Skip Internal Navigation
We are interested in mechanisms of chromosome segregation. Defects in chromosome segregation lead to aneuploidy, the condition of an abnormal number of chromosomes, which in turn may drive tumor formation. To elucidate these mechanisms, we study the mitotic roles of proteins associated with the interphase nuclear pore complex (NPC). NPCs are conduits for nucleocytoplasmic trafficking and organize other nuclear processes, such as gene expression and the repair of DNA damage. Surprisingly, many NPC components (nucleoporins) localize to mitotic kinetochores, the proteinaceous structures that attach sister chromatids to mitotic spindles and mediate their segregation into daughter cells at anaphase. The Ran GTPase and SUMO conjugation pathways are functionally and physically linked to each other and to NPCs. The Ran GTPase controls many cellular activities, including nucleocytoplasmic trafficking, nuclear assembly, and cell cycle progression. SUMO proteins are a family of ubiquitin-like proteins that become covalently conjugated to cellular targets. The pathways are indispensable for mitotic spindle assembly and kinetochore function. The ultimate goal of our studies is to understand the mitotic roles of these proteins, discover how they are coordinated, and determine how such coordination enhances the accurate distribution of chromosomes during mitosis.
SUMO-family small ubiquitin-like modifiers in higher eukaryotes
SUMO proteins are a family of ubiquitin-related proteins that become covalently linked to other cellular proteins. Human SUMO proteins have been implicated in a variety of cell functions, including nuclear trafficking, chromosome segregation, chromatin organization, transcription, and RNA metabolism. The three commonly expressed mammalian SUMO paralogs are called SUMO-1, SUMO-2, and SUMO-3. SUMO-2 and SUMO-3 are 96% identical while SUMO-1 is roughly 45% identical to either SUMO-2 or SUMO-3. This report collectively refers to SUMO-2 and SUMO-3 as SUMO-2/3. SUMO-1 is less abundant than SUMO-2/3, is conjugated to a distinct spectrum of targets, and exhibits different in vivo dynamics and responses to physiological stress such as heat shock. Unlike SUMO-2/3, SUMO-1 is concentrated at NPCs, the primary conduits of nucleocytoplasmic trafficking, reflecting the fact that it is the preferred conjugation partner of RanGAP1. RanGAP1 is the GTPase activator for Ran, a small GTPase that controls nuclear transport. SUMO-1 conjugation of RanGAP1 promotes its stable association to NPCs by binding to the large nucleoporin RanBP2.
The conjugation pathway for SUMO proteins is similar to the ubiquitin conjugation pathway: newly translated SUMO proteins are processed to reveal a di-glycine motif at their C-termini. The processing is mediated by a family of proteases called Ubiquitin-like protein proteases (Ulp) in yeast and Sentrin-specific proteases (SENP) in vertebrates (2). After processing, SUMO proteins undergo ATP-dependent formation of a thioester bond to their activating (E1) enzyme Aos1/Uba2. The activated SUMO proteins are transferred to form a thioester linkage with their conjugating (E2) enzyme Ubc9. Finally, an isopeptide bond is formed between SUMO proteins and substrates through the cooperative action of Ubc9 and protein ligases (E3). The linkage of SUMO proteins to their substrates can be severed by Ulps/SENPs, so it is likely that SUMO modification is highly dynamic in vivo. Ulp/SENPs play an important role in determining the spectrum of conjugated species because they directly regulate the production of free, conjugatable SUMO proteins and the half-life of conjugated species (2). There are six members of the Ulp/SENP family in mammals and five in amphibians (Xenopus laevis). We are systematically evaluating the physiological roles and regulation of these enzymes. Our recent observations have demonstrated that they play critical roles in ribosome biogenesis and mitotic kinetochore assembly.
Ribosome biogenesis occurs largely within the nucleolus. It incurs a major metabolic expense and is a critical point of cellular regulation during both cell growth and cancer progression. The process has been studied genetically in yeast but remains poorly characterized in metazoans. We examined the subnucleolar localization, behavior, and function of the two nucleolar SENPs: SENP3 and SENP5. Both enzymes colocalized with B23/nucleophosmin, an abundant 37-kD phosphoprotein that shuttles between the nucleolus and cytoplasm. B23/nucleophosmin is implicated in many cellular processes, including ribosome biogenesis and control of the Arf–MDM2–p53 pathway; it is often overexpressed in solid tumors and has been strongly linked to hematopoietic malignancies. We found that B23/nucleophosmin is essential for the stable accumulation of SENP3 and SENP5. After depletion of B23/nucleophosmin or co-depletion of SENP3 and SENP5, SUMO proteins accumulate within nucleoli. Importantly, depletion of SENP3 and SENP5 causes defects in ribosome biogenesis that are closely reminiscent of those observed in the absence of B23/nucleophosmin. Collectively, our findings indicate that control of SUMO deconjugation through SENP3 and SENP5 may be a major facet of B23/nucleophosmin function and that disruption of SUMOylation may significantly contribute to the phenotypes observed after B23/nucleophosmin loss.
Notably, SENP3 and SENP5 play slightly different roles in ribosome biogenesis, with SENP5 acting at the level of rRNA transcription and SENP3 acting at later stages of rRNA processing and ribosome assembly. To understand the role of SENP3 at a biochemical level, we characterized its interactions within Xenopus laevis egg extracts (XEEs). We found that SENP3 binds stably to three proteins that act as ribosome assembly factors. These SENP3-interacting proteins are essential for ribosome biogenesis and show sequence similarity to the yeast ribosome assembly factors Rix1p, Ipi1p, and Ipi3p, suggesting that they are previously unrecognized vertebrate homologs of the yeast Rix1 complex. SENP3 and the Rix1 complex bound to 60S ribosomal particles through B23/nucleophosmin under high Ran-GTP conditions. Under low Ran-GTP conditions, B23/nucleophosmin was lost, and SENP3 and the Rix1 complex bound instead to RanBP5, a nuclear import receptor. Despite no evidence that yeast Rix1 activity is integrated with Ulp1p function, our findings demonstrate that SENP3 and the vertebrate Rix1 complex are coupled both physically and functionally. Our data also suggest that B23/nucleophosmin and the Ran pathway regulate SENP3 and the Rix1 complex in a novel, antagonistic fashion.
We showed earlier that SENP6 (also called SUSP1) localizes within the nucleoplasm, where it plays a specialized role in dismantling highly conjugated SUMO-2/3 species. It was recently shown that poly-SUMO-2/3–conjugated species are frequently degraded through the action of RNF4, a ubiquitin ligase that targets them for proteasomal degradation. We analyzed the mitotic role of SENP6 in mammalian cells (1, 4). We found that SENP6–depleted HeLa cells have defects in metaphase chromosome congression and show characteristic changes in spindle morphology. We examined kinetochore composition to find the molecular basis of these phenotypes and found that a subset of the inner kinetochore proteins became undetectable at the kinetochores of SENP6-depleted cells, including components of the CENP-H/I/K and CENP-O complexes. At the same time, changes in outer kinetochore composition closely mimicked phenotypes observed after loss of CENP-H/I/K components. Importantly, we found that the CENP-H and CENP-I proteins were quantitatively degraded in the absence of SENP6 through a mechanism that requires both RNF4 and proteasome-mediated proteolysis. Together, these findings demonstrate a novel function of the SUMO pathway in inner kinetochore assembly, which finely balances the incorporation and degradation of components of the inner plate of the kinetochore.
Regulation of mitotic kinetochores by the Ran GTPase
We showed earlier that Ran has two important roles at mitotic kinetochores. First, it is essential for the regulation of the spindle assembly checkpoint (SAC). SAC is a cell cycle–regulatory pathway that prevents the onset of anaphase until all chromosomes are properly attached to the mitotic spindle and aligned on the metaphase plate. SAC accomplishes the task by monitoring the attachment of spindle microtubules to kinetochores. Failure of SAC is associated with chromosomal instability, leading to the production of aneuploid cells. Second, Ran is essential for assembly of microtubule fibers that attach kinetochores to spindle poles (k-fibers). We showed that RanBP2, a large Ran-binding nucleoporin, and RanGAP1, the GTPase-activating protein for Ran, are targeted to kinetochores during mitosis as a single complex that is both regulated by and important for stable kinetochore–microtubule association in mitotic spindles. We further found that Crm1, a Ran-GTP–binding nuclear export receptor, localizes to kinetochores in mammalian cells.
We are currently working on two aspects of Ran function at kinetochores. First, we are attempting to identify the effector for Ran that mediates its action on kinetochore-bound SAC proteins. We previously showed that elevated levels of Ran–GTP abrogate SAC-mediated mitotic arrest in XEEs and disrupt the kinetochore localization of SAC components, suggesting that the SAC is directly responsive to the overall concentration of Ran–GTP in that system. Our data strongly suggest that these effects are not mediated by known mitotic Ran effectors. We are currently examining other known Ran–GTP binding proteins, particularly nuclear transport receptors (karyopherins), to determine how they may contribute to SAC activation, and we are designing assays to identify novel components of this pathway.
Second, we are investigating how Crm1 promotes k-fiber assembly. We found that Crm1 localizes to kinetochores and that inhibition of Crm1 ternary complex formation by the inhibitor Leptomycin B (LMB) blocks kinetochore recruitment of RanGAP1/RanBP2. Crm1 requires neither ternary complex assembly nor microtubules for kinetochore binding. Our findings support a direct role of the RanGAP1/RanBP2 in correct k-fiber assembly and suggest that Ran has a kinetochore-associated effector pathway that can be clearly differentiated from Importin-beta–mediated inhibition of soluble spindle assembly factors. The component(s) at kinetochores that is(are) directly involved in Crm1 recruitment is a major focus of our ongoing studies, which include both an investigation of Ran-dependent interactions of Crm1 with kinetochore proteins and screens to identify mitotic-specific cargoes that may be controlled through Crm1 binding.
Mitotic roles of nuclear pore complex proteins
The relationship between mitotic kinetochores and NPCs is both surprisingly intimate and remarkably poorly understood. Interphase NPCs consist of around 30 proteins. During mitosis, metazoan NPCs disassemble into approximately a dozen subunits, many of which associate to kinetochores. Conversely, many kinetochore proteins bind to interphase NPCs. We hypothesize that NPCs act as both a conduit for nuclear trafficking and a scaffold for organization of cellular activities, including the SUMO and Ran pathways, by recruiting their components or targets through nucleoporin binding. When the NPC is disassembled during mitosis, many of the protein-protein interactions formed during interphase may persist, allowing the same mechanisms of regulation to be asserted in mitosis. A major goal of our research is to test such a hypothesis.
Our work has focused on two complexes of kinetochore-bound nucleoporins. The vertebrate Nup107–160 complex consists of Nup160, Nup133, Nup107, Nup96, Nup85, Nup43, Nup37, Sec13, and Seh1 and is broadly distributed on spindles during prometaphase. It remains kinetochore-bound throughout mitosis and shows enhanced accumulation on unattached kinetochores. The RRSU complex consists of RanBP2, SUMO-1–conjugated RanGAP1, and Ubc9. We reported earlier that the RanBP2 complex associates with kinetochores in a microtubule-dependent manner and that recruitment of RRSU to kinetochores requires Crm1, a Ran-dependent nuclear export receptor (see above). In the absence of kinetochore-bound RRSU, the assembly of microtubule-based kinetochore fibers is defective, and cells exhibit an elevated rate of chromosome mis-segregation.
Unattached kinetochores nucleate microtubules in a Ran-regulated manner; such microtubules promote assembly of kinetochore fibers (k-fibers) that connect kinetochores to spindle poles. We found that Nup107-160 interacts with the gamma-tubulin ring complex (gamma-TuRC), an essential and conserved microtubule nucleator, and that such an association recruits gamma-TuRC to unattached kinetochores (5). Nup107–160 and gamma-TuRC act cooperatively to promote spindle assembly through microtubule nucleation at kinetochores: HeLa cells lacking Nup107–160 or gamma-TuRC were profoundly deficient in kinetochore-associated microtubule nucleation. Moreover, co-precipitated Nup107–160/gamma-TuRC complexes nucleated microtubules in assays using purified tubulin. Although Ran did not regulate microtubule nucleation by gamma-TuRC alone, Nup107–160/gamma-TuRC complexes required Ran-GTP for microtubule nucleation activity. Collectively, our observations show that Nup107–160 promotes spindle assembly through Ran-GTP–regulated nucleation of microtubules by gamma-TuRC at kinetochores, and reveal an unexpected relationship between nucleoporins and the microtubule cytoskeleton. Notably, kinetochore recruitment of the RRSU complex is coupled with kinetochore-MT attachment and inhibits microtubule nucleation at kinetochores, possibly through suppression of Nup107–160/gamma-TuRC complexes. We are currently investigating the mechanism through which Ran regulates microtubule nucleation of gamma-TuRC associated with Nup107–160 at kinetochores.