Synthesis, F-18 Radiolabeling, and MicroPET Evaluation of 3-(2,4-Dichlorophenyl)-N-alkyl-N-fluoroalkyl-2,5-dimethylpyrazolo[1,5-a]pyrimidin-7-amines as Ligands of the Corticotropin-Releasing Factor Type-1 (CRF₁) Receptor

Abstract

A series of 3-(2,4-dichlorophenyl)-N-alkyl-N-fluoroalkyl-2,5-dimethylpyrazolo[1,5-a]pyrimidin-7-amines were synthesized and evaluated as potential positron emission tomography (PET) tracers for the corticotropin-releasing factor type-1 (CRF₁) receptor. Compounds 27,28,29, and 30 all displayed high binding affinity (≤ 1.2 nM) to the CRF₁ receptor when assessed by in vitro competition binding assays at 23 °C, whereas a decrease in affinity (≥ 10-fold) was observed with compound 26. The logP_7.4 values of [¹⁸F]26 – [¹⁸F]29 were in the range of ~2.2 – 2.8 and microPET evaluation of [¹⁸F]26 – [¹⁸F]29 in an anesthetized male cynomolgus monkey demonstrated brain penetrance, but specific binding was not sufficient enough to differentiate regions of high CRF₁ receptor density from regions of low CRF₁ receptor density. Radioactivity uptake in the skull, and sphenoid bone and/or sphenoid sinus during studies with [¹⁸F]28, [¹⁸F]28-d₈, and [¹⁸F]29 was attributed to a combination of [¹⁸F]fluoride generated by metabolic defluorination of the radiotracer and binding of intact radiotracer to CRF₁ receptors expressed on mast cells in the bone marrow. Uptake of [¹⁸F]26 and [¹⁸F]27 in the skull and sphenoid region was rapid but then steadily washed out which suggests that this behavior was the result of binding to CRF₁ receptors expressed on mast cells in the bone marrow with no contribution from [¹⁸F]fluoride.

1. Introduction

Corticotropin-releasing factor (CRF) is a 41-residue peptide⁵ that is the main regulator of the hypothalamic-pituitary-adrenal axis through its actions on CRF type-1 (CRF₁) receptors^{1, 6, 7} expressed in the pituitary⁸ which, in turn, signal for the release of adrenocorticotropic hormone.^9-13 Central nervous system (CNS) CRF has been linked to a variety of disorders including depression, stress, anxiety, post-traumatic stress disorder, and addiction.^14-24 CRF has been shown to be involved in the stress-induced phosphorylation of tau which implies a potential link between stress and Alzheimer's disease pathology.^{25, 26} It has also been suggested that CRF can play a neuroprotective role.^{25, 27-31} CRF is also found in the periphery where it is involved in inflammation^32-38 and cancer,^39-51 and it has been suggested that CRF may be one of the links between stress and cancer.^{48, 52-59}

Pharmacological manipulation of CNS CRF₁ receptors has been pursued as a means to treat stress, anxiety, and depression,^60-65 and this has resulted in the synthesis of numerous small-molecule CRF₁ receptor antagonists as potential therapeutics.^66-71 The development of therapeutic CRF₁ receptor antagonists can be aided by the availability of radiolabeled CRF₁ receptor antagonists that allow for imaging with positron emission tomography (PET) or single-photon emission computed tomography (SPECT). PET or SPECT imaging enables the measurement of receptor density thereby allowing for a determination of the levels of receptor expression in normal and altered psychiatric states, and how this changes during pharmacologic treatment.^72-74 Furthermore, utilization of PET or SPECT imaging in conjunction with a therapeutic allows for a determination of receptor occupancy, dose finding, and mechanism of action studies of the therapeutic.^75-79 Thus, there has been great interest in the development of radiolabeled CRF₁ receptor antagonists, and several examples (1 – 12)^80-89 are shown in Figure 1. Compounds [¹⁸F]1 and [¹²³I]2 had limited aqueous solubility (20% EtOH in saline was required to dissolve the compounds),⁸⁰ and biological evaluation in rats showed very low brain uptake indicating that the compounds did not diffuse through the blood-brain barrier (BBB).^90-93 In vivo biodistribution studies with [⁷⁶Br]3 in rats demonstrated brain penetrance, and in vitro autoradiography studies demonstrated specific binding.⁸² Compounds [¹¹C]6, [¹¹C]7, and [¹¹C]8 all penetrated the BBB of a baboon but did not display specific binding and were rapidly metabolized.^{84, 85} Compounds [¹⁸F]9 – [¹⁸F]11 and [¹¹C]12 have recently been reported, and these compounds were able to penetrate the BBB of a monkey but they did not display specific binding.^{87, 88} Thus, attempts to develop a viable CRF₁ receptor radiotracer have been hampered by low brain entry, little or no in vivo specific binding when brain entry is achieved, and rapid metabolism. A successful CRF₁ receptor radiotracer will, therefore, require improvements in binding affinity that will enable specific binding to be detected in vivo, a reduction in lipophilicity to increase brain penetrance and reduce non-specific binding, and increased metabolic stability. As part of an effort to develop a viable CRF₁ receptor PET tracer we have been investigating the effects of various N-fluoroalkyl groups (Scheme 1) on the binding affinity, lipophilicity, and PET imaging properties of 3-(2,4-dichlorophenyl)-N-alkyl-N-fluoroalkyl-2,5-dimethylpyrazolo[1,5-a]pyrimidin-7-amines.⁹⁴ This structural class^{94, 95} was chosen because the 2,4-dichloro-substituents would be resistant to metabolic transformation (unlike methyl-, alkyl ether-, and alkyl amino-substituents) and would be less lipophilic than bromo- or iodo-substituents. Furthermore, as shown in Scheme 1, the core molecule can be assembled and then a library of compounds can be prepared through a divergent synthetic approach.

Figure 1.

Examples of previously reported CRF₁ receptor radiotracers.

Scheme 1.

2. Results and Discussion

2.1. Chemistry

Amino-pyrazole 13^{89, 95, 96} (Scheme 1) was refluxed in ethyl acetoacetate (an alternative to using ethyl acetoacetate in toluene with acid catalyst⁹⁴ or in refluxing AcOH⁹⁶) to give 14 which was then reacted with POCl₃ to give 15. Compound 15 was reacted with primary alkylammonium chlorides or primary amines to give compounds 16 - 20, or with dialkylamines to give compounds 21 -25. Alcohols 24 and 25 were deprotonated with NaH in DMA and then reacted with 1-tosyloxy-2-fluoroethane to give compounds 26 and 27. Compounds 16 - 18 were deprotonated with NaH in DMA and then reacted with 1-bromo-4-fluorobutane to give compounds 28 - 30, or reacted with alkyl-ditosylates to give the radiolabeling precursors 31 - 34. Attempts to N-alkylate 19 with 1-tosyloxy-3-fluoropropane, and 20 with 1-tosyloxy-2-fluoroethane or 1,2-ditosyloxyethane did not give the desired products, presumably due to the formation of a cyclic-quaternary ammonium salt.⁹⁷ 1,4-Ditosyloxybutane (35) and 1,4-ditosyloxybutane-d₈ (35-d₈) were prepared by refluxing 1,4-dibromobutane or 1,4-dibromobutane-d₈, respectively, with AgOTs in CH₃CN (Scheme 2).^{98, 99}

Scheme 2.

2.2. In Vitro Competition Binding Assays

The binding affinities of compounds 21 – 30 at the CRF₁ receptor were determined using in vitro competition binding assays at 23 °C with transfected human CRF₁ receptors in HEK293T cells (Table 1). Compounds 21 and 22 have equal affinities indicating that the CRF₁ receptor does not discriminate between symmetrical and unsymmetrical N,N-dialkyl groups with an equal number of CH₂ groups (in this limited series of compounds). The binding affinity that we determined for 22 (K_i = 0.46 ± 0.06 nM) was similar to the previously reported value of K_i = 0.9 ± 0.1 nM.⁹⁶ Addition of a terminal fluorine atom to the butyl group of 22 to give 29 did not change the binding affinity, and replacement of the N-ethyl group of 29 with N-fluoroethyl to give 30 had a negligible effect on the binding affinity. Replacement of the ethyl group of 29 with a methyl group to give 28, in an attempt to reduce lipophilicity, resulted in a ~2.3-fold loss of affinity. The ethers 26 and 27 were also prepared in an effort to reduce lipophilicity. Insertion of an oxygen atom into the butyl group of 29 to give 27 resulted in a ~2.5-fold loss of affinity whereas a nearly 12-fold loss of affinity occurred upon changing 28 to 26.

Table 1.

Results of In Vitro Competition Binding Assays at 23 °C Using Transfected Human CRF₁ Receptors in HEK293T Cells.

compd	R1	R2	Ki (nM) ± SEMa)	n =
21	Pr	Pr	0.47 ± 0.08	3
22	Et	Bu	0.46 ± 0.06	3
26	Me	EtOEtF	12.4 ± 1.06	3
27	Et	EtOEtF	1.17 ± 0.07	3
28	Me	BuF	1.07 ± 0.21	4
29	Et	BuF	0.47 ± 0.10	5
30	FEt	BuF	0.59 ± 0.04	3

^a)vs. [¹²⁵I]-Tyr⁰-sauvagine.¹⁰⁰

2.3. Radiochemistry

Radiolabeling of [¹⁸F]28 was initially performed by a two-step method (Scheme 3) where 16 was deprotonated with NaH in anhydrous DMA, and then reacted with [¹⁸F]fluorobutyltosylate which was prepared from 35 in a Siemens/CTI chemical processing control unit (CPCU). Subsequently, a one-step method was developed which utilized the alkyltosylate precursors 31 – 34 (Scheme 4) in the CPCU. This shortened the radiosynthesis time by ~35 minutes and simplified the radiolabeling procedure because it moved the water-sensitive NaH deprotonation to the precursor synthesis step (Scheme 1). This also increased the radiochemical yield for [¹⁸F]28 (11% for the two-step method vs. 44% for the one-step method, see Experimental Sections 4.29 and 4.30). The crude radiotracers were purified by semi-preparative HPLC (we initially used MeOH in the solvent mixture but later replaced it with EtOH which successfully reduced radiolysis of the radiotracer^101-103 during purification and isolation), collected on a Waters tC₁₈ Sep-Pak, eluted from the Sep-Pak with EtOH, and collected in a sealed sterile dose vial as a 10%EtOH/saline solution. Radiochemical yields (decay-corrected) were then calculated at this point (see Experimental Sections 4.27 – 4.32). The solutions were then passed through successive 1 μm and 0.2 μm Acrodisc PTFE filters (pre-rinsed with EtOH) and collected in a sealed sterile dose vial as the formulated dose. The octanol/aqueous buffer partition coefficients (logP_7.4) of [¹⁸F]26 - [¹⁸F]29 (Table 2) were measured by the shake-flask method as previously reported by Wilson and Houle,^{104, 105} and were found to be between logP_7.4 = ~2.2 – 2.8 which is in the appropriate range for brain entry.^{90, 91, 93}

Scheme 3.

Scheme 4.

Table 2.

LogP_7.4 values.

compd	logP7.4 ± SD	n =
[18F]26	2.75 ± 0.02a	4
[18F]27	2.80 ± 0.01a	4
[18F]28	2.20 ± 0.02b	4
[18F]28	2.44 ± 0.09a	3
[18F]28-d8	2.49 ± 0.01a	4
[18F]29	2.81 ± 0.01a	4

^a1-Step radiolabeling method.

^b2-Step radiolabeling method.

2.4. MicroPET Imaging

Compound [¹⁸F]28 (prepared by the two-step radiolabeling method) was evaluated by microPET for brain uptake in an anesthetized male Sprague-Dawley rat. As shown in Figure S1 (Supplementary Material), brain penetrance was achieved but high uptake was also observed in the liver and Harderian glands.¹⁰⁶ The time-activity curves (TACs) in Figure 2 show that rapid brain uptake occurred (SUV^2-4 = 2 at 5 min) followed by a slow washout. Liver uptake was very high and rapid (SUV = 5.4 at 3 min) followed by a steady washout, whereas uptake in the Harderian glands continued throughout the course of the study. The lack of bone visualization suggested that [¹⁸F]28 was stable to defluorination in the rat. Based on this preliminary data, compound [¹⁸F]28 (prepared by the one-step radiolabeling method) was evaluated by microPET for brain uptake and specific binding in an anesthetized male cynomolgus monkey. The microPET images in Figure S2 (Supplementary Material) show that brain penetrance was achieved but the TACs in Figure 3 show that uptake of [¹⁸F]28 was equal across all brain regions including regions with high CRF₁ receptor density (cerebellum and cortex)^{85, 88, 107, 108} and regions of low CRF₁ receptor density (striatum and thalamus), thus suggesting little or no specific binding. This uniform brain distribution is similar to what was reported for [¹¹C]6, [¹¹C]7, [¹¹C]8, [¹⁸F]10, [¹⁸F]11, and [¹¹C]12.^{84, 85, 87} Furthermore, high radioactivity uptake in the skull, and sphenoid bone and/or sphenoid sinus, suggests defluorination. The deuterated version, [¹⁸F]28-d₈, was therefore prepared in an effort to reduce defluorination.^{109, 110} As shown by the TACs in Figure 4, uptake of [¹⁸F]28-d₈ in the skull and sphenoid region was reduced somewhat relative to [¹⁸F]28 but defluorination was not completely prevented. Skull uptake of [¹⁸F]28 and [¹⁸F]28-d₈ was rapid but then leveled off for [¹⁸F]28-d₈ and only slightly increased with time for [¹⁸F]28. Radioactivity uptake in the sphenoid region was also rapid for both [¹⁸F]28 and [¹⁸F]28-d₈, but then there was a brief washout followed by a slow increase. This seems to suggest that radioactivity uptake in the sphenoid region is not solely the result of metabolic defluorination to generate [¹⁸F]fluoride and, in fact, compounds [¹¹C]6, [¹¹C]7, and [¹¹C]8, which cannot generate [¹⁸F]fluoride, also show uptake in the same region.⁸⁵ Thus, some of the observed sphenoid uptake of [¹⁸F]28 and [¹⁸F]28-d₈ may be the result of binding in bone marrow. Mast cells, which are derived from bone marrow^{111, 112} and are also located intracranially,^113-115 have been shown to express CRF₁ receptors,^{43, 112, 116-124} and this would help to explain the uptake of [¹⁸F]28 and [¹⁸F]28-d₈ (and [¹¹C]6, [¹¹C]7, and [¹¹C]8) in the sphenoid region. Mast cells present in the skin also express CRF₁ receptors^125-127 which explains the radioactivity uptake observed in the skin in the coronal and sagittal images in Figures S2 and S3 (Supplementary Material). The pituitary, which is located in the hypophyseal fossa of the sphenoid bone, also contains CRF₁ receptors⁸ and the pituitary can be observed in the sagittal image in Figure S2 (Supplementary Material) but radioactivity spillover from the sphenoid region prevented an accurate generation of TACs for the pituitary. The brain uptake and washout of [¹⁸F]28-d₈ (Figure 5) was very similar to that of [¹⁸F]28 (Figure 3) which indicates that the deuterated N-fluorobutyl group did not alter the in vivo behavior of the radiotracer other than what is shown in Figure 4.

Figure 2.

MicroPET TACs of [¹⁸F]28 in the brain, liver, and Harderian glands of an anesthetized male Sprague-Dawley rat.

Figure 3.

MicroPET TACs of [¹⁸F]28 in the brain of an anesthetized male cynomolgus monkey.

Figure 4.

MicroPET TACs comparing the uptake of [¹⁸F]28 and [¹⁸F]28-d₈ in the skull, and sphenoid bone and/or sphenoid sinus of an anesthetized male cynomolgus monkey.

Figure 5.

MicroPET TACs of [¹⁸F]28-d₈ in the brain of an anesthetized male cynomolgus monkey.

The distribution of [¹⁸F]29 in the brain (Figure S4, Supplementary Material)) was similar to that of [¹⁸F]28 and [¹⁸F]28-d₈, but the overall amount of uptake was less (Figure 6) while the retention was slightly longer and the washout was slightly slower which may be the result of the greater lipophilicity of [¹⁸F]29 relative to [¹⁸F]28, or the ~2-fold stronger binding affinity of [¹⁸F]29 relative to [¹⁸F]28, or a combination of both factors. The skull uptake of [¹⁸F]29 (Figure 7) was less than that of [¹⁸F]28 and [¹⁸F]28-d₈ (Figure 4) which suggests that an N-ethyl group may be able to reduce metabolic defluorination through increased sterics relative to an N-methyl group. The uptake of [¹⁸F]29 in the sphenoid region (Figure 7) behaved similarly to [¹⁸F]28 and [¹⁸F]28-d₈ (Figure 4) with an initial rapid uptake followed by a quick partial washout and then a steady increase. As described above, the radioactivity uptake in the sphenoid region may initially be rapid uptake of [¹⁸F]29 by mast cells in the bone marrow followed by a brief washout and then a steady uptake of [¹⁸F]fluoride generated by metabolic defluorination. But, the continuous metabolic generation of [¹⁸F]fluoride would also be expected to be reflected by a continuous increase of radioactivity in bone¹⁰⁹ although that is not what is observed in the skull in Figure 7 with [¹⁸F]29 or in Figure 4 with [¹⁸F]28-d₈. Furthermore, as described above, uptake of [¹⁸F]fluoride cannot account for the observed uptake of [¹¹C]6, [¹¹C]7, and [¹¹C]8 in the sphenoid region.⁸⁵ The increase of radioactivity in the sphenoid region, after the initial brief washout, (Figures 4 and 7) corresponds to the beginning of radiotracer washout from the brain (Figures 3, 5, and 6) which suggests that the continuous radioactivity uptake (from ~20 – 115 min) in the sphenoid region is at least partly the result of intact radiotracer exiting the brain and is not solely the result of metabolically generated [¹⁸F]fluoride.

Figure 6.

MicroPET TACs of [¹⁸F]29 in the brain of an anesthetized male cynomolgus monkey.

Figure 7.

MicroPET TACs of [¹⁸F]29 in the skull, and sphenoid bone and/or sphenoid sinus of an anesthetized male cynomolgus monkey.

The distribution of [¹⁸F]26 and [¹⁸F]27 in the brain (Figures S5 and S6, respectively, Supplementary Material) was similar to that of [¹⁸F]28, [¹⁸F]28-d₈, and [¹⁸F]29. The TACs of [¹⁸F]27 in the brain (Figure 8) were also similar to those of [¹⁸F]28, [¹⁸F]28-d₈ , and [¹⁸F]29 with peak brain uptake achieved between ~5 – 15 min followed by a steady washout and with no discernable difference between regions with high CRF₁ receptor density (cerebellum and cortex)^{85, 88, 107, 108} and regions of low CRF₁ receptor density (striatum and thalamus) which suggests little or no specific binding. The TACs of [¹⁸F]26 in the brain (Figure 9) show that it entered the brain more rapidly and washed out more rapidly than [¹⁸F]27, [¹⁸F]28, [¹⁸F]28-d₈ , and [¹⁸F]29. This is further demonstrated by Figure S7 (Supplementary Material) which compares the whole-brain TACs (a single ROI was generated for the whole brain) for each of the radiotracers and which confirms the more rapid kinetics of [¹⁸F]26. Compound 26 has ~10 – 12-fold lower binding affinity than 27 or 28 and ~26-fold lower binding affinity than 29 (Table 1). Thus, the rapid washout of [¹⁸F]26 from the brain suggests that the higher binding affinity compounds, [¹⁸F]27, [¹⁸F]28, [¹⁸F]28-d₈ , and [¹⁸F]29, may actually be displaying a small amount of specific binding to the CRF₁ receptor which slows the washout from the brain. The rate of washout from the brain cannot be attributed to lipophilicity-induced non-specific binding because [¹⁸F]27 and [¹⁸F]29 have the same logP_7.4 value (Table 2) and are only slightly more lipophilic than [¹⁸ F]26, yet each compound has a different shaped whole-brain TAC in Figure S7. Therefore, it may be possible that [¹⁸F]27, [¹⁸F]28, [¹⁸F]28-d₈, and [¹⁸F]29 are binding to the CRF₁ receptor but that dissociation from the receptor is too rapid^{128, 129} for a difference between regions with high CRF₁ receptor density (cerebellum and cortex) and regions of low CRF₁ receptor density (striatum and thalamus) to be detected.

Figure 8.

MicroPET TACs of [¹⁸F]27 in the brain of an anesthetized male cynomolgus monkey.

Figure 9.

MicroPET TACs of [¹⁸F]26 in the brain of an anesthetized male cynomolgus monkey.

Figures 10 and 11 show the TACs for [¹⁸F]26 and [¹⁸F]27, respectively, in the skull and sphenoid region. Both [¹⁸F]26 and [¹⁸F]27 show rapid uptake and then a steady washout from these regions but do not show the steady continuous uptake that was observed with [¹⁸F]28, [¹⁸F]28-d₈ , and [¹⁸F]29 which supports the notion that some of the uptake of [¹⁸F]28, [¹⁸F]28-d₈, and [¹⁸F]29 in the sphenoid region, and possibly the skull, is due to binding to CRF₁ receptors expressed on mast cells in the bone marrow^{43, 112, 116} and not solely the result of metabolically generated [¹⁸F]fluoride. Furthermore, the lower binding affinity compound, [¹⁸F]26, washes out faster from the skull and sphenoid region than the higher binding affinity compound, [¹⁸F]27, which is in agreement with the idea that binding to the CRF₁ receptor expressed on bone marrow mast cells is being observed. Also, the rate of washout of [¹⁸F]27 from the sphenoid region changes significantly at ~20 min which corresponds to the beginning of washout of [¹⁸F]27 from the brain, while the change in washout of [¹⁸F]26 from the sphenoid region is less pronounced at ~20 min, which is also in agreement with the higher binding affinity compound, [¹⁸F]27, binding to the CRF₁ receptor expressed on mast cells in bone marrow.

Figure 10.

MicroPET TACs of [¹⁸F]26 in the skull, and sphenoid bone and/or sphenoid sinus of an anesthetized male cynomolgus monkey.

Figure 11.

MicroPET TACs of [¹⁸F]27 in the skull, and sphenoid bone and/or sphenoid sinus of an anesthetized male cynomolgus monkey.

The behavior of [¹⁸F]27, [¹⁸F]28, [¹⁸F]28-d₈, and [¹⁸F]29 in the brain of a cynomolgus monkey is similar to what was reported for [¹¹C]7 and [¹¹C]8 in the brain of a baboon,⁸⁵ and [¹⁸F]10 and [¹¹C]12 in the brain of a rhesus monkey⁸⁸ with regards to the shapes of the TACs, relatively even distribution throughout the brain, and an apparent lack of specific binding. All of these compounds displayed high binding affinities to the CRF₁ receptor when evaluated in in vitro binding assays and compounds [¹⁸F]10 and [¹⁸F]11 displayed specific binding in autoradiography studies with rhesus brain slices.⁸⁸ Furthermore, 8 blocked the binding of [¹²⁵I]sauvagine, [¹²⁵I]oCRF, and [³H]6 in rat autoradiography studies,^{130, 131} and numerous other studies have used autoradiography to confirm the presence of the CRF₁ receptor and to quantify its density.^{85, 88, 107, 108, 132-134} Thus, it is not clear why specific binding is not being observed during PET studies. It is possible that endogenous CRF is competing with the PET tracers for binding at the CRF₁ receptor during the PET studies or that there is interference from the isoflurane anesthesia which has already been proven to alter the results of some PET studies.^135-139 A speculative mechanism of how isoflurane, mediated through CNS mast cells, increases the concentration of endogenous CRF in the brain and which then competes with the PET tracer, is as follows: Mast cells¹¹¹ are located peripherally but also intracranially in the dura¹¹³ as well as in the brain, with high levels in the thalamus, hypothalamus, and grey matter, including the cortex;^{124, 140} they can also move from the blood to the brain,¹¹⁵ and they interact with CRF to regulate the permeability of the BBB.^{114, 141} They are involved in the allergic response, including the allergic response to anesthesia,¹⁴² and they can also be activated by non-allergic triggers,¹⁴³ including CRF.^{113, 118, 121, 122, 125, 126} Mast cells express CRF₁ receptors^{43, 112, 116, 118, 119} and also synthesize and secrete CRF,¹¹⁷ and thus the two are closely linked. Activation of mast cells results in the release of histamine and other chemical messengers,¹¹¹ and histamine has been shown to increase levels of CRF in rats.^{144, 145} This effect could be further enhanced by histamine receptor agonists or blocked by histamine receptor antagonists or anti-CRF.^{144, 146, 147} In dog brain, mast cell degranulation results in an increase in histamine release and a subsequent increase in plasma cortisol and ACTH.^{120, 148} This effect can be blocked by intracerebroventricular administration of anti-CRF or a histamine receptor antagonist, thus confirming the link between histamine and CRF release in the brain. In rats, isoflurane anesthesia has been shown to increase the concentration of histamine in the hypothalamus, and this was attributed to inhibition of histamine metabolism.¹⁴⁹ Furthermore, ketamine, an anesthetic commonly used to initially anesthetize non-human primates prior to using isoflurane during PET imaging, has been shown to stimulate the release of histamine from lung mast cells.¹⁴² Thus, we speculate that anesthesia could be increasing the histamine concentration in the brain through mast cell activation, and then the histamine causes an increase in CRF concentration. The increased CRF concentration would then compete directly with the PET tracer for binding at the CRF₁ receptor, or cause a downregulation or internalization of the receptor, effectively reducing the number of available CRF₁ receptors for PET tracer binding. Because high concentrations of CNS mast cells correspond to the same locations as high concentrations of CRF₁ receptor, a reduction in the number of available CRF₁ receptors caused by the anesthesia, and mediated through mast cells, would make PET tracer binding appear similar to brain regions with low mast cell and CRF₁ receptor density, and would explain the discrepancies between the PET data and the autoradiography data. This speculative mechanism could be tested by using an awake non-human primate, but restraint stress in rats has been shown to increase histamine levels and blood corticosterone.¹⁴⁷ Thus, histamine receptor blockade would still be necessary,^{144, 146, 147} but this histamine receptor blockade could potentially be employed in the presence of anesthesia,¹⁵⁰ thus eliminating the need for an awake non-human primate.

3. Summary

This work focused on the synthesis, binding affinity determination, F-18 radiolabeling, and microPET evaluation of a series of N-fluoroalkyl-pyrazolo[1,5-a]pyrimidin-7-amines as ligands of the CRF₁ receptor. Compounds 27, 28, 29, and 30 all displayed high binding affinity (≤ 1.2 nM) to the CRF₁ receptor when assessed by in vitro competition binding assays at 23 °C, whereas a decrease in affinity (≥ 10-fold) was observed with compound 26. A one-step F-18 radiolabeling method was developed which took less time and afforded higher radiochemical yields than the initial two-step method that was employed for [¹⁸F]28. The logP_7.4 values of [¹⁸F]26 – [¹⁸F]29 were in the range of ~2.2 – 2.8 which is appropriate for passive diffusion across the BBB. MicroPET evaluation of [¹⁸F]26 – [¹⁸F]29 in an anesthetized male cynomolgus monkey demonstrated that these compounds all penetrated the brain but did not display specific binding sufficient enough to differentiate regions of high CRF₁ receptor density from regions of low CRF₁ receptor density. This is similar to what has been reported for [¹¹C]6, [¹¹C]7, [¹¹C]8, [¹⁸F]10, [¹⁸F]11, and [¹¹C]12,^{84, 85, 87} and may be due to interference from anesthesia.^135-139 Radioactivity uptake in the skull and sphenoid region during studies with [¹⁸F]28, [¹⁸F]28-d₈, and [¹⁸F]29 was attributed to a combination of [¹⁸F]fluoride generated by metabolic defluorination of the radiotracer and binding of intact radiotracer to CRF₁ receptors expressed on mast cells in the bone marrow. Uptake of [¹⁸F]26 and [¹⁸F]27 in the skull and sphenoid region was rapid but then steadily washed out which suggests that this behavior was the result of binding to CRF₁ receptors expressed on mast cells in the bone marrow with no contribution from [¹⁸F]fluoride. This indicates that the [¹⁸F]fluoroethoxy ethyl group is more stable to defluorination than the [¹⁸F]fluorobutyl group and is a suitable substitute for the [¹⁸F]fluorobutyl group.

4. Experimental Section

4.1. General

NMR spectra were obtained on a Varian Mercury spectrometer at the specified frequencies. ¹H chemical shifts are referenced to internal TMS or residual CHCl₃ (7.26 ppm), and ¹³C chemical shifts are referenced to CDCl₃ (77.23 ppm). For ¹H NMR spectroscopy ~4 – 6 mg of sample was dissolved in ~1 mL CDCl₃, whereas for ¹³C NMR spectroscopy ~30 – 40 mg of sample was dissolved in ~1 mL CDCl₃. Solvents were from EMD and were used as received. 4-Ethylamino-1-butanol was purchased from TCI. Ethyl ammonium chloride was purchased from Alfa Aesar. 1,4-Dibromobutane-d₈ was purchased from CDN Isotopes, Inc. Diethyleneglycol p-toluenesulfonate, 1,4-dibromobutane, 1-bromo-4-fluorobutane, primary amines, dialkylamines, N,N-diisopropylethylamine, 2-fluoroethanol, 3-fluoropropanol, tosyl chloride, NaH (60% dispersion in mineral oil), and anhydrous DMA (100-mL septum-capped bottle) were purchased from Aldrich. 2-Fluoroethyl tosylate and 3-fluoropropyl tosylate were prepared from the corresponding alcohols. NaH was not rinsed with hexane prior to use. Dry silica gel purifications were performed by placing silica gel (Whatman Purasil 60Å, 230-400 mesh) in a medium-fritted filter tube (~31 cm total height, 24/40 ground glass joints, 16 cm h (frit to joint bottom) × 4 cm i.d.) or in a 60-mL medium-fritted filter funnel (43 mm h × 43 mm i.d.) and eluting under vacuum using a vacuum-takeoff adapter (24/40) and 125-mL flat-bottomed boiling flasks (24/40) for fraction collection. The same method was employed for vacuum flash chromatography except that the silica was pretreated with hexane (100 mL). Radial chromatography was performed with a Harrison Research Chromatotron using Analtech rotors (Silica Gel GF). Preparative-TLC was performed on Analtech Uniplate Silica Gel GF 20 × 20 cm, 2000 micron plates (catalog #02015). HRMS was performed by the Emory University Mass Spectrometry Center. Purity of target compounds was determined by elemental analysis (www.atlanticmicrolab.com). HPLC was performed with Waters XTerra (Prep-RP₁₈, 5 μm, 19 × 100 mm) and Waters NovaPak (C₁₈, 4 μm, 3.9 × 150 mm) columns. Radiochemistry development, logP determination, and rodent microPET imaging were performed at the Emory University Center for Systems Imaging (CSI) which houses a Siemens RDS 111 cyclotron and a Siemens Inveon MicroPET/CT. Non-human primate microPET imaging was performed at Yerkes National Primate Research Center which houses a PETNET facility (the source of H¹⁸F_(aq)) and a Siemens MicroPET Focus 220. One-step ¹⁸F-radiolabelings were performed in a Siemens/CTI Chemical Processing Control Unit (CPCU).

4.2. 3-(2,4-dichlorophenyl)-2,5-dimethylpyrazolo[1,5-a]pyrimidin-7-ol (14)

Compound 13 (3.90 g, 16.11 mmol) and ethyl acetoacetate (100 mL) were stirred at reflux under Ar_(g) for 20 h, then cooled to 0 °C, filtered, rinsed with cold EtOAc (25 mL × 2), hexane (25 mL), and dried under vacuum to afford 14 (3.51 g, 71%) as an off-white powder (insoluble in CHCl₃, CH₃OH, DMSO, acetone, H₂O).

4.3. 7-chloro-3-(2,4-dichlorophenyl)-2,5-dimethylpyrazolo[1,5-a]pyrimidine (15)

Compound 14 (1.35 g, 4.38 mmol) was suspended in 1,4-dioxane (30 mL) followed by addition of POCl₃ (2.1 mL, 22.5 mmol, 5.1 equiv.). The reaction mixture was stirred at reflux under Ar_(g) for 3 h, then cooled, and the solvent was removed azeotropically with heptane to give an orange/red syrup that was dissolved in CH₂Cl₂ (50 mL) and poured into ice-water (100 mL). The aqueous layer was basified to pH 8-9 with conc. NH₄OH_(aq), the layers were separated, and the aqueous layer was extracted with CH₂Cl₂ (20 mL × 2). The combined CH₂Cl₂ layers were dried over MgSO₄ and the solvent was removed to give an orange/red syrup that was redissolved in CH₂Cl₂ (~10 mL), poured onto dry silica (14 cm h × 4 cm i.d.), and eluted under vacuum: hexane (100 mL), %CH₂Cl₂/hexane – 25% (100 mL), 50% (100 mL), 75% (100 mL), CH₂Cl₂ (400 mL) to afford 15 (1.22 g, 85%) as a white solid: ¹H NMR (300 MHz, CDCl₃) δ 7.55 (s, 1 H), 7.34 (apparent s – overlapping resonances, 2 H), 6.83 (s, 1 H), 2.56 (s, 3 H), 2.45 (s, 3 H); ¹³C NMR (75 MHz, CDCl₃) δ 159.15, 154.54, 147.56, 137.77, 135.82, 134.47, 133.96, 129.93, 129.42, 127.36, 108.81, 107.66, 24.93, 13.93.

4.4. 3-(2,4-dichlorophenyl)-N,2,5-trimethylpyrazolo[1,5-a]pyrimidin-7-amine (16)

Compound 15 (293 mg, 8.97 × 10⁻ mol), MeNH₂•HCl (253 mg, 3.75 mmol, 4.2 equiv.), iPr₂NEt (2 mL, 11.5 mmol, 12.8 equiv.), and CH₃CN (20 mL) were stirred at reflux under Ar_(g) for 150 min. The solvent was removed to give an oil and then hexane was added and removed to give an off-white solid that was dried under vacuum. The solid was dissolved in CH₂Cl₂, poured onto dry silica (15 cm h × 4 cm i.d.), and eluted under vacuum: hexanes/EtOAc/NEt₃ v/v/v 95:4:1 (100 mL), 90:8:2 (100 mL), 75:20:5 (400 mL), 50:45:5 (100 mL), 20:75:5 (100 mL) to afford a white foam (295 mg). Purification by radial chromatography (4 mm silica): hexanes/EtOAc/NEt₃ v/v/v 98:1:1 (100 mL), 95:4:1 (100 mL), 90:8:2 (200 mL), 75:20:5 (125 mL) afforded 16 (285 mg, 99%) as a white foam: TLC R_f = 0.23 (silica, 75:20:5 v/v/v hexanes/EtOAc/NEt ); ¹H NMR (300 MHz, CDCl₃) δ 7.51 (d, 1 H, J = 2.1 Hz), 7.37 (d, 1 H, J = 8.1 Hz), 7.30 (dd, 1 H, J = 2.1 Hz, J = 8.1 Hz), 6.19 (br d (unresolved q), 1 H, J = 5.1 Hz), 5.81 (s, 1 H), 3.10 (d, 3 H, J = 5.1 Hz), 2.48 (s, 3 H), 2.35 (s, 3 H); ¹³C NMR (75 MHz, CDCl₃) δ 160.22, 152.27, 146.95, 146.83, 135.77, 134.26, 133.62, 130.57, 129.74, 127.21, 104.96, 85.47, 28.60, 25.55, 13.86; HRMS (APCI) [MH]⁺ Calcd for C₁₅H₁₅Cl₂N₄: 321.0668, found: 321.0670; HPLC: Waters XTerra t_R = 8.9 min (70:30:0.1 v/v/v MeOH/H₂O/NEt₃, 9 mL/min), t_R = 21.4 min (60:40:0.1 v/v/v MeOH/H₂O/NEt₃, 9 mL/min).

4.5. 3-(2,4-dichlorophenyl)-N-ethyl-2,5-dimethylpyrazolo[1,5-a]pyrimidin-7-amine (17)

Compound 15 (415 mg, 1.27 mmol), EtNH₃Cl (430 mg, 5.27 mmol, 4.1 equiv.), iPr₂NEt (3 mL, 17.2 mmol, 13.5 equiv.) and CH₃CN (25 mL) were stirred at reflux under Ar_(g) for 140 min, cooled, and the solvent was removed to give a faint brown syrup that was dried under vacuum to give a tan solid. The solid was dissolved in CH₂Cl₂, poured onto dry silica (14 cm h × 4 cm i.d.), and eluted under vacuum: hexane/EtOAc/NEt₃ v/v/v 95:4:1 (100 mL), 90:8:2 (100 mL), 75:20:5 (400 mL) to give an off-white foam (451 mg). Purification by radial chromatography (4 mm silica): hexane/EtOAc/NEt₃ v/v/v 98:1:1 (100 mL), 95:4:1 (200 mL), 90:8:2 (200 mL), 75:20:5 (125 mL) afforded a white foam (407 mg) that was further purified by radial chromatography (4 mm silica): CHCl₃ (200 mL) to give 17 (396 mg, 93%) as a white foam: ¹H NMR (300 MHz, CDCl₃) δ 7.51 (d, 1 H, J = 2.1 Hz), 7.37 (d, 1 H, J = 8.1 Hz), 7.30 (dd, 1 H, J = 2.1 Hz, J = 8.1 Hz), 6.13 (t, 1 H, J = 5.7 Hz), 5.82 (s, 1 H), 3.44 (dq, 2 H, J = 5.7 Hz, J = 7.2 Hz), 2.47 (s, 3 H), 2.35 (s, 3H), 1.42 (t, 3 H, J = 7.2 Hz); ¹³C NMR (75 MHz, CDCl₃) δ 160.14, 152.22, 147.01, 145.86, 135.77, 134.26, 133.60, 130.59, 129.74, 127.21, 104.96, 85.63, 37.02, 25.56, 14.60, 13.85; HRMS (APCI) [MH]⁺ Calcd for C₁₆H₁₇Cl₂N₄: 335.0825, found: 335.0826; HPLC: Waters Xterra t_R = 11.1 min (70:30:0.1 v/v/v MeOH/H₂O/NEt₃, 9 mL/min).

4.6. 3-(2,4-dichlorophenyl)-N-(2-fluoroethyl)-2,5-dimethylpyrazolo[1,5-a]pyrimidin-7-amine (18)

Compound 15 (271 mg, 8.30 × 10⁻⁴ mol), 2-fluoroethylamine•HCl (357 mg, 3.59 mmol, 4.3 equiv.), iPr₂NEt (1.3 mL, 7.5 mmol, 9.0 equiv), and CH₃CN (25 mL) were stirred at reflux under Ar_(g) for 7 h, and then the solvent was removed to give a light yellow oil. Hexane was added and removed to give a syrup that was dried under vacuum to give a light yellow solid. The solid was dissolved in CH₂Cl₂, poured onto dry silica (15 cm h × 4 cm i.d.), and eluted under vacuum: hexane (50 mL), %CH₂Cl₂/hexane – 25% (100 mL), 50% (100 mL), 75% (100 mL), CH₂Cl₂ (800 mL), hexane/EtOAc/NEt₃ v/v/v 75:20:5 (200 mL), 50:45:5 (200 mL), 20:75:5 (100 mL) to give 15 (24 mg, 9% recovery) and 18 as an off-white solid (284 mg). Purification by radial chromatography (2 mm silica): hexane/EtOAc/NEt₃ v/v/v 98:1:1 (100 mL), 95:4:1 (100 mL), 90:8:2 (200 mL), 75:20:5 (100 mL), 50:45:5 (50 mL) afforded 18 (257 mg, 88%) as a white foam: ¹H NMR (300 MHz, CDCl₃) δ 7.51 (d, 1 H, J = 2.1 Hz), 7.36 (d, 1 H, J = 8.1 Hz), 7.31 (dd, 1 H, J = 8.1 Hz, J = 2.1 Hz), 6.47 (t, 1 H, J = 6.0 Hz), 5.85 (s, 1 H), 4.72 (dt, 2 H,²J_HF = 46.8 Hz, J = 5.1 Hz), 3.73 (ddt, 2 H, ³J_HF = 25.7 Hz, J = 6.0 Hz, J = 5.1 Hz), 2.47 (s, 3 H), 2.36 (s, 3 H); ¹³C NMR (75 MHz, CDCl₃) δ 160.21, 152.52, 146.96, 145.84, 135.80, 134.25, 133.73, 130.45, 129.79, 127.26, 105.26, 85.70, 81.83 (d, ¹J_CF = 169.9 Hz), 42.55 (d, ²J_CF = 21.0 Hz), 25.62, 13.87; HRMS (APCI) [MH] Calcd for C₁₆H₁₆Cl₂FN₄: 353.0731, found: 353.0732.

4.7. 3-(2,4-dichlorophenyl)-2,5-dimethyl-N-propylpyrazolo[1,5-a]pyrimidin-7-amine (19)

Compound 15 (207 mg, 6.34 × 10⁻⁴ mol), PrNH₃Cl (255 mg, 2.67 mmol, 4.2 equiv.), iPr₂NEt (1.8 mL, 10.3 mmol, 16.2 equiv.), and CH₃CN (10 mL) were stirred at reflux under Ar_(g) for 135 min, then cooled, and the solvent was removed to give an off-white solid. The solid was dissolved in CH₂Cl₂, poured onto dry silica (8 cm h × 4 cm i.d.), and eluted under vacuum: hexane (50 mL), hexane/EtOAc/NEt₃ v/v/v 95:4:1 (50 mL), 90:8:2 (100 mL), 75:20:5 (300 mL) to afford a light yellow solid (223 mg). Purification by radial chromatography (2 mm silica): hexane/EtOAc/NEt₃ v/v/v 98:1:1 (100 mL), 95:4:1 (150 mL), 90:8:2 (100 mL) afforded 19 (213 mg, 96%) as a white solid: ¹H NMR (300 MHz, CDCl₃) δ 7.51 (d, 1 H, J = 2.1 Hz), 7.37 (d, 1 H, J = 8.1 Hz), 7.30 (dd, 1 H, J = 2.1 Hz, J = 8.1 Hz), 6.20 (t, 1 H, J = 5.7 Hz), 5.82 (s, 1 H), 3.36 (partially resolved dq, 2 H, J = 7.2 Hz), 2.46 (s, 3 H), 2.35 (s, 3 H), 1.80 (sextet, 2 H, J = 7.2 Hz), 1.08 (t, 3 H, J = 7.2 Hz); ¹³C NMR (75 MHz, CDCl₃) δ 160.13, 152.23, 147.03, 146.05, 135.79, 134.27, 133.61, 130.62, 129.76, 127.23, 104.98, 85.64, 44.01, 25.58, 22.59, 13.86, 11.69; HRMS (APCI) [MH]⁺ Calcd for C₁₇H₁₉N₄Cl₂: 349.0981, found: 349.0983.

4.8. N-butyl-3-(2,4-dichlorophenyl)-2,5-dimethylpyrazolo[1,5-a]pyrimidin-7-amine (20)

Compound 15 (244 mg, 7.47 × 10⁻⁴ mol), BuNH₂ (0.30 mL, 3.0 mmol, 4.0 equiv.), iPr₂NEt (0.30 mL, 1.7 mmol, 2.3 equiv.), and CH₃CN (20 mL) were stirred at reflux under Ar_(g) for 1 h, cooled, and the solvent was removed to give a yellow residue that was dried under vacuum. The solid was dissolved in CH₂Cl₂, poured onto dry silica (43 mm h × 43 mm i.d.), and eluted under vacuum: CH₂Cl₂ (650 mL) to afford 20 (247 mg, 91%) as a white foam: ¹H NMR (300 MHz, CDCl₃) δ 7.51 (d, 1 H, J = 2.1 Hz), 7.36 (d, 1 H, J = 8.1 Hz), 7.30 (dd, 1 H, J = 2.1 Hz, J = 8.1 Hz), 6.17 (t, 1 H, J = 5.4 Hz), 5.81 (s, 1 H), 3.39 (dt, 2 H, J = 5.4 Hz, J = 7.2 Hz), 2.46 (s, 3 H), 2.35 (s, 3 H), 1.76 (m, 2 H), 1.50 (m, 2 H), 1.01 (t, 3 H, J = 7.5 Hz); ¹³C NMR (75 MHz, CDCl₃) δ 160.11, 152.18, 147.00, 146.02, 135.76, 134.26, 133.58, 130.60, 129.72, 127.20, 104.94, 85.61, 41.96, 31.25, 25.55, 20.27, 13.90, 13.85; HRMS (APCI) [MH]⁺ Calcd for C₁₈H₂₁N₄Cl₂: 363.1138, found: 363.1140.

4.9. 3-(2,4-dichlorophenyl)-2,5-dimethyl-N,N-dipropylpyrazolo[1,5-a]pyrimidin-7-amine (21)

Compound 15 (155 mg, 4.75 × 10⁻⁴ mol), Pr₂NH (0.30 mL, 2.2 mmol, 4.6 equiv), iPr₂NEt (0.23 mL, 1.3 mmol, 2.7 equiv.), and CH₃CN (8 mL) were stirred at reflux under Ar_(g) for 3.5 h, then cooled, and the solvent was removed to give a residue that was dried under vacuum briefly. The residue was dissolved in CH₂Cl₂, poured onto dry silica (43 mm h × 43 mm i.d.), and eluted under vacuum: hexane (50 mL), hexane/EtOAc/NEt₃ v/v/v 95:4:1 (100 mL), 90:8:2 (100 mL), 75:20:5 (200 mL) to give a yellow oil (170 mg). Purification by radial chromatography (2 mm silica): hexane/EtOAc/NEt₃ v/v/v 98:1:1 (250 mL) gave a yellow oil (149 mg) that was repurified by radial chromatography (1 mm silica): hexane/EtOAc/NEt₃ v/v/v 98:1:1 (150 mL) to afford 21 (129 mg, 69%) as a white solid: ¹H NMR (300 MHz, CDCl₃) δ 7.50 (d, 1 H, J = 2.1 Hz), 7.36 (d, 1 H, J = 8.1 Hz), 7.29 (dd, 1 H, J = 2.1 Hz, J = 8.1 Hz), 5.80 (s, 1 H), 3.71 (t, 4 H, J = 7.5 Hz), 2.42 (s, 3 H), 2.34 (s, 3 H), 1.74 (sextet, 4 H, J = 7.5 Hz), 0.96 (t, 6 H, J = 7.5 Hz); ¹³C NMR (75 MHz, CDCl₃) δ 159.12, 151.57, 149.61, 148.78, 135.72, 134.32, 133.38, 131.06, 129.73, 127.17, 140.00, 92.05, 53.20, 25.29, 21.31, 14.13, 11.53; HRMS (APCI) [MH]⁺ Calcd for C₂₀H₂₅Cl₂N₄: 391.1451, found: 391.1456; Anal. Calcd for C₂₀H₂₄Cl₂N₄: C, 61.38; H, 6.18; N, 14.32; found: C, 61.65; H, 6.31; N, 14.16.

4.10. N-butyl-3-(2,4-dichlorophenyl)-N-ethyl-2,5-dimethylpyrazolo[1,5-a]pyrimidin-7-amine (22)

Compound 15 (105 mg, 3.21 × 10⁻⁴ mol), BuNHEt (0.20 mL, 1.5 mmol, 4.7 equiv.), iPr₂NEt (0.15 mL, 8.6 × 10⁻⁴ mol, 2.7 equiv.), and CH₃CN (5 mL) were stirred at reflux under Ar_(g) for 2.5 h, cooled, and the solvent was removed to give a residue that was dried under vacuum. The residue was dissolved in CH₂Cl₂, poured onto dry silica (43 mm h × 43 mm i.d.), and eluted under vacuum: hexane (50 mL), hexane/EtOAc/NEt₃ v/v/v 95:4:1 (50 mL), 90:8:2 (100 mL), 75:20:5 (200 mL) to give an off-white solid (96 mg). Purification by radial chromatography (1 mm silica): hexane/EtOAc/NEt₃ v/v/v 98:1:1 (150 mL) afforded 22 (90 mg, 72%) as a colorless syrup that became an off-white solid when stored in a freezer: ¹H NMR (300 MHz, CDCl₃) δ 7.50 (d, 1 H, J = 2.1 Hz), 7.36 (d, 1 H, J = 8.1 Hz), 7.29 (dd, 1 H, J = 2.1 Hz, J = 8.1 Hz), 5.83 (s, 1 H), 3.81 (q, 2 H, J = 6.9 Hz), 3.74 (t, 2 H, J = 7.8 Hz), 2.43 (s, 3 H), 2.34 (s, 3 H), 1.70 (pentet, 2 H, J = 7.8 Hz), 1.39 (sextet, 2 H, J = 7.5 Hz), 1.29 (t, 3 H, J = 6.9 Hz), 0.97 (t, 3 H, J = 7.5 Hz); ¹³C NMR (75 MHz, CDCl₃) δ 159.16, 151.65, 149.53, 148.73, 135.75, 134.32, 133.42, 131.02, 129.74, 127.17, 104.11, 92.32, 50.43, 45.74, 30.03, 25.28, 20.40, 14.08, 13.10; HRMS (APCI) [MH]⁺ Calcd for C₂₀H₂₅Cl₂N₄: 391.1451, found: 391.1459; Anal. Calcd for C₂₀H₂₄Cl₂N₄: C, 61.38; H, 6.18; N, 14.32; found: C, 61.63; H, 6.19; N, 14.28; HPLC: Waters Xterra t_R = 33.8 min (70:30:0.1 v/v/v MeOH/H₂O/NEt₃, 9 mL/min), t_R = 16.0 min (75:25:0.1 v/v/v MeOH/H₂O/NEt₃, 9 mL/min), Waters NovaPak t_R = 23.5 min (75:25:0.1 v/v/v MeOH/H₂O/NEt₃, 1.1 mL/min).

4.11. 4-((3-(2,4-dichlorophenyl)-2,5-dimethylpyrazolo[1,5-a]pyrimidin-7-yl)(ethyl)amino)butan-1-ol(23)

Compound 15 (106 mg, 3.25 × 10⁻⁴ mol), EtNHBuOH (0.15 mL, 1.16 mmol, 3.6 equiv.), iPr₂NEt (0.15 mL, 8.6 × 10⁻⁴ mol, 2.6 equiv.), and CH₃CN (10 mL) were stirred at reflux under Ar_(g) for 2.5 h, cooled, and the solvent was removed to give a faint yellow syrup. The syrup was dissolved in CH₂Cl₂, poured onto dry silica (43 mm h × 43 mm i.d.), and eluted under vacuum: hexane (50 mL), hexane/EtOAc/NEt₃ v/v/v 90:8:2 (50 mL), 75:20:5 (100 mL), 50:45:5 (100 mL), 20:75:5 (150 mL) to give a sticky, yellow residue (134 mg). Purification by radial chromatography (2 mm silica): hexane/EtOAc/NEt₃ v/v/v 90:8:2 (100 mL), 75:20:5 (100 mL), 50:45:5 (100 mL), 20:75:5 (50 mL) gave an off-white foam (123 mg) that was further purified by radial chromatography (1 mm silica): 1% MeOH/CHCl₃ (100 mL) to afford 23 (118 mg, 89%) as a white foam: ¹H NMR (300 MHz, CDCl₃) δ 7.50 (d, 1 H, J = 2.1 Hz), 7.35 (d, 1 H, J = 8.4 Hz), 7.29 (dd, 1 H, J = 2.1, J =8.4 Hz), 5.84 (s, 1 H), 3.87 (br s, 2 H), 3.73 (q, 2 H, J = 6.0 Hz), 3.72 (q, 2 H, J = 7.2 Hz), 2.43 (s, 3 H), 2.35 (s, 3 H), 2.01 (t, 1 H, J = 6.0 Hz), 1.88 (pentet, 2 H, J = 7.2 Hz), 1.65 (pentet, 2 H, J = 7.2 Hz), 1.30 (t, 3 H, J = 7.2 Hz); ¹³C NMR (75 MHz, CDCl₃) δ 159.36, 151.90, 149.60, 148.55, 135.76, 134.30, 133.50, 130.82, 129.70, 127.18, 104.24, 92.13, 62.15, 50.05, 45.73, 29.62, 25.17, 24.68, 13.94, 12.72; HRMS (APCI) [MH]⁺ Calcd for C₂₀H₂₅Cl₂N₄O: 407.1400, found: 407.1404.

4.12. 2-((3-(2,4-dichlorophenyl)-2,5-dimethylpyrazolo[1,5-a]pyrimidin-7-yl)(methyl)amino)ethanol (24)

Compound 15 (144 mg, 4.41 × 10⁻⁴ mol), CH₃CN (15 mL), iPr₂NEt (0.20 mL, 1.1 mmol, 2.5 equiv.), and MeNHEtOH (0.14 mL, 1.8 mmol, 4.1 equiv.) were stirred at reflux under Ar_(g) for 3 h, then cooled, and the solvent was removed to give a yellow syrup. Hexane was added and removed to give an off-white solid that was dried under vacuum. The solid was dissolved in CH₂Cl₂, poured onto dry silica (14 cm h × 4 cm i.d.), and eluted under vacuum: hexane/EtOAc/NEt₃ v/v/v 90:8:2 (100 mL), 75:20:5 (100 mL), 50:45:5 (100 mL), 20:75:5 (400 mL) to afford 24 (152 mg, 94%) as an off-white solid: ¹H NMR (300 MHz, CDCl₃) δ 7.52 (d, 1 H, J = 1.8 Hz), 7.34 (d, 1 H, J = 8.1 Hz), 7.31 (dd, 1 H, J = 1.8 Hz, J = 8.1 Hz), 6.23 (t, 1 H, J = 3.6 Hz), 5.99 (s, 1 H), 4.02 (partially resolved dt, 2 H, J = 4.5 Hz), 3.95 (br d, 2 H), 3.10 (s, 3 H), 2.47 (s, 3 H), 2.35 (s, 3 H); ¹³C NMR (75 MHz, CDCl₃) δ 160.49, 151.69, 149.83, 148.85, 135.73, 134.20, 133.87, 130.11, 129.81, 127.26, 104.92, 93.00, 59.76, 53.46, 37.89, 25.28, 13.74; HRMS (APCI) [MH]⁺ Calcd for C₁₇H₁₉Cl₂N₄O: 365.0930, found: 365.0931.

4.13. 2-((3-(2,4-dichlorophenyl)-2,5-dimethylpyrazolo[1,5-a]pyrimidin-7-yl)(ethyl)amino)ethanol (25)

Compound 15 (143 mg, 4.38 × 10⁻⁴ mol), CH₃CN (15 mL), iPr₂NEt (0.20 mL, 1.1 mmol, 2.5 equiv.), and EtNHEtOH (0.15 mL, 1.5 mmol, 3.4 equiv.) were stirred at reflux under Ar_(g) for 3.5 h, then cooled, and the solvent was removed to give a faint yellow liquid. Hexane was added and removed to give a faint yellow syrup that was dried under vacuum to give an off-white solid. The solid was dissolved in CH₂Cl₂, poured onto dry silica (14 cm h × 4 cm i.d.), and eluted under vacuum: hexane/EtOAc/NEt₃ v/v/v 90:8:2 (50 mL), 75:20:5 (50 mL), 50:45:5 (100 mL), 20:75:5 (200 mL) to afford 25 (162 mg, 98%) as a white foam: ¹H NMR (300 MHz, CDCl₃) δ 7.51 (d, 1 H, J = 1.5 Hz), 7.34 (d, 1 H, J = 8.1 Hz), 7.30 (dd, 1 H, J = 1.5 Hz, J = 8.1 Hz), 6.15 (t, 1 H, J = 3.9 Hz), 5.99 (s, 1 H), 4.08 (br d, 2 H), 3.98 (partially resolved dt, 2 H, J = 4.8 Hz), 3.56 (q, 2 H, J = 7.2 Hz), 2.46 (s, 3 H), 2.34 (s, 3 H), 1.36 (t, 3 H, J = 7.2 Hz); ¹³C NMR (75 MHz, CDCl₃) δ 160.18, 151.53, 149.36, 148.55, 135.71, 134.22, 133.79, 130.23, 129.78, 127.25, 104.68, 92.74, 59.72, 50.92, 44.58, 25.29, 13.75, 11.84; HRMS (APCI) [MH]⁺ Calcd for C₁₈H₂₁Cl₂N₄O: 379.1087, found: 379.1089.

4.14. 3-(2,4-dichlorophenyl)-N-(2-(2-fluoroethoxy)ethyl)-N,2,5-trimethylpyrazolo[1,5-a]pyrimidin-7-amine (26)

Compound 24 (126 mg, 3.45 × 10⁻⁴ mol) was dissolved in DMA (2 mL) under Ar_(g) and cooled to 0 °C. NaH (60%, 40 mg, 1.00 mmol, 2.9 equiv.) was added, the mixture was stirred at 0 °C under Ar_(g) for 30 min, and then a DMA (0.5 mL) solution of 2-fluoroethyl tosylate (207 mg, 9.48 × 10⁻⁴ mol, 2.7 equiv.) was added. The reaction mixture was warmed to ambient temperature, stirred for 16 h, then diluted with CH₂Cl₂ (1 mL) and hexane (3 mL), poured onto dry silica (43 mm h × 43 mm i.d.), and eluted under vacuum: hexane (25 mL), %CH₂Cl₂/hexane – 25% (50 mL), 50% (50 mL), 75% (50 mL), CH₂Cl₂ (150 mL). The desired fractions were combined, concentrated, dissolved in EtOAc (25 mL), washed with H₂O (10 mL × 3), sat. NaCl_(aq) (10 mL), and dried over MgSO₄. The solvent was removed to give a yellow syrup that was dissolved in CH₂Cl₂, poured onto dry silica (43 mm h × 43 mm i.d.), and eluted under vacuum: %CH₂Cl₂/hexane – 50% (50 mL), 75% (50 mL), CH₂Cl₂ (100 mL), hexane/EtOAc/NEt₃ v/v/v 75:20:5 (100 mL), 50:45:5 (100 mL), 20:75:5 (100 mL) to give a yellow syrup (131 mg). Purification by radial chromatography (2 mm silica): hexane/EtOAc/NEt₃ v/v/v 98:1:1 (100 mL), 95:4:1 (100 mL), 90:8:2 (100 mL), 75:20:5 (175 mL) gave a colorless syrup (110 mg) that was further purified by radial chromatography (2 mm silica): %CH₂Cl₂/hexane – 5% (100 mL), 10% (100 mL), 15% (100 mL), 25% (100 mL), 35% (100 mL), 50% (100 mL), 75% (100 mL), 85% (100 mL), CH₂Cl₂ (300 mL), %MeOH/CH₂Cl₂ – 1% (100 mL), 2% (50 mL) to afford 26 (64 mg, 45%) as a white solid: ¹H NMR (300 MHz, CDCl₃) δ 7.51 (d, 1 H, J = 2.1 Hz), 7.36 (d, 1 H, J = 8.4 Hz), 7.30 (dd, 1 H, J = 8.4 Hz, J = 2.1 Hz), 5.90 (s, 1 H), 4.46 (apparent dt, 2 H, ²J_HF = 47.7 Hz, ³J_HH = 4.2 Hz), 4.29 (partially resolved t, 2 H), 3.85 (t, 2 H, J = 5.1 Hz), 3.66 (apparent dt, 2 H, ³J_HF = 30.0 Hz, ³J_HH = 4.2 Hz), 2.45 (s, 3 H), 2.34 (s, 3 H); ¹³C NMR (75 MHz, CDCl₃) δ 159.63, 151.82, 149.67, 149.13, 135.73, 134.27, 133.54, 130.78, 129.74, 127.19, 104.38, 92.60, 83.18 (d, ¹J_CF = 168.2 Hz), 70.74, 70.26 (d, ²J_CF = 19.7 Hz), 52.08, 40.33, 25.26, 14.04; HRMS (NSI) [MH]⁺ Calcd for C₁₉H₂₂Cl₂FN₄O: 411.1149, found: 411.1156; Anal. Calcd for C₁₉H₂₁Cl₂FN₄O: C, 55.48; H, 5.15; N, 13.62; found: C, 55.51; H, 5.21; N, 13.70; HPLC: Waters XTerra t_R = 14.75 min (50:50:0.1 v/v/v EtOH/H₂O/NEt₃, 9 mL/min), t_R = 19.1 min (50:50:0.1 v/v/v EtOH/H₂O/NEt₃, 7 mL/min).

4.15. 3-(2,4-dichlorophenyl)-N-ethyl-N-(2-(2-fluoroethoxy)ethyl)-2,5-dimethylpyrazolo[1,5-a]pyrimidin-7-amine (27)

Compound 25 (158 mg, 4.17 × 10⁻⁴ mol) was flushed with Ar_(g), then dissolved in DMA (1.5 mL) and cooled to 0 °C. NaH (60%, 42 mg, 1.05 mmol, 2.5 equiv.) was added, the mixture was stirred at 0 °C under Ar_(g) for 30 min, and then a DMA (0.5 mL) solution of 2-fluoroethyl tosylate (248 mg, 1.14 mmol, 2.7 equiv.) was added. The reaction mixture was warmed to ambient temperature, stirred for 16 h, diluted with CH₂Cl₂ (1 mL) and hexane (3 mL), poured onto dry silica (43 mm h × 43 mm i.d.), and eluted under vacuum: hexane (25 mL), %CH₂Cl₂/hexane – 25% (50 mL), 50% (50 mL), 75% (50 mL), CH₂Cl₂ (200 mL). The desired fractions were combined, concentrated, dissolved in EtOAc (25 mL), washed with H₂O (10 mL × 3), sat. NaCl_(aq) (10 mL), and dried over MgSO₄. The solvent was removed to give a light yellow syrup that was dissolved in CH₂Cl₂, poured onto dry silica (43 mm h × 43 mm i.d.), and eluted under vacuum: hexane (50 mL), %CH₂Cl₂/hexane – 25% (50 mL), 50% (50 mL), 75% (100 mL), CH₂Cl₂ (125 mL), hexane/EtOAc/NEt₃ v/v/v 75:20:5 (300 mL), 50:45:5 (50 mL) to give an opaque syrup (164 mg). Purification by radial chromatography (2 mm silica): hexane/EtOAc/NEt₃ v/v/v 98:1:1 (200 mL), 95:4:1 (100 mL), 90:8:2 (100 mL), 75:20:5 (100 mL) gave a colorless syrup (144 mg) that was further purified by radial chromatography (2 mm silica): %CH₂Cl₂/hexane – 5% (100 mL), 10% (100 mL), 15% (100 mL), 25% (100 mL), 35% (100 mL), 50% (100 mL), 75% (200 mL), 85% (100 mL), CH₂Cl₂ (300 mL), %MeOH/CH₂Cl₂ – 1% (100 mL), 2% (50 mL) to afford 27 (89 mg, 50%) as a crystalline, white solid: ¹H NMR (300 MHz, CDCl₃) δ 7.50 (d, 1 H, J = 2.1 Hz), 7.35 (d, 1 H, J = 8.4 Hz), 7.29 (dd, 1 H, J = 8.4 Hz, J = 2.1 Hz), 5.91 (s, 1 H) 4.51 (dddd, 2 H, ²J_HF = 47.4 Hz, ³J_HH = 4.2 Hz, J = 1.2 Hz), 4.16 (t, 2 H, ³J_HH = 5.7 Hz), 3.87 (t, 2 H, ³J_HH = 5.7 Hz), 3.74 (q, 2 H, J = 7.2 Hz), 3.69 (apparent dt, 2 H, ³J_HF = 30.0 Hz, ³J_HH = 4.2 Hz), 2.44 (s, 3 H), 2.33 (s, 3 H), 1.31 (t, 3 H, J = 7.2 Hz); ¹³C NMR (75 MHz, CDCl₃) δ 159.37, 151.68, 149.46, 148.67, 135.70, 134.27, 133.46, 130.87, 129.72, 127.17, 104.18, 92.34, 83.18 (d, ¹J_CF = 168.2 Hz), 70.89, 70.38 (d, ²J_CF = 19.4 Hz), 49.99, 46.77, 25.28, 14.05, 12.57; HRMS (APCI) [MH]⁺ Calcd for C₂₀H₂₄Cl₂FN₄O: 425.1306, found: 425.1309; Anal. Calcd for C₂₀H₂₃Cl₂FN₄O: C, 56.48; H, 5.45; N, 13.17; found: C, 56.36; H, 5.48; N, 13.11; HPLC: Waters XTerra t_R = 12.2 min (55:45:0.1 v/v/v EtOH/H₂O/NEt₃, 9 mL/min), t_R = 20.5 min (50:50:0.1 v/v/v EtOH/H₂O/NEt₃, 9 mL/min).

4.16. 3-(2,4-dichlorophenyl)-N-(4-fluorobutyl)-N,2,5-trimethylpyrazolo[1,5-a]pyrimidin-7-amine(28)

Compound 16 (105 mg, 3.27 × 10⁻⁴ mol) was flushed with Ar_(g), then dissolved in DMA (2 mL) and cooled to 0 °C. NaH (60%, 38 mg, 9.50 × 10⁻⁴ mol, 2.9 equiv.) was added, the mixture was stirred at 0 °C under Ar_(g) for 30 min, and then 1-bromo-4-fluorobutane (0.12 mL, 1.12 mmol, 3.4 equiv.) was added. The reaction mixture was warmed to ambient temperature, stirred for 16 h, diluted with CH₂Cl₂ (1 mL) and hexane (2 mL), poured onto dry silica (43 mm h × 43 mm i.d.), and eluted under vacuum: hexane (50 mL), %CH₂Cl₂/hexane – 25% (50 mL), 50% (50 mL), 75% (50 mL), CH₂Cl₂ (200 mL). The desired fractions were combined, concentrated to a light yellow liquid, dissolved in EtOAc (25 mL), washed with H₂O (10 mL × 3), sat. NaCl_(aq) (10 mL), dried over MgSO₄, and then the solvent was removed to give a faint yellow syrup that was dried under vacuum (121 mg). Purification by radial chromatography (2 mm silica): hexane/EtOAc/NEt₃ v/v/v 98:1:1 (400 mL), 95:4:1 (100 mL), 90:8:2 (150 mL) gave a colorless syrup (104 mg) that was further purified by radial chromatography (2 mm silica): CHCl₃ (150 mL) to give a colorless syrup (100 mg). Purification by flash column chromatography (10 g silica, 90:8:2 v/v/v hexane/EtOAc/NEt₃) gave a colorless syrup (94 mg) that was further purified by radial chromatography (2 mm silica): CHCl₃ (125 mL) to give a white solid (89 mg). A final purification by radial chromatography (2 mm silica): %i-PrOH/hexane – 1% (100 mL), 2% (100 mL), 3% (100 mL) afforded 28 (87 mg, 67%) as a white solid: ¹H NMR (300 MHz, CDCl₃) δ 7.51 (d, 1 H, J = 2.1 Hz), 7.36 (d, 1 H, J = 8.4 Hz), 7.30 (dd, 1 H, J = 8.4 Hz, J = 2.1 Hz), 5.84 (s, 1 H), 4.51 (dt, 2 H, ²J_HF = 47.1 Hz, J = 6.0 Hz), 4.02 (br unresolved t, 2 H), 3.17 (s, 3 H), 2.45 (s, 3 H), 2.35 (s, 3 H), 1.85 (m, 2 H), 1.79 (dm, 2 H); ¹³C NMR (75 MHz, CDCl₃) δ 159.53, 151.86, 149.53, 149.20, 135.74, 134.27, 133.53, 130.81, 129.74, 127.18, 104.41, 92.29, 83.90 (d, ¹J_CF = 164.0 Hz), 52.27, 38.91, 27.78 (d, ²J_CF = 20.0 Hz), 25.25, 23.93, 14.04; HRMS (APCI) [MH]⁺ Calcd for C₁₉H₂₂Cl₂FN₄: 395.1200, found: 395.1203; Anal. Calcd for C₁₉H₂₁Cl₂FN₄: C, 57.73; H, 5.35; N, 14.17; found: C, 57.72; H, 5.39; N, 14.13; HPLC: Waters XTerra t_R = 15.0 min (70:30:0.1 v/v/v MeOH/H₂O/NEt₃, 9 mL/min), t_R = 15.7 min (55:45:0.1 v/v/v EtOH/H₂O/NEt₃, 9 mL/min), Waters Nova-Pak t_R = 7.8 min (75:25:0.1 v/v/v MeOH/H₂O/NEt₃, 1.1 mL/min).

4.17. 3-(2,4-dichlorophenyl)-N-ethyl-N-(4-fluorobutyl)-2,5-dimethylpyrazolo[1,5-a]pyrimidin-7-amine (29)

Compound 17 (113 mg, 3.37 × 10⁻⁴ mol) was flushed with Ar_(g) for 45 min, then dissolved in DMA (2 mL), and cooled to 0 °C. NaH (60%, 18 mg, 7.50 × 10⁻⁴ mol, 2.2 equiv.) was added, the mixture was stirred at 0 °C under Ar_(g) for 1 h, and then 1-bromo-4-fluorobutane (0.15 mL, 1.4 mmol, 4.2 equiv) was added. The reaction mixture was warmed to ambient temperature, stirred for 20 h, diluted with CH₂Cl₂ (1 mL) and hexane (2 mL), poured onto dry silica (43 mm h × 43 mm i.d.), and eluted under vacuum: hexane (25 mL), %CH₂Cl₂/hexane – 25% (50 mL), 50% (50 mL), 75% (100 mL), CH₂Cl₂ (150 mL) to afford a yellow syrup (126 mg). Purification by radial chromatography (2 mm silica): hexane/EtOAc/NEt₃ v/v/v 98:1:1 (200 mL), 95:4:1 (100 mL), 90:8:2 (100 mL), 75:20:5 (100 mL) afforded a colorless syrup (102 mg) that was again purified by radial chromatography (2 mm silica): %i-PrOH/hexane – 1% (100 mL), 2% (100 mL), 3% (100 mL) to give a colorless syrup (86 mg). Further purification by radial chromatography (2 mm silica): CHCl₃ (100 mL) afforded 29 (82 mg, 59%) as a colorless syrup: ¹H NMR (300 MHz, CDCl₃) δ 7.51 (d, 1 H, J = 2.1 Hz), 7.36 (d, 1 H, J = 8.4 Hz), 7.30 (dd, 1 H, J = 2.1 Hz, J = 8.4 Hz), 5.84 (s, 1 H), 4.51 (dt, 2 H, J = 6.0 Hz, ²J_HF = 47.1 Hz), 3.85 (br s, 2 H), 3.75 (q, 2 H, J = 7.2 Hz), 2.43 (s, 3 H), 2.34 (s, 3 H), 1.86 (m, 2 H), 1.79 (dm, 2 H, ³J_HF = 32.4 Hz), 1.30 (t, 3 H, J = 7.2 Hz); ¹³C NMR (75 MHz, CDCl₃) δ 159.27, 151.74, 149.50, 148.56, 135.73, 134.29, 133.47, 130.92, 129.74, 127.19, 104.23, 92.31, 83.94 (d, ¹J_CF = 164.2 Hz), 50.04, 45.71, 27.96 (d, ²J_CF = 20.0 Hz), 25.30, 24.27 (d, ³J_CF = 15.9 Hz), 14.07, 12.84; HRMS (APCI) [MH]⁺ Calcd for C₂₀H₂₄Cl₂FN₄: 409.1357, found: 409.1367; Anal. Calcd for C₂₀H₂₃Cl₂FN₄: C, 58.68; H, 5.66; N, 13.69; found: C, 58.38; H, 5.74; N, 13.41; HPLC: Waters Xterra t_R = 19.4 min (70:30:0.1 v/v/v MeOH/H₂O/NEt₃, 9 mL/min), t_R = 20.7 min (55:45:0.1 v/v/v EtOH/H₂O/NEt₃, 9 mL/min).

4.18. 3-(2,4-dichlorophenyl)-N-(4-fluorobutyl)-N-(2-fluoroethyl)-2,5-dimethylpyrazolo[1,5-a]pyrimidin-7-amine (30)

Compound 18 (42 mg, 1.19 × 10⁻⁴ mol) was dissolved in DMA (1 mL) under Ar_(g) and cooled to 0 °C. NaH (60%, 13 mg, 3.25 × 10⁻⁴ mol, 2.7 equiv.) was added, the mixture was stirred at 0 °C under Ar_(g) for 1 h, and then 1-bromo-4-fluorobutane (0.1 mL, 9.3 × 10⁻⁴ mol, 7.8 equiv.) was added. The reaction mixture was warmed to ambient temperature, stirred for 21 h, and then diluted with CH₂Cl₂ (1 mL) and hexane (3 mL). The mixture was poured onto dry silica (33 mm h × 33 mm i.d.) and eluted under vacuum: hexane (25 mL), %CH₂Cl₂/hexane – 25% (25 mL), 50% (25 mL), 75% (50 mL), CH₂Cl₂ (50 mL) to give a light yellow syrup (45 mg). Purification by radial chromatography (1 mm silica): hexane/EtOAc/NEt₃ v/v/v 98:1:1 (100 mL), 95:4:1 (100 mL), 90:8:2 (100 mL), 75:20:5 (100 mL) gave 18 (11 mg, 26% recovery) and 30 (29 mg) as a colorless syrup. Further purification by radial chromatography (1 mm silica): CHCl₃ (30 mL) afforded 30 (24 mg, 47%) as a colorless syrup: ¹H NMR (300 MHz, CDCl₃) δ 7.51 (d, 1 H, J = 1.8 Hz), 7.35 (d, 1 H, J = 8.1 Hz), 7.30 (dd, 1 H, J = 8.1 Hz, J = 1.8 Hz), 5.92 (s, 1 H), 4.79 (dt, 2 H, ²J_HF = 48.0 Hz, J = 4.5 Hz), 4.51 (dt, 2 H, ²J_HF = 47.1 Hz, J = 5.7 Hz), 4.28 (d of br s, 2 H, ³J_HF = 25.2 Hz), 3.72 (t, 2 H, J = 7.2 Hz), 2.45 (s, 3 H), 2.32 (s, 3 H), 1.89 (m, 2 H), 1.83 (dm, 2 H); ¹³C NMR (75 MHz, CDCl₃) δ 159.53, 151.94, 149.40, 148.33, 135.73, 134.26, 133.64, 130.67, 129.78, 127.24, 104.52, 92.70, 84.32 (d, ¹J_CF = 167.6 Hz), 83.87 (d, ¹J_CF = 164.2 Hz), 52.12, 51.53 (d, ²J_CF = 19.6 Hz), 27.89 (d, ²J_CF= 19.9 Hz), 25.34, 23.31 (d, ²J_CF = 4.0 Hz), 14.05; HRMS (NSI) [MH]⁺ Calcd for C₂₀H₂₃Cl₂F₂N₄: 427.1262, found: 427.1262; Anal. Calcd for C₂₀H₂₂Cl₂F₂N₄: C, 56.21; H, 5.19; N, 13.11; found: C, 56.07; H, 5.29; N, 13.02; HPLC: Waters XTerra t_R = 16.7 min (55:45:0.1 v/v/v EtOH/H₂O/NEt₃, 9 ml/min,).

4.19. 4-((3-(2,4-dichlorophenyl)-2,5-dimethylpyrazolo[1,5-a]pyrimidin-7-yl)(methyl)amino)butyl 4-methylbenzenesulfonate (31)

Compound 16 (28 mg, 8.72 × 10⁻⁵ mol) was flushed with Ar_(g), then dissolved in DMA (0.5 mL), and cooled to 0 °C. NaH (60%, 13 mg, 3.25 × 10⁻⁴ mol, 3.7 equiv.) was added, the mixture was stirred at 0 °C under Ar_(g) for 25 min, and then a DMA (1 mL) solution of 1,4-ditosyloxybutane (35) (174 mg, 4.37 × 10⁻⁴ mol, 5.0 equiv.) was added. The reaction mixture was warmed to ambient temperature, stirred for 19 h, then diluted with CH₂Cl₂ (1 mL) and hexane (1 mL), poured onto dry silica (33 mm h × 33 mm i.d.), and eluted under vacuum: hexane (10 mL), %CH₂Cl₂/hexane – 25% (10 mL), 50% (50 mL), 75% (50 mL), CH₂Cl₂ (25 mL). The desired fractions were combined, concentrated, dissolved in EtOAc (25 mL), washed with H₂O (5 mL × 3), sat. NaCl_(aq) (5 mL), and dried over MgSO₄. The solvent was removed to give a faint yellow oil that was dried under vacuum briefly to give an opaque residue that was purified by preparative-TLC (CH₂Cl₂) to give a faint yellow residue (27 mg). The residue was purified again by preparative-TLC, and then on Waters silica Sep-Pak Classics (2 in series): load – CH₂Cl₂ (1 mL), elution – hexane (1 mL), hexane/EtOAc/NEt₃ v/v/v 98:1:1 (3 mL), 95:4:1 (2 mL × 6), 90:8:2 (2 mL × 6), 75:20:5 (2 mL × 12) to afford 31 (13 mg, 27%) as a colorless residue: ¹H NMR (300 MHz, CDCl₃) δ 7.78 (d, 2 H, J = 8.1 Hz), 7.51 (d, 1 H, J = 2.1 Hz), 7.35 (d, 1 H, J = 8.4 Hz), 7.33 (d, 2 H, J = 8.1 Hz), 7.30 (dd, 1 H, J = 8.4 Hz, J = 2.1 Hz), 5.82 (s, 1 H), 4.09 (t, 2 H, J = 6.0 Hz), 3.95 (br s, 2 H), 3.11 (s, 3 H), 2.44 (s – overlapping resonances, 3 H + 3 H), 2.31 (s, 3 H), 1.76 (m – overlapping resonances, 2 H + 2 H).

4.20. 4-((3-(2,4-dichlorophenyl)-2,5-dimethylpyrazolo[1,5-a]pyrimidin-7-yl)(methyl)amino)butyl-1,1,2,2,3,3,4,4-d₈4-methylbenzenesulfonate (31-d₈)

Compound 16 (50 mg, 1.56 × 10⁻⁴ mol) was flushed with Ar_(g) for 1 h, then dissolved in DMA (1 mL) and cooled to 0 °C. NaH (60%, 25 mg, 6.25 × 10⁻⁴ mol, 4.0 equiv.) was added, the mixture was stirred at 0 °C under Ar_(g) for 30 min, and then 1,4-ditosyloxybutane-d₈ (35-d₈) (159 mg, 3.91 × 10⁻⁴ mol, 2.5 equiv.) was added. The reaction mixture was warmed to ambient temperature, stirred for 3 h, diluted with CH₂Cl₂ (1 mL) and hexane (1 mL), poured onto dry silica (33 mm h × 33 mm i.d.), and eluted under vacuum: hexane (10 mL), hexane/EtOAc/NEt₃ v/v/v 95:4:1 (25 mL), 90:8:2 (25 mL), 75:20:5 (200 mL). The desired fractions were combined, concentrated to a faint yellow liquid, dissolved in EtOAc (25 mL), washed with H₂O (10 mL × 2), sat. NaCl_(aq) (10 mL), and dried over MgSO₄. The solvent was removed to give a residue that was dried under vacuum (119 mg) and then purified by radial chromatography (2 mm silica): hexane/EtOAc/NEt₃ v/v/v 98:1:1 (100 mL), 95:4:1 (100 mL), 90:8:2 (100 mL), 75:20:5 (300 mL) afforded 16 (20 mg, 40% recovery) and 31-d₈ (21 mg, colorless residue) which was further purified by preparatory-TLC (75:20:5 v/v/v hexane/EtOAc/NEt₃) to afford 31-d₈ (15 mg, 17%) as an off-white foam: ¹ H NMR (300 MHz, CDCl₃) δ 7.78 (d, 2 H, J = 8.4 Hz), 7.51 (d, 1 H, J = 2.1 Hz), 7.35 (d, 1 H, J = 8.4 Hz), 7.33 (d, 2 H, J = 8.4 Hz), 7.30 (dd, 1 H, J = 8.4 Hz, J = 2.1 Hz), 5.81 (s, 1 H), 3.11 (s, 3 H), 2.44 (s – overlapping resonances, 3 H + 3 H), 2.31 (s, 3 H); HRMS (NSI) [MH]⁺ Calcd for C₂₆H₂₁D₈O₃N₄Cl₂S: 555.1834, found: 555.1832.

4.21. 4-((3-(2,4-dichlorophenyl)-2,5-dimethylpyrazolo[1,5-a]pyrimidin-7-yl)(ethyl)amino)butyl 4-methylbenzenesulfonate (32)

Compound 17 (36 mg, 1.07 × 10⁻⁴ mol) was flushed with Ar_(g), dissolved in DMA (0.5 mL), and cooled to 0 °C. NaH (60%, 14 mg, 3.50 × 10⁻⁴ mol, 3.3 equiv.) was added, the mixture was stirred at 0 °C under Ar_(g) for 25 min, and then a DMA (1 mL) solution of 1,4-ditosyloxybutane (35) (214 mg, 5.37 × 10⁻⁴ mol, 5.0 equiv.) was added. The reaction mixture was warmed to ambient temperature, stirred for 19 h, diluted with CH₂Cl₂ (1 mL) and hexane (1 mL), poured onto dry silica (33 mm h × 33 mm i.d.), and eluted under vacuum: hexane (10 mL), %CH₂Cl₂/hexane – 25% (10 mL), 50% (50 mL), 75% (50 mL), CH₂Cl₂ (50 mL). The desired fractions were combined, concentrated, and purified by preparative-TLC (CH₂Cl₂) to give a colorless residue (23 mg). The residue was purified again by preparative-TLC (75:20:5 v/v/v hexane/EtOAc/NEt₃), and then on Waters silica Sep-Pak Classics (2 in series): load – CH₂Cl₂ (1 mL), elution – hexane (2 mL), hexane/EtOAc/NEt₃ v/v/v 98:1:1 (3 mL), 95:4:1 (3 mL × 2), 90:8:2 (3 mL × 4), 75:20:5 (3 mL × 6) to afford 32 (15 mg, 25%) as a white foam that was used without further purification: ¹H NMR (300 MHz, CDCl₃) δ 7.78 (d, 2 H, J = 8.1 Hz), 7.50 (d, 1 H, J = 2.1 Hz), 7.35 (d, 1 H, J = 8.4 Hz), 7.33 (d, 2 H, J = 8.1 Hz), 7.29 (dd, 1 H, J = 8.4 Hz, J = 2.1 Hz), 5.82 (s, 1 H), 4.09 (t, 2 H, J = 6.0 Hz), 3.80 (br s, 2 H), 3.68 (q, 2 H, J = 7.2 Hz), 2.44 (s, 3 H), 2.43 (s, 3 H), 2.30 (s, 3 H), 1.76 (m – overlapping resonances, 2 H + 2 H), 1.26 (t, 3 H, J = 7.2 Hz); HRMS (NSI) [MH]⁺ Calcd for C₂₇H₃₁Cl₂N₄O₃S: 561.1494, found: 561.1523; [MH-pTsOH]⁺ Calcd for C₂₀H₂₃Cl₂N₄: 389.1300, found: 389.1297.

4.22. 2-(2-((3-(2,4-dichlorophenyl)-2,5-dimethylpyrazolo[1,5-a]pyrimidin-7-yl)(methyl)amino)ethoxy)ethyl 4-methylbenzenesulfonate (33)

Compound 16 (47 mg, 1.46 × 10⁻⁴ mol) was flushed with Ar_(g) for 30 min, then dissolved in DMA (0.5 mL) and cooled to 0 °C. NaH (60%, 14 mg, 3.50 × 10⁻⁴ mol, 2.4 equiv.) was added, the mixture was stirred at 0 °C under Ar_(g) for 30 min, and then a DMA (1 mL) solution of diethyleneglycol p-toluenesulfonate (212 mg, 5.11 × 10⁻⁴ mol, 3.5 equiv.) was added. The reaction mixture was warmed to ambient temperature, stirred for 16 h, then diluted with CH₂Cl₂ (1 mL) and hexane (2 mL), poured onto dry silica (33 mm h × 33 mm i.d.), and eluted under vacuum: %CH₂Cl₂/hexane – 25% (25 mL), 50% (25 mL), 75% (25 mL), CH₂Cl₂ (75 mL). The desired fractions were combined, concentrated to a faint yellow liquid, dissolved in EtOAc (25 mL), washed with H₂O (10 mL × 3), sat. NaCl_(aq) (10 mL), and dried over MgSO₄. The solvent was removed to give a faint yellow residue (216 mg) that was purified by radial chromatography (2 mm silica): %CH₂Cl₂/hexane – 25% (100 mL), 50% (100 mL), 75% (100 mL), CH₂Cl₂ (200 mL), %MeOH/CH₂Cl₂ – 1% (100 mL), 2% (100 mL) to give a white foam (45 mg). Further purification by radial chromatography (1 mm silica): CHCl₃ (100 mL) afforded 33 (42 mg, 51%) as a white foam: ¹H NMR (300 MHz, CDCl₃) δ 7.77 (d, 2 H, J = 8.4 Hz), 7.51 (d, 1 H, J = 2.1 Hz), 7.36 (d, 1 H, J = 8.4 Hz), 7.31 (d, 2 H, J = 8.4 Hz), 7.30 (dd, 1 H, J = 8.4 Hz, J = 2.1 Hz), 5.89 (s, 1 H), 4.21 (partially resolved t, 2 H), 4.08 (t, 2 H, J = 4.8 Hz), 3.76 (t, 2 H, J = 5.1 Hz), 3.61 (t, 2 H, J = 4.8 Hz), 3.16 (s, 3 H), 2.45 (s, 3 H), 2.43 (s, 3 H), 2.32 (s, 3 H); HRMS (NSI) [MH]⁺ Calcd for C₂₆H₂₉Cl₂N₄O₄S: 563.1281, found: 563.1277.

4.23. 2-(2-((3-(2,4-dichlorophenyl)-2,5-dimethylpyrazolo[1,5-a]pyrimidin-7-yl)(ethyl)amino)ethoxy)ethyl 4-methylbenzenesulfonate (34)

Compound 17 (55 mg, 1.64 × 10⁻⁴ mol) was flushed with Ar_(g) for 30 min, then dissolved in DMA (2 mL) and cooled to 0 °C. NaH (60%, 22 mg, 5.50 × 10⁻⁴ mol, 3.4 equiv.) was added, the mixture was stirred at 0 °C under Ar_(g) for 35 min, and then diethyleneglycol p-toluenesulfonate (242 mg, 5.84 × 10⁻⁴ mol, 3.6 equiv.) was added. The reaction mixture was warmed to ambient temperature, stirred for 16 h, then diluted with CH₂Cl₂ (1 mL) and hexane (3 mL), poured onto dry silica (43 mm h × 43 mm i.d.), and eluted under vacuum: hexane (25 mL), %CH₂Cl₂/hexane – 25% (50 mL), 50% (50 mL), 75% (50 mL), CH₂Cl₂ (100 mL). The desired fractions were combined, concentrated, dissolved in EtOAc (25 mL), washed with H₂O (10 mL × 3), sat. NaCl_(aq) (10 mL), and dried over MgSO₄. The solvent was removed to give a colorless syrup that was dried under vacuum to give an off-white solid (284 mg). Purification by radial chromatography (2 mm silica): hexane/EtOAc/NEt₃ v/v/v 98:1:1 (100 mL), 95:4:1 (100 mL), 90:8:2 (100 mL), 75:20:5 (300 mL) gave 17 (30 mg, 55% recovery) and 34 (35 mg) as a sticky, faint yellow residue. Further purification by radial chromatography (1 mm silica): CHCl₃ (100 mL) afforded 34 (26 mg, 27%) as a sticky, colorless residue: ¹H NMR (300 MHz, CDCl₃) δ 7.78 (d, 2 H, J = 8.4 Hz), 7.50 (d, 1 H, J = 2.1 Hz), 7.35 (d, 1 H, J = 8.4 Hz), 7.32 (d, 2 H, J = 8.4 Hz), 7.29 (dd, 1 H, J = 8.4 Hz, J = 2.1 Hz), 5.89 (s, 1 H), 4.11 (t, 2 H, J = 4.5 Hz), 4.08 (t, 2 H, J = 5.4 Hz), 3.77 (t, 2 H, J = 5.4 Hz), 3.68 (q, 2 H, J = 6.9 Hz), 3.64 (t, 2 H, J = 4.5 Hz), 2.43 (s – overlapping resonances, 3 H + 3 H), 2.31 (s, 3 H), 1.28 (t, 3 H, J = 6.9 Hz); ¹³C NMR (75 MHz, CDCl₃) δ 159.46, 151.70, 149.48, 148.61, 145.09,135.71, 134.29, 133.51, 133.15, 130.86, 130.04, 129.75, 128.11, 127.22, 104.21, 93.29, 71.00, 69.37, 68.70, 49.91, 46.83, 25.30, 21.85, 14.08, 12.54; HRMS (NSI) [MH]⁺ Calcd for C₂₇H₃₁Cl₂N₄O₄S: 577.1438, found: 577.1439.

4.24. 1,4-Ditosyloxybutane (35)

AgOTs (2.51 g, 8.99 mmol, 2.4 equiv) was dissolved in CH₃CN (60 mL) under Ar_(g) followed by addition of 1,4-dibromobutane (0.45 mL, 3.77 mmol). The reaction mixture was stirred at reflux under Ar_(g) for 17 h, then cooled to 0 °C, and filtered. The precipitate was rinsed with CH₃CN and dried under vacuum to afford a green powder (AgBr, 1.37g, 81%). The solvent was removed from the filtrate to give a white solid that was dried under vacuum, then suspended in CH₂Cl₂ (25 mL) and filtered. The precipitate was rinsed with CH₂Cl₂ and dried under vacuum to afford a white crystalline solid (AgOTs, 0.39 g, 16% recovery). The solvent was removed from the filtrate to give a colorless syrup that became an off-white solid under vacuum (1.54 g). The solid was dissolved in CH₂Cl₂, poured onto dry silica (8.5 cm h × 4 cm i.d.), and eluted under vacuum: hexane (25 mL), %CH₂Cl₂/hexane – 25% (50 mL), 50% (100 mL), 75% (100 mL), CH₂Cl₂ (350 mL) to afford 1-bromo-4-tosyloxy-butane (36) (94 mg, 8%) as a colorless syrup (spectroscopic data reported in Section 4.25), and 1,4-ditosyloxy-butane (35) (1.37 g, 91%) as a white solid: TLC R_f = 0.58 (silica, CH₂Cl₂); ¹H NMR (300 MHz, CDCl₃) δ 7.76 (d, 4 H, J = 8.1 Hz), 7.35 (d, 4 H, J = 8.1 Hz), 3.99 (m, 4 H), 2.46 (s, 6 H), 1.70 (m, 4 H); ¹³C NMR (75 MHz, CDCl₃) δ 145.14, 132.98, 130.11, 128.03, 69.58, 25.21, 21.85.

4.25. 1-Bromo-4-tosyloxy-butane (36)

(Obtained as a side product from Section 4.24): TLC R_f = 0.73 (silica, CH₂Cl₂); ¹H NMR (300 MHz, CDCl₃) δ 7.79 (d, 2 H, J = 8.1 Hz), 7.36 (d, 2 H, J = 8.1 Hz), 4.06 (t, 2 H, J = 6.0 Hz), 3.37 (t, 2 H, J = 6.0 Hz), 2.46 (s, 3 H), 1.91 (m, 2 H), 1.81 (m, 2 H); ¹³C NMR (75 MHz, CDCl₃) δ 145.11, 133.07, 130.11, 128.07, 69.57, 32.83, 28.69, 27.61, 21.86.

4.26. 1,4-Ditosyloxybutane-d₈ (35-d₈, TsO-CD₂CD₂CD₂CD₂-OTs)

1,4-Dibromobutane-d₈ (0.44 g, 1.96 mmol) was dissolved in CH₃CN (35 mL) followed by addition of AgOTs (1.31 g, 4.69 mmol, 2.4 equiv.). The reaction mixture was stirred at reflux under Ar_(g) for 24 h, then cooled to 0 °C, and filtered. The precipitate was washed with CH₃CN and dried under vacuum to afford a green solid (AgBr, 0.61 g, 83%). The solvent was removed from the filtrate to give a white solid that was dried under vacuum, then suspended in CH₂Cl₂, filtered, rinsed with CH₂Cl₂, and dried under vacuum to afford a white crystalline solid (AgOTs, 0.20 g, 15% recovery). The filtrate was concentrated, poured onto dry silica (6 cm h × 4 cm i.d.), and eluted under vacuum: hexane (25 mL), %CH₂Cl₂/hexane – 25% (50 mL), 50% (50 mL), 75% (50 mL), CH₂Cl₂ (400 mL) to afford 35-d₈ (0.72 g, 90%) as a white crystalline solid: ¹H NMR (300 MHz, CDCl₃) δ 7.76 (d, 4 H, J = 8.1 Hz), 7.35 (d, 4 H, J = 8.1 Hz), 2.46 (s, 6 H); ¹³C NMR (75 MHz, CDCl ) δ 145.11, 132.93, 130.08, 127.97, 68.83 (pentet, ¹J_CD = 22.7 Hz), 24.04 (pentet, ¹J_CD = 19.6 Hz), 21.81; HRMS (NSI) [MH]⁺ Calcd for C₁₈H₁₅D₈O₆S₂: 407.1433, found: 407.1433.

4.27. 3-(2,4-dichlorophenyl)-N-(2-(2-[¹⁸F]fluoroethoxy)ethyl)-N,2,5-trimethylpyrazolo[1,5-a]pyrimidin-7-amine ([¹⁸F]26)

H¹⁸F_(aq) (1786 mCi (66.1 GBq)) in H ¹⁸₂O was obtained from PETNET and collected in a V-vial housed in a Capintec detector. The radioactivity sample was divided and 901 mCi (33.3 GBq) was delivered under Ar_(g) pressure to a CPCU, collected on a trap/release cartridge, released with K₂CO_3(aq) (0.9 mg in 0.6 mL H₂O), and added to a solution of Kryptofix 222 (5 mg in 1 mL CH₃CN) in the CPCU side-arm reaction vessel. The vessel was placed in a 110 °C oil bath and the solvent was evaporated with an Ar_(g) flow. CH₃CN (3.5 mL) was added and evaporated in order to azeotropically dry the Kryptofix 222/K¹⁸F. A solution of 33 (4.8 mg) in CH₃CN (1 mL) was added to the vessel containing the Kryptofix 222/K¹⁸F, the mixture was heated at 110 °C for 15 min, and then cooled. HPLC solvent (3 mL) was added to the vessel and the solution was transferred under Ar_(g) pressure to a hot cell and collected in a conical tube. The CPCU vessel was rinsed with HPLC solvent (3 mL) and this was added to the conical tube. The solution was drawn into a 10-mL HPLC injector loop and then purified by semi-preparative HPLC (Waters XTerra, 50:50:0.1 v/v/v EtOH/H₂O/NEt₃, 7 mL/min, t_R (range) = 14 – 17 min). The desired fractions were combined, diluted 1:1 v/v with H₂O, loaded onto a Waters tC₁₈ Sep-Pak, and rinsed with 0.9 % NaCl_(aq) (30 mL). The radiotracer was eluted from the Sep-Pak with EtOH (1.5 mL) and collected in a sealed sterile vial containing 0.9 % NaCl_(aq) (13.5 mL) to give [¹⁸F]26 (204 mCi (7.5 GBq), 39.6% rcy decay-corrected). The solution was passed successively through a 1 μm filter and then a 0.2 μm filter (Acrodisc PTFE) under Ar_(g) pressure and collected in a sealed sterile dose vial (164 mCi (6.1 GBq), 82.4% recovery, decay-corrected; pH = 5.5). Radioactivity remaining on filters: ~35 mCi (~1.3 GBq). The total synthesis time was ~93 min from transfer of the H¹⁸F_(aq) to the CPCU. The radiotracer was analyzed by analytical HPLC (Waters Nova-Pak, 3.9 × 150 mm, 75:25:0.1 v/v/v MeOH/H₂O/NEt₃, 1 mL/min, t_R = 5.4 min) to determine the radiochemical purity (99.4%) and the specific activity (1.2 × 10⁴ Ci/mol).

4.28. 3-(2,4-dichlorophenyl)-N-ethyl-N-(2-(2-[¹⁸F]fluoroethoxy)ethyl)-2,5-dimethylpyrazolo[1,5-a]pyrimidin-7-amine ([¹⁸F]27)

H¹⁸F_(aq) (1929 mCi (71.4 GBq)) in H₂ ¹⁸₂O was obtained from PETNET and collected in a V-vial housed in a Capintec detector. Using Ar_(g) pressure 1849 mCi (68.4 GBq) was delivered to a CPCU, collected on a trap/release cartridge, released with K₂CO_3(aq) (0.9 mg in 0.6 mL H₂O), and added to a solution of Kryptofix 222 (5 mg in 1 mL CH₃CN) in the CPCU side-arm reaction vessel. The vessel was placed in a 110 °C oil bath and the solvent was evaporated with an Ar_(g) flow. CH₃CN (3.5 mL) was added and evaporated in order to azeotropically dry the Kryptofix 222/K¹⁸F. A solution of 34 (4.5 mg) in CH₃CN (1 mL) was added to the vessel containing the Kryptofix 222/K¹⁸F, the mixture was heated at 110 °C for 15 min, and then cooled. HPLC solvent (3 mL) was added to the vessel and the solution was transferred under Ar_(g) pressure to a hot cell and collected in a conical tube. The CPCU vessel was rinsed with HPLC solvent (3 mL) and this was added to the conical tube. The solution was drawn into a 10-mL HPLC injector loop and then purified by semi-preparative HPLC (Waters XTerra, 50:50:0.1 v/v/v EtOH/H₂O/NEt₃, 9 mL/min, t_R (range) = 16 – 18 min). The desired fractions were combined, diluted 1:1 v/v with H₂O, loaded onto a Waters tC₁₈ Sep-Pak, and rinsed with 0.9 % NaCl_(aq) (30 mL). The radiotracer was eluted from the Sep-Pak with EtOH (1.5 mL) and collected in a sealed sterile vial containing 0.9 % NaCl_(aq) (13.5 mL) to give [¹⁸F]27 (107.4 mCi (4.0 GBq), 10.4% rcy decay-corrected). The solution was passed successively through a 1 μm filter and then a 0.2 μm filter (Acrodisc PTFE) under Ar_(g) pressure and collected in a sealed sterile dose vial (56.3 mCi (2.1 GBq), 54% recovery, decay-corrected; pH = 5.5). Radioactivity remaining on filters: 47.5 mCi (1.8 GBq). The total synthesis time was ~98 min from transfer of the H¹⁸F_(aq) to the CPCU. The radiotracer was analyzed by analytical HPLC (Waters Nova-Pak, 75:25:0.1 v/v/v MeOH/H₂O/NEt₃, 1 mL/min, t_R = 7.25 min) to determine the radiochemical purity (99.5%) and the specific activity (>6.4 × 10⁵ Ci/mol).

4.29. 3-(2,4-dichlorophenyl)-N-(4-[¹⁸F]fluorobutyl)-N,2,5-trimethylpyrazolo[1,5-a]pyrimidin-7-amine ([¹⁸F]28) – Two-Step Radiolabeling.

H¹⁸F_(aq) (1186 mCi (43.9 GBq) at EOB) in H₂¹⁸O was delivered to a CPCU, collected on a trap/release cartridge, released with K₂CO_3(aq) (0.9 mg in 0.6 mL H₂O), and added to a solution of Kryptofix 222 (5 mg in 1 mL CH₃CN) in the CPCU side-arm reaction vessel. The vessel was placed in a 110 °C oil bath and the solvent was evaporated with an Ar_(g) flow. CH₃CN (3.5 mL) was added and evaporated in order to azeotropically dry the Kryptofix 222/K¹⁸F. A solution of 35 (5.8 mg) in CH₃CN (1 mL) was added to the vessel containing the Kryptofix 222/K¹⁸F, the mixture was heated at 110 °C for 10 min, and then cooled. EtOEt (~5 mL) was added, the solution was transferred under Ar_(g) pressure out of the reaction vessel, through a Waters silica Sep-Pak Classic, and collected in a 15-mL conical tube that had been placed in an adjacent hot cell. The CPCU reaction vessel was rinsed with EtOEt (~5 mL) and this was transferred through the Sep-Pak and into the conical tube. The tube was placed in a warm water bath and the EtOEt was evaporated with an Ar_(g) flow until ~2.5 mL remained (705 mCi (26.1 GBq), 82.5% rcy, decay-corrected from EOB, ~52 min). The tube was placed in a 110 °C oil bath and the remainder of the EtOEt was evaporated with an Ar_(g) flow.

NaH (95%, 1.1 mg, 4.6 × 10⁻⁵ mol, 4.6 equiv.) was placed in a 1-mL vial, capped with a septum, purged with Ar_(g) for 90 min, and then cooled to 0 °C. Compound 16 (4.0 mg, 1.0 × 10⁻⁵ mol) was placed in a 1-mL vial, capped with a septum, purged with Ar_(g) for 90 min, dissolved in DMA (0.3 mL), added to the cold NaH, and transferred to the hot cell. This DMA solution then stood at ambient temperature in the hot cell for ~10 – 15 min during the transfer of [¹⁸F]fluorobutyltosylate from the CPCU and evaporation of the EtOEt. The DMA solution was added to the conical tube containing [¹⁸F]fluorobutyltosylate, the mixture was heated at 110 °C for 5 min, and then cooled at 0 °C for 1 min (630 mCi (23.3 GBq)). HPLC solvent (0.5 mL) was added, the solution was drawn into a 10-mL HPLC loop, the tube was rinsed with HPLC solvent (0.5 mL, loaded onto the loop), rinsed with MeOH (0.5 mL, loaded onto the loop), and then the mixture was purified by semi-preparative HPLC (Waters XTerra + guard pack 19 × 10 mm, 70:30:0.1 v/v/v MeOH/H₂O/NEt₃, 9 mL/min, t_R (range) = 15 – 18 min). The desired fractions were combined, diluted 1:1 v/v with H₂O, loaded onto a Waters tC₁₈ Sep-Pak, and rinsed with 0.9 % NaCl_(aq) (15 mL). The radiotracer was eluted from the Sep-Pak with EtOH (1.5 mL) and collected in a sealed sterile vial containing 0.9 % NaCl_(aq) (13.5 mL) to give [¹⁸F]28 (65.8 mCi (2.4 GBq), 11.1% rcy, decay-corrected from EOB, ~110 min; 13.5% rcy, decay-corrected from measurement of [¹⁸F]fluorobutyltosylate, ~58 min). The solution was passed successively through a 1 μm filter and then a 0.2 μm filter (Acrodisc PTFE) under Ar_(g) pressure and collected in a sealed sterile dose vial (50.7 mCi (3.9 GBq), 81.6% recovery, decay-corrected; pH = 7.5). Radioactivity remaining on filters: 10.1 mCi (0.37 GBq). The total synthesis time was ~120 min from EOB. The radiotracer was analyzed by analytical HPLC (Waters Nova-Pak, 75:25:0.1 v/v/v MeOH/H₂O/NEt₃, 1 mL/min, t_R = 6.0 min) to determine the radiochemical purity (100%) and the specific activity (>9.9 × 10⁵ Ci/mol).

4.30. 3-(2,4-dichlorophenyl)-N-(4-[¹⁸F]fluorobutyl)-N,2,5-trimethylpyrazolo[1,5-a]pyrimidin-7-amine ([¹⁸F]28) – One-Step Radiolabeling

H¹⁸F_(aq) (1699 mCi (62.9 GBq)) in H₂¹⁸O was obtained from PETNET and collected in a V-vial housed in a Capintec detector. Using Ar_(g) pressure 1614 mCi (59.7 GBq) was delivered to a CPCU, collected on a trap/release cartridge, released with K₂CO_3(aq) (0.9 mg in 0.6 mL H₂O), and added to a solution of Kryptofix 222 (5 mg in 1 mL CH₃CN) in the CPCU side-arm reaction vessel. The vessel was placed in a 110 °C oil bath and the solvent was evaporated with an Ar_(g) flow. CH₃CN (3.5 mL) was added and evaporated in order to azeotropically dry the Kryptofix 222/K¹⁸F. A solution of 31 (4.7 mg) in CH₃CN (1 mL) was added to the vessel containing the Kryptofix 222/K¹⁸F, the mixture was heated at 110 °C for 15 min, and then cooled. HPLC solvent (3 mL) was added to the vessel and the solution was transferred under Ar_(g) pressure to a hot cell and collected in a conical tube. The CPCU vessel was rinsed with HPLC solvent (3 mL) and this was added to the conical tube. The solution was drawn into a 10-mL HPLC injector loop and then purified by semi-preparative HPLC (Waters XTerra, 70:30:0.1 v/v/v MeOH/H₂O/NEt₃, 9 mL/min, t_R (range) = 11.5 – 13.5 min). The desired fractions were combined, diluted 1:1 v/v with H₂O, loaded onto a Waters tC₁₈ Sep-Pak, and rinsed with 0.9 % NaCl_(aq) (25 mL). The radiotracer was eluted from the Sep-Pak with EtOH (1.5 mL) and collected in a sealed sterile vial containing 0.9 % NaCl_(aq) (13.5 mL) to give [¹⁸F]28 (428 mCi (15.8 GBq), 43.9% rcy decay-corrected). The solution was passed successively through a 1 μm filter and then a 0.2 μm filter (Acrodisc PTFE) under Ar_(g) pressure and collected in a sealed sterile dose vial (106 mCi (3.9 GBq), 25.8% recovery, decay-corrected; pH = 5.3). Radioactivity remaining on filters: 294 mCi (10.9 GBq). The total synthesis time was ~85 min from transfer of the H¹⁸F_(aq) to the CPCU. The radiotracer was analyzed by analytical HPLC (Waters Nova-Pak, 75:25:0.1 v/v/v MeOH/H₂O/NEt₃, 1 mL/min, t_R = 6.6 min) to determine the radiochemical purity (95.5%) and the specific activity (>1.1 × 10⁶ Ci/mol).

4.31. 3-(2,4-dichlorophenyl)-N-(4-[¹⁸F]fluorobutyl-1,1,2,2,3,3,4,4-d₈)-N,2,5-trimethylpyrazolo[1,5-a]pyrimidin-7-amine ([¹⁸F]28-d₈)

H¹⁸F_(aq) (1886 mCi (69.8 GBq)) in H₂¹⁸₂O was obtained from PETNET and collected in a V-vial housed in a Capintec detector. The radioactivity sample was divided and 892 mCi (33.0 GBq) was delivered under Ar_(g) pressure to a CPCU, collected on a trap/release cartridge, released with K₂CO_3(aq) (0.9 mg in 0.6 mL H₂O), and added to a solution of Kryptofix 222 (5 mg in 1 mL CH₃CN) in the CPCU side-arm reaction vessel. The vessel was placed in a 110 °C oil bath and the solvent was evaporated with an Ar_(g) flow. CH₃CN (3.5 mL) was added and evaporated in order to azeotropically dry the Kryptofix 222/K¹⁸F. A solution of 31-d₈ (4.8 mg) in CH₃CN (1 mL) was added to the vessel containing the Kryptofix 222/K¹⁸F, the mixture was heated at 110 °C for 15 min, and then cooled. HPLC solvent (3 mL) was added to the vessel and the solution was transferred under Ar_(g) pressure to a hot cell and collected in a conical tube. The CPCU vessel was rinsed with HPLC solvent (3 mL) and this was added to the conical tube. The solution was drawn into a 10-mL HPLC injector loop and then purified by semi-preparative HPLC (Waters XTerra, 55:45:0.1 v/v/v EtOH/H₂O/NEt₃, 9 mL/min, t_R (range) = 11 – 14 min). The desired fractions were combined, diluted 1:1 v/v with H₂O, loaded onto a Waters tC₁₈ Sep-Pak, and rinsed with 0.9 % NaCl_(aq) (25 mL). The radiotracer was eluted from the Sep-Pak with EtOH (1.5 mL) and collected in a sealed sterile vial containing 0.9 % NaCl_(aq) (13.5 mL) to give [¹⁸F]28-d₈ (208 mCi (7.7 GBq), 39.4% rcy decay-corrected). The solution was passed successively through a 1 μm filter and then a 0.2 μm filter (Acrodisc PTFE) under Ar_(g) pressure and collected in a sealed sterile dose vial (36.2 mCi (1.3 GBq), 18.1% recovery, decay-corrected; pH = 5.0). Radioactivity remaining on filters: 160.7 mCi (5.9 GBq). The total synthesis time was ~90 min from transfer of the H¹⁸F_(aq) to the CPCU. The radiotracer was analyzed by analytical HPLC (Waters Nova-Pak, 75:25:0.1 v/v/v MeOH/H₂O/NEt₃, 1 mL/min, t_R = 8.8 min) to determine the radiochemical purity (98.4%) and the specific activity (>5.4 × 10⁵ Ci/mol).

4.32. 3-(2,4-dichlorophenyl)-N-ethyl-N-(4-[¹⁸F]fluorobutyl)-2,5-dimethylpyrazolo[1,5-a]pyrimidin-7-amine ([¹⁸F]29)

H¹⁸F_(aq) (2.01 Ci (74.4 GBq)) in H₂¹⁸O was obtained from PETNET and collected in a V-vial housed in a Capintec detector. The radioactivity sample was divided and 840 mCi (31.1 GBq) was delivered under Ar_(g) pressure to a CPCU, collected on a trap/release cartridge, released with K₂CO_3(aq) (0.9 mg in 0.6 mL H₂O), and added to a solution of Kryptofix 222 (5 mg in 1 mL CH₃CN) in the CPCU side-arm reaction vessel. The vessel was placed in a 110 °C oil bath and the solvent was evaporated with an Ar_(g) flow. CH₃CN (3.5 mL) was added and evaporated in order to azeotropically dry the Kryptofix 222/K¹⁸F. A solution of 32 (3.6 mg) in CH₃CN (1 mL) was added to the vessel containing the Kryptofix 222/K¹⁸F, the mixture was heated at 110 °C for 15 min, and then cooled. HPLC solvent (3 mL) was added to the vessel and the solution was transferred under Ar_(g) pressure to a hot cell and collected in a conical tube. The CPCU vessel was rinsed with HPLC solvent (3 mL) and this was added to the conical tube. The solution was drawn into a 10-mL HPLC injector loop and then purified by semi-preparative HPLC (Waters XTerra, 70:30:0.1 v/v/v MeOH/H₂O/NEt₃, 9 mL/min, t_R (range) = 21 – 24 min). The desired fractions were combined, diluted 1:1 v/v with H₂O, loaded onto a Waters tC₁₈ Sep-Pak, and rinsed with 0.9 % NaCl_(aq) (25 mL). The radiotracer was eluted from the Sep-Pak with EtOH (1.5 mL) and collected in a sealed sterile vial containing 0.9 % NaCl_(aq) (13.5 mL) to give [¹⁸F]29 (178 mCi (6.6 GBq), 39% rcy decay-corrected). The solution was passed successively through a 1 μm filter and then a 0.2 μm filter (Acrodisc PTFE) under Ar_(g) pressure and collected in a sealed sterile dose vial (22.4 mCi (0.8 GBq), 13% recovery, decay-corrected; pH = 5.3). Radioactivity remaining on filters: 151 mCi (5.6 GBq). The total synthesis time was ~107 min from transfer of the H¹⁸F_(aq) to the CPCU. The radiotracer was analyzed by analytical HPLC (Waters Nova-Pak, 75:25:0.1 v/v/v MeOH/H₂O/NEt₃, 1 mL/min, t_R = 11.3 min) to determine the radiochemical purity (97.3%) and the specific activity (>2.8 × 10⁵ Ci/mol).

4.33. LogP_7.4Procedure

The logP_7.4 value of each radiotracer was measured by adapting the procedure of Wilson and Houle.^{104, 105} 1-Octanol (10 mL) and 0.02 M/pH 7.4 phosphate buffer (5 mL) were placed in a 60-mL separatory funnel and briefly shaken. Radiotracer (~75 – 125 μCi, formulated in 10% EtOH/saline) was added, the mixture was shaken by hand for 1 min, allowed to stand for 5 min, and then the aqueous layer was separated and discarded. Aliquots of the 1-octanol layer (2 mL) were pipetted into each of four 15-mL Corning polypropylene centrifuge tubes which contained phosphate buffer (2 mL). The tubes were shaken mechanically for 10 min and then centrifuged for 5 min. Aliquots (1 mL) of each layer were pipetted into separate glass culture tubes and counted on a gamma counter. Each culture tube was counted 3x (decay-corrected) and the average value for each culture tube was calculated using Microsoft Excel. The logP_7.4 value for each of the four centrifuge tubes was calculated as logP_7.4 = log (average culture tube radioactivity counts in octanol/average culture tube radioactivity counts in phosphate buffer). The average value and standard deviation of the four centrifuge tubes were calculated using Microsoft Excel.

4.34. CRF Receptor Binding Assays

4.34.1 Materials

[¹²⁵I]-Tyr⁰-sauvagine and [¹²⁵I]antisauvagine-30 radiotracers were purchased from Perkin Elmer (San Jose, California). Unlabeled sauvagine and antisauvagine-30 were purchased from the American Peptide Company (San Diego, CA).

4.34.2 Cell Culture and Membrane Preparation

Plasmids for human CRF₁ receptors were a generous gift from Dr. Bryan Roth (University of North Carolina, Chapel Hill, NC). Plasmids were sequenced to verify gene delivery of the desired receptor. HEK293T cells (ATCC, Manassas, VA) were grown in culture media (DMEM supplemented with 4.5 g/L glucose, L-glutamine, sodium pyruvate, 10% fetal bovine serum, 1% penicillin and streptavidin) in T-75 cm² sterile culture flasks to 75% confluency. Cells were transiently transfected by Fugene 6 lipofection (Roche, Mannheim, Germany). A negative control transfection with pUC19 DNA (Invitrogen, Carlsbad, CA) was used to assure a lack of endogenous receptor binding. The media was aspirated 24 hours after transfection and replaced with fresh media free of lipofectamine. The media was again aspirated 48 hours after transfection and the cell monolayer was washed once with sterile PBS. Cells were dislodged by triturating with 30% PBS, 0.53 mM EDTA and 70% culture media. Suspended cells were centrifuged at 200 × g for 5 minutes at 23 °C. The supernatant was aspirated and the cell pellets were frozen at −20 °C until the day of the assay.

On the day of the assay, cell pellets were washed once in binding assay buffer (55 mM HEPES, 10 mM MgCl₂, 2.2 mM EGTA, 0.1% bovine serum albumin, 1μg/mL aprotinin, pH 7.2). Pellets were homogenized in binding assay buffer with a polytron homogenizer on a speed setting of 1 for 30 seconds (Kinematica, Lucerne, Switzerland). A radioligand binding assay of the membrane homogenate was conducted to determine total binding and nonspecific binding with unlabeled sauvagine. The membrane homogenate was titrated based on the preliminary assay so that total binding would approximate 3,000 CPM per reaction.

4.34.3 Radioligand Binding Assays

Sauvagine was dissolved in degassed and sonicated 10 mM acetic acid. Sauvagine stocks (100 μM) were aliquoted and frozen at −80 °C. Compounds were dissolved in DMSO (1 – 3 μg/μL) and serially-diluted with binding assay buffer in silanized glass tubes. Radiotracers were diluted in binding assay buffer and counted on a Wallac LKB gamma counter (Perkin Elmer, Waltham, MA). Radiotracer concentrations were titrated based on gamma counts so radioligand assays would use approximately 80 pM of [¹²⁵I]sauvagine or 80 pM of [¹²⁵I]antisauvagine-30.

Binding assay buffer (250 μL), unlabeled ligand over a final concentration range of 10⁻¹³ – 10⁻⁵ M (50 μL), radiotracer (100 μL of [¹²⁵I]sauvagine or [¹²⁵I]antisauvagine-30), and membrane homogenate (100 μL) were added sequentially to a 96 deep well plate (Fisher Scientific, Hampton, NH). Plates were vortexed and gently centrifuged at 200 × g for 5 min at 23 °C to bring all liquids together in the well. Plates were incubated for two hours at 23 °C. During the incubation, 96-position FilterMat B filters were soaked in 0.1% polyethyleneamine. Binding reactions were terminated by adding 1 mL of 4 °C PBS and 0.01% Triton X-100. Membranes were filtered over FilterMat B filters with a 96-well FilterMate Universal Harvester (Perkin Elmer, Waltham, MA). Filters were washed 5 times with 1 mL of 4 °C PBS and 0.01% Triton X-100. Filters were dried overnight at 23 °C, placed in plastic bags with a MeltiLex solid scintillator, and melted at 70 °C for 30 minutes. Filters were then counted on a MicroBeta TriLux Microplate Scintillation Counter (Perkin Elmer, Waltham, MA). The counting efficiencies of the MicroBeta detector and LKB gamma counter are 49% and 42%, respectively. Nonspecific binding (~5 – 8% of total binding) was defined as 80 pM of [¹²⁵I]sauvagine or 80 pM of [¹²⁵I]antisauvagine-30 in the presence of 1 μM of sauvagine. Assays were performed in triplicate and independent experiments were performed for statistical replicates.

4.34.4 Data Analysis

Equilibrium dissociation constants were determined with a one-site competition nonlinear regression curve analysis in GraphPad Prism 4.0 (GraphPad Software, La Jolla, CA). The concentration of labeled ligand was set at 80 pM and the K_d was set at 0.5 nM.

4.35. MicroPET

Animal imaging procedures were approved by the Emory University IACUC. Non-human primate handling was performed by the Yerkes National Primate Research Center veterinary staff.

4.35.1. Rat MicroPET Imaging

MicroPET imaging was performed in a male Sprague-Dawley rat on a Siemens Inveon microPET/CT scanner. The rat was initially anesthetized with 3 – 4% isoflurane in an anesthesia chamber and then maintained with 1 – 1.5% isoflurane via nose cone. The rat was placed in the prone position on the microPET scanner bed and body temperature was maintained with a warm air-circulating heating pad. A transmission scan was performed with a Co-57 point source for attenuation correction. The radiotracer was administered via tail vein injection and the rat was imaged dynamically for 60 min.

4.35.2. Non-human Primate MicroPET Imaging

MicroPET imaging was performed in a male cynomolgus monkey using a Siemens MicroPET Focus 220. The monkey was fasted for 12 hours prior to receiving anesthesia. The monkey was initially anesthetized with 3 – 4 mg/kg Telazol, intubated, and then maintained with 1 – 2% isoflurane. Body temperature was maintained with a warming blanket. Respiratory rate, oxygen saturation, expired carbon dioxide, heart rate, blood pressure, and temperature were continuously monitored using a SurgiVet monitor. A transmission scan was performed with a Co-57 point source for attenuation correction. The radiotracer was administered via saphenous vein injection and imaging was performed dynamically for 120 min.

Abbreviations

CRF

corticotropin-releasing factor

CRF₁

CRF type-1 receptor ¹

PET

positron emission tomography

SPECT

single-photon emission computed tomography

BBB

blood-brain barrier

DMA

dimethylacetamide

PEG

polyethylene glycol

SUV

standard uptake value ^2-4

TACs

time-activity curves

ROIs

regions of interest

CPCU

chemical processing control unit

rcy

radiochemical yield

EOB

end-of-bombardment