The Role of Proteases in Plant Development

Maribel García-Lorenzo

Department of Chemistry, Umeå University Umeå 2007

i Department of Chemistry Umeå University SE - 901 87 Umeå, Sweden

Copyright © 2007 by Maribel García-Lorenzo ISBN: 978-91-7264-422-9 Printed in Sweden by VMC-KBC Umeå University, Umeå 2007

ii Organization Document name UMEÅ UNIVERSITY DOCTORAL DISSERTATION Department of Chemistry SE - 901 87 Umeå, Sweden Date of issue October 2007 Author Maribel García-Lorenzo

Title The Role of Proteases in Plant Development.

Abstract Proteases play key roles in plants, maintaining strict protein quality control and degrading specific sets of proteins in response to diverse environmental and developmental stimuli. Similarities and differences between the proteases expressed in different species may give valuable insights into their physiological roles and evolution. Systematic comparative analysis of the available sequenced genomes of two model organisms led to the identification of an increasing number of protease genes, giving insights about protein sequences that are conserved in the different species, and thus are likely to have common functions in them and the acquisition of new genes, elucidate issues concerning non-functionalization, neofunctionalization and subfunctionalization.

The involvement of proteases in senescence and PCD was investigated. While PCD in woody tissues shows the importance of vacuole proteases in the process, the senescence in leaves demonstrate to be a slower and more ordered mechanism starting in the chloroplast where the proteases there localized become important. The light-harvesting complex of Photosystem II is very susceptible to protease attack during leaf senescence. We were able to show that a metallo-protease belonging to the FtsH family is involved on the process in vitro. Arabidopsis knockout mutants confirmed the function of FtsH6 in vivo.

Key words: Comparative genomics, protease, PCD, leaf senescence, FtsH, Arabidopsis, Populus Language: English ISBN: 978-91-7264-422-9 Number of pages: 45 + 5 papers

iii

En este altar antiguo que levanté a lo alto de mis horas quiero subir, como polen nuevo me quiero esparcir, en un total abandono. Candiles de aceite habrá que encender, Sin llaves y a las puertas del instante estoy.

“Sin llaves” Manolo García & Quimi Portet

On this ancient altar that I have built at the summit of my days I yearn to climb, wanting to fling myself like new pollen, scattered in total abandon. Oil lamps shall be lit, Without keys, at the doors of instant I stand.

“Without keys” Manolo García & Quimi Portet

iv Table of Contents

1. List of Papers...... 2 2. Abbreviations...... 3 3. Introduction...... 4 4. Model organisms and comparative studies ...... 6 4.1 Cyanobacteria, model organisms for chloroplast studies...... 7 4.2 Comparative Genomics, a method to understand evolution...... 8 4.3 Genome duplication, the force that drives evolution ...... 8 4.4 Phylogenetic trees: picturing evolution...... 10 5. Plant Proteases ...... 11 5.1 Cytosolic proteases...... 11 5.1.1 Proteasome (ThrPro) ...... 11 5.1.2 Metacaspases (C14) ...... 13 5.2 Endoplasmic reticulum (ER) and ER-derived compartments ...... 14 5.3 Vacuolar proteases ...... 15 5.4 Mitochondria...... 17 5.4.1 ATP-independent proteases...... 18 5.4.2 ATP-dependent proteases...... 20 5.5 Chloroplast...... 22 5.5.1 ATP-independent proteases...... 22 5.5.2 ATP-dependent proteases...... 23 6. Proteases involved in senescence and PCD...... 26 6.1 Proteases during plant development (Paper IV)...... 26 6.2 Proteases involved in wood formation (Paper I)...... 29 6.3 FtsH proteases involved in the degradation of the photosynthetic light-harvesting antenna (Papers II and III) ...... 30 7. Conclusions and future perspectives ...... 32 8. Acknowledgements ...... 34 9. Reference list ...... 36

1 1. List of Papers

The thesis is based on the following publications listed below and will be referred to in the text by their corresponding Roman numerals

I. C. Moreau, N. Aksenov, M García-Lorenzo, B. Segerman, C. Funk, P. Nilsson, S. Jansson, H. Tuominen A genomic approach to investigate developmental cell death in woody tissues of Populus trees. Genome Biol. 2005; 6(4):R34

II. A. Zelisko, M. García-Lorenzo, G. Jackowski, S. Jansson, C. Funk AtFtsH6 is involved in the degradation of the light-harvesting complex II during high-light acclimation and senescence. Proc Natl Acad Sci U S A. 2005; 102(38):13699-704

III. M. García-Lorenzo, A. Zelisko, G. Jackowski, C. Funk Degradation of the main Photosystem II light-harvesting complex. Photochem Photobiol Sci. 2005; 4(12):1065-71

IV. M. García-Lorenzo, A. Sjödin, S. Jansson, C. Funk Protease gene families in Populus and Arabidopsis. BMC Plant Biol. 2006; 20,6:30

V. M. García-Lorenzo, A. Pruzinska, C. Funk ATP-dependent proteases in the chloroplast. Eva Kutejova (ed.) ATP-dependent proteases, Research Signpost, Kerala, India, accepted for publication

2 2. Abbreviations

AAA ATPase associated with various activites AspPro Aspartic protease CP Proteasome core particle CtpA C-terminal processing peptidase CysPro Cystein protease EGY Ethylene-dependent gravitrospism-deficient and yellow green ER Endoplasmic reticulum EST Expressed sequence tag HL High light HR Hypersensitive response Htr High temperature requirement I-CLiP Intra-membrane cleaving proteases IM Inner membrane IMP Inner membrane peptidase LHC Light Harvesting Complex LV Lytic vacuoles MetalloPro Metalloprotease MIP Mitochondrial intermediate protease MPP Mitochondrial processing peptidase OM Outer membrane Oma Overlapping with m-AAA protease PCD Programmed cell death PreP Presequence protease PSII Photosystem II PSV Protein storage vacuoles RIP Regulated intramembrane proteolysis ROS Reactive oxygen species RP Proteasome regulatory particle SAG Senescence associated gene SerPro Serine protease SPP Stromal processing peptidase TE Tracheary element ThrPro Threonine protease TIM Translocase of the inner membrane TOM Translocase of the outer membrane TPP Thylakoid processing peptidase Ub Ubiquitin VPE Vacuolar processing enzyme

3 3. Introduction Proteolysis is the mechanism by which the degradation of a protein occurs. This process is ruled by hydrolytic enzymes called proteases. Proteases are found in every cellular compartment. Their working mechanism varies very much among all families and groups of proteases. Some of them work on their own, some of them in cooperation with other proteases (e.g. cascades) and some of them form complexes and constitute an active proteolytic machine.

The cleavage of peptide bonds is not limited to the degradation of mature proteins to free amino acids, it also is relevant for modification and maturation of proteins. Therefore the proteolysis performed by proteases can be divided in two categories (Figure 1): 1. limited proteolysis: in which just some amino acids or a minor part of the protein are detached 2. unlimited proteolysis: in which a protein is totally degraded.

Limited proteolysis usually modifies proteins post-translationally at highly specific sites. This process is ruled by the so called processing peptidases and the purpose of the limited proteolysis is the activation and maturation of proteins or removal of signal or transit peptides (e.g. under subcellular targeting). Unlimited proteolysis results in the total degradation of damaged, misfolded and potentially harmful proteins, providing free amino acids that will be recycled for the synthesis of new proteins. The proteases performing this kind of proteolysis are considered housekeeping proteases; they “clean” the compartment of malfunctional proteins, maintain stoichiometric amounts of protein subunits or control metabolic pathways. Limited and unlimited proteolysis usually is performed by different proteases using different regulatory mechanisms [1]. Proteases are involved in key aspects of plant growth, development and defense, as well as in senescence and plant cell death, the ultimate fate of cells, organs or the whole organism. Changes in light quality and intensity, temperature, nutrient supply, drought stress and pathogen or herbivore attack generally lead to a reorganization of cell morphology, metabolism and even membrane structure. These modifications are the result of

4 intracellular responses including regulation of gene expression, changes in the rate of protein synthesis and degradation of altered, damaged or temporarily functional proteins [2].

A. + +

B. +

Figure 1.- Types of proteolysis. A. Limited proteolysis. B. Unlimited proteolysis

While proteases are involved in most cellular processes [3] proteolysis still is not a well understood process in plant biology. Very little is known about substrate specificity, physiological roles or cellular location of many of the putative proteases known or the processes they are involved in.

Putative proteases can be classified in exoproteases, acting on the termini of polypeptide chains, either carboxi- or aminopeptidases, and endopeptidases, cleaving in the interior of the polypeptide. Moreover they can be classified according to their catalytical site. This classification can be found at the MEROPS database [4]. In the following text I will use the names of these protease gene families:

1. Serine Proteases (SerPro) 2. Cystein Proteases (CysPro) 3. Aspartic Proteases (AspPro) 4. Metalloproteases (MetalloPro) 5. Threonine Proteases (ThrPro)

SerPro are characterized by the presence of a serine residue in the catalytical triad, His-Asp-Ser, at the active site of the enzyme. During protein hydrolysis an intermediate is formed between the substrate and the essential serine. CysPro form a covalent intermediate involving Cys

5 and His residues from the catalytical dyad. An additional Asn is required at the active site for the proper orientation of the His side chain [5]. The catalytical dyad of AspPro is formed by two highly conserved aspartates, which posses optimal activity at acidic pH. Metallo-Pro need a metal ion such as Zn2+, Mg2+ or Ca2+ at their active site, and ThrPro finally posses an N-terminal threonine at each catalytical site. This group of proteases is found exclusively within the biggest protease machine inside the cell: the proteasome.

4. Model organisms and comparative studies

The increasing amount of sequenced genomes provides a raising number of possible model organisms, which can be used for specialized research interests.

Table 1. Comparison of the genome sizes of different model organisms. Species Genome Size (Mbp) Escherichia coli (prokaryote) 4,6 Synechocystis sp. PCC 6803 (cyanobacterium) 3,5 Oryza sativa (angiosperm, monocotyledonous, plant) 430-450 Arabidopsis thaliana (angiosperm, dicotyledonous plant) 100-150 Populus sp., Salix (angiosperm, dictyledonous, trees) 450-550 Pinus sp. (gymnosperm, tree) 20000

The annual plant Arabidopsis thaliana (hereafter Arabidopsis) is the best- studied organism in plant science [6]; its genome was the first plant genome being sequenced. In nature Arabidopsis is nothing but a weed, and, as a weed, it is widespread. Its short life span and the possibility to transform this plant with Agrobacterium make it a perfect candidate for genomics studies. Since this model plant appeared in plant biological studies, the understanding of many processes also in plant biochemistry, physiology or ecology has improved very much.

Having genomic information of different annual plants, it was reasonable to await the sequencing of a tree genome: Populus has been completely

6 sequenced recently [7]. Trees do not constitute one unique clade within the evolutionary history of land plants, they are a polyphyletic group. Furthermore, woody plants have arisen many times during land plant evolution, so there are trees present in almost every plant phylogenetic group. Arabidopsis is a close relative to Populus and Salix, they are classified in the same group of Rosids [8], even closer related to each other, than to many other dicotyledonous annual plants. The phylogenetic proximity between Arabidopsis and Populus [8], the relatively small genome and the possibility of vegetative propagation of clone collections for experimentation made Populus a perfect candidate for becoming a model species within the woody plants. The major limitation of working with trees is their long life span and long generation time (they require long time before they start flowering). Populus, further more, is dioceous, making selfing and back-cross manipulation impossible. However, the aforementioned close relationship between Populus and Arabidopsis favors comparative functional studies as well as comparative genomics and has the potential to greatly facilitate studies on genome and gene family evolution.

4.1 Cyanobacteria, model organisms for chloroplast studies Cyanobacteria are perfect candidates to study plant evolution. Cyanobacteria are photosynthetic prokaryotes, having in their genome all the coding material for the photosynthetic apparatus, and therefore are very useful to study the photosynthetic process [9-12]. These ancient organisms have survived in very different conditions and environments and are believed to be the ancient origin of chloroplasts. It is accepted within the scientific community that plastids derive from a primary endosymbiotic event in which a non-photosynthetic eukaryote engulfed an ancient cyanobacterium. During evolution genes of the pre-plastids were either retained, lost or transferred to the nucleus. This vertical transfer of typical plastid genes to the nuclear genome includes genes coding for proteins with important plastid functions, like several photosynthetic proteins. Today more than 5000 nuclear encoded gene products are targeted to the chloroplast [10, 11]; while plastidic genomes encode only around 200 proteins (depending on the organism); a cyanobacterium´s prokaryotic chromosome encodes

7 over 3000 open reading frames. The process of gene transfer to the nucleus might have involved duplication of each plastid gene; after transfer the nuclear copy of the gene possibly also produced the functional product. The redundant proteins in the cytosol acquired then new functions, or, were - with appropriate targeting sequences - directed to other compartments [10, 11].

4.2 Comparative Genomics, a method to understand evolution The organization of genes within a genome is highly conserved over long evolutionary periods. Thus, the availability of an increasing number of sequenced genomes gives us the opportunity to understand genetic relationships between different species. A large-scale, multigene-based comparison between two species can give hints about differences in the evolutionary history and even about protein function in these species due to orthology.

Information resulting from the genome sequencing projects is available via internet in databases with various specificity concerning organisms or protein classes. However, different databases unfortunately only contain limited consensus regarding the annotation, gene names or assigned functions [13]. Additionally, similar genes not necessarily have to be paralogs [14]. Therefore certain routines have to be used when doing a comparative genomics approach, such as the search for protein domains or signatures typical for the group of proteins to study. By comparing a gene in different genomes we have the possibility to observe, how selection has worked upon it; mutations, gene duplications or gene transfer from organelles to the nucleus are just a few happenings that rule evolution and speciation.

4.3 Genome duplication, the force that drives evolution Gene duplication is considered to be the most important step for the origin of genetic novelties [13], including new gene functions and expression patterns [15]. The strongest evidence that gene duplication has occurred and has major importance during evolution is the widespread

8 existence of gene families [16]. Gene duplications can involve either only part of a gene or the whole genome [13].

The evolutionary pathway after gene duplication from complete redundancy of the new gene copy to two functional, maintained products is still not understood [15]. Most duplicated genes undergo a period of relaxed selection early in their history until they either are eliminated or fixated [15]. These new genes generated after the duplication then can diverge from the parental gene function (Figure 2). One of the copies can acquire a novel and beneficial function (neofunctionalization) that may be preserved by natural selection; or it may be differently expressed, performing the same function but at a different site of the cell or in a different organ of the organism (subfunctionalization). It also might just be silenced by degenerative mutations resulting in a pseudogene [15, 17].

Gene Subfunctionlization Duplication

Neofunctionalization

Duplication Nonfunctionalization eliminated Figure 2.- Fate of duplicated genes. The majority of the gene duplications become eliminated, but when the new gene product is fixated both gene copies can experience different expression (subfunctionalization). One of both copies can acquire a new function (neofunctionalization), or one of the genes can be inactivated (nonfunctionalization), resulting in a pseudogene. White boxes indicate loss of function.

9

The duplicated genes may undergo mutations during evolution forming clusters of similar genes belonging to the same functional class [18]. Homologous genes are genes that share a common ancestor. Two types of homologous genes are differentiated (Figure 3): 1. paralogs: similar genes in the same species resulting from a gene duplication 2. orthologs: homologous genes in different species that have the same function.

Ancestral Gene

Duplication

T Paralogous genes i m Speciation e

Orthologous genes

Figure 3.- Gene homology. An ancestral gene undergoes gene duplication resulting in two paralog genes. After a speciation event, both genes with same origin conserving the same function are called orthologs.

To proof if two genes indeed are paralogs, a comparison of the sequences at protein level is most appropriate [13], as it neglects silent mutations in the coding regions. For this reason we compared protein sequences in our studies.

4.4 Phylogenetic trees: picturing evolution Phylogenetic trees illustrate a hypothetical evolutionary relationship among the genes, proteins and/or organisms that are included in the study. Collected sequences from the different databases first have to be aligned. In this procedure either whole gene/protein sequences can be

10 used or certain conserved domains. For the phylogenetic analysis various programs are available that are based on different assumptions for calculating the tree. Usually similar tree topologies can be obtained when using different matrixes, but still one should keep in mind that the results are only hypothetical.

5. Plant Proteases

Proteolytic enzymes are involved, directly or indirectly, in every aspect of plant physiology and development [19]. The removal of signal or transit peptides or the degradation of damaged, misfolded or unassembled polypetides form part of the quality control of a cell. The attack of proteases to proteins is thought to be highly specific. Their substrate recognition depends mainly on the localization of the protease, its substrate specificity and the localization of the target protein. Substrate specificity is defined by structural properties of the active site of the protease [20]. However, localization of substrate and protease are factors that should not be neglected. Therefore this overview about plant proteases is structured according to the cell compartments. Many substrates or protease functions are still not identified; moreover, the cross-talk among the different cell compartments is mainly unknown. Still the reader might get an idea about the high degree of regulation that is needed by the cell to develop or respond to environmental stimuli.

5.1 Cytosolic proteases The cytosol is the internal fluid of a cell. Besides accommodating the different organelles, a communication between the different compartments has to be established and many metabolic reactions happen here.

5.1.1 Proteasome (ThrPro) The proteasome is described as being responsible for most of the cytosolic and nuclear protein degradation. This protein complex is found in the cytosol and nucleus of almost all eukaryotic cells [21]. The

11 structure of the 26S proteasome is highly conserved in eukaryotes and it is formed by two subcomplexes (Figure 4): 1. a core particle (CP), 20S proteasome, forming a barrel-shaped structure responsible for the proteolysis of the substrate. This CP is formed by 4 rings. In the two inner rings, called
-rings, reside the catalytical activity. The two outer rings are called -rings and work as gates for the entrance of the substrate to the hydrolytic chamber [21]; 2. the regulatory particle (RP), 19S proteasome, formed by a lid and base. The base interacts with the -rings of the 20S proteasome and uses ATP to unfold the substrate and to pull it into the 20S chamber. The lid confers the specificity [22, 23].

Most of the protein subunits that form the proteasome are encoded by gene pairs, causing structural heterogeneity in the complex. Moreover, it has been observed in Arabidopsis that the 20S proteasome is proteolytical active and degrades proteins in an ubiquitin (Ub) and ATP independent manner under moderate oxidative stress [21, 24, 25]. Furthermore, the possibility of hybrid proteasome formation has been described, in which some of the RP subunits could be replaced by other chaperones [26, 27]. Most of the proteins degraded by the proteasome follow the Ub pathway in which the proteins are targeted for degradation by a polyUb chain. Misfolded or damaged proteins that lost their tertiary structures follow this degradation pathway. That is why the proteasome is known as one of the contributors for cell quality control [28]. Optimal levels of proteasome activity are required for plant development [21]. Downregulation of the gene expression of proteasomal subunits has been shown in different animal and plant studies. It seems that under these conditions other proteases have the major proteolytic activity, such as caspase-like proteins (see below). Plant cells seem to have many proteasome 20S inhibitors [29, 30]. Most probably these inhibitors are stored in vacuoles and will be released to the cytoplasm under hypersensitive response (HR) or programmed cell death (PCD) accelerating cell death processes [21].

12 ATP- and Ubiquitin dependent pathway

UBP

ATP + +

Ub-Targeted protein Peptides Ub

19S 26S

ring

ATP- and Ubiquitin independent pathway ß rings

ring

20S

+

20S Target protein Peptides

Figure 4.- Possible degradation pathways mediated by the proteasome in plants. In the ATP and Ub dependent pathway, the targeted protein with polyUb tail is recognized by the lid of the 26S proteasome. After proteolysis a specific Ub- protease (UBP) degrades the polyUb tail into Ub subunits that will be recycled in the targeting of other proteins. The ATP- and Ub independent pathway is ruled by the 20S proteasome, substrate proteins are not targeted with the polyUb tail and ATP is not required.

5.1.2 Metacaspases (C14) Metacaspases constitute, together with the VPEs (vacuolar processing enzymes) (C13) and saspases (S8), the group of caspase-like proteases in plants. Metacaspases are cytosolic plant proteins, whose primary and tertiary structures resemble animal caspases [31]. Although they do not hydrolyze animal caspase substrates, they are also CysPro and are involved in PCD. Their autocatalytic activation and the possible cascade mechanisms in which they are involved suggest a similar function in plant and animal cells [32]. The increased expression of metacaspase genes under different stress conditions in different plant species implies proteolysis in the cytosol during PCD [33, 34], [35] and references therein.

13 5.2 Endoplasmic reticulum (ER) and ER-derived compartments The ER is formed by an extensive network of membranous tubules and cisternae. Almost all of the proteins and lipids of the endomembrane cell system are produced in the ER membrane. Moreover, the ER is the organelle where most secretory proteins are folded for subsequent delivery to their site of action [36]. The ER quality control system monitors misfolded or unassembled proteins preventing them from reaching their final destination. These aberrant proteins will be extracted from the ER and tagged by a polyUb tail for degradation by the proteasome in the cytosol [37]. ER is the major component of the secretory pathway. Vesicles are secreted from the ER to transport proteins to their final destinations. Some of these vesicles contain hydrolytic enzymes that will be active in other organelles: 1. KDEL vesicles accumulate CysPro having this KDEL-ER retention signal. They might be responsible for protein mobilization in graminea seed germination and they conduct their content to the vacuole (see below) [38, 39]. 2. Ricinosomes were discovered first in the endosperm of castor bean (Ricinus communis), but later described also in other species and other senescing tissues (see [40] for review). They accumulate KDEL tailed-CysPro belonging to the family of papain-like (C1) proteases that are involved in PCD of the senescing endosperm [41, 42]. Preproteases in these vesicles become active after destabilization of the tonoplast after lowering of the pH value. They are able of autocatalysis to reach the most active stage [40] and are therefore also called “suicide bombs”: under senescing conditions the membrane destabilizes and liberates papain-like proteases that are able to perform unlimited proteolysis of the cytoplasmic components. The products will then be recycled in the surviving parts of the plant [40]. 3. ER bodies are the largest vesicles derived from the ER [36]. They are most prevalent in epidermal cells of seedlings and hypocotyls after germination and disappear right before the senescence of these tissues. They were also observed in root cells of Arabidopsis [43], which are most sensitive to environmental stress in plants

14 [39, 44]. In rosette leaves of Arabidopsis they were only detected after wounding, never at normal conditions. These bodies accumulate the proteases RD21 (C1) and TVPE (C13). KDEL- tailed U-glucosidase (PYK10) is the major component in ER bodies; U-glucosidases produce toxic compounds to pathogens [43, 45].

5.3 Vacuolar proteases One plant cell structure essential for protein metabolism is the vacuole. Many proteases are found in the lumen of plant vacuoles [46, 47]. Protein degradation may happen directly after the protein has entered the lumen of the vacuole or the vacuole may act as a short- or long-time storage of proteins. The amino acids resulting from this degradation will be recycled for metabolic processes outside of the vacuole. The origin of the vacuoles is uncertain. It has been observed that early during cell mitosis of root tip cells, pro-vacuolar vesicles allocate in both daughter cells [48]; but a de novo synthesis of vacuoles has been suggested after cell specialization [49], either from the trans-Golgi network or from the endoplasmic reticulum [48]. Different types of vacuoles originate from similar basic structures, the pro-vacuoles. Vacuole differentiation is controlled by the cytoplasm; their differentiation is dependent on the proteins that they allocate coming form the ER, on developmental processes and environmental changes [47, 50, 51]. The best studied differentiation of the vacuole is the transformation from protein storage vacuoles (PSV) to lytic vacuoles (LV) during seed development and the reversed process during seed germination and seedling growth [47]. Different type of vacuoles can also be present in a cell at the same time, e.g. during leaf senescence [52], establishing a physical separation of different physiological functions. Even intra-vacuolar compartments within one vacuole have been observed, keeping e.g. lytic enzymes separated from storage proteins [47, 53]. Protein degradation for vacuole differentiation can happen on both sides of the vacuolar membrane, at the lumen as well as in the cytosol. In bacteria and animals intra-membrane-cleaving proteases (I-CliP) have been described in the vacuole membrane and homologous genes have also been found in plants [54], however, no activity has been proven yet.

15 Protease activity and protein mobilization are spatially and temporally highly regulated in the vacuole [55]: removal of intra-vacuolar compartments or changes in pH values activate proteases; processing peptidases with high specificity regulate proteases with a broader spectrum performing unlimited proteolysis [56-58]. Proteins imported into the vacuole can also undergo conformational changes e.g. due to the lumenal acidic pH, exposing cleaving sites to the active proteases [59]. Some biological functions performed by vacuolar proteases are: 1. Remobilization of nitrogen which will be used for the formation of new proteins during seed germination. Proteases involved are phytepsin (A1), aleurain (C1) and other CysPro. 2. Defence against herbivorous and pathogen attack, being an important process within the hypersensitive response (HR). There are two different strategies for the roles of protease: either activation of proteases that degrade the pathogens trying to penetrate in the wounded tissues or activation of compounds that inhibit pathogen proteases [47]. Proteases involved: VPEs (C13), subtilisin-like SerPro (S8) and carboxypeptidases. 3. Senescence previous to the point-of-no-return. It has been reported that proteases accumulate in vacuoles during senescence. Their function has not entirely resolved, they may just accumulate, so that in case the senescence becomes irreversible they can perform degradation. Proteases involved: senescence associated gene SAG12 (C1), VPEs (C13), carboxypeptidases, RD21 (C1). 4. PCD when cytosolic proteins are targeted to the vacuole for unlimited degradation. At this latest stage the vacuole will disrupt its membrane and the hydrolytic enzymes will be released in the cytoplasm, degrade the cell corpse and recycles cell material [60, 61]. Involved proteases: AspPro, aleurains (C1), VPEs (C13). Proteomic studies on Arabidopsis vacuoles have been performed [46] and over 20 proteases were identified. Proteases localized in vacuole seem to share some characteristics even though they belong to different proteolytic families: the activity at low pH, a certain autocatalytical power and the expression under similar conditions make the vacuole an organelle to take into account in stress response or nitrogen remobilization [5, 62-64]. Little is known about substrate specificity of these proteases. VPEs seem to have a processing role in vacuoles, they might be involved in nitrogen remobilization in storage vacuoles and,

16 maybe, in other protease activation [64]. But the facts that VPEs have substrate homology with mammalian caspases and are stress and senescence dependent regulated, demonstrate the versatility of these enzymes. C1 proteins are widely present in different organelles. Their presence has even been suggested in mitochondria and chloroplast [5] but their functional role in these organelles is unknown. They have seemed to be involved in nitrogen remobilization, plant suicide programs and stress responses [62, 65]. The facts that SAG12 is the only Arabidopsis protease with expression restricted to leaf senescence [5], and that SAG12 mutants do not show delayed senescence phenotypes [47] suggest a certain redundant function within this wide family.

5.4 Mitochondria Mitochondria have for a long time been viewed as the power station of the cell. Only recently it became obvious that these organelles function much more diverse and now they are recognized as the key compartments for integration of cellular responses to a variety of extra- and intracellular stresses. By activating reaction cascades that could lead to cellular senescence or PCD they might determine the fate of the cell [66]. Mitochondria are not synthesized de novo in the cell, but maternal inherited, originating from growth and division of pre-existing mitochondria [67]. The evolutionary origin of mitochondria as a process of endosymbiosis between an anoxigenic and an oxygenic organism has been accepted. Provided with its own genome, encoding proteins that form the core complexes of the respiratory chain complexes and the ATP-synthase, the majority of the mitochondrial proteins are now encoded by the nucleus [68]. Mitochondrial protein import requests effective processing peptidases which will activate or mature proteins and enzymes. Moreover, a protein quality control system is required for housekeeping the proteins that are already there and respond to different stimuli (Figure 5). Many of the studies on mitochondrial proteases were performed in yeast, which posses clear homologues to other eukaryotic proteases.

17 5.4.1 ATP-independent proteases Processing peptidases After import proteins targeted to mitochondria have to be processed by cleavage of their presequence. This processing is carried out by the mitochondrial processing peptidase (MPP) from the M16 family, an integral part of the cytochrome bc1 in plant mitochondria [69]. MPP is a heterodimer, whose core is formed by two to four V and U subunits. While a Zn-binding motif confers the protease activity in the U subunits, the V subunits enable substrate specificity [70]. MPP cleaves the majority of the mitochondrial proteins imported from the cytosol [67]. Some protein precursors targeted to mitochondrial matrix or the inner membrane are processed in two steps. The first cleavage is performed by an M3 protein, the mitochondrial intermediate protease (MIP), and only the second cleavage by MPP. Inner membrane peptidase (IMP) belonging to the S26 family, cleaves the presequence of mitochondrial precursor proteins delivered to the intermembrane space [67]. In contrast to yeast mitochondria very few homologues of these proteases are found in plants (for Arabidopsis see [4]). Oma1 (M48) is a membrane metalloprotease that forms homo-oligomeric complexes, probably of six subunits. Its name stands for overlapping with m-AAA protease; which is done in an ATP-independent manner. Oma1 has been described to degrade m-AAA substrate also in the absence of the m-AAA protease in vivo and in vitro [71]. Homologues to Oma1 are present in eukaryotic cells, bacteria and archaebacteria, and it has been suggested to be the founding member of a conserved family of membrane metallopeptidases [71]. Presequence protease (PreP) of the M16 family is a zinc metalloprotease encharged of degrading the transit peptide of processed proteins after it has been removed from the imported protein [72, 73].

Rhomboid proteases Regulated intramembrane proteolysis (RIP) is a mechanism for controlling production of signaling molecules [74]. Rhomboids (S54) are a novel group of I-CLiPs described as one of the performers of RIP [75]. Rhomboid proteases are described as membrane proteins with around seven predicted transmembrane (TM) domains [75] cleaving theirs substrates intramembranal [76, 77]. Although they belong to the family of

18 SerPro, they do not possess the typical catalytical triad, but instead posses a catalytical dyad indicating a convergent evolutionary pathway instead of a relation to the other members. Mitochondrial rhomboids have been shown to be associated with mitochondrial biogenesis [78] and direct regulation of apoptosis by controlling the release of cytocrome c [79].

Deg proteases Deg proteases (S1) are relatives to the bacterial HtrA (high temperature requirement). This group of SerPro is involved in tolerance against various stresses and pathogenicity [80]. In E. coli Deg protease activity was found to be dependent on high temperature, while, at low temperature, the chaperone function predominates. This protease is able to degrade unfolded or misfolded proteins. It contains a PDZ domain, important for protein-protein interaction as well as substrate specificity or assembly of multimeric complexes [81]. Function and substrates of higher plant mitochondrial Deg proteases remain unknown.

Cytosol

TOM

TIM i-AAA

MPP m-AAA PreP 

Lon IMP MIP

ClpXP

Oma1

Rhomboid

Deg

IM OM

Figure 5.- Mitochondrial proteases. Scheme of the different proteolytic fates a protein can undergo after it has been imported into mitochondria. The different mitochondrial proteases and their possible localization are presented. IM: Inner membrane. OM: outer membrane. TIM: translocase of the inner membrane. TOM: translocase of the outer membrane.

19 5.4.2 ATP-dependent proteases The quality control of mitochondrial proteins relays mainly on the ATP- dependent proteolytic machinery, which is responsible for the degradation of abnormal or damaged proteins [82]. Proteases belonging to this machinery are multi-oligomeric complexes that form ring-shaped structures. The AAA (ATPase associated with various activities) family contains homologous ATPase domains, called AAA domain, providing the energy needed to unfold and translocate the substrate into the proteolytic chamber through narrow pores at the centre of the complex [83, 84]. The three groups of ATP-dependent protease multigene families are highly conserved to their bacterial ancestors [85, 86]. In Lon and FtsH proteases reside proteolytic domain and ATP binding domain in the same polypeptide chain, while Clp proteases posses proteolytic subunits that are different from the subunits with ATP-domain [66].

FtsH proteases The M41 multigene family comprises TM metalloproteases, including the FtsH proteases. FtsH forms hexameric complexes degrading proteins from the mitochondrial matrix, also called m-AAA, or the IM, called i- AAA, depending to which side the ATP-binding motifs face [87]. These proteases have been extensively studied in yeast. The m-AAA protease is a hetero-oligomeric hexamer formed by the proteins Yta10p and Yta12p, while Yta11p forms the homo-oligomeric i-AAA protease. FtsH has been shown to remove non-assembled or misfolded membrane proteins and mediates the assembly of respiratory chain complexes [87, 88]. Moreover, it has been suggested to be involved in the communication between mitochondria and the nucleus, controlling protein expression in relation to the functional state of the respiratory chain [89]. Arabidopsis has 12 putative and active FtsH proteases. Three of the active proteases are targeted to mitochondria: FtsH3, FtsH4 and FtsH10; FtsH11 is dual targeted to mitochondria and the chloroplast [66, 90, 91]. Membrane topology studies claim that FtsH3 and FtsH10 build m-AAA proteases while FtsH4 as well as FtsH11 form homo-oligomeric i-AAA protease complexes [91]. T-DNA knockout mutants of these four FtsHs have related FtsH3, FtsH4 and FtsH10 to the mitochondrial biogenesis of the oxidative phosphorylation system [87], similar to their yeast relatives [66, 92, 93].

20 Apparently FtsH11 does not have the same role as the other mitochondrial FtsHs, its function in plant mitochondria is still unclear.

Clp proteases Clp proteolytic complex (S14) is a barrel-shaped structure in which the ATP-binding site and the proteolytic core are the result of different gene products. Two heptameric rings form the proteolytic chamber and are surrounded on each side by two hexameric rings giving the substrate recognition, ATP-binding and chaperon functions. This multisubunit complex is localized in mitochondrial matrix. Due to its similarity to the proteasome 26S, the Clp complex might fullfil a similar function in mitochondria and chloroplasts. In Arabidopsis Clp proteins belong to a multigene family of almost 30 members of chaperones and proteases, most of them showing a high sequence similarity with their bacterial orthologs [9]. There are 6 genes coding for subunits with proteolytical properties ClpP1-6 [9, 81]. All of them share the typical catalytic triad of SerPro, however, only ClpP2 is present in mitochondria forming a homo-oligomeric structure that constitutes the proteolytical core [94]. The chaperone rings in mitochondria of Arabidopsis seem to be formed exclusively by ClpX, a nuclear encoded class II chaperone with just one ATP-binding site [9, 94, 95]. Function of the Clp complexes has been associated with stress response, degradation of misfolded and abnormal proteins and gene regulation via the proteolysis of transcription factors [94].

Lon proteases Lon proteases (S16) have been described as hexameric rings in the soluble parts of the cell. Initially they were thought to be exclusively in mitochondria, supported by the fact that there have been no genes coding for Lon found in cyanobacteria [9, 81]. Similar to FtsH their AAA domain and the proteolytic part are localized on the same polypeptide chain. Even though Lon are SerPro they do not have the typical catalytical triad of this group, instead a Ser-Lys dyad may constitute the proteolytic core. Four of the Arabidopsis Lon proteases (AtLon1-4) are thought to be active due to their high homology to E.coli. Phylogenetic analyses placed AtLon2 in peroxisomes, while the other three proteases were thought to

21 have mitochondrial localization [96]. In silico combined with GFP analysis supported the localization of AtLon1 in mitochondria; AtLon4 is double targeted to mitochondria and chloroplast. These results were confirmed in tobacco plants [96].

5.5 Chloroplast Differentiation of plastids in higher plants is a light-dependent and organ- specific process. In light plastids start to develop the thylakoid membrane to allocate chlorophyll-binding protein complexes that will perform the photosynthetic process [97, 98]. In darkness the green chloroplast looses chlorophyll and turns into an etioplast. Amyloplasts are localized in storage tissues and are not pigmented. Chromoplasts with their high amounts of carotenoids are found in fruits and tissues responsible for pigment synthesis or storage. The differentiation of plastids is mediated by the protein content and in this process proteases seem to play a special role [95]. They are not only involved in the biogenesis and maturation of plastids but also in the maintenance of the quality control of this organelle. Similar to mitochondria also the chloroplast evolved from more simple bacteria and therefore also the plastidic proteases are very related to their bacterial ancestors [99]. However, compared to those ancestral prokaryotes photosynthetic organisms contain a significantly higher number of protease subunits and isoforms indicating diversification, neo- and subfunctionalization during evolution [95]. This higher complexity probably was acquired together with the photosynthetic apparatus and contributes to the adaptation to oxygenic photosynthesis in plastids [99].

5.5.1 ATP-independent proteases Processing peptidases Processing peptidases performing limited proteolysis for maturation or activation of enzymes belong to the group of ATP-independent proteases. In the chloroplast this role is carried out in the stroma by the stromal processing peptidase (SPP) belonging to the S49 family, and PreP, which is double targeted to both mitochondria and the chloroplast [100]. In the thylakoid lumen the thylakoid processing peptidase (TPP) (S26 family) and the C-terminal processing peptidase (CtpA, S41) are localized.

22 RIP proteases RIP in chloroplasts might be performed by two types of proteases. Two of the rhomboids annotated in the Arabidopsis genome are predicted to be localized in chloroplast (PaperIV) and might be involved in leaf development and senescence. Further on, a novel family of chloroplastic metalloproteases with similar localization and structure to rhomboids has recently been described to be localized in the thykaloid membrane. This EGY (ethylene-dependent gravitrospism-deficient and yellow green) protease belongs to the M50 MEROPS family. It consists of eight putative TM domains spanning in the thylakoid membrane [101]. Its function has been related to thylakoid and grana development, to the degradation of the chlorophyll a/b binding proteins of the photosynthetic light harvesting complex and, possibly, to regulation of lipid biosynthesis in plant cells [101].

Deg proteases The D1 protein of Photosystem II (PSII) has been described as target of some of the cyanobacterial or chloroplastic Deg and FtsH proteases [102- 106]. A Deg protease might perform the first cleavage of the D1 protein during D1 turnover and FtsH then degrades the resulting fragments [105, 107, 108]. Four Deg proteases are imported into the chloroplast of Arabidopsis. They have been shown to be attached to the thylakoid membrane on both sides, the luminal and stromal one. In vitro experiments indicated the stromal Deg2 to be responsible for the initial D1 cleavage, however these result could not be verified in vivo [109]. Instead the luminal located Deg1, Deg5 and Deg8 proteases seem to initiate D1 degradation, facilitating a further degradation by FtsH [106, 110].

5.5.2 ATP-dependent proteases FtsH proteases Similar to bacterial FtsHs the chloroplastic FtsH proteases consist of two N-terminal membrane-spanning helices exposing the ATP-ase and Zn2+ binding domains to the stroma [111]. Like the bacterial members of this family plastidic FtsHs have highly conserved AAA domains probably forming hexameric complexes a similar hexamer complex formation for plastids as described in the crystal structure for E.coli FtsH [112]. In E. coli FtsH degrades short-lived regulatory proteins in the cytosol as well

23 as membrane proteins. Although FtsH possesses only a limited ability to unfold substrate proteins [113], it has a unique ability to dislocate substrate membrane proteins out of the membrane and can initiate its degradation from either the N- or the C-terminus of the substrate [114]. The translocation of the substrate proteins to the proteolytically active site requires ATP hydrolysis and is achieved by unfolding of the substrate proteins through the ring structure.

In the chloroplast of Arabidopsis this group consists of 12 putative active proteases and 4 inactive FtsHs (FtsHi), lacking the zinc-binding domain. Characteristic for the FtsHs in Arabidopsis is the high homology of protease pairs, indicating a recent gene duplication. The tight homology between mitochondrial FtsH3 and FtsH10 and a less pronounced homology of FtsH4 and FtsH11 has already been mentioned. While the function of FtsH11 in mitochondria is not understood, its copy targeted to the chloroplast has recently been found to be involved in thermotolerance [115]. Other plastidic FtsH pairs are FtsH1/5, FtsH2/8 and FtsH7/9. FtsH6 and FtsH12 are located in the chloroplast (Paper IV), but do not form homologous pairs [90, 95] (Paper III, Paper V). Chloroplastic FtsHs have been shown to be involved in biogenesis of the organelle, and degradation of PSII proteins. Mutations in genes coding for FtsH5 (var1) and FtsH2 (var2) in Arabidopsis developed leaf- variegated phenotypes [90, 116, 117]. While the cotyledons in these mutants appear normal, the first true leaves are white and the rosette leaves show a variegated phenotype consisting of white and green sectors formed within the same leaf. During leaf development, the proportion of green sectors increases. The green tissues contain morphologically normal chloroplasts; white sectors, however, are heteroplastidic and contain normal chloroplast together with abnormal plastids [118]. Complementation tests [90] showed that the two subunits in a pair can substitute for each other to some extent. Thus, FtsH1 compensated the lack of FtsH5 and FtsH8 compensated for FtsH2 indicating an overlapping function [119, 120]. It has been proposed that FtsH1, 2, 5 and 8 form hetero-oligomeric complexes. The presence of all subunits is necessary to be above a certain threshold level for complex formation and normal chloroplast development [119, 120]. When the number of complexes does not reach this level, chloroplast function is impaired and the white sectors are formed.

24

FtsH1 and FtsH2 are also involved in turnover of the Photosystem II D1 protein and the PSII repair process. If impaired D1 degradation causes leaf variegation so far is not known. Several other photosynthetic proteins have been shown to be the substrate of FtsH proteases beside D1: D2 [121], the Rieske Fe-S protein of the cytochrome b6f complex [122] and also the light-harvesting complex of Photosystem II, which was shown to be degraded by FtsH6 (see Paper II).

Clp proteases ClpP proteins in Arabidopsis plastids are formed by nuclear and plastidic encoded gene products. These complexes in the chloroplast are hetero- oligomeric and larger than the homo-oligomeric complexes formed in mitochondria. The different plastidic ClpP proteins participating in the proteolytic core in Arabidopsis are ClpP1 (the only subunit chloroplastic encoded in the chloroplast) and ClpP3-6. Moreover there are four non- proteolytic subunits that play a structural role in the core formation. ClpR1-4, highly homologues to the ClpPs, they lack one or several of the catalytic residues in the catalytical triad. These proteins are exclusive in photosynthetic organisms [94, 123]. Additionally to these proteins a subunit unique in land plants, with unknown function, participates in the formation of the core [123]. These proteins are the ClpS. According to 3D modelling, these proteins might assist the substrate in entering the proteolytic pore and the resulting products to move out of the Clp complex [94]. The chaperone function of the chloroplastic Clp proteases is performed by ClpC and ClpD forming hexameric rings regulating the Clp proteolytic activity.

Mutations in ClpP and ClpC genes result in chlorotic phenotypes [124- 126]. Regreening of the ClpP mutant chlorotic plants was, as in the case of the var mutants, following leaf age and seems to be dependent on the upregulation of other ClpP and chaperone genes that could compensate for missing subunits or inactive complexes. Mutation on ClpR genes resulted in abnormal chloroplast development [127, 128]. In response to this phenomena, proteases and chaperone levels where adjusted either to compensate for lacking of functional Clp complexes [127] or to reduce the protein import efficiency [128].

25 6. Proteases involved in senescence and PCD

PCD is a process ruling the removal of redundant, misplaced or damaged cells and is essential for development and maintainance of multicellular organisms [33], especially in response to different stresses and HR [129, 130]. It is often associated with specific biochemical and morphological features such as the production of reactive oxygen species (ROS), activation of caspase-like proteases, cytochrome c release from mitochondria, cytoplasm shrinkage, chromatin condensation and DNA fragmentation [131, 132]. In plants, PCD has been observed during tracheary element differentiation [133], aerenchyma formation [134], HR and as a consequence of several biotic and abiotic stresses [129]. In animals mitochondria have been postulated to be the initiator of PCD. Metabolic perturbations let the mitochondrial membrane loose its integrity and certain elicitors are released into the cytosol switching on the cell death process [135]. In plants, however, mitochondria remain functional until the final stages of leaf senescence; instead it seems that the chloroplast regulates the senescence process[136].

Even though the definition of senescence and PCD is controversial debated [60] I will use these terms according to [137]: senescence is a reversible process at cellular or organ-level. Only from the “point of no return”, in which the cell is committed to die [138] PCD will start at cellular level. Therefore it is possible to say that during senescence conditions the purpose of the cell is not to reach death, but rather to start a defence mechanism to protect it from adverse conditions or exogenous attack [139]. An important function of senescence is to reallocate nutrients, particularly nitrogen, to other parts of the plant before the specific structure is degraded.

6.1 Proteases during plant development (Paper IV) As mentioned above, proteases are present in every cellular mechanism and perform important functions. However, substrate specificity, activation or function of most of the proteases in a specific organism

26 remains unknown. We will not know if all the hypothetical proteases annotated in different databases are active enzymes before we have found their substrates. A comparative genomic approach can give us hints of protein sequences conserved in different species and, via orthology, possible functions. Additionally one can learn more about the formation of gene families after gene duplication and possible neo- or subfunctionalitation of the newly developed genes. A comparison of all genes coding for proteases in the two organisms Arabidopsis and Populus clearly showed duplications in the poplar genome after evolutionary divergence of both species (Figure 6). Proteases in Populus might have acquired new functions or locations leading to different regulative pathways in the tree compared to an annual species like Arabidopsis.

In this study I concentrated mainly on the gene families that have been shown to be involved in leaf senescence. The cytosolic papain-like proteases as well as several chloroplastic localized proteases are thought to be senescence mediators: the first ones because they are localized mainly in the vacuole, the ultimate protease storage organelle, which is thought to work as a late PCD performer and the second ones because the chloroplast initiates plant senescence. In this context it is interesting In this study I concentrated mainly on the gene families that have been shown to be ito note that the seven proteases with highest transcription in the whole expressed sequence tag (EST) library are found in the leaf senescence library; all of them correspond to chloroplast or papain-like proteases (see Paper IV, Figure 8). Expression pattern analysis clustered the Populus proteases dependent on their transcription during leaf development, allowing conclusions on their functions. Similar to the previous results also these data (Paper IV, Figure 9) demonstrate the upregulation of vacuolar proteases (papain-like and VPEs) as well as chloroplastic proteases (FtsH and Clp) during leaf senescence conditions. It should be noted again, that gene expression does not necessarily equals

27 Arabidopsis Populus T1 T2 T3 C1 C12 C19 C54 C65 C13 C14 C48 C15 C26 C56 S1 S16 S49 S54 S8 S9 S10 S28 S33 S12 S26 S14 S41 S59 M48 M67 M1 M3 Protease Classes: M8 M10 ThrPro M41 M14 CysPro M16 M17 SerPro M24 M18 MetalloPro M20 M28 AspPro M38 M22 M50 A1 A11 A22 8 6 4 20 0 0 20 4 60 80

Figure 6.- Classification and comparison of proteases in Arabidopsis and Populus. The different colors indicate the different protease classes: threonine proteases (T), cysteine proteases (C), serine proteases (S), metalloproteases (M) and aspartic proteases (A). Each class can be divided into different families according to MEROPS, the family number is indicated between the Arabidopsis and Populus charts.

protease activity. Translational and post-translational modifications may regulate maturation of these proteases; they might be dependent on the expression of other proteins involved in the maturation process or on substrate availability.

28 6.2 Proteases involved in wood formation (Paper I) The tracheary elements (TE) in angiosperm trees are differentiated from lateral meristem of the stem, the vascular cambium. This differentiation includes morphological cell changes, followed by extensive synthesis of the secondary cell walls. Thickening of the wall adds strength and rigidity and prevents TEs to collapse during the high pressure from fluid transport they will expose to later on. While during leaf PCD nucleus and vacuole degenerate not before complete degradation of the chloroplast, in TEs organelles are degraded after vacuole breakage and release of the proteases [140]. The cell content was shown to be degraded very rapidly: in Zinnia the process from cell thickening to vacuole collapse and autolysis was reported to take six hours [141]. Proteases involved in this process are CysPro and the proteasome, which functions – probably due to the rapid process - until the vacuole disrupts. In paper I analysis of protease EST abundance and their distribution in the different libraries, especially in the wood cell death library, was performed in Populus. The abundance of ESTs in special libraries very likely correlates to the level of the corresponding mRNA and maybe even protein. Proteases present in the EST libraries of fiber death were compared to leaf senescence, virus/fungi infected leaves and tension wood libraries since the cells have similar faiths in these processes. It could be shown that the protease expression in wood cell death seems to follow a more common regulation, similar to the other cell processes and also cold stress (Paper IV, Figure 8). No protease was found to be uniquely expressed just in the fiber death library. However, one Deg protease was observed to be present exclusively in the leaf senescence library. The presence of papain-like CysPro and VPEs ESTs confirmed their importance in the different cell death processes. Moreover, SerPro and AspPro seem to be involved in the different cell death scenarios and also in fiber death. Our data show that only a few proteases belonging to a limited number of families are involved in the degradation of cells leading to different kinds of cell death. While different stimuli might trigger various responses leading to cell death, the ultimate roles are played by the same actors.

29 6.3 FtsH proteases involved in the degradation of the photosynthetic light-harvesting antenna (Papers II and III) The light harvesting complex of PSII (LHC II) is located in the thylakoid membrane, where it collects energy from sunlight and transfers it to the PSII reaction centre. The functional unit of LHC II is a trimer, representing various permutations of Lhcb1-3 apoproteins, that bind chlorophylls a and b and carotenoids. The genes coding for the three apoproteins are typically found as multiple copies in the genomes of higher plants. Beside the energy transfer, LHC II is involved in non- photochemical quenching, the distribution of excitation energy between PSII and photosystem I (PSI) and in preventing damage to the photosynthetic machinery under HL conditions. Depending on the light conditions expression and degradation of these proteins is highly regulated. Different proteases have been shown to be responsible for the degradation of LHCII during various conditions (Paper IV). While a serine protease was shown to be involved in high- light acclimation [142] (see Paper IV) an intrinsic, ATP-dependent metalloprotease was demonstrated to be responsible for the degradation of the subunits during senescence [142]. Lhcb3 seemed to be the LHC protein which was disappearing in a most rapid way during dark-induced senescence [142] (Paper II). The loss of chlorophyll in senescent leaves correlated with the loss in abundance of Lhcb3 content in thylakoid membranes (Paper II). Using the reverse genetics approach we were able to show that a member of the FtsH family in Arabidopsis, FtsH6, is involved in the degradation of Lhcb3 during senescence. Lhcb1 is the LHC protein that is most rapidly degraded under HL acclimation. We could conclude that also FtsH6 is involved in the degradation of Lhcb1. Gene expression of FtsH6 is low, but it is mainly expressed in leaves and upregulated under certain types of stresses and senescence [143]. Compared to all FtsHs of Arabidopsis, FtsH6 has highest homology to the Var1/Var2 group (FtsH1/5 and FtsH2/8), that are known to form hexamers and are involved in chloroplast biosynthesis as well degradation of the Photosystem II D1 protein. These 5 proteases therefore might have the same phylogenetic origin. This evolutionary ancestor probably was responsible for the degradation and maintenance of photosynthetic

30 proteins. After gene duplication FtsH6 was neofunctionalized and became more specialized for LHC proteins, which probably evolved at the same time.

Figure 7.- Schematic representation of the Arabidopsis FtsH6 protease subunit. The two TM helices are connected by a luminal loop, while ATP-binding domain and zinc-binding motif are facing the stroma. As the protease is present, but not active, in the thylakoid membrane of plants grown at normal light conditions, its proteolytic site might be shielded by unknown extrinsic proteins.

FtsH6 was found to have a true ortholog in Populus, supporting its important function. Further studies on substrate specificity and activity of FtsH6 will be necessary. Currently we postulate that FtsH6 activity is regulated by extrinsic factors (Figure 7), shielding its stromal located proteolytic site. Removal of these factors by another protease or environmental conditions FtsH6 either activates the LHCII degrading protease or itself performs the degradation.

31 7. Conclusions and future perspectives

Proteases are implicated in all cellular processes. Some of them are involved in maturation of other proteins and some others perform unlimited degradation to ensure maintenance of the protein complexes and contributing to the recycling of nitrogen. In this overview I tried to summarize the available data on function and localization of the different protease families with emphasis on proteases in higher plants, covering the most important cell organelles in the plant cell.

During my work with plant proteases several points became obvious that are summarized below: • Comparing the proteases encoded in the genomes of Arabidopsis and Populus I could show that despite a recent gene duplication in Populus the same protease families are present in a tree and an annual plant. Even though leaf senescence must be a more important process for a tree, Populus has not evolved a different set of proteases performing the process. However, neofunctionalization is a process currently taking place and we were able via phylogenetic analysis to find new proteases that appeared in Populus after gene duplication. • In Populus leaf senescence activates the same set of proteases as other processes leading to cell death. Even though the proteases might be the same, the program followed by the cell is different in leaf senescence compared to TEs differentiation: Leaf senescence starts in the chloroplasts, in woody tissues the cell thickening might be starting signal. • An ATP-dependent metalloprotease belonging to the family of FtsH, FtsH6, is involved in the degradation of the light- harvesting complex during leaf senescence in Arabidopsis and probably also in Populus. Some of the members of the FtsH family are known of participate in chloroplast biogenesis and PSII degradation in response to photoinhibition. FtsH6 might have acquired its function after neofunctionalitation after duplication of an ancient Var1/Var2 type protease.

32

FtsH is a protease family involved in important chloroplastic processes. Of the 12 active FtsHs in Arabidopsis only the four most strongly expressed members are studied extensively. Some data are available on FtsH6 and FtsH11. To learn more about these interesting proteases I would like to investigate their complex formation. Do all FtsH proteases form hexamers, are those hexamers homo- or hetero-oligomeric? What is the stoichiometry in these hexamers? Are different complexes formed in respond to different environmental stimuli? Furthermore, an interaction on gene and protein level has been demonstrated between FtsH proteases and other proteolytic complexes: Silencing of ClpC2 arrests the variegated phenotype caused in the FtsH2- deletion mutant; ClpR knock-out mutants respond with FtsH upregulation; or Deg1 mutations cause a decrease in the amount of FtsH. These cooperative pathways are described, but not understood. Answering these questions will be important for future research.

33 8. Acknowledgements

Coming to this point, it is time for acknowledging everybody that has passed by my side during these 5 years, quite a life and quite many people I got to know. Students, people in and outside of the Umeå University and people I have met at the different conferences have contributed to this work in a positive way. It has been 5 years of enriching experiences that I am grateful of having shared with all of them. This has been my university life in the last 5 years:

Mi familia, por estar siempre cerca, o por hacerme sentiros así. Álvaro told me first about Umeå. We spent together a lot of time and had fun. Umeå was never the same since he left. Christiane has been a very humane supervisor; she always has time for listening and helping in good and in not so good moments. She spreads around all her positive energy. Stefan has helped me very much with the articles and work. He is the best at talking about science for ordinary people and makes the best marinara meat at BBQ´s. Agnieszka rescued me from the computer-work, she showed me how much fun one can have working in the lab, she taught me very much and even, she let me be her radio the long working days in the coldroom =) Andreas let me bother him any time with my long protease lists and provided me with lots of valuable data. Olivier provided me with mitochondrial samples. He is positive, funny, always smiling, and appreciates the good “lomo” as a real Spaniard. The LeaSen group, for interesting discussions about leaf senescence Adriana is a nice lab-partner. She is enthusiastic with science and cheers me when I need it. W olfgang Schröder always has funny comments, especially about science, and the best way to keep such things in mind. Thomas Kieselbach is always positive at the PhD student seminars, has good and encouraging ideas. Arsenio, Marisa, Amaya, Jose, Miguel, Virginia, Aurora and Cristian made possible to talk about plant science in Spanish, helped with technical problems, personal worries or just spent time talking and laughing en español.

34 Ana saved my life many times with the teaching and helped me with the phylogeny and everyday little problems in an unfamiliar country. Anna-Märta helped me with translation, when I could not understand a word in Swedish, and legal formalities. Questions, wonderings and difficulties in preparing the courses were solved by the always helpful Viola and Katarina. Vladimir, Mantas (who makes me fly when we dance together, even though I do not control the landing yet), and Lena took me as if I was their little sister and had lots of fun with them in the beginning, they are good friends. Junko, a very good, kind and supportive friend, sometimes let me keep one of her secrets and then she could keep some of mine =) Zaki used to be as bad in Swedish as me. Later we discovered that that is not the only thing we have in common. He has taught me many English words and expressions I should never say at conferences. Barbara brought the glamour to the Biochemistry department, she knows how to combine high heels with a lab coat! Christina believed in me and now she would be happy because I reached the top of “her ladder”. Lars assigned me very much teaching that has been quite a fulfilling experience =) I learned a lot from it, very much about biochemistry, and very much about myself. The past and present members at Christiane and W olfgang´s groups made an enviable working environment. People at the UPSC, always helpful at work and funny at parties. With 3x J.Loring I have spent the best moments, food and games. It is always fun to play “washing machine”.

Tack Eivor, Roland, Maria, Enar och Robert för så trevliga tider och resor, alltid så god mat och för att visa mig den mysiga svenska kulturen.

Tack Markus, för alla stunder. Du gör mig GLAD.

35 9. Reference list

1. U Feller: Proteolysis. California: Elsevier Academic Press; 2004. 2. Z Adam: Chloroplast proteases: possible regulators of gene expression? Biochimie 2000, 82:647-54. 3. J Callis: Regulation of Protein Degradation. Plant Cell 1995, 7:845- 857. 4. ND Rawlings, FR Morton, AJ Barrett: MEROPS: the peptidase database. Nucleic Acids Res 2006, 34:D270-2. 5. EP Beers, AM Jones, AW Dickerman: The S8 serine, C1A cysteine and A1 aspartic protease families in Arabidopsis. Phytochemistry 2004, 65:43-58. 6. S Leonelli: Arabidopsis, the botanical Drosophila: from mouse cress to model organism. Endeavour 2007, 31:34-8. 7. GA Tuskan, S DiFazio, S Jansson, J Bohlmann, I Grigoriev, U Hellsten, N Putnam, S Ralph, S Rombauts, A Salamov, et al: The genome of black cottonwood, Populus trichocarpa (Torr. & Gray). Science 2006, 313:1596-1604. 8. HM Ku, T Vision, JP Liu, SD Tanksley: Comparing sequenced segments of the tomato and Arabidopsis genomes: Large-scale duplication followed by selective gene loss creates a network of synteny. Proceedings of the National Academy of Sciences of the United States of America 2000, 97:9121-9126. 9. A Sokolenko, E Pojidaeva, V Zinchenko, V Panichkin, VM Glaser, RG Herrmann, SV Shestakov: The gene complement for proteolysis in the cyanobacterium Synechocystis sp. PCC 6803 and Arabidopsis thaliana chloroplasts. Curr Genet 2002, 41:291-310. 10. JM Archibald: Jumping genes and shrinking genomes--probing the evolution of eukaryotic photosynthesis with genomics. IUBMB Life 2005, 57:539-47. 11. JA Raven, JF Allen: Genomics and chloroplast evolution: what did cyanobacteria do for plants? Genome Biol 2003, 4:209. 12. W Martin, T Rujan, E Richly, A Hansen, S Cornelsen, T Lins, D Leister, B Stoebe, M Hasegawa, D Penny: Evolutionary analysis of Arabidopsis, cyanobacterial, and chloroplast genomes reveals plastid phylogeny and thousands of cyanobacterial genes in the nucleus. Proc Natl Acad Sci U S A 2002, 99:12246-51. 13. WH Li, Z Gu, AR Cavalcanti, A Nekrutenko: Detection of gene duplications and block duplications in eukaryotic genomes. J Struct Funct Genomics 2003, 3:27-34. 14. G Theissen: Secret life of genes. Nature 2002, 415:741. 15. M Lynch, JS Conery: The evolutionary fate and consequences of duplicate genes. Science 2000, 290:1151-5.

36 16. M Hurles: Gene duplication: the genomic trade in spare parts. PLoS Biol 2004, 2:E206. 17. S De Bodt, G Theissen, Y Van de Peer: Promoter analysis of MADS- box genes in eudicots through phylogenetic footprinting. Mol Biol Evol 2006, 23:1293-303. 18. MC Riley, A Clare, RD King: Locational distribution of gene functional classes in Arabidopsis thaliana. BMC Bioinformatics 2007, 8:112. 19. A Schaller: A cut above the rest: the regulatory function of plant proteases. Planta 2004, 220:183-97. 20. M Ehrmann, T Clausen: Proteolysis as a regulatory mechanism. Annu Rev Genet 2004, 38:709-24. 21. J Kurepa, JA Smalle: Structure, function and regulation of plant proteasomes. Biochimie 2007. 22. J Hanna, D Finley: A proteasome for all occasions. Febs Letters 2007, 581:2854-2861. 23. J Smalle, RD Vierstra: The ubiquitin 26S proteasome proteolytic pathway. Annu Rev Plant Biol 2004, 55:555-90. 24. P Yang, H Fu, J Walker, CM Papa, J Smalle, YM Ju, RD Vierstra: Purification of the Arabidopsis 26 S proteasome: biochemical and molecular analyses revealed the presence of multiple isoforms. J Biol Chem 2004, 279:6401-13. 25. N Bader, T Grune: Protein oxidation and proteolysis. Biol Chem 2006, 387:1351-5. 26. Z Peng, G Serino, XW Deng: Molecular characterization of subunit 6 of the COP9 signalosome and its role in multifaceted developmental processes in Arabidopsis. Plant Cell 2001, 13:2393- 407. 27. X Huang, BK Hetfeld, U Seifert, T Kahne, PM Kloetzel, M Naumann, D Bech-Otschir, W Dubiel: Consequences of COP9 signalosome and 26S proteasome interaction. Febs J 2005, 272:3909-17. 28. AL Goldberg: Protein degradation and protection against misfolded or damaged proteins. Nature 2003, 426:895-9. 29. C Yang, V Kaushal, SV Shah, GP Kaushal: Mcl-1 is downregulated in cisplatin-induced apoptosis, and proteasome inhibitors restore Mcl-1 and promote survival in renal tubular epithelial cells. American Journal of Physiology-Renal Physiology 2007, 292:F1710-F1717. 30. JL Arbiser, XC Li, CF Hossain, DG Nagle, DM Smith, P Miller, B Govindarajan, J DiCarlo, KR Landis-Piwowar, QP Dou: Naturally occurring proteasome inhibitors from mate tea (Ilex paraguayensis) serve as models for topical proteasome inhibitors. J Invest Dermatol 2005, 125:207-12. 31. M Sanmartin, L Jaroszewski, NV Raikhel, E Rojo: Caspases. Regulating death since the origin of life. Plant Physiol 2005, 137:841- 7.

37 32. EaG Piszczek, W: Caspase-like proteases and their role in programmed cell death in plants. Acta Physiologiae Plantarum 2007, 29:391-398. 33. FA Hoeberichts, EJ Woltering: Multiple mediators of plant programmed cell death: interplay of conserved cell death mechanisms and plant-specific regulators. Bioessays 2003, 25:47-57. 34. E Lam, O del Pozo: Caspase-like protease involvement in the control of plant cell death. Plant Molecular Biology 2000, 44:417-428. 35. VI Rotari, R He, P Gallois: Death by proteases in plants: whodunit. Physiologia Plantarum 2005, 123:376-385. 36. I Hara-Nishimura, R Matsushima: A wound-inducible organelle derived from endoplasmic reticulum: a plant strategy against environmental stresses? Curr Opin Plant Biol 2003, 6:583-8. 37. C Hirsch, E Jarosch, T Sommer, DH Wolf: Endoplasmic reticulum- associated protein degradation--one model fits all? Biochim Biophys Acta 2004, 1695:215-23. 38. K Toyooka, T Okamoto, T Minamikawa: Mass transport of proform of a KDEL-tailed cysteine proteinase (SH-EP) to protein storage vacuoles by endoplasmic reticulum-derived vesicle is involved in protein mobilization in germinating seeds. J Cell Biol 2000, 148:453- 64. 39. RA Van Der Hoorn, JD Jones: The plant proteolytic machinery and its role in defence. Curr Opin Plant Biol 2004, 7:400-407. 40. C Gietl, M Schmid: Ricinosomes: an organelle for developmentally regulated programmed cell death in senescing plant tissues. Naturwissenschaften 2001, 88:49-58. 41. JS Greenwood, M Helm, C Gietl: Ricinosomes and endosperm transfer cell structure in programmed cell death of the nucellus during Ricinus seed development. Proceedings of the National Academy of Sciences of the United States of America 2005, 102:2238-2243. 42. M Schmid, DJ Simpson, H Sarioglu, F Lottspeich, C Gietl: The ricinosomes of senescing plant tissue bud from the endoplasmic reticulum. Proc Natl Acad Sci U S A 2001, 98:5353-8. 43. I Hara-Nishimura, R Matsushima, T Shimada, M Nishimura: Diversity and formation of endoplasmic reticulum-derived compartments in plants. Are these compartments specific to plant cells? Plant Physiol 2004, 136:3435-9. 44. R Matsushima, Y Hayashi, M Kondo, T Shimada, M Nishimura, I Hara-Nishimura: An endoplasmic reticulum-derived structure that is induced under stress conditions in Arabidopsis. Plant Physiol 2002, 130:1807-14. 45. R Matsushima, M Kondo, M Nishimura, I Hara-Nishimura: A novel ER-derived compartment, the ER body, selectively accumulates a beta-glucosidase with an ER-retention signal in Arabidopsis. Plant J 2003, 33:493-502.

38 46. C Carter, S Pan, J Zouhar, EL Avila, T Girke, NV Raikhel: The vegetative vacuole proteome of Arabidopsis thaliana reveals predicted and unexpected proteins. Plant Cell 2004, 16:3285-303. 47. K Muntz: Protein dynamics and proteolysis in plant vacuoles. J Exp Bot 2007, 58:2391-407. 48. F Marty: Plant vacuoles. Plant Cell 1999, 11:587-600. 49. B Hoh, G Hinz, BK Jeong, DG Robinson: Protein storage vacuoles form de novo during pea cotyledon development. J Cell Sci 1995, 108 ( Pt 1):299-310. 50. K Muntz: Deposition of storage proteins. Plant Mol Biol 1998, 38:77- 99. 51. MS Otegui, R Herder, J Schulze, R Jung, LA Staehelin: The proteolytic processing of seed storage proteins in Arabidopsis embryo cells starts in the multivesicular bodies. Plant Cell 2006, 18:2567-81. 52. MS Otegui, YS Noh, DE Martinez, MG Vila Petroff, LA Staehelin, RM Amasino, JJ Guiamet: Senescence-associated vacuoles with intense proteolytic activity develop in leaves of Arabidopsis and soybean. Plant J 2005, 41:831-44. 53. J Gillespie, SW Rogers, M Deery, P Dupree, JC Rogers: A unique family of proteins associated with internalized membranes in protein storage vacuoles of the Brassicaceae. Plant J 2005, 41:429-41. 54. A Weihofen, B Martoglio: Intramembrane-cleaving proteases: controlled liberation of proteins and bioactive peptides. Trends Cell Biol 2003, 13:71-8. 55. K Muntz, MA Belozersky, YE Dunaevsky, A Schlereth, J Tiedemann: Stored proteinases and the initiation of storage protein mobilization in seeds during germination and seedling growth. J Exp Bot 2001, 52:1741-52. 56. AD Shutov, H Baumlein, FR Blattner, K Muntz: Storage and mobilization as antagonistic functional constraints on seed storage globulin evolution. J Exp Bot 2003, 54:1645-54. 57. LW Jiang, AH Erickson, JC Rogers: Multivesicular bodies: a mechanism to package lytic and storage functions in one organelle? Trends in Cell Biology 2002, 12:362-367. 58. A Fath, P Bethke, J Lonsdale, R Meza-Romero, R Jones: Programmed cell death in cereal aleurone. Plant Mol Biol 2000, 44:255-66. 59. J Takano, K Miwa, L Yuan, N von Wiren, T Fujiwara: Endocytosis and degradation of BOR1, a boron transporter of Arabidopsis thaliana, regulated by boron availability. Proc Natl Acad Sci U S A 2005, 102:12276-81. 60. WG van Doorn, EJ Woltering: Senescence and programmed cell death: substance or semantics? J Exp Bot 2004, 55:2147-53. 61. AM Jones: Programmed cell death in development and defense. Plant Physiol 2001, 125:94-7.

39 62. EP Beers, BJ Woffenden, C Zhao: Plant proteolytic enzymes: possible roles during programmed cell death. Plant Mol Biol 2000, 44:399-415. 63. I Simoes, C Faro: Structure and function of plant aspartic proteinases. Eur J Biochem 2004, 271:2067-75. 64. K Yamada, T Shimada, M Nishimura, I Hara-Nishimura: A VPE family supporting various vacuolar functions in plants. Physiologia Plantarum 2005, 123:369-375. 65. S Gan, RM Amasino: Inhibition of leaf senescence by autoregulated production of cytokinin. Science 1995, 270:1986-8. 66. H Janska: ATP-dependent proteases in plant mitochondria: W hat do we know about them today? Physiologia Plantarum 2005, 123:399- 405. 67. O Gakh, P Cavadini, G Isaya: Mitochondrial processing peptidases. Biochim Biophys Acta 2002, 1592:63-77. 68. JL Boore: Animal mitochondrial genomes. Nucleic Acids Res 1999, 27:1767-80. 69. HP Braun, UK Schmitz: Purification and sequencing of cytochrome b from potato reveals methionine cleavage of a mitochondrially encoded protein. FEBS Lett 1993, 316:128-32. 70. P Luciano, S Geoffroy, A Brandt, JF Hernandez, V Geli: Functional cooperation of the mitochondrial processing peptidase subunits. J Mol Biol 1997, 272:213-25. 71. M Kaser, M Kambacheld, B Kisters-Woike, T Langer: Oma1, a novel membrane-bound metallopeptidase in mitochondria with activities overlapping with the m-AAA protease. J Biol Chem 2003, 278:46414-23. 72. S Bhushan, B Lefebvre, A Stahl, SJ Wright, BD Bruce, M Boutry, E Glaser: Dual targeting and function of a protease in mitochondria and chloroplasts. EMBO Rep 2003, 4:1073-8. 73. S Bhushan, A Stahl, S Nilsson, B Lefebvre, M Seki, C Roth, D McWilliam, SJ Wright, DA Liberles, K Shinozaki, et al: Catalysis, subcellular localization, expression and evolution of the targeting peptides degrading protease, AtPreP2. Plant Cell Physiol 2005, 46:985-96. 74. MS Brown, J Ye, RB Rawson, JL Goldstein: Regulated intramembrane proteolysis: A control mechanism conserved from bacteria to humans. Cell 2000, 100:391-398. 75. MM Kanaoka, S Urban, M Freeman, K Okada: An Arabidopsis Rhomboid homolog is an intramembrane protease in plants. FEBS Lett 2005, 579:5723-8. 76. EV Koonin, KS Makarova, IB Rogozin, L Davidovic, MC Letellier, L Pellegrini: The rhomboids: a nearly ubiquitous family of intramembrane serine proteases that probably evolved by multiple ancient horizontal gene transfers. Genome Biol 2003, 4:R19.

40 77. M Freeman: Proteolysis within the membrane: rhomboids revealed. Nat Rev Mol Cell Biol 2004, 5:188-97. 78. GA McQuibban, S Saurya, M Freeman: Mitochondrial membrane remodelling regulated by a conserved rhomboid protease. Nature 2003, 423:537-41. 79. S Cipolat, T Rudka, D Hartmann, V Costa, L Serneels, K Craessaerts, K Metzger, C Frezza, W Annaert, L D'Adamio, et al: Mitochondrial rhomboid PARL regulates cytochrome c release during apoptosis via OPA1-dependent cristae remodeling. Cell 2006, 126:163-75. 80. T Clausen, C Southan, M Ehrmann: The HtrA family of proteases: Implications for protein composition and cell fate. Molecular Cell 2002, 10:443-455. 81. Z Adam, I Adamska, K Nakabayashi, O Ostersetzer, K Haussuhl, A Manuell, B Zheng, O Vallon, SR Rodermel, K Shinozaki, et al: Chloroplast and mitochondrial proteases in Arabidopsis. A proposed nomenclature. Plant Physiol 2001, 125:1912-8. 82. A Mogk, D Dougan, J Weibezahn, C Schlieker, K Turgay, B Bukau: Broad yet high substrate specificity: the challenge of AAA+ proteins. J Struct Biol 2004, 146:90-8. 83. A Lupas, JM Flanagan, T Tamura, W Baumeister: Self- compartmentalizing proteases. Trends Biochem Sci 1997, 22:399-404. 84. GR Smith, B Contreras-Moreira, XD Zhang, PA Bates: A link between sequence conservation and domain motion within the AAA plus family. Journal of Structural Biology 2004, 146:189-204. 85. L Van Dyck, T Langer: ATP-dependent proteases controlling mitochondrial function in the yeast Saccharomyces cerevisiae. Cellular and Molecular Life Sciences 1999, 56:825-842. 86. M Kaser, T Langer: Protein degradation in mitochondria. Seminars in Cell & Developmental Biology 2000, 11:181-190. 87. M Kolodziejczak, A Kolaczkowska, B Szczesny, A Urantowka, C Knorpp, J Kieleczawa, H Janska: A higher plant mitochondrial homologue of the yeast m-AAA protease. Molecular cloning, localization, and putative function. J Biol Chem 2002, 277:43792-8. 88. I Arnold, T Langer: Membrane protein degradation by AAA proteases in mitochondria. Biochim Biophys Acta 2002, 1592:89-96. 89. I Arnold, M Wagner-Ecker, W Ansorge, T Langer: Evidence for a novel mitochondria-to-nucleus signalling pathway in respiring cells lacking i-AAA protease and the ABC-transporter Mdl1. Gene 2006, 367:74-88. 90. W Sakamoto: Leaf-variegated mutations and their responsible genes in Arabidopsis thaliana. Genes Genet Syst 2003, 78:1-9. 91. A Urantowka, C Knorpp, T Olczak, M Kolodziejczak, H Janska: Plant mitochondria contain at least two i-AAA-like complexes. Plant Mol Biol 2005, 59:239-52.

41 92. DJ Kominsky, MP Brownson, DL Updike, PE Thorsness: Genetic and biochemical basis for viability of yeast lacking mitochondrial genomes. Genetics 2002, 162:1595-604. 93. L Atorino, L Silvestri, M Koppen, L Cassina, A Ballabio, R Marconi, T Langer, G Casari: Loss of m-AAA protease in mitochondria causes complex I deficiency and increased sensitivity to oxidative stress in hereditary spastic paraplegia. J Cell Biol 2003, 163:777-87. 94. JB Peltier, DR Ripoll, G Friso, A Rudella, Y Cai, J Ytterberg, L Giacomelli, J Pillardy, KJ van Wijk: Clp protease complexes from photosynthetic and non-photosynthetic plastids and mitochondria of plants, their predicted three-dimensional structures, and functional implications. J Biol Chem 2004, 279:4768- 81. 95. W Sakamoto: Protein Degradation Machineries in Plastids. Annu Rev Plant Biol 2006. 96. O Ostersetzer, Y Kato, Z Adam, W Sakamoto: Multiple intracellular locations of lon protease in arabidopsis: evidence for the localization of AtLon4 to chloroplasts. Plant Cell Physiol 2007, 48:881- 5. 97. E Lopez-Juez, KA Pyke: Plastids unleashed: their development and their integration in plant development. Int J Dev Biol 2005, 49:557-77. 98. E Miura, Y Kato, R Matsushima, V Albrecht, S Laalami, W Sakamoto: The balance between protein synthesis and degradation in chloroplasts determines leaf variegation in Arabidopsis yellow variegated mutants. Plant Cell 2007, 19:1313-28. 99. Z Adam, A Rudella, KJ van Wijk: Recent advances in the study of Clp, FtsH and other proteases located in chloroplasts. Curr Opin Plant Biol 2006, 9:234-40. 100. A Stahl, S Nilsson, P Lundberg, S Bhushan, H Biverstahl, P Moberg, M Morisset, A Vener, L Maler, U Langel, et al: Two novel targeting peptide degrading proteases, PrePs, in mitochondria and chloroplasts, so similar and still different. Journal of Molecular Biology 2005, 349:847-860. 101. G Chen, YR Bi, N Li: EGY1 encodes a membrane-associated and ATP-independent metalloprotease that is required for chloroplast development. Plant J 2005, 41:364-75. 102. P Silva, E Thompson, S Bailey, O Kruse, CW Mullineaux, C Robinson, NH Mann, PJ Nixon: FtsH is involved in the early stages of repair of photosystem II in Synechocystis sp PCC 6803. Plant Cell 2003, 15:2152-64. 103. M Singh, Y Yamamoto, K Satoh, EM Aro, E Kanervo: Post- illumination-related loss of photochemical efficiency of Photosystem II and degradation of the D1 protein are temperature-dependent. J Plant Physiol 2005, 162:1246-53. 104. M Lindahl, C Spetea, T Hundal, AB Oppenheim, Z Adam, B Andersson: The thylakoid FtsH protease plays a role in the light-

42 induced turnover of the photosystem II D1 protein. Plant Cell 2000, 12:419-31. 105. K Haussuhl, B Andersson, I Adamska: A chloroplast DegP2 protease performs the primary cleavage of the photodamaged D1 protein in plant photosystem II. Embo J 2001, 20:713-22. 106. X Sun, L Peng, J Guo, W Chi, J Ma, C Lu, L Zhang: Formation of DEG5 and DEG8 complexes and their involvement in the degradation of photodamaged photosystem II reaction center D1 protein in Arabidopsis. Plant Cell 2007, 19:1347-61. 107. C Spetea, T Hundal, F Lohmann, B Andersson: GTP bound to chloroplast thylakoid membranes is required for light-induced, multienzyme degradation of the photosystem II D1 protein. Proc Natl Acad Sci U S A 1999, 96:6547-52. 108. PF Huesgen, H Schuhmann, I Adamska: The family of Deg proteases in cyanobacteria and chloroplasts of higher plants. Physiologia Plantarum 2005, 123:413-420. 109. PF Huesgen, H Schuhmann, I Adamska: Photodamaged D1 protein is degraded in Arabidopsis mutants lacking the Deg2 protease. FEBS Lett 2006, 580:6929-32. 110. E Kapri-Pardes, L Naveh, Z Adam: The Thylakoid Lumen Protease Deg1 Is Involved in the Repair of Photosystem II from Photoinhibition in Arabidopsis. Plant Cell 2007:In press. 111. K Ito, Y Akiyama: Cellular functions, mechanism of action, and regulation of FtsH protease. Annu Rev Microbiol 2005, 59:211-31. 112. S Krzywda, AM Brzozowski, C Verma, K Karata, T Ogura, AJ Wilkinson: The crystal structure of the AAA domain of the ATP- dependent protease FtsH of Escherichia coli at 1.5 A resolution. Structure 2002, 10:1073-83. 113. C Herman, S Prakash, CZ Lu, A Matouschek, CA Gross: Lack of a robust unfoldase activity confers a unique level of substrate specificity to the universal AAA protease FtsH. Mol Cell 2003, 11:659-69. 114. S Chiba, Y Akiyama, K Ito: Membrane protein degradation by FtsH can be initiated from either end. J Bacteriol 2002, 184:4775-82. 115. J Chen, JJ Burke, J Velten, Z Xin: FtsH11 protease plays a critical role in Arabidopsis thermotolerance. Plant J 2006, 48:73-84. 116. MR Aluru, F Yu, A Fu, S Rodermel: Arabidopsis variegation mutants: new insights into chloroplast biogenesis. J Exp Bot 2006, 57:1871-81. 117. F Yu, A Fu, M Aluru, S Park, Y Xu, H Liu, X Liu, A Foudree, M Nambogga, S Rodermel: Variegation mutants and mechanisms of chloroplast biogenesis. Plant Cell Environ 2007, 30:350-65. 118. M Chen, M Jensen, S Rodermel: The yellow variegated mutant of Arabidopsis is plastid autonomous and delayed in chloroplast biogenesis. J Hered 1999, 90:207-14.

43 119. F Yu, S Park, SR Rodermel: The Arabidopsis FtsH metalloprotease gene family: interchangeability of subunits in chloroplast oligomeric complexes. Plant J 2004, 37:864-76. 120. F Yu, S Park, SR Rodermel: Functional redundancy of AtFtsH metalloproteases in thylakoid membrane complexes. Plant Physiol 2005, 138:1957-66. 121. J Komenda, HA Hassan, BA Diner, RJ Debus, J Barber, PJ Nixon: Degradation of the Photosystem II D1 and D2 proteins in different strains of the cyanobacterium Synechocytis PCC 6803 varying with respect to the type and level of psbA transcript. Plant Mol Biol 2000, 42:635-45. 122. O Ostersetzer, Z Adam: Light-stimulated degradation of an unassembled Rieske FeS protein by a thylakoid-bound protease: the possible role of the FtsH protease. Plant Cell 1997, 9:957-65. 123. JB Peltier, J Ytterberg, DA Liberles, P Roepstorff, KJ van Wijk: Identification of a 350-kDa ClpP protease complex with 10 different Clp isoforms in chloroplasts of Arabidopsis thaliana. Journal of Biological Chemistry 2001, 276:16318-16327. 124. B Zheng, TM MacDonald, S Sutinen, V Hurry, AK Clarke: A nuclear- encoded ClpP subunit of the chloroplast ATP-dependent Clp protease is essential for early development in Arabidopsis thaliana. Planta 2006, 224:1103-1115. 125. LLE Sjogren, TM MacDonald, S Sutinen, AK Clarke: Inactivation of the clpC1 gene encoding a chloroplast H sp100 molecular chaperone causes growth retardation, leaf chlorosis, lower photosynthetic activity, and a specific reduction in photosystem content. Plant Physiology 2004, 136:4114-4126. 126. LLE Sjogren, TM Stanne, B Zheng, S Sutinen, AK Clarke: Structural and functional insights into the chloroplast ATP-dependent Clp protease in Arabidopsis. Plant Cell 2006, 18:2635-2649. 127. A Rudella, G Friso, JM Alonso, JR Ecker, KJ van Wijk: Downregulation of ClpR2 leads to reduced accumulation of the ClpPRS protease complex and defects in chloroplast biogenesis in Arabidopsis. Plant Cell 2006, 18:1704-1721. 128. S Koussevitzky, TM Stanne, CA Peto, T Giap, LLE Sjogren, YD Zhao, AK Clarke, J Chory: An Arabidopsis thaliana virescent mutant reveals a role for ClpR1 in plastid development. Plant Molecular Biology 2007, 63:85-96. 129. EP Beers, JM McDowell: Regulation and execution of programmed cell death in response to pathogens, stress and developmental cues. Curr Opin Plant Biol 2001, 4:561-7. 130. B Katoch, S Sebastian, S Sahdev, H Padh, SE Hasnain, R Begum: Programmed cell death and its clinical implications. Indian J Exp Biol 2002, 40:513-24. 131. A Danon, P Gallois: UV-C radiation induces apoptotic-like changes in Arabidopsis thaliana. FEBS Lett 1998, 437:131-6.

44 132. A Zuppini, L Navazio, P Mariani: Endoplasmic reticulum stress- induced programmed cell death in soybean cells. J Cell Sci 2004, 117:2591-8. 133. H Kuriyama, H Fukuda: Developmental programmed cell death in plants. Curr Opin Plant Biol 2002, 5:568-73. 134. AH Gunawardena, DM Pearce, MB Jackson, CR Hawes, DE Evans: Characterisation of programmed cell death during aerenchyma formation induced by ethylene or hypoxia in roots of maize (Zea mays L.). Planta 2001, 212:205-14. 135. A Jones: Does the plant mitochondrion integrate cellular stress and regulate programmed cell death? Trends in Plant Science 2000, 5:225-230. 136. BF Quirino, YS Noh, E Himelblau, RM Amasino: Molecular aspects of leaf senescence. Trends Plant Sci 2000, 5:278-82. 137. H Thomas, HJ Ougham, C Wagstaff, AD Stead: Defining senescence and death. J Exp Bot 2003, 54:1127-32. 138. WG van Doorn: Plant programmed cell death and the point of no return. Trends Plant Sci 2005, 10:478-83. 139. V Buchanan-Wollaston, S Earl, E Harrison, E Mathas, S Navabpour, T Page, D Pink: The molecular analysis of leaf senescence--a genomics approach. Plant Biotechnol J 2003, 1:3-22. 140. H Fukuda: Programmed cell death of tracheary elements as a paradigm in plants. Plant Mol Biol 2000, 44:245-53. 141. A Groover, AM Jones: Tracheary element differentiation uses a novel mechanism coordinating programmed cell death and secondary cell wall synthesis. Plant Physiol 1999, 119:375-84. 142. A Zelisko, G Jackowski: Senescence-dependent degradation of Lhcb3 is mediated by a thylakoid membrane-bound protease. J Plant Physiol 2004, 161:1157-70. 143. P Zimmermann, M Hirsch-Hoffmann, L Hennig, W Gruissem: GENEVESTIGATOR. Arabidopsis microarray database and analysis toolbox. Plant Physiol 2004, 136:2621-32.

45