The nuclear aod-1 gene of Neurospora crassa encodes the alternative oxidase and is induced when the standard cytochrome-mediated respiratory chain of mitochondria is inhibited. To study elements of the pathway responsible for alternative oxidase induction, we generated a series of mutations in the region upstream from the aod-1 structural gene and transformed the constructs into an aod-1 mutant strain. Transformed conidia were plated on media containing antimycin A, which inhibits the cytochrome-mediated electron transport chain so that only cells expressing alternative oxidase will grow. Using this functional in vivo assay, we identified an alternative oxidase induction motif (AIM) that is required for efficient expression of aod-1. The AIM sequence consists of two CGG repeats separated by 7 bp and is similar to sequences known to be bound by members of the Zn(II)2Cys6 binuclear cluster family of transcription factors. The AIM motif appears to be conserved in other species found in the order Sordariales.

---

MITOCHONDRIA play important roles in many aspects of cell function (WU and MORRIS 1998; BARTLETT and EATON 2004; TRIFUNOVIC et al. 2004; GREEN 2005; NICHOLLS 2005; LILL and MÜHLENHOF 2006) and contain their own mtDNA, which encodes a few proteins of the organelle. However, the vast majority of mitochondrial proteins are the products of nuclear genes (ATTARDI and SCHATZ 1988). It has been shown in many systems that the expression of nuclear-encoded mitochondrial proteins is altered when mitochondrial function is compromised (BUTOW and AVADHANI 2004; LISTER et al. 2004; BISWAS et al. 2005; RHOADS et al. 2006). Thus, the functional state of the organelle is monitored and communicated to the nucleus by one or more signal transduction systems. The mechanism by which mitochondria-to-nucleus communication occurs has been termed "the retrograde response." One of the best-studied examples of a retrograde response exists in Saccharomyces cerevisiae where a complex system involving a number of positive and negative controlling molecules has been described (BUTOW and AVADHANI 2004; LIU and BUTOW 2006).

In Neurospora crassa, one aspect of the retrograde response can be studied by examining the induction of the alternative oxidase, encoded by the nuclear aod-1 gene. Alternative oxidase accepts electrons from ubiquinol and donates them directly to molecular oxygen. Under standard growth conditions, alternative oxidase is not present in N. crassa. However, when the cytochrome-mediated electron transport chain is inhibited through growth of cells in the presence of drugs such as antimycin A, which inhibits complex III (ZHANG et al. 1998), the gene is strongly induced and the protein can be found in mitochondria (LI et al. 1996; TANTON et al. 2003). The alternative pathway bypasses complexes III and IV of the standard electron transport chain and results in a net decrease in ATP production. However, under conditions where it is induced, alternative oxidase ensures that electron flow will continue, allowing the recycling of electron carriers and energy production via complex I (LAMBOWITZ et al. 1972).

Alternative oxidase has long been known to exist in higher plants, many fungi, and some protists (VANLERBERGHE and MCINTOSH 1997; JOSEPH-HORNE et al. 2001; VEIGA et al. 2003; CHAUDHURI et al. 2006). More recently, the enzyme has been found in certain bacteria (STENMARK and NORDLUND 2003; MCDONALD and VANLERBERGHE 2005) and a few animal phyla (MCDONALD and VANLERBERGHE 2004). The control of alternative oxidase production and activity is achieved through many different mechanisms in these organisms. In plants, alternative oxidase is typically encoded by multiple genes. Transcript levels are influenced by both developmental signals and stress responses (KEARNS et al. 1992; WHELAN et al. 1996; FINNEGAN et al. 1997; SAISHO et al. 1997, 2001; VANLERBERGHE and MCINTOSH 1997; CONSIDINE et al. 2001; DJAJANEGARA et al. 2002; KARPOVA et al. 2002; THIRKETTLE-WATTS et al. 2003; DOJCINOVIC et al. 2005; ZARKOVIC et al. 2005; CLIFTON et al. 2006; RHOADS et al. 2006). In addition, the majority of plant alternative oxidases have two conserved cysteine residues in the N terminus of the protein, which are involved in the post-translational regulation of alternative oxidase activity (UMBACH et al. 2006). Disulphide bond formation between cysteine residues on adjacent subunits of an alternative oxidase homodimer results in an inactive complex. Conversely, the enzyme is activated if the reduced forms of these cysteine residues interact with -keto acids. In trypanosomes, the half-life of alternative oxidase mRNA dramatically changes during differentiation, indicating post-transcriptional regulation (CHAUDHURI et al. 2002). Nuclear run-on and transcript analysis studies from Magnaporthe grisea showed that the alternative oxidase gene is constitutively transcribed. In noninducing conditions, the transcript is actively degraded by one or more factors that are sensitive to cycloheximide (YUKIOKA et al. 1998b). Nuclear run-on experiments in N. crassa have demonstrated that alternative oxidase is transcribed at a low constitutive level under normal growth conditions, despite the fact that transcript and protein are typically not observed in cells grown under these conditions (TANTON et al. 2003). However, addition of antimycin A to the growth medium resulted in a dramatic increase in the rate of alternative oxidase mRNA production, suggesting that, in N. crassa, significant regulation occurs at the level of transcription. The pathway(s) for transducing signals to the nucleus for alternative oxidase expression appear to be complex. In Candida albicans, it was suggested that a histidine kinase may be involved in the expression of alternative oxidase (HUH and KANG 2001). A search for regulatory mutants based on the expression of a reporter gene in N. crassa led to the identification of four novel genes (aod-4, aod-5, aod-6, and aod-7), along with one previously identified gene (aod-2), involved in the regulation of alternative oxidase expression (DESCHENEAU et al. 2005). A similar approach identified several genes required for the expression of the Arabidopsis thaliana enzyme. Most mutants were impaired in their ability to induce the alternative oxidase transcript in response to either antimycin A or the TCA cycle inhibitor monofluroacetate. However, one mutant defective in the ability to respond to antimycin A inhibition could still respond to monofluroacetate inhibition, providing evidence that separate signaling pathways exist (ZARKOVIC et al. 2005).

Analysis of the alternative oxidase promoter has been performed in plant species such as Sauromatun guttatum, Glycine max, and A. thaliana (RHOADS and MCINTOSH 1993; THIRKETTLE-WATTS et al. 2003; DOJCINOVIC et al. 2005). Although several putative promoter elements have been uncovered through comparative analysis, the functionality of these elements has not yet been examined. Systematic deletions within the A. thaliana AOX1a promoter have identified a 93-bp region necessary for induction of alternative oxidase in Arabidopsis (DOJCINOVIC et al. 2005). Sequence analysis of the M. grisea alternative oxidase promoter region revealed the presence of several potential regulatory elements. One strain that had a reduced ability to express the transcript under inducing conditions was shown to contain a mutation between the putative TATA box and the beginning of the coding sequence (YUKIOKA et al. 1998a). In N. crassa, it has been shown that the promoter elements required for proper expression of alternative oxidase are contained in a region 255 bp upstream of the aod-1 transcription start site (TANTON et al. 2003). Here, we report further analysis of the 255-bp region and identify an alternative oxidase induction motif (AIM) that is required for maximal expression of the enzyme under inducing conditions.

N. crassa strains and growth conditions:

The N. crassa strain, 7207 (aod-1 pan-2 A), harbors a frameshift mutation in aod-1. Neither aod-1 mRNA nor the alternative oxidase protein are detectable in the strain, even when cells are grown under inducing conditions (LI et al. 1996; TANTON et al. 2003). We suspect that the frameshift mutation leads to rapid degradation of the transcripts. Growth and handling of this Neurospora strain were performed as previously described (DAVIS and DE SERRES 1970). Antimycin A at 0.5 µg/ml was used to inhibit the cytochrome-mediated electron transport chain and to induce expression of alternative oxidase in transformation experiments. Bleomycin was used at 1.0 µg/ml. Strains of different Neurospora species were obtained from the Fungal Genetics Stock Center (FGSC): N. sitophila (FGSC no. 4202), N. intermedia (FGSC no. 6605), and N. tetrasperma (FGSC no. 2511).

Plasmid construction:

Generation of the plasmids pMMAX and pMCMAX was previously described (TANTON et al. 2003). Both plasmids contain the alternative oxidase structural gene with 374 bp of downstream sequence and a bleomycin resistance gene (AUSTIN et al. 1990) in a pUC119 vector. pMMAX contains 255 bp of sequence upstream from the previously identified aod-1 transcription start site (LI et al. 1996), while pMCMAX contains only 10 bp of the upstream sequence. To generate plasmids pP-1–pP-4, single-stranded DNA of pMMAX was subjected to site-directed mutagenesis using the primer FNA410, which restored a BglII site at position –255 that was lost in the original cloning procedure that created pMMAX. Using single-stranded DNA from this restored pMMAX (rpMMAX), a second BglII site was introduced at various locations within the aod-1 upstream region using one of four additional mutagenic primers, FNA395–FNA 398. The resulting plasmids were digested with BglII and then religated, so that small regions of aod-1 upstream sequence located between the BglII sites were removed. Diagrams of these constructs are shown in Figure 1A. The procedures for generating and mutagenizing single-stranded DNA were as described previously (KUNKEL et al. 1991).

View larger version (52K): In this window In a new window Download PPT slide   FIGURE 1.— A 52-bp region upstream of the aod-1 gene is required for efficient induction of alternative oxidase. (A) Schematic of the aod-1 gene and flanking sequences present in each of six constructs used in deletion analysis. Each construct contains all exons and introns of the aod-1 gene, as well as 374 bp downstream of the stop codon. The different constructs contain various amounts of sequence upstream of the previously identified +1 transcription start site. The two diagonal lines through the aod-1 gene indicate that the relative size of the gene is larger than shown. (B) Conidia from strain 7207 were transformed with the various deletion constructs and a "no DNA" negative control. The resulting transformants were plated on medium containing either bleomycin or antimycin A. The bleomycin plates were used to confirm that the transformation efficiencies for each of the six constructs were similar, as each contains a bleomycin resistance cassette. Antimycin A plates were used to assay for induction of the alternative oxidase protein, which would be produced only from constructs containing sufficient promoter sequence.

 
All other plasmids used in this study were generated through PCR mutagenesis of rpMMAX. Phosphorylated mutagenic primer (200 ng) was added to 17 µl of 1.1x PCR reaction mix (50 mM KCl, 10 mM Tris–Cl, pH 8.5, 1.5 mM MgCl₂, 0.1 mg BSA, 0.2 mM dNTPs), 1 µl of 10 mM NAD, 2 µl of rpMMAX (0.5 µg), 0.5 µl of DMSO, 1 µl (2.5 units) of Pfu polymerase, and 0.3 µl (12 units) of Taq ligase. The reaction was placed in a thermocycler and heated to 95° for 5 min. This was followed by 30 cycles of 95° for 1 min, 55° for 1 min, and 65° for 10 min. After PCR amplification, 0.5 µl (10 units) of DpnI was added and the mixture was incubated at 37° overnight. The addition of DpnI enriches the mixture for newly synthesized mutant DNA by degrading the methylated wild-type template. The entire solution was then used to transform Escherichia coli (XL-2) competent cells, which were subsequently spread over three Luria broth plates containing ampicillin. Plasmid DNA was isolated from several colonies and sequence analysis was performed to identify plasmids containing the desired mutations. For linker-scanning mutagenesis, 14-bp sequences were replaced in eight different regions of upstream sequence by the sequence ACGAGGATCCTAGC. The replacement sequence contained a BamHI site to facilitate identification of mutant plasmids, which were named pLSM-1–pLSM-8.

The mutagenic primers used in this study, as well as the plasmids resulting from mutagenesis, are listed in Table 1. All plasmids were examined by DNA sequence analysis to confirm the presence of the desired mutation. The DNA sequence of the entire aod-1 coding and upstream regions was determined for the wild-type plasmid rpMMAX and the mutant derivatives pP-2, pLSM2, pLSM6, pLSM7, pMCHA8b, pMCHA9b, pMCHA28b, pMCHA30, pMCHA66, and pMCHA67. Other than the changes designed for a specific plasmid, no deviations from the sequence in rpMMAX were seen in any plasmid with one exception: Plasmid pMCHA66 contained a single silent mutation in the aod-1 coding sequence.

View this table: In this window In a new window   TABLE 1 Mutagenic primers used in this study

 

Transformation of N. crassa:

Freshly isolated conidia were subjected to electroporation as described previously (TANTON et al. 2003). The original 45 µl of electroporation mix was diluted with 1 ml of 1 M sorbitol immediately following electroporation. After incubation at 30° for 60 min, 600 µl of the mixture was added to top-agar-containing antimycin A. After gentle, but thorough, mixing, the top agar was dispensed evenly over three plates containing sorbose medium and antimycin A. To measure transformation to bleomycin resistance, 90 µl of the electroporation plus sorbitol mixture was added to top agar containing bleomycin, which was distributed over three plates containing sorbose medium plus bleomycin after mixing. Transformation plates containing antimycin A or bleomycin were incubated for 5 days or 4 days, respectively, before photography.

DNA sequence from other species:

Genomic DNA was isolated as described previously (TANTON et al. 2003) from the N. intermedia, N. sitophila, and N. tetrasperma strains listed above. Primers for PCR amplification were designed from sequence near the amino terminal coding region of the aod-1 gene (NCU07953.2) and the 5'-end of the next open reading frame (NCU07954.2) identified at the Neurospora genome sequencing project, assembly 7 (GALAGAN et al. 2003). This gave rise to a 2.5-kb PCR product, which included the aod-1 upstream region. The product was sequenced directly using primers within the aod-1 coding region. Cloning of the Gelasinospora aod-1 region was previously described (TANTON et al. 2003). The exact species of Gelasinospora used was not known. The sequence of the Podospora anserina alternative oxidase region was taken from GenBank (accession no. AF321004). Relevant sequences from Chaetomium globosom, Fusarium graminearum, and M. grisea were obtained from the respective genome sequencing projects at the Broad Institute of Harvard and MIT (http://www.broad.mit.edu).

Determining transcript start and end sites:

Two methods were used to determine the start and end points of the aod-1 transcript. In the first, plaques containing aod-1 cDNA were selected from a -phage library constructed in the ZAPII vector system (Stratagene, LaJolla, CA) using poly(A) RNA isolated from wild-type N. crassa cells grown in the presence of chloramphenicol to induce alternative oxidase. Following in vivo excision to release the cloned cDNAs in plasmids, the upstream and downstream regions of the aod-1 insert were sequenced with appropriate primers and the start and end points of transcription were determined. In the second approach, DNA was extracted from an aliquot of the library and subjected to PCR using one primer near the N terminus of the aod-1 structural gene and another in a multiple cloning site of the phagemid just upstream from the site of insertion for the cloned cDNAs. Since the 5'–3' orientation of the inserts was directional during the cloning, these PCR products were predicted to contain the 5'-end of the cloned cDNAs. The PCR products were cloned and transformed into the E. coli strain XL-2 Blue. Individual plasmids were sequenced to reveal the start point of the cDNAs.

Other techniques:

E. coli transformation and plasmid isolation from E. coli cells (SAMBROOK and RUSSELL 2001) were performed as described previously. DNA sequencing was done using a BigDye Terminator version 3.1 cycle sequencing kit (Applied Biosystems, Foster City, CA) as per the supplier's instructions. Sequence was generated on either an ABI 3100 or a 3730 DNA sequence system by the Molecular Biology Services Unit, Department of Biological Sciences, University of Alberta.

Characterization of the aod-1 gene promoter through deletion analysis:

An earlier study (TANTON et al. 2003) showed that the promoter element(s) required for inducible expression of aod-1 are located between –255 and –10, relative to the previously determined +1 site of transcription (LI et al. 1996). To identify the location of potential regulatory elements more precisely, we constructed a series of plasmids, designated pP-1–pP-4, each carrying the aod-1 structural gene preceded by various lengths of upstream sequence (Figure 1A). These plasmids, along with the control plasmids rpMMAX (structural gene with upstream sequence extending to –255) and pMCMAX (structural gene with upstream sequence extending to –10), all contained a bleomycin resistance cassette, which was used as a control for successful transformation. All six constructs were independently transformed into conidia from the aod-1 mutant strain 7207 and the transformation mixture was spread onto medium containing bleomycin and medium containing antimycin A. Antimycin A inhibits complex III of the cytochrome-mediated respiratory chain resulting in induction of alternative oxidase in wild-type N. crassa cells, thereby allowing them to grow in the presence of the drug. However, cells that are unable to produce alternative oxidase cannot grow in antimycin A.

When conidia from 7207 were transformed with rpMMAX, pMCMAX, pP-1, pP-2, pP-3, or pP-4, a similar number of colonies appeared on plates containing bleomycin (Figure 1B), indicating similar transformation efficiencies for each plasmid. However, only conidia transformed with rpMMAX and pP-1 were able to produce robust growth when plated on media supplemented with antimycin A, as judged by the number and size of colonies that appeared on the plates (Figure 1B). A low number of small colonies, which may be due to a low level of basal transcription, are observed in pP-2 transformations. No colonies were observed following transformation with pP-3, pP-4, or pMCMAX. These results suggested that only rpMMAX and pP-1 contain sufficient promoter sequence to allow efficient expression of alternative oxidase. Thus, it appeared that a crucial regulatory element(s) required for expression of aod-1 was located within the 52-bp region present in pP-1, but absent in pP-2.

Linker-scanning mutagenesis:

To define further the region required for expression of alternative oxidase, linker-scanning mutagenesis was performed. PCR mutagenesis was employed with the plasmid rpMMAX to replace various 14-bp segments encompassing the 52-bp region of interest with a sequence of equal size (Figure 2A). A similar approach was taken to remove the putative TATA box sequence and a region directly upstream of the position previously identified as the +1 transcriptional start site by primer extension analysis (LI et al. 1996). The resulting plasmids were transformed into the aod-1 mutant strain 7207 and assayed for growth on plates containing either bleomycin or antimycin A.

View larger version (88K): In this window In a new window Download PPT slide   FIGURE 2.— Characterization of the aod-1 gene promoter through linker-scanning mutagenesis. (A) The aod-1 upstream sequence present in the plasmid rpMMAX. The 52-bp sequence found to be required for efficient induction of aod-1 is underlined. rpMMAX was subjected to linker-scanning mutagenesis to give eight plasmids (pLSM-1–pLSM-8). In each, a different 14-bp region, indicated above the sequence, was replaced with the sequence ACGAGGATCCTAGC. The bases shown in boldface represent the putative TATA box and the aod-1 start codon (ATG). The previously determined +1 transcription start site is indicated with an arrow. (B) Growth of 7207 conidia transformed with the various control and linker-scanning mutagenesis constructs. Transformation and plating were as in the legend to Figure 1B.

 
All the transformants displayed a similar amount of growth on bleomycin plates (Figure 2B). The growth patterns observed on plates containing antimycin A were as expected for the control plasmids rpMMAX and pMCMAX (Figure 2B). The construct lacking the sequences upstream of the putative +1 transcription start site was also able to rescue the aod-1 mutant strain (Figure 2B, pLSM 3). We initially considered this to mean that the upstream sequence context of the +1 site was unimportant for the initiation of transcription. However, further work showed that there are multiple sites that can serve as transcription initiation sites and that the originally identified +1 site is not a major initiation point (see below). Replacing the putative TATA box rendered the construct incapable of producing vigorous growth on antimycin A when transformed into the aod-1 mutant strain (Figure 2B, pLSM 2). This supports the notion that this sequence represents the authentic TATA box. A construct (pLSM-1) with a 14-bp region changed between the 52-bp region of interest and the TATA box restored the ability of strain 7207 to grow in the presence of antimycin A, demonstrating that no critical elements for transcription occur in the affected region. Five constructs (pLSM-4–pLSM-8) contained 14-bp linker-scanning mutations, which collectively span the 52-bp region of interest plus some flanking sequence. When these constructs were transformed into the aod-1 mutant strain, there were obvious differences in their ability to confer growth in the presence of antimycin A (Figure 2B). The pLSM-4, pLSM-5, and pLSM-8 constructs gave transformation results similar to the positive control rpMMAX. However, pLSM6 and pLSM7 were similar to pLSM2, being inefficient at conferring the ability to grow in the presence of antimycin A. The poor growth on these plates suggests that the regulatory element(s) responsible for alternative oxidase induction are contained within a 28-bp sequence that extends from –109 to –136.

Identification of an AIM:

Inspection of the sequence that constitutes the 28-bp region that is mutated in pLSM6 and pLSM7 identified two CGG repeats, separated by 7 bp. This seemed significant as pairs of CGG triplets arranged as inverted repeats, direct repeats, or everted repeats are known to act as binding sites for proteins belonging to the Zn(II)2Cys6 binuclear cluster (also known as C₆ zinc clusters) family of transcription factors (SCHJERLING and HOLMBERG 1996; SCHWABE and RHODES 1997; TODD and ANDRIANOPOULOS 1997). For example, the Hap1 protein of S. cerevisiae, which regulates a number of genes involved in respiration, binds direct CGG repeats separated by 6 bp (ZHANG and GUARENTE 1996; ZHANG and HACH 1999). To determine if these repeats were involved in the induction of alternative oxidase, PCR mutagenesis was employed to create plasmids carrying substitutions for each of the six nucleotides that compose the CGG repeats, as well as for four surrounding nucleotides (Figure 3A). These constructs were transformed into the aod-1 mutant strain 7207, and the resulting conidia were plated onto medium containing either bleomycin or antimycin A. A substitution in any of the four nucleotides in the region surrounding the CGG repeats did not affect the ability of that construct to promote expression of alternative oxidase, as indicated by the abundant growth on plates containing antimycin A (Figure 3B). Conversely, conidia transformed with constructs containing mutations to any of the six nucleotides present in the CGG repeats were greatly reduced in their ability to grow in the presence of antimycin A (Figure 3B).

View larger version (86K): In this window In a new window Download PPT slide   FIGURE 3.— Identification of an AIM. (A) PCR mutagenesis of rpMMAX was used to introduce various mutations in the 28-bp region required for alternative oxidase induction that was identified through linker-scanning mutagenesis. These mutations targeted the six nucleotides that form the two CGG trinucleotides (in boldface type), as well as four surrounding nucleotides. The names of the mutant plasmids are shown on the left and the altered bases in each are indicated under the wild-type sequence. Boxes indicate the mutations affecting the CGG triplets. (B) Growth of 7207 conidia transformed with the various control and mutant constructs. Transformation and plating were as in the legend to Figure 1B.

 
If the identified sequence is functionally significant, it should be conserved in closely related species. We isolated genomic DNA from N. intermedia, N. sitophila, and N. tetrasperma strains, amplified the aod-1 upstream region of each by PCR, and determined the sequence of the PCR products directly. Neurospora is a member of the order Sordariales, and the sequence of the upstream region from Gelasinospora spp., P. anserina, and C. globosum was also examined as these are more distantly related species within the same order (HUHNDORF et al. 2004). In all of these species, perfectly conserved CGG repeats and a TATA box region were identified in positions relatively similar to those in N. crassa (Figure 4). The CGG repeats were all separated by 7 bp. Together, these data suggest that we have identified an AIM, consisting of two CGGs separated by 7 bp, which is required for inducible expression of aod-1. Inspection of the sequence in F. graminearum and M. grisea, which are not included in the Sordariales, revealed no analogous sequence (not shown).

View larger version (16K): In this window In a new window Download PPT slide   FIGURE 4.— Conservation of AIM sequence in other species. Regions identified as potential homologs of the N. crassa AIM and TATA box regions in six other species of the order Sordariales are shown. The two CGG repeats and the completely conserved bases in the TATA box region are boxed. Completely conserved bases in the spacer between the CGG triplets are underlined. All bases that deviate from the N. crassa sequence are in lowercase. The ATG sequence on the right indicates the beginning of the aod-1 coding sequence. Nc, N. crassa; Ni, N. intermedia; Nt, N. tetrasperma; Ns, N. sitophila; Gs, Gelasinospora spp.; Pa, P. anserina; Cg, C. globosum.

 

Transcription start and end sites for the aod-1 gene:

The finding that the removal of the sequence immediately upstream of the previously identified +1 transcription start site had no effect on the ability of the mutant construct to impart robust growth in antimycin A led us to reexamine the position of the +1 site. We constructed a -phage cDNA library from poly(A) RNA isolated from N. crassa grown in the presence of chloramphenicol. Chloramphenicol is an inhibitor of mitochondrial translation (SCHLUNZEN et al. 2001), and the resulting deficiency in oxidative phosphorylation complexes of the mitochondrial inner membrane leads to induction of alternative oxidase. Twelve cDNA clones were isolated from the library, and the transcription start site was determined by DNA sequencing. In addition, we used a PCR approach to isolate 36 individual clones representing the 5'-end of alternative oxidase transcripts from the same library (see MATERIALS AND METHODS). As shown in Figure 5A, this analysis suggests that a number of positions can serve as the transcription start site of the gene. Surprisingly, none of the 20 different start sites identified matched the start site predicted by primer extension analysis in our previous study (LI et al. 1996). Nonetheless, since there appears to be no obvious single major transcription start site, we suggest that the previous site still be referred to as +1 to avoid confusion in the literature.

View larger version (5K): In this window In a new window Download PPT slide   FIGURE 5.— Identification of transcription start and stop sites for the aod-1 gene. (A) Cloned cDNAs and PCR products (see MATERIALS AND METHODS) were sequenced to determine the start site of transcription. The start site previously determined by primer extension analysis (LI et al. 1996) is shown with an arrow. The TATA box region is boxed. The region altered by linker-scanning mutagenesis in pLSM-3 is underlined. The ATG start codon for the protein coding sequence is in boldface type. Numbers beneath bases indicate the number of times that a position was identified as a transcription start site. (B) Cloned cDNAs were examined for transcription end points. The TGA stop codon of the protein coding sequence is in boldface type. Numbers under bases indicate the number of times that a position was identified as a transcription termination site.

 
We also examined eight of the phage cDNA clones for the position of the transcription termination site. A major site was identified at a position 161 bp downstream from the stop codon of the protein coding sequence (Figure 5B).

We have used a qualitative in vivo assay that assesses, by simple inspection of the number and size of colonies observed following transformation, the ability of aod-1 constructs to restore efficient growth to aod-1 mutant cells plated on antimycin A. Using this assay and a series of plasmids containing different mutations in the region upstream of the aod-1 gene, we have identified an AIM that is required for efficient expression of the aod-1 gene in N. crassa. The motif consists of one pair of CGG repeats separated by 7 bp. Pairs of CGG triplets are known to be recognized by members of the Zn2Cys6 binuclear cluster family of transcription factors. These proteins typically bind as a hetero- or homodimer, with each subunit interacting with one of the trinucleotide repeats (SCHJERLING and HOLMBERG 1996). Although most proteins belonging to this family bind to inverted CGG repeats, others are known to bind direct repeats or everted versions of the triplet (SCHWABE and RHODES 1997). The importance of each of the bases in the CGG triplets of the AIM was shown by the severely reduced growth of mutant cells in the presence of antimycin A following transformation by constructs bearing mutations at any of the six positions within the CGG triplets. Thus, it seems likely that the AIM sequence is bound by a Zn2Cys6 cluster transcription factor that controls the transcription of aod-1 in N. crassa. Although most transcription factors in this family are activators of transcription, some are known to act as repressors (TODD and ANDRIANOPOULOS 1997; AKACHE et al. 2001). Since loss of the site reduces alternative oxidase expression, we predict that the AIM binds an activator of transcription.

The N. crassa AIM sequence is entirely conserved in N. tetrasperma, N. intermedia, and N. sitophila. This is not compelling evidence for the importance of the region since there are only 2–4 base sequence differences between any of the four species in the region between the AIM and TATA boxes (not shown). Similarly, the Gelasinospora AIM is identical to that of N. crassa, but there are only 12 base sequence differences in the region between the AIM and the TATA box in these two species. However, in the more distantly related species P. anserina and C. globosum, the CGG triplets with their 7-bp spacer region and the TATA box region are well conserved compared to N. crassa, but the sequence between these regions is not (not shown). This suggests that the AIM sequence is functionally conserved within these species, all of which are in the order Sordariales (HUHNDORF et al. 2004). We found no similar sequence within a region 500 bp upstream of the alternative oxidase coding sequence in either F. graminearum or M. grisea, which are members of the orders Hypocreales and Xylariales, respectively (CARLILE and WATKINSON 1997). Thus, the AIM sequence is apparently not a universal element for the induction of alternative oxidase transcription in fungi. Furthermore, since the AIM sequence is likely to be bound by a Zn2Cys6 binuclear cluster transcription factor, which is unique to fungi, alternative oxidases in other organisms will require different factors for transcriptional control.

For many Zn2Cys6 binuclear cluster proteins, the length of the spacer region, rather than its sequence, is an important determinant for binding affinity (SCHWABE and RHODES 1997) while, in others, the spacer sequence plays a significant role (NOËL and TURCOTTE 1998). Our sequence comparison of the AIM between species suggests that there may be conservation of some bases within the 7-bp spacer between the two CGG triplets. However, a construct carrying a mutation at one of these conserved positions (pMCHA29, Figure 3) was similar to the wild-type construct in its ability to restore growth in the presence of antimycin A to aod-1 mutant cells. Perhaps a more quantitative assay would reveal minor differences of induction with our mutant spacer construct.

In addition to the discovery of the AIM sequence, our studies have provided evidence that a putative TATA box, identified previously by inspection of the sequence, is necessary for efficient induction. Removal of the TATA box sequence dramatically reduced the ability of transformants to grow on antimycin A. Furthermore, the TATA box sequence is well conserved in the other Sordariales species. We had previously identified a +1 transcription start site using primer extension analysis (LI et al. 1996). However, removal of the sequence immediately upstream of this site did not affect transcription. A reevaluation of the position of the +1 site revealed a large number of start sites clustered in a region of 40 bp immediately downstream of the originally identified site.

Mutation of any of the bases in the CGG triplets resulted in a dramatic reduction in the number and size of colonies formed on antimycin A plates. However, a low number of small colonies were observed using these constructs. Similarly, low numbers of small colonies were observed in transformations involving the pP-2 construct, which completely lacks the AIM region. We have previously noted that there is a low level of constitutive transcription from the aod-1 gene. In addition, evidence for post-transcriptional regulation in aod-1 expression was observed (TANTON et al. 2003). One explanation for the presence of some small colonies on antimycin plates with mutant constructs lacking the AIM sequence might be that the presence of antimycin A inactivates systems that prevent expression of the usual constitutive transcripts, but they occur at such a low level that only weak growth is observed. Another possibility is that an additional sequence contained within the upstream sequence present in pP-2 is able to stimulate a low level of transcription. A low number of small colonies were also observed in the presence of antimycin A in transformations involving the linker-scanning mutagenesis construct (pLSM-2) that removed the TATA box. This suggests that some low level of transcription can be initiated using sequence that lacks this site. Presumably, another sequence present in the construct inefficiently fulfills the role of the TATA box.

Future work will be aimed at identifying the protein(s) that interact with the AIM sequence and determining how their function is regulated to control alternative oxidase expression. Several regulatory mutants affecting aod-1 expression have been described (DESCHENEAU et al. 2005) and one or more of these may define a transcription factor, which is likely to be a Zn2Cys6 homo- or heterodimer that activates aod-1 expression by binding to the AIM sequence.

This work was funded by a Discovery Grant from the Natural Sciences and Engineering Research Council (NSERC) to F.E.N. C.C.L. and K.E.K. were supported by summer scholarships from NSERC and the Alberta Heritage Foundation for Medical Research.

AKACHE, B., K. WU and B. TURCOTTE, 2001 Phenotypic analysis of genes encoding yeast zinc cluster proteins. Nucleic Acids Res. 29: 2181–2190.[Abstract/Free Full Text]

ATTARDI, G., and G. SCHATZ, 1988 The biogenesis of mitochondria. Annu. Rev. Cell Biol. 4: 289–333.[CrossRef][Medline]

AUSTIN, B., R. M. HALL and B. M. TYLER, 1990 Optimized vectors and selection for transformation of Neurospora crassa and Aspergillus nidulans to bleomycin and phleomycin resistance. Gene 93: 157–162.[CrossRef][Medline]

BARTLETT, K., and S. EATON, 2004 Mitochondrial beta-oxidation. Eur. J. Biochem. 271: 462–469.[Medline]

BISWAS, G., M. GUHA and N. G. AVADHANI, 2005 Mitochondria-to-nucleus stress signaling in mammalian cells: nature of nuclear gene targets, transcription regulation, and induced resistance to apoptosis. Gene 354: 132–139.[CrossRef][Medline]

BUTOW, R. A., and N. G. AVADHANI, 2004 Mitochondrial signalling: the retrograde response. Mol. Cell 14: 1–15.[CrossRef][Medline]

CARLILE, M. J., and S. C. WATKINSON, 1997 The Fungi. Academic Press/Harcourt Brace, London/Boston/San Diego/New York/Sydney/Tokyo.

CHAUDHURI, M., R. SHARAN and G. C. HILL, 2002 Trypanosome alternative oxidase is regulated post-transcriptionally at the level of RNA stability. J. Eukaryot. Microbiol. 49: 263–269.[Medline]

CHAUDHURI, M., R. D. OTT and G. C. HILL, 2006 Trypanosome alternative oxidase: from molecule to function. Trends Parasitol. 22: 484–491.[CrossRef][Medline]

CLIFTON, R., A. H. MILLAR and J. WHELAN, 2006 Alternative oxidases in Arabidopsis: a comparative analysis of differential expression in the gene family provides new insights into function of non-phosphorylating bypasses. Biochim. Biophys. Acta 1757: 730–741.[Medline]

CONSIDINE, M. J., D. O. DALEY and J. WHELAN, 2001 The expression of alternative oxidase and uncoupling protein during fruit ripening in mango. Plant Physiol. 126: 1619–1629.[Abstract/Free Full Text]

DAVIS, R. H., and F. J. DE SERRES, 1970 Genetic and microbiological research techniques for Neurospora crassa. Methods Enzymol. 17A: 70–143.

DESCHENEAU, A. T., I. A. CLEARY and F. E. NARGANG, 2005 Genetic evidence for a regulatory pathway controlling alternative oxidase production in Neurospora crassa. Genetics 169: 123–135.[Abstract/Free Full Text]

DJAJANEGARA, I., P. M. FINNEGAN, C. MATHIEU, T. MCCABE, J. WHELAN et al., 2002 Regulation of alternative oxidase gene expression in soybean. Plant Mol. Biol. 50: 735–742.[CrossRef][Medline]

DOJCINOVIC, D., J. KROSTING, A. J. HARRIS, D. J. WAGNER and D. M. RHOADS, 2005 Identification of a region of the Arabidopsis AtAOX1a promoter necessary for mitochondrial retrograde regulation of expression. Plant Mol. Biol. 58: 159–175.[CrossRef][Medline]

FINNEGAN, P. M., J. WHELAN, A. H. MILLAR, Q. ZHANG, M. K. SMITH et al., 1997 Differential expression of the multigene family encoding soybean mitochondrial alternative oxidase. Plant Physiol. 114: 455–466.[Abstract]

GALAGAN, J. E., S. E. CALVO, K. A. BORKOVICH, E. U. SELKER, N. D. READ et al., 2003 The genome sequence of the filamentous fungus Neurospora crassa. Nature 422: 859–868.[CrossRef][Medline]

GREEN, D. R., 2005 Apoptotic pathways: ten minutes to dead. Cell 121: 671–674.[CrossRef][Medline]

HUH, W., and S. KANG, 2001 Characterization of the gene family encoding alternative oxidase from Candida albicans. Biochem. J. 356: 595–604.[CrossRef][Medline]

HUHNDORF, S. M., A. N. MILLER and F. A. FERNANDEZ, 2004 Molecular systematics of the Sordariales: the order and the family Lasiosphaeriaceae redefined. Mycologia 96: 368–387.[Abstract/Free Full Text]

JOSEPH-HORNE, T., D. W. HOLLOMAN and P. M. WOOD, 2001 Fungal respiration: a fusion of standard and alternative components. Biochim. Biophys. Acta 1504: 179–195.[Medline]

KARPOVA, O. V., E. V. KUZMIN, T. E. ELTHON and K. J. NEWTON, 2002 Differential expression of alternative oxidase genes in maize mitochondrial mutants. Plant Cell 14: 3271–3284.[Abstract/Free Full Text]

KEARNS, A., J. WHELAN, S. YOUNG, T. E. ELTHON and D. A. DAY, 1992 Tissue-specific expression of the alternative oxidase in soybean and siratro. Plant Physiol. 99: 712–717.[Abstract/Free Full Text]

KUNKEL, T. A., K. BEBENEK and J. MCCLARY, 1991 Efficient site-directed mutagenesis using uracil-containing DNA. Methods Enzymol. 204: 125–139.[Medline]

LAMBOWITZ, A. M., E. W. SMITH and C. W. SLAYMAN, 1972 Electron transport in Neurospora mitochondria: studies on wild type and poky. J. Biol. Chem. 247: 4850–4858.[Abstract/Free Full Text]

LI, Q., R. G. RITZEL, L. T. T. MCLEAN, L. MCINTOSH, T. KO et al., 1996 Cloning and analysis of the alternative oxidase gene of Neurospora crassa. Genetics 142: 129–140.[Abstract]

LILL, R., and U. MÜHLENHOF, 2006 Iron-sulfur protein biogenesis in eukaryotes: components and mechanisms. Annu. Rev. Cell Dev. Biol. 22: 457–486.[CrossRef][Medline]

LISTER, R., O. CHEW, M.-N. LEE, J. L. HEAZLEWOOD, R. CLIFTON et al., 2004 A transcriptomic and proteomic characterization of the Arabidopsis mitochondrial protein import apparatus and its response to mitochondrial dysfunction. Plant Physiol. 134: 777–789.[Abstract/Free Full Text]

LIU, Z., and R. A. BUTOW, 2006 Mitochondrial retrograde regulation. Annu. Rev. Genet. 40: 159–185.[CrossRef][Medline]

MCDONALD, A. E., and G. C. VANLERBERGHE, 2004 Branched mitochondrial electron transport in the animalia: Presence of alternative oxidase in several animal phyla. IUBMB Life 56: 333–341.[Medline]

MCDONALD, A. E., and G. C. VANLERBERGHE, 2005 Alternative oxidase and plastoquinol terminal oxidase in marine prokaryotes of the Sargasso Sea. Gene 349: 15–24.[CrossRef][Medline]

NICHOLLS, D. G., 2005 Mitochondria and calcium signalling. Cell Calcium 38: 311–317.[CrossRef][Medline]

NOËL, J., and B. TURCOTTE, 1998 Zinc cluster proteins Leu3p and Uga3p recognize highly related but distinct DNA targets. J. Biol. Chem. 273: 17463–17468.[Abstract/Free Full Text]

RHOADS, D. M., and L. MCINTOSH, 1993 The salicylic acid-inducible alternative oxidase gene aox1 and genes encoding pathogenesis-related proteins share regions of sequence similarity in their promoters. Plant Mol. Biol. 21: 615–624.[CrossRef][Medline]

RHOADS, D. M., A. L. UMBACH, C. C. SUBBAIAH and J. N. SIEDOW, 2006 Mitochondrial reactive oxygen species: contribution to oxidative stress and interorganellar signaling. Plant Physiol. 141: 357–366.[Free Full Text]

SAISHO, D., E. NAMBARA, S. NAITO, N. TSUTSUMI, A. HIRAI et al., 1997 Characterization of the gene family for alternative oxidase from Arabidopsis thaliana. Plant Mol. Biol. 35: 585–596.[CrossRef][Medline]

SAISHO, D., M. NAKAZONO, K. H. LEE, N. TSUTSUMI, S. AKITA et al., 2001 The gene for alternative oxidase-2 (AOX2) from Arabidopsis thaliana consists of five exons unlike other AOX genes and is transcribed at an early stage during germination. Genes Genet. Syst. 76: 89–97.[CrossRef][Medline]

SAMBROOK, J., and D. W. RUSSELL, 2001 Molecular Cloning. A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

SCHJERLING, P., and S. HOLMBERG, 1996 Comparative amino acid sequence analysis of the C6 zinc cluster family of transcriptional regulators. Nucleic Acids Res. 24: 4599–4607.[Abstract/Free Full Text]

SCHLUNZEN, F., R. ZARIVACH, J. HARMS, A. BASHAN, A. TOCILJ et al., 2001 Structural basis for the interaction of antibiotics with the peptidyl transferase centre in eubacteria. Nature 413: 814–821.[CrossRef][Medline]

SCHWABE, J. W. R., and D. RHODES, 1997 Linkers made to measure. Nat. Struct. Biol. 4: 680–683.[CrossRef][Medline]

STENMARK, P., and P. NORDLUND, 2003 A prokaryotic alternative oxidase present in the bacterium Novosphingobium aromaticivorans. FEBS Lett. 552: 189–192.[CrossRef][Medline]

TANTON, L. L., C. E. NARGANG, K. E. KESSLER, Q. LI and F. E. NARGANG, 2003 Alternative oxidase expression in Neurospora crassa. Fungal Genet. Biol. 39: 176–190.[CrossRef][Medline]

THIRKETTLE-WATTS, D., T. C. MCCABE, R. CLIFTON, C. MOORE, P. M. FINNEGAN et al., 2003 Analysis of the alternative oxidase promoters from soybean. Plant Physiol. 133: 1158–1169.[Abstract/Free Full Text]

TODD, R. B., and A. ANDRIANOPOULOS, 1997 Evolution of a fungal regulatory gene family: the Zn(II)Cys6 binuclear cluster DNA binding motif. Fungal Genet. Biol. 21: 388–405.[CrossRef][Medline]

TRIFUNOVIC, A., A. WREDENBERG, M. FALKENBERG, J. N. SPELBRINK, A. T. ROVIO et al., 2004 Premature ageing in mice expressing defective mitochondrial DNA polymerase. Nature 429: 417–423.[CrossRef][Medline]

UMBACH, A. L., V. S. NG and J. N. SIEDOW, 2006 Regulation of plant alternative oxidase activity: a tale of two cysteines. Biochim. Biophys. Acta 1757: 135–142.[Medline]

VANLERBERGHE, G. C., and L. MCINTOSH, 1997 Alternative oxidase: from gene to function. Annu. Rev. Plant Physiol. Plant Mol. Biol. 48: 703–734.[CrossRef][Medline]

VEIGA, A., J. D. ARRABACA, F. SANSONETTY, P. LUDOVICO, M. CORTE-REAL et al., 2003 Energy conversion coupled to cyanide-resistant respiration in the yeasts Pichia membranifaciens and Debaryomyces hansenii. FEMS Yeast Res. 3: 141–148.[CrossRef][Medline]

WHELAN, J., A. H. MILLAR and D. A. DAY, 1996 The alternative oxidase is encoded in a multigene family in soybean. Planta 198: 197–201.[Medline]

WU, G., and S. M. J. MORRIS, 1998 Arginine metabolism: nitric oxide and beyond. Biochem. J. 336: 1–17.[Medline]

YUKIOKA, H., S. INAGAKI, R. TANAKA, K. KATOH, N. MIKI et al., 1998a Analysis of alternative oxidase gene in mutants of Magnaporthe grisea deficient in alternative, cyanide-resistant, resipiration. Annu. Phytopathol. Soc. Jpn. 64: 282–286.

YUKIOKA, H., S. INAGAKI, R. TANAKA, K. KATOH, N. MIKI et al., 1998b Transcriptional activation of the alternative oxidase gene of the fungus Magnaporthe grisea by a respiratory-inhibiting fungicide and hydrogen peroxide. Biochim. Biophys. Acta 1442: 161–169.[Medline]

ZARKOVIC, J., S. L. ANDERSON and D. M. RHOADS, 2005 A reporter gene system used to study developmental expression of alternative oxidase and isolate mitochondrial retrograde regulation mutants in Arabidopsis. Plant Mol. Biol. 57: 871–888.[CrossRef][Medline]

ZHANG, L., and L. GUARENTE, 1996 The C6 zinc cluster dictates asymmetric binding by HAP1. EMBO J. 15: 4676–4681.[Medline]

ZHANG, L., and A. HACH, 1999 Molecular mechanism of heme signaling in yeast: the transcriptional activator Hap1 serves as the key mediator. Cell. Mol. Life Sci. 56: 415–426.[CrossRef][Medline]

ZHANG, Z., L. HUANG, V. M. SHULMEISTER, Y. I. CHI, K. K. KIM et al., 1998 Electron transfer by domain movement in cytochrome bc1. Nature 392: 677–684.[CrossRef][Medline]

Communicating editor: M. S. SACHS

This article has been cited by other articles:

				C. H. Sellem, E. Bovier, S. Lorin, and A. Sainsard-Chanet Mutations in Two Zinc-Cluster Proteins Activate Alternative Respiratory and Gluconeogenic Pathways and Restore Senescence in Long-Lived Respiratory Mutants of Podospora anserina Genetics, May 1, 2009; 182(1): 69 - 78. [Abstract] [Full Text] [PDF]

				M. S. Chae, C. E. Nargang, I. A. Cleary, C. C. Lin, A. T. Todd, and F. E. Nargang Two Zinc-Cluster Transcription Factors Control Induction of Alternative Oxidase in Neurospora crassa Genetics, December 1, 2007; 177(4): 1997 - 2006. [Abstract] [Full Text] [PDF]

International Access Link

Online ISSN: 1943-2361  Print ISSN: 0016-6731
Copyright 2009 by the Genetics Society of America
phone: 412-268-1812   fax: 412-268-1813   email: genetics-gsa{at}andrew.cmu.edu