissue injury, and also the therapeutic opportunities of TLR9 antagonism.

Disclosures

B.N.T., D.W.B., J.P.M., and K.P.G. are employees of Covidien, which generated the small-molecule TLR9 antagonists studied in this article. The other authors have no financial conflicts of interest.

Footnotes

  • This work was supported by Covidien Mallinckrodt; National Institutes of Health Grants R01DK076674-01A2 (to W.M.), T32 DK7356 (to R.H.), and K08DK092281 (to R.H.); a Veteran's Administration Merit Award (to W.M.); and the resources of the Yale Liver Center (supported by National Institutes of Health Grant P30DK34989).

  • The online version of this article contains supplemental material.

  • Abbreviations used in this article:

    ALT
    alanine aminotransferase
    APAP
    acetaminophen
    DPBS
    Dulbecco’s PBS
    ODN
    oligonucleotide
    PEC
    peritoneal elicited cell
    SEAP
    secreted placental alkaline phosphatase.

  • Received August 8, 2012.
  • Accepted February 12, 2013.

References

  1. ↵
    1. Plitas G.,
    2. B. M. Burt,
    3. H. M. Nguyen,
    4. Z. M. Bamboat,
    5. R. P. DeMatteo
    . 2008. Toll-like receptor 9 inhibition reduces mortality in polymicrobial sepsis. J. Exp. Med. 205: 1277–1283.
    OpenUrlAbstract/FREE Full Text
  2. ↵
    1. Imaeda A. B.,
    2. A. Watanabe,
    3. M. A. Sohail,
    4. S. Mahmood,
    5. M. Mohamadnejad,
    6. F. S. Sutterwala,
    7. R. A. Flavell,
    8. W. Z. Mehal
    . 2009. Acetaminophen-induced hepatotoxicity in mice is dependent on Tlr9 and the Nalp3 inflammasome. J. Clin. Invest. 119: 305–314.
    OpenUrlPubMed
  3. ↵
    1. Hoque R.,
    2. M. Sohail,
    3. A. Malik,
    4. S. Sarwar,
    5. Y. Luo,
    6. A. Shah,
    7. F. Barrat,
    8. R. Flavell,
    9. F. Gorelick,
    10. S. Husain,
    11. W. Mehal
    . 2011. TLR9 and the NLRP3 inflammasome link acinar cell death with inflammation in acute pancreatitis. Gastroenterology 141: 358–369.
    OpenUrlCrossRefPubMed
  4. ↵
    1. de Jong S. D.,
    2. G. Basha,
    3. K. D. Wilson,
    4. M. Kazem,
    5. P. Cullis,
    6. W. Jefferies,
    7. Y. Tam
    . 2010. The immunostimulatory activity of unmethylated and methylated CpG oligodeoxynucleotide is dependent on their ability to colocalize with TLR9 in late endosomes. J. Immunol. 184: 6092–6102.
    OpenUrlAbstract/FREE Full Text
  5. ↵
    1. Collins L. V.,
    2. S. Hajizadeh,
    3. E. Holme,
    4. I. M. Jonsson,
    5. A. Tarkowski
    . 2004. Endogenously oxidized mitochondrial DNA induces in vivo and in vitro inflammatory responses. J. Leukoc. Biol. 75: 995–1000.
    OpenUrlAbstract/FREE Full Text
  6. ↵
    1. Wei T.,
    2. J. Gong,
    3. F. Jamitzky,
    4. W. M. Heckl,
    5. R. W. Stark,
    6. S. C. Rössle
    . 2009. Homology modeling of human Toll-like receptors TLR7, 8, and 9 ligand-binding domains. Protein Sci. 18: 1684–1691.
    OpenUrlCrossRefPubMed
  7. ↵
    1. Peter M. E.,
    2. A. V. Kubarenko,
    3. A. N. Weber,
    4. A. H. Dalpke
    . 2009. Identification of an N-terminal recognition site in TLR9 that contributes to CpG-DNA-mediated receptor activation. J. Immunol. 182: 7690–7697.
    OpenUrlAbstract/FREE Full Text
  8. ↵
    1. Gursel I.,
    2. M. Gursel,
    3. H. Yamada,
    4. K. J. Ishii,
    5. F. Takeshita,
    6. D. M. Klinman
    . 2003. Repetitive elements in mammalian telomeres suppress bacterial DNA-induced immune activation. J. Immunol. 171: 1393–1400.
    OpenUrlAbstract/FREE Full Text
    1. Yu D.,
    2. D. Wang,
    3. F. G. Zhu,
    4. L. Bhagat,
    5. M. Dai,
    6. E. R. Kandimalla,
    7. S. Agrawal
    . 2009. Modifications incorporated in CpG motifs of oligodeoxynucleotides lead to antagonist activity of toll-like receptors 7 and 9. J. Med. Chem. 52: 5108–5114.
    OpenUrlCrossRefPubMed
  10. ↵
    1. Barrat F. J.,
    2. T. Meeker,
    3. J. Gregorio,
    4. J. H. Chan,
    5. S. Uematsu,
    6. S. Akira,
    7. B. Chang,
    8. O. Duramad,
    9. R. L. Coffman
    . 2005. Nucleic acids of mammalian origin can act as endogenous ligands for Toll-like receptors and may promote systemic lupus erythematosus. J. Exp. Med. 202: 1131–1139.
    OpenUrlAbstract/FREE Full Text
  11. ↵
    1. Barrat F. J.,
    2. R. L. Coffman
    . 2008. Development of TLR inhibitors for the treatment of autoimmune diseases. Immunol. Rev. 223: 271–283.
    OpenUrlCrossRefPubMed
  12. ↵
    1. Lau C. M.,
    2. C. Broughton,
    3. A. S. Tabor,
    4. S. Akira,
    5. R. A. Flavell,
    6. M. J. Mamula,
    7. S. R. Christensen,
    8. M. J. Shlomchik,
    9. G. A. Viglianti,
    10. I. R. Rifkin,
    11. A. Marshak-Rothstein
    . 2005. RNA-associated autoantigens activate B cells by combined B cell antigen receptor/Toll-like receptor 7 engagement. J. Exp. Med. 202: 1171–1177.
    OpenUrlAbstract/FREE Full Text
  13. ↵
    1. Christensen S. R.,
    2. J. Shupe,
    3. K. Nickerson,
    4. M. Kashgarian,
    5. R. A. Flavell,
    6. M. J. Shlomchik
    . 2006. Toll-like receptor 7 and TLR9 dictate autoantibody specificity and have opposing inflammatory and regulatory roles in a murine model of lupus. Immunity 25: 417–428.
    OpenUrlCrossRefPubMed
  14. ↵
    1. Santiago-Raber M. L.,
    2. I. Dunand-Sauthier,
    3. T. Wu,
    4. Q. Z. Li,
    5. S. Uematsu,
    6. S. Akira,
    7. W. Reith,
    8. C. Mohan,
    9. B. L. Kotzin,
    10. S. Izui
    . 2010. Critical role of TLR7 in the acceleration of systemic lupus erythematosus in TLR9-deficient mice. J. Autoimmun. 34: 339–348.
    OpenUrlCrossRefPubMed
  15. ↵
    1. Wang D.,
    2. L. Bhagat,
    3. D. Yu,
    4. F. G. Zhu,
    5. J. X. Tang,
    6. E. R. Kandimalla,
    7. S. Agrawal
    . 2009. Oligodeoxyribonucleotide-based antagonists for Toll-like receptors 7 and 9. J. Med. Chem. 52: 551–558.
    OpenUrlCrossRefPubMed
  16. ↵
    1. Franklin B. S.,
    2. S. T. Ishizaka,
    3. M. Lamphier,
    4. F. Gusovsky,
    5. H. Hansen,
    6. J. Rose,
    7. W. Zheng,
    8. M. A. Ataíde,
    9. R. B. de Oliveira,
    10. D. T. Golenbock,
    11. R. T. Gazzinelli
    . 2011. Therapeutical targeting of nucleic acid-sensing Toll-like receptors prevents experimental cerebral malaria. Proc. Natl. Acad. Sci. USA 108: 3689–3694.
    OpenUrlAbstract/FREE Full Text
  17. ↵
    1. Trieu A.,
    2. N. Bokil,
    3. J. A. Dunn,
    4. T. L. Roberts,
    5. D. Xu,
    6. F. Y. Liew,
    7. D. A. Hume,
    8. K. J. Stacey,
    9. M. J. Sweet
    . 2009. TLR9-independent effects of inhibitory oligonucleotides on macrophage responses to S. typhimurium. Immunol. Cell Biol. 87: 218–225.
    OpenUrlCrossRefPubMed
  18. ↵
    1. Babiuk S.,
    2. N. Mookherjee,
    3. R. Pontarollo,
    4. P. Griebel,
    5. S. van Drunen Littel-van den Hurk,
    6. R. Hecker,
    7. L. Babiuk
    . 2004. TLR9-/- and TLR9+/+ mice display similar immune responses to a DNA vaccine. Immunology 113: 114–120.
    OpenUrlCrossRefPubMed
  19. ↵
    1. Fernandes-Alnemri T.,
    2. J. W. Yu,
    3. P. Datta,
    4. J. Wu,
    5. E. S. Alnemri
    . 2009. AIM2 activates the inflammasome and cell death in response to cytoplasmic DNA. Nature 458: 509–513.
    OpenUrlCrossRefPubMed
    1. Barber G. N.
    2011. Innate immune DNA sensing pathways: STING, AIMII and the regulation of interferon production and inflammatory responses. Curr. Opin. Immunol. 23: 10–20.
    OpenUrlCrossRefPubMed
  21. ↵
    1. Unterholzner L.,
    2. S. E. Keating,
    3. M. Baran,
    4. K. A. Horan,
    5. S. B. Jensen,
    6. S. Sharma,
    7. C. M. Sirois,
    8. T. Jin,
    9. E. Latz,
    10. T. S. Xiao,
    11. et al
    . 2010. IFI16 is an innate immune sensor for intracellular DNA. Nat. Immunol. 11: 997–1004.
    OpenUrlCrossRefPubMed
  22. ↵
    1. Kono H.,
    2. K. L. Rock
    . 2008. How dying cells alert the immune system to danger. Nat. Rev. Immunol. 8: 279–289.
    OpenUrlCrossRefPubMed
  23. ↵
    1. Bamboat Z. M.,
    2. V. P. Balachandran,
    3. L. M. Ocuin,
    4. H. Obaid,
    5. G. Plitas,
    6. R. P. DeMatteo
    . 2010. Toll-like receptor 9 inhibition confers protection from liver ischemia-reperfusion injury. Hepatology 51: 621–632.
    OpenUrlCrossRefPubMed
  24. ↵
    1. Watanabe A.,
    2. A. Hashmi,
    3. D. A. Gomes,
    4. T. Town,
    5. A. Badou,
    6. R. A. Flavell,
    7. W. Z. Mehal
    . 2007. Apoptotic hepatocyte DNA inhibits hepatic stellate cell chemotaxis via toll-like receptor 9. Hepatology 46: 1509–1518.
    OpenUrlCrossRefPubMed
    1. Miura K.,
    2. Y. Kodama,
    3. S. Inokuchi,
    4. B. Schnabl,
    5. T. Aoyama,
    6. H. Ohnishi,
    7. J. M. Olefsky,
    8. D. A. Brenner,
    9. E. Seki
    . 2010. Toll-like receptor 9 promotes steatohepatitis by induction of interleukin-1beta in mice. Gastroenterology 139: 323–334, e7.
    OpenUrlCrossRefPubMed
    1. Meneghin A.,
    2. E. S. Choi,
    3. H. L. Evanoff,
    4. S. L. Kunkel,
    5. F. J. Martinez,
    6. K. R. Flaherty,
    7. G. B. Toews,
    8. C. M. Hogaboam
    . 2008. TLR9 is expressed in idiopathic interstitial pneumonia and its activation promotes in vitro myofibroblast differentiation. Histochem. Cell Biol. 130: 979–992.
    OpenUrlCrossRefPubMed
  27. ↵
    1. Anders H. J.,
    2. V. Vielhauer,
    3. V. Eis,
    4. Y. Linde,
    5. M. Kretzler,
    6. G. Perez de Lema,
    7. F. Strutz,
    8. S. Bauer,
    9. M. Rutz,
    10. H. Wagner,
    11. et al
    . 2004. Activation of toll-like receptor-9 induces progression of renal disease in MRL-Fas(lpr) mice. FASEB J. 18: 534–536.
    OpenUrlAbstract/FREE Full Text
  28. ↵
    1. Hemmi H.,
    2. O. Takeuchi,
    3. T. Kawai,
    4. T. Kaisho,
    5. S. Sato,
    6. H. Sanjo,
    7. M. Matsumoto,
    8. K. Hoshino,
    9. H. Wagner,
    10. K. Takeda,
    11. S. Akira
    . 2000. A Toll-like receptor recognizes bacterial DNA. Nature 408: 740–745.
    OpenUrlCrossRefPubMed
  29. ↵
    1. Goldin R. D.,
    2. I. D. Ratnayaka,
    3. C. S. Breach,
    4. I. N. Brown,
    5. S. N. Wickramasinghe
    . 1996. Role of macrophages in acetaminophen (paracetamol)-induced hepatotoxicity. J. Pathol. 179: 432–435.
    OpenUrlCrossRefPubMed
  30. ↵
    1. Dang S. C.,
    2. D. L. Jiang,
    3. M. Chen,
    4. D. Li,
    5. J. X. Zhang
    . 2010. Clodronate-containing liposomes attenuate lung injury in rats with severe acute pancreatitis. J. Zhejiang Univ. Sci. B 11: 828–835.
    OpenUrlCrossRefPubMed
  31. ↵
    1. Nakamichi I.,
    2. A. Habtezion,
    3. B. Zhong,
    4. C. H. Contag,
    5. E. C. Butcher,
    6. M. B. Omary
    . 2005. Hemin-activated macrophages home to the pancreas and protect from acute pancreatitis via heme oxygenase-1 induction. J. Clin. Invest. 115: 3007–3014.
    OpenUrlCrossRefPubMed
  32. ↵
    1. Yang K.,
    2. A. Puel,
    3. S. Zhang,
    4. C. Eidenschenk,
    5. C. L. Ku,
    6. A. Casrouge,
    7. C. Picard,
    8. H. von Bernuth,
    9. B. Senechal,
    10. S. Plancoulaine,
    11. et al
    . 2005. Human TLR-7-, -8-, and -9-mediated induction of IFN-alpha/beta and -lambda Is IRAK-4 dependent and redundant for protective immunity to viruses. Immunity 23: 465–478.
    OpenUrlCrossRefPubMed
  33. ↵
    1. Connolly D. J.,
    2. L. A. O’Neill
    . 2012. New developments in Toll-like receptor targeted therapeutics. Curr. Opin. Pharmacol. 12: 510–518.
    OpenUrlCrossRefPubMed
  34. ↵
    1. Calcaterra C.,
    2. L. Sfondrini,
    3. A. Rossini,
    4. M. Sommariva,
    5. C. Rumio,
    6. S. Ménard,
    7. A. Balsari
    . 2008. Critical role of TLR9 in acute graft-versus-host disease. J. Immunol. 181: 6132–6139.
    OpenUrlAbstract/FREE Full Text
  35. ↵
    1. McGill M. R.,
    2. M. R. Sharpe,
    3. C. D. Williams,
    4. M. Taha,
    5. S. C. Curry,
    6. H. Jaeschke
    . 2012. The mechanism underlying acetaminophen-induced hepatotoxicity in humans and mice involves mitochondrial damage and nuclear DNA fragmentation. J. Clin. Invest. 122: 1574–1583.
    OpenUrlCrossRefPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section