ABSTRACT
The recessive, temperature-sensitive bimAl mutation of Aspergillus nidulans blocks nuclei in metaphase at restrictive temperature. To determine whether the bimA product is essential, integrative transformation was used to create a mutation in the bimA gene. The mutation was maintained in a heterokaryon and the phenotype of spores produced by the heterokaryon was analyzed. Molecular disruption of the wild-type bimA gene is recessive in the heterokaryon and causes a metaphase block, demonstrating that bimA is an essential gene for mitosis. bimA was cloned by DNA-mediated complementation of its mutant phenotype at restrictive temperature, and the nucleotide sequence of a full-length cDNA was determined. A single large open reading frame was identified in the cDNA sequence, which predicts a protein containing 806 amino acid residues that is related (30.4 % identity) to the Schizosaccbaro-myces pombe nuc2+ gene product, which also is required for completion of mitosis. The sequence of the bimA gene indicates that it is a member of a family of mostly nuclear proteins that contain a degenerate 34 amino acid repeat, the TPR (tetratricopeptide repeat) gene family.
INTRODUCTION
The analysis of temperature-sensitive, fungal mutants defective in mitosis has proven to be extraordinarily useful in understanding how mitosis is regulated. A large number of genes involved in mitosis have been analyzed from the yeasts Saccharomyces cerevisiae (Hartwell and Weinert, 1989) and Schizosaccharomyces pombe (Lee and Nurse, 1988), and from the filamentous fungus Aspergillus nidulans (Morris, 1976; Morris et al. 1989). Not only has this been helpful in defining mitotic functions in these organisms, but it has become increasingly clear that many of the mitotic genes first characterized in fungi are not only ubiquitous, but have functions that are conserved in higher eukaryotes (Doonan and Morris, 1989).
The large number of temperature-sensitive cell cycle mutants that have been identified in Aspergillus fall into two categories: bims (blocked in mitosis), which show an elevated chromosome and spindle mitotic index at restrictive temperature, and nims (never in mitosis), which fail to enter mitosis at the restrictive temperature (Morris, 1976). nimA encodes a protein kinase involved in the positive regulation of mitosis (Osmani et al. 1987; Osmani et al. 19886), while bimE functions in a negative manner, preventing mitosis from occurring during interphase (Osmani et al. 1988a; Engle et al. 1990). bimG encodes a phosphoprotein phosphatase required for the completion of anaphase, providing direct evidence that protein dephosphorylation is essential to complete mitosis (Doonan and Morris, 1989) (see also the dis and bwsl genes of S. pombe; Booher and Beach, 1989; Ohkura et al. 1989). A third bim gene of Aspergillus, bimC, has recently been shown to encode a protein related to kinesin heavy chain (Enos and Morris, 1990), further demonstrating the utility of using dim-type mutants to isolate and study the essential components of the mitotic apparatus.
The bimA gene of A. nidulans was identified in the course of a screen designed to identify temperaturesensitive, conditionally lethal mutations that cause abnormalities in mitosis (Morris, 1976). In this paper we show that the bimAl mutation at restrictive temperature or molecular disruption of bimA causes a mitotic arrest with condensed chromatin and a persistant mitotic spindle. We also report on the molecular cloning of the bimA gene of A. nidulans by complementation of the temperature-sensitive bimAl mutant phenotype. The nucleotide sequence of the wild-type bimA gene shows that it is structurally related to a family of genes that exhibit a degenerate 34 amino acid repeat known as the TPR (tetratricopeptide) motif (Boguski et al. 1990; Hirano et al. 1990; Sikorski et al. 1990).
MATERIALS AND METHODS
Strains used were R153 (wA3, pyroA4), SO8 (bimAl, wA2, cnxE3, choAl, pyrG39 and possibly other markers not tested), SO10 (bimAl, wA2 and possibly other markers not tested), and GR5 (pyrG89, wA3, pyroAi). Strains were grown on YAG solid medium and YG broth as previously described (Osmani et al. 1987). Medium for pyrGrH strains S08 and GR5 was supplemented with 5mM uridine and 10 mM uracil (YAG+UU). For growth studies, the permissive and restrictive temperature was 32 °C and 42–44°C, respectively. Standard A. nidulans genetic techniques were employed (Clutterbuck, 1974; Pontecorvo et al. 1953).
Microscopy of Aspergillus
Conidia (asexual uninucleate spores) were grown on solid media overlaid with cellophane, fixed and stained as described previously (Doonan and Morris, 1989; Engle et al. 1988; Osmani et al. 1988b). For indirect immunofluorescence visualization of microtubules, samples were stained with anti-tubulin antibodies (YOL 1/34 generous gift from Dr J. Kilmartin, MRC, Cambridge, England) (Kilmartin et al. 1982) to visualize mitotic spindles and to determine the spindle mitotic index (SMI) (Morris, 1976). For nuclear staining, cells were fixed in 5% glutaraldehyde, 50 mM sodium phosphate (pH 6.8), 0.2% Triton X-100 and 0.25–0.5 μg ml−1 4′,6′-diamidino-2-phenylindole (DAPI), a fluorescent DNA binding dye. Samples were viewed in a Zeiss Research Microscope under epifluorescent illumination and micrographs were recorded on Kodak T-Max film rated at ASA 400 and developed in T-Max developer according to the manufacturer’s instructions (Kodak, Rochester, New York). The chromosome mitotic index (CMI, percentage nuclei with condensed chromatin) was determined by the method of Morris (1976) as described by Bergen et al. (1984).
Molecular cloning of the bimA gene
The bimA gene was cloned by complementation of bimAl using the SAOLIB1 A. nidulans genomic library (Osmani et al. 1987). One transformant (SO8TR1) was identified by Southern blot analysis (Southern, 1975), and was found to contain a single plasmid integrated into its genome, using radiolabeled pBR322 DNA as a hybridization probe (Feinberg and Vogelstein, 1983). Genomic DNA isolated from SO8TR1 (Raeder and Broda, 1985) was subjected to a partial BgZII digest followed by religation and marker rescue in Escherichia coli. One plasmid (pSO8TR1.6) was rescued that could complement the SO8 mutant strain. Various subclones of pSO8TR1.6 were tested for their ability to complement the temperature sensitivity of the bimAl mutation in strain SO8. One such complementating subclone (pKO3), a 6.6 kb (kilobase) EcoRI genomic fragment ligated into the EcoRI site of pUC18, was used to probe an A. nidulans genomic DNA library constructed in bacteriophage lambda Charon 4A (provided by Dr W. E. Timberlake, University of Georgia) as described by May et al. (1985). A 7.0 kb EcoRI genomic fragment was isolated that complemented the SO8 bimAl mutant (insert of pKO4). After subcloning into pUC18 or pRG3 (Waring et al. 1989), plasmids were tested for their ability to complement bimAl as follows: pKO4 contains a 7.0 kb EcoRI fragment ligated into the EcoRI site of pRG3; pKO5 contains a 3.6 kb Psil fragment ligated into the Psil site of pRG3; pKO6 contains a 2.1 kb Psil-BamHI fragment ligated into the Bs/I-BamHI site of pUC18; pKO8 contains a 0.9 kb BamHI fragment ligated into the BaznHI site of pUC18; pKOlO contains a 1.9 kb BgZII fragment ligated into the BamHI site of pUC18 (see Fig. 3A for complementation data of subclones). Northern blots were conducted, in conjunction with the subcloning experiments, in order to determine the number of RNA species per subclone. Northern blots were prepared initially as described by Osmani et al. (1987), but formaldehyde-RNA gels were subsequently employed (Maniatis et al. 1982). A complementing 3.6 kb Pstl bimA genomic fragment (insert of pKO5) was used as a hybridization probe to screen a cDNA library constructed in lambda Zap vector (Stratagene, La Jolla, California) prepared from Aspergillus nidulans total mycelial poly(A)+ mRNA (Osmani et al. 1987). Eleven positive clones were purified and all clones were identical as determined by restriction mapping and by sequence analysis of the 5’ and 3’ ends of the cDNA inserts. Following purification on cesium density gradients (Maniatis et al. 1982), all four cDNA clones tested (pKO23-pKO26) complemented the bimAl mutation in strain SO8. In addition, a 2.7 kb EcoRV-EcoRI fragment subcloned from pKO25 into the Smal-EcoRI site of pBluescript to yield pKO27 complemented the bimAl mutation. One complementing cDNA (pKO25) was sequenced completely on both strands using chainterminating dideoxynucleotide methods (Sanger et al. 1977) as modified by Biggin et al. (1983). M13 sequencing primers (New England Biolabs, Beverly, MA) and the Sequenase kit (United States Biochemical Corporation, Cleveland, Ohio) were used in sequencing single-stranded bacteriophage M13 DNA templates (Yanisch-Perron et al. 1985). Sequencing reactions were run on 8 % and/or 5 % polyacrylamide wedge gels, which permitted up to 500 bases per template to be read. Sequence data were analyzed with the aid of the IBI Pustell system (International Bidtechnologies Inc., New Haven, CT) using an IBM-compatible personal computer and with the GCG Wisconsin package. The predicted bimA protein sequence was compared with protein sequences in the National Biomedical Research Foundation Protein Identification Resource (NBRF/PIR) database using the IBI Pustell sequence analysis programs. A hydropathy plot of the predicted bimA protein sequence was obtained, after Kyte and Doolittle (1982).
Molecular disruption of bimA
As bimA was expected to be an essential gene, the heterokaryotic gene disruption technique described by Osmani et al. (1988a) was employed to determine the terminal phenotype of a nontemperature-sensitive mutation at the bimA locus. To achieve this, an internal 1788 bp (base-pair) PuuII-BamHI fragment of the bimA cDNA was cloned into the Smal-BamHI site of pRG3 (Waring et al. 1989). The resulting disrupter plasmid (pKO35), which when integrated by homologous recombination at the bimA locus will introduce 5′ and 3′ deletions into the bimA gene, was used to transform protoplasts of host strain GR5 to uridine prototrophy. Analysis of the resulting balanced heterokaryon was as described previously (Osmani et al. 1988a). Transformant DNA was isolated and subjected to Southern blot analysis using an 815 bp SacII-AccI fragment internal to the 1788 bp PuuII-BamHI fragment as a hybridization probe.
Results
The bimA gene product is required for progression through metaphase
The bimAl mutant allele of A. nidulans was initially isolated as a recessive, temperature-sensitive mutation that caused an increased mitotic index when cells were shifted to the restrictive temperature (Morris, 1976). In the R153 wild-type strain grown at the restrictive temperature (44°C), chromosome disjunction occurs normally (Fig. 1A-D; A and D show germlings with interphase nuclei, B and C show germlings with mitotic metaphase and telophase nuclei, respectively). However, when conidia of the temperature-sensitive SCIO bimAl mutant strain were incubated at the permissive temperature (32 °C) for 6h and shifted to the restrictive temperature (44°C) (Fig. 1E-G), or were germinated at 44°C (Fig. 1H), nuclei arrested at metaphase.
Nuclear division in R153 wild-type and SO10 òimAl strains of Aspergillus at restrictive temperature. Fluorescence micrographs of A. nidulans R153 wild-type and temperaturesensitive SO10 bimAl mutant strains incubated at permissive temperature (32 °C) for 6h on YAG agar medium and shifted to restrictive temperature (44 °C) or germinated at 44°C. DAPI staining at various times revealed that mitosis proceeds normally at restrictive temperature in the wild-type strain (A-D; A and D show germlings with interphase nuclei; B and C show germlings with metaphase and telophase nuclei, respectively). The metaphase-to-anaphase transition fails to occur in the SO10 bimAl mutant strain shifted to 44°C (E-G) or germinated at 44°C (H). ×2200.
Nuclear division in R153 wild-type and SO10 òimAl strains of Aspergillus at restrictive temperature. Fluorescence micrographs of A. nidulans R153 wild-type and temperaturesensitive SO10 bimAl mutant strains incubated at permissive temperature (32 °C) for 6h on YAG agar medium and shifted to restrictive temperature (44 °C) or germinated at 44°C. DAPI staining at various times revealed that mitosis proceeds normally at restrictive temperature in the wild-type strain (A-D; A and D show germlings with interphase nuclei; B and C show germlings with metaphase and telophase nuclei, respectively). The metaphase-to-anaphase transition fails to occur in the SO10 bimAl mutant strain shifted to 44°C (E-G) or germinated at 44°C (H). ×2200.
Conidia from R153 wild-type and SO10 mutant strains were grown for 7.5 h at permissive temperature (32°C), shifted to restrictive temperature, and the chromosome mitotic index (CMI, using DAPI stain) and the spindle mitotic index (SMI, using yeast anti-tubulin monoclonal antibody) were scored over a 6h period (Fig. 2). At permissive temperature, the SO10 mutant strain grows normally with a CMI of 2% or less. When shifted from permissive to restrictive temperature, the CMI and SMI in the mutant strain increased in a time-dependent manner while no significant increase in the mitotic indices was observed in the wild-type (Fig. 2). After 6.5 h at restrictive temperature, the chromosome mitotic index of the SC10 bimAl strain had risen to 70% and the SMI to 44%.
Chromosome mitotic index (CMI) and spindle mitotic index (SMI) of R153 wild-type (WT) and SO10 bimAl strains at restrictive temperature. Conidia were grown for 7.5 h at 32°C, at which time the cultures were shifted to restrictive temperature 44 °C (time 0). Samples were taken at 1 h intervals to determine the CMI CDAPI fluorescent DNA binding dye) and SMI (anti-tubuhn immunofluorescence).
Chromosome mitotic index (CMI) and spindle mitotic index (SMI) of R153 wild-type (WT) and SO10 bimAl strains at restrictive temperature. Conidia were grown for 7.5 h at 32°C, at which time the cultures were shifted to restrictive temperature 44 °C (time 0). Samples were taken at 1 h intervals to determine the CMI CDAPI fluorescent DNA binding dye) and SMI (anti-tubuhn immunofluorescence).
Molecular cloning of the bimA gene by complementation of bimAl
We have used DNA-mediated complementation of the bimAl mutant phenotype to clone the bimA gene. An Aspergillus genomic library was used to transform mutant strain SOS carrying bimAl to temperature insensitivity. Southern blot analysis showed that one transformant (SO8TR1) contained a single integrated plasmid from the library. To determine whether bimAl had been complemented by bimA, or an extragenic suppressor of bimAl, a two-step gene replacement (Miller et al. 1985) was performed using this strain. Progeny isolated from a selfcross of SO8TR1 that had looped out the transforming plasmid and retained the temperature-insensitive phenotype were back-crossed to a wild-type strain. No temperature-sensitive progeny were observed from the back-cross, demonstrating that the wild-type bimA gene had integrated at the bimAl locus.
Subcloning of bimA from SO8TR1
A plasmid carrying a wild-type copy of bimA (pKO27) was isolated from strain SO8TR1 (pSO8TR1.6). A 6.6 kb AcoRI fragment was subcloned from pSO8TR1.6 into the JECORI site of pUC18 to yield plasmid pKO3, which complemented the SO8 bimAl mutant strain. However, further attempts to subclone this 6.6 kb EcoRI fragment and retain complementation were unsuccessful. Therefore, the 6.6 kb E’coRI fragment was used as a hybridization probe to screen an Aspergillus genomic library to isolate the 7 kb insert of pKO4 (Fig. 3A). Northern blot analysis revealed that this fragment hybridized to three mRNA species of 4.4, 2.8 and 1.8kb in length (Fig. 3B). Subcloning experiments were conducted as outlined in Fig. 3A. A 3.6 kb Pstl genomic fragment (insert of pKO5) was used as a probe to screen a cDNA library and a cDNA copy of bimA was isolated that hybridized to a single mRNA species of 2.8 kb and complemented the SO8 mutant strain (Fig. 3).
Subcloning of bimA. (A) A restriction map of the genomic insert of pSO8TR1.6 is shown above subcloned plasmids. Horizontal lines indicate DNA fragments that were tested for complementation of temperature sensitivity after transformation of SO8 bimAl strains. Data concerning complementation of the temperature sensitivity of SO8 bimAl strain is indicated to the right. (B) RNA hybridization analysis. Hybridization probes for Northern blot analysis were isolated from pKO4 or from cDNA clone pKO27 as indicated. Radiolabeled plasmid DNA was used to probe poly(A)+ mRNA.
Subcloning of bimA. (A) A restriction map of the genomic insert of pSO8TR1.6 is shown above subcloned plasmids. Horizontal lines indicate DNA fragments that were tested for complementation of temperature sensitivity after transformation of SO8 bimAl strains. Data concerning complementation of the temperature sensitivity of SO8 bimAl strain is indicated to the right. (B) RNA hybridization analysis. Hybridization probes for Northern blot analysis were isolated from pKO4 or from cDNA clone pKO27 as indicated. Radiolabeled plasmid DNA was used to probe poly(A)+ mRNA.
Northern blots of RNA made throughout the cell cycle (Osmani et al. 1987) showed that the level of bimA mRNA does not change during the cell cycle (data not shown), suggesting that the level of the bimA message is not regulated in a cell cycle-dependent manner. In addition, Southern blot analysis conducted at low stringency demonstrated that bimA is a single copy gene in the Aspergillus genome (data not shown).
Nucleotide and predicted amino acid sequence of bimA
A 3028 bp cDNA, which could complement the SO8 bimAl mutant strain, was isolated from a lambda library and used as a probe on a Northern blot where it hybridized to a 2.8 kb message, indicating that the bimA cDNA is near full length (Fig. 3B). Dideoxy sequencing of the 3028 bp cDNA on both strands revealed a long open reading frame encoding 806 amino acids (Fig. 4). An initiation methionine in good context (Kozak, 1986) at nucleotide position 359 opens this reading frame, which extends for 2418 bp. The predicted molecular mass of this protein is 89 690 with an isoelectric point of 9.65. A hydropathy plot of the bimA protein (Kyte and Doolittle, 1982) revealed that it is primarily rich in hydrophilic residues with only a few hydrophobic stretches in the amino (residues 1–260) and carboxyl (residues 400–806) domains (data not shown). The ‘bimA’ domain (residues 261–399, see below) is strongly hydrophilic as is most of the carboxyl domain. A single methionine in poor context is present in the bimA cDNA 5′ to the open reading frame at nucleotide position −109 but its frame terminates after five amino acid residues. A polyadenylation signal is absent and codon usage is unbiased in the bimA gene product.
Nucleotide and predicted amino acid sequence of the bimA gene. A 3028 bp long cDNA was sequenced completely on both strands and found to contain an 806 amino acid open reading frame (ORF). Single amino acid designations are used for the predicted bimA polypeptide. The numbers indicated on the left refer to the first nucleotide of each line, while numbers on the right refer to the last amino acid of each line. The predicted gene product of bimA has a calculated molecular mass of 89 690 and an isoelectric pH of 9.65.
Nucleotide and predicted amino acid sequence of the bimA gene. A 3028 bp long cDNA was sequenced completely on both strands and found to contain an 806 amino acid open reading frame (ORF). Single amino acid designations are used for the predicted bimA polypeptide. The numbers indicated on the left refer to the first nucleotide of each line, while numbers on the right refer to the last amino acid of each line. The predicted gene product of bimA has a calculated molecular mass of 89 690 and an isoelectric pH of 9.65.
The bimA and S. pombe nuc2+ gene products are related
Computer analysis revealed significant similarity between the bimA gene protein and the nuc2+ product of S. pombe (Hirano et al. 1988), and CDC16 of S. cerevisiae (Icho and Wickner, 1987). The Aspergillus bimA sequence more closely resembles the nuc2+ gene product (Fig. 5). The NH2- and COOH-terminal protein sequences of bimA and nuc2+ align readily with the addition of a few small gaps (<3 amino acid residues) in nuc2+. Overall 245 of 806 residues (30.4%) of bimA and nuc2+ are identical while 361 of 806 are conserved (44.8%). The NH2 and COOH termini of bimA and nuc2+ products show the most conservation while a stretch of 139 amino acids unique to bimA (termed the ‘bimA’ domain) is located between the amino (N) and carboxyl (C) domains. Of the first 250 amino-terminal amino acids, 26.8% (67/250) are identical while 49% (175/355) of the carboxyl-terminal 355 amino acids are identical. If this comparison is extended to include conservative amino acid substitutions, the N and C domains of bimA and nuc2+ show 48% and 64.3% similarity, respectively.
The predicted bimA gene product (An) is related to the nuc2+ product of S. pombe (Sp). Dashes indicate gaps. Identical amino acid residues are represented by a perpendicular line, while a star indicates a conservative substitution. The 10 unit TPR domains are indicated by the bold underlines.
The predicted bimA gene product (An) is related to the nuc2+ product of S. pombe (Sp). Dashes indicate gaps. Identical amino acid residues are represented by a perpendicular line, while a star indicates a conservative substitution. The 10 unit TPR domains are indicated by the bold underlines.
bimA belongs to the TPR gene family (tetratrico peptide repeat)
bimA belongs to a family of genes (Boguski et al. 1990; Sikorski et al. 1990) characterized by 34-residue polypeptide repeats (TPR family), whose members include SSN6 (Schultz and Carlson, 1987; Schultz et al. 1990), SKI3 (Rhee et al. 1989), CDC16 (Icho and Wickner, 1987), STI1 (Nicolet and Craig, 1989), OMP (Hase et al. 1983; Riezman et al. 1983) and CDC23 (Doi and Doi, 1990; Sikorski et al. 1990) of S. cerevisiae, nuc2+ of S. pombe (Hirano et al. 1988, 1990), and crn of Drosophila (Boguski et al. 1990). The TPR domains of the bimA and nuc2+ proteins are composed of 10 unit repeats of precisely 34 residues, except for unit repeat 1 of bimA, which is 37 residues long (Fig. 6). One unit in the bimA and nuc2+ gene products is in the N domain (unit 0), while the remaining nine unit repeats (1–9) are in tandem in the C domain. Data from subcloning experiments (Fig. 3A) indicate that the temperature-sensitive bimAl. mutation is located in one of the carboxyl-terminal repeat units (between 4 and 9). The ten 34 amino acid residue unit repeats present in bimA and nuc2+ proteins are illustrated in Fig. 6A. Alignment of the proteins revealed a bimA and nuc2+ unified consensus sequence (Fig. 6B) (Hirano et al. 1990; Sikorski et al. 1990).
Location of the ten 34-residue polypeptide unit repeats in A. nidu.Lans bimA and S. pombe nuc2+ proteins. (A) Sketch showing location of the ten (0–9) 34-residue polypeptide unit repeats (TPR) in bim.A and nuc2+ proteins. All TPR unit repeats are 34 residues in length with the exception of unit repeat 1 of bim.A, wh.ich is 37 residues long. The unique bim.A domain of unknown function and the nuc2 DNA binding domain are indicated. Unit repeat 3 of both polypeptides (indicated by crosshatching), termed 34 v-repeat by Hirano et al. (1990), lacks the conserved motifs. TPR unit repeat 6 of bimA and nuc2 (indicated by stippling) contains a glycine in position 3. Gly3 in repeat 6 of nuc2 is thought to be essential. aa, amino acid residues. (B) Alignment of the amino acid sequences of the 10 TPR unit repeats (0-9) in the bim.A and nuc2 proteins. Residues in positions 6–7 (IAS) of the degenerate 37-residue unit repeat 1 of bimA were deleted to improve the alignment. (The actual sequence is …FKIASGY…) Numbers on the right indicate the number of residues from the NH2 terminus. Conserved residues are in bold type. The unified bimA and nuc2 consensus sequence is indicated at the bottom of the figure. Large hydrophobic amino acid residues in the consensus sequence are represented by stars.
Location of the ten 34-residue polypeptide unit repeats in A. nidu.Lans bimA and S. pombe nuc2+ proteins. (A) Sketch showing location of the ten (0–9) 34-residue polypeptide unit repeats (TPR) in bim.A and nuc2+ proteins. All TPR unit repeats are 34 residues in length with the exception of unit repeat 1 of bim.A, wh.ich is 37 residues long. The unique bim.A domain of unknown function and the nuc2 DNA binding domain are indicated. Unit repeat 3 of both polypeptides (indicated by crosshatching), termed 34 v-repeat by Hirano et al. (1990), lacks the conserved motifs. TPR unit repeat 6 of bimA and nuc2 (indicated by stippling) contains a glycine in position 3. Gly3 in repeat 6 of nuc2 is thought to be essential. aa, amino acid residues. (B) Alignment of the amino acid sequences of the 10 TPR unit repeats (0-9) in the bim.A and nuc2 proteins. Residues in positions 6–7 (IAS) of the degenerate 37-residue unit repeat 1 of bimA were deleted to improve the alignment. (The actual sequence is …FKIASGY…) Numbers on the right indicate the number of residues from the NH2 terminus. Conserved residues are in bold type. The unified bimA and nuc2 consensus sequence is indicated at the bottom of the figure. Large hydrophobic amino acid residues in the consensus sequence are represented by stars.
Molecular disruption of bimA
It is possible to disrupt essential genes in A. nidulans by DNA-mediated integration of disrupter plasmids and maintain nuclei that lack the essential gene in heterokaryons. Upon differentiation to produce asexual spores the heterokaryotic state is broken as the spores are uninucleate. Germination of spores from such heterokaryons permits the phenotypic analysis of a cell type lacking the essential function. This technique has been termed heterokaryotic gene disruption (Osmani et al. 1988). To determine whether bimA is essential, DNA-mediated transformation was used to disrupt this gene using the heterokaryotic gene disruption technique (Fig. 7A-D) as previously described (Osmani et al. 1988a; Oakley et al. 1990). Conidia derived from the primary transformants were screened on selective and nonselective media (Fig. 7C). Ten transformants appeared to be balanced heterokaryons containing both pyr+, bimA− transformed, and pyr−, bimA+ nontransformed, nuclei. These yielded haploid conidia that could not grow into colonies on selective media. Southern blot analysis (Fig. 7D) revealed that the disrupter plasmid pKO35 had integrated at the bimA locus in two heterokaryon transformants (GR5pKO35.67 and GR5pKO35.74). Germinating conidia derived from these heterokaryons and untransformed host strain GR5 were grown on nonselective (YAG+UU) agar medium. When germlings were fixed and stained with DAPI at hourly intervals and examined cytologically (Fig. 8A-B) to determine the chromosome mitotic index (CMI), two classes of germlings were observed: 71 % were multinucleate and identical to the untransformed GR5 host strain (pyr−, bimA+) while 29% were uninucleate with nuclei blocked in metaphase (pyr+, bimA−) on nonselective media. Therefore, molecular disruption of bimA produces a mutant phenotype similar to the original temperature-sensitive SO8 bimAl mutant. When mycelium derived from conidia of transformant strains GR5pKO35.67 and GR5pKO35.74 grown on nonselective media was harvested and the DNA analyzed by Southern blot (Fig. 7D), only a nontransformed hybridization pattern was observed. It is thus only transformed, disrupted nuclei that are lost from the heterokaryon during growth on nonselective media.
Heterokaryotic gene disruption of bimA and phenotypic analysis. (A) Partial restriction map of a 5′-3′ truncated 1.7 kb Pvull-BamHI b1mA cDNA fragment. Pv, Pvull; H, Hind.Ill; P, Pstl; Bg, Bgill; 82, Sacll; K, Kpnl; A, Accl; B, BamHI. Filled bar under maps represents tbe 816 bp Sacll-Accl fragment UBed as a hybridization probe. The filled region in disrupter plasmid pK035 indicates the 1.7 kb internal Pvull-BamHI fragment from tbe bimA cDNA cloned into the Smal-BamHI sites of pRG3. A partial restriction map of the bimA genomic locUB is shown; probe hybridizes to a 1.9 kb BgnI fragment. A restriction map of pK036 integrated site specifically at the bimA genomic locus is also shown; probe hybridizes to a 6.49 kb BgIII fragment. (B) Outline of heterokaryotic gene disruption technique. Some nuclei m protoplasts of GR6 pyr−,bimA+ host strain were transformed to pyr+,bimA− by the disrupter plasmid pK036, thereby giving rise to a heterokaryotic germling, which can then grow on selective medium to form a conidiating colony, pK035 carries tbe pyr4 gene of Neurospora as a selectable marker. At conidiogenesis individual nuclei are segregated from the heterokaryotic cell into a conidium, thus abolishing the heterokaryotic state and permitting phenotypic analysis of the disrupted allele. (C) Putative heterokaryot1c traruiformanta were screened on selective (YAG, top plate) and nonselective (YAG+UU, bottom plate) solid media in order to identify stable heterokaryons. (D) Southern blot analysis of Bgill-digested DNA (300 ng) isolated from untransformed GR5 host strain (lane 1); transformed stable heterokaryons GR6pK035.67 and GR5pK035.74 of GR6 contairu.ng the disrupter plasmid pK035 integrated at the bimA locus of some nuclei (lanes 2 and 3, respectively); mycelium derived from the conidia of stable heterokaryons GR5pK036.67 and GR6pK035.74 grown on nonselective media (lanes 4 and 6, respectively). The blot was probed with a radiolabeled 815 bp Sacll-Accl fragment of the btm.A cDNA.
Heterokaryotic gene disruption of bimA and phenotypic analysis. (A) Partial restriction map of a 5′-3′ truncated 1.7 kb Pvull-BamHI b1mA cDNA fragment. Pv, Pvull; H, Hind.Ill; P, Pstl; Bg, Bgill; 82, Sacll; K, Kpnl; A, Accl; B, BamHI. Filled bar under maps represents tbe 816 bp Sacll-Accl fragment UBed as a hybridization probe. The filled region in disrupter plasmid pK035 indicates the 1.7 kb internal Pvull-BamHI fragment from tbe bimA cDNA cloned into the Smal-BamHI sites of pRG3. A partial restriction map of the bimA genomic locUB is shown; probe hybridizes to a 1.9 kb BgnI fragment. A restriction map of pK036 integrated site specifically at the bimA genomic locus is also shown; probe hybridizes to a 6.49 kb BgIII fragment. (B) Outline of heterokaryotic gene disruption technique. Some nuclei m protoplasts of GR6 pyr−,bimA+ host strain were transformed to pyr+,bimA− by the disrupter plasmid pK036, thereby giving rise to a heterokaryotic germling, which can then grow on selective medium to form a conidiating colony, pK035 carries tbe pyr4 gene of Neurospora as a selectable marker. At conidiogenesis individual nuclei are segregated from the heterokaryotic cell into a conidium, thus abolishing the heterokaryotic state and permitting phenotypic analysis of the disrupted allele. (C) Putative heterokaryot1c traruiformanta were screened on selective (YAG, top plate) and nonselective (YAG+UU, bottom plate) solid media in order to identify stable heterokaryons. (D) Southern blot analysis of Bgill-digested DNA (300 ng) isolated from untransformed GR5 host strain (lane 1); transformed stable heterokaryons GR6pK035.67 and GR5pK035.74 of GR6 contairu.ng the disrupter plasmid pK035 integrated at the bimA locus of some nuclei (lanes 2 and 3, respectively); mycelium derived from the conidia of stable heterokaryons GR5pK036.67 and GR6pK035.74 grown on nonselective media (lanes 4 and 6, respectively). The blot was probed with a radiolabeled 815 bp Sacll-Accl fragment of the btm.A cDNA.
Phenotypic analysis of heterokaryotic transformants. (A) DAPl-stained conidia of stable heterokaryon stram GR5pKO35.74 germinated on nonselective medium. Arrows indicate uninucleate germlings blocked in mitotic metaphase caused by molecular disruption of bimA. Nontransformed conidia of host strain GR5 have formed germlings that are multinucleate with all of the nuclei in interphase. The punctate cytoplasmic fluorescence represents mitochondria. ×1600. (B) Chromosome condensation during germination of conidia from a transformed stable heterokaryon (GR5pKO35.74) in which the disrupter plasmid pKO35 had integrated site specifically at the bimA locus in some nuclei (approx. 71% nuclei pyr−, bimA+; 29% nuclei pyr+, bimA-) and the untransformed host strain GR5 (pyr−, bimA+) grown on nonselective (YAG+ UU) agar medium. Samples were fixed and stained with DAPI at hourly mtervals to determine the chromosome mitotic index (CMI).
Phenotypic analysis of heterokaryotic transformants. (A) DAPl-stained conidia of stable heterokaryon stram GR5pKO35.74 germinated on nonselective medium. Arrows indicate uninucleate germlings blocked in mitotic metaphase caused by molecular disruption of bimA. Nontransformed conidia of host strain GR5 have formed germlings that are multinucleate with all of the nuclei in interphase. The punctate cytoplasmic fluorescence represents mitochondria. ×1600. (B) Chromosome condensation during germination of conidia from a transformed stable heterokaryon (GR5pKO35.74) in which the disrupter plasmid pKO35 had integrated site specifically at the bimA locus in some nuclei (approx. 71% nuclei pyr−, bimA+; 29% nuclei pyr+, bimA-) and the untransformed host strain GR5 (pyr−, bimA+) grown on nonselective (YAG+ UU) agar medium. Samples were fixed and stained with DAPI at hourly mtervals to determine the chromosome mitotic index (CMI).
These results demonstrate that molecular disruption of the bimA gene is recessive in a heterokaryon and confirm that the bimAl mutant is most likely a loss of function mutation. Thus bimA is an essential gene required for mitosis.
DISCUSSION
The gene products of bimA and nuc2 are related bimAl is a temperature-sensitive, recessive mutation of A. nidulans that causes nuclei to be blocked in mitosis at restrictive temperature. We have used DNA-mediated transformation to clone the wild-type bimA: gene by complementation. The nucleotide sequence of a bimA cDNA was determined and found to encode a protein that resembles the nuc2 gene of S. pombe, which is also required for proper completion of mitosis (Hirano et al. 1988, 1990). bimA and nuc2 have a number of strong similarities. The temperature-sensitive nuc2-663 mu tation, like bimAl, arrests cells in mitosis at restrictive temperature with condensed chromatin and a short mitotic spindle. Gene-disruption experiments have shown that both are essential and in the case of bim.A that disruption phenocopies the temperature-sensitive bim.Al mutation. Moreover, bim.A and nuc2 both belong to a newly recognized family of genes characterized by multiple copies of a highly degenerate, 34 amino acid repeat (the tetratricopeptide repeat or TPR family) (Boguski et al. 1990; Sikorski et al. 1990), which has been predicted to form an amphipathic helix capable of llBSOciations between protein helices or with membranes (Hirano et al. 1990). In addition to the TPR family of proteins, paired amphipathic helix (PAR) motifs have been identified in the SIN3 gene product of S. cereuisiae (Wang et al. 1990) and in the myc family of helix-loop-helix DNA-binding proteins (Nicolet and Craig, 1989; Murre et al. 1989).
Within the TPR family, seven genes are known in addition to bim.A and nuc2. These are the S. cereuisiae genes CDC16 (lcho and Wickner, 1987), CDC23 (Doi and Doi, 1990; Sikorski et al. 1990), SKI3 (Rhee et al. 1989), SSN6 (Schultz and Carlson, 1987; Schultz et al. 1990), STil (Nicolet and Craig, 1989), OMP (Hase et al. 1983; Riezman et al. 1983) and the Drosophila gene crooked neck crn (Boguski et al. 1990). Temperature-sensitive mu tations ofCDC16 and CDC23, like bimA and nuc2, cause a mitotic block (blocking with a large bud) at restrictive temperature. CDC23 resembles bimA and nuc2 more closely than the other TPR genes and may have a similar function. Other TPR gene products appear to play a role in transcription regulation (Rhee et al. 1989; Schultz and Carlson, 1987; Schultz et al. 1990), while the Drosophil, a crooked neck (crn) protein is involved in neural develop ment (Boguski et al. 1990). The gene products of nuc2, CDC16 and SSN6 have been located in the nucleus, and SKI3 has been shown to l9calize, B-galactosidase in the nucleus (Rhee et al. 1989). There is some evidence that the nuc2 protein is localized at the nuclear scaffold (Hirano et al. 1988, 1990). However, not all TPR gene products have a restricted nuclear localization, since OMP binds to the outer mitochondrial membrane (Hase et al. 1983; Riezman et al. 1983).
The predicted gene products of bimA and nuc2 each have 10 TPR units, one unit (unit 0) in the N-terrninal portion of the molecule, and the remaining nine (units 1–9) arranged in tandem with no intervening sequences in the C-terminal region of the protein (Fig. 6A). In both cases all of the repeat units contain the conserved, degenerate 34 amino acid sequence (Fig. 6B), except for unit repeat 3 of both polypeptides, which is still 34 amino acids long but lacks the conserved motif. All of the TPR repeats have a high degree of identity between bimA and nuc2. In repeat 0, the identity (29 %) is for the most part restricted to the consensus sequence (Fig. 6), but the identity becomes more pronounced, 44 % and 35 %, in repeats 1 and 2, and becomes stronger still, ranging from 47 % to 62 % in repeat units 3–9.
Because bim.A and nuc2 share significant amino acid sequence conservation (30.4 % identity overall), and because the nuc2 and bimAl mutations cause mitotic arrest, they may have similar, if not identical, functions. As the nuc2 protein is thought to be a structural component of the nuclear scaffold, bimA may also identify a structural component of the mitotic apparatus. Indeed, there is a formal possibility that bimA is the nuc2 homolog in A. nidulans. There are, however, some notable differences between the two proteins that suggest caution in ascribing identical functions to their gene products. In nuc2 a stretch of 150 amino acids that contains a putative DNA binding region separates repeats O and 1 (Hirano et al. 1990). In bimA repeats 0 and 1 are separated by 316 amino acids, having no similarity to the comparable region of nuc2. Another difference is that the C terminus of bimA is conspicuously more acidic (aspartic and glutamic acids constitute 12 of the last 14 C-terminal amino acids of bimA). Lastly, preliminary experiments expressing the nuc2+ gene in Aspergillus did not comp lement the S08 bimAl mutant strain (O’Donnell and Morris, unpublished). Thus, we do not know whether bimA and nuc2 serve identical functions in A. nidulans and S. pombe.
ACKNOWLEDGEMENTS
We thank Drs D. B. Engle, G. S. May, R. T. Pu, S. Fidel, A. Enos, K. Kirk, M. Giligilnski and P. Doshi for technical advice and encouragment during this work, and Linda Smith and Michael Russo for their skilled assistance in preparing the drawings. Special thanks are due to Dr J. H. Doonan for constructive discussions and good humor. This work was supported by NIH grant GM 34711 to N.R.M. and by NSF grant BSR-8607440 to K.O.D.