Synthesis of deuterium-labelled amlexanox and its metabolic stability against mouse, rat and human microsomes

Abstract

As part of a program toward making analogues of amlexanox (1), currently under clinical investigation for the treatment of type 2 diabetes and obesity, we have synthesized derivative 5 in which deuterium has been introduced into two sites of metabolism on the C-7 isopropyl function of amlexanox. The synthesis of 5 was completed in an efficient three-step process utilizing reduction of key olefin 7b to 8 by Wilkinson’s catalyst to provide specific incorporation of di-deuterium across the double bond. Compound 5 displayed nearly equivalent potency to amlexanox (IC₅₀ 1.1 μM vs 0.6 μM, respectively) against recombinant human TBK1. When incubated with human, rat, and mouse liver microsomes, amlexanox (1) and d₂-amlexanox (5) were stable (t_1/2 > 60 min) with 1 showing marginally greater stability relative to 5 except for rat liver microsomes. These data show that incorporating deuterium into two sites of metabolism does not majorly suppress Cyp-mediated metabolism relative to amlexanox.

Graphical Abstract

A derivative of the TBK1 inhibitor amlexanox (1) has been synthesized with deuterium introduced into two sites of metabolism on the C-7 isopropyl function. d₂-Amlexanox (5) displayed nearly equivalent potency to amlexanox (IC₅₀ 1.1 μM vs 0.6 μM, respectively) against recombinant human TBK1. When incubated with human, rat, and mouse liver microsomes, 1 and 5 were stable (t_1/2 > 60 minutes) with 1 showing marginally greater stability relative to 5 except for rat liver microsomes.

1. INTRODUCTION

There is currently a worldwide epidemic of obesity, which is manifesting itself as a leading risk factor in industrialized countries for the development of type 2 diabetes, dyslipidaemia, non-alcoholic fatty liver disease, cardiovascular disease and some cancers.¹ Obesity results from a sustained imbalance between calorie intake and energy expenditure and is characterized by chronic, low-grade inflammation in liver and adipose tissues, which in turn attenuates the actions of multiple hormones to produce a dysregulated metabolic state.^2,3,4 Prior studies on the role of inflammation in the generation of insulin resistance and type 2 diabetes during obesity implicate an important role for the nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) pathway, revealing the role of two non-canonical inhibitors of kappa B (IκB) kinases, TANK-binding kinase 1 (TBK1) and inhibitor of nuclear factor kappa-B kinase subunit ε (IKKε), in insulin-independent pathways that promote energy storage and block adaptive energy expenditure during obesity.^5,6 Since none of the currently available and widely used treatments for obesity addresses the underlying energy imbalance associated with inflammation pathways,⁷ we initiated a high throughput screen for small molecule inhibitors of the TBK1 and IKKε kinases. One hit from our screen was amlexanox (1; Figure 1), an approved drug developed by Takeda that is used for the treatment of canker sores, asthma, and allergic rhinitis.^8,9 The compound is a modest inhibitor of TBK1 and IKKε and reproduced the effects of the IKKε knockout phenotype, which include weight loss in obese mice, improved insulin sensitivity, and reduced hepatic steatosis and inflammation. Amlexanox also increased the level of the second messenger molecule cAMP and triggered the release of the hormone interleukin-6 (IL-6) from fat cells thereby reducing the production of glucose. Two recent clinical trials indicated that amlexanox treatment effectively lowered glycated hemoglobin (HbA1c) in a subset of patients with obesity and type 2 diabetes.¹⁰ Despite these promising results, the future utility of amlexanox in the clinic for obesity and type 2 diabetes intervention may be limited by its solubility, modest potency, and/or metabolic profile.

Figure 1.

Structures of amlexanox (1), metabolites (2 – 4) and d₂-amlexanox (5)

Previous studies by Takeda investigators showed that the oral administration of amlexanox (1) in mice, rats, guinea pigs, and dogs generated in plasma a major mono-hydroxyl metabolite (2) from oxidation of the C-7 isopropyl substituent, and two minor metabolites 3 and 4^11,12 (Figure 1). The glucuronide of 1 was also identified in rats. In humans, the main metabolite was 2. The structures of metabolites 2 and 3 have been confirmed by synthesis.¹³

As part of a program toward making analogues of amlexanox,^14,15 we decided to synthesize a derivative 5 (Figure 1) in which deuterium is introduced into two sites of metabolism on the C-7 isopropyl function. The deuterium kinetic isotope effect associated with placing deuterium at the site of metabolic derivatization slows metabolic activation and thus is expected to increase the lifetime of the active drug in vivo.¹⁶ Herein we provide data of the activity of 5 relative to amlexanox (1) in inhibiting TBK1 and its metabolic stability against mouse, rat and human microsomes.

2. RESULTS AND DISCUSSION

Synthetic chemistry.

Initial approaches toward generating d₁-amlexanox, in which the benzylic C-7 methine was exchanged for deuterium, centered on catalytic exchange chemistry. Thus, reaction of amlexanox (1) under a variety of conditions (hydrogen gas, 10 – 30% palladium on activated carbon, methanol-d₄, 25–50°C) left starting material only. We then focused on incorporation of deuterium through room temperature hydrogenation of the olefin in compound 7a, made by known procedures,^13,17 with deuterium gas in the presence of 10% palladium on activated carbon in ethyl acetate. Thus, a control reduction of 7a with hydrogen gave cleanly the non-deuterated C-7 isopropyl product. Substitution of deuterium for hydrogen gas yielded a new set of products with mass spectrum peaks associated with incorporation of more than two atoms of deuterium. The ¹H NMR spectrum suggested that the predominant N-Boc carboxylic methyl ester product had incorporated four atoms of deuterium and ¹³C NMR experiments on the next step hydrolysis product (analogous to 5) indicated the presence of seven uniquely substituted C-7 methine carbons. This showed that deuterium had been incorporated non-specifically into both C-7 methyl groups as well as the methine carbon. There is literature precedence for such scrambling of isotopic incorporation into benzylic and allylic sites.¹⁸

After unsuccessfully surveying other heterogenous catalytic conditions, we then resorted to olefin reduction with a homogeneous catalyst (Scheme 1). Employing our recently published methodology of introducing C-7 substituents via late-stage coupling chemistry to make amlexanox analogues¹⁵, we reacted tricyclic bromide 6¹⁹ with 2-propenyl pinacol boronic ester under standard Suzuki coupling conditions to give 7b in 91% yield. Reaction with tris(triphenylphosphine)rhodium(I) chloride (Wilkinson’s catalyst) and deuterium gas, as previously described for similar reductions,^20,21 proceeded in high yield to give key intermediate 8 with specific incorporation of di-deuterium across the double bond. Standard hydrolysis conditions as described before¹⁵ then provided d₂-amlexanox (5) in 92% yield and 72% overall yield from starting ester 6. All compounds were rigorously purified by flash chromatography or crystallization, and their structural assignments are supported by diagnostic peaks in the ¹H and ¹³C NMR spectra and by high resolution mass spectrometry. Digital copies of NMR and mass spectra can be found in Supplementary Materials Online.

Scheme 1.

Synthesis of d₂-amlexanox (5)

The ¹H NMR spectra of the upfield region (δ 1–3 ppm) for the C-7 isopropyl substituent (Figure 2) clearly show clean, specific incorporation of di-deuterium into 5. Complete collapse of the symmetrical pentet at δ 3.05 ppm occurs in 1 and the symmetrical doublet at δ 1.25 ppm, integrating for six protons, changes to an unsymmetrical doublet in 5 integrating for five protons. High resolution mass spectra for 5 and 8 (Supplementary Materials Online) also support specific incorporation of di-deuterium.

Figure 2.

Comparison ¹H NMR spectra in DMSO-d₆ for C-7 substituent of amlexanox (1) and d₂-amlexanox (5).

Metabolic stability studies.

Amlexanox (1) and d₂-amlexanox (5) are moderately metabolized by human, rat and mice liver microsomes. The in vitro disappearance half-life (t_1/2) of both compounds in all three microsomes is over 60 min (Figure 3). During the incubation in human and mice liver microsomes, about 8% more unmetabolized amlexanox remains after 60 min compared to d₂-amlexanox. However, d₂-amlexanox (5) is more stable than amlexanox (1) in rat liver microsomes. Both compounds show greater stability in human liver microsomes relative to the other microsomes. Verapamil was employed as positive control compound in order to ensure that the microsomes and NADPH used in this study were active and that incubation conditions were appropriate to perform the studies. Based on the observed degradation patterns, amlexanox (1) and d₂-amlexanox (5) are considered to be stable against the three liver microsomes.

Figure 3.

Human (H), rat (R) and mouse (M) microsomal stability for amlexanox (1) and d₂-amlexanox (5)

Activity against TBK1 kinase.

d₂-Amlexanox (5) was investigated for inhibition of recombinant human TBK1 as described previously.¹⁵ The IC₅₀ values shown were determined by dose-response radiometric kinase assays with myelin basic protein as substrate, and are technically apparent IC₅₀ values. Each is an average of three separate determinations. Since the isopropyl group on amlexanox is solvent exposed,¹⁴ incorporation of deuterium is not expected to affect inhibitor potency. Indeed, the in vitro potencies of amlexanox (1) and d₂-amlexanox (5) against TBK1 are 0.6 ± 0.1 and 1.1 ± 0.3 μM, respectively (Figure 4 and Supplementary Materials Online). The difference is not statistically significant demonstrating that amlexanox can be deuterated without affecting inhibitor potency.

Figure 4.

In vitro potency of amlexanox (1) and d₂-amlexanox (5) against purified TBK1 as determine via radiometric kinase assay measuring phosphorylation of the substrate myelin basic protein. The goodness of fit (r²) for 1 and 5 is 0.97 and 0.91, respectively, with no statistically significant difference via Student’s two-tailed T-test (n=4, reported as mean ± SD).

3. CONCLUSIONS

As part of a program toward making analogues of amlexanox (1), which is being evaluated clinically for the treatment of type 2 diabetes and obesity, we decided to synthesize a derivative 5 to determine if placing deuterium at two sites of known metabolic activation of amlexanox might increase its half-life against three species of liver microsomes. Microsomes are rich in cytochrome (Cyp) P-450 enzymes, which are the major enzymes involved in drug metabolism and account for about 75% of total metabolism.²² The most common pathway of Cyp-mediated metabolism is hydroxylation of aliphatic carbons.

The data in Figure 3 show that amlexanox (1) and d₂-amlexanox (5) are relatively stable against all three liver microsomes (t_1/2 > 60 min) with amlexanox showing marginally greater stability relative to d₂-amlexanox except for rat liver microsomes. Hence, blocking two sites of metabolism with incorporation of a single deuterium on the C-7 methine carbon and another on a flanking methyl group of 5 does not majorly suppress Cyp-mediated metabolism relative to amlexanox. Metabolite identification experiments will be required to expand upon these findings. Additionally, synthetic chemistry needs to be developed that can specifically incorporate C-7 deuterium patterns (e.g., -CH(CH₃)(CH₂D) and -CD(CH₃)₂) aligned with known metabolites 2-4, and additional patterns (e.g., -CH(CD₃)₂, -CD(CD₃)₂) to more fully explore the deuterium isotope effect.

4. EXPERIMENTAL

4.1. Materials and instrumentation.

All starting monomers were obtained from commercial suppliers and were used without further purification. Routine ¹H NMR spectra were recorded at 400 or 500 MHz on a Varian 400 or 500 instrument, respectively, with CDCl₃ or DMSO-d₆ as solvent. Chemical shift values are recorded in δ units (ppm). High resolution mass spectrometry (HRMS) analysis was performed on an Agilent Q-TOF system. Analytical HPLC was performed on an Agilent 1100 series instrument with an Agilent Zorbax Eclipse Plus C18 (4.6 mm × 75 mm, 3.5 μm particle size) column with the gradient 10% acetonitrile/water (1 min), 10−90% acetonitrile/water (6 min), and 90% acetonitrile/water (2 min) flow = 1 mL/min. Thin-layer chromatography (TLC) was performed on silica gel GHLF plates (250 μm) purchased from Analtech. Column chromatography was carried out in the flash mode utilizing silica gel (220−240 mesh) purchased from Silicycle. Extraction solutions were dried over anhydrous sodium sulfate prior to concentration.

4.2. Ethyl 2-amino-5-oxo-7-(prop-1-en-2-yl)-5H-chromeno[2,3-b]pyridine-3-carboxylate (7b).

A solution of tripotassium phosphate (2.63 g, 12.4 mmol) in water (6 mL) was added to a nitrogen-degassed suspension of ethyl 2-amino-7-bromo-5-oxo-5H-chromeno[2,3-b]pyridine-3-carboxylate¹⁹ (6; 750 mg, 2.07 mmol), 4,4,5,5-tetramethyl-2-(prop-1-en-2-yl)-1,3,2-dioxaborolane (833 mg, 4.96 mmol), triphenylphosphine (271 mg, 1.03 mmol), Pd(OAc)₂ (60.3 mg, 0.27 mmol) in p-dioxane (60 mL) in a 500 mL RB flask. The mixture was heated at 100 −105°C under nitrogen for 2 h. After cooling, the mixture was diluted with water (50 mL) and extracted with dichloromethane (2 × 120 mL). The combined organic phases were dried (Na₂SO₄) and concentrated to leave an orange solid that was triturated sequentially with small volumes of ethanol and ethyl ether, collected, and dried to leave 7b (610 mg, 91%) as a white solid: mp 245–247°C. ¹H NMR (400 MHz, chloroform-d): δ 9.16 (s, 1H), 8.38 (s, 1H), 8.31 (d, J = 2.3 Hz, 1H), 7.86 (dd, J = 8.8, 2.4 Hz, 1H), 7.47 (d, J = 8.7 Hz, 1H), 5.86 (s, 1H), 5.49 (d, J = 1.3 Hz, 1H), 5.23 – 5.18 (m, 1H), 4.41 (q, J = 7.1 Hz, 2H), 2.24 (dd, J = 1.5, 0.8 Hz, 3H), 1.45 (t, J = 7.1 Hz, 3H). HRMS: m/z (M+H)⁺: calcd. for C₁₈H₁₆N₂O₄: 325.1183; Found, 325.1188.

4.3. Ethyl 2-amino-5-oxo-7-(propan-2-yl-1,2-d₂)-5H-chromeno[2,3-b]pyridine-3-carboxylate (8).

Tris(triphenylphosphine)rhodium(I) chloride (Wilkinson’s catalyst; 339 mg, 0.366 mmol) was completely dissolved in anhydrous dichloromethane (10 mL) and ethanol-d₁ (10 mL) in a 250 mL 3-neck RB flask under nitrogen. The flask with the resulting clear red brown solution was evacuated by a water pump and then flushed with deuterium gas in a balloon. This flushing/evacuation procedure was repeated three more times, and then the mixture was stirred under deuterium gas (balloon pressure) for 20 min at room temperature. Then compound 7 (360 mg, 1.11 mmol) dissolved in a mixture of warm dichloromethane (75 mL) and ethanol-d₁ (75 mL) was added by syringe. The resulting solution was stirred at room temperature for 4.5 h and then concentrated to a solid that was triturated in a small volume of benzene. The collected solids were further purified by flash silica gel column chromatography (elution with 0.5 – 1.0% methanol in dichloromethane). Product fractions were combined and concentrated to leave 8 (315 mg, 86%) as a while solid: mp 239–241°C. ¹H NMR (400 MHz, chloroform-d): δ 9.16 (s, 1H), 8.34 (s, 1H), 8.11 (d, J = 2.3 Hz, 1H), 7.59 (dd, J = 8.6, 2.3 Hz, 1H), 7.44 (d, J = 8.6 Hz, 1H), 5.84 (s, 1H), 4.41 (q, J = 7.1 Hz, 2H), 1.44 (t, J = 7.1 Hz, 3H), 1.30 (d, J = 6.5 Hz, 5H). HRMS: m/z (M+H)⁺ Calcd. for C₁₈H₁₆D₂N₂O₄: 329.1465; Found 329.1480.

4.4. 2-Amino-5-oxo-7-(propan-2-yl-1,2-d₂)-5H-chromeno[2,3-b]pyridine-3-carboxylic acid (d₂-amlexanox; 5).

To a suspension of 8 (460 mg, 1.4 mmol) in ethanol (80 mL) and water (10 mL) was added 1 N aq. NaOH (10 mL) and the mixture was stirred at 50°C for 90 min. TLC (5% methanol in dichloromethane) showed consumption of starting material. The mixture was filtered and the filtrate concentrated to a wet solid that dissolved in water (50 mL). The solution was acidified with 1 N aq. HCl to pH ~3 and the precipitated solids were stirred at room temperature for 20 h. After storage at room temperature for another 24 h, the solids were collected, washed with water, and dried under vacuum at 60°C for 4 h to leave 5 (385 mg, 92%) as an off-white solid: mp > 300°C (dec.). ¹H NMR (500 MHz, DMSO-d₆): δ 13.45 (s, 1H), 8.80 (s, 1H), 8.25 (b, 2H), 7.92 (d, J = 2.3 Hz, 1H), 7.73 (dd, J = 8.6, 2.4 Hz, 1H), 7.53 (d, J = 8.6 Hz, 1H), 1.24 (d, J = 8.6 Hz, 5H). ¹³C NMR (101 MHz, DMSO-d₆): δ 174.84, 167.35, 162.25, 161.46, 153.02, 144.93, 141.35, 133.71, 122.44, 120.91, 118.04, 106.23, 105.94, 33.70 – 31.07 (m), 23.64, 23.58 – 23.11 (m). HRMS: m/z (M+H)⁺ Calcd. for C₁₆H₁₂D₂N₂O₄: 301.1152; Found 301.1157. HPLC: t_R, 6.42, purity 98.3%

4.5. Metabolic stability studies.

The in vitro metabolism study of amlexanox and d₂-amlexanox were performed in human, rat and mice liver microsomes to evaluate the potential of each compound for cytochrome P450-mediated metabolism. The metabolic stability was assessed as described using pooled CD-1 mouse liver microsomes, pooled SD rat liver microsmes and pooled human liver microsomes (purchased from Xeno Tech). 1 μM of each compound was incubated with 0.5 mg/mL liver microsomes and 1.7 mM cofactor β-NADPH in 0.1 M phosphate buffer (pH = 7.4) containing 3.3 mM MgCl₂ at 37°C. The DMSO concentration was less than 0.1% in the final incubation mixture. At 0, 5, 10, 15, 30, 45, and 60 min of incubation, 40 μL of reaction mixture were removed and the reaction stopped immediately by adding 3-fold excess of cold acetonitrile containing 100 ng/mL of internal standard for quantification. The incubation without the addition of NADPH was used as a negative control. Verapamil were incubated similarly as a positive control substrate. The collected fractions were centrifuged at 15000 rpm for 10 min to collect the supernatant for LC−MS/MS analysis, from which the amount of remaining compound was determined. The natural log of the amount of compound remaining was plotted against time to determine the disappearance rate and the half-life of tested compounds. The LC−MS/MS method consisted of a Shimadzu HPLC system and chromatographic separation of tested compound was achieved using a Waters XBridge-C18 column (5 cm × 4.6 mm, 3.5 μm). An AB Sciex QTrap 5500 mass spectrometer equipped with an electrospray ionization source (Applied Biosystems, Toronto, Canada) in the negative-ion multiple reaction monitoring (MRM) mode for detection. The mobile phases were 0.1% formic acid in purified water (A) and 0.1% formic acid in acetonitrile (B). The gradient (B) was held at 10% (0–0.5 min), increased to 95% at 2 min, kept at isocratic 95% B for 1.0 min, and then immediately stepped back down to 10% for a 2 min re-equilibration. Flow rate was set at 1.0 mL/min. Amlexanox, d₂-amlexanox and internal standard 4-ASA were detected on the mass spectrometer by using multiple reaction monitoring transitions 297.1 → 253.2 m/z, 299.2 → 255.1 m/z and 152.0 → 108.0 m/z, respectively. Mass spectrometry parameter optimization was performed using an automated quantitative method provided by the manufacturer.

4.6. Kinase assay.

The assay was performed as described previously.¹⁵ Briefly, 50 nM purified TBK1 was incubated with 5 μM myelin basic protein and varying concentrations of inhibitor (from 10 μM to 0 μM). Reactions were initiated by the addition of 10 μM ATP spiked with [γ−³²P]-ATP and allowed to proceed for 30 min at room temperature. After quenching the reactions, radioactive phosphorylated substrate was quantified via phosphorimaging plates. Results were analyzed with GraphPad Prism 7 and fit to an inhibition dose-response sigmoidal curve with the Hill slope constrained to −1. Assays were repeated four independent times and the results reported as mean ± standard deviation. Statistical significance was determined via Student’s two-tailed T-test (p<0.05).