*3.2. nCeO2 Plant Internalization*

The early step of our study was devoted to verifying the entry of *n*CeO2 into plant tissues. Control and treated seedlings of *H. lanatus*, *L. flos-cuculi* and *D. tenuifolia* were subjected to the extraction procedure and further analyzed by sp–ICP–MS. Size distributions of ceria nanoparticles in stock solutions and in the seedlings treated with 20 mg L−<sup>1</sup> *n*CeO2 25 nm and 50 nm are reported in Figure 2.

**Figure 2.** Particle size distributions of, respectively, *n*CeO2 25 nm and *n*CeO2 50 nm stock solutions (open bars) and after enzyme treatment of seedlings of *H. lanatus*, *L. flos-cuculi* and *D. tenuifolia* (closed bars) treated with 20 mg L−<sup>1</sup> of *n*CeO2.

As expected, in control seedlings, *n*CeO2 was not detected, whereas in all treated species, (i) the presence of internalized *n*CeO2 was verified, and (ii) the *n*CeO2 have a different size distribution than stock solution suggesting aggregation phenomena between nanoparticles (Figure 2 and Table 2). Data from sp–ICP–MS analysis confirmed that *n*CeO2 underwent agglomeration. The increase of the median diameter of *n*CeO2 was evident for seedlings treated with *n*CeO2 25 nm, being 41.7 nm the average size of the nanoparticles extracted from seedlings (41 nm in *H. lanatus* and *L. flos-cuculi*, and 43 nm in *D. tenuifolia*). The mean size of particles extracted from seedlings treated with *n*CeO2 50 nm was 47 nm so in good agreemen<sup>t</sup> with the treatment (Figure 2 and Table 2).


**Table 2.** Most frequent size, mean size and number of peaks determined by sp–ICP–MS analysis after enzyme treatment of seedlings of *H. lanatus*, *L. flos-cuculi* and *D. tenuifolia* treated with 20 mg L−<sup>1</sup> of *n*CeO2 25 nm and 50 nm.

The sp–ICP–MS results also show that the most frequent size of nanoparticles taken up by plants is similar for monocotyledons and dicotyledons for *n*CeO2 50 nm, whereas, for 25 nm treatments, the most frequent diameter is smaller in *L. flos-cuculi* and *D. tenuifolia* (respectively 31 and 35 nm) than *H. lanatus* (40 nm) (Figure 2 and Table 2). Regarding this aspect, some authors evidenced a size-dependent uptake and translocation of *n*CeO2 in plants. In particular, *n*CeO2 having a diameter smaller than 50 nm are present in all plant tissues and pass from roots to the aerial parts without dissolution and transformation.

### *3.3. Seed Germination and Root Length*

A three-way ANOVA was run in order to have a general view regarding the effects of plant species, *n*CeO2 size and Ce concentration on (i) percentage of germination, (ii) root length and (iii) Ce concentration in plant tissues. There were significant three-way interactions for percentage of germination (*p* = 0.0463 \*) and root length (*p* = 0.0000 \*\*\*) (Table 3). Subsequently, the statistical analysis with two-way ANOVA within the species continued.

**Table 3.** Three-way ANOVA *p* values for the main effects of plant species, *n*CeO2 size and Ce concentrations and their interactions on the percentage of germination, seedling root length and Ce concentration in seedling of *H. lanatus*, *L. flos-cuculi* and *D. tenuifolia*.


ns: not significant at *p* ≤ 0.05; \*, \*\* and \*\*\* indicate significance at *p* ≤ 0.05, *p* ≤ 0.01 and *p* ≤ 0.001.

As shown in Figure 3, treatments improve the germination percentage in all the three species if compared with controls. Indeed, germination increases more than 20% in several treatments in *H. lanatus*, 15% in *L. flos-cuculi* and 10% in *D. tenuifolia*, with respect to the control.

The evaluation of the effects induced by *n*CeO2 of different sizes is the main objective of this study. Actually, we have not verified a clear trend related to the *n*CeO2 size. In fact, the nanoparticle dimensions had no influence on the germination of *L. flos-cuculi*, while in both the other species, in some cases, they did. We observed that the germination percentage is higher for seeds treated with *n*CeO2 25 nm (about +10%) compared with 50 nm. (Figure 3). A statistically significant difference was found in *H. lanatus* at 2 and 200 mg L−<sup>1</sup> (respectively, *p* = 0.0282 \* and *p* = 0.0132 \*) and *D. tenuifolia* at the at 0.2 and 2 mg L−<sup>1</sup> (respectively, *p* = 0.0072 \*\* and *p* = 0.0249 \*) (Figure 3). It is quite likely that in this species, the influence of the size of *n*CeO2 on germination could have been observed even at the highest concentration, but the high variability of the data influenced the response of the statistics (*p* = 0.0866) (Figure 3).

Previously we demonstrated some relationships between the germination process and *n*CeO2 size. Similar observations were carried out on the seedling root length (Figure 4).

**Figure 3.** Percentage of seed germination in *H. lanatus*, *L. flos-cuculi* and *D. tenuifolia*, grown in Petri dishes and treated with solutions of *n*CeO2 25 nm and *n*CeO2 50 nm at 0, 2, 20, 200 and 2000 mg <sup>L</sup>−1, respectively. The values are mean ± SD (standard deviation) of 3 replicates. Statistical significance of the treatments for each Ce concentration is reported: (i) figure upper part→comparison between *n*CeO2 25 nm and 50 nm; (ii) figure lower part→comparison between all treatments. Different letters indicate statistical differences. ns = not significant, \* *p* ≤ 0.05, \*\* *p* ≤ 0.01.

In this case, the results demonstrate that, regardless of the Ce concentration, root length was not influenced by the *n*CeO2 size. However, treatments stimulate the root growth in all three species, with a clear increase of length, in particular in *L. flos-cuculi* and *D. tenuifo*lia, if compared with control seedlings (Figure 4). Some differences in response to treatments are species-specific. The *n*CeO2 of both sizes do not have any stimulating effect on the length of the roots of *H. lanatus*, which resulted in insensitivity to the treatments even at higher concentrations. On the other hand, we observed an increase in root length in treated seedlings of the other species. This effect was particularly intense in *L. flos-cuculi,* where the length of the roots of treated seedlings on average has almost doubled (+90.1%) compared to the control (14.7 mm and 7.73 mm, respectively). The stimulating effect of *n*CeO2 demonstrated in *D. tenuifolia* is much less powerful but remarkable, where we found a 34% increase in root length compared to the control (22.8 mm and 16.9 mm, respectively).

**Figure 4.** Root length in seedlings of *H. lanatus*, *L. flos-cuculi* and *D. tenuifolia*, grown in Petri dishes and treated with solutions of *n*CeO2 25 nm and *n*CeO2 50 nm at 0, 2, 20, 200 and 2000 mg <sup>L</sup>−1, respectively. The values are mean ± SD (standard deviation) of 3 replicates. Statistical significance of the treatments for each Ce concentration is reported: (i) figure upper part→comparison between *n*CeO2 25 nm and 50 nm; (ii) figure lower part→comparison between all treatments. Different letters indicate statistical differences. ns = not significant, \* *p* ≤ 0.05, \*\* *p* ≤ 0.01.

### *3.4. Ce Concentration in Plant Seedlings*

To quantify the total content of Ce that was taken up by seedlings in the three plant species, we used an ICP–MS after the acid digestion of the samples. The elaborated data with the total concentration of Ce are presented in Figure 5.

**Figure 5.** Ce concentration in seedlings of *H. lanatus*, *L. flos-cuculi* and *D. tenuifolia*, grown in Petri dishes and treated with solutions of *n*CeO2 25 nm and *n*CeO2 50 nm at 20, 200 and 2000 mg <sup>L</sup>−1, respectively. The values are mean ± SD (standard deviation) of 3 replicates. Statistical significance of the treatments (ANOVA *p* value) for each Ce concentration is reported. Different letters indicate statistical differences (*p* ≤ 0.05).

The concentration of total Ce in seedling tissues of *H. lanatus*, *L. flos-cuculi* and *D. tenuifolia* shows a different magnitude of accumulation according to the treatments. In fact, a statistically significant effect of treatments (*p* < 0.05) was verified for all the species. As already reported in Table 2, the interaction "species x Ce concentration" was highly statistically significant (*p* = 0.0000 \*\*\*). With regard to the *n*CeO2 size, we observed that the seedlings treated with the smaller *n*CeO2 reveal a higher concentration of Ce in their tissues than the ones treated with the 50 nm nanoparticles. This occurred in particular in *L. flos-cuculi* and *D. tenuifolia* and at the two highest concentrations of treatments, but not in *H. lanatus*. In *L. flos-cuculi*, the total content of Ce corresponds to 165 and 128 mg kg−<sup>1</sup> DW at 200 mg <sup>L</sup>−1; 1616 and 1151 mg kg−<sup>1</sup> DW at 2000 mg L−<sup>1</sup> (*p* = 0.0134 \*), respectively at 25 nm and 50 nm. In *D. tenuifolia* we detected 189 and 114 mg kg−<sup>1</sup> DW at 200 mg <sup>L</sup>−1; 1841 and 1305 mg kg−<sup>1</sup> DW at 2000 mg L−<sup>1</sup> (*p* = 0.0465 \*), respectively at 25 nm and 50 nm (Figure 5).

The size of the nanoparticles in our study was 25 nm and 50 nm. By looking at Figure 2, it turns out that the stock solutions of the two nominal sizes are actually a mixture of nanoparticles of different dimensions, with 25 nm and 50 nm being the dimensions among the ones having the highest frequencies. This makes one conclude that both the dispersions contain nanoparticles that can potentially enter the plant roots. There are two main factors, among others, that can influence the uptake and translocation of the nanoparticles: (i) the size of the pores in the cell membrane; (ii) the tendency of the nanoparticles

to aggregate due to chemical interactions. Given the fact that the dimension of the nanoparticles at a given shape determines the surface to volume ratio, this can affect the entity of such aggregation (the smaller the dimension, the more likely the aggregation).

Tables S1 and S2 report the calculations of the *n*CeO2 25 nm and 50 nm ratios based on the assumptions that all the nanoparticles are spherical and equal in size (25 nm or 50 nm) in the dispersions as well as inside the seedlings and no aggregations occur. The theoretical ratio (Table S1) refers to an ideal scenario for which an equal mass of nanoparticles is taken up by both the plants exposed to the *n*CeO2 25 nm and the *n*CeO2 50 nm; the theoretical ratio is hence obtained by calculating the number of nanoparticles at a given size (25 nm or 50 nm) as the mathematical division of the total mass of Ce in the plant by the mass of a single nanoparticle. The observed ratio is calculated following the same procedure and assumptions, but considering the experimental mean Ce mass measured in planta for the different treatments (Table S2).
