cohesins and condensins and linking it all up

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.


The objective is to bring the reader up-to-date on the subject under review.  Please first consult major texts to understand the historical aspects of the subject, and then the most recent major reviews to discover the current state of the field.   Students should attempt to review critically some of the major research papers in the field during the previous approximately 5 years and to define the interesting questions that remain to be answered.  This more recent literature should be searched for new information and ideas and should form the main part of the students' review. Some reviews by their nature may be entirely historical and should be structured accordingly.

Cohesins and Condensins: Linking it all up.

What are they? When were they discovered? What were their first known roles? How has this notion of function evolved recently? Pre-date complexes such as histones allowing for evolution(condensing; architect of..)


Organisation of the interphase genome

Gene expression

Animal development


Associated diseases- down syndrome? Conrnelia de lange


The array of functions that Cohesins and Condensins are tied to is becoming increasingly diverse. It was originally thought that the only functions of these proteins was to perform roles in chromatin condensation and cohesion , however it is now evident that processes from genome organisation to gene expression are affected by these proteins.


Chromosomes must be dynamic to achieve the multitude of functions that they are responsible for, as it is through this dynamism that processes such as cell division and gene expression can precede correctly. A substantial part of the adaptability that chromosomes possess emanates from the way in which they organise themselves in different areas as well as at particular stages of the cell cycle. This adaptability is established, partly, by a group of proteins known as structural maintainence of chromosome (SMC) proteins. These proteins, which are ubiquitous and conserved from prokaryotes to eukaryotes, are chromosomal ATPases that use energy from the process of ATP hydrolysis to organise the genome of a cell and regulate its functions. In eukaryotes there are at least three complexes which take advantage of SMC proteins: Cohesin, Condesin and the SMC5/SMC6 complex.

The importance of cohesin and condensin was first realised in their roles in mediating chromosome organisation during the process of mitosis (Nasmyth et al,1997)(Hirano et al, 1997).During this intense stage the cell is presented with the complex task of compacting chromosomes almost 10,000 times in length while also conserving vital intra and inter connections between chromosomes. Such changes requires strict precision and regulation to ensure the information stored on the DNA is not compromised during the process. But besides these connections and compaction it is becoming increasingly evident that these SMC complexes have much more to offer than previously thought with an increasing repertoire of functions being attributed to their involvement.

equally as important to the cell division process is the presence and timely severence of connections between sister chromatids. It is removal of interchromatid links which allows the cell to proceed to duplicate sucessfully.

appropriate inter and intra connection of chromosomal DNA also plays a vital role in a cell's ability to execute its functions and proliferate. These connections vary in strength in a spatiotemporal manner with cell cycle stages such as anaphase requiring complete severance between sister chromatids to allow for cell duplication.

New roles for cohesin and condensin are continuously being elucidated and the clinical significance of their dysfunction is also becoming evident with conditions such as cornelia de lange . These roles and dysfunctions will be considered later but first to really appreciate how these protein complexes work it is essential to have an understanding of the structure at a molecular level.

The backbone structure of Cohesin and Condensin

One of central themes in biology is that form fits function; therefore it is inherent that to gain a full understanding of the functional and mechanistic aspects of cohesin and condensin we must first understand the structure of their core units, Structural Maintenance of Chromosome (SMC) proteins.

The identification of the first SMC protein was achieved in 1985(larionov et al,1985), a time when the cell division cycle (cdc) was being dissected through mutants that disrupted it. One important event of the cdc which could be monitored was chromosome non-disjunction, a process where duplicated chromosomes fail to segregate properly during mitosis. Using this trait as a marker the isolation of associated mutants could be achieved. SMC1-1 (Stability of Mini Chromosomes) was one such mutant which when present in a recessive manner, resulted in failed segregation 10 times more than wild type strains. This suggested that the difference between the mutant and wild type was an important factor in the cell division process.

However it was a number of years until the initial genetic research on this mutant was undertaken. It consisted of sequencing this SMC1 gene product (Strunnikov AV et al, 1993) and comparing it on a protein database (GCG package). Although basic, this approach yielded some promising results such as the indication of a NTP-binding domain at the amino terminal and an aspartic acid/alanine rich (DA box) carboxyl terminal. It was also noted that two regions of the polypeptide, which accounted for >50% of the total length, formed coiled coil structures. The presence of these coiled coil structures near a NTP-binding domain was reminiscent of motor proteins such as myosin or kinesin and was therefore the initial mechanism of function proposed for SMC proteins. However this theory was suspect as although there were structural similarities, there was no homology with any known mechanochemical domain; so if they were motor proteins they would denote a new class.

A subsequent theory which emerged in 1994( Saitoh et al) was the idea that SMC proteins functioned as ATPases. This was suggested due to the presence of homologous domains between SMC proteins and the family of putative ATPases. The homologous domains are the previously discussed NTP-binding domain and the DA box domain which were seen similar to Walker A and Walker B sites, respectively. However, there was also some discrepancy to this theory as in ATPase's polypeptide structure the relevant domains lay next to each other in order to bind and hydrolyse ATP. This is not the case in SMC proteins with the domains located at the termini. Two models were proposed in 1994 on how these domains could become juxtaposed and thus allow for ATP processing. One suggested the interaction of SMC proteins in an antiparallel manner while the other explored the possibility of folding in the centre of the polypeptide.Folding in the centre of the polypeptide would result in antiparallel coiled-coils which were known of, but none of the length the SMC's would result in . ( fig)

It was 1998 before convincing evidence was presented on the correct model of SMC protein. (Melby et al,1998). By studying the conformation of the proteins using electron microscopy it could be seen that an SMC monomer folds back on itself forming antiparallel coiled-coil interactions. This results in the formation of a "hinge" domain at one end and an ATP binding domain at the other. The resulting monomer then dimerises through the hinge region forming either homodimers with identical monomers (prokaryotes) or heterodimers with different SMC monomers (eukaryotes) (Fig ?). The dynamic nature of the hinge mediated dimerisation was also noted under EM as it was revealed that the hinge region was very flexible and could vary from at least 180° in which the terminal domains were separated by 100nm to 0° where the domains could interact. This dynamism is responsible for many theories on the exact mechanism of cohesin and condensin function but before this is focused on it is important to discuss the initial work which identified these complexes

The Identification of the SMC complexes: Cohesin and Condensin

During the mid 90's while the exact structure of SMC's were still being uncoverd, sister chromatid separation and the condensation of metaphase chromosomes still fascinated researchers. Fundamental advances in understanding these important and visually striking events of cell division both occurred in 1997 through pioneering work.


When a cell prepares for mitosis it has the complex task of compacting metaphase chromosomes almost 10,000 times in length. This is a dramatic morphological change and is equivalent to shrinking a kilometre of DNA to 10cm. It is no mean feat and requires strict precision and regulation to ensure the information stored on the DNA is not compromised during the process. But how is this precision and regulation mediated? Although the specific details are still elusive, a major breakthrough came with the characterisation and purification of protein complexes termed condensin.

Condensins were discovered through innovative research that took advantage of cell free extracts derived from Xenopus laevis eggs (Hirano et al,1994) (Hirano et al,1997). This method provided a strong tool for studying biochemical actions of mitotic chromosome assembly as it recapitulates cell-cycle events in vitro. The advantages it had over the standard biochemical methods for analyzing mitotic chromosomes were key. First, using the cell free method allowed for a much simpler protein composition as embryonic chromosome assembly contained a minimal number of polypeptides. Second, this system also allowed a method of directly monitoring, in vitro, the role of proteins involved chromosome architecture.

Using this system mitotic chromosomes were isolated from Xenopus egg extracts by sedimentation(Hirano et al,1994). Protein components of the chromosomes subsequently obtained were then run on an SDS-polyacryamide gel. With the exception of histones, a set of polypeptides was found in the high molecular weight range. These polypetides were named XCAP- C and XCAP- E (Xenopus Chromosome Associated Protein C and E) and had a molecular weight of ~150kDa. It was observed that when these proteins were forcebly dissociated from chromosomes there was a loss of structure which led to the postulation that they were involved in maintaining higher order chromosome organisation. Supporting evidence to this was that antibodies which blocked XCAP-E and XCAP-C could prevent chromosome condensation and allow decondensation of condensed chromosomes (Hirano et al,1994). Sequence analysis revealed that these proteins were both ATPases belonging to the SMC family. This was an intriging discovery and with the emerging evidence that there was a gene homologous to XCAP-C present in C.elegans involved in dosage compensation (Chaung et al, 1994) interest in the XCAPs gained momentum.

To investigate the roles of XCAP-E and XCAP-C further, complexes containing these proteins were affinity purified and fractionated (Hirano et al, 1997). The results showed two peaks on a sucrose density gradient with Svedberg(S) values of 13S and 8S, respectively. While the 8S peak contained only XCAP-E and XCAP-C, the 13S peak possesed three extra proteins named in accordance with their molecular weight in kDa: P100,P130 and P150. This led to speculation that there was a pentamic complex involved. So to test this hypothesis, affinity purification was undertaken with anti-P130 and as expected only the 13S complex was isolated. This complex was able to induce to formation of mitotic chromosomes in Xenopus sperm and was thus termed condensin. The 8S could not induce this feature. As XCAP-C and XCAP-E were two of the most abundanteproteins which copurifiediwith mitotic chromosomes it was important to question what the role of the three new proteins were. Were they also targeted to chromosomes or did they act as molecular chaperones or loading complexes? This question was also addressed in this 1997 paper (Hirano et al, 1997) and it was concluded that P100, P130 and P150 were identical to previously uncharacterised XCAPs: XCAP-G,XCAP-D2 and XCAP-H and that they targeted to chromosomes. This was remarkable research carried out by Hirano with regard to chromosome organisation that shed light on how SMC proteins could associate with different subunits which is how they could be responsible for an array of chromsome dynamics.

Today thanks to much of the work carried out in the early experiments we know that condensin in eukaryotes has three additional subunits termed CAPs (CAP-G,CAP-D2 and CAP-H) in conjuction to the SMC subunits. CAP-H is a member of the kleisin family* while the other two subunits contain HEAT (Huntingtin, elongation factor 3 (EF3), protein phosphatase 2A (PP2A), and the yeast PI3-kinase TOR1) repeats. These repeats are suspected to facilitate protein-protein interactions and are found in numerous chromosomal proteins. The three non-SMC subunits of cohesin have been shown to have a role in modulating the ATPase activity of the complex (Kimura & Hirano,2000) but the details of this remain elusive.

It was assumed up until 2003 that only one form of the condensin complex existed in organsims but this idea changed when a database search using the hCAP-D2 (human CAP-D2) sequence uncovered a related protein, termed hCAP-D3. This related protein was shown to have an array of four HEAT repeats in the C-terminal region so an antiboby was raised against the C-terminal of this peptide and used for immunoprecipitation. The results showed that the putative protein affiliated with the SMC subunits of condensin (SMC4 and SMC2) and also with two unknown non-SMC proteins, named CAP-H2 and CAP-G2. CAP-H2 is another menber of the kleisin family while CAP-G2, which also possessed HEAT repeats, was closely related to CAP-G. This complex was named condensin II and siRNA targeted depletion resulted in severe morphological defects in chromosome assembly in both HeLa cells and Xenopus cell free extracts (Ono et al, 2003). While yeast do not have condesin II, higher eukaryotes possess both condensin I and II, though in different ratios depending on the organism. It seems as if more dramatic chromosome condensation, such as in metazoans, requires this extra condesin complex and therefore it has been suggested that organsisms with larger chromosomes use this complex for additional organisation.(Hirano, 2005).


The proper segregation of sister chromatids is one of the most important and dramatic events in a cell's life cycle. When DNA is replicated during the Synthesis phase of the cell cycle, the newly formed sister chromatids must be seperated to opposing poles of the cell before division can occur. But how is this seperation negotiated in order to allow cytokinesis and avoid disastrous events such as anueploidy? Key to these processes is the tug of war between the splitting forces exerted by microtubules and cohesion between sister chromatids. It is the loss of this cohesion which is responsible for the progression of a cell through anaphase. But how is this cohesion attained in the first place? And what is responsible for the timely dissociation of sister chromatids? These questions had puzzled researchers since Walther Flemming recognised mitosis over 130 years ago and it was only recently that a fundamental adnance was achieved through the elucidation of a protein complex termed cohesin.

Cohesins were uncovered in a similar fashion to how SMC proteins were first recognised; through utilising yeast cdc mutants in which the desired trait was disrupted. By the late 1980's it was known that catenation occurred between sister chromatids during replication. However this feature was deemed not responsible for the observed cohesion as it was seen that chromosomes within cells that were arrested in a mitotic state by the spindle assembly checkpoint were not seperate(Koshland & Hartwell, 1987). This led to the speculation that it must be something other than linkages between intertwining DNA that controls this sister-sister cohesion and that it may be mediated by "one or more interesting proteins" (Koshland & Hartwell, 1987).

The next piece of the puzzle to fall into place stemmed from research into B-type cyclin proteolysis, an event which takes place around the same time as sister chromatid seperation. This degradation of B-type cyclins relies upon their multiubiquitation, which in turn, depends on a large multisubunit complex shown essential for the seperation of sister chromatids and so was subsequently named the anaphase promoting complex (APC) (King et al,1995; Irniger et al, 1995). B-type cyclins were ruled out of having any direct effect on sister chromatid cohesion as mutants which overproduced nondegradable B-type cyclins continued to allow segragation at anaphase (Holloway et al, 1993).

As APC is required for sister chromatid cohesion the task was to find proteins which might be subtrates of this complex and test if they have a relationship to chromatid segregation. This was undertaken by isolating mutants which allowed segregation even in the absence of APC function. When the mutants were isolated and analysed a number of genes were tied to chromatid cohesion (Michaelis et al, 1997). One of the genes was SMC1, which as previously mentioned was the first SMC protein recognised (Strunnikov et al, 1993). Another was linked to an open reading frame which was uncharacterised but shown to be another member of the SMC family and therefore was termed SMC3. A further important gene coded for a protein which showed some relation to the double-strand-break repair protein, Rad21. Due to its neccisity in keeping sister chromatids united it was termed SCC1 (sister chromatid cohesion 1) .It was also noted that SCC1 dissociated at anaphase and could therefore be the trigger for chromatid seperation. From a more comphrehensive analysis of mutatants by the same group in 1999 a protein term SCC3 (sister chromatid cohesion 3) was found to bind SCC1 (Toth et al, 1999). Together SMC1, SMC3, SCC1, SCC3 form what is now called cohesin.

This systematic identification of genes involved in chromatid cohesion in S. cervisiae was a stepping stone for chromosomal dynamics and answered some fundamental questions with regard to cellular biology. But with these answers came more questions such as how do these complexes opperate at a mechanistic level?

Today pg528

Mechanisms of action

With the form of SMC complexes disscussed we can now concentrate on how they carry out their actions from a mechanistic point of view. Numerous models have been put forward as to how cohesin and condensin interact with chromosomes in order to carry out their functions.

The architecture of SMC proteins allows for a variety of models in which intermolecular and intramolecular interactions could be responsible for the diverse array of functions the complexes undertake. As the SMC's N and C terminals can interact there is the two types of engagement which could occur: one would see the formation of a closed ring (intramolecular) while the other could result in various structures such as a filament, double ring or complex rosettes (FIG). The coiled-coil arms of SMC's may also interact in an ATP-independent manner which would also allow an array of structures.

With these possible interactions availible to SMC proteins it would not be surprising if cohesin and condensin did not share the same mechanism of action since they have some structural and substantial functional difference.

Condensin: mechanism of action

Early research on condesin demonstrated that the complex in Xenopus egg extracts displays DNA- stimulated ATPase activity. Using functional assays it was also deduced that the complex induces supercoiling of double stranded DNA in a ATP-hydrolysis-dependent manner. This supercoiling is only possible by the holocomplex and requires the phosphorylation of the three non-SMC subunits (Stray et al, 2005).

As already mentioned the SMC subunits of condensin are ATPases, which is a vital feature for proper function as shown from targeted mutagenesis of all known SMC2 ATPase domains (Hudson et al, 2008) . This study also showed that it is the binding of ATP but not the hydrolysis that was neccesary for the initial association between condensin and mitotic chromosomes in vivo. Analagous mutations in the cohesin complex result in failed binding to chromatin (Arumugam et al, 2003) highlighting how the different complexes while related may use different mechanisms to function.

One model proposes the mode of action was that when ATP bound condensin interacts with double stranded DNA it initiates ATP hydrolysis and thus opens the arms allowing for more stable chromatin binding. Following this two scenerios were presented in which the formation of superhelical tension or chiral loops could be formed depending on whether there was intramolecular or intermolecular interactions. These structure then be further organised by coiling and stacking resulting in the metaphase chromatid (FIG) (Hirano, 2006).

Contradicting evidence to stage3' in FIG is strong as electron microscopy performed in 2002 revealed that while cohesin could be seen to form open circle structures the predominant form of condensin was observed as being a lollipop-like structure with the arms engaged with each other in the middle (Anderson et al, 2002). Another recent study in which a biologically active but cleavable form of condensin was used to investigate the mechanism of action revealed that the complex remained intact despite disruption by SMC2 cleavage. This supports the lollipop-like structure hypothesis and since the SMC2 cleavage did not result in dissociation of the complex from chromosomes it suggests that it does not form a ring around the DNA (Hudson et al, 2008). This leaves the intermolecular model which is the strongest to date and relies on the multimerisation of condensin complexes but further genetic and biochemical research will have to be conducted before a universally accepted model of condensin mechanisms of action is achieved.

Cohesin: mechanism of action

Cohesins are essential for the linking of sister DNA duplexes and the way in which they achieve this is of great interest. Knowledge on the structure of cohesin has greatly aided the efforts to propose models for mechanism of its actions. As discussed we know from electron microscopy and biochemical studies that the SMC1 and SMC3 proteins dimerise through the hinge domains forming a V shaped dimmer. The discovery that the SCC1 protein mediates cohesion between sister chromatids through binding to the ATPase domains of each SMC subunit has led to the conclusion that cohesin forms a tripartite ring. This ring has been postulated to encapsulate the chromatin fibres holding them together and only releasing them with the cleavage of SCC1 by a cysteine protease termed separase. Once released microtubules can pull each sister to opposite spindle poles. Consistent with this ring idea, an experiment in which cohesin subunits were forcibly cleaved in vivo caused the loss of sister-sister cohesion while all the subunit interactions remained intact (Gruber et al, 2003).

Despite the ring model being simple in principle, it is difficult to test as biochemical analysis of the purified cohesin complex has not proceeded with the same ease as condensin. However progression has been made on how cohesin initially interacts with chromatids and it is now thought that the complex attaches to chromatin via a loading complex composed of two subunits, SCC2 and SCC4 (Giosk et al, 2000). After the loading onto chromatin there are two types of model on how the cohesin ring can tie the sister chromatids together; a strong or a weak ring. The strong version postulates that a single cohesin ring encompasses the two sister DNA strands where as the weak version presents the idea that two different rings interact, each one having trapped a sister chromatid. (Fig)

An argument against the ring models of DNA trapping is mainly topological; as it cannot account the results which show that the complex localises at specific sites in the genome (Huang et al, 2005). Although this is a valid criticism in the sense that the ring model does not explain these findings, it cannot be said that entrapment precludes site-specific localisation. One explanation for this sequence localisation could be that the rings move along the DNA until they come into contact with a sequence-specific DNA binding protein which they have an affinity for. Another explanation may be that the specific localisation is due directly to the action of the loading complex.

Uncovering the true mechanism of action for each complex, if indeed there is only one, will advance our understanding of cohesin and condensin especially with recent research linking them to an increasing number of nuclear functions. These newly characterised roles for cohesin and condensin are in diverse areas and so it is important to discuss them in order to understand the real grasp these complexes have at a molecular level.

Condensin in gene expression

The first evidence of

The Walker A and Walker B motifs are nucleotide binding domains located at the amino and carboxy terminals, repectively. In the centre of the polypeptide exists the hinge domain which is connected to each terminal via coiled coil motifs. There were two main models postulated in the mid 1990's (Saitoh et al, 1994) as to how the SMC could fold up and

at the hinge region to form an antiparallel coiled coil structure. However this was speculative as there was no known example of such a long antiparallel coiled coil It was first pformed by two long opposite ends of the polypeptide w the possibility there existed antiparallel folding of the middle coiled coil region to allow for the close proximity needed for terminal end interaction (Saitoh et al, 1994)

These five domains ultimately interact to form three separate regions: the head, hinge and arm..(fig).

Speculation of how exactly SMC proteins interact and form dates back t these of this This folding mechainism was elucidated by a series of biochemical experiments, with the most convincing evidence obtained from electron microscopy (Melby TE et al, 1998).

which revealed that the hinge region was very flexible and could vary from at least 180° in which the terminal domains were separated by 100nm to almost 0° where they were together (Melby TE et al, 1998). It was also this paper that the speculation of antiparallel coiled-coil arms (Saitoh et al, 1994) was strenghtened by clever manipulation of the NH2 domain which saw the addition of a small rod shaped molecule. This allowed for the distinction of whether it was parallel or anti-parallel coiled coils as if it was parallel there would be two protrusions on the same visible under EM

In the same way as the elucidation of the structure of DNA was instrumental in understanding its function, uncovering the architecture of the backbone of SMC complexes has revealed much


Cohesin and Codensin in Interphase genome organisation

Although The names "condensin" and "cohesin" were origanlly given to to the SMC proteins and their complexes because of their function in DNA condensation and sister chromatid cohesion it is now clear that a myriad of functions is undertaken by these complexes and that more may still be uncovered

They were both discovered independently in 1997 as pivotal players in the cell division process mitosis.


And therefore I have cured cancer.