Synthesis of optically active cyanohydrins from aromatic ketones: evidence of an increased substrate range and inverted stereoselectivity for the hydroxynitrile lyase from Linum usitatissimum

Article abstract of DOI:10.1016/j.tetasy.2006.12.026

The synthesis of (R)- and (S)-cyanohydrins from a range of aromatic methyl and ethyl ketones, including the first examples from substituted variants of phenylacetone, benzylacetone and propiophenone, is described. Commercially available hydroxynitrile lya

Full text of DOI:10.1016/j.tetasy.2006.12.026

Tetrahedron: Asymmetry 18 (2007) 208–214
Synthesis of optically active cyanohydrins from
aromatic ketones: evidence of an increased substrate range
and inverted stereoselectivity for the hydroxynitrile lyase
from Linum usitatissimum
Christopher Roberge,* Fred Fleitz, David Pollard and Paul Devine
Merck Research Laboratories, Merck & Co, Inc., Rahway, NJ 07065, USA
Received 20 November 2006; accepted 27 December 2006
Abstract—The synthesis of (R)- and (S)-cyanohydrins from a range of aromatic methyl and ethyl ketones, including the first examples
from substituted variants of phenylacetone, benzylacetone and propiophenone, is described. Commercially available hydroxynitrile lyase
(HNL) enzymes were used to catalyze the asymmetric addition of cyanide to the ketones, including the first successful application of the
flax HNL (LuHNL) to the synthesis of any aromatic (S)-cyanohydrin. Both reaction yields and stereoselectivities were shown to be influ-
enced by the carbon chain length between the ketone and phenyl functional groups, and the type of aromatic substitution present on the
starting material. Substrates converted with the greatest degree of productivity and selectivity were phenylacetones with large, electron
withdrawing meta-substituents, such as 3-Cl, 3-Br and 3-CF3 phenylacetones, from which cyanohydrins are formed with 93–99% ee and
61–71% yield.
ꢀ 2007 Elsevier Ltd. All rights reserved.
1. Introduction
ing materials.2,3 Asymmetric conversion of ketones to
tertiary cyanohydrins has however proven to be much
more challenging, in large part due to the lower reactivity
of ketone compounds relative to that of analogous alde-
hydes, and the higher steric congestion adjacent to the car-
bonyl group adding to the difficulty of enantiospecific
cyanide addition. The past decade has seen many advances
in the discovery and development of chemical catalysts
applicable to the asymmetric formation of cyanohydrins
from ketone starting materials.2,4 Despite these improve-
ments, there are still significant obstacles to the practical
application of these chemical catalysts in several cases.
There are often long synthetic routes, with as many as thir-
teen linear steps, necessary for the production of metallic
bifunctional catalysts.5 The reaction conditions frequently
involve extreme reaction conditions, with temperatures as
low as À80 ꢁC.4 Also, the catalyst loading is typically high,
with as much as 20 mol % of the ligand required for the
production of cyanohydrin with adequate chiral purity.2
Finally, most applications of chemical catalysts produce
the trimethylsilylated product rather than the true cyano-
hydrin. Protection of the alcohol functionality prevents
subsequent single-step acylations, esterifications or Mitsu-
nobu reactions that have been demonstrated in the syntheses
Asymmetric cyanation of carbonyl compounds has become
an increasingly important component of enantiospecific
organic synthesis schemes. The cyanohydrin products of
these reactions are of great industrial significance, because
of the large number and wide variety of biologically active
chiral compounds that can be synthesized from these 2-
hydroxynitriles with tertiary or quaternary stereocentres.1
By derivatizing either or both of the alcohol and nitrile
functional groups of these compounds, useful synthetic
building blocks including, but not limited to, ketones,
amines, esters and nitriles can be formed without a loss
of enantiomeric purity.
An increasing array of catalysts, including cyclic dipep-
tides, transition metal complexes and enzymes, have been
identified as being applicable for the stereospecific genera-
tion of cyanohydrins from a wide variety of aldehyde start-
*
Corresponding author. Tel.: +1 908 594 5059; fax: +1 908 594 5091;
e-mail: christopher_roberge@merck.com
0957-4166/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2006.12.026

C. Roberge et al. / Tetrahedron: Asymmetry 18 (2007) 208–214
209
of optically active a-amino and tetronic acids from cyano-
2. Results and discussion
hydrin building blocks.6–8
As a first step in screening a large number of aromatic
ketones to identify substrates suitable for enantiospecific
cyanohydrin formation with LuHNL biocatalysis, several
monosubstituted derivatives of acetophenone 1a (Scheme
1), phenylacetone 1b and benzylacetone 1c were screened
under a set of standard reaction conditions.16 The results
of this substrate screen are summarized in Table 1.
Alternative catalysts are the hydroxynitrile lyase (HNL)
enzymes.9–12 These biocatalysts are particularly useful in
synthesizing true cyanohydrins while functioning at rela-
tively mild conditions and exhibiting a broader substrate
range and greater degree of enantioselectivity than analo-
gous chemical catalysts. As with chemical catalysis, the
large majority of published examples of HNL-catalyzed
cyanohydrin formation involve the conversion of aromatic
or aliphatic aldehyde compounds that share much struc-
tural similarity with the enzymes’ natural substrates.10
Additionally, there is a growing body of literature describ-
ing processes using HNLs to synthesize aliphatic ketone
cyanohydrins.13–15 However, the current number of refer-
ences to the biocatalytic conversion of methyl phenyl or
ethyl phenyl ketones to cyanohydrins is surprisingly small.
Herein, we survey a greater spectrum of aromatic ketone
substrates for HNL biocatalysis than has been discussed
in the current literature, and demonstrate for the first time
that HNL isolated from flax (LuHNL) is capable of con-
verting this class of compounds into optically enriched cya-
nohydrins, some with ee values as high as 99%. Among
these are what we believe to be the first descriptions of chi-
ral cyanohydrin synthesis from substituted derivatives of
phenylacetone, benzylacetone and propiophenone. Surpris-
ingly, these reactions are shown to not demonstrate the
exclusive (R)-selectivity that has been a feature of all previ-
ously known transformations catalyzed by LuHNL. The
activities of two additional commercially available en-
zymes, PaHNL (isolated from almonds) and MeHNL (iso-
lated from cassava), towards these aromatic ketone
substrates are also discussed.
Examples:
CN
R2
R2
a: n = 0, R1 = H, R2 = CH3
b: n = 1, R1 = H, R2 = CH3
c: n = 2, R1 = H, R2 = CH3
n
n
OH
O
R1
R1
1
2
Scheme 1. Synthesis of cyanohydrins from methyl phenyl and ethyl
phenyl ketones.
The first major result of this substrate screen was that the
cyanohydrins formed with the LuHNL biocatalyst are of
an (S)-configuration. This is in contrast to all prior litera-
ture descriptions of the enzyme, involving the exclusive
transformation of aliphatic aldehydes and ketones into
(R)-cyanohydrins. Additionally, although recent discover-
ies in the field of chemocatalysis have led to processes of
varying practicality for the production of enantiomerically
enriched cyanohydrins from benzylacetone,17,18 aceto-
phenone19,20 and its substituted forms,19–21 we believe this
to be the first description of the formation of chiral cyano-
hydrins from substituted forms of phenylacetone.
Table 1. Conversion of methyl phenyl ketones with LuHNL at 20 ꢁCa
Entry
Ketone
Yieldb (%)
eeb (%)
Configurationc
1
2
3
4
5
R = H
R = 4-F
R = 4-Br
R = 4-CH3
R = 4-CH3O
6
0
0
0
0
<10
—
—
—
—
Racemic
O
—
—
—
—
R
R
6
7
8
9
10
11
12
13
14
R = 2-Br
R = 3-F
R = 3-Cl
R = 3-Br
R = 3-CH3
R = 3-CF3
R = 3-CH3O
R = 4-Br
3
47
33
62d
61
44
45
34
59
<10
59
78
93
91
95
77
76
66
Racemic
S
S
S
S
S
S
S
S
O
R = 4-CH3O
O
15
16
17
R = H
R = 4-OH
R = 4-CH3O
15
41
10
40
36
56
S
S
S
R
a In each reaction, 30 lL ketone substrate, 720 lL of 0.1 M citrate (pH 4.5), 100 lL LuHNL, 100 lL DIPE and 50 lL trimethylsilyl cyanide were
combined and incubated for 17 h at 20 ꢁC.
b Determined by normal phase chiral HPLC analysis.
c The absolute configuration was determined by comparison with reported data for (S)-cyanohydrins produced by MeHNL from acetophenone,
phenylacetone and benzylacetone.22
d Confirmed with isolated yield value from 1 g reaction.

210
C. Roberge et al. / Tetrahedron: Asymmetry 18 (2007) 208–214
0
-0.5
-1
A general trend that is apparent from the data involves the
effect of the distance between the phenyl and ketone func-
tional groups of a compound on the activity of the HNL
enzyme towards the substrate. The biocatalyst was shown
to be most active towards derivatives of phenylacetone (en-
tries 6–14), converting 30–65% of 8 out of 9 tested starting
materials to the (S)-cyanohydrin with 55–95% ee in less
than 1 day. Derivatives of benzylacetone (entries 15–17)
were found to be poorer substrates for the enzyme, yielding
conversion rates and product ee values of <40%. LuHNL
exhibited the poorest activity towards acetophenone and
its monosubstituted forms (entries 1–5), converting very
little of these compounds to cyanohydrin product with
negligible enantiomeric enhancement. As further specific
examples involving identical substitution patterns, the con-
version of 4-bromophenylacetone (entry 13) gave much
better results than that of 4-bromoacetophenone (entry
3), which did not show appreciable reactivity. Also, conver-
sion of 4-methoxyphenylacetone (entry 14) gave a signifi-
cantly higher reaction rate and cyanohydrin ee than
conversion of 4-(4-methoxyphenyl)-2-butanone (entry 17),
which in turn was a better substrate than 4-methoxyace-
tophenone (entry 5). These results share a similar pattern
with those previously published for biocatalysis with the
MeHNL enzyme, in which higher cyanohydrin yields and
ee values were obtained for the conversion of unsubstituted
phenylacetone than that of either the smaller acetophenone
or the larger benzylacetone,22 suggesting that there may be
an optimal ketone chain length for HNL biocatalysis.
50
60
-F
70
80
90
100
-CH3O
-Br
-Cl
-1.5
-2
-CH3
-2.5
-3
-CF3
Cyanohydrin ee (%)
Figure 1. Correlation between Taft steric parameter of aromatic methyl
ketone substitution and resulting ee of cyanohydrin formed via LuHNL
biocatalysts. Reaction conditions are the same as shown in Table 1.
of the substrate into the active site cavity of the enzyme,
enhancing the ability of the biocatalyst to discriminate
between enantiomers in the process of adding cyanide to
the ketone.
In addition to steric effects, the electronic nature of the sub-
stituents tested also likely plays a role in determining the
enantioselectivity of the reactions. For the ketones that
contain halogenated substituents, there is a monotonically
increasing trend in cyanohydrin ee as the electronic with-
drawing ability of the substituents is increased. This can
be seen from the data listed in Table 1 for entries 7–9
and 11, describing the production of 59%, 78%, 93% and
95% ee cyanohydrins from phenylacetones with 3-F, 3-Cl,
3-Br and 3-CF3 substitutions, respectively.
Other steric effects of the substrate on the behavior of the
LuHNL enzyme can be seen when considering the location
and size of the substitutions to the aromatic ring. When
bromine is present at the ortho-position of phenylacetone
(entry 6), there is no significant production of cyanohydrin.
Phenylacetone with bromine at the para-position (entry
13), though, is converted with an overnight yield of 34%
to cyanohydrin with an ee of 76%. An analogous observa-
tion was made for the conversion of aldehyde substrates by
both PaHNL and PmHNL (isolated from Japanese apri-
cot). In the hydrocyanation of aldehydes catalyzed by those
enzymes, the presence of substituents at the ortho-position
always resulted in lower reaction rates and enantioselectiv-
ities than those seen with identical substituents at either the
meta- or para-positions.23 The best results for LuHNL-cat-
alyzed hydrocyanation of brominated derivatives of phenyl-
acetone were seen for the case of the meta-substitution
(entry 9), which gave a yield nearly double that seen with
the para-substitution and an excellent ee of 93%. Similarly,
in comparing a pair of methoxy substituted phenylace-
tones, the ee of the cyanohydrin generated from the m-
CH3O ketone (entry 12) is higher than that generated from
the p-CH3O substrate (entry 14).
A further improvement to the biocatalytic reaction pro-
cesses of Table 1 is realized by decreasing the reaction tem-
perature from 20 to 5 ꢁC. The result is an increase in overall
chiral selectivity, as is seen in the data presented in Table 2.
It is likely that this temperature reduction acts to suppress
the kinetics of the background reaction of spontaneous,
racemic cyanide addition more than those of the enzymatic
reaction.26 Under the new conditions, LuHNL is seen in
particular to produce cyanohydrin from several of the aro-
matic methyl ketones (entries 3–7) at very high ee values of
93–99%.
This revised set of reaction parameters was then applied to
the conversion of methyl phenyl ketones by two additional
commercially available HNL enzymes—MeHNL and
PaHNL. Results from these conversions are given in Table
3. A comparison of the applicability of these two enzymes
along with that of LuHNL shows that for most cases, high-
er activities towards the starting materials make MeHNL
the preferred biocatalyst. Notable exceptions are the ben-
zylacetones, the cyanohydrins of which are still generated
at higher yields through MeHNL catalysis (Table 3, entries
15–17), but with much greater stereoselectivity through
LuHNL catalysis (Table 2, entries 9–11). For example,
LuHNL-catalyzed conversion of unsubstituted benzylace-
tone and 4-(4-methoxyphenyl)-2-butanone gives products
with very high ee values of 89% and 99%, respectively,
whereas the substrates are converted with good yields by
MeHNL but to cyanohydrin that is nearly racemic.
Within a given substitution position, the relative size of the
substituent also appears to influence the stereoselectivity of
the enzymatic reaction. This effect can be seen in a plot of
the Taft steric parameter24,25 versus cyanohydrin ee for all
substituents tested, as is shown in Figure 1, illustrating a
clear correlation between the substituent size and biocata-
lytic performance. One potential explanation for this phe-
nomenon is that larger groups may result in a tighter fit

C. Roberge et al. / Tetrahedron: Asymmetry 18 (2007) 208–214
Table 2. Conversion of methyl phenyl ketones with LuHNL at 5 ꢁCa
211
Entry
Ketone
Yieldb (%)
eeb (%)
Configurationc
1
2
3
4
5
6
7
8
R = 3-F
R = 3-Cl
R = 3-Br
R = 3-CH3
R = 3-CF3
R = 3-CH3O
R = 4-Br
20
24
37
40
19
31
9
83
89
99
97
96
99
93
84
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
O
R
R
R = 4-CH3O
30
O
9
10
11
R = H
R = 4-OH
R = 4-CH3O
2
13
1
89
75
99
(S)
(S)
(S)
a In each reaction, 30 lL ketone substrate, 720 lL of 0.1 M citrate (pH 4.5), 100 lL LuHNL, 100 lL DIPE and 50 lL trimethylsilyl cyanide were
combined and incubated for 17 h at 5 ꢁC.
b Determined by normal phase chiral HPLC analysis.
c Absolute configuration was determined by comparison with reported data for (S)-cyanohydrins produced by MeHNL from acetophenone, pheny-
lacetone and benzylacetone.22
Table 3. Conversion of methyl phenyl ketones with MeHNL and PaHNL at 5 ꢁCa
Entry
Ketone
MeHNL
eeb (%)
PaHNL
eeb (%)
Yieldb (%)
Configurationc
Yieldb (%)
Configurationc
1
2
3
4
5
R = H
R = 4-F
R = 4-Br
R = 4-CH3
R = 4-CH3O
9
0
0
0
0
54
—
—
—
—
(S)
—
—
—
—
8
0
0
0
0
97
—
—
—
—
(R)
—
—
—
—
O
R
R
6
7
8
9
10
11
12
13
14
R = 2-Br
R = 3-F
R = 3-Cl
R = 3-Br
R = 3-CH3
R = 3-CF3
R = 3-CH3O
R = 4-Br
44
70
71
61
65
67
53
77
62
97
79
93
93
88
97
92
90
47
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
(S)
1
2
1
0
0
0
0
3
2
86
20
44
—
—
—
—
29
<10
(R)
(R)
(R)
—
—
—
—
(R)
Racemic
O
R = 4-CH3O
O
15
16
17
R = H
R = 4-OH
R = 4-CH3O
61
54
68
7
70
14
(S)
(S)
(S)
2
0
0
55
—
—
(R)
—
—
R
a In each reaction, 30 lL ketone substrate, 720 lL of 0.1 M citrate (pH 4.5), 100 lL enzyme preparation, 100 lL DIPE, and 50 lL trimethylsilyl cyanide
were combined and incubated for 17 h at 5 ꢁC.
b Determined by normal phase chiral HPLC analysis.
c Absolute configuration was determined by comparison with reported data for (S)-cyanohydrins produced by MeHNL from acetophenone, phenylac-
etone and benzylacetone.22
In contrast to the LuHNL and MeHNL enzymes, PaHNL
exhibited very little activity towards the entire range of aro-
matic ketones that were examined. Only unsubstituted ace-
tophenone, which was described previously as converting
to approximately 75% ee (R)-cyanohydrin and 30% yield
through PaHNL catalysis,27 was transformed with >5%
yield in the experiments described in this work. The signif-
icantly higher cyanohydrin ee of 97% that was observed in
this case, albeit with a lower reaction yield, strongly sug-
gests that gains in the stereoselectivity of the process are
due to a decrease in the relative rate of the background
hydrocyanation brought about by the choice of reaction
conditions, rather than a change in the properties of the
biocatalyst.
In addition to investigating the formation of chiral cyano-
hydrins from methyl phenyl ketones, several examples of
conversion of ethyl phenyl ketones with the HNL enzymes
were also tested. Data from these experiments are given in
Table 4. The use of the LuHNL enzyme to produce cyano-
hydrins from substituted forms of propiophenone (entries
1–5) showed no measurable reaction for any of the sub-
strate variants, a finding that is not surprising given the
lack of reactivity of the enzyme towards the methyl ketone

212
C. Roberge et al. / Tetrahedron: Asymmetry 18 (2007) 208–214
Table 4. Conversion of ethyl phenyl ketones with LuHNL and MeHNL at 5 ꢁCa
Entry
Ketone
LuHNL
eeb (%)
MeHNL
eeb (%)
Yieldb (%)
Configurationc
Yieldb (%)
Configurationb
1
2
3
4
5
R = H
R = 4-F
R = 4-Br
R = 4-CH3
R = 4-CH3O
0
0
0
0
0
—
—
—
—
—
—
—
—
—
—
21
4
2
4
1
90
63
57
77
88
(S)
(S)
(S)
(S)
(S)
O
R
6
43
89
(S)
59
85
(S)
O
a In each reaction, 30 lL ketone substrate, 720 lL of 0.1 M citrate (pH 4.5), 100 lL enzyme preparation, 100 lL DIPE, and 50 lL trimethylsilyl cyanide
were combined and incubated for 17 h at 5 ꢁC.
b Determined by normal phase chiral HPLC analysis.
c Absolute configuration was determined by comparison with reported data for (S)-cyanohydrins produced by MeHNL from acetophenone, phenylac-
etone and benzylacetone.22
versions of these compounds (Table 1, entries 1–5). When
MeHNL was the biocatalyst used to transform the propi-
ophenone starting materials however, conversion was
observed.
4.2. General procedure for the enzymatic reaction
To 720 lL of 0.1 M aqueous citrate, pH 4.5, were added
100 lL diisopropyl ether, 100 lL liquid enzyme solution,
30 lL ketone substrate and 50 lL trimethylsilylcyanide.
The reaction was then stirred at 5 ꢁC for 17 h.
3. Conclusion
The stereospecific conversion of methyl phenyl and ethyl
ketones to cyanohydrins is of great synthetic utility. In
the past, the insignificant activity towards aromatic ketones
of PaHNL, which has to-date been the most thoroughly
studied and frequently applied enzyme for chiral cyano-
hydrin production, may have been an impediment to the fur-
ther study of this important class of biocatalytic reactions.
In this work hydroxynitrile lyases, including for the first
time LuHNL, have been shown to be very enantiospecific
and moderately active catalysts in the hydrocyanation of
a wide variety of aromatic ketones. It should be noted that
the results of this initial screening work have been obtained
with minimal process development, and the product ee and
yield values reported represent only the minima that can be
achieved. Further refinement of these processes may in-
volve the optimization of such parameters as operating
temperature and pH and the choice and concentration of
organic solvent and cyanide donor used in the biphasic
reaction mixtures. Another important route to explore in
process improvement is the expansion of the available bio-
catalyst library, especially through engineering enzymes
with the goal of improving activity towards ketones. As
future developments bring about these activity increases,
the application of HNL enzymes to organic synthesis
schemes should increase significantly as well.
4.3. Assay of the reaction mixture
To the 1 mL reactions were added 100 lL saturated ammo-
nium sulfate and 800 lL ethyl acetate. The mixtures were
vortexed and the organic layers removed and evaporated
under nitrogen. The samples were then resuspended in
600 lL isopropanol and assayed to determine both the
yield and ee by chiral normal phase HPLC.
A
250 mm · 4.6 mm Chiralpak AD-H column was used with
an eluant of 95:5 heptane/ethanol, a flow rate of 3 mL/min,
a temperature of 10 ꢁC and a detection wavelength of
210 nm.
4.4. A 1 g scale hydrocyanation of 3-bromophenylacetone
To 18 mL of 0.1 M aqueous citrate, pH 4.5, were added
2.5 mL diisopropyl ether, 2.5 mL LuHNL solution,
750 lL 3-bromophenylacetone and 1.25 mL trim-
ethylsilylcyanide. The reaction was then stirred at 3 ꢁC
for 24 h. The product was then extracted into solvent with
the addition of 2.5 mL saturated ammonium sulfate and
25 mL ethyl acetate while stirring. The organic layer was
evaporated under nitrogen, giving an oil. HPLC analysis
using isolated racemic cyanohydrin as a standard, showed
770 mg of the cyanohydrin was present in the oil, resulting
in a 60% yield on starting material.
4. Experimental
4.1. Enzymes and chemicals
4.5. 2-Hydroxy-3-(3-fluorophenyl)-2-methyl-propanenitrile
(Table 1, entry 7)
All ketone compounds and organic solvents were pur-
chased from either Sigma–Aldrich (St. Louis, MO) or Fish-
er Scientific (Hampton, NH). Liquid preparations of
MeHNL, PaHNL and LuHNL were purchased from Julich
Chiral Solutions (Julich, Germany).
1H NMR (CDCl3, 400 MHz): d = 1.68 (s, 3H), 2.97–
3.13 (m, 2H), 7.04–7.14 (m, 4H); HPLC: tR(ketone) = 2.5
min, tR[(S)-cyanohydrin] = 4.2 min, tR[(R)-cyanohydrin] =
12.2 min.

C. Roberge et al. / Tetrahedron: Asymmetry 18 (2007) 208–214
213
4.6. 2-Hydroxy-3-(3-chlorophenyl)-2-methyl-propanenitrile
(Table 1, entry 8)
tR(ketone) = 2.3 min, tR[(S)-cyanohydrin] = 5.4 min, tR[(R)-
cyanohydrin] = 6.3 min.
1H NMR (CDCl3, 400 MHz): d = 1.67 (s, 3H), 2.95–3.10
(m, 2H), 7.08–7.34 (m, 4H); HPLC: tR(ketone) = 2.5
min, tR[(S)-cyanohydrin] = 3.9 min, tR[(R)-cyanohydrin] =
9.6 min.
4.14. 2-Hydroxy-2-methyl-4-(4-hydroxyphenyl)-butane-
nitrile (Table 1, entry 16)
1H NMR (CDCl3, 400 MHz): d = 1.65 (s, 3H), 2.03–2.07
(m, 2H), 2.78–2.89 (m, 2H), 6.79 (d, 2H, J = 8.6 Hz),
7.08 (d, 2H, J = 8.4 Hz); HPLC: tR(ketone) = 1.6 min,
4.7. 2-Hydroxy-3-(3-bromophenyl)-2-methyl-propanenitrile
(Table 1, entry 9)
tR[(S)-cyanohydrin] = 11.0 min,
tR[(R)-cyanohydrin] =
27.2 min.
1H NMR (CD4O, 400 MHz): d = 1.49 (s, 3H), 2.95–
3.00 (m, 2H), 7.14–7.50 (m, 4H); HPLC: tR(ketone) = 2.6
min, tR[(S)-cyanohydrin] = 4.0 min, tR[(R)-cyanohydrin] =
10.2 min.
4.15. 2-Hydroxy-2-methyl-4-(4-methoxyphenyl)-butane-
nitrile (Table 1, entry 17)
1H NMR (CDCl3, 400 MHz): d = 1.65 (s, 3H), 2.04–
2.08 (m, 2H), 2.83–2.87 (m, 2H), 3.79 (s, 3H), 6.86 (d,
2H, J = 8.8 Hz), 7.16 (d, 2H, J = 8.8 Hz); HPLC: tR-
4.8. 2-Hydroxy-3-(3-methyl-phenyl)-2-methyl-propanenitrile
(Table 1, entry 10)
(ketone) = 3.2 min,
tR[(S)-cyanohydrin] = 11.2 min,
1H NMR (CDCl3, 400 MHz): d = 1.69 (s, 3H), 2.39 (s,
tR[(R)-cyanohydrin] = 16.1 min.
3H), 2.91–3.12 (m, 2H), 7.01–7.31 (m, 4H); HPLC:
4.16. 2-Hydroxy-3-phenyl-2-ethyl-propanenitrile (Table 4,
entry 6)
tR(ketone) = 1.9 min,
tR[(S)-cyanohydrin] = 3.1 min,
tR[(R)-cyanohydrin] = 5.7 min.
1H NMR (CDCl3, 400 MHz): d = 1.20 (s, 3H), 1.86–
1.92 (m, 2H), 2.90–3.20 (m, 2H), 7.21–7.42 (m, 4H);
HPLC: tR(ketone) = 1.9 min, tR[(S)-cyanohydrin] = 3.8
min, tR[(R)-cyanohydrin] = 7.4 min.
4.9. 2-Hydroxy-3-(3-trifluoromethyl-phenyl)-2-methyl-
propanenitrile (Table 1, entry 11)
1H NMR (CD4O, 400 MHz): d = 1.52 (s, 3H), 2.93–
3.26 (m, 2H), 7.46–7.63 (m, 4H); HPLC: tR(ketone) = 2.0
min, tR[(S)-cyanohydrin] = 2.6 min tR[(R)-cyanohydrin] =
4.6 min.
References
1. Kruse, C. G. In The Commercial Manufacture and Applica-
tions of Optically Active Compounds; Collins, A. N., Shel-
drake, G. N., Crosby, J., Eds.; Wiley: Chichester, UK, 1992;
pp 279–299.
2. Chen, F.-X.; Feng, X. Curr. Org. Synth. 2006, 3, 77.
3. North, M. Tetrahedron: Asymmetry 2003, 14, 147.
4. Garcia, C.; Martin, V. S. Curr. Org. Chem. 2006, 10, 1849.
5. Deng, H.; Isler, M. P. l.; Snapper, M. L.; Hoveyda, A. H.
Angew. Chem., Int. Ed. 2002, 41, 1009.
4.10. 2-Hydroxy-3-(3-methoxyphenyl)-2-methyl-propane-
nitrile (Table 1, entry 12)
1H NMR (CDCl3, 400 MHz): d = 1.67 (s, 3H), 2.93–3.11
(m, 2H), 3.83 (s, 3H), 6.76–7.32 (m, 4H); HPLC: tR(keto-
ne) = 3.6 min,
tR[(S)-cyanohydrin] = 7.3 min,
tR[(R)-
cyanohydrin] = 12.0 min.
6. Decicco, C. P.; Grover, P. Synlett 1997, 529.
7. Buhler, H.; Bayer, A.; Effenberger, F. Chem. Eur. J. 2000, 6,
2564.
4.11. 2-Hydroxy-3-(4-bromophenyl)-2-methyl-propanenitrile
(Table 1, entry 13)
8. Johnson, D. V.; Griengl, H. Tetrahedron 1997, 53, 617.
9. Fecther, M. H.; Griengl, H. Food Technol. Biotechnol. 2004,
42, 287.
10. Brussee, J.; van der Gen, A. In Stereoselective Biocatalysis;
Patel, R. N., Ed.; Dekker: New York, 2000; pp 289–320.
11. Effenberger, F. In Stereoselective Biocatalysis; Patel, R. N.,
Ed.; Dekker: New York, 2000; pp 321–342.
1H NMR (CD4O, 400 MHz): d = 1.48 (s, 3H), 2.93–3.01
(m, 2H), 7.23 (d, 2H, J = 8.4 Hz), 7.58 (d, 2H,
J = 8.4 Hz); HPLC: tR(ketone) = 2.9 min, tR[(S)-cyano-
hydrin] = 4.7 min, tR[(R)-cyanohydrin] = 7.1 min.
4.12. 2-Hydroxy-3-(4-methoxyphenyl)-2-methyl-propane-
nitrile (Table 1, entry 14)
12. Sharma, M.; Sharma, N. N.; Bhalla, T. C. Enzyme Microb.
Technol. 2005, 37, 279.
13. Effenberger, F.; Roos, J.; Kobler, C.; Buhler, H. Can. J.
Chem. 2002, 80, 671.
14. Kobler, C.; Bohrer, A.; Effenberger, F. Tetrahedron 2004, 60,
10397.
15. Kobler, C.; Effenberger, F. Tetrahedron: Asymmetry 2004, 15,
3731.
1H NMR (CDCl3, 400 MHz): d = 1.65 (s, 3H), 2.89–3.08
(m, 2H), 3.82 (s, 3H), 6.88 (d, 2H, J = 8.6 Hz), 7.26 (d,
2H, J = 7.8 Hz); HPLC: tR(ketone) = 3.6 min, tR[(S)-cya-
nohydrin] = 6.7 min tR[(R)-cyanohydrin] = 13.6 min.
16. The reaction medium was composed of a biphasic mixture of
aqueous citrate buffer and diisopropyl ether, the organic
solvent that has been demonstrated as being the most
generally applicable for use with HNL-catalyzed hydrocya-
nation.11 The pH of the buffer component of the reaction
mixture was set at 4.5, chosen to balance the requirements for
4.13. 2-Hydroxy-2-methyl-4-phenylbutanenitrile (Table 1,
entry 15)
1H NMR (CD4O, 400 MHz): d = 1.57 (s, 3H), 1.96–2.00
(m, 2H), 2.71–2.88 (m, 2H), 7.11–7.27 (m, 4H); HPLC:

214
C. Roberge et al. / Tetrahedron: Asymmetry 18 (2007) 208–214
minimizing both enzyme instability that is known to be more
21. Deng, H.; Isler, M. P.; Snapper, M. L.; Hoveyda, A. H.
Angew. Chem., Int. Ed. 2002, 41, 1009.
22. Buhler, H.; Effenberger, F.; Forster, S.; Roos, J.; Wajant, H.
ChemBioChem 2003, 4, 211.
23. Nanda, S.; Kato, Y.; Asano, Y. Tetrahedron 2005, 61,
10908.
24. Hansch, C.; Leo, A.; Hoekman, D. Exploring QSAR:
Hydrophobic, Electronic, and Steric Constants; American
Chemical Society: Washington, DC, 1995.
pronounced at pH < 5 (see Albrecht, J.; Jansen, I.; Kula,
M.-R. Biotechnol. Appl. Biochem. 1993, 17, 191), and the rate
of non-enzymatic spontaneous cyanide addition that is
increased at pH > 4 and leads to racemic product formation
(see Kragl, U.; Niedermeyer, U.; Kula, M.-R.; Wandrey, C.
Ann. NY Acad. Sci. 1990, 613, 617).
17. Chen, F.-X.; Qin, B.; Fen, X.; Zhang, G.; Jiang, Y.
Tetrahedron 2004, 60, 10449.
18. Hamashima, Y.; Kanai, M.; Shibasaki, M. J. Am. Chem. Soc.
2000, 122, 7412.
19. Fuerst, D. F.; Jacobsen, E. N. J. Am. Chem. Soc. 2005, 127,
8964.
20. Chen, F.-X.; Zhou, H.; Liu, X.; Qin, B.; Feng, X.; Zhang, G.;
Jiang, Y. Chem. Eur. J. 2004, 10, 4790.
25. Hansch, C.; Leo, A. Exploring QSAR: Fundamentals and
Applications in Chemistry and Biology; American Chemical
Society: Washington, DC, 1995.
26. Van Langen, L. M.; Selassa, R. P.; van Rantwijk, F.;
Sheldon, R. A. Org. Lett. 2005, 7, 327.
27. Lin, G.; Han, S.; Li, Z. Tetrahedron 1999, 55, 3531.