Collaborative Research: PlantSynBio: Identification and Design of Transcriptional Activation Domains Across Plant Species

Every plant begins life as a seed with one cell and one genome. In order for a plant to grow, the cells must divide and turn into different kinds of cells by turning on different genes. Leaf cells turn on leaf genes; root cells turn on root genes.

This process of turning genes on and off requires specialized proteins called transcription factors. Transcription factors also allow plants to turn genes on or off to respond to stresses like drought or pests and pathogens. Transcription factors turn genes on using specialized regions called “activation domains.” The first part of this project will use technologies we developed to identify activation domains on all the transcription factors of one widely studied plant species, Arabidopsis.

The data we collect will power our computational models for predicting activation domains in other plants. The final portion of this project will build synthetic transcription factors that can be used to engineer gene regulation in other plants. This research will create useful and powerful tools for plant biologists and plant breeders.

This collaboration between three research teams creates a unique interdisciplinary training environment for undergraduates, graduate students, and postdoctoral research fellows. Our team is committed to building a supportive environment that fosters Equity, Diversity and Inclusion. The three PIs come from backgrounds that have been traditionally excluded from science. Two of the PIs are building on the success of their NSF CAREER awards.

In plants, transcription factors control gene regulatory programs for development, growth and stress responses. Transcription factors have two functions: 1) to bind DNA sequences in the genome directly with a DNA binding domain (DBD) or through a partner DBD-containing protein and 2) to recruit transcriptional machinery. DBDs have been well characterized and can be predicted directly from amino acid sequence.

In contrast, the regions of transcription factors that bind coactivator complexes, activation domains, remain poorly characterized and cannot be predicted from amino acid sequence. For example, in Arabidopsis, there are 1,717 transcription factors, but only 8 known activation domains. As a consequence, when a new genome is sequenced, models for predicting DBDs can identify putative transcription factors, but there are no analogous models for predicting if these transcription factors are activators or repressors.

This project will use high throughput screening methods to identify activation domains on all Arabidopsis transcription factors. These data will train deep learning neural networks to predict activation domains from amino acid sequence and predict activation domains in other diverse plant species. The final portion of this project will create and validate synthetic transcription factors for engineering gene regulation in plants.

These tools will expand the synthetic biology toolbox for targeted hypothesis testing of metabolic processes, engineering regulatory networks, advancing agriculture and contributing to solutions that could address environmental problems.

This award was co-funded by the Plant Genome Research Program in the Division of Integrative Organismal Systems and the Systems and Synthetic Biology Cluster in the Division of Molecular and Cellular Biosciences.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.