Metaproteomics-guided selection of targeted enzymes for bioprospecting of mixed microbial communities

Abstract

Background: Hitherto, the main goal of metaproteomic analyses has been to characterize the functional role of particular microorganisms in the microbial ecology of various microbial communities. Recently, it has been suggested that metaproteomics could be used for bioprospecting microbial communities to query for the most active enzymes to improve the selection process of industrially relevant enzymes. In the present study, to reduce the complexity of metaproteomic samples for targeted bioprospecting of novel enzymes, a microbial community capable of producing cellulases was maintained on a chemically defined medium in an enzyme suppressed metabolic steady state. From this state, it was possible to specifically and distinctively induce the desired cellulolytic activity. The extracellular fraction of the protein complement of the induced sample could thereby be purified and compared to a non-induced sample of the same community by differential gel electrophoresis to discriminate between constitutively expressed proteins and proteins upregulated in response to the inducing substance.

Results: Using the applied approach, downstream analysis by mass spectrometry could be limited to only proteins recognized as upregulated in the cellulase-induced sample. Of 39 selected proteins, the majority were found to be linked to the need to degrade, take up, and metabolize cellulose. In addition, 28 (72%) of the proteins were non-cytosolic and 17 (44%) were annotated as carbohydrate-active enzymes. The results demonstrated both the applicability of the proposed approach for identifying extracellular proteins and guiding the selection of proteins toward those specifically upregulated and targeted by the enzyme inducing substance. Further, because identification of interesting proteins was based on the regulation of enzyme expression in response to a need to hydrolyze and utilize a specific substance, other unexpected enzyme activities were able to be identified.

Conclusions: The described approach created the conditions necessary to be able to select relevant extracellular enzymes that were extracted from the enzyme-induced microbial community. However, for the purpose of bioprospecting for enzymes to clone, produce, and characterize for practical applications, it was concluded that identification against public databases was not sufficient to identify the correct gene or protein sequence for cloning of the identified novel enzymes.

Keywords: Biofuel; Biogas; Bioprospecting; Cellulase; Enzyme discovery; Extracellular; Induction; Metaproteome; Microbial community; Protein inference.

Fig. 1

a Cellulase activity and biogas production in induced and non-induced samples. a Cellulase activity in the induced sample (filled circle) and the non-induced reference sample with the original chemically defined medium (open circle). Dotted arrows indicate samples compared over time in the induced sample, and over space against the non-induced reference sample. b Gas production in the two samples over the same time period, indicating viable microbial communities. The higher gas production in the cellulase-induced sample (filled circle) is simply due to higher organic load

Fig. 2

2-D DIGE gels. a The difference in protein expression pattern between the induced sample at 96 h (Ind_96 h, blue) and induced sample at 24 h (Ind_24 h, green). b The difference in protein expression pattern between the Ind_96 h sample and the non-induced reference sample at the same time point (Ref_96 h, magenta). c Multichannel fluorescent view of the 2-D DIGE gel of all three variants. d All spots in the 2-D DIGE master gel

Fig. 3

Preparative 2-D gel of the Ind_96 h sample for spot picking and tandem mass spectrometry of proteins identified as upregulated. Selected and analyzed spots are encircled and color coded for functional identification