Plant Assimilation Kinetics and Metabolism of 2 ... - ACS Publications

Article pubs.acs.org/est

Plant Assimilation Kinetics and Metabolism of 2‑Mercaptobenzothiazole Tire Rubber Vulcanizers by Arabidopsis Gregory H. LeFevre,†,‡,∥ Andrea C. Portmann,†,‡,# Claudia E. Müller,†,‡ Elizabeth S. Sattely,†,§ and Richard G. Luthy*,†,‡ †

ReNUWIt Engineering Research Center, ‡Department of Civil & Environmental Engineering, and §Department of Chemical Engineering, Stanford University, Stanford, California 94305, United States # Institute of Environmental Engineering, ETH Zürich, Zürich, Zürich 8093, Switzerland S Supporting Information *

ABSTRACT: 2-Mercaptobenzothiazole (MBT) is a tire rubber vulcanizer found in potential sources of reclaimed water where it may come in contact with vegetation. In this work, we quantified the plant assimilation kinetics of MBT using Arabidopsis under hydroponic conditions. MBT depletion kinetics in the hydroponic medium with plants were second order (t1/2 = 0.52 to 2.4 h) and significantly greater than any abiotic losses (>18 times faster; p = 0.0056). MBT depletion rate was related to the initial exposure concentration with higher rates at greater concentrations from 1.6 μg/L to 147 μg/ L until a potentially inhibitory level (1973 μg/L) lowered the assimilation rate. 9.8% of the initial MBT mass spike was present in the plants after 3 h and decreased through time. Insource LC-MS/MS fragmentation revealed that MBT was converted by Arabidopsis seedlings to multiple conjugated-MBT metabolites of differential polarity that accumulate in both the plant tissue and hydroponic medium; metabolite representation evolved temporally. Multiple novel MBT-derived plant metabolites were detected via LC-QTOF-MS analysis; proposed transformation products include glucose and amino acid conjugated MBT metabolites. Elucidating plant transformation products of trace organic contaminants has broad implications for water reuse because plant assimilation could be employed advantageously in engineered natural treatment systems, and plant metabolites in food crops could present an unintended exposure route to consumers.

■

INTRODUCTION Benzothiazoles are a class of trace organic contaminants present in potential reclaimed water sources. Reclaimed wastewater is being employed for landscape and crop irrigation,1 and captured stormwater is being used to recharge aquifers;2−4 therefore, understanding contaminant fate is essential for safe and efficient use of these resources. Benzothiazoles are found in stormwater and road dust from tire abrasion residuals on roads2,5−9 (measured up to 74 μg/L) as well as wastewater effluent from both municipal6 (1.7−2.2 μg/L) and industrial sources10 (>1 mg/L). 2-Mercaptobenzothiazole (MBT) is a high-volume production benzothiazole that is present in many environmental media10,11 due to its use as a vulcanization accelerator in rubber manufacture, including vehicular tires.10 Some rubber varieties are composed of up to 2% MBT; leaching from tires may be an important component of overall MBT discharge to the environment because it is not tightly bound in the rubber matrix.10 MBT is also used as a corrosion inhibitor in antifreeze coolant,12 industrial fungicides,6 and as a topical antifungal drug.13 MBT is a moderately polar organic compound (log KOW = 2.41) and is fairly water-soluble11 (120 mg/L at 24 °C). MBT is almost exclusively derived from © XXXX American Chemical Society

anthropogenic sources and occurs only rarely as a natural product.14 MBT and some MBT degradation products are toxic to fish, bacteria, and mammals, including humans where it has been identified as a contact allergen.6,15−18 MBT is listed as a nonmutagen but has shown carcinogenic response in rats at several target sites.17,19 The LC50 for Daphnia magna20 is 0.67 mg/L, and the EC50 for Vibrio f ischeri12 is 0.12 mg/L. MBT can have adverse effects on microorganisms including viruses, yeasts, and fungi.14,21 This may impact biological processes at wastewater treatment plants because MBT can inhibit ammonia oxidation during biological nitrification processes.14 Reemtsma et al.12 observed a 40−50% respiration inhibition of mixed cultures at 0.1 mg/L MBT after 10 days. MBT has been Special Issue: Jerry Schnoor’s Lasting Influence on Global and Regional Environmental Research Received: September 25, 2015 Revised: December 21, 2015 Accepted: December 23, 2015

A

DOI: 10.1021/acs.est.5b04716 Environ. Sci. Technol. XXXX, XXX, XXX−XXX

Article

Environmental Science & Technology

point. On each of the days following the pollutant spike, harvesting occurred, and the remaining plants continued to grow. Experiments were conducted under sterile conditions to exclude bacterial contribution to MBT degradation. Experimental treatments contained plants in hydroponic medium that were exposed to light and a known mass of MBT as described below. Negative controls tested the impact of photolysis on the degradation of MBT in the medium and included hydroponic medium, exposure to the same lighting conditions as for plant growth, and addition of a known mass of MBT but no plants. “Dark controls” tested for other abiotic processes, such as oxidation or hydrolysis, on the degradation of MBT and excluded plants and light. The positive control (plant with no MBT) determined if the treatment had a stimulating, inhibitory, or no effect on the plant growth and ensured seeds were viable. MBT was spiked at four different concentrations between 1 μg/L and 2000 μg/L, an environmentally relevant range of concentrations for stormwater and some industrial wastewaters.5,12,24 Plant Growth Conditions. The entire seed planting procedure was conducted under sterile conditions in a laminar flow biological safety hood washed with 70% ethanol and irradiated with UV light for 30 min prior to the initiation of work. All plant boxes, glass containers, and measuring devices were autoclaved at 121 °C. Wild-type Arabidopsis thaliana seeds were sterilized using a bleach solution (SI) and allowed to stratify overnight at 4 °C. Seeds were added to autoclaved Magenta boxes (Magenta Corp, n = 30 seeds per box) with 25 mL of filter-sterilized (0.22 μm PES, Corning) Murashige and Skoog (MS) basal medium (SI). Any clusters of seeds were dispersed by aspirating with a sterile pipet. The box lids were then partially closed and sealed by microporous tape (Micropore, 3M). The “dark control” boxes were wrapped in aluminum foil. All plant boxes were kept in a growth chamber (Percival) for the duration of the experiment with the temperature held at 22−23 °C and the humidity retained between 50 and 60%. The day-night cycle consisted of 16 h of fluorescent growth light from 8:00 a.m.−11:59 p.m. followed by 8 h of darkness from 12:00 a.m.−8:00 a.m. Analytical Methods. Hydroponic Medium Extraction. The procedure for solid phase extraction (SPE) followed Janna et al. (2011) with some modifications and has been described elsewhere.39 Briefly, 10 μL of the internal standard 2mercaptobenzothiazole-d4 (10 mg/L, Medical Isotopes Inc.) was added to each sample prior to processing. The cartridges (Oasis HLB 60 mg, Waters) were preconditioned first with 5 mL of methanol (ACS grade, Fisher Scientific) and then with 5 mL of Milli-Q water. The total sample volume prior to SPE was determined for mass balance purposes. Samples were loaded and allowed to percolate through the sorbent by gravity. After all samples had been added, the cartridges were rinsed with 5 mL of Milli-Q deionized water and dried under vacuum for approximately 2 h (or until fully dry). Samples were eluted from the cartridge into glass containers using 5 mL of a solvent mixture containing dichloromethane (Arcos Organics) with 3% methanol. The elution was evaporated to dryness using nitrogen gas and was finally redissolved in 1 mL of a solution consisting of 50% methanol (LC-MS grade, Fisher Scientific) and 50% water (Milli-Q). Plant Tissue Extraction. Details of the plant tissue extraction procedure are presented elsewhere39,40 and described fully in the SI as adapted for MBT. Briefly, each plant sample was freeze-dried overnight on a lyophilizer. A 1.0 mL solution of 1:1

investigated as a potential fertilizer additive to prevent loss of nitrogen from soils.14 MBT can inhibit tryptophan synthase and lactate dehydrogenase in Escherichia coli, as well as phenoloxidases in plants.14 Limited information is available on the fate of MBT in reclaimed water sources. Benzothiazoles are generally in the aqueous phase or bound to tire-wear particles because they are soluble in water and nonvolatile, thus limiting accumulation in sediments or the atmosphere.10 Tire particles and derivatives are known to accumulate in benthic organisms and plants.22 Hydrolysis of MBT is not expected because high pH conditions are required;20 however, oxidation23 and photolysis are relevant abiotic transformation mechanisms. MBT is poorly biodegradable in aerobic systems and is highly refractory to anaerobic treatment.10,12,24 Kloepfer et al.6 observed total benzothiazole removal efficiencies of 5−28% from multiple municipal wastewater treatment plants with effluents concentrations between 1.9−6.7 μg/L. Higher removal efficiencies have been observed in constructed treatment wetlands (83−90%) than conventional wastewater treatment plants.25 Based on the compound log KOW, established models26−28 would predict plant uptake of MBT. MBT will interact with vegetation in many engineered natural treatment systems and water reuse scenarios, such as stormwater bioretention cells, constructed treatment wetlands, or crop irrigation with recycled water. Plant processes are known to impact organic contaminant fate in vegetated stormwater treatment systems,29−31 and contaminants of emerging concern in reclaimed water can present a human exposure route if used for food crop irrigation.32,33 Greater understanding of plant assimilation of trace polar organic contaminants and possible subsequent transformation is needed,34 both of which are important for understanding fate, exposure routes, and potential remediation approaches. The objective of this research was to quantify 2-mercaptobenzothiazole (MBT) vegetative assimilation kinetics (i.e., removal by plants) and elucidate novel plant metabolites using the model plant Arabidopsis thaliana.

■

MATERIALS AND METHODS Chemicals. Important chemicals used in these experiments include the following: 2-mercaptobenzothiazole (MBT; Fluka Analytical) and deuterated 2-mercaptobenzothiazole-d4 (MBTd4; Medical Isotopes Inc.). All solvents and chemicals used for LC/MS analysis were of LC/MS grade. Experimental Design. Experimental Design. Wild-type Arabidopsis thaliana (Col-0 ecotype) plants were grown from seed in antiseptic hydroponic medium in translucent Magenta boxes (Figures S.1, S.2). Arabidopsis is commonly used as a model plant for many fundamental biological and metabolism experiments (e.g., refs 35−38) and was chosen for this work to enhance prediction of transformation products through knowledge of established metabolic pathways. Each experiment was composed of two periods: an initial growth period followed by a pollutant spike and harvesting period (Figure S.3). The growth period occurred from day 0 to day 14 during which the plants sprouted and grew without MBT present in the Magenta box. On day 14, the hydroponic medium was exchanged and refilled with medium containing MBT. The first day of the pollutant spike marked the start (t = 0) of the harvesting period. Three biological replicate boxes containing 30 seedlings each were harvested from the growth chamber per day; the entire contents was harvested for each replicate at each time B


Article


Figure 1. (A) Relative concentration of MBT in the hydroponic medium over time in the treatment (containing plants) and abiotic controls. Error bars indicate standard error of the mean; some error bars are obscured by data symbols. The first spike concentration was 18 μg/L MBT at t = 0 h in all conditions (treatment, negative control, and dark control) and the respike of 36 μg/L MBT on t = 95 h occurred only in the treatment condition. Note that treatment concentrations after t = 120 h are below the method detection limit (MDL) and are reported as 1/2 MDL (0.23 μg/L). (B) Mass balance of MBT in the plant tissue and hydroponic medium for the plant-box system (treatment) over a total of 7 days. Error bars indicate standard error of the mean; some error bars are obscured by data symbols. The first spike of 0.457 μg MBT occurred at t = 0 h. At t = 95 h, the media in the remaining plant boxes was exchanged and refilled with medium containing 0.893 μg MBT (spike 2), thus resulting in a total mass in each plant-box system of 1.350 μg MBT and the discontinuous representation of the media data but not plant data.

curve was used to account for surrogate recovery and matrix effects during ionization and was run with each sample analysis. Surrogate recovery was between 91 and 108% (described in the SI). The method detection limit41 (MDL) for MBT was 0.46 μg/L. Media from the negative controls and dark controls were directly injected and not subjected to SPE; the media in the treatment boxes were both directly injected and underwent SPE as a quality assurance measure. The plant tissue from treatment (all concentration levels) and positive control was extracted as described, and the extract was directly analyzed. Additional quality assurance procedures are described in the SI. Data Analysis. The following statistical tests were conducted to assess differences between sample sets: paired t tests, one-way ANOVA, linear regression, and the Extra Sum of Squares F-Test. All statistical analysis was conducted using GraphPad Prism (version 5.04). Pairing was based on sampling time (t = 0, t = 1, etc.). If not stated explicitly, all analyses were evaluated at the 95% confidence level (α = 0.05).

water and methanol, a methanol-rinsed stainless-steel homogenizing ball, and internal standard (MBT-d4) were added to the dry plant tissue, vortexed briefly, and then frozen at −80 °C. Samples were then homogenized for 5 min at 30 Hz using a ball homogenizer, followed by vortexing, sonication, and centrifugation to separate plant tissue and supernatant. The supernatant was removed and then filtered through a 0.2 μm PTFE syringe filter (13 mm, Fisherbrand). The aforementioned procedure was repeated an additional two times to create a sequential extraction. The internal standard surrogate recovery method (using MBT-d4) was used to account for sample loss during the extraction procedure and matrix effects associated with the plant tissue (i.e., ionization suppression or enhancement). All reported results are normalized to internal standard surrogate recovery. MBT Quantification with LC-MS/MS. High performance liquid chromatography (HPLC) (Shimadzu SCL-10A) was used for chromatographic separation of the samples. The autosampler model was a Shimadzu SIL-20A HT. Separation was performed on a Targa Sprite C18 column (2.1 mm × 40 mm, 5 μm; Higgins Analytical TR-0421-C185) at a total flow rate of 0.3 mL/min. Eluent A was HPLC grade water and eluent B was methanol (Fisher Scientific), each containing 0.4% formic acid (Fisher Scientific). The gradient flow regime was programmed to linearly increase from 0% to 10% B for 1 min and to 60% B during the next 2 min. The conditions stayed constant at 60% B for the next 2 min, then decreased to 30% B within 1.75 min, and finally decreased to 0% in 0.25 min. The method allowed for an equilibration time of 12 min between the samples. Sample injection volume was set to 10 μL. MBT and the internal standard MBT-d4 were detected by multiple reaction monitoring (MRM) mode using negative electrospray ionization mass spectrometry (ESI-MS) on an API 3000 triple-quadrupole mass spectrometer (Applied Biosystems API 3000). The nebulizer gas flow was set to 6000 mL/min. Two transitions for each target analyte were measured as a quality assurance measure. Details on MS, ionization parameters, and MS/MS fragments are in Tables S.1 and S.2. A six-point internal standard normalized external calibration

■

RESULTS AND DISCUSSION Plant MBT Assimilation and Mass Balance. The experimental systems containing Arabidopsis plants experienced significantly greater MBT concentration reductions in the hydroponic medium compared to any losses in the abiotic controls (>18 times faster, p = 0.0056; Figure 1), indicating that plants impacted the fate of MBT. Statistically significant changes from the initial concentration in the negative and dark controls (p ≤ 0.002) also occurred, presumably due to both MBT photolysis and other abiotic processes (e.g., hydrolysis or oxidation). Of the 0.457 μg total MBT mass input to each box system, 9.8% was present in approximately 0.04 ± 0.01 g dry plant tissue (0.73 ± 0.20 g fresh weight) after 3 h, at which time 26.3% of MBT mass remained in the medium (Figure 1). On day 3, the MBT concentration in the medium was below the MDL, and the MBT mass in the plants decreased to 2.0% of the total input. For each sample point, a mass balance on the whole plant-box system was conducted (Figure 1), indicating average MBT mass losses in the C


Article


Figure 2. Second-order degradation model fitted to the C/C0 (medium) planted treatment data by nonlinear regression for hydroponic systems with different initial MBT concentrations (black circle). Negative control data at the same initial concentration are also shown (blue diamond). Although abiotic MBT losses occur, the planted systems have significantly greater MBT losses in the medium compared to the abiotic controls. Error bars indicate standard error of the mean; some error bars are obscured by data symbols. The higher baseline for the lowest concentration (C0 = 1.59 μg/ L) was because the concentrations for this treatment were 98% for later time points. The MBT concentration depletion in the hydroponic medium and mass accumulation in the plants occurred through two sequential MBT spikes (C0 = 18 μg/L and 36 μg/L, respectively) into the system (Figure 1). In both spiking events, an increase in MBT mass in the plant tissue was observed immediately followed the spiking of the pollutant to the medium. A respike (0.893 μg MBT) resulted in a subsequent mass increase in the plants to 2.4% of total mass after 2 h. The initial MBT mass accumulation in the plant tissue following each spike was subsequently followed by significant decreases in MBT parent compound mass in the plant tissue (p = 0.0068 and 0.0016, respectively, for Spike 1 and Spike 2), likely due to plant transformation as described below. By day 7, the MBT mass in the plants had declined to 1.0% of total, whereas the concentration in the medium was below the MDL by day 5. MBT Depletion Kinetics are Concentration Dependent. The relative change of MBT concentration in the hydroponic medium was quantified through time for four different initial concentrations (Figure 2). Kinetics most closely followed a second-order model (eq S.1, Figure S.13) and were D


Article


the plant tissue observed for triclosan and benzotriazole was due to in-planta metabolism and is consistent with the results observed for MBT. Indeed, a possible explanation for the observed phenomena includes rapid partitioning of MBT into plant tissue with subsequent biotransformation. Plant assimilation kinetics in the present experiment system were comparable to or substantially greater than other environmentally relevant loss processes for MBT. Reported first-order photolysis half-lives are between 1.2 and 28.8 h and vary by season and presence of organic matter.11,49 Brownlee at al.11 suggested that photolysis rates of MBT followed a zeroorder reaction at higher concentrations, whereas at lower concentration levels or in the presence of UV-absorbing material (e.g., DOM), rates would follow a first-order or mixed-order reaction. Reemtsma et al.12 observed that in 28 days, 87% of the MBT remained unchanged, and only 10% was converted to 2-methylthiobenzothiazole in an aerobic activated sludge batch system. Thus, plant assimilation kinetics are highly relevant compared to other transformation processes in engineered and natural treatment systems. MBT Accumulation in Plant Tissue and Evidence of Metabolite Formation. The significantly greater depletion of MBT in the hydroponic medium containing plants compared to abiotic controls concurrent with minimal MBT mass accumulation in the plant tissues suggests that Arabidopsis plants metabolized MBT. Accumulation of MBT in plant tissues appears to be only temporary and does not explain substantial (up to 98%) mass losses revealed by the mass balance over the whole plant-box system (Figure 1). When the plant tissues and media samples from the highest MBT exposure concentrations (1,973 μg/L and 147 μg/L) were analyzed, multiple peaks in the chromatograms contained the same MS/MS signature as MBT with different retention times than MBT (Figure 4), but only one peak had the same retention time as the internal standard (MBT-d4; Figure S.8, Figure S.9). In all negative control samples, only one peak was present and was clearly identified as the MBT parent compound through coelusion (i.e., identical RT) with MBTd4 (Figure S.10). Thus, abiotic processes can be excluded as the source of the unidentified peaks, and the effects can be fully attributed to the plants. Because plant transformation products are typically larger molecules than the parent compound via conjugation,39,45,48 in-source fragmentation on the mass spectrometer can reveal the presence of plant metabolites that have different chromatographic retention times while still measuring the identical mass-to-charge ratio as the parent compound because the weakly conjugated bond separates prior to MS/MS analysis. For example, Holder et al.50 observed the glucuronide moiety was easily lost from soy genistein glucuronides through in-source fragmentation, and Justesen51 detected aglycone fragments produced by in-source fragmentation from flavonoid glycosides in herbs. Temporal evolution of MBT plant metabolites was evident (Figure 4), with relative representation shifting from MBT and a more polar metabolite to metabolites less polar than the parent compound. Based on the different retention times, metabolites were identified as M1 (RT = 4.1 min), M2 (RT = 4.6 min), and M3 (RT = 4.7 min). The time-dependent relative peak-area representation of MBT and metabolites, respectively, in plants and in media are shown Figure 4. Note that this figure shows only the relative temporal change in how MBT and the various metabolites are proportionally represented; differential ionization efficiency in the mass spectrometer between

The Extra Sum-of-Squares F test revealed that the assimilation rate constants were significantly different for the systems with varied initial concentrations (p = 0.0002). For the experiments with initial concentrations between 1.59 to 147 μg/L, the rate was greater for higher concentrations (Figure 3).

Figure 3. MBT loss rate constant (k) in the hydroponic medium is related to the initial concentration (C0). Higher initial concentrations result in a greater rate, suggestive of an enzymatic process, until a potentially toxic or inhibitory exposure concentration is reached. Error bars are the standard error of the best-fit rate constant value. Note that the x-axis (C0) is on log-scale.

Nevertheless, for the initial MBT concentration value above 147 μg/L, there was a decrease in the rate constant. There may be a critical concentration value for maximum MBT plant assimilation, above which the rate is reduced. This saturation effect is typical of enzyme-catalyzed reactions wherein the reaction rate increases with increasing substrate concentration until the enzyme becomes substrate saturated; therefore, the rate does not increase further.46 Our observation that the highest initial concentration exposure levels of MBT led to a lower assimilation rate may potentially be from toxic or inhibitory effects on the plants. There is evidence that tire wear compounds can be toxic to plants22,47 and that MBT can inhibit some plant enzymes.14 No short-term growth inhibition was observed for two different MBT concentration exposure experiments as measured by dry plant total biomass (Figure S.6, p > 0.05); however, conclusions regarding long-term toxic effects of MBT on plants cannot be made from the available data. Additionally, it should be noted that the rate constants reported are for a single initial spike to the plants and the rate constants may vary with subsequent exposure as described above. The observed assimilation kinetic results are consistent with recent work on plant assimilation of other trace organic contaminants.39,48 Arabidopsis plants have been shown to assimilate benzotriazole rapidly.39 Benzotriazole and MBT have structural similarities in that they are heterocyclic compounds with an available nitrogen-bonding site, but benzotriazole is different in that it does not contain a thiol (SH−) group. The second-order kinetics observed for MBT is also consistent with literature observations. Plant assimilation of triclosan is best described by a second-order rate model.48 Triclosan depleted rapidly in medium with carrot cell cultures from 17.2 μg to 0.9 μg after 24 h, with a corresponding half-life of 9 h. Accumulation of triclosan in the cell material occurred rapidly within 2 h and then decreased constantly after 5 days.48 Benzotriazole mass also decreased following an initial influx into the plant.39 Pollutant assimilation with subsequent loss in E


Article


Figure 4. Chromatograms (left) of MBT metabolite evolution in the plant tissue and hydroponic medium observed via in-source fragmentation on the mass spectrometer. M1 is a more polar metabolite that is rapidly formed and then transformed to a less polar metabolite M2 and M3. The retention time of the parent compound remained constant and was verified with a deuterated internal standard (see SI; RT = 4.5 min). Representation (right) of MBT and MBT metabolites on total mass in plants and hydroponic medium at an MBT initial concentration of 1973 μg/ L. Values are the mean of triplicate samples with standard error. Metabolite representation was determined by quantifying the peak area of the MBT molecule represented on the MS/MS following in-source fragmentation. Note that the metabolite distribution is based on peak area and thus is only intended to convey the temporal change in representation of MBT and the metabolites rather than on a mass basis (standards for quantification are not available).

and laccases secreted from roots can biotransform some xenobiotic substances on cell surfaces prior to entrance into roots cells.43,54 Although we cannot determine if MBT metabolites were produced in-planta and subsequently released, this phenomenon has been suggested for glycosylated plant metabolites of triclosan48 and benzotriazole.39 Release of conjugated pollutants from plants has implications for phytoremediation applications because the parent compound may not be detected but is present in a slightly altered form in the environment and may thus underestimate total pollutant loading to the environment. Understanding fundamental fate and transformation mechanisms is important to engineering treatment systems and environmental monitoring. Proposed Plant Transformation Products. We propose several probable MBT plant metabolites based on accurate mass LC-QTOF-MS targeted formula search (Table 1). Extracts from three MBT exposed plants (n = 3) were compared to plant extracts not exposed to MBT using the same chromatography and mass spectrometer conditions described elsewhere.39 We employed the “targeted formula search” feature of Agilent Mass Hunter seeking hypothesized structural analogues to newly discovered benzotriazole plant metabolites.39 Like benzotriazole, MBT appears to conjugate with glucose as a major metabolite, likely representing M1, the more polar metabolite (Figure 4, Table 1). The m/z for in positive ionization mode for the glycosylated MBT was 330.0453 amu,

molecules prevents direct comparisons on a mass-basis and must only be used to compare changes in representation;48 authentic standards would be required for true quantitation.39 Metabolite M1 is more polar than the parent compound, whereas M2 and M3 are less polar than MBT (i.e., greater RT). The chromatograms indicate that after only 3 h of exposure to MBT in solution, the plant extract contains metabolites (M1 and M2) in addition to MBT, while in the medium samples MBT is the only compound detected. In the plant samples the peaks for MBT and M1 gradually decreased from t = 3 h to t = 51 h, and M2 became the most pronounced peak. In the medium the first metabolites were detected after 12 h. Unlike the plant extracts, the medium samples contained an increasing peak for M3 that was not detected in the plants. Like the plant tissue samples, the peaks in the media samples for MBT and M1 continuously decreased between 12 and 51 h. M2, which was the most pronounced metabolite in the plant extracts, was measured in the media after 1 day but to a lesser extent than M3. Conjugated pollutant metabolites are generally assumed to be sequestered in cell wall vacuoles and not bioavailable because they cannot diffuse across membranes;43,52 however, we observed MBT metabolites present in both the plant tissue and hydroponic medium. One possible explanation for the observed metabolite formation in the hydroponic medium is transformation directly at the root surface. Root surfaces are active sites of high enzyme activity;53 for example, peroxidases F


Article


Table 1. Proposed Plant Metabolites of 2-Mercaptobenzothiazole Measured via Agilent Mass Hunter Targeted Formula Search

# “Score” reported is fit determined by Agilent Mass Hunter (max score = 100). Exact mass (expressed here to five decimal places; instrument computed average of three biological replicates) of the compound measured is compared to exact mass for the hypothesized formula; difference is provided (expressed in ppm and mDa). §The MBT parent compound measured under the same conditions is provided for reference. Parent compound retention time and m/z were verified with an authentic standard. PLEASE NOTE: The “Find by Formula” feature only determines presence of a given formula but cannot determine the structure. Limited MS/MS fragmentation was conducted (see the SI) on targeted metabolites; structures hypothesized from results on related compounds in the literature.39 RT values for MBT differ above from Figure 2 because chromatography conditions were different.

biosynthesis pathway;39 the results herein suggest that MBT may undergo similar transformation processes. Although MBT is more structurally different from indole than is benzotriazole, many plant amino acid transporters are known to be somewhat nonspecific57,58 and are conserved between Arabidopsis and other higher plants.57 For example, glutathione S-transferases are known to catalyze conjugation of xenobiotics for detoxification as well as play a role in plant hormones, including auxin.52 The hypothesized MBT plant metabolites presented herein are based on targeted searches for predicted products39 and known Arabidopsis metabolic pathways (e.g., plantcyc.org) but only limited MS/MS fragment analysis was performed, and thus the proposed identification of these metabolites should be viewed at a level 3 confidence as per the Schymanski et al. framework;55 further metabolomics-based approaches or compound synthesis would be required for more robust identification.

and the MS/MS analysis showed MBT (m/z = 167.9927 amu), thus indicating the loss of the glucose moiety (C6H12O5) during fragmentation (Figure S.11). Both M1 observed during in-source fragmentation on the LC-MS/MS and the proposed glycosylated MBT (extracted formula C13H15NS2O5) have shorter retention time than the parent compound (Table 1), indicative of increased polarity. Glycosylated benzotriazole, confirmed to a Level 1 Confidence55 in previous work,39 also has a lower RT than the parent compound. NH2− and SH− functional groups (both present in MBT) on xenobiotic molecules are known to trigger glucosyltransferases as a detoxification route.45 These functional groups permit direct and often rapid conjugation in Phase II metabolism,45 allowing bypass of the often rate-limiting56 Phase I metabolism (e.g., ring hydroxylation). The targeted search also included amino acid conjugated MBT metabolites that occurred in analogous forms for benzotriazole (Table 1). All proposed metabolites were present in the treatments (MBT exposed plants) and absent from the controls (MS/MS fragmentation data in the SI). Amino acid conjugation has been observed infrequently and is thought to be a side reaction rather than a main route of detoxification.45 Previous research demonstrated that benzotriazole acted as an indole mimic and can be incorporated into the tryptophan

■

ENVIRONMENTAL IMPLICATIONS

The rapid assimilation and metabolism of MBT have applications to phytoremediation and in water reuse, particularly in stormwater reclamation of urban street runoff or trace organics polishing of wastewater effluents. The pollution control benefits associated with plants in urban G



■

stormwater bioretention systems has been established for some pollutants.30,31,59,60 For example, floating treatment wetlands have been used as an engineering retrofit in stormwater retention ponds to create a hydroponic root zone in the water column to remove dissolved pollutants.61 MBT removal observed in our laboratory hydroponic systems may undergo similar processes using macrophyte floating treatment wetlands. Nevertheless, the results presented herein employ a model plant under sterile hydroponic conditions, and assimilation kinetics may be very different under field conditions. Although many relevant plant metabolic pathways are conserved across species62 (e.g., tryptophan biosynthesis), the presence of soils/ natural organic matter and bacteria, as well as varied plant species, will alter bioavailability and assimilation/transformation propensity and kinetics in natural environments. Furthermore, the degree of contact of the plant (roots only or roots and shoot/leaf) with dissolved pollutants will impact contaminant assimilation in the environment, as well as translocation to edible portions of crops. Field crops do, however, take up and metabolize some trace organic contaminants from soils.32,33,63 We present these results as a way of better understanding the processes by which plants can affect MBT as an emerging contaminant in engineered natural treatment systems for stormwater and reclaimed water but recognize the limits of a model plant system and that species-dependent contaminant interactions exist. Further research is required to comprehensively determine plant impacts to MBT fate in passive stormwater treatment facilities used to enhance water quality, particularly for native plant species typically selected and for particle-bound contaminants. MBT transformation products generated via plant metabolism vary considerably from reported microbial MBT transformation products, such as methylated and hydroxylated MBT metabolites in the natural environment and activated sludge.10−12,14,64 The results of this study demonstrate that the core structure of the MBT molecule is not altered during plant metabolism but rather conjugated to form other larger more complex compounds. Thus, measurement of the MBT parent compound alone in plant tissues or other environmental media may underestimate total mass because MBT plant metabolites are not targeted for analysis. The potential to underestimate trace organic contaminant loading is particularly important when reclaimed water is used for irrigation of food crops because parent compounds could be masked in the form of conjugated metabolites and the bioavailability/toxicity of these metabolites to consumers is often unknown. Assimilation and phytotransformation of organic contaminants is vital to understanding environmental fate,52,56,65−71 particularly as use of reclaimed water for irrigation and groundwater recharge expands in response to population growth, water scarcity, and drier climates.

■

Article

AUTHOR INFORMATION

Corresponding Author

*Phone: 650-721-2615. Fax: 650-725-8662. E-mail: luthy@ stanford.edu. Corresponding author address: 473 Via Ortega, Stanford University, Stanford, CA 94305, United States. Present Address ∥

Department of Civil and Environmental Engineering, University of Iowa, Iowa City, IA 52242

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors are affiliated with ReNUWIt, the engineering research center for Re-Inventing the Nation’s Urban Water Infrastructure (NSF ERC 1028968). The authors acknowledge financial support from the Stanford Woods Institute Environmental Ventures Program. A.C.P. was supported by the Zeno Karl Schindler foundation.

■

REFERENCES

(1) Bischel, H. N.; Simon, G. L.; Frisby, T. M.; Luthy, R. G. Management experiences and trends for water reuse implementation in northern California. Environ. Sci. Technol. 2012, 46, 180−188. (2) LeFevre, G.; Paus, K.; Natarajan, P.; Gulliver, J.; Novak, P.; Hozalski, R. Review of dissolved pollutants in urban storm water and their removal and fate in bioretention cells. J. Environ. Eng. 2015, 141, 04014050. (3) Davis, A. P.; Hunt, W. F.; Traver, R. G.; Clar, M. Bioretention technology: Overview of current practice and future needs. J. Environ. Eng. 2009, 135, 109−117. (4) Roy-Poirier, A.; Champagne, P.; Filion, Y. Review of bioretention system research and design: past, present, and future. J. Environ. Eng. 2010, 136, 878−889. (5) Grebel, J. E.; Mohanty, S. K.; Torkelson, A. A.; Boehm, A. B.; Higgins, C. P.; Maxwell, R. M.; Nelson, K. L.; Sedlak, D. L. Engineered infiltration systems for urban stormwater reclamation. Environ. Eng. Sci. 2013, 30, 437−454. (6) Kloepfer, A.; Jekel, M.; Reemtsma, T. Occurrence, sources, and fate of benzothiazoles in municipal wastewater treatment plants. Environ. Sci. Technol. 2005, 39, 3792−3798. (7) Zeng, E. Y.; Tran, K.; Young, D. Evaluation of potential molecular markers for urban stormwater runoff. Environ. Monit. Assess. 2004, 90, 23−43. (8) Ni, H.; Lu, F.; Luo, X.; Tian, H.; Zeng, E. Y. Occurrence, phase distribution, and mass loadings of benzothiazoles in riverine runoff of the Pearl River Delta, China. Environ. Sci. Technol. 2008, 42, 1892− 1897. (9) Reddy, C. M.; Quinn, J. G. Environmental chemistry of benzothiazoles derived from rubber. Environ. Sci. Technol. 1997, 31, 2847−2853. (10) De Wever, H.; Besse, P.; Verachtert, H. Microbial transformations of 2-substituted benzothiazoles. Appl. Microbiol. Biotechnol. 2001, 57, 620−625. (11) Brownlee, B. G.; Carey, J. H.; MacInnis, G. A.; Pellizzari, I. T. Aquatic environmental chemistry of 2-(thiocyanomethylthio)benzothiazole and related benzothiazoles. Environ. Toxicol. Chem. 1992, 11, 1153−1168. (12) Reemtsma, T.; Fiehn, O.; Kalnowski, G.; Jekel, M. Microbial transformations and biological effects of fungicide-derived benzothiazoles determined in industrial wastewater. Environ. Sci. Technol. 1995, 29, 478−485. (13) Bujdáková, H.; Kuchta, T.; Sidóová, E.; Gvozdjaková, A. AntiCandida activity of four antifungal benzothiazoles. FEMS Microbiol. Lett. 1993, 112, 329−334. (14) De Wever, H.; Verachtert, H. Biodegradation and toxicity of benzothiazoles. Water Res. 1997, 31, 2673−2684.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.est.5b04716. Additional method details, quality assurance/control, additional detailed results (PDF) H


Article


Environ. Sci. Technol. 2015. Just Accepted Manuscript. DOI: 10.1021/ acs.est.5b01546. (35) Chen, Y.; Xu, W.; Shen, H.; Yan, H.; Xu, W.; He, Z.; Ma, M. Engineering arsenic tolerance and hyperaccumulation in plants for phytoremediation by a PvACR3 transgenic approach. Environ. Sci. Technol. 2013, 47, 9355−9362. (36) Mashiguchi, K.; Tanaka, K.; Sakai, T.; Sugawara, S.; Kawaide, H.; Natsume, M.; Hanada, A.; Yaeno, T.; Shirasu, K.; Yao, H.; McSteen, P.; Zhao, Y.; Hayashi, K.; Kamiya, Y.; Kasahara, H. The main auxin biosynthesis pathway in Arabidopsis. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 18512−18517. (37) Su, Z.; Xu, Z.; Peng, R.; Tian, Y.; Zhao, W.; Han, H.; Yao, Q.; Wu, A. Phytoremediation of trichlorophenol by phase II metabolism in transgenic Arabidopsis overexpressing a populus glucosyltransferase. Environ. Sci. Technol. 2012, 46, 4016−4024. (38) Wang, J.; Koo, Y.; Alexander, A.; Yang, Y.; Westerhof, S.; Zhang, Q.; Schnoor, J. L.; Colvin, V. L.; Braam, J.; Alvarez, P. J. J. Phytostimulation of poplars and Arabidopsis exposed to silver nanoparticles and ag+ at sublethal concentrations. Environ. Sci. Technol. 2013, 47, 5442−5449. (39) LeFevre, G. H.; Müller, C. E.; Li, R. J.; Luthy, R. G.; Sattely, E. S. Rapid phytotransformation of benzotriazole generates synthetic tryptophan and auxin analogs in Arabidopsis. Environ. Sci. Technol. 2015, 49 (18), 10959−10968. (40) Müller, C. E.; LeFevre, G. H.; Timofte, A. E.; Hussain, F. A.; Sattely, E. S.; Luthy, R. G. Competing mechanisms for perfluoroalkylacid accumulation in plants revealed using an Arabidopsis model system. Environ. Toxicol. Chem. 2015, DOI: 10.1002/etc.3251. (41) Eaton, A. D.; Clesceri, L. S.; Grennber, A. E. Standard methods for the examination of water and wastewater; APHA, AWWA, WEF: Washington, DC, 1995; p 1082. (42) Ali-Zade, V.; Alirzayeva, E.; Shirvani, T. Plant resistance to anthropogenic toxicants: Approaches to phytoremediation. In Plant Adaptation and Phytoremediation; Ashraf, M., Ozturk, M., Ahmad, M. S. A., Eds.; Springer: Netherlands, 2010; pp 173−192. (43) Bártíková, H.; Skálová, L.; Stuchlíková, L.; Vokřaĺ , I.; Vaněk, T.; Podlipná, R. Xenobiotic-metabolizing enzymes in plants and their role in uptake and biotransformation of veterinary drugs in the environment. Drug Metab. Rev. 2015, 47, 374−387. (44) Campos, V.; Souto, L. S.; Medeiros, T. A. M.; Toledo, S. P.; Sayeg, I. J.; Ramos, R. L.; Shinzato, M. C. Assessment of the removal capacity, tolerance, and anatomical adaptation of different plant species to benzene contamination. Water, Air, Soil Pollut. 2014, 225, 1−12. (45) Schröder, P. Exploiting plant metabolism for the phytoremediation of organic xenobiotics. In Phytoremediation: Methods and Reviews; Willey, N., Ed.; Humana Press: 2007; Vol. 23, pp 251−263. (46) Rittmann, B.; McCarty, P. Environmental Biotechnology: Principles and Applications; McGraw Hill: Madison, WI, 2001. (47) He, G.; Zhao, B.; Denison, M. S. Identification of benzothiazole derivatives and polycyclic aromatic hydrocarbons as aryl hydrocarbon receptor agonists present in tire extracts. Environ. Toxicol. Chem. 2011, 30, 1915−1925. (48) Macherius, A.; Eggen, T.; Lorenz, W.; Moeder, M.; Ondruschka, J.; Reemtsma, T. Metabolization of the bacteriostatic agent triclosan in edible plants and its consequences for plant uptake assessment. Environ. Sci. Technol. 2012, 46, 10797−10804. (49) Malouki, M. A.; Richard, C.; Zertal, A. Photolysis of 2mercaptobenzothiazole in aqueous medium: Laboratory and field experiments. J. Photochem. Photobiol., A 2004, 167, 121−126. (50) Holder, C. L.; Churchwell, M. I.; Doerge, D. R. Quantification of soy isoflavones, genistein and daidzein, and conjugates in rat blood using LC/ES-MS. J. Agric. Food Chem. 1999, 47, 3764−3770. (51) Justesen, U. Negative atmospheric pressure chemical ionisation low-energy collision activation mass spectrometry for the characterisation of flavonoids in extracts of fresh herbs. Journal of Chromatography A 2000, 902, 369−379. (52) Burken, J. G. Uptake and metabolism of organic compounds: Green liver model. In Phytoremediation: Transformation and Control of

(15) Jung, J. H.; McLaughlin, J. L.; Stannard, J.; Guin, J. D. Isolation, via activity-directed fractionation, of mercaptobenzothiazole and dibenzothiazyl disulfide as 2 allergens responsible for tennis shoe dermatitis. Contact Dermatitis 1988, 19, 254−259. (16) Saha, M.; Srinivas, C. R.; Shenoy, S. D.; Balachandran, C.; Acharya, S. Footwear dermatitis. Contact Dermatitis 1993, 28, 260− 264. (17) Whittaker, M. H.; Gebhart, A. M.; Miller, T. C.; Hammer, F. Human health risk assessment of 2-mercaptobenzothiazole in drinking water. Toxicol. Ind. Health 2004, 20, 149−163. (18) Nawrocki, S. T.; Drake, K. D.; Watson, C. F.; Foster, G. D.; Maier, K. J. Comparative aquatic toxicity evaluation of 2(thiocyanomethylthio)benzothiazole and selected degradation products using Ceriodaphnia dubia. Arch. Environ. Contam. Toxicol. 2005, 48, 344−350. (19) Gold, L. S.; Slone, T. H.; Stern, B. R.; Bernstein, L. Comparison of target organs of carcinogenicity for mutagenic and non-mutagenic chemicals. Mutat. Res., Fundam. Mol. Mech. Mutagen. 1993, 286, 75− 100. (20) Hanssen, H. W.; Henderson, N. D. A review of the environmental impact and toxic effects of 2-MBT. Environmental Protection Division, Ministry of Environment BC Environment (British Columbia): 1991 (Oct). http://www.env.gov.bc.ca/eirs/ epd/environmental_protection_publications_index/epdStaticPageR. htm (accessed Dec 1, 2015). (21) De Wever, H.; Besse, P.; Verachtert, H. Microbial transformations of 2-substituted benzothiazoles. Appl. Microbiol. Biotechnol. 2001, 57, 620−625. (22) Wik, A.; Dave, G. Occurrence and effects of tire wear particles in the environment − A critical review and an initial risk assessment. Environ. Pollut. 2009, 157, 1−11. (23) Kloepfer, A.; Jekel, M.; Reemtsma, T. Determination of benzothiazoles from complex aqueous samples by liquid chromatography−mass spectrometry following solid-phase extraction. Journal of Chromatography A 2004, 1058, 81−88. (24) Fiehn, O.; Reemtsma, T.; Jekel, M. Extraction and analysis of various benzothiazoles from industrial wastewater. Anal. Chim. Acta 1994, 295, 297−305. (25) Matamoros, V.; Jover, E.; Bayona, J. M. Occurrence and fate of benzothiazoles and benzotriazoles in constructed wetlands. Water Sci. Technol. 2010, 61, 191−198. (26) Dettenmaier, E. M.; Doucette, W. J.; Bugbee, B. Chemical hydrophobicity and uptake by plant roots. Environ. Sci. Technol. 2009, 43, 324−329. (27) Limmer, M. A.; Burken, J. G. Plant translocation of organic compounds: Molecular and physicochemical predictors. Environ. Sci. Technol. Lett. 2014, 1, 156−161. (28) Burken, J. G.; Schnoor, J. L. Predictive relationships for uptake of organic contaminants by hybrid poplar trees. Environ. Sci. Technol. 1998, 32, 3379−3385. (29) LeFevre, G. H.; Hozalski, R. M.; Novak, P. J. Root exudate enhanced contaminant desorption: An abiotic contribution to the rhizosphere effect. Environ. Sci. Technol. 2013, 47, 11545−11553. (30) LeFevre, G. H.; Hozalski, R. M.; Novak, P. J. The role of biodegradation in limiting the accumulation of petroleum hydrocarbons in raingarden soils. Water Res. 2012, 46, 6753−6762. (31) LeFevre, G. H.; Novak, P. J.; Hozalski, R. M. Fate of naphthalene in laboratory-scale bioretention cells: implications for sustainable stormwater management. Environ. Sci. Technol. 2012, 46, 995−1002. (32) Goldstein, M.; Shenker, M.; Chefetz, B. Insights into the uptake processes of wastewater-borne pharmaceuticals by vegetables. Environ. Sci. Technol. 2014, 48, 5593−5600. (33) Malchi, T.; Maor, Y.; Tadmor, G.; Shenker, M.; Chefetz, B. Irrigation of root vegetables with treated wastewater: Evaluating uptake of pharmaceuticals and the associated human health risks. Environ. Sci. Technol. 2014, 48, 9325−9333. (34) Miller, E. L.; Nason, S. L.; Karthikeyan, K. G.; Pedersen, J. A. Root uptake of pharmaceutical and personal care product ingredients. I


Article

Environmental Science & Technology Contaminants; McCutcheonson, S. C., Schnoor, J. L., Eds.; J. Wiley and Sons: Hoboken, NJ, 2003; pp 59−84. (53) Chasov, A. V.; Alekseeva, V. Y.; Kolesnikov, O. P.; Minibayeva, F. V. Activation of extracellular peroxidase of wheat roots under the action of xenobiotics. Appl. Biochem. Microbiol. 2010, 46, 431−437. (54) Wang, G.; Li, Q.; Luo, B.; Chen, X. Ex planta phytoremediation of trichlorophenol and phenolic allelochemicals via an engineered secretory laccase. Nat. Biotechnol. 2004, 22, 893−897. (55) Schymanski, E. L.; Jeon, J.; Gulde, R.; Fenner, K.; Ruff, M.; Singer, H. P.; Hollender, J. Identifying small molecules via high resolution mass spectrometry: Communicating confidence. Environ. Sci. Technol. 2014, 48, 2097−2098. (56) Zhai, G.; Lehmler, H.; Schnoor, J. L. Inhibition of cytochromes p450 and the hydroxylation of 4-monochlorobiphenyl in whole poplar. Environ. Sci. Technol. 2013, 47, 6829−6835. (57) Rentsch, D.; Schmidt, S.; Tegeder, M. Transporters for uptake and allocation of organic nitrogen compounds in plants. FEBS Lett. 2007, 581, 2281−2289. (58) Fischer, W.; André, B.; Rentsch, D.; Krolkiewicz, S.; Tegeder, M.; Breitkreuz, K.; Frommer, W. B. Amino acid transport in plants. Trends Plant Sci. 1998, 3, 188−195. (59) Read, J.; Fletcher, T. D.; Wevill, T.; Deletic, A. Plant traits that enhance pollutant removal from stormwater in biofiltration systems. Int. J. Phytorem. 2010, 12, 34−53. (60) Szota, C.; Farrell, C.; Livesley, S. J.; Fletcher, T. D. Salt tolerant plants increase nitrogen removal from biofiltration systems affected by saline stormwater. Water Res. 2015, 83, 195−204. (61) Winston, R. J.; Hunt, W. F.; Kennedy, S. G.; Merriman, L. S.; Chandler, J.; Brown, D. Evaluation of floating treatment wetlands as retrofits to existing stormwater retention ponds. Ecol. Eng. 2013, 54, 254−265. (62) Radwanski, E. R.; Last, R. L. Tryptophan biosynthesis and metabolism: biochemical and molecular genetics. Plant Cell 1995, 7, 921−934. (63) Aryal, N.; Reinhold, D. M. Phytoaccumulation of antimicrobials from biosolids: Impacts on environmental fate and relevance to human exposure. Water Res. 2011, 45, 5545−5552. (64) Ellis, L. B. M.; Gao, J.; Fenner, K.; Wackett, L. P. The University of Minnesota pathway prediction system: predicting metabolic logic. Nucleic Acids Res. 2008, 36, W427. (65) Hyland, K. C.; Blaine, A. C.; Higgins, C. P. Accumulation of contaminants of emerging concern in food crops, part two: Plant distribution. Environ. Toxicol. Chem. 2015, 34, 2222−2230. (66) Hyland, K. C.; Blaine, A. C.; Dickenson, E. R.; Higgins, C. P. Accumulation of contaminants of emerging concern in food crops, part one: Edible strawberries and lettuce grown in reclaimed water. Environ. Toxicol. Chem. 2015, 34, 2213−2221. (67) Blaine, A. C.; Rich, C. D.; Hundal, L. S.; Lau, C.; Mills, M. A.; Harris, K. M.; Higgins, C. P. Uptake of perfluoroalkyl acids into edible crops via land applied biosolids: Field and greenhouse studies. Environ. Sci. Technol. 2013, 47, 14062−14069. (68) Blaine, A. C.; Rich, C. D.; Sedlacko, E. M.; Hundal, L. S.; Kumar, K.; Lau, C.; Mills, M. A.; Harris, K. M.; Higgins, C. P. Perfluoroalkyl acid distribution in various plant compartments of edible crops grown in biosolids-amended soils. Environ. Sci. Technol. 2014, 48, 7858−7865. (69) Card, M. L.; Schnoor, J. L.; Chin, Y. transformation of natural and synthetic estrogens by maize seedlings. Environ. Sci. Technol. 2013, 47, 5101−5108. (70) Phytoremediation: Transformation and Control of Contaminants; McCutcheon, S. C., Schnoor, J. L., Eds.; John Wiley and Sons: Hoboken, NJ, 2003; p 987. (71) Zhai, G.; Gutowski, S. M.; Lehmler, H.; Schnoor, J. L. Enantioselective transport and biotransformation of chiral oh-pcbs in whole poplar plants. Environ. Sci. Technol. 2014, 48, 12213−12220.

J


Plant Assimilation Kinetics and Metabolism of 2 ... - ACS Publications

Recommend Documents