Investigations on enzyme catalytic promiscuity: The first attempts at a hydrolytic enzyme-promoted conjugate addition of nucleophiles to α,β-unsaturated sulfinyl acceptors

Article abstract of DOI:10.1016/j.molcatb.2012.05.002

Looking for new examples of enzyme catalytic promiscuity, attempts were made to use hydrolytic enzymes as catalysts for a conjugate addition of nucleophiles toα,β-unsaturated sulfinyl derivatives. The addition of piperidine to phenyl vinyl sulfoxide in chloroform proceeded both in the enzyme-catalyzed and non-catalyzed process, while in the former case the reaction was 2.5-fold faster. On the contrary, the conjugate addition of benzenethiol to phenyl vinyl sulfoxide proceeded only in the presence of enzymes and when ethanol was used as solvent. In no case were the products enantiomerically enriched. However, the addition of benzenethiol to a better Michael acceptor, namely a cyclic α-sulfinylalkenylphosphonate, performed in the presence of various lipases under kinetic resolution conditions gave in certain instances both the product and the recovered substrate with up to 25% optical purity. Although the stereoselctivity and the rates of these reactions were quite low, this are the first examples of the lipase-catalyzed Michael addition of heteroatom nucleophiles to α,β-unsaturated heteroorganic acceptors. Some mechanistic considerations are presented.

Full text of DOI:10.1016/j.molcatb.2012.05.002

Journal of Molecular Catalysis B: Enzymatic 81 (2012) 25–30
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Catalysis B: Enzymatic
journal homepage: www.elsevier.com/locate/molcatb
Investigations on enzyme catalytic promiscuity: The first attempts at a hydrolytic
enzyme-promoted conjugate addition of nucleophiles to ␣,␤-unsaturated
sulfinyl acceptors
Lidia Madalin´ ska, Małgorzata Kwiatkowska, Tomasz Cierpiał, Piotr Kiełbasin´ ski∗
Centre of Molecular and Macromolecular Studies, Polish Academy of Sciences, Department of Heteroorganic Chemistry, Sienkiewicza 112, 90-363 Łód´z, Poland
a r t i c l e i n f o
a b s t r a c t
Article history:
Looking for new examples of enzyme catalytic promiscuity, attempts were made to use hydrolytic
enzymes as catalysts for a conjugate addition of nucleophiles to␣,␤-unsaturated sulfinyl derivatives. The
addition of piperidine to phenyl vinyl sulfoxide in chloroform proceeded both in the enzyme-catalyzed
and non-catalyzed process, while in the former case the reaction was 2.5-fold faster. On the contrary, the
conjugate addition of benzenethiol to phenyl vinyl sulfoxide proceeded only in the presence of enzymes
and when ethanol was used as solvent. In no case were the products enantiomerically enriched. However,
the addition of benzenethiol to a better Michael acceptor, namely a cyclic ␣-sulfinylalkenylphosphonate,
performed in the presence of various lipases under kinetic resolution conditions gave in certain instances
both the product and the recovered substrate with up to 25% optical purity. Although the stereoselctivity
and the rates of these reactions were quite low, this are the first examples of the lipase-catalyzed Michael
addition of heteroatom nucleophiles to ␣,␤-unsaturated heteroorganic acceptors. Some mechanistic
considerations are presented.
Received 23 April 2012
Accepted 2 May 2012
Available online 11 May 2012
Keywords:
Enzymes
Catalytic promiscuity
␣,␤-Unsaturated sulfinyl compounds
Catalysis
© 2012 Elsevier B.V. All rights reserved.
1. Introduction
An intriguing behavior of other types of proteases, belonging
to the family of serine hydrolases, is their ability to hydrolyze
Enzyme catalytic promiscuity, i.e. the ability of a single active
site of the enzyme to catalyze more than one reaction, has been a
subject of increasing research interest in the recent few years (for
recent overviews see [1–6]). Enzyme promiscuity is clearly advan-
tageous to chemists since it broadens the applicability of enzymes
in chemical synthesis. So far, several interesting examples of this
phenomenon have been reported.
Thus, selected aminopeptidases, besides their natural hydrolytic
activity toward peptides, act also as phosphodiesterases [7,8] and
phosphotriesterases [9–12]. The latter is of particular importance,
since these enzymes can hydrolyze unnatural substrates – triesters
of phosphoric acid and diesters of phosphonic acids – such as
organophosphorus pesticides or organophosphorus warfare agents
[13]. This means that they are capable of catalyzing the reaction that
does not exist in nature. It should be noted that all the enzymes pre-
sented above belong to the class of dimetalloenzymes, which would
to some extent explain their similar hydrolytic activity toward
entirely different substrates. The mechanism of their action always
rests upon the activation of a water molecule by the metal cations
present in the active sites.
sulfur nitrogen bond in N-acyl sulfinamides. When some N-acyl
arenesulfinamides are treated with a buffer in the presence of sub-
tilisin Carlsberg, the S N bond hydrolysis unexpectedly becomes
favored over the expected C N bond hydrolysis to give, under
kinetic resolution conditions, the corresponding sulfinic acids and
carboxamides, together with the enantiomerically enriched recov-
ered substrates. The proof for the direct involvement of the enzyme
is the formation of an intermediate – O-sulfinyl enzyme, sulfiny-
lated most probably on the active site serine [14]. Thus, also in
this case, the hydrolysis proceeds according to a similar mechanism
for both types of substrates, i.e. carboxylic and sulfinic esters, and
involves attack of the catalytically active serine on the electrophilic
center of the substrate.
A much more interesting phenomenon seems the ability of ser-
ine hydrolases to catalyze a carbon-carbon and carbon-heteroatom
bond formation, i.e. to exhibit a lyase activity. Lipases are the
enzymes for which a number of examples of such a promis-
cuous activity have been reported. In addition to their original
activity comprising hydrolysis of lipids and, generally, catal-
ysis of the hydrolysis or formation of carboxylic esters [15],
lipases have been found to catalyze also the carbon–carbon and
carbon-heteroatom bond forming reactions. The first example of a
lipase-catalyzed Michael addition of a number of nucleophiles to 2-
(trifluoromethyl)propenoic acid was described as early as in 1986
∗
Corresponding author. Tel.: +48 42 6815832; fax: +48 426803261.
E-mail address: piokiel@cbmm.lodz.pl (P. Kiełbasin´ ski).
1381-1177/$ – see front matter © 2012 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molcatb.2012.05.002

26
L. Madalin´ska et al. / Journal of Molecular Catalysis B: Enzymatic 81 (2012) 25–30
Scheme 1. Lipase-catalyzed Michael additions.
[16]. Michael addition of secondary amines to acrylonitrile is up
to 100-fold faster in the presence of various preparations of lipase
from Candida antarctica (CAL-B) than in the absence of a biocatalyst
[17]. In a similar way, lipases catalyze Michael addition of amines,
thiols [18], 1,3-dicarbonyl derivatives [19,20] and other C-acids [21]
to ˛,ˇ-unsaturated carbonyl compounds (Scheme 1). Interestingly,
all such reactions proceed without enantioselectivity, the excep-
tions being a CAL-B-promoted Michael addition of benzylamine
to methyl crotonate, which gives the adduct with enantiomeric
excess of 60% [22] and Michael additions of C-acids to nitroalkenes,
giving products with ee’s from 7 to 86% [23]. Similarly, the lipase-
catalyzed aldol condensation proceeds with low enantioselectivity,
giving the products with enantiomeric excess of 9 to 43% [24]. Other
types of hydrolytic enzymes, namely some serine proteases, also
catalyze Michael addition of substituted pyrimidines [25], imida-
zoles [26] and purines [27] to acrylates.
of their stereoselective recognition by a chiral enzyme leading
to enantiomerically enriched products. The present investigations
are a continuation of our previous studies which have revealed
that heteroatom stereogenic and prostereogenic centers are stere-
oselectively recognized by common hydrolytic enzymes allowing
for the synthesis of enantiomerically enriched heteroorganic com-
pounds [30–36].
2. Experimental
2.1. General
The synthesized products were purified by column chromatog-
raphy on silica gel. Solvents were dried using general procedures
On the basis of the quantum-chemical studies, it has been con-
cluded that it is the so called “oxyanion hole” of the enzyme
that binds the carbonyl oxygen or nitrile nitrogen, increasing
electrophilicity of the corresponding carbon atom and enhancing
the attack of a nucleophile, which is, in turn, activated by histi-
dine (Scheme 2) [18]. This model clearly shows that the catalytic
machinery involves a dyad of histidine and aspartate together with
the oxyanion hole. Hence, it does not involve serine, which is the
key amino acid in the hydrolytic activity of lipases, and, together
with aspartate and histidine, constitutes the active site catalytic
triad. This has been confirmed by constructing a mutant in which
serine was replaced with alanine (Ser105Ala), and finding that it
catalyzes the Michael additions even more efficiently than the wild-
type enzyme [28]. In a similar way, the catalytic activity of CAL-B
O
(Gln106)
O
(Thr40)
O
H
N
N
H
H
O
CO_
NH_
O
H2C
H
HO
S
O
Me
(Ser105)
and its Ser105Ala mutant in an aldol addition, thus another C
bond forming reaction, has been explained [29].
C
N
H
N
(Asp187)
Taking into account the above mechanism and, particularly, the
fact that the carbonyl oxygen atom (or the nitrogen atom of the
nitrile) is H-bound with the amino acids of the oxyanion hole, we
have decided to check whether the same interaction will take place
with the oxygen connected with a heteroatom, e.g. with sulfur
in sulfinyl derivatives. Such an approach seems quite justifiable
since the sulfinyl oxygen forms strong hydrogen bonds with H-
donors. Moreover, sulfinyl derivatives are tetrahedral which, in
contrast to the planar carbonyl or cyano groups, raises a possibility
CO_
(His224)
NH_
(Ala105)
Scheme 2. Hypothetical mechanism of the lipase-catalyzed Michael addition of
methanethiol to acrolein [18,29]. In parentheses aminoacids: serine present in the
active site of a native lipase and alanine replacing serine in the Ser105Ala mutant.

L. Madalin´ska et al. / Journal of Molecular Catalysis B: Enzymatic 81 (2012) 25–30
27
Table 1
Conjugate addition of benzenethiol to phenyl vinyl sulfoxide rac-1.
Table 2
Conjugate addition of benzenethiol to 4.
Entry Enzyme Solvent
Reaction time
[days]
Unreacted
substrate 1
yield [%]
Product 3
yield [%]
Entry Lipase
Solvent
CHCl3
Product Yield (NMR) [%] [␣]D
o.p. [%]
1
2
3
4
5
CAL-B
CRL
AK
4
6
7
5
61
12
20
7
N.i.
–
–
N.i.
–
–
1
2
3
4
5
6
7
8
9
MJL
PPL
CRL
CAL B
PS
BSP
AK
CCL
MJL
PPL
EtOH
EtOH
EtOH
EtOH
EtOH
EtOH
EtOH
EtOH
CH2Cl2
CH2Cl2
7
7
8
8
5
5
6
6
8
8
43
53
43
42
36
40
46
44
35
29
21
22
35
28
24
25
–4.2 22
4
6
7
5
46
9
14
23
+5.5
–
–
6.4
–
–
CHCl3
i-Pr2O
CH2Cl2
–5.0 26
4
6
7
5
4
20
25
48
N.i.
–
–
N.i.
–
–
No reaction
No reaction
10
–1.5
8
Enzyme: see Section 2.1.
CRL
AK
4
6
7
5
22
–
11
42
+1
–
–
1.2
–
–
–2.5 15
and distilled prior to use. The NMR spectra were recorded in CDCl3
with a ‘Bruker AC 200ꢀ spectrometer. The chemical shifts (ı) are
expressed in ppm, the coupling constants (J) are given in Hz. Optical
rotation values were measured on a Perkin–Elmer-241 photopo-
larimeter for the sodium D line at 20 ◦C. Mass spectra were recorded
with a Finnigan MAT 95 Voyager Elite spectrometer.
Enzymes: MJL: lipase from Mucor javanicus SIGMA, 300 U/mg
solid; PPL: lipase from porcine pancreas, SIGMA, 30–90 U/mg pro-
tein; CRL: lipase from Candida rugosa, SIGMA, 700 U/mg solid;
CAL-B (Novozym 435): lipase acrylic resin form Candida Antarctica,
SIGMA, 10 U/mg; PS: lipase form Pseudomonas cepacia, AMANO,
>30 U/mg; BSP: proteinase from Bacillus subtilis (Subtilisin Carls-
berg), SIGMA, 20 U/mg; AK: lipase from Pseudomonas fluorescens,
AMANO, >20 U/mg; CCL: lipase from Candida cylindracea (rugosa),
SIGMA, 25 U/mg lyophilized powder.
4
6
7
5
22
–
12
42
N.i.
–
–
N.i.
–
–
Me2CO/Et2O
1:1
–2.8 16
Lipase: see Section 2.1; [␣]D in acetone, c = 1; N.i. – not isolated; o.p. – optical purity
calculated by comparison of the optical rotations of the reaction products with those
of the authentic samples; the measurements were performed using the same solvent
and concentration [37,39].
2.4. Enzyme-catalyzed addition of benzenethiol to
2-phosphono-2,3-didehydrothiolane S-oxide 4
2.4.1. 2-(5ꢀ,5ꢀ-Dimethyl-1ꢀ,3ꢀ,2ꢀ-dioxaphosphorinanyl)-2,3-
didehydrothiolane1-oxide 4
2-(5ꢀ,5ꢀ-Dimethyl-1ꢀ,3ꢀ,2ꢀ-dioxaphosphorinanyl)-2,3-
didehydrothiolane 1-oxide 4 was prepared according to a known
method [37].
2.2. Enzyme-catalyzed conjugate addition of piperidine to phenyl
vinyl sulfoxide
2.4.2. General procedure
The 2-phosphono-2,3-didehydrothiolane S-oxide 4 (100 mg,
0.40 mmol) was dissolved in a dry solvent (5 mL). To this solu-
tion, an enzyme (100 mg) was added and after 15 min benzenethiol
(0.5 eq., 0.20 mmol). The reaction was monitored by 31P NMR spec-
troscopy and by TLC (CH2Cl2/MeOH 20:1) and stopped after 2
days. The enzyme was filtered off and the solvent was evapo-
rated. The residue was separated by column chromatography using
CH2Cl2/MeOH in gradient (100:1–20:1) as eluent. The products
were identified on the basis of their NMR and MS spectra, as 4
(recovered substrate) and 5. The results are shown in Scheme 5
and collected in Table 2.
Racemic sulfoxide 1 (100 mg, 0.66 mmol) was dissolved in dry
chloroform (5 mL). To this solution, protease from Bacillus subtilis
(100 mg) and piperidine (0.5 equiv., 0.33 mmol) were added. The
reaction was monitored by TLC (CH2Cl2/MeOH 40:1). The enzyme
was filtered off and the solvent was evaporated. The residue was
separated by column chromatography CH2Cl2/MeOH in gradient as
eluent. The yields of the product and recovered substrate are shown
in Scheme 3.
2-(N-piperidyl)ethyl phenyl sulfoxide 2: 1H NMR (CDCl3)
ı = 1.41–1.62 (m, 6H), 2.30–2.58 (m, 5H), 2.72–3.02 (m, 3H),
7.48–7.67 (m, 5H). MS (CI) m/z 237 (M + H).
2-[2ꢀ-(5,5-dimethyl-1,3,2-dioxaphosphorinanyl)]-2,3-didehydro-
thiolane 1-oxide (+)-(S)-4.
31P NMR (CDCl3) ı = 4.2; 1H NMR (CDCl3) ı = 1.05 (s, 3H, CH3),
1.24 (s, 3H, CH3), 3.00–3.65 (m, 4H, SCH2, CH2), 3.98–4.40 (m, 4H,
2 × OCH2), 7.55 (dt, J = 11.5, 2.3, 1H, C CH). 13C NMR (CDCl3) ı = 20.7
(s, CH3), 21.7 (s, CH3), 32.3 (d, J = 6.8, CCH3), 34.9 (d, J = 18.2, CH2),
52.3 (d, J = 7.0, CH2), 76.5 (d, J = 6.5, OCH2), 77.4 (d, J = 6.4, OCH2),
140.1 (d, J = 187.4, PC), 157.4 (d, J = 12.7, PC = CH). HRMS (CI): calcd.
for C9H16PSO4 (M + H): m/z 251.0506, found: m/z 251.0509.
The spectroscopic data were identical with those described pre-
viously [37].
2.3. Enzyme-catalyzed addition of benzenethiol to phenyl vinyl
sulfoxide. General procedure
To a solution of sulfoxide 1 (152 mg, 1 mmol) in 96% EtOH (5 mL)
an enzyme (10–100 mg) and benzenethiol (55 mg, 0.5 mmol) were
added. The mixture was stirred at room temperature for several
days (TLC control CH2Cl2: MeOH 20:1). Then the enzyme was fil-
tered off and washed with CH2Cl2. The solvents were evaporated
to give a crude mixture of the product 5 and the substrate 1, which
were separated using preparative TLC (CH2Cl2: MeOH 20:1). The
results are summarized in Table 1.
3-Phenylsulfanyl-2-[2ꢀ-(5,5-dimethyl-1,3,2-dioxaphosphori-
nanyl)]thiolane 1-oxide (−)-(1R,2S,3S)-5: 31P NMR (CDCl3) ı 13.2;
1H NMR (CDCl3) ı 0.99 (s, 3H, CH3), 1.22 (s, 3H, CH3), 2.77–2.81
(m, 2H, CH2), 3.00 (m, 1H, CHHSO), 3.20 (m, 1H, CHHSO), 3.31
(ddd, 4JHH = 1.1, 3JHH = 7.2, 2JHP = 14.6, 1H, PCH), 3.92–4.01 (m, 3H),
4.11–4.16 (m, 2H), 7.35–7.37 (m, 3H, Harom), 7.54–7.57 (m, 2H,
Harom); 13C NMR (CDCl3) ı 20.98 (CH3), 21.69 (CH3), 32.43 (d,
Phenyl 2-phenylthioetyl sulfoxide 3: 1H NMR (CDCl3):
ı = 2.88–3.34 (m, 4H, CH2CH2), 7.16–7.63 (m, 10H, aromat.).
MS (CI): m/z 263 (M + H). HRMS: calcd. for C14H15OS2 263.056435
found 263.05639.

28
L. Madalin´ska et al. / Journal of Molecular Catalysis B: Enzymatic 81 (2012) 25–30
O
O
O
S
protease from
S
S
N
Bacillus subtilis
+
+
CHCl3
N
H
1
2
rac-1
Entry Reaction time
[days]
Enzyme
Recovered substrate 1 Product of addition 2
(isolated yield [%])
(isolated yield [%])
1
2
5
2
None
Protease from
Bacillus subtilis
40
42
42
43
Scheme 3. Conjugate addition of piperidine to phenyl vinyl sulfoxide rac-1 (expected absolute configurations are shown arbitrarily).
J = 6.8, (CH3)2C), 34.77 (d, J = 7.8, CH2), 48.26 (CHS), 53.29 (d, J = 3.4,
CH2SO), 68.52 (d, J = 130.8, PCH), 77.05 (d, J = 6.6, CH2O), 77.33
(d, J = 6.5, CH2O), 128.57 (CHarom), 129.37 (2× CHarom), 133.50
(SCarom), 133.55 (2× CHarom); MS (CI) m/z 361 (M + H), 161 (100);
HRMS (CI) calcd. for C15H21PO4S2 (M + H): m/z 361.0697, found:
m/z 361.0706.
relatively good yields (Scheme 4, Table 1). Unfortunately, again they
were racemic (chiral HPLC: ee < 1%). Noteworthy, no reaction was
observed under these conditions in the absence of enzyme.
Although the lack of stereoselectivity in the above reactions was
somewhat disappointing, the results undoubtedly proved that the
additions were catalyzed by enzymes. A surprising effect of the
application of ethanol as solvent seems at first sight difficult to
explain. However, there are precedents described in the literature,
which show that alcohols may enhance the activity of lipases, par-
ticularly in their promiscuous catalytic action in a C-heteroatom
[22] and C C bond formation [40].
The spectroscopic data were identical with those described pre-
viously [38,39].
3-Phenylsulphanyl-2-(5ꢀ,5ꢀ-dimethyl-1ꢀ,3ꢀ,2ꢀ-dioxaphospho-
rinanyl)-2,3-didehydrothiolane 6: 31P NMR (CDCl3) ı = 4.65. 1H
NMR (CDCl3) ı = 1.06 (s, 3H, CH3), 1.20 (s, 3H, CH3), 2.79–2.89
(m, 2H, PhSCCH2), 3.20 (t, 2H, J = 8.60 Hz, SCH2), 4.05–4.21 (m,
4H, 2× OCH2), 7.31–7.34 (m, 3H, Ph), 7.45–7.50 (m, 2H, Ph). 13C
NMR (CDCl3) ı = 21.23 (s, CH3), 21.74 (s, CH3), 31.51 (d, J = 6.70 Hz,
CH2), 32.50 (d, J = 6.35 Hz, C(CH3)CH3), 42.78 (d, J = 6.90 Hz, CH2S),
76.90 (d, J = 29.35 Hz, 2× CH2O), 127.99, 128.49, 129.20, 132.12,
132.58, 133.03 (Ph), 143.03 (d, J = 10. 14 Hz, PhSC ). MS (CI) m/z
343 (M + H), 685 (2M + H); HRMS (EI): calcd. for C15H19PS2O3: m/z
342.051329, found: m/z 342.05178.
3.2. Michael additions to an ˛-sulfinylalkenylphosphonate
Since phenyl vinyl sulfoxide 1 proved to be a relatively weak
Michael acceptor, we decided to construct an alkenyl compound
bearing sulfinyl and phosphoryl moieties at the same sp2 carbon
atom. The presence of the two electron-withdrawing substituents
was expected to enhance electrophilicity of the ␤ carbon atom. We
decided to use as a model compound racemic 2-(5ꢀ,5ꢀ-dimethyl-
1ꢀ,3ꢀ,2ꢀ-dioxaphosphorinanyl)-2,3-didehydrothiolane 1-oxide 4,
because its triethylamine-catalyzed reaction with benzenethiol
had been earlier studied by us and the relative and absolute con-
figurations of both 4 and the conjugate addition product 5 were
determined [37–39].
3-Phenylsulphanyl-2-(5ꢀ,5ꢀ-dimethyl-1ꢀ,3ꢀ,2ꢀ-dioxaphospho-
rinanyl)-3,4-didehydrothiolane 7: 31P NMR (CDCl3) ı = 10.86. 1H
NMR (CDCl3) ı = 0.91 (s, 3H, CH3), 1.28 (s, 3H, CH3), 2.88–2.95 (m,
1H), 3.96–4.10 (m, 4H, 2× CH2O), 4.34–4.46 (m, 2H), 7.28–7.52 (m,
6H, Ph, SCH2CH); MS (CI) m/z 343 (M + H); HRMS (EI): calcd. for
C15H19PS2O3: m/z 342.051329, found: m/z 342.05124.
The enzyme-catalyzed additions to 4 were performed under
kinetic resolution conditions, using half molar amount of ben-
zenethiol and a number of commercially available lipases. The
reaction was monitored by 31P NMR spectroscopy of the crude sam-
ples taken out directly from the reaction solution and stopped after
two days. As a result, complex mixtures of products were usually
obtained from which only four compounds, among them the recov-
ered substrate 4 and the expected adduct 5, could be isolated and
characterized (Scheme 5, Table 2). The structure and absolute con-
figurations of 4 and 5 were ascribed by comparison with authentic
samples which were obtained by us earlier and described in details
in Refs. [37–39]. It should be added that compound 5 was obtained
as a single diastereomer which, according to Ref. [39], is the ther-
modynamically controlled product. Hence, comparison of a sample
of 5 with the sample of the same compound obtained earlier [39]
(for which an X-ray analysis was performed and both relative and
absolute configurations were determined) using TLC, HPLC, MS and
1H and 31P NMR, undoubtedly established its identity. Therefore,
the sign and value of the [␣]D of compound 5 allowed to ultimately
determine its absolute configuration and optical purity.
3. Results and discussion
3.1. Michael additions to ˛,ˇ-unsaturated sulfoxides
Commercially available racemic phenyl vinyl sulfoxide rac-1
was chosen as the first and simplest substrate. It was subjected to
the reaction with piperidine (half molar amount to ensure kinetic
resolution) in the presence of protease from Bacillus subtilis. After
completion of the reaction the mixture was separated by col-
umn chromatography to give both the unreacted substrate 1 and
the addition product 2 (Scheme 3). Surprisingly, neither one was
optically active. Moreover, the reaction performed under identical
conditions without enzyme, gave almost the same result, although
the identical conversion required a 2.5-fold longer reaction time.
In similar reactions, in which a half-molar amount of ben-
zenethiol was used as a nucleophile and the reaction was performed
in chloroform or dichloromethane in the presence of various
hydrolytic enzymes, no expected product of the conjugate addi-
tion could be detected. However, when chloroform was replaced
with ethanol, the desired 1,4 addition product 3 was formed.
Although the time which was needed to complete the conver-
sion was, in comparison with the analogous Michael additions
to ˛,ˇ-unsaturated carbonyl compounds [16,17], quite long, both
the unreacted substrate 1 and the product 3 were isolated in
In certain cases it was impossible to obtain the pure recov-
ered substrate 4 due to its decomposition during chromatography:
the isolated substance was contaminated with unidentified prod-
ucts of decomposition, which could not be removed or separated.
This caused severe problems and substantially lowered the isolated

L. Madalin´ska et al. / Journal of Molecular Catalysis B: Enzymatic 81 (2012) 25–30
29
Scheme 4. Conjugate addition of benzenethiol to phenyl vinyl sulfoxide rac-1 (expected absolute configurations are shown arbitrarily).
O
O
S
O
O
P
O
P
O
P
O
O
O
O
O
O
PhSH
Lipase
S
S
+
PhS
(+)-(1 )-4
S
-5
S S
-4
(-)-(1 ,2 ,3 )
rac
R
O
O
O
P
O
O
S
S
P
O
PhS
PhS
6
7
Scheme 5. Conjugate addition of benzenethiol to 2-(5ꢀ,5ꢀ-dimethyl-1ꢀ,3ꢀ,2ꢀ-dioxaphos-phorinanyl)-2,3-didehydrothiolane 1-oxide 4.
yields of all the products. Therefore, Table 1 contains only selected
examples and the yields, which were estimated on the basis of the
NMR spectra of the crude reaction mixtures. It should be added that
no formation of the products could be observed when the reaction
was performed in the absence of enzymes or in the presence of the
CAL-B which was denatured with urea [21].
Taking into account the hypothetical mechanism shown in
Scheme 2, the course of the reaction discussed above may be illus-
trated in the following way (Scheme 6). The sulfinyl oxygen atom
is bound within the “oxyanion hole” of enzyme active site using
hydrogen bonds. In turn, the nucleophilicity of the sulfur center
in the benzenethiol molecule is enhanced by the histidine of the
catalytic dyad. Though the latter interaction is identical as in the
case of the Michael addition of thiols to enones, the H-binding
of the sulfinyl oxygen atom must be different (weaker?) from
that of the carbonyl oxygen atom, which results in a less efficient
catalysis of the reaction by enzymes. It is known that the oxyan-
ion hole binds transition states better than the ground state. In
case of the lipase-catalyzed hydrolysis of esters, the intermediary
oxy anion is tetrahedral. Although the sulfinyl group is tetrahe-
dral as well, which would suggest that it should be bound equally
well, in contrast to the oxy anion it bears no negative charge on
the oxygen atom in the intermediate, which obviously decreases
the H-bonding strength. Moreover, for the Michael addition of
Inspection of Table 2 reveals that the expected adduct 5 was
formed in a stereoselective manner, although its optical purity
was quite low. The term “optical purity” was used here, instead
of the “enantiomeric excess”, to evaluate the stereoselectivity of
the reaction, since under no conditions could the enantiomers be
separated by chiral HPLC (cf. [37,39]). Nevertheless, isolation of
the optically active species undoubtedly proves that the enzymes
must have been involved in the process, since they were the only
source of chirality. The fact that the recovered substrate 4 was
also optically active and its absolute configuration at the sulfinyl
sulfur atom was opposite to the configuration of the sulfinyl moi-
ety in 5, clearly indicates that the result was an effect of kinetic
resolution. The next two products, namely 6 and 7, were also iso-
lated and their structures were ascribed on the basis of 1H and 31
P
(Gln106)
NMR and MS. The way of their formation is not known. Since each
of them contained the benzenesulfenyl moiety, originating from
benzenethiol, it seemed reasonable to assume that the adduct 5
was the primary intermediate. However, their formation, which
undoubtedly had to be a result of a simultaneous reduction and
dehydrogenation/oxidation of 5, remains unclear. To gain a better
insight into the way of their formation, some additional experi-
ments were performed. Thus, when pure 5 was treated with lipase
AK in chloroform, no changes were observed for over 15 days.
However, addition of benzenethiol to this mixture caused forma-
tion of compound 7. This ultimately proves that both 6 and 7
must be produced from 5 in a subsequent reaction, in which the
presence of benzenethiol seems crucial. When the procedure was
reversed, i.e. benzenethiol was first added to a solution of 5, no
changes could be observed until the lipase was added. Then, the
result was as above, which may be taken as proof for the cooper-
ative action of benzenethiol and the enzyme in the formation of
by-products.
O
O
(Thr40)
N
O
H
H
H
N
O
S
O
Ph
O
S
Ph
N
H
H
N
(Asp187)
(His224)
Scheme 6. Assumed mechanism of the conjugate addition of benzenethiol to 1.

30
L. Madalin´ska et al. / Journal of Molecular Catalysis B: Enzymatic 81 (2012) 25–30
nucleophiles to enones, the intermediary oxy anion is planar, which
additionally makes it fit better in the oxyanion hole by lowering
spatial requirements in comparison with the tetrahedral sulfinyl
intermediate. It must be stressed that such an explanation is only a
speculative one. To find a better insight into the real mechanism of
the reaction further studies will be continued, which will involve
attempts at the X-ray analysis of an enzyme-substrate complex as
well as molecular modeling. Nevertheless, the results presented
undoubtedly prove that the stereogenic sulfinyl group in 4 must be
recognized and stereoselectively bound in the chiral enzyme envi-
ronment which results in the stereoselective course of the addition.
This result does not exclude another possible mechanism, based on
a phenomenon called “alternate site promiscuity”, which assumes
that the reaction neither involves any of the catalytic amino acids
of the natural enzymatic process, nor appears to occur in the natu-
ral binding pocket (for a recent example see Reetz and co-workers
[41]).
[8] A. Ercan, H. Park, L.-J. Ming, Chem. Commun. (2000) 2501–2502.
[9] C.M. Hill, W.-S. Li, T.-C. Cheng, J.J. DeFrank, F.M. Raushel, Bioorg. Med. Chem.
Lett. 10 (2000) 1285–1288.
[10] C.M. Hill, F. Wu, T.-C. Cheng, J.J. DeFrank, F.M. Raushel, Bioorg. Chem. 29 (2001)
27–35.
[11] S.-C. Jao, L.-F. Huang, Y.-S. Tao, W.-S. Li, J. Mol. Catal. B: Enzym. 27 (2004) 7–12.
[12] L.-F. Huang, B. Su, S.-C. Jao, K.-T. Liu, W.-S. Li, Chembiochem 7 (2006) 506–514.
[13] T.-C. Cheng, J.J. DeFrank, in: B. Zwanenburg, M. Mikołajczyk, P. Kiełbasin´ ski
(Eds.), Enzymes in Action: Green Solutions for Chemical Problems, Kluwer
Academic Publishers, Dordrecht, 2000, pp. 243–261.
[14] P.F. Mugford, V.P. Magloire, R.J. Kazlauskas, J. Am. Chem. Soc. 127 (2005)
6536–6537.
[15] U.T. Bornscheuer, R.J. Kazlauskas, Hydrolases in Organic Synthesis. Regio-
and Stereoselective Biotransformations, Second Edition, Wiley-VCH GmbH &
Co./KGaA, 2006, pp. 61–183.
[16] T. Kitazume, T. Ikeya, K. Murata, J. Chem. Soc. Chem. Commun. (1986)
1331–1333.
[17] O. Torre, I. Alfonso, V. Gotor, Chem. Commun. (2004) 1724–1725.
[18] P. Carlqvist, M. Svedendahl, C. Branneby, K. Hult, T. Brinck, P. Berglund, Chem-
biochem 6 (2005) 331–336.
[19] M. Svedendahl, K. Hult, P. Berglund, J. Am. Chem. Soc. 127 (2005) 17988–17989.
[20] J.-M. Xu, F. Zhang, Q. Wu, Q.-Y. Zhang, X.-F. Lin, J. Mol. Catal. B: Enzym. 49 (2007)
50–54.
[21] G.A. Strohmeier, T. Sovic, G. Steinkellner, F.S. Hartner, A. Andryushkova,
T. Purkarthofer, A. Glieder, K. Gruber, H. Griengl, Tetrahedron 65 (2009)
5663–5668.
4. Conclusions
[22] J. Priego, M. Ortiz-Nava, M. Carillo-Morales, A. Lopez-Munguia, J. Escalante, E.
Castillo, Tetrahedron 65 (2009) 536–539.
Attempts at the use of hydrolytic enzymes as catalysts for
a conjugate addition of nucleophiles to ␣,␤-unsaturated sulfinyl
derivatives have been made for the first time. The studies proved
that the simple mechanism proposed for the analogous lipase-
catalyzed conjugated addition of nucleophiles to enones and
acrylonitrile cannot be directly applied. Although in both cases
the “oxyanion hole” of the enzymes is supposed to enhance elec-
trophilicity of the Michael acceptor by the formation of hydrogen
bonds, the non-ionic tetrahedral structure of the sulfinyl group
(contrary to the planar oxy anion intermediate in the case of the
Michael addition to enones) is probably responsible for a less
effective location of the heteroatom substrates within the space
available. Nevertheless, in the case of a doubly activated accep-
tor, bearing both the sulfinyl and phosphoryl moieties at the same
sp2 carbon atom, the reaction takes place in a weakly stereose-
lective manner, which can be taken as proof for the involvement
of the enzyme, being here the only source of chirality. However,
an alternative mechanism, resting upon the enzyme “alternate site
promiscuity”, must also be taken into consideration. Further studies
on this problem are under way.
[23] J.-F. Cai, Z. Guan, Y.-Y. He, J. Mol. Catal. B: Enzym. 68 (2011) 240–244.
[24] C. Li, X.-W. Feng, N. Wang, Y.-J. Zhou, X.-Q. Yu, Green Chem. 10 (2008) 616–618.
[25] Y. Cai, X.-F. Sun, N. Wang, X.-F. Lin, Synthesis (2004) 671-674.
[26] Y. Cai, S.-P. Yao, Q. Wu, X.-F. Lin, Biotechnol. Lett. 26 (2004) 525–528.
[27] J.-L. Wang, J.-M. Xu, Q. Wu, D.-S. Lv, X.-F. Lin, Tetrahedron 65 (2009) 2531–2536.
[28] M. Svedendahl, B. Jovanovic, L. Fransson, P. Berglund, ChemCatChem 1 (2009)
252–258.
[29] C. Branneby, P. Carlqvist, A. Magnusson, K. Hult, T. Brinck, P. Berglund, J. Am.
Chem. Soc. 125 (2003) 874–875.
[30] P. Kiełbasin´ ski, M. Mikołajczyk, in: T. Matsuda (Ed.), Biocatalysis, Elsevier, 2007,
pp. 159–203.
[31] P. Kiełbasin´ ski, M. Rachwalski, M. Mikołajczyk, M. Szyrej, M.W. Wieczorek, R.
Wijtmans, F.P.J.T. Rutjes, Adv. Synth. Catal. 349 (2007) 1387–1392.
[32] M. Rachwalski, M. Kwiatkowska, J. Drabowicz, M. Kłos, W.M. Wieczorek,
M. Szyrej, L. Sieron´ , P. Kiełbasin´ ski, Tetrahedron: Asymmetry 19 (2008)
2096–2101.
[33] S. Les´niak, M. Rachwalski, E. Sznajder, P. Kiełbasin´ ski, Tetrahedron: Asymmetry
20 (2009) 2311–2314.
[34] M. Rachwalski, S. Les´niak, P. Kiełbasin´ ski, Tetrahedron: Asymmetry 22 (2011)
1325–1327.
[35] S. Kaczmarczyk, M. Kwiatkowska, L. Madalin´ ska, A. Barbachowska, M. Rach-
walski, J. Błaszczyk, L. Sieron´ , P. Kiełbasin´ ski, Adv. Synth. Catal. 353 (2011)
2446–2454.
[36] M. Kwiatkowska, G. Krasin´ ski, M. Cypryk, T. Cierpiał, P. Kiełbasin´ ski, Tetrahe-
dron: Asymmetry 22 (2011) 1581–1590.
[37] P. Kiełbasin´ ski, P. Łyz˙ wa, M. Mikołajczyk, M. Gulea, M. Lemaire, S. Masson,
Tetrahedron: Asymmetry 16 (2005) 651–655.
[38] P. Łyz˙ wa, A. Jankowiak, M. Kwiatkowska, M. Mikołajczyk, P. Kiełbasin´ ski,
A. Betz, P.-A. Jaffres, A.-C. Gaumont, M. Gulea, Tetrahedron Lett. 48 (2007)
351–355.
References
[1] M. Svedendahl Humble, P. Berglund, Eur. J. Org. Chem. (2011) 3391–3401.
[2] E. Busto, V. Gotor-Fernandez, V. Gotor, Chem. Soc. Rev. 39 (2010) 4504–4523.
[3] Q. Wu, B.-K. Liu, X.-F. Lin, Curr. Org. Chem. 14 (2010) 1966–1988.
[4] K. Hult, P. Berglund, Trends Biotechnol. 25 (2007) 231–238.
[5] O. Khershonsky, C. Roodveldt, D.S. Tawfik, Curr. Opin. Chem. Biol. 10 (2006)
498–508.
[39] M. Gulea, M. Kwiatkowska, P. Łyz˙ wa, R. Legay, A.-C. Gaumont, P. Kiełbasin´ ski,
Tetrahedron: Asymmetry 20 (2009) 293–297.
[40] S.-J. Chai, Y.-F. Lai, J.-C. Xu, H. Zheng, Q. Zhu, P.-F. Zhang, Adv. Synth. Catal. 353
(2011) 371–375.
[41] A. Taglieber, H. Ho˝ benreich, J.D. Carballeira, R.J.G. Mondiere, M.T. Reetz, Angew.
[6] U. Bornscheuer, R.J. Kazlauskas, Angew. Chem. Int. Ed. 43 (2004) 6032–6040.
[7] H.I. Park, L.-J. Ming, Angew. Chem. Int. Ed. 38 (1999) 2914–2916.
Chem. Int. Ed. 46 (2007) 8597–8600.