Aspergillus nidulans apsA (anucleate primary sterigmata) encodes a coiled-coil protein required for nuclear positioning and completion of asexual development.

Many fungi are capable of growing by polarized cellular extension to form hyphae or by isotropic expansion to form buds. Aspergillus nidulans anucleate primary sterigmata (apsA) mutants are defective in nuclear distribution in both hyphae and in specialized, multicellular reproductive structures, called conidiophores. apsA mutations have a negligible effect on hyphal growth, unlike another class of nuclear distribution (nud) mutants. By contrast, they almost completely block entry of nuclei into primary buds, or sterigmata (bud nucleation), produced during development of conidiophores. Failure of the primary sterigmata to become nucleated results in developmental arrest and a failure to activate the transcriptional program associated with downstream developmental steps. However, occasionally in mutants a nucleus enters a primary bud and this event relieves the developmental blockage. Thus, there is a stringent developmental requirement for apsA function, but only at the stage of primary bud formation. apsA encodes a 183-kD coiled-coil protein with similarity to Saccharomyces cerevisiae NUM1p, required for nuclear migration in the budding process.

Low repetitive DNA content in Aspergillus nidulans.

DNA-DNA reassociation experiments show that the genome of Aspergillus nidulans consists of approximately 97 to 98 percent unique and 2 to 3 percent reiterated sequences. The reiterated DNA sequences have a complexity of about 11,000 base pairs and are repeated approximately 60 times per haploid genome. Ribosomal RNA-DNA hybridization experiments indicate that most of the repetitive DNA codes for ribosomal RNA.

Genetic engineering of filamentous fungi.

Filamentous fungi are important in medicine, industry, agriculture, and basic biological research. For example, some fungal species are pathogenic to humans, whereas others produce beta-lactam antibiotics (penicillin and cephalosporin). Industrial strains produce large amounts of enzymes, such as glucoamylase and proteases, and low molecular weight compounds, such as citric acid. The largest and most economically important group of plant pathogens are fungi. Several fungal species have biological properties and genetic systems that make them ideally suited for basic biological research. Recently developed techniques for genetic engineering of filamentous fungi make it possible to alter their detrimental and beneficial activities in novel ways.

Temporal and spatial controls of Aspergillus development.

The mechanisms regulating elaboration of the multicellular asexual reproductive apparatus, the conidiophore, of the filamentous ascomycete Aspergillus nidulans provide a model for the control of fungal development. Recent advances have been made on three fronts. First, new physical and chemical signals have been discovered that control commitment of cells to the conidiation pathway. Second, positively acting developmental regulatory genes have been cloned and characterized. In addition, evidence has been obtained for the existence of negatively acting regulatory loci. Finally, an anonymous developmentally regulated gene has been used to make a directed mutation that has revealed the morphogenetic function of the gene.

The Aspergillus nidulans abaA gene encodes a transcriptional activator that acts as a genetic switch to control development.

The Aspergillus nidulans abaA gene encodes a protein containing an ATTS DNA-binding motif and is required for the terminal stages of conidiophore development. Results from gel mobility shift and protection, missing-contact, and interference footprint assays showed that AbaA binds to the sequence 5'-CATTCY-3', where Y is a pyrimidine, making both major- and minor-groove contacts. Multiple AbaA binding sites are present in the cis-acting regulatory regions of several developmentally controlled structural genes as well as those of the upstream regulatory gene brlA, the downstream regulatory gene wetA, and abaA itself. These cis-acting regulatory regions confer AbaA-dependent transcriptional activation in a heterologous Saccharomyces cerevisiae gene expression system. From these observations, we propose that the AbaA transcription factor establishes a novel set of feedback regulatory loops responsible for determination of conidiophore development.

Functional organization of the Aspergillus nidulans trpC promoter.

We investigated the functional organization of the Aspergillus nidulans trpC promoter by the sequential removal of sequences upstream of the major trpC mRNA cap site (+1). DNA fragments containing promoter mutations were fused to the Escherichia coli lacZ gene, and a novel method was used to select for integration of the fusion gene at the Aspergillus argB locus. beta-Galactosidase assays and S1 nuclease protection experiments demonstrated that the promoter mutations affected gene expression in three ways: (i) 5' deletions up to -82 resulted in variable increases in beta-galactosidase activity, depending on the growth conditions; (ii) a deletion from -67 to -11 did not alter the level of beta-galactosidase activity, but did give rise to mRNAs with aberrant 5' ends; and (iii) a 5' deletion with an endpoint at -11 and an internal deletion from -142 to -11 abolished gene expression. These results indicate that sequences upstream of -82 reduce transcription of the trpC gene and that distinct DNA sequence elements are required for expression versus correct initiation of transcription of the trpC gene. The sequences essential for trpC expression do not include the common eucaryotic promoter elements CCAAT and TATAAA. To our knowledge, this is the first functional analysis of a promoter from a fungus other than Saccharomyces cerevisiae.

Aspergillus nidulans wetA activates spore-specific gene expression.

The Aspergillus nidulans wetA gene is required for synthesis of cell wall layers that make asexual spores (conidia) impermeable. In wetA mutant strains, conidia take up water and autolyze rather than undergoing the final stages of maturation. wetA is activated during conidiogenesis by sequential expression of the brlA and abaA regulatory genes. To determine whether wetA regulates expression of other sporulation-specific genes, its coding region was fused to a nutritionally regulated promoter that permits gene activation in vegetative cells (hyphae) under conditions that suppress conidiation. Expression of wetA in hyphae inhibited growth and caused excessive branching. It did not lead to activation of brlA or abaA but did cause accumulation of transcripts from genes that are normally expressed specifically during the late stages of conidiation and whose mRNAs are stored in mature spores. Thus, wetA directly or indirectly regulates expression of some spore-specific genes. At least one gene (wA), whose mRNA does not occur in spores but rather accumulates in the sporogenous phialide cells, was activated by wetA, suggesting that wetA may have a regulatory function in these cells as well as in spores. We propose that wetA is responsible for activating a set of genes whose products make up the final two conidial wall layers or direct their assembly and through this activity is responsible for acquisition of spore dormancy.

Upstream elements repress premature expression of an Aspergillus developmental regulatory gene.

The Aspergillus nidulans abaA gene regulates intermediate steps in asexual reproductive development and is itself developmentally regulated. An 822-base-pair DNA fragment from the abaA 5'-flanking region is sufficient to drive developmentally appropriate expression of the Escherichia coli lacZ gene. Deletion analysis showed that this fragment contains elements that repress transcription in vegetative cells and immature conidiophores and that activate transcription later during development. A 45-base-pair region encompassing the major and minor abaA transcription initiation sites contains directly repeated sequences related to the mammalian initiator (Inr) element (S. T. Smale and D. Baltimore, Cell 57:103-113, 1989). This element or sequences in the untranslated leader were sufficient for correct transcription initiation and for measurable developmental induction. Similar elements were present at or near the initiation sites of other developmentally regulated genes. We propose that the temporal and spatial specificity of expression of these genes results from modulation of the activity of Inr elements.

Electron microscopy of Achlya deoxyribonucleic acid sequence organization.

Electron microscopic analysis of reassociated deoxyribonucleic acid (DNA) from the aquatic fungus Achlya bisexualis revealed details of the sequence arrangement of the inverted repeats and both the highly and moderately repetitive sequence clusters. We used the gene 32 protein-ethidium bromide technique for visualizing the DNA molecules, a procedure which provides excellent contrast between single- and double-stranded DNA regions. Long (greater than 6-kilobase) DNA fragments were isolated after reannealing to two different repetitive C0t values, and the renatured structures were then visualized in an electron microscope. Our results showed that the inverted repeat sequences were short (0.5 kilobase, number-average) and separated by nonhomologous DNA of various lengths. These pairs of sequences were not clustered within the genome. Both highly repetitive and moderately repetitive DNA sequences were organized as tandem arrays of precisely paired, regularly repeating units. No permuted clusters of repeating sequences were observed, nor was there evidence of interspersion of repetitive with single-copy DNA sequences in the Achlya genome.

Saccharomyces cerevisiae centromere CEN11 does not induce chromosome instability when integrated into the Aspergillus nidulans genome.

We constructed Aspergillus nidulans transformation plasmids containing the A. nidulans argB+ gene and either containing or lacking centromeric DNA from Saccharomyces cerevisiae chromosome XI (CEN11). The plasmids transformed an argB Aspergillus strain to arginine independence at indistinguishable frequencies. Stable haploid transformants were obtained with both plasmids, and strains were identified in which the plasmids had integrated into chromosome III by homologous recombination at the argB locus. Plasmid DNA was recovered from a transformant containing CEN11, and the sequence of the essential portion of CEN11 was determined to be unaltered. The transformants were further characterized by using them to construct heterozygous diploids and then testing the diploids for preferential loss of the plasmid-containing chromosomes. The CEN11 sequence had little or no effect on chromosome stability. Thus, CEN11 does not prevent chromosomal integration of plasmid DNA and probably lacks centromere activity in Aspergillus spp.

Direct and indirect gene replacements in Aspergillus nidulans.

We performed three sets of experiments to determine whether cloned DNA fragments can be substituted for homologous regions of the Aspergillus nidulans genome by DNA-mediated transformation. A linear DNA fragment containing a heteromorphic trpC+ allele was used to transform a trpC- strain to trpC+. Blot analysis of DNA from the transformants showed that the heteromorphic allele had replaced the trpC- allele in a minority of the strains. An A. nidulans trpC+ gene was inserted into the argB+ gene, and a linear DNA fragment containing the resultant null argB allele was used to transform a trpC- argB+ strain to trpC+. Approximately 30% of the transformants were simultaneously argB-. The null argB allele had replaced the wild-type allele in a majority of these strains. The A. nidulans SpoC1 C1-C gene was modified by removal of an internal restriction fragment and introduced into a trpC- strain by transformation with a circular plasmid. A transformant containing a tandem duplication of the C1-C region separated by plasmid DNA was self-fertilized, and trpC- progeny were selected. All of these had lost the introduced plasmid DNA sequences, whereas about half had retained the modified C1-C gene and lost the wild-type copy. Thus, it is possible with A. nidulans to replace chromosomal DNA sequences with DNA fragments that have been cloned and modified in vitro by using either one- or two-step procedures similar to those developed for Saccharomyces cerevisiae.

brlA requires both zinc fingers to induce development.

Expression of the Aspergillus nidulans brlA gene induces a developmental pathway leading to the production of asexual spores. We have introduced mutations into brlA that are expected to disrupt either or both Cys2-His2 Zn(II) coordination sites postulated to exist in the BrlA polypeptide. The resultant brlA alleles fail to induce either the asexual reproductive pathway or the expression of development-specific genes. These data support the hypothesis that brlA encodes a nucleic acid-binding protein whose activity requires each of two zinc fingers.

Position-dependent and -independent mechanisms regulate cell-specific expression of the SpoC1 gene cluster of Aspergillus nidulans.

Many genes that are expressed specifically in the differentiating asexual spores (conidia) of Aspergillus nidulans are organized into clusters. We investigated the effects of altered chromosomal position on expression of a gene from the conidiation-specific SpoC1 gene cluster. The gene became deregulated when integrated at nonhomologous chromosomal sites, in that transcript levels were elevated in vegetative cells (hyphae) and variably altered in conidia. We also investigated the effects on expression of insertion of the nonregulated argB gene into the SpoC1 region. Levels of argB transcripts were markedly reduced in hyphae. The results suggest that a cis-acting regional regulatory mechanism represses transcription of SpoC1 genes in hyphae. They also indicate that expression of individual SpoC1 genes is modulated during conidiation by trans-acting factors. We propose that the two types of regulation act together to produce the major differences in transcript levels observed in hyphae versus conidia.

Isolation and physical characterization of three essential conidiation genes from Aspergillus nidulans.

We cloned and characterized three genes from Aspergillus nidulans, designated brlA, abaA, and wetA, whose activities are required to complete different stages of conidiophore development. Inactivation of these genes causes major abnormalities in conidiophore morphology and prevents expression of many unrelated, developmentally regulated genes, without affecting the expression of nonregulated genes. The three genes code for poly(A)+ RNAs that begin to accumulate at different times during conidiation. The brlA- and abaA-encoded RNAs accumulate specifically in cells of the conidiophore. The wetA-encoded RNA accumulates in mature conidia. Inactivation of the brlA gene prevents expression of the abaA and wetA genes, whereas inactivation of the abaA gene prevents expression of the wetA gene. Our results confirm genetic predictions as to the temporal and spatial patterns of expression of these genes and demonstrate that these patterns are specified at the level of RNA accumulation.

Differential effects of analogs of cycloheximide on protein and RNA synthesis in Achlya.

Analogs of the glutarimide antibiotic cycloheximide were tested for their effect on growth and incorporation of proline and uridine into acid-insoluble material in Achlya bisexualis. Each of the compounds tested had reduced antibiotic activity as compared to cycloheximide. The effects of the antibiotics on protein and RNA synthesis were varied. While cycloheximide inhibited both protein and RNA synthesis immediately, two of the analogs inhibited proline incorporation without effect on uridine incorporation, while three, each representing a modification of the hydroxyl of cycloheximide, stimulated uridine incorporation and either had no effect on or inhibited protein synthesis. These results indicate that the control of RNA synthesis by protein synthesis in Achlya can be released by glutarimide antibiotics.

Structure and regulated expression of the SpoC1 gene cluster from Aspergillus nidulans.

We have previously described the organization of a 13.3 kb region of the Aspergillus nidulans genome, designated SpoC1, coding for multiple poly(A)+ RNAs that accumulate in asexual spores but not in somatic cells. We have determined the limits of the SpoC1 gene cluster by investigating the transcriptional features of 53 kb of chromosomal DNA. This segment of the genome codes for at least 19 poly(A)+ RNAs, some of which are transcribed from overlapping regions. The area of developmental regulation is approximately 38 kb in length and is delimited by 1.1-kb direct repeats. With one exception, RNAs transcribed from the central part of the cluster appear late during conidiophore development and accumulate specifically in spores. The exceptional transcript appears earlier during development and accumulates specifically in cells of the conidiophore. In contrast, RNAs encoded at the borders of the cluster occur in both somatic cells and spores. The results indicate that if a chromatin-level control mechanism operates to regulate expression of the SpoC1 gene cluster, as previously suggested by us, additional levels of regulation must also exist.

Isolation and molecular characterization of the Aspergillus nidulans wA gene.

The walls of Aspergillus nidulans conidia contain a green pigment that protects the spores from damage by ultraviolet light. At least two genes, wA and yA, are required for pigment synthesis: yA mutants produce yellow spores, wA mutants produce white spores, and wA mutations are epistatic to yA mutations. We cloned wA by genetic complementation of the wA3 mutation with a cosmid library containing nuclear DNA inserts from the wild-type strain. The wA locus was mapped to an 8.5-10.5-kilobase region by gene disruption analysis. DNA fragments from this region hybridized to a 7500 nucleotide polyadenylated transcript that is absent from hyphae and mature conidia but accumulates during conidiation beginning when pigmented spores first appear. Mutations in the developmental regulatory loci brlA, abaA, wetA and apsA prevent wA mRNA accumulation. By contrast, yA mRNA fails to accumulate only in the brlA- and apsA- mutants. Thus, the level of wA transcript is regulated during conidiophore development and wA activation requires genes within the central pathway regulating conidiation.

Identification of Aspergillus brlA response elements (BREs) by genetic selection in yeast.

The brlA gene of Aspergillus nidulans plays a central role in controlling conidiophore development. To test the hypothesis that brlA encodes a transcriptional regulator and to identify sites of interaction for the BrlA polypeptide, we expressed brlA in Saccharomyces cerevisiae (yeast) strains containing Aspergillus DNA sequences inserted upstream of a minimal yeast promoter fused to the Escherichia coli lacZ gene. Initially, a DNA fragment from the promoter region of the developmentally regulated rodA gene was tested and shown to mediate brlA-dependent transcriptional activation. Two additional DNA fragments were selected from an Aspergillus genomic library by their ability to respond to brlA in yeast. These fragments contained multiple copies of a sequence motif present in the rodA fragment, which we propose to be sites for BrlA interaction and designate brlA response elements (BREs). DNA fragments containing BREs upstream of a minimal Aspergillus promoter were capable of conferring developmental regulation in Aspergillus. Deletion of BREs from the upstream region of rodA greatly decreased its developmental induction. Multiple copies of a synthetic oligonucleotide with the consensus sequence identified among the BREs mediated brlA-dependent transcriptional activation in yeast. The results show that a primary activity of brlA is transcriptional activation and tentatively identify sites of interaction for the BrlA polypeptide.

A large cluster of highly expressed genes is dispensable for growth and development in Aspergillus nidulans.

We investigated the functions of the highly expressed, sporulation-specific SpoC1 genes of Aspergillus nidulans by deleting the entire 38-kb SpoC1 gene cluster. The resultant mutant strain did not differ from the wild type in (1) growth rate, (2) morphology of specialized reproductive structures formed during completion of the asexual or sexual life cycles, (3) sporulation efficiency, (4) spore viability or (5) spore resistance to environmental stress. Thus, deletion of the SpoC1 gene cluster, representing 0.15% of the A. nidulans genome, had no readily detectable phenotypic effects. Implications of this result are discussed in the context of major alterations in gene expression that occur during A. nidulans development.

The use of simulated annealing in chromosome reconstruction experiments based on binary scoring.

We present a method of combinatorial optimization, simulated annealing, to order clones in a library with respect to their position along a chromosome. This ordering method relies on scoring each clone for the presence or absence of specific target sequences, thereby assigning a digital signature to each clone. Specifically, we consider the hybridization of oligonucleotide probes to a clone to constitute the signature. In that the degree of clonal overlap is reflected in the similarity of their signatures, it is possible to construct maps based on the minimization of the differences in signatures across a reconstructed chromosome. Our simulations show that with as few as 30 probes and a clonal density of 4.5 genome equivalents, it is possible to assemble a small eukaryotic chromosome into 33 contiguous blocks of clones (contigs). With higher clonal densities and more probes, this number can be reduced to less than 5 contigs per chromosome.