3350
`
`J. Med. Chem. 1993,36, 335G3360
`
`Synthesis and Anticonvulsant Activities of a-Heterocyclic
`a-Acetamido-N-benzylacetamide Derivatives
`
`Harold Kahn,'?? Kailash N. Sawhney,? Patrick Bardel: David W. Robertson,:>$ and J. David Leander:
`Department of Chemistry, University of Houston, Houston, Texas 77204-5641, and Lilly Research Laboratories,
`Eli Lilly and Company, Lilly Corporate Center, Indianapolis, Indiana 46285
`
`Received May 14,19930
`Earlier studies showed that (R,S)-a-acetamido-N-benzylacetamides (2) containing a five- and six-
`membered aromatic or heteroaromatic group appended at the C(a) site displayed outstanding
`activity in the maximal electroshock-induced seizure (MES) test in mice. An expanded set of
`C(a)-heteroaromatic analogues of 2 have been prepared and evaluated. The observed findings
`extended the structureactivity relationships previously discerned for this novel class of anti-
`convulsants and have validated previous trends. The a-furan-2-yl (4), a-0xaz0l-2-yl (18), and
`a-thiazol-2-y1(19) a-acetamido-N-benzylacetamides afforded excellent protection against MES-
`induced seizures in mice. The EDm and PI values for these adducts rivaled those reported for
`phenytoin. The outstanding properties provided by 4 led to an in-depth examination of the effect
`of structural modification at key sites within this compound on biological activity. The
`pharmacological data in this series indicated that stringent steric and electronic requirements
`existed for maximal activity and revealed the outstanding activity of (R)-(-)-a-acetamido-N-(4-
`fluorobenzyl)-a-(furan-2-yl)acetamide [(R)-301.
`
`by a-acetamido-N-benzyl-a-(furan-2-yl)acetamide (4)
`against MES seizures led to an in-depth examination of
`the effect of structural modification at key sites in 4 on
`biological activity (Table 11). Included in this study was
`also the preparation of several enantiopure congeners of
`(R)-4 to demonstrate that this absolute configuration
`afforded compounds with marked anticonvulsant activity.
`
`Non-naturally occurring amino acid derivatives con-
`stitute an increasing resource of new chemotherapeutic
`agents that include antibacterial and CNS agents and
`enzyme inhibitors.' In recent years, we have reported on
`the anticonvulsant properties of functionalized amino acid
`derivatives l.%' Our studies demonstrated that a-acet-
`amido-N-benzylacetamides (2)containing a five- and six-
`membered aromatic or heteroaromatic group appended
`at the C(a)-site afforded excellent protection against
`maximal electroshock (MES)-induced seizures in mice
`(Table 1 ) , 3 3 6 For example, the EDSO values against MES
`seizures for the racemic a-phenyl (3) (32.1 mg/kg) and
`a-furan-2-yl (4) (10.3 mg/kg) derivatives6 compared fa-
`vorably with phenobarbital (21.8 mg/kg) and phenytoin
`(9.5 mg/kg).8 Examination of the individual enantiomers
`of 3 and 4 demonstrated the importance of stereochemistry
`at the C(a) site in 2 on biological activity.s96 In both 3 and
`4, the (R)-stereoisomer was 10 times more potent in the
`control of MES seizures than the (S)-enantiomer. This
`difference in activity is the greatest eudismic ratio reported
`to date for MES-selective anticonvulsant agents.
`
`Selection of Compounds
`Our investigation proceeded in two stages. First, we
`determined the effect of the C(a)-heteroaromatic group
`in 2 on anticonvulsant activity (Table I). Amino acid
`(Table I) served as the reference com-
`derivatives
`pounds for this investigation. The placement of additional
`heteroatoms within these derivatives or the preparation
`of isomeric adducts led to multiple effects. These included
`perturbations in the electron density of the aromatic ring,
`changes in the spatial orientation of the nonbonding
`electrons, and alterations in the basicity and bioavailability
`of the drug candidates. These multifaceted electronic,
`structural, and physical effects complicated the interpre-
`tation of the biological data. Nonetheless, the pronounced
`improvement in MES-induced seizure protection previ-
`ously observed by the placement of an electron-rich
`aromatic ring at the C(a)-site in 2 (i.e., 3 (EDm = 32.1
`mg/kg) versus 4 (ED50 = 10.3 mg/kg); 7 (EDm = 44.8 mg/
`kg) versus 5 (ED60 = 16.1 mg/kg) versus 4 (ED60 = 10.3
`mg/ kg)) prompted us to provide additional documentation
`for this trend. The compounds selected for synthesis and
`evaluation were grouped into three categories. The first
`set (i.e., 9-12) included aza analogues of 5 where the C(a)-
`heteroaromatic ring was appended by a carbon-carbon
`bond. Compounds in this series were a-imidazolyl (9 and
`lo), a-triazolyl (ll), and a-tetrazolyl (12). The second
`category (i.e., 13-17) encompassed the isomeric C(a)-
`'Author to whom correspondence should be addreseed at the
`azaromatics where heteroaromatic attachment to the
`University of Houston.
`amino acid backbone occurred through a nitrogen-carbon
`t University of Houston.
`bond. Compounds evaluated were a-pyrrolyl(13), a-pyra-
`t Lay Research Laboratories.
`I Current address: Ligand Pharmaceuticals, 9393 Tome Center Dr.,
`zolyl (14), a-imidazolyl (15), a-triazolyl (16), and a-tet-
`Suite 100, San Diego, CA 92121.
`* Abstract published in Advance ACS Abstracts, October 1, 1993.
`razolyl(17) adducts. Finally, the third category (i.e., 18-
`1993 American Chemical Society
`0022-2623/93/1836-3350$04.00/0 0
`
`0
`I1
`R'CNH-C
`
`$ 0
`I
`I1
`-CNHR3
`1
`H
`1
`
`In the present study we report the synthesis and
`pharmacological activities of a carefully selected series of
`C(a)-heteroaromatic analogues of 2. Information is pro-
`vided on the effect of type, number, and site of heteroatom
`substitution within the C(a)-substituent on anticonvulsant
`activity (Table I). The outstanding properties provided
`
`IPR2014-01126- Exhibit 1025 p. 1
`
`

`
`a-Acetamido-N- benzylacetamide Derivatives
`
`Journal of Medicinal Chemistry, 1993, Vol. 36, No. 22 3381
`Table I. Physical and Pharmacological Data in Mice for C(cr)-Heteroaromatic a-Acetamido-N-benzylacetamidesa
`0
`$ 0
`I
`II
`I1
`cH,cNH-c
`- C N H C H , ~
`I
`H
`
`R*
`
`m d
`202-203
`
`178-179
`
`174-175
`
`MESC EDso
`32.1 (27.5-40.2)
`
`10.3 (9.1-11.6)
`
`HSd TDso
`
`>40 - 40
`
`16.1 (13.2-19.9)
`
`>30, C100
`
`PI*
`
`>3.9
`
`no.
`
`7f
`
`8f
`
`9
`
`10
`
`11
`
`12
`
`13
`
`14
`
`15
`
`16
`
`17
`
`18
`
`19
`
`20
`
`179-181
`
`-300
`
`167-169
`
`198-199
`
`228-230
`
`188-191 dec
`
`205-207
`
`44.8 (38.9-51.4)
`
`87.8 (69.9-150)
`
`>loo
`
`>loo
`
`>loo
`
`236-238
`
`>30, C100
`
`182-184
`
`158-160
`
`146-148
`
`146-148
`
`169-171
`
`164-166
`
`166-167
`
`164-166
`
`80.2 (66.6-100.6)
`
`16.5 (14.1-22.5)
`
`>loo
`
`>30, C100
`
`>300
`
`8
`
`8
`
`g
`
`g
`
`g
`
`g
`
`10.4 (9.2-11.6)
`
`38.6" (33.8-46.0)
`
`12.1 (9.5-14.5)
`
`69.1" (61.6-78.6)
`
`3.7
`
`5.7
`
`>loo, c300
`
`g
`
`9.5 (8.1-10.4)
`6.9
`65.g (52.5-72.1)
`phenytoin'
`21.8 (15.0-22.5)
`3.2
`69.e (62.8-72.9)
`phenobarbital'
`272 (247-338)
`1.6
`426" (369-450)
`valwoatei
`a The compounds were administered intraperitoneally. EDm and TDm values are in milligrams per kilogram. Number in parentheses are
`95% confidence intervals. A dose-responee curve waa generated for all compounds that displayed sufficient activity. The dose effect data
`for these compounds was obtained at 0.5 h ("time of peak effect") except for compound 19, which waa obtained at 0.25 h. b Melting points
`(OC) are uncorrected. MES = maximal electroshock seizure test. d HS TDm = neurologic toxicity determined from horizontal screen unless
`otherwise noted. PI = protective index (TDso/EDm). f Reference 6. 8 Not determined. Neurologic toxicity determined using the rotorod test.
`Reference 8.
`20) contained C(a)-mixed heteroaromatic systems. The
`three compounds prepared were a-oxazolyl(18), a-thiazolyl
`(19), and a-l,2,4-oxadiazolyl(20) derivatives. In all cases,
`the functionalized amino acids were synthesized as the
`racemates.
`The second phase of this study (Table 11) focused on
`cu-acetamido-N-benzyl-a-(furan-2-yl)acetamide (4), the
`most active compound evaluated in the initial study.
`Structural modifications were conducted at key sites in 4,
`including the furan ring (i.e., 21), the C(a)-position (i.e.,
`22), the two amide carbonyl groups (i.e., 23,24), and the
`
`N-benzyl substituent (i.e., 25-29) to discern how these
`changes influenced biological activity. Moreover, because
`of the substantial eudismic ratio observed with enanti-
`omers of 4, enantiopure (R)-isomers of several congeners
`in the present series were prepared (i.e., (R)-30-(R)-32) to
`demonstrate that this absolute configuration afforded
`compounds with marked anticonvulsant activity.
`
`Chemistry
`The novel a-heteroaromatic amino acid derivatives 9-20
`were prepared from either a-acetamido-N-benzyl-a-bro-
`
`IPR2014-01126- Exhibit 1025 p. 2
`
`

`
`3362 Journal of Medicinal Chemistry, 1993, Vol. 36, No. 22
`
`Kohn et 01.
`
`Table 11. Physical and Pharmacological Data in Mice for a-Acetoamido-N-be~yl-~-(furan-2-yl)ace~ide (4) Derivatives'
`
`Rb
`H
`
`Rc
`CHzCsHs
`
`R.
`--Q
`--Q
`CHzCeHs
`H
`- Q H CHzCsHs
`CHzCsHs
`
`no.
`
`4,
`
`(R1-4,
`(S)-U
`
`21a
`
`21b
`
`22
`23
`
`
`R
`Y
`
`X
`CHaCNH - C- C - NHR,
`l a
`II
`II
`I
`b
`Y
`0
`
`PI'
`>3.9
`
`7.2
`
`X
`0
`
` mPb
`178-179
`
`MESC EDm
`10.3 (9.1-11.6)
`
`- 40
`
`HSd TDm
`
`0
`
`0
`
`0
`
`0
`
`O
`s
`s
`
`0
`
`196-197
`
`3.3 (2.8-3.9)
`
`196-197
`0
`0 159-161
`0 130-132
`
`>25
`
`51.7 (44.4-59.9)
`
`89.8 (78.4-103.4)
`
`O
`
`0
`s
`
`A
`
`
`
`>300
`
`7&80
`
`99-101
`
`18.4 (15.9-22.0)
`>loo
`
`23.8
`
`>200
`
`g
`
`g
`
`g
`
`g
`
`8
`
`8
`
`- Q H
`
`CH2Cd-k
`
`- Q H
`3 CHS
`CHzCeHa
`- - Q H CHzCsHa
`-Q
`H
`CHzCsHs
`
`24
`25
`
`26
`
`27
`
`28
`
`29
`
`30
`(R)-30
`
`(R)-31
`
`(R)-32
`
`H
`
`CH,-Q-F
`
`- Q H CHZ C N - 0
`-Q
`H
`-g
`H
`- Q H C
`- Q H CH2-+,,
`- Q H cy--Q-cF3
`
`H
`
`,
`
`~
`
`0
`
`0
`
`0
`
`0
`
`0
`
`0
`
`0
`
`0
`
`172-174
`
`-30
`
`168-170
`
`>lo0
`
`210-212
`
`226-228
`
`>loo
`
`>lo0
`
`F
`
`0
`0
`188-190
` 0 0 205-207
`0
`210-212
`0
`0 193-195
`
`0
`
`12.7 (10.4-15.1)
`
`3.5 (2.9-4.4)
`
`43.6 (26.1-143)
`
`22.8(15.9-33.4)
`
`8
`
`8
`
`8
`144 (123-171)
`
`14.4 (7.3-28.9)
`
`11.3
`
`4.1
`
`g
`
`g
`
`
`
`'Bc NMFV
`
`6.9
`65.g (52.5-72.1)
`9.5 (8.1-10.4)
`phenytoin'
`3.2
`69.G (62.8-72.9)
`21.8 (15.0-22.5)
`phenobarbital'
`1.6
`42@ (369-450)
`272 (247-338)
`valproatei
`0 The compounds were administered intraperitoneally. EDw and TDm values are in milligrams per kilogram. Numben, in parentheses are
`95% confidence intervals. A dose-response curve was generated for all compounds that displayed sufficient activity. The dose effect data
`for these compounds was obtained at 0.5 h ("time of peak effect") except for compound 27 which was obtained at 1 h. b Melting points ('C)
`are uncorrected. MES = maximal electroshock seizure test. Compound was suspended in 30 5% PEG. d HS TDm = neurologic toxicity determined
`from horizontal screen unless otherwise noted. * PI = protective index (TDdEDw). f Reference 6.8 Not determined. * Thick oil. i Reference
`8. i Neurologic toxicity determined using the rotorod test.
`moacetamide7 (33) or cY-acetamido-N-benzyl-cY-cyano- Table 111. NMR Assignments for Substituted Imidazoles'
`acetamides (34). Addition of a tetrahydrofuran solution
`2
`of the C(2)-lithio salt of l-(diethoxymethyl)imidazole10
`N b N - R
`L f
`
`(36) to 33 prepared in situ afforded 9 after workup.
`Correspondingly, treatment of 33 with triethylamine
`4
`5
`followed bv introduction of the lithio salt of l-W,
`1H NMR*
`N-dimethyisulfamoyl)imidazolell (36) gave 37, which upon
`deprotection with acid furnished 10. The structure of 37
`has been tentatively assigned as the C(4bimidazole-
`substituted derivative based on a comparison of the NMR
`chemical shift values for 37 versus the parent heterocycle
`3611 and 1-(N,N-dimethylsulfamoyl)d-methylimidazole
`(38) (Table 111). Compound 38 was prepared by the
`addition of dimethylsulfamoyl chloride to 4-methylim-
`idazole (39) in the presence of triethylamine. NMR and
`TLC analyses of the crude reaction mixture indicated the
`presence of only one major compound, and the structure
`was confirmed by X-ray crystallography (Figure 1, sup-
`plementary material). Select lH-13C NMR decoupling
`experiments on 38 provided the assignments listed in Table
`
`7.13
`7.13
`
`121.96
`121.96
`7.13
`imidazole
`118.75
`130.45
`7.56
`36
`115.50
`140.26
`7.40
`37
`114.34
`138.85
`7.32
`38
`131.00
`6.75
`39
`118.18
`115.9
`130.g
`7.47
`7.09.
`4od
`All spectra were recorded in DMSO-& unlw otherwise indicated.
`The number in each entry is the chemical shift value (6) okerved
`in ppm relative to Me&. 1H NMR spectra were recorded at 300
`MHz. 1% NMR spectra were obtained at 75 MHz. d Spectra taken
`in CDCk,. e Reference 12b. f Reference 12a.
`I11 and permitted the assignments for the corresponding
`NMR resonances in 36 and 37. The lSC NMR amignments
`
`(1
`
`IPR2014-01126- Exhibit 1025 p. 3
`
`

`
`Journal of Medicinal Chemistry, 1993, Vol. 36, No. 22 3353
`a-Acetamido-N- benzylacetamide Derivatives
`Scheme I. Preparation of a-Acetamido-N-benzyl-a-methyl-a-(furan-2-yl)acetamide (22)
`0
`0
`I
`Br 0
`0
`It
`II
`II
`II
`* CH3CNH -C-COCHa
`CH3CNH -C-COCH:,
`I
`II
`kH3
`45
`
`HBr
`
`II -
`
`O
`
`
`
`0 II
`
`T
`
`CHaCNH-C-CR
`I
`
`for 36 were in agreement with the values reported by
`Chadwick and Ngochindoll and follow the pattern cited
`by Begtrup and co-workers for N-acetylimidazole12* (40).
`Our NMR decoupling experiments on 36, however, re-
`quired a reversal of the previously proposed C(4) and C(5)
`proton assignmente.ll The revised values mirrored the
`1H NMR pattern reported for N-acetylimidazole12b (40).
`The origin for the formation of the C(4)-imidazole-
`substituted derivative 37 has not been determined. Pre-
`vious studies have shown that treatment of the C(2)-lithio
`salt of 36 with alkyl halides furnished the C(2) substituted
`product, while addition of electrophiles to the C(2),C(5)-
`dilithio intermediate provided the C(5)-substituted adduct
`as a major product.ll
`
`$ 0
`0
`I
`II
`II
`CHsCNH-C -CNHCH2
`I
`H
`
`s $ = B r
`
`~ R ~ = C N
`
`Z! Ff = NAN--SOpN(CH&
`
`h=!
`41 R2 i.i C(0)NHp
`42 R2 = C(S)NH2
`
`R2 = C(NOH)NH2
`
`2
`
`R4 4
`
`N ~ N - R ~ %
`I R' - CH(OCH2CH3)2, R4 =H
`a R' = SOzN(CH&,
`39 R' - H, R4 I CH3
`3 R' = S02N(CH&, R4 = CH,
`
`= H
`
`4p R' = C(O)CH3, R4 I H
`
`Comparable protocols were employed for the prepara-
`tion of the N-substituted heteroaromatics 13-17 beginning
`with 33. Addition of an excess amount of the preformed
`potassium salt of pyrrole to 33 in tetrahydrofuran yielded
`13, while 14-17 were synthesized by initial treatment of
`33 with excess triethylamine at -78 "C followed by addition
`of the parent heterocycle.
`a- Acetamido-N-benzyl-a-cyanoacetamide (34) served as
`the starting point for the synthesis of 11, 12, and 18-20.
`Addition of formic hydrazide to 34 in basic ethanol gave
`11 upon workup, while treatment of 34 with KN3 and
`
`4§ R=OCH3
`42 R=OH
`triethylamine hydrochloride in l-methyl-2-pyrrolidinone
`afforded l2.'3 a-Oxazol-2-yl (18) and a-thiazol-2-yl (19)
`derivatives were prepared by initial conversion of 34 to
`the a-amidegb (41) and a-thioamide (42) adducts, respec-
`tively, and then these compounds were condensed with
`excess bromoacetaldehyde dimethyl acetal14 in dimethox-
`yethane. a-Oxadiazol-3-y1(20) derivative was generated
`in two steps from 34.15 Addition of NH20HqHCl to 34 in
`basic ethanol gave the a-carboxamide oxime derivative
`43. Treatment of 43 with trimethyl orthoformate and a
`catalytic amount of boron trifluoride etherate gave 20.
`Several synthetic protocols were utilized for the prep-
`aration of compounds 21-32. Catalytic hydrogenation (Hz,
`Pd/C) of (R,S)-4 gave the tetrahydrofuran-2-yl adduct 21.
`Fractional recrystallization of the product mixture from
`ethyl acetate provided diastereomers 21a and 21b. Syn-
`thesis of the a-methyl analogue 22 was achieved by a four-
`step procedure (Scheme I) beginning with methyl 2-ac-
`etamidoacrylate16 (44). Addition of HBr to 44 furnished
`45, which was directly treated with furan and ZnCl2 to
`give the a-amidoalkylation adduct 46.6J7 Hydrolysis of
`46 to the free acid 47, followed by treatment of 47 with
`benzylamine using the mixed carbonic anhydride coupling
`procedureeJ8 (i.e., isobutyl chloroformate, 4-methylmor-
`pholine), gave 22.
`The two thioamides 23 and 24 were prepared directly
`from 4 using Lawesson's reagent.lg Treatment of 4 with
`this thiation reagent (0.5 molar equiv) at room temperature
`yielded the monothio derivative 23. Elevation of the
`reaction temperature and the relative proportion of
`Lawesson's reagent (>1 molar equiv) to 4 gave the dithio
`product 24.
`Synthesis of 25, 26, and 29 was accomplished from
`racemic a-acetamido-a-(furan-2-yl)acetic acid (48), isobu-
`tyl chloroformate, 4-methylmorpholine, and the appro-
`priate amine or hydrazine, while use of (R)-a-acetamido-
`a-(furan-2-yl)acetic acid6 [ (R)-48] in this protocol with
`4-fluorobenzylamine, 4-methylbenzylamine, and 4-(trif-
`1uoromethyl)benzylamine furnished the three optically
`active N-benzylamides (R)-30-(R)-32, respectively. This
`coupling strategy previously provided enantiopure W - 4
`and (S)-4.6 Evidence that amide bond formation pro-
`
`IPR2014-01126- Exhibit 1025 p. 4
`
`

`
`3364 Journal of Medicinal Chemistry, 1993, Vol. 36, No. 22
`ceeded without racemization (<5%) was obtained by
`examining the lH NMR spectra (CDCls) of (R,S)-30 and
`(R)-30-(R)-32 both in the absence and the presence of
`saturating amounts of (R)-(-)-mandelic acid.20 Addition
`of this chiral solvating reagent to (R,S)-30 led to the
`appearance of two acetyl methyl signals (-A ppm 0.02)
`of equal intensity,2l while only a single acetyl methyl singlet
`was observed in the corresponding 'H NMR spectra for
`(R)-30-(R)-32. Similar results were earlier secured for (R)-
`4, (S)-4, and (R,S)-46 (see the supplementary material for
`appropriate 'H NMR spectra). Access to the starting
`material (R)-48 was readily achieved using the protocol
`advanced by Whitesides and co-workers.22 Treatment of
`racemic 48 with acylase I led to the selective hydrolysis
`of the (&amino acid derivative providing (R)-48 in 75%
`yield. Previously, (R)-48 was obtained by fractional
`
`48
`recrystallization of the corresponding diastereomeric salts
`formed with (R)-a-methylbenzylamine.6 The two pyridine
`N-oxide adducts 27 and 28 were prepared by treating 25
`and 26, respectively, with m-chloroperoxybenzoic acid.
`Pharmacological Evaluation
`The heteroaromatic amino acid derivatives 9-32 were
`tested for anticonvulsant activity using the procedures
`described by Krall and co-w~rkers?~ and these results were
`compared to the findings previously reported for 3-8.6 All
`compounds were administered intraperitoneally (ip) to
`mice. Tables I and I1 list the EDw values required to
`prevent toxic extension of the hind limbs in mice in the
`MES test by 9-32. Included in these tables are the median
`neurologically impairing dose (TDw) values using either
`the horizontal screen24 (HS) or the rotorod test.26 In most
`cases, the TDm's were only determined for those com-
`pounds that had good activity in the MES test. The
`protective index (PI = TDdEDw) for these adducts, where
`appropriate, is also shown in Tables I and 11.
`Our previous studies indicated that placement of
`electron-rich five- and six-membered aromatic and het-
`eroaromatic moieties at the a-site within functionalized
`amino acids 2 led to compounds providing excellent
`protection against MES-induced seizures in mice? More-
`over, we noted in this series that improved activity resulted
`by the positioning of a heteroatom two atoms removed
`from the C(a)-site. A similar result was observed in
`a-acyclic derivatives of 2.7 The pharmacological data
`obtained in this study provided evidence in support of
`these two structure-activity themes.
`Support for the beneficial value accrued by the place-
`ment of an electron-rich aromatic ring at the C(a)-position
`was obtained by the comparison of the EDm values in the
`MES-test for pyrrole 5 (EDm = 16.1 mg/kg) versus the
`azoles 9-12 (EDm > 30 mg/kg) (Table I). The data
`demonstrated that overall reduction of the electron
`excessive character of the C(a) ?r-aromatic system by
`heteroatom incorporation26 led to decreased biological
`activity despite the fact that additional nitrogen incor-
`poration often provided a substrate that contained two
`heteroatoms two atoms removed from the C(a)-site (i.e.,
`9, 11, 12).
`
`Kohn et al.
`
`Comparison of the pharmacological activities of the
`C-substituted azoles 5, 9-12 versus the N-substituted
`isomers 13-17 provided qualitative information concerning
`the importance of heteroatom substitution versus the C(a)-
`position. We observed a significant reduction in activity
`for 13 (EDw = 80.2 mg/kg) versus 5 (EDw = 16.1 mg/kg),
`and 17 (EDw > 300 mg/kg) versus 12 (EDm >30, <lo0
`mg/kg). In compound 5 one heteroatom exists two atoms
`removed from the C(a)-site, while in 13 there is none.
`Similarly, in 12 there are two heteroatoms two atoms
`removed from the C(a)-site, while in 17 there is only one.
`The delicate interplay of the ?r-electron character of the
`appended C(a)-heteroaromatic group, the site of the
`heteroatom incorporation, and the identity of the het-
`eroatom on anticonvulsant activity was reinforced by
`comparison of the biological activities of the a-oxazol-by1
`(18), a-imidazol-Byl(9), and a-thiazol-2-y1(19) derivatives.
`Of these three compounds, 18 was the most active (EDw
`= 10.4 mg/kg), displaying protection similar to that
`reported for phenytoin (ED50 = 9.5 mg/kg).* The slight
`decrease in protection in the MES test afforded by 19
`(EDw = 12.1 mg/kg) versus 18 paralleled the larger
`difference previously observed for a-furan-Byl(4) (EDm
`= 10.3 mg/kg) and a-thien-2-yl (7) (EDw = 44.8 mg/kg)
`adducts.6 Surprisingly, the a-imidazol-byl(9) derivative
`failed to protect the mice from MES-induced seizures at
`dosages of 100 mg/kg or less. Previously, we observed
`that the anticonvulsant activity of 2 decreased in pro-
`ceeding from oxygen to nitrogen to sulfur containing C(a)-
`heteroaromatic derivatives.6 The low potency of 9 may be
`a reflection in part of the increased basicity of this
`compound versus 18 and 19.2s
`The pyrazole derivative 14 provided protection in the
`MES test (EDm = 16.5 mg/kg) comparable to phenobar-
`bital (EDm = 21.8 mg/kg),8 and this compound was
`considerably more potent than the isomeric imidazoles 9,
`10, and 15. Our results do not provide information
`concerning the underlying factors that contribute to this
`difference in activity. We do note that pyrazoles are
`substantially less basic than imidazoles.26
`Inspection of the composite data set for analogues of
`a-acetamido-N-benzyl-a-(furan-2-yl)acetamide (4) re-
`vealed that most structural changes at the a-carbon, amide
`carbonyl, and N-benzylamide site in 4 led to decreased
`potency of the compounds as anticonvulsants (Table 11).
`This result is in agreement with previous findings dem-
`onstrating that stringent steric and electronic factors
`governed the anticonvulsant activities of this class of
`
`comp~unds?~~*~J Examination of the individual test results
`led to several important observations. First, reduction of
`the furan ring in 4 to the tetrahydrofuran analogues 21a
`and 21b led to a decrease, but not an abolition, of activity
`in the MES test (i-e., EDw < 90 mg/kg). The decreased
`activity of 21 versus 4 can be attributed to the loss of the
`aromatic ring at the a-carbon site, since previous findings
`have demonstrated that substantial improvement in
`activity accompanied the placement of a small aromatic
`group at this position? The potency of 21a and 21b was
`greater than that observed for 49 (EDm > 100 mg/kg).eb
`This observation provided support for our suggestion that
`increased anticonvulsant activity generally accompanied
`the placement of a substituted (alkylated) heteroatom two
`atoms removed from the amino acid a-carbon.7 Second,
`replacement of the a-carbon proton in 4 by a methyl group
`led to a sharp decrease in anticonvulsant activity of the
`drug candidate. This decreased potency in the MES test
`
`IPR2014-01126- Exhibit 1025 p. 5
`
`

`
`a-Acetamido-N-benzylacetamide Deriuatiues
`
`0 P o
`I
`II
`II
`CH,CNH-C--CNHCH,
`I
`H
`49 R2 = CHZOH
`a R2 = CH3
`
`CH3O
`0
`I ~ i i
`ii
`CH3CNH-C-CNHCH
`I
`CH3
`5l
`
`was surprising in light of the near equipotency previously
`observed for 50 (EDw = 51.0 rng/kgP versus 51 (EDm >
`40, < 100 mg/kg).27 Third, isosteric replacement of the
`amide carbonyl groups in 4 by a thioamide moiety resulted
`in decreased potency in the MES test. Of the two amide
`groups, modification of the benzylamide moiety (Le., 24)
`appeared to affect the MES activity more than modifi-
`cation of the acetamide site (i.e., 23). Fourth, alteration
`of the N-benzylamide group affected the pharmacological
`profile of the functionalized amino acid test candidate.
`Conversion of the N-benzylamide substituent in 4 to the
`3-pyridinylmethyl(25) or the corresponding N-oxide (27)
`led to small decreases in anticonvulsant activity, whereas
`the isomeric 4-pyridinylmethyl adduct (26) and N-oxide
`(28) were devoid of anticonvulsant activity at doses less
`than 100 mg/kg. Similarly, the 2'-pyridine hydrazide (29)
`displayed no protective effects in the MES test at doses
`of 100 mg/kg or less. Fifth, the pharmacological ste-
`reospecificity that distinguishes this novel class of anti-
`convulsant agenta4-6 was reaffirmed by the biological data
`obtained for (R)-30, (R)-31, and (R)-32. We noted a
`significant improvement in anticonvulsant activity of (R)-
`30 (EDw = 3.5 mg/kg) versus the corresponding racemate
`306 (EDm = 12.7 mg/kg). Moreover, the potency of (R)-30
`exceeded the value previously reported for phenytoin (EDw
`= 9.5 mg/kgh8
`Conclusions
`Synthetic protocols have been developed for the gen-
`eration of C(a)-heteroaromatic a-acetamido-N-benzylac-
`etamides. The pharmacological activities of these unique
`amino acid derivatives (i.e., 9-20) along with the modified
`analogues of a-acetamido-N-benzyl-a-(furan-2-yl)acet-
`amide (Le., 21-32) extended the structure-activity rela-
`tionships previously obtained for this class of anticon-
`vulsant agents."7 Significantly, the a-furan-2-yl (41,
`a-oxazol-2-yl (18), and a-thiazol-2-yl (19) a-acetamido-
`N-benzylacetamides afforded excellent protection to MES-
`induced seizures in mice. The observed EDw and PI values
`rivaled those reported for phenytoin.8 The experimental
`findings provided further documentation of the beneficial
`properties gained by the incorporation of aromatic groups
`at the C(a)-site and the importance of heteroatom location
`within the aromatic ring system for maximal biological
`activity. Protection against MES-induced seizures proved
`to be sensitive to changes at the C(a)-site in 2 and to
`modifications conducted at each of the other key functional
`groups in these compounds.
`Experimental Section
`Chemistry. General Methods. Melting points were deter-
`mined with a Thomas-Hoover melting point apparatus and are
`uncorrected. Infrared spectra (IR) were run on Perkin-Elmer
`
`Journal of Medicinal Chemistry, 1993, Vol. 36, No. 22 3355
`1330 and 283 spectrometers and were calibrated against the
`1601-cm-1 band of polystyrene. Absorption values are expressed
`in wavenumbers (cm-1). Proton (1H NMR) and carbon ('gc NMR)
`nuclear magnetic resonance spectra were taken on Nicolet NT-
`300 and General Electric QE-300 NMR instruments. Chemical
`shifts (6) are in parts per million (ppm) relative to Mersi, and
`coupling constants (J values) are in hertz. Low-resolution mass
`spectra (MS) were recorded at an ionizing voltage of 70 eV with
`a Varian MAT CH-5 spectrometer at the Lilly Research Lab-
`oratories. High-resolution electron-impact mass spectra were
`performed on a VG ZAB-E instrument by Dr. M. Moini at the
`University of Texas-Austin. Microanalyses were provided by
`the Physical Chemistry Department of the Lilly Research
`Laboratories. Ethyl acetamidocyanoacetate and Lawesson's
`reagent [2,4bis(4methoxyphenyl) - 1,3-dithia-2,4diphosphetane
`2,CdisulfideI were obtained from Aldrich Chemical Co., Md-
`waukee, WI. Thin-layer chromatography was performed on
`precoated silica gel GHLF microscope slidea (2.5 x 10 cm; Analtech
`No. 21521).
`Synthesis of a-Acetamido-N-benzy1-a-(imidazo1-2-y1)-
`acetamide (9). n-BuLi (2.5 M in hexane, 6.8 mL, 17.0 m o l )
`was added to a cooled (-46 "C) solution of 3S10 (2.90 g, 17.06
`mmol) in THF (45 mL) under Nz, and then stirred at -46 "C (15
`min). The lithio salt solution of 35 was then added dropwise (15
`min) into a cooled (-78 "C) THF solution (130mL) of 33 (prepared
`from a-acetamido-N-benzyl-a-eth~xyacetamide~~
`(2.00 g, 8.0
`mmol) and BBra (1 M in CH&12, 10 mL, 10.0 mmol)).' The
`reaction was stirred at -78 "C (1 h) and then quenched with a
`saturated aqueous NH4Cl (50 mL) solution. The mixture was
`stirred at room temperature (30 min) and made basic (pH 9.2)
`with aqueous KaCOa. The aqueous mixture was extracted with
`EtOAc (3 X 100 mL), and the combined extracts were dried (Nar
`SO4). The solvents were removed in uacuo, and the residue was
`purified by flash column chromatography on Si02 gel (2.5%
`MeOH/CHCb) to give 0.14 g (7%) of 9: mp 228-230 "c
`(recrystallized from EtOH); Rf 0.46 (10% MeOH/CHCb); IR
`(KBr) 3200 (br), 1610,1500 (br), 1430, 1350,740,680 cm-I; lH
`NMR (DMSO-de) 6 1.91 (8, C(O)CHa), 4.29 (d, J = 5.6 Hz, CH:),
`5.51 (d, J = 7.7 Hz, CH), 6.85 (br s, C4H), 7.05 (br 8, CsH), 7.18-
`7.30 (m, 5 PhH), 8.42 (d, J = 7.7 Hz, NH), 8.65 (t, J = 5.6 Hz,
`NH), 11.91 (br 8, NH);
`NMR (DMSO-de) 22.49 (C(O)CHa),
`42.21 (CH2), 51.62 (CH), 126.60 (Cd'), 126.98 (2Ca'or ~CS'), 127.21
`(C4),128.09 (2C2' or 2Ca'), 128.32 (CS), 139.05 (Ci'), 143.74 (Ca),
`168.12 (C(O)NH), 169.30 (C(0)CHa) ppm; maas spectrum, FD
`(relative intensity), 273 (M+ + 1, 65), 272 (M+, 100). Anal.
`(cdld%02) C, H, N.
`Synthesis of a-Acetarnido-N-benzyl-a-(imidazol-4-y1)-
`acetamide (10). A 75% aqueous EtOH (16 mL) solution of 37
`and the solution was heated to reflux (8 h). The reaction was
`neutralized with a saturated aqueous NaHCOa solution and the
`EtOH-Ha0 azeotrope removed by distillation in uacuo. The
`remaining aqueous layer was made basic (pH 10) with aqueous
`NaOH. The aqueous mixture was extracted with EtOAc (3 X 50
`mL), and the combined extracts were dried (NaaSO4). The
`reaction mixture was concentrated in Vacuo to give 0.35 g (57 %)
`of 10 mp 189-191 "C dec (recrystallized from acetone); Rf 0.19
`(10% MeOH/CHCla); IR (KBr) 3400, 3260, 1650, 1600, 1500,
`1430,1360,1330,730,710 cm-l; lH NMR (DMSO-dd 6 1.88 (8,
`C(O)CH*), 4.28 (d, J = 5.9 Hz, CHI), 5.38 (d, J = 6.8 Hz, CH),
`6.97 (br 8, CsH), 7.15-7.30 (m, 5 PhH), 7.60 (8, CzH), 8.26 (br 8,
`NH), 8.53 (br 8, NH), 12.01 (br 8, NH); W NMR (CDaOD) 22.45
`(C(O)CHs), 44.15 (CHa), 127.88 (C6 or c#), 128.01 (C49 or c3,
`128.37 (2Cr or 2C,*), 129.44 (2C+ or 2Cr), 136.88 (C:), 139.74
`(Cp), 172.13 (C(O)NH), 173.00 (C(0)CHa) ppm (a weak signal
`was observed at b 54 and has been tentatively attributed to CHI;
`mass spectrum, FD, 273 (M+ + 1). Anal. ( C ~ ~ H I ~ N ~ O Z )
`C, H, N.
`Synthesis of a-Acetamido-N-benzyl-a-( 1,2,4-triazol-3-y1)-
`acetamide (11). An ethanolic solution (250 mL) of 348 (3.00 g,
`13.0 mmol), formic hydrazide (1.60 g, 26.0 mmol), and &Cos
`(6.00 g, 2.90 mmol) was heated at reflux (20 h). The reaction
`mixture was allowed to cool and filtered, and the solvent was
`removed in uacuo. The residue was purified by flash column
`chromatography on Si02 gel using 13% MeOH/CHCb as the
`eluant to give 1.40 g (40%) of the desired product. Compound
`11 was purified by recrystallization from EtOH: mp 205-207 "C;
`Rf 0.35 (16% MeOH/CHCb); IR (KBr) 3285,3080,2930,1690,
`
`(0.85 g, 3.05 mmol) was acidified (pH - 1.5) with ethanolic HCl,
`
`IPR2014-01126- Exhibit 1025 p. 6
`
`

`
`3356 Journal of Medicinal Chemistry, 1993, Vol. 36, No. 22
`1650,1510 cm-l; lH NMR (DMSO-de) 6 1.92 (8, C(O)CH,), 4.30
`( d , J = 5.7 Hz, CHI), 5.62 ( d , J = 7.8 Hz, CHI, 7.18-7.32 (m, 5
`PhH), 8.53 (8, CsH), 8.56 (d, J 7.8 Hz, NH), 8.71 (t, J = 5.7
`Hz, NH), 13.98 (8, NH); '*C NMR (DMSO-de) 22.48 (C(O)CHs),
`42.41 (CHz), 51.30 (CH), 126.63 (Cd'), 127.08 (2C;'or2Ca'), 128.11
`(2Cz' or ~CS'), 139.05 (Cl'), 167.92 (C(O)NH), 169.32 (C(0)CHs)
`ppm (the two triazole carbon signals were not detected); mass
`spectrum, FD (relative intensity), 274 (M+ + 1, 1001, 273 (66).
`Anal. (CisHisNaOz) C, H, N.
`Synthesis of a-Acetamido-N-benzyl-a-(tetrazol-6-y1)-
`acetamide (12). A mixture of 34 (1.00 g, 4.33 mmol), KNs (1.70
`g, 20.96 mmol), and EtSN.HC1 (1.78 g, 13.0 mmol) in l-methyl-
`2-pyrrolidinone (125 mL) was stirred at 110 "C (7 h). After
`cooling, aqueous concentrated HCl(1 mL) was added, and the
`reaction mixture was filtered. The solvent was removed in uacuo