- Email: [email protected]
Paracoccidioides brasiliensis exoantigens: recognition by IgG from patients with different clinical forms of paracoccidioidomycosis
Microbes and Infection 5 (2003) 1205–1211 www.elsevier.com/locate/micinf
Original article
Paracoccidioides brasiliensis exoantigens: recognition by IgG from patients with different clinical forms of paracoccidioidomycosis Ademilson Panunto-Castelo a, Gilson Freitas-da-Silva b, Imaculada Conceição Bragheto a, Roberto Martinez b, Maria-Cristina Roque-Barreira a,* a
Departamento de Biologia Celular e Molecular e Bioagentes Patogênicos, Faculdade de Medicina de Ribeirão Preto, Universidade de São Paulo, Av. Bandeirantes 3900, 14049-900 Ribeirão Preto, SP, Brazil b Departamento de Clínica Médica, Faculdade de Medicina de Ribeirão Preto, Universidade de São Paulo, Ribeirão Preto, SP, Brazil Received 6 June 2003; accepted 17 July 2003
Abstract Serum antibodies against antigens of Paracoccidioides brasiliensis have been one of the major diagnostic indicators of paracoccidioidomycosis (PCM). In the present study, released antigen preparations (exoAg) obtained from P. brasiliensis isolates were characterized in terms of their protein components electrophoretically detectable and recognizable by sera (IgG) of patients. Among five different isolates (DGO, C-9, BAT, Pb-18 and B-339) the electrophoretic profiles of exoAg varied greatly. A total of 28 different components were detected, 11 of them shared by all isolates. The most representative preparation was BAT-exoAg, which presented the largest number of protein bands (23) and the highest frequency of reacting bands (19) with sera from patients with active PCM (n = 40). Six bands reacted with more than 20% of sera. Independently of clinical forms, the sera recognized the 43-kDa (97% of tested sera), 160-kDa (78% of tested sera) and 70-kDa (60% of tested sera) antigens more frequently. Sera from patients with severe forms of acute (n = 14) or chronic (n = 10) PCM recognized a greater number of antigens, with a higher frequency, than those from moderate forms. The most pronounced reduction in reactivity was provided by sera of patients that became asymptomatic at the beginning of treatment. Remnant reactivity with BAT-exoAg was detected after clinical recovery, especially with those of 43, 70 and 160 kDa. The latter presented a stable recognition frequency (60%) during the entire follow-up, allowing us to suppose that the IgG reactivity against the 160-kDa antigen constitutes a possible persistent marker of P. brasiliensis infection. © 2003 Éditions scientifiques et médicales Elsevier SAS. All rights reserved. Keywords: Paracoccidioidomycosis; Paracoccidioides brasiliensis exoantigen; Serological IgG response
1. Introduction Paracoccidioides brasiliensis is a dimorphic human pathogenic fungus that causes paracoccidioidomycosis (PCM), the most prevalent deep-seated mycosis in Latin America, occurring predominantly in Brazil, Venezuela and Colombia [1]. PCM presents a broad spectrum of clinical and pathological manifestations ranging from asymptomatic pulmonary infection to severely disseminated disease [2]. PCM infection can be asymptomatic and detected in healthy individuals who live in endemic areas and are positive to the paracoccidioidin skin test. The chronic progressive form of the disease (CF) predominantly affects adult males, with a high frequency of pulmonary, mucosal, cutaneous, and adre* Corresponding author. Tel.: +55-16-602-3062; fax: +55-16-633-6631. E-mail address: [email protected] (M.-C. Roque-Barreira). © 2003 E´ditions scientifiques et médicales Elsevier SAS. All rights reserved. doi:10.1016/j.micinf.2003.07.008
nal involvement. In contrast, the acute form (AF) affects young patients of both sexes equally, and is characterized by systemic lymph node involvement, hepatosplenomegaly, as well as other forms of visceral involvement. Patients with CF usually exhibit lower levels of specific antibodies and adequate cellular immune responses, while those with AF typically show high levels of specific antibodies, polyclonal activation of B cells, antigenemia, and impaired cellular immune responses [1,3,4]. Serum antibodies against antigens of P. brasiliensis constitute one of the major diagnostic indicators of PCM, and their detection is especially important in cases of patients with cryptic lesions in internal organs. In addition, serologic procedures have proven valuable in monitoring the PCM patient’s response to treatment [5–8]. Among the antigens used in serological methods, gp43 is the most frequently studied P. brasiliensis component [1,5,9–12]. Although the
1206
A. Panunto-Castelo et al. / Microbes and Infection 5 (2003) 1205–1211
assays applying gp43 have been shown to be useful for the diagnosis of PCM infection, other antigens should also be considered. The present study aimed to characterize a set of P. brasiliensis isolates in terms of released antigens detectable by SDS-PAGE and to assay the most representative preparation of the antigen spectrum against serum IgG of patients with different clinical forms of PCM. 2. Material and methods
(before treatment, in the first year and in the second year after the beginning of treatment) and used to follow-up the serological reactivity of serum IgG against exoAg derived from the BAT isolate of P. brasiliensis. Thirty-seven serum samples were obtained from a control group, which included nine patients with histoplasmosis, seven with aspergillosis, two with cryptococcosis, two with candidiasis, three with tuberculosis, and two with rheumatoid arthritis, as well as 12 healthy individuals. The ethics committee of the hospital approved the study. 2.3. Electrophoresis and Western blotting
2.1. P. brasiliensis isolates and preparation of fungus exoantigen (exoAg) Five P. brasiliensis isolates (DGO, C-9, BAT, Pb18 and B-339) were selected for this study. Three of them were isolated from lesions of patients with the acute (C-9 and BAT) or chronic (DGO) form of PCM followed-up at the Hospital das Clínicas de Ribeirão Preto. The other two (Pb18 and B-339) are known virulent isolates, largely used as sources of antigens by several laboratories. The isolates were grown in solid Fava Netto medium at room temperature (RT) and subcultured every 15 days, for mycelial form. Yeast cells were grown in the same medium, at 35 °C, by subculturing every 7 days. To prepare the exoAg of each isolate, the yeast cells were transferred from the solid medium to a liquid medium containing 3% (w/v) neopeptone, 1.8% (w/v) glucose, 0.009% (w/v) asparagine, and 0.125% (w/v) thiamine (Negroni, 1966), at 37 °C. After 14-day culture, the cells were treated with an aqueous thimerosal solution (1:10 000) and left to stand for 24 h at 4 °C. Yeast cells were counted in a Neubauer chamber and assayed for viability with diacetate of fluorescein and ethidium bromide. The supernatant was filtered through Whatman no. 1 paper and centrifuged at 1050 × g for 10 min, constituting the preparation of exoAg from P. brasiliensis. Phenyl-methyl-sulfonyl fluoride (PMSF, 1 mM) and N-a-tosyl-L-lysyl-chloromethyl ketone (TLCK, 100 µM) were added to the exoAg preparations. ExoAg was concentrated and dialyzed against phosphatebuffered saline (PBS, 0.01 M; pH 7.2) by ultra filtration in YM-10 membrane (Amicon Division, W.R. Grace & Co., Beverly, MA, USA). After protein dosage (Coomassie Plus Protein Assay Reagent, Pierce Chemical Co., Rockford, IL, USA), preparations were stored at –20 °C until use. 2.2. Serum samples We used serum samples from 60 patients with PCM followed up at the Hospital das Clínicas de Ribeirão Preto. Diagnosis of PCM was established by identification of yeast forms of P. brasiliensis in clinical specimens and serology. Two groups of patients were defined. The first group comprised 20 patients who had not received antimycotic therapy: seven CF, and 13 AF. The other group included 40 patients: 20 CF and 20 AF. Serum samples from patients of this group were collected during three different periods of the disease
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed with 5.5–15% gel using a Mini V-8.10 Vertical Gel Electrophoresis System (Gibco BRL, Gaithersburg, MD) [13]. Samples were suspended in 0.5 M Tris–HCl buffer, pH 6.5, with 2% (w/v) SDS and 2.5% mercaptoethanol. After running (80–120 mA, 200 V), the gels were silver stained according to the protocols to detect proteins [14] or oligosaccharides [15]. Molecular mass standards from Amersham Biotech were used. In some experiments, electrophoresed materials were blotted to nitrocellulose membranes, using the Mini V-8.10 Blot System, according to the manufacturer’s instructions (Gibco BRL). The nitrocellulose membranes containing electrophoresed exoAg were blocked with 5% (w/v) skim milk in PBS, for 1 h, at RT. The membranes were incubated with human sera diluted 1:100 in 1% skim milk (w/v) in PBS containing 0.05% (v/v) Tween-20 (PBS-T), for 2 h, at RT, then washed three times with PBS-T and developed with alkaline phosphatase-conjugated goat antiheavy chain of human IgG antibody (Sigma Chemical Co., St. Louis, MO, USA) for 1 h at RT. The reactions were visualized with 5-bromo-4-chloro3-indolyl phosphate/nitroblue tetrazolium substrate (Sigma). 3. Results We evaluated five different P. brasiliensis isolates to identify a preparation of exoAg suitable for serological study. After 14-day culture in liquid medium, all isolates (DGO, C-9, BAT, Pb-18 and B-339) contained more than 90% viable yeast forms and were able to form colonies in solid medium (from 1200 to 2000 CFU). They provided exoAg preparations with protein concentrations of 40–50 mg/ml. The comparative analysis of electrophoretic profiles of the exoAg preparations showed a large variation from isolate to isolate. Considering the set of exoAg preparations, a total of 28 different protein components were detected, according to their electrophoretical mobility on SDS-PAGE (Table 1). Eleven of these components, presenting apparent molecular masses of 70, 68, 64, 60, 50, 43, 40, 33, 30, 24 and 22 kDa, were shared by all isolates. The largest number of components was detected in the exoAg derived from BAT, which showed 23 protein components, distributed in a range of molecular
A. Panunto-Castelo et al. / Microbes and Infection 5 (2003) 1205–1211
1207
Table 1 Silver-stained bands on the SDS-PAGE of the exoAg preparations derived from five different P. brasiliensis isolates Silver-stained bands (kDa) 172 160 136 107 104 95 90 82 70 68 64 60 59 54 50 43 42 40 38 36 33 32 30 28 24 22 20 17 Ag/exoAg Recognized Ag (%)
Pb-18
Exoantigen from isolates B-339 DGO C-9 15 15a
BAT 16
0 0 3 3
3 3 4 0
0 4 0 0
6 7 9 3 12 3 4 3 6 4 20
0 3 0 0 3 3 0
0 3 0 3 0 0 4 4 0
0 20 9 (45)
20 15 (75)
0 4 0 3 3 3 0 0 4 0 0 0 0 0 3 0 3 0
19 8 (42.1)
0 3 0 6 3 3 3 0 0 0 0 0 3 0 3 3 3 0 3 0 21 11 (52.4)
0 3 3 3 8 3 6 6 0 6 3 20 0 8 3 3 3 6 4 3 3 0 23 19 (82.6)
Reactivity of the serum samples from PCM patients (n = 20) by immunoblotting with each component of the exoAg preparations. Filled cells represent silver-stained band. a Number of PCM patient serum samples (20) that recognized exoAg component.
masses from 17 to 160 kDa. Densitometric analysis of representative gels silver stained for protein detection, showed in Fig. 1, panel A, indicated that the 43-kDa component (gp43), the best-characterized antigen from P. brasiliensis in the literature [10], was detected in highest amounts in B-339, BAT and Pb18 isolates. PAS staining for sugar detection in SDS-PAGE gels developed gp43 as well as exoAg components with the highest molecular mass from the isolates DGO (172 kDa), C-9 (172 kDa) and BAT (160 kDa) (Fig. 1, panel B). The exoAg preparations were analyzed by Western blotting, using 20 sera from PCM patients. Among the 28 exoAg components identified by SDS-PAGE analysis (Table 1), 21 were reactive with IgG from at least three sera (Table 1), whereas no serum recognized the other seven proteins. ExoAg preparation from isolate BAT (BAT-exoAg) showed the highest number of reactive components compared with the other isolates, since among the 23 silver-stained bands on
Fig. 1. SDS-PAGE analysis of exoAg preparations obtained from five different isolates of P. brasiliensis: Pb18, lane 1; 339, lane 2; DGO, lane 3; C9, lane 4; BAT, lane 5. Panel A, the silver-stained gel for protein detection. Panel B, the PAS-stained gel for sugar detection. The molecular masses of standard markers are indicated on the left side of each panel.
SDS-PAGE of BAT-exoAg (Table 1), 19 (82.6%) were developed by IgG from patients with PCM on Western blotting (Table 1). The second and third greatest numbers of reactive components were provided by exoAg derived from isolates B-339 (75% of its 20 components were recognized) and C-9 (52% of its 21 components were recognized), respectively. A lower number of antigens were recognized by IgG in exoAg derived from the other two isolates (Pb-18, nine bands; DGO, eight bands). As shown in Table 1, gp43 was the most frequently recognized antigen, 100% detected in the isolates BAT and B-339 by IgG from patients with PCM. The high molecular mass antigens (172 or 160 kDa), contained in exoAg from three isolates (DGO, C-9 and BAT), also had a high frequency of recognition, since they reacted with more than 75% of the sera. Altogether, these results supported our option of using exoAg preparations derived from BAT (BATexoAg) in the ensuing steps of the study. We next evaluated, by Western blotting, the recognition of BAT-exoAg by 40 different sera from patients with untreated active PCM: 20 AF patients and 20 CF patients. It made
1208
A. Panunto-Castelo et al. / Microbes and Infection 5 (2003) 1205–1211
evident a great variability in antigen recognition. In general, sera from patients with acute PCM reacted more intensely and with a greater number of antigens than sera from patients with chronic PCM, presenting apparent molecular masses of 70, 68, 64, 60, 50, 43, 33 and 22 kDa. Panel A, Fig. 2 shows some Western blotting reactions of BAT-exoAg with sera of patients with acute (lanes 1–6) and chronic (lanes 7–12) PCM. Six antigen bands were recognized by more than 20% of sera from patients with acute and chronic PCM. The frequency of reactivity of each of these antigens is represented in panel B, Fig. 2. The highest recognition frequencies were detected towards components with molecular mass 43 kDa (97% of tested sera), 160 kDa (78% of tested sera) and 70 kDa (60% of tested sera). The reactivity frequency of serum samples collected before treatment with each BAT-exoAg component was analyzed according to the clinical classification of patients whose samples were obtained, i.e. severe or moderate forms of acute or chronic PCM (Fig. 3). Sera from patients with severe forms of acute (n = 14) or chronic (n = 10) disease recognized a greater number of antigens, with a higher frequency, than those from patients with moderate forms. Towards the sera of any clinical form, the most reactive antigens were those with apparent molecular mass 43, 160 and
Fig. 2. Recognition of proteins from BAT-exoAg by serum IgG of patients with PCM, evaluated by Western blotting. Panel A: lanes 1–6: representative reactivity profiles obtained with serum samples of patients with the AF of the disease. Lanes 7–12: reactivity profiles with serum from patients with the chronic form of the disease. Lane 13: reaction with normal human serum (negative control). Panel B: relative frequency of recognition of six proteins from BAT-exoAg, reactive with more than 20% of serum samples from patients with active PCM (acute and chronic forms of disease). A typical pattern of Western blotting reactivity is used to show the position of each antigen in the electrophoretic profile.
70 kDa. The lowest frequency of antigen recognition was observed with sera from patients with a moderate chronic form of PCM. From this group of patients we obtained the only serum sample unable to react with gp43, an antigen recognized 100% by all tested sera collected from all other forms of the disease, before treatment. Sera from patients with severe acute PCM were outstanding due to their ability to recognize all BAT antigens and by their high reactivity with the antigens of apparent molecular mass 38 and 60 kDa, which were less frequently recognized by the sera from patients with other clinical forms of PCM. We studied the frequency of recognition of BAT-exoAg by serum samples from patients with acute or chronic PCM collected before treatment and during the first and second year of follow-up. All the BAT-exoAg antigens were reactive with at least one serum sample collected from patients with acute PCM before treatment. As shown in Fig. 4, panel A, the 43-, 160- and 70-kDa antigens accounted for the highest reactivity frequencies, i.e. 100%, 95% and 65%, respectively. The sera collected during the first year of treatment provided a very similar frequency of antigen recognition. On the other hand, those collected in the second year recognized a smaller number of antigens and presented a lower frequency (75%, 60% and 35%, respectively) of reaction with 43-, 160- and 70-kDa antigens. There is a 30% lower frequency of BATexoAg antigen recognition by serum samples collected from patients in the second year of follow-up, than by those collected before treatment. The sera of untreated patients with chronic PCM did not recognize five antigens of BAT-exoAg preparation, such as those with 90, 33, 32, 22 and 20 kDa (data not shown). Again, the 43-, 160- and 70-kDa antigens had the highest frequency of recognition, although the proportions reached (95%, 60% and 55%, respectively) were lower than those observed with sera of non-treated patients with acute PCM. Among patients with acute or chronic PCM, the recognition frequency was almost the same when the reaction occurred with sera collected in the first year of treatment. On the other hand, the serum samples collected in the second year after the beginning of treatment from chronic PCM patients were less reactive against most exoAg-BAT antigens, such as the 43-kDa antigen (from 95% to 55%) and the 70-kDa antigen (from 55% to 30%). Interestingly, the frequency of IgG reactivity with the 160-kDa antigen was maintained, given that it was recognized by 60% of serum samples collected before treatment or during the first treatment year, and by 55% of serum samples obtained in the second treatment year. In general terms, we can assume that in patients with chronic PCM the frequency of antigen recognition by serum IgG antibodies is more time-maintained than in patients with acute disease. Reaction with gp43 was an exception, since it decreased in both clinical PCM forms. In view of the clinical observation that 60% of the patients followed became asymptomatic after 12 months of specific treatment, we evaluated the frequency of reactivity of each antigen of BAT-exoAg with 40 sera from patients treated
A. Panunto-Castelo et al. / Microbes and Infection 5 (2003) 1205–1211
1209
Fig. 3. Proportion of sera of patients presenting different clinical forms of PCM, before treatment, reacting with each protein antigen (identified by its apparent molecular mass) contained in exoAg-BAT. The clinical forms are chronic moderate (CM, n = 10), chronic severe (CS, n = 10), acute moderate (AM, n = 6) and acute severe (AS, n = 14).
during 1 year, 16 of them still symptomatic and 24 who became asymptomatic. Considering all possible reactions, the frequency of antigen recognition was 2.6 times lower with sera of asymptomatic patients. Among the antigens recognized by at least half of sera samples from symptomatic patients, the frequency of recognition by sera from asymptomatic patients decreased by 66% with reference to the 70-kDa antigen, 51% to the 43-kDa antigen and 46% to the 160-kDa antigen (data not shown). We also assayed the reactivity of BAT-exoAg against sera from nine ex-patients who had been treated and recovered from PCM. Four of them slightly recognized gp43. Four distinct sera recognized high molecular mass antigen(s). The remaining serum reacted very slightly with the 30-kDa antigen from the BAT-exoAg preparation (data not shown).
Fig. 4. Distribution of recognition rate of BAT-exoAg protein antigens by sera from patients with PCM. The serum samples were obtained from patients with acute (AF) (panel A) or chronic (CF) (panel B) PCM, collected before treatment (BF) (n = 20), in the first year after the beginning of treatment (AF) (1–11 months) (n = 20), or in the second year after the beginning of treatment (12–24 months) (n = 20). Only the antigens recognized by no less than 20% of sera from at least one of the groups were included in each panel.
In order to verify whether BAT-exoAg components could be reactive with sera IgG from patients with diseases other than PCM or from normal individuals, we developed Western blotting of BAT-exoAg with 37 different sera. Sera of patients with candidiasis, tuberculosis, or rheumatoid arthritis, and normal human sera recognize no exoAg-BAT antigens. Among the blotted antigens, only those with 43, 70 and 160 kDa were reactive with IgG from a few samples (Table 2). There was cross-reactivity of sera from histoplasmosis patients against the 43-, 70- and 160-kDa antigens, from aspergillosis patients against the 70- and 160-kDa antigens, and from cryptococcosis patients against the 160-kDa antigen. These cross-reactivities were not detected when serum samples were assayed diluted four times, a condition
1210
A. Panunto-Castelo et al. / Microbes and Infection 5 (2003) 1205–1211
Table 2 Sera from patients with histoplasmosis, aspergillosis or cryptococcosis can cross-react with at least one of the three antigens most recognized by sera from PCM patients Human sera PCM before treatment (n = 40) Histoplasmosis (n = 9) Aspergillosis (n = 7) Cryptococcosis (n = 2)
43 39 2 0 0
BAT-exoAg (kDa) 70 160 24 31 4 5 3 1 0 1
RIa 0.8 0.4 0.2 0.2
a The recognition index of each group of serum samples is the relation between the number of reactions that occurred and the total number of possible reactions. Sera from patients with candidiasis (n = 2), tuberculosis (n = 3), or rheumatoid arthritis (n = 2), or from normal individuals (n = 12) recognize none of the antigens.
that did not alter the reactivity of sera from patients with PCM with any of these three BAT antigens.
4. Discussion Electrophoretic analysis of five different isolates of P. brasiliensis showed that BAT, isolated from an acute PCM patient, has the highest number of detectable protein antigens. More than 80% of these proteins were recognized by serum IgG of patients with non-treated PCM, a proportion that was greater than that of B-339, an isolate largely utilized for serologic diagnostic of P. brasiliensis infection. The detection and titration of anti-P. brasiliensis antibodies have extensive applications in the diagnosis and follow-up of PCM, even if the methods used ordinarily, such as double immunodiffusion, contraimmunoelectrophoresis, immunofluorescence and ELISA, have low sensibility and specificity [7]. One of the biggest obstacles for the development of new reagents and assays is the large antigenic variability among P. brasiliensis isolates [16]. The exoAg preparation obtained from the BAT isolate may represent a good alternative for the standardization of new assays. Among the proteins detected on the BAT-exoAg preparation, almost 50% were shared by the other isolates and were distributed in a range of molecular mass from 22 to 70 kDa. In the literature, this range is recognized as containing the majority of P. brasiliensis antigens, including gp43, the most highly studied P. brasiliensis antigen [17,18]. High molecular mass antigens (160–172 kDa) were detected in preparations obtained from BAT and two other isolates, but not in isolates Pb18 or B-339. The broad aspect of the electrophoretic bands provided by those high molecular mass antigens suggests that they are heavily glycosylated components. The preliminary features of preparation, in relation to those of the other isolates, motivated us to utilize it to detail the study of antigen recognition by a larger number of sera from PCM patients. The highest frequencies of recognition by serum IgG were obtained with the 43-kDa (97%), 160kDa (78%) and 70-kDa (60%) antigens. High serum reactivities with the 43- and 70-kDa antigens from isolate B-339 have already been reported [5]. However, recognition of the
high molecular mass antigen by almost 80% of sera was not expected, because it has not been previously reported as an important P. brasiliensis antigen. Cross-reactions of 43-, 70and 160-kDa antigens with sera of patients with other mycoses, especially histoplasmosis, were observed and were abolished by diluting the serum samples, without loss of specific reactivity against PCM sera. Cross-reactivity of the 43- and 70-kDa antigens with histoplasmosis sera was previously observed; in the case of gp43, the cross-reactivity was attributed to the sugar moiety of the molecule [19]. Serum IgG from acute PCM patients reacted more intensely and recognized a higher number of electrophoretic fractions of the BAT-exoAg preparation. Disseminated lesions and intense activation of the humoral response, as largely demonstrated by other studies [20,21], characterize this clinical form of the disease. In contrast, sera from patients with mild chronic PCM, typically presenting fewer lesions, showed the lowest IgG reactivity with BAT-exoAg. Although sera of AF patients recognized the 38-kDa antigen more frequently and intensely than CF patients, it is not possible to discriminate recognition patterns from different clinical forms of PCM. A patent reduction in reactivity was observed among patients who became asymptomatic at the beginning of treatment, consistent with previous reports establishing an association of clinical recovery and decline in serologic response [9,22]. The decline was more pronounced and occurred early in acute PCM patients. Nevertheless, in all clinical forms, qualitatively similar recognition patterns were provided by sera from non-treated patients or from patients after 24 months of antifungic therapy. After 12–24 months of treatment, a time interval generally sufficient to achieve clinical recovery, the patients still presented important reactivity with fractions, especially with those of 43, 160 and 70 kDa. Less reactivity was detectable towards the fractions that were more poorly recognized in the active phase of infection. It was not possible to identify a specific antigen from BAT that could represent a marker for active PCM, i.e. with a high recognition index determined by serum samples collected from patients before treatment and with no reactivity against samples collected after clinical cure of the disease. Even so, the serologic follow-up of patients by Western blotting assay is useful, since the reduction in reactivity with the 43- and 70-kDa antigens and the tendency to negativity with the other antigens accompanied the clinical improvement of PCM. Among the proteins constituting the BAT-exoAg preparation, the 160-kDa antigens represent an exception to the general tendency of decreasing reactivity with serum IgG of chronic form patients having recovered. The stable recognition frequency (60%) of the 160-kDa band from the time of diagnosis until 24 months’ follow-up allows us to suppose that the IgG reactivity against this antigen constitutes a possible persistent marker of P. brasiliensis infection. The maintenance of low levels of serum IgG antifungal antigens, as occurred in our study regarding gp43, was previously ob-
A. Panunto-Castelo et al. / Microbes and Infection 5 (2003) 1205–1211
served by other authors, when using double diffusion to detect serum reactivity [23]. In conclusion, we demonstrated that isolate BAT provides an exoAg preparation that is representative of the high antigenic variability observed in several P. brasiliensis isolates. BAT antigens are highly reactive with serum IgG of patients with acute or chronic PCM, indicating its potential application in the diagnosis and follow-up of the disease. In addition to 43- and 70-kDa proteins, we identified a high molecular mass antigen that was highly and persistently recognized by serum antibodies, thereby justifying future efforts to accomplish its structural and biological characterization. Acknowledgements
[9]
[10]
[11]
[12]
[13] [14]
We are grateful to Mrs. Sandra M.O. Thomaz, Mrs. Maria Helena Borges Malta and Mrs. Margarida Maria Passeri for technical assistance and Mr. André Luiz Dardes for critical reading of the manuscript. We thank Dr. Constance Oliver and Dr. Maria Célia Jamur for helpful discussions. Research supported by FAPESP (00/09333-2), CNPq (350418/00-4) and CAPES grants.
[15]
References
[18]
[1] [2] [3]
[4] [5]
[6]
[7]
[8]
E. Brummer, E. Castaneda, A. Restrepo, Paracoccidioidomycosis: an update, Clin. Microbiol. Rev. 6 (1993) 89–117. M. Franco, Host–parasite relationships in paracoccidioidomycosis, J. Med. Vet. Mycol. 25 (1987) 5–18. G. Del-Negro, C.S. Lacaz, V.A. Zamith, A.M. Siqueira, General clinical aspects: polar forms of paracoccidioidomycosis, the disease in childhood, in: M. Franco, C.S. Lacaz, A. Restrepo-Moreno, G. DelNegro (Eds.), Paracoccidioidomycosis, CRC Press, Boca Raton, 1994, pp. 225–232. A. Restrepo, The ecology of Paracoccidioides brasiliensis: a puzzle still unsolved, Sabouraudia 23 (1985) 323–334. M.H. Blotta, Z.P. Camargo, Immunological response to cell-free antigens of Paracoccidioides brasiliensis: relationship with clinical forms of paracoccidioidomycosis, J. Clin. Microbiol. 31 (1993) 671–676. Z.P. Camargo, J.L. Gesztesi, E.C. Saraiva, C.P. Taborda, A.P. Vicentini, J.D. Lopes, Monoclonal antibody capture enzyme immunoassay for detection of Paracoccidioides brasiliensis antibodies in paracoccidioidomycosis, J. Clin. Microbiol. 32 (1994) 2377–2381. M.J.S. Mendes-Giannini, G. Del-Negro, A.M. Siqueira, Serodiagnosis, in: M. Franco, C.S. Lacaz, A. Restrepo-Moreno, G. Del-Negro (Eds.), Paracoccidioidomycosis, CRC Press, Boca Raton, 1994, pp. 345–357. C.P. Taborda, Z.P. Camargo, Diagnosis of paracoccidioidomycosis by dot immunobinding assay for antibody detection using the purified and specific antigen gp43, J. Clin. Microbiol. 32 (1994) 554–556.
[16]
[17]
[19]
[20]
[21]
[22]
[23]
1211
Z.P. Camargo, C. Unterkircher, L.R. Travassos, Identification of antigenic polypeptides of Paracoccidioides brasiliensis by immunoblotting, J. Med. Vet. Mycol 27 (1989) 407–412. P.S. Cisalpino, R. Puccia, L.M. Yamauchi, M.I. Cano, J.F. da Silveira, L.R. Travassos, Cloning, characterization, and epitope expression of the major diagnostic antigen of Paracoccidioides brasiliensis, J. Biol. Chem. 271 (1996) 4553–4560. M.R.E. Costa, C.S. Lacaz, M. Kawasaki, Z.P. Camargo, Conventional versus molecular diagnostic tests, Med. Mycol. 38 (Suppl. 1) (2000) 139–145. R. Puccia, S. Schenkman, P.A. Gorin, L.R. Travassos, Exocellular components of Paracoccidioides brasiliensis: identification of a specific antigen, Infect. Immun. 53 (1986) 199–206. U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature 227 (1970) 680–685. H. Blum, H. Beier, H.J. Gross, Improved silver staining of plant proteins, RNA and DNA in polyacrylamide gels, Electrophoresis 8 (1987) 93–99. C.M. Tsai, C.E. Frasch, A sensitive silver stain for detecting lipopolysaccharides in polyacrylamide gels, Anal. Biochem. 119 (1982) 115–119. M. Casotto, S. Paris, Z.P. Camargo, Antigens of diagnostic value in three isolates of Paracoccidioides brasiliensis, J. Med. Vet. Mycol. 29 (1991) 243–253. L.R. Travassos, Immunochemistry of Paracoccidioides brasiliensis antigens, in: M. Franco, C.S. Lacaz, A. Restrepo-Moreno, G. DelNegro (Eds.), Paracoccidioidomycosis, CRC Press, Boca Raton, 1994, pp. 67–86. R. Martins, S. Marques, M. Alves, D. Fecchio, M.F. Franco, Serological follow-up of patients with paracoccidioidomycosis treated with itraconazole using Dot-blot, ELISA and western-blot, Rev. Inst. Med. Trop. Sao Paulo 39 (1997) 261–269. R. Puccia, L.R. Travassos, Forty-three kilodalton glycoprotein from Paracoccidioides brasiliensis: immunochemical reactions with sera from patients with paracoccidioidomycosis, histoplasmosis, or Jorge Lobo’s disease, J. Clin. Microbiol. 29 (1991) 1610–1615. G.M. Del-Negro, C.N. Pereira, H.F. Andrade, S.A. Palacios, M.M. Vidal, C.E. Charbel, G. Benard, Evaluation of tests for antibody response in the follow-up of patients with acute and chronic forms of paracoccidioidomycosis, J. Med. Microbiol. 49 (2000) 37–46. J.P. Bueno, M.J.S. Mendes-Giannini, G.M. Del-Negro, G.M. Assis, C.K. Takiguti, M.A. Shikanai-Yasuda, IgG, IgM and IgA antibody response for the diagnosis and follow-up of paracoccidioidomycosis: comparison of counterimmunoelectrophoresis and complement fixation, J. Med. Vet. Mycol. 35 (1997) 213–217. M.J.S. Mendes-Giannini, J.P. Bueno, M.A. Shikanai-Yasuda, A.M. Stolf, A. Masuda, V. Amato-Neto, A.W. Ferreira, Antibody response to the 43 kDa glycoprotein of Paracoccidioides brasiliensis as a marker for the evaluation of patients under treatment, Am. J. Trop. Med. Hyg. 43 (1990) 200–206. M.F. Ferreira-da-Cruz, A.C.F. Vale, M.C. Espinera, B. Wanke, B.G. Castro, Study of antibodies in paracoccidioidomycosis: follow-up of patients during and after treatment, J. Med. Vet. Mycol. 28 (1990) 151–157.
Original article
Paracoccidioides brasiliensis exoantigens: recognition by IgG from patients with different clinical forms of paracoccidioidomycosis Ademilson Panunto-Castelo a, Gilson Freitas-da-Silva b, Imaculada Conceição Bragheto a, Roberto Martinez b, Maria-Cristina Roque-Barreira a,* a
Departamento de Biologia Celular e Molecular e Bioagentes Patogênicos, Faculdade de Medicina de Ribeirão Preto, Universidade de São Paulo, Av. Bandeirantes 3900, 14049-900 Ribeirão Preto, SP, Brazil b Departamento de Clínica Médica, Faculdade de Medicina de Ribeirão Preto, Universidade de São Paulo, Ribeirão Preto, SP, Brazil Received 6 June 2003; accepted 17 July 2003
Abstract Serum antibodies against antigens of Paracoccidioides brasiliensis have been one of the major diagnostic indicators of paracoccidioidomycosis (PCM). In the present study, released antigen preparations (exoAg) obtained from P. brasiliensis isolates were characterized in terms of their protein components electrophoretically detectable and recognizable by sera (IgG) of patients. Among five different isolates (DGO, C-9, BAT, Pb-18 and B-339) the electrophoretic profiles of exoAg varied greatly. A total of 28 different components were detected, 11 of them shared by all isolates. The most representative preparation was BAT-exoAg, which presented the largest number of protein bands (23) and the highest frequency of reacting bands (19) with sera from patients with active PCM (n = 40). Six bands reacted with more than 20% of sera. Independently of clinical forms, the sera recognized the 43-kDa (97% of tested sera), 160-kDa (78% of tested sera) and 70-kDa (60% of tested sera) antigens more frequently. Sera from patients with severe forms of acute (n = 14) or chronic (n = 10) PCM recognized a greater number of antigens, with a higher frequency, than those from moderate forms. The most pronounced reduction in reactivity was provided by sera of patients that became asymptomatic at the beginning of treatment. Remnant reactivity with BAT-exoAg was detected after clinical recovery, especially with those of 43, 70 and 160 kDa. The latter presented a stable recognition frequency (60%) during the entire follow-up, allowing us to suppose that the IgG reactivity against the 160-kDa antigen constitutes a possible persistent marker of P. brasiliensis infection. © 2003 Éditions scientifiques et médicales Elsevier SAS. All rights reserved. Keywords: Paracoccidioidomycosis; Paracoccidioides brasiliensis exoantigen; Serological IgG response
1. Introduction Paracoccidioides brasiliensis is a dimorphic human pathogenic fungus that causes paracoccidioidomycosis (PCM), the most prevalent deep-seated mycosis in Latin America, occurring predominantly in Brazil, Venezuela and Colombia [1]. PCM presents a broad spectrum of clinical and pathological manifestations ranging from asymptomatic pulmonary infection to severely disseminated disease [2]. PCM infection can be asymptomatic and detected in healthy individuals who live in endemic areas and are positive to the paracoccidioidin skin test. The chronic progressive form of the disease (CF) predominantly affects adult males, with a high frequency of pulmonary, mucosal, cutaneous, and adre* Corresponding author. Tel.: +55-16-602-3062; fax: +55-16-633-6631. E-mail address: [email protected] (M.-C. Roque-Barreira). © 2003 E´ditions scientifiques et médicales Elsevier SAS. All rights reserved. doi:10.1016/j.micinf.2003.07.008
nal involvement. In contrast, the acute form (AF) affects young patients of both sexes equally, and is characterized by systemic lymph node involvement, hepatosplenomegaly, as well as other forms of visceral involvement. Patients with CF usually exhibit lower levels of specific antibodies and adequate cellular immune responses, while those with AF typically show high levels of specific antibodies, polyclonal activation of B cells, antigenemia, and impaired cellular immune responses [1,3,4]. Serum antibodies against antigens of P. brasiliensis constitute one of the major diagnostic indicators of PCM, and their detection is especially important in cases of patients with cryptic lesions in internal organs. In addition, serologic procedures have proven valuable in monitoring the PCM patient’s response to treatment [5–8]. Among the antigens used in serological methods, gp43 is the most frequently studied P. brasiliensis component [1,5,9–12]. Although the
1206
A. Panunto-Castelo et al. / Microbes and Infection 5 (2003) 1205–1211
assays applying gp43 have been shown to be useful for the diagnosis of PCM infection, other antigens should also be considered. The present study aimed to characterize a set of P. brasiliensis isolates in terms of released antigens detectable by SDS-PAGE and to assay the most representative preparation of the antigen spectrum against serum IgG of patients with different clinical forms of PCM. 2. Material and methods
(before treatment, in the first year and in the second year after the beginning of treatment) and used to follow-up the serological reactivity of serum IgG against exoAg derived from the BAT isolate of P. brasiliensis. Thirty-seven serum samples were obtained from a control group, which included nine patients with histoplasmosis, seven with aspergillosis, two with cryptococcosis, two with candidiasis, three with tuberculosis, and two with rheumatoid arthritis, as well as 12 healthy individuals. The ethics committee of the hospital approved the study. 2.3. Electrophoresis and Western blotting
2.1. P. brasiliensis isolates and preparation of fungus exoantigen (exoAg) Five P. brasiliensis isolates (DGO, C-9, BAT, Pb18 and B-339) were selected for this study. Three of them were isolated from lesions of patients with the acute (C-9 and BAT) or chronic (DGO) form of PCM followed-up at the Hospital das Clínicas de Ribeirão Preto. The other two (Pb18 and B-339) are known virulent isolates, largely used as sources of antigens by several laboratories. The isolates were grown in solid Fava Netto medium at room temperature (RT) and subcultured every 15 days, for mycelial form. Yeast cells were grown in the same medium, at 35 °C, by subculturing every 7 days. To prepare the exoAg of each isolate, the yeast cells were transferred from the solid medium to a liquid medium containing 3% (w/v) neopeptone, 1.8% (w/v) glucose, 0.009% (w/v) asparagine, and 0.125% (w/v) thiamine (Negroni, 1966), at 37 °C. After 14-day culture, the cells were treated with an aqueous thimerosal solution (1:10 000) and left to stand for 24 h at 4 °C. Yeast cells were counted in a Neubauer chamber and assayed for viability with diacetate of fluorescein and ethidium bromide. The supernatant was filtered through Whatman no. 1 paper and centrifuged at 1050 × g for 10 min, constituting the preparation of exoAg from P. brasiliensis. Phenyl-methyl-sulfonyl fluoride (PMSF, 1 mM) and N-a-tosyl-L-lysyl-chloromethyl ketone (TLCK, 100 µM) were added to the exoAg preparations. ExoAg was concentrated and dialyzed against phosphatebuffered saline (PBS, 0.01 M; pH 7.2) by ultra filtration in YM-10 membrane (Amicon Division, W.R. Grace & Co., Beverly, MA, USA). After protein dosage (Coomassie Plus Protein Assay Reagent, Pierce Chemical Co., Rockford, IL, USA), preparations were stored at –20 °C until use. 2.2. Serum samples We used serum samples from 60 patients with PCM followed up at the Hospital das Clínicas de Ribeirão Preto. Diagnosis of PCM was established by identification of yeast forms of P. brasiliensis in clinical specimens and serology. Two groups of patients were defined. The first group comprised 20 patients who had not received antimycotic therapy: seven CF, and 13 AF. The other group included 40 patients: 20 CF and 20 AF. Serum samples from patients of this group were collected during three different periods of the disease
Sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) was performed with 5.5–15% gel using a Mini V-8.10 Vertical Gel Electrophoresis System (Gibco BRL, Gaithersburg, MD) [13]. Samples were suspended in 0.5 M Tris–HCl buffer, pH 6.5, with 2% (w/v) SDS and 2.5% mercaptoethanol. After running (80–120 mA, 200 V), the gels were silver stained according to the protocols to detect proteins [14] or oligosaccharides [15]. Molecular mass standards from Amersham Biotech were used. In some experiments, electrophoresed materials were blotted to nitrocellulose membranes, using the Mini V-8.10 Blot System, according to the manufacturer’s instructions (Gibco BRL). The nitrocellulose membranes containing electrophoresed exoAg were blocked with 5% (w/v) skim milk in PBS, for 1 h, at RT. The membranes were incubated with human sera diluted 1:100 in 1% skim milk (w/v) in PBS containing 0.05% (v/v) Tween-20 (PBS-T), for 2 h, at RT, then washed three times with PBS-T and developed with alkaline phosphatase-conjugated goat antiheavy chain of human IgG antibody (Sigma Chemical Co., St. Louis, MO, USA) for 1 h at RT. The reactions were visualized with 5-bromo-4-chloro3-indolyl phosphate/nitroblue tetrazolium substrate (Sigma). 3. Results We evaluated five different P. brasiliensis isolates to identify a preparation of exoAg suitable for serological study. After 14-day culture in liquid medium, all isolates (DGO, C-9, BAT, Pb-18 and B-339) contained more than 90% viable yeast forms and were able to form colonies in solid medium (from 1200 to 2000 CFU). They provided exoAg preparations with protein concentrations of 40–50 mg/ml. The comparative analysis of electrophoretic profiles of the exoAg preparations showed a large variation from isolate to isolate. Considering the set of exoAg preparations, a total of 28 different protein components were detected, according to their electrophoretical mobility on SDS-PAGE (Table 1). Eleven of these components, presenting apparent molecular masses of 70, 68, 64, 60, 50, 43, 40, 33, 30, 24 and 22 kDa, were shared by all isolates. The largest number of components was detected in the exoAg derived from BAT, which showed 23 protein components, distributed in a range of molecular
A. Panunto-Castelo et al. / Microbes and Infection 5 (2003) 1205–1211
1207
Table 1 Silver-stained bands on the SDS-PAGE of the exoAg preparations derived from five different P. brasiliensis isolates Silver-stained bands (kDa) 172 160 136 107 104 95 90 82 70 68 64 60 59 54 50 43 42 40 38 36 33 32 30 28 24 22 20 17 Ag/exoAg Recognized Ag (%)
Pb-18
Exoantigen from isolates B-339 DGO C-9 15 15a
BAT 16
0 0 3 3
3 3 4 0
0 4 0 0
6 7 9 3 12 3 4 3 6 4 20
0 3 0 0 3 3 0
0 3 0 3 0 0 4 4 0
0 20 9 (45)
20 15 (75)
0 4 0 3 3 3 0 0 4 0 0 0 0 0 3 0 3 0
19 8 (42.1)
0 3 0 6 3 3 3 0 0 0 0 0 3 0 3 3 3 0 3 0 21 11 (52.4)
0 3 3 3 8 3 6 6 0 6 3 20 0 8 3 3 3 6 4 3 3 0 23 19 (82.6)
Reactivity of the serum samples from PCM patients (n = 20) by immunoblotting with each component of the exoAg preparations. Filled cells represent silver-stained band. a Number of PCM patient serum samples (20) that recognized exoAg component.
masses from 17 to 160 kDa. Densitometric analysis of representative gels silver stained for protein detection, showed in Fig. 1, panel A, indicated that the 43-kDa component (gp43), the best-characterized antigen from P. brasiliensis in the literature [10], was detected in highest amounts in B-339, BAT and Pb18 isolates. PAS staining for sugar detection in SDS-PAGE gels developed gp43 as well as exoAg components with the highest molecular mass from the isolates DGO (172 kDa), C-9 (172 kDa) and BAT (160 kDa) (Fig. 1, panel B). The exoAg preparations were analyzed by Western blotting, using 20 sera from PCM patients. Among the 28 exoAg components identified by SDS-PAGE analysis (Table 1), 21 were reactive with IgG from at least three sera (Table 1), whereas no serum recognized the other seven proteins. ExoAg preparation from isolate BAT (BAT-exoAg) showed the highest number of reactive components compared with the other isolates, since among the 23 silver-stained bands on
Fig. 1. SDS-PAGE analysis of exoAg preparations obtained from five different isolates of P. brasiliensis: Pb18, lane 1; 339, lane 2; DGO, lane 3; C9, lane 4; BAT, lane 5. Panel A, the silver-stained gel for protein detection. Panel B, the PAS-stained gel for sugar detection. The molecular masses of standard markers are indicated on the left side of each panel.
SDS-PAGE of BAT-exoAg (Table 1), 19 (82.6%) were developed by IgG from patients with PCM on Western blotting (Table 1). The second and third greatest numbers of reactive components were provided by exoAg derived from isolates B-339 (75% of its 20 components were recognized) and C-9 (52% of its 21 components were recognized), respectively. A lower number of antigens were recognized by IgG in exoAg derived from the other two isolates (Pb-18, nine bands; DGO, eight bands). As shown in Table 1, gp43 was the most frequently recognized antigen, 100% detected in the isolates BAT and B-339 by IgG from patients with PCM. The high molecular mass antigens (172 or 160 kDa), contained in exoAg from three isolates (DGO, C-9 and BAT), also had a high frequency of recognition, since they reacted with more than 75% of the sera. Altogether, these results supported our option of using exoAg preparations derived from BAT (BATexoAg) in the ensuing steps of the study. We next evaluated, by Western blotting, the recognition of BAT-exoAg by 40 different sera from patients with untreated active PCM: 20 AF patients and 20 CF patients. It made
1208
A. Panunto-Castelo et al. / Microbes and Infection 5 (2003) 1205–1211
evident a great variability in antigen recognition. In general, sera from patients with acute PCM reacted more intensely and with a greater number of antigens than sera from patients with chronic PCM, presenting apparent molecular masses of 70, 68, 64, 60, 50, 43, 33 and 22 kDa. Panel A, Fig. 2 shows some Western blotting reactions of BAT-exoAg with sera of patients with acute (lanes 1–6) and chronic (lanes 7–12) PCM. Six antigen bands were recognized by more than 20% of sera from patients with acute and chronic PCM. The frequency of reactivity of each of these antigens is represented in panel B, Fig. 2. The highest recognition frequencies were detected towards components with molecular mass 43 kDa (97% of tested sera), 160 kDa (78% of tested sera) and 70 kDa (60% of tested sera). The reactivity frequency of serum samples collected before treatment with each BAT-exoAg component was analyzed according to the clinical classification of patients whose samples were obtained, i.e. severe or moderate forms of acute or chronic PCM (Fig. 3). Sera from patients with severe forms of acute (n = 14) or chronic (n = 10) disease recognized a greater number of antigens, with a higher frequency, than those from patients with moderate forms. Towards the sera of any clinical form, the most reactive antigens were those with apparent molecular mass 43, 160 and
Fig. 2. Recognition of proteins from BAT-exoAg by serum IgG of patients with PCM, evaluated by Western blotting. Panel A: lanes 1–6: representative reactivity profiles obtained with serum samples of patients with the AF of the disease. Lanes 7–12: reactivity profiles with serum from patients with the chronic form of the disease. Lane 13: reaction with normal human serum (negative control). Panel B: relative frequency of recognition of six proteins from BAT-exoAg, reactive with more than 20% of serum samples from patients with active PCM (acute and chronic forms of disease). A typical pattern of Western blotting reactivity is used to show the position of each antigen in the electrophoretic profile.
70 kDa. The lowest frequency of antigen recognition was observed with sera from patients with a moderate chronic form of PCM. From this group of patients we obtained the only serum sample unable to react with gp43, an antigen recognized 100% by all tested sera collected from all other forms of the disease, before treatment. Sera from patients with severe acute PCM were outstanding due to their ability to recognize all BAT antigens and by their high reactivity with the antigens of apparent molecular mass 38 and 60 kDa, which were less frequently recognized by the sera from patients with other clinical forms of PCM. We studied the frequency of recognition of BAT-exoAg by serum samples from patients with acute or chronic PCM collected before treatment and during the first and second year of follow-up. All the BAT-exoAg antigens were reactive with at least one serum sample collected from patients with acute PCM before treatment. As shown in Fig. 4, panel A, the 43-, 160- and 70-kDa antigens accounted for the highest reactivity frequencies, i.e. 100%, 95% and 65%, respectively. The sera collected during the first year of treatment provided a very similar frequency of antigen recognition. On the other hand, those collected in the second year recognized a smaller number of antigens and presented a lower frequency (75%, 60% and 35%, respectively) of reaction with 43-, 160- and 70-kDa antigens. There is a 30% lower frequency of BATexoAg antigen recognition by serum samples collected from patients in the second year of follow-up, than by those collected before treatment. The sera of untreated patients with chronic PCM did not recognize five antigens of BAT-exoAg preparation, such as those with 90, 33, 32, 22 and 20 kDa (data not shown). Again, the 43-, 160- and 70-kDa antigens had the highest frequency of recognition, although the proportions reached (95%, 60% and 55%, respectively) were lower than those observed with sera of non-treated patients with acute PCM. Among patients with acute or chronic PCM, the recognition frequency was almost the same when the reaction occurred with sera collected in the first year of treatment. On the other hand, the serum samples collected in the second year after the beginning of treatment from chronic PCM patients were less reactive against most exoAg-BAT antigens, such as the 43-kDa antigen (from 95% to 55%) and the 70-kDa antigen (from 55% to 30%). Interestingly, the frequency of IgG reactivity with the 160-kDa antigen was maintained, given that it was recognized by 60% of serum samples collected before treatment or during the first treatment year, and by 55% of serum samples obtained in the second treatment year. In general terms, we can assume that in patients with chronic PCM the frequency of antigen recognition by serum IgG antibodies is more time-maintained than in patients with acute disease. Reaction with gp43 was an exception, since it decreased in both clinical PCM forms. In view of the clinical observation that 60% of the patients followed became asymptomatic after 12 months of specific treatment, we evaluated the frequency of reactivity of each antigen of BAT-exoAg with 40 sera from patients treated
A. Panunto-Castelo et al. / Microbes and Infection 5 (2003) 1205–1211
1209
Fig. 3. Proportion of sera of patients presenting different clinical forms of PCM, before treatment, reacting with each protein antigen (identified by its apparent molecular mass) contained in exoAg-BAT. The clinical forms are chronic moderate (CM, n = 10), chronic severe (CS, n = 10), acute moderate (AM, n = 6) and acute severe (AS, n = 14).
during 1 year, 16 of them still symptomatic and 24 who became asymptomatic. Considering all possible reactions, the frequency of antigen recognition was 2.6 times lower with sera of asymptomatic patients. Among the antigens recognized by at least half of sera samples from symptomatic patients, the frequency of recognition by sera from asymptomatic patients decreased by 66% with reference to the 70-kDa antigen, 51% to the 43-kDa antigen and 46% to the 160-kDa antigen (data not shown). We also assayed the reactivity of BAT-exoAg against sera from nine ex-patients who had been treated and recovered from PCM. Four of them slightly recognized gp43. Four distinct sera recognized high molecular mass antigen(s). The remaining serum reacted very slightly with the 30-kDa antigen from the BAT-exoAg preparation (data not shown).
Fig. 4. Distribution of recognition rate of BAT-exoAg protein antigens by sera from patients with PCM. The serum samples were obtained from patients with acute (AF) (panel A) or chronic (CF) (panel B) PCM, collected before treatment (BF) (n = 20), in the first year after the beginning of treatment (AF) (1–11 months) (n = 20), or in the second year after the beginning of treatment (12–24 months) (n = 20). Only the antigens recognized by no less than 20% of sera from at least one of the groups were included in each panel.
In order to verify whether BAT-exoAg components could be reactive with sera IgG from patients with diseases other than PCM or from normal individuals, we developed Western blotting of BAT-exoAg with 37 different sera. Sera of patients with candidiasis, tuberculosis, or rheumatoid arthritis, and normal human sera recognize no exoAg-BAT antigens. Among the blotted antigens, only those with 43, 70 and 160 kDa were reactive with IgG from a few samples (Table 2). There was cross-reactivity of sera from histoplasmosis patients against the 43-, 70- and 160-kDa antigens, from aspergillosis patients against the 70- and 160-kDa antigens, and from cryptococcosis patients against the 160-kDa antigen. These cross-reactivities were not detected when serum samples were assayed diluted four times, a condition
1210
A. Panunto-Castelo et al. / Microbes and Infection 5 (2003) 1205–1211
Table 2 Sera from patients with histoplasmosis, aspergillosis or cryptococcosis can cross-react with at least one of the three antigens most recognized by sera from PCM patients Human sera PCM before treatment (n = 40) Histoplasmosis (n = 9) Aspergillosis (n = 7) Cryptococcosis (n = 2)
43 39 2 0 0
BAT-exoAg (kDa) 70 160 24 31 4 5 3 1 0 1
RIa 0.8 0.4 0.2 0.2
a The recognition index of each group of serum samples is the relation between the number of reactions that occurred and the total number of possible reactions. Sera from patients with candidiasis (n = 2), tuberculosis (n = 3), or rheumatoid arthritis (n = 2), or from normal individuals (n = 12) recognize none of the antigens.
that did not alter the reactivity of sera from patients with PCM with any of these three BAT antigens.
4. Discussion Electrophoretic analysis of five different isolates of P. brasiliensis showed that BAT, isolated from an acute PCM patient, has the highest number of detectable protein antigens. More than 80% of these proteins were recognized by serum IgG of patients with non-treated PCM, a proportion that was greater than that of B-339, an isolate largely utilized for serologic diagnostic of P. brasiliensis infection. The detection and titration of anti-P. brasiliensis antibodies have extensive applications in the diagnosis and follow-up of PCM, even if the methods used ordinarily, such as double immunodiffusion, contraimmunoelectrophoresis, immunofluorescence and ELISA, have low sensibility and specificity [7]. One of the biggest obstacles for the development of new reagents and assays is the large antigenic variability among P. brasiliensis isolates [16]. The exoAg preparation obtained from the BAT isolate may represent a good alternative for the standardization of new assays. Among the proteins detected on the BAT-exoAg preparation, almost 50% were shared by the other isolates and were distributed in a range of molecular mass from 22 to 70 kDa. In the literature, this range is recognized as containing the majority of P. brasiliensis antigens, including gp43, the most highly studied P. brasiliensis antigen [17,18]. High molecular mass antigens (160–172 kDa) were detected in preparations obtained from BAT and two other isolates, but not in isolates Pb18 or B-339. The broad aspect of the electrophoretic bands provided by those high molecular mass antigens suggests that they are heavily glycosylated components. The preliminary features of preparation, in relation to those of the other isolates, motivated us to utilize it to detail the study of antigen recognition by a larger number of sera from PCM patients. The highest frequencies of recognition by serum IgG were obtained with the 43-kDa (97%), 160kDa (78%) and 70-kDa (60%) antigens. High serum reactivities with the 43- and 70-kDa antigens from isolate B-339 have already been reported [5]. However, recognition of the
high molecular mass antigen by almost 80% of sera was not expected, because it has not been previously reported as an important P. brasiliensis antigen. Cross-reactions of 43-, 70and 160-kDa antigens with sera of patients with other mycoses, especially histoplasmosis, were observed and were abolished by diluting the serum samples, without loss of specific reactivity against PCM sera. Cross-reactivity of the 43- and 70-kDa antigens with histoplasmosis sera was previously observed; in the case of gp43, the cross-reactivity was attributed to the sugar moiety of the molecule [19]. Serum IgG from acute PCM patients reacted more intensely and recognized a higher number of electrophoretic fractions of the BAT-exoAg preparation. Disseminated lesions and intense activation of the humoral response, as largely demonstrated by other studies [20,21], characterize this clinical form of the disease. In contrast, sera from patients with mild chronic PCM, typically presenting fewer lesions, showed the lowest IgG reactivity with BAT-exoAg. Although sera of AF patients recognized the 38-kDa antigen more frequently and intensely than CF patients, it is not possible to discriminate recognition patterns from different clinical forms of PCM. A patent reduction in reactivity was observed among patients who became asymptomatic at the beginning of treatment, consistent with previous reports establishing an association of clinical recovery and decline in serologic response [9,22]. The decline was more pronounced and occurred early in acute PCM patients. Nevertheless, in all clinical forms, qualitatively similar recognition patterns were provided by sera from non-treated patients or from patients after 24 months of antifungic therapy. After 12–24 months of treatment, a time interval generally sufficient to achieve clinical recovery, the patients still presented important reactivity with fractions, especially with those of 43, 160 and 70 kDa. Less reactivity was detectable towards the fractions that were more poorly recognized in the active phase of infection. It was not possible to identify a specific antigen from BAT that could represent a marker for active PCM, i.e. with a high recognition index determined by serum samples collected from patients before treatment and with no reactivity against samples collected after clinical cure of the disease. Even so, the serologic follow-up of patients by Western blotting assay is useful, since the reduction in reactivity with the 43- and 70-kDa antigens and the tendency to negativity with the other antigens accompanied the clinical improvement of PCM. Among the proteins constituting the BAT-exoAg preparation, the 160-kDa antigens represent an exception to the general tendency of decreasing reactivity with serum IgG of chronic form patients having recovered. The stable recognition frequency (60%) of the 160-kDa band from the time of diagnosis until 24 months’ follow-up allows us to suppose that the IgG reactivity against this antigen constitutes a possible persistent marker of P. brasiliensis infection. The maintenance of low levels of serum IgG antifungal antigens, as occurred in our study regarding gp43, was previously ob-
A. Panunto-Castelo et al. / Microbes and Infection 5 (2003) 1205–1211
served by other authors, when using double diffusion to detect serum reactivity [23]. In conclusion, we demonstrated that isolate BAT provides an exoAg preparation that is representative of the high antigenic variability observed in several P. brasiliensis isolates. BAT antigens are highly reactive with serum IgG of patients with acute or chronic PCM, indicating its potential application in the diagnosis and follow-up of the disease. In addition to 43- and 70-kDa proteins, we identified a high molecular mass antigen that was highly and persistently recognized by serum antibodies, thereby justifying future efforts to accomplish its structural and biological characterization. Acknowledgements
[9]
[10]
[11]
[12]
[13] [14]
We are grateful to Mrs. Sandra M.O. Thomaz, Mrs. Maria Helena Borges Malta and Mrs. Margarida Maria Passeri for technical assistance and Mr. André Luiz Dardes for critical reading of the manuscript. We thank Dr. Constance Oliver and Dr. Maria Célia Jamur for helpful discussions. Research supported by FAPESP (00/09333-2), CNPq (350418/00-4) and CAPES grants.
[15]
References
[18]
[1] [2] [3]
[4] [5]
[6]
[7]
[8]
E. Brummer, E. Castaneda, A. Restrepo, Paracoccidioidomycosis: an update, Clin. Microbiol. Rev. 6 (1993) 89–117. M. Franco, Host–parasite relationships in paracoccidioidomycosis, J. Med. Vet. Mycol. 25 (1987) 5–18. G. Del-Negro, C.S. Lacaz, V.A. Zamith, A.M. Siqueira, General clinical aspects: polar forms of paracoccidioidomycosis, the disease in childhood, in: M. Franco, C.S. Lacaz, A. Restrepo-Moreno, G. DelNegro (Eds.), Paracoccidioidomycosis, CRC Press, Boca Raton, 1994, pp. 225–232. A. Restrepo, The ecology of Paracoccidioides brasiliensis: a puzzle still unsolved, Sabouraudia 23 (1985) 323–334. M.H. Blotta, Z.P. Camargo, Immunological response to cell-free antigens of Paracoccidioides brasiliensis: relationship with clinical forms of paracoccidioidomycosis, J. Clin. Microbiol. 31 (1993) 671–676. Z.P. Camargo, J.L. Gesztesi, E.C. Saraiva, C.P. Taborda, A.P. Vicentini, J.D. Lopes, Monoclonal antibody capture enzyme immunoassay for detection of Paracoccidioides brasiliensis antibodies in paracoccidioidomycosis, J. Clin. Microbiol. 32 (1994) 2377–2381. M.J.S. Mendes-Giannini, G. Del-Negro, A.M. Siqueira, Serodiagnosis, in: M. Franco, C.S. Lacaz, A. Restrepo-Moreno, G. Del-Negro (Eds.), Paracoccidioidomycosis, CRC Press, Boca Raton, 1994, pp. 345–357. C.P. Taborda, Z.P. Camargo, Diagnosis of paracoccidioidomycosis by dot immunobinding assay for antibody detection using the purified and specific antigen gp43, J. Clin. Microbiol. 32 (1994) 554–556.
[16]
[17]
[19]
[20]
[21]
[22]
[23]
1211
Z.P. Camargo, C. Unterkircher, L.R. Travassos, Identification of antigenic polypeptides of Paracoccidioides brasiliensis by immunoblotting, J. Med. Vet. Mycol 27 (1989) 407–412. P.S. Cisalpino, R. Puccia, L.M. Yamauchi, M.I. Cano, J.F. da Silveira, L.R. Travassos, Cloning, characterization, and epitope expression of the major diagnostic antigen of Paracoccidioides brasiliensis, J. Biol. Chem. 271 (1996) 4553–4560. M.R.E. Costa, C.S. Lacaz, M. Kawasaki, Z.P. Camargo, Conventional versus molecular diagnostic tests, Med. Mycol. 38 (Suppl. 1) (2000) 139–145. R. Puccia, S. Schenkman, P.A. Gorin, L.R. Travassos, Exocellular components of Paracoccidioides brasiliensis: identification of a specific antigen, Infect. Immun. 53 (1986) 199–206. U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature 227 (1970) 680–685. H. Blum, H. Beier, H.J. Gross, Improved silver staining of plant proteins, RNA and DNA in polyacrylamide gels, Electrophoresis 8 (1987) 93–99. C.M. Tsai, C.E. Frasch, A sensitive silver stain for detecting lipopolysaccharides in polyacrylamide gels, Anal. Biochem. 119 (1982) 115–119. M. Casotto, S. Paris, Z.P. Camargo, Antigens of diagnostic value in three isolates of Paracoccidioides brasiliensis, J. Med. Vet. Mycol. 29 (1991) 243–253. L.R. Travassos, Immunochemistry of Paracoccidioides brasiliensis antigens, in: M. Franco, C.S. Lacaz, A. Restrepo-Moreno, G. DelNegro (Eds.), Paracoccidioidomycosis, CRC Press, Boca Raton, 1994, pp. 67–86. R. Martins, S. Marques, M. Alves, D. Fecchio, M.F. Franco, Serological follow-up of patients with paracoccidioidomycosis treated with itraconazole using Dot-blot, ELISA and western-blot, Rev. Inst. Med. Trop. Sao Paulo 39 (1997) 261–269. R. Puccia, L.R. Travassos, Forty-three kilodalton glycoprotein from Paracoccidioides brasiliensis: immunochemical reactions with sera from patients with paracoccidioidomycosis, histoplasmosis, or Jorge Lobo’s disease, J. Clin. Microbiol. 29 (1991) 1610–1615. G.M. Del-Negro, C.N. Pereira, H.F. Andrade, S.A. Palacios, M.M. Vidal, C.E. Charbel, G. Benard, Evaluation of tests for antibody response in the follow-up of patients with acute and chronic forms of paracoccidioidomycosis, J. Med. Microbiol. 49 (2000) 37–46. J.P. Bueno, M.J.S. Mendes-Giannini, G.M. Del-Negro, G.M. Assis, C.K. Takiguti, M.A. Shikanai-Yasuda, IgG, IgM and IgA antibody response for the diagnosis and follow-up of paracoccidioidomycosis: comparison of counterimmunoelectrophoresis and complement fixation, J. Med. Vet. Mycol. 35 (1997) 213–217. M.J.S. Mendes-Giannini, J.P. Bueno, M.A. Shikanai-Yasuda, A.M. Stolf, A. Masuda, V. Amato-Neto, A.W. Ferreira, Antibody response to the 43 kDa glycoprotein of Paracoccidioides brasiliensis as a marker for the evaluation of patients under treatment, Am. J. Trop. Med. Hyg. 43 (1990) 200–206. M.F. Ferreira-da-Cruz, A.C.F. Vale, M.C. Espinera, B. Wanke, B.G. Castro, Study of antibodies in paracoccidioidomycosis: follow-up of patients during and after treatment, J. Med. Vet. Mycol. 28 (1990) 151–157.