Biotransformation of ethyl 2-(2′-nitrophenoxy)acetate to benzohydroxamic acid (D-DIBOA) by Escherichia coli

Automatable downstream purification of the benzohydroxamic acid D-DIBOA from a biocatalytic synthesis

Herbicides play a vital role in agriculture, contributing to increased crop productivity by minimizing weed growth, but their low degradability presents a threat to the environment and human health. Allelochemicals, such as DIBOA (2,4-dihydroxy-(2H)-1,4-benzoxazin-3(4 H)-one), are secondary metabolites released by certain plants that affect the survival or growth of other organisms. Although these metabolites have an attractive potential for use as herbicides, their low natural production is a critical hurdle. Previously, the synthesis of the biologically active analog D-DIBOA (4-hydroxy-(2H)-1,4-benzoxazin-3(4H)-one) was achieved, using an engineered E. coli strain as a whole-cell biocatalyst, capable of transforming a precursor compound into D-DIBOA and exporting it into the culture medium, although it cannot be directly applied to crops. Here a chromatographic method to purify D-DIBOA from this cell culture medium without producing organic solvent wastes is described. The purification of D-DIBOA from a filtered culture medium to the pure compound could also be automated. Biological tests with the purified compound on weed models showed that it has virtually the same activity than the chemically synthesized D-DIBOA.

Overexpression of the nitroreductase NfsB in an E. coli strain as a whole-cell biocatalyst for the production of chlorinated analogues of the natural herbicide DIBOA

Benzohydroxamic acids, such as DIBOA (2,4-dihydroxy-2 H)-1,4-benzoxazin-3(4 H)-one), are plant products that exhibit interesting herbicidal, fungicidal and bactericidal properties. A feasible alternative to their purification from natural sources is the synthesis of analogous compounds such as D-DIBOA (2-deoxy-DIBOA) and their chlorinated derivatives. Their chemical synthesis has been simplified into two steps. However, the second step is an exothermic reaction and involves hydrogen release, which makes this methodology expensive and difficult to scale up. The study reported here concerns the possibility of producing chlorobenzoxazinones by a whole-cell biocatalytic process using the ability of the engineered E. coli nfsB-/pBAD-NfsB to catalyse the synthesis of 6-Cl-D-DIBOA and 8-Cl-D-DIBOA from their respective precursors (PCs). The results show that this strain is able to grow in media that contain these compounds and to produce the target molecules with 59.3% and 46.7% biotransformation yields, respectively. Moreover, the strain is capable of processing non-purified PCs from the first chemical step to give similar yields to those obtained from the purified PCs. The kinetics of the reaction in vitro with purified recombinant NfsB nitroreductase were studied to characterise the catalysis further and evaluate the effects that several components of the non-purified PCs have on the process. The results revealed that the kinetics are that of an allosteric enzyme. The inhibitory effect of the substrate of the first step of the chemical synthesis, which is present in some non-purified PCs, was also demonstrated.

A genetically engineered Escherichia coli strain overexpressing the nitroreductase NfsB is capable of producing the herbicide D-DIBOA with 100% molar yield

Background

The use of chemical herbicides has helped to improve agricultural production, although its intensive use has led to environmental damages. Plant allelochemicals are interesting alternatives due to their diversity and degradability in the environment. However, the main drawback of this option is their low natural production, which could be overcome by its chemical synthesis. In the case of the allelochemical DIBOA ((2,4-dihydroxy-2H)-1,4-benzoxazin-3(4H)-one), the synthesis of the analogous compound D-DIBOA (2-deoxy-DIBOA) has been achieved in two steps. However, the scale up of this synthesis is hindered by the second step, which uses an expensive catalyst and is an exothermic reaction, with hydrogen release and a relatively low molar yield (70%). We have previously explored the “Green Chemistry” alternative of using E. coli strains overexpressing the nitroreductase NfsB as a whole-cell-biocatalyst to replace this second step, although the molar yield in this case was lower than that of the chemical synthesis.

Results

In this work, we engineered an E. coli strain capable of carrying out this reaction with 100% molar yield and reaching a D-DIBOA concentration up to 379% respect to the highest biotransformation yield previously reported. This was achieved by a screening of 34 E. coli mutant strains in order to improve D-DIBOA production that led to the construction of the ΔlapAΔfliQ double mutant as an optimum genetic background for overexpression of the NfsB enzyme and D-DIBOA synthesis. Also, the use of a defined medium instead of a complex one, the optimization of the culture conditions and the development of processes with several substrate loads allowed obtaining maxima yields and concentrations.

Conclusions

The high yields and concentrations of D-DIBOA reached by the microbial-cell-factory approach developed in this work will facilitate its application to industrial scale. Also, the use of an optimized defined medium with only an organic molecule (glucose as carbon and energy source) in its composition will also facilitate the downstream processes.

Electronic supplementary material

The online version of this article (10.1186/s12934-019-1135-8) contains supplementary material, which is available to authorized users.

Study of the role played by NfsA, NfsB nitroreductase and NemA flavin reductase from Escherichia coli in the conversion of ethyl 2-(2'-nitrophenoxy)acetate to 4-hydroxy-(2H)-1,4-benzoxazin-3(4H)-one (D-DIBOA), a benzohydroxamic acid with interesting biological properties

Benzohydroxamic acids, such as 4-hydroxy-(2H)-1,4-benzoxazin-3(4H)-one (D-DIBOA), exhibit interesting herbicidal, fungicidal and bactericidal properties. Recently, the chemical synthesis of D-DIBOA has been simplified to only two steps. In a previous paper, we demonstrated that the second step could be replaced by a biotransformation using Escherichia coli to reduce the nitro group of the precursor, ethyl 2-(2'-nitrophenoxy)acetate and obtain D-DIBOA. The NfsA and NfsB nitroreductases and the NemA xenobiotic reductase of E. coli have the capacity to reduce one or two nitro groups from a wide variety of nitroaromatic compounds, which are similar to the precursor. By this reason, we hypothesised that these three enzymes could be involved in this biotransformation. We have analysed the biotransformation yield (BY) of mutant strains in which one, two or three of these genes were knocked out, showing that only in the double nfsA/nfsB and in the triple nfsA/nfsB/nemA mutants, the BY was 0%. These results suggested that NfsA and NfsB are responsible for the biotransformation in the tested conditions. To confirm this, the nfsA and nfsB open reading frames were cloned into the pBAD expression vector and transformed into the nfsA and nfsB single mutants, respectively. In both cases, the biotransformation capacity of the strains was recovered (6.09 ± 0.06% as in the wild-type strain) and incremented considerably when NfsA and NfsB were overexpressed (40.33% ± 9.42% and 59.68% ± 2.0% respectively).