Variability among lipids affects cellular metabolism, signaling pathways, energy storage, and cell membrane structures. Therefore, it’s important for biomanufacturers to have an accurate, straightforward way to characterize lipids at the single-cell level. Existing methods, however, are not comprehensive and may not reflect cellular-level differences within any given population.
“This is partly because no validation or comparison of single-cell lipidomics methods has ever been performed,” Melanie J. Bailey, PhD, professor, University of Surrey, tells GEN.
Bailey and colleagues at the University of Surrey, recently developed a high-throughput microfluidics-based method to measure and analyze lipids in single living cells using liquid chromatography-mass spectrometry (LC-MS). With it, she says, they “identified approximately 200 lipid features in single cells, with the ability to isolate 96 cells in just a few minutes.” They compared this method to capillary sampling for cell isolation, which is slower, but preserves spatial information.
“Importantly, we found that microfluidics and capillary sampling detect very similar lipid profiles,” Bailey says. “The fact that two independent methods with different strategies for blank correction detect the same single-cell lipid profiles is an exciting step in the validation of single-cell lipidomics technologies.”
The microfluidics workflow “facilitates automated sampling, data acquisition, and analysis, and carries the additional advantage of chromatographic separation,” which reduces variations in detection and quantification, according to the team, writing in a recent paper.
In this method, spatial information is exchanged for speed and low cost. Challenges for single-cell lipidomics include efficiently isolating single live cells, extracting and separating lipids from minuscule sample volumes, and compatibility with LC-MS. They can be overcome, the scientists assure, by optimizing single-cell sampling techniques.
For example, Bailey and colleagues found high levels of lipid contamination in blank samples despite adhering to the manufacturers’ protocols. “Around 50% of the lipid features found in cells were also detected in the instrument blanks…,” they report. After optimization, only 15% of the lipid features found in the cells were found in the mobile phase blanks.
Optimization steps focused on exchanging the bovine buffer used to coat Petri dishes for a human buffer that was diluted 10-fold, and removing fetal bovine serum from the media. To preserve cell viability, a concentration of 150 nM of ammonium formate fixed to physiological conditions proved most effective.
Microfluidics compares favorably to capillary sampling. Specifically, “The average number of lipids detected per single cell is not statistically different,” the scientists report, although capillary sampling showed greater variation in the number of lipids recovered per single cell. That difference may have been caused by manually transferring cells into vials for capillary sampling, rather than the automatic transfer that occurs with microfluidics.
Future work may include extending this microfluidics workflow to adherent and suspension cells, as well as reducing noise from blanks, and limiting sample dilution and the effects of blank correction.
