Dectin-1 Is Not Required for Controlling Candida albicansColonization of the Gastrointestinal Tract

Simon Vautier,a Rebecca A. Drummond,a Pierre Redelinghuys,a Graeme I. Murray,b Donna M. MacCallum,a and Gordon D. Browna

Aberdeen Fungal Group, University of Aberdeen, Institute of Medical Sciences, Foresterhill, Aberdeen, United Kingdom,a and Pathology, Division of Applied Medicine,University of Aberdeen, Aberdeen, United Kingdomb

Candida albicans is normally found as a commensal microbe, commonly colonizing the gastrointestinal tract in humans. How-ever, this fungus can also cause mucosal and systemic infections once immune function is compromised. Dectin-1 is an innatepattern recognition receptor essential for the control of fungal infections in both mice and humans; however, its role in the con-trol of C. albicans colonization of the gastrointestinal tract has not been defined. Here, we demonstrate that in mice dectin-1 isessential for the control of gastrointestinal invasion during systemic infection, with dectin-1 deficiency associating with im-paired fungal clearance and dysregulated cytokine production. Surprisingly, however, following oral infection, dectin-1 was notrequired for the control of mucosal colonization of the gastrointestinal tract, in terms of either fungal burdens or cytokine re-sponse. Thus, in mice, dectin-1 is essential for controlling systemic infection with C. albicans but appears to be redundant forthe control of gastrointestinal colonization.

Candida albicans is a member of the natural human commensalmicrobiota found in the gastrointestinal and genitourinary

tracts. However, C. albicans can cause both mucosal and systemicinfections and is one of the most important nosocomial infec-tions, with mortality rates of around 40% for disseminated infec-tions (29). Systemic C. albicans infections are commonly thoughtto originate from the commensal mucosal microbiota dissem-inating to tissues following damage to the mucosal barrier,alterations in the host immune system, and C. albicans over-growth following immunosuppressive drug and/or antibiotictreatments (15). It is important to understand the host mech-anisms involved in maintaining the natural microbiota andcontrolling potential pathogens, such as C. albicans, if we are tolearn how to prevent commensal organisms from becomingopportunistic infectious agents.

The innate immune system recognizes C. albicans and otherpathogens via pattern recognition receptors (PRRs) which inter-act with ligands termed pathogen-associated molecular patterns(PAMPs). Of the four major families of PRRs, the Toll-like recep-tor (TLR) family is the best studied, with TLR2 and TLR4, amongothers, being involved in fungal recognition (1). However, anti-fungal immunity appears to be mediated primarily by members ofthe C-type lectin receptor (CLR) family, including dectin-1 anddectin-2 (7). The �-glucan receptor dectin-1, in particular, plays anonredundant role in host defense against a number of medicallyimportant fungal species, including C. albicans, Aspergillus fu-migatus, and Pneumocystis jirovecii (24, 26, 27). Dectin-1 is ex-pressed primarily on myeloid cells (neutrophils, macrophages,and dendritic cells) and is strongly expressed on immune cells inthe lamina propria of the gastrointestinal tract (21). In vitro ex-periments have shown that dectin-1 mediates a number of cellularfunctions, including phagocytosis, the respiratory burst, and theproduction of inflammatory mediators, including cytokines,chemokines, and lipids (6). In vivo studies in mice have demon-strated that dectin-1 is vital for host defense in models of systemiccandidiasis (26), although this can be fungal strain dependent (6,24). In humans, dectin-1 deficiency is associated with a predispo-sition to mucocutaneous candidiasis, as well as increased gastro-

intestinal (GI) colonization in immunocompromised individuals,highlighting the importance of this receptor in human mucosaldefense (9, 20).

Our previous work suggested that dectin-1 was required toprevent GI tract infection with C. albicans during systemic infec-tion in mice (26). In these studies, dectin-1 deficiency was associ-ated with increased fungal burdens and gross morphologicalchanges of the GI tract, including enlargement of the stomach anddiscoloration of the small intestines (26). Here we further inves-tigated the role of dectin-1 in controlling C. albicans GI infectionsin both systemic and oral infections.

MATERIALS AND METHODSMice. Eight- to 12-week-old female C57BL/6 and dectin-1�/� mice (26)backcrossed nine times onto a C57BL/6 background were bred and main-tained at the Medical Research Facility, University of Aberdeen. Mice wereseparately housed in groups in individually ventilated cages, unless oth-erwise indicated, and provided with food and water ad libitum. All exper-iments were repeated at least two times with five or more mice per timepoint. All experimentation conformed to the terms and conditions ofUnited Kingdom Home Office licenses for research on animals and theUniversity of Aberdeen ethical review committee.

C. albicans strains, culture media, and growth conditions. C. albi-cans strains SC5314 (26) and AM2003-013 and AM2003-016 (19) wereroutinely grown and maintained on yeast extract-peptone-dextrose(YPD) agar (Sigma-Aldrich). For inoculum preparation, a single colony

Received 26 May 2012 Returned for modification 13 June 2012Accepted 8 September 2012

Published ahead of print 17 September 2012

Editor: G. S. Deepe, Jr.

Address correspondence to Gordon D. Brown, gordon.brown@abdn.ac.uk.

D.M.M. and G.D.B. contributed equally to this article.

Supplemental material for this article may be found at http://iai.asm.org/.

Copyright © 2012, American Society for Microbiology. All Rights Reserved.

doi:10.1128/IAI.00559-12

The authors have paid a fee to allow immediate free access to this article.

4216 iai.asm.org Infection and Immunity p. 4216–4222 December 2012 Volume 80 Number 12

on February 13, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

was grown in Sabouraud broth (Oxoid) at 30°C for 24 h with shaking.Cells were washed twice in sterile phosphate-buffered saline (PBS) andcounted using a hemocytometer. Cell density was adjusted to the desiredinoculum level with PBS, and was confirmed by viable cell counts on agarplates.

Gastrointestinal model. The gastrointestinal model was essentially asdescribed previously (15) (see Fig. 4A). To reduce the commensal bacte-rial and endogenous fungal microbiota, mice were provided with sterilewater containing 2 mg/ml streptomycin (Invitrogen), 2,000 U/ml penicil-lin (Invitrogen), and 0.25 mg/ml fluconazole (Enzo) for 3 days and thenswitched to water containing streptomycin and penicillin for a further 24h. Mice were then provided with sterile water containing 1 � 107 CFU/mlC. albicans, 2 mg/ml streptomycin, and 2,000 U/ml penicillin for 5 days.After C. albicans exposure, mice were maintained on sterile water contain-ing 2 mg/ml streptomycin, 2,000 U/ml penicillin, and 0.2 mg/ml genta-micin (Invitrogen). In some experiments, no antibiotics were used at anytime point. To monitor colonization, stools were collected from individ-ual mice on day 5 postinfection and every 2 days thereafter. Stools werehomogenized in 1 ml PBS and serially diluted, and 25 �l of each dilutionwas plated on YPD agar containing 0.01 mg/ml vancomycin (Sigma) and0.1 mg/ml gentamicin. Plates were incubated overnight at 37°C underaerobic conditions and fungal content determined by viable cell count andexpressed as CFU per g. Mice were sacrificed at days 7, 14, and 21 afterexposure to C. albicans. Kidney, stomach, small intestine, cecum, andlarge intestine samples were harvested and washed 3 times with 1 ml sterilePBS to remove gut contents. Tissue weights were determined, and sampleswere transferred into tubes containing 0.5 ml PBS, 0.05% (vol/vol) TritonX-100, and complete mini-EDTA-free protease inhibitor cocktail (as perthe manufacturer’s instructions [Roche]). The tissues were then homog-enized, serially diluted, and plated on YPD as described above. Cell debriswas removed from the remaining tissue homogenates by centrifugation at13,000 rpm for 15 min at 4°C and stored at �80°C for subsequent cyto-kine analysis.

Systemic model. Mice were inoculated intravenously with 2 � 105

CFU C. albicans SC5314 in 100 �l sterile PBS via the lateral tail vein. Micewere monitored daily and were sacrificed at day 3 postinfection or whenjudged to be moribund. Tissues were collected and processed for fungalburden and cytokine analysis as described above.

Cytokine analysis. Cytokine levels were measured using the Bio-PlexPro Mouse 23-Plex kit (Bio-Rad) and analyzed on the Bio-Plex systemusing Bio-Plex Manager software as per the manufacturer’s instructions.Stored tissue sample homogenate supernatants were thawed and centri-fuged for 15 min at 13,000 rpm at 4°C to remove debris. For each test, 50�l of undiluted sample was used, and cytokine concentrations were nor-malized to sample protein concentrations (bicinchoninic acid [BCA] pro-tein assay kit; Pierce).

MPO activity. Myeloperoxidase (MPO) activity in stored tissue super-natants was determined using a myeloperoxidase activity assay kit(Abcam) as per the manufacturer’s instructions.

Bile acid analysis. Small intestinal contents were centrifuged at 13,000rpm. Twenty microliters of the supernatant was serially diluted and ana-lyzed for bile acid content using the Diazyme colorimetric total bile acidassay kit as per the manufacturer’s instructions.

Histology. Kidney, stomach, small intestine, cecum, and large intes-tine samples were removed from uninfected mice and infected mice,snap-frozen in OCT (Sakura), and sectioned. Sections were dehydratedwith xylene, rehydrated through a series of different ethanol solutions,and stained with hematoxylin and eosin (H&E) or periodic acid-Schiff(PAS) stain using conventional staining methods. All individual segmentsof the alimentary tract were evaluated for the presence and intensity ofinflammation and for the presence of fungi.

Statistical analysis. The two-tailed Student t test was used to comparethe two groups of mice. The Mann-Whitney test was used to comparenonnormally distributed data, as determined by the D’Agostino-Pearsonomnibus normality test. Survival data were assessed by the log rank test.

All statistical analyses were performed with GraphPad Prism softwareversion 5.04.

RESULTSDectin-1-deficient mice are more susceptible to systemic C. al-bicans infection. We had previously observed that dectin-1 defi-ciency correlated with GI tissue colonization during systemicinfection with C. albicans (26). To further explore this phenomenon,we systemically infected C57BL/6 wild-type and dectin-1-deficientmice with 2 � 105 CFU C. albicans. This dose led to significantlyenhanced mortality of the receptor-deficient mice, as we had ob-served previously in the 129SvEv strain background (26) (Fig. 1A).We examined stool fungal burdens over time (Fig. 1B and C) andassessed tissue fungal burdens at day 3 postinfection (Fig. 1D), atime point chosen as it was prior to the onset of mortality in thedectin-1�/� mice. Within 2 days postinfection, we observed sig-nificantly higher fungal burdens in the stools of the dectin-1-de-ficient animals, which further increased by day 3. The identity ofthe fungi in the stools was confirmed as C. albicans by 18S rDNAsequencing (data not shown). Furthermore, stools plated fromuninfected control animals showed no growth of fungal colonies(data not shown). Interestingly, although wild-type mice also hadC. albicans present in their stools, these fungal burdens did notchange during the course of the infection (Fig. 1C). Similarly, onday 3, infected dectin-1-deficient animals displayed significantlyhigher fungal burdens in the kidney and GI tissues than wild-typeanimals, similar to what we had observed previously (26) (Fig.1D). Thus, dectin-1 deficiency allows enhanced C. albicans infec-tion of the GI tract.

FIG 1 Dectin-1 deficiency increases susceptibility of the GI tract during sys-temic infection with C. albicans SC5314. (A) Dectin-1�/� (n � 9) and wild-type (wt) (n � 10) mice were systemically infected with 2 � 105 CFU of C.albicans SC5314 and monitored for survival over a period of 21 days, as de-scribed in Materials and Methods. (B) Cartoon representation of the systemicexperimental infection model with sampling points. (C) Stool fungal burdensin dectin-1�/� (n � 21) and wt (n � 22) mice following systemic infectionwith 2 � 105 CFU C. albicans SC5314. (D) Fungal burdens in GI tissues indectin-1�/� (n � 21) and wt (n � 22) mice on day 3 postinfection. Barsrepresent mean values. All data are representative of at least two independentexperiments. *, P � 0.05.

Role of Dectin-1 in Gastrointestinal Candidiasis

December 2012 Volume 80 Number 12 iai.asm.org 4217

on February 13, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

Dectin-1-deficient mice display dysregulated cytokine re-sponses in the GI tract following systemic infection. To investi-gate the mechanism(s) underlying the increased susceptibility of dec-tin-1�/� mice to systemic GI infection, we measured cytokine levelsin gastrointestinal tissues (Fig. 2). In the stomachs of infected dectin-1�/� mice we observed significantly increased levels of interleukin-6(IL-6), granulocyte colony-stimulating factor (G-CSF), gamma in-terferon (IFN-�), and MIP-1� (CCL4), and in the large intestine andcecum we observed increased levels of G-CSF and KC (CXCL1). In-terestingly, in the latter two tissues we also observed significantly re-duced levels of tumor necrosis factor (TNF) in the dectin-1�/� mice,and there were significantly reduced IL-1� levels in the large intestine.Thus, these data show that dectin-1 deficiency results in dysfunc-tional cytokine responses in the GI tissues during systemic infectionwith C. albicans and that dectin-1-mediated cytokine responses tothis pathogen are tissue specific.

Enhanced fungal burdens and dysregulated cytokine re-sponses do not correlate with generalized GI tissue inflamma-tion in dectin-1-deficient mice. The increased fungal burdensand dysregulated cytokine responses in the dectin-1�/� miceprompted us to examine the GI tissues for signs of pathology (Fig.3). Surprisingly, we observed no significant differences in inflam-matory cell recruitment or pathology in any of the tissues exam-ined (Fig. 3A). Consistent with these observations, no differencein the levels of myeloperoxidase (MPO) activity, a marker of neu-trophil influx, was detected in these tissues (Fig. 3B). PAS stainingfor fungi, however, did reveal localized invasive C. albicans lesionsin dectin-1�/� mice, which were not observed in the wild-typeanimals (Fig. 3C and data not shown).

We had previously observed that C. albicans infection of the GItract resulted in gross morphological changes, including enlargedstomachs and inflamed intestines (26). While we did observe en-largement of the stomach upon infection, more extensive analysisrevealed that this phenotype was not limited to dectin-1�/� miceand that it did not occur in all infected animals (data not shown).

However, the small intestines of dectin-1�/� mice macroscopi-cally appeared to be reproducibly more inflamed than those of thewild-type mice, which did not correlate with our histological anal-ysis described above. On closer examination, the contents of thesmall intestines in dectin-1�/� mice appeared more yellow, pro-ducing an inflamed appearance (see Fig. S1 in the supplementalmaterial). As dysregulated production of bile salts, which have agreen/yellow appearance, has been observed previously in othermodels of inflammation (5), we examined the level of bile acids inthe contents of the small intestine and observed that these weresignificantly increased in the dectin-1�/� mice (Fig. 3D). Thus,these results indicate that dectin-1 deficiency results in an en-hanced, but localized, susceptibility to C. albicans infection in theGI tract. In addition, dectin-1 deficiency has broader effects, in-cluding dysregulated production of bile salts.

Dectin-1 deficiency does not alter susceptibility to C. albi-cans colonization of the GI tract during oral infection. As dectin-1-deficient mice were susceptible to GI tract colonization duringsystemic infections, we investigated the possibility that this CLRwas also involved in controlling colonization of these tissues by C.albicans following oral infection. Mice are not normally colonizedby this fungal pathogen, so for these experiments, we made use ofan established model of GI colonization (15), whereby mice aretreated with antibiotics to displace the endogenous microbiotaand then orally infected with C. albicans via the drinking water(Fig. 4A). C. albicans colonization levels in the gastrointestinaltract were subsequently monitored by examining fungal burdensin the stools and in the tissues at specific time points (Fig. 4A). Theestablished level of GI colonization (107 CFU/g of stool) wasunaffected by the dose of C. albicans administered in the drinkingwater (see Fig. S2 in the supplemental material).

Initially we obtained extremely variable results with these ex-periments, where dectin-1 deficiency had no effect or resulted ineither significantly higher or lower GI fungal burdens (data notshown; see Fig. S3 in the supplemental material). Notably, we also

FIG 2 Dectin-1 deficiency leads to abnormal cytokine production in the GI tract during systemic infection with C. albicans. Cytokine levels in GI tissues indectin-1�/� and wild-type (wt) mice on day 3 postinfection are shown. The data shown are pooled from two independent experiments (n � 12 combined) andare normalized to protein concentration. Bars represent mean values. *, P � 0.05.

Vautier et al.

4218 iai.asm.org Infection and Immunity

on February 13, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

observed variation in results between different cages of the samegroups of animals (data not shown). As the microbial composi-tion of the GI tract is transferable between mice and influenced bycolonization with C. albicans (8, 17), we performed all subsequentexperiments with cohoused wild-type and knockout animals. Un-der these conditions, we reproducibly found no difference in thecolonization levels of the dectin-1�/� mice, compared to wild-type animals, in both stool burdens (Fig. 4B) and GI tissue fungalburdens (Fig. 4C) at any of the time points examined. Interest-ingly, however, in both wild-type and dectin-1�/� mice, dissemi-nation of C. albicans to the kidneys was observed, although thishad substantially decreased by the end of the experiment in bothgroups of animals (Fig. 4C). Importantly, none of the animalssuccumbed to the infection during these experiments, even withincreased infective doses (up to 108 CFU/ml) or after extendedtime periods (over 60 days). Similar results were also obtained indectin-1-deficient mice in the 129SvEv background (data notshown).

Strains of C. albicans have different propensities in their ability

to colonize the mucosae, and SC5314, in particular, is a poor col-onizer of these tissues (18). Therefore, to confirm our observa-tions, we also tested the role of dectin-1 using two additional clin-ical strains of C. albicans (AM2003-013 and AM2003-016), whichwere originally isolated from infected mucosae (19). However, aswe found with SC5314, dectin-1 deficiency had no effect on GIcolonization with either of these clinical strains (see Fig. S4 in thesupplemental material).

We have recently shown that dectin-1 deficiency can exacer-bate colitis and that this phenotype was dependent on the compo-sition of the microbiome (14). To exclude the possibility that al-terations in the microbiome following antibiotic treatment ofmice were masking a potential contribution of dectin-1, we alsoexamined GI colonization in mice which did not receive any an-tibiotics (see Fig. S5 in the supplemental material). In these exper-iments, C. albicans was rapidly cleared from the GI tract, but, as wehad observed in the antibiotic-treated mice, there was no effect ofdectin-1 deficiency. Interestingly, while C57BL/6 background

FIG 3 Dectin-1 deficiency does not have a generalized effect on gastrointestinal histopathology during systemic infection with C. albicans. (A) Histopathologyof GI tissues in dectin-1�/� and wild-type (wt) mice as shown by H&E staining. (B) Myeloperoxidase (MPO) activity in the stomach (stom), small intestine(s. int), cecum (caec.), and large intestine (l. int). (C) A representative localized lesion observed in the stomachs of dectin-1�/� mice, as visualized by PAS staining.(D) Bile acid concentrations in the small intestines of uninfected wild-type (n � 4), infected wild-type (n � 6), and dectin-1�/� mice (n � 6), as indicated. Allobservations were performed on day 3. See also Fig. S1 in the supplemental material.

Role of Dectin-1 in Gastrointestinal Candidiasis

December 2012 Volume 80 Number 12 iai.asm.org 4219

on February 13, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

mice rapidly cleared the infection, 129/Sv mice retained a low, butpersistent, level of infection throughout the course of the experi-ment (see Fig. S5 in the supplemental material).

Dectin-1 deficiency does not affect cytokine responses or tis-sue pathology during C. albicans GI colonization following oralinfection. Given our findings in the systemic model, we also exam-ined cytokine responses and histopathology in the gastrointestinaltissues from orally infected wild-type and dectin-1�/� mice (Fig. 5).Cytokine responses were examined at day 14 postinfection. Althoughwe observed increased levels of cytokines known to be important incontrolling fungal infections, including IL-17 and IL-22 (3), particu-larly in stomach tissue, there were no significant differences betweenwild-type and dectin-1-deficient animals (Fig. 5A). The one excep-

tion was IL-4, which was slightly, but significantly, higher in the stom-achs of dectin-1�/� mice. Histological examination of the GI tractalso revealed no difference between the two groups, at either day 7, 14,or 21 postinfection (Fig. 5B; see Fig. S6 in the supplemental material).Localized fungal invasion of the stomach, with resulting hyperplasiaof the squamous epithelium, was observed in both groups at days 7,14, and 21 postinfection (see Fig. S7 in the supplemental material).Thus, these data indicate that dectin-1 is not involved in controllingC. albicans colonization of the GI tract following oral infection.

DISCUSSION

C-type lectin receptors play a central role in host defense, mediat-ing and directing both innate and adaptive antifungal immunity

FIG 4 Dectin-1 deficiency does not influence gastrointestinal tract colonization by C. albicans following oral infection. (A) Cartoon representation of theexperimental oral infection model with sampling time points. (B) Stool fungal burdens in cohoused dectin-1�/� (n � 21) and wild-type (wt) (n � 20) micefollowing oral infection with C. albicans SC5314. (C) GI tissue fungal burdens in dectin-1�/� and wt mice on days 7, 14, and 21 (pooled data, n 16 per group).Bars represent mean values. See also Fig. S3 in the supplemental material.

Vautier et al.

4220 iai.asm.org Infection and Immunity

on February 13, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

(1, 28). Dectin-1, in particular, has an essential role in the defenseagainst a number of fungal pathogens, including C. albicans (26).Our previous work characterizing the role of this receptor hadindicated that dectin-1 was required for controlling infection ofthe gastrointestinal tissues during systemic candidiasis (26). Herewe confirmed these findings and demonstrated that loss of dec-tin-1 led to dysregulated cytokine responses and uncontrolledfungal growth in the GI tract at early time points postinfection.These results are consistent with our previously established rolefor dectin-1 in mediating protective innate responses to C. albi-cans, including cytokine production and fungal uptake and killing(26). Interestingly, whereas we had previously not observed anydifferences in cytokine levels in infected kidneys at this early timepoint (day 3) (26), we could show that dectin-1 was involved inregulating cytokine responses in the GI tract, although this wasvaried and tissue specific. Surprisingly, the dectin-1-dependentdefects were observed only in localized lesions and did not corre-late with generalized changes in tissue pathology or inflammation(including neutrophil recruitment). However, we did reproduc-ibly observe gross abnormalities of the small intestines of dectin-

1�/� animals, which were not reflected in histopathologicalchanges. Rather, these changes correlated with substantially in-creased production of bile acids; however, the underlying reasonsfor this require further exploration.

We also examined the role of dectin-1 in gastrointestinal col-onization by C. albicans, using an oral model of infection. Surpris-ingly, despite the essential requirement of this receptor duringsystemic disease, we observed no differences in GI tract coloniza-tion between wild-type and dectin-1-deficient animals. Notably,we achieved reproducible results only when the wild-type anddectin-1-deficient mice were housed together in the same cage.When the animals were caged separately, we obtained variableresults; in some experiments dectin-1 deficiency conferred sus-ceptibility, whereas in others it conferred enhanced resistance. Infact, we even observed differences between different cages of thesame strain. Although not formally demonstrated here, it is likelythat this variation is due to differences in the microbiotas of themice, which is known to be transferrable between animals and toinfluence colonization by C. albicans (8, 17). The lack of an in-volvement of dectin-1 in controlling GI colonization under

FIG 5 Dectin-1 deficiency does not influence cytokine responses or pathology of the gastrointestinal tract during colonization by orally administered C. albicans.(A) Cytokine levels in GI tissues in dectin-1�/� and wild-type (wt) mice (n � 5 per group). Bars represent mean values. *, P � 0.05. (B) Representative H&Estaining of GI tissues in dectin-1�/� and wild-type mice. All observations were performed on day 14. See also Fig. S4 and S5 in the supplemental material.

Role of Dectin-1 in Gastrointestinal Candidiasis

December 2012 Volume 80 Number 12 iai.asm.org 4221

on February 13, 2018 by guest

http://iai.asm.org/

Dow

nloaded from

steady-state conditions may reflect the expression of dectin-1 inthe lamina propria and not the epithelium (22). However, dec-tin-1 expression can be induced in epithelial tissues (25), and thisreceptor may play a role under certain inflammatory conditions,such as we have recently shown during colitis (14).

Our results differ from two recent reports which have impli-cated dectin-1 in the control of C. albicans infections in the GItract (2, 10). In the first study, mice with dectin-1-deficient mac-rophages (and neutrophils) presented with increased fungal bur-dens in the stomach and cecum following oral infection with C.albicans (10). In the second study, dectin-1 was found to contrib-ute to either resistance or susceptibility to infection, depending onmouse strain background, which was related to the dectin-1 iso-form expressed in these animals (2, 11). However, the infectionmodels used were significantly different from our own; in bothstudies, mice were not pretreated with antibiotics and only a singleinfective dose was administered (2, 10). However, we also foundno role for dectin-1 in mice which had not been treated withantibiotics. The differences in susceptibility observed in these ex-periments may also reflect the lack of cohousing between experi-mental groups and/or the different sources of these animals. In-deed, mice from different locations can differ substantially in theirmicrobiota (4).

While our results suggest that dectin-1 is not involved in con-trolling the colonization of the GI tract, this receptor is requiredfor controlling C. albicans infection at other mucosal sites. Inmouse models, dectin-1 is necessary for controlling infection inboth the oral and vaginal mucosae (2, 13). Furthermore, in hu-mans, individuals homozygous for a polymorphism which ren-ders them essentially dectin-1 deficient have enhanced suscepti-bility to chronic mucocutaneous candidiasis (CMC), whichaffects the skin, nail beds, and oral and vaginal mucosae (9). Im-portantly, these patients have defects in Th17 responses, which arecritical for the control of mucosal Candida infections (12). Con-sistent with these observations, we did not observe any dectin-1-dependent effects on the production of IL-17 or IL-22 in the GItract, and it is likely that other CLRs, such as dectin-2, may play themajor role in driving these responses in these tissues (16, 23).However, preliminary analysis suggests that dectin-2 is also notinvolved in controlling GI colonization following oral infection(see Fig. S8 in the supplemental material).

In conclusion, our data demonstrate that the innate responsestriggered by dectin-1 play an essential role in controlling C. albi-cans infection of the GI tract during systemic infection. In con-trast, this receptor appears to have little, if any, role in controllingcolonization of the GI tract following oral infection with this or-ganism. Elucidating the mechanisms involved in this process re-mains a priority if we are to fully understand how our immunesystem controls and responds to colonization by a pathogen whichis a significant cause of human morbidity and mortality.

ACKNOWLEDGMENTS

This study was supported by the University of Aberdeen and the Well-come Trust.

We thank S. Vicky Tsoni for performing the 129SvEv systemic infec-tions, Yoichiro Iwakura for the dectin-2 mice, and Julie Taylor and thestaff of our animal facility for the care and maintenance of our animals.

REFERENCES1. Brown GD. 2011. Innate antifungal immunity: the key role of phagocytes.

Annu. Rev. Immunol. 29:1–21.

2. Carvalho A, et al. 2012. Dectin-1 isoforms contribute to distinct Th1/Th17 cell activation in mucosal candidiasis. Cell Mol. Immunol. 9:276 –286.

3. Conti HR, Gaffen SL. 2010. Host responses to Candida albicans: Th17cells and mucosal candidiasis. Microbes Infect. 12:518 –527.

4. Denning TL, et al. 2011. Functional specializations of intestinal dendriticcell and macrophage subsets that control Th17 and regulatory T cell re-sponses are dependent on the T cell/APC ratio, source of mouse strain,and regional localization. J. Immunol. 187:733–747.

5. Dial EJ, Romero JJ, Villa X, Mercer DW, Lichtenberger LM. 2002.Lipopolysaccharide-induced gastrointestinal injury in rats: role of surfacehydrophobicity and bile salts. Shock 17:77– 80.

6. Drummond RA, Brown GD. 2011. The role of Dectin-1 in the hostdefence against fungal infections. Curr. Opin. Microbiol. 14:392–399.

7. Drummond RA, Saijo S, Iwakura Y, Brown GD. 2011. The role ofSyk/CARD9 coupled C-type lectins in antifungal immunity. Eur. J. Im-munol. 41:276 –281.

8. Elinav E, et al. 2011. NLRP6 inflammasome regulates colonic microbialecology and risk for colitis. Cell 145:745–757.

9. Ferwerda B, et al. 2009. Human dectin-1 deficiency and mucocutaneousfungal infections. N. Engl. J. Med. 361:1760 –1767.

10. Gales A, et al. 2010. PPARgamma controls dectin-1 expression requiredfor host antifungal defense against Candida albicans. PLoS Pathog.6:e1000714. doi:10.1371/journal.ppat.1000714.

11. Heinsbroek SE, et al. 2006. Expression of functionally different Dectin-1isoforms by murine macrophages. J. Immunol. 176:5513–5518.

12. Hernandez-Santos N, Gaffen SL. 2012. Th17 cells in immunity to Can-dida albicans. Cell Host Microbe 11:425– 435.

13. Hise AG, et al. 2009. An essential role for the NLRP3 inflammasome inhost defense against the human fungal pathogen Candida albicans. CellHost Microbe 5:487– 497.

14. Iliev ID, et al. 2012. Interactions between commensal fungi and theC-type lectin receptor Dectin-1 influence colitis. Science 336:1314 –1317.

15. Koh AY, Kohler JR, Coggshall KT, Van Rooijen N, Pier GB. 2008.Mucosal damage and neutropenia are required for Candida albicans dis-semination. PLoS Pathog. 4:e35. doi:10.1371/journal.ppat.0040035.

16. Leibundgut-Landmann S, et al. 2007. Syk- and CARD9-dependent cou-pling of innate immunity to the induction of T helper cells that produceinterleukin 17. Nat. Immunol. 8:630 – 638.

17. Mason KL, et al. 2012. Interplay between the gastric bacterial microbiotaand Candida albicans during postantibiotic recolonization and gastritis.Infect. Immun. 80:150 –158.

18. Naglik JR, Fidel PL, Jr, Odds FC. 2008. Animal models of mucosalCandida infection. FEMS Microbiol. Lett. 283:129 –139.

19. Odds FC, et al. 2007. Molecular phylogenetics of Candida albicans. Eu-karyot. Cell 6:1041–1052.

20. Plantinga TS, et al. 2009. Early stop polymorphism in human DECTIN-1is associated with increased candida colonization in hematopoietic stemcell transplant recipients. Clin. Infect. Dis. 49:724 –732.

21. Reid DM, Gow NA, Brown GD. 2009. Pattern recognition: recent in-sights from Dectin-1. Curr. Opin. Immunol. 21:30 –37.

22. Reid DM, et al. 2004. Expression of the �-glucan receptor, Dectin-1, onmurine leukocytes in situ correlates with its function in pathogen recog-nition and reveals potential roles in leukocyte interactions. J. Leukoc. Biol.76:86 –94.

23. Robinson MJ, et al. 2009. Dectin-2 is a Syk-coupled pattern recognitionreceptor crucial for Th17 responses to fungal infection. J. Exp. Med. 206:2037–2051.

24. Saijo S, et al. 2007. Dectin-1 is required for host defense against Pneumo-cystis carinii but not against Candida albicans. Nat. Immunol. 8:39 – 46.

25. Sun WK, et al. 2012. Dectin-1 is inducible and plays a crucial role inAspergillus-induced innate immune responses in human bronchial epithe-lial cells. Eur. J. Clin. Microbiol. Infect. Dis. 31:2755–2764.

26. Taylor PR, et al. 2007. Dectin-1 is required for beta-glucan recognitionand control of fungal infection. Nat. Immunol. 8:31–38.

27. Werner JL, et al. 2009. Requisite role for the dectin-1 beta-glucan recep-tor in pulmonary defense against Aspergillus fumigatus. J. Immunol. 182:4938 – 4946.

28. Wuthrich M, Deepe GS, Jr, Klein B. 2012. Adaptive immunity to fungi.Annu. Rev. Immunol. 30:115–148.

29. Zaoutis TE, et al. 2005. The epidemiology and attributable outcomes ofcandidemia in adults and children hospitalized in the United States: apropensity analysis. Clin. Infect. Dis. 41:1232–1239.

Vautier et al.

4222 iai.asm.org Infection and Immunity

on February 13, 2018 by guest

http://iai.asm.org/

Dow

nloaded from