SUMO proteomics in C. elegans: focus on meiosis and the DNA damage response

Project Information

CeleSUMOMS

Grant agreement ID: 297881

Closed project

Start date 17 September 2012

End date 16 September 2014

Funded under

• Specific programme "People" implementing the Seventh Framework Programme of the European Community for research, technological development and demonstration activities (2007 to 2013)

Total cost € 209 033,40

EU contribution € 209 033,40

Coordinated by

UNIVERSITY OF DUNDEE
United Kingdom

Final Report Summary - CELESUMOMS (SUMO proteomics in C. elegans: focus on meiosis and the DNA damage response)

SUMMARY

Small ubiquitin-related modifier (SUMO) conjugation is essential for development in mammals (1, 2) and in the nematode Caenorhabditis elegans (C. elegans) (3-6). Post-translational protein modifications including phosphorylation, ubiquitylation, and sumoylation are essential for mitotic progression. Early data indicated that SUMO is involved in cell cycle progression. SUMO was initially characterized as a suppressor of a mutation in the gene encoding the centromeric protein MIF2 (7), and the SUMO conjugating enzyme Ubc9 as well as the SUMO protease Ulp1 regulate cell cycle progression in yeast (8, 9). Several studies showed essential roles for sumoylation in controlling chromosome condensation and cohesion, kinetochore assembly and function, and spindle dynamics (10-22). The C. elegans embryo is a powerful model system for studying metazoan cell division and it has provided important mechanistic insights into cell cycle progression, particularly related to kinetochore function (23). We took advantage of C. elegans to establish the contribution of SUMO to a timely and accurate cell division, using the first mitotic embryonic division. SUMO conjugates accumulate on chromosomes during metaphase and decrease rapidly during anaphase. The accumulation of SUMO on the metaphase plate and proper chromosome alignment depend on the SUMO E2 conjugating enzyme UBC-9 and SUMO E3 ligase GEI-17, ortholog of the Siz/PIAS family of E3 ligases. Deconjugation is achieved by the SUMO protease ULP-4 and is crucial for correct progression through the cell cycle. Our results show that highly regulated and dynamic SUMO conjugation plays a major role in cell cycle progression.

RESULTS I: SUMO regulates cell cycle in Caenorhabditis elegans embryos

1) Knockdown of sumoylation pathway components affects chromosome dynamics. Using live cell imaging and the power of RNAi-mediated gene knockdown, we established that the sumoylation pathway regulates chromosome dynamics in C. elegans (Figure 1).
2) SUMO exhibits a distinct localisation pattern during mitosis. Using a combination of live cell imaging, fixed immunostaining, and the power of RNAi-mediated gene knockdown, we established that SUMO conjugates accumulate during metaphase for ~50 seconds. These results highlight the dynamic nature of SUMO localisation during the first embryonic cell cycle in C. elegans (Figure 2).
3) SUMO conjugation is rapidly and dynamically regulated during mitosis. To establish if SUMO staining depends on actual conjugation, we knocked down the SUMO E2 ubc-9. This led to the complete loss of the mCherry-SUMO signal at metaphase chromosomes but not within pronuclei or nuclei. Importantly, knockdown of ubc-9 also abolished endogenous SUMO staining at metaphase (Figure 3).
4) SUMO conjugation is mediated by the SUMO E3 ligase GEI-17PIAS. Driven by the precisely timed appearance of SUMO conjugates that takes place during mitosis, we turned our attention to putative SUMO E3 ligases. We focused in the PIAS ortholog, GEI-17, and the component of the SMC-5/6 complex, MMS-21. Altogether, SUMO conjugates accumulate on chromatin during metaphase in a manner dependent on not only UBC-9 but also on the E3 ligase GEI-17 (Figure 4).
5) SUMO is deconjugated by the SUMO protease ULP-4SENP6/7. To analyse the role of individual SUMO proteases in the accumulation and removal of SUMO from mitotic chromosomes, the expression of each of the four proteases was inhibited by RNAi and the chromatin association of mCherry-SUMO conjugates was followed by time-lapse microscopy. SUMO was less efficiently removed from its association with chromatin after depletion of ulp-4 between during anaphase. The ULP-4 catalytic domain is most closely related to those in the mammalian chain-editing enzymes SENP6/7 and, like SENP6/7, ULP-4 actively depolymerises SUMO chains in vitro and, like SENP6/7, was unable to process immature SUMO. ULP-4 localises around the metaphase plate and the pericentriolar region and in the surroundings of the central spindle (Figure 5).
6) SUMO conjugation affects AIR-2 localisation. We based our search for putative SUMO substrates on the available data for mitotic SUMO substrates or to proteins exhibiting a similar pattern of localisation. A strong candidate was the Aurora B ortholog air-2 (24-27). Analysis of endogenous SUMO and AIR-2 by immunostaining showed that they co-localise perfectly on aligned metaphase chromosomes and at anaphase onset. While AIR-2 localised to DNA during prometaphase and metaphase and then translocated to the spindle midzone, SUMO co-localised with AIR-2 when chromosomes are aligned on the metaphase plate and in early anaphase. Some degree of co-localisation was also observed in the spindle mid-zone in late anaphase/telophase. AIR-2 translocation to the spindle mid-zone coincided with the loss of SUMO staining during anaphase. To further characterise the SMO-1/AIR-2 interaction, we performed proximity ligation assays (PLA) (28). PLA is a technology that extends the capabilities of traditional immunoassays to include direct detection of proteins, protein interactions and modifications with high specificity and sensitivity, allowing detection of interaction distances of as little as ~30 nm (28, 29). A strong and specific PLA signal was detected within the metaphase plate (Fig. 5d). Interestingly, the PLA signal is lost during anaphase and is recovered in late anaphase/telophase within the spindle midzone. These results show that SUMO and AIR-2 reside in close proximity specifically on chromosomes during metaphase and on the central spindle during anaphase. In the absence of AIR-2, the SUMO signal associated with metaphase chromosomes was diminished by 80%. Depletion of ulp-4 prevented AIR-2 from localasing to the spindle mid-zone. While ulp-4 depletion impaired the localisation of AIR-2 in the spindle, depletion of ubc-9 increased AIR-2 levels in the spindle midzone.
7) AIR-2 is modified by SUMO in vitro in a GEI-17-dependent manner.
In vitro conjugation reactions showed that AIR-2, like its ortholog Aurora B, is modified by SUMO and the modification is stimulated by GEI-17 in a dose-response manner. Moreover, the L/A mutation of the SP-RING within GEI-17 drastically diminishes its SUMO E3 activity. The AIR-2 ortholog in mammals, Aurora B, was previously shown to be modified by SUMO at Lys 207 in mice and Lys 202 in humans (30, 31). As shown in Figure 6, not only did we detect SUMO conjugation at Lys 155 (equivalent to Lys 202/207 in Aurora B (30, 31)), but also at Lys 168 (equivalent to Lys 215/220 in Aurora B). Mutation of the two sites decreases the GEI-17 stimulated conjugation by ≥ 75% (Fig. 6e, ‘K155/168R’ vs. ‘wild type’). This putative new SUMO conjugation site resides within an inverted SUMO consensus motif (32).

RESULTS II: SUMO localisation during Caenorhabditis elegans meiotic progression.

We have found that SMO-1 concentrates to a very discrete region between the homologous chromosomes at metaphase of meiosis I (Figure 7). Then, SMO-1 concentrates in the mid-region once again in metaphase II. Live imaging experiments have been already set up as described (33). A montage of selected frames is shown for mCherry-SMO-1 expressing embryo from meiosis I to the end of the first mitosis (Fig 1D), showing that the exogenous protein behaves as the endogenous one. Importantly, knocking down smo-1 or ubc-9 lead to chromosome alignment and segregation defects (Figure 7).

RESULTS III: Proteomic identification of sumoylation sites in Caenorhabditis elegans.

We have generated C. elegans strains expressing His-tagged processed SMO-1(GG) under the control of its own promoter. I have engineered a processed SMO-1 mutant amenable for site identification by mass spectrometry: substituting Leu 88 for Lys (SMO-1 L88K, GG, Figure 8). This substitution is important as it enables distinction between ubiquitin and SUMO by the choice of the protease used for digestion: since the C-terminus of the mutant SUMO is KGG, cleavage with LysC leaves the GG remnant without affecting ubiquitin, whose C-terminus is RGG. Importantly, I have already shown that the L88K mutation behaves like the wild type protein in in vitro conjugation reactions (data not shown) and results from the Hay lab show this is also the case for human SUMO-2. The first set of experiments has been performed and we are optimizing the method at the moment to obtain the maximum number of sites.

REFERENCES

1. Nacerddine, K et al., Developmental cell 9, 769-779 (2005); 2. Wang, L et al., EMBO reports 15, 878-885 (2014); 3. Fernandez, AG et al., Genome research 15, 250-259 (2005); 4. Jones, D et al., Genome biology 3, RESEARCH0002 (2002); 5. Kamath, RS et al., Nature 421, 231-237 (2003); 6. Rual, JF et al., Genome research 14, 2162-2168 (2004); 7. Lapenta, V et al., Genomics 40, 362-366 (1997); 8. Li, SJ, Hochstrasser, M, Nature 398, 246-251 (1999); 9. Seufert, W et al., Nature 373, 78-81 (1995); 10. Azuma, Y et al., The EMBO journal 24, 2172-2182 (2005); 11. Bachant, J et al., Molecular cell 9, 1169-1182 (2002); 12. Biggins, S et al., Genetics 159, 453-470 (2001); 13. Bylebyl, GR et al., The Journal of biological chemistry 278, 44113-44120 (2003); 14. Cubenas-Potts, C et al., Molecular biology of the cell 24, 3483-3495 (2013); 15. Dieckhoff, P et al., Molecular microbiology 51, 1375-1387 (2004); 16. Fukagawa, T et al., Nucleic acids research 29, 3796-3803 (2001); 17. Klein, UR et al., Molecular biology of the cell 20, 410-418 (2009); 18. Meluh, PB, Koshland, D, Molecular biology of the cell 6, 793-807 (1995); 19. Mukhopadhyay, D et al., The Journal of cell biology 188, 681-692 (2010); 20. Stead, K et al., The Journal of cell biology 163, 729-741 (2003); 21. Strunnikov, AV et al., Genetics 158, 95-107 (2001); 22. Zhang, XD et al., Molecular cell 29, 729-741 (2008); 23. Cheeseman, IM, Desai, A, Nature reviews. Molecular cell biology 9, 33-46 (2008); 24. Collette, KS et al., Journal of cell science 124, 3684-3694 (2011); 25. Hagstrom, KA et al., Genes & development 16, 729-742 (2002); 26. Kaitna, S et al., Current biology : CB 12, 798-812 (2002); 27. Schumacher, JM et al., The Journal of cell biology 143, 1635-1646 (1998); 28. Soderberg, O et al., Nature methods 3, 995-1000 (2006); 29. Soderberg, O et al., Methods 45, 227-232 (2008); 30. Ban, R et al., Genes to cells : devoted to molecular & cellular mechanisms 16, 652-669 (2011); 31. Fernandez-Miranda, G et al., Journal of cell science 123, 2823-2833 (2010); 32. Tammsalu, T et al., Science signaling 7, rs2 (2014); 33. Sonneville, R et al., The Journal of cell biology 196, 233-246 (2012)