sign in sign up
<<
Home , DNA polymerase, PFP (enzyme), Pfu DNA polymerase, Phosphofructokinase 1, Polymerase

View metadata, citation and similar papers at core.ac.uk brought to you by CORE

provided by Elsevier - Publisher Connector

FEBS Letters 581 (2007) 2401–2410

Characterisation of the ATP-dependent family from Arabidopsis thaliana

Angelika Mustropha,1, Uwe Sonnewaldb, Sophia Biemeltb,* a Humboldt-University Berlin, Institute of Biology, AG Plant Physiology, Philippstrasse 13, 10115 Berlin, Germany b Friedrich-Alexander University Erlangen-Nuremberg, Department of , Staudtstrasse 5, 91058 Erlangen, Germany

Received 1 March 2007; revised 15 April 2007; accepted 16 April 2007

Available online 30 April 2007

Edited by Mark Stitt

addition to the widely distributed ATP-dependent phospho- Abstract Plants possess two different types of phosphofructo- , an ATP-dependent (PFK) and a -depen- (PFK, EC 2.7.1.11), in plants and a number of dent form (PFP). While plant PFPs have been investigated in a second type of phosphofructokinase is present detail, cDNA clones coding for PFK have not been identified using pyrophosphate (PPi) instead of ATP as phosphoryl in Arabidopsis thaliana. Searching the A. thaliana re- donor (pyrophosphate--6-phosphate-phosphotransfer- vealed 11 putative members of a phosphofructokinase gene fam- ase, PFP, EC 2.7.1.90). Phosphorylation of Fru6P catalyzed by ily. Among those, four sequences showed high homology to the PFK is virtually irreversible in vivo, while PFP reacts near alpha- or beta-subunits of plant PFPs. Seven cDNAs resulted equilibrium and catalyzes the reaction in both directions. In in elevated PFK, but not PFP activity after transient expression addition a third, ADP-dependent phosphofructokinase has in tobacco leaves suggesting that they encode Arabidopsis PFKs. been reported which seems to be specific for certain groups RT-PCR revealed different tissue-specific expression of the indi- of archaebacteria [3]. vidual forms. Furthermore, analysis of GFP fusion indi- cated their presence in different sub-cellular compartments. Since its discovery in 1980s, plant PFP has been character- 2007 Federation of European Biochemical Societies. Published ized in detail in a range of plant species at the biochemical by Elsevier B.V. All rights reserved. and molecular level (for example, Solanum tuberosum [4]; Ricinus communis [5]; Citrus · paradisii [6]; Saccharum officina- Keywords: Phosphofructokinase; Agrobacterium infiltration; rum [7]). Additionally to plants, the was also found in Arabidopsis thaliana several microorganisms, for example in Amycolatopsis methan- olica [8] and Entamoeba histolytica [9]. PFP generally consists of two different subunits, alpha and beta, which usually form a heterotetramer (e.g. [10,11]). The enzyme activity is highly regulated by fructose-2,6-bisphosphate (Fru2,6BP) [1,12],in 1. Introduction contrast to animals, where PFK is regulated by Fru2,6BP [13]. Even though PFP is highly regulated, its function is not is a central which is present in clearly resolved. Transgenic potato plants with more than all organisms. It provides substrates to fuel energy production 90% reduction in PFP activity did not provide evidence for a and anabolic processes of living cells. Glycolysis is also of role of PFP in controlling the rate of glycolysis [14]. However, importance for adaptations to stress conditions such as nutri- PFP might be involved in adaptation of plants to non-optimal ent limitations, cold, drought or oxygen deficiency. The regu- conditions, such as phosphate deficiency or low oxygen stress lation of the glycolytic process is essential for all types. [15–17]. Key regulatory in plant glycolysis are pyruvate While PFP has been studied in detail, plant PFK has re- and phosphofructokinase [1]. ceived only little attention. Besides extensive studies of Musa Phosphofructokinase catalyzes the interconversion of fruc- cavendishii PFKs [18–22], the enzyme was analysed in Daucus tose-6-phosphate (Fru6P) and fructose-1.6-bisphosphate. carota, Cucumis sativus, Solanum tuberosum, Ricinus commu- Whereas most glycolytic enzymes are remarkably conserved nis, and Spinacia oleracea [23–32] (see Table 1 for details). Nei- between different organisms, various types of phosphofructo- ther subunit composition nor the size of the was similar kinases exist with a very complex evolutionary history [2].In in all plant species tested. One major problem in PFK isolation was the instability of the enzyme during purification [22,27]. However, as a consistent result, cytosolic and plastidic iso- forms were found in many plant species (Table 1), as known *Corresponding author. Fax: +49 9131 8528254. for several glycolytic enzymes [1]. Very recently, a PFK was E-mail address: [email protected] (S. Biemelt). isolated from spinach leaves, and sequenced [31]. This is the 1Present address: Department of Botany and Plant Sciences, Center for first and only report of a plant PFK protein sequence up to Plant Cell Biology, University of California, Riverside, CA 92521, now. USA. However, while this work was started, neither the sequence nor the amino acid composition of any plant PFK Abbreviations: Fru2,6BP, fructose-2,6-bisphosphate; G3PDH, was known. The importance of this enzyme for regulation of -3-phosphate-dehydrogenase; PFK, ATP-dependent phospho- fructokinase; PFP, pyrophosphate-fructose-6-phosphate-phospho- prompted us to search for PFK-encoding se- ; PPi, pyrophosphate; TPI, triosephosphate- quences in plants. Therefore we have chosen the model plant

0014-5793/$32.00 2007 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2007.04.060 2402 A. Mustroph et al. / FEBS Letters 581 (2007) 2401–2410

Table 1 Overview of approaches to purify ATP-dependent PFKs from different plant species Plant species Organ Characteristics of the purified protein(s) Reference Cytosolic form Plastidic form Size of the Presence and/or Size of the Presence and/or protein MW size (kDa) of SU protein MW size (kDa) of SU (kDa) (kDa) Cucumis sativus Seed 180 40, 42, 47 180 39, 41.5, 53 [23] Daucus carota Root 5000 60, 68 [24] Glycine max Nodules 200, 2000 33, 55, 59, 62 [25] Lycopersicon esculentum Fruit 180 35 [26] Musa acuminata Fruit 210 Unstable 160 Unstable [22] Ricinus communis Seed Present Unstable 220 57 [27] Ricinus communis Seed Unknown 61.5 [28] Solanum tuberosum Tuber 200–800 46.3, 49.5, 50, 53 [29] Spinacia oleracea Leaf Unknown 50–70 [30] Spinacia oleracea Leaf 200 52 [31] Triticum aestivum Seed 182 [32] Data were taken from the references indicated. MW, molecular weight; SU, subunit composition.

Arabidopsis thaliana. The gene coding for PFK in Anabaena sion-DNA- (Finnzymes, Finland; PFK2, 5, 6). The result- (strain PCC7120, Gene Entry: all7335) was used to find homo- ing PCR products were sub-cloned into the following vectors: PFK1, logues in the Arabidopsis genome. Thereby seven Ara- 3, 4: pCR 2.1 (Invitrogen, Karlsruhe, Germany), PFK7: pCR Blunt (Invitrogen); PFK2, 5, 6: pGEMT (Promega). bidopsis genes were identified which were essentially the The amplified products were sequenced to verify the sequence iden- same as previously predicted by phylogenetic analysis [31,33]. tity. The corresponding cDNAs were cloned and tested for their The fragments were excised using the Asp 718 and BamHI (PFK1–4, PFP- and PFK activity by transiently over-expressing them 6, 7) or Asp718 and XbaI restriction sites (PFK5) and inserted into a Bin19-derived binary vector [35] between the CaMV 35 S promoter in tobacco leaves. Thereby we could experimentally prove that and octopin synthase signal. these seven open reading frames code for PFK in A. thaliana. For localisation studies GFP was C-terminally fused to fragments In addition, tissue specific expression and sub-cellular localiza- encoding putative PFKs as well as to the proposed signal peptide of tion of the PFK isoforms were analysed. PFK4 (PFK4-SP). To this end, the cDNA clones were excised without the stop codon using the Asp 718 and SalI restriction sites of PFK1–4 and the Asp 718 and SpeI restrictions sites of PFK5–7, respectively. The fragments were then cloned into the pBinGFP vector [36] which 2. Materials and methods had been digested accordingly. The binary constructs were transformed into Agrobacterium tum- 2.1. Plant material and growth conditions efaciens, strain CV58C1, carrying the virulence plasmid pGV2260 Both Nicotiana tabacum (cv. Samsun NN) and Nicotiana benthami- [37], which were used for transient expression of the genes in tobacco ana were grown in a greenhouse with 16 h supplemental light (ca. plants as described below. 250 lmol quanta PAR À2 sÀ1) followed by 8 h darkness. The tempera- ture regime followed the day/night cycle with 25 C/18 C. Arabidopsis 2.4. Agrobacterium mediated transient gene expression in tobacco plants plants (cv. Columbia) were grown in a growth chamber under 9-h-illu- For transient transformation, the method described by Bendahmane mination at 22 C. et al. [38] was used. Briefly, Agrobacterium tumefaciens cells harbour- ing the putative PFK cDNAs were grown overnight in 50 ml YEB 2.2. Sequence alignments medium with kanamycin, rifampycin, ampicillin, 1 mM MES pH 5.6 PFK and PFP protein sequences were obtained from public dat- and 20 lM acetosyringone. Cells were harvested by centrifugation at abases (http://www.ncbi.nlm.nih.gov/entrez, http://www.arabidopsis. 3500 · g and 4 C for 20 min. The pellet was washed with sterile water org/). Putative phosphofructokinase sequences from Oryza sativa and to remove excess of the medium. After a second centrifugation step, Populus trichocarpa were obtained by BLAST search of the respective the pellet was re-suspended in infiltration buffer containing 10 mM databases with the Arabidopsis protein sequences (http://www.tigr.org/ MgCl2, 10 mM MES pH 5.6 and 0.1 mM acetosyringone, and adjusted tdb/e2k1/osa1/, http://genome.jgi-psf.org/Poptr1_1/). Sequence align- to a final concentration of OD600 of 1.0. After incubation for 3 h at ment of the selected 68 amino acid sequences (Suppl. 1) was made room temperature, the suspension was infiltrated at the lower side of by means of the ClustalW tool (http://clustalw.genome.jp) using the a tobacco leaf using a 1 ml blunt end tip syringe. default settings. For determination of enzyme activity the Agrobacteria were infil- trated into leaves of Nicotiana tabacum plants. Samples were taken two days after infiltration and immediately frozen in liquid . 2.3. Cloning of putative PFK cDNA sequences Nicotiana benthamiana leaves were infiltrated to carry out localisation RNA was isolated from whole, flowering Arabidopsis plants accord- studies of GFP fusion proteins. ing to Logemann et al. [34]. Subsequently, cDNA was prepared from 20 lg of total RNA. To this end total RNA was treated with 10 U of RNase–free DNase (Roche, Basel, Switzerland) for 45 min at 2.5. PFK and PFP activity measurements 37 C followed by heat inactivation for 10 min at 65 C. 2.5 lgof Leaf discs were ground to a fine powder in liquid nitrogen and ex- DNase-treated RNA were reverse transcribed for 60 min at 37 Cin tracted in 50 mM HEPES–KOH, pH 6.8 containing 5 mM Mg acetate, a25ll reaction mix containing 1 U of M-MLV [H-] reverse transcrip- 5mM b-mercaptoethanol, 15% (v/v) glycerine, 1 mM EDTA, 1 mM tase (Promega, Madison, WI, USA), 20 lM of each dNTP, 2 lM oligo EGTA, 5 mM DTT and 0.1 mM Pefabloc proteinase inhibitor (Boeh- dT[30]V[G/C/A]Primer, and 40 U of RNase inhibitor (Roche). ringer Mannheim, Germany). The homogenate was centrifuged at One microgram of cDNA was used for PCR amplification of the 13000 · g at 4 C for 15 min. The resulting supernatant was used for respective genes using specific primers given in Table 2 by means of spectrophotometric determination of the activities of the phosphofruc- either Taq-DNA-polymerase (Takara Shouzo, Japan PFK1, 3, 4), tokinase enzymes at 340 nm using a UVIKON photometer (Kontron, Pfu-DNA-polymerase (Stratagene, The Netherlands; PFK7), or Phu- Germany). A. Mustroph et al. / FEBS Letters 581 (2007) 2401–2410 2403

Table 2 Oligonucleotide primers used for cloning of ATP-PFKs from Arabidopsis thaliana Gene Accession number Primer sequence PFK1 At4g29220 Fw GGTACCATGTCATCTTCTGTTCCAAAC Rev GGATCCTCAGTCGACCCATTTCATCGTCCCTGGTTCACC

PFK2 At5g47810 Fw GGTACCATGGCCGCCGAAACTAGCATC Rev GGATCCTTAGTCGACACCCTTAACGTTCGTTTCAAAATC

PFK3 At4g26270 Fw GGTACCATGAGTACTGTGGAGAGTAGC Rev GGATCCTCAGTCGACCTTGGTGATCTCTTTGGTGACTGG

PFK4 At5g61580 Fw GGTACCATGGAAGCTTCGATTTCGTTTC Rev GGATCCTTAGTCGACGATAGAAGAGATCTTCATGTTATC

PFK4 SP Rev GTCGACTCTGATGAGTCGTTTCTGGTG

PFK5 At2g22480 Fw GGTACCATGGATGCTCTTTCTCAGGCG Rev TCTAGATTAACTAGTGATGAAATCGGGTTCGCCCGTTGA

PFK6 At4g32840 Fw GGTACCATGGCTTCTAATGGCGTGGAC rev GGATCCTTAACTAGTGTCGGTAAACTCCGTGGTTCCTTT

PFK7 At5g56630 Fw GGTACCATGTCTAGTCCGAGAAGTAAC Rev GGATCCTTAACTAGTCTTAGTGACCTCTTTAGTCACAGG

Actin X63603 Fw ATGGCAGACGGTGAGGATATTCA Rev GCCTTTGCAATCCACATCTGTTG

Tubulin At5g62690 Fw CTCAAGAGGTTCTCAGCAGTA Rev TCACCTTCTTCATCCGCAGTT Start- and stop codons were marked in bold; sequences of restriction enzymes added to facilitate cloning were underlined. Actin primers were taken from Romeis et al. [60], tubulin primers were used according to Fridman and Zamir [61].

The activity of PFK (EC 2.7.1.11) was assayed in a reaction buffer using standard protocols. The number of cycles was chosen separately containing 0.1 M HEPES–KOH, pH 7.9, 2 mM MgCl2, 0.15 mM for each primer pair (PFK1, 3, 7: 31; PFK2, 4-6: 35; actin, tubulin: 28). NADH, 7.5 mM Fru6P, 1 U aldolase, triosephosphate-isomerase Actin and tubulin were used as controls for equal amount of cDNA (TPI) and glycerol-3-phosphate-dehydrogenase (G3PDH). The reac- (Table 2). tion was started by adding ATP to a final concentration of 2.5 mM. For determination of PFP (EC 2.7.1.90) activity the described by Gibbs et al. [39] was slightly modified. As reaction buffer 0.1 M Hepes–KOH, pH 7.9 supplemented with 2.6 mM Mg-acetate, 3. Results and discussion 0.15 mM NADH, 7.5 mM Fru6P, 1 lM Fru2,6BP, 1 U aldolase, TPI and G3PDH, was used. The reaction was started by adding sodium 3.1. Bioinformatic approach to find putative PFK genes in pyrophosphate (1 mM final concentration) to the reaction mixture. Arabidopsis thaliana The protein content was measured according to Bradford [40] and cal- culated based on a calibration curve using BSA as standard. To find putative PFK encoding genes in Arabidopsis thaliana the PFK gene from Anabaena (strain PCC7120, Gene Entry: all7335; http://cyano.genome.ad.jp/) was used to search the 2.6. Isolation of protoplasts and detection of GFP Two days after Agro-infiltration, leaf discs of tobacco were har- Arabidopsis genome. Thereby, 11 PFK-related sequences were vested. Protoplasts were isolated according to Bayley et al. [41]. Briefly, identified that showed high homology: At1g12000, At1g20950, leaves were cut into small pieces, which were incubated in the dark for At1g76550, At2g22480, At4g04040, At4g26270, At4g29220, 4 h in 10 ml of K3AS medium (1· MS salts, 3 mM CaCl2, 0.4 M su- At4g32840, At5g47810, At5g56630, and At5g61580. crose, pH 5.8) containing 1% cellulase, 0.5% driselase and 0.2% mac- These sequences were aligned exploiting the ClustalW algo- roenzyme (Serva, Germany). Released protoplasts were collected in 15-ml tubes, and centrifuged at 200 · g for 20 min at 4 C omitting rithm together with all known plant-specific PFP sequences as the break after end of the centrifugation time. Intact protoplasts that well as genes encoding PFKs and PFPs from different microor- were floating on the surface were collected in a fresh tube. The proto- ganisms and animals. Furthermore, putative phosphofructoki- plast solution was diluted in 4 vols. of W5 medium (0.15 M NaCl, nase genes from rice and Populus trichocarpa were selected 0.13 M CaCl2, 5 mM KCl, 5 mM , pH 5.8), and protoplasts were pelleted by centrifugation for 5 min at 200 · g. The supernatant based on their sequence homology from public databases was removed, and the pellet was used for GFP analysis. (http://www.tigr.org/tdb/e2k1/osa1/, http://genome.jgi-psf.org/ GFP signals were detected with the confocal laser scanning micro- Poptr1_1/) and added to our analysis. The protein sequences scope CLSM 510 META (Zeiss, Jena, Germany). Excitation wavelength used for this alignment can be found in supplement 1. The was 488 nm, emission of GFP and chlorophyll autofluorescence were alignment revealed two different groups of plant phosphofruc- detected at 510–525 nm and at 645–700 nm, respectively. tokinase genes (Fig. 1), essentially as described by Nielsen et al. [33] and Winkler et al. [31]. 2.7. Semi-quantitative PCR analysis The plant PFP genes from potato, Ricinus communis and For semi-quantitative RT-PCR total RNA was extracted from dif- ferent Arabidopsis tissues as described by Logemann et al. [34]. cDNA Citrus x paradisii [4–6] form one separate cluster together with synthesis was performed as described above. PCR was performed sequences coding for PPi-using enzymes from microorganisms 2404 A. Mustroph et al. / FEBS Letters 581 (2007) 2401–2410

Group II, clade „X“ Group I, ATP-PFK

PFK_human_m PFK_Mus_m So_PFK1 PFK_Mus_p PFK_Trypanosoma Os10g26570 PFK_human_p PFK_human_l Plant PFKs Os09g24910 PFK_Amy- PFK_Mus_l Os08g34050 PFK_Entamoeba colatopsis PFK_Propionibacterium Ptr_PFK7 PFK_yeast_B At2g22480 Ptr_PFK4 PFK_yeast_A At4g29220 Os06g05860 Ptr_PFK5 PFK_Ecoli_A At5g56630 At4g26270 Ptr_PFK6 PFK_Bacillus So_PFK2 PFK_Anabaena At4g32840 Os05g10650 PFP_Amycolatopsis Os01g09570 Os05g44920 Os01g53680 At5g61580 Ptr_PFK2 Ptr_PFK3

Os09g30240 Ptr_PFPB1 Ptr_PFPB2 Os04g39420 At5g47810 PFPB_Ricinus Ptr_PFK1 PFPB_Citrus At1g12000 PFPB_Potato Os06g13810 At4g04040

PFP_Entamoeba At1g76550 PFP_Propionibacterium At1g20950 Ptr_PFPA1 Os09g12650 Ptr_PFPA2 Os08g25720 PFPA_Ricinus PFPA_Citrus PFK_Ecoli_B PFPA_Potato Os06g22060 Plant PFPs Os02g48360 Group II, clade „long“

Fig. 1. Searching for phosphofructokinase genes in the Arabidopsis genome. The unrooted Neighbour-joining tree, that was calculated by use of the ClustalW method at http://align.genome.jp/ using the default parameters, shows sequence similarity for 68 amino acid sequences of different plant, animal or bacterial ATP-dependent PFKs as well as pyrophosphate-dependent phosphofructokinase (PFP) a- (PFPA) and b-subunits (PFPB). Black circles in the plant PFK-group mark plastidal isoforms, as predicted by TargetP (http://www.cbs.dtu.dk/services/TargetP/), grey circles mark putative plastidal isomforms with lower certainty. The amino acid sequences and their accession numbers can be found in Supplement 1. which have been designated as phosphofructokinase-group II, the group I of ATP-dependent PFKs found in animals and clade ‘‘long’’ [2,42,43]. Additionally this group contains several [44], than to the plant PFP genes (Fig. 1). This indi- rice and poplar homologues and four Arabidopsis genes cates that the seven Arabidopsis genes as well as the corre- (At1g20950, At1g76550, At1g12000, At4g04040) indicating sponding rice and poplar genes might code for PFKs. that these sequences might code for PPi-dependent phospho- Besides forming distant phylogenetic groups, biochemical . As expected, all plant species included in our and structural analysis revealed several amino acids essential analysis harbour genes for the a- and b-subunit of PFP form- for function of PFK and PFP. Shirakihara and Evans [45] de- ing two distinct subgroups (Fig. 1). scribed the crystalline structure for the E. coli PFK suggesting The known PFKs from human, mouse, yeast and different putative ATP and Fru6P binding sites. Furthermore, Moore bacteria form another coherent cluster (Fig. 1 [31]). These se- et al. [46] found residues, which are involved in PPi binding quences have been previously designated as PFK group I con- of PFP. Especially one putative ATP , the Gly104 taining exclusively ATP-dependent PFKs [44]. A different from E. coli PFK (in the GGDG-motif) was shown to be group comprises seven Arabidopsis sequences (At2g22480, important for using ATP as co-. In contrast, an Asp At4g26270, At4g29220, At4g32840, At5g47810, At5g56630, at this position hinders ATP binding in PFP [2,46]. Accord- and At5g61580), the two PFKs recently identified in spinach ingly, mutations in the corresponding positions in E. histoly- [33] and a number of rice and poplar sequences. This group ex- tica PFP (Asp175) resulted in changing from a PPi-binding to hibit homology to PFKs from microorganisms such as PFKs an ATP-binding capacity of the enzyme [47]. In addition, from E. histolytica, T. brucei and A. methanolica [2,42,43]. Gly or Lys residues in position 124 (according to E. coli The latter were assigned as PFK group II, clade ‘‘X’’ [42]. PFK) seem to be characteristic for PFK or PFP [2], but muta- Based on sequence similarity this group is more related to tion of this position in E. histolytica PFP resulted only in a A. Mustroph et al. / FEBS Letters 581 (2007) 2401–2410 2405 slight change in PPi binding [47]. However, a comprehensive Frc6P binding sites (marked with open circles) as suggested analysis performed by these authors revealed a Gly104, Lys124 by Shirakihara and Evans [45] are found in all sequences, ami- signature for clade X ATP-dependent PFKs. no acids crucial for ATP binding including Gly104 and Lys124 To prove whether these amino acids are also present in the which are specific for clade X PFKs are conserved in the pro- Arabidopsis sequences of interest, an amino acid alignment posed Arabidopsis PFKs. In contrast, the two beta-forms of was done with all 11 putative phopshofructokinase sequences PFP, which represent catalytic subunits [15,48], have Asp of Arabidopsis together with PFK from E. coli and both PFP and Lys at the corresponding positions, respectively as charac- and PFK from A. methanolica (Fig. 2). Whereas putative teristic for PFP (Fig. 2).

Fig. 2. Alignment of amino acid sequences of different PFKs and PFPs. The amino acid sequences of putative PFK and PFP encoding genes from Arabidopsis together with PFK A from E. coli, PFK and PFP from Amycolatopsis methanolica were analysed. The last part of the alignment (about 100 amino acids) was omitted for better overview. Closed circles show putative ATP-binding sites, open circles show putative Fru6P-binding sites [45]. Black arrows show putative sites involved in PPi binding [46]. The most important amino acid residues determining the ATP/PPi substrate specificity, Gly/Asp104 and Gly/Lys124 [47] are marked with white arrows. Identical residues are shown with black background; similar residues are shown with grey background. The bars represent the phosphofructokinase domain IPB000023A, B and C (http://blocks.fhcrc.org). 2406 A. Mustroph et al. / FEBS Letters 581 (2007) 2401–2410

Table 3 Characteristics of putative ATP-dependent PFK genes from Arabidopsis thaliana Name Accession number Size of mRNA (bp) Size of protein (kDa) Predicted sub-cellular localisation Number of ESTs PFK1 At4g29220 1422 52.0 Cytosol 28 PFK2 At5g47810 1335 49.2 Cytosol 20 PFK3 At4g26270 1470 53.6 Cytosol 19 PFK4 At5g61580 1593 58.5 Chloroplast 10 PFK5 At2g22480 1614 58.6 Chloroplast 15 PFK6 At4g32840 1389 50.8 Cytosol 12 PFK7 At5g56630 1458 53.5 Cytosol 22 For easier handling genes were arbitrary assigned to PFK1 to PFK7. The size of protein was calculated by means of the EXPASY translation tool (http://expasy.org/). The sub-cellular localisation was predicted by analysing the sequences using the prediction programs TargetP and ChloroP [54,55]. Numbers of ESTs were obtained from the TAIR database (www.arabidopsis.org).

Based on the bioinformatic and phylogenetic analysis we in at least three independent infiltration experiments. In addi- therefore speculated that seven sequences (At2g22480, tion stable tobacco transformants expressing PFK3 or PFK4 At4g26270, At4g29220, At4g32840, At5g47810, At5g56630, also showed significantly higher PFK activities compared to and At5g61580) might encode PFKs in Arabidopsis thaliana. the wildtype (data not shown). Therefore, we conclude that Hence, they were chosen for further analysis. For easier han- the putative PFK genes code for ATP-dependent phospho- dling the genes were numbered consecutively (Table 3). fructokinases in Arabidopsis. PFK2 and PFK5 induced only a weak increase in PFK activity in our transient expression assays compared to the 3.2. Cloning of putative PFK cDNAs and determination of their other five PFK isoforms. One reason for their lower activity activity could be that these proteins were poorly expressed in tobacco For cloning of the putative PFK genes, mRNA was isolated and/or less stable. Interestingly, the protein sequences of PFK2 from whole Arabidopsis plants at flowering time, and cDNA and PFK5 possess lower sequence similarity to the five other was synthesized. The cDNAs were amplified by PCR using pri- AtPFKs and both belong to small subgroups of plant-specific mer combinations given in Table 2, and after sub-cloning frag- PFKs as indicated by the phylogenetic tree (Fig. 1). It could be ments were inserted into the pBinAR vector [35]. These also possible that these two genes are pseudogenes, however, constructs were transformed into A. tumefaciens, which were this is less likely since both are expressed in A. thaliana as indi- used for transient expression in tobacco leaves by means of cated by the multitude of ESTs (Table 3) as well as by publicly Agro-infiltration. available and our own expression data (Fig. 4). An additional Two days after infiltration of tobacco leaves, samples were hint that members of these two PFK subgroups might encode taken and the activities of PFK and PFP were measured. active PFKs, comes from the high homology of AtPFK5 to For 5 of the 7 constructs (PFK1, 3, 4, 6, 7) a 3–5-fold higher SoPFK1, an active PFK isolated from spinach (Fig. 1 [31]). PFK activity was found in comparison to non-infiltrated Another possibility for their lower activity may be that these leaves or to the empty vector control (Fig. 3). For PFK2 and proteins are regulatory subunits or that they are under meta- 5, we found only a slightly higher PFK activity compared to bolic control. the controls. PFP activity was similar in leaves infiltrated with Previous observations suggested that plant PFKs are, unlike either of the constructs (Fig. 3). Similar results were obtained animal PFKs, insensitive to Fru2,6BP [24,49,50]. In order to test, whether or not PFKs investigated, particularly PFK2 and 5, are subject of by Fru2,6BP, increas-

] ing amounts of Fru2,6BP were added to the assay mix. While -1 250 addition of 1–25 lM Fru2,6BP to the PFP assay resulted in a PFK

* min significantly increased PFP activity, there was no change in 200 -1 PFP PFK activity in any of the seven isoforms (data not shown). Our data demonstrate that expression of a single PFK gene 150 from Arabidopsis in tobacco leaves is sufficient to induce higher PFK activity. Exceptions were found for PFK2 and PFK5 100 which caused only a slightly higher PFK activity in transient assays (Fig. 3) and might have other properties. However, 50 neither the formation of PFK oligomers nor the subunit com-

0 position is clear in plants (summarized in Table 1) even activity [nmol * mg protein N.t.WT PFK1 PFK2 PFK3 PFK4 PFK5 PFK6 PFK7 buffer empty though, these enzymes tend to form aggregates [24] and differ- vec t or ent subunits could be isolated from cucumber, soybean and Fig. 3. ATP-PFK and PFP activity of putative PFKs from Arabidopsis potato [23,25,29]. In contrast to that, an oligomer formation thaliana after transient expression in tobacco leaves. Two days after has been demonstrated for non-plant-PFKs. For instance, infiltration of tobacco leaves (Nicotiana tabacum) with Agrobacterium the PFK from E. coli was described to form a homotetramer tumefaciens strains carrying the binary constructs of genes most likely [45], while the human PFK assembles to tetramers consisting encoding for ATP-PFK enzyme extracts were prepared and the activities of PFK and PFP were measured. Data are means ± S.D. of of several subunits [51]. Thus, oligomer formation of different four samples. Similar results were obtained in independent Agro- plant PFK subunits remains to be elucidated in future experi- infiltration experiments. ments. A. Mustroph et al. / FEBS Letters 581 (2007) 2401–2410 2407

RyLmLFlSt to be induced following heat stress, whereas expression of both PFK2 and 4 seem to be activated by osmotic stress. This let us to suggest, that the different AtPFKs isoforms PFK 1 may have different functions in Arabidopsis during develop- ment and in adaptation towards changing environment. PFK 2 T-DNA or RNA interference lines might help to elucidate the function of the individual PFK isoforms.

PFK 3 3.4. Localisation of PFK isoforms Previous experiments proposed cytosolic and plastidal iso- forms of PFKs (summarized in Table 1). Therefore, we ana- PFK 4 lysed the localization of the phosphofructokinase proteins. Targeting predictions programs as ChloroP and TargetP [54,55] were exploited to analyse the sequences. Thereby a PFK 5 plastidal compartimentation was predicted for two of the seven PFK genes (At5g61580, At2g22480; Table 3). In accordance PFK 6 with this prediction, both sequences which were assigned to PFK4 and 5, respectively, carry an N-terminal extension com- pared to the other 5 sequences (Fig. 2). The mRNAs of these PFK 7 putative plastidal isoforms contained about 1600 bp encoding predicted proteins of 530 or 537 amino acids for PFK4 or PFK5 (Table 3), whereas the mRNAs of the other five se- Actin quences are about 1400 bp long. TargetP identified potential 54 and 52 residues containing transit peptides for PFK4 and PFK5, respectively. There are also several plastidal isoforms Tubulin predicted among the putative PFK encoding sequences from rice and poplar. Interestingly, the subgroup of plant PFKs, Fig. 4. Expression analysis of AtPFKs. RT-PCR-analysis was per- containing the AtPFK5 (At2g22480), comprises exclusively formed for the seven PFK genes using different tissues of Arabidopsis plastidal members. The only exception is SoPFK1 where no thaliana as template. Actin and tubulin were used to control that equal transit peptide is recognized by these programs. However, amounts of cDNA were used for PCR reactions. One representative result out of four independently performed experiments is shown. R, plastidal localization was proposed by Winkler et al. [31]. root; yL, young leaf; mL, mature leaf; St, stem; Fl, flower. For the remaining 5 Arabidopsis protein sequences no special localisation signal or signature could be recognised suggesting that they might be cytosolic. 3.3. Tissue specific expression of AtPFK In order to prove the subcellular localisation experimentally, Our transient assays revealed the induction of high PFK the complete Arabidopsis sequences without stop codon were activity for 5 of 7 putative ATP-PFK of Arabidopsis. In order cloned into pbinGFP vectors [36], and these constructs were to analyse tissue-specific expression of PFK isoforms, semi- transiently expressed in Nicotiana benthamiana. Two days after quantitative RT-PCR analysis was performed. As shown in Agro-infiltration of tobacco leaves, protoplasts were obtained Fig. 4, all PFKs were found to be expressed. PFK1 was highly and GFP fluorescence was analysed using confocal laser scan- expressed in all tissues, whereas PFK4 was present in all green ning microscopy (Fig. 5). PFKs 2, 3, 6 and 7 showed clear cyto- tissues only. Similar to PFK1, PFK5 showed consistent expres- solic localization, as suggested by the bioinformatic analysis sion in all tissues of Arabidopsis investigated, despite that only (Fig. 5B, C, G, H). Interestingly, nuclear localisation has been a weak induction of PFK activity could be shown in tran- described for some glycolytic enzymes [56]. No obvious nuclear siently transformed tobacco leaves (Fig. 3). PFK7 showed localisation signal could be detected in any of the AtPFKs by in highest expression in roots, while PFK6 was highly expressed silico analyses using the PSORT tool (http://www.psort.org/, in flowers. While no transcript accumulation could be detected [57]), and there was no clear cell biological evidence, that one for PFK3 in stems after 31 PCR cycles, it was found to be PFK isoform is specifically targeted to the nucleus. strongly expressed in roots and flowers. PFK2, which was PFK4 and PFK5 were found to be localised at the exterior one of the less active PFKs, showed very low expression in of chloroplasts (Fig. 5D and F). Here, the entire sequences most tissues (Fig. 4) indicated by very weak signals even after of PFK4 or PFK5 were fused to GFP leading to the deposition 35 PCR cycles. However, the analysis of microarrays revealed of small, GFP expressing particles at the outside of chloro- PFK2 transcript accumulation especially in ripe seeds plasts (Fig. 5E and F). This could be due to the strong expres- (www.csbdb.mpimp-golm.mpg.de [52]; www.genevestigator. sion in the transient system. Another possibility is that the ethz.ch [53]), which could be a hint for the involvement of fusion proteins are not efficiently targeted. Therefore, only PFK2 in seed maturation and/or germination. the N-terminal first 54 amino acid containing the predicted Comparing our results with data from gene expression transit peptide of PFK4 were cloned in front of the GFP. databases such as genevestigator or csb database [52,53] con- Using this construct for Agro-infiltration, a GFP signal could firms that all seven genes are expressed in a tissue-specific man- be detected inside of the chloroplasts (Fig. 5D) indicating that ner. Moreover, the in silico analysis of various microarrays the proteins are targeted to plastids. indicated that they respond differentially to changing environ- A slightly different localisation pattern was observed for mental conditions. For example, PFK1 and PFK3 were found PFK1. Transient expression in N. bethamiana and subsequent 2408 A. Mustroph et al. / FEBS Letters 581 (2007) 2401–2410

Fig. 5. Sub-cellular localization of PFKs. The entire coding sequences of AtPFK 1–7 were cloned without the stop codon into the vector pBinGFP, transformed into Agrobacterium, which were then infiltrated into Nicotiana benthamiana leaves. Protoplasts isolated two days after Agro-infiltration were inspected using a confocal laser scanning microscope to detect GFP fluorescence. A, PFK1; B, PFK2; C, PFK3; D, PFK4 – transit peptide GFP fusion protein; E, PFK4; F, PFK5; G, PFK6; H, PFK7; I, empty pBinAR vector. The bars represent 20 lm. Similar results were obtained in at least two different experiments. Red: chlorophyll autofluorescence, green: GFP fluorescence.

analysis of protoplasts displayed plasmamembrane bound 4. Conclusions localisation of PFK1 (Fig. 5A). In addition, this protein could be also detected in the cytosol where it formed small granules The systematic bioinformatic analysis of the Arabidopsis (data not shown). Transient expression in A. thaliana resulted genome led to the identification of seven sequences with high in a similar pattern. Sequence analysis using PROSITE (http:// homology to PFKs from Anabena and other microorganisms. www..org/prosite/, [58]) did not reveal sequence infor- Subsequent activity tests using transient Agrobacterium-medi- mation such as N-terminal myristylation sides indicative for ated expression revealed the induction of PFK activity for all plasma membrane anchors. of the tested constructs. Therefore, we suggest re-annotating Remarkably, Giege´ et al. [59] demonstrated the presence of these seven sequences to ATP-dependent phosphofructoki- glycolytic enzymes in a mitochondrial fraction of Arabidopsis nase. KO- and transgenic plants will help to elucidate cells, and postulated that glycolytic proteins are attached to the function of the individual isoforms during plant develop- the outside of mitochondria to ensure direct delivery of pyru- ment and in response to environmental stimuli. vate to support the respiration chain. In their study, three per- cent of total PFK activity was found to be associated with Acknowledgements: We thank Christiane Bo¨rnke and Heike Deppner mitochondria. Interestingly, GFP images obtained for PFK6 for cloning of the PFKs, Jasmin Drobietz for help during biochemical analysis. Bernhard Claus and Christina ¨hn are acknowledged for resemble somehow those of aldolase- and -YFP fusion their help with the Confocal Laser Scanning Microscope. A.M. thanks proteins from Giege´ et al. [59]. In addition to their presence in Prof. Bernhard Grimm for excellent support. the cytosol, aldolase and enolase were found associated with mitochondria indicated by the accumulation of circular or rod-shaped bodies that co-localise with Mitotracker. Hence, Appendix A. Supplementary data we cannot exclude that some of the cytosolic isoforms are also associated with the outer membrane of mitochondria as sug- Supplementary data associated with this article can be found, gested by Giege´ et al. [59]. in the online version, at doi:10.1016/j.febslet.2007.04.060. A. Mustroph et al. / FEBS Letters 581 (2007) 2401–2410 2409

References banana: role of magnesium on its kinetics and regulation. Plant Sci. 81, 29–36. [1] Plaxton, W.C. (1996) The organization and regulation of plant [22] Turner, W.L. and Plaxton, W.C. (2003) Purification and charac- glycolysis. Annu. Rev. Plant Physiol. Plant Mol. Biol. 48, 185– terization of pyrophosphate- and ATP-dependent phosphofructo- 214. kinases from banana fruit. Planta 217, 113–121. [2] Bapteste, E., Moreira, D. and Philippe, H. (2003) Rampant [23] Cawood, M.C., Botha, F.C. and Small, J.G.C. (1988) Properties horizontal gene transfer and phospho-donor change in the of the phosphofructokinase isoenzymes from germinating cucum- evolution of the phosphofructokinase. Gene 318, 185–191. ber seeds. J. Plant. Physiol. 132, 204–209. [3] Siebers, B. and Scho¨nheit, P. (2005) Unusual pathways and [24] Wong, J.H., Yee, B.C. and Buchanan, B.B. (1987) A novel type of enzymes of metabolism in Archae. Curr. Opin. phosphofructokinase from plants. J. Biol. Chem. 262, 3185–3191. Microbiol. 8, 695–705. [25] Vella, J. and Copeland, L. (1993) Phosphofructokinase from the [4] Carlisle, S.M., Blakeley, S.D., Hemmingsem, S.M., Trevanion, host fraction of soybean nodules. J. Plant. Physiol. 141, 398–404. S.J., Hiyoshi, T., Kruger, N.J. and Dennis, D.T. (1990) Pyro- [26] Isaac, J.E. and Rhodes, M.J.C. (1982) Purification and properties phosphate-dependent phosphofructokinase. Conservation of pro- of phosphofructokinase from fruits of Lycopersicon esculentum. tein sequence between the alpha- and beta- subunits and with the Phytochemistry 21, 1553–1556. ATP-dependent phosphofructokinase. J. Biol. Chem. 265, 18366– [27] Knowles, V.L., Greyson, M.F. and Dennis, D.T. (1990) Charac- 18371. terization of ATP-dependent fructose 6-phosphate 1-phospho- [5] Todd, J.F., Blakeley, S.D. and Dennis, D.T. (1995) Structure of transferase from leaf and endosperm tissues of Ricinus the genes encoding the a and b-subunits of castor pyrophosphate- communis. Plant Physiol. 92, 155–159. dependent phosphofructokinase. Gene 152, 181–186. [28] Podesta, F.E. and Plaxton, W.C. (1994) Regulation of cytosolic [6] Kapri, R., Dahan, E., Zehavi, U., Goren, R. and Sadka, A. (2000) carbon metabolism in germinating Ricinus communis cotyledons. Cloning and characterization of PPi-phosphofructokinase from 1. Developmental profiles for the activity, concentration, and citrus fruit. Acta Horticult. (ISHS) 535, 113–118. molecular structure of the pyrophosphate-dependent and ATP- [7] Suzuki, J., Mutton, M.A., Ferro, M.I.T., Lemos, M.V.F., dependent phosphofructokinase, phosphoenolpyruvate carboxyl- Pizauro, F.M., Mutton, M.J.R. and DiMauro, S.M.Z. (2003) ase and . Planta 194, 374–380. Putative pyrophosphate phosphofructose 1-kinase genes identified [29] Kruger, N.J. and Hammond, J.B.W. (1988) Molecular compar- in sugar cane may be getting energy from pyrophosphate. Genet. ison of pyrophosphate- and ATP-dependent fructose 6-phosphate Mol. Res. 2, 376–382. 1- from potato tuber. Plant Physiol. 86, 645– [8] Alves, A.M.C.R., Euverink, G.J.W., Santos, H. and Dijkhuizen, 648. L. (2001) Different physiological roles of ATP- and PPi-dependent [30] Hausler, R.E., Holtum, J.A. and Latzko, E. (1989) Cytosolic phosphofructokinase isoenzymes in the methylotrophic actino- phosphofructokinase from spinach leaves. I. Purification, char- mycete Amycolatopsis methanolica. J. Bacteriol. 183, 7231–7240. acteristics, and regulation. Plant Physiol. 90, 1498–1505. [9] Deng, Z., Huang, M., Singh, K., Albach, R.A., Latshaw, S.P., [31] Winkler, C., Delvos, B., Martin, W. and Henze, K. (2007) Chang, K.P. and Kemp, R.G. (1998) Cloning and expression of Purification, microsequencing and cloning of spinach ATP- the gene for the active PPi-dependent phosphofructokinase of dependent phosphofructokinase link sequence and function for Entamoeba histolytica. Biochem J. 329, 659–664. the plant enzyme. FEBS J. 274, 429–438. [10] Wong, J.H., Kiss, F., Wu, M.-X. and Buchanan, B.B. (1990) [32] Mahajan, R. and Singh, R. (1992) Properties of ATP-dependent Pyrophosphate fructose-6-P 1- from tomato phosphofructokinase from endosperm of developing wheat (Trit- fruit. Evidence for change during ripening. Plant Physiol. 94, 499– icum aestivum L.) grains. J. Plant Biochem. Biotechnol. 1, 45–48. 506. [33] Nielsen, T.H., Rung, J.H. and Villadsen, D. (2004) Fructose-2,6- [11] Teramoto, M., Koshiishi, C. and Ashihara, H. (2000) Wound- bisphosphate: a traffic signal in plant metabolism. Trends Plant induced respiration and pyrophosphate:fructose-6-phosphate Sci. 9, 556–563. phosphotransferase in potato tubers. Z. Naturforsch. 55, 953–956. [34] Logemann, J., Schell, J. and Willmitzer, L. (1987) Improved [12] Stitt, M. (1990) Fructose-2,6-bisphosphate as a regulatory mol- method for the isolation of RNA from plant tissues. Anal. ecule in plants. Annu. Rev. Plant Physiol. Plant Mol. Biol. 41, Biochem. 163, 16–20. 153–185. [35] Hofgen, R. and Willmitzer, L. (1988) Storage of competent cells [13] Dunaway, G.A. (1983) A review of animal phosphofructokinase for Agrobacterium transformation. Nucleic Acids Res. 16, 9877. isozymes with an emphasis on their physiological role. Mol. Cell. [36] Giese, J.O., Herbers, K., Hoffmann, M., Klosgen, R.B. and Biochem. 52, 75–91. Sonnewald, U. (2005) Isolation and functional characterization of [14] Hajirezaei, M., Sonnewald, U., Viola, R., Carlisle, S., Dennis, D. a novel plastidic from Nicotiana tabacum. FEBS Lett. and Stitt, M. (1994) Transgenic potato plants with strongly 579, 827–831. decreased expression of pyrophosphate: fructose-6-phosphate [37] Deblaere, R., Bytebier, B., De Greve, H., Deboeck, F., Schell, J., phosphotransferase show no visible phenotype and only minor Van Montagu, M. and Leemans, J. (1985) Efficient octopine Ti changes in metabolic fluxes in their tubers. Planta 193, 16–30. plasmid-derived vectors for Agrobacterium-mediated gene trans- [15] Theodorou, M.E., Cornel, F.A., Duff, S.M. and Plaxton, W.C. fer to plants. Nucleic Acids Res. 13, 4777–4788. (1992) Phosphate starvation-inducible synthesis of the alpha- [38] Bendahmane, A., Querci, M., Kanyuka, K. and Baulcombe, D.C. subunit of the pyrophosphate-dependent phosphofructokinase in (2000) Agrobacterium transient expression system as a tool for the black mustard suspension cells. J. Biol. Chem. 267, 21901–21905. isolation of disease resistance genes: application to the Rx2 locus [16] Mertens, E., Laroundelle, Y. and Hers, H-G. (1990) Induction of in potato. Plant J. 21, 73–81. pyrophosphate:fructose 6-phosphate 1-phosphotransferase by [39] Gibbs, J., Morrell, S., Valdez, A., Setter, T.L. and Greenway, H. anoxia in rice seedlings. Plant Physiol. 9, 584–587. (2000) Regulation of alcoholic fermentation in coleoptiles of two [17] Stitt, M. (1998) Pyrophosphate as an energy donor in the cytosol rice cultivars differing in tolerance to anoxia. J. Exp. Bot. 51, 785– of plant cells: an enigmatic alternative to ATP. Botanica Acta 111, 796. 167–175. [40] Bradford, M.M. (1976) A rapid and sensitive method for the [18] Salminen, S.O. and Young, R.E. (1975) The control properties of quantification of microgram quantities of protein utilizing the phosphofructokinase in relation to the respiratory climacteric in principle of protein–dye binding. Anal. Biochem. 72, 248–254. banana fruit. Plant Physiol. 55, 45–50. [41] Bayley, C.C., Morgan, M., Dale, E.C. and Ow, D.W. (1992) [19] Nair, P.M. and Darak, B.G. (1981) Identification of multiple Exchange of gene activity in transgenic plants catalyzed by the forms of phosphofructokinase in ripening dwarf cavendish Cre-lox site-specific recombination system. Plant Mol. Biol. 18, banana. Phytochemistry 20, 605–609. 353–361. [20] Surendranathan, K.K., Iyer, M.G. and Nair, P.M. (1990) [42] Mertens, E., Ladror, U.S., Lee, J.A., Miretsky, A., Morris, A., Characterization of a monomeric phosphofructokinase from Rozario, C., Kemp, R.G. and Muller, M. (1998) The pyrophos- banana - role of magnesium in its regulation. Plant Sci. 72, 27–35. phate-dependent phosphofructokinase of the protist, Trichomonas [21] Surendranathan, K.K., Iyer, M.G. and Nair, P.M. (1992) vaginalis, and the evolutionary relationships of protist phosp- Mechanism of action of a dimeric phosphofructokinase from hofructokinases. J. Mol. Evol. 47, 739–750. 2410 A. Mustroph et al. / FEBS Letters 581 (2007) 2401–2410

[43] Muller, M., Lee, J.A., Gordon, P., Gaasterland, T. and Sensen, [52] Steinhauser, D., Usadel, B., Luedemann, A., Thimm, O. and T.W. (2001) Presence of prokaryotic and eukaryotic species in all Kopka, J. (2004) CSB.DB: a comprehensive systems-biology subgroups of the PPi-dependent group II phosphofructokinase database. Bioinformatics 20, 3647–3651. . J. Bacteriol. 183, 6714–6716. [53] Zimmermann, P., Hirsch-Hoffmann, M., Hennig, L. and Gruis- [44] Siebers, B., Klenk, H.P. and Hensel, R. (1998) PPi-dependent sem, W. (2004) GENEVESTIGATOR: Arabidopsis Microarray phosphofructokinase from Thermoproteus tenax, an archeal Database and Analysis Toolbox. Plant Physiol. 136, 2621–2632. descendant of an ancient line in phosphofructokinase evolution. [54] Emanuelsson, O., Nielson, H. and von Heijne, G. (1999) ChloroP, J. Bacteriol. 180, 2137–2143. a neural network-based method for predicting chloroplast transit [45] Shirakihara, Y. and Evans, P.R. (1988) Crystal structure of the peptides and their cleavage sites. Protein Sci. 8, 978–984. complex of phosphofructokinase from with its [55] Emanuelsson, O., Nielsen, H., Brunak, S. and von Heijne, G. reaction products. J. Mol. Biol. 204, 973–994. (2000) Predicting subcellular localization of proteins based on their [46] Moore, S.A., Ronimus, R.S., Roberson, R.S. and Morgan, H.W. N-terminal amino acid sequence. J. Mol. Biol. 300, 1005–1016. (2002) The structure of a pyrophosphate-dependent phosphofruc- [56] Anderson, L.E., Wang, X. and Gibbons, J.T. (1995) Three tokinase from the lyme disease spirochete Borrelia burgdorferi. enzymes of carbon metabolism or their antigenic analogs in pea Structure 106, 659–671. leaf nuclei. Plant Physiol. 108, 659–667. [47] Chi, A.S. and Kemp, R.G. (2000) The primordial high energy [57] Horton, P., Park, K.J., Obayashi, T. and Nakai, K. (2006) Protein compound: ATP or inorganic pyrophosphate? J. Biol. Chem. 275, subcellular localization prediction with WoLF PSORT. In: 35677–35679. Proceedings of the 4th Annual Asian Pacific Bioinformatics [48] Yan, T.F.J. and Tao, M. (1984) Multiple forms of pyrophos- Conference APBC06, Taipei, pp. 39–48. phate-fructose-6-phosphate 1-phosphotransferase from wheat [58] Hulo, N., Bairoch, A., Bulliard, V., Cerutti, L., De Castro, E., seedlings. J. Biol. Chem. 259, 5087–5092. Langendijk-Genevaux, P.S., Pagni, M. and Sigrist, C.J. (2006) [49] Kruger, N.J. and Dennis, D.T. (1985) Reassessment of an The PROSITE database. Nucleic Acids Res. 34, D227–D230. apparent hyperactive form of phosphofructokinase from plants. [59] Giege, P., Heazlewood, J.L., Roessner-Tunali, U., Millar, A.H., Plant Physiol. 78, 645–648. Fernie, A.R., Leaver, C.J. and Sweetlove, L.J. (2003) Enzymes of [50] Botha, F.C. and Turpin, D.H. (1990) Molecular, kinetic, and glycolysis are functionally associated with the mitochondrion in immunological properties of the 6-Phosphofructokinase from the Arabidopsis cells. Plant Cell. 15, 2140–2151. green alga Selenastrum minutum: activation during biosynthetic [60] Romeis, T., Ludwig, A.A., Martin, R. and Jones, J.D.G. (2001) carbon flow. Plant Physiol. 93, 871–879. Calcium-dependent protein kinases play an essential role in a [51] Dunaway, G.A., Kasten, T.P., Nickols, G.A. and Chesky, J.A. plant defence response. EMBO J. 20, 5556–5567. (1986) Regulation of 6-phosphofructo-1-kinase [61] Fridman, E. and Zamir, D. (2003) Functional divergence of a during aging and development. Mech. Ageing Dev. 36, 13– syntenic invertase gene family in tomato, potato, and Arabidop- 23. sis. Plant Physiol. 131, 603–609.