Hyphae of higher fungi are compartmentalized by septa. These septa contain a central pore that allows for inter-compartmental and inter-hyphal cytoplasmic streaming. The cytoplasm within the mycelium is therefore considered to be a continuous system. In this study, however, we demonstrate by laser dissection that 40% of the apical septa of exploring hyphae of Aspergillus oryzae are closed. Closure of septa correlated with the presence of a peroxisome-derived organelle, known as Woronin body, near the septal pore. The location of Woronin bodies in the hyphae was dynamic and, as a result, plugging of the septal pore was reversible. Septal plugging was abolished in a ΔAohex1 strain that cannot form Woronin bodies. Notably, hyphal heterogeneity was also affected in the ΔAohex1 strain. Wild-type strains of A. oryzae showed heterogeneous distribution of GFP between neighbouring hyphae at the outer part of the colony when the reporter was expressed from the promoter of the glucoamylase gene glaA or the α-glucuronidase gene aguA. In contrast, GFP fluorescence showed a normal distribution in the case of the ΔAohex1 strain. Taken together, it is concluded that Woronin bodies maintain hyphal heterogeneity in a fungal mycelium by impeding cytoplasmic continuity.
Cellular heterogeneity within an isogenic cell population is common in prokaryotic and eukaryotic organisms. Heterogeneity of cells can be beneficial for organisms in many ways. For instance, it is important for cell specialization (Kaech and Wherry, 2007) and it may function in the survival of (members of) the population under adverse conditions (Nobile and Mitchell, 2007; Veening et al., 2008). It has been shown that hyphae within a fungal mycelium are also heterogeneous. Hyphal heterogeneity has been observed in the case of gene expression, growth and secretion (Wösten et al., 1991; Moukha et al., 1993a,b; Teertstra et al., 2004; Vinck et al., 2005; 2011; Masai et al., 2006; Levin et al., 2007a,b; Kasuga and Glass, 2008; Etxebeste et al., 2009; de Bekker et al., 2011a,2011b). For instance, only part of the hyphae at the periphery of a vegetative mycelium of Aspergillus niger secrete glucoamylase (Wösten et al., 1991). This is due to heterogeneous expression of glaA in this part of the colony. In fact, two populations of hyphae can be distinguished; one that highly and one that lowly expresses this gene (Vinck et al., 2005). Hyphae that show high glaA expression also highly express other genes encoding secreted proteins. Moreover, they have a high rRNA content and highly express the glyceraldehyde-3-phosphate dehydrogenase gene gpdA (Vinck et al., 2011). From these studies it was concluded that two populations of hyphae can be distinguished at the outer part of the vegetative mycelium, one with a ‘high’ and one with a ‘low’ transcriptional and translational activity. The low activity would be sufficient to support growth but a high activity would be needed to support secretion of high amounts of protein.
Hyphae of Ascomycota and Basidiomycota are compartmentalized by porous septa. The diameter of the septal pore varies between 50 and 500 nm, which allows for passage of cytosol and even organelles (Shatkin and Tatum, 1959; Moore and McAlear, 1962; Gull, 1978; Lew, 2005). Upon mechanical injury, septa of ascomycetes are plugged by Woronin bodies to prevent excessive cytoplasmic bleeding (Trinci and Collinge, 1974; Collinge and Markham, 1985; Jedd and Chua, 2000; Tenney et al., 2000; Soundararajan et al., 2004; Maruyama et al., 2005; Dhavale and Jedd, 2007). The Woronin bodies of Aspergillus nidulans are usually located at the septal pore or in apical regions (Momany et al., 2002). In the case of Neurospora crassa, they are tethered via WSC and Leashin to the cell cortex (Ng et al., 2009; Jedd, 2011). Woronin bodies originate from peroxisomes by Pex11- and WSC-mediated budding (Jedd and Chua, 2000; Liu et al., 2008; Escaño et al., 2009). Pex14 functions in biogenesis of Woronin bodies in N. crassa by playing a role in import of HEX1 in the organelle (Managadze et al., 2007). The lumen of Woronin bodies is filled with hexagonal rods of the HEX1 protein (Hoch and Maxwell, 1974; Jedd and Chua, 2000; Managadze et al., 2010). Deletion of HEX1 in N. crassa, Magnaporthe grisea and Aspergillus oryzae results in the absence of Woronin bodies (Jedd and Chua, 2000; Tenney et al., 2000; Soundararajan et al., 2004; Maruyama et al., 2005).
Here, it is shown that Woronin bodies of A. oryzae plug septa not only in damaged hyphae but also in intact growing hyphae. By doing so, they maintain hyphal heterogeneity in a fungal mycelium by impeding cytoplasmic continuity.
Septal closure during vegetative growth does not depend on environmental conditions
It was assessed whether septa in intact growing hyphae are open or closed. To this end, the wild-type A. oryzae RIB40 strain was grown for 2 days at 30°C in a glass bottom microscopy dish in CD + Met medium. The apical compartments of hyphae at the most outer part of RIB40 colonies were dissected using a UV laser and simultaneously it was monitored whether cytoplasm from the sub-apical, adjacent compartment was streaming through the septum of the damaged compartment. The septum was scored as ‘open’ when cytoplasm was streaming through the septum towards the ruptured compartment (see Movie S1). When cytoplasmic movement quickly ceased within seconds, the septum was scored ‘quickly closed’ (Movie S2). In the case cytoplasmic streaming was not observed, the septum was scored as ‘closed’ (Movie S3). Using these criteria it was found that 57.5 ± 10.6% of the apical septa of the wild-type A. oryzae strain were open (this included the septa that were scored quickly closed) (Fig. 1). In more detail, 5 ± 7.1%, 52.5 ± 17.7% and 42.5 ± 10.6% of the septa were scored as ‘open’, ‘quickly closed’ and ‘closed’ respectively. Similar percentages of open septa were found for the second and third septa (67.5 ± 3.5% and 50 ± 7.1% respectively; Fig. 1). To assess whether the septal plugging state of neighbouring septa correlates, the first three compartments of hyphae were sequentially dissected. It was observed that neighbouring septa of closed compartments can be either open or closed (Table 1). Taken together, these data show that there is no difference in septal plugging incidence between the first three septa of intact growing hyphae at the periphery of an A. oryzae RIB40 colony and that closure of neighbouring septa does not correlate.
Table 1. Absence of association between the plugging state of the three most apical septa with a growing hypha of A. oryzae
1 = septum open, 0 = septum closed.
Aspergillus oryzae RIB40 was grown on CD + Met medium for 2 days at 30°C, after which it was subjected to 4°C or 45°C, hypo- or hypertonic conditions, or pH 2.0 or 11.0. Alternatively, RIB40 was grown for 2 days with C or N limitation. Laser dissection showed that none of the conditions significantly influenced the septal plugging incidence of the first (i.e. the apical) septum (Fig. 2). After incubating the mycelium of RIB40 for 30 min at 4°C or 45°C, 72.5 ± 3.5% and 65 ± 0% of the septa were open respectively (Fig. 2). When it was subjected to 1 M MgSO4 or to H2O for 1.5 h, 57.5 ± 3.5% and 77.5 ± 3.5% of the septa were open respectively. Similarly, 62.5 ± 10.6% and 67.5 ± 10.6% of the septa were open after incubation of the mycelium at pH 2.0 or pH 11.0 for 1.5 h. These numbers were 57.5 ± 10.6% and 70 ± 7.1%, when the mycelium was exposed to carbon and nitrogen starvation for 2 days (Fig. 2).
Woronin bodies are responsible for septal closure during vegetative growth
The A. oryzae ΔAohex1 deletion strain NSRK-ΔHx5 was grown for 2 days at 30°C on M medium and was exposed to standard environmental conditions and to 4°C, 45°C, pH 2.0, pH 11.0, and to carbon and nitrogen starvation. In all cases, 100 ± 0% of the apical septa were open (Fig. 2). In contrast, on average 62 ± 3.5% of the septa of the control strain NSRKu70-1-1AS were open during these conditions. Notably, cytoplasmic movement was observed throughout at least seven compartments after damaging the apical compartment of leading hyphae of NSRK-ΔHx5. In contrast, cytoplasmic movement was only observed in the second and sometimes the third compartment of strain NSRKu70-1-1AS.
Septal closure in strain NSRK-ΔHx5 was partially rescued after introduction of a vector encompassing a gene encoding EGFP–AoHex1 under control of the amyB promoter. In the resulting strain 11NSR-NAGHs 90 ± 0%, 85 ± 7.1% and 85 ± 7.1% of the first, second and third septa of the leading hyphae, respectively, were open under standard growth conditions (Fig. 1). The partial complementation of AoHex1 by EGFP–AoHex1 may be caused by a lower binding of the fusion protein to WSC when compared to the native AoHex1 protein. This binding is needed for proper localization of the Woronin body at the septal pore (Ng et al., 2009). Septa of the 11NSR-NAGHs strain with a Woronin body localized at the septal pore (visualized by the reporter protein EGFP–AoHex1) were always closed (Fig. 3A). In contrast, in the absence of a Woronin body, the septum was open in 100% of the cases. These results show that Woronin bodies of A. oryzae close septa during vegetative growth under various environmental conditions. Confocal laser scanning microscope (CLSM) live cell imaging of three hyphae showed that Woronin body localization at the septum is dynamic. Woronin bodies moved away from the septum between 15 min and several hours of growth (Fig. 3B).
Heterogeneous distribution of GFP resulting from glaA- and aguA-driven expression is abolished in a ΔAohex1 strain
Constructs pAN52-10S65TGFPn/s (Siedenberg et al., 1999) and PaguAsGFP+ (Vinck et al., 2011) encompassing the sGFP gene under control of the glaA and aguA promoters of A. niger, respectively, were introduced in strains NSRKu70-1-1AS and NSRK-ΔHx5 of A. oryzae. Transformants were screened by fluorescence microscopy and two representative strains of each transformation were selected. These strains were called RB#140.1 and RB#140.2 (glaA, Δku70 background), RB#154.3 and RB#154.4 (aguA, Δku70 background), RB#141.3 and RB#141.4 (glaA, Δku70ΔAohex1 background), and RB#156.2 and RB#156.4 (aguA, Δku70ΔAohex1 background). Hyphal heterogeneity of GFP accumulation at the outer part of colonies of these strains was assessed by modelling log transformed fluorescence intensity distributions assuming the existence of two populations of hyphae (i.e. one with high and one with low GFP fluorescence) (Fig. 4). For this, 50–100 μm sections of apical compartments of hyphae were measured. The confidence intervals of the mean in populations with low (μ1) and high (μ2) GFP fluorescence were not overlapping in A. oryzae strains RB#140.1, RB#140.2; RB#154.3 and RB#154.4 that all contain Woronin bodies (Table 2). This showed that glaA- and aguA-driven expression of GFP is heterogeneous in these strains. In contrast, the two confidence intervals of the mean in strains RB#141.3 and RB#141.4, and RB#156.2 and RB#156.4, all having the Aohex1 deletion, did overlap. Thus, the assumption that GFP distribution is heterogeneous in these strains was falsified (Fig. 4, Table 2). Taken together, these data show that Woronin bodies maintain hyphal heterogeneity in the A. oryzae mycelium.
Table 2. 95% CI of the mean assuming a bimodal fluorescence distribution of glaA- or aguA-driven GFP expression after log transformation in hyphae of strains with and without Woronin bodies
Promoter used for GFP expression
n = sample size; CI (μ1) and CI (μ2) represent the confidence intervals of the lower and upper limits of populations 1 and 2 respectively. CI (pf1) represents the confidence interval of the lower and upper limits of the participation fraction of population 1.
Intercellular cytoplasmic connections have been identified in multicellular eukaryotic organisms belonging to the four classical kingdoms. In the case of animals, gap junctions provide the cytoplasmic contact. They mediate interchange of molecules < 1000 Da. Plasmodesmata in plants provide intercellular transport of, for instance, photoassimilates, mRNA and proteins. The pore sizes in septa of fungi are even larger. They allow inter-compartmental streaming of macromolecules and organelles. Gap junctions and plasmodesmata can regulate their pore size and in this way function in developmental processes (Heinlein, 2002; Norris et al., 2008). Little is known about closure of fungal septa. So far, it is generally believed that septa are open in intact fungal hyphae. In this view, the mycelium would consist of a continuous cytoplasm. Recently, it was shown that septa of Schizophyllum commune can already be plugged in intact growing hyphae (van Peer et al., 2009). It was demonstrated that apical septa are open in intact vegetative hyphae of this basidiomycete. In contrast, only 50% and 10% of the second and third septa are open respectively. Septal plugging was shown to be reversible and to depend on environmental conditions (van Peer et al., 2009). A strain in which the spc33 gene was inactivated did not form septal pore caps and, as a result, septal plugging was abolished (van Peer et al., 2010). In this study it was shown that septa of intact hyphae of the ascomycete A. oryzae can also be reversibly plugged. In this case, Woronin bodies are essential for closure of septa. These results refute the general view that a fungal mycelium consists of a continuous cytoplasm. It should be noted that it is not yet clear how general these results apply to other species of ascomycetes and basidiomycetes. We do demonstrate that septal closure results in heterogeneity of cytoplasmic composition of neighbouring hyphae within a mycelium of A. oryzae.
Using laser dissection it was shown that about 60% of the first three septa of hyphae at the periphery of an A. oryzae colony are open. The plugging state of the first septum did not correlate with that of the second or the third septum. The plugging state also did not depend on the environmental conditions. Septa were always closed when a GFP-tagged Woronin body had been localized at the septal pore. Conversely, septa were always open when a Woronin body was absent. This strongly indicated a role for Woronin bodies in plugging septa of intact vegetative hyphae. Indeed, all septa were open, irrespective of the growth conditions, in a ΔAohex1 strain that cannot form these organelles. Live cell imaging showed that Woronin body localization near septa was dynamic in intact vegetative hyphae. Taken together, these results imply that closure of septa in A. oryzae is a reversible process. In contrast to S. commune, there seems not to be a mechanism directing septal closure in intact vegetative hyphae. It seems to be a stochastic driven process.
The general accepted view that the cytoplasm is continuous in a fungal mycelium due to the porosity of septa (Jennings et al., 1974; Jennings, 1987) is in conflict with hyphal heterogeneity as has been observed in A. niger (Wösten et al., 1991; Vinck et al., 2005; 2011; Levin et al., 2007a,b; Etxebeste et al., 2009; de Bekker et al., 2011a) and A. oryzae (Maruyama et al., 2006). Septal plugging in intact hyphae, however, explains why hyphae can be heterogeneous with respect to RNA and protein composition. The fact that each septum has a chance of 40% to be closed implies that only in about 5% of the cases the cytoplasm of two hyphae is in physical contact when they are separated by six septa. Absence of Woronin bodies would result in a continuous cytoplasm and consequently hyphal heterogeneity would be abolished. This hypothesis was tested. It was shown that GFP distribution was heterogeneous between neighbouring hyphae of a wild-type A. oryzae when the reporter was expressed from the A. niger glaA and aguA promoters (Fig. 5A). Two populations of hyphae were distinguished: those with a high and those with a low GFP fluorescence. In contrast, populations of hyphae with low and high GFP content could not be shown to exist in the ΔAohex1 mutant (Fig. 5B). From this it is concluded that hyphal heterogeneity is abolished in strains that do not form Woronin bodies. We propose that heterogeneous gene expression still occurs in the ΔAohex1 mutant. However, since all septa are open, cytoplasmic streaming evenly distributes gene products between neighbouring hyphae.
So far, it is not clear why colonies send out exploring hyphae that are heterogeneous with respect to transcriptional and translational activity. Possibly, this increases the chance that (some of the) hyphae survive when a colony is exposed to stress conditions like the presence of antibiotics, reactive oxygen species or high temperature. Plugging of septa by Woronin bodies would maintain diversity of RNA and protein composition between the hyphae, and thereby would promote survival of the mycelium. In agreement with this, the A. niger ΔAnhex1 strain died when exposed for 3 days to 45°C, whereas the strain with Woronin bodies survived (Bleichrodt, 2012). This effect could not be explained by more excessive bleeding (data not shown).
Strains and growth conditions
Strains used in this study are listed in Table 3. RIB40 was used as wild-type strain (Machida et al., 2005). Strain NSRKu70-1-1A (Escaño et al., 2009) is a derivative of NSRKu70-1-1 (Δku70 niaD− sC− adeA−; Escaño et al., 2009) expressing the adeA selection marker gene. NSRKu70-1-1AS (Tanabe et al., 2011) is a derivative of NSRKu70-1-1A, which has been transformed with the sC selection marker gene. NS4 (niaD− sC−) (Yamada et al., 1997) is the parental strain of NSR13 (niaD− sC− adeA−; Jin et al., 2004). Inactivation of Aohex1 in strain NSRKu70-1-1A, or in NSR13, resulted in strains NSRK-ΔHx5 and NSR-ΔHx11 respectively. Strain 11NSR-NAGHs expresses EGFP–AoHex1 under control of the amyB promoter in strain NSR-ΔHx11. Strains RB#140.1, RB#140.2, RB#154.3 and RB#154.4 are derivatives of A. oryzae NSRKu70-1-1AS (Δku70). The former two express sGFP under the control of the A. niger glaA promoter, while the latter two express sGFP under the control of the A. niger aguA promoter. Similarly, RB#141.3 and RB#141.4 express sGFP under the control of the A. niger glaA promoter, while RB#156.2 and RB#156.4 express sGFP under the control of the A. niger aguA promoter, but these strains are derivatives of A. oryzae NSRK-ΔHx5 (Δku70 ΔAohex1).
To obtain spores for inoculation, A. oryzae was grown on 3.7% PDA (potato dextrose agar, Sigma Aldrich, http://www.sigmaaldrich.com). Spores were harvested in 0.9% NaCl (w/v) containing 0.05% (v/v) Tween-20 and diluted to a final concentration of 5 × 105 spores ml−1.
For microscopy, A. oryzae was grown in glass bottom dishes (MatTek, http://www.glass-bottom-dishes.com, P35G-1.5-20-C) essentially as described by van Peer et al. (2009). Strains with nitrate prototrophy were grown on CD + Met medium (0.3% NaNO3, 0.2% KCl, 0.1% KH2PO4, 0.05% MgSO4.7H2O, 0.002%, FeSO4.7H2O, 1% glucose, 0.0015% methionine, pH 5.5; Maruyama et al., 2010), while strains with nitrate auxotrophy (niaD−) were grown on M medium (0.2% NH4Cl, 0.1% (NH4)2SO4, 0.05% KCl, 0.05% NaCl, 0.1% KH2PO4, 0.05% MgSO4.7H2O, 0.002% FeSO4.7H2O, 1% glucose, pH 5.5; Ohneda et al., 2005). The glass bottom dishes were filled with 30 μl of CD + Met medium or M medium containing 1% agarose. To this end, the glass bottom dishes and the agar medium were pre-warmed at 50°C. Spores (250 spores in 0.5 μl) were placed in the middle of an 18 mm cover glass and placed upside down on the non-solidified agarose medium. This resulted in a 118 μm thin agarose layer. After solidifying the agarose medium, 2 ml of liquid medium was added on top of the culture.
For the septal plugging experiments, cultures were grown for 2 days at 30°C under water saturating conditions, after which they were either or not exposed to stress conditions. Temperature stress was imposed by incubation at 4°C or 45°C for 30 min. For pH stress, the liquid medium was replaced by CD + Met medium with a pH of 2.0 or 11.0 and subsequent incubation for 1.5 h. To this end, the pH was adjusted with HCl or NaOH respectively.
Starvation stress was imposed from t = 0 onwards during 2 days of growth. For carbon starvation, solid medium contained 0.2% glucose and liquid medium contained no glucose. For nitrogen starvation solid medium contained nitrate/ammonium (see above), but liquid medium did not contain a nitrogen source.
Inactivation of the Aohex1 gene
The 1.9 kb upstream flanking region of the Aohex1 open reading frame was amplified with primers aB4F and aB1R (Table S1) and using RIB40 genomic DNA as a template. The fragment was inserted into pDONRTM P4-P1R (Invitrogen, http://www.invitrogen.com) by BP recombination reaction generating the 5′ entry clone plasmid, pg5′Ahx1. The 1.6 kb downstream flanking region of Aohex1 was amplified with primers aB2F and aB3R (Table S1) and inserted into pDONRTM P2R-P3 (Invitrogen) by BP recombination reaction generating the 3′ entry clone plasmid, pg3′Ahx1. The 5′ and 3′ entry clones together with the centre entry clone pgEsC (containing the sC marker gene) were subjected with LR clonase in the presence of pDEST R4-R3 (destination vector) to obtain the final plasmid pgΔdAoHex1. The Aohex1 gene deletion fragment (∼ 6.5 kb) was amplified by PCR using the plasmid pgΔdAoHex1 as template and primers aB4F and aB3R. Strain NSRKu70-1-1A (Escaño et al., 2009) was transformed with the deletion fragment as described (Ohneda et al., 2005). Transformants with the sC+ phenotype were selected on M medium containing 1.5% agar and 2% glucose. Disruption of the Aohex1 gene in strain NSRK-ΔHx5 was confirmed by Southern blotting using restriction enzymes BamHI and EcoT221. Plasmid pgEsC containing the sC selection marker gene was introduced into A. oryzae strain NSRKu70-1-1A generating the control strain NSRKu70-1-1AS.
The Aohex1 deletion vector pgDAHx1 was also constructed using the Multisite Gateway cloning system (Invitrogen) (Mabashi et al., 2006). The upstream (2.0 kb) and downstream (2.0 kb) regions of the Aohex1 gene were amplified by PCR using the primer combinations 5-hex1-F and 5-hex1-R, and 3-hex1-F and 3-hex1-R respectively (Table S1). The amplified upstream and downstream regions of Aohex1 were introduced into pDNORP4-P1R and pDONRP2R-P3, respectively, with the Gateway BP clonase reaction. The resulting plasmids were subjected to Gateway LR clonase reaction together with the centre entry clone plasmid containing the A. oryzae adeA gene as a selection marker (Jin et al., 2007) and the destination vector pDESTR4-R3 (Invitrogen). The resulting plasmid pgDAHx1 was used as a template to amplify the deletion cassette by PCR with the primers 5-hex1-F and 3-hex1-R. The amplified deletion fragment was introduced into A. oryzae NSR13 according to Maruyama and Kitamoto (2011). A representative transformant was selected in M medium for adeA prototrophy and named NSR-ΔHx11. Disruption of the Aohex1 gene was confirmed by Southern analysis. For this, genomic DNA was digested with BamHI and NcoI and the 2.0 kb fragment of the Aohex1 upstream region was used as a probe.
Expression of EGFP–AoHex1
The expression vector pUNAGHs (Juvvadi et al., 2007) encompassing the fusion of EGFP and Aohex1 under the control of the amyB promoter was introduced into NSR-ΔHx11 using niaD as a selection marker according to Maruyama and Kitamoto (2011). Strain 11NSR-NAGHs (Table 3) showed fluorescence representative for the transformants obtained.
Expression of sGFP under control of the glaA and aguA promoters
Aspergillus oryzae strains NSRKu70-1-1AS and NSRK-ΔHx5 were co-transformed with pAN52-10S65TGFPn/s (Siedenberg et al., 1999) and pNR10 (Yoon et al., 2010) or with PaguA_sGFP+ (Vinck et al., 2011) and pNR10 as described (Punt and van den Hondel, 1992; de Bekker et al., 2009). These plasmids contain sGFP (S65T) under the regulation of the glaA (pAN52-10S65TGFPn/s) or aguA (PaguA_sGFP+) promoter of A. niger and niaD (pNR10) under the control of the amyB promoter of A. oryzae. Nitrate prototrophic (niaD+) strains were selected on MMS medium (minimal medium pH 6.0, 0.95 M sucrose and 1.5% agar; de Bekker et al., 2009).
The reporter protein EGFP–AoHex1 was monitored over periods of up to 16 h on a Zeiss CLSM (Zeiss LSM 5 PASCAL; Zeiss, http://www.zeiss.com) using a 488 nm laser and a LP505 filter. For live cell imaging, images were taken in the Z-plane in three slices (slice thickness 0.4 μm) every 15 min using a Plan-Neofluor 25×/0.8 Imm Corr objective. The pixel time was 2.51 μs, the laser power 0.125 mW and the pinhole 136 μm. Images were taken with a resolution of 512 × 512 pixels and exported as tif files using the Zeiss LSM Image Browser v4.2 (http://www.zeiss.co.jp). Composition and layout of exported images was made with GIMP v2.6 (http://www.gimp.org/).
Heterogenic expression of sGFP under the control of the glaA and aguA promoters
Aspergillus oryzae was grown as a sandwiched culture (de Bekker et al., 2011a) on minimal medium [0.6% NaNO3, 0.15% KH2PO4, 0.05% KCl, 0.05% MgSO4.7H2O, 0.2 ml l−1 Vishniac (per litre: 10 g of EDTA, 4.4 g of ZnSO4, 1.01 g of MnCl2, 0.32 g of CoCl2, 0.315 g of CuSO4, 0.22 g of ammonium heptamolybdate, 1.47 g of CaCl2 and 1.0 g of FeSO4; Vishniac and Santer, 1957), pH 6.0; de Vries and Visser 1999] containing 3% agar and 200 mM xylose (glaA repressing) or 50 mM glucose (aguA repressing). To this end, 2 μl containing 1000 spores was spotted at the centre of a polycarbonate membrane (76 mm; Profiltra, http://www.profiltra.nl) that was placed on top of the solidified medium. After 24 h a Lumox membrane (20 × 20 mm, manually cut; Greiner Bio-One, http://www.greinerbioone.com) was placed on top of the culture with the hydrophobic side facing the colony. After 42 h of growth, the sandwiched colony was transferred for 6 h to minimal medium plates containing 25 mM maltose (glaA inducing) for strains RB#140.1, RB#140.2, RB#141.3 and RB#141.4. For strains RB#154.3, RB#154.4, RB#156.2 and RB#156.4 the sandwiched colony was transferred for 6 h to minimal medium plates containing 25 mM xylose (aguA inducing). The Lumox membrane was removed and a piece of the polycarbonate membrane (approximately 10 × 10 mm) carrying the colony was cut and placed upside down in a glass bottom microscopy dish (MatTek, P35G-1.5-20-C) on a 20 μl drop of minimal medium. GFP fluorescence was monitored on the Zeiss LSM 5 system equipped with a Plan-Neofluar 16×/0.5 oil immersion objective. GFP was excited with a 488 nm laser and images were captured as a Z-stack of optical slices (pinhole 1–2 airy units; optimal interval 2.02 mm; 4× line average; 8 bit scan depth). Maximum intensity projections of the Z-sections (1024 × 1024 pixels) were used for further analysis. The fluorescence intensity was quantified by measuring the mean pixel value of hyphae using a macro in the KS400 software (Version 3.0; Carl Zeiss Vision, http://www.zeiss.de). Sections of 50–100 μm of leading hyphae were selected by hand and fluorescence was automatically quantified as the sum grey value per hypha with the background value from an equivalent area subtracted (Vinck et al., 2005). Signals were normalized with a custom Python script by dividing single hyphal fluorescence by the total fluorescence of all selected hyphae per image. To examine whether hyphal fluorescence followed a bimodal distribution, the normalized data were log transformed and subsequently modelled using five parameters (P, μ1, σ1, μ2 and σ2) as described (Vinck et al., 2005). The 95% confidence intervals (CI) of the parameters were estimated by bootstrapping (1000 replicates). Customs scripts in the Scilab programming language were used to fit the data (http://web.science.uu.nl/microbiology/images/fung/fittools.zip; http://web.science.uu.nl/microbiology/images/fung/manual%20fittools.pdf).
Analysis of plugging
Compartments were ruptured by laser dissection using the laser pressure catapulting function (LPC) of the P.A.L.M. laser dissecting microscope (Zeiss, http://www.zeiss.com). To this end, 60–70% of the power of the pulsed UV laser was used. Each experiment was carried out in duplo using 20 hyphae in each experiment. anova analysis was used with Bonferroni post hoc correction when multiple comparisons were made between treatments. anova analysis was used with Dunnet's post hoc correction when differences in septal plugging in a strain were assessed between a control condition and stress conditions. In all cases, a difference was assumed significant when P < 0.05.
This work was financed by the Netherlands Organization for Scientific Research (NWO) and by a Grant-in-Aid for Young Scientist from the Japan Society for the Promotion of Science (JSPS).