SEARCH

SEARCH BY CITATION

Keywords:

  • 3-ketoacyl-CoA thiolase;
  • Arabidopsis thaliana;
  • lipid metabolism;
  • peroxisomes;
  • germination;
  • seedling growth

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

3-ketoacyl-CoA thiolase (KAT) (EC: 2.3.1.16) catalyses a key step in fatty acid β-oxidation. Expression of the Arabidopsis thaliana KAT gene on chromosome 2 (KAT2), which encodes a peroxisomal thiolase, is activated in early seedling growth. We identified a T-DNA insertion in this gene which abolishes its expression and eliminates most of the thiolase activity in seedlings. In the homozygous kat2 mutant, seedling growth is dependent upon exogenous sugar, and storage triacylglycerol (TAG) and lipid bodies persist in green cotyledons. The peroxisomes in cotyledons of kat2 seedlings are very large, the total peroxisomal compartment is dramatically increased, and some peroxisomes contain unusual membrane inclusions. The size and number of plastids and mitochondria are also modified. Long-chain (C16 to C20) fatty acyl-CoAs accumulate in kat2 seedlings, indicating that the mutant lacks long-chain thiolase activity. In addition, extracts from kat2 seedlings have significantly decreased activity with aceto-acetyl CoA, and KAT2 appears to be the only thiolase gene expressed at significant levels during germination and seedling growth, indicating that KAT2 has broad substrate specificity. The kat2 phenotype can be complemented by KAT2 or KAT5 cDNAs driven by the CaMV 35S promoter, showing that these enzymes are functionally equivalent, but that expression of the KAT5 gene in seedlings is too low for effective catabolism of TAG. By comparison with glyoxylate cycle mutants, it is concluded that while gluconeogenesis from fatty acids is not absolutely required to support Arabidopsis seedling growth, peroxisomal β-oxidation is essential, which is in turn required for breakdown of TAG in lipid bodies.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Many seeds contain a store of lipid in the form of triacylglycerol (TAG), which provides carbon and energy to support post-germinative seedling growth (Bewley and Black, 1994). Utilization of this lipid requires lipase activity and β-oxidation of the fatty acids released to form acetyl CoA, which enters the glyoxylate cycle for conversion to 4-carbon acids and ultimately synthesis of sugars (Beevers, 1961). Recently it has been shown that in Arabidopsis plants lacking a functional glyoxylate cycle much of the TAG is respired during seedling growth (Eastmond et al., 2000a), showing that carbon can be diverted through the respiratory pathway. This observation raises questions about the importance of gluconeogenesis from fatty acids relative to utilization of other carbon sources in the seed, and about the normal contribution of β-oxidation to respiratory and gluconeogenic pathways. Furthermore, while current information indicates that peroxisomal β-oxidation is responsible for such lipid metabolism, the potential role of mitochondrial β-oxidation remains uncertain (Dieuaide et al., 1993; Harwood, 1988; Masterson and Wood, 2000). Mutants disrupted in specific steps in fatty acid catabolism could potentially be used to answer such questions.

In the β-oxidation spiral, long-chain fatty acids are catabolized to acetyl-CoA via the repeated cleavage of acetate units from their thiol ends. With each turn of the cycle the substrate chain length shortens by two carbon units. Consequently, the enzymes responsible for this process accommodate substrates of varying chain length. Peroxisomal β-oxidation is catalysed by three proteins: (i) acyl-CoA oxidase (ACX); (ii) the multifunctional protein (MFP) which exhibits 2-trans-enoyl-CoA hydratase, l-3-hydroxyacyl-CoA dehydrogenase, d-3-hydroxyacyl-CoA epimerase and Δ32-enoyl-CoA isomerase activities; and (iii) l-3-ketoacyl-CoA thiolase (KAT). Plants contain a family of ACX isozymes with different chain-length specificities (Hooks et al., 1996; Kirsch et al., 1986). Recently, four ACX isogenes encoding enzymes with long (ACX2), medium–long (ACX1), medium (ACX3), and short (ACX4) chain substrate specificities have been characterized from Arabidopsis (Eastmond et al., 2000b; Froman et al., 2000; Hayashi et al., 1999; Hooks et al., 1999). These genes all show increased expression during germination and post-germinative seedling growth. Multiple isozymes for MFP have also been reported in plants but, unlike the ACX isozymes, these share broad, overlapping substrate specificities (Behrends et al., 1988; Gühnemann-Schäfer and Kindl, 1995; Preisig-Müller et al., 1994). Of the two MFP genes that have been studied in Arabidopsis, only AtMFP2 shows a significant induction during germination and seedling growth (Eastmond and Graham, 2000). The second gene, AIM1, shows low expression during post-germinative growth and much stronger expression in older (>8 days) seedlings (Richmond and Bleecker, 1999). KAT cDNA clones from cucumber (Preisig-Müller and Kindl, 1993) and pumpkin (Kato et al., 1996), that are induced during germination, have also been described. An Arabidopsis KAT gene from chromosome 2 (herein referred to as KAT2) has been described in a study of mutants that have defects in peroxisomal β-oxidation (Hayashi et al., 1998). Further paralogues occur on chromosome 1 (KAT1) and chromosome 5 (KAT5) of Arabidopsis. There are no reports on their functional characterization. Information regarding substrate specificity of thiolase enzymes from plants is lacking, partly due to the difficulty in synthesizing the longer chain-length substrates.

Arabidopsis mutants disrupted in each of the three major steps of the β-oxidation pathway have been described (Eastmond et al., 2000b; Hayashi et al., 1998; Richmond and Bleecker, 1999). However, only one of these mutants, ped1 (peroxisome defective), shows a discernible seedling phenotype in that it requires an exogenous supply of sucrose for growth (Hayashi et al., 1998). ped1 seedlings also have abnormally large peroxisomes and lack KAT protein. In acx3 mutant seedlings, although dramatically reduced in C10 and C12 acyl-CoA oxidase activity, they do not show any growth defects and fatty acid breakdown is not affected (Eastmond et al., 2000b). The aim1 mutant shows abnormal inflorescence development (Richmond and Bleeker, 1999), but seedling growth is normal, presumably because the AtMFP2 gene encodes the predominant multifunctional protein responsible for fatty acid β-oxidation at this stage. However, seedlings of all three of these mutants exhibit varying degrees of resistance to 2,4-dichlorophenoxybutyric acid (2,4-DB), which is normally converted to the auxin analogue 2,4 dichlorophenoxyacetic acid (2,4D) by β-oxidation. Therefore β-oxidation, at least of this compound, is compromised to an extent in seedlings of all three mutants.

Of the β-oxidation mutants previously characterized, only ped1 showed a sucrose-dependent seedling growth phenotype which could be explained by a block in fatty acid β-oxidation. The ped1 locus maps close to the KAT2 gene, which contains a frameshift mutation in the mutant (Hayashi et al., 1998). This observation strongly suggests that the ped1 phenotype results from a defective KAT2 protein. The objective of the current work was to characterize the biochemical and ultrastructural changes caused by blocking peroxisomal fatty acid β-oxidation during the period of storage lipid mobilization in germinating seedlings. We therefore employed reverse genetics (Krysan et al., 1999) to isolate a kat2 mutant for characterization.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Isolation of a kat2 mutant

Arabidopsis plants containing random T-DNA insertions were screened using PCR with multiple primer combinations, for insertions in KAT genes. Of the 30 000 plants screened, one plant from the Versailles collection (Bechtold et al., 1993; Bouchez et al., 1993) was found to contain an insertion in the KAT2 gene, but no insertions in KAT1 and KAT5 genes were detected. DNA sequence analysis of the interrupted KAT2 gene showed that the T-DNA is inserted into intron 8 (Figure 1). PCR and DNA sequence analyses established that T-DNA sequences close to the left border are located towards the 5′ end of the gene, but that the extreme left border is absent. Furthermore, an unidentifiable sequence of 68 bp is located between KAT2 and T-DNA sequences (Figure 1). No PCR products were obtained using T-DNA primers in combination with primers specific for the 3′ end of the gene (KAT2-A2 or KAT2-A1), indicating that the T-DNA is truncated at this end. Absence of T-DNA borders and rearrangement of sequences upon insertion are not uncommon (McKinney et al., 1995; Sherson et al., 2000). Inheritance of kanamycin resistance (carried by the T-DNA) co-segregated with the interrupted kat2 gene, and inheritance ratios, indicated the presence of a single T-DNA insertion site (or more than one closely linked site). Southern blot experiments were consistent with the presence of only one or two T-DNA copies in the KAT2 gene, but the precise organization was not determined (results not shown). The heterozygous plant containing the T-DNA insertion was self-fertilized in order to isolate a homozygous kat2 mutant. The frequency of such homozygous kat2 mutants was very low, despite achieving a high frequency of seed germination by including sucrose in the medium. Like the ped1 mutant, germination of kat2 seed proceeds only as far as radicle emergence and then arrests, but seedling growth can be recovered by germination in the presence of 20 mm sucrose, and kat2 seedling growth is resistant to 2,4-DB (results not shown). These results are consistent with the proposal that the ped1 phenotype results from a mutation in the KAT2 gene (Hayashi et al., 1998).

image

Figure 1. Physical map of the kat2 gene.

The 14 exons (E) of the KAT2 gene are shown as black boxes. Locations of primers used for PCR amplification are shown as arrows. LB is the T-DNA left border; a question mark denotes an undefined end to the T-DNA insertion. Nucleotide sequence shows terminal three nucleotides of exon 8 (capitals); intron 8 sequence of KAT2 (lower case); and 68 nucleotides of unknown sequence (upper case, italics) adjoining left-border T-DNA sequence (upper case, bold). The LB primer sequence is underlined. Appropriate combinations of the primers shown could be used to demonstrate the presence of T-DNA in the KAT2 gene, the absence of a wild-type allele in homozygous kat2 mutant, and the absence of T-DNA in the wild-type segregant.

Download figure to PowerPoint

Enzyme activity and gene expression in the kat2 mutant

The activities of key enzymes of lipid metabolism were assayed at intervals during the germination and growth of mutant and wild-type seedlings in the presence of sucrose (Figure 2). These results show that thiolase activity increases in the wild type to a peak at 2.5 days after imbibition, and then declines. In contrast, thiolase activity in the mutant remains low throughout. It is not known if the low level of thiolase activity seen in kat2 is due to other 3-ketoacyl-CoA thiolases (EC: 2.3.1.16) or acetoacetyl-CoA thiolases (EC: 2.3.1.9), because the assay employs acetoacetyl CoA as substrate. However, results of Figure 2 establish that KAT2 is the major such thiolase acting during TAG metabolism in seedling growth. The activities of other enzymes involved in lipid metabolism were unchanged in kat2, including medium-chain acyl-CoA oxidase and glyoxylate cycle enzymes (Figure 2).

image

Figure 2. Activities of β-oxidation and glyoxylate cycle enzymes in seedlings of wild-type (●) and kat2 (○) mutant.

Seedlings were germinated on half-strength MS supplemented with 20 mm sucrose and grown at 20°C under continuous light (100 µmol m−2 sec−1). Seedlings were collected at regular intervals 0–5 days after imbibition, and enzyme activities assayed.

(a) Thiolase; (b) ACOX (using C12 : 0 acyl-CoA as substrate); (c) ICL; (d) MLS. Values shown are mean ± SE of three measurements on each, three batches of seedlings for thiolase and two batches of seedlings for ACOX, ICL and MLS.

Download figure to PowerPoint

To confirm that the KAT2 gene encodes a thiolase, a KAT2 cDNA lacking the putative N-terminal peroxisomal targeting sequence (PTS2) was expressed in Escherichia coli, and the recombinant protein purified (Figure 3a). This protein was shown to have acetoacetyl-CoA thiolase activity, with an apparent KM of 77.5 ± 10.5 µm and Vmax of 5000 ± 239 nmol min−1 mg−1 protein (Figure 3b). In comparison, the total acetoacetyl-CoA thiolase activity in extracts from 2-day-old seedlings had an apparent KM of 39.5 ± 7.4 µm and Vmax of 54.8 ± 2.8 nmol min−1 mg−1 protein (Figure 3c). Polyclonal antibodies were raised against this purified protein and their specificity was established using Western blotting of recombinant KAT1, KAT2 and KAT5 proteins. The antibodies reacted very strongly with KAT2 and KAT1, but recognition of KAT5 required an approximately fivefold greater amount of this protein to obtain an equivalent signal (results not shown). Western blot analysis of wild-type seeds and seedlings confirmed that KAT increases in amount to a peak at about day 2 following seed imbibition, and then decreases progressively (Figure 4a), consistent with assays of thiolase activity. However, no KAT is detected in the kat2 mutant, confirming that KAT2 is absent, and showing that KAT1 and KAT5 are not abundant in such seedlings.

image

Figure 3. Production of recombinant KAT2 and kinetic parameters of the enzyme.

(a) SDS–polyacrylamide gel electophoresis of total proteins from E. coli strain harbouring plasmid encoding KAT2, before (lane 1) and after (lane 2) induction with isopropyl-β-d-thiogalactopyranoside. The KAT2 protein was purified by two cycles of affinity chromatography and 2 µg analysed (lane 3).

(b) Apparent kinetic parameters of recombinant KAT2 protein. Enzyme activities with varying substrate concentrations are shown with Hanes plot inset.

(c) Apparent kinetic parameters of thiolases in crude extract of 2-day-old wild-type seedlings. Enzyme activities with varying substrate concentrations are shown with Hanes plot inset.

Download figure to PowerPoint

image

Figure 4. Gene expression in kat2.

(a) Western blots of KAT2 thiolase protein in wild-type (WT) and kat2 seedlings. Protein extracts were prepared from seedlings taken at regular intervals between 0 and 5 days after seed imbibition. For each sample, 10 µg of total protein was used for SDS–PAGE and immunoblot analysis using antibodies raised against KAT2.

(b) Expression of KAT and glyoxylate cycle genes in seedlings of kat2 mutant. Seeds of kat2 and wild type (WT) were germinated on half-strength MS medium and seedlings collected at daily intervals from 2 to 6 days as indicated. Total RNA was isolated and analysed by RT–PCR and Southern blot hybridization using the 32P-labelled cDNA probes shown. Filters were exposed to X-ray film for different lengths of time, sufficient to obtain a clear signal for each probe. In this particular experiment the kat2 sample from day 2 failed to produce PCR products.

Download figure to PowerPoint

Northern analysis failed to detect KAT2 mRNA in the mutant or KAT1 and KAT5 in either mutant or wild type (results not shown), so RT–PCR was employed for greater sensitivity. Analysis of gene expression using RT–PCR with wild-type and mutant seedlings showed that KAT2 mRNA is undetectable in the mutant, whereas KAT1 and KAT5 mRNAs are still present (Figure 4b). Detection of KAT1 and KAT5 mRNAs required a 48-fold longer exposure time than did KAT2, indicating that they are much less abundant in such seedlings (Figure 4b). Expression of KAT1, KAT5, malate synthase (MLS) and isocitrate lyase (ICL) genes are similar in mutant and wild type.

Lipid metabolism in kat2 seeds

The total lipid and TAG contents of seeds were assayed during seed germination and seedling growth in the presence of sucrose (Figure 5). Surprisingly, the TAG content of kat2 seeds is 25% lower than that of the wild type, and this is reflected in total lipid content (Figure 5a,b). This difference is also reflected in the weights of seeds harvested from wild-type and kat2 plants (results not shown), indicating that KAT2 has an important role in seed development. Secondly, TAG and the total amount of lipid decreases rapidly during growth of wild-type seedlings, but the amounts remain unchanged in kat2 (Figure 5a,b), even though seedling growth is the same in both due to inclusion of sucrose in the medium. Further analysis of the TAG shows firstly that the fatty acid composition of wild-type and kat2 seeds is the same (Figure 5c), despite a lower amount of TAG in kat2. Secondly, the amount of eicosenoic acid (20 : 1) which is specific for TAG (Lemieux et al., 1990) is almost completely depleted in wild-type seedlings, but remains unaltered in kat2. In contrast, the amount of 18 : 3 remains relatively high in the wild type and increases in the mutant, presumably reflecting membrane lipid synthesis. These observations show that KAT2 is required for TAG catabolism, and that products of other KAT genes cannot substitute for it during kat2 seedling growth.

image

Figure 5. Lipid content of wild-type and kat2 seeds and seedlings during germination and early post-germinative growth.

(a) Total fatty acid content of TAG in wild-type (●) and kat2 (○) seedlings.

(b) Total fatty acid content of total lipid in wild-type (●) and kat2 (○) seedlings.

(c) Profiles of individual fatty acids in total lipid in wild-type (black bars) and kat2 (shaded bars) in (A) dry seed; (B) day 0 after imbibition; (C) day 1; (D) day 1.5; (E) day 2; (F) day 3; (G) day 4; (H) day 5. In all cases growth conditions were as for Figure 2, and values shown are mean ± SE of measurements on three batches of 50 seedlings.

Download figure to PowerPoint

Fatty acids are activated to fatty acyl-CoA esters prior to β-oxidation. Given that we had found that disruption of the last step in β-oxidation blocks TAG breakdown, it was of interest to establish if the acyl-CoA profiles were also altered in the kat2 mutant. A novel method for extraction and quantification of acyl-CoA esters (Larson and Graham, 2001) was employed to determine the acyl-CoA profiles in kat2 and wild-type extracts from 2- and 5-day-old seedlings grown in the presence of sucrose (Figure 6). The level of 20 : 1 acyl-CoA is significantly greater in kat2 than in wild type after 2 days’ growth, and after 5 days the difference is more pronounced and can also be observed for 16 : 0, 18 : 0, 18 : 1, 20 : 0, 22 : 0 and 22 : 1 acyl-CoAs (Figure 6). This accumulation of long-chain acyl-CoAs in the kat2 mutant suggests a deficiency in long-chain 3-ketoacyl-CoA thiolase activity encoded by the KAT2 gene.

image

Figure 6. Acyl-CoA profiles for wild-type (black bars) and kat2 (shaded bars) seedlings.

(a) Two-day-old seedlings; (b) 5-day-old seedlings. Growth conditions were as for Figure 2. Values shown are mean ± SE of measurements on five separate batches of 20 mg FW seedlings.

Download figure to PowerPoint

Organelle structure in kat2

Light micrographs of cells of cotyledons from 5-day-old green seedlings clearly show the presence of lipid bodies in most kat2 cells, but not in wild-type cells (results not shown). Electron microscopy showed that the content of lipid bodies in kat2 cotyledons differed from one cell to another (Figure 7). A small proportion of cells distributed randomly throughout the section contained very large amounts of lipid (Figure 7e). Morphometric studies of micrographs showed that the lipid bodies range between 2 and 4% of cytoplasm area in most cells, but up to 33% in the lipid-rich cells (Table 1), whereas some cells contain only a few small lipid bodies, as in the wild type. While peroxisomes in wild-type cells are round or dumbbell-shaped and approximately 500 nm in diameter, those in kat2 range from 500 nm to 2.5 µm (Figure 7b–e), representing a volume range of up to 100-fold. In kat2 cells the peroxisomes are observed in close contact with the lipid bodies. Furthermore, structural changes are observed in these organelles. In lipid-rich cells the peroxisomes are 1.5–2 µm in diameter and their dense matrix is perforated with small cavities (Figure 7e). The presence of such peroxisomes is consistent with observations of etiolated cotyledons in ped1 made by Hayashi et al. (1998). Most of the other kat2 cells contain large peroxisomes, either spherical or irregularly shaped, with a homogeneous matrix lacking the perforated structure, but with rearrangements of the peroxisomal membrane in close contact with endoplasmic reticulum elements (Figure 7b,d). The size and structure of peroxisomes in the few lipid-depleted cells are very similar to those of the wild type.

image

Figure 7. Electron microscopy of 5-day-old cotyledon cells from wild-type and kat2.

(a) Wild type; (b–e) kat2. p, peroxisome; m, mitochondrion; c , chloroplast; lb, lipid body; cw, cell wall. The peroxisomal membrane is shown (arrows) closely associated with endoplasmic reticulum (b) , chloroplast (d) or lipid bodies (d). Scale bar, 500 nm.

Download figure to PowerPoint

Table 1.  Morphometric analysis of cotyledonary cells from wild-type and kat2 seedlings
CompartmentParameterWild typekat2 akat2 bkat2 c
  • Seedlings were 5 days old and grown under conditions described in Figure 2. Values are derived from 12 independent micrographs for each cell type. kat2 a and kat2 b cells represent a random selection typical of cells shown in Figure 7(b–d). kat2 c cells were chosen as typical of the lipid-rich cell type shown in Figure 7(e).

  • a

    As a percentage of cytoplasm area.

  • b

    Number of organelle sections per 1000 arbitrary units.

  • c Mean surface of an organelle section (µm2).

Lipid bodiesTotal surface of compartmenta0.062.424.0633.60
PeroxisomesNumberb11.606.863.6511.36
Surface of an organelle sectionc0.194.744.522.43
Total surface of compartmenta0.405.983.034.36
PlastidsNumberb80.5224.9442.4211.36
Surface of an organelle sectionc3.9413.755.235.83
Total surface of compartmenta58.3862.9740.6111.48
MitochondriaNumberb94.2328.5647.925.16
Surface of an organelle sectionc0.190.820.620.55
Total surface of compartmenta3.394.295.450.52

The mean area of the peroxisomal compartment in kat2 cells is eight to 15 times greater than that of wild-type cells, but peroxisome number is two- to threefold lower (Table 1). Remarkably, major changes in plastids and mitochondria are also seen in kat2 cells. In the lipid-rich cells containing perforated peroxisomes (Table 1, kat2 c) there is a dramatic decrease in organelle number. In the other cells the number of plastids and mitochondria is distinctly lower than in wild-type cells, but is compensated by an increase in organelle size which results in only a moderate change in the relative development of these compartments compared to the wild type.

Transgenic complementation of the kat2 mutant

To confirm that the phenotypes observed in the kat2 mutant are a consequence of the T-DNA insertion in the KAT2 gene, and to investigate why other putative thiolase genes apparently do not function in TAG utilization, transgenic complementation was undertaken. The CaMV 35S promoter was linked to full-length cDNAs encoding KAT2 and KAT5 which were then used to transform the kat2 mutant. As judged by the ability of transgenic seedlings to grow in the absence of exogenous sucrose, KAT2 cDNA readily complements kat2 (Table 2), confirming that the mutant seedling phenotype is attributable to lack of KAT2. A KAT2 cDNA in which the N-terminal 34 amino acid residues was deleted failed to complement the mutant (Table 2), even though the truncated protein was shown to have acetoacetyl-CoA thiolase activity (Figure 3). This indicates the importance of the type 2 peroxisomal targeting sequence (PTS2) at the N-terminus of KAT2 (Hayashi, 2000; Olesena et al., 1997). KAT5 cDNA could also complement kat2, though less readily than KAT2 cDNA (Table 2). These results indicate that KAT5 has a similar substrate specificity to that of KAT2, but is not expressed to a sufficiently high level in kat2 seedlings to substitute for KAT2. In the lines complemented with KAT2 and KAT5 cDNAs, microscopic analysis of 5-day-old cotyledons showed wild-type peroxisomes and a lack of lipid bodies (results not shown).

Table 2.  Complementation of kat2 by transformation with d35S::KAT cDNA constructs
Genetic backgroundConstructNumber of transgenic lines analysedPhenotypesa (%)Intermediary phenotypeHygromycin resistanced
Not germinatedkat2 phenotypebWT phenotypec
  • a

    Seeds germinated without sucrose.

  • b

    Growth stops after root emergence.

  • c

    Seedling fully established with green cotyledons.

  • d

    Calculated on the germinated seeds.

kat2  5.095.0000
kat2d35S::KAT235.7 ± 7.211.3 ± 7.274.3 ± 6.18.7 ± 7.185.7 ± 11.0
kat2d35S::ΔNKAT234.3 ± 2.595.7 ± 2.50078.3 ± 4.0
kat2d35S::KAT544.0 ± 3.244.8 ± 30.234.3 ± 32.117.0 ± 3.687.5 ± 10.3
Wild type  4.0095.01.00
Wild typed35S::KAT233.7 ± 2.1095.7 ± 1.50.7 ± 0.678.7 ± 7.0

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

Using a reverse genetics approach, we have isolated a kat2 knock-out mutant which shows sucrose-dependent and 2,4-DB-resistant seedling growth phenotypes. Furthermore, KAT activity is lost and peroxisome structure is altered in seedlings of the mutant. These phenotypes are the same as those observed in the ped1 mutant isolated by Hayashi et al. (1998). These authors further showed that the ped1 mutation maps close to the KAT2 gene, which contains a frameshift mutation leading to premature termination of translation and absence of detectable KAT2 protein. Our results are consistent with the proposal that the PED1 locus is most probably the KAT2 gene (Hayashi et al., 1998), and provide new information on the control of lipid body metabolism, peroxisome development, and the role of β-oxidation in seed development. We demonstrate that KAT2 has thiolase activity, that it requires the putative PTS2 targeting signal for function, and that KAT5 is functionally similar.

The most striking observation is that the kat2 mutation largely abolishes TAG mobilization and lipid body degradation during seedling growth. This observation implies that metabolism in the peroxisome influences metabolism in the lipid body. Specifically, lipase activity must be inhibited when fatty acid β-oxidation is blocked. The lipases responsible for TAG hydrolysis have not yet been identified. Any signal causing feedback inhibition of TAG metabolism in kat2 must occur very early during germination or post-germinative growth, as there is no appreciable decrease in TAG levels after germination. The mechanism for this is unknown, but we note that at day 2 only the C20 : 1 acyl-CoA levels are significantly higher in kat2, compared with the wild type. C20 : 1 fatty acid is found solely in TAG in plants (Lemieux et al., 1990), and it is tempting to speculate that the early accumulation of this acyl-CoA ester could be involved in the observed inhibition of TAG breakdown in kat2. The increase in acyl-CoA esters could also be involved in signalling the abnormal development of peroxisomes and other organelles in kat2.

Modifications in peroxisome size and number have been reported in yeast and mammalian cells. These modifications are due to mutations either in proteins involved in peroxisome biogenesis, or in peroxisomal β-oxidation activities including acyl-CoA oxidase and multifunctional enzyme type 2 (Chang et al., 1999; Smith et al., 2000; van Roermund et al., 2000). The results shown here, together with those of Hayashi et al. (1998), demonstrate that peroxisome size and structure are modified in Arabidopsis due to a defect in peroxisomal β-oxidation. This is most evident in peroxisomes in contact with very large lipid bodies (Figure 7e). However, there is marked heterogeneity from cell to cell, perhaps reflecting heterogeneity in gene expression and cell metabolism.

Our results show that KAT1, KAT5 or other thiolases do not substitute for KAT2 in fatty acid β-oxidation in kat2 mutant seedlings. In the case of KAT5, this appears to be simply because it is expressed at a much lower level, because when overexpressed using the 35S promoter it complements kat2. A T-DNA insertion has been found in the KAT5 gene, but no phenotype for the kat5 mutation has been reported (Ferreira da Rocha et al., 1996).

There are no published reports of the characterization of substrate specificities of plant 3-ketoacyl-CoA thiolases, reflecting the difficulties of preparing individual enzymes and of obtaining suitable substrates. The cDNAs encoding KAT1, KAT2 and KAT5 were expressed in E. coli, but only KAT2 gave a soluble protein from which thiolase activity could be assayed using acetoacetyl CoA as substrate (Figure 3). Several lines of evidence suggest that the KAT2 protein has broad substrate specificity. Firstly, gene expression analysis (Figure 4) indicates that KAT2 is the predominant ketoacyl-CoA thiolase gene expressed during the period of TAG breakdown in seedlings. Furthermore, the kat2 mutant lacks the typical peak in activity observed during seedling growth (Figure 2) when acetoacetyl CoA is used as substrate, and the overexpressed KAT2 protein exhibits acetoacetyl-CoA activity, demonstrating that this enzyme is active on short-chain ketoacyl-CoAs. Finally, the accumulation of long-chain acyl-CoAs in kat2 indicates that long-chain ketoacyl-CoA thiolase activity is missing. Therefore KAT2 appears to be involved in the β-oxidation of the full range of fatty acyl-CoAs, and its absence in kat2 blocks fatty acid β-oxidation. In contrast, the acx3 mutant which is disrupted in the medium-chain-specific ACX breaks down TAG as effectively as the wild type, presumably because the other isozymes can substitute for the missing activity.

We have previously suggested that products of β-oxidation such as acetate might activate expression of glyoxylate cycle genes during seedling growth (Graham et al., 1992). These genes in bacterial and fungal cells are activated by acetate (Bowyer et al., 1994; Maloy and Nunn, 1982). However, addition of exogenous acetate to cucumber cells failed to affect expression of ICL and MLS genes (Graham et al., 1994). The observation that glyoxylate cycle enzymes are synthesized in kat2 seedlings provides evidence that products of β-oxidation are not required to activate ICL and MLS genes, at least in large amounts or in continuous supply.

The absence of TAG metabolism, and the failure of kat2 seedlings to grow beyond radicle emergence, imply that such lipid metabolism is required to support seedling growth in Arabidopsis. In recent studies it has been found that absence of a functional glyoxylate cycle in isocitrate lyase and malate synthase mutants does not prevent TAG utilization or seedling growth (Eastmond et al., 2000a; Germain et al., 2000). In this case, the fatty acids are respired rather than serving as gluconeogenic substrates. One question which was raised by these studies is the route by which carbon from fatty acids is transferred to the mitochondrion for respiration in the absence of a functional glyoxylate cycle. Present studies with kat2 show that mitochondrial β-oxidation is not an option, as no depletion of TAG is observed when peroxisomal β-oxidation is blocked. Therefore it is assumed that in the absence of the glyoxylate cycle, peroxisomal β-oxidation takes place and acetate, citrate or isocitrate could be transferred to the mitochondrion. This conclusion, based on a genetic approach, is supported by physiological studies in lettuce and sunflower which suggested that citrate could be transferred from peroxisome to mitochondrion at stages of seedling growth before the glyoxylate cycle becomes established (Raymond et al., 1992; Salon et al., 1988). The key conclusion is that peroxisomal β-oxidation of fatty acids from TAG is required for seedling growth.

The small seed size and low TAG content of kat2 indicates that KAT2 has an important role in seed development. We know that in developing seeds, KAT2 (S. Footitt, J.H. Bryce and S.M. Smith, unpublished results) and KAT5 (Ferreira da Rocha et al., 1996) genes are expressed. Furthermore, β-oxidation of fatty acids is active in developing Arabidopsis seeds (Eccleston and Ohlrogge, 1998; Poirier et al., 1999), and the aim1 mutant, which is disrupted in the β-oxidation multifunctional protein, has seeds with modified lipid content (Richmond and Bleeker, 1999). Establishment of the importance of the KAT2 gene in β-oxidation now identifies this as a key target for control of the futile cycling that compromises yields of unusual fatty acids in transgenic crops (Eccleston and Ohlrogge, 1998).

Experimental procedures

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

DNA isolation, pooling strategy and mutant isolation by PCR screening

T-DNA-transformed Arabidopsis thaliana populations were obtained, DNA extracted and pooled, and PCR was performed exactly as described by Sherson et al. (2000). The following T-DNA primers were used: left border (LB) 5′-CTACAAATTGCCTTTTCTTATCGAC-3′; right border (RB) 5′-CTGATACCAGACGTTGCCCGCATAA-3′. Sense (S) or antisense (A) KAT2 gene-specific primers were: KAT2-S1 5′-CGAGAGACAACGCGTTCTTCTTGAGC-3′; KAT2-S2 5′-GAGTCCATGACTACCAATCCAATGCC-3′, KAT2-A1 5′-AGCTCATCAACTCCATCTCCTCTCTC-3′, KAT2-A2 5′-CCCAAGAGAAGCAAGAGTTGTGGTTG-3′. KAT2 cDNA was labelled using the random primer method and used to detect specific PCR products by Southern hybridization (Sambrook et al., 1989). The T-DNA/KAT2 junctions were PCR-amplified using the gene specific primer KAT2-S1 with the T-DNA LB primer. PCR products were purified and sequenced, and homozygous kat2 mutant and wild-type control were identified according to Sherson et al. (2000).

Gene expression analysis using reverse-transcriptase PCR

Oligonucleotides (25 bp) were designed to amplify part of the KAT, MLS, ICL and ACTIN-2 genes (sense primer in the coding region, antisense primer in the 3′ non-coding region). Total RNA was isolated using an RNeasy plant mini kit (Qiagen, Hilden, Germany) and cDNA synthesized from 2 µg starting total RNA, using Superscript II (RNase H-reverse transcriptase, Invitrogen Ltd, Paisly, UK) according to the manufacturer's protocol. cDNA (5 µl of a 10× dilution) was used in the PCR reaction (total volume 25 µl). Semi-quantitative cDNA amplification employed the following conditions: 95°C for 5 min, followed by 20 cycles of 95°C for 20 sec, 60°C for 20 sec, 72°C for 45 sec. As a negative control, PCR amplification was carried out on RNA to check genomic DNA contamination. PCR reactions (10 µl) were separated on a 1.2% (w/v) agarose gel (no band was visualized by ethidium bromide fluorescence) and transferred to Hybond N+ (Amersham, Braunschweig, Germany) using standard protocols (Sambrook et al., 1989). The hybridization probes were 32P-labelled cDNAs. The film exposure was 30 min for KAT2, MLS, ICL, ACTIN-2; and 24 h for KAT1 and KAT5.

Enzyme assays

Seedling extracts were prepared as described by Hooks et al. (1999). Acyl-CoA oxidase was assayed according to Hyrb and Hogg (1979); ICL, MLS and thiolase were assayed according to Cooper and Beevers (1969).

Preparation of KAT2, antibody and Western blot analysis

A KAT2 gene was obtained by PCR amplification (using Pfu DNA polymerase) of the A. thaliana (ecotype Columbia) thiolase cDNA clone 91A18T7 (accession AB008854) and cloning into the expression vector pET-15b (Novagen, Madison, Wisconsin USA), creating plasmid pET-KAT2. Escherichia coli strain BL21(DE3) containing pET-KAT2 was grown at 37°C in Luria–Bertani broth, and when cultures reached an OD600 of approximately 0.6, isopropyl-β-d-thiogalactopyranoside was added to a concentration of 0.4 mm to induce gene expression. Cultures were grown for an additional 4 h, then harvested by centrifugation. Recombinant His-tagged thiolase protein was isolated and purified according to Novagen protocols.

Purified KAT protein was used to raise antiserum in rabbits by the Scottish Antibody Production Unit. Pre-immune serum showed no cross-reactivity with KAT protein. Arabidopsis protein samples were prepared as described under Enzyme assays, and SDS–PAGE was performed according to Laemmli (1970). Total protein (10 µg per lane) was separated on a 7.5% polyacrylamide gel and blotted onto nitrocellulose membrane using a BioRad (Hemel Hempstead, Herts., UK) electroblotting system. The membrane was blocked with 3% (w/v) bovine serum albumin and 2% (w/v) non-fat milk in phosphate-buffered saline pH 7.5, and immunoblotted with a 1 : 2000 dilution of KAT2 antibody followed by a 1 : 30 000 dilution of peroxidase-conjugated goat anti-rabbit antibodies. Protein bands were visualized using a solution of FAST BCIP/NBT (Sigma, Poole, Dorset, UK).

Lipid analysis

Fatty acids were extracted and measured using the method of Browse et al. (1986). TAG was isolated by hydrophobic column chromatography (E. Rylott, T.R. Larson and I. Graham, unpublished results) before measurement of fatty acid content. Acyl-CoAs were extracted and quantified according to Larson and Graham (2001).

Electron microscopy

Five-day-old cotyledons of Arabidopsis seedlings grown on MS medium supplemented with 1% sucrose were sequentially fixed with 2.5% (w/v) glutaraldehyde, 1% (w/v) osmium tetroxide and 1% (w/v) tannic acid according to Carde (1987), and embedded in Epon. Ultrathin sections (45 nm) were collected on bare 600 mesh copper grids (Gilder Grids, Grantham, UK), stained with ethanolic uranyl acetate and aqueous lead citrate, and observed with an FEI CM10 electron microscope. For morphometric studies, cotyledons from wild-type and three different kat2 seedlings were analysed. For each sample, 12 random electron micrographs were analysed with Image-Pro Plus (Media Cybernetics, Silver Spring, USA). kat2 a and b images were derived from cells containing the subcellular organization typical of most cells as shown in Figure 7(a–d). kat2 c images were derived from cells with unusually high lipid body content, as represented by Figure 7(e). The area of the main cell compartments – cytoplasm (including nucleus), plastids, peroxisomes, mitochondria and lipid bodies – was measured for each micrograph. The mean area of an organelle section was calculated from the total area of this compartment by the number of sections.

Construction of transgenic plants

The KAT2 cDNA from pET-KAT2 was subcloned between the double 35S CAMV promoter (d35S) and the Nos terminator in the pBS-d35SNOS vector (V. Germain, unpublished results). An XbaI digest was used to transfer the entire cassette into pGreen 0129 (Hellens et al., 2000). Agrobacterium tumefaciens strain GV3101 (Koncz and Schell, 1986) transformed with pSoup (Hellens et al., 2000) was used to treat kat2 or wild-type plants by floral dip (Clough and Bent, 1998). Transformants were selected on 20 µg ml−1 hygromycin B and transferred to soil in the greenhouse. The transgenic plants were self-fertilized, and their progeny analysed for germination on medium with or without 1% (w/v) sucrose and for hygromycin (20 µg ml−1) resistance in the presence of 1% (w/v) sucrose. Seedling establishment was recorded after 6 days.

For complementation with ΔNKAT2, the truncated cDNA was generated by PCR on cDNA clone 91A18T7. For complementation with KAT5, a cDNA was obtained from RT–PCR carried out on total RNA from 5-day-old seedlings. Specific primers were designed using the KAT5 cDNA (accession AF062589). The cDNA was subcloned, as for KAT2 cDNA, in the appropriate vectors for expression in E. coli and for plant transformation.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References

We thank Martine Thomas, Patrick Benoist (University of Paris-Sud, Orsay), Luigi DeBellis and Laura Pistelli (University of Pisa), for sharing DNA preparations. We also thank Susan Forbes and Sandrine Fedou for technical support, and Alain Descamps for photographic work. This research was supported by BBSRC grant RSP07677 to S.M.S. and J.H.B., and by EC contract BIO4 CT96 0311 to S.M.S.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Results
  5. Discussion
  6. Experimental procedures
  7. Acknowledgements
  8. References
  • Beevers, H. (1961) Metabolic production of sucrose from fat. Nature, 191, 433436.
  • Bechtold, N., Ellis, J., Pelletier, G. (1993) In planta Agrobacterium-mediated gene transfer by infiltration of adult Arabidopsis thaliana plants. C.R. Acad. Sci. Paris, Life Sci. 316, 11941199.
  • Behrends, W., Engeland, K., Kindl, H. (1988) Characterisation of two forms of the multifunctional protein acting in fatty acid β-oxidation. Arch. Biochem. Biophys. 263, 161169.
  • Bewley, J.D. & Black, M. (1994). Seeds: Physiology of Development and Germination, 2nd edn. New York: Plenum.
  • Bouchez, D., Camilleri, C., Caboche, M. (1993) A binary vector based on basta resistance for in planta transformation of Arabidopsis thaliana. C.R. Acad. Sci. Paris, Life Sci. 316, 11881193.
  • Bowyer, P., De Ramon Lucas, J., Turner, G. (1994) Regulation of the expression of the isocitrate lyase gene (acuD) of Aspergillus nidulans. Mol. Gen. Genet. 242, 484489.
  • Browse, J., McCourt, P.J., Somerville, C.R. (1986) Fatty acid composition of leaf lipids determined after combined digestion and fatty acid methyl ester formation from fresh tissue. Anal. Biochem. 152, 141145.
  • Carde, J.P. (1987) Electron microscopy of plant cell membranes. In Plant Cell Membranes. Methods in Enzymology 148 (L.Packer and R.Douce, eds). New York: Academic Press , pp. 599–622.
  • Chang, C.C., South, S., Warren, D., Jones, J., Moser, A.B., Moser, H.W., Gould, S.J. (1999) Metabolic control of peroxisome abundance. J. Cell. Sci. 112, 15791590.
  • Clough, S.J. & Bent, A.F. (1998) Floral dip: a simplified method for Agrobacterium-mediated transformation of Arabidopsis thaliana. Plant J. 16, 735743.
  • Cooper, G. & Beevers, H. (1969) Mitochondria and glyoxysomes from castor bean endosperm. J. Biol. Chem. 244, 35073513.
  • Dieuaide, M., Couee, I., Pradet, A., Raymond, P. (1993) Effects of glucose starvation on the oxidation of fatty acids by maize root tip mitochondria and peroxisomes: evidence for mitochondrial fatty acid β-oxidation and acyl-CoA dehydrogenase activity in a higher plant. Biochem. J. 296, 199207.
  • Eastmond, P.J., Germain, V., Lange, P.R., Bryce, J.H., Smith, S.M., Graham, I.A. (2000a) Post-germinative growth and lipid catabolism in oilseeds lacking the glyoxylate cycle. Proc. Natl Acad. Sci. USA, 97, 56695674.DOI: 10.1073/pnas.97.10.5669
  • Eastmond, P.J. & Graham, I.A. (2000) The multifunctional protein AtMFP2 is coordinately expressed with other genes of fatty acid β-oxidation during seed germination in Arabidopsis thalaiana. Biochem. Soc. Trans. 28, 9599.
  • Eastmond, P.J., Hooks, M.A., Williams, D., Lange, P., Bechtold, N., Sarrobert, C., Nussaume, L., Graham, I.A. (2000b) Promoter trapping of a novel medium-chain acyl-CoA oxidase, which is induced transcriptionally during Arabidopsis seed germination. J. Biol. Chem. 275, 3437534381.
  • Eccleston, V.S. & Ohlrogge, J.B. (1998) Expression of lauroyl-acyl carrier protein thioesterase in Brassica napus seeds induces pathways for both fatty acid oxidation and biosynthesis and implies a set point for triacylglycerol accumulation. Plant Cell, 10, 613621.
  • Ferreira da Rocha, P.S.C., Topping, J.F., Lindsey, K. (1996) Promoter trapping in Arabidopsis– one T-DNA tag, three transcripts and two thiolase genes. J. Exp. Bot. 47, ss24.
  • Froman, B.E., Edwards, P.C., Bursch, A.G., Dehesh, K. (2000) ACX3, a novel medium-chain acyl-coenzyme A oxidase from Arabidopsis. Plant Physiol. 123, 733741.
  • Germain, V., Footitt, S., Dieuaide-Noubhani, M., Raymond, P., Renaudin, J.-P., Bryce, J.H., Smith, S.M. (2000) Role of malate synthase and the glyoxylate cycle in oilseed plants. Plant Mol. Biol. Rep. 18, S20S26.
  • Graham, I.A., Leaver, C.J., Smith, S.M. (1992) Induction of malate synthase gene expression in senescent and detached organs of cucumber. Plant Cell, 4, 349357.
  • Graham, I.A., Denby, K.J., Leaver, C.J. (1994) Carbon catabolite repression regulates glyoxylate cycle gene expression in cucumber. Plant Cell, 6, 761772.
  • Gühnemann-Schäfer, K. & Kindl, H. (1995) Fatty acid β-oxidation in glyoxysomes. Characterisation of a new tetrafunctional protein (MFP III). Biochim. Biophys. Acta, 1256, 181186.
  • Harwood, J.L. (1988) Fatty acid metabolism. Annu. Rev. Plant Physiol. 39, 101138.
  • Hayashi, M. (2000) Plant peroxisomes: molecular basis of the regulation of their functions. J. Plant Res. 113, 103109.
  • Hayashi, M., Toriyama, K., Kondo, M., Nishimura, M. (1998) 2,4-dichlorophenoxybutyric acid-resistant mutants of Arabidopsis have defects in glyoxysomal fatty acid acid β-oxidation. Plant Cell, 10, 183195.
  • Hayashi, H., De Bellis, L., Ciurli, A., Kondo, M., Hayashi, M., Nishimura, M. (1999) A novel acyl-CoA oxidase that can oxidise short-chain acyl-CoA in plant peroxisomes. J. Biol. Chem. 274, 1271512721.DOI: 10.1074/jbc.274.18.12715
  • Hellens, R.P., Edwards, E.A., Leyland, N.R., Bean, S., Mullineaux, P.M. (2000) pGreen: a versatile and flexible binary Ti vector for Agrobacterium-mediated plant transformation. Plant Mol. Biol. 42, 819832.DOI: 10.1023/a:1006496308160
  • Hooks, M.A., Bode, K., Couee, I. (1996) Higher-plant medium- and short-chain acyl-CoA oxidases: identification, purification and characterisation of two novel enzymes of eukaryotic peroxisomal β-oxidation. Biochem. J. 320, 607614.
  • Hooks, M.A., Kellas, F., Graham, I.A. (1999) Long-chain acyl-CoA oxidases of Arabidopsis. Plant J. 19, 113.
  • Hyrb, D.J. & Hogg, J.F. (1979) Chain-length specificities of peroxisomal and mitochondrial β-oxidation in rat liver. Biochem. Biophys. Res. Commun. 87, 12001206.
  • Kato, A., Hayashi, M., Takeuchi, Y., Nishimura, M. (1996) cDNA cloning and expression of a gene for 3-ketoacyl-CoA thiolase in pumpkin cotyledons. Plant Mol. Biol. 31, 843852.
  • Kirsch, T., Loffler, H.G., Kindl, H. (1986) Plant acyl-CoA oxidase: purification, characterisation and monomeric apoprotein. J. Biol. Chem. 261, 85708575.
  • Koncz, C. & Schell, J. (1986) The promoter of TL-DNA gene 5 controls the tissue-specific expression of chimeric genes carried by a novel type of Agrobacterium binary vector. Mol. Gen. Genet. 204, 383396.
  • Krysan, P.J., Young, J.C., Sussman, M.R. (1999) T-DNA as an insertional mutagen in Arabidopsis. Plant Cell, 11, 22832290.
  • Laemmli, U.K. (1970) Cleavage of structural proteins during the assembly of the head of the bacteriophage T4. Nature, 227, 680685.
  • Larson, T.R. & Graham, I.A. (2001) A novel technique for the sensitive quantification of acyl-CoA esters from plant tissues. Plant J. 25, 115125.DOI: 10.1046/j.1365-313x.2001.00929.x
  • Lemieux, B., Miquel, M., Somerville, C., Browse, J. (1990) Mutants of Arabidopsis with alterations in seed lipid fatty acid composition. Theor. Appl. Genet. 80, 234240.
  • Maloy, S.R. & Nunn, W.D. (1982) Genetic regulation of the glyoxylate shunt in Escherichia coli K-12. J. Bact. 149, 173180.
  • Masterson, C. & Wood, C. (2000) Mitochondrial β-oxidation of fatty acids in higher plants. Physiol. Plantarum, 109, 217224.
  • McKinney, E.C., Nazeem, A., Traut, A., Feldmann, K.A., Belostotsky, D.A., McDowell, J.M., Meagher, R.B. (1995) Sequence-based identification of T-DNA insertion mutations in Arabidopsis: actin mutants act2-1 and act4-1. Plant J. 8, 613622.
  • Olesena, C., Thomsena, K.K., Svendsenb, I., Brandta, A. (1997) The glyoxysomal 3-ketoacyl-CoA thiolase precursor from Brassica napus has enzymatic activity when synthesised in Escherichia coli. FEBS Lett. 412, 138140.DOI: 10.1016/s0014-5793(97)00766-7
  • Poirier, Y., Ventre, G., Caldelari, D. (1999) Increased flow of fatty acids toward β-oxidation in developing seeds of Arabidopsis deficient in diacylglycerol acyltransferase activity or synthesising medium-chain-length fatty acids. Plant Physiol. 121, 13591366.
  • Preisig-Müller, R. & Kindl, H. (1993) Thiolase mRNA translated in vitro yields a peptide with a putative N-terminal presequence. Plant Mol. Biol. 22, 5966.
  • Preisig-Müller, R., Gühnemann-Schäfer, K., Kindl, H. (1994) Domains of the tetrafunctional protein acting in glyoxysomal fatty acid β-oxidation. J. Biol. Chem. 269, 2047520481.
  • Raymond, P., Spiteri, A., Dieuaide, M., Gerhardt, B., Pradet, A. (1992) Peroxisomal β-oxidation of fatty acids and citrate formation by a particulate fraction from early germinating sunflower seeds. Plant Physiol. Biochem. 30, 153161.
  • Richmond, T.A. & Bleecker, A.B. (1999) A defect in β-oxidation causes abnormal inflorescence development in Arabidopsis. Plant Cell, 11, 19111923.
  • Salon, C., Raymond, P., Pradet, A. (1988) Quantification of carbon fluxes through the tricarboxylic acid cycle in early germinating lettuce embryos. J. Biol. Chem. 263, 1227812287.
  • Sambrook, J., Fritsch, E.F., Maniatis, T. (1989) Molecular Cloning. A Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press.
  • Sherson, S.M., Hemmann, G., Wallace, G., Forbes, S., Germain, V., Stadler, R., Bechtold, N., Sauer, N., Smith, S.M. (2000) Monosaccharide/proton symporter AtSTP1 plays a major role in uptake and response of Arabidopsis seeds and seedlings to sugars. Plant J. 24, 849858.DOI: 10.1046/j.1365-313x.2000.00935.x
  • Smith, J.J., Brown, T.W., Eitzen, A., Rachubinski, R.A. (2000) Regulation of peroxisome size and number by fatty acid β-oxidation in the yeast Yarrowia lipolytica. J. Biol. Chem. 275, 2016820178.
  • Van Roermund, C.W.T., Tabak, H.F., Van Den Berg, M., Wanders, R.J.A., Hettema, E.H. (2000) Pex11 plays a primary role in medium-chain fatty acid oxidation, a prcess that affects peroxisome number and size in Saccharomyces cerevisiae. J. Cell Biol. 150, 489498.