APPLIED MICROBIAL AND CELL PHYSIOLOGY

Aoyap1 regulates OTA synthesis by controlling cell redoxbalance in Aspergillus ochraceus

Massimo Reverberi & Katia Gazzetti & Federico Punelli &Marzia Scarpari & Slaven Zjalic & Alessandra Ricelli &Anna A. Fabbri & Corrado Fanelli

Received: 10 November 2011 /Revised: 17 February 2012 /Accepted: 17 February 2012# Springer-Verlag 2012

Abstract Among the various factors correlated with toxinproduction in fungi, oxidative stress is a crucial one. In relationto this, an important role is played by oxidative stress-relatedreceptors. These receptors can transduce the “oxidative mes-sage” to the nucleus and promote a transcriptional changetargeted at restoring the correct redox balance in the cell. InAspergillus parasiticus, the knockout of the ApyapA gene, ahomologue of the yeast Yap-1, disables the fungus’s capacityto restore the correct redox balance in the cell. As a conse-quence, the onset of secondary metabolism and aflatoxinssynthesis is triggered. Some clues as to the involvement ofoxidative stress in the regulation of ochratoxin A (OTA) syn-thesis in Aspergillus ochraceus have already been provided bythe disruption of the oxylipin-producer AoloxA gene. In thispaper, we add further evidence that oxidative stress is alsoinvolved in the regulation of OTA biosynthesis in A. ochra-ceus. In fact, the use of certain oxidants and, especially, thedeletion of the yap1-homologue Aoyap1 further emphasize therole played by this stress in controlling metabolic and morpho-logical changes in A. ochraceus.

Keywords Oxidative stress . Ochratoxin A . Aspergillusochraceus . Aoyap1

Introduction

The control of ochratoxin A (OTA)-producer fungi is ofgreat interest, since OTA represents a severe health hazardfor humans and animals (WHO 2001), owing to its geno-toxic effect and potential carcinogenicity (IARC 1993).Because of the hazard posed by this mycotoxin, the amountof OTA permitted in food has been regulated by Europeanlaw (EU 1881/2006). Studies on the toxicity of OTA inmammals have shown that OTA interferes with tRNA andinhibits protein synthesis, leading mainly to kidney mal-function and related pathologies (Cheeke 1998; Creppy etal. 1990; Stoev 1998). Among the OTA-producing fungi,Aspergillus ochraceus is a ubiquitous colonizer, in temper-ate countries, of dried foods, such as nuts and cereals (Vander Merwe et al. 1965; Pardo et al. 2004). It is known thatOTA, which belongs to the isocoumarin family, is synthe-sized from acetate and malonate molecules through thepolyketide pathway (Van der Merwe et al. 1965). Somesteps of its biosynthetic pathway in A. ochraceus haverecently been elucidated (O’Callaghan et al. 2003; 2006).Enzyme-driven lipoperoxidation process was demonstratedto be one of the determinant events, affecting OTA biosyn-thesis in A. ochraceus (Reverberi et al. 2010). The disrup-tion of the AoloxA gene was shown to disable the mutantstrain to synthesize OTA both in a liquid medium and inwheat seeds. These results led to the assumption that thelipoxygenase products and by-products, like oxylipins andsuperoxide anions, may regulate morphogenesis andOTA synthesis and may drive the interaction with the hostin A. ochraceus. However, how this “oxidative message” is

M. Reverberi (*) :K. Gazzetti : F. Punelli :M. Scarpari :A. A. Fabbri : C. FanelliDipartimento di Biologia Ambientale, University Sapienza,Largo Cristina di Svezia 24,00165 Rome, Italye-mail: massimo.reverberi@uniroma1.it

A. RicelliIstituto di Chimica Biomolecolare (CNR),Piazzale Aldo Moro 5,00189 Rome, Italy

S. ZjalicDepartment of Mediterranean Agriculture and Aquaculture,University of Zadar,M. Pavlinovica bb,23000 Zadar, Croatia

Appl Microbiol BiotechnolDOI 10.1007/s00253-012-3985-4

“perceived” and transformed into a metabolic and morpho-logical response by A. ochraceus was still elusive. Al-though, over recent years, several environmental factorshave been correlated with the regulation of morphologicaltransitions and secondary metabolism onset in fungi (Payneand Hagler 1983; Wilkinson et al. 2007; Reverberi et al.2010), it has emerged that it is oxidative stress that plays thepivotal role in controlling these events (Aguirre et al. 2005;Aguirre et al. 2006; Jayashree and Subramanyam 2000;Reverberi et al. 2008). The biosynthesis of different myco-toxins (such as aflatoxins, patulin, OTA, and some fusario-toxins) has a particular factor in common, namely, it can beaffected by reactive oxygen species (ROS) (Narasaiah et al.2006; Ponts et al. 2006; Reverberi et al. 2008, 2010; Tolainiet al. 2010). ROS are generated during metabolism, inmitochondria and peroxisomes, and by NADPH-oxidaseactivity and fatty acids oxidation (Halliwell and Gutteridge2007), and they play a role in modulating fungal develop-ment. A hyperoxidant environment can drive fungal metabol-ic responses towards differentiation processes, such asconidiogenesis, sclerotia, sexual structure formation, and my-cotoxin synthesis in different Aspergilli (Calvo et al. 2002)and Neurospora crassa (Aguirre et al. 2005). Conversely, themaintenance of a reductive environment can promote undif-ferentiated mycelial growth (Jayashree and Subramanyam2000). In undifferentiated growth, i.e., in trophophase, thecorrect redox balance is also maintained through oxidativestress-related transcription factors, such as Yap-1 in Saccha-romyces cerevisiae. This cytosolic receptor/nuclear transcrip-tion factor is able to promote the gene expression targeted atrestoring the oxidant/antioxidant balance in the cell (Estruch2000). In the mycotoxin-producer fungus, Aspergillus para-siticus, the ApyapA gene, Yap-1-homologue, is also correlat-ed to the “perception” of oxidants and to the control of redoxbalance. The deletion of the ApyapA gene in A. parasiticusdownregulates the antioxidant cell response and triggers theaflatoxin biosynthesis. Notably, an in silico analysis indicatesthe presence of Yap1-responsive elements in the aflatoxinregulator (aflR) and in nor-1 promoter (Roze et al. 2011).

In this paper, we investigate whether a mechanism ofoxidative stress signal perception, similar to that alreadydisplayed in A. parasiticus, mediated by Aoyap1, is alsoinvolved in OTA biosynthesis by A. ochraceus. The roleplayed by Aoyap1 is also indirectly demonstrated usingsome oxidants in the culture media.

Materials and methods

Fungal strains and culture conditions

The wild type (WT) A. ochraceus (Wilhelm) 14899 [newlyassigned by the Culture Collection Agro-Food Important

Toxigenic Fungi-Item, Institute of Science of Food Production(ISPA), National Research Council (CNR), Via Amendola,122/O, 70126 Bari, Italy (http://server.ispa.cnr.it/ITEM/Collection/)] was kindly supplied by Prof. P. Battilani, UniversitàCattolica di Piacenza, Italy. The ΔAoyap1 mutant strain wasobtained as described in “Creation of theΔAoyap1mutant”).The culture conditions were performed using a conidia sus-pension (1×106 in 0.2 ml of sterile distilled water) of the WTstrain or ΔAoyap1 mutant. The conidia were inoculated andcultured in static condition in potato dextrose broth (PDB) orin potato dextrose agar (PDA), with (selective medium) orwithout hygromycin B (Hyg B; Merck KGaA, Darmstadt,Germany) 150 μg/ml. The cultures were then incubated at25 °C in dark conditions. In another experiment, some oxi-dants (carbon tetrachloride, CCl4 30 mM; menadione, MEN0.1 mM; hydrogen peroxide, H2O2 10 mM; tert-butyl hydro-peroxide, t-BOOH 100 mM) were added to the culture medi-um. The concentrations of the different compounds werechosen referring to those used in previous papers (Kim et al.2004; Reverberi et al. 2008).

Creation of the ΔAoyap1 mutant

Aliquots of 100 mg of lyophilized mycelium were powderedin liquid nitrogen and treated for DNA extraction. Threedifferent cetyltrimethylammonium bromide buffers (2, 10,and 1% w/v) were used to extract DNA from both strains ofA. ochraceus followed by purification steps in organic sol-vents. Since no gene sequence coding for the oxidativestress transcription factors Yap-1-homologues, superoxidedismutases (sod), and/or catalases (cat) of A. ochraceusare reported in the National Center for Biotechnology Infor-mation (NCBI)’s GenBank, the primers used for amplifica-tion were designed on the basis of the consensus sequencesderived from the alignment of several fungal yap1, sod, andcat gene loci. Alignments were performed by employingTBLASTX 2.2.13 software of the NCBI website (www.ncbi.nlm.nih.gov/BLAST), and the conserved regions wereselected as suitable oligonucleotide primer sites. Degenerateoligonucleotide primers were synthesized (Table 1). Afteramplification, the expected bands at 750 bp (Aoyap1),233 bp (Aoczsod1), and 309 bp (AocatA) were sequencedand compared with other fungal yap1, sod, and cat genes.Some yap1, sod, and cat primers with higher specificitywere synthesized (Table 1) and used for creating the Aoyap1disruption cassette, for amplifying A. ochraceus DNA andcomplementary DNA (cDNA).

Plasmids and transformation

The genomic DNA of A. ochraceus was amplified withAoyapEco91I-forward (5′- GGT NAC CAT AGA CTGCAG GCG AT -3′) and AoyapStuI-reverse (5′- AGG CCT

Appl Microbiol Biotechnol

TGC CGG ATC TGG GAA GTC -3′) to obtain an Aoyap1afragment of 239 bp and with AoyapXbaI-forward (5′- TCTAGA CAC GCT CTC GCT TTT G -3′) and Aoyap1-reverse(5′- TCC ATC ATC CTC ATC GGA CT -3′) to obtain anAoyap1b fragment of 143 bp. The fragments obtained werecloned into the vector pAN7.1 (6.7 Kb) previously digestedwith Eco91I–StuI at 5′ and HindIII–XbaI to 3′, alongside thehygromycin cassette (hph box). A. ochraceusWTwas trans-formed with the resulting disruption cassette.

Selection of transformants

Putative transformants were selected on PDA containingHyg B 150 μg/ml, transferred to fresh selective medi-um, and allowed to sporulate. Single spores were iso-lated from each of the selected transformants andtransferred to fresh selective medium by three consecu-tive steps. Finally, 20 colonies were selected and furthersubcultured. The stability of these transformants wastested by two additional single-spore transfers at firston nonselective and then, again, on selective media.Transformants selection was performed in three steps

as follows. (1) Carry out PCR with selective primersdesigned on the hph box and on the Aoyap1 genesequence (expected fragment at ~1.7 Kb). (2) Checkfor the presence of the hygromycin B resistance cassettein the fungal genome, a test performed by Southern blothybridization with a specific probe obtained using primersHpF and TrR (Table 1). The presence or absence of theexpected fragment was tested using the Southern blot analysisof KpnI-restricted DNA (which did not restrict the Aoyap1fragment and the hph box). (3) The ability of the mutants togrow and sporulate on PDA with or without Hyg B wasconfirmed.

Southern hybridization

For Southern blot analysis, 10 μg of genomic DNA from A.ochraceusWTandΔAoyap1 strains was completely digestedwith KpnI (10 U) at 37 °C for 4 h in the manufacturer’s bufferat the recommended concentrations (Fermentas GMBH, StLeon, Germany). KpnI-digested DNA fragments were sepa-rated by electrophoresis for 3 h and 30 min at 40 Von a 0.8%(w/v) agarose gel in 1 M Tris-acetate-EDTA buffer. Digoxige-nin (DIG)-labeled HindIII cut lambda (Roche, Basel,Switzerland) was used as a molecular weight standard. Beforeblotting, the gel containing the DNAwas washed twice withdenaturation buffer (0.5 N NaOH and 1.5 M NaCl) for 30 minand twice with neutralization buffer (0.5 M Tris-HCl and 3 MNaCl, pH 7.5) for 15 min. The nucleic acids were transferred toa Hybond-N+ nylon membrane (Roche, Basel, Switzerland)using 10× SSC (1.5 M NaCl and 0.15 M sodium citrate) andfixed by UV illumination, in accordance with the Roche meth-od, after overnight blotting. Fluorescent DNA probes wereprepared according to the PCR DIG labeling mix method(Roche, Basel, Switzerland). The membranes were prehybri-dized according to the manufacturer’s instructions of the DIGdetection kit, at 65 °C in DIG Easy buffer (Roche, Basel,Switzerland); they were then hybridized for 12 to 16 h in thesame buffer containing 250 ng of freshly denatured DIG-labeled Aoyap1 or hph probes at 65 °C.

Analysis of fungal growth and conidiogenesis

Fungal growth in the whole time course of the experimentswas measured as dry weight (DW) by weighing the myce-lium after filtration (Millipore filters, 0.45 μm) and dryingfor 48 h at 80 °C. The number of conidia was measured inboth WT and ΔAoyap1 strain cultures at each time intervals(0–168 h postinoculation, hpi), after their collection in dis-tilled water with Triton X-100 (Sigma-Aldrich, St. Louis,USA) 0.001% w/v using a Thoma cell count chamber. Todetermine the quantity of ROS (O2

·− and hydroperoxides,ROOH) and to perform molecular analyses, filtered myceliawere lyophilized and weighted.

Table 1 Sequences of the primers used for DNA and cDNA amplifi-cation in A. ochraceus (Ao)

Primer name and sequences

Yapdeg-for (5′- CAG AGC C (CT)A (GC)TT CGA ACC G -3′)

Yapdeg-rev (5′- G(GC)G AGG T(AG)C CAC (AT)GG ATG A -3′)

AoyapEco91I-for (5′- GGT NAC CAT AGA CTG CAG GCG AT -3′)

AoyapStuI-rev (5′- AGG CCT TGC CGG ATC TGG GAA GTC -3′)

AoyapXbaI-for (5′- TCT AGA CAC GCT CTC GCT TTT G -3′)

Aoyap1-for (5′- CCA TGG AAC CCA CAT TAA GC -3′)

Aoyap1-rev (5′- TCC ATC ATC CTC ATC GGA CT -3′)

Aoyap1-rev1 (5′- CAC AAT TTT AAT GAG GCG GCT -3′)

Aoyap1-for1 (5′- TGT GAG ACG GAC TTC CCA GAT CC -3′)

Aoyap1-rev2 (5′- CGA TCC AGT GCT GTT CGG AGA CTG -3′)

Hph-rev1 (5′- ATT TGT GTA CGC CGC CCG ACA GT -3′)

HpF (5′- CAT GGATGC GAT CGC TGC G -3′)

TrR (5′- AGGTACCGTCTAGAAAGAAGGATTAACC -3′)

Ao18S-for (5′- TCA CCA GGT CCA GAC AAA AT -3′)

Ao18S-rev (5′- AGC TGA TGA CTC GTG CCT AC -3′)

Cat-for (5′- T(CT)CATCG(CT)TT(CT)GA(CT)CA(CT)G -3′)

Cat-rev (5′- AATC(ATC)GGGAACTT(GA)A(CT)(CG)G -3′)

Cu/Zn-sod-for (5′- GAC(TC)T(GT)GG(CT)AACTTC(AG)AG -3′)

Cu/Zn-sod-rev (5′- T(AG)CCAGTCTTCTTGGACTC -3′)

Aocat-for (5′- TAC ATC TTG ACT GCA AAC CC-3′)

Aocat-rev (5′- CAG GCC TGT CTT CAC AC -3′)

Aoczsod-for (5′- GAC CTG AGC AAC TTC AAG AC-3′)

Aoczsod-rev (5′- GGA GTC CAA GAA GAC TGG C -3′)

Ao18S-for1 (5′- GTC GTC TGA GGT CGA TTG TAT -3′)

Ao18S-rev1 (5′- TCG ATG CCG GAA CCA AG -3′)

Appl Microbiol Biotechnol

SEM analysis

Samples, i.e., conidiated mycelia and sclerotia, were put insidea low vacuum chamber (10–750 Pa) and observed through aSEM-VP EVO 50 XVP (Carl-Zeiss Electron MicroscopyGroup, Oberkochen, Germany). These samples were processedin VPSE mode using the detector for secondary electrons invariable pressure modality (Goldstein et al. 2003) with energyof 20 KeV. Samples were fixed on an aluminum sample holderstub with a 12.5-mm diameter (Agar Scientific, Stansted, UK).

Analysis of ochratoxin A

OTA analyses were carried out using HPLC as previouslyreported (Solfrizzo et al. 1998), with slight modifications(Eskola et al. 2002).

Detection of ROS by FOX1 and XTT analyses

Mycelia were lyophilized and then homogenized using liquidnitrogen, and 5 mg each were dissolved in 100 μl of 50 mMphosphate buffer pH 7.8 and then centrifuged for 5 min at14,000 rpm at 4 °C. The supernatant was recovered, and 10 μlof FOX1 reagent was added to 990 μl of supernatant. Totalhydroperoxides produced by A. ochraceus WT and mutantstrains grown in several culture conditions, as reported previ-ously, were analyzed in a spectrophotometric assay by mon-itoring the oxidation of xylenol orange (Sigma-Aldrich, St.Louis, USA) at 560 nm (ferrous ion oxidation xylenol orange,FOX-1). Sensitivity of the method was increased using tri-phenylphosphine and stabilizing the reagent at pH 1.7–1.8, asreported by Banerjee et al. (2003). Mycelia, lyophilized andthen homogenized in liquid nitrogen, were dissolved in 100 μlof 50 mM phosphate buffer pH 7.8 in the presence of 0.5 mMof 2,3-bis-(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazoli-um-5-carboxanilide (XTT; Sigma-Aldrich, St. Louis, USA).The samples, rapidly frozen in liquid nitrogen, were held inthe dark for 10 min and then centrifuged at 14,000 rpm for5 min. The quantity of O2

·−was detected in the supernatant byspectrophotometrically measuring (490 nm) the reduction of2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-5-[(phenylamino)carbonyl]-2H-tetrazolium hydroxide (XTT) to formazan,according to a published protocol with the modificationsreported by Castoria et al. (2003). Absorbance at 490 nmwas measured in a microplate reader (Quant Bio-Tek instru-ments, Vermont, USA). XTT levels were expressed as meanabsorbance values mg−1.

Spectrophotometric analysis of SOD and CAT activities

The activity of superoxide dismutase, pH 7.8 and 10.0 (EC1.15.1.1), and catalase (EC 1.11.1.6) was analyzed in thehomogenized mycelia of A. ochraceusWTand mutant strain

as previously described by Reverberi et al. (2005). Theactivity of both enzymes was analyzed using a spectropho-tometer as mixture of the different SOD’s or CAT’s iso-zymes present in the cell crude extract.

Expression of genes related to the antioxidant activities

Total RNA from WT strain and Aoyap1-disrupted mutant my-celia was extracted, as reported in Reverberi et al. (2010), at 12,24, 48, 72, and 168 hpi for the PDB cultures and used to developan Aoyap1, AocatA, and Aoczsod1 real-time PCR (RT-PCR)assay in A. ochraceus. Reverse transcriptase reactions wereperformed using 1 μg of total RNA treated with RNAse-freeDNAse I (Sigma-Aldrich, St. Louis, USA) and 200 U of Super-Script reverse transcriptase (Invitrogen, New York, USA);SYBR green amplification of cDNA was performed in a LineGeneK thermocycler (Bioer, Hangzhou, People’s Republic ofChina), as previously reported (Reverberi et al. 2010). More-over, gene expression in theWTstrain and theΔAoloxAmutantwere also measured by comparing mRNA levels in the differenttime intervals with their own basal expressions at the baseline,i.e., after conidia germination (time 0). The 18S ribosomal RNAwas used as the housekeeping gene to normalize the differencesin total RNA target input and quality and in RTefficiency, usingspecific primers as Ao18S (Table 1).

Statistical analysis

Data are presented as the value of mean ± SD of threeindependent measurements from three separate experiments.In each experiment, data sets were pooled and comparedusing student’s t test, and the differences were consideredsignificant when the p value was <0.05.

Results

Oxidant stressor treatment in A. ochraceus cultures

The mycelial growth and OTA production by A. ochraceus(Table 2) was analyzed in the presence of certain oxidantscompounds. Carbon tetrachloride, menadione, hydrogenperoxide, and tert-butyl hydroperoxide were separatelyadded in the cultures at 5 and 10 days postinoculation(dpi). The mycelial growth of WT was inhibited whenmenadione 0.1 mM was added to the culture medium, incomparison with control, at both the time points. t-BOOH100 mM inhibited the fungal growth up to 5 dpi, and then itwas higher than the control (70 mg DW/ml versus 42 mgDW/ml in the control). In the presence of CCl4, the growthwas similar (at 5 dpi) or higher (10 dpi) compared to theuntreated control. The OTA biosynthesis was inhibited inthe presence of MEN 0.1 mM and H2O2 10 mM, whereas

Appl Microbiol Biotechnol

toxin synthesis was stimulated markedly by the addition ofCCl4 and t-BOOH (~5 times and ~3 times, respectively, at10 dpi) in the culture media, in comparison with the control.

Creation of Aoyap1-disrupted mutant of A. ochraceus

To study the functional role of the Aoyap1 gene in A.ochraceus, some knockout mutants were created. The vector

used for the transformation (pAN7.1, Aoyap1) had twofragments of the Aoyap1 gene as inserts flanking thehygromycin B resistance cassette (hph) (239 bp at 5′ and143 bp at 3′) (Fig. 1a). The A. ochraceus transformants(#40) were grown on PDA supplemented with hygromycinB and screened to verify the vector insertion in the proto-plasts of A. ochraceus. ΔAoyap1 mutant strains were ana-lyzed using PCR, and those which demonstrated moleculardifferences in comparison with WT (i.e., expected ampliconat ~1.7 Kb) were selected (Fig. 1b). The phenotypical,molecular, and physiological characterization was carriedout on one of the selected mutant strains. The genomicorganization of Aoyap1 was studied by Southern blottinghybridization. The results obtained (Fig. 1c) showed thedifferent genomic organizations between WT and somemutant strains (#3). KpnI did not cut inside the sequenceof the probe (Fig. 1a), and this may provide an indication asto the number of copies or alleles of Aoyap1 present in thegenome of A. ochraceus. Owing to the insertion of the3,975 bp construct inside the Aoyap1 gene in the putativemutant strains, the hybridization signal displayed a highermolecular weight in comparison with the WT strain(Fig. 1c). The WT strain showed three hybridization signalsthat could correspond to three copies/alleles, respectively,

Table 2 Growth (mg of DW/ml) and OTA (μg/ml) production by A.ochraceus WT strain inoculated in PDB at 5 and 10 dpi at 25 °C

Treatments Fungal growth(mg DW/ml)

OTA (μg/ml)

5 dpi 10 dpi 5 dpi 10 dpi

Control 40.0±2.5 42.2±3.1 2.7±0.5 2.5±0.4

CCl4 30 mM 38.2±4.5 72.0±6.7 11.3±0.9 10.2±0.3

Menadione0.1 mM

28.5±3.5 34.0±4.5 0.1±0.05 0.1±0.04

H2O2 10 mM 70.0±8.2 65.2±10.5 1.3±0.4 1.2±0.2

t-BOOH100 mM

18.1±0.4 70.3±6.5 6.5±0.5 6.0±0.8

The data are the mean ± SD of three separate experiments

Fig. 1 a Aoyap1 locus of A. ochraceus WT strain; final disruptionconstruct containing hygromycin B and Aoyap1 gene StuI-Eco91I andHindIII-XbaI fragments used for transforming WT protoplast. GenomicDNAwas isolated from the wild-type strain, the Aoyap1 gene replacementtransformant, and digested with KpnI. b PCR amplification of genomicDNA isolated fromWT (lanes 1–2) and some hygromycin-selected colony(lanes 3–10) using primer combination Aoyap1-for and hph-rev1. c KpnI-

digested DNA of WT and three putative Aoyap1-disrupted mutants washybridized with Aoyap1 (lane 1, WT strain; lanes 2–4, PCR-selectedtransformants) or with hph (lane 5, WT strain; lanes 6–8, PCR-selectedtransformants) PCR DIG-labeled probes. d Relative expression of theAoyap1 gene in WT and in ΔAoyap1 mutant strains at different timepostinoculation (12–168 hpi)

Appl Microbiol Biotechnol

of ~9, ~4, and ~2 Kb; the high MW fragment, i.e., ~9 Kb,indicating it was probably the target of the knockout, since ashift of ~4 Kb corresponding to a fragment of ~13 Kb wasevident in the mutant strains. The hybridization with theprobe hph reported positive signals (at ~13 Kb) only in theΔAoyap1 strains. The expression of Aoyap1 mRNA wasanalyzed in the WT strain and in one of the Aoyap1-dis-rupted mutant strains. It emerged clearly that the Aoyap1mRNAwas downregulated in ΔAoyap1 in comparison withthe WT strain in all the time intervals considered (Fig. 1d)even if a basal expression of Aoyap1, apparently due to theother gene copies/alleles present in A. ochraceus genome(Fig. 1c), is still present in the mutant strain.

Phenotypic analysis of the mutant strains

ΔAoyap1 strain was first characterized at the morphologicallevel (Fig. 2a–d). In liquid media, the transformed straingrew less and formed a lower amount of conidia in compar-ison with the WT strain (Fig. 2e). In solid media (Fig. 2a),the WT grew similarly to the mutant strain but, at 168 hpi,1.1×108/ml±0.5×106 conidia in WT and 5.2×107/ml±2.5×

105 conidia inΔAoyap1 strain were produced at 168 hpi (p<0.001). As shown in Fig. 2b, c, the size of conidia wasdifferent in the WT (diameter 2.309±0.0015 μm) and in theΔAoyap1 strain (diameter 2.634±0.085 μm) (p<0.05); theconidia of the mutant strain were larger than those of the WTstrain. Furthermore, sclerotia were formed only in the mutantstrain, both in liquid and solid media (Fig. 2d).

Mycelial growth and OTA biosynthesis in A. ochraceus WTand ΔAoyap1 strains in the presence and in the absenceof CCl4

In Fig. 3a, b, the growth and OTA synthesis of WT andmutant strain in the presence and absence of CCl4 as oxidantstressor in different time intervals (12–168 hpi) are shown.In the WT strain, OTA was produced starting from24–48 hpi (Fig. 3a), although in small amount, whereas inthe presence of CCl4, its synthesis started later at 72 hpi;then it was stimulated (10 μg/ml in the mutant and 3 μg/mlin the WT), whereas the untreated samples were not stimu-lated (Fig. 3a). OTA biosynthesis (Fig. 3a) occurred earlierin the ΔAoyap1 strain in comparison with the WT, indeed,

b) c) d)b) c) d)

e

24 hpi 48 hpi 72 hpi 96 hpi 168 hpi

2 m 1 m 200 m

a

b c d

WT

Aoyap1

Fig. 2 a Agar plates (PDA)showing the different timing inconidia formation of WT andΔAoyap1 mutant strain atdifferent time intervals afterinoculation (24–168 hpi). SEMimages (14,000× magnification)of conidia derived from PDBcultures (b) of WT and (c) ofΔAoyap1 strain. d SEM image(150× magnification) ofsclerotia derived from PDAculture of ΔAoyap1 strain. eAmount of conidia formation inPDB cultures of WT andΔAoyap1 mutant strains atdifferent time intervals afterinoculation (24–168 hpi). Thedata are the mean ± SD of threedeterminations from threeseparate experiments. Asterisksin 2e indicate a level ofsignificance of p<0.001

Appl Microbiol Biotechnol

it was already detectable (1.12 μg/ml) at 12 hpi; moreover,OTA synthesis was significantly higher in the ΔAoyap1strain at 24 h compared to WT, and its maximum levelwas reached at 168 hpi (~20 μg/ml) (p<0.001). As regardsthe ΔAoyap1 strain, the presence of CCl4 inhibited OTAproduction and also delayed and reduced its growth (Fig. 3a,b) (p<0.001 up to 72 hpi; p00.072 at 168 hpi). The presenceof CCl4 strongly affected the growth in the WT (Fig. 3b), atleast up to 72 hpi (p<0.001 up to 72 hpi; p00.068 at 168 hpi).

Reactive oxygen species formation in A. ochraceus WTand ΔAoyap1 strains

To investigate whether the disruption of the Aoyap1 gene in A.ochraceus may disturb the redox balance, the formation ofsome reactive oxygen species in the WT and in the mutantstrains was monitored (Fig. 4a, b). The ROOH formation washigher in the mutant strain than that in theWT, at 24 and 36 hpi(Fig. 4a) (p<0.01), whereas at 48 and 72 hpi, the ROOHproduction is higher, even if (p00.062) nonsignificantly, incomparison with the WT. As regards the superoxide anion

(Fig. 4b), its level was higher in the mutant strain comparedwith the WT, from 24 to 168 hpi (p<0.01 at 72 and 168 hpi).The major differences between WT and ΔAoyap1 were ob-served from 24 hpi; notably, the O2

·− level in the mutant moresignificantly increased at 72 and 168 hpi (p<0.001).

Catalase and Cu,Zn-superoxide dismutase mRNA expressionand enzymatic activity in WT and ΔAoyap1 strains

To further investigate the role played by the Aoyap1 gene inthe regulation of the redox balance in A. ochraceus, westudied the expression of a catalase and a Cu,Zn-superoxidedismutase gene. Since these genes were not annotated inGenBank, we partly sequenced them using a degeneratedprimers—PCR-based approach. The 309 bp of AocatA(FJ460503) gene fragment showed a high percentage ofsimilarity (90%) to the catA gene of Aspergillus clavatus(XM001274803), encoding for a catalase, whereas the233 bp of Aoczsod1 (FJ460502) showed an aminoacidicsimilarity of 95% to the sod gene of Aspergillus fumigatus(AF281057), encoding for a Cu,Zn-superoxide dismutase.

Fig. 3 a Ochratoxin Aproduction in culture filtrate(μg/ml) of A. ochraceus WTand Aoyap1 strains inoculatedinto OTA-conducive medium(PDB) amended or not withCCl4 30 mM. b Mycelialgrowth (mg/ml DW) of WT andAoyap1 strains inoculated inPDB amended or not with CCl430 mM and incubated at 25 °Cfrom 12 up to 168 hpi. The dataare the mean ± SD of threedeterminations from threeseparate experiments. Asterisksin 3a and 3b indicate a level ofsignificance of p<0.001

Appl Microbiol Biotechnol

Both genes were, in general, downregulated in themutant strain in comparison with the WT in almost allthe time intervals considered (0–168 hpi) (Fig. 5a, b).Furthermore, in the mutant strain, the activity of CATenzyme, evaluated from crude extracts, was significantlylower (p<0.001) in comparison with the WT until48 hpi (Fig. 6a). Activity of SOD at different pH (7.8and 10.0) was studied from the cell crude extracts ofWT and ΔAoyap1 mycelia. The mutant strain presenteda lower (p<0.01) or similar level of SOD activities incomparison with the WT strain both at pH 7.8 and 10.0between 36 and 48 hpi (Fig. 6b, c). In general, CATand SOD activity at pH 7.8 and 10.0 were similar in allthe samples analyzed during the last phase of the ex-periment (from 72 hpi); this suggests that the differen-tial responses of the mycelia were displayed in the first3 days of incubation. The differences between RT-PCRdata and enzyme activity measurement (no decline inactivity but remarked reduction in expression level)might be explained by the assumption that severalenzymes could be measured together in the cell crudeextracts.

Discussion

This research adds new clues to the biology of A. ochraceusand, in particular, to oxidative stress-related factors such asproduction and OTA biosynthesis. This fungus is infamousfor its ability to produce OTA in various commodities suchas wheat, wine and coffee, and its ubiquity in temperateclimates makes it still more dangerous.

In this study, oxidative stressors such as t-butyl hydro-peroxide and the prooxidant carbon tetrachloride were used(Kim et al. 2004; Passi et al. 1985); these compounds areable to significantly induce OTA biosynthesis in A. ochra-ceus, even if to different extents. It had recently been sug-gested that oxidative stress is involved in the secondarymetabolism regulation in A. ochraceus (Reverberi et al.2010). Therefore, an Aoyap1-deleted mutant strain wasgenerated in order to provide further supporting evidenceof the existence of this relation. Yap1 is a stress-responsivecytoplasm/nuclear transcription factor, which is activated bya shift of the cell redox balance towards a hyperoxidantstatus in yeast and several filamentous fungi (Moye-Rowley2003; Aguirre et al. 2005; Reverberi et al. 2008; Molina and

Fig. 4 a ROOH and ionsuperoxide formation (b) inthe mycelia of A. ochraceusinoculated in PDB andincubated at 25 °C for differenttimes (12, 24, 36, 48, 72, and168 hpi). The data are themean ± SD of threedeterminations from threeseparate experiments. Asterisksin 4a and 4b indicate a level ofsignificance of p<0.01

Appl Microbiol Biotechnol

Kahmann 2007). Under such conditions, Yap1 migrates intothe nucleus and activates the transcription of several oxida-tive stress-responsive genes, such as sod, cat, and glutathi-one peroxidase (gpx). This response is aimed at restoring theproper redox balance in the cell.

It had already been demonstrated that Yap1 is involved inaflatoxin biosynthesis (Reverberi et al. 2008). Moreover, thefinding of Yap1-responsive elements (YRE) in the promoterof the aflatoxin gene cluster regulator, aflR, suggested thatoxidative stress can directly modulate the synthesis of thetoxin in A. parasiticus and indirectly modulate some otheraspects of the secondary metabolism, such as morphogenesisthrough ApYapA (Reverberi et al. 2008; Roze et al. 2011).

In A. ochraceus, the Aoyap1 gene contributes to control-ling the redox balance in the cell. In fact, the Aoyap1-disrupted strain shows a higher quantity of unscavengedROS, which is to be expected given that Aoczsod1 andAocatA genes and their related enzymatic activities aredownregulated. Thus, a ΔAoyap1 strain “experiencing” ahyperoxidant status over its entire life cycle produces ahigher amount of OTA compared to a WT strain. The

inability of the mutant to counteract oxidative processes isalso demonstrated by the results obtained by treating theΔAoyap1 strain with the potent prooxidant CCl4 (Halliwelland Gutteridge 2007). This strain exhibits the inhibition ofgrowth and OTA biosynthesis in these culture conditions. Itcan be hypothesized that ΔAoyap1 diverts a portion of theexcess oxidants towards the secondary metabolism machin-ery, in particular, towards OTA synthesis, to enable it tosurvive chronic hyperoxidant conditions. If a surplus ofoxidants is present in the environment, i.e., in the CCl4-treated samples, ΔAoyap1 is unable to control the oxidantexcess and “decides” to switch the secondary metabolismoff, favoring colony survival instead, as it is demonstratedby the fungal growth and OTA production in the ΔAoyap1and ΔAoyap1+ CCl4 samples (Fig. 3a, b).

Our study also indicates oxidative stress as affecting somedifferentiation processes in A. ochraceus, since sclerotia for-mation is stimulated only in the mutant strain, starting fromthe early stage of growth. In filamentous fungi, sclerotia areformed as a consequence of oxidative stress damage. Manyfungi, such as Sclerotium rolfsii, differentiate sclerotia in an

Fig. 5 a AocatA and (b)Aoczsod1 mRNA expression inΔAoyap1 strain relative to WTstrain gene expressionsinoculated in PDB andincubated at 25 °C for differenttimes (12, 24, 36, 48, 72, and168 hpi). The results are themean ± SD of threedeterminations from threeseparate experiments

Appl Microbiol Biotechnol

attempt to insulate themselves from environmental oxygen, asreported by Georgiou et al. (2006). In the mutant strain, thedownregulation of the antioxidant defenses caused by theabsence of the Aoyap1 gene creates an intermediate unstablestate in the fungal cell, which is the hyperoxidant state. In thisstate, it is probable that the fungal cell differentiates sclerotia

in order to reduce O2 and limit oxygen entrance, therebyreducing the risk of cell death (Georgiou et al 2006). Thisevent correlates oxidative stress to sclerotia formation as away of enabling the fungal cell to survive.

In our opinion, it is worth further exploring the role playedby oxidative stress modulation, since it is increasingly being

Fig. 6 a Catalase (CAT) and(b) superoxide dismutases(SOD) at pH 7.8 and (c)pH 10.0 activity expressedas unit/mg protein. The resultsare the mean ± SD of threedeterminations from threeseparate experiments. Asterisksin 6a indicate a level ofsignificance of p<0.001and in 6b–c, of p<0.01

Appl Microbiol Biotechnol

found as a “backgroundmotif” in fungal metabolism (Fox andHowlett 2008; Fisher 2008; Lledias et al. 1999; Pallardò et al.2009). The information reported herein adds a further piece tothe OTA “jig-saw puzzle,” namely, oxidative stress which isnow to be considered a common trait in the regulation of thebiosynthesis of different mycotoxins (it has already beendemonstrated as a significant factor in the biosynthesis ofaflatoxins and Fusarium toxins and even involved in patulinproduction).

All of these evidences suggest that oxidative stress shouldbe considered the principal target in strategies to control OTAsynthesis using environmental-friendly antioxidants. This willbe the subject of future studies.

References

Aguirre J, Rios-Momberg MD, Hewitt D, Hansberg W (2005) Reactiveoxygen species and development in microbial eukaryotes. TrendsMicrobiol 13:111–118

Aguirre J, Hansberg W, Navarro R (2006) Fungal responses to reactiveoxygen species. Med Mycol 44:101–107

Banerjee D, Madhusoodanan UK, Sharanabasappa M, Ghosh S, JacobJ (2003) Measurement of plasma hydroperoxide concentration byFOX-1 assay in conjunction with triphenylphosphine. Clin ChimActa 337:147–152

Calvo AM, Wilson RA, Bok JW, Keller NP (2002) Relationshipbetween secondary metabolism and fungal development. Micro-biol Mol Biol Rev 66:447–459

Castoria R, Caputo L, De Curtis F, De Cicco V (2003) Resistance ofpost-harvest biocontrol yeasts to oxidative stress: a possible newmechanism of action. Phytopathol 93:564–572

Cheeke PR (1998) Natural toxicants in feeds, forages and poisonousplants. Interstate, Danville

Commission of the European Community (2006) Regulation (EC) no.1881/2006 of 19 December 2006: setting maximum levels forcertain contaminants in foodstuffs. Off J Eur Union L 364:5–24

Creppy EE, Chakor K, Fisher MJ, Dirheimer G (1990) The mycotoxinochratoxin A is a substrate for phenylalanine hydroxylase inisolated rat hepatocytes and in vivo. Arch Toxicol 64:279–284

Eskola M, Kokkonen M, Rizzo A (2002) Application of manual andautomated systems for purification of ochratoxin A and zearale-none in cereals with immunoaffinity columns. J Agric Food Chem50:41–47

Estruch F (2000) Stress-controlled transcription factors, stress-inducedgenes and stress tolerance in budding yeast. FEMS Microbiol Rev24:469–486

Fisher R (2008) Sex and poison in the dark. Science 320:1430–1431Fox ME, Howlett BJ (2008) Secondary metabolism: regulation and

role in fungal biology. Curr Op Microbiol 11:481–487Georgiou CD, Patsoukis N, Papapostolou I, Zervoudakis G (2006)

Sclerotial metamorphosis in filamentous fungi is induced byoxidative stress. Integr Comp Biol 46:691–712

Goldstein J, Newbury D, Joy D, Lyman C, Echlin P, Lifshin E, SawyerL, Michael J (2003) Scanning electron microscopy and x-raymicroanalysis, 3rd edn. Springer, New York, p 689

Halliwell B, Gutteridge JMC (2007) Cellular responses to oxidative stress:adaptation, damage, repair, senescence and death. In: Halliwell B,Gutteridge JMC (eds) Free radicals in biology and medicine. OxfordUniversity Press, New York, pp 187–267

IARC (1993) Some naturally occurring substances: food items andconstituents, heterocyclic aromatic amines, and mycotoxins.IARC monographs on the evaluation of carcinogenic risk ofchemicals to humans, vol 56. International Agency for Researchon Cancer, Lyon, p 599

Jayashree T, Subramanyam C (2000) Oxidative stress as a prerequisitefor aflatoxin production by Aspergillus parasiticus. Free Rad BiolMed 29:981–985

Kim JH, Campbell BC, Yu J, Mahoney N, Chan KL, Molyneux RJ,Bhatnagar D, Cleveland TE (2004) Examination of fungal stressresponse genes using Saccharomyces cerevisiae as a model sys-tem: targeting genes affecting aflatoxin biosynthesis by Aspergil-lus flavus Link. Applied Microbiol Biotechnol 67:807–815

Lledias F, Rangel P, Hansberg W (1999) Singlet oxygen is part of ahyperoxidant state generated during spore germination. Free RadBiol Med 26:1396–1404

Molina L, Kahmann R (2007) An Ustilago maydis gene involved inH2O2 detoxification is required for virulence. Plant Cell 19:2293–22309

Moye-Rowley WS (2003) Regulation of transcriptional response tooxidative stress in fungi: similarities and differences. EukaryotCell 2:381–389

Narasaiah KW, Sashidhar RB, Subramanyam C (2006) Biochemicalanalysis of oxidative stress in the production of aflatoxin and itsprecursor intermediates. Mycopathol 162:179–189

O’Callaghan J, Caddick MX, Dobson ADW (2003) A polyketidesynthase gene required for ochratoxin A biosynthesis in Aspergil-lus ochraceus. Microbiol 149:3485–3491

O’Callaghan J, Stapleton PC, Dobson ADW (2006) Ochratoxin Abiosynthetic genes in Aspergillus ochraceus are differentiallyregulated by pH and nutritional stimuli. Fungal Genet Biol43:213–221

Pallardò FV, Markovic J, Garcìa JL, Viña J (2009) Role of nuclearglutathione as a key regulator of cell proliferation. Mol AspectsMed 30:77–85

Pardo E, Marin S, Ramos A, Sanchis V (2004) Occurrence of ochra-toxigenic fungi and ochratoxin A in green coffee from differentorigins. Food Sci Technol Int 10(1):45–49

Passi S, Fanelli C, Fabbri AA, Finotti E, Panfili G, Nazzarro-Porro M (1985) Effect of halomethanes on aflatoxin induc-tion in cultures of Aspergillus parasiticus. J Gen Microbiol131:687–691

Payne GA, Hagler WM (1983) Effect of specific aminoacids on growthand aflatoxin by Aspergillus parasiticus and A. flavus in definedmedia. Appl Environ Microbiol 46:805–812

Ponts N, Pinson-Gadais L, Verdal-Bonnin MN, Barreau C, Richard-Forget F (2006) Accumulation of deoxynivalenol and its15-acetylated form is significantly modulated by oxidative stressin liquid cultures of Fusarium graminearum. FEMS MicrobiolLett 258:102–107

Reverberi M, Fabbri AA, Zjalic S, Ricelli A, Punelli F, Fanelli C (2005)Antioxidant enzymes stimulation in Aspergillus parasiticus byLentinula edodes inhibits aflatoxin production. Appl MicrobiolBiotechnol 69:207–215

Reverberi M, Zjalic S, Ricelli A, Punelli F, Camera E, Fabbri C,Picardo M, Fanelli C, Fabbri AA (2008) Modulation of antiox-idants defence in Aspergillus parasiticus is involved in aflatoxinbiosynthesis: a role for ApyapA gene. Eukaryot Cell 7(6):988–1000

Reverberi M, Punelli F, Scarpari M, Camera E, Zjalic S, RicelliA, Fanelli C, Fabbri AA (2010) Lipoperoxidation affectsochratoxin A biosynthesis in Aspergillus ochraceus and itsinteraction with wheat seeds. Appl Microbiol Biotechnol85:1935–1946

Roze LV, Chanda A, Wee J, Awad D, Linz JE (2011) Stress-relatedtranscription factor AtfB integrates secondary metabolism with

Appl Microbiol Biotechnol

oxidative stress response in Aspergilli. J Biol Chem 286(40):35137–35148

Solfrizzo M, Avantaggiato G, Visconti A (1998) Use of various clean-up procedures for the analysis of ochratoxin A in cereals. JChromat 815:67–73

Stoev S (1998) The role of ochratoxin A as a possible cause of Balkanendemic nephropathy and its risk evaluation. Vet Hum Toxicol40:352–360

Tolaini V, Zjalic S, Reverberi M, Fanelli C, Fabbri AA, Ricelli A (2010)Lentinula edodes enhances the biocontrol activity of Cryptococcuslaurentii against Penicillium expansum contamination and patulinproduction in apple fruits. Int J Food Microbiol 138:243–248

Van der Merwe KJ, Steyne KJ, Fourie L, Scott DB, Theron JJ (1965)Ochratoxin A, a toxic metabolite produced by Aspergillus ochra-ceus Wilh. Nature 205:1112–1113

WHO (2001) Safety evaluation of certain mycotoxins in food. Fiftysixth report of the joint FAO/WHO expert committee on foodadditives. WHO Food Additives Series 47. World Health Organi-zation, Geneva, p 700

Wilkinson JR, Yu J, Bland JM, NiermanWC, Bhatnagar D, Cleveland TE(2007) Amino acid supplementation reveals differential regulationof aflatoxin biosynthesis in Aspergillus flavus NRRL 3357 andAspergillus parasiticus SRRC 143. Appl Microbiol Biotechnol 74(6):1308–1319

Appl Microbiol Biotechnol