Synthesis of constrained cycloalkyl analogues of glutamic acid with an ω-phosphonic acid function

Article abstract of DOI:10.1016/S0040-4039(02)01858-0

A general method based on the sequential reactivities of bis-bromocycloalkenes (3-5) is proposed for the preparation of phosphonocycloalkanes (1a/b-3a/b), representing structural constrained analogues of AP4. For the synthesis of an additional congested A

Full text of DOI:10.1016/S0040-4039(02)01858-0

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 43 (2002) 7659–7662
Synthesis of constrained cycloalkyl analogues of glutamic acid
with an v-phosphonic acid function
Bernard Bessie`res,a Ange`le Schoenfelder,a Ce´line Verrat,a Andre´ Mann,a,* Paul Ornsteinb and
Conception Pedregalc
aLaboratoire de Pharmacochimie de la Communication Cellulaire, UMR 7081 Faculte´ de Pharmacie, 74, route du Rhin BP 24,
F-67401 Illkirch, France
bLilly Research Laboratories, Indianapolis, IN 46285, USA
cLilly, SA. Avd. de la industria 30, 28108 Alcobendas, Madrid, Spain
Received 5 August 2002; accepted 2 September 2002
Abstract—A general method based on the sequential reactivities of bis-bromocycloalkenes (3–5) is proposed for the preparation
of phosphonocycloalkanes (1a/b–3a/b), representing structural constrained analogues of AP4. For the synthesis of an additional
congested AP4 analogue (4), an intramolecular cyclopropanation of a ketocarbene towards a vinyphoshonate, assisted by
Rh(OAc)2 was successfully experimented. © 2002 Published by Elsevier Science Ltd.
(S)-Glutamate (Glu) is an important neurotransmitter
in the vertebrates, and because of its intrinsic structural
flexibility, Glu binds productively to several receptor
subtypes, classified as the ionotropic and metabotropic
glutamate receptors (iGlu and mGlu).1 The mGlu
receptors themselves are divided in three groups (I, II
and III) depending on sequence homology and pharma-
cology. Among them Group III mGlu receptors are
auto-receptors and as such, promising targets for mod-
ulating the glutamatergic transmission. AP4, an v
isostere of Glu, is until now the most selective group III
ligand,1,2 but owing its moderate potency better ligands
are highly awaited. Interestingly LY354740,3 a con-
gested Glu analogue, has also some affinity to the
Group III mGlu receptors. These observations sug-
gested that the restriction of rotatable bounds in the
Glu backbone may have some virtue in the design of
new Group III ligands.
Therefore, we embarked in
a chemical program
towards the synthesis of conformational restricted ana-
logues of AP4: compounds (1a/1b, 2a/2b and 3a/b) in
which the AP4 skeleton is incorporated into cycloalkyl
frames, and compound (4), a phosphono isostere close
Scheme 1.
* Corresponding author.
0040-4039/02/$ - see front matter © 2002 Published by Elsevier Science Ltd.
PII: S0040-4039(02)01858-0

7660
B. Bessie`res et al. / Tetrahedron Letters 43 (2002) 7659–7662
1
to the structure of LY354740 (Scheme 1). Indeed con-
formational restriction, a popular strategy in medicinal
chemistry, is based on the optimization of the free
energy gained during ligand/receptor association by
introducing bulk, unsaturation or cyclic motives into
ligand. Similar attempts have been reported in the field
of the Glu ligands.4–8 In this letter, we present our
preliminary results towards the synthesis of compounds
1a/b, 2a/b, 3a/b and 4.
based on H and 13C NMR data, allowed the determi-
nation of the relative stereochemistry of the other cou-
ples of diastereomers. The multiplicity of the methine
proton at C-1 was an ideal probe. In all cases if for the
unlike (ul) diastereomers a pseudo triplet was observed,
for the like (l) diastereomers it was a doublet of dou-
blet. Therefore in comparing the respective NMR spec-
tra, the stereochemistry of the major diastereomers 11a,
12a, 13a (higher Rf) and of the minor diastereomers
11b, 12b,21 13b (lower Rf) was attributed as respectively
ul (R*S*) and l (R*R*). Of course in a backwards
sense the stereochemistry of the corresponding precur-
sors 8a/b–10a/b could also be assigned on the basis of
the Rf and mass balance: 8a–10a are the unlike (ul)
diastereomers and 8b–10b are the like (l) diastereomers.
The final step was realized by an hydrolytic cleavage in
acidic media, and a subsequent treatment with propyl-
ene oxide provided the zwitterions of 1a/b–3a/b in
crystalline form.
For the preparation of the first series of compounds, a
general route was prospected, based on the bis-bromo
intermediates 5–7. Obviously the two bromines will
have different reactivities: the allylic bromide would be
available for a nucleophilic substitution, and the vinylic
bromide possibly for
a phosphorylation reaction
assisted by palladium. These two reactions have some
precedents in acyclic series.9,10 We disclose our results
in Scheme 2. The bis-bromo adducts 5–7 could be
obtained following reported procedures.11–14 Cyclopen-
tenone was the starting material for acceding to 5;
cyclopentene and cyclohexene via their corresponding
dibromocyclopropyl adducts, followed by thermal ring
expansion, afforded the six and seven membered bis-
bromo adducts 6 and 7. For the introduction of the
amino acid appendage at the allyl position, the acetam-
ido route was the most convenient in our hands, even if
there was an extra decarboxylative step. Gratifyingly
the couple of the two diastereomers 8a/8b, 9a/b15,16 and
10a/b could be separated by chromatography on silica
gel, and a slight diastereoselectivity was observed, the
ratio of the two diastereomers in each set was nearly
3/1 (the major diastereomer having the higher Rf). The
relative stereochemistry was secured in the following
adducts (vide infra). The formation of the CꢀP bound
from the vinylbromides was the most troublesome step.
Under the classical conditions recommended for the
palladium assisted vinylic phosphorylation no transfor-
mation was observed.17 After some experimentation we
found that the switch from triethylamine to DABCO as
base,18,19 give a clean reaction, and the expected phos-
phonates 11a/b–13a/b could be obtained in satisfying
yields. The attribution of the relative stereochemistry of
the phosphono-amino acids was facilitated, because a
single crystal X-ray structure was obtained for 12a,20
and assigned as unlike (ul). Thereafter a correlation
The synthesis of the more rigid analogues 4a was
constructed in a different way using the property of
ketocarbenes to produce cyclopropanation (Scheme
3).22,23 Compound 17 was our key intermediate in this
sequence and the cyclopropanation was designed for
compound 16, an electron deficient alkene. Commer-
cially available acid 14 constituted the starting material
for the synthesis of 17, in an uneventful sequence
reported in Scheme 3. The transient aldehyde obtained
after ozonolysis was transformed into the vinylphos-
phonate 16 (E/Z: 95/5) by a Wittig reaction. Finally,
the intramolecular cyclopropanation was best per-
formed by injecting over a 2 h period (syringe pump) a
solution of 16 in acetonitrile to boiling benzene con-
taining Rh(OAc)2 as catalyst. Compound 17 was
obtained after chromatography in good yield, as a
single diastereomer.24 A series of extensive NMR exper-
iments have confirmed that the rigid cyclopentanone 17
has the exo geometry as desired. Its noteworthy that
few examples of a carbene cyclopropanation towards
electron deficient alkenes are reported. The ketone in 18
was then converted into the corresponding hydantoine.
Monn’s condition gave a mixture of the hydantoines
18a/b in a 4/1 ratio, as an inseparable mixture of
diastereomers.3,25 The stereochemistry was assigned on
the mixture by NMR experiments, and it turns out that
Scheme 2. Reagents and conditions: (i) diethyl acetamidomalonate, NaH, DMF, rt, 12 h; (ii) LiBr, H2O, DMF and chromatogra-
phy (for 8a/b: 70%, for 9a/b: 77%; for 10a/10b: 53%, over the two steps, see text); (iii) diethyl phosphite (1.4 equiv.), DABCO (3
equiv.), Pd(PPh3)4 (0.03 equiv.), toluene 110°C, 6 h (from 57 to 84%); (iv) HCl (6N) reflux 24 h, then propylene oxide, rt, 12 h
(from 35 to 58%).

B. Bessie`res et al. / Tetrahedron Letters 43 (2002) 7659–7662
7661
Scheme 3. Reagents and conditions: (i) t-butyl acetate, H2SO4, 24 h; (ii) O3, Me2S; (iii) [(MeO)2 PO]CH2, LDA, THF (37% over
the three steps); (iv) HCO2H, rt, 5 h; (v) (COCl)2, CH2Cl2, then CH2N2 in ether (75%); (vi) 16 added to boiling benzene with 0.05
equiv. of [RhOAc2]2 (65%); (vii) KCN, (NH4)2CO3 in H2O, rt, 3 days (68%); (viii) HCl (6N) sealed tube, 150°C, 24 h (56%)
the adduct 18a (the one with the carboxylate and the
phosphonates function with the exo,exo stereochem-
istry) was the major one, in line with our expectations.
In the final step harsh hydrolytic conditions were used
to obtain fully deprotected amino acid 4 as a single
compound after crystallization, with an HPLC purity
greater than 95%.26
7. Ruiz, M.; Ojea, V.; Fernandez, M. C.; Conde, S.; Diaz,
A.; Quintela, J. M. Synlett 1999, 1903–1906.
8. Ojea, V.; Ruiz, M.; Shapiro, G.; Pombo-Villar, E. J. Org.
Chem. 2000, 65, 1984–1995.
9. Leanna, M. R.; Morton, H. E. Tetrahedron Lett. 1993,
34, 4485–4488.
10. Crisp, G. T.; Glink, P. T. Tetrahedron 1994, 50, 2623–
2640.
In conclusion, we have developed two very simple and
expeditive routes towards the synthesis of rigid ana-
logues of AP4. The first sequence relays on the
chemoselectivity of bis-bromocycloalkenes, the second
on an intramolecular cyclopropanation of a ketocar-
bene towards a vinylphosphonate.20–23
11. Organ, M. G.; Miller, M.; Konstantinou, Z. J. Am.
Chem. Soc. 1998, 120, 9283–9290.
12. Ceylan, M.; Secen, H.; Sutbeyaz, Y. J. Chem. Res. Synop.
1997, 70–71.
13. Grigg, R.; Sridharan, V.; Stevenson, P.; Sukirthalingam,
S.; Worakun, T. Tetrahedron 1990, 46, 4003–4018.
14. Wessjohann, L. A.; Muhlbauer, A.; Sinks, U. Acta Chem.
Scand. 1997, 51, 1112–1115.
Acknowledgements
15. Analytical data for 9a: mp 82°C; Rf 0.43 (hex/ether: 6/4);
1H NMR (300 MHz, CDCl3, 25°C): l 6.25 (m, 1H), 6.08
(bd, J=8.5 Hz, 1H), 4.92 (dd, J=8.5 and 3 Hz, 1H), 4.22
(m, 2H), 3.23 (m, 1H), 2.07 (s, 3H), 2.05–1.91 (m, 3H),
1.69–1.59 (m, 3H), 1.29 (t, J=7.1 Hz, 3H); 13C NMR (75
MHz, CDCl3, 25°C): l 171.1, 169.9, 133.5, 121.8, 61.4,
54.5, 44.6, 28.0, 27.3, 23.1, 19.4, 13.8. Anal. calcd for
C12H18BrNO3: C, 47.38; H, 5.96; N, 4.60. Found: C, 47.3;
H, 5.8; N, 4.5%.
The authors are grateful to Drs. A. Rivera and J. F.
Espinosa (Lilly, SA, Spain) for the highfield NMR
analysis of the mixture of compounds 18a/18b.
References
1
16. Data for 9b: oil, Rf 0.35 (hex/ether: 6/4); H NMR (300
1. For excellent reviews, see: (a) Brau¨ner-Osborne, H.; Ege-
bjerg, J.; Nielsen, E. O.; Madsen, U.; Krogsgaard-Larsen,
P. J. Med. Chem. 2000, 43, 2609–2645; (b) Pin, J. P; de
Colle, C.; Bessis, A. S.; Acher, F. Eur. J. Pharmacol.
1999, 375, 277–294; (c) Knopfel, T.; Kuhn, R.; Allgeier,
H. J. Med. Chem. 1995, 38, 1417–1426.
MHz, CDCl3, 25°C): l 6.22 (m, 1H), 6.16 (bd, J=8.7 Hz,
1H), 5.01 (dd, J=8.7 and 4.5 Hz, 1H), 4.17 (q, J=7.1
Hz, 2H), 2.94 (m, 1H), 2.05–2.00 (m, 2H), 2.01 (s, 3H),
1.72–1.57 (m, 4H), 1.25 (t, J=7.1 Hz, 3H); 13C NMR (75
MHz, CDCl3, 25°C): l 171.1, 170, 133.4, 122.3, 61.3,
54.3, 44.1, 27.1, 25.8, 22.8, 20.0, 13.9.
2. Nakajima, Y.; Iwakabe, H.; Akazawa, C.; Nawa, H.;
Shigemoto, R.; Mizuno, N.; Nakanishi, S. J. Biol. Chem.
1993, 268, 11868–11873.
3. Monn, J. A.; Valli, M. J.; Massey, S. M.; Wright, R. A.;
Salhoff, C. R.; Johnson, B. G.; Howe, T.; Alt, C. A.;
Rhodes, G. A.; Robey, R. L.; Griffey, K. R.; Tizzano, J.
P.; Kallman, M. J.; Helton, D. R.; Schoepp, D. D. J.
Med. Chem. 1997, 40, 528–537.
4. Amori, L.; Costantino, G.; Marinozzi, M.; Pellicciari, R.;
Gasparini, F.; Flor, P. J.; Kuhn, R.; Vranesic, I. Bioorg.
Med. Chem. Lett. 2000, 10, 1447–1450 and references
cited therein.
5. Ma, D.; Zhu, W. J. Org. Chem. 2001, 66, 348–350.
6. Conway, S. J.; Miller, J. C.; Howson, P. A.; Clark, B. P.;
Jane, D. E. Bioorg. Med. Chem. Lett. 2001, 11, 777–780.
17. Hirao, T.; Masunaga, T.; Yamada, N.; Ohshiro, Y.;
Agawa, T. Bull. Chem. Soc. Jpn. 1982, 55, 909–913.
18. Murata, Y.; Woodward, R. M.; Miledi, R.; Overman, L.
E. Bioorg. Med. Chem. Lett. 1996, 6, 2073–2076.
19. Experimental procedure for the preparation of 12a: A
mixture of 9a (800 mg, 2.63 mmol, 1 equiv.), diethylphos-
phite (0.47 ml, 3.68 mmol, 1.4 equiv.) and DABCO (880
mg, 7.89 mmol, 3 equiv.) in dry toluene (27 ml) was
degassed for 5 min with Ar, then palladium tetrakis (100
mg, 0.0087 mmol) was added and the yellow solution was
heated at reflux. After 8 h the mixture became black and
a precipitate appeared. After filtration and evaporation of
the organic layer the residue was purified by silica gel
chromatography eluting with AcOEt/MeOH: 95/5. Com-
pound 12a (680 mg, 72%) was obtained as an oil. Analyt-

7662
B. Bessie`res et al. / Tetrahedron Letters 43 (2002) 7659–7662
ical data for 12a: mp 92°C; Rf 0.35 (AcOEt/MeOH: 95/5);
24. Data for 17: 1H NMR (300 MHz, CDCl3, 25°C): l
3.81–3.75 (m, rotamers of OMe, 6H), 2.52–2.45 (m, 1H),
2.27–2.12 (m, 5H), 1.27 (m, 1H); 31P (81 MHz in CDCl3):
l 27.8; MS (ESI-TOF) C8H13O4P: Mw 204.06; found
(M+H)+: 205.03.
1H NMR (300 MHz, CDCl3, 25°C): l 8.03 (1 s, bd, J=6.5
Hz, 1H), 6.77 (dt, J=3.8 Hz, 1H), 4.47 (dd, J=6.9 and
1.6 Hz, 1H), 4.20 (q, J=6.5 Hz, 2H), 4.08 (m, 4H), 2.92
(m, 1H), 2.29 (m, 2H), 2.21 (s, 3H), 2.41–1.63 (m, 4H),
1.42–1.25 (m, 9H); 13C NMR (75 MHz, CDCl3, 25°C): l
171.5, 170.1, 146.6 (d, J=7.2 Hz), 127.7 (d, J=180 Hz),
62.1 (d, J=22.5 Hz), 61.8 (d, J=22.5 Hz), 60.9, 56.0, 34.8,
25.7, 25.4, 25.3, 22.7, 16.4, 16.1, 14.0. Anal. calcd for
C16H28NO6P: C, 53.17; N, 7.80; H, 3.87. Found: C, 53.3;
N, 7.9; H, 3.8%.
25. The NMR data for the mixture 18a/b were collected at
500 MHz in DMSO at 30°C. The following experiments
were performed 1H,31P-decoupled, 13C, DEPT, COSY,
1D-TOCSY, HSQC, HMBC and 2D-NOESY. Inspection
of the 1H spectrum indicates that the sample is a 4:1
mixture. C-1 was easily assigned in the 13C NMR spec-
trum (large 13C–31P coupling constant). H-1 was identified
at 1.12 ppm through its coupling to C-1 as observed by
HSQC. In the 31P-decoupled proton spectrum, H-1
appears as a triplet with (J 4.0 Hz), indicating that H-2
and H-3 are trans to H-1 (the cis value would be much
larger). In the minor compound H-1 also appears as
triplet with a similar coupling constant, therefore the
stereochemistry at the cyclopropane ring is the same for
the two isomeric compounds 18a/b. The configuration at
C-6 has been elucidated through 1D-NOESY. A NOE
effect is detected between H-1 and N-H(7) for the major
isomer 18a but not for the minor, indicating that H(1) and
H(7) are on the same face in the major isomer (18a), and
opposite in the corresponding minor one (18b).
20. The X-ray structure and coordinates for compound 12a
have been deposited at the Cambridge Crystallographic
Database under the deposition number CCDC 185495.
Crystal data for 12a: C16H28NO6P, M=361.38, mono-
clinic, space group: P121/C1, a=12.7248 (5), b=9.3206
3
,
,
(6), c=16.8025 (7) A, U=1977.1 A , T=294 K, Z=4,
v(Mo K)=0.167 mm−1, 7698 measured reflections, 1962
unique (Rint=0.04). The final wR(F2) was 0.062 (all data).
21. Physical data 12b: mp 87°C; Rf 0.30 (AcOEt/MeOH:
95/5); 1H NMR (300 MHz, CDCl3, 25°C): l 6.95 (dt,
J=22.5 and 3.4 Hz, 1H), 6.83 (bd, J=7.8 Hz, 1H), 4.83
(dd, J=7.8 and 3.1 Hz, 1H), 4.21–4.02 (m, 6H), 2.92 (m,
1H), 2.20 (m, 2H), 2.01 (s, 3H), 1.76–1.73 (m, 4H),
1.38–1.15 (m, 9H); 13C NMR (75 MHz, CDCl3, 25°C): l
170.8, 169.9, 147.7 (d, J=7.5 Hz), 127.3 (d, J=177 Hz),
61.9 (d, J=5.5 Hz), 61.0, 55.2, 37.0, 36.9, 25.7, 25.6, 25.5,
22.9, 19.4, 16.1, 13.9. Anal. calcd for C16H28NO6P: C,
53.17; N, 7.80; H, 3.87. Found: C, 52.9; N, 7.9; H, 3.7%.
22. Padwa, A.; Hornbuckle, S. F. Chem. Rev. 1991, 91,
263–309.
26. Data for 4: 1H NMR (300 MHz, MeOH-d4, 25°C): l
2.43–2.25 (m, 1H), 2.25–2.00 (m, 4H), 1.72–1.58 (m, 1H),
1.38 (bs, 1H); 13C NMR (75 MHz, MeOH-d4, 25°C): l
171.5, 65.4, 29.60, 28.5 (d, J=3.5 Hz), 26.1 (d, J=4.5
Hz), 24.2 (d, J=3.5 Hz), 13.9 (d, J=187.2); MS (electro-
spray +) C7H12NO5P: Mw 221; found (M+H)+: 222; MS
(electrospray −) (M−H)+: 220.
23. Davies, H. M. L.; Antoulikanis, E. G. Organic Reactions;
Wiley: New York, 2001; Vol. 57, pp. 1–327.