R. Wahlström et al. / Carbohydrate Research 373 (2013) 42–51
43
presence of low amounts of ILs (10–50 mM) in the BGEs increased
the resolution between the saccharide peaks, but increasing the
content of IL further led to baseline fluctuation. Another observa-
tion was that ILs with long hydrocarbon chains may actually func-
tion as surfactants and reverse the electroosmotic flow. High
sensitivity saccharide analytical methods in high content IL matri-
ces have not been described previously.
Saccharidesarenon-ionic compounds intheir naturalstate. To en-
able the resolution of neutral, non-derivatized saccharides by electric
fields as in CE, alkaline borate buffers are used to form charged bo-
rate–saccharide complexes.10–12 The detection (usually measured
at 195 nm) is considerably improved by the formation of borate–sac-
charide complexes.12 Also indirect detection of saccharides in CE
analysis is possible. In this case, a UV absorbing compound (such as
sorbic acid13 or 2,6-pyridinedicarboxylic acid14) is added to the BGE
and the saccharides are analysed under basic conditions.
The resolving power of CE with pre-column derivatization has
previously been demonstrated for mixtures of monosaccharides,
including uronic and hexenuronic acids, and small xylo- and celloo-
ligomers, in aqueous solutions.10,11,15 Maltooligosaccharides with
degrees of polymerization (DP) of up to 13 have been separated
employing CE techniques.16 Good separation results for derivatized
saccharides have also been obtained employing micellar electroki-
netic capillary chromatography (MEKC).17 CE has been used in the
separation of monosaccharide mixtures in matrices containing N-
methylmorpholine-N-oxide (NMMO) used as industrial cellulose
solvent.18 In this study, monosaccharides were analysed in aqueous
matrices containing roughly 10% NMMO prior to derivatization and
analysis. The presence of NMMO was reported to interfere neither
with the derivatization reaction nor with analysis with CE.
Advantages of sample derivatization include a manifold in-
crease in detectability. Usually derivatized saccharides have
absorption maxima with wavelengths greater than those of unde-
rivatized analytes, which increases also the selectivity of detection.
Commonly encountered carbohydrate derivatization reagents are
for example, 4-aminobenzoic acid ethyl ester (ABEE),10,11,15 4-ami-
nobenzonitrile (ABN),15 6-aminoquinoline (6-AQ)19 and 8-amino-
naphthalene-1,3,6-trisulfonic acid (ANTS).16 The derivatization
proceeds via reductive amination and needs a free reducing end
of the analyte. Reductive amination works well for aldoses, but ke-
toses such as fructose are not well derivatized.10,16 Great excesses
of derivatization reagent are usually used. In the derivatization
method described by Dahlberg et al.,11 the derivatization reaction
is quenched by addition of alkaline borate buffer, which is sug-
gested to form highly water soluble saccharide–borate complexes
at the same time as the excess ABEE reagent is precipitated. Alka-
line borate buffers are generally employed as BGEs for the separa-
tion of ABEE, ABN and similar saccharide derivatives in CE, with
some variations in the alkalinity and borate concentration. Occa-
sionally, additives such as surfactants and alcohols are added to
the BGEs to improve resolution between adjacent peaks.15 Both
normal10 and reverse polarity15 modes have been employed.
Our work on developing IL compatible analytics for saccharide
identification and quantification was started to allow us to study
the action of hydrolytic enzymes on cellulose in imidazolium-based
ILs.20 The ionic liquids studied were 1,3-dimethylimidazolium
dimethylphosphate [DMIM]DMP and 1-ethyl-3-methylimidazolium
acetate [EMIM]AcO. The primary aim was to find a method that
allows the sensitive quantification of cellooligomers up to the size
of cellohexaose in these IL containing matrices. In this paper, we
discuss how the presence of [DMIM]DMP and[EMIM]AcO affect
the routine methods, such as DNS assay, different chromatography
methods and CE in saccharide analysis. We present an optimized
method for cellooligomer analysis in significant contents of ILs
(20–40% (v/v)) employing CE with pre-column derivatization.
The separation power of this method is demonstrated for both
mono- and oligosaccharides obtained from wood-derived biomass
and results for the quantification of the water soluble cellooligo-
mers glucose, cellobiose, cellotriose, cellotetraose, cellopentaose
and cellohexaose in four different matrices are presented. The
usefulness of the method is illustrated by two studies, in the first
of which the action of a commercial b-glucosidase preparation is
studied in [DMIM]DMP and [EMIM]AcO matrices on cellooligomer-
ic substrates, and in the second of which the partial enzymatic
hydrolysis of microcrystalline cellulose by an endoglucanase is
followed for different time points in IL matrices.
2. Materials and methods
2.1. Chemicals
[DMIM]DMP was prepared as described in the literature.21
[EMIM]AcO (purity >98%) was purchased from Ionic Liquid
Technologies (Heilbronn, Germany) and used without further puri-
fications. The halide content of the [EMIM]AcO determined by ion
chromatography was: chloride <100 mg/kg and bromide <50 mg/kg.
Cello-, manno- and xylooligomers in the range of bioses to hexa-
oses were purchased from Megazyme International (Wicklow,
Ireland). Boric acid, sodium hydroxide (NaOH), 1,5-dimethyl-1,5-
diaza-undecamethylene polymethobromide (hexadimethrine bro-
mide), xylose, galactose, mannose and arabinose were obtained
from Sigma–Aldrich (Steinheim, Germany). Glucose was from
VWR International (Leuven, Belgium), galacturonic acid (from citrus
origin) was purchased from BDH Chemicals (Poole, UK). Water was
obtained from a Milli-Q purification system (Millipore, Bedford, MA,
USA). For the preparation of 3,5-dinitrosalicylic acid (DNS) reagent
solution according to Sumner,22 DNS and potassium sodium tartrate
tetrahydrate were acquired from Merck (Darmstadt, Germany). All
chemicals were used as received if not otherwise stated.
b-Glucosidase (Novozym 188) was obtained from Novozymes
(Bagsvaerd, Denmark) and used as such. b-Glucosidase, xylanase
and endoglucanase activities were measured for the crude
b-glucosidase preparation and determined to be 5900, 2970 and
740 nkat/mL, respectively. The unit katal (kat) is defined by the
International Union for Pure and Applied Chemistry (IUPAC) as
the number of catalysed reactions per time unit as mol/s.23
b-Glucosidase activity was measured according to Bailey and
Linko24 and xylanase activity according to Bailey et al.25 but at
pH 5.0. Endoglucanase activity measurements were carried out
according to the HEC assay26 but using carboxymethylcellulose
(CMC) in buffer at pH 5.0.
2.2. Chromatography and DNS assay
Reversed-phase chromatography was carried out based on
experimental conditions described by Yasuno et al.27 Analyses
were carried out on a Dionex Ultimate 3000 HPLC system equipped
with a Phenomenex C-18 Gemini-NX 3
umn and a diode array detector. The eluent was a 0.2 M potassium
l
m 110A 150 ꢀ 2 mm col-
borate buffer at pH 9 with 5% MeOH.
DNS photometric assay was carried out according to the IUPAC
standard procedure26 with the DNS reagent solution prepared as
described by Sumner22 using a Hitachi U-2000 spectrophotometer
for absorption measurements at 540 nm. High-performance anion
exchange chromatography with pulsed amperometric detection
(HPAEC-PAD) was done according to our previously published in-
house method.28
2.3. Capillary electrophoresis
2.3.1. Derivatization prior to CE analysis
The saccharides were derivatized with 4-aminobenzonitrile
(ABN, samples in aqueous solution or containing [DMIM]DMP)