Scientists at the Joint BioEnergy Institute (JBEI) have introduced two groundbreaking methods to enhance the engineering of microbes for jet fuel production. These innovations could cut development timelines from years to mere weeks, significantly impacting the biofuel industry.
Héctor García Martín, director of Data Science and Modeling at JBEI, emphasized the potential of these approaches, stating, “If widely adopted, these methods could reshape the industry. Instead of taking a decade and hundreds of people to develop one new bioproduct, small teams could do it in a year or less.” One of these methods employs artificial intelligence and lab automation to boost isoprenol production five-fold, while the other utilizes a genetic biosensor, achieving a remarkable 36-fold increase in fuel titers.
Innovative Approaches to Biofuel Production
The primary focus of this research is isoprenol, a precursor that can be converted into DMCO, creating a jet fuel alternative with higher energy density than traditional petroleum-based fuels. As current battery technology does not provide the necessary energy density for aviation, this synthetic alternative aims to meet the industry’s growing demands.
The research highlights how automation and discovery-based sensing can optimize microbial systems for commercial applications. The teams employed two distinct engineering strategies to enhance bio-manufacturing efficiency. One method combines artificial intelligence with lab automation to rapidly test and refine genetic designs of biofuel-producing microbes. The second approach leverages a microbe’s inherent ‘bad habits’ as a powerful sensing tool, revealing previously hidden pathways that enhance production.
Automation and Sensing for Enhanced Efficiency
Led by Taek Soon Lee and Héctor García Martín, a team created an automated pipeline designed to minimize reliance on human intuition in metabolic engineering. The researchers developed a system that uses robotics to generate and evaluate hundreds of genetic designs simultaneously. A custom microfluidic electroporation device was introduced, capable of inserting genetic material into 384 strains of Pseudomonas putida in under one minute, a task that typically takes hours manually.
This rapid process fosters a continuous learning loop. Machine learning models analyze protein measurements and propose specific gene combinations for adjustment via CRISPR interference. By completing six engineering cycles within weeks, the team identified genetic combinations that significantly increased fuel concentration.
Meanwhile, a second team, led by Thomas Eng, tackled the issue of Pseudomonas putida consuming the isoprenol it produces. They uncovered two proteins the microbe uses to detect the fuel and reengineered this system into a biological biosensor. This innovative twist linked the sensor to essential survival genes, creating a system where only the microbes that produce the highest amounts of fuel can thrive.
The biosensor approach allowed the team to screen millions of variants without the need for manual measurements. It was discovered that high-producing strains can sustain their production by adjusting their metabolism to utilize amino acids once glucose supplies are depleted.
The research teams are now focused on transitioning these engineered microbial strains from laboratory environments to industrial fermentation systems. The dual strategies of depth-first AI optimization and breadth-first biosensor discovery create a robust framework applicable to various bio-based products.
As the demand for sustainable aviation fuel continues to rise, these advancements in microbial engineering represent a significant leap forward in the quest for environmentally friendly energy sources.
