|
Gene name - pavarotti
Synonyms - Cytological map position - 64A12 Function - cytoskeletal motor protein Keywords - mitosis, cytoskeletion |
Symbol - pav FlyBase ID: FBgn0011692 Genetic map position - 3-18--20 Classification - Kinesin motor domain, P-loop containing nucleotide triphosphate hydrolases Cellular location - nuclear |
Mutations in the Drosophila gene pavarotti result in the formation of abnormally large cells in the embryonic nervous system. In mitotic cycle 16, cells of pav mutant embryos undergo normal anaphase but then develop an abnormal telophase spindle and fail to undertake cytokinesis. The septin Peanut, actin, and the actin-associated protein Anillin do not become correctly localized in pav mutants. pav encodes a kinesin-like protein related to the mammalian MKLP-1. In cellularized embryos, the protein is localized to centrosomes early in mitosis, and to the midbody region of the spindle in late anaphase and telophase. Polo kinase associates with Pav, with which it shows an overlapping pattern of subcellular localization during the mitotic cycle and this distribution is disrupted in pav mutants. It is suggested that Pav is required both to establish the structure of the telophase spindle to provide a framework for the assembly of the contractile ring, and to mobilize mitotic regulator proteins (Adams, 1998).
Cytokinesis is the quintessential act of mitosis, whereby segregated daughter nuclei are partitioned into two separate cells. In eukaryotes, this is accomplished by an actin-myosin contractile ring that forms around the cell equator during mitosis and constricts inwards at telophase. In higher eukaryotes the correct positioning and assembly of the contractile ring requires the mitotic spindle, although the mechanism by which this occurs is a matter of controversy. Classical experiments suggest that the spindle poles are sufficient to stimulate cytokinesis. Two mechanisms have been put forward to demonstrate how spindle poles might induce cleavage furrows. In the first model, the poles signal directly to the equatorial cortex of the cell, perhaps by the action of astral microtubules. Alternatively, the astral relaxation model proposes that spindle poles induce a relaxation of the cell cortex nearest to the poles, leading to a tension differential between the poles and the equator that results in equatorial contraction. However, this model has been discredited by force measurements showing that tension increases at the equator without a concomitant decrease at the cell poles (Adams, 1998 and references therein).
There is growing evidence that the central spindle plays an essential role in the positioning and assembly of the contractile ring. The central spindle is composed of a dense network of overlapping antiparallel microtubules that forms between the separating daughter nuclei during anaphase. In cultured cells induced to develop multipolar spindles as result of treatment with low concentrations of the microtubule-destabilizing drug colcemid, the formation of a cleavage furrow absolutely depends on the presence of central spindle microtubules that are required throughout cytokinesis. Creation of a barrier between the central spindle and the cell cortex prevents cleavage, indicating that a signal for cytokinesis may emanate from the central spindle. Manipulation of grasshopper neuroblast spindles results in the formation of a cleavage furrow when the spindle is moved close to the cell cortex (Adams, 1998 and references therein).
In addition to the contribution made by the spindle, several proteins, known as the chromosomal passenger proteins, dissociate from chromosomes at the metaphase-anaphase transition to be deposited at the cell equator. The inner centromere proteins (INCENPs), for example, transfer to the central spindle and the cell cortex and are necessary for completion of cytokinesis, whereas TD-60 forms a disc structure in the position of the metaphase plate anticipating the cleavage plane. This disc is proposed to link the central spindle to the cell cortex (Adams, 1998 and references therein).
In Drosophila, mutational analysis has identified several genes encoding structural elements of the contractile ring essential for cytokinesis, such as myosin light chain, cofilin (Twinstar in Drosophila), profilin (Chickadee in Drosophila), and Peanut, a homolog of the yeast septins. Kinesin-like proteins (KLPs) are microtubule motor proteins responsible for many of the dynamic aspects of mitosis such as centrosome separation, spindle assembly, and anaphase movement of chromosomes (Adams, 1998).
pavarotti encodes a fly homolog of the mammalian mitotic kinesin-like protein-1 (MKLP-1). In pav mutants, anaphase B occurs normally but cytokinesis is defective, due to a disruption of the central spindle structure and subsequent failure to assemble a contractile ring. In mitosis, the Pav first associates with the centrosomes and then becomes localized at the spindle equator at late anaphase. Pav associates with Polo kinase and is required for its localization to the centrosome and central spindle. The Polo-like kinases are known to be required for centrosome function: Mutations in the Drosophila gene polo and its fission yeast homolog plo1 lead to spindle defects including the formation of monopolar structures. Injection of antibodies against human plk also result in the formation of monopolar structures in HeLa cells. An additional role for plks in cytokinesis is suggested by the late mitotic phenotype of fission yeast plo1 and its budding yeast counterpart cdc5. Multiple septation has been shown to result from plo1+ overexpression in fission yeast, or the expression of mammalian Plk in budding yeast. In light of the association of Polo kinase with PAV-KLP, the plks might also have a role in cytokinesis in metazoan cells that has been obscured by the earlier requirement for mitosis (Adams, 1998).
Pav mutant embryos show cytokinesis defects in virtually all dividing cells at cycle 16. As with other genes required for cell cycle progression in Drosophila, the lethal phase depends on the role and stability of the wild-type protein, the abundance of its maternal provision to the embryo, and the strength of the mutant allele. The majority of cell cycle proteins are supplied maternally, and the rate at which this supply is exhausted depends on the particular protein. The maternal supply of Pav appears to be depleted early in development, either by degradation or its expulsion from newly divided cells in the remnants of the midbody. Consequently, the lethal phase occurs in embryonic development leading to the defects in the PNS, embryonic tissues in which cells continue dividing beyond a sixteenth cell cycle. Lethality at this stage is a consequence of the inability of each mutant allele to produce functional gene product, since all alleles of pav show the same lethal phase, whether homozygous or hemizygous against a chromosome deficiency for the region (Adams, 1998).
As the cell progresses through its division cycle, the distribution of Pav follows a highly dynamic pattern that is very similar to that of MKLP-1 (Nislow, 1990). Both the mammalian and fly proteins show a punctate distribution in the region of the nucleus at interphase. They are present at the centrosomes from metaphase onwards and localize to the spindle midzone at late anaphase and telophase. Many proteins with roles in cytokinesis, including Pav and Peanut, become concentrated in midbodies, the remnants of which are expelled from the two daughter cells as they separate, and are seen as discrete spots of staining between interphase cells in cellularized embryos (Adams, 1998).
The pattern of distribution of Pav in syncytial embryos differs in some respects from that seen in individual cells, presumably reflecting the absence of cytokinesis. The protein still shows a pronounced association with the central region of the spindle at late telophase but then follows a pattern that suggests association with actin or actin-associated molecules. Although cytokinesis does not take place at this stage of development, the organization of the actin cytoskeleton is important in maintaining the distribution of and spacing between the rapidly increasing numbers of syncytial nuclei. Thus, an actin-containing structure analogous to the contractile ring forms at the position of the former metaphase plate, but in interphase this is redistributed to the cortex where it will form a 'cap' that sits between the nucleus and plasma membrane. Upon entry into mitosis the cortical actin is reorganized into the 'pseudocleavage furrows' that form a network capturing individual metaphase figures. In many respects these are analagous to cleavage furrows except that they are formed earlier in the mitotic cycle. Pav appears to concentrate in this network but is then liberated to associate with the spindle and move to the central region at late anaphase. It has been proposed that actin is continually binding to microtubules at this stage and moving toward their plus ends. One might speculate that Pav could provide the requisite motile force (Adams, 1998).
Mutations leading to cytokinesis defects have been described in genes encoding two other KLPs in Drosophila: KLP3A, and KLP38B. It is difficult to make comparisons between the mutant phenotypes, as in the case of KLP38B, a cytokinesis phenotype is inferred in larval neuroblasts that attain low levels of polyploidy, and from the observation of binucleate follicle cells in the ovary. KLP3A mutants also show a central spindle defect leading to a failure of cytokinesis, but this is seen in male meiosis. The differing lethal phases of the pav and KLP3A mutants suggest a specialized role for the two proteins for cytokinesis in different cell types, but an overlapping function of Pav in meiosis cannot be ruled out at the present time. Both Pav and KLP3A accumulate in the midbody at telophase, although Pav has the distinctive attribute of being associated with the centrosome (Adams, 1998 and references therein).
pav embryos initially show dramatic defects in the morphology of the central spindle at telophase. The simplest model for Pav function is that it participates in reorganizing the central spindle region after anaphase B has occurred, in a manner that is vital for assembly of the contractile ring. How the central spindle assists formation of the contractile ring is poorly understood. The septin Peanut, itself essential for cytokinesis in Drosophila, fails to localize to a ring-like structure in the pav mutants. This may be a direct consequence of the disrupted structure of the telophase spindle, since recent work with Xenopus shows that septins are microtubule-associated proteins, and that the injection of anti-septin antibodies inhibits contractile ring function. The septins may link the central spindle to the cell cortex, and consequently the delocalization of Peanut in pav mutants would disrupt the formation of the contractile ring. Anillin and actin also fail to associate with the equatorial region of the cell during anaphase and telophase, and consequently cells fail to divide and become tetraploid (Adams, 1998).
The role of the spindle poles in cytokinesis is less clear. Do they signal the onset of cytokinesis, or, rather, direct the positioning of cell cleavage? In Drosophila male meiosis, study of the male sterile mutation asterless reveals that in the absence of astral microtubules, cell cleavage occurs to completion but in the wrong place. A normal central spindle is seen in these mutants, indicating that in this system the central spindle may play some role in the recruitment of contractile ring components, whereas astral microtubules dictate the position of the cleavage furrow (Adams, 1998 and references therein).
The finding that pav is required for cytokinesis seems at odds with previous work suggesting that the mammalian homolog might function during either metaphase or anaphase. In vitro assays demonstrate an ability to cross-link and slide apart antiparallel microtubules, leading to the suggestion of a role in spindle pole separation during anaphase B, whereas the injection of anti-MKLP-1 antibodies into mitotic HeLa cells results in a metaphase arrest. The in vitro assay may not be sufficiently specific to dissect the exact role of MKLP-1, since the system lacks centrosomes and the ancillary mechanisms necessary to generate a contractile ring, both of which influence microtubule behavior during cytokinesis. Thus, the ability to cross-link and slide along microtubules in vitro may be a property required of the KLP when associated with contractile ring structures. In any event, the pav mutant phenotype shows no indication of a defect in anaphase B, and spindles extend to the same length as in wild-type embryos at late anaphase and telophase. Moreover, the rare binucleate cells that have lost pav function in the previous mitosis can perform anaphase perfectly well. Thus, if Pav does play a role in anaphase spindle elongation or spindle assembly, it would seem that this role is redundant and can be fulfilled by other KLPs. Although MKLP-1 is the most closely related klp to Pav, the sequence homology outside the conserved motor domain is not particularly high (20%), so there could be nonidentical functions. The idea that KLPs in the same subfamily can have distinct functions in different organisms is not without precedent. For example, the chromokinesin subfamily of DNA-binding KLPs involved in chromosome congression includes Drosophila KLP3A, which does not appear to bind DNA and is also required for cytokinesis. Nevertheless, because the size, domain structure, ability to associate with plks, and dynamic localization of pav-KP is so similar to MKLP-1, it would be surprising if it were not a functional homolog (Adams, 1998).
The presence of Pav and its mammalian counterparts at the centrosomes and the central region of the spindle raises the possibility that it may transport a signaling molecule required for either centrosome function, cytokinesis, or both. The serine-threonine protein kinase encoded by polo is an ideal candidate for such a molecule. polo mutants show abnormal centrosome behavior, including the formation of monopolar spindles, a phenotype that is also seen following the microinjection of anti-plk antibodies into mammalian cells. In addition to the formation of monopolar spindles, the additional consequence of disruption of the fission yeast homolog is to prevent cytokinesis (Ohkura, 1995). It is not clear whether the polo mutation results in cytokinesis defects in Drosophila because the mitotic cycle is blocked at an earlier stage. Nevertheless, the localization of polo kinase throughout the mitotic cycle would be consistent with a dual role for the enzyme similar to that demonstrated in fission yeast. The association demonstrated between Polo kinase and Pav leads to a speculation that in addition to a role in assembling the central region of the spindle, Pav may also be responsible for transporting Polo kinase from one set of centrosome-associated substrates to a second set of substrates in the midzone of the spindle as mitosis progresses (Adams, 1998).
The sequence of a 3.2-kb pav cDNA reveals an ORF of 2658 bases coding for an 886-amino-acid protein of 100 kD predicted molecular mass with homology to the KLP family of microtubule motor proteins. It contains a predicted microtubule motor domain of ~350 amino acids that includes a nuclear localization signal, an ATP binding site, and several motifs common to plus end-directed KLPs. A predicted coiled-coil region lies between residues 500 and 700, whereas residues 700-886 show homology only with the MKLP-1 subfamily of KLPs. Pav is most similar to this subfamily throughout its entire sequence, showing 34% identity with MKLP-1 as compared to 13% identity to other mitotic KLPs. The highest degree of identity (48%) lies in the motor and hinge domains (residues 1-450), although patches of identity extend throughout the entire carboxy-terminal region of the protein (residues 700-886), a segment of KLPs known to be highly diverged. The MKLP-1 subfamily of KLPs comprises human MKLP-1 and its Chinese hamster counterpart CHO-1, first identified as a mitotic spindle antigen (Adams, 1998).
date revised: 15 February 2002
Home page: The Interactive Fly © 1995, 1996 Thomas B. Brody, Ph.D.
The Interactive Fly resides on the
Society for Developmental Biology's Web server.