Antimicrobial activity of mannose-derived glycosides

Article abstract of DOI:10.1007/s00706-015-1530-8

A set of 14 synthetic mannolipid-mimicking O-mannosides, S-mannosides, and mannosylsulfones varying in aglycone length were evaluated for their antimicrobial activity towards yeast Candida albicans as well as Gram-positive and Gram-negative bacteria, Staphylococcus aureus and Escherichia coli, respectively. S. aureus was the most susceptible to dodecyl α-d-mannopyranoside showing MIC value of 78 μM, while dodecyl α-d-thiomannopyranoside was superior yeast inhibitor exhibiting twofold lower MIC value. On the other hand, E. coli was resistant to both dodecyl glycosides. Mannosides exposure on RAW 264.7 cell line murine macrophages revealed the tight structure - immunobiological activity pattern. No significant cytotoxic effect and suppression of proliferation as a consequence of cell injury following 24 h exposure with 1-100 μg/cm³ mannosides were observed.

Full text of DOI:10.1007/s00706-015-1530-8

Monatsh Chem (2015) 146:1707–1714
DOI 10.1007/s00706-015-1530-8
ORIGINAL PAPER
Antimicrobial activity of mannose-derived glycosides
1
2
2
2
Andrea Bilkova´
•
Ema Paulovi cˇ ov a´
•
Lucia Paulovi cˇ ov a´
•
Monika Pol a´ kov a´
Received: 3 March 2015 / Accepted: 6 July 2015 / Published online: 30 July 2015
Ó Springer-Verlag Wien 2015
Abstract A set of 14 synthetic mannolipid-mimicking
O-mannosides, S-mannosides, and mannosylsulfones
varying in aglycone length were evaluated for their
antimicrobial activity towards yeast Candida albicans as
well as Gram-positive and Gram-negative bacteria, Sta-
phylococcus aureus and Escherichia coli, respectively.
S. aureus was the most susceptible to dodecyl a-D-
mannopyranoside showing MIC value of 78 lM, while
dodecyl a-D-thiomannopyranoside was superior yeast
inhibitor exhibiting twofold lower MIC value. On the other
hand, E. coli was resistant to both dodecyl glycosides.
Mannosides exposure on RAW 264.7 cell line murine
macrophages revealed the tight structure—immunobiolog-
ical activity pattern. No significant cytotoxic effect and
suppression of proliferation as a consequence of cell injury
Keywords Mannoside Á Antimicrobial activity Á
Candida albicans Á Escherichia coli Á
Staphylococcus aureus
Introduction
Glycosides are compounds consisting of a saccharide core
linked to a simple or more complex aglycone. Such com-
pounds occur in nature and are a matter of interest because
many of them exhibit interesting biological activities [1, 2].
Although their spectrum is wide, there are only several
reports on antimicrobial activity of carbohydrate fatty acid
esters [3–5], and alkyl glycosides have also been scarcely
studied in this respect [6–8].
3
following 24 h exposure with 1–100 lg/cm mannosides
were observed.
The most often evaluated carbohydrate esters have been
those derived from lauric and palmitic acids. For example,
sucrose laurate inhibited the growth of Bacillus cereus a
food poisoning bacterium [5]. This ester along with mal-
tose and maltotriose laurates completely inhibited
Streptococcus sobrinus, a dental pathogen [3]. Escherichia
coli, Lactobacillus plantarum, and Bacillus spp. are even
more susceptible because they have also been affected by
maltose palmitate, in addition to the three oligosaccharide
laurates [4].
Graphical abstract
The best inhibition towards G+ and G- bacteria and yeast
among 14 tested mannosides
S. aureus CNCTC Mau 82/78 C. albicans CNCTC 59/91
E. coli CNCTC 327/73
OH
O
OH
O
OH
O
OH
O
HO
HO
HO
HO
HO
HO
HO
HO
HO
HO
HO
HO
O(CH2)11CH3
MIC 78µM
S(CH2)11CH3
MIC 39µM
O(CH2)9CH3
MIC 1.25mM
S(CH2)7CH3
MIC 1.25mM
Alkyl glycosides are attractive compounds because
many of them can be made from naturally occurring
renewable sources (such as sugar and fatty alcohols).
Moreover, chemical synthesis of alkyl glycosides as gly-
colipid mimetics is straightforward and requires only
limited number of reaction steps. Due to their biocompat-
ibility and biodegradability, the alkyl glycosides have
potential biological and pharmaceutical applications as
detergents, pharmaceuticals, agrochemicals, oral-care
&
Monika Pol a´ kov a´
chemonca@savba.sk
1
Department of Cellular and Molecular Biology of Drugs,
Faculty of Pharmacy, Comenius University, Kalin cˇ iakova 8,
8
32 32 Bratislava, Slovakia
2
Institute of Chemistry, Center for Glycomics, Slovak
Academy of Sciences, D u´ bravska cesta 9, 845 38 Bratislava,
Slovakia
123

1
708
A. Bilkov a´ et al.
products, and medical supplies [9–11]. Additional benefits
result from their low toxicity which is particularly relevant
for expanding their applicability in fields of food and
cosmetic industries as well as personal hygiene.
Alkyl glycosides target the cell membrane. Their
incorporation into the lipid bilayer may lead to either a
disruption or a modification of the membrane structure. It
is known that many of them are able to dissolve the lipid
membrane and in this way disrupts the cells. This causes
a reduction of the surface tension of the membrane that
allows water to flow into the cell and finally results in
lysis and bactericidal action. To make this action as
efficient as possible, correct balance between the hydro-
philic and hydrophobic parts of the alkyl glycoside is
essential [12]. The factors that determine the physical
and chemical properties of the alkyl glycosides are
structure of hydrophobic aglycone, the sugar core, and its
stereochemistry. Searching for the active alkyl glycosides
is of importance because they could be a valuable
environment-friendly alternative for some common
products.
Regarding the synthetic alkyl glycosides, octyl and
dodecyl 2-deoxy-b-D-arabino-hexopyranosides have been
reported to inhibit the growth of Enterococcus faecalis,
which was also highly susceptible to dodecyl 2,6-dideoxy-
a-L-arabino-hexopyranoside. The latter compound was
particularly active towards Bacillus sp. [6]. To the best of
our knowledge, antimicrobial potential of only three
mannopyranosides has been tested [8]. In this study,
a series of D-mannosides having lipid aglycone of the
length varying from 6 to 20 carbon atoms that is attached
by O-glycosidic, S-glycosidic, or sulfonyl linkage were
evaluated. Staphylococcus aureus, E. coli, and Candida
albicans as the representatives of Gram-positive bacteria,
Gram-negative bacteria, and yeasts, respectively, were
chosen with the aim to better understand the factors
affecting the antimicrobial activity and specificity of the
glycolipid-like compounds.
Fig. 1 The tested mannosides
which gave a doublet at d 4.73 for O-mannoside 3, while
for S-mannosides 11 and 12 was downfielded and found at
d 5.23 and 5.24, respectively. On the other hand, anomeric
carbon signals for 3 appeared at d 101.5, and for both 11
and 12 was upfielded to d 86.4.
The efficacy of the glycosides 1–14 towards the 3
above-mentioned microorganisms was expressed as MIC
value. The results are summarized in Table 1. The refer-
ence compounds, mannose, methyl a-D-mannopyranoside
as well as 1-dodecanethiol, were inactive, and 1-dodecanol
did not inhibit E. coli and showed MIC value of 1.25 mM
towards S. aureus and 10 mM towards C. albicans.
Escherichia coli was the least affected microorganism
by the tested compounds. However, elongation of the
aglycone chain of the O-glycosides from C-6 to C-10 led to
16-fold increase in inhibitory activity giving MIC value of
1.25 mM for 3, which was the most efficient O-mannoside.
In contrast, further elongation of the alkyl chain to C-12
resulted in a complete loss of activity. A similar trend was
also observed for S-glycosides. The lowest MIC value of
1.25 mM was determined for 10 bearing octyl aglycone,
which was fourfold and 32-fold more active than hexyl
thiomannoside 9 and decyl thiomannoside 11, respectively.
Also in this case, dodecyl derivative 12 was inactive.
The efficacy of S. aureus inhibition was also increasing
with the elongation of the aglycone alkyl chain. Dodecyl a-
Results and discussion
A list of 14 tested compounds is depicted in Fig. 1. We
have previously published the synthesis of 1, 2, 4–10, 13,
14 [13]. Mannosides 3, 11, and 12 were prepared by
a similar procedure (Scheme 1, ‘‘Experimental’’ section).
Briefly, SnCl -promoted glycosylation of the correspond-
4
ing alcohol with peracetylated mannose provided after
column chromatography O- and S-mannosides with a-se-
lectivity. Subsequent removal of the acetyl protective
groups afforded the target mannosides 3, 11, and 12 in
moderate yields. Their structures were confirmed on the
basis of a typical chemical shift [13] of the anomeric proton
1
23

Antimicrobial activity of mannose-derived glycosides
1709
Table 1 MIC values of the mannosides (mM)
having the aglycone chain length of C-16 and C-20 was
affected by their limited solubility that resembles the sol-
ubility of C-16 and C-20 ethers of methyl glycosides [14].
Considering the selectivity and potency of the tested
compounds, Gram-negative bacterium E. coli appeared to
be the most resistant to the inhibition. The most efficient
glycosides were decyl mannoside 3 and octyl thioman-
noside 10, both exhibiting MIC value of 1.25 mM. The
aglycone lengths of these glycosides are similar to that of
short-chain fatty acids and their carbohydrate esters [14],
which exhibited comparable low inhibitory activity
towards the Gram-negative bacteria. In contrast, longer
free fatty acids [15] and their carbohydrate esters [3, 4, 14]
are more efficient towards the Gram-positive bacteria,
which agrees well with our findings that dodecyl
thiomannoside 12 and, in particular, dodecyl mannoside 4
were the most efficient inhibitors of S. aureus. The sus-
ceptibility of C. albicans resembles that of S. aureus,
although thiomannoside 12 was the most potent inhibitor of
the yeast (Fig. 2). It is also noteworthy that among the
tested O-mannosides, dodecyl glycoside 4 has been previ-
ously found to be the best inhibitor of S. aureus, while
E. coli has been to the greatest extent affected by decyl
mannoside 3, and inhibitory activity of the both glycosides
towards C. albicans was the same [8]. The data reported
are qualitatively consistent with ours, although in our case
the influence of the aglycone length is much more pro-
nounced. These differences may be explained by the
different microbial strains used.
Compound
E. coli CNCTC S. aureus C. albicans
327/73 CNCTC Mau 82/78 CNCTC 59/91
1
2
3
4
20
n.i.
n.i.
2.5
1.25
n.i.
n.i.
n.i.
20
5
10
0.625
0.078
n.i.
0.625
0.078
0.31
0.31
n.i.
a
5
a
6
n.i.
7
8
9
n.i.
10
10
20
5
10
10
1
1
1
1
1
0
1
2
3
4
1.25
40
1.25
0.625
0.625
40
2.5
0.31
0.039
n.i.
n.i.
40
20
10
n.i.
-
4
-4
6.8 9 10
Ciprofloxacine \3 9 10
n.a.
n.a. not applicable, n.i. no inhibition observed at the highest con-
a
centration (40 mM) tested, or due to limited solubility of the
compound its highest concentration tested was 0.625 mM
D-mannopyranoside 4 showed the MIC value of 0.078 mM,
which is the lowest among all tested compounds towards
this bacterium. It was 64-fold more potent than octyl gly-
coside 2. The most efficient thioglycosides 11 and 12
showed the same MIC value of 0.625 mM, and hence, 12
was eightfold less active than its O-linked analog 4.
Candida albicans was the most susceptible to the stud-
ied compounds, and thiomannosides were more active than
mannosides. Moreover, the elongation of the aglycone
chain of both S- and O-glycosides resulted in a significant
increase in their efficacy. In this case, however, dodecyl
thiomannoside 12, exhibiting MIC value of 0.039 mM, was
twofold better inhibitor than dodecyl mannoside 4 having
MIC value of 0.078 mM. Both decyl S- and O-mannosides
were one order of magnitude poorer inhibitors than the
corresponding dodecyl counterparts, and further shortening
of the aglycone was accompanied with a concomitant
reduction of the inhibition efficiency. This trend resulted in
Based on the results of the growth inhibition of S.
aureus, E. coli, and C. albicans and resulting MICs, the
in vitro cytotoxicity and cell proliferation tests with RAW
264.7 cell line murine macrophages following the 24 h
exposure with the most effective glycosides i.e., 3, 4, 10,
and 12 (Fig. 2) have been performed. In comparison with
non-treated cells, the non-cytotoxic, rather stimulation
effect of alkyl (thio)mannosides used at the concentrations
3
5
ranged from 1 to 100 lg/cm /10 cells on cell proliferation
could be observed with RAW 264.7 cell line murine
macrophages, following 24 h exposure (Fig. 3). The only
exception represented mannoside 3, where the 95.3 %
(P \ 0.001) decrease of the proliferation with the highest
an inactivity of hexyl mannoside
concentration.
1
at 40 mM
3
concentration of 100 lg/cm has been revealed. Generally,
concentration-dependent proliferation activity was
The potency of sulfones 13 and 14 was reduced towards
all tested strains in comparison with the corresponding
thiomannosides 9 and 10. This biological activity reduction
caused by the oxidation of alkylthio group to sulfone was
only slightly compensated by the effect of alkylsulfonyl
aglycone prolongation.
3
observed, the highest concentrations of 10 and 100 lg/cm /
10 cells being stimulative to lesser degree as the dose of
5
3
5
1 lg/cm /10 cells. Concentration of mannosides higher
3
than 10 lg/cm exerted significant dose-dependent
decrease of cell proliferation if compared to those of 1 and
3
Furthermore, replacement of alkyl aglycone in com-
pounds 1 and 2 with cyclohexylalkyl one of similar length
10 lg/cm , but not in comparison with non-treated base-
line. The concentration-dependent kinetics of proliferation
index supports these findings (Fig. 3). The only different
pattern of proliferation was observed for thiomannoside 10
(compounds 7 and 8) did not result in improval of the
potency. Evaluation of another alkyl mannosides 5 and 6
123

1
710
A. Bilkov a´ et al.
S. aureus CNCTC Mau 82/78 C. albicans CNCTC 59/91
E. coli CNCTC 327/73
OH
O
OH
O
OH
O
OH
O
HO
HO
HO
HO
HO
HO
HO
HO
HO
HO
HO
HO
O(CH2)11CH3
MIC 78 µM
Fig. 2 The most potent mannosides against each of the tested microbial strains
S(CH ) CH3
O(CH2)9CH3
S(CH2)7CH3
2 11
4
12
3
10
MIC 39µM
MIC 1.25mM
MIC 1.25mM
reasonable to synthesize a set of model compounds varying
in the aglycone length and shape (alkyl and cycloalkyl)
because it helps to identify appropriate structure of aglycone
that is necessary for mannolipid inhibitory activity against
given bacterial strains. The microorganisms used in this
study are of importance to food and health areas. E. coli
induces water and food poisoning symptoms. S. aureus
infection results in a range of mild to severe infection as well
as food poisoning and toxic shock syndrome. Human
pathogen C. albicans is well-known causing agent of fungal
infection. The most susceptible to the tested mannolipids
were found to be S. aureus and C. albicans, both being of
significant medicinal importance. Against them, two man-
nolipids, 4 and 12, showed potent activity at micromolar
level. In addition, the latter exhibited also selectivity towards
the Gram-positive bacterium [comp 12: IC50 (S. aureus/IC50
(C. albicans) = 16, E. coli n.i.].
Fig. 3 The effect of mannosides’ exposure on RAW 264.7 cell line
murine macrophages. The experimental data are expressed as
geometric means of double measurements ± SD. Comparisons of
data groups were performed vs untreated baseline proliferation, and
statistical significance of differences is expressed as follows:
Therefore, the effort to find out as much as efficient and
selective microbial strains inhibitors based on saccharides
is ongoing interest due to their environmental acceptance.
Moreover, they represent suitable and safe alternative to
standard therapy, to which many of microbial strains have
become resistant over the years.
?
??
??
?
P \ 0.001, 0.001 \ P \ 0.01, 0.01 \ P \ 0.05. Experimen-
3
tal datasets comparisons with 1 (lg/cm ) dose treatment: P \ 0.001,
*
*0.001 \ P \ 0.01, *0.01 \ P \ 0.05; comparison between 10 and
3 ### ##
1
00 (lg/cm ) dose treatment:
P \ 0.001, 0.001 \ P \ 0.01,
#
0
.01 \ P \ 0.05. Differences were considered significant when
0
.01 \ P \ 0.05
The tested compounds provide starting templates for
further development of carbohydrate-based antimicrobial
agents. Obtained results encourage us to synthesize a larger
set of glycosides to further explore the influence of several
structural variations (such as e.g., aglycone length, carbo-
hydrate core and its size, anomeric configuration) on their
biological activity as well as to determine bacteriostatic
and bactericidal activity on a wider scale of microorgan-
isms that are of interest in various areas such as food,
cosmetics, health, and others. Work is currently under way
and results will be reported in due course.
with shortest aglycone length, where the highest concen-
3
tration of 100 lg/cm triggered the most effective
enhancement of proliferation. Based on the comparison of
the proliferation activity of O- and S-mannosides (4 and
1
2), both with dodecyl aglycone, the statistically more
significant concentration-dependent pattern (P \ 0.001)
has been revealed with the mannoside 12 (Fig. 3). The
growth inhibition experiments (Table 1; Fig. 2) resembled
also the tight structure—immunobiological activity rela-
tions of these mannosides. Evidently, the length of
aglycone and the type of glycosidic linkage (O- or S-) are
of great importance.
In summary, a correlation between increased hydropho-
bicity of mannosides and their antimicrobial activity was not
observed. However, the results strongly suggest that only
glycoside bearing aglycone of a certain length possesses
significant activity, and the type of glycosidic linkage (O- or
S-) also affects the potency and selectivity. Therefore, it was
Conclusions
A series of alkyl O- and S-mannosides were synthesized
and evaluated as inhibitors of three microbial strains. The
most efficient were those having dodecyl aglycone showing
MIC values at micromolar level. While S. aureus was the
1
23

Antimicrobial activity of mannose-derived glycosides
1711
most susceptible to O-mannoside, thiomannoside was
superior yeast inhibitor. Antimicrobial activity of glycol-
ipids is still an open field. Therefore, it is desirable to
investigate further potential of alkyl glycosides as antimi-
crobial agents.
(1 eq) was added. The reaction mixture was stirred for
10 min, and then, the corresponding alcohol (1.2 eq) was
added dropwise. The resulting mixture was stirred for
15 min, brought to r.t. and stirring was continued for 16 h.
3
The reaction mixture was diluted with 80 cm CH Cl and
2
2
3
poured into 100 cm ice-cold satd NaHCO under stirring.
3
3
The organic phase was separated, washed with 100 cm
Experimental
water, dried with anhydrous Na SO , filtered, and con-
4
2
centrated. The crude product was purified by flash
TLC was performed on aluminum sheets precoated with
silica gel 60 F254 (Merck). Flash column chromatography
was carried out on silica gel 60 (0.040–0.060 mm, Merck)
with distilled solvents (hexane, ethyl acetate, and metha-
chromatography (hexane:EtOAc 7:1 ? 3:1).
Decyl 2,3,4,6-tetra-O-acetyl-a-D-mannopyranoside (3A,
C H O )
2
4 40 10
Yield 0.44 g (36 %); colorless oil; [a] = ?18.0 (c = 0.5,
D
nol). Dichloromethane was dried (CaH ) and distilled
2
CHCl ) [Ref. [16] [a] = ? 45.0 (c = 1.56, CHCl ), Ref.
3
D
3
before use. All reactions containing sensitive reagents were
carried out under argon atmosphere. All products were
20
D
1
[
17] ½aŠ = ?37.8 (c = 1.0, CH Cl )];
H
NMR
2
2
(
400 MHz, CDCl ): d = 5.36 (dd, 1H, J = 3.4 Hz,
1
13
3
2,3
fully characterized with H NMR and C NMR in com-
bination with the 2D NMR techniques (COSY, HSQC),
which were recorded at 25 °C with a VNMRS 400 MHz
H-3), 5.27 (dd, 1H, J3,4 = 9.9 Hz, H-4), 5.23 (dd, 1H,
H-3), 4.80 (d, 1H, J1,2 = 1.7 Hz, H-1), 4.28 (dd, 1H,
J5,6b = 5.4 Hz, J6a,6b = 12.2 Hz, H-6b), 4.10 (dd, 1H,
J5,6a = 2.4 Hz, H-6a), 3.98 (ddd, 1H, H-5), 3.67 (dt, 1H,
J = 6.8 Hz, J = 9.6 Hz, OCH (CH ) CH ), 3.44 (dt, 1H,
Varian spectrometer. Chemical shifts are referenced to
1
either TMS (d = 0.00 ppm, CDCl for H) or HOD
3
1
4.87 ppm, CD OD for H), and to internal CDCl
2
2 8
3
(
3
3
J = 6.6 Hz, J = 9.6 Hz, OCH (CH ) CH ), 2.15, 2.10,
3
1
3
2
2 8
(
d = 77.23 ppm) or CD OD (49.15 ppm) for C. Optical
3
2.04, 1.99 (each s, each 3H, 49 CH CO), 1.61–1.27 (m,
3
rotations were measured on a Jasco P2000 polarimeter at
1
6H, OCH (CH ) CH ), 0.88 (t, 3H, J = 7.0 Hz, O(CH )
2 9-
2
2 8
3
2
0 °C. High-resolution mass determination was performed
1
3
CH ) ppm; C NMR (100 MHz, CDCl ): d = 170.8,
3
3
by ESI–MS on a Thermo Scientific Orbitrap Exactive
instrument operating in positive mode. For bacterial cul-
tures incubation, Incubator BD 23 (Binder GmbH,
Germany) was used. Turbidity of bacterial cultures was
determined spectrophotometrically at 600 nm (Lambda 35
UV/VIS Spectrophotometer, PerkinElmer, USA). Cipro-
floxacine was used as a reference.
1
6
6
70.3, 170.1, 170.0 (49 CH CO), 97.7 (C-1), 69.9 (C-2),
3
9.3 (C-3), 68.7, 68.6 (C-5, OCH (CH ) CH ), 66.5 (C-4),
2
2 8
3
2.7 (C-6), 32.1, 29.7 (29), 29.6, 29.5, 29.4, 26.3, 22.9
(
(
OCH (CH ) CH ), 21.1, 20.9 (39) (49 CH CO), 14.3
2 2 8 3 3
OCH (CH ) CH ) ppm; HRMS (ESI): m/z calcd for
2
2 8
3
?
C H O Na [M ? Na] 511.2519, found 511.2502.
24 40 10
Decyl 2,3,4,6-tetra-O-acetyl-1-thio-a-D-mannopyranoside
(11A, C H O S)
General procedure for glycosylation (Scheme 1)
2
4 40 9
Yield 0.70 g (54 %); yellow oil; [a] = ?68.0 (c = 0.5,
D
1
A stirred solution containing 1 g 1,2,3,4,6-penta-O-acetyl-
3
D-mannopyranose (2.56 mmol, 1 eq) in 20 cm CH Cl
CHCl ); H NMR (400 MHz, CDCl ): d = 5.34–5.25 (m,
3
3
4H, H-1, H-2, H-3, H-4), 4.40 (m, 1H, H-5), 4.32 (dd, 1H,
2
2
2
was cooled down on an ice bath, and 1 M SnCl in CH Cl
4
J5,6b = 5.3 Hz, J6a,6b = 12.2 Hz, H-6b), 4.09 (dd, 1H,
2
123

1
712
A. Bilkov a´ et al.
J5,6a = 2.2 Hz, H-6a), 2.68–2.53 (m, 2H, SCH (CH2)8-
2
Decyl 1-thio-a-D-mannopyranoside (11, C H O S)
16 32 5
CH ), 2.16, 2.09, 2.05, 2.00 (each s, each 3H, 49 CH CO),
3
Yield 0.12 g (89 %); colorless oil; [a] = ?160.0
3
D
1
(c = 0.5, methanol); H NMR (400 MHz, CD OD):
1
.63–1.26 [m, 16H, SCH (CH ) CH ], 0.88 [t, 3H,
2
2 8
3
3
1
J = 6.6 Hz, S(CH ) CH ] ppm; C NMR (100 MHz,
3
d = 5.24 (d, 1H, J1,2 = 1.4 Hz, H-1), 3.94–3.90 (m, 2H,
H-2, H-3), 3.84 (dd, 1H, J5,6a = 2.4 Hz, J6a,6b = 11.8 Hz,
H-6a), 3.78 (dd, 1H, 1H, J5,6b = 5.6 Hz, H-6b), 3.70–3.64
(m, 2H, H-4, H-5), 2.73–2.67 [m, 1H, SCH (CH ) CH ],
2
9
3
CDCl ): d = 170.2, 170.1, 170.0, 169.9 (49 CH CO), 82.7
3
3
(
(
(
(
C-1), 71.3 (C-2), 69.6 (C-3), 69.1 (C-5), 66.5 (C-4), 62.6
C-6), 32.0, 31.5, 29.7, 29.6, 29.5, 29.4, 29.2, 28.9, 22.8
S(CH ) CH ), 21.1, 20.9 (29), 20.8 (49 CH CO), 14.2
2
2 8
3
2.64–2.58 [m, 1H, SCH (CH ) CH ], 1.70–1.32 [m, 16H,
3
2
9
3
3
2
2 8
SCH (CH ) CH ) ppm; HRMS (ESI): m/z calcd for
2
SCH (CH ) CH ], 0.92 [t, 3H, J = 6.8 Hz, S(CH ) CH ]
2 2 8 3 2 9 3
13
2 8
3
?
C H O SNa [M ? Na] 527.2291, found 527.2285.
40
ppm; C NMR (100 MHz, CD OD): d = 86.4 (C-1), 74.8,
2
4
9
3
73.8, 73.2, 68.9 (C-2, C-3, C-4, C-5), 62.7 (C-6), 33.1,
Dodecyl
2,3,4,6-tetra-O-acetyl-1-thio-a-D-mannopyra-
31.9, 30.7 (39), 30.4, 30.3, 30.0, 23.7 (S(CH ) CH ), 14.4
2
9
3
noside (12A, C H O S)
9
(S(CH ) CH ) ppm; HRMS (ESI): m/z calcd for
2 9
2
6
44
3
?
Yield 0.63 g (46 %); yellow oil; [a] = ?72.0 (c = 0.5,
C H O SNa [M ? Na] 359.1868, found 359.1861.
16 32 5
D
2
D
0
1
CHCl ) [Ref. [18] ½aŠ = ?76.8 (c = 0.37, CHCl )]; H
3
3
Dodecyl 1-thio-a-D-mannopyranoside (12, C H O S)
1
8 36 5
NMR (400 MHz, CDCl ): d = 5.35–5.26 (m, 4H, H-1,
3
Yield 0.12 g (88 %); colorless oil; [a] = ?164.0
D
H-2, H-3, H-4), 4.39 (m, 1H, H-5), 4.32 (dd, 1H,
J5,6b = 5.3 Hz, J6a,6b = 12.2 Hz, H-6b), 4.08 (dd, 1H,
J5,6a = 2.3 Hz, H-6a), 2.68–2.53 [m, 2H, SCH (CH )
1
(
c = 0.5, methanol); H NMR (400 MHz, CD OD):
3
d = 5.23 (d, 1H, J1,2 = 1.4 Hz, H-1), 3.94–3.90 (m, 2H,
H-2, H-3), 3.84 (dd, 1H, J5,6a = 2.4 Hz, J6a,6b = 11.8 Hz,
H-6a), 3.78 (dd, 1H, 1H, J5,6b = 5.6 Hz, H-6b), 3.70–3.66
2
2 10-
CH ], 2.17, 2.10, 2.05, 2.00 (each s, each 3H, 49 CH CO),
3
3
1
.66–1.24 [m, 20H, SCH (CH ) CH ], 0.88 [t, 3H,
2 2 10 3
1
J = 6.7 Hz, S(CH ) CH ] ppm; C NMR (100 MHz,
3
(m, 2H, H-4, H-5), 2.73–2.66 [m, 1H, SCH (CH )10CH ],
2 2 3
2
11
3
2
.64–2.57 [m, 1H, SCH (CH ) CH ], 1.69–1.31 [m, 20H,
2
2 10
3
CDCl ): d = 170.7, 170.1, 170.0, 169.9 (49 CH CO), 82.7
3
3
SCH (CH ) CH ], 0.96 [t, 3H, J = 7.4 Hz, S(CH ) CH ]
2
2 10
3
2 11
3
(
C-1), 71.3 (C-2), 69.6 (C-3), 69.0 (C-5), 66.5 (C-4), 62.6
1
3
ppm; C NMR (100 MHz, CD OD): d = 86.4 (C-1), 74.8,
3
(
C-6), 32.0, 31.5, 29.8, 29.7 (29), 29.5 (29), 29.4, 29.3,
73.8, 73.2, 68.9 (C-2, C-3, C-4, C-5), 62.7 (C-6), 33.1,
2
8.9, 22.8 (S(CH ) CH ), 21.1, 21.0, 20.9, 20.8 (49
2 9 3
31.9, 30.8 (29), 30.7 (39), 30.5, 30.3, 29.9, 23.7
CH CO), 14.3 (SCH (CH ) CH ) ppm; HRMS (ESI): m/z
3
2
2 8
3
?
calcd for C H O SNa [M ? Na] 555.2604, found
(S(CH
)11CH ), 14.4 (S(CH ) CH ) ppm; HRMS (ESI):
2 3 2 9 3
2
6
44
9
?
m/z calcd for C H O SNa [M ? Na] 387.2181, found
1
8 36 5
5
55.2594.
387.2176.
General procedure for removal of acetyl groups
Evaluation of the antimicrobial activity
3
Acetylated glycoside (0.2 g) was dissolved in 5 cm
Preparation of bacterial cultures
3
MeOH, and 0.15 cm 1 M MeONa was added. The reac-
tion mixture was stirred for 16 h at r.t., filtered, and
concentrated. The crude product was purified by column
chromatography (EtOAc:MeOH 1:0 ? 10:1).
For testing of antimicrobial activity, bacteria E. coli
CNCTC 327/73, S. aureus CNCTC Mau 82/78, and yeast
C. albicans CNCTC 59/91 [all purchased from Czech
National Collection of Type Cultures, Czech Republic and
recommended for antimicrobial susceptibility [19] and
preservative efficacy tests [20] (European Pharmacopoeia)]
were used. Bacteria were grown aerobically in nutrient
broth (NB, Imuna, Slovakia) and yeast in Sabouraud dex-
trose broth (SDB, Difco, France) for 18 h at 37 °C or 48 h
at 24 °C, respectively. Cultures were then maintained at
Decyl a-D-mannopyranoside (3, C H O )
1
6 32 6
Yield 0.12 g (90 %); colorless oil; [a] = ?50.0 (c = 0.5,
D
1
methanol); H NMR (400 MHz, CD OD): d = 4.73 (d,
3
1
H, J1,2 = 1.7 Hz, H-1), 3.83 (dd, 1H, J5,6a = 2.4 Hz,
J6a,6b = 11.9 Hz, H-6a), 3.78 [dd, 1H, J2,3 = 3.3 Hz, H-2),
.76–3.67 (m, 3H, H-3, H-6b, OCH (CH ) CH ], 3.61 (t,
3
2
2 8
3
4
°C on appropriate solid medium: Endo agar (EA, Oxoid,
J3,4 = 9.5 Hz, J4,5 = 9.5 Hz, H-4), 3.52 (m, 1H, H-5),
England) for E. coli, blood agar (BA, Biomark, India) for S.
aureus, and Sabouraud dextrose agar (SDA, Difco, France)
for C. albicans. Working cultures were prepared by incu-
bation of single colony of each microorganism in NB
3
.41 [dt, 1H, J = 6.3 Hz, J = 9.6 Hz, OCH (CH ) CH ],
2 2 8 3
1
.64–1.29 [m, 16H, OCH (CH ) CH ], 0.90 [t, 3H,
2
2 8
3
1
J = 6.9 Hz, O(CH ) CH ] ppm; C NMR (100 MHz,
3
2
9
3
CD OD): d = 101.5 (C-1), 74.6 (C-5), 72.7 (C-3), 72.3 (C-
3
(
bacteria) or SDB (yeast) for 18 h at 37 °C or 48 h at
4 °C, respectively. A microbial suspension was prepared
2
3
1
), 68.6 (29) (C-4, OCH (CH ) CH ), 62.9 (C-6), 33.1,
2 2 8 3
2
0.7 (29), 30.6 (29), 30.5, 27.4, 23.7 (OCH (CH ) CH ),
3
2
2 8
in saline solution (0.85 % NaCl) according to McFarland
standard No 0.5 using Lambda 35 UV/VIS
4.4 (O(CH ) CH ) ppm; HRMS (ESI): m/z calcd for
2
9
3
?
C H O Na [M ? Na] 343.2097, found 343.2090.
16 32 6
1
23

Antimicrobial activity of mannose-derived glycosides
1713
Spectrophotometer (PerkinElmer, USA), to obtain a con-
was measured with a 96-microtitre plate computer-driven
luminometer Immunotech LM-01T (Immunotech, Prague,
Czech Republic). Light emission, expressed as relative
light units (RLU), was recorded continuously for one sec-
ond and evaluated on the basis of integral (peak) values.
Proliferation observed with unstimulated cells was con-
sidered to be the baseline value. The proliferation index
represents ratio of induced proliferation (stimulated cells)
to baseline (unstimulated cells) proliferation.
8
centration of ca 1.5 9 10 cfu/cm . After a dilution in
3
appropriate liquid medium (NB for bacteria and SDB for
7
3
yeast), a working concentration 1.5 9 10 cfu/cm was
prepared.
Antimicrobial activity assay
The antimicrobial activity assay was performed according to
Mal ´ı k et al. [21] with some modifications. Stock solutions
(
80 mM) of test mannosides and standards were prepared in
sterile 50 % MeOH (1:1 MeOH-distilled H O), immediately
2
Cell cultivation and activation
Ò
before use. Working test mannosides and standards were
prepared by serial dilution of stock solutions in sterile doubly
3
concentrated NB or SDB to a final volume 100 mm within
RAW 264.7 cell line murine macrophages (ATCC TIB-
TM
71 ) were cultured in complete DMEM for 24 h (until
approx 60 percentage confluence), after then, the aliquots
5
3
the 96-well microplates. Freshly prepared inoculum
3
5 mm ) of the tested microorganism was added to each
of 1 9 10 cells/cm /well (with 93.1 % viability) were
seeded in 24-well cell culture plate (Sigma, US), and
exposed to different concentrations of mannosides (3, 4,
(
appropriate well (bacteria into the plates with NB, C. albi-
cans into the plates with SDB). The final concentration of
3
10, and 12), ranged from 1 to 100 lg/cm /well. In vitro
exposure has been performed for next 24 h, respectively, at
5
each organism in each well was ca. 7.5 9 10 cfu/cm . The
3
concentration of tested compounds ranged from 40 mM to
37 °C in an atmosphere of 5 % CO and at 90–100 %
2
0
.3 lM, except of mannosides 5 and 6, which were tested in
relative humidity. Cell viability has been assessed using the
the range from 0.625 mM to 0.3 lM due to their low solu-
bility under experimental conditions used. Each
concentration was assayed in triplicates. For each test com-
pound and microorganism, the following controls were used:
blank, uninoculated media without test compound to account
for changes in the media during the experiment; negative
control, uninoculated media containing only the test com-
pound; positive control 1, inoculated media without
compound; positive control 2, inoculated media without
compound but including mannose to evaluate any effect of
the sugar alone; and the positive control 3, inoculated media
with serial dilution of 50 % MeOH without test compound,
thereby assessing any activity of the alcohol. The 96-well
plates were incubated aerobically for 24 h at 37 °C or 24 °C,
trypan blue exclusion method.
Computational and statistical analysis
Results from in vitro experiments were calculated as mean
values with ± standard error of the mean (s.e.m.). Nor-
mality of distribution was evaluated according to Shapiro–
Wilk’s test at the 0.05 level of significance. Statistical
comparison between individual groups was performed
using one-way analysis of variance (ANOVA) and post hoc
Bonferroni’s test. The results were significant if the dif-
ference between the analyzed groups equaled or exceeded
the 95 % confidence level (P \ 0.05). Statistics were per-
formed with Origin 7.5 Pro software (OriginLab
Corporation, Northampton, MA, USA).
3
in dependence if bacteria or yeast were grown. Then, 5 mm
of each well was inoculated on the appropriate agar plate (EA
for E. coli, BA for S. aureus, and SDA for C. albicans).
Bacteria and yeast were grown aerobically 24 h at 37 °C or
Acknowledgments This work was supported by the Slovak
Research and Development Agency (the project APVV-0484-12),
Scientific Grant Agency of the Ministry of Education of Slovak
Republic and Slovak Academy of Sciences (the project VEGA-2/
0064/15), and the Slovak State Programme Project No.
2
4 °C, respectively. The MIC was defined as the lowest
concentration of compound that inhibited the growth of
microorganism on agar plates for all parallel samples com-
pared to the positive control after 24 h.
2
003SP200280203. Dr. Vladim ´ı r Puchart is appreciated for helpful
discussion.
Immunobiological experiments
References
Proliferation
1
. K rˇ en V, Mart ´ı nkov a´ L (2001) Curr Med Chem 8:303
2. Grynkiewicz G, Szeja W, Boryski J (2008) Acta Pol Pharm
5:655
Proliferation and cytotoxicity of mammalian cell line RAW
6
2
64.7 in culture was evaluated based on the bioluminescent
3
. Devulapalle K, de Sugura AG, Ferrer M, Alcalde M, Mooser G,
Plou FJ (2004) Carbohydr Res 339:1029
measurement of adenosine triphosphate (ATP) levels,
TM
marker of metabolically active cells, using the ViaLight
4
. Ferrer M, Soliveri J, Plou FJ, L o´ pez-Cort e´ s N, Reyes-Duarte D,
Christensen M, Copa-Pati n˜ o J, Ballesteros A (2005) Enzyme
Microb Technol 36:391
plus kit (Lonza, Walkersville, Maryland, US) according to
the manufacturer’s instructions. The emitted light intensity
123

1
714
A. Bilkov a´ et al.
ˇ
ˇ
5
6
. Habulin M, Sabeder S, Knez Z (2008) J Supercrit Fluids 45:338
. Silva FVM, Goulart M, Justino J, Neves A, Santos F, Caio J,
Lucas S, Newton A, Sacoto D, Barbosa E, Santos MS, Rauter AP
14. Smith P, Nobmann P, Henehan G, Bourke P, Dunne J (2008)
Carbohydr Res 343:2557
15. Kabara JJ (2004) Mothers milk, the first nutraceutical. Presenta-
tion at Autism One, ‘‘Annual Autism Conference’’, May 27–30,
2004, Chicago
16. Limousin C, Petit LL (1997) J Carbohydr Chem 16:327
17. Klotz W, Schmidt RR (1993) Liebigs Ann Chem 6:683
18. Son SH, Tano C, Furuike T, Sakairi N (2009) Carbohydr Res
344:285
19. Clinical and Laboratory Standards Institute (2005) CLSI Methods
for determining bactericidal activity of antimicrobial agents;
Approved guideline CLSI M26-A. CLSI, Wayne
20. European Pharmacopoeia Commision (1997) Efficacy of
antimicrobial preservation. In: European Pharmacopoeia, 3rd
edn. Strasbourg, EP 2.6.12
(
2008) Bioorg Med Chem 16:4083
7
. Rauter AP, Lucas S, Almeida T, Sacoto D, Ribeiro V, Justino J,
Neves A, Silva FVM, Oliveira MC, Fereira MJ, Santos MS,
Barbosa E (2005) Carbohydr Res 340:191
. Matsumura S, Imai K, Yoshikawa S, Kawada K, Uchibori T
1990) J Am Oil Chem Soc 67:996
8
9
(
. Rybinski W, Hill K (1998) Angew Chem Int Ed 37:1328
1
1
0. Hill K, Rhode O (1990) Fett/Lipid 1:25
1. Holmberg K (2003) Novel surfactants. Preparation, application
and biodegradability, 2nd edn. Surfactant Science Series, vol 114.
Marcel Dekker, New York
2. Gareth T (2002) Medicinal chemistry—an introduction. Wiley,
West Sussex, p 125
1
1
21. Mal ´ı k I, Bukovsk y´ M, Andriamainty F, Gali sˇ inov a´ J (2012) Braz
J Microbiol 43:959
3. Pol a´ kov a´ M, Bel a´ nˇ ov a´ M, Petru sˇ L, Miku sˇ ov a´ K (2010) Carbo-
hydr Res 345:1339
1
23