Description of the mechanisms underlying

JO U R N A L OF P ROTE O MI CS ( 20 ) 3–2 Available online at www.sciencedirect.com ScienceDirect www.elsevier.com/locate/jprot Description of the mechanisms underlying geosmin production in Penicillium expansum using proteomics Marc Behr, Tommaso Serchi, Emmanuelle Cocco, Cédric Guignard, Kjell Sergeant, Jenny Renaut, Danièle Evers⁎ Centre de Recherche Public-Gabriel Lippmann, Département Environnement et Agro-biotechnologies, Belvaux, Luxembourg AR TIC LE I N FO ABS TR ACT Article history: A 2D-DIGE proteomics experiment was performed to describe the mechanism underlying the Received 20 March 2013 production of geosmin, an earthy-smelling sesquiterpene which spoils wine, produced by Accepted 24 October 2013 Penicillium expansum The strains were identified by sequencing of the ITS and beta-tubulin regions This study was based on a selection of four strains showing different levels of geosmin production, assessed by GC–MS/MS The proteomics study revealed the differential abundance of Keywords: 107 spots between the different strains; these were picked and submitted to MALDI-TOF–TOF MS Geosmin analysis for identification They belonged to the functional categories of protein metabolism, Penicillium expansum redox homeostasis, metabolic processes (glycolysis, ATP production), cell cycle and cell signalling Proteomics pathways From these data, an implication of oxidative stress in geosmin production may be Oxidative stress hypothesized Moreover, the differential abundance of some glycolytic enzymes may explain the MVA pathway different patterns of geosmin biosynthesis This study provides data for the characterisation of MEP pathway the mechanism and the regulation of the production of this off-flavour, which are so far not described in filamentous fungi Biological significance Green mould on grapes, caused by P expansum may be at the origin of off-flavours in wine These are characterized by earthy–mouldy smells and are due to the presence of the compound geosmin This work aims at describing how geosmin is produced by P expansum This knowledge is of use for the research community on grapes for understanding why these off-flavours occasionally occur in vintages © 2013 Elsevier B.V All rights reserved Introduction Wine is a product for which organoleptic quality is primordial Wine aroma results from the contribution of volatile compounds originating from the grape microflora and winemaking practices [1] Although in most cases aroma compounds confer a special, variety-specific, positive characteristic to a wine, several grape-derived aroma compounds may alter wine aroma in a negative way Over the last years, winegrowers have observed organoleptic defects in wine characterized by mushroom, mouldy, camphoric or earthy odours [2] The risk for such defects is high when grapes are infected by rots Indeed, they are produced by Botrytis cinerea (causal agent of grey mould), Penicillium expansum (causal agent of green mould), a wide- Abbreviations: ITS, internal transcribed spacer (of the rDNA); PTV, programmed temperature vaporization; MEP pathway, methylerythritol phosphate pathway; MVA pathway, mevalonate pathway; PH-like, pleckstrin homology-like ⁎ Corresponding author at: Centre de Recherche Public-Gabriel Lippmann, Département Environnement et Agro-biotechnologies, 41, rue du Brill, 4422 Belvaux Luxembourg Tel.: + 352 47 02 61 441; fax: + 352 47 02 64 E-mail addresses: behr@lippmann.lu (M Behr), serchi@lippmann.lu (T Serchi), cocco@lippmann.lu (E Cocco), guignard@lippmann.lu (C Guignard), sergeant@lippmann.lu (K Sergeant), renaut@lippmann.lu (J Renaut), evers@lippmann.lu (D Evers) 1874-3919/$ – see front matter © 2013 Elsevier B.V All rights reserved http://dx.doi.org/10.1016/j.jprot.2013.10.034 14 JO U R N A L OF PR O TE O MI CS 96 ( 20 ) –28 spread filamentous fungus responsible of fruit decay including grapes [3], or a combination of both [4] The compounds responsible for defects associated with mushrooms are C8 alcohols and ketones such as 1-octen-3-ol and 1-octen-3-one and they have been reported to be metabolites of various fungi; in oenology, 1-octen-3-ol has been associated with the presence of B cinerea on grapes [5] As for the earthy smell, La Guerche et al [2] identified the responsible molecule as (−)-geosmin The olfactory perception threshold for geosmin in wines is 50 ng/L [6] According to La Guerche et al [7], geosmin originates from the metabolism of P expansum on grapes pre-contaminated by B cinerea; thus, B cinerea would induce the production of geosmin by P expansum Little is known about the biosynthesis of geosmin by P expansum Although several papers report about geosmin production in Actinobacteria, certain Cyanobacteria, Myxobacteria and higher Fungi, little is known about the genes implicated in geosmin biosynthesis [8] As suggested by Bentley and Meganathan [9], geosmin would be derived from a sesquiterpenoid precursor and would be synthesized from farnesyl pyrophosphate The use of a geosmin overproducing strain of Streptomyces citreus producing also high levels of the sesquiterpene alcohol germacra-1-E,5E-dien11-ol (germacradienol), indicated that this compound might be a precursor of geosmin [10] Later studies showed that germacradienol production was indeed the committed step in geosmin production [11] According to Gust et al [3] presenting a study on Streptomyces, sesquiterpene synthase would be involved in an early step in geosmin biosynthesis The functional characterization of six sesquiterpene synthases in a basidiomycete called Coprinus cinereus has been described [12] As previously said, concerning P expansum, few data are available on geosmin synthesis in must or culture medium Dionigi [13] has studied the impact of copper sulphate addition in Czapek medium on the biosynthesis of geosmin Recently, a cytochrome P450 monooxygenase gene, gpe1, which may intervene during the transformation of farnesyl pyrophosphate to geosmin, has been described [14] Proteomics techniques are used more and more to unravel proteins implicated in different pathways [15] Here we report a comparative proteomic study aimed at identifying differentially expressed proteins in four P expansum strains producing geosmin in different amounts in culture medium Some proteins putatively implicated in geosmin production were identified and their possible implication in geosmin biosynthesis is discussed Material and methods 2.1 Cultivation of strains The potential of geosmin production was assessed in triplicate on fourteen Penicillium strains isolated in vineyards in the luxembourgish part of the Moselle valley Picking was done on mature berries between 2007 and 2010 Cultures were conducted on malt-agar medium in Petri dishes at 25 °C, based on initial monoconidia production The same conditions were applied for the assessment of geosmin production and the proteomic studies: 50 mL of malt-peptone media were inoculated with 50 μL of the conidia suspension (106 conidia/mL) in a 250 mL Erlenmeyer flask, plugged with cotton and placed at 120 rpm, at room temperature, during days Geosmin quantification was done in triplicate and protein extraction was done in quadruplicate 2.2 DNA extraction Pure isolates were sub-cultured in Potato Dextrose Broth (PDB) during one week at 25 °C on a rotary shaker set at 120 rpm The subsequent biomass was lyophilised prior to grinding with metallic beads DNA was extracted with the DNeasy Plant Mini Kit (Qiagen, Hilden, Germany), following instructions of the manufacturer DNA concentration was quantified with the NanoDrop (ND 1000, Thermo Scientific, Waltham, MA) The samples were stored at − 20 °C 2.3 Molecular identification of strains All the strains were identified with four sets of primers: β-tubulin BT1 (a and b), β-tubulin BT2 (a and b) [16], ITS1–ITS4 and ITS U5–ITS R2 [17] The PCR reaction mixture contained 10 μL of Finnzymes Taq Phusion-buffer Mastermix HF (Thermo Scientific, Waltham, MA); μL of each primer (10 μM); μL of DNA (100 ng/μL) and μL of UltraPure™ DNase/RNase-free distilled water (Invitrogen, Paisley, UK) for a final volume of 20 μL Amplification was performed on a Biometra T-professional thermocycler (Biometra, Goettingen, Germany) using the following programme: an initial denaturation at 98 °C (2 min) followed by 30 cycles with denaturation at 98 °C (15 s), annealing at 67 °C (β-tubulin) or 63 °C (ITSU5–ITSR2) or 64 °C (ITS1–ITS4) during 20 s and elongation at 72 °C during 20 s (β-tubulin, ITS1–ITS4) or 10 s (ITSU5–ITSR2); the final elongation was performed at 72 °C during 10 The size, quality and quantity of the amplicons were checked on 3% w/v agarose gel stained with ethidium bromide (1 h; 100 V) under UV transillumination PCR products were diluted in UltraPure™ DNase/RNasefree distilled water (Invitrogen, Paisley, UK) to reach a concentration of 10–20 ng/μL The sequencing PCR was achieved with these dilutions using Big Dye products (Applied Biosystems, Carlsbad, CA): × sequencing buffer (2 μL), Big Dye sequencing RR-100 (2 μL), primer 10 μM (0.32 μL), μL of the diluted PCR amplicons and 14.68 μL of UltraPure™ DNase/ RNase-free distilled water (Invitrogen, Paisley, UK) for a total volume of 20 μL Both strands of each amplicon were sequenced The PCR programme consisted in an initial denaturation at 96 °C (1 min) followed by 25 cycles of denaturation at 96 °C (10 s) and elongation at 60 °C (4 min) The PCR products were cleaned with the BigDye Xterminator Purification kit (Applied Biosystems, Carlsbad, CA) Sequencing was done on the Applied Biosystems 3130 Genetic Analyzer (Applied Biosystems) 2.4 Geosmin measurement Post culture geosmin abundance of fourteen strains was measured on sterile-filtered media (cellulose acetate membrane, 0.2 μm, Sartorius, Goettingen, Germany) When it was required, the cultivation medium was diluted up to 50 fold with fresh sterile malt-peptone broth 2-Ethoxy-3-isopropylpyrazine (IPEP) (TCI, Tokyo, Japan) was used as internal standard Geosmin standard [(±)-Geosmin solution, 100 μg/mL in methanol] was JO U R N A L OF P ROTE O MI CS ( 20 ) 3–2 obtained from Sigma-Aldrich (St Louis, MO) The determination of geosmin was performed by GC–MS/MS using a Trace GC Ultra coupled to a TSQ Quantum XLS tandem mass spectrometer (Thermo Scientific, Waltham, MA) Geosmin was preconcentrated using Headspace Solid-Phase MicroExtraction (HS-SPME) on a DVB/CAR/PDMS fibre (Supelco, SigmaAldrich, St Louis, MO) The extraction was fully automated using a PAL Combi-xt autosampler with the following programme: incubation at 70 °C during min, adsorption at 70 °C during 20 and desorption in the GC injector during Injection was done in PTV splitless mode running the following programme: 56 °C during 0.05 — ramp of 14.5 °C/s until 270 °C and hold at 270 °C for 17 The column was an Rxi-5Sil MS (20 m ∗ 0.18 mm ∗ 0.18 μm, Restek, Bellefonte, PA) and the GC oven was programmed as follows: 50 °C during min, ramp of 15 °C/min until 100 °C and 30 °C/min until 250 °C (hold for 4.7 min) Retention times of geosmin and IPEP were respectively 8.38 and 10.07 A first, semi-quantitative approach, based on the calculation of the geosmin to internal standard ratio, was used to select four strains within the fourteen, i.e P4, P8, P21 and P23 (Fig 1) P4 and P8 were selected as low geosmin producers whereas P21 and P23 were considered as high geosmin producers These four strains were identically resubmitted to cultivation and geosmin assessment to obtain quantitative data (using a calibration curve with geosmin standard) The case of P8 was ambiguous: its production during the second round was much higher than during the first one 2.5 Protein extraction and quantification Unless stated otherwise, all reagents used for extraction and subsequent separation were purchased from GE Healthcare 15 (GE Healthcare, Little Chalfont, UK) Mycelia which were used for protein extraction were the same than those produced for geosmin assessment Mycelia were removed from growing medium and dried on cellulose acetate membrane Afterwards mycelia were ground in liquid nitrogen and proteins were precipitated by addition of ice cold TCA/Acetone/DTT (20%/79.9%/0.1% v/v) overnight at − 20 °C The mixture was centrifuged at 30,000 g for 45 at °C The precipitate was washed three times with rinsing buffer (TCA 20%, acetone 80% v/v) and dried in vacuo Pellets were solubilised in mL of lysis buffer (Urea M, Thiourea M, CHAPS 4%, TRIS 30 mM and proteases inhibitors) for about 30 mg of mycelium Quantification of the extracted proteins was achieved by the Bradford method using a DU 800 spectrophotometer (Beckman Coulter, Brea, CA) 2.6 Labelling of proteins and 2D electrophoresis Prior to labelling, the pH of each sample was checked and, if necessary, adjusted to pH 8.5 30 μg of proteins were labelled and separated by 2D-DIGE as reported previously [18] Briefly, 240 pmol of dyes were incubated with the proteins for 30 on ice in the dark The reaction was then stopped by addition of μL of 10 mM lysine solution and incubation on ice for 10 additional minutes Samples were labelled either with Cy3 or Cy5; the internal standard, constituted by an equal amount of each sample, was labelled with Cy2 Dye swap was performed by labelling, in each group, biological replicates with Cy3 and the other biological replicates with Cy5 After labelling, one Cy3 labelled sample was combined with one Cy5 labelled sample and with the Cy2 labelled standard: the mixture was diluted to 450 mL with a solution containing Urea M, Thiourea M, CHAPS 4%, TRIS 30 mM, μL of Fig – Preliminary semi-quantitative assessment of geosmin production of 14 strains of Penicillium expansum by GC–MS/MS Values represent the geosmin/internal standard ratio ± SD n = biological replicates 16 JO U R N A L OF PR O TE O MI CS 96 ( 20 ) –28 Bio-Lyte pH 3–10 ampholyte buffer (Bio-Rad, Hercules, CA) and 2.7 μL of destreak reagent (GE Healthcare) and traces of bromophenol blue 24 cm pH 3–10 non-linear strips (Readystrip IPG, BioRad) were passively rehydrated at room temperature overnight and focused at 20 °C in an Ettan IPGphor III (GE Healthcare) system until reaching approximately 100 kVh Following the first dimension, equilibration of the strips was carried out firstly during 15 in the equilibration solution (provided by Serva, Heidelberg, Germany) containing 1% w/v DTT and then in equilibration solution supplemented with 2.5% w/v iodoacetamide solution (w/v) for 15 The second dimension was obtained by a run on a 12.5% pre-cast polyacrylamide gel (Serva, Heidelberg, Germany) Obtained gels were scanned at a spatial resolution of 100 μm with a 9400 Typhoon (GE Healthcare) using the following wavelengths: excitation at 488 nm, 532 nm, and 633 nm (Cy2, Cy3 and Cy5, respectively) and emission at 520 nm, 610 nm and 670 nm (Cy2, Cy3 and Cy5, respectively) Images of the gels were analysed by DeCyder 2D Differential Analysis v.7.0 software (GE Healthcare) Maps were calibrated, to obtain experimental molecular weight and isoelectric point estimations, using the Fusarium reference map which was produced earlier in our laboratory [19] Highlighted proteins of interest (fold change ±1.3; t-test ≤0.05) were picked, trypsin digested for h at 37 °C and then spotted on MALDI disposable targets by the Ettan Spot Handling workstation (GE Healthcare) Identification of the proteins was carried out using an AB SCIEX TOF/TOF 5800 System (AB SCIEX, Framingham, MA) MS spectra were internally calibrated using trypsin autocleavage signals In MS/MS mode an external calibration using fragmentation products of Glu-fibrinopeptide was done For MS, the recorded spectrum was the accumulation of 1500 shots, for MS/MS this was 3000 Spectra were acquired using an automated approach defined in the MALDI software (TOF/TOF Series Explorer™ V4.1.0, AB Sciex); for each spot the highest peaks in the raw MS spectrum were selected for fragmentation after exclusion of common contaminants, for instance peaks from trypsin autocleavage products or keratin Proteins were identified by searching with the MASCOT algorithm version 2.3 (Matrix Science, www.matrixscience com, London, UK) against the NCBI database (updated to the 18th of January 2013, with 1,585,852 sequences belonging to “Other Fungi”), using ProteinPilot™ Software version 4.0 (AB SCIEX) Searches were carried out allowing a mass window of 100 ppm for the precursor and 0.5 Da for fragment ion masses The search parameters allowed maximum two missed cleavages; carbamidomethylation of cysteine as fixed modification; oxidation of methionine and oxidation of tryptophan (single oxidation, double oxidation and kynurenine) as variable modifications Proteins with probability-based MOWSE scores (p ≤ 0.05) were considered to be successfully identified Results 3.1 Identification of the strains Consensus sequences have been produced and compared to reference strains through BLAST Each strain has been successfully amplified and sequenced by each primer set They were all identified as P expansum with an E value of and 100% of identity The amplicon sizes were 484 bp (BT1), 477 bp (BT2), 716 bp (ITS1–4) and 280 bp (ITS U5-R2) (Behr et al., Journal International des Sciences de la Vigne et du Vin, accepted manuscript) 3.2 Geosmin production of the strains In order to select the appropriate strains for the proteomic investigations, fourteen isolates were submitted to a preliminary geosmin assessment (Fig 1) Based on this screening, four strains, namely P4, P8, P21 and P23, were selected P4 and P8 were supposed to be low geosmin producers, while P21 and P23 exhibited a much higher production during the preliminary screening During the second assessment (Fig 2), a similar geosmin production was measured with the exception of P8, showing a production comparable to P21 and P23 Malt-peptone broth was found to be a very efficient medium for the induction of geosmin production The determination of geosmin content is made easier by the liquid form of the medium, which allows a direct extraction by HS-SPME, without the extraction step required by culture on solid media [2] As compared to similar studies realised on other medium (grape juice or malt-agar), the production was much higher Indeed, Morales-Valle et al [4] have described a maximum concentration around 600 ng/L; La Guerche et al [20] have reported, under the same conditions, a maximum concentration of 500 ng/L, while we have reached an average concentration superior to 5000 ng/L for P23 Usually, concentrations reached in wine are lower, generally around 100 ng/L, up to 300 ng/L [21,22] Despite the first objective to select two strains of each phenotype, we decided to keep P8 for the proteomic study, since the examination of such a profile may be interesting 3.3 Differentially expressed proteins 107 proteins were differentially expressed and were used to a Principal Component Analysis (PCA, Fig 3) and a Hierarchical Clustering (HC, Fig 4) A typical gel with differentially expressed proteins is shown (Fig 5) Information concerning the fold-change of the proteins is presented in Table In supplementary data S1, a summary of all relevant information for each identified protein, such as possible Fig – Quantitative assessment of geosmin production (ng/L of media ± SD) for the strains used in the proteomic study n = biological replicates JO U R N A L OF P ROTE O MI CS ( 20 ) 3–2 involvements of the proteins in metabolism as suggested by KEGG database http://www.genome.jp/kegg/pathway.html, is reported In supplementary data S2, detailed information about the identification of the differentially expressed proteins, including peptide mass fingerprinting (PMF) and MS/MS fragmentation data can be found 3.3.1 Clustering of the strains Only the proteins which have exhibited a fold change of at least ± 1.3 and a p value below 0.05 and resulted in a single identification are presented in Table Most of the proteins were presenting multiple isoforms with sometimes different behaviours in their relative fold-change It is clear that the two high geosmin producing strains P21 and P23 can be distinguished neither in the PCA nor in the HC, while the low geosmin producer P4 is well separated from the other strains The strain P8 was not consistent in its production of geosmin, so that we could not clearly put it in one of the two categories, and this can be seen in the proteomic profile, since in the PCA and in the HC this strain was separated from high producers and the low producer Hereafter, the classification of the differentially regulated proteins into different functional categories will be discussed 3.3.2 Proteins involved in metabolic processes Several proteins involved in metabolic processes were found to be differentially abundant between the four strains Some of the proteins are involved in carbohydrate catabolism, including enolase and phosphoglycerate kinase (PGKA) Spots containing these proteins did not show a uniform trend within the groups Several of them were more abundant in the strain P4 (3 isoforms of phosphoglycerate kinases PGKA), while those containing enolase were less abundant Fourteen isoforms of enolases were found; nine of them were found in their full length (spots n° 1638, 1643, 1655, 1656, 1663, 1667,1674, 1680 and 17 1686; 43 kDa, pI from 4.89 to 5.20) while the others were found to be degradation/processing products of enolase Spot n° 2340 contains the original N-terminus, in all the spots (spots n° 2767, 2768, 2800, 3174 and 4580) the original C-terminus was identified The peaks were extracted from the spectra and used for classification with Speclust (http://bioinfo.thep lu.se/speclust.html) [23] With the peaks-in-common tool, those peaks that were unique to each spectrum were isolated and studied Because the spectra were generally of low intensity, no differences between the molecular forms at the same molecular weight could be found nor could cleavage sites resulting in the observed forms be discerned The nine isoforms presenting an intact form did not display significant differences Four degraded forms were more abundant in the geosmin producing strains Importantly, acetyl-CoA C-acetyltransferase, the enzyme which constitutes the beginning of secondary metabolism starting from acetyl-CoA [24] was less abundant in P4 This enzyme catalyses the transfer of an acetyl group into acetyl-CoA, producing acetoacetyl-CoA Acetyl-CoA is mainly produced via pyruvate from the glycolysis, or by the β-oxidation of the fatty acids Acetoacetyl-CoA is the starting point for the synthesis of farnesyl diphosphate, the common molecule of pathways leading to sesquiterpenoid products, including geosmin An overview of the glycolytic, methylerythritol phosphate (MEP) and mevalonate (MVA) pathways is shown (Fig 6) Expressions of the ATP synthase enzymes were also changing from one isoform to the other: two were much more abundant in P4 and six were less abundant Other proteins related to ATP synthesis were also differentially abundant: two cytochrome C oxidases (subunit 5a) were less abundant in P4 Cytochrome C oxidase is involved in the ATP production by the mitochondrial respiratory chain Two enzymes may be considered both as metabolic and linked to phytopathogenicity: serine carboxypeptidase and proteinase A They are involved in Fig – PCA analysis of the strains (n = biological replicates) used in proteomic studies Left panel: score plot of spot maps Right panel: loading plot of proteins 18 JO U R N A L OF PR O TE O MI CS 96 ( 20 ) –28 Fig – Hierarchical clustering of four strains of Penicillium expansum Red, yellow, green and blue dots represent P4, P8, P21 and P23 respectively Pearson correlation coefficient was used in order to achieve the clustering nitrogen metabolism and also in the lysis of molecules encountered during infection of vegetal host cells (destructuration of cell wall, reaction to PR proteins) [25,26] The three spots corresponding to proteinase A were less abundant in P4, as were two spots containing serine carboxypeptidases In contrast, three other spots containing serine carboxypeptidases were significantly more abundant in P4 3.3.3 Proteins involved in protein synthesis and folding Enzymes involved in protein synthesis and folding were also a major point of differentiation between the groups One protein involved in DNA transcription (a nucleic acid binding protein) was more abundant in P4 Concerning protein synthesis, ribosomal protein S2 (RPS2) and one translation elongation factor containing a glutathione S-transferase Fig – Representative gel of Penicillium expansum proteome Whole protein extracts were labelled with CyDyes and separated in first dimension by 24 cm 3–10 non-linear strips and in second dimension by 12.5% polyacrylamide precast gels The numbers of picked spots are reported The presented image is the standard (Cy2 channel) of the gel which was used as master gel in the experiment Additional images, one representative of each experimental group, are presented in supplementary material S3 19 JO U R N A L OF P ROTE O MI CS ( 20 ) 3–2 Table – Differentially abundant proteins identified by MALDI-MS in the four Penicillium expansum strains Theor MW/pI — theoretical molecular weight (expressed Da) and isoelectric point (expressed in pH units), Exp MW/pI — experimental molecular weight (expressed in Da) and isoelectric point (expressed in pH units); P8/P4 column and followings: fold change relative to the reported strains with the corresponding p-value are reported: positive values (reported in green) when the numerator is up-regulated – negative values (reported in red) when the denominator is up-regulated UniProt access No Protein Existance (UniProt) Spot No Theor MW/pI Exp MW/pI Protein name / function 799 63209/ 4.52 66680/ 4.53 Amidase family protein K9GF02 Predicted 1638 47250/ 5.26 43255/ 5.15 Enolase BAC82549 B6H602 1643 47250/ 5.26 43322/ 5.2 Enolase BAC82549 1655 47250/ 5.26 47055/ 4.95 1656 47250/ 5.26 1663 P8/P4 P21/P4 P23/P4 P21/P8 P23/P8 P23/P21 –1.32; 0.014 1.04; 0.75 1.12; 0.053 1.38; 0.036 1.48; 0.0027 1.07; 0.43 Inferred from homology 1.00; 0.89 –1.19; 0.21 –1.51; 0.028 –1.19; 0.15 –1.52; 0.018 –1.27; 0.21 B6H602 Inferred from homology 1.00; 0.96 –1.21; 0.11 –1.43; 0.036 –1.22; 0.087 –1.43; 0.029 –1.18; 0.31 Enolase BAC82549 B6H602 Inferred from homology 1.32; 0.10 –1.26; 0.080 –1.29; 0.10 –1.66; 0.0036 –1.70; 0.0073 –1.02; 0.72 43054/ 5.05 Enolase BAC82549 B6H602 Inferred from homology 1.04; 0.67 –1.29; 0.15 –1.74; 0.0018 –1.34; 0.10 –1.82; 0.00098 –1.35; 0.16 47250/ 5.26 42392/ 4.73 Enolase BAC82549 B6H602 Inferred from homology 1.28; 0.047 –1.12; 0.17 –1.08; 0.50 –1.44; 0.0036 –1.39; 0.044 1.04; 0.88 1667 47250/ 5.26 42656/ 5.30 Enolase BAC82549 B6H602 Inferred from homology 1.05; 0.42 –1.12; 0.28 –1.30; 0.017 –1.18; 0.11 –1.37; 0.0034 –1.15; 0.25 1674 47250/ 5.26 42195/ 4.86 Enolase BAC82549 B6H602 Inferred from homology 1.09; 0.31 –1.18; 0.16 –1.43; 0.035 –1.29; 0.057 –1.56; 0.016 –1.21; 0.26 1677 47250/ 5.26 41869/ 4.61 Enolase BAC82549 B6H602 Inferred from homology 1.52; 0.0047 1.09; 0.29 1.06; 0.51 –1.39; 0.0097 –1.43; 0.0 093 –1.03; 0.71 1680 47250/ 5.26 42064/ 4.83 Enolase BAC82549 B6H602 Inferred from homology 1.17; 0.12 –1.22; 0.084 –1.39; 0.024 –1.43; 0.0098 –1.63; 0.0043 –1.14; 0.31 1686 47250/ 5.26 42260/ 4.78 Enolase BAC82549 B6H602 Inferred from homology 1.20; 0.21 –1.20; 0.23 –1.47; 0.054 –1.44; 0.015 –1.77; 0.0072 –1.23; 0.18 1693 44107/ 5.98 41869/ 5.51 Phosphoglycerate kinase pgkA B6H903 Predicted –1.12; 0.41 –1.50; 0.015 –1.47; 0.029 –1.34; 0.076 –1.31; 0.12 1.02; 0.93 2340 47250/ 5.26 30182/ 4.81 Enolase BAC82549 B6H602 Inferred from homology 1.46; 0.0096 1.72; 0.00080 1.62; 0.045 1.18; 0.15 1.11; 0.72 –1.06; 0.61 2361 34281/ 5.44 29948/ 4.99 B6GYI7 Inferred from homology 1.06; 0.54 1.50; 0.018 1.34; 0.11 1.41; 0.033 1.26; 0.20 –1.12; 0.50 2497 35821/ 8.44 28366/ 5.73 Malate dehydrogenase B6HDG8 Inferred from homology –1.05; 0.60 1.22; 0.088 1.24; 0.022 1.29; 0.083 1.31; 0.039 1.01; 0.82 2540 35821/ 8.44 27202/ 5.43 Malate dehydrogenase B6HDG8 Inferred from homology –1.43; 0.058 1.32; 0.10 1.23; 0.17 1.88; 0.027 1.76; 0.056 –1.07; 0.76 2767 47250/ 5.26 23768/ 4.53 Enolase BAC82549 B6H602 Inferred from homology 1.41; 0.024 1.45; 0.045 1.53; 0.0042 1.02; 0.95 1.08; 0.54 1.06; 0.65 2768 47250/ 5.26 23731/ 4.58 Enolase BAC82549 B6H602 Inferred from homology 1.22; 0.16 1.42; 0.0023 1.40; 0.00023 1.16; 0.23 1.15; 0.21 –1.02; 0.86 2769 41175/ 5.93 23548/ 4.65 Acetyl–CoA C– acetyltransferase B6HV94 Inferred from homology 1.16; 0.43 1.41; 0.011 1.49; 0.00089 1.22; 0.22 1.29; 0.096 1.06; 0.47 2787 44107/ 5.98 23621/ 7.33 Phosphoglycerate kinase B6H903 Inferred from homology –1.07; 0.66 –1.96; 0.012 –2.07; 0.0034 –1.83; 0.023 –1.93; 0.0078 –1.06; 0.86 2794 44107/ 5.98 23548/ 7.20 Phosphoglycerate kinase B6H903 Inferred from homology –1.05; 0.78 –1.72; 0.020 –1.71; 0.019 –1.64; 0.017 –1.63; 0.016 1.01; 0.97 2800 47250/ 5.26 23402/ 4.49 Enolase BAC82549 B6H602 Inferred from homology –1.64; 0.058 2.41; 0.00071 2.04; 0.029 3.95; 0.0012 3.34; 0.0080 –1.18; 0.41 2812 27620/ 5.45 23221/ 4.75 Triosephosphate isomerase K9GAV3 Inferred from homology –1.02; 0.74 1.46; 0.024 1.47; 0.0024 1.50; 0.20 1.50; 0.049 1.00; 0.88 Glucose / TCA cycle metabolism Glyoxysomal and mitochondrial malate dehydrogenase 20 JO U R N A L OF PR O TE O MI CS 96 ( 20 ) –28 Table (continued) Spot No Theor MW/pI Exp MW/pI Protein name / function UniProt access No Protein Existance (UniProt) P8/P4 P21/P4 P23/P4 P21/P8 P23/P8 P23/P21 2917 44107/ 5.98 22165/ 7.22 Phosphoglycerate kinase B6H903 Inferred from homology 1.03; 0.73 –1.62; 0.020 –1.48; 0.066 –1.68; 0.0031 –1.53; 0.022 1.10; 0.65 3131 44142/ 6.08 19823/ 6.61 Phosphoglycerate kinase K9H8W2 Inferred from homology –1.34; 0.064 –1.56; 0.010 –1.68; 0.0056 –1.17; 0.40 –1.26; 0.23 –1.08; 0.65 3174 47250/ 5.26 19307/ 4.53 Enolase BAC82549 B6H602 Inferred from homology 1.27; 0.25 –1.28; 0.12 –1.24; 0.082 –1.63; 0.041 –1.58; 0.031 1.03; 0.72 4042 35821/ 8.44 12389/ 5.94 Malate dehydrogenase B6HDG8 Inferred from homology –1.26; 0.061 1.17; 0.040 1.14; 0.46 1.48; 0.0096 1.44; 0.073 –1.02; 0.76 4318 36149/ 6.23 10642/ 4.40 Glyceraldehyde–3– phosphate dehydrogenase B6HI59 Inferred from homology 1.40; 0.020 1.62; 0.00053 1.59; 0.0056 1.16; 0.23 1.13; 0.40 –1.02; 0.82 4580 47250/ 5.26 9269/4 95 Enolase BAC82549 B6H602 Inferred from homology 1.32; 0.0010 1.52; 0.00022 1.79; 1.2e–006 1.15; 0.047 1.36; 0.00016 1.18; 0.010 4703 35821/ 8.44 28944/ 5.68 Malate dehydrogenase B6HDG8 Inferred from homology –1.05; 0.64 1.92; 0.0042 1.65; 0.0027 2.02; 0.0070 1.74; 0.0077 –1.16; 0.41 ATP synthesis 382 69037/ 4.47 83112/ 4.51 Lysophospholipase1 K9G2P1 Predicted –1.21; 0.25 2.93; 8.9e–005 3.17; 2.9e–005 3.56; 0.00036 3.85; 0.00020 1.08; 0.50 1335 55237/ 5.25 50357/ 4.25 F0F1 ATP synthase subunit beta B6HI25 Inferred from homology –8.15; 3.7e–006 –7.52; 1.1e–006 –7.94; 7.2e–005 1.08; 0.41 1.03; 0.95 –1.06; 0.63 1437 55237/ 5.25 47770/ 4.28 F0F1 ATP synthase subunit beta B6HI25 Inferred from homology –5.00; 0.00021 –2.75; 0.0023 –3.08; 0.0061 1.82; 0.022 1.62; 0.19 –1.12; 0.54 1525 55237/ 5.25 46311/ 4.74 F0F1 ATP synthase subunit beta B6HI25 Inferred from homology 1.33; 0.053 1.37; 0.036 1.36; 0.0072 1.03; 0.85 1.02; 0.78 –1.01; 0.97 1561 55237/ 5.25 44274/ 4.83 F0F1 ATP synthase subunit beta B6HI25 Inferred from homology 1.41; 0.00026 1.09; 0.56 –1.03; 0.65 –1.29; 0.056 –1.45; 0.0042 –1.13; 0.45 2254 53763/ 5.44 31668/ 4.66 F0F1 ATP synthase subunit beta Q0CFC5 Inferred from homology 1.60; 0.0014 1.25; 0.013 1.18; 0.25 –1.28; 0.017 –1.35; 0.046 –1.06; 0.54 2257 55237/ 5.44 31718/ 4.72 F0F1 ATP synthase subunit beta B6HI25 Inferred from homology 1.42; 0.00044 1.24; 0.14 1.14; 0.26 –1.15; 0.27 –1.25; 0.071 –1.09; 0.61 2976 55237/ 5.26 21455/ 5.27 F0F1 ATP synthase subunit beta B6HI25 Inferred from homology –1.32; 0.013 –1.13; 0.20 –1.09; 0.34 1.17; 0.11 1.22; 0.044 1.04; 0.63 3910 55237/ 5.25 13162/ 4.49 F0F1 ATP synthase subunit beta B6HI25 Inferred from homology 1.46; 0.051 1.39; 0.021 1.46; 0.0055 –1.05; 0.72 1.00; 0.93 1.05; 0.50 3919 16650/ 8.44 13020/ 4.40 F0F1 ATP synthase subunit beta B6HSK3 Inferred from homology 2.99; 6.3e–007 –1.06; 0.15 1.02; 0.98 –3.18; 3.1e–007 –2.94; 0.00011 1.08; 0.64 3924 17127/ 6.73 13020/ 6.10 Ribose 5–phosphate isomerase B B6HSH9 Predicted –2.27; 0.0039 –1.85; 0.013 –2.26; 0.0034 1.23; 0.36 1.01; 0.97 –1.22; 0.36 4169 17966/ 6.17 11662/ 4.35 Cytochrome c oxidase subunit 5a B6HI83 Predicted 1.73; 0.00023 1.16; 0.071 1.32; 0.0052 –1.48; 0.0014 –1.30; 0.012 1.14; 0.14 4177 17966/ 6.17 11411/ 4.41 Cytochrome c oxidase subunit 5a B6HI83 Predicted 1.43; 0.018 1.36; 0.021 1.27; 0.12 –1.06; 0.64 –1.13; 0.37 –1.07; 0.55 21 JO U R N A L OF P ROTE O MI CS ( 20 ) 3–2 Table (continued) Spot No Theor MW/pI Exp MW/pI Protein name / function UniProt access No Protein Existance (UniProt) P8/P4 P21/P4 P23/P4 P21/P8 P23/P8 P23/P21 Protein metabolism 509 67583/ 5.08 77749/ 4.87 Serine carboxypeptidase B6HNT3 Predicted –2.09; 0.00011 –1.35; 0.047 –1.38; 0.019 1.55; 0.020 1.51; 0.012 –1.02; 0.91 512 67583/ 5.08 77388/ 4.83 Serine carboxypeptidase B6HNT3 Predicted –2.26; 0.0014 –1.52; 0.015 –1.45; 0.013 1.49; 0.032 1.55; 0.014 1.04; 0.66 524 67583/ 5.08 76197/ 4.79 Serine carboxypeptidase B6HNT3 Predicted –1.90; 0.0057 –1.23; 0.23 –1.23; 0.24 1.54; 0.0065 1.54; 0.012 1.00; 0.97 983 67583/ 5.08 60005/ 4.06 Serine carboxypeptidase B6HNT3 Predicted –1.09; 0.44 1.34; 0.027 1.38; 0.0077 1.46; 0.0083 1.50; 0.0020 1.03; 0.70 1189 56761/ 4.55 54250/ 4.47 Protein Disulfide Isomerase (PDIa) K9GF84 Inferred from homology –1.39; 0.050 –1.00; 0.94 –1.13; 0.41 1.39; 0.018 1.23; 0.064 –1.13; 0.24 1539 61928/ 4.99 45598/ 4.56 Carboxypeptidase Y, putative K9GGQ9 Predicted 1.54; 0.0085 1.57; 0.0086 1.56; 0.0078 1.02; 0.87 1.01; 0.91 –1.01; 0.95 1555 40215/ 4.53 45316/ 4.50 UV excision repair protein RAD23 B6HH40 Predicted –1.01; 0.94 1.38; 0.027 1.44; 0.016 1.40; 0.029 1.45; 0.017 1.04; 0.59 1900 24487/ 5.24 37502/ 4.68 hypothetical protein PDIP_62460 K9FL12 Predicted 1.45; 0.0082 1.23; 0.11 1.28; 0.011 –1.18; 0.23 –1.13; 0.28 1.04; 0.65 1927 32517/ 4.79 37155/ 4.60 Ribosomal protein S2 (RPS2) B6H2I7 Inferred from homology 1.44; 0.012 1.66; 0.0049 1.77; 0.00045 1.16; 0.23 1.23; 0.021 1.06; 0.44 1984 28352/ 5.79 35853/ 5.78 Nucleic acid binding protein B6HQE9 Predicted –1.20; 0.26 –1.56; 0.068 –1.87; 0.0035 –1.30; 0.23 –1.56; 0.024 –1.20; 0.55 2104 43589/ 5.08 34169/ 4.48 Proteinase A – saccharopepsin B6H445 Inferred from homology –1.43; 0.13 1.32; 0.18 1.51; 0.052 1.90; 0.0019 2.17; 9.3e–005 1.14; 0.30 2702 25049/ 4.44 24785/ 4.38 Elongation factor, C – terminal, alpha helical domain of the GST family B6H4G8 Inferred from homology 2.00; 0.10 3.35; 0.0040 3.51; 0.0025 1.68; 0.098 1.75; 0.066 1.05; 0.76 2879 43589/ 5.08 22617/ 4.53 Proteinase A – saccharopepsin B6H445 Inferred from homology 1.77; 0.00049 1.58; 0.00077 1.48; 0.0031 –1.12; 0.25 –1.20; 0.11 –1.07; 0.47 3247 43589/ 5.08 18429/ 4.45 Proteinase A – saccharopepsin B6H445 Inferred from homology 1.80; 0.022 1.62; 0.0057 1.86; 0.00071 –1.11; 0.69 1.03; 0.72 1.15; 0.23 4706 78775/ 5.58 16975/ 6.56 Oligopeptidase B6HF99 Predicted 1.10; 0.45 –1.27; 0.11 –1.50; 0.0098 –1.39; 0.016 –1.65; 0.00094 –1.19; 0.065 Protein folding 940 69692/ 5.03 62378/ 5.02 Heat shock 70kDa protein 1/8 B6HPY0 Inferred from homology 1.03; 0.69 1.41; 0.024 1.36; 0.036 1.36; 0.024 1.32; 0.038 –1.03; 0.81 1302 61986/ 5.61 51144/ 4.84 Chaperonin HSP60 B6H9L7 Inferred from homology 1.40; 0.021 1.33; 0.071 1.54; 0.0095 –1.05; 0.61 1.10; 0.29 1.15; 0.22 2355 66776/ 5.22 29948/ 4.66 Heat shock protein 70 F2SIL3 Inferred from homology 1.58; 0.0072 1.96; 0.00035 2.27; 0.042 1.24; 0.087 1.44; 0.38 1.16; 0.88 2683 67030/ 5.32 24978/ 4.46 Heat shock 70kDa protein 1/8 B6HVA2 Inferred from homology 1.65; 0.011 1.75; 0.0012 1.76; 0.0043 1.06; 0.56 1.06; 0.66 1.00; 0.95 (continued on next page) 22 JO U R N A L OF PR O TE O MI CS 96 ( 20 ) –28 Table (continued) Spot No Theor MW/pI Exp MW/pI Protein name / function UniProt access No Protein Existance (UniProt) P8/P4 P21/P4 P23/P4 P21/P8 P23/P8 P23/P21 2888 73741/ 4.81 22407/ 4.66 Heat shock 70kDa protein B6H0S5 Inferred from homology 1.42; 0.0096 –1.09; 0.16 –1.06; 0.34 –1.54; 0.014 –1.51; 0.0097 1.02; 0.79 2901 22046/ 4.50 22303/ 4.35 Hsps_p23–like protein B6HKR2 Predicted 1.55; 0.023 1.22; 0.11 1.52; 0.044 –1.27; 0.080 –1.02; 0.86 1.25; 0.17 3319 67030/ 5.32 18005/ 5.10 Heat shock 70kDa protein 1/8 B6HVA2 Inferred from homology –1.04; 0.89 2.17; 0.0070 1.95; 0.032 2.27; 0.0028 2.04; 0.018 –1.11; 0.56 3367 19425/ 5.89 17482/ 5.21 Peptidyl–prolyl cis– trans isomerase NIMA–interacting Rotamase B6H262 Predicted –1.49; 0.052 –1.36; 0.066 –1.58; 0.036 1.09; 0.55 –1.06; 0.75 –1.16; 0.35 3430 69862/ 5.04 16870/ 5.05 Heat shock 70 kDa protein K9H0L4 –1.06; 0.46 1.43; 0.0070 1.52; 0.00024 1.52; 0.0050 1.61; 0.00037 1.06; 0.35 3673 18109/ 6.91 14763/ 6.85 Peptidyl–prolyl cis– trans isomerase B6HAJ7 Inferred from homology –1.32; 0.034 –1.36; 0.018 –1.68; 0.0026 –1.03; 0.76 –1.28; 0.059 –1.23; 0.074 4605 17531/ 6.43 9084/4 87 Chaperonin, putative K9GM71 Inferred from homology 1.65; 0.029 1.62; 0.00096 2.08; 0.00061 –1.02; 0.94 1.26; 0.24 1.28; 0.12 Inferred from homology Redox homeostasis Inferred from homology Inferred from homology Inferred from homology 1.02; 0.89 1.57; 0.0055 1.61; 0.00099 1.54; 0.012 1.58; 0.0035 1.03; 0.74 –1.02; 0.89 1.45; 0.018 1.32; 0.095 1.48; 0.0087 1.35; 0.063 –1.10; 0.49 1.04; 0.64 1.43; 0.020 1.35; 0.044 1.37; 0.011 1.29; 0.032 –1.06; 0.57 Q4WPF8 Inferred from homology –1.13; 0.34 1.23; 0.027 1.11; 0.22 1.40; 0.035 1.25; 0.25 –1.12; 0.27 Superoxide dismutase K9G7B6 Inferred from homology –1.57; 0.092 –2.48; 0.0032 –1.88; 0.012 –1.57; 0.18 –1.19; 0.65 1.32; 0.22 14924/ 5.34 Peroxiredoxin 5, atypical 2–Cys peroxiredoxin K9GMX2 Predicted –2.22; 0.00058 –2.07; 0.00037 –2.25; 6.9e–005 1.07; 0.61 –1.02; 0.98 –1.09; 0.50 10126/ 4.77 Thioredoxin K9F7X4 Inferred from homology –2.64; 0.022 –3.50; 0.0024 –4.36; 0.00085 –1.33; 0.47 –1.65; 0.17 –1.24; 0.26 800 79912/ 5.37 66165/ 5.54 Catalase B6H9T9 812 79912/ 5.37 66992/ 5.76 Catalase B6H9T9 819 79912/ 5.37 66784/ 5.88 Catalase B6H9T9 2407 40468/ 8.64 29124/ 5.13 Cytochrome c peroxidase, mitochondrial 3051 31795/ 5.64 20767/ 6.37 3665 19342/ 8.67 4412 12078/ 5.31 Cell cycle/cell signaling 289 60397/ 4.9 86936/ 4.46 Pleckstrin homology– like domain protein B6HV58 Predicted –1.52; 0.0051 –1.30; 0.084 –1.35; 0.15 1.17; 0.32 1.13; 0.82 –1.04; 0.72 306 60397/ 4.9 86131/ 4.53 Pleckstrin homology– like domain protein B6HV58 Predicted –1.64; 0.0029 –1.60; 0.012 –1.89; 0.0020 1.02; 0.94 –1.15; 0.21 –1.18; 0.30 307 60397/ 4.9 85997/ 4.67 Pleckstrin homology– like domain protein B6HV58 Predicted –1.57; 0.0097 –1.69; 0.018 –2.01; 0.018 –1.08; 0.50 –1.28; 0.19 –1.19; 0.40 309 60397/ 4.9 86131/ 4.59 Pleckstrin homology– like domain protein B6HV58 Predicted –1.76; 0.0013 –2.56; 0.0063 –3.16; 0.0022 –1.45; 0.095 –1.80; 0.025 –1.24; 0.54 314 60397/ 4.9 85864/ 4.63 Pleckstrin homology– like domain protein B6HV58 Predicted –1.66; 0.0012 –1.98; 0.0087 –2.58; 0.0015 –1.19; 0.26 –1.55; 0.029 –1.30; 0.32 23 JO U R N A L OF P ROTE O MI CS ( 20 ) 3–2 Table (continued) Spot No Theor MW/pI Exp MW/pI Protein name / function UniProt access No Protein Existance (UniProt) P8/P4 P21/P4 P23/P4 P21/P8 P23/P8 P23/P21 319 60397/ 4.9 85598/ 4.56 Pleckstrin homology– like domain protein B6HV58 Predicted –1.62; 0.0041 –2.27; 0.0042 –3.08; 8.2e–005 –1.40; 0.093 –1.90; 0.0014 –1.36; 0.22 497 60397/ 4.9 77990/ 4.66 Pleckstrin homology– like domain protein B6HV58 Predicted –1.34; 0.059 –1.51; 0.044 –1.79; 0.0050 –1.13; 0.35 –1.34; 0.030 –1.18; 0.39 505 60397/ 4.9 76909/ 4.74 Pleckstrin homology– like domain protein B6HV 58 Predicted –1.81; 0.0038 –1.35; 0.056 –1.51; 0.027 1.34; 0.074 1.20; 0.28 –1.12; 0.47 582 60397/ 4.9 74444/ 4.58 Pleckstrin homology– like domain protein B6HV58 Predicted –1.65; 0.0013 –1.92; 0.012 –2.13; 0.0034 –1.17; 0.31 –1.29; 0.10 –1.11; 0.70 595 56106/ 4.88 74560/ 4.62 Immunogenic protein A2Q9H3 Predicted –1.54; 0.0080 –1.68; 0.016 –2.09; 0.00033 –1.09; 0.57 –1.36; 0.057 –1.25; 0.32 1017 41394/ 4.75 59173/ 4.39 CDC/Septin GTPase family protein K9H369 Predicted –1.46; 0.053 1.13; 0.39 1.11; 0.29 1.66; 0.020 1.63; 0.011 –1.02; 0.95 1029 41394/ 4.75 59081/ 4.42 CDC/Septin GTPase family protein K9H369 Predicted –1.57; 0.0085 1.11; 0.40 1.07; 0.26 1.73; 0.0063 1.68; 0.0021 –1.03; 0.83 1044 41394/ 4.75 58443/ 4.45 CDC/Septin GTPase family protein K9H369 Predicted –1.75; 0.0017 1.09; 0.40 1.07, 0.30 1.92; 0.0026 1.88; 0.0013 –1.02; 0.89 1045 41394/ 4.75 58172/ 4.48 CDC/Septin GTPase family protein K9H369 Predicted –1.40; 0.028 1.15; 0.27 1.12; 0.17 1.62; 0.011 1.56; 0.0035 –1.03; 0.82 1786 32603/ 4.94 39472/ 4.44 Chitosanase of glycosyl hydrolase B6GXD4 Predicted 2.08; 0.087 6.49; 5.2e–005 7.28; 2.9e–005 3.11; 0.0066 3.49; 0.0040 1.12; 0.46 1892 32603/ 4.94 37386/ 4.44 Chitosanase of glycosyl hydrolase B6GXD4 Predicted 1.01; 0.88 1.65; 0.056 2.18; 0.0092 1.63; 0.088 2.15; 0.022 1.32; 0.067 2027 32603/ 4.94 35192/ 4.44 Chitosanase of glycosyl hydrolase B6GXD4 Predicted –1.01; 0.97 1.46; 0.025 1.68; 0.0032 1.47; 0.021 1.69; 0.0025 1.15; 0.24 2822 25180/ 8.00 23330/ 6.69 26S proteasome non– ATPase regulatory subunit Nas2, putative K9FWG5 Predicted –1.29; 0.0036 –1.46; 0.0054 –1.57; 0.00019 –1.13; 0.20 –1.22; 0.015 –1.07; 0.52 3080 20138/ 4.65 20258/ 4.42 Uncharacterized protein K9G5T5 Predicted 1.84; 0.062 1.81; 0.014 1.87; 0.0013 –1.01; 0.93 1.02; 0.77 1.03; 0.78 3138 20138/ 4.65 19670/ 4.42 Uncharacterized protein K9G5T5 Predicted 1.98; 0.013 1.87; 0.0015 2.14; 9.0e–005 –1.06; 0.91 1.09; 0.50 1.15; 0.16 3143 20138/ 4.65 19609/ 4.46 Uncharacterized protein K9G5T5 Predicted 1.78; 0.016 2.39; 0.0027 2.82; 0.0017 1.35; 0.19 1.59; 0.073 1.18; 0.48 3236 20138/ 4.65 18863/ 4.42 Uncharacterized protein K9G5T5 Predicted 2.00; 0.00089 1.78; 0.0046 2.00; 0.00040 –1.12; 0.31 –1.00; 0.95 1.12; 0.23 3492 19032/ 6.85 16228/ 4.39 Mismatched base pair and cruciform DNA recognition protein, putative K9GE01 Predicted –1.13; 0.40 1.43; 0.015 1.48; 0.0010 1.62; 0.038 1.67; 0.020 1.03; 0.71 1.89; 0.018 2.00; 0.0056 1.49; 0.10 1.58; 0.043 1.06; 0.70 Secondary metabolites biosyntesis 2805 25629 /5.72 23257/ 5.76 isoepoxydon dehydrogenase A1XDS5 Inferred from homology 1.27; 0.30 (continued on next page) 24 JO U R N A L OF PR O TE O MI CS 96 ( 20 ) –28 Table (continued) Spot No Theor MW/pI Exp MW/pI Protein name / function UniProt access No Protein Existance (UniProt) P8/P4 P21/P4 P23/P4 P21/P8 P23/P8 P23/P21 Unknown function 1395 24487/ 5.24 48442/ 4.47 hypothetical protein PDIP_62460 K9FL12 Predicted –1.05; 0.76 1.46; 0.041 1.40; 0.017 1.53; 0.046 1.47; 0.028 –1.04; 0.82 1414 24487/ 5.24 47622/ 4.5 hypothetical protein PDIP_62460 K9FL12 Predicted –1.05; 0.64 1.64; 0.010 1.63; 0.00023 1.72; 0.030 1.71; 0.012 –1.00; 0.92 3137 20550/ 9.17 /7.19 Conserved hypothetical protein B6H779 Predicted –1.79; 0.039 –6.19; 0.00027 –6.58; 1.4e–006 –3.46; 0.011 –3.69; 0.0028 –1.06; 0.97 3365 20550/ 9.17 17267/ 7.43 Conserved hypothetical protein B6H779 Predicted –1.92; 0.019 –4.48; 2.0e–005 –4.73; 0.00012 –2.33; 0.014 –2.46; 0.017 –1.06; 0.69 3722 16830/ 5.80 14446/ 5.04 Uncharacterized protein K9GK67 Predicted 1.25; 0.040 –1.13; 0.46 –1.26; 0.094 –1.41; 0.079 –1.58; 0.00078 –1.11; 0.76 4060 13188/ 5.64 12331/ 4.87 YjgF_YER057c_UK1 14 (unknown function) B6HKJ1 Predicted 1.52; 0.015 1.23; 0.0055 1.38; 0.00011 –1.24; 0.17 –1.10; 0.55 1.13; 0.062 4094 12308/ 4.95 12123/ 4.63 RNA binding protein, putative K9FZG7 Predicted –1.09; 0.47 –1.56; 0.017 –1.70; 0.00032 –1.42; 0.073 –1.55; 0.0092 –1.09; 0.63 domain were less abundant in P4 Six enzymes playing a role in protein folding were found: several HSPs 70 (chaperonin), a HSP 60 (chaperonin) and a HSP-p23-like (co-chaperonin of HSP 90 involved in protein folding and degradation) were less abundant in P4 on one side, and a rotamase containing a peptidylprolyl isomerase (PPIase) domain was more abundant on the other side HSPs 60 are ATP-dependent chaperones exclusively located in the mitochondria They are involved in the folding of the proteins of this organelle, as well as in the maintaining of mitochondria nuclei [27]; moreover these proteins are also involved in the protection of proteins containing a Fe–S cluster against reactive oxygen species (ROS) HSPs 70 are one of the main components of the protein folding machinery, they aid in forming the native conformation of proteins [26] but are likewise involved in proteasomal degradation [28] They are found in cytosol, endoplasmic reticulum and mitochondria and are activated by ATP or by nucleotides through the nucleotide exchange factors (NEFs), the nature of which depends on the subcellular location of the HSPs 70 3.3.4 Proteins involved in cell cycle and cell signalling pathways Some proteins relative to cell cycle and cell signalling pathways were found to be differentially abundant in the investigated strains Nine spots corresponding to a pleckstrin homology-like (PH) domain containing protein were found These spots were all more abundant in P4 According to Fugelstad et al [29], such proteins may bind F-actin and play a role in the regulation of the cell wall biosynthesis PH-like containing proteins are also involved in cell communication and host protein targeting [30] Concerning cell cycle, and more specifically cell division, a spot corresponding to a chitosanase, an enzyme which might hydrolyse chitosan of the cell wall during division, was found to be less abundant in P4 Liu et al [31] reported that chitosanase reduces the phytopathogenicity of Fusarium solani Indeed, the decrease of the disease index has confirmed the lower relative expression level of CSN1, the gene responsible for the chitosanase synthesis The reasons why fungi produce an enzyme which might disfavour themselves remain elusive 3.3.5 Redox homeostasis Proteins involved in the regulation of redox homeostasis were detected and significantly differed from one group to the other Three spots corresponding to catalase were significantly less abundant in P4, while three others corresponding to peroxiredoxin 5, thioredoxin and superoxide dismutase were more abundant Catalases were presumably associated with Cat of Aspergillus fumigatus (80% of identity with XP_748550.1) This large-size subunit catalase is highly produced in conidia, i.e during germination and initiation of growth [32] For phytopathogenic fungi, these developmental stages mainly occur during the infection of the host A higher catalase activity is necessary to counteract the reaction of the plant As for peroxiredoxins, they are activated by low concentrations of peroxides, which oxidise the active-site cysteine to sulfenic acid [33] 3.3.6 Unknown hypothetical proteins and proteins with unknown functions Four spots containing unknown hypothetical proteins (spots n° 1395, 1414, 3137 and 3365) were differentially abundant They were presenting high levels of homology with hypothetical proteins of Aspergillus and Penicillium species No conserved domains were found by BLAST algorithm The JO U R N A L OF P ROTE O MI CS ( 20 ) 3–2 25 Fig – Glycolytic, mevalonate and non-mevalonate pathways leading to the production of geosmin Green and red arrows represent respectively enzymes that are more or less abundant in geosmin producing strains (P8, P21 and P23) Names of differentially abundant proteins are indicated PEP: phosphoenolpyruvate MEP: methylerythritol phosphate pathway MVA: mevalonate pathway available knowledge on these proteins is too low to allow us even to speculate on their potential role in the biosynthesis of geosmin Three additional proteins were found (spots n° 3722, 4060 and 4094) but their roles are not yet really understood Spot n° 4060 was less abundant in P4 whereas spot n° 4094 was more abundant in P4 as compared to the second group Discussion In the perspective of linking geosmin biosynthesis to the previously described results, several points have to be taken into account Geosmin is a molecule of the sesquiterpenoid family, which is part of the secondary metabolism encountered in fungi The synthesis of such molecules is highly energy-consuming and thus requires high quantities of energy storage molecules such as ATP or NADPH Production of ATP via oxidative phosphorylation leads to the formation of ROS, susceptible to damage cellular organelles and biological molecules, which is generally counteracted by detoxifying enzymes, such as catalase and peroxiredoxins deputed to keep the redox homeostasis Regulation of fungal secondary metabolism is fine-tuned by several external factors, including pH, temperature and nutrition [34], thus it may be hypothesized that geosmin production has a high impact on the behaviour of the cell regarding redox homeostasis Changes in secondary metabolism in general, and in geosmin production particularly, can be considered as both a cause and a consequence of oxidative stress As the culture conditions were the same for all the isolates, the achieved results can be considered as intrinsic and constitutive differences between the strains Investigation on the enzymes of the glycolytic pathway (Fig 6) has revealed in P4 an increased abundance of phosphoglycerate kinase, one of the flux-determining enzymes regulating the glycolysis, suggesting that the ADP/ATP ratio is in favour of ATP production Considering that spots containing the full length chain of enolase were less abundant in P4, we may consider that phosphoglycerate produced by PGK is dispatched to other pathways than glycolysis, i.e production of ribulose biphosphate or glyoxylate As a result, production of acetyl-CoA by the glycolytic pathway may be lower in P4 Acetyl CoA is the first molecule of the MVA pathway, which leads to the biosynthesis of isopentenyl diphosphate (IPP), a precursor of farnesyl diphosphate (FPP) Finally, FPP is the substrate for sesquiterpenoid production, including geosmin It must be noted that pyruvate can also be transformed into IPP via the methylerythritol phosphate pathway (MEP) The pool of pyruvate may be higher in geosmin producing strains due to the higher enolase abundance Enolase was identified in several spots at a molecular weight far below the expected for the intact polypeptide chain, suggesting degradation or processing of the protein as was previously postulated for enolase in Penicillium chrysogenum [35] and in more distantly related fungi [36] Different and 26 JO U R N A L OF PR O TE O MI CS 96 ( 20 ) –28 contradictory hypotheses have been proposed for the observed degradation Incubation with PMSF indicated that the fragmentation could be related to proteasome activity [37], even proposing enolase, a proteasome interacting protein, as substrate or regulator of proteasome activity In an older study a contradictory result was obtained Larsen et al [38] did a detailed mapping of the different spots and found a specific cleavage site in some instances However the proposed cleavage sequence K/RxA only appears infrequently in the P chrysogenum protein on which our identification is based, and will at best only explain part of the observed fragmentation, furthermore no known protease with this specificity was found Despite the fact that the protease responsible for this degradation remains elusive, the same fragmentation of enolase was previously observed during the switch from a respiratory to a fermentative metabolism [36] Larsen et al [38] hypothesize that degradation of the enzymes involved in the latest stages of the glycolytic pathway (including enolase) might be a way of surviving unfavourable conditions, as was previously suggested [39] Several differentially expressed enzymes were related to peptide and protein metabolism The second group (P8, P21 and P23) has shown a significantly higher abundance of three proteinases (endopeptidase) and two serine carboxypeptidases (exopeptidase) The combination of the actions of these classes of enzymes allows to catabolise entire proteins into small polypeptides and finally single amino acids Catabolism of proteins thus appears to be higher in the high-geosmin-producer group Blomberg et al [30] have reported that proteins containing a PH-like domain are involved in, but not limited to, controlling and running enzymatic activities or regulation of nuclear transport All the ten spots which were identified as PH-like protein together with the protein rotamase, a protein the function of which is to optimise protein folding, were more intense in P4, thus indicating a higher transcriptional activity In contrast to these observations, the elongation factor 1-beta and the 40S ribosomal protein S0 were less abundant in P4 Therefore the protein synthesis does not exhibit a clear tendency allowing discriminating P4 from P8, P21 and P23 Under normal conditions, HSPs play major roles in protein folding, disaggregation and degradation [27] When the culture conditions become more stressful, their increased concentrations in the cell may indicate that they are part of the response to the caused imbalance HSP 70 is up-regulated during cellular response to stress, especially oxidative stress, as described by Jamieson et al [40] and reviewed by Mayer and Bukau [28] It is also reported that the SSA1 gene, encoding an isoform of HSP 70, is induced by oxidising agents such as H2O2 [41] Oxidative conditions may distort the conformation of many macromolecules, including nucleic acids and proteins, which thus require both protection/reaction against oxidative species (e.g catalase, superoxide dismutase) and reinforcement of molecules dedicated to maintaining structural conformation of the macromolecules like HSPs HSP 60 has been described as protector of enzymes containing a Fe–S cluster, avoiding release of free iron in the cell, as free iron catalyses oxidative reaction [42] HSP 60 is an important enzyme due to its localisation in the mitochondria, which is the main site of cellular ROS production via the electron transport chain Observation of the fold change values of ATP synthase indicates that in P4 five out of seven of these proteins were less abundant Two cytochrome c oxidases were also less abundant in P4 This may finally agree with the high quantity of ATP needed for the production of secondary metabolites Qin et al [3] report that the expression of HSP 60 is higher in P expansum culture submitted to borate causing an oxidative stress Mitochondrial ATP production may cause a comparable oxidative mechanism Interestingly, HSP 60 and HSP 70 intervene successively for the protein folding inside the mitochondria [42] We have also detected a significant fold change in the expression of a HSP-p23 like protein, which is a co-chaperone of HSP 90 [43] This HSP-complex is involved in the protection of the nucleic acids against oxidative stress that could result in the loss of chromosome fragments [44] This study shows that H2O2 causes major loss of chromosome fragments by inhibiting HSP 90 activity, in a comparable way as benomyl, a fungicide obstructing microtubule synthesis Importance of HSP 90 in cell viability and protection against oxidative stress is also proved by the targeting of this protein by antifungal agents as geldanamycin or radicicol [45] The detoxification of ROS is done by a group of enzymes that degrade these molecules and that have the potential to scavenge the free radicals that are formed during the process, preventing the formation of damage on biomolecules One of the key actors in this defensive mechanism is the protein family of the peroxiredoxins, which constitute the first step of the cellular response to oxidative stress Its synthesis occurs when a limited concentration of oxidant species is encountered in the cell [46], i.e at H2O2 concentration below 10 μM [32] Moreover, H2O2 might activate other cellular pathways: 2-cys peroxiredoxin has a chaperone activity and prevents ROS-linked damage to DNA in Saccharomyces cerevisiae [47] Another way by which the cell can react to oxidative stress, and in particular to H2O2, is overexpressing catalase, which catalyses the degradation of this dangerous molecule This enzyme exhibits a Km of 20–200 mM, which is 1000 to 10000 fold higher than the Km of peroxiredoxin [48] All together, the expression patterns of these anti-oxidative enzymes may suggest that P4 has a more favourable redox status The abundance of catalase, the most efficient enzyme for ROS detoxification, was lower At the same time, peroxiredoxin was more abundant, which may indicate a slight oxidative stress Indeed, catalytic cysteine in its hyperoxidised form (mechanisms explained in [33]) becomes catalytically inactive, even if it is still participating to ROS signalling and having a chaperone activity [49], pushing the cell to adapt its response to oxidative aggression with faster and more powerful systems Conclusion By comparing the proteome profile of P expansum strains with a different geosmin production under identical culture conditions, 107 proteins were found to be differentially abundant Functional classification has revealed that enzymes involved in redox homeostasis, protein folding and in the first steps of secondary metabolism may have an important role in the biosynthesis of geosmin P4 may present a lower oxidative status as compared to P8, P21 and P23 regarding the differential abundance of catalase and peroxiredoxin Other enzymes related to glycolysis, protein metabolism, cell cycle and ATP production have also been JO U R N A L OF P ROTE O MI CS ( 20 ) 3–2 described and analysed All together, these results constitute a partial explanation of i) the reasons and ii) the mechanism of geosmin production and provide a first insight in the comprehension of geosmin synthesis by filamentous fungi Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.jprot.2013.10.034 Acknowledgements All the authors would like to thank Sébastien Planchon for his valuable technical assistance and support REFERENCES [1] Ugliano M, Bartowsky EJ, McCarthy J, Moio L, Henschke PA Hydrolysis and transformation of grape glycosidically bound volatile compounds during fermentation with three Saccharomyces yeast strains J Agric Food Chem 2006;54:6322–31 [2] La Guerche S, Dauphin B, Pons M, Blancard D, Darriet P Characterization of some mushroom and earthy off-odors microbially induced by the development of rot on grapes J Agric Food Chem 2006;54:9193–200 [3] Qin G, Tian S, Chan Z, Li B Crucial role of antioxidant proteins and hydrolytic enzymes in pathogenicity of Penicillium expansum Mol Cell Proteomics 2007;6:425–38 [4] Morales-Valle H, Silva LC, Paterson RRM, Venancio A, Lima N Effects of the origins of Botrytis cinerea on earthy aromas from grape broth media further inoculated with Penicillium expansum Food Microbiol 2011;28:1048–53 [5] Darriet P, Pons M, Henry R, Dumont O, Findeling V, Cartolaro P, et al Impact odorants contributing to the fungus type aroma from grape berries contaminated by powdery mildew (Uncinula necator); incidence of enzymatic activities of the yeast Saccharomyces cerevisiae J Agric Food Chem 2002;50:3277–82 [6] Boutou S, Chatonnet P Rapid headspace solid-phase microextraction/gas chromatographic/mass spectrometric assay for the quantitative determination of some of the main odorants causing off-flavours in wine J Chromatogr A 2007;1141:1–9 [7] La Guerche S, Chamont S, Blancard D, Dubourdieu D, Darriet P Origin of (−)-geosmin on grapes: on the complementary action of two fungi, Botrytis cinerea and Penicillium expansum Antonie Van Leeuwenhoek 2005;88:131–9 [8] Gust B, Challis GL, Fowler K, Kieser T, Chater KF PCR-targeted Streptomyces gene replacement identifies a protein domain needed for biosynthesis of the sesquiterpene soil odor geosmin Proc Natl Acad Sci U S A 2003;100:1541–6 [9] Bentley R, Meganathan R Geosmin and methylisoborneol biosynthesis in Streptomycetes: Evidence for an isoprenoid pathway and its absence in non-differentiating isolates FEBS Lett 1981;125:220–2 [10] Pollak FC, Berger RG Geosmin and related volatiles in bioreactor-cultured Streptomyces citreus CBS 109.60 Appl Environ Microbiol 1996;62:1295–9 [11] Cane DE, Watt RM Expression and mechanistic analysis of a germacradienol synthase from Streptomyces coelicolor implicated in geosmin biosynthesis Proc Natl Acad Sci U S A 2003;100:1547–51 [12] Agger S, Lopez-Gallego F, Schmidt-Dannert C Diversity of sesquiterpene synthases in the basidiomycete Coprinus cinereus Mol Microbiol 2009;72:1307–8 27 [13] Dionigi CP The Effects of copper-sulfate on geosmin biosynthesis by Streptomyces tendae, Streptomyces albidoflavus, and Penicillium expansum Water Sci Technol 1995;31:135–8 [14] Siddique MH, Liboz T, Bacha N, Puel O, Mathieu F, Lebrihi A Characterization of a cytochrome P450 monooxygenase gene involved in the biosynthesis of geosmin in Penicillium expansum Afr J Microbiol Res 2012;6:4122–7 [15] Bhadauria V, Banniza S, Wang LX, Wei YD, Peng YL Proteomic studies of phytopathogenic fungi, oomycetes and their interactions with hosts Eur J Plant Pathol 2010;126:81–95 [16] Glass NL, Donaldson GC Development of primer sets designed for use with the PCR to amplify conserved genes from filamentous ascomycetes Appl Environ Microbiol 1995;61:1323–30 [17] La Guerche S, Garcia C, Darriet P, Dubourdieu D, Labarère J Characterization of Penicillium species isolated from grape berries by their internal transcribed spacer (ITS1) sequences and by gas chromatography–mass spectrometry analysis of geosmin production Curr Microbiol 2003;48:405–11 [18] Haas B, Serchi T, Wagner DR, Gilson G, Planchon S, Renaut J, et al Proteomic analysis of plasma samples from patients with acute myocardial infarction identifies haptoglobin as a potential prognostic biomarker J Proteomics 2011;75:229–36 [19] Pasquali M, Serchi T, Renaut J, Hoffmann L, Bohn T 2D DIGE reference map of a Fusarium graminearum nivalenol producing strain Electrophoresis 2013;34:505–9 [20] La Guerche S, De Senneville L, Blancard D, Darriet P Impact of the Botrytis cinerea strain and metabolism on (−)-geosmin production by Penicillium expansum in grape juice Antonie Van Leeuwenhoek 2007;92:331–41 [21] Darriet P, Pons M, Lamy S, Dubourdieu D Identification and quantification of geosmin, an earthy odorant contaminating wines J Agric Food Chem 2000;48:4835–8 [22] Franc C, David F, de Revel G Multi-residue off-flavour profiling in wine using stir bar sorptive extraction–thermal desorption–gas chromatography–mass spectrometry J Chromatogr A 2009;1216:3318–27 [23] Alm R, Johansson P, Hjernø K, Emanuelsson C, Ringnér M, Häkkinen J Detection and identification of protein isoforms using cluster analysis of MALDI-MS mass spectra J Proteome Res 2006;5:785–92 [24] Roze LV, Chanda A, Linz JE Compartmentalization and molecular traffic in secondary metabolism: a new understanding of established cellular processes Fungal Genet Biol 2011;48:35–48 [25] Ramirez-Zavala B, Mercado-Flores Y, Hernandez-Rodriguez C, Villa-Tanaca L Purification and characterization of a serine carboxypeptidase from Kluyveromyces marxianus Int J Food Microbiol 2004;91:245–52 [26] ten Have A, Espino JJ, Dekkers E, Van Sluyter SC, Brito N, Kay J, et al The Botrytis cinerea aspartic proteinase family Fungal Genet Biol 2010;47:53–65 [27] Panaretou B, Zhai C The heat shock proteins: their roles as multi-component machines for protein folding Fungal Biology Reviews 2008;22:110–9 [28] Mayer M, Bukau B Hsp70 chaperones: cellular functions and molecular mechanism Cell Mol Life Sci 2005;62:670–84 [29] Fugelstad J, Brown C, Hukasova E, Sundqvist G, Lindqvist A, Bulone V Functional characterization of the pleckstrin homology domain of a cellulose synthase from the oomycete Saprolegnia monoica Biochem Bioph Res Co 2012;417:1248–53 [30] Blomberg N, Baraldi E, Nilges M, Saraste M The PH superfold: a structural scaffold for multiple functions Trends Biochem Sci 1999;24:441–5 28 JO U R N A L OF PR O TE O MI CS 96 ( 20 ) –28 [31] Liu H, Zhang B, Li C, Bao X Knock down of chitosanase expression in phytopathogenic fungus Fusarium solani and its effect on pathogenicity Curr Genet 2010;56:275–81 [32] Hansberg W, Salas-Lizana R, Dominguez L Fungal catalases: function, phylogenetic origin and structure Arch Biochem Biophys 2012;525:170–80 [33] Wood ZA, Schröder E, Robin Harris J, Poole LB Structure, mechanism and regulation of peroxiredoxins Trends Biochem Sci 2003;8:32–40 [34] Fox EM, Howlett BJ Secondary metabolism: regulation and role in fungal biology Curr Opin Microbiol 2008;11:481–7 [35] Jami MS, Garcia-Estrada C, Barreiro C, Cuadrado AA, Salehi-Najafabadi Z, Martin JF The Penicillium chrysogenum extracellular proteome Conversion from a food-rotting strain to a versatile cell factory for white biotechnology Mol Cell Proteomics 2010;9:2729–44 [36] Kobi D, Zugmeyer S, Potier S, Jaquet-Gutfreund L Two-dimensional protein map of an “ale”-brewing yeast strain: proteome dynamics during fermentation FEMS Yeast Res 2004;5:213–30 [37] Trabalzini L, Paffetti A, Scaloni A, Talamo F, Ferro E, Coratza G, et al Proteomic response to physiological fermentation stresses in a wild-type wine strain of Saccharomyces cerevisiae Biochem J 2003;370:35–46 [38] Larsen MR, Larsen PM, Fey SJ, Roepstorff P Characterization of differently processed forms of enolase from Saccharomyces cerevisiae by two-dimensional gel electrophoresis and mass spectrometry Electrophoresis 2001;22:566–75 [39] Godon C, Lagniel G, Lee J, Buhler JM, Kieffer S, Perrot M, et al The H2O2 stimulon in Saccharomyces cerevisiae J Biol Chem 1998;273:22480–9 [40] Jamieson DJ, Rivers SL, Stephen DWS Analysis of Saccharomyces cerevisiae proteins induced by peroxide and superoxide stress Microbiol 1994;140:3277–83 [41] Stephen DWS, Rivers SL, Jamieson DJ The role of the YAP1 and YAP2 genes in the regulation of the adaptive oxidative stress responses of Saccharomyces cerevisiae Mol Microbiol 1995;16:415–23 [42] Cabiscol E, Bellí G, Tamarit J, Echave P, Herrero E, Ros J Mitochondrial Hsp60, resistance to oxidative stress, and the labile iron pool are closely connected in Saccharomyces cerevisiae J Biol Chem 2002;277:44531–8 [43] Felts SJ, Toft DO p23, a simple protein with complex activities Cell Stress Chaperon 2003;8:108–13 [44] Chen G, Bradford WD, Seidel CW, Li R Hsp90 stress potentiates rapid cellular adaptation through induction of aneuploidy Nature 2012;482:246–50 [45] Nicola A, Andrade R, Dantas A, Andrade P, Arraes F, Fernandes L, et al The stress responsive and morphologically regulated hsp90 gene from Paracoccidioides brasiliensis is essential to cell viability BMC Microbiol 2008;8:158–65 [46] Vivancos A, Jara M, Zuin A, Sanso M, Hidalgo E Oxidative stress in Schizosaccharomyces pombe: different H2O2 levels, different response pathways Mol Genet Genomics 2006;276:495–502 [47] Morgan BA, Veal EA Functions of typical 2-Cys peroxiredoxins in yeast peroxiredoxin systems In: Flohé L, Harris JR, editors Netherlands: Springer; 2007 [48] Díaz A, Valdés VJ, Rudino-Pinera E, Horjales E, Hansberg W Structure–function relationships in fungal large-subunit catalases J Mol Biol 2009;386:218–32 [49] Edgar RS, Green EW, Zhao Y, van Ooijen G, Olmedo M, Qin X, et al Peroxiredoxins are conserved markers of circadian rhythms Nature 2012;485:459–64  ... in the mitochondria They are involved in the folding of the proteins of this organelle, as well as in the maintaining of mitochondria nuclei [27]; moreover these proteins are also involved in the. .. biomolecules One of the key actors in this defensive mechanism is the protein family of the peroxiredoxins, which constitute the first step of the cellular response to oxidative stress Its synthesis occurs... All together, these results constitute a partial explanation of i) the reasons and ii) the mechanism of geosmin production and provide a first insight in the comprehension of geosmin synthesis