Proteome Profiling and Functional Classification of Intracellular Proteins from Conidia of the Human-Pathogenic Mold Aspergillus fumigatus Janka Teutschbein,†,‡ Daniela Albrecht,§ Maria Po ¨ tsch,† Reinhard Guthke,§ | | Vishukumar Aimanianda, Ce´cile Clavaud, Jean-Paul Latge´,| Axel A. Brakhage,†,‡ and Olaf Kniemeyer*,†,‡ Department of Molecular and Applied Microbiology, Leibniz Institute for Natural Product Research and Infection Biology - Hans-Kno¨ll-Institute (HKI), Jena, Germany, Department of Microbiology and Molecular Biology, Friedrich Schiller University, Jena, Germany, Research Group Systems Biology/Bioinformatics, Leibniz Institute for Natural Product Research and Infection Biology - Hans-Kno¨ll-Institute (HKI), Jena, Germany, and Aspergillus unit, Institut Pasteur, Paris, France Received November 22, 2009
Aspergillus fumigatus is a ubiquitously distributed filamentous fungus that has emerged as one of the most serious life-threatening pathogens in immunocompromised patients. The mechanisms for its pathogenicity are poorly understood. Here, we analyzed the proteome of dormant A. fumigatus conidia as the fungal entity having the initial contact with the host. Applying two-dimensional polyacrylamide gel electrophoresis (2-D PAGE), we established a 2-D reference map of conidial proteins. By MALDITOF mass spectrometry, we identified a total number of 449 different proteins. We show that 57 proteins of our map are over-represented in resting conidia compared to mycelium. Enzymes involved in reactive oxygen intermediates (ROI) detoxification, pigment biosynthesis, and conidial rodlet layer formation were highly abundant in A. fumigatus spores and most probably account for their enormous stress resistance. Interestingly, pyruvate decarboxylase and alcohol dehydrogenase were detectable in dormant conidia, suggesting that alcoholic fermentation plays a role during dormancy or early germination. Moreover, we show that enzymes for rapid reactivation of protein biosynthesis and metabolic processes are preserved in resting conidia, which therefore feature the potential to immediately respond to an environmental stimulus by germination. The generated data lay the foundations for further proteomic analyses and a better understanding of fungal pathogenesis. Keywords: Aspergillus fumigatus • human pathogen • data warehouse • conidia, germination • proteome • reference map • two-dimensional gel electrophoresis • virulence factors • allergens
Introduction Aspergillus fumigatus is a filamentous fungus ubiquitously distributed in the environment. In its natural habitat, the soil, it plays an essential role in recycling environmental carbon and nitrogen.1 Although a sexual cycle of A. fumigatus was discovered recently,2 the fungus mainly reproduces via asexual spores. Sporulation occurs abundantly, with thousands of conidia being released into the atmosphere by every conidiophore. The small diameter of the conidia enables them to reach the lung alveoli of humans once they are inhaled. It is estimated that every human inhales at least several hundred A. fumigatus conidia per day. For immunocompetent individuals, this rarely * To whom correspondence should be addressed. Mailing address: Leibniz Institute for Natural Product Research and Infection BiologyHans-Kno¨llInstitute (HKI), Beutenbergstrasse 11a, 07745 Jena, Germany. Phone: +49 3641 532 1071. Fax: +49 3641 532 0803. E-mail:
[email protected]. † Department of Molecular and Applied Microbiology, HKI. ‡ Friedrich Schiller University. § Research Group Systems Biology/Bioinformatics, HKI. | Institut Pasteur. 10.1021/pr9010684
2010 American Chemical Society
has any adverse effect, since the innate immune mechanisms eliminate the conidia relatively efficiently.3 Still, in these humans, A. fumigatus can provoke hypersensitivity disorders such as allergic bronchopulmonary aspergillosis (ABPA)4 or cause aspergilloma, that is, the fungus grows in the lung mainly on the surface of preexisting cavities and forms hyphal balls.5,6 In recent years, A. fumigatus has attracted more and more attention as the major cause of life-threatening lung infections in immunocompromised patients, known as invasive aspergillosis (IA). Since the number of those patients has increased during the last two decades, a dramatic increase of IA has been observed.6,7 Individuals with leukemia or AIDS and patients undergoing stem cell or solid organ transplantations are at particular risk.8 Due to difficult diagnosis, increasing appearance of resistance to antifungal agents and only limited improvements that have been made with preventive strategies and the development of new antifungal drugs, IA is still associated with mortality rates as high as 90%.3,8-11 The mechanisms leading to pathogenicity of A. fumigatus are poorly understood. Although some factors contributing to Journal of Proteome Research 2010, 9, 3427–3442 3427 Published on Web 05/27/2010
research articles virulence in the immunocompromised host have been identified,6,12,13 more molecules, in particular proteins, can be expected to play a role. The recent sequencing and annotation of the A. fumigatus genome14 has paved the way for studying the biology of this fungus on a postgenomic level, for example, by proteomic approaches. Lately, some efforts to characterize the A. fumigatus proteome by 2-D gel electrophoresis and mass spectrometry (MS) have been made.15-18 However, no comprehensive proteome map for conidia has been published yet. The conidium represents the fungal entity having the initial contact with the host immune system; hence, the identification of conidial proteins is of enormous importance to understand the interaction of A. fumigatus with the human host. Proteins present in dormant conidia are presumably required to confer resistance to stress and especially to host defense reactions but may also be immediately needed for the initial exit from dormancy and germination that controls the establishment of the fungus in the lung parenchyma. In a study on the transcriptome of dormant (and germinating) conidia, Lamarre et al. showed that conidia store prepackaged mRNAs of more than 25% of the genes spotted on the array and that early germination was associated to major transcriptional changes.19 To get a closer insight into the protein composition of dormant conidia, we created a 2-D reference map for intracellular conidial proteins from A. fumigatus. This constitutes the first global view on the proteome of resting conidia of an Aspergillus species. Our 2-D PAGE analysis revealed that many enzymes of essential primary metabolic pathways are preserved in dormant conidia and do not have to be synthesized de novo on the onset of germination. In comparison to the mycelial proteome, conidia are specifically enriched in enzymes of the melanin biosynthesis pathway, proteins of the ROI detoxifying system and the conidial hydrophobin RodA. Our study provides additional information for the understanding of conidial dormancy and a framework for further comparative proteomic analyses, such as the identification of fungal allergens as well as virulence determinants.18,20,21
Materials and Methods Strains and Culture Conditions. For all experiments, the A. fumigatus strain ATCC 46645 (LGC Standards, Wesel, Germany) was used.22 Asexual spores of the fungus (conidia) were plated on Aspergillus minimal medium (AMM)23 containing 50 mM glucose and cultivated for three days at 37 °C until sporulation occurred. For mycelium formation, conidia were inoculated at a final concentration of 2 × 106 spores/mL AMM containing 50 mM glucose and shaken at 37 °C with 200 rpm overnight. Extraction of Conidial Cellular Proteins for Gel Electrophoresis. Conidia were harvested from the AMM agar plates, separated from mycelial contaminations by filtration through a 40 µm nylon cell strainer (BD Biosciences, San Jose, CA) and resuspended in 0.9% [w/v] saline. After adding 0.1 volume of glass beads (Ø 0.5 mm), conidia were disrupted in a microdismembrator (Sartorius, Goettingen, Germany) at 2000 rpm for 10 min. For extracting the proteins, we applied trichloroacetic acid (TCA)/acetone precipitation followed by phenol extraction using protocols modified from Isaacson et al.24 In detail, proteins were precipitated by incubating the suspension of disrupted conidia with three volumes of 13.3% [w/v] TCA/ acetone containing 20 mM DTT overnight at -20 °C. The resulting pellet was washed twice with 90% [v/v] acetone containing 20 mM DTT and resuspended in extraction buffer (50 mM Tris-HCl pH 8.8, 5 mM EDTA, 100 mM KCl, 30% [w/v] 3428
Journal of Proteome Research • Vol. 9, No. 7, 2010
Teutschbein et al. Sucrose, 200 mM DTT). Proteins were extracted by Trisbuffered phenol (pH 8) and precipitated by incubating the phenolic phase with five volumes of 100 mM ammonium acetate in methanol overnight at -20 °C. The pellet was washed twice with methanol and once with acetone containing 20 mM DTT and subsequently resuspended in lysis buffer (7 M urea, 2 M thiourea, 2% [w/v] CHAPS, 30 mM Tris, 1% [w/v] Zwittergent, 0.8% [v/v] Pharmalyt 3-10, 20 mM DTT).25 After extraction in an ultrasound water bath for 10 min, followed by a freezing step (1 h at -70 °C) and centrifugation, the supernatant containing the proteins was used for 2-D gel electrophoresis. Protein concentration was determined according to the method of Bradford26 using BIO-RAD protein assay (BIO-RAD Lab., Hartfordshire, U.K.). 2-D Polyacrylamide Gel Electrophoresis (2-D PAGE). For separation in a nonlinear (NL) pH range of 3-11, 300 µg of proteins were loaded by anodic cup-loading on a 24 cm Immobiline DryStrip (GE Healthcare, Freiburg, Germany) that had been rehydrated overnight in 7 M urea, 2 M thiourea, 2% [w/v] CHAPS, 1% [w/v] Zwittergent, 1.2% [v/v] DeStreak Reagent (GE Healthcare), and 0.5% [v/v] IPG Buffer pH 3-11 NL (GE Healthcare). Proteins were separated according to their isoelectric points at 20 °C with a current of 50 µA/strip using the Ettan IPGphor II Isoelectric Focusing System (GE Healthcare). The following protocol was applied: 3 h at 300 V (gradient), 4 h at 600 V (gradient), 4 h at 1000 V (gradient), 4 h at 8000 V (gradient) and 24 000 Vh at 8000 V. For separation in ranges pH 3-7 NL, pH 7-11 NL and pH 6-9, higher protein amounts (500, 450 and 350 µg, respectively) were loaded, the appropriate IPG buffers were used (pH 4-7, pH 7-11 NL or pH 6-11) and isoelectric focusing was run 4 h at 300 V (gradient), 4 h at 600 V (gradient), 4 h at 1000 V (gradient), 6 h at 8000 V (gradient) and 48 000 Vh at 8000 V. Subsequently, proteins in the strip were reduced and alkylated in equilibration buffer (75 mM Tris-HCl pH 8.8, 30% [v/v] glycerol, 6 M urea, 2% [w/v] SDS) containing 1% [w/v] DTT or 2.5% [w/v] iodoacetamide, respectively. The equilibrated strip was transferred onto the top of a lab-cast 26 cm × 20 cm × 1 mm polyacrylamide gel (12.5% [w/v]) containing Rhinohide as gel strengthener (Invitrogen, Paisley, U.K.). Proteins were separated according to their molecular masses in SDS running buffer (25 mM Tris, 192 mM glycine, 0.1% [w/v] SDS) at 1 W/gel for 45 min and 13 W/gel for 4 h using the ETTAN DALTsix or DALTtwelve Large Vertical electrophoresis systems (GE Healthcare). Protein Visualization and Image Analysis. Proteins were visualized by “Blue silver” colloidal Coomassie staining using a protocol from Candiano et al.27 with slight modifications. In brief, gels were gently shaken for 2 h in fixing solution (40% [v/v] methanol, 7% [v/v] acetic acid), subsequently washed 15 min in distilled water, afterward stained overnight in 10% [v/v] o-phosphoric acid, 10% [w/v] ammonium sulfate, 0.12% [w/v] Coomassie Blue G-250 (Roth, Karlsruhe, Germany), and finally washed once in distilled water and twice in 25% [v/v] methanol for 30 min each. Image analysis was performed with Delta2D Version 3.6 (DECODON, Greifswald, Germany). Technical triplicates from each of the three biological replicates were analyzed and a fused image was created for spot detection. After filtering noise spots, spots were quantified using the % volume parameter (%V), that is, the relative volume with respect to all detected spots, and the ratio that compares mean %V of protein spots from biological replicates. Only spots found
Conidial Proteome Map of Aspergillus fumigatus
research articles
reproducibly in the replicate gels (with % volume ratios between 0.1 and 10) were considered in further analyses.
less preferred codons, while a value of 1.0 reflects the maximum codon use fit.
Protein Digestion and Sequence Analysis by MALDI-TOF/ TOF. Coomassie blue stained spots were manually excised from the gels, tryptically digested according to the protocol of Bruker (Bruker Daltonics, Bremen, Germany) based on the method described by Shevchenko et al.28 Extracted peptides were mixed with saturated R-cyano-4-hydroxycinnamic acid in acetonitril/ trifluoric acid 0.1% (1:2 [v/v]) and allowed to dry on an MTP 800/384 anchor chip target (Bruker Daltonics). Samples were measured on a Bruker Ultraflex I matrix-assisted laser desorption ionization - tandem time-of-flight (MALDI-TOF/TOF) device (Bruker Daltonics). MS spectra were identified by searching the NCBI database using the MASCOT interface (MASCOT 2.1.02; Matrix Science, London, U.K.). With respect to the sample preparation, fixed modification of cysteines to S-carbamidomethyl derivatives and variable methionine oxidation were defined for the database search. No missed cleavage was allowed, and peptide mass tolerance was set to 50 ppm. Hits were considered significant according to the MASCOT score (p e 0.05). Database searches were triggered and archived on a Proteinscape 1.3 database server (Bruker Daltonics, Bremen, Germany). Data Warehouse. All spot data were exported as XML-file out of Delta2D and imported together with the gel images into the data warehouse Omnifung (www.omnifung.hki-jena.de).29 Here, the conidial proteome maps build one project together with the recently published maps of Vo¨disch et al.17 This enhances a global view on and allows comparisons of important proteins in different parts of the fungus. All data of identified spots, including protein identification data and Supporting Information of this article, are available via public login (no password needed). Classification of Proteins and CAI Calculation. For classification of identified and predicted proteins two classification systems were used. Gene Ontology (GO) categorization for A. fumigatus proteins provided by the European Bioinformatics Institute (EBI, www.ebi.ac.uk/GOA/proteomes.html) was the first source of information.30 It comprises annotations for the fungal proteome divided into three branches: biological process, molecular function and cellular component. The second information source was the Pedant server of the Munich Information Center for Protein Sequences (MIPS, http://pedant.gsf.de:3345/pedant3htmlview/ pedant3view?Method)analysis&Db)p3_p131_Asp_fumig). Here, proteins of A. fumigatus are categorized according to the Functional Catalogue (FunCat).31 To characterize the gene expression and translation efficiency of identified proteins we calculated the codon adaptation index (CAI). Synonymous codons are not generally used at equal frequencies. The degree of codon bias is associated to the content of the isoacceptor tRNAs and to the level of gene expression. Whereas highly expressed genes tend to mainly use synonymous codons with the most abundant tRNA (preferred or “major” codons), weakly expressed genes show a more frequent use of the unpreferred or “minor” synonymous codons.32 The codon adaptation index (CAI), which can be calculated with the CAIJava program (www.ihes.fr/∼materials/description.html)33 using the A. fumigatus codon usage table from the Japanese Kazusa DNA Research Institute webpage (www.kazusa.or.jp/cgi-bin/ showcodon.cgi?species)330879&aa)1&style)N), estimates the degree of synonymous codon adaptation in a coding region compared with the optimal usage. Lower values indicate use of
2-D Differential Fluorescence Gel Electrophoresis (DIGE). Proteins from resting conidia (RC) were extracted as described above. Mycelium (M) was harvested by filtering through Miracloth (Calbiochem, Edison, NJ) and mycelial proteins were extracted like the conidial proteins. The pH of the samples was adjusted to 8.5 by the addition of 100 mM NaOH. Then, the samples were labeled with CyDye minimal dyes according to the manufacturer’s protocol (GE Healthcare) and as described in Lessing et al.34 Briefly, 50 µg of protein of either sample was labeled with 300 pmol of CyDye (dissolved in dimethyl formamide). Samples from RC and M were labeled either with Cy3 or Cy5. A pool of both samples was prepared, labeled with Cy2, and used as an internal standard. Samples were vortex mixed and incubated for 30 min in the dark on ice. The reaction was stopped by adding 1 µL of 10 mM L-lysine. An equal volume of 4× sample buffer (composition as for the lysis buffer mentioned above) was added. Equal amounts of each of the three label preparations were mixed and applied via anodic cup loading onto a 24 cm Immobiline DryStrip (pH 3-11 NL). Rehydration of strips, isoelectric focusing, equilibration, and electrophoresis were performed as described above. Proteins were visualized by analyzing the gels with a Typhoon 9410 scanner using a resolution of 100 µm. Spot detection of cropped images was performed with the DeCyder software package (version 6.5). The following parameters were applied: detection sensitivity, 2500 spots (excluding filter set); slope, >1.6; area, not filtered; peak height, not filtered; and volume, 0.3. However, no direct correlation between CAI values and protein abundance assessed by spot % volumes was observed. This is probably due to the fact that the level of protein expression is not only determined by the codon usage, but influenced by growth conditions, environmental stimuli and the developmental stage of the organism. For example, three hypothetical proteins that had not been identified in a recent study on the mycelial proteome of A. fumigatus17 were highly abundant in conidia (Table 1) despite their low CAI values. Nevertheless, our global CAI analysis indicates the general tendency that proteins encoded by genes with a higher CAI are more abundant and hence more easily identified on 2-D gels, as previously described in the literature.40,41 Only three proteins with CAIs
