Biocatalytic Process Optimization for the Production of High-Added-Value 6-O-Hydroxy and 3-O-Hydroxy Glycosyl Building Blocks

Article abstract of DOI:10.1002/cctc.201601632

A biocatalytic process to synthesize regioselective monohydroxy glycosyl building blocks has been optimized. Lipases immobilized on commercial supports were treated with water-soluble carbodiimide (EDC) at different concentrations. In the presence of cosolvents, the stability of lipases adsorbed on octyl-Sepharose improved after the EDC modification. The new Candida rugosa lipase (CRL) modified heterogeneous biocatalysts were tested in the production of 6-OH hydroxyl-tetraacetyl glucose by a regioselective mono-deacetylation in aqueous media. Improvements in activity and excellent regioselectivity were obtained for octyl-CRL-EDC10mM preparation, with 95 % isolated yield of product on a multimilligram scale. We also observed excellent recyclability. The C-6 alcohol was transformed to a C-3 alcohol by chemical migration, and both compounds were transformed successfully in the corresponding new trichloroacetimidyl glucoderivatives. Modified CRL biocatalysts were also tested in the selective deprotection of peracetylated thymidine, and octyl-CRL-EDC10mM showed excellent specificity and improved regioselectivity to produce 3-hydroxy-5-acetyl-thymidine, a precursor of azidethymidine (AZT), in 95 % yield. The new Rhizomucor miehei lipase (RML)-modified heterogeneous biocatalysts showed excellent regioselectivity and recyclability in the 3-OH mono-deprotection of peracetylated lactal.

Full text of DOI:10.1002/cctc.201601632

Accepted Article
Title: Biocatalytic process optimization for production of high-added
value 6-O-hydroxy and 3-O-hydroxy glycosyl building blocks
Authors: Ciara Callaghan, Martin Redmond, rOBSON c. Alnoch, Cesar
Mateo, Marco Filice, and Jose M Palomo
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemCatChem 10.1002/cctc.201601632
Link to VoR: http://dx.doi.org/10.1002/cctc.201601632
A Journal of
www.chemcatchem.org

10.1002/cctc.201601632
ChemCatChem
Biocatalytic process optimization for production of high-added value
6-O-hydroxy and 3-O-hydroxy glycosyl building blocks
Ciara Callaghan[a], Martin Redmond[a], Robson Carlos Alnoch[a], Cesar Mateo[a],
Marco Filice*[a,b] and Jose M. Palomo*[a]
[a] C. Callaghan, M. Redmond, R. C. Alnoch, Dr. C. Mateo, Dr. M. Filice, Prof. Jose
M. Palomo
Department of Biocatalysis. Institute of Catalysis (CSIC). Marie Curie 2. Cantoblanco.
Campus UAM, 28049 Madrid, Spain. Fax: +34-91-585-4760.
E-mail: josempalomo@icp.csic.es / marco.filice1@gmail.com
[b] Dr. M. Filice
Current address: Advanced Imaging Unit (CNIC), CIBERES, Melchor Fernandez
Almagro, 3, 28029, Madrid. Spain.
1
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
Abstract
A biocatalytic optimization process to synthesize regioselective monohydroxy glycosyl
building blocks has been performed. Lipases immobilized on commercial supports were
treated with water soluble carbodiimide (EDC) at different concentrations. In the
presence of co-solvents, the stability of lipases adsorbed on octyl-Sepharose improved
after the EDC modification. The new Candida rugosa lipase (CRL) modified
heterogeneous biocatalysts were tested in the production of 6-OH hydroxyl-tetraacetyl
glucose by a regioselective monodeacetylation in aqueous media. Improvements in
activity and excellent regioselectivity were obtained for octyl-CRL-EDC10mM
preparation, with 95% isolated yield of product in multimilligrams scale. Also excellent
recyclability was observed. The C-6 alcohol was transformed in C-3 by chemical
migration and both compounds were successfully transformed in the corresponding
novel trichloroacetimidyl glucoderivatives. Modified CRL biocatalysts were also tested
in the selective deprotection of peracetylated thymidine, and octyl-CRL-EDC10mM
showed excellent specificity and improved regioselectivity, producing 3-hydroxy-5-
acetyl-thymidine, precursor of azidethymidine (AZT), in 95% yield. The new RML
modified heterogeneous biocatalysts showed excellent regioselectivity and recyclability
in the 3-OH monodeprotection of peracetylated lactal.
2
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
Introduction
Glycoderivatives are a versatile class of molecules, useful as key intermediates in the
synthesis of novel bio-active compounds such as drugs and vaccines, for example,
which will allow human beings to overcome diseases, improving the quality of life of
patients, or even in the development of early diagnosis and more effective and targeted
therapeutic modalities.[1-5] Therefore, the development of more effective and economic
strategies for preparing these molecules is mandatory.
In particular, one of the key steps in glycochemistry is the efficient access to new
partially protected monosaccharides containing a strategically positioned free hydroxyl
group (a nucleophilic acceptor) as building blocks for a regio- and stereochemical
controlled synthesis of oligosaccharides and glycoderivatives. Different chemical
methods have been described in literature based on regioselective orthogonal protection
strategies and different combinations, there for to give access to these specific
monosaccharides.[6-7] However, the ideal strategy would be one where only a single
protecting group is used, ensuring a high conversion and good final yields. An
extremely high control of selectivity during the synthesis is mandatory in order to
reduce the synthetic steps and to obtain pure homogeneous compounds with the highest
economic efficiency.
In this way, regioselective deprotection of fully protected (with the same protecting
group) monosaccharides represents one of the best choices. However, the preparation of
these molecules by chemical methods is tedious and hard work (several steps, tedious
work-up, time-consuming purification, moderate yields),[8] because of the difficulty to
discriminate between groups with a very similar reactivity which does not completely
fulfill the requirement of a very regioselective methodology.
3
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
In recent years our group has developed a straightforward strategy to synthesize
diversely monodeprotected mono- and disaccharide building blocks from fully
acetylated carbohydrates catalyzed by immobilized lipases at very mild conditions.[9]
The strategy is based on the complex catalytic mechanism of lipases which implies
dramatic conformational changes of the enzyme molecule between a closed and an open
form.[10]
Indeed, in most cases the lipase adsorbed on a hybrophobic surface, fixing the open
conformation, resulting in the best catalyst in terms of reaction yields and
regioselectivity providing valuable intermediates, such as the C-6 or C-3
monodeacetylated molecule (key positions in the synthesis of biological relevant
oligosaccharides).[9a,11] High activity and regioselectivity have been obtained using
lipase from C. rugosa (CRL) adsorbed on commercial octyl-Sepharose as catalyst for
regioselective
one-pot
preparation
of
different
C-6
monodeprotected
monosaccharides[9a] and the lipase from R. miehei (RML) adsorbed on commercial
octyl-Sepharose for regioselective deprotection C-3 in peracetylated lactal.[12]
However, despite the extremely high catalytic parameters shown by these immobilized
lipases in these processes, a reversible immobilization was promoted in both cases. As
consequence, the enzyme catalyst can be desorbed from the heterogeneous support in
precise conditions (e.g. in the presence of hydrophobic or amphiphilic agents).[13]
Therefore, issues such as stability of the catalyst, reutilization, activity and control of
the excellent regioselectivity require to be optimized before a possible implementation
at industrial scale of this biocatalytic regioselective process.
The free-hydroxyl acetylated molecules can be easily convertible in substituted mono-
and polysaccharides useful in chemical, biological and medicinal studies. This process is
4
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
performed by a previous activation of the hydroxyl group. Many strategies have been
described in literature although in most cases by activation of the anomeric position of
the sugar. [14-15] Here, we produce selectively 6-O and 3-O hydroxy glycosyl building
blocks which can be a new alternative position for activation making a great relevance in
biomedical studies. The biological activity of natural products in many cases derives
from the sugar moieties and changes in a simple single position in the sugar structure
could be dramatically important such as occur in the different ABO blood groups.[16] The
chemical activation of anomeric hydroxyl group in sugar is usually performed by using
tricloroacetonitrile at base conditions generating the corresponding tricloroacetamidate
(TCA) derivative.[15c] This modification in other positions of the sugar will permit to
create a new class of neo-glycoconjugates and glycoderivatives.
Herein, the chemo-enzymatic regioselective monodeprotection process of different
peracetylated glycoderivatives catalyzed by CRL and RML lipases immobilized on
commercial octyl-Sepharose has been improved by a post-immobilization treatment
with water-soluble carbodiimide, enhancing the catalyst stability while conserving its
excellent regioselectivity. By this way, the production of different monodeprotected
building-blocks in multimilligram scale has been successfully achieved. Subsequently
to this biocatalytic optimization, the chemical activation of the produced free hydroxyl
group producing the novel trichloroacetimidyl glucoderivatives (6-OTCA and 3-OTCA)
was satisfactorily carried out.
5
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
Results and Discussion
Lipase from C. rugosa adsorbed on commercial octyl-Sepharose (octyl-CRL) or
Lewatit1600 (Lew-CRL) and lipase from R. miehei adsorbed on commercial octyl-
Sepharose (octyl-RML) were incubated with EDC solution (at different concentrations)
by a solid-phase reaction strategy in order to avoid the leakage of the enzyme from the
catalysts. The effect of this treatment in the enzyme activity of the immobilized
preparations was evaluated in the hydrolysis of p-nitrophenylbutyrate. The activity was
almost conserved in all cases after treatment with 10 mM of EDC (Lew-CRL suffered a
decrease of 20% value), whereas loss of almost 30-40% in activity was observed using
higher EDC concentrations (Table 1).
In order to choose the optimal modification for the particular biotransformation, the
stability of the different modified-EDC lipase preparations by incubation at different
concentrations of co-solvents was studied.
First, the stability of the CRL preparations was evaluated in the presence of 20% or
50% (v/v) acetonitrile (Figure 1 and 2). CRL is a very sensitive enzyme compared with
[13a,17]
other lipases
and 50% co-solvent represents a very harsh condition. After
incubation in an aqueous solution of 50% acetonitrile, an improvement on stability for
the modified octyl-CRL preparations was observed (Figure 1) whereas for Lew-CRL a
dramatic loss in stability was observed after 1 h incubation (less than 20% of remaining
activity) (Figure 1A). In the case of octyl-CRL, the non-modified preparation suffered a
strong activity lost caused by the release of the enzyme from the support at these
conditions (soluble enzyme is instantly inactive at these conditions).[13a] In fact in 1 h
less than 50% activity was recovered, whereas the EDC modified catalysts conserved
75% activity at the same time. For longer incubation time, also non-modified Lew-CRL
and octyl-CRL showed less than 10% activity after 5 h incubation. On the other hand, at
6
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
the same time, the EDC modified octyl-CRL catalysts conserved more than 25%
activity whereas the corresponding modified Lew-CRL catalysts were completely
inactive (Figure 1).
Considering this profile in very drastic conditions, the biocatalysts stability was studied
using 20% (v/v) co-solvent at 25ºC (Figure 2). At these conditions the variations
between the non-modified, 10mM EDC or 50 mM EDC were more apparent.
CRL immobilized on Lewatit was less stable than immobilized on octyl-Sepharose, and
again the EDC modification resulted negative for the enzyme stability in the former
(Figure 2A). The enzyme was completely inactive after 2 h in the 50 mM EDC
preparation and conserving 25% activity in the case of 10 mM EDC preparation,
whereas the 75% activity of Lew-CRL was conserved at this time.
The chemical modification of the octyl-CRL with 10 mM EDC generated once again
the best catalyst, maintaining almost complete activity after 50h incubation, whereas the
CRL preparation modified with 50 mM EDC was slightly better than the non-modified
which conserved 50% activity after 50 h (Figure 2B).
The stability of octyl-RML was also evaluated in the presence of 25% and 50% acetone
(v/v) at 25ºC (Figure 3). The non-modified catalyst showed a decrease of 40% activity
after incubation in 25% acetone after 25 h. The different EDC modified octyl-RML
catalysts showed an excellent stability conserving 100% activity at these conditions in
all cases (Figure 3A). When the RML biocatalysts were incubated at 50% acetone (v/v),
octyl-RML resulted inactive after 1h, whereas the EDC modified catalyst conversed
more than 50% activity at that time. The 25mM and 50mM EDC modified catalysts
were the most stable maintaining around 60% activity after 25 h incubation (Figure 3A).
Therefore, in the case of CRL, the chemical modification was positive for octyl-CRL
7
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
and negative for Lew-CRL and the best results were obtained using 10 mM EDC
modification conditions whereas for octyl-RML in all cases, the EDC modification
strongly improved the stability. Considering the stability and the remaining activity, 25
mM EDC modification was selected as optimal conditions.
RML and CRL adsorbed on commercial octyl-Sepharose can be completely desorbed
using 0.5% triton X-100.[13] The covalent attachment effect of this modification strategy
was evaluated. These modified biocatalysts were incubated in this detergent solution
and no releasing of the enzyme from the support was observed.
Therefore to reveal the reason of the covalent interaction between the enzyme and the
solid support, a FT-IR analysis of the dry octyl-Sepharose was performed (Figure 4).
The spectrum showed a broad peak around 3380 cm−1, which is due to the stretching
vibration of O-H bonds, while peaks around 2890 cm−1 are due to the C-H stretch. In the
range 1070–1140 cm−1, peaks related to C-O-C stretch vibration that represents the
glycosidic bond between monosaccharide units (typical of agarose structure) and ester
groups.[18] A peak of 933 cm−1 represents the C-O-C bridge of 3,6-anhydrogalactose
unit. Finally, a peak around 1635 cm−1 is clearly observed, which corresponds to the
CO2 stretch vibration of carbonyl group at carboxylic acid (Figure 4). [18] Thus this result
directly demonstrates the presence of carboxylic groups in the agarose beads which are
the responsible of the covalent attachment of the enzyme in the EDC treatment.
Furthermore, analyzing the enzyme structure (Figure 5) it has been observed how
different aminoacids such as lysines can be involved in the reaction with the activated
carboxylic groups of the support, especially in the open form fixed in the immobilized
enzyme on the octyl-Sepharose (Figure 5B).
8
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
In this way, in order to evaluate the applicability of this technique in terms of specificity
and regioselectivity, the different modified enzyme preparations were used as catalysts
in the monodeprotection of different peracetylated molecules (1, 6 and 9) at 25ºC and
pH 5.
In the course of different projects, we have determined that octyl-CRL is an optimal
catalyst for the regioselective hydrolysis of peracetylated glucose (1). These catalysts
selectively produced the 6-OH hydroxyl-tetraacetyl glucose (2) at very high excellent
yield with a simple work-up[9a] (Scheme1).
Consequently, the effect of the modification with EDC of this catalyst was tested in this
reaction (Table 1). Also Lew-CRL was evaluated in this reaction. The preparation of
immobilized lipase on this support have been described as very promising in some
cases[19-22] and indeed the widely used commercial biocatalyst Novozym435 is the
immobilized form of Candida antarctica B lipase on Lewatit1600.[23]
The octyl-CRL-EDC10mM preparation showed excellent specificity and
regioselectivity as the non-modified enzyme in the hydrolysis of 1 at pH 5, 25ºC in the
presence of 20% acetonitrile, being slightly more active than the former obtaining in
both cases >99% yield of 2 (Table 1, Scheme 1).
Surprisingly, the Lew-CRL preparation was around 3-fold more active than the octyl-
CRL in the deprotection although the enzyme in this form showed less specificity,
producing 85% yield of the desired monodeprotected product, and 15% of other
products (other regioisomers and hydrolyses in other positions) (Table 1). The
modification of the enzyme in this case did not alter any property of the enzyme.
At this point the recyclability of the regioselective octyl-CRL preparations was studied
(Figure 6). The results after 4 cycles of reaction were very different. The octyl-CRL
9
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
suffered a decrease in the activity with the cycles, conserving only 29% initial value
after 4 cycles. However, the octyl-CRL-EDC10mM preparation almost conserved total
activity after the second cycle and more than 70% after the fourth cycles (Figure 6). In
all cycles the regioselectivity of the enzyme was conserved.
Therefore, these results demonstrate that the modification permits to conserve the
excellent properties of the catalyst adding the higher stability to the enzyme, being the
last a relevant characteristic for a possible industrial application.
The final monodeprotected product 2 was obtained in each case with a 95% isolated
yield. This is a very interesting building block in carbohydrate chemistry. One of the
most useful activated sugars for oligosaccharides synthesis is the trichloroacetamidyl
ester (TCA) product. Therefore, the free OH of 2 was activated by using
trichloroacetonitrile and DBU as nucleophilic base at 0 ºC to produce the corresponding
6-OTCA-tetraacetyl-glucose 3 in excellent yields (>90%) (Scheme 1).
In order to widen the activation possibilities and the overall scope of our proposed
strategy, we decided to apply on product 1 an optimized chemo-enzymatic strategy
developed by ourselves permitting to achieve in one-pot glucoderivatives
regioselectively deprotected in other positions then 6-OH. [9a] In more details, the 3-OH
glucose 4 was obtained in 25% overall yields after incubation of 1 solution at alkaline
pH and chromatography purification.[9a] Also this building-block was used in the
activation synthesis with TCA. By this way, the novel 3-OTCA compound 5 was
obtained in excellent yields (>90%) (Scheme 1).
To assess the applicability of the proposed strategy, another interesting
biotransformation was tested: the selective deprotection of per-acetylated thymidine 6.
Indeed, the 3-hydroxy-5-acetyl-thymidine (7) is a key direct intermediate in the
10
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
synthesis of azidethymidine (AZT) (Scheme 2) also called zidovudine, the first
antiretroviral drug. The catalytic performance of the different CRL biocatalysts was
assessed in the hydrolysis of 6 at 25ºC and pH 5 (Table 3). Comparing octyl-CRL and
Lew-CRL, the latter was more active although slightly worst in term of specificity
producing less amount of 7 but similar in regioselectivity, producing in both cases 8%
of free hydroxyl in 5´(8) (Table 3). The EDC modification caused different effects
depending on the immobilized CRL preparation. The octyl-CRL-EDC10mM showed
excellent specificity and improved regioselectivity compared to the non modified
catalyst, producing 95% of 7 at full conversion (Table 3). However, Lew-CRL-
EDC10mM exhibited a loss in specificity in comparison with Lew-CRL, only 48%
yield of 7, although slight improvement of regioselectivity was observed (Table 3).
Therefore these examples have demonstrated that the EDC modification of CRL
immobilized on commercial octyl-Sepharose greatly improves the catalyst performance.
Furthermore, in order to broad the scope of the method, this treatment performed in
octyl-RML was also evaluated in term of specificity and regioselectivity of the enzyme
in the hydrolysis of peracetylated-lactal (9) (a lactose-derivative compound very
interesting in oligosaccharide synthesis) at pH 5 and 25ºC in the presence of 20%
acetone as co-solvent (Scheme 3). In this time the three modified RML catalysts with
different EDC modification (10, 25 and 50 mM) were tested.
In all cases the enzyme activity of the EDC modified catalyst was 3 times lower
compared to octyl-RML. However, the high regioselectivity of the octyl-RML was
conserved after the different EDC modifications which even slightly improved the
specificity of the enzyme producing more than 90% of C-3 monodeprotected product
10. Therefore the octyl-RML-EDC25mM was chosen as the best one considering
11
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
selectivity and stability, it was recycled in this reaction up to seven times and complete
activity of the enzyme was conserved (data not shown).
Conclusions
The solid-phase chemical modification by water-soluble carbodiimide of lipases
previously immobilized on octyl-Sepharose have permitted the obtainment of a more
stable and in some cases more active and more regioselective enzyme. This new catalyst
overcomes the disadvantage of reversibility of this immobilization technology,
generating heterogeneous biocatalysts, which couple the numerous advantages of the
octyl-Sepharose support to covalent immobilization, with an excellent recyclability of
the catalyst. Also this strategy is a suitable optimization because in all cases an excellent
regioselectivity in the monodeprotection of peracetylated molecules was conserved,
obtaining different glycosyl and nucleosidic building blocks in multimiligram scale.
Therefore, this strategy represents a simple and effective methodology for interesting
industrial application of the preparation of these molecules. This process could be scale
at grams and especially for the 6-OH monodeprotected tetraacetylated glucose it could
be possible to multigram scale.
12
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
Experimental
General
Lipase from Candida rugosa (CRL), per-O-acetylated α-glucose (1), per-O-acetylated
lactal (9), thymidine, N-(3-Dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride
(EDC), trichloroacetonitrile (TCA), Boron trifluoride ethyl etherate (BF3·OEt2), 1,8-
Diazabicyclo[5.4.0]undec-7-ene (DBU) were from Sigma Chem. Co (St. Louis, USA).
Lipase from Rhizomucor miehei (RML) was kindly donated by Novozymes (Denmark).
Lewatit® VP OC 1600, a macroporous DVB-crosslinked polymer in spherical bead
form based on methacrylic esters, was from Lanxess AG (Germany). Octyl-Sepharose
4BCL fast flow was from GE heathcare (Uppsala, Sweden). TLC analyses were run on
silica plates (Merck 60 F254). HPLC analyses were performed using HPLC-spectra P400
(Thermo Separation products). The column was a Kromasil-C18 (250-4.6 mm and 5
mm) from Analisis Vinicos (Tomelloso, Spain). Analyses were run at 25◦C using an L-
7300 column oven at flow of 1 mL/min and UV detector L-7400.
Synthesis of per-O-acetylated thymidine (6)
Thymidine (1g, 4.12 mmol) was dissolved in 6 mL acetonitrile and then triethylamine
(TEA, 4 equiv.), acetic anhydride (4 equiv.) and a catalytic amount of 4-
(dimethylamino) pyridine (DMAP) were added. The reaction was stirred at room
temperature until the reaction was complete, followed by TLC analysis (ethyl
acetate/hexane (7/3) as eluent (Rf= 0.32) and visualized by UV light (254 nm)). Then
the solution was diluted with chloroform and water (1:1) (6 mL). The organic phase was
separated, washed with water (4x 20mL) and dried (Na2SO4), filtered, and concentrated
in vacuo. NMR data are in agreement with the reported values.[24]
13
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
Standard enzymatic activity assay
In order to follow the immobilization process, the activities of the soluble lipases and
their immobilized preparations were analyzed spectrophotometrically measuring the
increment in absorbance at 348 nm (= 5.150 M-1cm-1) produced by the release of p-
nitrophenol (pNP) in the hydrolysis of 0.4 mM pNPB in 25 mM sodium phosphate
buffer at pH 7 and 25°C. To initialize the reaction, 0.05-0.2 mL of lipase solution (blank
or supernatant) or suspension was added to 2.5 mL of substrate solution. Enzymatic
activity was determined as μmol of hydrolyzed pNPB per minute per mg of enzyme
(IU) under the conditions described above.
Preparation of biocatalysts
Four grams of CRL commercial powder was dissolved in 130 mL of 50 mM sodium
phosphate pH 6 (solution of 0.5 mg lipase per mL) and slow stirring for 30 min at 4ºC.
Then, solution was centrifuged at 12,000 rpm during 30 min and the supernant was
recovered.
Six milliliters of liquid solution of RML (containing aprox. 6 mg of lipase)[12] was
dissolved in 114 mL of 25 mM sodium phosphate buffer at 25ºC.
Octyl-Sepharose support was added to the enzyme solution; 2 g in the 65 mL CRL
solution and 1g in 120 mL solution of RML. The other 65 mL of CRL solution was
added to 2 g of Lewatit1600. The reactions were incubated at 4ºC for CRL and 25ºC for
RML for 4 h. After this time, enzymatic activity of the suspension and supernatant were
determined using the pNPB assay to determine the immobilization yield and remaining
enzymatic activity in the immobilized preparation. After immobilization, the enzyme
derivative was recovered by filtration under vacuum, washed with abundant distilled
water and conserved dry on the fridge. In all cases more than 95% of enzymatic activity
14
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
was immobilized on the support and the immobilized enzyme conserved the initial
activity.
Finally for the modified biocatalysts, one gram of the immobilized CRL biocatalyst
(octyl-CRL, Lewatit-CRL) and 0.5 g of octyl-RML was incubated in 10 mL of distilled
water pH 4.8 containing EDC at different concentrations (10, 20, 25, 50 mM) for 2 h
and then the suspension were filtrated, washed with distilled water and stored in the
fridge.
Stability of the different biocatalysts in the presence of co-solvent
The different unmodified and EDC-modified biocatalysts were incubated in phosphate
buffer solution containing 25% or 50% (v/v) of co-solvent (acetone or acetonitrile). The
activity of the immobilized enzyme was evaluated during the time by using the pNPB
assay.
FT-IR Spectra
Octyl-Sepharose support, after dissolved in distilled water, was freeze-dried. Infrared
spectra were then recorded using a Fourier transform infrared (FT-IR) spectra were
obtained on a PerkinElmer Spectrum 400 Series spectrometer (PerkinElmer, USA);
each spectrum was obtained by averaging 32 interferograms with a resolution of 1 cm-1.
Biocatalytic deprotection of per-O-acetylated glucose (1)
Per-O-acetylated α-glucose (1) (144 mg, 0.37 mmol) was dissolved in 4 mL of
acetonitrile. This solution was slowly mixed with 16 mL of 50 mM sodium phosphate
buffer at pH 4.8 and then 1 g of CRL preparations was added. The reaction was
followed by TLC and HPLC analysis, maintaining the hydrolysis reaction under
15
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
mechanical stirring at 25°C and the pH value constant by automatic titration in order to
avoid the chemical acyl migration. TLC conditions were Hexane:AcOEt 5:5 v/v, Rf 2:
0.25, Rf1:0.56. HPLC conditions were: eluent water: ACN 7:3 v/v, pH 4.3 adjusted by
adding TFA). Rt1 =20 min, Rt2 =7.3 min at 225 nm
When complete monodeprotection was reached, the reaction mixture was filtered by
vacuum on a sintered glass filter. The catalyst was washed with distilled water (x 3
times) and reused in a second reaction. This procedure was repeated two more times (4
cycles).
To obtain the monodeprotected product, the 20 mL solution containing the 6-OH
product was saturated with NaCl (6 g) and extracted with ethyl acetate (3 x 20 mL). The
organic layers were combined, wash with saturated aqueous sodium hydrogen carbonate
(2 x 20 mL) and dried it by adding anhydrous sodium sulfate. Solvent was evaporated in
a rotary evaporator and crystallized with cold diethyl ether to obtain the product as a
1
white solid (2, 136 mg, 95% yield). H-NMR data are in agreement with the reported
values.[9a]
Synthesis of 6-trichloroacetamidyl-tetra-O-acetyl-glucopyranose (3)
100 mg of 2 (0.26 mmol) was dissolved in 10 mL of CH2Cl2 in an ice bath and 2.6
mmol of trichloroacetonitrile was added. Then 0.54 mmol of DBU was added to the
mixture drop by drop and incubated for 1 h or until disappearance of 2. The reaction
procedure was followed by TLC (Hexane:AcOEt 5:5 v/v). The product was purified by
1
‘Cal Bridge/ Mini-chromatography column (90% of 3, 0.23 mmol). H-NMR (400
MHz, CDCl3) δ 8.52 (s, 1H-NH), 6.35 (d, J = 3.7 Hz, 1H-1), 5.48 (t, J = 9.8 Hz, 1H-3),
5.19 (t, J = 9.8 Hz, 1H-4), 5.09 (dd, J = 10.3, 3.7 Hz, 1H-2), 4.39 (d, J = 11.9, 3.3 Hz,
2H-6a,b), 4.16 (m, 1H-5), 4.22 (m, 1H-5), 2.16-2.01 (4s, 12H-Ac).
16
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
Synthesis of 3-trichloroacetamidyl-tetra-O-acetyl-glucopyranose (5)
The 3-OH precursor 4 was synthesized as described elsewhere.[9a] 100 mg of 4 (0.26
mmol) were dissolved in 10 mL of CH2Cl2 in an ice bath and 2.6 mmol of
trichloroacetonitrile was added. Then 0.54 mmol of DBU was added to the mixture drop
by drop and incubated for 1 h or until disappearance of 4. The reaction procedure was
followed by TLC (Hexane:AcOEt 5:5 v/v). The product was purified by ‘Cal Bridge/
Mini-chromatography column (90% of 5, 0.23 mmol).1H-NMR (400 MHz, CDCl3) δ
8.52 (s, 1H-NH), 6.35 (d, J = 3.57 Hz, 1H-1), 5.78 (t, J = 10.0 Hz, 1H-3), 5.35 (t, J =
9.85 Hz, 1H-4), 5.28 (dd, J = 10.26, 3.79 Hz, 1H-2), 4.26 (dd, J = 12.3, 4.0 Hz, 1H-6a),
4.16 (m, 1H-5), 4.10 (dd, J = 12.27, 4.0 Hz, 1H-6b), 2.19-2.00 (4s, 12H-Ac).
Biocatalytic deprotection of per-O-acetylated thymidine (4)
Substrate 4 (40 mg, 0.12 mmol) was dissolved in a mixture of acetonitrile (5%, v/v)
in 10 mM sodium acetate at pH 5.0. 0.2 g of CRL preparations was added to 2 mL of
this solution at 25°C. The degree of hydrolysis was analyzed by TLC and reverse phase
HPLC. TLC conditions were: dichloromethane/methanol, 92.5/7.5 v/v, Rf4: 0.25,
Rf5:0.12, Rf6:0.10. HPLC conditions were: eluent was obtained by following gradient
program (A: mixture of acetonitrile (10%, v/v) in 10mM ammonium phosphate at pH
4.2; B: mixture of miliQ water (10%, v/v) in acetonitrile; method: 0–6 min 100% A, 6–
14 min 85% A to 15%B, 14–22 min 100% A, flow: 1.0 mL·min−1). UV detection was
performed at 260 nm. The monodeprotected 3-OH (5) and 5-OH (6) were used as pure
standards. The retention time was 2.4 min for Thymidine, 9.4 min for 5 and 10.2 min 6
and 19 min for 4.
The reaction was then scale-up, starting with 200 mg (0.61 mmol) of thymidine
peracetylated in 25 mL volume (containing 5% ACN) and 1.5 g of octyl-CRL-
17
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
EDC10mM. The full conversion was achieved after 162 h with the same
regioselectivity. The product was purified and characterized in agreement with the
previous report.[25]
Biocatalytic deprotection of per-O-acetylated lactal (7)
7 (0.01 mmol) was added to 2 mL solution of phosphate buffer (25 mM) with
acetonitrile (20%) at pH 5, 25ºC and the reaction was initialized by adding 0.4 g of
biocatalyst. The hydrolytic reaction was carried out under mechanical stirring, and the
pH value was controlled by automatic titration. Hydrolysis reactions were followed by
HPLC, where the eluent was an isocratic mixture of 40% acetonitrile in 10 mM
ammonium phosphate buffer at pH 3.8 and UV detection at 215 nm. The retention time
of the product 7 and 8 was 18.50 min and 11.10 min respectively in these conditions.
When complete monodeprotection was reached, the reaction mixture was filtrated by
vacuum on a sintered glass filter. The catalyst was washed with distilled water (x 3
times) and reused in a second reaction. This procedure was repeated six times (7 cycles).
The reaction was scale-up using 5 g of octyl-RML-EDC catalyst in the hydrolysis of
0.53 mmoles (300 mg) of peracetylated lactal (6) in 40 mL volume (containing 20%
ACN) at pH 5. After 24 h, full conversion was achieved and the aqueous solution was
filtrated and saturated with NaCl and then extracted with ethyl acetate (3x40mL). The
collected organic layers were dried over anhydrous Na2SO4 which was then removed by
filtration and concentrated under vacuum producing a white solid (7, 270 mg, 90%).
NMR data are in agreement with previously reported.[12]
18
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
Acknowledgments
The authors thank the Ramon Areces Foundation for financial support and support by
the Spanish National Research Council (CSIC). We also thank the European community
for an ERA HEI mobility grants (ERASMUS+) to Ciara Callaghan and Martin
Redmond, the Brazilian National Council for Scientific and Technological
Development (CNPq) for fellowship to Robson Alnoch and Dr. Martinez from
Novozymes for the generous gift of lipase. Finally, we thank Spanish Ministry of
Economy for research grant SAF2014-59118-JIN co-funded by Fondo Europeo de
Desarrollo Regional (FEDER).
19
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
References
[1]B. Palitzsch, N. Gaidzik, N. Stergiou, S. Stahn, S. Hartmann, B. Gerlitzki, N.
Teusch, P. Flemming, E. Schmitt, H. Kunz, Angew. Chem. Int. Ed. 2016, 55,
2894-2898.
[2]S. Ingale, M.A. Wolfert, J. Gaekwad, T. Buskas, G.-J. Boons, Nat. Chem. Biol.
2007, 3, 663-667.
[3]K. Ren, C. He, C. Xiao, G. Li, X. Chen, Biomaterials 2015,51, 238-249.
[4]S. Woerly, S. Fort, I. Pignot-Paintrand, C. Cottet, C. Carcenac, M. Savasta,
Biomacromolecules 2008, 9, 2329-2337.
[5] P.H. Seeberger, D.B. Werz, Nature 2007,446, 1046–1051.
[6]a) T.-Y. Huang, M.M.L. Zulueta, S.-C. Hung, Org. Biomol Chem., 2014, 12,
376-382; b) A. Français, D. Urban, J.-M. Beau, Angew. Chem. Int. Ed. 2007,
45, 8662-8665; c) M. Filice, J.M. Palomo, RSC Advances 2012, 2, 1729-1742.
[7]a) C.-C. Wang, J.-C. Lee, S.-Y. Luo, S.S. Kulkarni, Y.-W. Huang, C.-C. Lee,
K.-L. Chang, S.-C. Hung, Nature 2007,446, 7138; b) C.-C. Wang, S.S.
Kulkarni, J.-C. Lee, S.-Y. Luo, S.-C. Hung, Nat. Prot. 2008, 3, 97-113.
[8]G. L. Thanh, et al. J. Org. Chem. 2010, 75, 197.
[9]A) M. Filice, J. M. Guisan, M. Terreni, J.M. Palomo, Nat.Protoc. 2012, 7,
1783-1796; b) M. Filice, T. Bavaro, R. Fernandez-Lafuente, M. Pregnolato, J.
M. Guisan, J.M. Palomo, M. Terreni, Catal. Today 2009,140, 11-18; c) J. M.
Palomo, M. Filice, R. Fernandez-Lafuente, M. Terreni, J.M. Guisan, Adv.
Synth Cat. 2007,349,1969-1976.
[10]
L. Brady, A. M. Brzozowski, Derewenda, Z. S. Dodson, E. Dodson, G.
Tolley, S. Turkenburg, J. P. Christiansen, L. Huge-Jensen, B. Norskov, L.
Thim, L. Menge, Nature 1990, 43, 767-770.
20
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
[11]
a) M. Filice, J. M. Palomo, P. Bonomi, T. Bavaro, R. Fernandez-
Lafuente, J.M. Guisan, M. Terreni, Tetrahedron 2008, 64, 9286-9292.
[12]
M. Filice, R. Vanna, M. Terreni, J.M. Guisan, J.M.
Palomo, Eur. J. Org Chem.2009,20,3327-3329.
[13]
a) J.M. Palomo, G. Fernández-Lorente, C. Mateo, C. Ortiz, R.
Fernández-Lafuentel, J.M. Guisan, Enzyme Microb. Technol. 2002, 31,775-783;
b) J.M. Palomo, G. Muñoz, G. Fernández-Lorente, C. Mateo, M. Fuentes, J.M.
Guisan, R. Fernández-Lafuente, J. Mol. Cat B: Enzymatic.2003, 21,201-210.
[14]
B. Ernst, G.W. Hart, P. Sinay (eds). Carbohydrates in Chemistry and
Biology, Vol. 1 (Wiley-VCH, 2000).
[15]
a) C. Gouliaras, D. Lee, L. Chan, M.S. Taylor, J. Am. Chem. Soc. 2011,
133, 13926-13929; b) E. Dimitrijević, M.S. Taylor, ACS Catal. 2013, 3, 945-
962; c) M. J. McKay, H. M. Nguyen, ACS Catal. 2012, 2, 1563-1595.
[16]
W.B. Turnbull, B.L. Precious, S.W. Homans, J Am Chem Soc. 2004, 126,
1047-1054.
[17]
C.W. Rivero, J.M. Palomo, Catalysts. 2016, 6,115; doi:10.3390/
catal6080115.
[18]
a) F. Rossi, M. Santoro, T. Casalini, P.Veglianese, M. Masi, G. Perale,
Int. J. Mol. Sci. 2011, 12, 3394-3408; b) N. Ashwinkumar, S. Maya, R.
Jayakumar. RSC Adv. 2014, 4, 49547; c) R. Imani, S. H. Emami, S. Faghihi,
Phys. Chem. Chem. Phys. 2015, 17, 6328-6339.
[19]
a) A.R. Faustino, N.M. Osorio, C. Tecelão, A. Canet, F. Valero, S.
Ferreira-Dias, Eur. J. Lip. Sci. Technol., 2016,118, 532-544.
[20]
D.S. No, T. Zhao, J. Lee, J.-S. Lee, I.-H. Kim, J. Agri. Food Chem, 2013,
61, 8934-8940.
21
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
[21]
Z. Cabrera, G. Fernández-Lorente, R. Fernandez-Lafuente, J.M. Palomo,
J.M. Guisan, J.Mol Cat B: Enzym 2009,57, 171-176.
[22]
J. Brabcova, J. Blazek, M. Krecmerova, J. Vondrášek, J.M. Palomo, M.
Zarevucka, Molecules.2016, 21, 648; doi:10.3390/molecules21050648.
[23]
B. Chen, J.Hu, E. M. Miller, W. Xie, M. Cai, R. A. Gross,
Biomacromolecules 2008, 9, 463-471.
[24]
O. Romero, M. Filice, B. de las Rivas, C. Carrasco-Lopez, J. Klett, A.
Morreale, J.A. Hermoso, J.M. Guisan, O. Abian, J.M. Palomo,
Chem.Commun.2012, 72, 9053-9055.
[25]
T. Bavaro, S. Rocchietti, D. Ubiali, M. Filice, M. Terreni, M. Pregnolato,
Eur. J. Org. Chem. 2009, 12, 1967-1975.
22
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
Captions
Scheme 1. Chemobiocatalytic synthesis of glucose building blocks. (a) Biocatalyst,
20% acetonitrile/80% buffer at pH 4.8, 25ºC. (b) Biocatalyst, 20% acetonitrile/80%
buffer at pH 4.8, 25ºC and after complete conversion pH changed from 4.8 to 9. (c)
tricloroacetonitrile, DBU, 0ºC, CH2Cl2.
Scheme 2. Biocatalytic regioselective preparation of monodeprotected thymidine.
(a) Biocatalyst, 5% acetonitrile/95% acetate buffer at pH 5, 25ºC.
Scheme 3. Biocatalytic production of monodeprotected lactal. (a) Biocatalyst, 20%
acetonitrile/80% buffer at pH 5, 25ºC.
Figure 1. Stability of different CRL preparations in the presence of acetonitrile
50% (v/v) as co-solvent. A. Lew-CRL (triangles), Lew-CRL-EDC10mM (squares),
Lew-CRL-EDC50mM (rhombus). B. Octyl-CRL (triangles), octyl-CRL-EDC10mM
(squares), octyl-CRL-EDC50mM (rhombus).
Figure 2. Stability of different CRL preparations in the presence of acetonitrile
20% (v/v) as co-solvent. A. Lew-CRL (triangles), Lew-CRL-EDC10mM (squares),
Lew-CRL-EDC50mM (rhombus). B. Octyl-CRL (triangles), octyl-CRL-EDC10mM
(squares), octyl-CRL-EDC50mM (rhombus).
Figure 3. Stability of different RML preparations in the presence of acetone as co-
solvent. A. 25% ; B. 50% (v/v). Octyl-RML (rhombus), octyl-RML-EDC10mM
(squeares), octyl-RML-EDC25mM (triangles), octyl-RML-EDC50mM (asterisk).
Figure 4. FT-IR spectra of octyl-Sepharose.
23
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
Figure 5. A. Structure of CRL. B. Adsorption of the open form of CRL on the
octyl-Sepharose matrix. Lid oligopeptide (light blue), є-NH2 Lys (red), hydrophobic
residues (orange), serine active site (green). The structure of CRL was obtained from
the Protein Data Bank (pdb code: 1CRL) and the picture were created using Pymol v.
0.99
Figure 6. Comparison in reuse of octyl-CRL and octyl-CRL-EDC10mM in the
hydrolysis of 1. Octyl-CRL (blue), octyl-CRL-EDC10mM (red).
24
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
NH
OH
O
CCl3
a
AcO
AcO
AcO
AcO
O
O
c
AcO
AcO
OAc
OAc
2
3
OAc
AcO
AcO
O
AcO
OAc
1
OAc
OAc
AcO
AcO
HO
O
b
O
c
HN
O
AcO
AcO
CCl3
OAc
OAc
4
5
Scheme 1. Chemobiocatalytic synthesis of glucose building blocks. (a) biocatalyst, 20%
acetonitrile/80% buffer at pH 4.8, 25ºC. (b) biocatalyst, 20% acetonitrile/80% buffer at pH 4.8, 25ºC and
after complete conversion pH changed from 4.8 to 9. (c) tricloroacetonitrile, DBU, 0ºC, CH2Cl2.
25
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
O
N
O
N
O
N
NH
O
NH
NH
AcO
HO
AcO
O
O
O
O
O
a
+
OH
OAc
OAc
6
8
7
O
NH
N
O
HO
O
N3
AZT
Scheme 2. Biocatalytic regioselective preparation of monodeprotected thymidine. (a)
biocatalyst, 5% acetonitrile/95% acetate buffer at pH 5, 25ºC.
26
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
OAc
OAc
OAc
O
OAc
O
OAc
OAc
AcO
AcO
a
O
O
AcO
O
O
OAc
OAc
HO
10
9
Scheme 3. Biocatalytic production of monodeprotected lactal. (a) biocatalyst, 20% acetonitrile/80%
buffer at pH 5, 25ºC.
27
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
A
100
75
50
25
0
0
1
2
3
4
5
Time (h)
B
100
75
50
25
0
0
1
2
3
4
5
Time (h)
Figure 1. Stability of different CRL preparations in the presence of acetonitrile 50%
(v/v) as co-solvent. A. Lew-CRL (triangles), Lew-CRL-EDC10mM (squares), Lew-
CRL-EDC50mM (rhombus). B. Octyl-CRL (triangles), octyl-CRL-EDC10mM
(squares), octyl-CRL-EDC50mM (rhombus).
28
This article is protected by copyright. All rights reserved.

10.1002/cctc.201601632
ChemCatChem
A
100
75
50
25
0
0
1
2
3
4
5
Time (h)
B
100
75
50
25
0
0
10
20
30
40
50
Time (h)
Figure 2. Stability of different CRL preparations in the presence of acetonitrile 20%
(v/v) as co-solvent. A. Lew-CRL (triangles), Lew-CRL-EDC10mM (squares), Lew-
CRL-EDC50mM (rhombus). B. Octyl-CRL (triangles), octyl-CRL-EDC10mM
(squares), octyl-CRL-EDC50mM (rhombus).
29
This article is protected by copyright. All rights reserved.