Viability of mono-specie biofilm formed by the solvent producer Clostridium beijerinckii during continuous fermentation in packed bed bioreactor.

Butanol and Isopropanol are naturally produced by the bacteria C. beijerinckii. Those products are used in large field of applications such as fuel and bulk chemicals. Since butanol is toxic at small concentration for cells, bacterial growth and metabolism are inhibited during classical batch fermentation (1). These phenomena lead to the production of low solvent concentration (around 7 g.L^-1) and a low volumetric productivity (0,13 g.L^-1.h^-1) (2). Continuous fermentation can be performed in order to avoid product inhibition by  a continuous removal of fermentation broth. However, the solvent productive biomass is easily washout at high dilution rate because of the low maximum growth rate of the strain in this metabolism phase  (0,05 h^-1) (3). To overcome this issue, cell immobilization of  C. beijerinckii by biofilm formation on solid support is the best solution. As a result, the biomass residence time can be uncorrelated from the hydraulic residence time leading to a higher viable biomass concentration in the bioreactor and consequently a higher volumetric productivity (up to 5 g.L^-1.h-1 ) (4). Our study aimed  at evaluating biofilm viability which is an important parameter that is linked to process productivity and has been little studied in the case of the IBE fermentation (5).

In this study we developed two techniques to monitor biofilm viability during immobilized cell fermentation: Flow cytometry (FC) and PMA qPCR. After FC analysis, a high background noise due to the biofilm extra polymeric substance is obtained. Consequently, an enzymatic  sequential enzymatic biofilm deconstruction using Dnase I and Proteinase K was developed . This pre-treatment successfully lowered the background noise of this analysis. The suspensions obtained were stained with carboxyfluoresceine diacetate (cFDA) and propidium iodide (PI) which are indicators of cellular activity and alteration of membrane integrity, respectively,  and analyzed by flow cytometry. The percentage of viable cells obtained after pre-treatment compared to the control sample is increased from 2.6 ± 0.9 % to 22.8 ± 8.6% because of the background noise decrease. PMA-qPCR confirmed the results obtained by flow cytometry without using enzymatic pre-treatment. Although FC is less accurate than PMA-qPCR, this technique is less time-consuming, cheaper and reliable to study biofilm viability.

References