Atropisomeric transition state analogs

Article abstract of DOI:10.1002/1099-0690(200004)2000:7<1365::AID-EJOC1365>3.0.CO;2-X

Transition state mimicry is one of the most powerful concepts in enzyme inhibitor design and has led to the development of catalytic antibodies. Transition state analogs are compounds with a fixed shape that resemble the geometry and charge distribution of the transition state of a given reaction. Stabilization of a transition state like conformation is most often achieved by incorporating a ring system into the analog. We show herein that atropisomerism can be used as a new principle for enforcing a transition state like conformation. Atropisomerism relates to the existence of stereoisomers of structurally constrained molecules due to a frozen rotation about a single bond, as for example in binaphthol. The 1- aminomethylnaphthalene derivative 1 exhibits atropisomerism due to a frozen rotation about the C(1)-C(methylene) single bond, which holds the dihedral angle θ[C(2)-C(1)-C(methylene)N] close to 90°. Compound 1 mimics the transition state for hydride transfer between 1,4-dihydroquinolines 4 and acetone.

Full text of DOI:10.1002/1099-0690(200004)2000:7<1365::AID-EJOC1365>3.0.CO;2-X

FULL PAPER
Atropisomeric Transition State Analogs
Olaf Ritzeler,[a][°] Serge Parel,[a] Bruno Therrien,[a] Nicolas Bensel,[a] Jean-Louis Reymond,*[a]
and Kurt Schenk[b]
Keywords: Atropisomerism / Transition states / Enzyme inhibitors
Transition state mimicry is one of the most powerful concepts
in enzyme inhibitor design and has led to the development of
isomerism relates to the existence of stereoisomers of struc-
turally constrained molecules due to a frozen rotation about
catalytic antibodies. Transition state analogs are compounds a single bond, as for example in binaphthol. The 1-aminome-
with a fixed shape that resemble the geometry and charge
distribution of the transition state of a given reaction. Stabil-
thylnaphthalene derivative 1 exhibits atropisomerism due to
a frozen rotation about the C(1)–C(methylene) single bond,
ization of a transition state like conformation is most often which holds the dihedral angle θ[C(2)–C(1)–C(methylene)–
achieved by incorporating a ring system into the analog. We
show herein that atropisomerism can be used as a new prin-
N] close to 90°. Compound 1 mimics the transition state for
hydride transfer between 1,4-dihydroquinolines 4 and acet-
ciple for enforcing a transition state like conformation. Atrop- one.
Introduction
transition state analog. Thus, five- and six-membered rings
have been used to generate transition state analogs for the
rearrangement of peptide bonds,[8] epoxy alcohol[9] and car-
bocationic cyclizations,[10] for pericyclic reactions such as
oxy-Cope[11] rearrangements, selenoxide[12] and Cope[13]
eliminations, and for alkene[14] and bis(aryl)[15] isomeriz-
ation reactions. Similarly, bicyclic compounds have been
used to mimic transition state conformations for Claisen
rearrangements,[16] Diels–Alder cycloadditions,[17] a disfav-
ored Z-selective β-fluoro-elimination,[18] and for glycosidic
bond cleavage.[19] An α-keto-amide has been used to mimic
the transition state for cis–trans isomerization in prolyl-pep-
tide bonds.[20]
We report herein on a new principle for enforcing a trans-
ition state like conformation in a stable analog, namely the
use of atropisomerism. Atropisomerism relates to the exist-
ence of stereoisomers of structurally constrained molecules
due to a frozen rotation about a single bond, as for example
in binaphthol.[21] Atropisomerism has provided the basis for
the design of a number of highly enantioselective reagents
and catalysts.[22] We show here that atropisomerism allows
the design of transition state analogs for hydride transfer
between a 1,4-dihydroquinoline and a methyl ketone. The
transition state of this reaction is accurately mimicked using
N-oxide model compounds.
As first formulated by Pauling,[1] chemical catalysis may
be described as tight binding between a catalyst and the
transition state of a reaction.[2] This simple concept led to
the development of transition state mimicry as a powerful
paradigm in enzyme inhibitor design, as well as to the de-
velopment of catalytic antibodies, whereby stable transition
state analogs of chemical reactions are used in the creation
of new catalytic activity through immunization.[3,4]
A transition state is defined as a saddle point on the
Born–Oppenheimer hypersurface of a molecular system.[5]
Transition state analogs are compounds with a fixed shape
that resemble the geometry and charge distribution of a
given transition state.[6] The construction of such stable an-
alogs is generally difficult because transition state bond
lengths and geometries are inherently unstable owing to
their transient nature. Indeed, wherever possible one
chooses to mimic not the transition state itself but a high
energy intermediate along the reaction pathway, if such an
intermediate exists. This approach includes the use of phos-
phonates as stable analogs of the tetrahedral intermediate
in ester hydrolysis, and of polyhydroxylated five- and six-
membered ring compounds as analogs of the oxocarbonium
cation intermediate in glycosidic bond hydrolysis.[7]
Construction of a transition state analog almost always
requires the fixing of a reactive conformation and/or a relat-
ive arrangement of reactants corresponding to the trans-
ition state. The strategy most often used to represent such
conformations has been to incorporate ring systems in the
Results and Discussion
[a]
Department of Chemistry and Biochemistry, University of
Alcohol dehydrogenases are extremely important en-
zymes that catalyze reversible hydride transfer between the
4-position of a 1,4-dihydronicotinamide of an NADH (re-
duced nicotinamide adenine dinucleotide) or NADPH (re-
duced nicotinamide adenine dinucleotide phosphate) cofac-
tor and the carbonyl group of an aldehyde or ketone
Bern,
Freiestrasse 3, CH-3012 Bern, Switzerland
[b]
´
Laboratoire de Cristallographie, BSP-Dorigny, Universite de
Lausanne,
CH-1015 Lausanne, Switzerland
Present address: Hoechst Marion Roussel Deutschland GmbH,
D-65926 Frankfurt, Germany
[°]
Eur. J. Org. Chem. 2000, 1365Ϫ1372
WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2000
1434Ϫ193X/00/0407Ϫ1365 $ 17.50ϩ.50/0
1365

O. Ritzeler, S. Parel, B. Therrien, N. Bensel, J.-L. Reymond, K. Schenk
FULL PAPER
˚
state C(4)–C(carbonyl) distance of approximately 2.7 A re-
ported for the calculated transition state[26,27] is accurately
˚
mimicked by a 2.59 A separation between C(2) and the ni-
trogen atom of the N-oxide in this arrangement. That the
dihydropyridine in the chemical reaction may be replaced
with an aromatic ring in transition state analog A is sug-
gested by the fact that dihydropyridines are essentially
planar and are isosteric with a benzene ring in their ground
state.[29] This replacement was also suggested by the fact
that alcohol dehydrogenases are known to stabilize the
neutral, reduced form of the NAD cofactor within their
active site during hydride transfer.[30]
Due to the absence of cyclic constraints, the relative posi-
tion of the benzene ring and the N-oxide function in trans-
ition state analog A depends on the values of the dihedral
angles θ [C(1)–C(2)–C(methylene)–N] and φ [C(2)–C(me-
thylene)–N–O], which are two freely rotating axes. The de-
sired transition state geometry can only be approached if
the dihedral angle θ [C(1)–C(2)–C(methylene)–N] between
the benzylic bond and the plane of the aromatic ring is close
to 90°. Molecular modelling suggests that the energetically
most favorable conformations for A do indeed have such a
Scheme 1. Transition state and analogs for the alcohol dehydro- value of θ. Although A adopts satisfactory transition state
genase reaction
like conformations in modelling experiments, the rotation
about θ is almost unhindered.
A greater degree of conformational control, leading to
the desired geometries, can be enforced by restricting ac-
(Scheme 1).[23] The origin of this catalysis and the geometry
of the transition state have been the subject of intense sci-
cessible values of the dihedral angle θ by introducing atropi-
entific debate in recent years. We have therefore become in-
somerism. Due to the known atropisomerism of naph-
terested in preparing an alcohol dehydrogenase catalytic an-
thalene derivatives, the 2-naphthoic acid derivative 1, as
well as its close relative 2 and 3, can be expected to exist as
separate atropisomers with the dihedral angle θ [C(2)–C(1)–
C(methylene)–N] held close to 90°. These compounds ad-
tibody. This requires the design of a stable transition state
analog for this reaction.[24] The key role played by the relat-
ive arrangement of reactants at the transition state is sug-
gested by the observation that the catalytic ability of alco-
equately mimic the transition state for hydride transfer be-
hol dehydrogenases is highly sensitive to the correct align-
tween acetone and dihydroquinoline cofactors 4–6
ment of the reactants within their active sites.[25] Theoretical
(Scheme 2).[31] The N-oxide group might adopt a variety of
calculations on the transition states for the gas-phase[26] and
orientations by rotation about φ [C(2)–C(methylene)–N–O].
In the transition state, a similar degree of rotational free-
dom exists for rotation about the H–C(carbonyl) axis (φЈ)
for the enzymatic reaction[27] show that the carbonyl group
being reduced lies above the dihydropyridine ring at approx-
˚
imately 2.7 A from C(4) (Scheme 1). Interaction of the car-
and calculations suggest that, in enzyme active sites, this
bonyl oxygen atom with either a Zn(II) or a protonated
parameter is dictated by the interaction with the activating
histidine (BH in Scheme 1) is crucial for transition state sta-
group Zn(II) or BH.[27]
bilization in the enzymatic reaction.
Transition State Analog Design
The N-oxide group has been used as a transition state
analog of a polarized carbonyl moiety in raising antibodies
capable of catalyzing an enantioselective reduction of ke-
tones with cyanoborohydride,[28] a reaction akin to the hy-
dride transfer process with dihydropyridines. We reasoned
that transition state analogs for the hydride transfer process
should consist of an N-oxide held at an appropriate dis-
tance above a mimic of the dihydropyridine ring. A trans-
ition state analog for the reduction of acetone by a dihy-
dronicotinamide can be formulated as an N,N-dimethyl N-
oxide bound as benzylic substituents to ortho-toluamide,
the latter serving as a dihydronicotinamide mimic, giving
the hypothetical transition state analog A. The transition inolines and acetone and atropisomeric transition state analogs
Scheme 2. Transition state of hydride transfer between dihydroqu-
1366
Eur. J. Org. Chem. 2000, 1365Ϫ1372

Atropisomeric Transition State Analogs
FULL PAPER
Synthesis
hydride transfer between acetone and dihydroquinoline co-
factor 5 (Figure 1). Distance and angular data reported for
the formate dehydrogenase reaction[27c] were used to place
the carbonyl group of the substrate at the appropriate dis-
tance and angle in relation to the quinoline ring.
Naphthoic ester 7 was prepared from acetophenone and
diethyl succinate using known procedures and then bromin-
ated to give bromide 8 (Scheme 3).[32]
Figure 1. Model for transfer hydride transition state between dihy-
droquinoline 5 and acetone (left) and structure of analog 2 accord-
ing to X-ray structure data (right); tube models with N(1)–C(4)–
C(carbonyl) resp. C(4)–C(1)–N angles in degrees and the C(4)–
˚
C(carbonyl) and C(1)–N distance in A
As discussed above, the value of the dihedral angle φЈ for
rotation about the H–C(carbonyl) axis in the transition
state of the enzymatic hydride transfer process (Scheme 1)
is dictated by interaction with a BH residue on the enzyme.
Similarly, in a catalytic antibody raised against a conformer
of our transition state analog with a given angle φ [C(2)–
C(methylene)–N–O], the hydride transfer transition state
would be forced to adopt the corresponding value of the
dihedral angle φЈ, probably through interaction with a BH
residue on the antibody that would have been induced by
the N-oxide function of the transition state analog in the
course of immunization. To achieve such a situation in our
comparison, the carbonyl group in the transition state can
be oriented about φ to fit the observed orientation of the
N-oxide in transition state analog 2. Finally, the carboxyl
group of the naphthalene cofactor 5 is oriented with a 20°
out-of-plane tilt to fit the accepted model for hydride trans-
fers with such cofactors.[27]
Scheme 3. Synthesis of intermediate 8[32]
Selective aminolysis of bromide 8 with dimethylamine
gave 9. Aminolysis of 9 with ammonia or with further di-
methylamine gave 10 and 11, respectively. Oxidation of 9,
10, and 11 with m-CPBA gave 2, 1, and 3 (Scheme 4).
Figure 2. Electron density surface colored with electrostatic poten-
tial for model transition state (left) and transition state analog 2
(right); surface coloration goes from red (most negative) to blue
(most positive), set at the same values for both structures; the sur-
face was calculated at semiempirical level using the PM3 model
as implemented in SPARTAN v5 from Wavefunction, Inc. Irvine
(CA), USA
Scheme 4. Synthesis of model transition state analogs
Structure and Atropisomerism
The similarity between the transition state model and the
Compound 2 was crystallized and its structure was deter- analog can be appreciated by inspection of Figure 1 and 2.
˚
mined by X-ray crystallography. As expected, the benzylic The carbonyl group of the substrate, placed at 2.50 A from
N-oxide substituent lies above the plane of the naphthalene C(4) of dihydropyridine 5 at an angle of 116°, is mimicked
˚
ring, with θ [C(2)–C(1)–C(methylene)–N] ϭ 75°, showing a by the N-oxide group in 2, which is located at 2.58 A from
slight tilt towards the carbomethoxy group. For comparison C(1) at an angle of 146°. The electrostatic potential surfaces
purposes, we generated a model of the transition state for compare well; in particular, the N-oxide group is an excel-
Eur. J. Org. Chem. 2000, 1365Ϫ1372
1367

O. Ritzeler, S. Parel, B. Therrien, N. Bensel, J.-L. Reymond, K. Schenk
FULL PAPER
lent mimic for the polarized carbonyl group undergoing re- about the C(1)–C(methylene) bond were to be possible; its
duction. A noticeable discrepancy between the transition splitting into an AB system indicates that this rotation is
state and the analog concerns the orientation of the carb- frozen at lower temperatures. Signal splitting is observed for
oxyl substituent of the dihydroquinoline. It has been shown the methyl substituents of the N-oxide, which would also be
for dihydronicotinamide that at the transition state the car- expected on freezing the rotation about the C(1)–C(methy-
boxamide group is tilted out of the plane of the molecule lene) bond. The activation energy for this rotation can be
by approximately 20°, with the CϭO moiety pointing to- calculated from the NMR data (Table 1 and Figure 4).[37]
wards the carbonyl compound being reduced. The same ef- These results show that enantiomers of 1–3 exist as separate
1
fect is believed to govern stereoselectivity of hydride trans- atropisomers on the NMR time scale. In contrast, the H-
fer with chiral dihydropyridines[33,34] and dihydroquinol- and 13C-NMR signals for 12 and its methyl ester 9 do not
ines,[35] and also occurs with other substituents at the carb- show any splitting, even at low temperatures, suggesting
oxyl groups. This 20° out-of-plane tilt is not correctly that in these compounds the dimethylamino group rotates
mimicked by analog 2, where the carboxymethyl substituent more rapidly than the observable NMR time scale.
is tilted out of the plane by –40°, away from the reduction
Table 1. Atropisomerism of compounds 1–3 as measured by vari-
able-temperature 1H-NMR; spectra were recorded in CD3OD at
10 K intervals between 223 K and 323 K
direction. A systematic conformational search by molecular
modelling showed that this unfavorable orientation is in-
deed the most stable orientation and not an effect of crystal
packing. The same effect is predicted for compounds 1 and
3. Interestingly, an intramolecular H-bond, leading to a
correct orientation of the carboxyl group with regard to
transition state mimicry, is observed in the crystal structure
of amino acid 12 (Figure 3), which was obtained in crystal-
line form following hydrolysis of 9.
Compound Signal
Tc (°C)[a] ∆G (kJmol–1)[b] kcoal. (s–1)[c]
1
2
3
N(CH3)2
30
61
63
56
54
59
62
184
101
431
324
314
674
Ar–CH2–N 40
N(CH3)2
20
Ar–CH2–N 20
N(CH3)2 25
Ar–CH2–N 50
[a]
Temperature by which coalescence of 1H NMR signals is ob-
served. – [b] Calculated activation energy for equilibration of atropi-
somers: (G∆ ϭ RTc(22.96 ϩ ln(Tc/δγ)) (methyl signals) and (G∆ ϭ
[c]
RTc(23.76 – ln(kcoal./Tc)) (methylene signals). – Rate of equlibr-
ation at coalescence temperature. kcoal. ϭ 2.22δγ (methyl signals)
and kcoal. ϭ π(δγ2 ϩ 6J2AB)/1.414 (methylene signals).
Figure 4. NMR signal splittings; 1H-NMR methylene signals for
compound 1 (left) and 13C-NMR methylene signal for compound
2 (right) at representative temperatures
The height of the rotational barrier evaluated from the
NMR data is approximately 50–60 kJmol–1 for all three
compounds. The nature of the carboxyl substituent on the
naphthalene ring has relatively little influence on the rate
Figure 3. Amino acid 12 and X-ray strucure; the carboxyl group is
tilted by 28.2° relative to the aromatic plane due to the intra-
molecular H bond
In order to determine whether enantiomeric forms of 1– of equilibration between atropisomers, which suggests that
3 actually exist as atropisomers, their conformational isomerization might occur by rotation on the side of the
stabilities in solution were investigated by NMR spectro- aromatic group (θ ϭ 180°) rather than on the side of the
scopy.[36] 1H- and 13C-NMR spectra were recorded in meth- carboxyl group (θ ϭ 0°). To test these hypotheses, we car-
anol solution at various temperatures. In all cases, charac- ried out semiempirical calculations on 1–3. In all three
teristic line broadening and signal splitting was observed cases, rotation about the C(1)–C(methylene) bond was pre-
for the benzylic methylene proton resonances. This signal dicted to have a lower activation energy for rotation on the
would only appear as a singlet at high temperature if free side of the aromatic group, amounting to approximately 56
rotation of the methylene group and its N-oxide substituent kJmol–1 for 1, 50 kJmol–1 for 2, and 45 kJmol–1 for 3 (Fig-
1368
Eur. J. Org. Chem. 2000, 1365Ϫ1372

Atropisomeric Transition State Analogs
FULL PAPER
be linked to a carrier protein by means of an alkyl chain
attached to one of the methyl substituents of the amine ox-
ide function, thereby mimicking the transition state of the
reaction of an alkyl methyl ketone.
Conclusion
We have shown that atropisomerism can be exploited in
the design of conformationally stable transition state anal-
ogs. In the example presented here, restricted rotation al-
lows the positioning of an N-oxide above a naphthalene
ring to serve as a mimic for a polarized carbonyl group in
a hydride transfer process. The almost exact match of the
˚
C(4)–C(carbonyl) distance of 2.7 A in the transition state by
the C(1)–N distance of 2.59 A in our analogs is noteworthy.
Indeed, it is often almost impossible to design analogs dis-
˚
Figure 5. Simulated energy profile for atropisomerization of com-
pound 2; the calculated heats of formation are plotted against the
constrained dihedral angle θ ϭ C(2)–C(1)–C(methylene)–N; the
starting structure was constructed using the X-ray coordinates of playing accurate transition state distances. As discussed
molecule 2, followed by geometry optimization; from the obtained,
energy-minimized structures, the rotation barrier around the C(1)–
above, the fact that atropisomers may not be separable due
to rapid equilibration is probably not a handicap in the con-
text of transition state analogs. The rotational barriers of
approximately 15 kcal/mol measured here are sufficient to
ensure that only the more stable conformers displaying the
correct transition state geometry will be significantly popu-
lated in solution and thus recognized in a non-covalent
C(methylene) bond was analysed using the coordinate driving pro-
cedure of SPARTAN: the dihedral angle was constrained in 10°
increments until a complete rotation was achieved, both clockwise
and counterclockwise; each constrained structure was fully energy-
minimized before incrementing the dihedral angle; similar proced-
ures were carried out for compounds 1 and 3 (data not shown); all
calculations were performed at the semiempirical level using the
PM3 model as implemented in SPARTAN v5 from Wavefunction,
Inc., Irvine (CA), USA; the geometry of each structure was pre- binding interaction with a target protein. This design
optimized using the SYBYL forcefield
should be applicable to processes involving similar trans-
ition state geometries, such as the aldol reaction[38] and aro-
matic substitution reactions.
ure 5). It must be pointed out that the starting conforma-
tion in these calculations is chiral, so that clockwise and
counterclockwise rotations are diastereoisomeric and not
Experimental Section
superimposable. These calculations nicely support our
NMR measurements.
These measurements demonstrate that, in spite of rapid
equilibration, only atropisomeric conformers with the N-
General: Reagents were purchased from Aldrich or Fluka. – All
chromatography (flash) was performed on Merck silica gel 60
(0.040–0.063 mm). – Preparative HPLC was carried out with Fisher
Optima grade acetonitrile and ordinary deionized water using a
oxide placed above the naphthalene ring, corresponding to
Waters prepak cartridge 500 g installed on a Waters Prep LC 4000
the geometry of the hydride transfer process to be mim-
icked, are significantly populated in solution. The binding
energies associated with non-covalent interactions of trans-
ition state analogs with their typical target proteins, either
an enzyme if they are used as inhibitors or an antibody if
they are used as immunogenic haptens in the context of an
immunization experiment, are typically of the order of 10–
20 kcal/mol. Within this energy range, the energetic handi-
cap associated with recognition of a conformer less stable
than the most stable conformer by only a few kcal/mol
would be prohibitive. Here, recognition of unstable ‘‘in
plane’’ conformers of compounds 1–3 would be associated
with a 15 kcal/mol handicap for binding. In the context of
immunization experiments, the most stable conformers are
also the most populated, and will thus be most frequently
encountered by the immune system. Therefore, we believe
that our design principle may be exploited in the context of
transition state analogs despite the fact that the atropisom-
eric forms of our compounds are not separable. In the con-
system from Millipore, flow rate 100 mL/min., gradient ϩ 0.5%/
min CH3CN; detection by UV at 254 nm. – TLC was performed
on fluorescent F254 glass-backed plates. – NMR spectra were re-
corded with Bruker AM-250 (250 MHz) or AM-300 (300 MHz) in-
struments. – IR spectra were recorded with a Perkin–Elmer 1600
Series FT-IR instrument. – Melting points were determined with a
Büchi 510 melting point apparatus. – MS and HRMS data were
provided by Analytical Research Services (Dr. T. Schürch, Univer-
sity of Bern) and by the Scripps Research Institute MS facility (Dr.
G. Siuzdak). The procedure for the synthesis of compound 7 was
carried out in accordance with the original literature[32] without
significant changes, as detailed below.
Ethyl 4-Acetoxy-1-methyl-2-naphthoate: 3-Ethyloxycarbonyl-4-phe-
nyl-3-pentenoic acid, obtained from acetophenone and diethyl suc-
cinate, was refluxed for 12 h in acetic anhydride (90 mL, 95 mmol)
containing sodium acetate (9.00 g, 100 mmol). The excess acetic
anhydride was then removed by co-evaporation with toluene and
the dark-brown residue was treated with iced water (300 mL) and
extracted with diethyl ether (3 ϫ 250 mL). The combined ethereal
extracts were washed with satd. aq. Na2CO3 solution (100 mL),
text of an immunization experiment, the molecules could dried (Na2SO4), and the solvent was evaporated. Chromatography
Eur. J. Org. Chem. 2000, 1365Ϫ1372 1369

O. Ritzeler, S. Parel, B. Therrien, N. Bensel, J.-L. Reymond, K. Schenk
FULL PAPER
(hexane/ethyl acetate, 10:1) of the residue gave ethyl 4-acetoxy-1-
methyl-2-naphthoate (8.3 g, 63%) as a colorless solid. TLC (hexane/
ethyl acetate, 10:3): Rf (starting materials) ϭ 0.43 and 0.37, Rf
NEt3; 2. hexane/ethyl acetate, 10:2): Rf ϭ 0.33. – 1H NMR
(250 MHz, CDCl3): δ ϭ 8.38–8.22 (m, 2 H), 7.62–7.48 (m, 2 H),
6.97 (s, 1 H), 4.02 (s, 1 H), 3.97 (s, 3 H), 3.91 (s, 3 H), 2.25 (s, 6 H). –
(product) ϭ 0.57. – 1H NMR (250 MHz, CDCl3): δ ϭ 8.22 (m, 13C NMR (65.5 MHz, CDCl3): δ ϭ 170.1, 154.5, 133.4, 130.2,
1 H), 7.85 (m, 1 H), 7.68–7.53 (m, 3 H), 4.42 (q, J ϭ 7.1 Hz), 2.91 127.8, 127.1, 126.7, 126.3, 125.4, 122.1, 102.8, 55.9, 55.5, 52.1, 45.1,
(s, 3 H), 2.49 (s, 3 H), 1.43 (t, J ϭ 7.1 Hz). – 13C NMR (65.5 MHz,
29.5. – IR (CHCl3): ν˜ ϭ 2948, 2674, 2490, 1472, 1398, 1036
CDCl3): δ ϭ 169.4, 167.0, 145.2, 135.5, 133.9, 127.8, 126.9, 125.6, cm–1. – FAB-MS: m/z ϭ 296 [M ϩ Naϩ], 274 [M ϩ Hϩ].
121.4, 118.2, 61.1, 20.9, 15.6, 14.2. – HRMS: C16H16NO4ϩ [M ϩ
Hϩ]: calcd. 272.1049; found 272.1042.
1-(Dimethylamino)methyl-4-methoxy-2-naphthoamide (10): A solu-
tion of compound 9 (140 mg, 0.51 mmol) in 17 mL of NH3/MeOH
(8.3 ) was stirred at 100 °C for 3 d. Evaporation of the solvent
and chromatography (CH2Cl2/CH3OH, 10:1) of the residue gave 10
4-Hydroxy-1-methyl-2-naphthoic Acid: Ethyl 4-acetoxy-1-methyl-2-
naphthoate (3.3 g, 12 mmol) was dissolved in aqueous KOH (30%,
30 mL) and methanol (30 mL) and the resulting mixture was heated (110 mg, 83%) as a colorless solid. – TLC (CH2Cl2/methanol, 10:2):
under reflux for 4 h. The mixture was then cooled to 20 °C, diluted
with water (100 mL), and washed with CH2Cl2 (2 ϫ 100 mL). The
aqueous phase was acidified with 6  HCl and extracted with di-
ethyl ether (2 ϫ 100 mL) to give 4-hydroxy-1-methyl-2-naphthoic
acid in quantitative yield (2.50 g) as a white crystalline solid. – TLC
(CHCl3/AcOH, 20:1): Rf (starting material) ϭ 0.70, Rf (product) ϭ
0.15. – 1H NMR (250 MHz, CD3OD): δ ϭ 8.08 (m, 1 H), 7.85 (m,
1 H), 7.37–7.22 (m, 2 H), 6.92 (s, 1 H), 4.90 (br. s, 1 H), 2.56 (s,
3 H). – 13C NMR (65.5 MHz, CD3OD): δ ϭ 172.8, 152.5, 135.2,
129.9, 127.9, 127.7, 127.0, 126.0, 123.5, 15.4. – HRMS: C12H10O3ϩ
[M ϩ Hϩ]: calcd. 202.0630; found 202.0626.
Rf (9) ϭ 0.47, Rf (10) ϭ 0.28. – 1H NMR (250 MHz, CDCl3/
CD3OD): δ ϭ 8.22–8.13 (m, 2 H), 7.51–7.42 (m, 2 H), 7.04 (s, 1 H),
4.06 (s, 2 H), 3.51 (s, 1 H), 2.68 (s, 1 H), 2.34 (s, 6 H). – 13C NMR
(65.5 MHz, CDCl3/CD3OD): δ ϭ 172.5, 155.1, 135.2, 133.1, 127.8,
127.1, 126.3, 126.0, 125.9, 123.9, 122.4, 103.1, 54.9, 54.0, 42.7. –
MS (electrosprayϩ): m/z ϭ 517 [2M ϩ Hϩ], 259 [M ϩ Hϩ].
N-[(2-Carbamoyl-4-methoxy-1-naphthyl)methyl]-N,N-dimethyl-
amine Oxide (1): Compound 10 (105 mg, 0.41 mmol), m-CPBA
(189 mg, 0.82 mmol), and Na2CO3 (200 mg) were suspended in
5 mL of CH2Cl2/MeOH, 1:1, and the mixture was stirred at 20 °C
for 2.5 h, after which the reaction was complete. The mixture was
then filtered and the residue was washed with CH2Cl2/MeOH, 4:1
(2 ϫ 10 mL). Evaporation of the solvents and chromatography
Methyl 4-Methoxy-1-methyl-2-naphthoate (7): A mixture of 4-hy-
droxy-1-methyl-2-naphthoic acid (1.12 g, 5.55 mmol), dry K2CO3
(6.88 g, 49.78 mmol), and dimethyl sulfate (4.47 g, 35.44 mmol) in (CH2Cl2/MeOH, 10:1) of the residue gave 95 mg (85%) of 1 as an
1
anhydrous acetone (41.4 mL) was heated for 4 h under reflux. The
mixture was then allowed to cool to room temperature, water
(130 mL) was added, and the acetone was removed in vacuo. The
remaining aqueous phase was then extracted with diethyl ether
(2 ϫ 200 mL), the combined extracts were dried over sodium sulf-
oily product. – H NMR (300 MHz, CD3OD): δ ϭ 7.92–7.80 (m,
2 H), 7.65–7.02 (m, 2 H), 7.22 (s, 1 H), 5.38 (br. s, 2 H), 4.16 (s,
3 H), 3.43–3.17 (br. s, 6 H). – 13C NMR (72.5 MHz, CD3OD): δ ϭ
174.5, 158.0, 140.6, 135.2, 128.8, 127.5, 127.3, 126.5, 123.4, 116.3,
104.9, 66.1, 59.4 (br.), 58.9 (br.), 56.3. – HRMS: C15H18N2O3ϩ [M
ate, and the solvent was evaporated to yield 7 as a colorless solid ϩ Hϩ]: calcd. 275.1396; found 275.1405.
(1.27 g, 100%). – TLC (hexane/ethyl acetate, 10:3): Rf ϭ 0.41. – 1H
N-[(4-Methoxy-2-methoxycarbonyl-1-naphthyl)methyl]-N,N-
NMR (250 MHz, CDCl3): δ ϭ 8.31 (m, 1 H), 8.11 (m, 1 H), 7.62–
7.50 (m, 2 H), 7.12 (s, 1 H), 4.05 (s, 3 H), 3.88 (s, 3 H). – 13C NMR
(65.5 MHz, CDCl3): δ ϭ 169.3, 153.3, 133.6, 129.0, 127.4, 127.0,
126.9, 126.6, 125.1, 122.2, 103.4, 55.4, 52.1, 15.3. – IR (CHCl3):
ν˜ ϭ 3410, 3018, 2962, 2646, 2476, 1636, 1472, 1398, 1038 cm–1. –
HRMS: C14H14O3ϩ [M ϩ Hϩ]: calcd. 230.0943; found 230.0928.
dimethylamine Oxide (2): A similar procedure as described above,
but using neat dichloromethane (6 mL) and starting with com-
pound 9 (130 mg, 0.4756 mmol), yielded 130 mg (94%) of 2 as a
colorless solid; m.p. 132–133 °C. – TLC (CH2Cl2/methanol, 10:2):
Rf (9) ϭ 0.28, Rf (2) ϭ 0.39. – 1H NMR (300 MHz, CD3OD): δ ϭ
8.32 (d, J ϭ 8.5 Hz, 1 H), 8.18 (d, J ϭ 8.5 Hz, 1 H), 7.56–7.43 (m,
2 H), 7.14 (s, 1 H), 5.20 (br. s, 2 H), 3.93 (s, 3 H), 3.85 (s, 3 H),
3.08–2.85 (br. s, 6 H). – 13C NMR (72.5 MHz, CD3OD): δ ϭ 173.2,
160.4, 138.5, 138.2, 131.7, 130.5, 130.4, 129.2, 126.0, 122.3, 107.7,
67.4, 61.2, 58.9, 55.7. – IR (CHCl3): ν˜ ϭ 3020, 2928, 2400, 1752,
1590, 1464, 1370, 1106 cm–1. – HRMS: C16H30NO4ϩ [M ϩ Hϩ]:
calcd. 290.139280; found 290.139233.
Methyl 1-(Bromomethyl)-4-methoxy-2-naphthoate (8): Methyl 4-me-
thoxy-1-methyl-2-naphthoate (7) (2.00 g, 8.69 mmol), N-bromosuc-
cinimide (1.85 g, 10.4 mmol), and azobis(isobutyronitrile) (20 mg)
were suspended in CCl4 (70 mL) and the mixture was heated to
reflux. The reaction was monitored by TLC (hexane/ethyl acetate,
10:3) and terminated once all the starting material (Rf ϭ 0.7) had
been consumed (105 min.). The mixture was then cooled to 20 °C
and the solids were filtered off and washed with CCl4 (2 ϫ 15 mL).
Concentration of the combined organic phases yielded 8 in quantit-
ative yield (2.70 g) as a slightly yellow solid. This bromide (Rf ϭ
N-{[2-(NЈ,NЈ-Dimethylcarbamoyl)-4-methoxy-1-naphthyl]methyl}-
N,N-dimethylamine Oxide (3): A solution of compound 9 (210 mg,
0.77 mmol) in 15 mL of 5.6  dimethylamine in methanol con-
taining sodium cyanide (50 mg, cat.) was stirred for 7 d at 100 °C.
Evaporation of the solvent, work-up (aq. NaHCO3/CH2Cl2), and
chromatography (hexane/CH2Cl2/acetone, 80:80:1) of the residue
gave 11 as a colorless solid (90 mg, 40%). – TLC (hexane/ethyl acet-
ate, 10:3): Rf (9) ϭ 0.22, Rf (11) ϭ 0.11. Amine 11 (37 mg,
0.1292 mmol) was further oxidized with m-CPBA in 5 mL of
1
0.6) was used for the next step without further purification. – H
NMR (250 MHz, CDCl3): δ ϭ 8.33–8.21 (m, 2 H), 7.69–7.57 (m,
2 H), 7.15 (s, 1 H), 5.39 (s, 2 H), 4.03 (s, 3 H), 3.88 (s, 3 H). – 13C
NMR (65.5 MHz, CDCl3): δ ϭ 155.7, 132.3, 127.8, 127.3, 124.7,
122.2, 103.9, 55.7, 52.6, 27.3.
Methyl 1-(Dimethylamino)methyl-4-methoxy-2-naphthoate (9): A CH2Cl2 as described above. After 1 h, work-up and subsequent
solution of bromide 8 (1.50 g, 4.85 mmol) in 80 mL of anhydrous
methanol saturated with dimethylamine (5.6 ) was stirred for 5 h
at 20 °C. Evaporation of the solvent and chromatography (hexane/
chromatography (CH2Cl2/MeOH, 10:1) and preparative reversed-
phase HPLC gave 35 mg (90%) of 3 as an oily product. – TLC
(CH2Cl2/methanol, 10:2): Rf (11) ϭ 0.55, Rf (3) ϭ 0.11. – 1H NMR
CH2Cl2/acetone, 80:80:1) of the residue yielded 9 as a colorless (300 MHz, CD3OD, 296 K): δ ϭ 8.43–8.33 (m, 2 H), 7.79–7.64 (m,
product (1.19 g, 90%). – TLC (1. hexane/ethyl acetate, 10:2 ϩ 0.5
2 H), 7.06 (s, 1 H), 5.66 (br. s, 2 H), 4.10 (s, 3 H), 3.53–3.29 (br. s,
1370
Eur. J. Org. Chem. 2000, 1365Ϫ1372

Atropisomeric Transition State Analogs
FULL PAPER
[11]
[12]
[13]
[14]
[15]
6 H), 3.23 (s, 3 H), 2.95 (s, 3 H). – 13C NMR (72.5 MHz, CD3OD,
296 K): δ ϭ 129.7, 128.9, 127.7, 127.2, 125.9, 125.8, 122.1, 59.2
(br.), 56.5 (br.), 55.2, 34.0. – IR (CHCl3): ν˜ ϭ 3452, 3010, 2421,
1684, 1628, 1594, 1464, 1416, 1200 cm–1. – MS (EI-MS): m/z ϭ
257 [M – 45], 241 [M – 61], 227 [M – 75], 212 [M – 90].
A. C. Braisted, P. G. Schultz, J. Am. Chem. Soc. 1994, 116,
2211–2212.
Z. S. Zhou, N. Jiang, D. Hilvert, J. Am. Chem. Soc. 1997, 119,
3623–3624.
S. S. Yoon, Y. Oei, E. Sweet, P. G. Schultz, J. Am. Chem. Soc.
1996, 118, 11686–11687.
D. Y. Jackson, P. G. Schultz, J. Am. Chem. Soc. 1991, 113,
2319–2320.
T. Uno, J. Ku, J. R. Prudent, A. Huang, P. G. Schultz, J. Am.
Chem. Soc. 1996, 118, 3811–3817.
1-[(Dimethylamino)methyl]-4-methoxy-2-naphthoic Acid (12): An
analytical sample of methyl ester 9 was dissolved in 2-propanol/
dichloromethane/aq. ammonia (10:10:1) for crystallization pur-
poses. Under these conditions, the methyl ester was cleaved and
crystals of compound 12 could be collected after 3 days at 20 °C;
m.p. 224–226 °C (dec.). – 1H NMR (250 MHz, CD3OD): δ ϭ 8.32–
8.28 (m, 2 H), 7.72–7.53 (m, 2 H), 7.39 (s, 1 H), 4.69 (s, 2 H), 4.08
(s, 3 H), 2.83 (s, 6 H). – 13C NMR (65.5 MHz, CDCl3): δ ϭ 173.2,
156.6, 133.5, 128.4, 127.5, 127.1, 126.7, 123.8, 123.4, 118.7, 107.1,
56.5, 55.1, 42.2. – HRMS: C15H17NO3ϩ [M ϩ Hϩ]: calcd. 259.1208;
found 259.1189.
[16] [16a] P. A. Bartlett, C. R. Johnson, J. Am. Chem. Soc. 1985, 107,
[16b]
7792–7793. –
D. Y. Jackson, J. W. Jacobs, R. Sugasawara,
S. H. Reich, P. A. Bartlett, P. G. Schultz, J. Am. Chem. Soc.
[16c]
1988, 110, 4841–4842. –
M. R. Haynes, E. A. Stura, D.
Hilvert, I. A. Wilson, Science 1994, 263, 646–652.
[17] [17a]
D. Hilvert, K. W. Hill, K. D. Nared, M.-R. M. Auditor, J.
[17b]
Am. Chem. Soc. 1989, 111, 9261–9262. –
A. C. Braisted,
[17c]
P. G. Schultz, J. Am. Chem. Soc. 1990, 112, 7430–7431. –
V. E. Gouverneur, K. N. Houk, B. de Pascual-Teresa, B. Beno,
[17d]
K. D. Janda, R. A. Lerner, Science 1993, 262, 204–208. –
C. J. Suckling, M. C. Tedford, L. M. Bence, J. I. Irvine, W. H.
Stimson, J. Chem. Soc., Perkin Trans. 1 1993, 1925–1929. – [17e]
A. A. P. Meekel, M. Resmini, U. K. Pandit, Bioorg. Med.
Chem. 1996, 4, 1051–1057.
B. F. Cravatt, J. A. Ashley, K. D. Janda, D. L. Boger, R. A.
Lerner, J. Am. Chem. Soc. 1994, 116, 6013–6014.
H. Suga, N. Tanimoto, A. J. Sinskey, S. Masamune, J. Am.
Chem. Soc. 1994, 116, 11197–11198.
J. T. Yli-Kauhaluoma, J. A. Ashley, C.-H. Lo, J. Coakley, P.
Wirsching, K. D. Janda, J. Am. Chem. Soc. 1996, 118, 5496–
5497.
Crystallographic Data: Crystallographic data (excluding structure
factors) for the structures of 2 and 12 have been deposited with the
Cambridge Crystallographic Data Centre as supplementary pub-
lication nos. CCDC-125392 (2) and CCDC-121427 (12). Copies of
the data can be obtained free of charge on application to the
CCDC, 12 Union Road, Cambridge CB2 1EZ, U.K. [Fax: (in-
ternat.) ϩ44 (0)1223/336033; E-mail: deposit@ccdc.cam.ac.uk].
[18]
[19]
[20]
[21]
M. Oki, Top. Stereochem. 1983, 14, 1–79.
Acknowledgments
[22] [22a]
J. Clayden, Angew. Chem. 1997, 109, 986–987; Angew.
[22b]
We thank Dr. R. A. Lerner for his support and Dr. T. Ward for
helpful suggestions. Financial support for this work was provided
by the Deutsche Forschungsgemeinschaft and the Roche Founda-
tion (postdoctoral fellowships to O. R.), the Swiss National Science
Foundation, the Koordinationsgruppe für Forschungsfragen der
Basler chemischen Industrie (KGF), and the Wander Stiftung.
Chem. Int. Ed. Engl. 1997, 36, 949–951. –
J. Clayden, M.
Darbyshire, J. H. Pink, N. Westlund, F. X. Wilson, Tetrahedron
Lett. 1997, 38, 8587–8590.
[23] [23a]
´
M. G. Rossman, A. Liljas, C.-I. Bränden, L. J. Banaszak,
‘‘Evolutionary and structural relationship among dehydro-
genases’’, in The Enzymes, Academic Press, Inc., 1975, vol. 11,
[23b]
p. 61. –
N. J. Oppenheimer, A. L. Handlon, ‘‘Mechanism
of NAD-dependent enzymes’’, in The Enzymes, Academic
Press, Inc., 1992, vol. 20, p. 453.
[1]
J. Kraut, Science 1988, 242, 533 and refs. cited therein.
[24]
[25]
J. Schröer, M. Sanner, J.-L. Reymond, R. A. Lerner, J. Org.
Chem. 1997, 62, 3220–3229.
[2] [2a]
[2b]
J. L. Kurz, Acc. Chem. Res. 1972, 5, 1–9. –
O. S. Tee,
Adv. Phys. Org. Chem. 1994, 29, 1–85. – [2c] R. Wolfenden, Acc.
Chem. Res. 1972, 5, 10–18. – [2d] G. E. Lienhard, Science 1973,
180, 149–154. – [2e] J.-L. Reymond, Y. Chen, Isr. J. Chem. 1996,
36, 199–206.
A. D. Mesecar, B. L. Stoddard, D. E. Koshland, Jr., Science
1997, 277, 202–206.
[26] [26a] J. W. Verhoeven, W. van Gerresheim, F. M. Martens, S. M.
[26b]
van der Kerk, Tetrahedron 1986, 42, 975–992. –
Y.-D. Wu,
[3] [3a]
M. M. Mader, P. A. Bartlett, Chem. Rev. 1997, 97, 1281–
[26c]
K. N. Houk, J. Am. Chem. Soc. 1987, 109, 2226–2227. –
1301. – [3b] J. D. Stewart, S. J. Benkovic, Nature 1995, 375, 388–
O. Tapia, R. Cardenas, F. Colonna-Cesari, J. Am. Chem. Soc.
[3c]
391. –
A. J. Kirby, Angew. Chem. 1996, 108, 771–790; An-
[26d]
1988, 110, 4046–4047. –
J. Wilkie, I. H. Williams, J. Am.
[3d]
gew. Chem. Int. Ed. Engl. 1996, 35, 707–724. –
I. Fujii, F.
[26e]
Chem. Soc. 1992, 114, 5423–5425. –
Y.-D. Wu, D. K. W.
Tanaka, H. Miyashita, R. Tanimura, K. Kinoshita, J. Am.
Chem. Soc. 1995, 117, 6199–6209.
Lai, K. N. Houk, J. Am. Chem. Soc. 1995, 117, 4100–4108.
[27] [27a]
[4]
[5]
[6]
Ö. Almarsson, R. Karaman, T. C. Bruice, J. Am. Chem.
[27b]
[4a] P. G. Schultz, R. A. Lerner, Science 1995, 269, 1835–1842. –
[4b]
Soc. 1992, 114, 8702–8703. –
J. Am. Chem. Soc. 1993, 115, 2125–2138. –
Ö. Almarsson, T. C. Bruice,
[27c]
J.-L. Reymond, Top. Curr. Chem. 1999, 200, 59–93.
B. Schiøtt, Y.-
H. Primas, U. Müller-Herold in Elementare Quantenchemie, 2.
J. Zheng, T. C. Bruice, J. Am. Chem. Soc. 1998, 120, 7192–7200.
Aufl./2nd Ed., Teubner, Stuttgart, 1990, p. 189–190.
[28]
[29]
[30]
L. C. Hsieh, S. Yonkovich, L. Kochersperger, P. G. Schultz,
Science 1993, 260, 337–339.
H.-J. Hofmann, R. Cimiraglia, FEBS Lett. 1988, 241, 38–40
and refs. cited therein.
K. P. Nambiar, D. M. Stauffer, P. A. Kolodziej, S. A. Benner,
J. Am. Chem. Soc. 1983, 105, 5886–5890.
Compound types other than transition state analogs can be
utilized to prepare enzyme inhibitors and/or haptens inducing
catalytic antibodies. Examples include suicide inhibitors, bait
and switch, and reactive immunizations.
[7]
M. Bols, Acc. Chem. Res. 1998, 31, 1–8 and refs. cited therein.
[8] [8a]
L. J. Liotta, R. A. Gibbs, S. D. Taylor, P. A. Benkovic, S. J.
[31] [31a]
[8b]
A. Ohno, Y. Ishikawa, N. Yamazaki, M. Okamura, Y. Ka-
Benkovic, J. Am. Chem. Soc. 1995, 117, 4729–4741. –
O.
[31b]
wai, J. Am. Chem. Soc. 1998, 120, 1186–1192. –
A. Ohno,
Ersoy, R. Fleck, A. Sinskey, S. Masamune, J. Am. Chem. Soc.
1996, 118, 13077–13078.
Y. Ishihara, S. Ushida, S. Oka, Tetrahedron Lett. 1982, 31,
[31c]
[9] [9a]
3185–3188. –
hedron Lett. 1979, 1397–1400.
S. Shinkai, H. Hamada, O. Manabe, Tetra-
K. D. Janda, C. G. Shevlin, R. A. Lerner, Science 1993,
[9b]
259, 490–493. –
K. D. Janda, C. G. Shevlin, R. A. Lerner,
[32] [32a]
S. M. Abdel-Wahhab, L. S. El-Assad, J. Ind. Chem. Soc.
J. Am. Chem. Soc. 1995, 117, 2659–2560.
[32b]
[10] [10a]
1969, 46, 927–932. –
Minematsu, J. Chem. Soc., Perkin Trans. 1 1994, 1833–1845.
O. Muraoka, G. Tanabe, K. Sano, T.
T. Li, K. D. Janda, J. A. Ashley, R. A. Lerner, Science
[10b]
1994, 264, 1289–1293. –
Am. Chem. Soc. 1995, 117, 3308–3309. –
Janda, R. A. Lerner, Nature 1996, 379, 326–327. –
T. Li, S. Hilton, K. D. Janda, J.
[33]
[34]
[10c]
P. M. T. de Kok, L. A. M. Bastiaansen, P. M. van Lier, J. A. J.
M. Vekemans, H. M. Buck, J. Org. Chem. 1989, 54, 1313–1320.
Recent examples of stereoselective hydride transfer with pyrid-
ine cofactors:[34a] N. Kanomata, M. Suzuki, M. Yoshida, T.
Nakata, Angew. Chem. 1998, 110, 1506–1508; Angew. Chem.
T. Li, K. D.
[10d]
J. Has-
serodt, K. D. Janda, R. A. Lerner, J. Am. Chem. Soc. 1996, 118,
11654–11655 and 1997, 119, 5993–5998. – [10e] J. Hasserodt, K.
D. Janda, Tetrahedron 1997, 53, 11237–11255.
Eur. J. Org. Chem. 2000, 1365Ϫ1372
1371

O. Ritzeler, S. Parel, B. Therrien, N. Bensel, J.-L. Reymond, K. Schenk
FULL PAPER
[34b]
[36]
Int. Ed. Engl. 1998, 37, 1410–1412. –
N. Kanomata, T.
Related study on naphthylcarbinols: D. Casarini, L. Lunazzi,
A. Mazzanti, J. Org. Chem. 1997, 62, 3315–3323.
H. Günther, NMR-Spektroskopie, 3rd ed., Georg Thieme Ver-
lag, Stuttgart, 1992, pp. 304–322.
J.-L. Reymond, Angew. Chem. 1995, 107, 2471–2473; Angew.
Chem. Int. Ed. Engl. 1995, 34, 2285–2287.
Received June 28, 1999
Nakata, Angew. Chem. 1997, 109, 1263–1266; Angew. Chem.
[34c]
[37]
[38]
Int. Ed. Engl. 1997, 36, 1207–1211. –
S. Obika, T. Nishi-
yama, S. Tatematsu, K. Miyashita, C. Iwata, T. Imanishi, Tet-
[34d]
rahedron 1997, 53, 593–602. –
T. Eguchi, M. Fukuda, Y.
Toyooka, K. Kakinuma, Tetrahedron 1998, 54, 705–714.
[35]
A. Ohno, M. Ohara, S. Oka, J. Am. Chem. Soc. 1986, 108,
6438–6440.
[O99387]
1372
Eur. J. Org. Chem. 2000, 1365Ϫ1372