Mechanism of Plastid Division: From a Bacterium to an Organelle

Chloroplasts arose more than 1 billion years ago, when a free-living cyanobacterium became an endosymbiont in a eukaryotic host cell. Several lines of evidence indicate that all chloroplasts and their nonphotosynthetic relatives (plastids) are directly or indirectly derived from a single endosymbiotic event (Reyes-Prieto et al., 2007; Gould et al., 2008; Archibald, 2009; Fig. 1). The original endosymbiosis of the cyanobacterium gave rise to plastids (primary plastids) in the Archaeplastida, which are made up of the Glaucophya, Rhodophyta (red algae), and Viridiplantae (green algae and land plants). All other plastids, such as those found in kelps, dinoflagellates, malaria parasites (these three belong to the Chromalveolata), euglenids (Excavata), and chlorarachniophytes (Rhizaria), were established by subsequent endosymbiotic events in which eukaryotic algae had become integrated into other eukaryotes. Recent phylogenetic studies have classified eukaryotes into six supergroups: Amoebozoa, Opisthokonta (fungi and animals), Archaeplastida, Chromalveolata, Excavata, and Rhizaria. Thus, four of the six eukaryotic supergroups possess plastids (Reyes-Prieto et al., 2007; Gould et al., 2008; Archibald, 2009; Fig. 1). A schematic view of the origin and spread of plastids in the putative six eukaryotic supergroups (modified from Archibald, 2009). The broken lines denote uncertainty of branch positions in the tree. Red algae and groups containing plastids of red algal origin are shown in red. Viridiplantae (green algae and land plants) and groups containing plastids of green algal origin are shown in green. Arrows indicate the primary endosymbiotic event of a cyanobacterium (1st) and secondary endosymbiotic events (2nd). The position of secondary endosymbiotic events in the tree is still uncertain at present. Note that some dinoflagellates possess plastids of stramenopile, haptophyte, or green algal origin, although those are not indicted in the scheme. It is suggested that these dinoflagellates replaced the original red algal chloroplast with chloroplasts of other origin by another endosymbiotic event (Gould et al., 2008). The conversion of a cyanobacterium to plastids required several steps (Reyes-Prieto et al., 2007; Gould et al., 2008; Archibald, 2009). Theoretically, a specific relationship developed between an endosymbiont and host, such as a predator-prey relationship. Subsequently, mechanisms controlling metabolic exchange between the two developed. Most of the genes that were once in the endosymbiont genome were either lost or transferred into the host nuclear genome. As a result, the size of the plastid genome has been reduced to less than one-tenth that of the free-living cyanobacterial genome. Thus, the bulk of the plastid proteome consists of nucleus-encoded proteins that are translated on cytoplasmic ribosomes. In order to translocate proteins from the cytoplasm into chloroplasts, a protein import apparatus was developed at the envelope membrane. In addition, a mechanism that controls endosymbiotic cell division was developed such that each daughter cell would inherit the endosymbiont during cell division. Generally, algal cells contain either a single choloroplast or just a few chloroplasts. Thus, chloroplast division is synchronized with the host cell cycle such that the chloroplast divides before cytokinesis and is thus transmitted into each daughter cell. The requirement of division synchronization is supported by several other endosymbiotic relationships. For example, certain species of heterotrophic dinoflagellates engulf eukaryotic algae and use them as temporary chloroplasts (called “kleptoplasts”) for a period of days to weeks before digesting them. In contrast, a few dinoflagellate species maintain a eukaryotic algal unicell (i.e. containing the nucleus, mitochondria, Golgi apparatuses, etc.) as a permanent endosymbiont by synchronizing the endosymbiont cell division to the host cell cycle (Wouters et al., 2009; Imanian et al., 2010). A similar situation is also observed in other eukaryotic groups. Hatena arenicola has a transient green algal endosymbiont, and this photosynthetic endosymbiont is inherited by only one daughter cell during cell division. Daughter cells that have lost the endosymbiont again engulf the green alga (Okamoto and Inouye, 2005). H. arenicola (Katablepharidophyta) bears a close evolutionary relationship with cryptophytes, in which permanent chloroplasts of a red algal origin have been established by division synchronization (Fig. 1). There are also eukaryotes that possess permanent cyanobacterial endosymbionts, such as Paulinella chromatophora. This endosymbiont is consistently inherited by progeny cells as a consequence of tight synchronization of host and endosymbiotic cell cycles (Melkonian and Mollenhauer, 2005). Currently, it is largely unknown how endosymbiotic cell division is regulated by host cells in the diverse array of endosymbiotic relationships. However, studies over the last decade have rapidly provided information on the mechanism and the regulation of primary chloroplast division (Kuroiwa et al., 2008; Yang et al., 2008; Maple and Møller, 2010; Miyagishima and Kabeya, 2010; Pyke, 2010). In addition, several genome projects on the plastid-carrying eukaryotes have been initiated. These advances promise the eventual understanding of the mechanisms that coordinate the proliferation of both a host cell and an endosymbiont. The mechanism of primary plastid division, especially that of chloroplasts, has been summarized in recent reviews (Kuroiwa et al., 2008; Yang et al., 2008; Maple and Møller, 2010; Miyagishima and Kabeya, 2010; Pyke, 2010). Here, the mechanism of primary chloroplast division is briefly introduced, and then the common and diverse mechanisms underlying division in other types of plastids and photosynthetic endosymbionts will be discussed. Note that, in this review, the term “chloroplasts” is used to indicate photosynthetic plastids. Chloroplasts usually divide by binary fission (Fig. 2), but multiple fission also occurs in certain cases (Duckett and Ligrone, 1993). Earlier electron microscopic studies established that chloroplast division is performed by the simultaneous constriction of the inner and outer envelopes at the division site. Since the 1980s, electron-dense ring structures at the division site called plastid-dividing (PD) rings have been detected on both the inside and outside of the chloroplast, suggesting that chloroplast division is performed by constriction of the ring-like division complex (Kuroiwa et al., 1998). Chloroplast division by the division complex. A, Dividing chloroplasts in the moss P. patens (top panel) and localization of the FtsZ protein at the chloroplast division site in Arabidopsis (bottom panel). The green fluorescence shows the localization of FtsZ, and red shows the autofluorescence of chlorophyll. Bars = 10 μm. B, Diagram showing the chloroplast division process (top panel) and localization of components of the division complex (bottom panel). The suggested function of each component is described in the text and Table I. Recent studies, especially molecular genetic studies on leaf chloroplasts in Arabidopsis (Arabidopsis thaliana) and chloroplasts in the unicellular red alga Cyanidioschyzon merolae, have identified several components of the division complex. As observed by electron microscopy, the division complex encompasses both the inside and the outside of the two envelopes (Fig. 2). All of the known components are encoded in the host nuclear genome, except for a few components of cyanobacterial origin encoded in the chloroplast genome in certain algal lineages (Kuroiwa et al., 2008; Yang et al., 2008; Maple and Møller, 2010; Miyagishima and Kabeya, 2010; Pyke, 2010). Consistent with the endosymbiotic origin of plastids, the stromal portion of the chloroplast division complex is descended from the cyanobacterial cell division machinery (Figs. 2 and 3; Table I). As in bacteria, Filamenting temperature-sensitive mutant Z (FtsZ; a prokaryotic tubulin-like GTPase) self-assembles into a ring structure on the stromal side of the chloroplast division site and is involved in the division process (Osteryoung and Vierling, 1995; Osteryoung et al., 1998; Strepp et al., 1998; Mori et al., 2001; Vitha et al., 2001). In addition to FtsZ, certain other components, such as ACCUMULATION AND REPLICATION OF CHLOROPLAST 6 (ARC6; an ortholog of cyanobacterial division protein Ftn2; Vitha et al., 2003), are also descended from the cyanobacterial division machinery. However, many other components were lost after the endosymbiotic event (Figs. 2 and 3). Evolution of the primary chloroplast division machinery. Components of the division machinery in the different phylogenetic lineages are shown. The details are available in Yang et al. (2008) and Miyagishima and Kabeya (2010). The references are indicated in the text. The references are indicated in the text. Although plastids have retained some of the cyanobacterial division machinery, both duplication and differentiation of the cyanobacteria-descended components have played significant roles in the evolution of the plastid division machinery (Figs. 2 and 3; Table I). While cyanobacterial genomes encode a single FtsZ protein, two phylogenetically and functionally distinct FtsZ proteins evolved in algae and plants by gene duplication and differentiation (Osteryoung et al., 1998; Osteryoung and McAndrew, 2001; Stokes and Osteryoung, Miyagishima et al., proteins on the stromal side of the division site et al., 2001; et al., in Viridiplantae has retained a (Osteryoung and McAndrew, that is for to the inner protein et al., 2005). In contrast, the and not with et al., but with et al., a chloroplast division protein to Viridiplantae et al., also evolved by of It consists of an and the which is known to to in other The duplication of the FtsZ protein and subsequent of the also in red algae and in after the secondary endosymbiotic from the event that in Viridiplantae et al., and of which in a common of in in the division complex et al., 2009). FtsZ et al., and the 2 protein in the et al., 2008). In contrast, of to plants) FtsZ an with and is required for the of of to et al., 2009). In addition to the of cyanobacteria-descended some components of eukaryotic host origin have become integrated into the chloroplast division machinery (Figs. 2 and 3; Table is a of the eukaryotic of and on the side of the chloroplast division complex et al., Miyagishima et al., of this on the and it is suggested that constriction and of the in fission and 2008). is to is not involved in plastid division but is involved in cytokinesis in Arabidopsis and the et al., 2008). Thus, the in chloroplast division evolved from the of the eukaryotic host cell. and are to land plants and are required for the of et al., It that proteins the cyanobacteria-descended stromal complex and the complex et al., 2009; 2 and 3). However, it is not known proteins directly with or how the two are in which not have The references are indicated in the text. The references are indicated in the text. The ring has been detected by electron in several different lineages of algae and land plants to usually have two the outer ring on the side of the outer and the inner ring on the stromal side of the inner envelope (Kuroiwa et al., 1998). a ring is observed in the et al., 2001; et al., 2008). A recent that was performed by the plastid division complex in the red alga that the outer ring is a of et al., 2010). These are with the protein, which is to the eukaryotic proteins are known to the of the The of the outer ring have been observed in several different lineages (Kuroiwa et al., and of are encoded in land genomes et al., 2010). However, the genomes of other plastid-carrying such as green algae and the how the is is uncertain at present. Chloroplasts divide at the two daughter chloroplasts of size (Fig. 2). The mechanism for the of the FtsZ ring is also descended from the cyanobacterial machinery known as the (Figs. 2 and 3; Table I). The cyanobacterial and proteins et al., Miyagishima et al., et al., 2009). In bacteria, is an of FtsZ and a complex with in and in the of such that the FtsZ ring only at the position et al., The mechanism of the division site has also been the endosymbiotic event (Figs. 2 and 3; and and have been although certain algal lineages and the moss patens still possess a protein et al., 2008; Miyagishima and Kabeya, 2010). A recent that the function of in Viridiplantae et al., of chloroplast division, an of the FtsZ the of with et al., and et al., 2009). The mutant multiple FtsZ rings a single chloroplast, similar to the Arabidopsis mutant and and et al., CHLOROPLAST 1 a protein to land is also involved in the of the plastid FtsZ to the and the mutant multiple FtsZ rings in a single chloroplast et al., 2009). The is in algae and land but the genome of the red alga not encode of the proteins described Currently, it is not known how the FtsZ ring is at the in this in Arabidopsis et al., et al., and et al., et al., that the plastid division complex is from the inside to the outside of the chloroplast before the of division site in the order of the FtsZ inner outer and ring et al., et al., 2009; et al., 2010; Fig. 2). Thus, information is from the stromal complex that is descended from the cyanobacterial endosymbiont to the complex of host eukaryotic Currently, how the division complex the two envelopes is a of FtsZ self-assembles into rings inside of and of these in et al., 2008). The of and of the in fission in and 2008). Thus, FtsZ and in the of in plastid division. However, is a certain between FtsZ ring and the of plastid division site after the other components of the division complex have been in the division complex et al., A similar has been observed in cell division et al., et al., 2009). It is that the of the ring by the et al., but such of the has not been rings et al., Miyagishima et al., 2003), other components of the division which have been observed to rings by fluorescence In the chloroplasts in and or et al., still divide by constriction that which is to is not involved in plastid division as described studies in and in at a will on how the division complex plastids. In algae and chloroplasts are usually the only of plastid present. This is with the that the cells of and photosynthetic their Thus, from an evolutionary the chloroplast is the origin of plastids. plants have evolved a complex plastid differentiation in which all of the plastids, such as chloroplasts, and are derived from in cells and Pyke, 2005). For example, in the the division of which with cell division that cells have the of leaf cells cell division with the chloroplasts still to divide and the of chloroplasts cell and microscopic studies that all types of plastids are of division and However, in of recent advances in understanding the mechanism of chloroplast division, few studies have the mechanism the division of the plastids 2010). This is largely plastids are than chloroplasts and or electron In addition, some plastids with between them called and 2008). In some multiple FtsZ rings in a single and multiple fission has been observed et al., and 2009; Pyke, 2010). of plastids in chloroplast division and of the localization of the chloroplast division proteins have shown that plastid division division machinery similar to that in chloroplasts However, these have also certain between chloroplasts and the plastids For example, is in division in the mutant et al., Pyke, and the protein is not detected in the et al., 2009). Thus, land plants have evolved a plastid division mechanism that not The division is suggested by the of a of than those in the et al. et al. Osteryoung et al. et al. Miyagishima et al. et al. et al. The division is suggested by the of a of than those in the et al. et al. Osteryoung et al. et al. Miyagishima et al. et al. et al. In addition, be a plastid division mechanism that not the known plastid division complex. the FtsZ et al., and in in both chloroplast and division et al., 1995; Pyke, the photosynthetic cells in contain plastids with chloroplasts. These plastids in the mutant et al., 2009). A similar of plastid proliferation also occurs in a chloroplast division and of the plastids are observed and Pyke, A recent that Arabidopsis in which FtsZ proteins are are and the cells still contain one or two chloroplasts et al., 2009). This that the plastids, at in the FtsZ in the although the FtsZ rings are in the division that occurs in the et al., 2008; et al., 2009). these types of plastid proliferation have been only in studies are required to the of mechanisms of plastid In this the genomes of certain lineages of and not encode Most of these not have a recent studies that the cell division in of the which is involved in in eukaryotes. In addition, when the is by a cells are to divide in the of FtsZ and 2010). other than the described have also been in plastid division in the CHLOROPLAST 1 also called Maple et al., et al., et al., genes and and et al., or et al., gene is plastid division is and plastids are as in other plastid division it is not known these proteins are directly involved in chloroplast division machinery or how these proteins are to plastid division. et al., and et al., et al., are descended from the cyanobacterial endosymbiont. of the of are as as an of FtsZ in et al., 2010). However, is between and at the of primary structure and evidence that is a of proteins are of and plastid size and and are to both the and plastids and coordinate the cell cycle as as plastid division et al., 2005). A diverse array of eukaryotic lineages possess plastids are that are of secondary endosymbiotic origin (Fig. 1). In addition, plastids of endosymbiotic origin have been found in certain species of The secondary endosymbiosis of a red algal gave rise to plastids in Chromalveolata, such as cryptophytes, of the photosynthetic dinoflagellates, and and possess chloroplasts of a green algal The of how many endosymbiotic events have rise to this However, several have indicated a for the that the red algal endosymbiosis and subsequent into chloroplasts only once in the common of (Reyes-Prieto et al., 2007; Gould et al., 2008; Archibald, 2009). This also that several such as and lineages of dinoflagellates, have lost the plastids. is known how secondary plastids in the diverse eukaryotic lineages but a few studies of chloroplast division and recent genome projects of and that a of the chloroplast division machinery in these lineages is descended from a red algal endosymbiont (Fig. The chloroplasts in are by four The inner two are descended from the inner and the outer envelopes of the primary The two are to to the of the alga and the of the host (Fig. The is with the outer nuclear envelope directly or indirectly In certain lineages of the the outer ring has been observed at the of the et al., The genome of the a of the red algal in cryptophytes, an FtsZ protein et al., 1998). In addition, genome projects have the of genes FtsZ, and in the nuclear genomes of as as a and Kabeya, 2010). studies that these proteins are of a red algal origin et al., 2008). Thus, the division of the inner of in secondary chloroplasts the primary chloroplast division machinery that is descended from the red alga (Fig. Evolution of the secondary plastid division machinery. A, of cryptophytes, and of the photosynthetic dinoflagellates evolved from a red algal endosymbiont. the endosymbiotic the red algal other than the plastid were lost (top except that a of the red algal It is unknown how these secondary plastids but a few of the components of the division complex have been identified or (bottom panel). B, of the protein at the chloroplast division site in the red alga and in the The green fluorescence shows the localization of and red shows the autofluorescence of chlorophyll. Bars = 2 μm. Currently, it is not known how the outer of but of species that the division is by a mechanism that is distinct from the primary chloroplast division. In certain the outer of divides in the as the inner but the constriction of the outer that of the inner In addition, have been observed in the between the two of (called the at the division site 2005). In such as in the division of the inner and the outer is not Thus, a of chloroplasts with two inner a single outer of et al., such as malaria parasites also possess nonphotosynthetic plastids (called In this the plastid division simultaneous constriction of the four In to genomes not encode of the known primary plastid division A recent in the that the protein on the side of the division site (i.e. on the it is required for the division et al., 2009; Fig. is in but to the (i.e. is not a of the in the primary chloroplast et al., 2009). Thus, have lost the primary chloroplast division machinery and have developed mechanisms for division. It is not known how this or the of the primary chloroplast division machinery is to the of photosynthetic of chloroplast division in photosynthetic relatives of such as et al., 2008; Fig. would be to on this unicellular and algae have either one or just a few chloroplasts cell. Thus, chloroplasts divide during a specific of the host cell cycle before the host cell cytokinesis (Fig. in red algae et al., et al., a green alga et al., and a et al., that the regulation of algal chloroplast division on the cell of the chloroplast division division in algae and In unicellular which usually have just one or only a few chloroplasts chloroplast division occurs once cell cycle to In the of and cell is the in leaf cells to into chloroplasts. leaf cell division but nuclear in cells with a of leaf cell chloroplast division still cell division, but the division in an of the chloroplast In contrast, land cells contain of plastids, except that cells contain a single chloroplast or a few chloroplasts. In the is a between cell cycle and division and plastids divide in the cell chloroplasts in the moss P. Fig. 2). the plastid division in with cell in of plastid size and (Fig. studies have indicated that the chloroplast division and and Pyke, 2005). Currently, it is largely unknown how plastid division is regulated in land but a recent that proteins the of plastid division in land plants et al., 2009). The of proteins in the division complex during cell either an or a of the and in a or in the plastid division For example, the is in the of the The of but not other components of the division during leaf in a of the chloroplast division and an in chloroplast It is not known how proteins the chloroplast division but it that proteins the of division site constriction et al., 2009). plastids in other the chloroplasts of algae (called have retained a between the two envelope that is descended from the cyanobacterial endosymbiont. In plastids of other the between the inner and the outer envelopes at the division site is the as that in other of the In contrast, the inner envelope in by the of the than the outer envelope of the relationship between the cytoplasmic and the outer in cyanobacterial cell division. the between the two envelopes at the division site than that in other of the chloroplast in and et al., 2009). chloroplast division division machinery on the stromal side et al., and a structure similar to the inner ring has been detected by electron However, the outer ring is not and et al., 2009). In addition, all of the chloroplast division genes of host origin, are from the the and genes are and Kabeya, 2010). It that the outer envelope division in a mechanism that is similar to that in However, studies on division of the and the outer in are just 2010). from the genome, and studies of cell and outer division, will the relationship between the of the and the evolution of the plastid division machinery. have been detected in plastids other than but it be that some algal and genomes still encode cyanobacteria-descended proteins that are involved in or to in of the nucleus-encoded or genes chloroplast division in the moss P. patens et al., et al., 2009). The chloroplast genomes of certain green algal lineages still and However, the of these gene in plastids are Although in eukaryotes largely on chloroplasts, eukaryotic species contain or eukaryotic photosynthetic endosymbionts other than plastids. of that have more recent cyanobacterial such as et al., and P. and Mollenhauer, into how eukaryotes are to a permanent endosymbiont in a process of the evolutionary origin of plastids. P. a photosynthetic that evolved by the endosymbiosis of a from and more than plastids et al., 2005). The (called a is inherited by synchronization of the host and the endosymbiont cell it is not known how the division is by the host cell. Recent studies that the genome has been reduced with free-living but not as as the primary plastids et al., 2008). The genome still of the known cyanobacterial cell division the plastid genome. The are the of a cyanobacterial of and of the portion of of et al., 2008). These or other cyanobacterial cell division genes would be for transferred to the host genome and to be regulated by the host cell Recent of the P. nuclear genome the of photosynthetic genes that were transferred from the endosymbionts and 2009; et al., will how the host cell has established the of endosymbiont cell division in order to a permanent endosymbiotic relationship. eukaryotic lineages have photosynthetic by endosymbiotic of photosynthetic endosymbionts or plastids on the regulation of endosymbiont or plastid division by the host cell. Recent studies have significant in understanding of the mechanism and regulation of plastid division, but many still to be and are The the mechanism of plastid division are how the division complex and chloroplasts and how the is In of another is are common with to the mechanisms that endosymbiont or division. There the evolution of the division machinery is similar to that of plastids. evolved from an endosymbiont, and division a In addition, division in some eukaryotic lineages also the FtsZ protein of origin et al., studies not only to the understanding of division but also the mechanisms of fission and eukaryotic evolution by