Novel synthetic (1 → 6)-α-d-mannodisaccharide substrates support processive mannosylation catalysed by the mycobacterial cell envelope enzyme fraction

Article abstract of DOI:10.1039/c3ra43575j

Three new (1 → 6)-α-d-mannodisaccharides with cyclohexylalkyl or octylsulfonyl function like aglycone were synthesized and screened in the mycobacterial mannosyltransferase assay. 2-Cyclohexylethyl (1 → 6)-α-d-Man2 acted as the best acceptor substrate, wh

Full text of DOI:10.1039/c3ra43575j

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Novel synthetic (1 A 6)-a-D-mannodisaccharide
substrates support processive mannosylation catalysed
Cite this: RSC Advances, 2013, 3,
17784
by the mycobacterial cell envelope enzyme fraction3
ab
c
c
a
´
´
´
´
ˇ
´
Erika Lattova, Zuzana Svetlıkova, Katarına Mikusova, Helene Perreault
b
´
´
and Monika Polakova*
Three new (1 A 6)-a-D-mannodisaccharides with cyclohexylalkyl or octylsulfonyl function like aglycone
were synthesized and screened in the mycobacterial mannosyltransferase assay. 2-Cyclohexylethyl (1 A 6)-
a-D-Man2 acted as the best acceptor substrate, whereas the sulfonyl group significantly reduced the ability
of the mannodisaccharide to serve as the acceptor. Despite these differences, mass spectrometric analysis
confirmed the capability of all synthetic mannodisacharides to accept up to ten additional mannose units,
i.e. the transfer was not affected by the type of aglycone. The results reported here suggest that the
enzyme responsible for the consecutive mannose attachment is the processive a-mannopyranosyltransfer-
ase present in the cell-free system of the mycobacterial cell envelope.
Received 6th May 2013,
Accepted 22nd July 2013
DOI: 10.1039/c3ra43575j
www.rsc.org/advances
However, the first enzyme involved in biosynthesis of LAM/LM
mannan chains, mannopyranosyltransferase Rv2181 responsi-
Introduction
Mycobacteria represent a large group of microorganisms, with
more than 150 species including important human pathogens
such as M. tuberculosis and M. leprae (http://www.bacterio.
cict.fr/m/mycobacterium.html). The cell envelopes of these
bacteria contain significant amounts of unique lipids and
heteropolysaccharides that play critical roles in modulating
the host response during infection.1
Among them, lipoglycans termed lipoarabinomannan (LAM)
and lipomannan (LM) display several immunomodulatory
properties through interactions with different receptors of the
immune system.2 Mannan polymers of these molecules are
composed of linear chains of a-(1 A 6)-linked mannose units,
frequently branched with a-(1 A 2)-mannoses. These are
attached to a phosphatidyl-myo-inositol anchor through the
6-position of inositol carrying an additional mannose at the
2-position. Moreover, a LAM molecule contains between 50–80
ble for the addition of a-(1 A 2) branches, was identified only in
2006.5 This discovery was soon followed by recognition of the
first a-(1 A 6)-mannopyranosyltransferase from M. tuberculosis,
Rv2174 (MptA), involved in biosynthesis of the distal part of
LAM/LM.6,7 Another a-(1 A 6)-mannopyranosyltransferase,
MptB from Corynebacterium glutamicum, along with its homolog
in M. tuberculosis Rv1459c, was proposed to extend phosphati-
dylinositol dimannoside and thus synthesize the proximal part
of LAM/LM.8 Surprisingly, the M. smegmatis strain with a
disrupted copy of the Rv1459c homolog did not show any defect
in LAM/LM synthesis, so it is likely that the gene is redundant in
mycobacteria.8
A neoglycolipid acceptor a-D-Manp-(1 A 6)-a-D-Manp-O-C8
(octyl (1 A 6)-a-D-Man2) was used in the cell free assays to
confirm the activities of MptA and MptB,7,8 and exemplified
the value of synthetic acceptors for investigation of glycosyl-
transferases involved in biosynthesis of bacterial cell wall-
associated polymers.
arabinose units organized in distinct structural motifs.3
report describing the activity of a cell-free mannosyltransferase
(ManT) that could participate in biosynthesis of mycobacterial
mannan polymers was published more than two decades ago.4
A
The pilot study demonstrating the efficacy of synthetic
mannosides as acceptor substrates for mycobacterial ManTs
was published in 2001.9 In the assay with GDP-[14C]Man,
aDepartment of Chemistry, University of Manitoba, 144 Dysart Road, Winnipeg, MB,
synthetic (1
A 6)-a-D-mannodisaccharides with variable
R3T 2N2, Canada
b
aglycone structures (octyl, dec-9-enyl, thiooctyl) and dec-9-enyl
(1 A 6)-a-D-Man3 acted as acceptor substrates for ManT.
Thorough analysis of the reaction products revealed that each
of the active compounds accepted one or two mannose
residues attached through a-(1 A 6)-glycosidic linkage. The
transfer was sensitive to the presence of amphomycin, which
confirmed that the direct mannose donors for these ManTs
´
´
Institute of Chemistry, Center for Glycomics, Slovak Academy of Sciences, Dubravska
cesta 9, SK-845 38 Bratislava, Slovakia. E-mail: monika.polakova@savba.sk;
Fax: +421 2 59410 222; Tel: +421 2 59410 272
cDepartment of Biochemistry, Comenius University, Faculty of Natural Sciences,
´
Mlynska dolina, CH1, SK-842 15 Bratislava, Slovakia
Electronic supplementary information (ESI) available: Experimental. 1H and
3
13C NMR spectra of the targeted disaccharides 27–30. Supplementary Fig. S1–S7.
See DOI: 10.1039/c3ra43575j
17784 | RSC Adv., 2013, 3, 17784–17792
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Paper
Fig. 1 Schematic illustration of the reaction catalyzed by the polyprenolphosphomannose-dependent a-(1 A 6)-mannosyltransferase (ManT).
Polyprenolmonophospomannose synthase (PPM synthase) produces polyprenolphosphomannoses, direct donors for ManT from GDP-Man and polyprenolpho-
sphates.
were polyprenylphosphoryl mannoses formed in situ from
GDP-[14C]Man (Fig. 1).
mannopyranoside. On the other hand, oxidized thiooctyl
glycosides, i.e. sulfone derivatives exhibited significantly
decreased efficiency in the transfer reaction. This study
allowed us to identify new aglycone structures since so far
only two models having structures other than octyl aglycone
have been evaluated.9
In the present work we further investigated the effect of the
aglycone structure on the acceptor ability of the newly
synthesized disaccharide homologs 28–30 since the optimal
glycone size for the transferase is disaccharide. The efficiency
of compounds 28–30 to act as ManT acceptors was compared
with that of octyl (1 A 6)-a-D-Man2 27, which can be
considered a reference substrate (Fig. 2).
The cell-free ManT reaction was further explored with
mycobacterial membranes as a source of the enzymatic
activities and compounds derived from octyl (1 A 6)-a-D-
Man2 by modification at the saccharide unit by deoxygenation,
etherification or epimerization.10–14 In our recent study, we
reported a modification achieved by replacement of glycosidic
linkage with a triazole linker.15 Detailed investigation of the
reaction products in cell-free ManT assay showed that
conjugates consisting of two or three mannose units and octyl
aglycone were tolerated by the enzyme; however the effective-
ness was reduced in comparison to that observed for the
acceptor analogues containing a glycosidic linkage.
We have also previously described the synthesis of
mannose glycosides with a variable aglycone moiety: linear
C6, C8, C12, C16 and C20, cyclohexylalkyl (alkyl = methyl or
ethyl), linear SC6, SC8 and their oxidized forms as SO2C6,
SO2C8.16 Examination of these monosaccharide compounds in
the mycobacterial ManT assay indicated that glycosides with
cyclohexylalkyl aglycone acted as substrates in the ManT
reaction with an efficacy similar to that observed for octyl a-D-
Results and discussion
Synthesis
Synthesis of glycosyl acceptors 13–16 was commenced from
the corresponding mannose glycosides 1–416 over three steps
(Scheme 1). Selective protection of the primary hydroxyl group
at C-6 of the saccharide unit by reaction with TrCl was followed
by benzylation of the secondary OH groups. Then, removal of
the acid labile bulky trityl function by treatment with p-TsOH
in a solvent mixture of CH2Cl2 : MeOH gave the target
acceptors. Imidate 17, chosen as a donor, was synthesized
according to a procedure reported recently17 (Scheme 1).
In the next step, a BF3?OEt2-promoted glycosylation of the
prepared acceptors 13–16 with mannopyranosyl trichloroace-
timidate 17 provided the required protected disaccharides 18–
21 with full a-selectivity (Scheme 2). Disaccharide 21 with a
thiooctyl function as aglycone was next transformed into the
corresponding sulfone 22 by oxidation with mCPBA in 49%
yield (over two steps, coupling and oxidation). Subsequent
removal of the acetyl followed by benzyl protective groups
Fig. 2 Structures of the synthesized disaccharide models 27–30.
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 17784–17792 | 17785

View Article Online
Paper
RSC Advances
Scheme 1 Reagents and conditions: a) TrCl, py, 24 h, 55 uC; b) BnBr, NaH, 16 h, 0 uC to rt; c) p-TsOH, CH2Cl2 : MeOH, 35 min, rt.
under deprotection conditions (saponification and hydrogena-
tion) provided the desired disaccharides 27–30 in high yields.
Evidence for a-stereochemistry of the glycosidic linkages in the
target disaccharides 27–30 was confirmed by the JC-1,H-1
heteronucelar coupling constant for the anomeric carbon
atoms18 with values between 168.9 Hz and 170.1 Hz.
disaccharide 30 with alkylsulfonyl aglycone showed about 60%
decrease of incorporation of the radioactive [14C]Man when
compared to the reference disaccharide 27 (Fig. 3).
Interestingly, prolonged exposure of the TLC plate to the
autoradiography film revealed several additional products of
higher polarity, suggesting further step-wise additions of
[14C]Man to the acceptor substrates (data not shown).
1
Biological evaluation
To establish the identity of the products, large scale
enzymatic reactions with synthetic disaccharides 27–30 and
unlabeled GDP-Man were carried out. The reactions were
stopped by addition of ethanol, followed by centrifugation and
re-extraction of the resulting pellet with the same solvent. The
combined dried ethanol extracts were then analyzed by
MALDI-MS and MS/MS or with an ESI mass spectrometer
coupled to HPLC. These experiments enabled us to validate
the presence of oligosaccharide products formed in the
enzymatic reactions; products detected clearly confirmed
addition of one to ten mannose units to each disaccharide
acceptor substrate. A representative example, obtained for the
synthetic acceptor 29, is shown in Fig. 4. The uppermost panel
represents the HPLC chromatogram (Fig. 4a) and mass spectra
recorded from the peaks are shown in this figure as panels b1–
Once the disaccharides 27–30 were prepared, they were
screened for their ability to act as acceptor substrates for the
mannosyltransferases present in cell envelope fractions of
Mycobacterium smegmatis (mc2155). Each of the synthetic
disaccharides was incubated with this enzymatic fraction in
the presence of radioactive GDP-[14C]Man which served as an
indirect mannose nucleotide donor. TLC analysis of the
reaction products followed by autoradiography showed the
incorporation of the radioactive label into compounds 27–30.
Comparison of the spot intensity corresponding to the
monomannosylated adducts showed that octyl (1 A 6)-a-D-
Man2 27 and 2-cyclohexylethyl (1 A 6)-a-D-Man2 29 acted as
the best acceptor substrates with similar activities; cyclohex-
ylmethyl (1 A 6)-a-D-Man2 28 was about 20% less active and
Scheme 2 Reagents and conditions: a) BF3?OEt2, CH2Cl2, 30 min, 0 uC; b) mCPBA, CH2Cl2, 2 h, rt; c) MeONa, CH2Cl2 : MeOH, 16 h, rt; d) 10%Pd/C, H2, MeOH, 4 h, rt.
17786 | RSC Adv., 2013, 3, 17784–17792
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Paper
Fig. 3 Analysis of the products of the in vitro mannosyl transferase reaction with GDP [14C]Man and synthetic acceptors. Panel A, TLC analysis of the reaction products.
The bands were visualized by autoradiography. Panel B, quantification of the reaction products derived from synthetic acceptors corresponding to the addition of the
first [14C]Man using IMAGEJ (NIH) software. The values in the graph represent the means of duplicate reactions.
5. The largest chromatographic peak (elution time 39 min)
under ESI-MS conditions produced [M + Na]+ ions at m/z 475
and corresponded to the synthesized disaccharide 29 (Fig. 4,
b1). The next chromatographic peak with elution time 38 min
(peak 2, in Fig. 4a) gave molecular ions at m/z 637 and
indicated addition of one mannose residue to the parent
disaccharide 29 (mass increase of 162 Da; Fig. 4, b2). MS
spectra shown in Fig. 4 as b3–5 were recorded from smaller
chromatographic peaks 3–5 and provided evidence for addi-
tional consecutive mass increases of 162 Da indicating the
presence of di-, tri- and tetramannosylated products.
eliminate unwanted compounds and to obtain a fraction
enriched with mannosylated adducts (e.g. Supplementary data,
Fig. S4, ESI ). Although the products with more than seven
3
additional mannose units (n = 8–10) were generally detected as
minor peaks, still the ensuing increase in mass of 162 Da
provided good evidence for their presence. In the case of a
reaction mixture of 30, it was possible to detect products
indicating even 11 additional mannose residues (Fig. S5, ESI ).
3
For investigation of the glycosidic linkages in products
formed by ManT in the enzymatic reactions with non-radio-
active GDP-Man, each mannosylated sample was digested
enzymatically with jack bean a-mannosidase and then
analyzed by MALDI-MS, as described in the Experimental
section. Fig. 5 represents MS spectra obtained from the
reaction mixture with 28 before and after a-mannosidase
treatment. After 1 h of incubation with the enzyme, only peaks
corresponding to di- and mono-mannosylated adducts were
detected (Fig. 5b). When the reaction mixture was analyzed
after 3 h, only the monosaccharide, cyclohexylmethyl a-D-
mannopyranoside 2 was detected (m/z 299, Fig. 5c). Similar
results were observed for the products of the enzymatic
reactions with acceptors 29 and 30 (Supplementary data, Fig.
An LC-MS analytical approach was also carried out for the
identification of products in the reaction mixtures obtained
with the acceptors 28 and 30. Here again the same
phenomenon regarding processive extension of the disacchar-
ide chains was observed (Supplementary Fig. S1 and S2, ESI ).
3
LC-MS analysis confirmed 4-fold procession, as shown for 29.
In the case of 28, only 3 products of the enzymatic ManT
reaction were detected – at m/z 623.4, 785.5 and 1109.5,
corresponding to mono-, di- and tetramannosylated products,
respectively (Supplementary data, Fig. S1, b2–4, ESI ). The lack
3
of a trimannosylated product (expected value of m/z at 947)
might be explained by its own consumption in a subsequent
ManT reaction resulting in a product with higher mass.
Interestingly, in the reaction mixture with 30, despite reduced
efficiency due to the presence of the octylsulfonyl aglycone, the
processivity was not negatively affected. The products with
increasing masses of 162 Da indicated the ability of the
enzyme to extend even this poorer substrate (Supplementary
S6 and S7, ESI ).
3
The combination of a-mannosidase treatment with MS-
analysis confirmed that newly formed linkages in mannosy-
lated products have a-configuration and therefore the synthe-
sized disaccharides 28–30 are appropriate substrates for
mycobacterial a-ManT. Several lines of evidence, as reported
in our and other studies, demonstrate that the predominant
activity of the a-mannosyltransferase in extension of synthetic
acceptors in the mycobacterial cell-free system is responsible
for a-(1 A 6)-glycosidic bond formation.9,12,15,16 It was shown
that efficiency of ManT catalyzed mannosyl transfer is strongly
affected by the length of saccharide unit and the most active
acceptor substrates are octyl (1 A 6)-a-D-Man2 and octyl (1 A
6)-a-D-Man3.12 The former has been extensively used as an
acceptor in enzymology studies, e.g. during investigation of
compartmentalization of lipid biosynthesis in mycobacteria19
data, Fig. S2, ESI ).
3
The presence of additional higher mannosylated adducts (n
. 4) was detected by MALDI-TOF/TOF-MS. Similar results were
obtained for the reference compound 27 (Supplementary data,
Fig. S3, ESI ) and as well as for new acceptors. The peaks
3
corresponding to these multiple mannosylated products were
relatively small and some of them were accompanied with
peaks originating from the mycobacterial membrane fraction.
Therefore an additional SPE purification was carried out to
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 17784–17792 | 17787

View Article Online
Paper
RSC Advances
Fig. 4 LC-MS analysis of the enzymatic reaction mixture 29. Panel (a) HPLC profile. Panel (b) ESI mass spectra obtained from peaks 1–5 in figure (a); all peaks are as [M
+ Na]+.
17788 | RSC Adv., 2013, 3, 17784–17792
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Paper
Fig. 5 MALDI-MS spectra recorded from sample 28: a) before treatment with mannosidase (the presence of up to deca mannosylated products is shown in Fig. S4,
ESI3); b) after 1 h; and c) after 3 h incubation with mannosidase. Peaks corresponding to saccharide adducts are in blue color.
or identification of the activities of the mannosyltransferases
MptA and MptB.7,8 Our new model compounds 28–30, carrying
the optimal (1 A 6)-a-D-mannodisaccharide unit, have been
prepared with the aim to investigate the influence of the
aglycone structure on the ability of the synthetic dimannoside
analogs to serve as acceptors in the cell free ManT reaction.
Our results showed that the reference compound 27, as well as
the newly synthesized disaccharides 28 and 29 with
O-glycosidically linked aglycones of different types, but similar
in size, were generally good acceptor substrates for ManT.
Therefore, they both (preferably 29) can be utilized as an
efficient alternative of a so far commonly used acceptor
substrate 27. Similar data has also been previously reported for
disaccharide with slightly longer O-linked dec-9-enyl aglycone,
which acted as an acceptor substrate comparable with 27,
although in a limited concentration range.9 Another report
dealing with the synthesis and evaluation of disaccharides
comprising octyl aglycone and a modified disaccharide unit
did not provide better acceptor substrates than the reference
compound 27. While those being etherified or deoxyge-
nated11,12 were comparable or weaker than 27, epimerization
of the mannosyl unit to either talo- or gluco-configuration led
to a reduction or loss of activity.14 A substantial decrease in
activity has also been observed upon a change of the atom
connecting aglycone with mannose unit, i.e. upon a replace-
ment of oxygen with sulfur. Octyl thiomannodisaccharide was
a weaker acceptor substrate than reference 27 in cell free
ManT assay.9 Oxidation of the sulfur atom resulting in
disaccharide 30 carrying alkylsulfonyl aglycone neither recov-
ered nor reduced the acceptor ability in comparison with
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 17784–17792 | 17789

View Article Online
Paper
RSC Advances
thiomannodisaccharide. However, it is worth noting that in
contrast to our previous results with monosaccharide deriva-
tives,16 the presence of the sulfonyl aglycone in the disacchar-
ide acceptor 30 did not completely abolish the reaction,
probably due to a longer distance of the site of mannosylation
from the aglycone attachment.
VNMRS 400 MHz Varian or INOVA AV 600 MHz spectrometers.
Chemical shifts are referenced to either TMS (d 0.00, CDCl3 for
1
1H) or HOD (d 4.87, CD3OD for H), and to internal CDCl3 (d
77.23) or CD3OD (d 49.15) for 13C. The assignment of
resonances in the 1H and 13C NMR spectra was based on data
obtained from two-dimensional homonuclear and hetero-
nuclear shift correlation experiments. Optical rotations were
measured on a Perkin Elmer 241 polarimeter at 20 uC. The
purity (¢95%) of targeted compounds 27–30 was determined
by microanalyses using a Fisons EA-1108 instrument. Unless
otherwise stated, mass spectrometry-high-resolution mass
spectral (HRMS) and tandem mass spectrometry (MS/MS)
analyses were performed on a UltrafleXtremeTM (Bruker)
operated in the positive ion mode. 2,5-dihydroxybenzoic acid
(DHB, Sigma) was used as a matrix. The aqueous sample
solutions (y0.7 mL) were spotted onto a matrix predeposited
on the surface of a stainless steel MALDI target and were air-
dried. The instrument was calibrated externally over a mass
range 100–3000 Da. Individual parent ions of interest were
manually selected for the structural analysis.
Moreover, mass spectrometric analysis of ManT reaction
products of compounds 27–30 revealed a series of mannooli-
gosaccharides corresponding to the transfer of up to ten
mannosyl residues to the original synthetic substrates. This
transfer was independent of the type of mannose attachment
to the aglycone. These results suggest that the enzyme
responsible for mannose attachment is the processive a-man-
nopyranosyltransferase. So far, the addition of more than two
mannose units was reported only in one study, when using
free mannose as the acceptor substrate in the mycobacterial
cell free system indicated the formation of oligosaccharides
with up to 12 mannoses.4 However, characterization of the
products larger than trisaccharides was based only on the
elution profile of the radiolabeled adducts separated by gel
filtration chromatography.4 Although Berg et al.20 proposed
the involvement of an a-(1 A 6) processive ManT in LM/LAM
biosynthesis, such an enzyme has not been identified so far in
mycobacteria. Our results suggest that processive mannosyla-
tion of synthetic mannodisaccharide substrates does occur in
the cell free system derived from the mycobacterial cell
envelope and offers new opportunities in the search for the
specific enzyme to catalyse this reaction.
Chemistry
General procedure for removal of benzyl groups. Synthesis
of targeted disaccharides 27–30. Compound 23, 24, 25 or 26
(0.12 mmol, 1 eq.) was dissolved in MeOH (5 mL) and 10% Pd/
C was added. The solution was stirred under a H2 atmosphere
for 4 h, then the catalyst was filtered off and washed with
MeOH (10 mL). The solvent was evaporated and purification of
the crude product by column chromatography (CHCl3 : MeOH
10 : 1 A 3 : 1) yielded targeted disaccharides 27–30, which
were lyophilized before use in biological tests.
Conclusions
Octyl a-D-mannopyranosyl-(1 A 6)-a-D-mannopyranoside
(27). (50.2 mg, 92%); [a]D + 31 (c 0.8, H2O); lit11 [a]D + 27.3 (c
0.7, H2O); 1H NMR (400 MHz, CD3OD): d 4.82 (d, 1H, J19,29 = 1.7
Hz, H-19), 4.71 (d, 1H, J1,2 = 1.6 Hz, H-1), 3.91 (ddd, 1H, J = 10.7
Hz, J = 3.1 Hz, J = 2.0 Hz, H-6a), 3.86–3.81 (m, 2H, H-29, H-69a),
3.79 (dd, 1H, J2,3 = 2.8 Hz, H-2), 3.75–3.61 (m, 9H, H-3, H-4,
H-5, H-6b, H-39, H-49, H-59, H-69b, OCH2C7H15), 3.41 (dt, 1H, J
= 6.2 Hz, J = 9.6 Hz, OCH2C7H15), 1.63–1.54 (m, 2H,
OCH2CH2C6H13), 1.39–1.28 (m, 10H, O(CH2)2(CH2)5CH3), 0.90
(t, 3H, J = 6.8 Hz, O(CH2)7CH3). 13C NMR (100 MHz, CD3OD): d
101.7 (C-1, 1JC,H = 169.5 Hz), 101.5 (C-19, 1JC,H = 170.0 Hz), 74.4,
73.2, 73.0, 72.8 (C-3, C-39, C-5, C-59), 72.3(26) (C-2, C-29), 68.8
(C-4), 68.7(26) (C-49, OCH2C7H15), 67.5 (C-6), 63.0 (C-69), 33.2,
30.8, 30.7, 30.6, 27.6, 23.9 (OCH2(CH2)6CH3), 14.6
(O(CH2)7CH3). HRMS (MALDI): m/z 477.2336 MNa+; calcd
477.2312 for C20H38O11Na. Anal. Calcd for C20H38O11: C, 52.85;
H, 8.43. Found: C, 52.74; H, 8.53.
In this work, we have described the synthesis of three new (1
A 6)-a-D-Man2 model compounds 28–30 and their evaluation
in the mycobacterial ManT assay. The results showed that
these disaccharide models have been recognized by a-manno-
syltransferase as acceptor substrates with different efficacies.
We observed that the nature of the glycosidic atom connecting
the mannose unit to the aglycone had a higher impact on the
acceptor substrate ability than the structure of the lipid-like
aglycone. Interestingly, none of the new compounds affected
the processivity of the enzyme, since mass spectrometric
analysis revealed attachment of up to ten additional mannose
units to each original synthetic substrate.
Experimental
General methods
Cyclohexylmethyl a-D-mannopyranosyl-(1 A 6)-a-D-manno-
pyranoside (28). (50.5 mg, 96%); [a]D + 37 (c 0.7, H2O); 1H NMR
(400 MHz, CD3OD): d 4.82 (d, 1H, J19,29 = 1.6 Hz, H-19), 4.68 (d,
1H, J1,2 = 1.5 Hz, H-1), 3.90 (ddd, 1H, J = 10.7 Hz, J = 3.0 Hz, J =
2.1 Hz, H-6a), 3.85–3.81 (m, 2H, H-29, H-69a), 3.79 (dd, 1H, J2,3
= 2.6 Hz, H-2), 3.75–3.60 (m, 8H, H-3, H-4, H-5, H-6b, H-39,
H-49, H-59, H-69b), 3.51 (dd, 1H, J = 6.9 Hz, J = 9.4 Hz,
OCH2C6H11), 3.22 (dd, 1H, J = 5.9 Hz, J = 9.4 Hz, OCH2C6H11),
1.81–0.96 (m, 11 H, OCH2C6H11). 13C NMR (100 MHz, CD3OD):
All commercially available solvents were distilled, dried and
then distilled again immediately before use. Reactions
containing sensitive reagents were carried out under an argon
atmosphere. TLC was performed on aluminium sheets pre-
coated with a silica gel 60 F254 (Merck). Flash column
chromatography was conducted on a silica gel 60 (0.040–
0.060 mm, Merck) with distilled solvents (hexanes, ethylace-
tate, dichloromethane, chloroform, methanol). 1H and 13C
NMR spectra were recorded at 25 uC on a Bruker AM300,
1
1
d 101.8 (C-1, JC,H = 169.5 Hz), 101.5 (C-19, JC,H = 169.9 Hz),
17790 | RSC Adv., 2013, 3, 17784–17792
This journal is ß The Royal Society of Chemistry 2013

View Article Online
RSC Advances
Paper
74.5, 73.2, 73.0, 73.3, 72.8 (C-3, C-39, C-5, C-59, OCH2C6H11),
72.3(26) (C-2, C-29), 68.8(2x) (C-4, C-49), 67.5 (C-6), 63.0 (C-69),
39.4, 31.4, 31.2, 27.8, 27.1(26) (OCH2C6H11). HRMS (MALDI):
m/z 461.1996 MNa+; calcd 461.1999 for C19H34O11Na. Anal.
Calcd for C19H34O11: C, 52.05; H, 7.82. Found: C, 51.82; H,
7.97.
by 200 mM GDP-mannose and the final volume of the reaction
was adjusted with buffer A to 320 mL. These reactions were
carried out in triplicate. After 1 h incubation at 37 uC the
reactions were stopped by an addition of 1 mL, or 2 mL of 96%
ethanol, respectively. The radioactive reaction products were
obtained by n-butanol/water partitioning, as described
before.15,16 The butanol extracts were dried under nitrogen
and resuspended in n-butanol. 10% of these extracts were
analyzed by TLC on aluminum-coated silica 60 F254 plate
(Merck) developed in CHCl3/MeOH/conc NH4OH/H2O
(65 : 25 : 0.5 : 4; by vol.). The TLC plates were exposed to
X-ray film (Kodak Bio-Max MR) and the reaction products
arising from the synthetic acceptors were quantified by
IMAGEJ (NIH) software.
Ethanol extracts from the non-radioactive reactions of 27–30
were clarified by centrifugation and the resulting pellets were
re-extracted with 2 mL 96% ethanol. Ethanol extracts were
combined and dried under the stream of N2 at 37 uC.
LC-MS analysis of the products formed in non-radioactive
enzymatic reactions. On-line LC-MS experiments were per-
formed using a HPLC system (Varian, ProStar). Eluent A was
0.01 M formic acid in water and eluent B consisted of 0.01 M
formic acid in 100% acetonitrile. The proportion of B was
maintained at 10% between 5–10 min and then linearly
increased to 80% over 70 min. The reaction mixtures were
separated at a flow rate of 0.4 mL min21 on a reversed-phase
analytical column (Vydac 218-TP54 C18, 300-Å, Grace,
Hesperia, USA). The column effluent was introduced directly
into the mass spectrometer equipped with an electrospray
ionization source (Varian 500 LC-MS). The nebulizing gas
pressure was 50 psi and drying gas pressure 39 psi at 350 uC.
Mass spectra were recorded in the positive mode with a scan
rate of 3 s/scan.
2-Cyclohexylethyl a-D-mannopyranosyl-(1 A 6)-a-D-manno-
pyranoside (29). (51.0 mg, 94%); [a]D + 39 (c 0.7, H2O); 1H NMR
(400 MHz, CD3OD): d 4.82 (d, 1H, J19,29 = 1.6 Hz, H-19), 4.70 (d,
1H, J1,2 = 1.5 Hz, H-1), 3.90 (ddd, 1H, J = 10.8 Hz, J = 3.3 Hz, J =
2.1 Hz, H-6a), 3.86–3.81 (m, 2H, H-29, H-69a), 3.78 (dd, 1H, J2,3
= 2.6 Hz, H-2), 3.77–3.63 (m, 9H, H-3, H-4, H-5, H-6b, H-39,
H-49, H-59, H-69b, OCH2CH2C6H11), 3.43 (dt, 1H, J = 5.9 Hz, J =
9.8 Hz, OCH2C6H11), 1.81–0.88 (m, 13H, OCH2CH2C6H11). 13C
1
NMR (100 MHz, CD3OD): d 101.7 (C-1, JC,H = 168.9 Hz), 101.6
1
(C-19, JC,H = 169.3 Hz), 74.4, 73.3, 73.0, 72.8 (C-3, C-39, C-5,
C-59), 72.3, 72.2 (C-2, C-29), 68.8, 68.7 (C-4, C-49), 67.6 (C-6),
66.5 (OCH2CH2C6H11), 63.0 (C-69), 38.3, 36.0, 34.8, 34.4, 27.8,
27.6, 27.5 (OCH2CH2C6H11). HRMS (MALDI): m/z 475.2185
MNa+; calcd 475.2155 for C20H36O11Na. Anal. Calcd for
C20H36O11: C, 53.09; H, 8.02. Found: C, 53.32; H, 7.89.
Octyl a-D-mannopyranosyl-(1 A 6)-a-D-mannopyranosyl sul-
1
fone (30). (57.3 mg, 95%); [a]D + 45 (c 0.7, H2O); H NMR (400
MHz, CD3OD): d 4.90 (d, 1H, J1,2 = 1.4 Hz, H-1), 4.82 (d, 1H,
J19,29 = 1.7 Hz, H-19), 4.50 (dd, J2,3 = 3.7 Hz, H-2), 4.27 (ddd, 1H,
J4,5 = 9.8 Hz, J5,6a = 6.2 Hz, J5,6b = 2.0 Hz, H-5), 3.95 (dd, 1H, J3,4
= 9.2 Hz, H-3), 3.88 (dd, 1H, J6a,6b = 11.2 Hz, H-6a), 3.84–3.81
(m, 2H, H-29, H-69a), 3.78 (dd, 1H, H-6b), 3.73–3.61 (m, 5H,
H-4, H-39, H-49, H-59, H-69b), 3.25–3.13 (m, 2H, SO2CH2C7H15),
1.86–1.78 (m, 2H, SO2CH2CH2C6H13), 1.53–1.46 (m,
SO2(CH2)2CH2C5H11), 1.36–1.29 (m, 8H, SO2(CH2)3(CH2)4
CH3), 0.91 (t, 3H, J = 6.9 Hz, SO2(CH2)7CH3). 13C NMR (100
MHz, CD3OD): d 101.6 (C-19, 1JC,H = 169.5 Hz), 92.9 (C-1, 1JC,H
=
170.1 Hz), 78.2 (C-5), 74.6 (C-59), 73.0, 72.9, 72.2 (C-29, C-3,
C-39), 68.7 (C-49), 67.8, 67.7 (C-4, C-6), 66.9 (C-2), 63.0 (C-69),
51.3 (SO2CH2(CH2)6CH3), 33.1, 30.4, 30.2, 29.7, 23.8, 22.6
Linkage determination. To confirm the a-linkages of
mannose units in all enzymatic products, ethanol extracts
from the whole enzymatic reaction mixtures were first partially
purified by SPE on Starta-XC cartridges (30 mg/1 ml,
Phenomenex, Torrance, CA) with deionized water. All collected
fractions were evaporated and the volume was adjusted with
deionized water to 50 mL and then analyzed by MALDI-TOF/
TOF-MS. The fractions enriched with enzymatic adducts were
subjected to a-mannosidase (Jack beans, Sigma) treatment at
37 uC for 1–20 h according to the protocol supplied by
manufacturer. After the incubation, 1 mL of the digested
mixture was spotted on the MALDI target and analyzed under
the same conditions as the original non-digested reaction
mixtures.
(SO2CH2(CH2)6CH3),
14.6
(SO2CH2(CH2)6CH3).
HRMS
(MALDI): m/z 525.1985 MNa+; calcd 525.1982 for
C20H38O12SNa. Anal. Calcd for C20H38O12: C, 47.80; H, 7.62;
S, 6.38. Found: C, 47.94; H, 7.50; S, 6.22.
Biological evaluation
In vitro mannosyltransferase assay and preparation of the
enzyme reaction products for structural characterization.
Synthetic disaccharides 27–30 were tested for their ability to
serve as the acceptors for mycobacterial mannosyltransferases
in the in vitro reaction. The enzymatically active cell envelope
(membrane and cell wall) fractions of Mycobacterium smegma-
tis mc2155, which were grown in the nutrient broth (EM
Science) were prepared essentially as described previously.15,16
The composition of the reaction mixtures was as follows: 1.5
mg membrane protein, 1 mg cell wall protein, 0.05 mCi GDP-
[14C]mannose (Amersham, 275 mCi/mmol), 60 mM ATP, DMSO
Acknowledgements
This work was supported by the APVV-51-046505 and VEGA-2/
0159/12 grants, the Slovak State Programme Project No.
2003SP200280203, by FP7 EC grant no. 260872 (MM4TB) and
in final concentration of 0.8% (v/v),
4 mM synthetic
´
the SRDA grant DO7RP-0015-11. Drs Vladimır Puchart and
disaccharide, and buffer A (50 mM MOPS, pH 7.9, 10 mM
MgCl2, 5 mM b-mercaptoethanol) in the final volume of 160
mL. In the cold reactions for obtaining material for structural
characterization radioactive GDP-[14C]mannose was replaced
´
´
Jana Kordulakova are acknowledged for helpful discussions.
´ˇ
´
Martina Belanova is acknowledged for performing pilot cell
free experiments and Miroslav Brecik for IMAGEJ analysis.
This journal is ß The Royal Society of Chemistry 2013
RSC Adv., 2013, 3, 17784–17792 | 17791

View Article Online
Paper
RSC Advances
10 V. Subramaniam, S. S. Gurcha, G. S. Besra and T. L. Lowary,
Tetrahedron: Asymmetry, 2005, 16, 553.
11 V. Subramaniam, S. S. Gurcha, G. S. Besra and T. L. Lowary,
Bioorg. Med. Chem., 2005, 13, 1083.
12 P.-H. Tam, G. S. Besra and T. L. Lowary, ChemBioChem,
2008, 9, 267.
13 P.-H. Tam and T. L. Lowary, Carbohydr. Res., 2007, 342,
1741.
14 P.-H. Tam and T. L. Lowary, Org. Biomol. Chem., 2010, 8,
181.
References
1 M. Jankute, S. Grover, A. K. Rana and G. S. Besra, Future
Microbiol., 2012, 7, 129.
2 A. K. Mishra, N. N. Driessen, B. J. Appelmelk and G.
S. Besra, FEMS Microbiol. Rev., 2011, 35, 1126.
3 K. H. Khoo, E. Douglas, P. Azadi, J. M. Inamine, G. S. Besra,
ˇ
´
K. Mikusova, P. J. Brennan and D. Chatterjee, J. Biol. Chem.,
1996, 271, 28682.
4 K. Yokoyama and C. E. Ballou, J. Biol. Chem., 1989, 264,
21621.
´
´
´ˇ
´
ˇ
´
´
15 M. Polakova, M. Belanova, K. Mikusova, E. Lattova and
5 D. Kaur, S. Berg, P. Dinadayala, B. Gicquel, D. Chatterjee,
M. R. McNeil, V. D. Vissa, D. C. Crick, M. Jackson and P.
J. Brennan, Proc. Natl. Acad. Sci. U. S. A., 2006, 103, 13664.
6 D. Kaur, M. R. McNeil, K.-H. Khoo, D. Chatterjee, D.
C. Crick, M. Jackson and P. J. Brennan, J. Biol. Chem., 2007,
282, 27133.
7 A. K. Mishra, L. J. Alderwick, D. Rittmann, R. V. V. Tatituri,
J. Nigou, M. Gilleron, L. Eggeling and G. S. Besra, Mol.
Microbiol., 2007, 65, 1503.
8 A. K. Mishra, L. J. Alderwick, D. Rittmann, C. Wang,
A. Bhatt, W. R. Jacobs Jr., K. Takayama, L. Eggeling and G.
S. Besra, Mol. Microbiol., 2008, 68, 1595.
9 J. R. Brown, R. A. Field, A. Barker, M. Guy, R. Grewal, K.-
H. Khoo, P. J. Brennan, G. S. Besra and D. Chatterjee,
Bioorg. Med. Chem., 2001, 9, 815.
H. Perreault, Bioconjugate Chem., 2011, 22, 289.
´
´
´ˇ
´
ˇ
ˇ ´
16 M. Polakova, M. Belanova, L. Petrus and K. Mikusova,
Carbohydr. Res., 2010, 345, 1339.
´
´
17 F. S. Ekholm, M. Polakova, A. J. Pawłowicz and R. Leino,
Synthesis, 2009, 4, 567.
18 K. Bock and C. A. Pedersen, J. Chem. Soc., Perkin Trans. 2,
1974, 2, 293.
19 Y. S. Morita, R. Velasquez, E. Taig, R. F. Waller, J.
H. Patterson, D. Tull, S. J. Williams, H. Billman-Jacobe
and M. J. McConville, J. Biol. Chem., 2005, 280, 21645.
20 S. Berg, D. Kaur, M. Jackson and P. J. Brennan,
Glycobiology, 2007, 17, 35R.
17792 | RSC Adv., 2013, 3, 17784–17792
This journal is ß The Royal Society of Chemistry 2013