Targeting Fungal Genes by Diced siRNAs: A Rapid Tool to Decipher Gene Function in Aspergillus nidulans

Background Gene silencing triggered by chemically synthesized small interfering RNAs (siRNAs) has become a powerful tool for deciphering gene function in many eukaryotes. However, prediction and validation of a single siRNA duplex specific to a target gene is often ineffective. RNA interference (RNAi) with synthetic siRNA suffers from lower silencing efficacy, off-target effects and is cost-intensive, especially for functional genomic studies. With the explosion of fungal genomic information, there is an increasing need to analyze gene function in a rapid manner. Therefore, studies were performed in order to investigate the efficacy of gene silencing induced by RNase III-diced-siRNAs (d-siRNA) in model filamentous fungus, Aspergillus nidulans. Methodology/Principal Findings Stable expression of heterologous reporter gene in A. nidulans eases the examination of a new RNAi-induction route. Hence, we have optimized Agrobacterium tumefaciens-mediated transformation (AMT) of A. nidulans for stable expression of sGFP gene. This study demonstrates that the reporter GFP gene stably introduced into A. nidulans can be effectively silenced by treatment of GFP-d-siRNAs. We have shown the down-regulation of two endogenous genes, AnrasA and AnrasB of A. nidulans by d-siRNAs. We have also elucidated the function of an uncharacterized Ras homolog, rasB gene, which was found to be involved in hyphal growth and development. Further, silencing potency of d-siRNA was higher as compared to synthetic siRNA duplex, targeting AnrasA. Silencing was shown to be sequence-specific, since expression profiles of other closely related Ras family genes in d-siRNA treated AnrasA and AnrasB silenced lines exhibited no change in gene expression. Conclusions/Significance We have developed and applied a fast, specific and efficient gene silencing approach for elucidating gene function in A. nidulans using d-siRNAs. We have also optimized an efficient AMT in A. nidulans, which is useful for stable integration of transgenes.


Introduction
RNA interference (RNAi) is a sequence-specific post-transcriptional gene silencing mechanism mediated by small interfering RNAs (siRNAs) which are the products of a double-stranded RNA (dsRNA) by the action of RNase III type enzyme Dicer. These siRNAs will then be incorporated into the RNA-induced silencing complex (RISC) and reduces cognate mRNA levels [1,2]. In certain organisms, including fungi, the silencing phenomenon requires the action of RNA-dependent RNA polymerases (RdRPs), either to generate dsRNA from single-stranded RNA (ssRNA) or to amplify siRNA signals [3]. The specific silencing effect of RNAi on target genes makes it a powerful tool in functional genomic studies, drug target discovery and disease treatment [4,5,6,7,8]. The use of RNAi in mammalian cells for specific knockdown of target genes was limited before the discovery of synthetic siRNA-mediated gene silencing, as the introduction of long dsRNAs into the mammalian cells frequently triggers a lethal interferon response [9]. However, it has been found that siRNAs can specifically silence target genes without triggering interferon response [10], hence it becomes a promising approach for achieving RNAi in mammalian cells. The RNAi phenomenon in fungi ''Quelling'' was first demonstrated in ascomycete Neurospora crassa [11]. Targeting fungal genes by RNAi has been shown to be suitable for a multitude of fungal species [12].
The filamentous fungus A. nidulans is an important experimental model for studying eukaryotic cell biology [13]. It also serves as an important model organism for the genus Aspergillus, which includes industrially important fermentation organisms as well as serious human fungal pathogens. In addition, A. nidulans is an opportunistic human pathogenic fungus, which causes fatal invasive aspergillosis in patients suffering from chronic granulomatous disease (CGD) [14]. Moreover, A. nidulans has been a fungal model for RNAi using hairpin RNAi constructs [15,16] and chemically synthesized siRNAs as silencing inducers [17].
The technical advances in sequencing platforms have rapidly brought in the discovery of a vast number of fungal genes without any predicted functions. Sequencing projects of most of the pathogenic and industrially important fungal species have now been completed (http://www.broad.mit.edu/annotation/fungi/ fgi/). The next major challenge in understanding fungal biology is to translate this surplus genome sequence information into biological function information. Traditionally, the analysis of gene function requires the construction of mutant strains by disrupting or deleting the gene of interest. However, gene targeting in filamentous fungi is hampered by low frequencies of homologous recombination (HR) [18]. Recently, hairpin RNAi vectors and other plasmid-based silencing vectors have been exploited for studying gene function in multitude of filamentous fungi [19]. The current techniques available for construction of either conventional deletion cassettes or plasmid-based RNAi vectors are tedious and time consuming. Moreover, most of the fungi are recalcitrant for transformation.
Other than the plasmid-based RNAi vectors, the heterologous GFP reporter and endogenous genes coding for hydrophobins and a peroxiredoxin were silenced in Moniliophthora perniciosa transfected with in vitro transcribed respective dsRNAs. However, this method also requires protoplast preparations for dsRNA transfection [20]. Previously, it has been shown that the addition of chemically synthesized siRNA, targeting a key enzyme in polyamine biosynthetic pathway, to the culture medium, can result in specific suppression of the corresponding target gene with phenotypic consequences in A. nidulans. This method of small RNA uptake by fungal cells obviates the preparation of protoplasts for small RNA transfection [17].
Although chemically synthesized siRNAs are widely used in RNAi experiments, the in silico prediction and validation of siRNAs for a large number of RNAi targets is time consuming and laborious. Several guidelines and computational tools for designing effective siRNAs are available, but experiments reveal that a majority of the synthetic siRNAs designed against a target gene are not very satisfactory in terms of silencing potency and off-target effects [21,22]. Many synthetic siRNAs do not work at low concentrations and are sometimes used at 100 nM or even higher concentrations [21,23]. Synthetic siRNA that transfects at higher concentrations might aggravate the problem of off-target effects [24,25,26,27]. However, siRNAs can also be made from in vitro transcribed dsRNAs by RNase III family enzymes [28,29]. In this case, the resulting short RNA fragments contain several siRNAs against one target gene. A pool of d-siRNAs can sometimes be more effective and have fewer off-target effects than any single sequence synthetic siRNA [30,31]. However, thus far, the possibility of d-siRNA mediated targeting of fungal genes has not been investigated.
To determine whether the d-siRNA induced RNAi route is functional in A. nidulans and could be efficiently used in targeting specific genes, the fungal mycelia were treated with d-siRNAs corresponding to a heterologous GFP reporter and the endogenous target genes (AnrasA and AnrasB). The reduction of GFP fluorescence and mRNA abundance of the target genes were then assessed. The results showed that this approach is simple and efficient in specifically silencing the target genes in A. nidulans. The silencing potency of d-siRNA was found to be greater than the equal concentration of single sequence synthetic siRNA duplex targeting the corresponding gene. In addition, we have optimized the efficiency of Agrobacterium tumefaciens-mediated transformation of A. nidulans for stable integration of transgenes.

Construction of pCAMBIA1300-sGFP over-expression vector
The GFP over-expression vector was prepared by cloning an improved version of GFP gene, sGFP-TYG [32] in pCAM-BIA1300 binary vector under the control of A. nidulans TrpC promoter. In sGFP-TYG, a Ser 65 to Thr point mutation was introduced in the chromophore domain which resulted in brighter fluorescence with a single excitation and emission peak [33]. pCAMBIA1300-sGFP ( Figure 1) contains hygromycin B phosphotransferase gene (HPT) fused to CaMV35S promoter. To the best of our knowledge, CaMV35S viral promoter is being used for the first time in A. nidulans to drive the marker gene expression. Although, CaMV35S promoter is widely used in multitude of plant species for constitutive expression of transgenes, we observed minimal resistance to hygromycin B (100 mg/ml concentration) antibiotic in fungal transformants (fungal transformants resist upto 200 mg/ml hygromycin B concentration when HPT gene is fused to fungal-specific promoter). We found that 100 mg/ml hygromycin B is the optimal concentration for selection of fungal transformants (data not shown). We have also compared the HPT gene expression under the control of CaMV35S promoter with non-promoter expression in A. nidulans. We found that HPT gene expression without CaMV35S promoter was not sufficient to generate hygromycin B resistance (when selection medium was amended with 100 mg/ml hygromycin B) for A. nidulans transformants to undergo selection process.

Agrobacterium tumefaciens-mediated transformation (AMT) of A. nidulans
To evaluate the possibility of d-siRNA mediated gene silencing in A. nidulans, we used the reporter gene GFP as a target for silencing. Therefore, we have generated GFP over-expressing A. nidulans transformants through AMT. Since AMT is not reported in A. nidulans, we have optimized the conditions for efficient transformation based on previous protocols reported in closely related filamentous fungus A. fumigatus [34]. The GFP overexpression vector pCAMBIA1300-sGFP was mobilized into A. tumefaciens EHA105 strain and the presence of vector was confirmed by colony PCR using HPT marker gene and GFP specific primers (data not shown). Fungal conidia and bacteria (1:10) were co-cultivated for 48 h at 24uC on nylon membrane placed on induction medium amended with 100 mM acetosyringone. The nylon membranes were then transferred to selection medium containing 100 mg/l hygromycin B and incubated for another 3 days. Clear isolated fungal colonies appeared on nylon membranes after 3-4 days of incubation (Fig. S1A). Putative fungal transformants were sub-cultured for further growth (Fig. S1B) on selection medium and subjected to molecular analysis. We also transformed A. nidulans with pCAMBIA1300 vector containing HPT marker gene (not fused to any promoter) through AMT. However, we observed that no fungal transformants were appeared on nylon membranes placed on selection medium

Molecular analysis of fungal transformants and detection of GFP fluorescence
To evaluate the presence and expression of transgenes in putative fungal transformants, four randomly-selected hygromycin B resistant fungal colonies were subjected to PCR and semiquantitative RT-PCR analyses. PCR amplification with HPT and GFP specific primers (Table S1) was carried out on the total genomic DNA template extracted from putative fungal transformants which yielded the amplicons of expected sizes ( Fig. 2A and  2B). Further, confirmation of transgene integration was demonstrated by Southern hybridization in several A. nidulans transformants (Fig. 2C). The transgene expression in putative fungal transformants was confirmed by RT-PCR analysis using GFP specific primers (Fig. 2D). Fluorescence microscopy was carried out in order to detect green fluorescence in fungal transformants and one of the strong GFP over-expressing fungal transformant (AnGFP 101) was chosen for further experiments. Green fluorescence was detected in various developmental stages of AnGFP 101 A. nidulans transformant, such as in spores, mycelia and conidiophores (Fig. 3).

In silico prediction of siRNA candidates and preparation of diced siRNAs
In order to achieve the efficient gene silencing, target gene sequences were analyzed for the presence of candidate siRNAs using siRNA prediction software [35]. Several candidate siRNAs were identified in coding sequences of GFP reporter, AnrasA and AnrasB genes. To further scrutinize the silencing potency of probable siRNA candidates of all the target genes, in silico siRNA scoring was performed according to the parameters suggested by Reynolds et al. [36]. In silico siRNA scoring revealed that some of the potent siRNA candidates in all of the three target genes possess the score above 7 which is hypothetically sufficient for effective gene knockdown (Table S2A, S2B and S2C).
All the target genes were PCR amplified with gene-specific primers and cloned in pGEM-T easy TA cloning vector and confirmed by automated DNA sequencing. PCR amplification strategy was utilized for the generation of templates for dsRNA preparation for all the three target genes (Fig. 4A). T7 promoter sequence was added to both the forward and reverse gene specific primers and target genes were re-amplified from appropriate pGEM-T plasmids (Fig. 4C). In vitro transcription of all the three target genes were carried out using T7 RNA polymerase which recognizes the T7 promoter present at both the ends of target gene sequences and transcribes both the strands (partial dsRNAs forms due to complimentary base pairing). The intact dsRNAs were then produced by annealing the sense and antisense RNAs. Whereas for the preparation of unrelated dsRNA, Meloidogyne incognita acetylcholinesterase (MiAchE) gene cloned in pGEM-T easy plasmid (available in the lab) was used. MiAchE was PCR amplified using M13 primers from pGEM-T-MiAchE plasmid; this amplification adds the T7 promoter to one end and SP6 promoter to the other end of the MiAchE template (Fig. 4B). In vitro transcription of MiAchE was carried out using T7 RNA polymerase and SP6 RNA polymerase in separate reactions which transcribes sense and antisense RNAs respectively. MiAchE dsRNA was then produced by annealing the sense and antisense RNAs.
All the dsRNAs were treated with RNase-free DNase I in order to digest the DNA template in transcription reactions. The integrity of all the dsRNAs was confirmed through electrophoresis in 1.2 % agarose gel (Fig. 4D) and quantified by spectrophotometer. All the dsRNAs were digested with RNase III in separate reactions to prepare diced siRNA pools. The integrity was checked on 20% native PAGE (Fig. 4E) and quantified by spectrophotometer.   Treatment of fungal mycelia with GFP-d-siRNA resulted in reduction of GFP fluorescence and GFP mRNA abundance To demonstrate the possibility of d-siRNA-mediated gene silencing, we first targeted the heterologous GFP reporter gene in A. nidulans mycelia. We found that 25 nM final concentration of d-siRNA treatment was optimum for achieving efficient silencing through the direct siRNA uptake transfection method (data not shown) and the same concentration was utilized throughout the study. In order to observe the GFP reduction in GFP-d-siRNA treated samples, fluorescent microscopy analysis was performed. The GFP-d-siRNA treated mycelia showed strong reduction in GFP fluorescence whereas untreated mycelia showed the normal levels of fluorescence (Fig. 5A).
To ensure that the silencing occurred at the transcript level, the expression of GFP RNA was measured by quantitative RT-PCR (qRT-PCR) in GFP-d-siRNA treated, unrelated MiAchE-d-siRNA treated and untreated mycelia. The GFP mRNA abundance was found to be reduced by 89.60% in GFP-d-siRNA treated mycelia when compared to unrelated MiAchE-d-siRNA treated and untreated control mycelia (Fig. 5B).

d-siRNA-mediated silencing of the endogenous AnrasA and AnrasB genes hampered A. nidulans growth and development
To assess the possibility of silencing the endogenous genes by d-siRNAs in A. nidulans, we have chosen the Ras protein coding AnrasA and AnrasB genes, which are involved in conidial germination and asexual development, as targets. AnrasA deletion is lethal to A. nidulans, so it is impractical to isolate deletion strain. However, it has been shown that mutations in AnrasA gene cause aberrations in germination and asexual development in A. nidulans [37]. The regulated expression of rasA also plays a pivotal role in conidial germination, which is required to initiate both infection and asexual development in the opportunistic pathogen A. fumigatus [38]. In order to study crucial gene (where generation of deletion mutant is not possible) function in filamentous fungi, siRNA/d-siRNA mediated knockdown is a suitable method since it only transiently suppresses the target gene activity. The treatment of AnrasA-d-siRNA to the germinating spores resulted in reduction in hyphal elongation and mycelial formation in broth culture (Fig. 6A). We also observed significant reduction in radial growth in AnrasA-d-siRNA treated culture 48 h post-inoculation on ACM agar medium compared to controls (Fig. 6B). In addition, total hyphal mass was also significantly reduced by 43.80% in AnrasA-d-siRNA silenced lines as compared to controls after 48 h post-inoculation in ACM broth (Fig. 6C).
Another fungal specific Ras family homolog rasB plays a pivotal role in germination, growth rate, regulated hyphal morphology and virulence in A. fumigatus [39]. A. nidulans complete genome sequence and annotation revealed the presence of rasB homolog in the genome, and it was predicted to have a 723-base pair coding sequence. To date, there is no report on the characterization of rasB in A. nidulans. Therefore, we have chosen this rasB homolog as the second endogenous target for d-siRNA-mediated silencing. The knockdown of AnrasB resulted in abnormal hyphal morphol- ogy characterized by irregular apical branching. The AnrasB-d-siRNA treatment also resulted in malformed conidiophores (Fig. 7A). In addition, we have analyzed the radial growth and total biomass of the isogenic AnrasB-d-siRNA treated cultures grown for 48 h at 37uC. The radial growth assay performed on ACM solid agar medium revealed that AnrasB-d-siRNA treatment significantly hampers the radial outgrowth in AnrasB silenced lines as compared to controls (Fig. 7B). AnrasB-d-siRNA treatment did not significantly affect the total hyphal mass accumulation of silenced cultures grown in ACM broth for 48 h at 37uC (Fig. 7C).

AnrasA and AnrasB mRNA abundance decreases after treatment with d-siRNAs
To demonstrate the silencing of the endogenous AnrasA and AnrasB mRNAs, total RNA was isolated from AnrasA or AnrasB-d-siRNA treated, unrelated MiAche-d-siRNA treated and untreated control mycelia and cDNA was produced. The relative expression levels of target mRNAs were measured by qRT-PCR in AnrasA or AnrasB-d-siRNA silenced lines. The AnrasA mRNA abundance was found to be reduced by 96.07% in AnrasA-d-siRNA treated mycelia when compared to unrelated MiAche-d-siRNA treated and untreated control mycelia (Fig. 6D). Similarly, AnrasB mRNA level was also reduced by 93.53% in comparison to controls (Fig. 7D). d-siRNA is more potent in silencing target mRNA than chemically synthesized siRNA counterpart To determine the silencing potency of d-siRNA when compared to synthetic siRNA, we have chosen to target AnrasA gene in A. nidulans. Target gene sequence was in silico scrutinized for high score siRNA in order to attain maximum silencing potency [36]. However, in silico predicted siRNA might have different outcome with in vitro assays, even if it possesses a high score. So, we eliminated the three non-efficient chemically synthesized siRNAs (data not shown) and we compared the silencing potency of the best working (in vitro assay) synthetic AnrasA-siRNA versus its d-siRNA counterpart. In silico designed AnrasA-siRNAs were chemically synthesized and dissolved in nuclease-free water. A. nidulans conidia were grown for 6 h in ACM and subsequently treated with synthetic AnrasA-siRNA and diced AnrasA-siRNA with final concentrations of 25 nM for another 12 h. Mycelial tissue was then collected from synthetic or diced AnrasA siRNA treated samples and controls. Total RNA was isolated from frozen mycelial samples and cDNA was synthesized. To analyze the AnrasA mRNA abundance in control and siRNA treated samples, qRT-PCR was performed.
qRT-PCR results showed that the level of AnrasA transcript was reduced to 5.30% in AnrasA-d-siRNA silenced lines compared to controls. However, we found that the same concentration of single synthetic AnrasA-siRNA treatment reduced AnrasA transcript levels only to 44.38% (Fig. 8). These results suggest that d-siRNAs in vitro generated by RNase III enzyme is more potent in silencing endogenous genes than single synthetic siRNA in A. nidulans.

Expression profiles of other Ras family genes in AnrasA and AnrasB silenced lines exhibited no change in transcript levels
In order to check the specificity of d-siRNA-mediated gene silencing, we analyzed the expression profiles of closely related, non-intended targets of other Ras family genes in AnrasA and AnrasB-d-siRNA silenced lines. Expression profiles of AnrhbA, An4873, AnmedA and An7661 in both AnrasA and AnrasB-d-siRNA silenced lines, AnrasA expression in AnrasB-silenced line and viceversa were obtained through qRT-PCR analysis. We showed that the transcript levels of all genes tested were found to be normal as in untreated control ( Fig. 9A and 9B). However, transcript level of AnrhbA in AnrasA-d-siRNA silenced line showed slightly elevated level than in untreated control but no such elevation was observed in AnrasB silenced lines. We also showed that the transcript levels of all Ras family genes tested in un-diced AnrasA or AnrasB dsRNA treated lines were found to be normal as in untreated control ( Fig.  S2A and S2B). These results suggest that d-siRNA mediated gene silencing seem to be target-specific, as no other similar or closely related genes were affected.

Discussion
RNAi is an emerging and powerful reverse genetic approach to characterize gene functions in various filamentous fungi [12]. With the explosion of sequenced fungal genomes (www.broad.mit.edu/ annotation/fungi/fgi/), there is an increasing interest in looking beyond the genome and investigating the function of genes. Conventionally, deciphering gene function in filamentous fungi has been performed by targeted gene disruption either by insertion or knockout method. This strategy is time-consuming and not feasible for large scale functional genomic studies. The majority of Determination of the total biomass. Germinating conidia were treated with AnrasB-d-siRNA in ACM broth for 48 h and total fresh biomass formation was determined. The data represent the means of three replicates. Values were compared using t test. * Significant difference at P,0.05 as compared to untreated control. (D) Quantification of AnrasB silencing by qRT-PCR. Relative AnrasB expression was measured in AnrasB-d-siRNA treated mycelia and controls using comparative D cycle threshold (CT) method. AnrasB values were normalized to AnActin values. The data represent the means of three replicates. Values were compared using t test. * Significant difference at P,0.05 as compared to untreated control. doi:10.1371/journal.pone.0075443.g007 the genes present in the genome are critical for the survival of the fungus and deleting any gene for functional study may result in a lethal phenotype. Hence, an alternative method which can generate down-regulation of gene activity rather than complete deletion of target gene might be more informative [40]. Thus, siRNA-induced gene silencing offers a convenient method for targeting specific genes in filamentous fungi. Previous reports have shown that siRNA-induced gene silencing targeted against ornithine decarboxylase (ODC) gene, a key polyamine biosynthesis gene in A. nidulans. This revealed the importance of cellular polyamines for proper growth of germ tube and mycelia. It also demonstrated successful uptake of siRNA molecules (25 nM optimum concentration) by germinating fungal spores from culture medium, which ultimately supports the simple transfection of small RNA species into the germinating spores or mycelia [17]. Subsequently, another group also reported that this small RNA uptake mechanism can be exploited to induce siRNA-mediated silencing of endogenous genes in A. fumigatus [41]. In our first attempt, we employed this simple small RNA uptake mechanism for silencing of fungal genes by RNase III generated diced siRNAs in A. nidulans.
The evaluation of a new RNAi induction route is made more comfortable by silencing of a stably introduced reporter gene, which can facilitate the visible quantification of silencing effect. Hence, we optimized stable introduction of GFP reporter gene into A. nidulans genome by AMT. Although it has been previously demonstrated that Agrobacterium tumefaciens is capable of transforming various fungi including the Aspergillus species [34,42], the transformation conditions must be optimized because the transformation frequencies vary among fungal species and strains. We have prepared a transformation vector (pCAMBIA1300-sGFP) which contains the HPT as a marker gene driven by CaMV35S promoter and sGFP gene driven by fungal-specific TrpC promoter using pCAMBIA 1300 T-DNA backbone. The CaMV35S promoter is widely used in multitude of plant species for driving constitutive expression of transgenes. However, only one such report is available in phytopathogenic fungus Fusarium oxysporum where they have shown that CaMV35S promoter efficiently drives the marker gene expression [43]. Hence, we utilized the CaMV35S promoter to drive the HPT marker gene expression in A. nidulans. In consistent with the earlier results, A. nidulans transformants showed small colony size, indicating the CaMV35S promoter is not as efficient in driving selection marker gene expression as in plants. However, we showed that non-promoter expression of HPT marker gene (leaky expression of HPT gene) was not sufficient to generate A. nidulans transformants on selection medium containing 100 mg/l hygromycin B. This observation suggests that the CaMV35S promoter may be useful in driving the marker gene expression in A. nidulans as the small colony size of the transformants is not interfering with the study. Moreover, most of the A. tumefaciens binary vectors contain CaMV35S promoter for driving (hygromycin B) marker gene expression, the vectors can directly be used in transformation of A. nidulans without any further modifications.
We have successfully optimized the AMT in A. nidulans based on previous protocol described in A. fumigatus [34] with the transformation frequency of approximately 80 transformants for 10 7 conidia. The frequency of transformation was comparable to that of the transformation frequency attained in A. fumigatus in the previous study. Molecular characterization by PCR and RT-PCR analyses confirmed the presence and expression of transgene in some of the putative fungal transformants. Further, transgene integration in the transformants was confirmed by Southern hybridization. A. nidulans transformant under study was found to have stable expression of GFP in different developmental stages tested.
The silencing potency of siRNA is mainly dependent on how effectively the siRNA is incorporated into RISC [44,45,46]. Besides the available computational siRNA prediction tools, it is necessary to test the several synthetic siRNAs in order to identify a potent siRNA duplex. Synthetic siRNAs can produce substantial off-target effects as a result of cross hybridization between the seed region of the siRNA and the 39UTR of the off-target transcripts [28,47]. Recently, it has been shown that testing of multiple siRNAs and off-target effects can be circumvented by generation of siRNAs using RNase III or Dicer enzyme [28]. The digestion of in vitro transcribed dsRNA with RNase III produces a mixture of siRNAs covering a much larger region of the target sequence. This usually obviates the need to test multiple sequences for each target gene, making the silencing more effective and having fewer offtarget effects than any one single synthetic siRNA [30,31].
We provided evidences for an efficient gene silencing induced by RNase III generated siRNAs in A. nidulans. In the first step, we showed that the GFP reporter gene stably introduced into A. nidulans genome was silenced by treating with GFP-d-siRNAs. GFP fluorescence in mycelia treated with GFP-d-siRNA was drastically reduced when compared to untreated mycelia, indicating that the silencing by RNase III-diced-siRNA is highly efficient. This was supported by reduction in mRNA level of GFP (89.6%) in silenced lines in comparison to controls. Our results are in agreement with the previously reported GFP dsRNA or hairpin GFP construct induced silencing in multitude of fungi, such as Moniliophthora perniciosa [20], Colletotrichum lagenarium [48], Magnaporthe oryzae [40], Venturia inaequalis [49], Phytophthora infestans [50] and Coprinopsis cinerea [51].
To be useful as a rapid reverse genetic tool, d-siRNA-mediated gene silencing must work for endogenous genes. Towards this goal, we attempted to silence an endogenous Ras homolog AnrasA gene in A. nidulans. We reported that the AnrasA-d-siRNA treatment resulted in severe reduction in hyphal growth, reduced radial outgrowth and total biomass accumulation. The phenotypic abnormalities exhibited by the AnrasA-d-siRNA silenced mycelia were supported by strong reduction in corresponding mRNA abundance. To date, the isolation of a viable deletion mutant of rasA gene has not been reported in A. nidulans, suggesting that rasA activity is essential; thus, deletion would be lethal. Although rasA deletion mutant in A. nidulans has never been achieved, dominantactive or dominant-negative rasA mutant inhibits asexual development [37,52]. Our results demonstrate that functional analysis of essential genes may be possible with their transient downregulation using d-siRNAs.
To further assess the efficacy of d-siRNA-induced endogenous gene knockdown, we attempted the silencing of an uncharacterized Ras homolog, rasB, in A. nidulans. We found that downregulation of AnrasB by rasB-d-siRNA resulted in abnormal hyphal branching, malformed conidiophore and reduction in radial outgrowth, which was in consistent with the drastic reduction in mRNA levels (93.53% reduction) of rasB gene when compared to controls. Our results are in line with those obtained with rasB deletion mutant phenotype in A. fumigatus. The deletion of rasB in A. fumigatus has been reported to exhibit delayed growth, decrease in germination rate and abnormal hyphal morphology [39].
Silencing efficacy of siRNA is crucial for any successful RNAi experiment. In order to assess the silencing efficacy of the RNase III-diced-siRNA in A. nidulans, we treated the fungal mycelia with equal concentrations of synthetic siRNA or d-siRNA targeting an endogenous gene (AnrasA) separately. We found that the silencing potency of AnrasA by d-siRNA was relatively higher than that by synthetic siRNA. In mycelia treated with d-siRNA, AnrasA mRNA level was reduced to 5.30%, whereas, the same concentration of synthetic siRNA treatment resulted in reduction of mRNA level to only 44.38%. These results are in agreement with previous results obtained with RNase III-diced-siRNAs in cultured mammalian cells [29].
RNAi mainly relies on homology-dependent base pairing leading to cognate mRNA degradation and it may sometimes cause non-intended silencing effects as well. In order to investigate off-target effects caused by d-siRNA, we evaluated its effect on other closely related Ras gene family in AnrasA or AnrasB gene silenced lines. We found that silencing of AnrasA or AnrasB gene with respective d-siRNAs did not alter the expressions of other Ras family genes. However, we observed slightly elevated expression of some of the Ras family genes tested. The probable explanation for this observation is due to the down-regulation of important gene which might lead to elevation in the expression levels of some genes through unknown mechanism. Previously, it has been reported that silencing of heterologous GFP gene with GFP-dsRNA resulted in insignificant elevation of inf1 endogenous gene in P. infestans [50]. Though, the phylogenic and sequence analyses ( Fig.  S3 and S4) revealed that AnrasA and AnrasB were close homologs; the silencing of AnrasA did not alter the expression of AnrasB transcript level and vice-versa. Similarly, we also found that the treatment of un-diced AnrasA or AnrasB dsRNA did not alter the expression profiles of other Ras family genes. These results are in line with dsRNA-induced sequence specific silencing in P. infestans [50] and tobacco rattle virus mediated transient silencing of Manduca sexta midgut-based CYP6B46 gene [53]. Altogether, our data suggests that the silencing induced by d-siRNA is highly specific and did not affect even closely related genes of the same family.
We conclude that silencing of A. nidulans genes by RNase IIIdiced-siRNA is highly specific, efficient and does not rely on laborious methods. This method has a great potential to be used as a tool for deciphering the function of fungal genes. This method also facilitates the analyzing of essential genes in A. nidulans through down-regulation of the intended target. We succeeded in elucidating the function of an uncharacterized Ras homolog rasB gene in A. nidulans. The off-target effects of d-siRNA mediated gene silencing are undetectable but can further be reduced by prior off-target analysis before generation of d-siRNA pools. Using this simple and effective gene silencing method, rapid characterization of a large number of candidate genes could be possible in filamentous fungi.

Materials and Methods
Fungal culture and spore isolation A. nidulans strain (strain-b1 A1 ribo A1-wild yellow) was cultured on solid complete medium according to Barratt et al. [54] in 90615 mm Petri dishes. For spore collection, sterile doubledistilled water (8 ml) was added to Petri dish containing A. nidulans culture, and the surface was scraped gently with a sterile glass rod to release the spores. The resulting spore suspension was filtered through glass wool and centrifuged at 12000 rpm for 10 min at 4uC. The spores were washed with sterile water and finally resuspended in sterile nuclease-free water. For liquid growth, A. nidulans was cultured at 37uC upto the indicated time-points in ACM broth.

Development of pCAMBIA1300-sGFP construct
The pCAMBIA1300-sGFP expression vector was generated using the T-DNA vector pCAMBIA1300 as the skeleton. The sGFP gene was PCR amplified from the pMT-sGFP vector (procured from the Fungal Genetics and Stock Centre, USA) and cloned in pGEM T-easy vector (Promega, USA). Then, sGFP gene was sub-cloned in pSilent-1 vector in XhoI and KpnI restriction sites by removing the spacer DNA and finally the expression cassette was introduced into pCAMBIA1300 backbone at the XbaI restriction site (Fig. S5), which possess HPT gene as the antibiotic selection marker. For the comparison of HPT marker gene expression under the control of CaMV35S promoter with nonpromoter expression, the CaMV35S promoter fused to the HPT marker gene was released from the pCAMBIA1300 vector by digesting with SmaI restriction enzyme. Then the resulting linearized pCAMBIA1300 vector was self-circularized using T4 DNA ligase (Fermentas, USA).

Transformation of A. nidulans
AMT was carried out using the basic protocol as previously described in A. fumigatus [34] with some modifications. All the media used in various steps of transformation were prepared as mentioned in the previous protocols [34]. A. tumefaciens EHA105 strain harboring pCAMBIA1300-sGFP vector was grown in YEM broth supplemented with 50 mg/ml of kanamycin and 100 mg/ml rifampicin on a rotatory shaker (250 rpm) at 28u C for 12 h. One ml of the Agrobacterium culture was inoculated into 9.0 ml of induction medium (IM) with appropriate concentration of acetosyringone (AS) and incubated on a shaker (250 rpm) at 28uC for 6 h or until an optical density at 660 nm of 0.8 was reached. Fungal conidia and Agrobacterium were co-cultivated at a ratio of 1:10 respectively. A total of 100 ml of the Agrobacterium culture with OD = 0.8 was mixed with 100 ml of 10 7 fungal conidia and spread on to the nylon membrane (MDI, India) placed on IM-AS agar plate. Then, the IM-AS co-cultured plates containing the nylon membrane were incubated at 24u C in the dark for 48 h. In order to select the fungal transformants, nylon membranes were transferred to selection medium (SM) [ACM+100 mg/l hygromycin B+300 mg/l augumentin] plates and incubated at 37uC for 3 days.

DNA isolation, PCR and Southern hybridization
Several independent putative fungal colonies raised in AMT were maintained on SM agar plates with appropriate hygromycin B concentration. For genomic DNA isolation, untransformed wildtype strain and fungal transformants were inoculated on to the ACM and grown at 37uC for 48 h in a rotary shaker. Genomic DNA was isolated from untransformed control and four putative fungal transformants using fungal DNA isolation kit (Himedia Labs, India) according to the manufacture's instructions. The putative fungal transformants were screened for the presence of transgenes using HPT and GFP specific primers.
Transgene integration in several fungal transformants was further confirmed by Southern analysis. Genomic DNA (10 mg) was digested with XhoI restriction enzyme and separated on 0.8% agarose gel. Then, blots were prepared by standard protocol [55] using nylon membrane (MDI, India). The PCR amplified and purified 500 bp HPT DNA was radio labeled with radioactive a-p32-dCTP using Megaprime DNA labeling kit (G. E Healthcare, USA) according to the manufacturer's instructions. Pre-hybridization and hybridization were carried out as previously described [55].
In silico analysis and cloning of AnrasA and AnrasB partial genes The coding sequences of AnrasA and AnrasB were retrieved from Aspergillus genome database (www.aspgd.org/). The presence of candidate siRNAs was evaluated using an algorithm developed by Bingbing Yuan et al. [35]. Further, siRNA scoring was done as previously described [36].
Total RNA was isolated from A. nidulans sub-merged culture using fungal RNA isolation kit (Himedia, India) and cDNA was prepared using Superscript (Invitrogen, USA) according to manufacturer's instructions. The partial coding sequences of AnrasA and AnrasB were PCR amplified using respective genespecific primers (Table S1) from A. nidulans cDNA. The AnrasA and AnrasB PCR products were purified using PCR clean up system (Fermentas, USA) and cloned in pGEM-T easy (Promega, USA) vector and confirmed by automated DNA sequencing.

Double-standard RNA (dsRNA) preparation
The dsRNAs were synthesized via in vitro transcription with the highscribe T7 in vitro transcription system (Fermentas, USA) using PCR products as templates for sGFP, AnrasA and AnrasB target genes. The T7 promoter sequence was placed in front of both the gene specific forward and reverse primers which were used for PCR amplification of the template for dsRNA synthesis. In vitro transcription was carried out according to the manufacturer's instructions. For synthesis of unrelated dsRNA, Meladogina incognata acetylcholineesterase (MiAchE) was PCR amplified from M. incognata cDNA and cloned in pGEM-T easy vector (already available in the lab). MiAchE was re-amplified from pGEM T-MiAChE with M13 primers and PCR products were used separately in T7 and SP6 transcription systems in separate reactions. Annealing of single standard RNAs was carried out in TE buffer at 65uC followed by snap cooling. The dsRNA quality was monitored by agarose gel electrophoresis and the concentration was determined by spectrophotometry.
Preparation of diced-siRNA pools and chemically synthesized siRNA The dsRNAs were digested with Short Cut RNase III (New England Biolabs, USA) to prepare diced-siRNA pools according to the manufacturer's protocol. Briefly, 10 mg dsRNAs were digested with Short Cut RNase III in a 100 ml reaction at 37uC for 30 min. Then 10 ml of 10X EDTA (0.5 M) was added to stop the reaction, and the small RNAs were precipitated in presence of nuclease-free glycogen and one-tenth volume of 3 M Sodium acetate (pH 5.2) with 3 volumes of chilled ethanol. Finally purified d-siRNAs were dissolved in nuclease-free water and integrity was analyzed by 20% native PAGE.
The custom-designed and chemically synthesized AnrasA genespecific siRNA used in this study was procured from Sigma, USA. The stock solutions of 100 mM were prepared and stored at 280u C until use. The 59-39 sequences of the sense and antisense strands of AnrasA are given below: Sense: 59 CCAUGCGCGAA-CAAUAUAUTT 39Antisense: 59 AUAUAUUGUUCGCG- CAUGGTT 39 Treatment of A. nidulans mycelia with siRNAs A. nidulans mycelia was treated with synthetic siRNA or d-siRNAs by the procedure previously described [17,41] with minor modifications. Briefly 1610 8 spores/ml (instead of 1610 7 spores/ ml) were incubated at 37uC and 250 rpm in liquid ACM for 6 h to obtain germinating spores. Subsequently, siRNAs were added to the liquid medium at a final concentration of 25 nM. Germinating conidia were treated for another 12 h. Later, cultures were harvested and stored at 280uC until RNA extraction.

qRT-PCR analysis
Total RNA was extracted from siRNA treated and untreated samples by using fungal RNA isolation kit (Himedia labs, India) according to the manufacturer's instructions. cDNA was generated from total RNA using Superscript reverse transcriptase (Invitrogen, USA) according to manufacturer's instructions.
qRT-PCR was performed using SYBR Green I technology in Realplex 2 thermal cycler (Eppendorf, Germany). A master mix was prepared by using MESA Blue qPCR mix as a PCR core reagent (Eurogentec, USA). 1.5 ng of cDNA and 750 nM each of the specific primers were added in a final volume of 10 ml (Table  S1). The amplification reactions were carried out at 95uC 4 min, 40 cycles at 95uC 15 sec followed by 60uC for 1 min. Specificity of amplification was assessed by disassociation or melt curve analysis at 60-95uC after 40 cycles. After completion of qRT-PCR, data was analyzed by using Realplex 2 software. Relative quantification of mRNA levels was performed by the comparative D cycle threshold (CT) method using AnActin mRNA as an internal control.

Microscopic analysis of d-siRNA treated fungal culture
A. nidulans spores (final concentration of 1610 4 /ml per well) were incubated in 1 ml of ACM broth in a 24 well cell-culture plate and incubated at 37uC for 6 h in order to obtain germinating conidia. Subsequently, d-siRNAs directed against sGFP or AnrasA or AnrasB transcripts were added to the liquid medium to get a final concentration of 25 nM. Fungal cultures were treated for another 12 h. Pictures by microscopy were taken 18 h after seeding.

Radial outgrowth and total biomass accumulation assays
Radial outgrowth assay was performed as previously described [17] with some modifications. Briefly, 5 ml of 10 4 germinating spores ml 21 (instead of 3 ml of 10 8 germinating spores ml 21 ) was treated with respective d-siRNAs and were spotted in the center of ACM agar plates. The change in colony diameter was monitored after 48 h. Determination of total biomass was performed by inoculating 10 4 conidia into ACM broth in pre-weighed, sterile 2 ml microcentrifuge tubes and incubated for 6 h at 37uC at 250 rpm to obtain germinating conidia. Consequently, 25 nM final concentrations of d-siRNAs against target genes were added. After 48 h seeding, reduction in total biomass in d-siRNA treated samples was calculated as follows: percentage of total biomass reduction = 100-(Total biomass in d-siRNA treated lines/ Total biomass in control) X 100.  Figure S5 Cloning strategy for construction of pCAM-BIA1300-sGFP vector. The sGFP gene was PCR amplified from the pMT-sGFP vector and cloned in pGEM T-easy vector. Then, the sGFP gene was sub-cloned in pSilent-1 vector in XhoI and KpnI restriction sites by removing the spacer DNA and finally expression cassette was introduced into pCAMBIA1300 backbone at the XbaI restriction site. (TIF)

Supporting Information
Table S1 List of primers used in the study. All the primers used in the study were listed in the table along with their applications. The primers used in qRT-PCR analysis were designed using ''Primer Quest'' software (Integrated DNA Technologies, USA); remaining primers were designed using Primer3 version v.0.4.0 software. (DOC) Table S2 A: List of siRNA candidates predicted in AnrasA target gene. The partial sequences of the target gene was evaluated for presence of candidate siRNAs using the siRNA selection software. All the identified candidate siRNAs were scrutinized for silencing potency using the standard parameters. List of candidate siRNAs along with their probable mRNA target site and score of each individual siRNA were presented in the table. B: List of siRNA candidates predicted in AnrasB partial nucleotide sequence. The partial sequences of the target gene was evaluated for presence of candidate siRNAs using the siRNA selection software. All the identified candidate siRNAs were scrutinized for silencing potency using the standard parameters. List of candidate siRNAs along with their probable mRNA target site and score of each individual siRNA were presented in the table. C: List of siRNA candidates predicted in sGFP nucleotide sequence. The partial sequences of the target gene was evaluated for presence of candidate siRNAs using the siRNA selection software. All the identified candidate siRNAs were scrutinized for silencing potency using the standard parameters. List of candidate siRNAs along with their probable mRNA target site and score of each individual siRNA were presented in the table. (DOC)