Studies on pyrrolidinones. On the decarboxylation of pyroglutamic acids and N-acyl prolines in acidic media

Article abstract of DOI:10.1016/S0040-4020(02)01183-3

During attempted Friedel-Crafts cyclization of some arylmethyl pyroglutamic acids or of N-phenacyl prolines, decomposition of the activated form of the acid have been observed, giving new heterocyclic systems. This general decarboxylation occurred when there are difficulties to realize a Friedel-Crafts cyclization and is explained by geometrical or electronic considerations.

Full text of DOI:10.1016/S0040-4020(02)01183-3

Tetrahedron 58 (2002) 9239–9247
Studies on pyrrolidinones. On the decarboxylation of
pyroglutamic acids and N-acyl prolines in acidic media
†
a
,
‡
a
Rufine Akue-Gedu, Sahar Al Akoum Ebrik, Anne Witczak-Legrand,a Dominique Fasseur,a §
,
,
´
´
Samira El Ghammarti,a { Daniel Couturier,b Bernard Decroix,c Mohamed Othman,c
,
d
Marc Debacker and Benoıt Rigo
a
,
*
ˆ
a
´
Groupe de Recherche sur l’Inhibition de la Proliferation Cellulaire, EA 2692, Ecole des Hautes Etudes Industrielles,
13, rue de Toul, 59046 Lille, France
b
´
´
´
Laboratoire d’Ingenierie Moleculaire, Universite des Sciences et Technologies de Lille, 59655 Villeneuve D’Ascq, France
´
c
´
Laboratoire de Chimie, Faculte des Sciences et Techniques de L’Universite du Havre, 76058 Le Havre, France
dLASIR-HEI, UMR 8516, 13 rue de Toul, 59046 Lille, France
Received 16 April 2002; revised 3 July 2002; accepted 16 September 2002
Abstract—During attempted Friedel–Crafts cyclization of some arylmethyl pyroglutamic acids or of N-phenacyl prolines, decomposition of
the activated form of the acid have been observed, giving new heterocyclic systems. This general decarboxylation occurred when there are
difficulties to realize a Friedel–Crafts cyclization and is explained by geometrical or electronic considerations. q 2002 Elsevier Science Ltd.
All rights reserved.
1. Introduction
respectively; surprisingly, a low amount of decarboxylated
products 5 (7%) and 6 (17%) were also isolated.
Pyroglutamic acid A is an interesting starting block for
heterocyclic ketones1 and we2 and other authors3 have
already described the cyclization of many N-arylmethylpyro-
glutamic acids B (Scheme 1).
In an attempt to provide supporting evidence for a
decarboxylated structure of these products, compound
6 was prepared by an electrochemical synthesis4
(Scheme 2). In order to suppress the decarbonylation
process, some reactions were also tested without success
by using aluminum bromide instead of aluminum
chloride.5a
Cyclization of acid 1 has been described in 24% yield.3b
While attempting to repeat this reaction or to cyclize acid 2,
ketones 3 and 4 were obtained in 60 and 33% yields,
Scheme 1.
Keywords: decarbonylation; decarboxylation; pyroglutamic acid; N-acylproline; lactams; Friedel–Crafts reaction; N-acyliminium salts.
*
Corresponding author. Tel.: þ33-3-28384858; fax: þ33-3-28384804; e-mail: rigo@hei.fr
†
Present address: Hautes Etudes Industrielles, Lille, France.
´
‡
Present address: Center de Phytopharmacie, Universite de Perpignan, France.
´
´ ´
Present address: France Faculte de Medecine, Rabat, Maroc.
§
{ Present address: LSEO, Universite de Bourgogne, Dijon, France.
0040–4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(02)01183-3

´ ´
R. Akue-Gedu et al. / Tetrahedron 58 (2002) 9239–9247
9240
Scheme 2.
ketone.2,3 When the aromatic is deactivated (2) (Scheme 2)
or must be cyclized in the meta position of an ortho/para
directing group (1) (Scheme 2) the cyclization rate is lower,
and loss of carbon monoxide can occur. In other compounds
(7, 14) (Schemes 3 and 4), the aromatic nucleus is farther
away from the acid carbonyl group, thus decreasing the rate
of ketone formation below those of the decarboxylation
step. These observations were confirmed by literature
reports3c,8 on the cyclization of acids 16 and 18 (Scheme
5): compound 16, with electron rich aromatic nucleus gave a
seven ring ketone in good yield, while non-electron rich acid
18 gives only a 8% yield of ketone 19. We postulate that the
formation of an iminium salt (20) (Scheme 6) can explain
this low yield.
Scheme 3.
In another set of reactions, the Friedel–Crafts cyclization of
acid 7 could not be achieved: no cyclized product was
isolated, and only the unsaturated lactam 8 was obtained in
56% yield (Scheme 3).
In another situation, to obtain ketone 12, the aromatic ring
needs to rotate in order to place a reacting center (a for acid
11) (Scheme 7) next to the anhydride carbonyl group. By
contrast, in the initial conformation (see Fig. 1, the X-ray
structure of the methyl ester of acid 116), another reacting
center (b for acid 11) (Scheme 7) is well situated, below the
carbon of the potential iminium salt, thus providing
assistance for carbon monoxide evolution (Scheme 7).
2. Results and discussion
Decarboxylations leading to non-cyclic products have to be
compared with other reactions that we have previously
observed5b,6,7 or that were reported in literature3c,8
(Schemes 4 and 5).
Interestingly, the same decarboxylation reactions can occur
in the N-acylproline series. Heated in polyphosphoric acid,
the acids 21–24 yield the seven-membered ring ketones
25–2810 (Scheme 8). Under the same conditions,
decarboxylation of acid 29 lead also to the iminium salt
30 which cyclized, giving the lactam 3111 (Scheme 9).
For all of these decarboxylations, the initial mechanism is
not a tandem radical-oxidation of the compound9a but is
identical to another decarbonylation described for amino
acids:5,9b the activated form of the acid decompose by loss
of carbon monoxide (Scheme 6). The acyl iminium salt (20)
then either converts to unsaturated lactam (8), or reacts with
an aromatic, if present, to give an opened (10) or cyclized
(13, 15) product. This iminium salt can also be quenched at
the end of the reaction, giving products 5, 6 (Scheme 6); no
attempts were made to quench the salt 20 by another
nucleophile that water.
In order to understand these results, geometries of acids 7,
14, 16, 18, 32 and 33 were optimized by using ab initio DFT
calculations (6.31 Gp p ) using JAGUAR 3.512a and
SPARTAN 512b to graphically display the results. Among
these molecules, 16, 32 and 33 cyclize easily and with good
yields. On the other hand, 7 and 14 do not give a ketone
while 18 does it with only a very low yield (Scheme 10).
Thus, when an activated form of a pyroglutamic acid is
formed (and if necessary subjected to the action of a Lewis
acid), there is a competition between two reactions with
different kinetics; generally speaking, the fastest process is
the reaction of the activated form, to give an aromatic
Geometric considerations help to explain why 7 and 14 do
not cyclize, but fail to explain the difference in behavior
between 16 and 18.

´ ´
R. Akue-Gedu et al. / Tetrahedron 58 (2002) 9239–9247
9241
Scheme 4.
Scheme 5.

´ ´
R. Akue-Gedu et al. / Tetrahedron 58 (2002) 9239–9247
9242
Scheme 6.
There is no overlap of the LUMO for 16 or 18, but this is
observed for the LUMO(þ1) of 16. In spite of the fact that
the energies of LUMO and LUMO(þ1) are almost the same
in the case of 18, this overlap is not observed. It would thus
seem required that LUMO overlap be required to get
cyclization.
3. Synthesis of acids 7 and 2
Acid 7 used in the reaction of Scheme 3 was obtained by a
modification of a method from Roth:13 condensation of
pyroglutamic acid A with the hemiacetal 34 of methyl
glyoxylate14 yields lactam 35 as a 50/50 mixture of
diastereoisomers. In acidic medium, iminium salt 36 was
formed; steric interactions between the lactam carbonyl
group and the ester function of compound 36 explain the
formation of a single isomer of the iminium salt. Approach
of the benzodioxole is favored on the side opposed to the
Scheme 7. Position a and b of the mixte anhydride of acid 11.
Energies of the HOMO, LUMO and LUMO(þ1) orbitals
are reported in Table 1, and it can be seen that these values
do not show any significant trend that would allow us to
account for the difference in reactivity of these acids. One
must note however that the energies of the LUMO and
LUMO(þ1) are fairly close and in case of 18, they can be
considered as degenerate.
Graphical representation of the first two LUMOs are shown
in Fig. 2 for 32, 33, 16 and 18. It is clear that the shape of
the LUMO of 32 and 33 give a good indication that a ring
closure is likely to take place.
Figure 1. ORTEP diagram of acid 11 methyl ester.
Scheme 8.

´ ´
R. Akue-Gedu et al. / Tetrahedron 58 (2002) 9239–9247
9243
Scheme 9.
Scheme 10.
acid group, giving then acid 7, also as a single isomer13
(Scheme 11).
Table 1. Energies in kJ of HOMO and LUMOs orbitals obtained by ab
initio DFT calculations
Synthesis of acid 1 used in the reaction of Scheme 2 was
already described15 by saponification of the methyl ester 37
formed by reaction of the sodium salt of methyl pyro-
glutamate16 with the corresponding benzyl chloride.17
Acid 2 was obtained by using a similar reaction pathway
(Scheme 12).
Acid
HOMO
LUMO
LUMO(þ1)
14
16
18
2803
2781
2855
354
360
371
403
408
380

´ ´
R. Akue-Gedu et al. / Tetrahedron 58 (2002) 9239–9247
9244
4. Conclusion
It is thus important to know that the decarboxylation of
activated forms of pyroglutamic acids or of N-acylprolines
is not an ‘exceptional’ reaction caused solely by reagents
such as polyphosphoric acid5b or by the ring size of the
ketone to be formed,7 but that a carbon monoxide loss,
leading to an acyl iminium salt, can occur in numerous and
diverse situations.
5. Experimental
5.1. Materials
Melting points were determined with an Electrothermalw
apparatus and are uncorrected. Thin-layer chromatographies
were carried out on Merck F-254 silica glass plates. The IR
spectra were recorded on a ‘Perkin–Elmer’ 700 spectro-
meter and the NMR spectra on a Varian ‘Gemini 2000’ at
1
200 MHz for H and 50 MHz for 13C, using tetramethyl-
silane as an internal reference. Elemental analyses were
performed by the ‘Service Central de Microanalyses’
(CNRS, Vernaison, France). Pyroglutamic acids or esters
used were racemic.
5.1.1. N-(2,4-Dichlorobenzyl)pyroglutamic acid (2). A
stirred mixture of methyl N-(2,4-dichlorobenzyl)pyrogluta-
mate (60.4 g, 0.20 mol) and sodium hydroxide (12 g,
0.30 mol) in water (400 ml) was refluxed for 2 h. The
cooled solution was washed with dichloromethane, solvents
were evaporated and the solution was acidified with
concentrated hydrochloric acid, giving 91% of acid 2, mp
1618C (methanol); IR (KBr): n cm21 1730, 1640 (CvO),
1
1590, 1565, 1480 1460 (CvC); H NMR (CDCl3): d ppm
2.1–2.3 (m, 1H), 2.3–2.48 (m, 1H), 2.48–2.75 (m, 2H),
4.17 (dd, J¼8.9, 2.9 Hz), 4.28 (d, J¼15.3 Hz, 1H), 5.04 (d,
J¼15.3 Hz, 1H), 7.23 (dd, J¼8.3, 1.8 Hz, 1H), 7.30 (d,
J¼8.3 Hz, 1H), 7.40 (d, J¼1.8 Hz, 1H); 13C NMR (CDCl3):
d ppm 23.0, 29.1, 42.9, 59.0, 127.5, 129.4, 131.6, 134.5,
134.6, 174.7, 176.3. Anal. calcd for C12H11NO3Cl2: C,
Figure 2. Representation of LUMO orbitals of 16, 18, 32 and 33.
Scheme 11.

´ ´
R. Akue-Gedu et al. / Tetrahedron 58 (2002) 9239–9247
9245
Scheme 12.
50.02; H, 3.85; N, 4.86; O, 16.66. Found: C, 50.40, H, 3.92;
N, 4.77; O, 16.38.
was stirred at room temperature for 2 days. The mixture was
poured in a water/ice mixture and methylene dichloride was
added. The organic phase was washed with water until
neutralization, then with a potassium carbonate solution.
After drying (sodium sulfate) solvents were evaporated.
Crystallization of the residue in ethyl acetate gives ketone 4
(33%), mp 137–1408C (ethyl acetate); IR (KBr): n cm21
5.1.2. 7-Methoxy-1,2,3,5,10,10a-hexahydrobenz[f]indoli-
zine-3,10-dione (3) and 5-hydroxy-1-(4-methoxybenzyl)-
pyrrolidin-2-one (5). A stirred mixture of acid 1 (10 g,
0.04 mol) and thionyl chloride (7.5 ml, 0.1 mol) in dichloro-
ethane (200 ml) was refluxed for 30 min. Solvents were
evaporated, dichloroethane (100 ml) was added and the
solution was cooled with a water bath at 88C. Aluminum
chloride (21.4 g, 0.16 mol) was added (30 min) and the
mixture was stirred at 88C for 4 h. The mixture was poured
in a water/ice mixture and methylene dichloride was added.
The organic phase was washed with water until neutrali-
zation, then with a potassium carbonate solution. After
drying (sodium sulfate), solvents were evaporated and
the residue was purified by chromatography (ethyl
acetate/heptane, 50/50).
1
1678, 1630 (CvO); H NMR (CDCl3): d ppm 2.36–2.64
(m, 4H), 4.20 (d, J¼18.2 Hz, 1H), 4.26–4.33 (m, 1H), 5.41
(d, J¼18.2 Hz, 1H), 7.65 (d, J¼2.2 Hz, 1H), 8.0 (d, J¼
2.2 Hz, 1H); 13C NMR (CDCl3): d ppm 20.0, 29.5, 39.9,
60.7, 126.1, 132.6, 133.2, 134.2, 134.4, 135.8, 173.9, 191.5.
Anal. calcd for C12H9NO2Cl2: C, 53.36; H, 3.36; N, 5.19; O,
11.85. Found: C, 53.21; H, 3.36; N, 5.06; O, 12.27.
Chromatography (ethyl acetate/heptane, 50/50) of the
residue from crystallization of ketone 4 yields compound
6 (17%), mp 131–1338C (ethyl acetate), IR (KBr): n cm21
1
3175 (OH), 1645 (CvO), 1590, 1570 (CvC); H NMR
Alcohol 5 was obtained as a white solid, yield 7%, mp 129–
1308C (ethyl acetate); IR (KBr): n cm21 3175 (OH), 1645
(CvO), 1610, 1590, 1515, 1460 (CvC), 1H NMR (CDCl3):
d ppm 1.73–2.05 (m, 1H), 2.15–2.45 (m, 2H), 2.49–2.73
(m, 1H), 3.02 (bs, 1H, D2O), 3.79 (s, 3H), 4.16 (d, J¼
14.7 Hz, 1H), 4.77 (d, J¼14.7 Hz, 1H), 5–5.15 (m, 1H),
6.81–6.90 (m, 2H), 7.18–7.26 (m, 2H). Anal. calcd for
C12H15NO3: C, 65.14; H, 6.83; N, 6.33; O, 21.69. Found: C,
65.23; H, 6.78; N, 6.29; O, 21.30.
(CDCl3): d ppm 1.87–2.06 (m, 1H), 2.22–2.51 (m, 2H),
2.53–2.74 (m, 1H), 2.83 (bs, D2O exchangeable, 1H), 4.42
(d, J¼15.5 Hz, 1H), 4.79 (d, J¼15.5 Hz, 1H), 5.14 (d,
J¼5.0 Hz, 1H), 7.22 (dd, J¼8.3, 2.1 Hz, 1H), 7.30 (d,
J¼8.3 Hz, 1H), 7.40 (d, J¼2.1 Hz); 13C NMR (CDCl3): d
ppm 27.8, 28.5, 40.6, 82.4, 127.1, 129.0, 130.3, 132.8,
133.5, 133.8, 175.0. Anal. calcd for C11H11NO2Cl2: C,
50.79; H, 4.26; N, 5.38; O, 12.30. Found: C, 51.07; H, 4.44;
N, 5.20; O, 12.68.
Ketone 3 was obtained in 60% yield, mp 97–988C (acetone)
1
5.1.4. Electrochemical synthesis of 1-(2,4-dichloroben-
zyl)-5-hydroxypyrrolidin-2-one (6). A mixture of acid 2
(3 g, 0.011 mol) and sodium hydroxide (0.04 g, 0.0011 mol)
in water (15 ml) and tetrahydrofuran (15 ml) was stirred into
an undivided jacketed beaker equipped with six graphite-
rods anodes and six graphite-rods cathodes. The carbon-
rods (6.15 mm in diameter), immersed 5 cm into the
suspension, were spaced 10 mm apart. The current was
adjusted at 175 mA; the initial voltage was 9.3 V. After 5 h,
tetrahydrofuran was evaporated, methylene dichloride was
added and the organic phases were washed with an
hydrogenocarbonate solution. The organic phases were
dried (sodium sulfate) then evaporated, giving compound 6
as a white solid, yield 78%, identical to the by-product of the
Friedel–Crafts cyclization of acid 2.
(106–1078C3b); IR (KBr): n cm21 1690 (CvO); H NMR
(CDCl3): d ppm 2.28–2.64 (m, 4H), 3.86 (s, 3H), 4.30 (d,
J¼16.9 Hz, 1H), 4.2–4.4 (m, 1H), 5.22 (d, J¼16.9 Hz, 1H),
7.16 (dd, J¼8.7, 2.9 Hz, 1H), 7.25 (d, J¼8.7 Hz, 1H), 7.54
(d, J¼2.9 Hz, 1H). Anal. calcd for C13H13NO3: C, 67.52; H,
5.67; N, 6.06; O, 20.76. Found: C, 67.59; H, 5.42; N, 5.89;
O, 20.83.
5.1.3. 6,8-Dichloro-1,2,3,5,10,10a-hexahydrobenz[f](in-
dolizine-3,10-dione (4) and 1-(2,4-dichlorobenzyl)-5-
hydroxypyrrolidin-2-one (6). Trifluoroacetic anhydride
(7 ml, 0.046 mol) was added to a suspension of acid 2
(11 g, 0.039 mol) in dichloroethane (100 ml) and the
solution was refluxed for 1 h, giving the mixed anhydride,
1H NMR (CDCl3): d ppm 2.1–2.4 (m, 1H), 2.4–2.62 (m,
1H), 2.62–2.8 (m, 2H), 4.29 (dd, J¼9.7, 3.2 Hz, 1H), 4.39
(d, J¼15 Hz, 1H), 5.06 (d, J¼15 Hz, 1H), 7.31 (s, 1H), 7.31
(d, J¼1.5 Hz, 1H), 7.45 (d, J¼1.5 Hz, 1H). Solvents were
evaporated, dichloroethane (100 ml) was added and the
solution was cooled with a water bath at 88C. Aluminum
chloride (26 g, 0.195 mol) was added (1 h) and the mixture
5.1.5. Methyl 2-(1,3-benzodioxol-5-yl)-2-(2-oxoprolin-1-
yl)acetate (7). Triflic acid (1 ml, 1.71 g, 0.001 mol) was
added via syringe to a mixture of crude acid 35 (97.7 g,
0.45 mol) and benzodioxole (61.1 g, 0.5 mol) in trifluoro-
acetic acid (150 ml). The solution was stirred at room
temperature for three days, solvents were evaporated and

´ ´
R. Akue-Gedu et al. / Tetrahedron 58 (2002) 9239–9247
9246
the residue was dissolved in dichloromethane. The solution
was extracted with aqueous sodium carbonate, aqueous
phases were washed with dichloromethane then acidified
with dilute hydrochloric acid. Dichloromethane was added
and the organic phases were dried (sodium sulfate) then
evaporated giving a yellow oil, 41%, IR (KBr): n cm21 3450
(OH), 1750, 1700, 16960 (CvO), 1510 1495, 1450 (CvC);
1H NMR (CDCl3): d ppm 2.15–2.32 (m, 2H), 2.32–2.53
(m, 1H), 2.59–2.82 (m, 1H), 3.78 (s, 3H), 4.05–4.15 (m,
1H), 5.88 (s, 1H), 5.99 (s, 2H), 6.72–6.88 (m, 3H), 7.61 (bs,
1H), D2O exchangeable); 13C NMR (acetone-d6): d ppm
24.9, 29.3, 52.1, 58.2, 58.7, 101.9, 108.5, 108.8, 122.2,
129.2, 148.3, 148.4, 169.8, 173.3, 175.8. Anal. calcd for
C15H15NO7: C, 56.08; H, 4.71; N, 4.36; O, 34.86. Found: C,
56.32; H, 4.49; N, 4.33; O, 34.99.
2H), 3.72 (s, 3H), 4.05 (dd, J¼8.8, 3.4 Hz), 4.26 (d, J¼
16 Hz, 1H), 4.97 (d, J¼16 Hz, 1H), 7.18–7.30 (m, 2H), 7.39
(s, 1H); 13C NMR (CDCl3): d ppm 22.4, 28.5, 42.1, 51.8,
58.4, 126.9, 128.7, 130.9, 131.7, 133.5, 133.9, 171.4, 174.5.
Anal. calcd for C13H13NO3Cl2: C, 51.68; H, 4.34; N, 4.64;
O, 15.89. Found: C, 51.29, H, 4.42; N, 4.28; O, 15.97.
References
1. Rigo, B.; Cauliez, P.; Fasseur, D.; Sauvage, F. X. Trends
Heterocycl. Chem. 1991, 2, 155–204, and references cited
therein.
2. (a) Rigo, B.; Kolocouris, N. J. Heterocycl. Chem. 1983, 20,
893–898. (b) Rigo, B.; Gautret, P.; Legrand, A.; El
Ghammarti, S.; Couturier, D. Synth. Commun. 1994, 24,
2609–2615. (c) Rigo, B.; Barbry, D.; Couturier, D. Synth.
Commun. 1991, 21, 741–747. (d) Al Akoum Ebrik, S.;
Legrand, A.; Rigo, B.; Couturier, D. J. Heterocycl. Chem.
1999, 36, 997–1000.
5.1.6. Methyl 2-(1,3-benzodioxol-5-yl)-2-(2-oxo-2,5-di-
hydro-1H-pyrrol-1-yl)acetate (8). Trifluoroacetic anhy-
dride (1.7 ml, 2.5 g, 0.012 mol) was added to a mixture of
crude acid 7 (3.3 g, 0.010 mol) and 1,2-dichloroethane
(50 ml). Boron trifluoride etherate (5.2 ml, 5.9 g, 0.042 mol)
was added to the solution. After reflux for 3 h, solvents were
evaporated, dichloromethane was added, organic phases
were washed with sodium carbonate, then dried (sodium
sulfate). After evaporation, the residue was refluxed in ethyl
acetate for 2 h (acticarbon). After evaporation, the residue
was purified by chromatography (ethyl acetate), giving a
56% of compound 8 as a yellow oil, IR (KBr): n cm21 1750,
1700 (CvO) 1600, 1510, 1495 (CvC), 1H NMR (CDCl3):
(d ppm 3.70 (dt, J¼19.1, 1.8 Hz, 1H), 3.78 (s, 3H), 4.42 (dt,
J¼19.1, 1.8 Hz, 1H), 5.98 (s, 3H), 6.22 (dt, J¼6, 1.8 Hz,
1H), 7.6–7.8 (m, 3H), 7.14 (dt, J¼6, 1.8 Hz, 1H). Anal.
calcd for C14H13NO5: C, 61.09; H, 4.76; N, 5.09; O, 29.06.
Found: C, 60.81; H, 4.38; N, 5.41; O, 29.12.
3. (a) Martin, L. L.; Scott, S. J.; Agnew, M. N.; Setescak, L. L.
J. Org. Chem. 1986, 51, 3696–3700. (b) Martin, L. L.; Scott,
S. J.; Setescak, L. L.; Van Engen, D. J. Heterocycl. Chem.
´
1987, 24, 1541–1549. (c) Merour, J. Y.; Cossais, F.; Piroelle,
¨
´
S.; Mazeas, D. J. Heterocycl. Chem. 1994, 31, 87–92.
(d) Buckley, T. F.; Rapoport, H. J. Org. Chem. 1983, 48,
4222–4232. (e) Marchalin, S.; Decroix, B.; Morel, J. Acta
Chem. Scand. 1993, 47, 287–291. (f) Lee, G. S.; Cho, Y. S.;
Shim, S. C.; Kim, W. J.; Eibler, E.; Wiegrebe, W. Arch.
Pharm. 1989, 322, 607–611.
4. (a) Iwasaki, T.; Horikawa, H.; Matsumoto, K.; Miyoshi, M.
J. Org. Chem. 1979, 44, 1552–1554. (b) Rigo, B.; Lelieur,
J. P.; Kolocouris, N. Synth. Commun. 1986, 16, 1587–1591.
(c) Fasseur, D.; Rigo, B.; Cauliez, P.; Debacker, M.; Couturier,
D. Tetrahedron Lett. 1990, 31, 1713–1716. (d) Reactivities
of compounds 5,6 was found similar to those of other
a-hydroxylactams: for instance, refluxing 6 in methanol yields
5.1.7. Methyl 2-hydroxy-2-(2-oxoprolin-1-yl) acetate
(35). A stirred solution of pyroglutamic acid (15 g,
0.116 mol) and methyl 2-hydroxy-2-methoxy acetate (22)
(16.7 g, 0.139 mol) in anhydrous acetone (150 ml) was
refluxed for 60 h. Solvents were evaporated, dichloro-
methane (300 ml) was added, and the solution was extracted
three times with water (100 ml). The aqueous phases were
evaporated, giving a viscous yellow oil which was not
analyzed but directly used for the next step (97% yield); IR
1
quantitavely the methoxy analog, H NMR (CDCl3): d ppm
1.98–2.20 (m, 2H) 2.30–2.48 (d, J¼15.3 Hz, 1H), 4.75
(d, J¼15.3 Hz, 1H), 7.21 (dd, J¼8.2, 1.9 Hz, 1H), 7.29
(d, J¼8.2 Hz, 1H), 7.39 (d, J¼1.9 Hz); 13C NMR (CDCl3): d
ppm 23.8, 28.5, 41.1, 53.3, 89.6, 127.2, 129.2, 131.0, 132.8,
133.8, 134.1, 174.9
1
(KBr): n cm21 3400 (OH), 1740, 1700 (CvO), H NMR
5. (a) Rapoport, H. Lect. Heterocycl. Chem. IV 1978, S-47–S-58.
(b) Rigo, B.; Fasseur, D.; Cherepy, N.; Couturier, D.
Tetrahedron Lett. 1989, 30, 7057–7060.
(CDCl3): d ppm 2–2.3 (m, 1H), 2.3–2.8 (m, 3H), 3.79 (s,
1.5H), 3.83 (s, 1.5H), 4.28–4.41 (m, 0.5H), 4.43–4.58 (m,
0.5H), 5.73 (s, 0.5H), 5.89 (s, 0.5H), 6.6–7.3 (bs, 1H).
´
6. Legrand, A.; Rigo, B.; Henichart, J.-P.; Norberg, B.; Camus,
F.; Durant, F.; Couturier, D. J. Heterocycl. Chem. 2000, 37,
215–227.
´
7. (a) El Ghammarti, S.; Rigo, B.; Mejdi, H.; Henichart, J.-P.;
5.1.8. Methyl N-(2,4-dichlorobenzyl)pyroglutamate (37).
A solution of methyl pyroglutamate (73 g, 0.51 mol) in
toluene (400 ml) was added (1 h) to a well-stirred suspen-
sion of sodium hydride (13 g, 0.54 mol) in toluene (400 ml).
The mixture was refluxed for 1 h, N-methylpyrrolidinone
(200 ml) was added, then a solution of 2,4-dichlorobenzyl
chloride (97.7 g, 0.50 mol) in toluene was added dropwise.
The mixture was refluxed for 2 h. After cooling, water was
added and the aqueous phases were extracted with
dichloromethane. The organic phases were washed with
water, dried (sodium sulfate) then evaporated, giving a yield
of 75%, mp 678C (acetone); IR (KBr): n cm21 1740, 1700
Couturier, D. J. Heterocycl. Chem. 2000, 37, 143–150.
(b) Rigo, B.; El Ghammarti, S.; Couturier, D. Tetrahedron
Lett. 1996, 37, 485–486.
8. Yasuda, S.; Yamamoto, Y.; Yoshida, S.; Hanaoka, M. Chem.
Pharm. Bull. 1988, 36, 4229–4231.
9. (a) Boto, A.; Hernandez, R.; Suarez, E. Tetrahedron Lett.
2000, 41, 2899–2902, and references cited therein.
(b) Maksimov, V. I. Tetrahedron 1965, 21, 687–698.
(c) Hernandez, A.; Marcos, M.; Rapoport, H. J. Org. Chem.
1995, 60, 2683–2691.
1
(CvO), 1590, 1565, 1480 (CvC); H NMR (CDCl3): d
ppm 2.01–2.2 (m, 1H), 2.2–2.40 (m, 1H), 2.40–2.7 (m,
10. Othman, M.; Netchitailo, P.; Decroix, B. J. Heterocycl. Chem.
1997, 34, 225–231.

´ ´
R. Akue-Gedu et al. / Tetrahedron 58 (2002) 9239–9247
9247
11. Yasuda, S.; Yamada, T.; Hanaoka, M. Tetrahedron Lett. 1986,
27, 2023–2026.
¨
12. (a) Jaguar 4.1; Schrodinger Inc.: Portland, OR, USA, 1991–
16. Cauliez, P.; Rigo, B.; Fasseur, D.; Couturier, D. J. Heterocycl.
Chem. 1991, 28, 1143–1146.
17. We have described some more interesting syntheses of
N-arylmethylpyroglutamic acids or esters: (a) Marchalin, S.;
Kadlecikova, K.; Bar, N.; Decroix, B. Synth. Commun. 1998,
28, 3619–3624. (b) Rigo, B.; Couturier, D. J. Heterocycl.
Chem. 1985, 22, 207–208. (c) Rigo, B.; Gautret, P.; Legrand,
2000. (b) Spartan, version 5.1.3; Wavefunction Inc.: 18401
Von Karman Ave., Irvine, CA., 92715 USA.
13. Roth, E.; Altman, J.; Kapan, M.; Ben-Ishai, D. Tetrahedron
1995, 30, 801–810.
´
A.; Henichart, J.-P.; Couturier, D. Synlett 1997, 998–1000.
´
(d) Rigo, B.; Gautret, P.; Legrand, A.; Henichart, J.-P.;
14. Methyl 2-hydroxy-2-methoxyacetate, Acros Organics No.
27692.
15. Kolocouris, N.; Rigo, B. Chim. Chron., New Ser. 1982, 11,
309–317.
Couturier, D. J. Heterocycl. Chem. 1998, 35, 567–573, and
Refs. 2b,2d.