Fast non-invasive monitoring of microalgal physiological stage in photobioreactors through Raman spectroscopy

Microalgal bioprocesses are increasing in multiple industrial sectors for production at large scales. Nevertheless, classical sensors are still used but are not adapted for production monitoring, leading to monoparametric, invasive and time-consuming solutions. Future approaches should eliminate those weaknesses and take advantage of optical methods to optimize the monitoring. This work is focussed on the concrete application of Raman spectroscopy to characterize the physiological kinetics of a microalgal process. The monitoring of Chlamydomonas reinhardtii growth in photobioreactors led us to build its specific Raman spectral database. The complex Raman signatures acquired showed 35 reproducible bands for each day of growth, corresponding to the spectral fingerprints of the cell metabolites, such as chlorophyll a, beta-carotene, nucleic acids and lipids. In total, 2688 spectra were compiled in a database representing the cellular chemical fingerprints in three physiological stages and showed progressive variations between the days of acquisition. New data acquisitions allowed us to build a trial for blind validation and to characterize the bioprocesses with 89.2% accuracy. To complete the study, transcriptomic experiments following the transcript expression of two different metabolic pathways of the microalgae confirmed the cell physiology attributions made by the Raman spectroscopy. This work enables us to query the bioprocess status directly from the cells by attributing a spectrum to the current cell physiology.