The effect of natamycin on the transcriptome of conidia of Aspergillus niger.

The impact of natamycin on Aspergillus niger was analysed during the first 8 h of germination of conidia. Polarisation, germ tube formation, and mitosis were inhibited in the presence of 3 and 10 μM of the anti-fungal compound, while at 10 μM also isotropic growth was affected. Natamycin did not have an effect on the decrease of microviscosity during germination and the concomitant reduction in mannitol and trehalose levels. However, it did abolish the increase of intracellular levels of glycerol and glucose during the 8 h period of germination.Natamycin hardly affected the changes that occur in the RNA profile during the first 2 h of germination. During this time period, genes related to transcription, protein synthesis, energy and cell cycle and DNA processing were particularly up-regulated. Differential expression of 280 and 2586 genes was observed when 8 h old germlings were compared with conidia that had been exposed to 3 μM and 10 μM natamycin, respectively. For instance, genes involved in ergosterol biosynthesis were down-regulated. On the other hand, genes involved in endocytosis and the metabolism of compatible solutes, and genes encoding protective proteins were up-regulated in natamycin treated conidia.

INTRODUCTION

Conidia are stress-resistant dispersal vehicles that are produced by many fungal species. Fungi belonging to the genera Aspergillus and Penicillium produce large numbers of airborne conidia. These conidia easily contaminate and colonise food, which explains why Aspergillus and Penicillium are among the most important food-spoiling organisms. Preservatives as sorbic acid and natamycin (Plumridge , Stark 2007) prevent fungal growth in or on a food source. There are clear indications that dormant conidia are more resistant to antifungal compounds than growing hyphae. Dormant conidia of Aspergillus fumigatus survive concentrations of 50 μg/mL of the polyene antibiotic amphotericin B, but become sensitive to 20 and 1–2 μg/mL of the antifungal after 2 and 4 h of germination, respectively (Russel et al. 1975, 1977). Similarly, conidia of A. niger and Penicillium discolor survive a treatment with 45 μM of the polyene antibiotic natamycin, which equals ten times the minimal inhibitory concentration for germinating conidia. Notably, conidia start to germinate upon removal of the antibiotic (van Leeuwen ).

It is the aim of this study to evaluate the cellular mechanisms that explain these variations in antifungal sensitivity. Novel insights may lead to new prevention strategies of fungal contamination in agriculture and the food industry. As a model system the antifungal compound natamycin that is used in the food industry (Stark 2007) is used. In contrast to other polyene antifungals, natamycin does not induce membrane permeability (Te Welscher , van Leeuwen ). It does inhibit endocytosis in germinating conidia of P. discolor in a time and dose dependent manner (van Leeuwen ). Moreover, natamycin interferes with vacuole fusion in yeast cells as well as filamentous fungi (Te Welscher ). Very recent work has shown that natamycin also reversibly inhibits transport of different nutrient molecules into the cell (Te Welscher ).

In order to study the changes that occur in conidia that are challenged with antifungal compounds, the transcriptome of conidia of Aspergillus niger was studied in the presence of natamycin and compared with data of untreated germinating conidia. Recently, RNA profiles of dormant and germinating conidia of A. niger were reported (van Leeuwen ). It was shown that the RNA composition of dormant conidia was most distinct when compared to conidia that had been germinating for 2, 4, 6, and 8 h. Dormant conidia contain high numbers of transcripts of genes involved in formation of protecting components such as trehalose, mannitol, heat shock proteins and catalase. Transcripts of the functional gene classes protein synthesis, cell cycle and DNA processing and respiration were over-represented in the up-regulated genes after 2 h of germination, whereas metabolism and cell cycle and DNA processing were over-represented in the up-regulated genes after 4 h of germination. No functional gene classes were over- or under-represented in the differentially expressed genes after 6 and 8 h of germination. From these data it was concluded that the RNA profile of conidia changes especially during the first 2 h of germination and that this coincides with protein synthesis and respiration.

We here show that 3 and 10 μM natamycin hardly affect the transcriptome during the first 2 h of germination, but it inhibits further stages of germination as judged by several cellular parameters. The transcriptome after 8 h was less affected when spores were kept in 3 μM natamycin compared to those treated in a concentration of 10 μM. For instance, genes involved in endocytosis, and genes involved in protection of conidia were up-regulated. On the other hand, genes involved in ergosterol biosynthesis were down-regulated.

MATERIALS AND METHODS

Organism and growth conditions

The A. niger strain N402 (Bos ) and its derivative RB#9.5 were used in this study. The latter strain expresses a gene encoding a fusion of sGFP and the histone protein H2B under regulation of the mpdA promoter (R. Bleichrodt, unpubl. results). For spore isolation, strains were grown for 12 d at 25 °C on complete medium (CM) containing per liter: 1.5 % agar, 6.0 g NaNO3, 1.5 g KH2PO4, 0.5 g KCl, 0.5 g MgSO4, 4.5 g D-glucose, 0.5 % casamino acids, 1 % yeast extract and 200 μl trace elements (containing per liter: 10 g EDTA, 4.4 g ZnSO4·7H2O, 1.0 g MnCl2·4H2O, 0.32 g CoCl2·6H2O, 0.32 g CuSO4·5H2O, 0.22 g (NH4)6Mo7O24·4H2O, 1.5 g CaCl2·2H2O, and 1.0 g FeSO4·7H2O). Conidia were harvested in ice-cold ACES-buffer (10 mM ACES, 0.02 % Tween-80, pH 6.8), filtered through sterile glass wool, washed in ice-cold ACES-buffer and resuspended in CM (van Leeuwen ). The conidia were kept on melting ice until further processing on the same day. An aliquot of 3.109 conidia were added to 300 ml CM in 500 ml Erlenmeyers. Cultures were shaken at 125 rpm in the absence or presence of 3 or 10 μM natamycin. Stock solutions of natamycin (10 mM) were freshly made in 85 % DMSO (Brik 1981).

Transcriptome analysis

Data analysis was performed on biological triplicates, each based on three cultures. At each time point, 15 ml of each of the three cultures was pooled. The (germinating) conidia were pelleted at 1100 g at 5 °C for 5 min and immediately frozen in liquid nitrogen. RNA extraction, cDNA labeling, microarray hybridisation and data analysis were done as described (van Leeuwen ). The array data has been deposited in NCBI’s Gene Expression Omnibus (Edgar ) and is accessible through GEO Series accession number GSE36440 (www.ncbi.nlm.nih.gov/geo/).

HPLC analysis

Dormant, germinating or treated conidia (5·107 - 1·108) were frozen at -80 °C and homogenised with a Qiagen Tissuelyser® (2 min at 30 strokes /sec: Qiagen, Venlo, The Netherlands) using a stainless steel grinding jar cooled with liquid nitrogen. After an additional round of grinding with 1 ml milliQ, the samples were thawed and quickly transferred to a 2 ml Eppendorf tube. Samples were centrifuged at 4 °C for 30 min at 20.817 g. The supernatant was stored at -80 °C until analysis. Prior to HPLC analysis samples were filtered through an Acrodisc® 0.2 μm PTFE syringe filter (Sigma-Aldrich, Zwijndrecht, The Netherlands). A volume of 10 μl was subjected to HPLC analysis, using a Waters 717 plus autosampler equipped with a 515 HPLC pump with control module II (Waters Corporation, Etten-Leur, The Netherlands). The mobile phase consisting of 0.1 mM Ca EDTA in water was maintained at a flow rate of 0.5 ml/min. The Sugar Pak I Ca+ cation-exchange column was kept at 65 °C with a Waters WAT380040 column heater module (Laborgerätebörse GmbH, Burladingen, Germany). Sugars and polyols were detected with an IR 2414 refractive index detector (Waters Corporation, Etten-Leur, The Netherlands). As standards, trehalose, mannitol, D-(+)-glucose, glycerol, erythritol and arabitol were used (Sigma-Aldrich, Zwijndrecht, The Netherlands). Peak integrations and quantitative calculations were performed with the Waters Empower software (Waters Corporation, Etten-Leur, The Netherlands).

ESR spectroscopy

Germinating conidia were centrifuged at 8000 rpm for 2 min. The supernatant was discarded and the conidia were resuspended in 25 μl perdeuterated TEMPONE-potassium ferricyanide solution (1 mM and 120 mM, respectively). Micro-viscosity was determined and calculated as described in (van Leeuwen ).

Fluorescence microscopy

Samples of liquid cultures were placed on poly-l-lysine (Sigma) coated cover slips (van Leeuwen ). The medium was removed and the cover slips with the immobilised conidia were placed upside-down onto an object glass with a < 0.5 mm layer of 2 % water agar. Images were taken with a Zeiss Axioskop 2 plus microscope (Zeiss, Oberkochen, Germany) equipped with a HBO 100 W mercury lamp and a AxioCam MRc (Zeiss, Germany) camera using standard FITC (λ = 450–490 nm, FT510, LP520) filters.

RESULTS

Morphological responses to natamycin during conidial germination

Light microscopy showed that germination of A. niger conidia is inhibited in natamycin-treated conidia compared to untreated cells (Fig. 1A). Untreated conidia swell slightly during the first 2 h of germination. The surface area of the cells on the micrographs increased from 17 to 22.6 μm2 (Fig. 1B). The conidia enlarged gradually to 46 μm2 between 2- and 6 h and their volume further increased up to 8 h. At this stage, the variability in size of the cells was largely due to differences in germ tube emergence and growth. After 6 and 8 h, 10 % and 80 % of the conidia had started to form germ tubes, respectively (Fig. 1C).

Fig. 1.

Effect of natamycin on germination of A. niger conidia. Morphology (A), increase in surface area (B) and percentage of conidia forming germ tubes (C) in the absence or presence of 3 μM and 10 μM natamycin. Bar represents 10 μm.

Conidia that had been exposed to 3 μM natamycin showed a similar swelling as control cells during the first 2 h. The surface area of the cells on the micrographs increased from 17.7 to 22.1 μm2. Between 2- and 6 h, the surface area of the conidia enlarged to 35.6 μm2, which had further increased to 40.1 μm2 after 8 h of germination. Notably, polarisation and germ tube formation were not observed during the 8 h incubation time (Fig. 1C). The surface area of conidia that had been exposed to 10 μM natamycin for 2 h increased from 17.3 to 21.8 μm2 (Fig. 1B). After 3 h the conidia had reached a surface area of 23.4 μm2, which remained unchanged up to 8 h of incubation. Polarised cells and germ tubes were not formed throughout culturing (Fig. 1C). All considering, these results show that polarisation and germ tube formation are inhibited at 3 and 10 μM natamycin, while at 10 μM also isotropic growth is inhibited.

An A. niger reporter strain expressing a fusion of the H2B histone protein and the sGFP protein under control of the mpdA promoter was used to monitor nuclear division (Fig. 2). Dormant conidia of this RB#9.5 strain were, with 85 % (n = 86), binucleate (van Leeuwen ). Nuclear division was shown to occur prior to the formation of germ tubes namely between 6 and 8 h of germination. Conidia did not show mitosis after 8 h of incubation when exposed to 3 or 10 μM natamycin.

Fig. 2.

Number of nuclei in conidia of the A. niger strain RB#9.5 in the absence or presence of 3 μM and 10 μM natamycin as visualised with fluorescence microscopy. Bar represents 10 μm.

Intracellular viscosity and compatible solutes during germination

The dormancy of fungal spores has been correlated to the viscosity of the interior of the cell (Dijksterhuis ). Viscosity within A. niger conidia was analysed by electron spin resonance (ESR) using the spin probe perdeuterated TEMPONE (PDT). The spectra included a narrow and a broad component. The narrow component represents the cytoplasmic signal and was detectable after subtraction of the broad component originating from cell wall located PDT (Dijksterhuis , van Leeuwen ). The narrow spectrum showed a central line (h), flanked by a low-field line (h), high-field line (h) and an m-line (h). The latter represents melanin, which has its own paramagnetic properties (Fig. 3A, Dijksterhuis ). Central and high-field components were used to calculate rotational correlation time and the viscosity (η) based on the Stokes-Einstein equation. The effective viscosity of dormant conidia ranged between 3.14 and 3.26 cP. After 2 h of germination, viscosity decreased with 20–30% irrespective of the presence of natamycin (Fig. 3B). As germination proceeded, no further change in viscosity was observed in all cases. This was irrespective of the presence of natamycin. These results show that natamycin does not affect the decrease in cytoplasmic viscosity as observed during germination of conidia of A. niger.

Fig. 3.

Viscosity (A, B) and glucose, trehalose, glycerol and mannitol levels (C) in germinating A. niger conidia and in conidia treated with 3 μM and 10 μM natamycin. In (A), h and h represent the low-field and high-field lines of the electron spin resonance (ESR) signals, respectively, which are used to calculate micro-viscosity (B). h is the ESR signal from melanin which is present in the conidial cell wall.

A dormant conidium of A. niger contains on average 2.1 ± 0.4 and 1.2 ± 0.5 pg of the compatible solutes mannitol and trehalose, respectively (Fig. 3C). These values drop to ≤ 0.5 pg within the first 2 h of germination. In contrast, the level of glycerol increased in this time span from almost undetectable to 1.0 ± 0.0 pg per cell. Levels of glycerol remained unchanged until 6 h of germination, but had increased to 1.5 ± 0.1 pg after 8 h. The level of glucose slowly increased to approximately 0.3 ± 0.3 pg during the first 6 h of germination, after which it increased to 0.9 ± 0.1 pg per cell. Natamycin did not affect the degradation of compatible solutes as well as the appearance of glycerol in the cytoplasm during the first 2 h. After 8 h, however, no further increase in the glycerol level had occurred in natamycin-treated cells. Instead, glycerol levels decreased between 2 and 8 h in conidia treated with 10 μM natamycin. Furthermore, no increase in glucose levels was observed in the presence of natamycin after 8 h. Between 4 and 8 h mannitol increased to 1.0 ± 0.5 pg per cell when conidia were exposed to 10 μM natamycin. Taken together, natamycin decreases glycerol and glucose levels in conidia after degradation of mannitol and trehalose, In addition, mannitol levels stay higher and in 10 μM natamycin increase after 8 h.

Transcriptional profiling and comparison of gene expression

RNA from conidia that had been treated for 2 or 8 h with 3 or 10 μM natamycin was hybridised to whole genome microarrays. MAS5.0 detection calls showed that the number of genes with a present call was invariably higher upon treatment with natamycin as compared with the controls (Fig. 4). Untreated conidia showed a marked decrease in the number of expressed genes after 2 h of germination (i.e. from 4626 to 3557; see also van Leeuwen ). The decrease was much less in the presence of 3 μM natamycin (i.e. from 4626 to 4285), whereas at 10 μM natamycin the number of expressed genes had even increased (i.e. from 4626 to 4783) (Fig. 4). The number of genes that lost their transcripts during the first 2 h of germination dropped from 1986 (control) to 1519 (3 μM) and 1235 (10 μM), while the number of genes whose expression was activated increased (917, 1178 and 1392 genes, respectively). The conidia treated with 10 μM natamycin had 34 % more expressed genes than the controls after 2h of germination. A similar difference was also observed after 8 h. The number of transcripts reached 6210 (this is 43 % of all ORFs identified in A. niger) in the case of treatment with 10 μM natamycin and 5802 and 4913 in the case of conidia treated with 3 μM or untreated cells, respectively. The increase in expressed genes was significant in the case of moderately to highly expressed genes (Fig. 5).

Fig. 4.

Overview of the global changes in the transcriptome of conidia that had either or not been treated with 3 μM or 10 μM natamycin. Inside the conidia the number of expressed transcripts is given. Green and red numbers represent genes with an absent to present call and a present to absent call between two stages, respectively.

Fig. 5.

The number of expressed genes with hybridisation values between 100 and 1000 in treated conidia and in controls. The triplicates were tested by means of ANOVA and different lettering indicates a significant difference (p < 0.05).

Correlation of the RNA profiles showed that dormant conidia were most different compared to the other samples. Conidia treated for 8 h with 10 μM natamycin showed the strongest resemblance to dormant conidia (Fig. 6). The correlation between the profiles at t = 2 h and t = 8 h was 0.76 for the control, 0.61 for the samples treated with 3 μM natamycin and 0.57 for 10 μM natamycin. All considering, these data indicate that RNA profiles of natamycin-treated conidia change to a higher extent when compared to non-treated cells.

Fig. 6.

Correlation of the RNA profiles of dormant or germinating conidia and conidia which are kept in natamycin for 2 and 8 h.

Differential gene expression in natamycin-treated cells.

The number of genes that was more than 2-fold up-regulated within the first 2 h ranged between 947 and 1152 in the absence or presence of natamycin (Fig. 7). The number of down-regulated genes ranged between 1343 and 1968. When the profiles at t = 2 h were compared, 1 and 9 genes were ≥ 2-fold down- and up-regulated, respectively. Apparently, the changes that occur in the RNA profile during the first 2 h of germination are hardly affected by natamycin. Indeed, the correlation of the profiles at t = 2 h was between 0.87 and 0.9 (Fig. 6). In all cases, transcripts belonging to the functional categories protein synthesis, energy and cell cycle and DNA processing were over-represented in the up-regulated genes at t = 2 h (Table 1). The functional gene class cell cycle and DNA processing was over-represented in the up-regulated genes and the functional gene class transcription was over-represented in the down-regulated genes when the profiles of t = 2 and t = 8 h were compared (Table 2). This was irrespective of natamycin treatment. In conidia treated with 10 μM natamycin, the categories C-compound and carbohydrate utilisation and lipid and fatty acid breakdown were overrepresented in the up-regulated genes. At 8 h, 280 genes were ≥ 2-fold up- or down-regulated (i.e. 173 and 103, respectively) when germinating controls were compared to the non-germinating conidia that had been exposed to 3 μM natamycin for 8 h (Fig. 7). Changes were clearly more pronounced between the controls and conidia treated with 10 μM natamycin. In this case, 1713 and 873 genes up- and down-regulated, respectively. Indeed, the correlation in the RNA profile was higher between the control and 3 μM natamycin at t = 8 h than between the control and 10 μM natamycin (i.e. 0.8 and 0.71, respectively; Fig. 6). The fact that the 10 μM natamycin sample at t = 8 h is more different from the control than the 3 μM sample is also reflected in a Fisher exact test (Table 3).

Fig. 7.

Overview of the global changes in the transcriptome of conidia that had either or not been treated with 3 μM or 10 μM natamycin. The number of differentially expressed genes is indicated.

Table 1.

Over- (E) and under- (S) representation of functional gene classes in the pool of genes that were up- and down-regulated in conidia of A. niger that had been incubated for 2 h in medium with or without 3 or 10 μM natamycin.

Table 2.

Over- (E) and under- (S) representation of functional gene classes in the pool of genes that were up- and down-regulated in conidia of A. niger that had been incubated for 2 or 8 h in medium with or without 3 or 10 μM natamycin.

Table 3.

Over- (E) and under- (S) representation of functional gene classes in the pool of genes that were up- and down-regulated in conidia of A. niger that had been incubated for 8 h in medium with or without 3 or 10 μM natamycin.

Specific transcriptional changes associated to conidial germination

In the following paragraphs expression of selected groups of genes in conidia that had been incubated in medium with or without 10 μM natamycin will be discussed. The tables also show the values for dormant conidia and conidia treated for 8 h with 3 μM natamycin.

Ergosterol and desaturated fatty acids

Natamycin specifically binds to ergosterol. Ergosterol is formed from acetyl CoA, which involves 22 enzymes in S. cerevisae (Onyewu , Da Silva Ferreira , Mysyakina & Funtikova, 2007). Fourteen out of 24 genes with homology to ergosterol biosynthesis genes showed ≥ 2-fold lower expression in conidia that had been incubated in the presence of 10 μM natamycin when compared to the control (Table 4). The most severe down-regulation was observed for genes with similarity to HMG-CoA synthase (erg13, HMGS, An02g06320), erg1 (An01g03350), erg3 (An16g02930, An15g00150), erg5 (An01g02810), a gene similar to squalene monooxygenase (erg1, An03g03770), erg25 (An03g06410), and a lanosterol 14 alpha-demethylase like gene (erg11, An11g02230). These genes were 6.5-fold to 40 times down-regulated. In contrast, the HMG-CoA reductase (hmg1, An04g00610, Basson ) was 10-fold up-regulated in the presence of natamycin.

Table 4.

Transcript levels of genes involved in synthesis of ergosterol and desaturated fatty acids in dormant conidia and conidia that were incubated for 8 h in medium in the absence or presence of natamycin. The normalised average values of three independent experiments are given. White to black shading indicate expression levels from absent (12 units of expression) to > 2500 expression units. The value of gene expression is significantly differentially expressed (≥ 2-fold) compared to the 8 h old germling if the outline of the box is dashed. SS = strong similarity; S = similarity. Calb = Candida albicans; Gfuj = Gibberella Fujikuroi; Ncra = Neurospora crassa; Pita = Penicillium albicans; Scer = Saccharomyces cerevisiae; Spom = Schizosaccharomyces pombe.

Δ9-stearic acid desaturases (Wilson ) and Δ12-oleic acid desaturases (Calvo , Chang ) are important for the generation of desaturated fatty acids. As such, they influence the amount of (poly)unsaturated fatty acids in membranes. Transcripts of four desaturases (An07g01960, An12g09940, An08g05160 and An14g06980) are strongly down regulated in conidia exposed to natamycin. For instance, transcripts of the enzyme odeA (An08g05160) were down-regulated 25.8-fold.

Vesicle trafficking

Ergosterol is involved in fusion and fission events in fungal cells (Jin ) including endocytosis (Heese-Peck , van Leeuwen ) and vacuole fusion (te Welscher ). Earlier work (van Leeuwen , te Welscher ) has provided evidence that natamycin interferes with vesicle trafficking and fusion events in fungal cells. Twenty six out of 77 genes involved in vesicle recognition and fusion, endocytosis and vesicle secretion were ≥ 2-fold up-regulated in conidia that had been incubated for 8 h in 10 μM natamycin (Table 5). For instance, a gene (An12g07570) similar to synaptobrevin SNC2, a protein involved in vesicle recognition, is over 3.7-fold higher expressed in natamycin. Up-regulation was also observed for genes encoding a FTI1-like protein (An01g00170) and the endosomal protein SNF7 (An18g05430 and An04g05670, Weiβ ) that showed 2.1-, 4-, and 3.7-fold higher transcript levels. The gene encoding a VPS33-like protein (An02g05380, Subramanian ) that is active during both endosome and vacuole fusion, is 6.6-fold higher expressed in the presence of the anti-fungal compound. Genes involved in vesicle secretion and endocytosis were also up-regulated in natamycin exposed conidia including members of the actin-organising arp2/3 complex that is involved in vesicle uptake. Moreover, the Sec15p homologue An15g00010 that is predicted to be involved in exocytosis (Köhli ) was upregulated 4.0-fold.

Table 5.

Transcript levels of genes involved in trafficking, fission and fusion of vesicles in dormant conidia and conidia that had been incubated for 8 h in medium in the absence or presence of natamycin. The normalised average values of three independent experiments are given. White to black shading indicate expression levels from absent (12 units of expression) to > 2200 expression units. The value of gene expression is differentially expressed (≥ 2-fold) compared to the 8 h old germling if the outline of the box is dashed. SS = strong similarity; S = similarity; Hsap = Homo sapiens.

Membrane rafts (Martin & Konopka 2004, Malinska ) are small stabilised domains of the plasma membrane that contain ergosterol and capture specific landmark or transport proteins like Pma1. The A. niger gene An09g05950 has similarity to this protein and is 10-fold upregulated in the presence of natamycin. The gene encoding a Sur7-like protein (An07g6530) is 27 times up-regulated. Sur7 can be found in membrane rafts, but also in eisosomes. Eisosomes are large protein complexes underlying the plasma membrane that co-localise with sites of endocytosis (Walther , Fröhlich , Loibl ). A central protein in this complex is PilA, which is also observed in the plasma membrane of A. nidulans conidia (Vangelatos ). Dormant conidia of A. niger contain transcripts of genes with similarity to PilA (i.e. An11g0175 and An07g08890). In the presence of natamycin these genes are 8.9-fold and 5.4-fold up-regulated, respectively. Walther et al. (2006) reported a network of interactions of eisosome components with known endocytic effectors. Five genes of A. niger with similarity to proteins of this network, i.e. RVS 161 (An17g01970), RVS 167 (An09g04300), Sla2 (An11g10320), Pan1 (An13g00290) and ABP1 (An03g06960) were 2.1- to 3.7-fold up-regulated in the presence of natamycin.

Compatible solutes

Compatible solutes accumulate in conidia to protect proteins and membranes during drought and other stressors. Most of the trehalose-synthesising and degrading enzymes are expressed in natamycin-treated conidia. Gene An07g08720, which has strong similarity to trehalose-phosphate synthase and the acid trehalase encoding gene (An01g01540) were 4.7- and 3-fold up-regulated respectively, in the treated conidia (Table 6). Mannitol-synthesising and degrading enzymes (see also Ruijter , Aguilar-Osorio ) were also up-regulated in the presence of the anti-fungal (5.4-fold for mpdA, An02g05830 and 4.8-fold for mtdA, An15g05450). Moreover, a gene with similarity to a mannitol transporter (An02g06710) was 42 times up-regulated in the presence of natamycin.

Table 6.

Expression of genes involved in the synthesis of trehalose and mannitol in dormant conidia and conidia that had been incubated for 8 h in medium in the absence or presence of natamycin. The normalised average values of three independent experiments are given. White to black shading indicate expression levels from absent (12 units of expression) to > 4100 expression units. The value of gene expression is differentially expressed (≥ 2-fold) compared to the 8 h old germling if the outline of the box is dashed. SS = strong similarity. Anid = Aspergillus nidulans; Anig = Aspergillus niger; Smut = Streptococcus mutans.

Glyoxylate cycle

Several genes that encode proteins predicted to be involved in fermentation, gluconeogenesis and glyoxylate cycle show strong up-regulation in the presence of natamycin (Table 7). This included an alcohol dehydrogenase (An13g00950, 39-fold), D-lactate dehydrogenase (An11g09520, 12.4-fold); pyruvate decarboxylate (An 09g01030, 11.3-fold); isocitrate lyase (An01g09270, 81-fold) and a malate synthase gene (An15g01860, 52-fold). A gene with similarity to 2-methylisocitrate lyase (An12g07630) that could have a role in fatty acid oxidation (Upton & McKinney 2007) was 2.6-fold upregulated.

Table 7.

Expression of genes involved in glycolysis, fermentation and gluconeogenesis in dormant conidia and conidia that had been incubated for 8 h in medium in the absence or presence of natamycin. The normalised average values of three independent experiments are given. White to black shading indicate expression levels from absent (12 units of expression) to > 1700 expression units. The value of gene expression is differentially expressed (≥ 2-fold) compared to the 8 h old germling if the outline of the box is dashed. SS = strong similarity. Klac = Kluyveromyces lactus.

Heat shock proteins

The expression of a number of genes involved in cell protection are shown in Table 8. Some of the genes show strong up-regulation in the presence of natamycin. For instance, a gene with similarity to the protective LEA proteins (An02g07350, Browne , Chakrabortee ) was 16.1-fold up-regulated. Similarly, genes encoding dehydrin-like proteins (An13g01110 and An14g05070, Wong Sak Hoi ) and a small heat shock protein (hsp9p; An06p01610) were 22-, 101- and 14.6-fold up-regulated, respectively. Other up-regulated genes included putative catalases (An08g08920 and An01g01830, 14.4- and 19.5-fold), a gene predicted to be involved in glutathione synthesis (An09g06270, 7.2-fold) and a gene similar to a glutathione transferase (An16g06100, 47-fold). A number of genes predicted to encode chaperonins were significantly down-regulated. For example, An16g09260 predicted to encode a Dnak-type chaperonine was 8-fold down-regulated. The other down-regulated genes are similar to hsp10, hsp60, hsp70 and hsp78.

Table 8.

Transcript levels of genes involved in cell protection in dormant conidia and conidia that had been incubated for 8 h in medium in the absence or presence of natamycin. The normalised average values of three independent experiments are given. White to black shading indicate expression levels from absent (12 units of expression) to > 6700 expression units. The value of gene expression is differentially expressed (≥ 2-fold) compared to the 8 h old germling if the outline of the box is dashed. SS = strong similarity; S = similarity; WS = weak similarity. Ncra = Neurospora crassa; Zmay = Zea mays.

DISCUSSION

In this study the impact of natamycin on germination of conidia of A niger was analysed. In the absence of natamycin, conidia swell, initiate polarised growth and undergo one round of mitosis when they are incubated in medium for an 8 h period. Conidia were unable to initiate polarised growth in the presence of 3 μM natamycin, whereas 10 μM natamycin even blocked isotropic swelling. In addition, mitosis did not occur at both concentrations of the anti-fungal. Earlier studies have shown that conidia of Penicillium and Aspergillus are not killed by natamycin. They survived a period of 20 h in 45 μM natamycin (van Leeuwen ) and initiated germination upon removal of the compound. A similar response is observed in conidia of Penicillium paneum that are exposed to the self-inhibitor 1-octen-3-ol (Chitarra ). This component prevents germination of conidia at high densities, the so-called crowding phenomenon. It was shown that 1-octen-3-ol has a clear effect on the proteome of conidia (Chitarra et al. 2004, 2005).

Natamycin did not affect germination of conidia during the first 2 h of the process. Degradation of compatible solutes, the decrease in viscosity and swelling were similar to control conidia. Moreover, natamycin hardly affected the transcriptome during the first 2 h of incubation. The functional gene classes energy, protein synthesis and transcription were overrepresented in the up-regulated genes irrespective of the presence of the polyene antibiotic. It has been shown that ergosterol cannot be observed in the plasma membrane of P. discolor (van Leeuwen ) during early stages of germination. Absence of ergosterol would explain why we could not find an effect of natamycin during the first stages of germination of A. niger conidia.

Natamycin did affect the transcriptome of conidia after an 8 h exposure. This was most notable at 10 μM natamycin of the anti-fungal. Several genes involved in biosynthesis of ergosterol were down-regulated upon exposure to 10 μM natamycin. In fungi, sterols are asymmetrically distributed and can be found in membranes at sites of cytokinesis and polarised growth (Wachtler 2003, Martin & Konopka 2004). The decrease in expression of ergosterol biosynthesis genes after polyene treatment is also observed in the case of Saccharomyces cerevisiea (Zhang ) and Candida albicans (Liu ). This suggests that natamycin and other polyene antibiotic not only exert their effect by binding to ergosterol but also by reducing the concentration of the sterol in the cell. These effects would impact the formation of an ergosterol cap at the site of polarised growth, as observed in the fungal species P. discolor, A. niger, Fusarium oxysporum and Verticillium fungicola (van Leeuwen et al. 2008, 2010). This would explain why formation of germ tubes is abolished upon natamycin exposure.

Recently, it has been shown that natamycin also blocks growth of yeast and fungi via inhibition of amino acid and glucose transport across the plasma membrane (te Welscher ). In agreement, an up-regulation of transport proteins is observed when conidia are exposed to natamycin. This may be a strategy to try to counteract this effect of natamycin. Some of the most extremely up-regulated genes are An06g02270 (similar to an arabinose transport protein, 168-fold); An03g02190 (similarity to the sugar transporter Sut 1, 136-fold) and An13g00840 (similarity to amino acid protein GAP1, 132-fold).

Genes encoding proteins related to eisosomes were also over-expressed in natamycin-exposed conidia. Eisosomes are structures that are present in Aspergillus conidia (Vangelatos ) and that are associated with endocytosis and putative membrane rafts. Up-regulation of the genes related to eisosomes may be a way of the conidium to counteract the inhibition of endocytosis by natamycin (van Leeuwen ). Genes involved in the biosynthesis of protecting compounds and genes encoding protective proteins were also up-regulated in natamycin exposed conidia. For instance, one gene involved in trehalose biosynthesis and two genes in mannitol biosynthesis and degradation were up-regulated in conidia that had been treated for 8 h with 10 μM natamycin. Trehalose levels did not increase in natamycin-treated spores when compared to the control. This indicates that the compatible solute is used for energy generation (D’Enfert & Fontaine 1997) as a result of the activity of the acid trehalase, which showed up-regulation in 8 h-treated cells. In contrast, the level of mannitol inside treated cells had increased after 8 h, which correlates to a more marked upregulation of genes involved in mannitol metabolism compared to trehalose biosynthesis. Furthermore, genes encoding LEA-like proteins, dehydrins (Wong Sak Hoi ), Hsp9 (Sales ), catalase and a glutathion synthesising enzyme were up-regulated. These data indicate that a stress response is activated in natamycin-exposed conidia. This may also explain the up-regulation of genes of the glyoxylate cycle in natamycin-treated cells. The glyoxylate cycle is an important shunt of the citric acid cycle. It is involved in fatty acid and acetate metabolism and has a role in gluconeogenesis (Eastmond & Graham 2001). Conidia of Aspergillus fumigatus that are stressed due to exposure to neutrophils also show up-regulation of catalase, glutathione and glyoxylate cycle enzymes (Sugui ).

All considering, this study shows that natamycin does not have an impact on conidia during the first stages of germination. However, longer exposure to natamycin shifts the transcriptome to a state of survival with some similarities to the dormant conidium. The conidia respond to the presence of the anti-fungal compound by activating genes that are involved in stress resistance.