The pcz1 Gene, which Encodes a Zn(II)2Cys6 Protein, Is Involved in the Control of Growth, Conidiation, and Conidial Germination in the Filamentous Fungus Penicillium roqueforti

Proteins containing Zn(II)2Cys6 domains are exclusively found in fungi and yeasts. Genes encoding this class of proteins are broadly distributed in fungi, but few of them have been functionally characterized. In this work, we have characterized a gene from the filamentous fungus Penicillium roqueforti that encodes a Zn(II)2Cys6 protein, whose function to date remains unknown. We have named this gene pcz1. We showed that the expression of pcz1 is negatively regulated in a P. roqueforti strain containing a dominant active Gαi protein, suggesting that pcz1 encodes a downstream effector that is negatively controlled by Gαi. More interestingly, the silencing of pcz1 in P. roqueforti using RNAi-silencing technology resulted in decreased apical growth, the promotion of conidial germination (even in the absence of a carbon source), and the strong repression of conidiation, concomitant with the downregulation of the genes of the central conidiation pathway brlA, abaA and wetA. A model for the participation of pcz1 in these physiological processes in P. roqueforti is proposed.


Introduction
Zinc-binding proteins are one of the largest families of transcription regulators in eukaryotes, displaying structural and functional diversity. According to their zinc finger binding motifs, zinc-binding proteins are classified into three main groups: Cys 2 His 2 , Cys 4 , and Zn(II) 2 Cys 6 [1]. Interestingly, Zn(II) 2 Cys 6 proteins (hereafter C6) have been found only in fungi and yeasts, and they seem to be absent from bacteria, plants, and animals [2].
Although the C6 genes are abundant in fungal genomes, mainly from the phylum Ascomycota, comparatively few of them (approximately 30-40 genes) have been functionally characterized [2]. From these studies, mainly performed in Saccharomyces cerevisiae and model species from the genus Aspergillus, it has been deduced that C6 proteins are mainly involved in the regulation of the utilization of carbon and nitrogen compounds, the regulation of secondary metabolism, and the regulation of sexual and/or asexual development [2]. Known examples of C6 proteins participating in these processes are the S. cerevisiae Gal4 transcriptional activator, which is involved in galactose catabolism [3], the AflR protein controlling the expression of several genes involved in the production of the secondary metabolites aflatoxin and sterigmatocystin in Aspergilli [4], and the SfgA protein, a repressor that interacts with FluG, a protein necessary for the activation of conidiation in A. nidulans [5].
In contrast with yeast and Aspergillus, very few C6 proteins have been studied in Penicillium species. In the dimorphic fungus Penicillium marneffei (currently known as Talaromyces marneffei), disruption of the facB gene downregulates the expression of the gene encoding for isocitrate lyase, a key enzyme in the glyoxylate cycle [6]. In P. citrinum, disruption of the mlcR gene demonstrated its involvement in the biosynthesis of the secondary metabolite compactin [7]. Another C6 gene from P. citrinum (ariB) was disrupted, but its absence was not associated with any phenotypic effect [7]. To our knowledge, no other studies of functional characterization of any C6 protein in Penicillium species have been performed.
Penicillium roqueforti is a filamentous fungus that is important to the food industry because it is used in the ripening of blue-veined cheeses. However, despite its biotechnological importance, not much progress has been made in understanding the physiology of this fungus. Previously, we have made efforts to gain insight into the biology of P. roqueforti by studying the effect of a dominant active α subunit in subgroup I (Gαi) of a heterotrimeric G protein [8][9][10]. Heterotrimeric G proteins remain inactive when their three subunits (Gα and the βγ dimer) are together. In this inactive state, Gα keeps GDP (guanosine 5`-diphosphate) bound, but when a suitable stimulus is sensed, Gα exchanges GDP for GTP (guanosine 5`-triphosphate), resulting in its separation from the βγ dimer. Then, both the Gα and the βγ dimer become active, interacting with downstream effectors. Normally, activation ends when the GTPase activity in Gα hydrolyzes GTP to GDP, causing Gα and the βγ dimer to reassociate [11]. However, in fungi, the replacement of glycine with arginine at the appropriate position produces constitutive activation of Gα, thereby resulting in persistent and dominant active Gα signaling [12].
Transformants of P. roqueforti containing the dominant active Gαi protein described above showed several phenotypic alterations compared with the wild-type strain: a drastic reduction in conidiation, the ability to germinate their spores in the absence of a carbon source, and delayed apical growth [8][9][10]. Additionally, these transformants have increased concentrations of cAMP [9]. However, in addition to increases in this metabolite, other putative downstream effectors of Gαi have not been functionally described in P. roqueforti to date. We hypothesized that the effects of the Gαi protein on the phenotype of the fungus could be due (at least in part) to transcriptional changes in effector genes downstream of the Gαi protein. In this paper, we have characterized a novel gene encoding a putative C6 protein of unknown function, which we have named pcz1. We confirmed that the expression of pcz1 is downregulated in strains of P. roqueforti containing the dominant active Gαi subunit. More interestingly, a functional analysis of pcz1 suggests that it may be a positive regulator of conidiation, which is concomitant with the downregulation of the genes of the central conidiation pathway brlA, abaA and wetA. In addition, pcz1 may be a positive regulator of apical growth, but it may be a repressor of conidial germination.

Fungal strains and culture media
The wild type strain P. roqueforti CECT 2905 was kindly provided by Dr. Juan F. Martín (Inbiotec, León, Spain). P. roqueforti transformants pga5 and pga7, derived from strain CECT 2905 and containing the dominant active Gαi subunit, have been previously described [8].
Construction of plasmid pC6-RNAi for pcz1 silencing Plasmid pJL43-RNAi [14] was used to generate the pcz1 knockdown construct. This plasmid contains two promoters in opposite directions, which generate double-stranded RNA molecules (dsRNAs) from a small DNA fragment inserted in an NcoI site between these two promoters [14]. These dsRNAs are recognized and processed by the fungal RNA-silencing machinery, resulting in the specific degradation of target mRNAs [14]. Plasmid pJL43-RNAi has already been successfully used for the silencing of genes in P. roqueforti [15].
Transformation of P. roqueforti P. roqueforti transformants M9 and M11, with a silenced pcz1 gene, were obtained by introducing plasmid pC6-RNAi into strain CECT 2905 by protoplast transformation. Protoplast obtainment, transformation, and selection of transformants on Czapek-sorbitol medium containing phleomycin were carried out as described by Chávez et al. [16] except that 10 μg/mL phleomycin was used to select transformants. After transformation, conidia from selected colonies were subsequently transferred three times onto a suitable medium to stabilize the genotype and obtain homokaryotic strains. The same procedure was used to obtain a P. roqueforti transformant containing empty plasmid pJL43-RNAi, which was used as a control in the experiments.

DNA and RNA extractions, RT-PCR experiments and suppression subtractive hybridization (SSH) experiments
For DNA isolation, spores from P. roqueforti strains were inoculated into CM medium at 28°C for 24 hours at 200 r.p.m. in an orbital shaker. Mycelia were harvested by filtration and washed with 0.9% NaCl. DNA from each mycelium was isolated according to Bainbridge et al. [17].
For RNA isolation, mycelia growing under suitable conditions (see the respective Figure legend for details) were harvested and washed as above. Then, they were frozen in liquid nitrogen and ground in a mortar. The total RNA was extracted using the RNeasy Plant Minikit (Qiagen, Germany) according to the manufacturer's instructions, and treated with RNase-free DNase I (Roche, Germany). Total RNA was quantified in a MultiSkan GO quantification system using a μDrop plate (Thermo Scientific, Germany) according to the manufacturer's instructions. One μg of total RNA was used to synthesize cDNA using RevertAid Reverse Transcriptase (Thermo Scientific, Germany) according to the manufacturer's instructions.
For SSH experiments, spores from P. roqueforti CECT 2905 (wild-type strain) and P. roqueforti pga7 (containing the dominant active Gαi subunit) were inoculated into YES medium at 28°C for 10 days at 200 r.p.m in an orbital shaker. Mycelia were harvested and washed three times with PBS (137 mM NaCl, 2.7 mM KCl, 10 mM Na 2 HPO 4 , 1.8 mM KH 2 PO 4 , pH 7.4). Total RNA from each mycelium was isolated as above. PolyA+-RNA was purified from total RNA using the "Absolutely mRNA Purification" kit (Stratagene, USA), according to the manufacturer's specifications and used for SSH. SSH experiments were performed using the "PCR-Select cDNA subtraction" kit (Clontech, USA) and following the manufacturer's instructions exactly. cDNA from the wild-type strain was utilized as "tester" sample and cDNA from P. roqueforti pga7 was utilized as "driver" sample. The selectively amplified cDNA molecules were ligated into pGEM-T Easy and transformed into high efficiency competent E. coli cells. Analysis and selection of bacterial colonies containing plasmids suitable for sequencing were performed by PCR and dot blot assays exactly as described by Klagges et al. [20]. Selected plasmids were sequenced (both strands) by Macrogen (Seoul, Korea).

Quantitative reverse-transcriptase polymerase chain reaction (qRT-PCR) analysis
Gene expression was analyzed by qRT-PCR. Total RNA quantification and cDNA synthesis were performed as above. For details of specific primers used in each qRT-PCR experiment, see S1 Table. All primer sets exhibited suitable efficiency as required for the comparative Ct (ΔΔCt) method (S2 Table). Reactions were performed in 20 μl reaction volumes. Each reaction contained 10 μl of KAPA SYBR Fast qRT-PCR Master Mix 2x (Kapa Biosystems, USA), 0.4 μl of each primer (at a concentration of 10 μM each), 0.4 μl de 50x ROX High/Low, 6.8 μl of water, and 2 μl of cDNA. Quantification was carried out with a StepOne Real-Time PCR System (Applied Biosystems, USA) using the following conditions: 30 s at 95°C and 40 cycles of 3 s at 95°C and 30 s at 60°C. Appropriate negative controls were included. For each gene expression analysis, three replicates were performed. The data were analyzed according to the comparative Ct (ΔΔCt) method and were normalized to β-tubulin gene expression in each sample.

Measurement of fungal apical extension rates
Apical extension rates were determined as described by Ivey et al. [21]. Briefly, the diameter of the colony was measured daily in different culture media (see the respective Figure legend for details). In each case, the apical extension rate was obtained as a linear regression of colony diameter over time.

Measurement of conidial production kinetics
The production of conidia was measured according to García-Rico et al. [8]. Briefly, 100 μl of a conidial suspension (5 × 10 5 conidia/ml) was seeded in Petri dishes containing a suitable culture medium (see the respective Figure legend for details). Dishes were incubated at 28°C for 1, 3 or 5 days, and the conidia produced were collected by adding NT solution (0.9% NaCl, 0.05% Triton) and scratching the surface of the plate with an inverted Pasteur pipette. This procedure was repeated one time. Conidia obtained were counted in a Neubauer chamber. Values are expressed as conidia/mm 2 of surface.

Analysis of conidial germination
With slight modifications, the measurement of conidial germination was performed as previously described [10]. For each strain, three replicate flasks containing 2 x 10 5 conidia/ml of a suitable liquid media were incubated at 28°C for a variable amount of time (see the respective Figure legend for details about the medium and time). At regular intervals, samples of 10 μl were taken, observed in the microscope, and the number of germinated and non-germinated conidia was counted in 10 randomly-chosen fields (around 230 conidia were counted in each experiment). This procedure was repeated, in duplicate, for each flask (technical replicate). Conidia were considered germinated when the length of their germ tubes were the same size or longer than the diameter of the conidia. Data were plotted as the percentage of germination versus time.

Results
The pcz1 gene is downregulated in P. roqueforti carrying the dominant Gα subunit As was stated in Introduction, the effects of the Gαi protein on the phenotype of P. roqueforti could be due to transcriptional changes in effector genes downstream of the Gαi protein. Therefore, to begin unveiling genes from P. roqueforti that are regulated by Gαi, we performed SSH experiments (see Materials and methods for details). Among the sequences obtained from these experiments, a cDNA sequence that was putatively differentially expressed in wild-type P. roqueforti compared with transformants carrying the dominant Gαi protein (S1 Fig.) drew our attention. This sequence matched an unnamed ORF from P. roqueforti (Proq08g087160; [22]) that encodes a putative C6 protein (see below) of unknown function. A scan of the P. roqueforti genome indicates that only one copy of this ORF is present (data not shown). Because C6 proteins are exclusive to fungi and have been poorly studied, we decided to characterize this gene, for which the name pcz1 (for Penicillium C6 zinc domain protein 1) is proposed.
First, we confirmed that pcz1 was differentially expressed in P. roqueforti CECT 2905 with respect to the transformants pga5 and pga7, carrying the dominant Gαi protein. Fig. 1 shows the results of qRT-PCR experiments for this gene in these strains. As can be observed, and compared with the wild-type strain, pcz1 is downregulated in transformants carrying the dominant Gαi subunit. The transformants pga5 and pga7 showed approximately 12.6-and 5.2-fold decrease in pcz1 transcripts compared with P. roqueforti CECT 2905. These results suggest that the active Gαi subunit exerts a repressor effect on the expression of pcz1.
Analysis of Pcz1 deduced protein pcz1 has previously been annotated and encodes a 790 amino acid protein [22]. BlastP analysis of the deduced Pcz1 protein reveals the presence of orthologs with high levels of similarity in a wide range of species of filamentous fungi from the phylum Ascomycota, particularly the classes Eurotiomycetes, Sordariomycetes, Leotiomycetes and Dothideomycetes (data not shown).
The analysis of Pcz1 reveals that the putative Zn 2 (II)Cys 6 Fig.).

RNA-mediated gene-silencing of pcz1
To functionally characterize the role of pcz1, RNA-mediated gene silencing was employed. For this purpose, P. roqueforti CECT 2905 was transformed with plasmid pC6-RNAi (see Materials and methods). After transformation, 46 phleomycin-resistant transformants were obtained. Fifteen of them were randomly selected and subjected to a preliminary screening by RT-PCR (data not shown). Compared with the wild-type strain, two of these transformants (named M9 and M11) showed dramatic reductions in the levels of pcz1 transcript. To determine the amount of down-regulation in strains M9 and M11, qRT-PCR analyses were performed (Fig. 2). The transformants strains M9 and M11 showed approximately 39.4-and 31.4-fold decrease in pcz1 transcripts compared with wild-type P. roqueforti. The presence of the full silencing cassette in these transformants was also confirmed (Fig. 2). Taken together, these data confirm the successful knockdown of pcz1 in transformants M9 and M11. These transformants were used for further analysis.

pcz1 silenced transformants show abnormal phenotype
The phenotype of transformants M9 and M11 was observed in several culture media (Fig. 3). Whereas wild-type P. roqueforti or the strain containing empty pJL43-RNAi vector showed the typical deep green color associated with normal sporulation, transformants M9 and M11 appeared to be whiter (Fig. 3). This effect can be clearly observed in PDA, CYA, YES and Czapek media (Fig. 3). In addition, the colonies of these transformants were smaller than the wild-type strain (Fig. 3). Therefore, the silencing of pcz1 in the transformants altered the normal phenotype of P. roqueforti. It should be noted that on PW, PDA and YES media, the transformants M9 and M11 show slight differences in their phenotypes (Fig. 3). This could be due to different levels of decrease in pcz1 transcripts in the transformants (see above).  RNA-mediated silencing of pcz1 decreased apical extension in P. roqueforti Transformants M9 and M11 showed a reduced rate of growth in all the media tested (Fig. 4). For example, the rate of growth of transformants M9 and M11 in CYA ranged between 4.56 and 4.80 mm/day, while the wild-type strain showed a rate of approximately 6.96 mm/day (Fig. 4). A similar behavior was observed in all other media tested. Expressed as a percentage and depending on the medium, the transformants grew at a rate of between 65.5 and 83% of the wild-type strain. These results suggest that pcz1 plays a role in the vegetative growth of P. roqueforti, positively regulating this process.

RNA-mediated silencing of pcz1 strongly repressed conidiation in P. roqueforti
The silencing of pcz1 in P. roqueforti resulted in a drastic reduction in conidia formation (Fig. 5). For example, at 3 days of growth, transformants grown in Czapek (a minimal medium) produced between 5.4% and 9.4% of the number of conidia of the wild-type strain, whereas the same transformants grown on Power medium (a medium optimized for sporulation) produced up to 11.4% of the conidia produced by the wild-type strain. A similar behavior was observed at 5 days of growth. At this time, transformants grown on Czapek produced 5.8% and 11.7% of the number of conidia observed in the wild-type strain, whereas when grown on Power, they produced up to 19.1% of the conidia produced by the wild-type strain. These results indicate that the pcz1 allele has a strong and positive role in conidiation.
RNA-mediated silencing of pcz1 decreased the expression of brlA, abaA and wetA in P. roqueforti The current model of genetic control of conidiation (mainly based on studies performed in Aspergillus spp.) involves a network of regulators, which lead to a central conidiation pathway, composed of the genes brlA, abaA and wetA encoding transcription factors that control sporulation [23]. Because of the severe reduction of conidiation under pcz1 silencing, we quantified the expression of these genes by qRT-PCR (Fig. 6). Compared with wild-type P. roqueforti, transformants M9 and M11 showed a drastic reduction of brlA transcripts (6.4-and 6.3-fold decrease, respectively), abaA transcripts (5.5-and 6.5-fold decrease, respectively), and wetA transcripts (304-and 67-fold decrease, respectively), suggesting that Pcz1 regulates positively the expression of conidiation-specific genes.

RNA-mediated silencing of pcz1 promotes conidial germination in P. roqueforti
In addition to conidiation, pcz1 also demonstrated a role in the control of conidial germination in P. roqueforti. Fig. 7 shows the germination kinetics of the wild-type strain and the pcz1-silenced transformants. As expected, all strains followed a sigmoidal pattern of conidiation. However, in the silenced transformants, the germination process was earlier compared with the wild-type strain. Thus at 10 hours, whereas approximately 16.3% of wild-type conidia had germinated, 33.5 and 28.4% of conidia from strains M9 and M11 had germinated, respectively.   These differences were maintained throughout the exponential phase of germination, and at 12 hours, whereas conidial germination from strains M9 and M11 reached approximately 93.1%, the percentage of conidial germination of the wild-type strain was approximately 76.9%.

RNA-mediated silencing of pcz1 triggers conidial germination in the absence of a carbon source
Taking into account the differences observed in germination kinetics, and because conidial germination is triggered mainly by the presence of a carbon source [24,25], we tested whether the strains with attenuated pcz1 expression were able to germinate in a medium lacking a carbon source. Consistently, and for all the times assayed (12 to 48 hours), transformants M9 and M11 showed enhanced conidial germination compared with the wild-type strain. Approximately 12.7% of conidia from transformants were able to germinate after 48 h in minimal Czapek medium lacking a carbon source, whereas at the same time, 3.8% of conidia from the P. roqueforti wild type strain had germinated (Fig. 7). These results indicate that the attenuation of pcz1 allows germination in the absence of carbon source, suggesting that pcz1 is a negative regulator of conidial germination, probably intervening in a pathway that senses carbon sources.

Discussion
Penicillium roqueforti is used throughout the world in the production of ripened blue-veined cheeses, but unlike model fungal organisms (i.e., Aspergillus spp.) or other industrial filamentous fungi (i.e., P. chrysogenum, currently known as P. rubens), the functional characterization of genes from P. roqueforti using genetics strategies is still in its infancy. Thus, this work is one of the few examples of functional characterization of a gene in this fungus.
According to the data described in this work, RNA-mediated silencing of pcz1 strongly represses conidiation (Fig. 5) and decreases the expression of brlA, abaA and wetA (Fig. 6). Activation of brlA is a key step of conidiation. BrlA is a transcription factor that during the middle stage of conidiation activates abaA, an essential component for differentiation and function of phialides [23]. Also, BrlA activates wetA during the late phase of conidiation; wetA plays an important role for the synthesis of spore cell wall [23]. Accordingly, our results suggest that pcz1 stimulates sporulation through the positive regulation of conidiation-specific genes expression (Fig. 8). Further work is necessary to determine if this is the result of a direct transcriptional regulation of Pcz1 or of an indirect effect.
Upstream of the conidiation central pathway, there are a group of activators of BrlA (flbB-flbE), which are in turn negatively regulated by the repressor protein SfgA [5,23]. Interestingly, SfgA is a C6 protein, but conversely to Pcz1, SfgA is a repressor of conidiation [5]. In A. nidulans, overexpression of sfgA resulted in reduced levels of mRNA of brlA, suggesting that SfgA inhibits the expression of this gene [5]. Therefore, the opposite roles of SfgA and Pcz1 in conidiation can be explained by their opposite roles in the regulation of brlA expression. In addition to sfgA, three other C6 genes (oefC from A. nidulans and MoCOD1 and MoCOD2 from Magnaporthe oryzae) are involved in the control of conidiation [26,27]. However, their exact roles in the process have not yet been established. In Aspergillus, sfgA has also been proposed as a regulator of vegetative growth, and to date, it is the only C6 gene involved in the control of this process. SfgA negatively regulates FlbA, a regulator of G-protein signaling (RGS) that stimulates the intrinsic GTPase activity of a Gαi subunit. Thus, SfgA helps in the promotion of vegetative growth through Gαi signaling in Aspergillus [23,25]. Interestingly, the attenuation of pcz1 represses vegetative growth in P. roqueforti (Fig. 4), suggesting that Pcz1 is a positive regulator of this process. This behavior of pcz1  [8,9]. See the main text for details. agrees with the previous description of the role of G-protein signaling (particularly the Gαi subunit) in P. roqueforti and other fungi, such as P. chrysogenum, Fusarium sporotrichioides and F. fujikuroi [8,28,29]. In these organisms, and contrary to what has been described in Aspergillus, Gαi signaling has a negative effect on the apical growth rate. Because the Gαi subunit downregulates pcz1 in P. roqueforti (Fig. 1), we suggest a model relating both regulators and their effects on vegetative growth in P. roqueforti (Fig. 8). Regarding the model, we observed that depending on the culture medium used, the attenuation of the expression of pcz1 results in approximately 65-83% apical extension rate compared with wild-type P. roqueforti (Fig. 4). The effect produced by the transformation of P. roqueforti with the dominant active Gαi protein was previously estimated to be approximately 40-70% of the apical extension rate of the wild-type strain [8]. Therefore, pcz1 may account for all or most of the control of apical extension rate produced by the dominant active Gαi protein (Fig. 8).
To the best of our knowledge, only one C6 gene has been functionally related to conidial germination in fungi to date. In M. oryzae, the absence of the MoCOD1 gene severely reduced the germination rate of its conidia [27], indicating that this C6 gene may be a positive regulator of conidial germination. In our case, we observed that the attenuation of pcz1 increased the germination rate of conidia in P. roqueforti (Fig. 7), suggesting that contrary to MoCOD1, pcz1 may be a negative regulator of conidial germination. Importantly, this effect is maintained even in the absence of a carbon source (Fig. 7). Initiation and completion of germination require the sensing of external signals, particularly a carbon source, which is usually the signal for conidial germination [25]. The presence of organic molecules is sensed by a suitable receptor and transmitted through the Gα-cAMP signaling pathway to protein kinase A (PkaA), which phosphorylates target proteins, thus beginning the conidial germination process [25]. Interestingly in a previous report, it was demonstrated that the Gαi protein has a positive effect on conidial germination in the absence of a carbon source in P. roqueforti [10], and now we have shown that Gαi downregulates pcz1 (Fig. 1). Taken together, these results suggest a model where pcz1 may be participating in the Gαi-signaling pathway that senses carbon sources and triggers conidial germination (Fig. 8). Further work is necessary to validate this suggestion.
Using the data obtained in this work and previously published data [8,9] we propose an integrated working model of the role of Pcz1 in developmental processes in P. roqueforti (Fig. 8), whose roles in vegetative growth and conidial germination were discussed above. Another interesting aspect of this model is related to conidiation. In P. roqueforti, Gαi controls conidiation through two pathways: cAMP-dependent and cAMP-independent pathways [9]. Specifically in this fungus, it has been demonstrated that an artificial increase in intracellular cAMP levels had a minor effect on conidiation (14-20% reduction of conidiation) compared with the effect produced by transformation with the dominant active Gαi protein (>99% reduction of conidiation; [9]). Therefore, in P. roqueforti, the repression of conidiation by Gαi is mainly (approximately 80%) through the cAMP-independent pathway [9]. Interestingly, and depending on the transformant and medium used, the attenuation of pcz1 resulted in 80-94% less conidiation than in the parental strain (Fig. 5), which matches well with the magnitude of the expected impact of the cAMP-independent pathway. Therefore, and although further confirmatory experiments are necessary, it seems very probable that pcz1 may be part of the cAMP-independent pathway that controls conidiation (Fig. 8). The same dual control of conidiation by Gαi through cAMP-dependent and cAMP-independent pathways has been observed in A. nidulans and P. chrysogenum [9,30]. In this context, it will be interesting to analyze the role of the orthologues of pcz1 in these fungi in the future.
Supporting Information S1 Fig. Dot blot assay showing the differential expression of pcz1 in a subtractive library. The dot blot assay was carried out as described in Klagges et al. [20]. Briefly, the same nylon membrane containing purified plasmids from selected clones was hybridized with subtracted cDNA from wild-type P. roqueforti (A) or subtracted cDNA from P. roqueforti pga7 (B). Those clones harboring putative differentially expressed cDNAs in the wild-type strain should hybridize only to A. Numbers in red are nomenclature of the clones. Clone 321 was differentially expressed in the wild-type strain (highlighted in the red box) and contains pcz1 cDNA. C+: Positive control for hybridization. (TIF)  Table. Correlation coefficient (R 2 ), slope and efficiency of calibration curves obtained for the genes analyzed by qRT-PCR. (DOCX)