Scientists at the Department of Energy’s Oak Ridge National Laboratory (ORNL) have created a platform that can pinpoint genetic triggers that turn microbes into efficient factories for new chemicals and materials. The platform identifies specific genetic triggers for useful complex traits, supporting the design and reprogramming of microbes that exhibit targeted capabilities. Potential applications of such bacterial factories might include the breakdown and conversion of plant lignin into valuable products, or the uptake of critical minerals.
The team’s approach, which combines synthetic biology expertise, artificial intelligence, and statistical mapping techniques, enables rapid, precise reprogramming of bacteria as biotechnology tools, and builds on previous work by ORNL scientists, who adapted a technique called protoplast fusion to create the diverse microbial offspring needed for genetic mapping. “Unlike past approaches that study the effect of gaining or losing whole genes, the new approach lets us determine how small differences in the nucleotide sequence affect bacterial function,” said Josh Michener, PhD, project co-lead and Biological Systems Design group leader at ORNL. “Variations in strains at the nucleotide level have a huge impact on the resulting phenotype, especially when you’re engineering microbes with specific mutations. The method also lets us study natural mutations in parental strains that make them ideal biotechnology tools.”
Michener is co-senior and co-corresponding author of the team’s report in Nature Communications (“Genome shuffling enables quantitative trait locus mapping in Bacillus subtilis”) in which they reported on the development of their platform and its validation using gene editing in bacteria.
To determine which genes control certain characteristics in organisms, scientists have used a method called quantitative trait locus (QTL) mapping. QTL mapping involves analyzing the traits of lots of varied offspring from two distinct parents, and is a common approach in mapping the genes of other organisms such as plants. The method examines many genetic differences at once and precisely identifies candidate genes in a single workflow.
However, linking DNA sequences to observable physical traits in bacteria is challenging, the authors noted. “Even in the best-studied model bacteria, many genes have unknown functions, and little is known about the genetic networks underlying complex phenotypes or the functional effects of natural sequence variation in bacterial genes and regulatory elements.”
The problem with applying QTL mapping to bacteria is that these microorganisms reproduce asexually with limited genetic variation. “Quantitative trait locus (QTL) mapping generally relies on sexual recombination to break linkages between genes, yet bacteria rarely undergo sufficient homologous recombination to generate suitable mapping populations,” the team continued.
ORNL researchers overcame the hurdles associated with applying QTL to bacteria using protoplast fusion, a tool first developed in the 1970s. Using the fusion technique, researchers were able to cross Bacillus strains, producing a large population of genetically varied offspring, called recombinants. “We have previously shown that genome shuffling by protoplast fusion between genetically diverse Bacillus strains generates frequent, unbiased, genome-wide recombination that mimics the effects of sexual recombination,” they noted. Bacillus are model bacteria that serve as workhorses for fermentation, enzyme production, and plant growth and health. Referring to their newly published paper, the team added, “In this study, we leveraged protoplast fusion to establish a bacterial QTL mapping platform.”
Researchers measured properties of the bacteria and identified DNA variants that could explain the differences in those traits. The team tested the method across several other bacterial groups, demonstrating alternative genome shuffling methods that expand the tool’s usability on different types of microbes used as biotechnology tools. These included Clostridium thermocellum, a bacterium that tolerates industrial processes and is good at breaking down and fermenting plant cellulose. Also in Novosphingobium aromaticivorans, a bacterium that excels at breaking down aromatic compounds from plant lignin and converting the molecules into high-value chemicals. Also in Stutzerimonas stutzeri, a versatile bacterium used in applications such as bioremediation and to fix nutrients in soil, supporting plant growth and suppressing plant pathogens.
They validated their findings by using CRISPR gene editing tools to swap gene sections and confirm the effects in bacteria. “We have now, for the first time ever, put all these pieces together for a platform that gets results on complex genes-to-traits linkages much faster,” Michener said. “We built the genetically diverse bacteria population, identified DNA variants, and confirmed the work with gene editing.”
The authors added, “In contrast to traditional loss-of-function and gain-of-function genetic methods, our approach enables rapid detection of the effects of natural genetic variation in both coding and noncoding regions on bacterial phenotypes, beyond gene presence or absence.”
By creating such broad diversity in the bacterial offspring, scientists faced a challenge in the research: phenotyping all the progeny. They tackled it with automation and AI, setting up a robotic system to quickly and repeatedly place plates with precision so that high-resolution digital imaging could be accomplished at the same angle and lighting for comparable data between the recombinants. The phenotyping was accomplished 10 times faster with automation, the scientists noted.
Getting consistent data was crucial to the application of mathematical algorithms and the use of a computer vision model that processed the images and extracted traits, explained co-lead Dan Jacobson, ORNL computational systems biologist. “We built this project with a very multidisciplinary lineup,” Jacobson said. “The team did everything from building the robotics, conducting imaging and image processing, performing the statistical work, the mapping and assemblies, the genome shuffling work, growing these different isolates from the population and extracting DNA to send for sequencing, then growing them again for the phenotype assays. It’s an example of the kind of good collaboration that’s possible at a national lab, and how that research can enable whole new areas of inquiry across the nation’s science ecosystem.”
Scientists continue to deploy the method to study and engineer microbes for better manufacturing processes as part of the DOE Center for Bioenergy Innovation (CBI) at ORNL. The platform is also being used to study plant-associated microbes as part of the DOE Secure Ecosystem Engineering and Design Science Focus Area (SEED SFA), as well as by a program at Colorado State University studying airborne microbes.
The microbial QTL mapping platform is available for licensing at ORNL.
The post Bacterial Trait-Finding and Gene Mapping Platform Speeds Microbial Engineering for Biotech Applications appeared first on GEN – Genetic Engineering and Biotechnology News.
