New chromogenic substrates for feruloyl esterases

Article abstract of DOI:10.1039/b717742a

Indolyl and nitrophenyl 5-O-hydroxycinnamoyl-α-l-arabinofuranosides were prepared by chemo-enzymatic syntheses. These probes were designed as substrates to be used in assays of feruloyl esterase activity (EC 3.1.1.77). Color development in the assays only

Full text of DOI:10.1039/b717742a

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
New chromogenic substrates for feruloyl esterases
Laurence Marmuse,a Miche`le Asther,b Emeline Fabre,c David Navarro,b Laurence Lesage-Meessen,b
Marcel Asther,b Michael O’Donohue,c Se´bastien Forta and Hugues Driguez*a
Received 16th November 2007, Accepted 5th February 2008
First published as an Advance Article on the web 29th February 2008
DOI: 10.1039/b717742a
Indolyl and nitrophenyl 5-O-hydroxycinnamoyl-a-L-arabinofuranosides were prepared by
chemo-enzymatic syntheses. These probes were designed as substrates to be used in assays of feruloyl
esterase activity (EC 3.1.1.77). Color development in the assays only occurs when feruloyl esterase
activity releases an intermediate chromogenic arabinoside that is a suitable substrate for
a-L-arabinofuranosidase (EC 3.2.1.55), which in turn releases the free chromogenic group. The
usefulness of these compounds was evaluated in both qualitative solid media-based assays and
quantitative liquid assays that can be performed in microtiter plates using feruloyl esterases and
arabinofuranosidases from various origins.
Recently, the synthesis of 4-nitrophenyl 2- and 5-O-trans-feruloyl-
Introduction
a-L-arabinofuranosides was reported.8 The use of this substrate
Plant cell walls are mainly composed of cellulose microfibrils
relies on a tandem reaction involving the splitting of the ester bond
embedded in a matrix of hemicelluloses, pectins and lignins.
and almost concomitant release of 4-nitrophenol through addition
of a-L-arabinofuranosidase, which is added in large quantity.
hardwood, accounting for as much as 35% of the dry weight of
However, for routine use the addition of a detergent is required
Xylans are the most abundant hemicelluloses in cereal and
higher plants. Their repeating xylosyl units are generally linked
to improve solubility and assays are necessarily discontinuous, be-
cause basic pH conditions are required for accurate measurement
of 4-nitrophenolate absorbance. Therefore, 4-nitrophenyl 2- and 5-
O-trans-feruloyl-a-L-arabinofuranosides are not particularly con-
venient substrates for high-throughput screening assays, where it
is desirable to minimize the number of steps.
In the 1990s, numerous chromogenic and fluorogenic com-
pounds were synthesized for the detection of hydrolytic activities of
through b(1→4) glycosidic bonds and can be mainly substituted at
O-2 or O-3 positions by esters such as acetyl, feruloyl, p-coumaroyl
and D-galactosyl, L-arabinosyl or D-glucuronic acid residues.1 In
arabinoxylans, some L-arabinofuranosyl residues can be esterified
at O-5 by phenolic acids.2,3 Among these, trans-ferulic acid is
particularly important because it participates in cross-linking
between hemicelluloses, pectins and lignins via ester and ether
bonds.4 The resulting network enhances the mechanical strength
enzymes acting in large range of pH. For instance, Claeyssens et al.
demonstrated that 2-chloro-4-nitrophenyl and 4-methylumbel-
liferyl b-cello-oligosaccharides offer good sensitivity and stability
of cell walls and increases their resistance towards cellulases and
xylanases. To counter this cross-linking, many lignocellulolytic
for continuous assays of various cellulases.9 Similarly, 2-chloro-
4-nitrophenyl b-maltoheptaoside was prepared and included in a
commercial kit for a-amylase determination.10 The advantage of
these substrates resides in the fact that, unlike 4-nitrophenol whose
molar extinction coefficient is halved at pH 7 compared to that at
pH 9, the molar extinction coefficient of 2-chloro-4-nitrophenol
is almost pH-independent. Consequently, the sensitivity of ab-
sorbance measurements at neutral pH is increased and continuous
absorbance measurements are possible. Another important step
was the development of 5-bromo-4-chloro-3-indolyl glycosides
for assays in solid media. The well known 5-bromo-4-chloro-3-
indolyl b-galactoside (X-Gal) has proved to be an essential tool
in molecular biology for the detection of b-gal+ microrganisms
grown on solid agar media. This glycoside is colorless, but the
free aglycon is rapidly air-oxidized to form an indigo blue colored
dimer. Likewise, microorganisms expressing b-galactosidase grow
as indigo-blue-colored colonies on agar plates.11
microorganisms produce feruloyl esterases (FAE) that catalyze
the cleavage of ester linkages between ferulic acid and sugars.
Likewise, FAE act in synergy with other debranching enzymes
and depolymerizing cellulases and xylanases in order to bring
about complete degradation of biomass.5 Presently, the growing
interest in the development of industrial bioprocessing strategies
using engineered lignocellulolytic microorganisms is providing an
impetus to research aimed at improving the intrinsic properties of
FAE. Most FAE from eukaryotic and prokaryotic sources belong
to family 1 of the carbohydrate esterase classification.6 Recently,
on the basis of substrate specificities and growth requirements
of the producer microorganisms, this family has been divided
into four functional classes A–D. So far, FAE activity has been
monitored using a variety of natural and synthetic compounds.7
aCentre de Recherches sur les Macromole´cules Ve´ge´tales (CERMAV-
CNRS), BP53, 38041, Grenoble cedex 9, France. E-mail: Hugues.Driguez@
cermav.cnrs.fr; Fax: +33 476 547 203; Tel: +33 476 037 664
In the present work, we have designed new chromogenic
substrates that provide assays for feruloyl esterase activity either
on solid agar plate media or in continuous liquid-based reactions.
Because fluorogenic substrates require equipment that is not
always available in laboratories, especially on high-throughput
bUMR-1163 INRA de Biotechnologie des Champignons Filamenteux,
IFR86-BAIM. Universite´s de Provence et de la Me´diterrane´e, ESIL, 163
avenue de Luminy, CP 965, 13288, Marseille Cedex 09, France
cINSA/INRA UMR 792, 135 avenue de Rangueil, 31077, Toulouse cedex
04, France
1208 | Org. Biomol. Chem., 2008, 6, 1208–1214
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
platforms, we have concentrated on more universally applicable
substrates based on indolyl and 2-chloro-4-nitrophenyl moieties.
arabinofuranosidases, which was fixed at a value of 5.5. Raising
the pH to 6 (pH of buffer for FAEA, Fig. 1, well 3) diminished this
effect. Nevertheless, a very clear positive blue/indigo coloration
was obtained when substrates 1a or 1b were incubated with FAE
and AN-ABF or BL-ABF (Fig. 1, well 4). This visual change was
correlated with a substantial increase in absorbance at 620 nm,
an effect that was totally absent when only FAEA was deployed
(Fig. 1, well 3).
Results and discussion
Our first target molecules were the 5-bromo-4-chloro-3-indolyl
5-O-trans-feruloyl-a-L-arabinofuranoside 1a and the 5-bromo-4-
chloro-3-indolyl 5-O-trans-coumaroyl-a-L-arabinofuranoside 1b
(Scheme 1)
Fig. 1 Detection of combined action of a-L-arabinofuranosidase and
FAEA activities in the presence of substrate. Only reactions using BL-ABF
are shown. 1a. Well 1, substrate 1a alone; Well 2, substrate 1a +BL-ABF;
Well 3, substrate 1a + FAEA; Well 4, substrate 1a + BL-ABF + FAEA.
In a further test, the suitability of 1a for use in solid media for
the detection FAEA+ microorganisms was studied. Fig. 2 shows
that when BL-ABF is applied to solid LB agar medium containing
substrate 1a, upon which Escherichia coli FAEA+ cells have been
grown, a blue color is developed in the vicinity of the application
zone. This blue coloration is the result of three successive reactions:
i) the FAEA-catalyzed cleavage of the ester bond and the liberation
of the 5-bromo-4-chloroindoxyl a-L-arabinofuranoside, ii) the BL-
ABF-catalyzed cleavage of the pseudo-glycosidic bond and the
release of 5-bromo-4-chloroindoxyl-3-ol, and iii) the spontaneous
oxidation of the 5-bromo-4-chloroindoxyl-3-ol. It is important to
note that in Fig. 2 the blue color is quite diffuse. This is because
BL-ABF was applied as a liquid sample after bacterial growth and
expression of FAEA. In order to obtain single colony coloration it
would probably be necessary to create a tandem expression system
that would provide simultaneous intracellular production of both
a-L-arabinofuranosidase and FAEA or more simply to use a sterile
top agar containing a-L-arabinofuranosidase.
Scheme 1 Reagents and conditions: i) 1-acetyl-5-bromo-4-chloroindoxyl-
3-ol 3, NIS, Sn(OTf)2, CH2Cl2, 4 58%; ii) lipase Novozym 435, i-PrOH, 5
75%; iii) pyridine, DMAP, 4-acetoxyferuloyl chloride, 6a 82%, 4-acetoxy-
coumaroyl chloride, 6b 80%; iv) NaOMe, MeOH, 1a 74%, 1b 86%.
The synthesis of 5-bromo-3-indolyl a-L-arabinofuranoside in
48% yield was previously described by Berlin and Sauer starting
from benzoylated arabinosyl bromide as donor.12 Due to the
known instability of furanosyl halides, we decided to investigate
the glycosyl donor property of the acetylated thiophenyl arabi-
nofuranoside 2, a compound recently described as an efficient
glycoside donor13 (Scheme 1).
Coupling of the thiosugar
2 with 1-acetyl-5-bromo-4-
chloroindoxyl-3-ol 3 was achieved in the presence of thiophilic
promoter system NIS/Sn(OTf)2 in DCM at −10 ◦C. The expected
compound 4 was obtained in 58% yield. Compound 5, prepared
in high yield from 4 by selective O-deacetylation of its primary hy-
droxyl group using the Candida antarctica lipase Novozym 435 was
esterified with 4-acetoxyferuloyl chloride14 or acetoxycoumaroyl
chloride15 to give 6a or 6b in 82 and 80% yield respectively.
Catalytic O-deacetylation afforded the target compounds 1a and
1b, which are potentially useful substrates for assaying FAE and,
perhaps, for differentiating between enzymes that are more specific
towards methoxylated or hydroxylated hydroxycinnamic acids.
Compounds 1a and 1b were evaluated as substrates for FAEA
and FAEB from Aspergillus niger. First, a spectrophotometric
qualitative assay was developed using commercially available a-
L-arabinofuranosidase from Aspergillus niger (AN-ABF) and a-
L-arabinofuranosidase from Bifidobacterium longum (BL-ABF)
(Fig. 1). Both substrates were readily soluble in methanol or
DMSO and were stable for several hours in the assay conditions.
Moreover, controls indicated that in the absence of FAE, a-L-
arabinofuranosidase was not able to cleave 1a or 1b, although
a yellow coloration was observed (Fig. 1, well 2). This artifact
was caused by the pH of the reaction medium used for the a-L-
Fig. 2 Detection of FAEA activity on solid LB agar medium using the
chromogenic substrate 1a. The FAE+ microorganism was Escherichia coli
bearing a plasmid that expresses a thioredoxin–FAEA fusion protein.
In a second series of experiments, a structural analogue of
1a/1b in which the indolyl aglycon was replaced by 2-chloro-
4-nitrophenol was prepared. Unfortunately, we were unable to
condense the thioglycoside 2 with 2-chloro-4-nitrophenol, despite
the use of several promoters. Therefore, the synthesis of 7
(Scheme 2) was carried out essentially as recently described for
the preparation of 4-nitrophenyl a-L-arabinofuranoside.16
Glycosylation of acylated furanose
8 with 2-chloro-4-
nitrophenol was catalyzed by Lewis acid and gave benzoylated
glycoside 9 in 48% yield. O-Debenzoylation with 1 M NaOMe
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 1208–1214 | 1209
©

View Article Online
Scheme 2 Reagents and conditions: i) BF3·Et2O, CH2Cl2, 2-chloro-4-
nitrophenol, 61%; ii) NaOMe, MeOH, 72%; iii) Ac2O, pyridine, 93%;
iv) lipase Novozym 435, i-PrOH, 96%; v) Et3N, DMAP, 4-acetylferuloyl
chloride, CH2Cl2, 66%; vi) K2CO3, DCM–MeOH, 64%.
in MeOH afforded the expected compound 10 in 72% yield
after column chromatography on silica. O-Acetylation gave the
fully acetylated derivative 11, which was treated by the Candida
antarctica lipase to selectively remove the primary acetate. The
deprotected derivative 12 was obtained in 89% overall yield.
Esterification of the primary hydroxyl using 4-acetoxyferuloyl
chloride was achieved in 66% yield. De-O-acetylation of 13 was
a delicate issue, and several attempts were needed before the
isolation of the target molecule 7 in reasonably good yield (64%).
Compound 7 proved to be a suitable substrate for continuous
monitoring of FAE activity. This procedure involved a two-
step enzymatic process: first the enzymatic splitting of the ester
function by FAE, followed by the liberation of the 2-chloro-4-
nitrophenol through the action of an a-L-arabinofuranosidase. To
ascertain the usefulness of 7, the combined hydrolytic activity of
FAE and AN-ABF were evaluated by monitoring the evolution
of the substrate’s absorption spectra over time. At the end of
the reaction period, the spectra clearly indicated characteristic
adsorption peaks for free ferulic acid at 288 nm and 310 nm7 and
a peak at 405 nm, corresponding to free 2-chloro-4-nitrophenol
(Fig. 3). Importantly, because the absorbance of 2-chloro-4-
nitrophenol at 405 nm is sufficiently high at neutral pH, direct
Fig. 3 Changes in absorption spectra during hydrolysis of compound 7
by FAEA from A. niger according to time. FAEA (33 nM) 32 mU ml−1
(using MFA as substrate) was incubated with compound 7 (41.5 lM) in
100 mM MOPS Buffer pH 6.0 and AN-ABF (250 mU ml−1). Spectra were
recorded according to time every 2 minutes. ꢀ–ꢀ 0 min;
2 min; ꢀ–ꢀ
4 min; ᭹–᭹ 6 min; --- 8 min; ꢁ–ꢁ 10 min.
measurement without alkalinization of the medium is possible.
Therefore 7 presents at least two clear advantages: i) it can be
employed in a continuous high-throughput assay using routine
equipment, and ii) it avoids interference with the spectrum of
ferulic acid that absorbs strongly at alkaline pH. In addition, only
reasonable amounts of L-arabinofuranosidase are required since
the release of the 2-chloro-4-nitrophenol by this enzyme should
not be rate-limiting. Using an excess of L-arabinofuranosidase, a
linear relationship could be obtained between the rate of 2-chloro-
4-nitrophenol release and the amount of FAE (Fig. 4).
Catalytic parameters for FAEA and FAEB have been deter-
mined using 7, 14 (methyl ferulate) and 15 5-O-trans-feruloyl-a-L-
Araf as substrates. Both enzymes display higher catalytic efficien-
cies on the natural substrate 15 and the synthetic one 7 than on
Fig. 4 Time course of hydrolysis of compound 7 by FAEA followed by the release of chloronitrophenol at 405 nm. A: Kinetics were followed at different
FAEA concentrations: ᭺–᭺ (0.1 mU ml−1), ꢀ–ꢀ (0.2 mU ml−1); ꢂ–ꢂ (0.4 mU ml−1) in MOPS buffer pH 6.0 with 20 lM substrate and AN-ABF
250 mU ml−1 at 37 ◦C. B: Kinetics were followed in the same conditions but with FAEA (0.1 nM) using AN-ABF at different concentrations: ꢁ–ꢁ
(250 mU ml−1); ꢀ–ꢀ (500 mU ml−1); ᭺–᭺ (1 U ml−1).
1210 | Org. Biomol. Chem., 2008, 6, 1208–1214
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Table 1 Kinetic parameters for the hydrolysis of the synthesized substrate
7 (2-chloro-4-nitrophenyl 5-O-feruloyl-a-L-arabinofuranoside) compared
to 14 (methyl ferulate) and 15 (5-O-trans-feruloyl-a-L-Araf ) by the two
feruloyl esterases FAEA and FAEB from A. niger
useful for assays in solid medium where only the endpoint of the
reaction is observed. With regard to the 2-chloro-4-nitrophenyl
based substrates, these are particularly useful for one-step liquid
assays that can provide quite accurate data concerning reaction
rates. Together, these substrates will be valuable tools for a variety
of high throughput screening applications, for example for a two-
stage screening strategy involving initial agar plate screening to
eliminate inactive clones, followed by liquid screening in 96-well
microplate format to detect improved activity.
Catalytic efficiencies (kcat/Km)/M−1 s−1
Substrate
FAEA
FAEB
14
15
7
0.52 × 105
2 × 106
0.25 × 105
2.1 × 105
3.5 × 105
5 × 106
Experimental
General methods
Reactions were monitored by TLC using silica gel 60 F254
precoated plates (E. Merck, Darmstadt) and detection by charring
with sulfuric acid solution (H2SO4/MeOH/H2O 3:45:45). For
flash chromatography, Merck silica gel 60 was used. NMR spectra
were recorded on Bruker AC 300 or Bruker Avance 400 at
298 K. Proton chemical shifts are reported in ppm relative to
internal SiMe4 (0 ppm). Low-resolution FAB mass spectra were
recorded in the positive mode of an R1010C quadripolar mass
spectrometer (model 2000, Nermag, Reuil-Malmaison, France).
MALDI-TOF measurements were performed on a Bruker Dal-
tonics Autoflex apparatus. ES experiments were performed on
a Waters Micromass ZQ spectrometer. The spectrophotometric
measurements were recorded on UVIKON XS spectrometer
equipped with a thermostated cuvette holder.
the methyl ester derivative 14, thus confirming the importance of
the L-arabinofuranosyl moiety substrate recognition.17 Moreover,
no significant differences could be observed between 15 and 7,
indicating that the presence of the nitrophenyl aglycon on the
sugar does not have an inhibitory effect (Table 1). These results
are coherent with structural data that indicate that the xylosyl
moieties of natural arabinoxylans do not interact with active site
elements of FAEs.
Various strains of Aspergillus niger belonging to the CIRM-
BRFM Collection (Centre International de Ressources Micro-
biennes, Banque de Ressources Fongiques de Marseille, France)
chosen for their ability to produce different types of feruloyl es-
terases (FAEA and FAEB) were cultivated in previously described
conditions.18 The culture medium was analyzed using a TECAN
Freedom Evo automated platform. Both substrates 1a and 7 were
checked in these conditions and it was shown that they were
completely adaptable in microplate screening. Positives responses
were obtained with FAE-producing strains with both substrates.
Moreover, using compound 7 kinetics could be monitored in order
to deduce relative activity levels (results not shown).
Enzymes and standard assays
Pure feruloyl esterases FAEA and FAEB from Aspergillus niger
were obtained according to the procedures described respectively
by Record et al.19 and Levasseur et al.20 Methyl ferulate (MFA) was
purchased from Apin Chemicals and the feruloylated oligosaccha-
ride FA, 5-O-trans-feruloyl-a-L-Araf was purified according to a
known procedure.21 Esterase activities using MFA and FA were
assayed by a continuous spectrophotometric method as previously
described.7 The a-arabinofuranosidase from Aspergillus niger
was obtained from Megazyme (Bray, Ireland), whereas the a-
arabinofuranosidase from Bifidobacterium longum NCC2705 (BL-
ABF) is a recombinant His-tagged enzyme produced in-house. The
sequence encoding this enzyme was fused to the 6 × His tag in
pET28a. The resulting plasmid was used to transform Escherichia
coli BL21 DE3 cells and protein expression via the pT7 promoter
was performed according to established procedures. Likewise,
His-tagged BL-ABF was purified using His-trap chromatography
(GE Healthcare) following the manufacturer’s recommendations.
Activities of BL-ABF and AN-ABF were determined using
5 mM 4-nitrophenyl arabinofuranoside and 50 mM acetate buffer
(pH 5.8) at 37 ◦C.22 Activities were expressed in units (U), where
one unit is the amount of enzyme needed to liberate one lmol of
pNP per minute.
Conclusions
Over the last ten years, the areas of enzyme discovery and
enzyme engineering using molecular evolution technology have
strongly been developed. Likewise, high-throughput methods have
been vitally important for the screening of vast natural and
artificially generated biodiversity. Therefore, it is increasingly
important to devise enzyme assays that are both specific and simple
to implement. In this regard, soluble and stable chromogenic
substrates are highly desirable, because they are generally reliable
and quite simple to use.
Indolyl and 2-chloro-4-nitrophenyl are already present as the
aglycon group in several valuable substrates that have been widely
used in enzyme assays and various screens. In this work, we
have used these two familiar aglycons in order to design new
chromogenic compounds with great potential for the screening
of FAE activities. The syntheses that are described are quite
straightforward and employ enzymatic de-O-acetylation catalyzed
by Candida antarctica lipase. These probes are more water-soluble
than the corresponding 4-nitrophenyl derivatives, allowing their
use in a microtiter plate format without any use of detergent.
According to our evaluations, the indolyl-based compounds are
unsuitable for comparative activity assays, but are actually very
Detection of FAE activity using synthetic substrates
For the evaluation of the usefulness of substrate 1a for the
detection of FAE activity on solid media, Escherichia coli TOP10
cells bearing a plasmid derived from pBAD-Thio (Invitrogen, NL)
were used. In this plasmid, the FAEA coding sequence was fused
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 1208–1214 | 1211
©

View Article Online
=
to that of thioredoxin. After plating on LB agar plates containing
ampicillin (100 lg ml−1) and substrate 1a (60 lg ml−1) bacterial
colonies were allowed to develop overnight at 30 ◦C. After, a
10 lL aliquot (3 U) of BL-ABF was carefully applied to the agar
plate, which was incubated at 37 ◦C for 1 h in order to allow color
development.
H-5a, H-5b), 2.58 (3H, s, N–CH3C O), 2.15, 2.14 (6H, 2 × s,
13
=
CH3C O); C NMR (75 MHz, CDCl3): d 170.9, 169.8, 168.4,
=
(CH3C O), 139.6–109.2–139 (C-indoxyl), 105.4 (C-1), 85.2, 81.0,
76.8 (C-2, C-3, C-4), 62.0 (C-5), 24.0 (N-CH3C O), 20.95, 20.9,
=
+
+
=
(2 × CH3C O); m/z (ES) 521 [M + NH4] , 526 [M + Na] ; HRMS
calcd for C19H19NO8ClBrNa+ 525.9880; found 525.9874.
For 2-chloro-4-nitrophenyl 5-O-feruloyl-a-L-arabinofurano-
side, a stock solution of the substrate (20 mM) was prepared
in methanol. The incubation mixture contained 100 mM MOPS
buffer pH 6.0, 0.25 U a-L-arabinofuranosidase, substrate concen-
trations ranging from 1 to 50 lM and appropriate amounts of
FAEA or FAEB. Activities were followed at 37 ◦C by the release
of 2-chloro-4-nitrophenol at 405 nm according to time.
Assays using a high-throughput screening platform were per-
formed in 96-well microtiter plates using experimental conditions
identical to those previously described, but using a 200 lL
final volume. Additionally, culture supernatants were used as the
enzyme source for screening in the place of pure FAE.
N-Acetyl-5-bromo-4-chloro-3-indolyl 5-O-(4-acetoxyferuloyl)-
2,3-di-O-acetyl-a-L-arabinofuranoside 6a
To a solution of 5 (108 mg, 0.21 mmol) in pyridine (5 mL) were
added DMAP (5 mg, 0.04 mmol) and 4-acetoxyferuloyl chloride
(216 mg, 0.85 mmol). The reaction was stirred for 4 h at room
temperature and diluted with CH2Cl2, washed with H2O and the
organic phase was dried over Na2SO4. The crude product was
purified by flash column chromatography (toluene–ethyl acetate,
9:1→85:15 v/v) to give the title compound 6a as a white foam
1
(102 mg, 67%); [a]D −23 (c 0.5 in CHCl3); H NMR (400 MHz,
CDCl3): d 8.25 (1H, d, J 8 Hz, H-Ar), 7.68 (1H, d, J 16 Hz,
=
HC CH), 7.02–7.55 (5H, m, H-Ar), 6.40 (1H, d, J 14 Hz,
Phenyl 2,3,5-tri-O-acetyl-1-thio-a-L-arabinofuranoside 2
=
HC CH), 5.65 (1H, s, J1,2 <1 Hz, H-1), 5.43 (1H, J2,3 <1 Hz, H-2),
The title compound was prepared as recently described.13
5.16 (1H, d, J4,3 4 Hz, H-3), 4.58 (1H, dd, J4,5b 3.5 Hz, J5a,5b 11 Hz,
H-5b), 4.53 (1H, m, H-4), 4.51 (1H, dd, J4,5a 5.3 Hz, J5a,5b 11 Hz,
=
H-5a), 3.84 (3H, OMe), 2.55 (3H, s, CH3C O), 2.30 (3H, s, N-
N-Acetyl-5-bromo-4-chloro-3-indolyl 2,3,5-tri-O-acetyl-a-L-
arabinofuranoside 4
CH3C O), 2.13, 2.12 (6H, 2 × s, CH3C O); 13C NMR (75 MHz,
=
=
=
CDCl3): d 170.4, 169.6, 168.8, 168.3, 166.4 (C O), 151.6–109.2
(C-ferulate, C-indolyl), 105.4 (C-1), 82.7, 80.7, 76.9 (C-2, C-3, C-
The compounds 2 (225 mg, 0.60 mmol) and 3 (210 mg, 0.73 mmol)
were stirred in dry CH2Cl2 (4 mL) under Ar at −10 ◦C in presence
=
4), 63.3 (C-5), 56.0 (OMe), 24.0 (N–CH3C O), 20.9 (2x), 20.8,
˚
of activated 4 A molecular sieves (200 mg) for 10 min. Then
+
=
(CH3C O); m/z (MALDI) 744.02 [M + Na] ; HRMS calcd for
N-iodosuccinimide (165 mg, 0.73 mmol) and Sn(OTf)2 (50 mg,
0.122 mmol) were successively added to the reaction. After 20 min
the reaction was neutralized with Et3N. The mixture was filtered
through Celite and the filtrate was diluted with CH2Cl2 and
washed successively with a Na2S2O3 solution, then with water,
the organic phase was dried over Na2SO4. The crude product was
purified by flash column chromatography (toluene–ethyl acetate,
9:1→8:2 v/v) to give the title compound 4 as an amorphous
solid (204 mg 58%); m.p. 158–160 ◦C (MeOH); [a]D −49 (c
C31H29NO12ClBrNa+ 744.0459; found 744.0476.
N-Acetyl-5-bromo-4-chloro-3-indolyl 5-O-(4-acetoxycoumaroyl)-
2,3-di-O-acetyl-a-L-arabinofuranoside 6b
To a solution of 5 (115 mg, 0.23 mmol) and DMAP (5 mg,
0.04 mmol) in pyridine (5 mL), 4-acetoxy-coumaryl chloride
(160 mg, 0.69 mmol) was added portionwise. The reaction was
stirred for 2.5 h, diluted with CH2Cl2, washed with sat. aq.
NaHCO3 solution and H2O, and dried over Na2SO4. The crude
product was purified by flash column chromatography (toluene–
ethyl acetate, 9:1 v/v) to give the title compound 6b as a white
foam (130 mg, 80%); [a]D −28 (c 0.53 in CHCl3); 1H NMR
(300 MHz, CDCl3): d 8.24 (1H, d, J 8.8 Hz, H-Ar), 7.72 (1H, d, J
1
0.5 in CHCl3); H NMR (400 MHz, CDCl3): d 7.17–8.29 (3H,
m, H-indoxyl), 5.65 (1H, s, J1,2 <1 Hz, H-1), 5.44 (1H, d, J2,3
0.7 Hz, H-2), 5.11 (1H, d, J3,4 4 Hz, H-3), 4.31–4.50 (3H, m, H-
=
4, H-5a, H-5b), 2.59 (3H, s, N–CH3C O), 2.13, 2.15, 2.17 (9H,
3 × s, CH3C O); 13C NMR (75 MHz, CDCl3): d 170.70, 170.4,
=
=
=
169.7, 168.4 (CH3C O), 139.7–109.2 (C-indolyl), 105.5 (C-1),
82.7, 81.0, 76.8 (C-2, C-3, C-4), 63.2 (C-5), 24.1 (N–CH3C O),
16 Hz, HC CH), 7.54–7.11 (6H, m, H-Ar), 6.42 (1H, d, J 16 Hz,
HC CH), 5.67 (1H, s, J1,2 <1 Hz, H-1), 5.45 (1H, s, J2,3 <1 Hz,
H-2), 5.19 (1H, d, J4,3 4 Hz H-3), 4.49–4.63 (3H, m, H-4, H-5a, H-
=
=
+
=
20.9 (3 × CH3C O); m/z (FAB) 568 [M + Na] ; HRMS calcd for
13
C21H21NO9ClBrNa+ 567.9985; found 567.9972.
5b), 2.56 (3H, s, N-CH3C O), 2.30, 2.15 (6H, 2 × s, CH3C O);
=
=
C NMR (75 MHz, CDCl3): d 170.3, 169.6, 169.1, 168.3, 166.3
=
(C O), 152.4–109.2 (C-ferulate, C-indolyl), 105.4 (C-1), 82.7,
N-Acetyl-5-bromo-4-chloro-3-indolyl 2,3-di-O-acetyl-a-L-
arabinofuranoside 5
=
80.7, 76.8 (C-2, C-3, C-4), 63.2 (C-5), 24.0 (N-CH3C O), 21.2,
+
+
=
20.9 (×2) (CH3C O); m/z (ESI): 714.1 [M + Na ], 730.0 [M + K ];
To compound 4 (150 mg, 0.27 mmol) suspended in isopropanol
(10 mL) was added lipase Novozym 435 (1.5 g). The reaction was
slowly agitated at 40 ◦C for 8 h. The mixture was filtered, the solu-
tion was evaporated and the crude product was purified by column
chromatography (petroleum ether–ethyl acetate, 1:1→3:7 v/v) to
give the title compound 6 as a white foam (134 mg, 98%); [a]D −51
HRMS calcd for C30H27NO11ClBrNa+ 714.0353; found 714.0370.
5-Bromo-4-chloro-3-indolyl 5-O-feruloyl a-L-arabinofuranoside 1a
The compound 6a (100 mg, 0.14 mmol) was suspended in dry
MeOH (2 mL). The mixture was cooled with an ice-water bath
and 1 M NaOMe (150 lL) was added. After 2 h, the reaction was
neutralized with Amberlite H+ and filtered. The crude product
was purified by flash column chromatography (CH2Cl2–MeOH,
1
(c 0.45 in CHCl3); H NMR (300 MHz, CDCl3): d 8.25–7,16 (3
H, m, H-indolyl), 5.64 (1H, s, J1,2 <1 Hz, H-1), 5.47 (1H, d, J2,3
<1 Hz, H-2), 5.15 (1H, d, J3,4 4 Hz, H-3), 4.37–3.86 (3H, m, H-4,
1212 | Org. Biomol. Chem., 2008, 6, 1208–1214
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
95:5 v/v) to give the title compound 1a as a yellowish foam (57 mg,
74%); [a]D −85 (c 0.54 in CH3OH); 1H NMR (400 MHz, CD3OD):
as a yellowish solid (360 g, 72%); [a]D −201 (c 0.2 in CH3OH); 1H
NMR (300 MHz, CD3OD): d 8.23 (1H, d, J 2.9 Hz, H-Ar), 8.12
(1H, dd, J 9.3 Hz, J 2.9 Hz, H-Ar), 7.4 (1 H, d, J 9.3 Hz, H-Ar),
4.39 (1H, dd, J1,2 1.5 Hz, J2,3 3.9 Hz, H-2), 5.70 (1 H, d, J1,2, H-1),
4.01–4.12 (2H, m, H-3, H-4), 3.79 (1H, dd, J5,4 2.9 Hz, H-5b), 3.67
(1H, dd, J5,4 4.8 Hz, J5a,5b 12.2 Hz, H-5a); 13C NMR (75 MHz,
CD3OD): d 117.1–159.2 (C-Ar), 108.8 (C-1), 87.0, 83.8, 78.2 (C-2,
C-3, C-4), 62.6 (C-5); m/z (ES) 328 [M + Na]+; HRMS calcd for
C11H12NO7ClNa+ 328.0200; found 328.0106.
=
d 7.65 (1H, d, J 16 Hz, HC CH), 7.28 (1H, d, J 8 Hz, H-Ar),
7.17–7.05 (4H, m, H-Ar), 6.80 (1H, d, J 8 Hz, H-Ar), 6.40 (1H,
=
d, J 16 Hz, HC CH), 5.37 (1H, s, J1,2 <1 Hz, H-1), 4.48 (1H,
dd, J4,5a 2.8 Hz, J5a,5b 11.4 Hz, H-5b), 4.41–4.32 (3H, m, H-2, H-4,
H-5), 4.03 (1H, dd, J 3.7 Hz, J 6.2 Hz, H-3), 3.87 (3H, s, OMe);
13
=
C NMR (75 MHz, CD3OD): d 169.0 (C O), 150.5–111.4 (C-1,
C-ferulate, C-indolyl), 83.6, 83.3, 79.5, 64.9 (C-2, C-3, C-4, C-
5), 56.4 (OMe); m/z (ESI): 576.0 (M + Na+); HRMS calcd for
C23H21NO8ClBrNa+ 576.0036; found 576.0036.
2-Chloro-4-nitrophenyl 2,3,5-tri-O-acetyl-a-L-arabinofuranoside
11
5-Bromo-4-chloro-3-indolyl 5-O-coumaroyl a-L-arabinofuranoside
1b
2-Chloro-4-nitrophenyl a-L-arabinofuranoside 10 (360 mg,
1.18 mmol) was dissolved in pyridine (5 mL). The reaction was ice-
cooledandacetic anhydride (1mL) was added tothe mixture. After
addition, the reaction was stirred for 3 h at room temperature.
After concentration in vacuo and coevaporation with toluene,
the residue was purified by silica gel chromatography (petroleum
ether–ethyl acetate, 8:2 v/v) to afford the title compound 11 as
The compound 6b (110 mg, 0.16 mmol) was suspended in MeOH
(3 mL). The mixture was cooled with an ice-water bath and 1 M
NaOMe (170 lL) was added. After 90 min, the reaction was
neutralized with Amberlite H+ and filtered. The crude product
was purified by flash column chromatography (CH2Cl2–MeOH,
95:5 v/v) to give the title compound 1b as a yellowish foam
1
a syrup (470 mg, 93%); [a]D −121 (c 0.5 in CHCl3); H NMR
1
(72 mg, 86%); H NMR (300 MHz, CD3OD): d 7.58 (1H, d, J
(300 MHz, CDCl3): d 8.24 (1H, d, J 2.4 Hz, H-Ar), 8.08 (1H, dd, J
9.3 Hz, J 2.4 Hz, H-Ar), 7.23 (1H, d, J 9.3 Hz, H-Ar), 5.83 (1H, s,
H-1), 5.41 (1H, d, J2,3, H-2), 5.07 (1H, dd, J2,3 1 Hz, J3,4 3.8 Hz,
H-3), 4.37–4.40 (2H, m, H-4, H-5b), 4.23 (1H, dd, J4,5 6.8 Hz, J5a,5b
=
16 Hz, HC CH), 6.73–7.37 (7H, m, H-Ar), 6.28 (1H, d, J 16 Hz,
=
HC CH), 5.32 (1H, s, J1,2 <1 Hz, H-1), 4.44–4.25 (4H, m, H-2,
H-4, H-5a, H-5b), 3.97 (1H, dd, J 3.4 Hz, J 5.8 Hz, H-3), 13C
NMR (75 MHz, CD3OD): d 169.0 (C O), 161.3–111.4 (C-1, C-
12.1 Hz, H-5a), 2.12, 2.11, 2.06 (12 H, 3 × s, 3 × CH3C O); 13C
=
=
=
ferulate, C-indolyl), 83.7, 83.4, 79.6, 65.04 (C-2, C-3, C-4, C-5);
m/z (ESI): 546.0 [M + Na+], 562.0 [M + K+]; HRMS calcd for
C22H19ClBrNa+ 545.9931; found 545.9928.
NMR (75 MHz, CDCl3): d 170.3, 170.0, 169.4 (3 × CH3C O),
156.4–115.2 (C-Ar), 103.9 (C-1), 82.8, 80.5, 76.4 (C-2, C-3, C-4),
+
=
62.7 (C-5), 20.56 (×3) (CH3C O); m/z (ES) 453.9 [M + Na] ;
HRMS calcd for C17H18NO10ClNa+ 454.0517; found 454.0524.
2-Chloro-4-nitrophenyl 2,3,5-tri-O-benzoyl-a-L-arabinofuranoside
9
2-Chloro-4-nitrophenyl 2,3-di-O-acetyl-a-L-arabinofuranoside 12
To a solution of 8 (2.6 g, 5.17 mmol) in dry CH2Cl2 (35 mL)
containing activated 4 A molecular sieves (2.5 g) was added 2-
The compound 11 (470 mg, 1.09 mmol) was dissolved in iso-
propanol (25 mL) and lipase Novozym 435 (4.7 g) was added.
The reaction mixture was agitated at 120 rpm and at 37 ◦C for
48 h. The crude product was purified by column chromatography
(petroleum ether–ethyl acetate, 3:2 v/v) to give the title compound
˚
chloro-4-nitrophenol (1.80 g, 5.17 mmol). The reaction was then
cooled to −5 ◦C and BF3·Et2O (3.2 mL, 25.85 mmol) was added
dropwise to the mixture. After 45 min the reaction was warmed up
to room temperature and stirred for 6 h. The mixture was filtered
through Celite, and the solution was diluted with CH2Cl2, washed
successively with sat. aq. NaHCO3 solution and H2O, and dried
with Na2SO4. The crude product was purified by flash column
chromatography (petroleum ether–ethyl acetate, 8:2 v/v) to give
the title compound 9 as a white foam (1.56 g, 49%); [a]D −46 (c
0.5 in CHCl3); 1H NMR (300 MHz, CDCl3): d 7.58–8.64 (18H, m,
H-Ar), 6.56 (1H, s, H-1), 6.22 (1H, s, H-2), 6.06 (1H, d, J3,4 2.9 Hz,
H-3), 5.18–5.04 (3H, m, H-5, H-5ꢁ, H-4); 13C NMR (75 MHz,
1
12 as a syrup (410 mg, 96%); [a]D −77 (c 0.53 in CHCl3); H
NMR (300 MHz, CDCl3): d 8.24 (1H, d, J 2.4 Hz, H-Ar), 8.07
(1H, dd, J 9.3 Hz, J 2.4 Hz, H-Ar), 7.23 (1H, d, Jꢁ 9.3 Hz, H-
Ar), 5.82 (1H, s, H-1), 5.44 (1H, d, J2,3, H-2), 5.12 (1H, dd, J2,3
1.5 Hz, J3,4, H-3), 4.25 (1H, dd, J3,4 4.4 Hz, J4,5 8.3 Hz, H-4),
3.77–3.89 (2H, m, H-5a, H-5b), 2.11 (6H, s, CH3C O); 13C NMR
=
=
(75 MHz, CDCl3): d 170.5, 169.5 (2 × CH3C O), 156.5–115.3
(C-Ar), 103.9 (C-1), 85.3, 81.0, 76.3 (C-2, C-3, C-4), 61.6 (C-5),
+
=
20.6, 20.5 (2 × CH3C O); m/z (ES) 411.9 [M + Na] ; HRMS
CDCl3): d 166.1, 165.6, 165.3 (C O), 156.5–115.3 (C-Ar), 104.3
calcd for C15H16NO9ClNa+ 412.0411; found 412.0409.
=
(C-1), 83.7 (C-4), 81.6 (C-2), 77.1 (C-3), 63.3 (C-5); m/z (DCI) 635
[M + NH4]+; HRMS calcd for C32H24NO10ClNa+ 640.0986; found
640.0993.
2-Chloro-4-nitrophenyl 5-O-(4-acetoxyferuloyl)-2,3-di-O-acetyl-
a-L-arabinofuranoside 13
To a solution of 12 (395 mg, 1.01 mmol) and DMAP (14 mg,
0.12 mmol) in pyridine (5 mL), 4-acetoxyferuloyl chloride23
(200 mg, 0.77 mmol) was added portionwise. The reaction was
stirred at room temperature. for 3.5 h then coevaporated with
toluene. Then the residue was diluted with CH2Cl2 (100 mL) and
washed successively with HCl (1M), sat. aq. NaHCO3 solution,
H2O and dried with NaSO4. The crude product was purified by
column chromatography (petroleum ether– ethyl acetate, 6:4 v/v)
2-Chloro-4-nitrophenyl a-L-arabinofuranoside 10
Compound 9 (1.0 g, 1.62 mmol) was suspended in dry MeOH
(25 mL), the reaction was cooled with an ice-water bath and 1 M
NaOMe (350 lL) was added to the mixture. After 2 h, the reaction
was complete and was neutralized with silica. The crude product
was evaporated and purified by flash column chromatography
(CH2Cl2–MeOH, 10:0→98:2 v/v) to give the title compound 10
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 1208–1214 | 1213
©

View Article Online
to give the title compound 13 as a white foam (240 mg, 66%); [a]D
−42 (c 0.52 in CHCl3); 1H NMR (300 MHz, CDCl3): d 8.27 (1H,
d, J 2.4 Hz, H-Ar), 8.10 (1H, dd, J 9.3 Hz, J 2.4 Hz, H-Ar), 2.11,
and in part, by Grants 0401C0042 from ADEME and PNRB2005-
11 from ANR.
=
2.13, 2.28 (9 H, 3 × s, 3 × CH3C O), 3.82 (3H, s, OMe), 4.37–4.56
References
(3H, m, H-4, H-5a, H-5b), 5.17 (1H, d, J3,4 3.4 Hz, H-3), 7.65 (1H,
1 K. Mazeau, C. Moine, P. Krausz and V. Gloaguen, Carbohydr. Res.,
2005, 340, 2752–2760.
=
d, J 16 Hz, HC CH), 7.25 (1H, d, J 9.3 Hz, H-Ar), 7.00–7.01
=
(3H, m, H-Ar), 6.36 (1H, d, J 16 Hz, HC CH), 5.85 (1H, s, H-1),
2 L. Saulnier and J.-F. Thibault, J. Sci. Food Agric., 1999, 79, 396–
5.44 (1H, d, J2,3 1 Hz, H-2); 13C NMR (75 MHz, CDCl3): d 170.1,
402.
3 T. Ishii and T. Hiroi, Carbohydr. Res., 1990, 196, 175–183.
4 T. Ishii, Plant Sci., 1997, 127, 111–127.
5 M. G. Tabka, I. Herpoel-Gimbert, F. Monod, M. Asther and J. C.
Sigoillot, Enzyme Microb. Technol., 2006, 39, 897–902.
6 P. Couthino and H. Henrissat, 1999, http://afmb.cnrs-mrs.fr/CAZY/.
7 M.-C. Ralet, C. B. Faulds, G. Williamson and J.-F. Thibault, Carbohydr.
Res., 1994, 263, 257–269.
=
=
169.4, 168.7, 166.2 (3 × CH3C O, C O), 156.5–111.3 (C-Ar, 2 ×
=
HC CH), 104.0 (C-1), 82.98, 80.65, 77.4 (C-2, C-3, C-4), 62.9
=
(C-5), 55.9 (OMe), 29.7, 20.7, 20.6 (CH3C O); m/z (ES) 630.1
[M + Na]+. HRMS calcd for C27H26NO13ClNa+ 630.0990; found
630.0993.
8 V. Puchart, M. Vrsanska, M. Mastihubova, E. Topakas, C. Vafiadi,
C. B. Faulds, M. Tenkanen, P. Christakopoulos and P. Biely, J. Biotech-
nol., 2007, 127, 235–243.
9 M. Claeyssens and G. Aerts, Bioresour. Technol., 1992, 39, 143–146.
10 E. Farkas, L. Janossy, J. Harangi, L. Kandra and A. Liptak, Carbohydr.
Res., 1997, 303, 407–415.
11 S. Maas, Bio Techniques, 1999, 27, 1126–1128.
12 W. Berlin and B. Sauer, Anal. Biochem., 1996, 243, 171–175.
13 X. Zhu, S. Kawatkar, Y. Rao and G.-J. Boons, J. Am. Chem. Soc., 2006,
128, 11948–11957.
14 M. Mastihubova, V. Mastihuba, L. Kremnicky, J. L. Willet and G. L.
Coˆte´, Synlett, 2001, 10, 1559–1560.
15 M. Gao, B. H. Mock, G. D. Hutchins and Q.-H. Zheng, Bioorg. Med.
Chem., 2005, 13, 6233–6243.
16 C. Re´mond, M. Frechichi, N. Aubry, R. Plantier-Royon, C. Portella
and M. O’Donohue, Tetrahedron Lett., 2002, 43, 9653–9655.
17 C. Williamson, C. B. Faulds and P. A. Kroon,, Biochem. Soc. Trans.,
1998, 26, 205–209.
2-Chloro-4-nitrophenyl 5-O-feruloyl-a-L-arabinofuranoside 7
Compound 13 (160 mg, 0.26 mmol) was dissolved in CH2Cl2–
MeOH (4 mL) (1:1, v:v). The reaction was ice-cooled and K2CO3
(75 mg, 0.54 mmol) was added to the mixture. The reaction was
allowed to warm up to room temperature. After 6.5 h the reaction
was complete and neutralized with silica. The crude product was
purified by two flash columns chromatography (hexane–acetone,
8:2→7:3 v/v) and (CH2Cl2–MeOH, 98:2 v/v) to give the title
compound 7 as a yellowish solid (80 mg, 64%); [a]D −55 (c 1.2
in MeOH); 1H NMR (300 MHz, CD3OD): d), 8.23 (1H, d, J 2.9,
H-Ar), 8.1 (1H, dd, J 9.3 Hz, J 2.9 Hz, H-Ar), 7.70 (1H, d, J
=
16 Hz, HC CH), 7.36 (1H, d, J 9.3, H-Ar), 7.11 (1H, d, J 2 Hz,
H-Ar), 6.99 (1H, dd, J 2 Hz, Jꢁ 8 Hz, H-Ar), 6.75 (1H, d, J 8 Hz,
18 M. Asther, M. Haon, S. Roussos, E. Record, M. Delattre, L. Lesage-
Meessen, M. Labat and M. Asther, Proc. Biochem., 2002, 38, 685–
691.
19 E. Record, M. Asther, C. Sigoillot, S. Page`s, P. J. Punt, M. Delattre, M.
Haon, C. A. M. J. J. Van Den Hontel, J. C. Sigoillot, L. Lesage-Meessen
and M. Asther, Appl. Microbiol. Biotechnol., 2003, 62, 349–355.
20 A. Levasseur, I. Benoit, M. Asther, M. Asther and E. Record, Protein
Expr. Purif., 2004, 37, 126–133.
=
H-Ar), 6.32 (1H, d, J 16 Hz, HC CH), 5.72 (1H, d, J1,2 1.3 Hz,
H-1), 4.43–4.03 (5 H, m, H-2, H-3, H-4, H-5a, H-5b), 3.83 (3H, s,
OMe); 13C NMR (75 MHz, CD3OD): d 169.1 (C O), 159.2–108.9
=
=
(C-1, C-Ar, HC CH), 84.3, 84.0, 79.0 (C-2, C-3, C-4), 64.9 (C-
5), 56.7 (OMe); m/z (ES) 504.08 [M + Na]+. HRMS calcd for
C21H20NO10ClNa+ 504.0673; found 504.0674.
21 L. Saulnier, C. Marot, M. Elgorriaga, E. Bonnin and J.-F. Thibault,
Carbohydr. Polym., 2001, 45, 269–275.
22 T. Debeche, N. Cummings, I. Connerton, P. Debeire and M. J.
O’Donohue, Appl. Environ. Microbiol., 2000, 66, 1734–1736.
23 R. D. Hatfield, R. F. Helm and J. Ralph, Anal. Biochem., 1991, 194,
25–33.
Acknowledgements
Ms Patricia Chaud and Ste´phanie Boullanger are thanked for
technical assistance. This work was supported by CNRS, INRA
1214 | Org. Biomol. Chem., 2008, 6, 1208–1214
This journal is
The Royal Society of Chemistry 2008
©