Introduction
Dinitrotoluene (DNT) is a nitroaromatic compound extensively used in manufacturing munitions, polyurethane foams, automobile airbags, and other important chemical products (Rickert et al., 1984). Due to a history of long-term manufacturing, civilian and military use, and waste steam leaks, DNT has entered the soil and water, which threatens the health of residents and the environment around DNT manufacturing and processing facilities. DNT exists as six isomers, with 2,4- and 2,6-DNT being the major forms that are moderately to highly toxic to humans, animals, and plants (Gong et al., 2003). The mixture of 2,4- and 2,6-DNT has been classified as a Class B2 carcinogen to humans and animals by the United States Environmental Protection Agency (EPA). Among them, 2,4-DNT is predominantly employed as an intermediate for production of polyurethane (Lent et al., 2012). It is also a byproduct from the synthesis of 2,4,6-trinitrotoluene (TNT), which is one of the most widely used explosives used for civilian and military purposes around the world. In addition, 2,4-dinitrotoluene-3-sulfonate (DNTS) produced from TNT refining is a major toxic component in TNT red water and its contaminated soils (Tsai, 1991). Both 2,4-DNT and its sulfonate have long-term environmental persistence and high-energetic toxicity and mutagenicity to most organisms (Gilbert, 1977). Thus, they have been considered as the priority pollutants that must be removed from soil and groundwater contaminated by manufacturing and processing of polyurethane and TNT.
Biological remediation including phytoremediation and microbial remediation is a safe, economic, and highly efficient strategy for degradation of nitroaromatic compounds. Previous studies have suggested that the detoxification pathway of DNT and other nitroaromatic explosives in plants might go through three phases: functionalization (hydrolysis, oxidation, and reduction), transformation (conjugation by glycosyl transferases or glutathione transferases), and compartmentalization (transportation to cell wall or vacuole) (Rai et al., 2020; Rao et al., 2009). However, the detailed molecular mechanism of detoxification of these nitroaromatic compounds remains elusive in plants. Previous studies have suggested that the high toxicity of nitroaromatic compounds to plants results from the predominant accumulation of excessive reactive oxygen species (ROS) in plant tissues, which damages normal physiological activities in plant cells (Brentner et al., 2010; Johnston et al., 2015). Deficiency ofArabidopsis MONODEHYDROASCORBATE REDUCTASE 6 (MDHAR6) can reduce the production of ROS in the mitochondria and confer Arabidopsishigh tolerance to TNT (Johnston et al., 2015). In addition, overexpression of UDP-glycosyltransferases (UGT743B4 andUGT73C1 ) and glutathione transferases (GSTU24 and GSTU25 ) in Arabidopsis leads to accumulation of the conjugation products of TNT and reduces its cytotoxicity (Gandia-Herrero et al., 2008; Gunning et al., 2014). Although some efforts have been made to alleviate the pollution of TNT by phytoremediation over recent decades, only one bioengineering work has been made for DNT detoxification by expressing a cyanobacterial flavodoxin in tobacco plants (Tognetti et al., 2007).
In contrast, DNT detoxification microbes and their degradation mechanism have been investigated extensively. Many microbes includingEnterobacter cloacae ,Saccharomyces sp , Cadida sp , and white-rot fungus participate in DNT degradation, which provide enriched sources for identification and characterization of genes involved in detoxification of these nitroaromatic compounds (Koder and Miller, 1998; Ziganshin et al., 2007; Kist et al., 2020). Previous studies have suggested that a dioxygenase is responsible for the first step of the DNT degradation pathway. However, this dioxygenase is encoded by four genes (dntAa , Ab, Ac, Ad ) in Burkholderia cepacian R34, which makes it difficult to be applied for plant bioengineering (Johnson et al., 2002). Microbial type I nitroreductase (NR) is an oxygen-insensitive flavoprotein that catalyzes the NAD(P)H-dependent reduction of nitro groups to hydroxylamino and/or amino groups on nitroaromatic compounds. These type I NRs consist of two main groups, NfsA (group A) and NfsB (group B) (Roldán et al., 2008). Moreover, members of NfsB can reduce both 2,4- and 2,6-DNT (Williams et al., 2019), while members of NfsA only convert 2,4-DNT (Rich et al., 2018). NfsI is a member of NfsB NR identified fromEnterobacter cloacae (Zajc, 1999). The recombinant NfsI protein can convert both TNT and DNT in vitro (Bryant et al., 1991). Overexpression of NfsIsuccessfully improves the degradation efficiency of TNT in transgenic poplar, tobacco, and wheatgrass (Brentner et al., 2010; Zhang et al., 2017b; Zhang et al., 2019). These findings shed light on phytoremediation through transgenic plants engineered with bacterial nitroaromatic compound degradation genes. However, it is still unknown if the plants engineered withNfsI could detoxify DNT and its sulfonates.
Switchgrass (Panicum virgatum L.) is a perennial C4 tall grass that has been used as lignocellulosic feedstock for forage and biofuel production. Switchgrass can increase soil organic carbon and facilitate a larger microbe population in margin lands. Moreover, switchgrass is well adapted to various soil types with excellent drought, cold, saline, and heavy metal tolerance (Song et al., 2018). Therefore, switchgrass has great potential for phytoremediation because of its extensive root system, vigorous growth, high ability to stress tolerance, and low-input requirements (Rai et al., 2020). A recent study has shown that ectopic expression of bacterial flavodoxin-cytochrome P450 XplA coupled with flavodoxin reducatase XplB in switchgrass can dramatically improve the removal efficiency of RDX in transgenic plants (Zhang et al., 2017a). Unfortunately, overexpression of the bacterial NfsI in switchgrass cannot degrade TNT due to low transcription of NfsI(Zhang et al., 2017a).
Here, we synthesized a codon optimized bacterial NfsI for overexpression in switchgrass and studied its function in DNT and DNTS detoxification for the first time. The transgenic switchgrass withNfsI overexpression significantly alleviated the 2,4-DNT-induced root growth inhibition and reduced ROS content. The removal efficiency for DNT achieved 94.1% in NfsI overexpressing transgenic switchgrass plants, which was 1.7-fold higher than that of control plants. In contrast, overexpression of NfsI in switchgrass barely improved the removal capacity for DNTS, supporting the in vitro enzyme kinetics analysis that suggests that the recombinant NfsI has approximate 100-fold higher catalytic efficiency for 2,4-DNT than DNTS. Furthermore, overexpression of NfsI in switchgrass partially alleviated the impact of 2,4-DNT on expression profiling of genes involved in plant detoxification. Our work suggests that engineeringNfsI in plants may have great application potential for DNT phytoremediation in the future.