Heat-shock treatment applied to inocula for H2 production decreases microbial diversities, interspecific interactions and performance using cellulose as substrate.

The heat-shock pretreatment (HST) is a useful method to select for H 2 -producing inocula when soluble substrates are employed. However, the HST has proven to have negative effects on the H 2 production performance from lignocellulosic substrates. We hypothesize that the negative effect of HST on H 2 production from lignocellulosic substrates is due to the loss of species involved in cellulose solubilization. In the present study, we tested this hypothesis by applying a heat-shock pretreatment (105 °C/24 h) on the microbial community for producing hydrogen from microcrystalline cellulose. Specifically, we compared a microbial community treated with 2-bromoethanesulfonate (BES-treated control) versus a heat-shock pretreated microbial community. For both experimental treatments, we determined the major fermentation products (hydrogen, acetic, butyric, propionic, and isovaleric acids), as well as the diversity of bacteria and fungi using Illumina MiSeq of amplicons in five sampling points. We found that HST immediately reduced alpha diversity of microbial communities, being fungi more affected than bacteria. We also found that the bacterial reduction in Comamonas , Ureibacillus , and Aneurinibacillus was related to a low hydrogen production in the heat-shock pretreated community. Strictly anaerobic fungi such as Orpinomyces , Cyllamyces , and Neocallimastix , which are recognized by their role in solubilization of fibrous materials, were unable to survive the HST. The reconstructed bacterial network predicted positive interactions between cellulase-producing and hydrogen-producing families. We conclude that the HST did not promote the high microbial diversity required for hydrogen production from cellulose. • Two inoculum pretreatments were compared for H 2 production from crystalline cellulose. • Heat-shock treatment caused a lower diversity and productivity than chemical treatment. • The poor hydrogen performance was explained by the loss of diversity. • Positive bacterial interactions were only predicted for the chemical treated community. • The heat-shock treatment was inefficient to select a productive community. [ABSTRACT FROM AUTHOR]