A versatile microfluidic platform for the cultivation of cyanobacteria at single-cell resolution

The primary endosymbiosis theory states, that about one billion years ago a heterotrophic host organism incorporated a cyanobacterium, which subsequently evolved into the chloroplast we find in algae and plants today. However, the early steps in the establishment of endosymbiosis and the transition towards an organelle have not been experimentally tested yet. Furthermore, cyanobacteria are expected to play a key role a future bio-economy.

Establish microfluidic single-cell cultivation for high-throughput characterization of cyanobacteria. The technology will later be used to trap and observe single cells in picoliters of medium, mimicking the inclusion vacuole during early endosymbiosis.

Cyanobacteria are cultivated in a strict monolayer, therefore self-shading of cells is avoided. The use of the highly gas-permeable polydimethylsiloxane (PDMS) to fabricate cultivation chips allows cultivation without carbon limitation.

A method was developed to generate a linear light-intensity gradient along the microfluidic chip, allowing growth data to be generated at high-throughput. Cells were segmented using a deep learning based algorithm. Python scripts were used for cell segmentation and subsequent data analysis.

Dynamic CO2 control was enabled by stacking an additional PDMS layer on top of the thin film cultivation chip. CO2-depleted air was then perfused through the gas layer in countercurrent to the medium flow.

The cyanobacterial strains Synechococcus elongatus UTEX2973 and PCC7942 as well as Synechocystis sp. PCC6803 were used as model organisms. Using the image analysis pipeline, TB of image data containing millions of cells were analysed. It was shown that UTEX2973 grows without carbon limitation in the microfluidic cultivation system when operating it at ambient air. Using dynamic CO2 control, it was shown that carbon limitation first occurs at a CO2 concentration of 15 ppm. In comparison to UTEX2973, PCC7942 showed higher CO2-dependent growth rates. PCC6803 was the slowest cyanobacterial strain tested. No divergence of growth rates in the direction orthogonal to the light intensity-gradient was observed.

A versatile platform for the characterization of cyanobacteria at single cell resolution was developed and published1. In the near future, the platform will be used to test hypotheses related to endosymbiosis.

1: Witting et al., Lab Chip, p., 2024, doi: 10.1039/D4LC00567H.