The Critical Role of Arabidopsis Electron-TransferFlavoprotein:Ubiquinone Oxidoreductase duringDark-Induced Starvation W

Kimitsune Ishizaki,a Tony R. Larson,b Nicolas Schauer,c Alisdair R. Fernie,c Ian A. Graham,b

and Christopher J. Leavera,1

a Department of Plant Sciences, University of Oxford, Oxford OX1 3RB, United Kingdomb Department of Biology, Centre for Novel Agricultural Products, University of York, Heslington, York YO10 5YW,

United Kingdomc Max-Planck-Institut fur Molekulare Pflanzenphysiologie, 14476 Golm, Germany

In mammals, electron-transfer flavoprotein:ubiquinone oxidoreductase (ETFQO) and electron-transfer flavoprotein (ETF) are

functionally associated, and ETF accepts electrons from at least nine mitochondrial matrix flavoprotein dehydrogenases

and transfers them to ubiquinone in the inner mitochondrial membrane. In addition, the mammalian ETF/ETFQO system

plays a key role in b-oxidation of fatty acids and catabolism of amino acids and choline. By contrast, nothing is known of the

function of ETF and ETFQO in plants. Sequence analysis of the unique Arabidopsis thaliana homologue of ETFQO revealed

high similarity to the mammalian ETFQO protein. Moreover, green fluorescent protein cellular localization experiments

suggested a mitochondrial location for this protein. RNA gel blot analysis revealed that Arabidopsis ETFQO transcripts

accumulated in long-term dark-treated leaves. Analysis of three independent insertional mutants of Arabidopsis ETFQO

revealed a dramatic reduction in their ability to withstand extended darkness, resulting in senescence and death within 10 d

after transfer, whereas wild-type plants remained viable for at least 15 d. Metabolite profiling of dark-treated leaves of the

wild type and mutants revealed a dramatic decline in sugar levels. In contrast with the wild type, the mutants demonstrated

a significant accumulation of several amino acids, an intermediate of Leu catabolism, and, strikingly, high-level accu-

mulation of phytanoyl-CoA. These data demonstrate the involvement of a mitochondrial protein, ETFQO, in the catabolism

of Leu and potentially of other amino acids in higher plants and also imply a novel role for this protein in the chlorophyll

degradation pathway activated during dark-induced senescence and sugar starvation.

INTRODUCTION

In mammals, the nuclear-encoded mitochondrial protein, elec-

tron-transfer flavoprotein:ubiquinone oxidoreductase (ETFQO),

which is associated with the inner mitochondrial membrane,

accepts electrons from the electron-transfer flavoprotein (ETF)

localized in the mitochondrial matrix and reduces ubiquinone

(Ruzicka and Beinert, 1977; Beckmann and Frerman, 1985,

1987). ETF is the physiological electron acceptor for at least

nine mitochondrial matrix flavoprotein dehydrogenases; hence,

the ETF/ETFQO system can be thought of as a branch of the

electron transport system with multiple input sites from seven

acyl-CoA dehydrogenases and two N-methyl dehydrogenases

(Frerman, 1988; Frerman and Goodman, 2001). These dehydro-

genases include four chain length–specific acyl-CoA dehydro-

genases involved in straight-chain fatty acid b-oxidation:

isovaleryl CoA dehydrogenase, 2-methyl branched-chain acyl-

CoA dehydrogenase, and glutaryl-CoA dehydrogenase, which

play a role in the oxidation of amino acids; and sarcosine and

dimethylglycine dehydrogenases, which are involved in choline

metabolism (Frerman, 1988; Frerman and Goodman, 2001). The

mammalian mitochondrial proteins, ETF and ETFQO, are essen-

tial for the catabolism of fatty acids, several amino acids, and

choline and are important in supplying mitochondria with re-

spiratory substrates auxiliary to those derived from sucrose. In

humans, mutations in either ETF or ETFQO results in the fatal

genetic disease type II Glutaric acidemia (multiple acyl-CoA

dehydrogenase dysfunctional disease) (Frerman and Goodman,

2001).

In the higher plantArabidopsis thaliana, botha- andb-subunits

of ETF (the corresponding genes are At1g50940 and At5g43430,

respectively) were identified in mitochondria by the use of gel-

based or liquid chromatography tandem mass spectrometry

mitochondrial proteomic analysis (Heazlewood et al., 2004).

However, the ETFQO protein has not been identified in plant

mitochondria, although a unique homologue of the ETFQO gene

(At2g43400) has been identified in the Arabidopsis genome

(Arabidopsis Genome Initiative, 2000).

One of the functions of the ETF/ETFQO system in mammals is

to allow respiration of substrates other than glucose. In plants,

sucrose availability can be restricted during extended darkness

1 To whom correspondence should be addressed. E-mail chris.leaver@plants.ox.ac.uk; fax 44-01865-275144.The author responsible for distribution of materials integral to thefindings presented in this article in accordance with the policy describedin the Instructions for Authors (www.plantcell.org) is: Christpher J.Leaver (chris.leaver@plants.ox.ac.uk).W Online version contains Web-only data.Article, publication date, and citation information can be found atwww.plantcell.org/cgi/doi/10.1105/tpc.105.035162.

The Plant Cell, Vol. 17, 2587–2600, September 2005, www.plantcell.org ª 2005 American Society of Plant Biologists

(Graham et al., 1992); however, plants have apparently evolved

a range of metabolic responses that allow them to survive for

extended periods during dark-induced sucrose starvation. Such

metabolic responses lead to the mobilization of alternate cellular

components, which compensate for decreases in primary re-

spiratory substrate and maintain important biochemical pro-

cesses, thus allowing survival in adverse conditions (Aubert et al.,

1996). Genome-wide responses to sucrose starvation have been

recently highlighted by several microarray experiments, where it

was shown that a range of genes involved in metabolic pro-

cesses, such as lipid metabolism, chlorophyll degradation, pro-

tein degradation, and amino acid catabolism, are upregulated,

whereas genes encoding components of pathways such as

photosynthesis, carbon fixation, ribosome biogenesis, amino

acid biosynthesis, glycolysis, and many others are downregu-

lated (Contento et al., 2004; Lin and Wu, 2004; Thimm et al.,

2004; Buchanan-Wollaston et al., 2005).

In this study, we have investigated the role of ETFQO in

Arabidopsis during sucrose starvation induced by extended dark

treatment and have focused on the characterization of Arabi-

dopsis T-DNA knockouts of the gene encoding ETFQO at the

molecular, physiological, and metabolite levels. We have con-

firmed the mitochondrial location of the protein by means of

green fluorescent protein (GFP) localization experiments and

have performed expression analysis of ETFQO in Arabidopsis

seedlings following transfer to extended dark periods. Finally, we

have studied the consequences of these gene mutations on plant

viability following periods of extended darkness (up to 15 d) in

addition to studying their impact on cellular metabolism under

these conditions. The data are discussed in the context of

current models of mitochondrial metabolism in plant tissues.

RESULTS

Arabidopsis ETFQO Is a Unique Homologue of Human

ETFQO and Is Localized in Mitochondria

We have identified a homologue of ETFQO in the Arabidopsis

genome, designated as ETFQO (At2g43400), that encodes

a protein with 70% similarity to human ETFQO (accession

number Q16134). The mammalian ETFQO is a unique integral

membrane protein containing one equivalent of flavin adenine

dinucleotide (FAD) and a [4Fe-4S]2þ1þ cluster (Goodman et al.,

1994), characteristics shared with the Arabidopsis ETFQO. The

N-terminal sequence of the Arabidopsis ETFQO protein has

characteristics consistent with that of a mitochondrial transit

peptide and is predicted to target ETFQO to the mitochondrion

by the three targeting prediction programs Predotar (http://

genoplante-info.infobiogen.fr/predotar/), MitoProt II (Claros and

Vincens, 1996), and PSORT (Nakai and Horton, 1999), consistent

with the location of mammalian ETFQO in mitochondria. Align-

ment of the amino acid sequence of Arabidopsis ETFQO

to homologues from bacteria to human shows a high over-

all similarity, including a putative FAD binding domain and

[4Fe-4S]2þ1þ cluster domain (see Supplemental Figure 1 online).

Taken together, the above observations give strong support to

our claim that Arabidopsis ETFQO is a functional homologue of

ETFQO from other organisms.

In order to confirm the in vivo subcellular localization of

Arabidopsis ETFQO, a full-length cDNA fragment was fused to

the cDNA of GFP (Figure 1A). The cDNA fusion construct was

subcloned downstream of the cauliflower mosaic virus (CaMV)

35S RNA promoter and transformed into Arabidopsis. Five-day-

old seedlings of stable transformants were incubated with

MitoTracker Orange for specific staining of mitochondria. In-

spection of the seedlings transformed with the Pro35S:ETFQO:

GFP construct by confocal microscopy revealed that the GFP

fluorescence within root cells was restricted to numerous spher-

ical or elongated bodies, which correspond well with mitochon-

drial morphology (Figures 1B to 1D). The colocalization of the

MitoTracker Orange dye and the GFP signal in root cells

confirmed mitochondrial targeting of the ETFQO:GFP fusion

protein (Figures 1B to 1D). These data clearly demonstrate the

mitochondrial targeting of the ETFQO:GFP fusion protein in vivo

and thus establish the mitochondrial localization of the ETFQO in

Arabidopsis.

ETFQO Is Induced under Sucrose-Starved Conditions

The ETFQO gene is represented on the Affymetrix ATH1 Ge-

nome Array, and analysis of microarray data stored in the

GENEVESTIGATOR database (Zimmermann et al., 2004) re-

vealed that ETFQO transcripts are expressed constitutively at a

low level in all plant tissues and developmental states, including

leaf senescence (Buchanan-Wollaston et al., 2005); however, in-

triguingly, it was highly expressed in dark-induced senescent

leaves (Lin and Wu, 2004), senescent cell cultures (Buchanan-

Wollaston et al., 2005), and sucrose-starved cell cultures

(Contento et al., 2004). RNA gel blot analysis confirmed that

ETFQO expression was not significantly induced during normal

developmental senescence (data not shown).

To further investigate regulation of ETFQO, transcript levels

were measured in 4-week-old Arabidopsis leaves following

transfer to darkness by RNA gel blot analysis. During dark

adaptation, the transcripts from a light-regulated photosynthetic

gene, Cab4, gradually declined (Figure 2) (Fujiki et al., 2001),

whereas the ETFQO transcripts accumulated to very high levels

(Figure 2). As a positive control, the transcript levels of the

isovaleryl-CoA dehydrogenase gene (IVD) and the E1a subunit of

branched-chain a-keto dehydrogenase gene (BCKDH), which

have previously been shown to be induced by sucrose starvation

(Daschner et al., 2001; Fujiki et al., 2001), were analyzed. The IVD

and BCKDH transcripts also accumulated in the dark-adapted

leaves (Figure 2), suggesting that expression of the ETFQO gene

is also induced by sucrose starvation following prolonged

darkness.

Isolation of T-DNA InsertionMutants ofArabidopsis ETFQO

To investigate the in vivo functions of the ETFQO protein, three

independent Arabidopsis lines containing T-DNA or dissociation

(Ds) transposon element insertional mutations of ETFQO were

isolated. Segregation of the encoded resistance markers of all

three insertion elements were in good agreement with the 3:1

(resistant:susceptible) ratio, suggesting insertion at a single Men-

delian locus. Homozygous lines for each mutant were confirmed

2588 The Plant Cell

by genomic PCR and designated etfqo-1, etfqo-2, and etfqo-3,

respectively (Figure 3A), and the respective sites of insertion of

these mutants were confirmed by sequencing PCR products

amplified from each mutant.

As described above, ETFQO contains a putative [4Fe-4S]2þ1þ

cluster domain near its C terminus that is likely to be required for

electron flow from flavin to ubiquinone. In all three mutant lines,

the insertional element is positioned upstream of the putative

[4Fe-4S]2þ1þ cluster, and it would be anticipated that this would

severely compromise its electron transfer capacity and so result

in a complete loss of functional protein. RT-PCR using primer

pairs designated to span the T-DNA insertion site of each mutant

locus was used to investigate ETFQO transcription. The Arabi-

dopsis b-tubulin gene was used as a control to demonstrate the

integrity of the RNA preparation. ETFQO mRNAs were detected

in the wild type (Columbia-0 [Col-0] and Landsberg erecta

[Ler-0]) using the primer sets L1/R1 and L2/R2. In etfqo-1,

ETFQO transcripts were detected using the L2/R2 primer set;

however, no amplification products were observed for the trans-

cripts using the L1/R1 primer set. In etfqo-2 and etfqo-3, ampli-

fication products were not detected using either the L1/R1 or

L2/R2 primer set (Figure 3B). These results confirmed that trans-

cripts spanning the T-DNA insertion site, and downstream of

the T-DNA insertion, are absent in all these mutant lines.

Phenotype of etfqoMutant Lines

After the molecular identity of the T-DNA insertional mutants was

established, they were grown in soil under long-day conditions

alongside the corresponding wild-type controls. Under these

conditions, there was no visible phenotype in the mutants during

the early stage of vegetative growth; however, in the reproduc-

tive stage, etfqo-1 and etfqo-2 plants often produced shorter

siliques with a lower number of seeds than the corresponding

wild type, whereas etfqo-3 had siliques with length and number

of seeds similar to the wild type (see Supplemental Table 1

online). This difference might be due to the fact that the etfqo-3

mutant is in a different wild-type strain background (Ler-0) from

Figure 1. Mitochondrial Localization of an ETFQO:eGFP Fusion Protein in Arabidopsis Seedlings.

(A) The construct carrying the gene encoding ETFQO:eGFP fusion protein. The ETFQO cDNA encoding the full-length coding sequence except stop

codon (ETFQO, yellow box), which includes a sequence for the predicted mitochondrial target peptide (red box), is cloned upstream of the eGFP

reading frame (eGFP, green box). Expression of the resulting fusion protein is controlled by the CaMV 35S promoter (blue box) and the NOS terminator

(gray box).

(B) to (D) Images taken from Arabidopsis root cells expressing the ETFQO:eGFP fusion protein. (B), (C), and (D) are GFP signal (green), MitoTracker

Orange signal (blue), and merged image of (C) and (D), respectively.

Figure 2. Accumulation of Transcripts from the ETFQO in Dark-Treated

Arabidopsis Leaves.

RNA was isolated from leaves harvested at indicated times after the

onset of dark treatment. As control, we examined the expression of the

IVD gene, that of BCKDH E1a subunit gene, and that of the chlorophyll

a/b binding protein gene (Cab). Equal loading was confirmed by compar-

ing ethidium bromide staining of rRNA bands.

Functional Analysis of ETFQO in Arabidopsis 2589

etfqo-1 and etfqo-2 (Col-0). From crosses using wild-type pollen

as the male parent and etfqo-1 and etfqo-2 as female parents,

shorter siliques with lower seed set were produced, whereas

crosses using pollen from etfqo-1 and etfqo-2 to create crosses

with wild-type plants as female parent produced normal si-

liques with seed number similar to wild-type siliques, suggest-

ing an effect of the ETFQO defect on the development of

female gametophyte cells (see Supplemental Table 2 online).

When 5-week-old etfqo mutants grown under short-day con-

ditions were transferred to dark conditions, a dramatic pheno-

type became apparent. All of the homozygous mutant lines

started to wilt and show signs of senescence after 10 d of

continuous darkness and were apparently dead after 15 d of

continuous darkness, whereas both wild-type ecotypes are still

alive after 15 d of continuous darkness and show only limited

signs of senescence (Figure 4A). To further investigate this

apparent accelerated senescence in the etfqo mutants, we

measured two parameters related to the function of chloroplasts,

chlorophyll content and photochemical efficiency (maximum

variable fluorescence/maximum yield of fluorescence [Fv/Fm]),

as diagnostics for leaf senescence (Oh et al., 1996). During

extended dark conditions, chlorophyll content declined more

rapidly in etfqo mutants than in the wild type (Figure 4B), and

the decline of chlorophyll content was accompanied by a de-

crease in the photochemical efficiency of photosystem II (PSII)

(Fv/Fm) (Figure 4C). This result indicates that the process of se-

nescence is more rapid in etfqo mutants than wild-type plants

during extended dark conditions.

In order to confirm that the phenotypes of the mutant plants

were caused by mutations in ETFQO, a CaMV 35S promoter–

driven ETFQO open reading frame (ORF) was introduced into the

etfqo-1 mutant (designated as etfqo-1/OE line) and shown to

completely reverse the partial sterility and prevent the acceler-

ated senescence phenotype observed in the etfqo-1 mutant line

(Figure 4A; see Supplemental Table 1 online).

Leu Catabolism Is Blocked in etfqoMutants

In mammals, defects in ETFQO result in a functional deficiency of

mitochondrial flavoprotein dehydrogenases and accumulation of

their substrates (Frerman and Goodman, 2001). To gain further

insight into the role of the ETFQO protein inArabidopsis, we used

the method of Larson and Graham (2001) to analyze changes in

acyl-CoAs in the wild type and etfqo mutants during the

extended dark treatment. The results shown in Figure 5 demon-

strate a dramatic increase in the amount of five-carbon acyl-CoA

in the etfqo-1 and etfqo-2 mutants compared with the wild type

during the extended dark period. There were four candidates in

Arabidopsis for the five-carbon acyl-CoA compounds: isovaleryl-

CoA (3-methyl-butyryl-CoA), 2-methyl-butyryl-CoA, valeryl-CoA,

and acetoacetyl-CoA. However, the retention time of synthetic

acetoacetyl-etheno-CoA was earlier than the five-carbon acyl-

CoA peak (data not shown), eliminating acetoacetyl-CoA as a

candidate for the peak. Furthermore, given the improved reso-

lution among peaks in the short-chain acyl-CoA region by HPLC

using the Hypercarb porous graphitic carbon column, we could

successfully discriminate this five-carbon acyl-CoA peak from

valeryl-CoA and 2-methyl-butyryl-CoA, and identified this peak

as isovaleryl-CoA (Figure 5B). Isovaleryl-CoA is an intermediate

of the Leu catabolic pathway, being subsequently dehydrogen-

ated to 3-methylcrotonyl-CoA by isovaleryl-CoA dehydroge-

nase. In animals, ETF is the physiological electron acceptor of

isovaleryl-CoA dehydrogenase (Ikeda and Tanaka, 1983). These

results suggest a similar interaction of isovaleryl-CoA dehydro-

genase, ETF, and ETFQO in Arabidopsis.

Involvement of ETFQO in Phytol Degradation

Acyl-CoA profiles from both etfqo-1 and etfqo-2 mutants from

the 7- and 10-d extended dark treatments also showed dramat-

ically increased content of a compound that had the same mass

as the straight-chain eicosanoyl (20:0)-CoA, but which had

a different retention time from the 20:0-CoA peak in the standard

mixture. The retention time of this unknown acyl-CoA peak from

the etfqo-1 mutant was identical to synthetic phytanoyl-CoA

(Figure 6A), and both shared the same molecular mass and

diagnostic acyl-CoA–specific fragmentation pattern when ana-

lyzed by liquid chromatography–mass spectrometry (LC-MS)

(data not shown). Unfortunately, LC-MS fragmentation did not

provide any information on the branching pattern of the acyl

chain and so could not distinguish 20:0-CoA from phytanoyl-

CoA. Therefore, in order to further confirm the identity of this

unknown peak, it was purified in bulk, transmethylated, and

analyzed by gas chromatography–mass spectrometry (GC-MS).

The mass spectrum of the transmethylated peak was distin-

guishable from that of eicosanoic acid methyl ester but matched

Figure 3. Identification of etfqo-1, etfqo-2, and etfqo-3 Mutants.

(A) Genomic structure of ETFQO loci. Arrows represent positions of

primers used for genotype and RT-PCR analyses of wild-type and

mutant lines; closed boxes indicate exons. In etfqo-1 and etfqo-2, the

T-DNA is inserted in exon 4 and exon 12, respectively. In etfqo-3,

Ds transposon element is inserted in exon 1.

(B) RT-PCR analysis on total RNA from the wild type, Col-0, Ler-0, and

the mutant lines etfqo-1, etfqo-2, and etfqo-3, with primer sets indicated

on the left and the positions of each primer in ETFQO genomic loci

represented in (A).

2590 The Plant Cell

well with that of phytanic acid methyl ester, which displayed an

electron-impact fragmentation spectrum expected for methyl-

branched acyl chains (Figure 6B). From these results, the accu-

mulated unknown peak near 20:0 acyl-CoA was unequivocally

identified as phytanoyl-CoA. During extended dark treatment,

etfqo-1 and etfqo-2 mutants contained dramatically increased

concentrations of phytanoyl-CoA relative to the wild type (Fig-

ures 4B and 6A). However, analysis of leaf lipid fatty acids by GC

showed no measurable increase in phytanic acid during the ex-

tended dark treatment (data not shown). In humans, phytanoyl-

CoA is a known breakdown product of phytol, the hydrophobic

tail of chlorophylls a and b. That said, relatively little is currently

known about the exact mechanism of phytol or chlorophyll deg-

radation in plants (Matile et al., 1999; Hortensteiner, 2004);

however, our data suggest that Arabidopsis ETFQO is involved

in phytol degradation.

Metabolite Contents of ETFQOMutants during the

Extended Dark Condition

Further characterization of the accelerated senescence pheno-

type in etfqo mutants was performed using a nonbiased GC-MS

metabolic profiling protocol allowing the simultaneous determi-

nation of >100 compounds (Roessner et al., 2001) As was ex-

pected, the extended dark treatment led to a rapid decline in

sucrose and other sugar contents in both the wild type and

etfqo-1 (Figure 7). Most free amino acids increased significantly

in both the wild type and etfqo-1, including, Leu, Isle, Val, Asn,

Arg, Trp, and Phe, suggesting that protein degradation was

elevated under these conditions (Figure 7). The significantly

elevated levels of several amino acids, especially Leu and Trp, in

the etfqo mutants suggest involvement of ETFQO in their deg-

radation pathways. Analysis of tricarboxy acid (TCA) cycle

intermediates showed a dramatic increase in succinate, fuma-

rate, and malate in etfqo-1 at 7 and 10 d dark treatment (see

Supplemental Figure 2 online). While these accumulations are

striking, the exact mechanism underlying this phenomenon

cannot be elucidated from the results in this study. There are

two possible explanations for the progressive accumulation of

these metabolites. First, it is conceivable that the TCA cycle is

progressively upregulated in the mutant, during the course of the

extended darkness, in an attempt to compensate for the reduced

availability of respiratory substrate. Secondly, the accumulation

of TCA cycle intermediates may be a consequence of a general

downregulation of biosynthesis that would be anticipated un-

der conditions of carbon starvation. While we favor the second

Figure 4. Phenotype of etfqo Mutants in Extended Dark Treatment.

(A) Photos of 4-week, short-day grown Arabidopsis plants after further

growth for 0 and 10 d in extended darkness. The plant designated as

etfqo-1/OE is the etfqo-1 line transformed with the CaMV 35S promoter–

driven ETFQO ORF. The leaves of etfqo mutants, etfqo-1, etfqo-2, and

etfqo-3,wereyellowedanddehydrated following10dofgrowth indarkness

compared with the wild-type control, Col-0 and Ler-0, and etfqo-1/OE.

(B) Chlorophyll (line graph) and phytanoyl-CoA (bar graph) content of 9th

or 10th leaves of 4-week, short-day grown Arabidopsis plants of wild

type (Col-0), etfqo-1, and etfqo-2 after further growth for 0, 3, 7, and 10 d

in extended dark treatment. Values are means 6 SD for six independent

samplings. FW, fresh weight.

(C) Fv/Fm, maximum quantum yield of PSII electron transport (maximum

variable fluorescence/maximum yield of fluorescence) of 9th or 10th

leaves of 4-week, short-day grown Arabidopsis plant of the wild type

(Col-0), etfqo-1, and etfqo-2 after further growth for 0, 3, 7, and 10 d in

extended darkness. Values are means 6 SD for six independent

samplings.

Functional Analysis of ETFQO in Arabidopsis 2591

Figure 5. Accumulation of Isovaleryl-CoA in etfqo-1 and etfqo-2 Mutants in Extended Dark Treatment.

(A) Acyl-CoA profiles of the wild type (Col-0), etfqo-1, and etfqo-2 during the extended dark treatment. Values are means 6 SE for six samples of 10 mg

each, derivatized to their acyl-etheno-CoA esters, separated by HPLC, and detected fluorometrically. FW, fresh weight.

2592 The Plant Cell

hypothesis, further experimentation, including direct measure-

ments of fluxes within the constituent pathways, will ultimately be

required in order to fully comprehend these data. Interestingly,

Glu and pyroglutamic acid concentrations declined in etfqo-1 at

7 to 10 d, whereas they increased in the wild type (Figure 7; see

Supplemental Figure 2 online). It has been suggested, in sucrose

starvation experiments, that amino acid catabolism is activated

and supplies respiratory substrates to the TCA cycle (Contento

et al., 2004); however, in etfqo-1, one or more amino acid

catabolic pathways appear to be blocked. One possible conse-

quence of this block could be an overactivation of Glu catabolism

in order to compensate for the shortfall of respiratory substrate.

Furthermore, accumulation of uracil and ribose in etfqo-1 in the

10-d dark treatment suggests that RNA degradation was acti-

vated in etfqo mutants at this stage, and this is most probably

associated with the process of cell death (data not shown). The

above observations were all made by analyzing data on the basis

of specific metabolites; however, interesting features of the data

also emerge when the entire metabolite profiles are considered.

While the majority of the metabolite levels, in the etfqo-1 mutant,

exhibit monophasic behavior with respect to time in continuous

darkness, a subset of metabolites, such as Gln, Gly, oxyproline,

Pro, mannose and trehalose, exhibit biphasic behavior. It is

tempting to speculate that monophasic responses are observed

for the metabolites more intimately related to the reactions

catalyzed by ETFQO, while the biphasic responses are probably

secondary effects arising from the extreme stress that the etfqo

mutants are under following longer periods of darkness. How-

ever, further experimentation will be required to determine

whether this is indeed the case. Future work should also include

characterization of mannose and trehalose metabolism in these

lines, since both of these carbohydrates increase following

extensive periods in continuous darkness despite initially de-

creasing. It is important to note that although the data presented

here were solely determined from etfqo-1, essentially the same

results were obtained from analysis of a second mutant (etfqo-2).

DISCUSSION

The essential roles in central metabolism of mammalian ETF and

ETFQO have been well established; however, we have no

knowledge of the role of their homologues in higher plants. ETF

and ETFQO homologues were identified in the Arabidopsis

genome and also in the genomes and EST databases of many

other plant species, including, rice (Oryza sativa), tomato (Lyco-

persicon esculentum), barley (Hordeum vulgare), and wheat

(Triticum aestivum) (Plant Genome Database; http://

www.plantgdb.org/). This suggests that ETF and ETFQO are

widely expressed in higher plants and potentially play an impor-

tant role in plant metabolism. In this article, we describe an

investigation of the functional importance of ETFQO in Arabi-

dopsis and its role during dark-induced senescence.

In mammals, the reducing equivalents from various flavoprotein

dehydrogenases are transferred via ETF to ETFQO, which is

localized in the inner mitochondrial membrane, and the accepted

electrons are further passed from ETFQO into the electron trans-

port chain via coenzyme Q to the CoQH2-cytochrome c oxidore-

ductase (Complex III) (Parker and Engel, 2000). The deduced

amino acid sequence of Arabidopsis ETFQO showed high overall

similarity to that of human ETFQO, including its two putative

functional domains, the FAD binding domain, and the [4Fe-4S]2þ1þ

cluster domain (see Supplemental Figure 1 online) and a mito-

chondrial targeting peptide at the N terminus. We confirmed the

mitochondrial localization ofArabidopsis ETFQO by colocalization

of MitoTracker Orange with a fusion to GFP (Figure 1). In Arabi-

dopsis, a- and b-subunits of ETF have been found in mitochondria

(Heazlewood et al., 2004), and their amino acid sequences also

show high similarity to the both subunits of the human ETF (data

not shown). Hence, it is quite feasible that the basic function of

ETFQO in electron transfer to coenzyme Q in the mitochondrial

respiratory chain is conserved between plants and animals.

ETFQO Is Induced and Essential for Viability

in Sucrose-Starved Condition

Analysis of microarray data in public databases and RNA gel blot

analyses confirmed that ETFQO was constitutively expressed in

all tissues and stages of development in Arabidopsis plants and

was upregulated following transfer of mature plants to extended

periods of darkness over a period of 15 d (Figure 2). During

experimentally induced sucrose starvation, there is a marked

decrease in expression of many genes involved in carbohydrate

breakdown, glycolysis, the TCA cycle, mitochondrial electron

transport, and ATP synthesis (Buchanan-Wollaston et al., 2005).

This is associated with a decrease in respiratory energy metab-

olism and increased expression of a number of genes involved in

proteolysis, amino acid catabolism, and fatty acid degradation,

suggesting the induction of salvage mechanisms that release

amino acids and fatty acids from structural proteins and lipids as

an alternative source of respiratory substrate to maintain viability

of carbohydrate-starved cells (Contento et al., 2004; Lin and Wu,

2004; Thimm et al., 2004; Buchanan-Wollaston et al., 2005). The

increased expression ofETFQO in dark-adapted plants (Figure 2)

suggests a central role for ETFQO in supplying mitochondria with

an alternative source of respiratory substrates from proteolysis

and/or lipid degradation. The observation that the etfqo mutants

senesce and die before wild-type plants under extended dark

Figure 5. (continued).

(B) Identification of the increased short-chain acyl-CoA peak in etfqo-1. Short-chain acyl-CoAs were chemically synthesized from their respective fatty

acids, derivatized to their acyl-etheno-CoA esters, separated by HPLC on a Hypercarb porous graphitic carbon column, and detected by MS. The 5:0

acyl-etheno-CoA peaks were identified by their characteristic parent and daughter ion mass-to-charge (m/z) ratios of 876 and 452, respectively. Peaks were

aligned against added heptanoyl-etheno-CoA (7:0) internal standard (ISTD; m/z 904). Shown are the extracted ion chromatograms for m/z values of 876 and

904. i, Valeryl-etheno-CoA; ii, 2-methyl-butyryl-etheno-CoA; iii, isovaleryl (3-methyl butyryl)-etheno-CoA; iv, etfqo-1 extract at 10 d in extended darkness.

Functional Analysis of ETFQO in Arabidopsis 2593

Figure 6. Identification of Phytanoyl-CoA Peak.

(A) Comparison of phytanoyl-CoA peak in etfqo-1 with standards. HPLC profiles of acyl-etheno-CoA derivatives are shown for etfqo-1 extract at 10 d in

extended darkness (i), phytanoyl-CoA synthesized from phytanic acid by Pseudomonas acyl-CoA synthetase (ii), and the calibration standard mix (iii).

(B) Confirmation of phytanoyl-CoA identity by GC-MS. Collected fractions of the putative phytanoyl-CoA peak from HPLC runs of derivatized etfqo-1 extracts

at 10 d in extended darkness were subjected to alkaline transmethylation to form the methyl ester. Left panels show GC-MS chromatograms; right panels

show mass spectra for the corresponding peaks. i, eicosanoic acid methyl ester; ii, synthetic phytanic acid methyl ester; iii, transmethylated peak from etfqo-1.

2594 The Plant Cell

conditions (Figure 4A) taken together with their altered metabolic

profiles provides further evidence that ETFQO plays a key role in

supplying alternative respiratory substrates. However, it is pos-

sible that the accelerated senescence and early death of the

etfqo mutants (Figure 4) could be, at least in part, due to the

hyperaccumulation of toxic metabolites, since both branched-

chain amino acids and a-keto acids are known to be cytotoxic

compounds that can induce apoptosis in mammals (Eden and

Benvenisty, 1999; Ogier de Baulny and Saudubray, 2002). Thus,

our results suggest that ETFQO plays an essential role during

sucrose starvation in plants by facilitating the use of alternative

respiratory pathways either by supplying alternate substrates, by

promoting the metabolism of toxic products thereof, or both.

Contribution of ETFQO to the Reproductive Development

of Arabidopsis

The ecotype-specific reduced seed set and silique size of etfqo

mutants (see Supplemental Table 1 online) suggest an additional

role for ETFQO inArabidopsis. Numerous studies on cytoplasmic

male sterility have revealed the important role of the mitochon-

drion during pollen development (Hanson and Bentolila, 2004). In

Figure 7. Relative Levels of Amino Acids and Sugars in the Wild Type (Black Bars) and etfqo-1 Mutant (White Bars) during Extended Dark Conditions.

The y axis values represent metabolite level relative to the wild type. Data were normalized to mean response calculated for the 0-d dark-treated leaves

of the wild type (in case no response was detected at 0 d, normalization was done against 3-d dark-treated leaves of the wild type). Values presented are

mean 6 SE of determinations on six individual plants.

Functional Analysis of ETFQO in Arabidopsis 2595

this case, however, the reproductive phenotype was not asso-

ciated with pollen quality (see Supplemental Table 2 online).

Thus, the mitochondrion also appears to play an important role in

female gametophyte development as previously shown (Skinner

et al., 2001; Christensen et al., 2002).

Involvement of ETFQO in Leu Catabolism

In Arabidopsis, isovaleryl-CoA dehydrogenase is localized in

plant mitochondria (Daschner et al., 2001; Taylor et al., 2004),

and our metabolite profiling data (Figure 5) demonstrated

a marked accumulation of the isovaleryl-CoA dehydrogenase

substrate, isovaleryl-CoA, in etfqo mutants, suggesting dysfunc-

tion of isovaleryl-CoA dehydrogenase in these mutants. These

results further support the existence of the electron transfer

cascade from isovaleryl-CoA dehydrogenase to ETF, ETF to

ETFQO, and ETFQO to ubiquinone in the electron transport chain

of plant mitochondria as reported in mammals (Ikeda and

Tanaka, 1983), although the direct interaction of these four

components has yet to be confirmed.

The recent proteomic study on Arabidopsis mitochondria by

Taylor et al. (2004) revealed that the enzymes necessary for Leu

catabolism are all localized within plant mitochondria, confirming

the findings of 14C-Leu tracer experiments on isolated pea (Pisum

sativum) mitochondria (Anderson et al., 1998). In this study, our

demonstration of a marked accumulation of isovaleryl-CoA, as

a result of mutation in the mitochondrial protein ETFQO, provides

further support for the mitochondrial location of Leu catabolism in

Arabidopsis. By contrast, we did not detect an accumulation

of the intermediates of either Ile or Val catabolism, 2-methyl-

butyryl-CoA, and isobutyryl-CoA, respectively, in etfqo mutants

during the extended dark treatment, although it has been shown

in mammals that ETF/ETFQO is also involved in dehydrogenation

of 2-methyl-butyryl-CoA and isobutyryl-CoA (Ikeda et al., 1983).

Our data imply that the mitochondrial ETF/ETFQO is less involved

in Ile and Val catabolism in Arabidopsis than it is in Leu

catabolism. However, it is important to note that these amino

acids and several others accumulate in the etfqomutants relative

to the wild type (Figure 7). It is not clear why CoA intermediates for

Ile and Val were not detected. It is possible that they are not as

actively degraded as Leu; thus, intermediates are below the level

of detection. Alternatively, the CoA intermediates could be

actively and specifically deacylated to free fatty acids, but short

chain acyl-CoA thioesterases that would catalyze this process

have not been described in plants. Finally it is possible that the

breakdown of Val and Ile may not be dependent on the ETFQO,

and the dehydrogenation/oxidation step could be localized to the

peroxisomes, as is the case for straight-chain fatty acids in higher

plants. Such a hypothesis is consistent with the observations that

the peroxisomal short-chain acyl-CoA oxidase is able to oxidize

isobutyryl-CoA (Hayashi et al., 1999), and the demonstration that

several other enzymes involved in Ile and Val catabolism are

localized in peroxisomes in Arabidopsis (Lange et al., 2004;

Taylor et al., 2004). The mitochondrial isovaleryl-CoA dehydro-

genase does utilize 2-methyl-butyryl-CoA as a substrate in vitro;

however, the Km value of the enzyme for the isobutyryl-CoA is

;10 times larger than that for isovaleryl-CoA (Daschner et al.,

2001). Since the relative concentration of Val is only half that of

Leu, the large difference in Km values suggests that isovaleryl-

CoA dehydrogenase is not involved in Val oxidation in vivo. The

recent finding that the Arabidopsis mitochondrial branched-

chain aminotransferase (BCAT-1) is able to initiate degradation

of Leu, Ile, and Val (Schuster and Binder, 2005) along with

previous proteomic results (Taylor et al., 2004) and those of this

study provide an important foundation for understanding the

degradative pathways of branched-chain amino acids. Intrigu-

ingly, the etfqo mutants also displayed elevated accumulation of

aromatic amino acids, hinting that the ETF/ETFQO system may

also be involved in the metabolism of this family of amino acids.

The Involvement of ETFQO in the Phytol Catabolic

Pathway in Plants

Our observations show that there was a significant accumulation

of phytanoyl-CoA in etfqo mutants during extended dark treat-

ment (Figures 5 and 6), which parallels a decrease in chlorophyll

(Figure 4B). It is well known in humans as a breakdown in-

termediate of phytol, which is derived from chlorophyll in the

diet. In humans, phytol is oxidized to phytanic acid, subse-

quently converted into phytanoyl-CoA by a long-chain fatty acyl-

CoA synthetase, and further degraded in peroxisomes by the

a-oxidation system followed by b-oxidation in mitochondria

(Wierzbicki et al., 2002; Wanders et al., 2003). Disorders in

peroxisomal a-oxidation result in a syndrome called Refsum’s

disease, which results in a dramatic accumulation of phytanic

acid in plasma- and lipid-containing tissues (Wierzbicki et al.,

2002). In plants, phytol is derived from the first step of chlorophyll

degradation, which is catalyzed by chlorophyllase (EC 3.1.1.14)

(Tsuchiya et al., 1999); however, the latter enzymatic steps of

phytol degradation are poorly understood, especially in compar-

ison with the pathway in humans. Previous studies have sug-

gested that phytol degradation during leaf senescence occurs via

photooxidative conversion into various isoterpenoid compounds

(Rontani et al., 1996; Matile et al., 1999; Rontani and Aubert,

2005). This obviously cannot be the case for chlorophyll break-

down during the extended dark treatments in this study. There-

fore, increasing accumulation of phytanoyl-CoA in response to

the length of dark treatment in etfqo mutants suggests the

existence of an enzymatic pathway of phytol degradation in

Arabidopsis mitochondria. In the extended dark condition, chlo-

rophyll breakdown was greater in theetfqomutant than in the wild

type (Figure 4B), which could explain the dramatically increased

concentration of phytanoyl-CoA in etfqo mutants (Figure 4B).

However, it is also possible that ETFQO is directly involved in the

phytanoyl-CoA degradation pathway, and phytanoyl-CoA degra-

dation is blocked in etfqo mutants. In the analysis of leaf lipid

derived fatty acid content of etfqo mutants during the extended

dark treatment, no accumulation of phytanic acid, which is an up-

stream intermediate of phytanoyl-CoA, was observed (data not

shown), suggesting rapid conversion from phytol to phytanoyl-

CoA in Arabidopsis, at least under the conditions studied.

The Function of ETFQO in Mammals and Plants

In mammals, it has been shown that one of the main roles of

ETF/ETFQO is the reoxidation of four straight-chain acyl-CoA

2596 The Plant Cell

dehydrogenases that are essential for the first step of fatty acid

b-oxidation (Bartlett and Eaton, 2004). b-Oxidation of fatty

acids primarily occurs in the mitochondria in mammals, while

b-oxidation in higher plants occurs exclusively in peroxisomes

(Graham and Eastmond, 2002). Our observations are consis-

tent with a peroxisomal localization for b-oxidation in higher

plants in that straight-chain acyl-CoAs do not accumulate in

the etfqo mutants during the dark treatment (Figure 5). Fur-

thermore, measurement of the fatty acid content of etfqo

mutants and the wild type during germination indicated that

fatty acid breakdown was the same in etfqo mutants and the

wild type (data not shown). This result suggests that the

ETFQO is not involved in b-oxidation of straight-chain fatty

acids in Arabidopsis, supporting the postulate that b-oxidation

of straight-chain fatty acids is localized exclusively in peroxi-

somes in Arabidopsis.

In mammals, ETF and ETFQO are also involved in the catab-

olism of several amino acids via dehydrogenation of isovaleryl-

CoA dehydrogenase (Leu) catabolism, 2-methyl branched-chain

acyl-CoA dehydrogenase (Ile and Val), and glutaryl-CoA dehy-

drogenase (Lys, Hyl, and Trp) (Frerman and Goodman, 2001).

While the accumulation of isovaleryl-CoA in the etfqo mutants

showed the involvement of ETFQO in Leu catabolism via iso-

valeryl-CoA dehydrogenase as in mammals, the lack of accu-

mulation of 2-methyl-butyryl-CoA and isobutyryl-CoA does not

allow us to formally conclude a similar role for ETFQO in Ile and

Val catabolism. However, it should be noted that Ile and Val

accumulate dramatically in the etfqo mutants, and it is possible

that ETFQO is involved in catabolism of Ile and Val upstream of

the intermediates 2-methyl-butyryl-CoA and isobutyryl-CoA since

neither of these accumulates. The increase of Trp in GC-MS pro-

files of the etfqo mutants suggests involvement of ETFQO in Trp

catabolism as found in mammals. The accumulation of phytanoyl-

CoA and also amino acids in the etfqo mutants, as revealed by

GC-MS profiling, implies the involvement of ETFQO in the unique

metabolic processes that are not found in animals. As we dis-

cussed above, given the importance of the structure of the mam-

malian ETFQO to its basic functionality, the high amino acid

sequence conservation (70%) between mammalian and Arabi-

dopsis ETFQO suggests that the basic function of ETFQO as the

electron donor between ETF and ubiquinone is conserved. How-

ever, the specific cellular role will be dependent on the interacting

partners of the ETF. Thus, further understanding of these roles in

plants will require identification of the ETF interacting partners.

Conclusion

This study has revealed that ETFQO is essential for plants to

survive in sucrose-depleted conditions induced by extended

growth in the dark and is involved in the Leu catabolic pathway.

The study also shows that phytanoyl-CoA accumulates when

ETFQO is disrupted, thus implicating the mitochondrial ETF/

ETFQO in the degradation of the phytol tail of chlorophyll. The

phytol tail represents ;50% of the carbon in chlorophylls, the

other 50% belongs to the porphyrin ring. On a global scale,

the annual cycle of synthesis and breakdown of chlorophyll is

massive, and more work is now needed to fully understand the

role of the mitochondrial ETFQO in this important process. The

identification of the ETFQO protein in plants clearly demon-

strates the existence and importance of a novel mitochondrial

electron transport complex in higher plants. In mammals, ETF is

associated with ETFQO and several flavoprotein dehydroge-

nases (Parker and Engel, 2000). A similar relationship is likely to

function in higher plants with at least some interacting partners

conferring plant-specific roles, and it is on these that future

studies need to focus.

METHODS

Plant Material and Dark Treatment

Arabidopsis thaliana seeds were surface-sterilized and imbibed for 4 d at

48C in the dark on 0.8% (w/v) agar plates containing half-strength

Murashige and Skoog (MS) media (Sigma-Aldrich; pH 5.7). Seeds were

subsequently germinated and grown at 228C under long-day conditions

(16 h light/8 h dark) with 150 mmol m�2 s�1 white light. For examination of

phenotype in the long-day treatment, seedlings were transferred to soil 7

to 10 d after germination and placed in a growth chamber at 228C under

long-day conditions (16 h light/8 h dark). For dark treatments, 7- to 10-d-

old seedlings were transferred to soil and then grown at 228C under short-

day conditions (8 h light/16 h dark) for 4 weeks. Following bolting, plants

were grown at 228C in the dark in the same growth cabinet. The fully

expanded, 7th to 12th, rosette leaves were harvested at intervals of 0,

3, 7, and 10 d from control and dark-grown plants for the subsequent

experiments.

RNA Gel Blot Analysis

Total RNA was isolated from light-grown control, from dark-treated

leaves, and from cell suspension culture using the TRIzol reagent

(Invitrogen). cDNA fragments for ETFQO, IVD, the E1a subunit ofBCKDH,

chlorophyll a/b binding protein (Cab4), and nitrate reductase (NR2), were

amplified from Arabidopsis cDNA by PCR. RNA was fractionated by

electrophoresis on 1% (w/v) agarose gels, blotted onto positively charged

nylon membrane, and hybridized at 658C with the appropriate cDNA

probes labeled with [a-32P]dCTP using the MegaPrime kit (Amarsham-

Pharmacia). The hybridized membrane was washed for 1 h in a solution

containing 23 SSC and 0.1% (w/v) SDS at 258C followed by two washes

with 0.13 SSC and 0.5% SDS at 558C for 1 h.

Isolation of T-DNA Insertion Mutants and

Genotype Characterization

The three mutant lines, SALK T-DNA line SALK_007870 (etfqo-1, back-

ground ecotype Col), Syngenta Arabidopsis insertion line SAIL_91_E03

(etfqo-2, background ecotype Col), and John Innes Centre Gene Trap line

GT_3_4705 (etfqo-3, background ecotype Ler) were obtained from the

Nottingham Arabidopsis Stock Centre (University of Nottingham). The

mutants were isolated according to published procedures: SIGnAL

(Alonso et al., 2003), Syngenta SAIL line (Sessions et al., 2002), and

John Innes Centre Gene Trap line (Sundaresan et al., 1995). Genotypes of

the different knockout mutant lines were analyzed by PCR using primers

specific for the ETFQO ORF (for etfqo-1, L1 59-AAGGTGGTACCGTG-

CTTCAG-39 and R1 59-CACCACCTTCAGGCACTGTC-39; for etfqo-2 and

etfqo-3, L2 59-AAAGCTATCTTCTTCTTCTTCACCA-39 and R2 59-GGT-

TGCAATCCCAACAACTT-39) and primers specific for the insertion

elements (for etfqo-1, LBa1 59-TGGTTCACGTAGTGGGCCATCG-39;

for etfqo-2, TMR1_LB1 59-GCCTTTTCAGAAATGGATAAATAGCCTTG-

CTTCC-39; for etfqo-3, Ds3-1 59-ACCCGACCGGATCGTATCGGT-39 and

Ds5-1 59-GAAACGGTCGGGAAACTAGCTCTAC-39).

Functional Analysis of ETFQO in Arabidopsis 2597

Complementation of etfqo-1with the CaMV 35S Promoter–Driven

ETFQO ORF

The full-length ORF including stop codon of ETFQO was amplified by RT-

PCR from wild-type Arabidopsis (ecotype Col-0) and subcloned into the

pH2GW7 vector (www.psb.ugent.be/gateway/) (Karimi et al., 2002) using

the Gateway recombination system (Invitrogen). Heterozygous etfqo-1

mutant plants were subsequently transformed by Agrobacterium tume-

faciens–mediated gene transfer utilizing the floral dip method (Clough and

Bent, 1998) because of the partial sterility exhibited by the homozygous

etfqo-1 plant. Positive transformants were selected on half-strength MS

medium containing 0.8% (w/v) agar and 30 mg/mL hygromycin. Homo-

zygous etfqo-1 lines containing the CaMV 35S promoter–driven ETFQO

ORF were characterized by genomic PCR using the primer set L1 and R1

described above in order to check the transformed ETFQO ORF and the

primer set L1 and RI1 (59-TTGTGCAGTGGATGGTTTTG-39) in order to

confirm that there was no amplification from a disrupted copy of ETFQO.

The primer RI1 was designed on the basis of the intron sequence

of ETFQO in order to eliminate amplification from the transformed

ETFQO ORF.

Analysis of ETFQOmRNA Expression by RT-PCR

Total RNA was isolated from rosette leaves of 3 d dark-treated plants

using TRIzol reagent. First-strand cDNA was synthesized from 10 mg of

total RNA with Superscript II Rnase H� reverse transcriptase (Invitrogen)

and oligo(dT) primer. PCR amplification of ETFQO cDNA-specific se-

quence was performed using primers specific for the ORF, L1 and R1 and

L2 and R2, described above. PCR amplification of the cDNA encoding the

b-tubulin of Arabidopsis (TUB9) with a forward primer (59-GATATC-

TGTTTCCGTACCTTGAAGC-39) and a reverse primer (59-ACCGACTG-

TAGCATCTTGATATTGC-39) served as a control.

Subcellular Localization Experiments

In order to generate CaMV 35S promoter–driven ETFQO:eGFP fusion

protein constructs, the 1899-bp coding region of ETFQO was amplified

by RT-PCR from wild-type Arabidopsis (ecotype Col-0) with primers

flanked with attB1 and attB2 sites. The resultant sequence was then

cloned into the entry vector pDONR201 by Gateway recombination

(Invitrogen). This cloned sequence was then transferred from the

pDONR201 vector to the pK7WFG2 vector (www.psb.ugent.be/

gateway/) (Karimi et al., 2002) by Gateway recombination (Invitrogen).

Wild-type Arabidopsis Col-0 was subsequently transformed by

Agrobacterium-mediated gene transfer utilizing the floral dip method

(Clough and Bent, 1998). Positive transformants were selected on half-

strength MS medium containing 0.8% (w/v) agar and 50 mg/mL kana-

mycin. Following selection, transformants were transferred to soil and

seeds collected. The T2 generation was used for the observation of GFP

fluorescence. Five-day-old seedlings were harvested into a freshly made

staining solution of 200 nM CMTMRos (MitoTracker Orange; Molecular

Probes) in half-strength MS medium and allowed to stain for 15 min at

room temperature. After staining, the seedlings were submerged in a half-

strength MS solution for 10 min. The T2 seedlings were visualized by

confocal microscopy with a LSM510 confocal microscope (Zeiss) set to

measure an emission band of 475 to 525 nm for GFP (excitation 488,

emission 522) and an emission band of 565 to 615 nm for MitoTracker

Orange CMTMRos (excitation 551, emission 576). Excitation for GFP was

provided by the 458-nm line of a 25-mW argon laser and for MitoTracker

by a 543-nm HeNe laser. The software LSM examiner (Zeiss) was used for

postacquisition image processing. Several T2 generation lines for each

construct were investigated, and all showed the same fluorescence

pattern.

Measurement of Senescence Parameters

For chlorophyll content, 20 mg of leaf tissue were frozen and ground in

liquid nitrogen and extracted by heating in 1 mL of methanol at 658C for

20 min. After clarification by centrifugation, the chlorophyll concentration

per fresh weight of leaf tissue was calculated as described previously

(Porra et al., 1989). Photochemical efficiency was examined at the given

time points for the dark treatment, using attached leaves of the wild type

(Col-0) and etfqo mutants. The photochemical efficiency of PSII was de-

duced from the characteristics of chlorophyll fluorescence using a por-

table plant efficiency analyzer (Hansatech Instruments) (Oh et al., 1996).

The ratio of Fv to Fm, which corresponds to the potential quantum yield

of the photochemical reactions of PSII, was used as the measure of

the photochemical efficiency of PSII (Oh et al., 1996).

Acyl-CoA and Fatty Acid Profiling

Ten-milligram portions of leaf material were frozen in liquid nitrogen and

extracted for subsequent quantitative analysis of fluorescent acyl-

etheno-CoA derivatives by HPLC with a LUNA 150 3 2 mm C18(2) col-

umn (Phenomenex) using the gradient conditions described previously

(Larson and Graham, 2001; Larson et al., 2002). The lipid portion of the

acyl-CoA extracts was transmethylated for fatty acid analysis by GC with

flame ionization detection (Larson and Graham, 2001).

Commercially available phytanic acid (3,7,11,15-tetramethylhexade-

canoic acid) (Sigma-Aldrich) was used to synthesize phytanic acid methyl

ester by acidic transmethylation (Larson and Graham, 2001). Phytanoyl-

CoA, for use as an HPLC retention time standard, was enzymatically

synthesized from phytanic acid using Pseudomonas acyl-CoA synthe-

tase from Sigma-Aldrich (Taylor et al., 1990). Putative phytanoyl-CoA

peaks from HPLC runs of synthetic and leaf extracts were further

characterized by LC-MS to determine molecular weight and verify acyl-

CoA specific fragmentation patterns (Larson and Graham, 2001). Addi-

tional confirmation of phytanoyl-CoA identity was obtained by GC-MS

analysis of collected HPLC fractions transmethylated to their fatty acid

methyl esters. Putative phytanoyl-CoA fractions from 10-fold concen-

trated and derivatized extracts run on the HPLC were collected, dried

under vacuum, and transmethylated with 500 mL 0.1 M sodium meth-

oxide (Sigma-Aldrich) for 2 h at 858C. The formed methyl ester was

partitioned against 0.25 mL 0.9% (w/v) KCl and removed in 200 mL

hexane. The hexane phase was dried under vacuum and reconstituted in

20 mL fresh hexane, and a 2 mL aliquot was analyzed by GC-MS as

described previously (Tonon et al., 2005), with the exception that a ZB5

30 m 3 0.25 mm i.d. 3 0.25 mm film thickness capillary column (Pheno-

menex) was used.

Short branched-chain acyl-CoA esters were commercially unavailable,

and Pseudomonas acyl CoA synthetase was unable to catalyze the

formation of short branched-chain acyl-CoA esters from branched-chain

5-carbon acids (Sigma-Aldrich). Instead, these were synthesized by the

carbonyldiimidazole method (Kawaguchi et al., 1981). The acyl-CoA

products were derivatized to their etheno derivatives and characterized

by LC-MS. In order to increase the resolution between the C5 isomers,

a Hypercarb 100 3 3 mm porous graphitic carbon column (Thermo

Hypersil) was used in place of the LUNA 15032 mm C18(2) column, using

the same gradient program.

Extraction, Derivatization, and Analysis of Arabidopsis Leaf

Metabolites Using GC-MS

Metabolite analysis by GC-MS was performed by a method modified from

that described by Rossener-Tunai et al. (2003). Arabidopsis leaf tissue

(;180 mg) was homogenized using a mortar and pestle precooled with

liquid nitrogen and extracted in 1400mL of methanol, and 60mL of internal

standard (0.2 mg ribitol mL�1 water) was subsequently added as a

2598 The Plant Cell

quantification standard. The extraction, derivatization, standard addition,

and sample injection were exactly as described previously (Roessner-

Tunali et al., 2003). The GC-MS system and settings were exactly as

described by Roessner et al. (2001). Both chromatograms and mass

spectra were evaluated using the MASSLAB program (ThermoQuest),

and the resulting data were prepared and presented as described by

Roessner et al. (2001).

Accession Numbers

Sequence data from this article can be found in the GenBank/EMBL data

libraries under the following accession numbers: for Arabidopsis ETFQO,

At2g43400; IVD, At3g45300; BCKDH, At1g21400; Cab4, At3g47470; NR2,

At1g37130; TUB9, At4g20890; for rice OsETFQO, AAK39567; for human

HsETFQO, Q16134; for rice MmETFQO, Q921G7; for fly DmETFQO,

NP_610536; for worm CeETFQO, D88483; for yeast ScETFQO, Q08822;

for protobacteria LpETFQO, YP_095306.

ACKNOWLEDGMENTS

We thank Ian Moore and Ooi-kock Teh for assistance with the confocal

microscopy and Nicholas J. Kruger for help with the measure of the

photochemical efficiency of PSII. We also thank Lee J. Sweetlove for

invaluable suggestions. This work was supported in part by a grant from

the Biotechnology and Biological Science Research Council (to C.J.L.),

by a research fellowship of the Japan Society for the Promotion of

Science for Young Scientists (to K.I.), and by a grant from the German-

Israeli Cooperation Project, Deutsch-Israelisches-Projekt (to N.S. and

A.R.F.).

Received June 13, 2005; revised July 5, 2005; accepted July 5, 2005;

published July 29, 2005.

REFERENCES

Alonso, J.M., et al. (2003). Genome-wide insertional mutagenesis of

Arabidopsis thaliana. Science 301, 653–657.

Anderson, M.D., Che, P., Song, J., Nikolau, B.J., and Wurtele, E.S.

(1998). 3-Methylcrotonyl-coenzyme A carboxylase is a component of

the mitochondrial leucine catabolic pathway in plants. Plant Physiol.

118, 1127–1138.

Arabidopsis Genome Initiative (2000). Analysis of the genome se-

quence of the flowering plant Arabidopsis thaliana. Nature 408,

796–815.

Aubert, S., Gout, E., Bligny, R., Marty-Mazars, D., Barrieu, F.,

Alabouvette, J., Marty, F., and Douce, R. (1996). Ultrastructural

and biochemical characterization of autophagy in higher plant cells

subjected to carbon deprivation: Control by the supply of mitochon-

dria with respiratory substrates. J. Cell Biol. 133, 1251–1263.

Bartlett, K., and Eaton, S. (2004). Mitochondrial beta-oxidation. Eur. J.

Biochem. 271, 462–469.

Beckmann, J.D., andFrerman, F.E. (1985). Electron-transfer flavoprotein-

ubiquinone oxidoreductase from pig liver: Purification and molec-

ular, redox, and catalytic properties. Biochemistry 24, 3913–3921.

Buchanan-Wollaston, V., Page, T., Harrison, E., Breeze, E., Lim,

P.O., Nam, H.G., Lin, J.F., Wu, S.H., Swidzinski, J., Ishizaki, K., and

Leaver, C.J. (2005). Comparative transcriptome analysis reveals

significant differences in gene expression and signalling pathways

between developmental and dark/starvation-induced senescence in

Arabidopsis. Plant J. 42, 567–585.

Christensen, C.A., Gorsich, S.W., Brown, R.H., Jones, L.G., Brown,

J., Shaw, J.M., and Drews, G.N. (2002). Mitochondrial GFA2 is

required for synergid cell death in Arabidopsis. Plant Cell 14, 2215–

2232.

Claros, M.G., and Vincens, P. (1996). Computational method to predict

mitochondrially imported proteins and their targeting sequences. Eur.

J. Biochem. 241, 779–786.

Clough, S.J., and Bent, A.F. (1998). Floral dip: A simplified method

for Agrobacterium-mediated transformation of Arabidopsis thaliana.

Plant J. 16, 735–743.

Contento, A.L., Kim, S.J., and Bassham, D.C. (2004). Transcriptome

profiling of the response of Arabidopsis suspension culture cells to

Suc starvation. Plant Physiol. 135, 2330–2347.

Daschner, K., Couee, I., and Binder, S. (2001). The mitochondrial

isovaleryl-coenzyme a dehydrogenase of Arabidopsis oxidizes inter-

mediates of leucine and valine catabolism. Plant Physiol. 126,

601–612.

Eden, A., and Benvenisty, N. (1999). Involvement of branched-chain

amino acid aminotransferase (Bcat1/Eca39) in apoptosis. FEBS Lett.

457, 255–261.

Frerman, F.E. (1987). Reaction of electron-transfer flavoprotein ubiqui-

none oxidoreductase with the mitochondrial respiratory chain. Bio-

chim. Biophys. Acta 893, 161–169.

Frerman, F.E. (1988). Acyl-CoA dehydrogenases, electron transfer

flavoprotein and electron transfer flavoprotein dehydrogenase. Bio-

chem. Soc. Trans. 16, 416–418.

Frerman, F.E., and Goodman, S.I. (2001). Defects of electron transfer

flavoprotein and electron transfer flavoprotein-ubiquinone oxidore-

ductase: Glutaric acidemia type II. In The Metabolic and Molecular

Bases of Inherited Disease, C.R. Scriver, W.S. Sly, B. Childs, A.L.

Beaudet, and D. Valle, eds (New York: McGraw-Hill), pp. 2357–2365.

Fujiki, Y., Ito, M., Nishida, I., and Watanabe, A. (2001). Leucine and its

keto acid enhance the coordinated expression of genes for branched-

chain amino acid catabolism in Arabidopsis under sugar starvation.

FEBS Lett. 499, 161–165.

Goodman, S.I., Axtell, K.M., Bindoff, L.A., Beard, S.E., Gill, R.E., and

Frerman, F.E. (1994). Molecular cloning and expression of a cDNA

encoding human electron transfer flavoprotein-ubiquinone oxidore-

ductase. Eur. J. Biochem. 219, 277–286.

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., Leaver, C.J., and Smith, S.M. (1992). Induction of malate

synthase gene expression in senescent and detached organs of

cucumber. Plant Cell 4, 349–357.

Hanson, M.R., and Bentolila, S. (2004). Interactions of mitochondrial

and nuclear genes that affect male gametophyte development. Plant

Cell 16 (suppl.), S154–S169.

Hayashi, H., De Bellis, L., Ciurli, A., Kondo, M., Hayashi, M., and

Nishimura, M. (1999). A novel acyl-CoA oxidase that can oxidize

short-chain acyl-CoA in plant peroxisomes. J. Biol. Chem. 274,

12715–12721.

Heazlewood, J.L., Tonti-Filippini, J.S., Gout, A.M., Day, D.A.,

Whelan, J., and Millar, A.H. (2004). Experimental analysis of the

Arabidopsis mitochondrial proteome highlights signaling and regula-

tory components, provides assessment of targeting prediction pro-

grams, and indicates plant-specific mitochondrial proteins. Plant Cell

16, 241–256.

Hortensteiner, S. (2004). The loss of green color during chlorophyll

degradation—A prerequisite to prevent cell death? Planta 219,

191–194.

Ikeda, Y., Dabrowski, C., and Tanaka, K. (1983). Separation and

properties of five distinct acyl-CoA dehydrogenases from rat liver

Functional Analysis of ETFQO in Arabidopsis 2599

mitochondria. Identification of a new 2-methyl branched chain acyl-

CoA dehydrogenase. J. Biol. Chem. 258, 1066–1076.

Ikeda, Y., and Tanaka, K. (1983). Purification and characterization

of isovaleryl coenzyme A dehydrogenase from rat liver mitochondria.

J. Biol. Chem. 258, 1077–1085.

Karimi, M., Inze, D., and Depicker, A. (2002). GATEWAY vectors for

Agrobacterium-mediated plant transformation. Trends Plant Sci. 7,

193–195.

Kawaguchi, A., Yoshimura, T., and Okuda, S. (1981). A new method

for the preparation of acyl-CoA thioesters. J. Biochem. (Tokyo) 89,

337–339.

Lange, P.R., Eastmond, P.J., Madagan, K., and Graham, I.A. (2004).

An Arabidopsis mutant disrupted in valine catabolism is also com-

promised in peroxisomal fatty acid beta-oxidation. FEBS Lett. 571,

147–153.

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.

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.

Lin, J.F., and Wu, S.H. (2004). Molecular events in senescing Arabi-

dopsis leaves. Plant J. 39, 612–628.

Matile, P., Hortensteiner, S., and Thomas, H. (1999). Chlorophyll

degradation. Annu. Rev. Plant Physiol. Plant Mol. Biol. 50, 67–95.

Nakai, K., and Horton, P. (1999). PSORT: A program for detecting

sorting signals in proteins and predicting their subcellular localization.

Trends Biochem. Sci. 24, 34–36.

Ogier de Baulny, H., and Saudubray, J.M. (2002). Branched-chain

organic acidurias. Semin. Neonatol. 7, 65–74.

Oh, S.A., Lee, S.Y., Chung, I.K., Lee, C.H., and Nam, H.G. (1996). A

senescence-associated gene of Arabidopsis thaliana is distinctively

regulated during natural and artificially induced leaf senescence. Plant

Mol. Biol. 30, 739–754.

Parker, A., and Engel, P.C. (2000). Preliminary evidence for the

existence of specific functional assemblies between enzymes of the

beta-oxidation pathway and the respiratory chain. Biochem. J. 345,

429–435.

Porra, R.J., Thompson, W.A., and Kriedemann, P.E. (1989). Determi-

nation of accurate extinction coefficients and simultaneous equations

for assaying chlorophylls a and b extracted with four different sol-

vents: Verification of the concentration of chlorophyll standards by

atomic absorption spectroscopy. Biochim. Biophys. Acta 975, 384–394.

Roessner, U., Luedemann, A., Brust, D., Fiehn, O., Linke, T.,

Willmitzer, L., and Fernie, A. (2001). Metabolic profiling allows com-

prehensive phenotyping of genetically or environmentally modified

plant systems. Plant Cell 13, 11–29.

Roessner-Tunali, U., Hegemann, B., Lytovchenko, A., Carrari, F.,

Bruedigam, C., Granot, D., and Fernie, A.R. (2003). Metabolic

profiling of transgenic tomato plants overexpressing hexokinase

reveals that the influence of hexose phosphorylation diminishes

during fruit development. Plant Physiol. 133, 84–99.

Rontani, J.F., and Aubert, C. (2005). Characterization of isomeric allylic

diols resulting from chlorophyll phytyl side-chain photo- and autoxi-

dation by electron ionization gas chromatography/mass spectrome-

try. Rapid Commun. Mass Spectrom. 19, 637–646.

Rontani, J.F., Cuny, P., and Grossi, V. (1996). Photodegradation of

chlorophyll phytyl chain in senescent leaves of higher plants. Phyto-

chemistry 42, 347–351.

Ruzicka, F.J., and Beinert, H. (1977). A new iron-sulfur flavoprotein of

the respiratory chain. A component of the fatty acid beta oxidation

pathway. J. Biol. Chem. 252, 8440–8445.

Schuster, J., and Binder, S. (2005). The mitochondrial branched-chain

aminotransferase (AtBCAT-1) is capable to initiate degradation of

leucine, isoleucine and valine in almost all tissues in Arabidopsis

thaliana. Plant Mol. Biol. 57, 241–254.

Sessions, A., et al. (2002). A high-throughput Arabidopsis reverse

genetics system. Plant Cell 14, 2985–2994.

Skinner, D.J., Baker, S.C., Meister, R.J., Broadhvest, J., Schneitz,

K., and Gasser, C.S. (2001). The Arabidopsis HUELLENLOS gene,

which is essential for normal ovule development, encodes a mito-

chondrial ribosomal protein. Plant Cell 13, 2719–2730.

Sundaresan, V., Springer, P., Volpe, T., Haward, S., Jones, J.D.,

Dean, C., Ma, H., and Martienssen, R. (1995). Patterns of gene

action in plant development revealed by enhancer trap and gene trap

transposable elements. Genes Dev. 9, 1797–1810.

Taylor, D.C., Weber, N., Hogge, L.R., and Underhill, E.W. (1990). A

simple enzymatic method for the preparation of radiolabeled erucoyl-

CoA and other long-chain fatty acyl-CoAs and their characterization

by mass spectrometry. Anal. Biochem. 184, 311–316.

Taylor, N.L., Heazlewood, J.L., Day, D.A., and Millar, A.H. (2004).

Lipoic acid-dependent oxidative catabolism of alpha-keto acids in

mitochondria provides evidence for branched-chain amino acid

catabolism in Arabidopsis. Plant Physiol. 134, 838–848.

Thimm, O., Blasing, O., Gibon, Y., Nagel, A., Meyer, S., Kruger, P.,

Selbig, J., Muller, L.A., Rhee, S.Y., and Stitt, M. (2004). MAPMAN:

A user-driven tool to display genomics data sets onto diagrams

of metabolic pathways and other biological processes. Plant J. 37,

914–939.

Tonon, T., Qing, R., Harvey, D., Li, Y., Larson, T.R., and Graham, I.A.

(2005). Identification of a long-chain polyunsaturated fatty acid acyl-

coenzyme A synthetase from the diatom Thalassiosira pseudonana.

Plant Physiol. 138, 402–408.

Tsuchiya, T., Ohta, H., Okawa, K., Iwamatsu, A., Shimada, H.,

Masuda, T., and Takamiya, K. (1999). Cloning of chlorophyllase,

the key enzyme in chlorophyll degradation: Finding of a lipase motif

and the induction by methyl jasmonate. Proc. Natl. Acad. Sci. USA 96,

15362–15367.

Wanders, R.J., Jansen, G.A., and Lloyd, M.D. (2003). Phytanic acid

alpha-oxidation, new insights into an old problem: A review. Biochim.

Biophys. Acta 1631, 119–135.

Wierzbicki, A.S., Lloyd, M.D., Schofield, C.J., Feher, M.D., and

Gibberd, F.B. (2002). Refsum’s disease: A peroxisomal dis-

order affecting phytanic acid alpha-oxidation. J. Neurochem. 80,

727–735.

Zimmermann, P., Hirsch-Hoffmann, M., Hennig, L., and Gruissem,

W. (2004). GENEVESTIGATOR. Arabidopsis microarray database and

analysis toolbox. Plant Physiol. 136, 2621–2632.

2600 The Plant Cell

DOI 10.1105/tpc.105.035162; originally published online July 29, 2005; 2005;17;2587-2600Plant Cell

J. LeaverKimitsune Ishizaki, Tony R. Larson, Nicolas Schauer, Alisdair R. Fernie, Ian A. Graham and Christopher

during Dark-Induced Starvation Electron-Transfer Flavoprotein:Ubiquinone OxidoreductaseArabidopsisThe Critical Role of

 This information is current as of January 25, 2020

 

Supplemental Data /content/suppl/2005/07/27/tpc.105.035162.DC1.html

References /content/17/9/2587.full.html#ref-list-1

This article cites 54 articles, 24 of which can be accessed free at:

Permissions https://www.copyright.com/ccc/openurl.do?sid=pd_hw1532298X&issn=1532298X&WT.mc_id=pd_hw1532298X

eTOCs http://www.plantcell.org/cgi/alerts/ctmain

Sign up for eTOCs at:

CiteTrack Alerts http://www.plantcell.org/cgi/alerts/ctmain

Sign up for CiteTrack Alerts at:

Subscription Information http://www.aspb.org/publications/subscriptions.cfm

is available at:Plant Physiology and The Plant CellSubscription Information for

ADVANCING THE SCIENCE OF PLANT BIOLOGY © American Society of Plant Biologists