News J-WAFS supports interdisciplinary research that could launch a new approach to making agriculture more productive

Improving an inefficient enzyme could unlock potential for dramatic increases to crop yields.

By Robbie Wilson, PhD, Research Scientist MIT Department of Chemistry March 12, 2025

As J-WAFS highlights its impact on MIT water and food research to celebrate its 10th anniversary this month, one notable observation is the increase over the past ten years in research involving plant biology and genetic engineering to address the serious challenge that climate change presents to our agricultural productivity. MIT has long been a leader in synthetic biology and genetic engineering. J-WAFS is increasingly supporting the application of this expertise to food and agriculture challenges. In one notable project, MIT research scientist Robbie Wilson is working with Professor Matt Shoulders of the Department of Chemistry and an interdisciplinary team to advance a way to boost crop yields. Their J-WAFS Grand Challenge funding is supporting the innovative application of molecular biology, chemistry, and computer science to a long-standing challenge. In this piece, Wilson explains more about the research and their progress thus far.

 

Most people are aware that life is carbon-based. Our DNA, proteins, and cell membranes are built around carbon. Most are probably not aware that virtually all the carbon present in life (termed organic carbon) is dependent on an enzyme called RuBisCO (Ribulose-1,5-bisphosphate carboxylase/oxygenase). RuBisCO is responsible for one of the most critical biochemical processes—converting organic carbon (CO2) into sugar, and subsequently all of the other building blocks of the cell.

After billions of years in this role and given the sheer abundance of biomass in nature, one might expect RuBisCO to be one the fastest and most efficient enzymes on Earth. However, RuBisCO is not only metabolically slow, reacting with only a few molecules of CO2 each second, but it also reacts unproductively with oxygen, and is constantly in need of help by other proteins (chaperones). These limitations can fundamentally constrain photosynthesis in plants. This makes improving RuBisCO a desirable strategy for increasing plant biomass and crop yields.

Identifying mutations that can improve the function of RuBisCO is the central goal of the Enhanced Photosynthesis in Crops (EPiC) team at MIT, supported by the inaugural J-WAFS Grand Challenge grant. The researchers are using state-of-the-art protein engineering techniques drawn from biomedicine and artificial intelligence to design better crop RuBisCO in the laboratory.

The team developed a semi-continuous directed evolution platform for RuBisCO which has been used to evolve the fastest RuBisCO enzyme ever measured. Directed evolution is a powerful protein engineering tool that can search millions of changes in a molecule and filter them according to an improved function. In the case of the EPiC team, the evolved RuBisCO improved in its ability to function in the presence of inhibitory oxygen. A preprint paper (meaning it has not yet completed peer review) was recently published on this work.

“RuBisCO’s counterproductive reaction with oxygen instead of CO2 has long been viewed as a massive detriment to the enzyme, and our work begins to uncover ways to minimize the effects of said reaction,” says Julie McDonald, a PhD candidate in the MIT Biology Department and member of the EpiC team. “This study, among many other exciting new works, is the start of what we anticipate to be a revolution in RuBisCO research, where many more beneficial changes to RuBisCO can be discovered,” McDonald adds. 

In another preprint paper, mutations to RuBisCO have been demonstrated to improve the photosynthetic and water use efficiency of Arabidopsis thaliana, a small plant from the mustard family. The research was conducted by Associate Professor Wataru Yamori and his team in Japan, who used gene editing to install mutations into the RuBisCO gene located in the chloroplast genome. These mutations were previously found using directed evolution on RuBisCO from a cyanobacterium (Wilson et al. 2018. JBC). Using a similar process, these same principles are now being applied at MIT on plant enzymes to identify, predict, install, and screen improvements to RuBisCO’s kinetic activity using high-throughput pipelines.

Bringing the AI revolution to bear on difficult protein engineering challenges is one of the hottest topics in science at the moment. Improving the activity of RuBisCO is certainly up there on the list of ‘most challenging enzymes to engineer’ and a central focus of our J-WAFS Grand Challenge project. The EPiC team is applying AI techniques to RuBisCO to understand how the active site of the enzyme interacts with its substrates in order to identify regions that can be improved. 

RuBisCO engineering is part of a staged expression pipeline from bacteria to vascular plants to characterize RuBisCO variants and validate mutants. In 2024, we began examining how enzyme variants respond when returned to the chloroplast environment. In collaboration with Mary Gehring, a member of the EPiC team whose lab is in MIT’s Department of Biology and the Whitehead Institute, we are expressing these proteins in leaves using specialized lines of Nicotiana inside the MIT Greenhouse.

With another year left in the grant period, we look forward to continuing progress on this epic grand challenge. The support of J-WAFS has been enormously important, as it has provided funding of sufficient magnitude to enable this interdisciplinary, multi-PI team to come together to work on a new and innovative collaboration, addressing a hard but promising challenge. It has allowed enthusiastic and talented MIT graduate students and postdocs to apply themselves to this important work. And it has enabled the development of MIT as a hub for RuBisCO engineering.