The annual exposure values ranged from 0.32 × 10-3 mg for BPA-lettuce to 2.14 × 10-2 mg for DCL-collards for an average, 70 kg individual residing in the United States. To place these amounts in context, the values were then converted to either medical dose or 17β-estradiol equivalents. Both DCL and NPX are commonly available non-steroidal anti-inflammatory pharmaceuticals. Based on typical doses and the observed plant concentrations, an average individual would consume the equivalent of much less than one dose of these medicines in a year due to consumption of leafy vegetables, representing a very minor exposure to these PPCPs. However, it should be noted that DCL has proven ecotoxicity and NPX has shown toxicity in mixture with other pharmaceuticals , so a simple estimation may not encompass all possible human health effects. Both BPA and NP are industrial products known to have endocrine disrupting activity. Bonefeld-Jørgensen et al. calculated the Relative Potency of these compounds as compared to 17β-estradiol , an endogenous estrogen hormone, at activating estrogenic receptors. In Table 4.3, the exposure values of BPA and NP were estimated as E2-equivalents by dividing by their Relative Potency . When the calculated E2-equivalents of BPA and NP are compared with the Lowest Observable Effect Concentration for E2 , it is obvious that the even the highest expected annual exposure to these compounds by consuming leafy vegetables would not reach the LOEC. This rough calculation suggests that consumption of vegetables would be unlikely to influence an individual’s overall endocrine activity,container vertical farming though caution should be used when considering risk to susceptible population groups.

Moreover, it must be noted that the use of hydroponic cultivation likely resulted in greater plant accumulation of these PPCP/EDCs, in relation to soil cultivation, due to the absence of chemical sorption to soil organic matter and minerals. This likelihood, when coupled with the fact that most of the 14C in plant tissues was in the non-extractable form, implies that the risk from actual plant accumulation of these PPCP/EDCs by leafy vegetables grown in uncontaminated fields irrigated with reclaimed water may be negligibly small. On the other hand, bio-solids have been shown to contain some PPCP/EDCs at much higher concentrations than treated wastewater and plant uptake from soil amended with bio-solids may pose an enhanced human exposure risk. Also, given that many PPCP/EDCs may be preferentially distributed in plant roots as compared to above-ground tissues , the potential risk may be significantly greater for root vegetables such as carrots, radishes, and onions. The occurrence of these and other PPCP/EDCs in leafy and root vegetables should be evaluated in the field under typical cultivation and management conditions.Population growth, urbanization, and climate change have created unprecedented stress on water resources. The reuse of treated wastewater from wastewater treatment plants is increasing by 15% each year to help meet water needs . As of 2006, about 3.6 × 109 cubic meters of treated wastewater were reused in the U.S. each year for purposes including agricultural and landscape irrigation . Regulations on wastewater reuse are mostly concerned with pathogen and heavy metal contamination . However, numerous studies have shown that a variety of trace organic contaminants are present in treated wastewater, including pharmaceutical and personal care products and endocrine disrupting chemicals . Some PPCP/EDCs have unintended biological effects on nontarget organisms at low concentrations .The beneficial reuse of treated wastewater for agricultural irrigation introduces PPCP/EDCs into the soil environment, where they may be taken up by plants and cause human exposure by ingestion .

While a number of studies have examined the uptake potential of PPCP/EDCs, most studies only considered a few compounds, making it difficult to discern the underlying mechanisms. On the other hand, plant uptake has been extensively investigated for many pesticides and herbicides . Studies show that systemic pesticides are passively taken up through the transpiration stream , and greater transpiration leads to increased accumulation of non-ionic compounds . Many PPCP/EDCs are ionizable compounds that may exist partially as ions at an environmentally relevant pH . The ionic state of a compound greatly affects the compound’s interactions with plants, such as adsorption on root tissue, interaction with the cell membrane, and sequestration into plant compartments . In a recent study, Wu et al. examined multiple PPCP/EDCs and observed a strong correlation between plant bio-concentration of a compound and its pH-adjusted octanol-water partition coefficient , but did not address transpiration effects. Herklotz et al. and Shenker et al. suggested that movement through transpiration-driven mass flow of water was likely an important route for the uptake of carbamazepine, and Carter et al. suggested that transpiration differences between radish and ryegrass contributed to their differential uptake of carbamazepine, diclofenac,fluoxetine, and propanolol. However, to date researchers have yet to quantitatively evaluate the dependence of plant accumulation of PPCP/EDCs on transpiration. In this study, we measured plant accumulation and translocation of 16 PPCP/EDCs, including neutral and ionizable compounds, in 3 plant species grown hydroponically in nutrient solution. Plants were grown in growth chambers with different environment regimes to impose two distinct transpiration patterns. Losses of nutrient solution through transpiration were monitored throughout the 21 d incubation and the levels of PPCP/EDCs in plant tissues were measured at the end of cultivation. The effect of transpiration on bioconcentration or translocation was statistically evaluated for anionic, cationic, and neutral PPCP/EDCs. Knowledge of the interplay between transpiration and plant uptake is useful for identifying types of PPCP/EDCs, as well as weather conditions, that may have a relatively high tendency for plant accumulation and pose potential human health risks.

Three plant species were included in this evaluation. ‘Champion II’ tomato seedlings were purchased from Armstrong Growers and ‘Nevada’ lettuce seedlings were purchased from Do-Right’s Plant Growers at 3 weeks post seeding through a local nursery. ‘Danvers 126’ carrot was started from seed in commercial potting soil and seedlings were used at 26 d post-seeding. Two growth chambers with open circulating air were used in this study. One chamber was programmed to simulate a cool and humid environment with a day time temperature of 17 °C, followed by a night time temperature of 15 °C, while the relative air humidity was kept at 80%. The other growth chamber was programmed to simulate a warm and dry environment with a day time temperature of 27 °C, a night time temperature of 20 °C, with humidity at 50%. The cool-humid and warm-dry environments were used to induce distinctively different plant transpiration patterns. Both chambers received irradiation from a mix of incandescent and fluorescent bulbs, which gradually ramped over 7 h each day to a maximum light intensity of 300 µmol/m2 -sec2 which was maintained for 2 h before decreasing to darkness for a total daily photoperiod of 16 h. Six days before the start of the incubation,hydroponic vertical garden plants were carefully removed from their growth media, rinsed with DI water, inserted through jar lids, fitted with the foam collars, and placed in 2 L glass jars filled with fresh nutrient solution, at one plant per jar. After the plants were transferred to the growth chambers, jars were attached to a small pump system to aerate the solution with ambient air. After 3 d, plants were transferred into clean jars of fresh nutrient solution to replenish nutrients and minimize microbial growth. After a total of 6 d of acclimation, 4 replicates of each plant species in each chamber were randomly selected and transferred into clean jars with 1900 mL of fresh nutrient solution that was amended with 5 mL of a working solution of PPCP/EDCs prepared in ultrapure water. The nominal concentration was 1 µg/L for each compound in the nutrient solution, a level at the higher end of concentration ranges found in treated wastewater effluents . The actual chemical concentration of each compound was measured with solid-phase extraction, as described below. Plants were grown for 21 d in the growth chambers. Every 1 to 3 d, based on the amount of solution transpired, all plants were transferred to clean jars containing fresh solution fortified with PPCP/EDCs. At each solution exchange, the masses of fresh and used solutions from each container were gravimetrically measured to determine the exact amount of solution transpired by each plant. The total transpired mass was defined as the cumulative mass of nutrient solution removed from a jar throughout the 21 d incubation. Evaporation from jars was negligible due to use of fitted lids. The pH in the nutrient solution was measured at that time with pH paper; which was later used to calculate the average log Dow of each compound . At 21 d, all plants were removed from their treatment jars, rinsed with DI water, and separated into different parts.

For lettuce and tomato, plants were divided into leaf, stem, and root tissues. For carrot, plant was separated into leaf and root. Plant tissues were weighed, placed in self-sealing plastic bags, and then stored at -70 °C before analysis. To characterize the depletion of PPCP/EDCs in the nutrient solutions between solution exchange, solution samples were analyzed for levels of PPCP/EDCs on day 8 and 10. On day 8, freshly prepared nutrient solutions were analyzed for the initial chemical concentrations of PPCP/EDCs. To determine the masses of PPCP/EDCs remaining in the solution after 2 d of plant growth, the used nutrient solution from each plant container on day 10 was analyzed. To estimate the potential removal of PPCP/EDCs not attributable to plant uptake, triplicate jars of fortified nutrient solution without plants were included in each growth chamber for 2 d and then similarly analyzed. Prior to analysis, nutrient solution from each container was weighed and mixed by shaking, from which a 275 mL sub-sample was removed. The solution sample was extracted according to a previously published method . Briefly, 100 µL of surrogate solution was added to each sample. A Supelco Visiprep DL solid phase extraction manifold with disposable liners and HLB cartridges were used for extraction. Cartridges were sequentially conditioned with 5 mL each of MTBE, methanol, and water, and samples were loaded at 5 mL/min under vacuum. Sample vessels were rinsed with 200 mL of ultrapure water, and the rinsate was also passed through the cartridge. Sample cartridges were dried with nitrogen gas and then eluted with 5 mL each of 90/10 MTBE/methanol and methanol. The eluent was evaporated under a gentle stream of nitrogen at 40 °C to a volume of 400 µL and then transferred to a 2 mL glass vial. The condensing vessel was rinsed twice with 300 µL of methanol and the rinsate was added to the sample vial to make the final volume to be 1.0 mL for analysis.The extraction of plant tissue samples followed a previously published method . In brief, plant samples were removed from the freezer and immediately placed in a freeze-drier . Samples were dried for 16 h, or to dryness,and then weighed. Each plant sample was then finely ground in a stainless steel coffee grinder. The grinder was cleaned between samples using soap, water, and acetone. A 0.20 g aliquot was placed in a 50 mL polypropylene centrifuge tube and spiked with 100 µL surrogate solution. Samples were sequentially extracted with 20 mL MTBE, and then 20 mL acetonitrile, by sonication in a Fisher Scientific FS110H ultrasonic water bath for 20 min followed by centrifugation at 3000 rpm. The combined supernatant from each extraction step was decanted into a 60 mL glass tube and evaporated at 40 °C under a gentle flow of nitrogen to a volume of 0.5 mL. The residue was re-dissolved in methanol and then mixed in 55 mL ultrapure water. The SPE cartridges were conditioned with 5 mL methanol and then 5 mL water. Samples were passed through cartridges at 5 mL/min under vacuum, and then sample tubes were rinsed with 30 mL of ultrapure water, which was also passed through the cartridge. Sample cartridges were dried with nitrogen gas and then eluted with 7 mL methanol. The eluent was evaporated under a gentle stream of nitrogen at 40 °C to a volume of 200 µL and then transferred to a 2 mL glass vial. The condensing vessel was rinsed twice with 150 µL of methanol and the rinsate was added to the sample in the vial to create a final volume of 0.5 mL.