Skip Navigation


Glycobiology Advance Access originally published online on February 9, 2007
Glycobiology 2007 17(5):504-515; doi:10.1093/glycob/cwm012
This Article
Right arrow Abstract Freely available
Right arrow FREE Full Text (PDF) Freely available
Right arrow All Versions of this Article:
17/5/504    most recent
cwm012v2
cwm012v1
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 arrow Search for citing articles in:
ISI Web of Science (2)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Vukelic, Z.
Right arrow Articles by Zamfir, A. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Vukelic, Z.
Right arrow Articles by Zamfir, A. D.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

© The Author 2007. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oxfordjournals.org

Human gliosarcoma-associated ganglioside composition is complex and distinctive as evidenced by high-performance mass spectrometric determination and structural characterization

Zeljka Vukelic2, Svjetlana Kalanj-Bognar3, Martin Froesch4, Laura Bîndila4, Boris Radic5, Mark Allen6, Jasna Peter-Katalinic4 and Alina D. Zamfir1,7,8

2 Department of Chemistry and Biochemistry, Faculty of Medicine, University of Zagreb, Salata 3, 10000 Zagreb, Croatia
3 Croatian Institute for Brain Research, Faculty of Medicine, University of Zagreb, Salata 12, 10000 Zagreb, Croatia
4 Institute for Medical Physics and Biophysics, University of Münster, Robert-Koch-Street 31, D-48149 Münster, Germany
5 Department of Neurology, Clinical Hospital Dubrava, Avenija Gojka Suska bb, 10000 Zagreb, Croatia
6 Advion BioSciences, Ltd, Rowan House, 26-28 Queens Road, Hethersett Norwich Norfolk, NR9 3DB, UK
7 Department of Chemistry and Biology, University of Arad, Revolutiei Blvd. 1, RO-310139, Arad, Romania
8 Mass Spectrometry Laboratory, National Institute for Research and Development in Electrochemistry and Condensed Matter, Plautius Andronescu Str.1, 300224, Timisoara, Romania

1 To whom correspondence should be addressed; Tel: +40-356-437974; Fax: +40-256-204698; e-mail: zamfir{at}uav.ro

Received on December 5, 2006; revised on January 29, 2007; accepted on January 30, 2007


   Abstract
Top
 Abstract
Introduction
 Results
 Discussion
 Materials and methods
 Conflict of interest statement
 Acknowledgments
 References
 
Gangliosides (GGs), involved in malignant alteration and tumor progression/invasiveness, are considered as tumor biomarkers or therapeutic targets. Here, we describe the first systematic GG composition characterization in human gliosarcoma versus normal brain tissue using our recently developed mass spectrometry (MS) methods, based on nano-electrospray (nano-ESI), Fourier-transform ion cyclotron resonance (FT-ICR), and chip nano-ESI quadrupole time-of-flight (QTOF), complemented by thin-layer chromatographic (TLC) analysis and quantification. Combined MS enabled detection and structural assignment of 73 distinct GG species: many more than reported so far for investigated gliomas. Apart from the 7.4-times lower total GG content, gliosarcoma contained all major brain-associated species, however, in very altered proportions, exhibiting a highly distinctive pattern: GD3 (48.9%) > GD1a/nLD1 > GD2/GT3 > GM3 > GT1b > GM2 > GM1a/GM1b/nLM1 > LM1 > GD1b > GQ1b. MS also revealed abundant O-Ac-GD3; its sequencing provided structural evidence to postulate a novel O-Ac-GD3 isomer O-acetylated at the inner Neu5Ac-residue, previously not structurally confirmed. The high sensitivity and mass accuracy permitted the assignment of unusual minor species: GM4, Hex-HexNAc-nLM1, Gal-GD1, Fuc-GT1, GalNAc-GT1, O-Ac-GM3, di- O-Ac-GD3O-Ac-GD3, and O-Ac-GT3, not previously reported as glioma-associated. The gliosarcoma-expressed GA2 might represent a marker distinguishing astrocytic from oligodendroglial tumors. This is, to our knowledge, so far the most complete GG composition characterization of certain glioma, which demonstrates that our MS-based approach could provide essential structural information relevant to glycosphingolipid role(s) in brain tumor biology, differential diagnosis/prognosis and novel treatment concepts.

Key words: biomarkers and target molecules / ganglioside structures / gliosarcoma ganglioside composition / high-performance mass spectrometry / novel O-acetylated GD3 isomer


   Introduction
Top
 Abstract
 Introduction
Results
 Discussion
 Materials and methods
 Conflict of interest statement
 Acknowledgments
 References
 
Gliosarcoma, defined as a variant of glioblastoma multiforme, according to the World Health Organization classification (WHO, 2000), is a rare primary neoplasm of the central nervous system (CNS), a grade IV tumor. Its characteristic biphasic tissue pattern consists of alternating areas displaying glial and mesenchymal (sarcomatous) differentiation. The origin, monoclonal or biclonal, of its biphasic nature is still a subject of debate. Gliosarcoma accounts for about 2% of all the glioblastomas, usually affecting the adult population in the fourth to the sixth decade of life with males more frequently affected than females (male : female, 1.8 : 1) (Ohagaki et al. 2000). Gliosarcomas are usually located in the cerebral cortex involving the temporal, frontal, parietal, and occipital lobe in decreasing frequency. Invariably, the clinical history of the patient is short and the presenting symptoms depend upon the location of the tumor (Machucka et al. 2004). After drastic surgical resection and chemotherapy the prognosis is very poor. Median survival is usually 11.5 months with less than 10% survival after 2 years following diagnosis (Luterbach et al. 2001Go). One of the factors leading to therapeutic failure in the treatment of human anaplastic gliomas is the extensive infiltration of the tumor cells into the brain tissue, which makes them rather inaccessible to treatment methods (Kelly et al. 1984Go). Nowadays, one strategy investigated for treatment is to target the invading tumor cells by using specific binding ligands (Shukla and Krag 2006Go). Therefore, the critical points are to identify the tumor-specific target molecules and characterize their structures in detail.

As generally accepted, malignant transformation of cells is accompanied by aberrant cell surface—glycocalyx—molecular composition, particularly due to irregularities in glycoconjugate glycosylation pathways. Various glycosyl epitopes constitute tumor-associated antigens (Hakomori 2002Go; Fredman et al. 2003Go). Moreover, highly expressed, some of them promote invasion and metastases, whereas some others suppress tumor progression (Hakomori 2001Go). Among molecules bearing characteristic glycosyl epitopes causing such effects are gangliosides (GGs), sialic acid (SA) containing glycosphingolipids (GSLs) incorporated into outer leaflet of the cell-membrane bilayer and particularly enriched in microdomains.

Through intermolecular interactions, GGs directly participate in cell-to-cell and cell-matrix recognition and adhesion, and also, through glycosynapses, modulate signal transduction pathways (Hakomori 2004Go). GSL-dependent cross-talk between glycosynapses interfacing tumor cells with their host cells has been recognized as a basis to define tumor malignancy (Hakomori and Handa 2002Go). Certain GG species induce cell growth inhibition and cell differentiation or/and apoptosis (Malisan and Testi 2002Go). For instance, GM3 treatment reduces the cell number in primary cultures of high-grade human glioblastoma multiforme, ependymomas, mixed gliomas, astrocytomas, oligodendrogliomas, and gangliogliomas, as well as of the rat 9L cell gliosarcoma cell line, whereas it had little effect on the cell number in cultures of a normal human brain. Moreover, intracranial injection of GM3 in nude mice following implantation of rat 9L CNS tumor cells resulted in significantly longer symptom-free survival times (Noll et al. 2001Go). The oncogenic transformation and its reversion can be explained through the difference in GM3 containing-glycosynapse organization (Mitsuzuka et al. 2005Go).

GGs are considered as potential therapeutic targets for cancer treatment, primarily for production of anticancer polyvalent vaccines (Fernandez et al. 2003Go). Immunotherapy with an anti-GD2 monoclonal antibody has been shown as a promising strategy in the prevention of neuroblastoma relapse in an experimental metastatic model (Raffaghello et al. 2003Go). Several GG-based vaccines, in particular against melanoma, small cell lung carcinoma, and breast carcinoma, are currently under clinical trials (Chapman 2003Go; Krug 2004Go).

In the case of human gliomas, mono- and di-sialylated GGs have been primarily studied and suggested as associated species and/or antigens (Fredman et al. 2003Go). Although several molecules have been suggested as potential candidate glioma-antigens, no definitive glioma-specific antigens have been identified to date (Fine 2004Go).

In the last years, several biophysical methods complementary to immunochemical and immunohistochemical techniques, for the investigation of GG expression and structure in gliomas, have been introduced. GG profiling, their quantification and correlation to histomorphology and grading of human gliomas have been studied (Wagener et al. 1999Go) using a newly developed microbore high-performance liquid chromatography (HPLC). Steiner et al. (2003)Go demonstrated the use of infrared spectroscopy as a potential adjunct to histopathological diagnosis of human brain. The analysis showed alteration of the nature and amount of brain lipids, including GG, in tumor as compared with control tissue. In another recent study, Hedberg et al. (2001)Go determined the GG expression in human glioblastoma by confocal microscopy using antiGG monoclonal antibodies.

Mass spectrometry (MS) represents one of the most precise and sensitive analytical methods. Electrospray ionization (ESI) MS enables not only accurate molecular mass determination but, by employing tandem MS, the molecules may be sequenced to deduce their structure in detail. Potentials of ESI-MS for pico- and sub-picomolar level analyses and discovery of biologically relevant species in complex glycomixtures increased after the introduction of chip ionization methods. The development of microfluidics/ESI-MS is aiding significant progress due to numerous advantages of microchip technology such as sensitivity, high performance resulting in MS data rich in structural information. The potential of chip ESI-MS and tandem MS in glycolipidomics for biomedical research has been reported by our group (Zamfir et al. 2004Go, 2005Go). On the other hand, Fourier-transform ion cyclotron resonance MS (FT-ICR-MS) is currently emerging as a powerful tool in glycobiology due to its high resolving power and accurate mass determination (Froesch et al. 2004Go; Zaia 2004Go; Vukelic, Zamfir et al. 2005Go).

The aim of this study is the systematic characterization of GG composition in human gliosarcoma. Modern MS was complemented by high-performance thin layer chromatography (HPTLC) analysis to achieve the final goal-structural analysis of the tumor-associated structures, which might serve as potential diagnostic markers or specific target molecules for production of antitumor therapeutic agents. Chip–based nano-ESI quadrupole time-of-flight (QTOF) MS and ultrahigh resolution nano-ESI-FT-ICR-MS, previously optimized and introduced by us in glycolipidomics, are here applied for the first time for structural identification of GG species in human gliosarcoma.


   Results
Top
 Abstract
 Introduction
 Results
Discussion
 Materials and methods
 Conflict of interest statement
 Acknowledgments
 References
 
Ganglioside quantity and pattern of human gliosarcoma analyzed by HPTLC and laser densitometry
GG were extracted and purified from gliosarcoma tumor tissue sample weighing 0.87 g and in parallel from normal frontal cortex (0.93 g). The total GG content in gliosarcoma sample was 117.85 µg of GG-bound SAs per gram tissue wet weight (w.w.) (µg GG-SA/g), as determined by spectrophotometric method. The content was obviously much lower as compared with normal brain tissue, e.g. approximately 7.4 times less than in the analyzed normal adult cerebral cortex of the frontal lobe (Table I). Qualitative analysis of gliosarcoma GG pattern using HPTLC showed the fractions migrating as GM3, GM2, GM1, GD3, GD1a, GD2/GT3, GD1b, GT1b, and GQ1b (Figure 1A). Additional minor fractions (X1 and X2, marked by arrows) with TLC migrating properties of monosialo- and disialo-GG structures were observed as well. These minor structures were not present in the reference GG mixture obtained from normal adult cerebral tissue of frontal cortex (Figure 1B). Proportions of individual GG separated by HPTLC, as quantified by densitometric analysis, differed to a great extent between gliosarcoma and normal brain tissue (Table I). In gliosarcoma, the GG fraction (upper and lower band) with migration properties of GD3 was the major one, accounting for almost 50% of the total GG content. GM3, GD2/GT3, and the fraction corresponding to 3'-isoLM1 were, beside GD3, present in much higher relative proportions than in normal brain tissue (Table I), whereas proportions of more complex structures were lower in gliosarcoma, several fold (GM1a/GM1b, GD1a, and GD1b) to slightly (GM2, GT1b, and GQ1b).


View this table:
[in this window]
[in a new window]

  		Table I. Relative amounts of TLC-separated ganglioside fractions, expressed as percentage distribution of ganglioside bound-sialic acids, and total ganglioside concentrations, expressed as total amount of ganglioside bound-sialic acids in (microgram/gram) tissue w.w. in gliosarcoma versus normal frontal cerebral cortex.

 

Figure 1
View larger version (43K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 1. Compositions of GGs isolated from human gliosarcoma specimen (A) and normal human frontal cortex (B), as revealed by HPTLC followed by densitometric scanning.

 
Compositional analysis of gliosarcoma GGs by high-performance chip-based ESI-QTOF and ESI-FT-ICR-MS
Purified native GG mixture from gliosarcoma was analyzed by both chip-based nano-ESI-QTOF and nano-ESI-FT-ICR mass spectrometric approaches of screening using MS1 negative ion mode. The obtained mass spectra are shown in Figures 2 and 3, respectively. GG molecular ions detected by chip-based nano-ESI-QTOF-MS1 and in parallel by nano-ESI-FT-ICR-MS1 are listed in Table II, together with their putative structural assignment. Both spectra featured a rich molecular ion pattern, showing the presence of all chromatographically detected fractions as well as some individual glycan species, highly heterogeneous in their ceramide structures. Several species not recognized by HPTLC were also detected. GD3, particularly d18:1/18:0, d18:1/24:1, and d18:1/24:0 (m/z 1470.827, 1552.898, and 1554.918), were present with the highest abundance, but surprisingly a high abundance of counterpart O-acetylated GD3 derivatives, particularly d18:1/18:0 and d18:1/20:0 (m/z 1512.855 and 1540.890), was also observed. High intensity ions corresponding to GM3 and GD2 species carrying different ceramides were present as well. Several considerably abundant ions related to GM2, GM1, and/or their isomers nLM1 and LM1, as well as to GD1 species characterized by heterogeneity in composition of their ceramide moieties, were found; GT1 species were represented by low-abundance ions. Unusual ions at m/z 1881.24, 1910.146 (1909.25), and 1992.53, corresponding to Hex-HexNAc-nLM1 d18:1/16:0, d18:1/18:0, and d18:1/24:1, respectively, were also recognized. Some low-abundance ions, detected by chip-based nano-ESI-QTOF MS1, were putatively assigned to: GT3 d18:1/24:0 and d18:0/24:0 (m/z 1845.40 and 1847.42), Fuc-GT1 d18:1/16:0 and d18:0/16:0 [m/z 2245.02 and 2246.98 (2247.494)], GalNAc-GT1 d18:1/18:0 and d18:1/20:0 [m/z 2330.263 (2329.97) and 2357.90]. Besides O-acetylated GD3 species, which was structurally confirmed by MS/MS (see Structural characterization of gliosarcoma-associated GGs by tandem MS), additional minor ions indicating presence of several other O-acetylated derivatives were recognized, and the species were putatively assigned to: O-Ac-GM3 d18:1/20:0, d18:1/22:1, d18:1/22:0, d18:0/22:0, and d18:1/24:2 (m/z 1249.786 (1249.92), 1275.801 (1275.87), 1277.80, 1279.818 (1279.91), and 1301.828 (1323.823, [MNa-2H]–), respectively), di-O-Ac-GD3 18:1/20:0 (m/z 1582.63), O-Ac-GT3 18:1/24:1 and 18:1/24:0 (m/z 1885.27 and 1887.22); the presence of a rare GM3 18:1/23:0 and 20:1/23:0 species (carrying odd fatty acid residues), characterized by almost identical m/z value as the O-Ac-GM3 d18:1/20:0 and d18:1/22:0, respectively, cannot be excluded. However, O-Ac-GD3 d18:1/20:0 (detected at m/z 1540.890; calculated m/z 1540.874) shares almost identical m/z value with GD3 d18:1/23:0 (calculated m/z 1540.911), but its structure was nevertheless proved by MS/MS, and no ions indicating d18:1/23:0 ceramide residue were found. GQ1 species, were not detected by MS, although HPTLC analysis showed the presence of a certain GGs migrating in the vicinity of GQ1b, detected and generally occurring in the normal human brain (Figure 1A and 1B).


Figure 2
View larger version (44K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 2. Negative ion mode nano-ESI-Chip-QTOF-MS1 of the native gliosarcoma ganglioside mixture. ESI voltage, 1.60 kV; sampling cone, 80 V; acquisition, 2 min; average sample consumption, 0.5 pmol.

 

Figure 3
View larger version (19K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 3. Negative ion mode nano-ESI-FT-ICR-MS1 of the native gliosarcoma ganglioside mixture. Solvent, methanol; average sample concentration, 3 pmol µL–1; capillary exit voltage, –150 V.

 

View this table:
[in this window]
[in a new window]

  		Table II. Ganglioside and asialo-ganglioside speciesa in gliosarcoma brain tumor tissue detected by chipESI-QTOF-MS1 and nano-ESI-FT-ICR-MS1

 
Structural characterization of gliosarcoma-associated GGs by tandem MS
In order to provide a reliable structural identification, several detected species were subjected to structural analysis by nano-ESI-QTOF-MS/MS sequencing in the negative ion mode. Here, we present sequencing data confirming structural identification primarily of those species detected to be highly abundant in gliosarcoma versus normal human brain, and which have been so far recognized by other authors to represent brain tumor-associated antigens. As the first example, the ion at m/z 1554.93 corresponding to the molecular [M-H+]– ion of the GD3 (d18:1/24:0) was selected for MS/MS sequencing. The product ion spectrum (Figure 4) revealed that, due to a broader selection window, the ion at m/z 1552.94 assigned to the molecular [M-H+]– ion of the GD3 (d18:1/24:1) was selected and sequenced in parallel with its 2 µm larger analog. The sequences of both species were represented by the complete series of Y-type ions (Y0–Y3), confirming their ceramide compositions d18:1/24:0 and d18:1/24:1 (Y0 ions at m/z 648.62 and 646.62, respectively), and identical carbohydrate moieties, Neu5Ac-Neu5Ac-Gal-Glc, attached to defined ceramides. The characteristic disialo-group was additionally proved by several related ions of B-series (B2, B2-CO2, and B2-H2CO3 at m/z 581.18, 537.18, and 519.17, respectively), accompanied by a minor C2-ion at m/z 599.19.


Figure 4
View larger version (29K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 4. Negative ion mode nano-ESI–QTOF MS/MS of the [M-H]– ions at m/z 1552.94 and m/z 1554.93 corresponding to the GD3 (d18:1/24:1) and GD3 (d18:1/24:0), respectively. Solvent, methanol; ESI voltage; 1000–1250 V; for precursor ions isolation the LM and HM parameters were set to 3; collision energy; 30–45 eV; collision gas pressure: 15 psi; acquisition: 3 min; average sample consumption: 1 pmol. Inset, the fragmentation scheme of GD3.

 
The product ion spectrum of the parent ion at m/z 1673.93 fitting to the [M-H+]– ion of the GD2 (d18:1/18:0), represented in Figure 5, exhibited a fragmentation pattern relevant to deduce the structure of proposed GD2 species. The ceramide composition d18:1/18:0 was confirmed by the low-abundance Y0 ion at m/z 564.14 and indirectly from the double cleavage ion C4/B1{alpha} (or Y3{alpha}/Z0) at m/z 835.47 formed by the loss of both Cer and NeuAc residues. Additional Y-series fragment ions (Y2{alpha}/B2ß, Y2{alpha}, and Y3{alpha} at m/z 888.68, 1091.79, and 1382.87, respectively) provided the evidence for the sugar sequence GalNAc (Neu5Ac)2 Gal-Glc-Cer. The structure-specific disialo-fragment was represented by both B2{alpha} and C2{alpha} ions at m/z 581.20 and 599.34.68, accompanied by a sodiated B2{alpha}(Na) counterpart.


Figure 5
View larger version (29K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 5. Negative ion mode nano-ESI-QTOF-MS/MS of the [M-H]– ion at m/z 1673.93 corresponding to the GD2 (d18:1/18:0). Acquisition time: 7 min; average sample consumption: 2.5 pmol. Other conditions as in Figure 4. Inset, the fragmentation scheme of GD2.

 
To confirm the presence of O-acetylated GD3 derivatives, the [M-H+]– ion detected by chip nano-ESI QTOF MS1 at m/z 1540.88, assigned to O-Ac-GD3 d18:1/20:0 was further subjected to MS/MS analysis. As revealed by the product ion spectrum (Figure 6), an additional isolation and partial sequencing of the parent ion at m/z 1544.93 corresponding to GM1, nLM1 and/or LM1 (d18:1/18:0) occurred in parallel with the chosen ion at m/z 1540.96. The same occurred during the selection and fragmentation of the parent ion at m/z 1544.89 (Figure 7), resulting in a partial isolation and sequencing of two additional ions at m/z 1540.88 and 1542.87 (inset in Figure 7). The ion at m/z 1542.87 was recognized as a di-unsaturated (d18:1/18:1) analog of the monosialo-tetraose species at m/z 1544.89. Nevertheless, both product ion spectra (Figures 6 and 7) contained sufficient information to confirm structural features of the parent ions. The fragment ions related to O-Ac-GD3 d18:1/20:0, originating from the parent ion at m/z 1540.96, detected in both spectra are denoted by an asterisk. The Y2-ion at m/z 916.81 represents the evidence of the sequence Gal-Glc-Cer– carrying d18:1/20:0 ceramide. The prominent Y3-ion at m/z 1249.84, consistent with a structure Ac-O-NeuAc-Gal-Glc-Cer–, arising upon cleavage of the terminal NeuAc residue from the intact O-Ac-GD3 species, strongly indicates O-acetylation of the inner and not the terminal NeuAc residue. The ions B2 at m/z 623.22 (Figure 6) and B2/B1 at m/z 332.22 (Figure 7) correspond to O-Ac(NeuAc)2– and Ac-O-NeuAc–. Although these ions are not indicating which NeuAc residue carries the O-Ac group, they are of importance for confirming the O-Ac attachment to NeuAc. The deacetylated counterpart B- and C-type ions at m/z 599.21, 581.20, 537.20, and 290.09, representing different forms of NeuAc2–and NeuAc–fragments, were also abundant and accompanied by two analogous A3-type ring cleavage ions at m/z 701.54 and 729.54. The proposed sequence was additionally confirmed by the low-abundance ion at m/z 1207.74 assigned to the deacetylated Y3-ion sequence, Y3-Ac.


Figure 6
View larger version (34K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 6. Negative ion mode nano-ESI-QTOF-MS/MS of the [M-H]– ion at m/z 1540.96 corresponding to the O-Ac-GD3 (d18:1/20:0). ESI voltage, 1000–1250 V; for precursor ions isolation the LM and HM parameters were set to 3; collision energy: 25–40 eV; collision gas pressure: 5–10 psi.; acquisition time, 11 min; average sample consumption, 3.5 pmol. Inset, the fragmentation scheme of O-Ac-GD3.

 

Figure 7
View larger version (35K):
[in this window]
[in a new window]
[Download PowerPoint slide]
  		Fig. 7. Negative ion mode nano-ESI-chip-QTOF-MS/MS of the [M-H]– ion at m/z 1544.89 corresponding to a monosialotetraose (d18:1/18:0) species, presumably GM1, nLM1 and/or LM1. ESI voltage, 1.60 kV; collision energy: 45–85 eV; acquisition: 50 min; average sample consumption: 15 pmol. Upper inset, the fragmentation scheme of the terminally sialylated monosialotetraose species. Lower inset: zoomed region of the nano-ESI-Chip-QTOF-MS1, evidencing the low abundance of the ion at m/z 1544.89 comparing to the ion at m/z 1540.96.

 
The fragmentation pattern obtained by MS/MS sequencing of the [M-H+]–parent ion at m/z 1544.89 (Figure 7) contained sufficient information to deduce structural characteristics of the major monosialotetraose isobar/isomer, represented by a sequence NeuAc-Gal-HexNAc-Gal-Glc-Cer (d18:1/18:0). However, the collected HPTLC and MS data do not permit distinction between the possible oligosaccharide chain series: ganglio-, neolacto-, and lacto-series. Actually, based on our previous findings (Metelmann et al. 2001Go; Vukelic et al. 2001Go) and findings of other authors (Fredman 1994Go), as explained in the ‘Discussion section’, we suspect that at least three or even four monosialotetraose GG species, in particular nLM1, LM1, GM1b, and GM1a all carrying Cer (d18:1/18:0), were present and sharing the same molecular ion (m/z 1544.89 in this case). The proposed major sequence NeuAc-Gal-HexNAc-Gal-Glc-Cer (d18:1/18:0), corresponding to nLM1, LM1, and GM1b, was deduced from the complete, intensively abundant, Y-ion series (Y0–Y4), with particularly highly intensive diagnostically important Y2-ion at m/z 888.77 indicative of a free Gal-Glc-Cer– fragment, carrying no sialyl residue at the inner Gal. The ion series C3–C5, accompanied by B-type counterpart ions and A4-type ring cleavage ions, corroborate the proposed major sequence, although not being discriminative regarding the attachment site of the NeuAc residue. On the other hand, the low-abundance ion at m/z 1179.80, consistent with the NeuAc-Gal-Glc-Cer– (d18:1/18:0) fragment structure, represents an argument for the presence of a branched monosialotetraose GG species, such as GM1a. As already mentioned above, the presence of additional minor monosialotetraose containing Cer (d18:1/18:1) was recognized from the corresponding low-abundance Y0–Y2 ion series accompanied by the analogous Z1-ion, as indicated in the spectrum (Figure 7).


   Discussion
Top
 Abstract
 Introduction
 Results
 Discussion
Materials and methods
 Conflict of interest statement
 Acknowledgments
 References
 
Among various low- and high-grade glioma types, including glioblastoma multiforme that have been so far investigated regarding the GG composition (Fredman, 1994Go; Fredman et al. 1999Go; Wagener et al. 1999Go; Becker et al. 2000Go; Popko et al. 2002Go), none of them was classified as a gliosarcoma variant. Besides, the data upon glioma GG collected in previous studies has been obtained using HPTLC, immunochemical and immunohistochemical detection, and eventually HPLC. All of these methods showed limited potential concerning efficient separation, detailed profiling, and structural characterization of GG species in native mixtures, especially if less abundant. In many cases, procedures for GG purification involving alkali hydrolysis have been used, causing degradation of O-acetylated and lactonized derivatives. Only few individual species, detected by specific antibodies, have been structurally confirmed by MS or NMR (Fredman et al. 1988Go).

In this context, our study is the first one presenting in detail the GG pattern in a human gliosarcoma specimen and also the first one demonstrating that the total number of distinct species expressed in a certain glioma is actually much larger than the postulated before. Our approach combining sensitive and accurate MS methods with conventional HPTLC enabled the detection and structural assignment of 73 distinct species expressed in gliosarcoma, as well as characterization of structural features of single GG component in the complex native mixture.

The ion assignment and postulation of structures was carried out based on previous knowledge upon this type of substrates (Wagener et al. 1999Go; Vukelic et al. 2001Go), existing MS and MS/MS data upon the normal human brain GG expression and structure (Zamfir et al. 2004Go) and biosynthetical pathway criteria corroborated with the data acquired here by HPTLC, chipESI-QTOF-MS and MS/MS, and FT-ICR-MS.

The chromatographic and densitometric analysis indicated a several fold decrease (7.4x) of the total GG content in gliosarcoma versus normal human brain, which is in agreement with the data reported previously for other glioma types (Fredman 1994Go).

By combining the two different MS approaches, we collected evidence (Table II) of a much higher complexity of gliosarcoma-associated GG composition than initially detected by HPTLC (Figure 1 and Table I). This shows that all individual HPTLC separated bands contained several distinct species with similar migration properties, which were distinguished by MS. Nevertheless, the MS and HPTLC patterns were in excellent agreement. Most of particular glycan structures showed a high heterogeneity with respect to ceramide moiety. The HPTLC pattern of gliosarcoma GGs is to a certain extent comparable with the patterns of glioblastoma multiforme specimens reported by other authors (Fredman 1994Go; Becker et al. 2000Go). In the case of gliosarcoma, GD3 species carrying different ceramides were found to account for nearly 50% of the total GG content, being the most prominent fraction, and the only fraction expressed in higher absolute concentration than in normal human frontal cortex (approximately 128%). The relative abundance of GD1a fraction was observed to be higher, whereas the GD1b and the GT1b lower than in the grade-comparable reported glioma.

The GG composition of gliosarcoma was found to be highly altered in comparison with the composition of normal human brain (Figure 1 and Table I). The relative abundances of GD3, GD2/GT3, and GM3 fractions were 9.4-, 5.6-, and 3.1-folds higher, respectively, whereas GM1/nLM1, GD1/nLD1, and GD1b fractions showed 14.9-, 2.3-, and 6.7-fold lower relative abundance. All these species were detected by MS with comparable intensities of the corresponding molecular ions, while GD3 and GD2 were structurally confirmed as well. Further, at least two minor TLC fractions, X1 and X2, were observed, the first one corresponding, according to TLC-migration property, to isoLM1 species. By tandem MS (Figure 7), we provided the structural evidence upon its presence. This species was already reported as a characteristic glioma-associated structure, which was not detected so far in the adult human brain (Fredman 1994Go). Although the MS evidence does not permit distinguishing between monosialotetraose isomers bearing sialylation at the terminal Gal residue, on the basis of our previous data (Metelmann et al. 2001Go; Vukelic et al. 2001Go) and the HPTLC pattern, we suspect that beside isoLM1, two additional isomers nLM1 and GM1b also sialylated at the terminal Gal residue, are present as well. MS data also provided evidence for the presence of a monosialotetraose species carrying SA at the inner Gal, consistent with GM1a isomer. MS reveals also a quite prominent abundance of O-Ac-GD3 (d18:1/18:0) and (d18:1/20:0), and strongly supports the presence of several unusual minor species, such as GM4, Hex-HexNAc-nLM1, Gal-GD1, GT3, Fuc-GT1, GalNAc-GT1, O-Ac-GM3, di-O-Ac-GD3, and O-Ac-GT3, not previously reported as glioma-associated GGs. Most of these species represent the so-called fetal brain-associated GGs i.e. developmentally regulated antigens, which are only minor components of the normal brain as observed by us previously (Zamfir et al. 2004Go).

The results obtained by tandem MS are consistent with the presence of gliosarcoma-associated O-Ac-GD3 isomer bearing O-acetylation at the inner SA residue, a form that has not yet been structurally confirmed. We cannot claim, however, that this novel isomer is the sole O-Ac-GD3 in gliosarcoma. It has been observed that at physiological pH, the 7- and 8-O-acetyl esters tend to migrate to position 9 within the same (terminal) Neu5Ac residue (Varki and Diaz 1984Go). The potential inter-Neu5Ac migration of the O-acetyl group from the outer Neu5Ac (C-9, -8 or -7) to the inner Neu5Ac residue is less probable considering the longer distance. On the other hand, the MS detection of O-Ac-GM3 structure(s) corroborates the natural occurrence of the identified O-Ac-GD3 isomer. O-Ac-GM3 may serve as a natural precursor molecule for the synthesis of the postulated O-Ac-GD3 form by the same sialyltransferase II responsible for the sialylation of GM3. The fact that gliosarcoma, as the highest malignancy grade brain tumor, contains much higher amount of potentially proapoptotic GD3 than of the O-acetyl GD3 species supports the recent assumption (Kniep et al. 2006Go) that O-Ac-GD3 could by itself be responsible for the protection of tumor cells from apoptosis.

The classification of human gliomas is currently based on neuropathological criteria. Prognostic and therapeutic parameters are dependent upon whether the tumors are of astrocytic or oligodendroglial origin. GA1 was postulated as one of three markers characterizing the oligodendroglial versus astrocytic origin (Popko et al. 2002Go). By MS screening, we detected several HexNAc-Hex-Hex-Cer species differing in ceramide composition. Such a structure corresponds to GA2, asialo-GM2. Therefore, although the MS analysis cannot exclude the presence of Lc3Cer, we assume that gliosarcoma, classified as a tumor of astrocytic origin, expresses GA2, which might serve as an additional marker distinguishing it from oligodendroglial tumors.

GD2 is one of the GG epitopes used for production of antitumor vaccines being under clinical trials in treatment of recurrent or progressive gliomas (Becker et al. 2002Go). Also, GD3 has been reported to be responsible for the glioma cell proliferation and studied as a potential target for antibodies able to inhibit cell proliferation (Hedberg et al. 2000Go). This is in contradiction with the recent evidence postulating the proapoptotic effects of GD3 (Chen and Varki 2002Go), which might be reversed by 9-O-acetylation. Surprisingly, other O-acetylated GG have been lately described to have inhibitory/antiproliferative activity against glioma cell invasion. An efficient inhibitor of astroblast and astrocytoma division, O-acetylated GD1b, called neurostatin, carries the O-Ac group at the outer SA residue (Abad-Rodriguez et al. 1998Go). Recent studies suggested that the sugar structure and the O-acetylation cooperate in improving the antimitotic activity of O-acetylated compounds (Romero-Ramirez and Nieto-Sampedro 2004Go). In gliosarcoma tissue analyzed in this study, no MS evidence for the presence of O-acetylated GD1b structure was found, which correlates with proliferative and invasive nature of the tumor.

We suppose that the reduction in the total GG content and the altered pattern in gliosarcoma versus control tissue is the result of both a lower overall biosynthetic rate, due to change in expression of certain glycosyltransferases, and higher turnover rate. A higher expression of sialyltransferase II (GD3 synthase) and a lower expression of galactosyltransferase II could in part explain the very high GD3 and very low GM1a, GD1a, and GD1b abundances.

Worth examining in further studies is whether the gliosarcoma cell proliferation could be suppressed by: (i) manipulations such as down-regulation/inhibition of the sialyltransferase II and/or up-regulation of the galactosyltransferase II and (ii) treatment with GM1a and/or GD1a, found to be diminished in gliosarcoma. Besides, several identified unusual GG antigens, here demonstrated as gliosarcoma-associated, but less common for healthy tissues, particularly the novel O-Ac-GD3 isomer, intrude as potential candidates for production of more selective polyvalent immunotherapeutic agents against gliosarcoma.

In conclusion, by a three-stage investigation based on chipESI-QTOF MS and tandem MS and FT-ICR-MS, it was found that the total number of distinct individual GG species expressed in an analyzed gliosarcoma specimen is significantly higher than previously assessed and reported in other glioma specimens. Moreover, the unique feature of tandem MS, to provide structural data at high sensitivity, enabled reliable structural characterization of the gliosarcoma-associated GD3, characteristic O-Ac-GD3 form, GD2 and monosialotetraose species, presumably isoLM1, nLM1, GM1b, and GM1a. A number of unusual GG species, most of which considered as brain developmental antigens and not previously reported as glioma-associated, was also evidenced.

To our knowledge, such a detailed GG profiling followed by structural analysis has not been achieved before for any tumor entity. Therefore, we consider that this study provides the methodological platform for characterizing other glioma entities in order to identify biomarkers valuable in early diagnosis and therapy.


   Materials and methods
Top
 Abstract
 Introduction
 Results
 Discussion
 Materials and methods
Conflict of interest statement
 Acknowledgments
 References
 
Gliosarcoma and normal brain tissue characterization
Brain tumor localized in frontotemporal cortex of right hemisphere in male patient, age 47, was clinically diagnozed using computerized tomography and magnetic resonance. Brain tumor sample was obtained during surgical procedure. The histopathological diagnosis of gliosarcoma, grade IV (according to WHO classification, 2000) was confirmed using specific silver stain showing mesenchymal proliferation (Department of Neuropathology, Faculty of Medicine, University of Zagreb, Croatia). The sample of frontal cortex from a normal adult human brain (42 years; the person died in a traffic accident) was dissected to serve as a control; the brain was obtained from the Department of Forensic Medicine, Faculty of Medicine, University of Zagreb, Croatia. Tissue samples used for biochemical analysis were weighed and stored at –20°C after careful removal of blood vessels and necrotic elements.

Ganglioside GG extraction and purification
GG extraction was performed according to the method of Svennerholm and Fredman (1980)Go, as modified in our laboratory (Vukelic et al. 2001Go). Tissue sample was weighed and homogenized in ice-cold distilled water (W) as to obtain the 10% homogenate. Lipids were extracted twice using solvent mixture of chloroform (C):methanol (M) (1:2, by volume), followed by partition and repartition by adding M and W to a final volume ratio 1:1:0.8. Upper phases containing polar GSLs (GG) were collected. Dried GG extract was finally purified by dialysis (overnight, at 4 °C) and used for quantitative and qualitative GG analysis.

Quantitative and qualitative analysis of GGs in gliosarcoma
The purified native GG mixtures from gliosarcoma and from adult frontal cortex, serving as a control, were analyzed in parallel. Quantitative analysis of total GG concentration was performed according to the spectrophotometric method of Svennerholm (1957)Go, as modified by Miettinen and Takki-Luukkainen (1959)Go. The absorbances of samples and N-acetylneuraminic acid (SA) used as a standard in a range of known concentrations were determined at 580 nm; the concentrations of GG-bound sialic acidSAs (GG-SA) were expressed as microgram GG-SA per gram of fresh tissue w.w.

Qualitative analysis was performed by HPTLC separation of individual GG on silica gel plate. The samples of purified GG extracts were dissolved in an adequate volume of C:M (1:1) and aliquots containing 10 µg of GG-SA were applied to the plate. Also, the standard mixtures of GG were applied to the same plate: GM2 GG from the human lung tissue and GG mixture from the human cerebellum were prepared and purified in our laboratory. The plates were developed in a solvent system containing C, M, and 12 mM MgCl2 (58:40:9, by volume). After drying, plate was sprayed with resorcinol reagent and heated for 30–45 min until GG fractions appeared as bluish bands. Finally, HPTLC separated and visualized GG fractions were subjected to laser densitometric scanning (LKB 2202 Laser Ultrascan, LKB, Bromma, Sweden) at 580 nm, as described by (Trbojevic-Cepe and Kracun 1990Go), enabling relative quantification of individual GG, expressed as their relative proportion (%) in a sample.

Sample preparation for MS
For MS analysis, the stock solution of the GG sample was prepared by dissolving the dried material in M to be stored at –20 °C. Dilution of the stock solution with pure M yielded the working aliquots at the roughly estimated concentrations of about 10 pmol/µL for capillary-based nano-ESI-FT-ICR-MS and 2–3 pmol/µL for capillary-based and chip-based nano-ESI-QTOF-MS experiments. M was obtained from Merck (Darmstadt, Germany) and used without further purification.

Quadrupole time-of-flight mass spectrometry
All QTOF MS and tandem MS experiments were conducted on an orthogonal hybrid QTOF (QTOF Micromass, Manchester, UK) equipped with nano-ESI ion source. The QTOF mass spectrometer is interfaced to a personal computer running the MassLynx software to control the instrument, acquire, and process the mass spectra. All mass spectra were acquired in negative ion mode, which was previously shown to be best suited for GG analysis (Metelmann et al. 2001Go; Vukelic et al. 2001Go). The values of the nano-ESI source parameters such as ESI capillary, sampling cone potentials, and desolvation gas were optimized to give rise to an efficient ionization, to hinder the in source fragmentation and ensure a maximal ionic transfer into MS.

Tandem MS was performed by collision-induced dissociation (CID) at low energies using argon as a collision gas. For ion isolation, the LM and HM parameters were set to provide a fair compromise between the precursor ion isolation and measurement sensitivity. Collision energy and gas pressure were readjusted several times during the ongoing MS/MS experiment to enhance an optimal fragmentation of GG species. The product ion spectra were combined over scans acquired at variable collision energy within 25–85 eV range (Elab).

All mass spectra were calibrated using sodium iodide as a calibrant. The reference provided in negative ion mode a spectrum with a fair ionic coverage of the m/z range scanned in both MS and CID MS/MS experiments. The fragment ions were assigned according to the nomenclature of Domon and Costello (1988)Go.

Fourier-transform ion cyclotron resonance mass spectrometry
FT-ICR experiments were performed on a Bruker Apex II FT-ICR-MSF (Bruker Daltonik, Bremen, Germany) equipped with a 9.4 T superconducting actively shielded magnet (Magnex Scientific Ltd, Oxford, UK) and an InfinityTM cell. Gas-phase ions were generated from solution by nano-ESI in the negative ion mode using an Apollo ion source. In this configuration, the sample is introduced into the home-made noncoated glass capillaries, in which a stainless steel wire kept at the ground potential is inserted. The ESI parameters were optimized to allow an efficient ionization and to reduce the in-source fragmentation of molecular ions. The capillary exit voltage was varied within 100–300 V to ensure an optimal ion transfer into MS and to provide a high-ionic yield of the molecules. The ions were accumulated for 1–2 s in the hexapole located after the second skimmer of the ion source and then transferred into the ion cyclotron resonance (ICR) cell. Ions were trapped by the Sidekick method. All mass spectra were acquired in the broad band mode with 512 kpoints/scan and externally calibrated. For the calculation of the theoretical monoisotopic masses of ions, the values for atomic masses of the most abundant isotopes given by (De Laeter et al. 2003Go) were used. The mass of electron, generating the negative charge, was included as suggested by Mamer and Lesimple (2004)Go.

Automated chip-based nanoelectrospray
Fully automated chip-based nanoelectrospray was carried out on a NanoMateTM 100 incorporating ESI Chip technology (Advion BioSciences, Ithaca, USA) mounted to the QTOF mass spectrometer. The robot was controlled and manipulated by ChipSoft software operating under Windows system. The position of the electrospray chip was adjusted with respect to the sampling cone potential to give raise to an optimal transfer of the ionic species into the mass spectrometer. In order to prevent any contamination, a glass coated microtiter plate was used for all experiments. Five microliter aliquots of the working sample solutions were loaded onto the 96-well plate. The robot was programed to aspirate the whole volume of sample, followed by 2 µL of air into the pipette tip and then deliver the sample to the inlet side of the microchip. Electrospray was initiated by applying a voltage of 1.60 kV to the pipette tip and a head pressure of 0.4–0.6 psi. Following sample infusion and MS analysis, the pipette tip was ejected and a fresh tip and nozzle were used for each sample, thus preventing any cross-contamination or carry-over.


   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 Drs V. Hlavka and B. Kruslin (Department of Neuropathology, Clinical Hospital Center Rebro, and Croatian Institute for Brain Research, Faculty of Medicine, University of Zagreb) for providing the histopathological diagnosis of gliosarcoma. The valuable assistance of Dr Milica Trbojevic Cepe (Clinical Hospital Rebro, Zagreb) and Petra Golja (Pliva-Research Institute, Zagreb) is gratefully acknowledged. This work was supported by the Croatian Ministry of Science, Education and Sports through the project 0108120, the German Society for Research through the SFB 492, project Z2., and the Romanian Ministry of Education and Research through the "Research of Excellence", project CE.EX. 14/2005.


   Abbreviations
 
C, chloroform; CNS CID, collision-induced dissociation; GG, ganglioside; GG-SA, ganglioside-bound sialic acid; GSL, glycosphingolipid; HPTLC, high-performance thin-layer chromatography; M, methanol; nano-ESI-FT-ICR-MS, nano-electrospray ionization-Fourier transform ion cyclotron resonance mass spectrometry; QTOF MS, quadrupole time-of-flight mass spectrometry; W, water; w.w., wet weight


   References
Top
 Abstract
 Introduction
 Results
 Discussion
 Materials and methods
 Conflict of interest statement
 Acknowledgments
 References
 
Abad-Rodriguez J, Vallejo-Cremades M, Nieto-Sampedro M. Control of glial number: purification from mammalian brain extracts of an inhibitor of astrocyte division. Glia (1998) 23:156–168.[CrossRef][Web of Science][Medline]

Becker R, Rohlfs J, Jennemann R, Wiegandt H, Mennel HD, Bauer BL. Glycosphingolipid component profiles or human gliomas-correlation to survival time and histopathological malignancy grading. Clin Neuropathol (2000) 19:119–125.[Web of Science][Medline]

Becker R, Eichler MK, Jennemann R, Bertalanffy H. Phase I clinical trial on adjuvant active immunotherapy of human gliomas with GD2-conjugate. Br J Neurosurg (2002) 16:269–275.[CrossRef][Web of Science][Medline]

Chapman PB. Vaccinating against GD3 ganglioside using BEC2 anti-idiotypic monoclonal antibody. Curr Opin Investig Drugs (2003) 4:710–715.[Medline]

Chen HY, Varki A. O-acetylation of GD3: An Enigmatic Modification Regulating Apoptosis? J Exp Med (2002) 196:1529–1533.[Free Full Text]

De Laeter JR, Bohlke JK, De Bievre P, Hidaka H, Peiser HS, Rosman KJR, Taylor PDP. Atomic weights of the elements: Review 2000 (IUPAC technical report). Pure Appl Chem (2003) 75:683–800.[Web of Science]

Domon B, Costello CE. A systematic nomenclature for carbohydrate fragmentations in FAB-MS/MS spectra of glycoconjugates. Glycoconj J (1988) 5:397–409.[CrossRef][Web of Science]

Fernandez LE, Alonso DF, Gomez DE, Vazquez AM. Ganglioside-based vaccines and anti-idiotype antibodies for active immunotherapy against cancer. Expert Rev Vaccines (2003) 2:817–823.[CrossRef][Medline]

Fine HA. J. Toward a glioblastoma vaccine: promise and potential pitfalls (Editorial). Clin Oncol (2004) 22:1–4.

Fredman P, von Holst H, Collins VP, Granholm L, Svennerholm L. Sialyllactotetraosylceramide, a ganglioside marker for human malignant gliomas. J. Neurochem (1988) 50:912–919.[CrossRef][Web of Science][Medline]

Fredman P. Gangliosides associated with primary brain tumors and their expression in cell lines established from these tumors. Prog Brain Res (1994) 101:225–240.[Web of Science][Medline]

Fredman P, Mansson JE, Dellheden B, Bostrom K, von Holst H. Expression of the GM1-species, [NeuN]-GM1, in a case of human glioma. Neurochem Res (1999) 24:275–279.[CrossRef][Web of Science][Medline]

Fredman P, Hedberg K, Brezicka T. Gangliosides as therapeutic targets for cancer. BioDrugs (2003) 17:155–167.[CrossRef][Web of Science][Medline]

Froesch M, Bindila L, Baykut G, Allen M, Peter-Katalinic J, Zamfir AD. Coupling of fully automated chip electrospray to Fourier transform ion cyclotron resonance mass spectrometry for high-performance glycoscreening and sequencing. Rapid Commun Mass Spectrom (2004) 18:3084–3092.[CrossRef][Web of Science][Medline]

Hakomori S. Tumor-associated carbohydrate antigens defining tumor malignancy: basis for development of anti-cancer vaccines. Adv Exp Med Biol (2001) 491:369–402.[Web of Science][Medline]

Hakomori S. Glycosylation defining cancer malignancy: new wine in an old bottle. Proc Natl Acad Sci USA (2002) 99:10231–10233.[Free Full Text]

Hakomori S, Handa K. Glycosphingolipid-dependent cross-talk between glycosynapses interfacing tumor cells with their host cells: essential basis to define tumor malignancy. FEBS Lett (2002) 531:88–92.[CrossRef][Web of Science][Medline]

Hakomori S. Glycosynapses: microdomains controlling carbohydrate-dependent cell adhesion and signaling. An Acad Bras Cienc (2004) 76:553–572.[Web of Science][Medline]

Hedberg KM, Dellheden B, Wikstrand CJ, Fredman P. Monoclonal anti-GD3 antibodies selectively inhibit the proliferation of human malignant glioma cells in vitro. Glycoconj J (2000) 17:717–726.[CrossRef][Web of Science][Medline]

Hedberg KM, Mahesparan R, Read TA, Tysnes BB, Thorsen F, Visted T, Bjerkvig R, Fredman P. The glioma-associated gangliosides 3'-isoLM1, GD3 and GM2 show selective area expression in human glioblastoma xenografts in nude rat brains. Neuropathol Appl Neurobiol (2001) 27:451–464.[CrossRef][Web of Science][Medline]

IUPAC-IUB Joint Commission on Biochemical Nomenclature (JCNB). Nomenclature of glycolipids. Recommendations 1997. Eur J Biochem (1998) 257:293–298.[Web of Science][Medline]

Kelly KA, Kirkwood JM, Kapp DS. Glioblastoma multiforme: pathology, natural history and treatment. Cancer Treat Rev (1984) 11:1–26.[Web of Science][Medline]

Kniep B, Kniep E, Ozkucur N, Barz S, Bachmann M, Malisan F, Testi R, Rieber EP. 9-O-acetyl GD3 protects tumor cells from apoptosis. Int J Cancer (2006) 119:67–73.[CrossRef][Web of Science][Medline]

Krug LM. Vaccine therapy for small cell lung cancer. Semin Oncol (2004) 31:112–116.[CrossRef][Web of Science][Medline]

Luterbach J, Guttenberg R, Pagenstecher A. Gliosarcoma: a clinical study. Radiother Oncol (2001) 61:57–64.[CrossRef][Web of Science][Medline]

Machuca TN, Prevedello DM, Pope LZ, Haratz SS, Araujo JC, Torres LF. Gliosarcoma: report of four cases with immunohistochemical findings. Arq. Neuropsiquiatr (2004) 62:608–612.[Medline]

Malisan F, Testi R. GD3 ganglioside and apoptosis. Biochim Biophys Acta (2002) 1585:179–187.[Medline]

Mamer OA, Lesimple A. A common shortcoming in computer programs used to calculate weights in Daltons of ions having elemental compositions entered as input. J. Am Soc Mass Spectrom (2004) 15:626.[CrossRef][Web of Science][Medline]

Metelmann W, Vukelic Z, Peter-Katalinic J. Nano-ESI quadrupole time-of-flight mass spectrometry of gangliosides from human brain tissue. J. Mass Spectrom (2001) 36:21–29.[CrossRef][Web of Science][Medline]

Miettinen T, Takki-Luukkainen IT. Use of buthylacetate in determination of sialic acid. Acta Chem Scand (1959) 13:656–658.

Mitsuzuka K, Handa K, Satoh M, Arai Y, Hakomori S. A specific microdomain ("glycosynapse 3") controls phenotypic conversion and reversion of bladder cancer cells through GM3-mediated interaction of alpha3beta1 integrin with CD9. J Biol Chem (2005) 280:35545–35553.[Abstract/Free Full Text]

Noll EN, Lin J, Nakatsuji Y, Miller RH, Black PM. GM3 as a novel growth regulator for human gliomas. Exp Neurol (2001) 168:300–309.[CrossRef][Web of Science][Medline]

Ohgaki H, Biernat W, Reis R, Hegi M, Kleihues P. Gliosarcoma. In: Pathology and Genetic of Tumors of the Nervous System—Kleihues P, Cavenee WK, eds. (2000) 2nd ed. Lyon: IARC Press. 42–44.

Popko B, Pearl DK, Walker DM, Comas TC, Baerwald KD, Burger PC, Scheithauer BW, Yates AJ. Molecular markers that identify human astrocytomas and oligodendrogliomas. J Neuropathol Exp Neurol (2002) 61:329–338.[Web of Science][Medline]

Raffaghello L, Marimpietri D, Pagnan G, Pastorino F, Cosimo E, Brignole C, Ponzoni M, Montaldo PG. Anti-GD2 monoclonal antibody: a promising strategy in the prevention of neuroblastoma relapse. Cancer Lett (2003) 18:205–209.[CrossRef]

Romero-Ramirez L, Nieto-Sampedro M. Inhibiting human astrocytoma growth: structure-activity relationships in neurostatin related glycolipids. J Med Chem (2004) 47:4983–4984.[CrossRef][Web of Science][Medline]

Shukla GS, Krag DN. Selective delivery of therapeutic agents for the diagnosis and treatment of cancer. Expert Opin Biol Ther (2006) 6:39–54.[CrossRef][Web of Science][Medline]

Steiner G, Shaw A, Choo-Smith LP, Abuid MH, Schackert G, Sobottka S, Steller W, Salzer R, Mantsch HH. Distinguishing and grading human gliomas by IR spectroscopy. Biopolymers (2003) 72:464–471.[CrossRef][Web of Science][Medline]

Svennerholm L. Quantitative estimation of sialic acids II. A colorimetric resorcinol-hydrochloric acid method. Biochim Biophys Acta (1957) 24:104–111.

Svennerholm L. Chromatographic separation of human brain gangliosides. J Neurochem (1963) 10:613–623.[Web of Science][Medline]

Svennerholm L. Ganglioside designation. Adv Exp Med Biol (1980) 125:11.[Medline]

Svennerholm L, Fredman P. A procedure for the quantitative isolation of brain gangliosides. Biochim Biophys Acta (1980) 617:97–109.[Medline]

Trbojevic-Cepe M, Kracun I. Determination of gangliosides in human cerebrospinal fluid by high-performance thin-layer chromatography and direct densitometry. J Clin Chem Clin Biochem (1990) 8:863–872.

Vukelic Z, Metelmann W, Müthing J, Kos M, Peter-Katalinic J. Anencephaly: Structural characterization of gangliosides in defined brain regions. Biol Chem (2001) 382:259–274.[CrossRef][Web of Science][Medline]

Vukelic Z, Zamfir AD, Bindila L, Froesch M, Peter-Katalinic J, Usuki S, Yu RK. Screening and sequencing of complex sialylated and sulfated glycosphingolipid mixtures by negative ion electrospray Fourier transform ion cyclotron resonance mass spectrometry. J Am Soc Mass Spectrom (2005) 16:571–580.[CrossRef][Web of Science][Medline]

Varki A, Diaz S. The release and purification of sialic acids from glycoconjugates: methods to minimize the loss and migration of O-acetyl groups. Anal Biochem (1984) 137:236–247.[CrossRef][Web of Science][Medline]

Wagener R, Rohn G, Schillinger G, Schroder R, Kobbe B, Ernestus RI. Ganglioside profiles in human gliomas: quantification by microbore high performance liquid chromatography and correlation to histomorphology and grading. Acta Neurochir (Wien) (1999) 141:1339–1345.[CrossRef][Medline]

Zaia J. Mass spectrometry of oligosaccharides. Mass Spectrom Rev (2004) 23:161–227.[CrossRef][Web of Science][Medline]

Zamfir A, Vukelic Z, Bindila L, Peter-Katalinic J, Almeida R, Sterling A, Allen M. Fully-automated chip-based nanoelectrospray tandem mass spectrometry of gangliosides from human cerebellum. J Am Soc Mass Spectrom (2004) 15:1649–1657.[CrossRef][Web of Science][Medline]

Zamfir AD, Lion N, Vukelic Z, Bindila L, Rossier J, Girault HH, Peter-Katalinic J. Thin chip microsprayer system coupled to quadrupole time-of-flight mass spectrometer for glycoconjugate analysis. Lab Chip (2005) 5:298–307.[CrossRef][Web of Science][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:
17/5/504    most recent
cwm012v2
cwm012v1
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 arrow Search for citing articles in:
ISI Web of Science (2)
Right arrowRequest Permissions
Right arrow Disclaimer
Google Scholar
Right arrow Articles by Vukelic, Z.
Right arrow Articles by Zamfir, A. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Vukelic, Z.
Right arrow Articles by Zamfir, A. D.
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?