Merging the Structural Motifs of Functionalized Amino Acids and α-Aminoamides: Compounds with Significant Anticonvulsant Activities

Abstract

Functional amino acids (FAAs) and α-aminoamides (AAAs) are two classes of antiepileptic drugs (AEDs) that exhibit pronounced anticonvulsant activities. We combined key structural pharmacophores present in FAAs and AAAs to generate a new series of compounds and document that select compounds exhibit activity superior to either the prototypical FAA (lacosamide) or the prototypical AAA (safinamide) in the maximal electroshock (MES) seizure model in rats. A representative compound, (R)-N-4′-((3″-fluoro)benzyloxy)benzyl 2-acetamido-3-methoxypropionamide ((R)-10), was tested in the MES (mice, ip), MES (rat, po), psychomotor 6 Hz (32 mA) (mice, ip), and hippocampal kindled (rat, ip) seizure tests providing excellent protection with ED₅₀ values of 13, 14, ~10 mg/kg, and 12 mg/kg, respectively. In the rat sciatic nerve ligation model (ip), (R)-10 (12 mg/kg) provided an 11.2-fold attenuation of mechanical allodynia. In the mouse biphasic formalin pain model (ip), (R)-10 (15 mg/kg) reduced pain responses in the acute and the chronic inflammatory phases.

Epilepsy is a chronic disorder, characterized by recurrent, unprovoked seizures.¹ A seizure is defined as a discrete clinical event arising from transient, hypersynchronous, abnormal neuronal behavior. Epilepsy, then, is not a disease but rather a syndrome arising from a group of nonspecific, dysfunctional events in the brain. The treatment mainstay for patients with epileptic disorders has been the long-term and consistent administration of anticonvulsant drugs.^2,3 There are more than 40 pharmacologic therapies used for the treatment of epilepsy.⁴ Unfortunately, even when used optimally, these therapeutic interventions are ineffective for some 30% of patients. ⁵ Moreover, their use is associated, in more than 40% of patients, with untoward effects (e.g., drowsiness, dizziness, nausea, liver damage).⁶ The shortcomings of current regimens highlight the need for new, more effective agents.

We have previously reported that functionalized amino acids (FAAs,^a 1)^7–18 exhibit excellent anticonvulsant activities in various animal seizure models. Whole-animal pharmacological studies for 1 showed a unique profile, which indicated a novel mechanism of action. ^7–19 Similarly, studies have demonstrated that α-aminoamides (AAAs, 2) provide superb seizure protection. ^20,21 Representative examples for each class of compounds have advanced through clinical trials. Lacosamide ((R)-3),¹⁶ the 1 prototype, is a first-in-class antiepileptic drug (AED) that was recently introduced in the United States and Europe for adjuvant treatment of partial-onset seizures in adults.²² Safinamide ((S)-4) is a leading representative for 2. (S)-4 exhibited excellent protection in seizure models, and positive responses have been reported in recent phase III human clinical trials for the treatment of Parkinson disorders.^21,23,24

Recent electrophysiology studies using cultured rat cortical neurons demonstrated that (R)-3 selectively enhanced sodium channel slow inactivation in a time- and voltage-dependent manner, without affecting fast inactivation.²⁵ Similarly, examination of (R)-3 with recombinant Na_V 1.3 and 1.7 voltage-gated sodium channels expressed in HEK293 cells and of Na_V1.8-type TTX-R currents from DRG neurons showed that (R)-3 selectively modulated the slow inactivation state in each of these sodium channel subtypes.²⁶ (R)-3 is the only reported antiepileptic agent that selectively enhances slow inactivation without apparent interaction with fast inactivation gating.

Patch-clamp, whole-cell electrophysiological studies using hippocampal neurons demonstrated that (S)-4 inhibited tetrodotoxin-sensitive (TTX-S) fast Na⁺ currents in a concentration-dependent manner.^21a The inhibition was voltage dependent, showing an IC₅₀ of ~100 uM when currents were stimulated from a resting condition, while stronger inhibition (IC₅₀ = 33 uM) was observed when sodium currents were stimulated from a −60 mV depolarized membrane potential. Together, these findings indicated that (S)-4 exhibited a higher affinity for the sodium channel’s inactivated state.²⁴ Thus, the mechanisms associated with sodium channel inhibition for (R)-3 and (S)-4 are different.

At first glance, (R)-3 and (S)-4 appear structurally similar. Both have low molecular weights ((R)-3: MW =250; (S)-4: MW = 302 [free base]), each has a vicinal diamine backbone that contains a carbonyl (C=O) moiety, and each has one chiral center. In addition, both compounds contain an N-benzyl (PhCH₂)–type substituent. Pharmacologically, both (R)-3 and (S)-4 exhibited excellent seizure protection^16,19,27,28 in the maximal electroshock seizure (MES) animal model,²⁹ and electrophysiology studies demonstrated that both modulate sodium currents.^21a,25,26 Further inspection of (R)-3 and (S)-4 revealed stark differences in structure and function. First, (R)-3 is neutral and (S)-4 is basic. Second, the (R)-3 sequence of atoms attaches the N-benzyl moiety to an amide while in (S)-4 the substituted N-benzyl moiety attaches to an amine. Third, there is a different structure-activity relationship (SAR) in the MES test between the two classes of antiepileptic agents 1 and 2. For the 1 compounds, we observed a steady improvement in activity as the C(2) position was changed from H and CH₂OH to CH₃ to C₆H₅ to CH₂OCH₃.^8,16 Correspondingly, for 2, activity in the MES test increased as the C(2) position was changed from C₆H₅ to CH₂OH to H to CH₃.²⁰ Fourth, (R)-3 showed chiral specificity for function in the mouse (ip) (i.e., MES activity of (R)-3 vs (S)-3 is >22:1),¹⁶ but (S)-4 did not (i.e., MES activity of (S)-4 ≈ (R)-4).^20,30 Fifth, (S)-4 administration provided protection against several chemoconvulsants²⁷ while (R)-3 did not.¹⁹ Finally, (R)-3 enhanced the slow inactivation state of the voltage-gated sodium channel thereby selectively blocking the activity of chronically depolarized neurons^25,26 while (S)-4 inhibited TTX-S fast sodium currents.^21a

The structural differences between (R)-3 and (S)-4 likely account for their different modes of action and underscore that each compound has distinctive pharmacophores that contribute to drug function. In this study, we asked whether key structural units in 1 and 2 could be incorporated within a single compound to provide more effective anticonvulsant agents.³¹ Herein, we report the design, synthesis, and pharmacological evaluation of a series of compounds that conform to the general structure of I, and we document that they display excellent anticonvulsant activities.

RESULTS

Compound Design

In generating I, we used the generalized structure II for 1 and III for 2 (Figure 1). Both II and III include an “X-Y-Z” substituent at the N-benzyl 4′ site. “X-Y-Z” is a molecular unit 1–3 atoms long, or in the case I and III, it could also be a single bond. Our recent SAR study for the N-benzyl 4′ site in (R)-3 showed that the introduction of select substituents at this position provided compounds with excellent anticonvulsant properties.¹⁸ Similarly, the reported SAR for 2 demonstrated that significant anticonvulsant activity was observed with various linkers (“X-Y-Z”) at the N-benzyl position that bridged the two aromatic moieties in this agent.²⁰ The area of pharmacophore overlap in I is portrayed in the box, and this overlap permitted the incorporation of key pharmacophores found in II (1) and III (2) within a unified structure (Figure 1). For convenience, we divided I into three sectors: A, B, and C. Sector A is the structural motif seen in II (1), sector B is the linker unit “X-Y-Z” reported in both II (1) and III (2), and sector C is the terminal aromatic ring (Ar) found in III (2).

Figure 1.

Overlay of Pharmacophores in I

Choice of Compounds

Tables 1–3 lists the I compounds evaluated in this study. We prepared 20 compounds (5–22) in which we varied the structural units in sectors A–C and the chirality of the C(2) center in sector A. Since the number of structural permutations was large, we maintained the structural pattern constancy for two sectors as we changed the other.

Table 1.

Novel Neurological Agents: Structure-Activity Relationship of Sector A^a

Cpd No	R	Stereo	Mice (ip)b				Rat (po)f
			MES,c ED50	6Hz, ED50	Tox,d TD50	PIe	MES,c ED50	Tox,g TD50	PIe
5	H	-	>100, <300 [4.0] >30, <100 [2]		>300 [0.5 and 4.0]		31 [1.0] (18–53)	>500 [1.0]	>16
(R)-6	(R)-Me	R	>30, <100 [0.5 and 4]		>300 [0.5]		31 [4.0] (21–44)	>500 [4.0]	>16
(S)-6	(S)-Me	S	>100,< 300 [0.5] > 300 [4.0]		>300 [0.5]		> 30 [0.25 to 4.0]	> 30 [0.25 to 4.0]
(R)-7	(R)-i-Pr	R	>300 [0.5 and 4]		>300 [0.5 and 4]		>30	>30
(R)-8	(R)-t-Bu	R	>300 [0.5 and 4]		>300 [0.5 and 4]
(R,S)-9		R, S	28 [0.5] (20–36)		213 [2.0] (156–285)	7.6	>30 [4]	>30 [4]
(R)-10	(R)-CH2OMe	R	13 [0.25] (11–16)	~10 [0.25]	26 [0.5] (21–34)		14 [0.5] (6.1–27)	>500 [0.5]	>36
(S)-10	(S)-CH2OMe	S	>300		>300
(S)-4h			8.0 (7.0–9.1)		630 (560–700)
(R)-4h			7.2 (5.9–8.9)		580 (410–830)
(R)-3i			4.5 [0.5] (3.7–5.5)		27 [0.25] (26–28)		3.9 [2.0] (2.9–6.2)	>500
phenytoinj			9.5 [2.0] (8.1–10)		66 [2.0] (53–72)	6.9	30 [4.0] (22–39)		> 100
phenobarbitalj			22 [1.0] (15–23)		69 [0.5] (63–73)	3.2	9.1 [5.0] (7.6–12)	61 [0.5] (44–96)	6.7
valproatej			270 [0.25] (250–340)		430 [0.25] (370–450)	1.6	490 [0.5] (350–730)	280 [0.5] (190–350)	0.6

^aThe compounds were tested through the auspices of the NINDS ASP.

^bThe compounds were administered intraperitoneally. ED₅₀ and TD₅₀ values are in milligrams per kilogram.

^cMES = maximal electroshock seizure test.

^dTD₅₀ value determined from the rotorod test.

^ePI = protective index (TD₅₀/ED₅₀).

^fThe compounds were administered orally. ED₅₀ and TD₅₀ values are in milligrams per kilogram.

^gTox = behavioral toxicity.

^hRef ³⁰.

ⁱRef ¹⁶.

^jRef ³².

Table 3.

Novel Neurological Agents: Structure-Activity Relationship of Sector C^a

Cpd No	R′	Mice (ip)b				Rat (po)f
		MES,c ED50	6Hz, ED50	Tox,d TD50	PIe	MES,c ED50	Tox,g TD50	PIe
(R)-20	-	<3 [0.5]		>10, <30 [0.5]		<10 [0.25 to 2.0]	>10 [0.25 to 2.0]
(R)-21	2-F	>3, <10 [0.5]		>30, <100 [0.5]		11 [0.5] (7.9–13)	>500	>45
(R)-10	3-F	13 [0.25] (11–16)	~10 [0.25]	26 [0.5] (21–34)	2	14 [0.5] (6.1–27)	>500 [0.5]	>36
(R)-22	4-F	>10, < 30 [0.5]		>30, < 100 [0.5]		5.8 [0.5] (4.3–7.3)	>500	>86

^aThe compounds were tested through the auspices of the NINDS ASP.

^bThe compounds were administered intraperitoneally. ED₅₀ and TD₅₀ values are in milligrams per kilogram.

^cMES = maximal electroshock seizure test.

^dTD₅₀ value determined from the rotorod test.

^ePI = protective index (TD₅₀/ED₅₀).

^fThe compounds were administered orally. ED₅₀ and TD₅₀ values are in milligrams per kilogram.

^gTox = behavioral toxicity.

^hRef ³⁰.

ⁱRef ¹⁶.

^jRef ³².

For compounds listed in Table 1, sector A was varied while we restricted sectors B and C to an OCH₂ and 3-(fluoro)phenyl moiety, respectively, to match (S)-4. Specifically, in this set of compounds we evaluated key structural features important to the 1 SAR. We showed that 1 compounds containing a small R substituent provided excellent seizure protection in the MES test and that anticonvulsant activity typically improved when a substituted heteroatom was introduced one atom removed from the C(2) center. Moreover, anticonvulsant activity for the 1 compounds principally resided in the D-configuration.^{9,10,11,16,18} Thus, we progressively increased the R substituent in 5–8 from hydrogen to methyl to isopropyl to tert-butyl and then included in our selection list 9 and 10, compounds that contained a substituted heteroatom positioned one atom removed from C(2). For compounds 6 and 10, we prepared the R- (D-configuration) and the S- (L-configuration) stereoisomers to determine whether the anticonvulsant activities for I mirrored those found for 1, in which a clear stereochemical preference was observed,^9–11,16,18 or that for 2, in which both stereoisomers displayed comparable anticonvulsant activities.^20,30 For all compounds listed in Table 1, except 5 and 9, we prepared I as a single stereoisomer. Compound 5 had no chiral center, and the synthetic route for 9 provided the racemic mixture.

For compounds 10–19 listed in Table 2, we varied sector B’s “X-Y-Z” linker region and employed the (R)-3 structural motif for sector A and the (S)-4 3-(fluoro)phenyl unit for sector C. Following the SAR reported for (S)-4,²⁰ we incorporated one, two, and three atom linkers and varied the heteroatom content and degree of unsaturation of the linker. The choice of the “X-Y-Z” unit was also consistent with the SAR reported for (R)-3,¹⁸ in which we showed that incorporating select substituents (e.g., alkyl, substituted alkyl, vinyl, acetylenic, aryl) at the 4′ position of the N-benzyl ring gave compounds with superb seizure protection in the MES test²⁹ in rodents with some analogs having activities comparable with (R)-3 and established AEDs.³²

Table 2.

Novel Neurological Agents: Structure-Activity Relationship of Sector B^a

Cpd No	-X-Y-Z-	Mice (ip)b				Rat (po)f
		MES,c ED50	6Hz, ED50	Tox,d TD50	PIe	MES,c ED50	Tox,g TD50	PIe
(R)-11	-	>10, <30 [0.5]	<30 [0.5]	>100, <300 [0.5]		2.4 [1.0] (1–3.9)	>500	>250
(R)-12	-O-	<30 [0.5]		>30, <100 [0.5]		<10 [0.25–2.0]	>10 [0.25–4.0]
(R)-13	-(CH2)2-	>10, <30 [0.5]		>30, <100 [0.5]		<30 [1]	>30 [1]
(R)-14	-CH=CH-	>30, <100 [0.5]	< 30 [0.25]	>100, <300 [0.5]		~30 [1.0, 4.0]	>30 [0.25, 4.0]
(R)-15	-≡-	>30, <100 [1.0, 4.0]		>100, <300 [4.0]		1.4 [4.0] (0.7–2.2)	>63, <125 [4]
(R)-16	-CH2O-	5.9 [0.25] (4.3–7.3)		10 [0.25] (9.1–13)	1.8	19 [2] (13–25)	>400 [0.5]	>21
(R)-17	-N(H)CH2-	>10, <30 [0.5]		>30, <100 [0.5]
(R)-10	-OCH2-	13 [0.25] (11–16)	~10 [0.25]	26 [0.5] (21–34)	2	14 [0.5] (6.1–27)	>500 [0.5]	>36
(R)-18	-CH2OCH2-	>30, <100 [0.5]		>30, <100 [0.5]
(R)-19	-OCH2CH2-	>30, <100 [0.5, 4.0]		>30, <100 [0.5]

^aThe compounds were tested through the auspices of the NINDS ASP.

^bThe compounds were administered intraperitoneally. ED₅₀ and TD₅₀ values are in milligrams per kilogram.

^cMES = maximal electroshock seizure test.

^dTD₅₀ value determined from the rotorod test.

^ePI = protective index (TD₅₀/ED₅₀).

^fThe compounds were administered orally. ED₅₀ and TD₅₀ values are in milligrams per kilogram.

^gTox = behavioral toxicity.

^hRef ³⁰.

ⁱRef ¹⁶.

^jRef ³².

Finally, we evaluated the effect of the terminal aromatic ring by preparing compounds (R)-10, and (R)-20–(R)-22 (Table 3). Here, we varied sector C to include the unsubstituted aromatic compound (R)-20 and the three monofluorine–substituted regioisomers (R)-10, (R)-21, and (R)-22 and set sector A to match (R)-3 and sector B to match (S)-4. We also prepared compound 23, which contains the sector C structural motif found in many of our compounds.

Chemistry

Two similar routes were employed to prepare most I compounds in this study and depended only on the C(2)-R substituent. For most I derivatives that incorporated the (R)-3 framework in sector A, we used a recently reported procedure to prepare 4′-substituted N-benzylamide (R)-3 analogs¹⁸ beginning with either tert-Boc–protected (R)- or (S)-serine 24. The acid was coupled with the desired benzylamine 25 using the mixed anhydride method (isobutylchloroformate (IBCF), N-methyl morpholine (NMM)),³³ unless otherwise indicated, to give the N-benzyl amide 26 without racemization of the C(2) chiral center (Scheme 1). Subsequent methylation of the serine hydroxy group (CH₃I, Ag₂O) gave ether 27. Deprotection of the tert-butoxycarbonyl group with acid followed by acetylation of the amine with acetyl chloride and triethylamine gave the desired product, I, in 43–80% yield. Using this method, we prepared (R)-10, (S)-10, and (R)-11–(R)-22.

Scheme 1.

General Procedure for the Preparation of (R)- and (S)-N-(4′-Substituted)benzyl 2-Acetamido-3-methoxypropionamide Derivatives I

A similar procedure was used to prepare compounds 5, (R)-6–(R)-8 and (S)-6 (Scheme 2). We began with a commercially available amino acid. Converting the amino acid to either the N-acetyl (28, (R)-29, (S)-29) or the N-tert-butoxycarbonyl ((R)-30, (R)-31) derivative permitted mixed anhydride coupling with benzylamine 32 to give the amides (5, (R)-6, (S)-6, (R)-33, (R)-34). For (R)-33 and (R)-34 that were protected as the tert-Boc derivatives, acid deprotection gave the amine, which was directly reacted with acetyl chloride and base to give (R)-7 and (R)-8, respectively.

Scheme 2.

General Procedure for the Preparation of (R)- and (S)-N-(4′-(3-fluorobenzyloxy)benzyl)benzyl 2-Acetamido-2-(substituted)-acetamide Derivatives 5–8

We developed a different route for the C(2) pyridyl derivative (R,S)-9 (Scheme 3). Beginning with commercially available ethyl 2-(pyridin-2yl)acetate (35), basic hydrolysis provided acid 36, which was coupled with 4-(((3′-fluoro)benzyloxy)phenyl)methanamine (32) using O-(benzotriazol-1-yl)-N,N,N′,N′-tetramethyluronium tetrafluoroborate (TBTU) to give 37. Treatment of 37 with NaNO₂ in acetic acid yielded the oximes 38 as an ~1:1 mixture of syn- and anti-isomers. While oximes 38 could be separated by silica gel chromatography, the mixture was reduced to amine 39 with Zn dust in the presence of ammonium formate and then converted to racemic 9 with acetyl chloride and triethylamine.

Scheme 3.

Preparation of N-(4-(3-Fluorobenzyloxy)benzyl) 2-Acetamido-2-(pyridin-2-yl)acetamide ((R,S)-9)

The extended N-benzyl amide moiety within the I compounds was a key unit in our compounds. Thus, we used a series of methods to construct this moiety (Schemes 4–6). For many of the compounds, we reacted either 4-cyanophenol (40) with a substituted benzyl bromide (41–44) or a (cyano)aryl bromide (52, 53) with the substituted phenol (54) or aryl-substituted alcohol (55) under base conditions to give the ether (45–48, 56–58) (Scheme 4). Subsequent reduction (LiAlH₄) of the nitrile in 45–48 and 56–58 gave the requisite benzylamine (32, 49–51, 59–61) for the mixed anhydride coupling reaction. We prepared nitrile 64 from 4-(hydroxy)benzonitrile (40) and 3-(fluoro)phenethanol (62) using Mitsunobu coupling conditions³⁴ and then reduced nitrile 63 to benzylamine 64 with LiAlH₄. Correspondingly, we reacted 4-(bromo)benzonitrile (52) with 3-(fluoro)benzylamine (65) under Buchwald-Hartwig coupling conditions³⁸ to generate 66, which was reduced to amine 67 (Scheme 5). Finally, (R)-4-(iodo)benzyl 2-acetamido-3-methoxypropionamide¹⁸ ((R)-69) served as the starting material for I compounds (R)-11 and (R)-13–(R)-15 (Scheme 6). Coupling (R)-69 with 3-(fluoro)phenylboronic acid (70) under Suzuki coupling conditions³⁶ gave (R)-11. When trans-2-((3′-fluoro)phenyl)vinylboronic acid (71) was substituted for 3-(fluoro)phenylboronic acid (70), we obtained (R)-14 with an embedded trans-double bond. Reduction of (R)-14 (10% Pd/C, H₂) gave (R)-13. Sonogashira coupling³⁷ of (R)-69 with 3-(fluoro)phenylacetylene (72) afforded (R)-15. For (R)-17 synthesis, we coupled amine 67 with (R)-2-acetamido-3-methoxypropanoic acid ¹⁷ ((R)-68) using 4-(4,6-dimethoxy-1,3,5-triazin-2-yl)-4-methoxymorpholinium chloride (DMTMM)³⁸ (Scheme 5).

Scheme 4.

Benzylamine 32, 49–51, 59–61, and 64 Synthesis

Scheme 6.

Preparation of Compounds (R)-11, (R)-13, (R)-14 and (R)-15

Scheme 5.

Benzylamine 67 and Compound (R)-17 Synthesis

The enantiopurity of (R)-6–(R)-8, (R)-10–(R)-22, (S)-6, and (S)-10 was assessed by the detection of a single acetyl methyl signal in the ¹H NMR spectrum for each compound when a saturated solution of (R)-(−)-mandelic acid was added.³⁹ In the cases of (R)-10-(R)-22, we also observed a single O-methyl peak upon addition of (R)-(−)-mandelic acid.

We report, in the Experimental Section, the details (synthetic procedure, characterization) of the final step for all the compounds evaluated in the seizure models. In Supporting Information, we provide a synthetic scheme for each compound tested and the experimental procedures used and physical and full spectroscopic properties for all the synthetic compounds prepared in this study.

Pharmacological Activity

Compounds 5, (R)-6–(R)-8, (R,S)-9, (R)-10–(R)-22, (S)-6, and (S)-10 were tested for anticonvulsant activity at the Anticonvulsant Screening Program (ASP), which is part of the National Institute of Neurological Disorders and Stroke (NINDS) at the U.S. National Institutes of Health. Screening was performed using the procedures described by Stables and Kupferberg.⁴⁰ The pharmacological data from the MES test ²⁹ are summarized in Tables 1–3 and similar results obtained for (R)-3,¹⁶ (R)-4,³⁰ (S)-4,³⁰ and the clinical antiepileptic drugs (AEDs) phenytoin,³² valproate,³² and phenobarbital³² are given in Table 1. All compounds were administered intraperitoneally (ip) to mice and orally (po) to rats. Tables 1–3 lists the values that were determined to be protective in blocking hind limb extension induced in the electrically induced MES seizure model from the rodent identification studies. For compounds that showed significant activity, we report the 50% effective dose (ED₅₀) values obtained in quantitative screening evaluations. Also provided are the median doses for 50% neurological impairment (TD₅₀) in mice, using the rotorod test,⁴¹ and the behavioral toxicity effects observed in rats.⁴² TD₅₀ values were determined for those compounds that exhibited significant activity in the MES test. The protective index (PI = TD₅₀/ED₅₀) for each of these analogs is also listed. Select compounds were also evaluated in the psychomotor 6 Hz (32 mA) seizure models (mice, ip).⁴³ When the I derivatives were evaluated in the subcutaneous Metrazol (scMet) seizure model⁴⁴ none provided protection at doses up to 300 mg/kg at two time points (0.5 and 4 h) (data not shown). The absence of seizure protection in this assay is a hallmark of FAA activity ^7–17 and contrasted with the data reported for (S)-4.²⁷

The SAR data for I provided distinctive trends. In Table 1, we varied the C(2) R group in sector A while maintaining the (S)-4 structural components found in sectors B and C. Using the mice (ip) data, we observed that as the size of the C(2) alkyl group increased from methyl ((R)-6, MES ED₅₀ = >30, <100 mg/kg) to isopropyl ((R)-7, MES ED₅₀ = >300 mg/kg) to tert-butyl ((R)-8, MES ED₅₀ = >300 mg/kg), the anticonvulsant activity decreased. We also found that MES seizure protection significantly increased upon the inclusion of a C(2)-substituted heteroatom group one atom removed from the C(2) center ((R,S)-9, MES ED₅₀ = 28 mg/kg; (R)-10, MES ED₅₀ = 13 mg/kg versus (R)-6, MES ED₅₀ = >30, <100 mg/kg; (R)-7, MES ED₅₀ = >300 mg/kg; (R)-8, MES ED₅₀ = >300 mg/kg). Similar SAR patterns were observed for 1 compounds.^7,11–14 Furthermore, for both 6 and 10, we tested the (R)- and (S)-enantiomers and found that the principal activity resided in the (R)-stereoisomer (D-configuration) (MES ED₅₀ (mice, ip): (R)-6, >30, <100 mg/kg vs (S)-6, >100, <300 mg/kg, and (R)-10, 13 mg/kg vs (S)-10, >300 mg/kg). Both the C(2) SAR pattern and the stereochemical preference for the (R)-enantiomer for seizure protection strongly indicated that anticonvulsant activity in the MES test for I compounds resembled the activity in 1 compounds. (R)-10 was the most active among the sector A compounds, exhibiting an MES ED₅₀ = 13 mg/kg. This placed (R)-10 midway between phenytoin and phenobarbital in activity³² and approximately three times less active than (R)-3.¹⁶ The protective index (PI = TD₅₀/ED₅₀) in mice (ip) for (R)-10 was 2, which was lower than (R)-3 (PI = 5.2).¹⁶ When sector A I compounds were evaluated in the rat (po), the effect of C(2) R substitution on MES seizure protection was less noticeable. In this model, we found that (R)-10 (MES ED₅₀ = 14 mg/kg) was more active than either 5 (MES ED₅₀ = 31 mg/kg) or (R)-6 (MES ED₅₀ = 31 mg/kg) but that the glycine derivative 5 and the alanine analog (R)-6 showed equal seizure protection. Interestingly, none of the three compounds displayed behavioral toxicity, even at doses up to 500 mg/kg. The activity observed for (R)-10 in the rat (po) exceeded that of phenytoin and phenobarbital.³²

Further SAR information was gathered by varying the sector B linker in I (Table 2). For these compounds, we maintained the (R)-3 framework for sector A and the (S)-4 (3-fluoro)phenyl substituent for sector C. We observed differences between the mice (ip) and rat (po) data sets. In the mice, anticonvulsant activity in the MES test decreased as the linker changed from CH₂O ((R)-16) [MES ED₅₀ = 5.9 mg/kg] to OCH₂ ((R)-10) [MES ED₅₀ = 13 mg/kg] to no linker ((R)-11), O ((R)-12), CH₂CH₂ ((R)-13), N(H)CH₂ ((R)-17) [MES ED₅₀ = >10, <30 mg/kg] to C(H)=C(H) ((R)-14), C≡C ((R)-15), CH₂OCH₂ ((R)-18), OCH₂CH₂ ((R)-19) [MES ED₅₀ = >30, <100 mg/kg]. Correspondingly, in the rat (po), we observed that the anticonvulsant activity decreased by going from C≡C ((R)-15) [MES ED₅₀ = 1.4 mg/kg], no linker ((R)-11) [MES ED₅₀ = 2.4 mg/kg], to O ((R)-12) [MES ED₅₀ = <10 mg/kg], OCH₂ ((R)-10) [MES ED₅₀ = 14 mg/kg], CH₂O ((R)-16) [MES ED₅₀ = 19 mg/kg], CH₂CH₂ ((R)-13) [MES ED₅₀ = <30 mg/kg] to C(H)=C(H) ((R)-14) [MES ED₅₀ = ~30 mg/kg]. We were interested to find that (R)-15 and (R)-11 exhibited superb seizure protection in the MES test (rat, po) with ED₅₀ values of 1.4 mg/kg and 2.4 mg/kg, respectively. Moreover, (R)-11 displayed no evidence of behavioral toxicity at 500 mg/kg. Earlier, we reported the anticonvulsant activity of (R)-N-(biphenyl-4-yl)methyl 2-acetamido-3-methoxypropionamide ((R)-73).¹⁸ (R)-73, like (R)-11, exhibited outstanding seizure protection in the MES test (rat, po: MES ED₅₀ = 2.0 mg/kg; TD₅₀ = 49 mg/kg) but, unlike (R)-11, was found to have noticeable neurotoxicity. The precise role of the 3′-fluoro group in (R)-11 in modulating behavioral neurotoxicity is unclear but is under investigation. Finally, for (R)-15 we not only observed excellent anticonvulsant protection but also the duration of seizure protection extended for the entire 4 h testing period. The activities of (R)-11 and (R)-15 exceeded (R)-3 (ED₅₀ = 3.8 mg/kg)¹⁶ and other AEDs.³²

Sector C’s terminal aryl group was the last structural unit evaluated (Table 3). For this series of compounds, we maintained the (R)-3 framework for sector A, and the (S)-4 OCH₂ unit for sector B since both moieties provided I compounds with excellent activity (Tables 1, 2). We chose to evaluate the unsubstituted aryl compound, (R)-20 and the three mono-fluorophenyl derivatives, (R)-10, (R)-21, and (R)-22. For the three fluorinated compounds, we observed in mice (ip) that the 2″-fluoro derivative (R)-21 was the most active (MES ED₅₀ = >3, <10 mg/kg). Correspondingly, in the rat (po) model, the 4″-fluoro isomer (R)-22 was the most potent (MES ED₅₀ = 5.8 mg/kg) followed by the 2″-fluoro ((R)-21, MES ED₅₀ = 11 mg/kg) and the 3″-fluoro ((R)-10, MES ED₅₀ = 14 mg/kg) analogs. Removing the fluoro substituent to give unsubstituted (R)-20 increased seizure protection (MES ED₅₀ = <3 mg/kg) in mice (ip), making this agent among the most potent I compounds tested under these conditions. Collectively, these findings demonstrated that the terminal aryl ring in I can influence anticonvulsant activities, thus warranting our further SAR exploration of this unit. We asked whether the extended N-aryl substituent in I directly contributed to the observed anticonvulsant activity. Accordingly, we evaluated 23 in the MES test (mice, ip) and observed modest anticonvulsant activity (MES ED₅₀ = >30, <100 mg/kg).

The excellent activity for (R)-10 warranted its further pharmacological evaluation in other models. In the psychomotor 6 Hz (32 mA) seizure test,⁴³ (R)-10 exhibited an ED₅₀ = ~10 mg/kg in mice (ip) (Table 1), a value comparable with that observed for (R)-3 (ED₅₀ = 10 mg/kg).¹⁹ In the sensitive rat (ip) hippocampal kindled seizure test,^44,45 its ED₅₀ value was 12 mg/kg. This closely matched the value for (R)-3 (ED₅₀ = 14 mg/kg) in similar screening.¹² The hippocampal kindled seizure assay is a model of partial complex seizures or temporal lobe seizures, which are the most common and drug-resistant type of adult focal epilepsy.^45a,46

Further evaluations of (R)-10 were undertaken using the sciatic nerve ligation model,⁴⁷ which is used to predict potential efficacy against chronic neuropathic pain in humans. Recent pharmacological and clinical studies have documented that certain anticonvulsant compounds are effective in several different models of inflammatory and neuropathic pain.⁴⁸ Thus, it is not surprising that these antiepileptic agents are also used to manage neuropathic pain. ^48,49 Neuropathic pain results from excessive neuronal activity and damage resulting in dysfunction of neuronal pathways within both the peripheral and the central nervous systems.^48,49 The sciatic ligation model was used to assess the efficacy of (R)-10 in rats. Using this test, we showed that administration of (R)-10 (12 mg/kg; rats, ip) provided a 11.2-fold attenuation of mechanical allodynia at 1 h.

The formalin test, a chemically induced biphasic pain model, was also employed.⁵⁰ There are two phases of response measured in the test, early (acute) and late (chronic inflammatory). The acute phase results in a behavioral licking response, which is believed to be mediated by chemical activation of local C-fibres.⁵⁰ The late phase is likely due to the development of peripheral inflammation and central sensitization of dorsal horn neurons. Thus, this model is believed to provide preliminary information about the utility of the test candidate for the treatment of acute and chronic inflammatory pain. When evaluated in the formalin model, (R)-10 (15 mg/kg, mice, ip) significantly reduced the pain response in both the acute (35% of control; p <0.01) and the late (44% of control; p <0.01) phases. These results compared favorably with (R)-3, where comparable reduction in pain was observed only in the late phase when 16 mg/kg was administered.⁵¹

In 5–22, we constructed compounds with the principal structural motif seen in 1 and then extended the N-benzyl amide moiety to resemble 2. We briefly explored the reverse, where we began with the 2 core structure and then incorporated at C(2) 1 units shown to have excellent anticonvulsant activities. Accordingly, we prepared racemic 74 and 75 using the synthetic procedures shown in Schemes 7 and 8, respectively. Compound 74 contained the (R)-3 (C)2-methoxymethylene unit, and compound 75 had a 2-furanyl moiety. In the 1 series, racemic 3 and 88 displayed MES ED₅₀ values (mice, ip) of 8.3 and 10 mg/kg, respectively (Table 4). When tested, (R,S)-74 and (R,S)-75 displayed good-to-moderate anticonvulsant activity in mice (ip) ((R,S)-74, MES ED₅₀ = 30–100 mg/kg; (R,S)-75, MES ED₅₀ = 10–30 mg/kg), values that were higher (lower activity) than observed for (R,S)-3 and (R,S)-88 in this test (MES ED₅₀ = 8.3–10 mg/kg).^11,16 Interestingly, we observed little or no activity in the scMet seizure model for (R,S)-74 (scMet ED₅₀ = >300 mg/kg) and (R,S)-75 (scMet ED₅₀ = 100–300 mg/kg). This finding contrasted with the excellent activity reported for (S)-4 in this model (scMet ED₅₀ = 27 mg/kg).²⁷ We concluded that the SAR guidelines for 1^7–16 did not help improve the pharmacological activity of 2.

Scheme 7.

Preparation of 74

Scheme 8.

Preparation of 75

Table 4.

Pharmacological Data for α-Aminoamide Derivatives 74 and 75 and their Functionalized Amino Acid Counterparts 3 and 88^a

No.	R4	Mice (ip)b
		MES,c ED50	Tox,d TD50	PI e
(R,S)-74		>30, <100 [0.5]	>100, <300 [0.5]
(R,S)-75		>10, <30 [0.5]	>100, <300 [0.5]
(R,S)-3f		8.3 [0.5] (7.9–9.8)	43 [0.25] (38–47)	5.2
(R,S)-88g		10 (9.1–12)	~40	~4.0
(S)-4h		8.0 (7.0–9.1)	630 (560–700)	78
(R)-4h		7.2 (5.9–8.9)	580 (410–830)	88
phenytoini		9.5 [2] (8.1–10)	66 [2] (53–72)	6.9
phenobarbitalli		22 [1] (15–23)	69 [0.5] (63–73)	3.2
valproatei		270 [0.25] (250–340)	430 [0.25] (370–450)	1.6

^aThe compounds were tested through the auspices of the NINDS ASP.

^bThe compounds were administered intraperitoneally. ED₅₀ and TD₅₀ values are in milligrams per kilogram.

^cMES = maximal electroshock seizure test.

^dTD₅₀ value determined from the rotorod test.

^ePI = protective index (TD₅₀/ED₅₀).

^fRef ¹⁶.

^gRef ¹¹.

^hRef ³⁰.

ⁱRef ³².

DISCUSSION

In this study, we combined key pharmacophores found in 1 and 2 to provide I, anticipating that I might exhibit potent anticonvulsant activity. The structural design for I permitted the near seamless overlap of the pharmacophores. Moreover, the structural economy gained from this overlap permitted the design of a compact agent that conforms to Lipinski’s pharmacokinetic rules for low molecular weight, oral therapeutic agents.⁵² We expected several additional benefits from this design. First, the overlap eliminated the need for extraneous bridging units, which could confound receptor binding and provide unwanted sites for metabolism and toxicity. Second, we anticipated that I, like 1 and 2, ^22,24 would exhibit favorable CNS biodistribution.

We found that I did not prevent scMet-induced seizures in mice and that it showed a stereochemical preference for function, which corresponds to the D-amino acid, in the MES-induced seizure test in rodents. Moreover, anticonvulsant activity increased with placement of a heteroatom one atom removed from the C(2) atom in I (Table 1) These whole animal pharmacological responses are similar to those observed for 1 ((R)-3)^7–18 but not 2 ((S)-4).^20,30 Table 2 shows that the most potent I derivatives in the MES test were those in which “X-Y-Z” was a single bond ((R)-11), a rigid, linear unit ((R)-15), or a short moiety (i.e., O ((R)-12); “X-Y” [where X and Y are C(H), N(H), O, and CH₂] ((R)-10, (R)-13, (R)-14, (R)-16, (R)-17). Of these, (R)-11 and (R)-15 were the most potent anticonvulsants when tested in the rat (po) (MES ED₅₀ = 1.4–2.4 mg/kg).

The excellent anticonvulsant activity for several I compounds demonstrated that seizure protection could be improved over their 1 counterpart by incorporating an extended N-aryl substituent similar to that found in 2 ((S)-4). For example, in the MES test (mice, ip), the glycine FAA 89 exhibited weak anticonvulsant activity (MES ED₅₀ = >100, <300 mg/kg),⁸ but glycine 5 exhibited a MES ED₅₀ = >30, <100 mg/kg, mice (ip). When 5 was tested in the rat (po), the MES ED₅₀ = 31 mg/kg, and we observed no apparent neurological toxicity at doses of 500 mg/kg. Similarly, we found superb anticonvulsant activity for (R)-11 (MES ED₅₀ = 2.4 mg/kg) and (R)-15 (MES ED₅₀ = 1.4 mg/kg) in the rat (po), which surpassed that of (R)-3 (MES ED₅₀ = 3.9 mg/kg). We evaluated 23 that contained the same extended N-benzyl substituent that was incorporated in 5 and (R)-6–(R)-10 and observed only modest anticonvulsant activity in mice (ip) (MES ED₅₀ = >30, <100 mg/kg), indicating that this group alone was not responsible for the observed seizure protection. Finally, we found that several I compounds ((R)-6, (R)-15) exhibited extended duration of action in the MES model in mice (ip) and rat (po).

We tested (R)-N-4′-((3″-fluoro)benzyloxy)benzyl 2-acetamido-3-methoxypropionamide ((R)-10) in other neurological models. We first evaluated (R)-10 in two additional seizure models. In the psychomotor 6 Hz (32 mA) test in mice (ip) ⁴³ and the rat (ip) hippocampal kindled seizure test,⁴⁵ (R)-10 provided excellent protection, giving an ED₅₀ of ~10 and 12 mg/kg, respectively. Since several clinically available AEDs are used to treat pain disorders,^48,49 we evaluated (R)-10 in two animal pain models. When (R)-10 was tested at 12 mg/kg in the sciatic nerve ligation model⁴⁷ in rats (ip) we observed an 11.2-fold attenuation of mechanical allodynia at 1 h. In the biphasic, chemically induced formalin pain model in mice,⁵⁰ ip administration of (R)-10 (15 mg/kg) reduced pain response in both the acute (35% of control) and the late (44% of control) phases. These findings document the potential value of I to treat a range of neurological conditions.

CONCLUSIONS

Our findings demonstrate that the incorporation of key pharmacophores in 1 and 2 to provide I gave compounds of significant neurological interest. Using animal tests, we observed excellent seizure protection and pronounced pain reduction. The pharmacological basis for I anticonvulsant activity and pain protection has not been determined. Anticonvulsants, like many neurological agents,⁵³ typically exert their activity through multiple pathways.⁵⁴ Current studies are directed at optimizing sectors A–C in I and evaluating these compounds in electrophysiology, radioligand displacement, and functional assays to provide information about the mode(s) of action of these novel hybrids.

EXPERIMENTAL SECTION

General Methods

The general methods used in this study were identical to those previously reported¹⁸ and are summarized in the Supporting Information Section. All compounds were checked by TLC, ¹H and ¹³C NMR, MS, and elemental analyses. The analytical results are within +/− 0.40% of the theoretical value. The TLC, NMR and the analytical data confirmed the purity of the products was ≥95%.

Preparation of N-4′-((3″-Fluoro)benzyloxy)benzyl 2-Acetamidoacetamide (5)

A THF solution (120 mL) of 28 (1.50 g, 12.8 mmol) was stirred and cooled at −78 °C under Ar and then 4-methylmorpholine (NMM) (1.7 mL, 15.4 mmol) was added dropwise. After 2 min of stirring at this temperature, isobutylchloroformate (IBCF) (2.0 mL, 15.4 mmol) was added dropwise leading to the precipitation of a white solid. The reaction was allowed to proceed for additional 2 min and 32 (3.26 g, 14.1 mmol) was added portionwise at −78 °C. The mixture was allowed to stir at room temperature (2 h), and then the white solid filtered and the organic layer concentrated in vacuo. The solid was purified by flash column chromatography on silica gel with methanol/EtOAc (0/10 -> 5/5) as the eluant to obtain 5 as white solid (1.30 g, 31%): R_f = 0.11 (EtOAc); mp 155–156 °C; ¹H NMR (CDCl₃) δ 2.03 (s, 3H), 3.92 (d, J = 5.1 Hz, 3H), 4.38 (d, J = 5.7 Hz, 2H), 5.05 (s, 1H), 6.24–6.35 (br m, 2H), 6.92 (m, 2H), 6.97–7.04 (m, 1H), 7.11–7.21 (m, 4H), 7.30–7.38 (m, 1H); MS (M+H⁺)(ESI⁺) 331.1 [M + H⁺] (calcd for C₁₈H₁₉FN₂O₃H⁺ 331.1); Anal. (C₁₈H₁₉FN₂O₃•0.35 H₂O): C, H, F, N.

Preparation of (R)-N-4′-((3″-Fluoro)benzyloxy)benzyl 2-Acetamido-propionamide ((R)-6)

Employing the procedure to prepare 5 and using THF (100 mL), (R)-29 (1.50 g, 11.4 mmol), NMM (1.5 mL, 13.7 mmol), IBCF (1.8 mL, 13.7 mmol), and 32 (2.90 g, 12.6 mmol) gave (R)-6 that was purified by recrystallization (EtOAc) as a white solid (905 mg, 22%): R_f = 0.16 (EtOAc); mp 158 °C; [α]^26.2_D +27.3° (c 1, CHCl₃); ¹H NMR (CDCl₃) δ 1.37 (d, J = 6.9 Hz, 3H), 1.94 (s, 3H), 4.32 (d, J = 5.7 Hz, 2H), 4.46–4.56 (m, 1H), 5.02 (s, 2H), 6.41 (d, J = 7.5 Hz, 1H), 6.79–6.85 (br m, 1H), 6.87–6.92 (m, 2H), 7.00 (td, J = 2.4, 8.4 Hz, 1H), 7.10–7.20 (m, 4H), 7.30–7.37 (m, 1H); MS (M+H⁺)(ESI⁺) 345.2 [M + H ⁺] (calcd for C₁₉H₂₁FN₂O₃H⁺ 344.2); Anal. (C₁₉H₂₁FN₂O₃): C, H, F, N.

Preparation of (S)-N-4′-((3″-Fluoro)benzyloxy)benzyl 2-Acetamidopropionamide ((S)-6)

Employing the procedure of (R)-6 and using (S)-29 (1.50 g, 11.4 mmol), NMM (1.5 mL, 13.7 mmol), IBCF, 32 (2.90 g, 12.6 mmol) gave after workup and recrystallization (EtOAc) a white solid (1.50 g, 38%): R_f = 0.16 (EtOAc); mp 153–154 °C; [α]^26.2_D -28.1° (c 1, CHCl₃); ¹H NMR (CDCl₃) δ 1.38 (d, J = 7.2 Hz, 3H), 1.94 (s, 3H), 4.32 (d, J = 5.7 Hz, 2H), 4.46–4.57 (m, 1H), 5.02 (s, 2H), 6.38–6.46 (br d, 1H), 6.78–6.85 (br m, 1H), 6.89 (d, J = 8.4 Hz, 2H), 6.98–7.04 (m, 1H), 7.10–7.20 (m, 4H), 7.30–7.37 (m, 1H); MS (M+H⁺)(ESI⁺) 345.2 [M + H⁺] (calcd for C₁₉H₂₁FN₂O₃H⁺ 345.2); Anal. (C₁₉H₂₁FN₂O₃): C, H, F, N.

Preparation of (R)-N-4′-((3″-Fluoro)benzyloxy)benzyl 2-Acetamido-3-methylbutanamide ((R)-7)

Trifluoroacetic acid (2 mL) was added to a CH₂Cl₂ (10 mL) solution of solution (R)-33 (1.00 g, 2.3 mmol) at 0 °C and the solution was stirred at room temperature (16 h). The reaction solution was concentrated in vacuo, dried (30 min), and CH₂Cl₂ (20 mL) and a saturated aqueous Na₂CO₃ solution (20 mL) were added. The layers were separated and the aqueous layer was washed with CH₂Cl₂ (2 × 20 mL). The organic layers were combined and concentrated under vacuum.

The residue was dissolved in CH₂Cl₂ (20 mL) and Et₃N (0.49 mL, 3.5 mmol) and AcCl (200 μL, 2.8 mmol) were successively added at 0 °C. The mixture was stirred at room temperature (3 h), aqueous 10% citric acid (60 mL) was added, and the organic layer was separated. The aqueous layer was washed with CH₂Cl₂ (2 x 30 mL). All the organic layers were combined, washed with aqueous saturated NaHCO₃ (30 mL), and H₂O (30 mL), dried (MgSO₄), and concentrated in vacuo. The solid was recrystallized with EtOAc to obtain (R)-7 (585 mg, 68%) as a white solid: R_f = 0.26 (EtOAc); mp 199–200 °C; [α]^25.2_D +25.6° (c 0.5, MeOH); ¹H NMR (400 MHz, DMSO-d₆) δ 0.81 (d, J = 3.0 Hz, 3H), 0.83 (d, J = 3.0 Hz, 3H), 1.87 (s, 3H), 1.90–1.99 (m, 1H), 4.12–4.43 (dd, J = 6.4, 8.8 Hz, 1H), 4.19 (d, J = 5.8 Hz, 2H), 5.11 (s, 2H), 6.95 (d, J = 8.4 Hz, 2H), 7.11–7.19 (m, 3H), 7.24–7.28 (m, 2H), 7.39–7.46 (m, 1H), 7.87 (d, J = 8.8 Hz, 1H), 8.38 (t, J = 5.8 Hz, 1H); HRMS (M+Na⁺)(ESI⁺) 395.1747 [M + Na⁺] (calcd for C₂₁H₂₅FN₂O₃Na⁺ 395.1747); Anal. (C₂₁H₂₅FN₂O₃): C, H, F, N.

Preparation of (R)-N-4′-((3″-Fluoro)benzyloxy)benzyl 2-Acetamido-3,3-dimethylbutanamide ((R)-8)

Employing a procedure similar to (R)-7 and using (R)-N-4′-((3″-fluoro)benzyloxy)benzyl 2-amino-3,3-dimethylbutanamide (1.20 g, 3.3 mmol), CH₂Cl₂ (40 mL), Et₃N (0.93 mL, 6.6 mmol), and AcCl (0.34 mL, 4.8 mmol) gave after workup and purification by flash column chromatography on silica gel with EtOAc/hexanes (8/2 to 10/0) as the eluant (R)-8 as a white solid (1.03 g, 81%): R_f = 0.56 (EtOAc); mp 64–66 °C; [α]^27.0_D -15.6° (c 1, CHCl₃); ¹H NMR (400 MHz, CDCl₃) δ 0.99 (s, 9H), 1.89 (s, 3H), 4.42 (1/2 ABq, J = 5.2, 14.4 Hz, 1H), 4.34–4.40 (m, CH, 1H), 5.02 (s, 2H), 6.41 (br d, J = 8.8 Hz, 1H), 6.88 (d, J = 8.8 Hz, 2H), 6.85–6.96 (br m, 1H), 7.00 (t, J = 8.4 Hz, 1H), 7.11–7.18 (m, 4H), 7.30–7.36 (m, 1H); HRMS (M+Na⁺)(ESI⁺) 409.1903 [M + Na⁺] (calcd for C₂₂H₂₇FN₂O₃Na⁺ 409.1903); Anal. (C₂₂H₂₇FN₂O₃): C, H, F, N.

Preparation of N-4′-((3″-Fluoro)benzyloxy)benzyl 2-Acetamido-2-(pyridin-2-yl)acetamide ((R,S)-9)

To a solution of 38 (1.81 g, 4.77 mmol, 1 equiv) in MeOH (95 mL) was added ammonium formate (1.21 g, 19.08 mmol, 4 equiv) as a solid and then the reaction mixture was stirred at room temperature (5 min). Zn dust (Sigma-Aldrich <10 micron, 1.20 g, 19.08 mmol, 4 equiv) was added and the reaction heated at reflux (6 h), and then maintained at room temperature (16 h). The reaction mixture was filtered through Celite^®. The filtrate was concentrated and the residue was dissolved in CH₂Cl₂ (100 mL). The CH₂Cl₂ layer was washed with a brine (2 × 100 mL), dried (Na₂SO₄), and concentrated in vacuo. The crude 39 was used without further purification for the next step: R_f = 0.00 (EtOAc).

Compound 39 (4.77 mmol, 1 equiv) was dissolved in CH₂Cl₂ (100 mL) and then triethylamine (0.8 mL, 5.72 mmol, 1.2 equiv) and AcCl (0.4 mL, 5.72 mmol, 1.2 equiv) were carefully added at 0 °C and the resulting solution was stirred at room temperature (2 h). An aqueous saturated NaHCO₃ solution (100 mL) was added and the organic layer was extracted with CH₂Cl₂ (3 x 100 mL). The organic layers were combined, dried (Na₂SO₄), and concentrated in vacuo. The residue was purified by chromatography on silica gel with EtOAc/hexanes (7/3 to 10/0) as the eluent. The residue was recrystallized (EtOAc) to obtain (R,S)-9 as a white solid (935 mg, 48%): R_f = 0.47 (EtOAc); mp 154–155 ºC; ¹H NMR (CDCl₃) δ 2.14 (s, 3H), 4.29–4.41 (m, 2H), 5.03 (s, 2H), 5.56 (d, J = 6.0 Hz, 1H), 6.87 (d, J = 9.0 Hz, 2H), 6.98–7.43 (m, 10H), 7.70 (dt, J = 1.6, 7.8 Hz, 1H), 8.50–8.52 (m, 1H); Anal. (C₂₃H₂₂FN₃O₃): C, H, F, N.

Preparation of (R)-N-4′-((3″-Fluoro)benzyloxy)benzyl 2-Acetamido-3-methoxypropionamide ((R)-10)

A saturated HCl solution in dioxane (1 mmol/2 mL, 21.75 mL) was added to (R)-N-4′-((3″-fluoro)benzyloxy)benzyl 2-N-(tert-butoxycarbonyl)amino-3-methoxypropionamide (4.70 g, 10.9 mmol) at 0 °C and the solution was stirred at room temperature (4 h). The reaction solution was concentrated in vacuo and dried (30 min).

Employing a procedure similar to (R)-7, and using the residue, CH₂Cl₂ (40 mL), Et₃N (4.47 mL, 32.6 mmol), and AcCl (1.16 mL, 16.30 mmol) gave after workup and recrystallization (EtOAc) (R)-10 (2.60 g, 65%) as a white solid: R_f = 0.29 (7/3 hexanes/EtOAc); mp 152 °C; [α]^24.5_D -18.9° (c 1, CHCl₃); ¹H NMR (CDCl₃) δ 2.03 (s, 3H), 3.37 (s, 3H), 3.43 (dd, J = 7.2, 9.0 Hz, 1H), 3.79 (dd, J = 3.9, 9.0 Hz, 1H), 4.40 (d, J = 5.7 Hz, 2H), 4.49–4.55 (m, 1H), 5.05 (s, 2H), 6.43 (br d, J = 7.2 Hz, 1H), 6.64–6.83 (br m, 1H), 6.89–7.05 (m, 3H), 7.10–7.22 (m, 4H), 7.31–7.38 (m, 1H); HRMS (M+H⁺)(ESI⁺) 375.1720 [M + H⁺] (calcd for C₂₀H₂₃FN₂O₄H⁺ 375.1720); Anal. (C₂₀H₂₃FN₂O₄): C, H, F, N.

Preparation of (S)-N-4′-((3″-Fluoro)benzyloxy)benzyl 2-Acetamido-3-methoxypropionamide ((S)-10)

A saturated HCl solution in dioxane (1 mmol/2 mL, 20.8 mL) was added to (S)-N-4′-((3″-fluoro)benzyloxy)benzyl 2-N-(tert-butoxycarbonyl)amino-3-methoxypropionamide (4.50 g, 10.4 mmol) at 0 °C and the solution was stirred at room temperature (4 h). The reaction solution was concentrated in vacuo and dried (30 min).

Employing a procedure similar to (R)-7, and using the residue, CH₂Cl₂ (40 mL), Et₃N (4.3 mL, 31.2 mmol), and AcCl (1.1 mL, 15.6 mmol) gave after recrystallization (EtOAc) (S)-10 (3.10 g, 80%) as a white solid: R_f = 0.29 (7/3 hexanes/EtOAc); mp 149–150 °C; [α]^24.5_D +18.8° (c 1, CHCl₃); ¹H NMR (CDCl₃) δ 2.02 (s, 3H), 3.36 (s, 3H), 3.43 (dd, J = 7.5, 9.1 Hz, 1H), 3.79 (dd, J = 4.2, 9.1 Hz, 1H), 4.40 (d, J = 5.7 Hz, 2H), 4.50–4.55 (m, 1H), 5.05 (s, 2H), 6.47 (br d, J = 6.0 Hz, 1H), 6.70–6.79 (br m, 1H), 6.90–7.05 (m, 3H), 7.10–7.22 (m, 4H), 7.31–7.38 (m, 1H); HRMS (M+H⁺)(ESI⁺) 375.1720 [M + H⁺] (calcd for C₂₀H₂₃FN₂O₄H⁺ 375.1720); Anal. (C₂₀H₂₃FN₂O₄): C, H, F, N.

Preparation of (R)-N-(3″-Fluorobiphenyl-4-yl)methyl 2-Acetamido-3-methoxypropionamide ((R)-11)

To a flame-dried Schlenck tube, under Ar, containing a dioxane (22.5 mL) solution of (R)-69¹⁸ (1.50 g, 4.0 mmol), palladiumtetrakis(triphenylphosphine) (464 mg, 0.402) and 3-fluorophenylboronic acid (70) (670 mg, 4.80 mmol) was added an aqueous solution (9 mL) of Cs₂CO₃ (2.60 g, 8.0 mmol). The mixture was stirred at reflux (16 h). Then MeOH and silica gel were added, and the volatiles were concentrated in vacuo. The residue was purified by flash chromatography on silica gel with EtOAc/MeOH (10/0 to 9/1) as the eluant to obtain (R)-11 (0.95 g, 60%) as a yellowish solid. To remove traces of palladium impurities, the solid was treated with 6.00 g of resin scavenger (SPM32, PhosPhonics) in CH₂Cl₂. The mixture was stirred at room temperature (2 h), filtered, and the filtrate evaporated under vacuum to obtain 800 mg (58%) of (R)-11 as a white solid: R_f = 0.22 (EtOAc); mp 170–172 °C; [α]^25.3_D = −8.1° (c 0.5, CHCl₃); ¹H NMR (CDCl₃) δ 2.03 (s, 3H), 3.39 (s, 3H), 3.47 (d, J = 7.5, 9.3 Hz, 1H), 3.81 (d, J = 3.9, 9.3 Hz, 1H), 4.45–4.55 (m, 2H), 4.56–4.63 (m, 1H), 6.53 (br d, J = 6.6 Hz, 1H), 6.93–7.07 (m, 2H), 7.23–7.51 (m, 5H), 7.53 (d, J = 8.1 Hz, 2H); HRMS (M+Cs⁺)(ESI⁺) 477.0591 [M + Cs⁺] (calcd for C₁₉H₂₁FN₂O₃Cs⁺ 477.0587); Anal. (C₁₉H₂₁FN₂O₃): C, H, F, N.

Preparation of (R)-N-4′-((3″-Fluoro)phenoxy)benzyl 2-N-Acetamido-3-methoxypropionamide ((R)-12)

A saturated HCl solution in dioxane (1 mmol/2 mL, 16.7 mL) was added to an Et₂O (8 mL) solution of (R)-N-4′-((3″-fluoro)phenoxy)benzyl 2-N-(tert-butoxycarbonyl)amino-3-methoxypropionamide (3.50 g, 8.4 mmol) at 0 °C and the solution was stirred at room temperature (16 h). The reaction solution was concentrated in vacuo and dried (30 min).

Employing a procedure similar to (R)-7, and using the residue, CH₂Cl₂ (40 mL), Et₃N (3.52 mL, 25.1 mmol), and AcCl (0.91 mL, 12.5 mmol) gave after workup and purification by flash column chromatography on silica gel with EtOAc as the eluant (R)-12 as a white solid (1.30 g, 43%): R_f = 0.45 (EtOAc); mp 125–126 °C; [α]^25.3_D –14.8° (c 1, CHCl₃); ¹H NMR (300 MHz, CDCl₃) δ 2.04 (s, 3H), 3.39 (s, 3H), 3.45 (dd, J = 7.5, 9.3 Hz, 1H), 3.81 (dd, J = 4.2, 9.3 Hz, 1H), 4.45 (d, J = 6.0 Hz, 2H), 4.53–4.59 (m, 1H), 6.48 (br d, J = 6.0 Hz, 1H), 6.68 (dt, J = 2.4, 10.2 Hz, 1H), 6.74–6.89 (m, 3H), 6.99 (d, J = 9.0 Hz, 2H), 7.21–7.34 (m, 3H); HRMS (M+H⁺)(ESI⁺) 361.1564 [M + H⁺] (calcd for C₁₉H₂₁FN₂O₄H⁺ 361.1563); Anal. (C₁₉H₂₁FN₂O₄): C, H, F, N.

Preparation of (R)-N-4′-((3″-Fluoro)phenethyl)benzyl 2-Acetamido-3-methoxypropionamide ((R)-13)

Pd/C (18 mg) was added to an EtOH solution of (R)-14 (180 mg, 0.49 mmol), and the mixture was stirred at room temperature under H₂ (1 atm) (36 h). The reaction mixture was filtered through a pad of Celite^®, and the pad was washed successively with EtOH and CH₂Cl₂. The filtrate was concentrated under vacuum to obtain (R)-13 (170 mg, 94%) as a white solid: R_f = 0.29 (EtOAc); mp 134–136 °C; [α]^24.4_D = −12.3° (c 0.48, CHCl₃); ¹H NMR (400 MHz, CDCl₃) δ 2.03 (s, 3H), 2.90 (s, 4H), 3.38 (s, 3H), 3.43 (dd, J = 7.6, 9.2 Hz, 1H), 3.81 (dd, J = 4.0, 9.2 Hz, 1H), 4.40–4.47 (m, 2H), 4.50–4.55 (m, 1H), 6.41–6.47 (br m, 1H), 6.68–6.75 (br m, 1H), 6.84–6.94 (m, 3H), 7.11–7.25 (m, 5H);HRMS (M+H⁺)(ESI⁺) 373.1927 [M + H⁺] (calcd for C₂₁H₂₅FN₂O₃H⁺ 373.1927); Anal. (C₂₁H₂₅FN₂O₃•0.32H₂O): C, H, N.

Preparation of (2-R,E)-N-4′-((3″-Fluoro)styryl)benzyl 2-Acetamido-3-methoxypropionamide ((R)-14)

To a flame-dried Schlenck tube, under Ar, containing a dioxane (22.5 mL) solution of (R)-69¹⁸ (1.50 g, 4.0 mmol), palladiumtetrakis(triphenylphosphine) (464 mg, 0.402) and trans-2-((3-fluoro)phenyl)vinylboronic acid (71) (800 mg, 4.82 mmol) was added an aqueous solution (9 mL) of Cs₂CO₃ (2.60 g, 8.0 mmol). The mixture was stirred at reflux (16 h). Then, MeOH and silica gel were added. The volatiles were concentrated in vacuo and the residue was purified by flash chromatography on silica gel with EtOAc/MeOH (10/0 to 9/1) as the eluant to obtain (R)-14 (0.90 g, 60%) as a yellowish solid. To remove traces of palladium impurities, the solid was treated with 6.00 g of resin scavenger (SPM32, PhosPhonics) in CH₂Cl₂. The mixture was stirred at room temperature (2 h), and filtered, the filtrate evaporated under vacuum to obtain 560 mg (37%) of (R)-14 as a white solid: R_f = 0.53 (EtOAc); mp 206–208 °C; [α]²⁷_D = −20.6° (c 0.5, CHCl₃); ¹H NMR (CDCl₃) δ 2.04 (s, 3H), 3.40 (s, 3H), 3.40–3.48 (m, 1H), 3.83 (d, J = 3.9, 8.7 Hz, 1H), 4.47–4.56 (m, 3H), 6.41–6.49 (br d, 1H), 6.75–7.02 (br t, 1H), 6.92–7.01 (m), 7.07 (d, J = 2.7 Hz), 7.18–7.35 (m), 7.47 (d, J = 8.4 Hz) (10H); LRMS (M+Na⁺)(ESI⁺) 393.1 [M + Na⁺] (calcd for C₂₁H₂₃FN₂O₃Na⁺ 393.1); Anal. (C₂₁H₂₃FN₂O₃): C, H, F, N.

Preparation of (R)-N-4′-(((3″-Fluoro)phenyl)ethynyl)benzyl 2-Acetamido-3-methoxypropionamide ((R)-15)

To an anhydrous THF (70 mL) solution of (R)-69¹⁸ (2.60 g, 7.0 mmol) were sequentially added triethylamine (0.95 mL, 14.0 mmol), 3-(fluoro)phenylacetylene (72) (1.20 ml, 10.37 mmol), dichlorobis(triphenylphosphine)palladium (II) (491 mg, 0.70 mmol), and CuI (200 mg, 0.1.05 mmol) under Ar. The mixture was stirred at room temperature (16 h), and then MeOH and silica gel were added. The volatiles were concentrated in vacuo and the residue was purified by flash chromatography on silica gel with EtOAc/MeOH (9/1) as the eluant to obtain (R)-15 (2.40 g, 93%) as a yellowish solid. To remove traces of palladium impurities, the solid was treated with 21.00 g of resin scavenger (SPM32, PhosPhonics) in CH₂Cl₂. The mixture was stirred at room temperature (2 h), and filtered, the filtrate evaporated under vacuum. The solid was recrystallized with EtOAc to obtain 1.20 g (46%) of (R)-15 as a white solid: R_f = 0.26 (EtOAc); mp 200–202 °C; [α]²⁴_D = −2.6° (c 0.5, CHCl₃); ¹H NMR (DMSO-d₆) δ 1.88 (s, 3H), 3.27 (s, 3H), 3.48–3.57 (m, 2H), 4.33 (d, J = 6.1 Hz, 2H), 4.45–4.53 (m, 1H), 7.25–7.32 (m, 3H), 7.38–7.53 (m, 5H), 8.13 (d, J = 6.3 Hz, 1H), 8.56 (br t, J = 6.1 Hz, 1H); HRMS (M+H⁺)(ESI⁺) 369.1614 [M + H⁺] (calcd for C₂₁H₂₁FN₂O₃H⁺ 369.1614); Anal. (C₂₁H₂₁FN₂O₃): C, H, F, N.

Preparation of (R)-N-4′-(((3″-Fluoro)phenoxy)methyl)benzyl 2-Acetamido-3-methoxypropionamide ((R)-16)

A saturated HCl solution in dioxane (1 mmol/2 mL, 10.2 mL) was added to (R)-N-4′-(((3″-fluoro)phenoxy)methyl)benzyl 2 -N-(tert-butoxycarbonyl)amino-3-methoxypropionamide (2.20 g, 5.1 mmol) at 0 °C and the solution was stirred at room temperature (2 h). The reaction solution was concentrated in vacuo and dried (30 min) to provide (R)-2-amino-N-4′-(((3″-fluoro)phenoxy)methyl)benzyl-3-methoxypropionamide hydrochloride as a white solid (1.80 g, quant.).

Employing a procedure similar to (R)-7, and using triethylamine (1.5 mL, 5.2 mmol), acetyl chloride (380 μL, 10.7 mmol), CH₂Cl₂ (20 mL), and (R)-N-4′-(((3″-fluoro)phenoxy)methyl)benzyl 2-amino-3-methoxypropionamide hydrochloride (1.30 g, 3.5 mmol) gave after workup and recrystallization (EtOAc) (R)-16 (900 mg, 68%) as a white solid: R_f = 0.18 (EtOAc); mp 140–142 °C; [α]^26.9_D -21.0° (c 1, CHCl₃); ¹H NMR (CDCl₃) δ 2.04 (s, 3H), 3.41 (s, 3H), 3.43 (dd, J = 7.5, 9.2 Hz, 1H), 3.82 (dd, J = 3.9, 9.2 Hz, 1H), 4.47–4.59 (m, 3H), 5.03 (s, 2H), 6.38–6.43 (br d, 1H), 6.64–6.78 (m, 4H), 7.14–7.30 (m, 3H), 7.40 (d, J = 8.4 Hz, 2H); LRMS (M+Na⁺)(ESI⁺) 397.1 [M + Na⁺] (calcd for C₂₀H₂₃FN₂O₄H⁺ 397.1); Anal. (C₂₀H₂₃FN₂O₄): C, H, F, N.

Preparation of (R)-N-4′-((3″-Fluoro)benzylamino)benzyl 2-Acetamido-3-methoxypropionamide ((R)-17)

Compound 67. HCl (293 mg, 1.1 mmol) was added to a THF (10 mL) solution of the (R)-68 (161 mg, 1.0 mmol) and the mixture was stirred at room temperature (5 min) and then NMM (121 μL, 1.1 mmol) was added. The mixture was stirred at room temperature (5 min) and DMTMM (332 mg, 1.2 mmol) was added, and the mixture was stirred at room temperature (16 h). The white precipitate was filtered and the filtrate was concentrated in vacuo. The residue was purified by flash column chromatography on silica gel with EtOAc/hexanes (5/5) to EtOAc/acetone (5/5) as the eluant to obtain (R)-17 as a yellow solid (140 mg, 35%): R_f = 0.37 (EtOAc); mp 78–81 °C; [α]^26.9_D −15.0° (c 0.5, CHCl₃); ¹H NMR (CDCl₃) δ 2.03 (s, 3H), 3.36 (s, 3H), 3.40 (dd, J = 7.2, 9.0 Hz, 1H), 3.82 (dd, J = 4.2, 9.0 Hz, 1H), 4.12–4.19 (br m, 1H), 4.31–4.37 (m, 4H), 4.46–4.52 (m, 1H), 6.38–6.45 (br m, 1H), 6.57 (d, J = 9.0 Hz, 3H), 6.91–6.89 (m, 1H), 7.05–7.15 (m, 4H), 7.27–7.34 (m, 1H); HRMS (M+H⁺)(ESI⁺) 374.1880 [M + H⁺] (calcd for C₂₀H₂₄FN₃O₃H⁺ 374.1879).

Preparation of (R)-N-4′-(((3″-Fluoro)benzyloxy)methyl)benzyl 2-Acetamido-3-methoxypropionamide ((R)-18)

A saturated HCl solution in dioxane (1 mmol/2 mL, 1.2 mL) was added to (R)-N-4′-(((3″-fluoro)benzyloxy)methyl)benzyl 2-N-(tert-butoxycarbonyl)amino-3-methoxypropionamide (1.10 g, 5.8 mmol) at 0 °C and the solution was stirred at room temperature (16 h). The reaction solution was concentrated in vacuo and dried (30 min).

Employing a procedure similar to (R)-7, and using the residue, CH₂Cl₂ (20 mL), Et₃N (1.40 mL, 9.8 mmol), and AcCl (356 μL, 4.9 mmol) gave after workup and recrystallization (EtOAc) (R)-18 (450 mg, 47%) as a white solid: R_f = 0.26 (EtOAc); mp 140–142 °C; [α]^25.2_D −21.0° (c 0.5, CHCl₃); ¹H NMR (CDCl₃) δ 2.04 (s, 3H), 3.39 (s, 3H), 3.43 (dd, J = 7.8, 9.0 Hz, 1H), 3.82 (dd, J = 3.9, 9.0 Hz, 1H), 4.48 (d, J = 6.0 Hz, 2H), 4.48–4.56 (m, 1H), 4.54 (s, 2H), 4.55 (s, 2H), 6.42 (br d, J = 6.6 Hz, 1H), 6.71–6.79 (br t, 1H), 6.96–7.15 (m, 3H), 7.24–7.35 (m, 5H); Anal. (C₂₁H₂₅FN₂O₄): C, H, F, N.

Preparation of (R)-N-4′-((3″-Fluoro)phenethoxy)benzyl 2-N-Acetamido-3-methoxypropionamide ((R)-19)

A saturated HCl solution in dioxane (1 mmol/2 mL, 10.0 mL) was added to an Et₂O (5 mL) solution of (R)-N-4′-((3″-fluoro)phenethoxy)benzyl 2-N-(tert-butoxycarbonyl)amino-3-methoxypropionamide (2.20 g, 5.0 mmol) at 0 °C and the solution was stirred at room temperature (16 h). The reaction solution was concentrated in vacuo and dried (30 min).

Employing a procedure similar to (R)-7, and using the residue, CH₂Cl₂ (30 mL), Et₃N (2.1 mL, 15.0 mmol), and AcCl (0.54 mL, 7.5 mmol) gave after workup and recrystallization (EtOAc) (R)-19 as a white solid (1.30 g, 66%): R_f = 0.28 (EtOAc); mp 147–148 °C; [α]^25.2_D −16.6° (c 0.5, CHCl₃); ¹H NMR (400 MHz, CDCl₃) δ 2.02 (s, 3H), 3.08 (t, J = 7.1 Hz, 2H), 3.36 (s, 3H), 3.39–3.44 (m, 1H), 3.79 (dd, J = 4.8, 9.6 Hz, 1H), 4.15 (t, J = 7.1 Hz, 2H), 4.33–4.44 (m, 2H), 4.49–4.54 (m, 1H), 6.44 (br d, J = 6.4 Hz, 1H), 6.65–6.73 (br t, 1H), 6.84 (d, J = 8.0 Hz, 2H), 6.90–7.06 (m, 3H), 7.16 (d, J = 8.0 Hz, 2H), 7.23–7.29 (m, 1H); HRMS (M+Na⁺)(ESI⁺) 411.1696 [M + H⁺] (calcd for C₂₁H₂₅FN₂O₄Na⁺ 411.1697); Anal. (C₂₁H₂₅FN₂O₄): C, H, F, N.

Preparation of (R)-N-4′-(Benzyloxy)benzyl 2-N-Acetamido-3-methoxypropionamide ((R)-20)

A saturated HCl solution in dioxane (1 mmol/2 mL, 24.1 mL) was added to an Et₂O (10 mL) solution of (R)-N-4′-(benzyloxy)benzyl 2-N-(tert-butoxycarbonyl)amino-3-methoxypropionamide (5.00g, 12.1 mmol) at 0 °C and the solution was stirred at room temperature (16 h). The reaction solution was concentrated in vacuo and dried (30 min).

Employing a procedure similar to (R)-7, and using the residue, CH₂Cl₂ (60 mL), Et₃N (5.1 mL, 36.3 mmol), and AcCl (1.4 mL, 18.8 mmol) gave after workup and recrystallization (EtOAc) (R)-20 as a white solid (2.60 g, 60%): R_f = 0.28 (EtOAc); mp 149 °C; [α]^25.1_D −26.8° (c 0.5, CHCl₃); ¹H NMR (400 MHz, CDCl₃) δ 2.02 (s, 3H), 3.36 (s, 3H), 3.39–3.44 (br m, 1H), 4.02–3.79 (dd, J = 4.2, 9.4 Hz, 1H), 4.36–4.44 (m, 2H), 4.48–4.55 (m, 1H), 5.05 (s, 2H), 6.42 (br d, J = 6.0 Hz, 1H), 6.64–6.71 (br m, 1H), 6.93 (d, J = 7.8 Hz, 2H), 7.18 (d, J = 7.8 Hz, 2H), 7.29–7.44 (m, 5H); HRMS (M+Na⁺)(ESI⁺) 379.1634 [M + Na⁺] (calcd for C₂₀H₂₄N₂O₄Na⁺ 379.1634); Anal. (C₂₀H₂₄N₂O₄): C, H, N.

Preparation of (R)-N-4′-((2″-Fluoro)benzyloxy)benzyl 2-Acetamido-3-methoxypropionamide ((R)-21)

A saturated HCl solution in dioxane (1 mmol/2 mL, 11.57 mL) was added to (R)-N-4′-((2″-fluoro)benzyloxy)benzyl 2-N-(tert-butoxycarbonyl)amino-3-methoxypropionamide (2.50 g, 5.8 mmol) at 0 °C and the solution was stirred at room temperature (16 h). The reaction solution was concentrated in vacuo and dried (30 min).

Employing a procedure similar to (R)-7, and using the residue (1.70 g, 5.1 mmol), CH₂Cl₂ (20 mL), Et₃N (2.10 mL, 15.3 mmol), and AcCl (550 μL, 7.6 mmol) gave after workup and recrystallization (EtOAc) (R)-21 (1.25 g, 65%) as a white solid: R_f = 0.28 (EtOAc); mp 173–174 °C; [α]^24.6_D −20.7° (c 1, CHCl₃); ¹H NMR (400 MHz, CDCl₃) δ 2.03 (s, 3H), 3.37 (s, 3H), 3.42 (dd, J = 7.6, 9.0 Hz, 1H), 3.79 (dd, J = 4.0, 9.0 Hz, 1H), 4.34–4.44 (m, 2H), 4.49–4.54 (m, 1H), 5.12 (s, 2H), 6.43 (br d, J = 6.4 Hz, 1H), 6.66–6.72 (br t, 1H), 6.94 (d, J = 8.0 Hz, 2H), 7.06–7.21 (m, 4H), 7.28–7.34 (m, 1H), 7.49 (td, J = 1.6, 7.6 Hz, 1H); HRMS (M+H⁺)(ESI⁺) 375.1720 [M + H⁺] (calcd for C₂₀H₂₃FN₂O₄H⁺ 375.1720); Anal. (C₂₀H₂₃FN₂O₄): C, H, F, N.

Preparation of (R)-N-4′-((4″-Fluoro)benzyloxy)benzyl 2-Acetamido-3-methoxypropionamide ((R)-22)

A saturated HCl solution in dioxane (1 mmol/2 mL, 11.57 mL) was added to (R)-N-4′-((4″-fluoro)benzyloxy)benzyl 2-N-(tert-butoxycarbonyl)amino-3-methoxypropionamide (2.50 g, 5.8 mmol) at 0 °C and the solution was stirred at room temperature (16 h). The reaction solution was concentrated in vacuo and dried (30 min).

Employing a procedure similar to (R)-7, and using the residue (1.85 g, 5.6 mmol), CH₂Cl₂ (30 mL), Et₃N (2.36 mL, 16.8 mmol), and AcCl (608 μL, 8.4 mmol) gave after workup and recrystallization (EtOAc) (R)-22 (1.28 g, 61%) as a white solid: R_f = 0.22 (EtOAc); mp 166–167 °C; [α]^25.2_D −19.4° (c 1, CHCl₃); ¹H NMR (400 MHz, CDCl₃) δ 2.03 (s, 3H), 3.37 (s, 3H), 3.42 (dd, J = 7.6, 9.4 Hz, 1H), 3.79 (dd, J = 4.0, 9.4 Hz, 1H), 4.40 (d, J = 5.2 Hz, 2H), 4.49–4.54 (m, 1H), 5.01 (s, 2H), 6.40 (br d, J = 5.6 Hz, 1H), 6.62–6.69 (br t, 1H), 6.92 (d, J = 8.8 Hz, 2H), 7.07 (t, J = 8.8 Hz, 2H), 7.18 (d, J = 8.0 Hz, 2H), 7.37–7.41 (m, 2H); HRMS (M+H⁺)(ESI⁺) 375.1720 [M + H⁺] (calcd for C₂₀H₂₃FN₂O₄H⁺ 375.1720); Anal. (C₂₀H₂₃FN₂O₄): C, H, F, N.

Preparation of N-4′-((3″-Fluoro)benzyloxy)benzyl Acetamide (23)

Employing a procedure similar to (R)-7, and using 4-((3′-fluoro)benzyloxy)benzylamine (32) (1.00 g, 4.3 mmol), CH₂Cl₂ (40 mL), Et₃N (728 μL, 5.2 mmol), and AcCl (376 μL, 5.2 mmol) gave after workup and trituration (Et₂O) 23 (810 mg, 69%) as a white solid: R_f = 0.39 (EtOAc); mp 131–132 °C; ¹H NMR (400 MHz, DMSO-d₆) δ 1.84 (s, 3H), 4.16 (d, J = 5.6 Hz, 2H), 5.11 (s, 2H), 6.95 (d, J = 8.8 Hz, 2H), 7.11–7.18 (m, 3H), 7.24–7.29 (m, 2H), 7.40–7.45 (m, 1H), 8.21–8.24 (br t, 1H); HRMS (M+Na⁺)(ESI⁺) 296.1063 [M + Na⁺] (calcd for C₁₆H₁₆FNO₂Na⁺ 296.1062); Anal. (C₁₆H₁₆FNO₂): C, H, F, N.

Preparation of 2-(4′-((3″-Fluoro)benzyloxy)benzyl)amino-3-methoxypropionamide (74)

A solution of 79 (1.50 g, 4.32 mmol) in NH₃ (7 N in MeOH, 150 mL) was stirred at room temperature in a sealed tube (7 d). The solution was concentrated in vacuo, and the residue was recrystallized (EtOAc) to obtain 74 as a white solid (350 mg, 24%): R_f = 0.25 (EtOAc); mp 84–85 °C; ¹H NMR (300 MHz, CDCl₃) δ 3.31–3.36 (m, 4H), 3.60 (d, J = 5.7 Hz, 2H), 3.71 (½ AB_q, J = 12.9 Hz, 1H), 3.78 (½ AB_q, J = 12.9 Hz, 1H), 5.06 (s, 2H), 5.40–5.44 (br s, 1H), 7.10 (d, J = 9.0 Hz, 2H), 6.96–7.05 (br dt, 1H), 7.13–7.25 (m, 4H), 7.31–7.38 (m, 1H); M_r (+ESI) 355.16 [M+Na]⁺ (calcd for C₁₈H₂₁FN₂O₃Na⁺ [M+355.14]⁺). Anal. (C₁₈H₂₁FN₂O₃): C, H, F, N.

Preparation of 2-(4′-((3″-Fluoro)benzyloxy)benzyl)amino-2-(furan-2-yl)acetamide Oxalate (75)

A solution of 87 (880 mg, 0.05 mmol) in NH₃ (7 N in MeOH, 88 mL) was stirred at 4 °C (16 h). The solution was concentrated in vacuo, and the residue was dissolved in THF (2.3 mL) to obtain a 1 N solution. To this solution, oxalic acid (2 N in THF, 4.6 mL) was added. After standing at room temperature (16 h) the precipitate was collected, dried, and recrystallized with i-PrOH. The white solid was recrystallized (absolute EtOH) to obtain 75 as a white solid (520 mg, 51%): R_f = 0.15 (EtOAc); mp 194–195 °C; ¹H NMR (DMSO-d₆) δ 3.75 (s, 2H), 4.54 (s, 1H), 5.14 (s, 2H), 6.44–6.45 (m, 1H), 6.48–6.50 (m, 1H), 7.51 (d, J = 9.0 Hz, 2H), 7.03–7.19 (m, 1H), 7.26–7.31 (m, 3H), 7.41–7.51 (m, 2H), 7.71–7.73 (m, 2H); M_r (+ESI) 355.13 [M+H]⁺ (calcd for C₂₂H₂₁FN₂O₇H⁺ 355.15 [M+H]⁺). Anal. (C₂₂H₂₁FN₂O₇): C, H, F, N.

Pharmacology

Compounds were screened under the auspices of the National Institutes of Health’s Anticonvulsant Screening Program. Experiments were performed in male rodents [albino Carworth Farms No. 1 mice (intraperitoneal route, ip), albino Spague-Dawley rats (oral route, po)]. Housing, handling, and feeding were in accordance with recommendations contained in the ‘Guide for the Care and Use of laboratory Animals’. Anticonvulsant activity was established using the MES test,²⁹ 6 Hz,⁴² hippocampal kindled seizure,⁴⁵ and the scMet test,⁴⁴ according to previously reported methods.¹⁶

Formalin Test^50a

The formalin test involved injection of 0.5% formalin into the mouse hind paw. Injection leads to a biphasic behavioral response characterized by licking of the affected paw. The number of licks is measured as a proxy for perceived pain. The first phase is termed the “acute” phase, and the second phase is termed the “inflammatory” phase. Each trial involved 16 animals, 8 controls given an ip injection of vehicle and 8 given the test compound at a specified dose. The amount of time that each animal spends licking is monitored at 2 min intervals, and monitoring continues for 45 min. Plots of time licking versus time provides a biphasic response and permits the area under the curve (AUC) to be determined for each animal for the acute and inflammatory phases. The AUC for each compound-treated animal is compared to the average result from the control group, yielding an average percent of control (reported with the SEM and p value).

Partial Sciatic Ligation Model⁴⁷

Rats are anesthetized with sodium pentobarbital and the depth of anesthesia monitored by their response to a tail pinch and observation of the depth of respiration. After surgical exposure of the sciatic nerve, the nerve is slightly elevated, and approximately one-third to one-half of the nerve is tied off. Typically, the surgical procedure is done on the right side while a sham surgery is performed on the left hind leg where the sciatic nerve is only exposed. After recovery (7 d), the animals are tested for the development of mechanical allodynia. The animals are placed in a bottomless plexiglass box placed on a wire mesh (1/4″) platform. After a 30–60 min acclimation period, a baseline mechanical sensitivity is determined by applying a series of calibrated Von Frey fibres perpendicularly to the plantar surface of each hind paw and holding it in place for 6 sec with enough force to slightly bend the fibre. After a positive response (withdrawal of the foot) is observed a weaker fibre is applied until a 50% threshold for withdrawal can be determined. The allodynic threshold is then re-determined after ip administration of the test compound. Testing is conducted at the time-to-peak effect of the compound in the MES test.