Identification and characterization of Arabidopsis indole-3-butyric acid response mutants defective in novel peroxisomal enzymes

Abstract

Genetic evidence suggests that indole-3-butyric acid (IBA) is converted to the active auxin indole-3-acetic acid (IAA) by removal of two side-chain methylene units in a process similar to fatty acid beta-oxidation. Previous studies implicate peroxisomes as the site of IBA metabolism, although the enzymes that act in this process are still being identified. Here, we describe two IBA-response mutants, ibr1 and ibr10. Like the previously described ibr3 mutant, which disrupts a putative peroxisomal acyl-CoA oxidase/dehydrogenase, ibr1 and ibr10 display normal IAA responses and defective IBA responses. These defects include reduced root elongation inhibition, decreased lateral root initiation, and reduced IBA-responsive gene expression. However, peroxisomal energy-generating pathways necessary during early seedling development are unaffected in the mutants. Positional cloning of the genes responsible for the mutant defects reveals that IBR1 encodes a member of the short-chain dehydrogenase/reductase family and that IBR10 resembles enoyl-CoA hydratases/isomerases. Both enzymes contain C-terminal peroxisomal-targeting signals, consistent with IBA metabolism occurring in peroxisomes. We present a model in which IBR3, IBR10, and IBR1 may act sequentially in peroxisomal IBA beta-oxidation to IAA.

Figure 1.—

ibr1 and ibr10 mutants display IBA- and 2,4-DB-resistant root elongation. (A) ibr mutant root elongation on IBA and IAA. Col-0 (Wt), ibr10-1, and six ibr1 alleles were plated on medium containing 0.5% sucrose with no hormone, 15 μm IBA or 120 nm IAA. Root length was measured after 7 days. Error bars show standard errors of mean root lengths (n ≥ 13). (B) ibr mutant root elongation in response to increasing IBA concentrations. Seeds were stratified for 3 days at 4° in 0.1% agar prior to incubation under white light for 2 days at 22°. Germinating seeds were transferred to medium containing 0.5% sucrose (no hormone) or supplemented with the indicated concentration of IBA. Roots were measured after 8 additional days of growth under yellow-filtered light. Error bars show standard errors of mean root lengths (n ≥ 10). ibr3-1 (Zolman et al. 2007) is included for comparison with ibr10 and ibr1. chy1-3 (Zolman et al. 2001a) is included as an IBA-resistant control. (C) ibr mutant root elongation on synthetic auxins. Roots were measured as in B. Error bars show the standard errors of mean root lengths (n ≥ 12). chy1-3 is included as a 2,4-DB-resistant control. (D) The ibr1 ibr3 ibr10 triple mutant remains more IBA responsive than the ibr3 chy1 double mutant. Root elongation was evaluated as in B. Error bars show standard errors of mean root lengths (n ≥ 8).

Figure 2.—

ibr1 and ibr10 mutants fail to induce lateral roots in response to IBA. (A) ibr mutant lateral root initiation. Seeds were stratified for 3 days at 4° in 0.1% agar prior to plating on medium containing 0.5% sucrose and incubating under white light for 4 days at 22°. Seedlings then were transferred to medium containing 0.5% sucrose with no hormone or supplemented with the indicated concentration of auxin. Roots were measured and lateral roots were counted after 4 additional days of growth. Data are presented as the mean number of lateral roots per millimeter of root length and error bars show the standard error of the means (n ≥ 8). chy1-3 is included as an IBA-resistant control. (B) Visualization of ibr10 lateral root initiation defects using the DR5-GUS reporter. Wild type or ibr10 containing the DR5-GUS transgene were germinated on medium containing 0.5% sucrose and grown for 4 days and then transferred to unsupplemented medium or medium containing IBA or NAA. After the indicated number of days, seedlings were removed from the medium, stained, and mounted for photography. Lateral root primordia of NAA-treated plants can be visualized in the higher magnification insets. Bars, 1 mm.

Figure 3.—

ibr1 and ibr10 mutants metabolize peroxisomal substrates similarly to wild type. (A) ibr mutant hypocotyl elongation in the absence of sucrose. Seeds were stratified for 4 days at 4° in 0.1% agar prior to plating on unsupplemented medium or medium containing 0.5% sucrose. Plates were placed under white light for 1 day at 22° and then transferred to the dark at 22° for 5 additional days. Error bars show standard errors of mean hypocotyl lengths (n ≥ 10). chy1-3 (Zolman et al. 2001a) is included as a sucrose-dependent control. (B) ibr mutant root elongation on propionate and isobutryate. Seeds were stratified for 3 days at 4° in 0.1% agar prior to incubation under white light for 2 days at 22°. Germinating seeds were transferred to medium containing 0.5% sucrose with or without 150 μm propionate or 350 μm isobutyrate. Roots were measured after 7 additional days of growth. Error bars show standard errors of mean root lengths (n ≥ 11). chy1-3 is included as a propionate- and isobutyrate-hypersensitive control (Lucas et al. 2007).

Figure 4.—

Positional cloning of IBR1 and complementation of ibr1 mutants. (A) Recombination mapping of ibr1 mutant alleles to chromosome 4. IBA-resistant F₂ plants were scored using PCR-based markers (above the bar) and the number of recombination events/total number of chromosomes scored is shown below the bar. The solid oval represents the chromosome 4 centromere. (B) Molecular lesions in ibr1 mutant alleles. Sequence analysis of the IBR1 (At4g05530) candidate gene (exons shown as solid boxes; introns shown as thin lines) using ibr1 mutant DNA revealed a C-to-T base-pair change in ibr1-1, altering a conserved Arg to Cys. ibr1-2 has a G-to-A base-pair change at the 3′ splice site of intron 2 (exon sequence is uppercase; intron sequence is lowercase). The ibr1-3 mutant allele has a 1-bp deletion, leading to a frameshift and premature termination at amino acid 95 (see Figure 5). The position of the ibr1-7 T-DNA insertion (SALK_010364) is shown by a triangle. ibr1-8 has a C-to-T base-pair change, altering a Ser residue to Phe. (C) PCR analysis of the ibr1-4 and ibr1-5 deletion alleles. Amplification reactions were performed on Col-0 (C), ibr1-2 (-2), ibr1-4 (-4), and ibr1-5 (-5) genomic DNA using primers upstream of IBR1, spanning exons 1 and 2, spanning exons 4–6, and downstream of the gene. Control reactions used primers amplifying IBR3 and IBR10. (D) Rescue of ibr1 defects in IBA-responsive root elongation inhibition by wild-type IBR1. Wild-type Col-0 (Wt), ibr1-1, ibr1-4, and ibr1-5 mutants and ibr1 lines containing either 35S-IBR1c (IBR1c) or pBIN-IBR1g (IBR1g) were germinated and grown on medium supplemented with 0.5% sucrose and 7.5 μm IBA for 8 days at 22°. Error bars indicate standard errors of the mean root lengths (n ≥ 12). (E) Rescue of ibr1 lateral root defects by wild-type IBR1. Lines used in D were assayed for lateral root production. Seeds were plated on medium with 0.5% sucrose and grown for 5 days under white light and then transferred to medium further supplemented with 10 μm IBA and grown for 4 additional days. Error bars indicate standard errors of the mean number of visible lateral roots per seedling (n ≥ 12).

Figure 5.—

IBR1 resembles short-chain dehydrogenase/reductase enzymes. An alignment comparing IBR1 to similar proteins from rice (Os), beetle (Tc; Tribolium castaneum), human (Hs), Tetrahymena (Tt), and the prokaryote Algoriphagus sp. PR1 (Al) was generated in the MegAlign program (DNAStar) using the Clustal W method. Amino acid residues identical in at least four sequences are against a solid background; hyphens indicate gaps introduced to maximize alignment. The ibr1 mutations are indicated above the sequence. The C-terminal PTS1 present in IBR1 and the three closest homologs is indicated by a shaded background; HsHEP27 is not targeted to peroxisomes (Pellegrini et al. 2002). Asterisks above the sequence denote three essential residues that compose the catalytic triad characteristic of SDR proteins (Kallberg et al. 2002).

Figure 6.—

Positional cloning of IBR10 and complementation of the ibr10 mutant. (A) Recombination mapping of ibr10 to the middle of chromosome 4, south of IBR1. IBA-resistant F₂ plants were scored using PCR-based markers (above the bar) and the number of recombination events/total number of chromosomes scored is shown below the bar. (B) Molecular lesion in ibr10-1. Sequence analysis of the IBR10 (At4g14430) candidate gene using mutant DNA revealed a 78-bp deletion in ibr10-1. This deletion removes 26 amino acids of the encoded protein (see Figure 7), but leaves the C terminus of the protein in frame. The position of the T-DNA insertion disrupting the upstream IBR10 homolog, AtHCD1/At4g14440, is indicated by a triangle. (C) Rescue of ibr10 defects in IBA-responsive root elongation inhibition by wild-type IBR10. Wild-type Col-0 (Wt), the ibr10 mutant, and ibr10 containing 35S-IBR10 were assayed as described in the legend for Figure 1B. Error bars show standard errors of mean root lengths (n ≥ 12). (D) Rescue of ibr10 lateral root defects by wild-type IBR10. Lines used in C were assayed for lateral root production, as described in the legend for Figure 2A. Data are presented as the mean number of lateral roots per millimeter of root length and error bars show the standard error of the means (n ≥ 12). (E) Root elongation of insertion mutants disrupting ibr10 homologs. T-DNA insertion mutants in the ibr10 homologs AtHCD1/At4g14440 (S_012852) and ECHIc/At1g65520 (S_036386) were assayed for IBA responses as described in the legend for Figure 1B. Error bars show standard errors of mean root lengths (n ≥ 11).

Figure 7.—

IBR10 resembles enoyl-CoA hydratases. An alignment comparing IBR10 to similar proteins from Arabidopsis (At), rice (Os), the bacterium Thermobifida fusca (Tf), Dictyostelium (Dd), and humans (Hs) was generated in the MegAlign program (DNAStar) using the Clustal W method. Amino acid residues identical in at least four sequences are against a solid background; hyphens indicate gaps introduced to maximize alignment. The ibr10-1 deletion is indicated above the sequence. The C-terminal PTS1 present in IBR10 and four of the homologs are indicated by a shaded background. The HsPECI homolog has an N-terminal extension compared to the Arabidopsis protein; for clarity, the first 95 amino acids of this protein are not shown. MFP2 is an Arabidopsis multifunctional protein acting in peroxisomal fatty acid β-oxidation that contains enoyl-CoA hydratase/isomerase activity in the first 200 amino acids of the protein (Richmond and Bleecker 1999); the C-terminal domains of MFP2 are not shown. A Glu residue important for activity in a Pseudomonas putida hydratase (Wong and Gerlt 2004), a yeast isomerase (Mursula et al. 2001), and a rat ECH (Holden et al. 2001) is indicated by an asterisk above the sequence.

Figure 8.—

Expression profiles of genes encoding IBR enzymes and enzymes acting in fatty acid β-oxidation. (A) A model for the conversion of IBA to IAA. General enzymatic requirements are shown to the right of the arrows along with the hypothesized positions for IBR3, IBR10, and IBR1 activity. (B) Relative expression levels (mRNA signal) for IBR3, IBR10, and IBR1 in several Arabidopsis tissues. Genevestigator (Zimmermann et al. 2004) data indicate that expression of the three IBR genes is similar among tissues. Error bars show the standard error of the mean expression levels; n = 121 for dry seeds, 26 for imbibed seeds, 944 for seedlings, 419 for young plants, 330 for roots, 619 for flowers, and 3 for senescing leaves. Data were retrieved from Genevestigator in June 2008. (C) Pathway of fatty acid β-oxidation, where R represents a side-chain extension of additional carbon units. In plant peroxisomes, step 1 is catalyzed by long-chain acyl-CoA synthetases (LACS6 and LACS7), step 2 is catalyzed by a family of acyl-CoA oxidases (ACX1-6), steps 3 and 4 are catalyzed by multifunctional proteins (AIM1 or MFP2), and step 5 is carried out by thiolase (PED1/KAT2, KAT1, and KAT5). (D) Relative expression levels (mRNA signal) for genes encoding fatty acid β-oxidation enzymes in several Arabidopsis tissues. Genevestigator data were collected as described in B.