Skip Navigation


Glycobiology Advance Access originally published online on November 21, 2006
Glycobiology 2007 17(3):304-312; doi:10.1093/glycob/cwl071
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
17/3/304    most recent
cwl071v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Duk, M.
Right arrow Articles by Lisowska, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Duk, M.
Right arrow Articles by Lisowska, E.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2006. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org

Structures of unique globoside elongation products present in erythrocytes with a rare NOR phenotype

Maria Duk2, Suddham Singh3, Vernon N. Reinhold3, Hubert Krotkiewski2, Ewa Kurowska4 and Elwira Lisowska1,2

2 Department of Immunochemistry, Ludwik Hirszfeld Institute of Immunology and Experimental Therapy, Polish Academy of Sciences, 53-114 Wroclaw, Poland
3 Department of Chemistry, University of New Hampshire, Durham, NH 03824, USA
4 Institute of Hematology and Blood Transfusion, 00-957 Warsaw, Poland

1 To whom correspondence should be addressed; Tel: +48-71-337 1172; Fax: +48-71-337 1382; e-mail: lisowska{at}iitd.pan.wroc.pl

Received on July 11, 2006; revised on October 24, 2006; accepted on November 16, 2006


   Abstract
Top
 Abstract
Introduction
 Results
 Discussion
 Materials and methods
 Conflict of interest statement
 Acknowledgments
 References
 
Rare polyagglutinable erythrocytes of NOR phenotype were found to contain two unique glycosphingolipids (designated NOR1 and NOR2). These components (not detected in normal erythrocytes) were reactive with Griffonia simplicifolia isolectin IB4 (GSL-IB4) and commonly present human anti-NOR antibodies. The NOR1 component has been reported to be a globoside containing a single galactose residue linked {alpha}1,4 to the terminal N-acetylgalactosamine. Here, we report the structural studies on a second glycolipid, NOR2, and a third novel component migrating in high-performance thin-layer chromatography (HPTLC) between NOR1 and NOR2. The structures were determined by a combination of ion trap sequential mass spectrometry (MALDI-QIT-TOF) and step-wise treatment with glycosidases, followed by identification of products on HPTLC plates with lectins and mouse monoclonal anti-NOR antibody. The NOR2 component was found to be a disaccharide extension of NOR1 with the following structure: Gal{alpha}1-4GalNAcß1-3Gal{alpha}1-4GalNAcß1-3Gal{alpha}1-4Galß1-4Glcß1-Cer. Treatment of NOR2 with {alpha}-galactosidase gave a glycolipid migrating between NOR1 and NOR2, which did not react with either GSL-IB4 or anti-NOR antibodies but did react with GalNAc-specific soybean agglutinin. This intermediate glycolipid (now designated NORint) was identified as a relatively abundant component of a neutral glycolipid fraction from NOR erythrocytes, suggesting its presence as a precursor to NOR2. The structure of NORint was also confirmed by sequential mass spectrometry studies. These results indicate that polyagglutination in NOR subjects is due to unique erythrocyte glycolipids that are synthesized by sequential addition of Gal{alpha}1,4 and GalNAcß1,3 to globoside.

Key words: antibodies / glycosphingolipids / lectins / sequential mass spectrometry / polyagglutination NOR


   Introduction
Top
 Abstract
 Introduction
Results
 Discussion
 Materials and methods
 Conflict of interest statement
 Acknowledgments
 References
 
Erythrocytes of a rare inheritable NOR phenotype were described for the first time by Harris et al. (1982Go). Agglutination of the erythrocytes by normal blood group-compatible human sera was strongly enhanced by pretreatment of the cells with proteases and inhibited by avian and sheep hydatic cyst fluid P1 glycoprotein. A second case of NOR polyagglutination was identified in a Polish family, and the erythrocytes were found to contain two major unique neutral glycosphingolipids, designated NOR1 and NOR2. These components were detected with Griffonia simplicifolia IB4 lectin (GSL-IB4) on high-performance thin-layer chromatography plates (HPTLC) and not observed in control erythrocytes (Kusnierz-Alejska et al., 1999Go). Further evidence for the involvement of NOR1 and NOR2 glycolipids in polyagglutination has been provided by staining the same glycolipids with anti-NOR antibodies isolated from human sera (Duk et al., 2001Go). Studies on the NOR1 component demonstrated the structure to be an {alpha}-galactosylated globoside, Gal{alpha}1-4GalNAcß1-3Gal{alpha}1-4Galß1-4Glcß1-Cer (Duk et al., 2001Go). This assigned structure was consistent with the recognition of NOR1 by GSL-IB4, a lectin specific for {alpha}-galactosyl residues (Goldstein and Winter, 1999Go). The structure of NOR1 was further confirmed by synthesis of Gal{alpha}1-4GalNAc and Gal{alpha}1-4GalNAcß1-3Gal (Westerlind et al., 2002Go), and using these oligosaccharides, the specificity of anti-NOR antibodies was determined (Duk, Westerlind, et al., 2003Go; Duk, Kusnierz-Alejska, et al., 2005Go). All natural human anti-NOR antibodies and two recently obtained mouse monoclonal anti-NOR antibodies (nor118, and nor87) reacted most strongly with the trisaccharide that represents the terminal sequence of the NOR1 glycolipid. Human anti-NOR represented two major types of subspecificity, for Gal{alpha}1-4GalNAc/Galß1-3Gal or Gal{alpha}1-4GalNAc, and both mouse monoclonal anti-NOR antibodies were specific for Gal{alpha}1-4GalNAc.

The aim of the present study was to structurally characterize the NOR2 glycolipid and its possible biosynthetic precursor, NORint. The results have shown NOR2 to be a disaccharide extension of the NOR1 glycolipid, with a terminally linked additional Gal{alpha}1-4GalNAcß1-3 unit. Moreover, the possible precursor, GalNAc-NOR1 (designated as NORint), has also been identified in NOR erythrocytes.


   Results
Top
 Abstract
 Introduction
 Results
Discussion
 Materials and methods
 Conflict of interest statement
 Acknowledgments
 References
 
Enzymatic degradation of NOR2 glycolipid
Reactivity of NOR1 and NOR2 glycolipids with anti-NOR antibodies suggested that the terminal sequence of both glycolipids is Gal{alpha}1-4GalNAcß1-3Gal. Therefore, the NOR2 sample was isolated and treated with coffee bean {alpha}-galactosidase, and part of this degalactosylated NOR2 product was treated with jack bean ß-N-acetylhexosaminidase (HexNAc). The three samples of NOR2, ‘sham’ treated, {alpha}-galactosidase treated, and sequentially treated with both enzymes, were fractionated by HPTLC, and the glycolipids were detected on plates with an anti-NOR MAb, nor118, and soybean agglutinin (SBA) (Fig. 1). An untreated total neutral glycolipid fraction from NOR erythrocytes served as a reference sample. Treatment with {alpha}-galactosidase totally transformed NOR2 into glycolipid migrating between NOR1 and NOR2 (designated NORint), detectable with GalNAc-specific SBA and not (Fig. 1) or weakly (Fig. 2) detectable with the anti-NOR antibody. This NORint band was also detected with SBA in the untreated total neutral glycolipid fraction from NOR erythrocytes (Fig. 1). Treatment of degalactosylated NOR2 (NORint) with ß-N-acetylhexosaminidase gave, in turn, the total transformation of NORint band into NOR1, detectable with the MAb, nor118. In conclusion, these results suggested that NOR2 is a NOR1 glycolipid elongated by Gal{alpha}-GalNAcß moiety.


Figure 1
View larger version (28K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1. Characterization of enzymatic degradation products of the isolated NOR2 glycolipid on the HPTLC plates. Left panel: orcinol staining; middle and right panels: detection of glycolipids with the anti-NOR MAb nor118 and SBA, respectively. Lanes A1 and NOR, total neutral glycolipid fractions from control A1 and NOR erythrocytes, respectively; 1, purified NOR2 control sample; 2, NOR2 treated with {alpha}-galactosidase; 3, NOR2, pretreated with {alpha}-galactosidase and then treated with ß-N-acetylhexosaminidase.

 

Figure 2
View larger version (21K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2. The {alpha}-galactosidase treatment of the isolated upper band of NOR2 glycolipid. T, total neutral glycolipid fraction from NOR erythrocytes; 1 and 2, ‘sham’-treated and {alpha}-galactosidase treated NOR2-upper band glycolipid, respectively. The glycolipids were detected on the HPTLC plates with anti-NOR MAb nor118, SBA and anti-Gb4 MAb HJ6.

 
The NOR2 glycolipid is usually detected with anti-NOR antibodies as a doublet of bands: a strong lower band and a weaker upper one (Fig. 1). Both bands must represent glycolipids that terminate with an {alpha}-galactose residue, because both disappeared after {alpha}-galactosidase treatment, and a single band of NORint appeared that failed to support a difference in the ceramide moiety. It was found that {alpha}-galactosidase treatment of the upper band of NOR2 did not give any glycolipid reactive with SBA or anti-NOR MAb, nor118, and the only detectable product was a globoside (Fig. 2). Our results suggested that while a major portion of further processed NOR1 is extended by the addition of a ß-GalNAc residue, a minor portion is modified by the addition of an {alpha}-Gal residue.

A new GalNAc-terminating glycolipid (NORint)
The detection NORint glycolipid (Fig. 1) raised the question whether it is also a unique component of NOR erythrocytes. To find an answer, total neutral glycolipids from NOR (blood group A2) erythrocytes and control A1 and A2 erythrocytes were probed on the HPTLC plate with SBA. As shown in Fig. 3, this lectin detected bands common for NOR and control erythrocytes (blood group A glycolipid and a weak band of globoside), and stained strongly an additional band, migrating between NOR1 and NOR2, but only in the NOR erythrocyte glycolipids. This latter SBA-reactive glycolipid (NORint) did not react with Helix pomatia agglutinin (HPA), which detected only blood group A glycolipids in control and NOR erythrocytes (Fig. 3). Interestingly, SBA detected only the faster migrating blood group A glycolipid, whereas HPA stained strongly two major bands; an explanation of this was outside the scope of our work. Both lectins, SBA and HPA, are specific for GalNAc, but HPA reacts only with {alpha}-linked GalNAc, whereas SBA recognizes both {alpha}- and ß-linked GalNAc residues (Wu et al., 2001Go). In view of this, the oligosaccharide chain of NORint is most likely terminated with ß-linked GalNAc residue. This was confirmed with the finding that isolated NORint, similarly to the degalactosylated NOR2, was digested with ß-N-acetylhexosaminidase, and the product of this reaction was the NOR1 glycolipid (Fig. 4). All these results indicated that NORint, a unique component of NOR erythrocytes, is the NOR1 glycolipid with additional ß-linked GalNAc residue and is identical with the product of digestion of the NOR2 glycolipid with {alpha}-galactosidase.


Figure 3
View larger version (47K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3. Unique SBA-reactive and HPA-nonreactive glycolipid (NORint) in NOR erythrocytes. Total neutral glycolipids from NOR erythrocytes and control A1 and A2 erythrocytes were fractionated by HPTLC and glycolipids were detected with lectins, GSL-IB4 (to indicate positions of NOR1 and NOR2 glycolipids), SBA and HPA.

 

Figure 4
View larger version (10K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4. Treatment of the isolated NORint glycolipid with ß-N-acetylhexosaminidase. T, total neutral glycolipid fraction from NOR erythrocytes; 1 and 2, control and ß-N-acetylhexosaminidase-treated NORint, respectively. The glycolipids were detected with SBA and anti-NOR MAb nor118.

 
Mass spectrometry analysis of NOR2 (MALDI-QIT-TOF)
In support of the enzymatic hydrolysis, the NOR2 sample was methylated and analyzed by MALDI-QIT-TOF mass spectrometry. In the spectrum, a major ion pair m/z 2226.2 (M + Na)+ and m/z 2204.2 (M + H)+ was observed, along with an unidentified ion m/z 2370.4 (Fig. 5a). This latter ion failed to fit any known glycosphingolipid composition and its ion current was inadequate for it to be evaluated further. One additional ion, m/z 1566.7, was judged to be a heptasaccharide fragment formed as a consequence of a neutral loss of the ceramide moiety. MALDI sources, lacking atmosphere pressure cooling, are recognized to be more energetic, which may account for this facile neutral loss. Isolation of the most abundant ion (M + Na)+, resonance activation, and CID provided the MS2 spectrum (Fig. 5b). The major product fragment, m/z 1566.7, was 449 amu larger than that observed for the NOR1 glycan (Duk et al., 2001Go), supporting the enzymatic hydrolysis studies indicating a Gal-GalNAc increment to the NOR1 nonreducing terminus. This was also supported by the fragment m/z 1760.7, a composition loss equal to Gal-GalNAc, consistent with the structure shown in Fig. 5c. To confirm the sequence of the reducing terminal tetrasaccharide, the glycan fragment m/z 1566.7 was isolated and analyzed by MS3 (Fig. 6a). All major fragments could be accounted for as rupture adjacent to HexNAc residues, which was anticipated. A single cross-ring fragment, m/z 778.6, confirmed the linkage information established earlier (Duk et al., 2001Go). The structural details within the terminal tetrasaccharide were further evaluated by MS4 analysis for the m/z 935.4 ion fragment (Fig. 6b). As was characteristic, facile HexNAc glycosidic bond rupture (m/z 486.3) coupled with cross-ring cleavages (m/z 329.1 and 778.6) defined the 4-O-linkages in the penultimate and reducing terminal residues. Isolation of the nonreducing terminal disaccharide fragment, m/z 486, and analysis by MS5 (Fig. 7a) showed these spectra to be identical with the Gal{alpha}1-4GalNAc library standard and the same fragment observed in the NOR1 isolate. A contrast of this Gal{alpha}1-4GalNAc fragment spectrum with those of the Galß1-3GalNAc and Galß1-4GlcNAc isomers (Fig. 7, a, b, and c, respectively) provided strong evidence for the identification of ion m/z 486.3 and some differential specificity of these isomers (Fig. 7, a and c). Such analysis has provided an opportunity to build a fragment library for detailed characterization of monomer stereochemistry and linkage isomers within numerous dimers (Ashline et al., 2005Go; Zhang et al., 2005Go). In this case, the presence of a m/z 329 fragment (marker for 4-linkage) and absence of m/z 384 (marker for 3-linkage) (Fig. 7) support the proposed 4-linkage structure. Equally informative was the MS3 spectrum of the m/z 935.4 ion (Fig. 6b), showing a fragment 14 amu lower than the terminal disaccharide, m/z 486.3. This we attribute to the penultimate, or internal, Gal-GalNAc fragment (see structure in Fig. 6b). This would be the expected disaccharide fragment (m/z 472.3, free hydroxyl on C-4) with a related cross-ring fragment at m/z 315.1, which indicates a 1-4 interresidue linkage. Thus, three cross-ring cleavage fragments, m/z 778.6, 329.1, and 315.1, confirm the internal 1-4 linkage for each Gal-GalNAc disaccharide (m/z 486.3 and 472.3). The terminal disaccharide linkage to the NOR1 moiety could not be confirmed by mass spectrometry. For NOR1 studies, a GalNAc1-3 linkage was established by finding that Gal residue is attached to globoside (Duk et al., 2001Go). The ion fragment, m/z 658.6, (Fig. 5b), detected previously in NOR1, suggests that this sample comprises a C18 sphingosine and C24 monounsaturated fatty acid. Sample size limitations did not allow further studies on this ion.


Figure 5
View larger version (31K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5. Analysis of the NOR2 glycolipid by mass spectrometry (A) Profile analysis (MS1) of the sample prepared by methylation and profiled by positive ion MALDI-QIT-TOF. Three components can be accounted for as the protonated and sodiated molecular ions and a heptasaccharide fragment, m/z 1566.7, induced by laser ionization. The ion at m/z 2370.4 did not fit any known GSL composition and the ion current was inadequate to study by MS2 analysis. (B) Isolation and CID of the principle profiled ion, m/z 2226.2, produced a major fragment, m/z 1566.7 accounted for as a neutral loss of the ceramide and a series of smaller fragments consistent with the linear topology of the carbohydrate moiety shown in (C). Internal fragments are observed at m/z 653.4, and 1102.5 which can be accounted for as secondary processes following loss of the ceramide, e.g., loss of the reducing terminal trisaccharide and terminal disaccharide. No major ions were detected below or above the scan intervals shown. The fragment m/z 658.6 can be considered the ceramide residue consistent with NOR1 samples and having C-18 sphingosine and C24 monounsaturated fatty acid.

 

Figure 6
View larger version (21K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6. MS3 and MS4 analyses of the NOR2 glycolipid fragments. (A) MS3 analysis by MALDI-QIT-TOF of major fragment observed in Fig. 5b, (m/z 1566.7, the heptasaccharide moiety following neutral loss of the ceramide). (B) MS4 analysis of major fragment observed in Fig. 6a, (m/z 935.4, the terminal tetrasaccharide ion). Fragment ion compositions are related to the structure drawn on the right. As expected, smaller fragments, in an effort to dissipate energy show greater fragmentation. Cross-ring cleavage fragments for 4-O-linkages, expected for the terminal and penultimate disaccharide moieties, m/z 329 and 778.5, and even for the internal monomer, m/z 315.1.

 

Figure 7
View larger version (34K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 7. MSn analysis by MALDI-QIT-TOF of Hex-HexNAc-1-ene isobars from: (A) NOR2 Fig. 6b, m/z 486.3, identical with similar ion isolated from NOR1 (Duk et al., 2001Go) and fragment library standard, Gal{alpha}1-4GalNAc-1-ene; (B) Galß1-3GalNAc-1-ene; and (C) Galß1-4GlcNAc-1-ene from standard fragment library. (Ashline et al., 2005Go; Zhang et al., 2005Go).

 
A careful screening of the MS spectra of glycolipids eluted from the NOR2 region of TLC plates (containing the double NOR2 band) did not show the presence of any additional NOR1-elongation product with carbohydrate moiety different from that present in NOR2. Probably, the amount of an additional glycolipid giving a weak upper band of NOR2 is too small to allow its detection by MS.

Mass spectrometry analysis of NORint
The NORint glycolipid, isolated from neutral NOR glycolipid fraction, was methylated and analyzed by MALDI-QIT-TOF mass spectrometry. The most abundant ion of MS1 spectrum (not shown), m/z 2021.3, was 245.2 Da higher in mass than NOR1 and 204.2 Da lower than NOR2, strongly suggesting this component to be structurally related to NOR1 and NOR2. Ion trap isolation, activation, and MS2 analysis provided the spectrum shown in Fig. 8a. The most abundant fragment, m/z 1361.5, indicated a loss of 659.8 Da, which was equal to the ceramide moiety identified in NOR1 and NOR2 samples, indicating that the mass modification was restricted to the glycan structure only. Isolation and MS3 analysis of this glycan, m/z 1361.5, provided the spectrum shown in Fig. 8b. Immediately apparent were the facile cleavages proximal to HexNAc residues that were identified in other NOR samples. The shared ion sequence, m/z 449, 653, 1102, and the absence of m/z 486 for a Hex-HexNAc unit positioned the modification to be a terminal HexNAc residue.


Figure 8
View larger version (38K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 8. MSn analysis of NORint glycolipid by MALDI-QIT-TOF. (A) MS2 analysis of the major ion, at m/z 2021.3, of the MS1 profile (not shown). This precursor was isolated and collisionally activated which provided the product spectrum. Neutral loss of the ceramide moiety following collisional activation provided a major fragment, m/z 1361.5. This latter fragment was isolated and analyzed by MS3 (B) providing ions corresponding to the sequence shown in (C).

 
Methylation analysis of globoside and NORint glycolipid
To establish the linkage of this terminal GalNAc to Gal in NORint by methylation analysis, it was necessary to take into account quantitative results, because globoside already contains one C4-substituted and one C3-substituted Gal residue. Therefore, the methylated alditol acetate derivatives of globoside and NORint (available in larger amounts than NOR2) were compared. If the linkage of the terminal GalNAc to Gal was 1-3, NORint should contain two C3-substituted and one C4-substituted Gal residue. The methylated and hydrolyzed globoside and NORint samples were analyzed by gas-liquid chromatography. The shown hexose regions of the chromatograms (Fig. 9) indicated that both compounds contained only C3- and C4-substituted Gal residues, and the ratio of C3- to C4-substituted Gal residue was increased in NORint, as compared to globoside. Despite that ratios of 2,4,6-Me-Gal to 2,3,6-Me-Gal in both compounds did not match exactly the theoretical values (most likely due to small amounts of samples available for analysis), the obtained results were in favor of 1-3 linkage of terminal GalNAc to Gal in NORint.


Figure 9
View larger version (23K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 9. The ‘hexose’-fragment of the gas-liquid chromatogram showing methylated derivatives of galactose and glucose derived from methylated and hydrolyzed globoside and NORint. Identification of the compounds was based on their MS spectra and elution time.

 
Periodate oxidation of NOR glycolipids
In looking for another micromethod that would corroborate the GalNAc-Gal linkage in the terminal part of NORint and NOR2, we checked the effect of periodate oxidation on the glycolipids. In the sequence GalNAcß1-?Gal{alpha}1-4GalNAcß1-, the disaccharide epitope (Gal-GalNAc-) recognized by monoclonal anti-NOR antibody should be resistant to periodate, only if its Gal residue is substituted at C3. Therefore, periodate oxidation followed by Smith degradation should destroy the epitopes in the NOR1 and NOR2 glycolipids and expose the epitope in the NORint glycolipid because the oxidized/reduced terminal GalNAc residue was removed. To appraise this protocol, the total neutral glycolipids from NOR erythrocytes were fractionated by HPTLC and blotted to the Immobilon P membrane. The membranes were successively overlaid with periodate, sodium borohydride, and sulfuric acid. The glycolipids were then detected with MAb nor118. The results were exactly as expected: the antibody detected the NOR1 and NOR2 glycolipids on the control membrane and stained only one band migrating as NORint on the processed membrane (Fig. 10). These results showed that the subterminal galactose residue in NORint is resistant to exhaustive periodate oxidation, supporting a 1-3 linkage of the terminal GalNAc residue.


Figure 10
View larger version (50K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 10. The effect of periodate oxidation on NOR-related glycolipids. Total neutral glycolipids from NOR erythrocytes were fractionated by HPTLC and blotted to Immobilon P membranes. Lane 1, control blot; lane 2, blot submitted to periodate oxidation/Smith degradation. The glycolipids were detected with anti-NOR MAb nor118.

 

   Discussion
Top
 Abstract
 Introduction
 Results
 Discussion
Materials and methods
 Conflict of interest statement
 Acknowledgments
 References
 
This report provides evidence that NOR erythrocytes contain three unusual glycosphingolipids, which are formed by consecutive attachment of galactose and N-acetylgalactosamine residues to globoside.

NOR1: Gal{alpha}1-4GalNAcß1-3Gal{alpha}1-4Galß1-4Glcß1-Cer

NORint: GalNAcß1-3Gal{alpha}1-4GalNAcß1-3Gal{alpha}1-4Galß1-4Glcß1-Cer

NOR2: Gal{alpha}1-4GalNAcß1-3Gal{alpha}1-4GalNAcß1-3Gal{alpha}1-4Galß1-4Glcß1-Cer

The structure of NOR1 was elucidated previously (Duk et al., 2001Go) and that of NOR2 has now been established by MSn studies to be Hex1-4HexNAc1-?Hex1-4HexNAc1-3Hex1-4Hex1-4Hex1-Cer. The ceramide ion, m/z 658.6, suggested a C18 sphingosine and a C24 monounsaturated fatty acid, as found for NOR1. Inter-residue linkages within the terminal Gal-GalNAc disaccharide (NOR1) and within each of the two Hex-HexNAc residues found in NOR2 were determined by very prominent cross-ring cleavages following MSn disassembly to dimers. Adding comparable energy to smaller oligomers greatly enhances fragmentation (decreased number of oscillators for energy dissipation) providing structural details not observed in larger ions (Ashline et al. 2005Go).

A combination of MSn studies with identification of enzymatic digestion products of NOR2 supplied further data. Defining the terminal hexose residue of NOR2 as {alpha}Gal was based on susceptibility to {alpha}-galactosidase and reactivity of this glycolipid with GSL-IB4 and anti-NOR antibodies (both recognizing oligosaccharide chains with terminal {alpha}Gal residue). The subterminal unit of NOR2 is ßGalNAc, since degalactosylated NOR2 (NORint) reacted with SBA, specific for {alpha}- and ß-linked GalNAc, and not with HPA, specific for {alpha}-linked GalNAc, and was transformed into NOR1 by ß-N-acetylhexosaminidase treatment. NORint exists in NOR erythrocytes as an intermediate compound between NOR1 and NOR2 glycolipids and was found to be absent in control erythrocyte glycolipids.

One detail that could not be established by MSn studies was the linkage position of terminal (NORint) or subterminal (NOR2) GalNAc residue to galactose. This problem could not be solved by NMR studies, because the amount of glycolipids required for this technique was not available. However, the 1-3 linkage was indicated by the results of periodate oxidation and supported by methylation analysis and by the reaction of the NOR2 glycolipid with human anti-NOR type 1 antibodies. These antibodies were shown to be strongly inhibited by the synthetic trisaccharide Gal{alpha}1-4GalNAcß1-3Gal and very weakly by Gal{alpha}1-4GalNAc and Gal{alpha}1-4Gal (the latter sequence terminating the P1 antigen). These antibodies strongly react with NOR1 and NOR2 glycolipids (Duk et al., 2005Go), suggesting the existence of the same terminal trisaccharide epitope in both glycolipids. Moreover, it is most likely that NORint glycolipid is synthesized with the participation of globoside synthase, which specifically transfers the GalNAc residue to C3 of terminal {alpha}Gal residue.

The most abundant glycolipids unique for NOR erythrocytes are NOR1 and NORint, because they are distinctly stained by orcinol in total neutral glycolipid pool fractionated by HPTLC (Fig. 1). The amount of NOR2 is much lower, it was not or hardly detected by orcinol staining, but its identification and structural studies were possible thanks to its strong reaction with GSL-IB4 and anti-NOR antibodies and high sensitivity of mass spectrometry. The NOR2 glycolipid is usually detected with anti-NOR antibodies as a doublet of bands, a strong lower band and a weaker upper one (Fig. 1). Transformation of this minor glycolipid to globoside by {alpha}-galactosidase treatment suggested that it is the globoside elongated with at least two {alpha}Gal residues. However, the amount of this additional glycolipid was too small to allow its detection by MS.

Identification of Polish case of polyagglutination as NOR (originally reported by Harris et al. in 1982Go) was initially based on serological properties of erythrocytes, including a weak inhibition of anti-NOR antibodies by P1 glycoprotein. We have recently found that erythrocytes from the original American NOR-positive blood donor contain the same unique glycolipids as those identified in our NOR case. In both samples, the patterns of glycolipid staining on HPTLC plates with GSL-IB4, anti-NOR MAb nor118, and SBA were identical, including the double NOR2 band detected by the antibody (Duk et al., 2006Go).

To the best of our knowledge, the Gal{alpha}1-4GalNAc sequence has not been reported in human and animal glycoconjugates, except in one of 25 oligosaccharide chains identified in amphibian egg jelly coat mucin (Mourad et al., 2001). This sequence is not common in bacteria either; it was reported only in the repeating units of lipopolysaccharide O-specific chains of two Proteus strains (Perepelov et al., 1999Go; Vinogradov et al., 1989Go). An enzyme transferring an {alpha}-galactose residue to the C4 position of GalNAc is unknown. Such transferase (GalTNOR) would be a key enzyme responsible for biosynthesis of NOR-related glycolipids. It seems to act in concert with normal ß1,3-N-acetylgalactosamine transferase (Gb4 synthase), to form several elongated forms of globoside. The GalTNOR may be the product of mutated gene encoding an existing transferase, or is a new enzyme not expressed or inactive in normal phenotypes. Further studies on the genetic background underlying the NOR phenotype are necessary to solve this problem.


   Materials and methods
Top
 Abstract
 Introduction
 Results
 Discussion
 Materials and methods
Conflict of interest statement
 Acknowledgments
 References
 
Erythrocytes
Erythrocytes of a healthy NOR-positive donor (TS) were serologically typed as A2, P1 (2). The control blood samples were supplied by the Regional Center of Blood Transfusion in Wroclaw.

Lectins and antibodies
Lectins, Griffonia simplicifolia IB4 isolectin (GSL-IB4), soybean agglutinin (SBA), and Helix pomatia agglutinin (HPA), purchased from Sigma, St. Louis, MO, were biotinylated as described previously (Duk et al., 1994Go). Murine monoclonal anti-Gb4 (IgM) antibody HJ6 (Madassery et al., 1991Go) was kindly provided by Dr. M.H. Nahm (Rochester, NY). Monoclonal anti-NOR antibody, nor118 (IgG1), specific for Gal{alpha}1-4GalNAc, was obtained by immunization of mice with the conjugate of Gal{alpha}1-4GalNAcß1-3Gal with human serum albumin, as described elsewhere (Duk et al., 2005Go).

Extraction and purification of glycosphingolipids
Isolation and fractionation of glycolipids were performed using the standard procedures described in detail in previous papers (Kusnierz-Alejska et al., 1999Go; Duk et al., 2001Go). Briefly, glycolipids were extracted from lyophilized membranes of NOR and control erythrocytes with chloroform/methanol; neutral glycolipids were separated from phospholipids and gangliosides, additionally purified in peracetylated form, de-O-acetylated, and desalted. To isolate individual glycolipids, neutral glycolipids from NOR erythrocytes were fractionated on the Kieselgel H type 60 (Merck, Darmstadt, Germany) column. Fractions were analyzed by HPTLC with the use of GSL-IB4 (NOR1 and NOR2) or SBA (NORint) and those containing the required glycolipid were pooled. Further purification of glycolipids from the column fractions was achieved by repeated preparative HPTLC and elution from the respective regions of the plate, as described by Duk et al. 2001Go).

Enzymatic degradation
The glycolipid preparations were treated with coffee bean {alpha}-galactosidase (0.4 U in a total volume of 700 µl, 72 h at 37 ° C) or with jack bean ß-N-acetylhexosaminidase (2 U in a total volume of 150 µl, 65 h at 37 ° C), both from Sigma. The NOR2 glycolipid sample was treated consecutively with both enzymes; after treatment with {alpha}-galactosidase, the glycolipids were isolated from the reaction mixture and a half of the portion was treated with ß-N-acetylhexosaminidase. Other details concerning the conditions of enzymatic treatment and processing the samples before and after treatment were based on the procedure reported by Ostrander et al. 1988Go) and described in detail by Duk et al. 2001Go). The control glycolipid samples were submitted to the same procedures, omitting enzymes.

HPTLC lectin and antibody assays
The glycolipid samples solubilized in chloroform/methanol (2 : 1, v/v) were applied to HPTLC plates (Kieselgel 60, Merck) and developed with chloroform/methanol/water (55 : 45 : 10, v/v/v). The dried plates were immersed in 0.05% polyisobutylmethacrylate (Aldrich, Steinheim, Germany) in hexane for 1 min, dried again, sprayed with TBS (0.05 M Tris buffer, 0,15 M NaCl, pH 7.4), and immersed in 5% human serum albumin (HSA) for 1 h (Magnani et al., 1982Go). The plates were overlaid successively with (i) biotinylated lectin solution (5 µg/ml) in TBS/0.05% Tween 20/1% HSA (TBS-T/HSA) for 1 h, (ii) ExtrAvidin-alkaline phosphatase conjugate (Sigma) diluted with TBS-T/HSA for 1 h, and (iii) a substrate (nitro blue tetrazolium/5-bromo-4-chloro-3-indolyl phosphate, from Sigma) solution. The immunostaining with monoclonal antibodies was done similarly, but instead of biotinylated lectins, the plates were overlaid with the nor118 antibody (1.5 h) or with the HJ6 antibody (overnight in refrigerator) and then with biotinylated goat antibodies against mouse Ig (Dako, Glostrup, Denmark) for 1 h. All incubations, if not stated otherwise, were done at room temperature.

Smith degradation of glycolipids on the blots
The procedure could not be performed directly on the HPTLC plates, because the silica layer was mechanically destroyed by NaBH4 or by acid treatment. Therefore, the glycolipids fractionated on the HPTLC plates were transferred to PVDF membranes (Immobilon P, Millipore, Carrigtwohill, Ireland) using the ‘ironing’ method (Taki and Ishikawa, 1997Go). The dried membranes were immersed for 1–2 seconds in 96% ethanol, washed with water, and overlaid successively with (i) 0.05 M sodium periodate in 0.05 M acetate buffer pH 4.5 for 48 h in refrigerator, (ii) 0.15 M NaBH4 in 0.2 M borate buffer pH 8.0 for 3 h at room temperature, and (iii) 0.025 M H2SO4 for 1 h at 80 ° C. After each incubation, the membranes were washed with water. The control membranes, after treatment with ethanol, were incubated in acetate buffer without periodate. The membranes were then incubated (overnight in refrigerator) in 5% bovine serum albumin in TBS, washed with TBS, and glycolipids were detected with the MAb nor118, using the same procedure as described for HPTLC plates.

Glycolipid methylation
Glycolipid methylation for MSn studies was achieved by dissolving samples in a suspension of NaOH/DMSO prepared by vortex mixing DMSO and powdered sodium hydroxide (Ciucanu and Kerek, 1984Go). After 1 h at room temperature, 50 µg of methyl iodide was added, and the suspension set for 1 h at room temperature with occasional vortexing. The methylated product was back extracted by adding 1 ml of chloroform, and the suspensions were washed four times with 2–3 ml of 30% acetic acid (Reinhold et al., 1996Go). Methylated samples were dried following chloroform extraction and stored at –20 ° C. The samples were re-dissolved to a concentration of 10 µM in a 1 mM solution of sodium acetate in 70 : 30 methanol/water just prior to analysis.

Glycolipid methylation for gas-liquid chromatography was done by a similar procedure (Gunnarsson, 1987Go). Briefly, a dried sample was dissolved in 0.2 ml of dry DMSO by stirring for 30 min. A portion of 12–14 mg of finely powdered NaOH was added, followed by 30 µl of methyl iodide, and the suspension was stirred for 2 h. It was neutralized with 1 M acetic acid and diluted with 0.2 ml water. Finally, permethylated glycolipids were extracted in chloroform, washed with water and evaporated.

MALDI-QIT-TOF mass spectrometry
The methylated samples were analyzed by MALDI-QIT-TOF (Axima-IT, Shimadzu Biotech, Kratos Analytical, Ltd., Manchester, UK). This ion trap MS was configured with MALDI for high throughput analysis and coupled with a TOF analyzer for improved mass resolution. The source was decoupled from the analyzer and transparent to laser irradiation effects that are nearly at a perpendicular angle to the sample stage. The 384 spotting plates were irradiated with an LSI N2 laser (337 nm, ~3 ns pulse width, 300mJ max.) and argon was used in ion trap CID studies. The methylated glycolipids were resuspended in methanol and mixed with 5 µl DHB (2,5-dihydroxybenzoic acid, gentisic acid) matrix and spotted (0.5–1 µl) on the target plate and allowed to dry. Laser power was set to 70 mW, mass range 50–5000 atm, using 10 shots per profile, 100 profiles per sample in reflectron mode.

Gas-liquid chromatography
The sample of methylated glycolipids was hydrolyzed in 4 M trifluoroacetic acid for 4 h at 100 ° C and derivatized into alditol acetates. Partially methylated alditol acetates were analyzed in a Hewlett-Packard 5890 gas chromatograph, equipped with a mass selective detector 5971A and HP-1 capillary column (0.2 mm x 12 m). A temperature gradient of 150–230 ° C (8 ° C/min) was used. Quantification of methylated sugar derivatives was based on the total ion current (TIC). The peaks were identified by mass spectra and elution time.


   Conflict of interest statement
Top
 Abstract
 Introduction
 Results
 Discussion
 Materials and methods
 Conflict of interest statement
Acknowledgments
 References
 
None declared.


   Acknowledgments
Top
 Abstract
 Introduction
 Results
 Discussion
 Materials and methods
 Conflict of interest statement
 Acknowledgments
References
 
We thank Dr. S. Bochenek (Regional Center of Blood Transfusion, Wroclaw) for the samples of normal human erythrocytes, Dr. M.H. Nahm (University of Rochester, Rochester, NY) for the HJ6 monoclonal antibody, and Dr. Rachelle Martin (Shimadzu Biotech, Manchester, UK) for preliminary analysis. This work was supported by grant 3 P05A 109 24 of the State Committee for Scientific Research (KBN, Warsaw) (MD and EL), and in part, by NIH Grant Numbers GM45054 and RR16459, the latter from the BRIN Program of the National Center for Research Resources (VNR).


   Abbreviations
 
CID, collision induced dissociation; Gb4Cer, GalNAcß1-3Gal{alpha}1-4Galß1-4Glcß1-Cer (globoside, P antigen); GSL-IB4, Griffonia simplicifolia isolectin IB4; HexNAc, N-acetylhexosamine; HPA, Helix pomatia agglutinin; HPTLC, high-performance TLC; MALDI, matrix-assisted laser desorption ionization; MSn, mass spectrometry; where n represents the number of times MS was performed, ; QIT, quadrupole ion trap; SBA, soybean agglutinin; TLC, thin-layer chromatography; TOF, time of flight


   References
Top
 Abstract
 Introduction
 Results
 Discussion
 Materials and methods
 Conflict of interest statement
 Acknowledgments
 References
 
Ashline D, Singh S, Hanneman A, Reinhold VN. (2005) Congruent strategies for carbohydrate sequencing. 1. Mining structural details by MS(n). Anal Chem 77:6250–6262.[Medline]

Ciucanu I and Kerek F. (1984) A simple and rapid method for the permethylation of carbohydrates. Carbohydr Res 131:209–217.[CrossRef]

Duk M, Kusnierz-Alejska G, Korczagina EY, Bovin NV, Bochenek S, Lisowska E. (2005) Anti-{alpha}-galactosyl antibodies recognizing epitopes terminating with {alpha}1,4-linked galactose: human natural and mouse monoclonal anti-NOR and anti-P1 antibodies. Glycobiology 15:109–118.[Free Full Text]

Duk M, Lisowska E, Moulds JJ. (2006) Polyagglutinable NOR red blood cells found in an American family and in a Polish family have the same unique glycosphingolipids. Transfusion 46:1264–1265.[CrossRef][ISI][Medline]

Duk M, Lisowska E, Wu JH, Wu AM. (1994) The biotin/avidin-mediated microtiter plate lectin assay with the use of chemically modified glycoprotein ligand. Anal Biochem 221:266–272.[CrossRef][ISI][Medline]

Duk M, Reinhold BB, Reinhold VN, Kusnierz-Alejska G, Lisowska E. (2001) Structure of a neutral glycosphingolipid recognized by human antibodies in polyagglutinable erythrocytes of the rare NOR phenotype. J Biol Chem. 276:40574–40582.[Abstract/Free Full Text]

Duk M, Westerlind U, Norberg T, Pazynina G, Bovin NV, Lisowska E. (2003) Specificity of human anti-NOR antibodies, a distinct species of "natural" anti-{alpha}-galactosyl antibodies. Glycobiology 13:279–284.[Abstract/Free Full Text]

Goldstein IJ and Winter HG. (1999) The Griffonia simplicifolia I-B4 isolectin. A probe for {alpha}–D-galactosyl end groups. Subcell Biochem 32:127–140.[Medline]

Gunnarsson A. (1987) N- and O-alkylation of glycoconjugates and polysaccharides by solid base in dimethyl sulphoxide/alkyl iodide. Glycoconjugate J. 4:239–245.

Harris PA, Roman GK, Moulds JJ, Bird GW.G, Shah NG. (1982) An inherited RBC characteristic, NOR, resulting in erythrocyte polyagglutination. Vox Sang. 42:134–140.[ISI][Medline]

Kusnierz-Alejska G, Duk M, Storry JR, Reid ME, Wiecek B, Seyfried H, Lisowska E. (1999) NOR polyagglutination and Sta glycophorin in one family. Relation of NOR polyagglutination to terminal {alpha}-galactose residues and abnormal glycolipids. Transfusion 39:32–38.[CrossRef][ISI][Medline]

Madassery JV, Gillard B, Marcus DM, Nahm MH. (1991) Subpopulations of B cells in germinal center. III. HJ6, a monoclonal antibody, binds globoside and a subpopulation of germinal center B cells. J Immunol 147:823–829.[Abstract]

Magnani JL, Nilsson B, Brockhaus M, Zopf D, Steplewski Z, Koprowski H, Ginsburg V. (1982) A monoclonal antibody-defined antigen associated with gastrointestinal cancer is a ganglioside containing sialylated lacto-N-fucopentaose II. J Biol Chem. 257:14365–14369.[Abstract/Free Full Text]

Mourad R, Morelle W, Neveu A, Strecker G. (2001) Diversity of O-linked glycosylation patterns between species. Characterization of 25 oligosaccharide chains from oviductal mucins Rana ridibunda. Eur J Biochem. 268:1990–2003.[ISI][Medline]

Ostrander GK, Levery SB, Eaton HL, Salyan MEK, Hakomori S, Holmes EH. (1988) Isolation and characterization of four major neutral glycosphingolipids from the liver of the English sole (Parophrys vetulus). J Biol Chem. 263:18716–18725.[Abstract/Free Full Text]

Perepelov AV, Ujazda E, Senchenkova SN, Shashkov AS, Kaca W, Knirel YA. (1999) Structural and serological studies on the O-antigen of Proteus mirabilis O14, a new polysaccharide containing 2-[(R)-1-carboxyethylamino]ethyl phosphate. Eur J Biochem. 261:347–353.[ISI][Medline]

Reinhold VN, Reinhold BB, Chan S. (1996) Carbohydrate sequence analysis (ES-MS/CID/MS). Methods Enzymol. 271:377–402.[ISI][Medline]

Taki T and Ishikawa D. (1997) TLC blotting: application to microscale analysis of lipids and as a new approach to lipid-protein interaction. Anal Biochem. 251:135–143.[CrossRef][ISI][Medline]

Vinogradov EV, Kaca W, Knirel YA, Rozalski A, Kochetkov NV. (1989) Structural studies on the fucosamine-containing O-specific polysaccharide of Proteus mirabilis O19. Eur J Biochem. 180:95–99.[ISI][Medline]

Westerlind U, Hagback P, Duk M, Norberg T. (2002) Synthesis and inhibitory activity of a di- and a trisaccharide corresponding to an erythrocyte glycolipid responsible for the nor polyagglutination. Carbohydr Res. 337:1517–1522.[CrossRef][ISI][Medline]

Wu AM, Song S, Tsai M, Herp A. (2001) A guide to the carbohydrate specificities of applied lectins – 2. Adv Exp Med Biol. 491:551–585.[Medline]

Zhang H, Singh S, Reinhold VN. (2005) Congruent strategies for carbohydrate sequencing. 2. FragLib: an MS(n) spectral library. Anal Chem 77:6263–6270.[Medline]


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?



This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
17/3/304    most recent
cwl071v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Duk, M.
Right arrow Articles by Lisowska, E.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Duk, M.
Right arrow Articles by Lisowska, E.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?