The oral LD50 in mice was found to be similar for BPA and its products , but lower for the product HA . The transformation products of DCL for which LD50 values were available generally had higher LD50 values. However, it must be noted that these threshold values were for acute exposures and may have little relevance to effects at low levels that are typical of environmental contamination. The coupled use of 14C labeling and chromatographic analysis in this study allowed a comprehensive investigation of transformation and removal pathways of four common PPCP/EDCs in soil. The results showed that the primary decontamination mechanisms may vary with compounds. In this study, formation of bound residue was the predominant removal process for BPA and NP, while mineralization was significant for DCL and NPX. In addition, extractable residues consisted of both the parent compound and multiple transformation products, and the relative contribution of the parent varied with compound and incubation time. The abundance of transformation products detected in all soil treatments highlights the importance of a more comprehensive evaluation of PPCP/EDC transformation and fate processes, in order to improve risk assessments of ecosystem and human health effects due to the reuse of treated wastewater and bio-solids. Treated wastewater, commonly called reclaimed or recycled water, is a valuable water source in arid and semi-arid areas where fresh water sources are becoming increasingly scarce due to urbanization and climate change . Reclaimed water may have many beneficial applications,vertical growing including agriculture irrigation and landscape irrigation. In the state of California, these irrigation uses account for 37% and 18%, respectively, of the 650,000 acre-feet per year of water reuse .
State policy calls to increase the use of reclaimed water to more than 2.5 million acre-feet per year by 2030 . Accompanying increased reuse, the presence and environmental risks of unregulated organic contaminants in reclaimed water are drawing attention . Pharmaceutical and personal care products and endocrine disrupting compounds are typically anthropogenic chemicals with known biological effects that may interfere with normal metabolism and behaviors of organisms . Many PPCP/EDCs are routinely found in reclaimed water , as well as in surface water and groundwater impacted by wastewater treatment plant effluent. When reclaimed water is used for irrigation, the associated PPCP/EDCs may interact with the soil matrix and may contaminate groundwater and food crops . Accumulation of PPCP/EDCs into food crops that are consumed fresh, such as many leafy vegetables, is relevant due to the likelihood of unintentional human exposure. If research demonstrates that accumulation of PPCP/EDCs by crops is unlikely to result in human health risks, this will provide scientific basis to promote use of reclaimed water, as well as enhance positive public perception of water reuse. Many factors influence the uptake of organic compounds into plants, such as by affecting diffusion through cell membranes. Briggs et al. suggested that chemical hydrophobicity is an important factor affecting uptake by diffusion and that chemicals with a log Kow of 1 – 3.5 have the greatest plant uptake potential because lipid and aqueous solubility are balanced . In addition to hydrophobicity, molecular ionization has also been shown to influence plant accumulation, such as of herbicides . Charged molecules may have a reduced potential for plant uptake, since ionization may reduce their ability to permeate cell membranes . However, the role of ionization is poorly understood and exceptions have been noted .
To date only a handful of studies have considered plant uptake of PPCP/EDCs . While these studies have clearly shown the ability for plants to take up PPCP/EDCs, the state of knowledge is limited to a few compounds or plant types. Due to the analytical challenges of detecting chemicals at trace levels in plant matrices, most studies also relied on the use of artificially high concentrations, with a few exceptions . In this study, we comparatively determined the accumulation of four commonly occurring PPCP/EDCs, i.e., bisphenol A , diclofenac , naproxen , or nonylphenol , at relevant environmental levels into two leafy vegetables, lettuce and collards, and examined the composition and distribution of accumulated residues. These compounds have been frequently detected in reclaimed water and surface water , and have different ionization states at neutral pH. To achieve realistically low concentrations while affording quantitative measurement, 14C-labeled compounds were used. Results were used to infer effects of plant type and compound characteristics on plant accumulation and estimate probable human intakes.Experimental treatments were created, in triplicate, for each 14C-PPCP/EDC with lettuce or collards plants. A spiked, no-plant control for each PPCP/EDC and a non-spiked control with a collards plant were also included in triplicate. The experiments were conducted in a growth chamber programmed for a 16 h light/8 h dark cycle, with constant 65% relative air humidity and a gradual increase and then decrease of photosynthetic photon flux density that peaked daily at 350 µmol/m2 s. Growth chamber air was freely exchangeable with ambient air. Plant seedlings were removed from packaging and soil, rinsed with deionized water, and placed in jars of continuously aerated nutrient solution, one plant per 2 L jar. Plants were suspended in the nutrient solution by means of a nonreactive foam collar around the stem that secured the plant in an opening in the lid.
After 3 d, the jars and nutrient solution were exchanged with clean jars and fresh nutrient solution to restore nutrient levels and reduce microbial load in the solution. After 6 d of acclimation under the prescribed conditions, plants of similar size for each species were randomly chosen and transferred into new jars containing nutrient solution spiked with14C-BPA, DCL, NPX or NP at, respectively, 46.4, 237.4, 178.2, or 110.4 ng/L . These concentrations are representative of concentrations measured in reclaimed water . Every 3 d after the initial treatment, all plants were transferred into clean jars with fresh,vertical farm system spiked nutrient solution that replicated their initial nutrient and PPCP/EDC conditions. Plants were grown for a total of 21 d in spiked solutions, a total growth time that represents commercial growth to a “market size”. Following 21 d of hydroponic cultivation, plants were sacrificed for analysis of 14C accumulation and distribution. Each whole plant was rinsed with DI water, and then separated into roots, stems, new leaves, and original leaves. Original leaves were designated as leaves present on the seedling at the beginning of the experiment. Individual plant samples were placed in pre-weighed metal screen pouches, weighed to determine wet weight, and dried at 50 °C for 60 h. After drying, each plant sample was weighed to measure the dry weight, and then chopped and mixed in a stainless steel coffee grinder. The grinder was rinsed between samples with DI water and methanol to prevent cross contamination. Multiple 150 mg sub-samples of each plant sample were analyzed until standard deviation of the sub-samples was below 20%, due to notable variation in plant tissue activity. sub-samples were combusted on an OX-500 Biological Oxidizer at 900 °C for 4 min, and the evolved 14CO2 was trapped in 15 mL of Harvey Carbon-14 cocktail . The 14C was measured on a Beckman LS 5000 TD Liquid Scintillation Counter . Recovery was 91 – 96% for spiked standards, which was used to correct for the actual activity. The activity and weight of the sub-samples were used to determine the total radioactivity accumulated in different tissues of each plant. Analysis of 14C by combustion provided information on total residue in plant tissues. To better understand the nature of the residue, plant samples were solvent extracted using a method modified from Wu et al. . The fractions of 14C in solvent-extractable and non-extractable forms were separately determined. Briefly, 400 mg sub-samples of the dried, ground plant matter were freeze-dried for 12 h, weighed, and extracted in polypropylene tubes by sequential sonication and centrifugation with 20 mL methyl tertbutyl ether and then again with 20 mL acetonitrile. The combined extracts were evaporated under nitrogen to less than 1 mL, and mixed with 5 mL methanol and 20 mL water.
A 6 mL aliquot of the extract was taken for analysis by LSC to determine the fraction of activity as extractable residue. Selected 150 mg sub-samples of the solvent-extracted plant matter were combusted on the Biological Oxidizer as described above to determine the fraction of 14C present as non-extractable residue. When nutrient solution and jars were exchanged, the volume of remaining nutrient solution in each jar was gravimetrically determined. A 9 mL aliquot of the solution was mixed with 13 mL Ultima Gold scintillation cocktail and the 14C was quantified by LSC. Water loss from evaporation during each 3 d period was found to be negligible in the no-plant control containers. It is likely that microbial activity in the nutrient solution may have resulted in transformation of the spiked 14C-compounds and that plants may have accumulated both parent PPCP/EDCs and transformation products. To discern the contribution of transformation products to plant accumulation, the used nutrient solution from day 21 was preserved with 2 g sodium azide and 100 mg ascorbic acid, extracted, and fractionated using high performance liquid chromatography . Solutions from 14C-BPA, DCL, or NPX treatments were first filtered through a Whatman #4 filter paper and then passed through a HLB solid phase extraction cartridge . Before use, the cartridges were sequentially conditioned with 5 mL each of MTBE, methanol , and water. The filtered solution was drawn through the conditioned HLB cartridges under vacuum and followed by 50 mL deionized water. A sub-sample of the filtrate that passed through the cartridge was collected for analysis by LSC to quantify 14C that was notretained by the cartridge. The cartridges were dried with nitrogen gas, and then sequentially eluted with 5 mL of MeOH:MTBE and 5 mL MeOH. The collected eluent was dried under nitrogen to 100 µL. The concentrated eluent was transferred to an HPLC vial equipped with a 250 µL insert. The condensing vial was rinsed with 130 µL of methanol, and the rinsate and 20 µL of non-labeled parent standard were added to the HPLC vial. Preliminary experiments showed that the recovery of this extraction procedure from the initial solution to HPLC analysis was 81.5 ± 7.1% for BPA, 85.8 ± 2.5% for DCL, and 74.0 ± 1.9% for NPX. Nutrient solutions from the 14C-NP treatment were extracted by a simple liquid-liquid extraction method. Each nutrient solution sample was shaken with 50mL hexane for 30min, and then the upper layer of the sample was transferred to a centrifuge tube and centrifuged at 3500 rpm for 30 min to reduce emulsification. The hexane phase was transferred to a 15 mL glass tube, concentrated under nitrogen to 300 µL, and transferred to an HPLC vial. The condensing vial was rinsed with 180 µL of methanol, and the rinsate and 20 µL of non-labeled NP standard were added to the HPLC vial. The recovery of this extraction method from the initial solution to HPLC analysis for NP was determined to be 66.8 ± 12.0%. An aliquot of the finalized sample was injected into an Agilent 1100 Series HPLC equipped with a Dionex Acclaim 120 C18 RP column . Column temperature was maintained at 35 °C. Mobile phase was created from ultra-pure water with 0.2% acetic acid and acetonitrile . Flow rate and mobile phase mix were 1.25 mL/min and 60:40 for BPA, 1.25 mL/min and 47:53 for DCL, 1.60 mL/min and 60:40 for NPX, and 1.0 mL/min and 20:80 for NP. Ultraviolet detection was set at 280, 284, 278, and 280 nm, respectively. Retention times were 13.3, 11.6, 13.1, and 11.0 min, respectively, for BPA, DCL, NPX, and NP. The column eluent was fractionally collected in 1 min increments into 7 mL glass tubes using an automated fraction collector and the 14C in each elution sample was measured by LSC. The distribution of 14C in the HPLC eluent as a function of run time was used to infer the fractions of parent and transformation products in the nutrient solution. Young plants of lettuce and collards were grown for 21 d in nutrient solution containing one of the four 14C-labeled PPCP/EDCs.