|Home | About | Journals | Submit | Contact Us | Français|
Candida species are the most common source of nosocomial invasive fungal infections. Previous studies have indicated that T-helper immune response is the critical host factor for susceptibility to Candida infection. The transcription factor GATA-3 is known as the master regulator for T-helper type 2 (Th2) differentiation. We therefore investigated the role of GATA-3 in the host defense against systemic Candida infection using GATA-3-overexpressing transgenic mice. The survival of GATA-3-overexpressing mice after Candida infection was significantly lower than that of wild-type mice. Candida outgrowth was significantly increased in the kidneys of GATA-3-overexpressing mice, compared with wild-type mice. The levels of various Th2 cytokines, including interleukin-4 (IL-4), IL-5, and IL-13, were significantly higher while the level of Th1 cytokine gamma interferon was significantly lower in the splenocytes of GATA-3-overexpressing mice after Candida infection. Recruitment of macrophages into the peritoneal cavity in response to Candida infection and their phagocytic activity were significantly lower in GATA-3-overexpressing mice than in wild-type mice. Exogenous administration of gamma interferon to GATA-3-overexpressing mice significantly reduced Candida outgrowth in the kidney and thus increased the survival rate. Administration of gamma interferon also increased the recruitment of macrophages into the peritoneal cavity in response to Candida infection. These results indicate that overexpression of GATA-3 modulates macrophage antifungal activity and thus enhances the susceptibility to systemic Candida infection, possibly by reducing the production of gamma interferon in response to Candida infection.
Candida albicans is an opportunistic fungal pathogen causative of the most widespread nosocomial bloodstream infections worldwide (25, 35). Candida is part of the normal microbial flora in human beings and domestic animals and is associated with mucosal surfaces of the oral cavity, gastrointestinal tract, and vagina. However, under conditions of immune dysfunction, Candida switches from a commensal to a pathogenic organism capable of infecting a variety of tissues and can subsequently cause fatal systemic diseases (2, 7).
Several host factors involved in the innate and adaptive immune systems are required for the systemic control of Candida infection. In mice, resistance and susceptibility to systemic Candida infection may be linked, at least in part, to the dissimilar expansion of functionally distinct CD4-positive T-cell subsets. Previous studies have demonstrated that resistance to Candida infection is associated with T-helper type 1 (Th1) immunity, whereas type 2 (Th2) immunity is associated with susceptibility to symptomatic infection (28, 29). Activation of innate immunity, such as the ability of phagocytic cells to inhibit fungal growth, is required for the induction of Th1 cells (19, 20). The Th1 cytokine gamma interferon (IFN-γ) is required for optimal activation of phagocytes such as neutrophils and macrophages, collaborates in the generation of protective antibody response, and favors the development of a Th1 protective response (9). Early in Candida infection, another Th1 cytokine, tumor necrosis factor alpha (TNF-α), is also essential for the successful control of the infection and the resulting protective Th1-dependent immunity (28). In contrast, Th2 cytokines, such as interleukin-4 (IL-4) and IL-10, inhibit Th1 development and deactivate phagocytic effector cells.
Th1 and Th2 cells are differentiated from common T precursor cells (1, 16), and their differentiation requires the activity of distinct transcription factors. Among a variety of key molecules governing Th1/Th2 differentiation, GATA-3 has been implicated in Th2 commitment (22, 34, 37, 39). Under physiological conditions, GATA-3 is selectively expressed in Th2 but not Th1 cells (37, 38, 39). Transgenic and retroviral expression of GATA-3 induces a Th2 cytokine profile in Th1 cells (37, 39), while dominant-negative GATA-3 downregulates Th2 clones (38). GATA-3 is now recognized as a master regulator for Th2 cell differentiation. It is therefore likely that GATA-3 is a critical host factor in the determination of susceptibility to systemic Candida infection.
We recently established GATA-3-overexpressing transgenic (GATA-3-tg) mice. In the present study, we investigate the role of GATA-3 in the host defense against disseminated candidiasis using these mice.
GATA-3-tg mice were generated as previously described (36). GATA-3-tg mice were backcrossed with C57BL/6 mice for eight generations, and the sufficiency of backcross was confirmed by analyzing the polymorphism of microsatellite DNAs. The genetic background was uniformized to C57BL/6 by backcrossing for eight generations. C57BL/6 wild-type (WT) mice were purchased from Charles River Breeding Laboratories (Kanagawa, Japan). All mice used in this study were 8 to 12 weeks old and were maintained in our animal facilities under specific-pathogen-free conditions. All animal studies were approved by the Institutional Review Board.
C. albicans was obtained from the American Type Culture Collection (Manassas, VA). Mice were infected by intraperitoneal inoculation of 1 × 108 Candida blastoconidia in 500 μl of sterile pyrogen-free phosphate-buffered saline (PBS) isolated from fungal cultures made in YM broth (Difco Laboratories, Detroit, MI) for 48 h at 24°C. The survival of mice after C. albicans infection was assessed daily for 56 days. At 1, 3, 7, 30, and 56 days after the inoculation, the blood and organs (i.e., lung, heart, liver, spleen, and kidney) were removed aseptically. Quantification of C. albicans was performed by plating several dilutions of homogenized organs in Sabouraud dextrose agar. The results were expressed as log10 CFU per organ. For the histological analysis, the tissues were immediately fixed in formalin. Sections (4 μm) of paraffin-embedded tissues were stained with periodic acid-Schiff (PAS) reagent.
The peritoneal cavity was lavaged with five sequential 2-ml aliquots of PBS. The cells were centrifuged and resuspended in PBS. Cells were counted using a hemocytometer, and a differential cell count was performed by standard light microscopic techniques. The percentage of macrophages and neutrophils phagocytosing C. albicans was evaluated by staining with PAS reagent. A minimum of 300 cells were evaluated in each preparation. To obtain peritoneal macrophages, the cells were stained with anti-F4/80 antibody (BD PharMingen, San Diego, CA) and were sorted with a MoFlo flow cytometry system using Summit software (Beckman Coulter, Inc., Brea, CA).
The expression levels of IFN-γ, TNF-α, IL-4, IL-5, IL-13, monocyte chemotactic protein 1 (MCP-1), keratinocyte-derived chemokine (KC), and macrophage inflammatory protein 2 (MIP-2) genes were determined by quantitative real-time reverse transcription-PCR (RT-PCR) using ready-made fluorogenic probes and primers (Applied Biosystems, Foster City, CA). Expression levels for amplicons were quantified using the threshold cycle (ΔΔCT) method according to the manufacturer's protocols. The expression levels were normalized against 18S rRNA.
Seven days after C. albicans inoculation, the spleen was removed, minced, and passed through a fine steel mesh to obtain a homogeneous cell suspension. Erythrocytes were hemolyzed by using NH4Cl solution. The residual cells were then filtered through a 20-μm nylon mesh. Cells were stained with anti-T-cell receptor β (anti-TCRβ), anti-CD4, and anti-CD8 (BD PharMingen) and anti-CXCR3 and anti-CCR3 (R&D Systems, Inc., Minneapolis, MN). After being stained, the cells were analyzed by flow cytometry using a FACSCalibur with CellQuest software (Biosciences, San Jose, CA).
Levels of IFN-γ production in splenocytes were determined by flow cytometric intracellular cytokine analysis as previously described (21). Briefly, the cells were resuspended in RPMI 1640 containing 10% fetal calf serum (FCS), incubated with phorbol myristate acetate (PMA; 50 ng/ml; Sigma-Aldrich) and ionomycin (500 ng/ml; Sigma) for 2 h, and then incubated with brefeldin A (10 μg/ml; Sigma) for 2 h at 37°C. The cells were stained with phycoerythrin (PE)-conjugated anti-mouse IFN-γ antibody (BD PharMingen) and fixed with 2% paraformaldehyde-PBS solution. Data were analyzed by flow cytometry.
GATA-3-tg mice were daily treated with 8 × 106 U of murine recombinant IFN-γ (BD PharMingen) or diluent subcutaneously for 14 consecutive days.
Data were expressed as the means ± standard errors of the means (SEM). Comparisons of data among the experimental groups were performed using the analysis of variance and Scheffe's test. The survival curves were analyzed using the log-rank test. Values of P < 0.05 were considered to be statistically significant.
We first assessed the survival rate following infection of WT mice and GATA-3-tg mice with C. albicans. The mortality of GATA-3-tg mice was significantly higher than that of WT mice after either intraperitoneal inoculation or intravenous inoculation of C. albicans. Sixty percent of GATA-3-tg mice died while 25% of the WT mice died within 56 days after the intraperitoneal inoculation (Fig. (Fig.1A).1A). Intravenous fungal inoculation caused early death in both genotypes of mice, compared with intraperitoneal inoculation. All GATA-3-tg mice died within 15 days while 50% of the WT mice survived to 18 days after the intravenous inoculation (Fig. (Fig.1B).1B). No mice died in the saline-administered control group of either genotype (Fig. 1A and B).
We next assessed the growth and systemic spread of C. albicans in WT mice and GATA-3-tg mice. The outgrowth of C. albicans was most frequent in the kidney of GATA-3-tg mice after infection by either the intraperitoneal or intravenous route. Candida CFU per tissue were significantly higher in the kidneys of GATA-3-tg mice than in those of WT mice at 56 days post-intraperitoneal inoculation (Fig. (Fig.2A).2A). Candida outgrowth was detected in all kidneys and 5/10 lungs of GATA-3-tg mice, whereas it was detected in none of the kidneys and 1/10 lungs of WT mice at that time point (Table (Table1).1). C. albicans was detected in all lungs and kidneys of both genotypes of mice at 14 days post-intravenous inoculation (Table (Table1).1). However, Candida CFU per tissue were significantly higher in both the lungs and kidneys of GATA-3-tg mice than in those of WT mice at that time point (Fig. (Fig.2B2B).
Histopathological analysis also revealed severe renal damage with bilateral hydronephrosis in GATA-3-tg mice after C. albicans infection via either an intraperitoneal or an intravenous route (Fig. (Fig.2C,2C, GATA3-tg). Marked fungal growth with hyphal formation was observed with the infiltration of numerous inflammatory cells into the pelvis of GATA-3-tg mice (Fig. (Fig.2C,2C, inset). In contrast, the damage was much less severe and no obvious hydronephrosis was observed in the kidneys of WT mice infected via either an intraperitoneal or an intravenous route (Fig. (Fig.2C,2C, WT).
To clarify the contribution of GATA-3 overexpression to Th1/Th2 balance after C. albicans infection, we evaluated the expression of both Th1 and Th2 cytokines in the splenocytes of WT mice and GATA-3-tg mice after fungal inoculation. The expression of the Th1 cytokine IFN-γ was elevated in the splenocytes of WT mice after the inoculation. The level of IFN-γ was significantly higher in the splenocytes of WT mice than in those of GATA-3-tg mice at 3, 7, and 30 days postinoculation (Fig. (Fig.3;3; IFN-γ). The expression of TNF-α was elevated in the splenocytes of GATA-3-tg mice 3 days postinoculation, and the level of TNF-α was significantly higher in WT mice than in GATA-3-tg mice at that time point (Fig. (Fig.3,3, TNF-α). On the other hand, the expression of Th2 cytokine IL-4 was elevated in the splenocytes of GATA-3-tg mice but not those of WT mice after C. albicans inoculation. The IL-4 level in GATA-3-tg mice was significantly higher than that in WT mice at 3, 7, and 30 days postinoculation (Fig. (Fig.3,3, IL-4). The levels of IL-5 and IL-13 were also significantly higher in the splenocytes of GATA-3-tg mice than in those of WT mice at 7 and 30 days postinoculation (Fig. (Fig.3,3, IL-5 and IL-13). These results indicate that GATA-3-tg mice are highly biased toward Th2 due to impaired induction of IFN-γ and enhanced induction of IL-4, IL-5, and IL-13 after Candida infection.
Previous studies demonstrated that IFN-γ plays an important role in protection against Candida infection. We assessed the IFN-γ production ability of the splenocytes of WT mice and GATA-3-tg mice. We first evaluated the ratio of CD4- to CD8-positive T cells in the spleens of both WT mice and GATA-3-tg mice before and 7 days postinoculation. The ratio of CD4- to CD8-positive cells was not different between the genotypes irrespective of Candida infection (Fig. (Fig.4A).4A). In both genotypes of mice, the number of CD4-positive cells was three to four times greater than that of CD8-positive cells. We then assessed the proportion of IFN-γ-producing cells among CD4- and CD8-positive cells. The proportion of IFN-γ-producing cells in CD4-positive cells increased after Candida inoculation in WT mice but not in GATA-3-tg mice. The proportion was significantly higher in WT mice than in GATA-3-tg mice before and 7 days postinoculation (Fig. (Fig.4B,4B, left panel). The proportion of IFN-γ-producing cells in CD8-positive cells was not different between the genotypes irrespective of Candida infection (Fig. (Fig.4B,4B, right panel). We then assessed the proportion of CXCR3-positive cells and CCR3-positive cells in CD4-positive cells; these are the cell surface markers of Th1 and Th2 cells, respectively. The proportion of CXCR3-positive cells was not different between the genotypes before Candida inoculation. The proportion of CXCR3-positive cells decreased significantly in GATA-3-tg mice but not in WT mice 7 days postinoculation (Fig. (Fig.4C,4C, left panel). Therefore, the proportion of CXCR3-positive cells was significantly higher in WT mice than in GATA-3-tg mice at that time point (Fig. (Fig.4C,4C, left panel). The proportion of CCR3-positive cells was not different between the genotypes either before or 7 days postinoculation (Fig. (Fig.4C,4C, right panel). We further assessed the proportion of IFN-γ-producing cells in CXCR3-positive cells. The proportion increased after Candida infection in WT mice but not in GATA-3-tg mice. The proportion was significantly higher in WT mice than in GATA-3-tg mice 7 days postinoculation (Fig. (Fig.4D4D).
We then assessed inflammatory cell infiltration into the peritoneal cavity, the site of infection. The number of total lavageable cells was increased in WT mice but not in GATA-3-tg mice 7 days postinoculation, compared with the number before inoculation (Fig. (Fig.5A,5A, total cells). The number of total lavageable cells was significantly higher in WT mice than in GATA-3-tg mice at that time point (Fig. (Fig.5A,5A, total cells). The number of macrophages was also significantly higher in WT mice than in GATA-3-tg mice 7 days postinoculation (Fig. (Fig.5A,5A, macrophages). No difference in the number of neutrophils or lymphocytes was observed between genotypes (Fig. (Fig.5A,5A, neutrophils and lymphocytes). Since the number of macrophages was increased in response to Candida infection in the peritoneal cavity of WT mice, we next assessed the expression of chemokines in peritoneal macrophages. Although the expression of MCP-1, a CC chemokine, was elevated in peritoneal macrophages of both genotypes 7 days postinoculation, the expression level was significantly higher in peritoneal macrophages of WT mice than in those of GATA-3-tg mice (Fig. (Fig.5B,5B, MCP-1). Similarly, the expression levels of KC and MIP-2, macrophage-derived CXC chemokines, were significantly higher in peritoneal macrophages of WT mice than in those of GATA-3-tg mice 7 days postinoculation (Fig. (Fig.5B,5B, KC and MIP-2).
We further assessed phagocytic activity of peritoneal macrophages and neutrophils histologically. The engulfment of C. albicans by both macrophages and neutrophils was observed most frequently at 3 days postinoculation in both genotypes of mice (Fig. (Fig.6A).6A). However, the numbers of macrophages and neutrophils phagocytosing C. albicans were significantly higher in GATA-3-tg mice than in WT mice both 3 days and 7 days postinoculation (Fig. 6B and C).
Because the production of IFN-γ was significantly lower in the spleen and its CD4- and CXCR3-positive cells in GATA-3-tg mice, we next assessed whether exogenous supplementation with IFN-γ would reduce the development of disseminated candidiasis. Mouse recombinant IFN-γ was administered to GATA-3-tg mice subcutaneously for 14 consecutive days. We first evaluated the survival rate following infection of GATA-3-tg mice with C. albicans with or without IFN-γ administration. The survival rate was significantly higher in GATA-3-tg mice treated with IFN-γ than in the vehicle-treated controls. Fifty-five percent of vehicle-treated GATA-3-tg mice died within 56 days postinoculation (Fig. (Fig.7A).7A). However, the mortality was reduced to 30% in GATA-3-tg mice treated with IFN-γ (Fig. (Fig.7A).7A). No mice died in the saline-administered control group of either genotype (Fig. (Fig.7A).7A). We next assessed the growth and systemic spread of C. albicans in GATA-3-tg mice with or without IFN-γ administration at 56 days postinoculation. The growth of C. albicans was most frequently observed in the kidney irrespective of IFN-γ administration (Fig. (Fig.7B).7B). However, the degree of Candida outgrowth was much less severe in GATA-3-tg mice treated with IFN-γ than in those without IFN-γ administration (Fig. (Fig.7B).7B). We further assessed inflammatory cell infiltration into the peritoneal cavity at 7 days after Candida inoculation with or without IFN-γ administration. The numbers of total lavageable cells and macrophages were significantly higher in GATA-3-tg mice treated with IFN-γ than in those not receiving IFN-γ administration (Fig. (Fig.7C,7C, total cells and macrophages). No significant difference in the number of neutrophils and lymphocytes was observed between IFN-γ-treated and vehicle-treated mice (Fig. (Fig.7C,7C, neutrophils and lymphocytes).
In the present study, we demonstrated for the first time that systemic infection with C. albicans is much more severe in GATA-3-tg mice than in WT mice of the same background. Analysis of cytokines revealed that overexpression of GATA-3 shifted the Th1/Th2 balance toward Th2 after Candida infection, whereas the balance was shifted to Th1 in WT mice by the same infection. GATA-3 is a member of the GATA family of zinc finger transcription factors, which bind the GATA consensus motif (11). As stated above, GATA-3 is best known to function as a master regulator of Th2 cell differentiation. In addition to Th2 cell differentiation, GATA-3 plays a central role in Th2 cytokine production. It has been demonstrated that antisense GATA-3 inhibited the expression of all Th2 cytokine genes in the Th2 clone D10 (39). In transgenic mice, elevated GATA-3 in CD4-positive T cells caused Th2 cytokine gene expression in developing Th1 cells (39). GATA-3 has been reported to transactivate the IL-5 promoter with only limited effects on IL-4 gene transcription (24, 37, 38). It has also been reported that GATA-3 regulates the locus accessibility of the IL-4 and IL-13 genes with chromatin remodeling (15, 33). These findings suggest that GATA-3 allows the expression of Th2 cytokines by functioning as a transcription factor as well as by modifying the chromatin structure of these cytokines.
In the present study, the IFN-γ level was significantly lower in the spleen of GATA-3-tg mice than in that of WT mice after Candida infection. Several studies have demonstrated that GATA-3 not only transactivates Th2 cytokines but also suppresses Th1 cytokine expression. It was reported that GATA-3 significantly downregulated IFN-γ production during in vitro Th1 differentiation of naïve CD4-positive T cells through downregulation of IL-12 receptor β2 and IFN-γ production (8, 22). In contrast, IFN-γ production in CD4+ T cells of GATA-3-deficient mice was increased even under a Th2 condition (23). GATA-3 suppresses IFN-γ gene transcription via the downregulation of signal transducer and activator of transcription 4 (STAT4) (12). In addition to suppression of IFN-γ gene transcription, GATA-3 inhibits IFN-γ production indirectly by inhibiting Th1 cell differentiation via repression of T-bet activation (10). Thus, it is reasonable to conclude that GATA-3 overexpression suppresses IFN-γ production in addition to upregulating Th2 cytokine production in the present model.
In the present study, exogenous administration of IFN-γ improved the degree of systemic Candida infection in GATA-3-tg mice. These findings suggest that the aggravation of Candida infection in GATA-3-tg mice is at least partly due to the reduction of IFN-γ production in response to Candida infection. Correspondingly, it has been reported that IFN-γ-knockout mice are highly susceptible to disseminated candidiasis induced by intraperitoneal inoculation of C. albicans blastoconidia with impairment of macrophage candidacidal activity (13). There is abundant evidence that IFN-γ is the major cytokine involved in activation of macrophages and neutrophils to a candidacidal stage (3, 9). The administration of IFN-γ to mice beneficially influences the course of experimental infection with various pathogens, such as Toxoplasma gondii (32), Leishmania major (17), Trypanosoma cruzi (6), and Pneumocystis murina (30). In most models, macrophages and/or neutrophils are the main effector cells of IFN-γ-stimulated microbicidal activity. Macrophages in particular play a key role in the innate immunity to Candida infections by engulfing, killing, and processing the pathogen for presentation to T cells. Consistent with these results, the phagocytic activity of macrophages and neutrophils was reduced in GATA-3-tg mice in the present study. In addition to phagocytosis, macrophages modulate immune and inflammatory responses by producing a number of cytokines, CC chemokines, and CXC chemokines.
MCP-1, also called CCL2, is a member of the CC chemokine family. MCP-1-mediated monocyte recruitment is essential for defense against various bacterial, protozoal, and fungal pathogens (31). KC and MIP-2, also called CXCL1 and CXCL2, respectively, are the members of the CXC chemokine family that are the most critical for neutrophil recruitment to a site of infection (14). In the present study, consistently, the recruitment of macrophages into the peritoneal cavity, at the site of Candida infection, and the induction of MCP-1, KC, and MIP-2 in these cells were diminished in GATA-3-tg mice. Furthermore, administration of IFN-γ to GATA-3-tg mice increased the number of peritoneal macrophages. Thus, overexpression of GATA-3 may reduce macrophage antifungal activity by reducing the production of IFN-γ in response to Candida infection.
TNF-α is another Th1 cytokine produced by a variety of cells. It has been demonstrated that the administration of exogenous TNF-α enhances host resistance against systemic Candida infection, although TNF-α did not demonstrate a direct anticandidal effect in vitro (18). Endogenously produced TNF-α was also shown to have a beneficial effect in the host defense against Candida infection, since treatment of mice with anti-TNF-α antibody resulted in higher colony formation of C. albicans in the kidney and spleen early in the infection (18). The present finding that the expression of TNF-α was significantly lower in the spleens of Candida-susceptible GATA-3-tg mice than in those of WT mice is consistent with these previous findings.
To generate GATA-3-tg mice, we inserted a full-length murine GATA-3 cDNA into a VA CD2 transgene cassette (36). The VA vector has been reported to directly express the inserted cDNA in all single-positive mature T cells of transgenic mice (40). Since GATA-3 is a T-cell-specific transcription factor, transgenic mice are thought to be adequate for use in evaluating T-cell-mediated regulation of Candida infection. In the present study, intracellular cytokine analysis revealed that the suppression of IFN-γ production in GATA-3-tg mice occurred in CD4-positive T cells. These findings suggest that the overexpressed GATA-3 gene actually acts on CD4-positive T cells in the present model.
Coordination between innate immunity and adaptive immunity is critical for host defense against fungal infections (4, 20). A variety of cells, including macrophages, neutrophils, natural killer cells, and natural killer T (NKT) cells, may participate in the nonspecific clearance of fungi as innate immune cells. On the other hand, Th1 cells and IFN-γ are clearly required for the expression of adaptive anticandidal responses (5, 26, 27). In the present analysis of peritoneal lavage fluids, macrophages, neutrophils, and NKT cells were predominant 1 day to 3 days following Candida inoculation (data not shown). On the other hand, CD4-positive T cells were predominant 5 days after Candida inoculation. These results indicate that adaptive immunity becomes predominant 5 days after the inoculation. In the present study, we consistently observed that GATA-3-tg mice had already begun to die at 5 days after Candida inoculation.
In conclusion, we demonstrated that mice overexpressing GATA-3 are highly susceptible to systemic Candida infection by both an intraperitoneal and an intravenous route. Overexpression of GATA-3 reduces IFN-γ production in CXC3-positive Th1 cells in response to Candida infection. IFN-γ insufficiency may impair macrophage anticandidal activity and thereby enhance the susceptibility to systemic Candida infection. We demonstrate here that GATA-3 is an important host factor in the regulation of susceptibility to Candida infection and that the determination of GATA-3 activation, as well as IFN-γ expression, in Th cells might serve as a useful clinical marker for predicting the severity of Candida infection.
This work was supported by a Grant-in-Aid for Scientific Research (C) in Japan from the Society for the Promotion of Science.
Editor: G. S. Deepe, Jr.
Published ahead of print on 15 March 2010.