![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Previous Article | Next Article 
Molecular and Cellular Biology, December 2007, p. 8522-8532, Vol. 27, No. 24
0270-7306/07/$08.00+0 doi:10.1128/MCB.01007-07
Copyright © 2007, American Society for Microbiology. All Rights Reserved.
,
Seth Noone,1,
Jill M. Henry,1,¶
Karin Gaudenz,1
Brian Sanderson,1
Chris Seidel,1 and
Jennifer L. Gerton1,2*
Stowers Institute for Medical Research, Kansas City, Missouri 64110,1 Department of Biochemistry and Molecular Biology, University of Kansas Medical Center, 4011 Wahl Hall East, 3901 Rainbow Blvd., Kansas City, Kansas 661602
Received 7 June 2007/ Returned for modification 25 July 2007/ Accepted 25 September 2007
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
In eukaryotic cells, cohesion is mediated by several evolutionarily conserved proteins. The complex itself is composed of two SMC (structural maintenance of chromosomes) subunits, Smc1 and Smc3, and two non-SMC subunits, Mcd1/Scc1 and Scc3. Together, these subunits form a large ring-shaped complex that is essential for cohesion between sister chromatids (14, 16). The loading of cohesin onto chromosomes in G1 is dependent on the Scc2/4 complex (5, 40). Cohesion is established during DNA replication (42). Cohesion is dissolved at the metaphase-to-anaphase transition by separase, which cleaves Mcd1, resulting in the movement of sister chromatids into separate daughter cells (41, 43). Although the molecular structure of cohesin has been well studied, exactly how this ring-shaped complex interacts with DNA remains unclear. Furthermore, since cohesin is fairly abundant and is present on chromosomes throughout most of the cell cycle, it is interesting to speculate how this complex might accommodate other processes that take place on DNA, such as transcription.
While the cohesin complex reproducibly associates with the same regions of the genome in each cell cycle, cohesin does not appear to bind to a particular DNA motif (13, 25). Cohesin association follows a common pattern on all chromosomes, with regions of intense association in the pericentric/centromere regions and shorter, less intense associations distributed in a semiperiodic manner throughout chromosome arms (3, 13, 25). Intriguingly, cohesin association shows a strong bias toward intergenic regions, where transcription from surrounding units is converging (13, 25). Furthermore, transcript elongation through a cohesin-associated region on a plasmid disrupts the association (13). These results suggest that cohesin may be localized by RNA polymerase sliding or pushing cohesin rings during transcription (25, 35), resulting in a conservation of cohesion. An alternative possibility is that cohesin complexes are assembled and disassembled at loci according to their transcriptional status, with active transcription units forming zones of inhibition for cohesin assembly. According to current models, this could result in the destruction of cohesion, since cohesion must be established concurrent with replication, although recent evidence suggests that cohesion can be established without replication (45). The connection between cohesin location and transcription is not understood.
We were interested in the relationship between cohesin location and transcription. We manipulated the transcriptional program of the cell and followed changes in cohesin association. We found that transcriptional induction could disrupt cohesin association. However, at most loci we did not see evidence for an accumulation of cohesin at the 3' ends of induced genes, which could indicate sliding. In addition, we observed that cohesin could associate with transcriptionally repressed regions, a result not easily explained by the sliding model. Scc2, a component of the Scc2/4 cohesin loading complex (5), is not required for transcription-based association/disassociation. Consistent with a previous report (17), we found that Mcd1 that associates with chromosomes during G2/M does not participate in functional cohesion. Taken together, these results suggest that cohesion may be degraded by transcription. The flexibility of the pattern of cohesin binding suggests that its association depends on features of chromatin which reflect the local transcriptional status.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-factor (final concentration, 1 µg/ml) or in metaphase by the addition of nocodazole (final concentration, 15 µg/ml), followed by 1 to 2 h of incubation at 30°C, depending on the strain. For the experiment in Fig. 3, cultures were arrested with nocodazole for 1 h and washed twice with warm YP plus nocodazole to remove any residual YPD. YPgal and nocodazole were added, and cells were incubated for an additional 6 h to ensure complete activation of the galactose promoter.
|
|
Global gene expression analysis. Aliquots (50 ml) of culture were pelleted at room temperature, flash frozen with liquid nitrogen, and stored at –80°C. Yeast total RNA was extracted from cell pellets by the acid-phenol method followed by ethanol precipitation (36). Polyadenylated RNA was prepared from total RNA by purification with an oligo(dT) cellulose column as described previously (7). RNA quality was assessed on a Bioanalyzer 2100 machine (Agilent). Cy5 and Cy3 dyes were indirectly incorporated into cDNA by using aminoallyl-dUTP (Ambion). Aminoallyl-dUTP was incorporated during reverse transcription of 2 µg of polyadenylated RNA, with priming by a dT18 and a dN9 oligomer. Reverse transcription, coupling, and hybridization protocols were followed as listed on the DeRisi lab website (http://derisilab.ucsf.edu/microarray/protocols.html). Comparative hybridization was carried out using a common polyadenylated RNA reference pool for all samples. Samples were dye swapped, and Cy5- and Cy3-labeled probes were hybridized together to poly-L-lysine-coated microarrays printed with PCR-amplified fragments containing all known and predicted S. cerevisiae open reading frames (ORFs) and with custom tiling oligomers of 70 nucleotides in length. Microarrays were scanned with a GenePix 4000B laser scanner (Axon Instruments, Foster City, CA), and the images were analyzed and stored in Acuity software (Molecular Devices Corporation). Data presented were normalized so that the mean of the ratios of medians of all features on the microarray was equal to 1. The data were further filtered to remove any flagged features, features that did not have a feature-to-background intensity of >150, or features with a correlation coefficient of <0.5. Duplicate and dye-swapped data were combined based on their median values and displayed as log2-normalized binding ratios.
ChIP.
For each chromatin immunoprecipitation (ChIP) experiment presented in this study, at least two independent experiments were performed, with similar results. We previously presented evidence that a cohesin ChIP performed with or without antibody on an untagged strain resulted in a stochastic signal (13). Cell cycle arrests were verified by FACS analysis (data not shown). Since cohesin is known to bind strongly to the centromere, as a positive control we compared the levels of enrichment of CEN3 DNA by semiquantitative PCR for each ChIP sample and for a control ChIP performed without antibody, and in every case, there was a significant enrichment for the CEN3 sequence when antibody was included. When the DNA recovered from a ChIP performed without antibody was amplified and hybridized to a microarray, no distinct pattern was observed (see Fig. 4). ChIP methods have been published previously and were performed as described previously (13, 23, 27). The control PCR product for each ChIP was amplified using a primer pair that amplified an 
280-bp region specific to the S. cerevisiae centromere III region. Primer sequences are shown in Table S7 in the supplemental material.
|
Quantitative PCR analysis. Primers used for quantitative PCR are listed in Table S7 in the supplemental material. The real-time reactions were performed using an iCycler machine (Bio-Rad) and iQ SYBR green supermix (Bio-Rad). Dynamic well factors were used, and cycling parameters were as follows: 94°C for 30 s, 50°C for 30 s, and 72°C for 60 s, repeated 40 times. A melting curve analysis was performed, starting at 50°C for 20 s and increasing at a rate of 0.5°C/cycle for 80 cycles. Each run contained a relative standard curve utilizing a serial dilution (1e–1, 1e–2, 1e–3, 1e–4, and 1e–5) of the IP total control DNA and CEN3 primers. Data analysis was performed with iCycler v3.0 software, and the cycle threshold (CT) was manually set at a threshold of 500.
Microscopy and one-spot-two-spot analysis. Medium was supplemented with 0.002% adenine to reduce autofluorescence and with CuSO4 to a final concentration of 500 µM to induce LacI-green fluorescent protein (LacI-GFP) from the copper promoter. One milliliter of culture was fixed with 4% paraformaldehyde-3.4% sucrose for 15 min, washed twice with 1 ml of phosphate-buffered saline, and then resuspended in 200 µl of phosphate-buffered saline. An aliquot was used to prepare a slide for fluorescence microscopy on an Axiovert microscope. Twelve to 18 z sections were collected as 0.225- to 0.3-µm slices to ensure that we could visualize fluorescence throughout the cell. The exposure time was fixed at 200 ms. Aliquots were stained with DAPI (4',6'-diamidino-2-phenylindole) for the images shown in Fig. 5 and in Fig. S1 in the supplemental material. Image processing was carried out using AxioVision 4.5 software. Counting of GFP spots was performed manually.
|
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The pattern of cohesin association showed several significant changes during amino acid starvation. Overall, the association with many ribosomal protein genes was significantly enhanced (P < 0.001), presumably because transcription of ribosomal protein genes is shut down during amino acid starvation (10). Two examples are presented in Fig. 1D and E. Cohesin association with many genes involved in amino acid biosynthesis was significantly depleted (P < 0.001), presumably because the transcription of these genes was induced (10). Four examples are presented in Fig. 1A to C and F.
|
The centromeric and pericentric regions of each chromosome are normally associated with large amounts of cohesin (3, 13). Although cohesin association with these regions may be subject to different parameters from those for arm cohesin, we found that transcriptional induction caused cohesin to be depleted from these domains, as shown by the disruption of Mcd1 at CIT2, which was transcriptionally induced by amino acid starvation and is located 6 kb from CEN3 (Fig. 1B). When transcription at this region was shut off by transfer back to rich medium, we observed that Mcd1 association increased without any significant change in neighboring regions. The pattern of disappearance and reappearance of Mcd1 in the face of transcriptional induction and repression, respectively, is most consistent with a dynamic association of Mcd1 with chromosomes.
We also examined the pattern of Mcd1 association at regions where transcription was repressed by amino acid starvation. The majority of genes in this class are ribosomal protein genes. We found that if a particular ribosomal protein gene or a neighboring region was associated with Mcd1 in rich medium, this association increased during amino acid starvation (Fig. 1D and E). For instance, the intergenic region where transcription converges from RPL2B and VID28 was a peak of cohesin association in rich medium. When transcription of RPL2B was repressed, more Mcd1 appeared to associate with its promoter and the ORF located 5', i.e., FAF1. Following a return to rich medium, this association was disrupted and Mcd1 assumed a preinduction profile. RPL14B is located immediately adjacent to CEN8, and Mcd1 association increased for this ORF and its promoter during amino acid starvation and then decreased upon transfer back to rich medium. Thus, in terms of Mcd1 association, cohesin at arm regions and pericentric domains appears to respond similarly to transcriptional activation or repression.
One possibility is that Mcd1 responds to transcription by disassociation but that the other subunits do not. A more detailed study of the behavior of each subunit of the complex will be important in the future to elucidate the effect of transcription on the entire complex, although the results shown in Fig. 6 support the idea that the association of the Smc1 subunit is similarly altered by transcription.
|
A culture growing in YPraff was arrested in G1 by using alpha factor and then released in the presence of nocodazole and either raffinose or galactose, followed by global mapping of cohesin by ChIP chip analysis. In raffinose, cohesin was found in the promoter region and ORF of GAL2, as previously observed (Fig. 2A, red arrow). When galactose was included during S phase, cohesin was instead located 3' of GAL2, in SRL2 (Fig. 2A, black arrow), similar to the pattern observed previously when cells were continuously cultured in galactose (13). This demonstrates that the transcriptional profile during cohesion establishment can affect the pattern of cohesin association.
|
Cohesin localization and galactose-induced transcription. We were interested in understanding the mechanism by which transcription caused a new peak of cohesin association 3' of GAL2, in SRL2. One possibility is that the movement of RNA polymerase II through the region slides cohesin rings to the 3' end of the gene. A second possibility is that the association of the complex is dynamic and that cohesin is removed from the GAL2 ORF and loaded on SRL2. We tested if the cohesin associated with this locus following galactose induction was "old" cohesin or "new" cohesin.
To accomplish this, we used a haploid strain (1827-11B) in which the endogenous copy of Mcd1 was tagged with 18 copies of Myc (Mcd1-18Myc) and an ectopic galactose-inducible copy was tagged with 6 copies of hemagglutinin (HA) (gal-Mcd1-6HA). In order to verify that Mcd-18Myc or Mcd1-6HA did not cause a defect in cohesion, we performed a growth assay comparing strains containing these alleles to a wild-type strain. We found no growth differences (see Fig. S1A in the supplemental material) and no difference in the ability to maintain cohesion (see Fig. S1B in the supplemental material). Strain 1827-11B was initially grown in glucose-containing medium to ensure that no GAL1-10-Mcd1-6HA protein was being made. Cells were arrested using nocodazole, and YPD was replaced with YPgal to induce the transcription of GAL2 and GAL1-10-MCD1-6HA. Western analysis was used to follow the expression of the two epitope-tagged Mcd1 proteins, ChIP was performed to follow their association with DNA, and microarrays were used to follow gene expression. GAL1-10-Mcd1-6HA was strongly expressed by 4 hours according to Western blotting (Fig. 3A). Mcd1-18Myc levels decreased slightly over the course of the arrest in galactose; similar results were obtained when cultures were arrested in glucose, demonstrating that the decrease was not due to the expression of the Mcd1-6HA subunit (data not shown). Expression of mRNA from GAL2 displayed similar kinetics, that is, maximal expression by 4 h (Fig. 3B). ChIP followed by real-time PCR demonstrated that histone H3 was strongly depleted from the GAL2 locus by 6 h (see Fig. S2 in the supplemental material). FACS analysis confirmed the G2/M arrest throughout the experiment (Fig. 3C).
We collected the pattern of Mcd1 association by microarray analysis (for the patterns at 0 and 6 h at GAL2, see Fig. 3G; for the patterns at all time points, see Fig. S3 in the supplemental material) and used real-time PCR to quantify the amounts of Mcd1-6HA and Mcd1-18Myc associated with the region upstream of GAL2 and SRL2 and with CEN3 and FMP32 (cohesin sites where transcription was not affected in galactose compared to that in glucose) in a nocodazole-arrested culture prior to (t = 0 h) and following 6 h of induction with galactose (Fig. 3E and F). For the Mcd1-18Myc present at 6 h, we cannot distinguish whether it was the same Mcd1 that was present at the region at time zero or had been exchanged for Mcd1 present in the soluble nuclear pool. While Mcd1-18Myc was still present at high levels at 6 h (Fig. 3E), a significant amount of Mcd1-6HA was newly associated with these regions (Fig. 3F). These results demonstrate that (i) Mcd1-6HA can associate with chromatin in the absence of DNA replication at SRL2 as well as in the region upstream of GAL2; (ii) Mcd1-6HA can associate with regions where transcription is not affected, such as CEN3 and FMP32; and (iii) Mcd1-6HA localizes to the same regions as Mcd1-18Myc. These results are consistent with de novo association of Mcd1 with chromosomes in G2/M, as previously shown (17).
While some cohesin may be pushed into SRL2 from GAL2 by RNA polymerase II (since we cannot distinguish old versus new Mcd1-18Myc), there is definitely new loading of cohesin (Mcd1-6HA). Taken together, our results suggest that in some cases of transcriptional induction, such as that of GAL2, it is possible that cohesion is maintained locally but that at other loci, such as MET17 and CIT2, where cohesin completely disappears upon transcriptional induction, cohesion may be degraded locally.
When we looked at the pattern of Mcd1-6HA association throughout the genome, we saw that it was present at most sites bound by Mcd1-18Myc (Fig. 4; see Fig. S4 in the supplemental material). Since the number of genes at which cohesin is located and where transcription varies fivefold or more in glucose versus galactose is small (13), transcriptional changes cannot account for the global loading of Mcd1-6HA that we observed. We concluded that the pattern of cohesin association can be altered by transcription but that the new association of Mcd1-6HA with chromosomes does not require transcription. Because Mcd1-6HA associates with the same sites as Mcd1-18Myc, this association appears to be specific to particular preferred regions (not a nonspecific process).
Mcd1 that associates with chromosomes in G2/M is not functional for cohesion. We asked whether the new Mcd1 that associates with chromosomes in metaphase participates in cohesion. A previous experiment conducted by Haering and colleagues demonstrated that a noncleavable version of Mcd1 expressed in G2/M could associate with chromosomes but did not prevent chromosome segregation (17). We constructed strains containing a temperature-sensitive allele of Mcd1 (scc1-173) along with galactose-inducible Mcd1-6HA. In addition, to visualize chromatid cohesion, we incorporated LacI-GFP and lacO repeats at one of two locations, namely, an arm site near HIM1 on chromosome IV and a telomere on chromosome IV (2). Each strain was grown in YPraff medium and arrested with nocodazole. Galactose was added to half of the culture. The arrest was maintained during the 3-h induction (Fig. 5A). We monitored whether GAL1-10-Mcd1-6HA, expressed during G2/M (Fig. 5B), could rescue cohesion when cultures were shifted to the nonpermissive temperature of 37°C for 1 h by counting the number of cells with one or two GFP spots (Fig. 5B).
We found that the expression of GAL1-10-Mcd1-6HA did not prevent sister separation at either the arm site or the telomere site (Fig. 5C). The observation that telomere cohesion was more affected than arm cohesion by the shift to 37°C is consistent with a recent report that cohesion is the only force keeping telomeres together (1). Although the numbers of cells with two GFP spots were similar for the raffinose-plus-galactose-induced culture versus the raffinose-only culture prior to the temperature shift, we were surprised to observe that the expression of Mcd1-6HA increased the cohesion defect twofold at both the arm and telomere sites monitored (P < 0.001 for the arm and P = 0.005 for the telomere). One possible explanation for the increased cohesion defect in the galactose-induced cultures is that the expression of Mcd1-6HA increased the turnover of Mcd1 in cohesin complexes and the turnover rendered the complexes nonfunctional.
In order to assay whether newly expressed Mcd1 could be found to interact with other components of the cohesin complex, we checked whether GAL1-10-Mcd1-6HA was able to coimmunoprecipitate with Smc3-3FLAG when Mcd1-6HA was expressed in nocodazole-arrested cultures grown continuously in YPgal or when galactose was added only during the arrest. We found in both cases that anti-HA antibody was able to coimmunoprecipitate Smc3-3FLAG (Fig. 5D). In addition, in a similar experiment, we subjected the immunoprecipitated material to mass spectrometry, and both Smc1 and Smc3 were detected (data not shown). These results suggest that when Mcd1-6HA is induced during metaphase, it is able to be incorporated into a cohesin complex. Taken together, our results indicate that cohesin complexes that associate with chromatin outside of S phase have the same binding site preferences as those that load during DNA replication but do not efficiently mediate cohesion.
Scc2 is not required for cohesin association or disassociation in G2/M. Mutations in the human homolog of Scc2/Nipped B have been linked with Cornelia de Lange syndrome and mild precocious sister chromatid separation (20, 22, 39). In budding yeast, Scc2/4 has been shown to be required for the association of the cohesin complex with chromatin following G1 arrest by cellular fractionation and by ChIP (5, 40). It has been proposed that cohesin inhibits long-range transcriptional activation of the Drosophila cut gene and that Scc2/Nipped B facilitates activation by regulating cohesin-chromosome binding (9). We asked if the Scc2/4 complex was required for the alterations observed in response to transcriptional changes that occur during amino acid starvation in G2/M-arrested cells.
To test the requirement for Scc2 in transcription-induced changes in cohesin, the scc2-4 temperature-sensitive mutant strain K8250 was grown in YPD at 23°C (permissive temperature) and arrested in G2/M. Half of the culture was shifted to 33°C for 30 min to inactivate scc2-4, as previously described (5, 37). Half of each culture was then subjected to amino acid starvation while being maintained in metaphase arrest. We analyzed the pattern of Smc1-6HA association at the locations previously identified to have the most significant changes during amino acid starvation. We found that changes occurred similarly in some genes at both the permissive and nonpermissive temperatures. For example, the association with several genes occurred normally (PRS1/RPL17A, RPL14A, and RPL2B) (Fig. 6B), and disassociation occurred normally at MET17 (Fig. 6A). First, this result confirms that Smc1 association, in addition to that of Mcd1, can respond to transcription. Second, it demonstrates that Scc2 is not essential for disassociation (Fig. 6A) or association (Fig. 6B) of cohesin in G2/M-arrested cells.
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Sliding of cohesin versus loading and unloading. If cohesin can slide locally along chromosomes, then a region where Mcd1 is depleted by transcription will be accompanied by a new peak downstream of transcription, due to sliding. Alternatively, Mcd1 may simply dissociate, in which case no new peak would be observed. In the regions presented in Fig. 1A and B, Mcd1 was depleted over regions that were transcriptionally induced, but there is no evidence of a new peak located 3' of transcription. In general, we observed that when transcription was shut off by transfer back to rich medium, Mcd1 resumed its preinduction profile. It seems difficult to reconcile how transcriptional repression could slide cohesin rings to their original location unless one postulates that diffusion of rings can occur in the absence of transcription. These observations suggest that at least for some locations, cohesin is not slid along DNA by RNA polymerase II but, rather, can disassociate from the DNA completely and can reassociate following transcriptional shutoff. Our results for GAL2 are consistent with either sliding or loading/unloading.
Cohesin and chromatin. Several models for the association of the cohesin complex with DNA have been proposed. Most models do not require that the cohesin ring is in direct contact with DNA (14, 16-18, 29, 31). However, cohesin reproducibly associates with the same regions of the genome, although there is no obvious sequence specificity to this interaction. It is unclear how many cohesin rings are present at an associated region. Cohesin can be cross-linked to DNA by formaldehyde, which suggests that it is intimately associated with DNA. We propose that the cohesin complex itself prefers to associate with particular chromosomal regions. These preferences do not seem to require the Scc2/4 loading complex. The cohesin complex may be able to distinguish one or more chromatin features, which likely reflects the local transcription status.
The transcriptional status of a region is reflected in several chromatin features, including (i) histone variants, (ii) nucleosome density, (iii) histone modifications, and (iv) supercoiling. Although histone variants (26, 32, 47) and nucleosome density (24, 46) have been mapped genome-wide in yeast, there is nothing obvious that distinguishes intergenic regions between converging transcriptional units with regard to these two chromatin features. The following correlation between a particular histone modification and cohesin association has been demonstrated: when a double-strand break is made, H2A is phosphorylated and cohesin associates with this region to facilitate double-strand break repair (44). Finally, there is some evidence that the promoter regions of genes may transiently accumulate negative supercoils during transcription (21). In this case, terminator regions may accumulate positive supercoils. This could potentially be a preferred substrate for cohesin association. Ivanov and Nasmyth (19) precipitated cohesin associated with a plasmid. The majority was supercoiled, unlike the plasmid precipitated via control proteins (19). In summary, at this time there is not enough evidence to suggest all of the specific features of chromatin that are recognized by cohesin, but it seems likely that local chromatin structure, including histone modification and topology, could play a role in localizing the complex (33).
Cohesin association with DNA may be dynamic. When we induce Mcd1 in G2/M, we observe that this Mcd1 associates with chromosomes at the usual locations. However, newly associated Mcd1 does not appear to participate in functional cohesion. A previous study found that inducing a noncleavable subunit of Mcd1 during metaphase, which did associate with chromosomes, did not have any effect on chromosome segregation, prompting the authors to suggest that Mcd1 does not turn over in the cohesin complex (17). An alternate interpretation that is more consistent with our observations is that Mcd1 can be exchanged but that this renders the complex nonfunctional for cohesion. Although cohesin that associates with chromosomes in G2/M is apparently nonfunctional for cohesion, it was recently shown that functional cohesion can be established in G2 in response to a double-strand break (37). Photobleaching studies recently identified a significant pool of cohesin that interacts dynamically with chromatin in interphase (11), although the nature and function of this cohesin are unknown.
Cohesion must persist from the time of replication until sisters separate from each other prior to cell division. While it might seem counterintuitive to suppose that the association of cohesin with chromosomes could be dynamic, cohesin is located throughout chromosome arms, with a potential site approximately every 11 kb in yeast (3, 13). This level of redundancy could permit any individual complex to be disrupted to allow for transcription without compromising the integrity of cohesion for the entire chromosome. Overexpression of Mcd1 may have artificially increased the turnover of this subunit and rendered a higher-than-normal percentage of complexes nonfunctional, resulting in a cohesion defect. Since we have demonstrated that transcription can disrupt Mcd1 association, it may be that transcription increases subunit turnover and has the effect of locally disrupting cohesion.
The length of time over which the cohesin complex acts may be on the order of an hour (yeast) to 40 years (human oocytes). In the latter case, a lack of subunit turnover would require that the complex established during fetal development in an oocyte would provide cohesion 40 years later, which seems unlikely. Over time, if cohesin subunits turn over, due to transcription or otherwise, the complex may no longer mediate cohesion. If cohesion degrades over time, this could potentially explain why chromosome nondisjunction is more likely as a woman ages.
The link between cohesin, transcription, and Scc2. It is becoming increasingly apparent that cohesin has roles in both cohesion and transcription. The ability of cohesin to accommodate changing transcriptional programs may be especially important during development in more complex organisms. Mutations in Scc2/Nipped B, Smc1, and Smc3 in humans result in a constellation of developmental defects known as Cornelia de Lange syndrome (6, 22, 30). This developmental disorder is probably not the result of defects in cohesion; rather, this syndrome is most likely due to transcriptional defects during embryogenesis (8). Dorsett et al. proposed that Nipped B (the fly homolog of Scc2) aids in the removal of cohesin to allow for long-range promoter-enhancer interactions and subsequent transcription (34). Scc2 is also required for loading in G1 (5). However, we demonstrate that cohesin can respond to transcription in the absence of Scc2 function, suggesting that Scc2 is not required for local association or disassociation in response to transcription in G2/M-arrested yeast cells. There are many reasons why our scc2-4 mutant may not recapitulate the Nipped B mutant phenotype, with the most obvious being that the mutations are different. Regardless, since mutations in cohesin subunits and a loading factor may cause a transcriptional defect in humans, it is important to understand the underlying molecular defect in transcription caused by these mutations. Yeast will likely prove a powerful model system with which to study the link between cohesin and transcription in molecular detail.
Conclusions. In summary, we present evidence that cohesin localization is determined in part by the transcriptional status of a locus. Cohesin association with chromosomes may be dynamic in order to accommodate transcriptional changes. We speculate that this may be accompanied by subunit turnover that leads to local cohesion defects. We propose that certain genomic regions may contain chromatin features, such as histone modifications, that make them preferred locations for cohesin association. The gamut of features that are recognized by cohesin remains to be defined.
|
ACKNOWLEDGMENTS |
|---|
We thank Sue Jaspersen, Kim Nasmyth, Marc Gartenberg, and Paul Megee for strains and Sue J. for help with microscopy. We thank Jeff Haug and the Cytometry facility for FACscan analysis. We thank A. McNairn for helpful discussions and S. Savage for technical assistance.
|
FOOTNOTES |
|---|
Published ahead of print on 8 October 2007.
Supplemental material for this article may be found at http://mcb.asm.org/.
Present address: SAFC Biosciences, Cell Sciences and Development, 2909 Laclede Ave., St. Louis, MO 63103.
Present address: Department of Biochemistry and Molecular Genetics, University of Colorado Health Sciences, Aurora, CO 80045.
¶ Present address: School of Pharmacy, Creighton University, Omaha, NE 68178.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
2. Bhalla, N., S. Biggins, and A. W. Murray. 2002. Mutation of YCS4, a budding yeast condensin subunit, affects mitotic and nonmitotic chromosome behavior. Mol. Biol. Cell 13:632-645.
3. Blat, Y., and N. Kleckner. 1999. Cohesins bind to preferential sites along yeast chromosome III, with differential regulation along arms versus the centric region. Cell 98:249-259.[CrossRef][Medline]
4. Bohlander, S. K., R. Espinosa III, M. M. Le Beau, J. D. Rowley, and M. O. Diaz. 1992. A method for the rapid sequence-independent amplification of microdissected chromosomal material. Genomics 13:1322-1324.[CrossRef][Medline]
5. Ciosk, R., M. Shirayama, A. Shevchenko, T. Tanaka, A. Toth, and K. Nasmyth. 2000. Cohesin's binding to chromosomes depends on a separate complex consisting of Scc2 and Scc4 proteins. Mol. Cell 5:243-254.[CrossRef][Medline]
6. Deardorff, M. A., M. Kaur, D. Yaeger, A. Rampuria, S. Korolev, J. Pie, C. Gil-Rodriguez, M. Arnedo, B. Loeys, A. D. Kline, M. Wilson, K. Lillquist, V. Siu, F. J. Ramos, A. Musio, L. S. Jackson, D. Dorsett, and I. D. Krantz. 2007. Mutations in cohesin complex members SMC3 and SMC1A cause a mild variant of Cornelia de Lange syndrome with predominant mental retardation. Am. J. Hum. Genet. 80:485-494.[CrossRef][Medline]
7. DeRisi, J. L., V. R. Iyer, and P. O. Brown. 1997. Exploring the metabolic and genetic control of gene expression on a genomic scale. Science 278:680-686.
8. Dorsett, D. 2004. Adherin: key to the cohesin ring and Cornelia de Lange syndrome. Curr. Biol. 14:R834-R836.[CrossRef][Medline]
9. Dorsett, D., J. C. Eissenberg, Z. Misulovin, A. Martens, B. Redding, and K. McKim. 2005. Effects of sister chromatid cohesion proteins on cut gene expression during wing development in Drosophila. Development 132:4743-4753.
10. Gasch, A. P., P. T. Spellman, C. M. Kao, O. Carmel-Harel, M. B. Eisen, G. Storz, D. Botstein, and P. O. Brown. 2000. Genomic expression programs in the response of yeast cells to environmental changes. Mol. Biol. Cell 11:4241-4257.
11. Gerlich, D., B. Koch, F. Dupeux, J. M. Peters, and J. Ellenberg. 2006. Live-cell imaging reveals a stable cohesin-chromatin interaction after but not before DNA replication. Curr. Biol. 16:1571-1578.[CrossRef][Medline]
12. Gerton, J. L., J. DeRisi, R. Shroff, M. Lichten, P. O. Brown, and T. D. Petes. 2000. Inaugural article: global mapping of meiotic recombination hotspots and coldspots in the yeast Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA 97:11383-11390.
13. Glynn, E. F., P. C. Megee, H. G. Yu, C. Mistrot, E. Unal, D. E. Koshland, J. L. DeRisi, and J. L. Gerton. 2004. Genome-wide mapping of the cohesin complex in the yeast Saccharomyces cerevisiae. PLoS Biol. 2:E259.[CrossRef][Medline]
14. Gruber, S., C. H. Haering, and K. Nasmyth. 2003. Chromosomal cohesin forms a ring. Cell 112:765-777.[CrossRef][Medline]
15. Guacci, V., D. Koshland, and A. Strunnikov. 1997. A direct link between sister chromatid cohesion and chromosome condensation revealed through the analysis of MCD1 in S. cerevisiae. Cell 91:47-57.[CrossRef][Medline]
16. Haering, C. H., J. Lowe, A. Hochwagen, and K. Nasmyth. 2002. Molecular architecture of SMC proteins and the yeast cohesin complex. Mol. Cell 9:773-788.[CrossRef][Medline]
17. Haering, C. H., D. Schoffnegger, T. Nishino, W. Helmhart, K. Nasmyth, and J. Lowe. 2004. Structure and stability of cohesin's Smc1-kleisin interaction. Mol. Cell 15:951-964.[CrossRef][Medline]
18. Huang, C. E., M. Milutinovich, and D. Koshland. 2005. Rings, bracelet or snaps: fashionable alternatives for Smc complexes. Philos. Trans. R. Soc. Lond. B 360:537-542.[CrossRef][Medline]
19. Ivanov, D., and K. Nasmyth. 2005. A topological interaction between cohesin rings and a circular minichromosome. Cell 122:849-860.[CrossRef][Medline]
20. Kaur, M., C. DeScipio, J. McCallum, D. Yaeger, M. Devoto, L. G. Jackson, N. B. Spinner, and I. D. Krantz. 2005. Precocious sister chromatid separation (PSCS) in Cornelia de Lange syndrome. Am. J. Med. Genet. A 138:27-31.[Medline]
21. Kouzine, F., J. Liu, S. Sanford, H. J. Chung, and D. Levens. 2004. The dynamic response of upstream DNA to transcription-generated torsional stress. Nat. Struct. Mol. Biol. 11:1092-1100.[CrossRef][Medline]
22. Krantz, I. D., J. McCallum, C. DeScipio, M. Kaur, L. A. Gillis, D. Yaeger, L. Jukofsky, N. Wasserman, A. Bottani, C. A. Morris, M. J. Nowaczyk, H. Toriello, M. J. Bamshad, J. C. Carey, E. Rappaport, S. Kawauchi, A. D. Lander, A. L. Calof, H. H. Li, M. Devoto, and L. G. Jackson. 2004. Cornelia de Lange syndrome is caused by mutations in NIPBL, the human homolog of Drosophila melanogaster Nipped-B. Nat. Genet. 36:631-635.[CrossRef][Medline]
23. Laloraya, S., V. Guacci, and D. Koshland. 2000. Chromosomal addresses of the cohesin component Mcd1p. J. Cell Biol. 151:1047-1056.
24. Lee, C. K., Y. Shibata, B. Rao, B. D. Strahl, and J. D. Lieb. 2004. Evidence for nucleosome depletion at active regulatory regions genome-wide. Nat. Genet. 36:900-905.[CrossRef][Medline]
25. Lengronne, A., Y. Katou, S. Mori, S. Yokobayashi, G. P. Kelly, T. Itoh, Y. Watanabe, K. Shirahige, and F. Uhlmann. 2004. Cohesin relocation from sites of chromosomal loading to places of convergent transcription. Nature 430:573-578.[CrossRef][Medline]
26. Li, B., S. G. Pattenden, D. Lee, J. Gutierrez, J. Chen, C. Seidel, J. Gerton, and J. L. Workman. 2005. Preferential occupancy of histone variant H2AZ at inactive promoters influences local histone modifications and chromatin remodeling. Proc. Natl. Acad. Sci. USA 102:18385-18390.
27. Meluh, P. B., and D. Koshland. 1997. Budding yeast centromere composition and assembly as revealed by in vivo cross-linking. Genes Dev. 11:3401-3412.
28. Michaelis, C., R. Ciosk, and K. Nasmyth. 1997. Cohesins: chromosomal proteins that prevent premature separation of sister chromatids. Cell 91:35-45.[CrossRef][Medline]
29. Milutinovich, M., and D. E. Koshland. 2003. Molecular biology. SMC complexes—wrapped up in controversy. Science 300:1101-1102.
30. Musio, A., A. Selicorni, M. L. Focarelli, C. Gervasini, D. Milani, S. Russo, P. Vezzoni, and L. Larizza. 2006. X-linked Cornelia de Lange syndrome owing to SMC1L1 mutations. Nat. Genet. 38:528-530.[CrossRef][Medline]
31. Nasmyth, K. 2005. How might cohesin hold sister chromatids together? Philos. Trans. R. Soc. Lond. B 360:483-496.[CrossRef][Medline]
32. Raisner, R. M., P. D. Hartley, M. D. Meneghini, M. Z. Bao, C. L. Liu, S. L. Schreiber, O. J. Rando, and H. D. Madhani. 2005. Histone variant H2A.Z marks the 5' ends of both active and inactive genes in euchromatin. Cell 123:233-248.[CrossRef][Medline]
33. Riedel, C. G., J. Gregan, S. Gruber, and K. Nasmyth. 2004. Is chromatin remodeling required to build sister-chromatid cohesion? Trends Biochem. Sci. 29:389-392.[CrossRef][Medline]
34. Rollins, R. A., M. Korom, N. Aulner, A. Martens, and D. Dorsett. 2004. Drosophila nipped-B protein supports sister chromatid cohesion and opposes the stromalin/Scc3 cohesion factor to facilitate long-range activation of the cut gene. Mol. Cell. Biol. 24:3100-3111.
35. Ross, K. E., and O. Cohen-Fix. 2004. Molecular biology: cohesins slip sliding away. Nature 430:520-521.[Medline]
36. Spellman, P. T., G. Sherlock, M. Q. Zhang, V. R. Iyer, K. Anders, M. B. Eisen, P. O. Brown, D. Botstein, and B. Futcher. 1998. Comprehensive identification of cell cycle-regulated genes of the yeast Saccharomyces cerevisiae by microarray hybridization. Mol. Biol. Cell 9:3273-3297.
37. Strom, L., H. B. Lindroos, K. Shirahige, and C. Sjogren. 2004. Postreplicative recruitment of cohesin to double-strand breaks is required for DNA repair. Mol. Cell 16:1003-1015.[CrossRef][Medline]
38. Tanaka, T., J. Fuchs, J. Loidl, and K. Nasmyth. 2000. Cohesin ensures bipolar attachment of microtubules to sister centromeres and resists their precocious separation. Nat. Cell Biol. 2:492-499.[CrossRef][Medline]
39. Tonkin, E. T., T. J. Wang, S. Lisgo, M. J. Bamshad, and T. Strachan. 2004. NIPBL, encoding a homolog of fungal Scc2-type sister chromatid cohesion proteins and fly Nipped-B, is mutated in Cornelia de Lange syndrome. Nat. Genet. 36:636-641.[CrossRef][Medline]
40. Toth, A., R. Ciosk, F. Uhlmann, M. Galova, A. Schleiffer, and K. Nasmyth. 1999. Yeast cohesin complex requires a conserved protein, Eco1p(Ctf7), to establish cohesion between sister chromatids during DNA replication. Genes Dev. 13:320-333.
41. Uhlmann, F., F. Lottspeich, and K. Nasmyth. 1999. Sister-chromatid separation at anaphase onset is promoted by cleavage of the cohesin subunit Scc1. Nature 400:37-42.[CrossRef][Medline]
42. Uhlmann, F., and K. Nasmyth. 1998. Cohesion between sister chromatids must be established during DNA replication. Curr. Biol. 8:1095-1101.[CrossRef][Medline]
43. Uhlmann, F., D. Wernic, M. A. Poupart, E. V. Koonin, and K. Nasmyth. 2000. Cleavage of cohesin by the CD clan protease separin triggers anaphase in yeast. Cell 103:375-386.[CrossRef][Medline]
44. Unal, E., A. Arbel-Eden, U. Sattler, R. Shroff, M. Lichten, J. E. Haber, and D. Koshland. 2004. DNA damage response pathway uses histone modification to assemble a double-strand break-specific cohesin domain. Mol. Cell 16:991-1002.[CrossRef][Medline]
45. Unal, E., J. M. Heidinger-Pauli, and D. Koshland. 2007. DNA double-strand breaks trigger genome-wide sister-chromatid cohesion through Eco1 (Ctf7). Science 317:245-248.
46. Yuan, G. C., Y. J. Liu, M. F. Dion, M. D. Slack, L. F. Wu, S. J. Altschuler, and O. J. Rando. 2005. Genome-scale identification of nucleosome positions in S. cerevisiae. Science 309:626-630.
47. Zhang, H., D. N. Roberts, and B. R. Cairns. 2005. Genome-wide dynamics of Htz1, a histone H2A variant that poises repressed/basal promoters for activation through histone loss. Cell 123:219-231.[CrossRef][Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
