Toll-Like Receptor 2 Suppresses Immunity against Candida
albicans through Induction of IL-10 and Regulatory T Cells
Mihai G. Netea,2*§Roger Sutmuller,2†Corinna Hermann,¶Chantal A. A. Van der Graaf,*§
Jos W. M. Van der Meer,*§Johan H. van Krieken,‡Thomas Hartung,¶Gosse Adema,†and
Bart Jan Kullberg3*§
Toll-like receptor (TLR) 2 and TLR4 play a pivotal role in recognition of Candida albicans. We demonstrate that TLR2?/?mice
are more resistant to disseminated Candida infection, and this is associated with increased chemotaxis and enhanced candidacidal
capacity of TLR2?/?macrophages. Although production of the proinflammatory cytokines TNF, IL-1?, and IL-1? is normal,
IL-10 release is severely impaired in the TLR2?/?mice. This is accompanied by a 50% decrease in the CD4?CD25?regulatory
T (Treg) cell population in TLR2?/?mice. In vitro studies confirmed that enhanced survival of Treg cells was induced by TLR2
agonists. The deleterious role of Treg cells on the innate immune response during disseminated candidiasis was underscored by
the improved resistance to this infection after depletion of Treg cells. In conclusion, C. albicans induces immunosuppression
through TLR2-derived signals that mediate increased IL-10 production and survival of Treg cells. This represents a novel mech-
anism in the pathogenesis of fungal infections. The Journal of Immunology, 2004, 172: 3712–3718.
disseminated candidiasis has changed little despite the availability
of new antifungal drugs (1, 2). Despite the importance of Candida
albicans in human pathology, relatively little is known of the
mechanisms through which this fungus is recognized by the im-
mune cells and triggers the host defense. We have recently shown
that the Toll-like receptors (TLR)42 and TLR4 are important rec-
ognition receptors for Candida, and we have demonstrated in-
creased susceptibility of TLR4-deficient mice to disseminated can-
didiasis (3). In human blood cells, TLR2-derived signals were
shown to contribute to production of proinflammatory cytokines
induced by C. albicans blastoconidia (3). This finding could imply
that TLR2 is a receptor involved in anticandidal host defense.
These data are sustained by the finding that TLR2 is involved in
the recognition of zymosan (a cell wall particle of the yeast Sac-
charomyces), leading to proinflammatory cytokine production (4).
In murine experimental infection, such as Staphylococcus au-
reus sepsis or pneumococcal meningitis, absence of TLR2 has
been accompanied by increased mortality, although the precise
mechanisms of TLR2 involvement in host defense have not been
identified (5–7). Cytokine release, activation of leukocyte migra-
ungal infections in general, and acute disseminated candi-
diasis in particular, are severe infections occurring mainly
in immunocompromised hosts. Mortality associated with
tion, and direct antimicrobial action have been implicated as po-
tential defense mechanisms triggered by TLRs (8).
Based on our initial in vitro data, we hypothesized that TLR2
knockout (TLR2?/?) mice would display a reduced activation of
innate immunity during infection with C. albicans, and would
prove more susceptible to disseminated candidiasis. However, the
results of the present study reveal that the absence of TLR2 leads
to increased resistance to candidiasis. We demonstrate that this is
associated with decreased release of anti-inflammatory, but not
proinflammatory cytokines, improved leukocyte recruitment to the
site of infection and candidacidal activity, and decreased numbers
of CD4?CD25?regulatory T cells (Treg).
Materials and Methods
TLR2?/?mice and TLR2?/?control littermates (20–25 g, 6–8 wk old)
were kindly provided by Tularik (San Francisco, CA). TLR4-deficient
ScCr mice were from a local colony at Nijmegen University, and control
TLR4-competent C57BL/10J mice were obtained from The Jackson Lab-
oratory (Bar Harbor, ME). The mice were fed sterilized laboratory chow
(Hope Farms, Woerden, The Netherlands) and water ad libitum. The ex-
periments were approved by the ethics committee on animal experiments
of Nijmegen University.
C. albicans infection model
C. albicans UC 820, a strain well described earlier (9), has been used in all
experiments. A lethal experimental model of disseminated candidiasis was
used to assess mortality, in which TLR2?/?and TLR2?/?mice were in-
jected i.v. with C. albicans (either 5 ? 106or 5 ? 105CFU/mouse) in a
100 ?l vol of sterile pyrogen-free PBS. Survival was assessed daily for
For the assessment of fungal growth in the organs, a nonlethal experi-
mental model of disseminated candidiasis was used, in which TLR2?/?
and TLR2?/?mice were injected i.v. with C. albicans (1 ? 105CFU/
mouse). In the nonlethal model, subgroups of five animals were killed on
day 1 or 7 of infection. To assess the tissue outgrowth of the microorgan-
isms on these days, the liver, the left kidney, and the brain of the sacrificed
animals were removed aseptically, weighed, and homogenized in sterile
saline in a tissue grinder. The number of viable Candida cells in the tissues
was determined by plating serial dilutions on Sabouraud dextrose agar
plates, as previously described (10). The CFU were counted after 24 h of
incubation at 37°C, and expressed as log CFU/g tissue. From the same
animals, the right kidneys were fixed in formaldehyde (4%) and embedded
Departments of *Medicine,†Tumor Immunology, and‡Pathology, University Med-
ical Center St. Radboud, Nijmegen, The Netherlands;§Nijmegen University Center
for Infectious Diseases, Nijmegen, The Netherlands; and¶Department of Biochemical
Pharmacology, University of Konstanz, Konstanz, Germany
Received for publication September 25, 2003. Accepted for publication January
The costs of publication of this article were defrayed in part by the payment of page
charges. This article must therefore be hereby marked advertisement in accordance
with 18 U.S.C. Section 1734 solely to indicate this fact.
1This study was partly supported by an International Sepsis Forum grant (to M.G.N.).
2M.G.N. and R.S. contributed equally to this work.
3Address correspondence and reprint requests to Dr. Bart Jan Kullberg, Department
of Medicine (541), University Medical Center St. Radboud, P.O. Box 9101, 6500 HB
Nijmegen, The Netherlands. E-mail address: B.Kullberg@aig.umcn.nl
4Abbreviations used in this paper: TLR, Toll-like receptor; PMN, polymorphonuclear
cell; Treg, regulatory T cells.
The Journal of Immunology
Copyright © 2004 by The American Association of Immunologists, Inc.0022-1767/04/$02.00
in paraffin, and serial sections were examined microscopically after stain-
ing with periodic acid Schiff and H&E.
Recruitment of polymorphonuclear cells (PMN) and
To investigate the recruitment of PMN and monocytes/macrophages at the
site of Candida infection, groups of five TLR2?/?and TLR2?/?mice were
injected i.p. with 107C. albicans organisms, in a volume of 100 ?l. The
experiments on leukocyte recruitment were performed using heat-killed C.
albicans, and not the live microorganisms, to avoid the bias induced by the
differential candidacidal function of the leukocytes in the knockout mice
and the wild-type controls. In pilot experiments, the maximum PMN in-
filtration after i.p. injection of heat-killed Candida was found after 4 h,
whereas the maximal monocyte/macrophage infiltration was found after
72 h. After 4 or 72 h, peritoneal cells from TLR2?/?and TLR2?/?mice
were collected in sterile saline containing 0.38% sodium citrate, and the
total cell number was counted in a hemocytometer. The percentage and the
absolute numbers of neutrophils and macrophages were determined in
Giemsa-stained cytocentrifuge preparations.
Phagocytosis and killing of C. albicans by macrophages and
Either resident peritoneal macrophages or exudate peritoneal PMN were
obtained, and phagocytosis and killing were performed by a modification
of a method described earlier (11). Exudate peritoneal phagocytes from
groups of five TLR2?/?and TLR2?/?mice were elicited by an i.p. injec-
tion of 10% proteose peptone. Cells were collected in separate sterile tubes
by washing the peritoneal cavity with 4 ml of ice-cold PBS containing 50
U/ml heparin, 4 h (65 ? 11% PMN) or 72 h (84 ? 6% macrophages) after
injection. Phagocytes were centrifuged (10 min; 3600 rpm; 2250 ? g),
counted in a hemocytometer, and resuspended in RPMI 1640 Dutch mod-
ification (with 20 mM HEPES, without glutamine; ICN Biomedicals,
Eschwege, Germany) supplemented with 5% heat-inactivated FCS, 1%
gentamicin, 1% L-glutamine, and 1% pyruvate. The processes of phago-
cytosis and intracellular killing were studied in an adherent monolayer of
phagocytes, as previously described (12). The percentage of phagocytized
microorganisms was defined as (1 – (number of uningested CFU/CFU at
the start of incubation)) ? 100.
Killing of C. albicans by phagocytes was assessed in the same mono-
layers (12). After removal of the nonphagocytized Candida blastoconidia,
200 ?l of culture medium, consisting of Sabouraud in MEM (50% v/v),
was added to the monolayers. After 3 h of incubation at 37°C in air and 5%
CO2, the wells were gently scraped with a plastic paddle and washed with
200 ?l of distilled H2O to achieve lysis of phagocytes. The percentage of
yeast killed by the phagocytes was determined as follows: (1 – (CFU after
incubation/number of phagocytized CFU)) ? 100. Phagocyte-free incuba-
tions of blastoconidia were included as a control for yeast viability.
Extracellular killing of C. albicans hyphae was determined by a mod-
ification of the XTT (2,3-bis-(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetra-
zolium-5-carboxanilide) (Sigma-Aldrich, St. Louis, MO) dye assay, as de-
scribed elsewhere (13).
In vitro cytokine production
Groups of five TLR2?/?and TLR2?/?mice were killed, and resident
peritoneal macrophages were harvested by injecting 4 ml of sterile PBS
containing 0.38% sodium citrate (11). After centrifugation and washing,
the cells were resuspended in RPMI 1640 containing 1 mM pyruvate, 2
mM L-glutamine, 100 ?g/ml gentamicin, and 2% fresh mouse plasma (cul-
ture medium). Cells were cultured in 96-well microtiter plates (Greiner,
Alphen a/d Rijn, The Netherlands) at 105cells/well, in a final volume of
200 ?l. The cells were stimulated with either control medium or heat-killed
(1 h, 100°C) C. albicans at 1 ? 107CFU/ml. After 24 h of incubation at
37°C, the plates were centrifuged (500 ? g, 10 min), and the supernatant
was collected and stored at ?80°C until cytokine assays were performed.
For the assessment of IFN-? and IL-10 production capacity, primed
spleen cells from mice on day 7 of infection with 1 ? 105CFU/mouse of
C. albicans were stimulated in vitro with heat-killed Candida. Spleen cells
were obtained by gently squeezing spleens in a sterile 200-?m filter cham-
ber. Microscopic examination of Giemsa-stained cytospin preparations
showed that these cells consisted of 95% lymphocytes, 2% monocytes, and
3% granulocytes. The cells were washed and resuspended in RPMI 1640
and counted in a Bu ¨rker counting chamber, and the number was adjusted
to 5 ? 106/ml. One milliliter of the cell suspension was stimulated with
1 ? 107heat-killed C. albicans yeasts (E:T ratio, 2:1). Measurement of
IFN-? and IL-10 concentrations was performed in supernatants collected
after 48 h of incubation at 37°C in 5% CO2in 24-well plates (Greiner).
IL-1?, IL-1?, and TNF-? were determined by specific RIAs (detection
limit 20 pg/ml), as previously described (14). IFN-? and IL-10 concentra-
tions were measured by a commercial ELISA (BioSource International,
Camarillo, CA; detection limit 16 pg/ml), according to the instructions of
CD4?CD25?Treg cells in TLR2?/?mice
To assess the presence of Treg cells in the circulation of TLR2?/?and
TLR2?/?mice, 50 ?l of blood was collected from the tail of 10 deficient
and 10 control mice. Analysis of cell surface markers on blood lympho-
cytes was performed using a FACSCalibur (BD Biosciences, Alphen aan
de Rijn, The Netherlands) and CellQuest software. Blood (50 ?l) was
collected in heparin-coated tubes, and erythrocytes were lysed using stan-
dard laboratory protocols. The remaining lymphocytes were washed and
incubated with anti-CD25 FITC (clone 7D4; BD PharMingen, San Diego,
CA) and anti-CD4 allophycocyanin (BD PharMingen) and subsequently
analyzed. Data are represented as the percentage of CD4?CD25?cells of
the total CD4?T cell population.
In addition to TLR2?/?mice, we also investigated the role of TLR4 in
the generation of Treg cells, by assessing the CD4?CD25?T cell popu-
lation in TLR4-deficient ScCr mice and control C57BL/10J mice.
T cell survival assay
Spleens were mashed into single cell suspensions using a nylon filter.
CD4?T cells were presorted by magnetic purification using anti-CD4
(L3T4) microbeads and positively separated on large separation columns
(both from Miltenyi Biotec, Auburn, CA). CD4?CD25?and CD4?CD25?
T cell subsets were obtained by flow cytometry purification of the presorted
T cells: CD4 cells were stained with allophycocyanin-conjugated CD4
mAb (BD Biosciences) and FITC-conjugated CD25 (PC61; BD Bio-
sciences). Cell sorting was performed on a Coulter Elite cell sorter (Corixa,
Seattle, WA). The purity of each cell preparation was ?95%. Purified
CD4?T cell subsets were subsequently cultured for 3 days in complete
medium (Iscove’s IMDM, 9% FCS) supplemented with, if indicated, 0.5
?g/ml highly purified Escherichia coli LPS (Sigma-Aldrich), 10 ?g/ml
peptidoglycan (Sigma-Aldrich), or 5 Cetus U/ml IL-2. After 3 days of
culture, all surviving cells were harvested using PBS/1 mM EDTA, stained
for CD4 and CD25 using the aforementioned Abs, and subsequently ana-
lyzed on a flow cytometer. Data indicate the average number and SEM of
all surviving CD4?cells of triplicate wells.
The role of CD4?CD25?Treg cells in host defense against
The possible effect of Treg cells on the anticandidal host defense was
investigated by depletion of Treg cell population using a protein G-purified
rat anti-mouse CD25 (300 ?g, clone PC61; American Type Culture Col-
lection, Manassas, VA) administered by i.p. injection 10 days before in-
fection with 1 ? 105CFU/mouse C. albicans. Rat IgG was used as control
in a separate group of mice. On days 1 and 7 of infection, Candida out-
growth in the organs of mice receiving anti-CD25 or control Abs was
assessed, as described above.
The mortality in the two groups was compared by Kaplan-Meier logarithm
of rank test. The differences in the other parameters between groups were
analyzed by Mann-Whitney U test, and where appropriate by Kruskal-
Wallis ANOVA test. The level of significance between groups was set at
p ? 0.05. All experiments were performed at least twice, and the data are
presented as cumulative results of all experiments performed.
TLR2?/?mice are less susceptible to disseminated C. albicans
TLR2 is one of the cellular receptors engaged by C. albicans (3).
To determine the role of TLR2 in the host defense against C.
albicans, we infected TLR2?/?mice with Candida and compared
their susceptibility to infection with that in wild-type mice. Sur-
prisingly, TLR2?/?mice survived longer after injection of a lethal
amount of C. albicans (5 ? 106CFU/mouse) than control mice
(p ? 0.05, Fig. 1A). All deaths occurred during the first 10 days of
infection, and no mortality was registered thereafter. When the
3713The Journal of Immunology
mice were injected with a lower inoculum of 5 ? 105CFU/ml, no
mortality was recorded in either group (data not shown). The target
organ for C. albicans growth in murine candidiasis, the kidneys,
showed no differences in fungal outgrowth between deficient and
control mice on day 1. However, on day 7 of infection, TLR2?/?
mice were found to have a 100-fold decreased load of C. albicans
in their kidneys (p ? 0.01), compared with TLR2?/?mice (Fig.
1B). A similar tendency, although not significant, was apparent for
the outgrowth of C. albicans in the liver and brain (Fig. 1B). His-
tology of the kidneys showed no differences in the inflammatory
infiltrate and the degree of Candida growth on day 1 of infection.
In contrast, on day 7, control TLR2?/?mice displayed invasive
growth of large amounts of C. albicans, especially in the pyelum,
compared with larger inflammatory infiltrates and fewer Candida
organisms in the TLR2?/?mice (Fig. 2).
Monocyte recruitment to the site of infection is increased in
To investigate the recruitment of PMN and macrophages to the site
of a C. albicans infection, groups of TLR2?/?and TLR2?/?mice
were injected i.p. with 107heat-killed C. albicans microorganisms,
and exudate peritoneal neutrophils or macrophages were harvested
4 and 72 h later, respectively. As shown in Fig. 3A, there was
significantly more influx of monocytes/macrophages into the peri-
toneal cavity of TLR2?/?mice than in that of TLR2?/?mice
(p ? 0.05), whereas the recruitment of PMN did not differ be-
tween the groups.
Increased candidacidal capacity of macrophages from TLR2?/?
Phagocytosis of C. albicans by PMN or macrophages of TLR2?/?
mice was similar to that of TLR2?/?mice (Fig. 3B). Both the
intracellular killing of Candida blastoconidia and the extracellular
killing of Candida hyphae by macrophages of TLR2?/?mice were
better than by macrophages of control TLR2?/?mice (Fig. 3, C
and D). In contrast, neutrophils of TLR2?/?and TLR2?/?mice
were equally potent to kill both conidia and hyphae of Candida
(Fig. 3, C and D).
TLR2?/?mice induce lower IL-10, but higher IFN-? release,
upon stimulation with C. albicans
To investigate the role of TLR2 in the stimulation of cytokines by
C. albicans, we stimulated peritoneal macrophages of TLR2?/?
and control TLR2?/?mice with heat-killed Candida blastospores
in vitro. Cytokine production by unstimulated macrophages of
both mouse strains was below the detection limit for all cytokines
studied (data not shown). Candida-stimulated production of TNF,
IL-1?, and IL-6 was only 20–30% reduced in macrophages iso-
lated from TLR2?/?mice compared with control TLR2?/?mice
TLR2?/?mice. TLR2?/?(F) and TLR2?/?(‚) mice were injected i.v.
with either 5 ? 105(A) or 1 ? 105(B) CFU/mouse C. albicans. Survival
(A) was assessed daily for 14 days in a lethal model of disseminated can-
didiasis: all mortality occurred during the first 10 days, and no further
deaths were recorded therafter. In a sublethal model (B), subgroups of
TLR2?/?(?) and TLR2?/?(f) mice were sacrificed on day 7 postinfec-
tion to assess the outgrowth of the yeasts in the organs. Data represent
means ? SEM of 10 mice per group. ??, p ? 0.01 by Kaplan-Meier
logarithm of rank test (A) and Mann-Whitney U test (B).
Decreasedsusceptibilityto invasivecandidiasis in
infected with C. albicans. Control TLR2?/?(A and B) and TLR2?/?(C
and D) mice were infected i.v. with 1 ? 105CFU of C. albicans, and
subgroups of five animals were sacrificed to assess the outgrowth of the
yeasts in the right kidneys. Serial sections were examined microscopically
after staining with periodic acid Schiff and H&E. On day 7 after infection,
control TLR2?/?mice displayed invasive growth of large amounts of C.
albicans both in the renal tissue (A) and pyelum (B), accompanied by
moderate inflammatory infiltrates. In contrast, very little growth of Can-
dida was found in the TLR2?/?mice, which had an almost normal kidney
architecture (C), whereas a severe inflammatory infiltrate was present
around the pyelum (D). Magnification: ?400. Arrows indicate C. albicans
Histology of the kidneys of TLR2?/?and TLR2?/?mice
3714 TLR2 IN DISSEMINATED CANDIDIASIS
(p ? 0.05, Fig. 4A). In contrast, the synthesis of the anti-
inflammatory cytokine IL-10 was significantly lower in TLR2?/?
macrophages stimulated with Candida blastospores than in con-
trols (42 ? 16 vs 98 ? 22 pg/ml, p ? 0.05). Moreover, the IL-10
production by spleen cells isolated from C. albicans-infected
TLR2?/?mice on day 7 of infection was only 25–30% of that by
TLR2?/?spleen cells when stimulated with Candida, whereas
IFN-? release was 3-fold higher than in TLR2?/?splenocytes
CD4?CD25?Treg cell numbers are decreased in TLR2?/?
Recent data have implicated a role for Treg cells in the immune
response to C. albicans (15), and it has been suggested that IL-
10-mediated signals are involved in the generation of Treg cells
(15). In contrast to TLR4?/?mice, which showed normal numbers
of Treg cells, there was a 50% decrease of the Treg population in
uninfected TLR2?/?mice compared with their wild-type litter-
mates (p ? 0.05; Fig. 5A). There were no differences either in the
total leukocyte counts of TLR2?/?and TLR2?/?mice, or in
the absolute numbers of CD4?T lymphocytes (21.1 vs 19.7% of
the total lymphocyte count; in absolute numbers, 1234 ? 342 vs
1102 ? 298 cells/mm3). This resulted in significantly decreased
numbers of Treg cells in the TLR2?/?mice (TLR2?/?vs
TLR2?/?mice: 10.5 vs 5.4%; 129 ? 35 vs 64 ? 16 cells/mm3,
p ? 0.05).
TLR2-mediated signals promote Treg cell survival
To establish the link between TLR2 and the number of Treg cells,
we analyzed T cell survival of purified CD4?CD25?and
CD4?CD25?subsets from wild-type animals. Three days of cul-
ture in the presence of IL-2 resulted in a 2-fold increase in T cell
survival as compared with the medium control for both CD25?
(Fig. 5B), as well as CD25?T cells (data not shown). Interestingly,
culture of T cells in medium supplemented with the TLR2 ligand
peptidoglycan also improved Treg cell survival (Fig. 5B). Surpris-
ingly, the addition of highly purified LPS did not have a significant
effect, indicating that in this particular setting, stimulation through
TLR2, but not TLR4, prolongs Treg cell survival in vitro (Fig. 5B).
TLR2?/?(f) mice were injected i.p. with 107heat-killed C. albicans mi-
croorganisms, and exudate peritoneal neutrophils or macrophages were
harvested 4 and 72 h later, respectively. B, PMN or macrophages from
TLR2?/?(?) or TLR2?/?(f) mice were incubated for 15 min with C.
albicans (E:T ratio 20:1), and the percentage of phagocytized microorgan-
isms was calculated as (1 – (number of uningested CFU/CFU at the start of
incubation)) ? 100. C and D, Intracellular killing of C. albicans blasto-
spores (C) or extracellular killing of C. albicans hyphae (D), by neutrophils
or macrophages of TLR2?/?mice (?) and TLR2?/?mice (f) were as-
sessed after 3 h of incubation. Data represent the mean ? SD for two
experiments with 10 mice/group. ?, p ? 0.05 by Mann-Whitney U test.
Function of PMN and macrophages. A, TLR2?/?(?) and
Naive murine peritoneal macrophages (A) or spleen cells on day 7 of in-
fection with 1 ? 105C. albicans (B) were harvested from TLR2?/?mice
(?) and TLR2?/?mice (f). The cells were stimulated with 107CFU/ml
heat-killed C. albicans. The proinflammatory cytokines TNF, IL-1?, and
IL-6 (A) or the Th cytokines IFN-? and IL-10 (B) were measured 24 and
48 h later, respectively. Data represent means ? SEM of 10 mice. ?, p ?
0.05 by Mann-Whitney U test.
Production of cytokines in TLR2?/?and TLR2?/?mice.
3715 The Journal of Immunology
Treg cell depletion decreases susceptibility to disseminated
Treg cells are known to produce large quantities of IL-10 and to
decrease cellular defense, and both mechanisms may decrease in-
nate resistance to C. albicans. We tested this hypothesis by deple-
tion of Treg cells with an anti-CD25 mAb. In previous experi-
ments, anti-CD25 Ab administration resulted in the depletion of
?90% of the CD4?, CD25?T cell population. As resting con-
ventional T cells (both CD4 and CD8 T cells) are CD25 negative,
the majority of the depleted CD25-expressing T cells will have the
suppressor phenotype. At this moment, CD25 is the best cell sur-
face marker for Treg in naive mice. As shown in Fig. 5C, depletion
of Treg cells in normal TLR2?/?mice resulted in a 10-fold de-
crease of fungal outgrowth in the kidneys on day 7, but not day 1
of infection, showing the crucial role of Treg cells in suppressing
host defense in normal TLR2?/?mice.
Our results show that the absence of TLR2-mediated signaling
results in an increased resistance to disseminated candidiasis.
Whereas production of TNF and IL-1 is normal, IL-10 synthesis is
severely impaired in TLR2?/?mice. The decreased production of
IL-10 is associated with an increased production of IFN-?, a di-
minished generation of Treg cells, and improved candidacidal
function of macrophages. This implies that C. albicans evades host
defense through TLR2-mediated signals.
We have recently demonstrated that recognition of the C. albi-
cans cell wall Ags involves both TLR2 and TLR4, and the absence
of TLR4 signals increases susceptibility to disseminated candidi-
asis (3). Which C. albicans cell wall components stimulate TLR2
and TLR4 is unknown as yet, although there are indications that a
mannan component is recognized by TLR4 (16). Because blockade
of TLR2 in human mononuclear cells resulted in a 40–50% de-
crease of the production of proinflammatory cytokines in vitro (3),
our initial hypothesis was that a similar defect in TLR2?/?mice
would lead to an increased susceptibility to systemic C. albicans
infection, as both TNF and IL-1 are central for an effective anti-
fungal defense (17). Surprisingly, the production of TNF and IL-1
was only marginally impaired in TLR2?/?mice, suggesting that
either other TLRs or receptors other than TLR such as mannose
receptor (18), ?-glucan receptor (19), complement receptor 3 (20),
or dendritic cell-specific intercellular adhesion molecule-grabbing
nonintegrin (21) are also involved in Candida-induced proinflam-
matory cytokine release. Moreover, the slight decrease in the TNF
and IL-1 release in TLR2?/?mice did not have functional conse-
quences, implying that the remaining production of these cytokines
is sufficient for effective host defense. It is also important to men-
tion the cooperation between TLR2 and the ?-glucan receptor in
recognition of C. albicans (22, 23), and it is tempting to speculate
on the role of ?-glucan receptors in the findings described in
The increased resistance to disseminated candidiasis was ac-
companied by increased capacity of TLR2?/?monocytes to mi-
grate to the site of infection and to kill Candida, whereas these
functions of neutrophils were similar in the control and TLR2-
deficient mice. This differential effect of TLR2 deficiency on the
function of macrophages and neutrophils is in line with the rela-
tively late effect on the fungal outgrowth in the kidneys (day 7), as
neutrophils are the main cell population involved in the antican-
didal defense during the first days of infection, and macrophages in
the later phase.
To investigate the mechanisms behind the enhanced function of
macrophages in the TLR2?/?mice, we hypothesized that TLR2-
mediated signals induce a suppressive signal, which is absent in
the knockout mice. IL-10 is known to have strong inhibitory ef-
fects on the defense against C. albicans infection, as treatment of
mice with anti-IL-10 Abs induces protection (24), and IL-10?/?
mice display an increased resistance to disseminated candidiasis
(25). Indeed, C. albicans-stimulated IL-10 production by either
peritoneal macrophages from naive TLR2?/?mice or spleen cells
from Candida-primed TLR2?/?mice was severely impaired.
Moreover, this was accompanied by increased IFN-? production,
very likely secondary to the absence of IL-10. These data indicate
that TLR2 mainly induces anti-inflammatory signals during inva-
sive candidiasis, and this may represent a mechanism to evade host
defense. This new paradigm is sustained by the findings of in vitro
studies suggesting that TLR2 mediates signals prone to induce
Th2-type responses (26), as well as a recent study showing that
Yersinia enterocolitica uses TLR2-dependent IL-10 release as a
mechanism of immunosuppression. In accordance with this obser-
vation, TLR2?/?mice are more resistant to Y. enterocolitica in-
fection (27). However, TLR2 seems to provide beneficial signals
in other types of infection such as staphylococcal sepsis (5) or
pneumococcal meningitis (6, 7), although the mechanisms respon-
sible for the protection in these models are unclear.
In addition to its direct inhibitory effects on the function of
macrophages and neutrophils, recent data suggest that IL-10 also
plays a central role in the shape of adaptive immunity. It has been
shown recently that dendritic cell-derived IL-10 is necessary for
proper development of a subset of CD4?CD25?T lymphocytes
(Treg) involved in regulation of effector T lymphocytes (15); in
turn, secretion of IL-10 and TGF-? by Treg cells is responsible for
The percentage of CD4?CD25?Treg cells from the CD4?T cell popu-
lation in mice deficient (f) in either TLR2 or TLR4 was assessed by FACS
analysis, and compared with control mice (?). B, Survival of Treg cells in
the presence of IL-2, the TLR2 agonist peptidoglycan, or the TLR4 agonist
LPS. C, The role of Treg cells in disseminated candidiasis was investigated
by depletion of Treg cells induced by a rat anti-CD25 mAb 10 days before
infection of the mice with 1 ? 105CFU/mouse C. albicans. On day 7 of
infection, Candida outgrowth was assessed in the organs of mice receiving
either anti-CD25 or control Abs. Data represent means ? SEM of 10 mice.
?, p ? 0.05 by Mann-Whitney U test.
Number of CD4?CD25?Treg cells and TLR2 signals. A,
3716 TLR2 IN DISSEMINATED CANDIDIASIS
their effects on innate and adaptive immunity (28). Because we
observed a strong defect in the release of IL-10 in the TLR2?/?
mice, we have investigated both the presence of Treg cells in their
circulation and the role of these cells during disseminated candi-
diasis. In contrast to TLR4?/?mice, which displayed a normal
population of Treg cells, TLR2?/?mice displayed a 50% decrease
in the Treg cell population. In addition, TLR2 ligation by pepti-
doglycan enhanced survival of Treg cells, sustaining the hypoth-
esis that TLR2-mediated signals are crucial for the homeostasis of
Treg cells. We could further demonstrate a deleterious role of these
cells during disseminated candidiasis by depletion of CD25?cells
in normal mice: this led to increased resistance to C. albicans
infection, as shown by the 10-fold decrease of the fungal out-
growth in the kidneys.
The demonstration of decreased IL-10 production, a reduced
Treg cell population in TLR2?/?mice, and beneficial effects of
Treg cell depletion on disseminated candidiasis has two important
conceptual consequences. First, these data strongly suggest that
TLR2-mediated signals, very likely through IL-10 production, are
crucial for the generation of Treg cells. This hypothesis is sus-
tained by recent studies demonstrating that IL-10 is needed for
generation of Treg cells during a mucosal model of Candida in-
fection (15). Along the same line, TLR2-mediated signals induced
by schistosomal lyso-phosphatidylserine lead to the development
of Treg cells (29). Based on our data and the latter studies, we
propose that TLR2-mediated IL-10 release is a stimulatory signal
for generation and function of Treg cells. Others have described a
role for the TLR4/TLR9-mediated IL-6 production in rendering
CD4?cells unresponsive to the Treg-mediated immunosuppres-
sion (30). Thus, different TLRs may modulate the adaptive im-
mune response through either stimulation or inhibition of Treg cell
functions. Despite the immunosuppressive effects during dissem-
inated candidiasis, Treg cells have protective effects in other in-
fections such as experimental models of pneumococcal meningitis,
in which mortality is induced by overwhelming inflammation (6).
In addition, Treg cells have proved to be beneficial in mucosal and
cutaneous infections, and are essential components of the memory-
protective immunity to C. albicans (15) or Leishmania major in-
fection (31). CD4?CD25?Treg cells are thought to be crucial in
maintaining resistance to reinfection through suppression of effec-
tor CD4?CD25?T cells. Thereby, the latter cells are no longer
able to eliminate commensal microorganisms from sites of colo-
nization (31). Conversely, our results could imply that the effects
of Treg cells are deleterious when a pathogen subsequently pen-
etrates the mucosa and disseminates through the bloodstream.
In addition, our data provide an answer to the paradox between
the concept that Th1 cytokines are beneficial to host defense, and
the observation that nude mice and mice depleted of CD4?cells
are relatively resistant to disseminated candidiasis (32–34). Our
finding that selective depletion of CD4?CD25?Treg cells bene-
ficially influences the outcome of disseminated candidiasis may
explain the increased resistance of nude (no T cells) and CD4?-
depleted mice (including CD4?CD25?Treg) to Candida infection
(32–34). Of note, it has been demonstrated that depletion of CD8?
cells has no effects on the susceptibility for infection (34).
In conclusion, TLR2?/?-deficient mice are relatively resistant
to a disseminated infection with C. albicans. This reduced suscep-
tibility to candidiasis is associated with impaired IL-10 production
and a decreased number of Treg cells in TLR2 knockout animals,
accompanied by more IFN-? production and effective antifungal
properties of TLR2?/?macrophages. Thus, C. albicans evades
host defense through TLR2-derived signals, and this represents a
novel pathogenetic mechanism in fungal infections.
1. Pfaller, M. A., R. N. Jones, G. V. Doern, A. C. Fluit, J. Verhoef, H. S. Sader,
S. A. Messer, A. Houston, S. Coffman, R. J. Hollis, and SENTRY Participant
Group (Europe). 1999. International surveillance of bloodstream infections due to
Candida species in the European SENTRY Program: species distribution and
antifungal susceptibility including the investigational triazole and echinocandin
agents. Diagn. Microbiol. Infect. Dis. 35:19.
2. Edmond, M. B., S. E. Wallace, D. K. McClish, M. A. Pfaller, R. N. Jones, and
R. P. Wenzel. 1999. Nosocomial bloodstream infections in United States hospi-
tals: a three-year analysis. Clin. Infect. Dis. 29:239.
3. Netea, M. G., C. de Graaf, A. Vonk, I. Verschueren, J. W. M. Van der Meer, and
B. J. Kullberg. 2002. The role of Toll-like receptors in the defense against dis-
seminated candidiasis. J. Infect. Dis. 185:1483.
4. Underhill, D. M., A. Ozinski, A. M. Hajjar, A. Stevens, C. B. Wilson,
M. Bassetti, and A. Aderem. 1999. The Toll-like receptor 2 is recruited to mac-
rophage phagosomes and discriminates between pathogens. Nature 401:811.
5. Takeuchi, O., K. Hoshino, and S. Akira. 2000. TLR2-deficient and MyD88-
deficient mice are highly susceptible to Staphylococcus infection. J. Immunol.
6. Koedel, U., B. Angele, T. Rupprecht, H. Wagner, A. Roggenkamp, H.-W. Pfister,
and C. J. Kirschning. 2003. Toll-like receptor 2 participates in mediation of
immune response in experimental pneumococcal meningitis. J. Immunol.
7. Echchannaoui, H., K. Frei, C. Schnell, S. L. Leib, W. Zimmerli, and
R. Landmann. 2002. Toll-like receptor 2-deficient mice are highly susceptible to
Streptococcus pneumoniae meningitis because of reduced bacterial clearing and
enhanced inflammation. J. Infect. Dis. 186:798.
8. Qureshi, S. T., and R. Medzhitov. 2003. Toll-like receptors and their role in
experimental models of microbial infection. Genes Immun. 4:87.
9. Lehrer, R. I., and M. J. Cline. 1969. Interactions of Candida albicans with human
leukocytes and serum. J. Bacteriol. 98:996.
10. Kullberg, B. J., J. W. Van ’t Wout, and R. Van Furth. 1990. Role of granulocytes
in enhanced host resistance to Candida albicans induced by recombinant inter-
leukin-1. Infect. Immun. 58:3319.
11. Kullberg, B. J., J. W. Van’t Wout, C. Hoogstraten, and R. Van Furth. 1993.
Recombinant interferon-? enhances resistance to acute disseminated Candida
albicans infection in mice. J. Infect. Dis. 168:436.
12. Vonk, A. G., C. W. Wieland, M. G. Netea, and B. J. Kullberg. 2002. Phagocytosis
and intracellular killing of Candida albicans blastoconidia by neutrophils and
macrophages:a comparison of different
J. Microbiol. Methods 49:55.
13. Vonk, A. G., M. G. Netea, J. H. van Krieken, J. W. M. Van der Meer, and
B. J. Kullberg. 2002. Delayed clearance of intraabdominal abscesses caused by
Candida albicans in tumor necrosis factor-? and lymphotoxin-? deficient mice.
J. Infect. Dis. 186:1815.
14. Netea, M. G., P. N. M. Demacker, B. J. Kullberg, O. C. Boerman, I. Verschueren,
A. F. H. Stalenhoef, and J. W. M. Van der Meer. 1996. Low-density-lipoprotein
receptor deficient mice are protected against lethal endotoxinemia and severe
Gram-negative infections. J. Clin. Invest. 97:1366.
15. Montagnoli, C., A. Bacci, S. Bozza, R. Gaziano, P. Mosci, A. H. Sharpe, and
L. Romani. 2002. B7/CD28-dependent CD4?CD25?regulatory T cells are es-
sential components of the memory-protective immunity to Candida albicans.
J. Immunol. 169:6298.
16. Tada, H., E. Nemoto, H. Shimauki, T. Watanabe, T. Mikami, T. Matsumoto,
N. Ohna, H. Tamura, K. Shibata, S. Akashi, et al. 2002. Saccharomyces cerevi-
siae- and Candida albicans-derived mannan induced production of tumor necro-
sis factor ? by human monocytes in a CD14- and Toll-like receptor 4-dependent
manner. Microbiol. Immunol. 2002:503.
17. Netea, M. G., L. J. H. Van Tits, J. H. A. J. Curfs, F. Amiot, J. F. G. M. Meis,
J. W. M. Van der Meer, and B. J. Kullberg. 1999. The increased susceptibility of
TNF? LT? double knock-out mice to systemic candidiasis is due to defective
recruitment and phagocytosis by neutrophils. J. Immunol. 163:1498.
18. Yamamoto, Y., T. W. Klein, and H. Friedman. 1997. Involvement of mannose
receptor in cytokine interleukin-1? (IL-1?), IL-6, and granulocyte-macrophage
colony-stimulating factor responses, but not in chemokine macrophage inflam-
matory protein 1? (MIP-1?), MIP-2, and KC responses, caused by attachment of
Candida albicans to macrophages. Infect. Immun. 65:1077.
19. Brown, G. D., P. R. Taylor, D. M. Reid, J. A. Willment, D. L. Williams,
L. Martinez-Pomares, S. Y. C. Wong, and S. Gordon. 2002. Dectin-1 is a major
?-glucan receptor on macrophages. J. Exp. Med. 196:407.
20. Forsyth, C. B., and H. L. Mathews. 1996. Lymphocytes utilize CD11b/CD18 for
adhesion to Candida albicans. Cell. Immunol. 170:91.
21. Cambi, A., K. Gijzen, J. de Vries, R. Torensma, B. Joosten, G. J. Adema,
M. G. Netea, B. J. Kullberg, L. Romani, and C. G. Figdor. 2003. The C-type
lectin DC-SIGN (CD209) is an antigen-uptake receptor for Candida albicans on
dendritic cells. Eur. J. Immunol. 33:532.
22. Brown, G. D., J. Herre, D. L. Williams, J. A. Willment, A. S. Marshall, and
S. Gordon. 2003. Dectin-1 mediates the biological effects of ?-glucans. J. Exp.
23. Gantner, B. N., R. M. Simmons, S. J. Canavera, S. Akira, and D. M. Underhill.
2003. Collaborative induction of inflammatory responses by dectin-1 and Toll-
like receptor 2. J. Exp. Med. 197:1107.
24. Romani, L., P. Puccetti, A. Mencacci, E. Cenci, R. Spaccapelo, L. Tonnetti,
U. Grohmann, and F. Bistoni. 1994. Neutralization of IL-10 up-regulates nitric
oxide production and protects susceptible mice from challenge with Candida
albicans. J. Immunol. 152:3514.
3717The Journal of Immunology
25. Vasquez-Torres, A., J. Jones-Carson, R. D. Wagner, T. Warner, and E. Balish. Download full-text
1999. Early resistance of interleukin-10 knock-out mice to acute systemic can-
didiasis. Infect. Immun. 67:670.
26. Re, F., and J. L. Strominger. 2001. Toll-like receptor 2 (TLR2) and TLR4 dif-
ferentially activate human denderitic cells. J. Biol. Chem. 276:37692.
27. Sing, A., D. Rost, N. Tvardovskaia, A. Roggenkamp, A. Wiedemann,
C. J. Kirschning, M. Aepfelbacher, and J. Heesemann. 2002. Yersinia V-antigen
exploits Toll-like receptor 2 and CD14 for interleukin-10-mediated immunosup-
pression. J. Exp. Med. 196:1017.
28. Maloy, K. J., L. Salaun, R. Cahill, G. Dougan, N. J. Saunders, and F. Powrie.
2003. CD4?CD25?Tr cells suppress innate immune pathology through cyto-
kine-dependent mechanisms. J. Exp. Med. 197:111.
29. Van der Kleij, D., E. Latz, J. F. H. M. Brouwers, Y. C. M. Kruize, M. Schmitz,
E. A. Kurt-Jones, T. Espevik, E. C. de Jong, M. L. Kapsenberg, D. T. Golenbock,
et al. 2002. A novel host-parasite lipid cross talk: schistosomal lyso-phosphati-
dylserine activates Toll-like receptor 2 and affects immune polarization. J. Biol.
30. Pasare, C., and R. Medzhitov. 2003. Toll pathway-dependent blockade of
CD4?CD25?T cell-mediated suppression by dendritic cells. Science 299:1033.
31. Belkaid, Y., C. A. Piccirillo, S. Mendez, E. M. Shevach, and D. L. Sacks. 2002.
CD4?CD25?regulatory T cells control Leishmania major persistence and im-
munity. Nature 420:502.
32. Fulurija, A., R. B. Ashman, and J. M. Papadimitriou. 1997. Increased tissue
resistance in the nude mice against Candida albicans without altering strain-
dependent differences in susceptibility. J. Med. Vet. Mycol. 35:197.
33. Tavares, D., P. Ferreira, and M. Arala-Chaves. 2000. Increased resistance to
systemic candidiasis in athymic or interleukin-10-depleted mice. J. Infect. Dis.
34. Coker, L. A., III, C. M. Mercadal, B. T. Rouse, and R. N. Moore. 1992. Differ-
ential effects of CD4?and CD8?cells in acute, systemic murine candidosis.
J. Leukocyte Biol. 51:305.
3718 TLR2 IN DISSEMINATED CANDIDIASIS