Synthesis, conformation, and stereodynamics of a salt of 2-{[2-(3,4-dichlorophenyl)-ethyl]propylamino}-1-pyridin-3-ylethanol

Article abstract of DOI:10.1021/jo051455f

A novel synthetic route was developed for 2-{[2-(3,4-dichlorophenyl)ethyl] propylamino}-1-pyridin-3-ylethanol (4). A dynamic process due to nitrogen inversion at the central amine nitrogen has been identified by NMR spectroscopy for the dihydrobromide sal

Full text of DOI:10.1021/jo051455f

The new method is a reductive alkylation of the secondary
amine2 6 using the selective hydride reducing agent sodium tri-
acetoxyborohydride.5 The second method gives 4 in higher yield
and with fewer side products (total yield: 59%) (Scheme 2).
The dihydrobromide salt 5 is prepared in both cases (Figure 1).
Synthesis, Conformation, and Stereodynamics of
a Salt of 2-{[2-(3,4-Dichlorophenyl)-
ethyl]propylamino}-1-pyridin-3-ylethanol
Tina Korosˇec,† Jozˇe Grdadolnik,‡ Urosˇ Urleb,†
Darko Kocjan,† and Simona Golicˇ Grdadolnik*,‡
Drug DiscoVery, Lek Pharmaceuticals d. d., VeroVsˇkoVa 57,
SI-1526 Ljubljana, SloVenia, and Laboratory for Molecular
Modeling and NMR Spectroscopy, National Institute of
Chemistry, HajdrihoVa 19, SI-1001 Ljubljana, SloVenia
simona.grdadolnik@ki.si
ReceiVed July 14, 2005
FIGURE 1. Primary structure of 5 illustrating the atom nomenclature
used in this study.
The observed extensive line broadening in the proton NMR
spectra of 5 measured in DMSO-d6 at room temperature
indicates the existence of a dynamic process which could be
quantitatively examined by NMR spectroscopy.
A novel synthetic route was developed for 2-{[2-(3,4-
dichlorophenyl)ethyl]propylamino}-1-pyridin-3-ylethanol (4).
A dynamic process due to nitrogen inversion at the central
amine nitrogen has been identified by NMR spectroscopy
for the dihydrobromide salt of this compound. The confor-
mational properties of the diastereomeric pair were deter-
mined by analysis of NOE connectivities and MO calcula-
tions.
The stereodynamics of tertiary amines is a complex process
because the inversion of nitrogen and the rotation around various
single bonds are potentially dynamic processes. Although the
inversion rates of amines in crowded molecules are usually
higher than the rotation rates around single bonds, it is often
difficult to distinguish between the two processes. Sometimes
the introduction of appropriate substituents influences the mag-
nitude of the inversion or rotation barrier and leads to identifica-
tion of a particular dynamic process. In open-chain amines,
observation of nitrogen inversion is especially difficult because
of low activation energies. However, the presence of acidic med-
ia can significantly increase the barrier to nitrogen inversion
due to N-atom protonation and proton exchange, thus enabling
detection of this dynamic process under experimental conditions
appropriate for the application of high-resolution NMR tech-
niques.6
A series of novel derivatives of pyridylethanol(phenethyl)-
amines was synthesized in order to explore the pharmacophoric
role of the pyridine nucleus within the tertiary arylalkylamine
structure. Such compounds display high affinities for various
receptors, a notable example is the inhibition of cholesterol
biosynthesis by 2-{[2-(3,4-dichlorophenyl)ethyl]propylamino}-
1-pyridin-3-ylethanol (4) in a cell assay.1 In a continuation of
our previous work, we have developed an additional synthetic
route which considerably improves the overall synthetic yield.
The dihydrobromide salt of 4 has shown some interesting
NMR spectral properties. Here we report on a novel synthetic
route for the above compound and a detailed investigation of
the dynamic and conformational properties of its salt form using
NMR spectroscopy and MO calculations.
The dynamic behavior of 5 was further studied in CD3OD to
enable measurements at lower temperatures. Splitting of broad
resonances was expected upon cooling the sample, because
internal motion is halted on the NMR time scale at sufficiently
low temperature, and therefore, separate signals for the present
isomers could be observed. Surprisingly, the splitting of
resonances in the proton spectrum of 5 in CD3OD could already
be observed at 25 °C (Figure S1, Supporting Information). In
its salt form, racemic 4 shows distinctive diastereomeric effects.
We suppose that nitrogen inversion is the mechanism responsible
for such a dynamic process. A significantly increased barrier
to nitrogen inversion, which usually cannot be observed at room
temperature in open-chain amines, can be attributed to the
presence of HBr. Signal intensities of one diastereomer are
slightly weaker, with approximately 10% lower integral values,
which indicates a 10% lower population of this diastereomer.
In the original route,1 amide 3 was prepared by coupling
secondary amine2 2 with the corresponding carboxylic acid using
EDC and HOBT activation.3 Tertiary amide 3 was further
reduced by a borane-dimethyl sulfide complex4 to provide the
free base 4 (total yield: 26%) (Scheme 1).
† Lek Pharmaceuticals d. d.
‡ National Institute of Chemistry.
(1) Rode, B.; Rozman, D.; Fon Tacer, K.; Kocjan, D. PCT WO 2004/
007456 A1.
1
The H and 13C chemical shift assignment is available in the
(2) Polo, F. L. Farm. Ed. Sc. 1963, 18, 972-980.
(3) Feng, D. M.; Gardell, S. J.; Lewis S. D.; Bock, M. G.; Chen, Z.;
Freidinger, R. M.; Naylor-Olsen, A. M.; Ramjit, H. G.; Woltmann, R.;
Baskin, E. P.; Lynch, J. J.; Lucas, R.; Shafer, J. A.; Dancheck, K. B.; Chen,
I. W.; Mao, S. S.; Krueger, J. A.; Hare, T. R.; Mulichak, A. M.; Vacca, J.
P. J. Med. Chem. 1997, 40, 3726-3733.
Supporting Information.
(5) Abdel-Magid, A. F.; Carson, K. G.; Harris, B. D.; Maryanoff, C. A.;
Shah, R. D. J. Org. Chem. 1996, 61, 3849-3862.
(6) Oki, M. In Application of Dynamic NMR Spectroscopy to Organic
Chemistry; Marchand, A. P., Ed.; Methods in Stereochemical Analysis 4;
VCH Publishers: Deerfield Beach, FL, 1985; pp 339-356.
(4) Brown, H. C.; Choi, Y. M.; Narasimhan, S. J. Org. Chem. 1982, 47,
3153-3163
10.1021/jo051455f CCC: $33.50 © 2006 American Chemical Society
Published on Web 12/15/2005
792
J. Org. Chem. 2006, 71, 792-795

SCHEME 1. Synthesis of Compound 5 (Route A)
SCHEME 2. Synthesis of Compound 5 (Route B)
in the preparation of NMR samples. Benzenesulfonic acid is
added in steps of 2 mM to the 10 mM solution of 4 in CD3OD.
Splitting of resonances appears at a 1:1 concentration ratio of
acid relative to 4 (Figure S4, Supporting Information). The same
resonances are affected as for 5 in CD3OD. This is a clear
indication that nitrogen inversion is the responsible dynamic
mechanism. At sufficient acid concentration, the on/off rate of
ligation of the acid to the nitrogen lone pair is reduced and the
barrier to nitrogen inversion is increased to an extent which
enables the observation of separate signals of each diastereomer
in the NMR spectra at relatively high temperatures.
Conformational properties of 5 were examined by 2D NOESY
and ROESY NMR methods and MO calculations. Although the
rate of exchange between diastereomers of 5 in CD3OD is slow
on the NMR time scale at 25 °C and allows observation of
separate signals in the NMR spectra, common NOE peaks for
both diastereomers are observed in the NOESY spectra at this
temperature. NOE measurements were repeated at lower tem-
perature in order to reduce the rate of internal motions. The
exchange rate is sufficiently reduced by cooling the sample to
0 °C, since averaging of NOEs is eliminated and each diaster-
eomer has individual NOE cross-peaks (Figure S5, Supporting
Information). Thus, observation of conformational differences
between diastereomers is enabled at this temperature.
The only difference between the diastereomers that can be
observed in the NOESY and ROESY spectra are the NOE cross-
peaks of the H7′ protons (Figure S6, Supporting Information).
One diastereomer has an NOE cross-peak between the H7′ and
H2′′ protons, while the other has it between H7′ and H7 protons.
This can be attributed to the close proximity of the H7 proton
with propyl chain in one and with the ethyl chain in the other
diastereomer and indicates a different spatial arrangement of
the ethanol, propyl and ethyl chains in diastereomers of 5. Due
to the extensive overlap of signals in the region of the H8 and
H8′ protons, it is not possible to determine any other specific
NOE for the particular arrangement of above-mentioned mo-
lecular segments. The same exclusive NOE connectivities of
each diastereomer of 4 are also observed in the solution of CD3-
OD and benzenesulfonic acid at 0 °C.
The large chemical shift differences of H2 and H6 protons
of the dichlorophenyl ring between diastereomers cannot be
explained on the basis of NOE connectivities. The H2 and H6
protons have NOE cross-peaks only with nearby protons of the
ethyl chain (H7 and H8). NOE cross-peaks of these protons
are of similar intensities for both diastereomers, which indicates
a similar orientation of the dichlorophenyl ring relative to the
ethyl chain.
Chemical exchange signals between the two sets of signals
are observed in the NOESY (Figure S2, Supporting Information)
and ROESY spectra, which confirm the existence of two
exchanging isomers of 5. In addition, the variable-temperature
measurements in CD3OD show behavior typical of resonances
undergoing exchange processes. The broadening and coales-
cence of signals was observed on heating of the sample (Figure
S3, Supporting Information).
Inspection of the carbon chemical shift differences between
the diastereomers supports nitrogen inversion as the responsible
exchange process. Namely, the resonances of the carbons (C8
and C1′′) near the nitrogen atom are largely split. The behavior
of the proton resonances is more complex. Large splitting of
the resonances of protons H1′′, H8, and H8′ near the amine
nitrogen is in good agreement with nitrogen inversion. On the
other hand, comparable splitting of the ortho protons of the
dichlorophenyl ring (H2, H6) is unexpected, if it is only the
configuration at the amine nitrogen that is responsible for the
chemical shift differences between the two diastereomers. The
resonances of the pyridine ring are not split. We suppose that
the two diastereomers have different overall conformations. The
whole exchange process between diastereomers consists of acid
dissociation, nitrogen inversion, nitrogen reprotonation, and
conformational reorganization.
Barriers (∆Gq) of internal motion in 5 were estimated from
the differences in the chemical shifts of the two exchange sites
at the coalescence temperature. The coalescence of N-bonded
carbons (C8 and C1′′) is reached at 60 °C, and the barrier is
16.0 kcal mol-1. The extensive overlap of resonances of the
protons nearest to the amine nitrogen (H8, H8′, H1′′) prevents
their application for determination of inversion barriers. The
signals in the nonoverlapped regions of the proton spectra, which
are remote from the amine nitrogen, provide similar barriers of
internal motion. The barriers for the H3′′, H7′, and the ortho
protons of the dichlorophenyl ring (H2, H6) are 16.2 (42 °C),
16.4 (40 °C), and 16.6 (52 °C) kcal mol-1, respectively. These
results indicate that the exchange process observed in the
resonances throughout the molecule is caused mainly by nitrogen
inversion at the central amine nitrogen.
The MO calculated structures of diastereomers of amine
N-protonated species 4 with total charge ) +1 fall into two
distinctive conformational families with respect to the torsional
angles τ1 (C7-C8-N-C8′) and τ2 (C8-N-C8′-C7′) ) -80/
The observation of nitrogen inversion at unusually high temp-
eratures is further confirmed by titration studies with benzene-
sulfonic acid which is used instead of HBr due to its convenience
J. Org. Chem, Vol. 71, No. 2, 2006 793

plates (0.25 mm), and components were visualized with ultraviolet
light. Column chromatography was carried out on silica gel 60
(particle size 240-400 mesh). Melting points were determined on
a hot stage microscope and are uncorrected. IR spectra were
recorded with a FT-IR spectrometer. Samples were prepared
between NaCl windows or as KBr pellets. MS spectra were
1
measured by ESI, FAB, and HRMS methods. H NMR and 13C
NMR experiments for the identification of compounds were carried
out at 300 MHz in a CDCl3 or DMSO-d6 solution with TMS as
the internal standard. The NMR studies of conformation and
stereodynamics were carried out at 600 MHz in a CD3OD solution.
Route A (Scheme 1). 2-Propylamino-1-pyridin-3-ylethanol (2).
NaBH4 (1.0 g, 26.4 mmol) was added to a solution of compd 1
(2.0 g, 7.2 mmol) in anhyd EtOH (40 mL). The reaction mixture
was stirred at rt for 2 h. The mixture was filtered, and PrNH2 (1.4
mL, 17.0 mmol) was added to the filtrate liquid. The solution was
heated to reflux and refluxed for 5 h. The EtOH was removed by
distillation. The resulting pale yellow solid was dissolved in CHCl3
(40 mL), the insoluble part was filtered off, and the filtrate was
concentrated by distillation under reduced pressure. The product
was chromatographed on silica (MeOH/EtOAc, 10:2, and MeOH/
NH2OH, 9:1) yielding a yellow oil: yield 0.7 g, 54% (93% pure
FIGURE 2. Computed structures of diastereomers of amine N-
protonated species 4: (A) conformational family τ1/τ2 ) -80/150, (B)
conformational family τ1/τ2 ) -160/90.
150 and -160/90, respectively, with closely spaced potential
energy levels.7 Molecular orbital calculations in vacuo indicate
that the OH‚‚‚NH+ interaction is a dominant feature in both
families and that it favors closed structures with the phenyl ring
positioned in a π-electron density interaction with the NH+
group (τ3 (C1-C7-C8-N) ) gauche). Structures with the
phenyl ring in an extended conformation (τ3 ) trans) are only
1.5 kcal/mol higher in energy. These structures are more similar
to those in polar solvents such as CD3OD since H7 protons
exhibit strong NOE effects.
It was quite interesting to find that both NOE connectivities
H7′-H7 and H7′-H2′′ correspond to the conformational family
τ1/τ2 ) -80/150 for each diastereomer which exhibit shorter
interatomic distances between respective H-atoms than those
in the family τ1/τ2 ) -160/90 (Figure 2A). On the other hand,
the chemical shift differences in protons 2 and 6 between the
diastereomers can be explained by the different steric position
of the pyridine ring with respect to the phenyl ring within the
conformational family τ1/τ2 ) -160/90 (Figure 2B). MO
calculations confirm the different steric position of H7′ protons
relative to the propyl and ethyl chain in each diastereomer
proposed on the basis of NOE connectivities. Simultaneous
observation of the different NOE pattern of H7′ protons and
the chemical shift differences in protons 2 and 6 of the
dichlorophenyl ring between the diastereomers can be explained
by the fast conformational exchange between the two confor-
mational families τ1/τ2 for each diastereomer in CD3OD.
We have shown an interesting case of a diastereomeric effect
in the salt form of a racemic tertiary arylalkylamine and we
have put forward some NMR spectroscopic evidence to
demonstrate the mechanisms underlying this effect.
1
by area % HPLC analysis); H NMR (300 MHz, CDCl3) δ 0.91
(3H, t, J ) 7.5 Hz), 1.47-1.59 (2H, m), 2.16-2.93 (3H,m), 2.90
(1H, dd, 1H, JRH′′-âH ) 3.5 Hz, Jg ) 11.9 Hz), 3.45 (2H, br s),
4.82 (1H, dd, JâH-RH′ ) 9.4 Hz, JâH-RH′′ ) 3.4 Hz), 7.25 (1H, m),
7.73 (1H, td, 1H, J ) 7.6 Hz, J ) 1.9 Hz), 8.45 (1H, dd, J ) 4.9
Hz, J ) 1.5 Hz), 8.55 (1H, d, J ) 2.25 Hz); Rf (MeOH/NH2OH,
9:1) ) 0.70; FT-IR (NaCl) 2963, 1664, 1593, 1470, 1421, 1095,
1053, 903, 809, 719, 634 cm-1; FAB MS m/z 181 [MH+]; HRMS
m/z calcd for C10H16N2O 181.1341, found 181.1345.
2-(3,4-Dichlorophenyl)-N-(2-hydroxy-2-pyridin-3-ylethyl)-N-
propylacetamide (3). HOBT (0.41 g, 3.0 mmol) was added to a
solution of compd 2 (0.55 g, 3.05 mmol) and 3,4-dichlorophenyl-
acetic acid (0.63 g, 3.0 mmol) in DMF (5 mL). The pH of the
solution was adjusted to 8 by adding N-methylmorpholine. EDC
(0.6 g, 3.1 mmol) was added. After the reaction mixture was stirred
at rt overnight, the solvent was evaporated under reduced pressure
and the residue dissolved in EtOAc (20 mL). The organic layer
was washed with aq saturated NaHCO3 (20 mL) and NaCl (20 mL)
solution and dried (Na2SO4). The solvent was evaporated under
reduced pressure yielding a colorless oil which was chromato-
graphed on silica (MeOH/EtOAc, 2:10): yield 0.92 g, 82% (99%
1
pure by area % HPLC analysis); H NMR (300 MHz, CDCl3) δ
0.91 (3H, t, J ) 7.5 Hz), 1.51-1.63 (2H, m), 3.08-3.33 (2H, m),
3.51 (1H, dd, JRH′′-âH ) 2,6 Hz, Jg ) 14,1 Hz), 3.68-3.75 (3H,
m), 4.82 (1H, br s), 5.03 (1H, dd, JâH-RH′ ) 9.4 Hz, JâH-RH′′ ) 3.4
Hz), 7.11 (1H, dd, J ) 8.3 Hz, J ) 2.25 Hz), 7.25-7.44 (3H, m),
7.73 (1H, td, J ) 7.9 Hz, J ) 1.9 Hz), 8.45 (1H, dd, J ) 4.75 Hz,
J ) 1.5 Hz), 8.55 (1H, d, 1H, J ) 2.1 Hz); Rf (CH2Cl2/MeOH,
10:0.5) ) 0.15; FT-IR (NaCl) 3368, 2965, 2875, 1636, 1472, 1420,
1230, 1131, 1079,1031, 801, 715 cm-1; EI MS m/z 366 [M+], FAB
MS m/z 367 [MH+]; HRMS m/z calcd for C18H20Cl2N2O2 [M+]
366.0902, found 366.0910.
2-{[2-(3,4-Dichlorophenyl)ethyl]propylamino}-1-pyridin-3-
ylethanol (4). The solution of compd 3 (0.77 g, 2.1 mmol) in dry
THF (10 mL) was heated to reflux, and a 2 M solution of borane-
dimethyl sulfide complex in diethyl ether (3.30 mL, 6.3 mmol) was
added in drops over a 15 min period, allowing dimethyl sulfide to
distill off. The reaction mixture was refluxed for 10 h. The THF
solution was then hydrolyzed during addition of 6 N HCl (0.35
mL, 2.1 mmol). After 30 min, the clear solution obtained was cooled
to room temperature and neutralized with 6 N NaOH (0.6 mL, 3.2
mmol). The reaction mixture was stirred at rt for another 1 h. EtOAc
(20 mL) was added, and the organic layer was washed with aq
saturated NaHCO3 (20 mL) and NaCl (20 mL) solution and dried
(Na2SO4). The solvent was evaporated under reduced pressure to
give a product which was further chromatographed on silica
Experimental Section
General Methods. Starting materials, reagents, and solvents were
purchased from commercial suppliers and used without further
purification. Analytical TLC was performed on silica gel (60 F 254)
(7) We also performed conformational analysis of the diprotonated
species and found that the second protonation does not affect the
characteristic conformational families. However, it does cause some changes
in the distribution of the conformational states. In addition, pKa’s were
determined by Across Barriers (7.66, 4.12), which show that the pyridine
nucleus is weakly basic. Therefore, monoprotonated species could be
predominant, particularly in the organic solvent.
794 J. Org. Chem., Vol. 71, No. 2, 2006

(MeOH/EtOAc, 2:10) to give a colorless oil: yield 0.43 g, 58%
(95% pure by area % HPLC analysis); 1H NMR (300 MHz, CDCl3)
δ 0.91 (3H, t, J ) 7.5 Hz), 1.48-1.53 (2H, m), 2.46-2.80 (8H,
m), 4.63 (1H, dd,1H, JâH-RH′ ) 10.5 Hz and JâH-RH′′ ) 3.6 Hz),
7.03 (1H, dd, 1H, J ) 8.2 Hz, J ) 2.1 Hz), 7.26-7.38 (3H, m),
7.71 (1H, td, J ) 7.8 Hz, J ) 1.7 Hz), 8.52 (1H, dd, J ) 4.75 Hz,
J ) 1.5 Hz), 8.56 (1H, d, J ) 2.2 Hz); Rf (MF: MeOH/EtOAc,
2:10) ) 0.46; FT-IR (NaCl) 3321, 2963, 2958, 1472, 1425, 1397,
1131, 1030, 819, 715 cm-1; EI MS m/z 353 [MH+], FAB MS m/z
353 [MH+]; HRMS m/z calcd for C18H23Cl2N2O [MH+] 353.1187,
found 353.1186.
in CD3OD. Sample concentrations were 25 mM and 10 mM. 2D
spectra were measured at 25 and 0 °C. The proton chemical shifts
were assigned following a standard procedures using homonuclear
DQF-COSY,8 TOCSY,9,10 and NOESY11 experiments in combina-
1
tion with an H-13C HSQC12 experiment. Carbons were assigned
using the 1H-13C HSQC spectrum. Typically, the 2D proton spectra
were acquired in the phase sensitive mode with 4096 data points
in the t2 dimension, 4-32 scans, 256-512 complex points in the
t1 dimension, and a relaxation delay of 1-2 s. The 1H sweep width
was 6600 Hz. NOESY and ROESY13 spectra were measured with
150 and 300 ms mixing times. The 1H-13C HMQC spectrum was
acquired as a matrix of 512 × 128 complex points with 4 scans
and a relaxation delay of 1 s. The 13C sweep width was 25000 Hz.
Data were processed and analyzed using a FELIX software package
from Accelrys. Spectra were zero-filled two times and apodized
with a squared sine bell function shifted by π/2 in both dimensions.
A linear prediction of the data was applied in the carbon dimension.
The chemical shift differences of the two exchange sites were
determined by line shape simulation performed using GRAMS/32
AI software from Thermo Galactic. For accurate determination of
the frequencies at coalescence the second derivatives were used.
Band components with known frequencies were adjusted using a
band fitting algorithm to minimize the difference between the
experimental and calculated band shape. A mixture of Lorentzian
and Gaussian band shape was used throughout. The only restriction
in the band fitting algorithm was the relative intensity of the band
component, which was assumed to obey the expected theoretical
values.
2-{[2-(3,4-Dichlorophenyl)ethyl]propylamino}-1-pyridin-3-
ylethanol Dihydrobromide (5). The solution of tertiary amine 4
(0.4 g, 1.1 mmol) in acetone (3.5 mL) was cooled on ice bath. A
0.97 mL (0.22 g HBr, 2.8 mmol) solution of HBr in EtOH was
added dropwise. After the precipitate appeared, approximate 2.5
mL of diethyl ether was added. The reaction mixture was stirred
on an ice bath for 2 h. The white precipitate was filtered and
successively washed with diethyl ether: yield 0.46 g, 79% (99%
pure by area % HPLC analysis); mp 130-132 °C; Rf (MF: MeOH/
1
EtOAc, 10:2) ) 0.6; H NMR (300 MHz, DMSO-d6, 50 °C) δ
0.94 (3H, t, J ) 7.1 Hz), 1.80 (2H, m), 3.10-3.70 (8H, m), 5.54
(1H, m), 7.39 (1H, br), 7.59 (1H, d, J ) 8.2 Hz), 7.70 (1H, br),
8.11 (1H, dd, J ) 8.1 Hz, J ) 5.6 Hz), 8.72 (1H, dt, J ) 8.2 Hz,
J ) 1.7 Hz), 8.93 (1H, dd, J ) 5.7 Hz, J ) 1.4 Hz), 9.09 (1H, d,
J ) 1.8 Hz); FT-IR (KBr) 3411, 3168, 2953, 2684, 1626, 1536,
1472, 1347, 1210, 1128, 1028, 805, 680 cm-1; EI MS m/z 353
[MH+], FAB MS m/z 353 [MH+]; HRMS m/z calcd for C18H22-
Cl2N2O [MH+] 353.1187, found 353.1188. Anal. Calcd for C18H24-
Br2Cl2N2O‚4/5H2O: C, 40.83; H, 4.87; N, 5.29. Found: C, 40.86;
H, 4.74; N, 5.21.
MO Calculations. Conformational analysis was performed using
a Spartan software package running on a Silicon Graphics computer.
All rotatable bonds were systematically varied and conformational
energies calculated at the AM1 level. Relevant minima were also
calculated at the ab initio HF level using a 3-21g* and 6-31g* basis
set. The calculations with the 6-31g* basis set were performed using
Gaussian 98 software.
Route B (Scheme 2). 2-[2-(3,4-Dichlorophenyl)ethylamino]-
1-pyridin-3-ylethanol (6). NaBH4 (2.0 g, 52.8 mmol) was added
to a solution of compd 1 (4.0 g, 14.4 mmol) in anhyd EtOH (60
mL). The reaction mixture was stirred at rt for 2 h. The mixture
was filtered, and 3,4-dichlorophenethylamine (3.9 mL, 26.0 mmol)
was added to the filtrate liquid. The solution was heated to reflux
and refluxed overnight. The EtOH was removed by distillation. The
resulting pale yellow solid was dissolved in CHCl3 (60 mL), the
insoluble part was filtered off, and the filtrate concentrated by
distillation under reduced pressure. The product was chromato-
graphed on silica (MeOH/EtOAc, 10:2 to gradient elution MeOH/
EtOAc, 1:1) to give a yellow oil: yield 3.18 g, 71%; (96% pure
by area % HPLC analysis); 1H NMR (300 MHz, CDCl3) δ 2.66-
3.01 (6H, m), 4.71 (1H, dd, JâH-RH′ ) 9.2 Hz, JâH-RH′′ ) 3.6 Hz),
7.03 (1H, dd, J ) 8.2 Hz, J ) 2.0 Hz), 7.26-7.38 (3H, m), 7.70
(1H, td, J ) 7.8 Hz, J ) 1.8 Hz), 8.51 (1H, dd, J ) 4.7 Hz, J )
1.7 Hz), 8.56 (1H, d, J ) 2.15 Hz); Rf (MF: MeOH/EtOAc, 10:2)
) 0.3; FT-IR (NaCl) 3411, 2934, 2835, 1602, 1561, 1472, 1413,
1130, 1031, 818 cm-1; EI MS m/z 311 [MH+], FAB MS m/z 311
[MH+]; HRMS m/z calcd for C15H17Cl2N2O [MH+] 311.0718, found
311.0725.
2-{[2-(3,4-Dichlorophenyl)ethyl]propylamino}-1-pyridin-3-
ylethanol Dihydrobromide (5). Secondary amine 6 (0.5 g, 1.6
mmol) and CH3CH2CHO (0.2 mL, 2.4 mmol) were dissolved in
1,2-dichloroethane (10 mL) and then treated with NaBH(OAc)3
(0.53 g, 2.5 mmol). The mixture was stirred at rt under an Ar
atmosphere for 2 h. The reaction mixture was quenched by adding
aq saturated NaHCO3 (20 mL) solution, and the product was
extracted with EtOAc (20 mL). The EtOAc extract was dried
(NaSO4), and the solvent was evaporated under reduced pressure
to give the crude free base 4, which was chromatographed on silica
(MeOH/EtOAc, 10:2) to give a yellow oil: yield 0.47 g, 83% (97%
pure by area % HPLC analysis). The product was converted to the
dihydrobromide salt 5 (the procedure described under method A
was followed).
Acknowledgment. We thank Silva Zagorc and Sandi
Borisˇek for technical assistance. This work was supported by
Lek Pharmaceuticals d.d., the Ministry of Higher Education,
Science, and Technology of Slovenia, and the European
Community (STEROLTALK project contract No. LSHG-CT-
2005-512096).
1
Supporting Information Available: (1) Table of the H and
13C chemical shift assignments for 5 in CD3OD, (2) proton spectrum
of 5 in CD3OD, (3) expanded region of the NOESY spectra of 5 in
CD3OD showing the exchange cross-peaks, (4) temperature de-
pendence of selected 1H NMR signals of 5 in CD3OD, (5)
1
dependence of selected H NMR signals of 4 upon titration with
benzenesulfonic acid, (6) expanded regions of the NOESY spectra
of 5 measured at 25 and 0 °C in CD3OD, (7) expanded region of
the ROESY spectra of 5 measured at 0 °C in CD3OD, and (8)
compact Gaussian 98 output of structures presented in Figure 2.
This material is available free of charge via the Internet at
http://pubs.acs.org.
JO051455F
(8) Rance, M.; Sorensen, O. W.; Bodenhausen, G.; Wagner, G.; Ernst,
R. R.; Wuethrich K. Biochem. Biophys. Res. Commun. 1983, 117, 479-
485.
(9) Braunschweiler, L.; Ernst, R. R. J. Magn. Reson. 1983, 53, 521-
528.
(10) Bax, A.; Davis, D. G. J. Magn. Reson. 1985, 65, 355-360.
(11) Jeener, J.; Meier, B. H.; Bachmann, P.; Ernst, R. R. J. Chem. Phys.
1979, 71, 4546-4553.
(12) Willker, W.; Leibfritz, D.; Kerssebaum, R.; Bermel, W. Magn.
Reson. Chem. 1993, 31, 287-292.
NMR Studies of Conformation and Stereodynamics. NMR
spectra of 4 and 5 were recorded on a 600 MHz NMR spectrometer
(13) Griesinger, C.; Ernst, R. R. J. Magn. Reson. 1987, 75, 261-271.
J. Org. Chem, Vol. 71, No. 2, 2006 795