DNA affinity purification sequencing and transcriptional profiling reveal new aspects of nitrogen regulation in a filamentous fungus

Significance

Microorganisms have evolved transcriptional networks to prioritize utilization of available nutrient sources. For filamentous fungi, such as Neurospora crassa, this entails distinguishing between a variety of organic and inorganic nitrogen sources. Here, we transcriptionally profiled the response of N. crassa to a variety of nitrogen sources and used DNA affinity purification sequencing to characterize the role of regulatory genes and their direct downstream targets. We identified a transcription factor responsible for regulating genes involved in amino acid and mannose metabolism. By comparing the genes regulated by transcription factors that regulate specific nitrogen utilization pathways and transcription factors that regulate utilization of all nitrogen sources that require metabolic processing before utilization, we revealed aspects of the nitrogen regulatory network.

Abstract

Sensing available nutrients and efficiently utilizing them is a challenge common to all organisms. The model filamentous fungus Neurospora crassa is capable of utilizing a variety of inorganic and organic nitrogen sources. Nitrogen utilization in N. crassa is regulated by a network of pathway-specific transcription factors that activate genes necessary to utilize specific nitrogen sources in combination with nitrogen catabolite repression regulatory proteins. We identified an uncharacterized pathway-specific transcription factor, amn-1, that is required for utilization of the nonpreferred nitrogen sources proline, branched-chain amino acids, and aromatic amino acids. AMN-1 also plays a role in regulating genes involved in responding to the simple sugar mannose, suggesting an integration of nitrogen and carbon metabolism. The utilization of nonpreferred nitrogen sources, which require metabolic processing before being used as a nitrogen source, is also regulated by the nitrogen catabolite regulator NIT-2. Using RNA sequencing combined with DNA affinity purification sequencing, we performed a survey of the role of NIT-2 and the pathway-specific transcription factors NIT-4 and AMN-1 in directly regulating genes involved in nitrogen utilization. Although previous studies suggested promoter binding by both a pathway-specific transcription factor and NIT-2 may be necessary for activation of nitrogen-responsive genes, our data show that pathway-specific transcription factors regulate genes involved in the catabolism of specific nitrogen sources, while NIT-2 regulates genes involved in utilization of all nonpreferred nitrogen sources, such as nitrogen transporters. Together, these transcription factors form a nutrient sensing network that allows N. crassa cells to regulate nitrogen utilization.

Sensing available nutrients and efficiently utilizing them is a challenge common to all organisms. In fungi, accurate nutrient sensing is important in the establishment of fungal colonies and in continued, rapid fungal growth for the exploitation of environmental resources. During fungal pathogenesis, mutations in nutrient sensing and utilization pathways cause reduced virulence in plant and human pathogenic fungi (1–3). Saprophytic and plant pathogenic fungi harvest their nutrients from plants, in which carbon is abundant. However, nitrogen is often a limiting nutrient source for fungal cells (4).

Fungi are capable of scavenging nitrogen from a variety of sources, including both inorganic and organic nitrogen sources. Utilization of different nitrogen sources requires the activation of a number of different transporters and metabolic enzymes. The fungal cell regulates the production of these proteins to ensure that they are only produced when necessary for nitrogen utilization. The nitrogen sources glutamine, glutamate, and ammonium are preferred in filamentous fungi and are utilized first in a mixture of nitrogen sources (5). Fungi are also capable of utilizing nitrate, nitrite, purines, amides, most amino acids, and proteins as nitrogen sources. Uptake and catabolism of these secondary nitrogen sources is highly regulated and requires the transcription of specific permeases and catabolic enzymes. The repression of these genes when preferred nitrogen sources are present is known as nitrogen catabolite repression (6).

Early work in the model filamentous fungus Neurospora crassa identified two genes that are required for nitrogen catabolite repression: nit-2 (NCU09068) and nmr (NCU04158) (Fig. 1). The nit-2 gene encodes a transcription factor that is required for the expression of genes involved in the utilization of nitrate and a number of amino and nucleic acids (7, 8). NMR represses the activity of NIT-2 when preferred nitrogen sources are available (9–11). Much of the work exploring regulation of this pathway focused on the nitrate reductase gene nit-3 (NCU05298). Activation of nit-3 transcription requires two transcription factors: NIT-2 and NIT-4 (NCU08294) (12) (Fig. 1). NIT-4 is a pathway-specific transcription factor that specifically activates genes necessary for nitrate utilization (13). In the presence of nitrate and absence of any preferred nitrogen sources, NIT-4 and NIT-2 promote expression of nit-3 (12). When nitrate is present in combination with a preferred nitrogen source, such as ammonium, NMR represses the activity of NIT-2, and nit-3 is not expressed (14).

Fig. 1.

Regulation of nit-3 expression in N. crassa. Both the transcription factors NIT-2 and NIT-4 are required for nit-3 expression. In the presence of nitrate, NIT-4 promotes nit-3 expression. NIT-2 promotes nit-3 expression in the absence of preferred nitrogen sources such as ammonium. In the presence of preferred nitrogen sources, NMR represses the activity of NIT-2, and nit-3 is not expressed (6).

Utilization of nonpreferred organic nitrogen sources such as amino and nucleic acids is also thought to be regulated by NIT-2 in concert with pathway-specific transcription factors that activate genes necessary to utilize specific nitrogen sources (6). In N. crassa, only a small number of these pathway-specific transcription factors have been identified, such as the transcription factor required for purine utilization, pco-1 (NCU07669) (15). Although NIT-2 works in concert with these pathway-specific transcription factors to promote utilization of nonpreferred nitrogen sources, we do not understand how these transcription factors regulate genes required for nitrogen utilization on a global scale.

In nature, N. crassa utilizes dead plant material, in which carbon and nitrogen sources occur in the matrix of the plant cell wall (16). We hypothesized that the integration of carbon and nitrogen metabolism via nutrient sensing networks must occur for optimal growth on these substrates. By combining RNA sequencing (RNAseq) data from 12 different nitrogen sources with transcriptional profiling of N. crassa on 40 different carbon sources (17), we identified and characterized a pathway-specific transcription factor, amn-1 (NCU00445), that is required for the utilization of a number of amino acids, as well as for expression of genes in response to the simple sugar mannose. DNA affinity purification sequencing (DAPseq) (18) of AMN-1, NIT-2, and NIT-4 enabled us to identify the direct targets of these transcription factors. Although previous studies suggested promoter binding by both a pathway-specific transcription factor and NIT-2 may be necessary for activation of nitrogen responsive genes (12, 19), our data suggest a model in which pathway-specific transcription factors activate genes that encode enzymes required for utilization of a specific nitrogen source, while NIT-2 regulates genes that are required for general nitrogen metabolism, such as nitrogen transporters.

Results

The Zinc Binuclear Cluster Transcription Factor NCU00445 Regulates Genes Involved in Mannan and Mannose Utilization.

To identify genes involved in nutrient sensing in filamentous fungi, we exposed N. crassa to a diverse set of carbon sources and measured the transcriptome using RNAseq (17, 20). Transcription of the conserved zinc binuclear cluster transcription factor NCU00445 was more than fourfold higher during exposure to 2 mM mannose than 2% sucrose and less strongly induced on a number of other hemicellulosic components (17). RNAseq data from wild-type cells exposed to mannose showed that the expression of 91 genes was at least fourfold higher than during exposure to carbon starvation (SI Appendix, Fig. S1A and Dataset S1) (17). If NCU00445 is a positive regulator of the mannose response, we would expect the expression of these genes to be reduced in cells lacking NCU00445 as compared to wild-type cells during exposure to mannose. Indeed, the expression of 83 of these 91 genes was at least fourfold down-regulated in ΔNCU00445 cells as compared to wild-type (SI Appendix, Fig. S1 and Dataset S1).

The transcriptional response of cells lacking NCU00445 during exposure to mannose led us to hypothesize that NCU00445 might play a role in mannose utilization. Mannose-6-phosphate isomerase is the only enzyme necessary for mannose utilization (21, 22). The N. crassa genome contains two mannose-6-phosphate isomerase genes, man-2 (NCU02322) and man-3 (NCU07165). The expression of man-2 was reduced sixfold in ΔNCU00445 cells as compared to wild-type during exposure to mannose; the expression of man-3 was essentially unchanged (SI Appendix, Fig. S2A). When wild-type and ΔNCU00445 cells were inoculated into media containing mannose as the carbon source, the growth of ΔNCU00445 cells was indistinguishable from that of wild-type cells, indicating that NCU00445 was not required for mannose utilization (SI Appendix, Fig. S2 B and C and Dataset S2).

Mannose is a building block of the complex carbohydrate mannan, which is a component of the fungal cell wall and hemicellulose in the plant cell wall. Although the N. crassa genome contains enzymes necessary for mannan degradation, N. crassa is unable to grow on mannan alone (21, 23). The ability to grow on mannan is enabled by constitutive expression of the transcription factor, CLR-2 (NCU08042), which binds the promoter and regulates the expression of an extracellular endo-β-1,4-mannanase gh5-7 (NCU08412) and predicted β-mannosidase gh2-1 (NCU00890) (17, 21, 24). The expression of both gh5-7 and gh2-1 was more than threefold lower in ΔNCU00445 cells than wild-type under mannose conditions. However, unlike with clr-2, increasing the expression of NCU00445 by 2.4-fold by placing it under the control of the gpd-1 (NCU01528) promoter did not allow N. crassa cells to grow on mannan (SI Appendix, Fig. S2 C and D and Dataset S2).

Deletion of NCU00445 caused the expression of 335 genes to be altered by more than fourfold as compared to wild-type during exposure to mannose (SI Appendix, Fig. S1 and Dataset S1). However, aside from genes involved in mannose or mannan utilization, only 15 additional carbohydrate active enzymes were regulated by NCU00445 in response to mannose, the majority of which were involved in hemicellulose degradation (Dataset S1).

NCU00445 Is Required for Utilization of a Variety of Amino Acids as a Nitrogen Source.

To identify direct targets of NCU00445, we performed DAPseq, a technique that uses DNA sequencing to identify target sites in genomic DNA bound by in vitro synthesized transcription factors (18). Using DAPseq, we identified 296 NCU00445 DNA binding sites within 3,000 base pairs upstream of the translational start site of 290 genes (Dataset S3). Surprisingly, none of the promoters of genes involved in mannose or mannan utilization were bound by NCU00445. Additionally, Kyoto Encyclopedia of Genes and Genomes (KEGG) categories associated with carbon metabolism were not enriched in the set of 290 genes whose promoters were bound by NCU00445 (SI Appendix, Fig. S3A and Dataset S3).

To identify cellular processes directly regulated by NCU00445, we used functional enrichment analysis (25). Of the 335 genes that were at least fourfold differentially expressed between wild-type and ΔNCU00445 cells during exposure to mannose, 40 had promoters directly bound by NCU00445 (SI Appendix, Fig. S4 and Dataset S3). NCU00445 also bound its own promoter. Because we used transcriptional profiling of cells lacking NCU00445 to identify genes that were directly regulated by NCU00445, we were unable to conclusively determine whether NCU00445 regulated its own expression; we included NCU00445 in our list of NCU00445 direct targets. Functional enrichment analysis of the entire set of 335 genes (plus NCU00445 itself) showed enrichment for genes involved in valine, leucine, and isoleucine degradation and biosynthesis; alanine, aspartate, and glutamate metabolism; tryptophan metabolism and degradation of aromatic compounds; pentose and glucuronate interconversions; and biosynthesis of secondary metabolites (SI Appendix, Fig. S3B). Among the 41 genes whose promoters were bound by NCU00445, enrichment for valine, leucine, and isoleucine degradation and biosynthesis; tyrosine metabolism; and thiamine metabolism was observed (SI Appendix, Fig. S3C). Because mannose may not be the only condition in which NCU00445 plays a role, we investigated the KEGG pathways that were enriched among all 290 genes whose promoters were bound by NCU00445 and found enrichment for branched-chain amino acid degradation; tyrosine, phenylalanine, cysteine, methionine, and thiamine metabolism; and ubiquinone and other terpenoid–quinone biosynthesis (SI Appendix, Fig. S3A). These data indicated a potential role for NCU00445 in directly regulating amino acid metabolism.

Since the NCU00445 mutant grew without amino acid supplements (SI Appendix, Fig. S2 B and C and Dataset S2), we hypothesized that NCU00445 may be required for utilization of amino acids as a nitrogen source. To test this hypothesis, we grew wild-type and ΔNCU00445 cells on media containing individual amino acids as the sole nitrogen source. As a control, we included Δnit-2 cells, which are not able to utilize many amino acids as a nitrogen source, since wild-type cells are not able to utilize every nitrogen source equally well (8) (Fig. 2A and SI Appendix, Fig. S5 A and B and Dataset S2). Biomass measurements of the NCU00445 deletion mutant in 3 mL and 100 mL cultures showed ΔNCU00445 cells had growth defects on proline, phenylalanine, tryptophan, isoleucine, leucine, and threonine (Fig. 2A and SI Appendix, Fig. S5 A and B and Dataset S2). Expressing NCU00445 from the gpd-1 promoter in cells lacking NCU00445 mitigated growth defects on proline, isoleucine, and tryptophan (SI Appendix, Figs. S2D and S5B and Dataset S2). Because NCU00445 was required for amino acid catabolism, we named NCU00445 amn-1 for “amino acid utilization-1.”

Fig. 2.

AMN-1 is required for utilization of a variety of amino acids. (A) Mycelial dry weight of wild-type, Δamn-1, and Δnit-2 cells inoculated into 100 mL media containing the indicated nitrogen source (the concentration of all nitrogen sources was 50 mM, except for 1% bovine serum albumin [BSA]). Asterisks indicate mycelial dry weights that are statistically significantly different from that of wild-type cells; *P_adj < 0.05. (B) Hierarchical clustering of the expression level of the set of genes in a KEGG pathway associated with nitrogen metabolism in wild-type cells exposed to the indicated nutrient condition (SI Appendix, Table S1) (22). Genes in a particular KEGG pathway are indicated with colored bars from left to right: nitrogen metabolism (light green); alanine, aspartic acid, and glutamic acid metabolism (red); arginine and proline metabolism (dark blue); glycine, serine, and threonine metabolism (cyan); branched-chain amino acid metabolism (magenta); aromatic amino acid metabolism (yellow); cysteine and methionine metabolism (dark green); histidine metabolism (purple); lysine metabolism (orange); and nucleic acid metabolism (brown).

AMN-1 Regulates Genes Encoding Enzymes Required for the Catabolism of Amino Acids.

To further investigate the role of AMN-1 in amino acid utilization, we first transcriptionally profiled the response of wild-type cells to a panel of 12 nitrogen conditions to identify genes involved in the utilization of amino acids. These nitrogen conditions included the following: nitrogen starvation, ammonium nitrate, preferred nitrogen sources (ammonium, glutamate, and glutamine), and nonpreferred nitrogen sources (nitrate, alanine, arginine, proline, glycine, isoleucine, and tryptophan) covering a range of metabolic pathways (22).

Analysis of the expression of genes involved in nitrogen metabolism and the metabolism of amino acids across the nitrogen panel revealed that many genes involved in the metabolism of a specific nitrogen source were up-regulated on that nitrogen source (Fig. 2B and Dataset S1). For example, the expression of the nitrate reductase nit-3, the nitrite reductase nit-6 (NCU04720), the nitrate transporter nit-10 (NCU07205), and the FMN-dependent 2-nitropropane dioxygenase npd-1 (NCU03949) all showed increased expression levels on nitrate. Similarly, the expression of genes involved in tryptophan and quinate metabolism, such as the kynureninase gene kyn-1 (NCU09183) and the indoleamine 2,3-dioxygenase genes iad-1 (NCU09184) and iad-2 (NCU01402), was highest on tryptophan (Dataset S1).

Exposure to glycine resulted in the upregulation of a large number of genes. These included genes involved in glycine metabolism such as gly-3 (NCU02727), gyd-1 (NCU02475), oxD (NCU06558), and NCU03761 but also genes involved in the metabolism of a variety of other amino acids (Fig. 2B and Dataset S1) (26). This result may indicate that since cells rarely see glycine in isolation from other amino acids, the presence of glycine activated expression of genes involved in a variety of amino acid utilization pathways. We asked whether genes from a variety of metabolic pathways were expressed during exposure to other amino acids. Indeed, when the expression pattern of genes in KEGG metabolic pathways was analyzed, it became clear that genes in a KEGG pathway associated with the metabolism of a particular amino acid were not specifically up-regulated in response to that nutrient (Fig. 2B). Additionally, hierarchical clustering of gene expression across our panel of nitrogen sources did not show discrete clusters of genes in a particular KEGG pathway (Fig. 2B).

To identify genes specifically regulated by AMN-1, we exposed the Δamn-1 mutant to media containing proline and the aromatic and branched-chain amino acids for which Δamn-1 cells showed the most severe growth defects, tryptophan and isoleucine, respectively (Fig. 2A and SI Appendix, Fig. S5 A and B and Dataset S2). To control for growth defects of Δamn-1 cells under these conditions, we grew wild-type and Δamn-1 cells in media containing ammonium nitrate as the nitrogen source, washed the mycelial mass in media lacking a nitrogen source, and transferred the mycelial cell mass into media containing the experimental nitrogen source for 4 h prior to harvesting for RNAseq.

We first investigated mechanisms of AMN-1–mediated regulation of proline catabolism. Functional enrichment analysis of the 57 genes that were at least fourfold differentially expressed between Δamn-1 and wild-type cells during exposure to proline showed significant enrichment for valine, leucine, and isoleucine degradation; β-alanine metabolism; tyrosine metabolism; phenylalanine metabolism; and glycerolipid metabolism (SI Appendix, Fig. S6A and Dataset S1). Eight of the 57 genes had promoters bound by AMN-1, including the homogentisate 1,2-dioxygenase NCU05499 and the dimethylallyl tryptophan synthase NCU12075, both predicted to be involved in aromatic amino acid metabolism, and the promoters of three genes predicted to be involved in branched-chain amino acid metabolism: the branched-chain α-keto acid dehydrogenase E2 component NCU02704, the isovaleryl-CoA dehydrogenase NCU02126, and the methylcrotonoyl-CoA carboxylase β subunit mcc-2 (NCU02127) (Figs. 3 and 4A and SI Appendix, Figs. S4 and S5C and Dataset S1).

Fig. 3.

NIT-2 regulates transporters and genes associated with general nitrogen metabolism, while pathway-specific transcription factors activate genes that are required for utilization of a specific nitrogen source. The plot was built with Circos, version 0.69 (70), to display the regulation of genes whose promoters were bound and whose expression was regulated by AMN-1, NIT-2, or NIT-4 when cells were exposed to the indicated conditions. Regulated genes were at least fourfold differentially expressed between wild-type and Δamn-1, Δnit-2, or Δnit-4 cells. Genes encoding transporters are highlighted in yellow. Solid lines indicate genes that were activated by AMN-1, NIT-2, or NIT-4. Dotted lines indicate genes that were repressed by AMN-1, NIT-2, or NIT-4.

Fig. 4.

Regulation of amino acid utilization by AMN-1 and NIT-2 occurs through different genetic pathways. (A–C) Differential expression analysis of Δamn-1 relative to wild-type cells after a shift to (A) 50 mM proline, (B) 50 mM tryptophan, and (C) 50 mM isoleucine. Green circles indicate genes whose promoters were bound by AMN-1. Dotted lines indicate fourfold change in expression. (D) Heatmap of the expression level of the 176 genes that were at least fourfold differentially expressed between wild-type and Δnit-2 cells exposed to 50 mM proline, 50 mM isoleucine, or 50 mM tryptophan plus nit-2 itself. Purple bars indicate genes whose promoters were bound by NIT-2. Green bars indicate genes that were differentially expressed between wild-type and Δnit-2 cells exposed to proline plus nit-2. Pink bars indicate genes that were differentially expressed between wild-type and Δnit-2 cells exposed to tryptophan plus nit-2. Yellow bars indicate genes that were differentially expressed between wild-type and Δnit-2 cells exposed to isoleucine plus nit-2. Numbers above each column indicate the total number of genes in each category.

The remaining genes whose promoters were bound by AMN-1 included the proline oxidase pro-7 (NCU02936), which is predicted to catalyze the first step in proline degradation. S. cerevisiae cells lacking the pro-7 ortholog, PUT1 are unable to utilize proline as a nitrogen source (27). In N. crassa, cells lacking pro-7 showed a severe growth defect when provided proline as the sole nitrogen source (SI Appendix, Fig. S7 A and B and Dataset S2). We hypothesized that the reduced expression of pro-7 in the amn-1 deletion mutant might be at least partially responsible for the inability of Δamn-1 cells to utilize proline. To test this, we asked whether expression of pro-7 under the regulation of the constitutive gpd-1 promoter restored growth in cells lacking amn-1 (SI Appendix, Fig. S7C and Dataset S2). Indeed, P_gpd-1-pro-7 Δamn-1 cells grew significantly better than the Δamn-1 mutant when proline was provided as the sole nitrogen source, indicating that reduced expression of pro-7 in cells lacking amn-1 was at least partially responsible for the inability of Δamn-1 cells to utilize proline (SI Appendix, Fig. S7 A and B and Dataset S2).

Although cells lacking amn-1 had a severe growth defect on media containing tryptophan as the sole nitrogen source, only 16 genes were at least fourfold differentially expressed in Δamn-1 cells as compared to wild-type (Fig. 4B and SI Appendix, Fig. S5C and Dataset S1). Of these 16 genes, the promoters of 12 were bound by AMN-1 (Figs. 3 and 4B and SI Appendix, Figs. S4 and S5C). Five of the genes whose promoters were bound by AMN-1 and whose expression was regulated by AMN-1 in response to tryptophan were predicted to be involved in aromatic amino acid metabolism: the flavoprotein oxygenase fpo-1 (NCU00236), the fumarylacetoacetase fah-1 (NCU05537), the homogentisate 1,2-dioxygenase NCU05499, the 4-hydroxyphenylpyruvate dioxygenase hpd-1 (NCU01830), and a hypothetical protein NCU01829 predicted to be involved in tyrosine catabolism. N. crassa cells lacking fah-1 had a slight growth defect on tryptophan (SI Appendix, Fig. S7 D and E and Dataset S2). However, deletion of NCU05499 did not significantly affect growth on tryptophan, suggesting that the mechanism by which AMN-1 regulated tryptophan metabolism may be through a complex combination of genes (SI Appendix, Fig. S7 D and E and Dataset S2). The remaining genes whose promoters were bound by AMN-1 and regulated by AMN-1 in response to tryptophan were involved in either proline metabolism (pro-7 and the pyrroline 5-carboxylate dehydrogenase pcd-2 [NCU03076]) or branched-chain amino acid metabolism (the branched-chain amino acid aminotransferase val-1 [NCU04292], the 2-oxoisovalerate dehydrogenase β subunit ovd-2 [NCU03913], mcc-2, NCU02704, and NCU02126) (Figs. 3 and 4B and SI Appendix, Figs. S4 and S5C and Dataset S1).

Cells lacking amn-1 also had a growth defect when branched-chain amino acids were provided as the nitrogen source (Fig. 2A and SI Appendix, Fig. S5 A and B and Dataset S2). The expression of 17 genes was at least fourfold differentially expressed between wild-type and Δamn-1 cells during exposure to media containing isoleucine as the nitrogen source, and the promoters of 14 of these genes were bound by AMN-1 (Figs. 3 and 4C and SI Appendix, Figs. S4 and S5C and Dataset S1). Six of these 14 genes play a role in branched-chain amino acid metabolism, including the 2-oxoisovalerate dehydrogenase α and β subunits ovd-1 (NCU09864) and ovd-2, the methylcrotonoyl-CoA carboxylase α and β subunits mcc-1 (NCU00591) and mcc-2, NCU02704, and NCU02126. Additionally, both pro-7 and pcd-2, necessary for proline catabolism, and fah-1, hpd-1, NCU05499, and NCU01829, involved in aromatic amino acid metabolism were regulated by AMN-1 in response to isoleucine and had promoters bound by AMN-1 (Figs. 3 and 4C and SI Appendix, Figs. S4 and S5C and Dataset S1).

To comprehensively identify genes regulated directly by AMN-1, we cross referenced the 290 genes with AMN-1 binding sites in the promoter region identified by DAPseq with genes that were differentially expressed by at least fourfold across our RNAseq experiments. Although it is likely that AMN-1 regulates additional genes during exposure to conditions or at time points we did not test by RNAseq, we identified 43 genes that were regulated by AMN-1 during exposure to mannose, proline, tryptophan, or isoleucine and whose promoters were bound by AMN-1 (Fig. 3 and SI Appendix, Fig. S4 and Dataset S3). Supporting our hypothesis that AMN-1 is required for utilization of a variety amino acids, functional analysis of these 43 genes (plus amn-1 itself) showed enrichment for genes involved in valine, leucine, and isoleucine degradation and biosynthesis, tyrosine metabolism, and thiamine metabolism (SI Appendix, Fig. S3C). Using the 45 promoter binding sites found in these 44 genes, we identified a consensus binding motif KCGGYTWKYRKCGGCHWW for AMN-1 (SI Appendix, Fig. S8A). This motif provided additional specificity to the NNCGGNNNNN motif identified in a broad survey of transcription factor binding motifs (28). As expected, AMN-1 only bound a small subset of the 2,189 locations where the KCGGYTWKYRKCGGCHWW motif was identified in N. crassa promoters (Dataset S4) (29).

The Nitrogen Catabolite Repressor NIT-2 Regulates Genes Necessary for Nitrogen Import and Enzymes Responsible for Amino Acid Utilization.

NIT-2 is a conserved GATA-type transcription factor responsible for regulating genes involved in utilizing nonpreferred nitrogen sources in filamentous fungi (6). Although initially identified for its role in promoting nitrate utilization in N. crassa, early studies also found NIT-2 is necessary for growth when a variety of amino and nucleic acids are provided as the nitrogen source (8). Unlike cells lacking nit-2, the Δamn-1 mutant exhibited normal growth when nitrate was provided as the nitrogen source (Fig. 2A and Dataset S2). Similarly, amn-1 is not required for growth on nucleic acids (SI Appendix, Fig. S5 A and B and Dataset S2). When exposed to amino acids, there was significant overlap in the substrates that caused growth defects of Δamn-1 cells as compared to Δnit-2 cells but also a number of notable differences. In particular, Δnit-2 cells had a more severe growth phenotype than Δamn-1 cells during exposure to glutamic acid, glycine, isoleucine, leucine, valine, phenylalanine, lysine, methionine, serine, and threonine while the reverse was true for tryptophan (p_adj < 0.05) (Fig. 2A and SI Appendix, Fig. S5 A and B and Dataset S2). These data indicated that NIT-2 has a broader role in nitrogen metabolism than AMN-1.

Because of the expanded role of NIT-2 relative to AMN-1 with respect to nitrogen metabolism, we hypothesized that AMN-1 and NIT-2 may not act through the same pathway to regulate nitrogen metabolism. To test this hypothesis, we performed RNAseq on cells lacking nit-2 under the same nitrogen conditions that we tested Δamn-1 cells (proline, tryptophan, and isoleucine) as well as exposure to nitrate and ammonium nitrate. To complement these transcriptional profiling experiments, we also performed DAPseq on NIT-2 to identify genes whose promoters were bound by NIT-2. Using DAPseq, we identified 354 NIT-2 binding sites in the putative promoter regions of 344 genes (Dataset S3). By comparing these 344 genes to the differentially expressed genes (at least fourfold) between wild-type and Δnit-2 cells during exposure to nitrate, ammonium nitrate, proline, tryptophan, or isoleucine, we identified the direct regulation of 33 of these genes by NIT-2 (Fig. 3 and Datasets S1 and S3).

The expression of 176 genes differed by at least fourfold in wild-type as compared to Δnit-2 cells during exposure to media containing proline, tryptophan, or isoleucine as the nitrogen source (Fig. 4D and SI Appendix, Fig. S9 A–C and Dataset S1). Functional analysis of these 176 genes showed enrichment for genes involved in the metabolism of a number of amino acids including glycine, serine, threonine, tyrosine, phenylalanine, arginine, proline, and β-alanine. We also observed an enrichment for glyoxylate and dicarboxylate metabolism as well as for carbon and nitrogen metabolism generally (SI Appendix, Fig. S6C). Of these 176 genes, the promoters of 22 were bound by NIT-2 (Figs. 3 and 4D and SI Appendix, Fig. S9 A–C and Datasets S1 and S3). Fully half of these 22 genes encoded transporters, suggesting that NIT-2 may regulate nitrogen metabolism primarily by regulating import of nitrogen substrates (Fig. 3 and SI Appendix, Fig. S9 A–C and Dataset S1).

To further explore these data, we first analyzed the 13 genes directly regulated by NIT-2 in response to proline, tryptophan, and isoleucine. Consistent with the role of NIT-2 functioning as a transcriptional activator (6), the expression of all 13 genes was down-regulated in the Δnit-2 mutant as compared to wild-type (Figs. 3 and 4D and Dataset S1). Nine of these 13 genes encoded for transporters, including the general amino acid permease aap-22 (NCU04435) and the oligopeptide transporter opt-1 (NCU09773) (Fig. 3 and Dataset S1). The expression of both of these transporters was reduced by more than 80-fold to a fragments per kilobase of transcript per million mapped reads value (FPKM) of less than 1 in cells lacking nit-2 as compared to wild-type cells when exposed to proline, tryptophan, or isoleucine (Dataset S1). This dramatic reduction in expression may at least partially explain the inability of Δnit-2 cells to grow on many amino acids as the sole nitrogen source. Previously posited to be regulated by NIT-2 (30, 31), the uracil permease uc-5 (NCU07334) promoter was bound by and transcription of uc-5 was activated by NIT-2. NIT-2 also directly regulated the purine permease pup-2 (NCU06918) and the H+/nucleoside cotransporter NCU08148 (Fig. 3 and Dataset S1). The regulation of these three nucleotide transporters may at least partially explain the role of NIT-2 in utilizing nucleotides as a nitrogen source (6). Urea uptake is also under the control of nitrogen catabolite repression (32); the urea transporter urt (NCU09909) was directly regulated by NIT-2 (Fig. 3 and Dataset S1). Consistent with its role in regulating the utilization of nonpreferred nitrogen sources in the presence of preferred nitrogen sources such as ammonium (6), the ammonium transporter tam-4 (NCU01065) was also directly regulated by NIT-2. Additionally, NIT-2 regulated the expression and bound the promoters of the high-affinity nickel transporter trm-34 (NCU08225) and the major facilitator superfamily (MFS) transporter NCU08407 (Fig. 3 and Dataset S1).

Four other genes down-regulated in Δnit-2 cells as compared to wild-type cells during exposure to proline, tryptophan, and isoleucine and whose promoters were bound by NIT-2 encoded enzymes important for amino and nucleic acid degradation. Unlike AMN-1, which mainly regulated the expression of enzymes important for utilizing a particular amino acid or class of amino acids, NIT-2 regulated the transcription of genes that encoded enzymes involved in more general metabolic pathways (Fig. 3). These included the agmatinase put-3 (NCU01348), an enzyme involved in the catabolism of amino groups, and the formamidase fma-1 (NCU02361) which participates in glyoxylate and dicarboxylate metabolism and may play a role in tryptophan metabolism. NIT-2 also directly regulated two genes involved in nucleic acid metabolism: the thymine dioxygenase uc-3 (NCU06416) and the uracil-5-carboxylate decarboxylase uc-7 (NCU06417) (33, 34) (Fig. 3 and Dataset S1).

Nine genes whose promoters were bound by NIT-2 were differentially expressed in a Δnit-2 mutant as compared to wild-type when exposed to one or two amino acids as nitrogen sources (Figs. 3 and 4D and Dataset S1). Several of these genes play a general role in amino acid utilization, including the general amino acid permease pmg (NCU03509), the amino acid-polyamine-organocation (APC) permease aap-9 (NCU07175) (35), and the Glu/Leu/Phe/Val dehydrogenase am (NCU01195) (32, 36, 37). Two genes played a role in purine catabolism: the allantoinase aln-1 (NCU02296) and the uracil phosphoribosyltransferase uc-8 (NCU06261) (38). Taken together, these data indicated that NIT-2 primarily functions to regulate genes, such as transporters, that have a general effect on the metabolism of a variety of amino acids rather than regulating genes directly involved in the utilization of specific amino acids.

NIT-2 Regulation in Response to Nitrate.

A large number of genes that did not have a direct connection to amino acid catabolism were directly regulated by NIT-2 in response to amino acids (Fig. 3 and Dataset S1). As NIT-2 regulates the utilization of nonpreferred inorganic nitrogen sources, such as nitrate (6), we hypothesized that the expression of these genes may also be activated by NIT-2 when cells are exposed to nitrate.

To test this hypothesis, we transferred wild-type and Δnit-2 cells to media containing nitrate and did RNAseq to measure global gene expression. The expression of 113 genes differed by at least fourfold between wild-type and Δnit-2 cells 4 h postshift to nitrate (Fig. 5A and SI Appendix, Fig. S9D and Dataset S1). Functional analysis of these genes showed an enrichment for genes involved in nitrogen metabolism and ribosome biogenesis (SI Appendix, Fig. S6D). Over 40% of the genes regulated by NIT-2 during exposure to nitrate were also regulated by NIT-2 during exposure to proline, tryptophan, or isoleucine, including 13 genes that encoded for transporters (Dataset S1). To distinguish between genes that NIT-2 directly, as opposed to indirectly, regulated in response to nitrate, we compared the genes that were differentially expressed between wild-type cells and cells lacking nit-2 during exposure to nitrate with genes whose promoters were bound by NIT-2. We identified 22 genes directly regulated by NIT-2 in response to nitrate, all of which were down-regulated in Δnit-2 cells as compared to wild-type (Figs. 3 and 5A and SI Appendix, Fig. S9D and Datasets S1 and S3). These 22 genes included 12 of the 13 genes that were differentially expressed in Δnit-2 cells in response to proline, tryptophan, and isoleucine as compared to wild-type cells and whose promoters were bound by NIT-2, with only fma-1 not represented. The promoter of an additional gene, aln-1, was bound by NIT-2, and aln-1 was differentially expressed during exposure to nitrate, isoleucine, and tryptophan in Δnit-2 as compared to wild-type cells (Fig. 3 and Datasets S1 and S3).

Fig. 5.

NMR and NIT-4 activate genes during starvation conditions. (A) Differential expression analysis of Δnit-2 relative to wild-type cells after a shift to 50 mM nitrate. Green squares indicate genes whose promoters are bound by NIT-2. Purple pentagons indicate genes predicted to encode nitrogen transporters. Dotted lines indicate a fourfold change in expression. (B) Heatmap of the expression level of the 60 genes that were at least fourfold differentially expressed between wild-type and Δnmr cells exposed to 25 mM ammonium nitrate plus nmr itself. Upper heatmap includes genes repressed by NMR. Lower heatmap includes genes activated by NMR plus nmr itself. Purple bars indicate genes whose promoters were bound by NIT-2. Green bars indicate genes that were differentially expressed between Δnit-2 and wild-type cells exposed to 25 mM ammonium nitrate. Yellow bars indicate genes that were differentially expressed between Δnit-2 and wild-type cells exposed to 50 mM nitrate. Numbers above each column indicate the total number of genes in each category. (C) Differential expression analysis of Δnmr relative to wild-type cells after a shift to nitrogen starvation. Green squares indicate genes whose promoters were bound by NIT-2. Red circles indicate genes whose promoters were bound by NIT-4. Dotted lines indicate a fourfold change in expression. (D) Differential expression analysis of Δnit-4 relative to wild-type cells after a shift to 50 mM nitrate. Red circles indicate genes whose promoters were bound by NIT-4. Dotted lines indicate a fourfold change in expression. (E) Heatmap of the expression level of the set of genes that were at least fourfold differentially expressed between wild-type and Δnit-4 cells exposed to nitrogen starvation. Upper heatmap includes genes repressed by NIT-4. Lower heatmap includes genes activated by NIT-4. Green bars indicate genes whose promoters were bound by NIT-4. Yellow bars indicate genes differentially expressed between Δnmr and wild-type cells exposed to nitrogen starvation. Numbers above each column indicate the total number of genes in each category.

We investigated the nine genes whose promoters were bound by NIT-2 and whose expression was regulated by NIT-2 in response to nitrate but not proline, tryptophan, or isoleucine. Two of these genes were specifically involved in nitrate utilization: the nitrite reductase nit-6 and the nitrate transporter nit-10 (6). The promoter of the ammonium transporter tam-1 (NCU03257) was also bound by NIT-2, and tam-1 expression was regulated by NIT-2 during exposure to nitrate. The majority of the remaining genes were involved in functions relating to the ribosome (Fig. 3 and Dataset S1). To our surprise, although previous studies identified NIT-2 binding sites in the nit-3 promoter, we did not find NIT-2 bound to the nit-3 promoter using DAPseq (12, 39). Since DAPseq is an in vitro method of identifying transcription factor binding sites, this may suggest that DNA modification or additional proteins are required to facilitate NIT-2 binding in some promoter regions.

Using the 38 NIT-2 binding sites in the promoters of these 33 directly regulated genes, we identified a consensus binding motif for NIT-2 (HGATAAGV) that had similarity to the binding motif identified in a broad survey of eukaryotic transcription factors (NNGATHNN) as well as other GATA transcription factor motifs, including the binding motif of the NIT-2 S. cerevisiae homolog Gln3 (28, 40–42) (SI Appendix, Fig. S8B). Prior work investigating the nucleotide sequence of the NIT-2 binding site in specific promoters, rather than globally, suggested a highly similar binding motif (43–45). This motif was present 804 times in the promoter regions of N. crassa genes, indicating that NIT-2 did not bind to every HGATAAGV motif in the genome (Dataset S4).

The Transcriptional Repressor NMR Acts as an Activator under Nitrogen Starvation Conditions.

Regulation of nitrate utilization is accomplished through the interactions of NIT-2, NIT-4, and NMR (6) (Fig. 1). Previous work characterizing the regulation of the nitrate reductase gene nit-3 showed that NIT-4 promotes expression of nit-3 in the presence of nitrate (46). If nitrate is present in the absence of a preferred nitrogen source, NIT-2 also promotes expression of nit-3. However, when preferred nitrogen sources are present, the activity of NIT-2 is repressed by NMR (11), and nit-3 is not expressed (6) (Fig. 1).

Our data showed that the expression of three genes required for nitrate utilization, nit-3, the nitrite reductase gene nit-6, and the nitrate transporter nit-10, was negligible in wild-type cells exposed to media containing the preferred nitrogen source ammonium (SI Appendix, Fig. S10A and Dataset S1). Exposure of wild-type cells to media containing nitrate increased expression of nit-3, nit-6, and nit-10 by over sixfold. Deletion of either nit-2 or nit-4 resulted in a more than fourfold drop in expression of nit-3, nit-6, and nit-10 under nitrate conditions. Upon exposure to media containing both ammonium and nitrate, expression of nit-3, nit-6, and nit-10 dropped by nearly twofold as compared to exposure to nitrate alone in wild-type cells in an NMR-dependent fashion (SI Appendix, Fig. S10A and Dataset S1). These data broadly supported the model of nitrate utilization developed using classical genetics (Fig. 1).

To investigate the role of the nitrogen repressor, NMR, we exposed wild-type cells and cells lacking nmr to media containing both nitrate and ammonium. The expression of 60 genes was at least fourfold differentially expressed between wild-type and Δnmr cells (Fig. 5B and SI Appendix, Fig. S10B and Dataset S1). Functional analysis showed an enrichment for genes involved in the metabolism of a number of amino acids including arginine, proline, glycine, serine, threonine, phenylalanine, tyrosine, alanine, aspartate, glutamate, β-alanine, and cyanoamino acids (SI Appendix, Fig. S10C). As expected, all 10 genes bound by NIT-2 and differentially expressed between wild-type and Δnmr cells during exposure to ammonium nitrate were more highly expressed in Δnmr cells, providing confirmation of the role of NMR in repressing NIT-2 activity (Fig. 5B and SI Appendix, Fig. S10B and Dataset S1). Although NMR is canonically only thought to act through NIT-2, only 11 genes were at least fourfold differentially expressed between both wild-type and Δnmr cells exposed to ammonium nitrate and wild-type and Δnit-2 cells exposed to nitrate, suggesting that NMR may have another role in transcriptional regulation (Fig. 5B and Dataset S1). For example, while many of the 60 genes regulated by NMR were predicted to play a role in nitrogen metabolism, there were also a significant number of genes involved in carbon metabolism, including genes encoding carbohydrate active enzymes and sugar transporters (Dataset S1). These data suggested a role for NMR in regulating carbon utilization as well as nitrogen utilization.

Our data supported the role of NMR as a transcriptional repressor. Of the 60 genes that were differentially expressed by at least fourfold between wild-type and Δnmr cells, 43 were more highly expressed in cells lacking nmr (Fig. 5B and SI Appendix, Fig. S10B and Dataset S1). However, the carbon catabolite repressor cre-1 (NCU08807) acts as a repressor when preferred carbon sources are available but as an activator during carbon starvation (47). We hypothesized that nmr might similarly function as an activator during nitrogen starvation. To test this hypothesis, we performed RNAseq on wild-type and Δnmr cells during exposure to media lacking a nitrogen source. In total, 70 genes were differentially expressed (at least fourfold) between wild-type and Δnmr cells, and of these, 60 were down-regulated in cells lacking nmr (Fig. 5C and SI Appendix, Fig. S10D and Dataset S1). Two of the three most highly down-regulated genes were nit-3 and nit-6, which were expressed more than 32-fold higher in wild-type cells than Δnmr cells during nitrogen starvation (SI Appendix, Fig. S10A and Dataset S1). Other genes down-regulated in the Δnmr mutant during nitrogen starvation included the nitrate transporter nit-10 and genes involved in amino acid metabolism. Only nine genes were differentially expressed in the Δnmr mutant as compared to wild-type during nitrogen starvation and in media containing ammonium nitrate (Dataset S1). This result could be because NMR plays a role in regulating nitrogen metabolism overall, not simply the interplay between nitrate and ammonium.

NIT-4 Directly Activates Genes Necessary for Nitrate Utilization in Response to Nitrate and Nitrogen Starvation.

Along with NIT-2, the transcription factor NIT-4 is required for expression of genes necessary to utilize nitrate (Fig. 1) (6). We hypothesized that the interplay of NIT-2 and NIT-4 may be similar to that of NIT-2 and AMN-1, with NIT-2 mainly directly regulating genes involved in general nitrogen metabolism while NIT-4 directly regulates the enzymes required for nitrate utilization. To test this hypothesis, we did RNAseq on wild-type and Δnit-4 cells after a shift to media containing nitrate as the nitrogen source. The expression of 76 genes was at least fourfold differentially expressed in Δnit-4 cells as compared with wild-type during exposure to nitrate (Fig. 5D and SI Appendix, Fig. S11A and Dataset S1). Functional analysis of these 76 genes showed enrichment for genes involved in nitrate and methane metabolism (SI Appendix, Fig. S11B).

NIT-4 is thought to act as a transcriptional activator (Fig. 1). However, only 21 of the 76 genes that were differentially expressed between wild-type and Δnit-4 cells during exposure to nitrate were down-regulated in the Δnit-4 mutant (Fig. 5D and SI Appendix, Fig. S11A and Dataset S1). To determine whether any of the genes repressed by NIT-4 were directly regulated, we performed DAPseq on NIT-4 and identified 29 NIT-4 binding sites within 3,000 base pairs upstream of translational start sites of 29 genes (Dataset S3). We compared the 29 genes with NIT-4 binding sites in their promoters with the list of genes that were differentially expressed in the nit-4 deletion mutant during exposure to nitrate. Only eight genes whose promoters were bound by NIT-4 were also differentially expressed in Δnit-4 cells as compared to wild-type exposed to nitrate (Figs. 3 and 5D and SI Appendix, Fig. S11A and Dataset S1). All eight of these genes were down-regulated in Δnit-4 cells as compared to wild-type, suggesting that the 54 genes that were up-regulated in cells lacking nit-4 were indirectly regulated, perhaps in response to nitrogen starvation (Fig. 5D and SI Appendix, Fig. S11A). Indeed, many of the genes that were repressed by NIT-4 in response to nitrate were up-regulated in wild-type cells exposed to nitrogen starvation (SI Appendix, Fig. S11A). The eight directly regulated genes included nit-3, nit-6, and nit-10 as well as the FMN-dependent 2-nitropropane dioxygenase npd-1, the flavohemoglobin protein fhb-2 (NCU10051), and the cyanamide hydratase NCU04466, which are all important in nitrogen metabolism. NIT-4 also bound and regulated the HPP family protein NCU09160 (Fig. 3 and Dataset S1); HPP family proteins have been reported to have nitrite transport activity in cyanobacteria (48). We used the NIT-4 binding sites located in these eight genes to identify the consensus binding motif KCCGCGGAGARAG for NIT-4, which shows some similarity to NTCCGCGGVN, the NIT-4 binding motif identified in a survey of eukaryotic transcription factors (28) (SI Appendix, Fig. S8C). The KCCGCGGAGARAG NIT-4 binding motif was present 1,912 times in N. crassa promoters, indicating that NIT-4 did not bind every location in the genome where its binding motif was present (Dataset S4).

A comparison of NIT-4– and NIT-2–regulated genes did not show as stark of a contrast as when comparing NIT-2– and AMN-1–regulated genes. Half of the genes that were regulated by NIT-4 in response to nitrate were also regulated by NIT-2 under ammonium nitrate, nitrate, proline, tryptophan, or isoleucine conditions (Dataset S1). These coregulated genes included nit-3 and nit-6 as well as the nitrate transporter nit-10 (Dataset S1). A closer examination showed that many genes whose expression was activated by NIT-2 were repressed by NIT-4, including three amino acid or oligopeptide transporters, suggesting that much of the overlap in regulation may be due to the cell experiencing nitrogen starvation conditions (SI Appendix, Figs. S9D and S11A and Dataset S1).

Our data suggested that NMR, which lacks a DNA binding domain, functioned as an activator during nitrogen starvation. However, it was not clear what transcription factor(s) might act downstream of NMR. Since NIT-4 activated the transcription of a number of genes that were up-regulated in response to nitrogen starvation, we asked whether NIT-4 might play a role in transcriptional activation during nitrogen starvation by doing RNAseq on wild-type and Δnit-4 cells. The expression of 85 genes was at least fourfold differentially expressed between wild-type and Δnit-4 cells during nitrogen starvation (Fig. 5E and SI Appendix, Fig. S11C and Dataset S1). Functional analysis showed an enrichment for genes involved in the metabolism of a number of amino acids including branched-chain amino acids, aromatic amino acids, glycine, serine, and threonine as well as other genes involved in nitrogen metabolism (SI Appendix, Fig. S11D). Over three-quarters of these genes were down-regulated in Δnit-4 cells (Fig. 5E and SI Appendix, Fig. S11C and Dataset S1). Notably, all seven genes whose promoters were bound by NIT-4 and were regulated by NIT-4 in response to nitrogen starvation were down-regulated in Δnit-4 cells, including five of the six most highly down-regulated genes (Fig. 5E and SI Appendix, Fig. S11C and Dataset S1). All seven of these genes were also directly regulated by NIT-4 in response to nitrate (Fig. 3). These data suggested that NIT-4 acts as a transcriptional activator and may function downstream of NMR in response to nitrogen starvation.

Discussion

N. crassa is a saprotrophic fungus that gets the majority of its nutrients from plant biomass. In this environment, nitrogen is a limiting nutrient (4). Genes involved in nitrogen metabolism are regulated by the interplay of transcription factors that respond to a particular nitrogen source and transcription factors that activate genes involved in the utilization of nonpreferred nitrogen sources when no preferred nitrogen sources are present (6). Through a combination of RNAseq and DAPseq, we compared the genes regulated by pathway-specific transcription factors and those regulated through nitrogen catabolite repression by NIT-2. Our data indicate that pathway-specific transcription factors generally directly regulate genes that encode for enzymes necessary for metabolism of a specific nitrogen source, while NIT-2 directly regulates genes that encode for transporters and genes necessary for the metabolism of a broad range of nitrogen sources.

AMN-1 Encodes a Zinc Binuclear Cluster Transcription Factor that Is Required for Amino Acid Utilization.

Although pathway-specific transcription factors regulating nitrogen utilization have been characterized for a few nitrogen substrates, including nitrate and purine catabolism, regulation of the catabolism of many nonpreferred nitrogen sources is poorly understood in filamentous fungi. Our data indicate that AMN-1 plays a role in directly regulating genes involved in the catabolism of proline, aromatic amino acids, and branched-chain amino acids (Fig. 3 and SI Appendix, Fig. S4). Interestingly, AMN-1 also indirectly regulates the expression of genes in response to mannose, although AMN-1 is not required for mannose utilization (SI Appendix, Figs. S1 and S2).

AMN-1 is conserved among ascomycete fungi (SI Appendix, Fig. S12). Although AMN-1 is required for the utilization of a number of amino acids, homologs of AMN-1, such as Aro80 in S. cerevisiae, are required only for aromatic amino acid utilization (49) (Fig. 2A and SI Appendix, Fig. S5 A and B). In the human pathogens Aspergillus fumigatus and Penicillium marneffei, the amn-1 homolog hmgR regulates a gene cluster required for tyrosine catabolism (50, 51). While the homologs of these genes are not contained in a gene cluster in the N. crassa genome, AMN-1 binds the promoters of all but one of the six genes in the A. fumigatus cluster. The P. marneffei gene cluster contains an additional two genes, one of which encodes for a putative α-1,2-mannosidase (51). This observation may indicate a conserved role for AMN-1 regulation of genes involved in mannose or mannan utilization across ascomycete fungi.

Unlike Aro80 and HmgR, AMN-1 also plays a role in regulating genes involved in branched-chain amino acid and proline metabolism (Fig. 3 and SI Appendix, Fig. S4). In fungi, both branched chain and aromatic amino acids can be catabolized through the Ehrlich pathway (52). However, proline is catabolized through a separate pathway and regulated by Put3/PrnA in S. cerevisiae and Aspergillus nidulans, respectively (53, 54). The prnA gene in A. nidulans is part of a proline utilization gene cluster containing four genes required for proline utilization (55). AMN-1 binds the promoters of orthologs of three of these four genes, although they are not clustered in the N. crassa genome.

The majority of genes whose promoters are bound by AMN-1 are regulated in response to mannose rather than proline, aromatic amino acids, or branched-chain amino acids (Fig. 3 and SI Appendix, Fig. S4). Many of the amino acids available for N. crassa utilization in the wild are likely part of proteins secreted either from plant or fungal cells and are likely to be glycosylated. We postulate that N. crassa may use mannose as a signal to indicate the presence of glycosylated proteins to activate genes involved in amino acid catabolism. Further work will be necessary to investigate the connection and conservation of mannose and amino acid catabolism in filamentous fungi.

Nitrogen Catabolite Repression Regulates Transporters, while Pathway-Specific Transcription Factors Regulate Catabolic Enzymes.

In N. crassa, nitrogen catabolite repression is regulated by NIT-2 (7). Utilization of nonpreferred nitrogen sources requires the activity of both NIT-2 and pathway-specific transcription factors (6). Prior studies suggested that binding of NIT-2 and a pathway-specific transcription factor is required for activation of gene transcription (12, 19). Here, a global analysis of NIT-2 promoter binding sites combined with the promoter binding sites of the pathway-specific transcription factors AMN-1 and NIT-4 indicated that while there are isolated incidences of gene coregulation by NIT-2 and either NIT-4 or AMN-1, the set of genes whose promoters were bound by NIT-2 was broadly separate from the set of genes whose promoters were bound by the pathway-specific transcription factors NIT-4 and AMN-1 (Fig. 3).

The pathway-specific transcription factors AMN-1 and NIT-4 generally regulated genes that encode enzymes involved in the catabolism of a particular nitrogen source, while NIT-2 regulated genes that encode for proteins required for utilization of all nitrogen sources. In particular, NIT-2 directly regulated the expression of a number of transporters that import nitrogen sources, including amino acids, nucleic acids, ammonium, nitrate, and urea, into the cell (Fig. 3). This role is similar to that of the carbon catabolite repressor CRE-1, which binds the promoters of genes encoding a number of transporters involved in importing different carbon sources into the cell (17). Regulation of transporters by major transcriptional regulators of both carbon and nitrogen catabolite repression indicates that controlling import of nutrients into the cell is one of the major mechanisms by which fungi prioritize the utilization of preferred nutrient sources over nonpreferred nutrient sources.

Nitrogen Utilization Genes Are Activated by NMR and NIT-4 in Response to Starvation.

The role of regulators of carbon and nitrogen catabolite repression has mainly been explored in conditions in which nutrients are present (6, 56). However, major regulators of carbon and nitrogen catabolite repression also appear to play a role in gene regulation during starvation conditions. In A. nidulans, the NIT-2 ortholog AreA activates the expression of target genes during nitrogen starvation (57). CRE-1, canonically thought of as a transcriptional repressor, activates the expression of genes required for utilization of nonpreferred carbon sources during carbon starvation (47). In a similar fashion, the nitrogen catabolite repressor NMR activated the expression of genes necessary to utilize nonpreferred nitrogen sources during nitrogen starvation (Fig. 5C and SI Appendix, Fig. S10D). Unlike CRE-1, NMR does not have a DNA binding domain, so NMR must work through transcription factors, such as NIT-2, to activate the expression of downstream targets (58). Our work suggests it may also act upstream of NIT-4 during nitrogen starvation. It will be the role of future studies to investigate potential biochemical interactions between NMR, NIT-4, and other downstream transcription factors.

Characterizing the regulatory network through which fungal cells control nitrogen utilization is important in understanding how fungi interact with their environment, including how saprophytic fungi recycle nutrients and the mechanisms by which fungi infect both plants and animals (1–3). Transcriptional profiling combined with transcription factor promoter binding data are a powerful tool to help elucidate regulatory networks, as it enabled us to make and test hypotheses addressing the role of transcription factors in nutrient sensing. We used these genomic tools to describe a network of transcription factors that regulate genes in response to changes in nitrogen conditions.

Studies in S. cerevisiae have shown that additional transcription factors, such as Gcn4 which regulates genes in response to amino acid starvation, are also involved in regulating nitrogen catabolite repression (59). Although we did not see evidence of direct regulation of the GCN4 ortholog, cpc-1 (NCU04050), by NIT-2, NIT-4, or AMN-1 in the conditions tested here, we expect future studies to place cpc-1 and other nitrogen regulators in the nitrogen sensing network. Additionally, our data indicating a role for AMN-1 in regulating genes involved in both mannose and amino acid utilization suggest a connection between nitrogen and carbon metabolism (Fig. 4 and SI Appendix, Fig. S1). Previous studies have suggested that NIT-2 and its ortholog AreA may also play a role in plant cell wall deconstruction (60, 61). Future studies dissecting the connections between carbon and nitrogen sensing networks will elucidate how fungal cells respond to and utilize nutrient sources in a variety of environments.

Materials and Methods

N. crassa Strains and Culturing.

Strains used in this study are listed in SI Appendix, Table S2. All strains were derived from the wild-type reference strain FGSC 2489 using standard genetic techniques and confirmed by PCR and DNA sequencing (62, 63). N. crassa cultures were grown on Vogel’s minimal medium (VMM) (64). In all experiments in which the nitrogen source was not ammonium nitrate, Vogel’s salts (64) were made without ammonium nitrate, and nitrogen sources were added as described in SI Appendix, Table S1. Thus, the indicated nitrogen source was the only nitrogen source present in the experiment. Nitrogen starvation indicates that Vogel’s salts were made without ammonium nitrate, and no nitrogen source was added. Carbon sources were added at 2% wt/vol, and nitrogen sources were added at 50 mM unless otherwise noted. Specifics of the carbon and nitrogen sources used can be found in SI Appendix, Table S1. Cells were grown from freezer stocks on VMM + sucrose + 1.5% agar (Thermo Fisher Scientific) slants for 2 d at 30 °C in the dark and 4 to 8 d at 25 °C in constant light prior to inoculation into the indicated media at 10⁶ conidia/mL. All chemicals were purchased from Sigma-Aldrich unless otherwise noted.

Gene Expression Analysis, RNAseq, and Transcript Abundance.

RNA extraction, library preparation, and sequencing were modified from Wu et al. (2020) (17) qRT-PCR was performed using the EXPRESS One-Step SYBR GreenER kit (Life Technologies). For details, see SI Appendix, SI Materials and Methods.

Statistical Significance Tests.

For RNAseq data, experiments had at least three biological replicates, and statistical significance was determined using Cufflinks version 2.2.1 (65). The exact number of replicates for all RNAseq experiments is shown in Dataset S1 and SI Appendix, Table S3. For all growth and qRT-PCR experiments, at least three biological replicates were done. The exact number of replicates for all growth and qRT-PCR experiments is shown in Dataset S2. Biological replicates refer to independent cultures inoculated on the same or independent days. Statistical significance was determined using a two-tailed homoscedastic (equal variance) Student’s t test with a Benjamini–Hochberg multiple hypothesis correction. In bar graphs, bars indicate the mean of biological replicates and dots indicate individual biological replicates.

DAPseq.

DAPseq was done as described in Wu et al. (2020) (17). For details, see SI Appendix, SI Materials and Methods.

DNA Binding Consensus Motif Generation.

Motif discovery was performed using Multiple Expectation maximizations for Motif Elicitation (MEME) version 5.1.0 (66). Identification of motif locations was performed using Find Individual Motif Occurrences (FIMO) v5.2.0 (67) For details, see SI Appendix, SI Materials and Methods.

Mycelial Dry Weight.

To measure mycelial dry weight, 10⁶ conidia/mL were inoculated in liquid VMM with the indicated carbon or nitrogen source either in 100 mL cultures in 250 mL flasks or 3 mL cultures in round-bottomed 24-well plates. Media in which the carbon source is specified had 25 mM ammonium nitrate as the nitrogen source. Media in which the nitrogen source is specified had 2% sucrose as the carbon source. Cells were grown at 25 °C in constant light with constant shaking at 200 rpm. The mycelial cell mass was harvested onto Whatman Grade 1 filter paper and dried by vacuuming away the media. Mycelial cell masses were then further dried in a 65 °C drying oven for 2 d prior to weighing the dried mycelia.

Functional Enrichment Analysis and Gene Annotation.

Functional enrichment analysis was done using the FungiFun2 online resource tool (https://elbe.hki-jena.de/fungifun/) with KEGG as the classification ontology (22, 25). The gene to category associations were tested for overrepresentation using hypergeometric distribution with Benjamini–Hochberg correction for false discovery rate.

Gene annotations were pulled from FungiDB (https://fungidb.org/) or inferred from homology to characterized genes in related fungi.

Data Availability

RNAseq data used in this study were deposited in the Gene Expression Omnibus (GEO) at the National Center for Biotechnology Information (NCBI) and are accessible through GEO series accession number GSE150256 (68). Processed RNAseq data are available in Dataset S1. DAPseq data used in this study were deposited in the NCBI Sequence Read Archive (SRA) and are accessible through SRA series accession number PRJNA436200 (69). Processed DAPseq data are available in Dataset S3. The numerical values used to generate all mycelial dry weight and qRT-PCR graphs are shown in Dataset S2. Strains constructed in this study are available from the Fungal Genetics Stock Center (www.fgsc.net).