Skip to main page content

• HOME
• Current Issue
• Archive
• Alerts
• About ASM
• Contact us
• Tech Support
• Journals.ASM.org

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Carvalho, K. C. Articles by Puccia, R. Search for Related Content PubMed PubMed Citation Articles by Carvalho, K. C. Articles by Puccia, R.

 Previous Article  |  Next Article 

Clinical and Vaccine Immunology, April 2008, p. 622-629, Vol. 15, No. 4
1071-412X/08/$08.00+0     doi:10.1128/CVI.00437-07
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

Use of Recombinant gp43 Isoforms Expressed in Pichia pastoris for Diagnosis of Paracoccidioidomycosis

K. C. Carvalho ,^, M. C. Vallejo, Z. P. Camargo, and R. Puccia^*

Department of Microbiology, Immunology and Parasitology, Federal University of São Paulo, Rua Botucatu 862, oitavo andar, São Paulo, SP 04023-062, Brazil

Received 30 October 2007/ Returned for modification 5 December 2007/ Accepted 22 January 2008

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
gp43 is the main diagnostic antigen for paracoccidioidomycosis (PCM). In vitro, gp43 expression in supernatant fluids of Paracoccidioides brasiliensis cultures can be unstable, and its regulation is poorly understood. We have been able to express soluble recombinant gp43 (gp43r) isoforms as N-mannosylated proteins secreted in the supernatants of Pichia pastoris cultures induced with methanol. They were secreted as major components from day 2 of induction and could be purified with affinity columns containing anti-gp43 monoclonal antibodies. We have expressed P. brasiliensis GP43 (PbGP43) sequences from genotypes A, D, and E, and the correspondent gp43r isoforms (gp43r A, -B, and -C, respectively; 200 ng) were compared to native gp43 in immunodiffusion (ID) and dot blot assays. Among 90 PCM patient sera showing ID-positive reactions with purified native gp43, 100% were positive with gp43rD and gp43rE and 98% reacted with gp43rA. Of these sera, 78 were tested in dot blot assays at a 1:1,000 dilution, and 100% reacted with all recombinant isoforms. In ID assays, the specificity was 100%, since 40 sera from patients with related mycoses and 30 sera from healthy individuals did not react with any of the antigens. In dot blot assays, 100% specificity for PCM occurred when cross-reactive mannose epitopes were neutralized with 10 mM metaperiodate or eliminated through deglycosylation. However, a 1:1,000 serum dilution was already discriminatory for most sera. We suggest that P. pastoris recombinant gp43, especially isoforms D and E, may replace the native antigen in ID and dot blot assays for diagnosis and prognosis of PCM. Regulated expression of large amounts of antigen in nonpathogenic yeast would justify its preferred use.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
gp43 is a secreted glycoprotein from Paracoccidioides brasiliensis. It is the best-studied fungal molecule and the main antigen for diagnosis and prognosis of paracoccidioidomycosis (PCM) described so far (7, 20, 25, 37). PCM is a granulomatous systemic mycosis that occurs as active disease in 1 to 2% of infected people, whose number is estimated to be 10 million throughout areas of endemicity in Latin America (32). Adult PCM is the commonest form and gradually affects the lungs, mainly of male adults; it can be unifocal or disseminate to any organ, generally involving mucous membranes and skin. Acute and subacute PCM progress rapidly and spread through the lymphatic system; they are characteristic of children and young adults of both sexes (14). Histoplasmosis, blastomycosis, and coccidioidomycosis are related endemic mycoses caused by dimorphic ascomycetes genetically close to P. brasiliensis (30). Cellular immunity is protective against these mycoses, whereas high antibody titers against fungal antigens suggest heavy fungal loads during severe stages of disease.

Apart from being a major P. brasiliensis antigen for the humoral response, gp43 also contains T-cell epitopes that elicit protective cellular immunity in experimentally infected animals (17, 36, 37) and displays adhesive capacity toward proteins associated with the extracellular matrix (15, 21). The open reading frame of the P. brasiliensis GP43 (PbGP43) gene lies within a 1,329-bp genomic fragment that has a single 78-bp intron (11). The full protein bears 416 amino acids, of which the first 35 correspond to a signal peptide. Mature or exocellular gp43 contains a single N-glycosylation site at sequon NRT, which is occupied by high-mannose chains composed, on average, of 13 or 14 nonphosphorylated mannose residues and one terminal β-galactofuranose residue (β-Galf) (3). The β-Galf residue is -1,6 linked to one of the -1,2-linked manopiranose residues added to the fixed Man₇GlcNAc₂ core. This chain can be removed enzymatically with endoglycosidase H (endo H) or N-glycosidase F, metabolically with tunicamycin, and chemically with trifluoromethanosulfonic acid, generating a faster-migrating single band in sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels (27). Exocellular gp43 is composed of a mixture of isoforms that have close but distinct isoelectric points (pI), as originally observed in P. brasiliensis isolate Pb339 (25). Moura-Campos et al. (23) described the existence of four gp43 pI profiles for eight fungal isolates, suggesting that differences can occur both within and among isolates. The antigen from one isolate was peculiarly basic (>8.5), while gp43 proteins from the others had pIs varying from 5.8 to 7.2.

Morais et al. (22) found a total of 21 informative substitution sites concentrated in exon 2 of the PbGP43 gene, defining genotypes A to E (29). So far, all six samples bearing genotype A have been grouped in a cryptic phylogenetic group, PS2, according to a multilocus study with 65 isolates conducted by Matute et al. (19). Genotype A sequences translate basic gp43 and are highly polymorphic (up to 15 substitution sites) compared with the others. Isolates from PS2 were less virulent than others in the B10.A mouse model of PCM (10).

Serological diagnosis and prognosis of PCM are important tools for clinicians. The most popular test is double immunodiffusion (ID) for its simplicity, high sensitivity, and specificity (9, 31). Whole P. brasiliensis extracellular antigen preparations are commonly used, where gp43 is the antigenic component responsible for ID specificity and positivity over 85% (5, 7, 9, 25). False-negative reactions have been found for patients with intense pulmonary infection and immune depression (5, 9). Purification of gp43 was facilitated after the production of monoclonal antibodies (15, 27), and the antigen, purified or not, has been tested in a variety of immunodiagnostic tests for detection of patient antibodies (4, 8, 26, 35, 37). Antigen detection in sera can also be used for diagnosis of PCM (18). Although gp43 is quite specific for PCM when presented to antibodies in solution (7, 26), in capture enzyme-linked immunosorbent assays (ELISAs) (8), and in dot blot assays (35), cross-reactions in ELISA can be frequent (1, 26), particularly with β-Galf residues, which are probably more exposed when the antigen is immobilized on plastic.

Regulation of gp43 expression is poorly understood (10) and varies with the isolate (34). However, instability in gp43 expression can occur even in good producers (e.g., strain Pb339). Therefore, standardized whole-antigen preparations are useful for diagnosis/prognosis (4, 7) but are not necessarily reproducible. On the other hand, differences in reactivity related to the gp43 isoforms have already been reported (34). For these reasons, the use of soluble recombinant gp43 (gp43r) would be a step forward not only in the diagnosis of PCM but also in other structural and biological studies.

In the present communication, we report the extracellular expression of soluble gp43r in the yeast Pichia pastoris. Purified gp43r was tested for reactivity with sera from patients with PCM and other mycoses. We chose ID and dot blot assays to compare the reactions with those of purified gp43 from Pb339 because they are simple and fairly specific tests according to the literature (7, 35). Dot blotting is a sensitive enzymatic assay (35) and can eventually be quantified (2). We managed to express gp43 isoforms which are characteristic of Pb339 from genotypes A, D, and E (gp43A, -B, and -C, respectively) but not C (22, 29). The expressed isoform from genotype A has a calculated pI of 8.3, while the translated sequences from genotypes D and E have pIs of 6.8 and 7.1, respectively (22). Our results suggest that gp43r produced in P. pastoris may be used in the diagnosis of PCM, especially those isoforms from genotypes D and E.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
Synthesis of PbGP43 cDNA and expression in P. pastoris. The PbGP43 coding sequences corresponding to genotypes A, D, and E (29) were reverse transcribed from DNA-free total RNA isolated from P. brasiliensis Pb3 (PbGP43 genotype A), Pb10 (PbGP43 genotype D), and Pb14 (PbGP43 genotype E) (22). Total RNA was extracted from logarithmically growing yeast cells cultivated at 36°C, with shaking, in 0.5% yeast extract, 0.5% peptone-casein, and 1.5% glucose. The cell pellet was mechanically disrupted by being vortexed with glass beads (425 to 600 µm; Sigma) in the presence of TRIzol reagent (Invitrogen), and total RNA was isolated according to the instructions of the manufacturer. cDNA of the PbGP43 gene was obtained by reverse transcription-PCR (RT-PCR), using the ThermoScript RT-PCR system (Gibco) and the specific 3' antisense primer 491 (5'-ACGTCGACTCACCTGCATCCACCATACTT-3'), which has a SalI site immediately after the TGA stop codon. The second strand of the full open reading frame was elongated, using standard PCR conditions, with the 5' sense primer 490 (5'-GTCAGATCTATCATGAATTTTAGTTCCTTAAC-3'), containing a BglII site before the ATG start codon, and with antisense primer 491. The 490/491 PCR fragment (1,250 bp) was extracted from agarose gels by using a Sephaglass kit (Amersham Pharmacia), cloned into the pGEM-T Easy vector (Promega), and sequenced. cDNA corresponding to processed (secreted) gp43 was synthesized by PCR, using the pGEM-T clone as the template; the upstream primer 690 (5'-CAGTCGACAAGCAGGATCAGCAATATAT-3'), containing a SalI site before the alanine codon GCA; and the downstream primer 691 (5'-GCGGTACCTCACCTGCATCCACCATA-3'), where a KpnI site lies immediately after the TGA stop codon. The 690/691 fragment (1,145 bp) was purified, cloned into the pGEM-T Easy vector, and subcloned into the SalI/KpnI sites of pHIS1 (33), which we previously used to express gp43(His₆) in bacteria (not published). Although the purified recombinant gp43(His₆) isoforms expressed in bacteria have not been of further use because of solubilization problems, cloning into pHIS1 was essential for subcloning the 690/691 insert into the EcoRI/NotI sites of a pPIC9 plasmid (Invitrogen) for expression in P. pastoris. The correct frame at the 5' end was verified by sequencing recombinant pPIC9, which was subsequently used to transform P. pastoris G115 according to the manufacturer's instructions.

A total of 10 recombinant clones from each PbGP43 genotype were randomly selected for induction of gp43r expression with methanol. For that purpose, individual colonies were grown for 2 days, with shaking, at 30°C in BMGY (buffered minimal glycerol complex medium), composed of 1% yeast extract, 2% peptone, 0.1 M potassium phosphate, pH 6.0, 1.34% YNB (Difco), 1% ammonium sulfate, and 1% glycerol. Cell pellets were suspended in one-half the initial volume of BMMY (BMGY with 1% methanol instead of glycerol) and incubated at 30°C with intense shaking for 6 days. Methanol (1%) was added daily, and aliquots of the culture were collected every 2 days to monitor secretion of gp43r.

Screening of positive P. pastoris clones and purification of gp43r. We used dot blotting to screen for P. pastoris gp43r-producing clones upon induction with methanol. Cell-free supernatants (5 µl) were dot blotted onto nitrocellulose membranes and tested for reactivity with rabbit polyclonal anti-gp43 serum (1:5,000), as described below. Purified native gp43 (gp43n) was used as a positive control, and the supernatant from an induced P. pastoris culture expressing an irrelevant protein was used as a negative control. Expression of gp43r by positive clones was then verified in SDS-polyacrylamide-stained gels and Western blots. We chose the best-producing clone of each isoform to induce larger cultures of P. pastoris for 2 days for purification purposes. To purify gp43n and gp43r, we used affinity columns of Affi-Gel 10 bound to MAb17c, which is an anti-gp43 monoclonal antibody that recognizes all gp43 isoforms (8; Rosana Puccia and Kátia C. Carvalho, international patent application PCT/BR2007/000258, Fundação de Amparo à Pesquisa, 27 September 2007; Rosana Puccia and Kátia C. Carvalho, Brazilian patent application priority PI 0604717-3, Brazilian Patent Office, 29 September 2006).

ID assay. ID assays were performed on microscope slides covered with a layer of 1% agarose gel diluted in 0.85% NaCl, as described previously (25). Antigens (200 ng) and sera were tested in a volume of 10 µl per well. Reactions were incubated overnight at room temperature in a moist chamber; the slides were then washed once in 5% sodium citrate and five times in 0.85% NaCl. The last wash was extended overnight to minimize background staining. Agarose was dehydrated over the slides in an oven, and the reactions were stained with Coomassie brilliant blue.

Dot blotting. In order to test the usefulness of gp43r in dot blots, antigens (200 ng) were spotted onto nitrocellulose membranes (1.5 to 2 µl), while for screening purposes, 5 µl of P. pastoris culture supernatant was used. The membranes were left to dry in the air for 5 min and then quenched with sodium phosphate-buffered saline (PBS) containing 5% skim milk (Nestlé) (PBS-M) overnight at 4°C. The membranes were washed three times (10 min each) with PBS-T (PBS containing 0.1% Tween 20) and incubated with rabbit polyclonal anti-gp43 serum (1:5,000) to screen for positive P. pastoris clones or with patient sera (1:300 or 1:1,000). Sera were diluted in PBS-M and incubated with shaking for 1 h at room temperature with the membranes, which were then washed three times in PBS-T and incubated with secondary antibodies (peroxidase-labeled goat anti-human immunoglobulin G [IgG] or anti-rabbit immunoglobulin, used at 1:2,000; Amersham Biosciences) for 1 h at room temperature, with shaking. The membranes were washed three times, and the reactions were developed with diaminobenzidine (Sigma). Preimmune serum or sera from healthy individuals were used as negative controls.

SDS-PAGE and Western blotting. Protein/glycoprotein profiles were visualized upon silver impregnation and/or Coomassie brilliant blue staining after electrophoresis in SDS-PAGE gels under reducing conditions (16). For Western immunoblotting, the samples were transferred from SDS-PAGE gels to nitrocellulose membranes, and the reactions were performed with patient sera as described previously (13). Immunocomplexes were evidenced with goat anti-human IgG-peroxidase conjugates (Amersham Biosciences) and developed with diaminobenzidine (Sigma).

Sera. Anti-gp43 rabbit immune serum and anti-gp43 monoclonal antibodies were previously obtained by Puccia et al. (27, 28). Patient sera and sera from healthy blood donors were obtained from São Paulo Hospital. Some PCM patient serum samples were kindly provided by M. H. S. L. Blotta, Unicamp, São Paulo, Brazil. We tested a total of 90 PCM serum samples from 78 diagnosed patients (with either the juvenile or adult form), most of whom were undergoing treatment. Samples from the same PCM patients were collected at different times of treatment. According to the records, original ID titers for PCM patients were between 1:1 and 1:4, and a few were between 1:8 and 1:64 when sera were tested against whole exocellular P. brasiliensis antigens from Pb339 (7).

Protein content. The gp43 protein content was estimated using (i) a modified version of the Bradford methodology (6), (ii) optical density readings (1.0 A₂₈₀ unit = 0.45 mg/ml), and (iii) visualization in Coomassie brilliant blue-stained SDS-PAGE gels.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
We expressed three gp43 isoforms (gp43rA, gp43rD, and gp43rE) as soluble glycoproteins in the culture supernatants of selected recombinant P. pastoris clones. They were expressed as major components migrating at about 45 kDa in SDS-PAGE gels from the second day after induction with methanol (Fig. 1A). The peak of expression generally occurred between days 4 and 6 of culture. The major component was identified as gp43r in Western blots incubated with polyclonal anti-gp43 rabbit antibodies (not shown).

View larger version (34K): [in this window] [in a new window]

FIG. 1. (A) SDS-PAGE profiles of silver-stained total supernatant fluids from recombinant P. pastoris cultures (gp43rD) induced with methanol for 2 to 6 days. gp43rD is indicated by an arrow. C, supernatant from P. pastoris, containing vector alone and induced for 2 days. (B) Purified gp43r isoforms A, D, and E before (–) and after (+) treatment with endo H (EH). gp43n is shown for comparison. The gel was stained with Coomassie brilliant blue stain. Molecular masses corresponding to standard protein bands are shown in kDa. The faster-migrating band seen in endo H-treated gp43rD and gp43rE corresponds to the enzyme.

The amount of gp43r expressed by P. pastoris varied with the isoform. Upon induction of medium-scale yeast cultures with methanol for 2 days, we obtained about 7 mg/liter of purified gp43rD and gp43rE and over 100 mg/liter of purified gp43rA. We chose to work with cultures induced for 2 days to minimize the presence of P. pastoris molecules. However, in order to increase the yield of gp43rD and -E, we might process longer-induced cultures in future purifications. Purification was achieved through affinity chromatography with columns containing anti-gp43 monoclonal antibody (MAb17c), and the purified recombinant isoforms can be seen in Fig. 1B in comparison with gp43n. In Fig. 1B, a smear is particularly visible for gp43rD before treatment, suggesting hyper- and heterogeneous glycosylation by P. pastoris. The smear is more evident in overloaded silver-stained gels (not shown) and can be seen in the Western blot in Fig. 3C. After enzymatic deglycosylation with endo H, all recombinant isoforms migrated slightly slower than gp43n, probably due to the addition of 10 to 14 vector amino acids at the N terminus. Although we have not sequenced this end of the gp43r isoforms, the constructions predict the addition of a YVEFKGLRRQ peptide sequence from pHIS1 and, possibly, the repeat EAEA from pPIC9 before that. This repeat is at the C terminus of the -mating-factor signal peptide and is left in the recombinant proteins produced in P. pastoris after Kex2 cleavage. Peptide EAEA is supposedly cleaved by a dipeptidyl aminopeptidase, Ste13, but cleavage may not occur when the protein is overexpressed (12).

View larger version (61K): [in this window] [in a new window]

FIG. 3. Representative dot blot reactions of PCM patient (1:1,000) (A) and heterologous-protein (1:300 or 1:1,000) (B) serum samples with gp43n and gp43r isoforms A, D, and E (200 ng per dot) before (ct) and after (per) treatment with 10 mM metaperiodate. Some sera were also tested against endo H-deglycosylated antigens (EH). (C) Immunoblotting profiles of gp43rD, treated (per) or not (ct) with 10 mM sodium metaperiodate, as tested with a control PCM patient serum and two histoplasmosis antigen-reactive sera (Hc1 and Hc2), at a 1:1,000 dilution.

We initially verified the reactivity of each gp43 isoform by double ID assay, which is the test used most often for diagnosis and prognosis of PCM because of its simplicity, sensitivity, and specificity. We started by testing a small sample of PCM patient sera that reacted with gp43n and obtained positive results using 2 µg of all the isoforms. We next titrated the antigen (not shown) in order to use the smallest possible amount to guarantee positive results with pure sera. Based on the results, we chose 200 ng of antigen to screen a larger serum sample. We assayed a total of 90 sera from 78 different patients (Table 1). The samples were tested with gp43rA, gp43rD, and gp43rE in comparison with gp43n. In Fig. 2A, we exemplify these reactions by showing the results for 18 PCM sera. The positive reactions between sera g, h, p, and o and gp43rD were considered weak, while that between gp43rD and serum n was negative (but positive in a second test). Among the negative results, only two were later confirmed, specifically those with isoform gp43rA (Table 1). However, these two sera produced weak reaction bands with the other antigens. The intensities of the precipitation lines were generally similar among isoforms and gp43n, but differences eventually occurred, as represented by those formed between serum a or g and gp43rE. In parallel, we tested sera from patients with histoplasmosis (n = 15), aspergillosis (n = 15), candidiasis (n = 5), and Jorge Lobo's disease (n = 5), in addition to 30 sera from healthy donors. The antibodies in none of these sera produced ID precipitation bands with gp43r or gp43n.

View this table: [in this window] [in a new window]

TABLE 1. Positive reactions between recombinant gp43r isoforms A, D, and E and serum samples from patients with PCM or other mycosesa

View larger version (77K): [in this window] [in a new window]

FIG. 2. Representative ID reactions with recombinant isoforms gp43rA, gp43rD, and gp43rE and control gp43n (200 ng; central well). (A) Undiluted PCM patient serum samples (a to s; 90 were tested) reactive with gp43n. Overall results are shown in Table 1. (B) Titrated PCM patient serum samples (from 1:1 [well 1] in twofold steps to a 1:32 dilution [well 6]). We show the reactions for PCM patient sera t, u, and v (out of eight sera tested).

In order to test the prognostic value of gp43r in ID assays, we titrated (1:1 to 1:32) eight serum samples. As represented by three sera in Fig. 2B, the results among gp43n and gp43r were generally similar, although some differences in band intensity could eventually be observed. Overall, our results showed that for PCM diagnosis using ID, the positivity of gp43rD and gp43rE was 100% comparable to that of gp43n, while for gp43rA it reached 98% (Table 1). The specificity was 100% with all isoforms. We concluded that gp43r produced in P. pastoris has ID diagnostic and prognostic value for PCM. A mixture of isoforms is not necessary, and isoform gp43rE apparently produced results most similar to those obtained with gp43n.

We then tested the usefulness of gp43r in dot blotting. Among ID-positive PCM patient serum samples, we assayed 78 with gp43rA, gp43rD, gp43rE, and gp43n (200 ng), treated or not with 10 mM sodium periodate. A total of 33 sera were also tested with the corresponding endo H-treated antigens. At a 1:1,000 serum dilution, the reactions were generally similar among all antigens, treated or not, as shown in the representative dot blots of Fig. 3A. About 13 samples presented weaker dots, as exemplified by sera PCM5 and PCM6. Reaction with PCM5 was the weakest among PCM sera at 1:1,000. Sera from patients with other mycoses were tested at 1:300, and at this dilution about 68% were positive with all the gp43r isoforms, while 20% were positive with gp43n (Table 1). At 1:1,000, cross-reactivity with gp43n was abolished. Dilution to 1:1,000 was also able to eliminate most of the cross-reactivity with gp43r: among 13 positive sera at 1:300, only 3 had dots as intense as or more intense than those for PCM5 (e.g., in Fig. 3B, those for candidiasis and aspergillosis), while 6 dots were completely abolished and 4 were extremely weak (e.g., in Fig. 3B, those for Jorge Lobo's disease). False-positive reactions with sera at a 1:300 or 1:1,000 dilution were totally abolished upon antigen treatment with 10 mM metaperiodate (Fig. 3B), indicating that they were due to cross-reactive carbohydrate epitopes. We also assayed 30 sera from healthy individuals at 1:300, and 26 did not react with any antigen tested, while in four cases there were extremely weak dots visible with gp43r (similar to those in Fig. 3B at 1:1,000 for Jorge Lobo's disease), which disappeared after metaperiodate treatment. The nature of the above-mentioned cross-reactions is exploited in Fig. 3C. Control PCM patient serum reacted intensely with gp43n and gp43r, and treatment with metaperiodate did not qualitatively affect this reactivity. When two histoplasmosis sera were tested under the same conditions, there was no reactivity with treated antigens. Overmannosylation is highly suggested by the smear seen for gp43rD. Our results suggest that gp43r isoforms from P. pastoris are as successful as gp43n in dot blot diagnosis of PCM; however, 100% specificity can be achieved only upon neutralization of carbohydrate epitopes with 10 mM metaperiodate or by deglycosylation, while a test dilution at 1:1,000 eliminated most of the cross-reactions for the samples tested here.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 
We have been able to express soluble gp43r isoforms as glycosylated proteins secreted in the supernatants of P. pastoris cultures induced with methanol. The glycoproteins were secreted as major components from day 2 of induction and could be purified in affinity columns containing anti-gp43 monoclonal antibodies. We expressed PbGP43 sequences from genotypes A, D, and E (29) and compared their performances with that of gp43n in immunodiagnostic tests. The correspondent gp43 isoforms were useful in both ID and dot blot assays to detect 100% of PCM sera that were reactive with the native antigen. With gp43r, we obtained total specificity for PCM patient sera in ID assays, while in dot blots, total specificity for PCM patient sera could be achieved when carbohydrate epitopes were previously neutralized by treatment with 10 mM metaperiodate or by enzymatic deglycosylation with endo H. In most cases, however, a serum dilution of 1:1,000 was enough to discriminate between PCM and heterologous sera.

We have previously made several efforts to express gp43r in soluble form for use not only as an antigen in immunodiagnostic tests but also as a tool in diverse biological and structural analyses. Our attempts included expression in both bacteria and, unsuccessfully, Saccharomyces cerevisiae. For bacteria, we reported the production of insoluble whole and truncated gp43 fused to glutathione S-transferase (13). In immunoblotting, whole gp43-glutathione S-transferase reacted specifically with 27 PCM patient sera, while reactivity with N-terminal, middle, and C-terminal fragments varied. Other vectors used for IPTG (isopropyl-β-D-thiogalactopyranoside)-induced expression of gp43 in bacteria included pET23-a (Novagen) and pHIS1 (33), which resulted in production of insoluble His-tagged proteins of limited antigenic capacity after solubilization with 8 M urea and renaturation using standard protocols (not published).

For production of gp43 in P. pastoris, we subcloned PbGP43 genotypes A, D, and E from pHIS1 into both pPIC9K and pPIC9 shuttle vectors, but we managed to express only the gene cloned in pPIC9. We also tried to express gp43 isoform C from Pb339, but so far this has been unsuccessful. Vectors pPIC9 and pPIC9K differ basically in the resistance marker for kanamycin, which is absent in pPIC9. The selection of transformed bacteria relies on ampicillin resistance, and screening for recombinant yeasts on histidine-deprived selection media depends on vector-carried HIS4, coding for histidinol dehydrogenase. The PbGP43 cDNA was cloned into pPIC9 between the promoter and stop sequences from AOX1, which codes for alternative oxidase, and in frame with the -factor leader sequence for secretion of large amounts of heterologous proteins upon methanol induction. P. pastoris can use methanol as a sole carbon source. We expressed gp43 in strain GS115 (his4) transformed with recombinant pPIC9 restricted with SacI at the 5' AOX1 promoter. Therefore, we obtained a His⁺ Mut⁺ phenotype, where the plasmid is integrated in tandem into the yeast genome upon homologous recombination at the promoter site. In that situation, the AOX1 gene is not replaced, hence preserving high methanol utilization (12). We have not estimated the integration copy number for our recombinant strains. Nevertheless, we obviously selected for the highest gp43 producer of each isoform based on the intensities of the reactions with anti-gp43 polyclonal rabbit serum during the screening process of culture supernatants. Under our induction conditions, gp43rA was secreted about 15 times more abundantly than gp43rD or gp43rE. This result suggests that PbGP43(rA) was integrated into the P. pastoris genome at higher copy numbers.

gp43r is glycosylated in P. pastoris. This yeast species is able to add both N- and O-linked oligosaccharide chains to proteins (12). The carbohydrate chains are constituted solely of mannose residues, and phosphorylation can occur in -1,2- or -1,6-linked mannose residues. Differently from S. cerevisiae, terminal -1,3-linked mannose is apparently not found in P. pastoris due to a lack of the appropriate transferase. The oligosaccharide chains added to recombinant proteins in P. pastoris are more commonly of the classical short high-mannose type (Man_8-14GlcNAc); however, there are examples of hyperglycosylation (12). gp43r produced in this yeast migrates roughly at a rate similar to that of gp43n in SDS-PAGE gels, but a slower-migrating long smear can also be seen. This profile suggests that the recombinant protein has the gp43 NRT sequon occupied mainly by short high-mannose chains, while part has been overmannosylated with longer and heterogeneous chains, all of which are susceptible to endo H. The presence of O-linked sugars or phosphorylated mannose residues has not been evidenced or investigated.

Differences in oligosaccharide composition or the presence of vector amino acids in the N termini of gp43r isoforms did not change their reactivity with PCM sera in ID assays compared with gp43n reactivity. This was demonstrated by the similar profiles generally seen in precipitation bands when sera were tested either undiluted or titrated. On the other hand, ID false-positive reactions were not observed either. Where false-negative reactivity is concerned, isoform gp43rA, which is the most polymorphic and renders basic gp43, was not recognized by two sera reactive with gp43n and recombinant isoforms D and E. Therefore, in future surveys, we will concentrate on the use of isoform D or E. In a capture ELISA test using anti-gp43 MAb17c as the capture antibody, basic gp43 isoforms (genotype A) isolated from Pb1925 (Pb2) recognized fewer PCM patient sera from both adult (71%) and juvenile (56%) patients than did purified gp43 from Pb339 (8). However, the performances showed by both antigens were similar when MAb8a was used. These data suggest that the epitope recognized by MAb17c is shared by different gp43 isoforms and is hence of special relevance in diagnosis.

It is noteworthy that in the present study we have not included PCM sera that can be falsely negative in ID assays. We worked exclusively with a random sample of PCM sera that reacted with purified gp43 in this test. Neves et al. (24) investigated the nature of 28 proven PCM cases, all of the unifocal pulmonary form, whose sera rendered false-negative ID results with whole secreted antigens from Pb339. They demonstrated by ELISA that in these patients the anti-gp43 antibodies were preferentially of the IgG2a isotype, with low avidity for carbohydrate epitopes. It is worth mentioning that in ELISA, gp43 carbohydrate epitopes are particularly accessible to antibody recognition, prompting unwanted levels of cross-reactivity (1, 26) that are not characteristic of immunoprecipitation tests (7), capture ELISA (8), or dot blots (35). PCM sera bearing antibodies preferentially directed to gp43 carbohydrate epitopes are exceptions and cannot guarantee serological diagnosis of PCM. It would be interesting to test false-negative sera in dot blot, quantitative dot blot (2), and ID assays with both native and recombinant purified gp43 under conditions that could favor recognition of peptide-specific epitopes, even at low titers.

Although overmannosylation of gp43r did not interfere with the specificity of ID results for PCM, the specificity of dot blot reactions was compromised: false-positive reactions were detected in dot blots performed at 1:300 and, at a smaller percentage, at a 1:1,000 serum dilution; however, the reactivity was completely abolished when the antigens were previously treated with metaperiodate or endo H. On the other hand, PCM patient sera reacted equally well at 1:1,000 with gp43n and gp43r. None became negative against treated antigens, corroborating our previous observation that anti-gp43 antibodies from PCM patients are preferentially directed to peptide epitopes (26). The rate of false-negative reactions with gp43n in dot blots was 20% in this work and only 4.3% when Taborda and Camargo (35) standardized the test for diagnosis and prognosis of PCM. They tested 64 sera from non-PCM mycosis patients and 50 sera from healthy individuals; treatment with 10 mM metaperiodate blocked cross-reactions. In the present work, higher percentages of cross-reactivity with gp43r in dot blots were expected due to the ubiquitous nature of mannose epitopes among pathogenic fungi. In gp43n, the main source of cross-reactivity seems to be a unique terminal β-Galf residue, at least when ELISA is considered (1, 26).

Overall, our results suggest that gp43r from P. pastoris may replace gp43n in dot blots and ID assays for detection of PCM; although a serum dilution of 1:1,000 eliminated most dot blot cross-reactions, neutralization of carbohydrate epitopes is recommended to avoid false-positive results with this test. We tested three isoforms, among which gp43rD or gp43rE can apparently be used without the need for a mixture of isoforms. The isoform corresponding to Pb339 (genotype C), which is largely used for antigen production, has not been expressed successfully in yeast so far. The advantage of gp43r from P. pastoris relies mainly in reproducibility for the production of large amounts of a known sequence of gp43, which is expressed in culture supernatants under inducible conditions in nonpathogenic, fast-growing yeast. Although posttranslational modifications in yeast can be advantageous, we are working on the expression of deglycosylated gp43r, using a point mutation at the glycosylation site. That might render a more specific antigen, but conformational problems could lead to decreased sensitivity.

 
This work was supported by FAPESP, CNPq, and CAPES.

We thank Luiz S. Silva for technical assistance.

 
* Corresponding author. Mailing address: Disciplina de Biologia Celular, UNIFESP, Rua Botucatu 862, oitavo andar, São Paulo, SP 04023-062, Brazil. Phone: 55-11-5084-2991. Fax: 55-11-5571-5877. E-mail: rpuccia{at}unifesp.br

Published ahead of print on 30 January 2008.

K. C. Carvalho and M. C. Vallejo contributed equally to this work.

Present address: Ludwig Institute for Cancer Research, São Paulo, Brazil.

Top ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

 

1
1. Albuquerque, C. F., S. H. da Silva, and Z. P. Camargo. 2005. Improvement of the specificity of an enzyme-linked immunosorbent assay for diagnosis of paracoccidioidomycosis. J. Clin. Microbiol. 43:1944-1946.[Abstract/Free Full Text]
2
2. Almeida, I. C., E. G. Rodrigues, and L. R. Travassos. 1994. Chemiluminescent immunoassays: discrimination between the reactivities of natural and human patient antibodies with antigens from eukaryotic pathogens, Trypanosoma cruzi and Paracoccidioides brasiliensis. J. Clin. Lab. Anal. 8:424-431.[Medline]
3
3. Almeida, I. C., D. C. Neville, A. Mehlert, A. Treumann, M. A. Ferguson, J. O. Previato, and L. R. Travassos. 1996. Structure of the N-linked oligosaccharide of the main diagnostic antigen of the pathogenic fungus Paracoccidioides brasiliensis. Glycobiology 6:507-515.[Abstract/Free Full Text]
4
4. Blotta, M. H., and Z. P. Camargo. 1993. Immunological response to cell-free antigens of Paracoccidioides brasiliensis: relationship with clinical forms of paracoccidioidomycosis. J. Clin. Microbiol. 31:671-676.[Abstract/Free Full Text]
5
5. Blotta, M. H., R. L. Mamoni, S. J. Oliveira, S. A. Nouer, P. M. Papaiordanou, A. Goveia, and Z. P. Camargo. 1999. Endemic regions of paracoccidioidomycosis in Brazil: a clinical and epidemiologic study of 584 cases in the southeast region. Am. J. Trop. Med. Hyg. 61:390-394.[Abstract]
6
6. Bradford, M. M. 1976. Rapid and sensitive method for quantitation of microgram quantities of protein utilizing principle of protein-dye binding. Anal. Biochem. 72:248-254.[CrossRef][Medline]
7
7. Camargo, Z., C. Unterkircher, S. P. Campoy, and L. R. Travassos. 1988. Production of Paracoccidioides brasiliensis exoantigens for immunodiffusion tests. J. Clin. Microbiol. 26:2147-2151.[Abstract/Free Full Text]
8
8. Camargo, Z. P., J. L. Gesztesi, E. C. Saraiva, C. P. Taborda, A. P. Vicentini, and J. D. Lopes. 1994. Monoclonal antibody capture enzyme immunoassay for detection of Paracoccidioides brasiliensis antibodies in paracoccidioidomycosis. J. Clin. Microbiol. 32:2377-2381.[Abstract/Free Full Text]
9
9. Camargo, Z. P., and M. F. Franco. 2000. Current knowledge on pathogenesis and immunodiagnosis of paracoccidioidomycosis. Rev. Iberoam. Micol. 17:41-48.[Medline]
10
10. Carvalho, K. C., L. Ganiko, W. L. Batista, F. V. Morais, E. R. Marques, G. H. Goldman, M. F. Franco, and R. Puccia. 2005. Virulence of Paracoccidioides brasiliensis and gp43 expression in isolates bearing known PbGP43 genotype. Microbes Infect. 7:55-65.[CrossRef][Medline]
11
11. Cisalpino, P. S., R. Puccia, L. M. Yamauchi, M. I. Cano, J. F. da Silveira, and L. R. Travassos. 1996. Cloning, characterization, and epitope expression of the major diagnostic antigen of Paracoccidioides brasiliensis. J. Biol. Chem. 271:4553-4560.[Abstract/Free Full Text]
12
12. Daly, R., and M. T. Hearn. 2005. Expression of heterologous proteins in Pichia pastoris: a useful experimental tool in protein engineering and production. J. Mol. Recognit. 18:119-138.[CrossRef][Medline]
13
13. Diniz, S. N., K. C. Carvalho, P. S. Cisalpino, J. F. Silveira, L. R. Travassos, and R. Puccia. 2002. Expression in bacteria of the gene encoding the gp43 antigen of Paracoccidioides brasiliensis: immunological reactivity of the recombinant fusion proteins. Clin. Diagn. Lab. Immunol. 9:1200-1204.[CrossRef][Medline]
14
14. Franco, M., M. T. Peracoli, A. Soares, R. Montenegro, R. P. Mendes, and D. A. Meira. 1993. Host-parasite relationship in paracoccidioidomycosis. Curr. Top. Med. Mycol. 5:115-149.[Medline]
15
15. Gesztesi, J. L., R. Puccia, L. R. Travassos, A. P. Vicentini, J. Z. de Moraes, M. F. Franco, and J. D. Lopes. 1996. Monoclonal antibodies against the 43,000 Da glycoprotein from Paracoccidioides brasiliensis modulate laminin-mediated fungal adhesion to epithelial cells and pathogenesis. Hybridoma 15:415-422.[Medline]
16
16. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227:680-685.[CrossRef][Medline]
17
17. Marques, A. F., M. B. da Silva, M. A. Juliano, L. R. Travassos, and C. P. Taborda. 2006. Peptide immunization as an adjuvant to chemotherapy in mice challenged intratracheally with virulent yeast cells of Paracoccidioides brasiliensis. Antimicrob. Agents Chemother. 50:2814-2819.[Abstract/Free Full Text]
18
18. Marques da Silva, S. H., A. L. Colombo, M. H. Blotta, J. D. Lopes, F. Queiroz-Telles, and D. C. Pires. 2003. Detection of circulating gp43 antigen in serum, cerebrospinal fluid, and bronchoalveolar lavage fluid of patients with paracoccidioidomycosis. J. Clin. Microbiol. 41:3675-3680.[Abstract/Free Full Text]
19
19. Matute, D. R., J. G. McEwen, R. Puccia, B. A. Montes, G. San Blas, E. Bagagli, J. T. Rauscher, A. Restrepo, F. Morais, G. Nino-Vega, and J. W. Taylor. 2006. Cryptic speciation and recombination in the fungus Paracoccidioides brasiliensis as revealed by gene genealogies. Mol. Biol. Evol. 23:65-73.[Abstract/Free Full Text]
20
20. Mendes-Giannini, M. J. S., J. P. Bueno, and M. A. Shikanai-Yasuda. 1990. Antibody response to 43 kDa glycoprotein of Paracoccidioides brasiliensis as a marker for the evaluation of patients under treatment. Am. J. Trop. Med. Hyg. 43:200-206.[Abstract/Free Full Text]
21
21. Mendes-Giannini, M. J., P. F. Andreotti, L. R. Vincenzi, J. L. da Silva, H. L. Lenzi, G. Benard, R. Zancope-Oliveira, H. L. Matos Guedes, and C. P. Soares. 2006. Binding of extracellular matrix proteins to Paracoccidioides brasiliensis. Microbes Infect. 8:1550-1559.[CrossRef][Medline]
22
22. Morais, F. V., T. F. Barros, M. K. Fukada, P. S. Cisalpino, and R. Puccia. 2000. Polymorphism in the gene coding for the immunodominant antigen gp43 from the pathogenic fungus Paracoccidioides brasiliensis. J. Clin. Microbiol. 38:3960-3966.[Abstract/Free Full Text]
23
23. Moura-Campos, M. C., J. L. Gesztesi, A. P. Vincentini, J. D. Lopes, and Z. P. Camargo. 1995. Expression and isoforms of gp43 in different strains of Paracoccidioides brasiliensis. J. Med. Vet. Mycol. 33:223-227.[Medline]
24
24. Neves, A. R., R. L. Mamoni, C. L. Rossi, Z. P. Camargo, and M. H. Blotta. 2003. Negative immunodiffusion test results obtained with sera of paracoccidioidomycosis patients may be related to low-avidity immunoglobulin G2 antibodies directed against carbohydrate epitopes. Clin. Diagn. Lab. Immunol. 10:802-807.[CrossRef][Medline]
25
25. Puccia, R., S. Schenkman, P. A. Gorin, and L. R. Travassos. 1986. Exocellular components of Paracoccidioides brasiliensis: identification of a specific antigen. Infect. Immun. 53:199-206.[Abstract/Free Full Text]
26
26. Puccia, R., and L. R. Travassos. 1991. 43-Kilodalton glycoprotein from Paracoccidioides brasiliensis: immunochemical reactions with sera from patients with paracoccidioidomycosis, histoplasmosis, or Jorge Lobo's disease. J. Clin. Microbiol. 29:1610-1615.[Abstract/Free Full Text]
27
27. Puccia, R., and L. R. Travassos. 1991. The 43-kDa glycoprotein from the human pathogen Paracoccidioides brasiliensis and its deglycosylated form: excretion and susceptibility to proteolysis. Arch. Biochem. Biophys. 289:298-302.[CrossRef][Medline]
28
28. Puccia, R., D. T. Takaoka, and L. R. Travassos. 1991. Purification of the 43 kDa glycoprotein from exocellular components excreted by Paracoccidioides brasiliensis in liquid culture (TOM medium). J. Med. Vet. Mycol. 29:57-60.[Medline]
29
29. Puccia, R., J. G. McEwen, and P. S. Cisalpino. Diversity in Paracoccidioides brasiliensis. The PbGP43 gene as a genetic marker. Mycopathologia, in press.
30
30. Rappleye, C. A., and W. E. Goldman. 2006. Defining virulence genes in the dimorphic fungi. Annu. Rev. Microbiol. 60:281-303.[CrossRef][Medline]
31
31. Restrepo, A., M. Restrepo, F. de Restrepo, L. H. Aristizabal, L. H. Moncada, and H. Velez. 1978. Immune responses in paracoccidioidomycosis. A controlled study of 16 patients before and after treatment. Sabouradia 16:151-163.
32
32. Restrepo-Moreno, A. 1994. Ecology of Paracoccidioides brasiliensis, p. 121-130. In M. Franco, C. S. Lacaz, A. Restrepo-Moreno, and G. Del Negro (ed.), Paracoccidioidomycosis. CRC Press, Boca Raton, FL.
33
33. Sheffield, P., S. Garrard, and Z. Derewenda. 1999. Overcoming expression and purification problems of RhoGDI using a family of "parallel" expression vectors. Protein Expr. Purif. 15:34-39.[CrossRef][Medline]
34
34. Souza, M. C., J. L. Gesztesi, A. R. Souza, J. Z. Moraes, J. D. Lopes, and Z. P. Camargo. 1997. Differences in reactivity of paracoccidioidomycosis sera with gp43 isoforms. J. Med. Vet. Mycol. 35:13-18.[Medline]
35
35. Taborda, C. P., and Z. P. Camargo. 1994. Diagnosis of paracoccidioidomycosis by dot immunobinding assay for antibody detection using the purified and specific antigen gp43. J. Clin. Microbiol. 32:554-556.[Abstract/Free Full Text]
36
36. Taborda, C. P., M. A. Juliano, R. Puccia, M. Franco, and L. R. Travassos. 1998. Mapping of the T-cell epitope in the major 43-kilodalton glycoprotein of Paracoccidioides brasiliensis which induces a Th-1 response protective against fungal infection in BALB/c mice. Infect. Immun. 66:786-793.[Abstract/Free Full Text]
37
37. Travassos, L. R., C. P. Taborda, L. K. Iwai, E. Cunha-Neto, and R. Puccia. 2004. The gp43 from Paracoccidioides brasiliensis: a major diagnostic antigen and vaccine candidate, p. 279-296. In J. E. Domer and G. S. Kobayashi (ed.), Mycota XII, human fungal pathogens. Springer-Verlag, Berlin, Germany.

---

Clinical and Vaccine Immunology, April 2008, p. 622-629, Vol. 15, No. 4
1071-412X/08/$08.00+0     doi:10.1128/CVI.00437-07
Copyright © 2008, American Society for Microbiology. All Rights Reserved.

This Article Abstract Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints and Permissions Copyright Information Books from ASM Press MicrobeWorld Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Carvalho, K. C. Articles by Puccia, R. Search for Related Content PubMed PubMed Citation Articles by Carvalho, K. C. Articles by Puccia, R.