Intellectual merit. Microtubules (MTs) are the primary components of crucial subcellular structures including the intracellular transport network and the mitotic spindle that separates the chromosomes during cell division. A central problem in cell biology is to understand how these MT-based structures form, are dynamically maintained, and drive the organization of the rest of the cell. Classically, these questions have been addressed by identifying and characterizing the proteins that regulate MT dynamics. While this approach has been powerful, it is not sufficient: the MT cytoskeleton is a complex system that exhibits behaviors ("emergent properties") not straightforwardly predictable from analysis of the individual components. Such a system level problem requires system level approaches: mathematical and computational modeling. The goals of this project are to develop, utilize, and experimentally test two computational models of MT dynamics, and to use these models to investigate the function and mechanism of MT binding proteins. These models will be built at two scales: a mesoscopic (medium scale) model that will be used to investigate the intrinsic properties of a system of dynamic MTs in a cell-like environment, and a molecular scale model that will be used to develop hypotheses about the mechanisms of dynamic instability and its alteration by MT binding proteins. The long-term goal of this work is to develop a predictive and quantitative understanding of the MT cytoskeleton and its regulation by MT binding proteins, which will impact fields ranging from systems biology to nanotechnology. Broader impacts. An important part of this project is to develop a freely disseminated software suite "MT Toolbox" (MTT), which will include analysis tools and instructional electronic tutorials. MTT will have two implementations: 1) a web-based interface that will allow scientists and students at remote sites to submit jobs for running the models and their analysis tools on our server; 2) a freely disseminated software suite containing all programs with online tutorials, user, and programmer's guides. The flexible models and tutorials produced through this project will allow researchers to develop and test specific hypotheses about the mechanisms of MT dynamics, which will in turn help design and direct future experiments. More broadly, it will help students and researchers at all levels gain an intuitive understanding of dynamic MT systems. The project will educate three graduate students and two undergraduates who will benefit from interdisciplinary training in biology and computational modeling. Projects related to the proposed research will be incorporated in The Research Experience for Teachers at Notre Dame (RET@ND) program as well as the Notre Dame McNair Program, which promotes graduate and doctoral studies for minority students.
BACKGROUND Microtubules are protein-based polymers that exist in all eukaryotic cells. Microtubules act as train-tracks for movement of intracellular cargo. However, unlike normal train tracks, microtubules are constantly being laid down and picked up in a surprising behavior known as dynamic instability. Dynamic instability is important because it allows the microtubules to explore cellular space and find their cargo. It also helps the cell change shape in response to extracellular signals, which is essential for processes as varied as movement of amoebas and development of animals. Cells regulate dynamic instability by using proteins that bind to the microtubules and alter their stability. Although microtubule dynamic instability and its importance are well established, the molecular-scale mechanisms of dynamic instability and its alteration by microtubule binding proteins are poorly understood. More specifically, there are varied conceptual models (ideas) for how microtubules and their regulators work, but it has been hard to distinguish between them because it has been difficult to make experimentally testable predictions. It is important to establish a molecular-level understanding of how microtubules work because doing so could lead to better agricultural pesticides and medicines (some of the best herbicides and anti-fungal agents target microtubule assembly). Moreover, having a deeper understanding of microtubules should provide insight into other dynamic biological polymers because some of the principles identified should be general. Finally understanding how biological polymers work may have applications in bioengineering and synthetic biology. INTELLECTUAL MERIT The purpose of the work supported by this NSF grant was to build a computational model of microtubule assembly and use it to bridge the gap between the conceptual models for how microtubules work and the experiments that could test the predictions of these models. More specifically, we generated a novel detailed computational model that is based on established microtubule structure and biochemistry. We then used this model to conduct in silico experiments and compared the results to real experiments, with the goal of using this comparison to support, refute, or refine specific conceptual models. Our work was focused both on understanding microtubules themselves and on how microtubule binding proteins alter microtubule dynamics. Significant advances include: - We used analysis of the behavior of the simulated microtubules to question several commonly held assumptions about microtubule assembly and dynamics, including the idea that microtubules grow as open sheets.1 - We used the computational model and associated analyses to propose that previously unrecognized cracks exist between subunits near the tip, and that fluctuations in these cracks play a fundamental role in dynamic instability.1-3 This work was important because it both contradicted some commonly accepted ideas and provided a new idea about the mechanism of microtubule dynamics. - We used a combination of experiments, mathematical modeling, and computer simulations to explain how microtubule binding proteins work, with particular advances on the Alzheimer's associated protein Tau4 and the cancer associated protein stathmin.5 - This model and associated experiments form the foundation for new work on how groups of microtubule binding proteins work together to control microtubule dynamics. BROADER IMPACTS Key impacts include: - Dr. Goodson continued to teach her biology course for freshman engineers and physical scientists (>500 students), and published a paper describing this course so that others could use it as a model.6 - To help students and researchers gain insight into protein binding behavior (a confusing basic concept), we wrote a software package called MTBindingSim, and published two papers on how to use it.7-8 Drs. Goodson and Alber co-organized Joint Notre Dame- Argonne National Laboratory Workshop on Physical Approaches to Studying the Cytoskeleton and Cell Motility, Chicago, IL, March 14, 2013: https://www3.nd.edu/~icsb/Workshop_2013/ - Drs. Alber and Goodson continued developing an interactive web site with electronic tutorials designed to help students at various levels understand and utilize stochastic and multiscale approaches in computational biology: http://www3.nd.edu/~biocomp/ - Eight graduate students (half female) and eight undergraduates (five female, including one underrepresented minority) were involved in various aspects of the project, as were three high school teachers and two high school students, including one underrepresented minority. Students at all of these levels obtained training in interdisciplinary science. The high school teachers used their experience with us to publish a paper in a teaching journal.9 - Dr. Goodson continued outreach efforts, including giving yearly guest lectures in a math class at Trinity High School and judging for the Regional Siemens science competition. She also co-authored an article for a scientific society newsletter explaining to scientists at different career stages how to get involved in K-12 science education.10 1. MolBiolCell 23, 642-656 (2012). 2. SoftMatter 10, 2069-2080 (2014). 3. PhysRevE 83, 041905 (2011). 4. MolBiolCell 23, 4796-4806 (2012). 5. PNAS 110, 20449-20454 (2013). 6. CellMolBioeng 6, 460-468 (2013). 7. Bioinformatics 28, 441-443 (2012). 8. MethodsCellBiol 115, 375-384 (2013). 9. MTA Journal Fall2010 31-35 (2010). 10. ASCB Newsletter 33, 1,6-7 (2010).
