Abstract
The fl (fluffy) gene of Neurospora crassa encodes a binuclear zinc cluster protein that regulates the production of asexual spores called macroconidia. Two other genes, acon-2 and acon-3, play major roles in controlling development. fl is induced specifically in differentiating tissue during conidiation and acon-2 plays a role in this induction. We examined the function of fl by manipulating its level of expression in wild-type and developmental mutant strains. Increasing expression of fl from a heterologous promoter in a wild-type genetic background is sufficient to induce conidiophore development. Elevated expression of fl leads to induction of development of the acon-2 mutant in nitrogen-starved cultures, but does not bypass the conidiation defect of the acon-3 mutant. These findings indicate that fl acts downstream of acon-2 and upstream of acon-3 in regulating gene expression during development. The eas, con-6, and con-10 genes are induced at different times during development. Morphological changes induced by artificially elevated fl expression in the absence of environmental cues were correlated with increased expression of eas, but not con-6 or con-10. Thus, although inappropriate expression of fl in vegetative hyphae is sufficient to induce conidial morphogenesis, complete reconstitution of development leading to the formation of mature conidia may require environmental signals to regulate fl activity and/or appropriate induction of fl expression in the developing conidiophore.
NEUROSPORA crassa grows rapidly to colonize the local environment. Following initial colonization, the mat of fungal mycelium gives rise to aerial hyphae that produce asexual spores called macroconidia. The macroconidia provide a means for rapid wind and splash dispersal to new sites of colonization. In the laboratory, macroconidiation (hereafter, conidiation) is induced when the fungus becomes desiccated and when it is starved for carbon or nitrogen (Turian and Bianchi 1972; Turian 1973).
Several genetic loci that are specifically required for conidiation have been identified. Two mutants, aconidiate-2 (acon-2) and fluffyoid (fld), produce aerial hyphae but fail to initiate the transition from filamentous to budding growth due to an early block in development (Matsuyamaet al. 1974; Springer and Yanofsky 1989). Two additional mutant strains, aconidiate-3 (acon-3) and fluffy (fl), are blocked during the transition to budding growth (Lindegren 1933; Matsuyamaet al. 1974). The flY allele is partially active and displays a phenotype similar to fld. The conidial separation mutants, csp-1 and csp-2, produce chains of budded cells but these do not mature to release free conidia (Selitrennikoffet al. 1974).
In addition to these loci that are involved in conidiophore development, other genes that are preferentially expressed during conidiation have been identified (Berlin and Yanofsky 1985; Robertset al. 1988; Roberts and Yanofsky 1989; Springer and Yanofsky 1992). The functions of one class of these genes (con genes) are unknown, but it appears that in some cases they encode stress-response proteins (Lee and Ebbole 1998). The eas gene is also regulated in a complex manner and encodes a hydrophobin that is found on the outer surface of aerial hyphae and conidia. In addition to developmental control, many of these genes are regulated directly by other factors such as light, the circadian clock, and nutrient availability (Lauter and Yanofsky 1993; Corrochanoet al. 1995; Bell-Pedersen 1998; Lee and Ebbole 1998).
The eas, con-6, and con-10 genes also are repressed during vegetative mycelial growth. This repression is mediated by rco-1 (Yamashiroet al. 1996). rco-1 is a homolog of Saccharomyces cerevisiae TUP1, a gene involved in transcriptional repression of several classes of genes (Keleheret al. 1992). rco-1 mutants are female sterile, grow more slowly than the wild type, and have a conidial maturation defect similar to that of csp-1 and csp-2 mutants.
fl is the only regulatory gene of those described above that has been characterized. fl encodes a binuclear zinc cluster protein that resembles the Gal4 class of transcription factors (Bailey and Ebbole 1998). fl is induced during synchronous development caused by nitrogen starvation or by transfer of the mycelium from liquid culture to air. Induction occurs at the time corresponding to the transition from filamentous to budding growth (Bailey and Ebbole 1998).
To better assess the genetic requirements for fl expression, we examined expression of fl in different tissues and in developmental mutants. fl is expressed at a basal level in vegetative hyphae and expression is induced during conidiation specifically in differentiating conidiophores. Induction of fl expression was observed in the developmental mutants fld and acon-3 under nitrogen-starvation conditions. These induced levels are not seen in the acon-2 and flL (null allele) strains. A heterologous promoter was used to elevate expression of fl in wild-type and developmental mutant strains. Wild-type strains expressing high levels of fl mRNA from a constitutive promoter sporulated in the absence of environmental cues inducing development. We measured the expression patterns of con genes in developmental mutants expressing fl to clarify the previously developed time line relating gene expression with stages of morphogenesis (Lauter and Yanofsky 1993). This study supports the view that acon-2 precedes fl and that acon-3 acts downstream of fl to regulate gene expression and morphogenesis.
MATERIALS AND METHODS
Fungal strains: From the Fungal Genetics Stock Center (FGSC), Department of Microbiology, University of Kansas Medical Center, we obtained strains acon-2 a (FGSC no. 3263), acon-3 a (FGSC no. 5074), fld a (FGSC no. 7023), flL a (FGSC no. 46), csp-1 a (FGSC no. 2555), 74OR23-1VA (FGSC no. 2489), and 74-OR6a (FGSC no. 4200). The acon-2 allele is temperature sensitive (conidiating at 25° but not at 34°). The rco-1 strain (RCO353) has been described previously (Lee and Ebbole 1998).
Growth conditions: To examine gene expression during aerial development, Vogel’s minimal medium N with 1.5% sucrose as carbon source (Davis and de Serres 1970) was inoculated with 1 × 105 conidia/ml of strain 74-OR23-1VA. After 20 hr growth at 34° (200 rpm), the culture was harvested onto 90-mm diameter circles of Whatman no. 1 filter paper. The harvested pads were inverted on top of a second filter paper on medium N with 0.45% agar. The cultures were placed at 34° in the light. The mycelium must first grow through filter paper to gain exposure to air, and aerial hyphae are first observed 7 hr after harvesting. Synchronous conidiation is obtained under these conditions and budding growth occurs 6 hr after aerial hyphae are first observed (13 hr after harvesting). The upper filter paper layer was peeled away and the aerial growth was harvested from the surface of the filter paper with a razor blade. The remaining mycelium was harvested separately for RNA isolation as described (Springeret al. 1992).
To examine nitrogen-starvation-induced development, strain 74-OR23-1VA was inoculated at a concentration of 1 × 105 conidia/ml into four 250-ml flasks containing 100 ml of modified medium N with 50 mm NH4Cl as the nitrogen source and grown for 20 hr. The cultures were then washed twice and each 100-ml culture was divided among three flasks containing medium N either without an added nitrogen source for nitrogen starvation or with 50 mm NH4Cl or 50 mm KNO3 as the nitrogen source. Mycelia from each of the different media were collected after 1, 6, 12, and 24 hr for RNA extraction.
Mycelial fragments were used as inoculum in experiments involving aconidial strains. Initial cultures were inoculated with a loop of aerial hyphae and grown for 20 hr prior to fragmention in a Waring blender (two times 15 sec). Twenty-milliliter aliquots were transferred into 50 ml of fresh medium, grown for 20 hr, and this step was repeated to increase the mycelial mass. Cultures were then washed twice with sterile water, harvested, and divided into six equal portions. One of the portions was frozen immediately, and the remaining five portions were used to inoculate five 125-ml flasks containing 50 ml of medium N without nitrogen source. Samples were harvested after 4, 6, 8, 10, and 12 hr of incubation, frozen with liquid nitrogen, and used for RNA extraction.
RNA extraction and analysis: RNA extraction and Northern blot analyses were performed as previously described (Madiet al. 1994). A 1.5-kb SpeI fragment located within the coding region of the fl gene was used as the probe in these experiments. Probes for eas, con-10, and con-6 were prepared from cDNA fragments isolated from plasmids pEAS, pBW100, and pCON6-6, respectively (Madiet al. 1994). A cosmid containing the rRNA repeat region was used to prepare a probe to detect ribosomal RNA. The N. crassa actin cDNA (Tinsleyet al. 1998) was provided by M. Plamann (University of Missouri, Kansas City). Because actin mRNA is subject to significant regulation at the mRNA level, we used actin and ribosomal RNA as a means to assess RNA quality and as a rough guide to assess RNA loading for RNA blot experiments. For this reason we did not attempt to quantitate levels of induction of gene expression. However, all hybridizations for a given probe were performed at the same time and the membranes were exposed to film for the same lengths of time and qualitative estimates of gene expression levels were consistent in replicate experiments.
Plasmid construction and transformation: A plasmid containing the fl gene under the control of the cross-pathway control-1 (cpc-1) promoter and the trpC terminator was used to examine the effect of increased fl expression. This plasmid was constructed in three steps. First, pLBS5 was constructed by ligating an ∼0.95-kb EcoRI-HindIII fragment containing the trpC terminator from pCZT (Madiet al. 1994) into pCB1004 (Carrollet al. 1994) at the EcoRI-HindIII site. An N. crassa cpc-1 promoter fragment was amplified from genomic DNA using a primer (5′-TTGGCACATGGAATGGAC-3′) that is upstream of an NheI site and a primer (5′-GTAAGCTAGCTTG ATGGATGCTTCTCG-3′), containing an incorporated NheI site, that is downstream of the transcriptional start site of cpc-1. The 1.2-kb amplified fragment containing the cpc-1 promoter was digested with NheI and then cloned into the SpeI site of pLBS5. This fragment was sequenced to verify that no mutations were introduced into the DNA during amplification. The resulting plasmid was named pLBS6 and was used as the control plasmid in these experiments. The fl fragment was also amplified from genomic DNA using a primer (5′-CTC TCCCGGGCTTCCAGGTTGCTTCAGA-3′) upstream of the translational start site (nucleotides 453–470) and a primer (5′-ACCGCCCGGGCAAATGCCTTAGGTGAAC-3′) downstream of the last codon (nucleotides 3269–3286; Bailey and Ebbole 1998). This fragment was cloned and sequenced to verify its integrity. The fl gene was excised by digestion with SmaI and then ligated into the SmaI site of pLBS6. The resulting plasmid, pLBS7, was used to overexpress fl in our experiments.
pLBS6 and pLBS7 were used to transform protoplasts of 74OR23-IVA, acon-2, acon-3, and flL strains (Vollmer and Yanofsky 1986). Homokaryotic strains were obtained by serial passage on medium containing 200 μg/ml hygromycin. The naming convention for overexpression strains designates the parental strain, the plasmid used for transformation, and the number of individual transformants (e.g., 74-LBS7-15 is pLBS7 transformant no. 15 of strain 74OR23-IVA).
Overexpression experiments: Minimal medium was inoculated with conidia of the transformed wild-type strains (74-LBS7-5, 74-LBS7-7, 74-LBS7-15, and 74-LBS6-1) and the transformed acon-2- strains (ACON2-LBS7-4, ACON2-LBS7-10, ACON2-LBS7-13, and ACON2-LBS6-4). These strains were grown for 16 hr, harvested, and washed with 1× medium N salts without nitrogen source. The cultures were then divided, and half of the sample was transferred into fresh medium N and the other half was transferred to medium N without nitrogen source. These cultures were grown for 12 hr, harvested, and frozen with liquid nitrogen for RNA preparation.
Overexpression experiments with aconidial strains were performed by inoculating minimal medium with the transformed flL strains (FL-LBS7-2, FL-LBS7-12, and FL-LBS6-13) and the transformed acon-3 strains (ACON3-LBS7-1, ACON3-LBS7-3, and ACON3-LBS6-1). These cultures were grown for 20 hr and then blended with a Waring blender. Ten milliliters of the blended culture was used to inoculate fresh minimal medium. These cultures were grown for 16 hr, harvested, washed, and divided into medium N and medium N without nitrogen source as described for the wild-type and acon-2 transformants.
RESULTS
fl expression during aerial development and nitrogen starvation: We showed previously that expression of fl is induced at approximately the time that budding growth initiates during conidiation (Bailey and Ebbole 1998). Chains of conidia develop at the tips of aerial hyphae and not within the mycelial mat from which these aerial hyphae emerge. Therefore, we examined the spatial pattern of fl expression to determine if the increase in fl mRNA occurs specifically in aerial hyphae.
Aerial hyphae and the underlying mycelium were harvested separately at different stages of development (see materials and methods). Basal levels of fl mRNA were maintained in the mycelial samples throughout development (Figure 1). Induction of fl mRNA in the aerial hyphae was observed early in development, when the majority of the tips of aerial hyphae had not yet differentiated and only a fraction of these initiated budding growth. con-6 and con-10 expression was also limited to the developing aerial hyphae (Figure 1). These results are consistent with those previously observed with con-10 (Springeret al. 1992). Relatively high levels of con-6 expression were reached by 13 hr, a time when budding growth is just commencing. con-10 expression was still limited at this time but is found in greater abundance at later time points and when production of free conidia is observed. Thus, induction of fl expression is conidiation specific in the aerial hyphae where conidiation-specific genes are expressed. Actin mRNA is typically used as a control for a gene with constant expression levels. However, steady-state actin mRNA levels are regulated during aerial development with mRNA levels declining in the mycelium and rising in the rapidly growing aerial hyphae. Although actin is appropriate as a loading control for aerial development in this experiment, the combination of actin and rRNA provides a good estimate of RNA loading and quality.
—Localization of fl, con-10, and con-6 induction to aerial hyphae. N. crassa was grown for 20 hr and harvested to induce synchronous conidiation. Aerial hyphae (A) and mycelium (M) samples were harvested separately for RNA analysis. Photographs were taken of aerial hyphae samples at two stages during development. Samples taken early in development (13 hr) had swollen tips with limited initiation of budding growth. Samples taken late in development (20 hr) contained chains of budding cells and free conidia. T = 0 designates the time of harvest from liquid cultures. Actin (act-1) mRNA and rRNA levels are shown to assess RNA quality and loading.
Uniform conidiation was also observed to commence in liquid cultures 6 hr after transfer to medium lacking a nitrogen source (not shown) and a strong induction of fl mRNA was observed by 6 hr (Figure 2A). fl expression was detected in low abundance in samples that were grown in media containing ammonium or nitrate as the nitrogen source. Examination of the time course of gene expression in medium lacking a nitrogen source showed that fl mRNA levels begin to rise by 2 hr and are strongly induced by 4 hr (Figure 2B). No increase in fl mRNA was detected at 1 hr of incubation; however, eas mRNA levels increased significantly by 1 hr. con-6 was induced by nitrogen-starvation-induced development starting ∼4 hr after inoculation and con-10 was induced by 6 hr (Figure 2B). Induction of eas, con-6, and con-10 was dependent on fl (see below) and was not a response to nitrogen starvation.
Expression of fl in developmental mutants: Since fl mRNA levels are induced during development, we examined fl expression in developmental mutants to examine epistatic relationships. Six genes (acon-2, fl, fld, acon-3, csp-1, and csp-2) that are required for conidial development have been identified. Only fl has been characterized and is known to have null alleles (such as flL) and a partially active allele, flY (Bailey and Ebbole 1998). Another regulator of conidiation-specific genes, rco-1, encodes a cell-type-specific repressor of ccg-1 and con-10. Shortly after induction of development, repression of the con-10 gene mediated by rco-1 is no longer detected (Lee and Ebbole 1998).
—Induction of gene expression in response to nitrogen source. (A) Expression of fl was determined in strains grown in medium N containing 50 mm ammonium chloride (NH4), 50 mm potassium nitrate (NO3), or no nitrogen source (no N). RNA for Northern blot analysis was isolated from samples taken 0, 1, 6, 12, and 24 hr after inoculation. act-1 (actin) was used as a probe to assess mRNA loading. (B) Expression of fl, eas, con-6 and con-10, and act-1 at hourly intervals during the first 6 hr after transfer to nitrogen-starvation medium.
The basal level of fl mRNA (0 hr) in the acon-2 and flL mutants was similar to wild type (Table 1, Figure 3). After transfer to medium lacking a nitrogen source, induction of fl mRNA was reduced in the acon-2 and flL mutants. This suggests that acon-2, and possibly fl, contributes to induction of fl mRNA. We note, however, that induction of fl mRNA in the flY mutant was comparable to that in the wild type (Table 1). Since the csp-1 mutation disrupts development at a very late stage of conidiophore maturation, it is not surprising that fl expression in this mutant was similar to the wild type (Table 1). The basal level of fl expression was not elevated in the rco-1 mutant, indicating that fl is not subject to repression by rco-1 and induction of fl was not markedly altered. Induction of fl was observed in the fld and acon-3 strains, suggesting that these genes do not act to regulate fl transcription (Figure 3).
con-6, con-10, and eas expression: We also assessed the roles of acon-2, acon-3, fld, fl, csp-1, and rco-1 genes in regulating eas and con gene expression by Northern blot analysis. As expected for csp-1, the expression patterns of eas, con-6, and con-10 resembled those of the wild type (Table 1, Figures 1 and 3). eas, con-6, and con-10 expression is not appreciably induced in the flL or acon-2 strains, although a small amount of con-10 expression was detected in the flL strain in this experiment. Delayed expression of eas was observed in the flY, acon-3, and fld strains (Table 1, Figure 3). con-6 and con-10 were not expressed in the acon-3 strain and reduced levels of con-6 and con-10 mRNA were observed in the fld strain (Table 1, Figure 3). Expression of these genes in the wild type and of these developmental mutants was also assessed during aerial development (data not shown) and found to be similar both to previously published results (Robertset al. 1988; Lauteret al. 1992; White and Yanofsky 1993) and to our results with nitrogen-starvation-induced cultures, with one exception. When we examined a time course of eas expression in cultures undergoing aerial development, we found that eas expression was induced in the acon-3 strain similarly to the wild-type strain, whereas eas expression is delayed in the nitrogen-starved acon-3 strain (Figure 3). Other than this exception, the epistatic relationships between these regulatory genes and the timing of eas, con-6, and con-10 expression during nitrogen-starvation-induced conidiation are similar to those in studies examining aerial development.
Gene expression patterns in developmental mutants
In the absence of rco-1, a number of nutrient and environmental stress conditions were found to regulate con-10 expression independently of developmental induction (Lee and Ebbole 1998). Basal levels of eas, con-6, and con-10 transcripts were elevated in the rco-1 sample (0 hr; Figure 3). Thus, eas and con-6 also appear to be subject to rco-1-mediated repression. It is unknown why con-6 expression declined after transfer to nitrogen-starvation conditions in the rco-1 mutant.
Elevated expression of fl induces conidiation: Because we observed elevated levels of fl mRNA in aerial hyphae during conidiophore formation, we wanted to determine if elevated expression of fl in vegetative hyphae was sufficient to induce conidiation. We used the cpc-1 promoter to drive expression of fl in these experiments. The cpc-1 gene is moderately expressed during germination and mycelial growth with reduced levels of expression during conidiation (Ebboleet al. 1991; Paluh and Yanofsky 1991; Sachs and Yanofsky 1991). A 1.2-kb fragment of the cpc-1 promoter was used to create a transcriptional fusion to drive expression of the fl gene. This plasmid, pLBS7, was used to generate a series of strains bearing single ectopically integrated copies of the plasmid. The vector lacking the fl gene was used as a control (pLBS6).
—Gene expression patterns in developmental mutants. Developmental mutants (csp-1, rco-1, fld, acon-3, acon-2, and fl) were inoculated into medium N for 20 hr. The mycelia were then harvested, washed, and transferred to medium N with no nitrogen source. Samples were harvested immediately (0) or incubated for 4, 8, and 12 hr prior to RNA isolation and Northern blot analysis using fl, eas, con-6, con-10, and act-1 as probes. rRNA bands from ethidium-bromide-stained agarose gels are shown to assess RNA loading and quality.
We examined 10 independent isolates of 74-ORS6a transformed with pLBS7. Of the 10 strains, 8 produced elevated levels of fl transcript (not shown), indicating integration of intact copies of the gene expressed from the cpc-1 promoter. These eight pLBS7 transformants produced conidiophores in liquid culture without starvation. Conidiophore development was observed in minimal medium after a 10-hr period of incubation but was more pronounced with longer incubation or nitrogen starvation. None of the control strains transformed with pLBS6 conidiated in minimal medium and all conidiated in medium lacking nitrogen, as was the case for the wild type. Strains 74-LBS6-1 and 74-LBS7-15 are representative of these transformed wild-type strains (Figure 4). We noted no deleterious effects of fl expression on growth rate of the transformants.
The flL strain that was transformed with pLBS7, FL-LBS7-2, produced conidia on solid minimal medium, indicating that the fl mutation was complemented by the plasmid (not shown). Conidiophore development in the FL-LBS7-2 strain was also observed when it was grown in liquid minimal medium and nitrogen-starvation medium (Figure 5). In contrast to the 74-LBS7-15 strain, the conidiation in the nitrogen-starved culture resembled that of the minimal medium culture and free conidia were not present at 10 hr. The negative control strain, FL-LBS6-13, was aconidial on solid media and it produced only vegetative hyphae when it was grown in nitrogen-starvation medium (Figure 5).
The acon-2 transformant containing pLBS7, ACON2-LBS7-4, conidiated when it was grown on solid medium at the nonpermissive temperature, while the control strain (ACON2-LBS6-15) was aconidial (not shown). When ACON2-LBS7-4 was grown in minimal medium, it did not produce conidiophores but displayed a slightly irregular morphology (Figure 6). The ACON2-LBS7-4 transformant produced conidiophores and free conidia when it was grown in the nitrogen-starvation medium. Thus, with exposure to air or nitrogen starvation the acon-2 mutation could be bypassed by the cpc-1-fl fusion gene. The ACON2-LBS6-15 negative control strain grew as vegetative hyphae under these conditions (Figure 6).
—Heterologous expression of fl induces conidiophore development in the wild-type genetic background. Transformants of strain 74-ORS6a containing pLBS6 (74-LBS6-1) and pLBS7 (74-LBS7-15) were transferred to medium N (+N) and medium lacking a nitrogen source (-N) for 10 hr. Cultures shown in Figures 4, 5, 6, 7 were harvested for RNA preparation after 12 hr growth in minimal medium and subjected to Northern blot analysis using fl, eas, con-6, con-10, and rRNA probes.
—Heterologous expression of fl in the flL genetic background. Transformants of the flL strain containing pLBS6 and pLBS7 were transferred to medium N (+N) and medium lacking a nitrogen source (-N) for 10 hr.
The acon-3 strain containing the fl expression plasmid (ACON3-LBS7-3) was indistinguishable from the control strain when it was grown in minimal medium. This was also typical of growth in nitrogen-starvation medium (not shown). ACON3-LBS7-3 did produce conidiophores on solid medium after a long delay (2–4 days; not shown), indicating that elevated expression of fl can eventually bypass this allele of acon-3. However, the conidiophores displayed a moderate conidial separation defect indicative of slow or incomplete maturation. The ACON3-LBS6-1 strain did not produce conidia under any condition.
Gene expression in pLBS7 transformants: Strain 74-LBS7-15 produced elevated levels of fl mRNA that exceeded the level found in 74-LBS6-1, even following induction by nitrogen starvation (Figure 7). In the wild-type strain transformed with either pLBS6 or pLBS7, we found that the eas, con-6, and con-10 genes were induced in the nitrogen-starved cultures that displayed abundant conidiation (Figure 4). However, in the 74-LBS7-15 strain grown in minimal medium, eas was strongly induced, while con-6 and con-10 expression was not detected (Figure 7). This result is surprising since the culture appeared to have reached a morphological stage where con-6 and con-10 expression would be anticipated (Figures 1 and 4).
—Heterologous expression of fl in the acon-2 genetic background. Transformants of the acon-2 strain containing pLBS6 and pLBS7 were transferred to medium N (+N) and medium lacking a nitrogen source (-N) for 10 hr.
The FL-LBS6-13 control strain produced a basal level of fl expression that was not markedly elevated in nitrogen-starvation medium and, as expected, this fl mutant did not express any of the conidiation-induced genes. The level of fl expression in the FL-LBS7-2 cultures was significantly elevated over that in the FL-LBS6-13 strain (Figure 7). In FL-LBS7-2, eas was expressed under both culture conditions, but significant levels of con-6 or con-10 were not induced even with nitrogen starvation (Figure 7).
As expected for the control strain, ACON2-LBS6-4, there was no increase in fl mRNA levels during nitrogen starvation and development was not observed. fl mRNA levels were elevated in the ACON2-LBS7-10 strain (Figure 7). The requirement for acon-2 to activate eas is completely bypassed by elevated fl mRNA levels. However, even with nitrogen starvation, little con-6 or con-10 expression was observed. Thus, gene expression patterns in ACON2-LBS6-4 and ACON2-LBS7-4 were very similar to those of the corresponding fl mutant strains (Figure 7).
For the ACON3-LBS6-1 control strain, nitrogen starvation induced fl and eas expression but not con-6 or con-10 expression. In the ACON3-LBS7-3 strain, expression of the cpc-1/fl fusion led to expression of eas in minimal medium and in nitrogen-starvation medium but expression of con-6 and con-10 was not observed (Figure 7).
—Effect of heterologous expression of fl on gene expression in wild-type and developmental mutants.
DISCUSSION
fl is expressed at a low level in mycelia, but is induced in the developing aerial hyphae after the initiation of conidiation. Our current model supposes that FL is a transcription factor since the closest homologs are related to the Gal4p class of transcriptional activators. fl is required for the induction of con-6 and con-10 that occurs at the time at which fl mRNA levels are induced. Since acon-3 is required for con-6 and con-10 induction, even in the presence of the cpc-1-fl fusion gene, fl either is required to activate acon-3 or works with acon-3 directly to induce con-6 and con-10.
Nitrogen-starvation-induced development was used in most of our experiments because of the strong synchrony of development. This also provided an opportunity to compare gene expression during nitrogen-starvation-induced development with gene expression during aerial development since expression of con-6 and con-10 in aerial conidiation has been characterized in developmental mutants previously (Robertset al. 1988; White and Yanofsky 1993). The most significant difference was that eas expression was delayed in nitrogen-starved cultures of the acon-3 mutant whereas no delay in eas expression was observed in the acon-3 mutant during aerial development. This may reflect the interplay of developmental and environmental regulation of gene expression during conidiation.
fl is required for eas expression during development. However, eas mRNA levels increase more rapidly than those of fl mRNA, suggesting that the basal level of fl expression in vegetative hyphae is sufficient to induce eas in the presence of appropriate environmental cues. This suggests the possibility of a post-transcriptional step for regulating fl protein activity in vegetative hyphae. Since acon-2 is also required for eas expression in wild-type cells, one possibility is that acon-2 protein is responsible for activating the basal level of FL activity in response to environmental cues. The finding that eas is expressed in the acon-2 mutant bearing the cpc-1-fl fusion is not inconsistent with this view since increased fl expression may increase total FL activity and bypass the requirement for acon-2 (and environmental cues) with respect to eas expression. Several members of the binuclear zinc cluster family of transcription factors are responsive to nutritional conditions (Wanget al. 1999). Alternatively, a small but undetectable increase in fl mRNA may be responsible for induction of eas early during development.
—Proposed role of regulators in development and gene expression. RCO1 (a homolog of S. cerevisiae Tup1p) functions in vegetative hyphae to repress expression of the developmentally regulated genes eas, con-6, and con-10. acon-2 functions to increase fl activity by promoting transcription of the fl gene. Nitrogen-starvation medium triggers a rapid FL- and ACON2-dependent induction of eas. Increased expression of fl from a heterologous promoter is sufficient to induce eas expression in the absence of ACON2 or nitrogen starvation. fl expression triggers the switch to conidiophore development and budding growth. ACON3 is required for full entry into budding growth and is for induction of con-6 and con-10. Tissue-specific induction of fl expression in developing conidiophores may be important for proper induction of con-6 and con-10 since wild-type levels of con-6 and con-10 gene expression are observed only when the endogenous fl gene is present in strains that are also undergoing nitrogen starvation.
Our data suggest a model in which fl is responsible for induction of eas early in development (Figure 8). Because of the rapid response of eas to developmental signals and the dependence on fl for expression, we suggest that eas is likely to be a direct target for the FL transcription factor. acon-3 appears to act downstream of fl to regulate con-6 and con-10. These genes are normally repressed in vegetative hyphae by rco-1. acon-2 is required for full transcriptional activation of fl, but as discussed above, acon-2 may also have other activities important for activation of the basal level of FL protein. The role of fld is unclear: it does not affect transcriptional induction of fl, but the delay in downstream target gene expression suggests that it is required for efficient activation of gene expression or is possibly required for derepression of eas, con-6, and con-10 (perhaps by affecting repression by rco-1).
Although our results confirm that fl plays a key role in developmental regulation, molecular analysis of the acon-2, fld, and acon-3 genes is required to define the regulatory pathway further. Recent progress in genome sequencing (Galaganet al. 2003) should greatly facilitate cloning of these key genes. Preliminary restriction fragment length polymorphism mapping experiments using markers near to where acon-2 has been previously mapped (ro-2, phe-2, tyr-1) show that acon-2 lies ∼5 cM away (on the basis of 80 segregating progeny). Unfortunately, the adjacent supercontig in the current physical map displayed no linkage to acon-2 (our unpublished observation).
There are aspects of fl function that we could not resolve with our experimental approach of expressing fl from the cpc-1 promoter. In addition to the ectopic copy of cpc-1-fl, 74-LBS7-15 has a functional endogenous copy of fl. The cpc-1 gene is expressed at lower levels during conidiation (Sachs and Yanofsky 1991), and thus the promoter does not provide elevated expression in developing conidiophores as does the endogenous fl promoter and this may explain why the development of FL-LBS7-2 appears to be significantly delayed relative to 74-LBS7-15 under nitrogen-starvation conditions. Additionally, some features of the normal developmental program may not be executed properly in the absence of appropriate environmental cues, for example, release of rco-1-mediated repression. This may explain why con-6 and con-10 expression is not observed in cultures that have reached a morphological stage at which con-6 and con-10 expresssion is always observed in wild-type strains.
In some respects fl plays a role in conidiation analogous to Aspergillus nidulans brlA (Adamset al. 1998). brlA is necessary for conidiophore morphogenesis and encodes a transcription factor that is induced during conidiation. Both fl and brlA are expressed constitutively at low levels and their induction is coincident with the onset of morphogenesis (Boylanet al. 1987). Forced expression of brlA using the alcA promoter in vegetative hyphae results in an incomplete form of conidiation where viable conidia bud from the tips of the hyphae grown in liquid cultures (Adamset al. 1988). This is similar to our observations in N. crassa with the cpc-1-fl fusion. However, overexpression of brlA causes a cessation of hyphal growth (Adamset al. 1988), while fl activates development without interfering with vegetative growth. fl expression does not, on its own, act as a switch between growth and development as is apparently the case for brlA.
Although there may be analogous roles for regulators of conidiation in these two fungal species, the regulators are not homologous. brlA encodes a Cys2-His2 zinc-finger protein in contrast to the binuclear zinc cluster encoded by fl. Furthermore, a homolog of A. nidulans flbD has been cloned from N. crassa, and this gene, rca-1, complements the sporulation defect of the A. nidulans mutant. However, the rca-1 mutant of N. crassa is not defective in spore development and plays a different role in the biology of the organism (Shenet al. 1998). The lack of homology in the regulatory genes involved in conidiation in these two fungi points to independent evolutionary origins for this developmental process. It seems likely that during the evolution of N. crassa and A. nidulans different members of the available pool of regulatory genes in the common ancestor were recruited to regulate the emerging developmental processes resulting in asexual sporulation.
Acknowledgments
This work was supported by National Science Foundation (NSF) grant MCB-9974608 and NSF training grant DGE-935-4891 to the Program for the Biology of Filamentous Fungi.
Footnotes
-
Communicating editor: M. Sachs
- Received May 9, 2003.
- Accepted September 8, 2003.
- Copyright © 2004 by the Genetics Society of America
