STAG3, a novel gene encoding a protein involved in meiotic chromosome pairing and location of STAG3-related genes flanking the Williams-Beuren syndrome deletion

Mammalian STAG3 is a cohesin specific to sister chromatid arms in meiosis I

Cohesins, which have been characterized in budding yeast and Xenopus, are multisubunit protein complexes involved in sister chromatid cohesion. Regulation of the interactions among different cohesin subunits and the assembly/disassembly of the cohesin complex to chromatin are key steps in chromosome segregation. We previously characterized the mammalian STAG3 protein as a component of the synaptonemal complex that is specifically expressed in germinal cells, although its function in meiosis remains unknown. Here we show that STAG3 has a role in sister chromatid arm cohesion during mammalian meiosis I. Immunofluorescence results in prophase I cells suggest that STAG3 is a component of the axial/lateral element of the synaptonemal complex. In metaphase I, STAG3 is located at the interchromatid domain and is absent from the chiasma region. In late anaphase I and the later stages of meiosis, STAG3 is not detected. STAG3 interacts with the structural maintenance chromosome proteins SMC1 and SMC3, which have been reported to be subunits of the mitotic cohesin complex. We propose that STAG3 is a sister chromatid arm cohesin that is specific to mammalian meiosis I.

The cohesin complex: sequence homologies, interaction networks and shared motifs

BACKGROUND

Cohesin is a macromolecular complex that links sister chromatids together at the metaphase plate during mitosis. The links are formed during DNA replication and destroyed during the metaphase-to-anaphase transition. In budding yeast, the 14S cohesin complex comprises at least two classes of SMC (structural maintenance of chromosomes) proteins - Smc1 and Smc3 - and two SCC (sister-chromatid cohesion) proteins - Scc1 and Scc3. The exact function of these proteins is unknown.

RESULTS

Searches of protein sequence databases have revealed new homologs of cohesin proteins. In mouse, Mmip1 (Mad member interacting protein 1) and Smc3 share 99% sequence identity and are products of the same gene. A phylogenetic tree of SMC homologs reveals five families: Smc1, Smc2, Smc3, Smc4 and an ancestral family that includes the sequences from the Archaea and Eubacteria. This ancestral family also includes sequences from eukaryotes. A cohesion interaction network, comprising 17 proteins, has been constructed using two proteomic databases. Genes encoding six proteins in the cohesion network share a common upstream region that includes the MluI cell-cycle box (MCB) element. Pairs of the proteins in this network share common sequence motifs that could represent common structural features such as binding sites. Scc2 shares a motif with Chk1 (kinase checkpoint protein), that comprises part of the serine/threonine protein kinase motif, including the active-site residue.

CONCLUSIONS

We have combined genomic and proteomic data into a comprehensive network of information to reach a better understanding of the function of the cohesin complex. We have identified new SMC homologs, created a new SMC phylogeny and identified shared DNA and protein motifs. The potential for Scc2 to function as a kinase - a hypothesis that needs to be verified experimentally - could provide further evidence for the regulation of sister-chromatid cohesion by phosphorylation mechanisms, which are currently poorly understood.

Together until separin do us part

Loss of sister-chromatid cohesion triggers chromosome segregation. Several recent reports show that the protease Esp1 cleaves the cohesin subunit Scc1/Mcd1 to induce sister-chromatid segregation in yeast and vertebrates. This finding indicates that cohesin cleavage may control sister-chromatid separation in all vertebrates.

The structural maintenance of chromosomes (SMC) family of proteins in mammals

Structural maintenance of chromosomes (SMC) family proteins play critical roles in chromosome structural changes. SMC proteins are known to be involved in two major chromosome structural organization events required for mitotic segregation of chromosomes: mitotic chromosome condensation and sister chromatid cohesion. In eukaryotes, two separate sets of SMC heterodimers form the cores of two distinct multiprotein complexes termed 'condensin' and 'cohesin', each specialized for condensation or cohesion, respectively. It is clear that both condensin and cohesin are conserved in mammals, including humans. The mammalian complexes demonstrate dynamic changes in intracellular distribution in a cell cycle-dependent manner. At any point in the cell cycle, the intracellular localization of the majority of mammalian cohesin and condensin appears to be complementary. Cohesin is associated with chromatin in interphase, while condensin is largely cytoplasmic. Similarly, in mitosis, cohesin is mostly excluded from chromosomes while condensin is distinctly bound to them. Cell cycle-dependent targeting of the two complexes appears to play a major role in regulating their cell cycle-specific activities, and how this redistribution is controlled is an area of active research. Finally, there is evidence that SMC proteins may be involved in DNA recombination and repair. This review focuses on what we have learned about SMC family proteins in humans and other mammalian species in comparison to those in lower eukaryotes. The authors present their own views with regard to some of the major outstanding questions surrounding the nature and functions of the SMC family of proteins.

Characterization of vertebrate cohesin complexes and their regulation in prophase

In eukaryotes, sister chromatids remain connected from the time of their synthesis until they are separated in anaphase. This cohesion depends on a complex of proteins called cohesins. In budding yeast, the anaphase-promoting complex (APC) pathway initiates anaphase by removing cohesins from chromosomes. In vertebrates, cohesins dissociate from chromosomes already in prophase. To study their mitotic regulation we have purified two 14S cohesin complexes from human cells. Both complexes contain SMC1, SMC3, SCC1, and either one of the yeast Scc3p orthologs SA1 and SA2. SA1 is also a subunit of 14S cohesin in Xenopus. These complexes interact with PDS5, a protein whose fungal orthologs have been implicated in chromosome cohesion, condensation, and recombination. The bulk of SA1- and SA2-containing complexes and PDS5 are chromatin-associated until they become soluble from prophase to telophase. Reconstitution of this process in mitotic Xenopus extracts shows that cohesin dissociation does neither depend on cyclin B proteolysis nor on the presence of the APC. Cohesins can also dissociate from chromatin in the absence of cyclin-dependent kinase 1 activity. These results suggest that vertebrate cohesins are regulated by a novel prophase pathway which is distinct from the APC pathway that controls cohesins in yeast.

Fine-scale comparative mapping of the human 7q11.23 region and the orthologous region on mouse chromosome 5G: the low-copy repeats that flank the Williams-Beuren syndrome deletion arose at breakpoint sites of an evolutionary inversion(s)

Williams-Beuren syndrome (WBS) is a developmental disorder caused by haploinsufficiency for genes deleted in chromosome band 7q11.23. A common deletion including at least 16-17 genes has been defined in the great majority of patients. We have completed a physical and transcription map of the WBS region based on analysis of high-throughput genome sequence data and assembly of a BAC/PAC/YAC contig, including the characterization of large blocks of gene-containing low-copy-number repeat elements that flank the commonly deleted interval. The WBS deletions arise as a consequence of unequal crossing over between these highly homologous sequences, which confer susceptibility to local chromosome rearrangements. We have also completed a clone contig, genetic, and long-range restriction map of the mouse homologous region, including the orthologues of all identified genes in the human map. The order of the intradeletion genes appears to be conserved in mouse, and no low-copy-number repeats are found in the region. However, the deletion region is inverted relative to the human map, exactly at the flanking regions. Thus, we have identified an evolutionary inversion with chromosomal breakpoints at the sites where the human 7q11.23 low-copy-number repeats are located. Additional comparative mapping suggests a model for human chromosome 7 evolution due to serial inversions leading to genomic duplications. This high-resolution mouse map provides the framework required for the generation of mouse models for WBS mimicking the human molecular defect.

Identification and characterization of SA/Scc3p subunits in the Xenopus and human cohesin complexes

A multisubunit protein complex, termed cohesin, plays an essential role in sister chromatid cohesion in yeast and in Xenopus laevis cell-free extracts. We report here that two distinct cohesin complexes exist in Xenopus egg extracts. A 14S complex (x-cohesin(SA1)) contains XSMC1, XSMC3, XRAD21, and a newly identified subunit, XSA1. In a second 12.5S complex (x-cohesin(SA2)), XSMC1, XSMC3, and XRAD21 associate with a different subunit, XSA2. Both XSA1 and XSA2 belong to the SA family of mammalian proteins and exhibit similarity to Scc3p, a recently identified component of yeast cohesin. In Xenopus egg extracts, x-cohesin(SA1) is predominant, whereas x-cohesin(SA2) constitutes only a very minor population. Human cells have a similar pair of cohesin complexes, but the SA2-type is the dominant form in somatic tissue culture cells. Immunolocalization experiments suggest that chromatin association of cohesin(SA1) and cohesin(SA2) may be differentially regulated. Dissociation of x-cohesin(SA1) from chromatin correlates with phosphorylation of XSA1 in the cell-free extracts. Purified cdc2-cyclin B can phosphorylate XSA1 in vitro and reduce the ability of x-cohesin(SA1) to bind to DNA or chromatin. These results shed light on the mechanism by which sister chromatid cohesion is partially dissolved in early mitosis, far before the onset of anaphase, in vertebrate cells.