Public Abstract

DE-SC0025892: AI-empowered high-resolution multimodal imaging of microbes under extreme conditions

Award Status: Active

Institution: Rensselaer Polytechnic Institute, Troy, NY
UEI: U5WBFKEBLMX3
DUNS: 002430742

Most Recent Award Date: 05/01/2025
Number of Support Periods: 1
PM: Sammak, Paul

Current Budget Period: 02/01/2025 - 01/31/2026
Current Project Period: 02/01/2025 - 01/31/2027
PI: Royer, Catherine

Supplement Budget Period: N/A

Public Abstract

Amazingly, the vast majority of the microbial biomass on Earth is found is environments we humans would consider extreme. These extreme conditions include the high temperatures (>100 degrees C) and high pressures (up to 500 atmospheres) found at oceanic hydrothermal vents, the highly acidic and high temperatures found in hot springs, such as in Yellowstone National Park, the high pressures (up to 1100 atmospheres) and cold temperatures (3 degrees C) in the deepest ocean trenches, and even the high pressures and temperatures inside the continental crust (>3000 atmospheres). We know that the fragile molecules of life are extremely sensitive to extreme conditions such as these. It is therefore astounding that these microbes not only live but can thrive under such conditions. Given their preference, or even requirements for growth under extreme conditions such organisms have not been studied in-depth. This is because such studies necessitate complex equipment to attain the extreme conditions required by these microbes. In particular, very little live cell imaging of such organisms has been reported, given the complexities of high pressure, extreme temperature imaging. Many of these extremophile microbes are methanogenic archaea or sulfate reducing or oxidizing organisms that contribute significantly to geochemical carbon and sulfur cycles on Earth. Despite their promise for methane mitigation, bioremediation, or industrial catalysis applications, little is known at the single cell level about the dependence and dynamics of their growth and metabolism on environmental conditions. Analyses of these organisms in bulk conditions obscures the underlying cell-to-cell variability. Moreover, how growth conditions affect their cell size and metabolism, and thus their fitness, cannot be deeply understood from measurements of bulk populations. The objective of this project is to implement and optimize an experimental prototype imaging system for high-resolution, quantitative single cell imaging of extremophile organisms over a broad range of temperatures and pressure. We will develop and implement Artificial Intelligence (AI) assisted image analysis approaches in order to test our hypothesis growth and metabolism of these organisms is tuned to the environmental conditions in which they are found in nature. We have developed a specialized temperature regulated capillary-based system to be able to image single cells at maximum optical resolution employing multiple highly quantitative imaging approaches such as particle counting, intracellular diffusion and fluorescence lifetime based metabolic fingerprinting. For this unique hybrid imaging system, we will design, validate and optimize AI-based image analysis techniques for both image formulation and quantitative analysis results based on prior work on the development of deep learning-based methods for image denoising and clustering. These pioneering single cell high resolution imaging studies of extremophile organisms under conditions of extreme pressure and temperature will provide unique insights into the evolution of life on Earth, with implications for the search for life elsewhere, as well as potential applications in environmental remediation and biotechnology.

Amazingly, the vast majority of the microbial biomass on Earth is found is environments we humans would consider extreme. These extreme conditions include the high temperatures (>100 degrees C) and high pressures (up to 500 atmospheres) found at oceanic hydrothermal vents, the highly acidic and high temperatures found in hot springs, such as in Yellowstone National Park, the high pressures (up to 1100 atmospheres) and cold temperatures (3 degrees C) in the deepest ocean trenches, and even the high pressures and temperatures inside the continental crust (>3000 atmospheres). We know that the fragile molecules of life are extremely sensitive to extreme conditions such as these. It is therefore astounding that these microbes not only live but can thrive under such conditions. Given their preference, or even requirements for growth under extreme conditions such organisms have not been studied in-depth. This is because such studies necessitate complex equipment to attain the extreme conditions required by these microbes. In particular, very little live cell imaging of such organisms has been reported, given the complexities of high pressure, extreme temperature imaging. Many of these extremophile microbes are methanogenic archaea or sulfate reducing or oxidizing organisms that contribute significantly to geochemical carbon and sulfur cycles on Earth. Despite their promise for methane mitigation, bioremediation, or industrial catalysis applications, little is known at the single cell level about the dependence and dynamics of their growth and metabolism on environmental conditions. Analyses of these organisms in bulk conditions obscures the underlying cell-to-cell variability. Moreover, how growth conditions affect their cell size and metabolism, and thus their fitness, cannot be deeply understood from measurements of bulk populations. The objective of this project is to implement and optimize an experimental prototype imaging system for high-resolution, quantitative single cell imaging of extremophile organisms over a broad range of temperatures and pressure. We will develop and implement Artificial Intelligence (AI) assisted image analysis approaches in order to test our hypothesis growth and metabolism of these organisms is tuned to the environmental conditions in which they are found in nature. We have developed a specialized temperature regulated capillary-based system to be able to image single cells at maximum optical resolution employing multiple highly quantitative imaging approaches such as particle counting, intracellular diffusion and fluorescence lifetime based metabolic fingerprinting. For this unique hybrid imaging system, we will design, validate and optimize AI-based image analysis techniques for both image formulation and quantitative analysis results based on prior work on the development of deep learning-based methods for image denoising and clustering. These pioneering single cell high resolution imaging studies of extremophile organisms under conditions of extreme pressure and temperature will provide unique insights into the evolution of life on Earth, with implications for the search for life elsewhere, as well as potential applications in environmental remediation and biotechnology.