Mechanisms behind Smc5/6 transcriptional regulation and cohesin-mediated chromosome organisation

The eukaryotic family of Structural Maintenance of Chromosomes (SMC) complexes are multi-subunit, ring-like protein complexes and include condensin, cohesin and the Smc5/6 complex (Smc5/6). Considered as an entity, this family is essential for most chromosome-based processes, and protects against diverse human syndromes and diseases. Condensin’s main role is to compact mitotic chromosomes, while cohesin creates sister chromatid cohesion, i.e., holds sister chromatids together from their formation during S-phase until their separation in anaphase. In contrast, the main role of Smc5/6 is currently unknown. The Smc5/6 complex is involved in DNA repair, chromosome replication, segregation and telomere maintenance but the mechanism behind these activities remains poorly understood. Previous studies using the budding yeast Saccharomyces cerevisiae show that Smc5/6 colocalises with cohesin between convergently transcribed genes on replicated chromosomes and that this association depends on cohesin and sister chromatid cohesion. However, while cohesin is evenly distributed along chromosomes, Smc5/6 binding positively correlates with increasing chromosome length, and augments after inhibition of the Top2 topoisomerase, suggesting a connection to DNA supercoiling and/or chromatid entanglements. Recently, Smc5/6 was also shown to specifically inhibit transcription of small circular viral genomes and plasmids, demonstrating that Smc5/6 suppresses transcription. How Smc5/6 recognises the viral genome and plasmids and inhibits transcription is currently not established.

Knowing that highly expressed convergent genes generate high levels of supercoiling, we used a previously created system that allows induction of transcriptional superhelical stress. Using this system, we have shown that induction of high transcription from a convergently oriented gene pair induces Smc5/6 association to the intervening intergenic region. The association of Smc5/6 to this site is cohesin- and DNA-damage independent, and is most likely driven by accumulation of transcription-induced DNA supercoiling alone. To gain insight into Smc5/6 transcriptional function, we depleted Smc5/6 by auxin-dependent protein degradation, and found that this leads to an increase of newly-induced transcription. Smc5/6 removal, however, has no effect on constitutively active genes, thereby recapitulating the reported effects on gene expression from newly-deposited viruses. The increased transcription detected after Smc5/6-depletion is strongly reduced by Top2 inhibition, suggesting that Smc5/6 negatively regulates Top2 function. Together, our data indicate that Smc5/6-mediated repression of viral transcription depends on the accumulation of the complex on positively supercoiled DNA. There, Smc5/6 represses de novo transcription, by preventing supercoil release by Top2. Altogether, the findings presented in project I provide new information on Smc5/6 transcriptional function, and show that our system can be used to examine how Smc5/6 antagonises viral infections. Apart from its role in sister chromatid cohesion, recent in vitro studies showed that the cohesin complex is also able to gradually enlarge DNA loops, that is, to perform loop extrusion. This process is essential for mitotic chromosomes compaction, and folds the interphase genome into regions that preferentially interact, called topologically associated domains (TADs). Functionally, TADs are suggested to regulate transcription by facilitating or inhibiting interactions between enhancers and promoters. TADs formation likely occurs by extruding DNA through the ring-like structure of cohesin and TAD boundaries are often defined by cohesin chromosomal binding sites. Previous reports suggest that transcription is able to push cohesin along chromosomes, and that cohesin associates with replication forks. However, if transcription and replication regulate loop-extruding cohesin and if they have any active role in chromosome folding, remains unclear. This was examined in project II, where the transcription inhibitor thiolutin was used to evaluate the effect on cohesin loop extrusion. This established that transcription removal increases the length of DNA loops and diminishes the boundaries of TAD-like structures in yeast. Moreover, highly induced stress genes also create new cohesin-mediated loop boundaries. In addition, treatment with hydroxyurea, which leads to stalled replication forks, known to be bound by cohesin, also creates boundaries for loop-extruding cohesin.

Together, our results show that both transcription and replication can act as barriers of loop-extruding cohesin and restrict the size of the cohesin-mediated loops. Together, this sheds additional light in the relationship between the regulation of loop extrusion, chromosome folding, transcription and replication, improving our knowledge of the three-dimensional organisation of chromosomes in the nucleus. In project III, the budding yeast Saccharomyces cerevisiae was used to examine and provide insight into the bacterial toxin ExoS from the human pathogen Pseudomonas aeruginosa.2022