By continuing to browse this site you agree to us using cookies as described in About Cookies
Wiley Online Library is migrating to a new platform powered by Atypon, the leading provider of scholarly publishing platforms. The new Wiley Online Library will be migrated over the weekend of March 17 & 18. You should not experience any issues or loss of access during this time. For more information, please visit our migration page: http://www.wileyactual.com/WOLMigration/
Proteome profiling of the dimorphic fungus Penicillium marneffei extracellular proteins and identification of glyceraldehyde-3-phosphate dehydrogenase as an important adhesion factor for conidial attachment
Susanna K. P. Lau,
State Key Laboratory of Emerging Infectious Diseases, Research Centre of Infection and Immunology and Carol Yu Centre for Infection, University of Hong Kong, China
Department of Microbiology, University of Hong Kong, China
S.K.P. Lau, State Key Laboratory of Emerging Infectious Diseases, Department of Microbiology, University of Hong Kong, University Pathology Building, Queen Mary Hospital, Hong Kong
Despite being the most important thermal dimorphic fungus causing systemic mycosis in Southeast Asia, the pathogenic mechanisms of Penicillium marneffei remain largely unknown. By comparing the extracellular proteomes of P. marneffei in mycelial and yeast phases, we identified 12 differentially expressed proteins among which glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and heat shock protein 60 (HSP60) were found to be upregulated in mycelial and yeast phases respectively. Based on previous findings in other pathogens, we hypothesized that these two extracellular proteins may be involved in adherence during P. marneffei–host interaction. Using inhibition assays with recombinant GAPDH (rGAPDH) proteins and anti-rGAPDH sera, we demonstrated that adhesion of P. marneffei conidia to fibronectin and laminin was inhibited by rGAPDH or rabbit anti-rGAPDH serum in a dose-dependent manner. Similarly, a dose-dependent inhibition of conidial adherence to A549 pneumocytes by rGAPDH or rabbit anti-rGAPDH serum was observed, suggesting that P. marneffei GAPDH can mediate binding of conidia to human extracellular matrix proteins and pneumocytes. However, HSP60 did not exhibit similar inhibition on conidia adherence, and neither GAPDH norHSP60 exhibited inhibition on adherence to J774 or THP-1 macrophage cell lines. This report demonstrates GAPDH as an adherence factor in P. marneffei by mediating conidia adherence to host bronchoalveolar epithelium during the early establishment phase of infection.
Penicillium marneffei is the most important thermal dimorphic fungus causing respiratory, skin and systemic mycosis in Southeast Asia [1-4]. P. marneffei was first discovered in Chinese bamboo rats, Rhizomys sinensis, and subsequently isolated from other species of bamboo rats in the Rhizomyinae subfamily [5, 6]. After its discovery in 1956, only 18 cases of human diseases were reported until 1985 . The emergence of the HIV pandemic in the 1980s was met with a surge of reports of HIV-associated P. marneffei infections in Southeast Asia where the fungus is endemic. In northern Thailand, penicilliosis is the third most common indicator disease of AIDS, after tuberculosis and cryptococcosis . In Hong Kong, about 10% of HIV patients are infected with P. marneffei, which represents the sixth leading cause of death [8, 9]. Aside from endemic regions, cases of imported P. marneffei infections have also been reported from various countries around the world [10, 11]. Besides HIV-positive patients, P. marneffei infections are increasingly reported in other immunocompromised patients, such as transplant recipients, patients with systemic lupus erythematosus and those on corticosteroid therapy [12-15].
Despite its medical importance, the mode of transmission and pathogenic mechanisms of P. marneffei remain largely unknown. Besides bamboo rats, P. marneffei has been isolated from soil samples collected from the burrows of these animals [5, 16]. It has also been observed that the rainy season and an agricultural occupation are risk factors for the disease . As for other thermal dimorphic fungi, inhalation of fungal conidia carried by airborne soil is thought to be the route of transmission for P. marneffei. One of the first steps towards establishment of infection in the lungs leading to subsequent dissemination to other organs is the adhesion of fungal pathogens to host cells. Extracellular matrix (ECM) proteins have been implicated to facilitate the attachment of a variety of pathogens, such as Aspergillus fumigatus conidia, to host tissues [18, 19]. It has been demonstrated that P. marneffei conidia can bind fibronectin and laminin [20, 21], as well as ECM-associated glycosaminoglycans, chondroitin sulfate B, heparin and highly sulfated chitosan CP-3, which are major constituents of various tissues especially the basal lamina . Although such interaction is postulated to facilitate the attachment of P. marneffei conidia to the bronchoalveolar epithelium of the host, the surface molecules involved remain unknown.
In 2002, we started the P. marneffei genome project which has expedited studies on the biology, epidemiology and virulence factors of this dimorphic fungus [23-28]. In this study, using the available genome sequence data, we attempted to characterize the extracellular proteomes of P. marneffei in mycelial and yeast phases. Twelve of 52 differentially expressed proteins observed in two-dimensional gel electrophoresis (2-DE) were identified by peptide mass fingerprinting (PMF) using MALDI-TOF MS. Upon annotation of these 12 identified proteins, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and heat shock protein 60 (HSP60) were noted to be involved in adherence for various pathogens including fungi. Since GAPDH and HSP60 were upregulated in the mycelial and yeast phases of P. marneffei respectively, we hypothesized that they may serve similar functions in the context of P. marneffei–host interaction in the mycelial and yeast phases respectively. Using inhibition assays by rGAPDH proteins and anti-rGAPDH antibodies, we demonstrated for the first time that P. marneffei GAPDH can mediate the binding of conidia to fibronectin and laminin and in vitro cultured human pneumocytes, suggesting that the protein may play an important role in the establishment of disease.
Analysis of extracellular proteomes of P. marneffei in yeast and mycelial phases
To study the extracellular proteomes of P. marneffei in mycelial and yeast phases, extracellular proteins from P. marneffei cultured at 25 °C and 37 °C respectively were subjected to 2-DE and compared. Approximately 81–102 spots per gel were identified within the target pI range 3–10. Only spots with at least a two-fold difference in their spot volumes between mycelial and yeast phases, which is statistically significant (P >0.05), were considered differentially expressed. A total of 21 ‘mycelia-upregulated’ spots and 31 ‘yeast-upregulated’ spots were observed, among which four ‘mycelia-upregulated’ and eight ‘yeast-upregulated’ spots were successfully identified by PMF (Table 1). Sequence analysis showed that these proteins were involved in various functions including cellular metabolism, although signal peptides were only identified in three of these 12 extracellular proteins. Among the mycelia-upregulated proteins one was identified as a homolog of GAPDH, whereas among the yeast-upregulated proteins one was identified as a homolog of the antigenic mitochondrial protein HSP60. Since these two proteins are known to play a role in adherence for various pathogens including fungi [29-34], they were subjected to further investigations on their possible role in the adherence of P. marneffei during pathogen–host interaction.
Table 1. Differentially expressed extracellular proteins in mycelial and yeast phases of P. marneffei. NA, not applicable (ratio not calculated as the protein spot was only present in one phase but not the other phase); MW, theoretical molecular weight determined from amino acid sequences; pI, theoretical isoelectric point determined from amino acid sequences. Spot no. refers to protein spot numbers in Fig. 1. Locus tag gives the Genbank accession no. of the locus tag in the P. marneffei genome. MOWSE score was determined by the ms-fit program (http://prospector.ucsf.edu). Normalized spot volume ratio was determined by imagemaster platinum 6.0, indicating the abundance change of P. marneffei proteins grown at yeast against mycelia phase. signalp 3.0 was used to predict the presence of signal peptide (3). Y denotes the presence of signal peptide whereas N denotes the absence of signal peptide. secretomep was used to predict non-classical protein secretion (4). Protein that is predicted to be non-classically secreted has a score exceeding the normal threshold of 0.5. Y indicates the protein is predicted to be secreted whereas N indicates the protein is not predicted to be secreted. P value refers to the significance of the difference of spot volumes (pixel intensity × area) of the particular spot quantified on the gel between the yeast and mycelial phase, as determined by Student's t test.
Theoretical MW (kDa)/pI
Sequence coverage (%)
Yeast/mould spot volume ratio
Upregulated in mycelial phase
Putative uncharacterized protein
Glyceraldehyde 3-phosphate dehydrogenase
Upregulated in yeast phase
Antigenic mitochondrial protein HSP60
Malate dehydrogenase, NAD-dependent
Other mycelia-upregulated proteins include phosphoglycerate kinase (PGK) and uricase. Apart from its role in glycolysis, PGK has been found in the extracellular proteome of other fungi including Histoplasma capsulatum and was implicated in cell wall localization in Candida albicans [35-37]. Uricase is an enzyme that converts urate to allantoin. Other yeast-upregulated proteins include various metabolic enzymes. Spermidine synthase is involved in the metabolism of polyamine which has been reported to be related to dimorphic transition, cell differentiation or sporulation in other fungi such as Ustilago maydis [38, 39]. Although dihydrolipoamide dehydrogenase is a component of the mitochondrial pyruvate dehydrogenase complex and the α-ketoglutarate dehydrogenase complex in the tricarboxylic acid cycle, its presence in culture supernatant in other fungi is not unprecedented [36, 40]. Saccharopine dehydrogenase is involved in the α-aminoadipate pathway for l-lysine biosynthesis. Fructose-bisphosphate aldolase, a key enzyme in gluconeogenesis, has been found in the cell wall of fungi including C. albicans and Saccharomyces cerevisiae and is suggested to be a potential antigen that reacts with sera from patients with candidiasis [41, 42]. While NAD-dependent malate dehydrogenase is another enzyme involved in glycolytic/tricarboxylic acid /glyoxylate cycles, it has also been identified in the extracellular proteomes of other pathogenic fungi such as Fusarium graminearum  and as an immunoreactive molecule in the secretome of A. fumigatus .
Reactivities of polyclonal rabbit anti-rGAPDH and anti-rHSP60 sera
To study the potential role of GAPDH and HSP60 in fungal–host adherence, rGAPDH and recombinant HSP60 (rHSP60) were cloned, expressed and purified in Escherichia coli. To obtain polyclonal anti-rGAPDH and anti-rHSP60 antibodies for subsequent experiments, rabbits were immunized with rGAPDH and rHSP60 respectively. Sera from rabbits immunized with rGAPDH and rHSP60 were demonstrated to be reactive against P. marneffei rGAPDH and rHSP60 respectively by western blot analysis (Fig. 1A). Blots treated with sera from rabbits immunized with rGAPDH (lanes 1 and 2) or rHSP60 (lanes 3 and 4) revealed reactive protein bands at around 37 kDa or 64 kDa respectively in lanes run with total protein extracts from P. marneffei (lanes 1 and 3), which are comparable to the electrophoretic mobility of the dominant protein bands in lanes run with rGAPDH (lane 2) and rHSP60 (lane 4) samples.
Antigenicity of native GAPDH and HSP60 in mice infected with P. marneffei
To test the antigenicity of native GAPDH and HSP60, mice were challenged with P. marneffei conidia or yeast. Western blot analysis showed that sera from mice experimentally infected with P. marneffei conidia was reactive against rGAPDH (Fig. 1B, lane 8), while no reaction could be observed between rGAPDH and sera from yeast-infected mice (Fig. 1B, lane 9), suggesting that native GAPDH in the conidial phase of P. marneffei is immunogenic in mice. The results also supported that GAPDH was produced by conidia of P. marneffei during in vivo infection. Native HSP60, on the other hand, did not appear to be an immunogenic protein in mice, since no reaction or only very faint bands could be observed between rHSP60 and sera from conidia- or yeast-infected mice (Fig. 1B, lanes 12, 13).
Adhesion of P. marneffei conidia to immobilized fibronectin and laminin is mediated by GAPDH
Since P. marneffei infection is most probably acquired through inhalation of airborne conidia which are small enough to reach the alveoli, instead of the much larger hyphal filaments of mycelia, we hypothesize that conidial adhesion to host tissues may be mediated by GAPDH. Also because P. marneffei conidia are known to bind fibronectin and laminin, which are important tissue components, we examined the role of GAPDH in mediating the adhesion of conidia to these ECM proteins. Adhesion of conidia to fibronectin or laminin appears to be dependent on GAPDH, as adhesion of conidia to the ECM proteins was inhibited by the addition of rGAPDH (Fig. 2A) or rabbit anti-rGAPDH polyclonal antibody (pAb) (Fig. 2B) in a dose-dependent manner. Conidial adhesion to fibronectin and laminin was significantly inhibited by rGAPDH at a concentration of 100 μg·mL−1, with 82% and 78% reduction in adherent conidia respectively (P <0.05). Similarly, rabbit anti-rGAPDH serum at a concentration of 1 : 50 significantly reduced conidial adhesion to fibronectin and laminin by 68.3% and 68% respectively (P <0.01), while rabbit anti-rGAPDH serum at a concentration of 1 : 10 significantly reduced conidial adhesion to fibronectin and laminin by 92% and 91% respectively (P <0.01). Since rGAPDH and anti-rGAPDH is intended to mimic the two sides of the conidia–ECM binding interface, the fact that both rGAPDH and anti-rGAPDH can inhibit the adherence of conidia to fibronectin and laminin suggests that this binding interaction is GAPDH-specific. Moreover, rHSP60 did not exhibit significant inhibition on the conidia adherence to fibronectin and laminin, with only 21.8% and 28.7% reduction respectively (not statistically significant) using 100 μg·mL−1 rHSP60. The role of secreted HSP60 in yeast adherence to ECM could not be evaluated, since it was found that P. marneffei yeast cells did not exhibit adherence characteristics like conidia (data not shown).
Adhesion of P. marneffei conidia to cells mediated by GAPDH
Since GAPDH appears to mediate conidial adhesion to ECMs, the possible role of GAPDH in the adhesion of conidia to human cells was assessed using the same strategy as the ECM-based assays. As in ECM-based assays, a dose-dependent inhibition of conidial adherence to A549 pneumocytes could be observed, with statistically significant inhibition being observed with rGAPDH at a concentration of 100 μg·mL−1 and rabbit anti-rGAPDH serum at 1 : 10, where adherence in both cases was reduced by ~ 57% (P <0.01) (Fig. 3). However, GAPDH or HSP60 is unlikely to be involved in the interaction between P. marneffei (conidia or yeast respectively) and macrophages since no observable differences could be observed in adhesion inhibition assays performed using J774 and THP-1 macrophage cell lines (data not shown).
Differential mRNA expression of gapdh in mycelial and yeast phases
Transcriptional analysis showed that the mRNA level of gapdh was higher in the mycelial than yeast phase by ~ 19.37 (P =0.002). This suggested that the ‘mycelia-upregulated’ expression of GAPDH observed in the extracellular proteome is probably a result of upregulated transcription in the mycelial phase.
Surface localization of GAPDH on P. marneffei conidia by indirect immunofluorescence
To examine the cellular distribution of GAPDH, paraformaldehyde-fixed P. marneffei conidia were examined by indirect immunofluorescence light microscopy with rabbit anti-rGAPDH serum. GAPDH was mainly localized on the conidial surface (Fig. 4).
This report demonstrates GAPDH as a potential virulence factor in P. marneffei by mediating conidia adherence to host tissues during the early stages of infection. Attachment to host is the first critical step in establishment of infection by pathogens. P. marneffei infection is thought to be acquired through inhalation of airborne conidia which are small enough to reach the alveoli. Thus, the attachment of conidia to respiratory epithelium can help avoid entrapment by mucus and removal by ciliary action. Fibronectin and laminin are ECM proteins that are known to be ligands that mediate adherence to host tissues by various fungal pathogens such as A. fumigatus [44, 45], C. albicans , H. capsulatum  and Paracoccidioides brasiliensis . Laminin, a heavily glycosylated protein present in basement membranes and in the lungs, can be exposed after tissue damage. Although it has been shown that P. marneffei conidia can bind to fibronectin and laminin probably via a common adhesion molecule [20, 21], the identity of fungal protein(s) involved in this interaction was unknown. In this study, we used fibronectin and laminin, A549 human alveolar epithelial cells, J774 murine and THP-1 human macrophage cell lines as analogues to the host environment to test for the potential role of GAPDH and HSP60 of P. marneffei as adhesion factors. The adherence of P. marneffei conidia to ECM glycoproteins was inhibited in a dose-dependent manner with increasing concentrations of soluble rGAPDH protein or anti-rGAPDH sera (i.e. by blocking putative GAPDH interacting sites on both sides of the interaction interface) (Fig. 5). A similar trend was also observed in a conidial adherence study using A549 cells. HSP60, mainly present in the yeast phase of P. marneffei, was shown to have little role in conidial adherence, which in turn provides supporting evidence for the specificity of the results observed in GAPDH mediated adherence assays. By comparing the maximal percent reduction in GAPDH mediated conidial adherence in the presence of inhibitors, it seems that conidial adhesion was more severely inhibited under the same inhibitor concentration in the ECM based assay as in the cell based assay. Therefore, it may be possible that other factors are involved in the adherence of conidia to A549 cells in a non-GAPDH dependent manner. Nevertheless, the present results suggest that GAPDH may mediate the adherence of P. marneffei conidia to the host ECM proteins and bronchoalveolar epithelium, a critical step during the early phase of infection. However, attempts to verify the role of GAPDH in mice pathogen-challenge studies using gapdh knockdown mutants of P. marneffei were unsuccessful, as all mutants exhibited severely stunted growth probably due to the housekeeping cellular function played by GAPDH (unpublished data).
The identification of GAPDH as mediator for P. marneffei–host adherence supports the multi-functionality of this housekeeping enzyme as observed in other pathogens. GAPDH is best known as a cytoplasmic glycolytic enzyme that is responsible for the sixth step of glycolysis by catalyzing the phosphorylation of glyceraldehyde-3-phosphate to d-glycerate 1,3-bisphosphate. This housekeeping enzyme has become increasingly recognized to be presented to the cell surface through unconventional secretory pathways and can serve various cellular functions in pathogens such as C. albicans, Streptococcus pyogenes, Staphylococcus aureus and Staphylococcus epidermidis, where it may be involved in host–pathogen interactions [49-52]. GAPDH has also been demonstrated to be a transferrin binding protein for S. epidermidis and S. aureus [53, 54], and involved in mediation of signaling for host–bacterial cross-talk for enteropathogenic E. coli strains . In C. albicans, GAPDH has been found to serve multiple functions as a plasminogen binding protein, and as a highly immunogenic cell-wall-associated protein which binds fibronectin and laminin [30, 31, 33]. In another thermal dimorphic fungal pathogen, P. brasiliensis, secreted GAPDH has also been shown to play a role in the adherence of its yeast forms to fibronectin, laminin, as well as A549 pneumocytes . Since host adherence represents the crucial first step in the pathogenic process, the identification of GAPDH as adherence factors in various pathogens may have significant implications with regard to development of antimicrobial agents and vaccines.
As an intracellular pathogen, interaction of P. marneffei with host macrophages also represents a crucial step to host invasion. In this study, the parasitic yeast form of P. marneffei does not appear to exhibit adherence to ECM glycoproteins. This is not unexpected, since during infection the yeast forms of P. marneffei are predominantly located intracellularly within host macrophages [7, 11, 56-58] and are probably not as important as conidia for initiation of infection by cell adhesion. However, we found that neither GAPDH nor HSP60 was involved in P. marneffei conidia–macrophage or yeast–macrophage adherence respectively, using murine macrophage and human monocyte cell lines. While it remains a possibility that the upregulation of HSP60 in the yeast phase of P. marneffei may be part of a mild heat stress response, further investigations should be performed to explore its other possible functions in the yeast phase such as intracellular survival within macrophages.
The inhibition of conidial adherence by anti-rGAPDH serum suggested a cell-surface localization pattern of GAPDH on P. marneffei. Cell-surface localization of GAPDH has been demonstrated in studies on unconventionally secreted GAPDH of C. albicans [31, 33], P. brasiliensis  as well as various bacterial pathogens, where the protein may serve a non-glycolytic function. The presence of GAPDH in culture filtrates of P. marneffei suggested that the protein may be secreted in extracellular space and is present in extracellular space at higher levels in mycelial than yeast phase. Our results using indirect immunofluorescence microscopy labeled with anti-GAPDH serum also supported that GAPDH is mainly localized on the conidial surface (Fig. 4). Although gapdh is considered to be a constitutively expressed housekeeping gene, there is increasing evidence showing that its expression can be highly regulated to serve different functions under different conditions [59, 60]. Cell-wall-associated GAPDH activity has been found to increase in response to starvation and temperature upshift in the yeast Saccharomyces cerevisiae . In P. brasiliensis, a higher expression of GAPDH was found in the yeast phase . In contrast, GAPDH was found to be present in the extracellular proteome only in the mycelial phase of P. marneffei in the present study. Moreover, the mRNA level of gapdh was ~ 20-fold higher in the mycelial than yeast phase, supporting ‘mycelia-upregulated’ transcription. Our results are in line with a previous study showing that the expression of GAPDH was downregulated in the yeast phase as well as during macrophage infection by P. marneffei . However, a recent proteomic study on P. marneffei using whole cell extracts identified GAPDH to be expressed at equivalent levels in developing yeast and mould phases . Since the present study suggested that GAPDH is apparently secreted by mycelia but also found on the conidial surface of P. marneffei, further studies are required to determine if the ‘secreted’ GAPDH detected in the secretome was due to accidental shedding from the conidial surface and to better define the amount and different roles of secreted, cytoplasmic and extracellular GAPDH in P. marneffei.
This study demonstrates the utility of proteomic approaches in studying pathogenesis in dimorphic fungi. Previous studies on comparative proteomics of P. marneffei, which mainly focused on whole cell proteins, have identified differentially expressed proteins that may be involved in cellular development and dimorphism [63, 64]. However, no further downstream experiments were shown to demonstrate the function of these proteins in the two studies. The present report represents the first attempt to study the extracellular proteomes of P. marneffei. Further characterization of other proteins in different growth phases may provide important clues to their roles in dimorphism or pathogenicity.
Materials and methods
P. marneffei strain and growth conditions
P. marneffei strain PM1 was isolated from a patient suffering from culture-documented penicilliosis in Hong Kong. Informed consent was not obtained from the patient from whom the initial isolate was collected in 1992. P. marneffei was grown on Sabouraud dextrose agar (Oxoid, Cambridge, UK) at 37 °C for 10 days for yeast cultures or at room temperature for 7 days for mould cultures for the collection of conidia as described previously . Yeast cells and conidia were collected by scraping and resuspension in 0.1% Tween-20 with phosphate-buffered saline (NaCl/Pi) followed by three washes in NaCl/Pi before use. Cells were enumerated using a hemocytometer. For preparation of liquid cultures for extracellular protein extraction, cells were resuspended in MilliQ water to obtain a concentration of McFarland 1. 5 mL of this inoculum was added to 500 mL BHI broth (Oxoid, Basingstoke, UK) for incubation at 37 °C for yeast phase and at 25 °C for mycelial phase for 3 days, with shaking at 250 rpm. Cultures were prepared in triplicate.
Analysis by 2-DE of extracellular proteins from P. marneffei grown in yeast and mycelial phases
Extracellular proteins of P. marneffei were prepared according to published protocols with modifications . Culture supernatant was filtered through a 0.22 μm pore size polyethersulfone low protein binding membrane filter (Corning Inc., New York, NY, USA). The filtrate was mixed with 0.5 mm phenylmethanesulfonyl fluoride (PMSF) and 1 mm EDTA, and concentrated by ultrafiltration via an Amicon stirred ultrafiltration cell (Millipore, Bedford, MA, USA) with 10 000 MWCO Ultracel YM regenerated cellulose ultrafiltration membrane (Millipore) at 4 °C. Samples were further concentrated using an Amicon Ultra-15 centrifugal filter device equipped with a 10 000 MWCO Ultracel YM regenerated cellulose membrane, and purified using 2-D Clean-up Kit (Amersham Biosciences Inc., Piscataway, NJ, USA). Purified proteins were dissolved in sample buffer containing 7 m urea, 2 m thiourea and 4% CHAPS. Protein concentration was determined according to the method proposed by Bradford .
2-DE and protein analysis were performed as described previously with modifications . Briefly, immobilized pH gradient (IPG) strip (Bio-Rad Laboratories, Hercules, CA, USA) (17 cm) with pH 3–10 was hydrated overnight in rehydration buffer containing 7 m urea, 2 m thiourea, 4% CHAPS, 1% IPG buffer pH 3–10 (GE Healthcare, Little Chalfont, UK) and 60 mm dithiothreitol (DTT) with 60 μg of purified extracellular protein. The first dimension, isoelectric focusing (IEF), was carried out in a Protean IEF cell electrophoresis unit (Bio-Rad Laboratories) for 100 000 V h. IPG strips were then equilibrated in 1% DTT in equilibration buffer containing 50 mm Tris/HCl, 6 m urea, 30% glycerol, 2% SDS and 0.002% bromophenol blue for 15 min, followed by incubation in 2.5% iodoacetamide in equilibration buffer for 15 min. Protein separation in the second dimension was performed in 18 × 20 cm 12% SDS/PAGE utilizing Bio-Rad Protean II xi unit (Bio-Rad Laboratories) at 16 V for 30 min followed by 24 V for 5 h. 2D gels were stained with silver and colloidal Coomassie Blue G-250 respectively for qualitative and quantitative analysis, and scanned with ImageScanner (GE Healthcare). ImageMaster 2D Platinum 6.0 (GE Healthcare) was used for image analysis. Protein spots were manually excised from gels and subjected to in situ digestion with trypsin, and peptides generated were analyzed using a 4800 Plus MALDI TOF/TOF Analyzer (Applied Biosystems, Foster City, CA, USA). Proteins were identified by PMF using the ms-fit software (http://prospector.ucsf.edu) and an in-house sequence database of P. marneffei PM1 proteins generated using available P. marneffei genome sequence data (Genbank accession no. AGCC00000000 and ABAR00000000) . At least three independent experiments for each growth condition were performed to ensure reproducibility of results.
The identified proteins were used as input in blastp searches against the NCBI non-redundant database and were analyzed for protein family assignments, functional classification, conserved domains, transmembrane regions and signal peptides using interproscan , signalp version 3.0  and secretomep version 2.0 (http://www.cbs.dtu.dk/services/SecretomeP/) .
Cloning and expression of GAPDH and HSP60 in Escherichia coli
Cloning and expression of GAPDH and HSP60 was performed using protocols as described previously [65, 72]. Total RNA was extracted from a yeast culture of P. marneffei using RiboPure-Yeast (Ambion, Foster City, CA, USA). The complete coding sequence of gapdh and hsp60 was PCR amplified from cDNA using primers based on P. marneffei genome sequence data. Forward and reverse primer pairs LPW 8137 (5′-CTAGCTAGCATGGTTACCAAGGTATGTCG-3′) and LPW 8138 (5′-CTAGCTAGCCTAAGCGTTGCCGTCG-3′), and LPW 8139 (5′-CTAGCTAGCATGCAGCGCGCTTTCT-3′) and LPW 8140 (5′-CTAGCTAGCTTAGAAGCCCATGCCAC-3′), were used for the amplification of gapdh and hsp60 respectively. PCR was performed in a 25 μL reaction mixture containing cDNA, 1 × iProof HF buffer (20 mm Tris/HCl pH 7.4, 0.1 mm EDTA, 1 mm DTT, 100 mm KCl, 0.5% Tween-20, 0.5% Nonidet P40, 200 μg·mL−1 BSA and 50% glycerol), 200 μm of each dNTP, forward and reverse gene-specific primers and 0.02 U iProof High Fidelity DNA polymerase (Bio-Rad Laboratories) and subjected to an initial 98 °C for 30 s for denaturation, followed by 40 cycles of 98 °C for 10 s, 55 °C for 30 s and 72 °C for 50 s, with a final extension at 72 °C for 10 min in an automated thermal cycler (Applied Biosystems). The underlined sequence indicated in the LPW primers encodes for NheI sites to be incorporated onto both ends of the PCR products for cloning into the pET28b(+) plasmid (Novagen Inc., Madison, WI, USA) in-frame and downstream of the series of six histidine residues. The fusion constructs were then transformed into E. coli BL21 Gold (DE3) competent cells (Strategene, La Jolla, CA, USA) by electroporation. Transformed bacteria were grown in LB medium supplemented with kanamycin (100 μg·mL−1) at 37 °C until the optical density at 600 nm reached 0.6, followed by induction with isopropyl-β-d-thiogalactopyranoside at a final concentration of 0.1 mm. After 4 h, cells were collected by centrifugation and the pellet was resuspended in 4 mL of buffer containing 20 mm Tris/HCl, 0.5 mm NaCl and 20 mm imidazole (GE Healthcare) at pH 8, and sonicated on ice for 45 cycles of 10 s pulses at 10 amplitude microns (SoniPrep 150, Sanyo, Chatsworth, CA, USA) to obtain cell lysate. Supernatant was obtained from the cell lysate after centrifugation. The (His)6-tagged rGAPDH and rHSP60 proteins contained in the lysate supernatant were subsequently purified using the Ni2+-loaded HiTrap Chelating System (GE Healthcare). Approximately 3 mg of purified protein was routinely obtained from 1 L of E. coli carrying the fusion plasmid.
This study was approved and performed in strict accordance with local ordinance and the recommendations by the Committee on the Use of Live Animals in Teaching and Research (CULATR) at the University of Hong Kong. Rabbit pAb against rGAPDH and rHSP60 were produced using previously described protocols with modifications [65, 73]. Briefly, 300 μg of purified recombinant proteins was mixed with an equal part of complete Freund's adjuvant and injected subcutaneously into 10-week-old New Zealand White rabbits, followed by two boosts with incomplete Freund's adjuvant every 2 weeks thereafter. Rabbit sera were collected 2 weeks after the third injection. Mouse sera used for testing the immunogenicity of native P. marneffei GAPDH and HSP60 were produced by intravenous challenge of BALB/c (H-2d) mice at 6–8 weeks of age with either 6 × 106 yeast or 8 × 106 conidia suspended in NaCl/Pi as described previously [72, 74]. Mouse sera were collected 4 weeks after challenge.
Extraction of total proteins from P. marneffei
Total protein extracts were prepared by cell disruption using 212–300 μm glass beads. Briefly, P. marneffei was harvested and washed with a buffer containing 10 mm Tris/HCl (pH 7.5), 1.0 mm EDTA and 0.5 mm PMSF. Cells were then resuspended in a buffer containing 7.0 m urea, 2.0 m thiourea and 4% CHAPS dissolved in dH2O and subjected to bead bashing at 10 s intervals for a total of 80 s. The crude supernatant was aspirated from the glass beads and centrifuged at 16 500 g for 1 h at 20 °C to sediment cell debris. The resulting supernatant was then collected as total protein extracts and stored at −80 °C until use.
Western blot analyses
Purified rGAPDH and rHSP60 proteins and total protein extracts were subjected to SDS/PAGE and electroblotted onto a nitrocellulose membrane (Bio-Rad) as described previously [65, 73]. The blot was then incubated with 1 : 3000 mouse anti-His IgG (GE Heathcare), pre-immune or immune mouse serum, followed by a 1 : 3000 horseradish peroxidase (HRP) conjugated goat anti-mouse IgG (H+L) antibody (Zymed, San Francisco, CA, USA). Alternatively, for testing of rabbit pAb against rGAPDH and rHSP60, the blot was incubated with rabbit sera followed by 1 : 4000 HRP-conjugated goat anti-rabbit IgG (H+L) antibody (Zymed). Final detection was performed with an ECL fluorescence system (GE Healthcare).
Adhesion of P. marneffei to fibronectin and laminin with protein or antibody inhibition
Adhesion assays to fibronectin or laminin were performed as described previously [19, 20, 33, 44]. BSA, fibronectin from human plasma and laminin derived from Engelbreth-Holm-Swarm murine sarcoma basement membrane were all purchased from Sigma-Aldrich Co. (St Louis, MO, USA) and stored in aliquots at −80 °C until use. Fibronectin or laminin were immobilized onto 96-well plates (Maxisorp, Nunc-Immuno, Denmark) by adding 100 μg·mL−1 solutions of the proteins to wells and were incubated for 1 h at 37 °C followed by three washes with NaCl/Pi. Wells were blocked with 0.1% BSA and incubated overnight at 4 °C and washed again with NaCl/Pi.
For protein inhibition assays, 105P. marneffei conidia were added and incubated for 1 h at 37 °C in the presence of 0, 1, 10 and 100 μg·mL−1 of rGAPDH or rHSP60 protein in NaCl/Pi or in 100 μg·mL−1 BSA as irrelevant protein control. For antibody inhibition assays, P. marneffei were resuspended to 107 mL−1 and pre-incubated in 1 : 100, 1 : 50 and 1 : 10 dilutions of rabbit pAb against rGAPDH or rHSP60 in NaCl/Pi for 1 h at 37 °C with shaking at 250 rpm, and washed three times with NaCl/Pi before adding 105 conidia per well. P. marneffei in NaCl/Pi only or pre-incubated with 1 : 10 dilution of pre-immune rabbit sera were used as negative controls. Unattached P. marneffei cells were then removed by three 1 min washes in 0.1% Tween-20 in NaCl/Pi and attached conidia were fixed in 4% paraformaldehyde for 15 min at room temperature. Finally, the plates were observed under a phase-contrast microscope at 400× magnification (Nikon Eclipse TS100) and the number of attached P. marneffei was enumerated by taking the average count from five random fields chosen from the center, top, bottom, left and right of each well. All adhesion assays were performed in triplicate.
Adhesion of P. marneffei to pneumocytes and macrophages
The cell-based adherence assays are a variant of the ECM-based assays wherein the fibronectin and laminin was replaced with different cell lines as described previously [29, 62]. A549 human alveolar epithelial cells and J774 murine macrophage-like cells were maintained in DMEM supplemented with 10% heat-inactivated fetal bovine serum (FBS) and incubated at 37 °C in a chamber with 5% CO2. THP-1 human monocytic macrophage cells were maintained in RPMI 1640 medium supplemented with 10% FBS and incubated as above. Approximately 2 × 104 cells were seeded onto 96-well tissue culture plates 24 h prior to experiment to achieve monolayers. Additionally, THP-1 cells were differentiated into mature adherent macrophages by treatment with phorbol myristate acetate (PMA) (Sigma-Aldrich Co.) at a final concentration of 0.32 μm during the 24 h incubation period before addition of conidia. Cells were washed three times in FBS-free culture medium before addition of P. marneffei with protein or antibody inhibition and incubation using conditions described above in fibronectin and laminin adhesion assays. P. marneffei in NaCl/Pi only, BSA or pre-incubated with a 1 : 10 dilution of pre-immune rabbit sera were used as negative controls. Medium appropriate to the respective cell lines were used for all washing and protein and P. marneffei incubation steps. Unattached P. marneffei cells were removed by three 1 min washes with culture medium. The macrophages or pneumocytes were lysed by repetitive pipetting with 100 μL of NaCl/Pi with 1% Triton X-100 added per well to free the attached conidia from cells. Finally, the lysis mixture containing the released conidia was serially diluted with NaCl/Pi and plated onto Sabouraud dextrose agar plates and incubated at 37 °C for up to 10 days. Colony forming units were enumerated and compared between different groups. All adhesion assays were performed in triplicate.
Quantitative real-time PCR
Total RNA was extracted using RiboPure-Yeast (Ambion). The RNA was eluted in 100 μL of RNase-free water and was used as the template for real-time RT-PCR. Reverse transcription was performed using the SuperScript III kit (Invitrogen, Carlsbad, CA, USA). Real-time RT-PCR assays for gapdh mRNA level were performed, as described previously with modifications , using primers (LPW21406 5′-AAGGTTGGCATCAACGGTT-4′ and LPW21407 5′-AGGCAGCGTAGTGAGTCTCA-3′). cDNA was amplified in a LightCycler 2.0 instrument (Roche, Basel, Switzerland) with 20 μL reaction mixtures containing FastStart DNA Master SYBR Green I Mix reagent kit (Roche), 2 μL of cDNA, 2 mm magnesium chloride and 1 mm primers at 95 °C for 10 min, followed by 50 cycles of 95 °C for 10 s, 55 °C for 5 s and 72 °C for 10 s. Each amplification was followed by dissociation curve analysis to ensure that a single-size product was amplified. In addition, amplification products were run on an agarose gel to confirm the correct product size. All experiments were performed in three independent replicates. Levels of gapdh mRNA were normalized to actin and fold-change comparisons were performed between yeast and mycelial phases. Fold change was calculated as 2−ΔΔCt using actin as endogenous control .
Indirect immunofluorescent microscopy was performed as described previously with modifications . P. marneffei conidia were resuspended in 4% paraformaldehyde and fixed for 1 h at 4 °C. After washing with NaCl/Pi, fungal cells were resuspended in rabbit immune and pre-immune sera diluted to 1 : 10 in NaCl/Pi with 1% BSA at a final concentration of 108 cells·mL−1. After overnight incubation at 4 °C on a rocking platform followed by two washings with NaCl/Pi, fungal cells were resuspended in fluorescein isothiocyanate conjugated goat anti-rabbit IgG (Sigma-Aldrich Co.) diluted to 1 : 20 in NaCl/Pi with 1% BSA and incubated for a further 1 h at 37 °C and washed twice more in NaCl/Pi. Finally, stained cells were resuspended in 100 μL NaCl/Pi and wet-mounted onto slides with glass coverslips and examined by immunofluorescence microscopy.
Statistical analyses of the data from fungal adherence and quantitative RT-PCR studies were performed using two-tailed Student's t test (IBM SPSS Statistics 19).
Nucleotide sequence accession number
The nucleotide sequences of the gapdh and hsp60 genes of P. marneffei have been deposited with GenBank under accession no. JQ231120 and JQ231121 respectively.
This work is partly supported by the Research Grant Council Grant, University Grant Council; Committee for Research and Conference Grant, Strategic Research Theme Fund, and University Development Fund, University of Hong Kong; the HKSAR Research Fund for the Control of Infectious Diseases (commissioned study) of the Health, Welfare and Food Bureau; the Shaw Foundation; and Providence Foundation Limited in memory of the late Dr Lui Hac Minh. The authors declare no conflict of interest.