was 52 f 2 nmol of O2 min-' mg-l.
`2. Enzyme Assays. Enzyme activity was determined by
`measuring oxygen consumption in solution with a Yellow Springs
`polarographic electrode (Clark oxygen electrode, YSI-5331) in
`conjunction with a Gilson oxygraph (Model K-1C) as previously
`de~cribed.~' The YSI reaction chamber was modified by uni-
`formly rounding the bottom to permit smaller sample volumes.
`The temperature of the reaction chamber was maintained a t 37
`& 1 "C with a Haake FG water circulator. The enzyme reaction
`was initiated by the addition of sodium arachidonate solution (5
`
`(37) Rome, L. H.; Lands, W. E. M. Prostaglandins 1975, 10, 813.
`
`567
`
`J. Med. Chem. 1987,30,567-574
`mg/mL; Nu-Check Preps, Elysian, MN) to provide a 100 WM final
`concentration (K, = 5.9 pM)30 in the 2-mL reaction chamber. All'
`inhibitors were added as 10 or 100 mM solutions in Me2S0 for
`enzymes assays. The levels of Me,SO used in the enzyme-in-
`hibition experiments had no effect on enzyme activity. Initial
`enzyme velocities (dO,/dt) were obtained by measuring the slopes
`of the resulting oxygen concentration vs. time curves and are
`reported as a percent of uninhibited control.
`Acknowledgment. We gratefully acknowledge the
`partial support of this work through a National Institute
`of Health Pre-doctoral Training Grant Award to J.L.V. for
`the period 1984-1985.
`
`Functionalized DL-Amino Acid Derivatives. Potent New Agents for the Treatment
`of Epilepsy
`
`Judith D. Conley' and Harold Kohn*
`Department of Chemistry, University of Houston-University Park, Houston, Texas 77004. Received September 17, 1986
`
`Structural analogues of the potent known anticonvulsant agent N-acetyl-DL-alanine N-benzylamide (la) have been
`prepared (16 examples). The pharmacological activities of these products were evaluated in the maximal electroshock
`seizure (MES), the subcutaneous pentylenetetrazole seizure threshold (sc Met), and the rotorod (Tox) tests. The
`median effective doses (ED50) and the median toxic doses (TD50) for the most active compounds by both intra-
`peritoneal and oral administration are reported. The most active compounds were N-acetyl-DL-phenylglycine
`N-benzylamide ( l a ) and N-acetyl-DL-alanine N-m-fluorobenzylamide (lm) along with the parent compound la.
`The ED50 values in the MES test for these three compounds compared well with phenobarbital, while their high
`TD50 values contributed to their large protective indexes, which approached that of phenytoin. When tested against
`four convulsant agents, compounds l a and Id displayed activity profiles significantly different from those reported
`for conventionally used antiepileptic drugs.
`
`Amino acids and their derivatives have not had a sig-
`nificant impact in the development of new agents for the
`treatment of epilepsy. The lack of interest in amino acid
`type compounds stems from the inability of many of these
`polar compounds to readily penetrate the blood-brain
`barriers2 Despite this concept, several types of amino acids
`and their derivatives have demonstrated the ability to
`prevent chemically, audiogenically, and photically induced
`seizures. These include derivatives of alicyclic and aro-
`matic amino acids,3 phosphono derivatives of aliphatic
`amino acids,4 N-benzoyl- and N-phenylacetylglycine
` amide^,^ and structural analogues of the inhibitory neu-
`rotransmitter y-aminobutyric acid (GABA).6 The en-
`
`dogenous neuropeptides Met- and Leu-enkephalin have
`also exhibited anticonvulsant activity in a variety of test
`animals and may play an important role in the prevention
`of a static convulsive state or in the maintenance of normal
`brain f ~ n c t i o n . ~
`Inspection of chemotherapeutic agents possessing central
`nervous system (CNS) depressant and anticonvulsant
`activity reveals a common structural pattern (Figure 1).
`Three functionalities are prevalent in many of these com-
`pounds: (1) a vicinal diamine linkage, (2) an oxygen atom
`on the ethylene chain bridging the two amino groups, and
`(3) an aromatic ring one carbon removed from an amino
`residue.8 Representatives of this structural design are
`
`(1) Abstracted from the Ph.D. dissertation of this author. Addi-
`tional structure proof, discussion, and experimental and spec-
`tral data may be found in this reference.
`(2) Callery, P. S.; Geelhaar, L. A.; Nayar, M. S. B.; Stogniew, M.;
`Rao, K. G. J. Neurochem. 1982,38, 1063-1067.
`(3) Alexander, G. J.; Kopeloff, L. M. Neuropharmacology 1977,
`16, 405-409. Crider, A. M.; Tita, T. T.; Tschappart, K. D.;
`Hinko, C. N.; Seibert, K. J. Pharm. Sci. 1984, 73,1612-1616.
`Hinko, C. N.; Seibert, K.; Crider, A. M. Neuropharmacology
`1984,23, 1009-1014. Crider, A. M.; Tita, T. T.; Wood, J. D.;
`Hinko, C. N. J. Pharm. Sci. 1982, 71,1214-1219. Yunger, L.
`M.; Fowler, P. J.; Zarevics, P.; Setler, P. E. J. Pharmacol. Exp.
`Ther. 1984, 228, 109-115. Beitz, A. J.; Larson, A. A. Eur. J.
`Pharmacol. 1985,114, 181-187. Matsui, Y.; Deguchi, T. Life
`Sci. 1971, 20, 1291-1296. Erez, U.; Frenk, H.; Goldberg, 0.;
`Cohn, A.; Teichberg, V. I. Eur. J. Pharmacol. 1985,110,31-39.
`(4) Croucher, M. J.; Collins, J. E.; Meldrum, B. S. Science
`(Washington, D.C.) 1982,216,899-901. Cates, L. A.; Li, V-S.;
`Hu, Z-S.; Lehmann, J.; Coyle, J. T.; Ferkany, J. W. J. Pharm.
`S C ~ . 1984, 73, 1550-1553.
`(5) Takahashi, T.; Ogui, K.; Fujimura, H.; Satoda, I.; Fukui, T.;
`Yamamato, Y. Swiss Patent 393 355, Oct 30,1965. Thorne, D.
`E. US. Patent 3 657 341, April 18, 1972.
`
`(6) Krogsgaard-Larsen, P.; Scheel-Kruger, J.; Kofod, H. GABA-
`Neurotransmitters, Pharmacochemical, Biochemical, and
`Pharmacological Aspects; Academic: New York, 1979. Enna,
`S. J. Biochem. Pharrnacol. 1981,30,907-913. Scheechter, P.
`J.; Tranier, Y.; Jung, M. J.; Sjoerdsma, A. J. Pharmacol. Exp.
`Ther. 1977,201,606-612. Scheechter, P. J.; Tranier, Y.; Jung,
`M. J.; Bohlen, P. Eur. J. Pharmacol. 1977,45, 319-328. Gal-
`zigna, L.; Garbin, L.; Bianchi, M.; Marzotto, A. Arch. Znt.
`Pharmacodyn. 1978,235,73-85. Chambon, J. P.; Molimard,
`J. C.; Calassi, R.; Maruani, J.; Rodier, D.; Sigault, G.; Legris,
`R.; Roncucci, R.; Bizisre, K. Arzneim.-Forsch. 1984, 34,
`1017-1021.
`(7) Zetler, G. Eur. J. Pharmacol. 1980,65,297-300. Meldrum, B.
`S.; Menini, C.; Stutzmann, J. M.; Naquet, R. Brain Res. 1979,
`170,333-348. Tortella, F. C.; Cowan, A.; Adler, M. W. Life Sci.
`1981,29, 1039-1045. Tortella, F. C.; Cowan, A. Life Sci. 1982,
`31, 2225-2228.
`(8) Other models have been proposed. For a comprehensive dis-
`cussion as well as review, see: Camerman, A.; Camerman, N.
`In Antiepileptic Drugs: Mechanism of Action; Glaser, G. H.,
`Penry, J. K., Woodbury, D. M., Eds.; Raven: New York, 1980;
`pp 223-231. Wong, M. G.; Defina, J. A,; Andrews, P. A. J.
`Med. Chem. 1986,29, 562-572.
`
`0022-2623/87/1830-0567$01.50/0 0 1987 American Chemical Society
`
`IPR2014-01126- Exhibit 1017, p. 1
`
`

`
`568 Journal of Medicinal Chemistry, 1987, Vol. 30, No. 3
`
`Conley and Kohn
`
`Table I. Phase I Pharmacological Evaluation of Functionalized DL-Amino Acid Derivatives 1"
`
`0
`II
`R ' - C N H ( C H ),,-
`
`R 2 0
`I
`II
`Ce C N H R
`I
`
`R 3
`
`1
`
`,
`
`R2
`
`Tox:
`sc Met:
`MES,c
`no.
`0.5 h
`0.5 h
`0.5 h
`ASPb
`R4
`R3
`R'
`n
`H
`CH,
`CH3
`la
`0
`I
`Bnf
`0
`3
`1
`H
`H
`I1
`Bn
`2
`0
`CH3
`2
`lb
`0
`H
`I1
`Bn
`CH(CH3)z
`CH,
`2
`0
`0
`0
`I C
`H
`I
`Bn
`CH3
`Id
`4
`0
`0
`2
`C6H5
`I1
`Bn
`H
`CH,CH2SCH3
`CH3
`le
`0
`2
`0
`0
`I11
`Bn
`H
`Bn
`CH3
`If
`0
`0
`0
`0
`I11
`Bn
`CH3
`1g
`0
`0
`0
`0
`C6H5
`C6H5
`I11
`Bn
`H
`H
`CH3
`lh
`0
`0
`0
`1
`I1
`Bn
`H
`CH3
`CH3
`li
`0
`0
`2
`1
`"The following code has been adopted: 0 = no activity at dose levels of 600 mg/kg; 1 = noticeable activity at dose levels of 600 mg/kg; 2
`= noticeable activity at dose levels of 300 mg/kg; 3 = noticeable activity at dose levels of 100 mg/kg; 4 = noticeable activity at dose levels
`of 30 mg/kg. bASP Results Classification. MES = maximal electroshock seizure test. d~~ Met = subcutaneous pentylenetetrazole (Me-
`trazol) seizure test. 'Tox = neurologic toxicity (the rotorod test). fBn = benzyl.
`
`Table 11. Phase I Pharmacological Evaluation of Functionalized DL-Amino Acid Derivatives ln
`0
`R 2 0
`I1
`I
`I1
`I
`
`R'-CNH(CH*),,-
`
`Ca-CNHR4
`
`R 3
`
`1
`
`R4
`
`Tox,~
`sc Met,d
`MES,"
`R2
`R'
`no.
`0.5 h
`0.5 h
`0.5 h
`ASPb
`R3
`n
`H
`CH3
`CH3
`I
`Bnf
`la
`0
`0
`3
`1
`CH,
`CH3
`CH,
`H
`111
`0
`1j
`0
`0
`0
`CH3
`I11
`CH(CeH&
`H
`0
`0
`0
`CH3
`lk
`0
`I1
`CH3
`CH3
`CH&sH,-m-OCH,
`H
`2
`0
`0
`11
`0
`I
`CHZCGH4-m-F
`H
`3
`0
`0
`CH3
`CH3
`lm
`0
`In
`I11
`CHoCONHBn
`H
`CH?
`CH,
`0
`0
`0
`0
`"The following code has been adopted: 0 = no activity at dose levels of 600 mg/kg; 1 = noticeable activity at dose levels of 600 mg/kg; 2
`= noticeable activity at dose levels of 300 mg/kg; 3 = noticeable activity at dose levels of 100 mg/kg; 4 = noticeable activity at dose levels
`of 30 mg/kg. bASP Results Classification. MES = maximal electroshock seizure test. d s ~ Met = subcutaneous pentylenetetrazole (Me-
`trazol) seizure test. e T ~ x = neurologic toxicity (the rotorod test). fBn = benzyl.
`
`Figure 1. Structural unit present in many anticonvulsants.
`
`substituted hydantoins, piperazines, and benzodiazepines.
`Recognition of this empirical blueprint in anticonvulsant
`drugs led to the hypothesis that functionalized amino acids
`should provide a rich source for future antiepileptic agents.
`This rationale was supported by a recent report from this
`laboratory that N-aCetyl-DL-alanine N-benzylamide (la)
`displayed potent anticonvulsant a ~ t i v i t y . ~
`
`l a
`In this paper, the syntheses, physical properties, and
`anticonvulsant activities of a select series of functionalized
`amino acid derivatives are described. Evidence is pres-
`ented that these simple compounds comprise a new and
`
`~~
`
`~~~
`
`(9) Cortes, S.; Liao, Z-K.; Watson, D.; Kohn, H. J. Med. Chem.
`1985, 28, 601-606.
`
`novel class of antiepileptic agents.
`Selection of Compounds
`N-Acetyl-DL-alanine N-benzylamide (la) served as the
`parent compound in this study (Tables 1-111). Systematic
`structural variations have been conducted at three sites:
`the a-carbon (Table I), the amide substituent (Table 111,
`and the N-acyl group (Table 111). In the selection of de-
`rivatives, we have attempted to adhere to the molecular
`blueprint cited previously and to test the validity of the
`empirical relationship between this molecular pattern and
`anticonvulsant activity. In all cases, where appropriate,
`the DL racemates were synthesized.
`In compounds lb-f, the a-carbon substituent has been
`systematically changed from methyl to hydrogen to iso-
`propyl to phenyl to a thio alkyl group to benzyl, while in
`compound l g both a-carbon sites have been substituted
`Interestingly, l g represents an
`with phenyl groups.
`open-chained analogue of the potent antiepileptic pheny-
`toin. Remaining compounds in this first category included
`the 0-amino acid derivatives (lh and li) in which the
`a-carbon moiety has been hoinologated by one carbon
`atom.
`The second category of substituents were structural
`variants of la in which the amide group has been altered.
`Included in this list were the N-methylamide (lj), the
`N-benzhydrylamide (lk), and the two derivatives of la in
`
`IPR2014-01126- Exhibit 1017, p. 2
`
`

`
`Functionalized DL- Amino Acid Derivatives
`
`Journal of Medicinal Chemistry, 2987, Vol. 30, No. 3 569
`
`Table 111. Phase I Pharmacological Evaluation of Functionalized DL-Amino Acid Derivatives 1"
`
`1
`
`R'
`
`Tax:
`sc Met!
`MES,"
`R4
`R3
`R2
`no.
`0.5 h
`0.5 h
`0.5 h
`ASPb
`n
`Bnf
`H
`I
`3
`0
`1
`CH3
`CH3
`l a
`0
`Bn
`I11
`0
`0
`H
`0
`CH3
`(CHdzCH
`lo
`0
`Bn
`H
`I1
`2
`2
`0
`CH3
`(CH3)sC
`0
`1P
`I11
`0
`Bn
`H
`CH3
`CH3CONHCH2
`0
`0
`0
`1q
`"The following code has been adopted: 0 = no activity at dose levels of 600 mg/kg; 1 = noticeable activity at dose levels of 600 mg/kg; 2
`= noticeable activity at dose levels of 300 mg/kg; 3 = noticeable activity at dose levels of 100 mg/kg; 4 = noticeable activity at dose levels
`MES = maximal electroshock seizure test. d~~ Met = subcutaneous pentylenetetrazole (Me-
`ASP Results Classification.
`of 30 mg/kg.
`trazol) seizure test. e Tox = neurologic toxicity (the rotorod test). f Bn = benzyl.
`
`Table IV. Selected Physical and Spectral Data of DL-Amino Acid N-Substituted Amides 4
`R 2 0
`l a
`II
`H2N- C -CNHR4
`
`A 3
`4
`
`13C NMRd
`'H NMR,dje
`a-c
`c=o
`CY-CH
`IRc
`mpb
`yielda
`R4
`R3 (R2)
`R2 (R3)
`no.
`H
`3.50 (9, 6.0, 1 H)
`oil
`50.4
`174.3
`Bnf
`1655,1525g
`40
`CH3
`4a
`3.15 (9, 2 H)
`44.7
`173.1
`1630, 1545 (br)
`Bn
`H
`44-48"
`36
`4b
`H
`3.17 (d, 4.0, 1 H)
`1650 (br), 1515
`oilh
`Bn
`H
`60.2
`174.6
`55
`(CH3)ZCH
`4~
`4.13 (9, 1 H)
`59.3
`173.3
`1670 (br), 1520 (br)
`oil
`82
`Bn
`H
`CsHS
`4d
`3.49 (dd, 8.5, 4.4, 1 H)
`56.4
`174.1
`Bn
`H
`1655, 1535
`64-65
`Bn
`4e
`60
`3.47 (9, 6.9, 1 H)
`1650 (br), 1550 (br)
`oil"
`CHs
`H
`50.7
`176.9
`33
`CH,
`4f
`nPurified yields (%) from the methyl ester hydrochloride 3. All compounds gave satisfactory analyses for C, H, N (f0.470) unless
`otherwise indicated. *Melting points ("C) are uncorrected.
`Infrared peak positions are recorded in centimeters (cm-') vs. the 1601-cm-'
`band in polystyrene. Solids were taken in KBr disks and oils were taken neat (NaC1). dNMR spectra were taken in CDC13 (in 6). eThe
`information in parentheses is the multiplicity of the signal, followed by the coupling constant in hertz (Hz), followed by the number of
`protons attributed to the signal. f Bn = benzyl. BReference 9. Elemental composition was verified by high-resolution mass spectroscopy,
`
`Scheme I
`
`which an electron-donating (11) or an electron-withdrawing
`(lm) group has been placed in the meta position of th;!
`aromatic ring. The remaining member of this class of
`compounds was the dipeptide In in which the amide
`substituent has been extended by a glycyl moiety.
`In the next group of drug candidates, the N-acyl sub-
`stituent in la has been modified. Compounds selected for
`synthesis included the dimethyl- and the trimethylacetyl
`derivatives (10 and lp) of la. As the amide substituent
`of la was extended with a glycyl moiety in the diDeDtide
`- - -
`In, the N-acyl group was lengthened with a glycyl group
`in the dipeptide 1% Of note, dipeptides In and 1q are
`isomeric.
`
`Chemistry
`The strategies employed in the synthesis of the racemic
`function&zed amino acid derivatives were patterned after
`procedures common to peptide synthesis.1° Two general
`methods were utilized for the preparation of these com-
`pounds as depicted in Scheme 1, No major effort was
`made to optimize the yields.
`In the first procedure (method A), the starting DL-amino
`acid 2 was initially converted to the corresponding methyl
`
`(io) Bodanszky, M.; Klausner, Y . S.; Ondetti, M. A. Peptide Syn-
`thesis, 2nd ed.; Wiley: New York, 1976.
`
`IPR2014-01126- Exhibit 1017, p. 3
`
`

`
`570 Journal of Medicinal Chemistry, 1987, Vol. 30, No. 3
`
`Conley and Kohn
`
`Table V. Selected Physical and Spectral Data of N-Acyl-DL-amino Acid N-Substituted Amides and Their Analogues 1
`R 2 0
`0
`I/
`l a
`/I
`C - NH - (CH 2)n-
`C - C - NH R 4
`I
`I
`R 3
`
`R'-
`
`1
`
`R'
`
`no. n
`la 0 CH3
`l b 0 CH3
`IC 0 CH3
`Id 0 CH3
`le 0 CH3
`
`If 0 CH3
`lg 0 CH3
`lh 1 CH3
`li 1 CH3
`l j 0 CH3
`l k 0 CH3
`11 0 CH3
`
`l m 0 CH3
`
`I n 0 CH3
`
`R2 (R3) R3 (R2)
`H
`CH3
`H
`H
`CH(CH3)2 H
`H
`C6H5
`CH2CH2- H
`SCH3
`Bn
`C6H5
`H
`CH3
`CH3
`CH3
`CH3
`
`CH3
`
`CH3
`
`H
`
`H
`
`R4
`
`Bng
`Bn
`Bn
`Bn
`Bn
`
`IRd
`method" yieldb mpC
`A
`60
`138-139 1660, 1515h
`A
`81
`140-142 1640, 1545
`A
`86
`192-193 1625, 1535
`A
`66
`202-203 1635, 1540
`B
`43
`134-135 1630, 1545
`
`I3C NMR'
`'H NMR,','
`R'-CO a-C C,-CO
`a-CH
`4.26-4.34 (m, 1 H) 168.9 48.2 172.3
`3.74 (d, 5.3, 2 H)
`168.9 42.4 172.3
`4.11 (d, 8.8, 1 H)
`169.2 57.8 171.1
`5.50 (d, 7.9, 1 H)
`168.9 56.3 169.9
`4.10-4.53 (m, 1 H) 169.5 52.0 171.4
`
`A
`B
`B
`B
`A
`B
`B
`
`B
`
`B
`
`84
`72'
`27
`79
`90
`67
`69
`
`54
`
`47
`
`161-162 1630, 1545
`189-190 1645, 1530
`166-167 1640, 1545
`130-131 1640, 1560
`158-159 1635, 1565
`193-194 1635, 1540
`111-113 1630, 1540
`
`4.36-4.72 (m, 1 H) 169.0 54.1 171.2
`169.2 68.8 170.6
`2.40 (t, 6.5, 2 H)
`169.3 35.4 170.4
`2.58 (q, 6.9, 1 H)
`169.5 42.0 174.3
`3.95-4.48 (m, 1 H) 169.1 48.2 172.8
`4.30-4.60 (m, 1 H) 169.0 48.1 171.8
`4.24-4.37 (m, 1 H) 169.1 48.3 172.5
`
`120-121 1645, 1545
`
`4.23-4.41 (m, 1 H) 169.6 48.5 172.8
`
`185-186 1685, 1640, 1545 3.93-4.38 (m, 1 H) 168.7 48.7
`
`172.8
`
`P
`Ph
`7
`Pharmacological Evaluation
`All N-acyl-DL-amino acid N-substituted amides 1 pre-
`pared in this study were submitted to the National In-
`stitutes of Health Antiepileptic Drug Development Pro-
`gram for pharmacological evaluation. Each compound was
`tested for anticonvulsant activity by using the procedures
`described by Krall et al.14
`The phase I test results are summarized in Tables 1-111.
`All compounds were administered intraperitoneally at
`three doses (30,100, and 300 mg/kg). The only exception
`was the parent compound la, which was evaluated at 600
`mg/kg as well. The smallest dose that produced activity
`was noted for separate tests involving maximal-induced
`
`H
`Bn
`C & , Bn
`H
`Bn
`H
`Bn
`H
`CH3
`H
`CH(CeH5)Z
`H
`CHzCsH4-
`m-OCH3
`CHzCsHd-
`m-F
`CHZCON-
`HBn
`H
`CH3
`4.03-4.48 (m, 1 H) 175.9 48.0 172.4
`10 0 (CH3)ZCH
`A
`40
`Bn
`164-165 1635, 1545
`4.23-4.42 (m, 1 H) 177.1 48.4 172.5
`H
`CH3
`I P 0 (CH&C
`40
`A
`123-124 1630, 1535
`Bn
`6 9 184-186 1685, 1640, 1545 4.00-4.18 (m, 1 H) 169.5 48.2 172.1
`H
`lq 0 CH3CONHCHz CH,
`B
`Bn
`"Compounds by method A were prepared from the DL-amino acid N-substituted amides, while those by method B from the N-acyl-DL-amino acids.
`All compounds gave satisfactory analyses for C, H, N (+0.4%) unless otherwise indicated. *The purified yields (a) are from the DL-amino acid
`N-substituted amides 4 for compounds synthesized by method A and from the N-aCetyl-DL-aminO acids 5 for compounds prepared by method B unless
`Melting points ("C) are uncorrected.
`Infrared peak positions are recorded in reciprocal centimeters (cm-') vs. the 1601-cm-'
`otherwise indicated.
`band in polystyrene and were taken in KBr disks. "11 NMR spectra were taken in MezSO-ds (in 8). /The information in parentheses is the
`multiplicity of the signal, followed by the coupling constant in hertz (Hz), followed by the number of protons attributed to the signal. gBn = benzyl.
`Reference 9. The yield is from N-acetyldiphenylglycine. Elemental composition was verified by high-resolution mass spectroscopy.
`ester hydrochloride 3 by the addition of SOClz to a sus-
`isolated but directly treated in situ with 1.1 equiv of the
`appropriate amine at -5 "c to produce the N-acyl-DL-
`pension of the DL-amino acid in cold methanol.ll Yielas
`amino acid N-substituted amides 1 in 43-79% yields.
`were quantitative, and no further purification was required.
`Compounds le, lh, li, lk-n, and lq were prepared by this
`Conversion of the DL-amino acid methyl ester hydro-
`route. N-Acetyldiphenylglycine N-benzylamide (lg) was
`chloride 3 to the corresponding N-substituted amide 4 yas
`synthesized by a slightly different method. Diphenyl-
`accomplished with an excess of the appropriate amine. At
`least 2 equiv of the amine was used in order to permit the
`glycine or N-acetyldiphenylglycine was heated with acetic
`isolation of 4 as the free base. Yields ranged from 33%
`anhydride at reflux (5-30 min) to give the oxazolone in-
`termediate 7 (80%).13 Treatment of the oxazolone with
`to 82% (Table IV). Reaction of the DL-amino acid N-
`benzylamine yielded product lg in 90% yield.
`substituted amide 4 with an acid anhydride or an acid
`halide produced the final product 1 in yields from 28%
`to 90% (Table V). This procedure was employed in the
`synthesis of la-d, If, lj, lo, lp, lr, and 1s.
`Low yields were encountered in the conversion of several
`DL-amino acid methyl ester hydrochlorides 3 to the cor-
`responding m-amino acid N-substituted amides 4, ne-
`cessitating the use of method B. In this route, the DL-
`amino acid or dipeptide 2 was initially acylated with the
`acid anhydride (1.1-3.0 equiv in refluxing acetic acid,
`dichloromethane, or water) to give the N-protected DL-
`amino acid 5 in moderate to high yields (56-99%).12
`Protection of the amino terminus permitted the subse-
`quent reaction with triethylamine and ethyl chloroformate
`to proceed at the terminal carboxyl group to generate the
`mixed N-acyl-DL-amino acid-carbonic ester anhydride
`intermediate 6. The activated mixed anhydride was not
`
`Y O "
`N T
`
`h
`
`
`
`~~~ (11) Brenner, M.; Huber, W. Helu. Chim. Acta 1953,36,1109-1115.
`Greenstein, J. P.; Winitz, M. Chemistry of the Amino Acids;
`Wiley: New York, 1961.
`(12) For N-acetyl-DL-a-amino acids, see: Greenstein, J. P.; Winitz,
`M. Chemistry o i the Amino Acids, Wiley: New York, 1961.
`For N-acetyl-DL-/?-amino acids, see: Fodor, P. J.; Price, V. E.;
`Greenstein, J. P. J. Biol. Chem. 1949, 178, 503-509.
`
`(13) Hohenlohe-Oehringen, K. Monatsch. Chem. 1962,93,639-644.
`(14) Krall, R. L.; Penry, J. K.; White, B. G.; Kupferberg, H. J.;
`Swinyard, E. A. Epilepsia 1978, 19, 409-428.
`
`IPR2014-01126- Exhibit 1017, p. 4
`
`

`
`Functionalized m-Amino Acid Derivatives
`
`Journal of Medicinal Chemistry, 1987, Vol. 30, No. 3 571
`
`Table VI. Phase I1 Pharmacological Evaluation of Functionalized DL-Amino Acid Derivatives 1"
`T O X , ~
`MES,b
`sc Met:
`ED50
`PIe
`TD50
`ED50
`compound
`76.54 (66.58-89.04)
`N-acetyl-DL-alanine N-benzylamide (la)
`5.93
`453.86 (416.56-501.01)
`f
`4.77
`96.92 (79.80-118.39)
`20.31 (16.85-24.45)
`N-acetyl-DL-phenylglycine N-benzylamide (la)
`f
`>6.46
`142.73 (61.53-237.97)
`77.38 (62.55-91.01)
`g
`N-aCetyl-DL-ahine N-rn-fluorobenzylamide (lm)
`6.89
`65.46
`9.50
`i
`phenytoinh
`3.17
`69.01
`13.17
`21.78
`phenobarbitalh
`<0.44
`440.83
`130.35
`ethosuximideh
`j
`1.57
`425.84
`148.59
`271.66
`valproateh
`" ED50 and TD50 are in mg/kg. Numbers in parentheses are 95% confidence intervals.
`sc
`MES = maximal electroshock seizure test.
`Met = subcutaneous pentylenetetrazole (Metrazol) seizure test. d T o ~ = neurologic toxicity (the rotorod test). e PI = protective index
`(TDBO/MES ED50). !The ED50 value was not computed for this substrate. gNo toxicity observed up to 500 mg/kg. hReference 15. &Not
`effective. 'No activity observed up to 1000 mg/kg.
`
`Table VII. Phase IV Pharmacological Evaluation of Functionalized DL-Amino Acid Derivatives 1"
`sc Met:
`MES,b
`PP
`ED50
`ED50
`comvound
`>8.15
`122.68 (106.82-138.40)
`266.29 (242.19-289.75)
`N-aCetyl-DL-alanine N-benzylamide (la)
`f
`5.17
`241.38 (194.39-284.08)
`N-acetyl-DL-phenylglycine N-benzylamide (la) 46.71 (30.76-76.40)
`g
`9.59
`86.71 (80.39-96.09)
`phenytoinh
`9.04 (7.39-10.62)
`1
`4.82
`96.78 (79.88-115.00)
`phenobarbitalh
`20.09 (14.78-31.58)
`12.59 (7.99-19.07)
`ethosuximideh
`<0.44
`879.21 (839.89-933.51)
`192.21 (158.59-218.44)
`j
`1.90
`1264.39 (800-2250)
`388.31 (348.87-438.61)
`664.80 (605.33-718.00)
`valvroateh
`"ED50 and TD50 are in mg/kg. Numbers in parentheses are 95% confidence intervals. bMES = maximal electroshock seizure test. csc
`Met = subcutaneous pentylenetetrazole (Metrazol) seizure test. d T o ~ = neurologic toxicity (the rotorod test). e PI = protective index
`(TD50/MES ED50). fNo toxicity was observed for doses up to 1000 mg/kg. gThe ED50 value was not computed for this substrate.
`protection up to 300 mg/kg. jNo protection up to 2000 mg/kg.
`hReference 15.
`
`T o x , ~
`TD50
`
`Table VIII. Phase V Pharmacological Evaluation of Functionalized DL-Amino Acid Dervatives 1"
`sc Bic,C
`sc pic:
`sc Met,b
`ED50
`ED50
`ED50
`204.66 (157.50-286.49) 133.61 (115.92-153.22) g
`g
`g
`g
`
`f
`g
`
`sc Strych,e
`ED50
`
`compound
`N-aCetyl-DL-alanine N-benzylamide (la)
`N-acetyl-DL-phenylglycine
`N-benzylamide (la)
`g
`phenytoinh
`i
`g
`g
`13.17 (5.87-15.93)
`37.72 (26.49-47.39)
`27.51 (20.88-34.82)
`95.30 (91.31-99.52)
`phenobarbitalh
`130.35 (110.99-150.45) 459.01 (349.92-633.13) 242.69 (227.84-255.22)
`ethosuximideh
`j
`148.59 (122.64-177.02) 359.95 (294.07-438.54) 387.21 (341.37-444.38) 262.96 (261.12-323.43)
`valproateh
`"ED50 and TD50 are in mg/kg. Numbers in parentheses are 95% confidence intervals. b s ~ Met = subcutaneous Metrazol test (CD97 =
`Bic = subcutaneous bicuculline test (CD97 = 2.70 mg/kg). d~~ Pic = subcutaneous picrotoxin test (CD97 = 3.15 mg/kg).
`85 mg/kg).
`Strych = subcutaneous strychnine test (CD97 = 1.20 mg/kg). fMaximum protection: 50% at 800 mg/kg. gThe ED50 value was not
`computed for this substrate. Reference 15. 'Maximum protection: 50% at 55-100 mg/kg. jMaximum protection: 62.5% at 250-1000
`m g / k
`convulsions (MES), subcutaneous Metrazol-induced con-
`vulsions (sc Met), and a rotorod toxicity test (Tox). The
`overall effect of the three tests were then given by one of
`four different ratings (ASP Results Classification I-IV).
`Compounds with a rating of I were designated as promising
`and were considered for phase I1 (quantification) testing
`(Table VI). This stage involved the same tests previously
`described, except under stricter monitoring of dosages and
`activity time spans and included an evaluation of the
`median effective dose (ED50) and the median toxic dose
`(TD50). If the anticonvulsant activity of the test com-
`pound was satisfactory, the amino acid derivative was then
`subjected to phase IV and V trials. Phase IV entailed the
`same tests described for phase I and 11, except the test
`compound was administered to mice orally (Table VII).
`The in vivo antiepileptic potential was further delineated
`in phase V (antiepileptic drug differentiation in mice), and
`the results are summarized in Table VIII. Phase V ex-
`amined the ability of the drug candidate to protect mice
`against seizures induced by a CD97 subcutaneous injection
`of Metrazol, bicuculline, picrotoxin, and strychnine. These
`convulsants have CD97 values of 85, 2.70, 3.15, and 1.20
`mg/kg, re~pective1y.l~
`
`Table I lists the pharmacological phase I results of the
`parent compound la and those analogues where only the
`a-carbon moiety has been modified. Evaluation of this set
`of results revealed several significant observations. First,
`the principal biological activity of these compounds resided
`in their ability to prevent seizures in the MES test. Sec-
`ond, reduced CNS activity was noted as the size of the
`substituent on the a-carbon atom in la was decreased from
`a methyl group to a hydrogen (lb) or increased to either
`an isopropyl group (IC) or a thio alkyl group (le). Each
`of these analogues possessed anticonvulsant activity but
`were not as effective as the parent compound. Third,
`pronounced activity was observed for N-acetyl-DL-
`phenylglycine N-benzylamide (la). Of note, N-acetyl-
`DL-phenylglycine N-benzylamide (la) contains two aro-
`matic rings both of which are one carbon atom removed
`from an amino residue. Fourth, loss of activity was ob-
`served when the phenyl group of Id was extended by a CH2
`to a benzyl moiety (If) and when both substituents on the
`a-carbon atom of the parent compound were replaced with
`phenyl groups (lg). Fifth, homologation of la and lb
`giving the corresponding p-alanine derivatives li and lh,
`respectively, led to reduced CNS activity. Both com-
`pounds possess a 1,3-diamine linkage.
`The second category of compounds tested for anticon-
`vulsant activity involved analogues of la where the benzyl
`moiety of the amide group was altered. The phase I results
`
`(15) Porter, R. J.; Cereghino, J. J.; Gladding, G. D.; Hessie, B. J.;
`Kupferberg, H. J.; Scoville, B.; White, B. G. Cleueland Clin.
`Q. 1984,51, 293-305.
`
`IPR2014-01126- Exhibit 1017, p. 5
`
`

`
`572 Journal of Medicinal Chemistry, 1987, Vol. 30, No. 3
`
`for these compounds are shown in Table 11. Assessment
`of these results revealed a significant pattern in the ability
`of these amino acid derivatives to prevent seizures. The
`anticonvulsant activity in the MES test was abolished
`when the size of the benzyl substituent in la was decreased
`to a methyl (lj) or increased to a benzhydryl (lk) or a
`glycine N-benzylamide (In) group. A similar trend can
`be ascertained by comparing the MES activities of N-
`acetyl-DL-phenylglycine N-benzylamide (ld) with N-
`acetyl-DL-phenylglycine N-rneth~lamide.~ Substitution on
`the benzyl group of the amide in la with either an elec-
`tron-donating (Le., m-methoxy) (11) or an electron-with-
`drawing (Le., m-fluoro) (lm) moiety led to compounds
`yielding different activities. Of the two, the latter analogue
`(lm) displayed higher activity in the MES test. Moreover,
`the toxicity of N-acetyl-DL-alanine N-m-fluorobenzylamide
`(lm) was lower than the parent compound, la. Finally,
`all drug candidates listed in Table 11 were devoid of activity
`in the sc Met test a t the administered doses.
`Table I11 lists the phase I results for la and its analogues
`where the N-acetyl moiety has been changed. Evaluation
`of this set of results revealed that any increase in the size
`of the methyl group of the N-acetyl moiety in la led to
`compounds that possessed decreased activity in the MES
`test. Furthermore, we note that compound lp, unlike the
`other derivatives examined, displayed significant activity
`in the sc Met test.
`To further probe the structural parameters needed for
`anticonvulsant activity, two of our most active compounds,
`la and Id, were compared to the known isomeric com-
`pounds, N-phenylacetyl-DL-alanine N-methylamide (lr)
`and N-phenylacetyl-DL-phenylglycine N-methylamide
`( l ~ ) . ~
`In this series, the methyl group of the N-acetyl
`moiety and the benzyl group of the N-benzylamide moiety
`have been interchanged. Transposition of these two ter-
`minal groups resulted in compounds that were devoid of
`anticonvulsant activity under the ADD screening protocol.
`The activity profiles exhibited by la and Id vs. l r and 1s
`are in agreement with our empirical molecular blueprint.
`The pharmacological activities for compounds Id and
`lm warranted their further evaluation in phase I1 trials.
`The data obtained from these studies are summarized in
`Table VI along with similar information for la and several
`proven antiepileptic drugs.16 The ED50 values in the MES
`test for N-acetyl-DL-phenylglycine N-benzylamide (ld) and
`N-acetyl-DL-alanine N-m-fluorobenzylamide (lm) com-
`pared well with phenobarbital. The high TD50 values of
`these amino acid analogues contributed significantly to
`their large protective indexes, which approach that of
`phenytoin.
`Phase IV tests provided information concerning the test
`candidate when administered orally (PO). The results
`observed for compounds la and Id are summarized in
`Table VI1 with similar data for currently used antiepileptic
`agents.15 The ED50 and the TD50 values disclosed both
`N-aCetyl-DL-alanine N-benzylamide (la) and N-acetyl-
`DL-phenylglycine N-benzylamide (ld) were less potent in
`the MES test and less toxic after oral administration than
`after intraperitoneal injection (Table VI). Significantly,
`compound la was effective in the sc Met test after oral
`administration but not after intraperitoneal injection
`(Table VI). A variety of phenomena (i.e., adsorption,
`metabolism) may be responsible for the differences ob-
`served in activity depending upon the mode of adminis-
`tration.
`Compounds la and Id were tested for the ability t o
`protect mice against a variety of chemically induced sei-
`zures. The findings from these tests (phase V evaluations)
`
`Conley and Kohn
`
`are summarized in Table VI11 along with similar results
`for phenytoin, phenobarbital, ethosuximide, and valproate.
`The mode of action of these convulsants (Metrazol, bicu-
`culline, picrotoxin, and strychnine) is different.15 N-
`Acetyl-DL-alanine N-benzylamide (la) was moderately
`effective in preventing seizures induced in mice by bicu-
`culline and picrotoxin, was partially effective (maximum
`protection, 50% at 800 mg/kg) in the sc Met test, and was
`ineffective against strychnine-induced convulsions. On the
`other hand, N-acetyl-~~-phenylglycine N-benzylamide (Id)
`did not provide mice with any protection against seizures
`induced by the four convulsants. The anticonvulsant ac-
`tivity profiles of both la and Id differ significantly from
`those of phenytoin, phenobarbital, ethosuximide, and
`valproate.
`Conclusions
`The pharmacological profiles exhibited by the func-
`tionalized DL-amino acid derivatives establishes a new class
`of anticonvulsant agents. The specific activities of these
`compounds in the MES, sc Met, and toxicity tests can be
`independently modulated by alteration of the substitution
`pattern at the a-carbon atom, the N-acyl, and the N-amido
`moieties. The observed structure-activity profile suggests
`that stringent steric and electronic requirements exist for
`maximal anticonvu