Negative regulation and developmental competence in Aspergillus

Asexual development (conidiation) in the filamentous fungus Aspergillus nidulans is governed by orchestrated gene expression. The three key negative regulators of conidiation SfgA, VosA, and NsdD act at different control point in the developmental genetic cascade. Here, we have revealed that NsdD is a key repressor affecting the quantity of asexual spores in Aspergillus. Moreover, nullifying both nsdD and vosA results in abundant formation of the development specific structure conidiophores even at 12 h of liquid culture, and near constitutive activation of conidiation, indicating that acquisition of developmental competence involves the removal of negative regulation exerted by both NsdD and VosA. NsdD’s role in repressing conidiation is conserved in other aspergilli, as deleting nsdD causes enhanced and precocious activation of conidiation in Aspergillus fumigatus or Aspergillus flavus. In vivo NsdD-DNA interaction analyses identify three NsdD binding regions in the promoter of the essential activator of conidiation brlA, indicating a direct repressive role of NsdD in conidiation. Importantly, loss of flbC or flbD encoding upstream activators of brlA in the absence of nsdD results in delayed activation of brlA, suggesting distinct positive roles of FlbC and FlbD in conidiation. A genetic model depicting regulation of conidiation in A. nidulans is presented.

A key event responding to the developmental inductive signal is activation of brlA, which encodes a C 2 H 2 zinc finger transcription factor (TF) (Fig. 1B) 5 . Further genetic and biochemical studies have identified the abaA and wetA genes as necessary regulators of conidiation. The abaA gene encodes a putative TF that is activated by brlA during the middle stages of conidiophore development after differentiation of metulae 6,7 . The wetA gene, activated by AbaA, functions in late phase of conidiation for the synthesis of crucial cell wall components and conidial metabolic remodeling 8,9 . These three genes have been proposed to define a central regulatory pathway that acts in concert with other genes to control conidiation-specific gene expression and determine the sequence of gene activation during conidiophore development and spore maturation [10][11][12] (reviewed in ref. 1). Subsequent studies have identified various upstream developmental activators (UDAs), fluG, flbA, flbB, flbC, flbD, and flbE that influence brlA expression (Fig. 1B) [13][14][15] . Mutations in any of these genes result in "fluffy" colonies that are characterized by undifferentiated cotton-like masses of vegetative cells (reviewed in ref. 1). Each of the FlbB, FlbC and FlbD proteins contains a DNA binding domain and they are shown to be direct activators of brlA expression 16,17 . The two genetic cascades composed of fluG →→ flbE/flbB→ flbD → brlA, and fluG →→ flbC → brlA were proposed, in which fluG functions upstream 18 .
Our studies to further understand the developmental control mechanisms have identified three key negative regulators of conidiation, SfgA, VosA, and NsdD [19][20][21] . The fluG suppressor sfgA is predicted to encode a Zn(II) 2 Cys 6 domain protein, and positioned between FluG and FLBs ( Fig. 1B) 16,22 . The velvet domain TF VosA and the GATA-type TF NsdD were isolated via gain-of-function genetic screens as repressors of conidiation 19,21 . VosA, which is activated by AbaA, governs spore maturation and exerts negative feedback regulation of brlA by binding to the 11 nucleotide VosA responsive element (VRE) in the brlAβ promoter 19,23 . NsdD, initially identified as a key activator of sexual fruiting 24 , was found to be also a key repressor of conidiation 21 . The deletion of nsdD bypasses the needs for FluG and all UDAs, but not brlA, for conidiation, indicating that NsdD acts downstream of UDAs and upstream or at the same level of brlA 25 .
In the present study, we further investigate negative regulation of conidiation and developmental competence. Through combinatorial genetic studies, we have found that VosA and NsdD are the major factors repressing brlA expression, and thereby influencing the acquisition of developmental competence in A. nidulans. We also report that the repressive role of NsdD in conidiation is conserved in other aspergilli. In A. nidulans conidia, NsdD directly binds to the brlAβ promoter region, which contains a GATAA sequence potentially interacting with NsdD. We also demonstrate that FlbC and FlbD are necessary for full activation of brlA even in the absence of nsdD. A working genetic model depicting the positive and negative regulations of brlA expression and conidiation in A. nidulans is presented.

Results
NsdD is a key factor determining the number of conidia. Previously, we showed that vosA and nsdD play an additive role in repressing conidiation and brlA expression in vegetative cells 21 . To further expand our understanding on the genetic interactions of the three negative regulators, we generated double mutants: ∆ nsdD ∆ vosA, ∆ sfgA ∆ nsdD and ∆ sfgA ∆ vosA. We then quantified the conidiation levels of FGSC4 (wild type; WT), ∆ sfgA, ∆ nsdD, ∆ vosA, ∆ nsdD ∆ vosA, ∆ sfgA ∆ nsdD and ∆ sfgA ∆ vosA strains by spreading conidia onto solid MMG and incubating for 2 days. As shown in Fig. 1C, the ∆ nsdD mutant produced ~2.3 fold more conidia than WT and other mutant strains (p < 0.001). The ∆ sfgA ∆ nsdD double mutant produced less number of conidia than the ∆ nsdD single mutant, but more than WT (p < 0.001). On the contrary, the ∆ sfgA ∆ vosA mutant produced a highly reduced number of conidia, ~6 fold less than WT and the ∆ vosA mutant. The ∆ nsdD ∆ vosA double mutant produced a similar number of conidia to WT. These results suggest that NsdD is a major determinant of the number of conidia being produced on solid culture condition. The ∆ sfgA ∆ vosA mutant exhibited a highly reduced number of conidia than each single mutant, suggesting that the ∆ sfgA and ∆ vosA mutations have synthetic negative effects on conidiogenesis.
NsdD and VosA cooperatively repress brlA expression and conidiation. One approach to investigate elevated or hyper activation of conidiation is to grow the strains in liquid shake culture and check conidiophore development and mRNA levels of brlA. Under this condition, WT hardly ever produces asexual developmental structure. When WT, ∆ nsdD, ∆ vosA, ∆ sfgA, ∆ sfgA ∆ vosA, ∆ sfgA ∆ nsdD and ∆ nsdD ∆ vosA strains were examined at 16 h liquid shake culture, only the ∆ nsdD ∆ vosA double mutant formed a high number of conidiophores ( Fig. 2A). We then examined the mRNA levels of brlA, abaA and wetA in WT and various mutant strains at 16 h of vegetative growth, and found that only the Δ nsdD Δ vosA double mutant showed a high level accumulation of brlA mRNA (Fig. 2B). Accumulation of abaA and wetA mRNA was consistent with the brlA mRNA expression pattern in the Δ nsdD Δ vosA mutant. These led us to determine the levels of brlA mRNA in WT, Δ nsdD, Δ vosA, and Δ nsdD Δ vosA strains in conidia and very early phases of growth (4~16 h of liquid culture). As shown in Fig. 2C, the Δ vosA mutant displayed a high level of brlA mRNA in conidia and somewhat reduced levels of brlA mRNA in vegetative cells, lacking further activation of brlA expression. On the contrary, the Δ nsdD Δ vosA mutant exhibited a high level of brlA mRNA in conidia, and began to show induced activation of brlA expression even at 6 h of liquid culture, and a sudden strong activation of brlA expression at 10 h and thereafter. In fact, the Δ nsdD Δ vosA mutant formed conidiophores as early as 12 h of liquid culture (data now shown). These findings indicate that NsdD and VosA are major negative regulators of brlA expression and conidiation, and that the removal of the repressive effects imposed by NsdD and VosA might be a key factor determining the developmental competence.
We also check timing of conidiation on solid air-exposed culture condition. Somewhat consistent with the above findings, the time required for the first conidiophore formation in a colony derived from a single conidium on solid medium was about 25 h in the Δ nsdD Δ vosA mutant (Fig. 2D). The Δ sfgA Δ nsdD and Δ nsdD mutants showed initial conidiophore development at 27 h. The Δ sfgA Δ vosA and Δ sfgA mutants formed the first conidiophore at 29 h. The Δ vosA showed elaboration of conidiophore at ~31 h, whereas WT formed the first conidiophore at 33 h. These results suggest collectively that negative regulation of brlA by both NsdD and VosA is a key attribute determining the developmental competence.

NsdD represses conidiation in A. flavus and A. fumigatus. All
Aspergillus species appear to have an ortholog of NsdD (AspGD; http://www.aspgd.org/). The predicted NsdD polypeptide, especially the GATA domain in the C-terminus, is highly conserved in A. flavus and A. fumigatus (Fig. 3A). We hypothesized that NsdD might play a similar repressive role in conidiation in these aspergilli.
In A. flavus, nsdD mRNA levels are high in conidia, and undulate during the lifecycle (Fig. 3B). The deletion of nsdD in A. flavus by replacing its coding region with the pyrG + marker from A. fumigatus, caused restricted colony growth coupled with abnormal conidiophores compared to WT (NRRL3357; Fig. 3C). The size of the ∆ nsdD conidiophores was averaged 45.81 μ m, whereas WT conidiophore size was averaged 125.6 μ m (P < 0.005; data not shown). This is consistent with the previous report demonstrating that NsdD is a major determinant of developmental morphogenesis 26 . We then examined levels of conidiation in varying ways, and found that the absence of nsdD resulted in hyper-active conidiation evidenced by the elaboration of a high number of conidiophores at 28 h of liquid shake culture (Fig. 3D), the formation of abundant conidiophores imbedded in agar at 28 h of solid culture (Fig. 3E), as well as enhanced production of conidia per plate (Fig. 3F). These results indicate that NsdD is a key repressor of conidiation in A. flavus.
In A. fumigatus, nsdD is somewhat constitutively expressed, and its mRNA levels are high in conidia (Fig. 4A). Similar to A. nidulans and A. flavus, the deletion of nsdD in A. fumigatus caused restricted hyphal growth (Fig. 4B), and early and uncontrolled activation of conidiation leading to the formation of conidiophores at 19 h liquid shake culture (Fig. 4C), and elaboration of a high number of conidiophores imbedded in agar at 28 h of solid culture (Fig. 4D), and enhanced production of conidia per plate (Fig. 4E). As the mycotoxin gliotoxin (GT)  biosynthesis is activated by BrlA 25,27 , the deletion of nsdD resulted in elevated production of GT (Fig. 4F). These results indicate that NsdD functions as a negative regulator of conidiation and GT production in the opportunistic human pathogen A. fumigatus.
NsdD directly binds to the brlAβ promoter region. We previously reported that, in the FluG-mediated conidiation pathway, NsdD functions downstream of FlbE/B/D/C and upstream of brlA 21 . In a simplistic interpretation, we hypothesized that NsdD directly binds to upstream of the brlA coding sequence. To test this hypothesis, we first scanned the brlA promoter region spanning 2 kb with JASPAR CORE database (http://jaspar.genereg.net) to search potential GATA-TF binding sites. NsdD contains a zinc finger GATA binding domain at 394 th -446 th amino acid at its C-terminus, which may interact with a core  Our previous study revealed that nsdD encodes two distinct transcripts designated as nsdDβ and nsdDα 21 . The nsdDα transcript specifically accumulates in conidia, whereas the nsdDβ constitutively accumulates throughout the lifecycle. Specific expression of nsdDα in conidia requires activity of both VosA and VelB during the formation of spores 23 . The nsdDβ and nsdDα transcripts contain 1,037 nt and 150 nt of 5′ untranslated region (UTR), respectively (Fig. 5B). Further analyses of cDNAs by RT-PCR indicate that these transcripts are predicted to encode the NsdDβ (461aa) and NsdDα (424aa) polypeptides, where NsdDα lacks the first 37 aa found in NsdDβ . To check the presence and expression levels of the predicted two NsdD proteins during the lifecycle, we carried out Western blot analysis employing a strain ectopically expressing NsdD::3XFLAG in Δ nsdD (TMK13, Table 1) and anti-FLAG antibody. We found that levels of both the NsdDβ and NsdDα proteins were very low in conidia, high in vegetative cells, then NsdDα became undetectable at 6 h post developmental induction (Fig. 5C). Employing the TMK13 strain and anti-FLAG antibody, we pulled-down the NsdD interacting DNAs in conidia, and PCR-amplified the five regions containing a core GATA site in the brlA promoter (ChIP-PCR). As shown in Fig. 5A,D, the three regions spanning − 1,950, − 1,013, and − 894 containing the GATAA sequence in the + strand gave rise amplicons. However, the two regions containing GATAA in the − strand were not enriched by NsdD-ChIP. Taken together, while the precise NsdD binding sequence should be identified and validated, multiple NsdD might occupy the brlAβ promoter region. It can be further proposed that binding of NsdD and VosA to the brlA promoter results in the full repressive control of brlA expression, and the developmental competence might be determined by the removal of these key direct negative regulators of brlA (see Discussion).

The positive roles of FlbC and FlbD in conidiation.
We previously showed that the deletion of nsdD suppressed the conidiation defects caused by the absence of FLBs 21 . We further asked whether the primary role of FLBs is to remove the NsdD-mediated negative regulation or they play distinct roles in activating conidiation. This was done by determining timing and levels of brlA in the Δ nsdD single and ∆ flbC ∆ nsdD and ∆ flbD ∆ nsdD double mutants. As shown in Fig. 6A, brlA accumulation was not observed in WT vegetative conditions, and was slightly increased 12 h post asexual developmental induction. The deletion of nsdD caused brlA mRNA accumulation at 36 h in submerged culture (vegetative), and at high levels at 12 and 24 h post induction. Importantly, we found that, even in the absence of nsdD, the deletion of flbC, or flbD resulted in significantly reduced and delayed accumulation of brlA mRNA at 24 and 48 h post induction. These results indicate that these positive regulators play distinct positive roles in activating brlA expression, and their activities are needed for full activation of brlA.

Discussion
Conidiation in Aspergillus occurs as an integral part of the life cycle primarily controlled by the intrinsic genetic program rather than as a response to unfavorable environmental conditions 1 . Neither the concentration of a limiting nutrient such as glucose or nitrogen source, nor continuous transfer to new medium modifies the timing with which cells become competent to develop 1,28 .
Given that the timing of competence acquisition is endogenous and genetically determined, one can ask how the fungus keeps track of the time that has transpired following germination. Timberlake 29 came up with an explanation for this by proposing a repressor of conidiation, which becomes diluted during early growth. In this model, a fixed amount of repressor would be produced during the final stages of conidium differentiation and stored in the spore. During the (~18 h of) vegetative growth, such a repressor is diluted to a critical concentration before development can proceed. Timberlake further speculated that it could be a negatively acting TF that prevents expression of genes required for conidiation, e.g., brlA, and mutational inactivation of such a repressor would be expected to lead to precocious development.
Indeed, collectively, our studies have revealed that there are at least three negative regulators of conidiation, and that a key event for the acquisition of developmental competence is to remove the repressive effects imposed by NsdD and VosA. We further have found that the positive upstream regulators are needed for maximum level conidiation, but not for the commencement of development. This is based on the fact that the deletion of nsdD could bypass the need for fluG, flbB, flbE, flbD, and flbC, but not brlA, in conidiation 21 . Importantly, for the first time, we demonstrated that NsdD physically binds to three different regions in the brlAβ promoter, further supporting the idea that NsdD directly (rather than indirectly) represses the onset of brlAβ expression and conidiation.
NsdD is a GATA TF with a highly conserved DNA-binding domain consisting a Cys2-Cys2 type IV zinc finger in its C-terminal basic region 24   composed of the GATAA core sequence. As shown in Fig. 5, the three regions containing GATAA in the + strand, but not the two regions with GATAA in the − strand, can be enriched by NsdD-ChIP. The GATAA sequence at − 1950 is positioned between the predicted binding sites for the two key UDAs, FlbB and FlbD 13,17 . The two sites at − 1013 and − 894 may overlap with the RNA polymerase binding region (the brlAβ transcript is marked by the arrow-head line in Fig. 5A). A revised VosA responsive element (VRE; TGGCTTGGGCTGG) is positioned at − 1,496 between the two predicted FlbC binding sites 16,23 . Thus, binding of multiple NsdD and one VosA-VelB in the promoter region of brlAβ may effectively inhibit the initiation of brlAβ transcription. In fact, as shown in Fig. 2A-D, the absence of both nsdD and vosA resulted in near constitutive expression of brlA throughout the life cycle, and abundant/precocious conidiophore development in liquid culture (as early as 12 h). However, the observation that about 25 h is required for the elaboration of the first conidiophore in the Δ nsdD Δ vosA mutant colony derived from a conidium (Fig. 2D) suggests that a single spore must undergo vegetative growth for a certain period even in the absence of the key negative regulators of conidiation.
The NsdD polypeptide(s) is highly conserved in most (if not all) Aspergillus species (Fig. 3A) and other fungi including Penicillium, Coccidioides, Ajellomyces, and Fusarium (not shown). Moreover, the A. fumigatus and A. flavus nsdD genes appear to encode two transcripts (and polypeptides; see the second Met position Fig. 3A,B  and 4A). We demonstrated that the role of NsdD in negatively controlling brlA and conidiation is conserved in these two species. In A. fumigatus, there are four GATAA sequences at − 2276, − 1148, − 976, and − 932, where the BrlA ATG is + 1. In A. flavus, two GATAA sequences are present at − 1413 and − 1389, where the BrlA ATG is + 1. In both cases, the deletion of nsdD resulted in precocious and enhanced conidiation, which is consistent with a previous report 31 . Early and increased production of GT in the A. fumigatus nsdD mutant can be explained by precocious and enhanced expression of BrlA, which in turn directly activates gliotoxin biosynthesis 27,32 .
Tight repression of brlA in a conidium and for a certain period of vegetative growth is important for the fitness of Aspergillus fungi. Adams and Timberlake 33 showed that overexpression of brlA in vegetative cells resulted in complete cessation of growth and generalized losses of protein and RNA. Collectively, we present a genetic model depicting the negative and positive regulations and the commencement of conidiation in A. nidulans (Fig. 6). During the formation of conidia, VosA and VelB are activated by AbaA 34 , which in turn activate expression of the lower transcript of nsdD in conidia 23 . VosA and multiple NsdD are bound to the upstream regulatory region of brlA, which confers full repression of brlA and conidiation. SfgA acts as an upstream negative regulator of conidiation functioning downstream of FluG 20 . During early phase of vegetative growth, FluG accumulates to a certain level, which then removes the repressive effects of SfgA, thereby allowing UDAs (FlbB/D and FlbC) to function. Acquisition of the developmental competence might also involve the translocation of FlbB to the hyphal tip, became transcriptionally competent, then entering into the nucleus 35 . In order for activated FlbB-FlbD 17 and FlbC to trigger brlA expression and conidiation, both NsdD and VosA need to be removed from the brlA promoter. Currently, we do not know how NsdD and VosA are displaced from the brlA promoter. One possible explanation is degradation of the VosA and NsdD proteins. Upon removal of NsdD and VosA coupled with the cooperative activity of FLBs, brlAβ is expressed above threshold, which then fully activates itself and brlAα, triggering development of conidiophores 21 . While not shown in the model, activated BrlA leads to expression of AbaA, which in turn activates expression of VosA, VelB, and WetA in phialides and conidia (see Fig. 1B). The VosA-VelB heterodimer shuts off expression of brlA and β-glucan biosynthetic genes, and activates genes associated with trehalose biogenesis and nsdDα in conidia, allowing full repression of brlA for next generation 19,34,36 .
Finally, while we presented a simplified single-path model for conidiation, it is important to note that that regulation of development is a complex multi-degree process involving both activation of the FluG-initiated conidiation pathway and inhibition of FadA-mediated G protein signaling pathway for vegetative growth [37][38][39][40][41] . In A. nidulans, various G protein mutants displayed precocious activation of conidiation [42][43][44] . Moreover, high level accumulation of brlA alone might not be sufficient to trigger conidiation as shown in our ricA study 45 . In A. fumigatus, various developmental regulators including velvet proteins, G-proteins, and RAS proteins govern conidiation (reviewed in ref. 46 and 47). Additional studies integrating genome-wide and systems analyses are in progress to better address the developmental control mechanisms in Aspergillus.

Methods
Fungal strains and culture conditions. The Aspergillus strains used in this study are listed in Table 1.
A. nidulans strains were grown on solid or in liquid minimal medium with 1% glucose (MMG) with supplements as described previously 48 at 7 °C. To determine the numbers of conidia in WT and mutant strains, approximately 10 5 spores were spread onto solid MMG and incubated at 37 °C for 2 days. The conidia were collected from the entire plate and counted using a hemocytometer. To check elaboration of conidiophores in liquid submerged culture, conidia (10 6 /ml) of individual strains were inoculated in liquid MMG and incubated at 37 °C, 220 rpm. For Northern blot analyses, samples were collected as described 49 . Briefly, for vegetative growth, conidia (10 6 /ml) of strains were inoculated in liquid MMG and cultured at 37 °C, 220 rpm. Samples of liquid submerged culture were collected at designated time points. Induction of asexual development or sexual development was done as described previously 49 .
A. flavus and A. fumigatus strains were grown on solid or in liquid MMG with 0.1% yeast extract (YE, v/v) and supplements as described 48,50 , at 30 °C and 37 °C, respectively. To check elaboration of conidiophores in liquid submerged culture, conidia (2 × 10 5 /ml) of individual strains were inoculated in liquid MMG with 0.5% YE and incubated, 220 rpm. For developmental induction, vegetative cells were collected and transferred to solid medium, and the culture plates were air exposed for asexual developmental induction or tightly sealed induction in dark condition as described 49 . Construction of A. nidulans strains. The oligonucleotides used in this study are listed in Table S1. The double joint PCR (DJ-PCR) method 51 was used to generate the ∆ sfgA ∆ vosA, ∆ nsdD ∆ vosA and ∆ sfgA ∆ nsdD mutants. Both 5′ and 3′ flanking regions of the sfgA and nsdD genes were amplified from genomic DNA of FGSC4 using OMK556;OMK557 and OM558;OMK559 (for sfgA), and OMK562;OMK563 and OMK564;OMK565 (for nsdD). The A. nidulans pyroA + marker was amplified with the primer pair ONK395;ONK396. The final DJ-PCR sfgA deletion construct was amplified with OMK560;OMK561, and the nsdD deletion construct was amplified with OMK566;OMK567. The sfgA deletion amplicon was introduced into THS15.1 to generate the ∆ sfgA ∆ vosA mutant. The nsdD deletion amplicon was introduced into THS15.1 and TNJ57 to generate the ∆ nsdD ∆ vosA and ∆ sfgA ∆ nsdD mutants, respectively. Protoplasts were generated using the Vinoflow FCE lysing enzyme (Novozymes) 52 . At least three independent deletion mutant strains were isolated. To complement ∆ nsdD and epitope-tag NsdD, the FGSC4 nsdD fragment including its 2kb 5′ and coding regions was amplified with the primer pair OMK574;OMK575, digested with PstI and NotI, and cloned into the pHS13 vector 34 , which contains 3/4 pyroA 53 , a 3xFLAG tag, and the trpC terminator. The resulting plasmid pMK20 was then introduced into the recipient ∆ nsdD strain TNJ108, and several TMK13 class transformants expressing the WT NsdD fused with the 3XFLAG tag under its native promoter have been isolated and confirmed. A. flavus and A. fumigatus strains. The nsdD gene was deleted in A. flavus NRRL3357.5 (pyrG − ) 54 and A. fumigatus AFU293.1 (pyrG − ) 55 employing DJ-PCR 51 . In A. flavus strain, the 5′ and 3′ flanking regions of the nsdD gene were amplified using A. flavus WT (NRRL3357) genomic DNA with the primer pairs ONK1037;ONK1038 and ONK1039;ONK1040. The A. fumigatus pyrG + marker was amplified from A. fumigatus WT (AFU293) genomic DNA with the primer pair OMK589;OMK590. The 5′ and 3′ flanking regions of nsdD were fused to the marker, and the resulting fusion product was further amplified by the nested primer pair ONK1041;ONK1042. The final deletion construct was introduced into A. flavus NRRL3357.5, and the ∆ nsdD mutant (LNJ11) was isolated and confirmed by PCR followed by restriction enzyme digestion 51 .

Construction of
In A. fumigatus strain, the 5′ and 3′ flanking regions of nsdD were amplified from A. fumigatus WT (AFU293) with the primer pairs ONK1043;ONK1044 and ONK1045;ONK1046. The A. nidulans pyrG + marker was amplified from FGSC4 genomic DNA with the primer pair OHS696;OHS697. The 5′ and 3′ flanking regions of nsdD were fused to the marker, and the fusion product was further amplified by the nested primer pair ONK1047;ONK1048. The final deletion construct was introduced into A. fumigatus AFU293.1, and the ∆ nsdD mutant (LNJ12) was isolated and confirmed by PCR followed by restriction enzyme digestion 51 . At least three independent deletion strains were isolated and confirmed.
Nucleic acid isolation and manipulation. Genomic DNA and total RNA isolation was carried out as previously described 18,56 . In Northern blot analyses, DNA probes were prepared by PCR amplification of the coding region of individual genes with appropriate oligonucleotide pairs using FGSC4 genomic DNA as template (Table S1). Probes were labelled with 32 P-dCTP (PerkinElmer) using Random Primer DNA Labeling Kit (Takara) and purified by illustra MicroSpin G-25 columns (GE Healthcare).
Scientific RepoRts | 6:28874 | DOI: 10.1038/srep28874 acetate-formic acid (5:4:1, v/v/v) as mobile phase, where the R(f ) value of GT was ~0.61. Photographs of TLC plates were taken following exposure to UV (365 nm) using a Sony DSC-T70 digital camera. This analysis was performed in triplicates.
Microscopy. The colony photographs were taken by using a Sony digital camera (DSC-F28). Photomicrographs were taken using a Zeiss M 2 Bio microscope equipped with AxioCam and AxioVision (Rel. 4.8) digital imaging software.