The cleavage furrow contains also scaffolding proteins, such as anillin and septins. Anillin associates with F-actin, myosin, septins, and activated RhoA, it localizes to the furrow at early stages of cytokinesis, and it is important for its stability, but it is not essential for its ingression
[5]. Instead, it might stabilize the furrow and be important for later stages of cytokinesis. Indeed, anillin remains in the cytoplasmic bridge, even after myosin and actin have dissociated. Another class of scaffolding proteins are septins: they are GTP-binding proteins that form filaments and localize to the cytokinetic ring
[6]. Several human septins localize to the central spindle and midbody during anaphase and cytokinesis. Septins participate in the regulation of actin and microtubule dynamics, directly interact with anillin, and allow for the full activation of myosin that is necessary for cytokinesis. Interestingly, septins may form a barrier that restricts the diffusion of membrane proteins in the furrow since they directly bind the plasma membrane
[7].
Scission or abscission is the final step of cytokinesis and requires remodeling of the plasma membrane to separate sister cells. The evolutionary conserved Endosomal Sorting Complex Required for Transport (ESCRT) proteins drive this process
[20]. Abscission occurs after CR has completely contracted, myosin II and filamentous actin have dissociated, and the furrow has formed an intracellular bridge rich in antiparallel microtubules with a diameter of 1 μm, the midbody. This structure is essential for proper cytokinesis completion that requires a plasma membrane remodeling that is driven by Anillin and septins, possibly fusing transport vesicles. Interestingly, a centrosome component, centriolin, has a role in vesicle targeting and fusion in the bridge
[21]. Mammalian centriolin is the homologue of budding yeast Nud1 and of fission yeast Cdc11 that are MEN/SIN components that regulate mitotic exit and cytokinesis. Likely, all of these proteins are involved in cytokinesis completion. In order to complete abscission, the activity of microtubule-severing proteins cuts the interdigitating MTs in the midbody, causing its breakage
[22].
3. Cytokinesis in Budding Yeast
Cytokinesis can be studied in model organisms since several proteins and most basic cytokinesis processes are evolutionarily conserved.
Table 1 shows a list of evolutionarily conserved proteins. Saccharomyces cerevisiae is a unicellular eukaryotic organism that offers many experimental advantages, among which: a fast cell cycle, easy and quick growth in solid and liquid medium, cellular morphology that changes during different cell cycle phases, haploid and diploid cycle, easy genetic manipulation, and analysis. An interesting feature is that S. cerevisiae divides asymmetrically, with the daughter cell being generated as a bud from the mother cell body. During late G1 phase of the cell cycle, an intense protein relocalization to a specific place of the cell cortex induces a strong polarization of the cells and culminates in bud emergence.
Table 1. Evolutionarily conserved proteins involved in cytokinesis.
Generic Name |
S. cerevisiae |
C. elegans |
D. melanogaster |
Humans |
Structural proteins |
|
|
|
Septins |
Cdc3, Cdc10, Cdc11, Cdc12, Shs1 |
UNC-59, UNC-61 |
Peanut, Sep1,2,4,5 |
Septin 1,4,5,6,7,8,10,11,14 |
Cdc42 pathway |
Cdc42, Cdc24 |
CDC-42 |
Cdc42 |
Cdc42 |
Actin |
Act1 |
ACT1, 3, 5 |
Actin (multiple genes) |
Actin (6 genes) |
Myosin |
Myo1, Mlc1, Mlc2 |
NMY-2, MLC-4 |
Zipper, Mlc-c, Spaghetti Squash |
Myosin II, Myosin ELC, Myosin RLC |
Rho proteins and regulators |
Hof1, Rho1 |
RHO-1, ECT-2 |
RhoA, Pbl |
RhoA, ECT2 |
Centriolin |
Nud1 |
CEP110 |
Centriolin |
CNTRL (Cep110) |
Formin |
Bni1, Bnr1 |
CYK-1 |
Diaphanous |
Dia1 |
IQGAP |
Iqg1 |
PES-7 |
- |
IQGAP1, IQGAP2, IQGAP3 |
Survivin |
Bir1 |
BIR-1 |
Scapolo |
Survivin |
Anillin |
Bud4 |
ANI-1, ANI-3 |
Scraps |
hAnillin |
Regulatory elements |
|
|
|
Aurora B |
Ipl1 |
AIR-2 |
Aurora B |
Aurora B |
Polo kinase |
Cdc5 |
PLK-1 |
Polo |
Plk1 |
MEN/SIN pathway |
Cdc14, Cdc15, Dbf2, Cbk1 |
CDC-14, WTS-1, YAP-1, EGL-44 |
Cdc14, Hippo, Mats, Warts |
Cdc14A, Cdc14B, MST2, LATS1, LATS2 |
The Rho-type GTPase Cdc42 concentrates at the site of bud emergence, together with its GEF Cdc24 and its GAP, and activates its effectors that organize the actin cytoskeleton to drive polarized growth and regulate the bud neck formation
[23].
3.1. Bud Neck Formation
The bud neck is the site that separates the mother cell from the daughter cell, it is not only a spatial point, but it has a very complex structure (
Figure 1). In late G1, the first recruited proteins are septins: 5 GTP-binding proteins (Cdc3, Cdc10, Cdc11, Cdc12, and Shs1) that form linear apolar filaments that assemble into 3D structures that are highly dynamic during mitosis
[24]. Septins filaments associate with the cell membrane and they form a ring around the bud neck during bud emergence and their function is essential in maintaining bud neck structure. During the progression of the cell cycle, the septin ring changes its shape becoming a rigid hourglass structure, and then splitting into two rings at the final steps of cytokinesis
[25]. Septins rearrangements are regulated by protein-protein interactions and by phosphorylation/dephosphorylation and sumoylation events
[26]. Septin ring acts as a platform for other proteins and a signaling landmark for actin cytoskeleton filaments that drive mitotic spindle positioning and orientation. In addition, several protein kinases that are important for cell cycle progression and cell cycle chekpoints are associated with septins and their localization at the bud neck depends on septins
[27]. It is important to point out that septins also act as a cortical barrier to block movement of membrane proteins thus helping to build cellular compartmentalization
[28].
Figure 1. Schematic representation of the sequential localization at the division site of important cytokinesis players. CAR: contractile actomyosin ring.
During late G1 and S phase, septins recruit myosin II heavy chain, Myo1, and the light chain Mlc2, the formin Bnr1
[29], and the scaffold protein Bni4. After anaphase onset Hof1, Iqg1, Inn1, Cyk3, the regulatory chain Mlc1, and the formin Bni1 are localized at the bud neck and in late anaphase a functional actomyosin ring is formed, thanks to actin recruitment (
Figure 1)
[30].
3.2. CAR Formation and Contraction
Budding yeast cytokinesis requires the formation and contraction of an actomyosin ring, even if the bud neck is very narrow (1 μm). The contractile actomyosin ring (CAR) is a structure composed by structural and regulatory proteins, the most important being actin and myosin.
Actin is encoded by the essential gene ACT1, it is an ATP-binding protein that can form filaments that have a polarity that is built during their formation. In yeast cells, actin can form three types of structures: actin patches, actin cables, and an actin ring
[23]. Actin patches are important for cell polarity formation and maintenance. Actin cables serve for intracellular transport of vesicles, organelles, and mRNA, and for mitotic spindle alignment. The actin ring is a structure that forms transiently at the mother-daughter neck, it binds myosin, and is important for cytokinesis.
The second essential element of CAR is the class II myosin heavy chain, hereafter called myosin, being encoded by MYO1 gene. Myosin is the motor that slides actin filaments and induces ring contraction
[31]. The catalytic domain is located in the N-terminus, while the C-terminus contains a coiled-coiled domain that can be recruited to the site of division and it is sufficient for CAR constriction and cytokinesis. Myo1 is regulated by two light chains, an essential light chain (Mlc1) and a regulatory light chain (Mlc2). Myo1 is recruited to the emergent bud site, being present as a ring at the bud neck until cytokinesis is completed and then disassembled after contraction. In late anaphase, a complete CAR is formed around the bud neck (
Figure 2A), and, after completion of previous cell cycle events, it can contract. This process does not need Myo1 motor activity and it involves a decrease in Myo1 levels and actin depolymerization regulated by cofilin, like in animal cells (see paragraph 3).
Figure 2. Bud neck structure during cytokinesis. Top panels: side view, bottom panels: top view. (A) in late anaphase septin ring and the contractile ring (CAR) are assembled at the bud neck. (B) in late anaphase the chitin synthase Chs2 is recruited to the bud neck and the primary septum (PS, grey area) is synthetized centripetally. (C) just before cell separation PS deposition is completed leading to physical dissociation of mother and daughter cytoplasms.
After mitotic exit, these events concomitantly occur: CAR contracts symmetrically, membrane invaginates, and a septum is formed in a centripetal way.
3.3. Cell Wall Deposition
Since yeast cells have a cell wall, in order to complete cell separation, two single plasma membranes and a septum must be formed. Specific proteins are expressed in order to achieve this task (
Table 2). Even if budding yeast division site is very thin, cell division involves membrane deposition and vesicle recruitment at the division site. In concomitance with CAR contraction, new membrane is deposited, and this ensures that the chitin septum forms correctly and efficiently. One important role of membrane deposition is to deliver integral membrane proteins that forms the septum: the chitin synthases, transmembrane proteins that polymerize chitin chain that is extruded by the plasma membrane.
Table 2. Budding yeast proteins that are involved in the final steps of cytokinesis.
Name |
Functions |
Inn1 |
Couples membrane ingression with CAR contraction, regulates chitin synthase Chs2 |
Cyk3 |
Interacts with Hof1 and regulates chitin synthase Chs2 |
Chs1 |
Chitin synthase, important for septum repair |
Chs2 |
Chitin synthase, mayor role in chitin primary septum synthesis |
Chs3 |
Chitin synthase |
Chs4 |
Regulator of chitin synthase Chs3, interacts with Bni4 |
Bni4 |
Required for correct septum formation |
Fks1 |
Catalytic subunit of 1,3-beta-D-glucan synthase, involved in secondary septum synthesis |
Cts1 |
Endochitinase, digestion of chitin in the primary septum |
Eng1 |
Glucanase, digestion of the secondary septum |
Scw11 |
Glucanase, digestion of the secondary septum |
Cbk1/Mob2/Ace2 |
RAM signaling pathway, required for asymmetric daughter-specific transcription of chitinase and glucanases |
S. cerevisiae expresses three chitin synthases (Chs1, 2, and 3) with different roles. Chitin synthase Chs1 is a repair enzyme, while chitin synthase Chs3 is recruited to the bud neck during bud emergence and it is important for the integrity of the division site [
40]. Chs3 activation requires its binding to the regulatory subunit Chs4 that physically interacts with the scaffold protein Bni4. The chitin synthase Chs2 is stored in the endoplasmic reticulum and it is delivered to the bud-neck during late anaphase in secretory vesicles [
41], it is inserted in the plasma membrane and then builds the primary septum (PS), essentially a chitin disk (
Figure 3) [
42]. PS is deposited in a centripetal way (
Figure 3B).
Figure 3. After mitotic exit, the chitin synthase Chs2 is translocated from endoplasmic reticulum (ER).
Recent findings indicate that Inn1 and Cyk3 regulate Chs2 catalytic domain, thus controlling chitin deposition during cytokinesis
[32]. As PS is a fragile structure, after its deposition (
Figure 2C), 1,3-beta-D-glucan synthases synthesize secondary septa (SS) on either side of the PS
[33]. SS are glucan-rich and they have a molecular structure similar to the cell wall.
Since Chs2, Chs3, and Fks1 are localized at the division site before mitotic spindle disassembly, their activation must be tightly controlled to avoid premature spindle breakage; this task is reached by multiple regulations, among which a balance between enzyme localization and their endocytosis
[34].
3.4. Cell Separation
The final step of cytokinesis is cell separation, the event that irreversibly divides the mother from the daughter cell. Cell separation requires the partial degradation of the septum, which occurs from the daughter cell side and leaves a sign, called bud scar, on the mother cell surface (
Figure 4B). In order to degrade PS and SS partially, endochitinase Cts1 and glucanases Eng1 and Scw11 are activated specifically from the daughter side of the bud neck (
Figure 4A)
[35].
Figure 4. (A) Schematic representation of cell separation: after primary septum (PS) and secondary septa (SS) formation, Cts1, Eng1 and Scw11 activity at the daughter bud neck allows for cell separation. (B) Budding yeast exponentially growing cells stained with Calcofluor White. On the surface of mother cells these are several scars, chitin-containing rings (arrows), originated during cell division. New born daughter cells do not show bud scars.
The activity of these hydrolytic enzymes is tightly regulated: they are specifically transcribed in the daughter cell and are correctly delivered at the daughter bud neck by the Golgi pathway. The regulatory proteins Cbk1 and Mob2 localize to the neck and to the daughter nuclei at the end of mitosis
[36], in particular they form a complex with the transcriptional factor Ace2 and prevent its export from the nucleus, thus ensuring daughter-specific expression of cell separation factors
[37]. In addition, Cts1 is post-translationally modified and delivered to the daughter bud neck through the Golgi pathway
[38]. Recently, a new level of Cts1 regulation has been described: upon incomplete septation, Fir1 inhibits Cbk1 activity, thus blocking production and secretion of septum degrading enzymes
[39]. This mechanism inhibits cell separation until septation is properly executed and, therefore, enforces cytokinesis order of events.
Cell separation is the very last event of the cell cycle and it can have great implications in cell survival. Indeed, separation of daughter from the mother must occur in correlation with other cell cycle events and a well-organized trilaminar septum structure must be formed before cell separation machinery is activated in order to obtain vital progeny. In case of overactivity of Cts1, a hole in the cell wall can be formed, in this case Chs1 is activated and repairs the cell wall at the daughter side
[40].
To sum up, the successful completion of cell separation depends upon the precise construction of the trilaminar septum in coordination with the division cycle, and then on the temporal and spatial regulation of chitinase and glucanase action.
4. Checkpoint Pathways that Control Cytokinesis
Cytokinesis must be tightly coordinated with the nuclear cycle in order to maintain genetic stability. Eukaryotic cells have evolved several checkpoints and checkpoint-like mechanisms that preserve the integrity of cell division.
The mitotic checkpoint is an evolutionarily conserved signaling pathway that blocks mitosis progression in case of problems in the mitotic spindle, in chromosome attachment to MTs, or in presence of mistakes in chromosome segregation. In budding yeast, the division site is determined before mitotic spindle formation, which implies that the bipolar spindle must be properly positioned at the bud neck and aligned perpendicularly with respect to the division axis before anaphase onset. Two pathways direct spindle positioning, the dynein pathway and Kar9 pathway, and the spindle orientation checkpoint (SPOC) blocks mitotic exit and cytokinesis in the case of defects
[41]. If the checkpoint fails, cytokinesis occurs, even if the nucleus divides into the mother cell, thus causing the formation of aneuploid cells. The target of the SPOC is Tem1, the G protein at the top of MEN pathway.
Mitotic Exit Network (MEN) in S. cerevisiae and Septation Initiation Network (SIN) in S. pombe are two signaling pathways that coordinate mitosis progression from anaphase onset to cell separation. MEN is a signaling cascade with the GTPase Tem1 at the top, several protein kinases, among which the Hippo-like kinase Cdc15 and the LATS-like kinase Dbf2, and the final target is the phosphatase Cdc14. MEN pathway leads to complete Cdc14 activation that causes a decrease in mitotic kinase activity (Cdk1), triggering mitotic exit
[42]. Before anaphase onset, MEN components localize at the spindle pole bodies (the yeast centrosomes), but later they are found at the bud neck, where they ensure that mitotic exit only occurs after acto-myosin ring contraction and septum deposition
[43]. In addition, in telophase, Cdc14 dephosphorylates cytoplasmic and bud-neck associated targets of Cdk1 and contributes to reorganize actin cytoskeleton and target secretion vesicles to the bud neck
[44][45]. In S. pombe the SIN controls CAR contraction and septum deposition, rather than mitotic exit and blocks septation in the case of problems
[46]. Homologs of Cdc14, MEN, and SIN components are also found in other yeasts: Candida albicans, Aspergillus nidulans, and Ashbya gossypii. Budding yeast Cdc14 is conserved in higher eukaryotes, however these homologs do not seem to control mitotic exit, rather they are involved in other processes, such as DNA replication, DNA damage repair, nuclear organization, mitotic entry, mitotic spindle assembly, and cytokinesis
[47]. In Drosophila and human cells, MEN homologs are part of the Hippo pathway that controls centrosome duplication, cell proliferation, and apoptosis
[48][49]. The conserved Hippo pathway inhibits cell proliferation by the activation of LATS1 and LATS2 kinases and by p53 stabilization (see below).
Mammalian cells have a checkpoint that arrest cells in mitosis in response to several problems in order to preserve genetic stability and cell survival, rarely cells can escape this arrest and enter mitosis without chromosome segregation, thus becoming tetraploid
[50][51]. Importantly, another checkpoint control is active in human cells to recognize cytokinesis failure and induce proliferation arrest in G1. Indeed, in tetraploid cells, arising either by failure of mitotic spindle or of cytokinesis, the checkpoint induces a p53-dependent cell cycle arrest, thus preventing the proliferation of aneuploid cells and carcinogenesis
[52]. However it is not clear how, at which stage and by which signal (from extra chromosomes or extra centrosomes) p53 is activated by tetraploidy. Interestingly, cytokinesis failure activates the Hippo pathway: LATS2, which is an important kinase of Hippo pathway, translocates from centrosomes to the nucleus and stabilizes p53 in the presence of additional centrosomes
[53][54]. In addition, extra centrosomes can activate RAC1, which is known to antagonize RhoA, which leads to LATS2 activation
[55]. If p53 function is lost, cells can override the cell cycle block that is induced by tetraploidization, accumulate chromosome aberrations, and start to proliferate without control.
Cytokinesis can be blocked even when the furrow started to ingress: the abscission checkpoint delays cytokinesis completion and can even drive the regression of the cleavage furrow in the presence of chromatin bridges or lagging chromosomes
[56]. The abscission delay is dependent upon the localization of the Aurora B kinase at the midbody. Aurora B translocates from kinetochores to the division site after anaphase onset. At the midbody, activated Aurora B phosphorylates the ESCRT-III subunit Chmp4c, but the molecular mechanism that delays abscission is not completely understood.
Human cells are well equipped with several pathways that block tetraploid cells proliferation since cytokinesis failure can be tumorigenic. However tetraploidization is used by evolution and it can be important for certain tissues to gain new advantageous traits, such as resistance to drugs or stimuli (see also paragraph 6).
Interesting data suggest new connections between cytokinesis and DNA damage, in particular cytokinesis might be regulated in response to DNA damage. Damaged DNA should not be segregated, so, in this case, cytokinesis must be blocked to prevent the cleavage furrow from cutting damaged DNA. Proteins that are involved in DNA repair, such as BRCA2 and BCCIP, may be directly involved in cytokinesis, as their deficiency induces cytokinetic abnormalities
[57]. The polo kinase Plk1 is an important regulator of mitosis and cytokinesis and it is also a modulator of the DNA damage checkpoint
[58]. The budding yeast DNA damage checkpoint kinase Rad53 localizes at the division site and associates with septins
[59]. Other experiments suggest that DNA damage pathways may regulate cytokinesis proteins by modulating their the expression or their post-translational modifications, as p53 and Rb pathways inhibit the expression of cytokinetic proteins, such as Plk1, ECT2, anillin, and survivin
[60][61], and BRCT inhibits Aurora B kinase activity by Poly(ADP-ribosyl)ation in response to DNA strand breaks
[62]. It is reasonable that DNA damage, and perhaps incompletely replicated DNA, activates pathways that promote DNA repair, arrests cell cycle progression, and blocks cytokinesis completion.