The Plant Journal (2006) 45, 930–941 doi: 10.1111/j.1365-313X.2005.02650.

The Arabidopsis thaliana multifunctional protein gene (MFP2)

of peroxisomal b-oxidation is essential for seedling
establishment
Elizabeth L. Rylott1, Peter J. Eastmond1, Alison D. Gilday1, Steve P. Slocombe2, Tony R. Larson1, Alison Baker2,
Ian A. Graham1,*
1
CNAP, Department of Biology, University of York, PO Box 373, York YO10 5YW, UK, and
2
Centre for Plant Sciences, University of Leeds, Leeds LS2 9JT, UK

Received 11 October 2005; revised 16 November 2005; accepted 22 November 2005.

*
For correspondence (fax þ44 (0)1904 32 8762; e-mail iag1@york.ac.uk).

Summary
The multifunctional protein (MFP) of peroxisomal b-oxidation catalyses four separate reactions, two of which
(2-trans enoyl-CoA hydratase and L-3-hydroxyacyl-CoA dehydrogenase) are core activities required for the
catabolism of all fatty acids. We have isolated and characterized five Arabidopsis thaliana mutants in the MFP2
gene that is expressed predominantly in germinating seeds. Seedlings of mfp2 require an exogenous supply of
sucrose for seedling establishment to occur. Analysis of mfp2-1 seedlings revealed that seed storage lipid was
catabolized more slowly, long-chain acyl-CoA substrates accumulated and there was an increase in
peroxisome size. Despite a reduction in the rate of b-oxidation, mfp2 seedlings are not resistant to the
herbicide 2,4-dichlorophenoxybutyric acid, which is catabolized to the auxin 2,4-dichlorophenoxyacetic acid by
b-oxidation. Acyl-CoA feeding experiments show that the MFP2 2-trans enoyl-CoA hydratase only exhibits
activity against long chain (C18:0) substrates, whereas the MFP2 L-3-hydroxyacyl-CoA dehydrogenase is active
on C6:0, C12:0 and C18:0 substrates. A mutation in the abnormal inflorescence meristem gene AIM1, the only
homologue of MFP2, results in an abnormal inflorescence meristem phenotype in mature plants (Richmond
and Bleecker, Plant Cell 11, 1999, 1911) demonstrating that the role of these genes is very different. The mfp2-1
aim1 double mutant aborted during the early stages of embryo development showing that these two proteins
share a common function that is essential for this key stage in the life cycle.

Keywords: b-oxidation, Arabidopsis, multifunctional protein, hydratase, dehydrogenase.

Introduction

During germination and early post-germinative growth in
oilseeds, fatty acids derived from storage triacylglycerol are
activated to acyl-CoA and then converted to sucrose by the
sequential actions of the b-oxidation pathway and glyoxy-
late cycle in the peroxisome, and the gluconeogenic path-
way in the cytosol. This provides metabolic energy and
carbon skeletons for germination and early post-germina-
tive seedling growth (Kindl, 1987). As well as an essential
role in oilseed germination, peroxisomal b-oxidation is also
required as a salvage pathway of fatty acids derived from
primary galactolipids during foliar senescence (reviewed in
Graham and Eastmond, 2002) and is induced to supply res-
piratory substrates in carbohydrate-deprived maize root tips
(Dieuaide et al., 1992; Hooks et al., 1995). At lower levels,
b-oxidation is also a constitutive property of plant tissues,
possibly acting in membrane lipid turnover. Additional
physiological roles for b-oxidation include the synthesis for
fatty acid-derived signals such as jasmonic acid (Cruz
Castillo et al., 2004; Pinfield-Wells et al., 2005; Theodoulou
et al., 2005; Wasternack and Parthier, 1997), traumatin
(Farmer, 1994) and auxin, because b-oxidation converts
indole-3-butyric acid to indole-3-acetic acid (Zolman et al.,
2000). UV light has been shown to increase the expression of
an acyl-CoA oxidase (ACX2) gene of b-oxidation, suggesting
that the pathway may be induced to provide acetyl-CoA
substrate for secondary metabolism (Logemann and
Hahlbrock, 2002; Logemann et al., 2000). A mutation in the
b-oxidation abnormal inflorescence meristem gene AIM1 in

930 ª 2006 The Authors

Journal compilation ª 2006 Blackwell Publishing Ltd
 b-oxidation in Arabidopsis mfp2 mutant seedlings 931
Arabidopsis thaliana suggests a novel role for b-oxidation in
floral development (Richmond and Bleecker, 1999). Finally,
an absolute requirement for short-chain acyl-CoA oxidase
activity during embryo development has been demonstra-
ted (Rylott et al., 2003).
Plant peroxisomal b-oxidation involves enzymes from
three gene families: (i) acyl-CoA oxidases (ACX), (ii) multi-
functional proteins (MFPs) and (iii) L-3-ketoacyl-CoA thiolas-
es. Together these enzymes catalyze the complete
degradation of both saturated and unsaturated long-chain
fatty acyl-CoAs to acetyl-CoA via the repeated cleavage of
acetate units from the thiol end of the fatty acid (reviewed in
Graham and Eastmond, 2002). Genes from each of these
families have been characterized in Arabidopsis. These
include four ACX isogenes, encoding enzymes with long
(ACX2), medium-long (ACX1; Hooks et al., 1999), medium
(ACX3; Eastmond et al., 2000b; Froman et al., 2000) and
short-chain (ACX4; Hayashi et al., 1999) substrate specifici-
ties, a thiolase gene (PED1; peroxisome defective; Germain
et al., 2001; Hayashi et al., 1998) and the MFP gene AIM1
(Richmond and Bleecker, 1999). AIM1 (GenBank accession
number AF123253) and MFP2 (AC016827) are the only MFP
genes in the Arabidopsis genome, and the AIM1 and MFP2
proteins share 58% identity. Both AIM1 and MFP2 have
putative type 1 peroxisomal targeting signals (PTS-1; S,K,L
for AIM1 and S,R,L for MFP2) and MFP2 has been shown to
be targeted to the peroxisome in vivo using a GFP fusion
construct (Cutler et al., 2000).
MFP activity has been studied in cucumber, where four
different isozymes have been reported (Behrends et al.,
1988; Gühnemann-Schäfer and Kindl, 1995a,b). MFP I, II and
III activities are present in germinating seedlings, while MFP
IV is a leaf peroxisomal form. MFP II has been cloned and the
protein contains four domains having 2-trans-enoyl-CoA
hydratase, L-3-hydroxyacyl-CoA dehydrogenase, D-3-hyd-
roxyacyl-CoA epimerase and D3,D2-enoyl-CoA isomerase
activities (Preisig-Müller et al., 1994). The Arabidopsis
MFP2 protein has extensive homology to MFP IV over all
four of these domains (Eastmond and Graham, 2000).
Complete b-oxidation of saturated acyl-CoAs and those with
a double bond at the D2 position in the trans configuration
require only the 2-trans-enoyl-CoA hydratase (EC 4.2.1.17)
and L-3-hydroxyacyl-CoA dehydrogenase (EC 1.1.1.35) activ-
ities. However, for 4-cis-unsaturated fatty acids, D-3-hydrox-
yacyl-CoA epimerase and D3,D2-enoyl-CoA isomerase
activities are also required (Gerhardt, 1992).
Arabidopsis mutants disrupted in each of the three major
steps of the b-oxidation pathway have been described,
including ped1/kat2 (Germain et al., 2001; Hayashi et al.,
1998), acx1and acx2 (Adham et al., 2005; Pinfield-Wells
et al., 2005), acx3 (Adham et al., 2005; Eastmond et al.,
2000b), acx4 (Rylott et al., 2003), acx5 (Adham et al., 2005)
and aim1 (Richmond and Bleecker, 1999). These mutants
exhibit a range of phenotypes during germination and early
ª 2006 The Authors
Journal compilation ª 2006 Blackwell Publishing Ltd, The Plant Journal, (2006), 45, 930–941
post-germinative growth, including the requirement of
exogenous sucrose for seedling establishment (ped1/kat2),
resistance to the herbicide and auxin precursor 2,4-dic-
hlorophenoxybutyric acid (2,4-DB; ped1/kat2, acx3, acx4,
aim1), enlarged peroxisomes (ped1/kat2) and accumulation
of acyl-CoAs (ped1/kat2, acx3, acx4). The AIM1 gene is
expressed predominantly in silique, flower and older
(>8 days) seedlings and at only very low levels during early
post-germinative growth (Richmond and Bleecker, 1999). In
contrast, MFP2 shows a significant induction during germi-
nation and is highly and predominantly expressed during
early post-germinative seedling growth (Eastmond and
Graham, 2000). Therefore, we chose to investigate the role
of MFP2 in seedling and organelle development during
germination and early post-germinative growth and the
combined role of both MFPs throughout development.
Results
Mutant isolation and genotypic characterization
Two insertional mutants in the MFP2 gene were isolated
from T-DNA populations. The mfp2-1 mutant was identified
from the population described by Sussman et al. (2000) as
outlined in Experimental procedures and the mfp2-2 mutant
was obtained from the SALK collection (Alonso et al., 2003).
The mfp2-1 and mfp2-2 mutants were backcrossed to wild
type [Wassilewskija (WS) and Columbia 0 (Col0), respect-
ively] and the F2 population was shown to segregate 3:1 for
kanamycin resistance. In both alleles, PCR-based techniques
showed that the T-DNA insertion in MFP2 co-segregated
with kanamycin resistance. Sequencing of the left border
revealed that the T-DNA in mfp2-1 is situated in an exon of
MFP2, 1880 bp 3¢ from the ATG, between the hydratase and
dehydrogenase domains (Figure 1). The mfp2-2 mutant has
Figure 1. Schematic representation of the MFP2 gene and MFP2 protein
showing the location of mfp2 mutations.
The 18 exons (E) of the MFP2 gene are shown as black boxes. Locations of
mfp2-1 mfp2-2, mfp2-3, mfp2-4 and mfp2-5 mutations are shown as grey
arrows.
932 Elizabeth L. Rylott et al.

a T-DNA in the first intron, 283 bp 3¢ from the ATG, with the (a)
left border pointing towards the 5¢ end of MFP2 (Figure 1).
Three additional mfp2 alleles were subsequently identi-
fied in a forward genetic screen designed to identify genes
involved in storage reserve mobilization during post-germi-
native growth (PJE, unpublished data). The screen was
based on a short hypocotyl phenotype in dark-grown
seedlings that could be rescued by exogenous sucrose,
essentially similar to the phenotype reported previously for
the phosphoenolpyruvate carboxykinase (pck1) mutant
(Penfield et al., 2004; Rylott et al., 2004) and isocitrate lyase
(icl) mutant (Eastmond et al., 2000a). These additional mfp2
alleles are in a Col0 genetic background and were generated
by ethyl methyl sulfonate (EMS) mutagenesis. Sequencing
revealed that mfp2-3, mfp2-4 and mfp2-5 all contain point (b)
mutations in the MFP2 coding sequence (Figure 1). Both
mfp2-2 and mfp2-3 contain mutations in the middle of the
hydratase domain (þ1235 and þ835 bp, respectively). In
mfp2-3, a C to T mutation changes Gln152 (cag) to a
premature stop codon (tag) and, in mfp2-4, an A to G
mutation converts Gly115 (gga) to Arg (aga). An A to G
mutation at þ1994 bp also leads to an amino acid substitu-
tion between the hydratase and dehydrogenese domains in
mfp2-5 where Gly301 (gga) is changed to Glu (gaa).

Phenotypic analysis

To investigate if the disruption of MFP2 impaired germina- WT mfp2-1

tion and growth in the mfp2 mutants, seedling establish-
ment was examined (Eastmond et al., 2000a; Hayashi et al.,
1998). On medium containing 20 mM sucrose, seedling
growth occurred normally in wild type and mfp2-1 in both
the light (not shown) and dark (Figure 2a). Normal seedling
growth was also observed in wild-type seedlings grown
without sucrose (results not shown). In the light, on medium
without sucrose, mfp2 growth arrested 3–4 days after ger-
mination, the cotyledons failed to expand and no true leaves
were produced (results not shown). In the dark, hypocotyl
growth in the mfp2 seedlings was significantly less than wild
types (Figure 2b; Table 1). Only homozygous mfp2 seed-
lings exhibited a sucrose-dependent seedling phenotype
(data not shown), demonstrating that the mutation is
recessive. Both light- and dark-grown mfp2 seedling phe-
notypes could be rescued by transferring the seedlings to
medium containing 20 mM sucrose, where the seedlings
grew and established photosynthetic competency. Follow-
ing seedling establishment, all mfp2 plants grew normally
and were phenotypically indistinguishable from wild-type
plants.
Activity of MFP2
To establish the effect of the mutation at the enzymatic
level, the activities of 2-trans-enoyl-CoA hydratase and
Figure 2. Hypocotyl length in 5-day-old mfp2-1 and wild-type seedlings
grown in the dark. Scale bar ¼ 10 mm.
(a) Seedlings grown on medium containing 20 mM sucrose.
(b) Seedlings grown on medium without sucrose.
L-3-hydroxacyl-CoA dehydrogenase were measured in wild-
type and mfp2-1 seedlings from 0 to 5 days after imbibition
(Figure 3). Whilst hydratase activity, using the C4:0 substrate
crotonyl-CoA, was unaltered from wild-type levels in the
mfp2-1 mutant (Figure 3a), dehydrogenase activity, meas-
ured using acetoacetyl-CoA, was reduced by >98% of wild
type (Figure 3b). Similar results were seen for mfp2-2, mfp2-
3, mfp2-4 and mfp2-5 (Table 1). In wild-type seedlings, the
hydratase and dehydrogenase activities in 0- to 3-day-old
seedlings were approximately equal, with both activities
peaking in 3-day-old seedlings. However, in older seedlings,
the dehydrogenase activity declined whilst the hydratase
activity remained high; after 5 days, the hydratase activity
was seven times higher than dehydrogenase activity
(Figure 3). As MFP2 hydratase and dehydrogenases are both
encoded on the same polypeptide, this lack of fixed stoichi-
ometry suggests the presence of additional hydratases with
activity to short-chain (C4:0) substrate. Richmond and
Bleecker (1999) reported that purified recombinant MFP2 had
a lower affinity for short-chain enoyl-CoA than AIM1 and,

ª 2006 The Authors

Journal compilation ª 2006 Blackwell Publishing Ltd, The Plant Journal, (2006), 45, 930–941
 b-oxidation in Arabidopsis mfp2 mutant seedlings 933

Table 1 Hypocotyl length, 2,4-DB resist-

ance and multifunctional protein (MFP) Hypocotyl Root length on Activity
activities of wild type and mfp2 mutants length (cm) 2,4-DB (cm) (nmol gfwt)1 min)1)

Mean SE 1.5 lM SE 3 lM SE Hydratase SE Dehydrogenase SE

Columbia 0 1.93 0.19 0.26 0.05 0.1 nd 833 61.4 749.9 126
(Col0)
Wassilewskija 1.89 0.17 0.23 0.07 0.1 nd 1145.4 87.8 854.2 172
(WS)
mfp2-1 0.78 0.11 0.19 0.08 0.1 nd 901.1 54.8 7.4 42
mfp2-2 0.73 0.15 0.24 0.06 0.1 nd 761.8 50.6 11.9 8.9
mfp2-3 0.71 0.09 0.21 0.09 0.1 nd 659.4 37.7 16.9 13
mfp2-4 0.73 0.1 0.3 0.11 0.1 nd 590.6 51.6 8.5 4.1
mfp2-5 0.79 0.12 0.22 0.06 0.1 nd 578.3 37.8 19.4 14
kat2 0.12 0.02 2.23 0.15 2.16 0.2 nd nd
acx3 1.78 0.13 1.65 0.09 0.25 0.1 nd nd

Hypocotyl length of seedlings grown in the dark on 1/2 MS and root length of seedlings grown in
the light on 1/2 MS plus 20 mM sucrose and 2,4-DB were measured after 5 days. Multifunctional
protein activities were measured using C4:0 substrates, in extracts of 2-day-old seedlings grown
in the light on medium containing 20 mM sucrose. Values are the mean
SE of measurements
made on three separate extracts from seedlings; nd, not determined.

extracts of 2-day-old wild-type and mfp2 seedlings. As enoyl-

Figure 3. Enzyme activity in wild-type and mfp2-1 seedlings grown on
20 mM sucrose, in the light.
Wild type (closed circles) and mfp2-1 (open circles). Values are the mean
SE
of measurements made on three separate protein extracts.
(a) 2-trans-enoyl-CoA hydratase activity.
(b) L-3-hydroxacyl-CoA dehydrogenase activity.
combined with the results here, this indicates that MFP2 may
have predominantly long-chain activity. To investigate this,
we used mass spectrometry to monitor the production of
b-oxidation intermediates in assays containing crude plant
ª 2006 The Authors
Journal compilation ª 2006 Blackwell Publishing Ltd, The Plant Journal, (2006), 45, 930–941
CoA substrates of longer chain lengths than C4:0 are not
commercially available, we used acyl-CoAs as substrates,
coupling the reaction through endogenous ACX activity. The
activities of ACX in 2-day-old seedlings of mfp2-1 were not
significantly different to wild-type levels (data not shown).
Figure 4 shows the relative abundance of C12:0 acyl-CoA
substrate and 2-trans-enoyl-CoA, 3-hydroxyacyl-CoA and 3-
ketoacyl-CoA b-oxidation intermediates, including the
appearance of C10:0 acyl-CoA. A schematic diagram of the b-
oxidation pathway intermediates is shown in Figure 5. After
40 min of incubation with C12:0 acyl-CoA, accumulation of
C12:0 2-trans-enoyl-CoA was significantly higher in mfp2-1,
mfp2-3 and mfp2-4 extracts (6.8-, 7.4- and 2.1-fold, respect-
ively) than wild types (1.5- and 0.8-fold; Figure 5a). Similar
increases in mfp2 3-hydroxyacyl-CoA intermediates, relative
to wild-type levels were also observed (Figure 5b). The levels
of 3-ketoacyl-CoA intermediate, the substrate for thiolase,
were correspondingly lower than wild type for mfp2 seed-
lings (Figure 5c). These data indicate that C12:0 dehydroge-
nase activity was disrupted in the mfp2 mutants resulting in
the accumulation of substrates upstream of the dehydroge-
nase step and a reduction in the levels of 3-ketoacyl-CoA
products downstream. A similar increase in 2-trans enoyl-
CoA and 3-hydroxyacyl-CoA accumulation was seen in the
mfp2-1 mutant when C6:0 acyl-CoA was supplied as sub-
strate (Figure 5d–f). When long-chain (C18:0) substrate was
supplied, mfp2-1 accumulated 2.3-fold more 2-trans-enoyl-
CoA than wild type and, as with feeding C6:0 and C12:0,
accumulation of the dehydrogenase product, 3-ketoacyl-
CoA, was dramatically reduced in the mutant compared to
wild type. However, unlike with C6:0 and C12:0 feeding, 3-
hydroxyacyl-CoA levels in mfp2-1 did not accumulate with
C18:0 feeding any more than in wild type. We conclude that
934 Elizabeth L. Rylott et al.
Figure 4. Relative abundance of b-oxidation intermediates in an ACX assay
mix containing extracts of light-grown 2-day-old seedlings after 40 mins of
incubation with C12:0 acyl-CoA substrate.
(a) Wild type.
(b) mfp2-1.
this is due to compromised long-chain hydratase activity in
mfp2-1, in addition to compromised broad range dehydrog-
enase activity.
Storage lipid breakdown, acyl-CoA levels and peroxisome
morphology
To determine whether the sucrose-dependent seedling
phenotype in the mfp2 mutants correlates with a reduced
ability to breakdown storage lipid, we measured fatty acid
levels in germinating mfp2-1 seedlings. Total fatty acid lev-
els were reduced by 35% in mfp2-1 seeds and the rate of lipid
catabolism during early post-germinative growth was also
less than wild type, so that by day 5, three times more fatty
acid remained in mfp2-1 seedlings. To investigate the rate of
storage TAG breakdown in mfp2-1 seedlings, we measured
the levels of eicosenoic acid (C20:1), which is specific to
storage TAG, in Arabidopsis (Lemieux et al., 1990). In mfp2-1
seeds, C20:1 levels were 41% less than in wild-type seeds
and the rate of catabolism was also reduced such that 5-day-
old mfp2-1 seedlings contained six times more C20:1 than
wild-type seedlings (Figure 6a).
We have developed a methodology for the measurement
of acyl-CoAs (Larson and Graham, 2001; Larson et al., 2002)
and shown that the b-oxidation mutants kat2 and acx4
accumulate acyl-CoAs during early post-germinative growth
Journal compilation ª 2006 Blackwell Publishing Ltd, The Plant Journal, (2006), 45, 930–941
Figure 5. Schematic diagram of b-oxidation pathway and abundance of b-
oxidation CoA ions during b-oxidation assays on extracts from 2-day-old light-
grown seedlings using acyl-CoA as substrate, as determined by mass
spectrometry.
Black bars ¼ mean abundance between 0 and 5 min incubation; grey
bars ¼ mean abundance between 35 and 40 min. Values are the mean
SE
of measurements made on three separate seedling extracts.
(a, d) Abundance of CoA ions derived from 2-trans-enoyl-CoA.
(b, e) Abundance of CoA ions derived from 3-hydroxylacyl-CoA.
(c, f) Abundance of CoA ions derived from 3-ketoacyl-CoA.
(Germain et al., 2001; Rylott et al., 2003). Using this meth-
odology, we observed a 2.5-fold increase in the quantity of
long-chain (C20:0–C20:3) acyl-CoAs in 5-day-old seedlings of
mfp2-1 relative to wild type (Figure 6b). Seedlings of ped1/
kat2, acx3, acx4 and aim1 all show varying levels of
resistance to 2,4-DB (Germain et al., 2001; Hayashi et al.,
1998; Eastmond et al., 2000b; Richmond and Bleecker, 1999;
Rylott et al., 2003). Richmond and Bleecker (1999) deter-
mined 2,4-DB resistance in aim1 seedlings using segrega-
ting F2 populations to overcome the effects of both
abnormal maternal tissues on embryo development and
the severely reduced fertility of aim1 homozygotes. Inter-
estingly, although mfp2-1 seedlings accumulate long-chain
acyl-CoAs, demonstrating a reduction in b-oxidation flux,
none of the mfp2 mutants was resistant to 2,4-DB (Table 1).
Seedlings of ped1/kat2 exhibit enlarged and abnormal
peroxisomes in 5-day-old cotyledons, some of which con-
tain unusual membrane inclusions (Germain et al., 2001).
We examined the ultrastructure in micrograph sections
through 4-day-old cotyledons of wild-type and mfp2-1
seedlings (Figures 7a,b). The number of peroxisomes per
ª 2006 The Authors
 b-oxidation in Arabidopsis mfp2 mutant seedlings 935
Figure 6. Lipid and acyl-CoA levels in wild-type and mfp2-1 seedlings grown,
in the light, on medium containing 20 mM sucrose.
(a) Eicosenoic acid (C20:1) levels in wild-type and mfp2-1 seedlings, 0–5 days
after imbibition (DAI).
(b) Acyl-CoA levels in wild-type and mfp2-1 seedlings. (0–5 DAI); black bars,
wild type; white bars, mfp2-1. The values shown are the mean
SE of
measurements on five batches of seedlings.
micrograph was 2.6-fold lower than wild-type levels; how-
ever, there was a 4.6-fold increase in individual peroxisomal
surface areas, demonstrating that mfp2-1 has fewer, larger
peroxisomes than wild type (Figure 7c).
Analysis of crosses between mfp2-1 and aim1
To investigate the relative roles of MFP2 and AIM1 in Ara-
bidopsis, we produced reciprocal crosses between mfp2-1
and aim1. Despite extensive testing using PCR-based
techniques, no mfp2-1-aim1 double mutant plants were
identified. The ratio of heterozygous (AIM1aim1) to wild-type
(AIM1AIM1) plants in progeny from selfed mfp2-1mfp2-1
ª 2006 The Authors
Journal compilation ª 2006 Blackwell Publishing Ltd, The Plant Journal, (2006), 45, 930–941
(a)
(b)
(c)
Figure 7. Electron microscopy sections from 5-day-old cotyledons grown in
the light on 1/2 MS containing 20 mM sucrose.
P, peroxisome; CP, chloroplast; LB, lipid body. Bar, 2 lm.
(a) Wild type.
(b) mfp2-1.
(c) Measurements of mean number of peroxisome sections per micrograph
and surface area of peroxisome sections per cell section (lm2); black bars,
wild type; grey bars, mfp2-1. Values were derived from micrographs from
three seedlings for each line.
AIM1aim1 plants recorded in Table 2(a), is close to the 2:1
ratio expected if homozygous aim1 embryos were non-viable
in the mfp2-1 homozygous background. Examination of
seeds in siliques from mfp2-1mfp2-1-AIM1aim1 plants at the
mid- to late-cotyledonary stage revealed a number of shriv-
elled seeds, whilst seeds in the siliques at the earlier heart and
936 Elizabeth L. Rylott et al.

Table 2 Phenotyps of seedlings and embryos segregating for AIM1/aim1 genotypes in mfp2 background
(a) AIM1genotypes, as determined by PCR, of seedling from mfp2-1 homozygous, AIM1 heterozygous parent plants. The seedlings were grown
in the dark on medium containing 1/2 MS without sucrose and hypocotyl length was measured after 5 days.
(b) Number of aborted and normal embryos found in siliques examined at the mid- to late-cotyledonary stage of embryo development in
siliques from mfp2-1 homozygous, AIM1 heterozygous plants. Numbers in parentheses are percentage per silique.
(a)

AIM1 genotype of progeny from mfp2mfp2-AIM1aim1 parent

Hypocotyl length Wild type (AIM1) Heterozygous Homozygous (aim1)

Short (>5 mm) 45 11 0

Very short (>5 mm) 31 113 0

(b)

Genotype Wild-type embryo Aborted embryo Total no. of embryos

mfp2 565 (88.3) 75 (11.7) 640

mfp2mfp2-AIM1aim1 466 (72.7) 175 (27.3) 641
Wild-type Wassilewskija (WS) 372 (92.8) 29 (7.2) 401

globular stages of embryo development appeared normal
(Table 2b) indicating that the embryos had aborted during
late morphogenesis/early maturation (Jurgens and Mayer,
1994). These results suggest that the combined removal of
both MFP2 and AIM1 activities is lethal to the embryo at an
early stage of development. The level of MFP activity was
assayed in developing wild-type Arabidopsis embryos. Both
hydratase and dehydrogenase activity was detectable from
the heart stage of embryo development onwards, peaking in
dry seed (data not shown). Approximately two-thirds, (72%)
of the progeny from selfed plants that were homozygous for
mfp2-1, but heterozygous for aim1, had a hypocotyl length
<5 mm, whilst the remainder had hypocotyl lengths between
5 and 8 mm. Genotyping by PCR revealed that 80% of the
seedlings with hypocotyls >5 mm were wild type for AIM1
(expected ratio 33% homozygous AIM1AIM1, 67% hetero-
zygous AIM1aim1 v2 ¼ 53.9, P < 0.1) and 78% of the seed-
lings with hypocotyls <5 mm were AIM1aim1 heterozygotes
(expected ratio 67% heterozygous AIM1aim1, 33% homo-
zygous AIM1AIM1, v2 ¼ 9.0, P < 1.0; Table 2a). PCR analysis
confirmed that all the seedlings were homozygous for mfp2-
1. This indicates that mfp2-1mfp2-1-AIM1aim1 seedlings are
more compromised in hypocotyl length than mfp2-1 single
mutants. Seed set from plants that were homozygous for
aim1 but heterozygous for mfp2 was low and of poor quality
due to severely reduced fertility and a similar abnormal
floral development phenotype to that seen in aim1 plants
(Richmond and Bleecker, 1999).
Discussion
Phenotypic characterization of mfp2
We have identified and characterized five Arabidopsis
mutants disrupted in the gene encoding the MFP (MFP2) of
peroxisomal b-oxidation. The mfp2 mutants show a sucrose-
dependent seedling growth phenotype, lack hydroxyacyl-
CoA dehydrogenase activity and have reduced levels of long-
chain enoyl-CoA hydratase activity. Further characterization
revealed a reduction in storage triacylglycerol breakdown,
accumulation of long-chain acyl-CoAs during early post-
germinative growth and enlarged peroxisomes. We have
previously shown that the acx3 and acx4 mutants of b-oxi-
dation are individually unaffected in storage lipid catabolism
due to the overlapping substrate specificity of the corres-
ponding gene products (Eastmond et al., 2000b; Rylott et al.,
2003). This overlapping substrate specificity also accounts for
the ability of acx3 and acx4 seedlings to germinate and
establish in the absence of an exogenous supply of sucrose
(Eastmond et al., 2000b; Rylott et al., 2003). Under the growth
conditions used, ACX4 and ACX3 complement the acx3 and
acx4 mutations, respectively. Despite this, both acx3 and
acx4 seedlings accumulate medium- and short-chain acyl-
CoAs, respectively, and exhibit resistance to 2,4-DB (1.5 lM)
demonstrating that both mutants have reduced rates of b-
oxidation. The sucrose-dependent phenotype, accumulation
of long-chain acyl-CoAs and increased resistance to 2,4-DB
(>5 lM) in ped1/kat2 seedlings demonstrates a more severe
restriction in the rate of lipid catabolism than in the acx3 and
acx4 seedlings. Here, we have shown that the severity of the
sucrose-dependent mfp2 seedling phenotype falls between
that of the ped1/kat2 phenotype and the acx3 or acx4 phe-
notypes. This is likely to be because additional hydratases
and dehydrogenases are contributing to the MFP activity
during seedling establishment.
There are several candidate genes that might supply the
hydratase activity seen in the mfp2 mutants. The hydratase
region of AIM1 shares 80% identity with that of MFP2. In
addition to AIM1, there are several other genes in the
Arabidopsis genome encoding putative enoyl-CoA hydra-

ª 2006 The Authors

Journal compilation ª 2006 Blackwell Publishing Ltd, The Plant Journal, (2006), 45, 930–941
 b-oxidation in Arabidopsis mfp2 mutant seedlings 937
tases, two with PTS-1s (Graham and Eastmond, 2002).
Richmond and Bleecker (1999) reported weak AIM1
expression in etiolated seedlings, with stronger expression
only detected in older (>8-day-old) seedlings. This suggests
that AIM1 does not contribute significantly to b-oxidation
during early post-germinative seedling growth. However,
our observation that mfp2-1mfp2-1-AIM1aim1 seedlings are
more compromised in hypocotyl elongation than mfp2-
1mfp2-1-AIM1AIM1 seedlings demonstrates that AIM1 does
supply significant MFP activity during seedling establish-
ment, at least in the mfp2-1 mutant background. We were
unable to obtain sufficient seeds to assay aim1 seedlings
due to both severely reduced fertility and abnormal floral
development, as reported by Richmond and Bleecker (1999).
Beyond seedling establishment, the mfp2 mutants show
no visible phenotype throughout the rest of the life cycle. This
suggests that, after seedling establishment, MFP2 activity is
sufficiently compensated for by AIM1 and or additional
hydratases. The severe phenotype of aim1 mutants in mature
plants and floral organs demonstrates the dominant role of
AIM1 during these phases (Richmond and Bleecker, 1999).
Chain length specificity of MFP2
Chain length specificity has been reported in cucumber MFP
hydratases; MFP IV exhibits higher activity towards shorter-
chain substrates, whilst MFP II shows no activity towards
C18:1 substrate (Gühnemann-Schäfer and Kindl, 1995b). We
used a novel mass spectrometery-based approach to meas-
ure the relative abundance of b-oxidation intermediates in
wild-type and mfp2 seedlings. This showed that the mfp2
mutant lacks both long-chain (C18:0) hydratase and short-,
medium- and long-chain (C6:0, C12:0 and C18:0)
dehydrogenase activities. This is in agreement with our
in vitro enzyme assays that showed that mfp2-1 extracts lack
C4:0 dehydrogenase activity but have wild-type levels of C4:0
hydratase activity. We therefore conclude that MFP2 exhibits
short-, medium- and long-chain dehydrogenase activity but
only long-chain hydratase activity. Using recombinant AIM1
and MFP2, Richmond and Bleecker (1999) reported that AIM1
has a higher affinity for short-chain (C4:0), crotonyl-CoA than
MFP2. The aim1 seedlings exhibit resistance to 2,4-DB
(Richmond and Bleecker, 1999), while mfp2 seedlings are
sensitive. AIM1 is therefore presumably active on 2,4-DB in
the mfp2 background. Our data from in vitro enzyme assays
on seedling extracts, sensitivity to 2,4-DB and long-chain
acyl-CoA accumulation seen in mfp2-1 seedlings fit with our
proposal that MFP2 hydratase has long-chain activity whilst
MFP2 dehydrogenase has broad range specificity.
Peroxisomal size and abundance
Peroxisome size and abundance in mammals and yeasts are
regulated by metabolic activity (Chang et al., 1999; van
ª 2006 The Authors
Journal compilation ª 2006 Blackwell Publishing Ltd, The Plant Journal, (2006), 45, 930–941
Roermund et al., 2000; Smith et al., 2000) and by the activity
of a peroxisomal membrane protein PEX11 (Erdman and
Blobel, 1995; Marshall et al., 1996; Schrader et al., 1998).
These two mechanisms appear to function independently (Li
and Gould, 2002). The metabolic control of peroxisome size
and abundance is also seen in plants as the ped1/kat2 mu-
tant (Germain et al., 2001; Hayashi et al., 1998) and the mfp2-
1 mutant described here both have a reduced number of
enlarged peroxisomes, which points to a conserved mech-
anism regulating peroxisome size and abundance.
The peroxisome ABC transporter 1 (pxa1)/peroxisomal
defective 3 (ped3)/comatose 1 (cts-1) mutants, which are
proposed to transport fatty acids or acyl-CoAs, exhibit a
similar phenotype to mfp2-1 in that seedlings that are
defective in the catabolism of storage lipid require an
exogenous supply of sucrose for seedling establishment
(Footitt et al., 2002; Hayashi et al., 2002; Zolman et al., 2001)
and accumulate acyl-CoAs (Footitt et al., 2002). However,
unlike mfp2-1, which shows an increase in peroxisome size,
cts peroxisomes appear normal (Footitt et al., 2002; Hayashi
et al., 2002). Based on the assumption that b-oxidation
mutants accumulate acyl-CoAs inside the peroxisome whilst
cts accumulates acyl-CoAs in the cytosol, Graham et al.
(2002) proposed that the increase in peroxisomal size seen in
the ped1 and kat2 mutants might be due specifically to the
peroxisomal concentration of acyl-CoAs. Here, our results
on the characterization of mfp2-1 are in agreement with this
proposal.
In addition to acyl-CoA accumulation, acyl-CoA chain
length also appears to play an important role in regulating
peroxisome size. Whilst seedlings of acx3 and acx4, like
mfp2-1 seedlings, accumulate acyl-CoAs (Eastmond et al.,
2000b; Rylott et al., 2003), they exhibit wild-type peroxisome
morphology. However, unlike mfp2-1, these mutants accu-
mulate short- and medium-chain-length acyl-CoAs, whereas
mfp2-1 accumulates specifically long-chain acyl-CoAs. The
acx1-acx2 mutant, which accumulates long-chain acyl-CoAs,
also exhibits enlarged peroxisomes (Pinfield-Wells et al.,
2005). It has been suggested that it is not the increase in acyl-
CoA concentration per se that results in increased peroxi-
somal size, but the accumulation of long-chain acyl-CoAs as
seen in ped1/kat2 that are likely to be specifically involved in
regulating this process (Germain et al., 2001; Graham et al.,
2002). Here, our results demonstrating the accumulation of
long-chain acyl-CoAs and enlarged peroxisomes in mfp2-1
seedlings reinforce this theory. There is evidence that
increased matrix protein content can increase peroxisome
size in yeasts (Smith et al., 2000), and fatty acids positively
regulate transcription of b-oxidation genes and PEX11 via
the Oaf1p/Pip2p (Saccharomycres cerevisiae; Rottensteiner
et al., 1997) and peroxisome proliferator activated receptor a
(mammals; Dreyer et al., 1992) transcription factors. The
Arabidopsis genome lacks obvious homologues of either of
these transcription factors suggesting that plants may have
938 Elizabeth L. Rylott et al.
evolved an as yet unidentified mechanism for acyl-CoA
mediated regulation of gene expression.
Crossing mfp2 with aim1
The ratios of heterozygous to wild-type plants for AIM1
recorded in Table 2(a) and (b), is close to the 2:1 ratio
expected if homozygous aim1 embryos were lethal in the
mfp2-1-homozygous background. Furthermore, the per-
centage of aborted embryos in siliques of plants with mfp2-
1mfp2-1 AIM1aim1 genotypes is close to the 25% expected if
double mfp2-1-aim1 mutant embryos were non-viable. We
have recently shown that acx3-acx4 double mutants also
abort at an early stage of embryo development (Rylott et al.,
2003) whilst acx1-acx2 mutants are viable (Adham et al.,
2005; Pinfield-Wells et al., 2005) demonstrating that short-
chain b-oxidation activity is essential for embryo develop-
ment. The removal of ACX3 and ACX4 activity through gen-
eration of a double mutant cannot be compensated for by one
of the other ACX proteins because these do not exhibit
activity with short-chain acyl-CoAs. (However, the thiolase
gene family encodes enzymes with broad substrate specif-
icity and differing expression patterns, and the ped1/kat2
phenotype is limited to the germination and early post-ger-
minative growth stages.) We conclude that the non-
viable embryos seen in the siliques of plants with mfp2-
1mfp2-1-AIM1aim1 phenotypes are aim1-mfp2-1 double
mutants and that a complete block in b-oxidation capacity is
lethal during early embryo development. We have detected
both hydratase and dehydrogenase activity in developing
embryos with activity peaking in dry seed. Significant levels
of hydratase activity are still present in imbibed seeds
(Figure 3a). A similar peak in MFP activity has also been
reported in Brassica napus embryos and a discrepancy
between hydratase and dehydrogenase activities was also
observed (Chia et al., 2005). This discrepancy is further evi-
dence that more than one enzyme is contributing to the MFP2
hydratase and dehydrogenase activities. The early stage
embryo lethality observed in the double mutant could be due
to one of a number of factors including the accumulation, to
toxic levels of acyl-CoAs or fatty acids, sequestration of CoA
to the acyl-CoA pool or the disruption in the production of
fatty acid or lipid-based signalling molecules that are critical
for embryogenesis. Oilseed rape plants engineered to pro-
duce increased levels of medium-chain fatty acids accumu-
lated high (60 mol percent) levels of C12:0 acyl-CoA in mature
seeds, which appeared normal and were viable (Eccleston
and Ohlrogge, 1998; Larson et al., 2002). Furthermore, ped1/
kat2 seedlings, which are effectively blocked in early post-
germinative b-oxidation activity, accumulate significantly
more (threefold) long-chain acyl-CoAs and have reduced
storage lipid levels, without any detrimental effects on em-
bryo development. These observations suggest that an
accumulation of acyl-CoAs is not toxic to the plant cells.
Journal compilation ª 2006 Blackwell Publishing Ltd, The Plant Journal, (2006), 45, 930–941
Peroxisomal b-oxidation may be required for the production
of signalling molecules, for example in the synthesis of jas-
monic acid, a wound-related signalling molecule (Miersch
and Wasternack, 2000). Although there are no reported roles
for jasmonic acid in embryo development, a ped1 ped3
double mutant has wavy leaves and dwarf, sterile inflores-
cences with abnormal structure (Hayashi et al., 2002). This is
similar to the phenotype of aim1, which exhibits abnormal
floral development, severely reduced fertility, and produces
seed of variable size and shape (Richmond and Bleecker,
1999). Our results demonstrating non-viability of aim1mfp2-1
embryos, together with results from the aim1, and ped1 ped3
mutants, strengthen the argument of a role for b-oxidation in
the generation of signals associated with development
throughout the plant life cycle. In addition, our results con-
solidate previous studies on the b-oxidation mutant ped1/
kat2, and acx3 and acx4 together with cts to reinforce the role
of peroxisomal long-chain acyl-CoAs in the regulation of
peroxisomal development.
Experimental procedures
Plant material and growth conditions
Seeds were surface sterilized and germinated in continuous light on
0.8% (w/v) agar plates containing 1/2 Murashige and Skoog (1962)
(MS) medium (plus 20 mM sucrose where indicated) at 20
C
following 4 days at 4
C in the dark. For experiments with etiolated
seedlings, plates were exposed to 1 h white light and then trans-
ferred to the dark.
Mutant isolation
The mfp2-1 mutant was isolated from the T-DNA mutagenized
A. thaliana (ecotype WS) population at the University of Wisconsin-
Madison Biotechnology Center (Sussman et al., 2000) using prim-
ers to the MFP2 gene 5¢-ATCCTCCCGTCAATTCTCTATCCTTCGAC-3¢
and 5¢-GGGAGCGCTCTGTAATACAAATGGAAGAA-3¢. Sequencing
of the region flanking the left border of the T-DNA insert using the
left border primer JL202 5¢-CATTTTATAATAACGCTGCGGACATC-
TAC-3¢ revealed the terminal 31 nucleotides of exon nine, 21 nu-
cleotides of unidentified sequence and 113 nucleotides of adjoining
left border T-DNA sequence. The mfp2-2 mutant was obtained using
insertion mutant information from the SALK collection (http://
signal.salk.edu/cgi-bin/tdnaexpress; SALK_098016; Alonso et al.,
2003). Additional alleles of mfp2 were identified by performing
complementation crosses on a collection of mutants that are
impaired in post-germinative growth (PJE, unpublished data). This
collection was isolated by screening an EMS mutagenized M2
population of Col0 from Lehle Seeds (Round Rock, TX, USA) for
seedlings with a sugar-dependent hypocotyl elongation phenotype
that phenocopied pck1 (Penfield et al., 2004; Rylott et al., 2004) and
icl mutants (Eastmond et al., 2000a).
Tissue extraction and subcellular fractionation
Tissue extracts were prepared from approximately 50-mg sam-
ples of seedlings as described by Hooks et al. (1999). For crude
ª 2006 The Authors
 b-oxidation in Arabidopsis mfp2 mutant seedlings 939
subcellular fractionation experiments, approximately 200 mg of
2-day-old seedlings were homogenized in 1.5 ml of a medium
containing 150 Tricine/KOH pH 7.5, 1 mM EDTA and 0.5 M sucrose
using a bench top homogeniser (PowerGene 700; Fisher Scientific,
Loughborough, UK). The homogenate was centrifuged at 300 g for
10 min, the supernatant was carefully removed to a fresh tube and
was then centrifuged at 30 000 g for 30 min, and the resulting
supernatant and pellet were assayed.
Enzyme assays, and fatty acid and acyl-CoA measurement
The 2-trans-enoyl-CoA hydratase and L-3-hydroxacyl-CoA dehy-
drogenase activities were assayed using crotonyl-CoA and aceto-
acetyl-CoA, respectively, as substrates, according to the method of
Gerhardt (1983). Isocitrate lyase was assayed according to the
method of Carpenter and Beevers (1959). The b-oxidation interme-
diates were measured in an assay mix adapted from Hryb and Hogg
(1979) and Gerhardt (1983), in 1-ml volumes containing 175 mM
Tris-HCl, pH 8.5, 50 lM flavin-adenine dinucleotide, 5.3 U peroxi-
dase, 5 lg bovine serum albumin, 1 mM sodium azide, 750 lM NAD,
7.5 mg homogenized 2-day-old seedlings (14 mg for C18:0 assays).
Assays were initiated by adding 50 lM acyl-CoA and immediately
infused at room temperature (20
C) over 40 min into an LCQ mass
spectrometer (ThermoFinnigan, Manchester, UK), fitted with an
electrospray ionization source operating in positive ionization mode
(source 5 kV, capillary temperature 300
C, capillary voltage 30 V;
capillary offset 20 V, sheath gas 60 units, aux gas 26 units). The
assay mix was added at 3 ll min)1 into a stream of acetonitrile:
water (1:1) containing 0.2% formic acid at a flow rate
of 0.5 ml min)1. M þ H ions were monitored over the m/z range of
700–1000 for the expected parent masses for each acyl-CoA
(mass ¼ n), enoyl-CoA (n)2), hydroxyacyl-CoA (n þ 16), ketoacyl-
CoA (n þ 14) and the next acyl-CoA (n)28) in the b-oxidation series.
Each of these masses were cyclically selected for fragmentation
(collisional energy 30%, isolation width 3 m/z) to yield the m/z 428
product ion, indicative of CoA. The intensity of this product ion was
monitored to calculate substrate and product concentrations over
the time course of the assay. Ion intensities were averaged over 0–
5 min and between 35 and 40 min.
Fatty acids were measured using the method of Browse et al.
(1986) and in vivo acyl-CoA profiles from plant extracts measured
using the method of Larson and Graham (2001) with modifications
(Larson et al., 2002).
Electron microscopy
Four-day-old cotyledons of wild-type and mfp2-1 seedlings were
grown at 22
C with 18 h light photoperiod, on 1/2 MS medium
containing 20 mM sucrose, then fixed and embedded as described
in Germain et al. (2001). Electron micrographs were taken of eight
cell sections from each of three seedlings for both wild type and
mfp2-1 and cell and organelle sectional areas determined using
IMAGEJ (version 1.27 z) software.
Acknowledgements
We gratefully acknowledge Todd Richmond for providing aim1
seed. ELR and SPS were funded by the European Union (QLRT1999-
00213). PJE was funded by the Biotechnology and Biological Sci-
ences Research Council via a David Phillips Research Fellowship
(87/JF/16985).
ª 2006 The Authors
Journal compilation ª 2006 Blackwell Publishing Ltd, The Plant Journal, (2006), 45, 930–941
References
Adham, A.R., Zolman, B.K., Millius, A. and Bartel, B. (2005) Mutations
in Arabidopsis acyl-CoA oxidase genes reveal distinct and over-
lapping roles in beta-oxidation. Plant J. 41, 859–874.
Alonso, J.M., Stepanova, A.N., Leisse, T.L. et al. (2003) Genome-
Wide Insertional Mutagenesis of Arabidopsis thaliana. Science,
301, 653–657.
Behrends, W., Engeland, K. and Kindl, H. (1988) Characterization of
two forms of the multifunctional protein acting in fatty acid beta-
oxidation. Arch Biochem. Biophys. 263, 161–169.
Browse, J., Mc Court, P.J. and 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, 141–145.
Carpenter, W.D. and Beevers, H. (1959) Distribution and properties
of isocitrate in plants. Plant Physiol. 34, 403–409.
Chang, C.-C., South, S., Warren, D., Jones, J., Moser, A.B., Moser,
H.W. and Gould, S.J. (1999) Metabolic control of peroxisome
abundance. J. Cell Sci. 112, 1579–1590.
Chia, T.Y., Pike, M.J. and Rawsthorne, S. (2005) Storage oil break-
down during embryo development of Brassica napus (L.). J. Exp.
Bot. 56, 1285–1296.
Cruz Castillo, M., Martinez, C., Buchala, A., Metraux, J.P. and
Leon, J. (2004) Gene-specific involvement of b-oxidation in
wound-activated responses in Arabidopsis. Plant Physiol. 135,
85–94.
Cutler, S.R., Ehrhardt, D.W., Griffitts, J.S. and Somerville, C.R.
(2000) Random GFP::cDNA fusions enable visualization of sub-
cellular structures in cells of Arabidopsis at a high frequency.
Proc. Natl Acad. Sci. USA 97, 3718–3723.
Dieuaide, M., Brouquisse, R., Pradet, A. and Raymond, P. (1992)
Increased fatty acid b-oxidation after glucose starvation in maize
root tips. Plant Physiol. 99, 595–600.
Dreyer, C., Krey, G., Keller, H., Givel, F., Helftenbein, G. and Wahli,
W. (1992) Control of the peroxisomal b-oxidation by a novel
family of nuclear hormone receptors. Cell, 68, 879–887.
Eastmond, P.J. and Graham, I.A. (2000) The multifunctional protein
AtMFP2 is coordinately expressed with other genes of fatty acid b-
oxidation during seed germination in Arabidopsis thaliana. Bio-
chem. Soc. Trans. 28, 95–99.
Eastmond, P.J., Germain, V., Lange, P.R., Bryce, J.H., Smith, S.M.
and Graham, I.A. (2000a) Postgerminative growth and lipid cata-
bolism in oilseeds lacking the glyoxylate cycle. Proc. Natl Acad.
Sci. USA 97, 5669–5674.
Eastmond, P.J., Hooks, M.A., Williams, D., Lange, P., Bechtold, N.,
Sarrobert, C., Nussaume, L. and 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, 34375–34381.
Eccleston, V.S. and 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,
613–622.
Erdman, R. and Blobel, G. (1995) Giant peroxisomes in oleic acid-
induced Saccharomyces cerevisiae lacking the peroxisomal
membrane protein Pmp27p. J. Cell Biol. 128, 509–523.
Farmer, E.E. (1994) Fatty acid signalling in plants and their associ-
ated microorganisms. Plant Mol. Biol. 26, 1423–1437.
Footitt, S., Slocombe, S.P., Larner, V., Kurup, S., Wu, Y., Larson, T.,
Graham, I., Baker, A. and Holdsworth, M. (2002) Control of ger-
mination and lipid mobilization by COMATOSE, the Arabidopsis
homologue of human ALDP. EMBO J. 21, 2912–2922.
940 Elizabeth L. Rylott et al.
Froman, B.E., Edwards, P.C., Bursch, A.G. and Dehesh, K. (2000)
ACX3, a novel medium-chain acyl-coenzyme A oxidase from
Arabidopsis. Plant Physiol. 123, 733–742.
Gerhardt, B. (1983) Localisation of b-oxidation enzymes in peroxi-
somes isolated from non-fatty plant tissues. Planta, 159, 238–246.
Gerhardt, B. (1992) Fatty acid degradation in plants. Prog. Lipid Res.
31, 417–446.
Germain, V., Rylott, E.L., Larson, T.R., Sherson, S.M., Bechtold, N.,
Carde, J.P., Bryce, J.H., Graham, I.A. and Smith, S.M. (2001)
Requirement for 3-ketoacyl-CoA thiolase-2 in peroxisome devel-
opment, fatty acid b-oxidation and breakdown of triacylglycerol in
lipid bodies of Arabidopsis seedlings. Plant J. 28, 1–12.
Graham, I.A. and Eastmond, P.J. (2002) Pathways of straight and
branched chain fatty acid catabolism in higher plants. Prog. Lipid
Res. 41, 156–181.
Graham, I.A., Li, Y. and Larson, T.R. (2002) Acyl-CoA measurements
in plants suggest a role in regulating various cellular processes.
Biochem. Soc. Trans. 30, 1095–1099.
Gühnemann-Schäfer, K. and Kindl, H. (1995a) Fatty acid b-oxidation
in glyoxysomes. Characterization of a new tetrafunctional protein
(MFP III). Biochim. Biophys. Acta. 1256, 181–186.
Gühnemann-Schäfer, K. and Kindl, H. (1995b) The leaf peroxisomal
form (MFP IV) of multifunctional protein functioning in fatty acid
b-oxidation. Planta, 196, 642–646.
Hayashi, H., Toriyama, K., Kondo, M. and Nishimura, M. (1998) 2,4-
Dichlorophenoxybutyric acid-resistant mutants of Arabidopsis
have defects in glyoxysomal fatty acid b-oxidation. Plant Cell, 10,
183–195.
Hayashi, H., De Bellis, L., Ciurli, A., Kondo, M., Hayashi, M. and
Nishimura, M. (1999) A novel acyl-CoA oxidase that can oxidise
short-chain acyl-CoA in plant peroxisomes. J. Biol. Chem. 274,
12715–12721.
Hayashi, M., Nito, K., Takei-Hoshi, R., Yagi, M., Kondo, M., Suenaga,
A., Yamaya, T. and Nishimura, M. (2002) Ped3p is a peroxisomal
ATP-binding cassette transporter that might supply substrates for
fatty acid b-oxidation. Plant Cell Physiol. 43, 1–11.
Hooks, M.A., Bode, K. and Couee, I. (1995) Regulation of acyl-CoA
oxidases in maize seedlings. Phytochemistry, 40, 657–660.
Hooks, M.A., Kellas, F. and Graham, I.A. (1999) Long-chain acyl-CoA
oxidases of Arabidopsis. Plant J. 19, 1–13.
Hryb, D.J. and Hogg, J.F. (1979) Chain length specificities of per-
oxisomal and mitochondrial b-oxidation in rat liver. Biochem.
Biophys. Res. Commun. 87, 1200–1206.
Jurgens, G. and Mayer, U. (1994) Arabidopsis. In EMBRYOS,
Colour Atlas of Development (Bard, J.B.L., ed). London; Wolfe
Publishing pp. 7–21.
Kindl, H. (1987) b-oxidation of fatty acids by specific organelles. In
The Biochemistry of Plants, vol 9 (Stumpf, P.K., ed.). New York:
Academic press, pp. 31–50.
Larson, T.R. and Graham, I.A. (2001) Technical advance: a novel
technique for the sensitive quantification of acyl-CoA esters from
plant tissues. Plant J. 25, 115–125.
Larson, T.R., Edgell, T., Byrne, J., Dehesh, K. and Graham, I.A. (2002)
Acyl-CoA profiles of transgenic plants that accumulate medium-
chain fatty acids indicate inefficient storage lipid synthesis in
developing oilseeds. Plant J. 32, 519–527.
Lemieux, B., Miquel, M., Somerville, C. and Browse, J. (1990) Mu-
tants of Arabidopsis with alterations in seed lipid fatty acid
composition. Theor. Appl. Genet. 80, 234–240.
Li, X. and Gould, S.J. (2002) PEX11 promotes peroxisome division
independently of peroxisome metabolism. J. Cell Biol. 156, 643–
651.
Logemann, E. and Hahlbrock, K. (2002) Crosstalk among stress
responses in plants: pathogen defence overrides UV protection
Journal compilation ª 2006 Blackwell Publishing Ltd, The Plant Journal, (2006), 45, 930–941
through an inversely regulated ACE/ACE type of light-respon-
sive gene promoter unit. Proc. Natl Acad. Sci. USA 99, 2428–
2432.
Logemann, E., Tavernaro, A., Schulz, W., Somssich, I.E. and
Hahlbrock, K. (2000) UV light selectively co induces supply
pathways from primary metabolism and flavonoid secondary
product formation in parsley. Proc. Natl Acad. Sci. USA, 97, 1903–
1907.
Marshall, P.A., Krimkevich, Y., Lark, R., Dyer, J.M., Veenhuis, M. and
Goodman, J.M. (1996) Pmp27 promotes peroxisome prolifer-
ation. J. Cell Biol. 129, 345–355.
Miersch, O. and Wasternack, C. (2000) Octadecanoid and jasmonate
signalling in tomato (Lycopersicon esculentum Mill.) leaves:
endogenous jasmonates do not induce jasmonate biosynthesis.
Biol. Chem. 381, 715–722.
Murashige, T. and Skoog, F. (1962) A revised medium for rapid
growth and bioassay with tobacco tissue cultures. Physiol. Plant,
15, 473–496.
Penfield, S., Rylott, E.L., Gilday, A.D., Graham, S., Larson, T.R. and
Graham, I.A. (2004) Reserve mobilization in the Arabidopsis
endosperm fuels hypocotyl elongation in the dark, is independent
of abscisic acid, and requires PHOSPHOENOLPYRUVATE CARB-
OXYKINASE1. Plant Cell, 16, 2705–2718.
Pinfield-Wells, H., Rylott, E.L., Gilday, A.D., Graham, S., Job, K.,
Larson, T.R. and Graham, I.A. (2005) Sucrose rescues seedling
establishment but not germination of Arabidopsis mutants dis-
rupted in peroxisomal fatty acid catabolism. Plant J. 43, 861–872.
Preisig-Müller, R., Guhnemann-Schafer, K. and Kindl, H. (1994)
Domains of the tetrafunctional protein acting in glyoxysomal fatty
acid b-oxidation. Demonstration of epimerase and isomerase
activities on a peptide lacking hydratase activity. J. Biol. Chem.
269, 20475–20481.
Richmond, T.A. and Bleecker, A.B. (1999) A defect in b-oxidation
causes abnormal inflorescence development in Arabidopsis.
Plant Cell, 11, 1911–1923.
van Roermund, C.W.T., Tabak, H.F., van den Berg, M. and Wanders,
R.J.A. (2000) Pex11p plays a primary role in medium-chain fatty
acid oxidation, a process that affects peroxisome number and
size in Saccharomyces cerevisiae. J. Cell Biol. 150, 489–497.
Rottensteiner, H., Kal, A.J., Hamilton, B., Ruis, H. and Tabak, H.F.
(1997) A heterodimer of the Zn2Cys6 transcription factors Pip2
and Oaf1p controls induction of genes encoding peroxisomal
proteins in Saccharomycres cerevisiae. Eur. J. Biochem. 247, 776–
783.
Rylott, E.L., Rogers, C.A., Gilday, A.D., Edgell, T., Larson, T.R. and
Graham, I.A. (2003) Arabidopsis mutants in short- and medium-
chain acyl-CoA oxidase activities accumulate acyl-CoAs and re-
veal that fatty acid b-oxidation is essential for embryo develop-
ment. J. Biol. Chem. 278, 21370–21377.
Rylott, E.L., Gilday, A.D. and Graham, I.A. (2004) The gluconeogenic
enzyme phosphoenolpyruvate carboxykinase in Arabidopsis is
essential for seedling establishment. Plant Physiol. 131, 1834–
1842.
Schrader, M., Reuber, B.E., Morrell, J.C., Jiminez-Sanchez, G., Obie,
C., Stroh, T., Valle, D., Schroer, T. and Gould, S.J. (1998)
Expression of PEX11b mediates peroxisome proliferation in the
absence of extracellular stimuli. J. Biol. Chem. 273, 29607–29614.
Smith, J.J., Brown, T.W., Eitzen, G.A. and Rachubinski, R.A. (2000)
Regulation of peroxisomal size and number by fatty acid b-oxi-
dation in the yeast Yarrowia lipolytica. J. Biol. Chem. 275, 20168–
20178.
Sussman, M.R., Amasino, R.M., Young, J.C., Krysan, P.J. and
Austin-Phillips, S. (2000) The Arabidopsis knockout facility at the
University of Wisconsin-Madison. Plant Physiol. 124, 1465–1467.
ª 2006 The Authors
 b-oxidation in Arabidopsis mfp2 mutant seedlings 941

Theodoulou, F.L., Job, K., Slocombe, S.P., Footitt, S., Holdsworth,
M., Baker, A., Larson, T.R. and Graham, I.A. (2005) Jasmonic acid
levels are reduced in COMATOSE ATP-binding cassette trans-
porter mutants. Implications for transport of jasmonate precur-
sors into peroxisomes. Plant Physiol. 137, 835–840.
Wasternack, C. and Parthier, B. (1997) Jasmonate signalled plant
gene expression. Trends Plant Sci. 2, 302–307.
Zolman, B.K., Yoder, A. and Bartel, B. (2000) Genetic analysis of
indole-3-butyric acid responses in Arabidopsis thaliana reveals
four mutant classes. Genetics, 156, 1323–1337.
Zolman, B.K., Silva, I.D. and Bartel, B. (2001) The Arabidopsis pxa1
mutant is defective in an ATP-binding cassette transporter-like
protein required for peroxisomal fatty acid b-oxidation. Plant
Physiol. 127, 1266–1278.

ª 2006 The Authors

Journal compilation ª 2006 Blackwell Publishing Ltd, The Plant Journal, (2006), 45, 930–941