Introduction
Aflatoxin is a secondary metabolite produced mainly by Aspergillus flavus and Aspergillus parasiticus. As the most potent naturally-occurring toxic and carcinogenic substance, aflatoxin causes an estimated 28% of hepatocellular carcinoma (HCC), while HCC is the most common form of liver cancer in the world 39 and the case rate is very high in sub-Saharan Africa, the Western Pacific region and Southeast Asia, as well as in Central America. Individuals with liver damage due to hepatitis B virus (HBV) infection are particularly vulnerable to aflatoxin invasion 14. In addition, aflatoxin can lead to dysfunction of the immune system, dysplasia in children, and even death due to acute poisoning 10 18. Aflatoxin contamination occurs in a wide range of food and feed commodities, including wheat, maize, peanuts, rice, peanut oil, cotton seed, milk, nuts and dairy products 2. Therefore, aflatoxin not only poses a serious threat to human and animal health, but also causes huge economic losses.
The gene cluster involved in aflatoxin biosynthesis has been identified 3 5 9 13 2 41 44. Most gene functions have been clarified 1 9. AflR and AflS are two key transcription factors. The aflatoxin regulatory gene aflR activates the transcription of other structural genes in the aflatoxin biosynthesis pathway by encoding a positive regulatory factor 6. AflS is adjacent to AflR and participates in the regulation of aflatoxin biosynthesis together with AflR. The combination of AflS and AflR forms a complex, which is then bound together in the promoter region of each structural gene in the cluster 7. In addition to aflR and aflS, there are many regulatory genes involved in aflatoxin biosynthesis regulation outside the aflatoxin gene cluster. laeA and veA encode global transcription factors that regulate the biosynthesis of many secondary metabolites, such as aflatoxin, sterigma-tocystin, and penicillin 4 11 12 19.
There is extensive evidence that secondary metabolism is associated with oxidative stress in filamentous fungi and plants 16 7 36. Based on this view, different oxidative stimuli, such as peroxides and diamide, can activate a variety of transcription factors, and many transcription factors have been proved to be involved in regulating secondary metabolism in yeast, fungi and plants. Within this network, the well-known Ap-1 transcription factor Yap-1 participated in the cellular response to oxidative stress signal in Saccharomyces cerevisiae25 33.
Like the Yap-1 roles in yeast 26, several Yap-1 homologue transcription factors have been identified in filamentous fungi, and they are usually associated with resistance to H 2O 2 or antifungals. In the rice blast fungus Magnaporthe oryzae, Moap1 mediates the oxidative stress response and is necessary for conidia formation, apical growth and pathogenicity 15. Afyap1 in Aspergillus fumigates was found to be associated with tolerance to oxidative stress 29. NapA and RsmA affect stress response, sexual development and secondary metabolism in Aspergillus nidulans42. In Aspergillus ochraceus, Aoyap1 not only participated in the oxidative stress response, but also regulated ochratoxin A biosynthesis. Similarly, to Aoyap1, ApyapA in A. parasiticus also participated in the oxidative stress response and in the modulation of aflatoxin biosynthesis 30. These findings suggested a probable similar link between the oxidative stress response and mycotoxin biosynthesis. However, under the oxidative stress condition, the mechanism of yap-1 homologue gene in the regulation of aflatoxin biosynthesis in A. flavus is not clear.
In this paper, afap1, the homologue of yap-1, was suggested to encode protein containing conserved bZIP domains based on the NCBI BLAST analysis. We engineered genetically modified strains of A. flavus lacking afap1 and showed the key role played by afap1 in response to oxidative stress and in the regulation of aflatoxin biosynthesis.
Materials and methods
Strains and growth conditions
The toxigenic A. flavus CA14PTs (Δku70, ΔniaD) and recipient (Δku70, ΔniaD, ΔpyrG) strains were obtained from Dr. Perng Kuang Chang, United States Department of Agriculture, New Orleans, USA. The strain A. nidulans WJAO1 was obtained from Prof. Shihua Wang, Fujian Agriculture and Forestry University, Fuzhou, China.
Strains were activated on potato dextrose agar (PDA) plates (20g/l dextrose, 200g/l peeled potatoes and 20 g/l agar) at 28 °C in the dark for 3 days for conidia production. Coni-dial suspensions were collected from sporulated cultures of fungi on PDA plates by surface washing with sterile deionized water containing 0.1% Tween-20. The number of conidia in the suspensions was counted using a hemocytometer and diluted to 10 6CFU/ml with 0.1% Tween-20 solution. Conidia were cultivated in 50 ml YES medium (150g/l sucrose, 20 g/l yeast extract and 1 g/l MgSO 47H 20 and solid medium supplemented with 16 g/l agar) and grown at 28 °C on a rotary incubator in the dark for AFB1 concentration detection and mycelia collection. The recipient strain was grown in broth containing yeast, glucose, trace element solution, uracil, uridine (YGTUU) (20g/l glucose, 5 g/l yeast extract, 1 ml trace element solution per liter of medium, 1 g/l uracil and 1 g/l uridine and solid medium supplemented with 15 g/l agar) at 28°C for mycelial growth and conidia production. Czapek-Dox medium (Difco) supplemented with 3% sucrose was used for mutant selection.
Hydrogen peroxide sensitivity analysis
Five microliters of conidia (10 6CFU/ml medium) from each strain were incubated in YES solid medium supplemented with different concentrations of H 20 2 (0, 5, 10, 20, and 40mmol/l) for the oxidative stress response assay. All plates were cultivated in the dark at 28 °C for 5 days, and colonies were photographed. Meanwhile, 5 ml of conidial suspension (10 6CFU/ml medium) was cultured in 50 ml YES liquid medium supplemented with the same concentrations of H 20 2 for AFBi concentration and the mycelial dry weight analysis. All experiments were performed in triplicate in three independent experiments.
AFB 1 concentration and fungi mycelial dry weight analysis
Mycelia were collected and mycelial dry weight was measured after drying in a dryer (HASUC, Inc., Shanghai, China) at 65°C for 72 h. AFB1 concentration in the culture filtrate was determined by extracting metabolites from the filtrate using methanol, followed by purification using an immunoaffinity column (Romer Labs, Inc., Tulln, Austria) according to the manufacturer's instructions. AFBi concentration was detected by high-performance liquid chromatography (HPLC; Agilent Series 1260; Agilent Technologies, Santa Clara, CA, USA). HPLC was performed on an Agilent C18 Zorbax XDB column (150 mm x 4.6 mm x5mm, Agilent Technologies), and detection was performed using a fluorescence detector (Agilent 1260; Agilent Technologies) with an excitation wavelength of 360 nm and an emission wavelength of 440 nm at 30°C. The mobile phase consisted of methanol/H 2O (7:3, v/v) injected at a flow rate of 1 ml/min.
RNA extraction and quantification of gene expression by quantitative real-time PCR (qRT-PCR)
Total RNA was extracted from fungal mycelia collected from YES liquid medium using a RNeasy mini kit (Qiagen, Germany) according to the manufacturer's instructions. RNA samples were treated with DNA-free DNase. The purity and concentrations of RNA were determined by measuring the absorbance of samples at 260 and 280nm using spectrophotometric quantification in a Beckman DU800 spectrophotometer (Beckman, USA). qRT-PCR was carried out in triplicate in 20-μl volumes using Power SYBR green master mix (Applied Biosystems, USA) in an ABI 7500 Real-Time PCR System (Applied Biosystems, USA). The thermal-cycling program was set as follows: 95 °C for 5 min, followed by 40 cycles of 95 °C for 25 s, 55 °C to 60 °C for 25 s (optimized for each primer pair), and 72 °C for 35 s, with a melting-curve stage at 95 °C for 15 s, 60 °C for 1 min, 95 °C for 30 s, and 80 cycles of 60 °C for 15 s. Gene-specific primers were designed for each target gene using Primer 5.0 ( http://www.premierbiosoft.com/primerdesign/) ( Table 1). As an endogenous control, primers 18S-F (5-GCTCTTTTGGGTCTCGTAATTGG-3) and 18S-R (5-CGCTATTGGAGCTGGAATTACC-3) were used based on previous studies in order to cover 154 bp of the 18S RNA gene 21. Samples from each of the three biological replicates were assayed in triplicate, and data were analyzed using the ABI 7500 SDS program (Applied Biosystems, USA) by the 2-44ct method.
Identification of Afap1
The sequence of Afap1 and its homologues in S. cerevisiae, A. parasiticus, A. nidulans and Aspergillus fumigatus were used as input for BLAST in the National Center for Biotechnology Information (NCBI) database ( https://blast.ncbi.nlm.nih.gov/Blast.cgi), to identify sequences with high similarities in the translated genome of A. flavus. Multiple sequence alignments were carried out using DNAssist 2.2.
Construction of the mutant strain
The deletion-mutant strain (Δafap1) was constructed as previously described 8 37. A homologous transformation system for A. flavus with the pyrG gene as selection marker was used in this study 35. The pyrG gene encodes an orotidine-5 -phosphate decarboxylase, which is a key gene in the synthesis of uracil nucleotides. The recipient strain CA14PTs ( Δku70, ΔniaD, ΔpyrG) with the pyrG deletion cannot grow on the transformant selection medium Czapek-Dox without adding uracil and uridine, while homologous transformants carrying the pyrG gene instead of afap1 could survive on the selection medium. The detailed procedure is as follows: Genomic DNA was extracted from mycelia grown for 5 days in 50 ml of YES liquid medium using benzyl chloride 38 45. For the homologous fragments, the 5' and 3' regions of afap1 (1159 and 1046 bp, respectively) were amplified with specific primer pairs Up-F/R and Down-F/R ( Table 1), which contain sequences that overlap the marker gene, and were verified by sequencing. The 1600 b p <i>pyrG gene was amplified with primer pairs pyrG-F/R from A. nidulans WJAO1 genomic DNA. The PCR-fusion product was constructed and transformed into recipient strain protoplasts using polyethylene glycol buffer (15mM KCl, 20 mM CaCl2 and 1MTris-HCl buffer, pH 7.5), and 500g/l PEG 4000. The cell suspension was plated on Czapek-Dox medium at 28 °C in the dark for 5 days. Putative mutants were confirmed by PCR using primer pairs Middle-F/R and Up-F/pyrG-R and sequencing analysis.
Southern blot
Transformants that passed the PCR pre-screening were further checked by Southern blot analysis, using the DIG system (Roche, Germany) in accordance with a previously described protocol 24. Ten μg genomic DNA was digested with Hindi!! (Takara, Japan), and then electrophoresed on a 1% agarose gel to separate by size. A sheet of nylon membranes (Hybond N+, Pharmacia, USA) was placed on top of the gel for DNA transference. The hybridization probe with DIG-labeled was synthesized with PCR DIG probe synthesis kit (Roche, Germany) by following the manufacturer's protocol. The probe matched the downstream sequence of the homology arm in the homologous recombination fragment was generated by PCR amplification using primers 5 -AACGTGGTTGTATTTGCCCC-3' and 5 -GCTCTTGGACAATGCTCTCG-3'.
Results
Effect of different oxidative stress on A. flavusgrowth and AFB 1 production
Oxidative stress is a very important environmental stimulus for fungi. To evaluate the effect of oxidative stress on the growth and AFB 1 production of A. flavus, H 2O 2 solutions at different concentrations (0, 5, 10, 20, and 40mmol/l) were added to YES plates and liquid culture medium, respectively. The growth of A. flavus CA14PTs was significantly inhibited by the increased H 2O 2 and were completely inhibited at 40mmol/l ( Fig. 1A). TheAFB 1 concentration in YES broth was increased after the treatment of H 2O 2 at the concentration of 5, 10 and 20mmol/l, respectively ( Fig. 1B). There is no obvious increase for AFBi concentration at 20mmol/l H 2O 2 compared with 10mmol/l. In addition, the expression levels of key aflatoxin biosynthetic structural genes (aflJ, aflM, aflO, aflP and aflX) were up-regulated by the treatment of 10mmol/l H 2O 2 according to qRT-PCR ( Fig. 1C). The data based on the results indicated that oxidative stress could affect strain growth and stimulate aflatoxin biosynthesis.
Identification of Afap1, a Yap-1 homologue in A. flavus
By the NCBI BLAST analysis, Afap1 was identified as a putative bZIP transcription factor. Alignment of the Afap1 protein sequence to those of Yap-1 ( S. cerevisiae S288c), ApyapA (A. parasiticus SU-1), NapA (A. nidulans FGSC A4) and Afyap1 (A. fumigatus Af293) ( Fig. 2) showed two conserved domains: A C-terminal nuclear export signal (NES) embedded in a characteristic cysteine-rich domain (c-CRD) and a N-terminal basic leucine zipper domain (bZIP domain). Afap1 has lower homology (16.22% similarity) with its yeast orthologues, but it has higher homology with other filamentous ascomycetes. The conserved bZIP domain and cystein-rich domain (CRD) suggested that Afap1 has a similar role in response to oxidative stress and toxin biosynthesis in A. flavus as other homologue proteins.
Confirmation of Aafap1 deletion mutants
To analyze the role of afap1 in oxidative-stress response and aflatoxin biosynthesis, Aafap1 mutants were generated using homologous recombination ( Fig. 3A). Two transformants (i.e., Aafap1-1 and Aafap1-2) were selected for further PCR verification. A 700bp fragment, encoding for partial ORF of afap1, could be amplified with primers Middle-F/R in CA14PTs but not in the positive transformants. A 2800 bp fragment, encoding for pyrG and upstream of afap1, could only be amplified with primers Up-F/pyrG-R in positive transformants. As shown in Figure 3B, only the 2800 bp fragment was observed in Aafap1-1 and Aafap1-2. Further sequencing analysis showed that the gene afap1 was exactly replaced by pyrG in these two mutants. Additionally, southern blot hybridization revealed a 3.4kb fragment and a 2.2 kb fragment in CA14PTs and the afap1 mutants when digested with HindIII, respectively ( Fig. 3C). It is confirmed that there are sequence differences between the afap1 mutants and CA14PTs as expected. Combined with the above homologous recombination strategy and PCR analysis, it was shown that afap1 was properly deleted and mono-copy.
Effects of afap1 deletion on sensitivity to oxidative stress
CA14PTs and the Aafap1 mutants were incubated in YES plates supplemented with different concentration of H2O2 ( Fig. 4A). The growth rates of the CA14PTs and Aafap1 mutants were similar on YES plates without H 2O 2. However, the growth of Aafap1 mutants was completely inhibited by 5mmol/l H 2O 2. In contrast, the growth of CA14PTs was not inhibited even at 10mmol/l H 2O 2. Meanwhile, AFBi concentration of the Aafap1 mutants was significantly decreased by around 75%, which compared to CA14PTs ( p < 0.01) ( Fig. 4B). In addition, the expression levels of the key transcription factors genes (tcsA, bos1, atfB and srrA) related to the oxidative-stress response and aflatoxin biosynthetic genes (aflD, aflB, aflR, aflS, aflM and aflP) in YES liquid medium were detected using qRT-PCR. The expression of tcsA, bos1, srrA, aflB and aflR was significantly up-regulated in the Aafap1 mutants compared to CA14PTs, and the expression of aflM and aflP were significantly down-regulated. The expression of atfB was up-regulated in the Aafap1 mutants, although the difference was not significant ( Fig. 5).
Discussion
Oxidative stress is one of the earliest responses and a common cell defense mechanism in living things. Cellular response to oxidative stress plays a crucial role in plants, vertebrates and fungi; it enables the cell to survive a variety of extra- and intracellular oxidative stressors. The classical review of the oxidative stress response in fungi was developed based on research in yeast which showed that regulation of defense-related antioxidant genes contributed to the survival of the organism. The regulation of secondary metabolism is closely linked to the cellular response to oxidative stress in filamentous fungi and contributes to the complexity of the response 16. However, this response is most complicated and robust than that of yeast in response to various environmental conditions.
Previous reports strongly suggested that several transcription factors associated with the Stress Activated Protein Kinase/Mitogen-Activated Protein Kinase (SAPK/MAPK) pathway coordinate the transcriptional level of secondary metabolism genes and antioxidant enzymes, thereby controlling the metabolic processes in cellular stress response. Ap-1 family is one of the most important transcription factors. Ap-1 family have many homologous proteins in S. cerevisiae33 and Aspergillus spp. 30 31; however, its role in toxin biosynthesis and virulence is divergent 28. In A. parasiticus and A. ochraceus, deletion of apyapA and aoyap1 resulted in increases of aflatoxin and ochratoxin, respectively 30 32. In Fusarium graminearum, the Afgap1 mutant showed higher sensitivity to oxidative stress (H2O2) and higher level of trichothecene concentration associated with overexpression of TRI genes. However, the activation mechanism of toxin accumulation in response to oxidative stress was not observed 25. In contrast, in A. nidulans, deletion or overexpression of napA led to a decreased tolerance to oxidative stress and sterigmatocystin synthesis 42. In this study, the growth of CA14PTs was significantly inhibited following treatment with 20mmol/l H 2O 2, whereas growth of the knockout mutants was completely inhibited following treatment with only 5 mmol/l H 2O 2 due to the lack of a key transcription factor Afap1 related to the oxidative-stress response. The effect of deletion of afap1 is similar with napA and is contrary to aoyap1, apyapA and fgap1.
Hong et al. (2013) proposed that SrrA (SrrA recruits AP-1) and AtfB combined with the promoter regions of aflatoxin biosynthetic genes to help their induction by transcription factor AflR 17. Moreover, AtfB has been proved to combine with the promoter regions of aflatoxin biosynthetic genes including aflB (fas-1), aflD (nor-1), aflM (ver-1) and aflP (omtA), which carry CRE sites 17 34. In addition, the sensor kinases TcsA transmit oxidative stress signals through SrrA and/or SskA response regulators 23, and then cooperate with Ap-1 against oxidative stress in other Aspergillus spp. 16. Therefore, a similar pathway may have A. flavus since oxidative stress signals were transmitted through sensor kinases (ortholog of TcsA or Bos1 in yeast) to AtfB-SrrA-Afap1 homologous complex, and then induced aflatoxin biosynthesis ( Fig. 6). In this study, the expression of tcsA, bos1, srrA and aflR was up-regulated in the Aafap1 mutants compared to CA14PTs. aflB, encoding fatty acid synthase and being close to aflR in the aflatoxin gene cluster, was also up-regulated in the Aafap1 mutants. However, the expression of aflM and aflP, two downstream structural genes, were significantly down-regulated in the Aafap1 mutants, which resulted in a down-regulation of aflatoxin production. On the other hand, interestingly, the gene expression of aflS was down-regulated in the knockout mutants. Down-regulation of aflS led to the decreased production of AflS protein, which is beneficial for some potential suppressors to bind to AflR in place of AflS. Consequently, the transcription of the aflatoxin biosynthesis gene, which relies on AflS-AflR, would be reduced and aflatoxin biosynthesis would decrease. There were similar findings in some previous studies 20 40 43.
The present study revealed that oxidative stress inhibited the growth of toxigenic strains and was completely inhibited at 40mmol/l H 2O 2. However, the AFB 1 concentration was increased until 10mmol/l. According to the NCBI BLAST analysis, transcription factor Afap1 has the conserved protein domains of other AP-1 homologue proteins. Deletion of afap1 resulted in an increase in sensitivity to oxidative stress and a decrease in aflatoxin production in A. flavus. These results suggested that afap1 plays a key role in tolerance to oxidative stress and promoted aflatoxin production in A. flavus.