A study on the enzyme catalysed enantioselective hydrolysis of methyl 2-methyl-4-oxopentanoate, a precursor of chiral γ-butyrolactones

Article abstract of DOI:10.1080/10242422.2018.1502274

Porcine pancreas lipase (PPL) resolution of the α-methyl group of racemic methyl 2-methyl-4-oxopentanoate, a valuable synthetic precursor of fragrances and marine natural products, was enhanced by salt modulation of the enzymatic hydrolysis. For the enantioselective hydrolysis of the title ester, PPL was selected from a series of esterases and lipases, and its enantioselectivity was evaluated by changing the reaction medium parameters. The use of 1.6 mol?L^–1 sodium sulfate in phosphate buffer (pH 7.2) improved the enantioselectivity allowing the formation of methyl (2R)-(+)-2-methyl-4-oxopentanoate and (2S)-(–)-2-methyl-4-oxopentanoic acid with an enantiomeric excess of >99% and 71%, respectively. The study showed that a modulation of PPL enantioselectivity could be achieved by using kosmotropic salts in the reaction media. The present method consists of a practical and low-cost option to improve enzymatic kinetic resolution reactions.

Full text of DOI:10.1080/10242422.2018.1502274

Biocatalysis and Biotransformation
ISSN: 1024-2422 (Print) 1029-2446 (Online) Journal homepage: http://www.tandfonline.com/loi/ibab20
A study on the enzyme catalysed enantioselective
hydrolysis of methyl 2-methyl-4-oxopentanoate, a
precursor of chiral γ-butyrolactones
Edgard A. Ferreira, Sindy V. A. Rodezno, Álvaro T. Omori & Rodrigo L. O. R.
Cunha
To cite this article: Edgard A. Ferreira, Sindy V. A. Rodezno, Álvaro T. Omori & Rodrigo L. O. R.
Cunha (2018): A study on the enzyme catalysed enantioselective hydrolysis of methyl 2-methyl-4-
oxopentanoate, a precursor of chiral γ-butyrolactones, Biocatalysis and Biotransformation, DOI:
10.1080/10242422.2018.1502274
To link to this article: https://doi.org/10.1080/10242422.2018.1502274
View supplementary material
Published online: 15 Nov 2018.
Submit your article to this journal
Article views: 7
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=ibab20

BIOCATALYSIS AND BIOTRANSFORMATION
https://doi.org/10.1080/10242422.2018.1502274
RESEARCH ARTICLE
A study on the enzyme catalysed enantioselective hydrolysis of methyl
2-methyl-4-oxopentanoate, a precursor of chiral c-butyrolactones
a
Edgard A. Ferreiraa,b , Sindy V. A. Rodeznoa , Alvaro T. Omori
and Rodrigo L. O. R. Cunhaa
ꢀ
a
b
^
ꢀ
Centro de Ciencias Naturais e Humanas, Universidade Federal do ABC, Santo Andre, Brasil; Escola de Engenharia, Universidade
~
Presbiteriana Mackenzie, Sao Paulo, Brasil
ABSTRACT
ARTICLE HISTORY
Received 18 May 2018
Accepted 9 July 2018
Porcine pancreas lipase (PPL) resolution of the a-methyl group of racemic methyl 2-methyl-4-
oxopentanoate, a valuable synthetic precursor of fragrances and marine natural products, was
enhanced by salt modulation of the enzymatic hydrolysis. For the enantioselective hydrolysis of
the title ester, PPL was selected from a series of esterases and lipases, and its enantioselectivity
was evaluated by changing the reaction medium parameters. The use of 1.6 mol L–1 sodium sul-
fate in phosphate buffer (pH 7.2) improved the enantioselectivity allowing the formation of
methyl (2R)-(þ)-2-methyl-4-oxopentanoate and (2S)-(–)-2-methyl-4-oxopentanoic acid with an
enantiomeric excess of >99% and 71%, respectively. The study showed that a modulation of
PPL enantioselectivity could be achieved by using kosmotropic salts in the reaction media. The
present method consists of a practical and low-cost option to improve enzymatic kinetic reso-
lution reactions.
KEYWORDS
Enzymatic kinetic resolution;
Hofmeister effect; salting-
out; medium engineering;
interfacial enzymes
Introduction
ubiquitous in biological and chemical sciences as rec-
ognized by Franz Hofmeister at the end of 19th cen-
tury (Kunz, Henle, et al. 2004). Despite being a subject
of continuous debate (Kunz, Lo Nostro, et al. 2004; Lo
Nostro and Ninham 2012; Salis and Ninham 2014;
Jungwirth and Cremer 2014), the enzyme activity
modulation by salts as result of the direct interaction
among ions and the protein surface has been shown
for Aspergillus niger (Pinna et al. 2005) and Candida
rugosa lipase (Salis et al. 2007), alkaline phosphatase
(Yang et al. 2010), human kallikrein-related peptidase
3 (KLK3 or PSA) (Andrade et al. 2010; Huang et al.
2001) and HIV protease (Heyda et al. 2009) in aque-
ous media.
Hofmeister effects are related to the interactions of
kosmotropic and chaotropic ions with enzymes or
solvent that changes their structure, stability and cata-
lytic properties (Baldwin 1996; Zhang and Cremer
2006; Peng et al. 2016; Endo et al. 2016, 2018). The
intent of this work consists in the optimization of the
enantioselective hydrolysis of methyl 2-methyl-4-oxo-
pentanoate (methyl 2-methyllevulinate), a known pre-
cursor of chiral 2,4-disubstituted c-butyrolactones of
synthetic utility as building blocks for the synthesis of
Enzymes have an important catalytic role in contem-
porary organic synthesis and allow access to a pleth-
ora of enantiomerically enriched compounds (Zheng
and Xu 2011; Clouthier and Pelletier 2012; Reetz 2013;
Nestl et al. 2014). In spite of this, biocatalysts are use-
ful tools for synthesis even for industrial processes
where problems related to scale-up of the reactions
can be circumvented (Wells et al. 2012). The utility of
biocatalysts for synthetic purposes resides in their abil-
ity to display promiscuity when subjected to react
with different substrates of their natural substrate in
some cases (Hult and Berglund 2007). Such reactions
have been optimized for efficiency improvement. This
goal is reached by managing the reaction media con-
ditions or by modification of the biocatalyst itself by
molecular biology techniques in the directed evolution
of enzymes (Bornscheuer et al. 2012). The former strat-
egy is straightforward because the optimization
involves the variation of reaction conditions such as
substrate concentration, pH, temperature and the use
of co-solvents (Faber 2011).
A medium variable that modulate enzyme activity
is the use of salt additives. Salt or ion effects are
^
Centro de Ciencias Naturais e Humanas, Universidade Federal do ABC, Santo
CONTACT Rodrigo L. O. R. Cunha
Andre, Brasil
rodrigo.cunha@ufabc.edu.br
ꢀ
Supplemental data for this article can be accessed here.
ß 2018 Informa UK Limited, trading as Taylor & Francis Group

2
E. A. FERREIRA ET AL.
jasplakinolide (Xu et al. 2013), geodiamolides A and B
(Hirai et al. 1990), milbemycin b3 (Barrett et al. 1986)
cepacia lipase, porcine pancreas lipase 100–400
unit mg–1, Amano Lipase sp. > 500 U g–1, Candida
rugosa lipase >100 U g–1 and Rhizopus niveus lipase
3.6 U g–1 were purchased from Sigma-Aldrich Co. The
esterase from Sacharomyces cerevisiae 2.2 U mg–1,
esterase from horse liver 0.52 U mg–1, esterase from
Candida lipolytica 0.10 U mg–1, esterase from Rhyzopus
oryzae 151 U g–1 and esterase from Mucor miehei 1.1
U mg–1 were purchased from Fluka Analytical.
Analytical thin-layer chromatography (TLC) was per-
formed with aluminum-backed silica plates coated
with a 0.25 mm thickness of silica gel 60 F254 (Merck),
exposure to vanillin solution and heating. Standard
chromatographic purification methods were followed
using 35–70 mm (240–400 mesh) silica gel purchased
from Sigma-Aldrich.
€
and amphinolide B1 (Furstner et al. 2009). The access
to chiral 2-methyl levulinate derivatives was demon-
strated by a stereoselective hydrogenation transfer
from a Hantzsch ester catalyzed by a chiral binaphtol
phosphate (Martin and List 2006) or by the stereospe-
cific ene reductase activity of Old Yellow Enzyme
(Korpak and Pietruszka 2011) on the same substrate,
2-methyl-4-oxopent-2-enoate (Figure 1).
In such scenario, a biocatalytic access of both enan-
tiomers of the title c-ketoester would be valuable for
synthetic purposes.
Here, we report the optimization of the enantioselec-
tive kinetic resolution of lipases and esterases from dif-
ferent sources on methyl 2-methyl-4-oxopentanoate
hydrolysis by screening medium conditions. Kosmotropic
salt effects improved the biocatalyst performance and
afforded (2R)-(þ)-methyl 2-methyl-4-oxopentanoate and
(2S)-(–)-methyl 2-methyl-4-oxopentanoic acid.
Nuclear magnetic resonance spectra were recorded
on a Bruker Ultrashield 300 spectrometer at 300 MHz
1
1
for H- and 75 MHz for 13C-NMR. The H-NMR chem-
ical shifts (d) are reported in ppm relative to the TMS
peak and 13C-NMR chemical shifts (d) are reported in
ppm relative to CDCl3. The coupling constants (J) are
given in hertz (Hz). Optical rotations were determined
on a JASCO DIP-378 polarimeter (Sodium D line at 589
nm). The enzymatic kinetic resolutions were con-
ducted at Thermomixer Comfort from Eppendorf
except during scale-up reactions that were conducted
in an orbital shaker (Solab). Reactions under micro-
wave irradiation were performed in a CEM Discover at
Materials and methods
General
All commercially available chemicals were purchased
from Sigma-Aldrich (USA) and used without further
purification. Solvents were purchased from Synth
(Brazil) and used without further purification. CAL-B
(Candida antactica lipase ꢀ10,000 U g–1), Pseudomonas
ꢁ
50 W and a maximum temperature of 75 C.
Figure 1. Approaches for the preparation of chiral 2-methyl-4-oxopentanoic acid derivatives.

BIOCATALYSIS AND BIOTRANSFORMATION
3
The ratio of nitro esters was determined in a Varian
type 450-GC flame ionization detector (FID) equipped
with a Varian CP-Sil 5 CB capillary column (30 m ꢂ
0.25 mm ꢂ0.25 um) using nitrogen as the carrier gas.
The conditions used for separation are: injector 270
ꢁC, detector 270 ꢁC, hydrogen 300 mL min–1, air 30
mL min–1. The column flow was 1.05 mL min–1 with a
Effect of addition of salts to reaction medium and
temperature
Racemic methyl 2-methyl-4-oxopentanoate (5) (50 mg)
and porcine pancreas lipase (PPL) (25 mg) were added
to 5 mL of a 0.1 mol L–1 sodium phosphate buffer solu-
tion (pH 7.2) in a 15 mL Falcon tube containing different
salts as depicted in Table 2. These mixtures were submit-
maximum temperature of 250 ꢁC. The temperature
ꢁ
ted to orbital stirring (500 rpm) at 25 C for 2 and 4 h.
–1
ꢁ
ꢁ
program was 50 C for 5 min increasing 10.0 C min
to 250 C.
After this period, each reaction mixture was filtered
through a cotton plug, and the pH was adjusted to 1.0
and extracted with AcOEt (3 ꢂ 2 mL). The organic
phases were combined, dried over MgSO4, filtered and
aliquots were analyzed by a chiral gas chromatography
to determine the enantiomeric excess of products and
unreacted starting materials (Table 1).
ꢁ
The enantiomeric excess of compounds was deter-
mined in a Shimadzu type GC-2010 flame ionization
detector (FID) chromatograph fitted with a Chirasil-
DEX CB b-cyclodextrin (25 m ꢂ 0.25 mm, 25 lm)
fused silica capillary column using nitrogen as the car-
rier gas. The conꢁditions used for the chiral separation
The procedure described above was also applied in
ꢁ
are: injector 250 C, detector 250 C, split 29.9, hydro-
ꢁ
ꢁ
different temperatures (32 C and 40 C) (Table 2).
gen 400 mL min–1 and air 30 mL min–1. The column
flow was 1.05 mL min–1 at a maximum temperature of
170 ꢁC. The temperature program was 50 ꢁC for 14
Enzymatic kinetic resolution of racemic methyl 2-
methyl-4-oxopentanoate (5) by PPL
–1
ꢁ
ꢁ
min increasing to 0.5 C min to 67 C and then 10
–1
ꢁ
ꢁC min to 160 C.
Racemic methyl 2-methyl-4-oxopentanoate (5) (50 mg)
and PPL (25 mg) were added to 5 mL of a 0.1 mol L–1
sodium phosphate buffer solution containing 1.6
mol L–1 of sodium sulfate at pH 7.2 in a 15 mL Falcon
tube. This mixture was submitted to orbital stirring
Purification of racemic keto ester was performed in a
€
Kugelrohr Buchi B-580. The enantiomeric excess of keto
ester and keto acid were calculated according to equa-
tion: ((E1 – E2)/(E1 þ E2)) ꢂ 100, where E1 and E2 are
the ratio of enantiomers of each compound (5 and 6) as
determined by chiral GC analyses. The conversions were
ꢁ
(500 rpm) at 32 C for 10 h. Next, the solution was fil-
tered in a cotton plug, and the pH was adjusted to
10–11 using a 2 mol L–1 solution of NaOH and
extracted with AcOEt (4 ꢂ 2 mL). The organic phases
were combined, dried over MgSO4 and the solvent
removed under reduced pressure to afford methyl
(2R)-(þ)-2-methyl-4-oxopentanoate (R)-5 in >99% ee
calculated according to equation: Conv. = eee/(eea
þ
eee) ꢂ 100. The enantiomeric ratios (E) were calculated
according to equation: E = {ln[eeP(1 ꢃ eeS)]/(eeP þ
eeS)}/{ln[eeP(1 þ eeS)]/(eeP þ eeS)}(Faber 2011).
25
as a colorless oil ½aꢄD = þ12.5 (c 0.8, CHCl3). The car-
Screening of enzymes for kinetic resolution of
racemic methyl 2-methyl-4-oxopentanoate (5)
boxylic acid (S)-6 was recovered from the aqueous
solution by saturating it with NaCl, and the pH was
adjusted to 1.0 using 5 mol L–1 HCl solution and wash-
ing with AcOEt (5 ꢂ 2 mL). The organic phases were
combined, dried over MgSO4 and solvent removed
under reduced pressure to afford the (2S)-(–)-2-methyl-
4-oxopentanoic acid with 71% ee (71%).
The compound 5 was synthesized according to a pre-
viously reported procedure with some modifications
(see supporting information).
Racemic methyl 2-methyl-4-oxopentanoate (5) (50
mg) and 25 mg of selected enzyme were added to 5
mL of a 0.1 mol L–1 sodium phosphate buffer solution
(pH 7.2) in a 15 mL Falcon tube. These mixtures were
This reaction was also performed using 500 mg of
(rac)-5 and PPL (250 mg) in 50 mL of a 0.1 mol L–1
sodium phosphate buffer solution containing 1.6 mol L–1
ꢁ
submitted to orbital stirring (500 rpm) at 25 C for 2
ꢁ
h. After this period, each reaction mixture was filtered
through a cotton plug, and the pH was adjusted to
1.0 and extracted with AcOEt (3 ꢂ 2 mL). The organic
phases were combined, dried over MgSO4, filtered and
aliquots were analyzed by a chiral gas chromatog-
raphy to determine the enantiomeric excess of prod-
ucts and unreacted starting materials (see supporting
information S21).
of sodium sulfate at pH 7.2 at 32 C. In this condition,
the reaction was performed in Erlenmeyer flask using
orbital stirring at 250 rpm affording methyl (2R)-(þ)-2-
methyl 4-oxopentanoate (R)-5 0.165 g (33% yield, >99%
ee) and (2S)-(–)-2-methyl-4-oxopentanoic acid (S)-6 0.271
g (60% yield, 71% ee), CG tR [(R)-5] = 44.21 min, [(S)-5] =
44.92 min, [(R)-6] = 44.21 min and [(S)-6] = 44.92 min,
Chirasil-DEX CB b-cyclodextrin (25 m ꢂ 0.25 mm, 25

4
E. A. FERREIRA ET AL.
Table 1. Effect of salts in PPL-catalyzed hydrolysis of (rac)-5.
2 hours
4 hours
Salt additive
Medium Ionic Strength (mol L–1
)
eee (%)
eea (%)
conv. (%)
E
eee (%)
eea (%)
conv. (%)
E
–
0.246
1.835
2.030
3.500
4.950
1.835
1.998
NS
9.0
9.4
13.2
12.5
21.5
1.8
19.4
–
26.4
23.3
–
59.5
59.7
68.7
67.7
80.4
3.1
90.4
–
87.3
92.1
–
10.0
10.4
14.2
13.5
22.5
2.8
17.7
–
23.2
20.2
–
4.3
4.3
6.1
5.9
11.3
1.1
24.0
–
13.8
13.7
24.4
21.2
46.5
0.8
20.7
–
35.6
50.2
–
62.3
67.1
71.7
74.1
84.4
5.6
87.8
–
86.0
92.6
–
14.8
13.7
25.3
22.2
47.5
1.8
19.1
–
29.3
35.2
–
4.9
5.8
7.7
8.3
18.7
1.1
18.8
–
NaCl (1.6 mol L–1
)
Na2SO4 (0.6 mol L–1
Na2SO4 (1.1 mol L–1
Na2SO4 (1.6 mol L–1
)
)
)
KF (1.6 mol L–1
)
K2SO4 (0.6 mol L–1
K2SO4 (1.0 mol L–1
)
)
Sodium citrate (0.6 mol L–1
Sodium citrate (1.0 mol L–1
Sodium citrate (1.6 mol L–1
)
)
)
3.554
5.778
NS
19.1
30.3
–
18.8
43.2
–
Conditions: Phosphate buffer 0.1 M (5.0 mL/pH 7.2), temperature ¼ 25 ꢁC, substrate ¼ 50 mg, enzyme ¼ 25 mg. eea ¼ enantiomeric excess of carboxylic
acid. eee ¼ enantiomeric excess of keto ester. NS ¼ Not soluble in phosphate buffer 0.1 M.
were added to 5 mL of a 0.1 mol L–1 sodium phos-
Table 2. Temperature effect on PPL-catalyzed hydrolysis of
(rac)-5.
phate buffer solution of phosphate containing 1.6
mol L–1 Na2SO4 at pH 7.2 and 32 ꢁC. This mixtuꢁre
was submitted to orbital stirring (500 rpm) at 32 C
for 9 h.
2 hours
4 hours
Temp. (ꢁC) eee (%) eea (%) Conv. (%)
E
eee (%) eea (%) Conv. (%)
E
25
32
40
21.5
32.8
43.5
80.4
81.5
79.3
22.5 11.3 46.5
28.7 13.4 64.9
35.4 13.3 64.5
84.4
83.1
80.4
47.5 18.7
43.8 21.1
44.5 17.8
Next, the mixture was filtered with a cotton plug,
and the pH adjusted to 11 using a 2.0 mol L–1 solution
of NaOH and extracted with AcOEt (4 ꢂ 2 mL) to
remove the unreacted ester. The carboxylic acid was
then recovered from the aqueous solution by saturat-
ing it with NaCl, and the pH was adjusted to 1–2
using 5 mol L–1 HCl and washed with AcOEt (5 ꢂ 2
mL). The organic phases were combined, dried over
MgSO4 and solvent removed under reduced pressure
to afford the (2S)-(–)-2-methyl-4-oxopentanoic acid (S)-
Conditions: Phosphate buffer 0.1 M (5.0 mL/pH 7.2), temperature ¼ 25 ꢁC,
substrate ¼ 50 mg, enzyme ¼ 25 mg. eea ¼ enantiomeric excess of car-
boxylic acid. eee ¼ enantiomeric excess of keto ester.
lm), temperature gradient: 50 ꢁC (ꢁ14 min iso), 0.5
–1
–1
ꢁ
ꢁ
ꢁC min to 67 C, 10 C min to 160 C.
Recycling of (2S)-(–)-2-methyl-4-oxopentanoic acid
(S)-6 (71% ee)
25
To a 100 mL one neck round-bottomed flask a DMF (55
mL) solution of enantioenriched (2S)-(–)-2-methyl-4-oxo-
pentanoic acid (71% ee) (2.60 g, 20 mmol) was cooled to
6 with excellent enantiomeric excess (>99%), ½aꢄD
25
–18.1 (c 0.8, CHCl3), literature ½aꢄD –21.1 (c 0.8, CHCl3)
(Hoffman and Kim 1995).
ꢁ
0 C. Anhydrous potassium carbonate (5.53 g, 40 mmol)
This reaction was also performed in 500 mg scale
of enantioenriched methyl (2S)-(–)-methyl 2-methyl-4-
oxopentanoate (S)-5 and PPL (250 mg) in 50 mL of a
0.1 mol L–1 sodium phosphate buffer solution contain-
was slowly added to this solution and allowed to react
for 15 min before addition of methyl iodide (60 mmol,
3.12 mL) in one portion. The reaction mixture was stirred
ing 1.6 mol L–1 of sodium sulfate at pH 7.2 at 32 C
ꢁ
ꢁ
at 0 C for 30 min and then for another 1.5 h at room
temperature. Next, water was added until complete dis-
solution of K2CO3. This solution was extracted with ethyl
acetate as needed to extract all c-keto esters. (The extrac-
tion was monitored by TLC using 50% ethyl acetate as
the eluent). The organic phases were combined, dried
over MgSO4, filtered and concentrated under reduced
pressure to afford methyl (2S)-(–)-2-methyl-4-oxopenta-
noate (71% ee) as colorless oil in 85% yield (2.45g). The
enantiomeric excess was unchanged as determined by
chiral gas chromatography.
for 9 h.
Here, the reactions were made in Erlenmeyer flasks
using orbital stirring at 250 rpm to afford unreacted
ester (0.113 g) and (2S)-(–)-2-methyl-4-oxopentanoic
acid (S)-6 (0.267 g; 61% yield, > 99% ee).
1H-NMR: (300 MHz, CDCl3) d 3.03–2.87 (m, 1H),
2.55–2.48 (m, 1H), 2.17 (s, 3H), 1.22 (d, J = 6.9 Hz,
3H); 13C-NMR (75 MHz, CDCl3): 207.3 (C=O), 181.3
(C=O), 46.3 (CH2), 34.6 (CH), 29.9 (CH3), 16.8 (CH3);
MS (70 eV, EI): m/z (%) 130 (1.6), 115 (3), 112 (11),
87 (7), 70 (14), 61 (15), 58 (7), 45 (15), 43 (100), 42
(13), 41 (10), 39 (5), CG tR [(S)-6] = 44.92 min,
Chirasil-DEX CB b-cyclodextrin (25 m ꢂ 0.25 mm, 25
lm), temperature gradient: 50 ꢁC (14 min iso), 0.5
ꢁC min–1 to 67 ꢁC, 10 ꢁC min–1 to 160 ꢁC, Rf 0.52
(AcOEt/MeOH, 3:2).
Enzymatic kinetic resolution of methyl (2S)-(–)-2-
methyl-4-oxopentanoate (S)-5 (71% ee) by PPL
The enantioenriched (2S)-(–)-methyl 2-methyl-4-
oxopentanoate (S)-5 (50 mg) and 25 mg of PPL

BIOCATALYSIS AND BIOTRANSFORMATION
5
Figure 2. Synthesis of racemic keto ester 5.
pressure to afford 0.19 g of (2R)-(þ)-2-methyl-4-oxo-
pentanoic acid (R)-6 in 72% yield; CG tR [(R)-6] = 44.21
min, Chirasil-DEX CB b-cyclodextrin (25 m ꢂ 0.25 mm,
Preparation of methyl (2S)-(–)-2-methyl-4-
oxopentanoate (S)-5
The same methodology as used for (S)-6 was applied
here. To a 25 mL one neck round-bottomed flask, a
DMF (10 mL) solution of (2S)-(–)-2-methyl-4-oxopenꢁta-
noic acid (S)-6 (0.52 g, 4.0 mmol) was cooled to 0 C.
Anhydrous potassium carbonate (1.38 g, 10 mmol)
was added to this solution portion-wise and after add-
ition, the mixture was allowed to react for 15 min
before addition of methyl iodide (16 mmol, 1.00 g,
0.44 mL) inꢁ one portion. The reaction mixture was
stirred at 0 C for 30 min and then for more 1.5 h at
room temperature. Next, water was added until com-
plete dissolution of K2CO3, and this solution was
extracted with ethyl acetate as much as necessary to
extract all c-keto ester. (The extraction was monitored
by TLC using 50% ethyl acetate as the eluent). The
organic phases were combined, dried over MgSO4, fil-
tered and concentrated under reduced pressure to
afford (2S)-(–)-methyl 2-methyl-4-oxopentanoate (S)-5
as a colorless oil in 80% yield (0.46 g). The enantio-
meric excess was unchanged as determined by chiral
gas chromatography analysis; CG tR [(S)-5] = 44.92
min, Chirasil-DEX CB b-cyclodextrin (25 m ꢂ 0.25 mm,
ꢁ
25 lm), temperature gradient: 50 C (14 min iso), 0.5
25
ꢁC min–1 to 67 ꢁC, 10 ꢁC min–1 to 160 ꢁC, ½aꢄD
=
þ17.6 (c 0.8, CHCl3).
Results and discussion
Preparation of racemic substrates
The title compound was prepared via a two reactions
sequence starting from the condensation of methyl
methacrylate and nitroethane followed by a Nef reac-
tion of the intermediate c-nitroester (Figure 2). The
first reaction was a DBU-catalyzed Michael addition of
the nitroethane (1) to neat methyl methacrylate (2)
under microwave irradiation. This resulted in a mixture
of the c-nitro ester (3) and dimethyl 2,4,6-trimethyl-4-
ꢀ
~
nitro-heptanedionate (4) (Escalante and Dıaz-Coutino
2009). The second step was adapted from literature
(Ballini 1993) where the c-nitro ester (3) was subjected
to a Nef reaction to afford the desired racemic methyl
2-methyl-4-oxopentanoate (5) in good overall yield.
The Nef reaction was carried out using the crude mix-
ture of 3 and 4 because the diester 4 is unreactive
toward Nef reaction conditions and is thus easily sepa-
rated from 5. The amount of base used to perform
the Nef reaction had to be increased 10-fold to
achieve moderate to good yields. The desired keto-
ester 5 was easily separated from 4 by a horizontal
distillation.
ꢁ
25 lm), temperature gradient: 50 C (14 min iso), 0.5
25
–1
–1
ꢁC min to 67 C, 10 C min to 160 C, ½aꢄD = –12.9
ꢁ
ꢁ
ꢁ
(c 0.8, CHCl3).
Preparation of (2R)-(þ)-2-methyl-4-oxopentanoic
acid (R)-6
To a 50 mL one neck round-bottomed flask, a solution
of (2R)-(þ)-methyl-2-methyl-4-oxopentanoate (R)-5
(0.288 g, 2 mmol) was prepared in THF:H2O (10 mL:3
mL). To this solution was added LiOH (0.09 g, 2.1
mmol) with stirring for 90 min. THF was removed
under reduced pressure and an additional 10 mL of
water was added. The pH was adjusted to 1 and
extracted with AcOEt (4ꢂ). The organic phases were
reunited and washed with saturated NaCl, dried with
MgSO4, and the solvent removed under reduced
Screening of hydrolases for best resolution of
(rac)-5
Using the keto ester (rac)-5, a series of lipases and
esterases was used in standard hydrolysis conditions
(Delinck and Margolin 1990; Thodi et al. 2009) to
screen the best hydrolysis enantioselectivities (Figure
3, Supporting Information Table S21). Although the
short reaction time invariably contributed to the

6
E. A. FERREIRA ET AL.
observed low E-values, a close inspection on the
obtained results revealed two potential enzymes –
porcine pancreas lipase (PPL) and the esterase from
horse liver. In our adopted approach, (rac)-5 was more
quickly hydrolyzed by the horse liver esterase
(Supporting Information Table S21, entry 7), however
moderate enantioselectivity in favor of the acid (S)-6
was observed. The evaluation of the enzyme catalyzed
hydrolyses were made by chiral gas-chromatography
as indicated in the experimental section. Only CAL-B
showed the opposite enantioselectivity for the
hydrolysis as observed by the chromatograms. In con-
trast, the PPL showed better performance, a better E-
value and a good ratio for the enantiomeric excess of
the acid and the ester (Supporting Information Table
S21, entry 3). Although only Candida antarctica lipase
B showed an opposite selectivity for the hydrolysis of
(rac)-5 (Supporting Information Table S21, entry 1), the
poor conversion led us to discard this enzyme for fur-
ther studies.
denaturation and a decrease in the oil–water interface
extension. Such effects are related to the well-known
interfacial action of PPL (Schmid and Verger 1998;
Kapoor and Gupta 2012).
Salt effects on PPL-catalyzed hydrolysis
enantioselectivity
As this kinetic resolution reaction was not optimized
yet, we decided to examine the effect of the presence
of salts in the reaction media due to the interfacial
mechanism of lipase catalysis. Kosmotropic salts
modulate water surface tension by increasing the salt-
ing-out of hydrophobic substances and decreases pro-
tein solubility. This behavior would be positive for the
studied reaction (Andrade et al. 2010). Such effects on
solute–solvent interactions biased by salts, known as
Hofmeister effect are seldom considered in biocatalytic
reactions for synthetic purposes. Thus, we decided to
explore this approach using two kosmotropic salts,
sodium chloride and sodium sulfate which led to a
positive process improvement (Table 1). The addition
of salts increased the ionic strength of medium and
Study of reaction conditions on the hydrolysis of
(rac)-5 promoted by PPL
2–
suppressed the effect of HPO4 however, the same
The influence of other reaction conditions in PPL-cata-
lyzed hydrolysis was examined to improve the hydroly-
sis stereoselectivity of racemic-5 in an extended reaction
time of 4 h. This included the pH of the 0.1 mol L–1
phosphate buffer. We used values of 6.2, 7.2 and 8.0
(see Supporting Information Table S22). The enantiose-
lectivity in pH 7.2 was slightly better both after 2 and 4
h of reaction. In this case, the enantiomeric excess of
the acid (S)-6 was 62% after 4 h of reaction.
Co-solvents normally improve the solubility of
water-immiscible substrates for enzyme-mediated
transformations (Carrea and Riva 2000; Adamczak and
Krishna 2004; Shen et al. 2008). Therefore, we exam-
ined the influence of the co-solvents in the enantiose-
lective hydrolysis of rac-5. Six organic solvents were
evaluated using 10% of their content in 0.1 mol L–1
phosphate buffer (pH 7.2). No co-solvent could
improve the reaction performance (see Supporting
Information Table S23). The solvents led to a decrease
in the enantioselectivity factor, conversion and the
enantiomeric excess of (S)-6. This is likely a dilution
effect for water immiscible solvents. Nevertheless,
water soluble solvents promote both enzyme
concentration of sodium sulfate doubled the conver-
sion and increased the enantiomeric excess of (S)-6
and the enantioselectivity factor in a two-fold factor.
These results may be explained by the differences
in the reaction medium in both cases as can be
noticed by the ionic strengths’ of each reaction as
long the dynamics of substrate and enzyme partitions
also changes. Although the concentration of sodium
sulfate was not further increased due to the limit of
its solubility in the phosphate buffer, the concentra-
tion decrease of this salt led to an adverse effect in
kinetic resolution efficiency.
Next, we performed the reaction using potassium
sulfate as a salt additive. In this case, the increase
regarding the conversion from 2–4 h was meager. This
result prompted us to discard this salt. Interestingly,
the cation exchange from sodium to potassium led to
a noticeable decrease in reaction rate which may be
explained by the increased chaotropic character of the
softer potassium ion. If our initial hypothesis were cor-
rect, an enhanced chaotropic salt would impair the
reaction efficiency. In fact, we selected potassium
fluoride as
a
negative control which sternly
Figure 3. Enzymatic hydrolysis of c-keto ester (rac)-5.

BIOCATALYSIS AND BIOTRANSFORMATION
7
handicapped both the reaction conversion and the
enantioselectivities as seen in Table 1.
Temperature effect on optimized PPL-catalyzed
hydrolysis of (rac)-5
The temperature effect was also studied, and it showed
ꢁ
a slight increase in the enantiomeric ratio (E) at 32 C
(Table 2). Despite the expected increase in the reaction
conversion as the temperature rises to 40 ꢁC, the
enantiomeric excess of the desired acid decreased.
ꢁ
Thus, the optimal reaction temperature was 32 C. The
E-value of this condition (21.1) was the best so far and
presented good conversion of racemic ester (44%).
Figure 4. Time course of PPL-catalyzed (rac)-5 hydrolysis in
preparative scale. (ꢅ: (R)-5 ee; : (S)-6 ee; ᭡: ester conversion).
᭺
Enzymatic kinetic resolution of methyl 2-methyl-4-
oxopentanoate
With the optimized reaction conditions determined,
the PPL-catalyzed enantioselective hydrolysis of (rac)-5
was carried out and monitored for 10 h. All efforts
were made to increase the enantiomeric excess of the
(2S)-2-methyl-4-oxopentanoic acid (S)-6, but it was not
possible to increase its enantiomeric excess over 95%,
desired for synthetic purposes. Alternatively, the
hydrolysis reaction was allowed to surpass the 50%
conversion where a decrease on (S)-6 enantiomeric
excess was expected. Despite this, a higher enantio-
meric excess of the less PPL-reactive ester enantiomer,
(R)-5, was >99% after 10 h (Figure 4).
At this point, the enantiomeric excess of the (2S)-2-
methyl-4-oxopentanoic acid, (S)-6, was 71% and its
enantiopurity was improved after a second PPL-cata-
lyzed kinetic resolution reaction using the enantioen-
riched methyl ester. It took 6 h to provide the
Figure 5. Time course of PPL-catalyzed hydrolysis of enantio-
enriched (S)-5 in preparative scale. The starting ee for (S)-5
was 71%. The red dotted line indicates the maximum amount
of acid (S)-6 starting that can be obtained from enantioen-
᭺
riched (S)-5 with 71% ee; this is 85.5%. ( : (S)-5 ee; ᭿: (R)-5
ee; ꢅ: (S)-6 ee; ᭡: ester conversion).
Figure 6. CD spectra of enantiomerically pure acids(R)-(þ)-6 and(S)-(–)-6 and esters(R)-(þ)-5 and (S)-(–)-5.

8
E. A. FERREIRA ET AL.
V. Comasseto, A. A. dos Santos and J.H.G. Lago for mater-
ial analyses.
carboxylic acid with high enantiomeric excess (>99%)
(Figure 5). After this period, the reaction rate was
slower, and the most abundant ester enantiomer is
the enantiomer (R)-5 (see Supporting Information for
the complete dataset).
Disclosure statement
The chiroptical properties of both enantiomers of 2-
methyl-4-oxopentanoic acid (6), and 2-methyl-4- oxo-
pentanoate (5) were fully characterized by polarimetry
and by circular dichroism. The R enantiomers of 5 and
6 were levorotatory, and their enantiomers were dex-
tro rotatory. In chloroform, enantiomerically pure acid
(R)-(þ)-6 showed first and second positive Cotton
effects at 230 and 266 nm, respectively, but the corre-
sponding third Cotton effect was negative at 299 nm.
The circular dichroism spectrum of enantiomeric (S)-
(–)-6 in chloroform was symmetrical to those of the
acid (R)-(þ)-6 (Figure 6(A)). The CD spectra of enantio-
meric esters were acquired in chloroform solution. The
ester (R)-(þ)-5 showed first and second positive
Cotton effect at 228 and 264 nm, and the third Cotton
effect was negative at 297 nm; not surprisingly the
spectrum of ester (S)-(–)-5 was symmetrical to those of
its enantiomer (Figure 6(B)).
No potential conflict of interest was reported by the authors.
Funding
~
This work was supported by grants provided by Fundac¸ao
ꢁ
~
de Amparo a Pesquisa do Estado de Sao Paulo (FAPESP)
[2009/00617-2] and Conselho Nacional de Pesquisa (CNPq)
~
[487012/2012-7]. EAF and SVAR thank Coordenac¸ao de
ꢀ
Aperfeicoamento de Pessoal de Nıvel Superior (CAPES) and
~
ꢀ
Coordenac¸ao de Aperfeic¸oamento de Pessoal de Nıvel
Superior and Organization of American States cooperation
program (CAPES-OEA) for scholarships.
ORCID
Edgard A. Ferreira
Sindy V. A. Rodezno
1123-6548
http://orcid.org/0000-0002-8109-156X
http://orcid.org/0000-0003-
ꢀ
Alvaro T. Omori
http://orcid.org/0000-0002-0115-3485
Rodrigo L. O. R. Cunha
0203-3143
http://orcid.org/0000-0002-
Conclusions
Taken together, the results in this work showed that
sodium sulfate, a kosmotropic salt, exerted a positive
effect in the performance of a porcine pancreas lipase,
an interfacial biocatalyst, in the kinetic resolution reac-
tion of the methyl ester of 2-methyl-levulinic acid. Thus,
the method studied herein allowed access to both
enantiomers of synthetically valuable 2-methyl-levulinic
acid derivatives in high optical purities in a two-step
biocatalytic and sustainable process; it complements
the previously reported stereospecific biocatalytic reac-
tion which uses old yellow protein and related ene-
reductases for the same purpose. Also, the approaches
to improve biocatalytic reactions related to reaction
media variables, salt effects are seldom explored. The
addition of salts is an affordable approach to modulate
biocatalysts action to improve processes. Nonetheless,
the exact mechanism of the salts, i.e. Hofmeister effect,
is still a complex subject, since it modulates both the
dynamics between reagents and the macromolecular
catalysts leading to improved reaction conditions and
thus it is worth to be studied more systematically when
one designs new enzyme-catalyzed reactions.
References
Adamczak M, Krishna SH. 2004. Strategies for improving
enzymes for efficient biocatalysis. Food Technol
Biotechnol. 42:251–264.
Andrade D, Assis DM, Lima AR, Oliveira JR, Araujo MS, Blaber
SI, Blaber M, Juliano MA, Juliano L. 2010. Substrate specifi-
city and inhibition of human kallikrein-related peptidase 3
(KLK3 or PSA) activated with sodium citrate and glycosa-
minoglycans. Arch Biochem Biophys. 498:74–82.
Baldwin RL. 1996. How Hofmeister ion interactions affect
protein stability. Biophys J. 71:2056–2063.
Ballini R. 1993. 5-Nitro-1-pentene as a precursor for the syn-
thesis of allylrethrone. Synthesis. 7:687–688
Barrett AGM, Carr RAE, Attwood SV, Richardson G, Walshe
NDA. 1986. Total synthesis of (þ)-milbemycin.beta.3. J Org
Chem. 51:4840–4856.
Bornscheuer UT, Huisman GW, Kazlauskas RJ, Lutz S, Moore
JC, Robins K. 2012. Engineering the third wave of
biocatalysis. Nature. 485:185–194.
Carrea G, Riva S. 2000. Properties and synthetic applications
of enzymes in organic solvents. Angew Chem Int Ed Engl.
39:2226–2254.
Clouthier CM, Pelletier JN. 2012. Expanding the organic tool-
box: a guide to integrating biocatalysis in synthesis. Chem
Soc Rev. 41:1585–1605.
Delinck DL, Margolin AL. 1990. Synthesis of chiral building
blocks for selective adenosine receptor agents.
Lipase-catalyzed resolution of 2-benzylpropanol and
2-benzylpropionic acid. Tetrahedron Lett. 31:6797–6798.
Acknowledgements
The authors would like to thank Profs. L. Juliano and T. J.
Brocksom for the critical reading of this manuscript, Profs. J.

BIOCATALYSIS AND BIOTRANSFORMATION
9
Endo A, Kurinomaru T, Shiraki K. 2016. Hyperactivation of
a-chymotrypsin by the Hofmeister effect. J Mol Catal B
Enzym. 133:S432–S438.
Martin NJA, List B. 2006. Highly enantioselective transfer
hydrogenation of a,b-unsaturated ketones. J Am Chem
Soc. 128:13368–13369.
Nestl BM, Hammer SC, Nebel BA, Hauer B. 2014. New gener-
ation of biocatalysts for organic synthesis. Angew Chem
Int Ed Engl. 53:3070–3095.
Peng D, Zhang C, Welborn M, Shepherd JJ, Zhu T, Voorhis
TV, Pentelute BL. 2016. Salt effect accelerates site-selective
cysteine bioconjugation. ACS Cent Sci. 2:637–646.
Pinna MC, Salis A, Monduzzi M, Ninham BW. 2005.
Hofmeister series: the hydrolytic activity of Aspergillus
niger lipase depends on specific anion effects. J Phys
Chem B. 109:5406–5408.
Reetz MT. 2013. Biocatalysis in organic chemistry and bio-
technology: past, present, and future. J Am Chem Soc.
135:12480–12496.
Salis A, Bilanicova D, Ninham BW, Monduzzi M. 2007.
Hofmeister effects in enzymatic activity: weak and strong
electrolyte influences on the activity of Candida rugosa
lipase. J Phys Chem B. 111:1149–1156.
Salis A, Ninham BW. 2014. Models and mechanisms of
Hofmeister effects in electrolyte solutions, and colloid and
protein systems revisited. Chem Soc Rev. 43:7358–7377.
Schmid RD, Verger R. 1998. Lipases: interfacial enzymes with
attractive applications. Angew Chem Int Ed Engl.
37:1608–1633.
Shen LL, Wang F, Mun HS, Suh M, Jeong JH. 2008. Solvent-
dependent reactivity in porcine pancreatic lipase
(PPL)-catalyzed hydrolysis. Tetrahedron: Asymmetry. 19:
1647–1653.
Thodi K, Barbayianni E, Fotakopoulou I, Bornscheuer UT,
Constantinou-Kokotou V, Moutevelis-Minakakis P, Kokotos
G. 2009. Study of the removal of allyl esters by Candida
antarctica lipase B (CAL-B) and pig liver esterase (PLE).
J Mol Catal B Enzym. 61:241–246.
Wells AS, Finch GL, Michels PC, Wong JW. 2012. Use of
enzymes in the manufacture of active pharmaceutical
ingredients-a science and safety-based approach to
ensure patient safety and drug quality. Org Process Res
Dev. 16:1986–1993.
Endo A, Kurinomaru T, Shiraki K. 2018. Hyperactivation of serine
proteases by the Hofmeister effect. Mol Catal. 455:32–37.
ꢀ
~
Escalante J, Dıaz-Coutino FD. 2009. Synthesis of c-nitro ali-
phatic methyl esters via michael additions promoted by
microwave irradiation. Molecules. 14:1595–1604.
Faber K. 2011. Biotransformations in organic chemistry: a
textbook. Heidelberg: Springer-Verlag.
€
Furstner A, Bouchez LC, Morency L, Funel JA, Liepins V,
ꢀ
Poree FH, Gilmour R, Laurich D, Beaufils F, Tamiya M.
2009. Total syntheses of amphidinolides B1, B4, G1, H1
and structure revision of amphidinolide H2. Chemistry.
15:3983–4010.
ꢀ
ꢀ
Heyda J, Pokorna J, Vrbka L, Vacha R, Jagoda-Cwiklik B,
ꢀ
Konvalinka J, Jungwirth P, Vondrasek J. 2009. Ion
specific effects of sodium and potassium on the catalytic
activity of HIV-1 protease. Phys Chem Chem Phys.
11:7599–7604.
Hirai Y, Yokota K, Yamaaki T, Momose T. 1990. A total syn-
thesis of (þ)-Geodiamolides a and b, the novel cyclodesi-
peptides. Heterocycles. 30:1101–1119.
Hoffman RV, Kim HO. 1995. The stereoselective synthesis of
2-alkyl.gamma.-keto acid and heterocyclic ketomethylene
peptide isostere core units using chiral alkylation by
2-triflyloxy esters. J Org Chem. 60:5107–5113.
Huang X, Knoell CT, Frey G, Hazegh-Azam M, Tashjian AH,
Hedstrom L, Abeles RH, Timasheff SN. 2001. Modulation of
recombinant human prostate-specific antigen: activation
by hofmeister salts and inhibition by azapeptides.
Appendix: thermodynamic interpretation of the activation
by concentrated salts. Biochemistry. 40:11734–11741.
Hult K, Berglund P. 2007. Enzyme promiscuity: mechanism
and applications. Trends Biotechnol. 25:231–238.
Jungwirth P, Cremer PS. 2014. Beyond Hofmeister. Nat
Chem. 6:261–263.
Kapoor M, Gupta MN. 2012. Lipase promiscuity and its bio-
chemical applications. Proc Biochem. 47:555–569.
Korpak M, Pietruszka J. 2011. Chemoenzymatic one-pot syn-
thesis of c-butyrolactones. Adv Synth Catal. 353:1420–1424.
Kunz W, Henle J, Ninham BW. 2004. ‘Zur Lehre von der
Wirkung der Salze’ (about the science of the effect of
salts): Franz Hofmeister’s historical papers. Curr Opin
Colloid Interface Sci. 9:19–37.
Xu YY, Liu C, Liu ZP. 2013. Advances in the total synthesis of
cyclodepsipeptide (þ)-Jasplakinolide (Jaspamide) and its
analogs. Curr Org Synth. 10:67–89.
Yang Z, Liu XJ, Chen C, Halling PJ. 2010. Hofmeister effects
on activity and stability of alkaline phosphatase. Biochim
Biophys Acta. 1804:821–828.
Kunz W, Lo Nostro P, Ninham BW. 2004. The present state of
affairs with Hofmeister effects. Curr Opin Colloid Interface
Sci. 9:1–18.
Zhang Y, Cremer PS. 2006. Interactions between macromole-
cules and ions: the Hofmeister series. Curr Opin Chem
Biol. 10:658–663.
Lo Nostro P, Ninham BW. 2012. Hofmeister phenomena:
an update on ion specificity in biology. Chem Rev.
112:2286–2322.
Zheng GW, Xu JH. 2011. New opportunities for biocatalysis:
driving the synthesis of chiral chemicals. Curr Opin
Biotechnol. 22:784–792.