BIOLOGICAL OVERVIEW

Centromeres are epigenetically defined by CENP-A-containing chromatin and are essential for cell division. Previous studies suggest asymmetric inheritance of centromeric proteins upon stem cell division; however, the mechanism and implications of selective chromosome segregation remain unexplored. This study shows that Drosophila female germline stem cells (GSCs) and neuroblasts assemble centromeres after replication and before segregation. Specifically, CENP-A deposition is promoted by CYCLIN A, while excessive CENP-A deposition is prevented by CYCLIN B, through the HASPIN kinase. Furthermore, chromosomes inherited by GSCs incorporate more CENP-A, making stronger kinetochores that capture more spindle microtubules and bias segregation. Importantly, symmetric incorporation of CENP-A on sister chromatids via HASPIN knockdown or overexpression of CENP-A, either alone or together with its assembly factor CAL1, drives stem cell self-renewal. Finally, continued CENP-A assembly in differentiated cells is nonessential for egg development. This work shows that centromere assembly epigenetically drives GSC maintenance and occurs before oocyte meiosis (Dattoli, 2020).

Stem cells are fundamental for the generation of all tissues during embryogenesis and replace lost or damaged cells throughout the life of an organism. At division, stem cells generate two cells with distinct fates: (1) a cell that is an exact copy of its precursor, maintaining the 'stemness,' and (2) a daughter cell that will subsequently differentiate. Epigenetic mechanisms, heritable chemical modifications of the DNA/nucleosome that do not alter the primary genomic nucleotide sequence, regulate the process of self-renewal and differentiation of stem cells. In Drosophila male germline stem cells (GSCs), before division, phosphorylation at threonine 3 of histone H3 (H3T3P) preferentially associates with chromosomes that are inherited by the future stem cell (Xie, 2015). Furthermore, centromeric proteins seem to be asymmetrically distributed between stem and daughter cells in the Drosophila intestine and germline. These findings support the 'silent sister hypothesis', according to which epigenetic variations differentially mark sister chromatids driving selective chromosome segregation during stem cell mitosis. Centromeres, the primary constriction of chromosomes, are crucial for cell division, providing the chromatin surface where the kinetochore assembles. In turn, the kinetochore ensures the correct attachment of spindle microtubules and faithful chromosome partition into the two daughter cells upon division. Centromeric chromatin contains different kinds of DNA repeats (satellite and centromeric retrotransposons) wrapped around nucleosomes containing the histone H3 variant centromere protein A (CENP-A). Centromeres are not specified by a particular DNA sequence. Rather, they are specified epigenetically by CENP-A. Centromere assembly, classically measured as CENP-A deposition to generate centromeric nucleosomes, occurs at the end of mitosis (between telophase and G1) in humans. Additional cell cycle timings for centromere assembly have been reported in flies. Interestingly, Drosophila spermatocytes and starfish oocytes are the only cells known to date to assemble centromeres before chromosome segregation, during prophase of meiosis I. These examples show that centromere assembly dynamics can differ among metazoans and also among different cell types in the same organism (Dattoli, 2020).

A key player in centromere assembly in vertebrates is HJURP (holliday junction recognition protein), which localizes at centromeres during the cell cycle window of CENP-A deposition. Furthermore, centromere assembly is regulated by the cell cycle machinery. In flies, deposition of CID (the homologue of CENP-A) requires activation of the anaphase promoting complex/cyclosome (APC/C) and degradation of CYCLIN A (CYCA). In humans, centromere assembly is antagonized by Cdk1 activity, while the kinase Plk1 promotes assembly. Additionally, the CYCLIN B (CYCB)/Cdk1 complex inhibits the binding of CENP-A to HJURP, preventing CENP-A loading at centromeres. To date, little is known about centromere assembly dynamics and functions in stem cell asymmetric divisions. Drosophila melanogaster ovaries provide an excellent model to study stem cells in their native niche. In this tissue, germline stem cells (GSCs) are easily accessible and can be manipulated genetically. Moreover, centromere assembly mechanisms in GSCs and their differentiated cells, cystoblasts (CBs), could be used to epigenetically discriminate between these two cell types. In Drosophila, CID binds to CAL1 (fly functional homologue of HJURP) in a prenucleosomal complex, and its localization to centromeres requires CAL1 and CENP-C (Dattoli, 2020).

This study investigated the dynamics of CENP-A deposition in Drosophila GSCs. GSC centromeres are assembled after replication, but before chromosome segregation, with neural stem cells following the same trend. Centromere assembly in GSCs is tightly linked to the G2/M transition. Indeed, CYCA localizes at centromeres, and its knockdown is responsible for a marked reduction of centromeric CID and CENP-C, but not CAL1. Surprisingly, excessive CID deposition is prevented by CYCB, through the kinase HASPIN. Superresolution microscopy analysis of GSCs at prometaphase and metaphase shows that CID incorporation on sister chromatids occurs asymmetrically, and chromosomes that will be inherited by the stem cell are loaded with more CID. Moreover, GSC chromosomes make stronger kinetochores, which anchor more spindle fibers. This asymmetric distribution of CID between GSC and CB is maintained also at later stages of the cell cycle, while it is not observed in differentiated cells outside of the niche. This study also found that the depletion of CAL1 at centromeres blocks GSC proliferation and differentiation. Notably, overexpression of both CID and CAL1, as well as HASPIN knockdown, promotes stem cell self-renewal and disrupts the asymmetric inheritance of CID. Conversely, overexpression of CAL1 causes GSC-like tumors. Finally, CAL1 and CID knockdown at later stages of egg development have no obvious effect on cell division, suggesting that these cells inherit CID from GSCs. Taken together, these findings establish centromere assembly as a new epigenetic pathway that regulates stem cell fate (Dattoli, 2020).

In this study a detailed characterization of centromere dynamics was performed throughout the cell cycle in Drosophila female GSCs. This analysis reveals that GSCs initiate CID incorporation after replication and that its deposition continues until at least prophase. Drosophila neural stem cells follow the same trend. Notably, this timing is different from existing studies in other metazoans. It was also found that CYCA, CYCB, and HASPIN are critically involved in CID (and CENP-C) loading at centromeres. According to the model, CYCA promotes centromere assembly, while CYCB prevents excessive deposition of CID, through the HASPIN kinase. Moreover, chromosomes that will be inherited by GSCs are labeled with a higher amount of CID and capture more spindle microtubules. Importantly, this study shows that overexpression of CAL1 and CID together, as well as HASPIN knockdown, promotes stem cell self-renewal, disrupting the asymmetric inheritance of CID. Depletion of CAL1 in stem cells blocks cell division, while CAL1 overexpression causes GSC-like tumors, highlighting its crucial role in cell proliferation. Three main points of discussion are raised: (1) the biological significance of centromere assembly in G2-M phase; (2) CAL1 is a cell proliferation marker; and (3) CID incorporation into centromeric chromatin occurs before meiosis (Dattoli, 2020).

According to the data, CID deposition occupies a wide window of time from after replication and early G2 phase to prophase. The assembly of GSC centromeres during the G2/M transition could be due to the contraction of the G1 phase, a characteristic of stem cells. Yet, in fly embryonic divisions, G1 phase is missing, and instead CID loading occurs at anaphase. Therefore, G2/M assembly might be a unique property of stem cells. This timing is also similar to the one found for Drosophila spermatocytes, which assemble centromeres in prophase of meiosis I. These spermatocytes undergo an arrest in prophase I for days, indicating a gradual loading of CID over a long period of time. Intriguingly, a similar phenomenon has been recently observed in G0-arrested human tissue culture cells and starfish oocytes. Given that GSCs are mostly in G2 phase, Drosophila stem cells might show similar properties to quiescent cells. According to the most recent models, there is a dual mechanism for CENP-A deposition: (a) a rapid pulse during G1 in mitotically dividing cells; and (b) a slow but constant CENP-A deposition in nondividing cells to actively maintain centromeres. Indeed, while previous studies in Drosophila NBs show a rapid pulse of CENP-A incorporation at telophase/G1, the majority of the loading could occur between G2 and prophase. The new results also support this model (Dattoli, 2020).

Incorporation of CID before chromosome segregation might reflect a different CYCLIN-CDK activity in these cells. For instance, it has been already shown that in Drosophila GSCs CYCLIN E, a canonical G1/S cyclin, exists in its active form (in combination with Cdk2) throughout the cell cycle, indicating that some of the biological process commonly occurring in G1 phase might actually take place in G2 phase. This is in line with the current functional findings, where depletion of CYCA causes a decreased efficiency in CID and CENP-C assembly. This study also found that this loss might be independent from CAL1. Surprisingly, correct CID deposition in GSCs also requires CYCB and HASPIN. Indeed, an inhibitory mechanism for CID deposition through CYCB has already been proposed in mammals (Stankovic, 2017). Interestingly, in Drosophila male GSCs, centromeric CAL1 is reduced between G2 and prometaphase (Ranjan, 2019), further suggesting a role for additional regulators of CID assembly, such as CYCA/B or HASPIN, at this time (Dattoli, 2020).

According to the current results, asymmetric cell division of GSCs is epigenetically regulated by differential amounts of centromeric proteins deposited at sister chromatids, which in turn can influence the attachment of spindle microtubules and can ultimately bias chromosome segregation. It is interesting to speculate on the temporal sequence of these events. Two scenarios can be proposed: (a) the nucleation of microtubules from the GSC centrosome requires bigger kinetochores; or (b) bigger kinetochores require a higher amount of spindle fibers to attach. The current results together with recent studies support the latter scenario. In fact, in Drosophila male GSCs, asymmetric distribution of centromeric proteins is established before microtubule attachment. Furthermore, microtubule disruption leaves asymmetric loading of CID intact, while it disrupts the asymmetric segregation of sister chromatids. The current data confirm this model, symmetric segregation of CID was observed upon HASPIN knockdown. Indeed, in vertebrates HASPIN knockdown causes spindle defects. Specifically, it was observed that a 1.2-fold difference in CID and CENP-C levels between GSC and CB chromosomes can bias segregation. While this difference is small, it fits with the observation that small changes in CENP-A level (on the order of 2-10% per day) impact on centromere functionality in the long run. In Drosophila male GSCs, an asymmetric distribution of CID on sister chromatids >1.4-fold was reported. This higher value might reflect distinct systems in males and females or the quantitation methods used. Importantly, CID asymmetry in males is established in G2/prophase, in line with the time window this study defines for CID assembly. Further support for unexpected CID loading dynamics comes from the finding that GSCs in G2/prophase contain ~30% more CID on average compared with S phase, indicating that CID is not replenished to 100% each cell cycle. Interestingly, the time course of H3T3P appearance during the GSC cell cycle closely follows the timing of CID incorporation, suggesting that the asymmetric deposition of CID might drive the differential phosphorylation of the histone H3 on sister chromatids. Finally, the results are in line with findings that the long-term retention of CENP-A in mouse oocytes has a role in establishing asymmetric centromere inheritance in meiosis (Dattoli, 2020).

These functional studies support a role for CAL1 in cell proliferation, with no apparent role in asymmetric cell division. Indeed, centromeric proteins have been already proposed as biomarkers for cell proliferation. Specifically, functional analysis of centromeric proteins, as well as the HASPIN kinase, allowed discrimination between the classic role of centromeres in cell division and a role in asymmetric cell division. In a favorite scenario, CAL1 is needed to make functional centromeres crucial for cell division, while the asymmetric distribution of CID sister chromatids regulates asymmetric cell division and might depend on other factors, such as HASPIN. However, it cannot be rule out that the effects on cell fate observed with the functional analysis might reflect alternative CAL1 functions outside of the centromere, for example due to changes in chromosome structure or gene expression (Dattoli, 2020).

Centromeres are crucially assembled in GSCs and therefore before meiosis of the oocyte takes place. Thus, it is possible that the 16-cell cysts inherit centromeric proteins synthesized and deposited in the GSCs, and the rate of new CID loading is reduced. This would explain why CAL1 function at centromeres is dispensable at this developmental stage (Dattoli, 2020).

Ultimately, the results provide the first functional evidence that centromeres have a role in the epigenetic pathway that specifies stem cell identity. Furthermore, these data support the silent sister hypothesis (Lansdorp, 2007), according to which centromeres can drive asymmetric division in stem cells (Dattoli, 2020).

Adaptive evolution of Cid, a centromere-specific histone in Drosophila

Centromeres are the chromosomal regions responsible for poleward movement at meiosis and mitosis, and are essential for the faithful segregation of genetic information. Centromeres of most organisms are embedded within constitutive heterochromatin, the condensed regions of chromosomes that account for a large fraction of complex genomes. Centromere function requires the coordination of many processes including kinetochore assembly, sister chromatid cohesion, spindle attachment and chromosome movement. Centromeric chromatin is distinguished from bulk chromatin, most conspicuously by the presence of centromere-specific histone H3 variants (CenH3). A Drosophila CenH3, Cid (for Centromere identifier), the Drosophila homolog of the CENP-A centromere-specific H3-like proteins, localizes exclusively to fly centromeres. The function of Cid is highly conserved. This chromatin component probably plays a key role in assembling the kinetochore at meiosis and mitosis. Thus, CenH3 could be expected to both interact with the underlying centromeric DNA, as well as interact with other proteins (including other CenH3 molecules) to provide the foundation for the kinetochore (Malik, 2002).

Remarkably, when the cid upstream promoter region drives expression of yeast, worm, and human centromeric histone proteins, localization is preferentially within Drosophila pericentric heterochromatin. Heterochromatin-specific localization also was seen for yeast and worm centromeric proteins constitutively expressed in human cells. Thus, these H3-like proteins from yeast and worms localize to pericentric heterochromatic regions surrounding fly and human centromeres. Preferential localization to heterochromatin in heterologous systems is unexpected if species specific centromere-specific or site-specific factors determine H3-like protein localization to centromeres. Rather, Cid is part of a specific conserved heterochromatic region surrounding centromeres (Henikoff, 2000).

The role of Drosophila CID in kinetochore formation, cell-cycle progression and heterochromatin interactions

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).

Immunolocalization experiments in mammals have shown that it is possible to distinguish between the inner and outer kinetochore by fluorescence microscope. To determine whether Cid is located in the inner kinetochore, Cid, spindle microtubules and many transient kinetochore proteins including Bub1, Zeste-white 10 (">ZW10 -- accepted FlyBase name: Mitotic 15) and Rod were simultaneously localized. Each of these proteins would be expected to localize to the outer kinetochore plate or the fibrous corona, similar to other transient kinetochore proteins localized in mammals (for example, Bub1, CENP-E, Dynein) (Blower, 2001).

Simultaneous detection of Cid with outer kinetochore proteins showed that Cid is consistently separated from ZW10, Rod and Polo kinase, and is located closer to the chromosome and further from the kinetochore microtubules than these proteins. Cid is also offset from Bub1 (a component of the spindle assembly checkpoint) at unattached kinetochores, but Cid and Bub1 show significant overlap. This result is consistent with studies in mammals, which suggest that Bub1 may be located at both the inner and outer kinetochore plates. These results show that Cid is located in or near the inner plate of the kinetochore in Drosophila and is likely to be associated closely with centromeric DNA (Blower, 2001).

Previous work has shown that the outer kinetochore proteins ZW10 and Dynein are present on fully functional Drosophila minichromosomes (that is, 100% transmission through mitosis and meiosis), as well as structurally acentric minichromosomes that lack detectable centromeric sequence (neocentromeres). To determine the relationship of Cid-containing chromatin to the functional centromere, the localization of Cid protein was examined in a series of minichromosome derivatives of decreasing size and meiotic transmission efficiency (Blower, 2001).

Cid is present on all minichromosome derivatives that contain a fully functional centromere (gamma238, 31E, 10B and J21A, indicating that Cid colocalizes with the molecular-genetically defined centromere. Cid also is present on all of the neocentromeric derivatives that lack centric heterochromatin, including the normal minichromosome centromere (26C, J19B), consistent with the localization pattern of outer kinetochore components ZW10 and Dynein. It is concluded that Cid localization is correlated with centromere function, regardless of the composition of the underlying DNA (Blower, 2001).

The presence of Cid at the functional minichromosome centromere shows that Cid localization is correlated with centromere activity. To test the role of Cid in mitosis directly, affinity-purified chicken anti-Cid antibodies were injected into early embryos that express histone H2A/green fluorescent protein (GFP) and chromosome behavior was observed using time-lapse microscopy. Rhodamine-labelled anti-Cid antibodies bind specifically to centromeres in all stages of the cell cycle, and show no nonspecific crossreactivity with histone H3 either in vivo or by Western blot. Injected antibody binds centromeres in a gradient in which more antibody binds close to the site of injection (Blower, 2001).

Injection of Cid antibodies into early embryos results in a range of phenotypes affecting both cell-cycle progression and mitotic chromosome segregation. The phenotypic series is consistent with a gradient of Cid inhibition mirroring the gradient of antibody concentration. Nuclei closest to the site of injection arrest in interphase (13%), whereas nuclei further from the site of injection delayed entering mitosis and exhibited different mitotic defects: specifically, entry into prophase condensation followed by a loss of condensation (3.6%; metaphase arrest [15%]); and various anaphase chromosome segregation defects (failure to move toward the poles at anaphase onset, unequal chromosome segregation, failure to maintain spindle contact and karyokinesis defects at telophase, 20%) (Blower, 2001).

Cid function in Drosophila Kc tissue-culture cells is also disrupted using RNAi. Cells were treated with dsRNA corresponding to the whole Cid transcript; cells were then fixed and monitored for levels of Cid protein and aberrant chromosome behaviour every 24 h after adding dsRNA. Cells in a given treated population display a variable penetrance of Cid inhibition, which results in different phenotypes (Blower, 2001).

Mitotic defects in Kc cells are consistent with those observed after antibody injection into embryos, including aberrant prometaphase congression, precocious sister chromatid separation, kinetochore microtubule capture and anaphase segregation. Cells treated with dsRNA were no longer dividing 8-10 d after RNAi, suggesting that interphase arrest results from complete Cid disruption; however, without live analysis it is difficult to differentiate between interphase arrest and a terminal phenotype resulting from massive chromosome segregation defects (Blower, 2001).

The results of Cid disruption by both antibody injection into embryos and RNAi in tissue-culture cells show that Cid is directly or indirectly required for many aspects of kinetochore-mediated chromosome movement as well as cell-cycle progression (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).

CENP-A has been proposed to be an epigenetic mark required for determining centromere identity and thus the assembly of the kinetochore. This study has shown that Cid, the Drosophila CENP-A homolog, is associated with the inner kinetochore and that Cid localization correlates with centromere function, specifically the presence of the molecular-genetically defined centromere and neocentromere DNA (Blower, 2001).

The presence of Cid on neocentromeres shows that these structurally acentric yet functional chromosomes have acquired centromeric chromatin, consistent with the presence of outer kinetochore components. Furthermore, the presence of Cid on neocentromeres shows that the location of centromeric chromatin is independent of sequence, and that centromeric chromatin can confer centromere identity on normally non-centromeric DNA -- a state that is then propagated faithfully through replication and division (Blower, 2001).

Cid is required for kinetochore assembly and cell-cycle progression in early Drosophila embryos and Kc tissue-culture cells. Mitotic defects have been observed in human cells after CENP-A antibody injection and in the mouse CENP-A knockout; however, the 'live studies' reported in this study have allowed an examination of the temporal and cytological effects of CENP-A/Cid disruption in greater detail. Both antibody injection into embryos and RNAi inhibition in Kc cells results in several defects expected for centromere dysfunction: specifically, aberrant prometaphase congression; chromosome attachment to spindle microtubules; entry into anaphase; anaphase poleward segregation,and failure to resolve properly at telophase (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).

Cid disruption may decrease the number or size of functional subunits; this is sufficient to cause defects in mitosis because mitotic defects are observed in cells with decreased but visible amounts of Cid. Disruption of the centromere/kinetochore substructure may be responsible for different degrees of mislocalization of outer kinetochore components, which results in the observed pleiotropy of mitotic phenotypes. Karyokinesis defects observed after Cid inhibition may be the result of the failure of chromosomal passenger proteins to localize to the kinetochore and consequently to the spindle and midbody. On the basis of the mislocalization of several outer kinetochore components, it is concluded that Cid is epistatic to transient kinetochore components; these results support the hypothesis that CENP-A proteins are involved directly in the epigenetic marking of the site of kinetochore formation, and show conclusively that proper kinetochore function is required for many cell-division processes (Blower, 2001).

More complete Cid disruption in embryos results in a severe interphase arrest phenotype, showing that Cid function is also required before entry into mitosis, which is similar to one of the phenotypes observed after injection of anti-CENP-A antibodies into HeLa cells. The use of real-time analysis allows for an unambiguous conclusion that the nuclei are arrested before entry into mitosis. The interphase arrest phenotype suggests that Cid functions in interphase, where it is constitutively bound to centromeres, and that there may be another cell-cycle checkpoint that monitors kinetochore assembly and blocks entry into mitosis if this process is compromised (Blower, 2001).

This putative 'kinetochore assembly' checkpoint is also likely to be responsible for the delay in entering mitosis observed in nuclei just distal to the injection site. It is possible that disruption of an interaction between Polo kinase and Cid is responsible for this cell-cycle arrest because Polo is required for entry into mitosis. Clearly, further investigation is necessary to elucidate the precise pathway involved in this arrest, and to determine whether there is a checkpoint that monitors kinetochore assembly before entry into mitosis (Blower, 2001).

Genetic and protein localization studies have implicated several proteins in regulating centromere function. Until now, it has been difficult to determine whether these proteins are involved in kinetochore formation and function, other centromere functions, or functions independent of the centromere/kinetochore. The structural and functional analyses presented in this study support the hypothesis that distinct spatial and functional domains exist in the centromere and adjacent regions. Previously, cytological studies in humans have revealed the presence of spatially distinct protein domains in the mitotic kinetochore (Blower, 2001).

This study extends these observations by investigating the domain structure of the kinetochore and the centromere region, including the flanking heterochromatin, and has provided data that reveal the functional separation and interdependence of these structural domains. Centromeric chromatin is the central and most essential component of the centromere region, and is required for entry into mitosis and chromosome movement during mitosis, as well as for recruiting proteins to the kinetochore and flanking domains. The domain organization of the centromere observed in Drosophila is similar to the situation in S. pombe, where different proteins occupy distinct subdomains within the centromere region, and are required for separable chromosome segregation processes (Blower, 2001).

Studies in a variety of organisms have indicated that the centromere region is a site of specialized sister cohesion, not only in meiosis I but also in mitosis. For example, normal homolog disjunction requires that Mei-S332 functions to maintain sister chromatid cohesion in the centric regions throughout meiosis I and until anaphase of meiosis II. Although Mei-S332 is involved in sister chromatid cohesion, it is not a cohesin; moreover, its localization differs from that observed in S. cerevisiae, where cohesins are concentrated at the centromere and associated with centromeric chromatin (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).

Heterochromatin encodes several inheritance functions, including homolog pairing in meiosis I, sister chromatid cohesion and interactions with anti-poleward forces. The conserved location of centromeres in heterochromatin suggests that heterochromatin proteins, such as Prod and HP1, may be required for establishing or maintaining centromeres. Neither Prod nor HP1 are detectable in Cid chromatin and prod and Su(var)2-5 mutations do not affect the localization of Cid. Furthermore, neither Prod nor HP1 localization is affected in anti-Cid-injected embryos or RNAi-treated Kc cells (Blower, 2001).

It is concluded that Prod and HP1 function in the pericentromeric regions to promote normal condensation and chromosome segregation -- processes distinct from the centromere/kinetochore. Although the kinetochore is typically embedded in large blocks of heterochromatin, evidence has been provideed that it may be structurally and functionally distinct from the closely juxtaposed pericentromeric or centric heterochromatin (Blower, 2001).

If centromere chromatin structure is distinct from centric heterochromatin, why are centromeres embedded in heterochromatin in almost all multicellular eukaryotes? Perhaps the flanking heterochromatin does provide an environment that is necessary for the formation of a centromere-specific higher order chromatin structure. In addition, although Prod and HP1 are not necessary for Cid localization, they may encode functions redundant with other heterochromatic proteins that establish or maintain proper kinetochore structure. It will be interesting to determine whether protein localization and mutant analyses with other centric heterochromatin proteins are consistent with the results for prod and Su(var)2-5 (Blower, 2001).

Drosophila CENP-A/Cid is required for kinetochore formation and mitotic function, cell-cycle progression and recruiting transient kinetochore components and a sister chromatid cohesion protein. In contrast, Cid and proteins that function in the pericentric heterochromatin are physically and functionally independent. These results support the hypothesis that CENP-A proteins are central to many mitotic processes, and may be a component of the epigenetic mark responsible for centromere identity and function. Future studies should investigate what mechanism is responsible for loading CENP-A specifically into CENP-A chromatin in a replication-independent manner, since this process may be the key to understanding maintenance of the epigenetic mark and centromere identity (Blower, 2001).

Centromeric nucleosomes induce positive DNA supercoils

Centromeres of higher eukaryotes are epigenetically maintained; however, the mechanism that underlies centromere inheritance is unknown. Centromere identity and inheritance require the assembly of nucleosomes containing the CenH3 histone variant (Centromere identifier) in place of canonical H3. Although H3 nucleosomes wrap DNA in a left-handed manner and induce negative supercoils, this study shows that CenH3 nucleosomes reconstituted from Drosophila histones induce positive supercoils. Furthermore, CenH3 likewise induces positive supercoils in functional centromeres in vivo, using a budding yeast minichromosome system and temperature-sensitive mutations in kinetochore proteins. The right-handed wrapping of DNA around the histone core implied by positive supercoiling indicates that centromere nucleosomes are unlikely to be octameric and that the exposed surfaces holding the nucleosome together would be available for kinetochore protein recruitment. The mutual incompatibility of nucleosomes with opposite topologies could explain how centromeres are efficiently maintained as unique loci on chromosomes (Furuyama, 2009).

CenH3 nucleosomes induce positive supercoils, both when D. melanogaster CID is reconstituted into nucleosomes in vitro, and when S. cerevisiae Cse4p is assembled at functional minichromosome centromeres in vivo. This behavior is in stark contrast to canonical nucleosomes, in which the left-handed wrapping leads to induction of negative supercoils in topological assays. These observations of positive supercoiling induced by CenH3 from eukaryotic taxa as different as animals and fungi can be explained by either of two general models: overtwisting with left-handed wrapping or right-handed wrapping (Furuyama, 2009).

In a covalently closed circle, overtwisting of DNA (positive ΔTw) causes compensatory negative writhe that is removed by topoisomerase, resulting in a net positive ΔLk after deproteination. If CenH3 nucleosomes are left-handed octamers (Wr = −1), ΔTw would need to be +2 in order to result in a ΔLk of +1 (ΔLk = ΔTw + ΔW). Although the reported value of ΔWr for left-handed octamers varies, this study used the most conservative cited value of −1 to calculate the degree of overtwisting consistent with left-handed wrapping. The change required in the helical periodicity of DNA (Δh) to gain ΔTw = +2 and cancel one negative writhe induced by a left-handed nucleosome can be calculated as Δh = −h²xΔTw/N, where h = N/Tw, where N is the number of base pairs wrapped around the nucleosome. If an octameric CenH3 nucleosome (N=150 bp) is assumed, Δh equals −1.47 for ΔTw = +2. This corresponds to a helical periodicity of 9.03 bp/turn (whereas h = 10.5 bp/turn for B-DNA free in solution). The situation is even more extreme for CenH3 hemisomes, which wrap 80-120 bp of DNA, because the same amount of twist must be taken up by the shorter span of DNA (helical periodicity of 7.74-8.66 bp/turn). These estimated values for helical twist are conservative in that they assume that the extra twist is distributed over the whole nucleosome, including the DNA that wraps H2A/H2B dimers, whereas in the crystal structure of the H3 nucleosome core particle the twist of DNA wrapping H2A/H2B is similar to that in free solution. In addition, DNaseI digestion of CID chromatin assembled in vitro resulted in a normal helical periodicity estimate of ~10 bp/turn, and electron microscopy of CID chromatin revealed a beads-on-a-string appearance, suggesting entry/exit crossing. Thus, existing data are inconsistent with positive ΔTw being the reason for the observed positive supercoiling (Furuyama, 2009).

The implausibility of such strongly overtwisted DNA wrapping around a left-handed nucleosome leads to the conclusion that positive supercoiling instead indicates a right-handed wrap. A right-handed nucleosome would satisfy the observed positive supercoiling of approximately one supercoil per CenH3 nucleosome without a significant change in B-DNA periodicity. Tetrameric archaeal nucleosomes also wrap DNA in a right-handed configuration, with a helical periodicity of 10-11bp/turn. Also, in the absence of H2A/H2B dimers, (H3/H4)₂ tetramers are capable of spontaneously shifting between both left- and right-handed configurations, presumably without significant changes in helical twist (Furuyama, 2009).

Histone octamers capable of wrapping DNA into a right-handed configuration have never been observed. Because H3/H4 tetramers can wrap DNA in either direction, it is the creation of a left-handed ramp by addition of two H2A/H2B dimers that is incompatible with the right-handed structure. The crystal structure of the H3 nucleosome (H2A'-H2B'-H4'-H3'-H3-H4-H2B-H2A plus DNA) reveals that the N-terminal helix of H3, as well as the C-terminus of H4, contact the C-terminal docking domain of H2A', which are essential interactions that hold the octamer together. In addition, the interaction between H2A and H2A' within the octamer through their Loop 1 regions hold together the two gyres of the DNA superhelix (Luger, 1997). These interactions that hold the octamer together are expected to be disrupted in a right-handed nucleosome because they would face away from each other in the right-handed structure; therefore, there is a strong structural basis for the absence of right-handed octameric nucleosomes in eukaryotes. Without altering the twist of DNA significantly, the only structures that yield ΔLk = +1 other than a right-handed octamer are right-handed hemisomes with right entry/exit crossing, and left-handed hemisomes with right entry/exit crossing. A single superhelical turn of DNA around a hemisome results in a closer physical distance between the entry/exit DNA compared to that in an octameric structure, which has an additional turn between the two entry/exit sites (see Structures and model of a right-handed hemisome). Therefore, it is structurally very difficult to make a left-handed hemisome with a right-handed crossing (Furuyama, 2009).

In budding yeast, various model of Cse4p nucleosomes have been suggested, including octamers (H2A/H2B/H4/Cse4p/Cse4p/H4/H2B/H2A), hemisomes (Cse4p/H4/H2B/H2A) and nucleosomes containing the non-histone Scm3 protein substituting for H2A/H2B dimers (H4/Cse4p/Cse4p/H4)(Scm3)_1-2. Given the current finding that Cse4p nucleosomes induce positive supercoils, it is unlikely that they can exist as octamers. Furthermore, the observation that Scm3 binds to the region of Cse4p required for the 4-helix bundle homodimerization interface of the octameric particle would a priori argue against a stable octameric particle. That leaves either Cse4p/H4/H2A/H2B hemisomes or Cse4p/H4/Scm3 particles as candidate yeast CenH3 nucleosomes. Both of these models are consistent with the localization of Cse4p to a small ~80bp CDEII region of CDE. It is attractive to suggest that right-handed hemisomes are conserved in all eukaryotes, because Cse4p can functionally replace human CENP-A (Furuyama, 2009).

There are several structural implications of right-handed hemisomes. The strong H3/H3 4-helix bundle at the dyad axis and the weak H4/H2B 4-helix bundles linking the central tetramer to flanking dimers precludes formation of H3/H4/H2B/H2A hemisomes, and indeed no stable H3 hemisomes have been observed. Therefore, the existence of CenH3 hemisomes suggests that CenH3 induces structural alterations that stabilize the tetrameric particle. The crossing of entry/exit DNA in the CenH3 hemisome may be an important feature, because it can potentially stabilize the hemisome. In contrast, the entry/exit DNA of H3 octameric nucleosomes does not cross most of the time, but rather is occupied by a linker histone. Consistent with this difference, the H1 linker histone is depleted from centromeric chromatin, and the H5 linker histone is incapable of associating with human CENP-A nucleosomes in vitro. In addition, surfaces involved in contacts within left-handed octameric nucleosomes will be exposed in right-handed hemisomes, such as the C-terminal docking domain of H2A and the N-terminal helix of H3. A right-handed configuration also changes the relative position of these domains. The combination of additional exposed surfaces and altered presentation of the same surfaces might provide essential interaction domains for kinetochore proteins to assemble functional centromeres (Furuyama, 2009).

The finding that CenH3 nucleosomes are right-handed also might help explain why key residues involved in H3/H3 4-helix bundle formation are invariant in CenH3s, despite considerable divergence elsewhere in the core. This observation suggests that the CenH3 dimerization interface is occupied under at least some circumstances. It is suggested that this interface has been retained to permit CenH3/H3 hybrid formation, which would result in left/right core particles that should be unable to stably wrap DNA. Misincorporation of CenH3 outside of centromeres occurs under many circumstances, yet is potentially catastrophic, causing dicentric formation, chromosome loss and dominant lethality. By retaining the ability to dimerize with H3, misincorporated CenH3s would predominantly form structurally defective nucleosomes, thus helping to maintain the extraordinary fidelity of centromere maintenance (Furuyama, 2009).

At the boundary between CenH3 and H3 nucleosomal arrays, the change in the direction of DNA around histones from left-handed to right-handed might also have profound implications for maintaining functional centromeres. The uniform packaging of H3 nucleosomes in pericentric heterochromatin, induced in part by the uniform size of centromeric satellite repeats, is expected to be disturbed by the sudden change in the direction of DNA wrapping around CenH3. This would result in a higher-order structural transition from near-crystalline rigid heterochromatin to less densely packaged centromeric chromatin as implied by the unusually long linker DNA found in Drosophila centromeric chromatin. The octameric form of canonical H3 nucleosomes is believed to represent a critical evolutionary leap in being able to more densely package the genome, whereas tetrameric archael nucleosomes fail to condense into a comparable higher order packaging. Therefore, the presence of a CenH3 hemisome array that packages DNA in a right-handed orientation and resists octameric packaging would provide a singular location that remains decondensed during mitosis and accessible to binding by kinetochore proteins. The mutual incompatibility of nucleosome cores that wrap DNA in opposite directions suggests a novel mechanism for perpetual maintenance of the centromere within a chromosomal landscape that is dominated by conventional chromatin (Furuyama, 2009).

ATP synthase F1 subunits recruited to centromeres by CENP-A are required for male meiosis

The histone H3 variant CENP-A epigenetically defines the centromere and is critical for chromosome segregation. This study reports an interaction between CENP-A and subunits of the mitochondrial ATP synthase complex in the germline of male Drosophila. Furthermore, knockdown of CENP-A, as well as subunits ATPsyn-alpha, ATPsyn-betalike (a testis-specific paralogue of ATPsyn-beta) and ATPsyn-gamma disrupts sister centromere cohesion in meiotic prophase I. This disruption is likely independent of reduced ATP levels. ATPsyn-alpha and -betalike were found to localise to meiotic centromeres; and this localisation is dependent on the presence of CENP-A. ATPsyn-alpha directly interacts with the N-terminus of CENP-A in vitro, and truncation of its N terminus perturbs sister centromere cohesion in prophase I. It is proposed that the CENP-A N-terminus recruits ATPsyn-alpha and -betalike to centromeres to promote sister centromere cohesion in a nuclear function that is independent of oxidative phosphorylation (Collins, 2018).

Meiosis is the specialised cell division cycle in which one round of DNA replication precedes two rounds of chromosome segregation that generate haploid gametes (eggs and sperm). Defects in meiosis lead to reduced fertility, sterility or aneuploidy in gametes or resulting zygotes. Centromeres, defined epigenetically by incorporation of the histone H3 variant CENP-A2 play a key role in coordinating meiotic chromosome segregation. Studies in plants suggest that CENP-A adopts meiosis-specific functions via its highly divergent N terminus. To investigate functions of the CENP-A N terminus in meiosis in an animal, this study used biochemical and genetic approaches in testis of Drosophila males. Unexpected functional links were uncovered between CENP-A, mitochondrial ATP synthase F1 subunits and sister centromere cohesion in meiosis. It is proposed that the CENP-A N-terminus recruits ATPsyn-α and -βlike, a testis-specific paralogue of -β, to centromeres to promote sister centromere cohesion in a novel nuclear function that is independent of canonical roles in oxidative phosphorylation (Collins, 2018).

In addition to an expected function in ATP synthesis, this paper reports a function for ATPsyn-α and ATPsyn-βlike in male meiosis and fertility. In testes depleted for ATPsyn-α or –βlike, prophase I cells accumulate prior to meiosis I, providing a possible explanation for observed sterility in previously isolated mutants. Given that canonical ATPsyn-β expression in testis is reduced compared to whole adults, ATPsyn-βlike might normally compensate for ATPsyn-β function. Moreover, although the expression pattern of ATPsyn-βlike is entirely consistent with a testis-specific function, ATPsyn-βlike expression is noted at larval and pupal stages (modENCODE RNA-Seq). This raises the possibility that ATPsyn-βlike adopts additional functions in development, which were not addressed in this study. In addition to its canonical role, a nuclear function is reported for ATPsyn-α and -βlike, in particular at centromeres. ATPsyn-α and -βlike localise to centromeres at meiotic prophase I, and this localisation requires CENP-A. Moreover, CENP-A, ATPsyn-α and ATPsyn-βlike are each required to maintain sister centromere cohesion at this stage. Remarkably, ATPsyn-βlike specifies the enrichment of the cohesion protector MEI-S322 to centromeres at prometaphase I, perhaps comparable to the Chromosome Passenger Complex subunit INCENP. In contrast, ATPsyn-α appears to have a distinct function in the nuclear and centromeric localisation of MEI-S332. MEI-S332 mislocalisation to global chromatin in ATPsyn-βlike-depleted nuclei is particularly striking and might be a consequence of a sustained prometaphase I arrest or indicates a more general function of ATPsyn-βlike on chromatin. In Drosophila males, an alternative cohesin complex made up of ORD, SOLO and SUNN maintains meiotic sister centromere cohesion at late prophase I S6. This study also found that CENP-A is required for centromere cohesion early in prophase I at S1/2a, prior to ORD, SOLO and SUNN. It is suggested that observed defects in cohesion lead to failed progression through meiosis I and ultimately reduced fertility or sterility. Intriguingly, depletion of ATP synthase F1 subunits also disrupts sister chromatid arm cohesion and 4th homologue pairing/cohesion, suggesting additional global functions outside of the centromere. The ATPsyn-α subunit directly interacts with the CENP-A N terminus, providing a first function for conserved B1 and B2 domains. It is proposed that ATPsyn-α recruits ATPsyn-βlike to centromeres. This functional analyses of flies expressing a GFP-tagged CENP-A lacking amino acids 1–118 show the CENP-A N terminus is not required for meiotic centromere localisation, different from plants. Instead, the fly CENP-A N terminus appears to be important for meiotic sister centromere cohesion, possibly via the recruitment of ATPsyn-α and ATPsyn-βlike (Collins, 2018).

These data support a model in which mitochondrial ATP synthase F1 subunits adopt nuclear functions that appear to be independent of ATP production. First, ATPsyn-α and ATPsyn-βlike interact with CENP-A/centromeres. Second, the severity of observed meiotic phenotypes does not correlate with ATP supply. Third, ATP depletion was not sufficient to induce a loss of cohesion. Finally, these findings are in line with an ATP independent requirement for ATPsyn-α, -β and -γ in germ line stem cell differentiation in Drosophila females. In conclusion, it is proposed that the CENP-A N-terminus recruits ATPsyn-α and ATPsyn-βlike to centromeres to promote sister centromere cohesion in a novel nuclear function that is independent of canonical roles in oxidative phosphorylation (Collins, 2018).

The checkpoint protein Zw10 connects CAL1-dependent CENP-A centromeric loading and mitosis duration in Drosophila cells

A defining feature of centromeres is the presence of the histone H3 variant CENP-A that replaces H3 in a subset of centromeric nucleosomes. In Drosophila cultured cells CENP-A deposition at centromeres takes place during the metaphase stage of the cell cycle and strictly depends on the presence of its specific chaperone CAL1. How CENP-A loading is restricted to mitosis is unknown. Overexpression of CAL1 is associated with increased CENP-A levels at centromeres and uncouples CENP-A loading from mitosis. Moreover, CENP-A levels inversely correlate with mitosis duration suggesting crosstalk of CENP-A loading with the regulatory machinery of mitosis. Mitosis length is influenced by the spindle assembly checkpoint (SAC), and this study found that CAL1 interacts with the SAC protein and RZZ complex component Zw10 and thus constitutes the anchor for the recruitment of RZZ. Therefore, CAL1 controls CENP-A incorporation at centromeres both quantitatively and temporally, connecting it to the SAC to ensure mitotic fidelity (Pauleau, 2019).

The formation of two genetically identical daughter cells with a correct and stable genome is of utmost importance during mitosis. Condensed chromosomes are attached and segregated to the opposing poles of the dividing cell at anaphase by the mitotic spindle. At the interface between the chromosomes and the spindle microtubules lies the kinetochore. This multi-protein complex is formed by the components of the KMN network (formed by the Knl1 complex, the Mis12 complex, and the Ndc80 complex) (Joglekar, 2017). The Ndc80 complex is mainly responsible for connecting microtubules with kinetochores while the Knl1 complex primarily coordinates the Spindle Assembly Checkpoint (SAC) (Musacchio, 2017). The SAC delays entry into anaphase until all chromosomes are properly attached and aligned at the metaphase plate. The metaphase to anaphase transition is controlled by the activation of Cdc20 of the APC/C, a multisubunit ubiquitin ligase that triggers the degradation of cell cycle regulators by the proteasome. The SAC proteins Mad2, BubR1, and Bub3 sequester Cdc20 by forming the Mitotic Checkpoint Complex (MCC) thereby preventing the activation of the APC/C. Besides, other proteins have been implicated in SAC activity including Bub1, Mad1, the Mps1 and Aurora B kinases, and the RZZ complex (formed by the three proteins Rough Deal (ROD), Zw10 and Zwilch (Musacchio, 2015). Finally, the Mis12 complex serves as a hub at the kinetochore interacting with all kinetochore complexes as well as with the centromere (Pauleau, 2019).

The kinetochore assembles on the centromere during mitosis, a highly specialized chromatin region that is defined by an enrichment of nucleosomes containing the histone H3 variant CENP-A, also called CID in Drosophila. In contrast to canonical histones, CENP-A deposition at centromeres is independent of DNA replication and is temporally restricted to a specific cell cycle stage, which varies between organisms: late telophase/early G1 in mammalian cultured cells, G2 in S. pombe and plants, and mitosis to G1 in Drosophila. The timing of CENP-A is particularly intriguing in Drosophila cultured cells as centromeric CENP-A is replenished during prometaphase-metaphase thus coinciding with kinetochore assembly. CENP-A loading requires the action of its dedicated chaperones: HJURP in humans, Scm3 in fungi and CAL1 in Drosophila (Chen, 2014, Erhardt, 2008, Schittenhelm, 2010). Deregulation of CENP-A and its loading machinery can result in the misincorporation of CENP-A into regions along the chromosome arms. Misincorporated CENP-A is usually rapidly degraded. If, however, CENP-A-containing nucleosomes remain at non-centromeric sites ectopic formation of functional kinetochores can occur that may lead to chromosome segregation defects and aneuploidy (Pauleau, 2019).

In Drosophila melanogaster, two other proteins are constitutively present at centromeres and essential for centromere function: the conserved protein CENP-C and the CENP-A chaperone CAL1. CENP-C has been shown to act as a linker between CENP-A nucleosomes and the Mis12 complex, therefore, providing a platform for kinetochore assembly (Przewloka, 2011). CENP-C is also implicated in CENP-A replenishment at centromeres during mitosis by recruiting CAL1 (Chen, 2009; Erhardt, 2008). CAL1 interacts with CENP-A in both pre-nucleosomal and nucleosomal complexes and is necessary for CENP-A protein stabilization via Roadkill-Cullin3-mediated mono-ubiquitination. Moreover, CAL1 has been previously shown to be the limiting factor for CENP-A centromeric incorporation in fly embryos. However, differences in centromere assembly have been reported between Drosophila cultured cells and embryos. Firstly, CENP-A loading has been shown to take place during mitotic exit in early embryos and prometaphase to early G1 in cultured cells. Second, CENP-C incorporates concomitantly to CENP-A in embryos while this time window seems to be larger in cultured cells. This study, therefore, set out to determine more precisely the function of CAL1 in CENP-A loading regulation in Drosophila cultured cells (Pauleau, 2019).

During the course of these studies, overexpression of CAL1 was found not only to increase endogenous and exogenous CENP-A abundance at centromeres, but also was found to uncouple CENP-A loading from mitosis. Strikingly, it was discovered a co-dependence of mitotic duration and accurate CENP-A loading that may be coordinated by an interaction of the CENP-A loading machinery with the SAC protein and RZZ subunit Zw10. These data suggest an intricate coordination of the spindle assembly checkpoint, CENP-A loading, and mitotic duration in order to safeguard accurate mitotic progression (Pauleau, 2019).

In Drosophila cells, CENP-A loading takes place primarily during prometaphase-metaphase. Additional turnover of CENP-A in G1 has been reported leading to the hypothesis that CENP-A could be further incorporated at this stage, which was not observe in FRAP experiments when centromeric CENP-A was bleached at the end of cytokinesis. However, the FRAP experiments and most importantly live staining of newly synthesized SNAP-CENP-A confirmed that the majority of CENP-A loading takes place during mitosis in Drosophila cultured cells (Pauleau, 2019).

In flies, CENP-A incorporation is controlled by its chaperone CAL1. It has been shown previously that co-overexpression of exogenous CENP-A and CAL1 leads to an increase of centromeric CENP-A in embryos. This study now shows that overexpression of CAL1 alone leads to increased endogenous CENP-A protein levels in Drosophila cultured cells. Ectopic incorporation of CENP-A, however, was never observed suggesting that CAL1 loads CENP-A exclusively to centromeres and that ectopic CENP-A incorporation in flies depends on alternative loading mechanisms similar to what has been suggested in human cells. Importantly, increased centromeric CENP-A levels following CAL1 overexpression correlated with faster mitosis. A similar acceleration of mitotic timing was observed when CENP-A was only mildly overexpressed, revealing a possible link between CENP-A loading and mitotic timing. Indeed, shortening mitosis duration by depleting Mad2 or BubR1 was associated with decreased CENP-A loading. However, just elongating the mitotic time window during which CENP-A can get loaded (Spindly or Cdc27 depletion, or by drug treatment) did not increase the amount of CENP-A incorporated at centromeres, showing that the length of mitosis alone is insufficient to control CENP-A amounts at centromeres. Rather, these experiments showed that only a defined amount of CENP-A can be incorporated at each mitosis probably correlating with CAL1's availability. Indeed, live analysis of CAL1-overexpressing cells allowed visualization of newly synthesized CENP-A incorporation to centromeres in all stages of the cell cycle. This strongly suggests that CAL1 controls CENP-A incorporation into centromeric chromatin both quantitatively and temporally. How exactly CENP-A levels at centromeres are sensed is unclear but this study identified the RZZ-component Zw10 as a new CAL1 interacting partner, which directly connects CENP-A loading to the SAC (Pauleau, 2019).

It has been proposed that SAC activation is a 2-steps process: at the end of G2-beginning of mitosis, before the kinetochores are assembled, cytosolic Mad1-Mad2 dimers initiate MCC formation inhibiting APC/CCdc20 and determining the timing of mitosis. After nuclear envelope disassembly, kinetochore-dependent MCC are generated and regulated by kinetochore-microtubules attachment. Therefore, the following model is suggested: efficient CENP-A loading by CAL1 during mitosis recruits Zw10 up to a threshold, which is sensed by the SAC. Low CENP-A levels at centromeres could lead to more cytosolic Mad2 thereby keeping the timer active longer. Higher CENP-A levels at centromeres during early mitosis would accelerate the recruitment of RZZ and consequently Mad2 to the kinetochores or capture microtubules more efficiently, therefore, releasing the timer and shortening mitosis duration in cells where kinetochores attach properly to the spindle microtubules. Interestingly, Nocodazole treatment did not affect CENP-A loading confirming previous observations that kinetochore attachment to the microtubule spindle does not play a role in CENP-A loading. These results are pointing further to an additional function of the SAC independent of the control of microtubule attachment. Interestingly, recent evidence shows that RZZ together with Spindly plays a central role in kinetochore expansion during early mitosis to form a fibrous corona that then compacts upon microtubule capture. Whether and -if so- how the kinetochore expansion by RZZ and spindly is involved in CENP-A loading needs to be investigated in the future (Pauleau, 2019).

Many essential components of the SAC require outer kinetochore components for their localization to centromeric regions. However, several outer components are missing from the Drosophila kinetochore and even though Mad1/2 recruitment to the kinetochore depends on the RZZ complex, the factors necessary for the localization of the RZZ to kinetochores are unknown. This study has shown that RZZ localization to the kinetochores does not require KNL1Spc105R but depends on the centromeric proteins CAL1 and CENP-A. Therefore, it is proposed that the Drosophila outer kinetochore and components of the SAC assemble through two independent pathways: the CENP-C-KMN-Bub1-Bub3/BubR1 branch or the CAL1-RZZ-Mad1/2 branch. How those two pathways communicate for the formation of MCC complexes remains to be determined. One link may be the KMN complex since Mad2 is diminished in the absence of KMN proteins. Interestingly, Spc105R mutation does not affect SAC function in fly embryos suggesting that flies rely more on the RZZ-Mad1/2 branch to engage the SAC (Pauleau, 2019).

CENP-A expression and its stability together with its dependence on the low abundant and highly specific loading factor CAL1 and the connection to mitotic events are likely interconnected cellular surveillance mechanisms to avoid misincorporation of CENP-A and, therefore, securing genome stability. How CAL1 itself is regulated to obtain such specificity is currently unknown. In conclusion, this study has shown that there is direct crosstalk between the SAC and the maintenance of centromeric chromatin, ensuring mitotic fidelity not only by controlling microtubule attachment but also by regulating the accurate composition of centromeres (Pauleau, 2019).

Spt6 is a maintenance factor for centromeric CENP-A

Replication and transcription of genomic DNA requires partial disassembly of nucleosomes to allow progression of polymerases. This presents both an opportunity to remodel the underlying chromatin and a danger of losing epigenetic information. Centromeric transcription is required for stable incorporation of the centromere-specific histone dCENP-A in M/G1 phase, which depends on the eviction of previously deposited H3/H3.3-placeholder nucleosomes. This study demonstrates that the histone chaperone and transcription elongation factor Spt6 spatially and temporarily coincides with centromeric transcription and prevents the loss of old CENP-A nucleosomes in both Drosophila and human cells. Spt6 binds directly to dCENP-A and dCENP-A mutants carrying phosphomimetic residues alleviate this association. Retention of phosphomimetic dCENP-A mutants is reduced relative to wildtype, while non-phosphorylatable dCENP-A retention is increased and accumulates at the centromere. It is concluded that Spt6 acts as a conserved CENP-A maintenance factor that ensures long-term stability of epigenetic centromere identity during transcription-mediated chromatin remodeling (Bobkov, 2020).

The CENP-A nucleosome is considered to be the key epigenetic mark for centromere identity in most organisms. Accordingly, CENP-A and epigenetic marks in general should meet three requirements: (1) Template its own deposition, (2) be replenished in a cell cycle-controlled manner to counteract dilution by half in each S-phase and (3) be stably transmitted to the next cell generation (Bobkov, 2020).

New dCENP-A can be targeted to sites of previous CENP-A deposition by its chaperone CAL1, which is recruited to centromeres by dCENP-C. Loading of new CENP-A is restricted to mitosis and G1 and serves primarily to replenish CENP-A containing nucleosomes that became diluted by half during the preceding S-phase. During DNA replication the MCM2-7 replicative helicase along with other histone chaperones like HJURP, are instrumental for the stable transmission of parental CENP-A during S-phase (Bobkov, 2020).

Recent work has shown in Drosophila S2 cells that transcription at the centromere is required for stable nucleosome incorporation of new dCENP-A9. This finding could be explained by a model in which transcription-mediated chromatin remodeling is re-purposed to evict placeholder H3 nucleosomes to make room for deposition of new dCENP-A. However, the induction of nucleosome eviction during CENP-A loading also bears the danger of losing previously incorporated CENP-A. This study reports the identification of the transcription elongation factor and histone chaperone Spt6 as a new CENP-A maintenance factor, which safeguards previously deposited CENP-A during centromeric transcription (see Model showing histone dynamics during dCENP-A loading) (Bobkov, 2020).

Drosophila Spt6 localizes to centromeres during mitosis and G1, coinciding with the time window when transcription and dCENP-A loading occurs. The SH2 domain enables Spt6 to interact directly with RNAPII and it is therefore likely that recruitment of Spt6 to centromeres is a direct consequence of centromeric transcription. Because Spt6 prevents transcription-coupled loss of posttranslationally modified nucleosomes in gene bodies, whether Spt6 might act to maintain dCENP-A at the centromere was tested. Indeed, when Spt6 was depleted in Drosophila or human cells, the specific loss of old CENP-A was observed after passage through mitosis into G1 phase. This observation suggests that ongoing transcription evicts nucleosomes at centromeres and that Spt6 serves a conserved role to recycle CENP-A/H4 tetramers expelled by closely spaced polymerase complexes. A key point of this model is the transcription-mediated creation of nucleosomal gaps as a prerequisite for full incorporation of new dCENP-A. Consequently, the additional loss of nucleosomes in Spt6-depleted cells should create more opportunities to load new dCENP-A. Indeed, when an experimental system was used that provides elevated levels of transgenic, ready-made dCENP-A (TI-dCENP-AHA), a clear increase was observed in loading. This is further supported by the fact that the loss of total centromeric dCENP-A in Spt6-depleted cells is completely compensated under these conditions (Bobkov, 2020).

It is currently unknown if the mitotic defects observed upon Spt6 depletion by RNAi are a direct or indirect consequence of Spt6 removal. As cells can tolerate very low CENP-A levels at the centromere down to 10%, the 2-day depletion of Spt6 likely leaves sufficient dCENP-A for centromere function. Despite this, chromosome segregation might be compromised due to the specific loss of old nucleosomes with specific PTMs. PTMs relevant for centromere function have been identified on CENP-A and shown to affect CENP-A stability and correct mitotic progression. Moreover, methylation of lysine 20 on the associated H4 plays an essential role for kinetochore formation. Likewise, in addition to CENP-A nucleosomes, centromeres contain canonical H3 nucleosomes with a specific set of posttranslational modifications that might need to be retained. It is therefore postulated that Spt6 should be able to distinguish between placeholder nucleosomes that need to be removed and epigenetically marked nucleosomes that should be kept. As previously demonstrated for H3/H4 in budding yeast39, direct binding of a bacterially expressed N-terminal fragment of Spt6 (199-338) was observed with both H3/H4 and dCENP-AΔNT/H4 tetramers. In addition, full length dCENP-AFLAG and H3 co-IP with endogenous Spt6 from S2 cell extracts with comparable efficiency (Bobkov, 2020).

Interestingly, CENP-A is phosphorylated in various organisms including flies and humans and phosphorylation events have been linked to transcription-induced loss of centromeric CENP-A nucleosomes in mouse cells. To test whether phosphorylation of dCENP-A affects its maintenance, three previously identified phosphorylation sites were mutated in the N-terminal tail of dCENP-A (S20, S75 and S7757). Indeed, it was found that dCENP-A mutants carrying the phosphomimetic residue aspartate showed significantly reduced binding to Spt6, while the opposite was observed for the respective non-phosphorylatable alanine mutants. Furthermore, wild-type or non-phosphorylatable mutants of dCENP-A bound robustly to Spt6 when exposed to high salt washes while canonical H3 binding was abolished. This difference hints toward a mechanism how Spt6 distinguishes between the two histone H3-variants and allows selective retention of CENP-A, while placeholder nucleosomes are exchanged. Consistent with the observations described above, a pulse-chase experiment to follow the decline of old dCENP-A during cell division showed higher loss rate for the phosphomimetic dCENP-A construct. Interestingly, dCENP-A wild-type displayed less than perfect inheritance after two cell cycles (<25% expected for replicative dilution). In contrast S77A was on average more and S77D less stable than wild type, likely accounting for the accumulation of the non-phosphorylatable dCENP-A mutant at centromeres over time (Bobkov, 2020).

Taken together, it is proposed that the transcription-mediated eviction of centromeric nucleosomes affects both placeholder H3 and previously deposited CENP-A nucleosomes. However, loss of the centromeric mark is prevented by specific recycling of CENP-A through Spt6, potentially involving phospho-regulation of the CENP-A/Spt6 interaction. It is concluded that Spt6 acts as an important CENP-A maintenance factor and contributes to the long-term stability of the epigenetic centromere mark (Bobkov, 2020).

PROTEIN STRUCTURE

Amino Acids - 225

Structural Domains

Mammalia CENP-A and yeast Cse4p share with histone H3 the ~100-aa nucleosomal core, but have N-terminal tails that are completely dissimilar from one another and from H3. In CENP-A, centromeric localization maps to the core, which is about equally divergent from both Cse4p and H3. Except for CENP-A and Cse4p, only three GenBank entries were found to have comparably diverged H3-like cores and dissimilar N-terminal tails. A 225-aa ORF that fits this profile has been found within a 'sequencing in progress' database entry (AC005652) deposited by the Berkeley Drosophila Genome Project. The possibility was considered that this ORF encodes a centromeric protein. An antiserum against a peptide predicted from the ORF was used for immunocytochemistry to D. melanogaster Kc cells. Intense, point-like signals were observed over interphase nuclei and exclusively over centromeric constrictions of all Drosophila mitotic chromosomes in both Kc tissue culture cells and larval neuroblasts. Similar results were obtained with antibody raised against a peptide predicted from another N-terminal tail region of the ORF. Therefore, these epitopes are present on a normal centromere component that is present throughout the cell cycle. Because the 225-aa protein behaves as a Drosophila homolog of CENP-A and Cse4p, the gene was named cid (for centromere identifier) (Henikoff, 2000).

date revised: 25 September 2023

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.