Detailed glycan analysis of serum glycoproteins of patients with congenitaldisorders of glycosylation indicates the specific defective glycan processing stepand provides an insight into pathogenesis
Michael Butler2, D. Quelhas3, Alison J. Critchley2, HubertCarchon4, Holger F. Hebestreit2, Richard G. Hibbert2,Laura Vilarinho3, E. Teles5, Gert Matthijs6, Els Schollen6,Pablo Argibay2, David J. Harvey2, Raymond A. Dwek2,Jaak Jaeken4, and Pauline M. Rudd1,2
2The Glycobiology Institute, Department of Biochemistry, OxfordUniversity, South Parks Road, Oxford, OX1 3QU, U.K.; 3MedicalGenetics Institute, Clinical Biology Department, PracËa Pedro Nunes, 88,4050 Porto, Portugal; 4Centre for Metabolic Disease, University ofLeuven, Leuven, Belgium; 5Hospital S. Jo~ao, Pediatrics Department,Porto, Portugal; and 6Center for Human Genetics, University HospitalGasthuisberg, Herestraat 49, B-3000 Leuven, Belgium
Received on January 15, 2003; revised on April 2, 2003; accepted onMay 1, 2003
The fundamental importance of correct protein glycosylationis abundantly clear in a group of diseases known as congenitaldisorders of glycosylation (CDGs). In these diseases, manybiological functions are compromised, giving rise to a widerange of severe clinical conditions. By performing detailedanalyses of the total serum glycoproteins as well as isolatedtransferrin and IgG, we have directly correlated aberrantglycosylation with a faulty glycosylation processing step. Inone patient the complete absence of complex type sugars wasconsistent with ablation of GlcNAcTase II activity. Inanother CDG type II patient, the identification of specifichybrid sugars suggested that the defective processing step wascell type±specific and involved the mannosidase III pathway.In each case, complementary serum proteome analysesrevealed significant changes in some 31 glycoproteins, includ-ing components of the complement system. This biochemicalapproach to charting diseases that involve alterations inglycan processing provides a rapid indicator of the nature,severity, and cell type specificity of the suboptimal glycanprocessing steps; allows links to genetic mutations; indicatesthe expression levels of proteins; and gives insight into thepathways affected in the disease process.
Key words: CDGs/glycosylation/IgG/proteome/transferrin
Introduction
Congenital disorders of glycosylation (CDGs) are a groupof genetic disorders characterized by low activity of one ofthe many enzymes, transporters, or other functional pro-teins required in the glycosylation processing pathway (Keir
et al., 1999; Yarema and Bertozzi, 2001). Diagnosis of thesedisorders is complicated by the fact that, although the clin-ical manifestations are varied and severe, the pathogenesisalone is insufficient to determine and classify the underlyingcause of the disease. There is an urgent need for new stra-tegies to address this problem in CDGs and in other patho-logies for which similar symptoms may reflect differentdisease processes. Previously, in a group of autoimmunediseases, alterations in glycosylation have been shown toprovide specific biochemical disease markers (Parekhet al., 1985; Watson et al., 1999) and also suggest underlyingcauses of pathogenesis. In a patient with a novel form ofCDG, we demonstrate a strategy that can be used to iden-tify the compromised enzymatic step, point to cell type±specific glycan processing pathways, allow links to geneticmutations, and indicate proteins and pathways affected inthe disease process.
CDGs can be divided into two groups based on the regionof the glycosylation pathway affected (Freeze, 2001). Group Idisorders result from defects in the assembly of the dolichol-phosphate-linked oligosaccharide precursor (Carchon et al.,1999) in the cytosol and the endoplasmic reticulum (ER)(Jaeken et al., 1997, 1998; Imbach et al., 1999; Korner et al.,1999; Schachter and Jaeken, 1999; Dupre et al., 2000).CDG-II disorders arise from defects in the processing ofthe glycans in the ER and the Golgi (Schachter and Jaeken,1999). They result in apparently normal levels of glycan siteoccupancy; however, the glycoproteins contain aberrantglycan structures. Genetic deficiencies in O-glycosylationhave been reported and in some cases associated with con-genital muscular dystrophy (Quintin et al., 1990).
As a rule, in CDGs neurological symptoms, such as psy-chomotor retardation and hypotonia, are common(Schachter et al., 1998; de Lonlay et al., 2001), stronglysuggesting that normal complex-type glycans are essentialfor proper neurological development. The scarcity of indi-viduals identified with these disorders (some 280 known asof 2001) probably indicates underdiagnosis as well as a lowsurvival rate of affected embryos (Jaeken and Matthijs,2001). Indeed, mice in which the Mgat1 gene was deleteddied after 9 days in embryo due to defects in vascularizationand neural tube closure (Ioffe and Stanley, 1994). More-over, deletion of the Mgat2 gene, coding for GlcNActransferase-II (GlcNAcT-II), caused frequent postnatallethality, and 99% of the mice died within the first weekafter birth (Wang et al., 2000). The Mgat2 (coding forGlcNAcT-II)-null mice have similar phenotypes to CDG-IIa patients and therefore provide an insight into the diseasepathogenesis. However, some of the clinical findings inmice are not observed in humans, for example, anemia,
1To whom correspondence should be addressed; e-mail:[email protected]
Glycobiology vol. 13 no. 9 # Oxford University Press 2003; all rights reserved. 601
Glycobiology vol. 13 no. 9 pp. 601±622, 2003DOI: 10.1093/glycob/cwg079
at Pennsylvania State University on February 23, 2013
http://glycob.oxfordjournals.org/D
ownloaded from
thrombocytopenia, glomerulonephritis, reticulocytosis,and kidney dysfunction. With age the mice develop signsof autoimmune disease, such as increasing amounts of auto-antibodies in the serum. These phenotypic differences couldbe due to species-specific variations or may be associatedwith aging, in which case these symptoms may develop laterin the CDG-IIa patients (Wang et al., 2001).
In general, CDGs are diagnosed from the analysis of amarker protein, usually serum transferrin (Tf ) (Yamashitaet al., 1993). A characteristic isoelectric focusing (IEF)banding pattern of normal Tf indicates that the predomin-ant glycoprotein structure contains four sialic acid residues(S4) attached to two biantennary complex glycans on eachof the two N-glycan sites (Asn413 and Asn611). A shiftof this band pattern toward the cathode is indicative of aCDG and results from a decrease in sialic acid. In CDG-I,this reduction correlates with reduced glycan site occupancyand an increased proportion of Tf molecules with only oneN-glycan or none at all. In CDG-II, aberrant glycan pro-cessing gives rise to a predominant S2 band. Althoughvaluable as a diagnostic, Tf contains a very restricted setof biantennary glycans, and IEF provides only limitedinformation about the deficient enzyme activity. In addi-tion, neither deglycosylation nor altered glycan processingof Tf has been shown to have any effect on iron binding,recognition by reticulocytes, or iron transport into the cell(Spik et al., 1988; Hershberger et al., 1991; Mason et al.,1993) and, to date, has not been linked directly with anydisease pathogenesis.
In this article we describe a strategy by which the detailedanalysis of the oligosaccharides attached to Tf, IgG, andtotal serum glycoproteins allows the faulty step in theglycosylation processing pathway to be identified. Coupledwith proteomics, which showed that a wide range of pro-teins were affected, this strategy provides a basis for diag-nosis, investigating the cause of the aberrant glycosylation,and identifying potential disease targets, ultimately openinga way to establishing a link with pathogenesis and a deeperinsight into roles for glycosylation. Proof of principle wasfirst demonstrated for a patient with a known deficiency inthe Mgat2 gene, and the strategy was then applied to apatient with liver disease of unknown cause with a type 2Tf IEF pattern.
Background to specific CDG patients
Patient HE is a girl with a typical CDG-Ia presentation.Symptoms include liver dysfunction, neurological disease,susceptibility to infection, and dysmorphic features. Thereis a confirmed deficiency in the phosphomannomutaseenzyme that converts Man-6-phosphate to Man-1-phosphate in a reaction essential for the synthesis of theoligosaccharide precursor (Glc3Man9GlcNAc2) prior to itstransfer to the protein.
Patient VT was diagnosed with CDG-IIa at 9.5 years(Jaeken et al., 1994). This syndrome is characterized bygrowth retardation, psychomotor retardation, and facialdysmorphism. However, unlike patients with CDG-Ia hehad no cerebellar hypoplasia and no peripheral neuropathy.He suffered from recurrent infections and, regrettably, diedin 2002 from a severe pulmonary infection. The affected
enzyme is UDP-N-acetylglucosaminyl:a6-D-mannoside-b-1,2-N-acetylglucosaminyl-transferase-II (GlcNAcT-II) thatcatalyzes an essential step in the conversion of an oligoman-nose to a complex type N-glycan. The human gene forGlcNAcT-II (Mgat2) has been cloned and localized tochromosome 14q21 (D'Agostaro et al., 1995; Tan et al.,1995). Southern blot analysis shows that there is only onecopy of Mgat2 in the human genome and the entire codingregion is a single exon with an open reading frame of 1341bp (Tan et al., 1996). VT is homozygous for the A1467Gpoint mutation in the Mgat2 gene resulting in a H262Rpoint mutation in the GlcNAcT-II protein.
Patient AC has severe liver disease without neurologicalinvolvement. Tf IEF was performed because an entero-hepatic presentation is associated with CDG-Ib (phospho-mannoisomerase deficiency). In fact, she showed a type IIsialotransferrin pattern, but the affected enzyme was notidentified.
Results
Tf and IgG were prepared as described in the Materials andmethods section. The concentration of IgG in serum frompatient VT (6 mg/ml) was considerably lower than that ofAC (10 mg/ml), although both were within the normalrange for children between 6 and 15 years of age (5.4±16.1 mg/ml; KIDS Foundation of New Zealand, availableat www.pidsnz.co.nz/home.htm). Sugars were releaseddirectly from the proteins in sodium dodecyl sulfate(SDS)±polyacrylamide gel electrophoresis (PAGE) gelbands, labeled with 2-aminobenzamide (2-AB, see Materialsand methods) and analyzed as described.
Glycan profiles of serum Tf
Figure 1 shows the normal phase (NP) high-performanceliquid chromatography (HPLC) glycan profiles of Tfderived from four serum samples. The two predominantpeaks (peaks 28 and 33, Figure 1 and Table I; Figure 2explains the annotations) in the control were monosialy-lated (A2G2S1) and disialylated (A2G2S2) biantennarycomplex structures. The corresponding IEF pattern of Tf(inset) indicated a predominant S4 band consistent with afully glycosylated Tf molecule containing A2G2S2 glycanstructures. As expected, the glycan profile of the CDG-I Tf(patient HE) was not significantly different from the con-trol, despite a major cathodic shift in the IEF bands of theprotein. This is consistent with the established finding thatCDG-I affects the number of glycans attached to the pro-tein but does not affect the processing of the glycans oncethey are attached.
Altered glycan processing has also been reported forCDG-Ia patients (Mills et al., 2001); however,this was notobserved from our data of patient HE. In the CDG-IIaglycan profile (patient VT) there were no detectablepeaks corresponding to those in the control sample.Matrix-assisted laser desorption/ionization time-of-flight(MALDI-TOF) mass spectrometry (MS) and exoglycosi-dase digestions coupled with HPLC indicated that the singlepeak (peak 13, Figure 1 and Table I) with a glucose unit (GU)of 6.79 had a composition of (Hex)4(GlcNAc)3(Neu5Ac)1
M. Butler et al.
602
at Pennsylvania State University on February 23, 2013
http://glycob.oxfordjournals.org/D
ownloaded from
consistent with a truncated monosialylated structure. Thisconsists of a chitobiose core with a GlcNAcGalNeu5Acantenna on the a1,3 arm and a single mannose on thea1,6 arm, (A2G1S1; data not shown). The correspondingIEF pattern (Figure 1, VT) showed a predominant S2 band,consistent with the presence of Tf containing the truncatedglycan at both glycosylation sites.
The fourth glycan profile (patient AC; Figure 1) is of aTf sample from a patient with a hitherto undesignatedCDG-II. There was a prominent peak (35, GU� 8.41) con-taining the hybrid structure GlcNAc2Man5GlcNAcGal-Neu5Ac in the HPLC profile and peaks in the MALDIprofile from this compound both with and without someresidual peptide as the result of incomplete hydrazinolysis.In addition, both HPLC and MALDI profiles containedother hybrid and complex structures (Table I). The MALDIMS spectra of both the hybrid glycan and the glycopeptidecontained ions corresponding to the presence of a freeacid group and its sodium salt. The MALDI tandem MS(MS/MS) fragmentation spectrum of the free acid con-tained a prominent ion produced by loss of sialic acid(291 mass units). This hybrid structure was also found with-out a sialic acid (Neu5Ac) terminating the 3-antenna(GU� 7.46). Its structure was confirmed by enzyme arrayanalysis; in addition, the presence of the unusual Man5
hybrid moiety was confirmed by MALDI MS using atandem quadrupole time-of-flight (Q-TOF) instrument toobtain fragmentation (Figure 3). The spectrum containedan ion at m/z 833.3 ([Hex]5), consistent with a hybrid struc-ture, and a prominent peak at 365 mass units less thanthe molecular ion indicating that the a1,3 arm containsHexNAc and hexose, which were subsequently identifiedby exoglycosidase digestion as GlcNAc and galactose.
Glycan profiles of serum IgG
In normal serum IgG, approximately 5% of light chains areglycosylated (Youings et al., unpublished data); however,no light chain glycosylation was detected in either patient.Figure 4 shows the serum IgG profiles for patients VT andAC and for a pooled control serum preparation. In contrastto control IgG, in which 19 glycans were clearly identified,consistent with previous studies (Jefferis et al., 1990;Wormald et al., 1997) (Table II), IgG from VT containsan extremely restricted set of glycan structures, none ofwhich are present in control IgG. More than 70% of theheavy chain glycan pool consists of a single core fucosylatedmonoantennary sialylated glycan structure (FcA1G1S1;peak 17 in Table II). The absence of complex type glycosy-lation is consistent with the ablation of the activity ofGlcNAcT-II, and the data indicate that normal processingof IgG has terminated in the Golgi following the removal ofthe mannose residues to generate the trimannosyl core andthe processing of the a1,3 arm by GlcNAcT-I. This defi-ciency would be expected to decrease the stability of IgGtoward proteases, (Leatherbarrow and Dwek, 1984) todecrease the affinity for the Fcg receptor on monocytes(Leatherbarrow et al., 1985), and to provide a potentialepitope for the mannose receptor and, if multiply presented,for mannose-binding lectin (Malhotra et al., 1995). Thepredominance of a single IgG glycoform is in strong con-trast with control IgG, where many differently glycosylatedvariants are present (Figure 4, control).
IgG heavy chains from patient AC were also aberrantlyglycosylated. Compared with the control, there was anoverall increase in galactosylation leading to decreasedamounts of nongalactosylated (G0) and monogalactosy-lated (G1) biantennary-type glycans and an increase insialylation (Figure 4). Seventy percent of glycans in thecontrol samples were of the G0 and G1 type, compared to56% in patient AC. Sialylated structures account for 14% inthe control and 26% in patient AC.
Glycan profiles of whole serum
Figure 5 (top) shows the NP HPLC profiles of the 2-AB-labeled N-glycans released from serum samples from VT,
Fig. 1. Isoelectric and NP HPLC glycosylation profiles of transferrinpurified from control and CDG serum. Tf samples were purifiedfrom serum by immunoaffinity chromatography. The glycans wereremoved enzymatically, labeled with 2-AB, and examined by NP HPLC.The resulting profiles covering the range of GU values from 5 to 13 frompatients HE, AC, VT, and the control are shown. The inserts show theIEF patterns obtained for the corresponding Tf. The degree of sialylationof each band from the IEF pattern is also shown.
Fig. 2. Diagrammatic representation of glycans. Each monosaccharide isrepresented by a distinct shape. This does not relate to the actualstructure. The angle of the line indicates the linkage position at reducingterminus. The type of line represents anomercity. N-acetylation isrepresented by a filled shape.
Proteomics and glycan analysis of CDGs
603
at Pennsylvania State University on February 23, 2013
http://glycob.oxfordjournals.org/D
ownloaded from
Table I. Analysis of transferrin glycans by HPLC and MALDI MS
All sialylated structures were found as a range of sodium salts. Some structures were not detected by MALDI, due tolow abundance or loss of sialic acids. Ax, N2M3 core� xGlcNAc;M, mannose; G, galactose; S, sialic acid; B, bisectingGlcNAc.1Peak assignment in Figure 1.2Monisitopic mass of the [M�Na]� ion of the non-2-AB-labeled glycan. All measured masses were within� 0.1 massunits of the calculated value.3Glycans analyzed from transferrin purchased from Sigma.4Mean values of glycans analyzed from transferrin extracted from three serum samples from healthy individuals.mMonosodium salt of sialic acid.dDisodium salt of sialic acid.
M. Butler et al.
604
at Pennsylvania State University on February 23, 2013
http://glycob.oxfordjournals.org/D
ownloaded from
AC, and a control preparation by optimized hydrazinolysis.Hydrazinolysis was used because in control experimentswhen glycans were released by peptide N-glycosidase F(PNGase F), it was clear that not all sugars were releasedby this enzyme even under reducing conditions (data notshown). In a typical control serum taken from healthyindividuals, 29 peaks with GU values from 5.5 to 10.6were assigned N-glycan structures, and no significant dif-ferences in glycosylation were observed. (4 healthy indivi-duals and 3 pools of serum from 30 healthy individuals wereanalyzed). The structural assignment of these peaks wasbased on their elution positions measured in GU values(Figure 5, top; Table III) compared with standard glycansand confirmed by systematic digestion of the glycans byexoglycosidase arrays (Figure 5, bottom) (Guile et al.,1996), MALDI TOF, and electrospray (ESI) MS (Table III).Seventy-four percent of the structures were biantennary,and 26% were triantennary. Fluorescent labeling of theglycan pool with 2-AB is nonselective, and integration ofthe peaks in the HPLC data gives quantitative data fromwhich the relative proportions of the glycan structures canbe determined. The major biantennary structure was non-fucosylated and disialylated (A2G2S2, peak 33 in Table III)whereas the major triantennary structure was nonfucosy-lated and disialylated (A3G2S2, peak 36 in Table III). Therewas a significant proportion (64%) of core fucosylatedstructures.
The CDG-II serum (VT) contained none of the peaksidentified in the control sample (Figure 5; Table III). Thepredominant structures (peaks 13 and 17) were the fucosy-lated and nonfucosylated sialylated hybrid glycans(FcA1G1S1 and A1G1S1) derived from the trimannosyl-chitobiose core (M3N2) that is the natural substrate
for GlcNAcT-II. HPLC and MALDI MS indicated thepresence of some digalactosylated (peaks 27, 30) and dis-ialylated glycans (peaks 32, 37). These additional galactoseand sialic acid residues may be attached to the bisectingGlcNAc, which is not normally found on glycans derivedfrom serum proteins. This type of structure was identified(Wang et al., 2001) in tissue samples from Mgat2-null mice.The elongation on bisecting GlcNAc in the absence of the6-arm branch suggests that there is usually restricted assessof galactosyl transferase to the bisecting GlcNAc residue.
Fig. 3. MS fragmentation of the hybrid glycan from AC obtained with the MALDI QTOF instrument. The fragmentation nomenclature is asproposed by Domon and Costello (1988).
Fig. 4. NP HPLC glycosylation profiles of IgG purified from control andCDG serum. IgG samples were isolated from serum using Protein Gaffinity chromatography. The glycans were removed enzymatically,labeled with 2-AB, and separated on NP HPLC. The resulting profiles ofcontrol and patients VT and AC are shown.
Proteomics and glycan analysis of CDGs
605
at Pennsylvania State University on February 23, 2013
http://glycob.oxfordjournals.org/D
ownloaded from
Table II. Analysis of IgG glycans by HPLC and MALDI MS
M. Butler et al.
606
at Pennsylvania State University on February 23, 2013
http://glycob.oxfordjournals.org/D
ownloaded from
In contrast to VT, and predictably because only siteoccupancy is disturbed in CDG-I, the serum glycans frompatient HE were not significantly different from the control(data not shown).
The serum profile from patient AC is shown in Figure 5(top). Of particular significance is the presence of oligoman-nose-type glycans, such as M5N2 at GU 6.2 (peak 7,Tables I and III). The Tf glycan profile shows a 10-foldincrease in the level of the oligomannose glycan M5N2(peak 7, GU 6.2) and also the presence of hybrid structures(peaks 23 and 35 GU 7.46 and 8.41, Table I). Although thehybrid structures may be masked by other large peaks athigher GU values in the serum glycan profile, their presenceis confirmed by corresponding peaks from MS and enzymearray analysis. In patient AC, IgG secreted from B cellscontains the same glycans as normal IgG; however, therelative proportions of the sugars are different (Figure 4,Table II). These data indicate that the altered glycosyla-tion is not confined to serum glycoproteins predominantly
synthesized in the liver. In addition, it is clear that Bcells contain some compensatory pathway because nooligomannose or hybrid-type glycans were identifiedon IgG.
The glycan pools from serum from patients VT and ACand from healthy controls were resolved on the basis ofcharge using weak anion exchange chromatography(Figure 6, top). The extent of sialylation was determinedby comparison with standard fetuin sugars. The variouslycharged fractions from patient AC were resolved by NPHPLC to confirm the assignment of the sialylated structuresin Table III (Figure 6, bottom). Consistent with thecomplete ablation of GlcNAcT-II activity and the NPHPLC data, which indicated the absence of complex-typeglycans, the glycoproteins from the VT serum containedonly neutral and monosialylated glycans. AC contained sig-nificantly more neutral structures, consistent with theNP HPLC data, which suggested that this patient has amannosidase deficiency.
Table II. continued
Ax, N2M3 core � xGlcNAc; N, GlcNAc; M, mannose; G, galactose; S, sialic acid; B, bisecting GlcNAc;Fc, fucose.1Peak assignment in Figure 4.2Calculated monoisotopic mass of the [M � Na]� ion. Measured masses were all within� 0.2 mass units.
Proteomics and glycan analysis of CDGs
607
at Pennsylvania State University on February 23, 2013
http://glycob.oxfordjournals.org/D
ownloaded from
A summary of the characteristics of each glycan profilederived from these three patients and a control samplealong with the confirmed or suggested enzyme defects andrelated clinical diagnosis is presented in Table IV.
Differential proteome analysis of the serum proteins fromVT, AC, and control
Further insight into the changes in protein glycosylationand protein expression in patients VT and AC was achievedby differential proteome analysis of the serum proteins bytwo-dimensional gel electrophoresis (2-DE). Representative2-DE images of proteins separated on 3±10 nonlinear pHgradients are shown in Figure 7 for the two patients and
a serum control pool prepared from 28 healthy normalindividuals (blood group AB).
Extensive mapping of the serum proteome enabled thedetailed analysis of the expression profiles of all proteinisoforms of transferrin detectable by 2-DE in the molecularweight range of 10±140 kDa and 3±10 pI. These spots areindicated in Figure 8A±C. The majority of Tf molecules arepresent as differently charged isoforms of a 77-kDa protein.Lower-molecular-weight isoforms of 60 kDa, 41 kDa,35 kDa, 21 kDa, and 15±17 kDa represent truncated mole-cules that can be present in normal serum, probably byproteolysis in vivo. Post±sample collection protein digestionis unlikely to account for these truncated molecules becauseproteases were very effectively controlled by serum proteaseinhibitors and sample storage time at low temperature waskept to a minimum. Analysis of the two patients' seraindicates the selective presence of some of these low-molecular-weight Tf molecules. The data are summarizedschematically in Figure 8, B3. In the serum of patient VT,compound 11 is present in addition to the main 77-kDa Tfarray (compounds 1±6), whereas in the serum of patient AC,compounds 12 and 16 are present (see Figure 7). The differ-ent abundance of these Tf fragments might well arisefrom different susceptibility to degradation of the proteinisoforms due to changes in glycosylation of the proteins.Differences in protein isoform expression of the 77-kDaisoforms as resolved on the 2-DE are shown in Figure 8C.These differences correlate well with the IEF patterns for Tfshown in Figure 1 in that a shift of isoform S4 to S2 is clearlyvisible for Tf from patient VT and even more pronounced,alongside a general reduction of Tf levels, in patient AC.
Differential analysis of the total 2-DE protein spotpatterns provides a more comprehensive insight into thechanges in protein expression and modification induced bythe genetic defect in VT and by the putatively low activity ofmannosidase III in AC. These analyses reveal changes inprotein expression in response to the mutation per se and/orthe disease pathogenesis. Changes in glycosylation can alsobe revealed, provided they give rise to a change in thephysicochemical characteristics of the proteins (molecularweight and isoelectric point) that can be spatially resolvedby the pH and gel gradient.
The differences in serum protein expression and spotposition detected for patients VT and AC versus controlare listed in Table V. A range of proteins, including clust-erin, apolipoprotein E and L, Tf, a1-antitrypsin, comple-ment C3 and C4, antithrombin III, fetuin, and a-1 acidglycoprotein, known to be affected in CDG-Ia patients,are missing or modified in their isoelectric point and mole-cular weight in the two patients (Yuasa et al., 1995; Henryet al., 1997, 1999). As in CDG-I, clusterin is absent in theseCDG-II patients. Additional proteins have been identifiedby our analysis that to our knowledge have not beenreported to be affected by CDG. These molecules are theinter±alpha trypsin inhibitor heavy chain 4, complementfactors B, complement factor H±related protein I, plasmaglutathione peroxidase, leucine-rich a-2-glycoprotein,serum paraoxonase arylesterase 1, plasma retinol bindingprotein, tetranectin, and transthyretin. These moleculesare all glycosylated, and many play important roles ininflammation.
Fig. 5. (Top) NP HPLC glycosylation profiles of whole serum glycanspurified from control and CDG serum. The glycans from serum samplesof patients VT, AC, and control were released by hydrazinolysis, labeledwith 2-AB, and separated by NP HPLC. The resulting profile from GUvalues 5±13 are shown. (Bottom) Exoglycosidase enzyme array of controlwhole serum. The glycan structures were assigned using exoglycosidaseenzyme arrays. The digest of control whole serum is shown, containingArthrobacter sialidase (ABS), bovine testes b-galactosidase (BTG),bovine kidney fucosidase (BKF), and Streptococcus pneumoniaeb-N,acetylhexosaminidase (SPH). The assignment of FcA2G2S2 (peak39) is highlighted.
M. Butler et al.
608
at Pennsylvania State University on February 23, 2013
http://glycob.oxfordjournals.org/D
ownloaded from
Table III. Analysis of glycans released from whole serum using HPLC and MALDI MS
Proteomics and glycan analysis of CDGs
609
at Pennsylvania State University on February 23, 2013
http://glycob.oxfordjournals.org/D
ownloaded from
Table III. continued
M. Butler et al.
610
at Pennsylvania State University on February 23, 2013
http://glycob.oxfordjournals.org/D
ownloaded from
Changes observed in the protein pattern only for VTare found in antithrombin III, plasma glutathioneperoxidase, leucine-rich a-2-glycoprotein, serum para-oxonase arylesterase 1, plasma retinol binding protein,and tetranectin. Interestingly, posttranslational modifica-tion of the prominent a-1 acid glycoprotein appears notto be affected in this patient. For patient AC, specificchanges are observed for the proteins transthyretin, car-boxypeptidase N, complement factor B, fetuin A, b2-glycoprotein 1, complement C3, and serum amyloid P. Incontrast to the reduced molecular weight of serum amyloidP found in CDG-I, the expression level of this proteinappears to be down-regulated in this potentially new formof CDG-II.
Mutation analysis of ManII and ManIIx on AC
Man IIa and Man IIx were sequenced at the cDNA levelfor patient AC and her parents. No pathogenic mutationswere found. Two silent polymorphisms were found in thecoding sequence of ManIIa: the patient is heterozygous fora 1299 G4A transition and homozygous for a 1449 G4Atransition. In addition, the patient is heterozygous for asequence variant 1235 A4G that results in a conservedsubstitution at the protein level (Q412R) in ManIIx. Thisvariation was found in the normal population with afrequency of 27%. Thus these genes are probably not thecause of the deficiency in this patient.
Discussion
Since the first identification of a CDG in 1980, there hasbeen a growing list of cases that can be ascribed to various
mutations of enzymes associated with protein glycosylation(Jaeken et al., 1980). However, to date only 9 defectiveenzymes have been identified from the 50 or so enzymesinvolved in the synthetic and processing steps of N-glyco-sylation (Freeze, 2001; Jaeken and Matthijs, 2001). Thissuggests that mutations leading to decreased activity ofmany of these enzymes may be incompatible with embryo-nic development and survival.
Strategy for identifying glycosylation processingdisorders
Until now, diagnosis of CDGs has been based on an alteredIEF profile of Tf as a marker protein. However, this tech-nique is inadequate for complete characterization of theaberrant glycan structures, let alone pinpointing the defec-tive processing step or the glycoproteins that might beaffected. The strategy described here combines proteomics,which suggests potential disease targets and links to patho-genesis, with the detailed glycan analysis of a pool ofproteins, such as serum. Even the analysis of the glycanstructures of a specific isolated protein, such as IgG or Tf,can give a good indication of the key abnormal glycans, aswas evident in our patient samples. Glycan analysis is usefulfor identifying glycosylation disorders for three reasons.First, a comparison of the glycan profiles of glycoproteinsfrom a patient with those from a healthy control can high-light unexpected glycan structures and lead to the identifi-cation of a defective enzyme or blocked processing step inthe patient. Second, the extent of the glycosylation changesmay provide an insight into the severity of the enzymedeficiency or block. Third, the glycan profile of proteinsexpressed in different cells, such as Tf and IgG expressed in
Table III. continued
Ax, N2M3 core � xGlcNAc; N, GlcNAc; M, mannose; G, galactose; S, sialic acid; B, bisecting GlcNAc;Fc, fucose. All sialylated structures were found as a range of sodium salts. Some structures were notdetected by MALDI due to low abudance or loss of sialic acids.1Peak assignment in Figure 5.2Calculated monoisotopic mass of the [M �Na]� ion. Measured masses were all within� 0.2 mass units.mMonosodium salt of sialic acid.dFound as disodium salt of sialic acid.
Proteomics and glycan analysis of CDGs
611
at Pennsylvania State University on February 23, 2013
http://glycob.oxfordjournals.org/D
ownloaded from
hepatocytes and B cells, respectively, can indicate the cellspecificity of the disorder.
Serum proteome of patients VT and AC suggests diseasetargets
The 2-DE of serum glycoproteins from patients VT and ACand from a healthy control were compared to obtain anoverview of the range of proteins directly or indirectlyaffected by structural changes in the glycans. In bothpatients the comparison with control serum revealed thatmany proteins were unaffected; however, in addition totransferrin and immunoglobulin, a range of serum glyco-proteins was identified which did have altered expressionlevels and physicochemical characteristics (Table V).Probably more proteins where aberrant glycosylation
does not lead to a significant change in molecular size orcharge are also affected. Other proteins, including manycomplement components, were not detected on the CDGproteographs. This suggests that their expression has beendirectly or indirectly down-regulated as a consequence ofthe glycosylation defect. It is also possible that in somecases terminal Man or exposed GlcNAc residues on unpro-cessed sugars may have led to rapid clearance of some glyco-proteins, thus lowering their concentration in the serum.
Glycan and proteome analysis of serum glycoproteins inpatient VT is consistent with ablation of GlcNAcT-IIenzyme activity and recurrent infections
In the case of patient VT, a single mutation in the Mgat2gene leads to a completely defective GlcNAcT-II enzyme.This is a stringent mutation that leads to a loss of >98% ofthe normal enzyme activity of GlcNAcT-II in fibroblastsand mononuclear blood cells isolated from patient VT(Jaeken et al., 1996). The glycan analysis of serum glyco-proteins (Figure 5, Table III) was consistent with ametabolic block in glycan processing that leads to the accu-mulation of truncated glycan structures, which are theexpected substrates of the GlcNAcT-II enzyme. The 2-DEanalysis shows a decreased negative charge (more basic pI )for some multisialylated proteins that most probablyreflects missing sialyl groups.
Not surprisingly, the serum proteome analysis does notidentify a protein factor that could directly be associatedwith the clinical characteristics of the syndrome (growthretardation, mental retardation, and facial dysmorphism)because these may be expected to involve proteins expressedin tissues other than the liver and B cells from which serumproteins are mainly derived. However, the expression levels,serum concentrations, structure, and possibly the functionof proteins that play a vital role in blood coagulation(antithrombin), metabolism (e.g., lipid transport andexchange), and immunity (complement factors) appears tobe affected with the obvious consequences for the fitness ofthe patient.
The severe recurrent infections suffered by patient VTand eventual death by severe pulmonary infection maywell have been the clinical manifestations of aberrant struc-tures and levels of components of the immune system. Thelevel of IgG in the serum (~6 mg/ml) was at the low end ofthe normal range and may have contributed to the diseasepathogenesis. Over 70% of the heavy chain glycan poolcontained a single core fucosylated monoantennary sialy-lated glycan structure (FcA1G1S1, peak 17) that may com-promise both the stability and receptor-mediated functionsof IgG (Jefferis et al., 1998; Ghirlando et al., 1999). Incontrast to normal serum IgG (Youings et al., 1996),no light chain glycosylation was detected in the IgG ofpatient VT, suggesting either defective glycan attachmentor that the IgG was derived from only a subset of B cells.Low levels of antibodies have been reported in the mousemodel of CDG-IIa, in which a partial block in pre±B celldevelopment leads to reduced levels of mature B cells(Wang et al., unpublished data). Additional galactose andsialic acid residues may be attached to the bisecting GlcNAcresidue (peaks 27, 30, 32, 37; Figure 5) which is not normally
Fig. 6. (Top) WAX glycosylation profiles of whole serum purified fromcontrol and CDG serum. The glycans from serum samples of patientsVT, AC, and control were released by hydrazinolysis, labeled with 2-AB,and separated over 22 min by WAX HPLC. The serum profiles werecompared with a fetuin standard in which the peaks for neutral mono-,di-, tri-, and tetrasialylated glycans could be identified. (Bottom) Analysisof WAX fractions from serum from the control and patient AC toconfirm assignment of sialylated structures (Table III). Fractionscorresponding to each of the sialylated groups of glycans obtained fromserum proteins were separated further by normal phase HPLC.
M. Butler et al.
612
at Pennsylvania State University on February 23, 2013
http://glycob.oxfordjournals.org/D
ownloaded from
elongated in the serum (Wang et al., 2001). This type ofstructure was identified in tissue samples from Mgat2-nullmice and supports the use of the mouse model for studyingthe human disease.
Patient VT represents a clear example of how a single pointmutation in the gene of a glycosylation enzyme (GlcNAcT-II) can result in structural changes to a range of glyco-proteins that are expressed in various tissues, leading tosevere multisystemic clinical symptoms. The severity of symp-toms caused by a defect in this enzyme is also clear fromthe low survival rates in both human patients and animalmodels. Mgat2-null mice have a reported survival of only 1%into the second week of postnatal life (Wang et al., 2001).
Glycan analysis is consistent with a block in thea-mannosidase III pathway in patient AC
In patient AC, a different set of aberrant glycans accumu-late. These are pentamannosyl hybrid type glycans with
varying lengths of a1-3-linked antenna. The data indicatethat GlcNAcT-I can modify the a1,3 arm and, subsequently,GalTase and sialylTase can further process the GlcNAcresidue, implying that the later part of the glycan processingpathway is unaffected. The defect appears to be the inabilityto cleave two terminal mannose residues with a1- and a1-6linkages to the a1-6-linked antenna of the core structure.
There are two major classes of mannosidase enzymesinvolved in N-glycosylation processing (Moremen et al.,1994). Class 1 a-mannosidases specifically cleave a1,2-linked terminal mannose residues and typically trim Man9-GlcNAc2 to a Man5GlcNAc2 in the ER and Golgi. Class 2a-mannosidases cleave terminal mannose residues from awider range of linkages, including a1,3; a1,6; and a1,2. Theclass 2 mannosidases localized to the Golgi typically cleavea1,3 and a1,6 linked mannose residues, a prerequisite forprocessing the a1,6 antenna of the core N-glycan structure.
The glycan analysis of glycoproteins from patient AC isconsistent with the low activity of a class 2 a-mannosidase
Fig. 7. Serum proteins from patients VT and AC as separated by 2-DE. Separation was carried out on 3±10 nonlinear immobilized pH gradients (pI) and9±16% SDS±PAGE gradient gels. The serum proteins from pooled serum of 30 healthy individuals is shown as control. The spots were visualized by afluorescent dye (OGT MP17), and the images were obtained by a scanner.
Table IV. Patients' samples analyzed and the associated clinical manifestations
Sample Glycan structures Suggested (s) or confirmed(c) enzymic defect
Phenotypic characteristics
Control serum All complex structures; ratio ofdi-/mono-sialylated structures� 2.2
Normal Normal
HE Structural pattern identical to control CDG- Ia; PMM2deficiency (c)
Dysmorphic features; liver dysfunction;susceptibility to infection;neurological disease
VT All hybrid structures;no complex structures
CDG-IIa; GlcNAcTIIdeficiency (c)
Dysmorphic features; severe psychomotorretardation; no peripheral neuropathy
AC 47% complex structures;53% hybrid structures;type 2 IEF Tf pattern
CDG- IIx;a-mannosidase II/III (s)
Mainly severe liver disease withoutneurological involvement
Proteomics and glycan analysis of CDGs
613
at Pennsylvania State University on February 23, 2013
http://glycob.oxfordjournals.org/D
ownloaded from
that cleaves mannose residues in 1,3 and 1,6 linkages(Figure 9).
The most well-characterized class 2 enzyme is the a3/6-mannosidase II (MII), which initiates the first committedstep of the conversion of a hybrid to a complex glycan(Schachter, 2001). The enzyme requires the addition of aterminal GlcNAc residue prior to mannose cleavage(GlcNAcMan5GlcNAc2 to GlcNAcMan3GlcNAc2). Prioraddition of the terminal GlcNAc is affected by the enzymeGlcNAcT-I. Evidence for the accumulation of pentaman-nosyl hybrid-type glycans following mutation of the man-nosidase II gene has been reported for ricin-resistant babyhamster kidney cells in culture (Hughes and Feeney, 1986).The knockout mouse lacking a functional mannosidase IIgene showed a phenotype in homozygotes of dyserythropoi-esis similar to the pathology of human congenital dysery-thropoietic anemia (HEMPAS) (Chui et al., 1997). The MIIknockout mice also showed a late-onset autoimmune dis-ease similar to human systemic lupus erythematosus (Chuiet al., 1997). Although normal erythropoiesis was inhibitedin these mice, the nonerythroid cells continued to producenormal complex N-glycans, suggesting the existence ofalternative mannosidase activities.
Two alternative routes of mannose cleavage in glycanprocessing have been described and suggest a network ofbiosynthetic pathways associated with the mannosidaseenzymes (Figure 9). The human a mannosidase IIx (Mx)is an enzyme closely related to MII. The gene encodingMx (Man2a2) has been identified in the human genomeby cross-hybridization with the cDNA from human MII(Misago, 1995; Oh-Eda et al., 2001). Loss of activity ofthis enzyme in Mx knockout mice results in ineffective
spermatogenesis and infertility of the males (Fukuda andAkama, 2002). Mannose cleavage can also be provided bya3/6-mannosidase III (MIII) that has a GlcNAcT-I-inde-pendent activity and converts Man5GlcNAc2 to Man3-
GlcNAc2. This is described as a cobalt-activated enzymethat has been detected in rat liver Golgi, in baby hamsterkidney cells, and significantly in the nonerythroid cells ofMII-null mice (Moremen, 2002).
The role of each of these three alternative class 2 manno-sidase enzymes (MII, Mx, and MIII) in glycosylationprocessing is unclear. Although each can trim terminalmannose groups from a hybrid glycan, the substrate speci-ficities differ. It is also unlikely that all the enzymes areexpressed in a single cell type and tissue- and species-specific expression of distinct mannosidases has been shown(Moremen et al., 1994).
The existence of multiple forms of the mannosidasesallows the possibility that a compensatory mechanism canlimit potential damage caused by a mutation in one indivi-dual enzyme. Also the clinical symptoms of any defect arelikely to depend on the relative distribution of enzyme iso-forms in different tissues (Moremen et al., 1994). The rela-tively mild condition of anemia that occurs in the MII-nullmice can be explained by the processing of normal complexglycans in nonerythroid cells, through the apparent activityof MIII detected in these cells (Chui et al., 1997; Moremen,2002). The enzymatic activity determined for Mx appears tobe minimal and is less likely to account for a compensatorymechanism in these animals (Fukuda and Akama, 2002).The relatively mild pathological effects of the loss of humanMII in HEMPAS could also be explained by a compensa-tory mechanism of activity offered by an alternative man-nosidase isoform.
The unexpected hybrid glycans identified in the serumproteins of AC are the natural substrates for mannosidaseII, IIx, or III, and their presence suggests the location of anenzyme defect in the processing pathway. However, theputative metabolic block is incomplete because there isstill a significant proportion (50%) of normal complex gly-cans, both in the Tf sample and the serum sample (Tables Iand III). These observations could be explained by at leastone mutation that decreases the enzyme activity but doesnot ablate it. In both the Tf and whole-serum protein gly-cans, the a1,3 antenna of the core structure is processed bythe addition of GlcNAc, galactose, and sialic acid, indicat-ing that the later part of the glycan processing pathway isunaffected. The addition of a bisecting GlcNAc throughthe action of GlcNAcT-III can also block the processingof biantennary complex glycans, but an up-regulation ofthis enzyme is unlikely to be a cause of the hybrid glycansbecause no bisecting GlcNAc residues were detected in Tf(AC� 0%, control� 0%) and similar amounts in IgG(AC� 41%, control� 46%), and only low levels were pre-sent in the serum glycan profile (AC� 9%, control� 25%)(Figures 1, 4, 5 and Tables I, II, III).
The affected mannosidase gene in AC is unlikely to beaMII because there is no evidence of the pathologicalsymptoms characteristic of HEMPAS. Therefore, we pro-posed that a partial loss of mannosidase activity is a resultof a defect in an alternative mannosidase isozyme, probablymannosidase III or mannosidase IIx. However, genetic
Fig. 8. Analysis of Tf isoforms in normal human serum and in the serumfrom patients VT and AC. On a zoomed region of the 3±10 pH gradient2D gel of pooled normal human serum all protein spots relating totransferrin are indicated (A). The relevant gel regions for the controlserum (B1) and the serum of patient AC (B2) are magnified andassembled in a collage, allowing translation of spot positions on the2-DE gel into a 1D IEF pattern (B3). The resulting band pattern is shownfor control and both patient (VT and AC) sera with the thickness andheight of the graphical bar representing protein quantities as suggestedby 2-DE spot position and intensity. The differential expression of themain 77-kDa Tf isoform array as resolved by 2-DE is shown for controland the patients AC and VT in (C).
M. Butler et al.
614
at Pennsylvania State University on February 23, 2013
http://glycob.oxfordjournals.org/D
ownloaded from
Tab
leV
.C
ha
ng
esin
pro
tein
exp
ress
ion
an
dp
hysi
coch
emic
al
pro
per
ties
(MW
,pI
)o
bse
rved
by
2-D
Ep
rote
om
ean
aly
sis
of
seru
mfr
om
pati
ents
VT
an
dA
Cin
com
pari
son
toa
po
ole
dse
rum
of
28
hea
lth
yn
orm
al
ind
ivid
ua
ls
Natu
reo
fch
an
ge
Sp
ot
po
siti
on
Pro
tein
na
me
Acc
essi
on
nu
mb
erV
TA
CV
TA
CC
om
men
t
Co
mp
lem
ent
4g
am
ma
-ch
ain
P0
10
28
Iso
form
mis
sin
gM
issi
ng
two
iso
form
sS
18
[S18]
S18
[S18]
Act
ivati
on
of
com
ple
men
t(c
lass
ical
path
way);
med
iato
ro
flo
cal
infl
am
mato
ryp
roce
sses
no
tgly
cosy
late
d
Co
mp
lem
ent
C3
P0
10
24
Inte
nsi
tyd
ow
nF
GH
17�a
/b19
[FG
H17�a
/b19]
Act
ivati
on
of
com
ple
men
t(c
lass
ical
an
dalt
ern
ate
path
way);
med
iato
ro
flo
cal
infl
am
mato
ryp
roce
sses
;N
-gly
cosy
lati
on
Co
mp
lem
ent
fact
or
BP
00
751
Mis
sin
gH
MW
arr
ay
Q-T
7[Q
-T7]
Alt
ern
ate
com
ple
men
tp
ath
way;
N-g
lyco
syla
tio
n
Co
mp
lem
ent
fact
or
H±
rela
ted
pro
tein
1Q
03
59
1M
issi
ng
Mis
sin
gQ
-T16/1
7[Q
-T16]
Q-T
16/1
7[Q
-T16/1
7]
Invo
lved
inco
mp
lem
ent
regu
lati
on
an
dli
pid
met
ab
oli
sm;
N-g
lyco
syla
tio
n
Alp
ha
-1-a
cid
icg
lyco
pro
tein
P0
27
63
pI
shif
t,M
Wre
du
ced
DE
F15/1
6[B
C14/1
5]
Pro
bab
lyin
vo
lved
inacu
tep
hase
react
ion
;N
-gly
cosy
late
d
Clu
ster
inP
10
909
Sh
ifte
dM
issi
ng
F16
[F16]
F16
[F16]
Fu
nct
ion
un
clea
r;acu
tep
hase
pro
tein
;N
-gly
cosy
late
d,
mo
no
-/b
i-/t
risi
alo
bia
nte
nn
ary
Inte
r-a
lph
atr
yp
sin
inh
ibit
or
hea
vy
cha
in4
Q1
46
24
Mis
sin
gE
8[F
8]
Acu
tep
hase
react
ion
s;in
crea
sed
aft
erd
iffe
ren
tsu
rgic
al
trau
ma;
N-
an
dO
-gly
cosy
lati
on
Inte
r-a
lph
a-t
ryp
sin
inh
ibit
or
hea
vy
cha
inH
4H
MW
Q1
46
24/
Q1
51
35
Mis
sin
gF
8[F
8]
Acu
tep
hase
react
ion
s;in
crea
sed
aft
erd
iffe
ren
tsu
rgic
al
trau
ma;
N-
an
dO
-gly
cosy
lati
on
Inte
r-a
lph
a-t
ryp
sin
inh
ibit
or
hea
vy
cha
inH
4L
MW
Q1
46
24/
Q1
51
35
Mis
sin
gL
19
[L18]
Acu
tep
hase
react
ion
s;in
crea
sed
aft
erd
iffe
ren
tsu
rgic
al
trau
ma;
N-
an
dO
-gly
cosy
lati
on
Ca
rbo
xyp
epti
da
seN
P1
51
69
Mis
sin
gQ
/R14
[Q/R
14]
Pro
tect
sth
eb
od
yfr
om
vaso
act
ive
an
din
flam
mato
ryp
epti
des
(kin
ins,
an
ap
hyla
toxin
s);
the
two
inact
ive
of
the
fou
rsu
bu
nit
sare
N-g
lyco
syla
ted
Proteomics and glycan analysis of CDGs
615
at Pennsylvania State University on February 23, 2013
http://glycob.oxfordjournals.org/D
ownloaded from
Ta
ble
V.
con
tin
ued
Natu
reo
fch
an
ge
Sp
ot
po
siti
on
Pro
tein
na
me
Acc
essi
on
nu
mb
erV
TA
CV
TA
CC
om
men
t
Alp
ha
-1-a
nti
try
psi
nP
01
009
MW
red
uct
ion
by
3k
Da
E/F
14
[F13/1
4]
Pro
tect
slo
wer
resp
irato
rytr
act
again
stp
rote
oly
tic
des
tru
ctio
nb
yh
um
an
leu
ko
cyte
elast
ase
;ass
oci
ate
dw
ith
risk
of
dev
elo
pin
gch
ron
ico
bst
ruct
ive
pu
lmo
nary
dis
ease
;N
-gly
cosy
lati
on
An
tith
rom
bin
III
P0
10
08
Mis
sin
g[I
11]
Reg
ula
tes
blo
od
coagu
lati
on
casc
ad
e;N
-gly
cosy
lati
on
Ap
oli
po
pro
tein
EP
02
649
Mis
sin
gR
18
[R18]
Tra
nsf
eran
dca
tab
oli
smo
fli
po
pro
tein
s;sy
nth
esiz
edw
ith
siali
caci
datt
ach
edb
yO
-gly
cosi
dic
lin
kage
an
dis
sub
seq
uen
tly
des
ialy
late
din
pla
sma
Ap
oli
po
pro
tein
L1
O1
479
1M
issi
ng
Mis
sin
gT
18
[T18]
S17
[T18]
Lip
idex
chan
ge
an
dtr
an
spo
rt;
po
ten
tial
N-g
lyco
syla
tio
n
Tra
nsf
erri
nP
02
787
pI
shif
t,H
MW
arr
ay
Q-T
9[Q
-T9]
Iro
nb
ind
ing
tran
spo
rtp
rote
in;
N-
an
dO
-gly
cosy
lati
on
Tra
nst
hy
reti
nP
02
766
MW
red
uce
dK
18
[K18]
Th
yro
idh
orm
on
e-b
ind
ing
pro
tein
.P
rob
ab
lytr
an
spo
rts
thyro
xin
efr
om
the
blo
od
stre
am
toth
eb
rain
;gly
cosy
lati
on
no
td
escr
ibed
Pla
sma
reti
no
l-b
ind
ing
pro
tein
P0
27
53
Inte
nsi
tyd
ow
nI2
4[I
24]
Del
iver
sre
tin
ol
fro
mli
ver
sto
res
top
erip
her
al
tiss
ue;
def
icie
ncy
cau
ses
nig
ht
vis
ion
pro
ble
ms;
un
kn
ow
ngly
cosy
lati
on
Ha
pto
glo
bin
LM
WP
00
737
Mis
sin
gI/
J25/2
6�2
8[I
/J25/2
6�2
8]
Co
mb
ines
wit
hfr
eep
lasm
ah
emo
glo
bin
pre
ven
tin
glo
sso
fir
on
thro
ugh
the
kid
ney
s;N
-gly
cosy
lati
on
Ha
pto
glo
bin
P0
07
37
Mis
sin
gL
MW
fragm
ents
pI
shif
t;M
Wre
du
ced
HM
Warr
ay
[I�L
28]
H-M
15/1
6[E
-H15]
Co
mb
ines
wit
hfr
eep
lasm
ah
emo
glo
bin
pre
ven
tin
glo
sso
fir
on
thro
ugh
the
kid
ney
s;N
-gly
cosy
lati
on
Ser
um
pa
rao
xo
na
se/a
ryle
ster
ase
1P
27
169
Mis
sin
gG
14
[G14]
Hyd
roly
zes
bro
ad
spec
tru
mo
fo
rgan
op
ho
sph
ate
sub
stra
tes
an
daro
mati
cca
rbo
xyli
caci
des
ters
;N
-gly
cosy
lati
on
M. Butler et al.
616
at Pennsylvania State University on February 23, 2013
http://glycob.oxfordjournals.org/D
ownloaded from
Tab
leV
.co
nti
nu
ed
Natu
reo
fch
an
ge
Sp
ot
po
siti
on
Pro
tein
na
me
Acc
essi
on
nu
mb
erV
TA
CV
TA
CC
om
men
t
Pla
sma
glu
tath
ion
ep
ero
xid
ase
P2
23
52
Mis
sin
gK
/L23
[K23]
Red
uct
ion
of
hyd
rogen
per
oxid
ean
dli
pid
per
oxid
esp
rote
ctin
gce
lls
an
dp
rote
ins
fro
mo
xid
ati
ve
dam
age;
Un
kn
ow
ngly
cosy
lati
on
Ser
um
am
ylo
idP
P0
27
43
Inte
nsi
tyd
ow
nb
oth
iso
form
sJ-
K21
[J-K
21]
Inte
ract
sw
ith
DN
Aan
dh
isto
nes
;m
ay
scaven
ge
nu
clea
rd
ebri
sfr
om
dyin
gce
lls;
N-g
lyco
syla
tio
n
Bet
a-2
-gly
cop
rote
inI
P0
27
49
Mis
sin
gb
asi
cis
ofo
rms
S12
[S12]
Bin
ds
vari
ou
sn
egati
vel
ych
arg
edco
mp
ou
nd
s(e
.g.,
hep
ari
n,
ph
osp
ho
lip
ids)
;N
-an
dO
-gly
cosy
lati
on
Fet
uin
AP
02
765
Mis
sin
gso
me
iso
form
sF
12
[E12]
Pro
mo
tes
end
ocy
tosi
s;o
pso
nic
pro
per
ties
;in
flu
ence
sm
iner
al
ph
ase
of
bo
ne;
N-
an
dO
-gly
cosy
lati
on
Tet
ran
ecti
nP
05
452
Sh
ifte
d(0
.1to
aci
dic
)I2
4[I
/J24/2
5]
C-t
yp
ele
ctin
;b
ind
sp
lasm
ino
gen
;m
ay
be
invo
lved
inp
ack
ing
of
mo
lecu
les
tob
eex
ocy
tose
d;
O-g
lyco
syla
ted
Leu
cin
e-ri
cha
lph
a-2
-gly
cop
rote
inP
02
750
Mis
sin
gE
14
[E13/1
4]
Un
kn
ow
nfu
nct
ion
;N
-an
dO
-gly
cosy
lati
on
Un
kn
ow
np
rote
ina
rra
y?
New
/in
ten
sity
up
I12
Un
kn
ow
np
rote
ina
rra
y?
New
/in
ten
sity
up
U-V
695
kD
a,
pI
7.0
Un
kn
ow
np
rote
ina
rra
y?
New
/in
ten
sity
up
F/G
17/1
8
Un
kn
ow
np
rote
ina
rra
y?
New
/in
ten
sity
up
Za
23
Un
kn
ow
np
rote
ina
rra
y?
New
/in
ten
sity
up
T/U
6/7
Entriesunder
spotpositionrelate
tothegridsuperim
posedonto
the2-D
Egelim
ages
fortherespectivepatientin
Figure
7.
Co
ord
inate
sin
bra
cket
sre
late
toth
ere
late
dsp
ot
on
the
2-D
Egel
ob
tain
edfo
rth
eco
ntr
ol.
Co
mm
ents
on
pro
tein
fun
ctio
nan
dgly
cosy
lati
on
wer
ere
trie
ved
fro
mth
eS
wis
s-P
rot
data
base
(avail
ab
leat
ww
w.e
xp
asy
.ch
,re
lease
40.1
3o
f27
Marc
h2002).
Proteomics and glycan analysis of CDGs
617
at Pennsylvania State University on February 23, 2013
http://glycob.oxfordjournals.org/D
ownloaded from
sequence analysis of lymphocytes from AC showed that themannosidase II and mannosidase IIx genes were apparentlynormal. This leads to the conclusion that the defect causingaberrant glycans in patient AC is likely to be in the enzymemannosidase III, the gene for which has not yet beencloned.
A compensatory enzymatic mechanism may explain theseverity of the disorder
Compensatory mechanisms have also been described forother mutations in the glycosylation pathway (De Praeteret al., 2000). Here a defect in glucosidase I leads to the up-regulation of a Golgi a1,2-endomannosidase (class I) thatenables the normal requirement for glucose removal priorto the action of an a1,2-exomannosidase to be bypassed(Lubas and Spiro, 1987; Volker et al., 2002).
However, there is no evidence for a similar compensatorymechanism or bypass pathway for a defect in the GlcNAcT-IIenzyme. We might speculate that the reason for this is thatthe gene is more recent in evolutionary development andconsequently confined to mammalian cells that are able toprocess complex glycans. Significantly, there is only onecopy of the Mgat2 gene, and it has an uninterrupted readingframe and no reported homologous sequences with otherproteins. The consequence is that a single point mutation inthis gene has serious physiological effects that give rise to asevere clinical condition.
According to the serum proteome, the proposed enzymedefect in patient AC appears to directly or indirectly affectthe structure and/or serum concentrations of specificglycoproteins including transthyretin, carboxypeptidase N,complement factor B, fetuin A, b2-glycoprotein 1, comple-ment C3, and serum amyloid P, alongside the commonCDG marker proteins of clusterin, Tf, and apolipoproteinE and L. Although it cannot be ruled out that the absence ofthese proteins is a secondary consequence of the diagnosedsevere liver disease, at the least these data provide a link topathogenesis.
In conclusion, we have demonstrated that glycan analysisof a pool or of specific proteins can allow the assignment ofunexpected glycan structures. This information can lead tothe identification of a defective glycosylation processingstep. In addition, the glycan profile of proteins expressed
in different cells can indicate the cell specificity of thedisorder, whereas the extent of glycosylation changesprovides an insight into the severity of the processingblock. To complement this information, 2-DE of serumprovides an overview of the range of proteins directly orindirectly affected by structural changes in the glycans,potentially providing insights into pathogenesis and diseasetargets.
Materials and methods
Extraction of Tf
Serum (15 ml) was loaded onto a column containinganti-human Tf IgG Sepharose 4B. After washing, the Tfwas eluted with 0.2 M glycine/HCl pH 2.8 and saturatedwith iron using a ferric chloride solution. The sample wasthen concentrated on a dialysis membrane (Hershbergeret al., 1991).
Extraction of IgG
Serum samples were dialyzed against 0.1% trifluroaceticacid (TFA) overnight. IgG was isolated from the serumsamples using a Protein G affinity column. (MabTrap kit,Amersham Biosciences, Uppsala, Sweden). Serum wasapplied at a flow rate of 1 ml/min to the column that hadbeen equilibrated using the binding buffer supplied with thekit. The column was washed until the absorbance at 280 nmwas zero. Elution buffer was added, and IgG was collectedinto tubes prepared with a neutralizing buffer. The isolatedIgG was dialyzed against 0.1% TFA overnight, lyophilizedfor 24 h, and then dissolved in double-distilled water to aconcentration of 1 mg/ml.
In-gel release of Tf glycans
Tf samples (5±10 mg) were applied to individual gel lanes forseparation by SDS±PAGE with precast gels (NuPage 10%with 4-morpholine propane sulfonic acid, pH 7, as runningbuffer). Coomassie blue staining resulted in a distinct bandat approximately 80 kDa, as determined from molecularweight markers. The protein bands from three identicallanes were removed from the stained gel by scalpel andtreated with PNGase F to remove the attached glycans(Rudd et al., 2001).
In-gel release of IgG glycans
IgG samples (70 mg) were reduced using 0.5 M dithiothreitol(DTT), alkylated using 100 mM iodoacetamide, andthen separated by SDS±PAGE (freshly made 12.5% gel).Coomassie blue staining showed distinct bands at approxi-mately 50 kDa and 28 kDa, corresponding to IgG heavyand light chains, respectively. The gel bands were cut outand washed alternatively in sodium bicarbonate buffer(20 mM, pH 7) and acetonitrile. The glycans were releasedfrom the protein in the gel bands by PNGaseF digestionovernight at 37�C. The glycans were extracted from the gelpieces with water and sonication.
Fig. 9. A network of biosynthetic pathways deploying alternativeisozymes of mannosidase.
M. Butler et al.
618
at Pennsylvania State University on February 23, 2013
http://glycob.oxfordjournals.org/D
ownloaded from
Glycan removal from serum proteins
Hydrazinolysis. Each dialyzed serum sample (50 ml) wassubjected to hydrazinolysis to remove the N-glycans. Afterextensive lyophilization, the samples were sealed in glasstubes with 0.1 ml anhydrous hydrazine under an atmo-sphere of argon. The tubes were heated to 85�C at a rate of10�C per h and held at this temperature for 12 h. The tubeswere then cooled, and the hydrazine was allowed to evapo-rate. Toluene (250 ml) was added and evaporated (�5).
Reacetylation. Each sample was then dissolved in 200 ml0.2 M sodium acetate pH 8 and reacetylated for 60 min atroom temperature by the addition of 50 ml acetic anhydride.The sample was then desalted by passage through a 0.5 mlDowex AG50 X12 column and elution with 4 � 0.5 mlwater. The sample was dried and redissolved in 100 mldistilled water.
Glycan purification. The reacetylated glycans werepurified by applying the aqueous solution to 1-ml columnsof processed glass beads (Oxford GlycoSciences, Oxford,U.K.) held in BioRad (Hercules, CA) disposable columns.The applied samples were washed successively with 5� 5 mldouble-distilled water, 2 � 5 ml ethanol, 6 � 5 ml butanol,and 8:2:1 (v/v/v) butanol/ethanol/water. The glycans werefinally eluted into 4 � 0.5 ml water. The samples were driedand redissolved in 100 ml double-distilled water in prepara-tion for labeling with 2-AB.
Fluorescent labeling with 2-AB
The glycans were fluorescently labeled with 2-AB by reduc-tive amination according to the method described by Biggeet al. (1995) using the Oxford GlycoSciences Signal labelingkit (OGS, Abingdon, Oxon, U.K.).
Glycan analysis
HPLC analysis was by normal phase (Glycosep N). Struc-tural assignment of the peaks arising from this separationwas made following subsequent treatment with multienzymearrays of exoglycosidases (Guile et al., 1996). Confirmationof structures was obtained by MS (MALDI and ESI).
HPLC
NP HPLC was performed according to the low salt buffersystem as previously described (Guile et al., 1996) using a4.6 � 250 mm Glycosep-N column (OGS). The system wascalibrated using an external standard of hydrolyzed and2-AB-labeled glucose oligomers to create a dextran ladder.Weak anion exchange (WAX) HPLC (Guile et al., 1994)was performed using a Vydac 301VHP575 7.5 � 50 mmcolumn (Anachem, Bedfordshire, U.K.) according to themodified methodology (Zamze et al., 1998).
MALDI MS
Positive-ion MALDI TOF mass spectra were recorded witha Micromass TofSpec 2E reflectron-TOF mass spectro-meter (Micromass, Manchester, U.K.) fitted with delayedextraction and a nitrogen laser (337 nm). The accelerationvoltage was 20 kV; the pulse voltage was 3200 V; and the
delay for the delayed extraction ion source was 500 ns.Samples were prepared by adding 0.5 ml of an aqueoussolution of the sample to the matrix solution (0.3 ml of asaturated solution of 2,5-dihydroxybenzoic acid [DHB] inacetonitrile) on the stainless steel target plate and allowing itto dry at room temperature. The sample/matrix mixture wasthen recrystallized from ethanol (Harvey, 1993). All mea-sured monoisotopic masses were within 0.2 Da from thecalculated values.
MALDI MS/MS data
MALDI MS/MS data were recorded with a MicromassQ-TOF mass spectrometer fitted with an experimentalMALDI ion source (Harvey et al., 2000). The sample, in1 ml water, was mixed with 2,5-DHB (0.9 ml of a saturatedsolution in acetonitrile) on the MALDI probe and allowedto dry. The probe was introduced into the ion source of themass spectrometer via a vacuum lock, and the laser wasfired at 10 Hz. Signals were accumulated for 5 s for eachspectrum, and positive ion spectra were accumulated until asatisfactory signal:noise ratio was obtained. For MS/MS(collision-induced dissociation) spectra, the precursor ion([M�Na]�) was selected with a mass window of about 3 Da,argon was used as the collision gas, and the collision energywas set to record fragments across the entire mass range.
HPLC LC-ESI-MS
ESI±liquid chromatography/MS (LC/MS) data wereobtained with a Waters CapLC HPLC system interfacedwith a Micromass Q-TOF mass spectrometer fitted with aZ-spray electrospray ion source and operated in positive ionmode. A 1 � 150 mm microbore NP HPLC column waspacked with stationary phase material from a Glycosep Ncolumn (Oxford GlycoSciences). The operating conditionsfor the mass spectrometer were: source temperature 100�C;desolvation temperature 120�C; desolvation gas flow 200L/h; capillary voltage 3000 V; cone voltage 30 V; TOFsurvey scan time 1 s, mass range 50±2300; TOF MS/MSscan time 1 s, survey scan 950±1600 with detection massrange 50±3500; and mass selection resolution about 3 Da.The MS to MS/MS automatic switching was initiated whenthe nominated peak intensity rose above 4. Switching backto MS mode occurred after 30 s or when the peak intensityfell below 1. The mass of this peak was then ignored by theautomatic switching routine for 30 s. An automatic collisionenergy (CE) profile was used with CEs of 14±32 V. Resultsfrom the different CEs were combined for data evaluation.
2-DE
Two hundred micrograms of the respective human serumwere dissolved in 375 ml of sample buffer (5 M urea, 2 Mthiourea, 0.002 M tributyl-phosphine, 0.0065 M DTT,0.065 M 3-[(3-cholamidopropyl)dimethylammonio]-1-pro-panesulfonate, 0.15 M NDSB-256) and were subjected to2-DE. Ampholytes were added to the sample at 0.9%Servalyte 3±10, 0.45% Servalyte 2±4 and 9±11. ImmobilizedpH gradient gels (Immobiline DryStrip 3±10 NL) wererehydrated in the sample and IEF was carried out at70 kVh at 17�C according to the method described by
Proteomics and glycan analysis of CDGs
619
at Pennsylvania State University on February 23, 2013
http://glycob.oxfordjournals.org/D
ownloaded from
Sanchez et al. (1997). Following focusing, the immobilizedpH gradient gel strips were immediately equilibrated for 15min in 6 M urea/2% (w/v) SDS, 2% (w/v) DTT, 50 mM Tris,pH 6.8, 30% glycerol. Proteins were separated in the seconddimension at 30 mA per gel and 20�C on 9±16% T, 2.7% CSDS±PAGE gels chemically bound to the rear glass plate inan electrophoresis tank similar to that described by Amessand Tolkovsky (1995). Following electrophoresis, the gelswere fixed in 40% (v/v) ethanol:10% (v/v) acetic acid andstained with the fluorescent dye OGT MP17 according tothe method previously described (Hassner et al., 1984).Sixteen-bit monochrome fluorescence images at 200 mmresolution were obtained by scanning the gels with anApollo II linear fluorescence scanner (OGS).
Image analysis
Scanned images were processed with a custom version ofMELANIE II. The resolved protein features were checked,and gel and staining artifacts were removed manually.From triplicates of 2-DE gel images obtained for eachsample, master images were generated representing featuresthat were present on at least two out of three gels. The 2-DEgel for the normal human serum had previously been cali-brated for molecular weight and isoelectric points andmapped extensively by nanoelectrospray MS/MS followingin-gel protein trypsinolysis (manuscript submitted for pub-lication). This image served as a template for the differentialimage analysis and protein spot identification.
The resolved protein features were checked manually,quantified on the basis of fluorescence signal intensity bysumming pixels within each feature boundary, and recordedas a percentage of the total feature intensity on the image.
Mutation analysis of ManII and ManIIx on AC
RNA was isolated from an EBV-transformed lymphoblast cellline from the patient and the parents with Trizol (Invitrogen,Merelbeke, Belgium) according to the manufacturer's pro-tocol. cDNA was prepared with Superscript II ReverseTranscriptase (Invitrogen) and oligo(dT)12±18 priming.The complete coding region of the mannosidase IIA (Gen-Bank accession number NM_002372) and mannosidase IIx(NM_006112) was sequenced on this cDNA in both direc-tions. The primer sequences are available upon request.
Acknowledgments
We thank R. Bateman and R. Tyldesley of Micromass foraccess to the MALDI-Q-TOF mass spectrometer and theHigher Education Funding Council for England and theBiotechnology and Biological Sciences Research Councilfor providing funds to purchase the other mass spectro-meters. We thank David Chittenden and Chris Brock forrunning and curating the proteome gels. P.M.R. andR.A.D. thank H.C.L. for her constant inspiration.
Abbreviations
2-AB, 2-aminobenzamide; CE, collision energy;CDG, congenital disorder of glycosylation; 2-DE,
two-dimensional gel electrophoresis; DTT, dithiothreitol;DHB, dihydroxybenzoic acid; ER, endoplasmic reticulum;ESI, electrospray ionization; GU, glucose unit; HEMPAS,hereditary erythroblastic multinuclearity with a positiveacidified serum lysis test (human congenital dyserythro-poietic anemia type II); HPLC, high-performance liquidchromatography; IEF, isoelectric focusing; LC/MS, liquidchromatography/mass spectrometry; MALDI, matrix-assisted laser desorption/ionization; MS, mass spectro-metry; MS/MS, tandem mass spectrometry; NP, normalphase; PAGE, polyacrylamide gel electrophoresis; PNGaseF, peptide N-glycosidase F; Q, quadrupole; Tf, transferrin;SDS, sodium dodecyl sulfate; TFA, trifluoroacetic acid;TOF, time-of-flight; WAX, weak anion exchange.
References
Amess, B. and Tolkovsky, A.M. (1995) Programmed cell death insympathetic neurons: a study by two-dimensional polyacrylamide gelelectrophoresis using computer image analysis. Electrophoresis, 16,1255±1267.
Bigge, J.C., Patel, T.P., Bruce, J.A., Goulding, P.N., Charles, S.M., andParekh, R.B. (1995) Nonselective and efficient fluorescent labeling ofglycans using 2-amino benzamide and anthranilic acid. Anal. Biochem.,230, 229±238.
Carchon, H., Van Schaftingen, E., Matthijs, G., and Jaeken, J. (1999)Carbohydrate-deficient glycoprotein syndrome type IA (phosphoman-nomutase-deficiency). Biochim. Biophys. Acta, 1455, 155±165.
Chui, D., Oh-Eda, M., Liao, Y.F., Panneerselvam, K., Lal, A., Marek,K.W., Freeze, H.H., Moremen, K.W., Fukuda, M.N., and Marth, J.D.(1997) Alpha-mannosidase-II deficiency results in dyserythropoiesisand unveils an alternate pathway in oligosaccharide biosynthesis. Cell,90, 157±167.
D'Agostaro, G.A., Zingoni, A., Moritz, R.L., Simpson, R.J.,Schachter, H., and Bendiak, B. (1995) Molecular cloning andexpression of cDNA encoding the rat UDP-N-acetylglucosamine:alpha-6-D-mannoside beta-1,2-N-acetylglucosaminyltransferase II.J. Biol. Chem., 270, 15211±15221.
de Lonlay, P., Seta, N., Barrot, S., Chabrol, B., Drouin, V., Gabriel, B.M.,Journel, H., Kretz, M., Laurent, J., Le Merrer, M., and others. (2001)A broad spectrum of clinical presentations in congenital disorders ofglycosylation I: a series of 26 cases. J. Med. Genet., 38, 14±19.
De Praeter, C M., Gerwig, G.J., Bause, E., Nuytinck, L.K.,Vliegenthart, J.F., Breuer, W., Kamerling, J.P., Espeel, M.F.,Martin, J.J., De Paepe, A.M., and others. (2000) A novel disordercaused by defective biosynthesis of N-linked oligosaccharides due toglucosidase I deficiency. Am. J. Hum. Genet., 66, 1744±1756.
Domon, B. and Costello, C.E. (1988) A systematic nomenclature forcarbohydrate fragmentations in FAB-MS/MS spectra of glycoconju-gates. Glycoconj. J., 5, 397±409.
Dupre, T., Barnier, A., De Lonlay, P., Cormier Daire, V., Durand, G.,Codogno, P., and Seta, N. (2000) Defect in N-glycosylation of proteinsis tissue-dependent in congenital disorders of glycosylation Ia.Glycobiology, 10, 1277±1281.
Freeze, H. H. (2001) Update and perspectives on congenital disorders ofglycosylation. Glycobiology, 11, 129R±143R.
Fukuda, M.N. and Akama, T.O. (2002) In vivo role of alpha-mannosidaseIIx: ineffective spermatogenesis resulting from targeted disruption ofthe Man2a2 in the mouse. Biochim. Biophys. Acta, 1573, 382±387.
Ghirlando, R., Lund, J., Goodall, M., and Jefferis, R. (1999) Glycosyla-tion of human IgG-Fc: influences on structure revealed by differentialscanning micro-calorimetry. Immunol. Lett., 68, 47±52.
Guile, G.R., Wong, S.Y., and Dwek, R.A. (1994) Analytical andpreparative separation of anionic oligosaccharides by weak anion-exchange high-performance liquid chromatography on an inertpolymer column. Anal. Biochem., 222(1), 231±235.
Guile, G.R., Rudd, P.M., Wing, D.R., Prime, S.B., and Dwek, R.A.(1996) A rapid high-resolution high-performance liquid
M. Butler et al.
620
at Pennsylvania State University on February 23, 2013
http://glycob.oxfordjournals.org/D
ownloaded from
chromatographic method for separating glycan mixtures and analyzingoligosaccharide profiles. Anal. Biochem., 240, 210±226.
Harvey, D.J. (1993) Quantitative aspects of the matrix-assisted laserdesorption mass spectrometry of complex oligosaccharides. RapidCommun. Mass Spectrom., 7(7), 614±619.
Harvey, D.J., Bateman, R.H., Bordoli, R.S., and Tyldesley, R. (2000)Ionisation and fragmentation of complex glycans with a quadrupoletime-of-flight mass spectrometer fitted with a matrix-assisted laserdesorption/ionisation ion source. Rapid Commun. Mass Spectrom.,14(22), 2135±2142.
Hassner, A., Birnbaum, D., and Loew, L.M. (1984) Charge-shift probes ofmembrane potential: synthesis. J. Org. Chem., 49, 2546±2551.
Henry, H., Tissot, J.D., Messerli, B., Markert, M., Muntau, A.,Skladal, D., Sperl, W., Jaeken, J., Weidinger, S., Heyne, K., andBachmann, C. (1997) Microheterogeneity of serum glycoproteins andtheir liver precursors in patients with carbohydrate-deficient glyco-protein syndrome type I: apparent deficiencies in clusterin and serumamyloid P. J. Lab. Clin. Med., 129, 412±421.
Henry, H., Froehlich, F., Perret, R., Tissot, J.D., Eilers-Messerli, B.,Lavanchy, D., Dionisi-Vici, C., Gonvers, J.J., and Bachmann, C.(1999) Microheterogeneity of serum glycoproteins in patients withchronic alcohol abuse compared with carbohydrate-deficient glyco-protein syndrome type I. Clin. Chem., 45, 1408±1413.
Hershberger, C.L., Larson, J.L., Arnold, B., Rosteck, P.R. Jr.,Williams, P., DeHoff, B., Dunn, P., O'Neal, K.L., Riemen, M.W.,Tice, P.A., and others. (1991) A cloned gene for human transferrin.Ann. NY Acad. Sci., 646, 140±154.
Hughes, R.C. and Feeney, J. (1986) Ricin-resistant mutants of baby-hamster-kidney cells deficient in alpha- mannosidase-II-catalyzedprocessing of asparagine-linked oligosaccharides. Eur. J. Biochem.,158, 227±237.
Imbach, T., Burda, P., Kuhnert, P., Wevers, R.A., Aebi, M., Berger, E.G.,and Hennet, T. (1999) A mutation in the human ortholog of theSaccharomyces cerevisiae ALG6 gene causes carbohydrate-deficientglycoprotein syndrome type-Ic. Proc. Natl Acad. Sci. USA, 96,6982±6987.
Ioffe, E. and Stanley, P. (1994) Mice lacking N-acetylglucosaminyltrans-ferase I activity die at mid- gestation, revealing an essential role forcomplex or hybrid N-linked carbohydrates. Proc. Natl Acad. Sci. USA,91, 728±732.
Jaeken, J. and Matthijs, G. (2001) Congenital disorders of glycosylation.Annu. Rev. Genomics Hum. Genet., 2, 129±151.
Jaeken, J., Vanderschueren-Lodeweyckx, M., Casaer, P., Snoeck, L.,Corbeel, L., Eggermont, E., and Eeckels, R. (1980) Familialpsychomotor retardation with markedly fluctuating serum proteins,FSH and GH levels, partial TBG-deficiency, increased serumarylsulphatase A and increased CSF protein: a new syndrome?Pediatr. Res., 14, 179.
Jaeken, J., Schachter, H., Carchon, H., De Cock, P., Coddeville, B., andSpik, G. (1994) Carbohydrate deficient glycoprotein syndrome type II:a deficiency in Golgi localised N-acetyl-glucosaminyltransferase II.Arch. Dis. Child., 71, 123±127.
Jaeken, J., Spik, G., and Schachter, H. (1996) Carbohydrate-deficientglycoprotein syndrome. Type II: An autosomal recessive disease dueto mutations in the N-acetylglucosaminyltransferase II gene. InJ. Montreuil, J.F.G. Vliegenthart, and H. Schachter (eds.), Glyco-proteins and disease. Elsevier Science, Amsterdam, pp. 457±467.
Jaeken, J., Matthijs, G., Barone, R., and Carchon, H. (1997) Carbohy-drate deficient glycoprotein (CDG) syndrome type I. J. Med. Genet.,34, 73±76.
Jaeken, J., Matthijs, G., Saudubray, J. M., Dionisi-Vici, C., Bertini, E.,de Lonlay, P., Henri, H., Carchon, H., Schollen, E., andVan Schaftingen, E. (1998) Phosphomannose isomerase deficiency: acarbohydrate-deficient glycoprotein syndrome with hepatic-intestinalpresentation. Am. J. Hum. Genet., 62, 1535±1539.
Jefferis, R., Lund, J., Mizutani, H., Nakagawa, H., Kawazoe, Y.,Arata, Y., and Takahashi, N. (1990) A comparative study of the N-linked oligosaccharide structures of human IgG subclass proteins.Biochem. J., 268(3), 529±537.
Jefferis, R., Lund, J., and Pound, J.D. (1998) IgG-Fc-mediated effectorfunctions: molecular definition of interaction sites for effector ligandsand the role of glycosylation. Immunol. Rev., 163, 59±76.
Keir, G., Winchester, B.G., and Clayton, P. (1999) Carbohydrate-deficientglycoprotein syndromes: inborn errors of protein glycosylation. Annu.Clin. Biochem., 36, 20±36.
Korner, C., Knauer, R., Stephani, U., Marquardt, T., Lehle, L., and vonFigura, K. (1999) Carbohydrate deficient glycoprotein syndrometype IV: deficiency of dolichyl-P-Man:Man(5)GlcNAc(2)-PP-dolichylmannosyltransferase. EMBO J., 18, 6816±6822.
Leatherbarrow, R.J. and Dwek, R.A. (1984) Binding of complementsubcomponent C1q to mouse IgG1, IgG2a and IgG2b: a novel C1qbinding assay. Mol. Immunol., 21, 321±327.
Leatherbarrow, R.J., Rademacher, T.W., Dwek, R.A., Woof, J.M.,Clark, A., Burton, D.R., Richardson, N., and Feinstein, A. (1985)Effector functions of a monoclonal aglycosylated mouse IgG2a:binding and activation of complement component C1 and interactionwith human monocyte Fc receptor. Mol. Immunol., 22, 407±415.
Lubas, W.A. and Spiro, R.G. (1987) Golgi endo-alpha-D-mannosidasefrom rat liver, a novel N-linked carbohydrate unit processing enzyme.J. Biol. Chem., 262(8), 3775±3781.
Malhotra, R., Wormald, M.R., Rudd, P.M., Fischer, P.B., Dwek, R.A.,and Sim, R.B. (1995) Glycosylation changes of IgG associated withrheumatoid arthritis can activate complement via the mannose-bindingprotein. Nature Med., 1, 237±243.
Mason, A.B., Miller, M.K., Funk, W.D., Banfield, D.K., Savage, K.J.,Oliver, R.W., Green, B.N., MacGillivray, R.T., and Woodworth, R.C.(1993) Expression of glycosylated and nonglycosylated humantransferrin in mammalian cells. Characterization of the recombinantproteins with comparison to three commercially available transferrins.Biochemistry, 32, 5472±5479.
Mills, P., Mills, K., Clayton, P., Johnson, A., Whitehouse, D., andWinchester, B. (2001) Congenital disorders of glycosylation type Ileads to altered processing of N-linked glycans, as well as under-glycosylation. Biochem. J., 359, 249±254.
Misago, M., Lia, Y.F., Kudo, S., Eto, S., Mattei, M.G., Moremen, K.W.,and Fukuda, M.N. (1995) Molecular cloning and expression ofcDNAs encoding human a-mannosidase II and a novel a-mannosidaseIIx isozyme. Proc. Natl Acad. Sci. USA, 92, 11766±11770.
Moremen, K.W. (2002) Golgi alpha-mannosidase II deficiency invertebrate systems: implications for asparagine-linked oligosaccharideprocessing in mammals. Biochim. Biophys. Acta., 1573, 225±235.
Moremen, K.W., Trimble, R.B., and Herscovics, A. (1994) Glycosidasesof the asparagine-linked oligosaccharide processing pathway. Glyco-biology, 4, 113±125.
Oh-Eda, M., Nakagawa, H., Akama, T.O., Lowitz, K., Misago, M.,Moremen, K.W., and Fukuda, M.N. (2001) Overexpression of theGolgi-localized enzyme alpha-mannosidase IIx in Chinese hamsterovary cells results in the conversion of hexamannosyl-N- acetylchito-biose to tetramannosyl-N-acetylchitobiose in the N-glycan-processingpathway. Eur. J. Biochem., 268, 1280±1288.
Parekh, R.B., Dwek, R.A., Sutton, B.J., Fernandes, D.L., Leung, A.,Stanworth, D., Rademacher, T.W., Mizuochi, T., Taniguchi, K.,Matsuta, K., and others. (1985) Association of rheumatoid arthritisand primary osteoarthritis with changes in the glycosylation pattern oftotal serum IgG. Nature, 316, 452±457.
Quintin, E., Gladen, A., Roden, L., and Kresse, H. (1990) A genetic defectin the biosynthesis of dermatan sulfate proteoglycan: galactosyltrans-ferase 1 deficiency in fibroblasts from a patient with progeroidsyndrome. Proc. Natl Acad. Sci. USA, 67, 1342±1346.
Rudd, P.M., Colominas, C., Royle, L., Murphy, N., Hart, E., Merry, A.H.,Hebestreit, H.F., and Dwek, R.A. (2001) A high-performance liquidchromatography based strategy for rapid, sensitive sequencing of N-linked oligosaccharide modifications to proteins in sodium dodecylsulphate polyacrylamide electrophoresis gel bands. Proteomics, 1,285±94.
Sanchez, J.C., Rouge, V., Pisteur, M., Ravier, F., Tonella,L., Moosmayer, M., Wilkins, M.R., and Hochstrasser, D.F. (1997)Improved and simplified in-gel sample application using reswelling ofdry immobilized pH gradients. Electrophoresis, 18, 324±327.
Schachter, H. (2001) Congenital disorders involving defective N-glycosy-lation of proteins. Cell Mol. Life Sci., 58, 1085±1104.
Schachter, H. and Jaeken, J. (1999) Carbohydrate-deficient glycoproteinsyndrome type II. Biochim. Biophys. Acta, 1455, 179±192.
Proteomics and glycan analysis of CDGs
621
at Pennsylvania State University on February 23, 2013
http://glycob.oxfordjournals.org/D
ownloaded from
Schachter, H., Tan, J., Sarkar, M., Yip, B., Chen, S., Dunn, J., andJaeken, J. (1998) Defective glycosyltransferases are not good for yourhealth. Adv. Exp. Med. Biol., 435, 9±27.
Spik, G., Coddeville, B., and Montreuil, J. (1988) Comparative study ofthe primary structures of sero-, lacto- and ovotransferrin glycans fromdifferent species. Biochimie, 70, 1459±1469.
Tan, J., D'Agostaro, G.A.F., Bendiak, B., Reck, F., Sarkar, M.,Squire, J.A., Leong, P., and Schachter, H. (1995) The human UDP-N-acetylglucosamine:alpha-6-D-mannoside-beta-1,2-N-acetylglucosa-minyltransferase II gene (Mgat2): cloning of genomic DNA, localiza-tion to chromosome 14q21, expression in insect cells and purificationof the recombinant protein. Eur. J. Biochem., 231, 317±328.
Tan, J., Dunn, J., Jaeken, J., and Schachter, H. (1996) Mutations in theMGAT2 gene controlling complex N-glycan synthesis cause carbohy-drate-deficient glycoprotein syndrome type II, an autosomal recessivedisease with defective brain development. Am. J. Hum. Genet., 59,810±817.
Volker, C., De Praeter, C.M., Hardt, B., Breuer, W., Kalz-Fuller, B., VanCoster, R.N., and Bause, E. (2002) Processing of N-linked carbohy-drate chains in a patient with glucosidase I deficiency (CDG type IIb).Glycobiology, 12, 473±483.
Wang, Y., Tan, J., Campbell, R.M., Ditto, D., Le, D., Schachter, H., andMarth, J.D. (2000) Investigating a model of human CDG type IIreveals a heterogenic disease with cell-specific defects and frequentperinatal lethality. Glycobiology, 10, 1131±1132.
Wang, Y., Tan, J., Sutton-Smith, M., Ditto, D., Panico, M.,Campbell, R.M., Varki, N.M., Long, J.M., Jaeken, J., Levinson,S.R., and others. (2001) Modeling human congenital disorder ofglycosylation type IIa in the mouse: conservation of asparagine-linked
glycan-dependent functions in mammalian physiology and insightsinto disease pathogenesis. Glycobiology, 11, 1051±1070.
Watson, M., Rudd, P.M., Bland, M., Dwek, R.A., and Axford, J.S. (1999)Sugar printing rheumatic diseases: a potential method for diseasedifferentiation using immunoglobulin G oligosaccharides. ArthritisRheum., 42, 1682±1690.
Wormald, M.R., Rudd, P.M., Harvey, D.J., Chang, S.C., Scragg, I.G.,and Dwek, R.A. (1997) Variations in oligosaccharide-protein inter-actions in immunoglobulin G determine the site-specific glycosylationprofiles and modulate the dynamic motion of the Fc oligosaccharides.Biochemistry, 36, 1370±1380.
Yamashita, K., Ideo, H., Ohkura, T., Fukushima, K., Yuasa, I., Ohno, K.,and Takeshita, K. (1993) Sugar chains of serum Tf from patients withcarbohydrate deficient glycoprotein syndrome. Evidence of aspara-gine-N-linked oligosaccharide transfer deficiency. J. Biol. Chem., 268,5783±5789.
Yarema, K.J. and Bertozzi, C.R. (2001) Characterizing glycosylationpathways. Genome Biol., 2(5), REVIEWS0004.
Youings, A., Chang, S.C., Dwek, R.A., and Scragg, I.G. (1996)Site-specific glycosylation of human immunoglobulin G is alteredin four rheumatoid arthritis patients. Biochem. J., 314, 621±630.
Yuasa, I., Ohno, K., Hashimoto, K., Iijima, K., Yamashita, K., andTakeshita, K. (1995) Carbohydrate-deficient glycoprotein syndrome:electrophoretic study of multiple serum glycoproteins. Brain Dev., 17,13±19.
Zamze, S., Harvey, D.J., Chen, Y.J., Guile, G.R., Dwek, R.A., andWing, D.R. (1998) Sialylated N-glycans in adult rat brain tissueÐawidespread distribution of disialylated antennae in complex and hybridstructures. Eur. J. Biochem., 258(1), 243±270.
M. Butler et al.
622
at Pennsylvania State University on February 23, 2013
http://glycob.oxfordjournals.org/D
ownloaded from