BSCB Newsletter, Winter 2011

abcam: Chromatin, Replication and Chromosomal Stability
20-21 June 2011. Werner Gren Centre, Stockholm, Sweden.

The second abcam: Chromatin, Replication and Chromosomal Stability was held in June in Stockholm, organised by Anja Groth (University of Copenhagen), Catherine Green (University of Cambridge) and Camilla Sjögren (Karolinska Institute), following the previous successful meeting in 2009 in Copenhagen.

I was fortunately able to attend this meeting through BSCB Honor Fell Travel Funding, and amazingly my work was selected for oral presentation; my first talk at a conference.

The conference kicked off with a fascinating broad ranging talk from one of the keynote speakers, Helen Blau (Stanford University), who highlighted the importance of a correct demethylation programme during reprogramming and the effect of 'stiffness' on the regenerative ability of Muscle Stem Cells (MuSCs). Although reprogramming of cells can also be achieved through iPS cell generation or nuclear transfer, her lab uses the method of cell fusion to investigate the mechanisms involved during reprogramming, specifically those of DNA demethylation, an 'epigenetic bottleneck'. She described their discovery of Activation Induced Cytidine Deaminase (AID) expression in heterokaryons, and the subsequent elucidation that demethylation during reprogramming is achieved through nucleotide base replacement rather than direct demethylation, as was previously thought. Prof. Blau ended the talk with an example of how much a cell's environment can alter its phenotype. She used a Goldilocks analogy to describe how MuSCs grown on the much too 'stiff' tissue culture plastic are unable to regenerate in vivo but those grown on a 'comfy bed' of PolyEthyleneGlycol (the same stiffness as muscle), have a greater capacity for regeneration, thereby conveying some of the previously ignored 3D requirements of cells.

The first session covered Replication, Chromosome Structure and Cellular Memory and I thankfully had an early talk slot. This meant I would be able to concentrate fully on the later talks rather than worrying about my own. Speaking in a session with well-known cell cycle personalities was rather intimidating however, my talk went well and I was able to speak with several people in breaks who asked interesting questions and offered helpful suggestions. Although it was a bit scary, I would definitely recommend pushing yourself as a PhD student and trying to get an opportunity to talk about your work.

From this session I found Marcel Méchali's (Institute of Human Genetics, CNRS) talk particularly interesting. He described some of the features his group are finding in higher eukaryotic DNA replication origins, by mining a large data set. It has long been known that higher eukaryotes, unlike yeast, do not have a strict DNA sequence that specifies origins. Work from his group showed that origins are enriched just before or after transcription start sites but not at the site itself, and that there is in fact some sequence impact, with origins having a TG/CA bias. He also described their 'flexible replicon model' where 4-5 origins are grouped in a replicon from which one origin will be stochastically activated and then silence the others in that replicon. The question everyone wanted to know when I spoke to him after was, how does this happen? I hope we soon find out!

After lunch, and meeting and chatting with various people, the second session on 'Chromatin Replication and Histone Dynamics' started. It covered topics from a potential histone modification-based therapeutic against Candida infections, to how stalled replication forks are resolved. I particularly enjoyed the talk from Patrick Varga-Weisz (Babraham Institute) telling us about how pericentromeric and centromeric boundaries are maintained. The heterochromatin found at these points has a complex combination of specific histone modifications and recruitment of additional proteins. The correct disruption and then reassembly of these structures must be undertaken during each cell cycle and this maintenance is critical for genomic stability and chromosome segregation. Varga-Weisz described the work from his lab on the role of the chromatin remodeler SMARCAD1 during this process, which appears to complement those roles performed by histone modifying enzymes such as Histone Deacetylases (HDACs) and Histone Methyltransferases (HMTs). SMARCAD1 interacts with many proteins, including some involved in DNA replication, repair, silencing and heterochromatin maintenance, and localises to pericentromeric heterochromatin during late S-phase when it is replicated. Although initially found to be required for ES cell maintenance, the knock out mouse generated was viable with only a small number of defects. This paradox is undergoing further investigation. However, it was clear that depletion of SMARCAD1 by siRNA led to a global increase in euchromatin marks and corresponding decrease in heterochromatin marks through apparent failure to correctly recruit histone modifying proteins which interact with SMARCAD1; and led to an increase in mitotic defects. This was the first of several fascinating talks on how chromatin is faithfully maintained after DNA replication.

On the Tuesday, session three covered 'Initiation, Timing and Epigenetic States'. The first talk by David Gilbert (Florida State University) made sure we were awake and had brains engaged as he discussed replication timing. In order to investigate the importance of when particular regions of the genome are replicated (early, mid or late S-phase) his lab has produced genome wide profiles for replication-timing from a wide number of cell lines, cells at different stages of differentiation and also for various human pathological conditions. Changes in replication-timing can affect half of the genome but surprisingly, correlated only slightly with transcriptional status and epigenetic marks. The factor that correlated strongest was long-range chromatin interactions suggesting importance of spatial organisation. Using the huge change in global replication timing between the early and late epiblast (which is not accompanied by a significant change in the transcriptional programme) work from his group showed that most genes which change their replication timing at this transition, move from early to late replication, and are linked to increased compaction of chromatin. Echoing the model of DNA fractal globules by Erez Lieberman-Aiden and Nynke van Berkum, early and late replicating genomic regions appear to segregate, with like associating with like. Consistent with this, regions of the genome that change in timing of replication are of the size 400-800kb suggesting this is the domain size for a region of the genome which is replicated at the same time. Interestingly the genes which change replication timing status correlated with genes which are difficult to reprogramme, clearly impacting on attempts to improve iPS efficiency and further highlighting the importance of spatial organisation.

The final session was entitled 'Replisome Structure, Fork Progression and Repair' and covered some of the recent data about a wide range of mechanisms involved in maintaining faithful DNA replication. I found the talk on how DNA replication machinery deals with the predicted bulky DNA tertiary structure at G-quadruplex (G4) motifs by Virginia Zakian (Princetown University) particularly fascinating. She presented work that replication through large protein complexes or DNA structures such as G-quadruplexes are facilitated by the helicases Rrm3 and Pif1 (S. cerevisiae) (or related Pfh1 in S. pombe and Pif1 in H. sapiens). From genome wide ChIP, ~25% of G4 motifs were bound by Pif1 and DNA replication dramatically slowed in and around these regions in pif1 mutated cells. Startlingly, knocking down Pif1 by RNAi also led to a huge mutation rate at these sites, 20% of Gs became mutated and 97% of these sites were no longer predicted to form G4 structures. The predicted G4 structures therefore do appear to form in vivo and be resolved by Pif1 to prevent them causing problems for DNA replication machinery and subsequent fork stalling, breakage and mutations

For the last talk of the day, the second keynote speaker, Michael O'Donnell (Rockefeller University and HHMI), gave us a different perspective, focussing on the bacterial replisome. It was fascinating to hear the story of the third polymerase, about the flexibility of polymerases and how the replisome varies its composition as required. In vitro di-polymerase and tri-polymerase replisomes have similar rates for DNA synthesis but the three polymerase version has several advantages. Firstly the processivity is much greater due to more contact with the lagging strand, also no gaps are left on the lagging strand unlike those seen when only two polymerases are permitted. The specific polymerases found in the replisome however, varied dramatically, with pol III found under normal conditions but replaced by pol II and pol IV during times of DNA damage. These alternative polymerases slow the helicase dramatically and are stable, presumably allowing time for DNA repair. It is sometimes too easy to ignore bacteria within the cell biology field, but this definitely showed how much we can learn about mechanism from bacteria.

As well as attending this illuminating conference and meeting other scientists from across the world I also managed to have a look around Stockholm. The city is beautiful, spread across 14 islands, so there are boats and bridges everywhere; not without reason is it known as the Venice of the North. I also saw some of the distinctive, colourful and very pretty wooden houses on the 13,000 islands of the archipelago and of course went to the Nobel Museum and saw one of the famous medals!

I would like to thank the organisers for a fantastic conference and also the BSCB for their generous funding which allowed me to attend this stimulating event.

Rosemary H C Wilson,
University of York