Derivatives of dibenzothiophene for PET imaging of α7-Nicotinic Acetylcholine Receptors

Abstract

A new series of derivatives of 3-(1,4-diazabicyclo[3.2.2]nonan-4-yl)dibenzo[b,d]thiophene 5,5-dioxide with high binding affinities and selectivity for α7-nicotinic acetylcholine receptors (α7-nAChRs) (K_i = 0.4 – 20 nM) has been synthesized for PET imaging of α7-nAChRs. Two radiolabeled members of the series [¹⁸F]7a (K_i = 0.4 nM) and [¹⁸F]7c (K_i = 1.3 nM) were synthesized. [¹⁸F]7a and [¹⁸F]7c readily entered the mouse brain and specifically labeled α7-nAChRs. The α7-nAChR selective ligand 1 (SSR180711) blocked the binding of [¹⁸F]7a in the mouse brain in a dose-dependent manner. The mouse blocking studies with non-α7-nAChR CNS drugs demonstrated that [¹⁸F]7a is highly α7-nAChR selective. In agreement with its binding affinity the binding potential of [¹⁸F]7a (BP_ND = 5.3 – 8.0) in control mice is superior to previous α7-nAChR PET radioligands. Thus, [¹⁸F]7a displays excellent imaging properties in mice and has been chosen for further evaluation as a potential PET radioligand for imaging of α7-nAChR in non-human primates.

Introduction

Cerebral neuronal nicotinic cholinergic receptors (nAChRs) are ligand-gated ion channels composed of α (i.e., α2–α10) and β (i.e., β2–β4) subunits that can assemble in multiple combinations of pentameric structures. Among the many nAChRs subtypes in the human CNS, heteropentameric α4β2-nAChRs and homopentameric α7-nAChRs are predominant.^{1, 2} α7-nAChRs are composed of five identical α7 subunits, and each subunit provides an orthosteric binding site for its neurotransmitter, acetylcholine.³ Many lines of evidence associate α7-nAChRs with the pathophysiology of a variety of disorders such as schizophrenia and AD, anxiety, depression, traumatic brain injury, multiple sclerosis, inflammation and drug addiction.^4–12

Clinical experiments with α7-nAChR agonists have demonstrated that selective activation of the receptor is a viable approach towards improving cognitive performance in patients with schizophrenia.^{13, 14}

Because of the importance of the α7-nAChR in human neurophysiology and as a potential drug target, synthesis and pre-clinical examination of α7-nAChR subtype selective compounds receive substantial interest in industry and academia.^{9, 14} A number of α7-nAChR drugs are currently in various stages of the development for treatment of a variety of disorders including schizophrenia, AD, multiple sclerosis, depression, asthma and type-2 diabetes.^15–17

In vivo imaging and quantification of α7-nAChR binding in humans would provide a significant advance in the understanding of α7-nAChR-related CNS disorders and could also facilitate novel α7-nAChR drug development. Positron emission tomography (PET) is the most advanced technique to quantify neuronal receptors and their occupancy in vivo, and the development of a suitable PET radiotracer for α7-nAChRs would be of particular interest.

Many lead structures of α7-nAChR ligands have been identified within various structural classes. A number of these ligands have been radiolabeled for PET ([¹⁸F], [¹¹C]) and single-photon emission computed tomography (SPECT) ([¹²³I]) (Table 1) and studied in mice, pigs and non-human primates as potential α7- nAChR probes.^18–29 Most of these radioligands entered the animal brain, but manifested relatively low specific binding (see for review^30–32) and insufficient BP_ND values (BP_ND < 1) (Table 1). [¹¹C]CHIBA-1001 is the only α7-nAChR PET radioligand so far that has been studied in human subjects,²⁵ but it also exhibits low specific binding (Table 1).

Table 1.

In vitro properties and binding potential in cortex (BP_ND) of the previously published PET/SPECT radioligands for imaging of α7-nAChR.

Radioligand	α7-nAChR, Ki, nM	BPNDa		References
		Mice	Monkey or pig
	0.26	0.6	-	18
	n/a	~ 0	-	21
	10.8	0.2 – 0.5	0.3	22
	11	0.6 – 0.7	0.5	22
	0.24, 1.53	0.5	-	23
	46, 120, 193	~ 0.1	0.6	19, 25, 38, 39
	40.6	0.4	0.4	20
	0.092	low brain uptake	low brain uptake	23
	0.5 – 0.6	1.9	-	24
	2.5	low brain uptake	-	28
	24.9	-	0.7	19
	2.2	-	~ 1	26
	11.6	0.4	0.8	29
	0.2	0.8	-	27, 40

^aThe BP_ND values in cortex were taken directly from the corresponding references or estimated as V_T/V_ND − 1 or (cortex uptake/cerebellum uptake) − 1.^{41, 42}

Due to the exceptionally low concentration (B_max) of cerebral α7-nAChR binding sites in the human (5–15 fmol/mg protein)³³ and animal brain (1.5 – 12 fmol/mg tissue),^{34, 35} a PET radioligand with high specific brain uptake for this receptor subtype must exhibit very high binding affinity and selectivity, along with other important properties (e.g., lipophilicity, polar surface area, suitability for radiolabeling) in an appropriate range (see for details^{30, 32, 36, 37}). The general aptness of a PET radioligand for quantitative imaging studies is defined by a conventional criterion B_max/K_D ≥ 10.³⁷ This equation predicts that a picomolar range of the binding affinity is required for a good α7-nAChR PET radioligand (K_D ≤ 0.15 – 1.2 nM), whereas the most previously published α7-nAChR radioligands exhibited nanomolar binding affinities (Table 1). It is noteworthy, however, that the inhibition binding assays of the published compounds have been performed under a variety of assay conditions and, thus, the values of K_i listed in Table 1 may not be directly comparable to one another (see Discussion for details).

Recently Abbott Laboratories has reported 3-(1,4-diazabicyclo[3.2.2]nonan-4-yl)dibenzo[b,d]thiophene 5,5-dioxide 5 (Figure 2) as an α7-nAChR selective antagonist with extraordinarily high binding affinity, K_i = 0.023 nM.⁴³ We envisioned the synthesis of fluoro-derivatives of 5 producing a new set of compounds with similar or even better binding affinities with the potential for radiolabeling with [¹⁸F] for PET imaging. The rationale for the design of fluorine-bearing analogs of compound 5 has been strengthened by reports that fluorine substituents can increase the metabolic stability and the rate and extent of blood-brain barrier penetration of radiotracers.⁴⁴

Figure 2.

3-(1,4-diazabicyclo[3.2.2]nonan-4-yl)dibenzo[b,d]thiophene 5,5-dioxide 5, an α7-nAChR antagonist with very high binding affinity⁴³ that was used as the lead compound in this report.

In this study, we describe the design, synthesis and in vitro and in vivo characterization in mice of a series of high α7-nAChR binding affinity derivatives of 5 as potential probes for PET imaging of α7-nAChR receptor.

Results and Discussion

Chemistry

Synthesis of α7-nAChR ligands

The fluoro-derivatives 7a–e of 3-(1,4-diazabicyclo[3.2.2]nonan-4-yl)-dibenzo[b,d]thiophene 5,5-dioxide 5 were synthesized via the Buchwald-Hartwig cross-coupling reaction between the respective fluoro-bromo compounds 6a–e with 1,4-diazabicyclo[3.2.2]nonane (Scheme 1).

Scheme 1.

(a) Pd₂(dba)₃, rac-BINAP, toluene, 1,4-diazabicyclo[3.2.2]nonane, Cs₂CO₃, 85 °C, 24 h.

The nitro-derivatives of (1,4-diazabicyclo[3.2.2]nonan-4-yl)-dibenzo[b,d]thiophene 5,5-dioxide 10 and 11 were synthesized similarly starting with respective nitro-bromo-dibenzothiophene derivatives 8 and 9 (Scheme 2). Reduction of nitro groups in 10 and 11 with iron powder gave corresponding anilines 12 and 13 in high yield. Diazotization-iodination of 12 and 13 yielded corresponding iodides 14 and 15 (Scheme 2).

Scheme 2.

(a) Pd₂(dba)₃, rac-BINAP, toluene, 1,4-diazabicyclo[3.2.2]nonane, Cs₂CO₃, 85 °C, 24 h; (b) iron powder, NH₄Cl, THF, MeOH, water, 80 °C, 3 h; (c) (i) 4N H₂SO₄, CH₃CN, NaNO₂, −5 °C, 30 min; (ii) NaI, CuI, water 70 °C, 30 min.

Synthesis of intermediate compounds

The synthesis of intermediate fluoro-bromide 6a was performed in four steps (Scheme 3). Coupling of commercially available 4-bromo-2-fluoro-nitrobenzene 16 and 2-fluoro-thiophenol 17 gave nitrodiaryl thioether 18 that was reduced to aniline 19. Aniline 19 was treated with sodium nitrite at 0 °C in the presence of hydrochloric acid and sodium tetrafluoroborate to yield a corresponding diazonium tetrafluoroborate derivative (not shown). The intramolecular deazotisation/cyclization of the diazonium salt in the presence of copper (I) oxide and 0.1 N sulfuric acid afforded fluoro-bromo-dibenzothiophene derivative 20, that in turn was oxidized with hydrogen peroxide to 6a in high yield (Scheme 3).

Scheme 3.

Reagents and conditions: (a) Cs₂CO₃, DMF, 5 h; (b) iron powder, NH₄Cl, THF, MeOH, water, 80 °C, 3 h; (c)(i) NaNO₂, 50% HCl, 0–5 °C, 30 min; (ii) NaBF₄, 0–5 °C, 30 min; (iii) Cu₂O, 0.1 N H₂SO₄, 35–40 °C 30 min. (d) 30% H₂O₂, acetic acid, 60 °C, 24 h.

The fluoro-bromo isomers 6b and 6c were synthesized in four steps via the commercially available dibenzo[b,d]thiophene 5,5-dioxide 21 and 2-nitrodibenzo[b,d]thiophene 22, respectively (Scheme 4). In brief, nitration of compound 21 and oxidation of compound 22 gave compounds 23 and 24, respectively. Bromination of compounds 23 and 24 provided mono-bromo derivatives 25 and 9 that sequentially were reduced to anilines 26 and 27, respectively, in high yields. The anilines 26 and 27 were converted to fluorides 6b and 6c in moderate yields by the Schiemann reaction via the corresponding intermediate diazonium fluoroborates (structures not shown). The diazonium salts precipitated in the reaction mixture and were isolated by filtration in high yields.

Scheme 4.

Reagents and conditions: (a) H₂SO₄, HNO₃, 0°C; (b) 30% H₂O₂, acetic acid, 60 °C, 24 h; (c) NBS, conc. H₂SO₄ 24 h; (d) SnCl₂•2H₂O, 37% HCl, HOAc, 100 °C, 60 min or iron powder, NH₄Cl, THF, MeOH, water, 80 °C, 3 h; (f) (i) 48% HBF₄, 0–5 °C 10 min; (ii) NaNO₂, 0–5 °C, 1 h; (iii) xylene, 135 °C, 30 min.

The brominated isomers 6d and 6e were prepared by bromination of 4-fluorodibenzo[b,d]thiophene 5,5-dioxide 29 starting with 4-fluorodibenzo[b,d]thiophene 28.⁴⁵ Oxidation of 28 with hydrogen peroxide gave dioxide 29 in nearly quantitative yield. Bromination of 29 with one equivalent NBS in H₂SO₄ afforded two isomeric bromides: 6d as the main product in 24% yield as well as 6e as a minor product in 13% yield. A substantial amount of compound 29 (about 50%) was recovered from the reaction mixture. Isomers 6d and 6e were readily separated by silica gel chromatography.

3-Bromo-6-nitrodibenzo[b,d]thiophene 5,5-dioxide 8 was synthesized in two steps: (1) oxidation of 4-nitrodibenzo[b,d]thiophene 30⁴⁶ gave 4-nitrodibenzo[b,d]thiophene 5,5-dioxide 31 in 90% yield; (2) bromination of compound 31 provided compound 8 as the only product in 77% yield. (Scheme 6)

Scheme 6.

Reagents and conditions: (a) 30% H₂O₂, acetic acid, 60 °C, 24 h; (b) NBS, H₂SO₄, 24 h.

In vitro inhibition binding assay

The results of the α7-nAChR in vitro inhibition binding assays for compounds 7a–e, 10, 11, 14 and 15 are shown in Table 2. In order to determine α7-nAChR selectivity of new compounds vs. other nAChR subtypes, binding assays for the main cerebral heteromeric nAChR subtypes (α2β2-, α2β4-, α3β2-, α3β4-, α4β2-, α4β4-) were also performed (Table 2). In addition, because α7-nAChR shares 30% homology with the 5-HT₃ receptor and first generation α7-nAChR radioligands exhibited low α7-nAChR/5-HT₃ selectivity,²⁷ the in vitro binding affinity at the 5-HT₃ receptor was also determined for selected compounds of our series (Table 2).

Table 2.

Inhibition in vitro binding affinities (K_i, nM) of the new series 7a–e, 10, 11, 14 and 15 and reference compounds toward α7-nAChR, heteromeric nAChR subtypes and 5-HT₃.

Compound	α7-nAChRa	Heteromeric nAChR subtypes b						5-HT3 c	Selectivity
		α2β2	α2β4	α3β2	α3β4	α4β2	α4β4		α7/α4β2	α7/5HT3
7a	0.37, 0.45	>10000	4000	1000	709	562	1000	230	1370	561
7b	1.02, 1.37	ntd	nt	nt	nt	nt	nt	nt	-	-
7c	1.32, 1.35	1000	8000	2000	5000	885	3000	505	663	378
7d	1.83, 2.45	292	838	678	3000	141	1000	nt	66	-
7e	17.8, 20.3	>10000	562	2000	261	4000	251	nt	210	-
10	0.34, 0.35	nt	nt	nt	nt	nt	nt	nt	-	-
11	3.41, 6.21	nt	nt	nt	nt	nt	nt	nt	-	-
14	0.93, 1.93	nt	nt	nt	nt	nt	nt	nt	-	-
15	6.46, 8.77	784	6000	1000	9000	477	5000	nt	63	-

^aRat cortical membranes, radiotracer [¹²⁵I]α-bungarotoxin (0.1 nM), K_D = 0.7 nM

^bInhibition in vitro binding assay of all heteromeric nAChR subtypes was performed with stably transfected HEK293 cells and [³H]epibatidine (0.5 nM), K_D = 0.021 nM (α2β2-nAChR), K_D = 0.084 nM (α2β4-nAChR), K_D = 0.034 nM (α3β2-nAChR), K_D = 0.29 nM (α3β4-nAChR), K_D = 0.046 nM (α4β2-nAChR), K_D = 0.094 nM (α4β4-nAChR).⁵⁵

^cHuman 5-HT₃ recombinant/HEK293 cells, radiotracer [³H]GR65630 (0.35 nM), K_D = 0.5 nM

^dnt = not tested

The α7-nAChR assays for 7a–e, 10, 11, 14 and 15 were performed using a commercial assay consisting of rat cortical membranes (rich in α7-nAChR) in competition against 0.1 nM [¹²⁵I]α-bungarotoxin, an α7-nAChR antagonist with a K_D of 0.7 nM. These assays were performed independently in duplicate, each twice (Table 2). Assays for two reference compounds, methyllycaconitine (MLA), a conventional reference α7-nAChR antagonist, and compound 5,⁴³ a lead of our series, were also performed (Table 3).

Table 3.

Inhibition in vitro binding affinities (K_i, nM) of reference compounds toward α7-nAChR. The binding assay conditions are the same as those in the Table 2.

Compound	α7-nAChR
MLA	2.91 ± 0.76 (n = 9)
2	20.4
3	38.0
4	3.3
5	0.30, 0.50

The new series of fluoro isomers 7a–d exhibited high binding affinity at α7-nAChRs with K_i values in the range of 0.3 – 2.5 nM, whereas the binding affinity of isomer 7e was lower (Table 2). The K_i values of the fluoro derivatives 7a–d (Table 2) were better than that of the conventional reference α7-nAChR ligand MLA (Table 3). Among all fluoro isomers compound 7a manifested the best α7-nAChR binding affinity that was an order of magnitude better than MLA and at least comparable to the non-fluorinated lead 5 (Tables 2, 3).

Within the series 7a–e, two fluoro derivatives 7a and 7c were selected for further evaluation. This selection was based on the high α7-nAChR binding affinity and selectivity of 7a and 7c (see Table 2) and the suitability of these compounds for radiolabeling with [¹⁸F]. The radiolabeling of [¹⁸F]7a and [¹⁸F]7c was anticipated to be accomplished by a direct nucleophilic substitution (S_NAr) with [¹⁸F]fluoride via the nitro 10 and 11 or iodo derivatives 14 and 15, respectively. The leaving nitro- in 10 and 11 or iodo-groups in 14 and 15 are activated for S_NAr fluorination by the powerful electron-withdrawing SO₂Ar on the ortho and para positions, respectively.^47–50 We did not find in the literature an example of fluorination of nitro- or iododibenzothiophene 5,5-dioxides, but the structural analog of 11, 4,4'-sulfonylbis(p-nitrobenzene), has been converted to the corresponding fluoro-derivative with good yield.⁵¹

The fluoro derivative 7b that also exhibited high α7-nAChR binding affinity was not selected for further studies because the activating SO₂Ar was located on the meta position to the leaving group and direct radiolabeling of [¹⁸F]7b via its nitro or iodo-derivative was less likely. A detailed search of the literature for S_NAr reactions for 3-nitrodibenzo[b,d]thiophene 5,5-dioxide, a potential precursor for [¹⁸F]7b, or its structural analogs (1-nitro-3-(phenylsulfonyl)benzene, etc.) did not reveal any previous publications.

The potential precursors 10, 11, 14 and 15 for ¹⁸F-fluorination of [¹⁸F]7a and [¹⁸F]7c were studied in the same α7-nAChR inhibition binding assay. The nitro compounds 10 and 11 exhibited comparable α7-nAChR binding affinities to those of the corresponding fluorides 7a and 7c, whereas the binding affinities of iodo-derivatives 14 and 15 were lower.

Currently, there is no conventional in vitro competition binding assay for α7-nAChR. Different research groups use different radioligands ([¹²⁵I]α-bungarotoxin, [³H]α-bungarotoxin, [³H]MLA, [¹²⁵I]iodo-MLA, [³H]A-585539, etc) and different sources of receptor tissue (cell lines, brain, adrenal glands) under different conditions for this assay.^{26, 29, 52–54} It is not surprising that the difference in the K_i values for the same compound under different assay conditions can exceed an order of magnitude.^{53, 54} Therefore, a direct comparison of K_i values of the previously published α7-nAChR ligands with compounds of our new series is not practical.

For the purpose of comparison, we determined the K_i values of three most recently published α7-nAChR PET radioligands [¹¹C]2,²⁶ [¹⁸F]3 ²⁹ and [¹⁸F]4²⁷ (Table 3) under the same assay conditions as that of our series (Table 2). It was noteworthy that the α7-nAChR binding affinities of the best compounds of our series 7a and 7c were substantially better than those of the previous radioligands.

Heteromeric nAChR subtypes assays (α2β2-, α2β4-, α3β2-, α3β4-, α4β2-, α4β4-nAChR) were performed in our labs using membrane preparations from HEK293 cells expressing the transfected nAChR under test in competition with 0.5 nM [³H]epibatidine to investigate the specificity of the ligand for each receptor (Table 2).

The heteromeric nAChR K_i values of the tested compounds 7a, 7c–7e and 15 were substantially greater than the corresponding α7-nAChR K_i’s, indicating a high α7-/heteromeric-nAChR subtype selectivity of all studied compounds (Table 2). Thus, the fluoro isomer 7a with the best α7-nAChR binding affinity also manifested an excellent selectivity versus heteromeric nAChR including the main cerebral subtype α4β2-nAChR (Table 2). Interestingly, the α7/α4β2 selectivity of iodo derivative 15 is ten times lower than the corresponding fluoro derivative 7c.

5-HT₃ assay. The in vitro binding affinity of the most promising members of the series, compounds 7a and 7c, at the 5-HT₃ receptor were determined commercially using membrane preparations from HEK293 cells expressing transfected human 5-HT₃R in competition with 0.35 nM [³H]GR65630, a 5-HT₃R antagonist with a K_D of 0.5 nM. The assay demonstrated that fluoro compounds 7a and 7c manifest relatively low 5-HT₃ binding affinities and they are highly α7-nAChR/5HT₃ selective (Table 2).

Lipophilicity of 7a and 7c

Lipophilicity (logD_7.4) is considered an important property of CNS radioligand because it has been linked to the blood-brain barrier permeability and non-specific binding.^{35, 37, 38} The lipophilicity values for 7a and 7c (logD_7.4 = 2.0) were calculated with ACD Labs Structure Designer Suite (ACD Labs, Toronto) and fall within the conventional range for CNS PET radioligands.

Radiochemistry

We radiolabeled the fluoro isomers 7a and 7c that exhibited highest binding affinity within the series with fluorine-18. The radiosyntheses were performed remotely in one step by 1,10-diaza-4,7,13,16,21,24-hexaoxabicyclo[8.8.8]hexacosane (Kryptofix-222^®) - assisted radiofluorination of the respective nitro- 10 and 11 (Scheme 7) or iodo-precursors 14 and 15 using a radiochemistry synthesis module (Microlab, GE) followed by the semi-preparative HPLC separation and formulation of [¹⁸F]7a and [¹⁸F]7c as sterile apyrogenic solutions in 7% ethanolic saline.

Scheme 7.

Radiosynthesis of [¹⁸F]7a and [¹⁸F]7c

It is noteworthy that the radiotracer product yields from iodo-precursors 14 and 15 were substantially lower than those of the nitro-precursors 10 and 11. The conventional Kryptofix-222/potassium carbonate assisted radiofluorination of both iodo-derivatives 14 and 15 in DMSO at 130–180 °C produced [¹⁸F]7a and [¹⁸F]7c with radiochemical yields below 0.5% and this radiosynthesis pathway was not optimized further (not shown).

The radiofluorination of nitro-derivatives 10 or 11 (Scheme 7) in the presence of Kryptofix-222/potassium carbonate at 160 °C produced [¹⁸F]7a or [¹⁸F]7c in a slightly better yield (2–3%). Further optimization of this radiosynthesis suggested that both final products [¹⁸F]7a or [¹⁸F]7c rapidly decomposed in the DMSO reaction solution in the presence of highly basic K₂CO₃, but the radiochemical yield was improved if the less basic potassium oxalate was used. In the presence of potassium oxalate, the final products [¹⁸F]7a and [¹⁸F]7c were prepared under similar reaction conditions with comparable radiochemical yields of 16 ± 6% (n=14) (nondecay-corrected), with specific radioactivities in the range of 330 – 1260 GBq/µmol (9–34 Ci/µmol), and a radiochemical purity greater than 99%. The nitro-precursors 10 and 11 that exhibited substantial α7-nAChR binding affinity (Table 2) were fully separated by preparative HPLC and were not detected by analytical HPLC in the final products [¹⁸F]7a and [¹⁸F]7c (see Table 6 for the HPLC details).

Table 6.

HPLC conditions for [¹⁸F]7a and [¹⁸F]7c

	Column	Mobile phase	Flow rate, mL/min	Product retention time, min	Nitro precursor retention time, min
[18F]7a, preparative	XBridge C18 column, 10 µm (250 × 10 mm)	CH3OH/CH3CN/H2O/Me3N 260:120:620:2	12	32	21
[18F]7a, analytical	XBridge C18 column, 5 µm (250 × 4.6 mm)	CH3CN/H2O/Et3N 390:610:1	2	7.4	5.5
[18F]7c, preparative	XBridge C18 column, 10 µm (150 × 10 mm)	CH3CN/H2O/NH3 280:720:1	10	20	27
[18F]7c, analytical	XBridge C18 column, 3.5 µm (100 × 4.6 mm)	CH3CN/H2O/NH3 380:620:1	2	3.4	5.2

Biodistribution Studies of [¹⁸F]7a and [¹⁸F]7c in mice

Baseline studies in mice

Radioligands [¹⁸F]7a and [¹⁸F]7c were evaluated in mice as potential PET tracers for imaging α7-nAChRs. After intravenous injection, [¹⁸F]7a and [¹⁸F]7c exhibited robust initial brain uptake followed by washout. The highest accumulation of radioactivity of both radioligands occurred in the superior/inferior colliculus, hippocampus and frontal cortex. Moderate uptake was observed in thalamus and striatum and the lowest radioactivity was seen in cerebellum (Figs. 4–6). This distribution of radioactivity was similar to the previously published in vitro data on the distribution of α7-nAChRs in rodents.^{56, 57} The clearance rate of [¹⁸F]7a and [¹⁸F]7c from cerebellum was higher than that from any other regions studied. The ratios of tissues to cerebellum increased steadily over the 90 minutes, reaching values of 10 for [¹⁸F]7a and 4.5 for [¹⁸F]7c. The better ratios for compound [¹⁸F]7a vs. [¹⁸F]7c are in agreement with in vitro α7-nAChR binding affinity of these compounds (Table 2).

Figure 4.

Regional distribution of [¹⁸F]7a (Left) and [¹⁸F]7c (Right) in CD-1 mice. Data: mean %injected dose/g tissue ± SD (n = 3). Abbreviations: Coll = superior and inferior colliculus; Hipp = hippocampus; FrCtx = frontal cortex; Rest = rest of brain; Th = thalamus; Str = striatum; CB = cerebellum

Figure 6.

Blocking of [¹⁸F]7a and [¹⁸F]7c with α7-nAChR – selective ligands in CD-1 mice. Left panel: Dose dependent blockade of [¹⁸F]7a (0.07 mCi, specific radioactivity = 7900 mCi/µmol, i.v.) accumulation by intravenous co-injection with 1 (doses 0.02 mg/kg, 0.2 mg/kg, 1 mg/kg, 3 mg/kg) in the mouse brain regions 90 min after the injection. *P ≤ 0.01, significantly different from controls (ANOVA). Right panel: Dose dependent blockade of [¹⁸F]7c (0.07 mCi, specific radioactivity = 11000 mCi/µmol, i.v.) accumulation by intravenous co-injection with 5 (doses 0.001 mg/kg, 0.0045 mg/kg, 0.014 mg/kg) in the mouse brain regions 90 min after the injection. *P < 0.01, significantly different from controls; and **P = 0.06, insignificantly different from control (ANOVA). Data: mean %injected dose/g tissue ± SD (n=3). Abbreviations: Coll = superior and inferior colliculus; Hipp = hippocampus; Ctx = cortex; Str = striatum; Th = thalamus; Rest = rest of brain; CB = cerebellum.

Specificity and selectivity of [¹⁸F]7a and [¹⁸F]7c binding in the mouse brain

A conventional in vivo blockade methodology with CNS drugs is used here for demonstration of specificity and selectivity at the α7-nAChR receptor in the mouse brain. A self-blockade study with a non-radioactive form of a radioligand estimates whether or not the binding is specific. A blockade study with a drug that is highly selective at the target binding site is expected to show the selectivity and specificity of the radioligand binding. A dose-escalation blockade with such a target selective drug provides further evidence of the radioligand specificity and selectivity and it is useful for demonstration of the radioligand suitability for evaluation of conventional drug candidates. In addition, blockade with CNS drugs that do not bind at the target site provide more evidence of the radioligand selectivity versus other cerebral binding sites.

Self-blockade studies of [¹⁸F]7a with 7a (Fig. 5-Left) and [¹⁸F]7c with 7c (Fig. 5-Right) demonstrated a reduction of the radioligand uptake in most brain regions except the cerebellum, a region with low density of α7-nAChRs. The studies showed that accumulation of [¹⁸F]7a and [¹⁸F]7c radioactivity in the mouse brain was specific. When the specific binding of the radioligands in the hippocampus and colliculus was estimated by using the radioactivity concentration in the blocked cerebellum as nonspecific binding, the specific binding value amounted to 94% and 80% and the baseline-to-blockade ratio in the α7-nAChR-rich regions was 13 and 5 for [¹⁸F]7a and [¹⁸F]7c, respectively. This result also demonstrated that [¹⁸F]7a exhibited a higher level of specificity and greater uptake in the mouse brain versus [¹⁸F]7c.

Figure 5.

Self-blockade study of [¹⁸F]7a and [¹⁸F]7c in CD-1 mice. Left panel: Inhibition of [¹⁸F]7a (0.07 mCi, specific radioactivity = 9200 mCi/µmol, i.v.) accumulation by intravenous co-injection with 7a (0 mg/kg (white) and 0.3 mg/kg (black)) in the mouse brain regions 90 min after the injection. *P < 0.01, significantly different from controls; **P = 0.04, insignificantly different from controls (ANOVA). Right panel: Inhibition of [¹⁸F]7c (0.07 mCi, specific radioactivity = 12000 mCi/µmol, i.v.) accumulation by intravenous co-injection with 7c (0 mg/kg (white) and 0.2 mg/kg (black)) in the mouse brain regions 90 min after the injection. *P ≤ 0.01, **P = 0.015, significantly different from controls; and ***P = 0.5, insignificantly different from controls (ANOVA). Data: mean %injected dose/g tissue ± SD (n=3). Abbreviations: Coll = superior and inferior colliculus; Hipp = hippocampus; FrCtx = frontal cortex; Str = striatum; Rest = rest of brain, CB = cerebellum.

Neither behavioral nor locomotor activity changes were observed in the mice in the blockade studies with 7a (0.3 mg/kg, i.v.) or 7c (0.2 mg/kg, i.v.).

Blocking with selective α 7-nAChR ligands. A blockade study of [¹⁸F]7a with 1, a selective α7-nAChR partial agonist with a K_i of 22 nM,⁵⁸ showed a dose dependent blockade in all regions studied. However, in the α7-nAChR-poor cerebellum, the blockade was significant only with the highest dose of 1 (3 mg/kg) (Fig. 6-Left). A similar dose-response study was performed with [¹⁸F]7c using compound 5, a selective α7-nAChR antagonist, as a blocker (Fig. 6-Right). These studies confirmed that the in vivo binding of [¹⁸F]7a and [¹⁸F]7c was specific and mediated by α7-nAChR. The dose-escalation response demonstrated that both radioligands are suitable tools for evaluation of new α7-nAChR drug candidates.

It is noteworthy that the doses of 1 that significantly blocked the [¹⁸F]7a binding in CD1 mice were comparable to the doses of 1 that significantly improved cognitive deficit in the various rodent models of schizophrenia.^{59, 60} This finding suggests that [¹⁸F]7a is suitable radioprobe for in vivo studies in mice with pharmacologically relevant doses of α7-nAChR drugs.

Because the lowest regional uptake of [¹⁸F]7a and [¹⁸F]7c was seen in the cerebellum the regional BP_ND values in mice were approximated for a single time point measurement (90 min) as BP_ND = (regional uptake/cerebellum uptake) – 1⁴² (Table 4). The substantially higher BP_ND values for [¹⁸F]7a are in agreement with greater binding affinity of this compound versus [¹⁸F]7c (Table 2; also see Fig. 9). The BP_ND values for both radioligands [¹⁸F]7a and [¹⁸F]7c were superior to all previously published α7-nAChR PET radioligands (Table 1)

Table 4.

Approximate BP_ND values (unitless) of [¹⁸F]7a and [¹⁸F]7c in the mouse brain regions. Data: mean ± SD (n = 6)

Region	Coll	Hipp	Ctx
Compound
[18F]7a	8.0 ± 1.6	5.5 ± 1.7	5.3 ± 1.2
[18F]7c	2.0 ± 0.5	3.1 ± 0.7	2.0 ± 0.3

Abbreviations: Coll = superior and inferior colliculus; Hipp = hippocampus; Ctx = cortex

Figure 9.

Correlation of the BP_ND^cortex (unitless) vs. 1/K_i (nM⁻¹) of α7-nAChR PET radioligands [¹¹C]2, [¹⁸F]3, [¹⁸F]4, [¹⁸F]7a and [¹⁸F]7c (y = 1.91x + 0.52, R² = 0.98). The BP_ND values are shown in the Tables 1 and 3; the SD values are available for [¹⁸F]7a and [¹⁸F]7c only. All K_i values were obtained in this study under the same binding assay conditions (Tables 2 and 3).

Blocking with nicotine and α4β2-nAChR selective cytisine. The blockade of [¹⁸F]7a in CD1 mouse brain with cytisine, a partial nicotinic agonist selective for α4β2-nAChR and other β2/β4-containing heteromeric nAChR subtypes while exhibiting low α7-nAChR binding affinity,^{52, 55, 61} showed insignificant reduction of radioactivity accumulation in all regions studied (Fig. 7). This result demonstrated that [¹⁸F]7a manifested insignificant binding at α4β2-nAChRs in the mouse brain.

Figure 7.

Blockade of [¹⁸F]7a accumulation in CD-1 mouse brain regions by injection of cytisine (1 mg/kg, s.c.) and nicotine (5 mg/kg, s.c.) (both 90 min after the injection). Data: mean %injected dose/g tissue ± SD (n = 3). Abbreviations: Coll = superior and inferior colliculus; Hipp = hippocampus; Ctx = cortex; CB = cerebellum; Rest = rest of brain. The effect of cytisine was insignificant in all regions studied (P > 0.05, asterisk is not shown). The difference between control and nicotine was significant (*P ≤ 0.01) in all regions, except CB (**P = 0.9) (ANOVA). The study demonstrates that [¹⁸F]7a does not bind in vivo at the main cerebral α4β2-nAChR subtype and it is suitable for nicotine blockade studies.

The blockade study of [¹⁸F]7a with nicotine that binds at all nAChR subtypes including α7-nAChR⁵² showed significant blockade in all regions, except the nAChR-poor cerebellum. This suggests that [¹⁸F]7a can be used for nicotine addiction or smoking studies in mice. The lesser blockade of [¹⁸F]7a with nicotine (Fig. 7) in comparison with 1 (Fig. 6) is due to the rather modest binding affinity of nicotine at α7-nAChR (K_i = 610 nM).⁵²

Blocking with non-a7-nAChR CNS ligands. For determination of in vivo selectivity of [¹⁸F]7a for α7-nAChRs vs. several major CNS receptor systems, we compared the regional distribution (Fig. 8) of the radiotracer in control CD-1 mice vs. mice preinjected with various CNS active drugs or the positive control 1 (see Table 5 for the drug list). None of the drugs except 1 reduced accumulation of radioactivity when compared with controls (Fig. 8). The absence of blockade with the 5-HT₃-selective drug ondansetron was especially remarkable because α7-nAChR ligands often bind to this receptor subtype. This finding suggests that in the mouse brain the radioligand [¹⁸F]7a was selective for α7-nAChRs versus several major cerebral binding sites.

Figure 8.

Effect of various CNS drugs (Table 5) on accumulation of [¹⁸F]7a in CD-1 mouse brain regions 90 min after injection of tracer expressed as %ID/g tissue. Abbreviations: Coll = superior and inferior colliculus; Hipp = hippocampus; Ctx = cortex; CB = cerebellum; REST = rest of brain. Data are mean ± SD (n=3). *P < 0.01, significantly different from controls. Columns that do not include the asterisk are insignificantly different from controls (P > 0.05) (ANOVA, single-factor analysis). The graph demonstrates that unlike the positive control (1) all non-α7-nAChR CNS drugs do not have an effect on the cerebral uptake of [¹⁸F]7a and the radiotracer is α7-nAChR selective in vivo.

Table 5.

CNS drugs (2 mg/kg, s.c.) for α7-nAChR selectivity studies in mice

Drug	Target receptor	Dose (mg/kg)	Time of administration before radiotracer, min
1	Selective α7-nAChR partial agonist	2	10
Ondansetron	Selective 5-HT3 antagonist	2	10
SCH23390	D1- and D5-antagonist and 5-HT1C/2C agonist	2	10
Spiperone	D2-like and 5-HT2A receptor antagonist	2	10
Ketanserin	5-HT2/5-HT2C antagonist	2	10
Naltrindole	Selective δ-opioid antagonist	2	10

5-HT = 5-hydroxytryptoamine (serotonin)

Comparison of imaging properties of [¹⁸F]7a and [¹⁸F]7c with previous α7-nAChR PET radioligands

Binding potential (BP_ND), a measure of in vivo specific binding and one of the most important imaging characteristics of a PET radioligand, is defined as the ratio of B_max (receptor density) to K_D (radioligand equilibrium dissociation constant) or the product of B_max and binding affinity.^{41, 62} Therefore, the binding affinities (1/K_i) of α7-nAChR radioligands should correlate linearly with their BP_ND values.

The comparison of all previously published α7-nAChR radioligands (Table 1) revealed little correlation between 1/K_i and BP_ND (R² = 0.05, not shown). It was likely that the lack of correlation was due to the wide variability in binding assay conditions for these compounds when performed by various research groups (see Discussion above).

When the α7-nAChR binding assay for the radioligands is performed under the same assay conditions (Tables 2, 3), the binding affinities 1/K_i correlate linearly (Fig. 9) with the cortical BP_ND’s of [¹⁸F]7a and [¹⁸F]7c (Table 4) and [¹¹C]2, [¹⁸F]3 and [¹⁸F]4 (Table 1). This finding may explain why the specific binding of the very high affinity radioligands [¹⁸F]7a and [¹⁸F]7c is superior to the previous α7-nAChR radioligands with lower binding affinities. This result emphasizes further the importance of high binding affinity for the imaging properties of α7-nAChR radioligands.

Conclusion

A series of 3-(1,4-diazabicyclo[3.2.2]nonan-4-yl)dibenzo[b,d]thiophene 5,5-dioxide derivatives with high binding affinities for α7-nAChRs (K_i = 0.4 – 20 nM) has been synthesized with potential application for PET imaging of α7-nAChRs. Two members of the series, 7a and 7c, with the best α7-nAChR binding affinities (K_i = 0.4 and 1.3 nM, respectively) and high selectivity vs. other nicotinic subtypes and 5-HT₃, were radiolabeled with ¹⁸F.

[¹⁸F]7a and [¹⁸F]7c readily entered the mouse brain and specifically and selectively labeled cerebral α7-nAChR receptors. The binding potential (BP_ND) values in mouse cortex of [¹⁸F]7a, [¹⁸F]7c and previously published α7-nAChR radioligands correlated linearly with their binding affinities (1/K_i) when the binding affinity values were determined under the same assay conditions. In agreement with the binding affinity of [¹⁸F]7a its BP_ND value in mice was substantially better than those of the previous α7-nAChR radioligands.

The best PET radioligand of this new series [¹⁸F]7a exhibits excellent α7-nAChR imaging properties in the mouse brain. Therefore, [¹⁸F]7a holds promise as a highly specific PET radioligand for quantification of α7-nAChR receptors and further evaluation of this radioligand in baboon PET studies is underway.

Experimental Section

All reagents were used directly as obtained commercially unless otherwise noted. Reaction progress was monitored by TLC using silica gel 60 F254 (0.040−0.063 mm) with detection by UV. All moisture-sensitive reactions were performed under an argon atmosphere using oven-dried glassware and anhydrous solvents. Column flash chromatography was carried out using E. Merck silica gel 60F (230–400 mesh). Analytical thin-layer chromatography (TLC) was performed on aluminum sheets coated with silica gel 60 F₂₅₄ (0.25 mm thickness, E. Merck, Darmstadt, Germany). Melting points were determined with a Fisher-Johns apparatus and were not corrected. ¹H NMR spectra were recorded with a Bruker-400 NMR spectrometer at nominal resonance frequencies of 400 MHz, in CDCl₃ or DMSO-d₆ (referenced to internal Me₄Si at δ_H 0 ppm). The chemical shifts (δ) were expressed in parts per million (ppm). First order J values were given in Hertz. Splitting patterns are described as singlet (s), doublet (d), triplet (t), quartet (q), and broad (br). High resolution mass spectra were recorded utilizing electrospray ionization (ESI) at the University of Notre Dame Mass Spectrometry facility. All compounds that were tested in the biological assays were analyzed by combustion analysis (CHN) to confirm the purity >95%. Elemental analyses were determined by Galbraith Laboratories, Inc. (Knoxville, TN). The HPLC system consisted of two Waters model 600 pumps, two Rheodyne model 7126 injectors, an in-line Waters model 441 UV detector (254 nm), and a single sodium iodide crystal flow radioactivity detector. All HPLC chromatograms were recorded with Varian Galaxy software (version 1.8). The analytical and semi-preparative chromatographies were performed using Waters XBridge C-18 10 µm columns (analytical 4.6 × 250 mm and preparative 10 × 250 mm).

A dose calibrator (Capintec 15R) was used for all radioactivity measurements. Radiofluorination was performed with a modified GE MicroLab radiochemistry box.

Chemistry

Typical procedure for reduction of nitro derivatives to anilines 12, 13, 19, 26, 27

A mixture of nitro compound (1 mmol), iron powder (4 mmol), ammonium chloride (1.2 mmol) in methanol (6 mL), THF (6 mL) and water (2 mL) was heated to reflux (80 °C) for 3 hours. The resulted mixture was diluted with ethanol and concentrated and dried under vacuum. The residue was purified by silica gel column chromatography (CHCl₃/i-PrOH/Et₃N 10:1:0.1 to 10:30:4) to give the corresponding aniline derivative.

Typical procedure for bromination

N-bromosuccinimide (NBS) (1 mmol) was added to a solution of the starting 1,4-dibenzothiophene derivative (1 mmol) in concentrated H₂SO₄ (3.6 mL) at room temperature. After 24 h, the solution was carefully poured into ice/water. The solids were filtered and washed with water and methanol. The obtained solids were recrystallized from 95% EtOH to afford the bromo compounds.

Typical procedure for oxidation of 1,4-dibenzothiophene derivatives

1,4-Dibenzothiophene derivative (1 mmol) was dissolved in glacial acetic acid (2.8 mL) at room temperature. Aqueous hydrogen peroxide (30%, 1.4 mL) was added in small portions to the stirred solution. The addition of H₂O₂ resulted initially in some precipitation. The mixture was stirred at 60 °C for 24 h, then cooled to room temperature. The solid was filtered off, sequentially washed with 70% aqueous acetic acid, then 30% aqueous acetic acid, then water and dried to afford the title compound.

3-Bromo-6-fluorodibenzo[b,d]thiophene 5,5-dioxide (6a)

The typical procedure for oxidation of 1,4-dibenzothiophene was followed, starting with 20 (600 mg, 2.13 mmol). The title compound 6a (648 mg, 97%) was obtained as white crystals. ¹H NMR (CDCl₃, 400 MHz) δ 7.97 (s, 1H), 7.81 (dd, J=12.0, 1.8 Hz, 1H), 7.68 (d, J=8.0 Hz, 1H), 7.66 (dd, J=8.0, 4.0 Hz, 1H), 7.59 (d, J=8.0 Hz, 1H), 7.24 (t, J=8.0 Hz, 1H).

3-Bromo-7-fluorodibenzo[b,d]thiophene 5,5-dioxide (6b)

A mixture of 26 (620 mg, 2 mmol) and 48% tetrafluoroboric acid (HBF₄) (4 mL) was stirred at 0–5 °C for 10 min. A cold solution of sodium nitrite (204 mg in 0.8 mL of water, 3 mmol) was added dropwise with stirring. After stirring for 1 h at 0–5 °C the precipitated intermediate diazonium tetrafluoroborate was collected by filtration, washed with cold tetrafluoroboric acid (5%) and water, Et₂O and dried under vacuum. The diazonium tetrafluoroborate was boiled in xylene (135 °C) for 120 min. The solvent was evaporated under reduced pressure. The residue was extracted with mixture of chloroform and water. The chloroform layer was separated and concentrated. The residue was chromatographed on silica gel using hexanes-EtOAc (4:1) as eluent to give 6b as a pale yellow solid (330 mg, 53%). ¹H NMR (CD₃Cl, 400 MHz) δ 7.96 (d, J=2.0 Hz, 1H), 7.81-7.77 (m, 2H), 7.64 (d, J=8.0 Hz, 1H), 7.54 (dd, J=8.0, 4.0 Hz, 1H), 7.40 – 7.35 (m, 1H)

7-Bromo-2-fluorodibenzo[b,d]thiophene 5,5-dioxide (6c)

A mixture of 27 (310 mg, 1 mmol) and 48% tetrafluoroboric acid (HBF₄) (2 mL) was stirred at 0 –5 °C for 10 min. A cold solution of sodium nitrite (102 mg, 1.5 mmol) in 0.4 mL of water was added dropwise with stirring. After stirring for 1 h at 0 –5 °C, the precipitated diazonium tetrafluoroborate was collected by filtration, washed with cold tetrafluoroboric acid (5%) and water, Et₂O and dried under vacuum. The diazonium tetrafluoroborate was boiled in xylene (135 °C) for 30 min. The solvent was evaporated under reduced pressure. The residue was treated with chloroform and water. The chloroform layer was separated and concentrated. The residue was chromatographed on silica gel using hexanes-EtOAc (4:1) as eluent to give 6c as a pale yellow solid (156 mg, 50%). ¹H NMR (DMSO-d₆, 400 MHz) δ 8.38 (d, J=4.0 Hz, 1H), 8.23-8.18 (m, 2H), 8.13-8.07 (m, 2H), 7.54 (t, J=8.0 Hz, 1H).

3-Bromo-4-fluorodibenzo[b,d]thiophene 5,5-dioxide (6d) and 1-bromo-4-fluorodibenzo[b,d]thiophene 5,5-dioxide (6e)

The typical procedure for bromination was followed, starting with 29 (905 mg, 3.86 mmol). Separation of the crude reaction product by silica gel chromatography using hexanes/ethyl acetate (5:2) yielded two isomers 6d (285 mg, 0.91 mmol, 23.6%) and 6e (160 mg, 0.51 mmol, 13%). The isomer 6e was in the first chromatography fraction, whereas 6d was in the second fraction.

6d: R_f=0.31 (hexanes/EtOAc 2:1); ¹H NMR (CDCl₃, 400 MHz) δ 7.86-7.80 (m, 3H), 7.70 (t, J=8.0 Hz, 1H), 7.63 (d, J=8.0 Hz, 1H), 7.49 (d, J=8.0 Hz, 1H).

6e : R_f=0.5 (hexanes/EtOAc 2:1); ¹H NMR (CDCl₃, 400 MHz) δ 8.94 (d, J=8.0 Hz, 1H), 7.90 (d, J=8.0 Hz, 1H), 7.83-7.79 (m, 1H), 7.75 (dd, J=8.0, 4.0 Hz, 1H), 7.67 (d, J=8.0 Hz, 1H), 7.11 (t, J=8.0 Hz, 1H).

Typical procedure for Buchwald-Hartwig cross-coupling reaction

3-(1,4-Diazabicyclo[3.2.2]nonan-4-yl)-6-fluorodibenzo[b,d]thiophene 5,5-dioxide (7a)

A catalyst solution was prepared by mixing tris(dibenzylideneacetone)dipalladium (Pd₂(dba)₃; 58 mg, 0.063 mmol; Aldrich) and racemic BINAP (39 mg, 0.125 mmol; Strem) in toluene (4 mL) and heating the mixture to 90 °C for 15 min. The solution was cooled, and then added to a mixture of 1,4-diazabicyclo[3.2.2]nonane (200 mg, 1.58 mmol) and 6a (0.492 g, 1.58 mmol), in toluene (12 mL). Cs₂CO₃ (766 mg, 2.4 mmol; Aldrich) was then added, and the reaction mixture was flushed with nitrogen and heated overnight at 80–85 °C. After cooling to room temperature, the mixture was concentrated and purified by silica gel flash chromatography (CHCl₃:i-PrOH:Et₃N 10:1:0.2). The title compound 7a (227 mg, 40% yield) was obtained as yellow solid. ¹H NMR (DMSO-d₆, 400 MHz) δ 7.89 (d, J=8.0 Hz, 1H), 7.77 (d, J=8.0 Hz, 1H), 7.73 (t, J=8.0 Hz, 1H), 7.29-7.24 (m, 2H), 7.12 (d, J=8.0 Hz, 1H), 4.19 (s, 1H), 3.70-3.67 (m, 2H), 2.98-2.91 (m, 4H), 2.88-2.82 (m, 2H), 1.99 (m, 2H), 1.72-1.66 (m, 2H); HRMS calculated for C₁₉H₂₀FN₂O₂S ([M+H]) 359.1224; found, 359.1240.

Preparation of 7a•p-TSA salt

A mixture of 7a (30 mg, 0.084 mmol) and p-toluenesulfonic acid monohydrate (19 mg, 0.099 mmol) was stirred in EtOAc-EtOH (2 mL, 10:1) at room temperature for 2 h. The resulting solid was collected, washed with EtOAc-EtOH (2 mL, 10:1) and EtOAc (3 mL) and dried under vacuum to afford the title compound as a yellow solid (32 mg, 72% yield). ¹H NMR (DMSO-d₆, 400 MHz) δ 10.10 (s, 1H), 7.99 (d, J=8.0 Hz, 1H), 7.85 (d, J=8.0 Hz, 1H), 7.79-7.73 (m, 1H), 7.49-7.45 (m, 3H), 7.32 (t, J=8.0 Hz, 1H), 7.25 (dd, J=8.0, 4.0 Hz, 1H), 7.12 (br s, 1H), 7.10 (br s, 1H), 4.47 (s, 1H), 3.95-3.93 (m, 2H), 3.49-3.39 (m, 6H), 2.29 (s, 3H), 2.19 (m, 2H), 2.05 (m, 2H); Elemental Analysis for C₂₆H₂₇FN₂O₅S₂, Calcd: C, 58.85; H, 5.13; N, 5.28; Found: C, 58.57; H, 5.04; N, 5.18.

4-(7-Fluorodibenzo[b,d]thiophen-3-yl)-1,4-diazabicyclo[3.2.2]nonane 5,5-dioxide (7b)

The typical procedure for Buchwald-Hartwig cross-coupling reaction was followed, starting with 6b (0.2 g, 0.64 mmol), the title compound 7b was obtained as a yellow solid (104 mg, 0.29 mmol, 45% yield). ¹H NMR (DMSO-d₆, 400 MHz) δ 7.99 (dd, J=8.0, 4.0 Hz, 1H), 7.89 (dd, J=8.0, 3.0 Hz, 1H), 7.86 (d, J=8.0 Hz, 1H), 7.55 (m, 1H), 7.25 (d, J=4.0 Hz, 1H), 7.11 (dd, J=8.0, 4.0 Hz, 1H), 4.17 (s, 1H), 3.66 (m, 2H), 2.99-2.91 (m, 3H), 2.87-2.82 (m, 3H), 2.00-1.97 (m, 2H), 1.71-1.65 (m, 2H). Elemental Analysis for C₁₉H₁₉FN₂O₂S • 0.1 H₂O, Calcd: C, 62.37; H, 5.17; N, 7.58; Found: C, 62.25; H, 5.44; N, 7.19.

7-(1,4-Diazabicyclo[3.2.2]nonan-4-yl)-2-fluorodibenzo[b,d]thiophene 5,5-dioxide (7c)

The typical procedure for Buchwald-Hartwig cross-coupling reaction was followed, starting with 6c (0.226 g, 0.72 mmol) and the title compound 7c (140 mg, 54% yield) was obtained as a yellow solid. ¹H NMR (DMSO-d₆, 400 MHz) δ 7.93-7.87 (m, 3H), 7.26-7.21 (m, 2H), 7.12 (d, J=8.0 Hz, 1H), 4.20 (s, 1H), 3.71-3.68 (m, 2H), 3.01-2.83 (m, 6H), 2.00 (br s, 2H), 1.73-1.67 (m, 2H); HRMS calculated for C₁₉H₂₀FN₂O₂S ([M+H]) 359.1224; found, 359.1241.

TSA salt: ¹H NMR (DMSO-d₆, 400 MHz) δ 10.08 (s, 1H), 8.00-7.94 (m, 3H), 7.48 (d, J=8.0 Hz, 1H), 7.43 (m, 2H), 7.32-7.25 (m, 2H), 7.12 (br, 1H), 7.10 (br, 1H), 4.48 (s, 1H), 3.94 (m, 2H), 3.47-3.38 (m, 6H), 2.29 (s, 3H), 2.19 (m, 2H), 2.06-1.99 (m, 2H); Elemental Analysis for C₂₆H₂₇FN₂O₅S₂ •0.75 H₂O, Calcd: C, 57.39; H, 5.28; N, 5.15; Found: C, 57.22; H, 5.11; N, 5.12.

3-(1,4-Diazabicyclo[3.2.2]nonan-4-yl)-4-fluorodibenzo[b,d]thiophene 5,5-dioxide (7d)

The typical procedure for Buchwald-Hartwig cross-coupling reaction was followed, starting with 6d (0.246 g, 0.78 mmol) and the title compound 7d was obtained as yellow solid (170 mg, 0.47 mmol, 60% yield). Free base: ¹H NMR (DMSO-d₆, 400 MHz) δ 8.06 (d, J=8.0 Hz, 1H), 7.91 (d, J=8.0 Hz, 1H), 7.81 (d, J=8.0 Hz, 1H), 7.75 (t, J=8.0 Hz, 1H), 7.54 (t, J=8.0 Hz, 1H), 7.36 (t, J=8.0 Hz, 1H), 3.82 (s, 1H), 3.43-3.40 (m, 2H), 3.03-3.00 (m, 2H), 2.93-2.89 (m, 4H), 2.00-1.97 (m, 2H), 1.74-1.66 (m, 2H); HRMS calculated for C₁₉H₂₀FN₂O₂S ([M+H]) 359.1224; found, 359.1246; TSA salt: ¹H NMR (DMSO-d₆, 400 MHz) δ 10.16 (s, 1H), 8.12 (br s, 1H), 7.94-7.90 (m, 2H), 7.78 (d, J=8.0 Hz, 1H), 7.59 (d, J=8.0 Hz, 1H), 7.50-7.40 (m, 3H), 7.12 (m, 2H), 4.01 (s, 1H), 3.54-3.38 (m, 6H), 2.30 (s, 3H), 2.19 (s, 2H), 2.07 (s, 2H), 1.09-1.03 (m, 2H); Elemental Analysis for C₂₆H₂₇FN₂O₅S₂ •0.5 H₂O, Calcd: C, 57.87; H, 5.23; N, 5.19; Found: C, 58.21; H, 5.56; N, 4.88.

1-(1,4-Diazabicyclo[3.2.2]nonan-4-yl)-4-fluorodibenzo[b,d]thiophene 5,5-dioxide (7e)

The typical procedure for Buchwald-Hartwig cross-coupling reaction was followed, starting with 6e (0.112 g, 0.36 mmol). The title compound 7e was obtained as a yellow solid (52 mg, 0.15 mmol, 40% yield). ¹H NMR (CDCl₃, 400 MHz) δ 8.50 (d, J=8.0 Hz, 1H), 7.84 (d, J=8.0 Hz, 1H), 7.68 (t, J=8.0 Hz, 1H), 7.55 (t, J=8.0 Hz, 1H), 7.43 (dd, J=8.0, 4.0 Hz, 1H), 7.14 (t, J=8.0 Hz, 1H), 3.66-3.63 (m, 1H), 3.29-3.21 (m, 5H), 3.14-3.09 (m, 2H), 2.16-2.10 (m, 2H), 1.87-1.71 (m, 3H); HRMS calculated for C₁₉H₂₀FN₂O₂S ([M+H]) 359.1224; found, 359.1215. Elemental Analysis for C₁₉H₁₉FN₂O₂S •1.5 H₂O, Calcd: C, 59.20; H, 5.75; N, 7.27; Found: C, 58.90; H, 5.76; N, 7.10.

3-Bromo-6-nitrodibenzo[b,d]thiophene 5,5-dioxide (8)

The typical procedure for bromination was followed, starting with 31 (1.96 g, 7.5 mmol) and compound 8 was obtained as a pale brown solid (1.73 g, 77%). ¹H NMR (DMSO-d₆, 400 MHz) δ 8.70 (d, J=8.0 Hz, 1H), 8.45-8.43 (m, 2H), 8.28 (d, J=8.0, Hz, 1H), 8.14-8.09 (m, 2H). HRMS calculated for C₁₂H₆BrNNaO₄S ([M+Na]⁺) 361.9093; found, 361.9080.

7-Bromo-2-nitrodibenzo[b,d]thiophene 5,5-dioxide (9)

The typical procedure for bromination was followed, starting with 24 (1.82 g, 6.95 mmol) and compound 9 (2.1 g, 89%) was obtained as a pale yellow solid. ¹H NMR (DMSO-d₆, 400 MHz) δ 9.10 (s, 1H), 8.44-8.47 (m, 3H), 8.33 (d, J=8.0 Hz, 1H), 8.11 (dd, J=8.0, 4.0 Hz, 1H).

3-(1,4-Diazabicyclo[3.2.2]nonan-4-yl)-6-nitrodibenzo[b,d]thiophene 5,5-dioxide (10)

The typical procedure for Buchwald-Hartwig cross-coupling reaction was followed starting with 8 (0.129 g, 0.38 mmol). Note: the reaction mixture was heated at 105 °C for 48 h. The title compound 10 was obtained as a reddish solid (80 mg, 55% yield). ¹H NMR (DMSO-d₆, 400 MHz) δ 8.40 (d, J=4.0 Hz, 1H), 8.15 (d, J=8.0 Hz, 1H), 7.97 (d, J=8.0 Hz, 1H), 7.94 (d, J=8.0 Hz, 1H), 7.26 (d, J=4.0 Hz, 1H), 7.15 (d, J=4.0 Hz, 1H), 4.21 (s, 1H), 3.71 (m, 2H), 3.00-2.85 (m, 6H), 2.00 (s, 2H), 1.71 (m, 2H); HRMS calculated for C₁₉H₂₀N₃O₄S ([M+H]) 386.1169; found, 386.1150; Elemental Analysis for C₁₉H₁₉N₃O₄S•H₂O, Calcd: C, 56.56; H, 5.25; N, 10.42; Found: C, 56.65; H, 4.99; N, 10.50.

7-(1, 4-Diazabicyclo[3.2.2]nonan-4-yl)-2-nitrodibenzo[b,d]thiophene 5,5-dioxide (11)

The typical procedure for Buchwald-Hartwig cross-coupling reaction was followed, starting with 9 (1.83 g, 5.38 mmol). The title compound 11 was obtained as a reddish solid (0.836 g, 61% yield). ¹H NMR (DMSO-d₆, 400 MHz) δ 8.77 (s, 1H), 8.20-8.12 (m, 3H), 7.32 (d, J=4.0 Hz, 1H), 7.16 (dd, J=8.0, 4.0 Hz, 1H), 4.23 (s, 1H), 3.72 (m, 2H), 3.00-2.88 (m, 6H), 2.00 (br s, 2H), 1.74-1.69 (m, 2H); HRMS calculated for C₁₉H₂₀N₃O₄S ([M+H]) 386.1169; found, 386.1152. Elemental Analysis for C₁₉H₁₉N₃O₄S • 1.25 H₂O, Calcd.: C, 55.94; H, 5.31; N, 10.30; found: C, 55.98; H, 5.17; N, 10.15.

6-Amino-3-(1,4-diazabicyclo[3.2.2]nonan-4-yl)-dibenzo[b,d]thiophene 5,5-dioxide (12)

The typical procedure for reduction of nitro derivatives was followed starting with 10 (0.34 g, 0.88 mmol) and compound 12 (146 mg, 46%) was obtained as a yellow solid. ¹H NMR (DMSO-d₆, 400 MHz) δ 7.73 (d, J=12.0 Hz, 1H), 7.25 (t, J=8.0 Hz, 1H), 7.11 (br s, 1H), 7.05 (d, J=8.0 Hz, 1H), 6.97 (d, J=4.0 Hz, 1H), 6.62 (d, J=8.0 Hz, 1H), 5.87 (br s, 2 H), 4.17 (s, 1H), 3.66 (m, 2H), 2.98-2.91 (m, 6H), 2.01 (s, 2H), 1.72 (m, 2H).

2-Amino-7-(1,4-diazabicyclo[3.2.2]nonan-4-yl)dibenzo[b,d]thiophene 5,5-dioxide (13)

The typical procedure for reduction of nitro derivatives was followed starting with 11 (0.68 g, 1.76 mmol) and compound 13 was obtained as a yellow solid (585 mg, 93%). ¹H NMR (DMSO-d₆, 400 MHz) δ 7.67 (d, J=12 Hz 1H), 7.43 (d, J=12 Hz 1H), 7.27 (s, 1H), 7.14 (d, J=8.0, 4.0 Hz, 1H), 6.91 (s, 1H), 6.53 (d, J=8.0, 4.0 Hz, 1H), 6.17 (s, 2H), 4.40 (s, 1H), 3.87 (br s, 3H), 3.07 (m, 1H), 2.15 (br s, 3H), 2.02 (br s, 3H), 1.20 (m, 2H).

3-(1,4-Diazabicyclo[3.2.2]nonan-4-yl)-6-iododibenzo[b,d]thiophene 5,5-dioxide (14)

Compound 12 (143 mg, 0.4 mmol) was dissolved in a mixture of 4 N H₂SO₄ (0.8 mL) and CH₃CN (1 mL), and the solution was cooled to −5 °C. Sodium nitrite (55 mg, 0.8 mmol) dissolved in H₂O (0.5 mL) was added dropwise at the same temperature. After stirring for 60 min a solution of diazonium salt was formed. To a mixture of CuI (268 mg, 1.4 mmol) and saturated KI solution (2.5 mL) at 70 °C was added above prepared solution of diazonium salt dropwise over 10 min and further stirred at 70 °C for 30 min. The reaction mixture was cooled and 28% ammonia solution was added (2 mL). The aqueous suspension was repeatedly extracted with CHCl₃ and the combined organic layers were washed with brine (10 mL), dried (Na₂SO₄) and concentrated in vacuo. The crude product was purified by flash chromatography on silica gel (CHCl₃/i-PrOH/Et₃N 10:1:0.1 to 3:1:0.2) to give 14 (28 mg, 15%). ¹H NMR (DMSO-d₆, 400 MHz) δ 7.96 (d, J=8.0 Hz, 1H), 7.92 (d, J=4.0 Hz, 1H), 7.81 (d, J=8.0 Hz, 1H), 7.38 (t, J=8.0 Hz, 1H), 7.32 (d, J=4.0 Hz, 1H), 7.18 (d, J=8.0 Hz, 1H), 4.34 (s, 1H), 3.82 (m, 2H), 3.22-3.18 (m, 6H), 2.09 (m, 2H), 1.87 (m, 2H); Elemental Analysis for C₁₉H₁₉IN₂O₂S •2.5 H₂O, Calcd: C, 44.63; H, 4.73; N, 5.48; Found: C, 44.88; H, 4.41; N, 5.48.

7-(1,4-Diazabicyclo[3.2.2]nonan-4-yl)-2-iododibenzo[b,d]thiophene 5,5-dioxide (15)

Compound 13 (285 mg, 0.8 mmol) was dissolved in a mixture of 4 N H₂SO₄ (1.5 mL) and CH₃CN (2 mL), and the solution was cooled to –5 °C. NaNO₂ (110 mg, 1.6 mmol, 2 equiv) dissolved in H₂O (1 mL) was added dropwise at the same temperature. After stirring for 60 min a solution of diazonium salt was formed. To a mixture of CuI (536 mg, 2.8 mmol, 3.5 eqiv.) and saturated KI solution (2.5 mL) at 70 °C was added above prepared solution of diazonium salt dropwise over 10 min and further stirred at 70 °C for 30 min. The reaction mixture was cooled and sat. NH₄OH was added (4 mL). The aqueous suspension was repeatedly extracted with CHCl₃ and the combined organic layers were washed with brine (10 mL), dried (Na₂SO₄) and concentrated in vacuo. The crude product was purified by flash chromatography on silica gel (CHCl₃/i-PrOH/Et₃N 10:1:0.1 to 3:1:0.2) to give 15 (75 mg, 20%). ¹H NMR (DMSO-d₆, 400 MHz) δ 8.40 (d, J=4.0 Hz, 1H), 7.93 (d, J=8.0 Hz, 1H), 7.78 (dd, J=8.0, 1.8 Hz, 1H), 7.60 (d, J=12.0 Hz, 1H), 7.25 (d, J=1.8 Hz, 1H), 7.11 (dd, J=8.0, 4.0 Hz, 1H), 4.20 (s, 1H), 3.70 (m, 2H), 2.98-2.88 (m, 6H), 2.00 (s, 2H), 1.72 (m, 2H); HRMS calculated for C₁₉H₂₀IN₂O₂S ([M+H]) 467.0285; found, 467.0306; Elemental Analysis for C₁₉H₂₁IN₂O₃S, Calcd: C, 47.12; H, 4.37; N, 5.78; Found: C, 47.24; H, 4.53; N, 5.87.

(5-Bromo-2-nitrophenyl)(2-fluorophenyl)sulfane (18)

Cesium carbonate (4.3 g, 13.2 mmol) was added to a solution of 4-bromo-2-fluoro-nitrobenzene 16 (2.42, 11 mmol, Aldrich) and 2-fluorobenzene thiol 17 (1.4 g, 11 mmol, Aldrich) in DMF (60 mL) and the mixture was stirred for 5 h at room temperature. Water (200 mL) and ethyl acetate (100 mL) were added, the organic layer was separated and washed sequentially with water (100 mL) then brine (100 mL). The organic phase was separated, dried and concentrated to yield a yellow solid that was purified by silica gel chromatography (Hexanes/EtOAc 8:1 to 3:1) to give 18 (2.88 g, 80%). ¹H NMR (CDCl₃, 400 MHz) δ 8.17 (d, J=8.0 Hz, 1H), 7.68-7.60 (m, 2H), 7.41-7.29 (m, 3H), 6.95 (s, 1H).

4-Bromo-2-((2-fluorophenyl)thio)aniline (19)

The typical procedure for reduction of nitro derivatives was followed, starting with 18 (3.2 g, 9.75 mmol) and the title compound 19 was obtained as a brown solid (2.46 g, 85%). ¹H NMR (CDCl₃, 400 MHz) δ 7.59 (d, J=4.0 Hz, 1H), 7.33 (dd, J=8.0, 4.0 Hz, 1H), 7.21-7.15 (m, 1 H), 7.10-7.00 (m, 2H), 6.92-6.87 (m, 1H), 6.69 (d, J=8.0 Hz, 1H), 4.37 (br s, 2H).

3-Bromo-6-fluorodibenzo[b,d]thiophene (20)

Compound 19 (1.18 g, 3.96 mmol) was dissolved in 37% HCl (11 mL) and the solution was cooled below 5 °C. To this reaction mixture, sodium nitrite (408 mg, 5.93 mmol) was added slowly at temperature below 5 °C. After addition, the reaction was stirred for 30 min below 5 °C. Then sodium tetrafluoroborate (865 mg, 7.92 mmol) was added and the reaction mixture was stirred for another 30 min at temperature below 5 °C. This reaction solution was then added to the stirred solution of copper (I) oxide (1.14 mg, 7.92 mmol) in 0.1 N sulfuric acid (390 mL) at 35–40 °C. The reaction mixture was stirred for 15–30 min. Ethyl acetate was added to the reaction mixture and the mixture was filtered to remove inorganic compound. The filtrate was then extracted with ethyl acetate (3 × 120 mL). The organic extract was washed with water followed by brine and then concentrated under vacuum. The residue was purified by silica gel chromatography (hexanes) to give 20 (600 mg, 54%). ¹H NMR (CDCl₃, 400 MHz) δ 8.04 (d, J=4.0 Hz, 1H), 8.02 (d, J=8.0 Hz, 1H), 7.93 (d, J=8.0 Hz, 1H), 7.62 (dd, J=8.0, 4.0 Hz, 1H), 7.47 (ddd, J=12.0, 8.0, 4.0 Hz, 1H), 7.22 (t, J=8.0 Hz, 1H).

3-Nitrodibenzo[b,d]thiophene 5,5-dioxide (23)

Dibenzo[b,d]thiophene 5,5-dioxide 21 (10 g, 46 mmol, Aldrich) was slowly added to a stirred mixture of glacial acetic acid (34 mL) and sulfuric acid (96%, 34 mL). The slurry was stirred and red fuming nitric acid (36 mL) was added dropwise over a period of 90 min at temperature −5°C – 5°C. The slurry was stirred for another 30 min and poured over ice. The precipitate was filtered, rinsed with water and dried at room temperature. The crude product was recrystallized with acetonitrile to give 23 as yellow crystals (8.7 g, 72%).¹H NMR (DMSO-d₆, 400 MHz) δ 8.84 (d, J=8.0 Hz, 1H), 8.65 (dd, J=8.0, 2.0 Hz, 1H), 8.50 (d, J=8.0, Hz, 1H), 8.39 (d, J=8.0, Hz, 1H), 8.13 (d, J=8.0, Hz, 1H), 7.93 (t, J=8.0, Hz, 1H), 7.81 (t, J=8.0, Hz, 1H).

2-Nitro-dibenzo[b,d]thiophene 5,5-dioxide (24)

The typical procedure for oxidation of 1,4-dibenzothiophene was followed starting with 22 (489 mg, 2.13 mmol, Oakwood Chemical) and the title compound 24 (510 mg, 90%) was obtained as white crystals. ¹H NMR (CDCl₃, 400 MHz) δ 8.66 (d, J=4 Hz 1H), 8.43 (dd, J=8, 4 Hz, 1H), 8.04 (d, J=8 Hz, 1H), 7.96 (d, J=8 Hz, 1H), 7.92 (d, J=8 Hz, 1H), 7.79 (t, J=8 Hz, 1H), 7.69 (t, J=8.0 Hz, 1H).

3-Bromo-7-nitrodibenzo[b,d]thiophene 5,5-dioxide (25)

The typical procedure for bromination was followed starting with 23 (2.59 g, 9.9 mmol) and brown solid was obtained and recrystallized with benzene to yield 25 as a yellow solid (1.73 g, 51%). ¹H NMR (DMSO-d₆, 400 MHz) δ 8.89 (d, J=4.0 Hz, 1H), 8.67 (dd, J=8.0, 3.0 Hz, 1H), 8.52-8.49 (m, 2H), 8.35 (d, J=8.0, Hz, 1H), 8.15 (dd, J=8.0, 2.0 Hz, 1H).

7-Bromodibenzo[b,d]thiophen-3-amine 5,5-dioxide (26)

A solution of stannous chloride dihydrate (12.4 g, 56 mmol) in 37% hydrochloric acid (21 mL) was added to a mixture of 25 (1.7 g, 5 mmol) in glacial acetic acid (50 mL). The reaction mixture was stirred at 100 °C for 60 min and cooled down to 5 °C. The precipitate was filtered off, rinsed with water on the filter and dispersed in water. The dispersion was made basic (pH=10) by addition of an excess of 1M sodium hydroxide and stirred for 3 h. The precipitate was filtered off, rinsed with water and dried overnight on the filter to yield 26 (0.7 g, 45%) as a pale white solid. ¹H NMR (DMSO-d₆, 400 MHz) δ 8.12 (s, 1H), 7.87-7.77 (m, 3H), 6.95 (s, 1H), 6.87 (dd, J=8, 4 Hz, 1H), 6.20 (br s, 2H).

2-Amino-7-bromodibenzo[b,d]thiophene 5,5-dioxide (27)

The typical procedure for reduction of nitro derivatives was followed staring with 9 (0.60 g, 1.76 mmol). The title compound 27 (496 mg, 91%) was obtained as a white solid. ¹H NMR (DMSO-d₆, 400 MHz) δ 8.16 (d, J=4 Hz, 1H), 7.93(d, J=8.0 Hz, 1H), 7.87 (d, J=12 Hz, 1H), 7.56 (d, J=8 Hz, 1H), 7.08 (s, 1H), 6.71 (d, J=8 Hz, 1H), 6.36 (br s, 2H).

4-Fluorodibenzo[b,d]thiophene 5,5-dioxide (29)

The typical procedure for oxidation of 1,4-dibenzothiophene was followed starting with 4-fluorodibenzo[b,d]thiophene 28⁴⁵ (1.62 g, 8 mmol). The title compound 29 (1.8 g, 96%) was obtained as white crystals. ¹H NMR (CDCl₃, 400 MHz) δ 7.85 (d, J=8.0 Hz, 1H), 7.82 (d, J=8.0 Hz, 1H), 7.71-7.57 (m, 4H), 7.20 (t, J=8.0 Hz, 1H).

4-Nitrodibenzo[b,d]thiophene 5,5-dioxide (31)

The typical procedure for oxidation of 1,4-dibenzothiophene was followed starting with 30⁴⁶ (1.08 g, 4.71 mmol). The final compound 31 (1.1g, 90%) was obtained as pale yellow crystals. ¹H NMR (DMSO-d₆, 400 MHz) δ 8.69 (d, J=8.0 Hz, 1H), 8.42 (d, J=8.0, Hz, 1H), 8.33 (d, J=8.0, Hz, 1H), 8.11 (t, J=8.0 Hz, 1H), 8.07 (d, J=8.0, Hz, 1H), 7.88 (t, J=8.0 Hz, 1H), 7.77 (t, J=8.0 Hz, 1H). HRMS calculated for C₁₂H₇NNaO₄S ([M+Na]) 283.9988; found, 283.9994.

Radiosynthesis of [¹⁸F]7a and [¹⁸F]7c

The same radiolabeling method was used for both radioligands [¹⁸F]7a and [¹⁸F]7c. A solution of the [¹⁸F]fluoride, 15–20 mg of Kryptofix 222^®, and 1–2 mg of K₂C₂O₄ in 1 mL of 50% aqueous acetonitrile was added to a reaction vessel of a GE MicroLab box. The mixture was heated at 120–135 °C under a stream of argon, while water was evaporated azeotropically after the additions of 2 mL of CH₃CN. A solution of the corresponding nitro precursor (10 or 11) (2 mg) in anhydrous DMSO (0.8 mL) was added to the reaction vessel and heated at 160 °C for 12 min. The reaction mixture was cooled, diluted with 0.7 mL of water, injected onto the reverse-phase semi-preparative HPLC column (Table 6). The radioactive peak was collected in 50 mL of HPLC water. The water solution was transferred through an activated Waters C-18 Oasis HLB light solid-phase extraction (SPE) cartridge. After washing the SPE with 10 mL saline, the product was eluted with a mixture of 1 mL of ethanol and 0.04 mL 1N HCl through a 0.2 µM sterile filter into a sterile, pyrogen-free multi-dose vial and 10 mL of 0.9% saline and 0.05 mL sterile 8.4% solution sodium bicarbonate was added through the same filter. The final products [¹⁸F]7a or [¹⁸F]7c were then analyzed by analytical HPLC (Table 6) using a UV detector at 340 nm to determine the radiochemical purity and specific radioactivity at the time synthesis ended. The total synthesis time including QC was 70–80 min.

In vitro binding assay

α7-nAChR assay with rat brain membranes

The assay was done commercially by Caliper PerkinElmer (Hanover, MD). In brief, rat cortical membranes were incubated with [¹²⁵I]α-Bungarotoxin (K_D = 0.7 nM) at a concentration of 0.1 nM in a buffer consisting of 50 mM Tris, 120 mM NaCl, 5 mM KCl, 2 mM CaCl₂, 1 mM MgCl₂, 0.003 mM atropine sulfate at pH 7.4 for 150 min at 0 °C.⁶³ The binding was terminated by rapid vacuum filtration of the assay contents onto GF/C filters presoaked in PEI. Radioactivity trapped onto the filters was assessed using a gamma-counter. Nonspecific binding was defined as that remaining in the presence of 1 µM α-Bungarotoxin. The assays were done two times independently, each in duplicate, at multiple concentrations of the test compounds. Binding assay results were analyzed using a one-site competition model, and IC₅₀ curves were generated based on a sigmoidal dose response with variable slope. The K_i values were calculated using the Cheng-Prusoff equation. Methyllycaconitine (MLA) was used as a reference compound in all assays.

HEK 293 Cell culture and stable transfections (heteromeric nAChR)

HEK 293 cells (ATCC CRL 1573) were maintained at 37 °C with 5% CO₂ in a humidified incubator. Growth medium for the HEK 293 cells was the minimum essential medium supplemented with 10% fetal bovine serum, 100 units/mL penicillin G, and 100 µg/mL streptomycin. Transfections of these cells, and selection and establishment of stable cell lines were carried out as described previously.^{52, 55}

Membrane homogenate preparation (heteromeric nAChR)

Membrane homogenates for ligand binding assays were made as described previously.^{52, 55} Briefly, cultured cells at >90% confluency were removed from the culture flask (80 cm²) with a disposable cell scraper and placed in 10 mL of 50 mM Tris.HCl buffer (pH 7.4, 4 °C). The cell suspension was centrifuged at 1,000 × g for 5 min and the pellet was collected. The cell pellet was then homogenized in 10 mL buffer with a polytron homogenizer for 20 s and centrifuged at 35,000 g for 10 min at 4 °C. Membrane pellets were resuspended in fresh buffer.

Binding to heteromeric nAChR

subunit combinations, which represent possible heteromeric nAChRs, was measured with 0.5 nM [³H]epibatidine in HEK cells expressing these subunits (K_D = 0.021 nM (α2β2-nAChR), K_D = 0.084 nM (α2β4-nAChR), K_D = 0.034 nM (α3β2-nAChR), K_D = 0.29 nM (α3β4-nAChR), K_D = 0.046 nM (α4β2-nAChR), K_D = 0.094 nM (α4β4-nAChR)).⁵⁵ Aliquots of the membrane homogenates containing 30 to 200 µg protein were used for the binding assays, which were carried out in a final volume of 100 µL in borosilicate glass tubes. After incubation at 24 °C for 2 h, the samples were collected with a cell harvester (Brandel M-48) onto Whatman GF/C filters pre-wet with 0.5% polyethylenimine. After harvesting the samples, the filters were washed three times with 5 mL of 50 mM Tris.HCl buffer, and then counted in a liquid scintillation counter. Nonspecific binding was measured in samples incubated in parallel containing 300 µM nicotine for [³H]epibatidine binding. Specific binding was defined as the difference between total binding and nonspecific binding. Data from these competition binding assays were analyzed using Prism 5 (GraphPad Software, San Diego, CA).

5-HT₃(h) binding assay

The assay was done commercially by Caliper PerkinElmer (Hanover, MD) using recombinant HEK293 cells and 0.35 nM [³H]GR65630 (K_D = 0.5 nM).

Biodistribution studies in CD-1 mice

Baseline study

Male, CD-1 mice weighing 25–30 g from Charles River Laboratories, (Wilmington, MA) were used for biodistribution studies. The animals were sacrificed by cervical dislocation at various times following injection of [¹⁸F]7a or [¹⁸F]7c (70 µCi, specific radioactivity 8000–12000 mCi/µmol, in 0.2 mL saline) into a lateral tail vein; three animals per time point. The brains were rapidly removed and dissected on ice. The brain regions of interest were weighed and their radioactivity content was determined in an automated γ-counter with a counting error below 3%. Aliquots of the injectate were prepared as standards and their radioactivity content was counted along with the tissue samples. The percent of injected dose per gram of tissue (%ID/g tissue) was calculated. All experimental protocols were approved by the Animal Care and Use Committee of the Johns Hopkins Medical Institutions.

Self-blockade of [¹⁸F] 7a binding with 7a

In vivo saturation blockade studies were done by i.v. co-administration of the radiotracer [¹⁸F]7a (70 µCi, SA = 9200 mCi/µmol, 0.2 mL) with various doses of “cold” 7a per animal (0 µg (vehicle); 0.0048 µg; 7.2 µg). Compound 7a was dissolved in saline at pH = 5.5. Ninety min after administration of the tracer and blocker, brain tissues were harvested, and their regional radioactivity content was determined. The self-blockade of [¹⁸F]7c with 7c was done similarly.

Blockade of [¹⁸F] 7a binding with 1

In vivo α7-nAChR receptor blocking studies were done by intravenous (i.v.) co-administration of the radiotracer [¹⁸F]7a (70 µCi, SA = 7900 mCi/µmol, 0.2 mL) with various doses of 1 (0 µg (vehicle); 0.02 mg/kg; 0.2 mg/kg, 1 mg/kg and 3 mg/kg). Three animals per dose were used. 1 was dissolved in a vehicle (saline:alcohol (9:1) at pH=5.5). Ninety min after administration of the tracer, brain tissues were harvested, and their regional radioactivity content was determined. The dose-dependent blockade study of [¹⁸F]7c with 5 was done the same way.

Blockade of [¹⁸F] 7a with nicotine and cytisine

In vivo CB1 receptor blocking studies were carried out by subcutaneous (s.c.) administration of (−)nicotine tartrate (5 mg/kg) or cytisine (1 mg/kg) followed by i.v. injection of the radiotracer [¹⁸F]7a (70 µCi, specific radioactivity ~14000 mCi/µmol, 0.2 mL) 5 min thereafter. The drugs were dissolved in saline and administered in a volume of 0.1 mL. Control animals were injected with 0.1 mL of saline. Ninety min after administration of the tracer, brain tissues were harvested, and their radioactivity content was determined.

Blockade of [¹⁸F] 7a with non-α7-nAChR drugs

In vivo receptor blocking studies were performed by administration of six drugs (Table 5), followed by i.v. injection of the radiotracer [¹⁸F]7a (70 µCi, specific radioactivity ~14000 mCi/µmol, 0.2 mL). The drugs (2 mg/kg, s.c.) were dissolved in a vehicle (saline:DMSO 5:1) and administered in a volume of 0.1 mL. Control animals were injected with 0.1 mL of the vehicle solution. Ninety min after administration of the tracer, brain tissues were harvested, and their radioactivity content was determined.

Scheme 5.

Reagents and conditions: (a) 30% H₂O₂, acetic acid, 60 °C, 24 h; (b) NBS, H₂SO₄, 24 h.

List of non-standard abbreviations and acronyms

BINAP

2,2'-bis(diphenylphosphino)-1,1'-binaphthyl

B_max

receptor density in tissue

BP_ND

binding potential

EtOAc

ethylacetate

EtOH

ethanol

Et₂O

diethylether

Et₃N

triethylamine

K_D

dissociation constant of the radioligand-receptor complex

nAChR

nicotinic acetylcholine receptor

i-PrOH

isopropyl alcohol