MITOTIC REGULATION IN HIGHER EUKARYOTES BY RAN AND SUMO-1
, Head, Section on Cell Cycle Regulation
Alexei Arnaoutov, PhD, Visiting Fellow
Maiko Furuta, PhD, Visiting Fellow
Ai Kametaka, PhD, Visiting Fellow
Ram Kumar Mishra, PhD, Visiting Fellow
Debaditya Mukhopadhyay, PhD, Visiting Fellow
Yonggang Wang, PhD, Visiting Fellow
Hyun-Joo Yoon, PhD, Visiting Fellow
Chawon Yun, PhD, Visiting Fellow
Maia Ouspenskaia, DVM, Biologist/Technician
Yekaterina Boyarchuk, BA, Graduate Student1
SUMO family small ubiquitin-like modifiers in higher eukaryotes
SUMO proteins, which are ubiquitin-related proteins that become covalently linked to other cellular proteins, have been implicated in a variety of cell functions, including nuclear trafficking, chromosome segregation, chromatin organization, transcription, and RNA metabolism. Of the three commonly expressed mammalian SUMO paralogues, SUMO-2 and SUMO-3 are 96 percent identical while SUMO-1 shares about 45 percent identity with either SUMO-2 or -3. Many lines of evidence argue that vertebrate SUMO paralogues are functionally distinct, although their individual roles are not yet fully understood. The conjugation pathway for SUMO proteins is similar to the ubiquitin conjugation pathway: SUMO proteins are processed by ubiquitin-like proteases/sentrin-specific proteases (Ulps/SENPs) to reveal a diglycine motif at their C-termini. After processing, SUMO proteins undergo conjugation through the successive action of an activating enzyme (E1), a conjugating enzyme (E2), and typically a SUMO protein ligase (E3). SUMO proteases can sever the linkage of SUMO proteins to their substrates, making it likely that SUMO modification is highly dynamic in vivo. As with processing, SUMO deconjugation is mediated by Ulps/SENPs.
Ulp/SENPs play an important role in determining the spectrum of conjugated species by directly regulating the production of free SUMO proteins and the half-life of conjugated species. We wish to determine the evolutionary relationships among Ulp/SENPs as well as the proteases’ enzymatic specificities and localization. To that end, we cloned cDNAs encoding each of the human and Xenopus laevis Ulp/SENP proteins and raised antibodies against them. Humans have six Ulp/SENPs (SENP1, 2, 3, 5, 6, and 7) while Xenopus has five (xUlp1, SENP3, 5, 6, and 7).
To assay the paralogue specificity of mammalian Ulp/SENPs under comparable conditions, we collaborated with Keith Wilkinson to examine the proteases’ reactivity with HA-tagged SUMO-1 and SUMO-2-vinyl sulfones (HA-SUMO1-VS and HA-SUMO2-VS). These reagents covalently react with active-site cysteine residues of SUMO proteases in a manner reflecting the selectivity of the enzymes to individual paralogues. Green fluorescent protein (GFP) fusions with all SENPs reacted with HA-SUMO2-VS, but only GFP-SENP1 and GFP-SENP2 showed substantial reactivity with HA-SUMO1-VS. Notably, the reaction of HA-SUMO1-VS with GFP-SENP1 was more efficient than with GFP-SENP2, consistent with earlier observations of the binding of these proteases to individual paralogues. In conjunction with earlier reports, our findings indicate that SENP1 and SENP2 are the primary enzymes responsible for SUMO-1 metabolism while all SENP/Ulps act on SUMO-2 and -3.
We are systematically examining the localization and function of individual Ulp/SENPs. SENP6 (also called SUSP1) localizes within the nucleoplasm. SENP6 depletion from cell lines expressing GFP fusions to individual SUMO paralogues caused redistribution of GFP-SUMO-2 and GFP-SUMO-3 into promyelocytic leukemia protein (PML) bodies as well as PML body enlargement and increased PML body number. We did not observe comparable redistribution of GFP-SUMO-1. While SENP6 has a strong paralogue preference for HA-SUMO2-VS, we were surprised to find that it displayed little activity for SUMO-2 or SUMO-3 processing or for the deconjugation of substrates containing fewer than three SUMO-2/3 moieties. By contrast, SENP6 was highly active against model substrates containing chains of SUMO-2 produced in a bacterial expression system. Together, our findings argue that SENP6 may play a specialized role in dismantling highly conjugated SUMO-2 and SUMO-3 species. Our findings are particularly interesting in light of the established role of a related enzyme Ulp2p in chain editing in yeast. Likewise, the other vertebrate Ulp2p-like enzyme SENP7 is a nucleoplasmic SUMO-2/3–specific isopeptidease, although we have not yet determined whether it acts preferentially on chains.
We examined which mammalian Ulp/SENPs localize to nuclear pore complexes (NPCs). Earlier, we and others had shown that SENP2 binds the nucleoplasmic side of NPCs; still others reported various localization patterns for SENP1 fusion proteins. Using specific anti–SENP1 antibodies, we found that endogenous SENP1 localizes to the inner face of NPCs in HeLa cells. The differences between our data and previous reports may reflect overexpression of SENP1 fusion proteins. We found that SENP1-NPC association occurred through saturable binding sites. The localizations of SENP1 and SENP2 indicate that both enzymes capable of efficient SUMO-1 deconjugation are restricted to NPCs, suggesting that SUMO-1 plays a role that is associated with or regulated by NPCs.
Mukhopadhyay D, Ayaydin F, Kolli N, Tan S-H, Anan T, Kametaka A, Azuma Y, Wilkinson KD, Dasso M. SUSP1 antagonizes formation of highly SUMO-2/3 conjugated species. J Cell Biol 2006;174:939-49.
Mukhopadhyay D, Dasso M. Modification in reverse: the SUMO proteases. Trends Biochem Sci 2007;32:286-95.
Quimby BB, Yong-Gonzalez V, Anan T, Strunnikov AV, Dasso M. The promyelocytic leukemia protein (PML) stimulates SUMO conjugation in yeast. Oncogene 2006;25:2999-3005.
Regulation of mitotic kinetochores by the Ran GTPase
The Ran GTPase is required for many cellular functions, including nucleocytoplasmic trafficking, spindle assembly, and cell cycle control. Ran’s nucleotide exchange factor RCC1 binds to chromatin throughout the cell cycle. Ran’s GTPase activating protein RanGAP1 localizes to the cytosolic face of the NPC during interphase through association with RanBP2, a large nucleoporin. The interphase distribution of Ran regulators leads to a high concentration of Ran-GTP in nuclei and low levels of Ran-GTP in the cytosol. The major effectors for Ran are a family of Ran-GTP binding proteins that were discovered as nuclear transport receptors. These receptors are collectively called Karyopherins. To date, two Karyopherins have been shown to act as Ran effectors during mitosis: Importin-beta and Crm1. Our studies have been particularly concerned with Ran functions at kinetochores. Kinetochores are proteinaceous structures that assemble at the centromere of each sister chromatid during mitosis and serve as sites of spindle microtubule (MT) attachment. Kinetochore attachment is monitored through the spindle assembly checkpoint (SAC), which prevents mitosis until all chromosomes are attached and aligned onto the metaphase plate. Components of the SAC include Mad1, Mad2, Mps1, Bub1, Bub3, BubR1, and CENP-E.
We found that Crm1 localizes to kintochores. We inhibited Crm1 ternary complex formation with cargo and Ran-GTP by using leptomycin B (LMB), a highly specific chemical inhibitor, and found that it blocked kinetochore recruitment of RanGAP1/RanBP2. Crm1 itself requires neither ternary complex assembly nor MTs for kinetochore binding. Centromeres of LMB-treated HeLa or U2OS cells were under increased tension, with the cells showing a decreased distance between their spindle poles. In LMB-treated cells, kinetochores dramatically failed to maintain discrete end-on attachments to single bundles of spindle microtubules (k-fibers) and showed a resultant elevation in chromosome mis-segregation. These findings have several important implications. First, they directly link the RanGAP1/RanBP2 to correct k-fiber assembly. Second, they suggest that Ran has a kinetochore-associated effector pathway that can be clearly differentiated from Importin-beta–mediated inhibition of soluble spindle assembly factors. Third, they show that several Karyopherins act as Ran effectors during mitosis; in principal, it is possible that other members of the Karyopherin family may also act during mitosis. The component(s) at kinetochores that is directly involved in Crm1 recruitment is a major focus of our ongoing studies.
We also examined the function of SAC components within mitotic cells, particularly the role of Bub1 kinases at the inner centromeres (ICs). Protein complexes of the IC regulate sister chromatid cohesion and modulate MT attachment. The proteins include the chromosomal passenger complex (CPC), mitotic centromere-associated kinesin (MCAK), and Shugoshin (Sgo). The CPC consists of the Aurora B kinase, INCENP, Survivin, and Dasra/Borealin. Aurora B phosphorylates and inhibits the microtubule depolymerase MCAK, thereby controlling the polymerization/depolymerization state of microtubules to achieve correct end-on attachment of microtubules to kinetochores. The CPC localizes along prophase chromosomes and concentrates at the IC in prometaphase and metaphase. The SAC component Bub1 is required for outer kinetochore assembly and for IC recruitment of Sgo. We found that Bub1 plays a central role in IC formation, acting at several points in the assembly pathway. First, Bub1 controls CPC localization to the IC. In the absence of Bub1, the CPC binds to chromosome arms but does not accumulate in the IC. Although the kinase activity of Aurora B was not lost under these circumstances, CPC stability was markedly altered. Second, as in earlier reports from others, we found that Bub1 mediates xSgo recruitment to the IC. In addition, we found that Bub1 acts primarily by promoting xSgo binding to mitotic chromatin and can accomplish this function even when it is not stably associated with mitotic chromosomes or kinetochores. Third, in contrast to chromatin binding of xSgo, Bub1 by itself is insufficient to direct xSgo to the IC in the absence of the CPC. Together, our findings suggest that Bub1 regulates localization of IC components through mechanisms that are both CPC-dependent and -independent. Remarkably, the kinase activity of Bub1 is essential for its roles in IC assembly in contrast to its roles at outer kinetochores.
Arnaoutov A, Dasso M. Ran-GTP regulates kinetochore attachment in somatic cells. Cell Cycle 2005;4:1161-5.
Boyarchuk EY, Nikol’skii NN, Dasso M, Arnaoutov AM. Assembly of correct kinetochore architecture in Xenopus egg extract requires transition of sperm DNA through interphase. Cell Tissue Biol 2007;1:80-8.
Boyarchuk Y, Salic A, Dasso M, Arnaoutov A. Bub1 is essential for assembly of the inner centromere. J Cell Biol 2007;176:919-28.
Dasso M. Ran at kinetochores. Biochem Soc Trans 2006;34:711-5.
Mitotic roles of NPC proteins
Trafficking between the nucleus and cytoplasm occurs through NPCs. During mitosis, metazoan NPCs disassemble into approximately a dozen subunits. Two of these subunits are targeted to mitotic kinetochores. First, the RanBP2 complex associates with kinetochores in an MT-dependent manner such that the complex consists of RanBP2 (a large nucleoporin also known as Nup358), SUMO-1–conjugated RanGAP1 (the activating protein for the Ran GTPase), and Ubc9 (the sole conjugating enzyme for the SUMO family of ubiquitin-like modifiers). Second, the nine-protein vertebrate Nup107–160 complex associates with kinetochores throughout mitosis in an MT-independent manner. The Nup107–160 complex includes Nup160, Nup133, Nup107, Nup96, Nup85, Nup43, Nup37, Sec13, and Seh1. During telophase, Nup107–160 is targeted to chromosomes and acts in a critical and early fashion during NPC re-assembly.
Our earlier studies suggested that mitotic RanBP2 complex targeting occurs through the RanBP2 protein and that its recruitment is both regulated by and essential for correct MT-kinetochore attachment. As discussed above, we further found that the Crm1 nuclear export receptor localizes to kinetochores and promotes binding of the RanBP2 complex. In addition, we examined the mitotic role of Nup107–160 complex in collaboration with Douglass Forbes and Beatriz Fontoura, using HeLa tissue culture cells and Xenopus egg extracts (XEEs). In HeLa cells, the complex localized not only to kinetochores but also to prometaphase spindle poles and proximal spindle fibers, reminiscent of the localization of some SAC components. Expanded crescents of the Nup107–160 complex formed around unattached kinetochores, similar to the observed accumulation of dynamic outer kinetochore proteins. In mitotic XEEs, the Nup107–160 complex localized throughout reconstituted spindles. When the complex was depleted from XEEs, the SAC remained intact and responsive to Ran-GTP concentration, but spindle assembly was defective. MT nucleation around sperm centrosomes appeared normal, but the MTs quickly disassembled, leaving largely unattached sperm chromatin. Notably, Ran-GTP could still promote assembly of MT asters in Nup107–160-depleted extracts, indicating that Nup107–160 is not essential for soluble spindle assembly factor (SAF) activation. Together, our findings showed that the Nup107–160 complex is dynamic in mitosis and promotes spindle assembly in a context that is distinct from intact interphase NPCs.
Our future studies will focus on several issues, including the component(s) at kinetochores that are directly involved in Crm1 recruitment, the relationship between the Nup107–160 and RanBP2 complexes during mitosis, and how the complexes together regulate the attachment of microtubules to kinteochores and the spindle assembly checkpoint. We are also examining the mitotic localization and function of other NPC components.
Orjalo A, Arnaoutov A, Shen Z, Boyarchuk Y, Zeitlin S, Fontoura B, Briggs S, Dasso M, Forbes D. The Nup107-160 nucleoporin complex is required for correct bipolar spindle assembly. Mol Biol Cell 2006;17:3806-18.
Prunuske AJ, Liu J, Elgort S, Joseph J, Dasso M, Ullman KS. Nuclear envelope breakdown is coordinated by both Nup358/RanBP2 and Nup153, two nucleoporins with zinc finger modules. Mol Biol Cell 2006;17:760-9.
Rundle NT, Nelson J, Flory MR, Joseph J, Th’ng J, Aebersold R, Dasso M, Andersen RJ, Roberge M. An ent-kaurene that inhibits mitotic chromosome movement and binds the kinetochore protein RanBP2. ACS Chem Biol 2006;1:443-50.
1 Graduate Partnerships Program
COLLABORATORS
Beatriz M.A. Fontoura, PhD, University of Texas Southwestern Medical Center, Dallas, TX
Douglass J. Forbes, PhD, University of California, San Diego, CA
Nikolay N. Nikol’skii, PhD, Institute of Cytology, Russian Academy of Sciences, St. Petersburg, Russia
Michel Roberge, PhD, University of British Columbia, Vancouver, British Columbia, Canada
Adrian Salic, PhD, Harvard Medical School, Boston, MA
Alexander V. Strunnikov, PhD, Program in Cellular Regulation and Metabolism, NICHD, Bethesda, MD
Katharine S. Ullman, PhD, Huntsman Cancer Institute, University of Utah, Salt Lake City, UT
Keith D. Wilkinson, PhD, Emory University, Atlanta, GA
For further information, contact mdasso@helix.nih.gov.
