Monosaccharide Composition of Mucin 159
159
14
Monosaccharide Composition of Mucins
Jean-Claude Michalski and Calliope Capon
1. Introduction
Mucin oligosaccharides are constructed by monosaccharide addition to form com-
mon cores. This architecture limits the number of constituent monosaccharides.
Monosaccharides commonly found in mucins may be divided into neutral (galactose
[Gal]; fucose [Fuc]), hexosamines (N-acetylgalactosamine [GalNAc]; N-acetyl-
glucosamine [GlcNAc]), and acidic compounds (sialic acids [NeuAc]). Additive het-
erogeneity comes from the possible substitution with aglycone residues such as sulfate,
phosphate, or acetate groups. Prior to their analysis, monosaccharides must be released
from the oligosaccharide chain by acidic hydrolysis. Monosaccharide composition can
also be achieved on free oligosaccharide-alditols released from the native glycopro-
tein by reductive alkaline treatment (β-elimination). In this case, GalNAc is converted
into N-acetylgalactosaminitol (GalNAc-ol). Different methods are available for the
analysis of monosaccharides depending mainly on the amount of material available.
Several techniques, such as gas-liquid chromatography (GLC) or high-performance
liquid chromatography (HPLC), allow both quantitative and qualitative analysis of
monosaccharide mixtures. Other chromatographic or electrophoretic procedures are
described herein, but these only allow a rapid qualitative analysis of samples. Single
separated monosaccharides may be further identified by physicochemical methods
such as mass spectrometry (MS) or nuclear magnetic resonance.
1.1. Release and Identification of Sialic Acids
Sialic acids constitute a family of nine-carbon carboxylated sugars found in the
external position on glycan chains. The diversity of sialic acids is generated by the
presence of various substituents present on carbon 4, 5, 7, 8, and 9. The substituent on
carbon 5 can be an amino, an acetamido, a glycolyl-amido, or a hydroxyl group and
defines the four major types of sialic acids: neuraminic acid (NeuAc), N-acetyl-
neuraminic acid (Neu5Ac), N-glycolylneuraminic acid (Neu5Gc), and keto-deoxy-

nonulosonic acid (Kdn), respectively. Substituents of the hydroxyl groups present on
From:
Methods in Molecular Biology, Vol. 125: Glycoprotein Methods and Protocols: The Mucins
Edited by: A. Corfield © Humana Press Inc., Totowa, NJ
160 Michalski and Capon
carbons 4, 7, 8, and 9 can be acetyl, lactyl, methyl, sulfate, or phosphate, anhydro
forms can also occur (Fig. 1) (1,2). Most of the substituents, largely O-acetyl groups
are quite labile during acid or alkaline hydrolysis methods generally utilized for the
release of monosaccharides. Consequently, the study of sialic acid must be generally
considered independently of other monosaccharides. The study of sialic acid modifi-
cations has been attempted after release and purification by improving the methods to
avoid any destruction, and is achieved either with low concentrated acid solutions (3)
or with enzymatic hydrolysis.
Many techniques for detection and quantification of sialic acids have been described
(1). These techniques differ widely in the initial purification of sialic acids from other
biological contaminants. One of the most widely used assays is the detection of free
Neu5Ac and Neu5Gc acids by the thiobarbituric acid assay (TBA). Free sialic acids
react with periodate under acidic conditions to produce β-formylpyruvic acid, which
condenses with TBA to produce a purple chromogen (λ
max
= 549 nm). The assay is
sensitive to 1 nmol, but 2-deoxy-sugars interfere because they also condense with
TBA to give a chromophore with a slightly lower λ
max
(532 nm). Powell and Hart (4)
have introduced an HPLC adaptation of the periodate–TBA assay sensitive to 2 pmol,
and requiring no prior purification of released sialic acids. The characterization of
released sialic acids can be achieved by chromatography: thin-layer chromatography,
GLC (3), or HPLC (5–8). The last technique has higher sensitivity and resolving
power. We have reported the HPLC separation of sialic acid quinoxalinones (8) that

allows the detection of sialic acids at the femtomole level.
Fig. 1. The sialic acids. The nine-carbon backbone common to all known sialic acids may be
substituted by R1 or R2 substituents, giving a family of more than 30 different compounds.
Monosaccharide Composition of Mucin 161
1.2. Analysis of Monosaccharides by GLC
GLC methods for identification of monosaccharides are powerful and extremely sen-
sitive. Detection is usually by means of a flame ionization detector (FID), but sensitivity
may be increased by coupling the gas chromatograph to an MS instrument (electron or
chemical impact). Prior to analysis, monosaccharides must be released by hydrolysis of
the oligosaccharides or the glycoproteins and converted to a volatile derivative (9,10).
1.3. Separation of Monosaccharides by HPLC
HPLC has been widely used because of the advantages of allowing rapid and direct
quantification of underivatized or derivatized samples and the ability to characterize
samples through coelution with samples of known structures or through retention time
comparison. Separation methods are based on anion-exchange (11), size exclusion
(12), ion suppression (13), reversed-phase (14), and, most recently, high-performance
anion-exchange chromatography (HPAEC).
HPAEC takes advantage of the weakly acidic nature of carbohydrates to give highly
selective separations at high pH using strong anion-exchange pellicular resins (15). In
HPAEC, strong alkaline solutions, usually NaOH, are used as eluent. Under these condi-
tions, the hydroxyl groups of carbohydrates are converted to oxyanions with pKa values in
the range of 12–14. The anomeric hydroxyl group of the reducing sugar is more acidic than
the others but each of the hydroxyl groups is characterized by a different pKa value (16);
thus, the modification of some of the hydroxyl groups should greatly influence the elution
positions (separation of anomeric and positional isomers) (17,18). Monosaccharides
released from glycoproteins by the previously mentioned hydrolysis methods can be rap-
idly separated in less than 30 min. Because their molar responses are different, a calibra-
tion curve must be established for each monosaccharide. When coupled with pulsed
amperometric detection (PAD), HPAEC allows direct quantification of underivatized
monosaccharides or carbohydrates at low picomole levels (10–50 pmol) with minimal

sample preparation and purification. PAD utilizes a repeating sequence of three potentials.
The most important potential is E1, the potential at which the carbohydrate oxidation cur-
rent is measured. Potential E2 is a more positive potential that oxidizes the gold electrode
and completely removes the carbohydrate oxidation products. The third potential, E3,
reduces the oxidized surface of the gold electrode in order to allow detection during the next
cycle at E1. The three potentials are applied for fixed periods referred to as t
1
, t
2
, and t
3
.
1.4. Electrophoretic Separation of Monosaccharides
Since the early 1990s, capillary electrophoresis has become a good alternative and
rapid procedure for analytical separation of microquantities of carbohydrate com-
pounds including monosaccharides (19). Separation of native monosaccharides is gen-
erally difficult owing to the lack of ionized groups and to their low extinction
coefficients, which do not allow direct ultraviolet (UV) absorbance detection. Conse-
quently, separation generally requires precolumn derivatization with reagents that con-
tain a suitable chromophoric or fluorophoric group in order to facilitate separation and
increase the sensitivity of detection. As described under HPLC, the most common
tagging methods are based on the reductive amination procedure, wherein the reduc-
ing end of the sugar reacts with the primary amino group of the chromophore (20).
162 Michalski and Capon
Different chromophores such as 2-aminopyridine (20), 8-aminonaphthalene-1,3,6-
trisulfonic acid (ANTS) (21), ethyl-4-aminobenzoate, and 4-aminobenzonitrile (22),
have been used for electrophoretic separation of monosaccharides.
2. Materials
2.1. Release and Identification of Sialic Acids
1. 1000 mol wt cutoff dialysis tube (Bioblock Scientific, Illkirch, France).

2. Dowex AG 50 W × 8 (H
+
) (Bio-Rad, Hercules, CA).
3. Dowex AG 3 × 4A (HCOO
–
) (Bio-Rad).
4. Neuraminidase from Vibrio cholerae or Clostridium perfringens (Boehringer Mannheim,
Indianapolis, IN).
5. HPLC equipment with fluorescent detector.
6. Lichrosorb RP 18 HPLC column (5-µm resin, 250 × 4.6 mm) equipped with an RT30-4
Lichrosorb RP18, 7-µm guard cartridge (Merck, Darmstadt, Germany).
7. Stock solution: 2.35 mL of phosphoric acid (85%), and 28.1 g of sodium perchlorate in 1
L of distilled water.
8. Working solution: water:methanol:2X buffer stock (2:3:5).
9. 1,2-Diamino-4,5-methylene dioxybenzene (DMB) (Merck).
10. C18 column (250 × 4.6 mm, particle size 5 µm) (Beckman, Fullerton, CA).
11. DMB–sialic acid HPLC solvents
a. Solvent A: methanol:water (7:93 v/v).
b. Solvent B: acetonitrile:methanol:water (11:7:82 v/v/v).
2.2. Analysis of Monosaccharides by GLC
1. Gas-liquid chromatograph fitted with an FID.
2. Magnesium turnings (Acros Organics, Geel, Belgium).
3. Sodium chloride and sulfuric acid (Sigma, St. Louis, MO).
4. Meso-inositol (Sigma).
5. Silver carbonate (Sigma).
6. Acetic anhydride (Sigma).
7. Heptane (Acros Organics, Sunnyvale, CA).
8. Bis-silyltrifluoroacetamide (BSTFA) (Pierce, Austin, TX).
9. Silicone OV 101 (BP1 phase, SGE).
10. Sodium borohydride (Merck).

11. Pyridine (Merck).
12. Dichloromethane (Merck).
13. Silicone BP 70 (SGE).
14. Helium gas (Air Liquide, Paris, France).
2.3. HPLC Separation Using Amino-Bonded Silica
1. HPLC apparatus equipped with a gradient system.
2. Refractive index detector.
3. Kromasil–NH
2
5 µm column (250 × 4.6 mm) (Alltech, Deerfield, IL).
2.4. Reversed-Phase HPLC of Pyridylamino-Monosaccharides
1. Analytical ODS (C18) 5-µm HPLC column (4.6 × 250 mm) (Zorbax, Interchim,
Montluçon, France).
2. 0.25 M sodium citrate buffer, pH 4.0, containing 1% acetonitrile (Merck).
Monosaccharide Composition of Mucin 163
2.5. HPAEC-PAD
All eluents and chemical products must be of the highest purity available.
1. Gradient pump module (Dionex Bio-LC apparatus, Sunnyvale, CA).
2. A model PAD-2 detector equipped with a gold working electrode. The following pulse
potentials and durations are used for detection: E1 = 0.05 V (t
1
= 360 ms); E2 = 0.70 V (t
2
= 120 ms); E3 = –0.50 V (t
3
= 300 ms) The response time is set to 3 s.
3. Eluent Degas module to sparge and pressurize the eluents with helium (Dionex).
4. Postcolumn with a DQP-1 single-piston pump (Dionex).
5. CarboPac PA-1 column (4 × 250 mm) (Dionex).
6. CarboPac PA-1 guard (4 × 50 mm) (Dionex).

7. CarboPac MA-1 column (4 × 250 mm) (Dionex).
8. CarboPac MA-1 guard (4 × 50 mm) (Dionex).
9. 18 M Ω deionized water (Milli-Q Plus System, Millipore, Bedford, MA).
10. NaOH 50% solution with less than 0.1% sodium carbonate (Baker, Deventer, The Netherlands).
11. Anhydrous sodium acetate (Merck).
12. Acetic acid (glacial, HPLC grade; Merck).
13. Eluents containing sodium acetate should be filtered through 0.45-µm nylon filters
(Millipore) prior to use.
14. Solvents for separation of neutral monosaccharides, hexosamines, and uronic acids (see
Subheading 3.3.3.2.).
a. Eluent 1: Deionized water.
b. Eluent 2: 25 mM NaOH and 0.25 mM sodium acetate.
c. Eluent 3: 200 mM NaOH and 300 mM sodium acetate.
d. Eluent 4: 125 mM NaOH and 10 mM sodium acetate.
15. Solvents for HPAEC separation of sialic acids (see Subheading 3.3.3.3.).
a. Eluent 1: Deionized water.
b. Eluent 2: 5 mM NaOAc.
c. Eluent 3: 5 mM acetic acid (glacial, HPLC grade; Merck).
16. Solvents for separation of a mixture of unreduced and reduced monosaccharides (see
Subheading 3.3.3.4.).
a. Eluent 1: Deionized water.
b. Eluent 2: 1.0 M NaOH.
17. Neu5Ac and Neu5Gc acid (Sigma).
18. A mixture of sialic acids released from bovine submaxillary gland mucin (BSM) (Sigma)
(see Subheding 3.1.1.).
2.6. Electrophoretic Separation of Monosaccharides
1. Capillary zone electrophoresis apparatus fitted with a UV detector (Beckman).
2. Capillary tube (50 µm id × 65 cm) (Beckman). A part of the polyimine coating on the
capillary tube is removed by burning at a distance of 15 cm from the cathode, to allow UV
detection.

3. 2-Aminoacridone (AMAC) (Lambda Fluoreszentechnologie GmbH, Graz, Austria) made
up to 0.1 M in acetic acid:dimethylsulfoxide (DMSO, Acros Organics, Sunnyvale, CA)
(3:17 v/v). The solution is stored at –70°C.
4. 1 M sodium cyanoborohydride (Merck) in water. This solution is made fresh for each
experiment.
164 Michalski and Capon
3. Methods
3.1. Release and Identification of Sialic Acids
Figure 2 illustrates the separation of the different sialic acid species obtained after
hydrolysis of BSM. The different sialic acids may be characterized according to their
specific retention times. Additionally, each sialic acid may be characterized by MS
analysis (8).
3.1.1. Chemical Hydrolysis of Sialic Acids
1. Suspend 1–10 mg of mucins in 5 mL of 2 M acetic acid in a Teflon-capped reaction tube.
2. Hydrolyze for 5 h at 80°C.
3. Dialyze the solution for 24 h against 20 vol of water (1000 mol wt cutoff tubing).
4. Lyophilize the diffusate. Direct analysis can be made at this stage.
5. Further purify sialic acids as follows:
a. Redissolve the dialysate in 1 mL of water.
b. Load the sample on a Dowex AG 50W × 8 (H
+
) (Bio-Rad) column (10 mL).
c. Wash the column with 100 mL of water.
d. Lyophilize the effluent.
e. Resuspend the lyophilysate in 1 mL of water.
Fig. 2. HPLC separation of sialic acid quinoxalinones obtained after mild acid hydrolysis of
BSM. Ac, acetyl; Lt, lactyl; Gc, glycolyl.
Monosaccharide Composition of Mucin 165
f. Load the sample on a Dowex AG 3 × 4A (HCOO
–

) (Bio-Rad) column (1 mL).
g. Wash the column successively with 7 mL of 10 mM , 7 mL of 1 M and 7 mL of 5 M
formic acid.
h. Pool the fractions and lyophilize.
3.1.2. Enzymatic Release of Sialic Acid
1. Resuspend 1–10 mg of mucin in 2 mL of 100 mM HEPES-KOH, pH 7.0, 150 mM NaCl,
0.5 mM MgCl
2
, and 0.1 mM CaCl
2
.
2. Add 200 mU/mL of V. cholerae or 40 mU/mL C. perfringens enzyme.
3. Introduce the solution in a dialysis tube (1000 mol wt cutoff) and dialyze against 5 mL of
the same solvent at 37°C overnight.
4. Collect the filtrate and purify the sialic acid as in Subheading 3.1.1.
3.1.3. TBA-HPLC Quantification of Sialic Acids
3.1.3.1. TBA R
EACTION
1. Sialic acids are released from mucins by mild acid hydrolysis as described in Subhead-
ing 3.1.1. The TBA assay is performed essentially according to Warren (23).
2. Place 40 µL of free sialic acid solution (10–100 pmol/100 µL in water) in an Eppendorf tube.
3. Add 20 µL of sodium periodate (128 mg of sodium metaperiodate, 1.7 mL of phosphoric
acid, and 1.3 mL of water).
4. After 20 min at room temperature, add slowly 0.1 mL of 10% sodium arsenite in 0.1 N
H
2
SO
4,
and 0.5 M Na
2

SO
4
.
5. When the solution appears yellow-brown, gently vortex the tubes.
6. Add 0.6 mL of 0.6% TBA (0.6 g of TBA [Sigma] in 0.5 M Na
2
SO
4
[Merck]).
7. After mixing, cap the tubes and heat at 100°C for 15 min.
8. Chill the tubes on ice and centrifuge before HPLC analysis.
3.1.3.2. HPLC A
NALYSIS
(F
IG
. 3)
1. Equilibrate the column in the working solution.
2. Elute in the isocratic mode at a flow rate of 1 mL/min.
3. Run UV detection at 549 nm.
4. Quantify the sialic acid by integrating the surface of the sialic acid. Obtain the chromophore
peak calibration curve with pure sialic acid solution (1–5 µg of sialic acid in 40 µL of water).
5. Wash the column extensively with 50% acetonitrile in water after use.
3.1.4. Characterization and Quantification of Sialic Acids by HPLC
3.1.4.1. D
ERIVATIZATION WITH
DMB
1. Heat sialic acid samples released by mild hydrolysis in 7 mM DMB, 0.75 M β-
mercaptoethanol, and 18 mM sodium hydrosulfite in 1.4 M acetic acid (100–200 µL) for
2.5 h in the dark.
2. Inject 10 µL of the reaction mixture on the C18 column.

3.1.4.2. E
LUTION BY
HPLC
1. Equilibrate the column in 65% solvent A–35% solvent B.
2. Elute using a linear gradient from 65% A/35% B to 100% B over 60 min followed by
isocratic elution by 100% B for 10 min at a flow rate of 1 mL/min.
3. Achieve on-line fluorescent detection at an emission wavelength of 448 nm and excita-
tion wavelength of 373 nm with a response time of 0.5 s.
166 Michalski and Capon
3.2. Analysis of Monosaccharides by GLC
Complete hydrolysis of oligosaccharide chains may be obtained using concentrated
acid solutions.
3.2.1. Trifluoroacetic Hydrolysis
1. Dissolve the oligosaccharide-alditol sample or the native glycoprotein in 0.5 mL of a 4 M
solution of trifluoroacetic acid (TFA) or a mixture of formic acid:water:TFA (3:2:1 v/v/v).
2. Heat at 100°C for 4 h in Teflon-capped tubes.
3. After hydrolysis, remove the acid by repeated evaporation under reduced pressure. Evapo-
ration is completed by the addition of ethanol.
3.2.2. Formic Acid–Sulfuric Acid Hydrolysis
1. Dissolve oligosaccharide-alditols or native glycoproteins in 0.5 mL of 50% aqueous for-
mic acid and hydrolyze for 5 h at 100°C in a Teflon-capped tube.
2. Repeat step 1 using 0.25 M aqueous sulfuric acid for 18 h at 100°C.
3. Neutralize the hydrolysate with barium carbonate powder, filter, and concentrate to dry-
ness (see Note 1).
3.2.3. Methanolysis
Methanolysis is a widely used method for hydrolysis of both oligosaccharides and
native glycoproteins.
Fig. 3. HPLC analysis of TBA chromophores. 2, NeuAc chromophore; 3, deoxyhexose chro-
mophore.