Determination of Aberrant O-Glycosylation in the IgA1 HingeRegion by Electron Capture Dissociation Fourier Transform-IonCyclotron Resonance Mass Spectrometry*

Received for publication, October 5, 2004, and in revised form, January 28, 2005Published, JBC Papers in Press, February 22, 2005, DOI 10.1074/jbc.M411368200

Matthew B. Renfrow‡¶¶, Helen J. Cooper‡§, Milan Tomana¶, Rose Kulhavy¶, Yoshiyuki Hiki�,Kazunori Toma**, Mark R. Emmett‡ ‡‡, Jiri Mestecky¶, Alan G. Marshall‡ ‡‡, and Jan Novak¶§§

From the ‡National High Magnetic Field Laboratory, Florida State University, Tallahassee, Florida 32310-4005,¶Departments of Microbiology and Medicine, University of Alabama at Birmingham, Birmingham, Alabama 35294,� Division of Nephrology, Department of Medicine, Fujita Health University, School of Medicine, Toyoake, 470-1192 Japan,**Research Department, The Noguchi Institute, Tokyo, 173-0003 Japan, and the ‡‡Department of Chemistry, FloridaState University, Tallahassee, Florida 32306

In a number of human diseases of chronic inflamma-tory or autoimmune character, immunoglobulin mole-cules display aberrant glycosylation patterns of N- orO-linked glycans. In IgA nephropathy, IgA1 moleculeswith incompletely galactosylated O-linked glycans inthe hinge region (HR) are present in mesangial immu-nodeposits and in circulating immune complexes. It isnot known whether the Gal deficiency in IgA1 proteinsoccurs randomly or preferentially at specific sites. Todevelop experimental approaches to address this ques-tion, the synthetic IgA1 hinge region and hinge regionfrom a naturally Gal-deficient IgA1 myeloma proteinhave been analyzed by 9.4 tesla Fourier transform-ioncyclotron resonance mass spectrometry. Fourier trans-form-ion cyclotron resonance mass spectrometry offerstwo complementary fragmentation techniques for anal-ysis of protein glycosylation by tandem mass spectrom-etry. Infrared multiphoton dissociation of isolated my-eloma IgA1 hinge region peptides confirms the aminoacid sequence of the de-glycosylated peptide and posi-tively identifies a series of fragments differing in O-glycosylation. To localize sites of O-glycan attachment,synthetic IgA1 HR glycopeptides and HR from a natu-rally Gal-deficient polymeric IgA1 myeloma proteinwere analyzed by electron capture dissociation and ac-tivated ion-electron capture dissociation. Multiple sitesof O-glycan attachment (including sites of Gal defi-ciency) in myeloma IgA1 HR glycoforms were identified(in all but one case uniquely). These results representthe first direct identification of multiple sites of O-gly-can attachment in IgA1 hinge region by mass spectrom-etry, thereby enabling future characterization at themolecular level of aberrant glycosylation of IgA1 in dis-eases such as IgA nephropathy.

Several human diseases of autoimmune or chronic inflam-matory character exhibit abnormal glycosylation of serum pro-teins, including immunoglobulins (1–7). The distinctive carbo-hydrate side chains of IgA1 molecules play a pivotal role in thepathogenesis of IgA nephropathy (IgAN)1 (8–10). IgA1 containsa hinge region (HR) between the first and second heavy chainconstant region domains with a high content of proline (Pro),serine (Ser), and threonine (Thr) (Fig. 1). Several groups haveidentified three to five O-linked glycan chains within the IgA1HR (11–15). In normal human serum IgA1, glycosylated siteshave been localized to Ser/Thr residues 225, 228, 230, 232, and236 by N-terminal sequencing methods (14) showing that Ser/Thr residues 228, 230, and 232 were occupied in most IgA1molecules (14). IgA1 O-linked glycans consist of GalNAc with a�1,3-linked Gal (11, 14, 16). Sialic acid (NeuAc) may be at-tached to GalNAc by an �2,6-linkage or to Gal by an �2,3-linkage (11, 14, 16). Carbohydrate composition of O-linkedglycans in the HR of normal human serum IgA1 is variable,and the prevailing forms include Gal-GalNAc disaccharide andits mono- and di-sialylated forms (11, 12, 14). A variant withterminal GalNAc or sialylated GalNAc is rare for normal serumIgA1 (12, 14) but is more common in IgAN patients (8, 10,17–20).

Recent reports from several laboratories support the earlierfindings that O-linked glycans in the HR of some IgA1 mole-cules in the circulation of IgAN patients are deficiently galac-tosylated (8, 17–22). The Gal-deficient IgA1 in the circulation isexclusively present in circulating immune complexes and ismostly a J-chain-containing polymer (20). In the absence ofGal, the terminal sugar is GalNAc (19, 20). Subsequently, theseaberrant O-glycans or HR glycopeptides (23, 24) are recognizedby naturally occurring antibodies with anti-glycan or anti-HRpeptide specificities (20, 24), and thus circulating immune com-plexes are formed (25). It is hypothesized that these Gal-defi-cient IgA1-containing circulating immune complexes are notefficiently cleared in IgAN patients and thus deposit in themesangium where they bind to and activate the resident me-sangial cells, inducing cellular proliferation and matrix over-production (9, 26). In fact, these Gal-deficient IgA1-containing

* This work was supported in part by National Science FoundationGrant CHE-99-09502, Florida State University, the National HighMagnetic Field Laboratory in Tallahassee, FL, and by National Insti-tutes of Health Grants DK57750, DK61525, and DE13694. The costs ofpublication of this article were defrayed in part by the payment of pagecharges. This article must therefore be hereby marked “advertisement”in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

§ Supported by the Wellcome Trust. Present address: School of Bio-sciences, University of Birmingham, Edgbaston, Birmingham B15 2TT,UK.

§§ To whom correspondence should be addressed: Dept. of Microbiol-ogy, University of Alabama, 845 19th St. S, BBRB 734 (Box 1),Birmingham, AL 35294. Tel.: 205-934-4480; Fax: 205-934-3894; E-mail:Jan_Novak@microbio.uab.edu.

¶¶ Present address: Dept. of Biochemistry and Molecular Genetics,University of Alabama at Birmingham, Birmingham, AL 35294.

1 The abbreviations used are: IgAN, IgA nephropathy; HR, hingeregion; FT-ICR, Fourier transform-ion cyclotron resonance; MS, massspectrometry; MS/MS, tandem mass spectrometry; IRMPD, infraredmultiphoton dissociation; ECD, electron capture dissociation; AI-ECD,activated ion-ECD; ESI, electrospray ionization; SWIFT, stored wave-form inverse Fourier transform; W, watt; CID, collision-induceddissociation.

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 280, No. 19, Issue of May 13, pp. 19136–19145, 2005© 2005 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

This paper is available on line at http://www.jbc.org19136

by guest on Novem

ber 27, 2020http://w

ww

.jbc.org/D

ownloaded from

complexes bind to mesangial cells more efficiently than uncom-plexed IgA1 or similarly sized IgA1-containing complexes fromhealthy controls (25). Furthermore, Gal-deficient IgA1 was de-tected in glomerular immune deposits (27, 28). These observa-tions support the hypothesis that aberrantly glycosylated IgA1-containing immune complexes participate in the pathogenesisof IgAN (4, 9, 22, 26).

Available techniques for analysis of IgA1 O-glycosylation,such as lectin binding assays, can identify the presence ofGal-deficient O-glycan chains (27–31). However, it is notknown whether the Gal deficiency in IgAN patients occursrandomly or preferentially at specific sites. Mass spectrometricanalysis of IgA1 HR O-glycosylation has characterized the het-erogeneity of glycoforms (15, 28, 30) but has not localizedspecific sites of O-glycosylation. Tandem mass spectrometry(MS/MS) has become a standard tool for the structural analysisof carbohydrates (for a review of complex carbohydrate massspectrometry see Ref. 32). Characterization of a carbohydrateor glycopeptide species is done tandem-in-time by isolation of agiven precursor ion followed by fragmentation and detection ofthe resulting fragment (product) ions. Collision-induced disso-ciation (CID) of carbohydrates can define the complete se-quence and linkages of the sugars in complex carbohydratestructures (32). Although localization of sites of glycan attach-ment in O-linked glycopeptides (32–35) and N-linked glycopep-tides (36–38) is possible by CID fragmentation, the task iscomplicated by the dominance of fragments resulting fromcleavage of glycosidic linkages rather than peptide backbonecleavages in the tandem (CID) MS/MS spectra. Also, as the sizeand number of glycan chains increase, so does the dominance ofglycosidic fragments (34).

Fourier transform-ion cyclotron resonance mass spectrome-try (FT-ICR MS) offers two complementary fragmentationtechniques for analysis of protein glycosylation by MS/MS (39).Infrared multiphoton dissociation (IRMPD) (40) induces selec-tive dissociation of the glycosidic bonds on N-linked glycans ofglycopeptides. Electron capture dissociation (ECD) fragmenta-tion, first demonstrated by Zubarev et al. (41) and reviewedrecently in Refs. 42 and 43, results in complementary cleavageof the backbone N–C-� bond with minimal loss of post-trans-lational modifications (44, 45). Haselmann et al. (46) unambig-uously localized four sites of O-glycosylation in a 6.8-kDa gly-copeptide by use of ECD fragmentation. Håkansson et al. (40,47) implemented both of these fragmentation techniques in asingle FT-ICR mass spectrometer configuration and success-fully demonstrated their use in analysis of glycoproteins. Thesestudies suggest ECD FT-ICR MS could be a valuable tool forthe analysis of Gal-deficient IgA1 HR.

Here we analyze O-glycans of synthetic IgA1 HR and anaturally Gal-deficient polymeric (p) IgA1 myeloma protein.We have shown previously (20) that this IgA1 can inhibitre-association of IgA1-containing immune complexes, suggest-ing that this IgA1 protein has properties similar to the aber-rantly glycosylated IgA1 present in the circulation of IgA ne-phropathy patients. We observed a heterogeneous populationof glycosylated IgA1 HR as in prior studies (15, 28, 30). FT-ICRMS/MS experiments localized sites of O-glycosylation in theIgA1 HR. Four and five glycan chains composed of either aGalNAc-Gal disaccharide or a GalNAc monosaccharide werelocalized to 5 of the 10 possible sites of O-glycosylation in themyeloma IgA1 HR. These analytical approaches will be used inthe future for characterization of O-glycans in IgA1 from IgANpatients.

MATERIALS AND METHODS

Synthesis of IgA1 HR Peptide and Glycopeptide Variants—A panel of19-mer hinge region peptide and glycopeptides, with an �-O-linkedGalNAc residue, corresponding to the amino acid sequence of the hu-man IgA1 HR were synthesized by and purchased from the PeptideInstitute Inc. (Osaka, Japan). Purity and molecular weight of the prep-arations were confirmed by high pressure liquid chromatography andmatrix-assisted laser desorption ionization time-of-flight mass spec-trometry (MALDI-TOF MS). The following HR peptide and glycopep-tides used: HP, VPSTPPTPSPSTPPTPSPS; 4-HP (GalNAc attached toThr4 and so on), VPS-(GalNAc)TPPTPSPSTPPTPSPS; 7-HP, VPSTPP-(GalNAc)TPSPSTPPTPSPS; 9-HP, VPSTPPTP-(GalNAc)SPSTPPTP-SPS; 11-HP, VPSTPPTPSP-(GalNAc)STPPTPSPS; 15-HP, VPSTPPTP-SPSTPP-(GalNAc)TPSPS; 4-15-HP, VPS-(GalNAc)TPP-(GalNAc)TP-(GalNAc)SP-(GalNAc)STPP-(GalNAc)TPSPS.

Isolation of IgA1 HR Glycopeptide from a Myeloma IgA1 (Mce)—HRfrom pIgA1 myeloma protein (Mce) was isolated as a tryptic-pepticfragment following the procedure described by Wolfenstein-Todel andco-workers (48, 49). The resultant preparation was lyophilized anddissolved in water before analyses.

Monosaccharide Compositional Analysis—The monosaccharidesfrom purified myeloma IgA1 HR were determined as trifluoroacetates ofmethylglycosides by gas chromatography as described previously (50).The analyses were performed with a gas chromatograph (model 5890,Hewlett-Packard, Sacramento, CA, equipped with a 25-m fused silica(0.22 mm inner diameter) OV-1701 WCOT column (Chrompack, Bridge-water, NJ) electron capture detector and a Hewlett-Packard model 3396integrator. About 10 �g of HR glycopeptide was used for the analysis.

Sample Preparation—For electrospray ionization (ESI) experiments,synthetic IgA1 HR samples (5 mM in 1:1 water/methanol, 2% aceticacid) were micro-electrosprayed (51) from an emitter consisting of a50-�m inner diameter fused silica capillary that had been mechanicallyground to a uniform thin walled tip (52) at a flow rate of 300 nl/min. Forseparated myeloma IgA1 HR glycopeptides, samples were desalted witha C18 ZipTip (Millipore, City, MA) into 30 �l of a 4:1 acetonitrile/watersolution containing 0.1% formic acid. Desalted samples were then mi-croelectrosprayed as above at a flow rate of 200–400 nl/min or from achip-based electrospray interface (NanoMate, Advion, Ithaca, NY).

FT-ICR MS of IgA1 HR Glycopeptides—IgA1 HR glycopeptides wereanalyzed with a home built 9.4 tesla ESI Q FT-ICR mass spectrometer(47, 53) under the control of a modular ICR data acquisition system (54,55). Ions were transported through a Chait-style atmosphere-to-vac-uum interface (56) and accumulated in a linear octopole ion trap (57),modified to allow improved ion ejection along the z axis (58). Analyteions were then transferred (1.0–1.4 ms) through an octopole ion guideto an open cylindrical ICR cell in which ions were captured by gatedtrapping. Ions were subjected to chirp (72–480 kHz at 150-Hz/�s)excitation and direct-mode broadband detection (1 M or 512 kWord timedomain data). Hanning apodization and one zero-fill were applied to alldata prior to fast Fourier transformation and magnitude calculation(59). Frequency domain spectra were frequency to m/z calibrated (60,61) externally from the measured ICR frequencies of Agilent calibrationmixture ions. Each displayed spectrum represents a sum of 25–50time-domain transients. Masses and m/z values were calculated withIsopro 3.1 (MS/MS software, members.aol.com/msmssoft/).

FT-ICR MS/MS of IgA1 HR Glycopeptides—Precursor ions weremass-selectively accumulated externally for 5–15 s (62, 63). Followingtransfer to the ICR cell (1.0–1.4 ms), stored waveform inverse Fouriertransform (SWIFT) (64, 65) ejection was applied for increased m/z

FIG. 1. IgA1 structural elements. IgA1 is both N- (filled circles) andO-glycosylated (open circles). The sites of O-glycosylation are in thehinge region between the first and second constant region on eachheavy chain of IgA1. The hinge region is a proline-rich segment with 10possible sites of O-glycan attachment (shown in red). Arrows identifythe tryptic-peptic fragment of HR used in this study.

IgA1 O-Glycosylation Sites Identified by FT-ICR MS 19137

by guest on Novem

ber 27, 2020http://w

ww

.jbc.org/D

ownloaded from

selectivity. An indirectly heated 10-mm diameter dispenser cathode(1109; Heat Wave, Watsonville, CA) mounted on the central axis of thesystem provided the electrons for ECD; IRMPD was performed with a40-W, 10.6-�m, CO2 laser (Synrad, Mukilteo, WA), fitted with a 2.5�beam expander. The laser beam is directed to the center of the cellthrough an off-axis BaF2 window (47). Activated ion (AI)-ECD wasperformed with the dispenser cathode and CO2 laser in tandem asdescribed below.

For ECD of synthetic IgA1 HR glycopeptides, quadrupole andSWIFT-isolated ion precursor populations were irradiated with elec-trons for 10–100 ms. During the ECD fragmentation event, the trapelectrodes were set at 10 V; the cathode was biased at �10 V, and theextraction grid was pulsed to �200 V. Immediately following the ECDevent, the trap electrodes were reset to 2 V, and potentials of 5 V for thegrid and 10 V for the cathode were applied for 1 ms to remove theremaining electrons. At all other times the cathode bias voltage was�0.1 V and the grid potential was �200 V. The cathode heating powerwas 11 W, corresponding to an electron emission current of �300 nA.Ions were frequency-sweep (“chirp”) excited (48–640 kHz, at 150-Hz/�s) and detected in direct mode (512 kWord time-domain data). Dis-played spectra represent the sum of 100–200 time-domain transients.The ECD FT-ICR MS/MS spectra were internally frequency to m/zcalibrated with respect to the precursor and the charge-reduced species.

For IRMPD of isolated myeloma IgA1 HR glycopeptides, quadrupoleand SWIFT-isolated ion precursor populations were photon-irradiatedfor 100 ms at 20% laser power (8 W). Ions were frequency-sweep excited(48–480 kHz, at 150-Hz/�s) and detected in direct mode (512 kWordtime-domain data). Displayed spectra represent the sum of 50 time-domain transients.

For AI-ECD of isolated myeloma IgA1 HR glycopeptides, quadrupoleand SWIFT isolated ion precursor populations were photon-irradiatedfor 300 ms at 16% (6.4 W) or 20% (8 W) laser power. Immediatelyfollowing photon irradiation, the isolated ion precursor populationswere irradiated with electrons for 10 or 20 ms. During the electronirradiation event, the trap electrodes were set to 4–10 V; the cathodewas biased at �4 or �5 V, and the extraction grid was pulsed to �5 V.Immediately following the ECD event, the trap electrodes were reset to2 V, and potentials of 5 V for the grid and �10 V for the cathode wereapplied for 100 ms to remove the remaining electrons (66, 67). At allother times the cathode bias voltage was �0.1 V, and the grid potential

was �200 V. The cathode heating power was 11 W, corresponding to anelectron emission current of �300 nA. Ions were frequency-sweep ex-cited (48–480 or 45–480 kHz, at 150-Hz/�s) and detected in direct mode(1 MWord time-domain data). Displayed spectra represent the sum of60–100 time-domain transients. The AI-ECD FT-ICR MS/MS spectrawere internally frequency to m/z calibrated with respect to the precur-sor, the charge-reduced species, and z5 ions. All FT-ICR MS and MS/MSspectra were analyzed by use of the modular ICR data acquisitionsystem analysis software package (55).

RESULTS

FT-ICR MS/MS of Synthetic HR Peptide and Glycopeptides—ECD fragmentation typically cleaves the peptide backboneN–C-� bond with minimal loss of post-translational modifica-tions. The exception to that rule is proline, which is cyclicaround the N–C-� bond and does not cleave into the character-istic N-terminal c and C-terminal z fragments (68) seen by ECD(69). The IgA1 HR is a proline-rich segment with 9–10 Ser/Thrresidues interspersed among 12 prolines. For that reason, pilotexperiments to determine the feasibility of characterizing IgA1HR glycosylation sites by ECD FT-ICR MS/MS were performedwith synthetic HR peptides. Fig. 2 shows the FT-ICR massspectra obtained following ECD of the peptides HP, 9-HP,4-HP, and 4-15-HP. Table I lists the identified fragments. Forpeptide HP (i.e. the unglycosylated peptide), c and/or z(�) frag-ments were detected for each amino acid except proline. Takentogether, the identified N-terminal c and C-terminal z(�) frag-ment ions provide amino acid sequence identification for theunglycosylated peptide HP (Fig. 2A). With the addition of asingle GalNAc (9-HP and 4-HP, Fig. 2, B and C) or multipleGalNAc residues (9–15-HP, Fig. 2D), the observed c and z(�)ions allow localization of the site of glycosylation based on theadditional mass of a GalNAc (203.08) in a given fragment ion.For peptide 9-HP, the observation of c8 without associatedGalNAc and [c10 � GalNAc]� unambiguously localizes the gly-can to Ser9. The remaining fragments (listed in Table I) elim-

FIG. 2. ECD FT-ICR positive ion mass spectra for synthetic HR peptides. A, VPSTPPTPSPSTPPTPSPS-NH2. All allowable cleavages aredetected, i.e. all N–C-� bonds except those of prolines are cleaved. B, VPSTPPTP-(GalNAc)SPSTPPTPSPS-NH2. The fragmentation pattern allowsfor unambiguous localization of glycosylation to Ser9; C, VPS-(GalNAc)TPPTPSPSTPPTPSPS-NH2. The fragmentation pattern allows localizationof glycosylation to Ser3/Thr4; D, VPS-(GalNAc)TPP-(GalNAc)TP-(GalNAc)SP-(GalNAc)STPP-(GalNAc)TPSPS-NH2. The fragmentation patternunambiguously localizes four of the glycosylations (to Thr7, Ser9, Ser11, and Thr15) and the final glycosylation to Ser3/Thr4. N-terminal c fragmentions and C-terminal z(�) fragment ions are indicated above and below each IgA1 HR segment, respectively.

IgA1 O-Glycosylation Sites Identified by FT-ICR MS19138

by guest on Novem

ber 27, 2020http://w

ww

.jbc.org/D

ownloaded from

inate the remaining serines and threonines as the site(s) ofglycan attachment. Similarly, the significant fragments de-tected for peptide 4-HP were [z17

� � GalNAc]�, z13, and [c6 �GalNAc]�. In contrast to MS/MS of the unglycosylated peptide,no z16 fragment was detected. The results allow location of theglycan to Ser3 or Thr4. Singly glycosylated peptides 7-HP and11-HP were also subjected to ECD analysis (data not shown).In both cases the ECD fragmentation pattern allowed for un-ambiguous localization of the glycan (Thr7 in 7-HP; and Ser11

in 11-HP). ECD analysis of the quintuply glycosylated peptide

4-15-HP (Fig. 2D) enabled unambiguous localization of four ofthe glycans to Thr7, Ser9, Ser11, and Thr15, and the final glycanto Ser3 or Thr4. These results demonstrate that direct assign-ment of native IgA1 HR O-glycosylation sites is possible byECD FT-ICR MS.

Analysis of Heterogeneity in Synthetic HR Glycopeptides—Toassess the ability of ECD fragmentation to distinguish betweenhomogeneous and heterogeneous glycosylation, mixtures of thesynthetic peptides were subjected to ECD and their fragmen-tation patterns examined. Fig. 3 shows the ECD FT-ICR mass

TABLE IFragment positive ions following ECD of the peptides, HP (VPSTPPTPSPSTPPTPSPS); 9-HP (VPSTPPTP-(GalNAc)SPSTPPTPSPS); 4-HP

(VPS-(GalNAc)TPPTPSPSTPPTPSPS); and 4-15-HP (VPS-(GalNAc)TPP-(GalNAc)TP(GalNAc)SP-GalNAc)STPP-(GalNAc)TPSPS)The reported m/z values are for the monoisotopic mass.

Peptide Measured Calculated Error Assignment

m/z m/z ppm

HP 533.283 533.283 0.0 �c11 � 2H�2�

596.343 596.341 3.4 c6623.300 623.300 0.0 �M � 2H � K�3�

636.275 636.276 �1.6 �z13� �2H � H � K�2�

637.293 637.283 15.7 �z13� � H � K�2�

699.836 699.837 �1.4 �c14 � H � K�2�

784.859 784.860 �1.3 �z16� � H � K�2�

794.445 794.441 5.0 c8798.886 798.887 �1.3 �c16 � H � K�2�

803.330 803.334 �5.0 �z8 �2H � K��

805.346 805.350 �5.0 �z8 � K��

828.374 828.376 �2.4 �z17� � H � K�2�

890.930 890.930 0.0 �c18 � H � K�2�

934.949 934.949 0.0 �M � 2H � K�2�

978.532 978.526 6.1 c101048.558 1048.532 24.8 b111065.564 1065.558 5.6 c111074.451 1074.451 0.0 �z11 �2H � K��

1076.466 1076.467 �0.9 �z11 � K��

1272.546 1272.552 �4.7 �z13-2H � K��

1274.559 1274.567 �6.3 �z13 � K��

9-HP 596.344 596.341 5.0 c6690.993 690.993 0.0 �M � 2H � K�3�

738.820 738.823 �4.1 �z13 � GalNAc � H � K�2�

794.445 794.441 5.0 c8801.375 801.377 �2.5 �c14 � GalNAc � H � K�2�

804.337 804.342 �6.2 �z8 � K��

886.400 886.400 0.0 �z16 � GalNAc � H � K�2�

900.427 900.427 0.0 �c16 � GalNAc � H � K�2�

992.467 992.470 �3.0 �c18 � GalNAc � H � K�2�

1036.487 1036.490 �2.9 �M � 2H � K�2�

1181.611 1181.605 5.1 c10 � GalNAc1268.636 1268.637 �0.8 c11 � GalNAc

4-HP 799.424 799.420 5.0 c6 � GalNAc801.375 801.377 �2.5 �c14 � GalNAc � H � K�2�

804.336 804.342 �7.5 �z8� � K��

900.424 900.427 �3.3 �c16 � GalNAc � H � K�2�

929.914 929.916 �2.2 �z17� � GalNAc � H � K�2�

992.469 992.470 �1.0 �c18 � GalNAc � H � K�2�

997.526 997.521 5.0 c8 � GalNAc1076.459 1076.467 �7.4 �z11 � K��

1181.608 1181.605 2.5 c10 � GalNAc1268.638 1268.637 0.8 c11 � GalNAc1274.559 1274.567 �6.3 �z13 � K��

4-15-HP 674.313 674.312 1.5 z5� � GalNAc

799.421 799.420 1.3 c6 � GalNAc949.113 949.113 0.0 �M � 3H�3�

970.477 970.473 4.1 z8 � GalNAc1200.604 1200.600 3.3 c8 � 2GalNAc1220.088 1220.087 0.8 Parent-2GalNAc1287.606 1287.608 �1.6 �c16 � 5GalNAc � 2H�2�

1321.623 1321.626 �2.3 Parent-GalNAc1379.658 1379.650 5.8 �c18 � 5GalNAc � 2H�2�

1423.670 1423.670 0.0 �M � 3H�2�

1587.797 1587.764 20.8 c10 � 3GalNAc1877.895 1877.876 10.1 c11 � 4GalNAc2173.055 2173.029 12.0 c14 � 4GalNAc

IgA1 O-Glycosylation Sites Identified by FT-ICR MS 19139

by guest on Novem

ber 27, 2020http://w

ww

.jbc.org/D

ownloaded from

spectrum obtained from a 1:1 mixture of peptides, 7-HP and9-HP. The fragments confirm the heterogeneity in the glycosy-lation pattern and localize sites of glycosylation to Thr7 andSer9. The c6 ions are exclusively nonglycosylated, thereby rul-ing out glycosylation at Ser3 or Thr4. c8 and [c8 � GalNAc]�

were detected, demonstrating heterogeneity in glycosylation atThr7. [c10 � GalNAc]�, but no c10, fragments were detected,demonstrating glycosylation at Ser9. There was no evidence forglycosylation at positions 11 and 15.

A 1:1:1 mixture of 7-GP, 11-GP and 15-HP was analyzed byECD FT-ICR MS (not shown). The results confirmed heteroge-neity in the glycosylation pattern. Glycosylation at positions 11and 15 was confirmed, as was the absence of glycosylation atposition 4. However, evidence for glycosylation at positions 9and 11 was ambiguous. Similarly, ECD analysis of a mixture ofall five peptides confirmed heterogeneity and the presence ofglycosylation at positions 4 and 15. The presence of glycosyla-tion at positions 7, 9, and 11 was ambiguous. The ambiguityassociated with heterogeneity can be attributed to the absenceof z fragments bracketing the residues of interest in thesemixtures. The results for 7-HP:9-HP were unambiguous be-cause the glycans are at adjacent glycan attachment sites.Thus the lack of z fragments did not interfere with the inter-pretation of the results. Similarly, unambiguous results wouldalso be expected for the following mixtures: 4-HP and 7-HP;9-HP and 11-HP; and 11-HP and 15-HP.

FT-ICR MS/MS Analysis of Isolated Myeloma IgA1 HR Gly-copeptides—HR glycopeptides were isolated from IgA1 my-eloma protein by trypsin-pepsin digestion of IgA1 as describedpreviously (49). Monosaccharide compositional analysis by gas-liquid chromatography determined the presence of GalNAc,Gal, and NeuAc (results not shown), indicating that HR wasfree of contaminating glycopeptides containing N-linked gly-cans. Isolated IgA1 HR was desalted (C18, ZipTip, Millipore)and analyzed by ESI FT-ICR MS. Fig. 4 (top) shows the massspectrum (10-s ion accumulation, sum of 25 time-domain tran-sients) of the desialylated IgA1 HR peptide separated from atrypsin-pepsin digest by size exclusion and affinity chromatog-raphy. The mass spectrum exhibits a series of triply chargedions (m/z 1479.011, 1533.028, 1600.033, 1654.737) separated by1/3 of the mass of hexose (162.05 Da) and N-acetylhexosamine(203.08 Da). Although no species corresponding to an unglyco-sylated IgA1 HR were detected, the series of ions suggests thatthe isolated IgA1 HR peptide is a heterogeneous mixture ofdifferentially glycosylated species, as seen in other myelomaIgA1 HR by mass spectrometry and lectin-binding assays (15,28, 30). The single species, m/z 1600.0333�, was isolated (Fig. 4(middle)) and fragmented by infrared multiphoton dissociation

(IRMPD, Fig. 4 (bottom)). The IRMPD product ion spectrumshows a series of triply and doubly charged ions correspondingto the loss of multiple Gal (162.05) and GalNAc (203.08) resi-dues from the isolated parent ion.

To establish the sequence of the isolated IgA1 myeloma HRglycopeptide, an equivalent sample was treated with neura-minidase and O-glycanase to remove all glycans. The predictedpeptide from a trypsin-pepsin digestion is 208HYTNPSQD-VTVPCPVPSTPPTPSPSTPPTPSPSCCHPRL246. However,two ion masses corresponding to IgA1 residues [Val216–Leu246]3� and [Val216–Leu246 � GalNAc]3� were observed. Fig.5 shows the IRMPD FT-ICR MS/MS product ion spectrum ofthe triply charged species of m/z 1046.166, [Val216–Leu246]3�.Eighteen (of 30) peptide backbone bonds are broken, confirm-ing the sequence of the isolated IgA1 HR as 216VTVPCPVPST-PPTPSPSTPPTPSPSCCHPRL246. From the known glycancomposition, the accurate mass of the glycosylated ion series inthe desialylated IgA1 HR sample (Fig. 4), and the confirmedamino acid sequence of the de-glycosylated IgA1 HR peptide(Fig. 5), glycan structures for the mixture may be inferred forthe dominant series of ions seen in Fig. 4 and are listed in TableII. ESI FT-ICR MS analysis of sialylated IgA1 HR indicated asimilar heterogeneous mixture of differentially glycosylatedspecies including multiple NeuAc residues (data not shown).For localization of sites deficient in Gal by ECD FT-ICR MS/MS, IgA1 HR trypsin-pepsin preparations were desialylated toreduce the IgA1 HR glycopeptide heterogeneity for eachsample.

Localizing Sites of O-Glycosylation by ECD FT-ICR MS/MS—Based on pilot studies with synthetic IgA1 HR and theidentification of multiply glycosylated variants in the isolatedmyeloma IgA1 HR, ECD for three IgA1 HR glycopeptidesserves to localize specific sites of O-glycan attachment. UnlikeECD fragmentation of the synthetic HR peptide, ECD of [IgA1HR � 4 GalNAc � 4 Gal]3� resulted in just a few detectableionic fragments, including two y-type ions ([y28 � 4 GalNAc �4 Gal]2� and [y26 � 4 GalNAc � 4 Gal]2�). Typically, y ionsrepresent a minor fragmentation channel in ECD (69).

Much more extensive c and z� ECD fragmentation may beachieved by prior or simultaneous activation by infrared radi-ation (at lower IR power than would be used for direct IRMPD)(45, 47, 70). Fig. 6 shows the mass spectrum obtained followingAI-ECD of a single IgA1 HR glycoform ([216VTVPCPVPST-

FIG. 3. ECD FT-ICR positive ion mass spectrum obtained froma 1:1 mixture of the peptides, 7-HP and 9-HP (VPSTPP-(GalNAc)-TPSPSTPPTPSPS-NH2 and VPSTPPTP-(GalNAc)SPSTPPTP-SPS-NH2). The two species have identical masses; however, the ECDfragmentation pattern reveals different sites of glycosylation. Absenceof C6�GalNAc and C10 serves to locate the sites of glycan attachment atThr7 and Ser9 within the mixture.

FIG. 4. ESI FT-ICR positive ion mass spectra of isolated IgA1HR peptide. Top, peptides released from IgA1 by trypsin-pepsin di-gest. The dominant series of triply charged ions are separated by 1/3 ofthe masses of GalNAc and Gal units. Middle, one of the triply chargedions seen in the top spectrum isolated externally by a quadrupole massfilter and internally by SWIFT excitation. Bottom, IRMPD MS/MSproduct ion spectrum of the isolated triply charged species. The spec-trum shows series of triply and doubly charged product ions correspond-ing to the loss of multiple GalNAc (open squares) and Gal (filled circles)residues, indicating the attachment of several O-glycan chains in thepeptide.

IgA1 O-Glycosylation Sites Identified by FT-ICR MS19140

by guest on Novem

ber 27, 2020http://w

ww

.jbc.org/D

ownloaded from

PPTPSPSTPPTPSPSCCHPRL246 � 4 GalNAc � 4 Gal � 3H]3�

ions after 300-ms photon irradiation followed by 20-ms electronirradiation). Along with the [y28 � 4 GalNAc � 4 Gal]2� and[y26 � 4 GalNAc � 4 Gal]2� fragments seen by ECD alone, 13additional fragments (c, z, and b) are seen. Taken together,these fragments enable unambiguous assignment of 4 O-glycanchains (each composed of a GalNAc-Gal disaccharide) to 4(Thr225, Thr228, Ser230, and Ser232) out of 10 possible sites inthe IgA1 HR peptide. These results are consistent with previ-ous reports localizing sites of O-glycosylation by other tech-niques (11, 14). Localization of O-glycosylation sites proceeds ina stepwise fashion by identification of fragments that comefrom the N terminus (b and c) followed by fragments that comefrom the C terminus (y and z) to identify (and eliminate) spe-cific serines and threonines with (or without) attached glycans.From the N terminus, detection of b7, c8, and c9 ions eliminatesThr217 and Ser224 as sites of O-glycan attachment. Fragments[b11 � GalNAc � Gal]� and [c12 � GalNAc � Gal]� localize oneglycan chain to Thr225. The detection of [c16 � 3 GalNAc � 3Gal]� indicates that 2 GalNAc and 2 Gal residues are distrib-uted between Thr228 and Ser230. The only remaining N-termi-nal fragment [c30 � 4 GalNAc � 4 Gal]2� indicates that theremaining two glycans are attached C-terminal to Pro231 (fivepossible sites). From the C terminus, the detection of z7, z9, z11,and z14 eliminates Ser240, Ser238, Thr236, and Thr233, respec-

tively, as sites of O-glycan attachment. The second and thirdglycan chains are localized to Ser232 and Ser230 by the detectionof [z15 � GalNAc � Gal]� followed by [z17 � 2 GalNAc � 2Gal]�, respectively. With the assignment of disaccharide chainsat Thr225 and Ser230, the remaining GalNAc and Gal residuesobserved in the [c16 � 3 GalNAc � 3 Gal]� fragment can beassigned only to Thr228. That assignment is consistent with thetwo remaining C-terminal fragments, [y26 � 4 GalNAc � 4Gal]2� and [y28 � 4 GalNAc � 4 Gal]2�, which indicate that theremaining 2 glycan chains are attached N-terminal to Ser230 (4possible sites), but because Thr217 and Ser224 are eliminated bythe presence of c8 and c9, and because the first glycan chain issimilarly assigned to Thr225, the only remaining site of glycanattachment is Thr228.

With the successful localization of four glycan chains in asingle ion species, we chose two other IgA1 HR glycoforms seenin the original series (see Fig. 4 and Table II) to fragment byAI-ECD. Based on the known structure of IgA1 O-glycans andthe fragment ions detected in the [Val216–Leu246 � 4 GalNAc �4 Gal]3� glycoform, we hypothesized that the predicted[Val216–Leu246 � 4 GalNAc � 3 Gal]3� and [Val216–Leu246 � 5GalNAc � 4 Gal]3� glycoforms would have O-glycan chainsdeficient in Gal. Fig. 7 shows the mass spectrum obtainedfollowing AI-ECD of the [IgA1 HR � 4 GalNAc � 3 Gal]3�

glycoform ([216VTVPCPVPSTPPTPSPSTPPTPSPSCCHP-RL246 � 4 GalNAc � 3 Gal � 3H]3� ions after AI-ECD frag-mentation). The fragmentation pattern (17 fragment ions) issimilar to that of the previous hinge region glycoform. ThreeGalNAc-Gal disaccharide chains are localized to Thr225, Thr228,and Ser230 as before. The difference between z15 and z14 corre-sponds to the mass of a serine residue � GalNAc, indicating aglycan chain composed of a GalNAc monosaccharide at Ser232,thus localizing a site of Gal deficiency.

Fig. 8 shows the mass spectrum obtained following AI-ECDof the [IgA1 HR � 5 GalNAc � 4 Gal]3� glycoform ([216VTVP-CPVPSTPPTPSPSTPPTPSPSCCHPRL246 � 5 GalNAc � 4Gal � 3H]3� ions after 300-ms photon irradiation, followed by10-ms electron irradiation). Although fewer fragment ions weredetected (11 fragments) compared with the other glycoforms,[c12 � GalNAc � Gal]� and [c14 � 2 GalNAc � 2 Gal]� fragmentions localize two glycan chains to Thr225 and Thr228. From the

TABLE IIDominant glycopeptide series (monoisotopic neutral masses) observed

in the tryptic-peptic preparation of IgA1 HR (Fig. 4A)Based on the determined amino acid sequence of the isolated IgA1 HR

(Fig. 5) and the known glycan compositions, the numbers of attachedGalNAc and Gal residues are predicted. Underlined glycoforms wereanalyzed by activated ion electron capture dissociation (AI-ECD) FT-ICR MS/MS.

Measured mass Calculatedmass Error Predicted no.

GalNAc residuesPredicted no.Gal residues

Da Da ppm

4434.011 4433.995 3.6 4 34596.062 4596.048 3.0 4 44799.094 4799.127 �6.9 5 44961.187 4961.180 1.4 5 55164.294 5164.259 6.7 6 5

FIG. 5. IRMPD FT-ICR MS/MS of de-glycosylated IgA1 HR [1046.166]3�

precursor ions. Eighteen (of 30) peptidebackbone bonds are broken, confirmingthe sequence of the isolated IgA1 HR as216VTVPCPVPSTPPTPSPSTPPTPSPSC-CHPRL246. N-terminal b fragment ionsand C-terminal y fragment ions are indi-cated in the IgA1 HR sequence.

IgA1 O-Glycosylation Sites Identified by FT-ICR MS 19141

by guest on Novem

ber 27, 2020http://w

ww

.jbc.org/D

ownloaded from

C terminus, the detection of z9 (without attached glycans)combined with the detection of [z14 � GalNAc]� localizes a thirdglycan chain to either Thr233 or Thr236. Subsequently the detec-tion of [z17 � 3 GalNAc � 2 Gal]� suggests that two additionalglycan chains are attached at the only two possible sites, Ser230

and Ser232, as seen in the [Val216–Leu246 � 4 GalNAc � 4 Gal]3�

glycoform (Fig. 4). Although the difference between z17 and z14

corresponds to the mass of SPS � 2 GalNAc � 2 Gal, theabsence of a [z15 � 2 GalNAc � 1 Gal]� prevents direct assig-nment of a glycan chain at each serine residue.

DISCUSSION

The present results represent the first direct identification ofmultiple sites of O-glycan attachment in the IgA1 HR by massspectrometry. A series of FT-ICR MS and MS/MS experimentsprovides a comprehensive characterization of IgA1 HRO-glycosylation.

FT-ICR MS Characterization of IgA1 HR—IgA1 HR-contain-ing trypsin-pepsin fragment is a proline-rich sequence with acluster of 10 serines and threonines that are potential sites ofO-glycan attachment (Val216–Leu246) (71, 72). Because c and zfragments rarely arise within proline (69), initial ECD FT-ICRMS/MS experiments were performed with a synthetic version

of IgA1 HR. ECD FT-ICR MS/MS of synthetic IgA1 HR showedthat ECD fragmentation results in enough products to localizeeach possible site of O-glycan attachment (Fig. 2). ECD ofseveral versions of synthetic IgA1 HR with a single GalNAcmonosaccharide attached at various sites can distinguish be-tween homogeneous and heterogeneous populations of glycosy-lated IgA1 HR (e.g. Fig. 3).

For HR isolated from IgA1 myeloma protein, IRMPD success-fully confirms the sequence of the de-glycosylated hinge region.However, IRMPD of the glycosylated IgA1 HR demonstrates thecomplexity of identifying multiple sites of O-glycan attachmentwithin a single glycoform. Although the loss of multiple GalNAcand Gal residues is easily recognized (Fig. 4, bottom), identifyingsites of O-glycan chain attachment and which chain is Gal-defi-cient by IRMPD or other slow heating methods (i.e. CID) isimpossible for several reasons. 1) the IgA1 HR glycan chains areall similar or identical in composition. 2) The 10 serines andthreonines are clustered. 3) The HR amino acid sequence is atandem repeat (49). All of these factors prevent the selectiveremoval of a single glycan chain by slow heating methods andsubsequent determination of its site of attachment by neutralloss. Only with the glycan chains intact could the site deficient inGal be identified. Based on our pilot experiments, ECD fragmen-

FIG. 6. ESI-activated ion ECD FT-ICR MS/MS spectrum of a population of quadrupole- and SWIFT-isolated [216VTVPCPVPST-PPTPSPSTPPTPSPSCCHPRL246 � 4 GalNAc � 4 Gal � 3H]3� ions from the O-glycosylated IgA1 HR peptide. Fifteen (of 30) peptidebackbone bonds are cleaved. Sites of O-glycosylation were identified from series of differentially glycosylated product ions. Squares, GalNAc.Circles, Gal. The product ions uniquely localize four GalNAc-Gal disaccharides to Thr225, Thr228, Ser230, and Ser232 (in red) and eliminate theremaining six Ser/Thr residues (in blue) as sites of O-glycan attachment. N-terminal fragment ions (c and b) and C-terminal fragment ions (z and y)are indicated above and below the IgA1 HR sequence, respectively.

IgA1 O-Glycosylation Sites Identified by FT-ICR MS19142

by guest on Novem

ber 27, 2020http://w

ww

.jbc.org/D

ownloaded from

tation of the HR thus offered the greatest promise for localizationof sites of O-glycan attachment.

Identification of IgA1 HR O-Glycosylation Sites—The patho-genesis of IgAN is characterized by immunodeposits in therenal messangium. These immunodeposits contain IgA1 withincompletely galactosylated O-linked glycans in the HR (27,28). It is not known whether the Gal deficiency in IgAN occursrandomly or at specific sites. In analysis of IgA1 HR O-glyco-sylation, previous studies have used a combination of tech-niques to assess sites of O-glycan attachment (N-terminal se-quencing and IgA specific proteases (14)), sites deficient in Gal(glycan specific lectins � specific and nonspecific proteases(27–31)), and O-glycan heterogeneity (MALDI-TOF MS (28,30)). The present results show that FT-ICR MS and MS/MS canachieve the equivalent of all of these analyses from a singlepreparation of IgA1 HR.

The HR glycopeptide, Val216–Leu246, was isolated from IgA1by a trypsin-pepsin digest. The initial FT-ICR MS spectrumshowed a series of species corresponding to the predicted massof Val216–Leu246 with the addition of varying numbers of Gal-NAc and Gal residues (Fig. 4, top). This series is consistentwith previous mass analyses of the heterogeneity of O-glycosy-lation in the HR (15, 28, 30). Three IgA1 HR glycopeptides werechosen for further characterization by AI-ECD FT-ICR MS/MS.The AI-ECD product ion spectrum for the predicted [Val216–Leu246 � 4 GalNAc � 4 Gal]3� (Fig. 6) reveals four GalNAc �Gal disaccharide chains unambiguously localized to Thr225,

Thr228, Ser230, and Ser232. For a second glycoform (Fig. 7), fourO-glycan chains could again be localized to the same sites, witha monosaccharide (GalNAc) instead of a disaccharide attachedat Ser232. The AI-ECD product ion spectrum of a third glyco-form (Fig. 8) localizes the same sites of attachment for fourGalNAc-Gal disaccharide chains plus a fifth GalNAc monosac-charide at either Thr233 or Thr236.

Baenziger and Kornfeld (11) reported that IgA1 myelomaprotein (Oso, different from the IgA1 myeloma protein reportedhere) contained O-glycans at Ser residues 224, 230, 232, 238,and 240 with a GalNAc monosaccharide at Ser224. Mattu et al.(14) localized sites of attachment to HR residues Thr225, Thr228,Ser230, Ser232, and Thr236 from pooled human serum IgA1, witha fraction of the population not glycosylated at residues Thr225

and Thr236. In our analysis, Thr225 in the three HR glycoformsalways contained an attached GalNAc-Gal disaccharide. Thesedifferences could be attributed to the difference in source of theIgA1 HR. One distinct difference is that N-terminal sequencingmethods sequence the entire population of IgA1 HR isolatedfrom IgA1, whereas we isolate individual HR glycoforms andsequence by AI-ECD fragmentation. It is also worth noting thatmethods established previously rely on the depression or ab-sence of expected signal at particular cycles in the amino acidsequencing (14), whereas an AI-ECD product ion spectrumdirectly identifies sites with and without an attached glycan.Recently, it was discovered that a single enzyme, GalNAc-transferase 2, is responsible for addition of GalNAc to IgA1 HR

FIG. 7. ESI AI-ECD FT-ICR MS/MS spectrum of quadrupole-and SWIFT-isolated [216VTVPCPVPSTPPTPSPSTPPTPSPSCCH-PRL246 � 4 GalNAc � 3 Gal � 3H]3� ions from the O-glycosylated IgA1 HR peptide. Fifteen (of 30) peptide backbone bonds are broken(notation is as in Fig. 6). The product ions uniquely localize 3 GalNAc-Gal disaccharides to Thr225, Thr228, and Ser230, and a GalNAc to Ser232.

IgA1 O-Glycosylation Sites Identified by FT-ICR MS 19143

by guest on Novem

ber 27, 2020http://w

ww

.jbc.org/D

ownloaded from

sites (73). Although there is a clear site preference, the enzymeis capable of adding GalNAc to all potential sites in IgA1 HR.Thus, the determination of which of the potential sites in IgA1HR are glycosylated is likely related to the properties andregulation of this single enzyme. Differential regulation of thisenzyme in different IgA1-producing cells may therefore resultin the heterogeneity of sites with attached glycans. We believethat the FT-ICR MS/MS techniques described here will beapplicable to the analysis of O-linked glycans of IgA1 isolatedfrom patients with diseases such as IgAN.

Conclusion—The present approach represents a new level ofanalysis of aberrant O-glycosylation found in IgA-associateddiseases. Previous mass spectrometric studies of IgA1 hetero-geneity concluded that the three glycoforms characterized hereare found in normal human serum IgA1 (15). However, assign-ment of glycan structures and attachment sites in previousstudies was based on existing knowledge of IgA1 O-glycanstructure and not direct analysis of individual IgA1 glycoformsas reported here. This study demonstrates the wealth of struc-tural information that FT-ICR MS and MS/MS analysis canprovide from a single preparation, instead of multiple tech-niques and preparations required to assess IgA1 O-glycan het-erogeneity and localize sites of attachment. Furthermore, ourFT-ICR MS/MS analysis of individual IgA1 HR glycoformsrepresents the first direct identification of sites deficient in Gal.The ability to analyze comprehensively individual IgA1 HR

glycoforms from a single source should lead to an accurate andcomplete profile of IgA1 HR glycoforms from various sources,including IgA1 myeloma protein as well as IgA1 isolated fromnormal human serum, IgAN patients, and IgA1 immunodepos-its in the mesangium. The successful identification of sites ofattachment of O-glycans in the HR of a IgA1 myeloma proteinenables characterization at the molecular level of the aberrantglycosylation of IgA1 from IgAN patients, and gives rise to newopportunities to understand the pathogenesis of diseases withaltered glycosylation pattern of immunoglobulins or otherglycoproteins.

Acknowledgments—M. B. R., H. J. C., M. R. E., and A. G. M. thankMichael J. Chalmers, Melinda A. McFarland, Christopher L.Hendrickson, and John P. Quinn for valuable discussions. We thankRhubell Brown and Stacy Hall for excellent technical assistance.

REFERENCES

1. Rudd, P., Elliott, T., Cresswell, P., Wilson, I. A., and Dwek, R. A. (2001) Science291, 2370–2376

2. Dwek, R. A. (1995) Science 269, 1234–12353. Mullinax, F., and Mullinax, G. L. (1975) Arthritis Rheum. 18, 417–4184. Mestecky, J., Novak, J., Julian, B. A., and Tomana, M. (2002) Nephrology 7,

S92–S995. Mestecky, J., Tomana, M., Matousovic, K., Konecny, K., and Julian, B. A.

(1997) Nephrology 3, 85–896. Tomana, M. (1996) in Glycoproteins in Disease (Montreuil, J., Vliegenhart,

J. F. G., and Schachter, H., eds) pp. 291–298, Elsevier, Amsterdam7. Kobata, A. (1998) Glycoconj. J. 15, 323–3318. Mestecky, J., Tomana, M., Crowley-Nowick, P. A., Moldoveanu, Z., Julian,

B. A., and Jackson, S. (1993) Contrib. Nephrol. 104, 172–182

FIG. 8. ESI AI-ECD FT-ICR MS/MS spectrum of quadrupole- and SWIFT-isolated [216VTVPCPVPSTPPTPSPSTPPTPSPSCCH-PRL246 � 5 GalNAc � 4 Gal � 3H]3� peptide ions from the Gal-deficient O-glycosylated IgA1 HR peptide. Eleven (of 30) peptide backbonebonds are cleaved (notation is as in Fig. 6); 2� � second harmonic signal from the precursor ion. Sites of GalNAc-Gal O-glycosylation at Thr225,Thr228, Ser230, and Ser232 are identified uniquely, and the site of Gal-deficient O-glycosylation is limited to two possible sites, Thr233 or Thr236.

IgA1 O-Glycosylation Sites Identified by FT-ICR MS19144

by guest on Novem

ber 27, 2020http://w

ww

.jbc.org/D

ownloaded from

9. Novak, J., Julian, B. A., Tomana, M., and Mestecky, J. (2001) J. Clin. Immu-nol. 21, 310–327

10. Smith, A. C., and Feehally, J. (2003) Springer Semin. Immunopathol. 24,477–493

11. Baenziger, J., and Kornfeld, S. (1974) J. Biol. Chem. 249, 7270–728112. Field, M. C., Dwek, R. A., Edge, C. J., and Rademacher, T. W. (1989) Biochem.

Soc. Trans. 17, 1034–103513. Iwase, H., Tanaka, A., Hiki, Y., Kokubo, T., Karakasa-Ishii, I., Kobayashi, Y.,

and Hotta, K. (1996) J. Biochem. (Tokyo) 120, 393–39714. Mattu, T. S., Pleass, R. J., Willis, A. C., Kilian, M., Wormald, M. R., Lellouch,

A. C., Rudd, P. M., Woof, J. M., and Dwek, R. A. (1998) J. Biol. Chem. 273,2260–2272

15. Novak, J., Tomana, M., Kilian, M., Coward, L., Kulhavy, R., Barnes, S., andMestecky, J. (2000) Mol. Immunol. 37, 1047–1056

16. Field, M. C., Amatayakul-Chantler, S., Rademacher, T. W., Rudd, P. M., andDwek, R. A. (1994) Biochem. J. 299, 261–275

17. Allen, A. C., Harper, S. J., and Feehally, J. (1995) Clin. Exp. Immunol. 100,470–474

18. Hiki, Y., Horii, A., Iwase, H., Tanaka, A., Toda, Y., Hotta, K., and Kobayashi,Y. (1995) Contrib. Nephrol. 111, 73–84

19. Tomana, M., Matousovic, K., Julian, B. A., Radl, J., Konecny, K., andMestecky, J. (1997) Kidney Int. 52, 509–516

20. Tomana, M., Novak, J., Julian, B. A., Matousovic, K., Konecny, K., andMestecky, J. (1999) J. Clin. Investig. 104, 73–81

21. Andre, P. M., Le Pogamp, P., and Chevet, D. (1990) J. Clin. Lab. Anal. 4,115–119

22. Coppo, R., and Amore, A. (2004) Kidney Int. 65, 1544–154723. Kokubo, T., Hiki, Y., Iwase, H., Tanaka, A., Nishikido, J., Hotta, K., and

Kobayashi, Y. (1999) Nephrol. Dial. Transplant. 14, 81–8524. Kokubo, T., Hashizume, K., Iwase, H., Arai, K., Tanaka, A., Toma, K., Hotta,

K., and Kobayashi, Y. (2000) Nephrol. Dial. Transplant. 15, 28–3325. Novak, J., Vu, H. L., Novak, L., Julian, B. A., Mestecky, J., and Tomana, M.

(2002) Kidney Int. 62, 465–47526. Julian, B. A., and Novak, J. (2004) Curr. Opin. Nephrol. Hypertens. 13,

171–17927. Allen, A. C., Bailey, E. M., Brenchley, P. E. C., Buck, K. S., Barrat, J., and

Feehally, J. (2001) Kidney Int. 60, 969–97328. Hiki, Y., Odani, H., Takahashi, M., Yasuda, Y., Nishimoto, A., Iwase, H.,

Shinzato, T., Kobayashi, Y., and Maeda, K. (2001) Kidney Int. 59,1077–1085

29. Hiki, Y., Kokubo, T., Iwase, H., Tanaka, A., Toma, K., Hotta, K., and Koba-yashi, Y. (1998) in 8th International IgA Nephropathy Symposium, Noord-wijkerhout, Netherlands, May 10–13

30. Hiki, Y., Tanaka, A., Kokubo, T., Iwase, H., Nishikido, J., Hotta, K., andKobayashi, Y. (1998) J. Am. Soc. Nephrol. 9, 577–582

31. Allen, A., and Feehally, J. (1998) Adv. Exp. Med. Biol. 435, 175–18332. Zaia, J. (2004) Mass Spectrom. Rev. 23, 161–22733. Medzihradszky, K. F., Gillece-Castro, B. L., Settineri, C. A., Townsend, R. R.,

Masiarz, F. R., and Burlingame, A. L. (1990) Biomed. Environ. Mass Spec-trom. 19, 777–781

34. Medzihradszky, K. F., Gillece-Castro, B. L., Townsend, R. R., Burlingame,A. L., and Hardy, M. R. (1996) J. Am. Soc. Mass Spectrom. 7, 319–328

35. Alving, K., Paulsen, H., and Peter-Katalinic, J. (1999) J. Mass Spectrom. 34,395–407

36. Zhu, X., Borchers, C., Bienstock, R. J., and Tomer, K. B. (2000) Biochemistry39, 11194–11204

37. Ritchie, M. A., Gill, A. C., Deery, M. J., and Lilley, K. (2002) J. Am. Soc. MassSpectrom. 13, 1065–1077

38. Nemeth, J. F., Hochgesang, G. P., Jr., Marnett, L. J., Caprioli, R. M., andHochensang, G. P., Jr. (2001) Biochemistry 40, 3109–3116

39. Marshall, A. G., Hendrickson, C. L., and Jackson, G. S. (1998) Mass Spectrom.Rev. 17, 1–35

40. Håkansson, K., Cooper, H. J., Emmett, M. R., Costello, C. E., Marshall, A. G.,and Nilsson, C. L. (2001) Anal. Chem. 73, 4530–4536

41. Zubarev, R. A., Kelleher, N. L., and McLafferty, F. W. (1998) J. Am. Chem. Soc.120, 3265–3266

42. Zubarev, R. A. (2003) Mass Spectrom. Rev. 22, 57–7743. Cooper, H. J., Håkansson, K., and Marshall, A. G. (2005) Mass Spectrom. Rev.

24, 201–22244. Kelleher, R. L., Zubarev, R. A., Bush, K., Furie, B., Furie, B. C., McLafferty,

F. W., and Walsh, C. T. (1999) Anal. Chem. 71, 4250–425345. Chalmers, M. J., Håkansson, K., Johnson, R., Smith, R., Shen, J., Emmett,

M. R., and Marshall, A. G. (2004) Proteomics 4, 970–98146. Haselmann, K. F., Budnik, B. A., Olsen, J. V., Nielsen, M. L., Reis, C. A.,

Clausen, H., Johnsen, A. H., and Zubarev, R. A. (2001) Anal. Chem. 73,2998–3005

47. Håkansson, K., Chalmers, M. J., Quinn, J. P., McFarland, M. A., Hendrickson,C. L., and Marshall, A. G. (2003) Anal. Chem. 75, 3256–3262

48. Wolfenstein-Todel, C., Frangione, B., and Franklin, E. C. (1972) Biochemistry11, 3971–3975

49. Frangione, B., and Wolfenstein-Todel, C. (1972) Proc. Natl. Acad. Sci. U. S. A.69, 3673–3676

50. Tomana, M., Prchal, J. T., Garner, L. C., Skalka, H. W., and Barker, S. A.(1984) J. Lab. Clin. Med. 103, 137–142

51. Emmett, M. R., and Caprioli, R. M. (1994) J. Am. Soc. Mass Spectrom. 5,605–613

52. Quinn, J. P., Emmett, M. R., and Marshall, A. G. (1998) in 46th ASMSConference on Mass Spectrometry and Allied Topics, Orlando, FL, May31–June 4, 1998, p. 1388, American Society for Mass Spectrometry, SantaFe, NM

53. Senko, M. W., Hendrickson, C. L., Pasa-Tolic, L., Marto, J. A., White, F. M.,Guan, S., and Marshall, A. G. (1996) Rapid Commun. Mass Spectrom. 10,1824–1828

54. Senko, M. W., Canterbury, J. D., Guan, S., and Marshall, A. G. (1996) RapidCommun. Mass Spectrom. 10, 1839–1844

55. Blakney, G. T., van der Rest, G., Johnson, J. R., Freitas, M. A., Drader, J. J.,Shi, S. D. H., Hendrickson, C. L., Kelleher, N. L., and Marshall, A. (2001) in49th ASMS Conference on Mass Spectrometry and Allied Topics, Chicago,May 27–31, 2001, American Society for Mass Spectrometry, Santa Fe, NM

56. Chowdhury, S. K., Katta, V., and Chait, B. T. (1990) Rapid Commun. MassSpectrom. 4, 81–87

57. Senko, M. W., Hendrickson, C. L., Emmett, M. R., Shi, S. D.-H., and Marshall,A. G. (1997) J. Am. Soc. Mass Spectrom. 8, 970–976

58. Wilcox, B. E., Hendrickson, C. L., and Marshall, A. G. (2002) J. Am. Soc. MassSpectrom. 13, 1304–1312

59. Marshall, A. G., and Verdun, F. R. (1990) Fourier Transforms in NMR, Optical,and Mass Spectrometry: A User’s Handbook, Elsevier, Amsterdam

60. Ledford, E. B., Jr., Rempel, D. L., and Gross, M. L. (1984) Anal. Chem. 56,2744–2748

61. Shi, S. D.-H., Drader, J. J., Freitas, M. A., Hendrickson, C. L., and Marshall,A. G. (2000) Int. J. Mass Spectrom. 195, 591–598

62. Hendrickson, C. L., Quinn, J. P., Emmett, M. R., and Marshall, A. G. (2000) in48th ASMS Conference on Mass Spectrometry and Allied Topics, LongBeach, CA, June 11–15, 2000, American Society for Mass Spectrometry,Santa Fe, NM

63. Belov, M. E., Nikolaev, E. N., Anderson, G. A., Udseth, H. R., Conrads, T. P.,Veenstra, T. D., Masselon, C. D., Gorshkov, M. V., and Smith, R. D. (2001)Anal. Chem. 73, 253–261

64. Marshall, A. G., Wang, T.-C. L., and Ricca, T. L. (1985) J. Am. Chem. Soc. 107,7893–7897

65. Guan, S., and Marshall, A. G. (1996) Int. J. Mass Spectrom. Ion Processes 157,5–37

66. Hendrickson, C. L., Blakney, G. T., Chalmers, M. J., MacKay, C. L., MccFar-land, M. A., Quinn, J. P., Renfrow, M. B., and Marshall, A. G. (2004) in 52ndASMS Conference on Mass Spectrometry and Allied Topics, Nashville, TN,May 23–27, 2004, American Society for Mass Spectrometry, Santa Fe, NM

67. McFarland, M. A., Chalmers, M. J., Quinn, J. P., Hendrickson, C. L., andMarshall, A. G. (2004) in 52nd ASMS Conference on Mass Spectrometry andAllied Topics, Nashville, TN, May 23–27, 2004, American Society for MassSpectrometry, Santa Fe, NM

68. Roepstorff, P., and Fohlman, J. (1984) Biomed. Mass Spectrom. 11, 601–60169. Cooper, H. J., Håkansson, K., Marshall, A. G., Hudgins, R. R., Haselmann,

K. F., Kjeldsen, F., Budnik, B. A., Polfer, N. C., and Zubarev, R. A. (2003)Eur. J. Mass Spectrom. 9, 221–222

70. Horn, D. M., Ge, Y., and McLafferty, F. W. (2000) Anal. Chem. 72, 4778–478471. Mestecky, J., Moro, I., Kerr, M. A., and Woof, J. M. (2005) in Mucosal Immu-

nology (Mestecky, J., Bienenstock, J., Lamm, M. E., Mayer, L., McGhee,J. R., and Strober, W., eds) pp. 153–181, Elsevier/Academic Press,Amsterdam

72. Mestecky, J., and Russell, M. W. (1986) Monogr. Allergy 19, 277–30173. Iwasaki, H., Zhang, Y., Tachibana, K., Gotoh, M., Kikuchi, N., Kwon, Y. D.,

Togayachi, A., Kudo, T., Kubota, T., and Narimatsu, H. (2003) J. Biol.Chem. 278, 5613–5621

IgA1 O-Glycosylation Sites Identified by FT-ICR MS 19145

by guest on Novem

ber 27, 2020http://w

ww

.jbc.org/D

ownloaded from

Kazunori Toma, Mark R. Emmett, Jiri Mestecky, Alan G. Marshall and Jan NovakMatthew B. Renfrow, Helen J. Cooper, Milan Tomana, Rose Kulhavy, Yoshiyuki Hiki,

SpectrometryCapture Dissociation Fourier Transform-Ion Cyclotron Resonance Mass

Glycosylation in the IgA1 Hinge Region by ElectronO-Determination of Aberrant

doi: 10.1074/jbc.M411368200 originally published online February 22, 20052005, 280:19136-19145.J. Biol. Chem. 

  10.1074/jbc.M411368200Access the most updated version of this article at doi:

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/280/19/19136.full.html#ref-list-1

This article cites 64 references, 9 of which can be accessed free at

by guest on Novem

ber 27, 2020http://w

ww

.jbc.org/D

ownloaded from