Screening of metagenomic and genomic libraries reveals three classes of bacterial enzymes that overcome the toxicity of acrylate.

Andrew R J Curson,Rolf Daniel,Kathryn Mcinnis,Andrew W B Johnston,Margaret Wexler,Jonathan D Todd,Sonja Voget,Oliver J Burns

doi:10.1371/journal.pone.0097660

Abstract

Acrylate is produced in significant quantities through the microbial cleavage of the highly abundant marine osmoprotectant dimethylsulfoniopropionate, an important process in the marine sulfur cycle. Acrylate can inhibit bacterial growth, likely through its conversion to the highly toxic molecule acrylyl-CoA. Previous work identified an acrylyl-CoA reductase, encoded by the gene acuI, as being important for conferring on bacteria the ability to grow in the presence of acrylate. However, some bacteria lack acuI, and, conversely, many bacteria that may not encounter acrylate in their regular environments do contain this gene. We therefore sought to identify new genes that might confer tolerance to acrylate. To do this, we used functional screening of metagenomic and genomic libraries to identify novel genes that corrected an E. coli mutant that was defective in acuI, and was therefore hyper-sensitive to acrylate. The metagenomic libraries yielded two types of genes that overcame this toxicity. The majority encoded enzymes resembling AcuI, but with significant sequence divergence among each other and previously ratified AcuI enzymes. One other metagenomic gene, arkA, had very close relatives in Bacillus and related bacteria, and is predicted to encode an enoyl-acyl carrier protein reductase, in the same family as FabK, which catalyses the final step in fatty-acid biosynthesis in some pathogenic Firmicute bacteria. A genomic library of Novosphingobium, a metabolically versatile alphaproteobacterium that lacks both acuI and arkA, yielded vutD and vutE, two genes that, together, conferred acrylate resistance. These encode sequential steps in the oxidative catabolism of valine in a pathway in which, significantly, methacrylyl-CoA is a toxic intermediate. These findings expand the range of bacteria for which the acuI gene encodes a functional acrylyl-CoA reductase, and also identify novel enzymes that can similarly function in conferring acrylate resistance, likely, again, through the removal of the toxic product acrylyl-CoA.

Highlights

IntroductionAcrylate is a well-known compound, largely due to its use as a major chemical feedstock for acrylyl polymers in paints and other products of the petrochemical industries
Acrylate is a well-known compound, largely due to its use as a major chemical feedstock for acrylyl polymers in paints and other products of the petrochemical industries. Though, it only occurs in significant amounts in few, specific niches. This is because it is the product of a particular catabolic reaction, namely the cleavage of dimethylsulfoniopropionate (DMSP), a highly abundant (,109 tons produced per annum) osmoprotectant and anti-stress compatible solute that is made by many marine photosynthetic phytoplankton, macroalgal seaweeds and a few angiosperms [1,2,3,4]
The most frequently encountered metagenomic genes that corrected the acrylate sensitivity of the E. coli AcuI2 mutant were themselves related to acuI

Summary

Introduction

Acrylate is a well-known compound, largely due to its use as a major chemical feedstock for acrylyl polymers in paints and other products of the petrochemical industries. The DMSP that is released when plankton die or are grazed can be catabolised by marine bacteria and some fungi, which produce enzymes that are generically termed ‘DMSP lyases’ which yield acrylate plus DMS [1]. When we introduced plasmids containing the metagenomic and B. megaterium arkA genes (in pBIO2167 and pBIO2195 respectively) into E. coli, there was no such effect; the two derivatives grew in 20 mM triclosan, the same maximal level as for E. coli with the empty vector. Taken together, these observations show that most, but not all, strains of the Bacillaceae contain the enzyme ArkA. In light of the reaction that is catalysed by bona fide FabK, it seemed likely that ArkA reduces acrylyl-CoA to propionyl-CoA when exogenous acrylate is present, the same bioconversion as that effected by AcuI (Figure 3)

Methods

Results

Conclusion