Engineering of an enzyme cocktail for biodegradation of petroleum hydrocarbons based on known enzymatic pathways and metagenomic techniques

Abstract

Hydrocarbon pollution is becoming a growing environmental concern in South Africa and globally. This inadvertently supports the need to identify enzymes for their targeted degradation. The search for novel biocatalysts such as monooxygenases, alcohol dehydrogenases and aldehyde dehydrogenases, have relied on conventional culture-based techniques but this allows sourcing of the biomolecules from only 1-10 % of the microbial population leaving the majority of the biomolecules unaccounted for in 90-99 % of the microbial community. The implementation of a metagenomics approach, a culture-independent technique, ensures that more or less than 100 % of the microbial community is assessed. This increases the chance of finding novel enzymes with superior physico-chemical and catalytic traits. Hydrocarbon polluted soils present a rich environment with an adapted microbial diversity. It was thus extrapolated that it could be a potential source of novel monooxygenases, alcohol dehydrogenases (ADH) and aldehyde dehydrogenases (ALDH) involved in hydrocarbon degradation pathways. Therefore, the aim of the study was to extract metagenomic DNA from hydrocarbon contaminated soils and construct a metagenomic fosmid library and screen the library for monooxygenases, alcohol dehydrogenases (ADH) and aldehyde dehydrogenases (ALDH). Accordingly, the fosmid library was constructed from metagenome of hydrocarbon-contaminated soil. Then the library was functionally screened using hexadecane, octadecene and cyclohexane as substrates and fifteen positive clones were selected. The fosmid constructs of the positive clones were sequenced using PacBio next generation sequencing platform. The sequences were de novo assembled and analysed using CLC Genomic Workbench. The open reading frames (ORF) of the contigs were identified by blasting the contigs against uniport database. Accordingly, four novel genes namely amo-vut1, aol-vut3, dhy-sc-vut5 and dhy-g-vut7 that showed close similarity with our target enzymes were further analysed in silico and codon-optimized as per Escherichia coli codon preference. The codon adjusted sequences were synthesised and cloned into pET30a(+) expression vector. However, it is worth noting that expression of amo-vut1 was not successful since it was later identified to be a multi-pass member protein, which made it insoluble despite the use of detergent to the effect. There is a need to meticulously genetically engineer amo-vut1 to remove the signal and other membrane-bound peptides while maintaining its activity. Yet the other three constructs were successfully transformed and expressed in E. coli BL21 (DE3). The enzymes were purified and characterized and cocktail for hydrolysis of hexanol was succesfully engineered based on AOL-VUT3, DHY-SC-VUT5 and DHY-G-VUT7. Therefore, novel enzymes were mined from metagenome of fossil-oil contaminated soil and effective hydrocarbon-degrading enzyme cocktails containing their combination were successfully engineered.