Effect of the 3- and 4-methyl groups on the opioid receptor properties of N-substituted trans-3,4-dimethyl-4-(3-hydroxyphenyl)piperidines.

N-substituted trans-3,4-dimethyl-4-(3-hydroxyphenyl)piperidines (2a,b) are opioid receptor antagonists where the antagonist properties are not due to the type of N-substituent. In order to gain a better understanding of the contribution that the 3- and 4-methyl groups make to the pure antagonist properties of 2a,b, we synthesized analogues of 2a,b that lacked the 4-methyl (5a,b), 3-methyl (6a,b), and both the 3- and 4-methyl group (7a,b) and compared their opioid receptor properties. We found that (1) all N-methyl and N-phenylpropyl substituted compounds were nonselective opioid antagonists (2) all N-phenylpropyl analogues were more potent than their N-methyl counterparts, and (3) compounds 2a,b which have both a 3- and 4-methyl substituent, were more potent antagonists than analogues 5a,b, 6a,b, and 7a,b. We also found that the removal of 3-methyl substituent of N-methyl and N-phenylpropyl 3-methyl-4-(3-hydroxyphenyl)piperazines (8a,b) gives (4a,b), which are opioid antagonists.

Introduction

In many classes of opioid compounds, small changes in structure modify the extent to which the opioid ligand exhibits agonist and/or antagonist properties.[1,2] In the morphine family and in fused-ring opioids such as the 4,5-epoxymorphinon-3-one,[3] morphinan,[4] benzomorphan,[5] and isoquinoline[6] series, N-substituent variation modulates relative agonist/antagonist potency. The N-methyl analogues are almost always pure agonist, while the N-allyl and N-cyclopropylmethyl analogues usually have antagonist properties. For example, naloxone (1a) and naltrexone (1b) are pure opioid receptor antagonists (Chart 1).[1] In contrast to these polycyclic structures, pure opioid receptor antagonist properties of N-substituted trans-3,4-dimethyl-4-(3-hydroxyphenyl)piperidines (2) (Chart 1) discovered by Zimmerman and co-workers[7,8] were a consequence of substitution at the 3-position on the piperidine ring rather than substitution at the nitrogen. A number of subsequent SAR studies have shown that all N-substituted analogues of 2, including the N-methyl (2a) analogue, are pure opioid receptor antagonists.[7−9] The N-substituent on 2 apparently only affects the antagonist potency and opioid receptor selectivity.[9−14] In other SAR studies, Zimmerman reported that replacing the 4-methyl group with a larger substituent led to compounds with both opioid receptor agonist and antagonist properties. For example, N-methyl-trans-3-methyl-4-propyl-4-(3-hydroxylphenyl)piperidine (3) (Chart 1) showed both agonist and antagonist effects.[8] In contrast to numerous studies on the opioid antagonist properties of the N-substituted trans-3,4-dimethyl-4-(3-hydroxyphenyl)piperidines 2,[7−14] the opioid receptor properties of N-substituted 4-(3-hydroxyphenyl)piperidines lacking a 4-methyl and/or a 3-methyl group on the piperidine ring have received little, if any, study. We recently reported that the N-methyl- and N-phenylpropyl-(S)-(3)-methyl-4-(3-hydroxyphenyl)piperazines 4a and 4b (Chart 1), respectively, which do not have a 4-methyl substituent, were nonselective, potent, pure opioid antagonists.[15] These results suggested that N-methyl- and N-phenylpropyl-3-methyl-4-(3-hydroxyphenyl)piperidines 5a and 5b could also be pure opioid receptor antagonists (Chart 1). In order to gain a better understanding of the contributions that 3- and 4-methyl groups make to the pure antagonist properties of 4-(3-hydroxyphenyl)piperidine and the 3-methyl group to the antagonist properties of 4-(3-hydroxyphenyl)piperazine compounds, we synthesized 2a,b, 4a,b, 5a,b, 6a,b, 7a,b, and 8a,b (Chart 1) and compared their opioid receptor properties as determined by the in vitro [35S]GTPγS functional assay. We found that (1) with the exception of 6a, all compounds studied were nonselective pure opioid antagonists at the μ, κ, and δ receptors or inactive, and even compound 6a was only a very weak agonist at the δ receptor and an antagonist at the μ and κ receptors; (2) similar to 2b, all N-phenylpropyl analogues were more potent antagonists than their N-methyl counterparts; and (3) analogues 2a,b, which have both a 3- and a 4-methyl substituent, were more potent antagonists than analogues that lacked a 3-, 4-, or both methyl substituents.

Chart 1

Structures of Compounds 1a,b, 2a,b, 3, 4a,b, 5a,b, 6a,b, 7a,b, 8a,b, and 22

Chemistry

Compounds 5a and 5b were prepared from a common intermediate (9)[16] (Scheme 1). Following deprotection and dehydration with refluxing hydrochloric acid in glacial acetic acid, the intermediate from 9 was reduced via catalytic hydrogenation to afford 5a as a mixture of geometric isomers. Careful chromatography and trituration of the resulting oil afforded a solid that proved to be the pure cis-isomer as determined by 1H NMR spectral analysis. In order to prepare the secondary amine 10, the dehydrated, hydrogenated intermediate from 9 was treated with 1-chloroethyl chloroformate (ACE-Cl) in refluxing chloroform and then refluxed in ethanol to decompose the intermediate chloroethyl carbamate. Reductive amination of 10 with 3-phenylpropanal afforded 11, which was deprotected with boron trichloride to afford 5b as a mixture of isomers. The major, cis-isomer (as determined by 1H NMR spectral analysis), selectively crystallized as the tosylate salt from isopropanol and diethyl ether.

Scheme 1

Compound 6b was prepared according to Scheme 2. The aryllithium reagent prepared from 3-bromoisopropoxybenzene (12)[16] was added to N-methyl-4-piperidinone. The resulting alcohol 13 was dehydrated with p-toluenesulfonic acid in refluxing toluene to afford the tetrahydropyridine 14. Deprotonation with n-butyllithium gave the blood-red anion which was quenched with dimethylsulfate. The resulting enamine was reduced with sodium borohydride to afford 15. As with the preparation of 5b, reaction with ACE-Cl was followed by refluxing 5b in methanol to afford the secondary amine 16. Reductive amination of 16 with 3-phenylpropanal afforded 17, which was treated with boron trichloride in methylene chloride to yield 6b.

Scheme 2

The preparation of 7b is shown in Scheme 3. Catalytic hydrogenation of the readily available intermediate 18(17) reduced both the olefin and cleaved the N-benzyl group, affording an intermediate of acceptable purity for reductive amination with 3-phenylpropanal to give 19. Intermediate 19 was deprotected with boron tribromide in methylene chloride to afford 7b.

Scheme 3

As shown in Scheme 4, the N-arylation[15] of N-methylpiperazine 20 with 3-bromoanisole afforded 21. Deprotection with boron tribromide gave compound 8a.

Scheme 4

Biology

All compounds were initially screened for intrinsic and antagonist activity at 10 μM in the [35S]GTPγS binding assay at the human μ, κ, and δ opioid receptors overexpressed in CHO cells. Compounds identified as agonists were evaluated in opioid receptor-appropriate assay using eight different concentrations selected to provide clear indication of the upper and lower asymptotes of the concentration–response curve. The Emax and EC50 were calculated, and the Emax was reported as a percentage of the Emax of the agonist standard (DAMGO, μ; DPDPE, δ) run on the same assay plate. Measures of functional antagonism and selectivity were obtained by measuring the ability of test compounds to inhibit stimulated [35S]GTPγS binding produced by the selective agonist DAMGO (μ), DPDPE (δ), or U69,593 (κ). Agonist concentration–response curves were run in the presence or absence of a single concentration of test compound. At least two different concentrations of test compound were used in these experiments, and these had to cause a minimum 4-fold shift in the agonist EC50 before a Ke was calculated. The Ke values were calculated using the formula Ke = [L]/(DR – 1), where [L] is the concentration of test compound and DR is the ratio of agonist EC50 value in the presence or absence of test compound.

Results and Discussion

The N-substituted 3,4-dimethyl-4-(3-hydroxyphenyl)piperidines (2) are a class of pure opioid antagonists discovered by Zimmerman and co-workers.[7−9] In 1978, Zimmerman reported that the addition of a trans-3-methyl group to the opioid agonist 1,4-dimethyl-4-(3-hydroxyphenyl)piperidine (6a) gave trans-N-methyl-3,4-dimethyl-4-(3-hydroxyphenyl)piperidine (2a), which was a pure opioid antagonist. The discovery of the structurally unique, pure antagonist 2a was highly interesting, since prior to this discovery, all pure opioid antagonists were N-allyl or N-cyclopropylmethyl analogues of opioid agonists such as naloxone (1a) and naltrexone (1b).[3]

Resolution of the N-substituted trans-3,4-dimethyl-4-(3-hydroxyphenyl)piperidines (2) showed that both the (3R,4R)- and (3S,4S)-isomers were pure opioid antagonists with the (3R,4R)-isomer being a more potent antagonist than the (3S,4S)-isomer.[9] However, even small alterations of the trans-3,4-dimethyl-4-(3-hydroxyphenyl)piperidine structure imparted opioid agonist activity to the molecule in animal antinociceptive tests. For example, cis-N-methyl-3,4-dimethyl-4-(3-hydroxyphenyl)piperidine (22) (Chart 1) has mixed agonist–antagonist properties. In addition, when the 4-methyl group of 2a was replaced with a 4-propyl group, the resulting trans-N-methyl-3-methyl-4-propyl-4-(3-hydroxypiperidine) (3) was a mixed agonist–antagonist.[8] To our knowledge no systematic study using the same assay has been conducted on the N-substituted 3,4-dimethyl-4-(3-hydroxyphenyl)piperidine class of opioid ligands having the 3-methyl-, 4-methyl-, or both methyl substituents removed to determine the effect that the contribution of the 3- and 4-methyl substituents of the piperidine ring has on antagonist properties at each of the opioid receptors.

In this study, we compared the opioid receptor properties of 4a,b and 5a,b which have only a 3-methyl group, 6a,b which have only a 4-methyl group, and 7a,b and 8a,b which have neither methyl group to those of 2a,b which have both the 3- and 4-methyl groups present. Each compound was tested for its opioid receptor agonist and antagonist properties using the [35S]GTPγS binding assay to gain information about the contribution of the 3- and 4-methyl substituents on the piperidine ring of 2a,b to the pure opioid antagonist properties of these compounds. Zimmerman and co-workers reported that 2a had an AD50 of 0.74 mg/kg as an antagonist of morphine-induced antinociception at the μ receptor and an AD50 of >5.0 mg/kg of κ agonist U50,488-induced antinociception in the mouse writhing test.[9] Zimmerman and co-workers had previously reported that 2a did not have any opioid receptor agonist activity in the mouse writhing test and the rat tail heat analgesic test.[7,8] We found that 2a had Ke values of 29.3, 681, and 134 nM at the μ, δ, and κ opioid receptors, respectively, with no opioid agonist efficacy at 10 μM at all three opioid receptors in a [35S]GTPγS assay. In a follow-up study to the report by McElvain and Clemens[18] that 6a was a morphine-like opioid agonist, Zimmerman and co-workers reported that 6a had ED50 = 3.4 mg/kg in the mouse writhing test.[7] We found that 6a was a weak agonist at the δ opioid receptor with ED50 = 8300 nM and Emax = 64% and was an antagonist at the μ and κ receptor with Ke values of 974 and 477 nM, respectively. In our study, we even found that the N-methyl analogue 7a, which does not have a 3- or 4-methyl, was still a pure but weak opioid antagonist with Ke value of 2700 nM at the κ receptor, was inactive at the μ and δ receptors, and had no agonist activity at 10 μM at all three receptors.

Zimmerman and co-workers reported that changing the N-methyl group in 2a to the N-phenylpropyl group present in 2b resulted in a much more potent antagonist. Compound 2b had an AD50 value of 0.28 mg/kg for μ antagonism of morphine-induced antinociception in the mouse writhing test compared to 0.74 mg/kg for 2a in the test.[9] Similar to 2a,b, we also found that changing the N-methyl group in 5a, 6a, and 7a to give the N-phenylpropyl analogues 5b, 6b, and 7b resulted in greatly increased antagonist potency, particularly at the μ and κ receptors with no agonist efficacy in the [35S]GTPγS binding assay for all three compounds. Compound 5b had Ke values of 11.9 and 16.5 nM at the μ and κ receptors, respectively, and was 105- and 79-fold more potent than 5a as an antagonist at the μ and κ receptors (Table 1). Compound 6b with Ke values of 5.5 and 21.3 nM at the μ and κ receptors, respectively, was 177 and 22 times more potent than 6a as an antagonist at the μ and κ receptors. Unlike 6a, which is an agonist at the δ receptor, 6b is an antagonist at the δ receptor with Ke = 178 nM (Table 1). Compound 7a is a weak antagonist at the κ receptor and has no antagonist properties at the μ and δ receptors. In contrast to the lack of antagonist properties of 7a at the μ and δ receptors, 7b is a weak antagonist at the μ and δ receptors with Ke values of 75.6 and 418 nM. Surprisingly, 7b, with Ke = 5.8 nM at the κ receptor, showed 13- and 72-fold selectivity for the κ relative to the μ and δ opioid receptors, respectively.

Table 1

Inhibition of Agonist-Stimulated [35S]GTPγS Binding in Cloned Human μ, δ, and κ Opioid Receptorsa

		Ke (nM)
compd	R	μ, DAMGO	δ, DPDPE	κ, U69,593
2a	CH3	29.3 ± 3.4b	681 ± 241b	134 ± 27.1
2b	C6H5(CH2)3	0.1 ± 0.02b	0.9 ± 0.32b	0.88 ± 0.17
4a	CH3	508 ± 26	IAd	194 ± 32
4b	C6H5(CH2)3	0.88 ± 0.03	13.4 ± 4.2	4.09 ± 0.79
5a	CH3	1248 ± 423	IAd	1307 ± 447
5b	C6H5(CH2)3	11.9 ± 2	46 ± 18	16.5 ± 5
6a	CH3	974 ± 230	agonistc	477 ± 150
6b	C6H5(CH2)3	5.5 ± 0.59	178 ± 47	21.3 ± 9.0
7a	CH3	IAd	IAd	2700 ± 1300
7b	C6H5(CH2)3	75.6 ± 21	418 ± 120	5.8 ± 2.0
8a	CH3	4300 ± 1700	1600 ± 480	IAd
8b	C6H5(CH2)3	8.47 ± 1.42	34.3 ± 5.8	36.8 ± 16.8

With the exception of 6a, all compounds had no agonist activity at 10 μM.

Data taken from ref (15)

Compound 6a had ED50 = 8300 ± 2500 nM with Emax = 64% ± 5% of DPDPE max.

Compounds that at 10 μM caused less than a 4× shift in the agonist EC50 were considered inactive (IA). The data represent the mean ± SE from at least three independent experiments.

In a previous study, we reported that both the N-methyl- and N-phenylpropyl-3-methylpiperazine analogues 4a and 4b were potent, pure opioid receptor antagonists.[15] The N-methyl compound 4a has Ke values of 508 and 194 at the μ and κ receptors, respectively, and no antagonism at the δ receptor, whereas the N-phenylpropyl 4b has Ke values of 0.88, 13.4, and 4.09 nM at the μ, δ, and κ receptors, respectively (Table 1). In this study, we found that N-methyl- and N-phenylpropyl-4-(3-hydroxyphenyl)piperidines (8a,b), which like 7a,b do not have a 3- or 4-methyl substituent, were pure opioid antagonists. Compound 8b, with Ke values of 8.47, 34.3, and 36.8 nM at the μ, δ, and κ opioid receptors, was 507- and 47-fold more potent at the μ and δ opioid receptors than 8a. Compounds 4a,b and 8a,b had no opioid agonist efficacy at 10 μM.

Previous 1H NMR and single crystal X-ray structural studies have suggested an equatorial orientation for the 4-(3-hydroxyphenyl) group in the N-substituted trans-3,4-dimethyl-4-(3-hydroxyphenyl)piperidine (2) class of opioid antagonists.[19,20] The pure opioid receptor antagonists’ activity of this class of compounds has been attributed to the compounds having the 4-(3-hydroxyphenyl) group in the equatorial orientation. The fact that compounds 5a,b, 6a,b, 7a,b, and 8a,b are all opioid receptor antagonists suggests that the 4-(3-hydroxyphenyl) group in each of these compounds is in an orientation similar to that found in the N-substituted 3,4-dimethyl-4-(3-hydroxyphenyl)piperidine class of opioid antagonists.

In summary, a study of the in vitro functional opioid agonist–antagonist properties of the N-substituted 4-(3-hydroxyphenyl)piperidines 2a,b, 5a,b, 6a,b, and 7a,b revealed that all these N-substituted 4-(3-hydroxyphenyl)piperidines were opioid receptor antagonists at the μ and κ receptors. In addition, all compounds with the exception of the 1,4-dimethyl-4-(3-hydroxyphenyl)piperidine (6a) were also antagonists at the δ opioid receptor. The very low δ opioid receptor efficacy of ED50 = 8300 nM and antagonist properties at the μ and κ receptors suggest that the analgesic activity seen in the animal studies reported by McElvain and Clemens[18] and Zimmerman and co-workers[7] for 6a might be due to the interaction with a target other than the opioid receptors or metabolism to an active metabolite. These studies show that neither the 3-methyl nor 4-methyl substituents on the piperidine rings of 2a,b are required to obtain potent μ and κ opioid receptor antagonists.

The study of the in vitro functional efficacy properties of the N-methyl- and N-phenylpropyl-4-(3-hydroxyphenyl)piperazines (8a and 8b, respectively) revealed that both compounds were opioid receptor antagonists at the μ, δ, and κ receptors. Thus, the 3-methyl substituent present in 4a and 4b is not needed to obtain pure opioid antagonists in the N-substituted 4-(3- hydroxyphenyl)piperazine class of compounds.

Even though the 3- and 4-methyl substituents of 2a,b are not required for obtaining pure opioid receptor antagonism, the presence of these two methyl substituents does increase the potency as an opioid receptor antagonist relative to the potency of 5a,b, 6a,b, and 7a,b. However, compounds 5a,b, 6a,b, and 7a,b, particularly 6a,b and 7a,b which are not chiral, are much simpler to synthesize than 2a,b. Compounds 4a,b, which have a 3-methyl substituent on the piperazine ring, are also more potent than 8a,b which lack a 3-methyl substituent. Since modification of the N-substituent in 2 led to alvimopam, a drug on the market for the treatment of GI motility disorders; JDTic, a potent and selective κ opioid receptor antagonist that is active in several animal models of CNS disorder; and LY255582, which was developed to treat obesity; it will be interesting to see if additional structural modifications of 5a,b, 6a,b, and 7a,b as well as 4a,b and 8a,b will lead to new opioid receptor antagonists with potential for drug development.

Experimental Section

General Procedures

All solvents were dried prior to use according to known procedures; all reagents were obtained from commercial sources or were synthesized from literature procedures and were used without further purification unless otherwise noted. Air-sensitive reactions were performed under slight positive pressure of nitrogen. Room temperature is assumed to be between 20 and 25 °C. Evaporation of solvents was accomplished under reduced pressure at less than 40 °C, unless otherwise noted. Melting points were taken on a Mel-Temp apparatus and are uncorrected. Chromatography solvent systems are expressed in v:v ratios or as % v. CMA80 refers to a solution of CHCl3–MeOH NH4OH−aq (80:18:2). Thin layer chromatography was performed on aluminum oxide IB-F plated from J. T. Baker (Phillipsburg, NJ) or silica gel 60 F254 plates from EMD (Gibbstown, NJ). Chromatograms were visualized under UV light at 254 nm. 1H NMR spectra were obtained at 300 MHz on a Bruker DPX300 spectrometer. 13C NMR spectra were obtained at 75 MHz on a Bruker DPX300 spectrometer. Chemical shift values for 1H were determined relative to an internal tetramethylsilane standard (0.00 ppm). Chemical shift values for 13C were determined relative to solvent (CDCl3 = 77.23 ppm). Elemental analyses were performed by Atlantic Microlabs, Norcross, GA. Purity of compounds (>95%) was established by elemental analysis.

3-[(3R,4R)-1,3,4-Trimethylpiperidin-4-yl]phenol (2a) and 3-[(3R,4R)-3,4-Dimethyl-1-(3-phenylpropyl)piperidin-4-yl]phenol (2b)

Compounds 2a and 2b were prepared according to a literature method.[9]

3-[(2S)-2,4-Dimethylpiperazin-1-yl]phenol (4a) and 3-[(2S)-2-Methyl-4-(3-phenylpropyl)piperazin-1-yl]phenol (4b)

Compounds 4a and 4b were synthesized as previously reported.[15]

3-(1,3-Dimethylpiperidin-4-yl)phenol (5a) Hydrobromide

A solution of racemic 1,3-dimethyl-4-[3-(propan-2-yloxy)phenyl]-4-piperidinol (9)[16] (4.20 g, 15.9 mmol) was dissolved and refluxed in 12 N HCl (5 mL) and AcOH (10 mL) for 24 h. The solution was concentrated to dryness and then dissolved in dilute NH4OH, extracted with EtOAc, dried (Na2SO4), and concentrated to afford an oil (3.23 g). A portion of the resulting oil (1.62 g) in EtOH (100 mL) with 20% Pd(OH)2 on carbon (0.15 g) was shaken under 50 psi of H2 for 12 h. The solution was filtered through Celite, concentrated to a residue, and subjected to chromatography on silica gel using a gradient up to 50% CMA80 in CH2Cl2. The resulting oil was triturated to a solid with CH2Cl2 and a trace of MeOH, filtered, and washed with cold Et2O to afford 5a (1.24 g, 76% over two steps), which proved to be the pure cis-isomer by NMR analysis. The solid was dissolved and then concentrated from MeOH and 48% HBr to afford the hydrobromide salt: mp 245–247 °C; 1H NMR (DMSO-d6) δ 9.32 (s, 1H), 8.99 (bs, 1H), 7.12 (t, 1H, J = 7.8 Hz), 6.67–6.55 (m, 3H), 3.54, 3.21 (m, 3H), 3.12–2.85 (m, 2H), 2.79 (d, 3H, J = 4.4 Hz), 2.37–2.31 (m, 2H), 1.94–1.76 (m, 1H), 0.76 (d, 3H, J = 7.4 Hz); 13C NMR (DMSO-d6) δ 157.3, 143.6, 129.2, 117.6, 113.9, 113.3, 59.2, 54.2, 43.4, 40.3, 32.6, 21.5, 11.3; MS (ESI) m/z 206.2 (M + H)+. Anal. (C13H20BrNO) C, H, N.

3-[3-Methyl-1-(3-phenylpropyl)piperidin-4-yl]phenol (5b) Tosylate

A solution of 11 (246 mg, 0.70 mmol) in CH2Cl2 (7 mL) in a salt water–ice bath was treated with BCl3 (7 mL, 1 M in CH2Cl2). After 30 min, the solution was washed with saturated NaHCO3 (5 mL). The organic layer was dried (Na2SO4) and concentrated. The residue was purified by chromatography on silica gel using EtOAc and then 20% CH3OH in CHCl3 to obtain 174 mg (80%) of 5b as a colorless oil that proved to be a mixture of the cis and trans geometric isomers (4:1). The tosylate salt of the major, cis isomer crystallized from i-PrOH/Et2O had mp 123–125 °C. 1H NMR (CD3OD) δ 7.71 (d, 2H, J = 8.1 Hz), 7.33–7.16 (m, 7H), 7.13 (t, 1H, J = 7.8 Hz), 6.69–6.59 (m, 3H), 3.62 (d, 1H, J = 12.0 Hz), 3.49 (d, 1H, J = 12.5 Hz), 3.32–3.21 (m, 1H), 3.18–2.97 (m, 4H), 2.70 (t, 2H, J = 7.5 Hz), 2.44–1.96 (m, 4H), 2.35 (s, 3H), 1.89 (d, 1H, J = 14 Hz), 0.82 (d, 3H, J = 7.4 Hz); 13C NMR (CD3OD) δ 158.7, 144.6, 141.7, 141.5, 130.5, 129.9, 129.7, 129.5, 127.5, 127.0, 119.3, 115.1, 114.7, 98.2, 59.8, 58.4, 54.8, 42.7, 34.7, 33.6, 26.8, 23.2, 21.3, 11.7. Anal. (C28H35NO4S) C, H, N.

3-(1,4-Dimethylpiperidin-4-yl)phenol (6a) Hydrochloride

Compound 6a was synthesized as described by McElvain and Clemens.[18]1H NMR (CDCl3) δ 7.15 (t, 1H, J = 7.9 Hz), 6.81 (d, 1H, J = 8.0 Hz), 6.76 (t, 1H, J = 1.7 Hz), 6.62 (dd, 1H, J = 7.9, 1.9 Hz), 2.67–2.43 (m, 4H), 2.31 (s, 3H), 2.21–2.08 (m, 2H), 1.86–1.73 (m, 2H), 1.20 (s, 3H); 13C NMR (CDCl3) δ 157.1, 129.5, 117.1, 113.4, 113.2, 52.1, 45.7, 36.3, 35.5; MS (ESI) m/z 206.1 (M + H)+. The free base was converted to 6a·HCl as white needles from methanol/ether: mp 187–189 °C. Anal. (C13H20ClNO·0.25H2O) C, H, N.

3-[4-Methyl-1-(3-phenylpropyl)piperidin-4-yl]phenol (6b) Hydrochloride

A solution of 17 (98 mg, 0.45 mmol) in CH2Cl2 (5 mL) was treated with BCl3 (5 mL, 1 M in CH2Cl2) at −78 °C. When the mixture was warmed to room temperature, the reaction was quenched with aqueous piperazine and the mixture was refluxed for 30 min. The cooled solution was extracted with CH2Cl2. The combined organic layers were washed with water, dried (Na2SO4), and concentrated. The residue was subjected to chromatography on silica gel using a gradient of CMA80 in CH2Cl2 to afford 6b as an oil: 1H NMR (CDCl3) δ 7.27–7.08 (m, 6H), 6.83 (d, 1H, J = 7.9 Hz), 6.76–6.73 (m, 1H), 6.59 (dd, 1H, J = 7.9, 2.0 Hz), 5.87 (bs, 1H), 2.41–2.33 (m, 2H), 2.61–2.18 (m, 6H), 2.17–2.05 (m, 2H), 1.90–1.69 (m, 4H), 1.17 (s, 3H); 13C NMR (CDCl3) δ 156.5, 141.9, 129.4, 128.5, 128.3, 125.8, 117.6, 113.5, 113.1, 58.4, 50.2, 36.6, 36.1, 33.9, 28.2; MS (ESI) m/z 310.6 (M + H)+. The free base was converted to 32.5 mg (32%) of 6b·HCl as a pale yellow powder from methanol/ether: mp 47–51 °C (fusion). Anal. (C21H28ClNO·1.25H2O) C, H, N.

3-(1-Methylpiperidin-4-yl)phenol (7a) Hydrochloride

Compound 7a was synthesized as described by McElvain and Clemens.[18]1H HMR (CDCl3) δ 7.12 (t, 1H, J = 7.8 Hz), 6.63–6.66 (m, 2H), 6.58 (s, 1H), 3.02 (d, 2H, J = 11.7 Hz), 2.39–2.30 (m, 1H), 2.32 (s, 3H), 2.08 (t, 2H, J = 12.0 Hz), 1.73 (q, 2H, J = 13.1 Hz), 1.60 (d, 2H, J = 12.7 Hz); 13C NMR (CDCl3) δ 157.6, 147.7, 129.7, 119.1, 114.2, 113.2, 56.3, 46.2, 42.2, 32.9; MS (ESI) m/z 192.1 (M + H)+. Concentration from HCl in CH3OH gave 7a·HCl: mp 203–206 °C. Anal. (C12H18ClN2O) C, H, N.

3-[1-(3-Phenylpropyl)piperidin-4-yl]phenol (7b) Hydrochloride

A solution of 19 (1.0 g, 3.2 mmol) in CH2Cl2 (20 mL) at −78 °C was treated with BBr3 (1 M in CH2Cl2, 6.78 mL). After warming to room temperature and being stirred for 2 h, the mixture was again cooled to −78 °C, treated with MeOH (20 mL), and then allowed to warm to room temperature. The solution was evaporated, the residue dissolved in MeOH (20 mL), then evaporated. The residue was purified by chromatography on silica gel using CMA80/CH2Cl2 (1:1) to afford 0.51 g (54%) of 7b as a colorless oil. 1H NMR (CDCl3) δ 7.36–7.17 (m, 5H), 7.12 (t, 1H, J = 7.7 Hz), 6.77–6.60 (m, 3H), 3.66 (d, 2H, J = 12.1 Hz), 3.21–3.01 (m, 4H), 2.89–2.78 (m, 1H), 2.74 (t, 2H, J = 7.54 Hz), 2.18–1.87 (m, 6H); 13C NMR (DMSO-d6) δ 159.8, 146.4, 141.5, 130.8, 129.7, 129.5, 127.5, 118.7, 114.9, 114.6, 57.9, 54.4, 40.8, 33.6, 31.9, 26.9; ESI MS (M + H)+ 296.0. The hydrochloride salt prepared by adding HCl (1 M in Et2O) to a solution of the free base in Et2O gave 7b·HCl: mp 206–207 °C. Anal. (C20H26ClNO) C, H, N.

3-(4-Methylpiperazin-1-yl)phenol (8a) Dihydrochloride

A solution of 21 in CH2Cl2 (10 mL) was treated with BBr3 (15 mL, 1 M in CH2Cl2) at −78 °C. After warming to room temperature, the mixture was concentrated to a residue, dissolved in aqueous piperazine (10 mL), then refluxed for 1 h. The cooled solution was extracted with EtOAc (3 × 25 mL). The combined organics were washed with water, dried (Na2SO4), and concentrated. The residue was dissolved in CH3OH, acidified with HCl (1 M in Et2O), and concentrated to yield 489 mg (54%) of 8a·HCl: 1H NMR (CDCl3) δ 11.4 (bs, 1H), 8.83 (bs, 2H), 7.04 (t, 1H, J = 8.1 Hz), 6.50–6.41 (m, 2H), 6.35 (d, 1H, J = 8.1 Hz), 3.72 (d, 2H, J = 8.8 Hz), 3.45 (d, 2H, J = 6.4 Hz), 3.14 (d, 4H, J = 8.6 Hz), 2.78 (s, 3H); 13C NMR (CDCl3) δ 158.3, 150.5, 129.8, 107.8, 107.1, 103.4, 51.8, 45.6, 41.8; MS (ESI) m/z 193.2 (M + H)+. Mp 216–220 °C (fusion). Anal. (C11H18Cl2N2O·0.5H2O) C, H, N.

3-[4-(3-Phenylpropyl)piperazin-1-yl]phenol (8b)

Compound 8b was previously synthesized and reported.[15]

3-Methyl-4-[3-(propan-2-yloxy)phenyl]piperidine (10) Hydrochloride

A solution of racemic 9 was dehydrated according to literature procedure.[16] A sample of this material (5.01 g, 20.4 mmol) in MeOH (60 mL) with 10% Pd on carbon (0.50 g) was shaken under 50 psi of H2 for 48 h. The suspension was filtered through Celite and concentrated to provide a residue which was carried forward without further purification. The residue was dissolved in CHCl3 (200 mL), combined with 1-chloroethyl chloroformate (25.1 g, 0.176 mmol) and NaHCO3 (14.0 g, 167 mmol), and refluxed 72 h, with additions of ACE-Cl (7.8 g, 55 mmol after 12 h; 3.9 g, 27 mmol after 18 h). The mixture was concentrated and then dissolved in a minimum of EtOH at reflux. Upon cooling of the mixture, 0.92 g of 10·HCl (17% over two steps) was collected by filtration. 1H NMR (CD3OD) δ 7.23 (t, 1H, J = 7.9 Hz), 6.83–6.70 (m, 3H), 4.59 (septet, 1H, J = 6.0 Hz), 3.56–3.25 (m, 3H), 3.25–3.02 (m, 2H), 2.44–2.29 (m, 1H), 2.26 (m, 1H), 1.96–1.84 (m, 1H), 1.30 (d, 6H, J = 6.0 Hz), 0.84 (d, 3H, J = 7.3 Hz); 13C NMR (CD3OD) δ 129.1, 119.0, 115.0, 113.43, 69.5, 49.7, 44.3, 41.6, 32.2, 21.0, 20.9, 10.1.

3-Methyl-1-(3-phenylpropyl)-4-[3-(propan-2-yloxy)phenyl]piperidine (11)

A solution of 10·HCl (343 mg, 1.27 mmol) in 1,2-dichloroethane (4.5 mL) was treated with NEt3 (362 μL, 2.6 mmol), 3-phenylpropanal (190 μL, 1.1 mmol), and NaBH(OAc)3 (393 mg, 1.85 mmol). After being stirred for 2 h, the mixture was poured into saturated NaHCO3 (10 mL) and extracted with CH2Cl2 (3 × 10 mL). The organic layer was dried (Na2SO4) and concentrated. The residue was purified by chromatography on silica gel with 0–40% EtOAc in hexanes to yield 246 mg (55%) of 11 as a colorless oil. 1H NMR (CDCl3) δ 7.34–7.13 (m, 6H), 6.80–6.67 (m, 3H), 4.54 (septet, 1H, J = 6.0 Hz,), 3.07–2.93 (m, 1H), 2.90–2.57 (m, 4 H), 2.45–1.95 (m, 6H), 1.91–1.72 (m, 2H), 1.69–1.56 (m, 1H), 1.33 (d, 6H, J = 6.0 Hz), 0.83 (d, 3H, J = 7.0 Hz).

1-Methyl-4-[3-(propan-2-yloxy)phenyl]piperidin-4-ol (13)

A solution of n-BuLi (8.7 mL, 2.5 M in hexanes, 22 mmol) was added dropwise to 1-bromo-3-(1-methylethoxy)benezene (12)[16] (5.25 g, 24.4 mmol) in THF (14 mL) at −78 °C. After 1 h, N-methyl-4-piperdinone (2.49 g, 22.0 mmol) was added dropwise at −78 °C. The mixture was allowed to warm to room temperature overnight and then was chilled to 0 °C and added to 6 M HCl (8 mL) and concentrated. The aqueous emulsion was extracted with hexane. The organic layer was discarded. The aqueous layer was adjusted to pH 13–14 with 2 M NH4OH and extracted with hexane. The combined hexane layers were dried (Na2SO4) and concentrated to afford 2.29 g (42%) of 13. 1H NMR (CDCl3) δ 7.25 (t, 1H, J = 7.9 Hz), 7.09–7.02 (m, 2H), 6.81–6.76 (m, 1H), 4.56 (septet, 1H, J = 6.0 Hz), 2.79–2.70 (m, 2H), 2.51–2.38 (m, 2H), 2.35 (s, 3H), 2.17 (td, 2H, J = 12.8, 4.2 Hz), 1.80–1.70 (m, 2H), 1.33 (d, 6H, J = 6.0 Hz).

1-Methyl-4-[3-(propan-2-yloxy)phenyl]-1,2,3,6-tetrahydropyridine (14)

A toluene (15 mL) solution of 13 (2.29 g, 9.2 mmol) was refluxed with TsOH·H2O (3.50 g, 18.4 mmol) for 3 h. The cooled solution was extracted with water, and the toluene layer was discarded. The aqueous layer was adjusted to pH 13–14 with 2 M NaOH and then extracted with hexane. The combined hexane layers were washed with 2 M NaOH, dried (Na2SO4), and concentrated to afford 1.67 g (79%) of 14. 1H NMR (CDCl3) δ 7.25–7.15 (m, 1H), 6.96 (d, 1H, J = 8.0 Hz), 6.91 (s, 1H), 6.77 (dd, 1H, J = 8.1, 2.4 Hz), 6.05 (m, 1H), 4.55 (septet, 1H, J = 6.1 Hz), 3.13–3.08 (m, 2H), 2.69–2.63 (m, 2H), 2.61–2.53 (m, 2H), 2.40 (s, 3H), 1.33 (d, 6H, J = 6.0 Hz).

4-Dimethyl-4-[3-(propan-2-yloxy)phenyl]piperidine (15)

A solution of n-BuLi (4.5 mL, 2.5 M in hexanes, 11.3 mmol) was added dropwise to a solution of 14 (1.67 g, 7.22 mmol) in THF (17.5 mL) maintained between −10 and −20 °C. After 15 min, the solution was cooled to −50 °C and dimethyl sulfate (0.77 mL, 8.1 mmol) was slowly and cautiously added. The mixture was stirred an additional 30 min. Then 2 M NH4OH (10 mL) was added. The resulting mixture was extracted with hexane. The hexane layer was washed with water, dried (Na2SO4), and concentrated to a residue. The residue was dissolved in CH3OH (20 mL), cooled in an ice bath, and treated with NaBH4 (0.42 g, 11 mmol). The mixture was stirred for 3 h at room temperature and then was quenched with the addition of acetone and saturated NaHCO3. The concentrated residue was dissolved in water and EtOAc. The aqueous layer was extracted again with EtOAc before the combined organic layer was washed with water and then concentrated to afford 1.45 g (81%) of 15. 1H NMR (CDCl3) δ 7.22 (t, 1H, J = 8.0 Hz), 6.91 (d, 1H, J = 7.8 Hz), 6.88 (s, 1H), 6.72 (dd, 1H, J = 8.0, 2.2 Hz), 4.54 (septet, 1H, J = 6.0 Hz), 2.54–2.33 (m, 4H), 2.56 (s, 3H), 2.20–2.08 (m, 2H), 1.81–1.70 (m, 2H), 1.34 (d, 6H, J = 6.1 Hz), 1.21 (s, 3H).

4-Methyl-4-[3-(propan-2-yloxy)phenyl]piperidine (16)

A sample of 15 (1.44 g, 5.8 mmol) was concentrated thrice from toluene and then dissolved in 1,2-dichloroethane (8.7 mL). A freshly distilled aliquot of 1-chloroethyl chloroformate (1.81 mL, 17.4 mmol) was added under an inert atmosphere, and the resulting black solution was refluxed overnight. The concentrated residue was then dissolved in CH3OH and refluxed for 1 h. The concentrated residue was dissolved in 2 M NaOH and extracted with CH2Cl2. The combined organic layers were dried (Na2SO4), concentrated, and subjected to chromatography on silica gel using a gradient of CMA80 in DCM to afford 677 mg (50%) of 16. 1H NMR (CDCl3) δ 7.23 (t, 1H, J = 8.0 Hz), 6.94–6.86 (m, 2H), 6.72 (dd, 1H, J = 8.1, 2.3 Hz), 4.55 (septet, 1H, J = 6.0 Hz), 2.97–2.77 (m, 4H), 2.09–1.97(m, 2H), 1.74–1.59 (m, 2H), 1.34 (d, 6H, J = 6.1 Hz), 1.24 (s, 3H).

4-Methyl-1-(3-phenylpropyl)-4-[3-(propan-2-yloxy)phenyl]piperidine (17)

A solution of 16 (105 mg, 0.45 mmol) and 3-phenylpropanal (78 mg, 0.54 mmol) in trifluoroethanol (3 mL) was stirred for 15 min at room temperature before NaCNBH3 (60 mg, 0.9 mmol) was added. After 18 h, the solution was concentrated. The residue was taken up in dilute NH4OH and extracted with CH2Cl2. The concentrated organic layer was subjected to chromatography on silica gel using a gradient of CMA80 in CH2Cl2 to afford 98 mg (62%) of 17. 1H NMR (CDCl3) δ 7.30–7.13 (m, 6H), 6.91 (s, 1H), 6.90–6.86 (m, 1H), 6.71 (dd, 1H, J = 8.1, 2.0 Hz), 4.53 (septet, 1H, J = 6.1 Hz), 2.66–2.06 (m, 10H), 1.88–1.70 (m, 4H), 1.33 (d, 6H, J = 6.1 Hz), 1.20 (s, 3H).

4-(3-Methoxyphenyl)-1-(3-phenylpropyl)piperidine (19)

A solution of 1-benzyl-4-(3-methoxyphenyl)-1,2,5,6-tetrahydropyridine (18)[17] (1.14 g, 4.1 mmol) in EtOH (25 mL) with 20% Pd(OH)2 on carbon (0.50 g) was shaken under 40 psi of H2 for 18 h. The suspension was filtered through Celite, and the resulting solution was treated with 3-phenylpropanal (550 mg, 4.1 mmol). After the mixture was stirred for 30 min, NaBH3CN (0.75 g, 12 mmol) was added. After 18 h, the mixture was evaporated and the residue treated with saturated NaHCO3 (20 mL) and extracted with EtOAc (3 × 25 mL). The organic layer was dried over Na2SO4, filtered, and evaporated to obtain a yellow residue. The residue was purified by chromatography on silica gel with CMA80/CH2Cl2 (1:1) to afford 1.10 g (87%) of 19 as a colorless oil. 1H NMR (CDCl3) δ 7.28–7.14 (m, 6H), 6.79 (m, 3H), 3.79 (s, 3H), 3.07 (d, 2H, J = 15 Hz), 2.65 (t, 2H), 2.48 (m, 1H), 2.40 (t, 2H), 2.01 (m, 2H), 1.81 (m, 6H); 13C NMR (CDCl3) δ 159.8, 148.3, 142.4, 129.3, 128.4, 128.3, 125.8, 119.3, 112.7, 111.3, 58.6, 55.2, 54.4, 42.9, 33.9, 33.4, 28.9; ESI MS (M + H)+ 310.5.

1-(3-Methoxyphenyl)-4-methylpiperazine (21)

A solution of 1-methylpiperazine (20) (500 mg, 5 mmol), 3-bromoanisole (0.94 mL, 7.5 mmol), and KO-t-Bu (842 mg, 7.5 mmol) in toluene (20 mL) was degassed with nitrogen before Pd(t-Bu3P)2 (13 mg, 0.025 mmol) was added. The mixture was refluxed overnight. Chromatography on silica using a gradient of CMA80 in DCM afforded 687 mg (66%) of 21.