Online Laboratory Investigations

Critical Care Medicine www.ccmjournal.org e191

Objective: To study the correlation between fungal colonization and bacterial pneumonia and to test the effect of antifungal treatments on the development of bacterial pneumonia in colonized rats.Design: Experimental animal investigation.Setting: University research laboratory.

Subjects: Pathogen-free male Wistar rats weighing 250–275 g.Interventions: Rats were colonized by intratracheal instillation of Candida albicans. Fungal clearance from the lungs and immune response were measured. Both colonized and noncolonized ani-mals were secondarily instilled with different bacterial species (Pseudomonas aeruginosa, Escherichia coli, or Staphylococcus aureus). Bacterial phagocytosis by alveolar macrophages was evaluated in the presence of interferon-gamma, the main cyto-kine produced during fungal colonization. The effect of antifungal treatments on fungal colonization and its immune response were assessed. The prevalence of P. aeruginosa pneumonia was com-pared in antifungal treated and control colonized rats.Measurements and Main Results: C. albicans was slowly cleared and induced a Th1-Th17 immune response with very high inter-feron-gamma concentrations. Airway fungal colonization favored the development of bacterial pneumonia. Interferon-gamma was able to inhibit the phagocytosis of unopsonized bacteria by alveo-lar macrophages. Antifungal treatment decreased airway fungal colonization, lung interferon-gamma levels and, consequently, the prevalence of subsequent bacterial pneumonia.Conclusions: C. albicans airway colonization elicited a Th1-Th17 immune response that favored the development of bacterial pneu-monia via the inhibition of bacterial phagocytosis by alveolar mac-rophages. Antifungal treatment decreased the risk of bacterial pneumonia in colonized rats. (Crit Care Med 2013;41:e191–e199)Key Words: alveolar macrophages; bacterial pneumonia; Candida albicans; interferon-gamma; phagocytosis; Pseudomonas aeruginosa

Ventilator-associated pneumonia (VAP) is a major concern in patients under mechanical ventilation in ICUs (1–3). In spite of intense efforts to control

the known risk factors of VAP, these infections remain frequent with high morbidity and possible attributable

Airway Fungal Colonization Compromises the Immune System Allowing Bacterial Pneumonia to Prevail

Damien Roux, MD, PhD1,2,3; Stéphane Gaudry, MD1,2,3; Linda Khoy-Ear, MD1,2;

Meryem Aloulou, PhD4,5; Mathilde Phillips-Houlbracq, MSc1,2; Julie Bex1,2;

David Skurnik, MD, PhD6; Erick Denamur, MD, PhD1,2; Renato C. Monteiro, MD, PhD4,5,7;

Didier Dreyfuss, MD1,2,3; Jean-Damien Ricard, MD, PhD1,2,3

1Institut National de la Santé et de la Recherche Médicale, INSERM U722, Paris, France.

2Univ Paris Diderot, Sorbonne Paris Cité, UMRS-722, Site Xavier Bichat, Paris, France.

3AP-HP, Hôpital Louis Mourier, Service de Réanimation Médico-chirurgicale, Colombes, France.

4INSERM UMR-S699, Paris, France.5Inflamex Laboratory of Excellence, Paris Diderot University, Sorbonne Paris Cité, Paris, France.

6Division of Infectious Diseases, Department of Medicine, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA.

7Immunology Laboratory, Bichat Hospital, Assistance Publique-Hôpitaux de Paris, Paris, France.

This work was performed at Université Paris Diderot.

Drs. Gaudry, Khoy-Ear, and Aloulou, and Mr. Phillips-Houlbracq contrib-uted equally.

Supported, in part, by the French Society of Intensive Care (SRLF) (Drs. Roux and Phillips-Houlbracq), the Société de Pneumologie de Langue Française (Dr. Gaudry), the Fonds d’Etudes et de Recherche du Corps Médical des Hôpitaux de Paris (Dr. Khoy-Ear), and the Legs Poix (Dr. Ricard). INSERM U722 laboratory received a research grant from Pfizer.

Dr. Ricard received grant support from Pfizer; is a board member and lectured for Covidien; and received support for travel from Fisher&Paykel. Dr. Roux received grant support from the French Society of Intensive Care. Dr. Gaudry lectured for Lob Conseils and received support for travel from Issis Medical. Drs. Aloulou and Bex received funding from from INSERM. Dr. Skurnik received grant support from the Hood Foundation and Hearst Foundation. Dr. Monteiro received funding from ANR. Dr. Dreyfuss received grant support from Pfizer. The remaining authors have disclosed that they do not have any potential conflicts of interest.

Address requests for reprints to: Jean-Damien Ricard, MD, PhD, Service de Réanimation, Hôpital Louis Mourier–178, rue des Renouillers, 92700 Colombes, France. E-mail: jean-damien.ricard@lmr.aphp.fr

Critical Care Medicine

0090-3493

10.1097/CCM.0b013e31828a25d6

41

9

e191

e199

2013

Copyright © 2013 by the Society of Critical Care Medicine and Lippincott Williams & Wilkins

DOI: 10.1097/CCM.0b013e31828a25d6

Mythili

xxx

September

2013

Roux et al

e192 www.ccmjournal.org September 2013 • Volume 41 • Number 9

mortality (4, 5). While bacterial species are responsible for 99% of VAP (1), the role of fungal pathogens is unclear in immunocompetent hosts (6). Candida species are detected in over 50% of ventilated critically ill patients (7, 8), but this microorganism is uncommonly responsible for lung infections in these patients (8–10). Thus, the very existence of true Candida pneumonia in critically ill patients is still questioned (11, 12). However, Azoulay et al (13) observed an association between airway colonization with Candida species and subsequent Pseudomonas aeruginosa VAP, raising the idea that interactions between bacterial and fungal species could be more important than previously considered (14). The latter concept prompted us to investigate a potential interaction between Candida albicans airway colonization and P. aeruginosa pneumonia (15). We observed that while an inoculum of P. aeruginosa caused pneumonia in only 3% of control rats, 34% of rats colonized with C. albicans developed pneumonia. We also noted a 60% decrease in reactive oxygen species production by alveolar macrophages (AMs) in the presence of C. albicans, suggesting an effect of C. albicans on the lung innate immune response. However, a formal causal link between C. albicans colonization and P. aeruginosa pneumonia and/or general bacterial pneumonia required further investigation. Recently, Hamet et al (7, 16) reported Candida colonization as a risk factor for VAP with multidrug resistant bacteria, highlighting the fact that this fungal colonization can no longer be overlooked.

Thus, the aims of the present study were 1) to confirm our previous results showing the facilitating effect of C. albicans airway colonization on the development of P. aeruginosa pneu-monia and to emphasize this finding by studying two other major bacterial pathogens responsible for VAP (Staphylococcus aureus and Escherichia coli) in the same animal model; 2) to investigate the mechanism of this effect, focusing on bacte-rial phagocytosis by AMs in the context of a high interferon-gamma (INF-γ) pulmonary concentration induced by fungal airway colonization; 3) to evaluate the consequences of our findings in a potential clinical situation by assessing the effect of an antifungal treatment on the development of bacterial pneumonia in the context of fungal airway colonization.

MATERIALS AND METHODS

AnimalsExperiments were performed on pathogen-free male Wistar rats weighing 250–275 g in compliance with the recommenda-tions from the French Ministry of Agriculture and approved by the French Veterinary Services. Transglottic instillations of rats were performed as described previously (15, 17).

MicroorganismsThe ATCC 10231 C. albicans, originally isolated from a patient with bronchomycosis, was used in the study. Three bacterial species were used: P. aeruginosa PAO1, S. aureus NCTC 8325, and E. coli P5.21 (B2 phylogenetic group, serogroup O6 [18]). C. albicans (3 × 106 colony-forming unit [CFU]/instillation)

was prepared as described previously (15). Bacteria were grown overnight in Luria Bertani broth, washed with saline, and diluted to the chosen concentration. The inocula of P. aeruginosa (6 × 106 CFU), S. aureus (3 × 107 CFU), and E. coli (2 × 106 CFU) were chosen to induce a bacterial pneumonia in less than 50% of the animals (preliminary data). To induce pneumonia in more than 80% of the animals, an inoculum of 8 × 107 CFU of P. aeruginosa was selected.

Fungal Airway Colonization, Immune Response, and Development of Subsequent P. aeruginosa PneumoniaAirway fungal colonization was obtained with a single intratra-cheal instillation of 3 × 106 CFU of C. albicans. Fungal clearance and the local immune response were assessed over time. Animals were killed 4 hours, 24 hours, 72 hours, 7 days, and 14 days after instillation, and C. albicans count was performed in homoge-nized lungs to evaluate spontaneous clearance. Levels of interleu-kin (IL)-2, INF-γ, IL-10, IL-4, and IL-17 in homogenized lungs were measured using a rat-specific ELISA kit (R&D and eBiosci-ence, France) at 24 and 72 hours after the fungal instillation (or a control instillation with saline) to evaluate the Th1, Th2, and Th17 local immune response in the presence of C. albicans.

To measure the effect of the fungal airway colonization on the development of bacterial pneumonia, the predetermined bacterial inoculum was instilled 24 or 72 hours after C. albi-cans, and the lungs were analyzed as described previously (15). A bacterial pneumonia was defined as the co-occurrence of a macroscopic pulmonary inflammation and a bacterial count above 104 CFU per instilled lung (15).

Bacterial Phagocytosis by Alveolar MacrophagesPhagocytosis was evaluated as described previously (19). Briefly, AMs were recovered from the bronchoalveolar lavage of nor-mal rats and incubated overnight with recombinant rat INF-γ (eBiosciences, Paris, France) at 10 ng/mL (20) or a control solu-tion (phosphate buffered saline + bovine serum albumin 1%). Commercially available Texas red-coupled bacteria (E. coli and S. aureus, Molecular Probes, Leiden, The Netherlands) were individually added at 37°C for 30 minutes (107 bacteria per 5 × 105 cells), with or without previous opsonization. Macro-phage membranes were stained with CD11b-FITC on ice for 20 minutes before confocal microscopy. Images were taken with a LSM 510 confocal microscope (Carl Zeiss, Le Pecq, France). The intensity of the red fluorescence inside each AM was quantified.

Antifungal TreatmentTwo antifungal treatments were chosen: fluconazole and amphotericin B. The minimum inhibitory concentrations of the C. albicans strain were tested by e-test (AB Biodisk, Solna, Sweden). For in vivo experiments, the daily doses were 10 mg/kg for fluconazole (21) and 1 mg/kg for amphotericin B (22, 23) through intraperitoneal injection. Treatment or control (saline solution) doses were started immediately after the fungal instillation and for the following 2 days. The effect of antifungal treatment or control was assessed 72 hours after

Online Laboratory Investigations

Critical Care Medicine www.ccmjournal.org e193

fungal instillation through quantitative fungal cultures and measuring INF-γ levels in the lungs.

Before testing an antifungal treatment on the development of P. aeruginosa pneumonia in colonized animals, we evaluated its direct effect on the development of the same pneumonia in the absence of C. albicans. Noncolonized rats were similarly treated; after 3 days, the P. aeruginosa inoculum (8 × 107 CFU) induced pneumonia in most of the rats.

To test the effect of the antifungal treatment on P. aerugi-nosa pneumonia during fungal airway colonization, the bacte-ria (6 × 106 CFU) were instilled 72 hours after C. albicans (24 hr after the third and last dose of antifungal treatment) and the lungs were evaluated 24 hours after the bacterial challenge.

StatisticsThe results are presented as median (interquartile range [IQR] 25; IQR, 75). The statistical analyses were performed with GraphPad Prism 4.00. Categorical data were compared by the chi-square test or Fisher exact test when an expected value was less than five. Continuous variables were compared using the Mann-Whitney U test for two groups, or Kruskal-Wallis test for more than two groups with the Dunn’s multiple compari-son posttest if statistical significance was found. A p value of less than 0.05 was considered significant.

RESULTS

Fungal Colonization Facilitates the Development of Subsequent P. aeruginosa PneumoniaA new simplified model of fungal airway colonization was tested and found to be responsible for an increased prevalence of P. aeruginosa pneumonia, confirming our previous observation (15). In this model, instead of three successive fungal instilla-tions of a clinical strain, fungal airway colonization was obtained through a single instillation of a laboratory strain of C. albicans.C. albicans Colonization. After intratracheal instillation, C. albicans was slowly cleared from the lungs (Fig. 1A) while the body weights of the colonized rats increased, similar to the uninstilled animals (Fig. 1B). As previously reported (15), only rare spots of peribronchial inflammation were caused by the Candida colonization (data not shown).

P. aeruginosa Pneumonia. Twenty-four hours after fun-gal colonization, instillation of the well-described laboratory strain P. aeruginosa PAO1 (24) induced significantly more bac-terial pneumonia as compared to control rats (14 of 32 vs 5 of 32, respectively; chi-square: p < 0.05, Table 1). An example of the macroscopic aspect of pneumonia is presented in Figure 2. Bacterial counts and lung weights were significantly higher in the colonized group as compared to the control group (Mann-Whitney U test: p < 0.05 for each; Fig. 3A).

Because high levels of C. albicans were still present in the lungs, 72 hours after the fungal instillation (Fig. 1), we tested a delayed P. aeruginosa instillation at 72 hours. Again, an increased prevalence of P. aeruginosa pneumonia (13 of 31 vs 6 of 32 in the colonized group and the control group, respectively; chi-square: p < 0.05; Table 1) was observed. In addition, the bacterial counts tend to be higher in the colonized group, and lung weights sig-nificantly differed between the two groups (Mann-Whitney U test: p = 0.06 and p < 0.05, respectively; Fig. 3B).

Effect of Fungal Airway Colonization on Other Bacterial Pneumonia (S. aureus and E. coli)After confirmation of the facilitating effect of fungal airway col-onization on the development of P. aeruginosa pneumonia with a simplified model of colonization, we investigated whether similar effects could be obtained with two other major bacterial pathogens involved in lung infection: S. aureus and E. coli (1, 2). Both microorganisms were individually instilled 24 hours after the fungal inoculum; bacterial pneumonias were more frequent in the colonized groups than in controls (14 of 15 vs 8 of 18 for S. aureus, chi-square: p < 0.01, and 6 of 20 vs 0 of 20 for E. coli, Fisher exact test: p < 0.05; Table 1). As previously shown, bacte-rial counts and lung weights were higher in the fungal colonized groups (Mann-Whitney U test: p < 0.001 and p < 0.01 for S. aureus and E. coli, respectively; Fig. 3, C and D).

Fungal Airway Colonization and Immune Response in Colonized LungsTo understand the mechanism linking fungal colonization to the development of subsequent bacterial pneumonia, we decided to study the immune response associated with fungal airway colonization. As shown previously, after a single intratracheal

instillation of C. albicans, the fungus was slowly cleared from the lungs over time. This airway colonization was found to be associated with significant increases in the levels of Th1-specific INF-γ and Th17-specific IL-17 whereas the Th2-specific IL-10 and IL-4 levels remained unchanged (Kruskal-Wallis test: p < 0.001, p < 0.05, p = 0.49, and p = 0.07, respectively; Fig. 4). Conversely, IL-2 decreased significantly (Kruskal-Wallis test: p < 0.01; Fig. 4). At 24 hours, this

Figure 1. Airway colonization with Candida albicans. Rats were intratracheally instilled with C. albicans (3 × 106 colony-forming unit [CFU]). At different time points, animals were killed and lungs homogenized. A, Fungal count was determined at 4 hr, 24 hr, 72 hr, 7 days, and 14 days after instillation. B, Body weight variation was compared at each time point with normal rats. Median and interquartile range are represented.

Roux et al

e194 www.ccmjournal.org September 2013 • Volume 41 • Number 9

cytokine profile of increased INF-γ and IL-17 and decreased IL-2, when compared to the control group with Dunn’s post hoc test (p < 0.001, p < 0.01, and p < 0.05, respectively), was consistent with a Th1-Th17 immune response. At 72 hours, only a tendency to an increased INF-γ level was observed (Fig. 4).

Phagocytosis by Alveolar Macrophages in the Presence of INF-γBecause high levels of INF-γ lead to a decrease in the expres-sion of scavenger receptors on AMs (20, 25), the major increase in INF-γ levels observed after fungal colonization led us to the hypothesis that an inhibition of the bacterial phagocytosis by AMs could explain the in vivo results. In the absence of stable fluorescent P. aeruginosa strain, com-mercially available red-labeled E. coli and S. aureus were tested in an in vitro model of bacterial phagocytosis (19) by AMs in the presence or absence of INF-γ.

Following an overnight culture with INF-γ, the phago-cytosis of unopsonized bacteria (red-fluorescent E. coli and S. aureus) by AMs (green-fluorescent membrane) was markedly decreased (Fig. 5A). In fact, the mean intracellu-lar fluorescence (red fluorescence measured inside the green membrane) was decreased by more than 90% in the presence of INF-γ as compared to control AMs (Mann-Whitney U test, p < 0.0001 for E. coli and S. aureus; Fig. 5B). As expected, opsonization of the bacteria, which allowed the Fcγ receptors

to recognize the bacteria and bypassed the need for scavenger receptors, allowed efficient phagocytosis in the presence of INF-γ with a 22% increase in the mean intracellular fluores-cence when opsonized E. coli were used (Mann-Whitney U test, p < 0.05; Fig. 5, A and B).

Effect of Antifungal Treatment on C. albicans Airway Colonization and Subsequent P. aeruginosa PneumoniaAfter studying the mechanism of action of fungal colonization on bacterial pneumonia, we investigated the effect of antifungal therapy on 1) the fungal count in the lung, 2) the local immune response raised during the course of the colonization, and 3) its potential to prevent bacterial pneumonia among colonized animals. The strain of C. albicans was susceptible to fluconazole and amphotericin B with minimum inhibitory concentrations equal to 0.50 and 0.047 µg/mL, respectively. In vivo, both treatments decreased fungal counts in lungs in comparison to the control group (Kruskal-Wallis test, p < 0.001; Fig. 6A) with a greater effect observed with amphotericin B than fluconazole (Dunn’s post hoc test, p < 0.01). Interestingly, the pulmonary levels of INF-γ were dramatically decreased by antifungal treatments as compared to the control group (Kruskal-Wallis test, p < 0.01; Fig. 6B). INF-γ levels in the treated groups did not differ from those in noncolonized rats (Dunn’s post hoc test, p > 0.05 between each treated groups and noncolonized group).

TABLE 1. Prevalence of Bacterial Pneumonia in Colonized and Control Rats

Bacterial StrainTime After

Colonization (hr)

No. With Pneumonia/No. of Rats

p (Chi-Square or Fisher Exact Testa)

C. albicans Airway Colonization Control

Pseudomonas aeruginosa 24 14/32 5/32 < 0.05

P. aeruginosa 72 13/31 6/32 < 0.05

Escherichia coli 24 6/20 0/20 < 0.01a

Staphylococcus aureus 24 14/15 8/18 < 0.01

Figure 2. Macroscopic aspects of Pseudomonas aeruginosa instilled lungs. A, Normal macroscopic aspect of lungs from a noncolonized rat that received 6 × 106 colony-forming unit (CFU) of P. aeruginosa PAO1 and did not develop pneumonia. B, Macroscopic aspect of lungs colonized with Candida albicans that develop pneumonia after an instillation of the same bacterial inoculum as in A. Note the marked hepatization of one pulmonary lobe.

Online Laboratory Investigations

Critical Care Medicine www.ccmjournal.org e195

In contrast, IL-17 concentrations were not significantly affected by antifungal treatments (Kruskal-Wallis test, p = 0.92; Fig. 6C).

Finally, we tested the effect of antifungal treatment in our model of P. aeruginosa pneumonia after fungal colonization. First, to avoid any direct effects of the antifungal treatment on P. aeruginosa pneumonia, we tested this treatment in a model of P. aeruginosa pneumonia in noncolonized animals using a higher bacterial inoculum. In the absence of fungal colonization, the antifungal treatment had no effect on the development of P. aeruginosa pneumonia with 100% (10 of 10) pneumonia in the treated group and 90% (9 of 10) in the control group (Fisher exact test, p = 1). The bacterial counts and lung weights did not differ between the treated and control groups (57,600 [37,750; 104,000] vs 46,000 [20,200; 80,650] CFU/instilled lung, Mann-Whitney U test: p = 0.35, and 1,917 [1,827; 1,993] vs 1,838 [1,673; 2,048] mg, Mann-Whitney U test: p = 0.53, respectively).

Second, in fungal-colonized animals, the antifungal treatment decreased the prevalence of P. aeruginosa pneumonia as compared to the untreated group (7 of 47 vs 15 of 46, respectively, chi-square: p < 0.05). Again, bacterial counts and lung weights were significantly lower in the treated group (Mann-Whitney U test: p < 0.05 and p < 0.001, respectively; Fig. 6D).

DISCUSSIONThis report confirms the increased prevalence of murine P. aeruginosa pneumonia observed in the context of C. albi-cans airway colonization (15). It extends this observation to two other important bacterial pathogens: S. aureus and E. coli; these three groups of bacterial pathogens account for up to 70–80% of the pathogens responsible for VAP (26, 27). As one possible contributing factor, we found that the high INF-γ

Figure 3. Effect of airway colonization by Candida albicans on the development of bacterial pneumonia. At day 0, anesthetized rats received either C. albicans (3 × 106 colony-forming unit [CFU], colonized group, closed squares and circles) or saline (control group, open squares and circles) in the trachea. Then a bacterial instillation was performed in a main bronchus: Pseudomonas aeruginosa (6 × 106 CFU) was instilled either 24 or 72 hr after the fungus (panel A or B, respectively); Escherichia coli (3 × 107 CFU, panel C) or S. aureus (2 × 106 CFU, panel D) were instilled 24 hr after the fungus. Rats were killed 24 hr after the bacterial instillation and lungs were aseptically removed. Bacterial counts (square) and lung weights (circle) were significantly higher in colonized groups (black) as compared to respective control groups (open). Bars represent the median. Mann-Whitney test: *p < 0.05; **p < 0.01; ***p < 0.001. Results are from at least two independent experiments.

Roux et al

e196 www.ccmjournal.org September 2013 • Volume 41 • Number 9

concentration observed in the lungs during fungal airway colo-nization inhibited bacterial phagocytosis by AMs. In addition, we showed that an antifungal treatment was able to normalize INF-γ levels and remove the facilitating effect of fungal colonization on bacterial pneumonia in vivo. We hypothesize that the presence of C. albicans in the airways could induce an immune response that may in turn decrease the normal antibacterial function of the immune cells, thus allowing bacterial pathogens to initiate an actual infection instead of being cleared by the immune system.

In this work, we first developed a simplified model of air-way colonization with a single instillation of C. albicans. This model fulfilled the predefined criteria of colonization, with no signs of patent infection (only mild airway inflammation was observed histologically, data not shown) and slow fungal clearance from the lungs (Fig. 1A), while the body weights of the colonized rats increased similarly to normal rats (Fig. 1B). This fungal airway colonization was responsible for a Th17 immune response with increased levels of IL-17 in the lungs and a decrease in IL-2, a cytokine known to repress the Th17 response (28). We also observed a very significant increase in the production of INF-γ, which is a major cytokine of the Th1 lineage. Interestingly, increased levels of INF-γ have recently been associated with a Th17 immune response (29–32). INF-γ could be secreted by the AMs themselves (33), by natural killer T cells (34), and also by INF-γ/IL-17 double producer T cells (29, 31, 32). The absence of increases in IL-4 and IL-10 concen-trations in the lungs indicated that the Th2 immune response was totally absent in our model. Thus, our model of fungal colonization induced a Th1-Th17 immune response with very high levels of INF-γ, as observed in other murine model of candidiasis (35) and in ex vivo human experiments (32). Interestingly, this Th1-Th17 combined immune response, which arose after immunization with the candidal Als3p adhe-sin (rAls3p-N), is required to protect against C. albicans (30).

AMs represent the primary immune cell line against inhaled pathogens that reach the alveoli. Because bacteria are

not initially opsonized by specific antibodies, internalization of the pathogen by professional phagocytes relies on scaven-ger receptors (36, 37). It has been shown that the expression of the major scavenger receptor of AMs (macrophage receptor with collagenous structure [38, 39]) decreases in the presence of INF-γ in both mice (20) and humans (25). Keeping in this line of reasoning, our finding of an increase in INF-γ levels, induced by fungal colonization, is interesting. Indeed, our in vitro model of phagocytosis showed that INF-γ strongly inhib-ited bacterial phagocytosis by AMs when the bacteria were not opsonized by specific antibodies. Thus, our results are in accor-dance with decreased AM phagocytosis due to a decrease in the expression of the AM scavenger receptors after activation by INF-γ as previously shown (20). Conversely, we observed an increase in bacterial phagocytosis by INF-γ-stimulated AMs after opsonization of the bacteria. This was expected because the expression of Fcγ receptors (especially Fcγ-R1), which bind opsonized bacteria, increases after INF-γ stimulation (40). This finding complements the previous observation that C. albicans could modulate the innate immune response mediated by AMs through a decreased production of reactive oxygen species (15).

The antifungal treatment of the colonized animals significantly decreased the risk for P. aeruginosa pneumonia in treated rats as compared to untreated colonized animals. While examining the mechanism for this result, we noticed a significant decrease in the INF-γ concentration in the lungs of treated rats. This observation confirmed the correlation between the increased production of INF-γ and an increased risk of bacterial pneumonia. The absence of IL-17 decrease after antifungal treatment underlines the com-plex interaction between fungal antigens and immune receptors, and thus may highlight the multiple immune pathways respon-sible for cytokine secretion after fungal recognition (41).

Our main hypothesis for the increased prevalence of murine bacterial pneumonia in the context of C. albicans airway colonization is based on a three-step mechanism. First, a Th1-Th17 immune response with an increase in the production of IL17 and INF-γ is induced by C. albicans colonization, without activation of the Th2 lineage. Second, as a consequence, the host antibacterial defense is impaired by the increased levels of INF-γ, with concurrent inhibition of the phagocytosis of nonopsonized bacteria by AMs, possibly through a decreased expression of scavenger receptors. Finally, these bacteria that have evaded the host immune defenses can more readily provoke pneumonia. Once again, this remains a hypothesis. It is however in agreement with the murine model developed by Sun and Metzger (20). They found an increased prevalence of Streptococcus pneumoniae pneumonia following a marked Th1 immune response with high levels of INF-γ observed during a Myxovirus influenzae infection in mice. Furthermore, these authors showed that INF-γ neutralization had a protective effect on the development of S. pneumoniae pneumonia, similar to the protective effect we observed during the antifungal treatment that normalized INF-γ levels in the lungs (Fig. 6).

In contrast to our results, a beneficial effect of short-term airway colonization by C. albicans on P. aeruginosa lung infection was recently reported (42). Interestingly, this in vivo model was developed with BALB/c mice. Contrary to our rat model, these

Figure 4. Cytokine levels in lungs before and after colonization with Candida albicans. Lungs of control rats or rats colonized with C. albicans (3 × 106 colony-forming unit [CFU] intratracheally) were homogenized. Interferon (INF)-γ, interleukin (IL)-2, IL-17, IL-10, and IL-4 concentrations were measured 24 and 72 hr after fungal colonization or a saline instillation (control). Median and interquartile range are represented. Kruskal-Wallis test: *p < 0.05; **p < 0.01; ***p < 0.001.

Online Laboratory Investigations

Critical Care Medicine www.ccmjournal.org e197

mice characteristically develop a Th2 immune response with increased IL-4 and IL-10 cytokines after intranasal instillation or mucosal challenge with C. albicans (43, 44). The pattern of cytokine production during fungal colonization thus appears to be animal model dependent. In our study, we observed a specific Th1-Th17 response, the same immune response that protects against candidiasis in humans (32, 45–47).

Strength and LimitsThere is very limited data on the kinetics of Candida airway colonization in ICU. In addition, there is no published model

of C. albicans airway coloniza-tion apart from our initial one (15). Hence, it was difficult for us to set the exact timing between Candida instillation and secondary bacterial chal-lenge. We adjusted our initial model by increasing the dura-tion of Candida colonization so as to approximate the clini-cal setting. Second, to comply with ethical regulations on limitation of animal experi-mentation, we did not repeat experiments performed in our previous article clearly indicat-ing that the observed increase in susceptibility to bacterial infection was not a nonspe-cific inflammatory response (15). Indeed, administration of killed C. albicans elicited no inflammatory response and no increase in bacterial count.

A reasonable proportion of the colonized population showed CFU results similar to the noncolonized group (Fig. 3). As we were using outbred rats (Wistar), a possible explanation was a differential host response within animals. However, the cytokine concentrations were homogeneously distributed within the colonized group, suggesting certain homogeneity in the host response. Furthermore, as the bacteria were instilled in a main bronchus, pneumonia was predominantly unilateral. Thus, the volume of the infected lung was frequently modest as compared to the whole lung volume that was analyzed,

resulting in a moderate increase in the CFU count of animals developing pneumonia. Finally, we observed a greater effect of fungal colonization on the development of pneumonia for both E. coli and S. aureus as compared to P. aeruginosa (Fig. 3), this could be due to the low virulence of the PAO1 strain in our model. Indeed, the well-described laboratory strain PA01 does not produce the effector protein ExoU, a major virulence factor in P. aeruginosa lung infection (24).

Our results provide a plausible explanation for the asso-ciation between airway colonization with Candida species and subsequent P. aeruginosa VAP in ICU patients (13): a reaction

Figure 5. Effect of interferon-gamma (INF-γ) on bacterial phagocytosis by alveolar macrophages. Alveolar macrophages recovered from bronchoalveolar lavage of normal rats were incubated with recombinant rat INF-γ or its control. After incubation with Texas Red-coupled bacteria (red-labeled unopsonized Escherichia coli and Staphylococcus aureus, or opsonized E. coli), cell surface was stained in green with a biotin primary antibody anti-rat CD11b. Images were taken with a confocal microscope. A, In the absence of INF-γ, unopsonized or opsonized bacteria were internalized (red fluorescence inside green-stained cell surface). After INF-γ incubation, alveolar macrophages were unable to internalize unopsonized bacteria. This was not the case after bacterial opsonization. B, Mean red fluorescence intensity (MFI) inside green-stained alveolar macrophages was dramatically decreased after overnight culture with INF-γ when unopsonized E. coli or unopsonized S. aureus were used. Inversely, INF-γ slightly increased MFI when using opsonized E. coli. Mann-Whitney test: *p < 0.05; ***p < 0.001. Median and interquartile range are represented.

Roux et al

e198 www.ccmjournal.org September 2013 • Volume 41 • Number 9

of the host immune system to the presence of the nonpatho-genic Candida leading to a reduced ability to clear pathogenic bacteria. The finding that antifungal treatment normalized INF-γ levels in colonized airways and allowed the host immune system to efficiently protect against subsequent bacterial chal-lenge is in concurrence with the interesting results of Nseir et al (48), who found a link between antifungal treatment and a decreased prevalence of P. aeruginosa pneumonia among venti-lated patients with airway colonization by Candida spp. Overall, our results may have some clinical consequences, given the high proportion of ventilated ICU patients with C. albicans airway colonization (7, 8) and the growing concern regarding the impact of this colonization on bacterial pneumonia (7, 13, 48).

CONCLUSIONSOur results suggest that C. albicans airway colonization could facilitate the development of bacterial pneumonia. In vivo, this effect was blunted by antifungal treatment. A possible

mechanism for these findings is the inhibition of AM phagocy-tosis in relation with a Th1-Th17 immune response triggered by the fungal colonization.

ACKNOWLEDGMENTSWe thank Gerald B. Pier for reviewing the manuscript.

REFERENCES 1. Chastre J, Fagon JY: Ventilator-associated pneumonia. Am J Respir

Crit Care Med 2002; 165:867–903 2. Markowicz P, Wolff M, Djedaïni K, et al: Multicenter prospective study

of ventilator-associated pneumonia during acute respiratory dis-tress syndrome. Incidence, prognosis, and risk factors. ARDS Study Group. Am J Respir Crit Care Med 2000; 161:1942–1948

3. Hubmayr RD, Burchardi H, Elliot M, et al; American Thoracic Soci-ety Assembly on Critical Care; European Respiratory Society; Euro-pean Society of Intensive Care Medicine; Société de Réanimation de Langue Française: Statement of the 4th International Consensus Conference in Critical Care on ICU-Acquired Pneumonia–Chicago, Illinois, May 2002. Intensive Care Med 2002; 28:1521–1536

Figure 6. Effect of antifungal treatment on airway fungal colonization and on Pseudomonas aeruginosa pneumonia. All rats were colonized with an intratracheal instillation of Candida albicans (3 × 106 colony-forming unit [CFU]). They received amphotericin B, fluconazole, or placebo daily by intraperitoneal injection just after fungal instillation and during two more days. A, Twenty-four hours after the third IP injection, fungal counts were significantly decreased by antifungal treatments with a greater effect of amphotericin B (Kruskal-Wallis test, p < 0.0001, Dunn’s multiple comparison test). B, Lung levels of interferon-gamma (INF-γ) decreased significantly and were not different from levels observed in noncolonized rats (Kruskal-Wallis test, p < 0.01, Dunn’s multiple comparison test). C, Interleukin (IL)-17 lung levels did not exhibit any significant differences with or without antifungal treatments (Kruskal-Wallis test, p = 0.82). D, Colonized animals that received either three daily doses of amphotericin B or control by intraperitoneal injection were instilled with P. aeruginosa (6 × 106 CFU) into a main bronchus 24 hr after the third injection. Rats were killed 24 hr after the bacterial instillation and lungs evaluated. Bacterial counts (squares) and lung weights (circles) were significantly lower in the treated group (gray) as compared to the control group (black, Mann-Whitney test). Bars represent the median on A and D; median and interquartile range are represented on B and C. *p < 0.05; **p < 0.01; ***p < 0.001.

Online Laboratory Investigations

Critical Care Medicine www.ccmjournal.org e199

4. Bouadma L, Deslandes E, Lolom I, et al: Long-term impact of a multi-faceted prevention program on ventilator-associated pneumonia in a medical intensive care unit. Clin Infect Dis 2010; 51:1115–1122

5. Nguile-Makao M, Zahar JR, Français A, et al: Attributable mortality of ventilator-associated pneumonia: Respective impact of main character-istics at ICU admission and VAP onset using conditional logistic regres-sion and multi-state models. Intensive Care Med 2010; 36:781–789

6. American Thoracic Society; Infectious Diseases Society of America: Guidelines for the management of adults with hospital-acquired, ventilator-associated, and healthcare-associated pneumonia. Am J Respir Crit Care Med 2005; 171:388–416

7. Hamet M, Pavon A, Dalle F, et al: Candida spp. airway colonization could promote antibiotic-resistant bacteria selection in patients with suspected ventilator-associated pneumonia. Intensive Care Med 2012; 38:1272–1279

8. Meersseman W, Lagrou K, Spriet I, et al: Significance of the isolation of Candida species from airway samples in critically ill patients: A pro-spective, autopsy study. Intensive Care Med 2009; 35:1526–1531

9. el-Ebiary M, Torres A, Fàbregas N, et al: Significance of the isola-tion of Candida species from respiratory samples in critically ill, non-neutropenic patients. An immediate postmortem histologic study. Am J Respir Crit Care Med 1997; 156(2, Part 1):583–590

10. Rello J, Esandi ME, Díaz E, et al: The role of Candida sp isolated from bronchoscopic samples in nonneutropenic patients. Chest 1998; 114:146–149

11. Eggimann P, Garbino J, Pittet D: Management of Candida species infections in critically ill patients. Lancet Infect Dis 2003; 3:772–785

12. Ricard JD, Roux D: Candida pneumonia in the ICU: Myth or reality? Intensive Care Med 2009; 35:1500–1502

13. Azoulay E, Timsit JF, Tafflet M, et al; Outcomerea Study Group: Can-dida colonization of the respiratory tract and subsequent pseudomo-nas ventilator-associated pneumonia. Chest 2006; 129:110–117

14. Peleg AY, Hogan DA, Mylonakis E: Medically important bacterial-fun-gal interactions. Nat Rev Microbiol 2010; 8:340–349

15. Roux D, Gaudry S, Dreyfuss D, et al: Candida albicans impairs mac-rophage function and facilitates Pseudomonas aeruginosa pneumo-nia in rat. Crit Care Med 2009; 37:1062–1067

16. Ricard JD, Roux D: Candida colonization in ventilated ICU patients: No longer a bystander! Intensive Care Med 2012; 38:1243–1245

17. Schortgen F, Bouadma L, Joly-Guillou ML, et al: Infectious and inflam-matory dissemination are affected by ventilation strategy in rats with unilateral pneumonia. Intensive Care Med 2004; 30:693–701

18. Levert M, Zamfir O, Clermont O, et al: Molecular and evolutionary bases of within-patient genotypic and phenotypic diversity in Esche-richia coli extraintestinal infections. PLoS Pathog 2010; 6:e1001125

19. Pinheiro da Silva F, Aloulou M, Skurnik D, et al: CD16 promotes Esche-richia coli sepsis through an FcR gamma inhibitory pathway that prevents phagocytosis and facilitates inflammation. Nat Med 2007; 13:1368–1374

20. Sun K, Metzger DW: Inhibition of pulmonary antibacterial defense by interferon-gamma during recovery from influenza infection. Nat Med 2008; 14:558–564

21. Marchetti O, Moreillon P, Glauser MP, et al: Potent synergism of the combination of fluconazole and cyclosporine in Candida albicans. Antimicrob Agents Chemother 2000; 44:2373–2381

22. Perfect JR, Durack DT: Comparison of amphotericin B and N-d-or-nithyl amphotericin B methyl ester in experimental cryptococcal meningitis and Candida albicans endocarditis with pyelonephritis. Antimicrob Agents Chemother 1985; 28:751–755

23. Witt MD, Imhoff T, Li C, et al: Comparison of fluconazole and amphotericin B for treatment of experimental Candida endocarditis caused by non-C. albicans strains. Antimicrob Agents Chemother 1993; 37:2030–2032

24. Stover CK, Pham XQ, Erwin AL, et al: Complete genome sequence of Pseudomonas aeruginosa PAO1, an opportunistic pathogen. Nature 2000; 406:959–964

25. Raju B, Hoshino Y, Kuwabara K, et al: Aerosolized gamma interferon (IFN-gamma) induces expression of the genes encoding the IFN-gamma-inducible 10-kilodalton protein but not inducible nitric oxide synthase in the lung during tuberculosis. Infect Immun 2004; 72:1275–1283

26. Esperatti M, Ferrer M, Theessen A, et al: Nosocomial pneumonia in the intensive care unit acquired by mechanically ventilated versus nonven-tilated patients. Am J Respir Crit Care Med 2010; 182:1533–1539

27. Koulenti D, Lisboa T, Brun-Buisson C, et al; EU-VAP/CAP Study Group: Spectrum of practice in the diagnosis of nosocomial pneumo-nia in patients requiring mechanical ventilation in European intensive care units. Crit Care Med 2009; 37:2360–2368

28. Liao W, Lin JX, Leonard WJ: IL-2 family cytokines: New insights into the complex roles of IL-2 as a broad regulator of T helper cell differen-tiation. Curr Opin Immunol 2011; 23:598–604

29. Fenoglio D, Poggi A, Catellani S, et al: Vdelta1 T lymphocytes producing IFN-gamma and IL-17 are expanded in HIV-1-infected patients and respond to Candida albicans. Blood 2009; 113:6611–6618

30. Lin L, Ibrahim AS, Xu X, et al: Th1-Th17 cells mediate protective adap-tive immunity against Staphylococcus aureus and Candida albicans infection in mice. PLoS Pathog 2009; 5:e1000703

31. Tanaka S, Yoshimoto T, Naka T, et al: Natural occurring IL-17 produc-ing T cells regulate the initial phase of neutrophil mediated airway responses. J Immunol 2009; 183:7523–7530

32. Zielinski CE, Mele F, Aschenbrenner D, et al: Pathogen-induced human TH17 cells produce IFN-γ or IL-10 and are regulated by IL-1β. Nature 2012; 484:514–518

33. Darwich L, Coma G, Peña R, et al: Secretion of interferon-gamma by human macrophages demonstrated at the single-cell level after costimulation with interleukin (IL)-12 plus IL-18. Immunology 2009; 126:386–393

34. Mattarollo SR, Rahimpour A, Choyce A, et al: Invariant NKT cells in hyperplastic skin induce a local immune suppressive environment by IFN-gamma production. J Immunol 2010; 184:1242–1250

35. van de Veerdonk FL, Joosten LAB, Shaw PJ, et al: The inflammasome drives protective Th1 and Th17 cellular responses in disseminated candidiasis. Eur J Immunol 2011; 41:2260–2268

36. Kobzik L: Lung macrophage uptake of unopsonized environmental partic-ulates. Role of scavenger-type receptors. J Immunol 1995; 155:367–376

37. Peiser L, Gordon S: The function of scavenger receptors expressed by macrophages and their role in the regulation of inflammation. Microbes Infect 2001; 3:149–159

38. Arredouani MS, Palecanda A, Koziel H, et al: MARCO is the major binding receptor for unopsonized particles and bacteria on human alveolar macrophages. J Immunol 2005; 175:6058–6064

39. Palecanda A, Paulauskis J, Al-Mutairi E, et al: Role of the scavenger receptor MARCO in alveolar macrophage binding of unopsonized environmental particles. J Exp Med 1999; 189:1497–1506

40. Kårehed K, Dimberg A, Dahl S, et al: IFN-gamma-induced upregu-lation of Fcgamma-receptor-I during activation of monocytic cells requires the PKR and NFkappaB pathways. Mol Immunol 2007; 44:615–624

41. Netea MG, Maródi L: Innate immune mechanisms for recognition and uptake of Candida species. Trends Immunol 2010; 31:346–353

42. Ader F, Jawhara S, Nseir S, et al: Short term Candida albicans colo-nization reduces Pseudomonas aeruginosa-related lung injury and bacterial burden in a murine model. Crit Care 2011; 15:R150

43. Elahi S, Pang G, Clancy R, et al: Cellular and cytokine correlates of mucosal protection in murine model of oral candidiasis. Infect Immun 2000; 68:5771–5777

44. Londoño P, Gao XM, Bowe F, et al: Evaluation of the intranasal chal-lenge route in mice as a mucosal model for Candida albicans infec-tion. Microbiology 1998; 144(Part 8):2291–2298

45. Conti HR, Gaffen SL: Host responses to Candida albicans: Th17 cells and mucosal candidiasis. Microbes Infect 2010; 12:518–527

46. Gaffen SL, Hernández-Santos N, Peterson AC: IL-17 signaling in host defense against Candida albicans. Immunol Res 2011; 50:181–187

47. Van Der Meer JWM, van de Veerdonk FL, Joosten LAB, et al: Severe Candida spp. infections: New insights into natural immunity. Int J Antimicrob Agents 2010; 36(Suppl 2):S58–S62

48. Nseir S, Jozefowicz E, Cavestri B, et al: Impact of antifungal treatment on Candida-Pseudomonas interaction: A preliminary retrospective case-control study. Intensive Care Med 2007; 33:137–142