Public Abstract

DE-SC0020366: High-throughput determination of a subcellular metabolic network map of plants

Award Status: Inactive

Institution: Carnegie Institution of Washington, Washington, DC
UEI: ZQ12LY4L5H39
DUNS: 072641707

Most Recent Award Date: 05/26/2023
Number of Support Periods: 3
PM: Rabinowicz, Pablo

Current Budget Period: 09/15/2021 - 09/14/2024
Current Project Period: 09/15/2019 - 09/14/2024
PI: Rhee, Seung

Supplement Budget Period: N/A

Public Abstract

High-throughput determination of a subcellular metabolic network map of plants Seung Y. Rhee, Carnegie Institution for Science (Principal Investigator) David W. Ehrhardt, Carnegie Institution for Science (Co-Investigator) Markita del Carpio Landry, UC Berkeley (Co-Investigator) Jenny Mortimer, Lawrence Berkeley National Lab (Co-Investigator) Climate change is rapidly changing the habitability of many species on Earth. Plants have a big role in slowing down climate change as major carbon sequesters, solar energy harvesters, and food providers for many species including humans. Plant metabolism underpins many traits that improve plant productivity but there are many holes in our understanding of how it works as a system. To quantify, model, predict, and engineer desirable metabolic traits such as to maximize biomass production under suboptimal conditions or reallocation of biomass from carbohydrates to lipids, we must decode the complex metabolic networks. Subcellular compartmentation of metabolic reactions through the locations of enzymes is critical to understanding, modeling, and engineering plant metabolism. Yet, the localization of the majority of the predicted enzymes are not yet known. The paucity of experimentally validated information in most plants, especially in all the DOE flagship plants, severely limits scientists and engineers to assess the performance and translatability of computational tools and resources. In this proposed work, a trans-disciplinary team with expertise in plant cell biology, genomics, metabolic modeling, algorithm development, synthetic biology, geochemistry, nanotechnology, and analytical chemistry will develop an integrated pipeline that combines computational prediction, metabolic network modeling, and high-throughput experimental testing using state of the art technologies in live confocal imaging, nanomaterial-mediated plant transformation, and metabolic network modeling. Using the pipeline, the team will create a high-quality subcellular map of small molecule metabolism in Sorghum andBrachypodium. Using this network, the team will create accurately compartmentalized metabolic network models. The models will be experimentally validated by measuring a series of outputs in response to environmental challenges, and by knocking out gene expression in somatic tissue using carbon nanotube-mediated CRISPR technology. This project has several novel components that could have a wide impact to a range of fields such as Artificial Intelligence, Metabolic Engineering, Cell Biology, Computational Biology, Plant Science and Biochemistry. It will create a set of ground-truth data for subcellular locations of metabolic enzymes, develop a novel integrative pipeline for accelerating knowledge generation, further develop a novel plant transformation technology with potentially broad application, and yield important insight into the structure and function of metabolic networks in both a biofuel model plant and a biofuel crop. This will be the first time such a large-scale, high-resolution imaging-based localization information of metabolic enzymes is generated for any plant. This dataset will have long-lasting value for benchmarking machine learning algorithms for localization prediction and image annotation. Moreover, the cellular view of metabolism at this scale has not been approached before. This project has the potential to create a new window into our understanding of metabolism in the cellular context. Finally, the novel approach of using carbon nanotubes to transform somatic cells in the leaf meristem to generate clonal sectors and perform metabolic network modeling and validation can rapidly test the validity of the models and enable rapid cycles of metabolic engineering for important traits.

High-throughput determination of a subcellular metabolic network map of plants
Seung Y. Rhee, Carnegie Institution for Science (Principal Investigator)
David W. Ehrhardt, Carnegie Institution for Science (Co-Investigator)
Markita del Carpio Landry, UC Berkeley (Co-Investigator)
Jenny Mortimer, Lawrence Berkeley National Lab (Co-Investigator)

Climate change is rapidly changing the habitability of many species on Earth. Plants have a big role in slowing down climate change as major carbon sequesters, solar energy harvesters, and food providers for many species including humans. Plant metabolism underpins many traits that improve plant productivity but there are many holes in our understanding of how it works as a system. To quantify, model, predict, and engineer desirable metabolic traits such as to maximize biomass production under suboptimal conditions or reallocation of biomass from carbohydrates to lipids, we must decode the complex metabolic networks. Subcellular compartmentation of metabolic reactions through the locations of enzymes is critical to understanding, modeling, and engineering plant metabolism. Yet, the localization of the majority of the predicted enzymes are not yet known. The paucity of experimentally validated information in most plants, especially in all the DOE flagship plants, severely limits scientists and engineers to assess the performance and translatability of computational tools and resources.

In this proposed work, a trans-disciplinary team with expertise in plant cell biology, genomics, metabolic modeling, algorithm development, synthetic biology, geochemistry, nanotechnology, and analytical chemistry will develop an integrated pipeline that combines computational prediction, metabolic network modeling, and high-throughput experimental testing using state of the art technologies in live confocal imaging, nanomaterial-mediated plant transformation, and metabolic network modeling. Using the pipeline, the team will create a high-quality subcellular map of small molecule metabolism in Sorghum andBrachypodium. Using this network, the team will create accurately compartmentalized metabolic network models. The models will be experimentally validated by measuring a series of outputs in response to environmental challenges, and by knocking out gene expression in somatic tissue using carbon nanotube-mediated CRISPR technology.

This project has several novel components that could have a wide impact to a range of fields such as Artificial Intelligence, Metabolic Engineering, Cell Biology, Computational Biology, Plant Science and Biochemistry. It will create a set of ground-truth data for subcellular locations of metabolic enzymes, develop a novel integrative pipeline for accelerating knowledge generation, further develop a novel plant transformation technology with potentially broad application, and yield important insight into the structure and function of metabolic networks in both a biofuel model plant and a biofuel crop. This will be the first time such a large-scale, high-resolution imaging-based localization information of metabolic enzymes is generated for any plant. This dataset will have long-lasting value for benchmarking machine learning algorithms for localization prediction and image annotation. Moreover, the cellular view of metabolism at this scale has not been approached before. This project has the potential to create a new window into our understanding of metabolism in the cellular context. Finally, the novel approach of using carbon nanotubes to transform somatic cells in the leaf meristem to generate clonal sectors and perform metabolic network modeling and validation can rapidly test the validity of the models and enable rapid cycles of metabolic engineering for important traits.