Formate dehydrogenase in
Arabidopsis thaliana
: characterization
and possible targeting to the chloroplast
Bradley J.S.C. Olson
a, Maryanne Skavdahl
a, Ha˚kon Ramberg
b, John C. Osterman
b,
John Markwell
a,*
aDepartment of Biochemistry,Uni6ersity of Nebraska,Lincoln,NE68588-0664,USA bSchool of Biological Sciences,Uni6ersity of Nebraska,Lincoln,NE68588-0118,USA
Received 25 April 2000; received in revised form 28 June 2000; accepted 29 June 2000
Abstract
Formate dehydrogenase (E.C. 1.2.1.2) is a mitochondrial-localized NAD-requiring enzyme in green plants. The enzyme activity and corresponding mRNA in leaves ofArabidopsis thalianaare induced by treatment with one-carbon metabolites. The cDNA for theArabidopsisformate dehydrogenase is similar to that of other plants except for the N-terminal region, which is predicted to target chloroplasts as well as mitochondria. The specific of activity of the enzyme in isolated chloroplasts suggests it is targeted to both mitochondria and chloroplasts inArabidopsis. Formate dehydrogenase from Arabidopsis was partially purified andKm
values for formate and NAD+were determined to be 10 mM and 65mM, respectively; theK
ifor NADH was 17mM. We conclude
that formate dehydrogenase is normally present inArabidopsischloroplasts and that sensitivity to inhibition by NADH may play a role in whether cellular formate is assimilated or dissimilated. © 2000 Elsevier Science Ireland Ltd. All rights reserved.
Keywords:Formate; Formate dehydrogenase;Arabidopsis thaliana; Mitochondria; Chloroplast
www.elsevier.com/locate/plantsci
1. Introduction
The NAD-linked formate dehydrogenase (FDH; EC 1.2.1.2) represents a family of enzymes found in methylotrophic bacteria and yeast, as well as higher plants [1]. In methylotrophic microorgan-isms, this enzyme plays an essential catalytic role in the final step of one-carbon metabolic oxidation and the generation of reducing equivalents. In these organisms, FDH is inducible by growth on one-carbon molecules. FDH activity is normally under metabolic control; when sufficient reducing equivalents are available to the cell, the activity is inhibited to promote assimilation of carbon from the cellular pool of formaldehyde [2]. Under con-ditions in which more NADH is required to fuel respiration, the enzyme is active and catalyzes
dissimilation by the oxidation of formate to CO2.
FDH in plants is situated at the periphery of one-carbon metabolism rather than in the main-stream of energy generation. Plant FDH is local-ized in mitochondria [3] and has a homodimeric quaternary structure, with subunits :42 000 in
molecular mass. The primary sequence is related to that of the FDH from methylotrophic yeast and bacteria [4].
When leaves are incubated in the dark, most exogenously supplied formate is dissimilated to CO2 [5]. However, FDH is not locked into a
dissimilatory role. It has long been known that [14C]-formate is readily assimilated into
metabo-lites, such as serine in plants incubated in the light [6,7]. It has been proposed that FDH regulates the concentration of one-carbon metabolites, deter-mining whether they are assimilated or dissimi-lated in plant tissues [3,8]. While the presence of formate in plant cells has been proposed to result from the oxidation of glyoxylate [9] or as part of a
* Corresponding author. Tel: +1-402-4722924; fax: + 1-402-4727842.
E-mail address:[email protected] (J. Markwell).
stress-signaling pathway [10], formate has also been reported to be a product of CO2 reduction
during some conditions of photosynthesis [7,11]. In barley (Hordeum 6ulgare L.), root FDH mRNA is induced by conditions of iron deficiency or anaerobiosis [4]. In potato (Solanum tuberosum
L.), leaf FDH mRNA is induced by hypoxia, chilling, drought, mechanical wounding and pro-longed darkness [12]. The latter reference also reported that foliar FDH mRNA was induced by spraying with 10 mM formate or 20% methanol within 24 h of treatment.
The current report examines the degree to which the FDH enzymatic activity is induced in Ara
-bidopsis thaliana, a model plant for biochemical and genetic studies, by foliar spraying with formal-dehyde as well as with either methanol or formate. These data indicate that FDH is induced by treat-ment with all three one-carbon metabolites. We also cloned and sequenced the cDNA for the
Arabidopsis FDH and examined the kinetic char-acteristics of the enzyme to assess whether it may act as a redox-sensing switch that controls whether formate is assimilated or dissimilated.
2. Methods
2.1. Materials
A. thaliana (Columbia ecotype) plants were grown in a controlled environment chamber at 20°C, 50% relative humidity, a 16 h photoperiod and a photon (400 – 700 nm) fluence rate of :300 mmol s−1m−2. Soil was subjected to steam
steril-ization prior to use. In addition to regular water-ing, plants were treated weekly with a nutrient solution containing 200 mg g−1 nitrogen.
2.2. Plant treatments
Plants were treated with a hand-held sprayer and the indicated solution was applied until the leaf surface was wet. Experiments demonstrated no difference between untreated plants and plants sprayed with water. Methanol and formate were used for treatments at concentrations of 20% (v/v) and 10 mM, respectively, based on their use at this concentration in previous studies [12]. Formalde-hyde was also used at a concentration of 10 mM. At no time did these concentrations appear to cause damage or stress to the plants.
2.3. Enzyme assays
For in vitro assays, leaf tissue was homogenized in a Tenbroeck tissue grinder with a chilled solu-tion containing 50 mM Tris – Cl (pH 8.6), 10 mM NaCl, 1 mM MgSO4, 2% (w/v)
polyvinyl-pyrrolidone and 0.1% (v/v) 2-mercaptoethanol. The extract was filtered through Miracloth (Cal-biochem) and centrifuged for 10 min at 40 000×g
to remove cell debris and insoluble material. The supernatant fluid was fractionated by chromatog-raphy through a column of Sephadex G-25 equili-brated with 50 mM Tris – Cl (pH 8.6), 10 mM NaCl and 1 mM MgSO4.
Formate dehydrogenase activity was measured spectrophotometrically at 340 nm and 30°C in a volume of 1 ml containing 100 mM potassium phosphate (pH 7.0), 50 mM sodium formate and 1 mM NAD+. The response of activity to the
con-centration of both formate and NAD+was
hyper-bolic and followed the expected Michaelis – Menten kinetics. To estimate the kinetic constants, the data were analyzed by a Linweaver – Burk plot. To estimate the Ki for NADH, the concentration
of that metabolite was held constant while the concentration of NAD+ was varied, which
pro-duced a classical pattern of competitive inhibition. Formaldehyde dehydrogenase (EC 1.2.1.1) activity was measured spectrophotometrically at 340 nm and 30°C in a volume of 1 ml containing 50 mM Tris – Cl (pH 8.6), 2 mM reduced glutathione, 1 mM MgSO4, 10 mM NaCl, 1.3 mM formaldehyde
and 1 mM NAD+. For dehydrogenase enzymatic
rates to be linear with time and amount of extract added, it was necessary to use extracts chro-matographed on Sephadex G-25. The protein con-centration in the extracts was determined using a dye-binding assay [13] with bovine serum albumin as a standard protein.
2.4. FDH purification
For purification of the Arabidopsis FDH, leaf tissue was ground in 50 mM NaPO4 (pH 7.0)
1.62 and 453 nmol min−1mg−1of protein,
respec-tively. We presume that the FDH was not purified to homogeneity since its specific activity was sig-nificantly lower than the enzyme isolated from pea leaves by a multi-step purification protocol [15].
2.5. FDH cDNA
The potato FDH sequence (Accession No. Z21493) was used to screen theArabidopsisdbEST using BLAST [16]. Several ESTs were detected and one was selected (Clone 173L24) with homol-ogy to the N-terminal half of the potato sequence to ensure identification of a full-length cDNA.
2.6. RNA preparation and blot analysis
Arabidopsis leaf tissue was rapidly frozen in liquid nitrogen. Total RNA was extracted from 100 mg of frozen tissue using a QIAGEN (Chatsworth, CA) RNeasy kit. For each experi-mental treatment, triplicate 4-mg samples of total RNA were fractionated by electrophoresis on a formaldehyde gel [17] and blotted onto Hybond N+ membrane using an Ambion (Austin, TX)
NorthernMAX-Plus kit. The membranes were subsequently UV cross-linked. Transfer efficiency was assessed by UV shadowing of the membrane. An identical set of samples along with a known RNA standard (Sigma, St. Louis, MO) was frac-tionated on an identical formaldehyde gel. RNA was visualized by staining with ethidium bromide. Relative abundance of 25S and 18S rRNA bands was quantitated using the public domain NIH Image program to permit normalization of the amounts of RNA loaded.
The full-length cDNA forArabidopsisFDH was randomly labeled with [a-32P]dATP. The labeled
probe (2×106 dpm ml−1) was hybridized
overnight at 42°C to the membrane in ZIP-Hyb hybridization solution included with the Ambion kit. The membranes were washed twice with low-stringency solution at 25°C and then washed twice with high stringency wash solution at 50°C. Hy-bridization was detected by exposing the mem-branes to a phosphor screen (Molecular Dynamics, Sunnyvale, CA). Imaging of the signal was acquired with a computer-controlled Molecu-lar Dynamics Storm 860 at 50 micron resolution. Quantitation of the hybridization signals was done with Image QuaNT 5.0 build 050 (Molecular Dy-namics). The intensity of the signal for each band was normalized to the ethidium bromide fluores-cence signals as described above.
2.7. Chloroplast isolation
Intact chloroplasts were isolated on gradients of Percoll and twice washed as previously described [18].
3. Results
Initial studies confirmed the previous report [12] that FDH was induced by wounding. Sampling of the leaves with a leaf punch or by cutting with a razor blade caused as much FDH induction as any of the other treatments. As a consequence, induc-tion experiments were conducted such that no plant was sampled more than once. The data in Table 1 report the effect of spraying plants with either 20% (v/v) methanol, 10 mM formaldehyde or 10 mM sodium formate. This experiment was repeated three more times and FDH specific activ-ity increases following treatment were consistently 2.1- to 2.8-fold. These data are consistent with the previous report [12] that methanol and formate
Table 1
Effect of foliar application of one-carbon metabolites on induction of leaf formate dehydrogenase and formaldehyde dehydroge-nase specific activities
Formate dehydrogenase Formaldehyde dehydrogenase (nmol min−1mg−1of protein)
(nmol min−1mg−1of protein)
Treatment (h) 6 24 48 6 24 48
1.5 17.5 20.1
Untreated 1.4 1.7 22.6
1.9 2.4 4.2
Methanol 17.3 18.1 15.6
17.8 22.1
19.3 3.0
1.7 2.8
Formaldehyde
1.6 3.6 4.8 18.2 19.1
Fig. 1. Induction of FDH mRNA by foliar application of methanol. Plants were either treated with 20% methanol (M) or untreated (−), but otherwise treated in an identical man-ner. Total RNA was prepared from leaves harvested at the indicated time following treatment, subjected to formaldehyde agarose electrophoretic fractionation, blotted onto a nylon membrane and hybridized with a homologous FDH cDNA.
The sequence of a cDNA clone for the FDH from A. thaliana was analyzed. Clone 173L24 was from the lPRL2 cDNA library. This clone was sequenced (Accession No. AF217195) and found to be 1480 bp with an open reading frame encod-ing 384 amino acids. The predicted molecular weight was 42 263. The predicted translation product exhibited :83% identity and 86%
simi-larity to the amino acid sequences for potato, rice and barley. Most of the dissimilarity was located at the N-terminus of the protein (Fig. 2). This is the region that specifies the mitochondrial target-ing signal for the potato, rice and barley proteins. Computer analysis using PSORT [20] predicts that a mitochondrial targeting sequence exists in the first 27 amino acids. This analysis also predicts a chloroplast localization signal. The program ChloroP [21] predicts the same signal with a cleav-age site after amino acid 29. Such dual targeting in
Arabidopsis has been reported with ferrochelatase-I [22] as well as the methionyl- and histidyl-tRNA synthetases [23 – 25]. The N-terminal sequences from potato, rice and barley are not identified as chloroplast-like by either the ChloroP or PSORT analyses.
Another group has recently added two addi-tional Arabidopsis sequences (AF208028 and AF208029; actually the same sequence with differ-ent polyadenylation sites) which are iddiffer-entical to the nucleotide sequence for the coding region of the sequence that we report. A recently available genomic sequence (AP002468) is also identical in the coding regions. These data corroborate the sequence we reported and are consistent with the presence of a single FDH sequence in Arabidopsis. Southern blots of Arabidopsis genomic DNA also suggest a single FDH gene (data not shown).
Isolated intact chloroplasts prepared from leaves ofArabidopsisplants using Percoll gradients treatments increase the FDH mRNA levels in
potato leaves. We additionally demonstrated that formaldehyde, a one-carbon metabolite with an oxidation state intermediate to methanol and for-mate, is also able to induce FDH activity in
Arabidopsis leaves.
We also examined the induction of FDH mRNA by treatment of Arabidopsis plants with methanol (Fig. 1). Leaves of treated plants had increased levels of FDH mRNA. The increase relative to nontreated plants under identical condi-tions was 3.2-fold at 6 h, 9.5-fold at 24 h and 1.5-fold at 48 h post-treatment.
In contrast to the induction of FDH by the three one-carbon metabolites, formaldehyde dehy-drogenase activity was not significantly induced by treatment of Arabidopsis plants with methanol, formaldehyde or formate (Table 1). This is consis-tent with the previous determination that this ac-tivity is expressed in a constitutive fashion [19]. Since the specific activity of formaldehyde dehy-drogenase is much higher than even the induced levels of FDH, the induction of formaldehyde dehydrogenase to higher than normal specific ac-tivities may not be an advantage for the plants.
Fig. 3. Immunoblot of proteins fractionated from leaf extract or isolated chloroplasts by SDS-PAGE. The monoclonal anti-body used for detection was prepared against thea-subunit of the maize mitochondrial ATPase. The estimated molecular weight from marker proteins is indicated on the left of the figure.
shown) do not reveal the presence of mitochon-drial contamination.
The Arabidopsis FDH was purified :280-fold
by affinity chromatography and subjected to ki-netic analyses. The Km for formate was 10 mM
and the Km for NAD+ was 65 mM. Because
NADH is a substrate for the FDH catalyzing the reverse reaction (reduction of CO2to formate), we
anticipated that this cofactor would function as a competitive inhibitor of the oxidation of formate. This was indeed the case and the Ki for NADH
was :17mM. To demonstrate the extent to which
NADH is able to inhibit the enzymatic oxidation of formate, the rate of formate oxidation was monitored at a variety of ratios of NAD+/
NADH, as demonstrated by the data points in Fig. 4. In this experiment, the total concentration of NAD+ and NADH was maintained at 1 mM,
which approximates the concentration of these metabolites within plant mitochondria [26]. The solid line in Fig. 4 represents the calculated activ-ity based on the experimentally determined Km
and Ki values.
4. Discussion
The three one-carbon metabolites, methanol, formaldehyde and formate, all appear to be effec-tive at the induction of FDH inArabidopsisleaves. The induction is consistently measurable and pro-longed, lasting at least 48 h. The increase in FDH mRNA in leaves ofArabidopsisplants treated with methanol is consistent with the change in enzyme specific activity. These results add to our knowl-edge from the previous report using potato [12] that did not examine the effect of formaldehyde or demonstrate increases in FDH specific activity.
Methanol is a major volatile emission product of leaves and seeds [27] and is, therefore, a molecule to which plants are normally exposed. However, rather than methanol having a direct effect on induction of formate dehydrogenase, we feel it more likely that oxidized metabolites, such as formaldehyde or formate, may serve as signal-ing molecules. In addition to caussignal-ing induction at much lower concentrations, both have been pro-posed as possible signaling molecules for regula-tion of plant metabolic processes [10,12,28]. Further research will be required to elucidate the cellular elements involved in the response of FDH
Fig. 4. Assessment of the ability of increasing concentrations of NADH to inhibit the FDH activity. The enzyme was assayed with varying ratios of NAD+/NADH, expressed as
the fraction NAD+. The total concentration of NAD+plus
NADH was maintained at 1 mM. The data points indicate experimental results and the error bars represent the S.E.M. of triplicate determinations. The solid line represents the expected inhibition based on competitive inhibition with aKm
for NAD+of 65mM and aK
ifor NADH of 17mM.
contain FDH specific activity (1.9 – 2.2 nmol min−1 mg−1 protein for four separate
experi-ments) approximately equal to cellular extracts, consistent with this enzyme being present in the chloroplast. In contrast, experiments with Percoll purified chloroplasts gave activities of B0.1 nmol
min−1 mg−1 protein for pea and the activity was
undetectable in chloroplasts from tobacco leaves. Western blots of Percoll purified Arabidopsis
to one-carbon metabolites and to determine if this is part of a more generalized plant stress response. The cDNA sequence for theArabidopsisFDH is similar to that reported for other NAD+
-depen-dent FDH activities in plants and methylotrophic microbes. Interestingly, the N-terminus of these proteins is the location of the greatest difference in amino acid sequence. The prediction from com-puter assisted analyses that the Arabidopsis FDH could potentially be targeted to the chloroplast as well as the mitochondrion was surprising. Such dual-targeting is not without precedent [22], but the N-terminal sequences from potato, rice and barley are not identified as chloroplast-like by either the ChloroP or PSORT analyses. It has been documented that FDH is abundant in Ara
-bidopsis mitochondria [29] and our data suggest a high FDH specific activity in Arabidopsis chloro-plasts isolated without significant mitochondrial contamination. Further research will be required to study whether the Arabidopsis FDH is targeted to the chloroplast as well as the mitochondrion under all conditions or if the localization is regu-lated as part of the stress response involving this activity.
A chloroplastic location for FDH in some plant species could be significant because the metabolic activity of this compartment is very different from the mitochondrion. This could result in a different set of regulatory cues, producing altered propor-tions of assimilation and dissimilation. As an addi-tional consideration, the oxidation of formate to CO2 by FDH is normally considered to be an
irreversible process, but is this physiologically true? The standard midpoint potentials for the NAD+/NADH and CO
2/formate redox half
reac-tions are −0.32 and −0.42 V, respectively. From these midpoint potentials it can be calculated that the DG°% for the reduction of CO2 to formate by
FDH is +19 kJ mol−1. However, a chloroplastic
location for the FDH may provide a ready source of higher energy electrons than normally possible in the mitochondrion and facilitate the reduction reaction. Such a situation might explain how some conditions of photosynthesis can produce signifi-cant amounts of formate as primary products [7,11]. The reduction of CO2 to formate may be
facilitated by formate-utilizing reactions, which re-sult in very low cellular concentrations of formate [30]. It has recently been demonstrated in vitro that FDH can efficiently reduce CO2 if the
for-mate generated is removed by other enzymatic reactions [31].
The Km values determined with the Arabidopsis
FDH for formate (10 mM) and NAD+ (65 mM)
are higher than reported for the enzyme from other plant species. The Michaelis constants for formate and NAD+ have been reported to be 0.6
mM and 6 mM for soybean [14] and 2 mM and 22
mM for pea [15]. Since NAD+and NADH bind to
the same site in the enzyme, NADH is a competi-tive inhibitor for the reaction oxidizing formate. The Km of the FDH for NADH, presumed to be
equivalent to the Ki value (17 mM), is lower than
the Km for NAD+. These values are similar to
those reported for the mitochondrial glycine de-carboxylase activity [32].
It seems likely that the leaf FDH activity is subject to metabolic regulation, since assimilation of formaldehyde and formate into other organic compounds versus dissimilatory conversion to CO2, is markedly higher in the light than in the
dark [5,6]. We were able to demonstrate that increasing concentrations of NADH could cause inhibition of the ability of FDH to oxidize formate to CO2. In intermediary metabolism, the
competi-tion between pyridine nucleotide cofactors has been minimized by evolution of metabolically sen-sible affinities for NAD+ and NADH [33]. The
mitochondrial ratio of NAD+/NADH in plant cell
mitochondria isolated by nonaqueous fractiona-tion is reported to be \100 [26], but estimates
based on 31P NMR studies suggest much lower
ratios [34]. The FDH NADH Ki of 15 mM is
similar to the values found for plant mitochondrial glycine decarboxylase [32], isocitrate dehydroge-nase and a-ketoglutarate dehydrogenase [35], en-zymes with activities regulated by changes in mitochondrial NAD+/NADH ratios and
catalyz-ing bidirectional, rather than unidirectional, reactions.
Acknowledgements
References
[1] V.O. Popov, V.S. Lamzin, NAD+-dependent formate
dehydrogenase, Biochem. J. 301 (1994) 625 – 643. [2] R.S. Hanson, T.E. Hanson, Methanotrophic bacteria,
Microbiol. Rev. 60 (1996) 439 – 471.
[3] D.J. Oliver, Formate oxidation and oxygen reduction by leaf mitochondria, Plant Physiol. 68 (1981) 703 – 705. [4] K. Suzuki, R. Itai, K. Suzuki, H. Nakanishi, N.-K.
Nishizawa, E. Yoshimura, S. Mori, Formate dehydroge-nase, an enzyme of anaerobic metabolism, is induced by iron deficiency in barley roots, Plant Physiol. 116 (1998) 725 – 732.
[5] B. Halliwell, Oxidation of formate by peroxisomes and mitochondria from spinach leaves, Biochem. J. 138 (1974) 77 – 85.
[6] A.R. Krall, N.E. Tolbert, A comparison of the light dependent metabolism of carbon monoxide by barley leaves with that of formaldehyde, formate and carbon dioxide, Plant Physiol. 32 (1957) 321 – 326.
[7] S.S. Kent, Photosynthesis in the higher plant Vicia faba. II. The non-Calvin cycle origin of acetate and its metabolic relationship to the photosynthetic origin of formate, J. Biol. Chem. 247 (1972) 7293 – 7302.
[8] C. Colas des Francs-Small, F. Ambard-Bretteville, I.D. Small, R. Remy, Identification of a major soluble protein in mitochondria from nonphotosynthetic tissues as NAD-dependent formate dehydrogenase, Plant Phys-iol. 102 (1993) 1171 – 1177.
[9] I. Zelitch, Control of plant productivity by regulation of photorespiration, Bioscience 42 (1992) 510 – 516. [10] E. Tyihak, S. Rozsnyay, E. Sardi, G. Gullner, L. Trezl,
R. Gaborjanyi, Possibility of formation of excited form-aldehyde and singlet oxygen in biotic and abiotic stress situations, Acta Biol. Hungary 45 (1994) 3 – 10. [11] N.K. Ramaswamy, A.G. Behere, P.M. Nair, A novel
pathway for the synthesis of solanidine in the isolated chloroplast from greening potatoes, Eur. J. Biochem. 67 (1976) 275 – 282.
[12] C. Hourton-Cabassa, F. Ambard-Bretteville, F. Moreau, J.D. de Virville, R. Remy, C.C. Francs-Small, Stress induction of mitochondrial formate dehydrogenase in potato leaves, Plant Physiol. 116 (1998) 627 – 635. [13] M. Bradford, A rapid and sensitive method for the
quantitation of microgram quantities of protein, utiliz-ing the principle of protein-dye bindutiliz-ing, Anal. Biochem. 72 (1976) 248 – 254.
[14] M.P. Farinelli, D.W. Fry, K.E. Richardson, Isolation, purification and partial characterization of formate de-hydrogenase from soybean seed, Plant Physiol. 73 (1983) 858 – 859.
[15] L. Uotila, M. Koivusalo, Purification of formaldehyde and formate dehydrogenase from pea seeds by affinity chromatography and S-formylglutathione as the inter-mediate of formaldehyde metabolism, Arch. Biochem. Biophys. 196 (1979) 33 – 45.
[16] S.F. Altschul, W. Gish, W. Miller, E.W. Myers, D.J. Lipman, Basic local alignment search tool, J. Mol. Biol. 215 (1990) 403 – 410.
[17] J. Sambrook, E.F. Fritsch, T. Maniatis, Molecular Cloning. A Laboratory Manual, Cold Spring Harbor Press, Cold Spring Harbor, 1989.
[18] J. Markwell, B.D. Bruce, K. Keegstra, Isolation of a carotenoid-containing sub-membrane particle from the chloroplastic envelope outer membrane of pea (Pisum sati6um), J. Biol. Chem. 267 (1992) 13933 – 13937. [19] R. Dolferus, J.C. Osterman, W.J. Peacock, E.S. Dennis,
Cloning of theArabidopsisand rice formaldehyde dehy-drogenase genes: implications for the origin of plant ADH genes, Genetics 146 (1997) 1131 – 1141.
[20] K. Nakai, M. Kanehisa, A knowledge base for predict-ing protein localization sites in eukaryotic cells, Genom-ics 14 (1992) 897 – 911.
[21] O. Emannuelsson, H. Nielsen, G. von Heijne, ChloroP, a neural network-based method for predicting chloro-plast transit peptides and their cleavage sites, Prot. Sci. 8 (1999) 978 – 984.
[22] K.-S. Chow, D.P. Singh, J.M. Roper, A.G. Smith, A single precursor protein for ferrocheletase-I from Ara
-bidopsisis imported in vitro into both chloroplasts and mitochondria, J. Biol. Chem. 272 (1997) 27565 – 27571. [23] B. Menard, L. Marechal-Dourard, W. Sakamoto, A.
Dietrich, H. Winz, A single gene of chloroplast origin codes for mitochondrial and chloroplastic methionyl-tRNA synthetase in Arabidopsis thaliana, Proc. Natl. Acad. Sci. USA 95 (1998) 11014 – 11014.
[24] A. Akashi, O. Grandjean, I. Small, Potential dual target-ing of anArabidopsisarchaebacterial-like histidyl-tRNA synthetase to mitochondria and chloroplasts, FEBS Lett. 431 (1998) 39 – 44.
[25] I. Small, H. Wintz, K. Akashi, H. Mireau, Two birds with one stone: genes that encode products targeted to two or more compartments, Plant Mol. Biol. 38 (1998) 265 – 277.
[26] S. Kro¨mer, H.W. Heldt, Respiration of pea leaf mito-chondria and redox transfer between the mitomito-chondrial and extramitochondrial compartment, Biochim. Bio-phys. Acta 1057 (1991) 42 – 50.
[27] M. Nemecek-Marshall, R.C. MacDonald, J.J. Franzen, C.L. Wojciechowski, R. Fall, Methanol emission from leaves. Enzymatic detection of gas-phase methanol and relation of methanol fluxes to stomatal conductance and leaf development, Plant Physiol. 108 (1995) 1359 – 1368. [28] A.U. Igamberdiev, N.V. Bykova, L.A. Kleczkowski, Origins and metabolism of formate in higher plants, Plant Physiol. Biochem. 37 (1999) 503 – 513.
[29] L. Jansch, V. Kruft, U.K. Schmitz, H.P. Braun, New insights into the composition, molecular mass and stoi-chiometry of the protein complexes of plant mitochon-dria, Plant J. 9 (1996) 357 – 368.
[30] V. Prabhu, K.B. Chatson, G.D. Abrams, J. King, 13C
nuclear magnetic resonance detection of interactions of serine hydroxymethyltransferase with C1-tetrahydrofo-late synthase and glycine decarboxylase complex activi-ties inArabidopsis, Plant Physiol. 112 (1996) 207 – 216. [31] R. Obert, B.C. Dave, Enzymatic conversion of carbon
[32] J. Bourguignon, M. Neuburger, R. Douce, Resolution and characterization of the glycine-cleavage reaction in pea leaf mitochondria, Biochem. J. 255 (1988) 169 – 178. [33] D.E. Atkinson, Cellular Energy Metabolism and its
Regulation, Academic Press, New York, 1977.
[34] J.K.M. Roberts, S. Aubert, E. Gout, R. Bligny, R. Douce, Cooperation and competition between adenylate
kinase, nucleoside diphosphokinase, electron transport, and ATP synthase in plant mitochondria studied by
31P-nuclear magnetic resonance, Plant Physiol. 113
(1997) 191 – 199.
[35] R. Douce, Mitochondria in Higher Plants. Structure, Function and Biogenesis, Academic Press, New York, 1985.
