Study of pulmonary experimental paracoccidioidomycosis by analysis of bronchoalveolar lavage cells: resistant vs. susceptible mice

Mycopathologia 141: 79–91, 1998. © 1998 Kluwer Academic Publishers. Printed in the Netherlands.

79

Study of pulmonary experimental paracoccidioidomycosis by analysis of bronchoalveolar lavage cells: Resistant vs. susceptible mice F´atima Vilani-Moreno, Denise Fecchio, Maria Cristina Iwama de Mattos, Maura MoscardiBacchi, J´ulio Defaveri & Marcello Franco Departamento de Patologia, Faculdade de Medicina, Universidade Paulista J´ulio de Mesquita Fliho, UNESP, Botucatu, São Paulo, Brasil Received 3 March 1997; accepted in final form 29 April 1998

Abstract Adult Swiss (susceptible) and BALB/c (non-susceptible) mice were inoculated by the intravenous route with 1 × 106 yeast cells of Paracoccidioides brasiliensis, strain 18. Immunologic parameters, histopathology and features of the bronchoalveolar lavage (BAL) were evaluated at week 2, 4, 8 and 16 post-infection. The pulmonary infection was progressive in Swiss mice and regressive in BALB/c mice. The numbers of total cells, lymphocytes and polymorphonuclear neutrophils increased in BAL, as well as the percentages of giant cells, and CD4 and CD8 positive cells. The ultrastructural study of BAL cells revealed a predominance of macrophages and a frequency of 13.2% of type II pneumocytes. As the infection progressed, the number of fungal cells and spreading macrophages, as well as the stimulated release of H2 O2 by macrophages, increased. The animals exhibited an exacerbation of the humoral immune response and a depression of cellular immunity during the infection. There was a good correlation between the intensity and the pattern of the pulmonary histopathology and the cellular findings in the BAL. The present model reproduces some anatomoclinical patterns of the human disease and shows that BAL may be a useful tool in monitoring the pulmonary infection caused by P. brasiliensis. Key words: bronchoalveolar lavage, paracoccidioidomycosis

Introduction Paracoccidioides brasiliensis (P. brasiliensis) is the causative agent of paracoccidioidomycosis (PCM), a deep mycosis that frequently involves the lungs, mononuclear-phagocytic system and mucocutaneous areas. There is evidence that the respiratory route is the major portal of entry of the fungus, by inhalation of conidia or fragments of the mycelial form [1, 2]. Mice have been extensively used in the study of experimental PCM because of their susceptibility to infection. In the murine model of intravenously induced pulmonary PCM, the lungs are affected in a diffuse pattern [3]. The model reproduces the features of the chronic pulmonary form of the human disease, with an infection similar to the intranasal inoculation of conidia or intratracheal inoculation of yeast cells [1, 2, 4, 5]. These two latter models are closer to how individuals become infected with the fungus. How-

ever the production of conidia is an unstable feature of the mycelial form of the fungus and represents a serious health hazard to the researcher. On the other hand, the intratracheal inoculation frequently causes a concomitant bacterial infection which alters the nature of the inflammatory exudate present in the pulmonary alveoli. The bronchoalveolar lavage (BAL) represents a clinical-laboratory tool of great value for the study of the pathogenesis of several pulmonary diseases, as well as for their diagnosis, monitoring of treatment and assessment of cure [6–8]. Few studies are available about BAL in human and experimental PCM. They usually report an increased number of neutrophils and/or lymphocytes [9–11]. Thus, considering the scarcity of data about BAL during the evolution of PCM, the present study was conducted with the following objectives: (1) To phenotype, quantitate and

80 evaluate the functional activity of inflammatory cells in BAL in an experimental murine model, using susceptible and non-susceptible strains of mice, and (2) To correlate these features with the lung histopathology and the animals specific immune response.

Material and methods Experimental design Isogenic (BALB/c) and non-isogenic (Swiss) mice were infected with P. brasiliensis by the intravenous route and sacrificed 2, 4, 8 and 16 weeks later. At sacrifice, the lungs were removed and examined by conventional histopathologic methods and BAL. BAL was submitted to the following analyses: determination of total and differential number of inflammatory cells (BALB/c & Swiss), phenotyping of the mononuclear cell population (Swiss), ultrastructural identification of the cells (BALB/c), determination of number of fungi (BALB/c & Swiss) and evaluation of the functional status of alveolar macrophages (BALB/c & Swiss). The histopathological features were correlated with the humoral and cellular immune response of the animals (Swiss) and with the BAL findings (BALB/c & Swiss). Experimental infection Animals. The study was conducted on male Swiss and BALB/c mice aged 8–10 weeks, provided by the Animal House of UNESP, Botucatu Campus, and of the Lauro de Souza Lima Institute, Bauru, São Paulo, respectively. Inoculum. Strain 18 of P. brasiliensis from the fungal collection of the Department of Microbiology and Immunology, Faculty of Medicine, USP, cultured in liquid PYG medium (0.5% peptone, 0.5% yeast extract and 0.5% glucose) at 37 ◦ C in a rotary shaker (220 rpm) was used. After 7 days of culture, a suspension of yeast-like P. brasiliensis cells was prepared at a concentration of 4 × 106 fungi/ml. Dose and route of inoculation. The animals were infected intravenously with 1 × 106 viable fungi. Viability was determined by vital staining with ethidium bromide-fluorescein diacetate [12]. Sacrifice. The animals were sacrificed and exsanguinated by cardiac puncture 2 (Swiss = 22; BALB/c

= 12), 4 (Swiss = 22; BALB/c = 13), 8 (Swiss = 30; BALB/c = 15) and 16 (Swiss = 15) weeks after inoculation. Serum was stored at −20 ◦ C until use. Groups of non-infected mice were used as controls in each experimental time (Swiss = 35; BALB/c = 43). BAL preparation. After lung removal, the trachea was cannulated and 1 ml sterile 0.85% saline (SSE) was instilled and aspirated from the lungs with the aid of a syringe. The procedure was repeated 6 to 7 times. The cell suspension was centrituged and the cell pellet resuspended in 1 ml SSE. BAL evaluation. The following parameters were evaluated: 1. Total number of inflammatory cells. After diluting the cells suspension with 0.5% crystal violet in 30% glacial acetic acid, the total number of cells was counted in a Neubauer chamber. 2. Differential counts of inflammatory cells. 3 × 105 BAL cells in 300 µl were cytocentrifuged and stained with routine Shorr staining technique. The percentages of macrophages, lymphocytes, neutrophils and multinucleated giant cells (MGC) were determined by counting 300 cells with a 100× objective. The number of nuclei present in the MGC was determined in 50 cells and the results were divided into 3 groups: cells with 3 to 5 nuclei, cells with 6 to 10 nuclei, and cells with 11 or more nuclei [13]. 3. T. lymphocytes subsets. The determination was performed by an immunochemical method using the avidin-biotin-peroxidase complex (ABC). The cytocentrifugates were incubated with rat monoclonal antibodies against the following surface antigens of mouse T lymphocytes: Thy 1–2 (panT), L3T4 (CD4 -helper/inducer) and Lyt-2 (CD8suppressor/cytotoxic) (Becton Dickinson, Mountain View, California, USA). After incubation with the biotinylated secondary antibody (rabbit anti-rat IgG serum) (Vector Lab., Burligame, California, USA), the preparations were incubated with the ABC complex (Vector Lab.) and with AEC (3amino-9-ethylcarbazole) solution (Biomeda Co., Foster City, California, USA), and counterstained with Mayer hematoxylin. Positive cells were identified on the basis of red-brown staining of the cell membrane. The results are reported as percentage of lymphocytes of each subset for a total of 200 cells.

81 4. Macrophage percentage. The determination was performed by the same immunochemical method as used for lymphocytes. The cytocentrifugates were incubated with mouse monoclonal antihuman macrophage antibody (CD68, KP1) (Dako Lab., Santa Barbara, California, USA) and with the secondary antibody (horse anti-mouse IgG serum) (Vector Lab.). The preparations were then incubated with the ABC complex and with DAB solution (30 ,30 -diaminobenzidine tetrahydrochloride) (Sigma Chemical Co., St. Louis, Missouri, USA), and counterstained with Harris hematoxylin. Positive cells presented brown-stained cytoplasm. The results are reported as percentage of stained macrophages in a total of 200 cells. 5. Ultrastructure of BAL cells. After microcentrifugation of the cell suspension, the pellet was fixed in 2.5% glutaraldehyde, postfixed in 1.0% osmium tetroxide, dehydrated with acetone and embedded in epoxy resin. Sections 600 to 800 Angstrom in thickness were stained with a saturated solution of uranyl acetate and lead citrate and examined with a transmission electron microscope. Several fields in the same sections were photographed at 5000× magnification and the cells were identified in the photographs according to their morphology. 6. Number of fungi. Fungi were counted in 20 fields selected at random using a 40× objective for cytocentrifugates stained by the Gomori-Grocott technique. 7. Functional state of BAL macrophages. (a) Spreading. Alveolar macrophages were evaluated in terms of spreading ability by the technique of Rabinovitch et al. [14]. After allowing the cells to adhere to glass coverslips and incubating in 199 medium (Sigma Chemical Co.), adherent cells were fixed in 2.5% glutaraldehyde and examined by phase microscopy. The percentage of spread macrophages was determined in 100 cells. (b) Release of H2 O2 . BAL cell suspensions at the concentration of 1 × 106 cells/ml in phenol red solution containing peroxidase were added to tissue culture plates and incubated in the presence or absence of 0.1 mg/ml phorbol myristate acetate (PMA) (Sigma Chemical Co.), according to the technique of Russo et al. [15]. The reaction was stopped with 1 N NaOH and absorbance was determined using an automatic ELISA microreader with a 620

nm filter against a blank consisting of phenol red solution and 1N NaOH.

Histopathology. After collecting the lavage, the lungs were fixed in 10% formalin and fragments were prepared by routine techniques for paraffin embedding. Histological sections of 5 µm in thickness were stained with hematoxylin-eosin and the slides were read blindly and scored using a 0–4 + semiquantitative scale in terms of extent of the lesions, granuloma size and numbers of lymphomonuclear cells, polymorphonuclear neutrophils and intact fungal cells in the lesions. Immunologic response. The specific humoral immune response to P. brasiliensis was determined by ELISA and by agar gel microdiffusion (MID). The cell immune response was determined by the footpad test (FPT). The fungal antigen used (AgPb) was obtained by filtration and sonication of yeast-like forms of strains 18, 113 and 192 cultured as previously described [16]. The protein concentration of the antigen mixture was 6 mg/ml, as determined by the method of Lowry et al. [17]. (a) ELISA. We used the technique of Mendes-Giannini et al. [18]. Sera submitted to serial twofold dilutions starting from 1 : 100 were added to the wells of the plate which had been previously sensitized with AgPb and incubated. After washing with buffered saline containing 0.05% Tween 20 (SST-T) and incubating with goat anti-mouse IgG peroxidase-bound serum (Kirkegaard & Perry Lab, Gaithersburg, Maryland, USA), 2.2 mM orthophenylenediamine (OPD) as substrate and hydrogen peroxide were added. The reaction was stopped by the addition of 2 M H2 SO4 and enzyme activity was determined by spectrophotometry at 492 nm. (b) Immunodiffusion. Immunodiffusion was determined by the MID technique according to Silva et al. [19]. (c) FPT. We used the technique of Almeida et al. [20]. Twenty-four hours before sacrifice, 0.05 ml AgPb and 0.05 ml SSE were injected into the right hind footpad (test paw) and into the left hind footpad (control paw), respectively. After sacrifice, the paws were sectioned and weighed; the FPT indices were expressed as the difference in weight between the test and the control paw.

82

Figure 1. Box plots graph showing total numbers of BAL cells for 16 weeks in Swiss and BALB/c mice infected with P. brasiliensis. The lower boundary of the boxes indicates the 25th percentile, the line within the boxes marks the median value and the upper of the boxes indicates the 75th percentile. Error bars above and below the boxes indicate the 90th and 10th percentile.

Correlation between BAL features and histopathology We calculated the correlations of the following parameters: (i) percent BAL lymphocytes with the lymphomononuclear halo of the granulomas; (ii) percent BAL neutrophils with the number of neutrophils in the granulomas and (iii) total number of inflammatory cells in the BAL with the total index of histopathological lesion (IHL). This index was expressed as the sum of the points given to the extent of lesions, granuloma size, number of neutrophils and size of the lymphomononuclear halo of the granulomas.

Statistical analysis Data were analyzed statistically by the non-parametric Kruskal-Wallis test or by analysis of variance for a fully randomized experiment, depending on the type of variable. The level of significance was set at p < 0.05. The non-parametric Spearman rank correlation coefficient was also calculated for pairs of variables.

Correlation between the immunologic and histopathologic findings The levels of the immune response (ELISA; MID and FPT) were correlated with IHL, plus the number of intact fungal cells.

BAL evaluation

Results

Total number of inflammatory cells The total number of cells present in the lavages of the two mouse strains was higher in infected animals than in controls (p < 0.001). The number of cells did not change significantly during the course of infection (Figure 1).

83

Figure 2. Box plots graph showing differential percent countings of BAL cells for 8 weeks in Swiss and BALB/c mice infected with P. brasiliensis. The lower boundary of the boxes indicates the 25th percentile, the line within the boxes marks the median, and the upper of the boxes indicates the 75th percentile. Error bars above and below the boxes indicate the 90th and 10th percentile. M = macrophages; L = lymphocytes; PMN = polymorphonuclear neutrophils; G = giant cells.

84

Figure 3. Cytocentrifugate of BAL of a Swiss mouse at week 16. A. Note numerous neutrophils, macrophages, lymphocytes and multinucleated giant cell (Shorr, 200×). B. Detail showing giant cell with a fungal cell in the cytoplasm (arrows) (Shorr, 400×).

Differential counts of inflammatory cells Macrophages were the predominant cells in BALs of both Swiss and BALB/c mice; the number of lymphocytes, neutrophils and MGC increased during the course of infection (Figures 2 and 3). In BALB/c mice, the percentages of lymphocytes and neutrophils were similar to those detected in the controls during the 8th week of infection (Figure 2). Both Swiss and BALB/c mice presented a larger number of MGC with 3 to 5 nuclei throughout the experiment. Percentage of T lymphocytes and subsets Percent T lymphocytes underwent a slight fall in the lavages of Swiss mice during the 16th week of infection while the CD4+ and CD8+ subset percents remained constant. The CD4/CD8 ratios were constant throughout the infection and ranged from 1.4 to 1.8 (Table 1). Percentage of alveolar macrophages The percentage of cells with the morphological characteristics of macrophages that expressed the KP1 antigen was significantly increased in the BAL of infected Swiss mice compared to the controls (Figure 4). Ultrastructure of BAL cells Macrophages predominated in the BAL of both control and infected animals. Type II pneumocytes represented up to 13.2% of recovered cells (Figure 5). Number of fungi The number of fungi was constant throughout the experiment in the BAL of Swiss mice and decreased in

Table 1. Percentages of T lymphocytes (T lymp), T helper lymphocytes (TH lymp) and T suppressor/cytotoxic lymphocytes (TS/C lymp) in BAL of Swiss micea Sacrifice (week)

T lymp (%)

TH lymp (CD4+ cells) (%)

TS/C lymp (CD8+ cells) (%)

CD4/CD8 ratio

4 8 16

35.2 ± 2.3 33.4 ± 1.1 27.4 ± 2.3

22.8 ± 2.6 21.5 ± 1.3 19.0 ± 1.4

13.4 ± 1.1 12.2 ± 2.0 13.8 ± 1.5

1.7 ± 0.3 1.8 ± 0.4 1.4 ± 0.2

a Results are expressed as mean values ± standard deviation of 4–5 animals per group. Statistical analysis: T lymp: (4 = 8) > 16 (F = 21.20; p < 0.001) TH lymp: CD4+ : 8 = 4 and 16; (4 ≥ 16), (F = 4.13; 0.05 < p < 0.10) T S/C lymp: CD8+ : (4 = 8 = 16) (F = 1.18; p > 0.10)

BALB/c mice at the end of the experiment (week 8) (Figure 6). Functional activity of macrophage in BAL (a) Spreading. The percentage of spread macrophages in the BAL of BALB/c mice was higher in infected animals. During the course of the infection this percentage increased significantly up to the 4th week and plateaued up to the end of the experiment (day 0 = 8.2 ± 1.9; week 2 = 15.7 ± 2.7; week 4 = 29.0 ± 3.5; week 8 = 27.0 ± 2.2; 0 < 2 < 4 = 8; F = 66.71, p < 0.001). (b) H2 O2 Release. Spontaneous H2 O2 release in the two mouse strains did not increase throughout the infection (Figure 7). Under stimulation, Swiss mice released higher levels of H2 O2 which kept

85

Figure 4. Percent countings of KP1+ macrophages in BAL of a Swiss mouse. Results are expressed as means ± standard deviation of 4 to 8 animals per group.

Figure 5. Electron micrography: (A) Macrophage; (B) Type II pneumocyte (−1 µm).

86

Figure 6. Numbers of fungal cells in BAL of Swiss and BALB/c mice sacrificed at week 2, 4, 8 and 16. Results are expressed as means ± standard deviation of 4 to 9 animals per group.

a plateau from week 4 on. In BALB/c mice there was no stimulated H2 O2 release up to week 4, but a significant reduction was observed at week 8. Histopathology The pulmonary histopathology of Swiss mice revealed granulomatous and progressive infection in 82.4% of the animals inoculated. At week 2, the lesions were poorly circumscribed; at week 4, they were more compact, better demarcated and with numerous parasites; starting from the 8th week, they became extensive, more cellular and confluent (Figure 8). In BALB/c mice, pulmonary infection was not progressive. Although the lesions were extensive, confluent and with numerous fungi during the initial week, they decreased in size and contained many fragmented fungi; at week 8 most of the animals exhibited a process of resolution represented by small residual foci of infection. Immunologic evaluation (a) Humoral immunity: The humoral response evaluated by MID in Swiss mice revealed low antibody titers (median = 1 : 1, 1 : 2) throughout the

study period. ELISA revealed high titers (median = 1 : 12,800), which peaked during the 4th weak of infection (median = 1 : 51,200); during the 8th week the levels fell (1 : 12,800) and continued unchanged until the 16th week. (b) Cell immunity: The delayed hypersensitivity response was detected during the 2nd week (FPT mean index = 50 mg) of infection in Swiss mice and maintained a plateau until the end of the experiment (FPT mean index at week 16 = 70 mg). Correlation between BAL features and histopathology: Swiss mice showed a positive correlation between the percentages of neutrophils in the BAL and numbers of neutrophils in the granulomas (r = 0.34667; p < 0.05). BALB/c mice showed a positive correlation between total number of BAL cells and IHL (r = 0.47005; p < 0.05), between percent BAL lymphocytes and lymphomononuclear halo (r = 0.70479; p < 0.01) and between percentage and number of neutrophils in the granulomas (r = 0.51415; p < 0.05). Correlation between the immunologic and histopathologic findings: A positive correlation between total

87

Figure 7. Spontaneous and stimulated release of H2 O2 by alveolar macrophages in BAL of Swiss and BALB/c mice. Results are expressed as means ± standard deviation of the amount released by 105 cells of 8 to 20 animals per group.

88

Figure 8. Pulmonary histopathology of a Swiss mouse at week 8. Note confluent granulomas with prominent lymphomonuclear halo and numerous fungal cells (arrows) (HE; 100×).

IHL and MID (r = 0.67266; p < 0.01) and between IHL and ELISA (r = 0.35294; p < 0.05) was observed. A negative correlation was observed between IHL and FPT (r = −0.32424; p < 0.05).

Discussion In the present study, we obtained progressive granulomatous pulmonary infection in 82.4% of the Swiss mice (susceptible strain), a result similar to that obtained by other investigators who employed the intravenous route of inoculation [3, 21, 22]. In BALB/c mice, the pulmonary infection was non-progressive (non-susceptible strain). However they presented an active infection at the initial phase of the experiment, which indicated that these animals set up early mechanisms of resistance. The elucidation of these protective mechanisms would be of great interest for the treatment and control of PCM.

The cell yield of the BAL is variable and depends on the volume of the solution instilled and recovered, as well as on the type of solution employed for the procedure [23]. In our study, the total number of cells recovered increased significantly during infection as a consequence of lung involvement. In infected animal of both strains, macrophages were detected at higher percentages although we also observed an increase in the percentages of lymphocytes and neutrophils, as well as in the presence of MGC. In BALB/c mice, the percentages of macrophages, lymphocytes and neutrophils during the 8th week of infection were similar to those observed in the controls, in agreement with the regressive nature of the lesions. The results obtained for Swiss mice agree with those obtained by Ferreira [24], who reported an increase in mononuclear cells and neutrophils in the BAL of mice infected by the intravenous route and with those of Tani & Franco [25], who reported an increase in lymphocytes and neutrophils in the BAL

89 of hamsters inoculated by the intratracheal route. They differ somehow from preliminary human data which showed a predominant increase either of neutrophils [9] or lymphocytes [10]. In addition to identifying macrophages on the basis of their morphology by Shorr staining, we also performed immunochemical labelling using the monoclonal antibody KP1 (CD68). This procedure was justified by the fact that the type II pneumocytes recovered by the BAL are difficult to differentiate morphologically from macrophages when routine staining is used. The monoclonal antibody KP1 recognizes the intracellular glycoprotein present in the lysosomes of cells of the monocyte/macrophage lineage, which persists throughout the maturation of these cells and is increased in activated and mature macrophages [26]. Our results demonstrate that the percentage of KP1+ macrophages was increased during infection. At the beginning of the experiment, this percentage was quite reduced, in contrast to the results obtained with Shorr staining showing that approximately 91% of the recovered cells were macrophages. This result supports the interpretation that the glycoprotein recognized by the KP1 antibody is manifested at higher concentrations by activated macrophages, as observed during infection. To further define the percentage of macrophages and type II pneumocytes recovered by the BAL, we carried out an ultrastructural identification of BAL cells. This study revealed that the macrophages were the most numerous cells detected in the BAL of both control and infected BALB/c mice and that type II pneumocytes represented up to 13.2% of the total cells recovered. The percentage of T lymphocytes in the BAL of infected Swiss mice was elevated, with a predominance of CD4+ lymphocytes. The CD4/CD8 ratio was slightly reduced during the late phase of infection, perhaps owing to a reduction of the percentage of CD4+ lymphocytes. All indices were within the 1.3 to 2.1 range reported for normal mice, demonstrating that infection with P. brasiliensis did not change the ratio between these lymphocyte subsets [27–29]. Contrariwise, in human PCM, phenotyping of T lymphocyte subsets in the BAL and peripheral blood of patients revealed a decrease in the percentage of CD4+ T lymphocytes in both; the CD4/CD8 ratio in the BAL was 1.58 ± 0.31, similar to that of peripheral blood and to the values detected in mice [11]. The reduction of CD4+ T lymphocytes in peripheral blood may have occurred as the effect of immunosuppres-

sive factors released by the fungus. These cells did not seem to be trapped into the pulmonary lesions since the reduction was also observed in the lavage. MGC were frequent in the BAL of both strains of infected mice. In intratracheal murine infection with Cryptococcus neoformans the formation of these cells apparently depends on CD4+ T lymphocytes, which represents a possible mechanism of preventing the dissemination of the infectious agent [30]. These cells originate from macrophage fusion and the mechanisms of this fusion are little known, possibly involving the participation of IL-3, IL-4 and IFNgamma [13, 31]. In experimental murine PCM, the MGC of resistant mice present a larger number of nuclei (6 or more) compared to the cells of susceptible animals (3–5 nuclei) [24]. In the present study we did not observe differences in the number of giant cell nuclei between Swiss mice (progressive infection) and BALB/c mice (regressive infection), so that a correlation between this parameter and susceptibility/resistance to P. brasiliensis infection could not be established. The number of fungi in the BAL of Swiss mice with progressive pulmonary infection was always elevated, showing that this is good parameter to indicate active disease. Infection with P. brasiliensis did not promote an increase in spontaneous release of H2 O2 in either mouse strain, as also observed for peritoneal macrophages of resistant and susceptible mice infected by the peritoneal route [32]. Macrophage stimulation with PMA induced a great release of H2 O2 in Swiss mice with progressive infection, suggesting a correlation between impaired macrophage activation and active disease. The correlation between histopathology and immune response in infected animals corresponded to that observed in human PCM. The association between severity of infection, high antibody titers and depressed cell immunity observed in Swiss mice with progressive infection reproduced the picture commonly described in the disseminated forms of human disease [33, 34]. To what extent is the BAL representative of the histopathological alterations observed in the pulmonary alveoli? In our experimental model, the answer to this question was usually positive. The total number of cells and the number of lymphocytes (isogeneic mice) and the number of neutrophils (isogeneic and non-isogeneic mice) in the BAL reflected the corresponding histopathological changes. This finding is

90 very important because it demonstrates that the BAL can be used as a reliable and accurate tool for the evaluation of the inflammatory changes occurring in the pulmonary alveoli. Our findings agree with those reported for the murine model of pulmonary histoplasmosis and the infection with respiratory syncytial virus [35, 36]. To our knowledge, this comparative analysis has not yet been done in human pulmonary PCM. The present data, taken as whole, seem to indicate that the model of murine PCM induced by the intravenous route with analysis of the BAL is suitable for the evaluation of the intensity and pattern of pulmonary infection induced. The procedure may become an important auxiliary method for diagnosis and the evaluation of host response to the parasite in human PCM, especially in cases with isolate lung involvement by the mycosis.

8.

9.

10.

11.

12. 13.

14.

15.

Acknowledgements We would like to thank Dr. Elisa Gregorio for assistance in the ultrastructural study, Dr Paulo Curi for assistance in the statistical analysis and to technicians of the Cytopathology, Immunohistochemistry and Immunopathology laboratories for their help. This research was partly supported by FAPESP.

16.

References

19.

1. Franco M. Host-parasite relationship in paracoccidioidomycosis. J Med Vet Mycol 1987; 25: 5–18. 2. McEwen JG, Bedoya V, Patiño MM, Salazar ME, Restrepo A. Experimental murine paracoccidioidomycosis induced by the inhalation of conidia. J Med Vet Mycol 1987; 25: 165–75. 3. Moscardi-Bacchi M, Franco M. Experimental paracoccidioidomycosis in the mouse. III – Histopathological and immunological findings after intravenous infection in the presence or absence of previous immunization. Rev Soc Bras Med Trop 1985; 18: 101–8. 4. Bedoya V, McEwen JG, Tabares AM, Jaramillo FU, Restrepo A. Pathogenesis of paracoccidioidomycosis: A histopathological study of the experimental murine infection. Mycopathologia 1986; 94: 133–44. 5. Castaneda E, Brummer E, Pappagianis D, Stevens DA. Chronic pulmonary and disseminated paracoccidioidomycosis in mice: quantitation of progression and chronicity. J Med Vet Mycol 1987; 25: 377–87. 6. Hunninghake GW, Gadek JE, Kawanami O, Ferrans VJ, Crystal RG. Inflammatory and immune processes in the human lung in health and disease: evaluation by bronchoalveolar lavage. Am J Pathol 1979; 97: 149–206. 7. Daniele RP, Elias JA, Epstein PE, Rossman MD. Bronchoalveolar lavage: role in the pathogenesis, diagnosis, and management of interstitial lung disease. Ann Intern Med 1985; 102: 93–108.

17.

18.

20.

21. 22. 23.

24.

25. 26. 27.

Crystal RG, Reynolds HY, Kalica AR. Bronchoalveolar lavage. The report of an international conference. Chest 1986; 90: 122–31. Boscardin RN, Brandão H, Balla A. Bronchoalveolar lavage findings in pulmonary paracoccidioidomycosis. J Med Vet Mycol 1985; 23: 143–6. Carvalho CRR, Kairalla RA, Carvalho-Filho RS, et al. O lavado broncoalveolar nas pneumopatias intersticiais pulmonares. Rev Hosp Clin Fac Med Univ São Paulo 1987; 42:110–4. Tapia FJ, Goihman-Yahr M, Cáceres-Dittmar G, et al. Leukocyte immunophenotypes in bronchoalveolar lavage fluid and peripheral blood of paracoccidioidomycosis, sarcoidosis and silicosis. Histol Histopathol 1991; 6: 395–402. Calich VLG, Purchio A, Paula CR. A new fluorescent viability test for fungi cells. Mycopathologia 1978; 66: 175–7. McInnes A; Rennick DM. Interleukin 4 induces cultured monocytes/macrophages to form giant multinucleated cells. J Exp Med 1988; 167: 598–611. Rabinovitch M, Manejias RE, Russo M, Abbey EE. Increased spreading of macrophages from mice treated with interferon inducers. Cell Immunol 1977; 29: 86–95. Russo M, Teixeira HC, Marcondes MCG, Barbuto JAM. Superoxide-indepedent hydrogen peroxide release by activated macrophages. Brazilian J Med Biol Res 1989; 22: 1271–3. Silva MIC, Carvalho IM, Franco TN, et al. The use of a mixture of somatic and culture filtrate antigens in the evaluation of the immune response to Paracoccidioides brasiliensis. J Med Vet Mycol 1991; 29: 1–4. Lowry OH, Rosebrough NJ, Farr AL, Randall RJ. Protein measurement with the folin phenol reagent. J Biol Chem 1951; 193: 265–75. Mendes-Giannini MJS, Camargo ME, Lacaz CS, Ferreira AW. Immunoenzymatic absorption test for serodiagnosis of paracoccidioidomycosis. J Clin Microbiol 1984; 20: 103–8. Silva MIC, Chamma LG, Franco M. Reação de microimunodifusão em gel de ágar no diagnóstico sorológico da paracoccidioidomicose. Rev Inst Med Trop São Paulo 1989; 31: 40–3. Almeida AV, Fogaça J, Chamma LG, Fecchio D, Franco M. Optimization of a mouse immunization protocol with Paracoccidioides brasiliensis antigens. Mem Inst Oswaldo Cruz 1995; 90: 45–9. Mackinnon JE. Pathogenesis of South American blastomycosis. Trans R Soc Trop Med Hyg 1959; 53: 487–94. Linares LI, Friedman L. Experimental paracoccidioidomycosis in mice. Infect Immun 1972; 5: 681–7. Sugar AM, Brummer E, Stevens DA. Murine pulmonary macrophages: evaluation of lung lavage fluids, miniaturized monolayers, and candidacidal activity. Am Rev Respir Dis 1983; 127: 110–2. Ferreira CSA. Características imunopatológicas da paracoccidioidomicose experimental murina induzida por via intravenosa. [Dissertation]. São Paulo, Instituto de Ciências Biológicas, Universidade de São Paulo, 1993, 133 pp. Tani EM, Franco M. Pulmonary cytology in paracoccidioidomycosis. Acta Cytologica 1984; 28: 571–5. Weiss LM, Arber DA, Chang KL. CD68: a review. Appl Immunohistochem 1994; 2: 2–8. Takizawa H, Suko M, Kobayashi N, et al. Spontaneous regression in murine hypersensitivity pneumonitis: lack of immunological tolerance. Int Arch Allergy Appl Immunol 1989; 89: 173–80.

91 28. Kimpen JLL, Rich GA, Mohar CK, Ogra PL. Mucosal T cell distribution during infection with respiratory syncytial virus. J Med Virol 1992; 36: 172–9. 29. Freudenberg N, Konarske K, Riede UN, et al. Function and “homing” of the lung macrophages. I. Evidence of functional heterogeneity of mobile cells of the murine lung parenchyma in the bronchoalveolar lavage (BAL). Virchows Arch B Cell Pathol 1993; 64: 191–7. 30. Hill JO. CD4+ T cells cause multinucleated giant cells to form around Cryptococcus neoformans and confine the yeast within the primary site of infection in the respiratory tract. J Exp Med 1992; 175: 1685–95. 31. Most J, Neumayer HP, Dierich NMP. Cytokine-induced generation of multinucleated giant cells “in vitro” requires interferon-gamma and expression of LFA-1. Eur J Immunol 1990; 20: 1661–7. 32. Teixeira HC. Ativação de macrófagos peritoneais e de linfócitos B esplênicos em camundongos resistentes e suscetíveis durante a infeção por Paracoccidioides brasiliensis. [Thesis]. São Paulo, Instituto de Ciências Biomédicas, Universidade de São Paulo, São Paulo, 1991. 120 pp.

33.

Biagioni L, Souza MJ, Chamma LG, et al. Serology of paracoccidioidomycosis. II. Correlation between class specific antibodies and clinical forms of the disease. Trans R Soc Trop Med Hyg 1984; 78: 617–21. 34. Mota NGS, Rezkallah-Iwasso MT, Peraçoli MTS, et al. Correlation between cell-mediated immunity and clinical forms of paracoccidioidomycosis. Trans R Soc Trop Med Hyg 1985; 79: 765–72. 35. Baughman RP, Kim CK, Vinegar A, et al. The pathogenesis of experimental pulmonary histoplasmosis. Correlative studies of histopathology, bronchoalveolar lavage, and respiratory function. Am Rev Respir Dis 1986; 134: 771–6. 36. Vaux-Peretz F, Chapsal JM, Meignier B. Comparison of the ability of formalin-inactivated respiratory syncytial virus, immunopurified F, G and N proteins and cell lysate to enhance pulmonary changes in BALB/c mice. Vaccine 1992; 10: 113–8. Address for correspondence: Marcello Franco, M.D., Department of Pathology, Medical School, UNESP, Cep 18618-000, Botucatu, São Paulo, Brazil Phone: 00-55-14-8212121; Fax: 00-55-14-8212348