Shaping a reactor microbiome generating stable n-caproate productivity through Design-Build-Test-Learn approach

Abstract

n-Caproate production by reactor microbiomes is receiving more attention due to its high economic and environmental benefits. However, the ecological complexity within the bioreactor obstructs the development of functional microbiome shaping technology, which is expected to further unlock the potential of microbiomes. This study adopted the "Design-Build-Test-Learn approach" to develop a fast and reliable microbiome shaping technology. In the first step, 10 mM n-butyrate was a better electron acceptor than 10 mM acetate when 50 mM L-lactate was used as an electron donor, but the n-caproate selectivity was only 33.9%. This was ascribed to the high H-2 and CO2 partial pressure causing a high homoacetogenic activity of the microbiome, hindering n-caproate production. The second step was conducted by continuously removing headspace CO2, and as a result n-caproate selectivity increased to 68.6%. This process was repeated five times to apply selection pressures, and a stable n-caproate specificity (i.e., 57.3-62.6%) was achieved using 50 mM L-lactate and 10 mM n-butyrate. Finally, a n-caproate-producing microbiome with a high productivity (10.0 gCOD/L/day, 0.19 g/L/h) was shaped in a substantially shorter start-up period (84 h) than other studies had achieved. An 16S rRNA sequencing analysis of the shaped microbiome identified the abundance of C. carboxidivorans, which is a chain-elongating bacterial species.