I am from the University of Ulm in Germany and took part in the BRISK2 transnational program to conduct experiments in the labs of SINTEF in Trondheim.

My goal was to perform batch fermentations with recombinant and native propionate-producing anaerobic bacteria to produce propionate. Due to the COVID-19 pandemic it was not possible to travel to Trondheim, however, the team at SINTEF performed all planned experiments on my behalf and according to my recommendations.

There were frequent online discussions about the strains and their properties as well as the achieved results. After completion of the fermentations, withdrawn samples were shipped to my home facility, where I analyzed them using gas chromatography and high-performance liquid chromatography (Figure 1).

During my PhD thesis, I constructed propionate-producing clostridial strains, and identified a saccharolytic, natively propionate-producing Clostridium, which I characterized in serum bottle fermentations.

Now, growth and production behaviour of both recombinant and native producers should be investigated in pH-controlled fermentations, however, the University of Ulm does not have such technical opportunities. SINTEF is in possession of multiple bioreactors, where larger volumes can be processed. Therefore, I sent my strains to SINTEF, where they were cultivated in 0.5 l medium with glucose, xylose or a mixture of both sugars as the main carbon and energy source.

Regarding recombinant propionate production, a total of 3 strains were cultivated (production strain, wild type, and empty vector strain). Growth behaviour of the control strains turned out very similar to what I could observe in my experiments (Figure 2).

Product analyses in Ulm revealed, that no propionate could be produced with the recombinant strain. When checking, if plasmids were still present, it turned out that the strain had already lost the plasmids at the beginning of the fermentation, which explains the lack of propionate production. This can probably be attributed to the missing selective pressure.

In the second part of the study, a native producer should also be characterized, however, the strain never started to grow in the bioreactors. Although the planned experiments did not work out as expected, I still could gain valuable insights with regards to bacterial fermentation, e. g. growth behaviour of strains in larger volumes, effect of medium composition on bacterial growth, and how to develop a suitable scale-up strategy.