The pea mitochondrial nucleoside diphosphate kinase cleaves DNA and RNA

FEBS Letters 581 (2007) 3507–3511

The pea mitochondrial nucleoside diphosphate kinase cleaves DNA and RNA Jenni Hammargrena, Thalia Salinasb, Laurence Mare´chal-Drouardb, Carina Knorppa,* b

a Department of Plant Biology and Forest Genetics, Swedish University of Agricultural Sciences, Box 7080, S-750 07 Uppsala, Sweden Institut de Biologie Mole´culaire des Plantes, Unite´ Propre de Recherche 2357, Conventionne´e avec l’Universite´ Louis Pasteur (Strasbourg 1), Centre National de la Recherche Scientifique, 12 Rue du Ge´ne´ral Zimmer, 67084 Strasbourg Cedex, France

Received 30 May 2007; accepted 22 June 2007 Available online 2 July 2007 Edited by Ulf-Ingo Flu¨gge

Abstract Here, we present the characterization of a plant NDPK exhibiting nuclease activity. This is the first identification of a nuclease localised in the intermembrane space of plant mitochondria. The recombinant pea NDPK3 protein cleaves not only supercoiled plasmid DNA, but also highly structured RNA molecules such as tRNAs or the 3 0 UTR of the atp9 mRNA suggesting that the NDPK3 nuclease activity has a structural requirement. ATP inhibits this nuclease activity, while ADP has no effect. Furthermore, studies on NDPK mutant proteins indicate that the nuclease- and the kinase-mechanisms are separate.  2007 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. Keywords: NDPK; Mitochondria; Plant; Intermembrane space; Nuclease; tRNA

strated for bacterial NDPKs such as that of Escherichia coli [7] and Mycobacterium turberculosis [8]. Three NDPK isoforms are found in plants. NDPK1 resides in the cytosol, NDPK2 in the chloroplast stroma and NDPK3 in the chloroplast lumen and in the mitochondrial intermembrane space (IMS)/inner membrane [9–14]. In addition to their nucleotide producing function, plant NDPKs have been implicated in signal transduction events, e.g. in light response [15– 17]. It has also been shown that plant NDPK2 interacts with two oxidative stress regulated mitogen activated protein kinases [18]. However, the capability of the plant NDPKs to cleave nucleic acid has not previously been addressed. In this study, we investigate the nuclease activity of the recombinant pea NDPK3 protein. 2. Materials and methods

1. Introduction Nucleoside diphosphate kinases (NDPKs) are enzymes that catalyse the transfer of a phosphate between nucleoside triand di-phosphates via a ping-pong mechanism involving a conserved histidine. Although the NDPKs were long considered exclusively as house-keeping enzymes with a primary function as equilibrators of the nucleoside pool, a multitude of studies demonstrate additional functions in signalling cascades and in the interaction and modulation of the activity of other proteins [1,2]. Eight NDPK genes reside in the human genome (NDPK-H1NDPK-H8). NDPK-H1 is involved in tumour suppression through its action as a DNase during Granzyme A induced apoptosis [3]. NDPK-H2 is a proto-oncogene, which activates the transcription of c-myc gene by binding to a nuclease hypersensitive region (NHE) in the c-myc promoter [4]. Interestingly, studies have shown that while binding, NDPK-H2 also cleaves sequences within the NHE [5]. In addition, NDPKH-5, H-7 and H-8 have been shown to exhibit exonucleolytic nuclease activity [6]. Nuclease activity has also been demon* Corresponding author. Fax: +46 18 67 33 89. E-mail address: [email protected] (C. Knorpp).

Abbreviations: AIF, apoptosis-inducing factor; Endo G, endonuclease G; IMS, intermembrane space; NDPK, nucleoside diphosphate kinase; NHE, nuclease hypersensitive region; PCD, programmed cell death

2.1. Expression and purification of NDPK proteins Proteins were expressed in E. coli as described in Johansson et al. [19]. The His-tag was subsequently cleaved off using TeV protease (Invitrogen), and eliminated by purification on a His-trap column (Amersham Pharmacia Biotech). The His-tag containing TeV protease was also eliminated in this purification. Purified, cleaved proteins were equilibrated in 50 mM Tris–HCl, pH 7.4, 10% glycerol using PD-10 columns (GE Healthcare). 2.2. Production of DNA substrates pBluescript KS (Stratagene) plasmid was linearised by cleavage with XhoI (Fermentas). The following oligonucleotides were synthetically synthesized (Invitrogen). A: 5 0 -AGGATTCATTTAGGAGCGCTCTTTTCATCCAACTCGAAATATCGTAAGATAAGAAAGATC3 0 and B: 5 0 -ATCTTTCTTATCTTACGATATTTCGAGTTGGATGAAAAGAGCGCTCCTAAATGAATCCT-3 0 . Oligonucleotide A was used as the single stranded DNA substrate. To produce the double stranded substrate the A and B oligonucleotides were annealed by mixing equal molar amounts of each oligonucleotide. The mixture was heated at 80 C for 10 min, and then placed on ice. The annealed products were purified using the Qiagen gel purification kit (Qiagen). 2.3. Production of RNA substrates tRNAAla or tRNAMet in vitro transcripts were synthesised under conditions previously described [20]. A PCR product corresponding to the 3 0 UTR of the potato atp9 transcript fused to a T7 promoter was used as template in the synthesis of a transcript corresponding to the atp9 sequence from position +290 to +423 with respect to the first nucleotide of the start codon [21]. To obtain an a priori ‘‘unstructured’’ RNA substrate a Bluescript plasmid SK (Stratagene) linearised with HindIII (Fermentas) was purified using the Qiagen gel purification kit (Qiagen). Transcription from the purified product was performed using the MEGA script kit (Ambion), and the transcription product was phenol extracted and isopropanol precipitated before use.

0014-5793/$32.00  2007 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. doi:10.1016/j.febslet.2007.06.062

3508

J. Hammargren et al. / FEBS Letters 581 (2007) 3507–3511

2.4. In vitro cleavage assay Two hundred and fifty nanograms or 1 lg substrate was incubated with 600 ng recombinantly produced NDPK3 at 30 C for 2.5 h in the presence of 10 mM MgCl2. ATP (Sigma) or ADP (Sigma) were added to the reactions as described in the text. All reactions were run at least three times, using a minimum of two biological replicates (batches of purified protein). Cleavage of pBluescript DNA was tested in parallel to all other assays, in order to ascertain the activity of the NDPK3 protein. Recombinant his-tagged Adenylate Kinase (E.C.2.7.4.3), purified at the same time as the NDPK3 protein using the same method, was employed as a negative control. Quantifications of cleavage products in the gels were done using the Quantity One program (BioRad, v. 4.5.0).

3. Results 3.1. DNA cleavage by NDPK3 The nuclease activity of NDPK3 was investigated using an in vitro cleavage system as described in Section 2. First, the ability of NDPK3 to cleave and degrade a supercoiled plasmid was tested. This experiment demonstrates that NDPK3 is able to nick the plasmid, producing the open circular form (Fig. 1A). Some cleavage of the open circular form into the linear form was also observed, although this was not prominent in most assays (data not shown). The cleavage activity was found to be dependent on incubation time and enzyme amount in a linear fashion up to 2.5 h and 600 ng of enzyme using 1 lg substrate (data not shown). However, NDPK3 is not capable to further cleave and/or degrade the linear form of the plasmid (Fig. 1A). Furthermore, we show that NDPK3 is unable to cleave two small DNA molecules of 60 nucleotides, one single stranded (ssDNA), and one double stranded (dsDNA, Fig. 1B). 3.2. DNA cleavage by NDPK3 mutants S69 and H117 We have previously carried out studies using two mutant NDPK3 enzymes, H117D and S69A, which are devoid of kinase activity [19]. The H117 residue is crucial as it mediates the phosphotransfer in the kinase reaction, while the inactivity of the S69 mutant was ascribed to its inability to form hexamers [19]. In order to investigate whether similar mechanisms are involved in the DNA cleavage activity of NDPK3 as in

M

2000 bp

-

+

-

the phosphotransfer reaction we tested the ability of the H117D and the S69A mutants to cleave supercoiled plasmid DNA (Fig. 2). No difference in cleavage was observed between the wild-type protein and the mutants in either case, indicating that the NDPK kinase- and nuclease-mechanisms are separate. 3.3. Effect of ATP on NDPK3 DNA cleavage ATP has previously proved to be a potent inhibitor of NDPK nuclease activity [5,7,8]. In order to examine the effect of ATP on the activity of the pea mitochondrial NDPK we incubated the enzyme with pBluescript in the presence of increasing amounts of ATP as well as ADP, which was used as a control (Fig. 3). As expected ATP inhibited the activity of NDPK, whereas ADP had no effect. ATP inhibition is effective even at low concentrations, as shown by the 64% inhibition caused by the addition of 1 mM ATP. A 100% inhibition was obtained at 4 mM ATP. 3.4. RNA cleavage by NDPK3 The investigations of NDPK nuclease activity have to date been limited to the area of DNA degradation, although studies have demonstrated that NDPK can bind pyrimidine-rich RNA

M

-

WT

S69A H117D

oc lin superc

2000 bp

Fig. 2. Cleavage of plasmid DNA by NDPK3 mutants. 1 lg Bluescript plasmid was incubated with 600 ng NDPK wt, S69A or H117D. M is mass mix ruler (Fermentas),  indicates the absence of NDPK protein and + indicates the presence of NDPK protein. Oc is opencirclar, lin is linear and superc is supercoiled plasmids.

+

bp

M

ssDNA

dsDNA

-

-

+

+

oc lin superc 75 50 35

Fig. 1. Characterization of DNA cleavage. (A) Agarose gels showing 1 lg supercoiled and 1 lg linear pBluescript and (B) polyacrylamide gel showing 250 ng ssDNA and dsDNA. The substrates were incubated in the absence () and in the presence (+) of 600 ng NDPK3. The same purified NDPK batch was used in (A) and (B). M is mass mix ruler (Fermentas). Oc is opencirclar, lin is linear and superc is supercoiled plasmids.

J. Hammargren et al. / FEBS Letters 581 (2007) 3507–3511

3509

120 ATP ADP

Percent inhibition

100 80 60 40 20 0 0

1

2

3

4

synthesized tRNA molecules. Both the atp9 3 0 UTR transcript and the tRNAs have very highly folded secondary structures, containing, e.g. predicted stem–loops [20,21]. The ‘‘random’’ RNA molecule was not cleaved by NDPK3 (Fig. 4A), whereas incubation of NDPK3 with the 3 0 UTR of the atp9 RNA molecule produced a specific cleavage pattern of about seven fragments ranging from 35 nucleotides (nt) to 150 nt in size (Fig. 4B). In addition, tRNAAla and tRNAMet were cleaved by NDPK, yielding small fragments ranging from 20 to 45 nt in size (Fig. 4C). Interestingly, ATP inhibited the cleavage of tRNAAla (Fig. 4D), in agreement with results obtained using plasmid DNA (Fig. 3).

-20 (mM)

Fig. 3. Regulation of DNA cleavage activity by ATP and ADP. 1 lg Bluescript plasmid was incubated with 600 ng NDPK protein in the presence of 0, 1, 2, 3, or 4 mM ATP or ADP. Cleavage products were separated on agarose gels. Error bars refer to standard deviation between three biological replicates.

molecules [22]. We proceeded to investigate whether NDPK3 is also able to cleave RNA. For this purpose, three types of RNA molecules were used; first a single stranded RNA produced by in vitro transcription from the cloning cassette of pBluescript (‘‘random’’ RNA), second a transcript corresponding to the 3 0 UTR of the mitochondrial atp9 mRNA, and third in vitro

bp

M

-

4. Discussion In this study the first evidence of a plant NDPK harbouring nuclease activity is presented. We report that the pea mitochondrial NDPK3 cleaves a supercoiled plasmid substrate, but is unable to cleave the linear counterpart as well as small ‘‘random’’ DNA molecules (Fig. 1A–C). A similar scenario holds true in regards to RNA (Fig. 4A–C) where highly structured substrates (3 0 UTR of atp9 mRNA and tRNAAla, tRNAMet) are cleaved, while the ‘‘random’’ RNA molecule remains intact. Based on these findings we suggest that the actual identity of the nucleic acid is of less importance for the cleavage specificity of NDPK3, rather it is the secondary or tertiary structure of the

bp

+

M

-

+

200 100 75 75

35

bp

tRNAAla M +

tRNAMet +

bp

100 75

M

tRNAAla - + - + +

ATP NDPK

80

50 35

40

25

Fig. 4. Characterization of RNA cleavage activity. Polyacrylamide gels showing 250 ng (A) ‘‘random’’ RNA, (B) 3 0 -UTR RNA of the potato mitochondrial atp9 gene, (C) Arabidopsis tRNAAla and tRNAMet and (D) Arabidopsis tRNAAla. The substrates were incubated in the absence () or in the presence (+) of 600 ng of NDPK3. 2 mM ATP was used in D. M is ultra low range ruler (Fermentas). Arrows indicate cleavage products. Negative images of ethidium bromide stained gels are shown for clarity.

3510

nucleic acid molecule that is of greatest significance. It is interesting to note that the NHE element, which has been extensively studied as a substrate for the human NDPK nuclease activity [4,7,8] forms a defined secondary structure, the Gquadruplex [23]. Indeed, the specific structure of the G-quadruplex has previously been suggested to be significant for NHE binding and/or cleavage by NDPK [4]. Plant cytosolic tRNAAla is naturally imported into mitochondria, while tRNAMet is cytosol-specific [20]. Consequently, tRNAAla must cross the double mitochondrial membranes, as well as the IMS where NDPK3 is located, in order to reach the matrix. It is thus tempting to speculate that NDPK3 may be involved in the regulation of the plant mitochondrial tRNA import process by degrading, under particular circumstances, cytosolic tRNAs on their way to the matrix side. However, since both tRNAAla and tRNAMet were substrates of NDPK3 in vitro it is unlikely that this enzyme plays a role in the specificity of the tRNA import process. We have previously reported that the pea NDPK3 S69 residue is crucial for its ability to form hexamers [19]. Here, we show that this defect does not affect the ability of NDPK3 to cleave a supercoiled plasmid (Fig. 2). This result is in agreement with previous studies, which have shown DNA binding to be mainly associated with the dimeric form of NDPK [24]. Mesnildrey et al. also presented the NDPK dimers as the primary DNA binding oligomeric form, and suggested that the variation in ability between NDPKs from different species to bind DNA (i.e. single stranded pyrimidine-rich strands) is related to the relative stability of the oligomeric forms [25]. Also compatible with earlier studies using NDPK-H2 [26,27], we report that the H117 residue is not crucial for the cleavage activity of NDPK3. Conclusively, NDPK3 DNA cleavage is independent of a hexameric stucture or a functional phosphorylation site at H117. The ATP status regulates NDPK3 cleavage, with inhibition at concentrations similar to those shown for NDPK-H2 [27]. There is conflicting information regarding the cell concentration of ATP, with some studies reporting numbers in the lM range [28], and others in the mM range [29,30]. Complicating matters further, the ATP concentration is likely to vary between different subcellular compartments inside the mitochondria, as well as between locations within a specific compartment. This could mean that NDPK3 encounters relatively high ATP levels at some sites, which probably results in a complete or near complete inhibition of nuclease activity. One such place might be at the point of interaction with the ATP/ADP translocator [31]. Interestingly, one of the earliest signs of programmed cell death (PCD) is an impairment in oxidative phosphorylation [32], in addition to a lowered ATP production [33]. IMS proteins are released later during the course of PCD [34]. We suggest that NDPK3 located in the soluble IMS could be activated by a decreasing ATP concentration during plant PCD, released from mitochondria and translocated into the nucleus where it might nick nuclear DNA. This situation would be similar to that of animals where two IMS residing nucleases, endonuclease G (Endo G) and apoptosis-inducing factor (AIF), are released during PCD [35,36]. In Arabidopsis, nuclease activities with characteristics similar to AIF and Endo G have previously been observed in the IMS compartment [37], although the identity/-ies of the responsible protein/-s have not yet been identified. The reported activities are exonucleolytic, and can therefore not likely be ascribed to the nucle-

J. Hammargren et al. / FEBS Letters 581 (2007) 3507–3511

ase activity of the Arabidopsis mitochondrial NDPK3. Further investigations will elucidate the possible involvement of the plant mitochondrial NDPKs in PCD. Acknowledgements: The authors thank D. Gagliardi for the kind gift of the PCR template for the 3 0 UTR atp9 transcript. This study received financial support from the Helge Ax:son Johnson’s foundation and the Swedish Science Research Council.

References [1] Leung, S.M. and Hightower, L.E. (1997) A 16-kDa protein functions as a new regulatory protein for Hsc70 molecular chaperone and is identified as a member of the Nm23/nucleoside diphosphate kinase family. Journal of Biological Chemistry 31, 2607–2614. [2] Crawford, R.M. et al. (2006) NDPK-A (but not NDPK-B) and AMPK alpha1 (but not AMPK alpha2) bind the cystic fibrosis transmembrane conductance regulator in epithelial cell membranes. Cell Signalling 18, 1595–1603. [3] Fan, Z., Beresford, P.J., Oh, D.Y., Zhang, D. and Lieberman, J. (2003) Tumor suppressor NM23-H1 is a granzyme A-activated DNase during CTL-mediated apoptosis, and the nucleosome assembly protein SET is its inhibitor. Cell 112, 659–672. [4] Postel, E.H., Berberich, S.J., Flint, S.J. and Ferrone, C.A. (1993) Human c-myc transcription factor PuF identified as nm23-H2 nucleoside diphosphate kinase, a candidate suppressor of tumor metastasis. Science 261, 478–480. [5] Postel, E.H. (1999) Cleavage of DNA by human NM23-H2/ nucleoside diphosphate kinase involves formation of a covalent protein-DNA complex. Journal of Biological Chemistry 274, 22821–22829. [6] Yoon, J.H. et al. (2005) Characterization of the 3 0 fi 5 0 exonuclease activity found in human nucleoside diphosphate kinase 1 (NDK1) and several of its homologues. Biochemistry 44, 15774– 15786. [7] Levit, M.N., Abramczyk, B.M., Stock, J.B. and Postel, E.H. (2002) Interactions between Escherichia coli nucleoside-diphosphate kinase and DNA. Journal of Biological Chemistry 277, 5163–5167. [8] Kumar, P., Verma, A., Saini, A.K., Chopra, P., Chakraborti, P.K., Singh, Y. and Chowdhury, S. (2005) Nucleoside diphosphate kinase from Mycobacterium tuberculosis cleaves single strand DNA within the human c-myc promoter in an enzymecatalyzed reaction. Nucleic Acids Research 33, 2707–2714. [9] Hammargren, J., Sundstro¨m, S., Johansson, M., Bergman, P. and Knorpp, C. (2007) On the phylogeny, expression and targeting of plant nucleoside diphosphate kinases. Physiologia Plantarum 129, 79–89. [10] Tanaka, N., Ogura, T., Noguchi, T., Hirano, H., Yabe, N. and Hasunuma, K. (1998) Phytochrome-mediated light signals are transduced to nucleoside diphosphate kinase in Pisum sativum L. cv. Alaska. Journal of Photochemistry and Photobiology 45, 113– 121. [11] Yang, L.M. and Lamppa, G.K. (1996) Rapid purification of a chloroplast nucleoside diphosphate kinase using CoAaffinity chromatography. Biochimica Biophysica Acta 1294, 99– 102. [12] Spetea, C., Hundal, T., Lundin, B., Heddad, M., Adamska, I. and Andersson, B. (2004) Multiple evidence for nucleotide metabolism in the chloroplast thylakoid lumen. The Proceedings of the National Academy of Sciences (USA) 101, 1409–1414. [13] Struglics, A. and Ha˚kansson, G. (1999) Purification of a serine and histidine phosphorylated mitochondrial nucleoside diphosphate kinase from Pisum sativum. European Journal of Biochemistry 262, 765–773. [14] Sweetlove, L.J., Mowday, B., Hebestreit, H.F., Leaver, C.J. and Millar, A.H. (2001) Nucleoside diphosphate kinase III is localized to the inter-membrane space in plant mitochondria. FEBS Letters 508, 272–276. [15] Choi, G., Yi, H. and Lee, J. (1999) Phytochrome signalling is mediated through nucleoside diphosphate kinase-II. Nature 401, 610–613.

J. Hammargren et al. / FEBS Letters 581 (2007) 3507–3511 [16] Ryu, J.S. et al. (2005) Phytochrome-specific type 5 phosphatase controls light signal flux by enhancing phytochrome stability and affinity for a signal transducer. Cell 120, 395–406. [17] Zimmermann, S., Baumann, A., Jaekel, K., Marbach, I., Engelberg, D. and Frohnmeyer, H. (1999) UV-responsive genes of Arabidopsis revealed by similarity to the Gcn4-mediated UV response in yeast. Journal of Biological Chemistry 274, 17017– 17024. [18] Moon, H. et al. (2003) NDP kinase 2 interacts with two oxidative stress-activated MAPKs to regulate cellular redox state and enhances multiple stress tolerance in transgenic plants. The Proceedings of the National Academy of Sciences (USA) 100, 358–363. [19] Johansson, M., Mackenzie-Hose, A., Andersson, I. and Knorpp, C. (2004) Structure and mutational analysis of a plant mitochondrial nucleoside diphosphate kinase. Identification of residues involved in serine phosphorylation and oligomerization. Plant Physiology 136, 3034–3042. [20] Delage, L., Dietrich, A., Cosset, A. and Marechal-Drouard, L. (2003) In vitro import of a nuclearly encoded tRNA into mitochondria of Solanum tuberosum. Molecular Cell Biology 23, 4000–4012. [21] Gagliardi, D., Perrin, R., Marechal-Drouard, L., Grienenberger, J.M. and Leaver, C.J. (2001) Plant mitochondrial polyadenylated mRNAs are degraded by a 3 0 - to 5 0 -exoribonuclease activity, which proceeds unimpeded by stable secondary structures. Journal of Biological Chemistry 276, 43541–43547. [22] Hildebrandt, M., Lacombe, M.L., Mesnildrey, S. and Veron, M. (1995) A human NDP-kinase B specifically binds single-stranded poly-pyrimidine sequences. Nucleic Acids Research 23, 3858– 3864. [23] Ambrus, A., Chen, D., Dai, J., Jones, R.A. and Yang, D. (2005) Solution structure of the biologically relevant G-quadruplex element in the human c-MYC promoter. Implications for Gquadruplex stabilization. Biochemistry 44, 2048–2058. [24] Postel, E.W., Weiss, V.H., Beneken, J. and Kirtane, A. (1996) Mutational analysis of NM23-H2/NDP kinase identifies the structural domains critical to recognition of a c-myc regulatory element. Biochemistry 93, 6892–6897. [25] Mesnildrey, S., Agou, F. and Veron, M. (1997) The in vitro DNA binding properties of NDP kinase are related to its oligomeric state. FEBS Letters 418, 53–57. [26] Postel, E.H. and Ferrone, C.A. (1994) Nucleoside diphosphate kinase enzyme activity of NM23-H2/PuF is not required for its DNA binding and in vitro transcriptional functions. Journal of Biological Chemistry 269, 8627–8630. [27] Postel, E.H., Abramczyk, B.A., Gursky, S.K. and Xu, Y. (2002) Structure-based mutational and functional analysis identify

3511

[28]

[29]

[30]

[31]

[32]

[33]

[34]

[35] [36] [37]

human NM23-H2 as a multifunctional enzyme. Biochemistry 41, 6330–6337. Borisjuk, L., Rolletschek, H., Walenta, S., Panitz, R., Wobus, U. and Weber, H. (2003) Energy status and its control on embryogenesis of legumes: ATP distribution within Vicia faba embryos is developmentally regulated and correlated with photosynthetic capacity. Plant Journal 36, 318–329. Farre, E.M., Tiessen, A., Roessner, U., Geigenberger, P., Trethewey, R.N. and Willmitzer, L. (2001) Analysis of the compartmentation of glycolytic intermediates, nucleotides, sugars, organic acids, amino acids, and sugar alcohols in potato tubers using a nonaqueous fractionation method. Plant Physiology 127, 685–700. Gardestrom, P. and Wigge, B. (1988) Influence of photorespiration on ATP/ADP ratios in the chloroplasts, mitochondria, and cytosol, studied by rapid fractionation of barley (Hordeum vulgare) protoplasts. Plant Physiology 88, 69–76. Knorpp, C., Johansson, M. and Baird, A.M. (2003) Plant mitochondrial nucleoside diphosphate kinase is attached to the membrane through interaction with the adenine nucleotide translocator. FEBS Letters 555, 363–366. Vacca, R.A., de Pinto, M.C., Valenti, D., Passarella, S., Marra, E. and De Gara, L. (2004) Production of reactive oxygen species, alteration of cytosolic ascorbate peroxidase, and impairment of mitochondrial metabolism are early events in heat shock-induced programmed cell death in tobacco Bright-Yellow 2 cells. Plant Physiology 134, 1100–1112. Valenti, D., Vacca, R.A., de Pinto, M.C., De Gara, L., Marra, E. and Passarella, S. (2007) In the early phase of programmed cell death in Tobacco Bright Yellow 2 cells the mitochondrial adenine nucleotide translocator, adenylate kinase and nucleoside diphosphate kinase are impaired in a reactive oxygen species-dependent manner. Biochimica et Biophysica Acta 1767, 66–78. Munoz-Pinedo, C., Guio-Carrion, A., Goldstein, J.C., Fitzgerald, P., Newmeyer, D.D. and Green, D.R. (2006) Different mitochondrial intermembrane space proteins are released during apoptosis in a manner that is coordinately initiated but can vary in duration. The Proceedings of the National Academy of Sciences (USA) 103, 11573–11578. Li, L.Y., Luo, X. and Wang, X. (2001) Endonuclease G is an apoptotic DNase when released from mitochondria. Nature 412, 95–99. Susin, S.A. et al. (1999) Molecular characterization of mitochondrial apoptosis-inducing factor. Nature 397, 441–446. Balk, J., Chew, S.K., Leaver, C.J. and McCabe, P.F. (2003) The intermembrane space of plant mitochondria contains a DNase activity that may be involved in programmed cell death. Plant Journal 34, 573–583.