Synthesis of phenylglycine derivatives as potent and selective antagonists of group III metabotropic glutamate receptors

Article abstract of DOI:10.1016/S0960-894X(01)00052-X

The syntheses of a range of ring and α-substituted 4-phosphonophenylglycines are described. A brief discussion of the antagonist activities of compounds 4-10 on group I, II and III metabotropic glutamate (mGlu) receptors expressed in the neonatal rat spin

Full text of DOI:10.1016/S0960-894X(01)00052-X

Bioorganic & Medicinal Chemistry Letters 11 (2001) 777±780
Synthesis of Phenylglycine Derivatives as Potent and Selective
Antagonists of Group III Metabotropic Glutamate Receptors
Stuart J. Conway,a Jacqueline C. Miller,a Patrick A. Howson,a Barry P. Clarkb
and David E. Janea,*
aDepartment of Pharmacology, School of Medical Sciences, Bristol BS8 1TD, UK
bEli Lilly and Co., Erl Wood Manor, Windlesham, Surrey, GU20 6PH, UK
Received 27 November 2000; revised 8 January 2001; accepted 19 January 2001
AbstractÐThe syntheses of a range of ring and a-substituted 4-phosphonophenylglycines are described. A brief discussion of the
antagonist activities of compounds 4±10 on group I, II and III metabotropic glutamate (mGlu) receptors expressed in the neonatal
rat spinal cord is included. # 2001 Elsevier Science Ltd. All rights reserved.
It is now accepted that (S)-glutamate ((S)-Glu) is the
principle excitatory neurotransmitter in the vertebrate
central nervous system (CNS).1 Due to its inherent
structural ¯exibility, (S)-Glu acts at a wide variety of
glutamate receptor subtypes.1À3 These glutamate or
excitatory amino acid receptors can be divided into two
general classes; ionotropic glutamate (iGlu) receptors
and metabotropic glutamate (mGlu) receptors, on the
basis of molecular cloning, pharmacology and electro-
physiology.1,3 The iGlu receptors include the N-methyl-
d-aspartate (NMDA), (S)-a-amino-3-hydroxy-5-methyl-
4-isoxazole-propionic acid (AMPA) and kainic acid
receptor subtypes. These mediate fast synaptic respon-
ses via ligand gated ion channels.3 The mGlu receptors
are coupled to intracellular second messenger systems
via G-proteins and play a modulatory role in synaptic
transmission.1,2 To date, eight mGlu receptor subtypes
have been cloned and these have been divided into three
groups on the basis of amino acid sequence homology,
pharmacology and coupling to second messenger sys-
tems. Group I (mGlu1 and mGlu5) mGlu receptors are
selectively activated by (S)-3,5-dihydroxyphenylglycine
(DHPG). Group II mGlu receptors (mGlu2 and mGlu3)
are selectively activated by (2R,4R)-4-aminopyrrolidine-
2,4-dicarboxylate (APDC) whilst those in group III
(mGlu4,6,7,8) are selectively activated by (S)-2-amino-4-
phosphonobutanoic acid ((S)-AP4).2
Whilst in-roads have been made into the development
of group selective antagonists very few show useful
selectivity for group III mGlu receptors.2
One of the ®rst antagonists to show selectivity for mGlu
over iGlu receptors, (S)-a-methyl-4-carboxyphenyl-
glycine (MCPG 1, Fig. 1), displayed activity at all three
groups of mGlu receptors.1,2,4,5 Bioisosteric replacement
of the terminal carboxyl group of MCPG with a phos-
phono group6,7 to give a-methyl-4-phosphonophe-
nylglycine (MPPG, 2) led to improved selectivity for
group III over group II mGlu receptors, and abolished
activity on group I mGlu receptors (see Table 1). A further
improvement in potency and selectivity for group III
over group II mGlu receptors was obtained by sub-
stitution of the a-methyl group for an a-cyclopropyl
group to give (RS)-a-cyclopropyl-4-phosphonophenyl-
glycine (CPPG, 3) (see Table 1).8,9
In recent years there has been much interest in the
possible therapeutic applications of mGlu receptor
*Corresponding author. Tel.:+44-117-9287639; fax; +44-117-
9279839; e-mail: david.jane@bris.ac.uk
Figure 1. Phenylglycine based group III mGlu receptor antagonists.
0960-894X/01/$ - see front matter # 2001 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(01)00052-X

778
S.J.Conway et al./ Bioorg.Med.Chem.Lett.11 (2001) 777±780
ligands.1,2,10 Before the therapeutic potential and physi-
ological signi®cance of the mGlu receptor subtypes can
be fully explored, it is necessary to develop potent and
selective ligands for these receptors. Here, we report the
synthesis and initial pharmacological evaluation of
seven new phenylglycine derivatives and discuss struc-
tural features required for selective antagonist activity
at group III mGlu receptors.
obtained directly by hydrolysis in 6 N HCl (5±50%
yield). Steps d, e, f and g in Scheme 1 were carried out in
an acid digestion bomb (Parr Instruments, Illinois), as
optimal yields were obtained under these conditions.
Compound 10 was synthesized from 2-methoxyphenol
(guaiacol, 21, Scheme 2). Treatment of 21 with alumi-
nium chloride and 4-bromobutyryl chloride in nitro-
benzene initially esteri®ed the phenolic alcohol which
subsequently underwent an in situ Fries rearrangement
to yield the corresponding butyrophenone. This was
extracted into two equivalents of 2 M NaOH to simul-
taneously isolate it from the nitrobenzene and e€ect
cyclisation, yielding the cyclopropyl phenyl ketone (22)
in 32% overall yield. Tri¯ation of the phenol13 and
subsequent treatment with diethyl phosphite and a pal-
ladium catalyst14 yielded the 4-phosphono substituted
analogue (23). The ketone (23) was converted to a 50-
cyclopropyl-hydantoin (24) via the Bucherer±Bergs
reaction.15 Base hydrolysis of 24 gave the monoethyl
phosphonate ester of amino acid 25. This prior base
hydrolysis allowed a shorter treatment with acid for the
subsequent hydrolysis than would otherwise have been
required thus ensuring that opening of the cyclopropyl
ring did not occur. The diacid (10) was obtained by
hydrolysis of the monoester (25) in 6 N HCl for 3 h at
re¯ux (35% yield).
Chemistry
A convenient synthesis of the phenylglycine deriva-
tives 5±9 was devised starting from the appropriately
substituted acetophenones (13±16).
As one such acetophenone (13) was not commercially
available it was synthesized from 2-chlorophenol
(Scheme 1). The phenolic alcohol (11) was acylated
under Schotten±Baumen conditions11 to give 12 in 79%
yield, which was subsequently subjected to a Fries
rearrangement12 to give 13 in 59% yield.
Tri¯ation of the phenols (13±16)13 and subsequent
treatment with diethyl phosphite and a palladium cata-
lyst14 yielded the corresponding 4-phosphono-sub-
stituted acetophenones. These were converted to the 50-
methylhydantoins (17±20) via the Bucherer±Bergs reac-
tion.15 In the case of hydantoin (18), base hydrolysis
gave the monoethyl ester of the phosphonate (7) in 34%
yield. The monoethyl ester is exclusively formed due to
the formation of the monosodium salt of the mono-
ethoxyphosphono group, which prevented further
hydroxyl attack. For other hydantoins the monoesters
were not isolated as the diacids (5, 6, 8 and 9) were
Compound 4 was synthesized from 3-methyl-aceto-
phenone (26) (Scheme 3). Bromination of 26 and a sub-
sequent Arbuzov reaction16 furnished 27 in 73% overall
yield. The ketone (27) was converted to the hydantoin15
in excellent yield (97%) and acid hydrolysis gave the
amino acid 4 in 21% yield.17
Scheme 1. (a) NaOH soln. (10% w/v), crushed ice, acetic anhydride
79%; (b) Fries rearrangement, AlCl3, 120 ꢀC, 59%; (c) 4-DMAP, 2,6-
lutidine, tri¯uoromethanesulfonic anhydride, CH2Cl2, N2 atmosphere,
À78 ꢀC, 62%±93%; (d) diethyl phosphite, tetrakistriphenylphosphine
palladium(0), anhydrous Et3N, bomb, 80 ꢀC, 70±83%; (e) Bucherer±
Bergs reaction: (NH4)2CO3, KCN, 50% EtOH, bomb, 80 ꢀC, 37±82%;
(f) 4 N NaOH, bomb, 120 ꢀC, 34%; (g) 6 N HCl, bomb, 120 ꢀC, 5±50%.
Scheme 2. (a) (i) AlCl3, 4-bromobutyrylchloride, nitrobenzene; (ii) 2 N
NaOH, 32% overall yield; (b) 4-DMAP, 2,6-lutidine, tri-
¯uoromethanesulfonic anhydride, CH2Cl2, N2 atmosphere, À78 ꢀC,
51%; (c) diethyl phosphite, tetrakistriphenylphosphine palladium(0),
anhydrous Et3N, bomb, 80 ꢀC, 27%; (d) (NH4)2CO3, KCN, 50%
EtOH, bomb, 80 ꢀC, 34%; (e) 2 N NaOH, bomb, 120 ꢀC, 34%; (f) 6 N
HCl, re¯ux, 3 h, 35%.

S.J.Conway et al./ Bioorg.Med.Chem.Lett.11 (2001) 777±780
779
To examine the e€ect of 4 and 5 (1 mM) on direct
depolarisation of spinal motoneurones induced by the
group I mGlu receptor ligand (S)-3,5-DHPG experi-
ments were conducted in standard medium (excluding
Mg2+/AP5) that contained tetrodotoxin (TTX; 10À5
M
for 2 min, then 10À7 M continuously). The degree of
depolarisation of the motoneuron was measured by
peak amplitude. The agonist was applied in the absence
and presence of the antagonist (1 mM) and the ability of
the antagonist to reduce the peak amplitude of the
response was measured.
Scheme 3. (a) NBS, CCl4, hn; (b) P(OEt)3; (c) (NH4)2CO3, KCN, 50%
EtOH, 80 ꢀC, 24 h, 96.5% yield; (d) hydrolysis, 6 N HCl, bomb,
120 ꢀC, 24 h, 21% yield. Stages a and b overall yield 73%.
Pharmacology
All experiments were performed on isolated hemisected
spinal cords of 1±5 day Wistar rats of either sex.18
Recordings were taken from the ventral root following
stimulation of the corresponding dorsal root (30 V, 2
pulses minÀ1). To isolate the fast non-NMDA compo-
nent of the response (i.e., the fDR-VRP), DR-VRPs
were recorded in the presence of 2 mM MgSO4 and 50
mM (R)-2-amino-5-phophono-pentanoate (AP5).19 The
novel phenylglycines were tested for their ability to
antagonize either group II ((1S,3S)-ACPD) or ((2R,4R)-
APDC) used as the agonist challenge) or group III ((S)-
AP4 used as the agonist challenge) mGlu receptor ago-
nist-induced depressions of the fDR-VRP (see Table 1).20
Structure±activity conclusions
From the pharmacological data presented in Table 1 we
propose a number of structural features required for
optimal antagonist activity at group III mGlu receptors
present on primary a€erent terminals.
A terminal phosphono group confers selective activity
for group III versus group I mGlu receptors as com-
pounds 2±5 showed no appreciable antagonist activity
at group I mGlu receptors.
It would appear that a co-linear arrangement of the
bonds attaching the phosphonate and glycine unit to the
phenyl ring is necessary for optimal activity at group III
mGlu receptors, as the 3-phosphonomethyl analogue (4)
which has the same inter-acidic group chain length as
MPPG (2) was almost completely inactive.
Table 1. Pharmacological data given as apparent KDs (mM)ÆSEM
(values are as a result of three separate determinations) unless other-
wise stated
Compound Antagonist activity at mGlu receptors in rat spinal cord:
Substitution on the phenyl ring of MPPG at the 3 posi-
tion leads to similar antagonist activity at group III
mGlu receptors and this activity is independent of the
electronic e€ects of the substituent i.e., no real di€er-
ences between KD values for the 3-chloro (5) (electron
withdrawing) and the 3-methyl (8) (electron donating)
substituted phenylglycines. Thus, the group III mGlu
receptor binding site can accommodate substituents at
least as large as methoxy at the 3-position of the phenyl
ring of MPPG.
Group I
Group II
Group III
1 (MCPG)
2 (MPPG)
144Æ1221
479Æ375
113Æ136
538
227Æ125
9.2Æ0.36
1.78
n.e. at 1 mM6
3 (CPPG) 70Æ12% at 1 mMa, 8
4
5
6
7
8
9
10
62Æ1% antag.a
20Æ6 % pot.b n.e. at 200 mM
ne at 1 mM
726Æ125
6.8Æ1.5e
4.8Æ0.8
Ð
Ð
Ð
Ð
Ð
51Æ1%c
Ð
Ð
Ð
Ð
13Æ2%d
5.5Æ0.6
180Æ49
n.e. at 50 mM
However, the most signi®cant e€ect of 3-substitution, is
that of the reduction of activity at group II mGlu
receptors. Addition of a chloro group at the 3-position
of the phenyl ring of MPPG conferred ꢁ100-fold selec-
tivity for group III over group II mGlu receptors. A 3-
methoxy substituent conferred selectivity in excess of
200-fold for group III over group II mGlu receptors.
This is probably due to the steric bulk of the substituent
at the 3-position of MPPG being well tolerated by
group III but not group II mGlu receptors.
a% Antagonism of (1S,3R)-ACPD response at 200 mM, n=3.
b% Potentiation of (1S,3S)-ACPD response at 200 mM, n=4.
c% Antagonism of (2R,4R)-APDC response at 1 mM.
d% Antagonism of (S)-AP4 at 100 mM.
For potent group III mGlu receptor antagonists, con-
centration±response curves to (S)-AP4 and either
(1S,3S)-ACPD or (2R,4R)-APDC (to assess selectivity)
were constructed in the absence and presence of the
novel antagonist and an apparent KD value obtained. In
cases where antagonist potency was low, a one point
assay (performed in triplicate) was carried out in which
a single concentration of antagonist was applied and
observed for its ability to antagonize a single agonist
induced response (results being expressed as a percen-
tage antagonism).
Substitution at the 2-position of the phenyl ring as
exempli®ed by 9 reduces antagonist potency at group III
mGlu receptors. This may be due to the steric bulk at
the 2-position being unacceptable to group III mGlu
receptors, or more likely due to the 2-methyl group alter-
ing the preferred orientation of the glycine unit relative to
the phenyl ring forcing the 2-methyl group into excluded
space in the receptor. This is supported by the overlaid
minimum energy conformations of 6 and 9 showing the
two phenyl rings perpendicular to each other (Fig. 2).
Antagonist action on group II mGlu receptors was not
determined when little or no antagonist action of a
compound was observed when tested at 100 mM against
group III mGlu receptor agonist-induced responses.

780
S.J.Conway et al./ Bioorg.Med.Chem.Lett.11 (2001) 777±780
Acknowledgements
We would like to thank Eli Lilly and Co., the BBSRC
and the MRC for funding this work. We acknowledge
Miss Marie Woolley for the pharmacological
characterization of compound 4.
References and Notes
1. Conn, P. J.; Pin, J.-P. Annu.Rev.Parmacol.Toxicol. 1997,
37, 205.
2. Schoepp, D. D.; Jane, D. E.; Monn, J. A. Neuropharma-
cology 1999, 38, 1431.
3. Watkins, J. C.; Krogsgaard-Larsen, P.; Honore, T. Trends
Pharmacol.Sci. 1990, 11, 25.
Figure 2. Left: Overlay of 6 (yellow) and 10 (blue). Right: Overlay of 6
(yellow) and 9 (red). Optimised in Chem-X and displayed in Chem3D.
4. Watkins, J. C.; Collingridge, G. L. Trends Pharmacol.Sci.
1994, 15, 333.
5. Kemp, M. C.; Roberts, P. J.; Pook, P. C.-K.; Jane, D. E.;
Jones, A. W.; Jones, P. L. St. J.; Sunter, D. C.; Udvarhelyi,
P. M.; Watkins, J. C. Eur.H.Pharmacol±.Mol.Pharmacol.
1994, 226, 187.
6. Jane, D. E.; Pittaway, K.; Sunter, D. C.; Thomas, N. K.;
Watkins, J. C. Neuropharmacology 1995, 34, 851.
7. Jane, D. E.; Pittaway, K.; Sunter, D. C.; Thomas, N. K.;
Tse, H.-W. Phosphorous, Sulfur and Silicon. 1996, 109±110,
313.
8. Jane, D. E.; Thomas, N. K.; Tse, H.-W.; Watkins, J. C.
Neuropharmacology 1996, 35, 1029.
9. Toms, N. J.; Jane, D. E.; Kemp, M. C.; Beding®eld, J. S.;
Roberts, P. J. Br.J.Pharm. 1996, 119, 851.
10. Pellicciari, R.; Costantino, G. Curr.Opin Chem.Biol.
1999, 3, 433.
Both OH groups on the phosphonate moiety may be
needed for optimal binding to the group III mGlu
receptor, as the mono ethyl ester (7) was much less
active than the corresponding diacid (6). Alternatively,
steric hindrance due to the ethyl group may be responsible
for the reduction in activity.
Compound 10, an a-cyclopropyl-substituted analogue
of 6, was surprisingly inactive at concentrations up to
50 mM at group III mGlu receptors. One explanation for
this could be that CPPG and MPPG have di€erent
preferred orientations of the glycine unit relative to the
plane of the phenyl ring and therefore the 3-methoxy
substituent on the phenyl ring of 10 may probe excluded
space in the receptor binding site. When energy mini-
mised conformations of compounds 6 and 10 are over-
laid it can be seen that the two phenyl rings are
perpendicular to each other supporting this theory (Fig. 2).
11. (a) Schotten, A. Chem.Ber. 1884, 17, 2544. (b) Baumann,
E. Chem.Ber 1886, 3218.
12. Donnelly, J. A.; Murphy, J. J. J.Chem.Soc.(C)
2596.
1970,
13. Eli Lilly and Co., European Patent EP 621 A10807, 1995.
14. Lu, X.; Zhu, J. Synthesis 1987, 726.
15. Bucherer, H. T.; Lieb, B. A. J.Prakt.Chem. 1934, 141, 5.
16. (a) Michaelis, A.; Kaehne, R. Chemistry.Ber. 1898, 31,
Concluding Remarks
1048. (b) Arbuzov, A. E. J.Russ.Phys.Chem.Soc.
687.
1906, 38,
As the 3-chloro- and 3-methoxy-MPPG analogues (5
and 6, respectively) are at least 100-fold selective for
group III over group II mGlu receptors they are the
most potent selective group III mGlu receptor antago-
nists yet reported. As such they are likely to be useful
tools to elucidate the physiological roles of group III
mGlu receptors in the CNS. Recent work has suggested
that the depressant action of (S)-AP4 on synaptic
transmission in the spinal cord is mediated by mGlu8
receptors2,22 and therefore the compounds described in
this report are likely to be mGlu8 receptor antagonists.
We are currently characterising 5, 6 and 8 on all eight
mGlu receptor subtypes in order to determine their
subtype selectivity.
17. All compounds had correct 1H and 13C NMRs and ele-
mental analysis accurate to within 0.4%.
18. Evans, R. H.; Francis, A. A.; Jones, A. W.; Smith, D. A. S.;
Watkins, J. C. Br.J.Pharmacol. 1982, 75, 65.
19. Long, S. K.; Smith, D. A. S.; Siarey, R. J.; Evans, R. H.
Br.J.Pharmacol. 1990, 100, 850.
20. Jane, D. E.; Jones, P. L. St. J.; Pook, P. C.-K.; Tse, H.-W.;
Watkins, J. C. Br.J.Pharmacol. 1994, 112, 809.
21. Jane, D. E.; Jones, P. L. St. J.; Pook, P. C.-K.; Salt, T. E.;
Sunter, D. C.; Watkins, J. C. Neuropharmacology 1993, 32,
725.
22. Thomas, N. K.; Wright, R. A.; Howson, P. A.; Kingston,
A. E.; Schoepp, D. D.; Jane, D. E. Neuropharmacology 2000,
32, 725.