Advantages of pyrene derivatization to site-specific glycosylation analysis on MALDI mass spectrometry

Article abstract of DOI:10.1016/j.ijms.2012.08.006

Glycoproteomics involving the analysis of glycopeptides by matrix-assisted laser desorption/ionization mass spectrometry (MALDI-MS) is a new and attractive technique. However, quantitative performance in MALDI-MS is hampered by its poor reproducibility among laser shots. 2,5-Dihydroxybenzoic acid (DHBA) is a useful matrix for glycopeptides but forms highly heterogeneous crystals. In this study, we have investigated the distribution of significant signals generated from a sample of glycopeptides on the target plate using a MALDI imaging technique. MALDI images of glycopeptides, which have different glycans on the same peptide, in the Lys-C digests of bovine ribonuclease B were identical. Thus, all glycoforms on a given peptide can be detected at the same laser irradiation spot simultaneously, which offers a significant advantage over other techniques. A similar result was observed with glycopeptides of human serum immunoglobulin G. Interestingly, distinct MALDI images were observed for glycopeptides having different amino acid sequences, despite having an identical glycan structure. The common peptides, which were glycosylated or non-glycosylated, or sialylated or desialylated gave similar MALDI images. Taken together, our results suggest that sweet spot localization of glycopeptides is dependent on the peptide moiety rather than the glycan structure. Furthermore, introduction of pyrene group to glycopeptides which have different peptides result in a uniform MALDI image. It suggested that pyrene derivatization in MALDI-MS facilitates straightforward analysis of a glycopeptide mixture because the same mass spectrum can be obtained at every sweet spot in addition to increase in signal intensity. Thus, this study validates the use of MALDI-MS for site-specific glycoprofiling at the glycopeptide level.

Full text of DOI:10.1016/j.ijms.2012.08.006

International Journal of Mass Spectrometry 333 (2013) 8–14
Contents lists available at SciVerse ScienceDirect
International Journal of Mass Spectrometry
journal homepage: www.elsevier.com/locate/ijms
Advantages of pyrene derivatization to site-specific glycosylation analysis
on MALDI mass spectrometry
Takashi Nishikaze, Hisako Okumura, Hiroshi Jinmei, Junko Amano∗
Laboratory of Glycobiology, The Noguchi Institute, 1-8-1 Kaga, Itabashi, Tokyo 173-0003, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Glycoproteomics involving the analysis of glycopeptides by matrix-assisted laser desorption/ionization
mass spectrometry (MALDI-MS) is a new and attractive technique. However, quantitative performance
in MALDI-MS is hampered by its poor reproducibility among laser shots. 2,5-Dihydroxybenzoic acid
(DHBA) is a useful matrix for glycopeptides but forms highly heterogeneous crystals. In this study, we
have investigated the distribution of significant signals generated from a sample of glycopeptides on
the target plate using a MALDI imaging technique. MALDI images of glycopeptides, which have differ-
ent glycans on the same peptide, in the Lys-C digests of bovine ribonuclease B were identical. Thus, all
glycoforms on a given peptide can be detected at the same laser irradiation spot simultaneously, which
offers a significant advantage over other techniques. A similar result was observed with glycopeptides
of human serum immunoglobulin G. Interestingly, distinct MALDI images were observed for glycopep-
tides having different amino acid sequences, despite having an identical glycan structure. The common
peptides, which were glycosylated or non-glycosylated, or sialylated or desialylated gave similar MALDI
images. Taken together, our results suggest that sweet spot localization of glycopeptides is dependent
on the peptide moiety rather than the glycan structure. Furthermore, introduction of pyrene group to
glycopeptides which have different peptides result in a uniform MALDI image. It suggested that pyrene
derivatization in MALDI-MS facilitates straightforward analysis of a glycopeptide mixture because the
same mass spectrum can be obtained at every sweet spot in addition to increase in signal intensity. Thus,
this study validates the use of MALDI-MS for site-specific glycoprofiling at the glycopeptide level.
© 2012 Published by Elsevier B.V.
Received 17 February 2012
Received in revised form 2 August 2012
Accepted 2 August 2012
Available online 10 August 2012
Keywords:
Glycopeptide
Glycoform
MALDI
DHBA
Signal localization
1. Introduction
spectrometry (ESI-MS) [3] and matrix-assisted laser desorp-
tion/ionization mass spectrometry (MALDI-MS) [4–7] have been
In recent years, glycoproteomics has become an important
field of post proteomics approaches. A glycoprotein exists as a
mixture of different glycosylated species named glycoforms. To
clear this microheterogeneity is indispensable to characterization
of the glycoprotein because glycoforms vary biological activi-
ties. Furthermore, if a glycoprotein has multi glycosylation sites,
site-specific glycoprofiling, which provides a quantitative map of
glycans attached to a given protein glycosylation site, is an impor-
tant approach to clearly defining the function of a glycoprotein. It
is essential to analyze glycopeptides without releasing glycans to
prevent from destruction of the protein moiety. It is the best that
all the glycoforms can be measured at once to know quantitative
contents of glycoforms.
used to ionize fragile glycopeptides. MALDI-MS offers several
advantages over ESI-MS for the analysis of a small amount of sam-
ple. In particular, MALDI-MS facilitates (i) relatively simple spectral
interpretation; (ii) rapid analysis and (iii) repeated measurements
of the same sample. Thus, MALDI-MS has been widely used in a
variety of applications for the analysis of glycans and glycopeptides
[8–12].
Although MALDI-MS is a very sensitive and precise method, it
is difficult for quantification because ionization efficiency of ana-
lytes depends on their specific physicochemical properties. The
ionization reaction of a glycopeptide, including protonation and
deprotonation, mostly occurs on the peptide moiety. Thus, the sig-
nal intensity of a glycopeptide largely is dependent on the nature
of the peptide portion, but only weakly dependent on the nature
of glycan. Therefore, the glycoprofiling of glycopeptides based on
their signal intensities should be possible because glycans are
attached onto the same peptide. In fact, compared with differ-
ent methods from MS, good correlation between signal intensities
of glycopeptides and the actual amount of corresponding glyco-
form has been reported [13–15]. However, poor shot-to-shot and
Mass spectrometry (MS) combined with a soft ionization
technique has proven to be a powerful analytical tool for the
analysis of glycopeptides [1,2]. Both electrospray ionization mass
∗
Corresponding author. Tel.: +81 3 3961 3255; fax: +81 3 3964 5588.
E-mail address: amano@noguchi.or.jp (J. Amano).
1387-3806/$ – see front matter © 2012 Published by Elsevier B.V.
http://dx.doi.org/10.1016/j.ijms.2012.08.006

T. Nishikaze et al. / International Journal of Mass Spectrometry 333 (2013) 8–14
9
sample-to-sample reproducibility in MALDI-MS may cause a prob-
lem in such quantitative measurements.
RapiGest SF (Waters, Milford, MA) to a final concentration of 0.1% in
50 mM ammonium bicarbonate, the mixture was heated at 100 ◦C
for 5 min. After cooling, the mixture was incubated with 1 ␮g of
trypsin or Lys-C at 37 ◦C overnight. For chymotrypsin digestion,
IgG was incubated in 100 mM Tris–HCl (pH 8.0) containing 10 mM
CaCl2 and 10 mM dithiothreitol at 55 ◦C for 45 min. After cool-
ing, the sample was carbamidomethylated and denatured using
RapiGest SF as described above. The mixture was incubated with
2 ␮g of chymotrypsin at 25 ◦C overnight.
The poor reproducibility is considered to be particularly promi-
nent when the matrix crystallizes heterogeneously. Although
2,5-dihydroxy benzoic acid (DHBA) is one of the most widely
used matrices for ionizing glycans and glycopeptides, the matrix
gives highly heterogeneous crystals. Typically, the long needle-like
crystals exist mostly at the periphery, whereas thin microcrystals
are distributed in the central region under standard dried-droplet
preparation method. The appearance and intensity of analyte signal
is highly dependent on the spot of laser irradiation over the crystal-
lized matrix. Hence, significant variation in signal intensity is often
observed. We have to measure all area of the matrix crystal and the
signals obtained by all laser shots should be summed. To overcome
the signal heterogeneity, many approaches have been suggested,
including the production of a microcrystalline matrix [16], the fast
evaporation of matrix solvent [17] and the use of liquid matrices
[18]. However, these approaches can often promote undesirable
alkali-metal adduction to the analyte. Bouschen and Spengler have
suggested that relatively slow co-crystallization with matrix is an
efficient sample-cleaning step in dried-droplet preparation [19].
Here, we used a MALDI imaging technique to investigate the dis-
tribution of signal spots of glycopeptides within MALDI samples.
Originally, the MALDI imaging technique was applied to thin-layer
chromatographic separation procedures [20] and for studying bio-
logical samples such as tissues [21]. Recently, several researchers
have used MALDI imaging or a secondary ion mass spectrometry
(SIMS) based imaging technique to investigate signal distribution
in a sample prepared using the dried-droplet technique [19,22–24].
The results of these studies using lipids and peptides as analytes
suggest that analyte signal distribution in DHBA matrix is highly
heterogeneous, caused by segregation of each analyte in its physic-
ochemical property.
The tryptic digest of albumin was desalted using a PepCleanTM
C-18 Spin Column (Pierce, Rockford, IL), and the digests of RNaseB
and IgG were subjected to enrichment by hydrophilic interaction
using cellulose fibrous medium [12,25].
For desialylation, enriched glycopeptide fractions derived from
IgG were heated in 0.8% TFA at 80 ◦C for 45 min, and then dried
using a Speed Vac.
2.3. Preparation of glycopeptides NA2-IRNKS and A2-I*RNKS
Disialylglycopeptide A2-KVANKT, A2-IRNKS, NA2-IRNKS,
IRN(GlcNAc)KS and IRNKS were prepared as described previously
[12]. Stable isotope-labeled A2-I*RNKS was prepared using Fmoc-
Ile*-OH instead of Fmoc-Ile-OH. Ile*, in which six 13C and one 15
were substituted, is 7.017 Da heavier than non-labeled Ile.
N
2.4. On-plate pyrene derivatization
Some samples were directly derivatized by 1-pyrenyl dia-
zomethane (PDAM, Molecular Probes, Inc., Eugene, OR) on the
target plate as described previously [12]. The plate was well rinsed
with xylene to remove excess amount of PDAM.
2.5. Mass spectrometry and MALDI imaging
In this study, we found that the MALDI images of glycoforms on
the same peptide are identical to each other. We also demonstrate
that pyrene derivatization of glycopeptides improves homogeneity
and reproducibility of shot-to-shot spectra.
Samples for MALDI-MS were prepared using standard dried-
droplet technique. Analyte solution was first deposited on a
mirror-polished stainless steel MALDI target. DHBA was dissolved
in 60% acetonitrile/H2O to a concentration of 10 mg/mL and 0.6 ␮L
was applied onto the samples with or without on-plate pyrene
derivatization and then left to dry without active air flow. All the
above procedures were done in a clean room, where the temper-
ature (23◦) and humidity (50%) are controlled. Before measuring
MS, the samples on the target plate were observed using a confo-
cal laser microscope with 50× objective, LEXT OLS3100 (Olympus,
Tokyo, Japan). MALDI-TOF mass spectra were acquired using an
AXIMA-QIT instrument consisting of a quadrupole ion trap and
reflector time-of-flight analyzer (Shimadzu Biotech, Manchester,
UK). A nitrogen laser (337 nm) was used to irradiate the sample
for ionization. Spectra in positive- and negative-ion modes were
obtained with higher laser fluence than the threshold fluence for
[M+H]+ or [M−H]− ion detection. The samples were scanned by
successive 10 laser shots with a spot-to-spot center distance of
50 ␮m in each direction to obtain a MALDI image. The MS data was
converted to comma-separated values (CSV) and was visualized as
MALDI image using graphical software Graph-R.
2. Experimental
2.1. Materials
Human serum albumin and bovine pancreatic ribonuclease
B
(RNaseB) were purchased from Sigma–Aldrich (Steinheim,
Germany). Human immunoglobulin G (IgG) was purchased from
WAKO Pure Chemical, Inc. (Osaka, Japan). Trypsin Gold (Mass
Spectrometry Grade) and Lysyl Endpeptidase (Lys-C, Mass Spec-
trometry Grade) were purchased from Promega (Madison, WI)
and WAKO Pure Chemical, Inc., respectively. Chymotrypsin from
bovine pancreas (Sequencing Grade) was purchased from Roche
(Penzberg, Germany). Cellulose fibrous medium was purchased
from Sigma–Aldrich. The highly purified MALDI matrix chemicals,
DHBA was purchased from Shimadzu-Biotech (Kyoto, Japan). Ace-
tonitrile (LC/MS grade) and trifluoroacetic acid (TFA, HPLC grade)
were purchased from Kanto Chemical Co., Inc. (Tokyo, Japan). The
water used in all experiments was purified by using a NANOpure
DIAMOND Ultrapure Water System from Barnstead (Boston, MA).
All reagents were used without further purification.
3. Results and discussion
3.1. Heterogeneity of signals in a sample of peptide mixture using
DHBA as a matrix
2.2. Digestion of glycoproteins and protein
DHBA forms small and large needle-shaped crystals from the
rim of the target spot toward the center and the crystals is non
homogeneity as shown in Fig. 1. This MALDI sample contains tryptic
digest of human serum albumin consisted of more than 10 differ-
ent peptides. Mass spectra a, b and c in Fig. 1 were obtained from
Albumin, RNaseB or IgG was incubated in 10 mM ammonium
bicarbonate containing 10 mM dithiothreitol at 55 ◦C for 45 min.
After cooling, 5 ␮L of 135 mM iodoacetamide was added to the
mixture, which was then kept in the dark for 45 min. After adding

10
T. Nishikaze et al. / International Journal of Mass Spectrometry 333 (2013) 8–14
Fig. 1. Positive-ion mass spectra of peptides from a tryptic digest of human serum albumin at three positions (a)–(c) as highlighted in the confocal laser microscopy image of
the MALDI sample. Identified peptides are as follows; the m/z ion 876.5, LCTVATLR; m/z 880.4, AEFAEVSK; m/z 927.5, YLYEIAR; m/z 951.4, DLGEENFK; m/z 960.5, FQNALLVR;
m/z 984.5, TYETTLEK; m/z 1074.5, LDELRDEGK; m/z 1138.5, C (calbamidemethyl) C (calbamidemethyl) TESLVNR; m/z 1149.6, LVNEVTEFAK; m/z 1226.6, FKDLGEENFK.
three different locations (a, b and c in the same sample as shown in
the microscopy image, respectively). The signal intensity patterns
such as the ions at m/z 927.5, m/z 960.5 and m/z 1074.5 are sig-
nificantly different among the three spectra. This result indicates
that the signal intensity from each peptide varies depending on
the precise position on the matrix of crystals. Such signal hetero-
geneity in a MALDI sample may cause serious problems in terms of
quantification using MALDI-MS.
Hence, a reproducible glycoprofile can be obtained without scan-
ning all the area of the sample with the laser.
3.3. Effect of glycan structures on the same peptide on signal
distribution
Next, we investigated the influence of sialylation of glycopep-
tide on signal distribution because the character of glycopeptides
occasionally differs depending upon the presence/absence of sialic
acid. The preferential detachment of sialic acid residues from the
glycopeptides mostly occurs in the post-source region after the
ionization event. In particular, the sialic acid is easily lost in the
MALDI ion trap type of mass spectrometer such as AXIMA-QIT.
In the case of disialylated glycopeptides, the vast majority are
detected as monosialylated species and some become completely
desialylated. The signals arising from neutral glycopeptides and
those from desialylated glycopeptides, which are produced via sia-
lylated species, cannot be distinguished. We used a mixture of
non-labeled neutral glycopeptide NA2-IRNKS and stable isotope-
labeled disialylated glycopeptide A2-I*RNKS. A2-I*RNKS gave the
monosialylated ion [M-NeuAc-H]− at m/z 2536.1 and small amount
of the desialylated ion [M-2NeuAc-H]− at m/z 2244.9, but no intact
disialylated ion. The deprotonated form of NA2-IRNKS was detected
at m/z 2237.9, which could be distinguish from desialylated ion of
A2-I*RNKS (Fig. 3). Although [M−H]− of A2-I*RNKS could not be
detected, signal points of [M-NeuAc-H]− should be the same as
those of intact molecular ion of [M−H]− because desialylation is
3.2. Common signal distribution of glycopeptides from RNaseB
We used glycopeptides of RNaseB to investigate whether
signal distribution varies among glycopeptides. RNaseB is
a
well-characterized glycoprotein containing a single N-linked gly-
cosylation comprising a set of high-mannose-type glycans [14,26].
Lys-C digestion of RNaseB generated five glycopeptides, Man5–9-
SRNLTK as shown in Fig. 2. Interestingly, MS spectra (Fig. 2a–c)
obtained from three different positions (a,
b and c) in the
microscopy image gave the same pattern of signal intensities for the
five glycopeptides ions. Furthermore, the MALDI images of Man5-,
Man6- and Man7-SRNLTK are entirely consistent with each other,
as shown in Fig. 2, indicating that they had common “sweet spots”
in the MALDI sample. The detailed signal distribution obtained by
scanning with a spot-to-spot center distance of 20 ␮m showed
the same result (data not shown). This has a practical advantage
because all glycoforms can be detected from common sweet spots.

T. Nishikaze et al. / International Journal of Mass Spectrometry 333 (2013) 8–14
11
Fig. 2. Negative-ion mass spectra of glycopeptides of a Lys-C digest of RNaseB at three positions (a)–(c) as highlighted in the confocal laser microscopy image of the MALDI
sample and the MS images of Man5- (m/z 1690), Man6- (m/z 1852) and Man7- (m/z 2014) glycopeptides, respectively.
Fig. 3. The confocal laser microscopy image of the sample containing NA2-IRNKS and A2-I*RNKS, negative-ion mass spectrum and MS images. I* indicates a stable isotope-
labeled isoleucine residue. The ion at m/z 2238 is used as MS image of NA2-IRNKS. The ion at m/z 2536 (A1-I*RNKS) is used as MS image of A2-I*RNKS.

12
T. Nishikaze et al. / International Journal of Mass Spectrometry 333 (2013) 8–14
Fig. 4. The MS images of the sample containing glycopeptides derived from a tryptic digest of IgG. (a) Confocal laser microscopy image. (b) Negative-ion mass spectrum.
(c) Signal distributions of G0, G1 and G2 glycopeptides derived from IgG2 and IgG1. The following ions are used as each MS image; G0-IgG1, m/z 2632; G1-IgG1, m/z 2794;
G2-IgG1, m/z 2956; G0-IgG2, m/z 2600; G1-IgG2, m/z 2762; G2-IgG2, m/z 2924.
due to post-source decay. As shown in Fig. 3, positions where satis-
factory signals could be obtained are very similar with NA2-IRNKS
and A2-I*RNKS. These results indicate that sialylation of the gly-
copeptide does not affect signal distribution of the glycopeptides
with the same peptide structure.
From above results signal distribution of glycopeptide is con-
sidered to be affected by peptide structure rather than glycan
structure. We investigated MS images of a mixture containing
NA2-IRNKS, IRNKS with the innermost GlcNAc and IRNKS with-
out glycan. As expectedly, these three peptides showed the same
signal localization (data not shown). Therefore, glycan dose not pos-
itively participate in matrix interaction. Consequently, the signals
from a set of glycopeptides having the same peptide with micro-
heterogeneity of glycan structures can be captured on the same
location.
galactose residues with or without core-fucosylation [27]. Six gly-
copeptides of IgG1 or IgG2 were detected in the preparation used
in this study (Fig. 4).
As expected, G0, G1 and G2 glycopeptides derived from IgG2
gave quite similar signal distributions (Fig. 4). In the same way,
the three kinds of glycopeptides derived from IgG1 also gave iden-
tical signal distributions. By contrast, however, different signal
distributions were found between IgG1 glycopeptides and IgG2
glycopeptides. For example, the signal derived from G0 glycopep-
tide of IgG2 is not the same as that from IgG1 glycopeptide, even
though the glycan structure is identical. These results again con-
firm that signal distribution of glycopeptide is dependent on the
peptide sequences rather than glycan structures. The difference
in the amino acid sequences between these glycopeptides is the
substitution of two Tyr residues for Phe residues. Recently, it is
reported that Tyr-296 residue in IgG1 interacts with core fucose
[28]. IgG2 glycopeptides without such interaction may be different
three-dimensional structure from IgG1.
3.4. Signal distribution of a glycopeptide mixture derived from
human IgG
Using a mixture of NA2-IRNKS and NA2-KVNKT, both MS images
were the same (data not shown). These two glycopeptides are very
similar and hydrophilic. These results strongly support the idea that
the localization of intense signals from glycopeptides in the mixture
depends on each peptide moiety.
Chymotryptic digestion of IgG results in smaller glycopeptides
with the amino acid sequences of NSTY for IgG1 and NSTF for IgG2.
In their mixtured sample, different MS spectra were obtained from
Next, we investigated signal distribution of a mixture of gly-
copeptides, which have different peptides. Human IgG possesses
an N-glycosylation site in the Fc-domain region. There are several
subclasses of IgG. The amino acid sequences of tryptic peptides con-
taining the N-glycosylation site are EEQYNSTYR and EEQFNSTFR
for IgG1 and IgG2, respectively. N-Glycans on IgG are reported to
be complex-type bi-antennary structures with varying numbers of

T. Nishikaze et al. / International Journal of Mass Spectrometry 333 (2013) 8–14
13
Fig. 5. The negative-ion mass spectra and MS images of the sample containing glycopeptides derived from a chymotryptic digest of IgG with (d–f) and without (a–c) pyrene
derivatization. During ionization using DHBA matrix, the pyrene group is easily dissociated from glycopeptides, and almost glycopeptides can be detected as the underivatized
form [12]. Therefore, the ion at m/z 2089 as G1-IgG1 and the ion at m/z 2073 as G2-IgG1 should be detected regardless derivatization.
different spots. The MS spectrum from spot a mainly shows IgG2
glycopeptide ion at m/z 2073, while only the IgG1 glycopeptide ion
at m/z 2089 was found around spot b (Fig. 5a and b). MS images
of these ions show quite different signal distributions, as shown
in Fig. 5c. Interestingly, after pyrene derivatization of the same
mixture, signals of both glycopeptides widely located at the rim
and the number of spots with intense signals increased, as shown
in Fig. 5f. MS spectra from different spots show the same profile
(Fig. 5d and e). We already reported that pyrene derivatization of
the mixture of glycopeptides and peptides selectively enhances
ionization of glycopeptides but suppresses ionization of peptides
[12]. These effects are found in various glycopeptides which have
different amino acid sequences if the peptides are small (less than
10 amino acid residues) or hydrophilic. Pyrene derivatives have
high UV photoabsorption and this nature may contribute to high
ionization. In fact, laser fluence to appear ion signals for MALDI
samples of pyrene-derivatized glycopeptides was less than that for
MALDI samples without pyrene derivatization. In this study we dis-
cuss an advantage of pyrene derivatization from another point of
view. Glycopeptides that labeled with pyrene appears to have high
affinity for DHBA molecules and facilitate their incorporation into
the crystals at many points, which become sweet spots. One of the
reasons for high sensitivity by pyrene derivatization should be high
incorporation into the matrix crystals. This effect is evenly given to
glycopeptides and results in homogeneous signal distributions of
them.
[24]. These researchers have also reported that the segregation of
analytes correlates with their hydrophobicity, and to a lesser extent
analyte mass or mobility. Thus, analyte co-crystallization into the
matrix crystal is highly dependent on the physicochemical prop-
erties of the analyte. In this study, no segregation was observed
for all glycoforms of N-linked glycopeptides having the same pep-
tide moiety. These glycoforms are probably incorporated into DHBA
crystals without segregating from each other. During crystallization
of the sample, interaction between the growing matrix crystal and
co-crystallizing glycopeptide is governed by the physicochemical
properties of the peptide moiety of the glycopeptide.
We recently reported that 2,5-DHBA crystals on a target plate
obtained by dried-droplet preparation contain two polymorphs
and the signal inhomogeneity is highly related to both these crystal
structures of 2,5-DHBA and physico-chemical properties of analyte
[29]. At early stage of crystallization interaction between glycopep-
tides and matrix molecules makes an important role for generation
of sweet spot. On this condensation process glycopeptides with the
same peptide structure should act together. Furthermore, pyrene-
derivatized glycopeptides have an advantage of interaction with
matrix due to hydrophobicity and the result showed an increase of
sweet spots regardless of crystal forms.
4. Conclusions
The current study clearly demonstrates that signal distribution
of glycopeptide depends on its peptide moiety, not on the glycan
structures. MALDI images of glycopeptides having the same peptide
moiety are entirely consistent with each other. Hence, all glyco-
forms are simultaneously detected by laser irradiation to a common
“sweet spot” on the MALDI sample, offering a considerable practi-
cal advantage. Data acquisition is made faster and easier by the fact
that the common sweet spots result in similar spectra. Moreover,
pyrene derivatization of glycopeptides enhances hydrophobicity
and results in effective incorporation into the matrix crystals. As
In fact, MALDI signal distribution is not necessarily consistent
with the direct analyte distribution within the matrix crystal. The
signal intensity of analyte is affected by various factors, such as
purity of the sample and efficiency of ionization. In previous work,
MALDI imaging techniques have been used to study the distri-
bution of analyte in a MALDI sample [19,22,24]. Bouschen and
Spengler have shown segregation of peptide analytes in small DHBA
crystals in dried-droplet preparations [19]. Similar segregation of
peptides within a DHBA matrix has been reported by Qiao et al.

14
T. Nishikaze et al. / International Journal of Mass Spectrometry 333 (2013) 8–14
the results, the number of sweet spots increases along the rim and
MS spectra with high signal intense and homogeneity are easily
obtained. Little search for better crystals is necessary, and typi-
cal shot spectra of a MALDI sample is enough without all shots
over the sample, which often make low signal to noise. Thus, it
should be possible to obtain accurate and reproducible MS data for
microheterogeneity of glycans and site-specific glycoprofiling on
the glycoprotein.
[12] J. Amano, T. Nishikaze, F. Tougasaki, H. Jinmei, I. Sugimoto, M. Fujita, K. Osumi,
M. Mizuno, Analytical Chemistry 82 (2010) 8738.
[13] Y. Wada, P. Azadi, C.E. Costello, A. Dell, R.A. Dwek, H. Geyer, R. Geyer, K. Kakehi,
N.G. Karlsson, K. Kato, N. Kawasaki, K.-H. Khoo, S. Kim, A. Kondo, E. Lattova, Y.
Mechref, E. Miyoshi, K. Nakamura, H. Narimatsu, M.V. Novotny, N.H. Packer, H.
Perreanlt, J. Peter-Katalinic, G. Pohlentz, V.N. Reinhold, P.M. Rudd, A. Suzuki, N.
Taniguchi, Glycobiology 17 (2007) 411.
[14] M. Thaysen-Anderson, S. Mysling, P. Højrup, Analytical Chemistry 81 (2009)
3933.
[15] T. Nakamura, T. Nishikaze, H. Jinmei, F. Tougasaki, I. Sugimoto, J. Amano, Journal
of Glycomics & Lipidomics 1 (2011) 103.
[16] T. Nishikaze, J. Amano, Rapid Communications in Mass Spectrometry 23 (2009)
3787.
Acknowledgement
[17] O. Vorm, P. Roepstorff, M. Mann, Analytical Chemistry 66 (1994) 3281.
[18] Y. Fukuyama, S. Nakaya, Y. Yamazaki, K. Tanaka, Analytical Chemistry 80 (2008)
2171.
[19] W. Bouschen, B. Spengler, International Journal of Mass Spectrometry 266
(2007) 129.
This work is supported in part by SENTAN, JST (Japan Science
and Technology Agency).
[20] A.I. Gusev, O.J. Vasseur, A. Proctor, A.G. Sharkey, D.M. Hercules, Analytical
Chemistry 67 (1995) 4565.
References
[21] R.M. Caprioli, T.B. Farmer, J. Gile, Analytical Chemistry 69 (1997) 4751.
[22] R.W. Garden, J.V. Sweedler, Analytical Chemistry 72 (2000) 30.
[23] S.L. Luxembourg, L.A. McDonnel, M.C. Duursma, X. Guo, R.M.A. Heeren, Analyt-
ical Chemistry 75 (2003) 2333.
[24] H. Qiao, G. Piyadasa, V. Spicer, W. Ens, International Journal of Mass Spectrom-
etry 281 (2009) 41.
[25] Y. Wada, M. Tajiri, S. Yoshida, Analytical Chemistry 76 (2004) 6560.
[26] H.J. An, T.R. Peavy, J.L. Hedrick, C.B. Lebrilla, Analytical Chemistry 75 (2003)
5628.
[27] C. Huhn, M.H.J. Selman, L.R. Ruhaak, A.M. Deelder, M. Wuhrer, Proteomics 9
(2009) 882.
[1] W. Morelle, K. Canis, F. Chirat, V. Faid, J.-C. Michalski, Proteomics 6 (2006) 3993.
[2] D.S. Dalpathado, H. Desaire, Analyst 133 (2008) 731.
[3] J.B. Fenn, M. Mann, C.K. Meng, S.F. Wong, C.M. Whitehouse, Science 246 (1989)
64.
[4] M. Karas, D. Bachmann, F. Hillenkamp, Analytical Chemistry 57 (1985) 2935.
[5] M. Karas, D. Bachmann, U. Bahr, F. Hillenkamp, International Journal of Mass
Spectrometry and Ion Processes 87 (1987) 53.
[6] K. Tanaka, H. Waki, Y. Ido, S. Akita, Y. Yoshida, T. Yoshida, Rapid Communica-
tions in Mass Spectrometry 2 (1988) 151.
[7] M. Karas, F. Hillenkamp, Analytical Chemistry 60 (1988) 2299.
[8] D.J. Harvey, Mass Spectrometry Reviews 18 (1999) 349.
[9] D.J. Harvey, Mass Spectrometry Reviews 25 (2006) 595.
[10] D.J. Harvey, Mass Spectrometry Reviews 27 (2008) 125.
[11] J. Amano, M. Osanai, T. Orita, D. Osumi, Glycobiology 19 (2009) 601.
[28] T. Mizushima, H. Yagi, E. Takemoto, M. Shibata-Koyama, Y. Isoda, S. Iida, K.
Masuda, M. Satoh, K. Kato, Genes to Cells 16 (2011) 1071.
[29] T. Nishikaze, H. Okumura, H. Jinmei, J. Amano, Mass Spectrometry 1 (2012)
A0006.