On the structural diversity of Shiga toxin glycosphingolipid receptors in lymphoid and myeloid cells...

On the structural diversity of Shiga toxin glycosphingo-

lipid receptors in lymphoid and myeloid cells determined

by nanoelectrospray ionization tandemmass spectrometry

Petra Hoffmann1, Marcel Hulsewig1, Sevim Duvar2, Holger Ziehr2, Michael Mormann3,

Jasna Peter-Katalinic3, Alexander W. Friedrich1, Helge Karch1 and Johannes Muthing1*1Institute of Hygiene, University of Munster, D-48149 Munster, Germany2Fraunhofer Institute for Toxicology and Experimental Medicine, D-38124 Braunschweig, Germany3Institute of Medical Physics and Biophysics, University of Munster, D-48149 Munster, Germany

Received 8 February 2010; Accepted 19 May 2010

Shiga toxin (Stx, synonymous to verotoxin, VT) binds with high and low affinity to the globo-series

neutral glycosphingolipids (GSLs), globotriaosylceramide (Gb3Cer or Gala4Galb4Glcb1Cer, also

known as CD77) and globotetraosylceramide (Gb4Cer or GalNAcb3Gala4Galb4Glcb1Cer), respect-

ively, which represent the targets of Stxs on many different cell types. B-cell-derived Raji cells and

THP-1 cells of monocytic origin are widely used for the investigation of Stx-mediated cellular

response, because Stx is known to cause cell death in both cell lines. Despite their functional

importance, the Stx receptors of Raji and THP-1 cells have so far not been investigated. This

prompted us to explore the structures of their GSL receptors in detail by means of nanoelectrospray

ionization quadrupole time-of-flightmass spectrometry (nanoESI-QTOF-MS)with collision-induced

dissociation (CID) in conjunction with Stx1 as well as anti-Gb3Cer and anti-Gb4Cer antibodies.

Using the combination of a thin-layer chromatography (TLC) overlay assay and MS1 and MS2

analysis we identifiedGb3Cer (d18:1, C24:1/C24:0) as the prevalent Stx1-receptor accompanied by less

abundant Gb3Cer (d18:1, C16:0) in the neutral GSL fraction of Raji cells. The same Gb3Cer species

but with almost equal proportions of the C24:1/C24:0 and C16:0 variants were found in THP-1 cells. In

addition, unusual hydroxylated Gb3Cer (d18:1, C24:1/C24:0) and Gb3Cer (d18:1, C26:1) could be

identified in trace quantities in both cell lines. As themost obvious difference betweenRaji and THP-

1 cells we observed the expression of Gb4Cer in THP-1 cells, whereas Raji cells failed to express this

elongation product of Gb3Cer. Both short- and long-chain fatty acid carrying Gb4Cer (d18:1, C16:0)

andGb4Cer (d18:1, C24:1/C24:0), respectively, were the prevalent Gb4Cer variants. This first report on

the differential expression of Gb3Cer and Gb4Cer and their structural diversity in lymphoid and

myeloid cell lines supports the hypothesis that such heterogeneities might play a functional role in

the molecular assembly of GSLs in membrane organization and cellular signaling of Stx-susceptible

cells. Copyright # 2010 John Wiley & Sons, Ltd.

Glycosphingolipids (GSLs) of the globo-series namely globo-

triaosylceramide (Gb3Cer, Gala4Galb4Glcb1Cer) and globo-

tetraosylceramide (Gb4Cer, GalNAcb3Gala4Galb4Glcb1Cer)

are strongly implicated in infectious processes in humans

and animals as binding ligands of Shiga toxins (Stxs) released

from Stx-producing Escherichia coli (STEC).1,2 Gb3Cer, also

known as CD77, permits surface binding of Stx via the

pentameric B-subunit of the AB5 toxin, which moves in a

retrograde fashion to the endoplasmic reticulum,3 where the

catalytically active A1 subunit exerts its cytotoxic function by

inhibiting protein biosynthesis, a process which leads to

death of the cell.4,5 Besides its cytotoxic effects on endothelial

cells, Stx1 (synonymous to verotoxin 1, VT1) is known to

induce apoptosis in various types of B-cell-derived Burkitt’s

lymphoma cells.6–8 Consequently, the Stx receptor Gb3Cer

has been mooted as a therapeutic target in various types of

Burkitt’s lymphoma.9–11 Among the different types of leuko-

cytes and leukocyte-derived cell lines, globo-series neutral

GSLs are not restricted to cells of B cell origin but are also

expressed by immune cells of the monocytic lineage.12,13

Myeloid cells have been reported to interact with Stxs 14,15

and are hypothesized to play a functional role in the transfer

of Stxs from the gut into the bloodstream, where the toxins

target mainly Gb3Cer receptors on endothelial cells resulting

in systemic complications.16–18 However, the entry process

of Stxs into the circulation remains unclear and the

RAPID COMMUNICATIONS IN MASS SPECTROMETRY

Rapid Commun. Mass Spectrom. 2010; 24: 2295–2304

Published online in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/rcm.4636

*Correspondence to: J. Muthing, Institute of Hygiene, University ofMunster, Robert-Koch-Str. 41, D-48149 Munster, Germany.E-mail: [email protected]

Copyright # 2010 John Wiley & Sons, Ltd.

involvement of GSLs in the mechanism of toxin delivery

supposed to be triggered by myeloid cells is still a matter of

debate.19–21

The leukocyte-derived Raji and THP-1 cell lines, repre-

senting cells of the B cell andmonocytic lineage, respectively,

are known to express neutral GSLs of the globo-series 12,13,22

and this makes them interesting candidates for the investiga-

tion of Stx-mediated cellular response. In fact, Stx-mediated

induction of cell death upon interaction with cell-surface

GSLs and subsequent toxin internalization has been reported

for Raji and THP-1 cells.6,23,24 However, despite their bio-

logical importance, the structures of the GSL-ligands of Stxs

have not been investigated so far neither for Raji nor for THP-

1 cells. This failure prompted us to explore the Stx receptors

of both cell lines in detail by mass spectrometry in

conjunction with Stx1- and GSL-specific antibodies. Using

the combination of a thin-layer chromatography (TLC)

overlay assay and nanoelectrospray ionization quadrupole

time-of-flight mass spectrometry (nanoESI-QTOF-MS) with

collision-induced dissociation (CID),25 we report here for

the first time on the full structural characterization of the

functional Stx1 receptors of Raji and THP-1 cells. Besides

differential expression of Gb3Cer and Gb4Cer in the two

cell lines, the structural diversity particularly of the fatty

acid composition of the ceramide moieties was elucidated

which is supposed to play a functional role in the molecular

assembly of GSLs in membrane organization and cell sign-

aling 26 as well as for the retrograde transport to the

intracellular targets (the ribosomal RNA) of Stx-susceptible

cells.27,28

EXPERIMENTAL

Cell culture and cell productionPermanent Raji and THP-1 cell lines were obtained from the

American Type Culture Collection (ATCC, Manassas, VA,

USA). Raji cells (ATCC, CCL-86) represent a Burkitt’s

lymphoma and THP-1 cells (ATCC, TIB-202) a myeloid cell

line, equivalent to cells of B lymphocytic and monocytic

lineage, respectively. Both cell lines were originally grown in

5% (v/v) fetal calf serum containing Dulbecco’s modified

Eagle’s medium (DMEM)/Ham’s F-12 (1:1). They were sub-

sequently adapted to and propagated under serum-free

conditions in DMEM/Ham’s F-12 (1:1) supplemented with

insulin, transferrin, selenit and 0.4% (w/v; Raji) or 1% (w/v;

THP-1) Albumax II (Invitrogen, Karlsruhe, Germany) in a

humidified atmosphere with 5% (v/v) CO2 at 378C. Cellproduction was performed in a 5 L bioreactor (Biostat

B-DCUR twin-multifermenter-system, Sartorius Stedim Sys-

tems GmbH, Melsungen, Germany). The temperature was

set to 378C, pH to 7.2 and pO2 to 30% air saturation. Raji

and THP-1 cells were produced in perfusion and in repeated

batch mode, respectively. The cells were harvested by

centrifugation, washed twice with phosphate-buffered saline

(PBS) and stored at –808C until extraction.

Extraction, isolation and purification of neutralGSLs from cellsTotal GSLs were successively extracted from the cells with

methanol, chloroform/methanol (1:2, v/v), chloroform/

methanol (1:1, v/v), and chloroform/methanol (2:1, v/v).

Supernatants were removed after centrifugation, pooled and

dried by rotary evaporation. Coextracted phospholipids

were saponified with 1M aqueous NaOH (2 h, 378C) follo-wed by neutralization with 10MHCl. Salts were removed by

dialysis against deionized water. Neutral GSLs were then

separated from acidic GSLs by anion-exchange chromatog-

raphywith DEAE Sepharose CL-6B (GEHealthcare, Munich,

Germany). The neutral GSL-containing fraction was applied

to a Silica gel 60 (Merck, Darmstadt, Germany) column

and whole neutral GSLs were eluted with chloroform/

methanol (40:60, v/v).29 Non-GSL impurities were removed

by fractionation of peracetylated neutral GSLs on a Florisil

(Merck) column according to Saito and Hakomori,30 as

recently described in detail by Li and co-workers.31,32 After

deacetylation and subsequent dialysis the neutral GSL

mixture was dried and adjusted to defined volumes of

chloroform/methanol (2:1, v/v) corresponding to 2.5� 106

cells/mL.

Reference GSLsA preparation of neutral GSLs from human erythrocytes

served as reference mixture for solid-phase antibody and

toxin overlay assays (see below). The preparation contained

neutral GSLs of the globo-series, mainly lactosylceramide

(Lc2Cer), Gb3Cer, and Gb4Cer, each with varying fatty acid

chain length.25,33 Abbreviations and corresponding stru-

ctures of neutral GSLs used in this study: Lc2Cer or

Galb4Glcb1Cer; Gb3Cer or Gala4Galb4Glcb1Cer; Gb4Cer

or GalNAcb3Gala4Galb4Glcb1Cer. The nomenclature of the

GSLs follows the IUPAC-IUB recommendations 1997.34

Stx1, anti-Stx1, anti-GSL, and secondaryantibodiesStx1 and polyclonal chicken antibodies against Gb3Cer and

Gb4Cer were used for primary detection of Stx-receptors

Gb3Cer and Gb4Cer as previously described.18,35,36 The

monoclonal mouse IgG-antibody 109/4-E9b (Sifin, Berlin,

Germany) was employed for the detection of Stx1.25,36

Alkaline phosphatase-conjugated goat anti-mouse IgG and

IgM and rabbit anti-chicken IgY were from Dianova

(Hamburg, Germany) and used as secondary antibodies.

High-performance thin-layer chromatography(TLC)Neutral GSLs were applied to precoated silica gel 60 high-

performance TLC plates (10 cm� 10 cm, 200mm thickness,

no. 1.05633.0001; Merck, Darmstadt, Germany) with an

automatic applicator (Linomat IV, CAMAG, Muttenz,

Switzerland) and separated for 20min in a solvent system

composed of chloroform/methanol/water (120:70:17, each

by volume) supplemented with 2mM CaCl2. Plates were

either stained with orcinol or used for the TLC overlay assay.

Orcinol- and immunostained GSL bands (see ‘TLC overlay

assay’) were scanned with a CD60 scanner (Desaga, Heid-

elberg, Germany, software ProQuantR, version 1.06.000).

Bands were quantified in reflectance mode at l¼ 544 nm

(orcinol) and l¼ 630 nm (indolylphosphate) with a light

beam slit of 0.1mm� 2mm.

Copyright # 2010 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2010; 24: 2295–2304

DOI: 10.1002/rcm

2296 P. Hoffmann et al.

TLC overlay assayThe silica gel layer of the TLC plate was impregnated with

polyisobutylmethacrylate (Plexigum P28, Rohm, Darmstadt,

Germany) after TLC separation and the plate submitted to the

immunostaining procedure as previously described.18,35,36

Stx1 was used at a concentration of 0.2mg/mL. Anti-Stx1

antibody was applied at 2mg/mL. Primary anti-GSL and

secondary alkaline phosphatase-labeled antibodies were

used in 1:2000 dilution. Bound antibodies were visualized

by color development using 0.05% (w/v) 5-bromo-4-chloro-

3-indolyl phosphate p-toluidine salt (BCIP; Roth, Karlsruhe,

Germany) in glycine buffer.37

Extraction of GSLs from TLC platesSubsequent to TLC immunodetection the plastic fixative

(Plexigum) was eliminated by threefold dipping of the plate

into distilled chloroform.25 The silica gel of immunostained

GSL bands was scraped off the glass layer and transferred

into a glass column fitted with a glass microfibre filter pad.

GSLs were eluted with chloroform/methanol/water (30:60:

8, v/v/v) and the eluates dried under a stream of nitrogen

at 378C. The dried residues were redissolved in distilled

methanol and subjected to mass spectrometry without any

further purification.

Mass spectrometryElectrospray ionization mass spectrometry (ESI-MS) was

performed according to Meisen and collaborators 25 using a

quadrupole time-of-flight (QTOF) mass spectrometer equi-

pped with a nanospray manipulator (Micromass, Manche-

ster, UK). Capillaries were made in-house from borosilicate

glass. High voltage was applied via a steel wire to the sample

solution. Positive ion mode was used for all experiments.

The capillary voltage was set to 1100 V, and cone potential

40 V. ForMS2 experiments, the singly charged precursor ions

were selected with the first quadrupole. Collision-induced

dissociation (CID) was performed using argon as collision

gas, whereby collision gas pressure was set to 15 psi and

collision energy was varied between 30 and 90 eV. The

nomenclature introduced by Domon and Costello 38,39 and

Adams and Ann 40 was used for the assignment of the

fragment ions obtained by CID.

RESULTS

Antibody- and Stx1-mediated TLC overlay assaydetection of Gb3Cer and Gb4Cer in Raji andTHP-1 cellsPurified neutral GSLs of Raji and THP-1 cells were separated

by TLC and either detected by orcinol staining or submitted

to TLC overlay assays employing anti-Gb3Cer antibody,

anti-Gb4Cer antibody and Stx1/anti-Stx1, as shown from left

to right in Fig. 1. Conventional orcinol stain of the neutral

GSLs revealed a rather simple pattern for Raji but a much

more complex GSL mixture for THP-1 cells. Neutral GSLs

from human erythrocytes which served as reference for

Gb3Cer and Gb4Cer (¼ standard) were co-chromatographed

in all TLC assays. The anti-Gb3Cer and anti-Gb4Cer immu-

nostains resulted in clear detection of Gb3Cer as the sole

globo-series neutral GSL in Raji cells, whereas THP-1 cells

expressed both, Gb3Cer and Gb4Cer. Relative quantification

revealed a 1.5 times higher content of Gb3Cer in THP-1

compared to Raji cells. Both Gb3Cer and Gb4Cer, known as

the high and less-effective receptors of Stx1, respectively,

were detected in the Stx1/anti-Stx1 binding assay with

preference for positive Gb3Cer detection and low intensity

binding towards Gb4Cer as expected. The anti-Gb3Cer

binding patterns of both cell lines were almost identical to

those of Stx1/anti-Stx1, showing a dominant upper (87%)

and a weak lower Gb3Cer band (13%) in Raji and app-

roximately equal intensities of upper (44%) and lower band

Gb3Cer species (56%) in THP-1 cells being in good agreement

with the orcinol stains. The observed chromatographic

heterogeneities are due to differences in the ceramide

moieties, most likely due to the variation of fatty acid chain

lengths resulting in double bands harboring GSLs with a

long- (upper band) or a short-chain fatty acid (lower band).

The comparison of the anti-Gb4Cer and the Stx1 overlay

assays of THP-1 GSLs revealed a difference in Gb4Cer

detection whereby the intensities of the antibody-detected

bands coincided with those of the orcinol stain intensities,

whereas a preferred binding of Stx1 to the upper band

Gb4Cer, most likely carrying a long-chain fatty acid, could be

observed.

MS1 investigation of antibody- and Stx1/anti-Stx1-detected Gb3Cer from Raji cellsIn order to elucidate the structural diversity of the antibody-

and Stx1/anti-Stx1-detected Gb3Cer species in detail, we

initially acquired MS1 spectra from silica gel extracts of the

TLC overlay assay positive bands of Raji cells, as shown in

Fig. 2. The predominant [MþNa]þ cationized molecules at

m/z 1156.75/1158.76 detected in the extract of the antibody-

derived overlay assay (Fig. 2(a)) correspond to Gb3Cer

(d18:1, C24:1/C24:0) accompanied by monosodiated ions

Figure 1. Stx1- and antibody-mediated TLC overlay detec-

tion of Stx receptors in the neutral GSL fractions of Raji and

THP-1 cells. Orcinol stain: Neutral GSLs equivalent to

2.5� 107 Raji and THP-1 cells, respectively, and 5mg of

reference neutral GSLs from erythrocytes (standard, Std)

were applied for orcinol staining. GSL amounts corresponding

to 2.0� 106 and 5� 106 cells were employed for the antibody

and Stx1/anti-Stx1 overlay assays, respectively. Standard

GSL amounts were 2.5mg for Stx1/anti-Stx1, 0.5mg for

anti-Gb3Cer and 0.05mg for anti-Gb4Cer detection.


DOI: 10.1002/rcm

ESI-MS/MS of Stx receptors in lymphoid and myeloid cells 2297

with m/z values at 1130.79, 1102.75, 1074.68, and 1046.66 at

minor abundance, which were assigned to Gb3Cer variants

with constant sphingosine (d18:1) and C22:0, C20:0, C18:0,

and C16:0 fatty acid, respectively. These data fit very well to

the immunostained band intensities of the TLC overlay assay

(see insert of Fig. 2(a)). The MS1 spectrum obtained from the

silica gel extract of Stx1/anti-Stx1-detected Gb3Cer bands

(Fig. 2(b)) revealed essentially the same ionic species when

compared to those from antibody-detected Gb3Cer. Again,

ionized Gb3Cer (d18:1; C24:1/C24:0) represented the domi-

nant species which were flanked by Gb3Cer variants with

saturated fatty acids from C22:0 to C16:0 appearing with

lower signal intensities. The relative ion signal intensities

in the MS1 spectrum agreed with the binding pattern of

the Stx1/anti-Stx1-positive bands of the TLC overlay assay

(see insert of Fig. 2(b)). Due to a generally better signal-to-

noise ratio in the mass spectrum of Stx1/anti-Stx1-detected

Gb3Cer-species, probably due to 2.5-times higher amount of

GSL applied for the TLC overlay assay (see Fig. 1), two very

low abundance ions at m/z 1184.79 and 1172.74/1174.76

evidenced the presence of unusual Gb3Cer (d18:1; C26:1) and

the hydroxylated forms of Gb3Cer (d18:1; C24:1/C24:0),

which were not detectable in the antibody-mediated TLC

overlay assay. The detected ions from both overlay assays

and the proposed Gb3Cer structures are summarized in

Table 1.

Tandem mass spectrometry of Stx1/anti-Stx1-detected Gb3Cer from Raji cellsThe proposed structures based on the MS1 data were further

confirmed by low-energy CID mass spectrometry of Stx1/

anti-Stx1-detected major Gb3Cer (d18:1, C24:1/C24:0) and

minor Gb3Cer (d18:1, C16:0) from Raji cells (see Fig. 2(b)). As

an example we show the MS2 spectrum of Stx1/anti-Stx1-

positive Gb3Cer (d18:1, C16:0) obtained after selection of the

precursor [MþNa]þ ions at m/z 1046.66. The fragmentation

spectrum and the adjacent scheme are shown in Fig. 3(a) and

(b), respectively, where the fragments are assigned according

to the nomenclature of Domon and Costello 38,39 and Adams

and Ann.40 Full series of Y- and Z-type ions and B- and C-

type ions, resulting from the sequential loss of three hexose

moieties of Gb3Cer, were detected. Ring cleavage-generated0,2A2 and 0,2A3, and NII ions, indicative for the presence of

sphingosine (4-sphingenine, d18:1), were achieved and gave

rise to the complete structure (for full list of fragment ions,

refer to Table 2). The same set of fragment ions was obtained

from molecular ions of the major Stx1/anti-Stx1-binding

Gb3Cer species with C24 fatty acids (not shown). These are

exemplarily listed in Table 2 for the precursor ions at m/z

1156.77, which could be allocated to the structure of Gb3Cer

(d18:1, C24:1).

MS1 investigation of antibody- and Stx1/anti-Stx1-detected Gb3Cer from THP-1 cellsThe [MþNa]þ ions detected in the silica gel extracts of

antibody- and Stx1/anti-Stx1-binding Gb3Cer species of the

neutral GSL fraction from THP-1 cells were characterized

accordingly. The major signals obtained from the antibody-

Figure 2. NanoESI-QTOF-MS1 spectra of (a) antibody- and

(b) Stx1/anti-Stx1-detected Gb3Cer species from Raji cells.

The spectra were obtained from the silica gel extracts of TLC

overlay assays and recorded in the positive ion mode. The

corresponding overlay assays are shown in the inserts and

the framed dotted rectangles indicate the areas from which

the silica gel was scraped off. A synopsis of the m/z values of

[MþNa]þ ions and the proposed Gb3Cer structures are

provided in Table 1.

Table 1. Monosodiated ions of Stx1 receptor Gb3Cer

acquired from the MS1 spectra and proposed structures of

antibody- or Stx1-detected neutral GSLs from Raji cells

[MþNa]þ,m/z Antibody-detected

[MþNa]þ,m/z Stx1-detected Proposed structure

1046.66 1046.66 Gb3Cer (d18:1, C16:0)1074.68 1074.68 Gb3Cer (d18:1, C18:0)1102.75 1102.71 Gb3Cer (d18:1, C20:0)1130.79 1130.75 Gb3Cer (d18:1, C22:0)1156.75 1156.77 Gb3Cer (d18:1, C24:1)1158.76 1158.78 Gb3Cer (d18:1, C24:0)n.d.a 1172.74 Gb3Cer (d18:1, C24:1),

hydroxylatedn.d.a 1174.76 Gb3Cer (d18:1, C24:0),

hydroxylatedn.d.a 1184.79 Gb3Cer (d18:1, C26:1)

a n.d.: not detected.


DOI: 10.1002/rcm


derived MS1 spectrum were detected at m/z 1046.65 and

1156.75/1158.75 and gave evidence for Gb3Cer (d18:1, C16:0)

and Gb3Cer (d18:1, C24:1/C24:0), as demonstrated in

Fig. 4(a). Accompanying ions of minor abundance at m/z

1074.69 and 1130.73 could be attributed to Gb3Cer (d18:1,

C18:0) and Gb3Cer (d18:1, C22:0). The same set and

distribution of ionic species was found in the MS1 spectrum

achieved from the silica gel extract of Stx1/anti-Stx1-det-

ected Gb3Cer bands (Fig. 4(b)) showing the highest abun-

dance for ions derived from Gb3Cer (d18:1, C16:0) followed

by less abundant ions indicative for Gb3Cer (d18:1, C24:1/

C24:0) and minor Gb3Cer species with sphingosine (d18:1)

Figure 3. NanoESI-QTOF-MS2 spectrum of Stx1-detected

Gb3Cer (d18:1, C16:0) from Raji cells with m/z 1046.66 (a)

and the corresponding fragmentation scheme with the mol-

ecular structure (b). The B/C- and Y/Z-type ions released by

cleavage of glycosidic linkages and the A-type ions and the NII

fragment obtained by internal sugar and ceramide cleavage,

respectively, are marked with their corresponding m/z values.

The fragment ions are listed in Table 2. Signals labelled with

asterisks correspond to ionic glycine clusters [(Gly–

HþNa)nþNa]þ (n¼ 1–4) arising from fragmentation of

[(Gly–HþNa)9þGly2þNa]þ precursor ions being isobaric

with the precursor ions. Cluster ions are experimentally

derived from glycine buffer which was used in the TLC overlay

assays.

Table 2. Fragment ions and corresponding m/z values

attained from MS2 spectra of Stx1-detected major Gb3Cer-

species from Raji cells

Fragment ions

Gb3Cer(d18:1, C16:0)m/z 1046.66

Gb3Cer(d18:1, C24:1)m/z 1156.77

m/z values m/z values

0,2A2 305.10 305.100,2A3 467.19 467.13B1; C1 185.03; 203.07 185.03; 203.05B2, C2 347.10; 365.11 347.10; 365.11B3; C3 509.16; 527.18 509.16; 527.16Y0; Z0 560.50; n.d.a 670.63; 652.66Y1; Z1 722.55; 704.59 832.67; 814.66Y2; Z2 884.64; 866.63 994.75; 976.71NII 264.27 264.29



(b) Stx1/anti-Stx1-detected Gb3Cer species from THP-1

cells. The spectra were obtained from the silica gel extracts

of TLC overlay assays and recorded in the positive ion mode.

The corresponding overlay assays are shown in the inserts

and the framed dotted rectangles indicate the areas from

which the silica gel was scraped off. A synopsis of the m/z

values of [MþNa]þ ions and the proposed Gb3Cer structures

are provided in Table 3. Asterisks indicate coextracted minor

Hex2HexNAc-ceramides with m/z values at 1087.68 and

1143.72 (a) and at 1087.66, 1143.75, and 1197.76/1199.81

(b) which were tentatively identified as Lc3Cer (d18:1, C16:0),

Lc3Cer (d18:1, C20:0), and Lc3Cer (d18:1, C24:1/C24:0),

respectively.


DOI: 10.1002/rcm


and C18:0 and C22:0 fatty acid. Ions at m/z 1184.79 and

1172.74/1174.76 giving rise to less intense signals point to

uncommon Gb3Cer (d18:1, C26:1) and hydroxylated species

of Gb3Cer (d18:1, C24:1/C24:0), which were detected in

addition to the species found in the GSL extract of the

antibody-stained bandsmost likely due to higher amounts of

applied GSLs in the Stx1/anti-Stx1 overlay assay (see Fig. 1).

Additionally, minor [MþNa]þ ions revealed the presence of

Hex2HexNAc-ceramides in both samples. These molecules

which are marked with asterisks in the two spectra were

tentatively identified as Lc3Cer species with sphingosine

(d18:1) and various fatty acids (C16:0, C20:0, and C24:1/

C24:0). Lc3Cer is known to co-migrate with Gb3Cer, but does

not interfere with the antibody or Stx1/anti-Stx1 TLC overlay

detection. The identified ions from both overlay assays and

the proposed Gb3Cer structures are summarized in Table 3.

Interestingly, the high versus low signal intensities of the

molecular ions corresponding to Gb3Cer (d18:1, C16:0)

and Gb3Cer (d18:1, C24:1/C24:0), respectively, in the MS1

spectrum do not reflect the almost identical proportion of

these Gb3Cer variants which appeared as equally orcinol and

antibody-stained upper and lower bands in the TLC assays

(see Fig. 1 and the insert of Fig. 4(a)).

Tandem mass spectrometry of Stx1/anti-Stx1-detected Gb3Cer from THP-1 cellsTo finally confirm the Stx-receptors of THP-1 cells, complete

compositional analysis was performed by ESI-QTOF-MS2

analysis for the main Stx1-binding species (see Fig. 4(b)) with

short- and long-chain fatty acids assigned to Gb3Cer (d18:1,

C16:0) and Gb3Cer (d18:1, C24:1/C24:0). As an example for

CID investigations, the fragmentation spectrum of Gb3Cer

(d18:1, C24:1/C24:0) at m/z 1056.77/1158.78 is shown in

Fig. 5(a) and the associated fragmentation scheme together

with the Gb3Cer structure are shown in Fig. 5(b). Proposed

structures were unequivocally verified by the appearance of

the complete series of Y- and Z-type ions and B- and C-type

ions obtained from successive cleavage of two galactose

moieties and one glucose molecule of Gb3Cer. Combined

with observed internal ring cleavages of galactose and

glucose being apparent by 0,2A2 and0,2A3 ions, respectively,

and NII ions, being diagnostic for sphingosine (4-sphingen-

ine (d18:1), proved the complete structure of the Stx1

receptors Gb3Cer (d18:1, C16:0) and Gb3Cer (d18:1, C24:1/

C24:0). The detected fragment ions of the twomost abundant

species Gb3Cer (d18:1, C16:0) and of the species Gb3Cer

(d18:1; C24:1) are listed in Table 4.

MS1 investigation of antibody- and Stx1/anti-Stx1-detected Gb4Cer from THP-1 cellsTwo immunopositive Gb4Cer bands were identifiedwith the

anti-Gb4Cer antibody in the neutral GSL fraction of myeloid

THP-1 cells, whereas lymphoid Raji cells did not express any

Gb4Cer species (see Fig. 1). The antibody- and Stx1/anti-

Stx1-derived GSL extracts of THP-1 cells were submitted to

mass spectrometry and the MS1 spectra are shown, together

with the related TLC overlay stains, in Fig. 6(a) and (b),

respectively. In analogy to Gb3Cer expression, the [MþNa]þ

ions of the antibody-detected double band gave evidence for

the presence of Gb4Cer variants containing sphingosine

(d18:1) and the same set of fatty acids with chain lengths

ranging between 16 and 24 carbon atoms. The ions at m/z

1249.72 could be attributed to Gb4Cer (d18:1, C16:0) being

the most and those at m/z 1359.83/1361.84 to Gb4Cer (d18:1,

C24:1/C24:0) being the second most abundant ionic species

(Fig. 6(a)). Minor ions at m/z 1277.75 and 1333.81 evidenced

low expression of Gb4Cerwith C18:0 andC22:0 fatty acids. In

addition a variation of Gb4Cer (d18:1, C24:1) with a hydroxyl

Table 3. Monosodiated ions of Stx1 receptor Gb3Cer

acquired from the MS1 spectra and proposed structures of

antibody- or Stx1-detected Gb3Cer species from THP-1 cells



1046.65 1046.66 Gb3Cer (d18:1, C16:0)1074.69 1074.68 Gb3Cer (d18:1, C18:0)1130.73 1130.75 Gb3Cer (d18:1, C22:0)1156.75 1156.77 Gb3Cer (d18:1, C24:1)1158.75 1158.78 Gb3Cer (d18:1, C24:0)n.d.a 1172.74 Gb3Cer (d18:1, C24:1),

hydroxylatedn.d.a 1174.76 Gb3Cer (d18:1, C24:0),

hydroxylatedn.d.a 1184.79 Gb3Cer/CD77 (d18:1, C26:1)



Gb3Cer (d18:1, C24:1/C24:0) from THP-1 cells with m/z

1156.77/1158.78 (a) and the corresponding fragmentation

scheme of Gb3Cer (d18:1, C24:0) with the molecular struc-

ture (b). The B/C- and Y/Z-type ions released by cleavage of

glycosidic linkages and the A-type ions and the NII fragment

obtained by internal sugar and ceramide cleavage, respect-

ively, are marked with their corresponding m/z values. The

fragment ions are listed in Table 4.


DOI: 10.1002/rcm


group in the ceramide moiety and Gb4Cer (d18:1, C26:0)

were detected in trace quantities in the MS1 spectrum. The

silica gel extract from the Stx1/anti-Stx1 overlay assay

produced the same profile of molecular ions but with

differing relative abundances (Fig. 6(b)). While [MþNa]þ

ions of Gb4Cer (d18:1, C16:0) dominated in theMS1 spectrum

from the antibody-derived extract, ions of Gb4Cer (d18:1,

C24:1/C24:0) are prevalent in the spectrum deduced

from the Stx1/anti-Stx1-detected Gb4Cer band. This can

be explained by the obvious difference between the overlay

assay patterns in which the antibody-staining features two

bands of equal intensity, whereas Stx1 detects just the upper

Gb4Cer band. Therefore, mainly Gb4Cer speciesmigrating in

or close to the upper band, that is Gb4Cer species containing

long-chain fatty acids, were extracted from the silica gel with

Stx1 as a primary detection reagent. The identified ions from

both overlay assays and the proposed Gb4Cer structures are

summarized in Table 5.

Tandem mass spectrometry of Stx1/anti-Stx1-detected Gb4Cer from THP-1 cellsThe [MþNa]þ ions of Stx1/anti-Stx1-identified Gb4Cer

(d18:1, C16:0) andGb4Cer (d18:1, C24:1/C24:0) were selected

for CID MS. Full structures could be acquired from the MS2

spectra and the tandem mass spectrum obtained from the

molecular ions at m/z 1359.82/1361.83 of Gb4Cer carrying

C24:1/C24:0 fatty acid is shown in Fig. 7(a) along with the

associated fragmentation scheme and the proposed structure

in Fig. 7(b). The fragment ions originating from the sodiated

precursor ions at m/z 1359.82 revealed the entire series of B-

and C-type ions and, with the exception of the Z3 ion, the full

series of Y- and Z-type ions indicating the consecutive loss

of GalNAc, Gal, Gal, and Glc in this order from the non-

reducing end of the selected Gb4Cer species. Ring cleavage-

generated 0,2A3 ions were present as well as NII ions,

whereby the latter indicated sphingosine (4-sphingenine,

d18:1) as the long-chain aminodiol of the ceramide moiety

giving rise to the complete structure of Gb4Cer (d18:1, C24:1).

The collection of detected fragment ions of Gb4Cer (d18:1,

C24:1) and those of Gb4Cer (d18:1, C16:0) achieved from the

precursor ions at m/z 1249.72 (spectrum not shown) are

summarized in Table 6.


attained from MS2 spectra of Stx1-detected major Gb3Cer-

species from THP-1 cells

Fragment ions

Gb3Cer(d18:1, C16:0)m/z 1046.66

Gb3Cer(d18:1, C24:1)m/z 1156.77

m/z values m/z values

0,2A2 305.08 305.110,2A3 467.13 467.16B1; C1 185.03; 203.05 185.05; 203.05B2, C2 347.10; 365.11 347.10; 365.11B3; C3 509.16; 527.16 509.16; 527.18Y0; Z0 560.50; 542.49 670.66; 652.60Y1; Z1 722.58; 704.55 832.70; 814.69Y2; Z2 884.64; 866.59 994.75; 976.71NII 264.29 264.29


(b) Stx1/anti-Stx1-detected Gb4Cer species from THP-1

cells. The spectra were obtained from the silica gel extracts

of TLC overlay assays and recorded in the positive ion mode.

The corresponding overlay assays are shown in the inserts

and the framed dotted rectangles indicate the areas from

which the silica gel was scraped off. A synopsis of the m/z

values of [MþNa]þ ions and the proposed Gb4Cer structures

are provided in Table 5.

Table 5. Monosodiated ions of Stx1 receptor Gb4Cer ident-

ified in the MS1 spectra and proposed structures of antibody-

or Stx1-detected Gb4Cer species from THP-1 cells



1249.72 1249.72 Gb4Cer (d18:1, C16:0)1277.75 1277.74 Gb4Cer (d18:1, C18:0)1333.81 1333.80 Gb4Cer (d18:1, C22:0)1359.83 1359.82 Gb4Cer (d18:1, C24:1)1361.84 1361.83 Gb4Cer (d18:1, C24:0)1375.87 1375.86 Gb4Cer (d18:1, C24:1),

hydroxylated1387.90 1387.90 Gb4Cer (d18:1, C26:1)


DOI: 10.1002/rcm


DISCUSSION

Fatty acid heterogeneity of ceramides due to varying chain

lengths often results in double band patterns (mainly due to

C24 and C16 fatty acids) being typical for many mammalian

GSLs.41 In agreement with our results, Raji cells have been

previously found to exhibit a dominant upper Gb3Cer band

in TLC separation 13 identified by us as Gb3Cer (d18:1,

C24:1/C24:0). In addition we detected Gb3Cer (d18:1, C16:0)

in the lower band of the Gb3Cer doublet. The Gb3Cer MS1

spectra obtained from silica gel extracts of both the antibody-

and Stx1/anti-Stx1-positive bands were very similar regard-

ing the relative ion abundances of the various Gb3Cer species

and revealed identical m/z values of [MþNa]þ ions which

could be assigned to Gb3Cer with varying fatty acids. Their

proposed structures were further confirmed by MS2 analysis

of selected precursor ions. The preference of Gb3Cer with

mainly long-chain C24 fatty acids is an interesting charac-

teristic of the Raji cell line in view of findings of Sandvig and

coworkers who reported that the sensitivity of a human

carcinoma cell line increased simultaneously with the amo-

unt of Gb3Cer species with long-chain fatty acids.27

Additionally, we detected a considerable amount of mono-

unsaturated C24:1 fatty acid, which is likely to augment

binding of Stxs.42 Hydroxylated Gb3Cer species, which

could be detected only in very low quantities, were shown

to enhance binding of Stx1 and Stx2 43 but might not be the

reason for the cytotoxic effects of Stx on Raji cells due to

extremely low amounts. However, preferential expression

of Gb3Cer with long-chain fatty acids accompanied with

monounsaturated and hydroxylated Gb3Cer species in Raji

cells, that are supposed to mediate extended Stx cytotoxicity,

might explain the strong effect of apoptosis induction of Stx

towards Raji cells.6

A typical double band pattern with almost equal pro-

portions of upper and lower bands of Gb3Cer has been

previously reported for THP-1 cells in TLC overlay assay

studies.13,22 This is in accordance with our data concerning

orcinol staining as well as the antibody- and Stx1/anti-Stx-1-

mediated detection of various Gb3Cer species in TLC overlay

assays. Subsequent structural investigation by MS1 and MS2

analysis evidenced Gb3Cer (d18:1, C16:0) andGb3Cer (d18:1,

C24:1/C24:0) as the main variants detected in the GSL

extracts of the Gb3Cer doublets. In addition the spectro-

metric data revealed the presence of the same monounsa-

turated and hydroxylated Gb3Cer-species as determined in

Raji cells, also appearing in very low quantities in THP-1

cells. Minor ions in the MS1 spectra gave evidence for the

presence of Hex2HexNAc-Cer (most likely Lc3Cer) pointing

to the existence of neolacto-series neutral GSLs. Preliminary

identification deduced from the orcinol-stained chromato-

gram suggests neolactotetraosylceramide (nLc4Cer) as a

prominent neutral GSL of THP-1 cells. Thus, myeloid THP-1

cells might synthesize also neolacto-series GSLs which have

to be characterized in further investigations. However, as the

most obvious difference between lymphoid Raji andmyeloid

THP-1 cells we observed the expression of Gb4Cer in THP-1

cells, whereas Raji cells failed to express this elongation

product of Gb3Cer. In analogy to the biosynthesis of the

Gb3Cer precursor, the predominant Gb4Cer variants were

Gb4Cer (d18:1, C16:0) andGb4Cer (d18:1, C24:1/C24:0). Both

the short- and long-chain fatty acid carrying Gb4Cer species

were detectable in the anti-Gb4Cer antibody TLC overlay

assay, whereas only Gb4Cer (d18:1, C24:1/C24:0) was ‘recog-

nized’ under the given conditions using the Stx1/anti-Stx1


Gb4Cer (d18:1, C24:1/C24:0) from THP-1 cells with m/z

1359.82/1361.83 (a) and the corresponding fragmentation

scheme of Gb4Cer (d18:1, C24:0) with the molecular struc-

ture (b). The B/C- and Y/Z-type ions released by cleavage of

glycosidic linkages and the A-type ions and the NII fragment

obtained by internal sugar and ceramide cleavage, respect-

ively, are marked with their corresponding m/z values. The

fragment ions are listed in Table 6.


attained from MS2 spectra of Stx1-detected major Gb4Cer

species from THP-1 cells

Fragment ions

Gb4Cer(d18:1, C16:0)m/z 1249.72

Gb4Cer(d18:1, C24:1)m/z 1359.82

m/z-values m/z-values

0,2A3 n.d.a 508.04Hex 185.03; n.d.a 185.05; 203.04B1; C1 226.07; n.d.a 226.07; n.d.a

Hex2 347.10; 365.11 347.12; 365.11B2; C2 388.14; n.d.a 388.14; 406.15Hex3 509.19; 527.18 n.d.a; 527.18B3; C3 550.19; n.d.a 550.17; n.d.a

B4; C4 712.26; 730.29 712.29; 730.29Y0; Z0 n.d.; n.d.a 670.66; 652.66Y1; Z1 722.61; n.d.a 832.70; 814.69Y2; Z2 884.67; 866.73 994.75; 976.75Y3; Z3 1046.69; n.d.a 1156.79; n.d.a

NII 264.29 264.29



DOI: 10.1002/rcm


approach. Anyway, this data indicate the general potency of

Stx1 to interact not only with Gb3Cer, but also with Gb4Cer

albeit to a lesser extent. However, the involvement and/or

the impact of the structural heterogeneity of Gb3Cer and

Gb4Cer in apoptosis processes of THP-1 cells caused by Stx123,24 remain elusive.

CONCLUSIONS

Sophisticated MS-based technologies provide profound

knowledge about the composition of the oligosaccharide

and the ceramide lipid anchor of GSLs being crucial for

understanding their implications for functional glycosphin-

golipidomics of mammalian cells.31,32 Among numerous

strategies,44 the TLC overlay assay procedure either with

GSL-specific toxins or equivalent antibodies in conjunction

with nanoESI-QTOF-MS1 and subsequent tandem MS2

provides an excellent tool to unravel the structural diversity

of GSLs.41 In particular acyl chain lengths and modifications

of the lipid anchor (such as fatty acid hydroxylation or vari-

ations in the degree of unsaturation) are supposed to influe-

nce not only the molecular interaction of GSLs with binding

partners in lipid rafts,45,46 but also the mechanism of rece-

ptor-mediated endocytosis or membrane invaginations and

the ensuing subcellular route of GSL-toxin or GSL-virus

complexes.47,48 Thus, unravelling the structural features and

the intrinsic cellular heterogeneity is a fundamental pre-

requisite to gaining a deeper understanding of the functional

efficiency of GSLs, which remains so far unknown for

many types of GSL-protein interactions. Knowledge about

the full structural repertoire of a certain type of receptor

might be of benefit to develop novel therapies not only

for human leukaemia,7,49 but also of various types of solid

tumors showing enhanced expression of tumor-associated

GSLs which could in turn serve as targets for cancer

therapies.36,50

AcknowledgementsThis work was supported by grants from the ’Deutsche

Forschungsgemeinschaft’ (DFG)-funded International

Graduate School ’Molecular Interactions of Pathogens with

Biotic and Abiotic Surfaces’ (GRK 1409, collaboration

between the projects 3.10 of J.M. and 3.6 of H.K.), cooperative

projects MU845/4-1 (J.M.) and FR2569/1-1 (A.W.F.), and

Special Collaborative Program ‘Extracellular Matrix: Bio-

genesis, Assembly, and Cellular Interactions’ (SFB 492, pro-

ject Z2, M.M. and J.P.-K.). We thank Martina Bielaszewska

(Institute of Hygiene, University of Munster) for providing

Shiga toxin for the TLC overlay assays and Waltraud Walz-

Schmidt andM. Alexander Schmidt (Institute of Infectiology,

University of Munster, Germany) for stimulating discus-

sions. The laboratory help of Kirsten Tschapalda and Lena

Reimann is gratefully acknowledged.

REFERENCES

1. Smith DC, Lord JM, Roberts LM, Johannes L. Semin. Cell Dev.Biol. 2004; 15: 397.

2. Muthing J, Schweppe CH, Karch H, Friedrich AW. Thromb.Haemost. 2009; 101: 252.

3. Sandvig K. Toxicon 2001; 39: 1629.

4. Endo Y, Tsurugi K, Yutsudo T, Takeda Y, Ogasawara T,Igarashi K. Eur. J. Biochem. 1988; 171: 45.

5. Garred Ø, van Deurs B, Sandvig K. J. Biol. Chem. 1995; 270:10817.

6. Mangeney M, Lingwood CA, Taga S, Caillou B, Tursz T,Wiels J. Cancer Res. 1993; 53: 5314.

7. Mori T, Kiyokawa N, Katagiri YU, Taguchi T, Suzuki T,Sekino T, SatoN, Ohmi K,NakajimaH, Takeda T, Fujimoto J.Exp. Hematol. 2000; 28: 1260.

8. Tetaud C, Falguieres T, Carlier K, Lecluse Y, Garibal J,Coulaud D, Busson P, Steffensen R, Clausen H, JohannesL, Wiels J. J. Biol. Chem. 2003; 278: 45200.

9. LaCasse EC, Saleh MT, Patterson B, Minden MD, Gariepy J.Blood 1996; 88: 1561.

10. Gordon J, Challa A, Levens JM, Gregory CD, Williams JM,Armitage RJ, Cook JP, Roberts LM, Lord JM.Cell Death Differ.2000; 7: 785.

11. Johannes L, Romer W. Nature Rev. Microbiol. 2009; 8: 105.12. Kniep B, Monner DA, Schwulera U, Muhlradt PF. Eur. J.

Biochem. 1985; 149: 187.13. Stults CLM, Larsen RD, Macher B. Glycoconj. J. 1995; 12: 680.14. Griener TP, Mulvey GL, Marcato P, Armstrong GD. J. Med.

Microbiol. 2007; 56: 1423.15. Brigotti M, Carnicelli D, Ravanelli E, Barbieri S, Ricci F,

Bontadini A, Tozzi AE, Scavia G, Caprioli A, Tazzari PL.J. Leukoc. Biol. 2008; 84: 1019.

16. Bielaszewska M, Karch H. Thromb. Haemost. 2005; 94: 312.17. KarchH, Tarr PI, BielaszewskaM. Int. J. Med.Microbiol. 2005;

295: 405.18. Schweppe CH, Bielaszewska M, Pohlentz G, Friedrich AW,

Buntemeyer H, Schmidt MA, Kim KS, Peter-Katalinic J,Karch H, Muthing J. Glycoconj. J. 2008; 25: 291.

19. Brigotti M, Caprioli A, Tozzi AE, Tazzari PL, Ricci F, ConteR, Carnicelli D, Procaccino MA, Minelli F, Ferretti AVS,Paglialonga F, Edefonti A, Rizzoni G. J. Clin. Microbiol.2006; 44: 313.

20. Flagler MJ, Strasser JE, Chalk CL, Weiss AA. Infect. Immun.2007; 75: 760.

21. Geelen JM, van der Velden TJAM, van den Heuvel LPWJ,Monnens LAH. Pediatr. Nephrol. 2007; 22: 1181.

22. Ramkumar S, Sakac D, Binnington B, Branch DR, LingwoodCA. Glycobiology 2009; 19: 76.

23. Lee SY, Lee MS, Cherla RC, Tesh VL. Cell. Microbiol. 2008; 10:770.

24. Lee MS, Cherla RP, Leyva-Illades D, Tesh VL. Infect. Immun.2009; 77: 5233.

25. Meisen I, Friedrich AW, Karch H, Witting U, Peter-KatalinicJ, Muthing J. Rapid Commun. Mass Spectrom. 2005; 19: 3659.

26. Sonnino S, Prinetti A, Nakayama H, Yangida M, Ogawa H,Iwabuchi K. Glycoconj. J. 2009; 26: 615.

27. Sandvig K, Ryd M, Garred Ø, Schweda E, Holm PK, vanDeurs B. J. Cell Biol 1994; 126: 53.

28. Arab S, Lingwood CA. J. Cell. Physiol. 1998; 177: 646.29. Muthing J, Maurer U, Sostaric K, Neumann U, Brandt H,

Duvar S, Peter-Katalinic J, Weber-Schurholz S. J. Biochem.1994; 115: 248.

30. Saito T, Hakomori S. J. Lipid Res. 1971; 12: 257.31. Li Y, Teneberg S, Thapa P, Bendelac A, Levery SB, Zhou D.

Glycobiology 2008; 18: 158.32. Li Y, Zhou D, Xia C, Wang PG, Levery SB. Glycobiology 2008;

18: 166.33. Souady J, Soltwisch J, Dreisewerd K, Haier J, Peter-Katalinic

J, Muthing J. Anal. Chem. 2009; 81: 9481.34. Chester MA. Glycoconj. J. 1999; 16: 1.35. Distler U, Hulsewig M, Souady J, Dreisewerd K, Haier J,

Senninger N, Friedrich AW, Karch H, Hillenkamp F, Ber-kenkamp S, Peter-Katalinic J, Muthing J. Anal. Chem. 2008;80: 1835.

36. Distler U, Souady J, Hulsewig M, Drmic-Hofman I, Haier J,Friedrich AW, Karch H, Senninger N, Dreisewerd K, Ber-kenkamp S, Schmidt MA, Peter-Katalinic J, Muthing J. PLoSONE 2009; 4: e6813.

37. Muthing J. in Methods in Molecular Biology Vol. 76: Glycoana-lysis Protocols, Hounsell EF (ed). Humana Press Inc.: TotawaNJ, 1998; pp 183.

38. Domon B, Costello CE. Glycoconj. J. 1988; 5: 397.39. Domon B, Costello CE. Biochemistry 1988; 27: 1534.40. Adams J, Ann Q. Mass Spectrom. Rev. 1993; 12: 51.41. Muthing J, Distler U. Mass Spectrom. Rev 2010; 29: 425.


DOI: 10.1002/rcm


42. Kiarash A, Boyd B, Lingwood CA. J. Biol. Chem. 1994; 269:11138.

43. Binnington B, Lingwood D, Nutikka A, Lingwood CA.Neurochem. Res. 2002; 27: 807.

44. Levery SB. Methods Enzymol. 2005; 405: 300.45. Jarvis RM, Chamba A, Holder MJ, Challa A, Smith DC,

Hodgkin MN, Lord JM, Gordon J. Biochem. Biophys. Res.Commun. 2007; 355: 944.

46. Gehrmann M, Liebisch G, Schmitz G, Anderson R, SteinemC, De Maio A, Pockley G, Multhoff G. PLoS ONE 2008; 3:e1925.

47. Romer W, Berland L, Chambon V, Gaus K, Windschiegl B,Tenza D, Aly MRE, Fraisier V, Florent JC, Perrais D, Lamaze

C, Raposo G, Steinem C, Sens P, Bassereau P, Johannes L.Nature 2007; 450: 670.

48. Ewers H, Romer W, Smith AE, Bacia K, Dmitrieff S,Chai W, Mancini R, Kartenbeck J, Chambon V, BerlandL, Oppenheim A, Schwarzmann G, Feizi T, Schwille P,Sens P, Helenius A, Johannes L. Nature Cell Biol. 2010;12: 11.

49. Maloney MD, Binnington-Boyd B, Lingwood CA. Glycocon-jugate J. 1999; 16: 821.

50. Distler U, Souady J, Hulsewig M, Drmic-Hofman I, Haier J,Denz A, Grutzmann R, Pilarsky C, Senninger N, DreisewerdK, Berkenkamp S, Schmidt MA, Peter-Katalinic J, Muthing J.Mol. Cancer Ther. 2008; 7: 2464.


DOI: 10.1002/rcm


On the structural diversity of Shiga toxin glycosphingolipid receptors in lymphoid and myeloid cells...

Documents

Transcript of On the structural diversity of Shiga toxin glycosphingolipid receptors in lymphoid and myeloid cells...