Bovine serum albumin-catalysed cross aldol condensation: Influence of ketone structure

Article abstract of DOI:10.1016/j.procbio.2019.08.003

Bovine serum albumin (BSA) catalyses the cross aldol condensation and proved to be catalytically active at mild temperature and in ethanol, a cheap and green solvent, contrasting with the strong or expensive reaction media usually employed for this reaction. We herein report the reaction of a set of ketones (butanone, 3-pentanone, cyclopentanone and cyclohexanone) with benzaldehyde and p-nitrobenzaldehyde which provided high conversions (77–95%) of the corresponding enones (isolated in a range of yields from 19% to 74%). Parameters assayed to achieve these conversion values were solvent, ketone/aldehyde molar ratio and temperature. In this procedure only cyclohexanone gave the bis-enone, by-product of the conventional aldol condensation, in low amount even at high benzaldehyde/cyclohexanone molar excess. Under the assayed conditions null or low ketol amounts were observed, except for the reaction of cyclopentanone and p-nitrobenzaldehyde. Moreover, kinetic data of BSA-catalysed aldol condensation of cyclohexanone and p-nitrobenzaldehyde suggest an ordered bi bi mechanism for enone formation; an enamine mechanism involving residues of the catalytic cavity exhibiting abnormal pK_a values is also proposed.

Full text of DOI:10.1016/j.procbio.2019.08.003

Journal Pre-proof
Bovine serum albumin-catalysed cross aldol condensation: Influence of
ketone structure
´
Sebastian M. Ardanaz, Estefan´ıa L. Borucki, Ana J. Velez Rueda,
Gustavo Parisi, Adolfo M. Iribarren, Luis E. Iglesias
PII:
S1359-5113(19)30553-7
DOI:
https://doi.org/10.1016/j.procbio.2019.08.003
PRBI 11728
Reference:
To appear in:
Process Biochemistry
Received Date:
Revised Date:
Accepted Date:
14 April 2019
14 July 2019
7 August 2019
Please cite this article as: Ardanaz SM, Borucki EL, Velez Rueda AJ, Parisi G, Iribarren AM,
Iglesias LE, Bovine serum albumin-catalysed cross aldol condensation: Influence of ketone
structure, Process Biochemistry (2019), doi: https://doi.org/10.1016/j.procbio.2019.08.003
This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal
pertain.
© 2019 Published by Elsevier.

Bovine serum albumin-catalysed cross aldol condensation: Influence of
ketone structure
Sebastián M. Ardanaz1, Estefanía L. Borucki1, Ana J. Velez Rueda2, Gustavo
Parisi2, Adolfo M. Iribarren1 and Luis E. Iglesias1,*
1-Laboratory of Biocatalysis and Biotransformation, Department of Science and
Technology, National University of Quilmes, Roque Sáenz Peña 352, (1876) Bernal,
Provincia de Buenos Aires, Argentina.
2-Structural Bioinformatics Group, Department of Science and Technology, National
University of Quilmes, Roque Sáenz Peña 352, (1876) Bernal, Provincia de Buenos Aires,
Argentina.
* Corresponding author. Tel: + 54-11-4365-7100; fax: + 54-11-4365-7100 - E-mail:
leiglesias@unq.edu.ar
Graphical abstract
O
O
R1
R2
1 R1 = H , R2 = CH3
8 R1 = CH3, R2 = CH3
14 R1-R2 = (CH2)2
19 R1-R2= (CH2)3
R1
R2
R3
Bovine
serum
albumin
77-95% conversion
O
4 R1 = H , R2 = CH3, R3 = H
12 R1 = CH3, R2 = CH3, R3 = H
13 R1 = CH3, R2 = CH3, R3 = NO2
17 R1-R2 = (CH2)2, R3 = H
22 R1-R2 = (CH2)3, R3 = H
23 R1-R2 = (CH2)3, R3 = NO2
R3
EtOH / 45°C
2 R3 = H
9 R3 = NO2
1

Highlights
 Bovine serum albumin (BSA) catalyses the cross aldol condensation of a set of
ketones
 BSA is active at mild temperature and in ethanol, a cheap and green solvent
 Bis-enone is formed only from cyclohexanone and in low amount
 An ordered bi-bi mechanism for BSA-catalysed enone formation is proposed
Abstract
Bovine serum albumin (BSA) catalyses the cross aldol condensation and proved to be
catalytically active at mild temperature and in ethanol, a cheap and green solvent,
contrasting with the strong or expensive reaction media usually employed for this
reaction. We herein report the reaction of a set of ketones (butanone, 3-pentanone,
cyclopentanone and cyclohexanone) with benzaldehyde and p-nitrobenzaldehyde which
provided high conversions (77-95%) of the corresponding enones (isolated in a range of
yields from 19% to 74%). Parameters assayed to achieve these conversion values were
solvent, ketone/aldehyde molar ratio and temperature. In this procedure only
cyclohexanone gave the bis-enone, by-product of the conventional aldol condensation, in
low amount even at high benzaldehyde/cyclohexanone molar excess. Under the assayed
conditions null or low ketol amounts were observed, except for the reaction of
cyclopentanone and p-nitrobenzaldehyde. Moreover, kinetic data of BSA-catalysed aldol
condensation of cyclohexanone and p-nitrobenzaldehyde suggest an ordered bi bi
mechanism for enone formation; an enamine mechanism involving residues of the
catalytic cavity exhibiting abnormal pKa values is also proposed.
Key words: aldol condensation, biocatalysis, bovine serum albumin, catalytic
promiscuity, enones
2

1. Introduction
Enzymes are sustainable catalysts, which operate under mild reaction conditions in
natural and non-natural media, recognising structurally diverse substrates with high regio-
, chemo- and stereoselectivity. Owing to these features biocatalytic procedures are
nowadays useful tools in synthetic organic chemistry. In addition, the synthetic potential
of biotransformations is increasing thanks to the discovery of catalytic promiscuity,
another property of enzymes. Defined as the ability of an active site to catalyse chemically
different reactions to those occuring physiologically and involving different transitions
states [1-3], catalytic promiscuity is supposed to be an inherent property of enzymes and
not limited to exceptions [1,4]. The synthetic usefulness of the latter feature is evidenced
by examples reviewed in the last years, most of them dealing with hydrolases [5-8].
Furthermore, promiscuity is not restricted to enzymes but it seems to be a general property
of proteins [4], so that new synthetic opportunities may arise from proteins whose role in
vivo is not catalysis. However, this catalytic potential has not been much explored.
Albumin, the most abundant protein of mammalian blood plasma, is in vivo a carrier
protein; in vitro, commercially available serum bovine albumin (BSA) is able to catalyse
many chemically different reactions. It is considered as a promiscuous catalyst, whose
activity is attributed to an unusually acid lysine residue (Lys-222) placed in the
subdomain IIA of the protein [9,10].
Cross aldol addition and condensation are very known and useful carbon-carbon bond-
forming transformations because of the versatility of the resulting skeletons. Much of the
current research on cross aldol reactions aims at a stereoselective and sustainable access
to cross aldol addition products; since aldolases show limited substrate tolerance,
alternatives based on enzymatic catalytic promiscuity affording different degrees of
enantioselectivity have been explored [5-8]. In this regard, BSA catalyses the cross aldol
addition of acetone and 6-methoxy-2-naphtaldehyde in neutral aqueous solution,
displaying moderate enantioselectivity; the reaction follows Michaelis-Menten kinetics
as for an enzymatic process and the activity is conserved in a polypeptide derived from
the albumin site IIA sequence [11,12]. The latter results support that the observed
catalysis is nor unspecific or due to impurities in commercial albumin [9,11]. Besides, in
the study of the retroaldol reaction of methodol the experiments performed with two
different commercial preparations of BSA showed activity, while this was supressed by
3

acetylation of BSA, enforcing the evidence that the observed activity is not caused by a
contaminant [13].
Focusing now on ,-unsaturated carbonyl compounds, the aldol condensation
products obtained by dehydration of cross aldol adducts, their synthesis is traditionally
accomplished in strong alkali, which promotes the formation of subproducts such as bis-
enones [14,15] and Michael adducts [16]. Nevertheless, very few works report the
synthesis of ,-unsaturated carbonyl compounds exploiting biocatalytic promiscuity.
The examples in literature are a tandem aldol addition-dehydration using a D-
aminoacylase and imidazole, which yielded ,-unsaturated carbonyl products in
generally good yield [17], and the production of cinnamaldehyde in 50% yield from
acetaldehyde and benzaldehyde, promoved by an engineered 4-oxalocrotonate
tautomerase [18]. In regard to BSA, it catalyses the condensation of acetone with a set of
substitued benzaldehydes in ionic liquid, affording the corresponding enones in good and
high yields [19]. In a recent paper we have reported the BSA-catalysed preparation of
-unsaturated ketones from acetone and a set of substitued benzaldehydes, using cheap
and easily available reaction media such as water or ethanol as reaction medium. High
enone yields (88-97%) could be obtained except for the tested nitrobenzaldehydes, which
gave mixtures of the corresponding ketol and enone; moreover, it was shown that albumin
catalyses also the dehydration step of the aldol condensation [20]. To our best knowledge,
only acetone has been reported as donor in BSA-catalysed aldol condensations and this
prompted us to extend our procedure to other ketones, in order to get knowledge on the
scope of the reaction. Prediction of the substrate performance is usually quite limited in
biocatalysis; a recent representative example related to promiscuous aldol activity,
dealing with the application of deep eutectic solvents (DES) as medium for porcine
pancreatic lipase (PPL)-catalysis, has reported that for a given aldehyde the reaction
affords ketol (aldol addition adduct) as the major product, but the amount of formed enone
(,-unsaturated carbonyl) product is dependent on the assayed ketone [21]. In addition,
the prediction of biocatalyst and solvent behaviour is also limited; a very recent paper has
reported that BSA-catalysed reaction of acetone with p-nitrobenzaldehyde in DES affords
unselectively or with low selectivity a mixture of ketol and enone [22], contrasting with
the results obtained with PPL [21].
Considering the previous discussion, which stressed the limited prediction of substrate
performance, the potential of catalytic promiscuity to provide a mild access to synthetic
4

useful ,-unsaturated ketones and the interest on procedures involving green and
inexpensive reaction media, in this paper we report the results of BSA-catalysed cross
aldol condensation of a set of ketones (Scheme 1).
2. Materials and Methods
2.1. General
Bovine serum albumin (BSA) was purchased from Sigma–Aldrich (fraction V, a
purified and non-immobilised preparation) and used without any treatment except for
control experiments. Substrates, reagents and solvents were of analytical grade and
obtained from commercial suppliers. For the biotransformations absolute ethanol was
employed; before use, benzaldehyde was filtered through basic aluminium oxide and then
distilled; butanone was dried over sodium carbonate, distilled and stored over molecular
sieves 3 Å.
Reactions were followed by thin-layer chromatography with Silicagel 60 F254 plates
(Merck); plates were developed employing hexane/ethyl acetate 70:30 v/v and developed
with UV light. Flash column chromatography was carried out using 0.040–0.063 mm
Silicagel 60 (Merck).
HPLC analyses were done with a C18 column (length: 150 mm; internal diameter:
4.6 mm; particle size: 5 m). Analysis of aliquots from biotransformations of 1 and 2 was
conducted applying a 5 min methanol/water gradient from 30:70 v/v (5 min) to 50:50 v/v
(5 min) followed by a 7 min gradient to pure methanol (2 min), at a flow rate of 0.9 mL
min−1 and detection at 274 nm. Reaction times were: 3: 16.3 min; 4: 22.1 min; 5: 14.8
min; 7: 22.3 min. Samples from biotransformations of 8, 14 and 19 were analysed by a
8 min gradient (the same solvents) from 50:50 v/v (2 min) to pure methanol (3 min), at
the same flow rate and detection at 254 nm. Retention times were: 10 + 11: 13.3 min; 12:
15.3 min; 13: 14.5 min; 15 + 16: 6.1 min; 17: 13.1 min; 18: 13.5 min; 20: 7.6 min; 21:
14.5 min; 22: 9.0 min; 23: 16.1 min.
1H-NMR and 13C-NMR spectra were recorded on a Bruker Avance II spectrometer at
500 MHz and 125 MHz, respectively; chemical shifts (δ) are reported in ppm from TMS
and CDCl3 as solvent. The multiplicity was indicated by the following abbreviations: s,
1
singlet; sbr, broad singlet; d, doublet; t, triplet; m, multiplet. Where needed, 2D H-13C
HSQC allowed the indicated signal assignments.
5

Kinetic studies were carried out by measuring absorption at 330 nm on a Biotek
Cytation 5 imaging reader.
2.2. BSA-catalysed cross aldol condensation of ketones (1, 8, 14 and 19) and
aldehydes (2 and 9)
2.2.1. General protocol in analytical scale
Typically, experiments consisted of a solution of the aldehyde (0.05 mmol) and the
ketone (its concentration is indicated in each case) in the reported solvent or mixture of
solvents (reaction total volume: 1 mL). BSA (54 mg) was added to the solution and the
mixtures were stirred at 200 rpm and the indicated temperature. The course of the
biotransformation was followed by TLC and HPLC of aliquots taken at different times,
which were centrifuged prior to analysis.
2.2.2. Preparative protocol for butanone and 3-pentanone
According to the previous procedure in analytical scale, preparative
biotransformations were carried out in reaction mixtures containing BSA (540 mg) and a
solution of the aldehyde (2, 9; 0.5 mmol) and the ketone (1, 8; its concentration is
indicated in each case) in the below indicated solvent or mixture of solvents (total reaction
total volume: 10 mL). The reaction mixtures were shaken at 45°C and 200 rpm during the
below indicated times, BSA was then filtered off and washed with ethanol (5 mL). The
solvents were eliminated in vacuo and the resulting residues purified by silicagel flash
column chromatography. The following products were isolated after elution with the
below detailed solvents, obtaining the indicated yields (for NMR characterisation of the
products, see Supplementary Material):
2.2.2.1. From butanone and benzaldehyde
(E)-1-Phenyl-1-penten-3-one (4) + (E)-3-methyl-4-phenyl-3-buten-2-one (7):
Reaction solvent and time: butanone/ethanol 20:80, 168 h; elution solvent:
toluene; yield: 33% (ratio 4/7 = 3.7, determined by the ratio of H5-4 and H1-7 1H NMR
signals); Rf = 0.52 (hexane/ethyl acetate 80:20).
Different preparative conditions were employed to isolate the ketols (for
structural elucidation); column chromatography afforded a mixture of 3 and 5 and pure
6:
6

1-Hydroxy-1-phenyl-3-pentanone (3) + Syn-4-hydroxy-3-methyl-4-phenyl-2-
butanone (5):
Reaction solvent and time: butanone/water 50:50, 168 h; elution solvent: toluene;
yield: 9% (ratio 5/3 = 2.1, determined by the ratio of H4-5 and H1-3 1H NMR signals);
Rf = 0.21 (hexane/ethyl acetate 80:20).
Anti-4-hydroxy-3-methyl-4-phenyl-2-butanone (6):
Reaction solvent and time: butanone/water 50:50, 168 h; elution solvent: toluene;
yield: 3%; Rf = 0.20 (hexane/ethyl acetate 80:20).
2.2.2.2. From 3-pentanone and benzaldehyde
(E)-2-Methyl-1-phenyl-1-penten-3-one (12):
Reaction solvent and time: butanone/ethanol 20:80, 48 h; elution solvent
hexane/ethyl acetate 98:2; yield: 31%; Rf = 0.60 (hexane/ethyl acetate 70:30).
2.2.2.3. From 3-pentanone and p-nitrobenzaldehyde
(E)-2-Methyl-1-(4-nitrophenyl)-1-penten-3-one (13):
Reaction solvent and time: butanone/ethanol 20:80, 96 h; elution solvent
hexane/ethyl acetate 95:5; yield: 19%; Rf = 0.31 (hexane/ethyl acetate 80:20).
By subsequent elution, ketols Syn-1-Hydroxy-2-methyl-1-(4-nitrophenyl)-3-
pentanone (10) + Anti-1-hydroxy-2-methyl-1-(4-nitrophenyl)-3-pentanone (11) were
obtained as a mixture:
Reaction solvent and time: butanone/ethanol 20:80, 96 h; elution solvent:
hexane/ethyl acetate 95:5; yield: 32% (ratio 10/11= 2.3, determined by the ratio of H1-10
and H1-11 1H NMR signals); Rf = 0.14 (hexane/ethyl acetate 80:20).
2.2.3. Preparative protocol from cyclohexanone and cyclopentanone
Adapting the procedure for analytical scale, preparative biotransformations were
conducted in reaction mixtures consisting of BSA (540 mg) and a solution of the aldehyde
(2, 9; 0.5 mmol) and the ketone (14, 19, 1.0 mmol) in ethanol (reaction total volume: 10
mL). These mixtures were shaken at 45°C and 200 rpm during the below indicated time,
then the work up described in Section 2.2.2. was applied. The following products were
isolated by silicagel flash column chromatography, eluting with the below detailed
solvents and affording the indicated yields (for NMR characterisation of the products, see
Supplementary Material):
2.2.3.1. From cyclohexanone and benzaldehyde
(E)-2-Benzylidenecyclohexanone (20):
7

Time: 120 h; elution solvent: hexane/ethyl acetate 98:2; yield: 26%; Rf = 0.58
(hexane/ethyl acetate 70:30).
Subsequent elution gave the corresponding bis-enone, (2E,6E)-2,6-
dibenzylidenecyclohexanone (22):
Time: 120 h; elution solvent: hexane/ethyl acetate 98:2; yield: 11%; Rf = 0.62
(hexane/ethyl acetate 70:30).
2.2.3.2. From cyclohexanone and p-nitrobenzaldehyde
(E)-2-(4-nitrobenzylidene)cyclohexanone (21):
Time: 96 h; elution solvent: toluene; yield: 74%; Rf = 0.26 (toluene).
Subsequent elution gave the corresponding bis-enone, (2E,6E)-2,6-bis(4-
nitrobenzylidene)cyclohexanone (23):
Time: 96 h; elution solvent: toluene; yield: 12%; Rf = 0.30 (toluene).
2.2.3.3. From cyclopentanone and benzaldehyde
(E)-2-Benzylidenecyclopentanone (17):
Time: 144 h; elution solvent: hexane/ethyl acetate 98:2; yield: 28%; Rf = 0.62
(hexane/ethyl acetate 80:20).
2.2.3.4. From cyclopentanone and p-nitrobenzaldehyde
(E)-2-(4-nitrobenzylidene)cyclopentanone (18):
Time: 48 h; elution solvent: hexane/ethyl acetate 94:6; yield: 22%; Rf = 0.48
(hexane/ethyl acetate 70:30).
Subsequent elution gave
nitrophenyl)methyl] cyclopentanone
nitrophenyl)methyl]cyclopentanone (16):
a
mixture of
ketols syn-2-[Hydroxy(4-
and anti-2-[Hydroxy(4-
(15)
Time: 48 h; elution solvent: hexane/ethyl acetate 90:10; yield: 31% (ratio 15/16 =
1
1.9, determined by the ratio of H1'-15 and H1'-16 H NMR signals); Rf = 0.26
(hexane/ethyl acetate 70:30).
2.3. Kinetics of BSA-catalysed condensation of cyclohexanone with p-
nitrobenzaldehyde
54 mg of BSA were added to a solution of cyclohexanone and p-nitrobenzaldehyde
in ethanol (1 mL total volume); the concentration of cyclohexanone was fixed at four
values (0.3, 0.6, 1 and 5 M) and in each case, p-nitrobenzaldehyde was tested at eight
concentrations (1, 5, 25, 50, 75, 100, 150 and 300 mM). The resulting systems were
shaken at 45ºC and initial velocities obtained by measuring during the first 10 h of
8

reaction the concentration of the produced enone. Control experiments were done with
the corresponding reaction mixtures without BSA. All the experiments were conducted
by triplicate and results are given in Figures 1 and 2.
2.4. In silico BSA titration and pKa shift analysis
For the theoretical titration of BSA, PROPKA software was employed [23]. To take
into account the contribution of the protein conformational diversity in the fluctuations
of the observed pKas shifts, all conformers avaible in Protein Data Bank database were
used for the calculations (PDB IDs: 4RO0_A,B; 4JK4_A,B; 4F5S_A,B and 3V03_A,B).
3. Results and discussion
3.1. BSA-aldol condensation of butanone (1)
BSA-catalysed reaction of butanone (Scheme 1) with benzaldehyde (2) was first
studied; due to the existence of two types of carbonyl -hydrogens in butanone, a mixture
of regioisomeric and diasteromeric ketols and enones is usually obtained by chemical
catalysis [24, 25]. Regarding catalytic promiscuity, as far as we know, the only works
that deal with butanone report its reaction with p-nitrobenzaldehyde, one applying pepsin
catalysis and conducting to a ketol (4-hydroxy-3-methyl-4-(4-nitrophenyl)-2-butanone)
in 25% yield [26], the other using a thermophilic esterase in heptane, giving a 69%
conversion towards the regioisomeric ketol 1-hydroxy-1-(4-nitrophenyl)-3-pentanone
(71% ee) [27].
When we assayed the reaction in neat butanone, no product was observed (Table 1,
Entry 1); since our previous work on BSA-catalysed aldol condensation of acetone with
a series of benzaldehydes showed that the conversion is very dependent on the content of
ethanol and water in the reaction mixture [20], the effect of these solvents was studied by
screening binary and tertiary solvent mixtures. Moreover, since in a previous study higher
conversions to enone were reached at 45 ºC [20], experiments were carried out at this
temperature. Among the assayed butanone-water mixtures (Entries 2-10), only butanone-
water 1:1 was biphasic and in all cases, low or very low conversions towards a mixture
of ketols (3, 5 and 6) and enones (4 and 7), whose major component was enone 4. With
butanone-ethanol mixtures (Entries 11-17 and 21) the reaction extent depended on
ethanol content: the lower ratios (Entries 11 and 12) afforded low conversions, while
quantitative benzaldehyde conversions were obtained at 50-90% ethanol (Entries 14-17
9

and 22). The main product was enone 4, formed in high conversions (81-88%) and
moderate regioselectivity (enone 4 / enone 7 ratio = 5-8); assayed tertiary mixtures of
butanone-water-ethanol (Entries 23-30) yielded mixtures of ketols and enones. Major
formation of the thermodynamically less stable enone 4, obtained from the less hindered
alpha-carbon of butanone carbonyl, could be explained by steric factors in the catalytic
subdomain of the protein. Control assays without biocatalyst and with biocatalyst
previously treated with urea at 100 ºC were done for the studied biotransformations and
representative results are given in Entries 19 and 20: no reaction takes place without BSA
(Entry 19) and previous treatment of BSA with urea at 100 ºC decreased dramatically its
activity (Entry 20), indicating that its tertiary structure is needed for catalysis. Moreover,
when porcine pancreatic lipase (PPL) was employed as biocatalyst (Entry 21), high
conversion to ketols but low to enones were observed, in contrast to BSA performance.
Results presented in Table 1 indicate that water does not favour conversion and
selectivity of the biotransformation, contrasting with the results obtained in BSA-
catalysed aldol condensation of acetone and benzaldehyde [20]. It is to be mentioned that
neither shorter reaction times nor carrying out the biotransformations at 30 °C were
satisfactory for selective production of ketols (a representative result is provided in Entry
18). Moreover, other solvents were tested (hexane, acetonitrile, isopropanol and 1-
butanol, in mixtures butanone/solvent 20:80). Apart from butanone/acetonitrile 20:80,
which gave low conversion (27% of enone 4 and 4% of enone 7 at 45 ºC and 48 h), no
reaction was observed with the other solvents, showing the dramatic influence of the
solvent on biocatalysed reactions. Thus, ethanol was chosen for subsequent experiments
with other ketones, taking also into account the advantages of a cheap and green solvent.
3.2. BSA-aldol condensation of cyclohexanone (19)
With regard to cyclohexanone, promiscuous enzymes have been reported to catalyse
the addition of this ketone to a set of benzaldehydes to give the corresponding ketols with
variable yields, diastereo- and enantioselectivity [6]: the group of Guan and He applied
proteases such as the acidic protease from Aspergillus usamii [28], chymopapain [29] and
tripsin [30], as well as a nuclease [31]; porcine pancreatic lipase was employed by the
group of Wang and Yu [32,33] and Gotor [21]. Interestingly, a recent work by Porto and
coworkers assessed the results obtained in Rhizopus niveus lipase-catalysed aldol addition
of cyclohexanone as an example of unspecific catalysis [34]. In addition to
cyclohexanone, cyclopentanone (14) also gave the corresponding ketols with variable
10

stereoselectivy [21, 28-32]. It is to be remarked that only one of the previously
commented papers [21] reported the formation of the ,-unsaturated ketones, found in
low percentages. Besides, it is to be stated that in the mentioned works [28-34] BSA had
been tested as a control experiment of the investigated promiscuous enzymes. In most
cases this control experiment was conducted taking the reaction of cyclohexanone with
p-nitrobenzaldehyde as model, obtaining the corresponding ketols with variable yields
(5-68%) and diastereoselectivity but without enantioselectivity. No report of enone
formation was indicated in the mentioned works.
Since cyclohexanone is not volatile, rendering product(s) isolation more complex, the
molar ratio of the substrates was varied in assays of BSA-catalysed reaction of
cyclohexanone with benzaldehyde (Scheme 1), in order to find the lowest ketone
concentration conducting to most satisfactory results (Table 2). Data presented in Entries
1-5 show that very high conversions to enone 20 were obtained (82-95%) even at a very
low excess of cyclohexanone (82%, Entry 1). Interestingly, no product was formed in
the reaction conducted in a mixture of substrates without addition of ethanol (Entry 6),
stressing again the influence of the solvent on these biotransformations. Additional
experiments were done to know if a reversal in the ratio of the substrates could favour the
production of the bis-,-unsaturated carbonyl compound (bis-enone). Entries 7-9 show
that enone 20 continued being the major product: this could be explained considering that
the conversion of enone to dienone may be affected by steric hindrance in the catalytic
subdomain of BSA.
3.3. BSA-aldol condensation of cyclopentanone (14) and 3-pentanone (8)
On the basis of the discussed results, ethanol and a ketone/aldehyde molar ratio = 2
were chosen as experimental conditions to assay cyclopentanone, which afforded the
corresponding enone 17 in high conversion (84%, Table 3, Entry 1). However, in this
case no bis-enone was formed, contrasting to the behaviour of cyclohexanone (15% of
bis-enone 22, Entry 3). 3-Pentanone, which to our best knowledge has not been reported
in experiments on aldol reactions catalysed by promiscuous biocatalysts, was tested under
the best experimental conditions found for butanone (Entry 4) and gave a high conversion
towards enone 12 (87%, Entry 2); again, no bis-enone was detected, suggesting that by-
product formation is dependent on the structure of the ketone. This suggestion is
renforced by results obtained with p-nitrobenzaldehyde (9, Entries 5-7): the bis-enone
was obtained once again only with cyclohexanone (10% of bis-enone 23, Entry 7).
11

Besides, while the enones of 3-pentanone (13) and cyclohexanone (21) were obtained in
high conversions (77% and 88% respectively, Entries 6 and 7), cyclopentanone (Entry
5) gave a mixture of diasteromeric ketols 15 and 16; the latter reaction profile has been
observed by us in the reaction of acetone with p-nitrobenzaldehyde [20].
Finally, some experiments were carried out to assay that BSA catalyses the step of
ketol dehydration. A mixture of ketols 15 and 16 in ethanol was shaken in the presence
of BSA at 45 °C (following the general analytical protocol given in Section 2.2.1) and
two control experiments were also conducted: one was carried out without BSA; the
other, using BSA pretreated with urea at 100 °C. After 48 h, these control experiments
did not show any reaction, while the former assay afforded 31% conversion to enone 18,
reaching 91% after 144 h. These results indicate that BSA also catalyses the dehydration
step of the aldol condensation, in agreement with the results found in BSA-catalysed
dehydration of the ketol from acetone and p-nitrobenzaldehyde [20].
3.3. Kinetics of BSA-catalysed condensation of cyclohexanone with p-
nitrobenzaldehyde
To have a mechanistic approach to the studied biotransformations, kinetics of BSA-
catalysed reaction of cyclohexanone and p-nitrobenzaldehyde in ethanol was studied as
reported in Section 2.3. Figure 1 shows that enone production is accelerated by BSA with
saturation kinetics; a Vmax = 25.24±1.55 mM/h and a Km = 38.53±8.53 mM were
determined. Kinetic experiments could be performed at cyclohexanone saturation but due
to the limited solubility of p-nitrobenzaldehyde, no saturation of this substrate could be
reached. Since the plots of 1/vo versus 1/[9] in Figure 2 are not parallel, a ping pong bi-bi
mechanism can be discarded, suggesting both a random sequential or an ordered
sequential bi-bi mechanism. Taking into account that an enamine mechanism has been
proposed for BSA-catalysed aldol addition of acetone and aromatic aldehydes [11], an
ordered sequential bi-bi mechanism could be proposed (Scheme 2). In this proposal, the
abnormally acid Lys-222 mentioned in the Introduction [9-11] could act as nucleophile
in the formation of the ketol, obtained as a racemic mixture, whose subsequent
dehydration to afford the enone could be favoured by Arg-280. We have studied pKas
shifts of residues which could explain their chemical behaviour; with this purpose, a
theoretical titration of BSA was made using Propka software with available conformers
of the protein (See Section 2.4). The pKas found for Lys-222 (Lys-221 according to our
PDB numeration, [20]) and Arg-280 were 7.78 ± 0.83 and 14.07 ± 0.19 respectively. The
12

proximity and localization in the same cavity of an unusually basic arginine and an
abnormally acid lysine could explain the proposed mechanism.
4. Conclusion
This work presents a deeper insight into bovine serum albumin (BSA) as a
promiscuous biocatalyst for the cross aldol condensation. High conversions towards
enones (77-95%) are reached and except for the reaction of cyclopentanone with p-
nitrobenzaldehyde, null or low ketol amounts were observed (0-13%). Bis-enones, by-
products of the conventional aldol condensation, are formed in low amounts only from
cyclohexanone, even at high benzaldehyde/cyclohexanone molar excess. Compared with
the conditions of the conventional aldol condensation (strong and hot basic media), BSA
provides a procedure for the aldol condensation at mild temperature. Moreover, since it
is catalytically active in ethanol, a cheap and green solvent, toxic or expensive solvents
and catalysts proposed in alternative strategies for aldol condensation can be
circumvented. Besides, kinetic experiments suggest an ordered bi-bi mechanism for BSA-
catalysed enone formation; in addition, a possible mechanism is proposed, involving
residues of the catalytic cavity of the protein exhibiting abnormal pKas values.
Acknowledgements
The authors acknowledge UNQ and ANPCyT, Argentina (PICT 2013-0232, LEI) for
financial support. GP, AMI and LEI are members of the Scientific Research Career of
CONICET (Argentina); SMA and AJVR are research fellows of CONICET.
References
[1] A. Babtie, N. Tokuriki, F. Hollfelder, What makes an enzyme promiscuous? Curr.
Opin. Chem. Biol. 14 (2010) 200-207.
[2] M. Svedendahl Humble, P. Berglund, Biocatalytic promiscuity, Eur. J. Org. Chem.
(2011) 3391-3401.
[3] S.D. Copley, Shining a light on enzyme promiscuity, Curr. Opin. Struct. Biol. 47
(2017) 167-175.
[4] O. Khersonsky, D.S. Tawfik, Enzyme promiscuity: A mechanical and evolutionary
perspective, Annu. Rev Biochem. 79 (2010) 471-505.
13

[5] Y. Miao, M. Rahimi, E.M. Geertsema, G.J. Poelarends, Recent developments in
enzyme promiscuity for carbon-carbon bond-forming reactions, Curr. Opin. Chem.
Biol. 25 (2015) 115-123.
[6] Z. Guan, L.Y. Li, Y.H. He, Hydrolase-catalysed asymmetric carbon-carbon bond
formation in organic synthesis, RSC Adv. 5 (2015) 16801-16814.
[7] M. López-Iglesias, V. Gotor-Fernández, Recent advances in biocatalytic promiscuity:
Hydrolase-catalysed reactions for nonconventional transformations, Chem. Rec. 15
(2015) 743-759.
[8] B.P. Dwivedee, S. Soni, M. Sharma, J. Bhaumik, J.K. Laha, U.C. Banerjee,
Promiscuity of lipase-catalysed reactions for organic synthesis: A recent update,
ChemistrySelect 3 (2018) 2441-2466.
[9] D.C.M. Albanese, N. Gaggero, Albumin as promiscuous biocatalyst in organic
synthesis, RSC Adv. 5 (2015) 10588-10598.
[10] K. Kikuchi, S.N. Thorn, D. Hilvert, Albumin-catalyzed proton transfer, J. Am.
Chem. Soc. 118 (1996) 8184-8185.
[11] F. Benedetti, F. Berti, S. Bidoggia, Aldolase activity of serum albumins, Org.
Biomol. Chem. 9 (2011) 4417-4420.
[12] I. Luisi, S. Pavan, G. Fontanive, A. Tossi, F. Benedetti, A. Savoini, E. Maurizio, R.
Sgarra, D. Sblattero, F. Berti, An albumin-derived peptide scaffold capable of binding
and catalysis, PLoS ONE 8 (2013) e56469.
[13] J. Schmidt, C. Ehasz, M. Epperson, K. Klas, J. Wyatt, M. Hennig, M. Forconi, The
effect of the hydrophobic environment on the retro-aldol reaction: comparison to a
computationally-designed enzyme, Org. Biomol. Chem. 11 (2013) 8419-8425.
[14] U.P. Kreher, A.E. Rosamilia, C.L. Raston, J.L. Scott, C.R. Strauss, Direct
preparation of monoarylidene derivatives of aldehydes and enolizable ketones with
DIMCARB, Org. Lett. 5 (2003) 3107-3110.
[15] Y. Wu, J. Hou, Y. Liu, M. Zhang, C.H. Tung, Y. Wang, Chemoselective Claisen-
Schmidt bis-substitutional condensation catalyzed by an alcoxy-bridge dinuclear
Ti(IV) cluster, Tetrahedron 72 (2016) 1511-1516.
[16] B. Lantaño, J.M. Aguirre, E.V. Drago, M. Bollini, D.J. de la Faba, J.D. Mufato,
Synthesis of benzylidenecycloalkan-1-ones and 1,5-diketones under Claisen-Schmidt
reaction: Influence of the temperature and electronic nature of arylaldehydes, Synth.
Commun. 47 (2017) 2202-2214.
14

[17] X. Chen, B.K. Liu, H. Kang, X.F. Lin, A tandem aldol condensation/dehydration co-
catalyzed by acylase and N-heterocyclic compounds in organic media, J. Mol. Catal.
B:Enzym. 68 (2011) 71-76.
[18] E. Zandvoort, E.M.Geertsema, W.J. Quax, G.J. Poelarends, Enhancement of the
promiscuous aldolase and dehydration activities of 4-oxalocrotonate tautomerase by
protein engineering, ChemBioChem 13 (2012) 1274-1277.
[19] N. Sharma, U.K. Sharma, R. Kumar, N. Katoch, R. Kumar, A.K. Sinha, First bovine
serum-albumin promoted synthesis of enones, cinnamic acids and coumarins in ionic
liquids: An insight into the role of protein impurities in porcine pancreas lipase for
olefinic bond fromation. Adv . Synth. Catal. 353 (2011) 871-878.
[20] S.M. Ardanaz, A.J. Velez Rueda, G. Parisi, A.M. Iribarren, L.E. Iglesias, A mild
procedure for enone preparation catalysed by bovine serum albumin in a green and
easily available medium, Catal. Lett. 148 (2018) 1750-1757.
[21] D. González-Martínez, V. Gotor, V. Gotor-Fernández, Application of deep eutectic
solvents in promiscuous lipase-catalysed aldol reactions, Eur. J. Org. Chem. (2016)
1513-1519.
[22] S. Milker, M. Pätzold, J.Z. Bloh, D. Holtmann, Comparison of deep eutetic solvents
and solvent-free conditions for aldol production, Mol. Catal. 466 (2019) 70-74.
[23] G. Kieseritzky, E.W. Knapp, Improved pKa prediction: Combining empirical and
semimicroscopic methods. J. Comput. Chem. 29 (2008) 2575-2581.
[24] F. Zhang, C. Li, C. Qi, Highly diastero- and enantioselective direct aldol reaction
under solvent free conditions, Tetrahedron:Asymmetry 24 (2013) 380-388.
[25] T. Müller, K. Djanashvili, J.A. Peters, I.W.C.E. Arends, U. Hanefeld, Tetrahedral
boronates as basic catalysts in the aldol reaction, Z. Naturforsch. 70 (2015) 587-595.
[26] C. Li, Y.J. Zhou, N. Wang, X.W. Feng, K. Li, X.Q. Yu, Promiscuous protease-
catalysed aldol reactions: A facile biocatalytic protocol for carbon-carbon bond
formation in aqeuous media, J. Biotechnol. 150 (2010) 539-545.
[27] W. Haoran, W. Zhi, Z. Hong, C. Ge, Y. Hong, W. Lei, Enzyme catalytic promiscuity:
asymmetric aldol addition catalyzed by a novel thermophilic esterase in organic
solvent, Green Chem. Lett. Rev. 7 (2014) 145-149.
[28] B.H. Xie, W. Li, H.H. Li, Z. Guan, Y.H. He, The asymmetric aldol reaction using
acidic protease from Aspergillus usamii, Tetrahedron 68 (2012) 3160-3164.
[29] Y.H. He, H.H. Li, Y.L. Chen, Y. Xue, Y. Yuan, Z. Guan, Chymopapain-catalysed
direct asymmetric aldol reaction, Adv . Synth. Catal. 354 (2012) 712-719.
15

[30] Y.L. Chen, W. Li, Y. Liu, Z. Guan, Y.H. He, Tripsin-catalysed direct asymmetric
aldol reaction, J. Mol. Catal. B:Enzym. 87 (2013) 83-87.
[31] H.H. Li, Y.H. He, Y. Yuan, Z. Guan, Nuclease p1: a new biocatalyst for direct
asymmetric aldol reaction under solvent-free conditions, Green Chem. 13 (2011)
185-189.
[32] Z.B. Xie, N. Wang, L.H. Zhou, F. Wan, T. He, Z.G. Le, X.Q. Yu, Lipase-catalysed
stereoselective cross-aldol reaction promoted by water, ChemCatChem 5 (2013)
1935-1940.
[33] Y. Zhang, N. Wang, Z.B. Xie, L.H. Zhou, X.Q. Yu, Ionic liquid as a recyclable and
efficient medium for lipase-catalysed asymmetric cross aldol reaction, J. Mol. Catal.
B:Enzym. 110 (2014) 100-110.
[34] W.G. Birolli, L.P. Fonseca, A.L.M. Porto, Aldol reactions by lipase from Rhizopus
niveus, an example of unspecific catalysis, Catal. Lett. 147 (2017) 1977-1987.
16

Figure 1. Plot of initial velocity (vo) versus p-nitrobenzaldehyde concentration ([9]) for
BSA catalysed aldol condensation of 9 and 5 M cyclohexanone (19)
Figure 2. Plots of the inverse of initial velocity (1/vo) versus the inverse of p-
nitrobenzaldehyde concentration (1/[9]) for BSA-catalysed aldol condensation of 9 and
cyclohexanone (19). The concentration of 19 was 5M (2a), 1M (2b) and 0.6 M (2c)
Scheme 1. BSA-catalysed aldol condensation of ketones 1, 8, 14 and 19 with aldehydes
2 and 9
Scheme 2. Proposed mechanism for BSA-catalysed aldol condensation of cyclohexanone
(19) and p-nitrobenzaldehyde (9)
17

Table 1. Solvent effect on BSA-catalysed reaction of butanone (1) and benzaldehyde (2)
(Scheme 1)
Ketol
5 Ketols 3+6 Enone 4 Enone
7
Butanone:water:ethano
l ratio
Entrya
conversio conversion conversion conversion
n (%)b
(%)b
0
(%)b
0
(%)b
0
1
100:0:0
98:2:0
0
3
1
2
2
1
1
3
2
1
0
1
2
1
1
1
1
1
0
0
0
0
4
3
4
4
7
2
2
5
20
11
11
11
6
7
3
96:4:0
3
3
4
94:6:0
3
3
5
92:8:0
3
3
6
90:10:0
50:50:0
25:75:0
10:90:0
5:95:0
2
2
7
2
4
1
8
5
19
4
7
9
4
1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
2
2
0
90:0:10
80:0:20
70:0:30
50:0:50
40:0:60
30:0:70
20:0:80
20:0:80c
20:0:80d
20:0:80e
20:0:80f
10:0:90
25:5:70
25:25:50
25:37:37
25:50:25
25:70:5
10:30:60
1
4
1
2
22
41
81
82
82
85
51
0
3
2
43
16
16
15
12
4
2
1
1
2
3
0
0
0
4
0
61
1
18
88
58
20
17
16
22
10
3
11
10
5
6
5
7
5
8
6
10
3
7
3
18

29
30
10:45:45
10:60:30
2
4
4
8
10
10
3
4
a Reactions were carried out at 45 ºC and 200 rpm, in the indicated mixture of solvents at the
b
reported ratios (See Section 2.2.1). Determined by HPLC at 48 h; no further conversion was
observed after this time. c At 30°C. d Without BSA. e BSA was pretreated with urea. Prior to the
assay, BSA (100 mg) was stirred in 8 M aqueous urea (2 mL) for 24 h at 100°C, then evaporated
under reduced pressure. To remove urea, the resulting residue was washed with acetone/water 1:1
(3 x 5 mL) and dried. f Porcine pancreatic lipase (PPL) was the biocatalyst.
19

Table 2. Effect of the substrates molar ratio on BSA-catalysed reaction of cyclohexanone
(19) and benzaldehyde (2) (Scheme 1)
Enone 20 Bis-enone 22
conversionb conversionb
Benzaldehyde:Cyclohexanone
Entrya
molar ratio
(%)
82
90
92
95
76
0
(%)
15
9
1
2
3
4
5
6
7
8
9
1:2
1:5
1:10
1:20
1:40
1:40c
2:1
5
3
15
0
82
49
42
18
24
31
10:1
20:1
a Reactions were carried out in ethanol at 45 ºC and 200 rpm (See Section 2.2.1).
b Determined by HPLC at 48 h; no further conversion was observed after this
time. c Without ethanol.
20

Table 3. BSA-catalysed reaction of ketones 1, 8, 14 and 19 with aldehydes 2 and 9
(Scheme 1)
Ketol
Dienone
Entrya Ketone
Aldehyde
Enone Conversion conversion conversion
(%)b
(%)b
(%)b
Cyclopentanone
1
2
3
4
5
6
7
Benzaldehyde (2)
84
0
0
17
12
20
4
(14)
3-Pentanone (8)
Cyclohexanone
(19)
87
0
0
2
2
2
82
0
15
0
Butanone (1)
85
3c
Cyclopentanone p-Nitrobenzaldehyde
35
48d
13e
0
0
18
13
21
(14)
(9)
3-Pentanone (8)
Cyclohexanone
(19)
77
0
9
88
10
9
a Reactions were carried out at 45 ºC and 200 rpm, at ketone:ethanol ratio 20:80 for 1 and 8, and
at 100 mM in ethanol for ketones 14 and 19 (See Section 2.2.1). Control experiments carried out
with BSA pretreated with urea (See footable of Table 2) gave conversions < 8%. b Determined by
c
d
HPLC at 48 h; no further conversion was observed after this time. Mixture of 3, 5 and 6.
Mixture of 15 and 16. e Mixture of 10 and 11.
21

30
20
10
0
vBSA (mM/h)
vuncat (mM/h)
0
100
200
300
400
[9] (mM)
Figure 1
22

0.4
0.3
0.2
0.1
0.0
0.6 M
1 M
5 M
0.00
0.01
0.02
0.03
0.04
0.05
1/[9] (mM-1)
Figure 2
23

O
OH
O
BSA
BSA
4
7
3
O
O
BSA
+
O
OH
O
1
2
5 (syn)
6 (anti)
O
OH O
O
O
BSA
BSA
+
R
R
R
2 (R=H)
9 (R=NO2)
10 (R=NO2, syn)
11 (R=NO2, anti)
12 (R=H)
13 (R=NO2)
8
O
O
O
OH
O
BSA
BSA
+
+
R
R
R
15 (R=NO2, syn)
16 (R=NO2, anti)
2 (R=H)
9 (R=NO2)
17 (R=H)
18 (R=NO2)
14
O
O
O
O
BSA
BSA
R
R
R
R
20 (R=H)
22 (R=H)
19
2 (R=H)
21 (R=NO2)
23 (R=NO2)
9 (R=NO2)
Scheme 1
24

Lys222
NH2
O
N
H
NH
Lys222
Lys222
H O
-
2
NH+
O–
H
O
NH
Lys222
NH
Lys222
+
Lys222
O–
H
H
NO2
NO2
NO2
NH+
O–
O
Lys222
OH
H2O
Lys222 NH2
+
H
H
NO2
NO2
O
O
H
HO–
H
H3N+ Arg280
H
+
H2N Arg280
HO
NO2
NO2
H2O
H2N Arg280
+
Scheme 2
25