Asymmetric synthesis of 2-amino-3-hydroxynorbornene-2-carboxylic acid derivatives

Article abstract of DOI:10.1021/jo010374q

The enantioselective synthesis of 2-amino-3-hydroxynorbornene-2-carboxylic acid derivatives (5) was studied using the Diels-Alder reaction between cyclopentadiene and different dienophiles, i.e., alkyl 5-oxo-2-phenyloxazol-4-methylenecarbonates (1) or 2-benzoylamino-3-alkoxycarbonyloxy-acrylates (12), operating with different Lewis acids and both with thermal and with ultrasound conditions. The enantioselective Synthesis of the exo/endo compounds 5c,d and 5′c,d was achieved starting from the chiral menthyl acrylates 12b,c using Mg(ClO₄)₂ as the catalyst and ultrasound. The cycloadducts were obtained in very good yield, in mild conditions, in short time, and in good diastereomeric excess (exo, 80%; endo, 87%). Finally, the use of alkylidene-oxazolones or acrylates and EtAlCl₂ or Mg(ClO₄)₂ as the catalyst allowed control of the cycloaddition reaction in favor of the exo or endo products.

Full text of DOI:10.1021/jo010374q

J . Org. Chem. 2001, 66, 6299-6304
6299
Asym m etr ic Syn th esis of
2-Am in o-3-h yd r oxyn or bor n en e-2-ca r boxylic Acid Der iva tives
Giorgio Abbiati,† Francesca Clerici,† Maria Luisa Gelmi,*,† Andrea Gambini,† and
Tullio Pilati‡
Istituto di Chimica Organica, Facolta` di Farmacia, Universita` di Milano, Via Venezian 21,
I-20133 Milano, Italy, and CNR Centro Studio delle Relazioni tra Struttura e Reattivita` Chimica,
via Golgi 19, I-20133 Milano, Italy
marialuisa.gelmi@unimi.it
Received April 11, 2001
The enantioselective synthesis of 2-amino-3-hydroxynorbornene-2-carboxylic acid derivatives (5)
was studied using the Diels-Alder reaction between cyclopentadiene and different dienophiles,
i.e., alkyl 5-oxo-2-phenyloxazol-4-methylenecarbonates (1) or 2-benzoylamino-3-alkoxycarbonyloxy-
acrylates (12), operating with different Lewis acids and both with thermal and with ultrasound
conditions. The enantioselective synthesis of the exo/endo compounds 5c,d and 5′c,d was achieved
starting from the chiral menthyl acrylates 12b,c using Mg(ClO4)2 as the catalyst and ultrasound.
The cycloadducts were obtained in very good yield, in mild conditions, in short time, and in good
diastereomeric excess (exo, 80%; endo, 87%). Finally, the use of alkylidene-oxazolones or acrylates
and EtAlCl2 or Mg(ClO4)2 as the catalyst allowed control of the cycloaddition reaction in favor of
the exo or endo products.
In tr od u ction
the â-position with a heteroatom,1,4-6 we have now
developed a new asymmetric synthesis which affords
2-amino-3-hydroxynorbornene-2-carboxylic acid deriva-
tives in very good yields and enantioselectivity. In fact,
through the use of new chiral building blocks 12b,c, it
has been possible to obtain the diastereomeric derivatives
exo- and endo-5/5′c,d with a good diastereomeric excess.
The chiral moiety elimination will allow obtaining the
enantiomeric 2-amino-3-hydroxynorbornenecarboxylic
acids.
Furthermore, the possibility of finding reaction condi-
tions to favor the formation of endo adduct with respect
to the exo one was considered. Many efforts were made
to attain these targets using the classical Diels-Alder
reaction between cyclopentadiene and different dieno-
philes and changing the reaction conditions (catalyst,
temperature). As known, different approaches to asym-
metric syntheses can be carried out using either chiral
catalysts or chiral starting materials. Both positive and
negative results of these researches are given below.
Recently,1 we reported that 2-amino-3-hydroxynor-
bornane-2-carboxylic acids were synthesized through a
Diels-Alder reaction, as a mixture of exo and endo
adducts in a 70:30 ratio. As reported,2 the skeleton of
these constrained amino acids and their particular func-
tionalization (i.e., the hydroxy group) make this class of
amino acids particularly interesting for biological impli-
cations.
The possibility of finding an asymmetric synthesis of
these amino acids attracted our attention considering
that the biological activity of R-amino acids is often
related to the substituent stereochemistry. Furthermore,
it is known that constrained amino acids incorporated
into bioactive peptides play an important role on the
change of the conformation of the peptides with conse-
quent change of their biological activity.3
As a part of our synthetic project on the preparation
of constrained carbocyclic amino acids functionalized at
† Universita` di Milano.
Resu lts
‡ CNR Centro Studio delle Relazioni tra Struttura e Reattivita`
Chimica.
A first approach to the asymmetric synthesis of 2-amino-
3-hydroxynorbornene-2-carboxylic acid derivatives was
found starting from oxazolone 1a , bearing the ethoxy-
carbonyl group on the methylene carbon, which was
reacted with cyclopentadiene (2) in dichloromethane at
0 °C in the presence of chiral catalyst (R)-1,1′-binaph-
thalene-2,2′-diol-titaniodibromide (BINOL-TiBr2) (25%).
The 1H NMR analysis of the crude reaction mixture
revealed the formation of two cycloadducts, exo-3a and
endo-3a , in a 64:36 ratio. Compounds 3a were not
isolated and were transformed, after solvent elimination,
(1) Clerici, F.; Gelmi, M. L.; Gambini, A. J . Org. Chem. 2000, 65,
6138.
(2) (a) Christensen, H. N.; Handlogten, M. E.; Lam, I.; Tager, H. S.;
Zand, R. J . Biol. Chem. 1969, 244, 1510. (b) Christensen, H. N.; Cullen,
A. M. J . Biol. Chem. 1969, 244, 1521. (c) Tager, H. S.; Christensen, H.
N. J . Am. Chem. Soc. 1972, 94, 968. (d) Zaleski, J .; Wilson, D. F.;
Erecinska, M. J . Biol. Chem. 1986, 261, 14091. (e) Panten, U.;
Zielmann, S.; Langer, J .; Zu¨nkler, B.-J .; Lenzen, S. Biochem. J . 1984,
219, 189. (f) Tellier, F.; Acher, F.; Brabet, I.; Pin, J .-P.; Azerad, R.
Bioorg. Med. Chem. 1998, 6, 195.
(3) (a) Toniolo, C. J anssen Chim. Acta 1993, 11, 10. (b) Kazmierski,
W. M.; Lipkowska, Z. U.; Hruby, V. J . J . Org. Chem. 1994, 59, 1789.
(c) Omstein, P. L.; Melikian, A.; Martinelli, M. J . Tetrahedron Lett.
1994, 35, 5759. (d) Colombo, L.; Di Giacomo, M.; Scolastico, C.;
Manzoni, L.; Belvisi, L.; Molteni, V. Tetrahedron Lett. 1995, 36, 625.
(e) Bisang, C.; Weber, C.; Inglis, J .; Schiffer, C. A.; van Gunsteren, W.
F.; J elesarov, I.; Bosshard, H. R.; Robinson, J . A. J . Am. Chem. Soc.
1995, 117, 7904. (f) Dubuisson, C.; Fukumoto, Y.; Hegedus, L. S. J .
Am. Chem. Soc. 1995, 117, 3697.
(4) Clerici, F.; Gelmi, M. L.; Pocar, D. J . Org. Chem. 1999, 64, 726.
(5) Clerici, F.; Gambini, A.; Gelmi, M. L. J . Org. Chem. 1999, 64,
5764.
(6) Clerici, F.; Gelmi, M. L.; Gambini, A. Tetrahedron, in press.
10.1021/jo010374q CCC: $20.00 © 2001 American Chemical Society
Published on Web 08/24/2001

6300 J . Org. Chem., Vol. 66, No. 19, 2001
Abbiati et al.
Sch em e 1a
F igu r e 1.
Ta ble 2. Diels-Ald er Rea ction s of 1a w ith 2 a n d Ch ir a l
Liga n d s 6-9 a t -20 °C
entry catalysta ligand t (h) exo/endoc endo ratiod exo ratiod
1
2
3
4
5
6
7
A
A
A
6
7
8
9
6
8
9
48
24
100
24
75
150
200
75:25
80:20
70:30
75:25
70:30
75:25
75:25
43:57
47:53
50:50
56:44
50:50
50:50
50:50
50:50
43:57
50:50
66:34
50:50
50:50
50:50
A
Bb
Bb
Bb
a
Reagents and conditions: (i) CH2Cl2, EtAlCl2; (ii) BINOL-
TiBr2, CH2Cl2, 0 °C; (iii) see Table 1; (iv) THF, HCl; (v) CH2N2,
CH2Cl2; (vi) CH2Cl2, EtAlCl2, 25 °C or CH2Cl2, Mg(ClO4), 95 °C.
a
b
A ) Mg(ClO4)2; B ) Ce(OTf)4‚H2O. Molecular sieves. c De-
Ta ble 1. Diels-Ald er Rea ction of 1a w ith 2 in th e
termined by 1H NMR. Determined by HPLC.
d
Absen ce of Ch ir a l Liga n d s
entry catalysta mol equiv T (°C) t (h) exo/endoc yield (%)
30 ratio, and the best yield was observed at room
temperature (entry 2). As shown in Table 1, the use of
LiClO4 (entries 4, 5) and Mg(ClO4)2 (entry 6) increased
the reaction yield giving the exo/endo-3a adducts in the
same 70:30 ratio. The best result was observed using Mg-
(ClO4)2 (entry 6). In fact, the reaction proceeded very
quickly (1 h) using only 0.2 equiv of catalyst per mole of
dienophile. The better efficiency of Mg(ClO4)2 with respect
to LiClO4 can be ascribed to the better solubility of the
former. Interesting results were observed using Ce(OTf)4,
which proved to be a very efficient catalyst (entry 10) also
at very low temperature. Furthermore, a different dis-
tribution of the exo/endo adducts was obtained by chang-
ing the amount of the catalyst (entries 7-10). In fact,
greater amounts of catalyst favored the formation of the
exo adduct. The use of Yb(OTf)3 did not give relevant
results.
Considering these data, we decided to test as catalysts
in the cycloaddition reaction in the presence of chiral
ligands the most efficient of the above, namely, Mg(ClO4)2
and Ce(OTf)4. Four chiral ligands were tried to promote
the asymmetric induction in the cycloaddition process,
i.e., the monosubstituted bis-oxazolines 6 and 8 and the
disubstituted bis-oxazolines 7 and 97 (Figure 1). As shown
(Table 2), the results in enantioselectivity were poor with
the magnesium salt and negative with the cerium salt.
Nevertheless, a reverted ratio in the formation of the exo
enantiomers and endo enantiomers was observed using
the magnesium salt when the ligands 7 and 9 (exo,
entries 2, 4) and 6 and 9 (endo, entries 1, 4) were used.
In a second approach, chiral dienophiles were used.
Two different methods were followed for the synthesis
of these starting materials.
1
2
3
4
5
6
7
8
9
A
A
A
B
B
C
0.2
0.2
0.2
5
40
25
-20
25
-20
25
-20
-50
-50
-20
-20
25
2
3
5
5
18
1
5
8
18
48
48
18
70:30
70:30
70:30
70:30
70:30
70:30
85:15
82:18
75:25
69:31
60
74
70
88
88
90
50
64
25
78
5
0.2
0.2
0.2
0.05
0.01
0.2
0.2
Db
Db
Db
Db
Eb
Eb
10
11
12
68:32
79
a
A ) EtAlCl2; B ) LiClO4; C ) Mg(ClO4)2; D ) Ce(OTf)4‚H2O;
E ) Yb(OTf)4‚H2O. Molecular sieves. c Determined by H NMR.
b
1
into the corresponding acids 4a by treating the crude
reaction mixture with acetic acid and a catalytic amount
of HCl. After chromatography, the fraction containing the
acid derivatives 4a was dissolved in dichloromethane and
treated with an excess of diazomethane, giving the
corresponding methyl esters 5a in 80% overall yield. The
HPLC analysis using a chiral column did not show
enantiomeric excess either in the couple of exo enan-
tiomers or in that of endo enantiomers (Scheme 1).
An alternative approach involved catalyzing the cy-
cloaddition reaction with a Lewis acid in the presence of
chiral bis-oxazolines 6-9. The use of these ligands
requires a catalyst different from ethylaluminum dichlo-
ride, which is usually the best catalyst for the cycload-
dition reactions involving alkylideneoxazolones. Accord-
ingly, the reaction between oxazolone 1a and cyclo-
pentadiene (2) was studied with different Lewis acids.
LiClO4, Mg(ClO4)2, Ce(OTf)4, and Yb(OTf)3 were tested
as catalysts in different reaction conditions (molar ratio,
temperature, presence of molecular sieves) operating in
dichloromethane as the solvent, and the results are
reported in Table 1 in comparison with EtAlCl2. The
standard reaction in the presence of EtAlCl2 (entries 1-3)
gave the corresponding cycloadducts exo/endo-3a in a 70:
To introduce a stereocenter in the carbonate group of
the oxazolone, the menthyl carbonate Z-1b was synthe-
(7) (a) Nishiyama, H.; Kondo, M.; Nakamura, T.; Itoh, K. Organo-
metallics 1991, 10, 500. (b) Carbone, P.; Desimoni, G.; Faita, G.;
Filippone, S.; Righetti, P. Tetrahedron 1998, 54, 6099.

Asymmetric Synthesis of Amino Acid Derivatives
J . Org. Chem., Vol. 66, No. 19, 2001 6301
Sch em e 2a
terified to 12a (95%) by reaction with diazomethane at
0 °C (Scheme 4).
The reactivity of compound 12a with cyclopentadiene
was tested with different Lewis acids (EtAlCl2, Mg(ClO4)2,
Ce(OTf)4, and Yb(OTf)3) and in different reaction condi-
tions (amount of catalyst, temperature). Compounds 5a
were obtained (Scheme 1), but a low reactivity of the
substituted acrylate double bond compared to that of
oxazolones 1 was observed confirming the data reported
in the literature. In fact, only EtAlCl2 and Mg(ClO4)2 were
efficient catalysts, but they worked only at higher tem-
peratures and afforded lower reaction yields (see Experi-
mental Section) compared to the corresponding reactions
with oxazolone 1a (see Table 1, entries 3, 6). However,
an interesting observation was made. With the use of
EtAlCl2 as the catalyst, a reversed exo/endo ratio was
obtained with respect to the corresponding reaction with
oxazolone 1a , which afforded (Table 1, entries 1-3) the
exo/endo compounds in a 70:30 ratio. Instead, when
acrylate 12a was used as the dienophile, the exo/endo
product ratio was 25:75. An explanation is given by
considering that EtAlCl2 coordinates the lactone group
in alkylideneoxazolones, whereas in the case of amino
acrylate derivatives, the coordination occurs at the amide
carbonyl group.9 The acrylate double bond is substituted
at C-3, and both the cis-standing carbonate group and
coordinated amido group (Scheme 5, intermediate B)
increase the steric hindrance of this side favoring the
approach of the diene from the other side so as to give
preferably the endo adduct. This is not true in the case
of the lactone (Scheme 5, intermediate A) in which the
bulky catalyst hinders the opposite side. However, sec-
ondary interactions also may be of importance in ex-
plaining these results.10
In view of the positive results of the cycloaddition
reaction of cyclopentadiene with acrylate 12a , the syn-
thesis of chiral acrylates 12b and 12c, i.e., (+)-menthyl
and (-)-menthyl esters, respectively, was faced. Many
problems were posed for the reasons reported above and
for the lower reactivity of menthol compared to methanol.
The difficulties were overcome by reacting hydroxyox-
azolone 10 with (+)-menthol or (-)-menthol and bis-
(dibutylchlorotin)oxide as catalyst.11 The menthyl 3-hy-
droxyacrylates 13a and 13b were obtained in 85% yield
and were reacted with ethyl choroformate in the presence
of triethylamine giving the corresponding chiral com-
pounds 12b and 12c (95%) (Scheme 4).
a
Reagents and conditions: CH2Cl2, TEA, -5 °C.
sized by the same procedure as that used in the prepara-
tion of the corresponding ethyl carbonate 1a . The chiral
compound 1b was obtained in good yield (78%) starting
from 4-hydroxymethylene-5(4H)-oxazolone (10) and (+)-
menthyl chloroformate at -5 °C in methylenechloride
and in the presence of triethylamine (Scheme 2).
Oxazolone Z-1b was reacted with cyclopentadiene (2)
in the presence of ethylaluminum dichloride at -20 °C.
The mixture was quenched with EtOH, and a catalytic
amount of AcOH/HBr was added. After chromatography,
the 1H NMR analysis of the fraction containing the
mixture of esters 5b and 5′b demonstrated that four
diastereomeric esters were formed. The exo/ endo ratio
was 60:40, and no significant diastereomeric excess was
observed in each exo and endo couple (Scheme 3).
From these data, we can conclude that the bulky
menthyl group increases the formation of the endo adduct
as compared to the ethyl group but the effect of the chiral
center on the diastereoselection is not relevant probably
because of its distance from the reaction site.
The above results proved that oxazolones 1 probably
are unsuitable starting materials for the enantioselective
synthesis of 2-amino-3-hydroxynorbornene-2-carboxylic
acids. Accordingly, our efforts were directed in the
synthesis of different substrates, i.e., the chiral amino-
acrylates 12b,c, functionalized at C-3 with the ethyl
carbonate group, which are the open-chain synthetic
equivalents of oxazolones.
However, this synthetic approach suffered from a
potential drawback. In fact, it is well-known that 3-sub-
stituted aminoacrylates do not react with dienes in the
Diels-Alder reaction.8 To verify the feasibility of this
method and to ascertain the reactivity of the double bond,
the achiral methyl 2-benzoylamino-3-ethoxycarbolyloxy-
acrylate 12a was first synthesized. Many approaches
were tried starting from oxazolone 1a . Because oxazo-
lones 1 can react with nucleophiles both at the lactone
carbonyl group and at the carbonate group and also at
the double bond by an addition-elimination mechanism,
a mixture of compounds was obtained after reaction of
methanol with 1a in the presence of acidic or basic
catalysts. Instead, the ester 12a was successfully ob-
tained by transforming oxazolone 1a into the correspond-
ing acrylic acid 11 (60%) with a catalytic amount of
hydrochloric acid in THF. Compound 11 was then es-
In principle, the cycloaddition reaction of cyclopenta-
diene with 12b or 12c can afford four diastereomers: two
couples of adducts exo-5c, exo-5′c, endo-5c, and endo-5′c
and two couples of adducts exo-5d , exo-5′d , endo-5d , and
endo-5′d , respectively (Scheme 3).
The cycloaddition reaction was catalyzed both with
EtAlCl2 and with Mg(ClO4)2. The first catalyst was
unsuccessful both at room temperature and at 45 ˚C; only
decomposition products were found. In contrast, Mg-
(ClO4)2 was effective at 95 °C (Table 3, entries 2, 3), and
the starting acrylates were partially transformed into the
corresponding menthyl 2-phenyloxazol-4-ylcarboxylate.
A remarkable improvement in the yield of the above
diastereomers was obtained when the cycloaddition reac-
(9) Cativiela, C.; Lopez, P.; Mayoral, J . A. Tetrahedron: Asymmetry
1991, 2, 1295.
(10) Bueno, M. P.; Cativiela, C.; Finol, C.; Mayoral, J . A. Can. J .
Chem. 1987, 65, 2182.
(11) Orita, A.; Mitsutome, A.; Otera, J . J . Org. Chem. 1998, 63, 2420.
(8) Cativiela, C.; Diaz de Villegas, M. D.; Mayoral, J . A. Tetrahedron
1993, 49, 677.

6302 J . Org. Chem., Vol. 66, No. 19, 2001
Abbiati et al.
Sch em e 3a
a
Reagents and conditions: (i) CH2Cl2, EtAlCl2, -20 °C; (ii) EtOH, AcOH/HBr; (iii) see Table 3.
Ta ble 3. Diels-Ald er Rea ction s of 2 w ith Ch ir a l Acr yla tes 12b,c
entry
reagent
catalyst
methoda
products
yieldb
exo/endoc
exo-5/exo-5′ (de)d
endo-5/endo-5′ (de)d
1
2
3
4
5
12c
12b
12c
12b
12c
EtAlCl2
)))
∆
5d /5′d
5c/5′c
5d /5′d
5c/5′c
5d /5′d
36
50
50
90
90
70:30
77:23
77:23
78:22
77:23
23:77 (55)
90:10 (79)
10:90 (79)
89:11 (78)
10:90 (79)
87:13 (74)
7:93 (87)
93:7 (87)
7:93 (87)
92:8 (86)
Mg(ClO4)2
Mg(ClO4)2
Mg(ClO4)2
Mg(ClO4)2
∆
)))
)))
a
b
d
∆: thermal reaction. ))): ultrasound reaction. Mixture of exo/endo isolated compounds. c Determined by 1H NMR. Determined by
HPLC.
Sch em e 4a
Sch em e 5
tiomer with respect to the other was produced from the
(+)-menthyl derivative 12b (exo-5c, [R]25 +63; endo-5c,
D
[R]25 +9) or from the (-)-menthyl derivative 12c (exo-
D
5′d , [R]25 -63; endo-5′d , [R]25 -9).
D
D
a
Reagents and conditions: (i) THF/H3O+; (ii) CH2Cl2, CH2N2,
0 °C; (iii) C6H6, bis(dibutylchlorotin)oxyde, (+)- or (-)-menthol,
reflux; (iv) CH2Cl2, ClCO2Et, TEA.
The absolute configuration (1S,2S,3R,4R) of the major
endo compound 5′c was unequivocally determined by
X-ray crystallography. Any attempt to prepare crystals
of the exo adducts failed.
tion was performed with ultrasound. Both EtAlCl2 and
Mg(ClO4)2 had a catalytic effect, but with the first
catalyst, a long reaction time (48 h) and a relatively low
yield (Table 3, entry 1) were observed. In contrast,
excellent yields (Table 3, entries 4, 5) were achieved with
the second catalyst. Furthermore, the reaction time was
shorter (15 h) than that obtained by the thermal method
(24 h).
The configuration of the major endo adduct 5′c, ob-
tained from the (+)-menthyl ester 12b, allows us to
conclude that the isopropyl group of the menthyl sub-
stituent should have a shielding effect on the CR-Si side
of the double bond favoring the attack on the CR-Re side
(Figure S3, Supporting Information).
To support this, theoretical studies12 to evaluate the
conformational energy of 12b were carried out through
The 1H NMR and HPLC analyses of both the crude
reaction mixtures obtained by the thermal method and
those obtained by ultrasound showed analogous results,
and in each exo and endo couple a different distribution
existed (Table 3). Furthermore, an excess of one enan-
(12) Calculations were done with the Hyperchem program (6.02
release), Hypercube, 1115 NW 4th Street, Gainesville, FL 32601, http://
www.hyper.com.

Asymmetric Synthesis of Amino Acid Derivatives
J . Org. Chem., Vol. 66, No. 19, 2001 6303
a conformational search at the MM+ level;13 the results
show that most of the conformers with lower steric energy
are characterized by the shielding of the isopropyl group
on the Si face of the alkene 12b. Moreover, the most
representative structures were minimized using the
AM114 semiempirical method and obtaining geometries
and energy differences between distinct conformers in
agreement with the data obtained at the MM+ level. In
particular, the semiempirical calculations confirmed that
the lowest energy conformer has a lower hindrance on
the Re face and is energetically more stable by about 4
kJ mol-1 compared to the lower energy structure in which
the isopropyl group shields the Re side.
7.30 (s, 1H), 4.80-4.60 (m, 1H), 2.20-0.80 (m, 18H). Anal.
Calcd: C, 67.89; H, 6.79; N, 3.77. Found: C, 67.80; H, 6.71;
N, 3.72.
(Z)-2-Ben zoyla m in o-3-eth oxyca r bon yloxya cr ylic Acid ,
11. Oxazolone 1a (700 mg, 2.7 mmol) was dissolved in THF
(30 mL). H2O (5 mL) and a catalytic amount of HCl (37%) were
added, and the reaction mixture was stirred at room temper-
ature for 30 min. The solvent was evaporated to dryness, and
the crude reaction mixture was taken up with CH2Cl2. The
pure acid 11 (430 mg, 60%) was separated as a solid and
filtered: mp 160 C°, dec (EtOH). IR (Nujol): 3270, 1780, 1700,
1
1650 cm-1. H NMR: δ 9.68 (s, 1H, exch), 7.96-7.47 (m, 5H),
7.86 (s, 1H), 4.28 (q, J ) 7.0, 2H), 1.26 (t, J ) 7.0, 3H). Anal.
Calcd: C, 55.90; H, 4.69; N, 5.02. Found: C, 55.78; H, 4.79;
N, 4.93.
The data reported in the literature15 for the cycload-
dition reactions of cyclopentadiene with (-)-menthyl
esters of unsubstituted aminoacrylates showed that the
H-1 and H-4 configurations of the norbornene ring are
opposite in the exo and endo series for the major isomers,
but for H-2, the same configuration holds. Furthermore,
the crowded side of the olefin in compounds 12b,c
corresponds to the chiral acrylates of the literature.
In agreement with the above considerations and with
theoretical results, we hypothesize for the major exo
adduct of the (+)-menthyl series (compound 5c) the 1R,
2S, 3R, 4S configuration.
In conclusion, many efforts were made to find an
asymmetric synthesis of exo- and endo-2-amino-3-hy-
droxynorbornenecarboxylic acids, and this aim has been
successfully attained.
As far as the stereoselectivity is concerned, some
approaches were not positive, but in all cases, a lot of
information was deduced in terms of catalyst choice and
reaction conditions. Mg(ClO4)2 is the best catalyst, and
the use of ultrasound allowed us to obtain the cycload-
ducts in very good yields, in mild conditions, and in a
short time.
Meth yl (Z)-2-Ben zoyla m in o-3-eth oxyca r bon yloxya cr y-
la te, 12a . Acid 11 (340 mg, 1.22 mmol) was suspended in CH2-
Cl2 (10 mL). The reaction was cooled at 0 °C, and an excess of
an ethereal solution of CH2N2 was added in small portions
until the solid was dissolved. After solvent evaporation, the
crude reaction mixture was chromatographed (CH2Cl2/Et2O,
100:1 to 50:1). The ester 12a was obtained in pure form (300
mg, 95%): mp 104 °C (CH2Cl2). IR (Nujol): 3270, 1770, 1730,
1
1660 cm-1. H NMR: δ 8.09 (s, 1H), 7.91-7.43 (m, 5H), 7.39
(s, 1H, exch), 4.33 (q, J ) 7.1, 2H), 3.84 (s, 3H), 1.32 (t, J )
7.1, 3H). 13C NMR: δ 165.2, 164.6, 151.3, 140.2, 133.5, 132.2,
128.7, 127.7, 113.4, 65.9, 52.6, 14.1. Anal. Calcd: C, 57.32; H,
5.16; N, 4.78. Found: C, 57.25; H, 5.10; N, 4.71.
(+)-Men t h yl (Z)-2-Ben zoyla m in o-3-h yd r oxya cr yla t e,
13a . Under nitrogen atmosphere, anhydrous oxazolone 10 (756
mg, 4 mmol) was suspended in anhydrous benzene (30 mL).
(+)-Menthol (416 mg, 2.67 mmol) and bis(dibutylchlorotin)-
oxide (243 mg, 0.44 mmol) were added, and the mixture was
refluxed for 24 h, after which the starting reagents were
dissolved. After solvent evaporation, the crude reaction mix-
ture was chromatographed on silica gel (CH2Cl2) giving ester
13a (800 mg, 85%) as an oil: [R]25 +51.
D
13a ,b. IR (Nujol): 3350, 1710, 1660, 1640 cm-1. H NMR:
1
δ 12.24 (d, J ) 12.0, 1H, exch), 8.54 (s, 1H), 7.86-7.42 (m,
6H), 4.89-4.76 (m, 1H), 2.19-0.69 (m, 18H). Anal. Calcd: C,
69.53; H, 7.88; N, 4.06. Found: C, 69.45; H, 7.91; N, 3.96.
(+)-Men th yl (Z)-2-Ben zoylam in o-3-eth oxycar bon yloxy-
a cr yla te, 12b. Under nitrogen atmosphere, the ester 13a (600
mg, 1.74 mmol) was dissolved in anhydrous CH2Cl2 (15 mL).
The solution was cooled at -10 °C, and ethyl chlorocarbonate
(0.195 mL, 2 mmol) was added. A solution of TEA (0.285 mL,
2 mmol) dissolved into CH2Cl2 (5 mL) was dropped in. After 3
h, the organic layer was washed with HCl (10 mL, 10%) and
dried over Na2SO4, and the solvent was evaporated. After
crystallization of the crude reaction mixture with i-Pr2O/n-
pentane, the pure compound 12b (690 mg, 95%) was isolated:
Finally, the use of alkylidene-oxazolones or acrylates
and EtAlCl2 or Mg(ClO4)2 as the catalysts allowed the
control of the cycloaddition reaction in favor of the exo
or endo products.
Exp er im en ta l Section
Gen er a l. Melting points are uncorrected. IR spectra of the
1
Nujol method were measured using NaCl plates. H and 13C
NMR were recorded in CDCl3 at 200 and 50 MHz, respectively,
with CHCl3 as internal standard. Coupling constants (J ) are
given in hertz. Ethanol-free CH2Cl2 was used in all experi-
ments. Oxazolones 1a , 3a , 4a ,1 and 1016 are known compounds.
(Z)-(+)-Men t h yl 5-Oxo-2-p h en yloxa zol-4-m et h ylen e-
ca r bon a te, 1b. Under nitrogen atmosphere, anhydrous ox-
azolone 10 (189 mg, 1 mmol) was suspended in anhydrous
CH2Cl2 (5 mL). The solution was cooled at -5 °C, and menthyl
chlorocarbonate (0.236 mL, 1.1 mmol) was added. A solution
of triethylamine (TEA) (0.155 mL, 1.1 mmol) in CH2Cl2 (5 mL)
was added dropwise. After 3 h, the organic layer was washed
with HCl (10 mL, 10%) and dried over Na2SO4, and the solvent
was evaporated. After crystallization of the crude reaction
mixture with i-Pr2O, the pure compound (Z)-1b (289 mg, 78%)
[R]25D +44. Mp 95 °C. IR (Nujol): 3300, 1770, 1730, 1655 cm-1
.
1H NMR: δ 8.05 (s, 1H), 7.91-7.44 (m, 5H), 7.36 (s, 1H, exch),
4.92-4.79 (m, 1H), 4.35 (q, J ) 7.0, 2H), 2.19-0.78 (m, 18H),
1.37 (t, J ) 7.0, 3H). 13C NMR: δ 165.0, 163.9, 151.5, 139.3,
133.8, 132.1, 128.7, 127.7, 113.7, 76.2, 65.8, 47.1, 40.8, 34.2,
31.5, 26.5, 23.6, 22.0, 15.5, 14.1. Anal. Calcd: C, 66.15; H, 7.49;
N, 3.36. Found: C, 66.04; H, 7.41; N, 3.29.
Gen er a l P r oced u r es for t h e Diels Ald er R ea ct ion .
Meth od a fr om 1a a n d BINOL-TiBr 2. Under argon atmo-
sphere, anhydrous (R)-(+)-1,1′-binaphthole (143 mg, 0.5 mmol)
was dissolved in anhydrous CH2Cl2 (15 mL) in the presence
of molecular sieves (3 g, 4 Å, powdered). The (i-PrO)2TiBr2 (163
mg, 0.5 mmol) was added, and the solution was stirred at room
temperature for 1 h. The solution was cooled at 0 °C, and
cyclopentadiene (2) (528 mg, 8 mmol) and oxazolone Z-1a (524
mg, 4 mmol) were added. Stirring was continued for 24 h. The
solution was evaporated, and the reaction mixture was taken
up with THF (10 mL). A catalytic amount of HCl (36%) was
added, and the mixture was stirred at room temperature for
4 h. After solvent elimination, the crude reaction mixture was
chromatographed (CH2Cl2/MeOH, 1:0 to 0:1). The fraction
corresponding to a mixture of the acids exo- and endo-4a was
dissolved in CH2Cl2 (10 mL), and an excess of CH2N2 was
added until the acids were transformed into the corresponding
was isolated: mp 98 °C. [R]25 +43. IR (Nujol): 1800, 1760,
D
1680 cm-1. H NMR: δ 8.20-7.40 (m, 5H), 8.10 (s, 1H, exch),
1
(13) MM+ is an extension of the MM2 force field developed by
Allinger. Allinger, N. L. J . Am. Chem. Soc. 1977, 99, 8127.
(14) Dewar, M. J . S.; Zoebish, E. G.; Healy, E. F.; Stewart, J . J . P.
J . Am. Chem. Soc. 1985, 107, 3902.
(15) (a) Cativiela, C.; Mayoral, J . A.; Pires, E.; Royo, A. J . Tetrahe-
dron 1995, 51, 1295. (b) Avenoza, A.; Cativiela, C.; Ferna´ndez-Recio,
M. A.; Peregrina, J . M. Tetrahedron: Asymmetry 1996, 7, 721.
(16) Blad, J . M.; Stammer, C. H.; Varughese, K. I. J . Org. Chem.
1984, 49, 1634.

6304 J . Org. Chem., Vol. 66, No. 19, 2001
Abbiati et al.
esters, exo- and endo-5a (4 h). The mixture of esters 5 was
obtained in 80% overall yield (exo/endo, 64:36) and was
analyzed by HPLC using an OD column (250 mm × 4.6 mm;
n-hexane/iPrOH, 85:15; T ) 30 °C; flow ) 0.8 mL/min, λ )
254).
Meth od b fr om 1a a n d Differ en t Lew is Acid s. Ox-
azolone Z-1a (52.2 mg, 0.2 mmol) was dissolved in anhydrous
CH2Cl2 (2 mL) under nitrogen atmosphere and under stirring.
Diene 2 (15 mg, 0.22 mmol) and Lewis acid were added.
Reaction conditions and results are given in Table 1. The crude
reaction mixture was quickly filtered on silica gel using
CH2Cl2 as the eluant, and the fraction corresponding to the
exo/endo adducts 3a was analyzed by 1H NMR to calculate
the exo/endo ratio. Yields and exo/ endo ratio are given in
Table 1.
It was possible to separate the diasteromeric compounds by
column flash chromatography on silica gel (230-400 mesh
ASTM; CH2Cl2/AcOEt, 50:1). In a typical procedure, a mixture
of compounds exo-5c, 5′c and endo-5c, 5′c (500 mg) were
separated (column, 3.5 mm × 25 cm; flow ) 20 mL/min) giving
four fractions containing exo-5c (200 mg), a mixture of exo-
5c, 5′c (150 mg), endo-5′c (120 mg), and a mixture of endo-5c,
5′c (40 mg). The fractions were analyzed by HPLC using a
silica Hipersil column (250 mm × 4.6 mm; CH2Cl2/AcOEt, 100:
3; T ) 30 °C, flow ) 0.8 mL/min, λ ) 254).
Met h yl (1R*,2S*,3R*,4S*)-2-Ben zoyla m in o-3-et h oxy-
ca r bon yloxy-bicyclo[2.2.1]ep t-5-en e-2-ylca r boxyla te, exo-
5a . Mp 170 °C (CH2Cl2/i-Pr2O). IR νmax: 3350, 1730, 1700, 1640
cm-1 1 H NMR: δ 7.80-7.41 (m, 5H), 6.57 (s, 1H, exch), 6.40-
.
6.27 (bs, 2H), 5.62 (d, J ) 3.9, 1H), 4.22 (q, J ) 7.0, 2H), 3.78
(s, 3H), 3.74 (bs, 1H), 3.27, (bs, 1H), 1.96, 1.75 (AB system,
J ) 10.2, 2H), 1.28 (t, J ) 7.0, 3H). Anal. Calcd: C, 63.49; H,
5.89; N, 3.36. Found: C, 63.40; H, 5.92; N, 3.41.
Meth od c fr om 1a a n d Lew is Acid s a n d Ch ir a l Liga n d s
6-9. A solution of oxazolone Z-1a (52.2 mg, 0.2 mmol)
dissolved in anhydrous CH2Cl2 (2 mL) was cooled at -20 °C
under nitrogen atmosphere and under stirring. Diene 2 (15
mg, 0.22 mmol), Mg(ClO4)2 or Ce(OTf)4‚H2O (0.022 mmol), and
the chiral ligand (6-9) (0.024 mmol) were added, and the
reaction was stirred for the time indicated in Table 2. The
Met h yl (1R*,2R*,3S*,4S*)-2-Ben zoyla m in o-3-et h oxy-
car bon yloxy-bicyclo[2.2.1]ept-5-en e-2-ylcar boxylate, en do-
5a . Mp 138 °C (CH2Cl2/i-Pr2O). IR νmax: 3350, 1730, 1700, 1640
cm-1. 1H NMR: δ 7.83-7.41 (m, 5H), 6.98 (s, 1H, exch), 6.40-
6.36, 6.23-6.18 (2 m, 2H), 5.17 (d, J ) 1.8, 1H), 4.22 (q, J )
7.3, 2H), 3.70 (s, 3H), 3.35 (bs, 1H), 2.96 (bs, 1H), 2.04, 1.81
(AB system, J ) 9.6, 2H), 1.28 (t, J ) 7.1, 3H). Anal. Calcd:
C, 63.49; H, 5.89; N, 3.36. Found: C, 63.43; H, 5.91; N, 3.40.
Eth yl 2-Ben zoyla m in o-3-(+)-m en th yloxyca r bon yloxy-
bicyclo[2.2.1]ept-5-en e-2-car boxylates, exo-5,5′b an d en do-
5,5′b. 1H NMR (mixture of diastereomers): δ 7.80-7.40 (m,
1
reaction was monitored by H NMR until consumption of the
starting oxazolone 1a was complete. The crude reaction
mixture was filtered on a Supelclean LC-SI column (3 mL)
and then analyzed by HPLC using a Chiralcel OF column (250
mm × 4 mm; n-hexane/iPrOH, 100:3; T ) 30 °C; flow ) 1 mL/
min, λ ) 254). The exo/endo ratio and the ratio of exo and endo
enantiomers are given in Table 2.
Meth od d Sta r tin g fr om Oxa zolon e 1b. Oxazolone Z-1b
(371 mg, 1 mmol) was dissolved in anhydrous CH2Cl2 (10 mL)
under nitrogen atmosphere and stirring. The solution was
cooled at -20 °C, and diene 2 (246 mg, 4 mmol) and EtAlCl2
(225 µL, 1.8 M in toluene) were added. After 20 h, the solvent
was evaporated, and EtOH (10 mL) and a mixture of AcOH/
HBr (0.1 mL) were added. Stirring was continued for 5 h, after
which the solvent was eliminated and the crude reaction
mixture was chromatographed on silica gel (CH2Cl2/Et2O, 1:0
to 10:1). A fraction containing the mixture of esters exo-5b,
exo-5′b and endo-5b, endo-5′b was obtained (85%), which was
analyzed by 1H NMR from which the exo/ endo ratio (60:40)
and diastereomeric excess (de) (exo-5b/exo-5′b, 1:1; endo-5b/
endo-5′b, 1:1) were determined.
5H), 6.94 (s, Hendo, exch), 6.90 (s, Hendo, exch), 6.53 (s, Hexo
,
exch), 6.51 (s, Hexo, exch), 6.35-6.15 (m, 2H), 5.64-5.23 (m,
1H), 4.56-4.47 (m, 1H), 4.28-4.11 (m, 2H), 3.70 (bs, Hexo), 3.30
(bs, Hendo), 3.20 (bs, Hexo), 2.91 (bs, Hendo), 2.80-0.62 (m, 23H).
(+)-Men th yl (1R,2S,3R,4S)-2-Ben zoyla m in o-3-eth oxy-
ca r bon yloxybicyclo[2.2.1]ep t-5-en e-2-ca r boxyla te, exo-
5c. Oil; [R]25 +63.
D
exo-5c/5′d . IR (Nujol): 3400, 1756, 1729, 1667 cm-1
.
1H
NMR: δ 7.73-7.40 (m, 5H), 6.56 (s, 1H, exch), 6.38-6.32 (m,
2H), 5.56 (d, J ) 3.7, 1H), 4.81-4.68 (m, 1H), 4.23 (q, J ) 7.3,
2H), 3.77 (bs, 1H), 3.27 (bs, 1H), 2.20-2.05 (m, 1H), 1.93 (d,
J ) 10.3, 1H), 1.90-0.85 (m, 19H), 1.30 (t, J ) 7.3, 3H). 13C
NMR: δ 171.5, 166.6, 153.8, 136.6, 136.2, 134.1, 131.6, 128.5,
127.0, 80.6, 75.7, 65.5, 64.5, 49.6, 47.1, 46.1, 43.9, 40.3, 34.3,
31.5, 25.8, 23.0, 22.1, 20.9, 15.8, 14.2. Anal. Calcd: C, 69.53;
H, 7.72; N, 2.90. Found: C, 69.45; H, 7.68; N, 2.85.
exo-5′c/5d . 1H NMR: significative signals at δ 5.56 (d, 1H),
4.13 (q, 2H).
Meth od e fr om Acr yla te 12a a n d EtAlCl2 or Mg(ClO4)2.
Acrylate Z-12a (245 mg, 1 mmol) was suspended in anhydrous
CH2Cl2 (5 mL) under nitrogen atmosphere and under stirring.
Diene 2 (264 mg, 4 mmol) and Lewis acid were added (EtAlCl2,
0.3 mmol, T ) 25 °C, t ) 3 h, 45%; Mg(ClO4)2, 1 mmol, T )
95 °C, t ) 40 h, 50%). The crude reaction mixture was
chromatographed on silica gel (n-pentane/AcOEt, 50:1 to 2:1),
and the fraction corresponding to the exo/ endo adducts 5a was
(+)-Men th yl (1S,2S,3R,4R)-2-Ben zoyla m in o-3-eth oxy-
ca r bon yloxybicyclo[2.2.1]ep t-5-en e-2-ca r boxyla te, en d o-
5′c. Mp 150 °C (iPr2O). [R]25 +9.
D
en d o-5′c/5d . 1H NMR: δ 7.81-7.40 (m, 5H), 6.90 (s, 1H,
exch), 6.38-6.30, 6.20-6.17 (m, 2H), 5.12 (d, J ) 2.2, 1H),
4.74-4.60 (m, 1H), 4.17 (q, J ) 7.3, 2H), 3.38 (bs, 1H), 2.95
(bs, 1H), 2.10-2.00 (m, 1H), 2.04 (d, J ) 9.2, 1H), 2.00-0.68
(m, 18H), 1.30 (t, J ) 7.3, 3H). 13C NMR: δ 171.1, 166.8, 153.8,
138.0, 134.3, 133.8, 131.6, 128.5, 127.0, 79.0, 75.4, 65.6, 64.4,
49.4, 47.9, 47.0, 45.8, 40.3, 34.3, 31.4, 25.7, 23.0, 22.0, 20.8,
15.8, 14.2. Anal. Calcd: C, 69.53; H, 7.72; N, 2.90. Found: C,
69.50; H, 7.74; N, 2.87.
1
analyzed by H NMR.
Meth od f: Th er m a l Meth od fr om Acr yla tes 12b,c a n d
Mg(ClO4)2. Acrylate Z-12b or 12c (500 mg, 1.2 mmol) and Mg-
(ClO4)2 (258 mg, 1.2 mmol) were suspended in anhydrous
toluene (5 mL) under nitrogen and stirring. Diene 2 (948 mg,
5.3 mmol) was added, and the reaction mixture was warmed
at 95 °C. After 24 h, the crude reaction mixture was chro-
matographed on silica gel (n-pentane/AcOEt, 50:1 to 2:1), and
the fraction corresponding to the exo/endo adducts 5c, 5′c or
5d , 5′d was analyzed by 1H NMR to calculate the exo/endo
ratio. By HPLC with a Chiral OF column (250 mm × 4 mm;
n-hexane/iPrOH, 85:15; T ) 25 °C; flow ) 0.8 mL/min, λ )
254), the diasteromeric excess was calculated. Yields, exo/ endo
ratio, and de are given in Table 3.
Meth od g: Ultr a sou n d Meth od fr om Acr yla tes 12b,c
a n d Lew is Acid s. In a sealed tube, acrylate Z-12b or 12c
(500 mg, 1.2 mmol), the diene 2 (948 mg, 5.3 mmol), and a
Lewis acid were suspended in anhydrous toluene (5 mL), and
the solution was sonicated (EtAlCl2, t ) 48 h; Mg(ClO4)2, t )
15 h). The crude reaction mixture was chromatographed on
silica gel, as described in method f, giving a mixture of exo/
endo adducts 5c, 5′c (520 mg) or 5d , 5′d (525 mg), which was
analyzed by 1H NMR and by HPLC as reported above. Yields,
exo/endo ratio, and de are given in Table 3.
en d o-5c/5′d . 1H NMR: δ 7.81-7.43 (m, 5H), 6.80 (s, 1H,
exch), 6.38-6.30, 6.23-6.18 (2 m, 2H), 5.19 (d, J ) 1.9, 1H),
4.68-4.55 (m, 1H), 4.21 (q, J ) 7.4, 2H), 3.33 (bs, 1H), 2.97
(bs, 1H), 2.10-0.70 (m, 20H), 1.27 (t, J ) 7.4, 3H).
Ack n ow led gm en t. We thank Prof. Giuseppe Faita
(University of Pavia, Italy) for chiral ligands and
MURST for financial support.
Su p p or tin g In for m a tion Ava ila ble: X-ray structure of
compound endo-5′c, computational chemistry for compound
12b, and specific rotation values for compounds 12c, 13b, exo-
5′d , and endo-5d . This material is available free of charge via
the Internet at http://pubs.acs.org.
J O010374Q