Extraordinary formation of a novel butadiene derivative, (Z,E)-1-(2-naphthylmethyloxycarbonyl-amino) -2,4-bis (4-nitrophenyl)butadiene, and subsequent isomerization into the corresponding (E,E)-isomer

Full text of DOI:10.1021/jo9802426

7516
J . Org. Chem. 1998, 63, 7516-7519
Notes
Sch em e 1
Extr a or d in a r y F or m a tion of a Novel
Bu ta d ien e Der iva tive,
(Z,E)-1-(2-Na p h th ylm eth yloxyca r bon yl-
a m in o)-2,4-bis(4-n itr op h en yl)bu ta d ien e,
a n d Su bsequ en t Isom er iza tion in to th e
Cor r esp on d in g (E,E)-Isom er
Min Li,† Leela Kar, and Michael E. J ohnson*
Center for Pharmaceutical Biotechnology and
Department of Medicinal Chemistry & Pharmacognosy
(M/ C 781), University of Illinois at Chicago,
Chicago, Illinois 60612-7231
Received February 10, 1998
Phosphorus-containing peptide analogues have been
widely utilized in designs of novel protease inhibitors
because of their ability to mimic the transition state of
enzymatic catalyses.
R-Amino phosphonates were most frequently incor-
porated in such designs;1-4 on the other hand, R-amino
phosphinates received less attention. In our ongoing
program of de novo design of thrombin inhibitors, we find,
as supported by molecular modeling, that the structural
features of R-amino phosphinates could be beneficial to
our goal of developing noncovalent active site inhibitors.
The R-amino phosphinates utilized in our design as the
key building blocks are N-CBZ-protected [1-amino-2-(4-
substituted-phenyl)ethyl]phenylphosphinates (1), which
were synthesized by adopting a procedure of Chen and
Dan5 (Scheme 1). In a typical preparation, benzyl
carbamate was treated with (4-nitrophenyl)acetaldehyde6
and dichlorophenylphosphine in acetic acid to give a good
yield (∼60%) of the desired 4-nitrophenyl compound 2.
When the benzyl carbamate was substituted by 2-naph-
thylmethyl carbamate in the same reaction under identi-
cal conditions, however, much less material was obtained.
Moreover, an unexpected compound 4 was formed as the
major product (a relative 80%, mol/mol) present in the
crude material, while the desired compound 3 was
produced in 20% yield. The major product was purified
careful examination of the 1H NMR spectrum (in DMSO-
d6) of 4 revealed the presence of an “impurity”, and it
was soon discovered that the peaks of this “impurity”
increased as those of 4 decreased (Figure 1) while the
NMR solution of the product stood at room temperature.
An important characteristic of this transformation (with
a half-life of ∼4 months) is that the total number of the
new peaks is equal to that of the old one (4) and, more
importantly, the new scalar coupling constants appear
to remain the same. This indicates that some key
structural features of the original product 4 were pre-
served and the transformation is an intramolecular
event, such as isomerization. Compound 4 appears to
be stable in its solid form when kept in the dark at room
temperature. However, under somewhat elevated tem-
perature, even the solid underwent transformation.
A scrutiny of several possible mechanisms led us to
the structure of (Z,E)-butadiene 4 shown in Figure 2 as
the original product, which slowly isomerizes into the
1
(E,E)-butadiene 5 (Figure 2). The results of 1D H NMR
(Figure 1) and 2D COSY experiments7 support the basic
configuration of the product (4) as a butadiene. The
important connectivities between neighboring protons of
1 and 2, 3, and 4 are clearly evidenced in the COSY
spectrum,7 and the same spectrum also shows that these
connectivities are preserved in the isomer (5), as indi-
cated by cross-peaks of 1′-2′ and 3′-4′. The coupling
constant between protons 3 and 4 was found to be 16 Hz,
leading to the assignment of a trans (E) configuration
for the second double bond. The ROESY experiments,7
on the other hand, provided crucial through-space cou-
pling data, rendering possible the unambiguous assign-
ment of the configurations and conformations for both 4
and its isomer, 5. Some key ROESY cross-peaks for
1
from the crude material and its H NMR (Figure 1, top)
showed a distinctively different pattern from that of the
expected one (3), and elemental analysis revealed a
composition of C28H21N3O6 with no phosphorus present,
in agreement with the mass spectrometric analysis (CI
mode) which showed an [M + 1]+ peak at m/ e 496. A
† Current address: Merck & Co., Inc., WP38-3, P.O. Box 4, West
Point, PA 19486.
(1) Fastrez, J .; J espers, L.; Lison, D.; Renard, M.; Sonveaux, E.
Tetrahedron Lett. 1989, 30, 6861-6864.
(2) Cheng, L.; Goodwin, C. A. Tetrahedron Lett. 1991, 32, 7333-
7336.
(3) Wang, C.-L. J .; Taylor, T. L.; Mical, A. J .; Spitz, S.; Reilly, T. M.
Tetrahedron Lett. 1992, 33, 7667-7670.
(4) Oleksyszyn, J .; Boduszek, B.; Kam, C.-M.; Powers, J . C. J . Med.
Chem. 1994, 37, 226-231.
(5) Chen, R.-Y.; Dan, S.-C. Phosphorus, Sulfur, Silicon 1991, 56,
39-48.
(7) The 2D COSY and ROESY experimental parameters and results
(Figures 3-5) are included in the Supporting Information for this
paper. However, some key ROESY cross-peaks observed in a mixture
of 4 and 5 are summarized in Table 1.
(6) Lethbridge, A.; Norman, R. O. C.; Thomas, C. B. J . Chem. Soc.,
Perkin 1 1973, 35-38.
S0022-3263(98)00242-4 CCC: $15.00 © 1998 American Chemical Society
Published on Web 09/30/1998

Notes
J . Org. Chem., Vol. 63, No. 21, 1998 7517
F igu r e 1. 1H NMR spectra of 4 in DMSO-d6 (Varian XL-300): top, immediately after dissolution (for assignment of protons, see
Figure 2); bottom, after the solution stood at room temperature for approximately 4 months.
on the basis of the coupling constant. Last, as compound
4 isomerized to 5, cross-peaks 1′-3′ and 2′-11′ occurred
while neither cross-peak 1′-11′ nor 2′-3′ was observed,
consistent with the proposed transformation. In both
isomers, cross-peaks 4-11 and 4′-11′ remained. It is
interesting to note that both 4 and 5 are locked in the
s-Z conformation.
The first striking feature of the proposed mechanism
(Scheme 2) is the role played by the naphthylmethyl
group that diverted the reaction course leading to the
predominant formation of an unexpected new structure
1
[a reexamination of the H NMR spectrum of the crude
product of 2 revealed that the corresponding benzyl-
butadiene compound 7 was also formed, but as a minor
product, with a relative 13% yield]. The importance of
the naphthalene ring may be related to the enamine form
of intermediate 8. Presumably, the naphthalene ring
facilitates the formation of the enamine 8a , a nucleophile,
possibly by an unconventional hydrogen-bonding interac-
tion8 involving its π orbital and the amide proton. Such
an interaction may be more favored with the naphthalene
ring than with the benzene ring, since the more extended
π orbital of the naphthalene ring may have a stronger
interaction with the amide proton. The second feature
F igu r e 2. Structures of compound 4 and 5 and proton
assignments.
is the formation of an electrophile 9 through the putative
enolization of (4-nitrophenyl)acetaldehyde; such enoliza-
tion has been suggested before by the present authors to
have played a central role in the oxidative degradation
of 4-nitrophenethyl alcohol by the Collins reagent.9 In
the formation of both the enamine 8a and enol 9, the key
nucleophile and electrophile in the proposed mechanism,
the nitro group plays a unique and critical role, probably
by inducing and/or stabilizing the more conjugated
enamine and enol forms through its remarkable reso-
Ta ble 1. Su m m a r y of Key ROESY Cr oss-P ea k s
compd 4
compd 5
1-11
2-3
4-11
1′-3′
2′-11′
4′-11′
compounds 4 and 5, respectively, are summarized in
Table 1. First of all, the presence of the cross-peak 1-11
in the ROESY spectrum established a Z configuration
for the first double bond. Second, the overall conforma-
tion of the product 4 is determined to be s-Z by the
occurrence of cross-peaks 2-3 and 4-11. The observa-
tion of these two cross-peaks also confirmed the previous
assignment of an E configuration for the first double bond
(8) Unconventional hydrogen bonding with a benzene ring acting
as hydrogen bond acceptor was observed experimentally,8a and theo-
retical calculation indicated that it was about half as strong as normal
hydrogen bonding.8b (a) Mitchell, J . B.; Nandi, C. L.; McDonald, I. K.;
Thornton, J . M.; Price, S. L. J . Mol. Biol. 1994, 239, 315-331. (b) Levitt,
M.; Perutz, M. F. J . Mol. Biol. 1988, 201, 751-754.
(9) Li, M.; J ohnson, M. E. Synth. Commun. 1995, 25, 533-537.

7518 J . Org. Chem., Vol. 63, No. 21, 1998
Notes
Sch em e 2
Sch em e 3
Varian Associates XL-300 NMR spectrometer. COSY and
ROESY experiments were conducted on a GE 500 MHz NMR
spectrometer. Mass spectra were obtained on a Finnigan MAT-
90 mass spectrometer. Elemental analysis was carried out by
Midwest Microlab, Indianapolis, IN. Thin-layer chromatography
(TLC) plates were obtained from E. M. Science.
nance effect. The function of dichlorophenylphosphine
is apparently to provide a leaving group for the hydroxyl,
resulting in the formation of a highly conjugated product
4; compound 4 could not be formed in the absence of the
former.
As for the subsequent isomerization, it appears that
the remarkable resonance effect of the nitro group once
again comes into play as illustrated in Scheme 3. The
first double bond of 4 should have significant single bond
character because of the contribution from the two
resonance forms. Therefore, isomerization may proceed
through the rotation of the first “double bond” of the
molecule (Scheme 3). The underlying driving force for
the isomerization has to be the energy difference between
the two isomers. However, it is not clear why the E,E
isomer would be more stable than the initial Z,E isomer.
2-Na p h th ylm eth yl Ca r ba m a te. The original procedure for
the preparation of tert-butyl carbamate10 was modified to afford
the title compound. Trifluoroacetic acid (1.56 mL, 21 mmol) was
added dropwise to a mixture of 2-naphthylmethanol (1.58 g, 10.0
mmol) and sodium cyanate (1.30 g, 20 mmol) in 12.5 mL of
benzene at room temperature. After being stirred at room
temperature overnight, the reaction mixture was combined with
100 mL of CH2Cl2 and the resulting mixture was washed with
2 N NaOH (3×). The aqueous phase was extracted with CH2-
Cl2 (4×), and the organic phases were combined and washed
again with water (3×) and brine (1×). The washed CH2Cl2
solution was dried and concentrated to about 17 mL. White
plates formed were collected by filtration (1.30 g, 65%). TLC:
Exp er im en ta l Section
Gen er a l Meth od s. All commercial chemicals were pur-
(10) Loev, B.; Kormendy, M. F.; Goodman, M. M. Organic Syntheses;
Wiley and Sons: New York, 1973; Collect. Vol. V, pp 162-166.
chased from Aldrich. 1H NMR spectra were acquired on a

Notes
J . Org. Chem., Vol. 63, No. 21, 1998 7519
Rf 0.50 (silica; ethyl acetate/hexane, 1/1, v/v). 1H NMR
(CDCl3): δ 7.9-7.8 (4 H, m), 7.6-7.45 (3 H, m), 5.26 (2 H, s),
4.80 (2 H, s, broad).
(1 H, d, J ) 11.2 Hz), 5.89 (1 H, d, J ) 15.8 Hz), 5.32 (2 H, s).
MS (CI): [M + 1]+, m/ e 496. Anal. Calcd for C28H21N3O6: C,
67.87; H, 4.27; N, 8.48. Found: C, 67.92; H, 4.41; N, 8.31.
(4-Nitr op h en yl)a ceta ld eh yd e. This compound was pre-
pared according to the method of Lethbridge et al.6 from the
oxidation, by lead tetraacetate, of 4-nitrostyrene which was in
turn synthesized from 2-(4-nitrophenyl)ethyl bromide.11 The
title compound was also prepared in lower yield from 4-nitroben-
zaldehyde by adapting a method for the synthesis of (4-
cyanophenyl)acetaldehyde.4 1H NMR (CDCl3): δ 9.83 (1 H, t, J
) 1.7 Hz), 8.23 (2 H, d, J ) 8.8 Hz), 7.41 (2 H, d, J ) 8.8 Hz),
3.89 (2 H, d, J ) 1.7 Hz).
(Z,E)-1-(2-Na p h th ylm eth yloxyca r bon yla m in o)-2,4-bis(4-
n it r op h en yl)b u t a d ien e. A reaction mixture of 2-naphthyl-
methyl carbamate (560 mg, 2.78 mmol), (4-nitrophenyl)acetal-
dehyde (476 mg, 2.88 mmol), and dichlorophenylphosphine (373
mol) in 2.5 mL of acetic acid was stirred at room temperature
overnight, and 7 mL of water was added. The resulting mixture
was heated at 100 °C for ∼3 min, and the supernatant was
decanted. The precipitate was washed with ether (2×) and
acetone/ether (1/5) to give 110 mg (15%, based on 4-nitrophen-
ylaldehyde) of a yellow powder. 1H NMR (DMSO-d6): δ 9.48 (1
H, d, J ) 11.2 Hz), 8.30 (2 H, d, J ) 8.7 Hz), 8.10 (2 H, d, J )
8.8 Hz), 7.91 (4 H, m), 7.64 (1 H, d, J ) 15.8 Hz), 7.63 (2 H, d,
J ) 8.8 Hz), 7.54-7.51 (3 H, m), 7.48 (2 H, d, J ) 8.7 Hz), 7.35
In another run (in which the procedure of Chen and Dan5 was
followed), after water was added, the resulting mixture was not
heated. Instead, the mixture was filtered and the precipitate
collected was washed only with ether. The 1H NMR spectrum
indicated that the dried precipitate contained approximately 80%
(mol/mol) of compound 4 and 20% of compound 3.
Ack n ow led gm en t. This work was partially sup-
ported by grants from the American Heart Association
of Metropolitan Chicago and the National Institutes of
Health. We thank Dr. Shantha D. Samarasinghe for
assistance with some of the COSY and ROESY experi-
ments.
Su p p or tin g In for m a tion Ava ila ble: Text describing 2D
NMR experiments and COSY and ROESY spectra (4 pages).
This material is contained in libraries on microfiche, im-
mediately follows this article in the microfilm version of the
article, and can be ordered from the ACS; see any current
masthead page for ordering information.
(11) Strassburg, R. W.; Gregg, R. A.; Walling, C. J . Am. Chem. Soc.
1947, 69, 2141-2143.
J O9802426