| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
J Biol Chem, Vol. 274, Issue 43, 30520-30526, October 22, 1999
-1,2-Linked Mannose Residues*
,,
From the Equipe Mixte de l'INSERM 99-15, Laboratoire
de Mycologie Fondamentale et Appliquée, Faculté de
Médecine, Centre Hospitalier Universitaire, Place de Verdun,
59045 Lille Cedex, France, § Laboratoire de Chimie
Biologique et Unité Mixte 111, CNRS, Université de Lille I,
59655 Villeneuve d'Ascq Cedex, France, and ¶ Zentrum für
Hygiene und Medizinische Mikrobiologie, Philipps-Universität
Marburg, Robert Koch Strasse 17, 35037 Marburg, Germany
 |
ABSTRACT |
|---|
 TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
In a series of studies, we have shown that
Candida albicans synthesizes a glycolipid,
phospholipomannan (PLM), which reacted with antibodies specific for
The yeast Candida albicans is a normal component of the
human endogenous microflora, but it can cause frequent and severe disseminated infections among hospitalized patients (1). Basic progress
has been made in the elucidation of C. albicans
characteristics linked to switching (2), dimorphism (3), adhesion (4), and enzyme secretion (5, 6) that could explain the mechanism by which
this fungus can be an opportunistic pathogen. However, the mechanisms
that direct susceptibility and resistance to C. albicans
infection are as yet unclear, and current extensive research concerns
C. albicans molecules interacting with the host immune system. Among these studies, research has gradually focused on In previous studies, we have shown by use of specific monoclonal
antibodies (16) that Strain, Culture Conditions, and Metabolic
Labelings--
C. albicans VW32 strain (serotype A) was
used throughout this study. For germ tubes preparation, C. albicans was first grown in Sabouraud dextrose agar for 18 h
at 28 °C and then transferred to RPMI 1640 medium (Roche Molecular
Biochemicals) for 4 h at 37 °C with mild shaking. Metabolic
labeling with [3H]mannose and [3H]inositol
was done in 50 ml of RPMI 1640 medium containing either 10-fold less
glucose and 200 µCi of [2-3H]mannose (Isotopchim,
Ganagobie-Peyruis, France) or 10-fold less vitamins and 100 µCi of
[3H]inositol (Isotopchim). The purification of high
amounts of PLM for physicochemical analysis was performed from yeast
cells grown at 28 °C in a bioreactor as described for the
preparation of phosphopeptidomannan (24).
Cell Extracts--
Whole cell extracts, designated as Fr. A,
were obtained following three cycles of French press disruption at a
pressure of 1400 kg/cm2 as described previously (25),
except that the extracts were not centrifuged before extensive dialysis.
Purification of PLM and Release of Its Glycan Moiety--
The
main purification steps are summarized in Fig.
1. Dried French press extracts (Fr. A),
either labeled or not, were subjected to three extractions with
chloroform/methanol (2:1, by volume), three extractions with
chloroform/methanol/water (10:10:1, by volume), and finally to three
extractions with chloroform/methanol/water (10:10:3, by volume;
C/M/W3). The three last extracts containing the PLM were pooled,
filtered on a GF/F membrane (Whatman) and designated as Fr. B. Fr. B
was then evaporated and partitioned at least three times between water
and water-saturated butanol. The water phase was designated as Fr.
C.
Improved purification of unlabeled PLM for physicochemical analyses was
performed from Fr. C using additional chromatographic steps. Fr. C was
dissolved in 0.1 M ammonium sulfate, 1% ethanol and
subjected to hydrophobic interaction chromatography on a
phenyl-Sepharose (Amersham Pharmacia Biotech) column (1 × 10 cm).
The column was successively washed with 1, 5, and 10% ethanol. PLM
elution was obtained with 40% ethanol. The PLM fractions were
concentrated, dissolved in C/M/W3, and subjected to chromatography on a
silica gel 60 column (1.5 × 20 cm) at 4 °C. Components were
first eluted with C/M/W3, and after the recovery of PLM, with
chloroform/methanol/water (2:3:1, by volume). These PLM fractions were
concentrated and frozen until used and were designated as Fr. D.
Fr. D was subjected to alkaline hydrolysis to release the PLM glycan
moiety. 2 mg of purified PLM (Fr. D) estimated by its sugar content
were incubated for 90 h at 37 °C in 1 N KOH, 30% methanol and then neutralized with 1 N acetic acid, dried,
and subjected to hexane extraction. The hydrolysate was then
solubilized in 1% propanol, 0.1 M ammonium acetate and
applied to a octyl-Sepharose (Amersham Pharmacia Biotech) column
(1 × 11 cm). Hydrolyzed PLM was eluted with the same buffer and
desalted on a Bio-gel P2 (Bio-Rad SA, Ivry sur Seine, France) column
(0.8 × 15 cm). This material was designated as Fr. Dg.
Thin Layer Chromatography (TLC) Analysis--
Thin layer
chromatography of unlabeled material was performed primarily using
aluminum-backed silica 60 TLC plates (Merck, Darmstadt, Germany).
Plates were developed at room temperature using
chloroform/methanol/water (2:3:1, by volume) as the solvent system, and
spots were revealed with an orcinol stain. TLC of radiolabeled extracts
were developed on glass-backed silica gel SI 60 plates at room
temperature using C/M/W3 or chloroform/methanol/0.1% KCl (2:3:1, by
volume) as solvent systems. Each lane was then scanned for
radioactivity profile with a Berthold LB 2842 automatic TLC scanner. To
further identify the PLM peaks on TLC, the radioactive areas were
scraped, and the glycolipids were extracted successively with methanol,
10% and C/M/W3 and then subjected to a dot-blot analysis involving the
monoclonal antibody (mAb) DF9-3 (25), a mouse IgM with specificity for
The glycan moiety (Fr. Dg) generated by alkaline hydrolysis of
[3H] inositol-labeled or unlabeled PLM was analyzed on
aluminum-backed silica gel 60 TLC plates developed in butanol/acetic
acid/water (20:8:17, by volume). Methanolysis of the Glycan Moiety of PLM--
Carbohydrate
analysis of Fr. Dg was performed by methanolysis (0.5 M HCl
in anhydrous methanol for 24 h at 80 °C) followed by
N-reacetylation and trimethylsilylation according to the
procedure of Kamerling et al. (26).
Methylation Analysis--
Fr. Dg was submitted to two runs of
permethylation (27). After methanolysis of the permethylated material
and O-acetylation of the resulting free hydroxyl groups, the
acetylated and methylated glycosides were analyzed by GC/MS according
to the method of Fournet et al. (28).
Inositol and Glucosamine Analysis--
Alkaline-treated PLM (Fr.
Dg) was submitted to a strong acid hydrolysis (6 M HCl,
20 h at 110 °C) according to Ferguson (29). After
trimethylsilylation, inositol was analyzed by GC/MS using the chemical
ionization mode with NH3 as the reactant gas. An aliquot of
the hydrolysate was submitted to N-reacetylation and trimethylsilylation for detection of glucosamine.
Nuclear Magnetic Resonance (NMR) Analysis--
The NMR
experiments were performed on a Bruker ASX 400 spectrometer in
D2O at 25 °C. Chemical shifts ( Mass Spectrometry--
Mass measurement was first performed by
matrix-assisted laser desorption and time-of-flight (MALDI-TOF) mass
spectrometry on a Vision 2000 instrument (Finnigan Mat, Hemel) in
reflection mode (nitrogen laser, 337 nm). Fr. Dg was dissolved in water
at a concentration of 50-100 pmol·µl
The electrospray (ES) ionization mass spectrometry was carried out on a
Quatro II triple-quadruple mass spectrometer (Micromass). Fr. Dg was
dissolved in 50% methanol at an approximate concentration of 50 pmol/µl, and the solution was infused into the electrospray ion
source by a Harvard syringe pump. The voltage difference between the
needle tip and the source electrode was TLC Analysis of C. albicans Extracts Showed that PLM Corresponds to
a Family of Glycolipids Labeled with [3H]Mannose and
[3H]Inositol--
Preliminary characterization of
C. albicans PLM was made through chloroform/methanol/water
extraction procedure and Western blot analysis (30). When the same
extraction procedure was applied to [3H]mannose-labeled
Fr. A and analyzed by thin layer chromatography (Fig.
2, a-c), PLM
corresponded to an heterogeneous peak with an average relative
migration (Rf) of 0.147, which was present only in
the more polar extract of the last extraction step (Fr. B) (Fig.
2c). This was confirmed by dot-blot analysis, which revealed
its reactivity with mAb DF9-3 (data not shown). Subsequent
butanol/water partitions of Fr. B resulted in the presence of PLM in
the water phase (Fr. C) (Fig. 2d) in contrast to other peaks
that were separated in the butanol phases (Fig. 2e).
The heterogeneity of the PLM preparation was clearly demonstrated by
thin layer chromatography analysis of [3H]mannose-labeled
Fr. C using 0.1% KCl instead of water in the solvent system (Fig.
3a). Moreover, the profiles of
[3H]mannose- and [3H]inositol-labeled Fr. C
(Fig. 3, a and b) were strikingly superimposed, with the exception of the peak at Rf 0.69, which was observed only with the inositol labeling. Analysis by dot-blot procedure of the reactivity of the peaks with mAb DF9-3 (Fig. 3c) confirmed that peaks 1-5, observed in both profiles,
were related to a family of glycolipids expressing
Alkaline Hydrolysis of Purified PLM Released a Soluble Fraction
Containing Mannose Residues and Inositol--
To obtain large
quantities of PLM for physicochemical analysis, Fr. C was prepared from
C. albicans grown in a bioreactor. Orcinol staining of this
fraction also revealed the main contaminant previously observed with
[3H]inositol labeling (Fig. 3b). Successive
chromatographic purifications of Fr. C on phenyl-Sepharose and silica
gel 60 columns were thus performed and resulted in a PLM preparation
(Fr. D) that produced a single spot on TLC (Fig.
4a, lane 1). Fr. D
displayed a low solubility in most solvent systems, leading to
opalescent solutions. This suggested a micellar conformation of PLM at
a high concentration. This hypothesis was confirmed by the exclusion of
concentrated Fr. D from the Ultrogel AcA34 column (Biosepra,
Cergy-Pontoise, France), for which the exclusion limit was 750 kDa
(data not shown). According to these properties, mass spectrometry and
NMR spectrometry analyses of Fr. D were unsuccessful, and assays were
therefore performed to obtain a soluble derivative from Fr. D. Treatment of Fr. D in 1 N KOH in 30% methanol for 90 h at 37 °C was found to gradually increase its solubility through
the release of a highly water-soluble fraction that no longer
interacted with an octyl-Sepharose column. TLC analysis of this
fraction (Fr. Dg) showed that it displayed an Rf
different from that of Fr. D (Fig. 4a, lanes 2 and 1). Fr. Dg behaved as a highly polymerized glycan when
compared with phosphopeptidomannan-derived PLM Alkali-released Fraction Is Composed Mainly of Linear Chains of
Homo- and heteronuclear two-dimensional NMR spectra of Fr. Dg are shown
in Fig. 5 and summarized in Table
I. The proton resonance region (Fig.
5a) is dominated by two major signals at Mass Spectrometry Revealed that the Heterogeneity of PLM
Alkali-labile Fraction Was Due to Different Chain Lengths of the
Phosphoinositolmannosides--
The MALDI-TOF mass spectrum (Fig.
6a), recorded in the negative
mode, exhibited a series of main peaks with an alternatively spaced
m/z ratio of 80 and 82 Da, which may be clustered
in different families (Table II). The
first family of pseudo-molecular ions with m/z
increments of 162 from 745 to 3499 Da corresponded to molecules
containing n mannoses + 1 inositol + 1 phosphate group, with
n varying from 3 to 19, whereas the second series with the same m/z increments of 162 Da from 663 to 3417 Da
corresponded to molecules containing n mannoses + 1 inositol + 2 phosphate groups, n also varying from 3 to 19. A third
series of minor peaks close to the second one ( Unlike In relation to the pathogenic character of C. albicans,
several groups have investigated the recognition of
In this study we analyzed PLM to obtain chemical evidence for the
presence of The average mass of PLM may be estimated, from the present study, to be
about 4 kDa. This mass is different from the former description of the
PLM as corresponding to a C. albicans 14-18-kDa antigen in
Western blotting (16). By using more reticulated gels (7-20%
acrylamide) and migration conditions favoring the progressive blockage
of the molecules in the gel rather than their migration speed, we
observed that the PLM relative molecular mass upon SDS-PAGE decreased
to 7 kDa (data not shown).
In Fig. 7 we suggest a structural model
for the PLM glycan moiety, based on the first chemical evidence for the
presence of 
-1,2-oligomannosides and was biosynthetically labeled by
[3H]mannose, [3H]palmitic acid, and
[32P]phosphorus. PLM has also been shown to be released
from the C. albicans cell wall and to bind to and stimulate
macrophage cells. In this study, we show by thin layer chromatography
scanning of metabolically radiolabeled extracts that the C. albicans PLM corresponds to a family of mannose and inositol
co-labeled glycolipids. We describe the purification process of the
molecule and the release of its glycan fraction through alkaline
hydrolysis. Analysis of this glycan fraction by radiolabeling and
methylation-methanolysis confirmed the presence of inositol and of
1,2-linked mannose units. NMR studies evidenced linear chains of
-1,2-oligomannose as the major PLM components. Mass spectrometry
analysis revealed that these chains were present in
phosphoinositolmannosides with degrees of polymerization varying from 8 to 18 sugar residues. The PLM appears as a new type of eukaryotic
inositol-tagged glycolipid in relationship to both the absence of
glucosamine and the organization of its glycan chains. This first
structural evidence for the presence of -1,2-oligomannosides in a
glycoconjugate other than the C. albicans
phosphopeptidomannan may have some pathophysiological relevance
to the adhesive, protective epitope, and signaling properties thus far
established for these residues.
INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
-1,2-linked oligomannosides. Oligomannnosides with this unusual type
of linkage were first described by Shibata et al. (7) as
associated with the C. albicans cell wall
phosphopeptidomannan by phosphodiester bridges. -1,2-oligomannosides
are immunogenic and elicit specific antibodies in animals (8-10) and
humans (11). Anti--1,2-oligomannosides antibodies have been shown to
be protective against C. albicans in rodent models of
systemic and vaginal candidosis (5, 12). -1,2-oligomannosides
derived from C. albicans phosphopeptidomannan have also been
shown to induce TNF1-
synthesis from cells of the macrophage lineage through a
phosphotyrosine kinase-dependent pathway (13) and to bind
to macrophage cell membranes (14, 15).
-1,2-oligomannosides are present (in the
absence of accessible -linked mannose residues) on a polydispersed
low molecular weight antigen and that this antigen is a glycolipid.
This glycolipid has been named a phospholipomannan (PLM) on the basis
of its composition (17). The PLM is a strong TNF- inducer in
vitro and in vivo (18). When C. albicans
comes into contact with macrophages, large amounts of PLM are rapidly shed by C. albicans, which trigger intense signaling and
secretory responses from these target cells (19). Similar signaling
events induced in host cells have been described as induced by
GPI-related glycolipids from pathogens of the genera
Leishmania, Trypanosoma, and
Mycobacteria (20-23). In this study, we have further
purified and chemically analyzed the C. albicans PLM to
establish the relationship of PLM with these microbiolglycolipids and
to provide a structural basis for the understanding of the
immunochemical and immunomodulatory properties of PLM.
MATERIALS AND METHODS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
View larger version (29K):
[in a new window]
Fig. 1.
Purification scheme of C. albicans phospholipomannan. Fr. A-D refer to
PLM-containing fractions with improved purification. Fr. Dg
corresponds to the glycan moiety of PLM released by alkaline hydrolysis
of Fr. D.
-1,2-linked oligomannosides (17). This mAb was kindly provided by
Dr. M. Borg-von-Zepelin (Göttingen, Germany).
-1,2-Linked oligomannosides (DP 13 and 14) prepared from the phosphopeptidomannan of the same C. albicans strain (24) were used as controls. Plates were then
analyzed by either fluorography or orcinol staining.
) were referenced to internal acetone ( = 2.225 ppm under experimental conditions). Two-dimensional homonuclear COSY (correlation spectroscopy) and HMQC
(heteronuclear multiple quantum correlation) experiments were performed
by using standard Bruker pulse programs.
1, and then 1 µl was mixed with 1 µl of matrix (3-aminochinoline in 1 mM CH3COONH4+) and
allowed to crystallize onto the target at room temperature.
3.2 kV.
RESULTS
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
View larger version (13K):
[in a new window]
Fig. 2.
Thin layer chromatography analysis of
C. albicans extracts metabolically labeled with
[3H]mannose. A [3H]Man-labeled French
press extract of C. albicans VW32 was successively extracted
with chloroform/methanol (2:1, by volume) (panel a),
chloroform/methanol/water (10:10:1, by volume) (panel b) and
chloroform/methanol/water (10:10:3, by volume) (Fr. B) (panel
c). The last extract was then partitioned between water (Fr. C)
(panel d) and water-saturated butanol (panel e).
Solvent system, C/M/W3. O, origin; F, solvent
front.
-1,2-oligomannosidic epitopes. Peaks 6 and 7 did not react with the
antibody. Moreover, these two peaks displayed properties quite
different from the others when tested on octyl-Sepharose column (data
not shown) and thus appeared unrelated to PLM.
View larger version (43K):
[in a new window]
Fig. 3.
Comparative TLC analysis of
[3H]mannose-labeled (panel a) and
[3H]inositol-labeled (panel b) Fr. C of C. albicans. Fr. C were obtained as
described in the legend to Fig. 1 from metabolically labeled cells and
were analyzed on the same TLC plate. Following radioactivity scanning
on a Berthold scanner, areas 1-7 of the silica gel corresponding to
the main peaks were scraped off the plates, extracted with solvents and
their reactivity to the mAb DF9-3 specific for -1,2-oligomannosides
was analyzed by a dot-blot procedure (panel c). Solvent
system, chloroform/methanol/KCl 0.1% (10:15:5, by volume).
O, origin; F, front.
-1,2-oligomannosides, with DP = 13 and 14 (Fig. 4a, lanes 4 and
3); these oligomannosides were used as controls in relation
to the expected glycan structure of Fr. Dg. Improved TLC analysis
following five consecutive migration runs revealed a co-migration of
this unlabeled Fr. Dg (Fig. 4b, lane 3) with a
Fr. Dg (Fig. 4b, lane 1) prepared from PLM of
C. albicans metabolically labeled with
[3H]inositol (Fig. 4b, lane 2).
These experiments, which showed the radiolabeling still present in Fr.
Dg, suggested that hydrolysis had occurred between inositol and the PLM
hydrophobic moiety.
View larger version (109K):
[in a new window]
Fig. 4.
TLC analysis of the alkali-released fraction
of PLM (Fr. Dg). In panel a, Fr. Dg
(lane 2) obtained following alkaline hydrolysis of PLM was
compared with Fr. D (unhydrolysed PLM) (lane 1) and
-1,2-linked oligomannosides released from C. albicans
phosphopeptidomannan with DP = 13 (lane 4) and 14 (lane 3) used as controls. In panel b,
radiolabeled Fr. Dg (lane 1) was prepared from C. albicans metabolically labeled with [3H]inositol and
compared with the unlabeled Fr. Dg (lane 3) used for the
physicochemical analyses. Unhydrolysed radiolabeled PLM (Fr. D)
(lane 2) and -1,2-linked oligomannoside with a DP = 14 (lane 4) were used as controls. All lanes were
revealed by an orcinol stain with the exception of lanes 1 and 2 of panel b, which were revealed by
fluorography. Solvent system, butanol/acetic acid/water (20/8/17, by
volume). Panels a and b correspond, respectively,
to one and five runs of migration in this solvent system.
-1,2-Linked Mannose Residues--
Gas chromatography analysis of
Fr. Dg subjected to either methanolysis or strong acid hydrolysis
revealed mannose as the major component and confirmed the presence of
inositol (less than 10% of the total material). Glucosamine was never
observed even in trace amounts (data not shown). Methylation analysis
of fraction Dg led to the characterization of
2-O-Ac,3,4,6-O-Me-Man and
2,3,4,6-O-Me-Man at an approximate ratio of 15 to 1. This
observation allowed us to define the glycan as a linear homopolymer of
1,2-linked mannose units.
5.046 and 4.384 ppm, which were assigned, respectively, to the H-1 and H-2 atom
resonances of -mannose. Three other resonances at 4.908, 4.934, and 5.05 ppm also possessed the characteristics of -mannose anomeric
protons. The set of these H-1 and H-2 resonances was identical to the
NMR data obtained for the -1,2-Man homopolymers released from
C. albicans phosphopeptidomannan by mild acid hydrolysis (24) and allowed us to ascertain the mannose units on the spectrum from
1 to n. However, the absence of reducing terminal -Man
can be notified in Fr. Dg and results in the modification of H-1 and H-2 resonances of Man 1 to Man 3. The integration value of the main
peak at 5.046, which included the additive contributions of Man 3 to Man n-1, also revealed a higher degree of polymerization of the -1,2-oligomannosides from PLM with an average DP of 14 mannose residues. Signals relative to inositol and glycerol units were
not detected, and the two minor atom resonances at 5.557 and 5.140 ppm, respectively, seemed to possess the characteristics of the
anomeric protons of - or -Man-1-phosphate
(3JH,P 
7.5), but these correlations remain
to be verified. Moreover, the heteronuclear NMR spectrum (Fig.
5b) clearly indicates that all C-2 atoms resonate at 80.6-80.9 ppm, with the exception of the C-2 atom of the mannose unit
n ( 71.66 ppm) in the terminal nonreducing position, and
definitively confirms the presence of a linear chain of -1,2-linked
mannose residues in PLM.
View larger version (45K):
[in a new window]
Fig. 5.
NMR spectra of Fr. Dg. Fr. Dg
was dissolved in D20 (99.95%) and analyzed in a 400-MHz
spectrometer in which acetone was used as the internal standard. The
H-1 and H-2 signals of the proton-NMR spectrum (panel a) and the heteronuclear two-dimensional spectrum
(panel b) evidenced linear chains of -1,2-linked mannose
residues.
Chemical shifts () of protons and carbons of the -1,2-linked
oligomannosides of Fr. Dg
m = +14) was also observed, but at the moment their masses did not fit to
any reasonable molar composition. This analysis revealed the
distribution and high degrees of polymerization of -Man residues
from PLM. These DP mainly comprised between 8 and 18 sugar units
because phosphoinositolmannosides with a mass lower than 1200 Da seemed
to be overestimated by this method according to the TLC analysis of the
chloroform/methanol/water extracts (Figs. 2 and 3) and NMR results. It
also evidenced the heterogeneity of PLM. These observations were
confirmed by ES mass spectrometry (Fig. 6b and Table
III), which indicated, in addition to the
previous phosphoinositolmannosides, the presence of a family of minor
molecules derived from the second family through the formation of a
presumed cyclic phosphodiester bridge during the alkaline hydrolysis
(m = 18).
View larger version (42K):
[in a new window]
Fig. 6.
MALDI-TOF and ES mass spectra of Fr. Dg generated by alkaline hydrolysis of PLM. The MALDI-TOF
spectrum (panel a) revealed three families of peaks (i.e
m/z 2527, 2607, and 2621). The
m/z increment between two consecutive peaks of
the same family is always 162. The ES mass spectrum (panel
b) contains five families of peaks (i.e m/z
1303, 1310, 1314, 1345, and 1375) with, for two consecutive peaks of
the same family, a m/z increment of 81 resulting
from the formation of double-charged pseudo-molecular ions and
consequently corresponding to an effective mass increment of 162 (see
Table III).
Examples of molar ratios deduced from pseudo-molecular ions detected by
negative ion MALDI-TOF mass spectrometry of Fr. Dg
Examples of molar ratios deduced from pseudo-molecular ions detected by
negative ion ES-mass spectrometry of Fr. Dg
DISCUSSION
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
-mannosides, which are widely expressed on
glycoconjugates, the rarity of -mannosides may be explained by a
less favorable stereochemistry. To date, homopolymers of -1,2-linked mannose have been chemically characterized only in imperfect yeasts of
the genus Candida (related to Ascomycetes). They are
associated to side chains of the cell wall phosphopeptidomannan of
C. albicans and C. tropicalis through
phosphodiester bridges (7, 31). NMR analysis of these residues released
from C. albicans phosphopeptidomannan by mild acid
hydrolysis has also shown changes in their ratios and degrees of
polymerization, depending on the strains (7, 31), the cell form (32,
33), and the growth conditions (34, 35).
-1,2-oligomannosides by immune systems. These molecules have been
shown to elicit specific antibodies in mice (36), rats (37), rabbits
(9), and humans (11, 38). The construction of neoglycolipids with
phosphopeptidomannan-released -1,2-oligomannosides has demonstrated
that they can act as epitopes for a large number of anti-C.
albicans phosphopeptidomannan monoclonal antibodies (16),
suggesting that C. albicans mannoglycoconjugate(s) expressing these residues are strong immunogens. Two
anti--1,2-oligomannosides monoclonal antibodies have been described
as protective against experimental C. albicans infection.
The first one, reacting with a -1,2-linked mannotriose, protected
mice in a systemic model of candidosis (12). The second one protected
rats in a model of vaginal infection (5). When we analyzed C. albicans molecules expressing -1,2-oligomannosidic epitopes, we
observed that all polyclonal or monoclonal antibodies specific for
these residues bound to a 14-18-kDa antigen that did not display
accessible -linked mannose residues (16). This antigen, named
phopholipomannan (17), is expressed only in C. albicans and
C. tropicalis, which are the most pathogenic
Candida species (25). PLM is synthesized by C. albicans including under growth conditions that prevent association of -1,2-linked oligomannosides to phosphopeptidomannan (30). Recent experiments have shown that C. albicans, in
contact with macrophages, shed large amounts of PLM, which triggers an intense phosphotyrosine kinase-dependent signaling pathway
and the secretion of TNF- (19). The synthesis of surface glycolipids that protect from host defenses and/or disturb host cell immune functions is recognized as a pathogenic characteristic of eukaryotic protozoa of the genera Leishmania and Trypanosoma
(20) and of prokaryotes of the genus Mycobacteria
(21-23).
-1,2-oligomannosides and to assess the possible structural relationships with surface glycolipids of these other microbial pathogens. Like these glycolipids, the C. albicans
PLM was metabolically labeled by [3H]mannose and
[3H]inositol, and both labeling profiles were
superimposed in a family of hydrophilic glycolipids with
-1,2-oligomannosidic epitopes. The physicochemical analysis of the
PLM sugar moiety confirmed the presence of mannose and inositol and
evidenced the absence of glucosamine. The absence of this residue is
consistent with the PLM resistance to nitrous acid treatment (29) and
its unlabeling with [3H]glucosamine (unpublished data).
Glucosamine linking inositol to the sugar moiety is a common feature of
GPI and GPI-related glycolipids (39) of eukaryotic cells. Its absence
has only been reported to date in lipoarabinomannan, a GPI-like
structure from prokaryotes of the genus Mycobacteria.
Another peculiarity of PLM lies in the exclusive presence of
-1,2-linked mannose residues in its sugar moiety, which were found
to be organized in linear chains with degrees of polymerization ranging
from 8 to 18. Confirmation of the probable presence of a
Man-1-phosphate linkage in the molecule, as deduced from NMR spectrum,
will require further studies.
-1,2-linked oligomannosides in a glycoconjugate other
than the yeast phosphopeptidomannan. Very little is known about
-1,2-mannosyltransferases of C. albicans, their
activation and substrate specificity, but the presence of such linear
chains of up to 18 mannose residues represent quite unusual structures
(24). It has been suggested that a consequence of coating parasite
surfaces with long sugar chains is the triggering of host effector
mechanisms at a distance too great for efficient antimicrobial activity
on the parasite. The recent demonstration for the presence of PLM at
the C. albicans cell wall surface (19) suggests that these
mechanisms may play a role during host-C. albicans
interaction. Moreover, -1,2-oligomannosides have been shown to act
as C. albicans adhesins for the macrophage membrane (15) and
to stimulate macrophages to produce high levels of TNF- (13, 18).
The stimulating activity of -1,2-oligomannosides was found to depend
on the length of the mannosyl chain and maximum activity was observed
for DPs of 8 or higher (13). Interestingly, these high DPs are present
mainly in the C. albicans PLM but correspond to minor
components among the -1,2-oligomannosides released from the mannan
of the same species (24).
View larger version (7K):
[in a new window]
Fig. 7.
Proposed structure for the family of
glycolipids found in PLM of C. albicans VW32, serotype
A. The structure of the glycan moiety was deduced from our
results, and its linkage to the lipid moiety was postulated both from
our results and from the usual structures of inositol and phosphorus
containing glycolipids; n may vary from 5 to 15. (a), the position of this branch in the molecule and the
percentage of molecules displaying this branch are still unknown.
In conclusion, we have shown that the pathogenic yeast C. albicans synthesizes inositol-labeled glycolipids that have glycan moieties devoid of glucosamine. These C. albicans
glycolipids are thus structurally more similar to lipoarabinomannans of
Mycobacteria than to the glycosylinositolphospholipids of
parasitic protozoa or the lipophosphoglycan of Leishmania.
Recently, as well as being B cell antigens, mannose sequences of
lipoarabinomannan have been implicated in the presentation to T cells
by CD1b nonclassical major histocompatibility complex molecules (40).
Whether or not this property is shared by C. albicans PLM
remains to be investigated. An important PLM structural peculiarity
lies in the presence of long chains of -1,2-linked mannose residues.
There is now considerable experimental evidence that these sugar
residues are involved in virulence and immunomodulation and can elicit
protective antibody responses. Therefore, PLMs are molecules that must
be considered for a comprehensive analysis of host-C.
albicans relationships. A complete elucidation of their structure
and biosynthetic pathways will be necessary to provide a structural
basis for understanding their immunochemical properties and some
aspects of the pathogenesis of C. albicans infections.
| |
ACKNOWLEDGEMENTS |
|---|
We gratefully acknowledge Dr. Margarete Borg-Von-Zepelin (University of Göttingen, Germany) for providing monoclonal antibody DF9-3. We thank Pr. Casadevall (Albert Einstein College of Medecine, Bronx, NY) for helpful improvements and correction of the manuscript and Pr. Ferguson (University of Dundee, Scotland) for constructive criticism.
| |
FOOTNOTES |
|---|
* The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
To whom correspondence should be addressed. Tel.:
33-3-20472629; Fax: 33-3-20472625; E-mail:
dan_poulain@compuserve.com.
| |
ABBREVIATIONS |
|---|
The abbreviations used are: TNF, tumor necrosis factor; PLM, phospholipomannan; TLC, thin layer chromatography; GC/MS, gas chromatography/mass spectrometry; MALDI-TOF, matrix-assisted laser desorption/ionization time-of-flight; mAb, monoclonal antibody; GPI, glycosylphosphatidylinositol; DP, degrees of polymerization; Fr., fraction.
| |
REFERENCES |
|---|
|
TOP
ABSTRACT
INTRODUCTION
MATERIALS AND METHODS
RESULTS
DISCUSSION
REFERENCES
|
|---|
| 1. | Pfaller, M. A., Jones, R. N., Messer, S. A., Edmond, M. B., and Wenzel, R. P. (1998) Diagn. Microbiol. Infect. Dis. 31, 327-332[CrossRef][Medline] [Order article via Infotrieve] |
| 2. | Soll, D. R. (1997) Microbiology 143, 279-288[Medline] [Order article via Infotrieve] |
| 3. | Leberer, E., Ziegelbauer, K., Schmidt, A., Harcus, D., Dignard, D., Ash, J., Johnson, L., and Thomas, D. Y. (1997) Curr. Biol. 7, 539-546[CrossRef][Medline] [Order article via Infotrieve] |
| 4. |
Gale, C. A.,
Bendel, C. M.,
McClellan, M.,
Hauser, M.,
Becker, J. M.,
Berman, J.,
and Hostetter, M. K.
(1998)
Science
279,
1355-1358 |
| 5. | De Bernardis, F., Boccanera, M., Adriani, D., Spreghini, E., Santoni, G., and Cassone, A. (1997) Infect. Immun. 65, 3399-3405[Abstract] |
| 6. | Ibrahim, A. S., Mirbod, F., Filler, S. G., Banno, Y., Cole, G. T., Kitajima, Y., Edwards, J. E., Jr., Nozawa, Y., and Ghannoum, M. A. (1995) Infect. Immun. 63, 1993-1998[Abstract] |
| 7. | Shibata, N., Ichikawa, T., Tojo, M., Takahashi, M., Ito, N., Ohkubo, Y., and Suzuki, S. (1985) Arch. Biochem. Biophys. 243, 338-348[CrossRef][Medline] [Order article via Infotrieve] |
| 8. |
Faille, C.,
Michalski, J.,
Strecker, G.,
Mackenzie, D. W.,
Camus, D.,
and Poulain, D.
(1990)
Infect. Immun.
58,
3537-3544 |
| 9. |
Shibata, N.,
Arai, M.,
Haga, E.,
Kikuchi, T.,
Najima, M.,
Satoh, T.,
Kobayashi, H.,
and Suzuki, S.
(1992)
Infect. Immun.
60,
4100-4110 |
| 10. |
Tojo, M.,
Shibata, N.,
Kobayashi, M.,
Mikami, T.,
Suzuki, M.,
and Suzuki, S.
(1988)
Clin. Chem.
34,
539-543 |
| 11. |
Poulain, D.,
Faille, C.,
Delannoy, C.,
Jacquinot, P. M.,
Trinel, P. A.,
and Camus, D.
(1993)
Infect. Immun.
61,
1164-1166 |
| 12. | Caesar-TonThat, T.-C., and Cutler, J. E. (1997) Infect. Immun. 65, 5354-5357[Abstract] |
| 13. | Jouault, T., Lepage, G., Bernigaud, A., Trinel, P. A., Fradin, C., Wieruszeski, J. M., Strecker, G., and Poulain, D. (1995) Infect. Immun. 63, 2378-2381[Abstract] |
| 14. |
Li, R. K.,
and Cutler, J. E.
(1993)
J. Biol. Chem.
268,
18293-18299 |
| 15. |
Fradin, C.,
Jouault, T.,
Mallet, A.,
Mallet, J.,
Camus, D.,
Sina , P.,
and Poulain, D.
(1996)
J. Leukocyte Biol.
60,
81-87[Abstract]
|
| 16. |
Trinel, P. A.,
Faille, C.,
Jacquinot, P. M.,
Cailliez, J. C.,
and Poulain, D.
(1992)
Infect. Immun.
60,
3845-3851 |
| 17. |
Trinel, P. A.,
Borg-von-Zepelin, M.,
Lepage, G.,
Jouault, T.,
Mackenzie, D.,
and Poulain, D.
(1993)
Infect. Immun.
61,
4398-4405 |
| 18. | Jouault, T., Bernigaud, A., Lepage, G., Trinel, P., and Poulain, D. (1994) Immunology 83, 268-273[Medline] [Order article via Infotrieve] |
| 19. | Jouault, T., Fradin, C., Trinel, P. A., Bernigaud, A., and Poulain, D. (1998) J. Infect. Dis. 178, 792-802[Medline] [Order article via Infotrieve] |
| 20. | Schofield, L., and Tachado, S. (1996) Immunol. Cell Biol. 74, 555-563[Medline] [Order article via Infotrieve] |
| 21. | Barnes, P. F., Chatterjee, D., Abrams, J. S., Lu, S., Wang, E., Yamamura, M., Brennan, P. J., and Modlin, R. L. (1992) J. Immunol. 149, 541-547[Abstract] |
| 22. | Dahl, K. E., Shiratsuchi, H., Hamilton, B. D., Ellner, J. J., and Toossi, Z. (1996) Infect. Immun. 64, 399-405[Abstract] |
| 23. | Yoshida, A., and Koide, Y. (1997) Infect. Immun. 65, 1953-1955[Abstract] |
| 24. | Trinel, P. A., Lepage, G., Jouault, T., Strecker, G., and Poulain, D. (1997) FEBS Lett. 416, 203-206[CrossRef][Medline] [Order article via Infotrieve] |
| 25. | Cantelli, C., Trinel, P. A., Bernigaud, A., Jouault, T., Polonelli, L., and Poulain, D. (1995) Microbiology 141, 2693-2697[Abstract] |
| 26. | Kamerling, J. P., Gerurg, G. J., Vliegenthart, J. F. G., and Clamp, J. R. (1975) Biochem. J. 151, 491-495[Medline] [Order article via Infotrieve] |
| 27. | Ciucanu, I., and Kerek, F. (1984) Carbohydr. Res. 131, 209-217[CrossRef] |
| 28. | Fournet, B., Strecker, G., Leroy, Y., and Montreuil, J. (1981) Anal. Biochem. 116, 489-502[CrossRef][Medline] [Order article via Infotrieve] |
| 29. | Ferguson, M. A. J. (1993) in Glycobiology: A Practical Approach (Fukuda, M. , and Kobata, A., eds) , pp. 349-384, IRL Press, Oxford, UK |
| 30. | Trinel, P., Cantelli, C., Bernigaud, A., Jouault, T., and Poulain, D. (1996) Microbiology 142, 2263-2270[Abstract] |
| 31. |
Kobayashi, H.,
Matsuda, K.,
Ikeda, T.,
Suzuki, M.,
Takahashi, S. H.,
Suzuki, A.,
Shibata, N.,
and Suzuki, S.
(1994)
Infect. Immun.
62,
615-622 |
| 32. | Shibata, N., Kobayashi, H., Tojo, M., and Suzuki, S. (1986) Arch. Biochem. Biophys. 251, 697-708[CrossRef][Medline] [Order article via Infotrieve] |
| 33. | Shibata, N., Fukasawa, S., Kobayashi, H., Tojo, M., Yonezu, T., Ambo, A., Ohkubo, Y., and Suzuki, S. (1989) Carbohydr. Res. 187, 239-253[CrossRef][Medline] [Order article via Infotrieve] |
| 34. | Kobayashi, H., Giummelly, P., Takahashi, S., Ishida, M., Sato, J., Takaku, M., Nishidate, Y., Shibata, N., Okawa, Y., and Suzuki, S. (1991) Biochem. Biophys. Res. Commun. 5, 1003-1009 |
| 35. | Okawa, Y., Takahata, T., Kawamata, M., Miyauchi, M., Shibata, N., Suzuki, A., Kobayashi, H., and Suzuki, S. (1994) FEBS Lett. 45, 167-171 |
| 36. | Cassone, A., Torosantucci, A., Boccanera, M., Pellegrini, G., C., P., and Malvasi, F. (1988) J. Med. Microbiol. 27, 233-238[Abstract] |
| 37. |
Hopwood, V.,
Poulain, D.,
Fortier, B.,
Evans, G.,
and Vernes, A.
(1986)
Infect. Immun.
54,
222-227 |
| 38. | Hernando, F., Cailliez, J. C., Trinel, P. A., Faille, C., Mackenzie, D., and Poulain, D. (1993) J. Med. Vet. Mycol. 31, 219-226[Medline] [Order article via Infotrieve] |
| 39. | McConville, M. J., and Ferguson, M. A. J. (1993) Biochem. J. 294, 305-324 |
| 40. | Prigozy, T. I., Seiling, P. A., Clemens, D., Stewart, P. L., Behar, S. M., Porcelli, S. A., Brenner, M. B., Modlin, R. L., and Kronenberg, M. (1997) Immunity 6, 187-197[CrossRef][Medline] [Order article via Infotrieve] |
This article has been cited by other articles:
|
|
 |
|
  B. Sendid, T. Jouault, R. Coudriau, D. Camus, F. Odds, M. Tabouret, and D. Poulain Increased Sensitivity of Mannanemia Detection Tests by Joint Detection of {alpha}- and {beta}-Linked Oligomannosides during Experimental and Human Systemic Candidiasis J. Clin. Microbiol., January 1, 2004; 42(1): 164 - 171. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. H. Jin and A. Seyfang High-affinity myo-inositol transport in Candida albicans: substrate specificity and pharmacology Microbiology, December 1, 2003; 149(12): 3371 - 3381. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Ibata-Ombetta, T. Idziorek, P.-A. Trinel, D. Poulain, and T. Jouault Candida albicans Phospholipomannan Promotes Survival of Phagocytosed Yeasts through Modulation of Bad Phosphorylation and Macrophage Apoptosis J. Biol. Chem., April 4, 2003; 278(15): 13086 - 13093. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Dromer, R. Chevalier, B. Sendid, L. Improvisi, T. Jouault, R. Robert, J. M. Mallet, and D. Poulain Synthetic Analogues of {beta}-1,2 Oligomannosides Prevent Intestinal Colonization by the Pathogenic Yeast Candida albicans Antimicrob. Agents Chemother., December 1, 2002; 46(12): 3869 - 3876. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P.-A. Trinel, E. Maes, J.-P. Zanetta, F. Delplace, B. Coddeville, T. Jouault, G. Strecker, and D. Poulain Candida albicans Phospholipomannan, a New Member of the Fungal Mannose Inositol Phosphoceramide Family J. Biol. Chem., September 27, 2002; 277(40): 37260 - 37271. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. A. Trinel, T. Jouault, J. E. Cutler, and D. Poulain {beta}-1,2-Mannosylation of Candida albicans Mannoproteins and Glycolipids Differs with Growth Temperature and Serotype Infect. Immun., September 1, 2002; 70(9): 5274 - 5278. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Poulain, C. Slomianny, T. Jouault, J. M. Gomez, and P. A. Trinel Contribution of Phospholipomannan to the Surface Expression of {beta}-1,2-Oligomannosides in Candida albicans and Its Presence in Cell Wall Extracts Infect. Immun., August 1, 2002; 70(8): 4323 - 4328. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. B. Herrero, D. Uccelletti, C. B. Hirschberg, A. Dominguez, and C. Abeijon The Golgi GDPase of the Fungal Pathogen Candida albicans Affects Morphogenesis, Glycosylation, and Cell Wall Properties Eukaryot. Cell, June 1, 2002; 1(3): 420 - 431. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Nitz, C.-C. Ling, A. Otter, J. E. Cutler, and D. R. Bundle The Unique Solution Structure and Immunochemistry of the Candida albicansbeta -1,2-Mannopyranan Cell Wall Antigens J. Biol. Chem., January 25, 2002; 277(5): 3440 - 3446. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
