Mutations in the Drosophila condensin subunit dCAP-G: defining the role of condensin for chromosome condensation in mitosis and gene expression in interphase

Chromosomes are dynamic structures that are reorganized during the cell cycle to optimize them for distinct functions. SMC and non-SMC condensin proteins associate into complexes that have been implicated in the process of chromosome condensation. The roles of the individual non-SMC subunits of the complex are poorly understood, and mutations in the CAP-G subunit have not been described in metazoans. Here we elucidate a role for dCAP-G in chromosome condensation and cohesion in Drosophila. We illustrate the requirement of dCAP-G for condensation during prophase and prometaphase; however, we find that alternate mechanisms ensure that replicated chromosomes are condensed prior to metaphase. In contrast, dCAP-G is essential for chromosome condensation in metaphase of single, unreplicated sister chromatids, suggesting that there is an interplay between replicated chromatids and the condensin complex. In the dcap-g mutants, defects in sister-chromatid separation are also observed. Chromatid arms fail to resolve in prophase and are unable to separate at anaphase, whereas sister centromeres show aberrant separation in metaphase and successfully move to spindle poles at anaphase. We also identified a role for dCAP-G during interphase in regulating heterochromatic gene expression.     

The aurora B kinase AIR-2 regulates kinetochores during mitosis and is required for separation of homologous Chromosomes during meiosis

BACKGROUND: Mitotic chromosome segregation depends on bi-orientation and capture of sister kinetochores by microtubules emanating from opposite spindle poles and the near synchronous loss of sister chromatid cohesion. During meiosis I, in contrast, sister kinetochores orient to the same pole, and homologous kinetochores are captured by microtubules emanating from opposite spindle poles. Additionally, mechanisms exist that prevent complete loss of cohesion during meiosis I. These features ensure that homologs separate during meiosis I and sister chromatids remain together until meiosis II. The mechanisms responsible for orienting kinetochores in mitosis and for causing asynchronous loss of cohesion during meiosis are not well understood. RESULTS: During mitosis in C. elegans, aurora B kinase, AIR-2, is not required for sister chromatid separation, but it is required for chromosome segregation. Condensin recruitment during metaphase requires AIR-2; however, condensin functions during prometaphase, independent of AIR-2. During metaphase, AIR-2 promotes chromosome congression to the metaphase plate, perhaps by inhibiting attachment of chromatids to both spindle poles. During meiosis in AIR-2-depleted oocytes, congression of bivalents appears normal, but segregation fails. Localization of AIR-2 on meiotic bivalents suggests this kinase promotes separation of homologs by promoting the loss of cohesion distal to the single chiasma. Inactivation of the phosphatase that antagonizes AIR-2 causes premature separation of chromatids during meiosis I, in a separase-dependent reaction. CONCLUSIONS: Aurora B functions to resolve chiasmata during meiosis I and to regulate kinetochore function during mitosis. Condensin mediates chromosome condensation during prophase, and condensin-independent pathways contribute to chromosome condensation during metaphase.     

A postprophase topoisomerase II-dependent chromatid core separation step in the formation of metaphase chromosomes

Metaphase chromatids are believed to consist of loops of chromatin anchored to a central scaffold, of which a major component is the decatenatory enzyme DNA topoisomerase II. Silver impregnation selectively stains an axial element of metaphase and anaphase chromatids; but we find that in earlier stages of mitosis, silver staining reveals an initially single, folded midline structure, which separates at prometaphase to form two chromatid axes. Inhibition of topoisomerase II prevents this separation, and also prevents the contraction of chromatids that occurs when metaphase is arrested. Immunolocalization of topoisomerase II alpha reveals chromatid cores analogous to those seen with silver staining. We conclude that the chromatid cores in early mitosis form a single structure, constrained by DNA catenations, which must separate before metaphase chromatids can be resolved.     

DNA catenations that link sister chromatids until the onset of anaphase are maintained by a checkpoint mechanism

Treatment of Allium cepa meristematic cells in metaphase with the topoisomerase II inhibitor ICRF-193, results in bridging of the sister chromatids at anaphase. Separation of the sisters in experimentally generated acentric chromosomal fragments was also inhibited by ICRF-193, indicating that some non-centromeric catenations also persist in metaphase chromosomes. Thus, catenations must be resolved by DNA topoisomerase II at the metaphase-to-anaphase transition to allow segregation of sisters. A passive mechanism could maintain catenations holding sisters until the onset of anaphase. At this point the opposite tension exerted on sister chromatids could render the decatenation reaction physically more favorable than catenation. But this possibility was dismissed as acentric chromosome fragments were able to separate their sister chromatids at anaphase. A timing mechanism (a common trigger for two processes taking different times to be completed) could passively couple the resolution of the last remaining catenations to the moment of anaphase onset. This possibility was also discarded as cells arrested in metaphase with microtubule-destabilising drugs still displayed anaphase bridges when released in the presence of ICRF-193. It is possible that a checkpoint mechanism prevents the release of the last catenations linking sisters until the onset of anaphase. To test whether cells are competent to fully resolve catenations before anaphase onset, we generated multinucleate plant cells. In this system, the nuclei within a single multinucleate cell displayed differences in chromosome condensation at metaphase, but initiated anaphase synchronously. When multinucleates were treated with ICRF-193 at the metaphase-toanaphase transition, tangled and untangled anaphases were observed within the same cell. This can only occur if cells are competent to disentangle sister chromatids before the onset of anaphase, but are prevented from doing so by a checkpoint mechanism.     

Regulation of sister chromatid cohesion between chromosome arms

Sister chromatid separation in anaphase depends on the removal of cohesin complexes from chromosomes. In vertebrates, the bulk of cohesin is already removed from chromosome arms during prophase and prometaphase, whereas cohesin remains at centromeres until metaphase, when cohesin is cleaved by the protease separase. In unperturbed mitoses, arm cohesion nevertheless persists throughout metaphase and is principally sufficient to maintain sister chromatid cohesion. How arm cohesion is maintained until metaphase is unknown. Here we show that small amounts of cohesin can be detected in the interchromatid region of metaphase chromosome arms. If prometaphase is prolonged by treatment of cells with microtubule poisons, these cohesin complexes dissociate from chromosome arms, and arm cohesion is dissolved. If cohesin dissociation in prometaphase-arrested cells is prevented by depletion of Plk1 or inhibition of Aurora B, arm cohesion is maintained. These observations imply that, in unperturbed mitoses, small amounts of cohesin maintain arm cohesion until metaphase. When cells lacking Plk1 and Aurora B activity enter anaphase, chromatids lose cohesin. This loss is prevented by proteasome inhibitors, implying that it depends on separase activation. Separase may therefore be able to cleave cohesin at centromeres and on chromosome arms.     

Laser microirradiation of kinetochores in mitotic PtK2 cells: chromatid separation and micronucleus formation

Irradiation of the kinetochore region of PtK2 chromosomes by laser light of 532 nm was used to study the function of the kinetochore region in chromosome movement and to create an artificial micronuclei in cells. When the sister kinetochores of a chromosome were irradiated at prometaphase, the affected chromosome detached from the spindle and exhibited no further directed movements for the duration of mitosis. The chromatids of the chromosome remained attached to one another until anaphase, at which point they separated. No poleward movement of the chromatids was observed, and at telophase they passively moved to one of the daughter cells and were enclosed in a micronucleus. The daughter cell containing the micronucleus was then isolated by micromanipulation and followed through subsequent mitoses. At the next mitosis, two chromosomes, each with two chromatids, condensed in the micronucleus. These chromosomes did not attach to the spindle and showed chromatid separation, but no poleward movements at anaphase. They were again enclosed in micronuclei at telophase. The third generation mitosis was similar to the second. Occasionally, both the irradiation-produced and naturally occurring micronuclei exhibited no chromosome condensation at mitosis. Feulgen-stained monolayers of PtK2 cells with naturally occurring micronuclei showed that some micronuclei stain positive for DNA and others do not. This finding raises questions about the fate of chromosomes in a micronucleus.     

A chromosome separation checkpoint

Here we discuss a “chromosome separation checkpoint” that might regulate the anaphase‐telophase transition. The concept of cell cycle checkpoints was originally proposed to account for extrinsic control mechanisms that ensure the order of cell cycle events. Several checkpoints have been shown to regulate major cell cycle transitions, namely at G1‐S and G2‐M. At the onset of mitosis, the prophase‐prometaphase transition is controlled by several potential checkpoints, including the antephase checkpoint, while the spindle assembly checkpoint guards the metaphase‐anaphase transition. Our hypothesis is based on the recently uncovered feedback control mechanism that delays chromosome decondensation and nuclear envelope reassembly until effective separation of sister chromatids during anaphase is achieved. A central player in this potential checkpoint is the establishment of a constitutive, midzone‐based Aurora B phosphorylation gradient that monitors the position of chromosomes along the spindle axis. We propose that this surveillance mechanism represents an additional step towards ensuring mitotic fidelity.     

Chromosome behavior after laser microirradiation of a single kinetochore in mitotic PtK2 cells

The role of the kinetochore in chromosome movement was studied by 532- nm wavelength laser microirradiation of mitotic PtK2 cells. When the kinetochore of a single chromatid is irradiated at mitotic prometaphase or metaphase, the whole chromosome moves towards the pole to which the unirradiated kinetochore is oriented, while the remaining chromosomes congregate on the metaphase plate. The chromatids of the irradiated chromosome remain attached to one another until anaphase, at which time they separate by a distance of 1 or 2 micrometers and remain parallel to each other, not undergoing any poleward separation. Electron microscopy shows that irradiated chromatids exhibit either no recognizable kinetochore structure or a typical inactive kinetochore in which the tri-layer structure is present but has no microtubules associated with it. Graphical analysis of the movement of the irradiated chromosome shows that the chromosome moves to the pole rapidly with a velocity of approximately 3 micrometers/min. If the chromosome is close to one pole at irradiation, and the kinetochore oriented towards that pole is irradiated, the chromosome moves across the spindle to the opposite pole. The chromosome is slowed down as it traverses the equatorial region, but the velocity in both half-spindles is approximately the same as the anaphase velocity of a single chromatid. Thus a single kinetochore moves twice the normal mass of chromatin (two chromatids) at the same velocity with which it moves a single chromatid, showing that the velocity with which a kinetochore moves is independent, within limits, of the mass associated with it.     

Human TopBP1 localization to the mitotic centrosome mediates mitotic progression

TopBP1 contains repeats of the BRCA1 C-terminal (BRCT) domain and plays important roles in DNA damage response, DNA replication, and other cellular regulatory functions during the interphase. In prometaphase, metaphase, and anaphase, TopBP1 localizes to the mitotic centrosomes, which function as spindle-poles for the bipolar separation of sister chromatids. The localization of TopBP1 to the mitotic centrosomes is mediated by amino acid residues 1259 to 1420 in the TopBP1 C-terminal region (TbpCtr). GST and DsRed2 tags fused to TbpCtr were localized in the mitotic centrosomes, thereby suggesting that TbpCtr functions as a mitosis-specific centrosome localization signal (CLS). Mutations of Ser 1273 and/or Lys 1317, which were predicted to interact with a putative phosphoprotein, inhibited CLS function. Ectopic expression of TbpCtr specifically eliminated endogenous TopBP1 from the mitotic centrosomes, whereas mutant TbpCtr derivatives, containing substitutions at Ser 1273 and/or Lys 1317, did not. The specific elimination of TopBP1 from the mitotic centrosomes prolonged the durations of prometaphase and metaphase and shortened the inter-kinetochore distances of metaphase sister chromatids while maintaining the spindle assembly checkpoint. These results suggest that the localization of TopBP1 to the mitotic centrosomes is necessary for proper mitotic progression.     

Scanning electron microscopy of mammalian chromosomes from prophase to telophase

Changes in the morphology of human and murine chromosomes during the different stages of mitosis have been examined by scanning electron microscopy. Two important findings have emerged from this study. The first is that prophase chromosomes do not become split into pairs of chromatids until late prophase or early metaphase. This entails two distinct processes of condensation, the earlier one starting as condensations of chromosomes into chromomeres which then fuse to form a cylindrical body. After this cylindrical body has split in two longitudinally, further condensation occurs by mechanisms that probably include coiling of the chromatids as well as other processes. The second finding is that the centromeric heterochromatin does not split in two at the same time as the rest of the chromosome, but remains undivided until anaphase. It is proposed that the function of centromeric heterochromatin is to hold the chromatids together until anaphase, when they are separated by the concerted action of topoisomerase II acting on numerous similar sites provided by the repetitive nature of the satellite DNA in the heterochromatin. A lower limit to the size of blocks of centromeric heterochromatin is placed by the need for adequate mechanical strength to hold the chromatids together, and a higher limit by the necessity for rapid splitting of the heterochromatin at anaphase. Beyond these limits malsegregation will occur, leading to aneuploidy. Because the centromere remains undivided until anaphase, it cannot undergo the later stage of condensation found in the chromosome arms after separation into chromatids, and therefore the centromere remains as a constriction.by U. Scheer     

Laser microirradiation of kinetochores in mitotic PtK2 cells

Irradiation of the kinetochore region of PtK₂ chromosomes by laser light of 532 nm was used to study the function of the kinetochore region in chromosome movement and to create artificial micronuclei in cells. When the sister kinetochores of a chromosome were irradiated at prometaphase, the affected chromosome detached from the spindle and exhibited no further directed movements for the duration of mitosis. The chromatids of the chromosome remained attached to one another until anaphase, at which point they separated. No poleward movement of the chromatids was observed, and at telophase they passively moved to one of the daughter cells and were enclosed in a micronucleus. The daughter cell containing the micronucleus was then isolated by micromanipulation and followed through subsequent mitoses. At the next mitosis, two chromosomes, each with two chromatids, condensed in the micronucleus. These chromosomes did not attach to the spindle and showed chromatid separation, but no poleward movements at anaphase. They were again enclosed in micronuclei at telophase. The third generation mitosis was similar to the second. Occasionally, both the irradiation-produced and naturally occurring micronuclei exhibited no chromosome condensation at mitosis. Feulgenstained monolayers of PtK₂ cells with naturally occurring micronuclei showed that some micronuclei stain positive for DNA and others do not. This finding raises questions about the fate of chromosomes in a micronucleus.     

Early chromosome condensation without cell fusion. I. Preliminary results with allogeneic and xenogeneic cells.

Early chromatin condensation in interphase cells (G1) of human peripheral blood lymphocytes has been induced without virus or cell fusion by exposure to allogeneic or xenogeneic mitotic cells. The event, although similar in some ways to the phenomenon described as "premature chromosome condensation," "chromosome pulverization," and "prophasing," differs in that it does not require the presence of viruses and cell fusion before mitosis proceeds in the G1 cell. Early chromatin condensation in interphase cells induced by mitotic cells only, consists of chromatids in the early or late G1 phase of the cell cycle that are not pulverized or fragmented at mitosis. Some of the chromosomes are twice as long as the metaphase chromosomes and exhibit natural bands. Almost twice as many of these bands are produced as by trypsin treatment of metaphase chromosomes. The nuclear membrane is intact and nucleoli are present, to which some chromosomes are attached. The DNA content of the precocious chromosomes in G1 is half the amount of the metaphase complement.     

Localization of anti-topoisomerase IIα antibodies under normal conditions and after artificial chromosome condensation and decondensation

By means of immunofluorescence method, localization of DNA-topoisomerase IIα (Topo IIα) in interphase nuclei and chromosomes at different stages of mitosis was studied in situ under normal conditions and after treatment with condensing and decondensing solutions. In non-isolated mitotic M-HeLa cell chromosomes, Topo IIα was uniformly distributed along chromatids after fixation and permeabilization in situ. After treatment of cells with decondensing solutions (10 mM Tris; 0.1 mM CaCl₂ in 10 mM Tris; 0.3 mM CaCl₂ in 10 mM Tris; 15% DMEM; 75 mM KCl), Topo IIα was evenly distributed along chromatids in prophase, prometaphase and metaphase; its concentration was the highest in the pericentromere region. After treatment of cells with condensing solutions containing 0.7 mM, 1 mM, 2 mM or 3 mM CaCl₂ in 10 mM Tris, Topo IIα was not detected in prophase, metaphase and anaphase. However, in late telophase anti-Topo IIα antibodies were found in reforming nuclei under identical conditions. After sequential treatment with condensing and decondensing solutions, the distribution patterns of Topo IIα in chromosomes were the same as after treatment with only decondensing solutions. In anaphase and telophase, Topo IIα was evenly distributed along chromatids, while in prophase, prometaphase and metaphase it was predominantly localized in the pericentromere region. After the treatment of cells with condensing solutions chromosome staining was not observed, apparently due to “masking” of binding sites for anti-Topo IIα antibodies. Homogenous distribution of Topo IIα along chromatids in non-isolated chromosomes was preserved after the treatment of cells with hypotonic solutions; however, under these conditions Topo IIα concentration was higher in centromeres.     

Heteropycnosis of an underreplicating chromosome

Drosophila nasutoides has an extraordinary genome since 62% of its DNA resides in chromosome4. This element mainly consists of constitutive heterochromatin which does not polytenize. Earlier studies of heterochromatin attributed little attention to the fact that condensed chromosomes often vary in condensation. This paper reports that chromosomes of the same complement display different degrees and kinetics of condensation. InD. nasutoides, even sex specific differences can be observed. The results of a comparative microphotometric study on neuroblast metaphases in both sexes revealed the following picture. The process of chromosome condensation is not restricted to mitotic prophase but continues into the metaphase. The mean condensation is not equal for all chromosomes. In the metaphase of the female, Feulgen density increases from theX chromosome, via3 and2, to chromosome4. In the male, the order isX, 2, 3, Y, and4. During the metaphase of the male, chromosomes condense with similar kinetics. In contrast, chromosomes of the female display asynchrony as monitored by area and length determinations. TheX chromosomes of the female probably have enhanced shortening during prophase. This would explain the metaphase of the female where theX chromosomes shorten less than the autosomes, and why each of theX chromosomes is 15% shorter than theX chromosome in the metaphase of the male. Further differences were observed in the longitudinal and lateral compaction of the chromosomes in males and females. The sex chromosomes and chromosome3 condense by shortening, while chromosomes2 and4 preferentially reduce their diameter. The large amount of DNA engaged in heteropycnosis and the isochromosome nature allow the identification of chromosome4 during interphase. At this stage, a new category of extreme DNA packaging was detected. The interphase density of chromosome4 can exceed that of metaphase by a factor of up to 8. Two events account for this high degree of condensation:(1) the homologues are particularly associated due to somatic pairing and (2) the arms are further tightened as a result of pericentric folding. The features of the isochromosome suggest that the interaction of chromatids during interphase is essentially caused by specific DNA sequences. The data confirm that heteropycnosis not only interferes with gene expression but also strongly inhibits DNA synthesis in endocycles.     

A complex rearrangement involving simultaneous translocation and inversion is associated with a change in chromatin compaction

Detailed fluorescence in situ hybridisation analysis of a previously described translocation revealed it to be a more complex rearrangement consisting of both a translocation and a paracentric inversion with an apparent coincident breakpoint at 16p13.3, t(14;16)(p32;p13.3) inv16(p13.3p12.1). This unusual three-breakpoint rearrangement was not obvious from examination of G-banding. Such rearrangements may be undiagnosed in cytogenetic studies. The presence of an interstitial deletion of 16p was unlikely as the rearranged chromosome contained probes distributed along the short arm of chromosome 16. Fluorescence in situ hybridisation studies suggested that the inverted segment was smaller in size than that on the normal chromosome. Measurements of distances between probes on metaphase chromosomes confirmed that there was differential compaction of the inverted portion on 16p. The inverted region was significantly reduced in size by 21% compared with the same region on the normal chromosome 16. The size reduction across the region was non-uniform, with one region showing a 55% increase in compaction. The change in compaction was also associated with a change in the lateral position of a probe on the chromatids. The finding that a single chromosome breakpoint can change the compaction of chromatin over an extensive region has implications for models of the structure of metaphase chromosomes. Possible explanations are either a localized severe disruption of DNA packaging over relatively short distances (hundreds of kilobases) or a more generalized change that extends over many megabases. These results raise the important possibility that chromosome breaks may result in a more global change in DNA compaction across large segments of a chromosome.     

A new chromosome model

We present a new model of the three-dimensional structure of chromosomes. With DNA and protein staining it could be shown by high-resolution scanning electron microscopy that metaphase chromosomes are mainly composed of DNA packed in "chromomeres" (coiled solenoides) and a dynamic matrix formed of parallel protein fibers. In the centromeric region, the chromomeres are less densely packed, giving insight into the matrix fibers. We postulate that chromosome condensation is achieved by the binding of solenoids to matrix fibers which have contact sites to one another and move antiparallel to each other. As condensation progresses, loops of solenoids accumulate to form additional chromomeres, causing chromosomes to become successively shorter and thicker as more chromomeres are formed. For sterical reasons, a tension vertical to the axial direction forces the chromatids apart. The model can simply explain the enormous variety of chromosome morphology in plant and animal systems by varying only a few cytological parameters. Primary and secondary constrictions and deletions are defined as regions devoid of chromomeres. Even in the highly condensed metaphase, all genes would be easily accessible.     

人染色体脆性位点部位的显微光谱学研究 Study of Microspectroscopy for the Position of the Fragile Sites in Human Chromosome

采用显微分光光度法,对染色体脆性位点的部位进行了显微光谱学研究。实验证明,带有脆点的染色体其DNA含量大多数趋向减少,少数略有增加,推测染色体脆性部位的产生是由于染色质DNA在高度凝缩形成中期染色体过程中超旋结构改变的结果。 The position of fragile sites in human chromosome was studied by means of the microspectroscopy. The results show that the amount DNA in chromosome with fragile sites decreases in most condition. We can suppose that the fragile sites of chromosome is caused by the superhelix structure changes of chromosome DNA during the formation of metaphase chromosome which is formed in high condensation.     

A role for Drosophila SMC4 in the resolution of sister chromatids in mitosis

BACKGROUND: Faithful segregation of the genome during mitosis requires interphase chromatin to be condensed into well-defined chromosomes. Chromosome condensation involves a multiprotein complex known as condensin that associates with chromatin early in prophase. Until now, genetic analysis of SMC subunits of the condensin complex in higher eukaryotic cells has not been performed, and consequently the detailed contribution of different subunits to the formation of mitotic chromosome morphology is poorly understood. RESULTS: We show that the SMC4 subunit of condensin is encoded by the essential gluon locus in Drosophila. DmSMC4 contains all the conserved domains present in other members of the structural-maintenance-of-chromosomes protein family. DmSMC4 is both nuclear and cytoplasmic during interphase, concentrates on chromatin during prophase, and localizes to the axial chromosome core at metaphase and anaphase. During decondensation in telophase, most of the DmSMC4 leaves the chromosomes. An examination of gluon mutations indicates that SMC4 is required for chromosome condensation and segregation during different developmental stages. A detailed analysis of mitotic chromosome structure in mutant cells indicates that although the longitudinal axis can be shortened normally, sister chromatid resolution is strikingly disrupted. This phenotype then leads to severe chromosome segregation defects, chromosome breakage, and apoptosis. CONCLUSIONS: Our results demonstrate that SMC4 is critically important for the resolution of sister chromatids during mitosis prior to anaphase onset.     

Kiss and break up—a safe passage to anaphase in mitosis and meiosis

Eukaryotic chromosomes have many challenges to overcome between DNA replication and sister chromatid segregation. If these challenges are not met, cell death or unregulated cell division (cancer) may result. During prophase, chromosomes condense, the nuclear membrane breaks down and cohesins are removed from chromosome arms. In prometaphase, initial spindle attachments are made by sister kinetochores followed by correction of erroneous attachments, centromere oscillation between spindle poles and congression towards the cell's equator. In metaphase, all chromosomes attain stable bipolar spindle attachments and align at the metaphase plate, ready for the metaphase–anaphase transition when all ties between sister chromatids are broken. This review concentrates on recent developments that have revealed the intricacies of these processes. We now know more about how the mechanisms of cohesin removal differ between prophase and the metaphase–anaphase transition, the processes for detection and correction of improper spindle-kinetochore attachments and the concept that tension between sister kinetochores is the driving factor for satisfying the spindle checkpoint. We are also beginning to gain some understanding of the mechanisms behind the co-segregation of sister chromatids at the first meiotic division. Review related to the 15th International Chromosome Conference (ICC XV), held in September 2004, Brunel University, London, UK     

Regulation of Centromeric Cohesion by Sororin Independently of the APC/C

Regulated separation of sister chromatids is the key event of mitosis. Sister chromatids remain cohered from the moment of DNA duplication until anaphase. Two known factors account for cohesion: DNA catenations and cohesin complexes. Premature loss of centromeric cohesion is prevented by the spindle checkpoint. Here we show that sororin, a protein implicated in promoting cohesion through effects on cohesin complexes, is involved in maintenance of cohesion in response to the spindle checkpoint. Sororin-depleted cells reach prometaphase with cohered sister chromatids and are able to form metaphase plates. However, loss of cohesion in anaphase is asynchronous and cells are unresponsive to the spindle checkpoint, accumulating with separated sisters scattered throughout the cytoplasm. These phenotypes are similar to those seen after Shugoshin depletion, suggesting that sororin and Shugoshin might act in concert. Furthermore, sororin-depleted and Shugoshin-depleted cells lose cohesion independently of the APC/C. Therefore, sororin and Shugoshin protect centromeric cohesion in response to the spindle checkpoint, but prevent the removal of cohesion by a mechanism independent of the APC/C.