bZIP transcription factors PcYap1 and PcRsmA link oxidative stress response to secondary metabolism and development in Penicillium chrysogenum

Abstract

Background: Reactive oxygen species (ROS) trigger different morphogenic processes in filamentous fungi and have been shown to play a role in the regulation of the biosynthesis of some secondary metabolites. Some bZIP transcription factors, such as Yap1, AtfA and AtfB, mediate resistance to oxidative stress and have a role in secondary metabolism regulation. In this work we aimed to get insight into the molecular basis of this regulation in the industrially important fungus Penicillium chrysogenum through the characterization of the role played by two effectors that mediate the oxidative stress response in development and secondary metabolism.

Results: In P. chrysogenum, penicillin biosynthesis and conidiation are stimulated by the addition of H₂O₂ to the culture medium, and this effect is mediated by the bZIP transcription factors PcYap1 and PcRsmA. Silencing of expression of both proteins by RNAi resulted in similar phenotypes, characterized by increased levels of ROS in the cell, reduced conidiation, higher sensitivity of conidia to H₂O₂ and a decrease in penicillin production. Both PcYap1 and PcRsmA are able to sense H₂O₂-generated ROS in vitro and change its conformation in response to this stimulus. PcYap1 and PcRsmA positively regulate the expression of brlA, the first gene of the conidiation central regulatory pathway. PcYap1 binds in vitro to a previously identified regulatory sequence in the promoter of the penicillin gene pcbAB: TTAGTAA, and to a TTACTAA sequence in the promoter of the brlA gene, whereas PcRsmA binds to the sequences TGAGACA and TTACGTAA (CRE motif) in the promoters of the pcbAB and penDE genes, respectively.

Conclusions: bZIP transcription factors PcYap1 and PcRsmA respond to the presence of H₂O₂-generated ROS and regulate oxidative stress response in the cell. Both proteins mediate ROS regulation of penicillin biosynthesis and conidiation by binding to specific regulatory elements in the promoters of key genes. PcYap1 is identified as the previously proposed transcription factor PTA1 (Penicillin Transcriptional Activator 1), which binds to the regulatory sequence TTAGTAA in the pcbAB gene promoter. This is the first report of a Yap1 protein directly regulating transcription of a secondary metabolism gene. A model describing the regulatory network mediated by PcYap1 and PcRsmA is proposed.

Keywords: DNA-binding proteins; Fungal morphogenesis; Oxidative stress defense; Reactive oxygen species; RsmA; Secondary metabolism; Transcriptional regulation; Yap1; bZIP transcription factor.

Fig. 1

Penicillin production of strain P. chrysogenum Wis54-1255 in cultures submitted to oxidative stress. Different concentrations of H₂O₂ were added to the MCFP medium at the beginning of the cultivation time. Cultures were performed in triplicate, error bars represent standard deviation

Fig. 2

EMSA of PcYap1 binding activity in the pcbAB-pcbC intergenic region. A Positive binding reaction of PcYap1 with the probe PTA1-WT containing the regulatory sequence TTAGTAA, and lack of binding reaction with the probe PTA1-M1 containing the mutant sequence CTAGTCA. B Binding reactions of PcYap1 and PcRsmA with the probes PTA1-WT and PTA1-M1. C Specificity test for the binding of PcYap1 to the TTAGTAA sequence. PTA-WT* is the labelled probe; PTA-WT and PTA-M1 are cold (unlabelled) probes, which were added in excess amounts of 50×, 100×, 150× and 200× as indicated at the top of the gel image. D Screening of the entire pcbAB-pcbC intergenic region to locate target sites of PcYap1; lanes 1 through 6 correspond to binding reactions between probes 1 through 6 with PcYap1, lane F5 is the free probe 5. E Location in the pcbAB gene promoter of the six probes used in D; below is indicated the position and the sequence of the upper strand of probes PTA1-WT and PTA1-M1

Fig. 3

Penicillin production in strains with knocked down expression of Pc-yap1 and Pc-rsmA (A) and strains overexpressing the respective genes (B); strain C1 contains the empty pGpdPki-RNAi vector used for gene silencing and strain C2 (pyrG) contains the empty pBKSpyrG vector used for gene overexpression. To test the effect of H₂O₂-generated oxidative stress, 100 mM H₂O₂ was added at the beginning of the cultivation time. In the panel at the right in A the extra bar (violet color) corresponds to the parental strain Wis54-1255 grown on medium without added H₂O₂. Bar sizes are the result of three biological replicas, error bars correspond to standard deviation. See “Materials and methods” section for details

Fig. 4

EMSA of the PcRsmA binding activity in the pcbAB-pcbC intergenic region. A The pcbAB-pcbC intergenic region is shown at the top, with the position of the six probes used for screening the DNA binding activity of PcYap1 (Fig. 2D) and PcRsmA. The position of the DNA binding sites of PcYap1 and PcRsmA is shown with a bar of red and violet color, respectively. The small lines above or below the bars indicate the position of the binding sequence in the upper or lower strand, respectively. Below it is shown the relative position of seven probes located within the DNA region corresponding to probe 5, with the sequence of the upper strand from the smaller probes at their right. B Screening of the entire pcbAB-pcbC intergenic region to locate target sites of PcRsmA; lanes 1 through 6 correspond to binding reactions between probes 1 through 6 with PcRsmA. At the right are shown confirmation binding reactions of PcYap1 and PcRsmA with probe 5, and a positive control reaction of PcRsmA with probe AflR-RsmA of 25 bp (Additional file 12), which belongs to a region in the promoter of the A. nidulans sterigmatocystin cluster gene aflR containing a TGACACA sequence proven to be bound by A. nidulans RsmA [44]. C Positive binding reaction of PcRsmA with the probe upPta1 and negative reaction with the probe dwPta1. D Location of the PcRsmA binding sequence; the probes containing the sequence TGAGACA (RsmA-2 and RsmA-2C) are bound by PcRsmA, whereas the probes lacking this sequence (RsmA-2A and RsmA-2B) are not. E Specificity test for the binding of PcRsmA to the TGAGACA sequence. We used the probe RsmA-2C-M1, containing the mutated sequence TGCGATA, to prove the specificity. RsmA-2C* and RsmA-2C-M1* are labelled probes; RsmA-2C and RsmA-2C-M1 are cold (unlabelled) probes, which were added in excess amounts of 50x, 100x, 150× and 200× as indicated at the top of the gel image

Fig. 5

EMSA of the PcAtf21 binding activity in the pcbAB-pcbC intergenic region and of the three bZIP transcription factors in a region of the penDE gene promoter. A The six probes (lanes 1 to 6) of the pcbAB-pcbC intergenic region (Fig. 4A) were tested for binding to the PcAtf21 transcription factor; a binding reaction with probe 5 and PcYap1 was used as positive control. B EMSA of the PcYap1, PcRsmA, and PcAtf21 binding activities with probe penDE-CRE123, whose sequence corresponds to a central region in the penDE gene promoter; probe norR4 was used as positive control for PcAtf21 binding (see text). C Sequences of the probes used for EMSA. Probe penDE-CRE123 contains a CRE site (highlighted in red) and two CRE-like sites upstream (highlighted in pink); the nucleotide changes at the CRE site in the probe penDE-CRE123-M1 are highlighted in blue

Fig. 6

Effect of PcYap1 and PcRsmA on oxidative stress response. Effect of Pc-yap1 and Pc-rsmA silencing (A) and overexpression (B) on the amount of naturally occurring and H₂O₂-generated ROS in mycelium from submerged cultures. Strain C1 contains the empty pGpdPki-RNAi vector used for gene silencing and strain C2 (pyrG) contains the empty pBKSpyrG vector used for gene overexpression. Results are expressed in relative fluorescence units (RFU) per gram of biomass (dry weight) (see “Materials and methods” section for details). C Effect of Pc-yap1 and Pc-rsmA silencing on the viability of conidia submitted to oxidative stress with different concentrations of H₂O₂. Results are expressed as percentage of conidia able to form colonies on plates with respect to the initial number counted under the microscope (see “Materials and methods” section for details). In all cases bar sizes are the result of three biological replicas, error bars correspond to standard deviation

Fig. 7

Mobility changes in SDS-PAGE of purified PcYap1 and PcRsmA incubated with H₂O₂ prior to loading on the gel. A An H₂O₂ concentration gradient was used to determine the amount of H₂O₂ producing changes in mobility (upper panels); incubation time was 15 min. In the lower panels, proteins were incubated with a concentration of 400 µM for different times; the three lanes at the right of each panel were loaded with a control (no incubation with H₂O₂) and the 400 µM H₂O₂-incubated protein simultaneously, with the purpose to showcase the different mobility of the bands corresponding to H₂O₂-treated and untreated proteins. Gels had an acrylamide concentration of 10%, the amount of protein used in each reaction was 25 µg. In all cases non-reducing conditions were used (see Materials and Methods and the text for details). B Comparison between reducing and non-reducing conditions in the mobility of the proteins after 0 to 30 min of incubation with 400 µM H₂O₂. Gels had an acrylamide concentration of 8%, the amount of protein used in each reaction was 15 µg

Fig. 8

Regulation of the expression of the Pc-yap1 gene by PcRsmA. A Semiquantitative RT-PCR using as template RNA extracted from mycelium grown for 36 h in submerged cultures (medium MPPY); primers were the same as indicated in the legend to Fig. 3. The left panel shows the intensity of the bands in an agarose gel loaded with the products of the RT-PCR reactions, and the right panel the densitometry analysis of the bands. The results were normalized with the bands of the constitutively expressed act gene, and the parental strain Wis54-1255 (lane W) was used as reference with a value for integrated optical density (IOD) of 100. R24: strain si-PcRsmA-24; R25: strain si-PcRsmA-25. B Northern blot with RNA extracted from mycelium grown on solid Power medium for 120 h; the probes used were a 458 bp fragment from the Pc-yap1 gene amplified by PCR with primers siYAP1-F and -R, and a 508 bp fragment from the act gene amplified by PCR with primers N-actA-1F and R (Additional file 12). At the right, the densitometry analysis was performed as described above, adjusting the IOD of strain Wis54-1255 to 1. (See “Materials and methods” section for details)

Fig. 9

Effect of H₂O₂-induced oxidative stress and of PcYap1 and PcRsmA attenuation on conidia production. A Conidia production in strain P. chrysogenum Wis54-1255 grown on Power medium with different concentrations of H₂O₂. The H₂O₂ was added at the moment of pouring the media onto the plates, next, 50 µL of a suspension with 1 × 10⁶ conidia/mL was spread onto the plates, which were incubated for 5 days at 28 °C. B Colonies of control strains, Pc-yap1 and Pc-rsmA knockdown strains grown on Power medium for 72 and 120 h at 28 °C. Four microliters of a solution with 1 × 10⁶ conidia/mL were inoculated at the center of Petri dishes. C Conidia production in strains with knocked down expression of Pc-yap1 or Pc-rsmA. Strain C1 contains the empty pGpdPki-RNAi vector used for gene silencing. To test the effect of H₂O₂-generated oxidative stress 100 mM H₂O₂ was added at the moment of pouring the media onto the plates (right panel). In A and C bar sizes represent the result of three biological replicas, error bars correspond to standard deviation. See “Materials and methods” section for details

Fig. 10

Regulation of the expression of the brlA gene by PcYap1 and PcRsmA. A Northern blot with RNA extracted from mycelium grown on solid Power medium for 120 h; the probes used were a 573 bp fragment from the brlA gene amplified by PCR with primers N-brlA-1F and R, and a 508 bp fragment from the act gene amplified by PCR with primers N-actA-1F and R (Additional file 12). The densitometry analysis was performed as described in the legend to Fig. 8. Strain Q204L was used as a control for brlA gene transcriptional repression (see text). B EMSA of the PcYap1 and PcRsmA binding activities with the probe AP1-brlA, whose sequence corresponds to a region located between − 58 and − 90 bp upstream the ATG start codon of the brlA gene promoter, and which contains an AP1 site (highlighted in red); this site is mutated in the AP1-brlA-Mut probe

Fig. 11

Proposed model for the regulation of penicillin biosynthesis and conidiation in response to oxidative stress by PcYap1 and PcRsmA. The model has been elaborated after the results obtained in this work complemented with previously published results with sufficient evidence to be applicable to this outline. In presence of naturally occurring or H₂O₂-generated ROS, PcYap1 undergoes conformational change, by direct sensing or mediated by peroxiredoxins (Prxs), and locates to the nucleus (see text). In these conditions, PcRsmA senses ROS, probably undergoes conformational change and locates to the nucleus by an unknown mechanism. PcYap1 binds a previously described regulatory sequence in the pcbAB gene promoter (TTAGTAA), while PcRsmA binds to a site 68 bp upstream from the PcYap1-binding site and to a second site in the penDE gene promoter. The two transcription factors positively regulate penicillin biosynthesis. PcRsmA positively regulates transcription of the Pc-yap1 gene. PcYap1 binds to an AP-1 site in the conidiation central regulatory pathway gene brlA and induces its transcription. PcRsmA also stimulates transcription of brlA, through direct (unproven) binding to the promoter or via regulation of Pc-yap1 transcription, or both. Upregulation of the brlA gene triggers conidiation. Both PcYap1 and PcRsmA participate in oxidative stress defense, reducing the amount of ROS in the cell; PcYap1 by directly controlling transcription of several oxidative stress response genes, and PcRsmA by putatively controlling expression of oxidative stress response genes or through regulation of Pc-yap1 transcription, or both. See text for additional details. Created with BioRender.com