Localization of Mei-S332 requires Centromere Identifier

Studies show that centromeres and flanking heterochromatin are physically and functionally separable protein domains that are required for different inheritance functions, and that Cid is required for normal kinetochore formation and function, as well as cell-cycle progression. Injection of Cid antibodies into early embryos, as well as RNA interference in tissue-culture cells, shows that Cid is required for several mitotic processes. Cid chromatin is physically separate from proteins involved in sister cohesion (MEI-S332), centric condensation (Prod, kinetochore function (Rough deal, Zeste-white 10 and Bub1) and heterochromatin structure (HP1). Cid localization is unaffected by mutations in mei-S332, Su(var)2-5 (HP1), prod or polo. Furthermore, the localization of Polo, the kinesin kinetocore motor CENP-meta, Rough deal (Rod), Bub1 and Mei-S332, involved in sister chromatid cohesion) depends on the presence of functional Cid. It is concluded that Cid directs kinetochore formation and function by forming a unique heterochromatin within the centromere (Blower, 2001).

Centromeres in most higher eukaryotes are embedded in centric heterochromatin, suggesting that both the structure and function of heterochromatin are required for centromere function. What are the structural relationships between centromeric chromatin, defined by Cid, and chromosomal proteins previously localized to the centromere region? This question was addressed using immunolocalization of three proteins and Cid on mitotic chromosomes from S2 and Kc tissue-culture cells (Blower, 2001).

Mei-S332 is required for sister chromatid cohesion during metaphase I of meiosis, and is present in the centromeric regions of meiotic and mitotic chromosomes. Simultaneous localization of Cid with Mei-S332 shows that Cid antibodies yield typical double-dot staining, whereas Mei-S332 is localized in two concentrated foci joined by a bridge of staining that connects the sister chromatids. Although Mei-S332 has been described as centromeric and possibly located to the inner kinetochore, the higher resolution localization of Mei-S332 presented in this study showed consistent offset of antibody staining to one side of the kinetochore and along the chromosome axis on all chromosomes. The offset localization is always to the same side of the kinetochore on each chromosome type. This is especially evident on the X chromosome, in which Mei-S332 is always located on the proximal long arm side of Cid, and for chromosomes 2 and 3, on the basis of colocalization with the sequence-specific satellite binding protein Prod. It is concluded that Mei-S332 is located near but not in the Cid chromatin, providing a physical basis for the previous observation that kinetochore function and Mei-S332-mediated cohesion can be separated using minichromosome derivatives (Blower, 2001).

proliferation disrupter (prod) mutant larval neuroblasts display hypo-condensation of the centromere region and metaphase/anaphase arrest. Consistent with the decondensation phenotype, the Prod protein localizes to the centromeric region of chromosomes 2 and 3 in mitosis, suggesting that it may be involved in kinetochore function on these chromosomes. However, simultaneous detection of Cid and Prod on mitotic chromosomes shows that Prod stains a more expansive portion of the chromosome than Cid, and is offset from the kinetochore in the same manner as Mei-S332. In fact, Prod and Mei-S332 are both localized to the same side of the kinetochore on chromosomes 2 and 3 (Blower, 2001).

HP1 mutants show dominant suppression of heterochromatin-induced position-effect variegation (PEV), and recessive telomere fusions and chromosome segregation defects. Human and mouse homologs of HP1 localize to the centromere region, and S. pombe Swi6, another chromodomain protein, is localized to the centromere and required for proper chromosome transmission. Simultaneous localization of Cid with HP1 revealed that HP1 is not present in centromeric chromatin in either interphase or metaphase. In metaphase chromosomes, HP1 is located throughout the pericentric heterochromatin, and is near but not in Cid chromatin (Blower, 2001).

It is concluded that Prod and HP1 are located in the pericentric heterochromatin and not in the centromeric chromatin. These results suggest that, although the centromere is embedded in large blocks of heterochromatin, centromeric chromatin is spatially separable from canonical centric heterochromatin (Blower, 2001).

Does the spatial separation of Cid chromatin, outer kinetochore proteins and centric heterochromatin proteins reflect functional independence? Cid localization was examined in larval neuroblasts from animals lacking either Prod, HP1, Mei-S332 or Polo kinase. In interphase nuclei and mitotic chromosomes from homozygous prod mutants, Cid remains localized in the typical punctate pattern observed in wild type, despite visible centromere region hypocondensation. Similarly, Cid was localized in the typical punctate pattern in interphase nuclei from homozygous mutant Su(var)2-5 (HP1) neuroblasts (Blower, 2001).

In mutant metaphase spreads exhibiting the Su(var)2-5 telomere fusion phenotype Cid still localizes in the characteristic double-dot pattern. Furthermore, Cid is also localized in the characteristic double dot pattern in homozygous mei-S332 mutant larval neuroblasts. Finally, in metaphases exhibiting circular spreads indicative of centrosome disorganization, characteristic of polo mutations, Cid remains localized in characteristic double dots. Thus, the analyses of Cid localization in mutant neuroblasts show that the assembly and maintenance of centromeric chromatin in interphase and metaphase is not dependent on the presence of proteins required for normal centromere region condensation (Prod), heterochromatin structure (HP1), centric cohesion (Mei-S332), or outer kinetochore function (Polo kinase) (Blower, 2001).

Although Cid localization is not dependent on the presence of Prod, HP1, Mei-S332 or Polo kinase, the mutant analyses did not determine whether the localization of these proteins depended on Cid. Therefore, Polo kinase, Mei-S332 and Prod localization were examined in embryos in which Cid function was inhibited. In embryonic nuclei close to the site of injection, where high levels of Cid antibody binding and the most severe mitotic defects are observed, Polo kinase localization is diffuse and apparently absent from kinetochores, as judged by counterstaining with Prod. Notably, in these same nuclei Mei-S332 was absent from the pericentromeric region, whereas Prod, a protein with sequence-specific satellite-binding properties, was still present in the pericentromeric region (Blower, 2001).

The localizations of Rod, CENP-meta outer kinetochore, CENP-E kinesin-like protein homolog, Polo kinase, Bub1 and Mei-S332, but not Prod or HP1, were also disrupted in Kc cells displaying mitotic defects as a result of RNAi inhibition of Cid expression. Quantitative deconvolution microscopy has revealed that transient kinetochore component recruitment is proportional to the amount of Cid present at the kinetochore, whereas Prod recruitment is independent of Cid levels. Thus, Cid function is required for the recruitment or maintenance of transient kinetochore components and a centric cohesion protein (Mei-S332), but is not required for the localization of Prod or HP1. These results also indicate that the pleiotropic mitotic defects observed in anti-Cid injection and RNAi are likely to be caused by a failure to recruit or bind transient kinetochore components and a centric cohesion protein (Blower, 2001).

The mislocalization of Rod, Bub1, CENP-meta, Polo and Mei-S332 in nuclei displaying missegregation phenotypes shows that the defects are correlated with aberrant kinetochore structure and the recruitment of transient kinetochore proteins and other centromere region proteins. These results extend the earlier observation that the inner kinetochore protein CENP-C is mislocalized in the CENP-A knockout mouse to the location of outer kinetochore proteins. Notably, the amount of outer kinetochore components present at the kinetochore is proportional to the amount of Cid, suggesting that the kinetochore may be composed of a repeated substructure (Blower, 2001).

Mei-S332 is required for proper chromosome inheritance in Drosophila, but surprisingly the chromosome still retains kinetochore protein localization and functions if Mei-S332 is eliminated. Despite the spatial and functional separation of Mei-S332 and Cid, Mei-S332 is dependent on functional centromeric chromatin for its recruitment to the centromere region: it is mislocalized in anti-Cid injected embryos and RNAi-treated Kc cells. Cid localization is not, however, dependent on the presence of Mei-S332. Therefore, Cid is epistatic to Mei-S332 in the pathway responsible for the assembly and/or maintenance of this physically and functionally distinct centromere region domain in mitosis. The relationship between Cid, kinetochore function and Mei-S332-mediated cohesion warrants further genetic and biochemical analyses. It will be particularly interesting to determine the significance of the consistent asymmetric positioning of Mei-S332 to only one side of the Cid chromatin, as well as its impact on Cid during meiosis, where mutant phenotypes are more severe (Blower, 2001).

Control of centromere localization of the Mei-S332 cohesion protection protein

In mitosis and meiosis, cohesion is maintained at the centromere until sister-chromatid separation. Drosophila Mei-S332 is essential for centromeric cohesion in meiosis and contributes to, though is not absolutely required for, cohesion in mitosis. It localizes specifically to centromeres in prometaphase and delocalizes at the metaphase-anaphase transition. In mei-S332 mutants, centromeric sister-chromatid cohesion is lost at anaphase I, giving meiosis II missegregation. Mei-S332 is the founding member of a family of proteins important for chromosome segregation. One likely activity of these proteins is to protect the cohesin subunit Rec8 from cleavage at the metaphase I-anaphase I transition. Although the family members do not show high sequence identity, there are two short stretches of homology, and mutations in conserved residues affect protein function. This study analyzes the cis- and trans-acting factors required for Mei-S332 localization. A striking correlation is found between domains necessary for Mei-S332 centromere localization and conserved regions within the protein family. Drosophila Mei-S332 expressed in human cells localizes to mitotic centromeres, further highlighting this functional conservation. Mei-S332 can localize independently of cohesin, assembling even onto unreplicated chromatids. However, the separase pathway that regulates cohesin dissociation is needed for Mei-S332 delocalization at anaphase (Lee, 2004).

The Mei-S332 family members share short conserved regions at the N and C termini, with a highly polar region in the middle of the protein. Given the limited sequence conservation among family members, the protein domains essential for centromere localization were delineated. Mei-S332 localization and dissociation from mitotic chromosomes coincides with times of sister-chromatid cohesion presence and release. A rapid assay was devised for Mei-S332 localization in mitotic Drosophila cultured cells. By fusing mei-S332 to an N-terminal gfp tag under a constitutive armadillo promoter, various deletion constructs of Mei-S332 were expressed in Drosophila S2 cells by transient transfection. The GFP-expressing cells were recovered by flow cytometry, stained with DAPI, and scored for Mei-S332 localization. A Mei-S332-GFP fusion protein with GFP at the same N-terminal site has been shown to be fully functional in vivo (Lee, 2004).

As controls for the accuracy of Mei-S332 localization in this system, full-length Mei-S332 fused to GFP was expressed. This fusion protein localizes to centromeres in 93% of transfected S2 metaphase cells, as confirmed by colocalization with anti-CID, a centromere-specific histone. As an additional control, a deletion was generated that truncates the same part of the C terminus that is missing in the Mei-S332 protein made in the mei-S3327 mutant. This allele (deletion 'a') abolishes localization on mitotic and meiotic chromosomes in vivo. In transfected S2 cells, the deletion 'a' protein failed to localize specifically onto chromosomes. Therefore, the S2 cell system faithfully reproduces aspects of in vivo Mei-S332 localization (Lee, 2004).

A series of deletions spanning the Mei-S332 protein were generated to define domains needed for chromosomal localization; the conserved N and C termini of the protein were found to be essential. The localization pattern of the deletion 'a' construct, which mimics the protein structure of Mei-S3327, was consistent with a role for the C terminus of the protein in chromosomal localization, as has been demonstrated in vivo. Point mutations located in the last 26 amino acids of the C terminus, such as alleles 2, 6, and 10, disrupt Mei-S332 localization in mitosis and meiosis. A large deletion of the N terminus (deletion 'b') demonstrated that a region at the N terminus also was essential for localization. To delineate further the domain necessary for localization, the region was split into two smaller deletions (deletions c and d). The essential region was found to included the conserved coiled-coil. The requirement for this domain was assessed by specifically deleting the coiled-coil region (deletion e) and it was found that this protein did not localize to chromosomes (Lee, 2004),

Focus was next placed on the central region of the protein, a region whose sequence is not conserved within the family, although in all family members it is marked by a bias toward charged residues. Almost all of the orthologs also show high probabilities of PEST sequences between the conserved N and C termini. The function of these regions has not been delineated. Allele mei-S3325, which changes Ser-277 to Phe, has only slight effects on meiotic chromosome segregation and does not perturb protein localization, hinting that this region might not be crucial for protein function. This region was examined with a number of deletions. Deletions h, i, and j together cover much of the middle part of the protein. Deletion h is missing both of the predicted PEST sequences, the first of which includes an acidic region in which 14 out of 26 residues are aspartate or glutamate. Deletion i removes most of the second predicted PEST sequence and the amino acid mutated in allele 5, and deletion j uncovers the region from the position of the allele 7 truncation to a site 40 amino acids from the C terminus of the protein. All three of these fusion proteins localized to metaphase chromosomes and compellingly confirmed the observation that mutation of the middle of the protein, such as allele 5, does not abolish localization. To prove further that the middle of the protein is dispensable for localization, deletion k was generated, that removes 161 amino acids from Mei-S332. Even a deletion of this size still allows the protein to localize. It is concluded that this central area is not necessary for localization (Lee, 2004).

It was noticed, however, that in some cases, the GFP signal seems to be spread further along the chromosomes rather than residing solely at the centromere. This might be due to increased protein expression, because the GFP signal in the cells was visibly brighter and there are high levels of the Mei-S332-GFP protein. To test this hypothesis, the 25% of cells with the highest levels of fluorescence were separated from cells that fell in the middle 50% of fluorescence levels and the localization of deletion proteins h and i was examined. It was found that all of the examples of GFP signal spreading were found in the high fluorescence population. This was also true for very rare cases of GFP spreading seen with the full-length construct. These results raise the possibility that the PEST domains contribute to instability and reduce protein levels. These findings also indicate that Mei-S332 does not have an intrinsic property of binding only to the centromere, and with increased protein levels can localize to the arms. The CENP-A-like protein CID is necessary for Mei-S332 localization. Perhaps when Mei-S332 is overexpressed, it saturates its centromeric anchors, possibly including CID, and thus can localize over a larger chromosomal region (Lee, 2004).

Whereas the conclusion that the C terminus of Mei-S332 is essential for centromere localization is consistent with the in vivo meiotic consequences of mutations in this domain, the results of deleting the N terminus provide new insights into its function. These experiments delineate the coiled-coil region as being crucial, yet it has been observed that the amino acid changes in alleles 3 and 8 predicted to disrupt coiled-coil structure do not impede localization in mitosis or meiosis in vivo. The amino acid mutated in allele 8 appears to be conserved and is usually a hydrophobic residue in the Mei-S332 family members [6 and 8]. There are several possible explanations for the discrepancy. The coiled-coil may be essential for localization, but perhaps these single point mutations, which have only moderate nondisjunction effects, are not disruptive enough to abolish chromosomal localization of Mei-S332. A stronger allele in the N terminus, mei-S3329, could not be tested for effects on localization in vivo because levels of mutant protein are low. Alternatively, a coiled-coil structure may not be necessary but the domain itself could be critical for localization (Lee, 2004).

Another deletion suggested a role for the interval immediately C-terminal to the coiled-coil region. However, when this region was divided into two smaller regions (deletions d and f), both regions were not essential for localization. Western blotting showed that deletion g did not destabilize the protein. It is possible that the areas removed by deletions d and f are redundant. A small percentage of deletion g metaphases do show some GFP foci on the chromosomes. Deletion g could also be disrupting protein structure by destabilizing the coiled-coil conformation and thereby indirectly impairing protein localization. Given the lack of sequence conservation in the region of deletion g, the latter explanation is favored (Lee, 2004).

These results highlight the role of conserved domains in the Mei-S332 protein family in centromere localization of the protein. In addition, it was learned that the middle of the protein is not involved in localization. While the central region of the protein is not needed for localization, it may be necessary for the function of the protein in protecting cohesion (Lee, 2004).

Given that mei-S332 is part of a conserved protein family, whether Drosophila Mei-S332 could be localized onto human prometaphase chromosomes was tested. mei-S332 fused to an N-terminal gfp under a constitutive promoter expressed in transiently transfected human 293T cells. Specific localization of Mei-S332 was seen on chromosomes in cells that were in mitosis. The foci of GFP colocalized with foci of anti-Bub3, a protein found at the kinetochore, showing that Mei-S332-GFP localizes to centromeres. Thus, the protein machinery that facilitates the localization of Mei-S332 to mitotic centromeres is conserved (Lee, 2004).

In S. cerevisiae and S. pombe, the Mei-S332 counterpart Sgo1 appears to protect centromeric cohesin from being cleaved in meiosis I, thereby preventing the release of sister-chromatid cohesion until meiosis II. The Rec8 subunit of meiotic cohesin was found to localize normally in sgo1 mutants in prophase I, but it has not yet been examined if Mei-S332 family members require cohesin for localization. The timing of Mei-S332 and Sgo1 delocalization relative to cohesin cleavage is important in evaluating whether these proteins protect cohesin merely by their presence at the centromere, or whether they can retain localization without actively protecting cohesin. A first step is to determine whether Mei-S332 can localize in the absence of cohesin. Because meiotic cohesin complexes have not yet been characterized in Drosophila and there are no known mutations in cohesin complex proteins, the S2 cell system was used to address this question by depleting the cohesin complex subunit Rad21 (also known as Scc1 or Mcd1) and examining endogenous Mei-S332 localization (Lee, 2004).

RNA interference has been shown to effectively deplete DRAD21. Using 600 bp dsRNA fragments to drad21, it was possible to deplete much of the protein as assayed by Western blotting. DRAD21 immunofluorescence also showed that DRAD21 levels are much lower. In DRAD21-depleted cells, chromosomes appeared disorganized and failed to congress to proper metaphase plates, showing that sister-chromatid cohesion indeed had been disrupted (Lee, 2004).

To determine if Mei-S332 depends on cohesin for localization, drad21 RNAi cells were stained with an anti-Mei-S332 antibody. Strikingly, Mei-S332 localized to chromosomes. In all metaphase figures at 66 hr and 90 hr after dsRNA addition, Mei-S332 localization on chromosomes was observed. Cells whose spindle and chromosome organization was sufficiently disrupted such that they could not be staged, indicative of inactivation of cohesin, definitively showed Mei-S332 on chromosomes. Because of the metaphase delay and chromosome disorganization in these cells, it was difficult to identify anaphases unambiguously, but in a small number of clear anaphases, Mei-S332 was not associated with chromosomes. Mei-S332 was eventually able to dissociate from chromosomes in those cells that managed to get through the cell cycle, because telophases did not show any specific Mei-S332 localization. These results demonstrate that the localization of Mei-S332 is independent of cohesin. It also appears that Mei-S332 is able to delocalize at anaphase from chromosomes in the absence of cohesin (Lee, 2004).

These data indicate that cohesin and Mei-S332 assembly onto chromosomes are controlled independently. The localization of Mei-S332 to metaphase chromosomes in DRAD21-depleted cells also demonstrates that Mei-S332 remains localized and does not simply fall off the chromosomes in the absence of cohesin. In addition, these findings are consistent with the observation that Mei-S332 localizes to metaphase chromosomes in Drosophila eco mutants. Drosophila Eco is a homolog of yeast Eco1/Eso1, a protein that is needed to properly establish cohesion between sister chromatids (Lee, 2004).

To test whether Mei-S332 can localize in the complete absence of sister-chromatid cohesion, Mei-S332 was examined in a double parked mutant. The Double Parked/Cdt1 proteins are essential factors in replication initiation. In double parked (dup) mutants, S phase in cycle 16 of Drosophila embryogenesis is blocked but chromosomes are able to condense and enter mitosis. Despite the absence of a replicated sister chromatid, Mei-S332 localizes onto the chromosomes. Although Mei-S332 is capable of chromosomal association, its localization was not entirely normal in that the Mei-S332 signal appeared spread, rather than in tight foci. It has been shown that on the single sister chromatids of dup mutants, a functional kinetochore is assembled, capable of merotelic microtubule attachments, with a normal localization pattern of Inner Centromere Protein (INCENP), the centromere histone CID, and the BUB1 spindle checkpoint component. Thus, Mei-S332 localization to chromosomes is completely independent of sister-chromatid cohesion, and cohesin is not required as an anchor (Lee, 2004).

It was asked whether the regulatory mechanisms for releasing sister-chromatid cohesion could affect Mei-S332. Mei-S332 dissociates from centromeres at the metaphase-anaphase transition, although these cytological studies do not have the resolution to determine whether this slightly precedes or is coincident with cohesin release. Some of the components for regulating cohesin release have been characterized molecularly. The separase protease has been shown to cleave the RAD21 subunit of cohesin at the metaphase-anaphase transition in yeast and metazoan cells, thereby allowing sister chromatids to separate. In Drosophila, separase activity is encoded by two proteins, Three rows (Thr) and Separase (Sse). Thr mutants are embryonic lethal, since the chromosomes are unable to separate in the 15th mitosis of embryogenesis. Centromeric attachments remain such that chromosomes rereplicate and produce diplochromosomes in the 16th cycle (Lee, 2004).

Mei-S332 localization was examined in embryos homozygous mutant for three rows. Mutant cells in which cohesin release failed are recognizable by the distinctive diplochromosomes. It was found that Mei-S332 localized to mitotic chromosomes as normally seen in wild-type embryos, but also Mei-S332 signal was seen in the nuclei of interphase cells. This dispersed localization in interphase nuclei was seen reproducibly. Such a pattern has never been observed before and suggests that Mei-S332 does not dissociate from chromosomes in anaphase and consequently persists into the next interphase. Although it seems likely that Mei-S332 remains on chromosomes in interphase because the signal is not diffuse throughout the nucleus, the possibility that Mei-S332 is nuclear but not bound to chromosomes cannot be excluded (Lee, 2004).

As an additional test of the possibility that the separase pathway can affect Mei-S332, embryos mutant for pimples (pim) were examined. Pimples corresponds to the securin molecule that inhibits separase until the proper time for sister-chromatid separation in many organisms. It has been demonstrated, however, that functional Pimples is needed also for active separase activity in Drosophila, because mutants in pimples display a failure to release sister-chromatid cohesion and lead to diplochromosomes. As in thr mutants, Mei-S332 localized to chromosomes in mitosis and showed nuclear staining in interphase in pim mutants, further providing evidence that Mei-S332 delocalization is dependent upon the separase pathway (Lee, 2004).

It is as yet unclear whether the requirement for separase activity to delocalize Mei-S332 is direct or indirect. There are two sites in Mei-S332 that fit the loose consensus of a separase cleavage site (EXXR). By Western blotting with polyclonal anti-Mei-S332 antibodies, no evidence of cleavage fragments has been seen in cell cycle-staged embryos. However, such an analysis might not be sensitive enough to reveal cleavage products if only a small pool of Mei-S332 is being cleaved. These putative separase cleavage sites are removed in deletions h and k that do not result in persistence of the protein into interphase. Thus, separase activity may be a prerequisite for delocalization of Mei-S332 in ways that do not involve direct cleavage of the protein. While it is possible that Mei-S332 delocalization requires cohesin release, this interpretation is not favored, given that Mei-S332 localizes independently of cohesin (Lee, 2004).

The persistent localization of Mei-S332 in separase mutants has the significant implication that anaphase-promoting complex/cyclosome (APC/C) activity is not sufficient to delocalize Mei-S332. The APC/C is required to degrade many target proteins, including securin (Pimples) and cyclin B, in order to allow chromatid separation. In studies of S. pombe Sgo1, the depletion of APC/C activity led to the persistence of elevated Sgo1 protein levels in meiosis. Because APC/C activity is unaffected in thr and pim mutants, as assayed by the normal degradation of cyclins A and B, yet Mei-S332 remains in interphase nuclei, APC/C activity is unlikely to function in delocalization. This does not exclude a role for the APC/C in degrading delocalized pools of Mei-S332 protein (Lee, 2004).

It is concluded that the conserved terminal domains of Mei-S332 are essential for centromere localization, but the Mei-S332 protein can localize onto chromosomes independently of sister-chromatid cohesion. Mei-S332 localizes onto the centromeres of chromosomes lacking the cohesin complex, and it can assemble even onto unreplicated sister chromatids. Thus, localization of Mei-S332 is not synonymous with the presence of cohesion, suggesting that the activity of the protein in protecting cohesion may be regulated such that localized Mei-S332 can be either active or inactive. The retention of Mei-S332 protein in separase mutants reveals a prerequisite for separase activity, directly or indirectly, in Mei-S332 delocalization. This further indicates that APC/C activity, independent of its role in activating the separase pathway, is not sufficient to delocalize the protein (Lee, 2004).

Regulation of centromere localization of the Drosophila Shugoshin MEI-S332 and sister-chromatid cohesion in meiosis

The Shugoshin (Sgo) protein family helps to ensure proper chromosome segregation by protecting cohesion at the centromere by preventing cleavage of the cohesin complex Some Sgo proteins also influence other aspects of kinetochore-microtubule attachments. Although many Sgo members require Aurora B kinase to localize to the centromere, factors controlling delocalization are poorly understood and diverse. Moreover, it is not clear how Sgo function is inactivated and whether this is distinct from delocalization. This study investigated these questions in Drosophila melanogaster, an organism with superb chromosome cytology to monitor Sgo localization and quantitative assays to test its function in sister-chromatid segregation in meiosis. Previous research showed that in mitosis in cell culture, phosphorylation of the Drosophila Sgo, MEI-S332, by Aurora B promotes centromere localization, whereas Polo phosphorylation promotes delocalization. These studies also suggested that MEI-S332 can be inactivated independently of delocalization, a conclusion supported in this study by localization and function studies in meiosis. Phospho-resistant and phospho-mimetic mutants for the Aurora B and Polo phosphorylation sites were examined for effects on MEI-S332 localization and chromosome segregation in meiosis. Strikingly, MEI-S332 with a phospho-mimetic mutation in the AuroraB phosphorylation site prematurely dissociates from the centromeres in meiosis I. Despite the absence of MEI-S332 on meiosis II centromeres in male meiosis, sister chromatids segregate normally, demonstrating that detectable levels of this Sgo are not essential for chromosome congression, kinetochore biorientation, or spindle assembly (Nogueira, 2014).

Protein Interactions

Phosphorylation of Mei-S332 by Polo kinase

Because Polo kinase may regulate the function of Mei-S332 directly through phosphorylation, it was first asked if Mei-S332 was phosphorylated in embryo extracts. Embryo extracts were utilized because large quantities of protein from cells in mitosis could be easily isolated and analyzed biochemically. Because Mei-S332 resolves into a doublet of bands by SDS-PAGE analysis (LeBlanc, 1999), it was determined whether this doublet was due to phosphorylation. Wild-type embryo extracts were treated with lambda protein phosphatase and the migration pattern of Mei-S332 was analyzed on Western blots. Mei-S332 shifted to the slower migrating form upon phosphatase treatment, and this conversion was blocked by incubation with sodium vanadate and sodium phosphate, lambda phosphatase inhibitors. By contrast, sodium fluoride, a weak inhibitor of lambda phosphatase, failed to block this shift. These results indicate that Mei-S332 is a phosphoprotein and that the faster migrating form is the phosphorylated form. Although it is more common for phosphoproteins to have a reduced electrophoretic mobility, there are precedents for phosphorylated protein forms migrating faster. From this analysis, it cannot be determined if there are multiple phosphorylation events responsible for the faster migrating form of Mei-S332, especially given that there are at least 53 possible phosphorylation sites in Mei-S332 (Clarke, 2005).

Whether Mei-S332 centromere localization correlates with phosphorylation state was examined by preparing extracts from embryos in specific stages of mitosis. Early embryos undergoing rapid S/M cycles were isolated, fixed, stained with DAPI to visualize the chromosomes, and micromanipulated to gather embryos with nuclei in interphase, metaphase, and anaphase, and then extracts were prepared from the staged embryos. This analysis revealed that Mei-S332 was in the apparent phosphorylated state in interphase, the dephosphorylated state in metaphase, and the phosphorylated state in anaphase. Therefore, Mei-S332 appears to be in the dephosphorylated state when it is localized to centromeres and phosphorylated when it is not at centromeres. This correlation suggests that Mei-S332 could be phosphorylated at the metaphase/anaphase transition. This analysis cannot distinguish whether the anaphase and interphase phosphorylated forms of Mei-S332 are the same. Indeed, it seems unlikely that the identical phosphorylation state of Mei-S332 persists from anaphase until interphase of the next cell cycle (Clarke, 2005).

To determine if Mei-S332 must be localized to centromeres in order for phosphorylation to occur, mei-S332⁶ mutant embryos were used in which the mutant protein fails to localize to centromeres during mitosis (Tang, 1998). Homozygous mei-S332⁶ early embryos were micromanipulated according to cell cycle stage and compared to wild-type extracts. Phosphorylation of Mei-S332⁶ protein was still detectable as in wild-type embryos, indicating that centromeric localization is not required for the phosphorylation of Mei-S332 (Clarke, 2005).

To show directly that Mei-S332 can be phosphorylated, GFP- and myc-tagged forms of the protein, previously shown to be functional in vivo, were immunoprecipitated. Radiolabeled phosphate was added to the immunoprecipitates, and whether Mei-S332 could be phosphorylated was examined. In immunoprecipitates, a ³²P-labeled protein band was observed and confirmed to be the Mei-S332-GFP or myc fusion protein by Western blotting. These results suggest that a kinase that associates with Mei-S332 is capable of phosphorylating Mei-S332 in vitro (Clarke, 2005).

The observations that Polo antagonizes Mei-S332 and is needed for its delocalization from the centromere coupled with the finding that Mei-S332 becomes phosphorylated in anaphase suggested that Polo phosphorylates Mei-S332 at the metaphase/anaphase transition. Therefore, tests were performed to see whether Mei-S332 phosphorylation was affected in the polo mutants. Protein extracts from polo⁹ and polo¹⁰ homozygous and heterozygous third instar larval brains were analyzed for the Mei-S332 phosphorylation state by Western blotting. These hypomorphic mutant alleles are lethal at this stage, when no maternal protein pools persist. In homozygous mutants, a high percentage of the cells in brains are arrested in metaphase. In both the heterozygous and homozygous polo⁹ mutants, Mei-S332 was predominately in the dephosphorylated, slower migrating form of the protein. The levels of phosphorylated MEI-S322 were markedly reduced in polo⁹ homozygous mutant neuroblasts, and levels were reduced to a lesser extent in the polo⁹ heterozygote. It is a possibility that Mei-S332 is mostly dephosphorylated because a large percentage of cells are in metaphase in polo⁹ mutants. In the weaker polo¹⁰ heterozygous and homozygous mutants, Mei-S332 was predominately in the phosphorylated form. These results are consistent with Polo kinase affecting Mei-S332 phosphorylation in vivo, the possibility that electrophoretic variants of Mei-S332 are present in the polo mutant strains cannot be excluded (Clarke, 2005).

An in vitro Xenopus egg extract system was used to test whether Polo could phosphorylate Mei-S332 directly. In vitro transcribed and translated (IVT) Mei-S332 protein undergoes multiple mobility shifts when incubated with mitotic extracts as compared to interphase extracts. These mitotic extracts have active APC/C and thus are in a state equivalent to anaphase, when Mei-S332 is expected to be phosphorylated. These events were due to the phosphorylation of Mei-S332, because treatment with lambda protein phosphatase abolished all mobility shifts. Phosphorylation in vitro shifts Mei-S332 to a slower migrating form, whereas in vivo Mei-S332 is shifted to a faster migrating form. Despite this difference, these results indicate that Mei-S332 is phosphorylated in a cell cycle-dependent manner in vitro, similar to what is observed in Drosophila embryos (Clarke, 2005).

Advantage of these in vitro observations was taken to determine if the Xenopus Polo kinase, Plx1, was responsible for Mei-S332 phosphorylation in the extracts. The Xenopus interphase and anaphase extracts were immunodepleted with antibodies against Plx1 or a control antibody and then Mei-S332 IVT protein was incubated in these depleted extracts. Importantly, depleting for Plx1 in Xenopus extracts does not affect Cdc2 or Aurora B kinase activities. All of the mitotic Mei-S332 phosphorylation mobility shifts were present in the control-depleted sample, but the slowest migrating form of phosphorylated Mei-S332 was no longer present in the Plx1-depleted sample. The immunodepleted extracts were probed by Western blot for Plx1, confirming the extent of the depletion. These results demonstrate that Xenopus Plx1 kinase is required for one phosphorylated form of Drosophila Mei-S332 in an anaphase state in vitro. The other in vitro mitotic Mei-S332 phosphorylation events likely result from one or more kinases other than Plx1. It remains to be seen whether these additional phosphorylation events are relevant in vivo (Clarke, 2005).

Plx1-dependent phosphorylation of Mei-S332 results in a slower migrating form of the protein, whereas in vivo the Polo-dependent phosphorylation of Mei-S332 results in a faster migrating form. Because it appears that other kinases are phosphorylating Mei-S332 in vitro, these modifications may alter the mobility of Mei-S332 IVT protein as compared to the in vivo form. Additionally, other posttranslational modifications, such as ubiquitination, may be differentially present on Mei-S332 between the in vitro and in vivo forms, thereby affecting the mobility of the protein. These possibilities are supported by the fact that the unphosphorylated Mei-S332 IVT protein has a significantly slower mobility than both protein forms in vivo. The in vitro Xenopus extract system demonstrates that a vertebrate Polo kinase is required for phosphorylation of Drosophila Mei-S332 in an anaphase state (Clarke, 2005).

A consensus motif of Ser-pSer/pThr-Pro/X defines a Polo box domain (PBD) binding site on Polo substrates. The PBD is conserved in Polo-like kinases from many organisms, and it is proposed to be critical for proper binding of Polo to the substrates and release of inhibition of the Polo kinase domain. In substrates, the central serine/threonine residue of the PBD binding site must be phosphorylated to ensure proper binding of Polo. This motif was found in two sites in Mei-S332: SSP with the central serine at residue 234 and STP with the central threonine at residue 331. To test the relevance of these two PBD binding motifs in Mei-S332, the central residue in each motif was mutated to an alanine to make single and double mutants (Clarke, 2005).

To determine if Plx1-dependent phosphorylation of Mei-S332 was likely to be the result of a direct interaction between Polo and Mei-S332, these PBD binding site mutant constructs were transcribed and translated in vitro, and resulting proteins were incubated with control- and Plx1-immunodepleted Xenopus mitotic extracts. Mitotic shifts of all three mutant proteins were demonstrated to be due to phosphorylation by treatment with lambda protein phosphatase. The Plx1-dependent phosphorylation was unaffected by the S234A mutation but was abolished with the T331A and double mutations. Thus, at least the T331 residue and possibly the combination of the T331 and S234 are required for Plx1-dependent phosphorylation to occur in mitotic extracts (Clarke, 2005).

To test more directly if Mei-S332 could be a substrate of Polo kinase, it was asked whether the Drosophila Polo PBD and the Xenopus Plx1 PBD interact with Mei-S332. GST pull-down experiments were carried out using purified GST, GST-Polo PBD (residues 298-576), and GST-Plx1 PBD proteins expressed in bacteria and lysates from transfected S2 cells. N-terminal GFP-tagged Mei-S332 was set under the control of the constitutive armadillo promoter and transiently expressed in Drosophila S2 cells. GFP-Mei-S332 showed strong binding to both the GST-Polo and GST-Plx1 PBD compared to GST alone (Clarke, 2005).

It was next asked if the binding of the PBD to Mei-S332 was affected in T331A, S234A, and double mutant-transfected S2 lysates. The double mutant protein was decreased in binding to Polo PBD by 50% and to Plx1 PBD by 24%. The S234A and T331A mutant proteins also showed a decrease in binding to Polo and Plx1 PBDs, but to a lesser extent. Together, these results suggest that Mei-S332 is a direct substrate of Polo kinase in vivo and that the interaction is mediated through the PBD of Polo kinase (Clarke, 2005).

Whether the S234 and T331 sites were required in vivo for proper Mei-S332 localization was tested. N-terminal GFP-tagged Mei-S332 wild-type and mutant constructs were transiently expressed in S2 cells. Mei-S332 localization was monitored by GFP fluorescence in interphase and mitotic cells to determine if Mei-S332 was localized properly during the cell cycle. As expected, wild-type Mei-S332 associated with chromosomes in metaphase and delocalized in anaphase. All three mutant proteins localized normally to metaphase chromosomes in comparison to wild-type. However, the T331A and the double mutant proteins remained localized to chromosomes in anaphase, telophase, and interphase in a higher percentage of cells than in wild-type transfected cells (Clarke, 2005).

Quantification of these S2 results revealed that the S234A mutant is the least severe in the failure to dissociate Mei-S332. This mutant shows a similar number of anaphase cells with Mei-S332 localized to centromeres to that of wild-type and only a modest increase in the number of telophase and interphase cells with Mei-S332 chromosomal localization. However, T331A and the double mutant shows a more severe effect on Mei-S332 delocalization. In most cases, these Mei-S332 mutant proteins appeared to localize to entire chromosomes. It is possible that this spreading out of Mei-S332 away from centromeres may be due in part to a metaphase delay, since previously it has been shown that in metaphase-arrested cells, additional Mei-S332 loads onto and persists at noncentromeric sites (Lee, 2004; Tang, 1998). However, centromeric foci of Mei-S332 are clearly visible in a small number of anaphase cells for all three mutants. Overexpression of Mei-S332, particularly in tissue culture cells, can result in protein localization along the chromosomes not solely at the centromeres, and this most likely explains the persistence of the wild-type protein on chromosomes in some anaphase and telophase cells. The increased frequencies of chromosomal localization of mutant GFP-Mei-S332 in anaphase and telophase cells relative to wild-type GFP-Mei-S332 is not due to higher levels of expression of the mutant proteins than the wild-type following transfection. In contrast, Western blotting showed that the wild-type protein is more highly expressed than the mutants. The possibility cannot be excluded that the mutant forms of GFP-Mei-S332 reload onto chromosomes in telophase, as opposed to persisting through the metaphase/anaphase transition. These results suggest that phosphorylation of at least residue T331 and possibly both T331 and S234 are critical for proper delocalization of Mei-S332 from chromosomes during anaphase. Further, these results are consistent with Polo kinase binding to Mei-S332 via phosphorylated PBD binding sites and helping Mei-S332 to be released from centromeres through phosphorylation (Clarke, 2005).

INCENP and Aurora B promote meiotic sister chromatid cohesion through localization of the Shugoshin MEI-S332 in Drosophila.

The chromosomal passenger complex protein INCENP is required in mitosis for chromosome condensation, spindle attachment and function, and cytokinesis. INCENP has an essential function in the specialized behavior of centromeres in meiosis. Mutations affecting Drosophila incenp profoundly affect chromosome segregation in both meiosis I and II, due, at least in part, to premature sister chromatid separation in meiosis I. INCENP binds to the cohesion protector protein MEI-S332, which is also an excellent in vitro substrate for Aurora B kinase. A MEI-S332 mutant that is only poorly phosphorylated by Aurora B is defective in localization to centromeres. These results implicate the chromosomal passenger complex in directly regulating MEI-S332 localization and, therefore, the control of sister chromatid cohesion in meiosis (Resnick, 2006).

This analysis of Drosophila incenp mutants reveals for the first time a crucial role for INCENP in regulating centromeric cohesion during the reductional division of meiosis. INCENP influences the localization and/or function of MEI-S332: precocious sister chromatid separation is observed at the centromeres in the mutants, the distribution of MEI-S332 is abnormal when INCENP levels are decreased, INCENP can bind MEI-S332 in vitro, the protein is phosphorylated in vitro by Aurora B, and MEI-S332 localization to centromeres in mitosis is perturbed when its preferred Aurora B phosphorylation site is mutated (Resnick, 2006).

The QA26 incenp mutation perturbs chromosome condensation and causes precocious separation of the sister chromatids in spermatocytes. Quantitative genetic nondisjunction tests showed that chromosome segregation fails in both meiosis I and II, and that these nondisjunction events are consistent with premature separation of sister chromatids and random segregation in both meiotic divisions. This genetic analysis is likely to underestimate the true rates of nondisjunction because many of the defects caused by loss of passenger function (e.g., defective spindle organization or cytokinesis) would not yield functional gametes, thereby preventing the scoring all of the nondisjunction events. Although the aberrant condensation in prophase and prometaphase I made direct visualization of the onset of loss of cohesion difficult, completely separated sister chromatids could unambiguously be seen in mutant anaphase I cells, confirming one mechanism that contributes to the genetic nondisjunction phenotype (Resnick, 2006).

In C. elegans meiosis, the chromosome passenger complex is necessary for chiasma resolution. If chromosomal passengers were to participate both in regulation of centromeric cohesion as well as processing of chiasmata in C. elegans, essential roles in the latter might obscure roles in the former. In Drosophila male meiosis, there is no synapsis of homologs or recombination. Rather, segregation of homologous chromosomes is regulated via specific pairing sites. The analysis of passenger function was therefore simplified in Drosophila males, where chiasmata do not form (Resnick, 2006).

The MEI-S332-related yeast Shugoshin proteins are critical for the maintenance of the meiotic-specific cohesin subunit Rec8 at centromeres during anaphase Interestingly, no Rec8 homolog has yet been found in Drosophila. The only Drosophila meiotic kleisin, C(2)M, is a component of the synaptonemal complex and has been shown to have an earlier role in female and male meiosis. Thus, what MEI-S332 protects at centromeres in meiosis remains unclear. In mitosis, ablation of MEI-S332 does not lead to premature loss of the mitotic cohesin Rad21 (Resnick, 2006).

In both incenp mutants, impaired INCENP function results in a failure of MEI-S332 localization to centromeres in meiosis. This presumably leads to defects in the protection of cohesion at sister centromeres and contributes to the observed increase in meiotic nondisjunction. The failure to localize MEI-S332 in the incenp mutants is not a general secondary effect of prophase I condensation defects or of premature sister chromatid separation prior to the onset of anaphase I: ord mutants, which display both of those phenotypes, localize MEI-S332 normally. Although the data support a role for MEI-S332 in the increased nondisjunction in incenp mutants, mei-S332 mutants predominantly lead to meiosis II nondisjunction, whereas the incenp alleles show defects in both meiotic divisions. Thus, INCENP must be required for additional functions beyond its role in MEI-S332 regulation described in this study (Resnick, 2006).

One mechanism by which INCENP could promote MEI-S332 function is through its role in establishing or maintaining the specialized chromatin structure around centromeres. The chromosomal passenger complex is involved in regulation of chromatin remodeling complexes like ISWI, and it interacts with histone and nonhistone proteins from the pericentric heterochromatin. Recent studies show a direct link between Aurora B activity and regulation of HP1 localization in mitosis, suggesting a possible role in the regulation of heterochromatin structure. Since heterochromatin is important for cohesin binding to centromeres, it is possible that modifications of both MEI-S332 and the underlying heterochromatin are important for stabilizing centromeric cohesion during meiosis I (Resnick, 2006).

Alternatively, INCENP could act as a platform for the regulation of MEI-S332 at centromeres. The direct binding between INCENP and MEI-S332 could target MEI-S332 to heterochromatin, or it could help to direct its regulation by protein kinases. MEI-S332 binds better in vitro to a mixture of INCENP and Aurora B than to INCENP alone, suggesting that the interaction is strengthened by phosphorylation of either INCENP or MEI-S332. In addition to its role in binding and activating Aurora B, INCENP that has been phosphorylated by CDK1 can bind to Plk1, the human homolog of POLO kinase (Resnick, 2006).

Binding to phosphorylated INCENP is required to target Plk1 to centromeres in mitosis. Thus, INCENP could potentially coordinate the functions of POLO and Aurora B, both of which have been implicated in the regulation of cohesin (and also in the regulation of MEI-S332 in the case of POLO). These kinases have been shown to cooperate in the release of arm cohesion in chromosomes assembled in Xenopus extract. In contrast to Aurora B, however, POLO promotes the dissociation of MEI-S332 from centromeres during mitosis and meiosis. In polo mutants, MEI-S332 persists on the centromere, and mutation of two POLO box domains disrupts POLO binding and phosphorylation of MEI-S332 in vitro, as well as MEI-S332 dissociation from the centromeres (Resnick, 2006).

Together, these observations suggest that INCENP may act to integrate the various pathways controlling MEI-S332 function in meiosis I. Early in meiosis I, INCENP/Aurora B complexes may stabilize centromeric MEI-S332 through direct binding or modification of the underlying chromatin as described above. Similar to what happens in mitosis, CDK1 could phosphorylate INCENP at the POLO binding site, and phosphorylation-dependent binding of POLO to INCENP could target the kinase to the centromere. This binding might also render the kinase unavailable to phosphorylate MEI-S332. During the metaphase-anaphase I transition, INCENP remains on the centromeres and might therefore prevent MEI-S332 from being phosphorylated by POLO. At the onset of anaphase II, however, as INCENP transitions off the centromere, POLO may be free to phosphorylate MEI-S332, thereby releasing it from centromeres, allowing the release of sister chromatid cohesion (Resnick, 2006).

INCENP is emerging as a key regulator of kinase signaling pathways in mitosis. The present study reveals that this versatile protein may have a similar role in meiosis and may use its interactions with Aurora B and POLO to coordinate the specialized behavior of sister chromatids in meiosis I (Resnick, 2006).

Home page: The Interactive Fly © 1997 Thomas B. Brody, Ph.D.

The Interactive Fly resides on the
Society for Developmental Biology's Web server.