Fungal development and secondary metabolism is intimately associated via activities of the fungi-specific velvet family proteins. Here we characterize the four velvet regulators in the opportunistic human pathogen Aspergillus fumigatus. The deletion of AfuvosA, AfuveA and AfuvelB causes hyperactive asexual development (conidiation) and precocious and elevated accumulation of AfubrlA during developmental progression. Moreover, the absence of AfuvosA, AfuveA or AfuvelB results in the abundant formation of conidiophores and highly increased AfubrlA mRNA accumulation in liquid submerged culture, suggesting that they act as repressors of conidiation. The deletion of AfuvosA or AfuvelB causes a reduction in conidial trehalose amount, long-term spore viability, conidial tolerance to oxidative and UV stresses, and accelerated and elevated conidial germination regardless of the presence or absence of an external carbon source, suggesting an interdependent role of them in many aspects of fungal biology. Genetic studies suggest that AfuAbaA activates AfuvosA and AfuvelB expression during the mid to late phase of conidiation. Finally, the AfuveA null mutation can be fully complemented by Aspergillus nidulansVeA, which can physically interact with AfuVelB and AfuLaeA in vivo. A model depicting the similar yet different roles of the velvet regulators governing conidiation and sporogenesis in A. fumigatus is presented.
Aspergillus fumigatus is a saprophytic ascomycete fungus that plays a crucial role in nutrient recycling. It is also the most prevalent airborne fungal pathogen (Latge, 1999; 2001). This airborne fungus propagates in the environment by producing a massive number of asexually derived spores called conidia. Inhaled conidia are normally removed by the innate immune system including alveolar macrophages. However, inhalation of conidia by immune-compromised individuals causes serious invasive pulmonary aspergillosis with a human mortality rate of at least 50% (Latge, 1999; Dagenais and Keller, 2009; McCormick et al., 2010; Cramer et al., 2011). Moreover, the airborne fungal spores contain potent allergens to which certain people respond with exaggerated hypersensitive reaction causing allergic bronchopulmonary aspergillosis (Stevens et al., 2003; Tillie-Leblond and Tonnel, 2005).
Asexual development (conidiation) in Aspergillus involves the formation of elaborate multicellular conidia-bearing structures called conidiophores (Adams et al., 1998; Yu, 2010). The asexual reproductive cycle of A. fumigatus can be divided into vegetative growth and developmental phases. Vegetative growth begins with the spore germination leading to the formation of a network of undifferentiated interconnected hyphae known as the mycelium. After a certain period of hyphal proliferation, in response to appropriate stimuli (e.g. exposure to air or nutrient deficiency), some of the hyphal cells cease normal growth and initiate conidiation, which can be subdivided into conidiophore formation and spore maturation. Conidiophores consist of a thick-walled foot cell, stalk, vesicle, phialides (no metulae) and (up to 50 000) conidia (Adams et al., 1998; Rhodes and Askew, 2010; Yu, 2010). The process of conidiation is a precisely timed and genetically programmed event responding to internal and external signals (Mirabito et al., 1989; Adams et al., 1998). Our previous studies demonstrated that BrlA, AbaA and WetA control spatial and temporal expression of conidiation-specific genes during conidiophore development and spore maturation in A. fumigatus (Mah and Yu, 2006; Tao and Yu, 2011). The roles of these central regulatory components are conserved in two distantly related fungi: Aspergillus nidulans (Ani; model) and A. fumigatus (Afu; pathogen) (Mah and Yu, 2006; Yu et al., 2006; Yu, 2010; Tao and Yu, 2011).
Fungal conidia contain high levels (up to 15% of the dry mass) of trehalose (α-d-glucopyranosyl-α-d-glucopyranoside), which plays a crucial role in long-term spore viability and tolerance against environmental stresses including heat, oxidative and osmotic stresses (Sussman and Lingappa, 1959; Winkler et al., 1991; Elbein et al., 2003; Paul et al., 2008). Importantly, trehalose biosynthesis is involved in the virulence of several pathogenic fungi including A. fumigatus, Candida albicans and Cryptococcus neoformans (Alvarez-Peral et al., 2002; Petzold et al., 2006; Martinez-Esparza et al., 2007; Al-Bader et al., 2010; Puttikamonkul et al., 2010). Due to the lack of orthologues involved in trehalose biosynthesis in humans and the importance of trehalose biosynthesis in fungal pathogenesis, proteins associated with trehalose biosynthesis have been considered as potential antifungal drug targets. Recent study demonstrated that AfuWetA plays an essential role in trehalose biogenesis in conidia of A. fumigatus (Tao and Yu, 2011).
Our previous studies have demonstrated that AniVosA plays a key role in trehalose biogenesis in conidia, negative feedback regulation of conidiation and conidial maturation in A. nidulans (Ni and Yu, 2007). AniVosA and three other proteins (AniVeA, AniVelB and AniVelC) share high levels of similarity (Ni and Yu, 2007), and together they define the velvet family regulators. The velvet regulators are highly conserved, are only found in filamentous ascomycetes and basidiomycetes and all contain the velvet domain (Ni and Yu, 2007; Bayram and Braus, 2012). They play differential roles in governing development, sporogenesis and secondary metabolism by forming the multiple velvet complexes such as VelB–VeA–LaeA, VelB–VosA and VelB–VelB in A. nidulans (Bayram et al., 2008a; Sarikaya Bayram et al., 2010; Park et al., 2012). The velvet regulators, particularly VeA and VelB homologues, have been characterized in various fungi and shown to be involved in controlling development, secondary metabolism and pathogenicity (reviewed in Calvo, 2008; Bayram and Braus, 2012). However, these crucial regulators have not been characterized in A. fumigatus.
In this study, we characterize the four velvet genes by genetic and biochemical approaches in A. fumigatus, and present evidence that they play pivotal roles in governing development, trehalose biosynthesis and conidial germination. Somewhat distinct from A. nidulans, the deletion of AfuveA or AfuvelB results in hyperactivation of conidiation, leading to abundant formation of conidiophores in liquid submerged culture and near constitutive accumulation of high levels of AfubrlA mRNA in vegetative growth, suggesting that AfuVeA and AfuVelB are the primary negative regulators of conidiation during vegetative growth. It appears that the feedback control of development upon completion of conidiophore is conferred by AfuVelB and AfuVosA, which are activated by AfuAbaA. In addition, both AfuVosA and AfuVelB participate in coupling trehalose biogenesis in conidia and sporogenesis, and AfuVosA, AfuVeA and AfuVelB play a differential role in controlling conidial germination. Finally, we show that the AfuveA null mutation is complemented by AniveA and AniVeA physically interacts with AfuVelB and AfuLaeA in A. fumigatus. We propose a genetic model depicting the velvet-mediated developmental regulation in A. fumigatus.
Summary of A. fumigatus velvets
The AfuvosA gene (EF544392, AFUG_4G10860) was previously identified by a genome search (Ni and Yu, 2007). The ORF of AfuvosA consists of 2012 bp nucleotides (nt) with 10 exons, predicted to encode a 435 aa length protein (Fig. 1A). As shown in Fig. 1B, levels of AfuvosA transcript are particularly high in conidia (C) and during the early vegetative growth. Then, the levels drop quickly during late vegetative growth (vegetative 9–24 h) and increase during late phase of conidiation (asexual 12, 24, 48 and 72). The AfuVosA protein contains the velvet domain in the N-terminal half and a putative NLS-pat7 (239 PVKRQRT 245; http://psort.hgc.jp/form2.html). Searching the Pfam database revealed that the AfuVosA C-terminal portion is similar to the PAT1 protein, a topoisomerase II-associated deadenylation-dependent mRNA-decapping factor, necessary for accurate chromosome transmission and mRNA turnover in Saccharomyces cerevisiae (Fig. 1C) (Wang et al., 1996; Marnef and Standart, 2010).
The AfuveA gene (CAE47975; AFUA_1G12490) was previously reported by Krappmann et al. (2005): the ORF of AfuveA consists of 1713 bp, interrupted by one intron (70 nt), encoding a 570-amino-acid length protein (63.0 kDa). Northern blot analysis revealed that AfuveA mRNA is detectable at 3 h of vegetative growth, maintained at high levels during vegetative growth and the early phase (6–24 h) of conidiation, then reduced (Fig. 1B). Protein database searches revealed that AfuVeA contains the velvet domain and a putative classical bipartite nuclear localization signal (NLS) in the N-terminal region (Ni and Yu, 2007; Stinnett et al., 2007). Whereas AniVeA contains a putative export signal (NES) (Bayram et al., 2008b), no canonical NES motif is found in AfuVeA (http://www.cbs.dtu.dk/services/NetNES/). The Web-based program ePESTfind (http://emboss.bioinformatics.nl/cgi-bin/emboss/epestfind) predicted a putative PEST sequence for rapid degradation at the N terminus of AfuVeA.
The AfuvelB gene (EAL87909; AFUA_1G01970) is composed of a 1255 bp ORF with three (64, 58 and 62 nt) introns (Fig. 1A). As shown in Fig. 1B, levels of the AfuvelB transcript(s) are high in conidia (C) and during the late phase of conidiation (asexual 24, 48 and 72 h) when conidia differentiate and become mature. Levels of the AfuvelB transcript(s) quickly drop during vegetative growth and early asexual developmental induction. The AfuvelB gene appears to encode two overlapping transcripts that are detectable during the late phase of conidiation and sporogenesis. Previously, Bayram and Braus proposed that VelB contains two velvet domains in A. nidulans (Bayram and Braus, 2012). Our further analyses of the velvet domain using ClustalW and MEME (Bailey et al., 1994) revealed that it contains three conserved motifs (boxes in Fig. 1C). The N-terminal and C-terminal parts of AfuVelB include motif 1 and motifs 2&3, respectively, separated by 90 aa residues (Fig. 1D), suggesting that AfuVelB contains one velvet domain.
The ORF of AfuvelC (EAL89811; AFUA_4G09770) consists of 1584 bp, interrupted by one intron (144 nt), and is predicted to encode a 479-amino-acid length protein (52.0 kDa) containing the velvet domain in the C-terminal region (Fig. 1A and C). The AfuVelC protein contains one putative PEST sequence (ePESTfind) in the N-terminal region (Fig. 1C). We examined mRNA levels of AfuvelC and found that levels of AfuvelC mRNA are low during vegetative growth, increase at 6 and 12 h of conidiation and are undetectable in conidia (Fig. 1B).
Differential roles of velvets in asexual development
To investigate the roles of the Afuvelvet genes in A. fumigatus, we generated Afuvelvet null mutants (ΔAfuvelvet) by replacing the entire coding region of each Afuvelvet gene with the A. nidulans pyrG+ or argB+ marker. We also generated complemented strains for each Afuvelvet deletion by introducing the wild-type (WT) allele (Fig. 2A). Multiple ΔAfuvelvet and complemented strains displaying identical phenotypes were isolated and further examined. The first noticeable changes were that, when point-inoculated on solid medium, the deletion of AfuveA and AfuvelB resulted in a significant reduction (P <0.05) of radial colony growth (∼ 80% of WT) and enhanced accumulation of green pigment(s), implying hyperactive conidiation (Fig. 2A and B).
When we further examined the effects of Afuvelvet during asexual developmental induction, we found that, while WT and ΔAfuvelC strains began to show formation of conidiophores at 5 h, the ΔAfuvosA, ΔAfuveA and ΔAfuvelB mutants started to produce conidiophores at 3–4 h after induction of development (data not shown). Further examination of mRNA levels of AfubrlA and AfuabaA in WT and ΔAfuvelvet strains revealed precocious and elevated accumulation of these two key developmental activators by the absence of AfuveA, AfuvelB or AfuvosA. As shown in Fig. 2C, in WT accumulation of AfubrlA mRNA increased from 6 h, peaked at 12 h, decreased after 24 h and became almost undetectable at 48 h post developmental induction. AfubrlA mRNA in the ΔAfuvosA mutant started to accumulate at 0 h, increased from 6 h, peaked at 24 h and decreased after 48 h. The deletion of AfuveA resulted in near constitutive accumulation of high levels of AfubrlA mRNA at 0–48 h post developmental induction. AfubrlA mRNA levels in ΔAfuvelB strains were high during early (0 h) asexual development and somewhat declined during late conidiation. In the ΔAfuvelC mutant it appears that levels of AfubrlA mRNA decreased compared to WT. AfuabaA mRNA levels were high in the ΔAfuveA, ΔAfuvelB and ΔAfuvosA mutants, but low in the ΔAfuvelC mutant (Fig. 2C). These results indicate that Afuvelvet regulators play differential roles in (negatively) controlling conidiation and expression of AfubrlA during the progression of development.
Negative regulation of conidiation by AfuVosA, AfuVeA and AfuVelB
To further investigate the developmental regulatory roles of Afuvelvet, conidia of WT, ΔAfuvelvet and complemented strains were inoculated into liquid minimal medium with glucose (MMG) with 0.1% yeast extract (YE), and examined for the conidiophore formation. As shown in Fig. 3A, whereas WT, ΔAfuvelC and complemented strains did not elaborate conidiophores, the absence of AfuvosA caused swelling of the hyphal tips and production of some conidiophores. Furthermore, the deletion of AfuveA and AfuvelB resulted in the formation of a high number of conidiophores and also green conidia in liquid submerged culture, suggesting that AfuVeA and AfuVelB play a primary role in proper repression of conidiation during vegetative proliferation (Fig. 3A). We further examined whether the deletion of Afuvelvet affected AfubrlA mRNA levels via Northern blot analyses. As shown in Fig. 3B, AfubrlA mRNA in the ΔAfuvosA mutant was detectable at 24 h of liquid submerged culture conditions, and accumulated even higher levels in the ΔAfuveA and ΔAfuvelB vegetative cells (see V24 and V48 in Fig. 3B).
To test a genetic relationship between AfuVelB and AfuVeA or AfuVelB and AfuVosA, we generated the ΔAfuveA ΔAfuvelB and ΔAfuvelB ΔAfuvosA double mutants and examined conidiophore formation and AfubrlA mRNA levels. Similar to the AfuveA and AfuvelB single deletion mutants, both ΔAfuveA ΔAfuvelB and ΔAfuvelB ΔAfuvosA double mutants exhibited abundant formation of conidiophores and green conidia, and enhanced accumulation of AfubrlA in liquid submerged culture (Fig. 3B). These corroborate the idea that AfuVeA and AfuVelB are the principal negative regulators of conidiation during vegetative growth in A. fumigatus, and they may play an interdependent role in repression of development, most likely by forming a heterocomplex(es). We further propose that AfuVosA and AfuVelB play a primary role in negative feedback regulation of conidiation and brlA upon completion of conidiogenesis in A. fumigatus (see Discussion).
VosA and VelB are required for proper trehalose biogenesis in Afu conidia
Previous studies demonstrated that both VelB and VosA play an essential role in trehalose biogenesis in conidia and sporogenesis in A. nidulans (Ni and Yu, 2007; Sarikaya Bayram et al., 2010). To test a potential role of AfuVelvet regulators in spore maturation and trehalose biosynthesis, viability of conidia, trehalose amount and tolerance of conidia to oxidative and UV stresses were examined and compared. For the viability, the ability of 30 day-old WT and ΔAfuvelvet conidia to generate colonies was examined. The results revealed that the ΔAfuvosA and ΔAfuvelB conidia exhibited ∼ 40% reduced viability compared to conidia of WT and other strains (data not shown). To test whether the AfuVelvet regulators are needed for the proper accumulation of trehalose in spore, trehalose concentrations in 2 day-old fresh conidia of WT, ΔAfuvelvet and complemented strains were measured. In the ΔAfuvosA and ΔAfuvelB mutants conidia, trehalose content was decreased compared to that of WT, whereas amounts of trehalose of conidia of ΔAfuveA and ΔAfuvelC strains were comparable to that of conidia of WT, suggesting that AfuVelB and AfuVosA are required for the proper biogenesis of trehalose in conidia (Fig. 4A).
Because trehalose serves as a vital protectant against various environmental stresses, we further questioned whether the absence of AfuvosA and AfuvelB resulted in increased sensitivity of the conidia against oxidative and UV stresses. We found that the ΔAfuvosA and ΔAfuvelB mutant conidia exhibited reduced tolerance to oxidative and UV stresses (Fig. 4B and C).
Our previous studies suggested that AniVelB primarily interacts with AniVosA in asexual and sexual spores, and AniVelB and AniVosA play an interdependent role in trehalose biosynthesis in conidia (Sarikaya Bayram et al., 2010; Park et al., 2012). To examine the relationship between AfuVosA and AfuVelB, we checked trehalose amount and tolerance of spores against oxidative and UV stresses, and found that there were no differences between single and double deletion mutants (Fig. 4). Taken together, these results suggest that both AfuVosA and AfuVelB are required for proper trehalose biosynthesis and stress tolerance of spores, and they play an interdependent (not additive) role in spore maturation.
The roles of Afuvelvet controlling conidial germination
In our previous study, we demonstrated that the absence of AnivelB or AnivosA resulted in elevated conidial germination in A. nidulans, regardless of the presence or absence of an external carbon source (Park et al., 2012). To examine the roles of Afuvelvet in spore germination, we first inoculated conidia of WT, ΔAfuvelvet and complemented strains in liquid MMG with 0.1% YE and examined germ tube formation. As shown in Fig. 5A, WT, ΔAfuvelC and all complement strains exhibited about 30% conidial germination at 8 h and near 100% germination at 13 h in liquid submerged culture. On the other hand, the ΔAfuvosA, ΔAfuveA and ΔAfuvelB mutants showed 60–70% conidial germination at 8 h and near 100% germination at 11 h in liquid medium. Moreover, the ΔAfuveA ΔAfuvelB and ΔAfuvelB ΔAfuvosA double mutants exhibited increased conidial germination rates similar to individual single deletion strains. To test further, conidia of WT, ΔAfuvelvet and complemented strains were inoculated in liquid medium without an external carbon source and observed for germination. We found that no conidia of all strains showed any sign (swelling) of germination until 48 h (data not shown).
We then examined germination rates of conidia of WT, ΔAfuvelvet and complemented strains on solid MMG or MMG with 0.1% YE. In both media, similar to liquid medium, ΔAfuveA, ΔAfuvelB and ΔAfuvosA single mutants exhibited enhanced conidial germination rates compared to WT (data not shown). Finally, we examined germination rates in the absence of external carbon source by inoculating conidia of WT, ΔAfuvelvet and complemented strains on solid MM with 1.6% agarose as a solidifying agent. As shown in Fig. 5B, at 24 h after inoculation, whereas only 10% of WT, ΔAfuveA, ΔAfuvelC and complemented strain conidia showed germling formation, 80% of the ΔAfuvosA or ΔAfuvelB single mutant conidia and the ΔAfuveA ΔAfuvelB and ΔAfuvelB ΔAfuvosA double mutant conidia germinated (P < 0.05). These indicate that AfuvosA, AfuveA and AfuvelB are associated with controlling conidial germination, and that AfuvosA and AfuvelB play an interdependent role in proper regulation of spore germination in the absence of external carbon source.
AfuAbaA is necessary for expression of AfuvelB and AfuvosA
Previously, we showed that AniAbaA binds to the promoter regions of AnivelB and AnivosA, and activates AnivelB and AnivosA expression during the mid-late phase of conidiation in A. nidulans (Park et al., 2012). We analysed the promoter regions of the AfuvosA, AfuveA and AfuvelB genes for the presence of the predicted AbaA response elements (AREs, 5′-CATTCY-3′, where Y is a pyrimidine; Andrianopoulos and Timberlake, 1994) and TEC1p, the Saccharomyces cerevisiae homologue of AniAbaA, binding sequence (Gavrias et al., 1996; Heise et al., 2010) using the jaspar program (http://jaspar.genereg.net/cgi-bin/jaspar_db.pl?rm=browse&db=core&tax_group=fungi; Vlieghe et al., 2006). As shown in Fig. 6A, one to three predicted AREs are present in the promoter regions of the AfuvosA, AfuveA and AfuvelB genes. We further hypothesized that AfuAbaA participates in the regulation of AfuvosA, AfuveA and AfuvelB expression during conidiation and examined mRNA levels of these genes in WT and ΔAfuabaA strains. The absence of AfuabaA abolished and reduced accumulation of AfuvosA and AfuvelB transcripts, respectively, but did not affect AfuveA expression (Fig. 6B). However, when AfuAbaA was overexpressed under the control of the alcA promoter, levels of AfuveA and AfuvelB, but not AfuvosA, mRNA increased at 6 and 12 h in liquid MMT submerged culture, indicating that AfuAbaA is sufficient to induce the activation of AfuveA and AfuvelB (Fig. 6C). Taken together, these results suggest that AfuAbaA differentially functions in proper expression of AfuvosA, AfuveA and AfuvelB.
Function and interaction of VeA is conserved in two aspergilli
The AniVeA protein regulates sexual development and sterigmatocystin production by interacting with AniVelB and AniLaeA (Bayram et al., 2008a). Many VeA homologues have been shown to control fungal development and secondary metabolism in other filamentous fungi (Calvo, 2008; Bayram and Braus, 2012), and they are functionally conserved among different ascomycetes (Bayram et al., 2008b; Wiemann et al., 2010). To check the functional conservancy of VeA in two aspergilli, we generated strains expressing the AniVeA protein fused with a tap C-terminal tag in the ΔAfuveA background in A. fumigatus, and examined the phenotypes and interacting proteins in vivo. We found that the introduction of AniveA fully restored growth and development in the ΔAfuveA mutant, suggesting the functional conservancy of VeA (Fig. 7A). To further test whether AniVeA also interacts with AfuVelB or AfuLaeA, we used tandem affinity purification (TAP) (Bayram et al., 2008a) and found that the tagged AniVeA was co-purified with AfuVelB and AfuLaeA in vegetative cells (Fig. 7B and C). These data suggest that the function and interaction of VeA is conserved in two Aspergillus species.
In A. nidulans, the velvet regulators play differential roles controlling asexual and sexual development (Kim et al., 2002; Ni and Yu, 2007; Bayram et al., 2008a; Park et al., 2012). AniVelB and AniVeA are absolutely required for sexual development and AniVosA is necessary for proper ascosporogenesis (Ni and Yu, 2007; Bayram et al., 2008a). Our recent studies reveal that AniVelC is a positive regulator of sexual development (H-S. Park et al., unpublished). Additional studies demonstrated that AniVelB acts as a positive regulator of conidiation and AniVosA (either as homodimer or heterodimer with AniVelB) confers feedback regulation of AnibrlA and conidiation (Ni and Yu, 2007; Park et al., 2012). In A. fumigatus, as shown in Fig. 2C, the ΔAfuvosA, ΔAfuveA and ΔAfuvelB mutants exhibited considerably increased accumulation of AfubrlA and AfuabaA mRNAs under the conidiation-inducing conditions. In liquid submerged culture conditions, the ΔAfuvosA, ΔAfuveA and ΔAfuvelB mutants exhibited abundant formation of conidiophores at 24 h, whereas WT, ΔAfuvelC and complemented strains did not develop (Fig. 3A). The ΔAfuvosA, ΔAfuveA and ΔAfuvelB mutants accumulated AfubrlA mRNA at 24 h of vegetative growth, whereas no AfubrlA expression was detected in the WT, ΔAfuvelC and complemented strains (Fig. 3B). These imply that AfuVosA, AfuVeA and AfuVelB are necessary for normal progression of growth and proper-down regulation development in A. fumigatus.
We previously showed that the AniVosA and AniVelB proteins play a major role in trehalose biogenesis of conidia and completion of sporogenesis in A. nidulans (Ni and Yu, 2007; Sarikaya Bayram et al., 2010; Park et al., 2012). Similar to accumulation patterns of AnivosA and AnivelB mRNAs, AfuvosA and AfuvelB transcripts highly accumulate in the late phase of conidiation and mature conidia (Fig. 1B). The phenotypes of conidia caused by mutational inactivation of AnivosA or AnivelB are also similar to those of the ΔAfuvosA and ΔAfuvelB mutants (Figs 4 and 5). These results suggest that the roles of VelB and VosA in coupling trehalose biogenesis and conidiogenesis are conserved in two aspergilli. However, it is important to note that while the trehalose in the ΔAnivelB and ΔAnivosA conidia was absent (Ni and Yu, 2007; Sarikaya Bayram et al., 2010), the conidia of ΔAfuvelB and ΔAfuvosA strains contain a reduced (∼ 60% of WT) amount of trehalose. Also unlike in A. nidulans no significant differences in mRNA levels of AfutpsA, AfutpsB, AfuorlA between WT and ΔAfuvelB or ΔAfuvosA mutant strains were observed (data not shown), indicating that AfuVosA and AfuVelB are not required for proper activation of genes necessary for the accumulation of trehalose in conidia. Moreover, whereas the deletion of AnivosA resulted in high-level accumulation of AnibrlA mRNA in conidia (Ni and Yu, 2007), the deletion of AfuvosA, AfuvelB and/or AfuveA did not cause accumulation of AfubrlA in conidia (Fig. 3B), indicating that feedback regulation of conidiation in A. fumigatus may not be conferred by AfuVelvet regulators alone. Importantly, AfuWetA is required for trehalose biogenesis, conidial maturation and negative feedback regulation of AfubrlA (Tao and Yu, 2011). Tao and Yu showed that the lack of AfuwetA caused the total absence of trehalose in conidia, high-level accumulation of AfubrlA mRNA in conidia and reduced accumulation of AfuvosA mRNA. We propose that AfuWetA together with AfuVelvets confers trehalose biogenesis, the completion of conidiogenesis and feedback regulation of conidiation in A. fumigatus.
Our studies have revealed that the velvet proteins interact with each other and form complexes, including the VelB–VeA–LaeA, VelB–VelB and VosA–VelB complexes, which play differential roles in regulating development, sporogenesis and secondary metabolism in A. nidulans (Bayram et al., 2008a; Sarikaya Bayram et al., 2010). Results of our double mutant analyses are generally in agreement that AfuVelB and AfuVosA, similar to AniVelB and AniVosA, play an interdependent role in trehalose biosynthesis and conidial germination (Figs 4 and 5), indicating that AfuVosA–AfuVelB complex may also be a functional unit in A. fumigatus. Furthermore, our cross-complementation results demonstrate that the AfuveA null mutation is fully complemented by AniveA, and AniVeA physically interacts with AfuVelB and AfuLaeA and may form a heterotrimeric complex (Fig. 7). These results indicate that function and interaction of VeA is conserved in both aspergilli. The components of the AniVelB–AniVeA–AniLaeA complex are required for secondary metabolite production in various filamentous fungi (Kato et al., 2003; Bok and Keller, 2004; Calvo et al., 2004; Bayram et al., 2008a; Kale et al., 2008; Wiemann et al., 2010; Lee et al., 2012; Park et al., 2012). In addition, Bok and Keller previously demonstrated that gliotoxin production was reduced in the ΔAfulaeA mutants in A. fumigatus (Bok and Keller, 2004). Based on these data, we thought that the absence of AfuveA and AfuvelB may affect gliotoxin production and checked the gliotoxin accumulation in WT and mutants. Thin layer chromatography (TLC) analyses, however, indicated that WT, ΔAfuveA and ΔAfuvelB strains showed equal levels of gliotoxin under liquid submerged culture conditions (data not shown). We speculate two possible explanations for these results. First, this heterotrimeric complex is not associated with gliotoxin production and may be involved in other secondary metabolites. Second, a large accumulation of AfubrlA mRNA caused by the absence of AfuveA and AfuvelB may be sufficient to activate gliotoxin production. Several studies support the idea that AfuBrlA may play a vital role in co-ordinating gliotoxin production and conidiation in A. fumigatus (Gardiner and Howlett, 2005; Kwon-Chung and Sugui, 2009; Xiao et al., 2010). Additional molecular studies of the velvet complexes in A. fumigatus need to be carried out.
Multiple AREs are present in the promoter regions of the velvet genes in both aspergilli (Tao and Yu, 2011; Park et al., 2012). In our previous study, we demonstrated that AbaA positively regulates velB and vosA expression during the late phase of asexual development in A. nidulans, and that AbaA binds to the promoters of velB and vosA in vitro (Park et al., 2012). Likewise, AfuAbaA is necessary for proper expression of AfuvelB and AfuvosA during conidiation and in conidia (Fig. 6). Recently, we proposed a model that AfuvosA is primarily activated by AfuAbaA with the assistance of AfuWetA during late phase of conidiation (Tao and Yu, 2011). Collectively, we present working models for developmental regulation in A. nidulans and A. fumigatus (Fig. 8). In both aspergilli, the proposed brlA → abaA → wetA central regulatory pathway in conidiation is fully conserved (Adams et al., 1998; Mah and Yu, 2006; Yu, 2010; Tao and Yu, 2011). VosA and VelB are required proper regulation of conidial germination in both aspergilli and AfuVeA, together with AfuVosA and AfuVelB, plays a negative role in regulating conidial germination. The upstream regulators play similar yet slightly different roles, and are required for proper expression of brlA in both aspergilli (Mah and Yu, 2006; Xiao et al., 2010; Yu, 2010). During early asexual development AniVelB (activator) and AniVelC (repressor) indirectly affect the AnibrlA expression. While AniVelB acts as a positive regulator of the formation of conidiophores and AnibrlA expression in early phase of conidiation, AfuVelB together with AfuVeA acts as a key negative regulator of conidiation and AfubrlA expression during vegetative growth and conidiogenesis. For spore maturation, the two aspergilli share key regulators and the VelB–VosA complex may be a functional unit for controlling trehalose biosynthesis in conidia, tolerance of spores to environmental stresses and feedback regulation of conidiation. Regulation of vosA and velB mRNA expression in the late phase of conidiation is exerted by AbaA in both species. It is important to note that, while AniVosA and AniVelB play an essential role in trehalose accumulation in conidia, it appears that AfuWetA is the principal regulator of trehalose biogenesis and conidial wall maturation. WetA and VosA–VelB co-ordinately regulate the completion of spore maturation and trehalose biosynthesis in both aspergilli (Boylan et al., 1987; Marshall and Timberlake, 1991; Ni and Yu, 2007; Tao and Yu, 2011; Park et al., 2012). Finally, while both AniVelB and AniVeA are required for mycotoxin sterigmatocystin biosynthesis by forming the key heterotrimeric complex AniVelB–AniVeA–AniLaeA, AfuVelB and AfuVeA are not necessary for gliotoxin biosynthesis. The molecular mechanisms governing velvet-mediated regulation are being investigated and the outcomes will expand our knowledge in fungal biology.
Strains, media and culture conditions
Aspergillus fumigatus strains used in this study are listed in Table 1. The fungal strains were grown on solid or liquid MMG and 0.1% YE at 37°C, as previously described (Kafer, 1977; Mah and Yu, 2006). For auxotrophic mutants (strains AF293.1 and AF293.6) the medium was supplemented with 5 mM uridine + 10 mM uracil, and 0.1% arginine (Xue et al., 2004). For liquid submerged culture, about 5 × 105 conidia ml−1 were inoculated into liquid MMG with 0.1% YE and incubated at 37°C.
Table 1. Aspergillus strains used in this study
aThe 3/4 AfupyrG marker in pNJ25 causes targeted integration at the AfupyrG1 locus.
For Northern blot analysis, samples were collected as described (Seo et al., 2003; Ni and Yu, 2007). Briefly, for vegetative growth phases, conidia (5 × 105 conidia ml−1) of WT and mutant strains were inoculated in 100 ml of liquid MMG with 0.1% YE in 250 ml flasks and incubated at 37°C, 220 r.p.m. Individual mycelial samples were collected at designated time points from liquid submerged cultures. For synchronized asexual developmental induction, about 5 × 105 conidia ml−1 WT and relevant mutant strains were inoculated in 500 ml of liquid MMG with 0.1% YE and incubated at 37°C and 250 r.p.m. for 16 h (0 h for developmental induction). The mycelium was harvested by filtering through Miracloth (Calbiochem, CA), transferred to solid MMG with 0.1% YE and incubated at 37°C for air-exposed asexual developmental induction. Samples were collected at various time points post asexual developmental induction. To induce AfuabaA overexpression, MM with 100 mM threonine as sole carbon source (MMT) and 0.1% YE was used. Briefly, to overexpress AfuabaA under the control of the alcA promoter, strains were inoculated in liquid MMG, incubated at 37°C, 220 r.p.m. for 14 h, and the mycelial aggregates were collected, rinsed with liquid MMT, transferred into liquid MMG or MMT with 0.1% YE, and further incubated at 37°C, 220 r.p.m. All samples were collected, squeeze-dried, quick-frozen in liquid N2 and stored at −80°C subject to RNA isolation.
Generation of A. fumigatus mutants
The oligonucleotides used in the present study are listed in Table S1. The Afuvelvet genes were deleted in A. fumigatus AF293.1 (pyrG1) and AF293.6 (pyrG1; argB1) strains (Xue et al., 2004) employing double-joint PCR (DJ-PCR) (Yu et al., 2004). The deletion constructs containing the A. nidulans selective markers with the 5′ and 3′ franking regions of the Afuvelvet genes were introduced into the recipient strains AF293.1 or AF293.6 respectively (Szewczyk et al., 2006). The selective markers were amplified from FGSC4 genomic DNA with the primer pairs oBS8/oBS9 (AnipyrG+) and oKH60/oNK105 (AniargB+). The Afuvelvet mutants were isolated and confirmed by PCR, followed by restricted enzyme digestion (Yu et al., 2004). At least three deletion strains for each case were isolated.
To complement ΔAfuveA, ΔAfuvelB and ΔAfuvelC, single-joint PCR (SJ-PCR) method was used (Yu et al., 2004). The AfuveA, AfuvelB and AfuvelC genes' ORF with presumed promoter and terminator were amplified with specific primer pairs where the 3′ reverse primer carries overlapping sequences with the hygB gene's 5′ end. Amplification of the hygB gene was carried out with primer pairs where the 5′ forward primer carries overlapping sequences with each Afuvelvet gene's 3′ end. The final amplicons were amplified with nested primer pairs OHS335;OHS333 (AfuveA), OHS336;OHS333 (AfuvelB) and OHS336;OHS333 (AfuvelC) and introduced into strains THSF1.1, THSF2.1 and THSF3.1 respectively. For the complementation of the AfuvosA null mutation, two PCR amplicons, the AfuvosA gene region including its presumed promoter and terminator (amplified by OHS203;OHS204) and the marker for AnipyrG (OBS8;OBS9) were co-introduced into a ΔAfuvosA strain (THSF10.1). Multiple complemented strains were isolated and confirmed by PCR and Northern blot analyses. To test the ability of AniveA to complement ΔAfuveA, the plasmid pME3157 (Bayram et al., 2008a) containing AniveA::ctap tag fusion was introduced into the AfuveA deletion strain AfS13 (Krappmann et al., 2005) to yield AfGB11.
Nucleic acid isolation and manipulation
To isolate genomic DNA from A. fumigatus, about 106 conidia were inoculated in 2 ml of liquid MMG + 0.5% YE, and stationary cultured at 37°C for 24 h. The mycelial mat was collected, squeeze-dried and genomic DNA was isolated as described (Yu et al., 2004). All deletion mutants were confirmed by PCR amplification of the coding region of individual genes followed by restriction enzyme digestion amplicons.
Total RNA isolation and Northern blot analyses were carried out as previously described (Han et al., 2004; Mah and Yu, 2006). For total RNA isolation, the prepared sample was homogenized using a Mini Bead beater in the presence of 1 ml of TRIzol® reagent (Invitrogen) and 0.3 ml of silica/zirconium beads (Biospec). RNA extraction was performed according to the manufacturer's instruction (Invitrogen). For Northern blot analyses, total approximately 10 g (per lane) of total RNA isolated from individual samples was separated by electrophoresis using an 1% (v/w) agarose gel containing 3% (v/v) formaldehyde and blotted onto a Hybond-N membrane (Amersham). RNA blots were hybridized with individual probes using modified Church buffer as described (Yu and Leonard, 1995; Xiao et al., 2010). The DNA probes were prepared by PCR amplification of the coding regions of individual genes with appropriate oligonucleotide pairs using AF293 genomic DNA as template (Table S1). Probes were labelled with [32P]dCTP (PerkinElmer) using Random Primer DNA Labeling Kit (Takara) and purified by illustra MicroSpin G-25 columns (GE Healthcare).
Trehalose was extracted from conidia and analysed as described (d'Enfert and Fontaine, 1997; Ni and Yu, 2007). Two day-old conidia (2 × 108) were collected and washed with ddH2O. Conidia were resuspended in 200 μl of ddH2O and incubated 95°C for 20 min, and then the supernatant was separated by centrifugation and collected. The supernatant was then mixed with equal volume of 0.2 M sodium citrate (pH 5.5) and further incubated at 37°C for 8 h with or without 3 mU of trehalase from porcine kidney, which hydrolyses trehalose to glucose. The amount of glucose generated by trehalase was assayed with a glucose assay kit (Sigma) following manufacturer's instruction, and converted into the trehalose amount per 107 conidia (triplicates). Each sample not treated with trehalase served as a negative control.
Oxidative stress tolerance test
Oxidative stress tolerance test was carried out as described (Han et al., 2004; Sarikaya Bayram et al., 2010) with some modifications. Hydrogen peroxide sensitivity of conidia was tested by incubating 1 ml of conidial suspensions containing 105 conidia with varying concentrations (0, 1.25 or 2.50 mM) of H2O2 and incubated for 30 min at room temperature. Each conidia suspension was then diluted with ddH2O, and conidia were inoculated into solid MMG with 0.1% YE. After incubation at 37°C for 48 h, colony numbers were counted and calculated as a survival ratio of the untreated control.
UV light stress tolerance test
UV light tolerance test was carried out as described previously (Lima et al., 2005; Sarikaya Bayram et al., 2010) with some modifications. Two day-old fresh conidia were collected in ddH2O and plated out on solid MMG with 0.1% YE plates (100 conidia per plate). The plates were then irradiated immediately with UV using a UV cross-linker and incubated at 37°C for 48 h. The colony numbers were counted and calculated as a ratio of the untreated control.
Germination of conidia
Germination rates were measured as previously described with a slight modification (Ni et al., 2005). To examine germination levels in liquid media, conidia of WT and mutants were inoculated in 5 ml of liquid MMG with 0.1% YE, or liquid medium lacking a carbon source, and incubated at 37°C, 220 r.p.m. Levels of germination (isotropic growth and germ tube formation) were examined every 1 h after inoculation under a microscope. To examine germination levels on solid media without any carbon sources, conidia of WT and mutants were spread onto solid MM with 1.6% agarose and incubated at 37°C. Levels of germination were examined every 6 h after inoculation under a microscope.
The colony photographs were taken by using a Sony digital camera (DSC-F828). Photomicrographs were taken using a Zeiss M2 BIO microscope equipped with AxioCam and AxioVision digital imaging software (Zeiss).
Tandem affinity purification and LC-MS/MS protein identification
Tandem affinity purification method and peptide identifications were carried out as given in detail (Bayram et al., 2008a).
The authors thank Nicholas Rogall for critically reviewing this manuscript. This work was supported by the National Science Foundation (IOS-0950850), USDA CSREES Hatch (WIS01195) and UW Food Research Institute grants to J. H. Y. The work at Göttingen was supported by the Deutsche Forschungsgemeinschaft (DFG) to G. H. B. (SFB860, FOR1334). The work at KAIST was supported in part by the Intelligent Synthetic Biology Center of Global Frontier Project (2011-0031955) funded by the Ministry of Education, Science and Technology, Korea to S. C. K.