Natural metal oxides such as clays are known to strongly and preferentially sorb phosphate over other organic and inorganic ligands,and research has shown that metal oxide ENMs can also sorb phosphorus and thereby potentially affect its bioavailability in soils and other environmental media.Fourth, we hypothesized that higher light and lower nutrient conditions would be more physiologically stressful forClarkia plants, and that highly stressed plants would be most vulnerable to ENM toxicity. Additionally, since TiO2 and CeO2 are photoactive and produce ROS when exposed to light,we predicted that they would have the greatest effect in plants grown under high illumination by interfering with photosynthesis in leaves.Metals from ENMs were taken up into all tissues in all treatments, although the amounts depended on ENM type, soil ENM concentrations, growth condition , high light limited nutrient , low light excess nutrient , and low light limited nutrient, and tissue type. Mean tissue metal concentrations can be seen in Figure 1, and results from multiple regressions can be seen in Figures 2 4 and Table S1. In general, Ce and Ti were found in highest concentration in roots , while Cu was primarily found in leaves , although relatively high concentrations of Ti were also seen in stems . Background concentrations of Ti and Cu were found in all three tissues, while background Ce was only found in roots. Among plants in the Control group , it is likely that Ce was not found in stems or leaves because it was not present in the soil at concentrations as high as Ti , nor is it an essential micro-nutrient as is Cu. Of the three ENMs to which plants were exposed, those exposed to CeO2 and TiO2 followed the pattern of distribution described in our first hypothesis, with concentrations being consistently highest in the roots followed by leaves then stems . In Cu2-exposed plants, however, Cu concentrations were roughly an order of magnitude higher in leaves than in roots . Plants from all groups showed statistically significant positive correlations between exposure concentration and metal concentration in roots and,growing hydroponically with a few exceptions, tended to have the highest metal concentrations at the highest exposure level in all tissues. The most notable exceptions to this trend are the variable Ce and Ti content of leaves from plants grown under high light, excess nutrient and high light, limited nutrient conditions .

This reflects the high inter-leaf metal content variability for Ce and Ti and may be due to a randomized or patchy accumulation of these nanoparticles between leaves. There were no significant associations between leaf metal content and leaf node number, which is analogous to order of production . Since C. unguiculata leaves are produced in a temporal sequence along the height of the plant and are also larger lower on the plant, this indicates that ENM uptake into leaves was independent of both stage of growth and leaf size. Growth conditions also played a role in ENM uptake, with plants grown under high light accumulating more Ce and Ti in their leaves than those grown in low light and HL leaves accumulating more Ti than HE . Along with the increased transpiration rates seen in plants grown under high light , these findings validate our second hypothesis that high light plants would exhibit elevated uptake of ENMs to leaves due to increased transpiration. However, increased uptake of Cu into leaves and roots was found under low light conditions . These differences among ENM types in uptake and distribution are also likely to be due to differences in particle characteristics, particularly morphology and surface charge. The CeO2 ENMs we used had a moderately high aspect ratio and thus had a smaller minimum dimension, which may allow them to pass through narrow vascular tissues in the stem more easily than the spherical TiO2. Due to this physical size limitation, TiO2 may also aggregate in the conductive tissues of the stems at higher concentrations, causing the buildup seen in Figure 1.Table S2 shows that all three ENMs used here had a weak negative charge in potting soil pore solution, although this was likely due to the high ionic strength and organic content of this soil shielding the particle surfaces and not a result of a direct alteration of the ENM crystal surface. Wang et al.and Zhu et al.found that under hydroponic conditions, well-dispersed particles coated with positively charged polymers are more readily taken up into plant roots compared to those coated with negatively charged polymers , which had higher accumulation in leaves. The results seen here provide confirmation of the importance of surface charge in ENM uptake and distribution in plants under more environmentally relevant conditions, i.e., in soil and with poly disperse ENMs. In addition to its surface charge, the tendency of Cu2 to dissolve at low pH,such as is found in the soil used in this study , likely also contributes to its uptake behavior. Rhizosphere pH tends to be more acidic than the surrounding soil due to the release of protons by roots to stimulate and counterbalance the uptake of ions from the soil; one effect of this acidity may be to dissolve a portion of the Cu2.

Dissolved Cu would, in turn, encounter less size exclusion than ENMs and be retained less in the roots and stems in addition to being actively transported to the leaves, consequently making Cu uptake and translocation less dependent on plant transpiration than CeO2 or TiO2. Although Cu is an essential component of several enzymes and other compounds in chloroplasts and mitochondria,it can be toxic at higher concentrations.34 Last, although we predicted that P would be correlated with metal content in tissues due to physiochemical sorption of phosphate to the ENMs, it was only in root tissue of HL plants exposed to CeO2 ENMs that we found a relationship. At root Ce concentrations below 100 μg g 1 , P was positively associated with Ce , but this trend plateaued at higher concentrations. One possible explanation for this is that CeO2 ENMs adsorbed P from the soil and were then sorbed into/onto the plant roots, but at higher exposure concentrations, the soil was depleted of readily available P for the ENMs to adsorb. Previous studies using hydroponic systems have shown increased P uptake in maize exposed to ZnO ENMs and in spinach exposed to nZVI,although these results were due to the uptake of dissolved metal/phosphate complexes rather than ENM-sorbed P. Rui et al.observed the partial transformation of CeO2 ENMs into particulate CePO4 that were then taken up into hydroponically grown cucumber seedlings, although the general lack of correlation between tissue Ce and P concentrations suggests this process was not occurring to a significant extent in this study. Overall, our results did not indicate a significant relationship for C. unguiculata between ENM exposure and P bio-availability.We found that the physiological effects of ENM exposure on our test plants were strongly dependent on the environmental conditions under which plants were grown, namely, high light excess nutrient , high light limited nutrient , low light excess nutrient , and low light limited nutrient . By comparing photosyntheticrates and other physiological parameters of the zero concentration groups across growth conditions, we can establish baseline levels of stress for each condition, which can be used to explain the trends seen in ENM-exposed plants. On the basis of the responses to growing conditions of A, transpiration rate , intracellular CO2 , and quantum yield of CO2 assimilation in zero concentration groups , the relative rankings from most to least stressful growth condition appear to be HL > HE > LL ≈ LE. This ranking aligns with our hypothesis that higher light and lower nutrient conditions are the most stressful conditions imposed in this experiment. For plants exposed to these ENMs, few significant correlations between the physiological parameters measured and ENM exposure concentration were seen at the second or sixth week of exposure, and by the eighth week, all high light plants had reached the end of their life cycle and ceased photosynthesizing. However, at the fourth week of exposure we found that in HE plants exposed to CeO2 and TiO2, A and ΦCO2 decreased significantly and Ci increased significantly with increasing exposure concentration .

This supports our final hypothesis and indicates that these two photoactive ENMs reduce photosynthetic rate by interfering with the assimilation of CO2 required for photosynthesis,growing strawberries hydroponically which results in a buildup of CO2 within leaf cells. Additionally, there were no changes in ΦPSII in these plants, and this lack of correlation between ΦPSII and ΦCO2 could indicate that energy transfer from photosystem II to the Calvin cycle is being disrupted by the ENMs. This effect appears to be light-driven since no impacts of exposure concentration on any physiological parameters were seen in low light plants. High light conditions had the 2-fold impact of increasing particle uptake to leaves by increasing transpiration rates and possibly stimulating greater photo activity of TiO2 and CeO2. The disruption of energy transfer observed may be due to the absorption of electrons from photo system II by the ENM upon the creation of an e /hþ pair after excitation by a photon, or alternately through reactions with ROS produced by the ENM. Exposure to CeO2 had slightly weaker effects on physiological parameters than TiO2, and if the latter scenario is correct, this could be due to the lower relative ROS production rate of CeO2 compared to TiO2. Barhoumi et al. saw an inhibition of PSII and a corresponding increase in ROS inLemna gibba exposed to iron oxide ENMs, so similar phenomena may be occurring here. ROS production by TiO2 and CeO2 ENMs may also explain why no physiological effects were seen in HL plants, since plants upregulate antioxidant production at higher stress levels40 that may counteract ROS produced by these ENMs. Additionally, interference with photosynthetic mechanisms implies that CeO2 and TiO2 ENMs are able to penetrate or be actively transported not only into the leaf cells but also into the chloroplasts as well, and are able to intercalate themselves between thylakoid stacks to intercept electrons from PSII. Given that interthylakoid gaps can be on the order of 50 250 nm,individual particles or small aggregates would not necessarily be excluded based on size alone. White side et al.found uptake of NH2-coated quantum dots <15 nm in diameter into bluegrass chloroplasts, so it is plausible that at least primary particles of TiO2 and CeO2 were able to enter the chloroplasts of our model plant. Both of these ENMs have limited dissolution and have been shown to be taken up into plant tissues as nanoparticles,making it unlikely that any effects on photosynthesis are due to ionic Ti or Ce. A similar decrease in A was seen in HL plants after 4 weeks of exposure to Cu2, but without a corresponding change in ΦCO2 or Ci . By further decreasing the already low photosynthetic rate of HL plants, Cu2 had a larger relative impact than in HE, LE, or LL plants. This suggests that Cu2 may affect photosynthesis through a different mechanism than TiO2 and CeO2. Additionally, we found that the fraction of oxidized PSII reaction centers increased significantly with increasing exposure concentration . In healthy plants, qL is typically positively associated with photosynthetic production,38 but since we found a negative correlation between Cu2 exposure and photosynthesis , the increases in qL we observed were likely due to interference with the oxidation of the primary PSII quinine acceptor by light rather than increased photosynthetic efficiency. Others have found similar oxidation of PSII reaction centers in plants exposed to ionic copper due to interference with the photon antennae of PSII,which may indicate Cu2 toxicity seen in this study is due to Cu ions released from the Cu2 ENMs. In our system, Cu2 could be dissolved either in the rhizosphere and taken up as ionic Cu or be taken up into the plant in particle form and dissolve within the plant tissues. However, since these Cu2 particles has been shown to have increased dissolution at acidic pH and lower dissolution at basic pH,the majority of dissolution probably occurs in the soil rather than in the neutral or slightly basic conditions of cell or chloroplast interiors.Linear growth rates , maximum height, leaf production rate, leaf loss rate , maximum number of leaves, and week of maximum leaf production were calculated from physical measurements and are shown in Figures S1 S6. Few effects due to ENM exposure were seen under any growth condition, although LE plants exposed to Cu- 2 had reduced growth rates, leaf production rates, and maximum number of leaves with increasing exposure concentrations . Cu is an essential plant micro-nutrient but at high concentrations such as those observed in this experiment, Cu can decrease the uptake of other nutrients from the soil and disrupt nitrogen metabolism.Nutrient limitation caused by the presence of Cu2 may have been responsible for limiting the growth of LE plants . The lack of a growth response in HE plants exposed to CeO2 and TiO2 may be because, under high light conditions, reductions in CO2 assimilation have been shown to have minimal impacts on C gain.