Chiral β-lactam-based integrin ligands through Lipase-catalysed kinetic resolution and their enantioselective receptor response

Article abstract of DOI:10.1016/j.bioorg.2019.102975

Obtainment and testing of pure enantiomers are of great importance for bioactive compounds, because of the assessed implications of enantioselectivity in receptor-mediated responses. Herein we evaluated the use of biocatalysis to obtain enantiomerically pure β-lactam intermediates further exploited in the synthesis of novel integrin ligands as single enantiomers. From a preliminary screening on a set of commercially available hydrolases, Burkholderia Cepacia Lipase (BCL)emerged as a suitable and highly performing enzyme for the kinetic resolution of a racemic azetidinone, key intermediate for the synthesis of novel agonists of integrins. Upon optimization of the biocatalytic protocol in terms of enzymes, acylating agents and procedures, the two β-lactam enantiomers were obtained in excellent enantiomeric excesses (94% and 98% ee). Synthetic elaborations on the separated enantiomers allowed the synthesis of four chiral β-lactams which were evaluated in cell adhesion assays on Jurkat cell line expressing α₄β₁ integrin, and K562 cell line expressing α₅β₁ integrin. Biological tests revealed that only (S)-enantiomers maintained the agonist activity of racemates with a nanomolar potency, and a specific enantio-recognition by integrin receptors was demonstrated.

Full text of DOI:10.1016/j.bioorg.2019.102975

Accepted Manuscript
Chiral β-lactam-based integrin ligands through Lipase-catalysed kinetic reso-
lution and their enantioselective receptor response
Giulia Martelli, Paola Galletti, Monica Baiula, Luca Calcinari, Giacomo Boschi,
Daria Giacomini
PII:
S0045-2068(19)30115-4
https://doi.org/10.1016/j.bioorg.2019.102975
102975
DOI:
Article Number:
Reference:
YBIOO 102975
Bioorganic Chemistry
To appear in:
Received Date:
Revised Date:
Accepted Date:
22 January 2019
3 May 2019
4 May 2019
Please cite this article as: G. Martelli, P. Galletti, M. Baiula, L. Calcinari, G. Boschi, D. Giacomini, Chiral β-lactam-
based integrin ligands through Lipase-catalysed kinetic resolution and their enantioselective receptor response,
Bioorganic Chemistry (2019), doi: https://doi.org/10.1016/j.bioorg.2019.102975
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. 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.

Chiral -lactam-based integrin ligands through Lipase-catalysed kinetic resolution and their
enantioselective receptor response.
Giulia Martellia, Paola Gallettia, Monica Baiulab, Luca Calcinaria, Giacomo Boschia, Daria
Giacominia,*,daria.giacomini@unibo.it
aDepartment of Chemistry "G. Ciamician", University of Bologna, Via Selmi 2, 40126 Bologna, Italy.
bDepartment of Pharmacy and Biotechnology, University of Bologna, Via Irnerio, 48, 40126, Bologna, Italy.
*Corresponding author.
Graphical abstract
Highlights
Kinetic resolution of azetidinones by Lipases
Burkholderia cepacia lipase promoted the resolution in transesterification
Synthesis of enantiopure -lactams targeting integrins α4β1 and α5β1
Enantiospecific response of (S)-azetidinones in promoting cell adhesion
Abstract
Obtainment and testing of pure enantiomers are of great importance for bioactive compounds, because of
the assessed implications of enantioselectivity in receptor-mediated responses. Herein we evaluated the
use of biocatalysis to obtain enantiomerically pure -lactam intermediates further exploited in the
synthesis of novel integrin ligands as single enantiomers. From a preliminary screening on a set of
commercially available hydrolases, Burkholderia Cepacia Lipase (BCL) emerged as a suitable and highly
performing enzyme for the kinetic resolution of a racemic azetidinone, key intermediate for the synthesis
of novel agonists of integrins. Upon optimization of the biocatalytic protocol in terms of enzymes, acylating
agents and procedures, the two -lactam enantiomers were obtained in excellent enantiomeric excesses
(94% and 98% ee). Synthetic elaborations on the separated enantiomers allowed the synthesis of four
chiral -lactams which were evaluated in cell adhesion assays on Jurkat cell line expressing α4β1 integrin,
1

and K562 cell line expressing α5β1 integrin. Biological tests revealed that only (S)-enantiomers maintained
the agonist activity of racemates with a nanomolar potency, and a specific enantio-recognition by integrin
receptors was demonstrated.
Keywords: Biocatalysis, Kinetic resolution, Integrins, Cell adhesion, lactams, Azetidinones, Agonists
1. Introduction
Since the early discovery by A. Piutti of the enantioselectivity in receptor-mediated biological events [1, 2],
the demand for chiral molecules as active pharmaceutical ingredients (APIs) has continuously increased by
the pharmaceutical industry because enantiomers could have dramatically different biological activities
such as pharmacokinetic, pharmacology or toxicology [3, 4]. Two enantiomers of a molecule could in fact
display different spatial distributions on the same receptor or even bind two diverse receptors being
responsible of other biological pathways. Thus, it is of paramount importance to know the biological
response of single stereoisomers and to separately obtain two enantiomers of a new bioactive molecule. To
meet this challenge, a variety of technologies could be applied to the synthesis of chiral APIs, with
biocatalysis playing a prominent role [5, 6].
Biocatalysis has nowadays a widespread application in the synthesis of chiral compounds and offers
multiple advantages such as the reduction of side-products and synthetic steps because the milder reaction
conditions do not require protection/deprotection of sensible functional groups [7, 8]. Enzymes are highly
selective catalysts in terms of chemo-, regio-, and stereo-selectivity, and highly specific toward substrates
and reactions [9]. Biocatalysis indeed has many attractive features in the context of green chemistry and
sustainable development: enzymes derive from renewable resources and are biocompatible,
biodegradable, and inherently non-hazardous [10].
Hydrolases are the most studied enzymes and largely used in organic chemistry because of their stability
and their no need of cofactors [11, 12]. In addition, these enzymes are commercially available and
exploitable not only for hydrolysis or alcoholysis, but also for the reverse condensation reaction. In
particular, lipases (EC 3.1.1.3) have gained attention due to their robustness, wide tolerance to different
substrates, stability to organic solvents and the possibility to work at high temperatures [13, 14]. Moreover,
lipases are inexpensive and available from multiple sources, mainly of microbial origin [15]. Lipases catalyse
various processes such as esterification, transesterification, and hydrolysis and are particularly applied to
the synthesis of enantiopure compounds via kinetic resolutions (KR) [16-18]. Notwithstanding the intrinsic
limitation of a maximum theoretical yield of 50% for a single enantiomer, KR processes have special
relevance in the development of new bioactive compounds, allowing access to both enantiomers
separately so that their specific activities could be singularly acquired.
Recently, we reported the synthesis of a library of new -lactam compounds (azetidinones) that were
specifically designed by a structure-based strategy to target specific classes of integrin receptors [19, 20].
The biological activities of the new azetidinone derivatives were deeply evaluated by investigating their
2

effects on isolated receptors, integrin-mediated cell adhesion, and cell signalling. Among the new
derivatives, some ligands were very active and showed interesting affinities toward specific integrin
families. Some molecules were synthesized as pure enantiomers, other as racemic mixtures, and among the
latter emerged some very interesting ligands with nanomolar range activities (Fig. 1). Integrins are
transmembrane receptors, composed of α and β subunits that bind to specific ligands mediating cell-cell
and cell-extracellular matrix interactions. These abilities allow integrins to regulate crucial cellular
functions, such as adhesion, migration, growth and survival [21]. Their physiological importance is
underlined by several types of pathologies in which integrins play a pivotal role; therefore, integrins are
attractive pharmacological targets for the development of new drugs [22, 23].
The aim of the present work is to establish an enzymatic route to obtain single enantiomers of the most
active integrin ligands previously developed [19]. Since the -lactam intermediate 1 was a common building
block for the synthesis of all the racemic derivatives, we studied a biocatalytic strategy for obtaining both
its enantiomers. Subsequent synthetic elaborations on the two enantiomers of 1 provided 4 enantiopure
integrin ligands, whose biological effects were then evaluated in cell adhesion assays on Jurkat cell line
expressing α4β1 integrin, and on K562 cell line expressing α5β1 integrin, and compared to the racemic
analogues (Scheme 1).
2. Results and Discussion
Initial studies were focused on performing KR directly on -lactam 1 as ethyl, benzyl or tbutyl esters (1a-c).
A small panel of lipases, esterases and proteases, such as Candida Antarctica lipase A (CAL-A), Candida
Antarctica lipase B (CAL-B), Pig liver esterase (PLE) and α-chymotrypsin from bovine pancreas were
screened (Table 1). Racemic esters 1a-c were prepared from the commercially available 4-acetoxy-
azetidinone with a Reformatsky reagent as previously reported [19, 20].
The enzymatic KR on the three esters 1a-c was not satisfactory. With CAL A, CAL B, or PLE the process
mainly was not chemo-selective, since these enzymes hydrolysed the amide or the ester functions or both,
giving mixtures of by-products (entries 1, 2, 5-6, 10, and 11), with a final poor enantiomeric ratio of the
unreacted esters. In selected cases the reaction turned to be chemoselective but with no or poor
enantioselection (entries 3-4, 7, and 9, respectively) or very low conversion (entry 12). Only α-
chymotrypsine provided reasonable results with benzyl ester 1b, if considering both conversion and
enantiomeric ratio (entry 8).
Detection and isolation of the acid 2 from crude mixtures was a difficult issue in all cases because of its no
UV absorption above 200 nm and a high solubility in water. Furthermore, when isolated from the aqueous
phase through lyophilization, esterification of 2 to afford the ester 1b (a necessary step to complete the
synthesis of the final integrin ligands) failed.
We then changed the synthetic strategy and turned our attention to an alternative substrate to get an
enzymatic KR based on a transesterification process instead of a hydrolysis. Kaman et al. reported the KR of
some -lactam intermediates with an oxymethylene residue on the -lactam nitrogen atom [24].
3

Following this strategy, the racemic compound 1b was derivatized with an oxymethylene group using
paraformaldehyde and catalytic potassium carbonate under sonication (scheme 2) [25]. A screening of
some reaction parameters (amounts of paraformaldehyde and K2CO3) led to optimized conditions and
racemic -lactam 5 was obtained in quantitative isolated yields with 6% K2CO3 in a mixture of THF/H2O =
11:1 (Scheme 2).
The first screening for the KR of the racemic compound 5 was carried out in tbutylmethylether (TBME) as
solvent with vinyl acetate as acyl donor at room temperature. In general, the use of vinyl esters as acyl
donors could prevent the formation of competitive reactions, since the co-product vinyl alcohol
tautomerizes to acetaldehyde with a shift toward formation of the products, hence avoiding hydrolysis or
nucleophilic attack on the newly formed ester. A panel of commercially available lipases was screened in
order to find the best enzyme in terms of activity and stereoselectivity in the formation of the acylated
product 6 and in the enantio-enriched substrate 7 (Table 2). The variability in the enzyme amount used in
this screening derives from different commercial enzymes formulations. Some of the tested lipases (lipase
from porcine pancreas - crude type II, Rhizopus niveus, Candida lipolytica and lipase from wheat germ)
appeared totally inactive with no conversion after 4 days and were not included in the table. The same
negative response was observed with α-chymotrypsin and pig liver esterase (crude extract) that were
tentatively tested. Lipase from Candida Antarctica A showed a higher conversion compared to other
enzymes, but no selectivity (Table 2, entry 1). Other lipases (Table 2, entries 2-9) instead displayed low
conversions (6-19%) but in some cases interesting enantiomeric excesses: with porcine kidney acylase and
Mucor miehei lipase the ee% of product 6 was 84 and 87, respectively (Table 2, entries 6, 7). Lipase from
Aspergillus oryzae displayed a 57% conversion in 4 days (entry 11), but at the same time Burkholderia
cepacia lipase (BCL)(entry 12) afforded higher enantiomeric excess values for the enantioenriched
substrate 7. Lipase from Pseudomonas fluorescens (entry 13) displayed a higher conversion in shorter times
than BCL (24 h vs 4 days), demonstrating a high affinity and a good stereospecificity for the substrate 7 that
was obtained with a 91 ee%.
From data in Table 2, BCL and Pseudomonas fluorescens emerged as the most promising enzymes for the
KR of the racemic in terms of both conversion and ees 5 and were thus selected for further evaluations.
Specifically, a study was carried out for determining the evolution of the enantiomeric excess of substrate
and product vs conversion; accordingly, the biocatalytic reactions were set up with BCL or Pseudomonas
fluorescens lipase in TBME and monitored over the time. Vinyl acetate and vinyl butyrate were examined as
acyl donors in this study, and conversion plots are reported in Figure 2.
BCL showed the best results, allowing the obtainment of the enriched substrate 7 in good ee% at a 60%
conversion, whereas the ee% of the ester products was good only using vinyl butyrate to give enantio-
enriched butyrate 8 at around 40% conversion. Conversely, Pseudomonas Fluorescens led to worse results
than BCL: vinyl acetate provided a lower selectivity compared to vinyl butyrate, and enantiomeric excesses
in the 40-60% conversion range were not satisfying for both substrate and products. Encouraged by the
good results achieved with BCL, we further investigated the influence of anhydrides as acyl donors versus
the corresponding vinyl esters on the enantioselectivity of the KR with BCL (Table 3). Anhydrides are not so
commonly used in kinetic resolutions, but some literature papers reported the use of these agents as acyl
donors in the resolution of primary and secondary racemic alcohols [26].
4

All reactions were conducted in thermostat at 37 °C and stopped after a fixed time of 5.5 hours by filtration
of the enzyme and analysed by chiral HPLC. The temperature enhancement from 25 °C (rt) to 37 °C with
vinyl acetate allowed a 64% conversion in 5.5 hours (Table 3, entry 1), while a lower value (50%) was found
in 4 days at room temperature (see Table 2, entry 12). Moreover, vinyl acetate provided a higher reaction
rate among the tested acyl donors (Table 3, entry 1 vs 2-4) whereas acetic anhydride displayed too low
conversions for being further considered (Table 3, entry 3). Butyric anhydride showed very good results in
terms of both selectivity and conversion (Table 3, entry 4); in this case a classic kinetic resolution could be
performed: stopping the reaction at 50% conversion allowed to obtain good enantiomeric excesses for
both substrate and product, together with the maximum yield (around 50%). Unfortunately, the use of
butyric anhydride as acylating agent released an equivalent of butyric acid which was difficult to separate
from the enantiomerically enriched substrate 7 and product 8. For this reason, considering both reaction
outcome and product isolation, the results obtained with vinyl butyrate were considered the most
promising to allow a scale-up process.
A preparative KR was therefore carried out with vinyl butyrate and BCL in order to isolate a larger amount
of both the enantio-enriched substrate 7 and the ester 8 with the highest ee values (Scheme 3) [27]. A two-
steps enzymatic resolution process [9] was adopted: in the first step the enzymatic transesterification with
BCL and vinyl butyrate in TBME was stopped at a 38% conversion, the enzyme was filtered off, and the
enantio-enriched substrate 7 was separated from product 8 by flash-chromatography obtaining compound
8 with 90% ee and 33% of isolated yield.
The enriched substrate 7 obtained in the first step was then re-processed in a second step under the same
conditions (BCL, vinyl butyrate in TBME) up to 67% conversion; after the work-up, the reaction mixture was
finally purified to isolate substrate 7 with a good enantiomeric excess (94%) and 35% of isolated yields. By
the two-steps procedure good amounts of 7 and 8 were thus obtained and allowed determination of the
optical rotatory powers (Scheme 3).
The compounds (+)7 and (-)8 were then further elaborated to obtain the two enantiomers (+)9 and (-)9,
intermediates required for the synthesis of the final chiral integrin ligands. For this purpose, it was
necessary to eliminate the oxymethylene residue on the -lactam nitrogen atom. Some oxidative
conditions were tested (NaClO/NaClO2, TEMPO/NaIO4,) [28, 29], but only KMnO4 [30] was successful:
reaction on (+)7 was effective with a complete conversion and good yields affording (+)9 in 16 hours
without any purification needed (Scheme 4).
Transformation of (-)8 required the hydrolysis of the butanoate ester, but the use of some chemical
procedures (K2CO3 in methanol, TEA under MW irradiation, trifluoroacetic acid in CH2Cl2) to selectively
hydrolyse the butanoate ester over the benzylic one, failed, with detection of partial hydrolysis of the
benzylic ester or formation of by-products. It was then decided to exploit again the selectivity of the
biocatalysis with BCL under hydrolytic conditions (Scheme 4).
5

The process was carried out in MilliQ water at 37 °C with a minimal amount of acetonitrile to dissolve (-)8
(H2O/CH3CN=11:1). The progress of the reaction was again monitored through chiral HPLC. As expected, the
system showed a perfect selectivity toward the aliphatic ester vs the benzylic one, and alcohol (-)7 was
obtained from ester (-)8 in only one hour and good isolated yields. Remarkably, a further increase in ee%
was obtained: from 90% of ester (-)8 to a 98% ee of alcohol (-)7, due to the enantiomeric selection by the
enzyme. Treatment of (-)7 with KMnO4 in CH3CN afforded, also in this case, the chiral -lactam (-)9 in good
yields. The enantiomeric excesses of -lactams (+)9 and (-)9 were completely preserved, as detected by
chiral HPLC analysis after optimization of the analytical conditions (see Supplementary Material, Figure S1).
It was then necessary to assign the absolute configuration to the separated enantiomers (+)9 and (-)9, and
this was achieved through chemical correlation with a known reference compound(S)-2-(4-oxoazetidin-2-
yl)acetic acid, already reported in the literature obtained starting from (S)-aspartic acid (scheme 5) [31].
The two enantiomers (+)9 and (-)9 were thus submitted to hydrogenolysis to obtain the acids (-)10 and
(+)10. When comparing the specific rotatory powers of (-)10 and (+)10 with that of the reference
compound, it was possible to assign the absolute configurations of our acids as (-)(R)10 and (+)(S)-10 and,
consequently, the configurations of the benzylic ester precursors as (+)(S)-9 and (-)(R)-9. The enantio-
preference of BCL for the (R)-configuration was thus confirmed and in line with what reported in the
literature [30].
The enantiopure compounds (+)(S)-9 and (-)(R)-9 were then separately subjected to those synthetic steps
previously optimized on racemic precursors rac-A and rac-B and necessary for obtaining the chiral integrin
ligands [19, 20]. In scheme 6 the synthetic sequence for the synthesis of four enantiomerically pure integrin
ligands (S)-11, (R)-11, (S)-12, and (R)-12 is reported. The synthesis of (S)-12, and (R)-12 comprised four
steps: acylation of the -lactam nitrogen atom with o-tolylisocyanate, hydrogenolysis of the benzyl ester
obtaining (S)-11, and (R)-11, peptidic coupling with glycine benzylester, and final hydrogenolysis to get the
terminal carboxylic acid function. The optical purity was monitored all over the intermediates through
chiral HPLC analysis and it was consistently maintained up to the final products (S)-11, (R)-11, (S)-12, and
(R)-12.
3. Pharmacological evaluation
The four enantiomers (S)-11, (R)-11, (S)-12, and (R)-12 were finally tested in integrin-mediated cell
adhesion assays. Jurkat cell lines expressing leukocyte integrin 41 or K562 expressing 51 were adopted
for the biological evaluations, since the corresponding racemic derivatives rac-A and rac-B were selective
agonists of 41 or 51, respectively [19, 20]. Interestingly, only -lactams (S)-11 and (S)-12 showed a
concentration-dependent enhancement of cell adhesion in fibronectin-mediated tests, whilst (R)-11 and
(R)-12 were completely inactive (Figure 3). In details, compound (S)-11 showed an EC50 of 21 nM towards
41 integrin, maintaining the same potency as previously found for rac-A (12.9 nM vs 41 integrin, Figure
1); moreover, compound (S)-12 displayed a comparable potency towards 51 integrin (EC50 = 35 nM vs
51 integrin, Figure 3) in accordance with that obtained for rac-B (9.9 nM vs 51 integrin, Figure 1). The
6

potency of the racemic compounds was thus not impaired by the presence of inactive (R)-enantiomers. The
apparent decrease in EC50 numerical values for the (S)-enantiomers could be due to a not complete optical
purity (94% ee). Noteworthy, the (S)-enantiomers maintained the agonist activity of the racemates, and a
specific enantio-recognition by integrin receptors was then demonstrated.
A comparison between the two active enantiomers (S)-11 and (S)-12 and some β-lactam derivatives
previously reported [19], leads to the observation that they are among the most potent agonists for 41 or
51 integrin, respectively, and that the integrin activity vs the (S) configuration at the C-4 of the
azetidinone is consistent with that observed for chiral (S)--lactam integrin ligands obtained from L-aspartic
acid [19].
3. Conclusion
Biocatalysis has become nowadays a standard tool to produce fine chemicals and its application in the
pharmaceutical field heavily dominates. In fact, the use of enzymes in APIs synthesis is becoming an
alternative way to obtain enantiomerically pure products leading to many advantages such as mild work
conditions, reduction of extrasteps and less or no toxic wastes.
Having in hands a series of new -lactam compounds being active as integrin ligands [19], most of which
racemic at the C4 position of the ring, it was necessary to determine how two enantiomers could
differently interact with the receptor in terms of activity as agonists or antagonists, and potency. In this
work we evaluated the use of enzyme catalysis to obtain the two enantiomers of a -lactam intermediate
7, precursor of novel integrin ligands. Among a series of Lipases, Burkholderia Cepacia Lipase (BCL) was
selected as the most suitable enzyme for the kinetic resolution of the intermediate 7. After optimization of
the biocatalytic protocol (enzymes and acylating agents) and of the synthetic procedures, the two
enantiomers (+)S-9 and (-)R-9 were obtained in excellent enantiomeric excesses. Assignment of the
absolute configurations was assessed by chemical transformation into a known compound. Synthetic
elaboration of (+)S-9 and (-)R-9 allowed the obtainment of four chiral -lactams (S)-11, (R)-11, (S)-12, and
(R)-12 which were finally evaluated in cell adhesion assays. Pharmacological tests on Jurkat cell line
expressing α4β1 integrin and K562 cell line expressing α5β1 integrin, revealed that the activity as agonists
with a nanomolar potency was fully maintained only in case of (S)-enantiomers, while (R)-derivatives were
completely inactive, thus revealing an important stereochemical requirement for further developments of
the new -lactam-based integrin ligands.
4. Experimental
4.1 Material and Methods
Anhydrous solvents and reagents were obtained commercially and used as received. Deionized water was
obtained from a Millipore analytical deionization system (MilliQ). For TLC monitoring Merck 60 F254 plates
were used and for liquid chromatography Merck silica gel 200-300 mesh was used. 1H and 13C NMR spectra
were recorded with an INOVA 400 instrument with a 5 mm probe. All chemical shifts are quoted relative to
deuterated solvent signals (δ in ppm and J in Hz). FTIR spectra were recorded with Alpha FT IR Bruker
spectrometer. Polarimetric analyses were conducted on Unipol L 1000 Polarimeter at 598 nm. The purities
7

of the target compounds were assessed as being >95% using HPLC-MS. HPLC-MS: Agilent Technologies
HP1100 instrument, equipped with a ZOBRAX-Eclipse XDB-C8 Agilent Technologies column; mobile phase:
H2O/CH3CN, 0.4 mL/min, gradient from 30 to 80% of CH3CN in 8 min, 80% of CH3CN until 25 min, coupled
with an Agilent Technologies MSD1100 single-quadrupole mass spectrometer, full scan mode from m/z =
50 to 2600, in positive ion mode, ESI spray voltage 4500 V, nitrogen gas 35psi, drying gas flow 11.5mL/min,
fragmentor voltage 20 V. Chiral-HPLC: Agilent Technologies 1200 instrument equipped with a diode array
UV detector on Daicel Chiralcel columns (0.46 cm I.D. x 25 cm) with HPLC grade isopropanol and n-hexane
as eluting solvents.
The employed enzymes were: Burkholderia Cepacia lipase (BCL) powder, ≥30 U/mg by Sigma Aldrich;
Candida antartica lipase A (CAL-A), CLEA (Cross Linked Enzyme Aggregate), ≥10 U/mg, by Fluka (Sigma
Aldrich); Lipase crude from porcine pancreas, 187 U/mg, by Sigma Aldrich; Acylase from porcine kidney,
powder, 52 U/mg, by Serva Biochemical; Alpha-chymotripsine from bovine pancreas, powder, 42 U/mg, by
Serva Biochemical; Esterase crude from porcine liver, powder, 19 U/mg, by Sigma Aldrich; Lipase from
Aspergillus niger, powder, 0.94 U/mg, by Fluka; Lipase from hog pancreas, powder, 27 U/mg, by Fluka;
Lipase from Pseudomonas fluorescens, powder, 36 U/mg, by Fluka; Lipase from Rhizopus niveus, powder,
1.7 U/mg, by Fluka; Lipase from Rhizopus arrhizus, powder, 2.2 U/mg, by Fluka; Lipase from Mucor miehei,
powder, 1.4 U/mg, by Fluka; Lipase from Candida cylindracea, powder, 2.8 U/mg, by Fluka; Lipase from
Aspergillus oryzae, powder, 48 U/mg, by Fluka; Lipase from Penicillium roqueforti, powder, 0.65 U/mg, by
Fluka; Lipase from wheat germ, powder, 0.10 U/mg, by Fluka; Lipoprotein Lipase from Chromobacterium
viscosum, powder, 2962 U/mg, by Fluka; Lipase from Mucor javanicus, powder, 9.9 U/mg, by Fluka; Lipase
from Candida lipolytica, powder, 1.1 U/mg, by Fluka; Lipase From Candida antartica B, immobilized on
acrylic resin, 10 U/mg, by Sigma Aldrich.
4.2 Synthesis of racemic -lactams
Compounds 1a-c were synthesized as previously reported [19] and their spectroscopic data are in
accordance with those reported. Preparation of racemic esters (±)-6 and (±)-8 for the optimization of an
analytic method by chiral HPLC was reported in the Supplementary Material.
Benzyl 2-(1-(hydroxymethyl)-4-oxoazetidin-2-yl)acetate (5)
To a solution of compound 1b (55 mg, 0.25 mmol, 1 equiv) in THF (0.55 mL), paraformaldehyde (11 mg,
0.38 mmol, 1.5 equiv), K2CO3 (2 mg, 0.01 mmol, 0.3 equiv) and water (22 μL) were added. The system was
sonicated at room temperature. At completion (4 h, TLC monitoring), the reaction mixture was diluted with
EtOAc (2.5 mL), dried on anhydrous Na2SO4, filtered and concentrated. The crude was purified by flash-
chromatography (cyclohexane/EtOAc 35:65) yielding compound 5 as a colorless oil (62 mg, 99%).
1
IR (film, cm-1) 3406, 3033, 2953, 1736, 1397, 1175; H NMR (400 MHz, CDCl3) δ (ppm) 7.41 – 7.28 (m, 5H),
5.15 (s, 2H), 4.64 (d, JAB = 11.6 Hz, 1H), 4.60 (d, JAB = 11.6 Hz, 1H), 4.16 – 4.04 (m, 1H), 3.15 – 3.08 (dd, J =
5.3, 15.0 Hz 1H), 2.86 – 2.71 (m, 2H), 2.70 – 2.64 (dd, J = 2.4, 15.0 Hz, 1H); 13C NMR (100 MHz, CDCl3) δ
(ppm) 171.1, 166.6, 135.2, 128.6, 128.5, 128.3, 66.9, 64.1, 47.0, 42.6, 38.7; ESI-MS (Rt = 4.4 min) m/z 232
[M-OH]+.
4.3 Synthesis of enantio-enriched -lactam compounds
4.3.1. Enzymatic KR of esters 1a-c (data Table 1):
In a glass vial with a screw cap, to a solution of ester 1a-c (0.1 mmol) in 4 mL of phosphate buffer 0.1 M pH
7.5, the desired enzyme was added (in the amount specified in Table 1). The mixture was stirred at room
8

temperature under orbital shaking (700 rpm). The reaction was monitored at set time intervals with chiral
HPLC (reaction times specified in Table 1) for evaluate the substrate conversion and the enantiomeric
excess of unreacted esters 1a-c. At the end of the reaction, the residual solution was extracted with CH2Cl2
(3 × 5 mL), dried on anhydrous Na2SO4, filtered and concentrated. The aqueous phase was lyophilized and
the ratio of compounds 2/3/4 was analysed via 1H NMR.
4.3.2. Enzymatic KR of compound 5
Method A: in a glass vial with a screw cap, to a solution of alcohol 5 (20 mg, 0.08 mmol, 1 equiv), acyl donor
(6 equiv) and solvent (2 mL), the desired enzyme was added (in the amount specified in Tables 2 and 3).
The mixture was stirred at room temperature under orbital shaking (700 rpm) – as specified in Table 2 - or
at 37 °C under magnetic stirring – as specified in Table 3. The reaction was monitored at set time intervals
with chiral HPLC (reaction times specified in Tables 2 and 3) for evaluate the substrate conversion and the
enantiomeric excesses of unreacted alcohol 7 and ester products 6 or 8.
Method B: With the best conditions (TBME as solvent, BCL as enzyme, vinyl butyrate as acyl donor), the
kinetic resolution was scaled-up: alcohol 5 (150 mg, 0.6 mmol, 1 equiv), vinyl butyrate (457 µL, 3.6 mmol, 6
equiv), TBME (15 mL) and BCL (150 mg) were reacted at room temperature in a 30 mL vial with a screw cap.
At 38% conversion (chiral HPLC monitoring), solvent was evaporated and the crude was directly purified by
flash-chromatography (cyclohexane/EtOAc 40:60) to separate target ester (-)8 as a colorless oil (60 mg,
33%, ee = 90%, [α]D20 = – 15 (c = 1.0, CH2Cl2)) and residual alcohol (+)7, which was re-processed in the same
reaction conditions. At 67% conversion, work-up and purification were carried out as described above, and
alcohol (+)7 was isolated as a colorless oil (51 mg, 35%, ee = 94%, [α]D20 = + 13 (c = 1.0, CH2Cl2)).
Spectroscopic data of (+)7 and (-)8 were in fully accordance with those reported for their corresponding
racemic analogues 5 and (±)8 (Supplementary Material).
4.3.3. Enzymatic hydrolysis of compound (-)8
In a 10 mL vial with a screw cap, to a solution of ester (-)8 (200 mg, 0.8 mmol, 1 equiv) CH3CN and MilliQ
H2O (1:11 ratio, total volume 20 mL), Burkolderia Cepacia lipase was added (200 mg). The mixture was
stirred at 37 °C in thermostat. At completion (1 h, chiral HPLC monitoring), the aqueous mixture was
extracted with EtOAc (3 × 10 mL). The collected organic phases were dried over Na2SO4, filtered and
concentrated under vacuum. The crude was purified by flash-chromatography (cyclohexane/EtOAc 40:60)
20
to yield target alcohol (-)7 as a colorless oil (75%, ee = 98%, [α]D = – 16 (c = 1.0, CH2Cl2)). Spectroscopic
data of (-)7 were in fully accordance with those reported for its corresponding racemic analogue 5.
4.3.4. Oxidation with KMnO4 of the enantiomers (+)7 or (-)7
To a solution of the alcoholic compound (+)7 or (-)7 (1 equiv) in CH3CN (30 mL/mmol), KMnO4 (6 equiv) was
added portionwise at 0°C. The system was then warmed to room temperature and left under stirring
overnight. At completion (16 h, TLC monitoring), the reaction mixture was quenched with a saturated
solution of Na2S2O5 until complete decoloring. The mixture was then filtered and the acetonitrile
evaporated under reduced pressure. The residual aqueous solution was then extracted with EtOAc (3 × 10
mL), dried on anhydrous Na2SO4, filtered and concentrated. The desired product (+)9 or (-)9 was obtained
without further purification as a white solid.
(+)9: 91%, ee = 94%, [α]D20 = + 26 (c = 1.0, CH2Cl2)
(-)9: 90%, ee = 98%, [α]D20 = – 26 (c = 1.0, CH2Cl2).
Spectroscopic data of (+)9 and (-)9 were in fully accordance with those reported for their corresponding
racemic analogues [19].
9

4.3.5. Synthesis of compounds (S)10, (R)10, (S)11, (R)11, (S)12, and (R)12
Compounds (S)10, (R)10, (S)11, (R)11, (S)12, and (R)12 were synthesized according to the previously
reported procedures [19]; yields and spectroscopic data were in fully accordance with those reported for
their corresponding racemic analogues [19].
4.4. Analytical methods
The enantiomeric composition and conversions of azetidinones were determined by chiral HPLC analysis
and established at 210 nm. To evaluate the conversion, a chromatographic analysis was conducted also at
time 0, i.e. before inserting the enzyme into the reaction; areas of the substrate chromatographic peaks at
time 0 were then compared with those, progressively decreased, found over the time.
Samples for HPLC injection were prepared as follows:
Transesterification reactions:
-
in case of explorative reactions: 50 μL of the reaction mixture were sampled and the solvent was
evaporated under air flux. Then, the mixture was suspended in 350 μL of a solution of n-
hexane/isopropanol 1:1, filtered through regenerated cellulose syringe filters (diameter = 25 mm,
pore diameter = 0.2 μm) and injected in chiral HPLC.
-
in case of preparative reactions: 0.5 mL of the reaction mixture was filtered through regenerated
cellulose syringe filters (diameter = 25 mm, pore diameter = 0.2 μm); the organic solvent was dried,
re-suspended in a solution of n-hexane/isopropanol 1:1 and directly injected in chiral HPLC.
Hydrolysis reactions:
0.5 mL of the reaction mixture was extracted with EtOAc (0.5 mL); the organic solvent was dried,
re-suspended in a solution of n-hexane/isopropanol 1:1 and directly injected in chiral HPLC
-
In case of kinetic resolution performed on alcohol 5, we determined the best analytical conditions in order
to verify in a single HPLC analysis both the reaction conversion and the enantiomeric excesses of the
residual substrate 7 and the formed product 6 or 8 (see Supplementary Material, Figure S2).
Elution conditions:
Compounds 1a-c: Column: Chiralpak IA, solvent: n-hexane/isopropanol 80:20, flow: 0.5 mL/min,
temperature: 40 °C.
Compounds 5-8: Column: Chiralpak IC, solvent: n-hexane/isopropanol 50:50, flow: 0.5 mL/min from 0 to 18
min, 1 mL/min from 19 min, temperature: 40 °C.
Compound 9: Column: Chiralpak IC, solvent: n-hexane/isopropanol 60:40, flow: 0.5 mL/min, temperature:
40 °C.
4.5. Cell culture
Jurkat E6.1 and K562 cell lines were routinely grown in RPMI-1640 (Thermofisher) and glutamine with 10%
FBS. Cells were kept at 37°C under 5% CO2 humidified atmosphere; 40 h prior to experiments, K562 cells
were treated with 25 ng/mL PMA (Phorbol 12-myristate 13-acetate, Sigma Aldrich SRL, Milan, Italy) in order
to prompt differentiation and to augment cell surface α5β1 integrin expression. These two cell lines were
employed as a suitable model to study integrins as Jurkat cells express α4β1 integrin, while K562 cells
express α5β1 integrin. Both cell lines were obtained from American Type Culture Collection (ATCC, Rockville,
MD, USA).
10

4.6. Cell adhesion assays
The assays were performed as previously described in ref. 19, 20, 32, and 33. Briefly, for adhesion assay on
K562 cells 96-well plates (Corning Costar, Celbio, Milan, Italy) were coated with fibronectin (FN, 10 µg/mL)
overnight at 4 °C. K562 cells were pre-incubated with several concentrations of each compound or with the
vehicle (methanol) for 30 min at room temperature, plated (50,000 cells/well) on fibronectin coated wells
and incubated at room temperature for 1 h. After three washes with 1% BSA (bovine serum albumin) in PBS
(phosphate-buffered saline) to remove non-adherent cells, 50 μL hexosaminidase substrate were added to
each well. Afterwards 100 μL of stopping solution were used to stop the enzymatic reaction and
absorbance was measured at 405 nm in a Victor2 Multilabel Counter (PerkinElmer, Waltham, MA, USA). For
adhesion assays on Jurkat E6.1 cells, black 96-well plates were coated overnight at 4 °C with VCAM-1 (5
μg/mL). Cells were stained with CellTracker green CMFDA (12.5 μM, 30 min at 37 °C, ThermoFisher). Jurkat
E6.1 cells were pre-incubated with various concentrations of each compound or with the vehicle
(methanol) for 30 min at 37°C and then plated (500,000/well) on VCAM-1-coated wells for 30 min at 37 °C.
After three washes, Jurkat E6.1 cells were lysed with 0.5% Triton X-100 in PBS (30 min at 4 °C) and
fluorescence was measured (Ex485 nm/Em535 nm).
Experiments were repeated at least three times and carried out in quadruplicate. Data analysis and EC50
values were calculated using GraphPad Prism 5.0 (GraphPad Software, San Diego, CA, USA).
Acknowledgements
D.G. would like to thank Prof. K. Faber for fruitful discussions, and C. Vincenzi, and T. Flanigan for technical
assistance.
Funding
This research was supported by University of Bologna and Ministero dell'Università e della Ricerca (PRIN
2015 project 20157WW5EH).
References
[1] A. Piutti, Una nuova specie di asparagine, L’Orosi–Giornale di Chimica, Farmacia e Scienze Affini 9
(1886) 198–202.
[2] J. Gal, The Discovery of Stereoselectivity at Biological Receptors: Arnaldo Piutti and the Taste of the
Asparagine Enantiomers—History and Analysis on the 125th Anniversary, Chirality 24 (2012) 959-
976, https://doi.org/10.1002/chir.22071
[3] H.Y. Aboul-Enein, I.W. Wainer, The Impact of Stereochemistry on Drug Development and Use, 1997
Wiley-Interscience
[4] S. W. Smith, Chiral toxicology: it’s the same thing. . .only different, Toxicol. Sci. 110 (2009) 4-30,
https://doi.org/10.1093/toxsci/kfp097
[5] R.N. Patel, Biocatalysis for synthesis of pharmaceuticals, Bioorg. Med. Chem. 26 (2018) 1252-1274,
https://doi.org/10.1016/j.bmc.2017.05.023
11

[6] M.D. Truppo, Biocatalysis in the Pharmaceutical Industry: The Need for Speed, ACS Med. Chem.
Lett., 8 (2017) 476-480, https://doi.org/10.1021/acsmedchemlett.7b00114
[7] T. Classen, J. Pietruszka, Complex molecules, clever solutions – Enzymatic approaches towards
natural product and active agent syntheses, Bioorg. Med. Chem. 26 (2018) 1285-1303,
http://dx.doi.org/10.1016/j.bmc.2017.06.045
[8] R. A. Sheldon, D. Brady, The limits to biocatalysis: pushing the envelope, Chem. Commun., 54
(2018) 6088-6104, https://doi.org/10.1039/C8CC02463D
[9] K. Faber (Ed.), Biotransformations in Organic Chemistry (sixth ed.), Springer, Berlin-Heilderberg
(2011)
[10]R. A. Sheldon, J. M. Woodley, Role of Biocatalysis in Sustainable Chemistry, Chem. Rev. 118 (2018)
801-838, https://doi.org/10.1021/acs.chemrev.7b00203
[11] U.T. Bornscheuer,R.J. Kazlauskas, Hydrolases in Organic Synthesis: Regio- and Stereoselective
Biotransformations, 2nd ed. Blackwell Science Publ, Oxford.
[12] C.-P. Lin, X.-L.g Tang, R.-C. Zheng, Y.-G. Zheng, Efficient chemoenzymatic synthesis of (S)-α-amino-
4-fluorobenzeneacetic acid using immobilized penicillin amidase, Bioorg. Chem. 80 (2018) 174-179,
https://doi.org/10.1016/j.bioorg.2018.06.020
[13] G. Angajala, P. Pavan, R. Subashini, Lipases: An overview of its current challenges and prospectives
in the revolution of biocatalysis, Biocatal. Agric. Biotechnol. 7 (2016) 257-270,
https://doi.org/10.1016/j.bcab.2016.07.001
[14] A.C.L.M. Carvalho, T.S. Fonseca, M.C. De Mattos, M.C.F. De Oliveira, T.L.G. De Lemos, F. Molinari,
D. Romano, I. Serra, Recent advances in lipase-mediated preparation of pharmaceuticals and their
intermediates, Int. J. Mol. Sci., 16 (2015), 29682-29716 https://doi.org/10.3390/ijms161226191
[15] P. Majewska, M. Serafin, M. Klimek-Ochab, M. Brzezińska-Rodak, E. Żymańczyk-Duda, Lipases and
whole cell biotransformations of 2-hydroxy-2-(ethoxyphenylphosphinyl)acetic acid and its ester,
Bioorg. Chem. 66 (2016) 21-26, http://dx.doi.org/10.1016/j.bioorg.2016.02.011
[16] M. Rachwalski,. N. Vermue, F. P. J. T. Rutjes, Recent advances in enzymatic and chemical
deracemisation of racemic compounds, Chem. Soc. Rev., 42 (2013) 9268-9282,
https://doi.org/10.1039/c3cs60175g
[17] A. Ghanem, Trends in lipase-catalyzed asymmetric access to enantiomerically pure/enriched
compounds, Tetrahedron 63 (2007) 1721-1754, https://doi.org/10.1016/j.tet.2006.09.110
[18] M. Kapoor, N. Anand, S. Koul, S.S. Chimni, K.S. Manhas, C. Raina, R. Parshad, S.C. Taneja, G.N. Qazi,
Kinetic resolution of 1-chloro-3-(1-naphthyloxy)-2-propanol, an intermediate in the synthesis of b-
adrenergic receptor blockers, Bioorg. Chem. 31 (2003) 259-269, https://doi.org/10.1016/S0045-
2068(03)00050-6
[19] M. Baiula,P. Galletti, G. Martelli, R. Soldati, L. Belvisi, M. Civera, S.D.Dattoli, S.M. Spampinato, D.
Giacomini, New β-Lactam Derivatives Modulate Cell Adhesion and Signaling Mediated by RGD-
Binding and Leukocyte Integrins. J. Med. Chem. 59 (2016) 9721-9742, https://doi.org/
10.1021/acs.jmedchem.6b00576
[20] P. Galletti, R. Soldati, M. Pori, M. Durso, A. Tolomelli, L. Gentilucci, S.D. Dattoli, M. Baiula, S. M.
Spampinato, D. Giacomini, Targeting integrins v3 and 51 with new -lactam derivatives. Eur.
J. Med. Chem. 2014, 83, 284-293, https://doi.org/ 10.1016/j.ejmech.2014.06.041
[21] Hynes, R. O. Integrins: bidirectional, allosteric signaling machines. Cell 110 (2002) 673-687 ,
https://doi.org/ 10.1016/S0092-8674(02)00971-6
12

[22]A. Tolomelli , P. Galletti, M. Baiula, D. Giacomini, Can Integrin Agonists Have Cards to Play against
Cancer? A Literature Survey of Small Molecules Integrin Activators. Cancers 9 (2017) E78.
https://doi.org/10.3390/cancers9070078
[23]K. Ley, J. Rivera-Nieves, W.J. Sandborn,.S. Shattil, Integrin-based therapeutics: biological basis,
clinical use and new drugs, Nat. Rev. Drug. Discov. 15 (2016) 173-183 https://doi.org/
10.1038/nrd.2015.10
[24]J. Kaman, E. Forrò, F. Fülöp, Enzymatic resolution of alicyclic β-lactams, Tetrahedron: Asymmetry 1
(2000) 1593–1600, https://doi.org/10.1016/S0957-4166(00)00070-7
[25]E. Forrò, F. Fulop, Synthesis of 4-aryl-substituted β-lactam enantiomers by enzyme-catalyzed kinetic
resolution, Tetrahedron: Asymmetry 12 (2001) 2351-2358 https://doi.org/10.1016/S0957-
4166(01)00388-3
[26]T. Miyazawa, E. Kaito, T. Yukawa, Enzymatic resolution of 2-aryloxy-1-propanols via lipase-catalyzed
enantioselective acylation using acid anhydrides as acyl donors, J. Mol. Catal. B-Enzym. 37 (2005)
63-67, https://doi.org/10.1016/j.molcatb.2005.09.006
[27]J. M. Keith, J. F. Larrow, E. N. Jacobsen, Practical considerations in kinetic resolution reactions, Adv.
Synth. Catal., 343 (2001) 5-26, https://doi.org/10.1002/1615-4169(20010129)343:1<5::AID-
ADSC5>3.0
[28]P. Galletti, M. Pori, D. Giacomini, One-step oxidation of 2-arylpropanols to 2-arylpropionic acids:
improving sustainability in the synthesis of profens, Synlett 17 (2010) 2644-2648,
https://doi.org/10.1055/s-0030-1258580
[29]P. Galletti, G. Martelli, G. Prandini, C. Colucci, D. Giacomini, Sodium periodate/TEMPO as a selective
and efficient system for amine oxidation, RSC Adv. 8 (2018) 9723-9730.
https://doi.org/10.1039/C8RA01365A
[30] X-G. Li, L. T. Kanerva, Chemoenzymatic preparation of fluorine-substituted β-lactam enantiomers
exploiting Burkholderia cepacia lipase, Tetrahedron: Asymmetry, 18 (2007) 2468–2472,
https://doi.org/10.1016/j.tetasy.2007.10.003
[31]N. Ikota, H. Shibata, K. Koga, Stereoselective reactions. IX. Synthetic studies on optically active β-
lactams. I. Chiral synthesis of carbapenam and carbapenem ring systems starting from (S)-aspartic
acid Chem. Pharm. Bull. 33 (1985) 3299-3306. https://doi.org/10.1248/cpb.33.3299
[32] R. De Marco, G. Mazzotti , S.D. Dattoli, M. Baiula M, S. Spampinato, A. Greco, L. Gentilucci. 5-
aminomethyloxazolidine-2,4-dione hybrid α/β-dipeptide scaffolds as inductors of constrained
conformations: Applications to the synthesis of integrin antagonists, Biopolymers, 104 (2015) 636-
49, http://doi.org/10.1002/bip.22704
[33] S.D. Dattoli, R. De Marco, M. Baiula, S. Spampinato, A. Greco, A. Tolomelli A, L. Gentilucci Synthesis
and assay of retro-α4β1 integrin-targeting motifs. Eur J Med Chem 73 (2014) 225-32
http://doi.org/10.1016/j.ejmech.2013.12.009
Fig. 1. -Lactam–based ligands evaluated as racemic mixtures, and EC50 values as agonists of specific integrins (in brackets).
Fig. 2. KR of racemic -lactam 5 with Burkolderia Cepacea lipase (BCL) or Pseudomonas fluorescen,
vinyl acetate or vinyl butyrate were used as acyldonors. The configurations reported in the scheme
for 6, 7, and 8 were assigned as described in the text. Plots of enantiomeric excesses of enantio-
enriched substrate 7 (blue) and acetate 6 (red) or butyrate 8 (green) vs conversion are depicted.
13

Fig. 3. Effects of -lactams (S)-11 and (R)-11 in cell adhesion assays on Jurkat cell line expressing α4β1 integrin, and (S)-
12, and (R)-12 on K562 cell line expressing α5β1 integrin. Concentration-response curve showing the effects of (S)-11,
(R)-11, (S)-12, and (R)-12 on cell adhesion are shown. Results are expressed as the number of cells attached ±S.E.M
from quadruplicate wells and repeated at least three times. EC50 values ± SD are reported on the corresponding curves.
Scheme 1. Synthetic strategy to obtain enantiopure azetidinones to be evaluated as integrin ligands through enzymatic KR.
Scheme 2. N-oxymethylene insertion to obtain 5.
Scheme 3. Results of the kinetic resolution of racemic -lactam 5. Reagents and conditions: a) BCL, vinyl butyrate, TBME, 37 °C.
Isolated yields, ee%, and specific optical rotatory powers are reported.
Scheme 4. Synthesis of -lactams enantiomers (+)9 and (-)9. Reagents and conditions: a) KMnO4, CH3CN, 0 °C then rt, 16 h, b) BCL,
H2O/CH3CN 11:1, 37 °C, 1 h. Isolated yields, ee%, and specific optical rotatory powers are reported.
Scheme 5. Synthetic elaborations for the assignment of absolute configurations. Hydrogenolysis of (+)9 and (-)9 to
obtain (+)10 and (-)10. Assignment of the absolute configurations by comparison with the reference compound.
Reagents and conditions: a) H2, Pd/C (10%), THF/CH3OH 1:1, rt, 2 h.
Scheme 6. Synthetic sequence for the obtainment of the enantiomerically pure ligands (S)-11, (R)-11, (S)-12, and (R)-12.
Reagents and conditions: a) K2CO3, o-tolylisocyanate, CH3CN, rt, 2 h; b) H2, Pd/C (10%), THF/CH3OH 1:1, rt, 2 h; c) EDC,
TEA, DMAP, glycine benzylester∙PTSA, CH2Cl2, 0 °C then rt, 16 h.
14

Table 1. Enzymatic KR of the racemic compound 1 as ethyl, benzyl, or tbutyl esters.a
\fx
Entry
Compound
(R)
enzyme
Enzyme amount
(U)
Conversion %
(time)
2/3/4b
1 erc
1
2
3
4
5
6
7
8
9
1a (Et)
1a (Et)
1a (Et)
CAL A
CAL B
PLE
α-chym
CAL A
CAL B
PLE
α-Chym
CAL A
CAL B
PLE
45
150
170
84
15
150
34
84
45
150
170
84
93 (1h)
82 (5h)
100 (1h)
68 (5h)
40/0/60
37/0/63
100/0/0
100/0/0
72/0/28
68/0/32
100/0/0
100/0/0
100/0/0
72/28/0
18/82/0
traces/0/0
60/40
65/35
-
1a (Et)
-
1b (Bn)
1b (Bn)
1b (Bn)
1b (Bn)
1c (tBu)
1c (tBu)
1c (tBu)
1c (tBu)
95 (1h)
53/47
46/54
-
13/87
36/63
50/50
82/12
97 (24h)
100 (5h)
41 (24h)
49(1h)
36 (24 h)
43 (1h)
10
11
12
α-Chym
15 (24 h)
a Reaction conditions: ester 1a-c (0.1 mmol), PBS 0.1M pH 7.5 (4 mL), enzyme (U in table), rt
b Determined by 1H NMR of aqueous phases after solvent extraction and lyophilization.
c enantiomeric ratio (er) of the unreacted 1a-c determined by chiral HPLC.
Table 2. Screening of lipases in the KR of racemic -lactam 5.a
O
COOBn
COOBn
O
COOBn
OH
enzyme
+
+
N
OH
N
O
N
O
O
O
TMBE, rt
6
5
7
O
Entry
Lipases
Enzyme amount (U) Time Conv. (%)b
6 ee%b
29
7 ee%b
28
6
1
2
CaL A (CLEA)
Mucor javanicus
30
60
1 h
4 d
43
6
79
3
Penicillium roqueforti
Aspergillus niger
Rhizopus arrhizus
Porcine kidney acylase
Mucor miehei
6.5
4.5
22
4 d
7
62
6
4
24 h
24 h
44 h
24 h
24 h
24 h
4 d
10
14
15
18
19
19
26
59
8
5
77
17
21
24
18
13
20
6
300
14
84
7
87
8
Hog pancreas
135
28
79
9
Candida cylindracea
44
10
Chromobacterium viscosum
2962
39
15

11
12
13
Aspergillus oryzae
Burkholderia cepacia L (Amano PS)
Pseudomonas fluorescens
200
30
4 d
4 d
57
50
70
75
68
31
45
89
91
180
24 h
a
Reaction conditions: compound 5 (20 mg, 1 equiv), TBME (2 mL), vinyl acetate (6 equiv), enzyme (U in
table), rt
b
Determined by chiral HPLC analysis. The configurations reported in the table scheme for 6 and 7 were
assigned as described in the text.
Table 3. Kinetic resolution of -lactam 5 with different acyl donors in TBME at 37 °C.a
BCL
COOBn
OH
acyl donor
COOBn
COOBn
OH
+
N
N
O
R
N
TBME, 5.5 h
37 °C
O
O
O
5
O
7
6
8
R = CH3
R = CH2CH2CH3
Entry
Acyl donor
Conversion (%)b
7 (ee%)b
Products (ee %)
1
2
3
4
Vinyl acetate
64
88
6 (61)
Vinyl butyrate
Acetic anhydride
Butyric anhydride
46
12
53
61
26
91
8 (90)
6 (74)
8 (87)
a
Reaction conditions: compound 5 (20 mg, 1 equiv), TBME (2 mL), acyl donor (6 equiv),
BCL (20 mg), 37°C, 5.5 h
b Determined by chiral HPLC analysis and the configurations reported in the table scheme for 6, 7 and 8 were
assigned as described in the text.
16