Previous Article | Next Article ![]()
Eukaryotic Cell, October 2008, p. 1831-1846, Vol. 7, No. 10
1535-9778/08/$08.00+0 doi:10.1128/EC.00130-08
Copyright © 2008, American Society for Microbiology. All Rights Reserved.
,
Institut für Botanik der Westfälischen Wilhelms-Universität Münster, Schloβgarten 3, D-48149 Münster, Germany,1 Mycotoxin Research Group, USDA/ARS, 1815 N. University St., Peoria, Illinois 616042
Received 11 April 2008/ Accepted 4 August 2008
|
|
|---|
areA strains by use of a Fusarium verticillioides microarray representing
9,300 genes. Analysis of the array data revealed sets of genes significantly down- and upregulated in the areA mutant under both N starvation and N-sufficient conditions. Among the downregulated genes are those involved in nitrogen metabolism, e.g., those encoding glutamine synthetase and nitrogen permeases, but also those involved in secondary metabolism. Besides AreA-dependent genes, we found an even larger set of genes responding to N starvation and N-sufficient conditions in an AreA-independent manner. To study the impact of NMR on AreA activity, we examined the expression of several AreA target genes in the wild type and in areA and nmr deletion and overexpression mutants. We show that NMR interacts with AreA as expected but affects gene expression only in early growth stages. This is the first report on genome-wide expression studies examining the influence of AreA on nitrogen-responsive gene expression in a genome-wide manner in filamentous fungi. |
|
|---|
AreA is a member of the GATA-type transcription factor family and a functional homologue of two positively acting nitrogen regulators in Saccharomyces cerevisiae, Gln3 (40) and Gat1 (Nil1) (69). AreA homologues have been isolated from several filamentous fungi and include the Aspergillus nidulans AreA (34), Neurospora crassa NIT2 (21), Penicillium chrysogenum NreA (26), Fusarium oxysporum Fnr1 (17), and Magnaporthe grisea Nut1 (20). Most of our knowledge on the molecular mechanism of AreA-dependent nitrogen regulation derives from studies of A. nidulans and N. crassa, where AreA activity is negatively affected by binding to the nitrogen metabolite regulators NmrA and NMR1, respectively, when rich nitrogen sources are available. The deletion of the nmr gene in either fungus resulted in a significant derepression of AreA target genes under nitrogen-sufficient conditions (2, 51). In contrast, for F. fujikuroi, the NMR homologue was shown to play only a minor role in nitrogen regulation, at least in the regulation of GA biosynthesis genes, despite fully complementing the N. crassa nmr-1 mutant (39). Therefore, we postulate that F. fujikuroi must contain additional regulatory proteins that affect AreA activity and nitrogen metabolite repression of its target genes. These additional proteins may be involved in protein interactions with AreA or in the poorly understood processes that modulate AreA activity via the rapid degradation of the areA transcript in response to intracellular ammonium and glutamine (7, 41). In A. nidulans, AreA accumulates in the nucleus during nitrogen starvation and is exported from the nucleus and complexes with NmrA when rich nitrogen sources are added to starving mycelia (78). In S. cerevisiae, the target of rapamycin (TOR) kinase cascade regulates the cellular response to the nutrient status of the cell. The TOR kinase is active under conditions of nitrogen sufficiency and is inactivated by nitrogen starvation conditions or by the addition of rapamycin. The transcriptional activators Gln3 and Gat1 (Nil1) are phosphorylated in a TOR-dependent manner under nitrogen-sufficient conditions and are restricted to the cytoplasm after complexing with the regulator Ure2. Nitrogen starvation or the addition of rapamycin inhibits TOR, resulting in the dephosphorylation of the transcriptional activators, their release from Ure2, and their translocation to the nucleus, where they activate their target genes (reviewed in references 13 and 60).
Recently, we have shown that the TOR protein kinase in F. fujikuroi plays a role in the regulation of genes involved in GA and bikaverin biosynthesis, ammonium transport, ribosome biogenesis, protein synthesis, and autophagy (76). In contrast to what is seen for the yeast model system S. cerevisiae, the inhibition of TOR by rapamycin resulted in only partial derepression of AreA-regulated nitrogen metabolite repression genes, which suggests that there must be additional factors which inactivate AreA under conditions of nitrogen availability (76).
One strategy to learn more about the role AreA may play as a global nitrogen regulator is to compare the expression profiles of the wild-type (WT) and areA mutant strains after growth in nitrogen-limited and -sufficient conditions. Microarrays are a powerful tool to investigate the genome-wide regulation of genes. The recent availability of genomic data from Fusarium verticillioides, a close relative of F. fujikuroi that is estimated to share
93% nucleotide identity, provides an opportunity to explore gene expression changes in F. fujikuroi by a cross-species microarray analysis. Microarrays have already been successfully used for cross-species hybridization between different fungi. S. cerevisiae microarrays successfully identified differentially expressed genes of Pichia pastoris (63), N. crassa microarrays identified developmentally regulated genes and mating-type-dependent genes of Sordaria macrospora (46, 54), and finally, an A. niger macroarray identified growth-phenotype-related genes of Aspergillus oryzae (73).
The aim of this study was to discover a set of genes in F. fujikuroi which are significantly affected by nitrogen availability by cross-species hybridization using an oligonucleotide microarray derived from F. verticillioides expressed sequence tag (EST) data. By comparing the expression patterns of the WT and the areA mutant under conditions of nitrogen starvation and sufficiency, we were able to differentiate between AreA-dependent and non-AreA-dependent nitrogen-repressed and nitrogen-induced genes. The expression pattern of representative genes from each expression group was confirmed by Northern blot analysis and helped to provide an overview of the role AreA plays in nitrogen metabolite repression. Furthermore, we show that nitrogen-regulated genes can be differentiated into rapamycin (TOR)-dependent and -independent genes, confirming our previous suggestion (76) that TOR controls mainly AreA-independent genes (e.g., ribosome biogenesis and translation control genes) and only partially affects the expression of AreA target genes.
To further examine the role of NMR in nitrogen metabolism in F. fujikuroi, we compared the expressions of the newly identified, larger set of AreA target genes in the WT, nmr deletion, and nmr-overexpressing mutants and in areA deletion and areA-overexpressing strains. Although initial experiments did not indicate a significant effect of either deletion or overexpression of nmr on the expression of AreA target genes, we found that NMR has an inhibitory effect on AreA during early growth phases, as AreA target genes were upregulated in the areA mutant during the first 24 h of growth. We demonstrated by a yeast two-hybrid approach that AreA interacts with NMR, as has been shown previously for N. crassa and A. nidulans (35, 51), and that this interaction probably affects AreA activity mainly in early stages of growth.
|
|
|---|
areA-T19) and overexpression (glnAprom::areA-T19/66) strains were described before (81), as were the nmr deletion and overexpression strains, the
nmr-T8 and glnAprom::nmr-T1 strains (39). For RNA isolation, the F. fujikuroi strains were first cultivated for 2 days in 300-ml Erlenmeyer flasks with 100 ml Darken medium (14) [but with 2.0 g/liter L-glutamine instead of (NH4)2SO4] on a rotary shaker at 190 rpm at 28°C. One milliliter of this culture was then used to inoculate 100 ml of ICI medium containing 1.8 g L-glutamine/liter (23). Growth proceeded for 5 days, after which the mycelia were harvested, washed with deionized water, and added to fresh ICI medium without nitrogen for 4 h to induce starvation conditions. Then, glutamine was added to a final concentration of 9.2 g/liter to one half of the flasks, while the other flasks were kept under starvation conditions for a final 2 h of growth. For inhibition of the TOR kinase, rapamycin was added to a final concentration of 200 ng/ml after 1.5 h, with mycelia being harvested after an additional 30 min.
Bacterial strains and plasmids. Escherichia coli strain Top10 F' (Invitrogen, Groningen, The Netherlands) was used for plasmid propagation. Vectors for the GAL4 yeast two-hybrid analysis were prepared as follows. Vector pAD-nmr, containing the 1.4-kb full-length nmr cDNA, was created by cloning the amplicon generated with primers nmr-AD-for and nmr-AD-rev into the SalI/XbaI-digested pAD-GAL4-2.1 vector (Stratagene, La Jolla, CA). Vector pBD-areA, containing the 2.9-kb full-length areA cDNA, was created by cloning the amplicon generated with primers areA-F3 and areA-RTR1 into the SrfI site of the pBD-GAL4 Cam vector. Vector pBD-areA-ZF was created by cloning a 0.92-kb amplicon generated with primers areA-ZF and areA-RTR1 into the SrfI site of the pBD-GAL4 Cam vector. The 0.92-kb sequence corresponds to amino acids 662 to 968 of AreA.
PCR and RT-PCR. PCR mixtures contained 25 ng of template DNA, 50 ng of each primer, 0.2 mM deoxynucleoside triphosphates, and 1 U of Biotherm polymerase (Genecraft, Lüdinghausen, Germany). Reverse transcription-PCR (RT-PCR) was performed using the Superscript one-step RT-PCR with Platinum Taq kit (Invitrogen, Groningen, The Netherlands), 10 pM of oligo(dT) primer, and 1 µg of total RNA as the template.
Nucleic acid isolation and Northern blot analysis.
Plasmid DNA was extracted using the Qiaprep Miniprep kit (Qiagen, Hilden, Germany) essentially as described by the manufacturer. Total F. fujikuroi RNA was isolated using the RNAgents total RNA isolation kit (Promega, Mannheim, Germany). Formaldehyde gel electrophoresis, transfer to nylon membranes (Hybond N+; GE Healthcare, Little Chalfont, United Kingdom), and subsequent hybridization were accomplished following the protocol of Sambrook et al. (62). 18S and 28S rRNA bands were visualized by illumination with UV light (
= 256 nm) against the background of a silica thin-layer chromatography plate (Schleicher und Schuell, Dassel, Germany). PCR products using the primers pBS2-uni and pBS2-rev (Table 1) and templates derived from a plasmid cDNA library (75) or specifically created PCR products for genes not represented in this library were used as probes. The accession numbers of probe sequences are listed in Table 2.
|
View this table: [in a new window] |
TABLE 1. Oligonucleotides used in this study
|
|
View this table: [in a new window] |
TABLE 2. Genes differentially regulated in the microarray analysis
|
F. fujikuroi mycelia were harvested from two individual cultures and RNA was extracted as described above. Microarray hybridization, data acquisition, and initial analysis were conducted by NimbleGen. Normalized data from the probe sets of the eight arrays (two biological repetitions for each strain and condition) were compared using the Acuity 4.0 microarray analysis software package (Molecular Devices Corp., Sunnyvale, CA) and the Excel macro collection FiRe (22). Only oligonucleotide sets with at least twofold up- or downregulation in both replicate hybridizations compared between strains or treatments were considered to be differentially regulated. The F. verticillioides tentative consensus sequences and EST sequences (DFCI F. verticillioides Gene Index; Dana Farber Cancer Institute; http://compbio.dfci.harvard.edu/tgi/cgi-bin/tgi/gimain.pl?gudb=f_verticill) corresponding to hybridizing oligonucleotide sets were analyzed by BLASTN against the EST database of a plasmid F. fujikuroi cDNA library (75). BLASTN matches to this cDNA library with E values of
1.00 x 10–100 were regarded as homologous. Both F. verticillioides and F. fujikuroi ESTs were examined by BLASTX against the nonredundant protein sequence NCBI database.
DNA sequencing and sequence homology comparison. DNA sequencing of recombinant plasmid clones was accomplished with a Li-Cor 4000 automatic sequencer (MWG, Munich, Germany). Sequences from both DNA strands were generated using universal primers flanking the cloning site and/or specific primers obtained from MWG Biotech (Munich, Germany). DNA and predicted protein sequence alignments were done with DNA Star (Madison, WI). Sequence homology searches were performed using the BLAST program against the NCBI database (1).
Yeast two-hybrid analysis. Yeast two-hybrid analyses were performed following the manufacturer's instructions with the Hybri-ZAP-2.1 two-hybrid kit (Stratagene, La Jolla, CA) by use of the S. cerevisiae SMY3 strain (8) instead of the system's Y190 strain to avoid the reddish color of Y190 colonies included in the kit.
Nucleotide sequence accession numbers. XXX.
|
|
|---|
areA-T19) under nitrogen starvation and nitrogen-abundant growth conditions. To verify that the fungal transcriptional response under these conditions is consistent with what was found by previous work (e.g., in terms of expression pattern of nitrogen-repressed AreA target genes), we performed Northern blot analysis of total RNA from both strains and conditions with known nitrogen-regulated genes as probes. As previously observed, the GA biosynthetic genes cps/ks and P450-4, the bikaverin biosynthetic gene bik3, and the ammonium permease-encoding gene mepA (74) were expressed in the WT only under nitrogen starvation conditions and were barely expressed at all in the
areA mutant (see Fig. S1 in the supplemental material).
Preliminary analysis of the microarray data with Acuity 4.0's self-organizing map function led to the identification of 1,241 probe sets that appeared to be differentially regulated either according to the nitrogen status or in the
areA mutant compared to the WT. Subsequent analysis of this subset with FiRe (22) identified 144 probe sets that differed significantly between the two experimental replicates and were not considered further. Genes for which expression was significantly up- or downregulated in response to nitrogen availability in both the WT and the
areA mutant were defined as AreA-independent nitrogen-regulated genes (Table 2, categories Ia and IIa), whereas genes for which expression was up- or downregulated by nitrogen in an AreA-dependent manner were defined as AreA target genes (Table 2, categories Ib and IIb, c, and d). There were also genes which were differentially expressed in the
areA mutant compared to the WT, independent of nitrogen availability (categories III and IV).
Genes up- or downregulated in response to glutamine. We identified 300 probe sets representing 279 genes whose expression was upregulated (Table 2, category I; Fig. 1A) and 250 probe sets representing 223 genes whose expression was downregulated (Table 2, category II; Fig. 1B) by glutamine addition in both the WT and the areA mutant. Analysis of the biological function of the category I genes by use of the MIPS functional catalogue database (61) found that 58% (161 genes) are directly involved in de novo protein biosynthesis, whereas 36% (100 genes) play a role in ribosome biogenesis (predominantly encoding 40S and 60S ribosomal subunits) and 22% are involved indirectly in protein synthesis, e.g., genes coding for translation initiation and elongation factors (32 genes), general transcription (18 genes), RNA processing (5 genes), protein folding (4 genes), and protein targeting (2 genes) (Fig. 1A). Interestingly, we also identified two cross-pathway control (CPC) genes, cpc2 and cpc1. Cpc2 is a Gβ-like protein composed entirely of WD repeats first identified in N. crassa (42). The gene was shown to be expressed analogously to ribosome biogenesis genes (9, 82). Cpc1 is a bZIP transcription factor that acts as global regulator of amino acid biosynthesis in filamentous fungi (50). Beside these regulators, the induction of 22 amino acid biosynthesis genes, e.g., the argininosuccinate synthase gene arg1, and of 6 nucleotide metabolism genes was observed (Table 2, category I).
![]() View larger version (14K): [in a new window] |
FIG. 1. Functional distribution of nitrogen-regulated genes. (A) Pie chart representing the functions of genes upregulated by nitrogen addition. (B) Pie chart representing the functions of genes repressed by nitrogen addition. The division of genes into several functional categories was accomplished following the MIPS functional database catalogue (http://mips.gsf.de/proj/funcatDB). met., metabolism.
|
-aminobutyrate transaminase catalyzing the formation of succinate semialdehyde and glutamate (57).
Beside genes for the use of alternative nitrogen sources, we found genes with possible roles in the transport of external nitrogen sources or the degradation of nonessential proteins in the vacuole. Among the 14 genes (6%) involved in transport were those encoding all three ammonium transporters described for F. fujikuroi, namely, MepA, MepB, and MepC (74), an amino acid permease (Aap8) (B. Schönig and B. Tudzynski, unpublished data), a putative polypeptide transporter (Mtd1), and putative transporters for the secondary nitrogen sources allantoate and
-aminobutyrate. We also noted genes with possible roles in protein degradation, such as proteasome component genes, genes involved in protein ubiquitination (e.g., that for the polyubiquitin Ubi4), and genes for proteases like the fungal vacuolar protease A (Pep4), the endopeptidase B (Prb1), and carboxypeptidase Y (Prc1), putatively involved in posttranslational modification and activation of multiple vacuolar proteases (32). In addition, we identified 12 genes probably involved in protein localization, e.g., a homologue of VPS15, which plays a role in vacuolar protein sorting in S. cerevisiae (68).
Another group of glutamine-repressed genes (7%) encode signal transduction proteins, including components of the TOR signaling pathway (Fpr1, 14-3-3), the catalytic and regulatory subunits of the cyclic AMP-dependent protein kinase (Pka1), the signaling pathway of alternative carbon source utilization (Snf4 and CreA), and small GTPases (Rho1, Rab7, Rab11), as well as calmodulin (Cmd1) and another calcium binding protein. In addition, nine stress-responsive genes were identified, such as heat shock protein-encoding genes (e.g., the hsp70 gene) and the gene encoding the superoxide dismutase, sod. Finally, the following two glutamine-repressed genes are involved in secondary metabolism: cpr, encoding the NADPH-cytochrome P450 oxidoreductase donating electrons to the P450 enzymes from the GA biosynthetic pathway (37), and bik3, encoding the bikaverin-specific O-methyltransferase (Wiemann and Tudzynski, unpublished). A total of 33% share homology to genes with no known function.
The differential expression of several of these genes was verified in Northern blot experiments using probes derived from cDNA clones from an F. fujikuroi EST library (75) or PCR fragments generated from F. fujikuroi genomic DNA with primers derived from the F. verticillioides EST sequences (Fig. 2A and B; Table 2, categories I and II, respectively). The expression of all tested genes was consistent with the array data.
![]() View larger version (79K): [in a new window] |
FIG. 2. Genes with expression levels affected by nitrogen addition. (A) Northern blot experiments showing genes upregulated by glutamine addition. (B) Northern blot experiments showing genes downregulated by glutamine addition. The WT was grown for 5 days in synthetic ICI medium containing 1.8 g/liter glutamine and then shifted into ICI medium without nitrogen (no N). After 4 h, glutamine was added to one half of the flasks to a final concentration of 9.2 g/liter (Gln), and the mycelia were harvested after 2 h. Probes used for hybridization are listed in Table 2. 28S and 18S rRNA was used as the loading control.
|
![]() View larger version (26K): [in a new window] |
FIG. 3. Most nitrogen-repressed genes are independent of AreA. (A) Venn diagram demonstrating the share of AreA in the regulation of nitrogen metabolite repression. The set of genes downregulated in the areA mutant under nitrogen starvation conditions was compared with the set of genes repressed by the addition of glutamine to the media. "Classic" AreA target genes are genes repressed by nitrogen addition and downregulated in the areA mutant. (B) Comparison of the distribution of AreA-dependent and -independent genes into the functional categories shown in Fig. 1. Gray bars show genes repressed by nitrogen addition and downregulated in the areA mutant; white bars show genes repressed by nitrogen in an AreA-independent manner. The values for the gray bars and the white bars add up to the total number of nitrogen-repressed genes. met., metabolism.
|
![]() View larger version (62K): [in a new window] |
FIG. 4. Expression pattern of nitrogen-regulated and AreA-dependent genes. (A) Nitrogen-repressed genes not affected by the areA deletion. (B) Genes upregulated in the areA mutant under starvation conditions. (C) Genes downregulated in the areA mutant. The categories (Cat.) are as shown in Table 2. Strains and culture conditions were the same as described for Fig. 1. Probes used for hybridization are listed in Table 2. 28S and 18S rRNA was used as the loading control.
|
As mentioned above, the majority of nitrogen-repressed genes (161 of 223 genes) are AreA independent (Table 2, category IIa). These completely AreA-independent functional categories of genes include those with putative functions in protein degradation, modification, and localization, stress response, vitamin metabolism, and the utilization of alternative carbon sources and homeostasis (Fig. 3B).
Deletion of areA leads to the upregulation of amino acid biosynthetic genes.
We also identified a set of 108 genes whose expression is upregulated by the deletion of areA under conditions of either nitrogen starvation or nitrogen abundance (Table 2, category IV). To this group belong amino acid biosynthetic genes, e.g., asn2, encoding asparagine synthetase, arg1, encoding argininosuccinate synthase, several amino acid aminotransferases as well as Cpc1 and its putative coregulator Mbf (a homologue of the yeast multiprotein bridging factor Mbf1 [70]), and also the pyridoxine biosynthesis gene pdx1, a homologue of snz1 from S. cerevisiae, which has been reported to be under CPC in N. crassa, S. cerevisiae, and Candida albicans (77). In addition, this group also includes 11 tRNA synthetase-encoding genes for the synthesis of isoleucyl-, methionyl-, prolyl-, arginyl-, glutaminyl-, phenylalanyl-, threonyl-, aspartyl-, valyl-, and asparagyl-tRNAs and one gene encoding a DEAD/DEAH box helicase from the group of protein synthesis genes. Furthermore, 12 genes encoding enzymes involved in carbohydrate and energy metabolism are also upregulated in the
areA mutant; these genes include those for malate synthase and the glyoxylate cycle enzyme isocitrate lyase, which is known to link C and N metabolisms; the gene encoding the homologue of the Podospora anserina autophagy-specific transcription factor IDI4 (15); and the gdhA gene, encoding the NADP+-dependent glutamate dehydrogenase. The S. cerevisiae homologue gdh1 has also been reported to be under Gcn4 control (59). The expression pattern of an arbitrarily chosen subset of genes was confirmed by Northern blot analyses (Fig. 4B and data not shown).
Response of AreA target genes to rapamycin. Recently, we have shown that the F. fujikuroi TOR kinase is involved in the regulation of genes responding to changes in the nitrogen status of the cell, though to a much lesser extent than in yeast. Among the genes partially derepressed by the addition of rapamycin were some AreA target genes (e.g., GA biosynthetic genes and the GS-encoding gene glnA). In contrast, the expression levels of several genes involved in processes of de novo protein synthesis, such as ribosome biogenesis and translation initiation, were significantly reduced by rapamycin (76). In order to further explore the role of the TOR kinase in F. fujikuroi, we transferred mycelia grown under standard conditions for 3 days into medium with or without rapamycin (200 ng/ml) and compared the expression levels of genes from all four categories (Table 2) under both conditions (see Fig. S2 in the supplemental material).
We found that both nitrogen-induced (category I) and -repressed (category II) genes responded to rapamycin addition. Among the newly identified genes which are partially regulated by rapamycin are some of the typical AreA target genes, e.g., the xanthine dehydrogenase-encoding gene xdh1 and the peptide transporter-encoding gene mtd1, but also several nitrogen-regulated AreA-independent genes, e.g., the glutamate decarboxylase gene gad1. The most pronounced effect observed for rapamycin was for the set of genes which are involved in protein synthesis and which were shown to be induced by glutamine in an AreA-independent manner. These genes (e.g., eIF4A and several 60S and 40S ribosomal subunit genes) are significantly reduced by rapamycin-mediated inhibition of TOR (data not shown; also see Fig. S2 in the supplemental material), confirming previous results (76).
The role of NMR in AreA-dependent nitrogen regulation. In A. nidulans and N. crassa, the deletion of nmrA and nmr-1, respectively, led to a significant derepression of AreA target genes (2, 30). In F. fujikuroi, the deletion of the nmr gene resulted in higher sensitivity against chlorate (39) and higher resistance toward the TOR inhibitor rapamycin (76), suggesting that NMR is involved in nitrogen regulation in some way. Likewise, the ability of the F. fujikuroi nmr gene to complement an N. crassa nmr-1 mutant indicates that the protein shares significant functional similarity to the N. crassa Nmr1 protein. However, the AreA-targeted genes involved in GA biosynthesis were not derepressed in the nmr mutant (39).
To further analyze the relationship between AreA and NMR in F. fujikuroi, we compared the growth behavior of the WT with that of the areA or nmr deletion mutants and strains overexpressing areA or nmr on plates with glutamine as the preferred nitrogen source and with or without chlorate. Under nitrogen starvation conditions, AreA activates the expression of the nitrate reductase-encoding gene niaD (39). This enzyme catalyzes not only the reduction of nitrate to nitrite but also that of chlorate to chlorite, a toxic and growth-inhibiting metabolite. In mutants with a defect in nitrogen metabolite repression, the toxic effect of chlorate should be stronger, whereas mutants with a defect in AreA-mediated target gene activation should be more chlorate resistant than the WT. We found that the areA overexpression mutant, the glnAprom::areA mutant, shows a strong growth inhibition similar to that of the nmr deletion mutant, indicating a derepression of the AreA target gene, niaD. The overexpression of nmr (glnAprom::nmr strain) and the deletion of areA resulted in higher chlorate resistance, presumably caused by the repression of the nitrate reductase gene (Fig. 5). These results indicate that nitrogen metabolite repression could indeed be regulated by balanced expression levels of the two antagonists AreA and NMR, as proposed for A. nidulans (84).
![]() View larger version (80K): [in a new window] |
FIG. 5. Nitrate reductase activity is affected in areA and nmr mutant strains. All strains were grown in media containing 9.2 g/liter L-glutamine as the nitrogen source with (+) or without (–) potassium chlorate (10 mM). Pictures were taken after 2 (medium without chlorate) or 4 (chlorate medium) days of incubation at 28°C. OE::areA, areA overexpression mutant (areA transcription regulated by the glnA promoter); OE::nmr, nmr overexpression mutant (nmr regulated by the glnA promoter).
|
![]() View larger version (40K): [in a new window] |
FIG. 6. Yeast two-hybrid experiment showing that NMR interacts with the C terminus of AreA. Transformants were dropped in 10-fold dilution steps as indicated on SD medium without tryptophan and leucine (SD –W –L) and on SD without tryptophan, leucine, and histidine and with the addition of 50 mM 3-amino triazole (SD –W –L –H + 3-AT). The decreasing concentration is indicated by a triangle. The positive control was transformation with pBD-WT and pAD-WT (Stratagene). AD, transformation with pAD-GAL4-2.1 (Stratagene); AD-Nmr, transformation with pAD-Nmr, containing the full-length nmr cDNA; BD, transformation with pBD-GAL4 Cam (Stratagene); BD-AreA, transformation with pBD-AreA, containing the full-length areA cDNA; BD-AreA-ZF, transformation with pBD-AreA-ZF, containing the C-terminal ZF domain of AreA.
|
The WT,
areA, glnAprom::areA (areA overexpression [OE::areA]),
nmr, and glnAprom::nmr (OE::nmr) strains were cultivated as described in Materials and Methods, and the washed mycelia were transferred into media without nitrogen, with glutamine, or with nitrate. Nitrate (10 mM) was chosen as the alternative nitrogen source, as nitrate can be used only by strains with an active AreA due to the AreA-dependent expression of the niaD gene under starvation conditions (81).
We observed almost no effect of nmr deletion or overexpression on the expression of the secondary metabolism genes bik3 and cps/ks or on that of the gene cpc1, which is responsible for the regulation of amino acid biosynthesis genes. However, we did find that the transcript levels of glnA and meaB, both encoding proteins with a likely regulatory role in nitrogen metabolism, were significantly increased in the
nmr mutant incubated with glutamine (Fig. 7).
![]() View larger version (116K): [in a new window] |
FIG. 7. Expression of AreA target genes in areA and nmr mutant strains. Abbreviations for strains are the same as used in Fig. 5. Strains were incubated as described for Fig. 2. Sodium nitrate (NO3) was added to a final concentration of 10 mM. Probes used for hybridization are listed in Table 2 except the gene coding for the nitrate reductase, niaD (CAA62232
[GenBank]
), and the gene coding for the bifunctional ent-copalyl diphosphate/ent-kaurene synthase, cps/ks (Q9UVY5).
|
nmr and downregulated in the OE::nmr mutant with nitrate, while the niaD expression levels are almost identical under starvation conditions (Fig. 7). Interestingly, the expression of niaD depends on an active AreA but does not need nitrate as an inducing substrate, in contrast to what is seen for other fungi (Fig. 7). To determine if there is any effect of the nmr mutation on AreA target gene expression at an early growth stage, we performed a time course experiment. For this, the WT and mutant strains were cultivated in ICI medium with glutamine (0.9 g/liter), and the cultures were harvested after 24 and 48 h of growth instead of after 5 days. Northern analysis revealed that several AreA target genes (e.g., the GA biosynthetic genes ggs2 and cps/ks and the nitrogen metabolism genes mtd1, mepB, and glnA, encoding a peptide transporter, an ammonium transporter, and the GS, respectively) were already expressed after 24 h of incubation in the nmr mutant when nitrogen was still available (Fig. 8). After 48 h, these differences in expression levels were no longer evident, indicating that nitrogen sources were depleted and nitrogen repression was relieved.
![]() View larger version (104K): [in a new window] |
FIG. 8. The partial deregulation of AreA target genes in the nmr mutant is time dependent. areA and nmr mutant strains (Fig. 5) were grown for 24 h and 48 h, respectively, in synthetic ICI medium containing 0.9 g/liter L-glutamine as the nitrogen source. Probes used for hybridization are listed in Table 2, except the gene coding for the geranylgeranyldiphosphate synthase, ggs2 (CAA75568
[GenBank]
), and the gene coding for the bifunctional ent-copalyl diphosphate/ent-kaurene synthase, cps/ks (Q9UVY5).
|
|
|
|---|
areA mutant (see also Fig. 5), indicating that the uptake of glutamine by either the general amino acid permease or the glutamine-specific permease is not affected by the deletion of areA. In contrast to the A. nidulans
areA mutant, the F. fujikuroi
areA mutant is not able to grow on ammonium due the strict dependency of all three ammonium permease genes upon AreA (74).
Using this approach, we were able to delineate subsets of genes that were up- or downregulated in response to glutamine as a preferential nitrogen source or in the
areA mutant compared with what was seen for the WT. The differential expression of a representative subset of the genes has been confirmed by Northern blot analysis under the conditions used for the arrays.
Response of the cell to nitrogen-abundant conditions. Predominant among the group of genes upregulated in response to the addition of glutamine were genes involved in de novo protein synthesis: 61% of upregulated genes are involved in translation, ribosome biogenesis, amino acid and nucleotide biosynthesis, general transcription, RNA processing, and protein folding and targeting (Fig. 1A). Similar results were obtained for S. cerevisiae, where 67% of the genes highly expressed in response to nitrogen availability are also involved in amino acid and nucleotide metabolism, aminoacyl-tRNA synthesis, and protein biosynthesis (64). Homologues of several of these genes are also part of the data set of genes upregulated by glutamine in our study (e.g., the amino acid and nucleotide biosynthetic genes ade3, ade4, ade5,7, lys2, sam1, and trp5 and 34 protein biosynthesis genes which include ribosomal subunit genes and translation initiation factors) (64).
Many of these genes are well-known target genes of a complex transcriptional system, referred to as the CPC network, which enables yeast cells and filamentous fungi to respond to amino acid starvation (28, 43, 50). In general, starvation for a single amino acid results in specific uncharged tRNAs activating the sensor kinase Cpc3 followed by activation of the central transcriptional activator Cpc1, which induces genes involved in the biosynthesis of amino acids, purines, or aminoacyl-tRNA synthetases among others (reviewed in reference 33). In N. crassa, 16 out of 20 tRNA synthetases were identified as Cpc1 target genes by a microarray approach (77). A similar set of genes was also identified in our studies under glutamine sufficiency conditions and also in the
areA mutant, suggesting that an excess of a single amino acid also leads to the activation of the CPC network and that this network is repressed by AreA in F. fujikuroi.
An external amino acid imbalance in S. cerevisiae has been shown to upregulate a homologous group of genes regulated primarily by the functional homologue of Cpc1, the transcription factor Gcn4 (44). In contrast to what we found, glutamine did not induce the expression of Gcn4 target genes. A recent microarray study using 20 different nitrogen sources revealed that only growth on leucine, isoleucine, methionine, threonine, tyrosine, or tryptophan as a sole nitrogen source led to the upregulation of the Gcn4 target genes in yeast, whereas the expression of these genes is unaffected during growth on glutamine as well as on asparagine, serine, ammonium, alanine, arginine, and glutamate as nitrogen sources (24). In addition, Gcn4 was identified in this study as a nitrogen catabolite repression (NCR) target, with its expression decreased under nitrogen-abundant conditions. A putative cross talk between CPC and NCR in S. cerevisiae remains ambiguous, although there are indications for a role for Gcn4 in NCR by repressing Gln3 and for the repression of Gcn4 translation under nitrogen starvation conditions (25, 67). This is in clear contradiction to our finding that cpc1 is upregulated by glutamine addition and in the areA mutant (Fig. 4B). For filamentous fungi, nothing is known about the coordination of nitrogen metabolite repression and CPC. The increase in cpc1 expression could be explained by the fact that the addition of glutamine as the sole nitrogen source necessitates the synthesis of the other amino acids required for protein synthesis during cell growth. The unexpected increase in cpc1 expression in the F. fujikuroi areA mutant might be due to imbalances in the amino acid pool of the cell, as the downregulation of AreA target genes indirectly involved in amino acid metabolism could lead to a lack of single amino acids. The CPC in F. fujikuroi and its coherence with nitrogen metabolite repression are under investigation at the moment (Schönig and Tudzynski, unpublished).
Response of the cell to nitrogen starvation conditions. The biological functions of genes downregulated by glutamine addition are much more diverse than those of the upregulated genes (Fig. 1B). The most prominent subset of the 223 downregulated genes have functions in carbohydrate metabolism and energy generation (e.g., the pentose phosphate cycle, glycolysis, and the tricarboxylic acid cycle). For example, the expression of gdhA, encoding the NADP+-depending glutamate dehydrogenase, is 2.5-fold reduced in response to glutamine. This enzyme plays a key role in providing cells with glutamate. Other important groups include genes involved in protein degradation, modification, and localization (14%). Under conditions of nitrogen starvation, an important internal nitrogen resource is the bulk turnover of nonessential proteins in the vacuole and their subsequent degradation by vacuolar proteases (71). Accordingly, we identified homologues of the yeast proteases Pep4, Prb1, and Prc1 to be upregulated under conditions of nitrogen starvation.
Another protein degradation pathway is the turnover of proteins in the 26S proteasome, located in the nucleus and in the cytosol. It is elicited by environmental conditions like nutrient limitation, heat stress, and exposure to heavy metals (31) and is involved in the controlled degradation of key proteins of cell cycle control, signaling pathways, and general metabolism, where transcriptional regulators are degraded by the 26S proteasome (27). Various components of the ubiquitin/26S proteasome pathway, such as the polyubiquitin Ubi4, ubiquitin-conjugating and -activating enzymes, and catalytic and regulatory proteasome components, were shown to be upregulated under nitrogen starvation conditions in our study.
Nine genes encode proteins probably involved in stress response, like a hsp70 homologous gene and a gene coding for a manganese superoxide dismutase, sod. Recently, an interesting link between nitrogen regulation and regulation of the reactive oxygen species (ROS) scavenger machinery has been uncovered for the strawberry pathogen Colletotrichum acutatum. Both ROS and nitrogen deficiency play important roles in the initial infection process. Consequently, a coordinated increase of proteins involved in nitrogen transport and utilization and protein degradation on the one hand and the generation and elimination of ROS on the other hand has been found by proteome analysis (29). This could also be the case for the rice pathogen F. fujikuroi.
The group of transporter-encoding genes represents 6% of the genes that were downregulated under glutamine-sufficient conditions. Permeases play an important role in the utilization of different nitrogen sources. In plant pathogenic fungi, several genes of this group have been described as pathogenicity-related genes. The conditions on the plant surface are similar to those of nitrogen starvation, as the fungus has to penetrate and overcome plant defenses before getting access to the plant's carbon and nitrogen resources (11, 66, 72). In addition to nitrogen permease genes, other genes have been shown to be highly induced at the beginning of a plant infection in different fungi. In F. oxysporum, the genes encoding the amino acid permease Gap1 and the peptide transporter Mtd1 are induced (16). In M. grisea, genes encoding the vacuolar serine protease Spm1, an orthologue of S. cerevisiae Prb1, and the neutral trehalase 1, Nth1, are induced (18, 19). Homologues of these genes were also upregulated by nitrogen starvation conditions in our study. For some but not all phytopathogenic fungi, a role of AreA homologues in plant-fungus interaction has been shown, e.g., for Colletotrichum lindemuthianum, F. oxysporum, and M. grisea (16, 20, 52) but not for Cladosporium fulvum (53). The precise nature of the role AreA may play in pathogenesis in F. fujikuroi will require further study.
AreA regulates only a subset of nitrogen-repressed genes. One of the most striking findings was that AreA controls only about 20% of nitrogen-repressed genes (Fig. 3A). Most of these AreA target genes whose products belong to the functional groups of permeases, catabolic enzymes needed to degrade poor nitrogen sources and secondary metabolism (Fig. 3B). One reason for the relatively low number of "classic" AreA target genes that we observed might be that genes involved in the utilization and metabolization of alternative nitrogen sources need the inducing substrate in addition to an active AreA to be expressed. Thus, the prn cluster genes of A. nidulans involved in proline utilization may be induced only when no preferred carbon (glucose) or nitrogen (ammonium or glutamine) sources are present in the medium and when proline is available (58).
In contrast to these AreA target genes, the genes of other functional groups, e.g., those involved in protein modification, degradation, and localization, are not affected by the areA deletion. In S. cerevisiae, the vacuolar proteinase A (Pep4), vacuolar endoproteinase B (Prb1), vacuolar aminopeptidase I (Lap4), and vacuolar carboxypeptidase S (Cps1) proteases are controlled by the activators Gln3 and Nil1 in response to nitrogen starvation, whereas vacuolar carboxypeptidase Y (Prc1) is not affected by the deletion of the GATA factors or the nitrogen status of the cell (10). Homologues of Prb1, Pep4, and Prc1 were also part of the group of nitrogen-repressed genes in our study, although the expression of all of these genes does not depend on the GATA factor AreA, in contrast to their Gln3 dependence in S. cerevisiae.
Previous work has shown that the biosyntheses of the secondary metabolites GA and bikaverin are repressed by nitrogen in F. fujikuroi. The expression pattern of one gene in our microarray study supported this work. We found that the bikaverin biosynthesis gene bik3, encoding a putative O-methyltransferase with homology to AflJ from A. flavus, is repressed by glutamine in an AreA-dependent manner, although the expression of bikaverin biosynthetic genes is affected by AreA to a lesser extent than are the GA genes. Recently, the F. oxysporum homologue of bik3 has been identified as nitrogen starvation induced (16). The conservation of the bikaverin gene cluster between F. fujikuroi, F. verticillioides, and F. oxysporum (Wiemann and Tudzynski, unpublished) appears to mirror a conserved strategy of regulation by nitrogen status. We also found the expression of numerous other genes with predicted functions in secondary metabolism (e.g., those encoding oxidoreductases, transporters of the major facilitator superfamily, and monooxygenases) to be repressed in a similar manner.
In S. cerevisiae, two positively acting GATA factors, Gln3 and Gat1, are known to regulate the transcription of NCR-sensitive genes either alone or in combination with each other (69). By use of DNA microarrays, the expression profiles of a WT strain and of a gln3
gat1
double mutant strain grown under different nitrogen conditions have been examined. Of the 392 genes whose expression was dependent on the quality of the nitrogen source, only 91 genes were activated in a GATA factor-dependent manner (64), which is consistent with our findings. No Gat1 homologue is known to exist in filamentous fungi, indicating that the functions of Gln3 and Gat1 converge on AreA homologues in filamentous fungi.
The role of NMR in nitrogen regulation of F. fujikuroi. Previously, we have shown that NMR does not play a major role in regulating AreA activity in F. fujikuroi, at least with respect to GA biosynthetic genes. In contrast, NmrA and Nmr1 play essential roles in regulating AreA activity in A. nidulans and N. crassa, respectively. The ability of the F. fujikuroi NMR to fully complement the N. crassa nmr1 mutant (39) indicates that they share similar modes of action in both fungi. In this study, we wanted to examine the impact of the NMR protein on the expression of a larger set of genes and its possible interactions with AreA. We show that the F. fujikuroi NMR protein interacts with the C-terminal part of AreA (which includes the ZF domain) (Fig. 6), similarly to what is seen for N. crassa and A. nidulans (35, 51). We also found that the overexpression of NMR inhibits the expression of the AreA target gene niaD and is reflected by an increased chlorate resistance (Fig. 5), as already shown in our previous work (39). Consistently, the areA overexpression strains are more sensitive to chlorate, while the areA deletion strains are more resistant (Fig. 5). Taken together, these results point toward a direct cooperation of these two regulatory proteins in nitrogen metabolite regulation in F. fujikuroi as well.
However, in contrast to the clear differences in chlorate resistance between areA and nmr deletion mutants, the effect of nmr deletion or overexpression on AreA target gene expression is less obvious. For most of the AreA target genes, we did not observe a significant upregulation in the nmr deletion mutant in shift experiments. If NMR acted as the main inhibitor of AreA activity in F. fujikuroi in a manner similar to that seen for homologues NmrA and NMR1 in A. nidulans and N. crassa, the expression of AreA target genes would be expected to be clearly upregulated in
nmr mutants and downregulated in nmr-overexpressing mutants. This was not the case, confirming our previous results (39), which had shown that the expression of the GA biosynthetic genes is not affected by the nmr deletion. Thus, NMR seems to play a minor role in the regulation not only of GA biosynthesis genes but also of "classic" AreA target genes, such as the xanthine dehydrogenase gene xdh1, the nitrate reductase gene niaD, and the amino acid permease gene aap8 (Fig. 7). The minor role of the nmr mutation in the expression of niaD is in contrast to the higher sensitivity of the
nmr mutant and the higher resistance of the glnAprom::nmr mutant against chlorate (Fig. 5).
The clearest effect of nmr deletion and overexpression has been shown for the glnA and meaB genes, both of which were significantly upregulated with glutamine, in contrast to what was seen for the "classic" AreA target genes (Fig. 7), supporting our suggestion that the GS might play a regulatory role (75).
To examine if NMR exerted an effect on the expression of AreA target genes in a time-dependent manner, the WT and areA and nmr deletion and overexpression strains were grown in medium containing a limited amount of glutamine as the nitrogen source and harvested after only 24 h and 48 h of growth rather than after 5 days. Northern analysis showed that the expression of a number of AreA target genes was derepressed after 24 h in the
nmr mutant compared to what was seen for the WT. The difference in expression levels between the WT and the nmr mutant disappeared as soon as the nitrogen source was exhausted (48 h postinoculation), as indicated by the upregulation of nitrogen source transporters (e.g., the peptide transporter Mtd1) (Fig. 8). Thus, the F. fujikuroi NMR seems to negatively affect AreA activity under nitrogen-repressing growth conditions, which is fully consistent with the situation for A. nidulans and N. crassa (35, 51). However, the extents to which AreA target genes are derepressed in an nmr deletion mutant might differ for different fungi.
On the other hand, the immediate and efficient downregulation of AreA target genes after the addition of a preferred nitrogen source to nitrogen-starved WT and nmr mutant mycelia indicates that in addition to NMR, another regulator(s) which switches off the expression of AreA target genes under nitrogen-sufficient conditions must exist. The nature of this repressor and its supposed interaction with AreA will need to be clarified in the future. One putative additional repressor of AreA target genes may be the MeaB bZIP transcription factor, as its deletion leads to a derepression of AreA target genes in A. nidulans (55, 84) as well as in F. fujikuroi (A. Schmeinck, D. Wagner, and B. Tudzynski, unpublished data). In addition, we recently cloned the gene encoding the GATA-type transcription factor AreB, the orthologue of the S. cerevisiae repressor Dal80 (49) and the A. nidulans AreB (12). Its role in nitrogen regulation and as a putative counterpart of AreA in F. fujikuroi is currently under investigation (P. Rengers and B. Tudzynski, unpublished data).
An interesting finding that distinguishes F. fujikuroi from A. nidulans is that the meaB gene is downregulated in the areA mutant and that the meaB transcript size depends on nitrogen availability in the WT (the transcript size is significantly smaller under nitrogen starvation conditions) and on AreA. In contrast, the expression of the A. nidulans meaB gene is almost constitutive, and no changes in transcript size were found under various nitrogen conditions. Furthermore, it has been shown that MeaB in A. nidulans induces the expression of nmrA under nitrogen-sufficient conditions, thereby affecting the activity of AreA (84). In F. fujikuroi, meaB itself appears to be an AreA target gene in terms of its effects on general expression level and transcript size (Fig. 4C). While the loss of AreA leads to the exclusive presence of the larger transcript, the mutation of nmr resulted in an increased expression of the smaller transcript, indicating that AreA and NMR act opposingly on the expression of the small transcript of meaB. Analysis of F. verticillioides and F. fujikuroi EST collections revealed that there exist two groups of EST clones with alternative transcription start points. The exact mechanism by which AreA and NMR are involved in the regulation of the two alternative transcripts remains to be elucidated.
In summary, we show the utility of the F. verticillioides microarray to detect global gene expression changes in F. fujikuroi. We were able to identify numerous new candidate AreA target genes and found that the loss of this transcription factor resulted not only in a dramatic downregulation of "classic" AreA target genes but also in the upregulation of a new set of target genes that mainly encode genes involved in amino acid biosynthesis. Studies with rapamycin indicated that, in contrast to what is seen for S. cerevisiae, only some AreA target genes, e.g., the GS-encoding gene glnA and the peptide permease-encoding gene mtd1, are derepressed by rapamycin, and this only to some extent, whereas the global control of protein de novo synthesis functions in a TOR-dependent and AreA-independent manner in both organisms. We were able to show that NMR is a repressor of AreA target gene expression in a growth-phase-dependent manner and that it interacts directly with AreA. Our finding that the deletion of nmr did not prevent the strong repression of AreA target genes by glutamine suggests that additional regulators, e.g., MeaB and AreB, might play also a significant role in the regulation of AreA activity in F. fujikuroi.
The project was funded by the DFG (Tu101/7).
Mention of trade names or commercial products in this paper is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture.
Published ahead of print on 8 August 2008. ![]()
Supplemental material for this article may be found at http://ec.asm.org/. ![]()
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»