8
Biotransformation and Degradation of the Insensitive Munitions Compound, 3Nitro-1,2,4-triazol-5-one, by Soil Bacterial Communities Mark J. Krzmarzick, ,§ Raju Khatiwada, Christopher I. Olivares, Leif Abrell, Reyes Sierra-Alvarez, Jon Chorover, and James A. Field* ,Department of Chemical and Environmental Engineering, University of Arizona, Tucson, Arizona 85721-0011, United States Department of Soil, Water & Environmental Science, University of Arizona, Tucson, Arizona 85719, United States * S Supporting Information ABSTRACT: Insensitive munitions (IM) are a new class of explosives that are increasingly being adopted by the military. The ability of soil microbial communities to degrade IMs is relatively unknown. In this study, microbial communities from a wide range of soils were tested in microcosms for their ability to degrade the IM, 3-nitro-1,2,4-triazol-5-one (NTO). All seven soil inocula tested were able to readily reduce NTO to 3- amino-1,2,4-triazol-5-one (ATO) via 3-hydroxyamino-1,2,4- triazol-5-one (HTO), under anaerobic conditions with H 2 as an electron donor. Numerous other electron donors were shown to be suitable for NTO-reducing bacteria. The addition of a small amount of yeast extract (10 mg/L) was critical to diminish lag times and increased the biotransformation rate of NTO in nearly all cases indicating yeast extract provided important nutrients for NTO-reducing bacteria. The main biotransformation product, ATO, was degradable only in aerobic conditions, as evidenced by a rise in the inorganic nitrogen species nitrite and nitrate, indicative of nitrogen-mineralization. NTO was nonbiodegradable in aerobic microcosms with all soil inocula. INTRODUCTION Conventional munition compounds, such as 2,4,6-trinitroto- luene (TNT), hexahydro-1,3,5-trinitro-1,3,5-triazacyclohexane (RDX), and octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine (HMX), are often associated with problematic contamination issues and often degrade slowly, if at all, in environmental systems. 1 The toxicity and mutagenicity of these compounds is widely known, 24 and in soils with a long-term exposure to TNT, RDX, or HMX, a signicant loss in bacterial activity and fungal populations has been observed. 5 The heterocyclic compounds RDX and HMX have a greater propensity toward mineralization, 6 while TNT is likely to incorporate into bound residues. 7,8 Recently, insensitive munitions (IM)s are being proposed and tested as alternatives to TNT, RDX, and HMX due to their relative safety with regards to accidental explosions. 911 However, little is known concerning their fate and biodegradation potential in the environment. One of the IM compounds, 3-nitro-1,2,4-triazol-5-one (NTO), has been shown to have lower toxicity and mutagenicity properties and may be less harmful to human health than traditional explosives. 1215 However, NTO may be problematic in the environment due to its high solubility 16,17 and thus increased mobility into groundwater and through water systems. The biodegradation of NTO has been observed in preliminary studies. 16,18 Mammalian cytochrome P450 enzymes biotransformed NTO to both urazole or 3-amino-1,2,4-triazol- 5-one (ATO) under aerobic conditions, while anaerobically NTO is primarily reduced to ATO with only a minor yield of urazole. 18 A Bacillus licheniformis was isolated from NTO production wastewater and was shown to biotransform NTO to ATO with sucrose at pH 6. ATO was then ring cleaved at a higher pH of 8. 16,18 This bacterial process was reported to be oxygen insensitive and was carried out with high quantities of NTO, cell mass, and glucose. Although there is initial evidence of NTO biotransformation and biodegradation, studies are needed to evaluate the biodegradability of NTO in soils where residues of unexploded ordnance may end up as contamination in military ring ranges. We report the biodegradation of NTO and its main metabolite ATO by microbial communities in diverse soils under aerobic and anaerobic conditions. Several dierent electron donors and the presence and absence of a nutritional amount of yeast extract (YE) were also tested. This constitutes the rst study to determine if soil microbial communities can mineralize NTO nitrogen. MATERIAL AND METHODS Materials. NTO was purchased from Interchem (San Pedro, CA). The synthesis of ATO from NTO was adapted Received: January 29, 2015 Revised: April 1, 2015 Accepted: April 3, 2015 Published: April 3, 2015 Article pubs.acs.org/est © 2015 American Chemical Society 5681 DOI: 10.1021/acs.est.5b00511 Environ. Sci. Technol. 2015, 49, 56815688

Biotransformation and Degradation of the Insensitive ... RDX and HMX have a greater propensity toward mineralization,6 while TNT is likely to incorporate into bound residues.7,8 Recently,

Embed Size (px)

Citation preview

Page 1: Biotransformation and Degradation of the Insensitive ... RDX and HMX have a greater propensity toward mineralization,6 while TNT is likely to incorporate into bound residues.7,8 Recently,

Biotransformation and Degradation of the Insensitive MunitionsCompound, 3‑Nitro-1,2,4-triazol-5-one, by Soil BacterialCommunitiesMark J. Krzmarzick,†,§ Raju Khatiwada,‡ Christopher I. Olivares,† Leif Abrell,‡ Reyes Sierra-Alvarez,†

Jon Chorover,‡ and James A. Field*,†

†Department of Chemical and Environmental Engineering, University of Arizona, Tucson, Arizona 85721-0011, United States‡Department of Soil, Water & Environmental Science, University of Arizona, Tucson, Arizona 85719, United States

*S Supporting Information

ABSTRACT: Insensitive munitions (IM) are a new class ofexplosives that are increasingly being adopted by the military.The ability of soil microbial communities to degrade IMs isrelatively unknown. In this study, microbial communities froma wide range of soils were tested in microcosms for their abilityto degrade the IM, 3-nitro-1,2,4-triazol-5-one (NTO). Allseven soil inocula tested were able to readily reduce NTO to 3-amino-1,2,4-triazol-5-one (ATO) via 3-hydroxyamino-1,2,4-triazol-5-one (HTO), under anaerobic conditions with H2 asan electron donor. Numerous other electron donors were shown to be suitable for NTO-reducing bacteria. The addition of asmall amount of yeast extract (10 mg/L) was critical to diminish lag times and increased the biotransformation rate of NTO innearly all cases indicating yeast extract provided important nutrients for NTO-reducing bacteria. The main biotransformationproduct, ATO, was degradable only in aerobic conditions, as evidenced by a rise in the inorganic nitrogen species nitrite andnitrate, indicative of nitrogen-mineralization. NTO was nonbiodegradable in aerobic microcosms with all soil inocula.

■ INTRODUCTION

Conventional munition compounds, such as 2,4,6-trinitroto-luene (TNT), hexahydro-1,3,5-trinitro-1,3,5-triazacyclohexane(RDX), and octahydro-1,3,5,7-tetranitro-1,3,5,7-tetrazocine(HMX), are often associated with problematic contaminationissues and often degrade slowly, if at all, in environmentalsystems.1 The toxicity and mutagenicity of these compounds iswidely known,2−4 and in soils with a long-term exposure toTNT, RDX, or HMX, a significant loss in bacterial activity andfungal populations has been observed.5 The heterocycliccompounds RDX and HMX have a greater propensity towardmineralization,6 while TNT is likely to incorporate into boundresidues.7,8 Recently, insensitive munitions (IM)s are beingproposed and tested as alternatives to TNT, RDX, and HMXdue to their relative safety with regards to accidentalexplosions.9−11 However, little is known concerning their fateand biodegradation potential in the environment. One of theIM compounds, 3-nitro-1,2,4-triazol-5-one (NTO), has beenshown to have lower toxicity and mutagenicity properties andmay be less harmful to human health than traditionalexplosives.12−15 However, NTO may be problematic in theenvironment due to its high solubility16,17 and thus increasedmobility into groundwater and through water systems.The biodegradation of NTO has been observed in

preliminary studies.16,18 Mammalian cytochrome P450 enzymesbiotransformed NTO to both urazole or 3-amino-1,2,4-triazol-5-one (ATO) under aerobic conditions, while anaerobically

NTO is primarily reduced to ATO with only a minor yield ofurazole.18 A Bacillus licheniformis was isolated from NTOproduction wastewater and was shown to biotransform NTO toATO with sucrose at pH 6. ATO was then ring cleaved at ahigher pH of 8.16,18 This bacterial process was reported to beoxygen insensitive and was carried out with high quantities ofNTO, cell mass, and glucose.Although there is initial evidence of NTO biotransformation

and biodegradation, studies are needed to evaluate thebiodegradability of NTO in soils where residues of unexplodedordnance may end up as contamination in military firing ranges.We report the biodegradation of NTO and its main metaboliteATO by microbial communities in diverse soils under aerobicand anaerobic conditions. Several different electron donors andthe presence and absence of a nutritional amount of yeastextract (YE) were also tested. This constitutes the first study todetermine if soil microbial communities can mineralize NTOnitrogen.

■ MATERIAL AND METHODS

Materials. NTO was purchased from Interchem (SanPedro, CA). The synthesis of ATO from NTO was adapted

Received: January 29, 2015Revised: April 1, 2015Accepted: April 3, 2015Published: April 3, 2015

Article

pubs.acs.org/est

© 2015 American Chemical Society 5681 DOI: 10.1021/acs.est.5b00511Environ. Sci. Technol. 2015, 49, 5681−5688

Page 2: Biotransformation and Degradation of the Insensitive ... RDX and HMX have a greater propensity toward mineralization,6 while TNT is likely to incorporate into bound residues.7,8 Recently,

from Le Campion and Ouazzani.19 During the same procedure,3-hydroxyamino-1,2,4-triazol-5-one (HTO) was also synthe-sized. Details of ATO and HTO synthesis are described in theSupporting Information (SI). The end product, ATO, was awhite, or slightly yellow, powder. The structures of ATO andHTO were verified with high resolution quadrupole-time-of-flight mass spectrometry (QToF-MS, see below).All soils were collected within 20 cm from the surface. Roger

Road soil was collected from the University of Arizona CampusAgricultural Center in Tucson, AZ. Catlin agricultural soil wasfrom Illinois and was described previously.20 Maricopa soil wascollected from the University of Arizona Maricopa CountyAgricultural Station in Maricopa County, AZ. Camp Navajo(AZ), Camp Butner (NC), Camp Ripley (MN), and Florence(AZ) soils were collected by CH2M HILL at U.S. NationalGuard bases. All soils, with the exception of Catlin and RogerRoad soils, were immediately placed on ice and shippedovernight to the laboratory. Roger Road (AZ) soil wasimmediately transferred to the laboratory. All soils with theexception of Catlin were sealed to maintain original moisture,while Catlin soil was air-dried. All soils were sieved (2 mm) andstored at 4 °C until used in experiments.Solid Phase Characterization of the Soils. The seven

soils used in this study were characterized for various soilproperties (Table 1) including pH, total organic carbon(TOC), particle size distribution, and Brunauer−Emmett−Teller (BET) specific surface area. The mineralogy was alsocharacterized (Supporting Information (SI), Tables S1 and S2).The pH was measured on a 1:5 (w/w) mixture of soil andwater using a VWR Symphony pH Electrode (VWR Interna-tional, Randor, PA). External specific surface area wasdetermined using the BET dinitrogen gas adsorption method(Beckman Coulter SA-3100). TOC was calculated from thedifference between total carbon, and total inorganic carbon.Total carbon was determined by combustion at 900 °C, andtotal inorganic carbon was determined by phosphoric acidaddition followed by combustion at 200 °C using a Shimadzu5000A-SSM TOC Analyzer (Columbia, MD). For particle sizeanalysis, samples were pretreated for organic matter removaland analyzed with a fully automated Beckman Coulter LS 13320 Laser Diffraction Particle Size Analyzer (Fullerton, CA).Mineralogical analysis was conducted on both the bulk soil(powder) and oriented clays using a PANalytical X’Pert ProMPD X-ray Diffractometer (XRD) with Cu−Kα radiationsource for qualitative and quantitative identification of the soilminerals. Samples were pretreated for the organic matterremoval by oxidation with sodium hypochlorite. For orientedclay slides, the soil was dispersed using sodium hexametaphos-

phate and allowed to settle overnight. The clay suspension wasused for preparing the oriented aggregate for clay mineralidentification. Minerals were identified based on the expansionand contraction of d-spacing of clay minerals with varioustreatments. The treatments include saturation with MgCl2,ethylene glycol, KCl at 25 °C, and subsequent heating at 300and 550 °C. Quantitative phase analysis was performed usingthe Rietveld module in X’Pert High Score Plus softwarefollowing the methodology described previously.21

Degradation Assays. Anaerobic Assays. Anaerobic micro-cosms were incubated at 30 °C in 160 mL serum bottles withbutyl-rubber septa. Assays composed of 100 mL basal mineralmedium with trace elements as described previously22 with theonly exception that resazurin was not included here. Theheadspace of the microcosms (60 mL) was flushed with He/CO2 (80/20%). In microcosms with H2, 1 bar of H2/CO2 (80/20%) over pressure was added. The microcosms wereinoculated with 5 g L−1 soil dry weight (or autoclaved soilfor killed controls). For killed controls, soil was autoclaved for60 min for three consecutive days. Microcosms were autoclavedprior to addition of soil, and NTO or ATO was added viafiltering a 20 mM stock solution through a 0.22 μm syringefilter. Samples were taken with sterile 1 mL sterile syringesthrough the septum. Samples (0.5 mL) were immediatelyspiked with 1 mL of ascorbic acid (300 mg L−1). Samples werestored at −20 °C until analysis and centrifuged (10 min, 13,000× g) prior to analytical analysis.

Aerobic Assays. Aerobic microcosms were incubated at 30°C on a shaker table (180 rpm) in 200 mL flasks topped withcotton. Assays were composed of 100 mL of basal mineralmedium prepared as above with the exception of 1.1 g L−1

K2HPO4 and 1.7 g L−1 KH2PO4 in lieu of NaHCO3. Five g L−1

of dry weight soil (or autoclaved soil for killed controls) wasadded to microcosms, and NTO or ATO was added via a 20mM stock solution filtered through a 0.22 μm syringe filter.NH4Cl and/or YE were excluded in microcosms to furtherstudy ATO degradation (see below). Samples were taken fromsterile pipettes, and flasks with initial medium were weighed tocorrect the concentrations of analytes due to evaporation. Allmicrocosm conditions were run in duplicates. Samples (0.5mL) were immediately spiked with 1 mL of ascorbic acid (300mg L−1). Samples were stored at −20 °C and centrifuged (10min, 13,000 × g) prior to analytical analysis.Initial degradation experiments for NTO were conducted

under anaerobic and aerobic conditions for all seven soils at pH7.2. H2 was added in anaerobic microcosms. These experimentsincluded killed controls for all soils (aerobic and anaerobic),basal-medium only controls (aerobic and anaerobic), and for

Table 1. Selected Properties of the Soils Used in This Study

soil texture

(kg‑1) (%)

soils pH BET SAa (m2 g‑1) TNb TCc TOCd sand silt clay textural class

Camp Ripley (MN) 5.96 ± 0.06 1.72 ± 0.03 1.27 ± 0.20 12.55 ± 1.50 12.5 ± 1.50 78.20 13.99 7.82 loamy sandCamp Butner (NC) 6.36 ± 0.02 4.85 ± 0.07 1.33 ± 0.05 20.69 ± 1.20 20.69 ± 1.20 68.68 19.83 11.50 sandy loamFlorence (AZ) 6.96 ± 0.11 32.45 ± 1.73 0.81 ± 0.02 4.16 ± 0.20 4.16 ± 0.20 44.20 28.50 27.30 clay loamCamp Navajo (AZ) 6.32 ± 0.01 21.5 ± 0.56 3.65 ± 0.21 52.36 ± 3.70 52.36 ± 3.70 21.48 38.10 40.43 clayMaricopa (AZ) 7.75 ± 0.07 34.58 ± 1.70 0.80 ± 0.05 7.07 ± 0.40 4.65 ± 0.40 37.48 21.98 40.55 clayRoger Road (AZ) 7.75 ± 0.01 27.69 ± 0.80 1.54 ± 0.03 18.25 ± 0.10 7.07 ± 0.40 23.33 35.10 41.58 clayCatlin Soil (IL) 6.42 ± 0.06 5.05 ± 0.44 2.81 ± 0.18 45.44 ± 1.10 44.08 ± 1.10 13.50 54.98 31.53 silty clay loam

aBrunauer, Emmett, and Teller (BET) surface area. bTN = total nitrogen. cTC = total carbon. dTOC = total organic carbon.

Environmental Science & Technology Article

DOI: 10.1021/acs.est.5b00511Environ. Sci. Technol. 2015, 49, 5681−5688

5682

Page 3: Biotransformation and Degradation of the Insensitive ... RDX and HMX have a greater propensity toward mineralization,6 while TNT is likely to incorporate into bound residues.7,8 Recently,

selected anaerobic soils (Camp Ripley Camp Butner, CampNavajo, and Florence) endogenous controls (no externalelectron donor).Additional anaerobic experiments were conducted with

various electron donors (20 mM), with and without YE, withCamp Butner soil and included the following: acetate, lactate,ethanol, methanol, glucose, pyruvate, citrate, lactose, butyrate,propionate, and formate. H2-amended (0.8 bar) microcosmsand endogenous controls were also run in parallel.Lastly, we tested the variation of pH and glucose amendment

on NTO and ATO degradation. In these experiments, ATOdegradation was tested using both aerobic and anaerobicmicrocosms, while NTO degradation was tested only withaerobic microcosms. Assays were run with Camp Butner andCamp Navajo soils for each set (ATO, aerobic; ATO,anaerobic; and NTO, aerobic). Because previous researchshows that ATO can be ring cleaved at mildly alkaline pH,16

microcosms were operated at either pH 7.2 or 8.4 and with orwithout glucose (20 mM) for a total of four differenttreatments for each soil. NH4Cl was excluded from the basalmedium to quantify inorganic nitrogen resulting fromdegradation. An additional aliquot of glucose (20 mM) wasadded at Day 14 to microcosms with glucose. The pH wasmeasured and, if needed, adjusted every 7 days. Killed controlswere operated for both soils as well as basal-medium onlycontrols (at pH 7.2, no glucose). For ATO-aerobic assays,microcosms without ATO were run for both soils tocharacterize soil endogenous release of inorganic nitrogen.Analytical Methods. HPLC-DAD. NTO, HTO, and ATO

were analyzed using an Agilent 1200 series (Santa Clara, CA)HPLC-DAD. Samples were diluted (1:3) into 0.1% trifluoro-acetic acid (TFA) buffer prior to analysis. Injections (5 μL)were separated with a Hypercarb column23 (150 mm × 4.6mm, 5 μm pore size) at a temperature of 30 °C. The mobilephase (1 mL min−1) was operated under the following v/vratios of 0.1% TFA aqueous buffer and acetonitrile: 0−3 min100/0; 11 min 85/15; 15 min 50/50; 17 min 50/50; 19 min100/0; 20 min 100/0. NTO was detected at 15 min/340 nm,HTO at 13 min/360 nm, and ATO at 8.9 min/216.5 nm. Stocksolutions of NTO and ATO were prepared from dry powder,and standards were prepared from dilutions of this stocksolution in 0.1% TFA to final concentrations of 1.56, 3.13, 6.25,12.5, 25, and 50 mg L−1. HTO stock solution was preparedfrom a mixed solution of NTO, HTO, and ATO (from thesynthesis above). This stock solution was diluted in 0.1% TFAbuffer, and the concentration of HTO was determined via a

molar balance of the synthesis experiment: the HTO was equalto the amount of the initial NTO in the reaction microcosmminus the amount of NTO and ATO in the final mixture.Standards of HTO were prepared from this stock solution andwere at a final concentration of 0.1, 1.5, 2.0, 4.0, 8.0, and 16.0mg L−1. All standards were linear with respect to peak area.

QToF-MS. QToF-MS analysis was performed with aqueoussolutions by direct infusion on a TripleTOF 5600 QTOF-MS(AB Sciex, Framingham, MA). QToF-MS analysis was used toconfirm the MW of ATO and HTO. ATO ([M + H]+ =101.0447 detected, 1.1 ppm from expected) was verified viadirect infusion of a 1 mg L−1 solution in positive mode. HTO([M + H]+ = 117.0390 observed, 1.7 ppm from expected) wasverified as well. QToF-MS analyses were also used to screen formetabolites in selected microcosms. Microcosm contents werecentrifuged, as above, and diluted (1:10−1:20 final dilution inwater v/v) before infused directly into the QToF-MS. Spectrawere obtained in both positive and negative mode, and a massrange of 35−600 m/z was acquired. Analyst TF 15.1 andFormula Finder 2.02.0 were used to process data.

IC. Ammonium (NH4+), nitrite (NO2

−), and nitrate (NO3−)

were measured with ion chromatography (IC) for assaysinvestigating the aerobic degradation of ATO. The IC analyseswas performed on an ICS-3000 system (Dionex, Sunnyvale,CA) with a split flow for simultaneous anion and cation analysison an AG18 RFIC column (4 × 50 mm, Dionex) and IonPacCG16 RFIC column (3 × 50 mm, Dionex), respectively. Theeluent flow rate for anion analysis was 1 mL min−1 and forcation analysis 0.5 mL min−1. Standards were 3.125, 6.25, 12.5,25, 50, and 100 mg L−1 for NH4

+ and NO2− and 1.5625, 3.125,

6.25, 12.5, 25, and 50 mg L−1 for NO3−.

Data Analysis. Lag time, degradation rate, initial ATOyield, and fraction of NTO converted to ATO were calculatedfor the anaerobic microcosms. Spearman’s correlation testswere performed with Stata 10.1 software.

■ RESULTSSoil Characteristics. Soils collected and used in this study

covered a broad range of characteristics (Table 1 and Tables S1and S2 and Figure S1). Five unique textural classes were foundfor the seven soils (Table 1). The clay content ranged from 8%(Camp Ripley) to 42% (Roger Road), and the sand contentranged from 14% (Catlin) to 78% (Camp Ripley). The amountof secondary minerals also covered a broad range, from a low of0.62% (Camp Ripley) to a high of 44.0% (Catlin). The BETspecific surface area of Camp Ripley soil was lowest (1.72 m2

Figure 1. Anaerobic degradation of NTO to ATO in microcosms inoculated with (A) Camp Navajo (AZ) soil, and (B) Florence (AZ) soil in H2amended microcosms (squares) endogenous controls (triangles) and killed controls (circles). NTO concentrations are shown with solid symbolsand solid lines, and ATO concentrations are shown with open symbols, dotted lines. Error bars indicate standard deviation of duplicate microcosms.

Environmental Science & Technology Article

DOI: 10.1021/acs.est.5b00511Environ. Sci. Technol. 2015, 49, 5681−5688

5683

Page 4: Biotransformation and Degradation of the Insensitive ... RDX and HMX have a greater propensity toward mineralization,6 while TNT is likely to incorporate into bound residues.7,8 Recently,

g−1) and Maricopa was highest (34.58 m2g−1). Maricopa andRoger Road soils had pH in the alkaline range (7.75) whileFlorence soil had near neutral pH, and other soils were on theslightly acidic range (6−6.5). TOC content of the soils rangedfrom 4.16 to 52.36 g kg−1, with Camp Navajo soil having thelowest and Flagstaff soil having the highest TOC concentration.Total nitrogen, like TOC, was highest for the Flagstaff soil(3.65 g kg−1). Maricopa and Camp Navajo had the lowest (0.8g kg−1) values.Anaerobic Reduction of NTO to ATO. In anaerobic

microcosms with H2 added as an electron donor, NTO wasfully biotransformed (Figure 1, Table S3). No biotransforma-tion was observed in killed controls or noninoculated basalmedia. Endogenous controls (no external electron donor)displayed NTO bioconversions rates of 0.03 mM d−1 and 0.006mM d−1 in Camp Navajo soil and in Florence soil microcosms,respectively. When H2 was amended as an electron donor, thelag phase prior to NTO biotransformation was less than 5 daysin all soils except for Roger Road and Maricopa, who had lagphases of ∼12 days. This lag phase is putatively due to thegrowth of nitro-group reducing bacteria. Amendments with H2as electron donor increased biotransformation rates, rangingfrom 0.23 (Roger Road) to 1.25 mM d−1 (Camp Navajo). Noclear connection between biotransformation rates and TOC ortotal nitrogen of the soil was observed. There was a negativecorrelation with soils with alkaline pH that corresponded toslower biotransformation rates (Spearman’s ρ = −0.85, P =0.016).ATO was the dominant product from NTO biotransforma-

tion under anaerobic conditions. ATO formation was

concomitant to the removal of NTO, and the yield of ATOproduction as a fraction of NTO removed was stoichiometric,averaging 95.3 ± 9.4% in the H2 amended microcosms.Consequently, the final molar concentration of ATO in H2-amended microcosms was also nearly equivalent with the initialamount of NTO added to the microcosms (98.1 ± 4.7% amongthe 7 soils) since NTO was totally consumed. Thus, any loss ofNTO to sorption or other reactions was minimal. NTOconversion to ATO was dependent on the presence of anexternal electron donor; the conversion NTO to ATO was verylow in endogenous microcosms although the yield per unit ofNTO removed was similar to that observed in H2-amendedmicrocosms.The ability of Camp Butner soil microbial communities to

use various electron donors to degrade NTO was tested, bothwith and without a nutritional amount of YE (Figures 2 andS2). The electron donors stimulated biotransformation of NTOcompared to endogenous controls (there was only oneexception, methanol without the addition of YE). In almostall cases a nutritional quantity of YE (10 mg L−1) decreased thelag phase for NTO reduction. YE did not improve the rates ofbiotransformation if the electron donor could readily serve ascarbon source, such as the cases with the citrate or pyruvateamended microcosms (Figure S2 in SI). Conversely, YE wasresponsible for a major improvement in the rate for theelectron donors that did not include carbon such as with theH2-amended microcosms or with the microcosms with electrondonors that readily yield H2 upon anaerobic fermentation (e.g.,lactate).

Figure 2. Biotransformation of NTO to ATO in microcosms with H2 as an electron acceptor without YE (A) and with 10 mg L−1 YE (B), or with 20mM of citrate without YE (C) and with 10 mg L−1 YE (D), or with 20 mM pyruvate without YE (E) and with 10 mg L−1 YE (F). The concentrationof NTO (black squares) and ATO (triangles) are shown on the primary axis; the concentration of HTO (circles) is shown on the secondary axis.Error bars indicate standard deviations of duplicate microcosms.

Environmental Science & Technology Article

DOI: 10.1021/acs.est.5b00511Environ. Sci. Technol. 2015, 49, 5681−5688

5684

Page 5: Biotransformation and Degradation of the Insensitive ... RDX and HMX have a greater propensity toward mineralization,6 while TNT is likely to incorporate into bound residues.7,8 Recently,

With many of the electron donors, minor amounts of theintermediate HTO were observed (Figure 2, Table S4). HTOdid not accumulate, probably because it was quickly reduced toATO.Aerobic Degradation Microcosms. NTO was not

degraded under fully aerobic conditions with any of the soils(Table S4). In these microcosms, pH was maintained at 7.2 forat least 62 days (and up to 112 days). Two soils, Camp Butnerand Camp Navajo, were also tested for their ability toaerobically degrade NTO with glucose addition, increasedpH, and under nitrogen-limiting conditions (no exogenousnitrogen source). With these treatments, NTO was notdegraded with either soil.Degradation of ATO. The degradation of ATO was tested

under various conditions with Camp Butner and Camp Navajosoils under aerobic or anaerobic conditions, with and withoutglucose addition, and with neutral pH (7.2) or increased pH(8.5) (Figure 3). ATO did not degrade under any anaerobiccondition with either soil. Under aerobic conditions, ATO wasfound to degrade relatively slowly with both soils as evidenced

by loss of the parent compound and detection of inorganic Nspecies. With the Camp Butner soil, ATO was degraded in allaerobic microcosms tested, and the fastest degradationoccurred at neutral pH without glucose addition. With theCamp Navajo soil, the fastest degradation occurred with pH 8.5without glucose addition. At pH 7.2 and with glucose addition,the duplicate microcosms were dissimilar, with only onereplicate degrading the ATO. Generally, nitrite was found toinitially rise, followed by a permanent rise of nitrate, far abovethe background amount of inorganic nitrogen (0.5 mM). Incontrols without ATO amendment and with killed controls andnoninoculated media-only controls, inorganic nitrogen speciesremained very low and did not increase (Figure S3). Thus, thisincrease in inorganic nitrogen observed in live-soil inoculatedtreatments after the decrease in ATO concentration isindicative of the mineralization of the ATO in thesemicrocosms. Intermediate(s) between ATO and inorganicnitrogen species must exist as is seen from a lack of a molarbalance between the amount of degraded ATO and theproduction of inorganic nitrogen products. Samples from these

Figure 3. Degradation of ATO under aerobic conditions and release of nitrogen species: (A) Camp Butner Soil, pH 7.2; (B) Camp Navajo soil, pH7.2; (C) Camp Butner, pH 8.5; (D) Camp Navajo soil, pH 8.5; (E) Camp Butner soil, pH 7.2, with glucose; (F) Camp Navajo soil, pH 7.2, withglucose; (G) Camp Butner soil, pH 8.5, with glucose; (H) Camp Navajo soil, pH 8.5, with glucose. Solid triangles show ATO concentrations (as permol of N), squares correspond to ammonia concentration, circles correspond to nitrite concentration, diamonds correspond to nitrate concentration,and dashed line represents the sum of nitrogen species (both inorganic and ATO). Mineral medium used in these experiments initially contained 0.5mM of inorganic nitrogen (as ammonia). Error bars represent standard deviations of duplicate microcosms. Autoclaved controls and noninoculatedcontrols are available in the Supporting Information (Figure S3 in the SI).

Environmental Science & Technology Article

DOI: 10.1021/acs.est.5b00511Environ. Sci. Technol. 2015, 49, 5681−5688

5685

Page 6: Biotransformation and Degradation of the Insensitive ... RDX and HMX have a greater propensity toward mineralization,6 while TNT is likely to incorporate into bound residues.7,8 Recently,

microcosms were analyzed with QToF-MS, but no organicintermediates were detected.

■ DISCUSSIONLack of NTO Biodegradation under Aerobic Con-

ditions. Nitro groups are electron withdrawing making directoxidation of the molecule difficult. However, nitroaromaticcompounds with two or less nitro-groups have been degradedby several oxidative pathways.24 In other cases, aerobic bacteriawill reduce nitro-aromatics to aromatics with hydroxylamino-groups24,25 or to amino-groups,26,27 which are then furthermetabolized. Less is known concerning the aerobic biode-gradation of nitro-heterocyclic compounds that may alsorequire a reduction of the nitro group but may involve differentmechanisms. The reduction of NTO to ATO in this study wasonly achieved under anaerobic conditions, suggesting the needof an anaerobic step for bioremediation. On the other hand, LeCampion et al.16 and Richard and Weidhaas28 observedreduction of NTO in the presence of oxygen using highamounts of cells and rich organic broths as medium (e.g.,glucose). High levels of organic substrates in those experimentsmay have unknowingly produced O2-deficient conditions as theorganic substrates were consumed.Nitro-group containing organic compounds are likely difficult

to biodegrade aerobically in soils. Similar to our findings, thedegradation of the heterocyclic compound RDX was notdegraded by soil microbial communities under aerobicconditions.29 TNT is also not readily degraded aerobicallyexcept when coamended with other substrates which causenitro-group reduction to occur.30 In our study, however, theaddition of glucose as a cosubstrate did not aid in thebiodegradation of NTO under aerobic conditions.Anaerobic Biotransformation. The anaerobic reduction

of NTO to ATO occurred in all seven soils tested indicatingthat this pathway should be ubiquitous in soil environments ifsubjected to anaerobic conditions. The facile anaerobicreduction of nitro-groups is common.31−33 The completereduction of the nitro- to the amino-group in NTO occurredreadily with nearly stoichiometric amounts of ATO produced innearly all cases. Stoichiometric reduction of NTO to ATO wasobserved in previous literature with Bacillus licheniformis andmammalian liver microsomes18 and has been observed inenvironmental soils for RDX.29 The results here indicate thisreduction process is very ubiquitous.Nitroreductases. Specific nitroreductase enzymes have

been implicated in some aerobic nitroaromatic degradingbacteria.24,32 Less specific oxygen-insensitive nitroreductases ofenteric bacteria have been shown to reduce TNT andRDX.32,34,35 Hydrogenases and carbon monoxide dehydrogen-ase pyruvate:ferrodoxin oxidoreductase of Clostridia have beenshown to reduce TNT.32,33 Bulk reducing agents generated inanaerobic environments such as sulfide together with redoxmediating natural organic matter and ferrous iron adsorbed toiron oxide minerals are also known to reduce nitro-groupsabiotically.36,37 Though such a mechanism is theoreticallypossible in our experiments, the relatively low concentrations ofsulfate (0.4 mM) amended in the mineral medium would needto be cycled many fold in order to provide the reducingelectrons needed to transform the amount of NTO added (3.8mM). The IM compound, 2,4-dintroanisole (DNAN), wastransiently converted to both nitroso- and hydroxylamine-intermediates as it was being reduced to 2-amino-4-nitroanisoleby an aerobic soil bacteria, Bacillus sp. strain 13G.38

Physiology. An external electron donor was required forrapid and complete biotransformation. In previous studies, thereduction of nitro-organic explosive compounds also neededelectron donors, and, similar to this study, diverse sources ofelectron donors were found suitable.33,39−41 H2 or substratesthat generate H2 during their conversion were the best electrondonors in agreement with the findings in this work on NTO.YE provided a critical nutritional requirement to the soilmicroflora that enabled them to reduce NTO with a decreasedlag phase with nearly every electron donor tested and anincreased degradation rate with many of the electron donors.YE provided the greatest improvement of NTO degradationrates when H2 was used as an electron donor, indicating thatYE may have also served as a heterotrophic carbon source. Thishas been observed previously for chemolithotrophic microbialreactions where either pyruvate or YE was needed to stimulatethe reduction of perchlorate with elemental sulfur.42 YE hasalso been shown to increase the nitro-group reduction of TNTby a Pseudomonas strain.43

Degradation Products. The reduction of nitro-groupsoccurs via three steps of two electron reductions of nitro- tonitroso-, nitroso- to hydroxylamino-, and hydroxylamino- toamino-groups.32 In this study, NTO was reduced to ATO withsmall amounts of the hydroxylamino intermediate (HTO)being observed temporally in some assays (see Figure 4). The

HTO as an intermediate of NTO biotransformation has notbeen reported previously. TNT is often observed to degrade tomixed amounts of hydroxylamino- and amino-derivatives bybacteria,44−46 and RDX is most commonly associated with areduction to nitroso-intermediates.38,47,48 Hydroxylamino-intermediates are known to be reactive.3,49 In studies withDNAN, the mixture of hydroxylamino- and nitroso-intermedi-ates may lead to dimerization products when degraded bybacteria.22,38 A similar high level of reactivity was not observedin this study, since there was not a major loss in thestoichiometric yield of ATO in biodegradation experimentseven under conditions with the highest HTO intermediateconcentration or conditions with the most prolonged exposureto HTO. Additionally, coupling products were sought usingMS-QToF, yet these products were not found.ATO was never degraded under any of the anaerobic

conditions or soils used in this study, but degradation wasobserved under aerobic conditions. In the aerobic microcosms,the nitrogen balance was incomplete after the degradation ofATO and prior to the production of inorganic nitrogenproducts indicating that intermediates not measured nordetected in this study were produced. Heterocyclic explosivessuch as RDX and HMX become unstable after initial reductionof the nitro-groups due to the weak energy of carbon−nitrogenbond leading to abiotic hydrolysis.6,50,51 In contrast, ATO wasstable in anaerobic conditions and degradation in aerobic

Figure 4. Degradation pathway of NTO found in this study.

Environmental Science & Technology Article

DOI: 10.1021/acs.est.5b00511Environ. Sci. Technol. 2015, 49, 5681−5688

5686

Page 7: Biotransformation and Degradation of the Insensitive ... RDX and HMX have a greater propensity toward mineralization,6 while TNT is likely to incorporate into bound residues.7,8 Recently,

conditions occurred slowly. Previous research shows that ATOcan be ring cleaved at mildly alkaline pH.16 A higher pH didaccompany a faster degradation rate in the Camp Navajo soilwithout glucose addition, but otherwise an increase in pH wasnot accompanied by significantly better degradation of ATO.Le Campion et al.16,18 found ring cleavage products such asCO2 and urea. Likewise they putatively identified hydroxyurea.Urea and hydroxyurea could potentially represent N-com-pounds that were missing in the N-balance of this study.Additionally, denitrification may have occurred if there wereanaerobic niches in the aerobic assays, causing nitrogen to bereleased as nitrogen gas.A recent study28 reported aerobic conversion of NTO using

an enrichment culture exposed to the munitions formula IMX-101 as an N source and organic cosubstrates. They indicated anNTO degradation product lacking the nitro group; however,the LC-MS evidence provided did not support the proposedstructure.Environmental Implications. In soils, the degradation of

NTO may be stimulated by promoting an initial anaerobicphase to form ATO followed by an aerobic phase for ATObiodegradation. In aerobic conditions, our research suggeststhat NTO will be nonbiodegradable in soils. In anaerobicenvironments, it will most likely be readily converted to ATO,but ATO will then persist in anaerobic conditions. In aerobicconditions, ATO may be susceptible to mineralization depend-ing on the soil microbial community and pH conditions. Theaddition of cosubstrates (e.g., glucose) was not observed in ourstudy to necessarily enhance ATO degradation.

■ ASSOCIATED CONTENT*S Supporting InformationATO and HTO synthesis; soil characteristics; mineralogicalcontent of the oriented clays; concentrations of theintermediate 3-hydroxyamino-1,2,4-triazol-5-one; lag times,degradation rates, and ATO yields for anaerobic soil micro-cosms; and, summary of experiments where degradation ofATO or NTO was not observed. This material is available freeof charge via the Internet at http://pubs.acs.org.

■ AUTHOR INFORMATIONCorresponding Author*Phone: 1-520-621-2591. Fax: 1-520-621-6048. E-mail:[email protected].

Present Address§School of Civil and Environmental Engineering, OklahomaState University, 207 Engineering South, Stillwater, OK 74078.

NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSThis study was supported by the Strategic EnvironmentalResearch and Development Program (SERDP) project ER-2221. We thank Katerina Donstova for the Catlin soil andStefan Walston for the Maricopa soil. CIO was funded in partby the Mexican National Council for Science and Technology(CONACyT).

■ REFERENCES(1) Pennington, J. C.; Brannon, J. M. Environmental fate ofexplosives. Thermochim. Acta 2002, 384 (1), 163−172.

(2) Kaplan, D. L.; Kaplan, A. M. Mutagenicity of 2,4,6-trinitrotoluene-surfactant complexes. Bull. Environ. Contam. Toxicol.1982, 28 (1), 33−38.(3) Maeda, T.; Nakamura, R.; Kadokami, K.; Ogawa, H. I.Relationship between mutagenicity and reactivity or biodegradabilityfor nitroaromatic compounds. Environ. Toxicol. Chem. 2007, 26 (2),237−241.(4) Won, W. D.; Disalvo, L. H.; Ng, J. Toxicity and mutagenicity of2,4,6-trinitrotoluene and its microbial metabolites. Appl. Environ.Microbiol. 1976, 31 (4), 576−580.(5) Meyers, S.; Deng, S.; Basta, N.; Clarkson, W.; Wilber, G. Long-term explosive contamination in soil: Effects on soil microbialcommunity and bioremediation. Soil Sediment Contam. 2007, 16 (1),61−77.(6) Hawari, J. Biodegradation of RDX and HMX: From basicresearch to field application. In Biodegradation of nitroaromaticcompounds and explosives; Spain, J. C., Hughes, J. B., Knackmuss, H.J., Eds.; Lewis Publishers: Boca Raton, FL, 2000; pp 277−310.(7) Bruns-Nagel, D.; Knicker, H.; Drzyzga, O.; Butehorn, U.;Steinbach, K.; Gemsa, D.; von Low, E. Characterization of 15N-TNTresidues after an anaerobic/aerobic treatment of soil/molassesmixtures by solid-state15N NMR spectroscopy. 2. Systematicinvestigation of whole soil and different humic fractions. Environ. Sci.Technol. 2000, 34 (8), 1549−1556.(8) Bruns-Nagel, D.; Steinbach, K.; Gemsa, D.; Von Low, E.Composting (humification) of nitroaromatic compounds. In Biode-gradation of nitroaromatic comopunds and explosives; Spain, J. C.,Hughes, J. B., Knackmuss, H. J., Eds.; Lewis Publishers: Boca Raton,FL, 2000; pp 357−393.(9) Agrawal, J. P. Some new high energy materials and theirformulations for specialized applications. Propellants, Explos., Pyrotech.2005, 30 (5), 316−328.(10) Isler, J. The transition to insensitive munitions (IM). Propellants,Explos., Pyrotech. 1998, 23 (6), 283−291.(11) Sikder, A. K.; Sikder, N. A review of advanced high performance,insensitive and thermally stable energetic materials emerging formilitary and space applications. J. Hazard. Mater. 2004, 112 (1−2), 1−15.(12) Haley, M. V.; Kuperman, R. G.; Checkai, R. T. Aquatic Toxicityof 3-Nitro-1,2,4-Triazol-5-One; Edgewood Chemical Biological Center,U.S. Army Research, Development and Engineering Command,Aberdeen Proving Ground: Ft. Belvoir, 2009.(13) London, J. E.; Smith, D. M. Toxicological study of NTO; LosAlamos National Laboratory: 1985.(14) Reddy, G.; Song, J.; Kirby, P.; Lent, E. M.; Crouse, L. C. B.;Johnson, M. S. Genotoxicity assessment of an energetic propellantcompound, 3-nitro-1,2,4-triazol-5-one (NTO). Mutat. Res. 2011, 719(1−2), 35−40.(15) Sarlauskas, A.; Nemeikaite -Ce niene, A.; Anusevicius, Z.;Miseviciene, L.; Maroziene, A.; Markevicius, A.; Cenas, N. Enzymaticredox properties of novel nitrotriazole explosives implications for theirtoxicity. Z. Naturforsch., C: J. Biosci. 2004, 59 (5−6), 399−404.(16) Le Campion, L.; Vandais, A.; Ouazzani, J. Microbial remediationof NTO in aqueous industrial wastes. FEMS Microbiol. Lett. 1999, 176(1), 197−203.(17) Smith, M. W.; Cliff, M. D. NTO-based explosive formulations: Atechnology review; Aeronautical and Maritime Research Laboratory:Australia, 1999.(18) Le Campion, L.; Delaforge, M.; Noel, J. P.; Ouazzani, J.Metabolism of 14C-labelled 5-nitro-1,2,4-triazol-3-one (NTO):comparison between rat liver microsomes and bacterial metabolicpathways. J. Mol. Catal. B: Enzym. 1998, 5 (1), 395−402.(19) Le Campion, L.; Ouazzani, J. Synthesis of 5-amino-1,2,4-triazole-3-one through the nitroreduction of 5-nitro-1,2,4-triazole-3-one. Comparison between chemical and microbiological catalysis.Biocatal. Biotransform. 1999, 17 (1), 37−44.(20) Dontsova, K. M.; Hayes, C.; Pennington, J. C.; Porter, B.Sorption of high explosives to water-dispersible clay: influence of

Environmental Science & Technology Article

DOI: 10.1021/acs.est.5b00511Environ. Sci. Technol. 2015, 49, 5681−5688

5687

Page 8: Biotransformation and Degradation of the Insensitive ... RDX and HMX have a greater propensity toward mineralization,6 while TNT is likely to incorporate into bound residues.7,8 Recently,

organic carbon, aluminosilicate clay, and extractable iron. J. Environ.Qual. 2009, 38 (4), 1458−1465.(21) Perdrial, N.; Rivera, N.; Thompson, A.; O’Day, P. A.; Chorover,J. Trace contaminant concentration affects mineral transformation andpollutant fate in hydroxide-weathered Hanford sediments. J. Hazard.Mater. 2011, 197 (Journal Article), 119−127.(22) Olivares, C.; Liang, J.; Abrell, L.; Sierra-Alvarez, R.; Field, J. A.Pathways of reductive 2,4-dinitroanisole (DNAN) biotransformationin sludge. Biotechnol. Bioeng. 2013, 110 (6), 1595−1604.(23) Le Campion, L.; Adeline, M. T.; Ouazzani, J. Separation ofNTO related 1,2,4-triazole-3-one derivatives by a high performanceliquid chromatography and capillary electrophoresis. Propellants,Explos., Pyrotech. 1997, 22 (4), 233−237.(24) Ju, K. S.; Parales, R. E. Nitroaromatic compounds, fromsynthesis to biodegradation. Microbiol. Mol. Biol. Rev. 2010, 74 (2),250−272.(25) Marvin-Sikkema, F. D.; de Bont, J. A. M. Degradation ofnitroaromatic compounds by microorganisms. Appl. Microbiol.Biotechnol. 1994, 42 (4), 499−507.(26) Boopathy, R.; Melancon, E. Metabolism of compounds withnitro-functions by Klebsiella pnuemoniae isolated from a regionalwetland. Int. Biodeterior. Biodegrad. 2004, 54 (4), 269−275.(27) Wang, J.; Zhou, J.-T.; Zhang, A.-L.; Hong, L. U. Aerobicdegradation of nitrobenzene by Pseudomonas sp. JX165 and its intactcells. China Environ. Sci. 2001, 21, 144−147.(28) Richard, T.; Weidhaas, J. Biodegradation of IMX-101 explosiveformulation constituents: 2,4-dinitroanisole (DNAN), 3-nitro-1,2,4-triazol-5-one (NTO), and nitroguanidine. J. Hazard. Mater. 2014, 280,372−379.(29) Jayamani, I.; Manzella, M. P.; Cupples, A. M. RDX degradationpotential in soils previously unexposed to RDX and the identificationof RDX-degrading species in one agricultural soil using stable isotopeprobing. Water, Air, Soil Pollut. 2013, 224 (10), 1−15.(30) Lewis, T. A.; Ederer, M. M.; Crawford, R. L.; Crawford, D. L.Microbial transformation of 2,4,6-trinitrotoluene. J. Ind. Microbiol.Biotechnol. 1997, 18 (2−3), 89−96.(31) Razo-Flores, E.; Lettinga, G.; Field, J. A. Biotransformation andbiodegradation of selected nitroaromatics under anaerobic conditions.Biotechnol. Prog. 1999, 15 (3), 358−365.(32) Roldan, M.; Perez-Reinado, E.; Castillo, F.; Moreno-Vivian, C.Reduction of polynitroaromatic compounds: the bacterial nitro-reductases. FEMS Microbiol. Rev. 2008, 32 (3), 474−500.(33) Gorontzy, T.; Kuver, J.; Blotevogel, K. H. Microbial trans-formation of nitroaromatic compounds under anaerobic conditions. J.Gen. Microbiol. 1993, 139, 1331−1336.(34) Kitts, C. L.; Green, C. E.; Otley, R. A.; Alvarez, M. A.; Unkefer,P. J. Type I nitroreductases in soil enterobacteria reduce TNT (2,4,6,-trinitrotoluene) and RDX (hexahydro-1,3,5-trinitro-1,3,5-triazine).Can. J. Microbiol. 2000, 46 (3), 278−282.(35) Koder, R. L.; Haynes, C. A.; Rodgers, M. E.; Rodgers, D. W.;Miller, A. F. Flavin thermodynamics explain the oxygen insensitivity ofenteric nitroreductases. Biochemistry 2002, 41 (48), 14197−14205.(36) Dunnivant, F. M.; Schwarzenbach, R. P.; Macalady, D. L.Reduction of substituted nitrobenzenes in aqueous solutionscontaining natural organic matter. Environ. Sci. Technol. 1992, 26(11), 2133−2141.(37) Elsner, M.; Schwarzenbach, R. P.; Haderlein, S. B. Reactivity ofFe(II)-bearing minerals toward reductive transformation of organiccontaminants. Environ. Sci. Technol. 2004, 38 (3), 799−807.(38) Perreault, N. N.; Manno, D.; Halasz, A.; Thiboutot, S.;Ampleman, G.; Hawari, J. Aerobic biotransformation of 2, 4-dinitroanisole in soil and soil Bacillus sp. Biodegradation 2012, 23(2), 287−295.(39) Adrian, N. R.; Arnett, C. M.; Hickey, R. F. Stimulating theanaerobic biodegradation of explosives by the addition of hydrogen orelectron donors that produce hydrogen. Water Res. 2003, 37 (14),3499−3507.(40) Boopathy, R.; Kulpa, C. F.; Manning, J. Anaerobicbiodegradation of explosives and related compounds by sulfate-

reducing and methanogenic bacteria: A review. Bioresour. Technol.1998, 63 (1), 81−89.(41) Donlon, B. A.; Razo-Flores, E.; Lettinga, G.; Field, J. A.Continuous detoxification, transformation, and degradation of nitro-phenols in upflow anaerobic sludge blanket (UASB) reactors.Biotechnol. Bioeng. 1996, 51 (4), 439−449.(42) Ju, X.; Field, J. A.; Sierra-Alvarez, R.; Salazar, M.; Bentley, H.;Bentley, R. Chemolithotrophic perchlorate reduction linked to theoxidation of elemental sulfur. Biotechnol. Bioeng. 2007, 96 (6), 1073−1082.(43) Oh, B.-T.; Shea, P. J.; Drijber, R. A.; Vasilyeva, G. K.; Sarath, G.TNT biotransformation and detoxification by a Pseudomonasaeruginosa strain. Biodegradation 2003, 14 (5), 309−319.(44) Fleischmann, T. J.; Walker, K. C.; Spain, J. C.; Hughes, J. B.;Craig, A. M. Anaerobic transformation of 2,4,6-TNT by bovineruminal microbes. Biochem. Biophys. Res. Commun. 2004, 314 (4),957−963.(45) Preuss, A.; Fimpel, J.; Diekert, G. Anaerobic transformation of2,4,6-trinitrotoluene (TNT). Arch. Microbiol. 1993, 159 (4), 345−353.(46) Yin, H.; Wood, T. K.; Smets, B. F. Reductive transformation ofTNT by Escherichia coli: pathway description. Appl. Microbiol.Biotechnol. 2005, 67 (3), 397−404.(47) Khan, M. I.; Lee, J.; Park, J. Microbial degradation and toxicityof hexahydro-1,3,5-trinitro-1,3,5-triazine. J. Microbiol. Biotechnol. 2012,22 (10), 1311−1323.(48) McCormick, N. G.; Cornell, J. H.; Kaplan, A. M. Biodegradationof hexahydro-1,3,5-trinitro-1,3,5-triazine. Appl. Environ. Microbiol.1981, 42 (5), 817−823.(49) Hughes, J. B.; Wang, C.; Yesland, K.; Richardson, A.; Bhadra, R.;Bennett, G.; Rudolph, F. Bamberger rearrangement during TNTmetabolism by Clostridium acetobutylicum. Environ. Sci. Technol. 1998,32 (4), 494−500.(50) Bhushan, B.; Trott, S.; Spain, J. C.; Halasz, A.; Paquet, L.;Hawari, J. Biotransformation of hexahydro-1,3,5-trinitro-1,3,5-triazine(RDX) by a rabbit liver cytochrome P450: insight into the mechanismof RDX biodegradation by Rhodococcus sp. strain DN22. Appl. Environ.Microbiol. 2003, 69 (3), 1347−1351.(51) Halasz, A.; Hawari, J. Degradation routes of RDX in variousredox systems. In Aquatic Redox Chemistry; Tratnyek, P. G., Grundl, T.J., Haderlein, S. B., Eds.; 2011; Vol. 1071, pp 441−462.

Environmental Science & Technology Article

DOI: 10.1021/acs.est.5b00511Environ. Sci. Technol. 2015, 49, 5681−5688

5688