Download Organelle division: From coli to chloroplasts

Dispatch
R619
Organelle division: From coli to chloroplasts
Joe Lutkenhaus
FtsZ, an ancestral homolog of eukaryotic tubulin,
assembles into the cytokinetic Z ring that directs cell
division in bacteria. Recent results indicate that FtsZ
is also used for division by chloroplasts, though not
by mitochondria.
Address: Department of Microbiology, Molecular Genetics and
Immunology, University of Kansas Medical Center, Kansas City,
Kansas 66160, USA.
E-mail: [email protected]
Current Biology 1998, 8:R619–R621
http://biomednet.com/elecref/09609822008R0619
© Current Biology Publications ISSN 0960-9822
There is compelling evidence that the eukaryotic
organelles, chloroplasts and mitochondria, are evolutionarily derived from bacteria [1]. Chloroplasts are most closely
related to cyanobacteria, and mitochondria to members of
the alpha division of proteobacteria. For these organelles to
be distributed among their eukaryotic host cells during
cytokinesis, they must themselves undergo division.
Dumbbell-shaped chloroplasts that appear to be undergoing binary fission have long been observed in plant tissue.
The mechanism of organelle division, and the degree to
which this resembles the division process in bacteria, is
only beginning to be addressed [2]. Recent results are starting to shed light on organelle division; first, however, I shall
briefly review what we know about bacterial cell division.
Genetic studies of cell division in the bacterium Escherichia
coli have led to the identification of a dozen or so genes
that are specifically dedicated to this process [3]. Many of
these genes carry the designation fts, as the mutants that
led to their identification had a temperature-sensitive filamenting phenotype (growth for an extended period of
time in the absence of division). Among these genes, it
appears that only ftsZ is universally conserved among all
prokaryotic organisms. Homologs of ftsZ are present in all
prokaryotic genomes sequenced to date, including those of
various gram-negative, gram-positive and wall-less
(mycoplasmas) bacteria, as well as various archaea,
prokaryotic organisms occupying an evolutionary domain
distinct from bacteria and eukaryotes [4]. The sequences
of over 40 ftsZ homologs from a wide variety of prokaryotic
organisms have already been deposited in databases.
What do we know about the role of the FtsZ protein in
bacterial cell division? During the cell-division cycle, FtsZ
assembles into a cytokinetic ring, known as the Z ring,
that is present at the leading edge of the septum that
separates the two daughter cells [5–7]. This ring is
essential to bacterial cell division and disrupting the ring
immediately halts division [8]. Interestingly, an ftsZ
mutation that results in the formation of FtsZ spirals,
instead of rings, leads to spiral-shaped septa [9]. The Z
ring is a cytoskeletal element that is analogous to the contractile ring of eukaryotic cells: the Z ring determines the
septal plane in prokaryotic cells, and the contractile ring
determines the cleavage furrow in eukaryotic cells.
Although the rings are functionally equivalent, in that
both partition the replicated chromosomes and cytoplasm
into two compartments, their molecular natures are quite
different. The contractile ring is composed of actin filaments, whereas the Z ring is composed of FtsZ, almost
certainly the ancestral homolog of eukaryotic tubulins.
FtsZ and tubulins show only limited sequence similarity;
although this suggests that they might be related proteins,
the overall identity is only in the range of 10–18%, clearly
less than would be expected for homologous proteins [10].
Early this year, however, the structures of FtsZ and the αβ
tubulin dimer were reported [11,12]. The overall structures of the two proteins are nearly identical, and their
secondary structural elements are colinear over most of
their length. Moreover, the mechanism of GTP binding
and hydrolysis by the two proteins appears to be similar,
and distinct from that of other GTPases [13].
In addition to the similarities of their sequences and
structures, the proteins are also functionally similar. FtsZ
and the αβ tubulin dimer both assemble into protofilaments that are about 5 nanometres in diameter [10,14]. In
the case of tubulin, the protofilaments are further organized into microtubules, hollow cylinders that each consist
of 13 protofilaments. For FtsZ, however, it is not clear
what physiological structure the protofilaments assemble
into, as none has been observed in vivo. One possibility is
that the physiologically relevant structure is a sheet of
about 15–20 protofilaments that lies on the inner surface
of the cytoplasmic membrane [10,14]. Such a structure
would be difficult to see in thin sections. A hallmark of
microtubules is that they exhibit ‘dynamic instability’,
rapid changes in polymer length that are regulated by
GTP hydrolysis [15]. Importantly, FtsZ protofilaments are
also dynamic and regulated by GTP hydrolysis [16].
Although archaea are phylogenetically distinct from
bacteria, they have the typical cellular morphology of
prokaryotic organisms and share a common ancestor. It
was therefore gratifying to find that archaeal cells also
have FtsZ protein present in a Z ring localized at the
septum. This implies that FtsZ also plays an important
R620
Current Biology, Vol 8 No 17
part in the division of archaeal cells [17]. As organelles are
derived from bacteria, do they use a similar mechanism for
division? For mitochondria from at least one yeast species,
the answer is no, as the complete sequences of the nuclear
and mitochondrial genomes of Saccharomyces cerevisiae have
been determined and no ftsZ homolog is present.
For chloroplasts, however, the situation is different: the
discovery of an ftsZ-like sequence in a cDNA library from
Arabidopsis thaliana provided the first hint that chloroplasts might indeed use a similar division mechanism to
bacterial and archaeal cells [18]. The protein product of
this gene has a chloroplast targeting signal, and in vitro
expression experiments demonstrated that the protein is
indeed targeted to the chloroplast. The absence of an
efficient gene targeting system in Arabidopsis, however,
prevented an immediate rigorous test of the role of this
ftsZ gene in chloroplast division. Meanwhile, cDNAs
encoding portions of ftsZ homologs from other plant
species appeared in the databases.
Now evidence has been reported that FtsZ is indeed
required for chloroplast division in the moss Physcomitrella
patens [19]. Using the polymerase chain reaction (PCR)
with degenerate primers, a full-length cDNA encoding an
FtsZ homolog was obtained. Having the cloned gene in
hand, advantage was taken of the efficient homologous
recombination system in this moss to create ftsZ knockout
plants. The mutants were identified among the transformants by the striking phenotype of one huge chloroplast
per cell, instead of the usual fifty. As expected from S. cerevisiae genome sequence, mitochondria were unaffected.
Interestingly, the elongated morphology of the single
chloroplast in ftsZ mutant moss cells is reminiscent of the
filaments formed by E. coli cells lacking ftsZ. As the
mutant plants live, chloroplast division is obviously not
essential, but the question left unanswered is how the
chloroplast survives and is disseminated during cytokinesis of its host eukaryotic cell. One possibility is that the
chloroplast is trapped in the cleavage furrow and bisected
during cytokinesis, and that its ends are somehow
‘healed’. In this aspect, the phenotype is somewhat
reminiscent of ftsZ mutant cells of the filamentous
bacterium Streptomyces coelicolor [20]. Surprisingly, the
mutant S. coelicolor cells survive despite the absence of
septa. Even more surprisingly, the mutant can be
propagated by maceration of the mycelia, indicating that
the broken filaments are somehow able to heal their ends
at some frequency.
The importance of ftsZ in chloroplast division has now
also been documented in Arabidopsis (K. Osteryoung,
personal communication). Arabidopsis appears to have at
least three ftsZ genes, but the product of only one
appears to be targeted to the interior of the chloroplast.
Silencing of either of two of the Arabidopsis ftsZ genes,
using an antisense approach, led to a dramatic reduction
in chloroplast number, similar to the effect of inactivating
ftsZ on moss cells.
The requirement for ftsZ for chloroplast division is thus
confirmed, but many questions remain. What are the functions of the different FtsZ proteins? Do all plant cells
contain multiple ftsZ genes? Some bacteria and archaea
have more than one ftsZ gene, although no distinct function has been ascribed to the second copy. In plant cells,
could one FtsZ act within chloroplasts and the other in the
cytoplasm? Do the FtsZ proteins localize to the two concentric electron-dense rings seen at the septum in electron
micrographs of dividing chloroplasts [2]? These concentric
rings, one at the inner surface of the chloroplast envelope
and the other at the outer surface, have been implicated in
chloroplast division.
If, as shown by the work on ftsZ of the moss P. patens, and
by the additional plant ftsZ homologs in the databases,
chloroplasts indeed use FtsZ for division, why do
mitochondria not do the same? One possible answer is that
the acquisition of chloroplasts by eukaryotes is a more
recent evolutionary event than acquisition of mitochondria, so that chloroplasts may therefore have retained more
bacterial-like systems. The components of the mitochondrial division system are unknown, but they might be
related to some other process characteristic of eukaryotic
cells, such as vesicle budding. Another example of where
chloroplasts are more ‘bacteria-like’ than mitochondria
involves protein transport. Chloroplasts contain the bacterial SecA component of the protein secretion machinery,
whereas mitochondria appear not to — at least no secA
homolog was found in the yeast genome [21].
If ftsZ is so conserved during evolution of prokaryotic
cells, what about the other essential E. coli cell division
genes that have been identified? Many of these — ftsI,
ftsW, ftsN, ftsL and ftsQ — encode membrane proteins that
are involved, or thought to be involved, in biosynthesis of
septal cell-wall peptidoglycan [3]. It is not surprising,
then, that homologs of these genes are not present in
mycoplasma, which lack cell walls completely, or archaea,
which lack peptidoglycan [3]. More surprising is the
restriction of ftsA homologs to peptidoglycan-containing
bacteria. FtsA is a cytoplasmic protein and a member of
the DnaK/actin/hexokinase family of ATPases [22], and
interacts physically with FtsZ [23,24]. Such a limited distribution of FtsA is consistent with it having a role as a
molecular linker between the cell-wall biosynthetic
machinery and the Z ring.
Although FtsZ and tubulins are very similar in their core
regions, important for polymerization, they are quite different in other regions. This is especially true of their
Dispatch
carboxy-terminal regions, which interact with two completely different sets of proteins. With tubulin, the carboxyl tail lies on the outside of the microtubule, where it
can interact with accessory proteins, including various
motor proteins [15]. For FtsZ, the carboxyl tail is expected
to be on one surface of the protofilament, positioned to
interact with the membrane or accessory proteins. In the
case of peptidoglycan-containing bacteria, the accessory
proteins include FtsA and ultimately the peptidoglycan
synthesis machinery [3]. For prokaryotes and chloroplasts
that lack peptidoglycan, the accessory proteins are
expected to be quite different, perhaps explaining the
extreme variability of the carboxyl tail among evolutionary
distant FtsZ homologs [3]. What remains to be identified,
though, are conserved proteins that modulate the position
and dynamics of the Z ring in prokaryotic cells.
The work on FtsZ over the past few years has had a
unifying effect on biology. On the one hand, the many
similarities to tubulin — especially the regulation of
polymerization dynamics by GTP hydrolysis [16] —
indicate that, despite the sequence divergence, FtsZ is the
long sought ancestral homolog of this key component of
the eukaryotic cytoskeleton. On the other hand, FtsZ and
the Z ring appear to be used by all prokaryotic organisms,
regardless of their cell shape or cell envelope composition,
for cell division. The addition of chloroplasts to the list of
Z-users also has a unifying effect, as this organelle is evolutionarily derived from a prokaryote. The fact that mitochondria cannot be added is surprising considering that the
Z ring is so highly conserved; it will be of interest to learn
what mechanism has taken over this role in mitochondria.
References
1. Gray MW: Origin and evolution of organelle genomes. Curr Opin
Genet Dev 1993, 3:884-890.
2. Kuroiwa T, Kuroiwa H, Sakai A, Takahashi H, Toda K, Itoh R: The
division apparatus of plastids and mitochondria. Int Rev Cytol
1998, 181:1-41.
3. Lutkenhaus J, Addinall SG: Bacterial cell division and the Z ring.
Annu Rev Biochem 1997, 66:93-116.
4. Woese CR, Kandler O, Wheelis ML: Towards a natural system of
organisms: proposal for the domains Archaea, Bacteria and
Eucarya. Proc Natl Acad Sci USA 1990, 87:4576-4579.
5. Bi E, Lutkenhaus J: FtsZ ring structure associated with division in
Escherichia coli. Nature 1991, 354:161-164.
6. Addinall SG, Bi E, Lutkenhaus J: FtsZ ring formation in fts mutants.
J Bacteriol 1996, 178:3877-3884.
7. Sun Q, Margolin W: FtsZ dynamics during the division cycle of live
Escherichia coli cells. J Bacteriol 1998, 180:2050-2056.
8. Addinall SG, Cao C, Lutkenhaus J: Temperature shift experiments
with an ftsZ84 (Ts) strain reveal rapid dynamics of FtsZ
localization and indicate that the Z ring is required throughout
septation and cannot reoccupy division sites once constriction
has initiated. J Bacteriol 1997, 179:4277-4284.
9. Addinall SG, Lutkenhaus J: FtsZ spirals and arcs determine the
shape of the invaginating septa in some mutants of Escherichia
coli. Mol Microbiol 1996, 22:231-238.
10. Mukherjee A, Lutkenhaus J: Guanine nucleotide-dependent
assembly of FtsZ into filaments. J Bacteriol 1994, 176:2754-2758.
11. Lowe J, Amos L: Crystal structure of the bacterial cell-division
protein FtsZ. Nature 1998, 391:203-206.
12. Nogales E, Wolf SG, Downing KH: Structure of the alpha beta tubulin
dimer by electron crystallography. Nature 1998, 391:199-203.
R621
13. Nogales E, Downing KH, Amos LA, Lowe J: Tubulin and FtsZ form a
distinct family of GTPases. Nat Struct Biol 1998, 5:451-458.
14. Erickson HP, Taylor DW, Taylor KA, Bramhill D: Bacterial cell
division protein FtsZ assembles into protofilament sheet and
minirings, structural homologs of tubulin polymers. Proc Natl
Acad Sci USA 1996, 93:519-523.
15. Desai A, Mitchison TJ: Microtubule polymerization dynamics. Annu
Rev Cell Dev Biol 1997, 13:83-117.
16. Mukherjee A, Lutkenhaus J: Dynamic assembly of FtsZ regulated by
GTP hydrolysis. EMBO J 1998, 17:462-469.
17. Wang X, Lutkenhaus J: FtsZ ring: the eubacterial division
apparatus conserved in archaebacteria. Mol Microbiol 1996,
21:313-320.
18. Osteryoung KW, Vierling E: Conserved cell and organelle division.
Nature 1995, 376:473-474.
19. Strepp R, Scholz S, Kruse S, Speth V, Reski R: Plant nuclear gene
knockout reveals a role in plastid division for the homolog of the
bacterial cell division protein FtsZ, an ancestral tubulin. Proc Natl
Acad Sci USA 1998, 95:4368-4373.
20. McCormick JR, Su EP, Driks A, Losick R: Growth and viability of
Streptomyces coelicolor mutant for the cell division gene ftsZ. Mol
Microbiol 1994, 14:243-254.
21. Glick BS, Von Heijne G: Saccharomyces cerevisiae mitochondria
lack a bacterial-type sec machinery. Protein Sci 1996,
5:2651-2652.
22. Bork P, Sander C, Valencia A: An ATPase domain common to
prokaryotic cell cycle proteins, sugar kinases, actin and hsp70
heat shock proteins. Proc Natl Acad Sci USA 1994, 89:7290-7294.
23. Wang X, Huang J, Mukherjee A, Cao C, Lutkenhaus J: Analysis of the
interaction of FtsZ with itself, GTP and FtsA. J Bacteriol 1997,
179:5551-5559.
24. Ma X, Sun Q, Wang R, Singh G, Jonietz EI, Margolin W: Interactions
between heterologous FtsA and FtsZ proteins at the FtsZ ring.
J Bacteriol 1997, 179:6788-6797.