Progress 09/01/11 to 08/31/13
Outputs
Target Audience: Topp trained three undergraduate students, Victoria Arendt, Margaret Cinderella, and Michael McLennon, in root phenotyping and plant biology Changes/Problems:
Nothing Reported
What opportunities for training and professional development has the project provided? Training Activities Submitted an NSF Plant Genome Reserach Program grant as a co-PI to study the genetic basis of perenniality in close relatives of crop plants Professional Development Topp secured an independent researcher position (Assistant Member and Principal Investigator) at the Donald Danforth Plant Science Center in Saint Louis, MO, largely in part from training and research accomplishments funded under this project. Oral presentations in this funding period: Maize Genetics Conference; CEPCEB and Department of Botany and Plant Sciences seminar, University of California Riverside; Donald Danforth Plant Science Center How have the results been disseminated to communities of interest?
Nothing Reported
What do you plan to do during the next reporting period to accomplish the goals?
Nothing Reported
Impacts
What was accomplished under these goals?
Impact: Increases in the severity and unpredictability of harsh weather, coupled with an increasing global population, place an emphasis on discovering how crops interact with their environment so that we can rapidly develop new varieties that are robust to climate change. Root systems are responsible for water and nutrient uptake in plants, and thus represent a golden opportunity to develop crops that produce more food, fiber, and fuel, with less water, fertilizer, and other unsustainable inputs. However, roots are difficult to study since they grow underground and have complex shapes that can rapidly change. In this project, building off previous advances in root imaging, we developed a way to measure the growth of living roots using computer automation. We compared the root growth of corn varieties that are either very efficient, or inefficient, at collecting the key soil element Nitrogen (N), and found that increases in root branching were correlated with N uptake. We can use these data to breed varieties that need less N-fertilizer inputs and that can grow on marginal, low-N soils. We used the same system to compare corn varieties similar to those that were grown in the 1930's to modern kinds. Modern varieties can be grown about 8-fold more densely than historical varieties (resulting in ~8 times greater yield), and we demonstrated that changes in root systems likely played a big role in this historical trend. One important finding was that when historical varieties are grown closely together in the same container, they sense each other via root-to-root communication, retarding their growth, whereas modern varities grow equally as well in close pairs or alone. These studies revealed fundamental aspects of how plants talk to each other, and this knowledge can be applied to increase the planting density and yield of other crop species that are not as well adapted to high planting density. Accomplishments: Aim1: Quantify and characterize environmental responses of maize root systems a) BSSS lines - studying the effects on root growth of long-term reccurent selection for density adaptation in maize. In year 1 of the project, non-density adapted maize inbreds (BSSS cycle0 and BSCB cycle 0), as well as their hybrid were phenotyped in the 3D gel system, and compared to their density adapted (cycle 17) counterparts. Topp demonstrated that cycle 17 architectures were significantly deeper and narrower than cycle 0, despite similar biomass, ostensibly reflecting a selective pressure on root allocation during density adaptation. In year 2, Topp discovered that when grown two-to-a-pot, cycle 0 hybrids exhibited growth inhibition (biomass and architectural traits), whereas cycle17 hybrid roots when grown alone or together were indiscernable. These data strongly suggest a root-root communication mechanism, that has been altered in some way by artificial selection for density adaptation. Several follow up studies are nominated based on these results, including: understanding the intra/interspecific nature of the root-sensing phenomenon, dissecting its spatiotemporal parameters (including using PET imaging to follow dynamic carbon allocation), capturing transcriptional differences in cycle 0 and 17 genotypes upon root-sensing, and isolating and identifying the signalling compound that seems to be moving through the gel from one plant to another. b) IHP and ILP lines - studying the effects on root growth of long-term recurrrent selection for Nitrogen uptake and seed content in maize. In year 1, IHP and ILP lines were phenotyped and an architectural ideotype for Nitrogen uptake was demonstrated (larger, deeper, more highly branched). Subsequent studies in low Nitrogen conditions showed no discernable effect for either IHP or ILP, suggesting that perhaps the IHP genotype is constitutively 'on' for Nitrogen scavenging, whether it's abundant iin the soil or not, whereas the ILP genotype has lost it's ability to respond to N availability. In an effort to identify QTLs and eventually genes responsible for this phenomenon, 26 of 138 families from a high-resolution, multiply-intermated, IHP x ILP RIL population were phenotyped in both the gel system and from field samples, revealing great variation among and high heritability within families for root traits. These findings suggest a high-likelihood for mapping genes driving root architectures adapted to low-Nitrogen tolerance in future studies. Aim 2: Assemble an automated imaging platform to capture root system growth dynamics at fine spatiotemporal scales (deep phenotyping) a) Assemble and test the automated imaging platform. The system was redesigned from its original inception to incorporate superior (telecentric) optics and a more accurate (direct-drive) sample rotation. To double throughput, increase field of view, and due to the high cost of the telecentric optics, an automated linear motion capacity was also added to the design that would enable the imaging apparatus to move precisely back and forth between two imaging tanks. These critical improvements delayed the assembly and testing of the automated in situ imaging platform until very near the end of the project in summer of 2013. Since PI Topp had secured an independent position (asistant Member and Principal Investigator) at the Donald Danforth Plant Science Center, St. Louis, MO, in September of 2013, the system was not fully realized at Duke University, but instead is currently under construction in Topp's lab at the DDPSC. As planned, Topp will use this system to study the fine-scale dynamics of root growth x environment, for example as outlined in Aim 1a, and will attribute the conception of the device to this USDA-NIFA project. As another way of capturing fine scale root dynamics, Topp developed a collaboration with groups at Duke University and the Jefferson National Laboratory to apply Positron Emission Tomography (PET) to the study of carbon allocation to roots in real time. Radiolabelled CO2 was fed to plant leaves and the resulting flow of photosynthate was monitored by PET detectors and computationally reconstructed. By combining with 3D optical gel imaging, root morhpology and carbon allocation were overlayed and correlated, and furthermore the dynamic changes in response to biotic factors such as root competition, fungal interactions, and abiotic factors such as localized nutrient sources could be monitored. Topp developed a similar collaboration with Yuan-Chuan Tai, Department of Radiology, Washington University Medical School, St Louis, MO, in the summer of 2013 to conducted multiple daily studies correlating 3D root growth with carbon allocation. Aim 3: Model root system growth dynamics a) Model growth as a continuous function using skeletons extracted from 3D time series. Progress toward this aim advanced rapidly in year 1, as Topp manually generated several hour-by hour root growth data sets, which his collaborator Olga Symonova, IST-Austria, used to develop a computational technique, Root Growth Model (RGM), capable of quantifying the origin and growth dynamics of each root axis along the time axis. The software was nearing beta-testing phase and publication at a computer vision conference when Symonova took an extended ~1.5 year leave for the birth of her child. Despite the help of co-mentor Harer, the mathematics and computation of the RGM proved to be outside of Topp's capability, and thus remains unfinished until Symonova is back from leave. In the interim, Topp is working with Tai's group to develop a method to co-register optical and PET reconstructions taken twice daily over a week-long period, and measure the corresponding dynamic (growth and carbon allocation) changes.
Publications
- Type:
Journal Articles
Status:
Accepted
Year Published:
2013
Citation:
Topp, C. N., Iyer-Pascuzzi, A. S., Anderson, J. T., Lee, C.-R., Zurek, P. R., Symonova, O., et al. (2013). 3D phenotyping and quantitative trait locus mapping identify core regions of the rice genome controlling root architecture. Proceedings of the National Academy of Sciences of the United States of America, 110(18), E1695704. doi:10.1073/pnas.1304354110
- Type:
Journal Articles
Status:
Other
Year Published:
2014
Citation:
Topp, C.N., Edwards, J., Moose, S. (in preparation) The effects of long-term recurrent selection on maize root architecture.
|
Progress 09/01/11 to 08/31/12
Outputs
OUTPUTS: Progress toward each of the three Aims was made. For Aim 1, RSA of the Illinois High and Low Selection lines under standard conditions were analyzed at each of several time points from day 03 to day 12 of growth. Preliminary experiments to characterize their growth in Nitrogen depleted conditions were conducted. Iowa Stiff Stalk Synthetic (BSSS) cycle 0, cycle 17, c0 x c17 F1 hybrids, BSSS c0 x Corn Borer c0, and BSSS c17 x Corn Borer c17 were also characterized. These lines are currently being grown together to study the effects of planting density on root architecture. For Aim 2, a novel and fully automated root imaging platform that allows 24/7 continuous monitoring of growth was developed and is currently under construction. Preliminary time series data was captured manually and used to develop the software in Aim 3. For Aim 3, a collaboration was fostered with Olga Symonova and Herbert Edelsbrunner at the Institute of Science and Technology Austria to develop and implement Root Growth Model software for the analysis of time series data. These activities and products were described at several conferences and workshops, including: The Maize Genetics Meeting in Portland, Oregon; Plant Biology Department Seminar at Cornell University; Plant Molecular Biology Retreat Asheville, NC; Plant Growth Modelling Meeting Elche, Spain; Automated Imaging and High-Throughput Phenotyping Cold SPring Harbor, NY; and and the International Workshop on Image Analysis Methods for the Plant Sciences Nottingham, UK. Two Duke undergraduates-now-postgraduates are also being mentored in root growth analysis and software development. A two week section on "The Role of Biotechnology in Modern Agriculture" was developed and taught at North Carolina Central University, a historically African-American school. PARTICIPANTS: Victoria Arendt (undergradute) Duke. I supervised Victoria's honors thesis project in which she used her strong computational skills to develop root parsing software for 3D root models. I trained Victoria in wet lab skills necessary to conduct experiments that allowed her to ground-truth her program. Molly Cinderella (undergradute) Duke. Molly has been working with me to further refine Victoria's software as well as software related to the Root Growth Model. I am training her in root developmental biology to complement her strong programming and computational skills. Herbert Edelsbrunner (professor) and Olga Symonova (post-doctoral associate), Institute of Science and Technology Austria. I work with Olga to develop the conceptual framework for the Root Growth Model, including a week-long trip to the IST to work first-hand with the computational group there. Steve Moose (professor), University of Illinois Urbana-Champaign. Steve is a maize geneticist/ agronomist with whom I collaborate on the Illinois High and Low protein project. Jode Edwards (research scientist) USDA-ARS and (adjunct professor) Iowa State. Jode is a plant geneticist/breeder who conducts experiments on the effects of planting density on maize agricultural traits. TARGET AUDIENCES: North Carolina Central University BIOL4660 'The genetics of complex traits' students. I taught a four lecture course on 'The role of biotechnolgy in modern agriculture'. These upper-level undergraduate students are interested in pursuing careers in some aspect of biology. PROJECT MODIFICATIONS: Nothing significant to report during this reporting period.
Impacts
Comparisons between Illinois High and Low Protein maize lines provided clear quantitative evidence for RSA differences that are relatable to their function in Nitrogen uptake. Previous evidence suggests that increased Nitrogen uptake capacity in the IHP lines was not due to increased transporter activity, thus we have identified an architectural idiotype that reflects superior Nitrogen uptake. An existing cycle 90 IHP x ILP mapping population could be used to isolate QTLs and subsequently genes underlying this trait. For the BSSS maize lines, architectural differences that resulted from selection for high yield at high planting density were quantified for the first time. These results provide a baseline from which to compare ongoing functional growth response to density experiments, and support the popular hypothesis that root architectural changes underlie the historical ability of breeder to improve yields through selection for higher planting densities. The design and implementation of the automated real-time root imaging platform funded here is a groundbreaking achievement in that it will allow for the first time the continuous monitoring of 3D RSA in a crop plant. This tool will allow the growth and development of each root axis to be monitored, and thus allow us to determine how local growth decisions contribute to global architecture, as well as developmental responses to local nutrient and other cues. Concomitantly, the Root Growth Model software is a powerful infrastructure for quantifying the growth and developmental responses from the real-time imaging platform. I have gained key computational and mathematical knowledge during its collaborative development with Olga Symonova. In a similar vein, my mentorship of two Duke undergraduates with strong computational backgrounds has helped lead to their strong interests in plant biology.
Publications
- No publications reported this period
|