Directed evolution using Raman activated cell sorting

Current challenge

Directed evolution is a powerful methodology that can be used to develop new strains of microorganisms optimized for the alternative protein industry. Directed evolution involves growing cells for several generations and selecting the best performing cells for the trait of interest (e.g., higher protein production) in each generation for continued growth. The key hurdle in directed evolution is identifying a selection tool that quantitatively measures the trait of interest on the single-cell level with high throughput.

Proposed solution

Flow cytometry is a technique that can analyze cells on the single-cell level and sort them with high throughput. Cells are commonly processed at 1000 to 100,000 cells per second (events per second, EPS), making it an attractive tool for directed evolution. However, flow cytometry cell sorting commonly uses fluorescent signals rather than metabolite quantification, limitingresearchers to measuring cellular traits for which a fluorescent marker is available that doesn’t harm the cells.

Raman spectroscopy provides molecular fingerprint identification without fluorescent labeling. However, Raman spectroscopy has a weak signal, limiting throughput to 1 EPS. Several Raman methods were developed that can increase the signal and improve throughput. Coherent Raman spectroscopy (CRS) improves signal acquisition time by three to six orders of magnitude. 

The combination of CRS and flow cytometry can be used for cell sorting based on the level of key metabolites of interest, such as the amounts of lipids, proteins, and polysaccharides or even the quantification of  hemoglobin. This can facilitate directed evolution toward the measured parameters by isolating a fraction of the cell population with the highest signal of a desired metabolite in every generation. Raman flow cytometry has been used to analyze cell populations of microalgae, yeast, and mammalian stem cells and adipocytes. Lipids show a strong Raman signal, which researchers used to generate several papers on adipogenesis. For example, Nitta et. al. were able to sort and clone a rare microalgae strain with high fat content and sort differentiated mammalian adipogenic cells using CRS. Raman flow cytometry can also be used with imaging for the localization of metabolites inside cells.

Research opportunities

• Microbiology: Developing new strains of microorganisms that produce more fat, protein, polysaccharides, or iron-containing compounds.
• Machine learning: Optimizing the Raman signal to increase signal strength or to detect a broader array of molecules of interest for alternative protein applications.
• Microfluidics: Most systems rely on microfluidic devices to manipulate the cells (e.g., cell focusing or cell sorting). Improving the efficiency and integration of microfluidics and optics using Raman spectroscopy can improve processing speed and accuracy. Droplet microfluidics enables prolonged cell incubation steps prior to detection, detection of secreted molecules, and improved sorting throughput. Secreted molecule quantification is highly relevant for precision fermentation improvements.
• Optics: Increasing the optical spectrum from the high-wavelength region to the fingerprint region would allow for more precise detection of metabolites. Another research direction would be improving signal processing from microfluidic devices to increase the signal processing speed.

Anticipated impact

Raman-activated cell sorting can be used to optimize strains of microorganisms with optimal composition and other characteristics (such as protein secretion) for use in alternative protein applications. This can increase the production yields of precision fermentation and optimize composition for biomass fermentation by developing new strains of microorganisms through directed evolution.

Related efforts

• Review article: High-Throughput Raman Flow Cytometry and Beyond¹
• Webinar

GFI resources

The science of fermentation

Learn about the emerging role of microbial fermentation in building the next generation of alternative protein products.