Enzymatic synthesis of 3-aminopropyl-1- O -β- D -galactopyranoside catalyzed by Aspergillus oryzae β-galactosidase

Article abstract of DOI:10.3109/10242422.2015.1095678

Glycosidases represent excellent green chemistry alternatives as catalysts for the synthesis of glycosides, and in particular their stereoselectivity allows the production of anomerically pure glycosides, in only one reaction step using mild reaction conditions. Here, we report the enzymatic synthesis and structural characterization of 3-aminopropyl-1-O-β-D-galactopyranoside. Optimal reaction conditions for the transgalactosylation reaction were 100 mM lactose, 500 mM 3-amino-1-propanol and 24 h of incubation at 50 °C with 6 U/mL of β-galactosidase from Aspergillus oryzae. The fact that the synthesis of 1-propyl-2-O-β-D-galactopyranoside using 1-amino-2-propanol as acceptor was not achieved, and that N-glycoside formation was not observed, confirms the selectivity of β-galactosidase for the synthesis of O-glycosides, and particularly for primary alcohols. The synthesized galactosides were evaluated for their ability to interact with bovine spleen galectin-1 (Gal-1) by using the hemagglutination inhibition assay; results demonstrated that 3-aminopropyl-1-O-β-D-galactopyranoside may be considered as a functionalized galactose moiety more than an efficient Gal-1 inhibitor. The proposed approach constitutes a promising tool for the generation of glycopolymers and glyconanoparticles with potential applications in the development of biosensors as well as construction blocks in chemical synthesis.

Full text of DOI:10.3109/10242422.2015.1095678

Biocatalysis and Biotransformation
ISSN: 1024-2422 (Print) 1029-2446 (Online) Journal homepage: http://www.tandfonline.com/loi/ibab20
Enzymatic synthesis of 3-aminopropyl-1-O-β-D-
galactopyranoside catalyzed by Aspergillus oryzae
β-galactosidase
Cecilia Porciúncula González, Ernesto Rodríguez, Silvia Soule, Laura Franco
Fraguas, Beatriz M. Brena, Cecilia Giacomini & Gabriela Irazoqui
To cite this article: Cecilia Porciúncula González, Ernesto Rodríguez, Silvia Soule, Laura Franco
Fraguas, Beatriz M. Brena, Cecilia Giacomini & Gabriela Irazoqui (2015) Enzymatic synthesis
of 3-aminopropyl-1-O-β-D-galactopyranoside catalyzed by Aspergillus oryzae β-galactosidase,
Biocatalysis and Biotransformation, 33:4, 197-207, DOI: 10.3109/10242422.2015.1095678
To link to this article: http://dx.doi.org/10.3109/10242422.2015.1095678
Published online: 18 Nov 2015.
Submit your article to this journal
Article views: 8
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=ibab20
Download by: [University of Nebraska, Lincoln]
Date: 08 December 2015, At: 07:05

Biocatalysis and Biotransformation, 2015; 33(4):197–207
RESEARCH ARTICLE
Enzymatic synthesis of 3-aminopropyl-1-O-b-D-galactopyranoside
catalyzed by Aspergillus oryzae b-galactosidase
1
1
2
´
´
´
CECILIA PORCIUNCULA GONZALEZ , ERNESTO RODRIGUEZ , SILVIA SOULE ,
LAURA FRANCO FRAGUAS1, BEATRIZ M. BRENA1, CECILIA GIACOMINI1
& GABRIELA IRAZOQUI1
1
´
´
´
Catedra de Bioquımica, Dpto. Biociencias, Facultad de Quımica, UdelaR, Gral Flores 2124, Montevideo,
2
´
Uruguay and Laboratorio de Carbohidratos, DQO, Facultad de Ciencias/Facultad de Quımica, UdelaR, Alfredo
Navarro 3051, Montevideo, Uruguay
Abstract
Glycosidases represent excellent green chemistry alternatives as catalysts for the synthesis of glycosides, and in particular
their stereoselectivity allows the production of anomerically pure glycosides, in only one reaction step using mild reaction
conditions. Here, we report the enzymatic synthesis and structural characterization of 3-aminopropyl-1-O-b-D-
galactopyranoside. Optimal reaction conditions for the transgalactosylation reaction were 100 mM lactose, 500 mM
3-amino-1-propanol and 24 h of incubation at 50 ꢀC with 6 U/mL of b-galactosidase from Aspergillus oryzae. The fact that the
synthesis of 1-propyl-2-O-b-D-galactopyranoside using 1-amino-2-propanol as acceptor was not achieved, and that
N-glycoside formation was not observed, confirms the selectivity of b-galactosidase for the synthesis of O-glycosides, and
particularly for primary alcohols. The synthesized galactosides were evaluated for their ability to interact with bovine spleen
galectin-1 (Gal-1) by using the hemagglutination inhibition assay; results demonstrated that 3-aminopropyl-1-O-b-D-
galactopyranoside may be considered as a functionalized galactose moiety more than an efficient Gal-1 inhibitor. The
proposed approach constitutes a promising tool for the generation of glycopolymers and glyconanoparticles with potential
applications in the development of biosensors as well as construction blocks in chemical synthesis.
Keywords: ꢀ-galactosidase, galactoside, galectins, glycosidases, transglycosylation
Introduction
cells, and its participation in several processes related
to cancer development, including immune suppres-
sion, angiogenesis, hypoxia and metastasis has been
reported (Hasan et al. 2007; Ito et al. 2012). So, the
blocking of the CRD of Gal-1 with glycosides that
compete with its natural ligands may convert Gal-1
into a potential target for cancer therapy (Hasan et al.
The study of carbohydrate–lectin interactions is one
of the most promising research areas for unveiling
the mechanism of biological processes mediated by
glycans, for which the availability of glycosides and
oligosaccharide of defined structure is essential
(Ito et al. 2012; Ogiso et al. 2013; Sunasee and
Narain 2013).
`
2007; Giguere et al. 2008; Ito et al. 2012). In this
relation, several glycosides capable of inhibiting
binding between galectins and their natural ligands
have been reported (Rabinovich 2005; Ingrassia et al.
Lectins are proteins that possess a carbohydrate
recognition domain (CRD) with the ability to inter-
act specifically and reversibly with glycoconjugates.
Those with specificity towards b-galactosyl contain-
ing glycoconjugates are known as galectins (Varki
et al. 1999, Rabinovich 2005; Camby et al. 2006;
Boscher et al. 2011). Particularly galectin-1 (Gal-1)
is abundantly secreted by almost all malignant tumor
`
2006; Giguere et al. 2006, 2008; Hasan et al. 2007;
Ito et al. 2012).
Glycosidases are excellent alternatives as catalysts
for synthetic purposes as their stereoselectivity allows
the production of anomerically pure glycosides,
´ ´ ´
Correspondence: Gabriela Irazoqui, Catedra de Bioquımica, Dpto. Biociencias, Facultad de Quımica, UdelaR, Gral. Flores 2124, CC 1157, Montevideo,
Uruguay. Tel: +598 29241806. E-mail: mgidrv@fq.edu.uy
(Received 4 March 2015; accepted 14 September 2015)
ISSN 1024-2422 print/ISSN 1029-2446 online ß 2015 Taylor & Francis
DOI: 10.3109/10242422.2015.1095678

´
´
198 C. Porciuncula Gonzalez et al.
in only one reaction step using mild reaction
conditions (Ichikawa et al. 1992; Van Rantwijk
Enzymatic activity assay
Aliquots of 100 mL of suitably diluted A. oryzae
b-galactosidase solution were added to 2.0 mL of
25 mM ONPG in 50 mM sodium acetate buffer at
pH 5.5 (activity buffer), at room temperature. The
rate of formation of free o-nitrophenol (ONP) was
recorded spectrophotometrically at 405 nm using a
1 cm path length. One enzyme unit (U) was defined
as the amount of enzyme hydrolyzing 1 mmol of
substrate per minute in the above defined conditions.
The extinction coefficient used for ONP at pH 5.5
was 7.5 ꢁ 102 Mꢂ1 cmꢂ1. Enzyme activity was
expressed as U/mL.
´
et al. 1999; Bojarova and Kren 2008). These
enzymes catalyze the transfer of a glycosyl-moiety
from a donor compound (glycoside or oligosacchar-
ide) to a nucleophile that works as an acceptor
molecule.
The potential of transglycosylation reactions
catalyzed by enzymes in synthetic chemistry is
enormous, and has not been fully exploited so far.
However, these systems are complex and critically
dependant on the enzyme specificity, the acceptor
molecule and reaction conditions, making the syn-
thesis of new products a challenging objective.
All the enzymatic assays were performed in
triplicate.
Transgalactosylation
reactions
catalyzed
by
Aspergillus oryzae b-galactosidase have already been
applied to the synthesis of different products, mainly
galactooligosaccharides (GOS) (Albayrak & Yang
2002; Urrutia et al. 2013). In this work, we studied
the transgalactosylation activity of this enzyme using
two aliphatic alcoholamines (3-amino-1-propanol
and 1-amino-2-propanol) as acceptors and lactose
as the galactosyl donor. The biological activity as
galectin inhibitor of the synthesized galactosides was
evaluated using red blood cells hemagglutination
assay. A thorough structural analysis of the obtained
products was performed by spectrometric and spec-
troscopy methods.
Protein determination assay
Protein was determined by the bicinchoninic acid
(BCA) assay (Smith et al. 1985; Giacomini et al.
1998). All the protein assays were preformed in
triplicate.
Enzymatic synthesis of galactosides
To volumes of 10 mL of reaction mixtures contain-
ing lactose (100–347 mM) and 3-amino-1-propanol
hydrochloride or 1-amino-2-propanol hydrochloride
(100–500 mM) in 0.5 M sodium acetate buffer at pH
5.5, soluble A. oryzae b-galactosidase was added to a
final concentration of 6 U/mL. The reaction mix-
tures were incubated at 50 ꢀC and gently stirred.
Aliquots of the mixture were taken at regular
intervals and the reaction was stopped by heating at
100 ꢀC for 5 min. The samples were analyzed by thin
layer chromatography (TLC).
Experimental section
Materials
o-Nitrophenyl-b-D-galactopyranoside (ONPG), lac-
tose, galactose, 3-amino-1-propanol, orcinol,
b-galactosidase (b-D-galactoside galactohydrolase;
EC 3.2.1.23) from A. oryzae, fetuin from calf fetal
serum, 1-cyano-4- dimethylamino pyridinium tetra-
fluoroborate (CDAP-BF4), triethylamine and
DEAE-cellulose were purchased from Sigma-
Aldrich (St. Louis, MO). Chloroform, 2-amino-1-
propanol and ninhydrin were purchased from Merck
(Hohenbrunn, Germany). TLC plates were pur-
chased from Machery Nagel (Duren, Germany).
Sepharose 4B and PD-10 (Sephadex G25) columns
All the experiments were performed in triplicate.
TLC analysis
Analytical TLC was performed on 20 cm ꢁ 10 cm
ꢁ 0.25 mm precoated Alugram Silica Gel 60/UV254
indicator TLC plates. Aliquots of 5 mL of samples
(diluted 1:4) and standards (glucose, galactose and
lactose) were spotted onto the TLC plates and
developed to 10 cm in an 22.0 cm ꢁ 11.0 cm ꢁ 6.2
cm chamber (saturation time 30 min), using
MeOH:CHCl3:C3H6O:NH4OH (42:17:25:17) as
mobile phase. The TLC plates were air-dried,
sprayed with orcinol 0.2% (w/v) in EtOH:H2SO4
(90:10), then heated for 2 min at 110 ꢀC or sprayed
with ninhydrin spray reagent (0.2% (w/v) in
1-butanol:10% acetic acid (95:5) for free amino
were
purchased
from
GE
Health
Care
(Buckinghamshire, UK). BCA protein assay reagents
were purchased from Pierce (Rockford, IL).
Heparinized rabbit blood was supplied by the
Unidad de Reactivos y Biomodelos Experimentales,
Facultad de Medicina, Universidad de la Repu´blica.
´
Bovine Spleen was donated by Frigorıfico Carrasco,
Uruguay. All other chemicals used were of analytical
grade.
´
group detection (Garofalo et al. 2011). All the
experiments were performed in triplicate.

Enzymatic synthesis of 3-aminopropyl-1-O-ꢀ-D-galactopyranoside 199
Purification of the compounds
position in each structure was determined by detect-
ing interring cross-peaks in HMBC spectrum.
Synthesized galactosides were purified in two steps.
The first one consisted of size exclusion chromatog-
raphy in a Sephadex G10 column (1.6 cm id ꢁ 83 cm
height) and isocratically eluted with deionized water
at a flow rate of 0.06 mL/min (1.8 cm hꢂ1 linear flow
rate). Fractions were collected, lyophilized and
analyzed by TLC as described previously. The
second purification step consisted in a preparative
TLC performed in 20 cm ꢁ 10 cm ꢁ 0.25 mm pre-
coated with Alugram Silica Gel 60/UV254 indicator
TLC plates in the same conditions as the analytical
TLC. In order to identify the area of the plate
containing the compound of interest, the edges of the
plate were stained with orcinol. The silica from the
zone containing the compound of interest was
removed and the compound was extracted by son-
ication during 10 min in methanol:water (20:80 v/v)
solution. Then the suspension was centrifuged for
10 min at 8700g. The extraction procedure was
repeated three times. The supernatants were
pooled, the methanol was rotary evaporated and
the sample was further lyophilized and the purity was
checked by analytical TLC as described previously.
Determination of kinetic parameters
Kinetic parameters were determined using ONPG as
galactose donor (1.0–15.0 mM) in the presence of 3-
amino-1-propanol (100–1000 mM) in activity buffer
at 25 ꢀC. The rate of ONP formation was determined
as previously described. Results were analyzed using
Hanes linearization (Cornish-Bowden 1995).
All the activity assays were performed in triplicate.
Purification of Gal-1
Gal-1 was purified from bovine spleen as previously
reported by Ahmed et al. (1996) with the following
modifications. The bovine spleen membrane was
removed and the organ was cut into small pieces and
suspended in 75 mM sodium phosphate buffer pH
7.2 containing 75 mM NaCl, 2 mM EDTA, 4 mM
mercaptoethanol, 0.3M lactose (buffer B) in a ratio of
2 mL/g of spleen. The mixture was homogenized at
4 ꢀC and then was centrifuged at 26 000g for 60 min at
4 ꢀC. Aliquots of 30 mL of clarified extract twice
diluted in buffer were incubated with 30 g of DEAE-
cellulose previously swollen in 75 mM sodium phos-
phate buffer pH 7.2 containing 75 mM NaCl, 2 mM
EDTA, 4 mM mercaptoethanol (buffer A) under
mild stirring for 4 h at room temperature. The non-
bound material of the ionic exchange was gel filtered
in PD-10 columns pre-equilibrated in buffer A. The
void volume fraction containing lectin activity was
applied to the asialofetuin (ASF)–sepharose affinity
column, previously equilibrated in buffer A, at a flow
rate of 0.25 mL/min. The column was washed with
buffer A until the A280 of the eluate was lower than
0.05, then Gal-1 was eluted with buffer B. The ASF–
sepharose was synthesized in our laboratory as previ-
MS analysis
ESI-MS analyses were performed using a Thermo
LTQ Velos spectrometer equipped with a linear
detector. The electrospray-ionization ion source was
operated in the positive-ion mode. The capillary
conditions were: voltage 5 kV and temperature
275 ꢀC. The sample was diluted in a 30% acetonitrile
(v/v), 0.1% formic acid (v/v) aqueous solution and
introduced in direct injection mode at 4 mL/min.
NMR experiments
´
ously described by Pla et al. (2003).
The chemical structures of synthesized galactosides
1
were identified by an extensive H and 13C NMR
Hemagglutination assay (HAG)
study. NMR spectra were recorded at 30 ꢀC at 400
and 100 MHz, respectively, using deuterated metha-
nol (CD3OD) as solvent on a Bruker Avance DPX
400 NMR. 1H and 13C chemical shifts were
expressed in ppm using TMS (ꢁ ¼ 0.00) and
Me2CO (ꢁ ¼ 31.00) as internal reference. 2D experi-
ments [correlation spectroscopy (COSY), hetero-
nuclear single-quantum coherence (HSQC) and
heteronuclear multiple-bond correlation (HMBC)]
were carried out according to the standard pulse
sequences of the instrument software. Proton signals
were assigned based on COSY spectra. 13C signals
were assigned with HSQC spectra, based on the
assignment of proton signals. Finally, the linkage
Gal-1 activity was determined by the HAG assay
using rabbit red cells and estimated by the twofold
serial dilution assay (Franco Fraguas et al. 2003).
Erythrocytes were prepared from fresh blood col-
lected in heparin and washed four times with sodium
phosphate buffer 50 mM pH 7.4, containing 0.15 M
NaCl (PBS buffer) by centrifugation during 3 min at
1500g. Glutaraldehyde-fixed and trypsin-treated red
cells were prepared as described by Nowak et al.
(1976) and diluted to give a 4% suspension in PBS.
In the experiment, 25 ml NaCl 0.15 M, 25 ml 1% BSA
in 0.15 M NaCl, 25 ml Gal-1 and 25 ml 4% suspen-
sion of the glutaraldehyde-fixed and trypsin-treated

´
´
200 C. Porciuncula Gonzalez et al.
R
f
(A)
0.40
0.33
0.25
0.18
0.07
1
2
3
4
5
6
7
8
9
10
11
12
(B)
(C)
R
f
0.40
0.33
0.25
0.18
0.07
1
2
3
4
5
1
2
3
4
Figure 1. Time course reaction using lactose 200 mM, 6 U/mL at 50ꢀ C and pH 5.5, and different 3-amino-1-propanol concentrations. (A)
3-Amino-1-propanol 500 mM; Lanes: 1 – lactose standard (50 mM), 2 – 0 h, 3 – 0.25 h, 4 – 0.5 h, 5 – 1 h, 6 – 2 h, 7 – 3 h, 8 – 4 h, 9 – 5 h, 10 –
24 h, 11 – glucose standard (50 mM), 12 – galactose standard (50 mM). (B) 3-Amino-1-propanol 100 mM; Lanes: 1 – 0 h; 2 – 0.5 h; 3 – 1 h;
4 – 3 h; 5 – 24 h. (C) 3-Amino-1-propanol 250 mM; Lanes: 1 – 0.5 h; 2 – 1 h; 3 – 3 h; 4 – 24 h.
rabbit red cells were homogenized and incubated in a
U-shaped microtiter plate at room temperature
during 30 min and examined for agglutination.
Titer per mL was defined as the reciprocal of the
highest dilution giving visible agglutination.
Figure 1(A). As expected, the release of glucose
(Rf ¼ 0.40) by hydrolysis is proportional to the
decrease of lactose (Rf ¼ 0.18). However, during
the first five hours of reaction, the concentration of
galactose released (Rf ¼ 0.33) was notoriously less
pronounced than that of glucose. This is an indica-
tion of the coexistence of transgalactosylation and
hydrolysis reaction, as glucose is released in the first
reaction step of both processes, while galactose is
released during the second reaction step, only when
water is the galactosyl acceptor (Scheme 1). On the
other hand, two new spots with Rf of 0.25 (com-
pound 1) and 0.07 (compound 2) were observed and
they could not be assigned neither to the substrate
nor to lactose hydrolysis products, inferring that they
may correspond to transgalactosylation products. It
should be highlighted that upon 24 h of reaction the
intensity of the spot corresponding to compound 2
notoriously decreases, while that of compound 1
remains unchanged. This could point to enzymatic
hydrolysis of compound 2; on the other hand
compound 1 either could not be hydrolyzed by the
enzyme, or steady state between its synthesis rate and
its hydrolysis rate was achieved. In order to detect
the presence of free amino groups in the new
compounds, the TLC was developed with ninhydrin
(specific developer for primary amine groups). Only
compound 1 (Rf ¼ 0.25) and 3-amino-1-propanol
(Rf ¼ 0.67) were revealed by ninhydrin (data not
shown). This was the first evidence that compound 1
could be preliminarily identified as the galactoside:
3-aminopropyl-1-O-b-D-galactopyranoside.
All the experiments were performed in triplicate.
HAG inhibition assay
The ability of the synthesized galactosides to interact
with Gal-1 was determined by the HAG inhibition
assay. The lectin dilution used for the end point was
the highest dilution causing 50% HAG (dilution
before the previous causing visible HAG) defined
as HAG50% (Franco Fraguas et al. 2003).
The minimum inhibitory concentration (MIC) of
the galactoside that inhibits the HAG50% was
determined by using a serial twofold dilution,
starting from 100 mM galactoside, in the HAG
inhibition assay. Relative inhibitory power (RIP)
was defined as the ratio between the galactose MIC
and the galactoside MIC.
All the experiments were performed in triplicate.
Results and discussion
Synthesis of 3-aminopropyl-1-O-ꢀ-D-
galactopyranoside
The time course of the enzymatic reaction using
lactose (200 mM), 3-amino-1-propanol (500 mM)
and 6 U/mL of A. oryzae b-galactosidase in 0.5 M
sodium acetate buffer pH 5.5 at 50 ꢀC is shown in

Enzymatic synthesis of 3-aminopropyl-1-O-ꢀ-D-galactopyranoside 201

´
´
202 C. Porciuncula Gonzalez et al.
(A)
R
f
0.40
0.33
0.25
0.18
0.07
1
2
3
4
5
6
7
8
9
10
11
12
R
(B)
(C)
f
0.40
0.33
0.18
0.07
1
2
3
4
1
2
3
4
5
Figure 2. Time course reactions using different lactose concentrations, 6 U/mL at 50ꢀ C, pH 5.5, in the absence of 3-amino-1-propanol. (A)
Lactose 347 mM, Lanes: 1 – lactose standard (50 mM), 2–2 h of reaction synthesis with 500 mM 3-amino-1-propanol (as control), 3 – 0 h, 4
– 0.25 h, 5 – 0.5 h, 6 – 1 h, 7 – 2 h, 8 – 3 h, 9 – 4 h, 10 – 24 h, 11 – glucose standard (50 mM), 12 – galactose standard (50 mM). (B) Lactose
200 mM: Lanes: 1 – 0 h, 2 – 0.25 h, 3 – 0.5 h, 4 – 1 h, 5 – 24h. (C) Lactose 100 mM: Lanes: 1 – 0 h, 2 – 0.25 h, 3 – 0.5 h, 4 – 1 h, 5 – 24 h.
Synthesis reaction in the absence of the alcoholamine
Study of the optimal reaction conditions for the
synthesis of 3-aminopropyl-1-O-ꢀ-D-
galactopyranoside
The transgalactosylation reaction was evaluated in
the absence of 3-amino-1-propanol and for this
purpose the enzyme (6 U/mL) was incubated with
different lactose concentrations (100, 200 and
347 mM) in 50 mM sodium acetate buffer pH 5.5
at 50 ꢀC for 24 h. Spots corresponding to hydrolysis
products, glucose and galactose, as well as to
compound 2 could be observed for all the lactose
concentrations studied (Figure 2).
The transgalactosylation reaction for the synthesis of
3-aminopropyl-1-O-b-D-galactopyranoside was eval-
uated under different conditions. The synthesis of
both, compounds 1 (Rf ¼ 0.25) and 2 (Rf ¼ 0.07),
was achieved at all lactose concentrations studied
(100–347 mM) at a constant 3-amino-1-propanol
concentration (500 mM) (data not shown). These
results showed that the lactose concentration was not
a critical parameter for the galactoside synthesis, in
the concentration range studied. Nevertheless, an
increase in 3-amino-1-propanol concentration from
100 to 500 mM at a fixed lactose concentration
(200 mM) resulted in an increase in the amount of 3-
aminopropyl-1-O-b-D-galactopyranoside, shown by
the higher intensity of the TLC spots with Rf ¼ 0.25
(Figure 1). This performance shows that high con-
centrations of 3-amino-1-propanol favor transglyco-
sylation over hydrolysis.
The fact that compound 2 (Rf ¼ 0.07) could be
synthesized in the absence of 3-amino-1-propanol
denotes that it could be the result of a transgly-
cosylation reaction where lactose fulfilled the
double role of galactosyl donor and acceptor,
resulting in the formation of
a trisaccharide
(Scheme 1). Interestingly, this transglycosylation
reaction took place in aqueous medium even at
low lactose concentrations (Figure 2C). On the
other hand, the fact that compound 1 was not
generated in the absence of 3-amino-1-propanol,
reinforced the hypothesis that it could be the
3-aminopropyl-1-O-b-D-galactopyranoside. A point
to be considered is that total lactose conversion
was achieved after 24 h of reaction, when initial
lactose concentrations were 100 and 200 mM
(Figure 2C and 2B), and almost completely
achieved when lactose concentration was 347 mM
(Figure 2A).
The comparison between the reaction synthesis at
24 h in the absence of acceptor (Figure 2), and in the
presence of all the 3-amino-1-propanol concentra-
tions studied (Figure 1), at the same lactose con-
centration (200 mM), clearly showed that lactose
conversion was lower in the latter cases. In order to
evaluate if 3-amino-1-propanol was an enzyme
inhibitor, kinetic parameters in the presence

Enzymatic synthesis of 3-aminopropyl-1-O-ꢀ-D-galactopyranoside 203
of different concentrations of the aminoalcohol
(A)
(B)
Rf
(100–1000 mM) were determined. As the
enzyme kinetic parameters (KM ¼ 2.6 ꢃ 0.2 mM,
Kcat ¼ (5.7 ꢃ 0.5) ꢁ 103 minꢂ1) did not change sig-
nificantly in the presence of 3-amino-1-propanol
(data not shown) enzyme inhibition was discarded as
the cause of lactose conversion slowing-down.
On other hand, since b-galactosidase from
A. oryzae showed to be relatively stable in the most
drastic conditions assayed (50 ꢀC, 500 mM 3-amino-
1-propanol) keeping near to 50% of residual enzyme
activity upon 24 h, this could be the cause of the
decreased rate of the lactose conversion. Moreover,
given that the enzyme was completely stable at 50 ꢀC
during 24 h in the absence of the acceptor, this
decrease may be partially attributed to the presence
of the 3-amino-propanol in the bulk reaction.
0.40
0.33
0.25
0.18
0,07
1
2
3
4
5
6
7
Figure 3. Thin layer chromatography of purified compounds 1
and 2. (A) Lanes: 1 – purified compound 2; 2 – glucose standard
(50 mM); 3 – galactose standard (50 mM); 4 – lactose standard
(50 mM); 5 – 2 h of reaction synthesis with 347 mM lactose, 500
mM 3-amino-1-propanol, 50ꢀC, 6 U/mL; 6 – purified compound
1. All lines developed with orcinol. (B) Lane: 7 – purified
compound 1 developed with ninhydrin.
corresponding carbon signal (Table 1). This infor-
3
mation, together with the values of JH1,H2 for the
Despite this enzyme inactivation, 500 mM of
3-amino-1-propanol was selected as the optimal
condition as it was the one that provided the highest
production of galactoside. Taking into account that
the galactoside purification process was easier at low
residual lactose concentration, and considering that
initial lactose concentration was a parameter that had
low significance over the amount of the synthesized
galactoside, the optimal conditions selected for the
synthesis were 100 mM lactose, 500 mM 3-amino-1-
propanol, 6 U/mL, 50 ꢀC, 24 h.
anomeric proton and published chemical shift data
for sugar residues allowed the assignment of the spin
system for a specific sugar residue and the deter-
mination of the anomeric configuration. The data
showed that the galactose residue was a b-pyranoside
(Jansson et al. 1989; Agrawal 1992).
1
The H NMR spectrum of compound 1 showed,
in addition to the signals of the monosaccharide
residue, signals corresponding to the 3-amino-1-
propanol C2 protons bonded to the amino terminal
through a methylene carbon at ꢁH ¼ 1.87 ppm (m),
and two important signals at ꢁH ¼ 3.63 ppm (t) and
ꢁH ¼ 3.07 ppm (t) corresponding to two methylene
carbons.
Galactoside purification and structural
characterization
The purification of the galactoside was performed in
two chromatographic steps. The first one by size
exclusion chromatography using a Sephadex G10
column; here compound 2 was isolated while com-
pound 1 was co-eluted with lactose. In the second
step, compound 1 was purified by preparative TLC.
The purity of both compounds was analyzed by TLC
(Figure 3).
Moreover the 13C NMR spectrum showed signals
for nine carbons, three of which arose from the
aglycon moiety. Signals observed in the HSQC
spectrum indicated the presence of four methylene
and five methine carbons, which confirmed the
presence of the aminopropyl aglycon and a sugar
residue. The proton spin of the pyranoside ring was
assigned by H–H correlation COSY. These data
together with the proton–carbon correlation from
the HSQC experiment allowed the assignment of the
carbon signals. The absence of a correlation of the
proton signal at ꢁH ¼ 2.5 with a carbon signal
confirmed that it corresponds to the amino protons.
The glycosidic linkage between both molecules was
evidenced by the strong correlation observed among
the H1 proton ꢁH ¼ 3.63 and the anomeric carbon
(ꢁC ¼ 102.8) by the HBMC experiment.
Structural elucidation of the purified compounds
1 and 2 was performed by combining the use of
mono (1H NMR, 13C NMR) and bidimensional
(COSY, HSQC, HMBC) techniques of nuclear
magnetic resonance as well as mass spectrometry.
Structural characterization of 3-aminopropyl-1-O-
ꢀ-D-galactopyranoside
NMR analysis
1
The H NMR spectrum of compound 1 in water-d2
MS analysis
at 30 ꢀC showed only one anomeric proton at
ꢁH ¼ 4.32 ppm (J ¼ 8 Hz), confirming the presence
of a monosaccharide moiety. Proton decoupled
HSQC experiments allowed the assignment of the
Compound 1 ESI-MS spectrum revealed the pres-
ence of intense ions [M + H]+ and [M + Na]+ at m/z
238.12 and 260.13, respectively, which are in
accordance to the molecular weight of the proposed

´
´
204 C. Porciuncula Gonzalez et al.
Table 1. 1H and 13C NMR Chemical Shift Values of Compound 1, b-D-galactose and methyl-1-O-b-D-galactopyranoside.
Compound 1
Galactose
40
3-amino-1-Propanol
10
20
30
50
3.58 d
60
3.72
600
3.95 m
1
2
3
4.32 d
(8.0)a
102.80
3.42 dd
(8.0, 8.0)a
70.71
3.60 dd
(3.2, 3.6)a
72.67
3.83 d
(3.2)a
68.59
3.63 t
60.98
1.87 m
26.64
3.07 t
37.64
75.17
67.81
b-D-galactoseb
10
4.53
20
3.45
72.9
30
3.59
73.8
40
3.89
69.7
50
3.65
75.9
60
3.64
61.8
600
3.72
97.4
Methyl-1-O-b-D-galactopyranosideb
10
4.20
20
3.39
71.7
30
3.53
73.8
40
3.81
69.7
50
3.57
76.0
60
3.69
62.0
600
3.64
104.5
Multiplicity: d-doublet; dd- double doublet; t-triplet; m-multiplet.
aJ(Hz) in brackets.
bChemical shifts reported by Agrawal, 1992.
Figure 4. 3-Aminopropyl-1-O-b-D-galactopyranoside MS-analysis, (A) ion m/z ¼ 238 MS/MS spectrum; (B) ion m/z ¼ 260 MS/MS
spectrum.
galactoside. According to the Domon and Costello
(1988) nomenclature, we analyzed the MS–MS
spectrum of the [M + H]+ shown in Figure 4(A).
It can be observed that the ion m/z ¼ 76.10 (Y1)
generated due to the release of the galactose moiety
lacking the oxygen atom of the glycosidic bond

Enzymatic synthesis of 3-aminopropyl-1-O-ꢀ-D-galactopyranoside 205
Figure 5. b-D-galactopyranosyl-(1-4)-b-D-galactopyranosyl-(1-4)-a-D-glucose MS-analysis. (A) Compound 2 MS; (B) compound 2 ion
m/z ¼ 527 MS/MS.
corresponds to the molecular weight of the 3-amino-
1-propanol. The fragments m/z ¼ 202.09 (C1) and
m/z ¼ 220.13 were also identified. The latter was
assigned to the [M ꢂ NH3]+ ion which evidenced the
loss of the terminal amino as an ammonium mol-
ecule according to a polyfunctional compound
(McLafferty 1980). The [M + Na]+/MS–MS spec-
trum (Figure 4B) showed that the ions m/z ¼ 242.16
arose as a consequence of the loss of a water
molecule [M + Na ꢂ H2O]. The fragments assigned
to the split of the pyranoside ring 2.4A1
(m/z ¼ 83.08), 0.2A1 (m/z ¼ 143.0) and 0.2X1
(m/z ¼ 140.09) confirmed the presence of a galactose
moiety (McLafferty 1980; Hofmeister et al. 1991).
Finally, the glycosidic bond excision was supported
by the ions m/z ¼ 185.06 (B1) and m/z ¼ 202.09
(C1).
MS analysis
Compound 2 ESI-MS spectrum revealed the pres-
ence of an intense ion [M + Na]+ at m/z 527.17,
which is in accordance with the molecular weight of a
trisaccharide (Figure 5A). This was reinforced by the
MS–MS fragmentation pattern of this ion (Figure
5B). The generation of the ions corresponding to the
cleavage of the glycosidic bonds (m/z ¼ 203.00 (C1 o
Y2), m/z ¼ 365.09 (C2 o Y3), m/z ¼ 347.17 (B2),
m/z ¼ 509.09 (B3)), as well as those assigned to the
pyranoside rings internal cleavage (m/z ¼ 305.17
(
0.2A2), m/z ¼ 407.09
(
0.2X3
o
2.4A3) and
m/z ¼ 467.09 (0.2A3
o
2.4X3)), evidenced that the
trisaccharide was formed by a galactose bonded to
lactose moiety trough (1 ! 4) bond. The presence of
ion 0.2An in the absence of the ions 0.3An y 0.4An is
characteristic of (1 ! 4) bonds between monosac-
charide moieties (McLafferty 1980; Hofmeister et al.
Consequently the structure 3-aminopropyl-
was
´
1991; Matamoros Fernandez et al. 2004s).
1-O-b-D-galactopyranoside
compound 1.
deduced
for
NMR analysis
In order to confirm the structure proposed for
compound 2 according to MS–MS experiments we
included a brief analysis by NMR.
Structural characterization of ꢀ-D-galactopyranosyl-
(1-4)-ꢀ-D-galactopyranosyl-(1-4)-ꢂ-D-glucose

´
´
206 C. Porciuncula Gonzalez et al.
1
The H NMR spectrum of compound 2 showed
Table 2. Galectin-1 inhibitory properties of compounds 1 and 2.
MICa (mM)
Compound
RIPb
three signals corresponding to three anomeric pro-
tons at ꢁH ¼ 4.35 ppm (J ¼ 7.6 Hz), ꢁH ¼ 4.53 ppm
(J ¼ 8.0 Hz) and ꢁH ¼ 5.10 ppm (J ¼ 3.6 Hz), this
information is consistent with a trisaccharide struc-
ture assigned to the three anomeric protons. The
coupling constant evidenced the presence of two b
Galactose
Lactose
1
2
50.0
3.1
25.0
6.3
1
16
2
8
aMIC: Minimal inhibition concentration, defined as the minimal
inhibitor concentration that cause hemagglutination inhibition.
bRIP: Relative inhibitor, defined as the quotient between
Galactose MIC and compound MIC.
anomers (J ¼ 7.6, J ¼ 8.0) and one
a anomer
(J ¼ 3.6). The HSQC experiment showed the cor-
responding anomeric carbons at ꢁC ¼ 103.32,
ꢁC ¼ 95.64 and ꢁC ¼ 91.81 ppm, respectively.
This information, together with the values of
3JH1,H2 for each anomerics protons and published
chemical shift data for the sugars residues allowed
the determination of each anomeric configuration.
The data showed that galactose residues were
b-pyranosides forms; instead the glucoside residue
was a-pyranoside (Jansson et al. 1989; Agrawal
1992).
A. oryzae exhibited preference for primary hydroxyl
as previously reported (Woudenberg-van Oosterom
et al. 1998; Irazoqui et al. 2009).
Interaction between galactosides and bovine
spleen Gal-1
The interaction between compounds 1 and 2 and
Gal-1 was evaluated by using the HAG inhibition
assay. This assay is reported as a useful tool for the
study of carbohydrate–lectin interactions. The HAG
inhibition assay was performed as described in the
experimental section, in the presence of the potential
inhibitors; those with affinity for Gal-1 will bind to
its CRD, thus inhibiting the HAG. Minimal inhib-
ition concentration (MIC) and relative inhibitor
power (RIP) were determined (Table 2). The
MIC for 3-aminopropyl-1-O-b-D-galactopyranoside
(compound 1) and b-D-galactopyranosyl-(1 ! 4)-
b-D-galactopyranosyl–(1 ! 4)-a-D-glucopyranoside
(compound 2) was higher than the MIC for galact-
ose, being lower than for the case of lactose, a natural
ligand for Gal-1.
Based on the experimental data, the enzyme
mechanism (Scheme 1) as well as the fact that
galactosidases preserve the anomeric configuration,
we propose that the trisaccharide structure is b-D-
galactopyranosyl-(1 ! 4)-b-D-galactopyranosyl-(1
! 4)-a-D-glucopyranoside. The ability of A. oryzae
b-galactosidase to generate galactooligosaccharides
in aqueous medium with lactose as unique substrate
is well known. Urrutia et al. (2013) reported a
detailed characterization of the GOS synthesized at
40 ꢀC, using 400 g/L lactose and 15 U/mL of
enzyme. In this case, the main trisaccharides
formed were: Gal-b(1 ! 6)-Gal-b(1 ! 4)-Glc,
Gal-b(1 ! 3)-Gal-b(1 ! 4)-Glc and Gal-b(1 !
4)-Gal-b(1 ! 4)-Glc. The comparison with our
results suggested that the presence of high concen-
trations of 3-amino-1-propanol was a determinant
parameter since it led to the formation of a single
trisaccharide identified as Gal-b(1 ! 4)-Gal-b(1 !
4)-aGlc.
Conclusions
Transgalactosylation catalyzed by the A. oryzae
b-galactosidase demonstrated to be useful for the
enzymatic synthesis of 3-aminopropyl-1-O-b-D-
galactopyranoside. Optimal reaction conditions for
this synthesis reaction were achieved with 100 mM
lactose, 500 mM 3-amino-1-propanol and 6 U/mL
incubated during 24 h at 50 ꢀC.
Study of 1-amino-2-propanol as acceptor
The use of 1-amino-2-propanol as acceptor in the
transgalactosylation system was explored using iden-
tical reaction conditions as those previously described
for 3-amino-1-propanol. In this case, only one new
spot with rf 0.07 was observed when the reaction
performance was followed by TLC. This compound
was purified by molecular exclusion chromatography,
analyzed by mass spectrometry and NMR, confirm-
ing it was a trisaccharide with the same chemical
structure as compound 2 previously reported. No
additional spots which pointed to the synthesis of the
1-amino-2-propanol galactoside was observed, rein-
forcing the hypothesis that b-galactosidase from
The synthesis of 1-aminopropyl-2-O-b-D-galacto-
pyranoside was not achieved when using 1-amino-2-
propanol as acceptor and N-glycoside formation was
not observed. These results evidenced the selectivity
of b-galactosidase for the synthesis of O-glycosides as
well as the selectivity of the enzyme from A. oryzae
towards primary alcohols (Woudenberg-van
Oosterom et al. 1998; Irazoqui et al. 2009;
´
Porciu´ncula Gonzalez et al. 2013).

Enzymatic synthesis of 3-aminopropyl-1-O-ꢀ-D-galactopyranoside 207
`
Giguere D, Bonin M-A, Cloutier P, Patnam R, St-Pierre C, Sato
The synthesized galactoside did not show better
inhibitor properties for Gal-1 than the natural
inhibitors, yet the 3-aminopropyl-1-O-b-D-galacto-
pyranoside may be considered as a functionalized
galactose moiety. This is very interesting for the
generation of glycopolymers and glyconanoparticles
for the development of biosensors, as well as for
building blocks in chemical synthesis (Ito et al.
2012; Ogiso et al. 2013; Sunasee and Narain
2013).
S, Roy R. 2008. Synthesis of stable and selective
inhibitors of human galectins-1 and -3. Bioorg Med Chem
16:7811–7823.
Hasan SS, Ashraf GM, Banu N. 2007. Galectins – potential
targets for cancer therapy. Cancer Lett 253:25–33.
Hofmeister GE, Zhou Z, Leary J. 1991. Linkage position
determination in lithium-cationized disaccharides: tandem
mass spectrometry and semiempirical calculations.
Chem Soc 113:5964–5970.
J Am
Ichikawa Y, Look GC, Wong CH. 1992. Enzyme-catalyzed
oligosaccharide synthesis. Anal Biochem 202:215–238.
Ingrassia L, Camby I, Lefranc F, Mathieu V, Nshimyumukiza P,
Darro F, Kiss R. 2006. Anti-galectin compounds as potential
anti-cancer drugs. Curr Med Chem 13:3513–3527.
Irazoqui G, Giacomini C, Batista-Viera F, Brena BM, Cardelle-
Cobas A, Corzo N, Jimeno ML. 2009. Characterization of
galactosyl derivatives obtained by transgalactosylation of lactose
and different polyols using immobilized b-galactosidase from
Aspergillus oryzae. J Agric Food Chem 57:11302–11307.
Ito K, Stannard K, Gabutero E, Clark AM, Neo S-Y, Onturk S,
Blanchard H, Ralph SJ. 2012. Galectin-1 as a potent target for
cancer therapy: role in the tumor microenvironment. Cancer
Metastasis Rev 31:763–778.
Acknowledgements
´
Authors thank to Dra. Rosario Duran (Institute
Pasteur Montevideo, Uruguay) for performing MS
analyses.
Jansson PE, Kenne L, Widmalm G. 1989. Computer-assisted
structural analysis of polysaccharides with an extended version
of CASPER using 1H and 13C-NMR data. Carbohyd Res
188:169–191.
Declaration of interest
The authors report no conflicts of interest. The
authors alone are responsible for the content and
writing of this article.
´
Matamoros Fernandez L, Obel N, Scheller HV, Roepstorff P.
2004. Differentiation of isomeric oligosaccharide structures by
ESI tandem MS and GC–MS. Carbohyd Res 339:655–664.
McLafferty FW. 1980. Interpretation of mass spectra. 3rd ed.
California: University Science Books.
References
Nowak TP, Haywood PL, Barondes SH. 1976. Developmentally
regulated lectin in embryonic chick muscle and a myogenic cell
line. Biochem Biophys Res Commun 68:650–657.
Ogiso M, Kobayashi J, Imai T, Matsuoka K, Itoh M, Imamura T,
Okada T, Miura H, Nishiyama T, Hatanaka K, Minoura N.
2013. Carbohydrate immobilized on a dendrimer-coated col-
loidal gold surface for fabrication of a lectin sensing device
based on localized surface plasmon resonance spectroscopy.
Biosens Bioelectron 41:465–470.
Agrawal PK. 1992. NMR spectroscopy in the structural elucida-
tion of oligosaccharides and glycosides. Phytochemistry
31:3307–3330.
Ahmed H, Fink NE, Pohl J, Vasta, GR. 1996. Galectin-1 from
bovine spleen: biochemical characterization, carbohydrate spe-
cificity and tissue specific isoform profiles.
120:1007–1019.
J Biochem
Albayrak N, Yang ST. 2002. Production of galactooligosacchar-
ides from lactose by Aspergillus oryzae b-galactosidase immobi-
lized on cotton cloth. Biotechnol Bioeng 77:8–19.
Bojarova´ P, Kren V. 2008. Glycosidases: a key to tailored
carbohydrates. Trends Biotechnol 27:199–209.
Boscher C, Dennis JW, Nabi IR. 2011. Glycosylation,
galectins and cellular signaling. Curr Opin Cell Biol
23:383–392.
Camby I, Le Mercier M, Lefranc F, Kiss R. 2006. Galectin-1: a
small protein with major functions. Glycobiology 16:137R–
157R.
Cornish-Bowden A. 1995. Introduction of enzyme kinetics. In:
Cornish-Bowden A, Ed. Fundamentals of enzyme kinetics.
London: Portland Press; 31–32.
´
Pla A, Alonso E, Batista-Viera F, Franco Fraguas L. 2003.
Screening for carbohydrate-binding proteins in extracts of
Uruguayan plants. Braz J Med Biol Res 36:851–860.
Porciu´ncula Gonzalez C, Castilla A, Garofalo L, Soule S, Irazoqui
G, Giacomini C. 2013. Enzymatic synthesis of 2-aminoethyl
b-D-galactopyranoside catalyzed by Aspergillus oryzae b-galac-
tosidase. Carbohyd Res 368:104–110.
Rabinovich GA. 2005. Galectin-1 as a potential cancer target. Br J
Cancer 92:1188–1192.
Smith PK, Khron RI, Hermanson GF, Mallia AK, Gartner FH,
Provenzano MD, Fujimoto EK, Goeke NM, Olson BJ, Klenk
DC. 1985. Measurements of protein using bicinchoninic acid.
Anal Biochem 150:76–85.
´
´
Sunasee R, Narain R. 2013. Glycopolymers and glyconanoparti-
cles in biomolecular recognition processes and vaccine devel-
opment. Macromol Biosci 13:9–27.
Domon B, Costello CE. 1988. A systematic nomenclature for
carbohydrate fragmentations in FAB-MS/MS spectra of glyco-
conjugates. Glycoconjugate J 5:397–409.
Urrutia P, Rodriguez-Colinas B, Fernandez-Arrojo L, Ballesteros
AO, Wilson L, Illanes A, Plou FJ. 2013. Detailed analysis of
galactooligosaccharides synthesis with b-galactosidase from
Aspergillus oryzae. J Agric Food Chem 61:1081–1087.
Varki A, Cummings R, Esko J, Freeze H, Hart G, Marth J. 1999.
S-type lectins (Galectins). In: Varki A, Cummings R, Esko J,
Freeze H, Hart G, Marth J, Eds. Essentials of glycobiology.
New York: Cold Spring Harbor Laboratory Press; 417–427.
Van Rantwijk F, Woudenberg-Van Oosterom M, Sheldon RA.
1999. Glycosidase-catalysed synthesis of alkyl glycosides. J Mol
Catal B: Enzym 6:511–532.
Franco Fraguas L, Batista-Viera F, Carlsson J. 2003. Isolation of a
b-galactoside-binding lectin from cat liver. Braz J Med Biol Res
36:447–457.
´
´
Garofalo L, Vazquez D, Ferreira F, Soule S. 2011. Wheat flour
non-starch polysaccharides and their effect on dough rheo-
logical properties. Ind Crops Prod 34:1327–1331.
Giacomini C, Villarino A, Franco-Fraguas L, Batista-Viera F.
1998. Immobilization of b-galactosidase from Kluyveromyces
lactis on silica and agarose: comparison of different methods.
J Mol Catal B: Enzym 4:313–327.
`
GiguereD, SatoS, St-PierreC, SiroisS, RoyR. 2006. Aryl O- and
Woudenberg-van Oosterom M, Van Belle HJA, van Rantwijk F,
Sheldon RA. 1998. Immobilized b-galactosidases and their use
in galactoside synthesis. J Mol Catal A: Chem 134:267–274.
S-galactosides and lactosides as specific inhibitors of human
galectins-1 and -3: role of electrostatic potential at O-3. Bioorg
Med Chem Lett 16:1668–1672.