Biochemical and Bioinformatic Characterization of Type II Metacaspase Protein (TaeMCAII) from Wheat
- 878 Downloads
The biochemical analysis and homology modeling of a tertiary structure of a cereal type II metacaspase protein from wheat (Triticum aestivum), TaeMCAII, are presented. The biochemical characterization of synthetic oligopeptides and protease inhibitors of Escherichia coli-produced and purified recombinant TaeMCAII revealed that this metacaspase protein, similar to other known plant metacaspases, is an arginine/lysine-specific cysteine protease. Thus, a model of a plant type II metacaspase structure based on newly identified putative metacaspase-like template was proposed. Homology modeling of the TaeMCAII active site tertiary structure showed two cysteine residues, Cys140 and 23, in close proximity to the catalytic histidine, most likely participating in proton exchange during the catalytic process. The autoprocessing that leads to activation of TaeMCAII was highly dependent on Cys140. TaeMCAII required high levels of calcium ions for activity, which could indicate its involvement in stress signaling pathways connected to programmed cell death.
KeywordsType II wheat metacaspase Cysteine-dependent autoprocessing Calcium ions Programmed cell death
Programmed cell death (PCD) is a process of elimination of unwanted cells during the ontogenesis of organisms and in response to environmental stresses. It is common to all eukaryotic cells, including animal and plant cells. Plants and animals share many similarities in the morphological features and enzymatic machinery of PCD (Sun et al. 2012; Sanmartin et al. 2005). The initiators and executors of animal PCD are caspases, a family of cysteine-dependent proteases that cleave their substrates at the carboxyl-terminal side of aspartate residues. They are synthesized as inactive proenzymes that comprise an N-terminal prodomain together with one large and one small subunit. The crystal structures of caspases show that the active enzymes are heterotetramers that contain two small and two large subunits. The enzymes have two active sites that are found at opposite ends of the molecules. Both the small and large subunits participate in the formation of active site. Two residues, cysteine and histidine, are present in the active sites and participate in catalysis (Ho and Hawkins 2005; Cohen 1997). The activation of caspases during PCD processes such as apoptosis and autophagy results in the cleavage of important cellular proteins, including poly(ADP-ribose) polymerase and lamins, leading to the demise of the cell (Earnshaw et al. 1999).
The existence of distant caspase relatives named caspase-like proteases has been demonstrated previously in plant cells undergoing PCD (Sanmartin et al. 2005). Metacaspases, which are caspase-like enzymes, were discovered in silico in the Arabidopsis genome more than a decade ago, and they are present in protozoa, fungi and plants (Uren et al. 2000). Phylogenetic analysis has revealed that metacaspases are distant homologs and ancestors of animal caspases (Vercammen et al. 2007). These proteases, together with eukaryotic caspases, metazoan paracaspases, legumains, separases and the bacterial clostripains and gingipains, are classified as members of clan CD cysteine proteases (Bonneau et al. 2008). All proteins from this clan share a common structural feature, the presence of the caspase–hemoglobinase fold (Bonneau et al. 2008).
On the basis of metacaspase structure, they can be subdivided into two groups: type I and type II. Type I metacaspases possess an N-terminal prodomain with a Zn finger motif, which is absent in type II. The distinguishable feature of type II metacaspases is the presence of a linker region between the large and the small subunit (Piszczek and Gutman 2007). Until now, it has been shown for Arabidopsis thaliana and Picea abies type II metacaspases that they are synthesized as inactive zymogens, and that they are activated by autoprocessing, similar to effector caspases from mammals (Vercammen et al. 2004; Bozhkov et al. 2005). In contrast, type I metacaspases from Arabidopsis do not autoprocess, and most likely similar to initiator mammalian caspases, they require oligomerization for activity (Vercammen et al. 2004). Metacaspases and animal caspases contain a conserved catalytic His/Cys dyad in their active site, with the Cys residue acting as a nucleophile for substrate peptide bond hydrolysis (Piszczek and Gutman 2007). The striking difference between all discovered metacaspases and caspases is the formers’ preference for Arg or Lys residues in their substrates (Vercammen et al. 2004; Bozhkov et al. 2005). Metacaspase activities can be modified by post-translational modification and through the action of inhibitors. The zymogen of metacaspase 9 of A. thaliana (AtMC9), but not its active form, is modified post-translationally by S-nitrosylation (Belenghi et al. 2007). The same modification can also influence caspase activities (Lai et al. 2011). Furthermore, the activity of AtMC9 can be inhibited by serpin-1, an inhibitor of serine proteases (Vercammen et al. 2006). Apart from serine proteases, some serpins also inhibit caspase-1, 8 and 10 (Ye and Goldsmith 2001). Because of the mode of its inhibitory action on AtMC9, serpin-1 is also the first identified natural metacaspase substrate (Vercammen et al. 2006). Recently, an animal caspase-3 substrate, tudor staphylococcal nuclease (TSN), was discovered to be a natural substrate of Picea abies metacaspase mcII-Pa during both developmental and stress-induced programmed cell death (PCD) (Sunström et al. 2009).
Similar to caspases, metacaspases are multifunctional proteins that take part in the regulation and the execution of PCD, cell cycle control, aging and oxidative stress (Tsiatsiani et al 2011). A striking example of a multifunctional metacaspase is YCA1 from budding yeast (Saccharomyces cerevisiae), which was found to be a positive regulator and active player in oxidative stress and senescence-associated PCD in addition to being a cell cycle controller (Madeo et al. 2004). Moreover, the participation of YCA1 in the clearance of insoluble protein aggregates during physiological and stress conditions provides a vital non-death cell function for metacaspases (Lee et al. 2010). Another metacaspase, mcII-Pa, is involved in developmental PCD during the early stages of embryogenesis from the gymnosperm plant Norway spruce (Picea abies) (Bozhkov et al. 2005). Interestingly, AtMC8 metacaspase from A. thaliana, was found to be upregulated during oxidative stress PCD pathways that were induced by UV-C, H2O2 and methyl viologen (He et al. 2008).
To date, very few plant metacaspases have been identified and well characterized. Most biochemical and functional findings described so far come from Arabidopsis and Picea abies studies (Vercammen et al. 2004; Bozhkov et al. 2005). Here, we provide a biochemical analysis of the cereal metacaspase from Triticum aestivum, TaeMCAII, including molecular modeling of its tertiary structure. TaeMCAII cDNA encodes a protein of 405 amino acids with a molecular mass of 44 kDa and an isoelectric point of 5.29 (Piszczek et al. 2011).
Materials and Methods
Plants and Growth Conditions
Spring wheat (Triticum aestivum L.) seedlings were grown in Hoagland nutrient solution for 2 weeks under controlled conditions (temperature of 20°C, 16 h photoperiod, irradiance of 260 μmol m−2 s−1 and a relative humidity of 70–80 %). The plants were subjected to heat shock at a temperature of 50°C for 20 min to evoke PCD. Wheat leaves were collected 3 h after stress and were frozen in liquid nitrogen and stored at −80°C until the isolation of total RNA according to the method of Chomczyński and Sacchi (1987).
Cloning of Type II Metacaspase Open Reading Frame from Triticum aestivum
cDNA was synthesized using a reverse transcriptase system (Promega, Madison, WI) from an RNA template obtained from 2-week-old spring wheat seedlings (Triticum aestivum L.) that were exposed to heat shock (50 °C) for 20 min. The cDNA and forward and reverse primers with sequences of 5′-CGCAACATTGGATCCATGGGCCGCAAGCTCGCGCTCCTGGTGGGCATC-3′ and 5′-GATATTGATCTCGAGTCAGCAGATGAAAGCCACATGAACATGCTCATC-3′, respectively, were used for the PCR reaction using the KOD Hot Start polymerase (Novagen). The PCR reaction was performed under the following conditions: 95°C at 2 min, 35 cycles of 20 s at 95°C, 10 s at 60°C and 30 s at 70°C. The PCR product was cloned into the bacterial expression vector pET-28a(+) (Novagen), which resulted in an N-terminal fusion with an His6 epitope tag.
Production of TaeMCAII in Escherichia coli
A plasmid vector harboring the open reading frame of TaeMCAII was transformed into Escherichia coli strain BL21(DE3)pLysS. Bacterial cultures were grown in LB medium supplemented with 50 μg ml−1 kanamycin and 50 μg ml−1 chloramphenicol at 37°C until the culture reached an optical density of 0.6–0.8 at 600 nm. Protein expression was induced by the addition of 1 mM isopropyl-thiogalactoside (IPTG) for 3 h. The bacteria were centrifuged at 5,000 rpm at 4°C for 15 min. The pellet was resuspended in protein extraction buffer that contained 20 mM Tris-HCl (pH 8.0), 0.5 M NaCl, 5 % glycerol, 0.1 % Triton X, 5 mM β-mercaptoethanol and protease inhibitor Complete (Roche, Mannheim, Germany) and was then sonicated. The bacterial lysate was centrifuged at 20,000 rpm for 30 min at 4°C, and the supernatant was mixed with 1 ml Ni-NTA agarose resin (Invitrogen, La Jolla, CA) for 1 h with the addition of 10 mM imidazole to bind the recombinant protein. The Ni-NTA agarose column containing bound protein was then washed with protein extraction buffer supplemented with 20 mM imidazole five times. The protein was eluted with the same buffer containing 300 mM imidazole and dialyzed overnight against MTSB buffer containing 20 mM PIPES (pH 7.0), 2 mM EGTA, 2 mM MgSO4, 10 % glycerol, 50 mM NaCl and 2 mM DTT.
Cloning and Expression of TaeMCAII Cys140 and Cys23 Mutants
To investigate the involvement of the two cysteine residues that make up the active site of TaeMCAII, two mutant forms, TaeMCAIIC140A and TaeMCAIIC23A, were cloned and expressed in E. coli strain BL21(DE3). The expression vector carrying an insert of the TaeMCAII sequence as a template and oligonucleotide primers with the appropriate Cys replaced by Ala were used in the PCR reactions. The following pairs of mutagenic primers were designed (mutagenic bases in bold): 5′-GTCTCCGACTCAGCGCACAGTGGTGGC-3′ and 5′-GCCACCACTGTGCGCTGAGTCGGAGAC-3′ for the TaeMCAIIC140A mutant, and 5′-GCCGAGCTCAAGGGAGCGCACAACGACGTTGAC-3′ and 5′-GTCAACGTCGTTGTGCGCTCCCTTGAGCTCGGC-3′ for the TaeMCAIIC23A mutant. The PCR reaction with Pfu Turbo Polymerase (Stratagene, La Jolla, CA) was performed under the following conditions: 95 °C for 30 s, 16 cycles of 30 s at 95 °C, 1 min at 55 °C and 5 min 30 s at 68 °C. Before the transformation of competent E. coli DH5α cells, the PCR products were treated with DpnI for 1 h at 37°C to digest the parental DNA. After the transformations, the plasmid vectors were isolated from single colonies and sequenced to confirm the mutations. The plasmid vectors with Cys140 or 23 replaced by Ala were used for the transformation of E. coli strain BL21(DE3), and the production of mutated proteins of TaeMCAII was induced by 1 mM IPTG. Cysteine mutants of TaeMCAII were expressed and purified as described above for wild-type TaeMCAII.
Activity Assays of Wild-Type and Mutant TaeMCAII
The optimized buffer for wild-type and mutant TaeMCAII activity consisted of 100 mM HEPES (pH 7.0), 10 % glycerol, 10 mM DTT, 30 mM CaCl2 and 0.1 % CHAPS. The following buffer was used for determination of the pH profile of wild-type TaeMCAII: 50 mM acetic acid, 50 mM MES and 100 mM Tris. All assays were performed with 50 μM substrate and with 50 μg recombinant protein in 200 μl reaction mixtures using a colorimetric method. Metacaspase substrates, acetyl-valyl-arginyl-prolyl-arginyl-p-nitroanilide (Ac-VRPR-pNa) and acetyl-isoleucyl-arginyl-seryl-lysyl-p-nitroanilide (Ac-IRSK-pNa), were obtained from Biosyntan (Berlin, Germany), and caspase substrate acetyl-valyl-glutamyl-isoleucyl-aspartyl-p-nitroanilide (Ac-VEID-pNa) was purchased from Bachem (Bubendorf, Switzerland). The amount of hydrolyzed substrate was measured spectrophotometrically at 405 nm. For inhibition experiments, wild-type TaeMCAII was incubated with synthetic inhibitors, i.e., PMSF (1 mM), 1-trans-epoxysuccinylleucylamide-(4-guanido)-butane (E-64, 100 μM), chymostatin (100 μM), Nα-tosyl-l-lysine-chloromethyl ketone (TLCK, 100 μM) and leupeptin (10 μM) for 1 h at 30°C prior to the addition of the substrate. All tested inhibitors were purchased from Sigma-Aldrich (St. Louis, MO).
Structure Prediction and Sequence Analysis
Structural Models of TaeMCAII
Sequence alignments between TaeMCAII, human caspase 7 (PDB 1F1J) and Geobacter sulfurreducens unknown protein GSU0718 were produced by the FFAS03 structure prediction method (Rychlewski et al. 2000; Jaroszewski et al. 2005) and were adjusted manually to accommodate the predicted secondary structures. Three-dimensional structure models were constructed using the program MODELLER (Sali and Blundell 1993; Eswar et al. 2006) with the combined Modeller9v8 Python scripts (automodel mode), model-multichain and model-ligand. Out of the models that were presented by MODELLER, the one with the most favorable molpdf score was selected for further analysis. The MetaMQAP server (Pawłowski et al. 2008) was used to estimate the correctness of the 3D models using a number of model quality assessment methods in a meta-analysis.
Bioinformatics Identification of Potential Peptide Ligands for Modeled Structure of TaeMCAII
The list of oligopeptides as caspase and metacaspase ligands tested using bioinformatics tools
The set of potential caspase and metacaspase ligands for testing were selected based on literature data (Vercammen et al. 2006) and were prepared starting from raw amino acids using Maestro tools and the MacroModel 9.5 conformational search algorithm. More than 500 candidate ligand structures were obtained, which were screened and docked in the active site (substrate binding groove) of the modeled structure of TaeMCAII. All docking poses were scored, and the best scoring poses for each oligopeptide were saved.
Autoprocessing of TaeMCAII Depends on Cys140
TaeMCAII preferentially cleaves peptide bonds after arginine and to a lesser extent after lysine at the P1 position of the substrate.
Influence of Protease Inhibitors on TaeMCAII Activity
TaeMCAII Requires a High Concentration of Ca+2 for Activity
Millimolar ranges of Ca+2 concentrations were required for TaeMCAII to cleave its optimal peptide substrate, Ac-VRPR-pNa. Without calcium, TaeMCAII exhibited only 2 % of its optimal activity, which was reached in the presence of 30 mM Ca+2. A slight decrease in substrate hydrolysis by TaeMCAII was observed when this optimal calcium level was exceeded (Fig. 5a).
Structure Modeling and Analysis
Determining the appropriate template is crucial to obtaining accurate data from protein modeling. Plant metacaspases show rather moderate sequence similarity to animal caspases, and only for caspases can one find experimentally resolved structures in the Protein DataBank. However, poor similarity did not disturb the construction of plant metacaspase structural models based on human caspase-7 templates (Belenghi et al. 2007).
Attempting to obtain an accurate model of the newly sequenced Triticum aestivum type II metacaspase, we also started with the caspase-7 template, but decided to select another structure as the modeling template after running fold predictions methods. The chosen PDB record contained the structure of an unknown Geobacter sulfurreducens protein, described as GSU0716 (Northeast Structural Genomics target GsR13). This was indicated by almost all available fold prediction algorithms, including COMA, COMPASS, Phyre, FFAS03 and HHsearch, as the best scoring hit for the TaeMCAII sequence. Moreover, unknown protein GSU0716 has close homologs within the group of the bacterial members of the peptidase_C14 pfam00656 family and shows homology to plant metacaspases with e-values in the range of 7e−10 to 2e−09. Even a preliminary BLAST search followed by hmmpfam and HHsearch analysis revealed the presence of Pfam Peptidase C1 domain PF00656, which is called the caspase domain. After analysis of the profile–profile alignments of TaeMCAII and GSU0716 (based on the results of the FFAS03 method) (Fig. 6), we observed two highly conserved regions between TaeMCAII and GSU0716 near the catalytic dyad His/Cys (5- and 7-amino-acid-long motifs: YSGHG and SDSCHSG), which are also present in another plant metacaspases and aligned bacterial metacaspase-like sequences (Fig. 1), in addition to high conformity of the secondary structure assignments and predictions for the TaeMCAII and GSU0716 sequences.
However, the lack of exposure of key catalytic residues in the hypothetical binding groove could discourage the use of the 3BIJ structure as a metacaspase-like modeling template. Looking closer at the molecular surface of the GSU0716 putative substrate binding, we noticed that four presumably catalytic residues (Cys135, His84, Gly85 and Asp133 in GSU0716 numbering) were hidden or were not sufficiently exposed to make contact with the potential substrate. In the original GSU0716 structure, only His84 from the putative catalytic dyad His-Cys was partially exposed. The catalytic cleft was blocked by a short oligopeptide, IRYRA, that covered the surfaces of Cys135, Gly85 and Asp133 residues. However, after removing this fragment, the GSU0716 catalytic residues in the binding groove were exposed to make contact with the potential substrate.
The obtained Glide Scores (the general measure used for the evaluation of a given ligand pose in receptor binding sites) for the set of oligopeptides that were suspected to bind to the type II metacaspase active site indicated that the best potential ligands for the TaeMCAII enzyme were oligopeptides VRPR and IRSK, as these peptides resulted in the highest scores.
Very few plant metacaspases have been identified and well characterized (Vercammen et al. 2004, 2006; Bozhkov et al. 2005; Sunström et al 2009). This study describes the biochemical characterization and the results of homology modeling of a cereal type II metacaspase from wheat (Triticum aestivum), TaeMCAII. Modeling was based on a bacterial template, as there are no experimentally resolved tertiary structures of plant metacaspases in the Protein Databank. The reasons for choosing this specific template are described in the Results section.
Members of the CD clan of proteases, including metacaspases, are known to recognize the residues at the N-terminal side of the scissile bond, named the P1 residue, in their substrates (Tsiatsiani et al. 2011). Metacaspases, like caspases, recognize three or more amino acids residues N-terminal to the P1 position. This position is acidic in the case of caspase substrates and is basic for metacaspase substrates (Earnshaw et al. 1999). More in vitro evidence was provided here in that metacaspases cleave their substrates preferentially after the basic amino acids Arg or Lys (Fig. 3a). Like other known metacaspases, e.g., AtMC9 and mcIIPa, TaeMCAII exhibits much greater activity towards substrates that contain Arg at the P1 position than Lys (Fig. 3a). According to the homology model, which was based on bacterial template GSU0716, negatively charged residues (Asp138, Ser137, Asp26, Asp202, Thr199, Ser210, Asp78 and Ser215) near the TaeMCAII active site can facilitate the binding of positively charged ligands, i.e., peptides enriched in Arg or Lys (Fig. 7b). The substrate specificity of TaeMCAII towards Arg in the P1 position was also confirmed by inhibition studies. TaeMCAII activity was blocked completely by an arginyl inhibitor, leupeptin, and TLCK and chymostatin also sufficiently inhibited its activity (Fig. 5b). Nevertheless, MS/MS analysis revealed the autoprocessing sites of TaeMCAII as the C-terminal sides of Lys 148 and Arg 214, removing a linker region of 7,457 Da and separating two caspase-like subunits (Fig. 2a, b).
Up to now, all well-characterized metacaspases have had larger N-terminal subunits than C-terminal subunits, and correspond to p-20 and p-10 caspase subunits, respectively (Vercammen et al. 2004; Bozhkov et al. 2005). In contrast, the TaeMCAII N-terminal subunit was smaller than the C-terminal subunit (Fig. 2a). It should be noted that, similar to caspases and other known metacaspases, the catalytic dyad Cys140 and His87 of TaeMCAII are located in the N-terminal subunit (Fig. 6). The overproduction of TaeMCAII in E. coli resulted in its concomitant autoprocessing depending on the catalytic Cys140. This autoprocessing was completely abolished in the TaeMCAIIC140A mutant (Fig. 2a,b). This mutant form of TaeMCAII was also found to be inactive (Fig. 4). Processing is necessary for the activation of the zymogens of all known type II metacaspases and executioner caspases of animals (Tsiatsiani et al. 2011). Multi-sequence alignment (MSA) and a homology model confirmed the significance of Cys140 in the TaeMCAII active site (Figs. 1, 8). In bacterial and plant metacaspases, including TaeMCAII, three residues before the catalytic histidine (Tyr, Ser and Gly) and six residues surrounding the catalytic Cys140 (Ser, Asp, Ser and His, Ser and Gly) are strictly conserved (Fig. 1). These conserved motifs appeared to be specific only for plant metacaspases and bacterial metacaspase-like sequences, but are not observed in animal caspases (data not shown). MSA of the amino acid sequences of known plant metacaspases and TaeMCAII revealed the existence of a second highly conserved cysteine residue (Cys23 in TaeMCAII ) that is involved in catalytic process in metacaspases (Fig. 1). The obtained homology model of TaeMCAII based on the bacterial putative metacaspase-like protein helped explain the role of Cys23 in the catalytic process (Fig. 7b). According to this model, two conserved cysteines near the catalytic His87, Cys23 and Cys140 may participate in the nucleophile attack. The distances between the His87 ring imidazole N atom and the Cys thiol groups, crucial for the reaction mechanism, are 7.31 Å (Cys140) and 7.28 Å (Cys23). Both cysteines can donate a proton to the histidine ring (Fig. 8). Docking results show that the peptide substrate was bound in a position that enables both Cys23 and 140 to exchange a proton (Fig. 8).
As revealed by analysis of the TaeMCAIIC23A mutant, Cys23 is not required for activation of the wheat metacaspase, but is needed for its activity. The proper autoprocessing was found in the TaeMCAIIC23A mutant, but it exhibited very low activity compared to the wild-type protein (Figs. 2, 4). The implied functions of two corresponding catalytic Cys residues during autoprocessing and enzyme activity were also confirmed during the studies with AtMC9 (Belenghi et al. 2007). The biochemical studies and modeling results described here indicate the importance of both cysteines for the activity of TaeMCAII, but Cys140 is required for the autoactivation process.
On the basis of TaeMCAII’s neutral pH preference, one might suggest that TaeMCAII is a cytoplasmic proteinase (Fig. 3b). The majority of metacaspases that have been studied so far, including Arabidopsis metacaspases, (AtMC4 and AtMC9), Picea abies mcIIPa, Leishmania (LdMC1and LdMC2) and Trypanosoma (TbMCA2) metacaspases, require neutral or slightly basic pH for optimal activity in vitro (Vercammen et al. 2004; Bozhkov et al. 2005; Moss et al. 2007; Lee et al. 2007). To date, AtMC9 is the only known metacaspase that requires an acidic environment for its activation and action (Vercammen et al. 2004). Until now, an organelle localization was found only for Arabidopsis metacaspase AtMCP1b, and it was found in chloroplasts (Castillo-Olamendi et al. 2007). It would be useful to know in which organelles metacaspases are active in order to suggest their possible roles in signaling pathways in plant cells. It is interesting that TaeMCAII, similar to Arabidopsis and Picea abies metacaspases, had highest activity at millimolar levels of Ca+2 ions (Fig. 5a) (Vercammen et al. 2004; Bozhkov et al. 2005). Recently, it was demonstrated that the AtMC4 autoproteolytic process is Ca+2-dependent (Watanabe and Lam 2011). It should be noted that under many stress conditions, e.g., salinity, cold and heat stress, the concentration of cytosolic calcium increases dramatically (Xiong et al. 2002; Kang et al. 2011). It is also known that high concentrations of calcium ions is toxic for plant organelles; thus, there are some specific channels/pumps that regulate the movement of Ca+2 in and out of cells and organelles (Xiong et al. 2006). Some plant organelles are more tolerant of high levels of this ion than others. The highest concentration of Ca+2 ions, exceeding 50 mM, is found in the endoplasmic reticulum (ER). Plastids, mitochondria and vacuoles contain millimolar levels of Ca+2 but much less than the ER (Hepler 2005). The role of calcium ions in stress signaling in plants and in plant PCD has been demonstrated previously (Xiong et al. 2006; Hepler 2005). It should be noted that the very high levels of Ca+2 ions required for optimal activity of the wheat metacaspase TaeMCAII and other plant metacaspases, e.g., AtMC4 and mcIIPa, may play a significant role in the adaptation and acclimatization of these plants to stressed environmental conditions. The aim of our future investigations will be the localization of TaeMCAII in subcellular organelles and in vivo measurements of its catalytic activity under various stress conditions, followed by elevated cellular Ca+2 concentrations. The results should be highly relevant to agriculture, as they indicate possible signaling pathways that influence the acclimation of cereal plants to stress conditions.
This study was supported by the National Science Centre (grant number: N303 048837).
This article is distributed under the terms of the Creative Commons Attribution License which permits any use, distribution, and reproduction in any medium, provided the original author(s) and the source are credited.
- Bozhkov PV, Suarez MF, Filonova LH, Daniel G, Zamyatnin AA, Rodriguez-Nieto S, Zhivotovsky B, Smertenko A (2005) Cysteine protease mcII-Pa executes programmed cell death during plant embryogenesis. Plant Biol 102:14463–14468Google Scholar
- Eswar N, Marti-Renom MA, Webb B, Madhusudhan MS, Eramian D, Shen M, Pieper U, Sali A (2006) Comparative protein structure modeling with MODELLER. Current Protocols in Bioinformatics. Wiley, New York, Supplement 15:5.6.1–5.6.30Google Scholar
- Sunström JF, Vaculova A, Smertenko AP, Savenkov EI, Golovko A, Minina E, Tiwari BS, Rodriguez-Nieto S, Zamyatnin AA, Välineva T, Saarikettu J, Frilander MJ, Suarez MF, Zavialov A, Stähl U, Hussey PJ, Silvennoinen O, Sundberg E, Zhivotovsky B, Bozhkov PV (2009) Tudor staphylococcal nuclease is an evolutionarily conserved component of the programmed cell death degradome. Nat Cell Biol 11:1347–1354CrossRefGoogle Scholar
- Vercammen D, Van de Cotte B, De Jaeger G, Eeckhout D, Casteels P, Vanepoele K, Vandenberghe I, Van Beeumen J, Inze D, Van Breusegem F (2004) Type II metacaspases Atmc4 and Atmc9 of Arabidopsis thaliana cleave substrates after arginine and lysine. J Biol Chem 279:45329–45336PubMedCrossRefGoogle Scholar