Identification and Development of A New Positron Emission Tomography Ligand 4-(2-Fluoro-4-[¹¹C]methoxyphenyl)-5-((1-methyl-1H-pyrazol-3-yl)methoxy)picolinamide for Imaging Metabotropic Glutamate Receptor Subtype 2 (mGlu₂)

Abstract

Metabotropic glutamate receptor 2 (mGlu₂) is a known target for treating several central nervous system (CNS) disorders. To develop a viable PET ligand for mGlu₂, we identified new candidates 5a–5i as potent negative allosteric modulators (NAMs) of mGlu₂. Among these candidates, 4-(2-fluoro-4-methoxyphenyl)-5-((1-methyl-1H-pyrazol-3-yl)methoxy)picolinamide (5i, also named as [¹¹C]MG2-1812) exhibited high potency, high subtype selectivity, and favorable lipophilicity. Compound 5i was labeled with positron-emitting carbon-11 (¹¹C) to obtain [¹¹C]5i in high radiochemical yield and high molar activity by O-[¹¹C]methylation of the phenol precursor 12 with [¹¹C]CH₃I. In vitro autoradiography with [¹¹C]5i showed heterogeneous radioactive accumulation in the brain tissue sections, ranked in the order: cortex > striatum > hippocampus > cerebellum ≫ thalamus > pons. PET study of [¹¹C]5i indicated in vivo specific binding of mGlu₂ in the rat brain. Based on the [¹¹C]5i scaffold, further optimization for new candidates is underway to identify a more suitable ligand for imaging mGlu₂.

Graphical Abstract

INTRODUCTION

Glutamate is a key neurotransmitter that is required for many neuronal functions in the central nervous system (CNS). In particular, excitatory glutamatergic neurotransmission plays a vital role in the development of neuronal networks and in synaptic plasticity. Excess stimulation of glutamatergic neurons induces neurotoxicity, resulting in a variety of CNS disorders.¹ The actions of glutamate are mediated by the glutamate receptors, which are classified as ionotropic- and metabotropic receptors.² Of these two types of receptors, the metabotropic glutamate receptors (mGlus) are G protein-coupled receptors (GPCRs) that are categorized into three groups including eight subtypes.³ Among these subtypes, mGlu₂ and mGlu₃ are group II, G_i/o-type protein-coupled receptors, which downregulate cyclic adenosine monophosphate formation via inhibition of adenylyl cyclase.⁴ In particular, mGlu₂ is predominately localized in presynaptic terminals and involved in neuronal protection against excitotoxicity via negative regulation of the endogenous glutamate concentration.⁵ It has been reported that pharmacological modulation of mGlu₂ activity shows promising therapeutic effects in multiple CNS disorders, including Alzheimer’s disease,⁶ Parkinson’s disease,^7,8 schizophrenia,⁹ epilepsy,¹⁰ pain,¹¹ anxiety,¹² and depression.¹³ Because of this, mGlu₂ has been considered a potential target for drug discovery. However, direct evidence for the involvement of mGlu₂ in such CNS disorders has not been established due to the lack of in vivo imaging studies.

Positron emission tomography (PET) is an advanced imaging modality, which permits in vivo activity to be visualized using radiolabeled, targeted molecules, and provides detailed information about the relationship between brain function and disease. In 2012, an allosteric binding PET ligand derived from 7-(phenylpiperidinyl)-1,2,4-triazolo[4,3-a]pyridine was developed for mGlu₂ imaging for the first time.¹⁴ However, in the report, specific binding in the PET images was insufficient for vizualization.¹⁵ Inspired by the identification of VU6001192 (1, IC₅₀ = 207 nM),¹⁶ a potent negative allosteric modulator (NAM) of mGlu₂, our research group developed 7-((2,5-dioxopyrrolidin-1-yl)methyl)-4-(2-fluoro-4-[¹¹C]methoxyphenyl)-quinolone-2-carboxamide ([¹¹C]QCA, [¹¹C]2, IC₅₀ = 45 nM; Figure 1).¹⁷ In vitro autoradiography in rat brain sections showed that [¹¹C]QCA demonstrates a relatively high specific binding at locations that corresponding to mGlu₂-rich brain regions. However, in a PET imaging study, [¹¹C]QCA showed poor penetration into the rat brain, because of intensive interaction with drug efflux transporters located on the blood–brain barrier (BBB).¹⁷ Recently, the structure-activity relationship (SAR) studies using compounds 1 and 2 were conducted to improve their penetration into the brain.¹⁸ Among all the reported candidates, 4-(2-fluoro-4-methoxyphenyl)-5-((2-methylpyridin-4-yl)methoxy)picolinamide (3; Figure 1) and 4 (VU6001966) exhibited high affinities for mGlu₂ (IC₅₀ = 26 nM and 78 nM, respectively), low plasma protein binding, and high brain penetration.¹⁸ These attractive profiles motivated us to radiolabel 3 with ¹¹C. We have recently reported the radiosynthesis and evaluation of [¹¹C]3, and determined the heterogeneous distribution and high specific binding of [¹¹C]3 by in vitro autoradiography.¹⁹ As expected, the brain penetration of [¹¹C]3 was improved over [¹¹C]2; however, in vivo specific binding of [¹¹C]3 remained insufficient.

Figure 1.

Chemical structures of some modulators of mGlu₂.

To develop a useful radioligand for PET imaging of mGlu₂, using 3 and 4 as lead compounds, we synthesized nine novel compounds 5a–5i which are suitable for radiolabeling (Figure 2). These compounds were divided into two categories: 1) pyridine and pyrimidine analogs (5a–5f) with different substitution groups and positions; 2) pyrazole and 1,2,4-oxopyrazole analogs (5g–5i). We characterized their potency and selectivity towards mGlu₂. Among those compounds, compound 5i demonstrated the highest potency for mGlu₂ and a more than 450-fold selectivity for mGlu₂ over mGlu₃. This compound was labeled with ¹¹C and evaluated in vivo to study its potential as a PET ligand. In addition, we labeled benchmark compound 4 to compare PET imaging potentials between [¹¹C]4 and [¹¹C]5i.

Figure 2.

Chemical structures of the targeted compounds 5a–5i and the radioligands [¹¹C]5i and [¹¹C]4 for mGlu₂ in this study.

RESULTS AND DISCUSSION

Chemical Synthesis of a Focused Library of mGlu₂ Compounds

In this study, nine novel mGlu₂-targeted compounds were synthesized, as shown in Scheme 1. The starting material 5-fluoro-4-(2-fluoro-4-methoxyphenyl)picolinonitrile (6) was synthesized according to reported methods.¹⁸ The reaction of 6 with different benzyl alcohol compounds in the presence of sodium hydride as a base produced ethers 7a–7i, of which the crude 7e–7h were directly used for the next step, without further purification. Treatment of 7a–7i with potassium trimethylsilanolate (KOSiMe₃) or hydrogen peroxide (H₂O₂) produced the primary carboxamides 5a–5i in 20–69% yields from 6, respectively.

Scheme 1.

Synthesis of mGlu₂-targeted compounds 5a–5i. a) NaH, THF, 0°C to room temperature, 10–12 h; b) DMSO, NaOH, room temperature, 20 min.

Pharmacology and Lipophilicity

Compounds 5a–5i were screened in vitro for their potency and selectivity towards mGlu₂ and mGlu₃, according to our previously reported protocols.¹⁷ Pharmacological results are shown in Table 1. Briefly, we utilized a thallium flux assay in HEK293 cells expressing heteromeric G protein-coupled inwardly-rectifying potassium (GIRK) channels and human mGlu₂ or mGlu₃, to determine the potency and selectivity of our candidate compounds. The potency was expressed as the IC₅₀ for the inhibition of the glutamate EC₈₀ response (Figure 3).

Table 1.

In vitro potency and lipophilicity of compounds of 3, 4, and 5a–5i.

Compound R	Potency (IC50, μM)		cLogD
	for mGlu2 (n = 3)	for mGlu3 (n = 3)
	0.059 ± 0.008 (0.03)a)	> 10	3.12
	0.033 ± 0.001	> 10	3.02
	0.046 ± 0.010	> 10	3.41
	0.135 ± 0.012	> 10	3.36
	0.074 ± 0.015	> 10	3.41
	0.224 ± 0.032	> 10	3.12
	0.261 ± 0.026	> 10	2.12
	0.158 ± 0.042	> 10	2.54
	0.092 ± 0.030	> 10	2.39
	0.021 ± 0.001	> 10	2.72
	0.058 ± 0.08 (0.078)a)	> 10	2.12

Figure 3.

Concentration-inhibition relationships of compounds 5a–5i to glutamate responses reflecting mGlu₂ (A) and mGlu₃ (B) functions, compared to the lead compounds 3 and 4.

Under the present experimental conditions, the in vitro potency of compound 3 was determined to be 0.059 μM for mGlu₂; however, the value was previously reported to be 0.026 μM.¹⁸ We postulate that the discrepancy in the potency of compound 3 could be derived from different measurement methods as the group from Vanderbilt evaluated the activity of this compound in calcium mobilization assays utilizing promiscuous G protein coupling¹⁷ as opposed to our own efforts utilizing mGlu₂ or mGlu₃ coupling to GIRK channels. As shown in Figure 3A and Table 1, these compounds exhibited a wide range of inhibitory activities for mGlu₂ with IC₅₀ values of 0.021–0.261 μM. The compounds without a substituent (5a) or with a small fluorine atom (5b and 5d) in the pyridine ring showed a similar order of IC₅₀ values to 3 (0.033–0.074 μM). When the methyl group in 3 was changed to a methoxy group, the potency of 5c and 5e decreased to 0.135 μM and 0.224 μM, respectively. In place of the pyridine ring, the pyrimidine analog 5f reduced the potency to 0.261 μM. Hinted by the previously reported compound 4, which contains a pyrazole ring, we designed three five-membered heterocycle compounds at this time. To our surprise, the 4-pyrazole 5g and 3-oxopyrazole 5h analogs showed higher IC₅₀ values than those of 3 and 4. Among all synthesized compounds, the 5-pyrazole analog 5i showed the highest potency for mGlu₂, and the IC₅₀ value of 5i was lower than those of the two lead compounds 3 and 4. In addition, as shown in Table 1 and Figure 3B, 5a–5i did not show significant inhibitory effects towards mGlu₃ (IC₅₀ > 10 μM). These results revealed the excellent selectivity of 5a–5i for mGlu₂ over mGlu₃. The density (B_max) of mGlu_2/3 determined using [³H]LY341595 was reported to be 2000 fmol/mg protein of prefrontal cortex in postmortem brain of healthy subjects.²⁰ Thus, it was assumed that mGlu2 and mGlu₃ are also expressed in rat brain with high density.

We also confirmed the mechanism of action by which compound 5i acts as a NAM by performing a Schild analysis.²¹ The compound 5i dose-dependently right-shifted the concentration-response of glutamate toward mGlu₂ (Figure 4A) and decreased the maximal glutamate response, consistent with a noncompetitive mode of action. Moreover, consistent with the above in vitro studies, 5i had no effect on either the glutamate potency or the glutamate maximal response toward mGlu₃ (Figure 4B). These results also serve as an additional control demonstrating the mGlu₂ selectivity of 5i. Moreover, 5i did not show significant inhibition against major neuronal receptors and transporters in the CNS (Table S1 in Supplemental information).

Figure 4.

Assays to determine the antagonistic activity of 5i. The glutamate-induced increases in thallium flux were evaluated for both mGlu₂ (A) and mGlu₃ (B) in the absence or in the presence (30, 10, 3.3, and 1.1 μM, and 370, 123, and 41 nM) of 5i.

The lipophilicity of candidate compounds is a predictive factor in assessing brain permeability, with a preferred range of 1.0−3.5.^22,23 The cLogP values of the experimental compounds were predicted to be 2.31–3.41 using the Pallas 3.0 prediction software (Table 1). All the values for these compounds are in the range generally considered as being suitable for PET ligands.

Radiochemistry

Based on these results (Figures 3 and 4; Table 1), we found that 5i exhibited the highest potency for mGlu₂ among all analogs and reasonable lipophilicity, which warranted further radiolabeling and subsequent evaluation by in vivo PET imaging and ex vivo biodistribution studies. For comparison, we also labeled 4 to evaluate the potentials of [¹¹C]4 as a PET radioligand. To synthesize the two radioligands [¹¹C]5i and [¹¹C]4, we prepared the corresponding desmethyl precursors 12 and 17, and then reacted them individually with [¹¹C]CH₃I.

The desmethyl precursor 12 was synthesized as shown in Scheme 2. Removal of the methyl group in compound 5i was attempted using common demethylating reagents, such as boron tribromide, pyridine hydrochloride, and sodium sulfide. However, none of the conditions were sufficient to remove the methyl group in 5i, leaving decomposed 5i in the reaction mixture. As an alternative approach, we synthesized 11 from a known intermediate 8,¹⁹ in which the hydroxy group was protected by a methoxymethoxy group. The coupling reaction of 8 with pyrazole methyl alcohol (9) using sodium hydride as a base produced 10. The hydrolysis of 10 with KOSiMe₃ resulted in primary amide 11, followed by the removal of methoxymethoxy group with HCl to produce 12 with a 72% yield (from 10).

Scheme 2.

Synthesis of desmethyl precursor 12 for radiolabeling: a) NaH, THF, 0°C to room temperature, 12 h; b). DMSO, NaOH, 20 min, room temperature; c) MeOH, 40°C, 10 h.

The desmethyl precursor 17 was synthesized as shown in Scheme 3. We synthesized 15 via the coupling reaction between 13¹⁹ and trityl group-protected pyrazole methyl alcohol 14, which was prepared by the tritylation and LiAlH₄-reduction of 18. Hydrolysis of nitrile 15 with KOSiMe₃ resulted in amide 16. Subsequent removal of the trityl group in 16 produced 17 with 45% yield.

Scheme 3.

Synthesis of the desmethyl precursor 17 for radiolabeling: a) NaH, THF, 0°C to room temperature, 12 h; b). NaOH, DMSO, room temperature, 20 min; c) MeOH, 40°C, 10 h; d) NaH, THF, room temperature, 4 h; e) THF, room temperature, 3 h.

The radiosynthesis of [¹¹C]5i or [¹¹C]4 was performed by the reaction of the desmethyl precursor 12 or 17 with [¹¹C]CH₃I (Scheme 4). All processes from the production of [¹¹C]CH₃I to the formulation of radioactive products were conducted in an automated synthesis system developed in-house.^24,25 [¹¹C]CH₃I as a labeling agent was produced by reducing cyclotron-produced [¹¹C]CO₂ with LiAlH₄, followed by iodination with HI. The O-[¹¹C]methylation of 12 or 17 was performed with [¹¹C]CH₃I. After purification using semi-preparative HPLC and formulation, [¹¹C]5i or [¹¹C]4 was obtained as an injectable solution with >98% radiochemical purity. The decay-corrected radiochemical yields based on [¹¹C]CO₂ were 56.8 ± 8.9% (n = 20) for [¹¹C]5i, and 12.6 ± 7.9% for [¹¹C]4 (n = 15). The synthesis time was about 40 min from the end of bombardment (EOB). At the end of synthesis (EOS), the molar activity of [¹¹C]5i and [¹¹C]4 in the formulated solutions was measured as >74 GBq/μmol (n > 30). Neither products showed radiolysis at room temperature within 90 min of the formulation, indicating that excellent radiochemical stability was achieved for the period of at least one PET scan. The analytical results of both radioactive products complied with our in-house quality control/assurance specifications for radiopharmaceuticals in evaluation and application studies.

Scheme 4.

Radiosynthesis of [¹¹C]5i and [¹¹C]4. a) NaOH, DMF, 80 °C, 5 min; b) Cs₂CO₃, DMSO, 90 °C, 5 min.

In Vitro Autoradiography

To evaluate the specificity and selectivity of [¹¹C]5i and [¹¹C]4 for rodent mGlu₂, in vitro autoradiography using rat brain sections was performed with these NAMs (Figure 5). Figures 5A and B show representative autoradiographs of [¹¹C]5i and [¹¹C]4 using sagittal brain sections. In both control sections, heterogeneous radioactive accumulation was detected. High levels of radioactivity were determined in the striatum, cortex, hippocampus, and cerebellum, while in the thalamus and pons it was very low. Previously, biological evidence has indicated that the magnitude of mGlu₂ expression in the brain is in the following order: hippocampus > cortex > cerebellum.²⁶ This corresponds to the distribution pattern observed on the autoradiographs of both control sections. Pre-treatment with 4-(8-bromo-2,3-dihydro-2-oxo-1H-1,5-benzodiazepin-4-yl)-2-pyridinecarbonitrile^27,28 (MNI-137, 20; an mGlu_2/3-selective NAM) lead to a significant decrease of radioactive accumulation compared to the control sections, consistent with its high in vitro binding specificity.

Figure 5.

In vitro autoradiography with [¹¹C]5i and [¹¹C]4 using rat brain sections. Representative autoradiographs were obtained by incubating sagittal brain sections with [¹¹C]5i (A) and [¹¹C]4 (B) solutions including either DMSO (vehicle), unlabeled 5i or 4 (10 μM, self-blocking), or 20 (10 μM). Ratios of total binding against nonspecific binding on autoradiographs with [¹¹C]5i (C) and [¹¹C]4 (D) were estimated in the cortex, striatum, hippocampus, thalamus, cerebellum, and pons. ***p < 0.001, **p < 0.01, *p < 0.05 (vs. control).

Figures 5C and D show the ratio of total binding against nonspecific binding (radioactivity in the self-blocking group), which is an index reflecting the specific binding of a radioligand to its receptor. In the controls, the respective ratios of [¹¹C]5i and [¹¹C]4 were: 17.2 ± 3.0 and 5.5 ± 1.3 in the cortex; 8.7 ± 1.8 and 5.6 ± 1.7 in the striatum; 9.3 ± 2.2 and 4.3 ± 1.0 in the hippocampus; 2.5 ± 0.6 and 1.7 ± 0.6 in the thalamus; 9.1 ± 1.1 and 2.8 ± 0.7 in the cerebellum; and 2.2 ± 0.2 and 1.4 ± 0.5 in the pons. Treatment with NAM 20 significantly decreased these ratios in mGlu₂-rich brain regions (cortex, striatum, hippocampus, thalamus, and cerebellum; all p < 0.05). Thus, the effective blocking with a selective mGlu_2/3 dual NAM, coupled with in vitro binding selectivity of over mGlu₃, suggested good mGlu₂ target specificity of [¹¹C]5i in brain tissue binding.

Small-Animal PET Assessment Using Pgp/BCRP Double Knockout Mouse

Our previous PET study showed that the brain uptake of [¹¹C]2 and [¹¹C]3 was limited by P-glycoprotein (Pgp) and breast cancer resistant protein (BCRP).¹⁹ Prior to the PET assessment evaluating in vivo specific binding in the rat brain, we conducted whole-brain PET imaging using Pgp/BCRP double knockout (Pgp/BCRP-KO) mice and wild-type mice. As shown in Figure 6B, the maximum uptake of [¹¹C]5i in the whole brain was 1.16 standardized uptake value (SUV) in a wild-type mouse and 1.25 SUV in a Pgp/BCRP-KO mouse. In our previous paper, the maximum uptake of [¹¹C]3 was 0.52 SUV in wild-type and 0.97 SUV in Pgp/BCRP-KO mice.¹⁹ These results support the conclusion that [¹¹C]5i has a negligible interaction with ATP-binding cassette efflux pumps in the murine BBB. On the other hand, as shown in Figure 6D, the maximum uptake of [¹¹C]4 in the brain was 0.62 SUV in a wild-type mouse and 1.59 SUV in a Pgp/BCRP-KO mouse. These results suggest that brain uptake of [¹¹C]4 is probably limited by brain efflux transporters, similar to that reported for [¹¹C]3_.¹⁹ Because of its high-level specific binding to mGlu₂ in vitro and its moderate level of brain permeability (>1 SUV), we selected [¹¹C]5i for further in vivo PET evaluation to assess its potential for imaging mGlu₂ in the brain.

Figure 6.

Summed (0–60 min) PET/MRI-fused images and time-activity curves of [¹¹C]5i (A and B) and [¹¹C]4 (C and D) in the whole brains of wild-type and Pgp/BCRP-knockout mice.

Biodistribution Study

Prior to PET assessment using rat, an ex vivo biodistribution study using [¹¹C]5i was performed to evaluate pharmacokinetics in peripheral organs and brain uptake in mice. The radioactivity uptake values are expressed in Table 2 as the percentage of the injected dose per gram of wet tissue (%ID/g). At 1 min after the injection, high uptake values (>5%ID/g) were observed in the liver and kidney, and moderate uptake values (>3%ID/g) were seen in the heart, lung, pancreas, and small intestine. Subsequently, radioactivity in the blood, heart, lung, kidney, muscle, and brain rapidly decreased, while the intestinal radioactivity decreased slowly. On the other hand, high-level radioactive accumulation was retained in the pancreas up to 60 min after the injection, suggesting that radioactive metabolites of [¹¹C]5i might have unknown binding sites in the pancreas. Our pharmacokinetic results also suggest that the elimination of radioactivity was primarily a result of hepatobiliary and urinary excretion, as well as the intestinal reuptake pathway. In the mouse brain, the initial uptake of [¹¹C]5i was 2.11%ID/g, which was higher than that of [¹¹C]3 (1.06%ID/g).¹⁹ Moreover, radioactive accumulations of [¹¹C]5i in peripheral organs (heart, lung, and kidney) was much lower than that of [¹¹C]3 (>10%ID/g at 1 min). This suggests that [¹¹C]5i permits radiation exposure to be kept within safe levels, a favorable property for a PET ligand in putative clinical use.

Table 2.

Biodistribution of [¹¹C]5i in mice (n = 3, mean ± SD).

%ID/g Organ	Time after the injection (min)
	1	5	15	30	60
Blood	2.17 ± 0.25	1.48 ± 0.16	0.77 ± 0.06	0.65 ± 0.01	0.76 ± 0.13
Heart	4.04 ± 0.51	2.03 ± 0.12	1.11 ± 0.07	0.92 ± 0.09	0.86 ± 0.01
Lung	4.47 ± 0.89	2.61 ± 0.14	1.66 ± 0.12	1.32 ± 0.14	1.32 ± 0.09
Liver	5.58 ± 0.89	7.31 ± 0.83	5.70 ± 0.45	4.90 ± 0.44	4.50 ± 0.37
Pancreas	4.15 ± 0.09	3.61 ± 0.35	5.07 ± 0.38	5.24 ± 0.23	5.87 ± 0.32
Spleen	2.30 ± 0.16	1.70 ± 0.11	1.57 ± 0.20	1.77 ± 0.26	1.84 ± 0.36
Kidney	7.18 ± 0.30	3.74 ± 0.36	2.62 ± 0.14	2.06 ± 0.11	2.06 ± 0.08
Stomach	1.38 ± 0.33	1.18 ± 0.15	1.20 ± 0.31	1.98 ± 0.47	2.95 ± 1.53
S. Intestine	4.06 ± 0.16	5.34 ± 0.44	5.93 ± 0.49	5.51 ± 1.60	4.18± 0.77
L. Intestine	1.56 ± 0.19	1.99 ± 0.49	1.77 ± 0.37	1.74 ± 0.28	1.68 ± 0.20
Testis	0.78 ± 0.12	0.82 ± 0.05	0.51 ± 0.02	0.32 ± 0.04	0.31 ± 0.05
Muscle	2.24 ± 0.24	1.11 ± 0.05	0.69 ± 0.07	0.49 ± 0.06	0.49 ± 0.02
Brain	2.11 ± 0.08	1.09 ± 0.08	0.64 ± 0.02	0.63 ± 0.03	0.55 ± 0.10

PET Studies in Rats

Dynamic PET scans for rats were performed for 30 min. Figure 7 shows averaged PET/magnetic resonance imaging (MRI)-fused images scaled by the distribution volume ratio (DVR), time-activity curves (TACs) of [¹¹C]5i in the control (A and D), self-blocking (B and E), and the compound 4-treated (C and F) subjects (n = 4 per group), and respective quantitative data of DVR in the striatum, hippocampus, cerebral cortex, and cerebellum (G). In control subjects, the radioactivity in the brain regions peaked at 2 min and rapidly decreased after that. Although no significant differences were seen in the TACs among the cortex, striatum, hippocampus, cerebellum, and pons, the radioactive accumulations in the cortex and striatum were slightly higher than those in cerebellum and pons. As shown in Figure 7A, high signals were observed in the cerebral cortex and striatum of parametric PET images (scaled by the DVR, and using the TAC of pons as a reference). With pre-administration of unlabeled 5i (1 mg/kg) or compound 4 (1 mg/kg), there was a decrease with significant differences (p < 0.05) excepting cerebellum in heterogeneous uptake of [¹¹C]5i among different brain regions; instead, the distribution became homogenous across brain regions. Although treatment with elacridar (5 mg/kg), a potent inhibitor for Pgp and BCRP, slightly enhanced specific binding of [¹¹C]5i in cortex and striatum (Figure S1 in Supplemental information), overall, the ligand developed in this study, [¹¹C]5i, demonstrated good in vivo binding specificity towards mGlu₂.

Figure 7.

Averaged parametric PET/MRI-fused images, time-activity curves (n = 4, mean ± SD) of [¹¹C]5i in control (A and D), self-blocking (B and E), and 4-blocked (C and F) rat brains, and quantitative data for distribution volume ratio (DVR) in brain regions (G). The DVR was estimated using the TAC from the reference region (pons). ***p < 0.001, *p < 0.05 (vs Control)

Metabolite Analysis

To confirm whether radioactive metabolites were included in brain uptake of radioactivity, we performed ex vivo metabolite analysis with [¹¹C]5i using plasma and brain homogenates. Figure 8 shows the percentages of unchanged [¹¹C]5i (retention time, t_R = 5.5 min) accounting for the total radioactivity in plasma and brain homogenates at 5 min and at 20 min after the injection of [¹¹C]5i. The fractions corresponding to unchanged [¹¹C]5i in plasma were 60% and 17% at 5 min and at 20 min, respectively. Two polar radiolabeled metabolites (t_R = 2.4 and 4.3 min) were detected by reversed-phase HPLC. On the other hand, the parent form of [¹¹C]5i in brain homogenate was 96% at 5 min and 56% at 20 min. Overall, our results show reasonable in vivo stability for [¹¹C]5i in PET studies.

Figure 8.

Percentages (n = 3, mean ± SD) of the unchanged form of [¹¹C]5i in the plasma and brain homogenate at 5 min and at 20 min after the injection.

CONCLUSION

In this study, we identified compounds 5i and 4 as new PET ligand candidates for mGlu₂ imaging, and subsequently radiolabeled these compounds with ¹¹C in high radiochemical yields and observed high molar activity. Of the two candidates, [¹¹C]5i (also known as [¹¹C]MG2-1812) exhibited high potency, high subtype selectivity, and favorable lipophilicity, and further demonstrated high specific binding of mGlu₂ in vitro. Moreover, comprehensive in vivo evaluation, including PET measurements in rats and Pgp/BCRP-KO mice, and ex vivo whole-body distribution and radiometabolite analysis, indicated that [¹¹C]5i successfully visualized mGlu₂-positive brain regions in vivo. Further optimization of new chemical scaffold based on compound 5i to improve affinity and specific binding are underway to achieve optimal binding potential and brain kinetics. Comprehensive PET study and kinetic modeling studies in higher species is also necessary for potential human translation.

EXPERIMENTAL SECTION

All chemical reagents and organic solvents were purchased from Sigma-Aldrich (St. Louis, MO, USA), FUJIFILM Wako Pure Chem. (Osaka, Japan), or Nacalai Tesque (Kyoto, Japan) and used without further purification. The authentic compound 4 was synthesized according to the procedure reported previously.¹⁹ The commercially-available compound 20 (MNI-137) was purchased from Nacalai Tesque. ¹¹C was produced using a cyclotron (CYPRIS HM-18; Sumitomo Heavy Industries, Tokyo, Japan). ¹H NMR (300 MHz) spectra were recorded on a JEOL-AL-300 spectrometer (JEOL, Tokyo, Japan) using tetramethylsilane (TMS) as an internal standard. All chemical shifts (δ) are reported as the ppm downfield relative to the TMS signal. Signals are quoted as s (singlet), d (doublet), t (triplet), br (broad), or m (multiplet). High-resolution fast atom bombardment mass spectra (HRMS) were acquired using a JEOL NMS-SX102 102A spectrometer. Silica gel column chromatography was performed using Wakosil C-200 (FUJIFILM Wako Pure Chem.). HPLC separation and analysis were performed using a JASCO HPLC system (JASCO, Tokyo, Japan). All semi-preparative HPLC separations were performed using CAPCELL PAK C₁₈ columns (10 mm ID × 250 mm; Shiseido, Tokyo, Japan). All HPLC analyses were performed by using CAPCELL PAK C₁₈ columns (4.6 mm ID × 250 mm; Shiseido) to determine the chemical purities (>98%) of 5a–5i.

Chemical Synthesis

4-(2-Fluoro-4-methoxyphenyl)-5-((pyridin-4-yl)methoxy)picolinonitrile (7a).

NaH (60%, 24 mg, 0.6 mmol) was added to a solution of pyridin-4-ylmethanol in DMF (1.5 mL) at 0 °C. The mixture was warmed to room temperature and stirred for 15 min before being cooled to 0°C, and 5-fluoro-4-(2-fluoro-4-methoxyphenyl)picolinonitrile¹⁸ (6; 74 mg, 0.3 mmol) was added in DMF (1.5 mL) in 5 min. The mixture was stirred overnight before being quenched with water (10 mL) and extracted with ethyl acetate (10 mL, 3 times). The combined organic layers were washed with saturated NaCl, dried over MgSO₄, and concentrated in vacuo. The residue was purified by flash chromatography on silica gel (hexanes to ethyl acetate gradient column) to yield the compound 7a as a white solid (54 mg, 51%; R_f = 0.3; hexanes/ethyl acetate = 2:1). ¹H NMR (300 MHz, CDCl₃): δ = 8.82 (d, J = 6.0 Hz, 2H), 8.56 (s, 1H), 7.91 (d, J = 6.0 Hz, 2H), 7.71 (s, 1H), 7.32 (d, J = 8.5 Hz, 1H), 6.94–6.72 (m, 2H), 5.58 (s, 2H), 3.91 (s, 3H). ¹³C NMR (75 MHz, DMSO-d₆): δ = 162.2, 162.0, 160.3 (d, J = 247.8 Hz), 154.1, 148.8, 147.3, 137.8, 133.2, 132.5 (d, J = 4.9 Hz), 131.2, 125.7, 122.2, 117.9, 113.5 (d, J = 15.4 Hz), 111.2 (d, J = 2.8 Hz), 102.2 (d, J = 25.7 Hz), 69.2, 56.3.

4-(2-Fluoro-4-methoxyphenyl)-5-((2-fluoropyridin-4-yl)methoxy)picolinonitrile (7b).

The procedure described for the synthesis of 7a was applied to (2-fluoropyridin-4-yl)methanol to give 7b (54%) as a white solid. ¹H NMR (300 MHz, DMSO-d₆): δ = 8.66 (s, 1H), 8.25 (d, J = 5.1 Hz, 1H), 8.06 (s, 1H), 7.49 (t, J = 8.6 Hz, 1H), 7.28 (dt, J = 5.4, 1.6 Hz, 1H), 7.08–6.87 (m, 3H), 5.53 (s, 2H), 3.84 (s, 3H).¹³C NMR (75 MHz, DMSO-d₆): δ = 163.7 (d, J = 227.5 Hz), 162.0 (d, J = 13.8 Hz), 160.3 (d, J = 258.0 Hz), 154.0, 152.1 (d, J = 8.1 Hz), 148.4 (d, J = 15.4 Hz), 137.7, 133.3, 132.5 (d, J = 4.9 Hz), 131.2, 125.8, 120.0 (d, J = 3.9 Hz), 117.9, 113.5 (d, J = 15.6 Hz), 111.2 (d, J = 2.8 Hz), 107.2 (d, J = 39.2 Hz), 102.2 (d, J = 25.6 Hz), 68.8 (d, J = 3.3 Hz), 56.3.

4-(2-Fluoro-4-methoxyphenyl)-5-((2-methoxypyridin-4-yl)methoxy)picolinonitrile (7c).

The procedure described for the synthesis of 7a was applied to (2-methoxypyridin-4-yl)methanol to give 7c (40%) as a white solid. ¹H NMR (300 MHz, CDCl₃): δ = 8.37 (s, 1H), 8.16 (d, J = 5.3 Hz, 1H), 7.66 (s, 1H), 7.31 (t, J = 8.5 Hz, 1H), 6.89–6.80 (m, 2H), 6.77 (dd, J = 2.5, 11.9 Hz, 1H), 6.70 (s, 1H), 5.23 (s, 2H), 3.94 (s, 3H), 3.88 (s, 3H). ¹³C NMR (75 MHz, CDCl₃): δ = 164.5, 160.4 (d, J = 260.6 Hz), 153.7, 147.2, 147.1, 136.4, 134.0, 131.4 (d, J = 4.7 Hz), 130.6 (d, J = 2.0 Hz), 126.7, 117.2, 114.1, 113.3 (d, J = 15.7 Hz), 110.4 (d, J = 3.0 Hz), 108.1, 102.0 (d, J = 25.6 Hz), 69.3, 55.7, 53.7.

4-(2-Fluoro-4-methoxyphenyl)-5-((6-fluoropyridin-3-yl)methoxy)picolinonitrile (7d).

The procedure described for the synthesis of 7a was applied to (6-fluoropyridin-3-yl)methanol to give 7d (43%) as a white solid. ¹H NMR (300 MHz, CDCl₃): δ = 8.49 (s, 1H), 8.22 (d, J = 2.4 Hz, 1H), 7.76 (td, J = 8.0, 2.5 Hz, 1H), 7.65 (s, 1H), 7.25 (t, J = 8.5 Hz, 1H), 6.95 (dd, J = 8.5, 3.0 Hz, 1H), 6.82–6.66 (m, 2H), 5.25 (s, 2H), 3.85 (s, 3H). ¹³C NMR (75 MHz, CDCl₃): δ = 162.2, 160.3 (d, J = 256.9 Hz), 146.8 , 146.6, 140.4 (d, J = 8.3 Hz), 136.4, 134.2, 131.3 (d, J = 4.9 Hz), 130.6 (d, J = 1.9 Hz), 128.6 (d, J = 4.7 Hz), 126.9, 117.1, 113.2 (d, J = 15.4 Hz), 110.4 (d, J = 3.0 Hz), 110.1, 109.6, 102.0 (d, J = 25.5 Hz), 68.2, 55.7.

4-(2-Fluoro-4-methoxyphenyl)-5-((5-methyl-1,2,4-oxadiazol-3-yl)methoxy)-picolinonitrile (7g).

The procedure described for the synthesis of 7a was applied to (5-methyl-1,2,4-oxadiazol-3-yl)methanol to give 7g (45%) as a white solid. ¹H NMR (300 MHz, CDCl₃): δ = 8.59 (s, 1H), 7.67 (d, J = 1.3 Hz, 1H), 7.36 (t, J = 8.5 Hz, 1H), 6.88–6.67 (m, 2H), 5.33 (s, 2H), 3.86 (s, 3H), 2.61 (s, 3H). ¹³C NMR (75 MHz, CDCl₃): δ = 177.7, 165.6, 162.0 (d, J = 11.1 Hz), 160.3 (d, J = 250.4 Hz), 153.5, 135.7 (d, J = 244.2 Hz), 131.7 (d, J = 4.6 Hz), 130.7 (d, J = 2.7 Hz), 127.1, 117.1, 113.0 (d, J = 14.8 Hz), 110.4 (d, J = 3.0 Hz), 102.0 (d, J = 25.7 Hz), 62.3, 55.7, 12.4.

4-(2-Fluoro-4-methoxyphenyl)-5-((1-methyl-1H-pyrazol-3-yl)methoxy)picolinonitrile (7i).

The procedure described for the synthesis of 7a was applied to (1-methyl-1H-pyrazol-3-yl)methanol to give 7i (19 %) as a white solid. ¹H NMR (300 MHz, CDCl₃): δ = 8.59 (s, 1H), 7.60 (s, 1H), 7.27–7.33 (m, 2H), 6.70–6.79 (m, 2H), 6.23 (d, J = 2.2 Hz, 1H), 5.27 (s, 2H), 3.88 (s, 3H), 3.85 (s, 3H). ¹³C NMR (75 MHz, CDCl₃): δ = 167.07, 160.21 (d, J = 226.2 Hz), 156.18, 154.19, 146.83, 137.34, 131.65 (d, J = 5.0 Hz), 131.37, 130.48, 125.94, 117.49, 114.11 (d, J = 46.6 Hz), 110.22 (d, J = 3.1Hz), 105.32, 101.94 (d, J = 26.1 Hz), 65.56, 55.70, 38.95. HRMS (FAB): m/z calcd for [M+H] C₁₈H₁₆FN₄O₂, 339.1257, found 339.1271.

4-(2-Fluoro-4-methoxyphenyl)-5-((pyridin-4-yl)methoxy)picolinamide (5a).

Nitrile 7a (35 mg, 0.1 mmol) in DMSO (1 mL) was added to NaOH (6 mol/L, 50 μL). The solution was cooled to 0°C before the addition of H₂O₂ (30% in H₂O, 23 μL, 0.2 mmol). The mixture was stirred at room temperature for 20 min before being quenched with water (10 mL) and extracted with ethyl acetate (5 mL, 3 times). The combined organic layers were washed with saturated NaCl, dried over MgSO₄, and concentrated in vacuo. The residue was purified by flash chromatography on silica gel (hexanes to ethyl acetate gradient column) to yield the compound 5a as a white solid (25.2 mg, 70%; R_f = 0.3; hexane/ethyl acetate = 1:1). ¹H NMR (300 MHz, DMSO-d₆): δ = 8.62–8.53 (m, 2H), 8.46 (s, 1H), 8.01 (s, 1H), 7.91 (s, 1H), 7.56 (s, 1H), 7.48 (t, J = 8.6 Hz, 1H), 7.37–7.29 (m, 2H), 7.06–6.86 (m, 2H), 5.44 (s, 2H), 3.83 (s, 3H). ¹³C NMR (75 MHz, DMSO-d₆): δ = 166.0, 161.7, 161.6, 160.3 (d, J = 247.0 Hz), 153.6, 150.2, 145.8, 144.2, 134.4, 133.0, 132.3 (d, J = 5.0 Hz), 124.4, 121.8, 114.9 (d, J = 15.5 Hz), 114.3, 111.1, 102.2 (d, J = 25.8 Hz), 69.1, 56.2.

4-(2-Fluoro-4-methoxyphenyl)-5-((2-fluoropyridin-4-yl)methoxy)picolinamide (5b).

The procedure described for the synthesis of 5a was applied to 7b with H₂O₂ to give 5b (54%) as a white solid. ¹H NMR (300 MHz, CD₃OD): δ = 8.47 (s, 1H), 8.17 (d, J = 5.2 Hz, 1H), 8.05 (s, 1H), 7.41 (t, J = 8.5 Hz, 1H), 7.28 (d, J = 5.3 Hz, 1H), 7.02 (s, 1H), 6.96–6.80 (m, 2H), 5.43 (s, 2H), 3.88 (s, 3H). ¹³C NMR (75 MHz, CD₃OD): δ = 167.5, 163.9 (d, J = 238.3 Hz), 162.0 (d, J = 1.8 Hz), 160.4 (d, J = 257.1 Hz), 160.4 , 153.8, 152.6 (d, J = 8.2 Hz), 147.2 (d, J = 14.4 Hz), 143.3, 133.9 (d, J = 3.6 Hz), 131.5 (d, J = 5.0 Hz), 124.2 , 118.9 (d, J = 3.9 Hz), 114.6 (d, J = 15.9 Hz), 110.0 (d, J = 2.9 Hz), 106.5 (d, J = 38.4 Hz), 101.2 (d, J = 25.9 Hz), 68.4, 54.8.

4-(2-Fluoro-4-methoxyphenyl)-5-((2-methoxypyridin-4-yl)methoxy)picolinamide (5c).

The procedure described for the synthesis of 5a was applied to 7c with H₂O₂ to give 5c (60%) as a white solid. ¹H NMR (300 MHz, DMSO-d₆): δ = 8.44 (s, 1H), 8.14 (d, J = 5.3 Hz, 1H), 8.06–7.96 (m, 1H), 7.90 (s, 1H), 7.63–7.52 (m, 1H), 7.46 (t, J = 8.6 Hz, 1H), 6.99 (dd, J = 12.4, 2.5 Hz, 1H), 6.96–6.87 (m, 2H), 6.73 (s, 1H), 5.39 (s, 2H), 3.84 (s, 3H), 3.82 (s, 3H). ¹³C NMR (75 MHz, DMSO–d₆): δ = 166.0, 161.7 (d, J = 11.3 Hz), 160.3 (d, J = 246.8 Hz), 153.5, 149.1, 147.5, 144.2, 134.4, 133.0, 132.3 (d, J = 5.1 Hz), 124.4, 115.2, 114.9 (d, J = 15.7 Hz), 111.1 (d, J = 2.8 Hz), 108.0, 102.1 (d, J = 25.7 Hz), 68.9, 56.2, 53.6.

4-(2-Fluoro-4-methoxyphenyl)-5-((6-fluoropyridin-3-yl)methoxy)picolinamide (5d)

The procedure described for the synthesis of 5a was applied to 7d with H₂O₂ to give 5d (73%) as a white solid. ¹H NMR (300 MHz, acetone-d₆): δ = 8.56 (s, 1H), 8.30 (d, J = 2.4 Hz, 1H), 8.09–7.95 (m, 2H), 7.55–7.38 (m, 1H), 7.10 (dd, J = 8.4, 2.9 Hz, 1H), 6.94–6.80 (m, 2H), 5.45 (s, 2H), 3.87 (s, 3H). ¹³C NMR (75 MHz, acetone–d₆): δ = 165.5, 163.4 (d, J = 224.5 Hz), 161.7 (d, J = 11.2 Hz), 160.2 (d, J = 235.2 Hz), 146.9 (d, J = 15.7 Hz), 144.1, 141.2 (d, J = 8.1 Hz), 133.9, 133.4, 131.9 (d, J = 5.1 Hz), 130.3 (d, J = 4.4 Hz), 124.2 (d, J = 1.7 Hz), 114.9 (d, J = 15.7 Hz), 110.3 (d, J = 3.0 Hz), 109.3 (d, J = 38.3 Hz), 101.5 (d, J = 26.0 Hz), 67.8 (d, J = 1.7 Hz), 55.2.

4-(2-Fluoro-4-methoxyphenyl)-5-((1-methyl-1H-pyrazol-3-yl)methoxy)picolinamide (5i).

The procedure described for the synthesis of 5a was applied to 7i with KOSiMe₃ to give 5i (70%) as a white solid. ¹H NMR (300 MHz, CDCl₃): δ = 8.42 (s, 1H), 7.70 (brs, 1H), 7.28–7.34 (m, 2H), 6.74 (ddd, J = 2.5, 9.2, 17.3 Hz, 2H), 6.29 (d, J = 2.2 Hz, 1H), 5.56 (brs, 1H), 5.30 (s, 2H), 3.88 (s, 3H), 3.84 (s, 3H). ¹³C NMR (75 MHz, CDCl₃): δ = 166.72, 161.31 (d, J = 11.2 Hz), 160.04 (d, J = 249.7 Hz), 154.39, 147.53, 142.62, 143.02, 133.85, 131.73 (d, J = 5.0 Hz), 131.24, 124.00 (d, J = 1.9 Hz), 115.21 (d, J = 15.5 Hz), 109.89 (d, J = 3.1 Hz), 105.20 (d, J =1.9 Hz), 101.74 (d, J = 25.5 Hz), 65.48, 55.58, 38.84.

4-(2-Fluoro-4-methoxyphenyl)-5-((6-methoxypyridin-3-yl)methoxy)picolinamide (5e).

NaH (60%, 24 mg, 0.6 mmol) was added to a solution of (6-methoxypyridin-3-yl)methanol in DMF (1.5 mL) at 0°C. The mixture was warmed to room temperature and stirred for 15 min before being cooled to 0°C and 6 (74 mg, 0.3 mmol) in DMF (1.5 mL) was added in 5 min. The mixture was stirred overnight before being quenched with water (10 mL) and extracted with ethyl acetate (10 mL, 3 times). The combined organic layers were washed with saturated NaCl, dried over MgSO₄, and concentrated in vacuo. The residue was used for the next step without further purification. To a solution of the crude 4-(2-fluoro-4-methoxyphenyl)-5-((6-methoxypyridin-3-yl)methoxy)picolinonitrile (7e) in DMSO (1 mL), NaOH was added (6 mol/mL, 50 μL). The solution was cooled to 0°C before the addition of H₂O₂ (30% in H₂O, 23 μL, 0.2 mmol). The mixture was stirred at room temperature for 20 min before being quenched with water (10 mL) and extracted with ethyl acetate (5 mL, 3 times). The combined organic layers were washed with saturated NaCl, dried over MgSO₄ and concentrated in vacuo. The residue was purified by flash chromatography on silica gel (hexanes to ethyl acetate gradient column) to yield the compound 5e as a white solid (17.5 mg, 16% for two steps; R_f = 0.3; hexane/ethyl acetate = 1:1). ¹H NMR (300 MHz, DMSO-d₆): δ = 8.56 (s, 1H), 8.20 (d, J = 2.3 Hz, 1H), 8.01 (s, 1H), 7.87 (d, J = 5.3 Hz, 1H), 7.70 (dd, J = 8.6, 2.4 Hz, 1H), 7.55 (s, 1H), 7.41 (t, J = 8.6 Hz, 1H), 6.99–6.80 (m, 3H), 5.30 (s, 2H), 3.83 (s, 3H), 3.80 (s, 3H). ¹³C NMR (75 MHz, DMSO-d₆): δ = 166.1, 163.9, 161.5 (d, J = 11.3 Hz), 160.2 (d, J = 247.0 Hz), 153.9, 146.9, 143.9, 139.6, 134.6, 133.0, 132.3 (d, J = 5.1 Hz), 125.2, 124.4, 115.0 (d, J = 15.7 Hz), 111.0, 102.1 (d, J = 25.9 Hz), 68.4, 56.2, 53.7.

4-(2-Fluoro-4-methoxyphenyl)-5-((2-methoxypyrimidin-5-yl)methoxy)picolinamide (5f).

The procedure described for the synthesis of 5e was applied to (2-methoxypyrimidin-5-yl)methanol to give 5f (12%) as a white solid. ¹H NMR (300 MHz, DMSO-d₆): δ = 8.62 (s, 2H), 8.60 (s, 1H), 8.12–7.98 (m, 1H), 7.89 (s, 1H), 7.58 (s, 1H), 7.42 (t, J = 8.6 Hz, 1H), 7.04–6.80 (m, 2H), 5.34 (s, 2H), 3.91 (s, 3H), 3.81 (s, 3H). ¹³C NMR (75 MHz, DMSO-d₆): δ = 166.0, 165.4, 161.6 (d, J = 11.3 Hz), 160.2 (d, J = 247.1 Hz), 159.8, 153.7, 144.2, 134.7, 133.1, 132.3 (d, J = 5.2 Hz), 124.4, 123.7, 114.9 (d, J = 15.7 Hz), 111.0 (d, J = 2.8 Hz), 102.1 (d, J = 25.9 Hz), 66.2, 56.2, 55.2.

4-(2-Fluoro-4-methoxyphenyl)-5-((5-methyl-1,2,4-oxadiazol-3-yl)methoxy)picolinamide (5g).

The procedure described for the synthesis of 5e was applied to 5-methyl-1,2,4-oxadiazol-3-yl)methanol to give 5g (18%) as a white solid. ¹H NMR (300 MHz, DMSO-d₆): δ = 8.60 (s, 1H), 8.03 (d, J = 2.5 Hz, 1H), 7.89 (d, J = 1.1 Hz, 1H), 7.65–7.51 (m, 1H), 7.43 (t, J = 8.7 Hz, 1H), 7.01–6.84 (m, 2H), 5.52 (s, 2H), 3.82 (s, 3H), 2.59 (s, 3H). ¹³C NMR (75 MHz, DMSO-d₆): δ = 178.2, 166.6, 165.9, 161.6 (d, J = 11.3 Hz), 160.2 (d, J = 247.5 Hz), 153.3, 144.5, 135.1, 133.0, 132.3 (d, J = 5.0 Hz), 124.4, 114.7 (d, J = 15.2 Hz), 111.0 (d, J = 2.8 Hz), 102.2 (d, J = 25.8 Hz), 62.2, 56.2, 12.4.

4-(2-Fluoro-4-methoxyphenyl)-5-((1-methyl-1H-imidazol-4-yl)methoxy)picolinamide (5h).

The procedure described for the synthesis of 5e was applied to (1-methyl-1H-imidazol-4-yl)methanol to give 5h (15%) as a white solid. ¹H NMR (300 MHz, CDCl₃): δ = 8.55 (d, J = 2.9 Hz, 1H), 7.64 (d, J = 2.9 Hz, 1H), 7.44 (s, 1H), 7.20 (t, J = 8.4 Hz, 1H), 6.74 (dd, J = 21.3, 10.5 Hz, 2H), 6.31 (s, 1H), 5.24 (d, J = 2.9 Hz, 2H), 3.86 (t, J = 2.2 Hz, 3H), 3.78 (t, J = 2.2 Hz, 3H). ¹³C NMR (75 MHz, CDCl₃): δ = 162.1, 162.0, 160.2 (d, J = 249.8 Hz), 153.6, 138.2, 136.7, 135.4, 134.3, 131.3 (d, J = 4.9 Hz), 130.6 (d, J = 1.9 Hz), 126.9, 117.1, 113.2 (d, J = 15.6 Hz), 110.4 (d, J = 3.0 Hz), 107.8, 102.1, 101.2, 61.9, 55.7, 36.6.

4-(2-Fluoro-4-(methoxymethoxy)phenyl)-5-((1-methyl-1H-pyrazol-3-yl)methoxy)picolinonitrile (10).

The procedure described for the synthesis of 7a was applied to (1-methyl-1H-pyrazol-3-yl)methanol and 5-fluoro-4-(2-fluoro-4-(methoxymethoxy)phenyl)picolinonitrile¹⁹ (8) to give 10 (96%) as a white solid. ¹H NMR (300 MHz, CDCl₃): δ = 8.60 (s, 1H), 7.60 (s, 1H), 7.23–7.32 (m, 2H), 6.86–6.96 (m, 2H), 6.23 (d, J = 2.2 Hz, 1H), 5.27 (s, 2H), 5.21 (s, 2H), 3.88 (s, 3H), 3.51 (s, 3H). ¹³C NMR (75 MHz, CDCl₃): δ = 159.36, 154.18, 146.77, 137.37, 133.44, 131.62 (d, J = 5.0 Hz), 131.38, 130.45 (d, J = 2.5 Hz), 125.91, 117.45, 114.87 (d, J = 14.9 Hz), 112.13 (d, J = 3.7 Hz), 105.35, 105.34, 104.24 (d, J = 26.1 Hz), 94.43, 65.54, 56.28, 38.94. HRMS (FAB): m/z calcd for [M+H] C₁₉H₁₈FN₄O₃, 369.1363, found 369.1322.

4-(2-Fluoro-4-(methoxymethoxy)phenyl)-5-((1-methyl-1H-pyrazol-3-yl)methoxy)-picolinamide (11).

The procedure described for the synthesis of 5a was applied to 10 with KOSiMe₃ to give 11 (77%) as a white solid. ¹H NMR (300 MHz, CDCl₃): δ = 8.42 (s, 1H), 8.13 (s, 1H), 7.70 (brs, 1H), 7.28–7.34 (m, 2H), 6.84–6.91 (m, 2H), 6.24 (d, J = 2.2 Hz, 1H), 5.26 (s, 2H), 5.20 (s, 2H), 3.88 (s, 3H), 3.51 (s, 3H). ¹³C NMR (75 MHz, CDCl₃): δ = 166.62, 160.34 (d, J = 232.3 Hz), 154.40, 147.53, 142.63, 133.93 (d, J = 21.7 Hz), 131.73 (d, J = 5.0 Hz), 131.27, 124.93, 116.32, 111.95, 111.91, 105.27, 105.25, 104.04 (d, J = 25.5 Hz), 94.47, 65.52, 56.21, 38.87. HRMS (FAB) m/z calcd for [M+H] C₁₉H₂₀FN₄O₄, 387.1469, found 387.1419

4-(2-Fluoro-4-hydroxyphenyl)-5-((1-methyl-1H-pyrazol-3-yl)methoxy)picolinamide (12).

To a solution of 11 (200 mg, 0.52 mmol) in CH₃OH (10 mL), hydrochloric acid (1 mL, 2 mol/mL) was added. The reaction mixture was stirred at 40°C for 10 h. After cooling, MeOH was removed from the reaction mixture. The resultant residue was suspended in CH₂Cl₂ (10 mL) and triethylamine (1 mL) was added. After removal of the reaction solvent under reduced pressure, the resultant solid was filtered, washed with water, and dried to obtain 12 as a light gray solid (159 mg, 94%). ¹H NMR (300MHz, DMSO-d₆): δ = 10.15 (s, 1H), 8.58 (s, 1H), 7.99 (brs, 1H), 7.84 (s, 1H), 7.65 (d, J = 1.8 Hz, 1H), 7.52 (brs, 1H), 7.29 (t, J = 8.6 Hz, 1H), 6.68 (t, J = 9.3 Hz, 2H), 6.21 (d, J = 2.2 Hz, 1H), 5.25 (s, 2H), 3.81 (s, 3H). ¹³C NMR (75 MHz, CDCl₃): δ = 165.70, 164.26, 159.51 (d, J = 11.8 Hz), 158.12, 153.57, 144.95 (d, J = 259.1 Hz), 134.30 (d, J = 9.9 Hz), 132.72, 131.88 (d, J = 5.6 Hz), 124.02 (d, J = 30.4 Hz), 113.06 (d, J = 15.5 Hz), 111.66 (d, J = 2.5 Hz), 104.97, 102.67 (d, J = 24.2Hz), 70.83, 64.87, 38.46. HRMS (FAB) m/z calcd for [M+H] C₁₇H₁₆FN₄O₃, 343.1206, found 343.1128.

4-(4-Fluorophenyl)-5-((1-trityl-1H-pyrazol-3-yl)methoxy)picolinonitrile (15).

The procedure described for the synthesis of 7a was applied to 14 and 5-fluoro-4-(4-fluorophenyl)picolinonitrile¹⁹ (13) to yield 15 (77 %) as a white solid. ¹H NMR (300 MHz, CDCl₃): δ = 8.51 (s, 1H), 7.58 (s, 1H), 7.50–7.55 (m, 2H), 7.28–7.30 (m, 9H), 7.06–7.11 (m, 9H), 6.21 (d, J = 1.8 Hz, 1H), 5.32 (s, 2H). ¹³C NMR (75 MHz, CDCl₃): δ = 163.03 (d, J = 249.8 Hz), 153.42, 146.75, 142.85, 137.53 (d, J = 21.7 Hz), 133.90, 131.01 (d, J = 8.1 Hz), 129.98, 129.22, 127.80, 127.72, 127.66, 126.26, 117.49, 115.67, 115.38, 103.83, 78.83, 65.48. HRMS (FAB) m/z calcd for [M+H]+ C₃₅H₂₆FN₄O, 537.2091, found 537.2068.

4-(4-Fluorophenyl)-5-((1-trityl-1H-pyrazol-3-yl)methoxy)picolinamide (16).

The procedure described for the synthesis of 5a was applied to 15 with KOSiMe₃ to give 16 (48%) as a white solid. ¹H NMR (300 MHz, DMSO-d₆): δ = 8.56 (s, 1H), 8.00 (brs, 1H), 7.94 (s, 1H), 7.63–7.68 (m, 2H), 7.55 (brs, 1H), 7.18–7.33 (m, 12H), 6.99–7.02 (m, 6H), 6.38 (d, J = 1.5 Hz, 1H), 3.32 (s, 2H). ¹³C NMR (75 MHz, DMSO-d₆): δ = 165.74, 160.45, 152.76, 147.39, 143.72, 142.75, 136.27, 135.13, 133.51, 131.29, 131.19, 129.53, 127.76, 122.74, 115.33, 115.05, 104.74, 78.02, 64.89. HRMS (FAB) m/z calcd for [M+H] C₃₅H₂₈FN₄O₂, 555.2196, found 555.2200

4-(4-Fluorophenyl)-5-(1H-pyrazol-3-yl)methoxy)picolinamide (17).

The procedure described for the synthesis of 12 was applied to 11 with HCl to give 17 (93%) as a white solid. ¹H NMR (300 MHz, DMSO-d₆): δ = 12.86 (s, 1H), 8.63 (s, 1H), 8.00 (s, 1H), 7.94 (s, 1H), 7.68–7.72 (m, 3H), 7.54 (brs, 1H), 7.29 (t, J = 8.8 Hz, 2H), 6.34 (s, 1H), 5.36 (s, 2H). ¹³C NMR (75 MHz, DMSO-d₆): δ = 170.74, 165.93, 163.90, 153.12, 143.73, 136.27, 134.84, 131.43, 131.33, 124.55, 122.94, 115.58, 115.30, 96.91. HRMS (FAB) m/z calcd for [M+H] C₁₆H₁₄FN₄O₂, 313.1101, found 313.1138.

Ethyl-1-trityl-1H-pyrazole-3-carboxylate (19).

A THF (50 mL) solution of ethyl 1H-pyrazole-3-carboxylate (18; 1.10 g, 7.8 mmol) was added slowly to the mixture of NaH (640 mg, 16 mmol) in THF (5 mL). The mixture was stirred at room temperature for 30 min and a THF (10 mL) solution of trityl chloride (2.23 g, 8.0 mmol) was added and then stirred at 40°C for 4 h. The reaction mixture was quenched by brine and crude product was extracted with ethyl acetate (3 times). The organic layer was washed with brine and dried over MgSO₄ and evaporated. Purification by column chromatography on silica gel (n-hexane/ethyl acetate = 95/5 to 90/10) produced 19 (0.93 g, 31%) as a white solid. ¹H NMR (300 MHz, CD₃OD): δ = 7.45 (d, J = 2.6 Hz, 1H), 7.30–7.36 (m, 9H), 7.10–7.13 (m, 6H), 6.79 (d, J = 2.6 Hz, 1H), 4.32 (q, J = 7.0 Hz, 2H), 1.34 (t, J = 7.1 Hz, 3H). ¹³C NMR (75 MHz, CD₃OD): δ = 164.02, 144.90, 144.05, 135.27, 129.28, 129.14, 129.00, 108.40, 81.13, 62.02, 14.56. FAB-MS m/z, 383 [M+H].

(1-Trityl-1H-pyrazole-3-yl) methanol (14).

A solution of 19 (930 mg, 2.43 mmol) in THF (13 mL) was added slowly to the suspension of LiAlH₄ (184.4 mg, 4.86 mmol) in THF (3 mL) and stirred at 40°C for 3 h. The reaction mixture was quenched with ice-water and crude product was extracted with ethyl acetate (3 times). The organic layer was washed with brine, dried over MgSO₄, and evaporated. Purification by column chromatography on silica gel (n-hexane/ethyl acetate = 90/10 to 75/25) produced 20 (529 mg, 64%) as a white solid. ¹H NMR (300 MHz, CDCl₃): δ = 7.25–7.32 (m, 10H), 7.11–7.17 (m, 6H), 6.21 (d, J = 2.6 Hz, 1H), 4.67 (d, J = 5.5 Hz, 2H), 2.08 (t, J = 5.9 Hz, 1H). ¹³C NMR (75 MHz, CDCl₃): δ = 152.41, 143.20, 133.60, 130.11, 127.70, 127.67, 102.64, 78.51, 59.37. FAB-MS m/z, 341 [M+H].

Radiochemistry

4-(2-Fluoro-4-[¹¹C]methoxyphenyl)-5-((1-methyl-1H-pyrazol-3-yl)methoxy)picolinamide ([¹¹C]5i).

[¹¹C]CO₂ was produced in a CYPRIS HM-18 cyclotron (Sumitomo Heavy Industries). [¹¹C]CH₃I for radiolabeling was synthesized from [¹¹C]CO₂, as described previously.^24,25 Briefly, [¹¹C]CO₂ was bubbled into 0.4 mol/mL LiAlH₄ in anhydrous THF (300 μL). After evaporation of THF, the residue was treated with 57% hydroiodic acid (300 μL) to produce [¹¹C]CH₃I, which was distilled at 180°C and then transferred by nitrogen gas into a solution of 11 (1 mg) and NaOH (0.5 mol/L, 6 μL) in anhydrous DMF (300 μL) at −15 to −20 °C. After radioactivity reached a plateau, the reaction mixture was heated at 80°C for 5 min. After the reaction, CH₃CN/H₂O/Et₃N (40/60/0.1%, v/v/v, 500 μL) was added and this mixture was loaded into a preparative HPLC system (JASCO) for separation. The HPLC conditions were as follows: column, CAPCELL PAK C₁₈ (10 mm ID × 250 mm); mobile phase, CH₃CN/H₂O/Et₃N (40/60/0.1%, v/v/v); flow rate, 5 mL/min; UV, 254 nm. The fraction corresponding to [¹¹C]5i (t_R, 9.0 min) was collected into a flask containing polysorbate-80 (75 μL) and ethanol (150 μL), evaporated to dryness in vacuo, re-dissolved in 5 mL of sterile normal saline, and passed through a 0.22 mm Millipore filter for analysis and application experiments.

The radiochemical purity of [¹¹C]5i was assayed by analytical HPLC as follows: column, CAPCELL PAK C₁₈ (4.6 mm ID × 250 mm; Shiseido); UV, 254 nm; mobile phase, CH₃CN/H₂O/Et₃N (40/60/0.1%, v/v/v). The t_R for [¹¹C]5i was 7.0 min at a flow rate of 1.0 mL/min. The identity of [¹¹C]5i was confirmed by co-injection with authentic unlabeled 5i. The molar activity was determined by comparing the assayed radioactivity to the mass associated with the 5i UV peak at 254 nm. The result of a typical batch was as follows: synthesis time, 40 min from EOB; radiochemical yield (decay–corrected), 48% based on [¹¹C]CO₂; radiochemical purity, >99%; and molar activity at EOS, 85 GBq/μmol.

4-(2-Fluoro-4-(methoxymethoxy)phenyl)-5-((1-[¹¹C]methyl-1H-pyrazol-3-yl)methoxy)-picolinamide ([¹¹C]4).

The [¹¹C]CH₃I produced was distilled and then transferred by nitrogen gas into a solution of 17 (1 mg) and Cs₂CO₃ (5–6 mg) in anhydrous DMF (300 μL), at −15 to −20°C. After radioactivity reached a plateau, the reaction mixture was heated at 90°C for 5 min. After the reaction, CH₃CN/H₂O/Et₃N (35/65/0.1%, v/v/v, 500 μL) was added and this mixture was loaded into a preparative HPLC system (JASCO) for separation. The HPLC conditions were as follows: column, Triart ExRS C₁₈ column (10 mm ID × 250 mm; YMC Kyoto, Kyoto, Japan); mobile phase, CH₃CN/H₂O/Et₃N (35/65/0.1%, v/v/v); flow rate, 4 mL/min; UV, 254 nm. The fraction corresponding to [¹¹C]4 (t_R, 9.5 min) was collected into a flask containing polysorbate-80 (75 μL) and ethanol (150 μL), evaporated to dryness in vacuo, re-dissolved in 5 mL of sterile normal saline, and passed through a 0.22 mm Millipore filter for analysis and application experiments.

The radiochemical purity of [¹¹C]4 was assayed by analytical HPLC as follows: column, Triart C₁₈ (4.6 mm ID × 250 mm; YMC Kyoto); UV, 254 nm; mobile phase, CH₃CN/H₂O/Et₃N (35/65/0.1%, v/v/v). The t_R for [¹¹C]4 was 9.9 min at a flow rate of 1.0 mL/min. The identity of [¹¹C]4 was confirmed by co-injection with authentic unlabeled 4. The result of a typical batch was as follows: synthesis time, 45 min from EOB; radiochemical yield (decay-corrected), 13% based on [¹¹C]CO₂; radiochemical purity, 97%; and molar activity at EOS, 72 GBq/μmol.

Pharmacology

Cell line generation and thallium flux assays.

In order to generate stable human mGlu₂ and mGlu₃ cell lines for thallium flux assays, human mGlu₂ and mGlu₃ were prepared by PCR amplification of the entire coding sequence of each receptor and cloning into the pIRESpuro3 vector (Invitrogen, Carlsbad, CA, USA). For mGlu₂ and mGlu₃, the cloning sites were NheI/NotI. HEK-GIRK cells, generously provided by Lily Jan (University of California San Francisco, San Francisco, CA, USA), were transfected with 24 μg of DNA using Fugene6 (Promega), stable transfectants were selected with 1000 ng/mL puromycin dihydrochloride (Sigma-Aldrich), and polyclonal human mGlu₂/GIRK and mGlu₃/GIRK cell lines were established. Cells were maintained at 37°C in the presence of 5% CO₂ following selection in a solution containing: 45% DMEM; 45% Ham’s F12; 10% FBS; 100 units/mL penicillin/streptomycin; 20 mmol/L HEPES, pH 7.3; 1 mmol/L sodium pyruvate; 2 mmol/L glutamine; 700 μg/mL G418 (Mediatech Inc., Herndon, VA, USA), and 600 μg/mL puromycin (growth media). All cell culture reagents were purchased from Invitrogen unless otherwise noted.

Human mGlu₂ and mGlu₃ thallium flux in vitro assays.

Compound activity at mGlu₂ and mGlu₃ was assessed using thallium flux through GIRK channels, a method that has been previously described in detail.^29,30 Briefly, cells were plated into 384-well, black-walled, clear-bottomed poly-D-lysine-coated plates at a density of 15,000 cells/20 μL/well in DMEM containing 10% dialyzed FBS, 20 mmol/L HEPES, and 100 units/mL penicillin/streptomycin (assay media). Plated cells were incubated overnight at 37°C in the presence of 5% CO₂. The following day, the medium of the cells was exchanged to assay buffer [Hanks’ balanced salt solution (Invitrogen) containing 20 mmol/L HEPES, pH 7.3] using an ELX405 microplate washer (BioTek), leaving 20 μL/well. This was followed by the addition of 20 μL/well FluoZin2-AM (330 nmol/L final concentration) indicator dye (Invitrogen; prepared as a stock in DMSO and mixed in a 1:1 ratio with Pluronic acid F-127) in assay buffer. Cells were incubated for 1 h at room temperature, and the dye was exchanged to assay buffer using an ELX405, leaving 20 μL/well. For concentration-response curve experiments, compounds were serially diluted 1:3 into 10-point concentration response curves and were transferred to daughter plates using an Echo acoustic plate reformatter (Labcyte, Sunnyvale, CA, USA). Tested compounds were diluted to 2 times their final desired concentration in assay buffer (0.3% DMSO final concentration). Agonists were diluted in thallium buffer [125 mmol/L sodium bicarbonate (added fresh on the morning of the experiment), 1 mmol/L magnesium sulfate, 1.8 mmol/L calcium sulfate, 5 mmol/L glucose, 12 mmol/L thallium sulfate, and 10 mmol/L HEPES, pH 7.3] at 5 times the final concentration to be assayed. Cell plates and compound plates were loaded onto a kinetic imaging plate reader (FDSS 6000 or 7000; Hamamatsu Corporation, Bridgewater, NJ, USA). Appropriate baseline readings were taken (10 images at 1 Hz; excitation, 470 ± 20 nm; emission, 540 ± 30 nm) and test compounds were added in a 20 μL volume and incubated for approximately 1 h at room temperature before the addition of 10 μL of thallium buffer with or without an EC₈₀ concentration of the agonist glutamate. After the addition of agonist, data were collected for approximately an additional 2.5 min. Data were analyzed using Excel (Microsoft Corp., Redmond, WA, USA). The slope of the fluorescence increase beginning 5 s after thallium/agonist addition and ending 15 s after thallium/agonist addition was calculated, corrected to vehicle and maximal agonist control slope values, and plotted using either XLfit (ID Business Solutions) or Prism software (GraphPad Software, San Diego, CA, USA) to generate concentration-response curves. Potencies were calculated from fits using a four-point parameter logistic equation. Each concentration-response represents duplicate IC₅₀ determinations on a single day for each receptor, which were performed on three separate days to generate N=3 datasets to report average IC₅₀s +/− SEM.

Animals.

Male Sprague-Dawley rats (8–10 weeks old) and ddY mice (8 weeks old) were purchased from Japan SLC (Shizuoka, Japan). Male Pgp/BCRP-KO (FVB.129P2-Abcb1a^tm1BorAbcb1b^tm1BorAbcg2^tm1AhsN7) and wild-type (FVB) mice were obtained from Taconic Farm (Hudson, NY, USA). All animals were kept in a temperature-controlled environment with a 12-h light/dark cycle and fed a standard diet (MB-1/Funabashi Farm, Chiba, Japan). Animal experiments were performed according to the recommendations of the Committee for the Care and Use of Laboratory Animals in our institute and the ARRIVE guidelines (http://www.nc3rs.org/ARRIVE).

In Vitro Autoradiography.

Three rats were euthanized by decapitation under anesthesia (1.5% isoflurane) and their brains were quickly removed and frozen on powdered dry ice. Sagittal brain sections (20 μm) were cut at −20°C with a cryostat (HM560; Carl Zeiss, Oberkochen, Germany) and mounted on air plasma spray-coated glass slides (Matsunami, Tokyo, Japan). Brain sections were pre-incubated with a 50 mmol/L Trizma buffer solution containing 2 mmol/L MgCl₂ and 1.2 mmol/L CaCl₂ for 20 min at room temperature. After pre-incubation, the sections were incubated with fresh buffer containing [¹¹C]5i (53 MBq/L, 0.6 nmol/L) or [¹¹C]4 (130 MBq/L, 1.0 nmol/L) for 30 min at room temperature. To determine the specificity and selectivity of radioligands for mGlu₂, vehicle (DMSO), unlabeled compound (5i or 4, 10 μmol/L), or 20 (MNI-137, 10 μmol/L) were also co-incubated, respectively. After incubation, the sections were washed (2 min, 3 times) using fresh buffer, dried in air, and then exposed to imaging plates (BAS-MS2325; Fujifilm, Tokyo, Japan) for 60 min. Slides were analyzed using a Bio-imaging Analyzer System (BAS-5000; Fujifilm). The radioactivity in each section was quantified using the MultiGauge analysis software (version 2.3; Fujifilm) and obtained in terms of photo-stimulated luminescence per unit area (photo-stimulated luminescence/mm²). Regions of interest (ROIs) in autoradiographs were manually drawn around the cortex, striatum, hippocampus, thalamus, cerebellum, and pons. The ratio of total binding to nonspecific binding was estimated from the radioactivity in each autoradiograph divided by the radioactivity in autoradiographs of slides treated by self-blocking.

Small-Animal PET Studies.

Each mouse (n = 1, in each type) or rat (n = 4, in each group) was anesthetized with 1.5% (v/v) isoflurane. Prior to securing the animal in a custom-designed chamber placed in the center of a small-animal PET scanner (Inveon; Siemens Medical Solutions, Knoxville, TN, USA) and under anesthesia, a 29-gauge needle with 12–15 cm of a polyethylene 10 system prepared in-house was inserted into the tail vein of each mouse, and a 24-gauge intravenous catheter (Terumo Medical Products, Tokyo, Japan) was inserted into the tail vein of each rat.^31,32 A bolus of either [¹¹C]5i (29–31 MBq/0.1 mL per mouse; 30–40 MBq/mL per rat) or [¹¹C]4 (17 MBq/0.1 mL per mouse) was injected via the tail vein catheter, and then dynamic emission scans in three-dimensional list mode were conducted for 60 min (1 min × 4 frames, 2 min × 8 frames, and 5 min × 8 frames for each mouse) or 30 min (10 s × 12 frames, 20 s × 3 frames, 30 s × 3 frames, 1 min × 3 frames, 2 min × 3 frames, 150 s × 3 frames, and 5 min × 3 frames for each rat). Body temperatures were maintained at 37°C using a heated water circulation system (T/Pump TP401; Gaymar Industries, Orchard Park, NY, USA) during the PET scans. For the blocking study, the rat was pre-administered with unlabeled 5i (1 mg/kg) and unlabeled 4 (1 mg/kg).

Data Analysis.

The PET dynamic image data (slice thickness, 0.6 mm) were reconstructed by filtered back-projection using a Hanning filter with a Nyquist cutoff of 0.5 cycles per pixel. Summed PET images of each mouse brain and averaged DVR-scaled PET images of each rat brain were obtained and fused with an MRI template using the PMOD software package (version 3.4; PMOD Technology, Zurich, Switzerland). The TACs of [¹¹C]5i and [¹¹C]4 in wild-type and in Pgp/BCRP-KO mice were acquired from volume of interests (VOIs) in the whole brain, with reference to the MRI template. For rats, VOIs were located in the cortex, striatum, cerebellum, and pons. To estimate DVR as a quantitative index of specific radioligand binding, the Logan Reference method³³ was performed using the TAC of the pons as the reference. The radioactivity was decay-corrected to the injection time and expressed as the SUV, normalized to the injected radioactivity and body weight. SUV was calculated according to the following formula: SUV = (radioactivity per mL tissue/injected radioactivity) × body weight (g).

Biodistribution Study in Mice.

Each mouse (37.7 ± 0.4 g) was injected with a bolus of [¹¹C]5i (2.5 MBq/0.1 mL) via the tail vein. Three mice were sacrificed at each experimental time point (1, 5, 15, 30, and 60 min) after the injection by cervical dislocation. Whole brain, heart, liver, lung, spleen, testis, kidney, pancreas, stomach (including contents), small intestines (including contents), large intestines (including contents), muscle, and blood samples were removed quickly. The radioactivity in these tissues was measured with an autogamma scintillation counter (2480 Wizard²; ParkinElmer, Walthan, MA, USA) and expressed as %ID/g. All radioactivity measurements were corrected for decay.

Metabolite Analysis of Plasma and Brain Samples.

Following the intravenous injection of [¹¹C]5i (74 MBq/0.2 mL), the rats (n = 3 per time point) were sacrificed at 5 min and at 20 min under anesthesia (2% isoflurane). Blood and brain samples were quickly removed. The blood samples were centrifuged at 15,000 g for 2 min at 4°C to separate the plasma. The plasma (0.5 mL) was collected in a test tube containing CH₃CN (0.5 mL), and the resulting mixture was vortexed for 15 s and then centrifuged at 15,000 g for 2 min for deproteinization. The resulting supernatant was collected and then the pellet was resuspended in CH₃CN (0.2 mL). Subsequently, the mixture was centrifuged at 15,000 g for 2 min. The supernatant was collected. The brain samples were homogenized using a homogenizer (Silent Crusher S; Heidolph, Schwabach, Germany) in ice-cooled saline (2.0 mL). The resulting homogenate (0.5 mL) was added into a test tube containing CH₃CN (0.5 mL), vortexed, and centrifuged at 15,000 g for 2 min at 4°C. After the supernatant (0.5 mL) was collected, the pellet was resuspended with CH₃CN (0.2 mL) and then centrifuged at 15,000 g for 2 min. The resulting supernatant was collected. Subsequently, the mixture was centrifuged at 15,000 g for 2 min. an aliquot of the supernatant (0.2–0.7 mL) obtained from the plasma or from the brain homogenate was injected into the HPLC system with a radioactivity detector and analyzed using a CAPCELL PAK C₁₈ column with a mobile phase (CH₃CN/H₂O = 50/50, v/v) at a flow rate of 1.0 mL/min. The percent ratio of the unchanged form to total radioactivity (corrected for decay) on the HPLC chromatograms was calculated as % = (peak area of [¹¹C]5i/total peak area) × 100. The extraction efficiency of [¹¹C]5i was confirmed by counting radioactivity in each pellet and supernatant after the deproteinization of plasma and brain at each time point using an autogamma scintillation counter (2480 Wizard²).

Statistics.

All data are expressed as the mean ± standard deviation (SD). The differences between each radioligand were calculated using one- or two-way repeated measures analyses of variance. Post hoc analyses employed the Bonferroni method. Statistical significance (denoted with asterisks in the figures) was determined at a 95% confidence level (p < 0.05). All statistical data were analyzed using GraphPad Prism 5.

ABBREVIATIONS USED

CNS

central nervous system

DMF

N,N-dimethylformamide

DMSO

dimethyl sulfoxide

DVR

distribution volume ratio

EOB

end of bombardment

EOS

end of synthesis

MRI

magnetic resonance imaging

mGlus

metabotropic glutamate receptors

%ID/g

percentage of the injected dose per gram of wet tissue

PET

positron emission tomography

SUV

standardized uptake value

TAC

time–activity curve

t_R

retention time