Synthesis, nicotinic acetylcholine receptor binding, and antinociceptive properties of 2'-fluoro-3'-(substituted pyridinyl)-7-deschloroepibatidine analogues.

2'-Fluoro-3-(substituted pyridine)epibatidine analogues 7a-e and 8a-e were synthesized, and their in vitro and in vivo nAChR properties were determined. 2'-Fluoro-3'-(4″-pyridinyl)deschloroepibatidine (7a) and 2'-fluoro-3'-(3″-pyridinyl)deschloroepibatidine (8a) were synthesized as bioisosteres of the 4'-nitrophenyl lead compounds 5a and 5g. Comparison of the in vitro nAChR properties of 7a and 8a to those of 5a and 5g showed that 7a and 8a had in vitro nAChR properties similar to those of 5a and 5g but both were more selective for the α4β2-nAChR relative to the α3β4- and α7-nAChRs than 5a and 5g. The in vivo nAChR properties in mice of 7a were similar to those of 5a. In contrast, 8a was an agonist in all four mouse acute tests, whereas 5g was active only in a spontaneous activity test. In addition, 5g was a nicotine antagonist in both the tail-flick and hot-plate tests, whereas 8a was an antagonist only in the tail-flick test.

Introduction

Tobacco use continues to be the leading cause of preventable deaths in the United States as well as globally. Current statistics reveal that smoking-related diseases are responsible for nearly 6 million premature deaths globally annually.[1] In the United States, tobacco use is responsible for nearly 1 in 5 deaths that is approximately 443 000 premature deaths.[2] In 2010, an estimated 23% (58.3 million) of U.S. adults were current cigarette smokers.[3] Tobacco use increases the risk of multiple cancers such as cancers of the lung, mouth, nasal cavities, larynx, pharynx, esophagus, stomach, colorectum, liver, pancreas, kidney, bladder, uterine, cervix, ovary, and myeloid cells. Therefore, smoking accounts for at least 30% of all cancer deaths and 87% of lung cancer deaths.[4]

Due to the well-documented negative heath consequences, approximately 70% of smokers want to quit and about 40% try to quit every year. Of those who try to quit, only about 7% stay off nicotine for more than a year. The vast majority do not make it even a week without cigarettes. The major factor that is attributed to the initiation and sustaining of smoking is the presence of nicotine (1), the addictive substance in tobacco. Nicotine can produce a myriad of behavioral effects and is unquestionably one of the most popular and powerful reinforcing agents. Both the psychological and physiological effects of tobacco smoke are a result of nicotine’s activation of various nicotinic acetylcholine receptor (nAChR) subtypes. For example, nicotine interacts with α4β2-, α4β2α6*-, α4β2α5*, and α7-nAChR in the dopaminergic mesolimbic pathway, a brain system thought to mediate the pleasurable and rewarding effects of most substances of abuse, including nicotine.[5] In addition, currently, the few treatments for nicotine dependence include nicotine replacement therapies (NRT); the antidepressant buproprion[6,7] (2), which acts as a dopamine uptake inhibitor in addition to its properties as a nicotinic antagonist of α3β4- and α4β2-nAChRs;[8] and the FDA-approved varenicline[9,10] (3), which acts as a partial nicotine agonist at the α4β2 and a full agonist at the α3β4- and α7-nAChRs.[11] In addition, varenicline has affinity for α6β2*-nAChR equal to that at α4β2-nAChR, but functionally varenicline was more potent in stimulating α6β2* versus α4β2* mediated [3H]dopamine release from rat striatal synaptosomes.[12] However, side effects such as gastrointestinal disturbances (nausea and vomiting) and neuropsychiatric effects (trouble sleeping, unusual dreams, violent or suicidal ideation) were frequently reported with the use of varenicline. In addition, recent evidence suggests that varenicline produces increased risk of heart attack, stroke, and/or other cardiovascular problems.[13] Therefore, there is need for development of new and improved pharmacotherapies for smoking cessation.

The natural alkaloid epibatidine (4, exo-2-(2′-chloro-5′-pyridinyl)-7-azabicyclo[2.2.1]heptane) is an important lead structure in the development of pharmacotherapies for treating nicotine addiction as well as other central nervous system (CNS) disorders including Alzheimers and Parkinson’s diseases, pain, schizophrenia, anxiety, depression, and Tourette’s syndrome among others.[14] Since its isolation and structural determination in 1992,[15] epibatidine has drawn a lot of interest because of its very high affinity for the α4β2*-nAChRs.[16,17] In previous studies, we reported the synthesis, nAChR binding affinity, and pharmacological properties of a number of epibatidine analogues.[18,19] Interestingly, some analogues retained high affinity for nAChR but unlike epibatidine showed no agonist activity in the acute mouse antinociception test and were antagonists of nicotine-induced antinociception in these assays.[18−20] For example, we identified 2′-fluoro-3′-(4-nitrophenyl)deschloroepibatidine (5a), also referred to as RTI-7527-102 and 4-nitro-PFEB, as an nAChR ligand with a Ki value of 0.009 nM for inhibition of [3H]epibatidine binding. This compound also showed potent antagonism of nicotine-induced antinociception in the tail-flick and hot-plate tests in mice.[21] In a separate study, we showed that 5a was a competitive antagonist of human α4β2-nAChRs with a potency 17-fold higher than that of dihydro-β-erythroidine (6) with very low efficacy at α3β4- and α7-nAChRs.[22] In a more recent study, the α4β2-nAChR antagonist 5a attenuated the discriminative stimulus effects of nicotine, reduced nicotine’s ability to facilitate intracranial self-stimulation (ICSS), blocked conditioned place preference (CPP) produced by nicotine in mice, and dose-dependently blocked nicotine self-administration in rats.[23] Thus, 5a has both in vitro and in vivo properties thought to be favorable for a potential pharmacotherapy to treat smokers. However, the presence of a nitro-substituted phenyl group, a system that is associated with toxicity via partial reduction in vivo to the hydroxylamine, which can undergo metabolic activation to an electrophilic nitroso species of 5a, raises concern about its future development. In a recent study, we reported that replacement of the 4-nitro group in 5a by other strong electron-withdrawing groups led to compounds 5b–g that retained high affinity for α4β2-nAChRs and potent antagonist activity in the tail-flick test.[24]

In this study, we report the synthesis, nAChR binding, and pharmacological properties of compounds 7a–e and 8a–e. Compound 7a is a bioisosteric analogue of 5a where the nitrophenyl group has been replaced by a pyridine nitrogen. Compound 8a is a similar bioisosteric analogue of 5g, a compound that has a Ki value of 0.053 nM of affinity for inhibition of [3H]epibatidine binding and AD50 values of 0.5 and 130 μg/kg in the tail-flick and hot-plate tests.[24] The syntheses and evaluation of analogues 7b–e and 8b–e allowed a determination of the effects of electron-withdrawing and -donating groups on the pyridine ring. See Chart 1 for the structures of the compounds described in the above paragraphs.

Chart 1

Structures of Compounds 1–4, 5a–g, 6, 7a–e, and 8a–e

Chemistry

The synthetic route to the 7a–c and 8a–e analogues commenced with the intermediate 7-tert-butoxycarbonyl-2-exo-(2′-amino-3′-bromo-5′-pyridinyl)-7-azabicyclo[2.2.1]heptane (9) prepared in several steps from N-Boc pyrrole as reported in earlier work.[25,26] As outlined in Scheme 1, the Suzuki cross-couplings of the haloboronic acids, that is, 2-fluoropyridine-5-boronic acid, 2-fluoropyridine-4-boronic acid, 2-chloropyridine-5-boronic acid, and 2-chloropyridine-4-boronic acid with the 2′-amino-3′-bromo compound 9, carried out in the presence of palladium diacetate, tri-(o-tolyl)phosphine, and sodium carbonate, heated at 80 °C in 1,2-dimethoxyethane and water for 5 h furnished the bipyridine intermediates 11b, 11c, 12b, and 12c.[27] Suzuki cross-coupling of pyridine-4-boronic acid, pyridine-3-boronic acid, and 2-methoxypyridine-5-boronic acids with 9 in the presence of tetrakis(triphenylphosphine) palladium(0) as the catalyst, potassium carbonate as the base, and toluene (15 mL), ethanol (1.5 mL), and water (1.5 mL) as solvents and heating at reflux for 24 h in a sealed tube provided the cross-coupled products 11a, 11d, and 12a in good yields. Conversion of the amino group to the fluoro group along with a concomitant removal of the tert-butyloxycarbonyl protecting group in the intermediates 11a–d and 12a–c performed through the diazotization reaction with sodium nitrite in the presence of hydrogen fluoride in pyridine (70%) furnished the products 8a–e and 7a–c. Compound 8d was synthesized by subjecting 2-exo-(2′-fluoro-3′-bromo-5′-pyridinyl)-7-azabicyclo[2.2.1]heptane 10(21) to a Suzuki–Miyaura cross-coupling with 2-aminopyridine-5-pinacol boronic ester in the presence of tetrakis(triphenylphosphine) palladium(0), potassium carbonate, 1,4-dioxane, and water, heated at 110 °C in a sealed tube overnight. The reaction furnished the diamine 8d in a 67% yield (Scheme 1).

Scheme 1

The synthesis of the 2-exo-[2′-fluoro-3′-(2-aminopyridin-4-yl)-5′-pyridinyl]-7-azabicyclo[2.2.1]heptane 7d, was accomplished in a “one-pot” reaction that combined the borylation and the Suzuki–Miyaura steps (Scheme 2). The borylation reaction was accomplished using Buchwald’s dialkylphosphinobiphenyl ligand, 2-dicyclohexylphosphino-2′,4′,6′-triisopropylbiphenyl (XPhos), and tris(dibenzylideneacetone)dipalladium (0) as the catalytic system.[28,29] Cross-coupling of 2-amino-4-bromopryidine (13) and bis(pinacolato)diborane in the presence of XPhos, tris(dibenzylideneacetone)dipalladium (0), and potassium acetate heated at 110 °C in 1,4-dioxane converted 13 to the boronic ester, which was carried on to the next step directly by addition of 2′-fluoro-3′-bromo intermediate 10, tribasic potassium phosphate as base, and an additional 3 mol % of tris(dibenzylideneacetone)dipalladium (0) and heating at 110 °C for 18 h to provide 7d. Compound 7e was synthesized as shown in Scheme 3. Borylation of 14 was achieved by cross-coupling with bis(pinacolato)diborane heated at 80 °C in the presence of 1,1′-bis(diphenylphosphino)ferrocene-palladium(II)dichloride, potassium acetate, and dimethylformamide as the solvent to provide the pinacol boronic ester 15 (Scheme 3). Compound 15 was cross-coupled with 2′-fluoro-3′-bromo intermediate 10 to furnish 2-exo-[2′-fluoro-3′-(2-methoxypyridin-4-yl)-5′-pyridinyl]-7-azabicyclo[2.2.1]heptane (7e).

Scheme 2

Scheme 3

Biology

The inhibition of [3H]epibatidine binding at α4β2*-nAChRs were conducted as previously reported.[21] The binding assays were performed using tissue homogenates prepared from freshly collected cerebral cortices from adult male Sprague–Dawley rats. These homogenates were frozen at −80 °C until use. It should be noted that, although this brain region contains a variety of nAChRs, α4β2 is the predominant (>90%) subtype. The epibatidine analogues were tested for agonist and antagonist activity at rat α4β2-, α3β4-, and α7-nAChRs in an in vitro electrophysiology assay as previously described.[24] The epibatidine analogues were tested in mice for their effects on body temperature and two pain models (tail-flick and hot-plate assays) after acute administration as previously described.[21] For the antagonist experiments, male ICR adult mice were pretreated subcutaneously (sc) with either saline or epibatidine analogues 10 min before nicotine. Nicotine was administered at a dose of 2.5 mg/kg, sc (an ED84 dose), and mice were tested 5 min later. ED50 and AD50 values with 95% confidence limits were determined.

Results and Discussion

The nAChR binding affinities and the functional nicotinic pharmacological properties of 2′-fluoro-3′-(substituted pyridine)deschloroepibatidine analogues 7a–e and 8a–e were determined. The Ki values for the inhibition of [3H]epibatidine binding at the α4β2*-nAChRs for compounds 7a–e and 8a–e along with reference compounds nat-epibatidine (4), varenicline (3), and the lead compounds 5a and 5g are listed in Table 1. The reference standards nat-epibatidine and varenicline and lead compounds 5a and 5g have Ki values of 0.026, 0.12, 0.009, and 0.053 nM for the α4β2*-nAChR, respectively.[24] The pyridine bioisosteres 7a and 8a of 5a and 5b, respectively, had Ki values of 0.12 and 0.35 nM, respectively, for α4β2*-nAChR. Even though 7a and 8a have Ki values slightly less than 5a and 5g, respectively, their Ki values are still subnanomolar and thus have potent affinity for nAChRs labeled by [3H]epibatidine.

Table 1

Radioligand Binding and Efficacy Profile Data for 2′-Fluoro-3′-(substituted pyridine)deschloroepibatidine Analogues

				agonist activity at 100 μM (% of max ACh activity)d			antagonist activity at 100 μM (% ACh response remaining)e
compda	X	Y	α4β2* [3H]epibatidineb (Ki, nM) (Hill slope)	α4β2	α3β4	α7	α4β2	α3β4	α7
nicotinec			1.50 ± 0.30	40 ± 1	37 ± 3	43 ± 5	nd	nd	nd
nat-epibatidine			0.026 ± 0.002	131 ± 13	97 ± 4	150 ± 8	nd	nd	nd
varenicline			0.12 ± 0.02	13 ± 0.4	66 ± 4	74 ± 5	38 ± 2	nd	nd
5af	NO2	H	0.009 ± 0.001	0	4 ± 1	6 ± 1	6 ± 1	9 ± 2	55 ± 6
5gf	H	NO2	0.053 ± 0.004	1.0 ± 0.1	1.3 ± 0.2	6 ± 1	5 ± 1	2.1 ± 0.4	17 ± 3
7a	H		0.12 ± 0.03	1.5 ± 0.3	7 ± 1	8 ± 2	6 ± 1	34 ± 4	45 ± 4
7b	F		0.067 ± 0.01	3 ± 0.3	5 ± 0.4	2 ± 0.5	10 ± 1	27 ± 2	44 ± 4
7c	Cl		1.18 ± 0.14	1.2 ± 0.2	4 ± 0.4	3 ± 0.3	7 ± 1	20 ± 1	31 ± 3
7d	NH2		0.13 ± 0.005	3.9 ± 0.5	0	10 ± 4	8 ± 1	9 ± 1	32 ± 10
7e	CH3		0.04 ± 0.012	0	1.5 ± 0.2	2.1 ± 0.8	4 ± 1	6 ± 1	12 ± 4
8a	H		0.35 ± 0.038	2 ± 0.3	3 ± 0.5	7 ± 2	7 ± 1	15 ± 2	54 ± 9
8b	F		0.049 ± 0.02	9 ± 1	9 ± 1	8 ± 0.7	24 ± 4	73 ± 8	75 ± 18
8c	Cl		0.063 ± 0.08	1 ± 0.2	14 ± 1	0	8 ± 1	23 ± 2	32 ± 4
8d	NH2		0.25 ± 0.033	2.7 ± 0.3	1.3 ± 0.3	5 ± 2	7 ± 1	9 ± 1	50 ± 6
8e	CH3O		0.13 ± 0.027	5 ± 0.1	9 ± 2	22 ± 4	14 ± 1	23 ± 6	41 ± 2

All compounds were tested as their (±)-isomers.

The Kd for (±)-[3H]epibatidine is 0.02 nM.

Data taken from ref (21).

Assessed by comparing the current response to 100 μM of each compound to the mean current response of three preceding applications of ACh, applied at an EC20 concentration (20 μM for α4β2, 110 μM for α3β4) or an EC50 concentration (300 μM for α7) and expressed as a percentage of the maximal response to ACh.

Assessed by comparing the current response to an EC50 concentration of ACh (70 μM for α4β2, 200 μM for α3β4, 300 μM for α7) in the presence of 100 μM of each compound to the mean current response of three preceding applications of ACh alone.

Data taken from ref (24).

Substitution of the 4′-pyridyl and 3′-pyridyl rings of analogues 7a and 8a, respectively, with 3′- and 4′-substituents, respectively, had only small effects on α4β2*-nAChR binding affinity. The Ki values varied from 0.04 to 1.18 nM. With the exception of the 3′-chloro analogue 7c, all the 3′-substituted 4′-pyridyl analogues 7b–e had very high affinity for α4β2*-nAChRs. The Ki values ranged from 0.04 for 7e to 0.13 nM for 7d. Even the 3′-chloro substituted analogue had a Ki value of 1.18 nM. All the 4′-substituted 3′-pyridyl analogues had Ki values of 0.04–0.35 nM and thus very high affinity for α4β2*-nAChR. The presence of 3′-substituents on the 4′-pyridyl analogues 7b–e or 4′-substituents on the 3′-pyridyl analogues 8b–e did not show any clear structure affinity patterns. In the case of the 7a–e series, the two highest affinity compounds were the electron-withdrawing 3′-fluoro analogue 7b (Ki = 0.067) and the electron-donating methoxy analogue 7e (Ki = 0.04). For the 4′-substituted 3′-pyridyl analogues, the 4′-fluoro and 4′-chloro electron-withdrawing analogues 8b and 8c (Ki = 0.049 and 0.063 nM) had higher α4β2*-nAChR affinity than the 3′-amino and 3′-methoxy electron-donating analogues 8d and 8e (Ki = 0.25 and 0.13 nM). However, all four compounds have subnanomolar Ki values.

The receptor subtype selectivity of 7a–e and 8a–e was assessed in an electrophysiological assay using rat α4β2-, α3β4-, and α7-nAChRs expressed in Xenopus oocytes and assayed by a two-electrode voltage clamp. Compounds were compared to previously determined values for nicotine, nat-epibatidine, varenicline, and compounds 5a and 5g (Table 1). Current responses to a high concentration (100 μM) of each compound were compared to the maximum response that can be achieved with acetylcholine. All compounds differed dramatically from nat-epibatidine (a full agonist at α4β2-, α3β4-, and α7-nAChRs). Compounds 7a, 7c, 7e, 8a, and 8c displayed little or no agonist activity at α4β2 in this initial screen, while 7b, 7d, 8b, and 8d–e had a low level of agonist activity at this subtype. Compounds 7d and 8d had little or no agonist activity at α3β4, while 7a–c, 8a–c, and 8e had a low level of agonist activity at this subtype. At α7-nAChRs, compounds 7b, 7e, and 8c had little or no agonist activity, compounds 7a, 7c–d, 8a–b, and 8d displayed low levels of agonist activity, and compound 8e showed a moderate level of agonist activity (22 ± 4% of the maximal acetylcholine response). Compounds 7a–e and 8a–e all showed lower agonist activity at α4β2-, α3β4-, and α7-nAChR than did varenicline (a partial agonist at α4β2- and a full agonist at α3β4- and α7-nAChRs).

As an initial screen of antagonist properties, we measured the current response to an EC50 concentration of acetylcholine in the presence of 100 μM of each compound and compared this to a preceding current response to acetylcholine alone (Table 1). Compounds 7a–e and 8a–e antagonized, to varying extents, the α4β2-, α3β4-, and α7-nAChR subtypes in this preliminary screen. This contrasts with varenicline, which can antagonize α4β2 receptors but is a full agonist at α3β4- and α7-nAChRs. The ability of 8e to both activate and antagonize the α7-nAChR subtype indicates that 8e is a partial agonist at this receptor. The results from this initial screen suggested that some compounds in this series may be selective α4β2 antagonists, and 7a, 7c, and 8a were selected for more detailed studies.

We examined the subtype selectivity of antagonist activity of compounds 7a, 7c, and 8a in more detail by generating concentration–inhibition curves (Table 2) and compared the IC50 values to the lead nitro compounds 5a and 5g. Compound 7a, the bioisosteric analogue of 5a, where the nitro group has been replaced by a pyridine nitrogen, displayed an improved α4β2 selectivity. While 5a was 2.5-fold selective for α4β2 over α3β4 and 10-fold selective for α4β2 over α7, compound 7a was 5.7-fold selective for α4β2 over α3β4 and 54-fold selective for α4β2 over α7. Similarly, compound 8a, the bioisosteric analogue of 5g, where the nitro group has been replaced by a pyridine nitrogen, displayed an improved α4β2 selectivity. While 5g was nonselective between α4β2 and α3β4 and 5-fold selective for α4β2 over α7, compound 8a was 11-fold selective for α4β2 over α3β4 and 58-fold selective for α4β2 over α7. We also examined 7c but found it to be less selective for α4β2 than was 7a. None of these compounds was as potent an antagonist as varenicline at α4β2-nAChRs. The ability of these compounds to antagonize α3β4- and α7-nAChRs differs markedly from varenicline, which is a full agonist at these subtypes.

Table 2

Comparison of Antagonist Potency (IC50 Values) for Several Epibatidine Analogues at α4β2-, α3β4-, and α7-nAChRs

	antagonist activitya IC50 (μM)
compd	α4β2	α3β4	α7
varenicline	0.20 ± 0.03b	d	d
5a	3.2 ± 0.2c	7.9 ± 0.5c	32 ± 12c
5g	4.3 ± 0.6	3.9 ± 0.3	23 ± 5
7a	1.4 ± 0.1	8 ± 1	75 ± 16
7c	2.0 ± 0.4	8.8 ± 0.9	56 ± 10
8a	1.7 ± 0.2	18 ± 3	99 ± 24

Antagonist activity of 7a, 7c, and 8a was assessed in the in vitro electrophysiology assay at a range of concentrations to generate concentration–inhibition curves. Data were fit to the following equation: I = Imax/[1+(IC50/X)], where I is the current response at a compound concentration (X), Imax is the maximum current, IC50 is the compound concentration producing half-maximal inhibition of the current response, and n is the Hill coefficient.

Data taken from ref (11).

Data taken from ref (24).

Varenicline is an agonist at α3β4- and α7-nAChRs, with an EC50 of 55 ± 8 and 18 ± 6 μM, respectively (ref (11)).

The 3′-substituted 4′-pyridyl analogues 7a–e and the 4′-substituted 3-pyridyl analogues (8a–e) were evaluated for their in vivo nAChR properties in mice, and the results were compared to the properties of lead compounds 5a and 5g and varenicline (Table 3). Similar to 5a, the pyridine bioisostere 7a does not have agonist activity in the tail-flick and hot-plate tests but like 5a and varenicline did show activity in the hypothermia and spontaneous-activity tests. The ED50 values for 7a in the hypothermia and spontaneous-activity tests were 1.69 and 0.38 mg/kg compared to 0.21 and 0.22 mg/kg for 5a. Similar to 5a, 7a antagonized nicotine-induced antinociception in the tail-flick and hot-plate tests with AD50 values of 12 and 290 μg/kg, respectively, compared to 3 and 120 μg/kg for 5a. Thus, 7a is a very good bioisostere analogue of 5a. In contrast, the 3′-pyridyl compound 8a, which is the pyridine bioisostere of 5g, was an agonist in all four mice acute tests, whereas 5g was active only in the spontaneous-activity test. In addition, whereas 5g was an antagonist of nicotine-induced antinociception in the tail-flick and hot-plate tests with AD50 values of 0.5 and 130 μg/kg, respectively, 8a was an antagonist in only the tail-flick test with an AD50 value of 3 μg/kg. Somewhat surprisingly, the in vivo properties of 8a are quite different from those of 5g, and thus, even though it is an interesting partial agonist, it is not a good bioisosteric analogue of 5g in the mice in vivo test. These discrepancies between the in vivo and in vitro effects of 5g and 8a could also be related to many factors such as differences in the metabolic profile and brain penetrability of these two analogues as well as the fact that we used expressed receptors systems and not native receptors preparations. Furthermore, since the agonistic response of nicotine in these two tests is largely mediated by α4β2* nAChR subtypes, it is possible that 5g and 8a differ in their affinity/activity at the various α4β2* nAChR subtypes mediating their pharmacological responses. In addition, in vivo regulation of nicotinic receptors such as α4β2* nAChR subtypes by these two analogues may differ also. For example, it is possible that, since 5g was more potent than 8a as a functional blocker in the tail-flick and hot-plate tests, in vivo desensitization/blockade of α4β2* receptors by 5g is more pronounced. We recently reported that varenicline (3) and sazetidine, two α4β2* nicotinic partial agonists that differ in their desensitization properties, differ in their potency to act as functional antagonists of nicotine in these tests.[30,31]

Table 3

Antinociception, Hypothermia, and Spontaneous Activity Profile Data for 2′-Fluoro-3′-(substituted pyridine)deschloroepibatidine Analogues

			ED50 (mg/kg)				AD50 (μg/kg)
compda	X	Y	tail-flick	hot-plate	hypothermia	spontaneous activity	tail-flick	hot-plate
nicotineb			1.3 (0.5–1.8)	0.65 (0.25–0.85)	1.0 (0.6–2.1)	0.5 (0.15–0.78)
nat-epibatidine			0.006 (0.001–0.01)	0.004 (0.001–0.008)	0.004 (0.002–0.008)	0.001 (0.0005–0.005)
varenicline			11% @ 10	10% @ 10	2.8	2.1	0.2	470
5ac	NO2	H	5% @ 10	10% @ 10	0.21 (0.04–1.9)	0.22 (0.04 ± 1.2)	3 (0.8–45)	120 (10–900)
5gc	H	NO2	3% @ 10	20% @ 10	0% @ 10	6.5 (5.3 ± 8.3)	0.5 (0.3–5)	130 (50–290)
7a	H		13% @ 10	40% @ 10	1.69 (1.1–2.6)	0.38 (0.2–2.7)	12 (10–172)	290 (19–991)
7b	F		5% @ 10	18% @ 10	1.58 (0.97–2.1)	0.17 (0.08–1.5)	4 (0.1–70)	117 (110–1100)
7c	Cl		11% @ 10	19% @ 10	2.74 (1.89–3.5)	1.01 (0.27–3.7)	320 (45–3262)	1370 (180–1430)
7d	NH2		11% @ 10	12% @ 10	1.87 (0.1–35)	0.61 (0.04–9.1)	9 (0.4–19)	10% @ 10000
7e	CH3O		5% @ 10	10% @ 10	8.5 (1.9–38.6)	1.82 (0.4–8.4)	0.3 (0.02–5.7)	40% @ 10000
8a	H		4.9 (3.6–6.7)	5 (3.7–6.7)	3.7 (2.9–4.5)	0.69 (0.4–12.8)	3 (0.5–24)	10% @ 1000
8b	F		3.6 (2.7–4.7)	3.27 (2.1–5.3)	0.68 (0.52–1.1)	0.38 (0.13–1.1)	1% @ 100	1% @ 100
8c	Cl		10% @ 10	27% @ 10	3.11 (1.5–5.1)	1.58 (0.5–4.4)	9 (2–38)	2001 (297–3610)
8d	NH2		5% @ 10	8% @ 10	2.8 (2–3.8)	184 (0.5–6.3)	30 (3–35)	50% @ 10
8e	CH3O		4.22 (3–5.3)	1.72 (0.9–3.4)	0.77 (0.51–1.2)	0.53 (0.19–1.1)	21 (3–125)	0% @ 100

All compounds were tested as their (±)-isomers.

Data taken from ref (21).

Data taken from ref (24).

Similar to the unsubstituted analogue 7a, none of the 3′-substituted 4-pyridyl analogues 7b–e had any agonist activity in the tail-flick and hot-plate tests (Table 3). In the case of the 4′-substituted 3′-pyridyl analogues, the 4′-fluoro and 4′-methoxy analogues 8b and 8e had agonist activity in the tail-flick and hot-plate tests. Similar to 5a, 5g, and varenicline, 7b–e and 8b–e had activity in the hypothermia and spontaneous-activity tests.

Analogues 7b–e and 8c–e antagonized nicotine-induced antinociception in the tail-flick test (Table 3). Analogues 7b–c and 8c also antagonized nicotine-induced antinociception in the hot-plate test with AD50 values ranging from 117 to 1370 μg/kg. Compounds 7a and 7b, which have AD50 values of 290 and 117 μg/kg, respectively, in the hot-plate test, compared to 470 μg/kg for varenicline strongly suggest that these compounds have good brain penetration. Unlike any of the other pyridine substituted analogues, the 3′-fluoro-4′-pyridyl analogue 8b was an agonist in all four acute mouse tests and had no antagonist properties.

Calculated physicochemical properties such as lipophilicity (clogP), topological polar surface area (TPSA), and derived values such as logBB can be used as an indication of the potential of a compound for development as a CNS drug. These molecular descriptors were calculated for lead compounds 5a and 5g, bioisosteric analogues 7a and 8a, respectively, as well as compounds 7b–e, 8b–e and reference compounds nicotine, epibatidine, and varenicline (Table 4). In general, CNS drugs have a clogP in the range 2–4,[32] TPSA less than 76 Å,[33] and logBB greater than −1.[34] All of the compounds have clogP values within or close to the desirable range and logBB values between −0.54 and +0.08. In addition, all of the compounds have TPSA values of less than 76 Å. A comparison of the logP, TPSA, and logBB values of 1.99, 37.81, and −0.12, respectively, for bioisosteric analogues 7a and 8a to the corresponding values of 3.14, 70.74, and −0.43 for lead compounds 5a and 5g show that these two bioisosteric analogues (7a and 8a) have at least as favorable if not better calculated physicochemical properties than lead compounds 5a and 5g. In addition, both 7a and 8a have calculated logBB values somewhat better than that of varenicline.

Table 4

Calculated Physiochemical Properties of 5a, 5g, 7a–e, 8a–e, Nicotine, Nat Epibatidine, and Varenicline

compd	logPa	TPSAa	logBBb
nicotine	1.16	16.13	0.08
epibatidine	1.84	24.92	0.05
varenicline	1.01	37.81	–0.27
5a	3.14	70.74	–0.43
5g	3.14	70.74	–0.43
7a	1.99	37.81	–0.12
7b	2.52	37.81	–0.04
7d	1.75	63.83	–0.54
7c	2.81	37.81	0.01
7e	2.12	37.81	–0.10
8a	1.99	37.81	–0.12
8b	2.52	37.81	–0.04
8c	2.81	37.81	0.01
8d	1.75	63.83	–0.54
8e	2.42	47.04	–0.19

ChemAxon Calculator Plugins, Marvin 6.1.0, 2013.

logBB = −0.0148 × TPSA + 0.152 × clogP + 0.139 (from ref (34)).

In summary, 2′-fluoro-3-(substituted pyridine)epibatidine analogues 7a–e and 8a–e were synthesized and evaluated for the ability to inhibit [3H]epibatidine binding to nAChR, tested for agonist and antagonist activity at α4β2-, α3β4-, and α7-nAChR in an electrophysiology assay, and evaluated for agonist effects in the tail-flick, hot-plate, spontaneous-activity, and hypothermia tests in the mouse and as antagonists of nicotine-induced antinociception in the tail-flick and hot-plate tests in the mouse. A comparison of the nAChR binding and electrophysiology of bioisosteres 7a and 8a to those of the nitrophenyl lead compounds 5a and 5g, respectively, showed that 7a and 8a had in vitro nAChR properties similar to those of 5a and 5g but were more selective for the α4β2-nAChR relative to the α3β4- and α7-nAChRs than 5a and 5g. Similar to 5a, 7a did not have agonist activity in the tail-flick and hot-plate tests and like 5a was a potent antagonist of nicotine-induced antinociception in these two tests. Thus, 7a is a very good bioisosteric analogue of 5a. In contrast, 8a unlike 5g was an agonist in both the tail-flick and hot-plate tests and was an antagonist of nicotine-induced antinociception only in the tail-flick test, whereas 5g was an antagonist in both tests. Even though 8a is not a good bioisosteric analogue of 5g in the mice test, it is an interesting partial agonist. A comparison of the AD50 value of 7a to that of varenicline in the hot-plate test, strongly suggest that this compound penetrates the brain in mice. Calculated logBB values for 7a and 8a also suggest that this compound will have good blood–brain barrier penetration. Since both nAChR antagonists and partial agonists are of interest as possible pharmacotherapies for treating smokers, both 7a and 8a are candidates for development as pharmacotherapies to treat nicotine addiction.

Experimental Section

Melting points were determined on a Mel-temp (Laboratory Devices, Inc.) capillary tube apparatus. NMR spectra were recorded on a Bruker Avance 300 or AMX 500 spectrometer using tetramethylsilane as internal standard. Mass spectra were determined on a Perkin–Elmer Sciex API 150EX mass spectrometer outfitted with APCI and ESI sources. Melting point was determined on a Laboratory Devices MEL-TEMP II. Elemental analyses were carried out by Atlantic Microlab, Inc., Norcross GA. The purity of the compounds (>95%) was established by elemental analysis. Analytical thin-layer chromatography (TLC) was carried out on plates precoated with silica gel (60 F254). TLC visualization was accomplished with a UV lamp or in an iodine chamber. Purifications by flash chromatography were performed on a Combiflash Teledyne ISCO instrument.

Suzuki Cross-Coupling Reaction: General Procedure (Method A)

To a resealable reaction vessel under nitrogen was added 1.0 equiv of 7-tert-butoxycarbonyl-2-exo-(2′-amino-3′-bromo-5′-pyridinyl)-7-azabicyclo[2.2.1]heptane (9), Pd(OAc)2 (0.1 equiv), P(o-tolyl)3 (0.2 equiv), sodium carbonate (2.0 equiv) and the respective pyridinyl boronic acid (1.6 equiv), DME (6 mL), and water (0.7 mL). The mixture was degassed through bubbling nitrogen, sealed, and heated on a sand bath at 80 °C for 5 h. The mixture was cooled, poured into 20 mL of a saturated aqueous solution of NaHCO3, and extracted with EtOAc (3 × 30 mL). The combined organic layers were dried over MgSO4 and filtered through Celite, and the solvent was removed under reduced pressure. The resultant residue was purified on silica gel by flash chromatography eluted with CHCl3–MeOH (50:1 to 10:1).

Suzuki Cross-Coupling Reaction: General Procedure (Method B)

To a resealable reaction vessel under nitrogen was added 1.0 equiv of the 3′-bromo compound 9, Pd(PPh3)4 (10 mol %), K2CO3 (2.0 equiv) and the respective pyridinyl boronic acid (1.3 equiv), toluene (12 mL), ethanol (1.5 mL), and water (1.5 mL). The mixture was degassed through bubbling nitrogen, sealed, and heated on a sand bath at 110 °C. After 24 h, the mixture was cooled, poured into 30 mL of H2O, and extracted with EtOAc (3 × 30 mL). The combined organic layers were dried over MgSO4 and filtered through Celite, and the solvent was removed in vacuo. The resultant residue was purified by flash chromatography on a silica gel column using hexanes–isopropanol (80:20 to 25:75) or CHCl3–MeOH (30:1 to 10:1) as the eluent.

General Procedure C: Removal of the Boc-Protecting Group

A solution of the Boc-protected compound in methylene chloride (5 mL) was treated with TFA (1.5 mL) and stirred at room temperature overnight. In some cases the solution was heated at 40 °C for 2 h and then stirred at room temperature overnight. The solvent was then removed in vacuo, and the residue was treated with a solution of NH4Cl (20 mL) and extracted with CHCl3–MeOH (10%) (3 × 30 mL). The combined organic layers were dried over anhydrous Na2SO4, filtered, and concentrated in vacuo, and the residue was purified by flash chromatography using a silica gel column eluted with CHCl3–MeOH (10%) to provide the respective amine.

2-exo-[2′-Fluoro-3′-(pyridin-4-yl)-5′-pyridinyl]-7-azabicyclo[2.2.1]heptane (7a) Fumarate

A solution of 12a (378 mg, 1.03 mmol, 1.0 equiv) in 70% HF in pyridine (1.5 mL) was stirred at 0 °C for 30 min. Sodium nitrite (806 mg, 10 equiv) was added in small portions, and the reaction mixture was stirred at room temperature. After 1 h, the mixture was poured into a 1:1 aqueous solution of NH4OH–H2O (100 mL) and extracted with EtOAc (3 × 100 mL). The combined organic layers were dried over anhydrous Na2SO4, filtered through Celite, and concentrated in vacuo. The resultant residue was purified by flash chromatography on a silica gel column using CHCl3–MeOH as the eluent to provide 192 mg (69%) of 7a as a colorless oil. 1H NMR (300 MHz, CD3OD) δ 1.50–1.78 (m, 6H), 2.01–2.08 (dd, J = 9.0, 11.2 Hz, 1H), 3.02–3.07 (dd, J = 8.7, 5.2 Hz, 1H), 3.66 (s, 1H), 3.77 (br s, 1H), 7.70–7.73 (m, 2H), 8.13 (dd, J = 2.4, 9.4 Hz, 1H), 8.18 (s, 1H), 8.64 (d, J = 1.5 Hz, 1H), 8.65 (d, J = 1.5 Hz, 1H); 13C NMR (CD3OD) δ 30.0, 31.8, 41.1, 45.7, 57.9, 63.7, 121.2, 125.1, 141.3, 142.1, 144.2, 147.8, 148.0, 150.6, 158.6, 161.7; MS (ESI) m/z 270.2 (M + H)+.

A solution of 7a (302 mg, 1.12 mmol) in chloroform (2 mL) was placed in vial and treated with 1.1 equiv of fumaric acid (0.65 M in MeOH). After 24 h, the white solid obtained was recrystallized from a MeOH–Et2O mixture to provide the salt 7a·C4H4O4 as a white solid: mp 192–195 °C. 1H NMR (300 MHz, CD3OD) δ 1.86–2.22 (m, 6H), 2.44–2.51 (dd, J = 9.0, 11.0 Hz, 1H), 3.50–3.55 (m, 1H), 4.35 (br s, 1H), 4.56 (d, J = 3.9 Hz, 1H), 6.63 (s, 1H), 7.72–7.75 (m, 2H), 8.20 (dd, J = 2.4, 9.0 Hz, 1H), 8.27 (d, J = 2.4 Hz, 1H), 8.67 (m, 2H); 13C NMR (CD3OD) δ 27.0, 29.0, 37.7, 43.4, 60.2, 64.1, 121.6, 125.1, 136.2, 137.6, 141.3, 143.9, 147.8, 148.0, 150.7, 159.0, 162.2, 171.4; MS (ESI) m/z 270.1 [(M – fumaric)+, M = C16H16FN3·C4H4O4]. Anal. (C20H20FN3O4·0.25 H2O) C, H, N.

2-exo-[2′-Fluoro-3′-(2-fluoropyridin-4-yl)-5′-pyridinyl]-7-azabicyclo[2.2.1]heptane (7b) Fumarate

A solution of 12b (230 mg, 0.60 mmol, 1.0 equiv) in 70% HF in pyridine (1.5 mL) was stirred at 0 °C for 30 min. Sodium nitrite (413 mg, 10 equiv) was added in small portions, and the reaction mixture was stirred at room temperature. After 1 h, the mixture was poured into a 1:1 aqueous solution of NH4OH–H2O (100 mL) and extracted with EtOAc (3 × 100 mL). The combined organic layers were dried over anhydrous Na2SO4, filtered through Celite, and concentrated in vacuo. The resultant residue was purified by flash chromatography on a silica gel column using CHCl3–MeOH as the eluent to provide 121 mg (70%) of 7b as a colorless oil. 1H NMR (300 MHz, CDCl3) δ 1.56–1.68 (m, 6H), 1.92–1.98 (dd, J = 9.1, 11.2 Hz, 1H), 2.81–2.86 (m, 1H), 3.60 (s, 1H), 3.83 (br s, 1H), 7.17 (d, J = 1.0 Hz, 1H), 7.43 (ddd, J = 1.6, 4.9, 6.9 Hz, 1H), 8.15–8.19 (m, 2H), 8.23 (d, J = 5.3 Hz, 1H); 13C NMR (CDCl3) δ 30.4, 31.5, 40.7, 44.2, 56.4, 62.9, 109.0, 119.4, 121.1, 139.6, 141.5, 147.5, 157.1, 160.3, 162.6, 162.7; MS (ESI) m/z 288.3 (M + H)+.

A solution of 7b (141 mg, 0.49 mmol) in CH2Cl2 in a vial was treated with 1.2 equiv of fumaric acid (0.65 M) in MeOH, and the vial was allowed to stand in a refrigerator overnight. The excess solvent was then removed in vacuo from the salt and then redissolved in a minimal amount of MeOH, and the fumarate salt was recrystallized from MeOH using diethyl ether to provide 110 mg (55%) of the salt 7b·C4H4O4 as a white crystalline solid: mp 203–205 °C. 1H NMR (500 MHz, CD3OD) δ 1.87–2.20 (m, 5H), 2.45–2.50 (dd, J = 9.3, 13.2 Hz, 1H), 3.50–3.53 (m, 1H), 4.34–4.35 (br s, 1H), 4.56 (d, J = 3.9 Hz, 1H), 6.64 (s, 2H), 7.41 (s, 1H), 7.61–7.63 (m, 1H), 8.21 (dd, J = 2.4, 9.3 Hz, 1H), 8.28 (d, J = 1.0 Hz, 1H), 8.32 (d, J = 5.3 Hz, 1H); 13C NMR (CD3OD) δ 25.8, 27.8, 36.5, 42.2, 59.0, 62.8, 109.4, 121.6, 135.0, 136.5, 140.1, 147.2, 147.8, 158.3, 160.2, 163.4, 165.3, 170.2; MS (ESI) m/z 288.3 [(M – fumaric)+, M = C16H15F2N3·C4H4O4]. Anal. (C20H19F2N3O4) C, H, N.

2-exo-[2′-Fluoro-3′-(2-chloropyridin-4-yl)-5′-pyridinyl]-7-azabicyclo[2.2.1]heptane (7c) Fumarate

A solution of 12c (130 mg, 0.32 mmol, 1.0 equiv) in 70% HF in pyridine (1.5 mL) was stirred at 0 °C for 30 min. Sodium nitrite (224 mg, 10 equiv) was added in small portions, and the reaction mixture was stirred at room temperature. After 1 h, the mixture was poured into a 1:1 aqueous solution of NH4OH–H2O (100 mL) and extracted with EtOAc (3 × 100 mL). The combined organic layers were dried over anhydrous Na2SO4, filtered through Celite, and concentrated in vacuo. The resultant residue was purified by flash chromatography on a silica gel column using CHCl3–MeOH as the eluent to provide 86 mg (87%) of 7c as a colorless oil. 1H NMR (300 MHz, CDCl3) δ 1.54–1.67 (m, 6H), 1.92–1.98 (dd, J = 9.1, 11.2 Hz, 1H), 2.81–2.86 (m, 1H), 3.60 (s, 1H), 3.83 (br s, 1H), 7.46 (dd, J = 1.2, 5.2 Hz, 1H), 7.56 (s, 1H), 8.12–8.15 (m, 2H), 8.47 (d, J = 5.2 Hz, 1H); 13C NMR (CDCl3) δ 30.4, 31.6, 40.7, 44.3, 56.4, 62.9, 119.2, 122.1, 139.6, 141.5, 145.1, 147.2, 149.9, 152.1, 157.1, 160.3; MS (ESI) m/z 304.3 (M + H)+.

A solution of 7c (106 mg, 0.35 mmol) in CH2Cl2 in a vial was treated with 1.2 equiv of fumaric acid (0.65 M) in MeOH, and the vial was allowed to stand in a refrigerator overnight. The excess solvent was then removed in vacuo from the salt and then redissolved in a minimal amount of MeOH, and the fumarate salt was recrystallized from MeOH using diethyl ether to provide 62 mg (42%) of the salt 7c·C4H4O4 as a white crystalline solid: mp 193–194 °C. 1H NMR (500 MHz, CD3OD) δ 1.87–2.21 (m, 5H), 2.45–2.50 (dd, J = 9.2, 13.2 Hz, 1H), 3.50–3.53 (m, 1H), 4.34–4.35 (br s, 1H), 4.56 (d, J = 3.9 Hz, 1H), 6.63 (s, 2H), 7.67 (dd, J = 1.4, 9.3 Hz, 1H), 7.80 (s, 1H), 8.21 (dd, J = 2.4, 9.3 Hz, 1H), 8.28 (d, J = 2.4 Hz, 1H), 8.48 (d, J = 4.9 Hz, 1H); 13C NMR (CD3OD) δ 25.7, 27.8, 36.5, 42.2, 59.0, 62.9, 119.4, 122.7, 135.0, 136.6, 140.1, 145.4, 147.3, 149.9, 151.8, 158.3, 160.3, 170.1; MS (ESI) m/z 304.0 [(M – fumaric)+, M = C16H15ClFN3·C4H4O4]. Anal. (C20H19ClFN3O4·0.25 H2O) C, H, N.

2-Fluoro-3-(2′-amino-4′-pyridinyl)deschloroepibatidine (7d) Hydrochloride

A solution of 2-amino-4-bromopyridine (200 mg, 1.16 mmol, 1.0 equiv), bispinacolato diborane (307 mg, 1.21 mmol, 1.05 equiv), Pd2dba3 (36 mg, 0.035 mmol, 3 mol %), Xphos (88 mg, 0.185 mmol, 16 mol %), and KOAc (272 mg, 2.77 mmol, 2.4 mmol) in dioxane placed in a resealable pressure vessel was degassed through bubbling nitrogen for 40 min then heated at 110 °C for 4 h. A TLC check revealed that all the bromopyridine had been converted to the boronic ester. The reaction was allowed to cool to room temperature, and K3PO4 (613 mg, 2.89 mmol, 2.5 equiv), a solution of 10 (270 mg, 1.0 mmol, 0.87 equiv) in dioxanes, an additional 3 mol % of Pd2dba3, and H2O (1 mL) were added to the reaction. The mixture was degassed for 30 min and heated for 18 h at 110 °C. The reaction was cooled to room temperature and extracted with EtOAC (3 × 30 mL). The combined organic layers were dried over MgSO4 and filtered through Celite, and the solvent was removed in vacuo. Two purifications of the residue by flash chromatography through an ISCO column using CHCl3–MeOH (10:1) as the eluent provided 60 mg (21%) of 7d as a colorless oil. 1H NMR (300 MHz, CDCl3) δ 1.51–1.71 (m, 5H), 1.90–1.97 (m, 1H), 2.36 (br s, 1H), 2.80–2.85 (dd, J = 3.8, 5.0 Hz, 1H), 3.61 (s, 1H), 3.81 (d, J = 2.7 Hz, 1H), 4.66 (br s, 2H), 6.72 (s, 1H), 6.84 (d, J = 5.3 Hz, 1H), 8.02 (dd, J = 2.3, 9.5 Hz, 1H), 8.11 (s, 1H), 8.13 (d, J = 5.7 Hz, 1H); 13C NMR (CDCl3) δ 30.2, 31.4, 40.5, 44.3, 56.5, 62.9, 108.1, 113.9, 139.4, 140.6, 143.7, 145.9, 148.5, 157.4, 158.8, 160.6; MS (ESI) m/z 285.5 (M + H)+.

A solution of 7d (122 mg, 0.43 mmol) in chloroform in a vial was treated with a 2.0 equiv solution of HCl in diethyl ether and allowed to stand at room temperature. The excess solvent was filtered off, and the obtained salt was washed with ether and then dried to provide 94 mg of the salt 7d·HCl as a white solid: mp 205–208 °C. 1H NMR (300 MHz, CD3OD) δ 1.83–2.28 (m, 5H), 2.46–2.53 (dd, J = 3.8, 9.6 Hz, 1H), 3.52–3.57 (dd, J = 3.1, 5.5 Hz, 1H), 4.37 (d, J = 3.6 Hz, 1H), 4.59 (d, J = 2.7 Hz, 1H), 7.02–7.05 (dd, J = 1.6, 6.1 Hz, 1H), 7.10 (s, 1H) 7.98 (d, J = 6.1 Hz, 1H) 8.16 (dd, J = 2.3, 9.2 Hz, 1H) 8.28 (s, 1H); 13CNMR (CD3OD) δ 26.8, 28.9, 37.5, 43.3, 60.5, 64.2, 112.0, 113.8, 137.4, 141.3, 143.2, 148.0, 148.2, 158.8, 158.9, 162.1; MS (ESI) m/z 285.7 [(M – HCl)+, M = C16H17FN4·2HCl]. Anal. (C16H19Cl2FN4) C, H, N.

2-exo-[2′-Fluoro-3′-(2-methoxypyridin-4-yl)-5′-pyridinyl]-7-azabicyclo[2.2.1]heptane (7e) Fumarate

To a resealable reaction pressure vessel under nitrogen was added compound 10 (180 mg, 0.66 mmol, 1.0 equiv), compound 16 (188 mg, 0.80 mmol, 1.2 equiv), (Pd(PPh3)4 (38 mg, 0.03 mmol, 5 mol %), K2CO3 (184 mg, 1.33 mmol, 2.0 equiv), 1,4-dioxane (10 mL), and water (0.80 mL). The reaction mixture was degassed through bubbling nitrogen for 40 min, sealed, and heated over a sand bath at 110 °C for 18 h. After cooling, the solvent was removed under reduced pressure, and to the residue was added 20 mL of H2O. The organic product was extracted using EtOAc (3 × 30 mL). The combined organic layers were dried over MgSO4 and filtered through Celite, and the solvent was removed in vacuo. Purification by flash chromatography on silica gel using MeOH–CHCl3 as the eluent provided 100 mg (50%) of 7e as a colorless oil. 1H NMR (300 MHz, CDCl3) δ 1.51–1.68 (m, 5H), 1.89–1.96 (dd, 3.8, 9.6 Hz, 1H), 1.98 (broad signal 1H), 2.79–2.84 (dd, J = 3.4, 5.5 Hz, 1H), 3.59 (s, 1H), 3.81 (s, 1H), 3.96 (s, 3 H), 6.96 (s, 1H), 7.07–7.10 (dt, J = 5.3, 1.5 Hz, 1H), 8.06 (dd, J = 2.4, 9.6 Hz, 1H), 8.11 (s, 1H), 8.21 (d, J = 5.3 Hz, 1H); 13C NMR (CDCl3) δ 30.2. 31.4, 40.5, 44.3, 53.5, 56.4, 62.9, 110.5, 116.6, 139.6, 140.8, 144.6, 146.2, 146.4, 147.2, 160.5, 164.6; MS (ESI) m/z (300.4) (M + H)+.

A solution of 7e (156 mg, 0.52 mmol) in CH2Cl2 in a vial was treated with 1.2 equiv of fumaric acid (0.65 M) in MeOH, and the vial was allowed to stand in a refrigerator overnight. The excess solvent was then removed in vacuo from the salt that was then redissolved in a minimal amount of MeOH, and the fumarate salt was recrystallized from MeOH using diethyl ether to provide 164 mg (74%) of the salt 7e·C4H4O4 as a white solid: mp 160–164 °C. 1H NMR (300 MHz, CD3OD) δ 1.85–2.19 (m, 5H), 2.43–2.50 (dd, J = 9.3, 13.2 Hz, 1H), 3.48–3.53 (m, 1H), 3.96 (s, 3H), 4.34 (br s, 1H), 4.55 (s, 1H), 6.65 (s, 2H), 7.07 (s, 1H), 7.22 (dd, J = 1.2, 4.1 Hz, 1H), 8.12 (d, J = 9.2 Hz, 1H), 8.22–8.23 (m, 2H); 13C NMR (CD3OD) δ 26.9, 29.0, 37.7, 43.4, 54.2, 60.2, 64.1, 111.6, 117.9, 136.1, 137.5, 141.2, 145.9, 147.5, 147.6, 148.3, 162.2, 166.2, 171.1; MS (ESI) m/z 300.3 [(M – fumaric)+, M = C17H18FN3O·C4H4O4]. Anal. (C21H22FN3O5) C, H, N.

2-exo-[2′-Fluoro-3′-(pyridin-3-yl)-5′-pyridinyl]-7-azabicyclo[2.2.1]heptane (8a) Hemifumarate

A solution of 11a (394 mg, 1.08 mmol, 1.0 equiv) in 70% HF in pyridine (1.5 mL) was stirred at 0 °C for 30 min. Sodium nitrite (742 mg, 10 equiv) was added in small portions, and the reaction mixture was stirred at room temperature. After 1 h, the mixture was poured into a 1:1 aqueous solution of NH4OH–H2O (40 mL) and extracted with EtOAc (3 × 40 mL). The combined organic layers were dried over anhydrous Na2SO4, filtered through Celite, and concentrated in vacuo. The resultant residue was purified by flash chromatography on silica gel using CHCl3–MeOH as the eluent to provide 203 mg (70%) of 8a as a colorless oil. 1H NMR (300 MHz, CD3OD) δ 1.49–1.79 (m, 6H), 2.01–2.08 (dd, J = 9.1, 11.2 Hz, 1H), 3.02–3.07 (dd, J = 3.3, 5.4 Hz, 1H), 3.67 (s, 1H), 3.77 (br s, 1H), 7.54 (dd, J = 2.6, 7.8 Hz, 1H), 8.08–8.15 (m, 3H), 8.58–8.60 (d, 2H), 8.58 (d, J = 1.4 Hz, 1H), 8.80 (s, 1H); 13C NMR (CD3OD) δ 29.9, 31.8, 40.6, 41.1, 45.7, 57.8, 63.9, 121.1, 125.2, 138.4, 141.4, 142.0, 147.0, 150.1, 158.7, 161.8; MS (ESI) m/z 270.3 (M + H)+.

A solution of 8a (246 mg, 0.91 mmol) in chloroform (2 mL) was placed in vial and treated with 1.1 equiv of fumaric acid (0.65 M in MeOH). After 24 h, the white solid obtained was recrystallized from MeOH using Et2O to provide the salt 8a·0.5C4H4O4 as a white solid: mp 155–159 °C. 1H NMR (300 MHz, CD3OD) δ 1.86–2.22 (m, 6H), 2.44–2.51 (dd, J = 9.0, 11.0 Hz, 1H), 3.49–3.54 (dd, J = 3.0, 5.1 Hz, 1H), 4.35 (br s, 1H), 4.56 (d, J = 3.9 Hz, 1H), 6.63 (s, 1H), 7.56–7.60 (dd, J = 2.3, 7.5 Hz, 1H), 8.12–8.16 (m, 2H), 8.23 (s, 1H), 8.61 (dd, J = 1.4, 6.0 Hz, 1H), 8.81 (s, 1H); 13C NMR (CD3OD) δ 27.4, 29.4, 38.2, 43.8, 59.8, 64.0, 121.0, 125.4, 136.8, 138.2, 138.5, 141.4, 147.0, 147.2, 150.0, 159.1, 162.3, 171.5; MS (ESI) m/z 270.2 [(M – fumaric)+, M = C16H16FN3·0.5C4H4O4]. Anal. (C18H18FN3O2·0.5 H2O) C, H, N.

2-exo-[2′-Fluoro-3′-(6-fluoropyridin-3-yl)-5′-pyridinyl]-7-azabicyclo[2.2.1]heptane (8b) Hemifumarate

Compound 11b (250 mg, 0.65 mmol, 1.0 equiv) was placed in a plastic vessel and was treated dropwise with 1.5 mL of 70% HF in pyridine, and the mixture was stirred at 0 °C for 30 min. Sodium nitrite (449 mg, 10 equiv) was added in small portions, and the reaction mixture was stirred at room temperature. After 1 h, the mixture was poured into a 1:1 aqueous solution of NH4OH–H2O (40 mL) and extracted with EtOAc (3 × 40 mL). The combined organic layers were dried over anhydrous Na2SO4, filtered through Celite, and concentrated in vacuo. The resultant residue was purified by flash chromatography on silica gel using CHCl3–MeOH as the eluent to provide 170 mg (91%) of 8b as a colorless oil. 1H NMR (300 MHz, CDCl3) δ 1.54–1.70 (m, 6H), 1.92–1.99 (dd, J = 9.0, 11.2 Hz, 1H), 2.82–2.87 (m, 1H), 3.61 (s, 1H), 3.83 (br s, 1H), 7.04 (dd, J = 3.0, 8.4 Hz, 1H), 7.99–8.09 (m, 2H), 8.14 (br s, 1H), 8.42 (d, J = 0.8 Hz, 1H); 13C NMR (CDCl3) δ 30.3, 31.5, 40.6, 44.3, 56.4, 62.9, 109.3, 118.5, 139.5, 141.3, 145.8, 147.5, 157.3, 160.4, 161.7, 164.9; MS (ESI) m/z 288.3 (M + H)+.

A solution of 8b (198 mg, 0.69 mmol) in chloroform (2 mL) was placed in a vial and treated with 1.1 equiv of fumaric acid (0.65 M in MeOH). After 24 h, the white solid obtained was recrystallized from MeOH using Et2O to provide 200 mg (84%) of the salt 8b·0.5C4H4O4 as a white crystalline solid: mp 197–199 °C. 1H NMR (500 MHz, CD3OD) δ 1.81–2.15 (m, 5H), 2.38–2.43 (dd, J = 9.3, 13.2 Hz, 1H), 3.42–3.46 (m, 1H), 4.43 (br s, 1H), 6.57 (s, 1H), 7.21 (dd, J = 2.4, 8.3 Hz, 1H), 8.14 (dd, J = 2.4, 8.2 Hz, 1H), 8.21–8.25 (m, 2H), 8.48 (br s, 1H); 13C NMR (CD3OD) δ 27.5, 29.5, 38.3, 43.8, 59.9, 64.1, 111.0, 120.5, 137.0, 141.4, 143.8, 147.2, 148.8, 159.7, 161.6, 164.1, 166.0, 174.0; MS (ESI) m/z 288.3 [(M – fumaric)+, M = C16H15F2N3·0.5C4H4O4]. Anal. (C18H17F2N3O2) C, H, N.

2-exo-[2′-Fluoro-3′-(6-chloropyridin-3-yl)-5′-pyridinyl]-7-azabicyclo[2.2.1]heptane (8c) Fumarate

Compound 11c (300 mg, 0.75 mmol, 1.0 equiv) was placed in a plastic vessel and was treated dropwise with 3 mL of 70% HF in pyridine, and the mixture was stirred at 0 °C for 30 min. Sodium nitrite (559 mg, 10 equiv) was added in small portions, and the reaction mixture was stirred at room temperature. After 1 h, the mixture was poured into a 1:1 aqueous solution of NH4OH–H2O (40 mL) and extracted with EtOAc (3 × 40 mL). The combined organic layers were dried over anhydrous Na2SO4, filtered through Celite, and concentrated in vacuo. The resultant residue was purified by flash chromatography on silica gel using CHCl3–MeOH as the eluent to provide 142 mg (62%) of 8c as a colorless oil. 1H NMR (300 MHz, CDCl3) δ 1.54–1.71 (m, 6H), 1.92–1.98 (dd, J = 9.1, 11.2 Hz, 1H), 2.81–2.86 (m, 1H), 3.61 (s, 1H), 3.81 (br s, 1H), 7.42 (dd, J = 0.6, 8.3 Hz, 1H), 7.88 (ddd, J = 0.8, 4.1, 8.3 Hz, 1H), 8.06 (dd, J = 2.4, 9.6 Hz, 1H), 8.15 (br s, 1H), 8.58 (br s, 1H); 13C NMR (CDCl3) δ 30.4, 31.5, 40.6, 44.3, 56.4, 62.9, 118.5, 124.1, 129.2, 139.5, 141.3, 146.1, 149.2, 151.2, 157.3, 160.5; MS (ESI) m/z 304.3 (M + H)+.

A solution of 8c (138 mg, 0.46 mmol) in chloroform (2 mL) was placed in a vial and treated with 1.1 equiv of fumaric acid (0.65 M in MeOH). After 24 h, the white solid obtained was recrystallized from MeOH using Et2O to provide 105 mg (55%) of the salt of 8c·0.5C4H4O4 as a white crystalline solid: mp 194–195 °C. 1H NMR (500 MHz, CD3OD) δ 1.89–2.20 (m, 5H), 2.45–2.49 (dd, J = 9.2, 13.2 Hz, 1H), 3.49–3.52 (dd, J = 3.5, 9.5 Hz, 1H), 4.34 (br s, 1H), 4.56 (d, J = 3.5 Hz, 1H), 6.63 (s, 2H), 7.60 (d, J = 8.5 Hz, 1H), 8.09–8.15 (m, 2H), 8.23 (d, J = 2.4 Hz, 1H), 8.64 (br s, 1H); 13C NMR (CD3OD) δ 27.1, 29.1, 37.8, 43.5, 60.3, 64.2, 120.4, 125.8, 130.6, 136.3, 137.8, 141.3, 147.4, 150.6, 152.6, 159.8, 161.7, 171.5; MS (ESI) m/z 304.5 [(M – fumaric)+, M = C16H15ClFN3·C4H4O4]. Anal. (C20H19ClFN3O4) C, H, N.

2′-Fluoro-3′-(2″-amino-5″-pyridinyl)deschloroepibatidine (8d) Hydrochloride

To a resealable reaction pressure vessel under nitrogen was added 2-exo-(2′-fluoro-3′-bromo)-7-azabicyclo[2.2.1]heptane (10) (125 mg, 0.46 mmol, 1.0 equiv), Pd(PPh3)4 (27 mg, 5 mol %), K2CO3 (128 mg, 0.92 mmol, 2.0 equiv), 1,4-dioxane (10 mL), water (0.80 mL), and 2-aminopyridine-5-pinacolate boronic ester (122 mg, 0.55 mmol, 1.2 equiv). The mixture was degassed through bubbling nitrogen for 40 min and heated at 110 °C for 18 h. After cooling, the solvent was removed under reduced pressure, and to the residue was added 20 mL of H2O. The organic product was extracted using EtOAc (3 × 30 mL). The combined organic layers were dried over Na2SO4 and filtered through Celite, and the solvent was removed in vacuo. Purification by flash chromatography on silica gel using MeOH–CHCl3 as the eluent provided 88 mg (67%) of the desired product 8d as a colorless oil. 1H NMR (300 MHz, CDCl3) δ 1.47–1.67 (m, 5H), 1.85–1.92 (m, 2H), 2.76–2.80 (dd, J = 3.8, 5.0 Hz, 1H), 3.56 (s, 1H), 3.75 (d, J = 2.7 Hz, 1H), 4.82 (s, 2H), 6.53 (d, J = 8.6 Hz, 1H), 7.63 (dt, J = 1.9, 8.6 Hz, 1H), 7.87 (dd, J = 2.3, 9.5 Hz, 1H), 7.98 (s, 1H), 8.23 (s, 1H); 13CNMR (CDCl3) δ 30.2, 31.4, 40.5, 44.5, 56.4, 62.8, 108.2, 120.2, 138.0, 138.7, 140.7, 144.2, 147.9, 157.5, 158.3, 160.6; MS (ESI) m/z 285.7 (M + H)+.

A solution of the diamine 8d (217 mg, 0.76 mmol) in chloroform in a vial was treated with a 2.0 equiv solution of HCl in diethyl ether and allowed to stand at room temperature. The excess solvent was filtered off, and the obtained salt washed with ether and then dried to provide 246 mg (90%) of 8d·HCl as a white solid: mp 202–205 °C. 1H NMR (300 MHz, CD3OD) δ 1.88–2.24 (m, 5H), 2.44–2.52 (dd, J = 3.8, 9.6 Hz, 1H), 3.51–3.56 (dd, J = 3.1, 5.5 Hz, 1H), 4.37 (d, J = 3.4 Hz, 1H), 4.58 (d, J = 2.7 Hz, 1H), 7.11 (dd, J = 1.9, 8.2 Hz, 1H), 8.18–8.28 (m, 4H); 13CNMR (CD3OD) δ 26.8, 28.9, 37.6, 43.3, 60.5, 64.4, 114.7, 119.3, 120.4, 137.6, 140.6, 145.1, 147.2, 155.8, 158.9, 162.1; MS (ESI) m/z 285.6 [(M – HCl)+, M = C16H17FN4·2HCl]. Anal. (C16H19Cl2FN4·1.25 H2O) C, H, N.

2-exo-[2′-Fluoro-3′-(6-methoxypyridin-3-yl)-5′-pyridinyl]-7-azabicyclo[2.2.1]heptane (8e) Hemifumarate

Compound 11d (480 mg, 1.21 mmol, 1.0 equiv) was placed in a plastic vessel and was treated dropwise with 3 mL of 70% HF in pyridine, and the mixture was stirred at 0 °C for 30 min. Sodium nitrite (835 mg, 10 equiv) was added in small portions, and the reaction mixture was stirred at room temperature. After 1 h, the mixture was poured into a 1:1 aqueous solution of NH4OH–H2O (100 mL) and extracted with EtOAc (3 × 100 mL). The combined organic layers were dried over anhydrous Na2SO4, filtered through Celite, and concentrated in vacuo. The resultant residue was purified by flash chromatography on silica gel using CHCl3–MeOH as the eluent to provide 227 mg (94%) of 8e as a colorless oil. 1H NMR (300 MHz, CD3OD) δ 1.48–1.76 (m, 6H), 1.94–2.05 (m, 2H), 2.96–3.01 (dd, J = 3.4, 5.5 Hz, 1H), 3.65 (s, 3H), 3.77 (br s, 1H), 6.83 (d, J = 8.7 Hz, 1H), 7.88 (tt, J = 0.7, 1.7, 8.7 Hz, 1H), 7.99 (dd, J = 2.4, 9.6 Hz, 1H), 8.04 (d, J = 0.8 Hz, 1H), 8.34 (d, J = 1.6 Hz, 1H); 13C NMR (CD3OD) δ 29.9, 31.7, 40.9, 45.7, 54.3, 57.7, 63.7, 111.7, 121.3, 124.5, 140.8, 141.6, 145.8, 147.9, 158.6, 161.8, 165.5; MS (ESI) m/z 300.3 (M + H)+.

A solution of 8e (169 mg, 0.53 mmol) in CH2Cl2 in a vial was treated with a 1.2 equiv of fumaric acid (0.65 M) in MeOH, and the vial was allowed to stand in a refrigerator overnight. The excess solvent was removed in vacuo from the salt that was then redissolved in a minimal amount of MeOH, and the fumarate salt was recrystallized from MeOH using diethyl ether to provide 159 mg of the salt 8e·0.5C4H4O4: mp 193–195 °C. 1H NMR (300 MHz, methanol-d4) δ 1.80–2.15 (m, 6H), 2.36–2.43 (dd, J = 9.3, 13.2 Hz, 1H), 3.40–3.45 (m, 1H), 3.96 (s, 3H), 4.27 (br s, 1H), 4.42 (s, 1H), 6.61 (s, 1H), 6.91 (dd, J = 0.7, 7.6 Hz 1H), 7.95 (dt, J = 0.8, 2.4, 8.8 Hz, 1H), 8.06 (dd, J = 1.9, 8.8 Hz, 1H), 8.14 (d, J = 1.9 Hz, 1H), 8.41 (br s, 1H); 13C NMR (methanol-d4) δ 26.9, 29.0, 37.7, 43.4, 54.3, 60.2, 64.1, 111.7, 124.2, 136.2, 137.4, 140.6, 140.8, 145.8, 148.0, 159.1, 162.3, 165.8, 171.3; MS (ESI) m/z 300.5 [(M – fumaric)+, M = C17H18FN3O·0.5C4H4O4]. Anal. (C19H20FN3O3·0.25 H2O) C, H, N.

7-tert-Butoxycarbonyl-2-exo-[2′-amino-3′-(pyridin-3-yl)-5′-pyridinyl]-7-azabicyclo[2.2.1]heptane (11a)

A solution of compound 9 (322 mg, 0.87 mmol, 1.0 equiv), pyridine-3-boronic acid (140 mg, 1.14 mmol, 1.3 equiv), Pd(PPh3)4 (50 mg, 0.044 mmol, 5 mol %), and K2CO3 (242 mg, 1.75 mmol, 2.0 equiv) in toluene (10 mL), EtOH (2 mL), and water (2 mL) was degassed through bubbling nitrogen for 20 min. The mixture was sealed and heated over a sand bath at 110 °C for 22 h. After cooling to room temperature, 20 mL of H2O was added, and the organic product was extracted with EtOAc (3 × 30 mL). The combined organic layers were dried over MgSO4, filtered through Celite, and concentrated in vacuo. The resultant residue was purified by flash chromatography on silica gel using i-PrOH–hexanes as the eluent to provide 263 mg (82%) of 11a. 1H NMR (300 MHz, CDCl3) δ 1.41 (br s, 9H), 1.48–1.61 (m, 2H), 1.75–1.86 (m, 3H), 1.96–2.04 (m, 1H), 2.79–2.83 (dd, J = 3.8, 5.0 Hz, 1H), 4.16 (s, 1H), 4.35 (br s, 1H), 4.66 (s, 2 NH), 7.34 (d, J = 2.5 Hz, 1H), 7.38 (d, J = 4.9 Hz, 1H), 7.80 (dt, J = 7.9, 1.9 Hz, 1H), 7.96 (d, J = 2.2 Hz, 1H), 8.59 (dd, J = 4.9, 1.6 Hz, 1H), 8.69 (d, J = 1.8 Hz, 1H); 13C NMR (CDCl3) δ 28.3 (3 C), 28.8, 29.7, 40.3, 44.9, 55.9, 62.2, 79.5, 118.0, 123.6, 132.1, 134.1, 136.2, 136.9, 146.6, 148.9, 149.7, 154.5, 154.9; MS (ESI) m/z 367.6 (M + H)+.

7-tert-Butoxycarbonyl-2-exo-[2′-amino-3′-(6-fluoropyridin-3-yl)-5′-pyridinyl]-7-azabicyclo[2.2.1]heptane (11b)

A solution of compound 9 (241 mg, 0.65 mmol, 1.0 equiv), 5-fluoropyridine-4-boronic acid (148 mg, 1.05 mmol, 1.6 equiv), Pd(OAc)2 (15 mg, 0.065 mmol, 10 mol %), P(o-tolyl)3 (40 mg, 0.131 mmol, 20 mol %), and Na2CO3 (139 mg, 1.31 mmol, 2.0 equiv) in DME (8 mL) and water (0.9 mL) was degassed through bubbling nitrogen for 20 min. The mixture was sealed and heated over a sand bath at 80 °C for 5 h. After cooling to room temperature, the mixture was poured into a saturated aqueous solution of NaHCO3 (20 mL) and extracted with EtOAc (3 × 30 mL). The combined organic layers were dried over MgSO4, filtered through Celite, and concentrated in vacuo. The resultant residue was purified by flash chromatography on silica gel using EtOAc–hexanes as the eluent to provide 250 mg (99%) of 11b. 1H NMR (300 MHz, CDCl3) δ 1.39 (br s, 9H), 1.51–1.59 (m, 2H), 1.81–1.85 (m, 3H), 1.94–2.00 (m, 1H), 2.79–2.84 (m, 1H), 4.16 (s, 1H), 4.35 (br s, 1H), 4.70 (s, 2 NH), 7.02 (dd, J = 2.9, 8.4 Hz, 1H), 7.34 (d, J = 2.25 Hz, 1H), 7.91 (ddd, J = 2.5, 8.4, 16 Hz, 1H), 7.96 (d, J = 2.25 Hz, 1H), 8.28 (d, J = 2.4 Hz, 1H); 13C NMR (CDCl3) δ 28.2 (3 C), 28.8, 29.7, 40.3, 44.8, 55.9, 62.1, 79.5, 109.5, 116.8, 132.0, 136.9, 141.5, 146.8, 147.5, 154.6, 154.9, 161.3, 164.5; MS (ESI) m/z 385.5 (M + H)+.

7-tert-Butoxycarbonyl-2-exo-[2′-amino-3′-(6-chloropyridin-3-yl)-5′-pyridinyl]-7-azabicyclo[2.2.1]heptane (11c)

A solution of compound 9 (304 mg, 0.83 mmol, 1.0 equiv), 5-chloropyridine-4-boronic acid (208 mg, 1.32 mmol, 1.6 equiv), Pd(OAc)2 (19 mg, 0.083 mmol, 10 mol %), P(o-tolyl)3 (51 mg, 0.166 mmol, 20 mol %), and Na2CO3 (176 mg, 1.66 mmol, 2.0 equiv) in DME (6 mL) and water (0.7 mL) was degassed through bubbling nitrogen for 20 min. The mixture was sealed and heated over a sand bath at 80 °C for 5 h. After cooling to room temperature, the mixture was poured into a saturated aqueous solution of NaHCO3 (20 mL) and extracted with EtOAc (3 × 30 mL). The combined organic layers were dried over MgSO4, filtered through Celite, and concentrated in vacuo. The resultant residue was purified by flash chromatography on silica gel using EtOAc–hexanes as the eluent to provide 305 mg (99%) of 11c as a colorless oil. 1H NMR (300 MHz, CDCl3) δ 1.31 (br s, 9H), 1.43–1.50 (m, 2H), 1.72–1.76 (m, 3H), 1.85–1.92 (m, 1H), 2.70–2.74 (m, 1H), 4.06 (s, 1H), 4.26 (br s, 1H), 4.60 (s, 2 NH), 7.25 (d, J = 2.25 Hz, 1H), 7.30 (d, J = 8.2 Hz, 1H), 7.71 (dd, J = 2.5, 8.2 1H), 7.88 (d, J = 2.2 Hz, 1H), 8.38 (d, J = 2.2 Hz, 1H); 13C NMR (CDCl3) δ 28.3 (3 C), 28.8, 29.7, 40.3, 44.8, 55.9, 62.1, 79.5, 116.7, 124.2, 132.2, 133.1, 136.9, 139.0, 147.0, 149.5, 150.6, 154.4, 155.0; MS (ESI) m/z 401.5 (M + H)+.

7-tert-Butoxycarbonyl-2-exo-[2′-amino-3′-(6-methoxypyridin-3-yl)-5′-pyridinyl]-7-azabicyclo[2.2.1]heptane (11d)

A solution of 9 (337 mg, 0.92 mmol, 1.0 equiv), 2-methoxypyridine-5-boronic acid (182 mg, 1.2 mmol, 1.3 equiv), Pd(PPh3)4 (53 mg, 0.046 mmol, 5 mol %), and K2CO3 (253 mg, 1.83 mmol, 2.0 equiv) in toluene (12 mL), EtOH (2 mL), and H2O (2 mL) was placed in a resealable pressure vessel and degassed through bubbling nitrogen for 20 min. The vessel was sealed and placed on a sand bath that was heated at 110 °C overnight. After cooling to room temperature, H2O (20 mL) was added, and the organic product was extracted with EtOAc (3 × 30 mL). The combined organic layers were dried over anhydrous MgSO4, filtered through Celite, and concentrated in vacuo. The residue was purified by flash chromatography on silica gel using EtOAc–hexanes as the eluent to furnish compound 11d (310 mg, 92%) as a colorless oil. 1H NMR (300 MHz, CDCl3) δ 1.39 (br s, 9H), 1.50–1.59 (m, 2H), 1.76–1.88 (m, 3H), 1.91–1.98 (m, 1H), 2.77–2.81 (dd, J = 3.7, 5.0 Hz, 1H), 3.94 (s, 3H), 4.16 (s, 1H), 4.34 (br s, 1H), 4.78 (s, 2 NH), 6.79 (d, J = 8.5 Hz, 1H), 7.30 (d, J = 2.3 Hz, 1H), 7.65 (dd, J = 8.4, 2.4 Hz, 1H), 7.93 (d, J = 2.3 Hz, 1H) 8.22 (d, J = 2.3 Hz, 1H); 13C NMR (CDCl3) δ 28.1 (3 C), 28.4, 28.6, 40.1, 44.8, 53.3, 55.3, 62.1, 79.3, 110.8, 118.1, 126.9, 128.4, 131.8, 136.7, 138.9, 145.8, 146.6, 154.8, 163.5; MS (ESI) m/z 397.5 (M + H)+.

7-tert-Butoxycarbonyl-2-exo-[2′-amino-3′-(pyridin-4-yl)-5′-pyridinyl]-7-azabicyclo[2.2.1]heptane (12a)

A solution of compound 9 (354 mg, 0 0.96 mmol, 1.0 equiv), pyridine-4-boronic acid (154 mg, 1.25 mmol, 1.3 equiv), Pd(PPh3)4 (56 mg, 0.048 mmol, 5 mol %), and K2CO3 (266 mg, 1.92 mmol, 2.0 equiv) in toluene (10 mL), EtOH (2 mL), and water (2 mL) was degassed through bubbling nitrogen for 20 min. The mixture was sealed and heated over a sand bath at 110 °C for 22 h. After cooling to room temperature, 20 mL of H2O was added, and the organic product was extracted with EtOAc (3 × 30 mL). The combined organic layers were dried over MgSO4, filtered through Celite, and concentrated in vacuo. The resultant residue was purified by flash chromatography on silica gel using i-PrOH–hexanes as the eluent to provide 340 mg (97%) of 12a. 1H NMR (300 MHz, CDCl3) δ 1.39 (br s, 9H), 1.44–1.59 (m, 2H), 1.81–1.84 (m, 3H), 1.93–2.00 (m, 1H), 2.79–2.84 (dd, J = 3.8, 5.0 Hz, 1H), 4.16 (s, 1H), 4.36 (br s, 1H), 4.67 (s, 2 NH), 7.39–7.43 (m, 3H), 7.99 (d, J = 2.3 Hz, 1H), 8.66 (dd, J = 6.0, 1.5 Hz, 1H); 13C NMR (CDCl3) δ 28.3 (3 C), 28.8, 29.7, 40.4, 44.8, 55.8, 62.1, 79.5, 118.7, 123.4 (2 C), 132.2, 136.5, 146.4, 147.2, 150.5 (2 C), 153.9, 154.9; MS (ESI) m/z 367.6 (M + H)+.

7-tert-Butoxycarbonyl-2-exo-[2′-amino-3′-(2-fluoropyridin-4-yl)-5′-pyridinyl]-7-azabicyclo[2.2.1]heptane (12b)

A solution of compound 9 (319 mg, 0.87 mmol, 1.0 equiv), 2-fluoropyridine-4-boronic acid (196 mg, 1.39 mmol, 1.6 equiv), Pd(OAc)2 (20 mg, 0.087 mmol, 10 mol %), P(o-tolyl)3 (53 mg, 0.173 mmol, 20 mol %), and Na2CO3 (184 mg, 1.73 mmol, 2.0 equiv) in DME (6 mL) and water (0.7 mL) was degassed through bubbling nitrogen for 20 min. The mixture was sealed and heated over a sand bath at 80 °C for 5 h. After cooling to room temperature, the mixture was poured into a saturated aqueous solution of NaHCO3 (20 mL) and extracted with EtOAc (3 × 30 mL). The combined organic layers were dried over MgSO4, filtered through Celite, and concentrated in vacuo. The resultant residue was purified by flash chromatography on silica gel using EtOAc–hexanes as the eluent to provide 300 mg (92%) of 12b as a colorless oil. 1H NMR (300 MHz, CDCl3) δ 1.4 (br s, 9H), 1.52–1.59 (m, 2H), 1.82–1.84 (m, 3H), 1.94–1.98 (m, 1H), 2.79–2.84 (m, 1H), 4.16 (s, 1H), 4.36 (br s, 1H), 4.77 (s, 2 NH), 7.06 (s, 1H), 7.34 (ddd, J = 1.6, 5.13, 8.4 Hz, 1H), 7.41 (d, J = 2.3 Hz, 1H), 8.0 (d, J = 2.3 Hz, 1H), 8.26 (d, J = 5.16 Hz, 1H); 13C NMR (CDCl3) δ 28.3 (3 C), 28.8, 29.7, 40.5, 44.8, 55.9, 62.1, 79.7, 108.8, 121.1, 132.5, 136.5, 147.8, 148.3, 152.0, 153.7, 155.0, 162.8, 166.0. MS (ESI) m/z 385.3 (M + H)+.

7-tert-Butoxycarbonyl-2-exo-[2′-amino-3′-(2-chloropyridin-4-yl)-5′-pyridinyl]-7-azabicyclo[2.2.1]heptane (12c)

A solution of compound 9 (192 mg, 0.52 mmol, 1.0 equiv), 2-chloropyridine-4-boronic acid (131 mg, 0.83 mmol, 1.6 equiv), Pd(OAc)2 (12 mg, 0.052 mmol, 10 mol %), P(o-tolyl)3 (32 mg, 0.104 mmol, 20 mol %), and Na2CO3 (111 mg, 1.04 mmol, 2.0 equiv) in DME (6 mL) and water (0.7 mL) was degassed through bubbling nitrogen for 20 min. The mixture was sealed and heated over a sand bath at 80 °C for 5 h. After cooling to room temperature, the mixture was poured into a saturated aqueous solution of NaHCO3 (20 mL) and extracted with EtOAc (3 × 30 mL). The combined organic layers were dried over MgSO4, filtered through Celite, and concentrated in vacuo. The resultant residue was purified by flash chromatography on silica gel using EtOAc–hexanes as the eluent to provide 112 mg (54%) of 12c as a colorless oil. 1H NMR (300 MHz, CDCl3) δ 1.41 (br s, 9H), 1.49–1.61 (m, 2H), 1.77–1.83 (m, 3H), 1.94–2.00 (m, 1H), 2.78–2.83 (m, 1H), 4.16 (s, 1H), 4.36 (br s, 1H), 4.54 (s, 2 NH), 7.37 (dd, J = 1.4, 5.13 Hz, 1H), 7.40 (d, J = 2.22 Hz, 1H), 7.45 (s, 1H), 8.0 (d, J = 2.22 Hz, 1H), 8.44 (d, J = 5.10 Hz, 1H); 13C NMR (CDCl3) δ 28.3 (3 C), 28.8, 29.7, 40.5, 44.8, 55.9, 62.1, 79.6, 117.4, 122.0, 123.8, 132.4, 136.5, 147.8, 149.6, 150.2, 152.4, 153.8, 154.9; MS (ESI) m/z 401.3 (M + H)+.

Preparation of 2-Methoxypyidine-4-boronic Acid Pinacol Ester (15)

A solution of 4-bromo-2-methoxypyridine (14) (462 mg, 2.46 mmol, 1.0 equiv), bis(pinacolato)diboron (749 mg, 2.95 mmol, 1.2 equiv), PdCl2(dppf) (54 mg, 0.074 mmol, 3 mol %), and KOAc (724 mg, 7.37 mmol, 3.0 equiv) in DMF (6 mL) in a resealable pressure vessel was degassed through bubbling nitrogen for 20 min. The reaction mixture was sealed and heated over a sand bath at 85 °C overnight. After cooling to room temperature, the mixture was diluted with EtOAc and filtered through a plug of Celite and anhydrous Na2SO4. The solvent was removed in vacuo, and the residue was purified by flash chromatography on silica gel using EtOAc–MeOH as the eluent to provide 427.4 mg (74%) of 15 as a brownish oil. 1H NMR (300 MHz, CDCl3) δ 1.34 (s, 12H), 3.93 (s, 3H), 7.13 (s, 1H), 7.18 (d, J = 5.0 Hz, 1H), 8.18 (d, J = 5.0 Hz, 1H).

[3H]Epibatidine Binding Assay

The inhibition of [3H]epibatidine binding at rat brain α4β2*-nAChRs was conducted as previously reported.[21]

Electrophysiology

The electrophysiology assays with rat α4β2-, α3β4, and α7-nAChRs were conducted as previously described.[24]

In Vivo Test

The antinociception (tail-flick and hot-plate), locomotor, and body temperature tests were all conducted in mice as previously described.[21]