MALDI-TOF MS characterization of glycation products of whey proteins in a glucose/galactose model system and lactose-free milk

Article abstract of DOI:10.1021/jf104131a

The major modifications induced by thermal treatment of whey proteins α-lactalbumin (α-La) and β-lactoglobulin (β-Lg) in a model system mimicking lactose-free milk (L^- sugar mix) were investigated by matrix-assisted laser desorption ionization-

Full text of DOI:10.1021/jf104131a

ARTICLE
pubs.acs.org/JAFC
MALDI-TOF MS Characterization of Glycation Products of Whey
Proteins in a Glucose/Galactose Model System and Lactose-free Milk
Saverio Carulli,* Cosima D. Calvano,* and Francesco Palmisano
Dipartimento di Chimica and Centro di Ricerca Interdipartimentale S.M.A.R.T. Universitꢀa degli Studi di Bari “Aldo Moro”, Via Orabona
4, 70126 Bari, Italy
Monika Pischetsrieder
Department of Chemistry and Pharmacy, Food Chemistry, University of Erlangen-Nuremberg, Schuhstrasse 19, 91052 Erlangen,
Germany
ABSTRACT: The major modifications induced by thermal treatment of whey proteins R-lactalbumin (R-La) and β-lactoglobulin
(β-Lg) in a model system mimicking lactose-free milk (L- sugar mix) were investigated by matrix-assisted laser desorption
ionization-time-of-flight mass spectrometry (MALDI-TOF MS). The analysis of the intact R-La revealed species with up to 7 and
14 adducts from lactose and sugar mix, respectively, whereas for β-Lg 3 and up to 5 sugar moieties were observed in the case of
lactose and sugar mix experiments, respectively. A partial enzymatic hydrolysis with endoproteinase AspN prior to mass
spectrometric analysis allowed the detection of further modifications and their localization in the amino acid sequence. Using
R-cyano-4-chlorocinnamic acid as MALDI matrix, it could be shown that heating R-La and β-Lg with glucose or galactose led to the
modification of lysine residues that are not glycated by lactose. The higher glycation degree of whey proteins in a lactose-free milk
system relative to normal milk with lactose reflects the higher reactivity of monosaccharides compared to the parent disaccharide.
Finally, the analysis of the whey extract of a commercial lactose-free milk sample revealed that the two whey proteins were present as
three main forms (native, single, and double hexose adducts).
KEYWORDS: glycation, whey proteins, MALDI mass spectrometry, lactose-free milk, AspN
’ INTRODUCTION
lactosylation of β-Lgs and R-La in either model solutions or
commercial milk products.18-22
Because of their biochemical, nutritional, and functional
properties, the major whey proteins, R-lactalbumin (R-La) and
β-lactoglobulins (β-Lg) A and B, belong to the most valuable
food proteins.1 Whey proteins are extensively used as additives
(e.g., as foaming agents, emulsifiers, egg protein replacers, taste
enhancers) in the food industry and, more recently, their
introduction in clinical nutrition has also been proposed because
of their beneficial effects on human health.2,3 However, thermal
treatments, such as pasteurization and ultrahigh-temperature
(UHT) processes, required to reduce the bacterial load and
increase the milk’s shelf life, can alter some characteristics of the
whey proteins.4,5 Unfolding, glycation of lysine (and, to a lesser
extent, of arginine and cysteine) via the Maillard reaction,
oxidation and coaggregation with casein micelles are among
the most important alterations occurring during milk heating.6,7
As a consequence, a reduced bioavailability of important
amino acids such as lysine8 and a general deterioration of
whey proteins’ functional properties,9-11 with remarkable effects
also on technological processes such as cheesemaking,12 are
observed.
Some systematic investigations performed on whey protein
lactosylation occurring in pasteurized or UHT milk17-19 showed
that even a mild heat treatment such as pasteurization can lead to
an appreciable extent of lactosylation, involving one or two sites
per protein molecule.
On the other hand, a much higher degree of lactosylation was
reported for whey proteins subjected to severe thermal stress,
involving a maximum of 5 and 10 lysine residues for R-La and
β-Lgs, respectively, in infant formula powders20 and 16 residues
(i.e., 15 lysines plus a leucine at the amino terminus) for β-Lgs
following solid state lactosylation.23
These findings indicate that the exposure of lysine residues to
lactose originally embedded into the protein tertiary structure of
β-Lg is significantly increased by protein unfolding resulting
from thermal denaturation.
Very recently, Losito et al. studied the lactosylation of R-La
and β-Lgs A and B occurring upon 1-5 h of heating at 70, 80, and
90 °C in the presence of a lactose excess, by means of HPLC-ESI-
MS.24 A significant amount of mono- and bilactosylated forms of
the three proteins and their relative increase with heating
temperature and time were observed. Lactosylation sites were
Mass spectrometry (MS) techniques based on soft ionization
processes, such as matrix-assisted laser desorption ionization
(MALDI) and electrospray ionization (ESI), have been exten-
sively used13-17 to study the effects of lactosylation on the major
whey proteins. A þ324 Da shift in the molecular mass of the
intact protein gives an immediate indication of a single lactosyla-
tion event. Indeed, MS has provided evidence for single/multiple
Received: October 23, 2010
Accepted: January 19, 2011
Revised:
January 18, 2011
Published: February 14, 2011
r
2011 American Chemical Society
1793
dx.doi.org/10.1021/jf104131a J. Agric. Food Chem. 2011, 59, 1793–1803
|

Journal of Agricultural and Food Chemistry
ARTICLE
identified by analyzing the tryptic digests of the heated protein
solutions; moreover, a parallel, significant denaturation, invol-
ving partial tertiary structure unfolding, was also observed for
β-Lgs. In particular, several new lactosylated lysine sites were
identified for R-La, although their appearance was not accom-
panied by a significant modification of the protein tertiary
structure. On the contrary, interesting differences were observed
for the two variants of β-Lgs, suggesting a different denaturation
pathway for these proteins.
respectively, after a reaction time of 120 min. Lys256 and Lys420
appeared to be the most available sites for conjugation.33
However, to the best of our knowledge, no detailed study on
heat-induced whey protein modifications in lactose-hydrolyzed
milk or in model systems has been reported. Fenaille et al.34 have
shown MALDI-TOF MS spectra of glycated R-La in different
milk samples including one lactose-free powder infant formula.
This work aims at a systematic study of heat-induced mod-
ifications of whey proteins in a model system mimicking lactose-
free milk. Solutions of R-La and β-Lg were thermally treated in
the presence of lactose, glucose, galactose, or a sugar mix,
resembling the carbohydrate fraction of lactose-free milk. The
formation of protein modifications was monitored as a function
of heating duration. Partial enzymatic digestion of the whey
proteins followed by MALDI-TOF MS analysis allowed the
identification of glycation and oxidation products and the
modification sites in the amino acid sequence. Finally, a com-
mercial lactose-free milk was analyzed and its glycation state
assessed.
Furthermore, a recent study25 showed that, in addition to the
well-known glycation reactions, oxidation of whey proteins plays
a major role during the heating of model solutions containing
lactose. Partial enzymatic hydrolysis, followed by MALDI-TOF
MS, allowed the detection and localization of the most promi-
nent modifications in the amino acid sequence, namely, lactulo-
syl- and N-carboxymethyl-lysine formation, lysine oxidation to
aminoadipic semialdehyde, methionine oxidation to methionine
sulfoxide, cyclization of N-terminal glutamic acid to a pyrroli-
done, and oxidation of cysteine or tryptophan.
All of these studies focus on the investigation of modifications
in milk or in model systems simulating lactose-containing milk.
Lactose-free products are of emerging interest, as a response to
the growing importance of lactose intolerance. Lactose intoler-
ance is the inability to completely digest lactose;26 approximately
75% of the world’s population loses the ability to completely
digest a physiological dose of lactose after infancy. Intestinal
digestion of lactose involves its breakdown into glucose and
galactose (rapidly absorbed into the portal circulation) by a
membrane-bound lactase, located in the small intestine. In
subjects suffering lactose maldigestion, a portion of the lactose
load, not digested in the small intestine, passes into the large
intestine, where it is fermented by the colonic microflora,
producing short-chain fatty acids and gases, such as H2, CO2,
or CH4.27-30
’ EXPERIMENTAL PROCEDURES
Materials. Bovine R-lactalbumin type I (R-La), bovine β-lactoglo-
bulin (β-Lg), D-(þ)-glucose (Glu), D-(þ)-galactose (Gal), and D-(þ)-
lactose (Lac) were purchased from Sigma-Aldrich (Taufkirchen,
Germany). Sequencing grade endoproteinase AspN was obtained from
Roche (Mannheim, Germany). ClinProt magnetic beads MB-IMAC-Cu
and 2,5-dihydroxyacetophenone (DHAP) were purchased from Bruker
(Bremen, Germany). Cyanoacetic acid and p-chlorobenzaldehyde used
for R-cyano-4-chlorocinnamic acid (CClCA) synthesis were from abcr
GmbH & Co. KG (Karlsruhe, Germany). R-Cyano-4-hydroxycinnamic
acid (CHCA), diammonium citrate (DAC), and dithiothreitol (DTT)
were obtained from Fluka (Taufkirchen, Germany). Ammonium dihy-
drogenphosphate was from Acros (Geel, Belgium).
To cope with this problem many dairy industries market
lactose-free or so-called high-digestibility milk, in which lactose is
almost quantitatively converted by an enzymatic process into
glucose and galactose. Typically, the lactose content of untreated
milk is around 4.9 g/100 mL; after enzymatic digestion, lactose
content is reduced, by about 90%, to 0.5 g/100 mL with the
concomitant formation of glucose and galactose (2.2 g/100
mL).30 Hydrolysis with soluble or immobilized enzymes can be
performed either before or after heating (pasteurization, UHT).
The effects of intensity and sequence of heat and hydrolytic
treatments could be assessed by monitoring the glucidic fraction
(glucose, lactose, and galactose) and selected thermal treatment
markers (furosine, lactulose, and fructose). For lactose-hydro-
lyzed milk, a higher reactivity toward Maillard reaction was found
thatcouldbe ascribedtothehigherreactivityofthereducingmono-
saccharides glucose and galactose compared to lactose.31 Similar
conclusions could also be drawn by analyzing the decrease of
lysine availability that was greater for skim milk powder with
hydrolyzed lactose compared with the normal skim milk
powder.32
Synthesis of CClCA. CClCA was synthesized according to
a standard Knoevenagel condensation using cyanoacetic acid and
p-chlorobenzaldehyde.35 Ammonium acetate was used as a catalyst.
Two grams of cyanoacetic acid (1 equiv), 0.9 equiv of the benzaldehyde,
and 0.15 equiv of ammonium acetate were refluxed while stirring in
sufficient amounts of toluene (ca. 50 mL) . After quantitative separation
of the reaction water by a Dean-Stark apparatus (ca. 3 h), the reaction
mixture was cooled to 50 °C and filtered. The crude product was washed
with sufficient amounts of distilled water and purified by repeated
recrystallization.
Heating of Whey Proteins in Different Milk Models. R-La
(0.13 g/100 mL) and β-Lg (0.32 g/100 mL) were dissolved in together
with Glu, Gal, and Lac in phosphate-buffered saline (10 mM sodium
phosphate, 8 mM NaCl, pH 6.8). The sugars were either tested alone or
mixed in a composition mimicking that of typical lactose-free milk. For
single-sugar experiments, the model solutions had the following com-
position: lactose, 4.93 g/100 mL; glucose or galactose, 2.2 g/100 mL.
For experiments involving the sugar mix, the following concentrations
were used: lactose, 0.51 g/100 mL; glucose and galactose, 2.2 g/100 mL.
In the following, the different samples used will be referred to as “single
sugar (Glu, Lac, or Gal)”, “sugar mix (Glu þ Gal þ Lac)”, and “control
(no sugar)”.
For instance, an increased reactivity of the monosaccharides
has been already observed in a recent study on the glycation
process of the bovine serum albumin (BSA) reacted with
D-glucose, D-galactose, and D-lactose after dry-heating at 60 °C
(for 30-240 min). The molecular mass increase and glycation
sites of BSA were investigated by MS after digestion with trypsin
and chymotrypsin. D-Galactose was more reactive than D-glucose
or D-lactose, leading to the addition of 10, 3, and 1 sugar residues,
The model solution (600 μL) was heated in a thermoshaker at 60 °C,
and sample aliquots were taken after 0, 8, 46, 56, and 70 h and 3, 7, and 14
days, purified by OMIX C18 tips (Varian Inc., Palo Alto, CA), and
lyophilized. Sugar-free protein solutions were treated in the same way
and used as control. All experiments were performed in triplicate.
Although during industrial manufacturing higher temperatures are
normally reached in the lactose-free milk production, the thermal
1794
dx.doi.org/10.1021/jf104131a |J. Agric. Food Chem. 2011, 59, 1793–1803

Journal of Agricultural and Food Chemistry
ARTICLE
treatment was carried out at 60 °C to prevent and/or reduce protein
denaturation.
140 ns, respectively). Laser-desorbed positive ions were analyzed after
acceleration by 20 kV in the linear mode for the intact proteins and by 19
kV in the reflector mode for the peptide digest. External calibration was
performed using a mix of bovine R-La, chicken lysozyme, and bovine
β-Lg variants A and B for the intact protein and a mix of angiotensins I
and II, substance P, bombesin, ACTH clips 1-17 and 18-39, and
somatostatin 28 for the digest. For each displayed mass spectrum,
150 laser shots from several positions on the spot were averaged.
Protein Identification. The mass spectra obtained from enzy-
matic digest of whey proteins were compared with the peptide
sequences obtained in the nonredundant database Swiss-Prot/TrEMBL
utilizing the Peptide Mass software, available at www.expasy.org (version
06/11/2009). The assignment was performed by setting the following
parameters: monoisotopic peptide masses were indicated as [M þ H]þ
with “cysteines treated with nothing”. AspN and AspN þ N-terminal
Glu were selected as enzymes. One missed cleavage was considered.
Peptides with a mass higher than 750 Da were displayed.
Partial enzymatic protein hydrolysis was performed according to the
method of Meltretter et al.25
Sample Preparation for MALDI TOF Analysis. Before MALDI-
TOF MS analysis, 1 μL of 100 mM DTT (reducing agent) was added to a
5 μL aliquot of the protein solution, followed by incubation for 30 min at
room temperature. A 3 μL aliquot of the reduced sample was diluted 1:1
with 2% trifluoroacetic acid (TFA) and mixed with 3 μL of MALDI matrix
solution prepared by mixing a DHAP saturated ethanolic solution with 10
μM diammonium hydrogen citrate (4 þ 1 v/v). A 1 μL aliquot of the
sample-matrix solution was spotted twice onto a stainless steel target and
air-dried.
For peptide analysis, 1 μL of the protein digest was diluted with 14 μL
of a matrix consisting of a 1:1 mixture of a CHCA saturated solution in
50% acetonitrile/0.1% TFA and a 10 mM ammonium dihydrogen
phosphate solution in 50% acetonitrile/0.1% TFA. A 1 μL aliquot of
the final solution was spotted twice onto a stainless steel target and air-
dried.
A 10 mg/mL solution of CClCA, in 50% acetonitrile/0.1% TFA, was
also used as an alternative matrix to CHCA in peptide analysis.36 One
microliter of the protein digest was mixed with an equal volume of
CClCA solution; a 1 μL aliquot of the final solution was spotted onto a
stainless steel target, air-dried, and washed on target with 2 μL of distilled
water.
Extraction and Purification of Whey Proteins from Milk
Samples. A sample of bovine, lactose-free UHT milk marketed in Italy
was centrifuged at 3850 rpm for 60 min under controlled temperature
(4 °C) conditions. Milk fat floated spontaneously and then solidified on
the top layer so that it could be easily removed. Two milliliters of the
defatted milk were treated with 250 μL of sodium acetate buffer solution
(pH 4.2) to precipitate casein. After centrifugation at 3850 rpm for 5 min
at 4 °C, the casein pellet was discarded, and the remaining solution
(containing whey proteins) was filtered through PVDF filters (0.45 μm
porosity), adjusted to pH 6.8, and then subjected to MB-IMAC
purification protocol.
’ RESULTS AND DISCUSSION
Analysis of Intact Whey Proteins. To investigate the glyca-
tion reactions in a lactose-free milk model, solutions of whey
proteins were heated with lactose, glucose, and galactose, sepa-
rately or mixed, in concentrations typically occurring in milk
samples (sugar mix). Figure 1 shows the MALDI-TOF MS
spectra obtained for R-La heated at 60 °C for 0, 8, 46, 56, and
72 h in the presence of lactose (panel A), glucose (panel B), and a
sugar mix (panel C), respectively. Spectra resulting from longer
heating times, that is, 3, 7, and 14 days, are not displayed because
only broad and unresolved peaks were obtained, probably due to
severe modification, thermal degradation processes, or dehydra-
tion/oxidation processes of the condensed sugars.
The MALDI spectrum of native R-La showed (bottom figure
in panel A) a main signal at m/z 14177 and two shoulders at
higher masses shifted by 162 and 324 Da that can be likely
assigned to the glucosyl and lactosyl adducts of the native
standard protein. With increasing heating time, the formation
of two series of signals was clearly observed (see uppermost
spectrum in panel A). In particular, after 72 h of heating time, up
to seven lactose adducts (mass shift þ324) of the native and
glycated protein (marked as solid and open squares, respectively)
were formed. Similar experiments carried out by Meltretter
et al.25 gave slightly different results. However, in that case R-La
type III was used, which was contaminated by its truncated form in
which the C-terminal leucine was cleaved during the purification
process. Moreover, after 3 days of incubation in the presence of
lactose, only three adducts were clearly visible, instead of the seven
adducts observed here. This discrepancy is most likely explained
by the use of different heating conditions with a probably higher
heating efficiency achieved in the present work.
Similarly, incubation experiments with glucose (panel B) up to
72 h led to the formation of up to 14 adducts (mass shift þ162)
of native and glycated R-La (marked as solid and open triangles,
respectively). It is worth noting that the number of observed
adducts (14) is higher than the total number (i.e., 12) of lysine
residues, suggesting the possibility that R-La undergoes addi-
tional glycation probably at the N-terminus amino acid or at an
arginine residue, as already observed by Fenaille et al.37 For
sugars mix experiments (panel C), the observed behavior was
qualitatively similar even if one additional glyco adduct was
formed compared with single sugar experiments up to 56 h.
These results observed for R-La suggest that the modification
IMAC magnetic beads (5 μL) were pretreated with 50 μL of MB-
IMAC binding buffer following the manufacturer’s instructions. Then, 5
μL of the milk extract containing whey proteins was added, and the
solution was carefully mixed. The tube was then placed in a magnetic
beads separator, and the supernatant was carefully removed using a
pipet. The magnetic beads were washed three times with 100 μL of
washing buffer, and bound analytes were recovered using 10 μL of the
elution buffer.
MALDI-TOF MS. MS experiments were performed using a Micro-
mass M@LDI-LR time-of-flight mass spectrometer (Waters MS Tech-
nologies, Manchester, U.K.) or a Bruker Autoflex (Bruker Daltonik,
Bremen, Germany), both equipped with a nitrogen UV laser (337 nm
wavelength). The following voltages were applied (Micromass
M@LDI): pulse, 2610 V (1550 V); source, 15000 V (15000 V);
reflectron, 2000 V; MCP 1900 V (1900 V) for reflectron (linear) mode.
A time lag focusing (TLF) delay of 500 ns was used. The laser firing rate
was 5 Hz, and, unless otherwise specified, 80 laser shots, obtained by a
random rastering pattern, were used for each well. The resulting spectra
were averaged, background subtracted, and smoothed by a Savitzky-
Golay algorithm in reflectron mode and a mean algorithm in linear
mode. Mass calibration in reflectron mode was performed using a
peptide mixture composed of the fragment 18-39 (2465.199 Da) of
the adrenocorticotropic hormone (ACTH), renin (1758.933 Da), and
angiotensin (1296.687 Da). Mass calibration in linear mode was
performed using a protein mixture composed of insulin β-chain
(3497.0 Da), insulin (5735.0 Da), and cytochrome c (12361.1 Da).
Using a Bruker Autoflex mass spectrometer, measurements of intact
and digested proteins were carried out by delayed extraction (350 and
1795
dx.doi.org/10.1021/jf104131a |J. Agric. Food Chem. 2011, 59, 1793–1803

Journal of Agricultural and Food Chemistry
ARTICLE
Figure 1. MALDI-TOF mass spectra of R-La heated at 60 °C for different times in the presence of (A) lactose, (B) glucose, and (C) sugars mix
(D-glucose þ D-galactose þ D-lactose). Lactose adducts of the native and pre-existing glycated protein are marked as solid and open squares, respectively;
hexose adducts of the native and glycated protein are marked as solid and open triangles, respectively.
pattern is different from that observed with the sugars mix or
glucose (and galactose) alone and that glycation of R-La is more
pronounced than lactosylation, in the same conditions of time
and temperature.
degradation of protein, including denaturation and intermole-
cular coaggregation.
Typically, MALDI mass spectra of Amadori products between
β-lactoglobulin and carbohydrates were characterized by a broad
Gaussian peak shape because of the great heterogeneity of the
glycated forms of β-lactoglobulin.38-40
Galactose experiments (data not shown) provided, for both
whey proteins, results similar to those observed in the presence of
glucose.
To evaluate the exact number of carbohydrates covalently
linked to β-Lg, heating experiments were repeated at shorter
times (5 h) but at higher temperatures (80 °C).
The protein was heated without sugar (Figure 3A) and in the
presence of glucose (Figure 3B) and sugars mix (Figure 3C).
After 5 h of heating in the absence of added sugar, the two main
native variants A and B were still detected at m/z 18250 and
18336, respectively, but with lower absolute signal intensity. This
result is not unexpected because it has been reported24 that β-Lg
undergoes a partial degradation during heating at relatively high
temperatures, generating peptides with MW between 2000 and
10000 Da. MALDI-TOF MS spectra were acquired for heated
samples; large peptides in the mass range m/z 2000-15000 were
detected together with the less intense signal relevant to native
protein (data not shown). In the presence of glucose (Figure 3B),
heating for 5 h led to six additional adducts (mass shift þ162 Da)
Parallel experiments were carried out with β-Lg, the two main
variants, A and B, of which were observed at m/z 18250 and
18336, respectively, with a mass difference of 86 Da. Figure 2
shows the results obtained after heating in the presence of lactose
(panel A), glucose (panel B), and sugars mix (panel C).
After 8 h of incubation with lactose and glucose, one and two
newly formed adducts could be resolved, respectively, from the
native protein, whereas heating with the sugars mix led to the
addition of four sugar moieties. In our case, after 56 h of
incubation with glucose and sugars mix up to four and five
adducts (mass shift 162 Da), respectively, were detected; on the
other hand, during incubation with lactose up to three adducts
(mass shift 324 Da) were identified. After 72 h of heating, three
lactose adducts were still detectable over a much broader peak.
For glucose and sugars mix experiments, only a broad peak
was observed after heating for 72 h; a comparison with the
signal obtained after heating for 56 h suggests that the adducts
average number remained almost unchanged for glucose,
whereas it increased (likely up to 6) for the sugars mix together
with other severe modifications. In parallel, a decrease of the
signal intensity was observed, probably due to the thermal
1796
dx.doi.org/10.1021/jf104131a |J. Agric. Food Chem. 2011, 59, 1793–1803

Journal of Agricultural and Food Chemistry
ARTICLE
Figure 2. MALDI-TOF mass spectra of β-Lg heated at 60 °C for different times in the presence of (A) lactose, (B) glucose, and (C) sugars mix
(D-glucose þ D-galactose þ D-lactose). Numbers of lactose and hexose adducts of both β-Lg variants are indicated.
clearly detectable on both variants and up to 11 adducts in the
presence of the sugars mix (Figure 3C). Incubation for longer
times led again to broad signals with average molecular weights
around m/z 20500, indicating the addition of up to 15 glucose
residues (data not shown).
The formation of a higher number of glycation adducts upon
incubation at 80 °C compared to 60 °C can be explained, besides
a higher reaction rate, by tertiary structure modification occur-
ring on a major portion of β-Lg A and B populations at higher
temperatures. The modified proteins likely expose new basic
amino acids leading to a higher glycation degree.
In a next step, with the goal to get a more detailed insight into
the structures and site of the modifications, the modified proteins
were partially enzymatically digested prior to MS analysis.
MALDI-TOF MS Analysis of Whey Proteins after Partial
Enzymatic Hydrolysis. The native and heated whey proteins
were digested using endoproteinase AspN that specifically
cleaves peptide bonds on the N-terminal aspartic acid, which is
not affected by glycation.20,41
The use of CHCA as MALDI matrix led to unsatisfactory
results in terms of the number of glycation adducts that could be
detected (data not shown), likely because of changes in the
ionization efficiency originating from a reduced basicity of the
modified lysines.42 A recently introduced new matrix, namely,
CClCA, facilitates the ionization of less basic peptides compared
to the classical CHCA.36 Indeed, using CClCA, very informative
MALDI spectra (see, e.g., Figure 4) were obtained. Tables 1 and
2 summarize the detected modifications of R-La and β-Lg and
their localization in the amino acid sequence.
Similar to the R-La/lactose system, we could also show (see
Table 1) a double oxidation of W118/C120, a probable glycation
on N66/N71/N74, and K58/62 besides the modifications
already reported by Meltretter et al. (ultimate assignments of
these glycation sites would obviously require MS/MS
experiments).25
The main modifications produced by the Maillard reaction
involving glucose, galactose, and the “sugars mix” were oxidation,
generation of Nε-carboxymethyl-lysine (CML), lysine aldehyde
formation, and glycation. Concerning glycation of R-La, the
signals at m/z 1414.7 and 1784.9 were ascribed to glucose/
galactose adducts on fragments 1-10 (m/z 1252.6) and 1-13
(m/z 1622.9 Da), respectively, of the native peptides. The
glycation likely involved the lysine residue K5 as already reported
for lactose experiments.25 Moreover, for both native and mod-
ified peptides, byproducts with a mass difference of -18 Da were
observed due to the heating-induced side reaction of the free
amino group of the N-terminal glutamic acid with the side-chain
carboxyl residue to form a pyrrolidone with loss of H2O. In
addition, a modification of the native peptide at m/z 2188.2
(fragment 97-115) with a mass shift of 162 Da was detected in
the glucose, galactose, and sugars mix experiments, indicating a
hexose adduct at Lys98 or Lys114.
The formation of a glyco adduct observed (see Table 1) for the
fragment 64-77 (m/z 1531.6), where lysine is absent, deserves
1797
dx.doi.org/10.1021/jf104131a |J. Agric. Food Chem. 2011, 59, 1793–1803

Journal of Agricultural and Food Chemistry
ARTICLE
Figure 3. MALDI-TOF mass spectra of β-Lg heated at 80 °C for 5 h (A) in the absence of sugar, (B) in the presence of glucose, and (C) in the presence
of the sugars mix (D-glucose þ D-galactose þ D-lactose). The numbers of hexose adducts are indicated.
special interest. This is in agreement with the analyses of the
intact proteins, which revealed a number of hexose adducts
higher than the lysine residues, suggesting that a different residue,
for example, asparagine or glutamine, could be involved in the
glycation reaction. Galactose behavior was similar to glucose.
For β-Lg, a similar reactivity was observed with some differ-
ences between lactose and glucose/galactose experiments
(Table 2; Figure 4B). Sugar adducts of the native peptides at
m/z 2199.1 (fragment 33-52) and 3032.7 (fragment 137-162)
were observed for glucose, galactose, and the sugars mix experi-
ments. These glycation sites have been previously reported when
β-Lg was reacted with lactose.24,25 Additionally, an adduct was
detected at m/z 1265.7 (mass shift of þ162 Da) involving Lys8
or Leu1 on the fragment 1-10 (m/z 1103.6). Indeed, a leucine
residue (Leu1) is located on the N-terminus of the β-Lg
sequence where the -NH2 group is free and can react with
sugar, giving the Amadori product. When a mild thermal treat-
ment is adopted, the most probable glycation site is Leu1, but
for longer time and higher temperature, further modification
on Lys8 is also possible.18,43,44 Furthermore, in agreement
with Meltretter et al.,25 we could detect one oxidation on M7,
a double oxidation on W19 and/or M24, and the CML on K47
(see Table 2).
modified peptides were selected, and the MALDI signal inten-
sities of modified and native peptides were recorded. The adduct
percentage (%A) was calculated as
ꢀ
ꢀ
%A ¼ ½M =ðM þ MÞꢁ ꢂ 100
ð1Þ
where M and M* are the intensities of the unmodified and
modified peptides, respectively.
When both, native and glycated, forms occur (see above)
together with the pyrrolidone byproduct (pyrr, mass shift -18
Da), %A was calculated according to
ꢀ
ꢀ
ꢀ
ꢀ
%A ¼ ½ðM þ M pyrr=ððM þ M pyrrÞ
þ ðM þ MpyrrÞÞꢁ ꢂ 100
ð2Þ
Because other kinds of modifications such as oxidation,
generation of CML, and lysine aldehyde formation are negligible
compared to glycation, their contributions were omitted in the %
A calculation. With this approach, relative quantification of each
modification can be achieved, because the native peptide works
as an internal standard correcting for variations in ionization
efficiency of the single experiment. As a result, good reproduci-
bility among the triplicate experiments was achieved. This
method is particularly useful to record the time course of product
formation.
Figure 5 shows the glycation degree (%A) versus time (days)
for the native peptides at m/z 1252.6 (fragment 1-10 of R-La)
and m/z 2199.1 (fragment 33-52 of β-Lg) heated at 60 °C in the
presence of a single sugar. As can be seen, the %A value was
higher for glucose and galactose than for lactose, suggesting that,
Other previous observations16,18,43 indicated Lys91 and
Lys100 as the most probable glycation sites of β-Lg; this
discrepancy may be explained by the use of different heating
conditions and/or the use of different enzymatic digestion
conditions.
Study of Sugars Reactivity. To assess the different reactivity
of each sugar toward a specific site in the protein sequence, some
1798
dx.doi.org/10.1021/jf104131a |J. Agric. Food Chem. 2011, 59, 1793–1803

Journal of Agricultural and Food Chemistry
ARTICLE
Figure 4. MALDI-TOF mass spectra of the AspN digest of R-La (A) and β-Lg (B) heated for 56 h at 60 °C in the presence of galactose. The glycated
peptides are marked with an asterisk.
under the conditions used in this study, glucose and galactose
were more reactive than lactose. In particular, at the longest
observation time, glycation degrees of 25 and 35% were esti-
mated for the peptide at m/z 1252.6 in the presence of lactose
and glucose/galactose, respectively. Similar results, but with a
lower glycation degree, were obtained for β-Lg (ca. 16% for
glucose/galactose and ca. 8% for lactose).
As can be inferred from Figure 5, after a lag time (different for
R-La and β-Lg), most of the glycation process takes place
between the second and fourth days.
Analysis of a Commercial Lactose-free Milk Sample.
Finally, a commercial UHT lactose-free milk sample purified
by IMAC magnetic beads (see Experimental Procedures) was
analyzed. The main goal of this experiment was to confirm the
validity of our milk-simulating model solutions. The MALDI-
TOF mass spectra of R-La and β-Lg obtained after IMAC
cleanup are shown in Figure 6, panels A and B, respectively.
For R-La, two hexose adducts (mass shift þ162 Da) were
observed. With regard to β-Lg, three hexose adducts could be
detected for both protein variants.
It is worth noting that compared to our milk model sample the
real system shows a lower glycation degree. As discussed before,
this discrepancy is most likely explained by the use of different
heating conditions in laboratory studies compared to industrial
processing. In fact, UHT treatment is developed as a continuous
flow process, whereas our investigations are carried out in a bulk
static solution leading to a different heating transfer. Moreover, a
probable competition for glycation by caseins or a protection by
other milk components could contribute to decrease the glyca-
tion degree on whey proteins.
To the best of our knowledge, glycation of whey proteins in
lactose-free milk has not been investigated. Only one study on
lactose-free milk powder has reported the modification of R-La
with six hexose moieties.35 The lower glycation rate observed in
the present study reflects very well the thermal treatment, which
is less severe for UHT milk than for milk powders.
These results underscore once more that monitoring
of Amadori products in milk samples by MALDI-TOF
MS is a good indicator of protein damage during heating
processes.
Conclusions. The main modifications occurring in whey
proteins in a model system mimicking lactose-free milk were
investigated by means of MALDI-TOF MS analysis of intact and
partially digested proteins. Analysis of intact proteins reveals the
formation of hexose/lactose adducts during heating, whereas the
analysis of AspN digests, using R-cyano-4-chloro-cinnamic acid
as MALDI matrix, allowed the identification of several (including
new) glycation sites. In experiments mimicking lactose-free milk,
additional modification sites were evidenced (compared with the
lactose system) together with a higher glycation degree, indicat-
ing a higher reactivity of glucose and galactose compared to the
parent disaccharide (lactose). Thus, under the same heating
conditions, whey proteins in lactose-free milk should be more
severely damaged than in regular milk.
A bovine lactose-free milk sample has been analyzed
via MALDI-TOF MS after treatment with IMAC-Cu
1799
dx.doi.org/10.1021/jf104131a |J. Agric. Food Chem. 2011, 59, 1793–1803

Journal of Agricultural and Food Chemistry
ARTICLE
Table 1. Main r-La Modifications Observed after Thermal Treatment (See Text for Details)
Lac
Glc
Gal
m/z native peptide mass shift attributiona ΔM = þ324 ΔM = þ162 ΔM = þ162 mix position
peptide sequence
EVFRELK
sites
920.5
þ162
Glyco
NDb
b
b
ND [7-13]
K13
1034.5
þ16
þ32
Ox
b
b
b
b
b
b
b
b
b
b
b
b
[116-123] DQWLCEKL
[116-123] DQWLCEKL
[116-123] DQWLCEKL
W118, C120
W118, C120
K122
diOx
þ324|þ162 Glyco
1177.6
1252.6
þ162
Glyco
ND
b
ND
ND [87-96]
DDIMCVKKIL
K93, K94
-1
Lys
Pyr
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
[1-10]
[1-10]
[1-10]
[1-10]
EQLTKCEVFR
EQLTKCEVFR
EQLTKCEVFR
EQLTKCEVFR
K5
E1
K5
K5
-18
þ324|þ162 Glyco
þ306|þ144 Pyr-Glyco
1622.9
-1
Lys
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
b
[1-13]
[1-13]
[1-13]
[1-13]
[1-13]
[1-13]
EQLTKCEVFRELK
EQLTKCEVFRELK
EQLTKCEVFRELK
EQLTKCEVFRELK
EQLTKCEVFRELK
EQLTKCEVFRELK
K5, K13
E1
-18
þ58
þ40
Pyr
CML
Pyr-CML
K5, K13
K5, K13
K5, K13
K5, K13
þ324|þ162 Glyco
þ306|þ144 Pyr-Glyco
1531.6
1755.9
1817.9
2059.1
2174.0
2188.2
2517.2
þ324|þ162 Glyco
b
ND
b
b
b
b
b
b
b
b
b
ND
b
b
[64-77]
DQNPHSSNICNISC
EYGLFQINNKIWCK
N66, N71, N74
K58, K62
þ162
Glyco
ND [49-62]
þ324|þ162 Glyco
þ324|þ162 Glyco
ND [97-112] DKVGINYWLAHKALCS
K98, K108
b
b
b
b
b
[46-62]
[46-63]
DSTEYGLFQINNKIWCK
DSTEYGLFQINNKIWCKD
K58, K62
þ162
þ162
þ162
Glyco
Glyco
Glyco
ND
ND
ND
b
K58, K62
b
[97-115] DKVGINYWLAHKALCSEKL
K98, K108, K114
b
ND [14-36]
DLKGYGGVSLPEWVCTTFHTSGY K16
a Modification type: Glyco, glycation adduct; Pyr-Glyco, glycation adduct with water loss; Ox/diOx, mono/di oxidation; Lys, lysine aldehyde; Pyr,
pyrrolidone; CML, Nε-carboxymethyl-lysine; Pyr-CML, Nε-carboxymethyl-lysine with water loss. b ND, not detected.
Table 2. Main β-Lg Modifications in Lactose, Glucose, Galactose, and Sugars Mix Samples
Lac
Glu
Gal
m/z native peptide mass shift attributiona ΔM = þ324 ΔM = þ162 ΔM = þ162 mix position
peptide sequence
LIVTQTMKGL
sites
1103.6
-1
Lys
b
b
b
b
b
b
b
b
b
b
[1-10]
[1-10]
[1-10]
K8
þ16
þ162
Ox
b
NDb
LIVTQTMKGL
LIVTQTMKGL
M7
Glyco
L1, K8
1811.9
2199.1
3032.7
þ16
þ32
Ox
b
b
b
b
b
b
b
[11-27]
DIQKVAGTWYSLAMAAS
DIQKVAGTWYSLAMAAS
W19, M24
W19, M24
diOx
ND [11-27]
þ58
þ324|þ162
CML
Glyco
b
b
b
b
b
b
b
b
[33-52]
[33-52]
DAQSAPLRVYVEELKPTPEG
DAQSAPLRVYVEELKPTPEG
K47
K47
þ324|þ162
Glyco
b
b
b
b
[137-162] DKALKALPMHIRLSFNPTQLEEQCHI K138, K141
a Modification type: Glyco, glycation adduct; Ox/diOx, mono/di oxidation; Lys, lysine aldehyde; CML, Nε-carboxymethyl-lysine. b ND, not detected.
1800
dx.doi.org/10.1021/jf104131a |J. Agric. Food Chem. 2011, 59, 1793–1803

Journal of Agricultural and Food Chemistry
ARTICLE
Figure 5. Glycation rate of selected R-La (A, native peptide at m/z 1252.6) and β-Lg peptides (B, native peptide at m/z 2199.1).
magnetic beads. Experimental evidence showed that up to two
monosaccharide adducts can be detected in R-La, whereas up to
three monosaccharide adducts were observed in both β-Lg
variants.
This study demonstrates, once more, the usefulness of MAL-
DI-TOF MS as a simple and fast technique for the high sample
throughput analysis of heat-induced protein modifications in
different milk products.
1801
dx.doi.org/10.1021/jf104131a |J. Agric. Food Chem. 2011, 59, 1793–1803

Journal of Agricultural and Food Chemistry
ARTICLE
Figure 6. MALDI-TOF mass spectra of R-La (A) and β-Lg (B) extracted from a lactose-free milk sample and purified as described under Experimental
Procedures. Hexose adducts on both whey proteins are marked with solid triangles.
’ AUTHOR INFORMATION
capillary electrophoresis of the whey protein fraction compared with
other methods. Int. Dairy J. 2003, 13, 87–94.
Corresponding Author
(10) Finot, P. A. Metabolism and physiological effects of Maillard
reaction products. In The Maillard Reaction in Food Processing, Human
Nutrition and Physiology; Finot, P. A., Aeselbacher, H., Hurrell, R. F.,
Liardon, R., Eds.; Birkhaeuser: Boston, MA, 1990; pp 259-271.
(11) Horvat, S.; Jakas, A. Peptide and amino acid glycation: new
insights into the Maillard reaction. J. Pept. Sci. 2004, 10, 119–137.
(12) Singh, H.; Waungana, A. Influence of heat treatment of milk on
cheesemaking properties. Int. Dairy J. 2001, 11, 543–551.
(13) Marsilio, R.; Catinella, S.; Seraglia, R.; Traldi, P. Matrix-assisted
laser-desorption ionization mass-spectrometry for the rapid evaluation
of thermal-damage in milk. Rapid Commun. Mass Spectrom. 1995, 9,
550–552.
(14) French, S. J.; Harper, W. J.; Kleinholz, N. M.; Jones, R. B.;
Green-Church, K. B. Maillard reaction induced lactose attachment
to bovine β-lactoglobulin: electrospray ionization and matrix-assisted
laser desorption/ionization examination. J. Agric. Food Chem. 2002, 50,
820–823.
(15) Czerwenka, C.; Maier, I.; Pittner, F.; Lindner, W. Investigation
of the lactosylation of whey proteins by liquid chromatography-mass
spectrometry. J. Agric. Food Chem. 2006, 54, 8874–8882.
(16) Siciliano, R.; Rega, B.; Amoresano, A.; Pucci, P. Modern mass
spectrometric methodologies in monitoring milk quality. Anal. Chem.
2000, 72, 408–415.
(17) Losito, I.; Carbonara, T.; Monaci, L.; Palmisano, F. Evaluation
of the thermal history of bovine milk from the lactosylation of whey
proteins: an investigation by liquid chromatography-electrospray ioni-
zation mass spectrometry. Anal. Bioanal. Chem. 2007, 389, 2065–2074.
(18) Leonil, J.; Molle, D.; Fauquant, J.; Maubois, J. L.; Pearce, R. J.;
Bouhallab, S. Characterization by ionization mass spectrometry of
lactosyl β-lactoglobulin conjugates formed during heat treatment of
milk and whey and identification of one lactose-binding site. J. Dairy Sci.
1997, 80, 2270–2281.
*Phone: þ39-080-5443589. Fax: þ39-080-5442026. E-mail:
Saverio.Carulli@agenziadogane.it (S.C.); calvano@chimica.
uniba.it (C.D.C.).
Funding Sources
Partial funding from Universitꢀa degli Studi di Bari “Aldo Moro” is
acknowledged.
’ ACKNOWLEDGMENT
We gratefully acknowledge Gregor Vollmer for his technical
support, Dr. Jasmin Meltretter for the manuscript revision, and
Dr. Antonio Monopoli for the matrix synthesis.
’ REFERENCES
(1) Kinsella, J. E.; Whitehead, D. M. Proteins in whey: chemical,
physical and functional properties. Adv. Food Nutr. Res. 1989, 33, 343–438.
(2) Ewan, H.; Michael, B. Z. Functional properties of whey, whey
components, and essential amino acids: mechanisms underlying health
benefits for active people (review). J. Nutr. Biochem. 2003, 14, 251–258.
(3) Chatterton, D. E. W.; Smithers, G.; Roupas, P.; Brodkorb, A.
Bioactivity of β-lactoglobulin and R-lactalbumin - technological im-
plications for processing. Int. Dairy J. 2006, 16, 1229–1240.
(4) Fox, P. F. Heat-induced changes in milk preceding coagulation.
J. Dairy Sci. 1981, 64, 2127–2137.
(5) Singh, H.; Creamer, L. K. Heat stability of milk. In Advanced
Dairy Chemistry - 1. Proteins; Elsevier Applied Science: London, U.K.,
1992; pp 621-656.
(6) Mauron, J. The Maillard reaction in food; a critical review from
the nutritional standpoint. Prog. Food Nutr. Sci. 1981, 5 (1-6), 5–35.
(7) van Boekel, M. A. J. S. Effect of heating on Maillard reactions in
milk. Food Chem. 1998, 62, 403–414.
(19) Morgan, F.; Molle, D.; Henry, G.; Venien, A.; Leonil, J.; Peltre,
G.; Levieux, D.; Maubois, J. L.; Bouhallab, S. Nonenzymatic lactosyla-
tion of bovine β-lactoglobulin under mild heat treatment leads to
structural heterogeneity of the glycoforms. Biochem. Biophys. Res. Com-
mun. 1997, 236, 413–417.
(8) Mauron, J. Influence of processing on protein quality. J. Nutr. Sci.
Vitaminol. (Tokyo) 1990, 36 (Suppl. 1), S57–69.
(9) De Block, J.; Merchiers, M.; Mortier, L.; Braekman, A.; Ooghe,
W.; Van Rentergem, R. Monitoring nutritional quality of milk powders:
1802
dx.doi.org/10.1021/jf104131a |J. Agric. Food Chem. 2011, 59, 1793–1803

Journal of Agricultural and Food Chemistry
ARTICLE
(20) Marvin, L. F.; Parisod, V.; Fay, L. B.; Guy, P. A. Characterization
of lactosylated proteins of infant formula powders using two-dimen-
sional gel electrophoresis and nanoelectrospray mass spectrometry.
Electrophoresis 2002, 23, 2505–2512.
(41) Fenaille, F.; Morgan, F.; Parisod, V.; Tabet, J. C.; Guy, P. A.
Solid-state glycation of β-lactoglobulin monitored by electrospray
ionisation mass spectrometry and gel electrophoresis techniques. Rapid
Commun. Mass Spectrom. 2003, 17, 1483–1492.
(21) Galvani, M.; Hamdan, M.; Righetti, P. G. Two-dimensional gel
electrophoresis/matrix-assisted laser desorption/ionisation mass spec-
trometry of a milk powder. Rapid Commun. Mass Spectrom. 2000, 14,
1889–1897.
(22) Meltretter, J.; Birlouez-Aragon, I.; Becker, C. M.; Pischetsrieder,
M. Assessment of heat treatment of dairy products by MALDI-TOF-MS.
Mol. Nutr. Food Res. 2009, 53, 1487–1495.
(23) Fenaille, F.; Morgan, F.; Parisod, V.; Tabet, J. C.; Guy, P. A.
Solid-state glycation of β-lactoglobulin by lactose and galactose: locali-
zation of the modified amino acids using mass spectrometric techniques.
J. Mass Spectrom. 2004, 39, 16–28.
(42) Yeboah, F. K.; Yaylayan, V. A. Analysis of glycated proteins by
mass spectrometric techniques: qualitative and quantitative aspects.
Nahrung 2001, 45, 164–171.
(43) Morgan, F.; Bouhallab, S.; Mollꢀe, D.; Henry, G.; Maubois, J. L.;
Lꢀeonil, L. Lactolation of β-lactoglobulin monitored by electrospray
ionisation mass spectromety. Int. Dairy J. 1998, 8, 95–98.
(44) Morgan, F.; Molle, D.; Henry, G.; Venien, A.; Leonil, J.; Peltre,
G.; Levieux, D.; Maubois, J. L.; Bouhallab, S. Glycation of bovine
β-lactoglobulin: effect on the protein structure. Int. J. Food Sci. Technol.
1999, 34, 429–435.
(24) Losito, I.; Stringano, E.; Carulli, S.; Palmisano, F. Correlation
between lactosylation and denaturation of major whey proteins: an
investigation by liquid chromatography-electrospray ionization mass
spectrometry. Anal. Bioanal. Chem. 2010, 396, 2293–2306.
(25) Meltretter, J.; Seeber, S.; Humeny, A.; Becker, C. M.; Pischets-
rieder, M. Site-specific formation of Maillard, oxidation, and condensa-
tion products from whey proteins during reaction with lactose. J. Agric.
Food Chem. 2007, 55, 6096–6103.
(26) Harrington, L. K.; Mayberry, J. F. A re-appraisal of lactose
intolerance. Int. J. Clin. Practice 2008, 62, 1541–1546.
(27) Levitt, M. D. Production and excretion of hydrogen gas in man.
N. Engl. J. Med. 1969, 281, 122–7.
(28) Strocchi, A.; Corazza, G.; Ellis, C. J.; Gasbarrini, G.; Levitt,
M. D. Detection of malabsorption of low-doses of carbohydrate -
accuracy of various breath H2 criteria. Gastroenterology 1993, 105,
1404–10.
(29) Bond, J. H.; Levitt, M. D. Use of breath hydrogen (H2) to
quantitate small bowel transit time following partial gastrectomy. J. Lab.
Clin. Med. 1977, 90, 30–6.
(30) http://www.parmalat.it/prodotti/zymil/index.htm.
(31) Messia, M. C.; Candigliota, T.; Marconi, E. Assessment of
quality and technological characterization of lactose-hydrolyzed milk.
Food Chem. 2007, 104, 910–917.
(32) Rutherfurd, S. M.; Moughan, P. J. Effect of elevated tempera-
ture storage on the digestible reactive lysine content of unhydrolyzed-
and hydrolyzed-lactose milk-based products. J. Dairy Sci. 2008, 91,
477–482.
(33) Ledesma-Osuna, A. I.; Ramos-Clamont, G.; Vꢁazquez-Moreno,
L. Characterization of bovine serum albumin glycated with glucose,
galactose and lactose. Acta Biochim. Pol. 2008, 55, 491–497.
(34) Fenaille, F.; Parisod, V.; Visani, P.; Populaire, S.; Tabet, J. C.;
Guy, P. A. Modifications of milk constitutents during processing:
a preliminary benchmarking study. Int. Dairy J. 2006, 16, 728–739.
(35) Knoevenagel, E. Ber. Dtsch. Chem. Ges. 1898, 31, 2596–2619.
(36) Jaskolla, T. W.; Lehmann, W. D.; Karas, M. 4-Chloro-R-
cyanocinnamic acid is an advanced, rationally designed MALDI matrix.
Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 12200–12205.
(37) Fenaille, F.; Campos-Gimenez, E.; Guy, P. A.; Schmitt, C.;
Morgan, F. Monitoring of R-lactoglobulin dry-state glycation using
various analytical techniques. Anal. Biochem. 2003, 320, 144–148.
(38) Corzo-Martınez, M.; Moreno, F. J.; Olano, A.; Villamiel, M.
Structural characterization of bovine β-lactoglobulin-galactose/tagatose
Maillard complexes by electrophoretic, chromatographic, and spectro-
scopic methods. J. Agric. Food Chem. 2008, 56, 4244–4252.
(39) Van Teeffelen, A. M. M.; Broersen, K.; De Jongh, H. H. J.
Glucosylation of β-lactoglobulin lowers the heat capacity change of
unfolding; a unique way to affect protein thermodynamics. Protein Sci.
2005, 14, 2187–2194.
(40) Sanz, M. L.; Corzo-Martınez, M.; Rastall, R. A.; Olano, A.;
Moreno, F. J. Characterization and in vitro digestibility of bovine
β-lactoglobulin glycated with galactooligosaccharides. J. Agric. Food
Chem. 2007, 55, 7916–7925.
1803
dx.doi.org/10.1021/jf104131a |J. Agric. Food Chem. 2011, 59, 1793–1803