Stereochemical analysis of the enzymic reduction of the double bond of α- and β-substituted nitrostyrenes and α-ethoxycinnamaldehyde through deuterium labelling experiments

Article abstract of DOI:10.1002/ejoc.201000442

²H NMR studies and comparison with authentic labelled reference compounds prepared from the (Z)-acetamidocinnamic acid 10 and from (Z)-2-ethoxy-3-(4-methoxyphenyl) prop-2-en-1-ol (14), respectively, by catalytic syn reduction with deuterium gas

Full text of DOI:10.1002/ejoc.201000442

FULL PAPER
DOI: 10.1002/ejoc.201000442
Stereochemical Analysis of the Enzymic Reduction of the Double Bond of α-
and β-Substituted Nitrostyrenes and α-Ethoxycinnamaldehyde through
Deuterium Labelling Experiments
Elisabetta Brenna,*[a] Giovanni Fronza,*[b] Claudio Fuganti,[a] and Francesco G. Gatti[a]
Keywords: Isotopic labelling / Enzyme catalysis / Reduction
2H NMR studies and comparison with authentic labelled ref-
erence compounds prepared from the (Z)-acetamido-
cinnamic acid 10 and from (Z)-2-ethoxy-3-(4-methoxy-
the (Z)-α-ethoxycinnamaldehyde (3) leads to (2R,3R)-[2,3-
2H2]-3-(4-methoxyphenyl)-2-nitropropan-1-ol (5) and to
(1R,2S,3R)-[1,2,3-3H2]-2-ethoxy-3-(4-methoxyphenyl)propan-
phenyl)prop-2-en-1-ol (14), respectively, by catalytic syn re- 1-ol (6), respectively, thus supporting an anti mode of hydro-
duction with deuterium gas show that the baker’s yeast-me-
diated saturation in the presence of deuterated water of the
double bond of (E)-α-(hydroxymethyl)nitrostyrene (2) and of
gen addition across the activated double bond, with hydride
delivery in β-position respect to the activating group from the
upper side of the molecule.
penic aldehyde, the double bond of the cyclohexene ring, in
α position to the formyl group, is reduced with anti stereo-
chemistry in the case of one enantiomer, but in syn-fashion
in the other one.[3] The observation, acquired in the whole
cell system, has been subsequently confirmed using
NADPH and Ltb4 dehydrogenase derived from mouse as
a catalyst.[11] Also, cinnamaldehyde is transformed into 3-
phenylpropan-1-ol by anti formal addition of hydrogen to
the double bond,[2] while the opposite is true when 4-(4-
hydroxyphenyl)buten-2-one is converted into 4-(4-hy-
droxyphenyl)butan-2-one (raspberry ketone).[7]
Currently, the reductive transformation of a wide set of
substrates, already performed partially with baker’s yeast
and other microbial systems, is being studied using purified
enzymes, obtaining in some instances outstanding improve-
ments in terms of selectivity.[12] Reportedly,[13] the added
value of the enzymatic reduction of activated olefins is rep-
resented by the anti stereochemistry of hydrogen addition,
which is complementary to the syn addition promoted by
transition metal based homogeneous catalysts. The differen-
tiation between the two modes of addition can be achieved
using a tetrasubstituted olefin as a substrate. However, to
the best of our knowledge, the only two examples studied
in this field are the baker’s yeast reductions of (Z)-3-phenyl-
2-nitro-2-butene[14] and of (E)-2,3-dimethylcinnamal-
dehyde,[15] both leading to mixtures of diastereoisomers in
which the syn addition products prevail. This means that
of the rich array of enzymes presiding over the reductive
manipulation of activated double bonds in baker’s yeast
those particularly active towards the two mentioned materi-
als operate in syn fashion.
Introduction
For some years now we have been using baker’s yeast-
mediated transformations of compounds containing a car-
bonyl-activated double bond as a method for the prepara-
tion of enantiomerically enriched compounds,[1] including
materials that are chiral due to isotopic hydrogen substitu-
tion. In this context, the cryptic stereochemistry of the re-
duction of the double bond of α,β-unsaturated alde-
hydes,[2,3] ketones[4–7] and δ- and γ-lactones[8,9] has been ex-
amined in some detail by means of deuterium labelling.
The enzyme-mediated saturation of the activated double
bond is thought to occur by addition in β-position of an
hydrogen atom from the reduced cofactor, followed by in-
corporation in the α-place of a proton from the solvent
water. By performing the baker’s yeast reduction in the
presence of D2O – where exchange of the hydrogen atoms
of the reduced cofactor(s)[10] with those of the solvent takes
place – and/or by using specifically labelled substrates, it
has been possible to obtain stereochemical information rel-
ative to the reduction of unsaturated materials belonging to
quite different structural classes. In this context, it has been
shown that in the case of perillaldehyde, a chiral cyclic ter-
[a] Politecnico di Milano, Dipartimento di Chimica, Materiali ed
Ingegneria Chimica,
Via Mancinelli 7, 20131 Milano, Italy
Fax: +39-02-23993180
E-mail: elisabetta.brenna@polimi.it
[b] Istituto di Chimica del Riconoscimento Molecolare – CNR,
Dipartimento di Chimica, Materiali ed Ingegneria Chimica,
Via Mancinelli 7, 20131 Milano, Italy
Fax: +39-02-23993180
E-mail: giovanni.fronza@polimi.it
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201000442.
In this context it seemed of interest to develop an analyti-
cal procedure enabling the determination of the hidden
View this journal online at
wileyonlinelibrary.com
Eur. J. Org. Chem. 2010, 5077–5084
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5077

E. Brenna, G. Fronza, C. Fuganti, F. G. Gatti
FULL PAPER
stereochemistry of the enzymatic reduction of trisubstituted on -glucose. Derivative 5 was obtained as a unique trans-
olefins. This would be a useful tool to investigate those bio- formation product, shown to be a 85:15 mixture of two
reductions in which syn or anti hydrogen addition can af- enantiomers by HPLC analysis. However, conversion was
ford the two possible enantiomers of a chiral product. For too low to allow the recovery of a suitable quantity of com-
example it could be applied to elucidate the mode of action pound 5 for deuterium NMR analysis.
of the two sets of enzymes which are described to transform
When incubation was repeated on the free substrate,
(E)-1-phenyl-2-nitropropene into the two enantiomers of 1- added to fermenting baker’s yeast in ethanol solution, con-
phenyl-2-nitropropane,[16,17] providing that the substrate we version was complete, but the enantiomeric excess of the
choose as a model (see below) is transformed by the same resulting derivative 5 was found to be rather low (ee =
enzymes.
40%), anyway enough to perform deuterium analysis. The
The goal could be formally achieved only through iso- enantiomeric excess is often a function of dilution: if a lot
topic hydrogen substitution experiments and we now report of substrate is available in solution it can be transformed
on a study showing that satisfactory results can be obtained by competing enzymes working with lower stereoselectivity.
by performing the reductive transformations in the presence
Compound 5 was assigned (R) absolute configuration be-
of deuterated water, thus avoiding the troublesome synthe- cause it was converted by Zn/H2SO4/EtOH reduction[20]
sis of precursors which are regiospecifically labelled with into (R)-7 (Scheme 2), whose absolute configuration was
isotopic hydrogen.
known.[21] In a separate experiment, product 2 was submit-
At this exploratory stage, the enzymatic reactions were ted to catalytic hydrogenation with deuterium gas (AcOEt,
performed in fermenting baker’s yeast, since this whole cell moist 5% Pd/C) to provide the [2,3-2H2] derivative 8.
system contains all the three required enzymatic capacities,
consisting in the ability of a) exchanging the hydrogen
atoms of the reduced cofactor(s) with the deuterium atoms
of the solvent water,[10] b) saturating the double bond of the
activated olefin and c) reducing an aldehyde to the corre-
sponding alcohol, respectively.
Results and Discussion
Substrates of choice for this investigation were nitrostyr-
enes 1[17] and 2[18] and the structurally similar α-ethoxycin-
namaldehyde 3[19] (Scheme 1), not only because they are ef-
ficiently transformed by baker’s yeast into (R)-4, (R)-5 and
Scheme 2. i. Fermenting baker’s yeast + 1% D2O; ii. Zn/30%
(S)-6 of fair or good enantiomeric purity, respectively, but,
H2SO4/ethanol, 40–50 °C; iii. deuterium gas/10% Pd/C/ethyl acet-
most important, because products 4–6 display the 2H NMR
ate, room temp.
signals of the side chain hydrogen atoms with the dispersion
which is required to allow a proper analysis.
The natural abundance 2H NMR spectrum of 5 (Fig-
ure 1, a) shows separate signals for all the five hydrogen
atoms of the side chain. Subsequently, the exchangeability
of the hydrogen atoms of 5 with the solvent hydrogen atoms
was determined submitting the synthetic racemic material
to incubation at room temp. for three days with fermenting
baker’s yeast in the presence of ca. 1% deuterated water. In
this time interval, the pH of the mixture remained around
7. The spectrum of 5 recovered from the experiment (Fig-
ure 1, b) indicates, as judged from the increase of the signal
at δ = 4.36 ppm respect to that at δ = 3.39 ppm, relative to
the three deuterium atoms of the O-methyl group assumed
as internal standard (see Figure 1, a), incorporation of deu-
terium from the solvent in position α to the nitro group.
This result qualitatively confirms what observed when 2-
First, the (E) stereochemistry of the α-(hydroxymethyl)- nitro-1-phenylpropane was incubated with deuterated
nitrostyrene 2 was firmly established through a NOE ex- water.[16]
Scheme 1. Baker’s yeast-mediated asymmetric reduction of com-
pounds 1–3 to 4–6.
periment showing that the hydroxymethylene protons inter-
The baker’s yeast reduction of 2 was then performed in
act strongly with the ortho position of the phenylene ring, the presence of ca. 1% deuterated water, isolating labelled
2
while they do not display any contact with the vinylic hy- 5 with an ee value of ca. 40%. The H NMR spectrum of
drogen. Substrate 2 was adsorbed on nonpolar resins (XAD 5 obtained in these conditions (Figure 1, c) shows incorpo-
1180), and incubated with baker’s yeast actively fermenting ration of deuterium at positions 2 and 3 of the side chain.
5078
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5077–5084

Stereochemical Analysis of an Enzymatic Reduction
incorporation of deuterium in benzylic position as a conse-
quence of multiple isotope effects. Indeed, this species is
depleted because it derives from the reduced cofactor which
has been deuterium labelled through an enzyme-controlled
exchange with diluted deuterated water.[10] We were confi-
dent to be able to assign the deuterium benzylic signals of
2
5 comparing its H NMR spectrum (Figure 1, c) with that
of the chemically identical material produced from 2 by hy-
drogenation with deuterium gas in the presence of Pd/C in
2
the expectation of a syn process.[13] However, the H NMR
spectrum (Figure 1, d) of [2,3-2H2] 3-(4-methoxyphenyl)-2-
nitropropan-1-ol (8, Scheme 2), produced from 2 by treat-
ment with deuterium gas in AcOEt in the presence of moist
Pd/C, shows, in addition to the signal at δ = 4.36 ppm (posi-
tion 2), two peaks of equal intensity at δ = 2.62 and
2.81 ppm, respectively, for the two diastereotopic hydrogen
atoms in position 3. Also, there is a dramatic difference in
the extent of the labelling of the two positions (Figure 1,
d), specifically inverse respect to the one observed in com-
pound 5 bio-generated in baker’s yeast (Figure 1, c).
The labelling pattern of 8 could be the consequence of a
two steps mechanism, involving 1,4-deuterium addition
onto 2 to yield (Scheme 3) the nitronic acid derivative 9.
Subsequently, the sp2 carbon atom at position 2 of the latter
is protonated without steric control with the prevalent in-
1
corporation of H species from the medium. This hypothe-
sis explains both the lack of stereospecificity in the hydro-
genation and the difference in the deuterium incorporation
at the two sites and might be of relevance for the elucidation
of the mechanism of the enzymatic reduction of nitro al-
kenes.[16,22,23]
Figure 1. Deuterium NMR spectra in C6H6 of (a) unlabelled 5
(deuterium natural abundance spectrum) (b) 5 after incubation
with B. Y. in H2O/D2O (1%) (c) 5 obtained from 2 by B. Y. re-
duction in H2O/D2O (1%) (d) 8 obtained from 2 by catalytic re-
duction (D2-Pd/C). The deuterium chemical shifts of 5 in benzene
are slightly dependent on the sample concentration and thus the
signals show some variability throughout the spectra (within
0.1 ppm).
However, the signals at δ = 2.62 and 2.81 ppm, respectively,
relative to the diastereotopic hydrogen atoms in position 3,
appear in ca. 7:3 ratio, once depurated from the contri-
bution to the peak intensity of the natural abundance sig-
nal. This contribution was evaluated from the peak areas
Scheme 3. Steps in the chemical reduction with deuterium gas of 2
to provide randomly deuterated 8.
Thus, to unambiguously establish the relative stereo-
of the nuclei at C-1 and of the methoxy group. As expected, chemistry of the deuterium signals in positions 2 and 3 of
2
the intensity of the deuterium atom signal at δ = 4.36 ppm (R)-5 obtained in deuterated water, the H NMR spectrum
(position 2) is much higher than those at δ = 2.62 and of the derived amine 7 (Figure 2, a) was acquired and com-
2.81 ppm (position 3), in agreement with the fact that this pared with that (Figure 2, b) of the chemically identical syn-
hydrogen atom is thought to be delivered from the solvent thetic (2SR,3SR)-[2,3-2H2] amine 13. The latter was pre-
water in the reduction step and, moreover, it is incorporated pared (Scheme 4) from the (Z)-acetamido-cinnamic acid de-
from the medium in 5 once the reduction has taken place rivative 10[24,25] by catalytic hydrogenation with deuterium
(see Figure 1, c). It is instead reasonable to expect a lower gas in the presence of 10% Pd/C. The resulting (2SR,3SR)-
Eur. J. Org. Chem. 2010, 5077–5084
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5079

E. Brenna, G. Fronza, C. Fuganti, F. G. Gatti
FULL PAPER
[2,3-2H2] derivative 11, via the ester hydrochloride 12 and reductions performed in the presence of D2O as an analyti-
LiAlH4 reduction, provided the required stereospecifically cal tool for the determination of the stereochemical course
syn labelled amine 13.
of the process arose from experiments on the aldehyde (Z)-
3, which shows some structural similarity with nitrostyrene
derivative 2. The material, when administered to baker’s
yeast loaded onto an adsorption resin under conditions as-
suring a very low concentration of substrate in the fermen-
tation medium, is converted into enantiomerically pure
(2S)-6 (ee = 99%).[19] In the present work, using a reduced
volume due to the presence of deuterated water, at higher
substrate concentration, the reduced material 6 with ee =
92% was isolated.
Thus, aldehyde 3, prepared by MnO2 oxidation of allylic
alcohol 14,[19] administered to actively fermenting baker’s
yeast in the presence of ca. 4% deuterated water, afforded
labelled [1,2,3-3H2]-6, containing 96% (HPLC) of the (S)-
enantiomer (Scheme 5). In a separate experiment, allylic
alcohol 14 was converted into (2SR,3SR)-[2,3-2H2]-15 on
hydrogenation with deuterium gas in ethyl acetate in the
presence of wet 10% Pd/C.
Figure 2. Deuterium NMR spectra in CHCl3 of (a) 7 obtained from
5 by reduction with Zn/H2SO4 (b) 13 synthesized starting from 10
(see Scheme 4).
Scheme 5. i. MnO2, CH2Cl2, reflux, 3 h; ii. baker’s yeast/deuterated
water; iii. deuterium gas, ethyl acetate, 10% Pd/C, room temp.
The deuterium NMR spectrum of 6 (Figure 3, a) shows
significant signals for three of the five hydrogen atoms of
the side chain at δ = 3.52, 3.46 and 2.81 ppm, respectively.
Minor signals are at δ = 3.61 and 2.70 ppm. Comparison
of the spectrum of biogenerated 6 with that of syn-2,3-di-
deuterated 15 (Figure 3, b) clearly indicates that the two
materials are diastereoisomers. Thus, in 6 the deuterium
atom at position 3 is in anti relationship with that in 2. This
means that the biological hydrogenation of the double bond
Scheme 4. i. deuterium gas, 10% Pd/C, ethanol, room temp.; ii.
HCl (gas)/ethanol, reflux, 10 h; iii. aqueous NaHCO3/Et2O extrac-
tion, then LiAlH4 reflux, 12 h.
It thus clearly appears from the inspection of spectra in of 3 involves the addition of an hydrogen atom in position
parts a and b of Figure 2 that 7 is composed by the β respect to the activating carbonyl group from the upper
(2R,3R)-[2,3-2H2] diastereoisomer in ca. 7:3 mixture with part of the molecule, while α-protonation took place by
the (2S,3R)-[2,3-2H2] material.
96% in anti manner to yield (S)-6. The spectrum of 6 shows
Going back from 7 to 5 and to the (E)-unsaturated pre- deuterium labelling also in position 1, as expected for an
cursor 2 it follows that the enzymatic reduction of 2 to 5 alcohol generated in baker’s yeast from an aldehyde like 3.
involved the addition in position β respect to the nitro
In order to simplify the assignment and the quantitative
group of an hydrogen atom delivered from the upper face evaluation of the signals relative to position 1, the alcohol
of the molecule, while the protonation in α occurred preva- 6 was converted into the corresponding acetyl derivative
2
lently from the opposite site to yield ca. 70% of the (2R) and its H NMR spectrum is reported in Figure 3 (c). The
material.
signals for the two diastereotopic deuterium atoms in posi-
A further confirmation of the utility of 2H NMR studies tion 1 of the derivative of 6 appear at δ = 4.02 and 4.13 ppm
on the starting materials obtained in baker’s yeast-mediated in ca. 95:5 ratio, as that of the two deuterium atoms at
5080
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5077–5084

Stereochemical Analysis of an Enzymatic Reduction
= 4.47 ppm) and H-1b (δ = 4.68 ppm) of 4 show the follow-
ing homo- and heteronuclear vicinal coupling constants:
J(H1a,H2) = 8.7, J(H1b,H2) = 6.0, J(H1a,13CH3) = 3.3,
J(H1b,13CH3) = 5.3 Hz. Such values unequivocally indicate
that the most populated conformational state of the mole-
cule is that with the aromatic ring and the nitro group trans-
oriented where the protons H-1a and H-1b can be assigned
the pro-S and pro-R configuration, respectively.
Luckily enough, product 1, which was poorly reduced to
racemic 4 with Clostridium sporogenes extract,[17] in baker’s
yeast and D2O rapidly afforded 4, shown by HPLC analysis
to be a 98:2 mixture of two enantiomers. This material was
tentatively assigned the (R)-configuration because of the or-
der of elution on chiral HPLC, identical to that of (2R)-17
prepared from 16 with fermenting baker’s yeast in the pres-
2
ence of D2O (Scheme 6). H NMR studies show that the
recovered material (Figure 4, a) is (1RS,2R)-[1,2-2H2]-4
since there are two deuterium signals of equal intensity at δ
= 4.70 and 4.49 ppm relative to the hydrogen atoms in posi-
tion 1 and a signal at δ = 3.96 ppm for the deuterium atom
in position 2.
Figure 3. Deuterium NMR spectra in CHCl3 of (a) 6 obtained by
B. Y. reduction of the aldehyde 3 in H2O/D2O (4%) (b) 15 obtained
from the allylic alcohol 14 by catalytic reduction (D2-Pd/C) (c) ace-
tyl derivative of 6.
Scheme 6. i. Baker’s yeast/deuterated water; ii. NaBH4, EtOH; iii.
Baker’s yeast deuterated water.
In order to confirm these results, the reduction product
2
position 3. If we consider reasonable that the mode of yeast obtained from nitrostyrene 16 was submitted to H NMR
reduction of the carbonyl of the substituted cinnamal- analysis, and it was found to be (1RS,2R)-[1,2-2H2] 17 (Fig-
dehyde 3 is the same of cinnamaldehyde, i.e., addition onto ure 4, b) as indicated from the signals of equal intensity at δ
the carbonyl C atom of a pro-R deuterium atom mediated = 4.52 ppm and 4.47 ppm (position 1) and at δ = 3.60 ppm
by yeast alcohol dehydrogenase,[2] we can depict labelled 6 (position 2).
obtained in the baker’s yeast reduction of 3 in the presence
Finally, a blank experiment was performed, incubating
of deuterated water as substantially composed by the synthetic 17 with baker’s yeast in the presence of deuterated
(1R,2S,3R)-[1,2,3-2H3] diastereoisomer. In this instance, the water under the conditions used in the reduction of 1 and
2
stereopecific reduction of the aldehydic group of 3 by yeast 16. The H NMR spectrum (Figure 4, c) of the recovered
alcohol dehydrogenase with hydride delivery from deuter- product indicates that it is randomly labelled [1-2H]-2-(4-
ated reduced cofactor serves is an internal probe of the ste- methoxyphenyl)-1-nitropropane (17) with signals at δ =
reospecificity of the saturation of the double bond.
4.52 and 4.47 ppm (position 1). Thus, no mechanistic con-
The attention was finally drawn on the mode of re- clusions can be drawn from the mode of deuterium labelling
duction of 1 by baker’s yeast in the presence of deuterated at position 1 of the two nitro alkanes 4 and 17 generated in
water. The choice of 1 as a substrate was dictated from the the presence of deuterated water from 1 and 16, also at the
circumstance that the reported[17] spectrum of the saturated light of the results of the exchange experiments performed
racemic material 4 shows a signal pattern amenable for a early on synthetic racemic 5 (Figure 1, b) incubated with
full assignment. In fact the two methylene protons H-1a (δ deuterated water.
Eur. J. Org. Chem. 2010, 5077–5084
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5081

E. Brenna, G. Fronza, C. Fuganti, F. G. Gatti
FULL PAPER
tiomeric purity through anti addition of hydrogen atoms
across the double bond, with formal hydride delivery in po-
sition β with respect to the activating group from the “up-
per” side of the molecule.
Experimental Section
General Procedure for the Baker’s Yeast Reduction: In an open jar a
mixture was made up composed of baker’s yeast (200 g), -glucose
(50 g), tap water (1 L) and 99% deuterated water (20 mL). Under
mechanical stirring, at 35 °C, the indicated substrate (typically 1 g)
dissolved in the minimum amount of ethanol was added dropwise
within 10 min. After 24–48 h stirring at room temp., acetone, 0.5 L
was added at once. The mixture was stirred for 30 min and then
extracted three times with a total of 1.2 L of ethyl acetate/hexane,
9:1. The organic phase was evaporated to remove most of the ace-
tone and washed twice with brine. The residue obtained upon evap-
oration of the dried (Na2SO4) organic phase was column chromato-
graphed with increasing amounts of ethyl acetate in hexane. Typi-
cally, the recovery was 60–70%. When non-polar resins were used,
the substrate was dissolved in diethyl ether in 1:15 w/w ratio with
XAD-1180. The solvent was removed under reduced pressure and
the residue was added to fermenting Baker’s yeast.
(2R)-3-(4-Methoxyphenyl)-2-nitropropan-2-ol (5): Yeast reduction
of 2 yields 5 (60%) as a yellowish oil, which solidifies on standing.
1H NMR (400 MHz, C6D6): δ = 6.80 (m, 2 H, aromatic H), 6.66
(m, 2 H, aromatic H), 4.37 (m, 1 H, 2-H), 3.55 (dd, J = 12.2,
7.5 Hz, 1 H, 1-H), 3.40 (dd, J = 12.2, 3.4 Hz, 1 H, 1-H), 3.31 (s, 3
H, OCH3), 2.85 (dd, J = 14.1, 7.6 Hz, 1 H, 3-H), 2.64 (dd, J =
14.1, 6.9 Hz, 1 H, 3-H) ppm. 13C NMR (100.6 MHz, CDCl3): δ =
158.7, 129.8, 128.7, 114.1, 90.4, 62.4, 55.1, 34.9 ppm. 2H NMR
(76.7 MHz, C6H6): δ = 4.36 (2-D), 3.29 (OCH2D), 2.82 (3-D), 2.61
(3-D) (see Figure 1, c) ppm. [α]2D0 = +35.1 (c = 2.05, CHCl3), ee
(HPLC) = 70%. HRMS (ESI) 211.0849. C10H13NO4 requires
211.0844.
Figure 4. Deuterium NMR spectra of (a) 4 obtained from 1 by B.
Y. reduction in H2O/D2O (1%) (b) 17 obtained from 16 by B. Y.
reduction in H2O/D2O (1%) (c) 17 recovered after incubation of
unlabelled 2-(4-methoxyphenyl)-1-nitropropane with B. Y. in H2O/
D2O (1%).
[2,3-2H2]-3-(4-Methoxyphenyl)-2-nitropropan-2-ol (8): Product
2
Conclusion
(2 g, 10 mmol), in ethyl acetate (50 mL) was treated at room temp.
with deuterium gas in the presence of wet 5% Pd/C (150 mg). The
reaction was interrupted after the absorption of ca. 0.5 mol. equiv.
of gas. The residue obtained upon evaporation of the filtered solu-
tion was chromatographed, as above, to afford dideuterated 8 in
30% yield. 1H NMR (400 MHz, C6D6): δ = 6.78 (m, 2 H, aromatic
H), 6.76 (m, 2 H, aromatic H), 4.30 (m, 0.8 H, 2-H), 3.45 (m, 1 H,
1-H), 3.31 (m, 1 H, 1-H), 3.28 (s, 3 H, OCH3), 2.81 (m, 0.7 H, 3-
H), 2.59 (m, 0.7 H, 3-H) ppm. 2H NMR (76.7 MHz, C6H6): δ =
4.28 (2-D), 2.79 (D-3), 2.58 (3-D) (see Figure 1, d) ppm.
2
The present results demonstrate that H NMR studies
on materials obtained in the presence of deuterated water in
the enzymatic reduction of activated trisubstituted olefins
represent a direct analytical tool for the determination of
the steric course of the reduction process. Synthetic deuter-
ated materials obtained by syn metal-catalysed reduction
with deuterium gas were employed for comparison pur-
poses. A relevant prerequisite of the method is a dispersion
2
of the H NMR spectra signals of the relevant hydrogen
(2R)-2-Amino-3-(4-methoxyphenyl)propan-1-ol (7): (2R)-3-(4-Meth-
oxyphenyl)-2-nitropropan-1-ol (5) obtained in the yeast reduction
of 2 (0.6 g, 3 mmol), in ethanol (50 mL), in the presence of Zn
atoms large enough as to allow a proper analysis. For prac-
tical reasons, the procedure has been here applied using
baker’s yeast due to the presence in the whole cell system powder (5 g) under mechanical stirring was treated dropwise with
30% sulfuric acid (10 mL).[20] After 12 h stirring, the mixture was
of the capacity to exchange of an hydrogen of the reduced
filtered with suction and the solid washed repeatedly with ethanol.
cofactor for a deuterium atom. The operation with isolated
The organic phase was concentrated to a small volume, diluted
purified reducing enzymes would require the addition of the
with ice-water and extracted twice with ethyl acetate (discarded).
reduced deuterated cofactor or the presence of an enzy-
The cold aqueous phase was made strongly alkaline with 30%
matic machinery suitable for its generation. Finally, as far
NaOH and extracted with ethyl acetate (3ϫ100 mL). The residue
as the stereochemistry of the transformations here investi-
gated is concerned it is worth to mention that in baker’s
yeast 2 and 3, trisubstituted analogs of the tetrasubstituted
materials observed to be prevalently reduced in syn fash-
obtained upon evaporation of the dried organic extract was dis-
solved in ethyl acetate (5 mL), and treated with dry HCl in ethanol
(2 mL). After 24 h at 4 °C, upon filtration, the crystalline hydro-
chloride of (2R)-2-amino-3-(4-methoxyphenyl)propan-1-ol (10) was
ion,[14,15] are converted into (R)-5 and (S)-6 of good enan- obtained (0.42 g, 68%): 1H NMR (400 MHz, [D6]DMSO): δ = 7.18
5082
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5077–5084

Stereochemical Analysis of an Enzymatic Reduction
(m, 2 H, aromatic H), 6.86 (m, 2 H, aromatic H), 3.73 (s, 3 H,
added dropwise to LiAlH4 (4 g), in refluxing diethyl ether (100 mL)
OCH3), 3.50 (dd, J = 11.6, 3.5 Hz, 1 H, CHHOH), 3.38 (dd, J = under mechanical stirring in a N2 atmosphere. After 10 h under
11.6, 5.9 Hz, 1 H, CHHOH), 3.23 (m, 1 H, CHNH2), 2.89 (dd, J these conditions the cooled reaction mixture was treated with cau-
= 13.3, 5.3 Hz, ArCHH), 2.75 (dd, J = 13.3, 9.1 Hz, ArCHH) ppm. tion with a satd. solution of potassium sodium tartrate. The residue
13C NMR (100.6 MHz, [D6]DMSO) δ = 157.9, 130.1, 128.3, 113.8, obtained upon evaporation of the washed and dried organic phase
59.4, 54.9, 53.9, 33.7 ppm. [α]2D0 = –32.0 (c = 2, HCl 1 ) [lit. for
(S)-7 [α]2D0 = +48.8 (c = 2, aq. HCl)].[21] In the deuterated series
from yeast, the free amine 7 for the 2H NMR studies was obtained
as follows. The crude ethyl acetate extract of the Zn reduction,
dissolved in the minimum amount of ethyl acetate was treated with
an equal weigh of oxalic acid dissolved in ethyl acetate. The mixture
was left aside 24 h at room temp. The white precipitate was col-
lected and treated in a two phase system CH2Cl2/water with an
excess of aqueous potassium carbonate. The organic phase was sep-
arated, dried and evaporated to furnish the amine 7. 1H NMR
(500 MHz, CDCl3): δ = 7.11 (m, 2 H, aromatic H), 6.86 (m, 2 H,
aromatic H), 3.80 (s, 3 H, OCH3), 3.63 (dd, J = 10.7, 4.0 Hz, 1 H,
1-H), 3.37 (dd, J = 10.7, 7.3 Hz, 1 H, 1-H), 3.08 (m, 1 H, 2-H),
2.74 (dd, J = 13.7, 5.4 Hz, 1 H, 3-H), 2.48 (dd, J = 13.7, 8.6 Hz, 1
in a small volume of ethyl acetate was treated with oxalic acid (1 g)
in ethyl acetate. The separated crystals collected by filtration were
treated in CH2Cl2/water, as above, with potassium carbonate to
provide, on evaporation of the solvent, (2SR,3SR)-[2,3-2H2]-2-
amino-3-(4-methoxyphenyl)propan-1-ol (13), thick oil, 0.5 g (27%).
1H NMR (500 MHz, CDCl3): δ = 7.11 (m, 2 H, aromatic H), 6.86
(m, 2 H, aromatic H), 3.80 (s, 3 H, OCH3), 3.64 (m, 1 H, 1-H),
3.38 (m, 1 H, 1-H), 3.09 (m, 0.6 H, 2-H), 2.74 (m, 0.7 H, 3-H),
2
2.49 (m, 1 H, 3-D) ppm. H NMR (76.7 MHz, CHCl3): δ = 3.09
(2-D), 2.74 (3-D) (see Figure 2, b) ppm.
(2S)-2-Ethoxy-3-(4-methoxyphenyl)propan-1-ol (6): Yeast reduction
1
of 3 provided 6 as a yellow oil, in 65% yield. H NMR (400 MHz,
CDCl3): δ = 7.12 (m, 2 H, aromatic H), 6.83 (m, 2 H, aromatic H),
3.78 (s, 3 H, OCH3), 3.61 (dd, J = 11.0, 3.8 Hz, 1 H, H-1), 3.57–
3.49 (m, 3 H, H-2 + OCH2), 3.45 (dd, J = 11.0, 6.2 Hz, 1 H, H-
1), 2.81 (dd, J = 13.5, 5.9 Hz, 1 H, H-3), 2.70 (dd, J = 13.5, 6.9 Hz,
1 H, H-3), 2.33 (s br, 1 H, OH) ppm. 2H NMR (76.7 MHz, CHCl3)
δ = 3.79 (OCH2D), 3.61 (1-D), 3.52 (2-D), 3.46 (1-D), 2.81 (3-D),
2.71 (3-D) (see Figure 3, a) ppm. 13C NMR (100.6 MHz, CDCl3)
δ = 158.0, 130.2, 130.1, 113.7, 81.1, 65.2, 63.6, 55.2, 36.4, 15.5 ppm.
[α]2D0 = +3.8 (c = 3.1, CHCl3), ee (HPLC) = 92%; lit. ref. [19]: [α]2D0
= +4.0 (c = 3.5, CHCl3) for (S)-6 with ee (HPLC) = 99%. This
material (0.3 g) was treated with acetic anhydride (2 mL) and pyr-
idine (2 mL) overnight at room temp. The reagents were taken to
dryness at the rotary evaporator at 50 °C. The oily residue was kept
under high vacuum for 10 h to provide the acetyl derivative used
for the 2H NMR studies. 1H NMR (400 MHz, CDCl3): δ = 7.12
(m, 2 H, aromatic H), 6.81 (m, 2 H, aromatic H), 4.12 (dd, J =
11.5, 4.1 Hz, 1 H, 1-H), 4.00 (dd, J = 11.5, 5.7 Hz, 1 H, 1-H), 3.79
(s, 3 H, OCH3), 3.63 (m, 1 H, 2-H), 3.58–3.41 (m, 2 H, OCH2),
2.80 (dd, J = 13.9, 7.0 Hz, 1 H, 3-H), 2.74 (dd, J = 13.9, 5.9 Hz, 1
2
H, 3-H), 1.70 (s br, 1 H, OH) ppm. H NMR (76.7 MHz, CHCl3):
δ = 3.82 (OCH2D), 3.08 (2-D), 2.75 (3-D), 2.49 (3-D) (see Figure 2,
a) ppm.
(2SR,3SR)-[2,3-2H2]-2-Amino-3-(4-methoxyphenyl)propan-1-ol (13):
The acetamido-cinnamic acid 10 was obtained[24] from acetyl gly-
cine and anisaldehyde as sparingly soluble brownish solid. 1H
NMR (400 MHz, [D6]DMSO): δ = 7.59 (m, 2 H, aromatic H), 7.23
(s, 1 H, CH=C), 6.96 (m, 2 H, aromatic H), 3.79 (s, 3 H, OCH3),
1.99 (s, 3 H, NHCOCH3) ppm. 13C NMR (100.6 MHz, [D6]-
DMSO) δ = 168.8, 166.4, 159.8, 131.4, 126.1, 124.9, 116.9, 113.9,
55.1, 51.1, 22.3 ppm. A suspension of 10 (4.7 g, 50 mmol), in ethyl
acetate (50 mL) was treated whilst stirring at room temp. with deu-
terium gas until the gas absorption ceased. The filtered solution
was taken to dryness and the residue was crystallized from ethanol/
ethyl acetate to afford 2,3-dideuterated 2-(acetylamino)-3-(4-meth-
1
oxyphenyl)propionic acid (11) quantitatively. H NMR (400 MHz,
CDCl3): δ = 7.10 (m, 2 H, aromatic H), 6.86 (m, 2 H, aromatic H),
5.83 (s br, 1 H, NH), 4.82 (m, 0.6 H, 2-H), 3.80 (s, 3 H, OCH3),
3.18 (m, 0.65 H, 3-H), 3.10 (m, 1 H, 3-H), 2.00 (s, 3 H, COCH3)
ppm. 2H NMR (76.7 MHz, CHCl3) δ = 4.82 (2-D), 3.18 (3-D) ppm.
13C NMR (125.7 MHz, [D6]DMSO, the molecule was partially
deuterated and thus the signals of some C atoms are split in a few
peaks due to the deuterium isotope effect) δ = 173.2, 169.2, 157.9
130.0, [129.52, 129.49, 129.47 (C-1 aromatic quaternary C atom)],
113.6, 54.9, [53.73, 53.68, 53.43 (t, J = 22.0 Hz, C-2)], [35.97, 35.89,
35.61 (m br, C-3)], 22.3 ppm. Product 11 was treated at reflux under
mechanical stirring with ethanol saturated with HCl gas (50 mL).
After 10 h the reaction mixture was evaporated to dryness and the
residue was crystallized from hot ethanol to provide ethyl [2,3-2H2]-
2-amino-3-(4-methoxyphenyl)propionate hydrochloride (12) (3.2 g,
2
H, 3-H), 2.06 (s, 3 H, COCH3), 1.12 (t, J = 7.1 Hz, CH3) ppm. H
NMR (76.7 MHz, CHCl3): δ = 4.13 (1-D), 4.00 (1-D), 3.78
(OCH2D), 3.62 (2-D), 2.81 (3-D) (see Figure 3, c) ppm.
(2SR,3SR)-[2,3-2H2]-2-Ethoxy-3-(4-methoxyphenyl)propan-1-ol
(15): 2-Ethoxy-3-(4-methoxyphenyl)-2-propen-1-ol (14) (0.4 g, 2
mmol) in ethyl acetate (40 mL) was treated with deuterium gas at
room temp. in the presence of wet 10% Pd/C (100 mg). At the end
of the hydrogenation the reaction mixture was filtered through a
Celite® pad to provide on evaporation of the solvent the desired
compound quantitatively. 1H NMR (400 MHz, CDCl3): δ = 7.12
(m, 2 H, aromatic H), 6.84 (m, 2 H, aromatic H), 3.79 (s, 3 H,
OCH3), 3.63–3.42 (m, 4.2 H, 2-H, CH2OH, OCH2), 2.82 (m, 1 H,
3-H), 2.69 (m, 0.2 H, 3-H), 2.13 (s br, 1 H, OH), 1.18 (t, J = 7.1 Hz,
1
65% from 10). H NMR (400 MHz, CDCl3): δ = 8.70 (s br, 3 H,
2
CH3) ppm. H NMR (76.7 MHz, CHCl3): δ = 3.53 (2-D), 2.70 (3-
NH3+), 7.23 (m, 2 H, aromatic H), 6.84 (m, 2 H, aromatic H), 4.32
(m, 0.6 H, 2-H), 4.16 (q, J = 7.1 Hz, 2 H,OCH2), 3.76 (s, 3 H,
OCH3), 3.39 (m, 1 H, 3-H), 3.31 (m, 0.75 H, 3-H), 1.21 (t, J =
7.1 Hz, 3 H, CH3) ppm. 2H NMR (76.7 MHz, CHCl3) δ = 4.32
(2-D), 3.32 (3-D) ppm. 13C NMR (125.7 MHz, [D6]DMSO, the
molecule was partially deuterated and thus the signals of some C
atoms are split in a few peaks due to the deuterium isotope effect)
δ = 168.9, 158.5, 130.5, [126.49, 126.45 (C-1 aromatic quaternary
C atom)], 113.9, 61.4, 55.0, [53.36, 53.31, 53.06 (t, J = 22.0 Hz, C-
2)], [35.02, 34.94, 34.67 (m br, C-3)], 13.8 ppm. Product 12 (2.6 g,
10 mmol), was suspended in diethyl ether/water, 1:1 (100 mL) and
treated whilst stirring with satd. NaHCO3 solution at alkaline pH.
The organic phase was washed with brine, dried (Na2SO4) and
D) (see Figure 3, b) ppm.
(2R)-2-(2-Methoxyphenyl)-1-nitropropane (4):[17] The product was
obtained in baker’s yeast from 1 in 65% yield. 1H NMR (500 MHz,
CDCl3): δ = 7.26 (m, 1 H, aromatic H), 7.17 (m, 1 H, aromatic H),
6.94 (m, 1 H, aromatic H), 6.90 (m, 1 H, aromatic H), 4.69 (dd, J
= 11.9, 6.0 Hz, 1 H, 1-H), 4.47 (dd, J = 11.9, 8.7 Hz, 1 H, 1-H),
3.96 (m, 1 H, 2-H), 3.87 (s, 3 H, OCH3), 1.39 (d, J = 7.1 Hz, CH3)
ppm. 2H NMR (76.7 MHz, CHCl3) δ = 4.70 (1-D), 4.49 (1-D), 3.96
(2-D), 3.89 (OCH2D) (see Figure 4, a) ppm. 13C NMR
(100.6 MHz, CDCl3): δ = 157.1, 128.8, 128.5, 127.4, 120.9, 110.9,
80.5, 55.3, 33.5, 17.1 ppm. [α]2D0 = +6.4 (c = 2.1, CHCl3); ee (HPLC)
= 96%.
Eur. J. Org. Chem. 2010, 5077–5084
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5083

E. Brenna, G. Fronza, C. Fuganti, F. G. Gatti
FULL PAPER
(2R)-2-(4-Methoxyphenyl)-1-nitropropane (17):[17] The material ob-
tained in baker’s yeast from 16: H NMR (500 MHz, CDCl3): δ =
[8] G. Fronza, C. Fuganti, P. Grasselli, M. Barbeni, Tetrahedron
Lett. 1992, 33, 6375–6378.
[9] G. Fronza, C. Fuganti, P. Grasselli, A. Mele, A. Sarra, G. Alle-
grone, M. Barbeni, Tetrahedron Lett. 1993, 34, 6467–6470.
[10] H. Simon, M. Kellner, H. Guenther, Angew. Chem. Int. Ed.
Engl. 1973, 12, 146–147.
[11] D. J. Bougioukou, J. D. Stewart, J. Am. Chem. Soc. 2008, 130,
7655–7658.
[12] M. Hall, C. Stueckler, H. Ehammer, E. Pointner, G. Ober-
dorfer, K. Gruber, B. Hauer, R. Stuermer, W. Kroutil, P. Mach-
eroux, K. Faber, Adv. Synth. Catal. 2008, 350, 411–418.
[13] R. Stuermer, B. Hauer, M. Hall, K. Faber, Curr. Opin. Chem.
Biol. 2007, 11, 213–213.
1
7.15 (m, 2 H, aromatic H), 6.88 (m, 2 H, aromatic H), 4.51 (dd, J
= 11.8, 7.1 Hz, 1 H, 1-H), 4.46 (dd, J = 11.8, 8.0 Hz, 1 H, 1-H),
3.80 (s, 3 H, OCH3), 3.60 (sextet, J = 7.2 Hz, 1 H, 2-H), 1.37 (d, J
2
= 7.1 Hz, 3 H, CH3) ppm. H NMR (76.7 MHz, CHCl3) δ = 4.52
(1-D), 4.47 (1-D), 3.81 (OCH2D), 3.60 (2-D) (see Figure 4, b) ppm.
13C NMR (100.6 MHz, CDCl3): δ = 158.9, 132.9, 127.9, 114.4,
82.1, 55.3, 37.9, 18.8 ppm. [α]2D0 = +60.5 (c = 1.2, CHCl3), ee
(HPLC) = 88%; ref.[17]: [α]2D0 = +66.2 (c = 0.99, CHCl3, 30 °C) for
(R)-17 ee = 97%.
[14] Y. Kawai, Y. Inaba, N. Tokitoh, Tetrahedron: Asymmetry 2001,
12, 309–318.
(2RS)-2-(4-Methoxyphenyl)-1-nitro propane (17):[17] This material,
prepared as reported,[17] was recovered in 70% yield from the incu-
bation in fermenting baker’s yeast in the presence of deuterated
water. 2H NMR (76.7 MHz, CHCl3) δ = 4.52 (1-D), 4.47 (1-D),
[15] G. Fronza, C. Fuganti, S. Serra, Eur. J. Org. Chem. 2009, 6160–
6171.
[16] H. S. Toogood, A. Fryszkowska, V. Hare, K. Fisher, A. Roujei-
nikova, D. Leys, J. M. Gardiner, G. M. Stephens, N. S. Scrut-
ton, Adv. Synth. Catal. 2008, 350, 2789–2803; BY reduction of
other nitrostyrenes: H. Ohta, K. Ozaki, G.-I.-Tsuchihashi,
Chem. Lett. 1987, 191; H. Ohta, N. Kobayashi, K. Ozaki, J.
Org. Chem. 1989, 54, 1802–1804.
[17] A. Fryszkowska, K. Fisher, J. M. Gardiner, G. M. Stephens, J.
Org. Chem. 2008, 73, 4295–4298.
[18] N. Rastogi, I. N. N. Namboothiri, M. Cojocaru, Tetrahedron
Lett. 2004, 45, 4745–4748.
1
3.81 (OCH2D), 3.60 (2-D) (see Figure 4, c) ppm. The H and 13C
NMR spectra were in accordance with those of compound (2R)-
17.
Supporting Information (see also the footnote on the first page of
2
this article): General experimental methods; acquisition of the H
spectra; spectroscopic data of compounds 2, 3, 1 and 16.
[19] E. Brenna, C. Fuganti, F. G. Gatti, F. Parmeggiani, Tetrahe-
dron: Asymmetry, in press.
[20] F. W. Hoover, H. B. Hass, J. Org. Chem. 1947, 12, 506–509.
[21] M. Abarbri, A. Guignard, M. Lamant, Helv. Chim. Acta 1995,
78, 109–121.
[1] C. Fuganti, P. Grasselli, Chem. Ind. (London) 1977, 983.
[2] C. Fuganti, D. Ghiringhelli, P. Grasselli, J. Chem. Soc., Chem.
Commun. 1975, 846–847.
[3] G. Fronza, C. Fuganti, M. Pinciroli, S. Serra, Tetrahedron:
Asymmetry 2004, 53, 9383–9388.
[4] G. Fronza, C. Fuganti, P. Grasselli, A. Mele, J. Org. Chem.
1991, 56, 6019–6023.
[5] G. Fronza, C. Fuganti, P. Grasselli, S. Lanati, R. Rallo, S. Tchi-
libon, J. Chem. Soc. Perkin Trans. 1 1994, 2927–2930.
[6] G. Fogliato, G. Fronza, C. Fuganti, S. Lanati, R. Rallo, R.
Rigoni, S. Servi, Tetrahedron 1995, 51, 10231–10240.
[7] G. Fronza, C. Fuganti, M. Mendozza, R. Rallo, G. Ottolina,
D. Joulain, Tetrahedron 1996, 52, 4041–4052.
[22] Y. Meah, V. Massey, Proc. Natl. Acad. Sci. USA 2000, 97,
10733–10735.
[23] A. D. N. Vaz, S. Chakraborty, V. Massey, Biochemistry 1995,
34, 4246–4256.
[24] E. Erlenmeyer, Justus Liebigs Ann. Chem. 1893, 275, 1–8.
[25] M. Cutolo, V. Fiandanese, F. Naso, O. Sciacovelli, Tetrahedron
Lett. 1983, 24, 4603–4606.
Received: April 1, 2010
Published Online: July 26, 2010
5084
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5077–5084