Diels-Alder Reactions of Amino Acid-Derived Trienes

Article abstract of DOI:10.1021/jo990459f

Triene precursors (1a-e, 2a-k) were constructed for substrate-controlled asymmetric Diels-Alder reactions. Boc-L-phenylalanal and Boc-L-valinal were condensed with triethyl phosphonoacetate or 2-phosphonopropionate to generate the α,β-unsaturated esters as dienophiles. Removal of the Boc group to give free amines 4a-d, which after, or without N-benzylation, were treated with 3,5-hexadienoyl chloride to give 1a-e, or with 2,4-hexadienoyl chloride to afford 2a-f. The trienes 2g-i were prepared via reductive alkylation of amines 4a-i with 2,4-hexadienal. The secondary amide triene 1a failed to yield any Diels-Alder product when heated at 170°C. The tertiary amide trienes 1b-e produced in refluxing toluene the major cycloaddition products that were cis-fused and derived from the exo transition states. Trienes 2a-k underwent surprisingly facile Diels-Alder reactions to produce the major trans-fused isomers that were derived from the endo transition states. For trienes 2b-h and 2j,k, Diels-Alder reactions proceeded at room temperature. For the primary amide 2a, the Diels-Alder reaction proceeded smoothly in refluxing toluene. The tertiary amide triene 22 was constructed to have two electron-withdrawing ester substituents at the termini of the triene. The Diels-Alder reaction of 22 took place spontaneously at room temperature upon benzoylation of the secondary amine 21 and produced a single isomer derived from the endo transition state. 1,3-Allylic strain is discussed as an important factor in control of the diastereo-selectivity.

Full text of DOI:10.1021/jo990459f

5930
J . Org. Chem. 1999, 64, 5930-5940
Diels-Ald er Rea ction s of Am in o Acid -Der ived Tr ien es
William V. Murray,* Sengen Sun, Ignatius J . Turchi, Frank K. Brown, and
A. Diane Gauthier
The R. W. J ohnson Pharmaceutical Research Institute, 1000 Route 202, Raritan, New J ersey 08869
Received March 12, 1999
Triene precursors (1a -e, 2a -k ) were constructed for substrate-controlled asymmetric Diels-Alder
reactions. Boc-L-phenylalanal and Boc-L-valinal were condensed with triethyl phosphonoacetate
or 2-phosphonopropionate to generate the R,â-unsaturated esters as dienophiles. Removal of the
Boc group to give free amines 4a -d , which after, or without N-benzylation, were treated with
3,5-hexadienoyl chloride to give 1a -e, or with 2,4-hexadienoyl chloride to afford 2a -f. The trienes
2g-i were prepared via reductive alkylation of amines 4a -i with 2,4-hexadienal. The secondary
amide triene 1a failed to yield any Diels-Alder product when heated at 170 °C. The tertiary amide
trienes 1b-e produced in refluxing toluene the major cycloaddition products that were cis-fused
and derived from the exo transition states. Trienes 2a -k underwent surprisingly facile Diels-
Alder reactions to produce the major trans-fused isomers that were derived from the endo transition
states. For trienes 2b-h and 2j,k , Diels-Alder reactions proceeded at room temperature. For the
primary amide 2a , the Diels-Alder reaction proceeded smoothly in refluxing toluene. The tertiary
amide triene 22 was constructed to have two electron-withdrawing ester substituents at the termini
of the triene. The Diels-Alder reaction of 22 took place spontaneously at room temperature upon
benzoylation of the secondary amine 21 and produced a single isomer derived from the endo
transition state. 1,3-Allylic strain is discussed as an important factor in control of the diastereo-
selectivity.
In tr od u ction
The Diels-Alder reaction is a powerful method for
synthesis of many categories of cyclic compounds and
possesses considerable potential in synthesis of hetero-
cyclic libraries for drug screening. There have several
reports on trienes containing a nitrogen atom within the
triene tether.1 Our goal was to develop an efficient
method, based on the intramolecular Diels-Alder reac-
tion of novel amino acid derived trienes, for fast, mild,
and selective generation of large number of heterocyclic
compounds.
F igu r e 1.
The trienes (1a -e, 2a -k , Figures 1 and 2) have been
designed so that a stereogenic center at the allylic
position of the dienophile can provide steric discrimina-
tion of the diastereotopic faces of the dienophile.2 The
terminal ethyl ester group may facilitate the cycloaddi-
tion process compared with similar published amino acid
derived trienes.1 We report here our surprising observa-
tions of the facile intramolecular cycloadditions and our
investigation of substituent effects on the reactivity and
stereoselectivity.
(1) For the syntheses of other hydroisoquinone and hydroisoindole
derivatives via Diels-Alder reactions, see: (a) Gutierrez, A. J .; Shea,
K. J .; Svoboda, J . J . J . Org. Chem. 1989, 54, 4335. (b) Moriwake, T.;
Hamano, S.; Saito, S.; Torii, S. J . Org. Chem. 1989, 54, 4114. (c) Martin,
S.; Williamson, S. A.; Gist, R. P.; Smith, K. M. J . Org. Chem. 1983,
48, 5170. (d) Guy, A.; Lemaire, M.; Negre, M.; Guette, J . P. Tetrahedron
Lett. 1985, 26, 3575. (e) Takeuchi, H.; Fujimoto, T.; Hoshino, K.;
Motoyoshiya, J .; Kakehi, A.; Yamamoto, I. J . Org. Chem. 1998, 63,
7172.
(2) For examples of the discussion on 1,3-allylic strains in Diels-
Alder reactions, see: (a) Crisp, G. T.; Gebauer, M. G. J . Org. Chem.
1996, 61, 8425. (b) Roush, W. R.; Koyama, K.; Curtin, M. L.; Moriarty,
K. J . J . Am. Chem. Soc. 1996, 118, 7502. (c) Reetz, M. T.; Kayser, F.;
Harms, K. Tetrahedron Lett. 1992, 33, 3453.
F igu r e 2.
Resu lts a n d Discu ssion
Synthesis of the triene precursors shown in Scheme 1
involves Horner-Wadsworth-Emmons condensation of
Boc-L-valinal3 and Boc-L-phenylalanal3 with triethyl
phosphonoacetate or phosphonopropionate in the pres-
10.1021/jo990459f CCC: $18.00 © 1999 American Chemical Society
Published on Web 07/21/1999

Diels-Alder Reactions of Amino Acid-Derived Trienes
J . Org. Chem., Vol. 64, No. 16, 1999 5931
Ta ble 1. P r od u ct Ra tios Wer e Obta in ed by 1H NMR
In tegr a tion of th e Cr u d e Rea ction Mixtu r es, or Ba sed on
Isola ted Yield s of th e Isom er s. Tr ien es 2b-g a n d 2j Wer e
Not Ch a r a cter ized Du e to th e On goin g Diels-Ald er
Rea ction s, a n d th e Yield s for Th ese Rea ction s Ar e for
Tw o Step s
Sch em e 1a
reaction
time
yield
%
5:6:7:8 or
9:10:11:12
solvents/temperature
1a
1b
1c
1d
1e
2a
2b
2c
2d
2e
2f
170 °C
24 h
16 h
50 h
16 h
6 days
7 h
12 h
20 h
60 h
80 h
72 h
<6 h
15 days
18 h
0
toluene, reflux
toluene, reflux
toluene, reflux
toluene, reflux
toluene, reflux
CHCl3, rt
CHCl3, rt
CHCl3, rt
CHCl3, rt
CHCl3, rt
94
85
74
66
93
82
73
91
80
-
33:10:11:3
3:1:0:0
3:1:0:0
5:2:0:0
34:11:1:0
10:1:10:1
83:25:1:1
20:2:1:0
185:50:1:0
6:1:0:0
5:1:0:0
3:2:0:0
9:1:0:0
6:1:-:-
2g
2h
2j
CH2Cl2/TFA, rt
CHCl3, rt
CHCl3, rt
>66
84
61
67
a
Key: (a) (EtO)2P(O)dCR2COOEt, THF; (b) 20% TFA in
2k
CHCl3, rt
18 days
CH2Cl2; (c) C6H5CHO or 2,4-(MeO)2C6H3CHO, AcOH, NaCNBH3;
(d) (E,E)-2,4- or (E)-3,5-hexadienoyl chloride, 1.5 equiv, Et3N, THF;
(e) (E,E)-2,4-hexadienal, AcOH, NaBH(OAc)3; (f) BzCl, Et3N,
THF.
of reactivity of 1a . The lack of reactivity of the triene 1a
may be rationalized by the following computational
results. The energy of an extended, low energy conformer
of 1 (R1 ) Me, R2 ) R3 ) H) is calculated to be 4.6 kcal/
mol (B3LYP/6-31G(d)//RHF/3-21G) lower than a folded
(i.e., reactive) conformer. This folded, reactive conformer
is a local minimum that would lead to the product 5 (R1
) Me, R2 ) R3 ) H). Because of this large energy
difference between these conformers, compound 1 cannot
reach a large enough concentration of the reactive
conformer at equilibrium, and consequently, no Diels-
Alder reaction is observed. In contrast, introduction of
an NMe substituent as in 1 (R1 ) Me, R2 ) H, R3 ) Me)
leads to a difference of only 2.5 kcal/mol at the same level
of theory between the extended and the folded, reactive
conformers. In this case, the concentration of the reactive
conformer is high enough at equilibrium for reaction to
occur.5
For compounds 1c,d , only two diastereomers were
present in the crude reaction mixture as seen by 1H NMR
and were isolated by silica gel column chromatography
(Scheme 2, Table 1). The calculated relative activation
energies for the reaction of 1 leading to products 5-8 at
the B3LYP/6-31G(d)//RHF/3-21G level of theory ad-
equately rationalize the observed product distributions
assuming similar entropies of activation. In both cases
the exo/equatorial transition structure is calculated to
be the lowest energy with the endo/equatorial transition
structure the next lowest as observed experimentally.
Sch em e 2a
a
E ) COOEt.
ence of a base. Removal of the Boc protecting group
afforded amine derivatives 4a -d , which were acylated
using 2,4- or 3,5-hexadienoyl chloride to give 1a and 2a .
4a -d were alternatively benzylated by reductive alkyl-
ation to give 4e-i before installation of the dienes by
acylation to generate 1b-e and 2b-f. Compounds 4a ,
4d (TFA salts), and 4i were treated with 2,4-hexadienal
and a boron hydride in the presence of acetic acid to
produce amine-linked trienes (2g-i). Triene 2i was
benzoylated to afford the amide triene 2j.
Triene 1a containing an amide NH group did not
produce any cycloaddition products after reflux in toluene
for 48 h and decomposed completely under N2 after reflux
for 24 h in 1,3-dichlorobenzene. Triene precursor 1b
containing an N-benzyl group was converted to Diels-
Alder products in excellent yield after reflux in toluene
for 16 h. All four possible isomers were observed in a ratio
of 33:11:10:3 as shown in Scheme 2 and Table 1. The
calculated activation energies for the Diels-Alder reac-
tions of the trienes 1 to the cycloadducts 5-8 are
presented in the Supporting Information.4 These values
are obtained by taking the difference in energy between
an extended, low energy conformer of 1 and the corre-
sponding transition structure leading to the products
5-8. This treatment is not sufficient to explain the lack
The bond lengths of the forming bonds for the transi-
tion structures leading to the products 5-8 are given.6
The formation of the new bonds is asynchronous with the
C1-C2 (lactam ring) bond being shorter than the C3-
C4 (cyclohexene ring) bond by 0.092 Å (AM1) and 0.267
Å (RHF/3-21G) on average.
It has been reported that cis-fused hydroisoquinolone
derivatives are generally the major isomers in cycload-
ditions of 7-aza-1,3,9-decatrienes.1 Martin et al. reported
a case that the isomeric cycloaddition products can be
interconverted at the reaction temperature.1c Our trienes
(5) Brown, F. K.; Singh, U. C.; Kollman, P. A.; Raimondi, L.; Houk,
K. N.; Bock, C. W. J . Org. Chem. 1992, 57, 4862; Raimondi, L.; Brown,
F. K.; Gonzalez, J .; Houk, K. N. J . Am. Chem. Soc. 1992, 114, 4796.
(6) Supporting Information Table 3.
(3) Fehrentz, J .-A.; Castro, B. Synthesis 1983, 676.
(4) Supporting Information Table 2.

5932 J . Org. Chem., Vol. 64, No. 16, 1999
Murray et al.
F igu r e 3.
F igu r e 4.
Sch em e 3a
refluxed in toluene, however, the reaction was complete
in 6 h, affording endo selectivity of ∼3:1 and facial
selectivity of 34:1. The facial selectivity across the di-
enophile in these cases appears to be due to the pro-
nounced interaction between the axial R1 group and the
R2 methyl group in transition state 11 (Figure 4).
When the TFA salt of 4a (R1 ) Bn, R2 ) H) was
subjected to reductive alkylation conditions with 2,4-
hexadienal, triene 2g was not detected. Instead, the
isomeric cycloaddition products 9g and 10g were isolated
in a 5:1 ratio. Under the same reaction conditions, pure
amine triene 2i was isolated in a 73% yield from the TFA
salt of 4d and underwent the cycloaddition very slowly
(10% conversion in 10 days in CDCl3). The Diels-Alder
reaction of 2j was complete in 18 h after the addition of
benzoyl chloride to a THF solution of 2i containing excess
triethylamine. Amine triene 2h was purified in a 54%
yield from a reaction of 4i. The cycloaddition of 2h was
complete in 15 days (Scheme 3, Table 1). It is apparent
in each of these cases that either the interaction between
R1 and R3 in the transition state or the planarization of
the amine nitrogen to its requisite amide has a profound
effect on the rate of the Diels-Alder cycloaddition.
Triene 2k was prepared by alkylation of benzylamine
with 4-bromo-trans-crotonic ethyl ester, followed by N-
acylation with 2,4-hexadienoyl chloride. Cycloaddition of
2k took 18 days to complete in either CDCl3 or C6D6. This
result indicated the interaction between R1 and R3 was
a key factor to the rate acceleration.
a
E ) COOEt.
1b-d also gave the major cis-fused products 5, but
derived from the exo transition state (Figure 3). The
purified isomers from the cyclizations of trienes 1c-e
were not able to be interconverted when reheated at the
reaction temperature, suggesting a kinetically controlled
mechanism. This observation is surprising as kinetically
controlled Diels-Alder reactions usually produce major
isomers derived from endo transition states.7 All of the
Diels-Alder reactions described herein are calculated to
be exothermic by 9-44 kcal/mol depending on the level
of theory.8 This is consistent with the experimental
observations.
This exo selectivity may be a result of the overall
conformation that can afford a reaction path of relatively
low energy. The sterically more hindered isopropyl group
R1, or the size increase of R2 from proton to methyl, may
further destabilize the transition states (7, 8) where the
R1 group is axial (Figure 3). Overall, the exo transition
state with the equatorial R1 (5) is favored.
Compound 2b could not be synthesized in pure form
by acylating 4e with 2,4-hexadienoyl chloride at room
temperature due to the facile Diels-Alder cycloaddition
that occurred at room temperature. We chose not to
isolate the intermediate 2b, and the crude product 2b in
ether was washed with sodium carbonate and water and
was then allowed to stand in CDCl3 or C6D6 at room
temperature. The cycloaddition of 2b was found to be
complete in 12 h (Scheme 3, Table 1). The Diels-Alder
cycloaddition reactions of 2c-f also proceeded at room
temperature and produced two prominent diastereomers
in each case. 1H NMR showed endo/exo ratios of from 3:1
to 6:1. The Diels-Alder reaction of 2a containing a free
NH group was sluggish at room temperature. When
The calculated activation energies for the Diels-Alder
reactions of the trienes 2 to the cycloadducts 9-12 are
given.9 These values are 3-6 kcal/mol lower for the
amides than for the amines with the exception of the
amides where the nitrogen substituted with a hydrogen
(i.e., 2: R1 ) R2 ) Me, R3 ) H and R1 ) iPr, R2 ) Me, R3
) H) at the B3LYP/6-31G(d)//RHF/3-21G level. Also, the
i
amide 2 (R1 ) Pr, R2 ) Me, R3 ) Ac) has an activation
energy for the cycloaddition to 9-12 of about 2 kcal/mol
lower than the amines at the same level of theory. The
calculated values agree at least qualitatively with the
observed relative rates for the cycloadditions of 2 to 9-12.
Frontier MO theory is inadequate for rationalizing the
relative rates of the Diels-Alder reactions of the amides
vs the amines since the energy gaps for the frontier
orbitals are nearly equal in each case. Thus, subtle steric
effects in the transition structures are probably respon-
sible for these rate effects.6
As was the case for the Diels-Alder reactions of 1, the
calculated relative activation energies for the reaction of
(7) Atkinson, R. S. Stereoselective Synthesis; J ohn Wiley & Sons:
Chichester, 1995; pp 156-180.
(8) Supporting Information Tables 2 and 4.
(9) Supporting Information Table 4.

Diels-Alder Reactions of Amino Acid-Derived Trienes
J . Org. Chem., Vol. 64, No. 16, 1999 5933
Sch em e 4a
Sch em e 5a
a
Key: (a) (E)-BrCH2CHCHCOOEt, DMAP, THF, 40 h, 52%;
(b) PhCOCl, Et3N, THF, 0.5 h; (c) rt, CDCl3 or C6D6, 40 h, 92%
(based on 21), one isomer only.
a
Key: (a) (E)-(EtO)2P(O)dCHCHdCHCOOEt, THF; (b) 20%
1
detected by H NMR and was complete in 40 h in either
TFA in CH2Cl2; (c) C6H5CHO, AcOH, NaCNBH3; (d) (E)-crotonyl
chloride, Et3N, THF; (e) toluene, reflux, 4 h, 100%, 17:18:19:20 )
10:10:1:1.
CDCl3 or C6D6. A single diastereomer 23 was observed
and purified in 92% yield as the cycloaddition product.
This system eliminates the complimentary polarization
that could be in effect in the other examples and
demonstrates the requirement for a large group at R3 as
well as a planar nitrogen.
2 leading to cycloadducts 9-12 at the B3LYP/6-31G(d)//
RHF/3-21G level of theory give a good account of the
observed product distributions. In general, the transition
structures leading to the endo trans-fused isomers are
calculated to be the lowest energy species. Lower levels
of theory usually give less satisfactory results.
Amide triene 16 containing an ester-substituted diene
was synthesized (Scheme 4). The Horner-Wadsworth-
Emmons reaction between valinal and triethyl trans-4-
phosphonocrotonate gave an inseparable mixture of the
major trans,trans-diene 13 and a minor cis,trans-diene
isomer (10:1). The mixture was converted to triene 16 in
a pure form. The cycloaddition of 16 was sluggish at room
temperature (∼10% conversion in 1 week) and was
carried out at reflux in toluene for 4 h to afford four
isomers 17, 18, 19, and 20 in a ratio of 10:10:1:1.
It is notable that we are adding an electron-deficient
diene and an electron-deficient dienophile for 2a -f to
form the 5,6-fused bicycloadducts. Our initial hypotheses
for these facile Diels-Alder reactions involved the comple-
mentary dipole interaction between the diene and the
dienophile in an intramolecular fasion. The similar
intermolecular reaction between 2,4-dienoylate and acry-
late was apparently disfavored for the regio orientation
of the favorable dipole interaction, relative to the regio
orientation of apparent dipole repulsion.11 We believed
that matching the presumed negative carbon R to the
carbonyl group in the diene to the positive carbon â to
the carbonyl in the dienophile, as well as the positive
carbon δ to the carbonyl in the diene with the negative
carbon R to the carbonyl in the dienophile, we could
achieve reasonable orbital overlap, and this could explain
the facile nature of the observed Diels-Alder reactions.
To test this hypothesis, the diester triene 21 was syn-
thesized (Scheme 5). This compound proved unreactive
at room temperature until it was N-benzoylated. Its
cycloaddition was initiated upon the benzoylation as
For the Diels-Alder reactions of 16 to 17-20 and 22
to 23 we were unable to locate transition structures at
the RHF/3-21G level of theory using standard or newer
techniques such as quadratic synchronous transit meth-
ods. Transition structures were located using the AM1
Hamiltonian and single-point calculations using the
hybrid HF-DFT method, B3LYP/6-31G(d) were carried
out on these two systems. The activation energy for the
conversion of 16 to 17-20 is about 10-13 kcal/mol higher
than for the cycloaddition of 22 to 23. This is in qualita-
tive agreement with the experimental findings that the
triene 22 undergoes cycloaddition much more rapidly
than triene 16. This result can be rationalized based on
FMO theory.6,10 For 16 the interacting orbitals are the
HOMO - 1 localized on the diene and the LUMO + 1
localized on the olefinic double bond. The energy gap
between these orbitals is 10.1 eV (AM1). For 22, however,
the interacting orbitals are the HOMO localized on the
diene and the LUMO + 1 localized on the olefin. In this
case the gap is only 9.1 eV (AM1). Other factors such as
steric and electrostatic interactions may contribute to the
relative reactivity of 16 vs 22, but these would be more
subtle and difficult to visualize.
The relative activation energies for the formation of
17-20 accurately reflect the observed product distribu-
tions. The transition structures leading to 17 and 18 are
calculated to be 3-4 kcal/mol lower in energy than those
leading to 19 and 20 at the B3LYP/6-31G(d)//AM1 level
of theory. This is not the case for the Diels-Alder reaction
of triene 22. Experimentally, only cycloadduct 23 is
observed while the calculations suggest that a mixture
of 23 and 24 should be obtained. This incorrect result
may be a consequence of using the AM1 geometries for
the transition structures leading to 23 and 24 for the
single-point B3LYP/6-31G(d) energy calculations.
(10) Fleming, I. Frontier Orbitals and Organic Chemical Reactions;
J ohn Wiley & Sons: London, 1976.
(11) Overman, L. E.; Taylor, G. F.; Houk, K. N.; Domelsmith, L. N.
J . Am. Chem. Soc. 1978, 100, 3182. (b) Yedidia, V.; Leznoff, C. C. Can.
J . Chem. 1980, 58, 1144.
In contrast to 1b-e, trienes 2a -k , and 22 all cyclized
to produce endo-derived trans-fused isomers as the
major products. The diastereomeric ratio from the

5934 J . Org. Chem., Vol. 64, No. 16, 1999
Murray et al.
cycloaddition of triene 2b (9b/10b/11b/12b ) 10:1:10:1)
reflects a good endo/exo selectivity, but no diastereofacial
selectivity. This result indicates that the A1,3 interaction
between benzyl at R1 and hydrogen at R2 in transition
state 11 (Figure 4) is not significant, despite the axial
position of R1. As the size of the R2 group increased to a
methyl in 2c and 2e-f, the pathway through the transi-
tion states 11 and 12 are largely suppressed and good
stereofacial selectivity is achieved. As expected, the
reaction of 2d provided a good diastereofacial selectivity
due to the larger isopropyl (R1) and dimethoxybenzyl (R3)
groups, as well as good endo/exo selectivity. It appears
that the endo selectivity among trienes 2f, 2h , and 2j is
rate dependent, as evidenced by the observed endo/exo
ratios and reaction times of 2f (18 h, 9:1), 2h (15 days,
3:2), and 2j (80 h, 3:1).
The linker length between a diene and a dienophile
for a Diels-Alder reaction is believed to affect the orbital
overlap.12 Trienes 2b-k are more reactive toward Diels-
Alder reactions to form 5,6-fused cycloaddition products
at room temperature, while trienes 1b-e require heating
to afford 6,6-fused hydroisoquinones. The enhanced
reactivity of compounds 2a -k relative to compounds
1a -d may be attributed to the ring-size difference. That
is, the triene system for the smaller 5,6-fused bicycload-
ducts was able to organize more effectively for the
cycloadditions than the system for 6,6-fused rings. It was
also shown that the reactivity in these systems was very
sensitive to variations of substituents R1, R2, and R3. The
relative reactivity in 2a -j decreases when the R1 group
is changed from benzyl to sterically more hindered
isopropyl. In addition, when we tested triene 2k , which
bears a hydrogen at R1, the cyclization proceeded very
slowly (18 days). It appears that an R1 group of moderate
size is in the best position to accelerate the cycloaddition.
The reaction was slowed when R2 was changed from
hydrogen to a methyl group, although the larger R2 group
can enhance the diastereofacial selectivity across the
dienophile.
Triene 2j cyclized faster than 2f, and 2f cyclized faster
than 2h , suggesting a trend that the tertiary amine
reacts more slowly than the tertiary amides. The diene
that is free from the conjugation with the amide carbonyl
group in triene 2j is in the best position to cyclize with
the dienophile. Trienes 22 cyclized much faster than 16.
This result relates to the corresponding intermolecular
cycloaddition between acrylate and 2,4-pentadienoate,
which produces the ortho product preferentially via an
endo transition state.10,11
It is interesting that trienes 2a -f,k , 16, and 22
containing both electron-deficient diene and electron-
deficient dienophile cyclized under surprisingly mild
conditions. The complementary electron demand is not
required between the two reacting partners. There had
been occasional examples of facile Diels-Alder reactions
in the literature that do not appear to have complemen-
tary electronic demand. For example, the dimerization
of 2-methoxycarbonyl-1,3-butadiene took place spontane-
ously at room temperature.13 These examples were
generally limited to theoretical and mechanistic argu-
ments. We are interested in the utilizization of these
reactions in the stereoselective syntheses of heterocyclic
compounds. The mild reaction conditions can facilitate
our further implementation of solid-phase technology in
the construction of molecular libraries for biological
evaluations.
The bond lengths of the forming bonds for the transi-
tion structures leading to the products 9-12, 17-20, and
23-26 are given.14 The formation of the new bonds is
even more asynchronous than in the 6,6-systems as
expected with the C1-C2 (lactam ring) bond being shorter
than the C3-C4 (cyclohexene ring) bond by 0.37 Å (AM1)
and 0.5 Å (RHF/3-21G) on average.
Con clu sion
We have observed a class of cycloadditions of amino
acid derived trienes in the synthesis of novel isohydroin-
dole derivatives and found that the major products were
derived from endo transition states. The corresponding
6,6 fused hydroisoquinoline derivatives were formed in
refluxing toluene through exo transition states. The A1,3
interaction between R1 and R2 appears to have an
important impact on the diastereomeric distribution of
the products. A sterically more hindered R1 or R2 group
affords stronger discrimination of the two diastereotopic
faces of the dienophile for its access to the diene during
cycloaddition.
These reported Diels-Alder reactions to form isohy-
droindoles lack complementary electronic demand and
can take place under mild thermal conditions. Further-
more, the stereoselectivity of the cyclizations between an
electron-deficient diene and an electron-deficient dieno-
phile can reach up to 100% to give a single stereoisomer.
Recently, Houk and co-workers published several
reports of the successful application of DFT, especially
the hybrid HF-DFT gradient corrected methods, to the
study of pericyclic reactions.15 Based on our experimental
results, we have shown that the B3LYP energies with a
moderate-sized basis set using the RHF/3-21G-optimized
geometries gives excellent estimates of the relative
reactivity and product distributions of the intramolecular
Diels-Alder reactions studied experimentally herein.
We are investigating these reactions to generate
complex functionalized heterocyclic libraries.
Exp er im en ta l Section
Com p u ta tion a l Deta ils. The programs GAUSSIAN-94,
GAUSSIAN-98, and SPARTAN 5.0 were used for the ab initio
and DFT calculations.16,17 The CAChe (v. 3.8) worksystem18
was used for the AM1 calculations. Optimized geometries were
obtained at the AM1 and RHF/3-21G levels of theory. All
stationary points (minima and transition structures) were
characterized by calculation of their harmonic vibrational
frequencies at the corresponding level (AM1 or RHF/3-21G).
(13) Spino, C.; Crawford, J .; Cui, Y.; Gugelchuk, M. J . Chem. Soc.,
Perkin Trans. 2 1998, 1499 and references therein.
(14) Supporting Information Table 5.
(15) Yoo, X.; Houk, K. N. J . Am. Chem. Soc. 1997, 119, 2877. Wiest,
O.; Houk, K. N. Top. Curr. Chem. 1996, 183 (Density Functional
Theory IV), 1.
(16) Gaussian 94 and 98: Frisch, M. J .; Trucks, G. W.; Schlegel, H.
B.; Gill, P. M. W.; J ohnson, B. G.; Robb, M. A.; Cheeseman, J . R.; Keith,
T.; Petersson, G. Al-Laham, M. A.; Zakrzewski, V. G.; Ortiz, J . V.;
Foresman, J . B.; Cioslowski, J .; Stefanov, B. B.; Nanayakkara, A.;
Challacombe, M.; Peng, C. Y.; Ayala, P. Y.; Chen, W.; Wong, M. W.;
Andres, J . L.; Replogle, E. S.; Gomperts, R.; Martin, R. L.; Fox, D. J .;
Binkley, J . S.; Defrees, D. J .; Baker, J .; Stewart, J . P.; Head-Gordon,
M.; Gonzalez, C.; Pople, J . A.; Gaussian, Inc., Pittsburgh, PA, 1995,
1998.
(12) Boeckman, R. K., J r.; Demko, D. M. J . Org. Chem. 1982, 47,
1789.
(17) SPARTAN (v. 5.0), Wavefunction, Inc., Irvine, CA.
(18) CAChe Scientific, Oxford Molecular, Beaverton, OR.

Diels-Alder Reactions of Amino Acid-Derived Trienes
J . Org. Chem., Vol. 64, No. 16, 1999 5935
All minima had no negative eigenvalues of the Hessian and
no imaginary frequencies. All first-order saddle points (transi-
tion structures) had one negative eigenvalue of the Hessian
and one imaginary frequency. Single-point energies were
calculated using the Becke three-parameter hybrid Hartree-
Fock-DFT method19 with a 6-31G(d) basis set. ZPE corrections
scaled by 0.8929 were applied to the RHF/3-21G calculated
energies. B3LYP/6-31G(d)//RHF3-21G and B3LYP/6-31G(d)//
AM1 energies are uncorrected.
Gen er a l P r oced u r e for P r ep a r a tion of 3a -d a n d 13.
To triethyl phosphonoacetate, 2-phosphonopropionate, or
4-phosphono-(E)-crotonate (13.4 mmol) in dry THF (50 mL)
was added potassium tert-butoxide (12 mL, 12 mmol, 1 M in
THF) at 0-5 °C. The mixture was stirred at room temperature
for 1 h. N-Boc-L-phenylalanal or N-Boc-L-valine (16.5 mmol)
in THF (20 mL) was added slowly at 0-5 °C. Some inorganic
salt precipitated as sticky material. The mixture was stirred
at room temperature for 2 h. Sodium bicarbonate (5%, 100 mL)
was added to quench the reaction. The mixture was extracted
with ether (2 × 100 mL). The combined organic phase was
dried over sodium sulfate and concentrated in vacuo. The
residue was purified by column chromatography. The column
was packed with silica gel using a mixture of benzene and
hexanes (1:1). The sample was dissolved in the same solvent
mixture for loading and was then eluted with 5-10% ethyl
acetate in hexanes.
69% yield from column chromatography: 1H NMR (CDCl3, 300
MHz) δ 7.27 (dd, 1H, J ) 10.9 Hz, 15.3 Hz), 6.27 (dd, 1H, J )
11.0 Hz, 15.1 Hz), 5.99 (dd, 1H, J ) 6.2 Hz, 15.3 Hz), 5.86 (d,
1H, J ) 15.4 Hz), 4.65 (d, broad, 1H, J ) 8.2 Hz), 4.19 (q, 2H,
J ) 7.2 Hz), 1.76-1.85 (m, 1H), 1.45 (s, 9H), 1.29 (t, 3H, J )
7.2 Hz), 0.92 (d, 3H, J ) 6.8 Hz), 0.90 (d, 3H, J ) 6.8 Hz); 13
C
NMR (CDCl3, 75 MHz) δ 166.8, 155.3, 143.7, 141.8, 128.3,
121.0, 60.1, 57.3, 32.3, 28.3, 28.2, 18.6, 18.0, 14.1.
Gen er a l P r oced u r e for Rem ova l of Boc fr om 3a -d .
Compounds 3a -d and 13 were treated with 20% trifluoroacetic
acid in CH2Cl2 for 20 min. The mixture was evaporated,
coevaporated twice with chloroform, and dried in vacuo to give
TFA salts 4a -d and 14, which were not purified and charac-
terized, and were used directly for benzylation, 2,4-dimethoxy-
benzylation, or acylation.
Gen er a l P r oced u r e for Red u ctive Ben zyla tion a n d 2,4-
Dim eth yoxyben zyla tion . These reactions were performed
by strictly following the procedure described in ref 20.
(4S)-4-N-Ben zylam in o-5-ph en yl-2-tr a n s-pen ten oic acid,
eth yl ester (4e): 1H NMR (CD3OD) δ 7.10-7.29 (m, 10H), 6.76
(dd, 1H, J ) 8.2 Hz, 15.8 Hz), 5.75 (d, 1H, J ) 15.7 Hz), 4.14
(q, 2H, J ) 7.1 Hz), 3.75 (d, 1H, J ) 13.3 Hz), 3.56 (d, 1H, J
) 13.3 Hz), 3.45 (q, 1H, J ) 7.4 Hz), 2.72-2.91 (m, 2H), 1.25
(t, 3H, J ) 7.2 Hz); HRMS calcd for C20H24NO2 (MH+)
310.1807, found 310.1797.
(4S )-4-N -B e n zy la m in o -5-p h e n y l-2-m e t h y l-2-t r a n s-
p eten oic a cid , eth yl ester (4f): 1H NMR (C6D6) δ 6.97-7.24
(m, 10H), 6.81 (dd, 1H, J ) 1.2 Hz, 9.5 Hz), 3.94-4.07 (m,
2H), 3.51-3.60 (m, 2H), 3.36 (d, 1H, 13.5 Hz), 2.66 (dd, 1H, J
) 6.9 Hz, 13.3 Hz), 2.52 (dd, 1H, J ) 6.8 Hz, 13.3 Hz), 1.60 (d,
3H, J ) 1.0 Hz), 0.99 (t, 3H, J ) 7.2 Hz); 13C NMR (CDCl3, 75
MHz) δ 167.8, 139.8, 137.6, 129.5, 129.3, 128.4, 128.3, 127.9,
127.3, 126.9, 126.4, 60.6, 56.6, 51.4, 41.3, 14.2, 12.5; HRMS
calcd for C21H26NO2 (MH+) 324.1964, found 324.1958.
(4S)-4-(N-2,4-Dim eth oxyben zyl)am in o-5-m eth yl-2-tr a n s-
h exen oic a cid , eth yl ester (4g): 1H NMR (C6D6, 300 MHz)
δ 7.16 (d, 1H, J ) 8.0 Hz), 7.05 (dd, 1H, J ) 7.9 Hz, 15.9 Hz),
6.43 (d, 1H, J ) 2.2 Hz), 6.35 (dd, 1H, J ) 2.3 Hz, 8.2 Hz),
6.16 (d, 1H, J ) 15.9 Hz), 4.07 (q, 2H, J ) 7.2 Hz), 3.91 (d,
1H, J ) 13.3 Hz), 3.63 (d, 1H, J ) 13.4 Hz), 3.39 (s, 3H), 3.28
(s, 3H), 2.87 (t, 1H, J ) 7.0 Hz), 1.49-1.58 (m, 1H), 1.01 (t,
3H, J ) 7.3 Hz); 13C NMR (C6D6, 75 MHz) δ 166.1, 160.5, 159.0,
150.1, 130.6, 122.8, 121.6, 103.9, 99.0, 64.6, 60.0, 54.9, 54.7,
46.9, 32.7, 19.0, 18.6, 14.3; HRMS calcd for C18H28NO4 (MH+)
322.2018, found 322.2031.
(4S)-4-N-Ben zyla m in o-2,5-d im et h yl-2-tr a n s-h exa n oic
a cid , eth yl ester (4h ): 1H NMR (C6D6, 300 MHz) δ 7.08-
7.32 (m, 5H, 6.69 (dd, 1H, J ) 1.3 Hz, 10.1 Hz), 4.06 (q, 2H, J
) 7.1 Hz), 3.65 (d, 1H, J ) 13.4 Hz), 3.42 (d, 1H, J ) 13.5 Hz),
3.02 (dd, 1H, J ) 6.6 Hz, 10.2 Hz), 1.78 (d, 3H, J ) 1.3 Hz),
1.47-1.58 (m, 1H), 1.04 (t, 3H, J ) 7.1 Hz), 0.87 (d, 3H, J )
6.7 Hz), 0.79 (d, 3H, J ) 6.8 Hz); 13C NMR (C6D6, 75 MHz) δ
167.6, 143.6, 141.1, 129.9, 128.4, 127.0, 126.9, 60.8, 60.5, 51.5,
33.3, 19.3, 18.6, 14.3, 13.3; HRMS calcd for C17H26NO2 (MH+)
276.1964, found 276.1973.
(4S)-4-N-ter t-Bu toxyca r bon yla m in o-5-p h en yl-2-tr a n s-
p en ten oic Acid , Eth yl Ester (3a ). An oil that solidified
slowly was obtained in 75% yield: 1H NMR (CDCl3, 300 MHz)
δ 7.23-7.36 (m, 5H), 6.86-6.94 (m, 1H), 5.84 (d, 1H, J ) 17.3
Hz), 4.61 (m, 2H), 4.20 (q, 2H, J ) 7.2 Hz), 2.82-2.93 (m, 2H),
1.38 (s, 9H), 1.28 (t, 3H, J ) 7.2 Hz); 13C NMR (CDCl3, 75
MHz) δ 166.2, 155.0, 147.6, 136.5, 129.4, 128.6, 126.9, 121.1,
79.9, 60.4, 52.3, 40.9, 28.3, 14.2. Anal. Calcd for C18H25NO4:
C, 67.71; H, 7.84; N, 4.39. Found: C, 68.03; H, 7.86; N, 4.27.
4-S-4-N-ter t-Bu toxyca r bon yla m in o-5-p h en yl-2-m eth yl-
2-tr a n s-p en ten oic Acid , Eth yl Ester (3b). An oil that
solidified slowly was obtained in 26% yield. A side product,
from cyclization of the cis isomer by attack of Boc-protected
amino group on the ethyl ester to kick out the ethoxide group,
was obtained in 27% yield using 20% ethyl acetate in hexanes
as the eluent. For 3b: 1H NMR (CDCl3, 300 MHz) δ 7.16-
7.31 (m, 5H), 6.51-6.54 (m, 1H), 4.65 (m, 2H), 4.17 (q, 2H, J
) 7.0 Hz), 2.74-2.94 (m, 2H), 1.69 (d, 3H, J ) 1.2 Hz), 1.40
(s, 9H), 1.28 (t, 3H, J ) 7.0 Hz); 13C NMR (CDCl3, 75 MHz) δ
167.6, 154.8, 140.1, 136.6, 129.4, 128.3, 128.2, 126.5, 79.5, 60.5,
50.2, 41.1, 28.2, 14.1, 12.5; HRMS calcd for C19H28NO4 (MH+)
334.2018, found 334.2021. Anal. Calcd for C19H27NO4: C,
68.47; H, 8.11; N, 4.20; found: C, 68.44; H, 8.02; N, 4.11.
4-S-4-N-ter t-Bu t oxyca r b on yla m in o-5-m et h yl-2-tr a n s-
h exen oic Acid , Eth yl Ester (3c). An oil was obtained in 91%
yield: 1H NMR (CDCl3, 300 MHz) δ 6.51 (dd, 1H, J ) 5.6 Hz,
15.6 Hz), 5.57 (dd, 1H, J ) 1.5 Hz, 15.6 Hz, 4.80 (d, 1H, J )
9.0 Hz), 3.81 (q, 2H, J ) 7.1 Hz), 1.45-1.53 (m, 1H), 1.06 (s,
9H), 5.25 (t, 3H, J ) 7.1 Hz), 0.56 (d, 3H, J ) 6.7 Hz), 0.55 (d,
3H, J ) 6.8 Hz); 13C NMR (CDCl3, 75 MHz) δ 165.9, 155.2,
147.2, 121.1, 78.9, 60.0, 56.5, 32.0, 31.7, 28.0, 18.6, 17.8, 13.9;
HRMS calcd for C14H25NO4 (MH+) 272.1862, found 272.1871.
4-S-4-ter t-Bu toxyca r bon yla m in o-2,5-d im eth yl-2-tr a n s-
h exen oic Acid , Eth yl Ester (3d ). An oil was obtained in 56%
yield. A side product from cyclization of the cis isomer by attack
of Boc-protected amino group on the ethyl ester to kick out
the ethoxide group was obtained in 37% yield using 15% ethyl
acetate in hexanes as the eluent. For 3d : 1H NMR (CDCl3,
300 MHz) δ 6.51 (dd, 1H, J ) 1.1 Hz, 9.2 Hz), 4.58 (s, broad,
1H), 4.20 (q, 2H, J ) 7.2 Hz), 1.93 (d, 3H, J ) 1.0 Hz), 1.71-
1.81 (m, 1H), 1.43 (s, 9H), 0.95 (d, 3H, J ) 6.8 Hz), 0.91 (d,
3H, J ) 6.7 Hz); 13C NMR (CDCl3, 75 MHz) δ 168.0, 155.3,
139.9,129.6, 79.3, 60.6, 53.9, 32.9, 28.4, 18.5, 18.4, 14.2, 13.0;
HRMS calcd for C15H28NO4 (MH+) 286.2018, found 286.2014.
(6S)-6-N-ter t-Bu t oxyca r b on yla m in o-7-m et h yl-tr a n s,-
tr a n s-2,4-octa d ien oic Acid , Eth yl Ester (13). Obtained in
(4S)-4-(N-2,4-Dim eth oxyben zyl)a m in o-2,5-d im eth yl-2-
tr a n s-h exen oic a cid , eth yl ester (4i): 1H NMR (CDCl3, 300
MHz) δ 7.05 (d, 1H, J ) 8.1 Hz), 6.61 (dd, 1H, J ) 1.3 Hz, 9.9
Hz), 6.38-6.48 (m, 2H), 4.20 (q, 2H, J ) 7.1 Hz), 3.80 (s, 3H),
3.79 (s, 3H), 3.75 (d, 1H, J ) 13.6 Hz), 3.48 (d, 1H, J ) 13.4
Hz), 3.13 (dd, 1H, J ) 6.4 Hz, 9.9 Hz), 1.79 (d, 3H, J ) 1.2
Hz), 1.67-1.79 (m, 1H), 1.31 (t, 3H, J ) 7.1 Hz), 0.91 (d, 3H,
J ) 6.7 Hz), 0.86 (d, 3H, J ) 6.8 Hz); 13C NMR (CDCl3, 75
MHz) δ 167.9, 159.9, 158.5, 143.8, 130.3, 129.2, 120.7, 103.4,
98.3, 60.5, 60.3, 55.1, 55.0, 47.0, 32.7, 19.1, 18.4, 14.1, 13.1;
HRMS calcd for C19H30NO4 (MH+) 336.2175, found 336.2183.
(6S)-6-N-Ben zyla m in o-7-m et h yl-tr a n s,tr a n s-2,4-oct a -
1
d ien oic a cid , eth yl ester (15): H NMR (CDCl3, 300 MHz)
δ 7.21-7.37 (m, 6H), 6.24 (dd, 1H, J ) 11.0 Hz, 15.2 Hz), 5.94
(dd, 1H, J ) 8.4 Hz, 15.3 Hz), 5.86 (d, 1H, J ) 15.3 Hz), 4.21
(q, 1H, J ) 7.1 Hz), 3.80 (d, 1H, J ) 13.3 Hz), 3.59 (d, 1H, J
(20) Szardemings, A. K.; Burkoth, T. S.; Look, G. C.; Campbell, D.
A. J . Org. Chem. 1996, 61, 6720.
(19) Becke, A. D. J . Chem. Phys. 1993, 98, 5648.

5936 J . Org. Chem., Vol. 64, No. 16, 1999
Murray et al.
) 13.3 Hz), 2.90 (dd, 1H, J ) 6.0 Hz, 8.3 Hz), 1.68-1.79 (m,
1H), 1.30 (t, 3H, J ) 7.2 Hz), 0.93 (d, 3H, J ) 6.8 Hz), 0.88 (d,
3H, J ) 6.8 Hz); 13C NMR (CDCl3, 75 MHz) δ 167.0, 144.5,
144.2, 140.5, 129.9, 128.3, 128.0, 126.8, 120.5, 65.5, 60.2, 51.4,
32.6, 19.3, 18.4, 14.3; LC-MS calcd for C16H28NO4 (MH+) 298,
found 298.
Gen er a l P r oced u r e for Acyla tion w ith Hexa d ien oyl
Ch lor id e. An amine or benzylated amine compound (4a -i, 1
mmol) in dry THF (20 mL) was treated with triethylamine
(0.3 mL) and hexadienoyl chloride (0.2 mL, ∼1.6 equiv) at 0-5
°C. The mixture was stirred at room temperature for 20 min,
quenched with 10% sodium carbonate (50 mL), and stirred for
another 0.5 h, followed by extraction with ether (100 mL). The
ether layer was washed again with 10% sodium carbonate (40
mL), brine (50 mL), and water (2 × 50 mL), dried over sodium
sulfate, and evaporated. The residue was dried under high
vacuum for 0.5 h before column purification for 1a -e and 2a .
For 2b-f, the crude residue was not purified and was directly
dissolved in chloroform for the Diels-Alder reaction to com-
plete.
(4S)-4-[N-Ben zyl-N-(3,5-h exa d ien oyl)a m in o]-5-p h en yl-
2-tr a n s-p en ten oic a cid , eth yl ester (1b): 1H NMR (CDCl3,
300 MHz) δ 6.95-7.54 (m, 11H), 5.67-6.37 (m, 4H), 4.98-
5.17 (m, 3H), 4.43 (d, 1H, J ) 17.3 Hz), 4.16 (d, 1H, J ) 17.0
Hz), 4.13 (t, 2H, J ) 7.2 Hz), 2.88-3.21 (m, 4H), 1.24 (t, 3H,
J ) 7.1 Hz); 13C NMR (CDCl3, 75 MHz) δ 171.7, 165.8, 145.2,
137.1, 136.6, 136.3, 136.2, 134.4, 133.8, 129.1, 128.7, 128.4,
127.7, 127.5, 127.0, 126.6, 126.4, 125.3, 122.7, 116.9, 116.5,
60.3, 58.6, 50.2, 38.1, 37.7, 14.0; HRMS calcd for C26H30NO3
(MH+) 404.2226, found 404.2242.
(4S)-4-[N-Ben zyl-N-(3,5-h exa d ien oyl)a m in o]-5-p h en yl-
2-m eth yl-2-tr a n s-p en ten oic a cid , eth yl ester (1c): 1H
NMR (CDCl3, 300 MHz) δ 7.05-7.33 (m, 10H), 6.79 (d, 1H, J
) 9.5 Hz), 6.30 (dt, 1H, J ) 10.2 Hz, 10.2 Hz, 17.0 Hz), 5.96
(dd, 1H, J ) 10.4 Hz, 15.3 Hz), 5.67-5.80 (m, 1H), 5.34 (q,
1H, J ) 8.4 Hz), 5.00-5.5.16 (m, 2H), 4.49 (d, 1H, J ) 17.6
Hz), 4.35 (d, 1H, J ) 17.6 Hz), 4.10 (q, 2H, J ) 7.1 Hz), 3.03-
3.15 (m, 3H), 2.81 (dd, 1H, J ) 8.4 Hz, J ) 13.4 Hz), 1.59 (s,
3H), 1.20 (t, 1H, J ) 7.2 Hz); 13C NMR (CDCl3, 75 MHz) δ
171.5, 167.4, 137.4, 137.2, 136.4, 133.6, 130.9, 129.2, 128.7,
128.3, 127.4, 126.9, 126.5, 126.1, 116.4, 60.5, 55.2, 49.3, 39.0,
38.2, 14.0, 12.6; HRMS calcd for C27H32NO3 (MH+) 418.2382,
found 418.2388.
5.4 Hz), 1.73 (s, 3H), 1.26 (t, 3H, J ) 7.2 Hz); 13C NMR (CDCl3,
75 MHz) δ 167.6, 165.6, 141.4, 139.4, 137.8, 136.6, 129.9, 129.5,
129.3, 128.3, 126.6, 121.1, 60.6, 48.7, 40.6, 18.4, 14.0, 12.6;
HRMS calcd for C20H25NO3 (MH+) 328.1913, found 328.1907.
(4S)-4-(N-Ben zyl-N-h exa-2′,4′-dien yl)am in o-2,5-dim eth yl-
2-tr a n s-h exa n oic Acid , Eth yl Ester (2h ). To a solution of
4h (154 mg, 0.56 mmol) in methylene chloride (10 mL) was
added hexa-2,4-dienal (0.120 mL, 1.10 mmol), acetic acid (0.2
mL), and sodium triacetoxyborohydride (300 mg, 1.42 mmol)
in sequence under nitrogen. The mixture was stirred over-
night, quenched with 10% aqueous sodium bicarbonate, and
separated between 10% aqueous sodium bicarbonate and
methylene chloride. The organic phase was dried over sodium
sulfate and concentrated. The residue was purified by column
chromatography using a mixture of hexanes and benzene (1:
1) to give 2h (108 mg, 54%). Further elution with benzene gave
the cycloaddition products 9h (19 mg) and 10h (9 mg).
Compound 4h (21 mg, 14%) was recovered. For 2h : 1H NMR
(CDCl3, 300 MHz) δ 7.7.19-7.36 (m, 5H), 6.72 (dd, 1H, J )
11.0 Hz, 1.3 Hz), 5.97-6.16 (m, 2H), 5.50-5.69 (m, 2H), 4.23
(q, 2H, J ) 7.1 Hz), 3.91 (d, 1H, J ) 14.2 Hz), 3.19-3.39 (m,
2H), 3.00 (t, 1H, J ) 10.4 Hz), 2.83 (dd, 1H, J ) 14.4 Hz, 8.5
Hz), 1.80-1.88 (m, 1H), 1.74 (d, 3H, J ) 4.4 Hz), 1.73 (d, 3H,
J ) 1.1 Hz), 1.34 (t, 3H, J ) 7.3 Hz), 1.08 (d, 3H, J ) 6.5 Hz),
0.73 (d, 3H, J ) 6.6 Hz); 13C NMR (CDCl3, 75 MHz) δ 167.9,
140.4, 139.9, 132.2, 131.1, 130.7, 129.7, 128.3, 128.1, 128.0,
126.5, 63.0, 60.5, 53.6, 51.4, 29.7, 20.1, 19.9, 17.9, 14.2, 13.3.
(4S)-4-N-Hexa-2′,4′-dien yl-2,5-dim eth yl-2-tr a n s-h exan o-
ic Acid , Eth yl Ester (2i). Compound 3d (600 mg, 2.11 mmol)
was treated with 20% TFA in methylene chloride (30 mL) for
20 min. The mixture was evaporated to a residue, pumped
overnight, and treated by following the procedure to make 2h .
Compound 2i was purified by elution with 5-8% ethyl acetate
in hexane from column chromatography in 410 mg, 73%: 1H
NMR (CDCl3, 300 MHz) δ 6.52 (dd, 1H, J ) 10.2 Hz, 1.2 Hz),
5.99-6.13 (m, 2H), 5.53-5.67 (m, 2H), 4.20 (q, 2H, J ) 7.1
Hz), 3.18-3.32 (m, 2H), 3.04 (dd, J ) 13.8 Hz, 7.1 Hz), 1.85
(d, 3H, J ) 1.2 Hz), 1.72-1.80 (m, 1H), 1.74 (d, 3H, J ) 6.7
Hz), 1.31 (t, 3H, J ) 7.2 Hz), 0.95 (d, 3H, J ) 6.7 Hz), 0.88 (d,
3H, J ) 6.9 Hz); 13C NMR (CDCl3, 75 MHz) δ 167.9, 143.0,
131.8, 131.0, 129.5, 129.3, 128.6, 60.7, 60.5, 49.2, 32.9, 19.3,
18.3, 17.9. 14.2, 13.1.
(6S)-6-(N-Ben zyl-N-tr a n s-cr ot on yla m in o)-7-m et h yl-
tr a n s,tr a n s-2,4-octa d ien oic Acid , Eth yl Ester (16). Com-
pound 15 in dry THF was treated with triethylamine and
trans-crotonyl chloride by following the general procedure for
acylation with hexadienoyl chloride. The product was purified
by column chromatography and eluted with 5-10% ethyl
acetate in hexanes to give 16 in 84% yield: 1H NMR (CDCl3,
300 MHz) δ 6.91-7.32 (m, ∼7H), 5.89-6.22 (m, 3H), 5.78 (d,
1H, J ) 15.4 Hz), 4.46-4.58 (m, 3H), 4.17 (q, 2H, J ) 7.1 Hz),
2.05-2.17 (m, 1H), 1.79 (d, 3H, J ) 6.6 Hz), 1.27 (t, 3H, J )
(4S)-4-[N-(2,4-Dim eth oxyben zyl)-N-(3,5-h exa d ien oyl)-
a m in o]-5-m eth yl-2-tr a n s-h exen oic a cid , eth yl ester (1d ):
1H NMR (CDCl3, 300 MHz) δ 6.92 (d, 1H, J ) 8.0 Hz), 6.80
(dd, 1H, J ) 9.2 Hz, 15.8 Hz), 6.27-6.43 (m, 3H), 5.81-6.09
(m, 2H), 5.72 (d, 1H, J ) 15.7 Hz), 4.90-5.18 (m, 2H), 4.45 (d,
1H, J ) 17.1 Hz), 4.35 (d, 1H, J ) 17.2 Hz), 4.11 (q, 2H, J )
7.2 Hz), 3.81 (s, 3H), 3.79 (s, 3H), 3.12-3.27 (m, 3H), 2.17-
2.29 (m, 1H), 1.22 (t, 3H, J ) 7.1 Hz), 0.86 (d, 3H, J ) 6.8
Hz), 0.83 (d, 3H, J ) 6.8 Hz); 13C NMR (CDCl3, 75 MHz) δ
171.7, 165.9, 160.4, 157.7, 144.8, 136.5, 133.5, 128.8, 127.4,
123.7, 117.0, 116.1, 103.7, 98.3, 63.5, 60.0, 55.2, 55.0, 45.3, 37.9,
28.9, 20.1, 19.3, 14.0; HRMS calcd for C24H34NO5 (MH+)
416.2437, found 416.2453.
6.5 Hz), 0.91 (d, 3H, J ) 6.7 Hz), 0.97 (d, 3H, J ) 6.5 Hz); 13
C
NMR (CDCl3, 75 MHz) δ 167.3, 166.8, 143.7, 142.7, 140.1,
137.7, 131.5, 128.6, 127.4, 126.6, 122.5, 121.6, 64.6, 60.3, 49.4,
29.7, 20.2, 19.6, 18.2, 14.2; HRMS calcd for C22H30NO3 (MH+)
356.2226, found 356.2217.
(4S )-4-[N -Be n zyl-N -(3,5-h e xa d ie n oyl)a m in o]-2,5-d i-
1
m eth yl-2-tr a n s-h exen oic a cid , eth yl ester (1e): H NMR
Diels-Ald er Rea ction To Ma k e Com p ou n d s 5b, 6b, 7b,
a n d 8b. Triene 1b (92 mg, 0.23 mmol) in toluene (10 mL) was
heated under nitrogen at reflux for 16 h. Toluene was removed
in vacuo. The residue was purified by slow column chroma-
tography without air pressure: 8% ethyl acetate in hexanes
to elute the fast isomer (14 mg, 6b, pure); 10% ethyl acetate
in hexanes to elute a second fraction (5 mg, 8b, not pure,
containing other isomers); 15-20% ethyl acetate in hexanes
to elute a third fraction (66 mg as a mixture of 5b and 7b
with no separation, 1H NMR showed a ratio of 3:1). The overall
estimated ratio is 5b:6b:7b:8b ) 33:10:11:3, and the total yield
is 94%. The fraction for the mixture of 5b and 7b was purified
in a second round column chromatography using 12% ethyl
acetate in hexanes to elute slowly without air pressure. The
very initial fractions that were positive in an iodine chamber
were identified as 5b (6 mg). The last half portion of the eluent
was combined to give 16 mg (5b:7b ∼ 1:1), which was purified
(CDCl3, 300 MHz) δ 7.09-7.32 (m, 5H), 6.52 (d, 1H, J ) 10,3
Hz), 6.31 (dt, 1H, J ) 10.1 Hz, 10.1 Hz, 16.9 Hz), 5.78-6.03
(m, 2H), 4.96-5.17 (m, 3H), 4.60 (d, 1H, J ) 17.7 Hz), 4.46 (d,
1H, J ) 17.7 Hz), 4.00-4.12 (m, 2H), 3.14 (dd, 1H, J ) 7.1
Hz, 16.2 Hz), 3.03 (dd, 1H, J ) 6.3 Hz, 16.2 Hz), 1.95-2.05
(m, 1H), 1.89 (d, 3H, J ) 1.0 Hz), 1.15 (t, 3H, J ) 7.2 Hz),
0.94 (d, 3H, J ) 6.6 Hz), 0.86 (d, 3H, J ) 6.6 Hz); 13C NMR
(CDCl3, 75 MHz) δ 171.7, 167.4, 138.1, 137.4, 136.4, 133.6,
131.4, 128.5, 127.2, 127.1, 126.0, 116.3, 60.3, 58.0, 48.1, 38.0,
30.4, 19.4, 19.1, 14.0, 13.3; HRMS calcd for C23H32NO3 (MH+)
370.2382, found 370.2369.
(4S)-4-(N-2,4-Hexa d ien oyla m in o)-5-p h en yl-2-m eth yl-2-
tr a n s-p en ten oic a cid , eth yl ester (2a ): 1H NMR (CDCl3,
300 MHz) δ 7.08-7.29 (m, 6H), 6.57 (dd, 1H, J ) 1.3 Hz, 9.6
Hz), 6.00-6.20 (m, 2H), 5.73 (d, 1H, J ) 15.1 Hz), 5.04-5.14
(m, 2H), 4.15 (q, 2H, J ) 7.0 Hz), 2.99 (dd, 1H, J ) 6.0 Hz,
13.4 Hz), 2.83 (dd, 1H, J ) 7.5 Hz, 13.5 Hz), 1.81 (d, 3H, J )

Diels-Alder Reactions of Amino Acid-Derived Trienes
J . Org. Chem., Vol. 64, No. 16, 1999 5937
in a third round of column chromatography similarly to give
7b (2 mg) by collecting the very last small fraction.
173.5, 137.6, 136.5, 131.0, 129.0, 129.0, 128.9, 128.3, 128.0,
127.5, 125.8, 61.4, 55.0, 47.2, 45.2, 44.5, 38.8, 38.2, 37.4, 31.7,
16.2, 14.5; obsd NOE (CD3OD, 400 MHz) between H1 and Me8,
and between H4a and Me8; HRMS calcd for C27H32NO3 (MH+)
418.2382, found 418.2382.
For 5b: 1H NMR (CDCl3, 300 MHz) δ 7.11-7.33 (m, 10H),
5.66-5.74 (m, 1H), 5.52-5.61 (m, 1H), 3.29 (d, 1H, J ) 14.6
Hz), 3.79-3.90 (m, 2H), 3.59-3.70 (m, 2H), 3.02 (dd, 1H, J )
4.9 Hz, 13.9 Hz), 2.63-2.83 (m, 3H), 2.44 (dt, 1H, J ) 6.1 Hz,
10.3 Hz, 10.3 Hz, for H8), 2.26 (dd, 1H, J ) 9.1 Hz, 18.0 Hz),
2.09-2.17 (m, 2H), 1.96-2.08 (m, 1H); 13C NMR (CDCl3, 75
MHz) δ 174.1, 168.6, 137.6, 137.3, 129.6, 129.1, 128.9, 128.7,
128.6, 127.5, 126.8, 124.6, 60.4, 58.9, 48.6, 40.0, 38.6, 35.6, 34.9,
28.6, 27.8, 13.9; obsd NOE (CD3OD, 270 K, 400 MHz) between
H4a and H8a (may be between H7 and H8a due to overlapping
of H4a and H7), no NOE between H1 and H8a, H4a and H8,
H1 and H4a, H8 and H8a; HRMS calcd for C26H30NO3 (MH+)
404.2226, found 404.2242.
Diels-Ald er Rea ction To Ma k e Com p ou n d s 5d a n d 6d .
Triene 1d (148 mg, 0.36 mmol) in toluene (20 mL) was heated
under nitrogen at reflux for 16 h. Toluene was removed in
1
vacuo. H NMR of the mixture showed two isomeric products
in a 3:1 ratio. The residue was purified by slow column
chromatography without air pressure: a mixture of benzene
and hexanes (1:1) was used to pack the column and load the
sample; 10% ethyl acetate in hexanes to elute a fast unknown;
15% ethyl acetate in hexanes to elute the fast isomer 6d (28
mg, 19%), and the slow isomer 5d (82 mg, 55%).
For 6b: 1H NMR (CDCl3, 300 MHz) δ 7.08-7.34 (m, 10H),
5.65 (d, 1H, J ) 14.7 Hz), 5.59-6.64 (m, 1H), 5.39 (d, broad,
1H, J ) 9.6 Hz), 4.05-4.19 (m, 2H), 3.75 (d, 1H, J ) 14.8 Hz),
3.62 (dt, 1H, J ) 6.9 Hz, 3.6 Hz, 3.6 Hz, H1), 3.18 (dd, 1H, J
) 3.8 Hz, 14.1 Hz), 2.34-2.45 (m, 3H), 2.20-2.31 (m, 2H),
2.02-2.11 (m, 1H), 1.81 (dt, 1H, J ) 6.9 Hz, 10.8 Hz, 10.8 Hz,
for H8a), 1.21 (t, 3H, J ) 7.2 Hz), 1.06 (dd, 1H, J ) 13.7 Hz,
15.9 Hz for H7′); 13C NMR (CDCl3, 100 MHz) δ 175.3, 173.2,
137.5, 136.2, 131.0, 129.3, 129.1, 129.9, 128.5, 127.9, 127.4,
126.2, 61.2, 57.4, 47.0, 45.8, 41.7, 38.6, 37.9, 35.4, 30.3, 14.6;
obsd NOE (CDCl3, 400 MHz) between H1 and H4a, H4 and
H8a, between H1 and H8a, H8 and H8a, no NOE between H4a
and H8a; HRMS calcd for C26H30NO3 (MH+) 404.2226, found
404.2221.
For 5d : 1H NMR (CDCl3, 300 MHz) δ 7.20 (d, 1H, J ) 8.2
Hz), 6.39-6.44 (m, 2H), 5.25 (d, 1H, J ) 14.4 Hz), 3.96-4.07
(m, 2H), 3.75-3.88 (m, 1H), 3.79 (s, 3H), 3.78 (s, 3H), 3.10 (dd,
1H, J ) 1.0 Hz, 4.4 Hz, H1), 2.76 (s, broad, 1H, H4a), 2.60
(dd, 1H, J ) 7.4 Hz, 17.3 Hz), 2.32-2.51 (m, 2H), 2.07-2.28
(m, 3H), 1.94 (d, broad, 1H, J ) 18.1 Hz), 1.14 (t, 3H, J ) 7.2
Hz), 1.00 (d, 3H, J ) 7.3 Hz), 0.98 (d, 3H, J ) 7.0 Hz); 13C
NMR (CDCl3, 75 MHz) δ 174.4, 169.3, 160.2, 158.6, 132.0,
129.6, 125.0, 117.9, 104.3, 98.1, 62.1, 60.4, 55.32, 55.27, 41.6,
40.3, 36.1, 33.0, 31.5, 28.9, 26.8, 19.9, 18,4, 14.0; obsd NOE
(CDCl3, 400 MHz) between H1 and H7 and between H4a and
Me of i-Pr.
For 6d : 1H NMR (CDCl3, 300 MHz) δ 7.23 (d, 1H, J ) 9.0
Hz), 6.41-6.44 (m, 2H), 5.63-5.69 (m, 1H), 5.48 (d, broad, J
) 9.8 Hz), 5.04 (d, 1H, J ) 14.2 Hz), 4.03-4.20 (m, 2H), 3.95
(d, 1H, J ) 14.2 Hz), 3.85 (s, 3H), 3.79 (s, 3H), 3.77-3.80 (m,
1H, H1), 2.34-2.59 (m, 3H), 2.18-2.33 (m, 1H), 1.98-2.08 (m,
2H), 1.73 (m, 1H, J ) 7.2 Hz), 1.57 (dt, 1H, J ) 6.7 Hz, 10.5
Hz, 10.5 Hz, H8a), 1.28 (t, 3H, J ) 7.1 Hz), 1.09 (d, 3H, J )
7.3 Hz), 0.84 (d, 3H, J ) 6.7 Hz); 13C NMR (CDCl3, 75 MHz)
δ 174.6, 173.0, 160.0, 158.4, 131.8, 129.1, 125.5, 118.1, 103.6,
98.2, 63.3, 60.6, 55.2, 54.9, 46.4, 46.1, 44.6, 39.2, 35.8, 35.6,
30.0, 22.5, 15.7, 14.0; HRMS calcd for C24H33NO5 (MH+)
416.2437, found 416.2437.
Diels-Ald er Rea ction To Ma k e Com p ou n d s 5e a n d 6e.
Triene 1e (150 mg, 0.41 mmol) in toluene (20 mL) was heated
under nitrogen at reflux for 6 days. Toluene was removed in
vacuo. 1H NMR showed two isomers in a 2.5:1 ratio. The
residue was purified by slow column chromatography without
air pressure: a mixture of benzene and hexanes (1:1) was used
to pack column and load the sample; 5% ethyl acetate in
hexanes to elute the fast minor isomer 6e (31 mg, 21%); 10%
ethyl acetate in hexanes to elute the slow isomer 5e (67 mg,
45%).
For 7b: 1H NMR (CDCl3, 300 MHz) δ 6.92-7.33 (m, 10H),
5.62-5.74 (m, 1H), 5.46-5.54 (m, 1H), 5.34 (d, 1H, J ) 15.1
Hz), 3.84-3.93 (m, 2H), 3.72 (dt, 1H, J ) 4.3 Hz, 4.3 Hz, 8.6
Hz, H1), 3.10 (dd, 1H, J ) 3.7 Hz, 13.9 Hz), 3.00 (d, 1H, J )
15.1 Hz), 2.56-2.72 (m, 3H), 2.12-2.38 (m, 3H), 1.97 (dt, 1H,
J ) 4.1 Hz, 11.5 Hz, 11.5 Hz, H8a), 1.01 (t, 3H, J ) 7.1 Hz);
13C NMR (CDCl3, 100 MHz) δ 173.9, 169.7, 138.9, 137.4, 129.6,
129.1, 128.8, 128.3, 128.0, 127.8, 127.5, 127.1, 124.8, 60.9, 58.7,
49.2, 42.7, 42.5, 38.0, 37.5, 30.3, 29.6, 14.2; obsd NOE (CDCl3,
400 MHz) between H1 and H8a, between H4a and H-R1, no
NOE between H4a and H8a; 1H decoupling NMR (CDCl3, 400
MHz) at 1.97 gave J 8a-1 ) 4.3 Hz; HRMS calcd for C26H30NO3
(MH+) 404.2226, found 404.2239.
Diels-Ald er Rea ction To Ma k e Com p ou n d s 5c a n d 6c.
Triene 1c (116 mg, 0.28 mmol) in toluene (10 mL) was heated
under nitrogen at reflux for 50 h. Toluene was removed in
vacuo. The residue was purified by slow column chromatog-
raphy without air pressure: a mixture of benzene and hexanes
(1:1) was used to pack column and load the sample; 5% and
then 10% ethyl acetate in hexanes to elute the unreacted
starting material (9 mg, 8%); 15% ethyl acetate in hexanes to
elute the fast isomer 6c (24 mg, 21%, oil); 20% and 30% ethyl
acetate in hexanes to elute a second fraction 5c (74 mg, 64%,
oil).
For 5e: 1H NMR (CDCl3, 300 MHz) δ 7.24-7.31 (m, 5H),
5.64 (d, 1H, J ) 14.3 Hz), 5.27-5.37 (m, 2H), 4.00-4.09 (m,
2H), 3.62 (d, 1H, J ) 14.4 Hz), 3.02 (dd, 1H, J ) 3.0 Hz, 5.2
Hz, H1), 2.77 (s, broad, 1H, H4a), 2.61 (dd, 1H, J ) 5.4 Hz,
15.8 Hz), 2.50 (d, broad, 1H, J ) 6.6 Hz, H8a), 2.37 (dd, 1H, J
) 3.5 Hz, 15.7 Hz), 1.97-2.08 (m, 2H), 1.17 (t, 3H, J ) 7.1
Hz), 1.13 (d, 3H, J ) 7.1 Hz), 1.04 (d, 3H, J ) 6.9 Hz), 0.99 (s,
3H); 13C NMR (CDCl3, 75 MHz) δ 176.6, 170.9, 136.9, 129.8,
128.4, 127.9, 127.5, 127.4, 60.5, 59.5, 49.4, 45.5, 40.5, 36.9, 36.3,
31.7, 29.9, 23.7, 21.4, 18.6, 13.9; coupling constants, 0 Hz
between H4a and H8a, 5.2-6.6 Hz between H1 and H8a (not
accurate due to broadness); obsd NOE (CDCl3, 400 MHz)
between H1 and H7, H1 and Me8, H4a and H8a. HRMS calcd
for C23H32NO3 (MH+) 370.2382, found 370.2383.
For 5c: 1H NMR (CDCl3, 300 MHz) δ 7.18-7.38 (m, 10 H),
5.50 (d, 1H, J ) 14.4 Hz), 5.37-5.42 (m, 1H), 5.30 (d, broad,
1H, J ) 10.3 Hz), 4.02 (q, 2H, J ) 7.1 Hz), 3.46 (d, 1H, J )
14.4 Hz), 3.27-3.32 (m, 1H, H1), 3.03 (dd, 1H, J ) 6.5 Hz,
13.4 Hz), 2.79 (dd, 1H, J ) 5.0 Hz,13.4 Hz), 2.63 (s, broad,
2H, H4a and H8a), 2.24 (dd, 1H, J ) 2.8 Hz, 15.5 Hz), 2.04
(dd, 1H, J ) 4.4 Hz, 18.1 Hz), 1.91 (dd, 1H, J ) 4.3 Hz, 15.5
Hz), 1.16 (t, 3H, J ) 7.0 Hz), 0.76 (s, 3H), 0.63 (dd, 1H, J )
2.4 Hz, 17.9 Hz); 13C NMR (CDCl3, 75 MHz) δ 176.2, 170.3,
136.7, 136.6, 130.1, 129.6, 128.6, 128.5, 128.1, 127.6, 127.2,
127.0, 60.5, 54.9, 47.4, 45.1, 42.6, 40.5, 36.4, 31.3, 29.4, 23.2,
13.8; obsd NOE (CD3OD, 400 MHz) between H4a and H8a,
H1 and Me8, H1 and H7; HRMS calcd for C27H32NO3 (MH+)
418.2382, found 418.2388.
For 6e: 1H NMR (CDCl3, 300 MHz) δ 7.23-7.36 (m, 5H),
5.66 (d, 1H, J ) 13.5 Hz), 5.60-5.66 (m, 1H), 5.48 (d, broad,
1H, J ) 9.9 Hz), 3.97-4.13 (m, 2H), 3.75 (d, 1H, J ) 11.6 Hz),
3.40 (dd, 1H, J ) 0.9 Hz, 6.7 Hz, H1), 2.86 (d, broad, 1H, J )
17.4 Hz), 2.55 (d, 1H, J ) 13.1 Hz), 2.00-2.18 (m, 2H), 1.86
(dd, 1H, J ) 4.7 Hz, 18.3 Hz), 1.64 (dd, 1H, J ) 6.8 Hz, 11.2
Hz, H8a), 1.52 (m, 1H, J ) 7.0 Hz), 1.24 (t, 3H, J ) 7.1 Hz),
1.13 (d, 3H, J ) 7.3 Hz), 0.86 (d, 3H, J ) 6.7 Hz), 0.67 (s, 3H);
13C NMR (CDCl3, 100 MHz) δ 176.5, 173.5, 137.5, 128.6, 128.5,
127.7, 127.5, 60.9, 59.7, 49.4, 48.6, 44.1, 38.9, 36.9, 36.3, 31.9,
23.7, 15.9, 15.5, 13.9; obsd NOE (CDCl3, 400 MHz) between
H1 and Me8, H4a and Me8.
For 6c: 1H NMR (CDCl3, 300 MHz) δ 7.08-7.36 (m, 10 H),
5.64 (d, 1H, J ) 14.7 Hz), 5.55-5.60 (m, 1H), 5.36 (d, broad,
1H, J ) 10.7 Hz), 4.08-4.27 (m, 2H), 3.58 (d, 1H, J ) 14.5
Hz), 3.49-3.54 (m, 1H, H1), 3.07 (dd, 1H, J ) 3.7 Hz, 14.0
Hz), 2.86 (d, broad, 1H, J ) 19.6 Hz), 2.30 (dd, 1H, J ) 4.4
Hz, 14.0 Hz), 2.20 (dd, 1H, J ) 1.9 Hz, 15.6 Hz), 1.85-2.01
(m, 3H), 1.29 (t, 3H. J ) 7.2 Hz), 0.98 (dd, 1H, J ) 12.7 Hz,
15.4 Hz), 0.69 (s, 3H); 13C NMR (CDCl3, 100 MHz) δ 177.0,

5938 J . Org. Chem., Vol. 64, No. 16, 1999
Murray et al.
Diels-Ald er R ea ct ion To Ma k e Com p ou n d s 9a a n d
10a . Triene 2a (96 mg, 0.29 mmol) in toluene (10 mL) was
heated under nitrogen at reflux for 7 h. Toluene was removed
in vacuo. The residue was purified by slow column chroma-
tography without air pressure: the fast isomer was eluted with
0.5% methanol in methylene chloride to give 9a (65 mg, 68%).
The slow isomer was eluted with 1% methanol in methylene
chloride to give 10a (21 mg, 22%).
the procedure for synthesis of 9b and 11b. The cycloaddition
was found complete in 20 h. 1H NMR showed a mixture of
isomeric products in a ratio of 83:25:1:1. The residue was
purified by slow column chromatography without air pres-
sure: a mixture of benzene and hexanes (1:1) was used to pack
column and load the sample; 8% ethyl acetate in hexanes to
elute a fraction 9c (16 mg, pure) and a second fraction (79 mg)
as a mixture of 9c and 10c. The total recovered yield from 2c
was 73%. The second fraction was purified by a second round
of column chromatography: 0.25% and 0.5% ethyl acetate in
methylene chloride to elute slowly. The second isomer 10c was
the fast moving in this solvent system, and an early fraction
was collected to give 10c (4 mg).
For 9a : 1H NMR (CDCl3, 300 MHz) δ 7.14-7.35 (m, 5H),
6.09 (dt, 1H, J ) 2.1 Hz, 2.1 Hz, 9.7 Hz), 5.60 (dt, 1H, J ) 3.4
Hz, 3.4 Hz, 9.7 Hz), 5.23 (s, broad, 1H), 4.11-4.28 (m, 2H),
3.74 (ddd, 1H, J ) 3.4 Hz, 9.8 Hz, 12.3 Hz, H3), 3.12 (dd, 1H,
J ) 3.2 Hz, 13.9 Hz), 2.82 (dd, 1H, J ) 2.5 Hz, 13.2 Hz, H7a),
2.51 (dd, 1H, J ) 12.5 Hz, 13.2 Hz), 2.45 (dd, 1H, J ) 9.3 Hz,
13.2 Hz, H3a), 2.18-2.22 (m, 1H), 1.37 (s, 3H), 1.32 (t, 3H, J
) 7.1 Hz), 1.01 (d, 3H, J ) 7.3 Hz); 13C NMR (CDCl3, 75 MHz)
δ 175.0, 173.9, 137.9, 133.0, 128.9, 128.6, 126.9, 121.8, 60.6,
56.0, 48.9, 46.3, 43.2, 41.7, 40.5, 18.2, 17.9, 14.0; 1H NMR
coupling constants, 13.2 Hz between H3a and H7a, 9.3-9.8
Hz between H3 and H3a; HRMS calcd for C20H26NO3 (MH+)
328.1913, found 328.1915.
For 9c: 1H NMR (CDCl3, 300 MHz) δ 7.18-7.22 (m, 6H),
6.96-6.99 (m, 2H), 6.83-6.87 (m, 2H), 6.37 (dt, 1H, J ) 2.8
Hz, 2.8 Hz, 9.2 Hz), 5.62 (dt, 1H, J ) 3.0 Hz, 9.2 Hz), 4.98 (d,
1H, J ) 15.4 Hz), 4.02 (d, 1H, J ) 15.4 Hz), 3.99 (t, 2H, J )
7.1 Hz), 3.80 (ddd, 1H, J ) 3.7 Hz, 6.2 Hz, 10.1 Hz, H3), 3.00
(dd, 1H, J ) 3.8 Hz, 15.6 Hz), 2.71-2.81 (m, 2H), 2.45 (dd,
1H, J ) 10.0 Hz, 13.0 Hz, H3a), 1.88-2.04 (m, 1H), 1.16 (s,
3H), 1.14 (t, 3H, J ) 7.2 Hz), 0.99 (d, 3H, J ) 7.3 Hz); 13H
NMR (CDCl3, 75 MHz) δ 175.2, 173.3, 137.5, 136.6, 133.8,
128.8, 128.5, 128.3, 127.3, 127.0, 126.3, 125.9, 60.5, 58.0, 48.8,
46.2, 44.2, 43.2, 42.2, 37.1, 18.3, 16.3, 13.9; HRMS calcd for
For 10a : 1H NMR (CDCl3, 300 MHz) δ 7.13-7.34 (m, 5H),
5.93 (dt, 1H, J ) 3.3 Hz, 3.3 Hz, 9.9 Hz), 5.69 (d, 1H, NH),
5.50 (dt, 1H, J ) 2.1 Hz, 2.1 Hz, 9.9 Hz), 4.01-4.09 (m, 1H),
3.83-3.95 (m, 1H), 3.52 (t, broad, 1H, J ) 6.8 Hz, H3), 3.04
(dd, 1H, J ) 3.2 Hz, 8.9 Hz, H7a), 2.88 (d, 1H, J ) 8.9 Hz,
H3a), 2.72-2.82 (m, 2H), 2.63-2.72 (m, 1H), 1.22 (t, 3H, J )
7.2 Hz), 0.93 (s, 3H), 0.91 (t, 3H, J ) 7.3 Hz); 13C NMR (CDCl3,
75 MHz) δ 176.03, 175.97, 136.9, 132.3, 129.2, 128.6, 126.7,
121.5, 60.7, 55.5, 48.1, 44.9, 42.2, 39.2, 35.8, 15.9, 14.2, 10.1;
obsd NOE (CDCl3, 400 MHz) between H3a and H7a, H3 and
Me4, H3a and H5; HRMS calcd for C20H25NO3 (M+) 327.1834,
found 327.1829.
C
27H32NO3 (MH+) 418.2382, found 418.2381.
For 10c: 1H NMR (CDCl3, 300 MHz) δ 7.20-7.34 (m, 8H),
7.06 (d, 2H, J ) 6.7 Hz), 5.94 (dt, 1H, J ) 3.4 Hz, 3.4 Hz, 9.9
Hz), 5.42 (dt, 1H, J ) 2.6 Hz, 2.6 Hz, 9.2 Hz), 4.92 (d, 1H J )
14.3 Hz), 4.13 (d, 1H, J ) 14.5 Hz), 3.72-3.84 (m, 1H), 3.37-
3.47 (m, 2H), 2.78-2.85 (m, 2H), 2.69 (d, 1H, J ) 8.7 Hz), 2.53
(m, 1H), 1.03 (t, 3H, 7.2 Hz), 0.75 (d, 3H, J ) 7.3 Hz), 0.46 (s,
3H); 13C NMR (CDCl3, 75 MHz) δ 175.5, 173.1, 136.6, 135.8,
132.0, 129.32, 129.29, 128.5, 128.3, 127.8, 126.7, 122.1, 60.4,
58.7, 48.0, 45.0, 42.3, 40.0, 37.8, 35.7, 15.8, 13.9, 9.4; obsd NOE
between H3a and H5, H7a and H8, between Me4 and Me5,
H3 and Me4; HRMS calcd for C27H32NO3 (MH+) 418.2382,
found 418.2376.
Diels-Ald er Rea ction To Ma k e Com p ou n d s 9d . Com-
pound 4g (302 mg, 0.94 mmol) was used by following the
procedure for synthesis of 9b and 11b. The cycloaddition was
complete in 60 h. 1H NMR showed a very clean mixture of
three isomeric products in a ratio of 20:2:1. The major isomer
could not be isolated by three rounds of slow column chroma-
tography using different solvent systems. A total of 355 mg
product (91% from 4g) was recovered from column chroma-
tography. The stereochemistry of 9d was assigned using the
product mixture.
Diels-Ald er R ea ct ion To Ma k e Com p ou n d s 9b a n d
11b. The crude triene 2b, prepared from 4e (90 mg, 0.29 mmol)
by following the general procedure for acylation, was dissolved
in chloroform and kept in the dark. The Diels-Alder reaction
1
was monitored by H NMR (a small part of the solution was
taken in a separate flask; chloroform was blown away with
nitrogen; the residue was dissolved in CDCl3). The cycloaddi-
tion was complete in 12 h. 1H NMR showed an isomeric
product ratio of 10:10:1:1. The residue was purified by slow
column chromatography without air pressure: a mixture of
benzene and hexanes (1:1) was used to pack column and load
the sample; 5% ethyl acetate in hexanes to elute the fast
isomer 9b (46 mg, 39%); 8% ethyl acetate in hexanes to elute
a second pure isomer 11b (6 mg) and a third fraction of 11b
(44 mg) containing minor 10b and 12b.
For 9d : 1H NMR (CDCl3, 300 MHz) δ 7.05 (d, 1H, J ) 9.1
Hz), 6.42-6.46 (m, 3H), 5.64 (dt, 1H, J ) 3.0 Hz, 3.0 Hz, 9.1
Hz), 4.98 (d, 1H, J ) 15.4 Hz), 4.15 (q, 2H, J ) 7.1 Hz), 4.02
(d, 1H, J ) 15.5 Hz), 3.81 (s, 3H), 3.80 (s, 3H), 3.20 (dd, 1H J
) 2.2 Hz, 9.8 Hz, H3), 2.81 (t, 1H, J ) 9.2 Hz), 2.46-2.62 (m,
2H), 2.19-2.34 (m, 1H), 2.12 (dt, 1H, J ) 9.6 Hz, 9.6 Hz, 12.7
Hz, H3a), 1.27 (t, 3H, J ) 7.2 Hz), 1.11 (d, 3H, J ) 7.4 Hz),
0.84 (d, 3H, J ) 7.3 Hz), 0.79 (d, 3H, J ) 7.0 Hz); obsd NOE
(CDCl3, 400 MHz) between H3 and H4, H3 and H7a, H4 and
H5, H4 and H7a, weak between H4 and Me5; HRMS calcd for
For 9b: 1H NMR (CDCl3, 300 MHz) δ 6.50-7.36 (m, 10H),
6.15 (d, 1H, J ) 9.6 Hz), 5.55 (dt, 1H, J ) 3.3 Hz, 9.7 Hz),
5.11 (d, 1H, J ) 15.3 Hz), 3.96-4.16 (m, 3H), 3.68 (ddd, 1H, J
) 3.1 Hz, 5.8 Hz, 9.0 Hz, H3), 3.39 (dd, 1H, J ) 3.0 Hz, 15.6
Hz), 3.01 (dd, 1H, J ) 5.7 Hz, 15.7 Hz), 2.93 (dd, 1H, J ) 6.6
Hz, 11.0 Hz, H4), 2.70-2.87 (m, 2H), 2.12 (dt, 1H, 9.9 Hz, 11.2
Hz, 11.2 Hz, H3a), 1.21 (t, 3H, J ) 7.2 Hz), 0.95 (d, 3H, J )
7.2 Hz); 13C NMR (CDCl3, 75 MHz) δ 173.0, 172.6, 137.7, 136.7,
133.2, 129.5, 128.5, 128.3, 127.7, 127.2, 126.4, 123.0, 60.4, 59.8,
48.7, 47.7, 44.2, 39.6, 36.2, 34.1, 17.4, 14.2; obsd NOE: between
H3a and Me5, H4 and H5, H3 and H4, H3 and H7a, H3 and
H3a, H3a and H4; HRMS calcd for C26H30NO3 (MH+) 404.2226,
found 404.2241.
C
24H34NO5 (MH+) 416.2437, found 416.2455.
Diels-Ald er R ea ct ion To Ma k e Com p ou n d s 9e a n d
10e. Compound 4h (225 mg, 0.82 mmol) was used by following
the procedure for synthesis of 9b and 11b. The cycloaddition
completed in 80 h. 1H NMR showed a mixture of three isomeric
products in a ratio of 185:50:1. The residue was purified by
slow column chromatography without air pressure: a mixture
of benzene and hexanes (1:1) was used to pack column and
load the sample; 3-5% ethyl acetate in hexanes to elute. Three
fractions were collected to give a first fraction of 24 mg
(mixture), a second fraction of 203 g (mixture), and a third
fraction of 14 mg (pure 9e). The total recovery yield was 80%
based on 2e. The first fraction of 24 mg was again purified
using the same solvent system to give the fast moving isomer
10e (8 mg).
For 11b: 1H NMR (CDCl3, 300 MHz) δ_6.90-7.34 (m, 10
H), 6.12 (d, 1H, J ) 9.8 Hz), 5.58 (dt, 1H, J ) 3.4 Hz, 3.4 Hz,
9.8 Hz), 4.95 (d, 1H, J ) 14.9 Hz), 3.91-4.08 (m, 3H), 3.20 (d,
1H, J ) 14.9 Hz), 3.00 (dd, 1H, J ) 7.0 Hz, 12.0 Hz, H4), 1.36
(m, 1H, H7a), 2.76 (m, 1H, H5), 2.65-2.78 (m, 2H), 2.78 (dt,
1H, J ) 6.5 Hz, 12.6 Hz, 12.6 Hz, H3a), 1.19 (t, 3H, J ) 7.2
Hz), 0.89 (d, 3H, J ) 7.1 Hz); 13C NMR (CDCl3, 75 MHz) δ
172.8, 172.5, 138.1, 136.7, 133.0, 129.3, 128.7, 128.5, 128.1,
127.4, 126.7, 122.8, 60.4, 58.7, 45.2, 45.0, 42.6, 39.8, 34.2, 33.2,
17.0, 14.1; obsd NOE between H4 and H8, H7a and H8, H4
and H7a, H3a and Me5, H4 and H5; HRMS calcd for C26H30
-
NO3 (MH+) 404.2226, found 404.2238.
For 9e: 1H NMR (CDCl3, 300 MHz) δ 7.23-7.35 (m, 3H),
7.17 (d, 2H, J ) 6.7 Hz), 6.50 (dt, 1H, J ) 3.2 Hz, 3.2 Hz, 9.9
Hz), 5.66 (dt, 1H, J ) 2.9 Hz, 2.9 Hz, 10.0 Hz), 5.20 (d, 1H, J
Diels-Ald er R ea ct ion To Ma k e Com p ou n d s 9c a n d
10c. Compound 4f (95 mg, 0.29 mmol) was used by following

Diels-Alder Reactions of Amino Acid-Derived Trienes
J . Org. Chem., Vol. 64, No. 16, 1999 5939
) 15.4 Hz), 4.12 (q, 2H, J ) 7.2 Hz), 3.88 (d, 1H, J ) 15.5 Hz),
3.37 (dd, 1H, J ) 2.2 Hz, 10.0 Hz, H3), 2.68 (dq, 1H, J ) 2.7
Hz, 2.7 Hz, 2.7 Hz, 12.8 Hz, H7a), 2.42 (dd, 1H, J ) 10.0 Hz,
12.8 Hz, H3a), 1.89-2.03 (m, 2H), 1.25 (t, 3H, J ) 7.1 Hz),
1.15 (s, 3H), 1.04 (d, 3H, J ) 7.4 Hz), 0.78 (d, 3H, J ) 7.1 Hz),
0.76 (d, 3H, J ) 7.2 Hz); 13C (CDCl3, 75 MHz) δ 175.8, 173.9,
136.6, 134.1, 129.0, 128.6, 127.6, 127.3, 61.9, 60.6, 46.7, 46.4,
44.3, 43.7, 42.9, 27.0, 18.3, 17.4, 16.6, 15.5, 14.1; HRMS calcd
for C23H32NO3 (MH+) 370.2382, found 370.2379.
1.28 (t, 3H, J ) 7.1 Hz), 1.00 (s, 3H), 0.94 (d, 3H, J ) 6.8 Hz),
0.88 (d, 3H, J ) 7.3 Hz), 0.84 (d, 3H, J ) 6.9 Hz); 13C NMR
(CDCl3, 75 MHz) δ 177.3, 140.9, 130.9, 128.2, 128.1, 126.6,
125.7, 60.5, 59.3, 58.3, 48.2, 46.4, 39.5, 37.1, 32.8, 18.5, 17.9,
16.5, 14.1, 13.3; obsd NOE (CDCl3, 400 MHz) H3a-H5, H3a-
H7a, H3-Me4.
Diels-Ald er Rea ction To Ma k e Com p ou n d 9j. To the
primary amine triene 2i (110 mg, 0.415 mmol) in THF (10 mL)
were added Et3N (0.4 mL) and PhCOCl (70 mL, 0.60 mmol).
The mixture was stirred for 0.5 h, quenched with 5% NaHCO3
(50 mL), and extracted with ether. The ether layer was dried
over MgSO4 (0.5 h), evaporated, and pumped under high
vacuum for 1 h. A part of the residue was dissolved in CDCl3,
and the second part was dissolved in C6D6. The reaction was
monitored by 1H NMR and was found complete in 18 h in
either CDCl3, or C6D6 (the time was counted immediately after
For 10e: 1H NMR (CDCl3, 300 MHz) δ 7.29 (s, 5H), 6.03
(dt, 1H, J ) 3.4 Hz, 3.4 Hz, 9.9 Hz), 5.45-5.51 (m, 1H), 4.79
(d, 1H, J ) 14.4 Hz), 4.09-4.20 (m, 1H), 4.05 (d, 1H, J ) 14.5
Hz), 3.90-4.01 (m, 1H), 3.14 (dt, 1H, J ) 3.2 Hz, 3.2 Hz, 9.2
Hz, H7a), 3.09 (d, 1H, J ) 3.1 Hz, H3), 2.69-2.73 (m, 1H),
2.64 (d, 1H, J ) 9.2 Hz, H3a), 2.00-2.11 (m, 1H), 1.21 (t, 3H,
J ) 7.1 Hz), 0.75-0.84 (m, 9H), 0.53 (s, 3H); 13C NMR (CDCl3,
300 MHz) δ 176.2, 173.3, 136.2, 131.9, 129.2, 128.5, 127.8,
122.7, 62.8, 60.7, 48.6, 44.8, 41.9, 38.7, 36.0, 28.4, 18.2, 16.2,
15.6, 14.1, 9.2; obsd NOE (CDCl3, 400 MHz) between H3 and
Me4, H3a and H7a, Me4 and Me5, H5 and Me4; HRMS calcd
for C23H32NO3 (MH+) 370.2382, found 370.2371.
Diels-Ald er Rea ction To Ma k e Com p ou n d s 9f a n d 10f.
Followed exactly the same procedure to make 9b and 11b.
Proton NMR of the crude but clean Diels-Alder reaction
mixture showed two isomeric products 9f and 10f in a 6:1 ratio,
as compared with that of the crude reaction mixture of 9e and
10e.
Diels-Ald er R ea ct ion To Ma k e Com p ou n d s 9g a n d
11g. Compound 3a (110 mg, 2.11 mmol) after removal of Boc
group was used for reductive alkylation with 2,4-hexadienal
by following the procedure to make 2i and 2h . The reduction
was allowed for 6 h after addition of triacetoxyborohydride.
1H NMR of the crude mixture indicated the completion of the
cycloaddition of 2g, and showed two diastereomers (9g:11g )
5:1). The major and more polar isomer 9g was isolated in pure
form (68 mg, 66%), eluted with 20-30% ethyl acetate in
hexanes from column chromatography: 1H NMR (CDCl3, 300
MHz) δ 7.16-7.31 (m, 5H), 5.75 (dt, 1H, J ) 9.8 Hz, 1.7 Hz),
5.56 (dt, 1H, J ) 9.7 Hz, 3.3 Hz), 4.18 (q, 2H, J ) 7.1 Hz),
3.20-3.29 (m, 2H), 3.03 (dd, 1H, J ) 8.5 Hz, 7.0 Hz, H1), 2.90
(dd, 1H, J ) 11.3 Hz, 6.8 Hz, H4), 2.74-2.85 (m, 1H, H5), 2.59
(dd, 1H, J ) 12.5 Hz, 10.0 Hz, H-R1), 2.53 (dd, 1H, J ) 11.9
Hz, 8.6 Hz, H1′), 2.36-2.46 (m, 1H, H7a), 1.75 (q, 1H, J )
11.1 Hz, H3a), 1.30 (t, 3H, J ) 7.1 Hz), 1.02 (d, 3H, J ) 7.2
Hz); 13C NMR (CDCl3, 75 MHz) δ 173.4, 140.0, 132.9, 129.0,
128.2, 125.8, 61.3, 60.0, 48.2, 47.6, 45.2, 44.6, 41.5, 33.5, 29.6,
17.8, 14.2; HRMS calcd for C19H26NO2 (MH+) 300.1965, found
300.1953.
Diels-Ald er R ea ct ion To Ma k e Com p ou n d s 9h a n d
10h . Compound 2h (101 mg), after isolation from column
chromatography, was immediately dissolved in deuterated
chloroform. The cycloaddition was complete in 15 days as
monitored by 1H NMR and gave two isomers in a 3:2 ratio.
The mixture was eluted first with a mixture of benzene and
hexanes (1:1) to remove impurities and then with benzene to
give the minor isomer 10h (32 mg) and finally with 5% ethyl
acetate in hexanes to give the major isomer 9h (52 mg). Total
yield was 84%.
1
the addition of PhCOCl). The H NMR showed a mixture of 2
isomers in a ratio of 9:1. Column chromatography using 5-10%
ethyl acetate in hexanes gave the pure major isomer 9j (89
mg, 58%) and a fraction of mixtures (5 mg, 3%). The total yield
is 61%. For 9j: 1H NMR (C6D6, 300 MHz) δ 7.64-7.68 (m, 2H),
7.09-7.16 (m, 3H), 5.42-5.52 (m, 2H), 4.63 (d, broad, 1H, J )
8.2 Hz, H1), 3.84-3.95 (m, 2H), 3.45 (dd, 1H, J ) 6.6 Hz, 9.6
Hz), 2.84 (dd, 1H, J ) 10.3 Hz, 11.6 Hz), 2.56-2.68 (m, 1H),
2.43 (dd, 1H, J ) 10.0 Hz, 11.8 Hz, H3a), 1.72-1.80 (m, 2H),
1.21 (d, 3H, J ) 7.0 Hz), 1.17 (s, 3H), 0.90 (m, 9H); 13C NMR
(CDCl3, 75 MHz) δ 176.0, 171.0, 137.0, 134.8, 133.6, 130.3,
128.4, 128.3, 127.9, 127.6, 61.5, 60.5, 54.4, 49.5, 47.4, 42.0, 40.4,
30.7, 19.8, 18.8, 17.4, 16.7, 14.2; obsd NOE (C6D6, 400 MHz)
H3-Me4.
Com p ou n d s 9k . To benzylamine (1.60 mL, 14.6 mmol) in
methylene chloride (20 mL) were added ethyl 4-bromocrotonate
(1.0 mL, 7.3 mmol) and 4-(dimethylamino)pyridine (200 mg).
The mixture was stirred overnight and separated between
ether (200 mL) and water (200 mL). The ether layer was dried
over Na2SO4 and concentrated. The residue was purified by
column chromatography and eluted using 10-20% ethyl
acetate in hexanes to give ethyl 4-benzylaminocrotonate (0.81
g, 51%): 1H NMR (CDCl3, 300 MHz) δ 7.22-7.34 (m, 5H), 7.00
(dt, 1H, J ) 15.6 Hz, 5.3 Hz, 5.3 Hz), 6.01 (dt, 1H, J ) 15.6
Hz, 1.6 Hz), 4.18 (q, 2H, J ) 7.1 Hz), 3.78 (s, 2H), 3.39 (dd,
2H, J ) 1.5 Hz, 5.3 Hz); 13C NMR (CDCl3, 75 MHz) δ 166.3,
146.6, 139.7, 128.3, 128.0, 127.0, 121.5, 60.1, 53.1, 49.4, 14.1;
LC-MS calcd for C13H17NO2 (MH+) 220, found 220.
To ethyl 4-benzylaminocrotonate (125 mg, 0.571 mmol) in
THF (10 mL) were added triethylamine (0.2 mL) and 2,4-
hexadienoyl chloride (100 mL, ∼0.75 mmol). The mixture was
stirred for 0.5 h, quenched with 5% Na2CO3 (50 mL), and
extracted with ether (50 mL). The ether layer was dried over
MgSO4 (∼0.5 h), evaporated, and pumped under high vacuum
for 1 h. A part of the residue was dissolved in CDCl3, and the
second part was dissolved in C6D6. The reaction was monitored
by 1H NMR and was found complete in 18 days in either CDCl3
or C6D6. The 1H NMR showed a mixture of two isomers in a
ratio of 6:1. Column chromatography using a mixture of
methylene chloride and hexanes (3:2) gave the pure major
isomer 9k (109 mg, 61%) and a fraction of mixtures (11 mg,
6%). The total yield is 67%. For 9k : 1H NMR (CDCl3, 300 MHz)
δ 7.21-7.35 (m, 5H), 6.12 (d, 1H, J ) 9.9 Hz), 5.59 (dt, 1H, J
) 3.1 Hz, 3.1 Hz, 9.9 Hz), 4.58 (d, 1H, J ) 14.7 Hz), 4.37 (d,
1H, J ) 14.7 Hz), 4.08-4.21 (m, 2H), 3.54 (dd, 1H, J ) 6.3
Hz, 9.5 Hz), 2.98 (t, 1H, J ) 9.9 Hz), 2.82-2.89 (m, 2H), 2.69
(d, broad, 1H, J ) 12.5 Hz, H7a), 2.27 (dq, 1H, J ) 6.3 Hz,
11.3 Hz, 11.3 Hz, 11.3 Hz, H3a), 1.25 (t, 3H, J ) 7.1 Hz), 0.93
(d, 3H, J ) 6.9 Hz); 13C NMR (CDCl3, 75 MHz) δ 172.7, 172.6,
136.7, 133.3, 128.7, 128.1, 127.5, 122.4, 60.4, 49.3, 47.3, 47.0,
46.7, 35.8, 33.2, 16.9, 14.2; LC-MS calcd for C19H24NO3 (MH+)
314, found 314.
For 9h : 1H NMR (CDCl3, 300 MHz) δ 7.19-7.41 (m, 5H),
5.71 (dt, 1H, J ) 9.6 Hz, 2.0 Hz), 5.54 (dt, 1H, J ) 9.6 Hz, 3.0
Hz), 4.02-4.22 (m, 2H), 3.85 (d, 1H, J ) 13.4 Hz), 3.66 (d, 1H,
J ) 13.4 Hz), 2.61-2.68 (m, 2H), 2.52 (t, 1H, J ) 11.0 Hz,
H1), 2.22-2.32 (m, 1H, H7a), 2.16 (dd, 1H, J ) 11.5 Hz, 9.7
Hz, H3a), 2.12 (m, 1H, H5), 1.70 (m, 1H, isopropyl), 1.28 (t,
3H, J ) 7.1 Hz), 1.25 (s, 3H), 0.97-1.02 (5 peaks, 9H); 13C
NMR (CDCl3, 75 MHz) δ 176.2, 141.7, 133.6, 128.1, 128.0,
127.4, 126.5, 69.8, 63.7, 60.2, 55.5, 48.5, 46.8, 42.2, 37.5, 30.7,
22.4, 18.3, 18.1, 15.8, 14.2; HRMS calcd for C23H34NO2 (MH+)
356.2589, found 356.2578; obsd NOE (CDCl3, 600 MHz) H3-
Me4, H7a-Me4, H3-H5, H5-Me4.
Diels-Ald er Rea ction To Ma k e Com p ou n d s 17 a n d 18.
Triene 16 (96 mg, 0.33 mmol) in toluene (10 mL) was heated
under nitrogen at reflux for 4 h. Toluene was removed in
vacuo. 1H NMR showed the two major isomers in a ratio of
1:1. The residue was purified by slow column chromatography
using 2-3% ethyl acetate in methylene chloride to elute the
fast isomer 17 (39 mg, pure). A second fraction was then
For 10h : 1H NMR (CDCl3, 300 MHz) δ 7.18-7.34 (m, 5H),
5.62 (dt, 1H, J ) 9.9 Hz, 3.2 Hz), 5.36 (d, broad, 1H, J ) 9.9
Hz), 3.98-4.24 (m, 3H), 3.55 (d, 1H, J ) 13.6 Hz), 2.94 (t, 1H,
J ) 7.9 Hz), 2.74-2.90 (m, 3H), 2.37 (dd, 1H, J ) 8.5 Hz, 2.2
Hz, H3a), 2.16 (dd, 1H, 12.0 Hz, 8.2 Hz), 1.67-1.80 (m, 1H),

5940 J . Org. Chem., Vol. 64, No. 16, 1999
Murray et al.
collected to give a mixture of 17 and a minor isomer 19 or 20
(total 3 mg). A third fraction eluted with 5% ethyl acetate in
hexanes was obtained in 49 mg, which was a mixture of the
major isomer 18 and some starting material 16 (18:16 ) 3:2).
A fourth fraction eluted with 6% ethyl acetate in hexanes gave
a mixture (5 mg, not pure) of 18 containing a second minor
isomer 20 or 19. The total yield was 84% (Diels-Alder products
only) or 100% based on the recovery of 16. For 17 (structural
assignment was from 1H NMR coupling constants): 1H NMR
(CDCl3, 300 MHz) δ 7.15-7.34 (m, 5H), 6.12 (d, 1H, J ) 9.9
Hz), 5.60 (dt, 1H, J ) 3.2 Hz, 3.2 Hz, 9.8 Hz), 5.17 (d, 1H, J )
15.2 Hz), 4.10-4.22 (m, 2H), 3.74 (d, 1H, J ) 15.2 Hz), 3.25
(dt, 1H, J ) 3.3 Hz, 3.3 Hz, 6.8 Hz, H6), 3.16 (dd, 1H, J ) 3.2
Hz, 10.2 Hz, H3), 2.52 (t, 1H, J ) 11.8 Hz, H7a), 2.28-2.39
(m, 2H), 2.16-2.23 (m, 1H), 1.42 (d, 3H, J ) 6.8 Hz), 1.29 (t,
3H, J ) 7.1 Hz), 0.99 (d, 3H, J ) 7.1 Hz), 0.83 (d, 3H, J ) 6.8
Hz); 13C NMR (CDCl3, 75 MHz) δ 175.4, 172.9, 136.8, 130.4,
128.6, 127.9, 127.3, 127.1, 62.0, 60.7, 48.5, 45.8, 43.4, 39.4, 33.6,
27.0, 19.7, 15.2, 15.0, 14.3; HRMS calcd for C22H29NO3 (MH+)
356.2226, found 356.2240.
For 18, the fraction 4 containing a minor isomer was used
for NMR analyses: 1H NMR (CDCl3, 300 MHz) δ 7.22-7.32
(m, 5H), 5.59-5.76 (m, 2H), 5.05 (d, 1H, J ) 15.1 Hz), 4.12-
4.22 (m, 1H), 3.85 (d, 1H, J ) 15.0 Hz), 3.06 (t, 1H, J ) 3.8
Hz, H3), 2.79 (s, broad, 1H, H6), 2.64 (s, broad, 1H, H3a), 2.49-
2.58 (m, 1H, H7), 2.44 (dd, 1H, J ) 6.9 Hz, 8.0 Hz, H7a), 1.99-
2.08 (m, 1H), 1.28 (t, 3H, J ) 7.3 Hz), 1.25 (d, 3H, J ) 6.5
Hz), 0.92 (d, 3H, J ) 6.9 Hz), 0.78 (d, 3H, J ) 6.9 Hz); obsd
NOE (CDCl3, 400 MHz) H3a-H7a, H6-H7a, H3-H7.
Com p ou n d s 23. To amine 14 (224 mg, 1.14 mmol) in THF
(10 mL), were added ethyl 4-bromocrotonate (282 µL, 1.50
mmol) and 4-(dimethylamino)pyridine (70 mg, 0.57 mmol). The
mixture was stirred overnight and separated between ether
(100 mL) and water (100 mL). The ether layer was dried over
Na2SO4 and concentrated. The residue was purified by column
chromatography and eluted using 10% ethyl acetate in hexanes
to give 21 (184 mg, 52%): 1H NMR (CDCl3, 300 MHz) δ 7.28
(dd, 1H, J ) 11.0 Hz, 16.1 Hz), 6.96 (dt, 1H, J ) 5.3 Hz, 5.3
Hz, 15.8 Hz), 6.23 (dd, 1H, J ) 11.0 Hz), 5.83-6.01 (m, 3H),
4.16-4.24 (m, 4H), 3.40 (ddd, 1H, J ) 2.0 Hz, 5.8 Hz, 16.8
Hz), 3.25 (ddd, 1H, J ) 1.4 Hz, 5.8 Hz, 16.4 Hz), 2.91 (dd, 1H,
J ) 5.8 Hz, 8.3 Hz), 1.65-1.79 (m, 1H), 1.26-1.32 (m, 6H),
0.94 (d, 3H, J ) 6.7 Hz), 0.89 (d, 3H, J ) 6.8 Hz); 13C NMR
(CDCl3, 75 MHz) δ 167.0, 166.4, 146.9, 143.8, 143.6, 130.1,
121.4, 120.7, 65.7, 60.2, 47.8, 32.6, 19.3, 18.2, 14.23, 14.18; LC-
MS calcd for C17H28NO4 (MH+) 310, found 310.
The second amine triene 21 (85 mg, 0.275 mmol) in THF (5
mL) was treated by following the procedure to make 9j. A part
of the residue was dissolved in CDCl3, and the second part
was dissolved in C6D6. The reaction was monitored by 1H NMR
and was found complete in 40 h in either CDCl3 or C6D6 (the
time was counted immediately after the addition of PhCOCl).
1
The H NMR of the reaction mixture showed a single isomer.
Column chromatography using 5-10% ethyl acetate in hex-
anes gave the pure 23 (105 mg, 92%): 1H NMR (C6D6, 300
MHz) δ 7.59 (t, 2H, J ) 3.6 Hz), 7.02-7.06 (m, 3H), 5.84 (d,
1H, J ) 9.8 Hz), 5.61 (dt, 1H, J ) 3.5 Hz, 9.8 Hz), 4.28 (dd,
1H, J ) 6.2 Hz, 10.2 Hz), 4.09 (dd, 1H, J ) 6.0 Hz, 11.1 Hz,
H3), 3.70-3.92 (m, 4H), 3.60-3.64 (m, 1H, H6), 2.96-3.05 (m,
2H), 2.47 (dq, 1H, J ) 6.1 Hz, 11.4 Hz, 11.4 Hz, 11.4 Hz, H7a),
2.13 (dd, 1H, J ) 6.3 Hz, 11.5 Hz, H7), 2.02 (t, broad, 1H, J )
12.0 Hz, H3a), 0.85-0.99 (m, 9H), 0.78 (t, 3H, J ) 7.1 Hz);
13C NMR (CDCl3, 75 MHz) δ 171.8, 171.5, 170.6, 136.8, 131.1,
130.3, 128.1, 127.8, 124.8, 63.7, 61.1, 60.6, 55.4, 44.7, 44.5, 43.8,
39.9, 28.4, 19.4, 16.1, 14.1, 13.8; HRMS calcd for C24H32NO5
(MH+) 414.2281, found 356.2263.
Ack n ow led gm en t. We thank Drs. G. H. Kuo and
P. Connolly for helpful discussions, M. Evangelisto for
purification of compound 10b, and W. J . J ones for
HRMS. Elemental analysis was performed by Quantita-
tive Technologies, Inc., NJ .
Su p p or tin g In for m a tion Ava ila ble: The compound pu-
1
rity is exemplified by H and 13C NMR spectra for the Diels-
Alder reaction products of trienes 2a -f and 22. Calculated
structures (Cartesian coordinates) and energies of all ground
and transition structures are given. This material is available
free of charge via the Internet at http://pubs.acs.org.
J O990459F