Combining light microscopy, dielectric spectroscopy, MALDI intact cell mass spectrometry, FTIR spectromicroscopy and multivariate data mining for morphological and physiological bioprocess characterization of filamentous organisms.

Along with productivity and physiology, morphological growth behavior is the key parameter in bioprocess design for filamentous fungi. Lacking tools for fast, reliable and efficient analysis however, fungal morphology is still commonly tackled by empirical trial-and-error techniques during strain selection and process development procedures. Bridging the gap, this work presents a comprehensive analytical approach for morphological analysis combining automated high-throughput microscopy, multi-frequency dielectric spectroscopy, MALDI intact cell mass spectrometry and FTIR spectromicroscopy. Industrial fed-batch production processes were investigated in fully instrumented, automated bioreactors using the model system Penicillium chrysogenum. Physiological process characterization was based on the determination of specific conversion rates as scale-independent parameters. Conventional light microscopic morphological analysis was based on holistic determination of time series for more than 30 morphological parameters and their frequency distributions over the respective parameter range by automated high-throughput light microscopy. Characteristic protein patterns enriched in specific morphological and physiological states were further obtained by MALDI intact cell mass spectrometry. Spatial resolution of molecular biomass composition was facilitated by FTIR spectromicroscopy. Real-time in situ monitoring of morphological process behavior was achieved by linking multi-frequency dielectric spectroscopy with above outlined off-line methods. Data integration of complementing orthogonal techniques for morphological and physiological analysis together with multivariate modeling of interdependencies between morphology, physiology and process parameters facilitated complete bioprocess characterization. The suggested approach will thus help understanding morphological and physiological behavior and, in turn, allow to control and optimize those complex processes.