“We foresee that rapid data acquisition through O-PTIR imaging will significantly aid in understanding and managing phenotypic diversity in microbial cells by providing a detailed representation of individual cells for population statistics.”
FULL PUBLICATION >
Reporting in Microbial Cell Factories, researchers at Norwegian University of Life Sciences and Synchrotron SOLEIL demonstrated the use of optical photothermal infrared (O-PTIR) spectroscopy for assessing single-cell phenotypic diversity in microbial lipid production. Microbial production of single-cell oils (SCOs) utilizes diverse microorganisms including bacteria, yeasts, and microalgae, but minor physicochemical heterogeneity in bioreactor environments leads to phenotypic diversity at the single-cell level. This metabolic heterogeneity can impact production of desired lipid types due to variations in gene expression, enzyme activity, and environmental gradients. While flow cytometry, the current predominant tool, is used to investigate phenotypic heterogeneity, it has limitations including difficulty with cell aggregates and reliance on fluorescent markers that may not exist for all molecules. Importantly, specific markers capable of distinguishing free fatty acids (FFAs) from triacylglycerols (TAGs) are currently unavailable for flow cytometry and fluorescence microscopy.
The research team measured 40 individual Rhodotorula graminis yeast cells using O-PTIR, acquiring six single-point spectra per cell for a total of 240 spectra. Cell sizes were estimated from microscopy images and analyzed alongside O-PTIR spectral data. The analysis revealed that while most cells had similar chemical composition, several cells differed significantly from the population average, and numerous cells exhibited relatively large intra-cell chemical variability. Principal component analysis demonstrated that the main chemical differences correlated with cell size. Small cells showed higher protein content than mid-size and large cells, consistent with active growth phases. Large cells exhibited higher TAG-to-FFA ratios compared to mid-size cells, with the elevated FFA content in mid-size cells potentially representing a transitional metabolic phase.
The study demonstrated statistically significant size-dependent differences in cellular chemistry, with the ratio of intensities at 1748 and 1714 cm⁻¹ differing significantly among three cell size groups. The researchers identified characteristic wavenumbers for major constituents: 1748 cm⁻¹ for TAGs, 1714 cm⁻¹ for FFAs, and 1659 cm⁻¹ for proteins. Importantly, simulation of sparse wavenumber data acquisition showed that measurement at only these three wavenumbers plus one reference wavenumber at 1800 cm⁻¹ provided assessment of major chemical constituents. This sparse-mode approach would allow use of a more economical single-stage quantum cascade laser device covering only 1801-1520 cm⁻¹, rather than the four-stage broadband device used in the study.
O-PTIR spectroscopy provided label-free chemical characterization of individual cells at sub-micron spatial resolution (approximately 500 nm), overcoming the diffraction limit of conventional FTIR microspectroscopy which cannot measure individual yeast cells. The technique successfully differentiated two lipid classes (TAGs and FFAs) that cannot be distinguished by existing flow cytometry or fluorescence microscopy methods. The ability to rapidly acquire data through sparse-mode O-PTIR imaging at just four wavenumbers enables measurement of hundreds of cells for representative population statistics. Combined with automated bioreactor sampling and sample processing, this approach could enable characterization of phenotypic diversity within approximately one hour of harvesting, providing valuable insights for understanding and managing heterogeneity in industrial microbial lipid production processes.
Authors: Uladzislau Blazhko, Dana Byrtusová, Volha Shapaval, Achim Kohler, Christophe Sandt, and Boris Zimmermann
Faculty of Science and Technology, Norwegian University of Life Sciences, Norway
