Thioesters for the in vitro evaluation of agents to image brain cholinesterases

Abstract

Cholinesterases are associated with pathology characteristic of conditions such as Alzheimer’s disease and are therefore, considered targets for neuroimaging. Ester derivatives of N-methylpiperidinol are promising potential imaging agents; however, methodology is lacking for evaluating these compounds in vitro. Here, we report the synthesis and evaluation of a series of N-methylpiperidinyl thioesters, possessing comparable properties to their corresponding esters, which can be directly evaluated for cholinesterase kinetics and histochemical distribution in human brain tissue. N-methylpiperidinyl esters and thioesters were synthesized and they demonstrated comparable cholinesterase kinetics. Furthermore, thioesters were capable, using histochemical method, to visualize cholinesterase activity in human brain tissue. N-methylpiperidinyl thioesters can be rapidly evaluated for cholinesterase kinetics and visualization of enzyme distribution in brain tissue which may facilitate development of cholinesterase imaging agents for application to conditions such as Alzheimer’s disease.

Keywords::

• Butyrylcholinesterase
• acetylcholinesterase
• brain histochemistry
• neuroimaging
• Alzheimer’s disease

Abbreviations
4-[11C]AMP,4-(N-[11C],	=	methyl)piperidinyl acetate
4-[11C]PMP,	=	4-(N-[11C]methyl) piperidinyl propionate
11C-MP4B,	=	4-(N-[11C]methyl) piperidinyl n-butyrate
Aβ,	=	β-amyloid; AChE, acetylcholinesterase
AD,	=	Alzheimer’s disease
AV,	=	anteroventral nucleus
BuChE,	=	butyrylcholinesterase
DAB,	=	3, 3′-diaminobenzidine tetrahydrochloride
Km,	=	affinity constant
MMFF,	=	Merk Molecular Force Field
NFT,	=	neurofibrillary tangles.

Introduction

The cholinergic system is affected in a number of disorders, such as Alzheimer’s disease (AD)^1–3, multiple sclerosis^4,5 and primary brain tumors⁶. In the brain cholinergic system, two enzymes, acetylcholinesterase (AChE, EC 3.1.1.7) and butyrylcholinesterase (BuChE, EC 3.1.1.8), catalyze the hydrolysis of acetylcholine, thereby regulating the actions of this neurotransmitter⁷. Changes in the activity and location of cholinesterases in neurological disorders render these enzymes suitable targets for neuroimaging in the living brain, to facilitate early diagnosis and treatment monitoring. In AD, cholinesterases, particularly BuChE, are associated with characteristic pathological hallmarks of the disease, β-amyloid (Aβ) plaques and neurofibrillary tangles (NFTs)^8–16.

In earlier attempts to realize cholinesterase imaging, several ¹¹C-radiolabelled alkyl esters of N-methylpiperidinol were synthesized as AChE-specific imaging agents. Two such compounds, 4-(N-[¹¹C]methyl)piperidinyl acetate (4-[¹¹C]AMP) and 4-(N-[¹¹C]methyl) piperidinyl propionate (4-[¹¹C]PMP), were examined^17–21 and provided some reproduction of the known brain distributions of AChE^18,21. Specific BuChE visualization was also attempted with 4-(N-[¹¹C]methyl) piperidinyl n-butyrate (¹¹C-MP4B)²² but was unsuccessful²³. Alternate types of BuChE imaging ligands, derived from N-methylpiperidinol and N-methyl pyrrolidinol, were subsequently synthesized and tested in a normal rat model system²⁴. Using autoradiographic analysis these para-¹²³iodobenzoate esters were found to have entered the brain and their distribution recapitulated, for the most part, the known distribution of BuChE therein. Based on these observations, N-methylpiperidinyl compounds remain promising agents for the imaging of cholinesterases in the living brain.

In vivo testing of the N-methylpiperidinyl cholinesterase imaging agents, 4-[¹¹C]AMP, 4-[¹¹C]PMP and ¹¹C-MP4B, has been extensive. However, methodology for determination of their specificity and direct measurement of their actual affinity constants for AChE and BuChE has been difficult due to the lack of a suitable chromophore that would allow the use of sensitive spectrophotmetric methods. Tracking hydrolysis by radiolabelling reactants and products²⁵ or by employing the chromophore of m-nitrophenol to tag reactants in a spectrophotometric approach²² have both proved cumbersome for screening of compounds as potential imaging agents. Thus, a rapid in vitro evaluation method would be beneficial for the continued development of imaging agents derived from N-methylpiperidinol.

The Ellman spectrophotometric method²⁶, or variations thereof²⁷, remains the most rapid, reliable and sensitive means to measure AChE or BuChE activity. The principle on which this assay is based is that, thioesters, such as acetylthiocholine and butyrylthiocholine, should be kinetically analogous to their corresponding esters with respect to cholinesterase catalysis. A kinetic analysis²⁸ comparing benzoylcholine and benzoylthiocholine supports this notion. In addition to their use in Ellman cholinesterase assays, thioesters are employed for in vitro histochemical detection of cholinesterase activity in tissues. One of the commonly used histochemical techniques for detection of AChE and BuChE is the Karnovsky-Roots method²⁹. In this procedure, hydrolysis of acetylthiocholine and butyrylthiocholine generates a thiolate anion that leads to a precipitate in the area of enzyme activity. This precipitation facilitates detection of enzyme distribution under the microscope. Thus, in theory, any thioester susceptible to cholinesterase hydrolysis could be characterized using Ellman enzyme kinetics and should enable histochemical visualization of areas containing enzyme activity in brain tissue using the Karnovsky-Roots method.

Herein is described the synthesis, enzyme kinetics and histochemical evaluation, as cholinesterase substrates, of alkyl and aryl thioesters of N-methyl-4-piperidinethiol as an indirect means to evaluate potential cholinesterase imaging agents, such as the corresponding N-methyl-4-piperidinol esters.

Materials and methods

Materials and compound characterization

Materials used for synthesis and enzymatic evaluation of compounds are found in Supplementary Material. Compound characterization methods are also presented therein.

Brain tissues

Brain tissues used in this study were provided by the Maritime Brain Tissue Bank, (Halifax, Nova Scotia, Canada), following approval from the Capital Health Ethics Board, from the brain of a 90 year old female, removed within 20 h of death. The brain was immersion fixed in 10% formalin in 0.1 M phosphate buffer (pH 7.4) for approximately 94 h and cut in 1–2 cm slabs. These slabs were cryoprotected by immersion in increasing concentrations of sucrose, ranging from 10% to 40% in 0.1 M phosphate buffer (pH 7.4). The brain tissue was in each concentration of sucrose solution for approximately 48 h and was stored in 40% buffered sucrose (pH 7.4) with 0.6% sodium azide until used.

Synthesis of compounds

Synthesis of N-methylpiperidin-4-yl 4-cyanobenzoate (1)

Under an argon atmosphere, N-methylpiperidin-4-ol (0.92 g, 8.00 mmol) was dissolved in THF (15 mL). To this solution was added 1.6 M butyl lithium in hexanes (5.00 mL, 8.00 mmol) at −78°C followed by a solution of 4-cyanobenzoyl chloride (1.21 g, 7.30 mmol) in THF (20 mL) and the mixture was stirred at −78 °C for 16 h. Water (20 mL) was added to the reaction mixture and extracted with ethyl acetate (3 × 20 mL). The combined organic layers were dried over Na₂SO₄. The solvent was removed in vacuo to produce a yellow solid. The product was purified by silica gel chromatography (10:90 MeOH/CH₂Cl₂) to give a white solid (1.06 g, 59%). The solid was recrystallized from hexanes to afford white crystals. Analytical data: MP _(hexanes): 119–120˚C. IR (Nujol): 2228, 1718, 1278, 1122, 1030 cm⁻¹. ¹H NMR (CDCl₃): δ 1.92–2.03 (m, 2H), 2.10–2.18 (m, 2H), 2.44–2.56 (m, 2H), 2.42 (s, 3H), 2.72–2.87 (m, 2H), 5.08–5.19 (m, 1H), 7.77–7.79 (m, 2H), 8.15–8.17 (m, 2H). ¹³C NMR (CDCl₃): δ 30.9 (2), 46.2 (3), 52.9 (2), 71.4 (1), 116.4 (0), 118.0 (0), 130.1 (1), 132.2 (1), 134.5 (0), 164.3 (0). EI-MS m/z: 244 (M⁺, 44%), 130 (22%), 114 (20%), 102 (29%, 98 (80%), 97 (100%), 96 (80%), 82 (32%), 70 (14%), 57 (10%), 55 (29%). HRMS (EI): M⁺ found 244.1216, calcd for C₁₄H₁₆N₂O₂⁺ = 244.1206.

Synthesis of N-methylpiperidin-4-yl 4-fluorobenzoate (2)

Under an argon atmosphere, N-methylpiperidin-4-ol (0.94 g, 8.16 mmol) was dissolved in THF (15 mL). To this solution was added 1.6 M butyl lithium in hexanes (5.00 mL, 8.00 mmol) at −78 °C followed by a 4-fluorobenzoyl chloride (0.88 g, 7.30 mmol) and the mixture was stirred at −78 °C for 16 h. Water (20 mL) was added to the reaction mixture and extracted with ethyl acetate (3 × 20 mL). The combined organic layers were dried over Na₂SO₄. The solvent was removed in vacuo to produce a yellow solid. The product was purified by silica gel chromatography (10:90 MeOH/CH₂Cl₂) to give a pale yellow solid (0.72 g, 42%). The solid was recrystallized from hexanes to afford white crystals. Analytical data: MP _(hexanes): 48–49˚C. IR (Nujol): 1726, 1274, 1114, 852 cm⁻¹. ¹H NMR (CDCl₃): δ 1.83–1.88 (m, 2H), 2.00-2.05 (m, 2H), 2.32 (s, 3H), 2.33–2.40 (m, 2H), 2.63–2.77 (m, 2H), 4.96–5.10 (m, 1H), 7.08–7.13 (m, 2H), 8.04–8.08 (m, 2H). ¹³C-NMR (CDCl₃): δ 30.9 (2), 46.2 (3), 52.9 (2), 70.3 (1), 115.4 (J_CF = 22 Hz, 1) 126.9 (0), 132.0 (J_CF = 10 Hz, 1), 164.7 (J_CF = 254 Hz, 0), 166.7 (0). EI-MS m/z: 237 (MH⁺, 30%), 123 (30%), 114 (15%), 98 (53%), 97 (100%), 96 (82%), 95 (31%), 82 (35%), 75 (12%), 70 (14%), 55 (28%). HRMS (EI): M⁺ found 237.1173, calcd for C₁₃H₁₆FNO₂⁺ = 237.1160.

Synthesis of N-methylpiperidin-4-yl 4-iodobenzoate (3)

Synthesized according to a previously described procedure²⁴ and the analytical data was consistent and purity greater than 98%. Analytical data: MP: 128–130°C (Lit MP = 130°C²⁶). IR (Nujol): 1711, 1585, 1283, 1268, 1118, and 754 cm⁻¹. ¹H NMR (CDCl₃): δ 1.84–1.91 (m, 2H), 2.01–2.08 (m, 2H), 2.31–2.39 (m, 5H), 2.67–2.73 (m, 2H), 5.02–5.08 (m, 1H), 7.75 (d, J = 8.5 Hz, 2H), and 7.81 (d, J = 8.5 Hz, 2H). ¹³C NMR (CDCl₃): δ 31 (2), 46 (3), 53 (2), 70 (1), 100 (1), 130, 131, 138, and 166. EI-MS m/z: 345(M⁺, 16%), 231 (12%), 203 (9%), 114 (11%), 97 (100%), 82 (22%), 70 (7%), 55 (14%). HRMS (EI): M⁺ found 345.0233, calcd for C₁₃H₁₆NO₂I = 345.0226.

Synthesis of N-methyl-4-piperidinethiol

N-Methyl-4-piperidone (52.92 g, 469.7 mmol) was dissolved in isopropanol (200 mL). Hydrogen sulfide was bubbled through the solution using a glass frit at 0°C resulting in the precipitation of white solid, which was collected by suction filtration after 1 hour. The filtrate was again treated with hydrogen sulfide for an additional 30 min and all further white solid was collected by suction filtration for a total of 80.47 g. Sodium borohydride (16.62g, 439.3 mmol) was suspended in isopropanol (150 mL) followed by the slow addition of the white solid at 0˚C. The reaction mixture was returned to room temperature and stirred under an argon atmosphere for 16 h. Excess solvent was removed under reduced pressure to produce a white paste which was partitioned between 150 mL of both water and ether. The aqueous layer was extracted with ether (4 × 50 mL) and the combined organic layers washed with brine (2 × 30 mL) and dried over anhydrous sodium sulfate. The solvent was removed in vacuo to yield a colourless liquid which was distilled (BP: 168–172°C, lit. BP =  62°C at 0.8mm³⁰); to yield the product as a clear colourless oil (21.22 g, 34%). IR (Neat): 3338, 2944, 1278, 1132 cm^-1. ¹H NMR (CDCl₃): δ 1.55 (d, J = 7.0, 1H), 1.64–1.71 (m, 2H), 1.99–2.04 (m, 4H), 2.26 (s, 3H), 2.67–2.80 (m, 3H). ¹³C NMR (CDCl₃): δ 35.5 (1), 36.9 (2), 46.1 (3), 55.2 (2). EI-MS m/z: 131 (M⁺ 32%), 98 (100%), 96 (12%), 70 (18%), 55 (27%). HRMS (ESI): MH⁺ found 132.0841, calcd for C₆H₁₄NS⁺ = 132.0841.

Synthesis of (N-methylpiperidin-4-yl) 4-cyanobenzenecarbothioate (4)

Under an argon atmosphere, triethylamine (0.53 mL, 3.80 mmol) was added to dichloromethane (20 mL), followed by N-methyl-4-piperidinethiol (0.45 mL, 3.80 mmol) and 4-cyanobenzoyl chloride (0.629 g, 3.80 mmol). The reaction mixture was stirred for 24 h at room temperature. The reaction was extracted with saturated NaHCO₃ (3 × 20 mL) and the organic layer dried over Na₂SO₄. The solvent was removed in vacuo to yield a light yellow solid which was recrystallized from hexanes to afford pale yellow crystals (0.24 g, 24%). Analytical data: MP _(hexanes): 152–153°C. IR (Nujol): 2229, 1655, 1208, 860 cm^-1. ¹H NMR (CDCl₃): δ 1.82–1.89 (m, 2H), 2.10–2.15 (m, 2H), 2.25–2.33 (m, 2H), 2.35 (s, 3H), 2.74-2.85 (m, 2H), 3.72-3.82 (m, 1H), 7.74-7.76 (m, 2H), 8.02-8.04 (m, 2H). ¹³C NMR (CDCl₃): δ 32.1 (2), 40.4 (1), 46.3 (3), 55.1 (2), 116.5 (0), 117.8 (0), 127.6 (1), 132.4 (1), 140.3 (0), 190.2 (0). EI-MS m/z: 260 (M⁺, 4%), 130 (54%), 102 (20%), 98 (100%), 97 (81%), 96 (30%), 70 (15%), 55 (22%), 44 (10%), 42 (18%). HRMS (ESI): MH⁺ found 261.1056, calcd for C₁₄H₁₇N₂OS⁺ = 261.1056.

Synthesis of (N-methylpiperidin-4-yl) 4-fluorobenzenecarbothioate (5)

Under an argon atmosphere, triethylamine (0.53 mL, 3.80 mmol) was added to dichloromethane (20 mL), followed by N-methyl-4-piperidinethiol (0.45 mL, 3.80 mmol) and 4-fluorobenzoyl chloride (0.18 mL, 3.80 mmol). The reaction mixture was stirred for 4 d at reflux temperature. The reaction was extracted with saturated NaHCO₃ (3 × 20 mL) and the organic layer dried over Na₂SO₄. The solvent was removed in vacuo to yield a light yellow solid which was purified by silica gel chromatography (7% MeOH in CH₂Cl₂) and recrystallized from hexanes to afford pale yellow crystals (0.11 g, 28%). Analytical data: MP _(hexanes): 66–68°C. IR (Nujol): 1655, 1225, 1202, 1155, 916, 843 cm^-1. ¹H NMR (CDCl₃): δ 1.88–1.95 (m, 2H), 2.11–2.20 (m, 2H), 2.41 (s, 3H), 2.81–2.94 (m, 2H), 3.69–3.81 (m, 1H), 7.10–7.15 (m, 2H), 7.95–7.99 (m, 2H). ¹³C NMR (CDCl₃): δ 32.5 (2), 40.2 (1), 46.6 (3), 55.5 (2), 116.2 (J_CF = 22 Hz, 1), 130.2 (J_CF = 9 Hz, 1), 134.0 (0), 166.4 (J_CF = 255 Hz, 0), 190.4 (0). EI-MS m/z: 253 (M⁺, 3%), 130 (39%), 123 (24%), 98 (82%), 97 (100%), 96 (36%), 95 (24%), 82 (10%), 75 (10%), 70 (17%), 55 (21%), 42 (17%). HRMS (ESI): MH⁺ found 254.1009, calcd for C₁₃H₁₇FNOS⁺ = 254.1009.

Synthesis of (N-methylpiperidin-4-yl) 4-iodobenzenecarbothioate (6)

Under an argon atmosphere, triethylamine (0.53 mL, 3.80 mmol) was added to dichloromethane (20 mL), followed by N-methyl-4-piperidinethiol (0.45 mL, 3.80 mmol) and 4-iodobenzoyl chloride (1.01 g, 3.80 mmol). The reaction mixture was stirred for 24 h at reflux temperature. The reaction was extracted with saturated NaHCO₃ (3 × 20 mL) and the organic layer dried over Na₂SO₄. The solvent was removed in vacuo to yield a light yellow solid which was recrystallized from hexanes to afford pale yellow crystals (0.41 g, 30%). Analytical data: MP _(hexanes): 136–137°C. IR (Nujol): 1650, 1208, 824 cm^-1. ¹H NMR (CDCl₃): δ 1.80–1.87 (m, 2H), 2.09–2.14 (m, 2H), 2.25–2.32 (m, 2H), 2.34 (s, 3H), 2.74–2.83 (m, 2H), 3.67–3.78 (m, 1H), 7.64–7.67 (m, 2H), 7.79–7.82 (m, 2H). ¹³C NMR (CDCl₃): δ 31.9 (2), 39.6 (1), 46.0 (3), 54.9 (2), 100.8 (0), 128.2 (1), 136.2 (0), 137.6 (1), 190.6 (0). EI-MS m/z: 361 (M⁺, 1%), 230 (12%), 130 (30%), 98 (60%), 97 (100%), 96 (22%), 76 (14%), 70 (10%), 55 (12%), 42 (10%). HRMS (ESI): MH⁺ found 362.0070, calcd for C₁₃H₁₇INOS⁺ = 362.0070.

Synthesis of (N-methylpiperidin-4-yl) ethanethioate (7)

N-Methyl-4-piperidinethiol (0.5 mL, 4.2 mmol) was dissolved in dry CH₂Cl₂ (5 mL) under an argon atmosphere. To this was added acetyl chloride (0.90 mL, 13 mmol) and the reaction refluxed for 1.5 h. The solvent was removed in vacuo and the resulting oil dissolved in saturated NaHCO₃ (20 mL). This solution was extracted with CH₂Cl₂ (4 × 20 mL), dried over Na₂SO₄ and the solvent removed in vacuo to produce a clear colourless oil. This oil was purified by silica gel column chromatography (1:19 MeOH/CH₂Cl₂) to produce a clear colourless viscous liquid (0.69 g, 96%). Analytical Data: IR (Neat): 2941, 1686, 1448, 1219, 915 cm^-1. ¹H NMR (CDCl₃): δ 1.58–1.65 (m, 2H), 1.88–1.92 (m, 2H), 2.08–2.18 (m, 2H), 2.21 (s, 3H), 2.24 (s, 3H), 2.59–2.70 (m, 2H), 3.33–3.48 (m, 1H). ¹³C NMR (CDCl₃): δ 30.6 (3), 31.8 (2), 39.3 (1), 46.0 (3), 54.9 (2), 195.2 (0). EI-MS m/z: 173(M⁺, 17%), 130 (52%), 98 (100%), 97 (30%), 96 (33%), 70 (26%), 55 (45%). HRMS (ESI): MH⁺ found 174.0947, calcd for C₈H₁₆NOS⁺ = 174.0947.

Synthesis of (N-methylpiperidin-4-yl) propanethioate (8)

N-Methyl-4-piperidinethiol (0.5 mL, 4.2 mmol) was dissolved in dry CH₂Cl₂ (5 mL) under an argon atmosphere. To this was added propionyl chloride (0.94 mL, 11 mmol) and the reaction refluxed for 2 h. The solvent was removed in vacuo and the resulting oil dissolved in saturated NaHCO₃ (20 mL). This solution was extracted with CH₂Cl₂ (4 × 20 mL), dried over Na₂SO₄ and the solvent removed in vacuo to produce a clear colourless viscous liquid. This oil was purified by silica gel column chromatography (1:19 MeOH/CH₂Cl₂) to produce a clear colourless viscous liquid (0.45 g, 57%). Analytical data: IR (Neat): 2939, 2782, 1689, 1463, 1129, 937 cm^-1. ¹H NMR (CDCl₃): δ 1.19 (t, J = 7.5, 3H), 1.65-1.72 (m, 2H), 1.95–1.98 (m, 2H), 2.13–2.25 (m, 2H), 2.28 (s, 3H), 2.55 (q, J = 7.5, 2H), 2.63–2.83 (m, 2H), 3.41–3.54 (m, 1H). ¹³C NMR (CDCl₃): δ 9.6 (3), 32.1 (2), 37.5 (2), 39.1 (1), 46.2 (3), 55.1 (2), 199.8 (0). EI-MS m/z: 187 (M⁺, 17%), 130 (61%), 98 (100%), 97 (27%), 96 (28%), 70 (17%), 55 (20%). HRMS (ESI): MH⁺ found 188.1104, calcd for C₉H₁₈NOS⁺ = 188.1104

Synthesis (N-methylpiperidin-4-yl) butanethioate (9)

N-Methyl-4-piperidinethiol (0.5 mL, 4.2 mmol) was dissolved in dry CH₂Cl₂ (5 mL) under an argon atmosphere. To this was added butyryl chloride (1.3 mL, 13 mmol) and the reaction refluxed for 1.5 h. The solvent was removed in vacuo and the resulting oil dissolved in saturated NaHCO₃ (20 mL). This solution was extracted with CH₂Cl₂ (4 × 20 mL), dried over Na₂SO₄ and the solvent removed in vacuo to produce a clear colourless viscous liquid. This oil was purified by silica gel column chromatography (1:19 MeOH/ CH₂Cl₂) to produce a clear colourless oil (0.56 g, 67%). Analytical data: IR (Neat): 2938, 2782, 1686, 1447, 1129, 994 cm^-1. ¹H NMR (CDCl₃): δ 0.95 (t, J = 7.3, 3H), 1.62-1.71 (m, 4H), 1.89–2.01 (m, 2H), 2.08–2.22 (m, 2H), 2.25 (s, 3H), 2.49 (t, J = 7.3, 2H), 2.59–2.81 (m, 2H), 3.38–3.53 (m, 1H). ¹³C NMR (CDCl₃): δ 13.3 (3), 19.0 (2), 32.1 (2), 39.1 (1), 45.9 (2), 46.20 (3), 55.1 (2), 198.9 (0). EI-MS m/z: 201 (M⁺, 16%), 130 (65%), 98 (100%), 97 (28%), 96 (26%), 70 (17%), 55 (20%). HRMS (ESI): MH⁺ found 202.1260, calcd for C₁₀H₂₀NOS⁺ = 202.1260.

Enzyme kinetics

Measurement of aryl ester hydrolysis

Cholinesterase specificity and maximum absorbance change during hydrolysis for each benzoyl ester was determined spectrophotometrically, making use of differences in the aryl chromophores of substrates and products determined through repetitive absorbance scans (e.g. Figure 1). Briefly, 15 μL of AChE (2.5 U) or BuChE (5.4 U) dissolved in 0.1% gelatin_(aq), containing 0.01% sodium azide, and 1.44 mL of 0.1 M phosphate buffer (pH 7.4) were placed in a quartz cuvette of 1-cm path length. The reaction was commenced with the addition of 50 μL of 5 mM ester in 50% acetonitrile_(aq). The absorbance was scanned from 200–300 nm every 2 min for a total of 30 min using a Ultrospec 2100 pro UV/Visible Spectrophotometer (Biochrom) with Swift II software (Amersham). The wavelength corresponding to the maximum absorbance change during hydrolysis (235 nm for compounds 1-3) was used for the subsequent determination of BuChE affinity constants (K_m values, Table 1) using Lineweaver-Burk double reciprocal plots.

Table 1.  Affinity constants (K_m) and maximum velocity values (V_max) for N-methylpiperidinyl alkyl thioesters, N-methylpiperidinyl aryl esters and corresponding thioesters with butyrylcholinesterase (BuChE) and acetylcholinesterase (AChE). None of the aryl compounds (1–6) were hydrolyzed by AChE under the conditions used (X).

Compound	Structure	AChE		BuChE
		Km (µM)	Vmax (µM min^−1)	Km (µM)	Vmax (µM min^−1)
1		X	X	438 ± 75	45.0 ± 6.6
2		X	X	284 ± 28	24.7 ± 2.0
3		X	X	123 ± 42	21.0 ± 5.1
4		X	X	167 ± 17	30.7 ± 2.4
5		X	X	173 ± 13	27.6 ± 1.7
6		X	X	11 ± 1	9.7 ± 0.3
7		751 ± 26	18.3 ± 0.2	684 ± 54	16.7 ± 1.3
8		1108 ± 106	21.4 ± 2.2	302 ± 2	20.7 ± 0.1
9		X	X	129 ± 12	7.0 ± 0.3

Figure 1.  Repetitive absorbance scans for N-methylpiperidin-4-yl 4-cyanobenzoate in the presence of butyrylcholinesterase (BuChE) or acetylcholinesterase (AChE). Note change in absorbance over time when the compound was incubated with BuChE (left) reflecting hydrolysis of the compound by this enzyme. No change in absorbance with AChE (right) indicates no hydrolysis by this enzyme. The absorbance was measured every 2 min for 30 min.

Determination of affinity constants (K_m) and maximum velocity values (V_max) was accomplished by measuring change in absorbance per min (ΔA/min) at 235 nm, using a fixed amount of enzyme AChE (8.3 U) or BuChE (5.9 U) and varying amounts of compounds (1.67 × 10⁻⁴− 1.67 × 10⁻⁵ M), using a Spectronic 1001 (Milton Roy) UV-visible spectrophotometer. The plot of 1/v against 1/s gave K_m as the negative reciprocal of the intercept on the 1/s-axis and gave V_max as the reciprocal of the intercept of the 1/v-axis. As defined previously³¹, 0.1 U is the amount of cholinesterase that gives a ΔA/min of 1.0 in the presence of 1.6 × 10⁻⁴ M substrate (acetylthiocholine for AChE, butyrylthiocholine for BuChE).

Measurement of thioester hydrolysis

The kinetics of both alkyl and aryl thioester hydrolysis by AChE or BuChE was determined using a modification²⁷ of the method described by Ellman et al.²⁶. Briefly, 1.40 mL of buffered 5, 5′-dithio-bis(2-nitrobenzoic acid) solution (pH 7.4) and 0.05 mL of AChE (8.3 units) or BuChE (5.9 units), in 0.1% aqueous gelatin, were mixed in a quartz cuvette of 1-cm path-length and zeroed at 412 nm. The reaction was initiated by the addition of thioester in a 50% aqueous acetonitrile solution to give a final substrate concentration from 1.7 × 10⁻⁴ to 3.3 × 10⁻⁶ M. The reactions were performed at 23˚C. The rate of change of absorbance (ΔA/min), reflecting the rate of hydrolysis of the thioester, was recorded every 5 s for 1 min, using a Milton-Roy 1201 UV–vis spectrophotometer. The molar extinction coefficient for the Ellman product, 5-thio-2-nitrobenzoic acid, used to convert the change in absorbance at λ = 412 nm to moles of product was 14,150 M⁻¹ cm⁻¹. These experiments were performed at least in triplicate and the values averaged.

Structural analysis

Computational chemistry studies, to determine the most stable geometries of compounds, were carried out at the molecular mechanics level of theory using the Merck Molecular Force Field (MMFF), employing Spartan’06³².

Cholinesterase histochemistry with thioester substrates

Brain tissue containing the thalamus was cut on a Leica SM2000R microtome with a Physitemp freezing stage in 50 μm thick coronal sections. Sections were stored in 40% sucrose in 0.1 M phosphate buffer (pH 7.4) at −20˚C until used. A modified³³ Karnovsky-Roots method was employed for AChE and BuChE staining. Substrates used to provide staining were synthesized thioesters and the standards, butyrylthiocholine and acetylthiocholine in the absence of cholinesterase inhibitors. Briefly, tissue sections were rinsed in 0.1 M maleate buffer (pH 7.4) for 30 min and then placed in 0.1 M maleate buffer, pH 7.4, containing 0.15% hydrogen peroxide for 30 min to quench endogenous peroxidase activity. Following a second rinse in 0.1 M maleate buffer, pH 7.4, for 30 min, sections were incubated for 1 h (acetylthiocholine, [N-methylpiperidin-4-yl] ethanethioate (7)) or 18 h (butyrylthiocholine, [N-methylpiperidin-4-yl] butanethioate (9) and [N-methylpiperidin-4-yl] 4-cyanobenzenecarbothioate (4)) in a medium containing 0.5 mM sodium citrate, 0.47 mM cupric sulfate, 0.05 mM potassium ferricyanide and thioester substrate in 0.1 M maleate buffer (pH 7.4). Substrate concentrations were 4 mM for acetylthiocholine, butyrylthiocholine, (N-methylpiperidin-4-yl) ethanethioate, (N-methylpiperidin-4-yl) butanethioate and 0.5 mM for the less soluble (N-methylpiperidin-4-yl) 4-cyanobenzenecarbothioate. Sections were then rinsed for 30 min in distilled water and placed in 0.1% cobalt (II) chloride in water for 10 min. After a further rinse in distilled water, sections were placed in a 3, 3′-diaminobenzidine tetrahydrochloride (DAB) solution consisting of 1.39 mM in 0.1M phosphate buffer (pH 7.4). After 5 min, a solution of 0.15% hydrogen peroxide in distilled water was added at a ratio of 10:1 (DAB solution: hydrogen peroxide solution) and the reaction was carried out for approximately 2 min. Sections were then washed in 0.01 M acetate buffer, pH 3.3, mounted on slides, coverslipped and examined with brightfield microscopy.

Sections were photographed with an AxioCam HRc camera on a Zeiss Axioplan II microscope. The photographic plates were assembled using Adobe Photoshop 7.0. The images were colour balanced, contrast enhanced and brightness adjusted to match the background from different images.

Results and discussion

Organic synthesis

To synthesize N-methyl-4-piperidinethiol, N-methyl-4-piperidone was mixed with hydrogen sulfide and the resulting reaction produced a dithiol. This intermediate was then reduced with sodium borohydride (Scheme 1), in a reaction adapted from an earlier procedure³⁴. Thioesters were subsequently generated by esterification of this thiol with an acid chloride (Scheme 1). Benzoate esters were also synthesized, as previously described²⁴, (Scheme 1, Table 1, 1–3) in order to directly compare cholinesterase kinetic parameters for aryl thioesters and esters. Two classes of thioesters were synthesized, those from aryl acid chlorides (Table 1, 4–6), analogous to the benzoate esters, and those from alkyl acid chlorides (Table 1, 7–9), analogous to previously developed alkyl ester imaging agents^17,22.

Scheme 1.  Reaction scheme for the synthesis of (a) alkyl and aryl thioesters and (b) aryl esters.

Kinetic analysis

Thioester analogues of aryl esters such as benzoylcholine have been found²⁸ to be kinetically comparable to their ester counterparts for cholinesterase hydrolysis. Therefore, thioesters can provide a rapid, indirect enzyme kinetic screening method for analysis of potential ester imaging agents. All aryl and alkyl thioesters, producing a thiol product, upon hydrolysis, are amenable to kinetic analysis as cholinesterase substrates via an Ellman assay. Furthermore, since all aryl esters showed distinct UV spectral differences between reactant and product (Figure 1), they could be analyzed for comparison with thioester kinetic parameters by an alternate method employing changes in absorbance. Esters and thioesters of substituted benzoic acids gave comparable affinity constants and maximum velocity values (Table 1), in agreement with the earlier observation comparing benzoylcholine and benzoylthiocholine analogues²⁸. In general, the thioester derivatives had a slightly greater affinity towards BuChE than did the esters. In addition, none of the aryl thioesters, as observed earlier for aryl esters²⁴, were susceptible to hydrolysis by AChE (Figure 1).

Although it was not possible to directly obtain affinity constants for the esters of N-methylpiperidinol derived from acetic acid, propionic acid and butyric acid, it is reasonable to expect, from comparable studies with aryl derivatives (Table 1,²⁸), these values to be similar to those for the comparable thioesters (Table 1). This is supported by the observation that for AChE, like the analogous thiocholine esters, the acetyl ester of N-methyl-4-piperidinethiol (7, Table 1) had the greatest affinity, followed by the propionyl ester (8), while the butyryl derivative (9) was not hydrolyzed by AChE under these conditions. In addition, BuChE affinity increased with longer substrate side chains, consistent with the known preference of this enzyme for longer acyl chains^27,35. The observation that the thioester analogue of MP4B is BuChE-specific is also in keeping with the earlier report²⁵ that the ester is only hydrolyzed by BuChE. Of note is that the affinities of the acetyl (7) and propionyl (8) thioesters for BuChE are greater than for AChE. Even though the corresponding radiolabelled esters, 4-[¹¹C]AMP and 4-[¹¹C]PMP, have been developed as AChE imaging agents, these kinetic results suggest that BuChE is also capable of efficiently hydrolyzing these compounds and thus, both AChE and BuChE may be detected in molecular imaging studies with these agents, rendering them to be non-specific.

Structural analysis

The preferred geometries of the synthetic thioesters and their ester counterparts were computed at the molecular mechanics level in order to further explore any subtle differences observed in the cholinesterase kinetic affinities. In general, the preferred conformations for each oxygen and sulfur analogous pair were comparable (Figure S1). However, slight differences in molecular shape, because of the presence of the bulkier and electronically distinct sulfur atom, most likely contributed to observed differences in relative affinity constants for BuChE (Table 1) in each analogous pair. The structural similarity further suggests that thioesters are acceptable kinetic surrogates for corresponding ester imaging agents.

Histochemical evaluation of N-methyl piperidinyl thioesters

Distribution of an imaging agent within the living brain is dependent upon a number of factors. These include cerebral blood flow, penetration of the blood–brain barrier, interaction with the target molecules and fate of the metabolites. Thus, determining the in vitro distribution of the imaging agent in brain tissues would be valuable before advancing to in vivo radioimaging studies. The cholinesterases are commonly stained in brain tissues using thioesters such as butyrylthiocholine or acetylthiocholine^29,36. In order to explore cholinesterase specificity and to determine their potential to detect the distribution of cholinesterases in the brain, several of the thioesters prepared in this study were tested as histochemical substrates and compared with choline thioesters for staining human thalamus brain tissue (Figure 2). This region of the brain was chosen because it has been precisely mapped for cholinesterase activity and certain nuclei, such as the anteroventral (AV) nucleus (Figure 2A), exhibit distinct patterns of AChE and BuChE activity³⁶. In the AV nucleus of the thalamus, AChE activity is found only in the neuropil while BuChE activity is observed only in neurons. As can be seen in Figure 2, acetylthiocholine (Figure 2B) and N-methylpiperidinyl acetylthioester (7; Table 1; Figure 2C) show comparable neuropil staining typical of AChE activity in this nucleus. That is, both the choline and N-methylpiperidinethiol acetyl thioester substrates provided histochemical staining consistent with the previously described distribution of AChE in this region of the brain³⁶. This observation is also in agreement with the in vitro enzyme kinetic results indicating that AChE is capable of efficiently hydrolyzing the N-methylpiperidinyl acetylthioester (Table 1). N-methylpiperidinyl acetylthioester (7) is hydrolyzed by both AChE and BuChE (Table 1), similar to acetylthiocholine³⁷, using Ellman method. However, the modified Karnovsky-Roots method for histochemical staining requires shorter (1 h) incubation for staining AChE and longer (18 h) for BuChE. Based on this difference in the incubation time, staining in Figure 2C is predominantly for AChE. Similarly, to evaluate specific histochemical visualization of BuChE activity in the human thalamus, comparative staining with butyrylthiocholine, N-methylpiperidinyl butyrylthioester (9; Table 1) and cyanobenzoate thioester (4; Table 1) was carried out (Figure 2D, E, F). BuChE staining using butyrylthiocholine as substrate (Figure 2D) was comparable to that obtained with the butyryl (Figure 2E) and 4-cyanobenzoyl (Figure 2F) derivatives of N-methylpiperidinethiol. Furthermore, the BuChE staining profiles (Figure 2D, E, F) were distinct from those for AChE observed with the acetyl thioesters (Figure 2B and C) in the same region of the brain. High (>1 mM) thioester concentrations are typically required in the staining medium to achieve adequate visualization of cholinesterase activity. Because of this, solubility limitations prohibited use of the iodo- and fluorobenzoate thioesters in similar histochemical analysis to that provided by the cyanobenzoate derivative (Figure 2F). However, successful histochemical recapitulation of enzyme activity with an aryl thioester provides a critical proof-of-principle that analogous radiolabelled compounds are potential imaging agents to be highly sensitive detectors of cholinesterase activity in brain tissue.

Figure 2.  Histochemical staining of human brain tissue at the level of the thalamus in the coronal plane. (A) Parcellation of the thalamus in the region used to compare histochemical staining by various cholinesterase substrates; (B) Acetylthiocholine. (C) (N-methylpiperidin-4-yl) ethanethioate. Note that both of these substrates produced a similar pattern of staining recapitulating the known distribution of acetylcholinesterase in this region. (D) Butyrylthiocholine. (E) (N-methylpiperidin-4-yl) butanethioate. (F) (N-methylpiperidin-4-yl) 4-cyanobenzenecarbothioate. Note that these substrates produced a similar pattern of distribution that reflected the known distribution of butyrylcholinesterase in this region. AV: anteroventral nucleus; CL: central lateral nucleus; MD: mediodorsal nucleus; R: reticular nucleus; V: ventral nuclei. Scale bar = 200 µm.

Conclusion

This study indicates that N-methylpiperidinyl thioesters are effective surrogate substrates for the evaluation and development of corresponding ester compounds as AChE and BuChE imaging agents. These thioester analogues have comparable affinities towards AChE and BuChE to their ester counterparts. These similarities permit screening of corresponding ester substrates as potential radioligands for specific detection of AChE or BuChE activity. These thioester compounds can also be employed for histochemical detection of enzyme activity in brain tissue, which could then be used for comparison with the in vivo images generated by the analogous radiolabelled molecules. Development of compounds, screened in this manner, could facilitate discovery of suitable molecular imaging agents for the detection of AChE and BuChE in the living brain for application to neurological disorders involving the cholinergic system, such as AD.

Acknowledgements

The authors thank Eric Joy for technical assistance.

Declaration of interest

Portions of this work were supported by grants from Canadian Institutes of Health Research, Canada Foundation for Innovation, Natural Sciences and Engineering Research Council of Canada, Vascular Health and Dementia Initiative (through Canadian Institutes of Health Research, Heart & Stroke Foundation of Canada, the Alzheimer Society of Canada and Pfizer Canada Inc.), Capital District Health Authority Research Fund, Nova Scotia Health Research Foundation, Brain Tumour Foundation of Canada, Multiple Sclerosis Society of Canada, National Institutes of Health, Dalhousie Medical Research Foundation, Faculty of Medicine of Dalhousie University and the Committee on Research and Publications of Mount Saint Vincent University.