Assistant Professor, Eppley Institute
Research Interests
My main interest is eukaryotic cell division. Specifically, I hope to understand the regulation of molecular motors and microtubule-associated proteins (MAPs) in mitotic spindle formation.
Summary of Research
At the onset of mitosis in animal cells, the nuclear envelope breaks down, the nuclear lamina is depolymerized, and interphase chromatin is condensed into chromosomes. Concomitantly, cytoplasmic microtubules (Mts) are reorganized into a mitotic spindle apparatus, a highly dynamic structure required for the segregation of sister chromatids. Recent studies from several laboratories have led to the discovery that the small GTPase Ran regulates multiple aspects of spindle assembly in mitosis. The activated form of Ran (RanGTP) stimulates spindle assembly in Xenopus laevis egg extracts by releasing SAFs such as TPX2 and NuMA from the inhibitory binding of importina/b. My studies have since uncovered several novel aspects of the Ran-signaling pathway that lead to activation of Aurora A, a mitotic kinase which is essential for spindle assembly.
- Discovering a Ran-signal pathway mediated by Aurora A kinase
I found that Ran-GTP stimulates the interaction between TPX2 and Aurora A kinase by displacing importin a and b from TPX2. This enhanced interaction allows TPX2 to stimulate autophosphorylation and activation of Aurora A kinase in a MT-dependent manner. Furthermore, I showed that TPX2 and MTs could prevent phosphatase I from dephosphorylating and inactivating Aurora A. Based on this work, I proposed that Aurora A activation by RanGTP could establish a positive feedback loop to drive spindle assembly toward chromosomes where the RanGTP concentration is elevated. Previous studies have shown that Aurora A regulates spindle assembly by phosphorylating a number of SAFs, and my findings established a RanGTP-signaling pathway in mitosis that is mediated by Aurora A kinase (Fig. 1).
- Development of assays to further study spindle assembly mediated by the RanGTP-Aurora A pathway
To further understand how the RanGTP-Aurora A pathway regulates spindle assembly, I developed a technique to study how locally concentrated active Aurora A affects Ran-dependent spindle assembly. I found that magnetic beads coated with Aurora A function as microtubule organizing centers (MTOCs) in the presence of RanGTP inXenopus egg extracts. Aurora A-coated beads significantly increase the rate of bipolar spindle assembly in the presence of RanGTP. However, beads coated with XMAP215, a Xenopus microtubule-associated protein that can nucleate microtubule assembly, stimulate only aster formation but not spindle formation. Furthermore, Aurora A beads do not stimulate microtubule aster formation in vitro with pure tubulin, while XMAP215 beads do. I showed that Aurora A is able to recruit both microtubule nucleators and microtubule motors to the beads to stimulate microtubule nucleation and organization. A kinase dead mutant of Aurora A (Aura-AA) does not stimulate microtubule aster or spindle formation, further suggesting that Aurora A kinase activity is important in this process (Fig. 2).
- Identification of lamin B as a structural component of the long-sought-after spindle matrix
The ability of Aurora A beads to nucleate MTs and stimulate spindle assembly provides a powerful system to further study the role of Aurora A in mitosis. Moreover, this assay system provides a potent means to isolate the spindle structure and to identify components of the spindles. I was able to isolate spindle structures by using a magnet to retrieve spindles made by Aurora A beads. Spindles isolated by magnetic retrieval have fewer contaminants than spindles isolated by a traditional centrifugation method. More importantly, I was able to use this system to isolate the long-sought-after spindle matrix. Decades ago, a static spindle matrix was proposed to help organize and stabilize spindle microtubules during mitosis. However, the molecular nature of this structure has remained mysterious. Using the Aurora A beads, I have been able to isolate and characterize a matrix structure that remains associated with the beads after spindle MTs are depolymerized. I found that the intermediate filament protein lamin B, a major component of the interphase nuclear lamina, together with several SAFs, were part of this matrix. Lamin B, but not the SAFs, is required for the assembly of the matrix. The assembly of the lamin B matrix requires RanGTP in mitosis. Importantly, I found that reduction of lamin B results in defects in spindle assembly in both tissue culture cells and in Xenopus egg extracts. Moreover, disruption of lamin B assembly with mutant lamin B also disrupts spindle assembly in egg extracts. Since the lamin B matrix tethers a number of spindle assembly factors, including those suspected to interact with the putative spindle matrix, my studies identify lamin B as a structural component of the spindle matrix that can support assembly of the mitotic spindle.
Future directions
I will continue to utilize the assays I have developed to further identify and study new SAFs that are regulated by the Aurora-A kinase signaling pathway.
- Identification and functional characterization of new factors involved in spindle assembly
Considering its ability to stimulate mitotic spindle assembly, Aurora A is likely to regulate the activities of several microtubule nucleators and microtubule motors as well as stimulate spindle matrix formation. With the development of the Aurora A bead based spindle assembly assay as described above, I were able to isolate spindle structures,microtubule nucleators and microtubule motors that are recruited to Aurora A beads from Xenopus egg extracts. In addition to lamin B, I have also identified other novel components of the spindle matrix. Many of these components are established targets of anti-cancer therapies. Although it has not been possible to assemble spindles using lysates made from mammalian tissue culture cells, I have recently been able to use lysates made from mitotically arrested mouse embryonic stem cells and HeLa to develop assays similar to those using Xenopus egg extracts. This new technical development should allow me to mass spectrometry to identify candidate SAFs in mammalian system, as well as to gain insight into the biology of the spindle matrix. - Understand the regulation of assembly and disassembly of the spindle matrix
Interphase lamin B assembles into nuclear lamina, which maintains nuclear integrity and organizes chromosomes into distinct domains for gene expression. As cells enter mitosis, lamin B is phosphorylated by Cdc2 kinase leading to disassembly of nuclear lamina. I found that lamin B assembles into a matrix that associates with spindles and regulates spindle assembly in a RanGTP-dependent manner. This finding opens up a unique angle to study the mechanism of spindle assembly. I plan to further explore how the mitotic lamin B matrix is assembled and disassembled and how this, in turn, regulates spindle morphogenesis.
Since lamin B is a structural component of interphase nuclear lamina, the discovery of the mitotic lamin B matrix should provide new means to study how mitotic exit and spindle disassembly is coupled with nuclear reformation. Understanding how mitotic spindle lamin matrix is transformed into nuclear lamina should also provide new insight into how spindle disassembly could influence the organization of interphase chromatin structure and cell fate. Therefore, an in-depth analysis of the assembly and disassembly of the mitotic lamin B matrix will not only advance our understanding of the mechanism of spindle morphogenesis but also shed light on how spindle disassembly is coupled to nuclear reformation after mitosis. Although I will focus my effort on studying spindle assembly in the first two to three years in my own lab, I anticipate that new findings we make during this period will direct us to new areas that will permit us to study how spindle disassembly is coordinately regulated with other postmitotic processes. - Functional screening of small molecule inhibitors of Aurora A kinase
The crucial role of the mitotic spindle in cell division has made it an important target in cancer chemotherapy. In vitrospindle assembly in Xenopus egg extracts and mammalian cell lysates provide very powerful means of dissecting the roles of proteins involved in mitosis. It is therefore a very useful assay system for identification of pharmacological agents that could serve as potent anti-cancer therapies. Some of the most promising anti-cancer drug targets are mitotic kinases, such as Aurora A that regulate cellular progression through mitosis. Traditionally, screens for kinase inhibitors involve mixing purified active kinase with different chemical compounds and assaying for inhibition of kinase activity in vitro. Thus far, most chemical inhibitors inhibit kinase activity by binding to the ATP-binding site of the kinase. Since the activity of many kinases is regulated by their interacting proteins in the cell, it should also be possible to identify inhibitors that act on these kinase regulators. It is also interesting to note that some successful kinase inhibitors such as Gleevec bind to the inactive form of Src kinase and prevent its activation. These kinds of inhibitors would be easily missed in traditional drug screens that depend on in vitro kinase assays.
Several Aurora A kinase binding proteins, including TPX2, Ajuba, PPI, HURP, and p53, regulates Aurora A kinase activity. For example, TPX2 can lock Aurora A in the active conformation and protect it from dephosphorylation by PPI. The binding of TPX2 also blocks the inhibition of p53 on Aurora A. Therefore, developing assays that closely mimic the cellular environment of Aurora A will allow identification of additional Aurora A inhibitors that would be missed in traditional drug screens that depend on in vitro kinase assays. My strategy is based on my finding that Aurora A beads can stimulate spindle assembly in Xenopus egg extracts and that this stimulation is dependent on Aurora A kinase activity. I am now in a process of developing a high-throughput screening protocol to identify small molecules that inhibit Aurora A activity that might otherwise be missed in traditional drug screens that employ in-vitro kinase assays. This type of assay can identify inhibitors that block the ability of Aurora A kinase to stimulate spindle formation in the presence of its regulatory proteins. Inhibitors that are effective in the egg extracts, which closely mimics the physiological environment of the cell, would be most likely to be effective therapeutically in stopping the growth of cancer cells.
Selected Publications
Ming Ying Tsai, Ph.D.
