Differential Interaction of the Two Related Fungal Species albicans and Candida dubliniensis with Human Neutrophils This information is current as of September 26, 2021. Eliska Svobodová, Peter Staib, Josephine Losse, Florian Hennicke, Dagmar Barz and Mihály Józsi J Immunol 2012; 189:2502-2511; Prepublished online 30 July 2012; doi: 10.4049/jimmunol.1200185 Downloaded from http://www.jimmunol.org/content/189/5/2502

Supplementary http://www.jimmunol.org/content/suppl/2012/07/30/jimmunol.120018 http://www.jimmunol.org/ Material 5.DC1 References This article cites 67 articles, 20 of which you can access for free at: http://www.jimmunol.org/content/189/5/2502.full#ref-list-1

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision by guest on September 26, 2021

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2012 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

Differential Interaction of the Two Related Fungal Species and Candida dubliniensis with Human Neutrophils

Elisˇka Svobodova´,* Peter Staib,† Josephine Losse,* Florian Hennicke,† Dagmar Barz,‡ and Miha´ly Jo´zsi*

Candida albicans, the most common facultative human pathogenic is of major medical importance, whereas the closely related species Candida dubliniensis is less virulent and rarely causes life-threatening, systemic infections. Little is known, however, about the reasons for this difference in pathogenicity, and especially on the interactions of C. dubliniensis with the human immune system. Because innate immunity and, in particular, neutrophil granulocytes play a major role in host anti-

fungal defense, we studied the responses of human neutrophils to clinical isolates of both C. albicans and C. dubliniensis. Downloaded from C. dubliniensis was found to support neutrophil migration and fungal cell uptake to a greater extent in comparison with C. albicans, whereas inducing less neutrophil damage and extracellular trap formation. The production of antimicrobial reactive oxygen species, myeloperoxidase, and lactoferrin, as well as the inflammatory chemokine IL-8 by neutrophils was increased when stimulated with C. dubliniensis as compared with C. albicans. However, most of the analyzed macrophage-derived inflammatory and regulatory cytokines and chemokines, such as IL-1a,IL-1b,IL-1ra,TNF-a,IL-10,G-CSF,andGM-CSF, were less induced by C. dubliniensis. Similarly, the amounts of the antifungal immunity-related IL-17A produced by PBMCs http://www.jimmunol.org/ was significantly lower when challenged with C. dubliniensis than with C. albicans. These data indicate that C. dubliniensis triggers stronger early neutrophil responses than C. albicans, thus providing insight into the differential virulence of these two closely related fungal species, and suggest that this is, in part, due to their differential capacity to form hyphae. The Journal of Immunology, 2012, 189: 2502–2511.

ungi of the genus Candida are harmless commensals Despite their similarity, C. dubliniensis appears to be less virulent coexisting with humans. In the immunocompromised host, in animal and mucosal infection models (4–7) according to its in- F however, Candida spp. can cause superficial and even life- fection rate and the severity of disease. C. dubliniensis strains are by guest on September 26, 2021 threatening, systemic infections. Although Candida albicans is the more rapidly cleared form the site of infection and are less able to most common cause of candidiasis, the number of infections cause disseminated infections. The phylogenetic properties of C. caused by nonalbicans species such as Candida glabrata, Candida dubliniensis have been well described in the literature (2, 8). A DNA parapsilosis, and Candida dubliniensis, especially in HIV-infected hybridization array (9) and lately comparison of both genomes (10) and AIDS patients, is quite high (1). C. dubliniensis was first uncovered particular differences between C. dubliniensis and C. discovered in 1995 upon genetic characterization and is mainly albicans in regard to putative virulence genes, including those recovered from oral candidosis of HIV+ patients (2). C. dubliniensis encoding hyphae-associated factors such as specific secreted is the closest relative of C. albicans and is the only other Candida proteases or invasins. The ability of C. albicans to reversibly switch species able to form true hyphae and (3). between and filamentous growth forms has widely been docu- mented as an important virulence attribute of C. albicans, facilitating host invasion and colonization. C. dubliniensis also produces true *Junior Research Group Cellular Immunobiology, Leibniz Institute for Natural Prod- hyphae, however, with a far lesser efficiency than C. albicans,which uct Research and Infection Biology–Hans Kno¨ll Institute, D-07745 Jena, Germany; putatively contributes to its comparatively poor virulence (6, 11). † Junior Research Group Fundamental Molecular Biology of Pathogenic Fungi, Leib- The control of fungal infection in the human host critically niz Institute for Natural Product Research and Infection Biology–Hans Kno¨ll Insti- tute, D-07745 Jena, Germany; and ‡Institute for Transfusion Medicine, University depends on cells of the innate immunity. Neutrophil granulocytes Hospital of Jena, D-07747 Jena, Germany are major phagocytic cells that are recruited early to the infection Received for publication January 13, 2012. Accepted for publication July 3, 2012. site, where they activate a large repertoire of antimicrobial This work was supported in part by the Deutsche Forschungsgemeinschaft (DFG mechanisms including phagocytosis, the generation of reactive Grant STA 1147/1-1 to P.S.). oxygen species (ROS), the release of granule enzymes and anti- Address correspondence and reprint requests to Dr. Miha´ly Jo´zsi, Junior Research microbial peptides, and the formation of neutrophil extracellular Group Cellular Immunobiology, Leibniz Institute for Natural Product Research and traps (NETs) (12–14). Attracted monocytes possess less micro- Infection Biology–Hans Kno¨ll Institute, Beutenbergstrasse 11a, D-07745 Jena, Germany. E-mail address: [email protected] bicidal capacity but are important in alerting cells of the innate The online version of this article contains supplemental material. and adaptive immune system, and give rise to macrophages, im- Abbreviations used in this article: CFW, calcofluor white; DPBS, Dulbecco’s portant phagocytic and APCs. These innate immune cells recog- phosphate-buffered saline; LDH, lactate dehydrogenase; MDM, monocyte-derived nize fungal pathogens via three major receptor families (i.e., the macrophage; MOI, multiplicity of infection; MPO, myeloperoxidase; NET, neutro- TLRs, the lectin receptors, and the NOD-like receptors), leading phil extracellular trap; ROS, reactive oxygen species. to activation of signaling pathways that result in tightly regulated Copyright Ó 2012 by The American Association of Immunologists, Inc. 0022-1767/12/$16.00 and complementary immune responses (15, 16). www.jimmunol.org/cgi/doi/10.4049/jimmunol.1200185 The Journal of Immunology 2503

To date, little is known about the Candida species-specific sites, respectively, to result in pCdADH1E2. A SalI-BglII GFP frag- responses of human innate immune cells. Because C. dubliniensis ment from pNIM1 (28) was cloned in the XhoI-BglII digested plasmid has a reduced ability to cause severe (i.e., systemic) infections, pCdADH1E2, resulting in pCdADH1GFP1. Flanked by CdADH1 upstream and downstream sequences, pCdADH1GFP1 contains a cassette with the which are usually brought in context with C. albicans, one would GFP gene under control of PCdADH1,followedbyTCaACT1 termination expect a better clearance of the former by host immune cells. A sequences, and the SAT1 marker for transformant selection after trans- previous report found no difference in the phagocytosis, oxidative formation. Transformation by electroporation of C. dubliniensis Wu¨284 burst, or killing induced by these two fungal species in whole by use of the linear ApaI-SacII DNA cassette from pCdADH1GFP1, transformant selection on nourseothricin (Werner Bioagents, Jena, Ger- blood (17). Nevertheless, it has been shown that even though both many),andSouthernanalysisofgenomic DNA with ECL-labeled (GE Candida species release a chemotactic factor for neutrophils, Healthcare, Freiburg, Germany) upstream and downstream CdADH1 the chemotaxis toward C. dubliniensis is enhanced compared with probes for proving correct insertion of the GFP construct in the CdADH1 C. albicans (18). Studies using epithelial cell infection mod- locus of the resulting transformant CdADH1GFP1A were performed by els demonstrated that C. dubliniensis causes minimal invasion standard procedures as described previously (29). (5, 19) and cytokine production (20, 21) compared with C. Human cell culture and stimulation albicans, and that this effect is attributed to the ability of the latter to switch from yeast to hyphal morphology under cell culture Human neutrophil granulocytes were isolated from peripheral blood of healthy individuals as previously described (30). All blood donors gave conditions (21). informed consent. If not otherwise stated, neutrophils were cultured in It is important to understand their interaction with the host serum-free X-VIVO 15 medium (Lonza) at a concentration of 2.5 3 106 immune system in detail to decipher species-specific differences in cells/ml and cultivated at 37˚C in a humidified atmosphere containing the pathogenicity of these two closely related Candida species. 5% CO2. If not stated otherwise, neutrophils were incubated with Downloaded from Candida cells at a multiplicity of infection (MOI) of 1. Culture super- Therefore, we compared the activation of human neutrophil gran- natants were collected either after 1.5 h for ROS, myeloperoxidase ulocytes upon coincubation with C. albicans and C. dubliniensis (MPO), and lactoferrin measurement or after 20 h of coincubation for in vitro, using a set of previously characterized clinical isolates. IL-8 measurement. All stimulations for each donor were performed in We found that C. dubliniensis strains enhanced neutrophil mi- duplicates. gration and phagocytosis, as well as IL-8, ROS, and lactoferrin PBMCs were isolated by Ficoll-Hypaque (GE Healthcare) density gradient separation, and erythrocytes were lysed using a hypotonic salt release, but caused less extracellular trap formation and neutrophil solution. PBMCs were washed with DPBS and cultured in X-VIVO 15 http://www.jimmunol.org/ damage, in comparison with C. albicans strains. In addition, a medium containing 10% pooled normal human serum at a concentration of reduced ability of C. dubliniensis to stimulate the production of 5 3 106 cells/ml. For stimulation experiments, PBMCs were incubated in 3 4 IL-17 by mononuclear cells, a cytokine involved in neutrophil medium alone or in medium containing 1 10 live Candida cells/ml, and supernatants were collected after 7 d for IL-17 measurement. recruitment and activation, was observed. These results demon- Monocytes were obtained from PBMCs by positive selection with CD14 strating a differential interaction of the two Candida species with magnetic beads (Miltenyi Biotech, Bergisch Gladbach, Germany), fol- cells of the human innate immune system may, in part, explain the lowing the manufacturer’s instructions. In brief, PBMCs were resuspended differences in their virulence. in MACS buffer (DPBS with 0.5% BSA and 2 mM EDTA) containing CD14 microbeads. After incubation for 30 min at 4˚C, cells were washed and resuspended in MACS buffer and loaded onto MiniMACS LS Sepa- by guest on September 26, 2021 Materials and Methods ration Columns (Miltenyi Biotech). After washing out of nonlabeled cells + Candida strains with MACS buffer, monocytes were eluted and the purity of CD14 monocytes was determined by FACS. The C. albicans and C. dubliniensis wild-type strains that were used are Monocyte-derived macrophages (MDMs) were obtained by culturing listed in Table I. All strains were routinely maintained on yeast peptone isolated monocytes for 7 d in X-VIVO 15 medium supplemented with 10% dextrose agar (20 g peptone, 10 g yeast extract, 20 g glucose, 15 g agar per heat-inactivated FCS, 2 mM ultraglutamine (Lonza), and 50 mg/ml gen- liter). For experiments, cells were grown in yeast peptone dextrose medium tamicin sulfate (Lonza), and by addition of 500 IU/ml GM-CSF (Immu- at 30˚C with shaking for 16 h and washed twice in Dulbecco’s PBS notools, Friesoythe, Germany) every 48 h. MDMs were resuspended at 2 3 (DPBS; Lonza, Wuppertal, Germany) before use. In case of C. albicans, 106 cells/ml in serum-free X-VIVO 15 medium and cultivated at 37˚C in hypha formation was induced in RPMI 1640 medium containing L-gluta- a humidified atmosphere containing 5% CO2. Cells were infected with mine and 25 mM HEPES (Invitrogen, Karlsruhe, Germany) at 37˚C for 1 Candida at a MOI of 1 or 4, depending on experiment. Culture super- h. For C. dubliniensis, hypha formation was induced in water containing natants were collected 20 h later, and cytokine profiles were determined by 10% FCS (PAA Laboratories, Co¨lbe, Germany) at 37˚C for 1 h (11). The cytokine array and/or ELISA. hyphal morphology for both species was monitored microscopically to ensure a similar hyphal percentage and length. Neutrophil migration assays C. dubliniensis reporter strain construction Migration assays were performed as described previously (30). Neutrophils were labeled with calcein AM (Sigma-Aldrich, Taufkirchen, Germany), To visualize Candida cells during interaction with neutrophils, we also allowed to migrate toward Candida cells through a 3-mm-pore poly- used GFP-labeled fungal cells. For this purpose, C. albicans and C. dub- carbonate membrane of a transwell system (Corning Life Sciences, liniensis strains were used that express the GFP gene from the constitu- Amsterdam, The Netherlands) for 60 min and were quantified in the lower tively active C. albicans and C. dubliniensis ADH1 promoter, respectively. well. The C. albicans reporter strain SCADH1G4A carrying a PCaADH1-GFP fusion in one of the ADH1 alleles of the wild-type strain SC5314 had Neutrophil phagocytosis assay been described previously (26). Analogously, the C. dubliniensis strain 5 CdADH1GFP1A, which harbors a PCdADH1-GFP fusion in one of the Neutrophils (5 3 10 ) in X-VIVO 15 medium were added to coverslips, CdADH1 alleles of the wild-type strain Wu¨284, was constructed as follows. precoated with 0.01% poly-L-lysine, in 24-well plates and allowed to at- An ApaI-XhoI fragment with CdADH1 upstream sequences was amplified tach for 40 min at 37˚C. For some experiments, neutrophils were pretreated by PCR with the primers CdADH1-1 (59-ATATAGGGCCCCCGGGTT- with 15 mg/ml Cytochalasin D (Sigma-Aldrich) for 30 min at 37˚C before CCGTCATGCAAGCAAGC-39) and CdADH1-2 (59-ATATACTCGAGG- phagocytosis. Candida cells constitutively expressing GFP were added at TTTTTGTATTTGTATGACAATAACGTTTATTG-39). A PstI-SacII fragment a MOI of 2 and sedimented by centrifugation at 1400 rpm for 3 min. After containing CdADH1 downstream sequences was obtained by PCR with 40 min of coincubation at 37˚C and 5% CO2, plates were cooled on ice and the primers CdADH1-3 (59-GCTGAACCAAACTGCAGTGAAGCTGAC- nonphagocytosed Candida cells were stained with 10 mg/ml calcofluor TTG-39) and CdADH1-4 (59-GTGTCTTTTCTGTTACCGCGGTAAGA- white (CFW; fluorescent brightener 28; Sigma-Aldrich) for 20 min at room ACCTTTG-39). Genomic DNA from C. dubliniensis strain Wu¨284 was temperature. Cells were fixed with 4% paraformaldehyde (Carl Roth, used as a template for the PCR. The CdADH1 upstream and downstream Karlsruhe, Germany) for 15 min at 20˚C, coverslips were mounted on glass fragments were successively substituted for CaADH1 upstream and down- slides, and phagocytosis was analyzed using a 633 immersion oil objec- stream fragments in plasmid pADH1E1 (27) via the introduced restriction tive at room temperature with an LSM 710 confocal laser-scanning mi- 2504 NEUTROPHIL RESPONSES TO C. ALBICANS AND C. DUBLINIENSIS croscope running ZEN software (Carl Zeiss, Jena, Germany). At least four Results 3 3 of 3 3 microscopic fields (200 200 mm) were analyzed for the Human neutrophils migrate at a higher rate toward C. colocalization of GFP and CFW staining using an argon laser with a wavelength of 488 nm and corresponding filters. The percentage of dubliniensis phagocytosed cells was calculated from the GFP signal that does not Because both C. albicans and C. dubliniensis were shown to re- colocalize with the CFW signal. lease chemotactic factor(s) for neutrophils (18), we studied the Release of NETs migratory activity of neutrophils toward different strains of both Candida species (Table I) in a transwell assay. All studied strains Neutrophils were attached to coverslip as described earlier for the phagocytosis assay. Afterward, 1 3 106 GFP-expressing yeast cells or 100 of C. dubliniensis significantly enhanced the number of neutrophils nM PMA as a positive control were added. Fungal cells were sedimented that migrated through the porous membrane of the transwell by centrifugation at 1400 rpm for 3 min. The coverslips were then incu- system compared with the strains of C. albicans (Fig. 1). We did bated for 3 h at 37˚C in a humidified atmosphere containing 5% CO2. not observe significant differences in the growth rate between the Extracellular DNA was stained with 20 mg/ml propidium iodide (Sigma- studied strains in suspension cultures. During the 1-h incubation Aldrich) for 20 min at 37˚C. After washing gently with DPBS, the cov- erslips were placed on mounting solution on a glass slide, and NETs were time, the fungal cells did not multiply in the wells as monitored by analyzed using an LSM 710 confocal laser-scanning microscope (Carl microscopic analysis, and the C. albicans cells only started to Zeiss). For quantification of NETs, neutrophils were coincubated with germinate. Thus, the observed difference in supporting neutrophil Candida cells in white 96-well plates. Released DNA was stained with 5 migration was not due to difference in fungal mass among the mg/ml SYTOX Orange (Invitrogen), and the fluorescence was measured in a fluorescence reader (Tecan, Crailsheim, Germany) with emission and fungal species. absorption filters of 540 and 575 nm, respectively. C. dubliniensis is rapidly phagocytosed by human neutrophils Downloaded from Neutrophil damage assay To assess the phagocytosis of both Candida species, we coincu- Neutrophil granulocytes were seeded in a flat-bottom, 96-well plate at bated blood-derived human neutrophils with GFP-labeled yeast a concentration of 5 3 106 cells/ml. After addition of Candida cells at cells (C. albicans strain SCADH1G4A and C. dubliniensis strain a MOI of 1, the plate was incubated for 20 h at 37˚C and 5% CO2. For CdADH1GFP1A; for details, see Materials and Methods)ata control, neutrophils and Candida cells were incubated in medium only. MOI of 2. After 40 min, nonphagocytosed fungal cells were The plate was equilibrated to 20˚C, supernatants were collected in a white stained with CFW and the percentage of engulfed cells was de- http://www.jimmunol.org/ 96-well plate, and the lactate dehydrogenase (LDH) activity was analyzed ∼ using a fluorescence-based CytoTox-ONE Homogeneous Membrane In- termined microscopically. Under these conditions, 70% of the C. tegrity Assay kit (Promega, Mannheim, Germany) according to the man- dubliniensis cells were internalized compared with only ∼20% of ufacturer’s instructions. Experiments were performed in duplicates, and C. albicans cells (Fig. 2). As a negative control, neutrophils were the results are expressed as percentage of released LDH in the supernatant pretreated with Cytochalasin D before phagocytosis. Under this of lysed neutrophils. condition, almost no cells were internalized after the specified Release of ROS, MPO, and lactoferrin time point (data not shown). Neutrophils were coincubated in white 96-well plates (Greiner, Frick- Reduced ability of C. dubliniensis to induce NETs enhausen, Germany) in 200 ml medium/well with of the various C. by guest on September 26, 2021 albicans and C. dubliniensis strainsataMOIof1inX-VIVO15me- To analyze whether C. dubliniensis induces the production of dium at 37˚C. For some experiments, Candida cells were separated from extracellular traps by human neutrophils, similar to C. albicans neutrophils through a 0.4-mm-pore membrane in a 24 well-plate trans- (14), we coincubated GFP-expressing yeast cells of both Candida well system (Corning Life Sciences). ROS were measured after 90 min of coculture by incubating cells in X-VIVO 15 medium containing 10 mM dihydrorhodamine (Sigma-Aldrich) for 15 min at 37˚C followed by lysis of neutrophils with DPBS supplemented with 1% Triton X-100. Table I. C. albicans and C. dubliniensis clinical isolates used in this The fluorescence signal of the oxidized dihydrorhodamine was mea- study sured in a fluorescence reader (Tecan) with excitation and emission filters of 485 and 538 nm. MPO and lactoferrin were detected using a Strain Source Reference sandwichELISA.Inbrief,4mg/ml anti-MPO or anti-lactoferrin mAb (HyTest, Turku, Finland) was immobilized on microtiter plates (Nunc, C. albicans Wiesbaden, Germany) at 4˚C overnight. After blocking with 4% skimmed ATCC10261 ATCC Reference strain milk for 2 h, supernatants of cocultures, taken after 90 min, were added ATCC38696 ATCC Reference strain for 1 h at room temperature. Lactoferrin and MPO were detected using ATCC44807 ATCC Reference strain 50 ng/ml HRP-conjugated anti-lactoferrin IgG (antibodies-online, Aachen, ATCC44808 ATCC Reference strain Germany) and rabbit anti-MPO IgG (1:5000; HyCultec, Beutelsbach, DSM1386 DSM Reference strain Germany), followed by the corresponding secondary Ab. The ELISA SC5314 F. Mu¨hlschlegel 22 was developed using TMB PLUS substrate (Kem-En-Tec Diagnostics, WO-1 D. Soll 23 Taastrup, Denmark), and the absorbance was measured at 450 nm. MA13 Clinical isolate, Berlin 24 Wu¨212 Clinical isolate, Wu¨rzburg 24 Measurement of cytokine production Wu¨277 Clinical isolate, Wu¨rzburg 24 C. dubliniensis The production of IL-8 by human neutrophils was measured from H1 Clinical isolate, Berlin 25 supernatants of cocultures taken after 20 h using a commercial sandwich H12 Clinical isolate, Berlin 25 ELISA kit (ImmunoTools). IL-1b and IL-17 were quantified from culture VK9603-29 Clinical isolate, Berlin 25 supernatants of MDMs or PBMCs using commercial sandwich ELISA VK9603-34 Clinical isolate, Berlin 25 kits from ImmunoTools and eBioscience (Frankfurt, Germany), respec- VK9603-54 Clinical isolate, Berlin 25 tively. The amounts of other inflammation-related cytokines were VK9603-63 Clinical isolate, Berlin 25 measured after 24 h from the supernatants of Candida-macrophage Wu¨272 Clinical isolate, Wu¨rzburg 25 cocultures incubated at a MOI of 1 in X-VIVO 15 medium using Wu¨284 Clinical isolate, Wu¨rzburg 25 a Proteome Profiler Human Cytokine Array Kit (Panel A; R&D Systems, Wu¨294 Clinical isolate, Wu¨rzburg 25 Wiesbaden, Germany). Wu¨361 Clinical isolate, Wu¨rzburg 25 Wu¨367 Clinical isolate, Wu¨rzburg 25 Statistical analysis Wu¨370 Clinical isolate, Wu¨rzburg 25 Statistical analysis of the results was carried out using an unpaired two- ATCC, American Type Culture Collection; DSM, German collection of micro- tailed Student t test. A p value #0.05 was considered significant. organisms. The Journal of Immunology 2505

damage and escape from host phagocytes (31). The ability of C. dubliniensis to switch from the yeast to the hyphal form is re- duced, especially in a nutrient-rich environment (11), which can be found in the infection site in the human body. To test whether C. dubliniensis can cause damage to human neutrophils to the same extent as C. albicans, we determined the amounts of LDH released upon damage of the plasma membrane of the host cells, in culture supernatants after 20 h of coincubation of C. albicans cells (strains SC5314, ATCC38696, and ATCC44808) and C. dubliniensis cells (strains Wu¨272, Wu¨284, and H1) with neu- trophils. As a positive control, neutrophils were lysed, and as a negative control, LDH release was measured from neutrophils incubated in medium only. Even though C. dubliniensis was rapidly phagocytosed, the yeast cells were not able to switch into FIGURE 1. Enhanced neutrophil migration toward C. dubliniensis.Live hyphae as efficiently as C. albicans, as observed microscopically ∼ yeast cells were added to the lower chamber and calcein-labeled neutrophils (Supplemental Fig. 3), and caused only slight damage ( 20%) to to the upper chamber of the transwell system. The fluorescence of neu- neutrophils compared with the negative control representing dead trophils in the lower chamber was determined after 60 min of transmi- neutrophils after 20 h (∼16%). In contrast, all C. albicans strains gration. Transwells without yeasts were used as negative control. Results caused more damage (30–50% after 20 h) to human neutrophils Downloaded from shown are relative mean fluorescence intensities (MFI) 6 SD from at least (Fig. 5). Similarly, C. dubliniensis strains caused less neutrophil five experiments with different blood donors. The migration induced by damage compared with C. albicans strains when the fungal cells C. albicans strain SC5314 was set at 100%. The negative control was were added after 1-h hypha induction (data not shown). subtracted, and mean value of the population of both species was derived from values of all strains in this population. **p # 0.01, Student t test. Increased release of antimicrobial substances and IL-8 by

neutrophils stimulated with C. dubliniensis versus C. albicans http://www.jimmunol.org/ species with neutrophils for 3 h at 37˚C at a MOI of 2. Confocal laser-scanning microscopy of cultures stained with propidium io- Antimicrobial compounds and cytokines produced by the innate dide revealed that C. dubliniensis caused only little NET formation, immune cells are important factors that affect the outcome of whereas NETs were readily visible for neutrophils challenged with fungal infections. We therefore analyzed the production and/or C. albicans (Fig. 3). Neutrophils incubated without fungi did not release of ROS, MPO, lactoferrin, and the proinflammatory cy- produce NETs. As a positive control, NET formation was induced tokine IL-8 by human neutrophils coincubated with C. albicans or using PMA (data not shown). To quantify NET formation, we C. dubliniensis. Supernatants from neutrophils incubated in me- analyzed the DNA content in the supernatants of cocultures of dium only or with PMA (potent neutrophil stimulator) were used Candida cells with neutrophils using SYTOX Orange. Neutrophils as a negative and positive control, respectively. All analyzed by guest on September 26, 2021 efficiently formed NETs by 3 h after stimulation with C. albicans strains of C. dubliniensis caused an increased release of ROS, strains. However, significantly less extracellular DNA was detected MPO, lactoferrin, and IL-8 when compared with the strains of C. from neutrophil cultures stimulated with C. dubliniensis strains albicans (Table II). (Fig. 4), confirming the microscopy data. We have repeated this To analyze in more detail whether the morphology of the Candida assay with fungal cells for which hypha formation was induced for cells per se differentially influences this phenotype, hypha forma- 1 h at 37˚C before addition to the neutrophils (Supplemental Fig. 1). tion was induced for C. albicans strains SC5314, ATCC38696, C. dubliniensis strains again caused less NET formation compared ATCC44808, and WO-1, and for C. dubliniensis strains H1, with C. albicans strains (data not shown). Notably, after several Wu¨272, Wu¨284, and VK9603-29 for 1 h at 37˚C before incubation hours of coincubation with neutrophils, most C. dubliniensis cells with human neutrophils (Supplemental Figs. 1, 2). These fungal did not maintain the hyphal morphology (Supplemental Fig. 2), as cells caused an increased ROS and lactoferrin production compared also reported by Moyes et al. (21). with yeast cells; nevertheless, the elevation was more prominent in the case of C. albicans than C. dubliniensis (Fig. 6). The ability of C. dubliniensis to damage neutrophils is reduced Further, we determined whether the stimulation of neutrophils, One of the pathogenic attributes of C. albicans is the production and thus the induction of lactoferrin production, is dependent on and elongation of hyphae, through which the pathogen is able to a direct contact between Candida and neutrophils or is caused by

FIGURE 2. Phagocytosis of C. albicans and C. dubliniensis by human neutrophils. Blood-derived neutrophils were coincubated with GFP-expressing C. albicans and C. dubliniensis cells for 40 min at a MOI of 2. Extracellular fungal cells were stained with CFW and analyzed at 3630 magnification in a laser-scanning microscope. (A) Pictures represent overlay images of green (internalized fungal cells) and blue (nonphagocytosed fungal cells) fluores- cence channels with bright field. Scale bar, 10 mm. (B) The percentage of internalized fungal cells was quantified microscopically. Results shown are percentage of phagocytosed cells + SD from at least five experiments with different blood donors. ***p # 0.001, Student t test. 2506 NEUTROPHIL RESPONSES TO C. ALBICANS AND C. DUBLINIENSIS

FIGURE 3. Generation of NETs on stimulation with C. albicans and C. dubliniensis. Neutrophils were coincubated with GFP-expressing C. albicans or C. dubliniensis cells (green fluorescence) for 3 h at 37˚C. NETs were examined at 3630 mag- nification using propidium iodide to stain the DNA (red fluorescence). Images represent laser- scanning microscopy fluorescence, bright field, and overlay images. Scale bars, 10 mm. Images are representative of four experiments.

Candida products acting on neutrophils. For this purpose, we Because the production of IL-17, an important cytokine in host compared the neutrophil response when coincubated with Can- antifungal defense and a potent activator of neutrophils (32), is also Downloaded from dida cells directly and when separated by a porous membrane to mediated through the action of the proinflammatory macrophage- allow exchange of medium and products. Whereas C. albicans derived IL-1b, we further analyzed the IL-1b production. MDMs caused only a slightly elevated lactoferrin production upon direct were coincubated with yeast cells of C. albicans strains (SC5314 contact compared with separated cells, the induction of lacto- and ATCC38696) and C. dubliniensis strains (Wu¨272 and Wu¨284) ferrin production by C. dubliniensis was largely dependent on the for 20 h at a MOI of 4, and IL-1b was measured from the culture

Candida–neutrophil contact (Fig. 7). supernatants by ELISA. Both strains of C. albicans triggered http://www.jimmunol.org/ significantly higher IL-1b production than did the C. dubliniensis C. dubliniensis stimulates less production of fungal strains (Fig. 9), which remained in yeast form, compared with the immunity-related cytokines hypha-forming C. albicans (Supplemental Fig. 3). Because macrophages are important innate immunity cells that In addition, we have compared the amounts of IL-17A produced recruit additional cell types to the site of infection, we measured by PBMCs upon stimulation with C. albicans and C. dubliniensis. chemokines and cytokines from MDMs after coincubation with PBMCs at a concentration of 5 3 106/ml were coincubated with both Candida species for 24 h at a MOI of 1 using a commercially 1 3 104/ml yeast cells of C. albicans strains (SC5314, ATCC38696, available inflammation array. C. dubliniensis and C. albicans in- ATCC44808, and WO-1) and C. dubliniensis strains (H1, Wu¨272, duced several inflammatory molecules under these experimental Wu¨284, and VK9603-29) for 7d, and IL-17A was measured from by guest on September 26, 2021 conditions in comparison with MDMs incubated without fungal the supernatants by ELISA. C. albicans stimulated a higher IL- cells (Fig. 8). Whereas some molecules were detected in similar 17A production in PBMCs compared with C. dubliniensis (Fig. amounts for both C. dubliniensis and C. albicans, such as C5/C5a, 10). This assay was also performed with fungal cells stimulated growth-regulated oncogene-a (also known as CXCL1), RANTES for 1 h at 37˚C to produce hyphae. Again, C. albicans cells caused (CCL5), and MIP-1a (CCL3), a markedly reduced induction of G- more IL-17A release than C. dubliniensis cells, and the levels of CSF, GM-CSF, IL-1a, IL-1b, IL-1ra, IL-10, IL-16, macrophage IL-17A were increased for both fungal species compared with the migration inhibitory factor, serine protease inhibitor 1, TNF-a, yeast cells (Fig. 10). and soluble triggering receptor expressed on myeloid cells was detected when MDMs were exposed to C. dubliniensis. In con- Discussion trast, the chemoattractant IFN-g–induced protein 10 (also known C. albicans is the most virulent species of the genus Candida, and as CXCL10) was more strongly induced by the two tested causes not only superficial but also life-threatening systemic C. dubliniensis strains (Wu¨272 and Wu¨284) compared with infections. C. albicans and C. dubliniensis are very closely related C. albicans SC5314 (Fig. 8). and share numerous phenotypes; intriguingly, however, systemic

FIGURE 4. C. dubliniensis causes less extracellular trap formation by neutrophils. The release of NETs was quantified after 3 h of coincubation with Candida cells or PMA as a positive control. The fluorescence of released DNA stained with SYTOX Orange was mea- sured. C. dubliniensis strains (H1, Wu¨272, Wu¨284, and VK9603-29; white bars) caused less NET release than the C. albicans strains (SC5314, ATCC38696, ATCC44808, and WO-1; black bars). Data represent mean fluorescence intensity values (MFI) + SD of at least five experiments performed in duplicates with different cell donors. *p # 0.05, Student t test. The Journal of Immunology 2507

C. dubliniensis species specifically induces increased migration, uptake, and antimicrobial effects, but less formation of NETs by neutrophils as compared with C. albicans, thereby providing new insights into the differential virulence of the two species. Fur- thermore, our data suggest that this differential neutrophil inter- action is, in part, associated with the different capacity of the two Candida species to form hyphae. Previous studies demonstrated that C. albicans releases a che- motactic factor that can act through the formyl peptide receptor on neutrophils and induce migration (18, 33). Using the single-cell chemotaxis assay, Geiger et al. (18) showed that also C. dubliniensis releases chemokinetic and chemotactic factors for human neu- trophils, and that the conditioned media of C. albicans resulted in FIGURE 5. Less neutrophil damage is caused by C. dubliniensis. The a lower chemotactic index than did the conditioned supernatant of concentration of LDH, as a measure of neutrophil damage, was determined C. dubliniensis. In agreement with this, we detected a higher after 20 h of neutrophil-Candida coculture in the supernatants. C. dubliniensis migration of human neutrophils toward all tested isolates of C. strains (H1, Wu¨272, and Wu¨284; white bars) caused less neutrophil damage dubliniensis compared with isolates of C. albicans (Fig. 1). The than did the C. albicans strains (SC5314, ATCC38696, and ATCC4408; black bars). Results were normalized setting the lysis of all neutrophils to chemoattractant of C. albicans has been identified as a small 100%. Data represent mean percentage values of relative fluorescence peptide (18, 33), but there is evidence for the involvement of other Downloaded from intensities (% MFI) + SD of at least five experiments performed with factors in the neutrophil migration toward C. albicans, such as different cell donors. **p # 0.01, Student t test. soluble b-glucans (34). Although C. albicans hyphae induce a higher neutrophil motility than the yeast form (35), in our assays, C. dubliniensis infections have rarely been observed (2), and this C. dubliniensis caused a higher migration rate of neutrophils species proved to be less virulent than C. albicans in various in- than C. albicans (Fig. 1) despite its incapability of forming true

fection models (4–7). The differences in the interaction of both hyphae under the tested cell culture conditions, suggesting that http://www.jimmunol.org/ fungal species with human immune cells remain largely unknown. the chemoattractant produced by C. dubliniensis is not a hypha- Neutrophils play a major role in innate antifungal defense. specific product. Therefore, the understanding of neutrophil–Candida interaction is Neutrophils also internalized C. dubliniensis cells more effi- of particular importance. These observations instigated the com- ciently than cells of C. albicans (Fig. 2). In a previous study, parative analysis of the two species in the context of neutrophil Peltroche-Llacsahuanga et al. (17) found no significant difference interaction in this study. By this approach, we unraveled that in the phagocytosis between different isolates of both fungal

Table II. Increased release of ROS, MPO, lactoferrin, and IL-8 from neutrophils stimulated with C. dubliniensis compared with C. albicans by guest on September 26, 2021

ROS MPO Lactoferrin IL-8

Donors Strain Tested % RFI t Test % Conc. t Test % Conc. t Test % Conc. t Test C. albicans SC5314 8 100 100 100 100 ATCC10261 8 95 6 11 101 6 39 82 6 38 100 6 22 ATCC38696 3 93 6 3456 6*346 2 * 107 6 1* ATCC44807 3 106 6 3926 13 77 6 19 89 6 5 ATCC44808 8 120 6 20 119 6 22 164 6 78 114 6 17 DMS1386 3 92 6 1**696 9*696 12 * 88 6 3 WO-1 8 66 6 22 80 6 39 108 6 76 104 6 21 MA13 3 89 6 5716 13 80 6 13 86 6 4 Wu¨212 3 100 6 6866 3 100 6 30 86 6 6 Wu¨277 3 95 6 5876 11 98 6 6976 3 C. dubliniensis H1 8 143 6 24 ** 192 6 26 ** 355 6 146 *** 196 6 25 * H12 8 157 6 23 ** 188 6 18 ** 377 6 168 *** 195 6 35 * Wu¨272 8 159 6 36 *** 194 6 32 ** 340 6 149 ** 165 6 30 * Wu¨284 8 157 6 35 ** 209 6 20 ** 387 6 195 ** 176 6 30 * Wu¨294 3 124 6 11 127 6 12 * 209 6 56 144 6 9* Wu¨361 3 128 6 3 ** 133 6 18 * 220 6 47 134 6 4* Wu¨367 3 135 6 5 * 140 6 11 * 273 6 38 ** 151 6 14 ** Wu¨370 3 147 6 2 *** 137 6 3 * 249 6 30 ** 123 6 4** VK9603-29 3 129 6 9 * 144 6 20 * 291 6 106 121 6 2** VK9603-34 8 138 6 22 169 6 17 ** 309 6 128 *** 183 6 35 * VK9603-54 3 123 6 9 137 6 5 * 230 6 36 ** 129 6 8* VK9603-63 3 115 6 7 123 6 4 178 6 21 * 116 6 7* Neutrophils were coincubated with live yeast cells of the indicated strains at a MOI of 1, or left untreated as a negative control. ROS were measured intracellularly by detecting the fluorescence of ROS-converted fluorescent dye. MPO, lactoferrin, and IL-8 were measured from coculture supernatants by ELISA. Results shown are percentages of relative mean fluorescence intensities (RFI; for ROS) or concentrations (Conc.; for MPO, lactoferrin, IL-8) 6 SD from at least three experiments with different blood donors. Data were normalized to the value of C. albicans strain SC5314 (set to 100%) after the subtraction of the negative control. *p # 0.05, **p # 0.01, ***p # 0.001, Student t test. 2508 NEUTROPHIL RESPONSES TO C. ALBICANS AND C. DUBLINIENSIS

FIGURE 6. Hyphae of C. albicans augment the release of ROS to a higher extent than hyphal cells of C. dubliniensis. Neutrophils were coincubated with live yeast and hyphal cells at a MOI of 1, or left untreated as a negative control. ROS (A) were measured intracellularly by detecting the fluorescence of ROS-converted fluorescent dye. Lactoferrin (B) was measured from coculture supernatants by ELISA. Results shown are percentages of relative mean fluorescence intensities (ROS) or relative concentrations (lactoferrin) 6 SD from at least five experiments with different blood donors. Data were nor- malized to the value of C. albicans strain SC5314 (set to 100%) after the subtraction of the negative control. *p # 0.05, **p # 0.01, Student t test. n.s., Not significant.

species. This discrepancy is likely due to the use of whole hepa- ficiently as they do when in contact with C. albicans (Figs. 3, 4). Downloaded from rinized blood instead of purified human neutrophils. Phagocytosis Moreover, Yost et al. (49) demonstrated that ROS are necessary of fungi by neutrophils involves complement receptor 3, FcRs, and but not sufficient for the induction of NETs, but the exact mech- dectin-1 (36–38). In our serum-free assays, the participation of a anism of NETosis and all factors involved are yet unknown. One complement-dependent uptake or an Ab-mediated uptake through might speculate that because C. dubliniensis is more efficiently FcRs could be excluded. Dectin-1 is a major receptor for Candida phagocytosed than C. albicans, the formation of NETs would be

(39, 40), recognizing b-glucan components of the fungal cell wall dispensable; in contrast, C. albicans is also able to escape from the http://www.jimmunol.org/ and inducing phagocytosis, defense processes, and cytokine pro- neutrophils by producing hyphae, and the large hyphae of C. duction by neutrophils, but its important role in yeast recognition albicans can be prevented from spreading when trapped in the is contradictory (41). Further studies are needed to analyze NETs (14). whether species-specific differences in the expression or avail- Candida species can cause cell and tissue damage during in- ability of b-glucan might explain a more efficient uptake of C. teraction with the host. Although C. dubliniensis is able to damage dubliniensis compared with C. albicans. host cells (6), our results show that isolates of C. dubliniensis do The release of NETs is thought to be an immune defense strategy not cause as severe damage to human neutrophils in vitro as C. that prevents the distribution of pathogens, trapping them into albicans strains (Fig. 5). These data are in line with earlier reports a milieu of highly concentrated antimicrobial substances (42–44). showing that the ability of C. dubliniensis to disseminate and to by guest on September 26, 2021 NET formation is appreciated as a killing mechanism against cause cell damage in vitro and in vivo is reduced (4–7). This re- different morphotypes of fungi (14, 45), because neutrophils that duction in host cell damage is likely associated, in part, with the are unable to produce NETs lack the ability to inhibit fungal in- reduced hyphal formation of C. dubliniensis, because the hyphal fection efficiently (43, 46). However, the ability of NETs to di- morphology serves the penetration of the host cells through me- rectly kill pathogens such as C. albicans is not unequivocally chanical force of the hyphal tip (50). We have observed that under established (47). Even though NETosis requires autophagy and the the used cell culture conditions, C. dubliniensis strains were not activation of the respiratory burst machinery involving MPO and able to efficiently form hyphae, and even after induction of hypha the production of ROS (46, 48), our results show that despite the formation, this morphology was not maintained over time (Sup- higher MPO and ROS production by neutrophils stimulated with plemental Figs. 2, 3). Thus, both the ability to switch between the C. dubliniensis, these neutrophils do not undergo NETosis so ef- two morphological forms and to maintain hyphal growth seems to be pivotal for Candida cells to adapt to different host niches and survive in the host (51). In addition, the hyphal morphology is associated with the production of several virulence factors, such as proteases and adhesins, which contribute to the virulence in the host (51). Thus, the observed differences between C. dubliniensis hyphae compared with C. albicans hyphae in inflicting damage to neutrophils could be, in part, also due to differences in some of the virulence factors between these fungal strains. Upon the encounter with the fungus, neutrophils use a set of killing mechanisms and can further activate and recruit other im- mune cells by producing cytokines. Neutrophils produce micro- bicidal amounts of ROS such as superoxide, hydrogen peroxide, and hypochlorous acid, which damage fungi (13, 52). ROS pro- FIGURE 7. Stimulation of lactoferrin release from neutrophils by C. duction is catalyzed by the NADPH system and subsequently by dubliniensis is dependent on the host cell–Candida contact. Neutrophils MPO, a component of the azurophilic granules. In addition to the were coincubated with live yeast cells at a MOI of 1 either in a direct contact (black bars) or neutrophils were separated from the Candida cells oxidative mechanisms, neutrophils can fight bacteria and fungi through a porous membrane (striped bars). Lactoferrin was measured from through a set of antimicrobial substances such as lactoferrin, coculture supernatants by ELISA. Results shown are lactoferrin concen- defensins, and cathelicidins. We found that in comparison with trations, and data represent means + SD from three experiments with isolates of C. albicans, the studied C. dubliniensis isolates stim- different cell donors. ulated increased levels of ROS, MPO, and lactoferrin by human The Journal of Immunology 2509

FIGURE 8. Differential macrophage cytokine response to C. dubliniensis and C. albicans. MDMs were coincubated with live yeast cells of C. albicans (strain SC5314) and C. dubliniensis (strains Wu¨272 and Wu¨284) for 24 h at a MOI of 1, or left untreated as a negative control. Cytokine levels were measured from coculture supernatants using a cytokine array. Results shown represent means from two experiments with different blood donors. Data were normalized to the value of the arrays’ positive control (set to 100% after subtracting the background). GRO-a, Growth regulated oncogene-a; IP-10, IFN-g– induced protein 10; MIF, macrophage migration inhibitory factor; Serpin-1, serine protease inhibitor-1; sTREM, soluble triggering receptor expressed on myeloid cells.

neutrophils (Table II). These data point to an early and more differences in the chemokine and cytokine pattern induced by the Downloaded from pronounced activation of neutrophil antifungal activity by C. two Candida species. Because the production of several of these dubliniensis. These observations are in agreement with the study cytokines and chemokines was shown to be induced by dectin- of Vilela et al. (7) where a relatively strong inflammatory reaction 1– and/or dectin-2–mediated signaling (53, 54), the differential of mice infected with C. dubliniensis was seen compared with cytokine and chemokine induction by C. albicans and C. dubliniensis those infected with C. albicans. Furthermore, our results indicate may indicate their differential recognition via these host cell

that the strong stimulation of lactoferrin production in neutrophils pattern recognition receptors. Similarly, it has been shown that the http://www.jimmunol.org/ by C. dubliniensis requires the fungus–host cell contact (Fig. 7). more invasive C. albicans induces considerably higher production This suggests that either C. dubliniensis more strongly activates of IL-1a, IL-8 (20), G-CSF, GM-CSF, and IL-6 (21) in recon- neutrophils, for example, because of surface-expressed molecular stituted human epithelium than C. dubliniensis. patterns engaged with host cell receptors, or this species lacks certain Many studies have demonstrated that IL-17 is a key cytokine in factor(s) that counteract or suppress the reaction of the host cells. the host defense against C. albicans infection (55–58). This cy- We have also investigated the release of fungal immunity-related tokine is mainly produced by Th17 cells, which develop in the cytokines upon encounter with both Candida species. IL-8 is re- presence of IL-1b, IL-6, and TGF-b, whereas IL-23 is important leased by neutrophils and several other cell types in response to for the expansion, maintenance, and function of Th17 cells (32). fungal or bacterial infection and triggers the mobilization of in- C. albicans induces IL-17 production dependent on macrophage by guest on September 26, 2021 flammatory cells into the site of infection (52). Similar to the mannose receptor, and this IL-17 production is enhanced by the antimicrobial substances, the amount of IL-8 released from neu- TLR2/dectin-1 activation pathway (59). Because the hyphal form trophils was also significantly higher when coincubated with C. of C. albicans is recognized preferentially by TLR2 on mononu- dubliniensis than C. albicans. In addition, several differences in clear cells, in contrast with the yeast cells recognized mainly by the response of macrophages to the two related fungal species TLR4 (60), the morphological plasticity of Candida cells might be were observed. Most molecules detected with the used array were an important aspect in regulating IL-17 production. Furthermore, less induced by C. dubliniensis, such as G-CSF, GM-CSF, IL-1a, IL-1b is produced after the stimulation of the inflammasome (61), IL-1b, IL-1ra, IL-10, IL-16, serine protease inhibitor-1, TNF-a, which also responds to the morphology shift of C. albicans and is and soluble triggering receptor expressed on myeloid cell, but crucial in the secretion of IL-17. Indeed, it has been shown that C. IFN-g–induced protein 10 was enhanced, indicating important albicans hyphae specifically induce inflammasome activation in

FIGURE 10. PBMCs produce less IL-17A when stimulated with C. FIGURE 9. C. dubliniensis triggers less IL-1b production by human dubliniensis compared with stimulation with C. albicans. PBMCs were macrophages than C. albicans. MDMs were coincubated with live yeast coincubated with live yeast and hyphal cells for 7 d at a MOI of 0.002, or cells for 20 h at a MOI of 4, or left untreated as a negative control. IL-1b left untreated as a negative control. IL-17A was measured from coculture was measured from coculture supernatants by ELISA. Results shown are supernatants by ELISA. Results shown are relative IL-17A concentrations relative IL-1b concentrations + SD from at least five experiments with 6 SD from at least six experiments with different blood donors. Data were different blood donors. Data were normalized to the value of C. albicans normalized to the value of C. albicans strain SC5314 (set to 100%). *p # strain SC5314 (set to 100%). **p # 0.01, Student t test. 0.05, ***p # 0.001, Student t test. n.s., Not significant. 2510 NEUTROPHIL RESPONSES TO C. ALBICANS AND C. DUBLINIENSIS macrophages in a partly dectin-1–dependent manner, resulting in when encountering C. dubliniensis. Late responses, such as NETosis enhanced IL-1b secretion and induction of IL-17 production (53). and the release of cytokines by mononuclear cells, including IL-1b In this article, we demonstrate that the production of IL-1b is and IL-17A involved in antifungal immunity, are reduced, which reduced when MDMs are stimulated with C. dubliniensis in might indicate a partly dispensable late response because of the comparison with C. albicans (Fig. 9), supporting the important early neutrophil activation and fungal clearance. Future studies role of the morphological conversion. Furthermore, the release of should identify the fungal and host factors underlying this differ- IL-17 from PBMCs is also reduced for C. dubliniensis when ential neutrophil response. compared with C. albicans (Fig. 10). These differences are likely due to the fact that C. albicans readily formed hyphae under the Acknowledgments used culture conditions, whereas C. dubliniensis did not. IL-17 We thank Joachim Morschha¨user for providing C. albicans and C. dublin- plays an important role in the recruitment of neutrophils (62). iensis clinical isolates and C. albicans strain SCADH1G4A, as well as Although C. dubliniensis induced less IL-17A production in our plasmids pADH1E1 and pNIM1. We also thank Toni Kaulfuß and assays, measured after 7 d, it did enhance the migration of neu- Christina Taubert for technical assistance, and the Department of In- trophils directly and the production of IL-8, which also supports fection Biology (Hans Kno¨ll Institute, Jena, Germany) for access to neutrophil migration. The induction of these immediate and early their microscope. neutrophil responses, as well as a more efficient phagocytosis, and the limited induction of later responses, including production of Disclosures inflammatory cytokines and IL-17, suggest that partly because of The authors have no financial conflicts of interest. its restricted capacity to form hyphae, C. dubliniensis poses less Downloaded from danger to the host organism than C. albicans, which is reflected in the attenuated late host response. References Key virulence factors of Candida include the ability to switch 1. Sullivan, D. J., G. P. Moran, E. Pinjon, A. Al-Mosaid, C. Stokes, C. Vaughan, and D. C. Coleman. 2004. Comparison of the epidemiology, drug resistance between yeast and hypha morphology enabling the fungus in its mechanisms, and virulence of Candida dubliniensis and Candida albicans. spreading and invasion of the human host (50, 63), the production FEMS Yeast Res. 4: 369–376. 2. Sullivan, D. J., T. J. Westerneng, K. A. Haynes, D. E. Bennett, and of enzymes such as secreted aspartyl proteases and phospholipases http://www.jimmunol.org/ D. C. Coleman. 1995. Candida dubliniensis sp. nov.: phenotypic and molecular (64), and the expression of adhesins (65). Some of these virulence characterization of a novel species associated with oral candidosis in HIV- factors, found in C. albicans, however, are lacking or are reduced infected individuals. Microbiology 141: 1507–1521. in C. dubliniensis (9, 10, 66). Aside from C. albicans, C. dubliniensis 3. Staib, P., and J. Morschha¨user. 2007. formation in Candida albicans and Candida dubliniensis—an enigmatic developmental programme. is the only Candida species that has the capacity to produce true Mycoses 50: 1–12. hyphae (3), but with far less efficiency and under fewer conditions. 4. Koga-Ito, C. Y., E. Y. Komiyama, C. A. Martins, T. C. Vasconcellos, A. O. Jorge, Y. R. Carvalho, R. F. do Prado, and I. Balducci. 2011. Experimental systemic In our experiments, C. dubliniensis did not form filaments during virulence of oral Candida dubliniensis isolates in comparison with Candida coincubation with neutrophils in cell culture medium (Supple- albicans, Candida tropicalis and Candida krusei. Mycoses 54: e278–e285. mental Fig. 3). 5. Jayatilake, J. A. M. S., Y. H. Samaranayake, and L. P. Samaranayake. 2008. A comparative study of candidal invasion in rabbit tongue mucosal explants and by guest on September 26, 2021 We and others previously showed that neutrophils respond more reconstituted human oral epithelium. Mycopathologia 165: 373–380. strongly to C. albicans hyphae than to yeast cells (30, 35). In 6. Stokes, C., G. P. Moran, M. J. Spiering, G. T. Cole, D. C. Coleman, and agreement with this, we detected an enhanced production and D. J. Sullivan. 2007. Lower filamentation rates of Candida dubliniensis con- tribute to its lower virulence in comparison with Candida albicans. Fungal release of ROS and lactoferrin for neutrophils coincubated with Genet. Biol. 44: 920–931. hyphal cells of various C. albicans isolates in comparison with 7. Vilela, M. M. S., K. Kamei, A. Sano, R. Tanaka, J. Uno, I. Takahashi, J. Ito, K. Yarita, and M. Miyaji. 2002. Pathogenicity and virulence of Candida dubliniensis: yeast cells (Fig. 6). However, in case of the neutrophil challenge comparison with C. albicans. Med. Mycol. 40: 249–257. with both morphotypes of C. dubliniensis, the levels of released 8. Gilfillan, G. D., D. J. Sullivan, K. Haynes, T. Parkinson, D. C. Coleman, and ROS and lactoferrin remained almost the same. When coincu- N. A. Gow. 1998. Candida dubliniensis: phylogeny and putative virulence fac- tors. Microbiology 144: 829–838. bating neutrophils or PBMCs with fungal cells that were induced 9. Moran, G., C. Stokes, S. Thewes, B. Hube, D. C. Coleman, and D. Sullivan. for 1 h to form hyphae, increased amounts of ROS, lactoferrin, and 2004. Comparative genomics using Candida albicans DNA microarrays reveals IL-17A were detected for both species. Nevertheless, hyphae- absence and divergence of virulence-associated genes in Candida dubliniensis. Microbiology 150: 3363–3382. induced ROS, lactoferrin, and IL-17A production for isolates 10. Jackson, A. P., J. A. Gamble, T. Yeomans, G. P. Moran, D. Saunders, D. Harris, of C. dubliniensis did not reach the levels of those induced by M. Aslett, J. F. Barrell, G. Butler, F. Citiulo, et al. 2009. Comparative genomics of the fungal pathogens Candida dubliniensis and Candida albicans. Genome C. albicans hyphae. This could be because of the fact that Res. 19: 2231–2244. C. dubliniensis does not persist in the state of hyphal form when 11. O’Connor, L., N. Caplice, D. C. Coleman, D. J. Sullivan, and G. P. Moran. 2010. introduced to the nutrition-rich environment (21), as we also ob- Differential filamentation of Candida albicans and Candida dubliniensis Is gov- erned by nutrient regulation of UME6 expression. Eukaryot. Cell 9: 1383–1397. served during coincubation with neutrophils (Supplemental Fig. 2). 12. Nordenfelt, P., and H. Tapper. 2011. Phagosome dynamics during phagocytosis The lack of the filamentation in C. dubliniensis is supported by neutrophils. J. Leukoc. Biol. 90: 271–284. by expression of the NRG1 repressor and suppression of UME6 13. Robinson, J. M. 2009. Phagocytic leukocytes and reactive oxygen species. Histochem. Cell Biol. 131: 465–469. induction in a nutrient-dependent manner (11, 67). Thus, fila- 14. Urban, C. F., U. Reichard, V. Brinkmann, and A. Zychlinsky. 2006. Neutrophil mentation of C. dubliniensis can be increased after deletion of the extracellular traps capture and kill Candida albicans yeast and hyphal forms. Cell. Microbiol. 8: 668–676. NRG1 gene, which leads to an improved invasion of human epi- 15. Netea, M. G., G. D. Brown, B. J. Kullberg, and N. A. R. Gow. 2008. An inte- thelium and survival in macrophage cocultures in vitro (67), or in grated model of the recognition of Candida albicans by the innate immune a nutrient-poor environment (11). The role of hypha production by system. Nat. Rev. Microbiol. 6: 67–78. 16. Silva, M. T. 2010. When two is better than one: macrophages and neutrophils C. dubliniensis in the interaction with host immune cells could be work in concert in innate immunity as complementary and cooperative partners further studied using such genetically manipulated fungal cells, of a myeloid phagocyte system. J. Leukoc. Biol. 87: 93–106. although the conclusions would be limited because of the multiple 17. Peltroche-Llacsahuanga, H., N. Schnitzler, S. Schmidt, K. Tintelnot, R. Lu¨tticken, and G. Haase. 2000. Phagocytosis, oxidative burst, and killing effects of NRG1 in addition to influencing filamentous growth. of Candida dubliniensis and Candida albicans by human neutrophils. FEMS In conclusion, our data indicate that, in comparison with C. Microbiol. Lett. 191: 151–155. 18. Geiger, J., D. Wessels, S. R. Lockhart, and D. R. Soll. 2004. Release of a potent albicans, early responses such as neutrophil migration, phagocy- polymorphonuclear leukocyte chemoattractant is regulated by white-opaque tosis, and the release of antimicrobial substances are increased switching in Candida albicans. Infect. Immun. 72: 667–677. The Journal of Immunology 2511

19. Jayatilake, J. A. M. S., Y. H. Samaranayake, L. K. Cheung, and L. P. Samaranayake. infected lung tissue is dependent on invading neutrophils and influenced by 2006. Quantitative evaluation of tissue invasion by wild type, hyphal and SAP hydrophobin RodA. PLoS Pathog. 6: e1000873. mutants of Candida albicans, and non-albicans Candida species in reconstituted 46. Metzler, K. D., T. A. Fuchs, W. M. Nauseef, D. Reumaux, J. Roesler, I. Schulze, human oral epithelium. J. Oral Pathol. Med. 35: 484–491. V. Wahn, V. Papayannopoulos, and A. Zychlinsky. 2011. Myeloperoxidase is 20. Jayatilake, J. A. M. S., L. P. Samaranayake, Q. Lu, and L. J. Jin. 2007. IL-1alpha, required for neutrophil extracellular trap formation: implications for innate IL-1ra and IL-8 are differentially induced by Candida in experimental oral immunity. Blood 117: 953–959. candidiasis. Oral Dis. 13: 426–433. 47. Menegazzi, R., E. Decleva, and P. Dri. 2012. Killing by neutrophil extracellular 21. Moyes, D. L., C. Murciano, M. Runglall, A. Kohli, A. Islam, and J. R. Naglik. traps: fact or folklore? Blood 119: 1214–1216. 2012. Activation of MAPK/c-Fos induced responses in oral epithelial cells is 48. Remijsen, Q., T. Vanden Berghe, E. Wirawan, B. Asselbergh, E. Parthoens, specific to Candida albicans and Candida dubliniensis hyphae. Med. Microbiol. R. De Rycke, S. Noppen, M. Delforge, J. Willems, and P. Vandenabeele. 2011. Immunol. (Berl.) 201: 93–101. Neutrophil extracellular trap cell death requires both autophagy and superoxide 22. Gillum, A. M., E. Y. Tsay, and D. R. Kirsch. 1984. Isolation of the Candida generation. Cell Res. 21: 290–304. albicans gene for orotidine-59-phosphate decarboxylase by complementation of 49. Yost, C. C., M. J. Cody, E. S. Harris, N. L. Thornton, A. M. McInturff, S. cerevisiae ura3 and E. coli pyrF mutations. Mol. Gen. Genet. 198: 179–182. M. L. Martinez, N. B. Chandler, C. K. Rodesch, K. H. Albertine, C. A. Petti, 23. Slutsky, B., M. Staebell, J. Anderson, L. Risen, M. Pfaller, and D. R. Soll. 1987. et al. 2009. Impaired neutrophil extracellular trap (NET) formation: a novel “White-opaque transition”: a second high-frequency switching system in Can- innate immune deficiency of human neonates. Blood 113: 6419–6427. dida albicans. J. Bacteriol. 169: 189–197. 50. Brand, A. 2012. Hyphal growth in human fungal pathogens and its role in vir- 24. Staib, P., and J. Morschha¨user. 1999. Chlamydospore formation on Staib agar as ulence. Int. J. Microbiol. 2012: 517529. a species-specific characteristic of Candida dubliniensis. Mycoses 42: 521–524. 51. Kumamoto, C. A., and M. D. Vinces. 2005. Contributions of hyphae and hypha- 25. Morschha¨user, J., M. Ruhnke, S. Michel, and J. Hacker. 1999. Identification of co-regulated genes to Candida albicans virulence. Cell. Microbiol. 7: 1546– CARE-2-negative Candida albicans isolates as Candida dubliniensis. Mycoses 1554. 42: 29–32. 52. Romani, L. 2004. Immunity to fungal infections. Nat. Rev. Immunol. 4: 1–23. 26. Dabas, N., and J. Morschha¨user. 2007. Control of ammonium permease ex- 53. Cheng, S.-C., F. L. van de Veerdonk, M. Lenardon, M. Stoffels, T. Plantinga, pression and filamentous growth by the GATA transcription factors GLN3 and S. Smeekens, L. Rizzetto, L. Mukaremera, K. Preechasuth, D. Cavalieri, et al. GAT1 in Candida albicans. Eukaryot. Cell 6: 875–888. 2011. The dectin-1/inflammasome pathway is responsible for the induction of 27. Reuss, O., and J. Morschha¨user. 2006. A family of oligopeptide transporters is protective T-helper 17 responses that discriminate between yeasts and hyphae of Downloaded from required for growth of Candida albicans on proteins. Mol. Microbiol. 60: 795–812. Candida albicans. J. Leukoc. Biol. 90: 357–366. 28. Park, Y.-N., and J. Morschha¨user. 2005. Tetracycline-inducible gene expression 54. Sato, K., X. L. Yang, T. Yudate, J. S. Chung, J. Wu, K. Luby-Phelps, and gene deletion in Candida albicans. Eukaryot. Cell 4: 1328–1342. R. P. Kimberly, D. Underhill, P. D. Cruz, Jr., and K. Ariizumi. 2006. Dectin-2 is 29. Reuss, O., A. Vik, R. Kolter, and J. Morschha¨user. 2004. The SAT1 flipper, an a pattern recognition receptor for fungi that couples with the Fc receptor gamma optimized tool for gene disruption in Candida albicans. Gene 341: 119–127. chain to induce innate immune responses. J. Biol. Chem. 281: 38854–38866. 30. Losse, J., E. Svobodova´, A. Heyken, B. Hube, P. F. Zipfel, and M. Jo´zsi. 2011. 55. Huang, W., L. Na, P. L. Fidel, and P. Schwarzenberger. 2004. Requirement of Role of pH-regulated antigen 1 of Candida albicans in the fungal recognition interleukin-17A for systemic anti-Candida albicans host defense in mice. J.

and antifungal response of human neutrophils. Mol. Immunol. 48: 2135–2143. Infect. Dis. 190: 624–631. http://www.jimmunol.org/ 31. Sudbery, P. E. 2011. Growth of Candida albicans hyphae. Nat. Rev. Microbiol. 9: 56. Eyerich, K., S. Foerster, S. Rombold, H.-P. Seidl, H. Behrendt, H. Hofmann, 737–748. J. Ring, and C. Traidl-Hoffmann. 2008. Patients with chronic mucocutaneous 32. Gaffen, S. L., N. Herna´ndez-Santos, and A. C. Peterson. 2011. IL-17 signaling in candidiasis exhibit reduced production of Th17-associated cytokines IL-17 and host defense against Candida albicans. Immunol. Res. 50: 181–187. IL-22. J. Invest. Dermatol. 128: 2640–2645. 33. Edens, H. A., C. A. Parkos, T. W. Liang, A. J. Jesaitis, J. E. Cutler, and 57. Conti, H. R., F. Shen, N. Nayyar, E. Stocum, J. N. Sun, M. J. Lindemann, H. M. Miettinen. 1999. Non-serum-dependent chemotactic factors produced by A. W. Ho, J. H. Hai, J. J. Yu, J. W. Jung, et al. 2009. Th17 cells and IL-17 re- Candida albicans stimulate chemotaxis by binding to the formyl peptide re- ceptor signaling are essential for mucosal host defense against . ceptor on neutrophils and to an unknown receptor on macrophages. Infect. J. Exp. Med. 206: 299–311. Immun. 67: 1063–1071. 58. Zelante, T., A. De Luca, C. D’Angelo, S. Moretti, and L. Romani. 2009. IL-17/ 34. Sato, T., K. Iwabuchi, I. Nagaoka, Y. Adachi, N. Ohno, H. Tamura, K. Seyama, Th17 in anti-fungal immunity: what’s new? Eur. J. Immunol. 39: 645–648. Y. Fukuchi, H. Nakayama, F. Yoshizaki, et al. 2006. Induction of human neu- 59. van de Veerdonk, F. L., R. J. Marijnissen, B. J. Kullberg, H. J. P. M. Koenen, S.-

trophil chemotaxis by Candida albicans-derived beta-1,6-long glycoside side- C. Cheng, I. Joosten, W. B. van den Berg, D. L. Williams, J. W. M. van der Meer, by guest on September 26, 2021 chain-branched beta-glucan. J. Leukoc. Biol. 80: 204–211. L. A. B. Joosten, and M. G. Netea. 2009. The macrophage mannose receptor 35. Wozniok, I., A. Hornbach, C. Schmitt, M. Frosch, H. Einsele, B. Hube, J. Lo¨ffler, induces IL-17 in response to Candida albicans. Cell Host Microbe 5: 329–340. and O. Kurzai. 2008. Induction of ERK-kinase signalling triggers morphotype- 60. van der Graaf, C. A. A., M. G. Netea, I. Verschueren, J. W. M. van der Meer, and specific killing of Candida albicans filaments by human neutrophils. Cell. B. J. Kullberg. 2005. Differential cytokine production and Toll-like receptor Microbiol. 10: 807–820. signaling pathways by Candida albicans blastoconidia and hyphae. Infect. 36. Mayadas, T. N., and X. Cullere. 2005. Neutrophil beta2 integrins: moderators of Immun. 73: 7458–7464. life or death decisions. Trends Immunol. 26: 388–395. 61. Gross, O., H. Poeck, M. Bscheider, C. Dostert, N. Hannesschla¨ger, S. Endres, 37. Wellington, M., J. M. Bliss, and C. G. Haidaris. 2003. Enhanced phagocytosis of G. Hartmann, A. Tardivel, E. Schweighoffer, V. Tybulewicz, et al. 2009. Syk Candida species mediated by opsonization with a recombinant human antibody kinase signalling couples to the Nlrp3 inflammasome for anti-fungal host de- single-chain variable fragment. Infect. Immun. 71: 7228–7231. fence. Nature 459: 433–436. 38. Kennedy, A. D., J. A. Willment, D. W. Dorward, D. L. Williams, G. D. Brown, 62. Laan, M., Z. H. Cui, H. Hoshino, J. Lo¨tvall, M. Sjo¨strand, D. C. Gruenert, and F. R. DeLeo. 2007. Dectin-1 promotes fungicidal activity of human neu- B. E. Skoogh, and A. Linde´n. 1999. Neutrophil recruitment by human IL-17 via trophils. Eur. J. Immunol. 37: 467–478. C-X-C chemokine release in the airways. J. Immunol. 162: 2347–2352. 39. Taylor, P. R., S. V. Tsoni, J. A. Willment, K. M. Dennehy, M. Rosas, H. Findon, 63. Saville, S. P., A. L. Lazzell, C. Monteagudo, and J. L. Lopez-Ribot. 2003. K. Haynes, C. Steele, M. Botto, S. Gordon, and G. D. Brown. 2007. Dectin-1 is Engineered control of cell morphology in vivo reveals distinct roles for yeast and required for beta-glucan recognition and control of fungal infection. Nat. filamentous forms of Candida albicans during infection. Eukaryot. Cell 2: 1053– Immunol. 8: 31–38. 1060. 40. Drummond, R. A., and G. D. Brown. 2011. The role of Dectin-1 in the host 64. Naglik, J. R., C. A. Rodgers, P. J. Shirlaw, J. L. Dobbie, L. L. Fernandes-Naglik, defence against fungal infections. Curr. Opin. Microbiol. 14: 392–399. D. Greenspan, N. Agabian, and S. J. Challacombe. 2003. Differential expression 41. van Bruggen, R., A. Drewniak, M. Jansen, M. van Houdt, D. Roos, H. Chapel, of Candida albicans secreted aspartyl proteinase and phospholipase B genes in A. J. Verhoeven, and T. W. Kuijpers. 2009. Complement receptor 3, not Dectin- humans correlates with active oral and vaginal infections. J. Infect. Dis. 188: 1, is the major receptor on human neutrophils for beta-glucan-bearing particles. 469–479. Mol. Immunol. 47: 575–581. 65. Wa¨chtler, B., D. Wilson, K. Haedicke, F. Dalle, and B. Hube. 2011. From at- 42. Brinkmann, V., U. Reichard, C. Goosmann, B. Fauler, Y. Uhlemann, D. S. Weiss, tachment to damage: defined genes of Candida albicans mediate adhesion, in- Y. Weinrauch, and A. Zychlinsky. 2004. Neutrophil extracellular traps kill vasion and damage during interaction with oral epithelial cells. PLoS ONE 6: bacteria. Science 303: 1532–1535. e17046. 43. Fuchs, T. A., U. Abed, C. Goosmann, R. Hurwitz, I. Schulze, V. Wahn, 66. Spiering, M. J., G. P. Moran, M. Chauvel, D. M. Maccallum, J. Higgins, Y. Weinrauch, V. Brinkmann, and A. Zychlinsky. 2007. Novel cell death program K. Hokamp, T. Yeomans, C. d’Enfert, D. C. Coleman, and D. J. Sullivan. 2010. leads to neutrophil extracellular traps. J. Cell Biol. 176: 231–241. Comparative transcript profiling of Candida albicans and Candida dubliniensis 44. Papayannopoulos, V., K. D. Metzler, A. Hakkim, and A. Zychlinsky. 2010. identifies SFL2, a C. albicans gene required for virulence in a reconstituted Neutrophil elastase and myeloperoxidase regulate the formation of neutrophil epithelial infection model. Eukaryot. Cell 9: 251–265. extracellular traps. J. Cell Biol. 191: 677–691. 67. Moran, G. P., D. M. MacCallum, M. J. Spiering, D. C. Coleman, and 45. Bruns, S., O. Kniemeyer, M. Hasenberg, V. Aimanianda, S. Nietzsche, D. J. Sullivan. 2007. Differential regulation of the transcriptional repressor A. Thywissen, A. Jeron, J.-P. Latge´, A. A. Brakhage, and M. Gunzer. 2010. NRG1 accounts for altered host-cell interactions in Candida albicans and Production of extracellular traps against Aspergillus fumigatus in vitro and in Candida dubliniensis. Mol. Microbiol. 66: 915–929.