Epitope structure of the carbohydrate recognition domain of asialoglycoprotein receptor to a monoclonal antibody revealed by high-resolution proteolytic excision mass spectrometry

Article abstract of DOI:10.1007/s13361-010-0010-y

Recent studies suggest that the H1 subunit of the carbohydrate recognition domain (H1CRD) of the asialoglycoprotein receptor is used as an entry site into hepatocytes by hepatitis A and B viruses and Marburg virus. Thus, molecules binding specifically to the CRD might exert inhibition towards these diseases by blocking the virus entry site. We report here the identification of the epitope structure of H1CRD to a monoclonal antibody by proteolytic epitope excision of the immune complex and high-resolution MALDI-FTICR mass spectrometry. As a prerequisite of the epitope determination, the primary structure of the H1CRD antigen was characterised by ESI-FTICR-MS of the intact protein and by LC-MS/MS of tryptic digestmixtures. Molecularmass determination and proteolytic fragments provided the identification of two intramolecular disulfide bridges (seven Cys residues), and a Cys-mercaptoethanol adduct formed by treatment with β-mercaptoethanol during protein extraction. The H1CRD antigen binds to the monoclonal antibody in both native and Cys-alkylated form. For identification of the epitope, the antibody was immobilized on N-hydroxysuccinimide (NHS)-activated Sepharose. Epitope excision and epitope extraction with trypsin and FTICR-MS of affinity-bound peptides provided the identification of two specific epitope peptides (5-16) and (17-23) that showed high affinity to the antibody. Affinity studies of the synthetic epitope peptides revealed independent binding of each peptide to the antibody. American Society for Mass Spectrometry, 2011.

Full text of DOI:10.1007/s13361-010-0010-y

B American Society for Mass Spectrometry, 2011
J. Am. Soc. Mass Spectrom. (2011) 22:148Y157
DOI: 10.1007/s13361-010-0010-y
RESEARCH ARTICLE
Epitope Structure of the Carbohydrate
Recognition Domain of Asialoglycoprotein
Receptor to a Monoclonal Antibody Revealed
by High-Resolution Proteolytic Excision Mass
Spectrometry
Raluca Stefanescu,1 Rita Born,2 Adrian Moise,1 Beat Ernst,2 Michael Przybylski1
1Laboratory of Analytical Chemistry and Biopolymer Structure Analysis, Department of Chemistry, University of Konstanz,
78457 Konstanz, Germany
2Department of Molecular Pharmacy, Pharmacenter, University of Basel, 4056 Basel, Switzerland
Abstract
Recent studies suggest that the H1 subunit of the carbohydrate recognition domain (H1CRD) of the
asialoglycoprotein receptor is used as an entry site into hepatocytes by hepatitis A and B viruses and
Marburg virus. Thus, molecules binding specifically to the CRD might exert inhibition towards these
diseases by blocking the virus entry site. We report here the identification of the epitope structure of
H1CRD to a monoclonal antibody by proteolytic epitope excision of the immune complex and high-
resolution MALDI-FTICR mass spectrometry. As a prerequisite of the epitope determination, the
primary structure of the H1CRD antigen was characterised by ESI-FTICR-MS of the intact protein
and by LC-MS/MS of tryptic digest mixtures. Molecular mass determination and proteolytic fragments
provided the identification of two intramolecular disulfide bridges (seven Cys residues), and a
Cys-mercaptoethanol adduct formed by treatment with β-mercaptoethanol during protein extraction.
The H1CRD antigen binds to the monoclonal antibody in both native and Cys-alkylated form. For
identification of the epitope, the antibody was immobilized on N-hydroxysuccinimide (NHS)-activated
Sepharose. Epitope excision and epitope extraction with trypsin and FTICR-MS of affinity-bound
peptides provided the identification of two specific epitope peptides (5–16) and (17–23) that showed
high affinity to the antibody. Affinity studies of the synthetic epitope peptides revealed independent
binding of each peptide to the antibody.
Key words: Asialoglycoprotein receptor, H1CRD protein, Epitope structure, Proteolytic epitope
excision, High-resolution mass spectrometry
Abbreviations: FTICR-MS, Fourier transform ion cyclotron resonance mass spectrometry; MALDI,
matrix-assisted laser desorption/ionization; CRD, carbohydrate recognition domain; IAA, iodoaceta-
mide; SPPS, solid-phase peptide synthesis; DMF, dimethylformamide; NHS, N-hydroxysuccinimide;
PyBOP, (benzotriazol-1-yloxy)trispyrrolidinophosphonium hexafluorophosphate
Electronic supplementary material The online version of this article
Introduction
(doi:10.1007/s13361-010-0010-y) contains supplementary material, which
is available to authorized users.
he asialoglycoprotein receptor (ASGP-R) belongs to the
T
C-type (calcium-dependent) class of lectins [1] and is
located on hepatocytes of all mammalian organisms [2]. The
Received: 19 November 2009
Revised: 4 October 2010
Correspondence to: Michael Przybylski; e-mail: Michael.Przybylski@uni-
konstanz.de
Accepted: 11 October 2010
Published online: 20 January 2011

R. Stefanescu, et al.: Antibody Epitope Structure of H1CRD Protein
149
receptor is a hetero-oligomer consisting of various ratios of
the two homologous subunits H1 and H2, which exhibit 58%
sequence homology [3]. Both subunits are composed of a
cytosolic N-terminal domain, a single transmembrane seg-
ment, a stalk domain and a C-terminal carbohydrate
recognition domain (CRD) (Figure 1a). The carbohydrate
Carbohydrate recognition
Domain (CRD)
(a)
Stalk region
(coiled-coil)
H1
H2
recognition domain of subunit H1 (H1CRD) binds glyco-
proteins with terminal galactose (Gal) and N-acetylgalactos-
amine (GalNAc) motifs [4, 5]. The receptor provides
clearance of circulating ligand glycoproteins by receptor-
mediated endocytosis followed by degradation in the
lysosome, and it is then recycled to the cell surface.
Although both subunits are required for endocytosis of
ligands, the carbohydrate binding activity was shown to be
associated predominantly with H1 [6]. Recent studies
suggest that ASGP-R is used as an entry site into
hepatocytes by hepatitis B virus, Marburg virus and hepatitis
A virus [7, 8]. Thus, molecules that bind specifically to the
carbohydrate recognition domain (CRD) might exert inhib-
itory effects towards these diseases by blocking the virus
entry site. Knowledge of the epitope(s) of antibodies against
H1CRD may contribute to the development of specific
agents capable of blocking carbohydrate binding.
Hepatocyte
cell membrane
Cytoplasmic domain
H1CRD
(b)
H1CRD consists of the amino acid residues (147–290) of
full-length H1. The X-ray crystal structure of H1CRD
(available at the PDB accession number, 1dv8) has been
described for the amino acid residues 153–280, although the
structures at the N- and C-terminal sequences (147–152) and
(281–290) were not identified as a result of thermal disorder
[9]. According to the structural data, H1CRD is a globular
protein comprising six β strands (β1–β6) and two α helices
of 10 and 11 residues, respectively (Figure 1b). The protein
contains seven cysteine residues for which two disulfide
bridges have been described and a third disulfide bond
proposed from the X-ray structure, and three Ca2+ binding
sites. The carbohydrate binding site comprises the amino
acid residues Gln-239, Asp-241, Trp-243, Glu-252, Asp-253
and Asn-264 [10–12]. Although the carbohydrate binding
site resides within residues 239–264, the complete amino
acid sequence of the globular CRD structure is required as
an immunogen for antibody production.
In the present study we have identified the epitope
structure of a recombinant H1CRD antigen expressed in E.
coli [13] to a specific monoclonal antibody, using proteolytic
excision of the H1CRD immune complex and affinity mass
spectrometry. As a basis for the epitope identification, the
complete primary structure and intramolecular disulfide
bridges of H1CRD were characterised by high-resolution
FTICR mass spectrometry and LC-MS/MS. Affinity mass
spectrometry approaches using proteolytic epitope excision
and epitope extraction have been previously described in
detail and applied in epitope identifications [14–19]. In the
general analytical method [14–17, 19], the immobilised
ligand–binder (antigen–antibody) complex is subjected to
limited proteolytic digestion, followed by mass spectrometric
Figure 1. (a) Schematic representation of the asialoglyco-
protein receptor (ASGPR), illustrating the hetero-oligomer
composed of one H1 and one H2 subunit, and the four
domains of the H1 subunit; cytosolic, transmembrane, stalk
and carbohydrate recognition domains are indicated by
arrows. (b) Ribbon diagram of the X-ray crystal structure of
H1CRD (PDB accession number, 1dv8). The N- and C-
terminus are shown on the left side. α Helices are indicated
by red arrows and β strands by blue arrows; calcium ions are
indicated by black dots. (Figure prepared using the program
BallView 1.1.1.)
analysis of the eluted affinity-bound epitope peptide fragment
(s). In the proteolytic step, the epitope is protected from
digestion owing to the shielding of the ligand–binder inter-
action structure; hence, epitope excision on an immobilised
immune complex is essential for isolation, specific dissociation
and identification of the bound epitope. In a variation of this
approach (“epitope extraction”) the antigen ligand is first
subjected to proteolytic digestion, and the mixture of peptide
fragments is subsequently presented to the binder (antibody)
[19]. In alternative approaches [20–22], epitope identifications
have been performed by proteolytic digestion in solution
without immobilisation of the antigen–antibody complex, by
utilizing the specific signal depletion of the complex preserved
on the MALDI target. Here we show that the application of
both epitope excision MS and epitope extraction MS provided
the identification of a specific epitope structure at the N-

150
R. Stefanescu, et al.: Antibody Epitope Structure of H1CRD Protein
terminal part of H1CRD, which is independent and remote
from the carbohydrate binding site. In contrast to the binding of
carbohydrates to H1CRD that requires the presence of calcium
ions, antibody binding and epitope were found to be identical
with and without the presence of calcium. Moreover, we show
that the two synthesized N-terminal epitope peptides each
exhibit independent affinity to the antibody. The determination
of the epitope structure provides a molecular basis for
approaches towards the development of specific inhibitors of
the carbohydrate binding of H1.
Proteolytic Epitope Excision and Extraction
For epitope excision 50 μg solutions of H1CRD dissolved in
200 μL PBS (5 mM Na2HPO4, 150 mM NaCl, pH 7.5) with
and without the addition of 20 mM CaCl2 were applied to
the affinity matrix. The column was gently shaken for 2 h to
allow complete antigen binding, and unbound material was
removed by washing with 10 mL PBS. Remaining affinity-
bound antigen was digested for 2 h at 37 °C with 1 μg tosyl
phenylalanyl chloromethyl ketone (TPCK)-treated trypsin in
50 mM NH4HCO3, or for 20 h with endoprotease Glu-C.
Supernatant non-epitope containing peptides were removed by
washing with 50 mL PBS. The immune complex was then
dissociated by addition of 0.5 mL 0.1% trifluoroacetic acid
(TFA), pH ∼2. The column was gently shaken for 15 min at
20 °C and the released epitope peptides collected in an
Eppendorf cup for mass spectrometric analysis (Supplemen-
tary Figure 4). The column was regenerated by washing with
10 mL 0.1% TFA acid, pH ∼2, followed by 20 mL PBS.
Epitope extraction was performed in the same manner by
applying the proteolytic peptide mixture directly to the column.
Desalting and concentration of the sample were performed by
using ZipTip C18 pipette tips as previously described [24].
Experimental
Preparation of H1CRD
The recombinant carbohydrate recognition domain of H1
(H1CRD) was expressed in E. coli and purified by affinity
chromatography and size exclusion chromatography as
previously described [5, 23]. The cDNA comprised the
amino acid residues 147–290 of the H1 subunit of ASGPR
and an initiator methionine (see sequence numbering of the
recombinant H1CRD in the Supplementary Figure 1). N-
terminal Edman sequence determination of the purified
recombinant H1CRD, performed with a Procise-470 auto-
mated Edman sequencer (Applied Biosystems, Foster City,
CA), showed the lack of the initiator methionine.
Peptide Synthesis
Epitope peptides were synthesized by solid-phase peptide
synthesis (SPPS) according to the Fmoc/tBu strategy using a
semiautomated peptide synthesizer EPS 221 (Abimed,
Langenfeld, Germany). All syntheses were performed on a
NovaSyn TGR resin using N(α)-fluorenylmethyloxycar-
bonyl (Fmoc)-protected amino acids in a fivefold molar
excess relative to the amount of resin. Deprotection of the N
(α)-amino group was carried out by addition of 2% 1,8-
diazabicyclo[5.4.0]undec-7-ene (DBU) and 2% piperidine in
dimethylformamide (DMF). Activation was performed in
0.9 M (benzotriazol-1-yloxy)trispyrrolidinophosphonium
hexafluorophosphate (PyBOP) and 4 M N-methylmorpho-
line base (NMM) in DMF. Cleavage of peptides from the
resin was performed for 2–3 h at 20 °C using a solution
containing 95% TFA, 2.5% triethylsilane and 2.5% deion-
ized water.
Reduction and Alkylation of Disulfide Bonds
Disulfide bridge reduction was performed by addition of a
50-fold molar excess of dithiothreitol (DTT) per cysteine
residue followed by incubation at 56 °C for 1 h. The sample
was then alkylated by incubation for 1 h at 25 °C in the dark
with a 2.5-fold molar excess of iodoacetamide (IAA) over
DTT. Carboxymethylation was also carried out without prior
cysteine reduction in order to modify only cysteine residues
not involved in disulfide bridges.
Antibody Immobilization
Preparative reversed-phase high-performance liquid chro-
matography (RP-HPLC) purification of synthetic epitope
peptides was carried out on a Knauer HPLC system (Knauer,
Bad Homburg, Germany) using a preparative C18 column
GROM-SIL 120 ODS-4 SE with 10 μm silica, 250×20 mm,
120 Å pore size (Grom, Herrenberg-Kayh, Germany). HPLC
chromatograms were recorded by UV detection at 220 nm.
The monoclonal antibody clone B01.4 [23] was dissolved in
0.2 M NaHCO3 and 0.5 M NaCl coupling buffer (pH 8.3) to
a final concentration of 0.5 μg μL−1. A 200 μL aliquot of
this solution was added to 66 mg dry NHS-activated 6-
aminohexanoic acid-coupled Sepharose (Sigma, St. Louis,
MO) and the coupling reaction performed for 1 h at 20 °C.
The coupling product was loaded into a 0.8 mL micro-
column (Mobitec, Göttingen, FRG) and washed sequentially
with blocking (0.1 M aminoethanol, 0.5 M NaCl, pH 8.3)
and washing solution (0.2 M CH3COONa 0.5 M NaCl, pH
4.0). Each washing step was performed with 24 mL (four
times 6 mL) buffer and the procedure was repeated after 1 h
incubation in blocking buffer.
Mass Spectrometry
MALDI-FTICR mass spectrometry was performed with a
Bruker APEX II FTICR mass spectrometer equipped with a
7 T magnet (Bruker Daltonik, Bremen, Germany) using an

R. Stefanescu, et al.: Antibody Epitope Structure of H1CRD Protein
151
external Scout 100 MALDI ion source with fully automated
X-Y target stage with pulsed collision gas [25, 26]. The
pulsed nitrogen laser was operated at 337 nm and MALDI-
generated ions cooled with the pulsing collision gas and
directly captured into the hexapole ion guide. Ions generated
by 20–30 laser shots were accumulated in the hexapole for
0.5–1 s at 30 V and extracted at 15 V. The matrix solution
was prepared by dissolving 100 mg of 2,5-dihydroxybenzoic
acid (DHB) in 1 mL acetonitrile/0.1% TFA in water (2:1 v/v).
A 0.5 μL aliquot of matrix solution was deposited on the
MALDI target, mixed with 0.5 μL of sample solution and
allowed to dry in ambient air. Mass spectra were obtained by
using 10 laser shots for each scan and accumulating 32–64
scans. External calibration was performed by using the
monoisotopic masses of singly protonated ions of human
angiotensin I and II, bradykinin, neurotensin, bovine insulin β
chain (oxidized) and bovine insulin.
ESI-FTICR- MS was performed with the Bruker APEX II
FTICR mass spectrometer equipped with an APOLLO ESI
source as previously described [19]. The sample was
dissolved in a solution containing 50% methanol, 48% water
and 2% acetic acid, and introduced at 2 μL min−1 through a
PEEK capillary to the spraying needle using a syringe pump.
A voltage of −4200 V was applied between the metal-coated
entry of the glass capillary and the grounded spray needle.
The voltage applied at the end cap was −3800 V. Nitrogen
was employed as nebulizing gas to stabilise the spray, and
desolvation was facilitated by using N2 drying gas at 150 °C.
Before each measurement cycle, the source and the analyser
cell were quenched for 50 ms. The ESI ions were passed
through the glass capillary and from the capillary exit
through the skimmer entering into the hexapole ion guide.
The declustering potential (ΔCS) between capillary exit and
skimmer was 60 V. After a predefined accumulation time in
the hexapole (0.1–2 s) the voltage of the extraction plate was
reversed and ions were extracted for transmission to the ICR
cell. External calibration was carried out using monoisotopic
masses of angiotensin I fragment ions formed by in-source
fragmentation; typical calibration mass accuracy was
0.5 ppm at a resolving power of approximately 80,000.
LC/MS analysis was performed with an Agilent (Wald-
bronn, Germany) HP1100 liquid chromatograph for binary
gradient elution (model G1312A) coupled to a Bruker
Daltonik Esquire 3000+ ion trap mass spectrometer. A
binary gradient system consisting of solvent A (0.1%
aqueous formic acid) and solvent B (80% acetonitrile,
0.1% aqueous formic acid) was employed for LC separation;
the sample was dissolved in solvent A at an injection volume
of 5 μL. A 150 mm×4.6 mm×3 μm Discovery RP-18
column was used for peptide separation. All LC-MS data
1
GSERTCCPVN WVEHERSCYW FSRSGKAWAD ADNYCRLEDA HLVVVTSWEE
QKFVQHHIGP VNTWMGLHDQ NGPWKWVDGT DYETGFKNWR PEQPDDWYGH
GLGGGEDCAH FTDDGRWNDD VCQRPYRWVC ETELDKASQE PPLL
51
(a)
101
[M+11H]11+
1539.3553
[M+10H]10+
+
[M+12H]12
(b)
[M+9H]9+
[M+13H]13+
[M+8H]8+
Figure 2. (a) Amino acid sequence of recombinant H1CRD; (b) ESI-FTICR mass spectrum of H1CRD in 50% methanol and 1%
aqueous acetic acid at a concentration of 6 pmol μL−1, showing the 8+ to 13+ charged molecular ions. The isotope resolution of
the 11+ charged molecular ion is shown in the upper right panel. The isotopic structure of the [M+H]+ ion after deconvolution is
shown on the left side

152
R. Stefanescu, et al.: Antibody Epitope Structure of H1CRD Protein
Table 1. Comparison of monoisotopic molecular masses of recombinant H1-CRD determined by ESI-FTICR-MS and the calculated masses containing
different numbers of disulfide linkages and modifications
Number of disulfide bonds
Monoisotopic molecular weight [M+H]+
H1-CRD
−Met
−Met/+β-ME
Measured
Reduced state
1 S–S bond
2 S–S bonds
3 S–S bonds
16981.6263
16979.6107
16977.5951
16975.5795
16850.5858
16848.5702
16846.5546
16844.5390
16926.5841
16924.5685
16922.5529
16920.5373
16921.9093
were obtained in the positive ion mode. Mass spectra were
recorded in the full scan mode (m/z 200–1500). Ion source
parameters were 20 psi nebulizer gas and 10.0 L min−1
drying gas with a temperature of 300 °C. MS/MS experi-
ments were carried out in the autofragmentation mode.
yielding NH2-GSERTCCP- (data not shown) and by LC-
MS/MS (Supplementary Table 2; Supplementary Figures 3, 4).
The Cys-β-ME adduct was formed during extraction of the
protein from the cell lysate and purification in the presence of β-
ME [23].
The primary structure was further characterised by
digestion of native and alkylated H1CRD with trypsin and
LysC protease followed by MALDI-FTICR-MS and LC-
MS/MS of the proteolytic fragments. MALDI-FTICR mass
spectra of the tryptic digest of native and alkylated H1CRD
are shown in Figure 3b–d and the identified peptides
summarised in Table 2 and Supplementary Table 1. Twelve
tryptic fragments (T1–T12) are formed by complete diges-
tion of H1CRD (Figure 3a), all of which were identified with
the exception of the small peptides T1 and T4 (Table 2). The
MALDI-FTICR mass spectra of the tryptic peptide mixture
of native H1CRD revealed the fragments T5 and T11,
thereby allowing the presence of the disulfide bridge Cys-
35–Cys-130 to be established. In addition, the accurate
mass of T2 ([M+H]+, m/z 1470.615) was 2.02 lower than
the calculated mass with reduced Cys-6 and Cys-7 residues,
thus corresponding to a vicinal disulfide bridge (Cys-6–Cys-
7). The identification of the vicinal disulfide bridge in the
fragment T2 was ascertained by reduction and carbamido-
methylation of cysteine residues and MALDI-FTICR-MS
(Figure 3d). The alkylated tryptic peptides each contained a
single alkylated cysteine residue (T3, Cys-18; T10, Cys-122;
T11, Cys-130), whereas the fragment T2 contained two
alkylated cysteine residues (Cys-6, Cys-7), thus confirming
the vicinal disulfide bridge Cys-6–Cys-7 of H1CRD (Sup-
plementary Table 1). Additional confirmation was obtained
Results and Discussion
Characterisation of Primary Structure
and Disulfide Linkages of H1CRD
The H1CRD of the receptor protein was expressed in E. coli
and purified by affinity chromatography and size exclusion
chromatography as previously described [23]. Amino acid
residues in the sequence of the recombinant H1CRD were
numbered 1–144 as shown in Figure 2a and Supplementary
Figure 1, and this sequence numbering is used throughout.
As a precondition for the identification of the epitope, the
primary structure and disulfide linkages of H1CRD were
initially characterised by ESI-FTICR-MS of the intact
protein and LC-MS/MS of proteolytic digest mixtures. The
molecular mass determination of H1CRD was performed
(Figure 2b) using an aliquot of the stock solution diluted to a
final concentration of 6 μM. The determined experimental
mass of 16,921.9093 is well in agreement with the
calculated mass of the protein of 16,922.5529, without
methionine, containing two disulfide bridges, two reduced
cysteine residues and one cysteine–β-mercaptoethanol (Cys-
β-ME) adduct (Table 1). The lack of the initiator methionine,
a frequently observed modification of recombinant proteins
[27], was confirmed by N-terminal Edman sequence analysis
Table 2. Molecular masses and amino acid sequences of tryptic peptide fragments of native H1-CRD determined by MALDI-FTICR-MS
+
+
[M+H]exp
[M+H]calc
948.4033
Δm (ppm)
Tryptic fragment
Cysteine residue
Sequence
948.409
6.0
2.3
0.9
6
21
7.0
T3
T11
T5
T8
C18
C130
C35
17SCYWFSR23
1122.511
1184.480
1417.619
1451.679
1470.615
1122.5136
1184.4789
1417.6270
1451.6485
1470.6253
128WVCETELDK136
27AWADADNYCR36
76WVDGTDYETGFK87
117WNDDVCQRPYR127
5TCCPVNWVEHER16
T10
C122
C6/C7
T2
disulfide bridgea
1883.001
2303.980
1882.9545
2303.9694
24.6
4.6
T6
37LEDAHLVVVTSWEEQK52
T5/T11
C35/C130
C108
27AWADADNYCR36
disulfide bridgea
128WVCETELDK136
2698.298
3287.319
2698.3096
3287.3620
4.3
13.0
T7
T9
53FVQHHIGPVNTWMGLHDQNGPWK75
88NWRPEQPDDWYGHGLGGGEDCAHFTDDGR116
aDisulfide bridge between cysteine residues marked in bold typeface

R. Stefanescu, et al.: Antibody Epitope Structure of H1CRD Protein
153
(a)
C6 C7
C35
36
C130
136
C18
16
C108
C122
T10
1
4
23 26
52
75
87
116
127
144
T1
T2
T3 T4
T5+T6
T5
T6
T7
T8
T9
T11
T12
T8
(b)
(c)
T5+T11
T2
T7
T5+T6
T5+T11
T2
T8
T5
T3
T11
T9
T5
(d)
Figure 3. (a) Schematic representation of tryptic cleavage sites and peptide fragments, and location of cysteine residues and
disulfide bonds identified in the recombinant H1CRD. Tryptic fragments identified by mass spectrometry are shown by grey
bars. The two disulfide bridges identified are indicated by connector lines between the cysteine residues involved. (b, c) MALDI-
FTICR mass spectra of the peptide mixture obtained by tryptic digestion of native H1CRD (5 h, 37 °C, 1:50 enzyme to substrate
ratio). Fragments containing a disulfide bridge are indicated in red. The spectra were recorded by adjusting the hexapole
parameters for optimum sensitivity of the mass range selected: (b) time-of-flight delay 2500 μs, RF amplitude 2.5; (c) time-of-
flight delay 3500 μs, RF amplitude 3.5. (d) MALDI-FTICR-MS of the tryptic peptide mixture of reduced and alkylated H1CRD;
carbamidomethyl modification of cysteine residues are indicated by asterisks
by LC-MS/MS upon LysC digestion of the reduced and
alkylated H1CRD (Supplementary Table 2). The identifica-
tion of the disulfide bridge Cys-35–Cys-130 is in agreement
with the X-ray structure [9], although the Cys-6–Cys-7
linkage was not found by X-ray crystallography. The
formation of a disulfide bond between adjacent cysteine
residues is a relatively rare structural element, which is
usually associated with a β-turn of the protein backbone;

154
R. Stefanescu, et al.: Antibody Epitope Structure of H1CRD Protein
(a)
C6 C7
T2
C18
16
C35
36
C108
T9
C122
T10
C130
T11
1
4
23 26
T4
52
75
87
116
127
136
144
T1
T3
T5
T6
T7
T8
T12
(b)
(c)
T3*
T3
T2**
T2
T2*
T3
(d)
T3
948.4065
T2
1470.6280
Figure 4. (a) Location of the antibody epitope of H1CRD shown by grey bars; (b) MALDI-FTICR-MS of the tryptic epitope
peptides obtained by epitope excision MS of the immune complex with native H1CRD without the presence of CaCl2; (c)
MALDI-FTICR-MS of tryptic epitope peptides after epitope extraction MS of alkylated H1CRD; (d) MALDI-FTICR-MS of tryptic
epitope peptides by epitope excision MS of the immune complex in the presence of 20 mM CaCl2
vicinal disulfide linkages, however, have been identified and
structurally characterised in a number of proteins including
enzymes, receptors and toxins [28–30]. Under the conditions
of the MS analyses, no disulfide scrambling was observed;
the β-ME mixed disulfide adduct is labile under the
moderately acidic solution conditions used and was not
found.
antigen, with two identical sets of experiments with and
without the presence of calcium ions (Figure 4b, c). After
preparation of the affinity column with the immobilised
antibody, epitope excision was carried out with the anti-
body-bound H1CRD as described in the “Experimental”.
Following digestion with trypsin the supernatant containing
unbound peptide fragments was removed, the column
washed extensively with buffer under binding conditions,
and affinity- bound peptides were eluted under acidic
conditions.
Epitope Structure Identification of a Monoclonal
Antibody to H1CRD
The FTICR mass spectra of the affinity-bound epitope
peptide fraction provided the identification of two tryptic
On the basis of the primary structure characterisation,
proteolytic epitope excision and extraction mass spectrom-
etry were applied to the identification of the epitope
recognised by the monoclonal antibody (B01) against
H1CRD. Proteolytic excision MS analyses were carried out
both with the native H1CRD and the reduced/alkylated
5
peptides, TCCPVNWVEHER16 (T2) containing the vicinal
Cys-6–Cys-7 disulfide linkage, and 17SCYWFSR23 (T3)
(Figure 4a, b). These peptides identified the location of the
epitope at the N-terminal domain within residues 5–23 of
H1CRD. To investigate the possible role of the Cys-6 and

R. Stefanescu, et al.: Antibody Epitope Structure of H1CRD Protein
155
Figure 5. Structure model of H1CRD (ribbon diagram), and accessible surface area of the epitope identified by the tryptic
fragments T3 and T2. The identified epitope sequences are depicted in red, the carbohydrate binding site is shown in green and
the position of calcium ions in yellow. Structure models were prepared with the BallView v1.1.1. programme and are based on
the X-ray crystal structure of H1CRD (PDB entry 1DV8)
Cys-7 residues in the interaction with the antibody, epitope
excision MS was additionally performed with the alkylated
H1CRD and this led to the identification of the same epitope
peptides, T2 and T3. Moreover, epitope extraction mass
spectrometry, by subjecting a 50 μg aliquot of the mixture of
tryptic peptides of the alkylated H1CRD to the affinity
column, provided the identical N-terminal fragments T2 and
T3 (Figure 4c), consistent with the previous results; this
suggested that the Cys-6 and Cys-7 residues, albeit con-
tained in the isolated epitope peptide T2, are not involved in
the binding interaction.
Epitope excision MS performed with and without the
presence of calcium ions provided the identification of
identical epitope peptides, T2 and T3 (Figure 4b and c), a
result which is consistent with the binding site of the
antibody being remote from the carbohydrate binding.
The differences in abundances of the two affinity-eluted
epitope peptides were also identical to those found with
the epitope identification without the presence of calcium;
the different abundances may be due to differences in
relative hydrophobicities of the peptides as is frequently
observed in MALDI-MS, and/or differences in relative
binding affinities of the two peptides. The proteolytic
formation and identification of the peptide T2 showed that
the Arg-16 residue is not shielded by the antibody
binding, and suggested that Arg-16 is not involved in
the epitope interaction.
antibody–immune complex are not available. The amino
acid residues Thr-5 and Cys-6 have not been described in the
X-ray structure [9]. The position of Arg-16 together with
Glu-15 within the loop connecting the β1 strand 12VEH14
and the β2 strand 17SCYWF21 appears to favour the tryptic
cleavage observed at Arg-16. The first three residues of the
peptide T3 adjacent to Arg-16, Ser-17, Cys-18 and Tyr-19
are not accessible whereas the last four amino acids
20WFSR23 are exposed. Considering the tryptic cleavage
sites and the location of Cys residues, the epitope compris-
ing the peptides T2 and T3 may be formally described by the
8
two partial sequences PVNWVEH14 and 19YWFS22.
Affinity of Synthetic Epitope Peptides
to the Antibody
On the basis of the identification of the epitope structure, the
characterisation of the binding affinity of the peptides T2
and T3 was a further aim of this study. The peptides T2 and
T3 were synthesised as described in the “Experimental”,
with the cysteine residues protected by acetamidomethyla-
tion in order to prevent the formation of disulfide bridges
following deprotection and purification. The interaction of
the synthetic peptides with the immobilised antibody was
characterised by affinity mass spectrometry. Binding of a
mixture of the two peptides to the antibody and MALDI-
FTICR-MS of the affinity elution fraction provided the
recovery and identification of both peptides T2 and T3,
consistent with the results of the epitope extraction MS
(Supplementary Figure 2). Moreover, the affinity binding
and elution of each individual peptide and MALDI-FTICR-
A structure model based on the crystal structure of
H1CRD suggests that the amino acid residues of peptide
T2 are accessible at the protein surface (Figure 5); however,
surface accessibilities from crystal structure data of the

156
R. Stefanescu, et al.: Antibody Epitope Structure of H1CRD Protein
7. Dotzauer, A., Gebhardt, U., Bieback, K., Gottke, U., Kracke, A.,
MS analysis confirmed that each peptide was binding
independently to the antibody (data not shown).
Mages, J., Lemon, S.M., Vallbracht, A.: Hepatitis A virus-specific
immunoglobulin A mediates infection of hepatocytes with hepatitis A virus
via the asialoglycoprotein receptor. J. Virol. 74, 10950–10957 (2000)
8. Treichel, U., Meyer zum Buschenfelde, K.H., Stockert, R.J., Poralla, T.,
Gerken, G.: The asialoglycoprotein receptor mediates hepatic binding
and uptake of natural hepatitis B virus particles derived from viraemic
carriers. J. Gen. Virol. 75(Pt 11), 3021–3029 (1994)
Conclusions
Proteolytic epitope excision and extraction mass spec-
trometry of an antibody complex of the H1 subunit of the
carbohydrate recognition domain of the asialoglycoprotein
receptor H1CRD provide the identification of the epitope
comprising the N-terminal domain 5T–R23, with proteo-
lytic accessibility observed at the R-16 and R-23 residues.
The primary structure characterisation and identification
of disulfide linkages of H1CRD was found to be an
essential prerequisite for the identification of the antibody
epitope. The epitope recognised by the antibody is located
on the opposite side of the carbohydrate binding site
comprising the sequence 239Gln–Asn264, consistent with
the lack of a neutralising effect of the antibody on
carbohydrate binding; moreover, the epitope binding by
the antibody was ascertained to be independent from the
presence of calcium ions. These results confirm the
feasibility of the B01 antibody as a molecular tool for
the isolation and functional characterisation of carbohy-
drate complexes of H1CRD [5]. Furthermore, proteolytic
epitope excision mass spectrometry is shown here as a
highly efficient approach to reveal molecular details of
antigen–antibody interactions.
9. Meier, M., Bider, M.D., Malashkevich, V.N., Spiess, M., Burkhard, P.:
Crystal structure of the carbohydrate recognition domain of the H1
subunit of the asialoglycoprotein receptor. J. Mol. Biol. 300, 857–865
(2000)
10. Kolatkar, A.R., Leung, A.K., Isecke, R., Brossmer, R., Drickamer, K.,
Weis, W.I.: Mechanism of N-acetylgalactosamine binding to a C-type
animal lectin carbohydrate-recognition domain. J. Biol. Chem. 273,
19502–19508 (1998)
11. Kolatkar, A.R., Weis, W.I.: Structural basis of galactose recognition by
C-type animal lectins. J. Biol. Chem. 271, 6679–6685 (1996)
12. Seimetz, D., Frei, E., Schnolzer, M., Kempf, T., Wiessler, M.: One step
isolation of bovine asialoglycoprotein receptor and its characterization
by sequence analysis and MALDI mass spectrometry. Biosci. Rep. 19,
115–124 (1999)
13. Stokmaier, D., Khorev, O., Cutting, B., Born, R., Ricklin, D., Ernst,
T.O., Boni, F., Schwingruber, K., Gentner, M., Wittwer, M.,
Spreafico, M., Vedani, A., Rabbani, S., Schwardt, O., Ernst, B.:
Design, synthesis and evaluation of monovalent ligands for the
asialoglycoprotein receptor (ASGP-R). Bioorg. Med. Chem. 17,
7254–7264 (2009)
14. Suckau, D., Kohl, J., Karwath, G., Schneider, K., Casaretto, M., Bitter-
Suermann, D., Przybylski, M.: Molecular epitope identification by
limited proteolysis of an immobilized antigen-antibody complex and
mass spectrometric peptide mapping. Proc. Natl. Acad. Sci. U.S.A. 87,
9848–9852 (1990)
15. Papac, D.I., Hoyes, J., Tomer, K.B.: Epitope mapping of the gastrin-
releasing peptide/anti-bombesin monoclonal antibody complex by
proteolysis followed by matrix-assisted laser desorption ionization mass
spectrometry. Protein Sci. 3, 1485–1492 (1994)
16. Glocker, M.O., Nock, S., Sprinzl, M., Przybylski, M.: Characterization
of surface topology and binding area in complexes of the elongation
factor proteins EF-Ts and EF-Tu GDP from Thermus thermophilus. A
study by protein-chemical modification and mass spectrometry. Chem-
istry 4, 707–715 (1998)
17. Fiedler, W., Borchers, C., Macht, M., Deininger, S.O., Przybylski, M.:
Molecular characterization of a conformational epitope of hen egg white
lysozyme by differential chemical modification of immune complexes
and mass spectrometric peptide mapping. Bioconjug. Chem. 9, 236–241
(1998)
Acknowledgements
This study was partially supported by the Deutsche
Forschungsgemeinschaft (Bonn, Germany; PR-175-14-1),
the Roche Research Foundation (Grant for R. Born) and
by the University of Konstanz.
18. Macht, M., Marquardt, A., Deininger, S.O., Damoc, E., Kohlmann, M.,
Przybylski, M.: "Affinity-proteomics": direct protein identification from
biological material using mass spectrometric epitope mapping. Anal.
Bioanal. Chem. 378, 1102–1111 (2004)
19. Stefanescu, R., Iacob, R.E., Damoc, E.N., Marquardt, A., Amstalden,
E., Manea, M., Perdivara, I., Maftei, M., Paraschiv, G., Przybylski, M.:
Mass spectrometric approaches for elucidation of antigenantibody
recognition structures in molecular immunology. Eur. J. Mass Spec-
trom. 13, 69–75 (2007)
20. Kiselar, J.G., Downard, K.M.: Direct identification of protein
epitopes by mass spectrometry without immobilization of antibody
and isolation of antibody-peptide complexes. Anal. Chem. 71, 1792–
1801 (1999)
21. Morrissey, B., Downard, K.M.: Kinetics of antigen-antibody interac-
tions employing a MALDI mass spectrometry immunoassay. Anal.
Chem. 80, 7720–7726 (2008)
22. Zhao, Y., Muir, T.W., Kent, S.B., Tischer, E., Scardina, J.M., Chait, B.
T.: Mapping protein-protein interactions by affinity-directed mass
spectrometry. Proc. Natl. Acad. Sci. U.S.A. 93, 4020–4024 (1996)
23. Born, R.: dissertation, University of Basel; Born, R.; Stokmaier, D.;
Rabbani, R.; Ernst, B. 2009; unpublished results (2008)
References
1. Drickamer, K.: Two distinct classes of carbohydrate-recognition
domains in animal lectins. J. Biol. Chem. 263, 9557–9560 (1988)
2. Stockert, R.J.: The asialoglycoprotein receptor: relationships
between structure, function, and expression. Physiol. Rev. 75,
591–609 (1995)
3. Spiess, M., Lodish, H.F.: Sequence of a second human asialoglycopro-
tein receptor: conservation of two receptor genes during evolution.
Proc. Natl. Acad. Sci. U.S.A. 82, 6465–6469 (1985)
4. Halberg, D.F., Wager, R.E., Farrell, D.C., Hildreth, J.T., Ques-
enberry, M.S., Loeb, J.A., Holland, E.C., Drickamer, K.: Major and
minor forms of the rat liver asialoglycoprotein receptor are
independent galactose-binding proteins. Primary structure and
glycosylation heterogeneity of minor receptor forms. J. Biol. Chem.
262, 9828–9838 (1987)
24. Kussmann, M., Lassing, U., Sturmer, C.A., Przybylski, M.,
Roepstorff, P.: Matrix-assisted laser desorption/ionization mass
spectrometric peptide mapping of the neural cell adhesion protein
neurolin purified by sodium dodecyl sulfate polyacrylamide gel
electrophoresis or acidic precipitation. J. Mass Spectrom. 32, 483–
493 (1997)
5. Khorev, O., Stokmaier, D., Schwardt, O., Cutting, B., Ernst, B.:
Trivalent, Gal/GalNAc-containing ligands designed for the asialoglyco-
protein receptor. Bioorg. Med. Chem. 16, 5216–5231 (2008)
6. Shia, M.A., Lodish, H.F.: The two subunits of the human asialoglyco-
protein receptor have different fates when expressed alone in fibro-
blasts. Proc. Natl. Acad. Sci. U.S.A. 86, 1158–1162 (1989)

R. Stefanescu, et al.: Antibody Epitope Structure of H1CRD Protein
157
25. Baykut, G., Jertz, R., Witt, M.: Matrix-assisted laser desorption/
ionization Fourier transform ion cyclotron resonance mass spectrometry
with pulsed in-source collision gas and in-source ion accumulation.
Rapid Commun. Mass Spectrom. 14, 1238–1247 (2000)
26. Witt, M., Fuchser, J., Baykut, G.: Fourier transform ion cyclotron resonance
mass spectrometry with NanoLC/microelectrospray ionization and matrix-
assisted laser desorption/ionization: analytical performance in peptide mass
fingerprint analysis. J. Am. Soc. Mass Spectrom. 14, 553–561 (2003)
27. Frottin, F., Martinez, A., Peynot, P., Mitra, S., Holz, R.C., Giglione, C.,
Meinnel, T.: The proteomics of N-terminal methionine cleavage. Mol.
Cell. Proteomics 5, 2336–2349 (2006)
28. Carugo, O., Cemazar, M., Zahariev, S., Hudaky, I., Gaspari, Z.,
Perczel, A., Pongor, S.: Vicinal disulfide turns. Protein Eng. 16,
637–639 (2003)
29. Cemazar, M., Zahariev, S., Lopez, J.J., Carugo, O., Jones, J.A., Hore, P.
J., Pongor, S.: Oxidative folding intermediates with nonnative disulfide
bridges between adjacent cysteine residues. Proc. Natl. Acad. Sci. U.S.
A. 100, 5754–5759 (2003)
30. Wang, X., Connor, M., Smith, R., Maciejewski, M.W., Howden, M.E.,
Nicholson, G.M., Christie, M.J., King, G.F.: Discovery and character-
ization of a family of insecticidal neurotoxins with a rare vicinal
disulfide bridge. Nat. Struct. Biol. 7, 505–513 (2000)