INFECTION AND IMMUNITY, Feb. 1998, p. 800–806 0019-9567/98/$04.0010 Copyright © 1998, American Society for Microbiology
Vol. 66, No. 2
Protective Role of Gamma Interferon in Experimental Pulmonary Paracoccidioidomycosis ´ , CYNTHIA F. XIDIEH, LUZ E. CANO,† SUELY S. KASHINO, CELINA ARRUDA, DENISE ANDRE ´ IA A. C. VAZ, EVA BURGER, LUCIA M. SINGER-VERMES, CELIDE AND VERA L. G. CALICH* Departamento de Imunologia, Instituto de Cieˆncias Biome´dicas da Universidade de Sa ˜o Paulo, Sa ˜o Paulo, Brazil Received 14 April 1997/Returned for modification 21 May 1997/Accepted 23 October 1997
It has been well documented that IFN-g plays a pivotal role in host resistance against various pathogens through augmentation of the killing activity of macrophages (7, 15, 26, 30). IFN-g-activated macrophages presented an enhanced killing activity against P. brasiliensis conidia and yeast cells (6, 12). Mody et al. (30) demonstrated IFN-g-induced improvement of cryptococcocidal activity of rat alveolar macrophages. In addition, Salkowski and Balish (39) showed enhancement of natural killer (NK) cell activity by IFN-g during cryptococcal infection and impaired clearance of the fungus from the spleens, lungs, and livers of mice treated with anti-IFN-g monoclonal antibody (MAb). The availability of these reagents has facilitated many studies aimed at elucidating IFN-g-mediated immune mechanisms at the molecular level and at defining its in vivo physiologic role. The purpose of this work was to identify type 1 (IFN-g and interleukin-2 [IL-2]) and type 2 (IL-4, IL5, and IL-10) cytokines produced at the site of infection and to verify the effects of anti-IFN-g MAbs as an in vivo treatment in the murine pulmonary model of PCM. We studied the pulmonary infection, extrapulmonary dissemination, specific delayed-type hypersensitivity (DTH) reactions, and specific humoral responses in three groups of animals (untreated, treated with normal immunoglobulin G [IgG], and treated with anti-IFN-g MAbs) of each mouse strain (A/Sn and B10.A) at two periods post-i.t. infection (weeks 4 and 8). We demonstrated a mixed pattern of pulmonary cytokine secretion in both mouse strains, but the levels of IFN-g, IL-4, IL-5, and IL-10 were higher in the lungs of susceptible animals. We also verified that irrespective of the mouse strain, IFN-g plays an important role in resistance to P. brasiliensis infection, through its enhancement of the clearance of fungal cells and of cell-mediated immune responses and its regulatory effects on specific humoral immune responses. Furthermore, the proinflammatory activity of this cytokine appears to be crucial to the induction of circumscribed lesions in the lungs.
Paracoccidioidomycosis (PCM) is a human systemic mycosis caused by the thermally dimorphic fungus Paracoccidioides brasiliensis. Geographically, it is confined to Latin America, with its areas of endemicity extending from Central America to Argentina. It constitutes one of the most prevalent deep mycoses in this region (37). The great majority of the infected subjects develop an asymptomatic pulmonary infection, although some individuals present clinical manifestations which give rise to the polar forms of overt PCM, namely, hyperergic or localized and anergic or disseminated disease (31). Clinical and experimental data indicate that cell-mediated immunity plays a significant role in host defense against P. brasiliensis infection, whereas high levels of specific antibodies and polyclonal activation of B cells are associated with the most severe forms of the disease (2, 13, 34, 40). Using a murine model of intraperitoneally (i.p.) induced PCM, Calich et al. (9) showed that there were significant differences in susceptibility among inbred strains: A/Sn mice were found to be the most resistant, while B10.A animals were the most susceptible to P. brasiliensis infection. More recently, we developed a pulmonary PCM model employing the same inbred mouse strains but using the intratracheal (i.t.) route (11). It was observed that A/Sn mice developed a chronic benign, pulmonary-restricted PCM whereas B10.A mice developed a progressive disseminated disease. The results obtained suggested that resistance to PCM was associated with T-cell, macrophage, and B-cell activities that are known to be mediated by gamma interferon (IFN-g). * Corresponding author. Mailing address: Depto. de Imunologia, Instituto de Cieˆncias Biome´dicas, Universidade de Sa˜o Paulo, Av. Prof. Lineu Prestes 2415, Cep 05508-009, Sa˜o Paulo, SP, Brazil. Phone: 55-11-818 7397. Fax: 55-11-818 7224. E-mail:
[email protected] .usp.br. † Present address: Laborato ´rio de Micologia Experimental, Corporacio ´n para Investigaciones Biolo ´gicas, Medellı´n, Colombia. 800
Downloaded from http://iai.asm.org/ on August 26, 2015 by guest
We have developed a murine model of pulmonary infection by Paracoccidioides brasiliensis in which resistance was associated with immunological activities governed by gamma interferon (IFN-g). To better characterize this model, we measured type 1 and type 2 cytokines in the lungs and investigated the effect of endogenous IFN-g depletion by monoclonal antibodies in the course of infection of susceptible (B10.A) and resistant (A/Sn) mice. At weeks 4 and 8 after infection, lungs from susceptible animals presented levels of IFN-g, interleukin-4 (IL-4), IL-5, and IL-10 higher than those in resistant mice. In both mouse strains, neutralization of endogenous IFN-g induced exacerbation of the pulmonary infection, earlier fungal dissemination to the liver and spleen, impairment of the specific cellular immune response resulting in significantly lower delayed-type hypersensitivity reactions, and increased levels of immunoglobulin G1 (IgG1)- and IgG2b-specific antibodies. Histopathological analysis demonstrated that depletion of IFN-g changes the focal granulomatous lesions found in the lungs of B10.A and A/Sn mice into coalescent granulomata which destroy the pulmonary architecture. These results suggest that irrespective of the mouse strain, IFN-g plays a protective role and that this cytokine is one major mediator of resistance against P. brasiliensis infection in mice.
VOL. 66, 1998
PROTECTIVE ROLE OF IFN-g IN PARACOCCIDIOIDOMYCOSIS
801
TABLE 1. Cytokines present in lung tissue of susceptible (B10.A) and resistant (A/Sn) mice at weeks 4 and 8 after i.t. infection with 106 yeast cells of P. brasiliensisa Wk after infection
IFN-g (U/ml)
IL-2 (U/ml)
IL-4 (pg/ml)
IL-5 (pg/ml)
IL-10 (pg/ml)
B10.A
4 8
33.78b (4.25) 29.14b (3.27)
3.89 (0.85) 2.98 (0.49)
29.26b (10.45) 25.74b (4.71)
183.41b (25.14) 187.74b (21.69)
84.12b (15.47) 86.62b (8.40)
A/Sn
4 8
18.71 (3.31) 22.24 (1.84)
3.20 (0.73) 2.77 (0.65)
9.22 (5.99) 13.78 (4.45)
118.74 (19.99) 152.75 (13.55)
55.25 (10.11) 66.81 (7.06)
Mice
a Lungs from infected mice were collected and disrupted in 3.0 ml of RPMI 1640 medium, and supernatants were obtained for cytokine assays. Capture ELISA was performed as described in Materials and Methods. Numbers indicate the mean (standard deviation) for six mice tested at each time point. b Difference statistically significant; P , 0.05 compared to results obtained from A/Sn mice at the same periods after infection.
MATERIALS AND METHODS
RESULTS Determination of the cytokine profile in the lungs of infected B10.A and A/Sn mice. To examine whether type 1 (IFN-g and IL-2) and type 2 (IL-4, IL-5, and IL-10) cytokines were differentially secreted in the lungs of B10.A and A/Sn mice, we harvested lung cell supernatants at weeks 4 and 8 after P. brasiliensis infection and measured the presence of cytokines by capture ELISA. As can be seen in Table 1, both mouse strains showed a mixed type 1-type 2 cytokine response. However, at both periods assayed, susceptible animals presented
Downloaded from http://iai.asm.org/ on August 26, 2015 by guest
Fungus. P. brasiliensis 18, an isolate which is highly virulent (25), was used throughout this study. To ensure the maintenance of its virulence, the isolate was used after three serial animal passages (23). P. brasiliensis 18 yeast cells were then maintained by weekly subcultivation in semisolid Fava Netto’s culture medium (16) at 35°C and used at the 7th day in culture. The fungal cells were washed in phosphate-buffered saline (PBS; pH 7.2) and counted in a hemocytometer, and the concentration was adjusted to 20 3 106 fungal cells ml21. Viability of fungal suspensions, determined by Janus green B vital dye (Merck, Darmstadt, Germany) (5), was always higher than 80%. Animals. Unless otherwise stated, groups of 8 to 10 male mice (9 to 11 weeks old) from the susceptible (B10.A) and resistant (A/Sn) strains were used for each period of infection. BALB/c mice were used for the expansion of the anti-IFN-g hybridoma. All animals were bred at the University of Sa˜o Paulo animal facilities and provided with acidified water and sterilized food and bedding. Surgical i.t. inoculation. Mice were anesthetized by the i.p. route with a 0.4% solution of 2-(2,6-xylidine) 5,6-dihydro-4 H-1,3-thiasine hydrochloride (10 ml/kg of body weight; Rompun; Bayer of Brazil). After 10 min, the animals were injected i.p. with a 2.5% solution of chloral hydrate (10 ml/kg; Reagen, Quimibra´s, Indu ´strias Quı´micas, Rio de Janeiro, Brazil). Animals were restrained on a small board and infected with 106 fungal cells (P. brasiliensis 18) by surgical i.t. inoculation. For this purpose, a small incision was made through the skin over the trachea, and the underlying tissue was separated; a 30-gauge needle attached to a tuberculin syringe was inserted into and parallel with the trachea, and 50 ml of inoculum was dispensed into the lungs. The skin was then sutured, and mice were allowed to recover under a heat lamp. Treatment of mice with anti-IFN-g MAb. The anti-murine IFN-g (XMG1.2) hybridoma was kindly provided by Ises A. Abrahamsohn (Departamento Imunologia, ICB, University of Sa˜o Paulo, Sa˜o Paulo, Brazil) with permission from the original source (DNAX Research Institute of Molecular and Cellular Biology Inc., Palo Alto, Calif.). The monoclonal rat IgG1 antibody XMG1.2 was grown i.p. in pristane (Sigma Chemical Co., St. Louis, Mo.)-primed, sublethally irradiated (550 rads) BALB/c mice. MAb XMG1.2 was purified from ascites fluid as described by McKinney and Parkinson (29) and assayed for purity by sodium dodecyl sulfate-polyacrylamide gel electrophoresis. The biological activity of MAb XMG1.2 was tested by its in vitro inhibitory activity on nitric oxide production by concanavalin A-activated macrophages. Groups of B10.A and A/Sn mice were given 1 mg of MAb XMG1.2 or rat IgG (controls) in 1 ml of PBS (pH 7.2) by the i.p. route 1 day before i.t. inoculation with P. brasiliensis (day 21) and 0.5 mg of XMG1.2 or rat IgG weekly thereafter (days 7, 14, 21, etc.). Cytokine analysis. B10.A and A/Sn mice were i.t. infected with P. brasiliensis and at 4 and 8 weeks after infection (six mice per group) their right lungs were aseptically removed and disrupted in 3.0 ml of RPMI 1640 medium (Gibco BRL). Supernatants were separated from cell debris by centrifugation at 2,000 3 g for 15 min, passed through 0.22-mm-pore-size filters (Millipore), and stored at 270°C. The levels of IL-2, IL-4, IL-5, IL-10, and IFN-g were measured by capture enzyme-linked immunosorbent assay (ELISA) using antibody pairs purchased from PharMingen (San Diego, Calif.). The ELISA procedure was performed according to the manufacturer’s protocol, with slight modifications. Briefly, ELISA 96-well microtiter plates (Corning Laboratory Sciences Co., New York, N.Y.) were coated with purified MAb, and nonspecific binding sites were blocked with 2% gelatin (Difco Laboratories, Detroit, Mich.) in PBS (pH 7.0) for 3 h at room temperature. The supernatants were added in triplicates, and biotinylated MAb was used as the secondary antibody. Finally, wells were incubated with a streptavidin-biotin-horseradish peroxidase complex (Vectastain ABC kit; Vector Laboratories, Inc., Burlingame, Calif.) for 30 min at room temperature. The final reaction was developed with 10 mg of o-phenylenediamine (Merck) in 10 ml of 0.03 M citric acid–0.04 M sodium phosphate solution, supplemented with 5 ml of 30% hydrogen peroxide (Merck), for 40 min at room temperature. The enzymatic reaction was stopped with 4 N sulfuric acid, and A490 was read on a Dynatech MR 5000 Microplate Reader (Dynatech Laboratories, Inc., Chantilly, Va.). The concentrations of cytokines were determined with reference to a standard curve for serial twofold dilutions of murine recombinant cytokine
(PharMingen). The lower limits of detection of the recombinant standard curves were 0.78 U/ml for IL-2, 7.8 pg/ml for IL-4, 7.8 pg/ml for IL-10, 7.8 pg/ml for IL-5, and 0.2 U/ml for IFN-g. Assay for organ CFU. The number of viable microorganisms in different organs in infected mice was determined by CFU counts. At two times postinfection (4 and 8 weeks), 8 to 10 infected B10.A and A/Sn mice of three groups (untreated or treated with normal IgG or anti-IFN-g) were sacrificed, and their lungs, livers, and spleens were aseptically removed, weighed, and homogenized in 5 ml of sterile PBS by means of a tissue grinder. The cellular suspensions were washed three times in PBS, and the final pellets were resuspended in 1 ml of PBS. Aliquots (100 ml) of each homogenate were plated on brain heart infusion agar (Difco) which contained 4% (vol/vol) normal horse serum (Instituto Butantan, Sa˜o Paulo, Brazil) and 5% P. brasiliensis 192 culture filtrate; the latter constituted the source of growth-promoting factors (41). When necessary, dilutions were made in sterile PBS. Plates were incubated at 35°C, and colonies were counted daily until no increase in counts was observed. The numbers (log10) of viable P. brasiliensis per gram of tissue are expressed as the means 6 standard errors. DTH assay. The DTH reactions of the same animals used for the CFU determinations were evaluated by the footpad test as described by Fazioli et al. (17). Briefly, 24 h before the animals were sacrificed, the footpad thicknesses were measured with a dial caliper (precision, 0.01 mm; Mitutoyo Corporation, Tokyo, Japan); immediately thereafter, the mice were challenged by injection of 25 ml of Fava Netto’s antigen (16) derived from P. brasiliensis 18 (protein concentration, 310 mg ml21). The footpad thickness was measured 24 h later, and swelling was expressed in millimeters. Noninfected control mice were also submitted to the footpad test. Specific antibody levels. Specific antibody levels (total Ig, IgM, IgG1, IgG2a, IgG3, and IgG2b) were measured by a previously described ELISA (11), using a cell-free antigen (10) prepared from a pool of different P. brasiliensis isolates (339, 265, and 18). The average of the optical densities obtained by using sera from control mice (PBS inoculated), diluted 1:20, was considered the cutoff for each isotype. Optical densities for each dilution of experimental sera were compared to the control values. The titer for each sample was expressed as the reciprocal of the highest dilution which presented absorbance higher than the cutoff. Histopathologic analysis. Groups of four to five untreated, IgG-treated, and IFN-g-depleted P. brasiliensis-infected A/Sn and B10.A mice were killed at week 4 after infection. Lungs were collected, fixed in 10% formalin, and embedded in paraffin. Five-micrometer sections were stained by the hematoxylin-eosin (HE) method. Pathologic changes were analyzed based on the number, size, morphology, and cell composition of granulomatous lesions, number of fungi, and intensity of the inflammatory infiltrates. Statistical analysis. The number of CFU/gram of tissue, DTH reactions, antibody titers, and pulmonary cytokines were analyzed by a nonparametric method (Kruskal-Wallis), and then by the Dunn multiple comparisons method; the two-way analysis of variance was used to verify differences in cytokine secretion (45). A P value of ,0.05 was considered significant.
802
CANO ET AL.
significantly higher levels of IFN-g, IL-4, IL-5, and IL-10 than resistant mice. Effect of in vivo neutralization of IFN-g on the course of P. brasiliensis infection. The evolution of the disease in untreated and treated B10.A and A/Sn mice was monitored by CFU counts in the lungs, spleen, and liver at weeks 4 and 8 after infection (Fig. 1). Rat IgG treatment had no effect on the degree of infection of both B10.A and A/Sn mice compared with the infected untreated mice. The in vivo depletion of IFN-g for 4 weeks induced an earlier dissemination of fungal cells to the liver and spleen in both mouse strains and exacerbation of the pulmonary infection in B10.A animals. When depletion of endogenous IFN-g was accomplished by treatment for 8 weeks, progression of both pulmonary and extrapulmonary infection with elevated CFU numbers was observed, and the numbers of viable yeast cells were significantly higher (P , 0.05) than at week 4 postinfection. Effect of anti-IFN-g MAb on the DTH response. To assess the possible effect of anti-IFN-g treatment on the development of DTH, mice challenged i.t. with 106 P. brasiliensis yeast cells (untreated or treated with control normal IgG or XMG1.2 antibodies) were injected intrafootpad with 25 ml of a soluble P. brasiliensis antigen 4 and 8 weeks after infection. The increase in footpad swelling was measured 24 h later (Fig. 2). Untreated and normal IgG-treated A/Sn mice demonstrated at both postinfection periods a marked increase in the footpad thickness in the limb that had been injected with antigen. The mean values (millimeters) of footpad swelling of untreated A/Sn mice were 0.201 6 0.014 and 0.155 6 0.018 at weeks 4 and 8 after infection, respectively. Similar reactions were ob-
served in normal IgG-treated A/Sn mice (0.182 6 0.009 and 0.192 6 0.018 at weeks 4 and 8). On the other hand, the footpad swelling was significantly reduced in infected A/Sn mice treated with XMG1.2 MAb at both postinfection periods (0.031 6 0.003 and 0.031 6 0.008, respectively). The B10.A mice treated with anti-IFN-g MAb for 4 weeks had very low DTH values (0.017 6 0.004) comparable only to those of the control, noninfected animals (0.021 6 0.003) and significantly lower than those for untreated B10.A-infected mice (0.079 6 0.013). The same pattern of DTH responses was observed when the MAb treatment was prolonged for 8 weeks (Fig. 2). Effect of anti-IFN-g treatment on serum levels of P. brasiliensis-specific antibodies. We next measured the levels and determined the isotype profiles of specific antibodies in the sera of B10.A and A/Sn mice infected with P. brasiliensis and treated with control IgG or XMG1.2 antibodies. At 4 and 8 weeks after infection, sera were collected and assayed for their contents of total Ig, IgM, IgG1, IgG2a, IgG2b, and IgG3. Figure 3 shows that in both mouse strains, treatment with normal rat IgG for 4 weeks induced a tendency toward increased levels of antibodies. However, significantly increased levels of total Ig and IgG1 were observed in B10.A mice treated with normal IgG or anti-IFN-g antibodies (titers of about 1:640). Additionally, the in vivo depletion of IFN-g produced enhancement of total Ig (1:2,228), IgG1 (1:1,470), IgG2a (1:2,560), and IgG2b (1:1,689) levels in the A/Sn mice. When treatment was given for 8 weeks, a few alterations in the antibody levels were also noticed (Fig. 4). There were significant increments in IgG1 and IgG2b isotype levels in both mouse strains. Effect of IFN-g depletion on pulmonary lesions. The increased fungal burden and extrapulmonary dissemination observed in IFN-g-depleted mice suggested that this cytokine could influence the morphology of pulmonary lesions. To investigate this possibility, we analyzed lung sections from control and anti-IFN-g-treated mice at week 4 postinfection. Pul-
FIG. 2. Effect of anti-IFN-g treatment on DTH responses of resistant (A/Sn) and susceptible (B10.A) mice infected i.t. with 106 P. brasiliensis yeast cells. Animals were untreated (h), treated with normal rat IgG (o), or treated with rat anti-mouse IFN-g MAb (■). At weeks 4 and 8 after infection, mice (8 to 10 per group) were injected intrafootpad with a soluble yeast antigen 24 h before measurement of the footpad response. The bars depict means 6 standard errors of footpad swelling. The horizontal lines denote means 6 2 standard deviations (confidence interval, 95%) of noninfected A/Sn and B10.A mice submitted to the footpad test (n 5 39). p, significantly different (P , 0.05) from untreated group; 1, significantly different (P , 0.05) from normal IgG group.
Downloaded from http://iai.asm.org/ on August 26, 2015 by guest
FIG. 1. Adverse effects of neutralizing endogenous IFN-g in genetically resistant (A/Sn) and susceptible (B10.A) mice to P. brasiliensis infection. Shown is recovery of CFU from lungs (o) liver (u), and spleen (s) of i.t.-infected mice either untreated, treated with normal rat IgG, or treated with rat anti-mouse IFN-g MAb. The bars depict means 6 standard errors of log10 CFU obtained from groups of 8 to 10 mice at weeks 4 and 8 after infection. p, significantly different (P , 0.05) from untreated group; 1, significantly different (P , 0.05) from normal IgG group.
INFECT. IMMUN.
VOL. 66, 1998
PROTECTIVE ROLE OF IFN-g IN PARACOCCIDIOIDOMYCOSIS
of IFN-g in the site of infection do not determine the resistance phenotype to P. brasiliensis infection, but the amount of IL-4, IL-5, and IL-10 or another inhibitory cytokine such as transforming growth factor b could be linked to susceptibility. Our results do not demonstrate a polarized Th1/Th2 activation. On the contrary, IFN-g and Th2 cytokines were equally produced with different net results in resistant and susceptible mice. Since A/Sn mice infected i.t. with P. brasiliensis present a restricted pulmonary infection, low mortality rates, persistent DTH reactions, bronchoalveolar macrophage activation, and predominance of IgG2a and IgG3 antibodies (11), we can infer that the lower levels of type 2 cytokines at the site of infection result in a prevalent effect of IFN-g. On the other hand, the higher levels of pulmonary IFN-g in B10.A mice appear to confer local protection in controlling fungal growth, but the concomitant presence of high levels of type 2 cytokines appears to be sufficient to downmodulate macrophage activation and DTH responses, resulting in dissemination of yeast cells to extrapulmonary sites. In agreement with the disease-exacerbating effect of Th2 cytokines, previous studies in our laboratory demonstrated a protective effect of murine recombinant IL-12 on the i.t. infection of B10.A mice as revealed by the absence of fungal dissemination to liver and spleen and significant decreases in the levels of pulmonary IL-4, IL-5, and IL-10 (3). The high levels of pulmonary IFN-g were not accompanied by increased amounts of IL-2, another Th1 cytokine. As CD41 and CD81 T cells and NK cells are able to produce IFN-g (4, 20, 33), further studies are needed to elucidate the phenotype of the IFN-g-secreting cells. Furthermore, characterization of
monary lesions of untreated susceptible and resistant mice showed similar patterns: isolated or aggregated granulomatous lesions of various sizes (Fig. 5A), composed of macrophages often transformed into epithelioid cells and fungi, surrounded by a lymphoid sheath (Fig. 5B). Neutrophils were frequent and usually oriented around fungal cells, forming rosette-like structures. Some lesions presented a central necrotic area rich in P. brasiliensis yeast cells. The fungi in the granulomas either showed their typical morphology with multiple budding or were degenerated. Despite the similarity in morphology, lesions from resistant animals presented more necrotic areas and tended to be more circumscribed. Treatment with normal rat IgG had no significant effect on the morphology of the lesions. In contrast, neutralization of endogenous IFN-g led to striking differences: in both mouse strains, the pulmonary parenchyma was almost totally substituted by huge granulomatous lesions (Fig. 5C), containing innumerable yeast cells (Fig. 5D). The most relevant difference between B10.A and A/Sn strains was the increased amount of eosinophils in the former and the higher number of necrotic lesions in the latter. DISCUSSION This study demonstrates for the first time that irrespective of the genetic background of mouse strain, IFN-g exerts a protective role against pulmonary P. brasiliensis infection. At weeks 4 and 8 after infection, B10.A and A/Sn mice showed a mixed response where IFN-g and type 2 cytokines appeared concomitantly. However, the levels of IFN-g, IL-4, IL-5, and IL-10 were higher in the lungs of B10.A mice. Thus, the levels
FIG. 4. Levels of P. brasiliensis-specific antibodies in sera of resistant (A/Sn) and susceptible (B10.A) mice at 8 weeks after i.t. infection with 106 yeast cells. Infected mice were untreated (h), treated with normal rat IgG (o), or treated with rat anti-mouse IFN-g MAb (■). Sera were assayed for total Ig, IgM, IgG1, IgG2a, IgG2b, and IgG3 by using an isotype-specific ELISA as detailed in Materials and Methods. The bars depict geometric means 6 standard errors of serum titers (8 to 10 animals per group). p, significantly different (P , 0.05) from untreated group.
Downloaded from http://iai.asm.org/ on August 26, 2015 by guest
FIG. 3. Levels of P. brasiliensis-specific antibodies in sera of resistant (A/Sn) and susceptible (B10.A) mice at 4 weeks after i.t. infection with 106 yeast cells. Infected mice were untreated (h), treated with normal rat IgG (o), or treated with rat anti-mouse IFN-g MAb (■). Sera were assayed for total Ig, IgM, IgG1, IgG2a, IgG2b, and IgG3 by using an isotype-specific ELISA as detailed in Materials and Methods. The bars depict geometric means 6 standard errors of serum titers (8 to 10 animals per group). p, significantly different (P , 0.05) from untreated group.
803
804
CANO ET AL.
INFECT. IMMUN.
Th1 and Th2 cytokines at other lymphoid tissues is necessary to extend our present findings. Several experimental findings obtained in this study show that IFN-g plays a critical role in host resistance against P. brasiliensis. In both mouse strains, treatment of i.t.-infected mice with anti-IFN-g MAb prevented the clearance of yeast cells from the lungs and promoted their dissemination to the liver and to the spleen. Thus, IFN-g depletion increased more than 1,000-fold the pulmonary fungal burden of B10.A mice and 10-fold the already high number of fungal cells in the lungs of A/Sn mice, which became indistinguishable from those in susceptible animals. Furthermore, in vivo IFN-g neutralization dramatically altered the histological pattern of pulmonary lesions of B10.A and A/Sn mice, which lose their ability to circumscribe fungal cells at the site of infection. The compact granulomata composed of packed epithelioid cells and restricted number of fungi were replaced by a fungus-rich diffuse inflammation obliterating the normal pulmonary structure. This picture is more severe than those observed in T-celldeficient animals (BALB/c nu/nu) or mice with impaired macrophage activity (8, 24) after infection with the same P. brasiliensis isolate. Previous investigations have shown that depletion of endogenous IFN-g by in vivo administration of anti-IFN-g MAb
exacerbates bacterial (7, 35), protozoan (44), and fungal (1, 27, 38) infections. Protection against these diseases correlates directly with expression of cell-mediated immune responses and parasite killing by activated macrophages. One mechanism by which IFN-g might mediate resistance to P. brasiliensis is by activating macrophages to inhibit or kill the fungus. Human macrophages cultivated for 3 to 7 days and treated for 3 days with cytokines or IFN-g could be activated restraining P. brasiliensis multiplication in long-term cultures (32). Moreover, this modulation by IFN-g or cytokines could be inhibited by the presence of antibody to IFN-g. Accordingly, it was demonstrated that treatment of murine pulmonary or peritoneal macrophages with IFN-g potentiated their anti-Paracoccidioides activity (6, 12). In a previous work, we also demonstrated a clear association between macrophage activation, as measured by hydrogen peroxide release, and resistance to fungal challenge (11). Altogether, these results suggest that IFN-g plays a role in macrophage activation and inhibition of P. brasiliensis multiplication. The murine inflammatory response that results in DTH is transferred by Th1 cells when injected with the appropriate antigen (14). Issekuts et al. (22) demonstrated that IFN-g is a potent stimulator of lymphocyte migration into the skin and plays a major role in lymphocyte recruitment into DTH reac-
Downloaded from http://iai.asm.org/ on August 26, 2015 by guest
FIG. 5. Representative histopathology of lesions caused by P. brasiliensis in lungs of B10.A mice showing isolated granulomatous foci scattered in the parenchyma (A [HE; magnification, 340]), one typical granuloma (B [HE; magnification, 3100]); extensive granulomatous lesions replacing the normal pulmonary morphology (C [HE; magnification, 340]); and presence of innumerable yeast cells D [HE, magnification, 3100]). (A and B) P. brasiliensis-infected mice; (C and D) Anti-IFN-g-treated and P. brasiliensis-infected mice.
VOL. 66, 1998
PROTECTIVE ROLE OF IFN-g IN PARACOCCIDIOIDOMYCOSIS
ACKNOWLEDGMENTS We are grateful to A. Restrepo for helpful suggestions and thank S. M. D. Moraes, T. A. Costa, and B. P. Albe for technical assistance. S. S. Kahino and C. Arruda are recipients of fellowships from Fundac¸˜ao de Amparo a` Pesquisa do Estado de Sa˜o Paulo (FAPESP). L. E. Cano was supported by Ph.D. fellowship 92/0962-9 from the Fundac¸˜ao de Amparo `a Pesquisa do Estado de Sa˜o Paulo. This work was supported by grants from FAPESP and Conselho Nacional de Pesquisas. REFERENCES 1. Aguirre, K., E. A. Havell, G. W. Gibson, and L. L. Johnson. 1995. Role of tumor necrosis factor and gamma interferon in acquired resistance to Cryptococcus neoformans in the central nervous system of mice. Infect. Immun. 63:1725–1731.
2. Arango, M., and L. Yarzabal. 1982. T-cell dysfunction and hyperimmunoglobulinemia E in paracoccidioidomycosis. Mycopathologia 79:115–123. 3. Arruda, C., R. A. Fazioli, S. S. Kashino, F. R. F. Nascimento, K. C. Rocha, A. T. Costa, M. Russo, and V. L. G. Calich. 1997. In vivo treatment with rIL-12 protects susceptible mice from Paracoccidioides brasiliensis pulmonary infection, abstr. 283, p. 144. In Abstracts of the 13th Congress of the International Society of Human and Animal Mycology. 4. Bancroft, G. L., R. D. Schreiber, and E. R. Unanue. 1991. Natural immunity: a T cell independent pathway of macrophage activation, defined in scid mice. Immunol. Rev. 124:5–24. 5. Berliner, M. D., and M. E. Reca. 1966. Vital staining of Histoplasma capsulatum with Janus Green B. Sabouraudia 5:26–29. 6. Brummer, E., L. H. Hanson, A. Restrepo, and D. A. Stevens. 1988. In vivo and in vitro activation of pulmonary macrophages by IFN-g for enhanced killing of P. brasiliensis or B. dermatitides. J. Immunol. 140:2786–2787. 7. Buchmeier, N. A., and R. D. Schreiber. 1985. Requirement of endogenous interferon-gamma production for resolution of Listeria monocytogenes infection. Proc. Natl. Acad. Sci. USA 82:7404–7408. 8. Burger, E., M. Miyaji, A. Sano, V. L. G. Calich, K. Nishimura, and H. L. Lenzi. 1996. Histopathology of paracoccidioidomycotic infection in athymic and euthymic mice: a sequential study. Am. J. Trop. Med. Hyg. 55:235–242. 9. Calich, V. L. G., L. M. Singer-Vermes, A. M. Siqueira, and E. Burger. 1985. Susceptibility and resistance of inbred mice to Paracoccidioides brasiliensis. Br. J. Exp. Pathol. 66:585–594. 10. Camargo, Z. P., C. P. Taborda, E. G. Rodrigues, and L. R. Travassos. 1991. The use of cell-free antigens of Paracoccidioides brasiliensis in serological tests. J. Med. Vet. Mycol. 29:31–38. 11. Cano, L. E., L. M. Singer-Vermes, C. A. C. Vaz, M. Russo, and V. L. G. Calich. 1995. Pulmonary paracoccidioidomycosis in resistant and susceptible mice: relationship among progression of infection, bronchoalveolar cell activation, cellular immune response, and specific isotype patterns. Infect. Immun. 63:1777–1783. 12. Cano, L. E., R. Arango, M. E. Salazar, E. Brummer, D. A. Stevens, and A. Restrepo. 1992. Killing of P. brasiliensis conidia by pulmonary macrophages and the effects of cytokines. J. Med. Vet. Mycol. 30:161–168. 13. Castan ˜ eda, E., E. Brummer, D. Pappagianis, and D. A. Stevens. 1988. Impairment of cellular but not humoral immune response in chronic pulmonary and disseminated paracoccidioidomycosis in mice. Infect. Immun. 56:1771– 1777. 14. Cher, D. J., and T. R. Mosmann. 1987. Two types of murine helper T-cell clones. II. Delayed-hypersensitivity is mediated by TH1 clones. J. Immunol. 138:3688–3894. 15. Denis, M. 1991. Interferon-gamma treated murine macrophages inhibit growth of tubercle bacilli via the generation of reactive nitrogen intermediates. Cell. Immunol. 132:150–157. 16. Fava Netto, C., V. S. Vegas, I. M. Scianname´a, and D. B. Guarnieri. 1969. Antı´geno polissacarı´dico do Paracoccidioides brasiliensis. Estudo do tempo de cultivo do P. brasiliensis necessa´rio ao preparo do antı´geno. Rev. Inst. Med. Trop. Sa˜o Paulo 11:177–181. 17. Fazioli, R. A., L. M. Singer-Vermes, S. S. Kashino, E. Burger, M. F. de Franco, M. Moscardi-Bacchi, and V. L. G. Calich. 1994. Delayed-type hypersensitivity response in an isogenic murine model of paracoccidioidomycosis. Mycopathologia 126:137–146. 18. Finkelman, F. D., A. Svetic, I. Gresser, C. Snapper, J. Holmes, P. P. Trotta, I. M. Katona, and W. C. Gause. 1991. Regulation by IFN-a of immunoglobulin isotype selection and lymphokine production in mice. J. Exp. Med. 174:1179–1188. 19. Finkelman, F. D., I. M. Katona, T. R. Mossman, and R. L. Coffman. 1988. IFN-g regulates the isotypes of Ig secreted during in vivo humoral immune responses. J. Immunol. 140:1022–1027. 20. Fong, T. A., and T. R. Mossman. 1990. Allo-reactive murine CD81 T cell clones secrete the Th1 pattern of cytokines. J. Immunol. 144:1744–1752. 21. Isakson, P. C., E. Pure, E. S. Vitetta, and P. H. Krammer. 1982. T cell derived B-cell differentiation factor(s). Effect on the isotype switch of murine cells. J. Exp. Med. 155:734–748. 22. Issekuts, T. B., J. M. Stoltz, and P. Van der Meide. 1988. Lymphocyte recruitment in delayed-type hypersensitivity. The role of IFN-gamma. J. Immunol. 140:2989–2993. 23. Kashino, S. S., L. M. Singer-Vermes, V. L. G. Calich, and E. Burger. 1990. Alterations in the pathogenicity of one Paracoccidioides brasiliensis isolate do not correlate with its in vitro growth. Mycopathologia 111:173–180. 24. Kashino, S. S., R. A. Fazioli, M. Moscardi-Bachi, M. Franco, L. M. SingerVermes, E. Burger, and V. L. G. Calich. 1995. Effect of macrophage blockade on the resistance of inbred mice to Paracoccidioides brasiliensis infection. Mycopathologia 130:131–140. 25. Kashino, S. S., V. L. G. Calich, E. Burger, and L. M. Singer-Vermes. 1985. In vivo and in vitro characteristics of six Paracoccidioides brasiliensis strains. Mycopathologia 92:173–178. 26. Lucchiari, M. A., M. Modolell, K. Eichmann, and C. A. Pereira. 1992. In vivo depletion of interferon-gamma leads to susceptibility of A/J mice to mouse hepatitis virus 3 infection. Immunobiology 185:475–482. 27. Magee, D. M., and R. A. Cox. 1995. Roles of gamma interferon and inter-
Downloaded from http://iai.asm.org/ on August 26, 2015 by guest
tions. In addition, IFN-g has been shown to up-regulate intercellular adhesion molecules (ICAM-1) on endothelial cells promoting lymphocyte-endothelial cell binding (36). In the present investigation, the administration of anti-IFN-g antibodies induced in both mouse strains a depressed cellular immune response as revealed by the diminished DTH reactivity. The observed DTH anergy could be ascribed either to a deficient priming of naive T cells to Th1 responses or to an impaired lymphocyte migration and homing at the site of antigen inoculation. Independent of the mechanism, IFN-g depletion results in decreased systemic inflammatory activity and a more severe pathology. When we analyzed humoral immune responses, we observed that both treatments used (normal IgG and anti-IFN-g) stimulated the production of higher antibody levels. A recent publication (43) reported that high concentrations of normal IgG induce the activation of CD41 T cells and B cells. This fact could explain, at least partially, the increments in the antibody levels observed in this study. The depletion of IFN-g caused in susceptible and resistant mice higher production of IgG1 and IgG2b antibodies, indicating an enhanced production or activity of IL-4 and transforming growth factor b, which positively control B-cell switches to IgG1 and IgG2b isotypes, respectively (21, 28). The detected increase of IgG2a antibodies in the serum of anti-IFN-g-treated resistant animals at week 4 after infection appears to be paradoxical since it occurred in an environment of neutralization of IFN-g, a cytokine which is known to induce IgG2a production (19, 42). However, Snapper and Paul (42) described the potentiating effect of low levels of IFN-g on IgG2a secretion. Therefore, in the present work a partial depletion of this cytokine might explain the higher levels of IgG2a-specific antibodies. The increase of this isotype could also be explained by the action of other mediators such as IFN-a, which was previously shown to stimulate a B-cell switch to IgG2a production (18). In the disseminated form of human PCM, high levels of specific antibodies and depressed cellular immunity are associated with histopathological features showing loose granulomatous inflammation containing large numbers of fungal cells (31). Our present work, using the natural route of infection, showed that depletion of IFN-g induces in resistant and susceptible animals a pathology which mimics the severest forms of human PCM. It is apparent from the pulmonary model of PCM that the presence of IFN-g is essential for protective immunity and optimal elimination of P. brasiliensis. In the human pathology, it may be possible to enhance protective immune responses by passive administration of recombinant IFN-g or other immunotherapeutic procedures which downmodulate the synthesis of Th2 cytokines. However, further studies on the effects of immunoregulatory cytokines are needed to better understand the resistance mechanisms to pulmonary PCM.
805
806
28.
29. 30. 31. 32. 33.
35.
leukin-4 in genetically determined resistance to Coccidioides immitis. Infect. Immun. 63:3415–3419. McIntire, T. M., D. R. Klinman, P. Rothman, M. Lugo, J. R. Dash, J. J. Mond, and C. M. Snapper. 1993. Transforming growth factor b1 selectively stimulates immunoglobulin G2b by lipopolysaccharide-activated murine B cells. J. Exp. Med. 177:1031–1037. McKinney, M. M., and A. Parkinson. 1987. A simple, non-chromatographic procedure to purify immunoglobulins from serum and ascites fluid. J. Immunol. Methods 96:271–278. Mody, C. H., C. L. Tyler, R. G. Sitrin, C. Jackson, and G. B. Toews. 1991. Interferon-g activates rat alveolar macrophages for anti-cryptococcal activity. Am. J. Respir. Cell. Mol. Biol. 5:19–26. Montenegro, M. R., and M. Franco. 1994. Pathology, p. 131–147. In M. Franco, C. S. Lacaz, A. Restrepo-Moreno, and G. Del Negro (ed.), Paracoccidioidomycosis. CRC Press, Inc., Boca Raton, Fla. Moscardi-Bacchi, M., E. Brummer, and D. A. Stevens. 1994. Support of Paracoccidioides brasiliensis multiplication by human monocytes or macrophages: inhibition by activated phagocytes. J. Med. Microbiol. 40:159–164. Mossman, T. R., H. Cherwinski, M. W. Bond, M. A. Giedlin, and R. L. Coffman. 1986. Two types of murine helper T clones. I. Definition according to profiles of lymphokine activities and secreted proteins. J. Immunol. 136: 2348–2357. Mota, N. G. S., M. T. Rezkallah-Iwasso, M. T. S. Perac¸oli, R. C. Audi, R. P. Mendes, J. Marcondes, S. A. Marques, N. L. Dillon, and M. F. Franco. 1985. Correlation between cell-mediated immunity and clinical forms of paracoccidioidomycosis. Trans. R. Soc. Trop. Med. Hyg. 79:765–772. Nakane, A., S. Nishikawa, S. Sasaki, T. Miura, M. Asano, M. Kohanawa, K. Ishiwata, and T. Minagawa. 1996. Endogenous interleukin-4, but not interleukin-10, is involved in suppression of host resistance against Listeria monocytogenes infection in gamma interferon-depleted mice. Infect. Immun. 64: 1252–1258.
Editor: S. H. E. Kaufmann
INFECT. IMMUN. 36. Pober, J. S., M. A. Gimbrone, Jr., L. A. LaPierre, D. L. Mendrick, W. Fiers, R. Rothlein, and T. A. Springer. 1986. Overlapping patterns of activation of human endothelial cells by interleukin-1, tumor necrosis factor and immune interferon. J. Immunol. 137:1893–1899. 37. Restrepo, A. 1985. The ecology of P. brasiliensis: a puzzle still unsolved. J. Med. Vet. Mycol. 23:323–334. 38. Romani, L., E. Cenci, A. Menacci, R. Spaccapelo, U. Grohmann, P. Puccetti, and F. Bistoni. 1992. Gamma interferon modifies CD41 subset expression in murine candidiasis. Infect. Immun. 60:4950–4957. 39. Salkowski, C. A., and E. Balish. 1991. A monoclonal antibody to gamma interferon blocks augmentation of natural killer cell activity induced during systemic cryptococcosis. Infect. Immun. 59:486–493. 40. Singer-Vermes, L. M., C. B. Caldeira, E. Burger, and V. L. G. Calich. 1993. Experimental murine paracoccidioidomycosis: relationship among dissemination of the infection, humoral and cellular immune responses. Clin. Exp. Immunol. 94:75–79. 41. Singer-Vermes, L. M., M. C. Ciavaglia, S. S. Kashino, E. Burger, and V. L. G. Calich. 1992. The source of the growth-promoting factor(s) affects the plating efficiency of Paracoccidioides brasiliensis. J. Med. Vet. Mycol. 30:261–264. 42. Snapper, C. M., and W. Paul. 1987. Interferon-g and B cell stimulatory factor-1 reciprocally regulate Ig isotype production. Science 236:944–947. 43. Sundblad, A., M. A. R. Marcos, E. Malanchere, A. Castro, M. Haury, F. Huetz, A. Nobrega, A. Freitas, and A. Coutinho. 1994. Observation on the mode of actions of normal immunoglobulin at high doses. Immunol. Rev. 139:125–158. 44. Suzuki, Y., and J. S. Remington. 1990. The effect of anti-IFN-g antibody on the protective effect of Lyt-21 immune T cells against toxoplasmosis in mice. J. Immunol. 144:1954–1956. 45. Zar, Y. H. 1984. Biostatistical analysis. Prentice-Hall Inc., Englewood Cliffs, N.J.
Downloaded from http://iai.asm.org/ on August 26, 2015 by guest
34.
CANO ET AL.
