Single-molecule live cell imaging of the Smc5/6 DNA repair complex

Background:

Our chromosomes (linear DNA molecules) are arranged and folded into a small confined space; the nucleus. How chromosomes fold, their placement in the nucleus and their ability to interact with each other (collectively their ‘higher-order’ structure) all contribute to modulating gene expression. In Eukaryotes, several complexes act in nuclear chromosome organisation including four Structural Maintenance of Chromosomes (SMC) complexes; the dosage dependent complex, cohesin, condensin and the Smc5/6 complex. The Smc5/6 complex, whose diverse roles include associating with double strand breaks (DSB) to assist homologous recombination (HR) repair, telomeres, the ribosomal DNA (rDNA) array and replication forks, is composed of Smc5 and Smc6 (1). Individually, both proteins look like long ‘arms’ meeting at a circular doughnut shaped region (‘hinge’) at one end (thought to interact with DNA) and a globular region at the other end which contains ATPase activity; little is known about the role of ATPase activity in the Smc5/6 complex. At the globular ATPase region, a sub-complex can interact to support Smc5/6 functionality: Kleisin (Nse4) and two KITE proteins (Nse1 and Nse3). Other Nse proteins also interact, including Nse2 (interacts with Smc5’s arm) and Nse5 and Nse6 (in yeast) (2). Despite Smc5/6 having important chromatin-based roles, much of the dynamics of its interaction with the chromatin remains unknown. Here, the authors used sophisticated microscopy and Smc5/6 mutants to understand these reaction dynamics.

 

Key Findings: 

1. PALM microscopy can be used to track SMC complexes

To understand dynamic interactions between Smc5/6 and the chromatin, the authors performed single molecule tracking using photoactivated Localisation Microscopy (PALM) in live cells, which allows the capture of precise spatial information regarding the origin of the fluorescent signal (3). Briefly, when the sample is pulsed stochastically with an ultraviolet (UV) laser, the photoconvertible fluorophore mEos3 will ‘photoconvert’ from a green fluorescent signal to a red fluorescent signal. This process is repeated until all the mEos3 molecules have converted. Stochastic photoconversion of a small number of molecules at any one time means each fluorophore emission can be detected, the centre point of each ‘spot’ calculated and the information used to pinpoint the signal origin, within ~ 20nm. Here the authors fused mEos3 to the kleisin subunits of each SMC complex. Using S. pombe condensin kleisin (Cnd2; 4), as a proof-of-principle to monitor SMC complex behaviour at the chromatin in live cells and newly developed software ‘Spot-On’ (5) they tracked single molecules of Cnd2 in the nucleus and on the chromatin. Molecules bound to chromatin diffuse less than unbound molecules. This approach is known as single-particle tracking PALM (sptPALM).

Figure 1. Fig. 1A illustrates the core composition of fission yeast SMC complexes. Fig. 2B the fraction of bound molecules of each protein to the chromatin (measured by sptPALM using the ‘Spot On’ software) for both the Smc5/6 (Nse4-mEos3) and the Cohesin complex (Rad21-mEos3). Fig. 2C sptPALM images showing the locations of Nse4-mEos3 and Rad21-mEos3 single-molecules in the nucleus (white dashed lines). Fig. 3D the fraction of bound molecules of each protein to the chromatin (measured by sptPALM using the ‘Spot On’ software) for Smc5/6 (Nse4-mEos3) in wild type cells and in cells lacking either Brc1 or Nse6. Figures made available under a CC-BY-NC-ND 4.0 International license.

2. Cohesin and Smc5/6 interact differently and distinctly with chromatin

Using the above approach, the authors investigated the association of cohesin and Smc5/6 to the chromatin by fusing Nse4 (for Smc5/6) and Rad21 (for cohesin) to mEos3. Using their modelling software, the show Smc5/6 and cohesin interact differently with the chromatin, with cohesin binding more to the chromatin than Smc5/6 and the cohesin complex mostly formed nuclear foci where as Smc5/6 formed foci and showed a diffuse localisation.

 

3. Nse6 likely loads or stabilises Smc5/6 on chromatin

To ask what factors might recruit Smc5/6 to the chromatin, the authors investigated two factors: Nse6 and Brc1. First, using sptPALM, they show Nse6 likely acts to load or stabilise the Smc5/6 complex. Next, they ask whether Nse6 or Brc1 are needed to localise Smc5/6 on the chromatin by deleting Nse6 or Brc1 in the Nse4-mEos3 cell line. Deletion of either gene disrupted the wild-type localisation of Nse4 with both genes resulting in a loss of Smc5/6 association to the chromatin. However, this loss was greatest when Nse6 was deleted supporting its role as a loader or stabiliser of Smc5/6. When cells were exposed to a genotoxin (methyl methanesulfonate; MMS) in wild-type cells, Smc5/6 was found to associate to chromatin, whereas in both deletion mutants, no chromatin association was detected (though this was independent of MMS treatment) supporting the role of Smc5/6 being loaded in response to DNA damage.

 

4. Smc5/6 needs ATPase activity to fully associate with chromatin

Next, the authors investigated if the ATPase activity of Smc5/6 was needed for chromatin association. By disrupting the ‘arginine-finger’ region needed for ATP interactions in either Smc5 or Smc6 in the Nse4-mEos3 cell line then performing sptPALM, they reveal 1) both mutants were sensitive to replication stress exposure, with the Smc6 mutant more so, 2) both mutants show less association to the chromatin with the Smc6 mutant showing a more extreme phenotype, 3) both mutants were largely unable to recruit Smc5/6 to the chromatin in the presence of MMS treatment. In all, the ATPase activity is needed for Smc5/6 recruitment to chromatin and individually, Smc5 and Smc6 show differences in their ATP binding capacity.

 

5. Smc5/6 interacts with ssDNA to stop chromosomes rearranging

The Smc5/6 complex can bind both ds- and ss-DNA. By reducing the ability of their Nse4-mEos3 cell line to bind dsDNA (by disrupting Nse3), they reveal the Nse3-dsDNA interaction is likely needed for Smc5/6 recruitment to chromatin. Next, they asked if the same was true for ssDNA. By mutating regions in Smc5 and Smc6 needed for ssDNA binding, they show Smc5/6-ssDNA interactions are not needed for Smc5/6 to be recruited to chromatin but under DNA damage conditions (via MMS exposure), these Smc5/6 ssDNA-binding mutants were largely unable to associate with chromatin. This suggest ssDNA may ‘keep’ Smc5/6 on the chromatin during a damage event and to bring in other factors after Smc5/6 loading to regulate repair. Next, the authors asked if Smc5/6 doesn’t need ssDNA for recruitment to chromatin, then why may it need to interact with ssDNA? Using their mutants and combining them with a previously described system in yeast for monitoring homologous recombination (HR) dynamics they investigate this. In short, when the assay is induced, an arrested replication fork occurs within a defined genomic location. Replication can only then continue once another region (ura4) is used to restart replication. This reaction is HR dependent and can result in genomic instability due to large scale genomic rearrangements (i.e a non-homologous allele is used for restart) or when the re-started replication fork makes errors as it moves. Finally, if HR fails to occur, the cells are non-viable. They reveal, the ssDNA-binding deficient mutant presented with chromosome rearrangements indicating Smc5/6 may bind ssDNA to stop these rearrangements from occurring thereby regulating HR rather than controlling it.

 

What I liked about this preprint:

I really enjoyed the use of sptPALM by the authors. By focusing on one technique and using it to address several questions, they reveal important findings about how Smc5/6 interacts with chromatin, including under damaging conditions. Here, they have really shown the resolution power of PALM and how it can be applied to successfully solve protein dynamics in live cells.

 

Questions for the authors:

1. Have you performed sptPALM on Brc1 to localise this protein in the cell?
2. Smc5/6 may also have important roles during S-phase. Do you plan to study the dynamics of its interaction by sptPALM within this particular stage?
3. Here you have used MMS to cause DNA damage. Would you expect similar dynamics of Smc5/6 following exposure to, for example, a replication poison like Hydroxyurea, which would result in ssDNA accumulation and the stalling of DNA replication? Or another genotoxin such as phleomycin, which induces DSBs?
4. You mention a possible pathway independent to Brc1 due to the weaker effect on chromatin association observed compared to Nse6 deletion. Do you think it is possible that Nse6 could instead compensate for the loss of Brc1 rather than a separate pathway operating?

 

References:

1. Aragon, L. The Smc5/6 Complex: New and Old Functions of the Enigmatic Long-Distance Relative. Ann. Rev. Gen. 52 (2018).
2. Adamus, A., Lelkes, E., Potesil, D., Ganji, S.R., Kolesar, P., Zabrady, K., Zdrahal, Z., and Palecek, J.J. Molecular insights into the architecture of the human Smc5/6 complex. JMB 13 (2020).
3. Sydor, A.M., Czymmek, K.J., Puchner, E.M., and Mennella, V. Super-Resolution microscopy: from single molecules to supramolecular assemblies. Trends in Cell. Bio. 25 (2015).
4. Sutani, T., Yuasa, T., Tomonaga, T., Dohmae, N., Takio, J., and Yanagida, M. Fission yeast condensing complex: essential roles of non-SMC subunits for condensation and Cdc2 phosphorylation of Cut3/SMC3. Genes Dev. 13 (1999).
5. Hansen, A.S., Woringer, M., Grimm, J.B., Lavis, L.D., Tjian, R., and Darzacq, X. Robust model-based analysis of single particle tracking experiments with Spot-On. eLife (2018).

Tags: chromatin, dna repair, microscopy, palm, single-molecule, smc, smc5/6

doi: https://doi.org/10.1242/prelights.22354

Read preprint (No Ratings Yet)