|
|
||||||||





*
Division Infectious Diseases and AIDS, Department of Medicine, and
Eijkman-Winkler Laboratory, University Medical Center, Utrecht, The Netherlands; and
Department of Chemistry, Georgia State University, Atlanta, GA 30303
| Abstract |
|---|
|
|
|---|
| Introduction |
|---|
|
|
|---|
Disseminated cryptococcosis is characterized by the presence of high
levels of capsular glucuronoxylomannan (GXM), galactoxylomannan
(GalXM), mannoprotein (MP)-1, and MP-2 in the CSpF and serum of
affected patients (7, 8). Titers in both serum and CSpF
from AIDS patients reach levels equivalent to several hundred
micrograms per milliliter, in exceptional cases rising to 20 mg/ml
(9). GXM, GalXM, and MP have been shown to induce the
production of the early proinflammatory cytokines TNF-
and IL-1
by peripheral blood monocytes (10, 11, 12) and PMN
(13) in the presence of pooled human serum (PHS). TNF-
then plays a pivotal role (14) in initiating a protective
cell-mediated immune response. In healthy subjects, TNF-
enables the
host to overcome the antiphagocytic properties of the fungus, which
usually suffices to clear systemic infection (15). In
patients in whom this accurate response is hampered, the disease
disseminates and induces inflammation. This reaction theoretically
could be a very purulent one, because the induced TNF-
has several
properties that strongly promote PMN extravasation. First, TNF-
can
cause profound systemic vasodilatation and subsequent hypotension via
the induction of PGs (16). Hypotension is a prerequisite
to allow the transient L-selectin (CD62L)-mediated adherence of
leukocytes to endothelial cells (17, 18, 19). In addition,
TNF-
and IL-1
can induce the expression of adhesion molecules on
endothelial and glial cells (4). Furthermore, TNF-
and
IL-1
activate mononuclear cells, endothelial cells, and astrocytes
to produce IL-8 (20) in the brain of patients with
meningitis.
Despite the elevated IL-8 CSpF-serum ratio (13, 21) and
the elevated serum levels of TNF-
and IL-1
, the CSpF of patients
with cryptococcal meningitis typically contains few mononuclear cells
and virtually no PMN. To date, GXM is considered to account for this
inhibition of leukocytosis. In a mouse model study, Dong and Murphy
(22) showed that the inhibition of PMN infiltration
elicited by i.v. administration of the cryptococcal culture filtrate
(CneF) is due mainly to GXM. Using a modified Boyden chamber, we
previously demonstrated that GXM significantly inhibited PMN migration
toward IL-8 (23). In addition, we confirmed Dong and
Murphys earlier observation that GXM mediates the chemotactic
activity for PMN of whole encapsulated yeast cells and unfractionated
CneF derived from these cells (24). Moreover, we found a
significant inverse correlation between the GXM ratio (serum-CSpF) and
the CSpF leukocyte count in patients with cryptococcosis
(25). More recently, we showed that GXM delays
translocation of PMN across the blood-brain barrier in a rabbit BM
model (26). Therefore, the initial aim of this study was
to investigate the molecular mechanism by which GXM prevents PMN
migration toward chemoattractants.
Several studies have investigated the mechanisms regulating the passage
of leukocytes through endothelial cells and their infiltration at
inflammatory foci. The exact nature of the signaling mechanisms in
brain inflammation still remains to be elucidated, but undoubtedly
involves TNF-
and IL-1
, chemoattractants (IL-8,
platelet-activating factor (PAF)), as well as the expression of
adhesion molecules and proteases that together promote cell recruitment
and vascular permeability (6). Initially, localized
inflammation results in hypotension allowing selectin-mediated
tethering and rolling of PMN along the vessel wall (17, 18). Rolling precedes a further functional up-regulation of PMN
following exposure to proinflammatory cytokines and chemoattractants
(27), resulting in firmer integrin-mediated adherence and
shedding of CD62L (17, 27). Further stimulation of PMN
phenotypically high in CD11b/CD18 (CR3, Mac-1) initiates migration
through the interendothelial junctions (28). Finally,
chemotactic gradients guide PMN to the site of infection.
Currently, the discrepancy between raised cytokine levels and hampered
PMN influx into the brains of patients is explained by the observation
that GXM is able to induce shedding of CD62L and TNF-
R (TNF-R)
molecules from PMN (29). The key observation underlying
this study, i.e., a quantitatively similar prevention of PMN migration
by CneF harvested from GXM-producing and nonproducing (
CneF)
strains, seriously questions the opinion that GXM is the sole
cryptococcal Ag preventing extravasation. Therefore, we set out to
characterize the component of
CneF responsible for the inhibition.
To assess this, we first optimized the current protocol for isolating
capsular Ags and purified the most important constituents of
CneF
(GalXM, MP-1, MP-2, and MP-4) to near homogeneity (30). We
found that MP-4 was primarily responsible for the inhibition of PMN
migration. We then investigated the intrinsic chemotactic capacity of
MP-4 and analyzed its ability to influence the expression of PMN
surface receptors involved in PMN migration. As a result of the action
of MP-4, the surface expression of CD62L was down-regulated, and both
TNF-R p55 and p75 were shed into the surrounding medium. Furthermore,
MP-4-induced signaling caused Ca2+ transients
that could regulate inflammatory reactions by desensitizing
chemoattractant receptors.
| Materials and Methods |
|---|
|
|
|---|
The cryptococcal culture filtrate Ags
CneF and CneF were
prepared from cultures of the non-GXM-producing C.
neoformans mutant strain, CAP67 (E. S. Jacobson, Medical
College of Virginia, Richmond, VA), or the isogenic capsular
strain NIH B3501, respectively. Inocula (108 CFU)
of these strains were added to 500 ml RPMI 1640 supplemented with 10
µg/ml gentamicin and allowed to propagate for 5 days at 37°C. Next,
supernatant fluids were isolated by centrifugation and concentrated by
ultrafiltration (Stirred Cell Concentrator; Millipore, Bedford, MA;
3.5-kDa cutoff). The concentrated (
)CneF was filtered, sterilized,
and adjusted to 0.1 mg/ml protein, as determined by the bicinchoninic
acid (Pierce, Rockford, IL) protein concentration assay.
CneF used
in the assays was diluted 3-fold. MPs were purified from
CneF, as
described recently (30). Briefly,
CneF was dialyzed
against excess mQ water, freeze dried, and applied to a 100-ml Con A
Sepharose-4B column. The column was washed and stepwise eluted with
-methyl-D-mannose pyranoside
(
mDm). GalXM flows through this column,
whereas MP-1 and MP-2 are eluted at 0.2 M and MP-4 at 0.4 M
mDm. These fractions were concentrated (MP-4)
or further separated (MP-1 and MP-2) by DE52 or source Q anion exchange
columns. All purified components were finally dialyzed against excess
PBS and kept frozen at -20°C.
Maintenance of endotoxin-free conditions
Preparations of the various cryptococcal components were negative for endotoxin contamination using a Limulus assay (Kabi Diagnostica, Mölndal, Sweden) with a sensitivity of 100 pg/ml Escherichia coli LPS. Nevertheless, all experiments were conducted at least once in the presence of 10 µg/ml polymyxin B sulfate to neutralize any undetected LPS contamination.
Isolation of human PMN
PMN were isolated as previously described (31). Briefly, blood from healthy volunteers was collected into Vacuette tubes containing sodium heparin, diluted with an equal volume of pyrogen-free PBS, and centrifuged through a gradient of Ficoll and Histopaque. PMN were collected from the Histopaque phase, briefly shocked with mQ water, washed, and suspended at 5 x 106 cells/ml in RPMI 1640 supplemented with 0.05% human serum albumin (HSA) (RPMI-HSA).
Transwell assay for measurement of chemotaxis
PMN were labeled by incubating them for 30 min at 20°C with 3.3 µM 2',7'-bis(2-carboxyethyl)-5-(6)-carboxyfluorescein acetoxymethyl ester (BCECF-AM; Molecular Probes Europe, Leiden, The Netherlands) in RPMI-HSA. Cells were washed and suspended in HBSS-HSA; 200-µl aliquots were incubated with an equal volume of each of the cryptococcal Ags being assessed for chemoattractive activity or buffer control for 30 min at 37°C under constant agitation. Just before the end of this incubation, Transwell filters (pore size 3 µm; Costar type 3415, Cambridge, MA) were prewetted in HBSS/HSA. To each bottom compartment we added 600 µl of the indicated chemoattractants: human rIL-8 (10 nM; PeproTech, Rocky Hill, NJ), fMLP (10 nM; Sigma, St. Louis, MO), or PAF (100 nM; Calbiochem, Cambridge, MA). Filter compartments containing 100 µl PMN suspension were placed on top of the bottom compartments. Migration was allowed to proceed for 1 h at 37°C (5% CO2) and analyzed by reading the fluorescence of the wells with a Cytofluor II multiwell platereader (PerSeptive Biosystems, Farmingham, MA) equipped with a 485-nm excitation and a 530-nm emission filter. Control wells containing BCECF-AM-labeled cells were included to obtain the maximal fluorescence value. The fluorescence values of the samples are expressed relative to this maximum. Fluorescence was linear with cell number, with a detection limit of 2500 cells (1% of input), as determined by serial dilution.
Immunostaining and FACS analysis of PMN Ag expression
To measure the expression of selected surface Ags, 5 x 105 PMN were suspended in 40 µl RPMI-HSA containing purified cryptococcal Ags at the designated concentrations. Mixtures were incubated for 30 min at 37°C on a rotation wheel after they had been heated for 3 min in a water bath at 37°C. Then 10 µl RPMI-HSA containing 0.250.5 µg FITC-conjugated mAbs directed against CD11b (IgG1; 44a, American Type Culture Collection, Manassas, VA), CD18 (IgG2a; BD Biosciences, Mountain View, CA), or CD62L (IgG2a; Leu-8; BD Biosciences) was added, and the incubation was continued for 30 min on ice. Heparinized or EDTA-treated whole blood samples were analyzed using an identical staining protocol with a final incubation step in lysing solution (BD Biosciences) to eliminate red cells. Neutrophils were distinguished from other leukocytes by differences in forward and side scatters in FACS analyses.
Expression levels of chemoattractant and TNF receptors on PMN were measured using the same methodology. We determined the expression of IL-8R type A and PAF-R using mAbs to CDw128A of the IgG2b subclass and to PAF-R (IgG), both from Alexis (San Diego, CA). Surface expression levels of C5a-R and fMLP-R were determined using an anti-CD88 mAb (IgG2a; Serotec, Oxford, U.K.) and anti-fMLP-R mAb (IgG1; BD PharMingen, San Diego, CA), respectively. Because these different mAbs were not FITC labeled, binding was visualized by incubation with FITC-conjugated goat F(ab')2 anti-mouse IgG (DAKO, Carpinteria, CA) as a secondary reagent to recognize the mAbs. To quantify the surface expression of the two receptors for TNF (p55 and p75), we used FITC-labeled Abs of IgG1 and IgG2a subclasses, respectively, which were purchased from R&D Systems (Minneapolis, MN). Isotype-matched conjugated mAbs (mouse IgG1 FITC and mouse IgG2a/b R-PE) not specifically reacting with PMN were purchased from DAKO and used as controls to exclude nonspecific or Fc-related binding. The labeled cells were washed twice and analyzed with a FACS or incubated with the secondary Abs before analysis.
Recording of cytosolic Ca2+ concentration
To measure chemoattractant-induced Ca2+ fluxes, PMN were loaded with 2 µM Fluo-3-AM in RPMI-HSA for 15 min at 37°C under agitation, washed with buffer, and suspended at 106/ml in RPMI-HSA. The fluorescence (530 nm) of each 0.5-ml sample was measured to determine the basal Ca2+ level. Because Ca2+ fluxes are very rapid and transient, 5 µl reagent was added under vortexing, and the sample was analyzed immediately by FACS. For each sample, nine measurements (0 through 8) of 2000 cells were performed, each requiring an average time of 10 s for sampling and data saving before the next acquisition was started. Samples were analyzed after gating the PMN population, thereby excluding cell debris and nonspecific staining.
TNF-R ELISA
Soluble TNF-R p55 and p75 shed from the surface of PMN were measured using a commercially available human ELISA kit (R&D Systems), according to the manufacturers protocol. The minimum detectable doses of soluble TNF-R for these assays were typically less than 3 pg/ml for p55 or 1 pg/ml for p75. All data are expressed as means ± SEM.
Generation of Abs directed against MP-4 and immunodetection
A rabbit antiserum directed against MP-4 was prepared by primary
s.c. injection of two New Zealand White rabbits with 100 µg MP-4 in
CFA, followed by a booster of 50 µg MP-4 in IFA. The serum
specifically recognized MP-4 migrating between 25 and 28 kDa, as shown
(see Fig. 5
A) by standard Western blotting and ECL detection
(Amersham Pharmacia Biotech, Uppsala, Sweden). When indicated, samples
were preadsorbed to rabbit IgG purified from preimmune serum using a
protein G-Sepharose 4 Fast Flow column (Amersham Pharmacia Biotech).
Shortly, isolated preimmune IgG was coupled to CNBr-activated Sepharose
4B (Amersham Pharmacia Biotech) beads. Immobilized IgG was added to the
patient serum, and CSpF samples and nonspecific IgG-Ag interactions
were allowed to occur for 30 min at room temperature (batch
binding).
|
| Results |
|---|
|
|
|---|
IL-8, fMLP, and PAF are examples of inflammatory mediators causing
neutropenia because of rapid sequestration of PMN within the
postcapillary venules of target organs. We evaluated the relative
potency of capsular Ags to inhibit the migration of PMN toward these
chemoattractants. Therefore, we first purified GalXM and three
different MP from
CneF. Next, we confirmed that PMN were able to
migrate toward IL-8, PAF, or fMLP in a dose-dependent manner (data not
shown). In our initial experiments, we obtained almost identical
results with CneF and
CneF, which was unexpected because GXM is
considered to be the Ag exclusively responsible for the inhibition of
PMN migration into the CSpF. We were able to detect a statistically
significant inhibition of migration by adding
CneF to PMN migrating
toward fMLP (18% inhibition, p = 0.002) or IL-8 (23%
inhibition, p = 0.025; Fig. 1
A). We then added PMN
incubated with MP-4 or other capsular Ags at the indicated optimal
concentrations to the upper chamber and compared the inhibitory effect
on PMN migration for each of them. GalXM (89%), GXM (46%), and MP-4
(98%) all caused significant (p < 0.01)
inhibition of migration toward IL-8, with MP-4 being the most potent
(Fig. 1
A). GXM, GalXM, and MP-4 all impaired migration
toward fMLP, albeit to a somewhat lesser extent, with migration being
inhibited by 66% (p = 0.05), 60%
(p = 0.006), and 56%
(p = 0.0004), respectively. Migration toward
PAF was completely blocked by MP-4 (p = 0.015)
and significantly inhibited by MP-2 (51%, p = 0.007),
GalXM (28%, p = 0.0003), and GXM (70%,
p = 0.004). Because this inhibition could be caused by
down-regulation of the receptors for these chemoattractants by MP-4, we
analyzed the expression of the receptors for IL-8, fMLP, and PAF on
PMN. There were no significant changes in the expression of these
receptors, although the number of fMLP and PAF receptors was slightly
increased and the number of IL-8Rs slightly decreased (not shown).
Addition of polymyxin B to exclude any possible influence of trace
amounts of LPS yielded identical results.
|
To determine the nature of the potent blocking effect of MP-4 on
PMN migration, we assayed the intrinsic chemoattractive activity of
MP-4. MP-4 attracted PMN in a concentration-dependent manner (Fig. 1
B). Maximum migration (61% of input; p =
0.007) was reached at an MP-4 concentration of 30 µg/ml, whereafter
migration started decreasing, probably due to saturating diffusion of
MP-4 into the upper chamber. This observation was confirmed by the
conventional under-agarose assay (32) for detection of
chemotaxis (data not shown). In this assay, the potency of MP-4 at 200
µg/ml equaled that of the positive control (PHS). Chemotaxis toward
GXM (1 mg/ml) could not be detected with this less-sensitive assay.
Importantly, the under-agarose assay permits measurement of both
chemotaxis and spontaneous migration. The directed migration toward
MP-4 observed in this assay demonstrates that MP-4 is truly a
chemoattractant and not simply acting as a chemokinetic agent.
Furthermore, when 0.1 ng to 10 µg LPS/ml was added to the medium, in
both assays no migration of PMN was observed, thus ruling out the
possibility that the effects of MP-4 were caused by LPS contamination
(data not shown).
Ca2+ mobilization data indicate cross-desensitization of chemoattractant receptors by MP-4
The above results indicated that MP-4 might exert its effect by
acting as a chemoattractant or by mimicking the action of
chemoattractants. Therefore, we determined whether MP-4 could trigger
an intracellular Ca2+ signal, as occurs when
chemoattractants bind to their G protein-coupled receptors. Fig. 2
A shows a steep increase in
intracellular Ca2+ concentration in
Fluo-3-AM-loaded PMN following stimulation by MP-4.
|
Effects of cryptococcal Ags on the expression of CD11b/18 and L-selectin in the presence or absence of serum and bivalent cations
Stimulation of leukocytes with different chemoattractants such as
fMLP, C5a, and LTB4 causes a rapid decrease in the number of surface
CD62L molecules and a concomitant inverse regulation of
Mac-1/CR3 (27). Given the observed functional
resemblance of MP-4 and other chemoattractants, we investigated whether
MP-4 resembles these true-type chemoattractants in this respect.
Culture of isolated PMN with purified MP-4 at 2
µg/106 cells resulted (Fig. 3
, C and D) in a
significant decrease in the amount of cell surface CD62L (68%;
p = 6E-5) and in up-regulation of surface Mac-1
(2.2-fold; p = 7E-4). MP-4 was more potent than GXM or
CneF in evoking this response (Fig. 3
, A and B).
The MP-4 response almost equaled that induced by fMLP, which caused a
decrease in surface CD62L of 74% and an almost 3-fold increase in
Mac-1 expression.
CneF caused a significant down-regulation of CD62L
(47%, p = 4E-9) and up-regulation of CD11b (64%,
p = 9E-8). Neither GXM, GalXM (not shown), MP-1 (not
shown), MP-2 (not shown), nor CneF induced significant CD62L loss or
CD11b gain in the absence of serum. These data provide evidence that
MP-4 is the capsular component that has the greatest effect on the
expression of CD62L and Mac-1.
|
CneF, and GXM. Under
these circumstances,
CneF (75%; p = 0.02), GXM
(149%; p = 0.02), and MP-4 (195%; p =
0.008) significantly increased the expression of CD11b (Fig. 3MP-4 induces shedding of TNF-R from the surface of PMN
Results showed that MP-4 desensitized PMN, which explains the
inhibited PMN influx into the brains of patients with meningitis. In
conflict with this clinical observation is the induction of TNF-
expression observed in vitro (10, 11, 12, 13) because TNF-
is
generally assumed to promote extravasation. Because at least two
reports (34, 35) demonstrated a substantial reduction in
PMN influx at the site of infection after neutralization of TNF, we
investigated the possible involvement of TNF scavenging in reducing
CSpF leukocyte counts. To examine the possible role of TNF-R in
cryptococcal meningitis, we measured the influence of capsular Ags on
the generation of soluble TNF-R for the two types of TNF-R expressed on
the surface of PMN: p55 and p75 (36). With a soluble TNF-R
ELISA kit, the medium from freshly isolated PMN cultured for 1 h
at 37°C contained clearly detectable levels of TNF-R (108 and 186
pg/ml p55 and p75, respectively), suggesting that cell surface
proteolysis is part of the normal turnover of TNF-R. Stimulation of PMN
with MP-4 resulted in a significant increase of both soluble TNF-R p55
(4-fold; p = 0.0002) and p75 (1.8-fold;
p = 0.002) in the supernatants (Fig. 4
A). None of the other tested
capsular components had an effect on TNF-R shedding. Because soluble
TNF-R compete with TNF-
at the cellular receptor, we also measured
the density of p55 and p75 on PMN after treatment with MP-4. We
observed a significant (p = 0.01)
down-regulation of surface p55 on PMN after addition of MP-4. p75 was
also down-regulated; however, due to the large variability between
individual donors, this was not significant (Fig. 4
B).
|
Using a rabbit polyclonal antiserum directed against MP-4, we
measured the levels of MP-4 present in the serum and CSpF of patients
suffering from cryptococcal meningitis by standard Western blot
analysis. As shown in Fig. 5
A,
we analyzed and compared the amount of MP-4 detectable in serum
collected from a cryptococcosis patient and PHS. Although the
nonadsorbed antiserum directed against MP-4 is cross-reactive to
several serum proteins, we clearly detected a specific band at the
putative m.w. of MP-4 in patient, but not pooled serum. The identity of
this putative MP-4 band was confirmed in Fig. 5
B,
lanes 14, in which we applied purified MP-4 in a
concentration of 30, 10, 3, and 1 µg/ml, respectively, as a
reference. In this panel, all samples were preadsorbed to preimmune
IgG, which removed all cross-reactive proteins, except for one doublet
at a high m.w. position. In lanes 514, we analyzed and
compared the amounts of MP-4 detectable in paired samples of CSpF and
serum collected from HIV-infected cryptococcosis patients (5, 6, 7, 8, 10, 11, 12, 13) or a non-HIV-infected patient (9, 14). MP-4
is clearly detected in the serum of these patients at a level strongly
exceeding that present in CSpF, providing evidence for the formation of
a gradient of MP-4 over the blood-brain barrier. The amount of MP-4
present in the serum of these patients was calculated relative to the
standards and ranged between 10 and 55 µg/ml, respectively, by
densitometric scanning of the autoradiograph. The concentration present
in CSpF varied between <1 and 7 µg/ml.
| Discussion |
|---|
|
|
|---|
MP-4 strongly inhibited PMN migration toward fMLP, PAF, and IL-8 (Fig. 1
A). Conversely, MP-4 was found to be a potent
chemoattractant in vitro (Fig. 1
B). The paradox emerging in
this study can be explained by a phenomenon called
cross-desensitization of chemoattractant receptors, as initially
described by Sabroe et al. (33). This hypothesis was
confirmed by our observation that preincubation of PMN with MP-4
impaired Ca2+ mobilization by fMLP, IL-8, and
C5a, but not PAF (Fig. 2
B). Although speculative at present,
this difference in the regulation of chemoattractant-mediated responses
is likely to be related to G protein usage, as suggested by the
distinct pertussis toxin sensitivity of the receptors for IL-8, fMLP,
and C5a vs PAF, with the former being sensitive and the latter at least
partially resistant (38, 39).
Thus, the pathophysiological role of MP-4 is to functionally mimic the
action of chemoattractants, resulting in a premature activation of PMN.
Such a role for MP-4 is consistent with the observed loss of CD62L and
concomitant up-regulation of CD11b/18 (Fig. 3
), because
chemoattractants affect the expression of these molecules (27, 40, 41). Kishimoto et al. (27) showed that PMN shed
CD62L from the cell surface within minutes after activation with
chemotactic factors. Both the expression and activity of CD11b/18 were
greatly increased. This work founded the current view that by the time
chemoattractants arrive in the circulation, CD62L-mediated neutrophil
rolling has already taken place and the integrin-mediated attachment
takes over. In addition, Luscinskas et al. (40) showed
that IL-8, fMLP, and C5a, besides inhibiting the attachment of PMN to
cytokine-activated endothelial monolayers, promoted the rapid
detachment of tightly adherent PMN from activated endothelial cells,
and abolished transendothelial migration. Finally, in IL-8-transgenic
mice, but not in nontransgenic littermates, PMN migration into the
inflamed peritoneal cavity was severely inhibited (41).
Thus, although fMLP, IL-8, and C5a have been characterized primarily as
chemoattractants, they can exert a wide range of modulatory effects on
PMN-endothelial adhesive interactions.
The association between MP-4 and down-regulation by shedding of surface
TNF-R (Fig. 4
) is also in agreement with the proposed role of MP-4 as
chemoattractant. Several in vitro studies (42, 43)
demonstrated that within minutes of chemoattractant incubation, both
TNF-R p55 and p75 were down-regulated from the surface of PMN. At the
same time, soluble TNF-R appeared in supernatants, in amounts
proportional to the extent of down-regulation. This suggests that
shedding is the major mechanism leading to the loss of p55 and p75 upon
chemoattractant activation. Our demonstration of increased amounts of
shed receptors in conditioned medium of neutrophils exposed to MP-4
supports this hypothesis. In vivo, scavenging of TNF-
by soluble
TNF-R is likely to contribute to the observed lack of leukocytes in the
CSpF of patients for at least two reasons. First, others and we have
shown that cryptococcal Ags stimulate the production of TNF-
by
leukocytes in the presence of serum (12, 13). Second,
TNF-
is known to play a crucial role in the control of organ
infiltration. TNF-
activates the endothelium to cause leukocyte
adherence (4, 44) and triggers leukocyte infiltration of
lung, liver, and kidney in control mice, but not p55-deficient mice
(44). Finally, TNF-
can cause profound systemic
vasodilatation and subsequent hypotension (16), allowing
CD62L-mediated adherence of leukocytes to the vessel wall
(17, 18, 19). Together, these in vivo data indicate that the
presence of TNF-
positively correlates with leukocyte organ
infiltration.
In a series of studies by Dong and Murphy and our group (22, 23, 24, 25, 45), GXM has been shown to possess anti-inflammatory
properties that prevent PMN from accumulating in C.
neoformans-infected tissues. The present work shows that MP-4
shares most of these properties with GXM, but is more potent. Although
analyzed before (22, 24, 45), no direct effects of MP on
PMN migration have been reported. These conflicting results can be
explained by differences in the purification strategy used in this
study. Previously, MP were eluted from the Con A column using a 0.2 M
mDm step elution protocol. Recently, however,
we reported that MP-4 elutes from this column at 0.4 M
-mDm (30), thus indicating that
MP-4 was never present in the previously used MP preparations. Because
analysis of MP-4 reveals a signal eluting between polysaccharide
calibration standards of 18 and 26 kDa in gel filtration chromatography
(not shown), this component also was not present in
CneF used in the
studies mentioned above, because the culture filtrates were
concentrated using a 30-kDa cutoff cassette.
Our analysis of the concentrations of MP-4 present in the serum of patients with cryptococcal meningitis has revealed a concentration range matching that required to obtain the described in vitro effects, thereby confirming the biological significance of our data. The concentrations required are high when compared with serum concentrations for mannoproteins from other fungi, but are in line with experiments, both in vitro and in vivo, performed with other mannoproteins from C. neoformans (46, 47). The relatively high concentrations of MP-4 detected in four of five patients can be partly explained by the fact that these patients were all newly admitted patients for which treatment had not been started at the moment of sampling. To further explain the relatively high concentrations required for cryptococcal mannoproteins, experiments are in progress to determine the biological activity and half-life of MP-4 in vivo as well as in vitro.
In summary, we have shown that MP-4 has potent anti-inflammatory activity that may be responsible for the poor cellular infiltration of PMN into the CSpF of cryptococcosis patients. Given our hypothesis that MP-4 exerts its action largely by its intrinsic chemoattractive properties, the balance between proinflammatory and anti-inflammatory effects will depend critically on the temporal and spatial distribution of MP-4. Interestingly, even in patients with cryptococcal meningitis, MP-4 concentrations in serum strongly exceed the levels detected in CSpF. We hypothesize that this distribution might be due to active drainage of MP-4 from CSpF to blood, resulting in increasing blood concentrations of MP-4 in the course of the disease. Active drainage will prevent both the migration of PMN toward MP-4 present in the CSpF, as well as the proper response to extravascular chemoattractants because of the premature activation of PMN. Alternatively, MP-4 expression might be strongly up-regulated during cryptococcal trafficking in the vascular compartment, before the brain section is reached. Environmentally induced expression of capsule genes is not unprecedented. Several nutrients, e.g., iron and CO2, have been demonstrated to affect the composition of the cryptococcal capsule (48). We are currently testing the ability of MP-4 to limit neutrophil infiltration in vivo using a rabbit model for BM.
| Footnotes |
|---|
2 Abbreviations used in this paper: BM, bacterial meningitis;
mDm,
-methyl-D-mannose pyranoside; BCECF-AM, 2',7'-bis(2-carboxyethyl)-5-(6)-carboxyfluorescein acetoxymethyl ester; CneF, cryptococcal culture filtrate; CSpF, cerebrospinal fluid; GalXM, galactoxylomannan; GXM, glucuronoxylomannan; HSA, human serum albumin; MP, mannoprotein; PAF, platelet-activating factor; PHS, pooled human serum; PMN, polymorphonuclear neutrophil; CD62L, L-selectin; TNF-R, TNF-
R. ![]()
Received for publication February 1, 2001. Accepted for publication July 18, 2001.
| References |
|---|
|
|
|---|
in human leukocytes stimulated by Cryptococcus neoformans. Infect. Immun. 62:1975.
in human peripheral blood mononuclear cells by the mannoprotein of Cryptococcus neoformans involves human mannose binding protein. J. Immunol. 159:2979.[Abstract]
in human peripheral blood mononuclear cells. Infect. Immun. 65:272.[Abstract]
interferon in acquired resistance to Cryptococcus neoformans in the central nervous system of mice. Infect. Immun. 63:1725.[Abstract]
and granulocyte-macrophage colony-stimulating factor for phagocytosis of Cryptococcus neoformans. Eur. J. Immunol. 22:1447.[Medline]
2 integrins in vivo. Proc. Natl. Acad. Sci. USA 88:7538.
: tissue distribution, biochemistry, and function of a natural adherence molecule (ICAM-1). J. Immunol. 137:245.[Abstract]
. Nature 381:77.[Medline]
This article has been cited by other articles:
![]() |
Y. Luo, B. M. Isaac, A. Casadevall, and D. Cox Phagocytosis Inhibits F-Actin-Enriched Membrane Protrusions Stimulated by Fractalkine (CX3CL1) and Colony-Stimulating Factor 1 Infect. Immun., October 1, 2009; 77(10): 4487 - 4495. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Grijpstra, G. J. Gerwig, H. Wosten, J. P. Kamerling, and H. de Cock Production of Extracellular Polysaccharides by CAP Mutants of Cryptococcus neoformans Eukaryot. Cell, August 1, 2009; 8(8): 1165 - 1173. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Jongerius, J. Kohl, M. K. Pandey, M. Ruyken, K. P.M. van Kessel, J. A.G. van Strijp, and S. H.M. Rooijakkers Staphylococcal complement evasion by various convertase-blocking molecules J. Exp. Med., October 1, 2007; 204(10): 2461 - 2471. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Pietrella, C. Corbucci, S. Perito, G. Bistoni, and A. Vecchiarelli Mannoproteins from Cryptococcus neoformans Promote Dendritic Cell Maturation and Activation Infect. Immun., February 1, 2005; 73(2): 820 - 827. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. M. Ellerbroek, D. J. Lefeber, R. van Veghel, J. Scharringa, E. Brouwer, G. J. Gerwig, G. Janbon, A. I. M. Hoepelman, and F. E. J. Coenjaerts O-Acetylation of Cryptococcal Capsular Glucuronoxylomannan Is Essential for Interference with Neutrophil Migration J. Immunol., December 15, 2004; 173(12): 7513 - 7520. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. Postma, M. J. Poppelier, J. C. van Galen, E. R. Prossnitz, J. A. G. van Strijp, C. J. C. de Haas, and K. P. M. van Kessel Chemotaxis Inhibitory Protein of Staphylococcus aureus Binds Specifically to the C5a and Formylated Peptide Receptor J. Immunol., June 1, 2004; 172(11): 6994 - 7001. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. K. Mansour, L. E. Yauch, J. B. Rottman, and S. M. Levitz Protective Efficacy of Antigenic Fractions in Mouse Models of Cryptococcosis Infect. Immun., March 1, 2004; 72(3): 1746 - 1754. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Bose, A. J. Reese, J. J. Ory, G. Janbon, and T. L. Doering A Yeast under Cover: the Capsule of Cryptococcus neoformans Eukaryot. Cell, August 1, 2003; 2(4): 655 - 663. [Full Text] [PDF] |
||||
![]() |
P. M. Ellerbroek, A. I. M. Hoepelman, F. Wolbers, J. J. Zwaginga, and F. E. J. Coenjaerts Cryptococcal Glucuronoxylomannan Inhibits Adhesion of Neutrophils to Stimulated Endothelium In Vitro by Affecting Both Neutrophils and Endothelial Cells Infect. Immun., September 1, 2002; 70(9): 4762 - 4771. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |