Evaluation of Apical and Molecular Effects of Algae Pseudokirchneriella subcapitata to Cerium Oxide Nanoparticles

Cerium oxide engineered nanoparticles (nCeO2) are widely used in various applications and are, also, increasingly being detected in different environmental matrixes. However, their impacts on the aquatic environment remain poorly quantified. Hence, there is a need to investigate their effects on non-target aquatic organisms. Here, we evaluated the cytotoxic and genotoxic effects of <25 nm uncoated-nCeO2 on algae Pseudokirchneriella subcapitata. Apical (growth and chlorophyll a (Chl a) content) and genotoxic effects were investigated at 62.5–1000 µg/L after 72 and 168 h. Results demonstrated that nCeO2 induced significant growth inhibition after 72 h and promotion post 96–168 h. Conversely, nCeO2 induced enhanced Chl a content post 72 h, but no significant changes were observed between nCeO2–exposed and control samples after 168 h. Hence, the results indicate P. subcapitata photosynthetic system recovery ability to nCeO2 effects under chronic-exposure conditions. RAPD-PCR profiles showed the appearance and/or disappearance of normal bands relative to controls; indicative of DNA damage and/or DNA mutation. Unlike cell recovery observed post 96 h, DNA damage persisted over 168 h. Thus, sub-lethal nCeO2-induced toxicological effects may pose a more serious threat to algae than at present anticipated.


Introduction
Cerium oxide engineered nanoparticles (nCeO 2 ) are a class of emerging contaminants with unique properties when compared to their bulk counterpart. These properties include redox activity, scavenging of free radicals, and inhibition of biofilm formation [1]. As a result, nCeO 2 find widespread applications, e.g., in sunscreens as UV absorbent [2], fuel additives [3], catalysis [4,5], biomedicine [6], and nano-pharmacy [7]. Estimates indicate that nCeO 2 global production to be 100 and 1000 tons/year [8] and may have increased to 7500-10,000 tons/year [9,10]. Since nCeO 2 is widely used as an additive for diesel fuels, for instance, it is, therefore, among likely engineered nanoparticles (ENPs) of ecological concern in the natural environment [11]. In addition, nCeO 2 has been quantified in various environmental matrixes using modelling (e.g., 0.1 µg/L in surface waters [12] and experimental (0.4-5.2 ng/L in surface waters [13]) approaches. However, the environmental implications of nCeO 2 remain poorly understood. Yet, it is among the top ten priority ENPs identified both for the evaluation of human health and environmental safety effects [14].
Following the release of nCeO 2 into the ecosystems, they may accumulate in the aquatic systems linked to their transformation processes including aggregation, sedimentation, and low degradation rate. This, in turn, increases their uptake, accumulation, and bio-magnification in the food chain [15] and inevitable interactions with different classes of aquatic organisms [16]. Therefore, understanding the impact of nCeO 2 on aquatic biota especially organisms that represent the base of the trophic chain is of ecological significance.
Algae are unicellular organisms and primary producers that play a vital role in the structure and functioning of ecosystems and are susceptible test species to environmental herein we investigated the impact of nCeO 2 on freshwater microalgae P. subcapitata at low exposure concentrations in the µg/L range as their use increases; thus, they are likely to be found in the actual environment. The study specific objectives were to assess the effects of nCeO 2 (at concentrations of 62.5-1000 µg/L) on P. subcapitata: (i) at apical endpoints including growth effect and Chl a content and (ii) on DNA integrity using RAPD-PCR analysis in 10% Blue Green algae-11 (BG-11) medium after 72 and 168 h. The toxicological effects observed from the molecular and apical endpoints assessments were compared or linked to gain better understanding on the effects of nCeO 2 on algae under chronic exposure conditions.

Characterization of nCeO 2
Uncoated nCeO 2 in dispersion (<25 nm particle size, 10 wt.% in H 2 O) were purchased from Sigma Aldrich (Johannesburg, South Africa). Size and morphology were characterized as previously described by Mahaye [36]. Hydrodynamic diameter (HDD) and zeta (ζ) potential of nCeO 2 in de-ionized water (DI water) (15 MΩ/cm) and 10% Blue Green algae-11 media (herein referred to as BG-11 media) [37] were measured using dynamic light scattering (DLS) (Malvern Zetasizer Nano ZS, Malvern, UK). Measurements for the ζ-potential and HDD were taken at 0, 2, 6, 24, 48, and 72 h in triplicate. The HDD and ζ-potential measurements in DI water and BG-11 media were only done at 1000 µg/L nCeO 2. This is because nCeO 2 at concentrations <1000 µg/L were below the detection limit using the Zetasizer.

Preparation of Exposure Media and Concentrations
Algal experiments were conducted in BG-11 media (media preparation and composition details are listed in Section SI-1 and Table S1, respectively, in the supporting information) following Direct Estimation of Ecological Effects Potential (DEEEP) toxicity testing protocols [37]. The media was stored at 4 • C under dark conditions before use. The nCeO 2 exposure concentrations of 62.5, 125, 250, 500, and 1000 µg/L were prepared in BG-11 media in triplicate and ultra-sonicated for 30 min before carrying out the exposure experiments.

Test Organisms
The Algaltoxkit F TM kit (MicroBioTests Inc., Gent, Belgium) was purchased from Tox-Solutions (Johannesburg, South Africa). Algaltoxkit F contains all the materials needed to perform a 72 h growth inhibition tests with the freshwater microalgae P. subcapitata (former names: Selenastrum capricornutum and Raphidocelis subcapitata). This algal toxicity test was carried strictly in adherence to ISO Standard 8692 and OECD Guideline 201 protocols. The de-immobilization of P. subcapitata from the beads was performed in accordance with the manufacturer's instructions. In this study, stock cultures were incubated under controlled conditions (temperature: 25 ± 1 • C; light intensity: 6000 Lux; 12:12 h light: dark cycle and shaken continuously at 100 rpm) for 5-7 d to obtain exponentially growing P. subcapitata.

Cytotoxic Effects of nCeO 2 on P. subcapitata
Preparation of the algal test was performed as outlined in Mahaye et al. [27]. Reference tests aimed to ascertain the sensitivity of algae growth are described in Section SI-2. Inoculum was prepared by harvesting exponentially growing P. subcapitata cells. Cells from a 5-7 d old stock culture prepared in Section 2.3 were transferred as 1 mL volume into Eppendorf tubes and centrifuged at 10,000 rpm for 10 min. The supernatant was decanted, and the algal cells were re-suspended in 0.1 mL phosphate-buffered saline (PBS). The centrifugation and decanting steps were repeated twice. The volume of stock culture required and the cell density of algal inoculum required per experiment in test and control wells were calculated using the following expression: no. o f f lasks used × vol f lask × 200000 cells/mL Cell density (cells/mL) in the stock culture (1) where vol/flask is the volume of test solution per flask and cells/mL is the cell density in the inoculum given by the following expression [38]: Cells/mL = e lnλ 684 +16.439 1.0219 (2) where λ 684 is the optical density (OD) at 684 nm.
In the final step, algal cells were re-suspended and mixed well in 10% BG-11 media and the cell density in the inoculum was measured before the experiment was initiated. For each test, 200,000 cells/mL sample was required. Tests were carried out in 2 mL volumes in 24well microplates with 1.8 mL test sample (or de-ionised water for the control), with 0.2 mL of the inoculum and algal medium. Thereafter, it was incubated at the same conditions as the stock culture for 168 h. P. subcapitata exposures to nCeO 2 were conducted following the standard algal test of 72 [37], or 96 h [39] with slight modifications. First, the exposure time was increased from 96 to 168 h to gain insights on likely effects under chronic conditions. The standard US EPA flask test method [39] yielded inadequate biomass for genotoxicity analysis as it requires only 10,000 cells/mL as initial inoculum. Thus, to generate sufficient biomass for genotoxicity analysis, we used the DEEEP toxicity testing protocol [37]. This protocol requires an inoculum of 200,000 cells/mL, which, in turn, generated adequate biomass for DNA damage analysis [27].
For negative controls, exposures for algae were done without nCeO 2 . All experiments were done in triplicate. Exponentially growing P. subcapitata were exposed to five concentrations of nCeO 2 (62.5, 125, 250, 500 and 1000 µg/L) for 168 h, in a 24-well microplate system, under defined conditions outlined in Section 2.3. Exposure concentrations were selected based on the detected or predicted environmental concentrations from the previous studies. For example, nCeO 2 concentrations of 0.3-230 µg/L [40] and 0.04-0.27 µg/L [41] were reported in freshwater. The ENPs' concentrations in freshwater are predicted to reach six-fold higher by the year 2050 [42]. Thus, in this study, the selected exposure concentrations cover both current and plausible future predicted concentrations of nCeO 2 in freshwater systems.
After the experiments were initiated, the cell density (in the form of optical density) was measured at 684 nm every 24 h for 168 h using a microplate reader (FLUOstar Omega BMG Labtech, Ortenberg, Germany). Briefly, the wavelength of 684 nm used here was adopted from Rodrigues et al. [38] and has been successfully used on ENPs-exposed P. subcapitata studies [27,43,44]. After 72 and 168 h exposure periods, Chl a content was determined following a protocol by Harris [45]. Briefly, 1 mL of the control and exposed algal cells were centrifuged for 10 min at 13,000 rpm, and the pellet was washed using DI water. The algal cells pellet was suspended in 95% ethanol, vortexed for 2 min, kept at 4 • C for 30 min and centrifuged at 13,000 rpm for 2 min. The supernatant was analysed for Chl a content using a UV-Vis spectrophotometer (HACH, Loveland, CO, USA) at wavelengths of 665 and 649 nm. The content of Chl a was then calculated using the expression: Chl a = 13.70A 665 − 5.76A 649 (3) where A 665 and A 649 are the OD values (n =3) at wavelengths of 665 nm and 649 nm, respectively.

DNA Damage and Estimation of Genomic Template Stability
DNA isolation, visualization, and amplification were done as described in Mahaye et al. [27], and details are set out in Section SI-3. Briefly, exponentially growing P. subcapitata were exposed to three nCeO 2 concentrations at 62.5, 250, and 1000 µg/L as described in Section 2.3. RAPD-PCR data analysis was performed by comparing the PCR product profiles for nCeO 2-treated sample with the control samples. The genomic template stability percentage (GTS%) was calculated using the following Equation [46]: where a is the average number of RAPD polymorphic bands detected in ENPs-treated samples and n is the total bands in the controls. Polymorphisms in RAPD profiles include deletion of a normal band and induction of a new band in comparison to the control RAPD profiles. GTS percentage of nCeO 2 -treated samples was calculated and changes of genomic stability were expressed as a percentage of controls set at 100%.

Data Analysis
All measurements were performed in triplicate, and the results were expressed as mean ± standard deviation (SD). Statistical analysis was performed using GraphPad Prism Software version 9.3.0 (GraphPad Software, San Diego, CA, USA). One-way analysis of variance (ANOVA) followed by Dunnett's post hoc test was used to evaluate statistical differences between nCeO 2 -exposed samples and the controls. Differences between samples were considered statistically significant when p ≤ 0.05.

Characterization of nCeO 2
nCeO 2 had non-uniform triangular, tetrahedral, and hexagonal shapes (Figure S1a), with diameters of 15-50 nm due to the asymmetry of the morphology. Although most nCeO 2 were <25 nm, larger and compact crystalline structures were also observed ( Figure S1b). Figure S1c depicts the particle size distribution of nCeO 2 at 1000 µg/L in BG-11 media measured using DLS. The presence of agglomerates >25 nm showed that the primary particle sizes could not be attained even after ultrasonication, as previously documented in other works [47][48][49][50]. nCeO 2 aggregated immediately following introduction into both DI water and BG-11 media ( Figure 1A). After 24-72 h exposure, aggregation was higher in BG-11 media compared to DI water ( Figure 1A). The higher aggregation was likely due to high ionic strength of BG-11 media relative to DI water as previously documented [51,52].
In an earlier work, uncoated nCeO 2 sized 28 nm was observed to form aggregates of 200-300 nm in ultrapure water [49]. Here, the observed aggregation in BG-11 media show a good agreement with the behaviour of uncoated-nCeO 2 as documented in other algal ecotoxicity media, e.g., synthetic freshwater algal media [31,53], OECD TG 201 [54] and Dutch Standard (DS) medium [55]. For example, 20 nm uncoated-nCeO 2 immediately agglomerated to 218 nm in DS medium, which is ten-fold higher than the primary size [55]. Negative ζ-potential values for 1000 µg/L nCeO 2 were observed in both media types over 72 h and at a narrow range of −8 to −16 mV ( Figure 1B). These low ζ-potential values indicate nCeO 2 instability and are consistent with the rapid agglomeration observed in both DI water and BG-11 media ( Figure 1A). This is because ζ-potential values should be ±30 mV to stabilize ENPs suspensions [56,57]. ζ potential at 1 000 µ g/L of nCeO2 in DI water and BG−11 media measured using DLS over 72 h. Data are presented as mean ± standard deviation (SD). Differen symbols denote significant differences (p < 0.05) between DI water and BG-11 media per time period analysed using Two-way ANOVA with Tukey's multiple comparisons test.
In an earlier work, uncoated nCeO2 sized 28 nm was observed to form aggregates o 200-300 nm in ultrapure water [49]. Here, the observed aggregation in BG-11 media show a good agreement with the behaviour of uncoated-nCeO2 as documented in other alga ecotoxicity media, e.g., synthetic freshwater algal media [31,53], OECD TG 201 [54] and Dutch Standard (DS) medium [55]. For example, 20 nm uncoated-nCeO2 immediately ag glomerated to 218 nm in DS medium, which is ten-fold higher than the primary size [55] Negative ζ-potential values for 1 000 µg/L nCeO2 were observed in both media types ove 72 h and at a narrow range of −8 to −16 mV (Figure 1b). These low ζ-potential values indi cate nCeO2 instability and are consistent with the rapid agglomeration observed in both DI water and BG-11 media (Figure 1a). This is because ζ-potential values should be ±30 mV to stabilize ENPs suspensions [56,57].

Effect of nCeO2 on Algal Growth
A positive control was performed using potassium dichromate (K2Cr2O7) as a refer ence toxicant, and results are presented in Figure S2. Results in Figure 2 demonstrates the growth effect of nCeO2 on algae over 168 h. Remarkably, nCeO2 had no significant effec on algal growth following exposure after 24 and 48 h but induced significant growth in hibition after 96 h relative to the controls. Growth promotion was, however, observed af ζ potential at 1000 µg/L of nCeO 2 in DI water and BG−11 media measured using DLS over 72 h. Data are presented as mean ± standard deviation (SD). Different symbols denote significant differences (p < 0.05) between DI water and BG-11 media per time period analysed using Two-way ANOVA with Tukey's multiple comparisons test.

Effect of nCeO 2 on Algal Growth
A positive control was performed using potassium dichromate (K 2 Cr 2 O 7 ) as a reference toxicant, and results are presented in Figure S2. Results in Figure 2 demonstrates the growth effect of nCeO 2 on algae over 168 h. Remarkably, nCeO 2 had no significant effect on algal growth following exposure after 24 and 48 h but induced significant growth inhibition after 96 h relative to the controls. Growth promotion was, however, observed after 96 h and significantly higher after 144 and 168 h, compared to the controls. Any modification of algae growth may, subsequently, affect higher trophic levels [58]. For instance, they may lead to altered species composition and habitat structure [59] and, as a result, compromise ecological integrity. Among ecological functioning, aspects that may be adversely affected include the extinction of sensitive algal species and macrophytes, or higher growth may outcompete other biological life forms with consequent undesirable perturbations on the food chain and nutrient recycling, among others. 82.7%, respectively, in natural oligotrophic seawater and nutrient-enriched media. The observed cell decline was attributed to the extensive aggregation behaviour of nCeO2 upon entry into natural seawater and hetero-aggregation with algae [60]. In addition, direct contact of nCeO2 with algae was reported previously to be responsible for toxicity and to cause membrane damage of P. subcapitata [53]. Further, an increase in intracellular reactive oxygen species (ROS) was observed in algae [19]. Intracellular ROS plays a role in the inhibition of photosynthesis and can indicate oxidative damage [61]. Previous studies have demonstrated the absence of uptake of uncoated and agglomerated nCeO2 in algae [31,61]. The lack of ENPs uptake was linked to the formation of agglomerates that exceeded the pore sizes (ranges between 5 and 20 nm) of the algal cell wall [62], which, in turn, impeded plausible uptake by algae. Here, the observed agglomerates (up to 918 ± 74 nm) exceeded the algal cell wall pore sizes and, therefore, uptake was reasonably unlikely. Thus, growth inhibition observed was possibly due to the entrapment of algal cells by ENPs agglomerates. As a result, this may have reduced the light and nutrients' availability to the entrapped algal cells with concomitant growth inhibition. Previously, algal growth inhibition was observed to result from physical removal due to co-aggregation and co-sedimentation with nCeO2, as opposed to the toxicological and cell death effect [60].  [60]. In addition, direct contact of nCeO 2 with algae was reported previously to be responsible for toxicity and to cause membrane damage of P. subcapitata [53]. Further, an increase in intracellular reactive oxygen species (ROS) was observed in algae [19]. Intracellular ROS plays a role in the inhibition of photosynthesis and can indicate oxidative damage [61].
Previous studies have demonstrated the absence of uptake of uncoated and agglomerated nCeO 2 in algae [31,61]. The lack of ENPs uptake was linked to the formation of agglomerates that exceeded the pore sizes (ranges between 5 and 20 nm) of the algal cell wall [62], which, in turn, impeded plausible uptake by algae. Here, the observed agglomerates (up to 918 ± 74 nm) exceeded the algal cell wall pore sizes and, therefore, uptake was reasonably unlikely. Thus, growth inhibition observed was possibly due to the entrapment of algal cells by ENPs agglomerates. As a result, this may have reduced the light and nutrients' availability to the entrapped algal cells with concomitant growth inhibition. Previously, algal growth inhibition was observed to result from physical removal due to co-aggregation and co-sedimentation with nCeO 2 , as opposed to the toxicological and cell death effect [60]. nCeO 2 effects on algae have been observed to be concentration-and exposure durationdependent [60,63,64]. For example, growth inhibition of the algae Microcystis aeruginosa following exposure to < 25 nm nCeO 2 (1, 10 and 50 mg/L) in BG-11 media over 72 h was observed to be exposure-duration dependent [64]. No significant differences were observed between the controls and nCeO 2-treated samples after 24 h. However, after 48 h at concentrations of 1 and 10 mg/L nCeO 2 , results indicated algal growth promotion, but 50 mg/L induced significant growth inhibition of M. aeruginosa [64]. After 72 h, no algal growth was observed at 1 mg/L, increased significantly at 10 mg/L, but a significant inhibition was apparent at 50 mg/L nCeO 2 [64]. Deng and colleagues reported similar results, where they observed induction of growth promotion on marine diatom Phaeodactylum tricornutum at low nCeO 2 concentrations of ≤5 mg/L, whereas growth inhibition at ≥10 mg/L was documented [63].
Herein, the observed recovery of algal population under extended exposure conditions may be attributed to a decrease in nCeO 2 concentrations bioavailable for algae ( Figure S3) as ENPs formed aggregates ( Figure 1A and Figure S1) and underwent sedimentation in BG-11 media over time. Furthermore, recovery of the populations was attributed to the algae defence mechanisms in response to ENPs exposure. For example, algae employ a variety of defence mechanisms including activation of the antioxidative defence system to eliminate reactive oxygen species (ROS) [65,66], excretion of biomolecules to form a protective layer [67], and intracellular processes to decrease the cellular content of ENPs [68]. Thus, the findings herein and others demonstrates that the likely environmental risk of nCeO 2 on algae appear to be low at the morphological level even under extended exposure conditions.

Effect of nCeO 2 on Chl a Content
Photosynthesis is a key process in algae and quantified as Chl a content-an efficient indicator for physiological health status of algal cells [69,70]. Figure 3 demonstrates Chl a content of P. subcapitata for nCeO 2 -and non-exposed samples after 72 and 168 h. Contrary to the algal growth inhibition observed up to 72 h (Figure 2), findings in Figure 3 demonstrate that nCeO 2 enhanced Chl a content (p < 0.05) compared to the controls over the same period, but remarkably independent of the exposure concentration. In an earlier work, increase in Chl a content relative to controls were observed on C. reinhardtii following exposure to 4 nm-sized uncoated-nCeO 2 at 0.1-50 mg/L [16]. The observed increase in Chl a was associated with an interruption of the electron transport at the acceptor side of photosystem PSII [71,72]. Furthermore, other metal oxide ENPs, e.g., nZnO and nTiO 2 , were observed to enhance algal growth and Chl a content in Picochlorum sp. [73] and P. subcapitata [74][75][76]. The basis for Chl a content promotion was plausibly due to the conversion of other forms of pigments (e.g., Chl b content) into Chl a content as a response to ROS following exposure to ENPs [77]. Gui et al. [78] reported significant increases in Chl content after plant exposure to 10, 50, and 100 mg/kg nCeO 2 after 40 d. On day 50, only 50 and 100 mg/kg nCeO 2 concentrations increased the Chl content. Similar to our findings after 168 h, at the harvest stage, all of nCeO 2 treatments had no more significant difference [78]. After nCeO 2 (200 mg/L) exposure for 1 w, the Chl a and Chl b contents of rice seedlings did not show any significant changes relative to the control [79].
Toxics 2023, 11, x FOR PEER REVIEW 9 of 15 was observed at higher concentrations of 10-100 mg/L [19]. The primary cause of the observed photosynthetic inhibition was due to excessive levels of ROS formation, which, in turn, induced oxidative damage as evidenced by lipid peroxidation data [19,69]. Furthermore, ENPs bound onto algal membranes were observed to induce a shading effect or membrane damage and, in turn, inhibit the photosynthesis process [80][81][82]. To date, physical restraints and oxidative stress were reported to be mechanisms responsible for ENPs toxicity to algae [83]. The entrapment of algal cells by large ENP aggregates not only reduces light available for photosynthesis, but also prevents uptake of nutrients [70]. Among available ENP-toxicity mechanisms, a large number of studies indicated oxidative stress as the dominant toxicity mechanism of ENPs to algae [84]. For instance, Chen et al. [85] conducted a meta-analysis study, and the results showed that the level of ROS significantly increased by 90% in the presence of ENPs, indicating the accumulation of excess ROS in algal cells which ultimately caused oxidative stress. Additionally, ENP-induced ROS accumulation was not significantly influenced by ENP surface modification (p = 0.103) but was strongly influenced by the ENP type (p = 0.044), ENP dose (p = 0.001), and algae species (p < 0.001). Findings of the current study point to the need to consider longterm exposure conditions, as the results of nCeO2 on algae appear to be exposure time dependent. Figure 3. Chl a content of P. subcapitata at different nCeO2 exposure concentrations. Results are reported as mean ± standard deviation where n = 3 and asterisk denotes significant differences (p ≤ 0.05) between exposed and non-exposed samples after 72 h.

DNA Damage and Estimation of Genomic Template Stability
Cytotoxicity study (growth effect and Chl a content) results did not differ as a function of nCeO2 exposure concentration compared to the control (Figures 2 and 3). Thus, Figure 3. Chl a content of P. subcapitata at different nCeO 2 exposure concentrations. Results are reported as mean ± standard deviation where n = 3 and asterisk denotes significant differences (p ≤ 0.05) between exposed and non-exposed samples after 72 h.
After 168 h, no significant changes in Chl a content were observed between nCeO 2exposed and control samples. These findings indicated that the photosynthetic system of P. subcapitata can tolerate the presence of nCeO 2 under chronic exposure conditions. In contrast to growth promotion post 96 to 168 h, a reduction in Chl a content was observed after 168 h compared to 72 h. Similarly, the findings of Zhao et al. [64] showed that 10 mg/L nCeO 2 promoted algal growth, but it was also accompanied by a slight inhibition of photosynthetic yield. In addition, exposure of P. subcapitata to <50 nm uncoated-nCeO 2 at 0.01-100 mg/L for 72 h showed a dual response, firstly, with 20-50% stimulation in Chl a content at lower concentration range of 0.01-1 mg/L and, secondly, a significant inhibition was observed at higher concentrations of 10-100 mg/L [19]. The primary cause of the observed photosynthetic inhibition was due to excessive levels of ROS formation, which, in turn, induced oxidative damage as evidenced by lipid peroxidation data [19,69]. Furthermore, ENPs bound onto algal membranes were observed to induce a shading effect or membrane damage and, in turn, inhibit the photosynthesis process [80][81][82]. To date, physical restraints and oxidative stress were reported to be mechanisms responsible for ENPs toxicity to algae [83]. The entrapment of algal cells by large ENP aggregates not only reduces light available for photosynthesis, but also prevents uptake of nutrients [70]. Among available ENP-toxicity mechanisms, a large number of studies indicated oxidative stress as the dominant toxicity mechanism of ENPs to algae [84]. For instance, Chen et al. [85] conducted a meta-analysis study, and the results showed that the level of ROS significantly increased by 90% in the presence of ENPs, indicating the accumulation of excess ROS in algal cells which ultimately caused oxidative stress. Additionally, ENPinduced ROS accumulation was not significantly influenced by ENP surface modification (p = 0.103) but was strongly influenced by the ENP type (p = 0.044), ENP dose (p = 0.001), and algae species (p < 0.001). Findings of the current study point to the need to consider long-term exposure conditions, as the results of nCeO 2 on algae appear to be exposure time dependent.

DNA Damage and Estimation of Genomic Template Stability
Cytotoxicity study (growth effect and Chl a content) results did not differ as a function of nCeO 2 exposure concentration compared to the control (Figures 2 and 3). Thus, genotoxicity studies using RAPD-PCR method were conducted at 62.5, 250, and 1000 µg/L nCeO 2 representing the lowest, median, and highest concentrations, correspondingly. The results in Figure 4 show the RAPD-PCR profiles of isolated genomic DNA from nCeO 2 -treated and untreated samples. These profiles were also used to analyse GST% (Equation (2)). A negative control (no DNA) was included to ascertain whether any band observed was attributable to DNA amplification.
RAPD-PCR profiles for nCeO 2 treated algae using OPB1 primer were markedly different from those of the controls (Figure 4a). Modifications in DNA were in the form of appearance of two new clear bands at ±200-500 bp and disappearance of a normal clear band observed at ±900 bp in the controls (Figure 4a). Specifically, we observed various size ranges for nCeO 2 treated samples compared to the controls. Notably, the DNA strands in the form of clear bands for nCeO 2 treated samples were shorter (±200-500 bp) compared to the controls (±900 bp), indicating that the DNA of the control samples was more intact compared to one from nCeO 2 treated algae. The observed DNA modifications were neither concentration nor time dependent, indicating that nCeO 2 -induced 72 h-DNA damage persisted over 168 h. The observed modifications of RAPD-PCR profiles were likely due to one or a combination of variant events, e.g., DNA adducts, DNA breakage and mutation (e.g., point mutations and large rearrangements) [46,86]. The OPB14 primer produced similar RAPD profiles for controls and nCeO 2 -treated algal DNA irrespective of exposure concentration and time (Figure 4b), with a GTS of 100%. Similarly, RAPD profile analysis after exposure of Pseudomonas putida to aluminium oxide ENPs (nAl 2 O 3 ) showed no difference to the control, pointing to the induction of the DNA repair mechanisms [87]. Furthermore, the findings demonstrated primer-dependent genotoxicity. Previously, expo-sure of P. putida bacteria to <50 nm nAl 2 O 3 using four primers, e.g., OPA2, OPA10, OPA9, and OPA18 also showed primer-dependent DNA damage [87]. Results obtained using primer OPA2, demonstrated the most significant mutagenic action of nAl 2 O 3 , whereas the OPA10 and OPA18 primers RAPD band profiles showed the least mutagenic effect with small variations between ENPs-treated samples and control.
to one or a combination of variant events, e.g., DNA adducts, DNA breakage and mutation (e.g., point mutations and large rearrangements) [46,86]. The OPB14 primer produced similar RAPD profiles for controls and nCeO2-treated algal DNA irrespective of exposure concentration and time (Figure 4b), with a GTS of 100%. Similarly, RAPD profile analysis after exposure of Pseudomonas putida to aluminium oxide ENPs (nAl2O3) showed no difference to the control, pointing to the induction of the DNA repair mechanisms [87]. Furthermore, the findings demonstrated primer-dependent genotoxicity. Previously, exposure of P. putida bacteria to <50 nm nAl2O3 using four primers, e.g., OPA2, OPA10, OPA9, and OPA18 also showed primer-dependent DNA damage [87]. Results obtained using primer OPA2, demonstrated the most significant mutagenic action of nAl2O3, whereas the OPA10 and OPA18 primers RAPD band profiles showed the least mutagenic effect with small variations between ENPs-treated samples and control. Similarly, using the OPB1 primer, Mahaye et al. [27] observed DNA bands characterized by various size ranges compared to the controls following exposure of P. subcapitata to citrate-and branched-polyethyleneimine-nAu at 62.5-1 000 µ g/L for 168 h. The OPB14 primer produced similar RAPD-PCR profiles, irrespective of nAu coating type and exposure duration or concentration. The results of Mahaye et al. [27] indicated that DNA stability decreased after 72 h and increased after 168 h. Thus, they were indicative of likely DNA damage recovery over a long-term exposure period. However, herein, findings for nCeO2 demonstrated persistent DNA damage under extended exposure conditions. This is critical, as genotoxic effects may be, subsequently, transmitted to future generations with deleterious implications such as a compromised defence towards pests or an inability to adopt adverse environmental conditions.
In turn, this may affect survival and reproduction of algae, thus, compromising ecological balance as algae are food source for higher organisms in the food web. For example, transfer of metal oxide ENPs from algae to daphnia [88] or algae to fish [89] have been Similarly, using the OPB1 primer, Mahaye et al. [27] observed DNA bands characterized by various size ranges compared to the controls following exposure of P. subcapitata to citrate-and branched-polyethyleneimine-nAu at 62.5-1000 µg/L for 168 h. The OPB14 primer produced similar RAPD-PCR profiles, irrespective of nAu coating type and exposure duration or concentration. The results of Mahaye et al. [27] indicated that DNA stability decreased after 72 h and increased after 168 h. Thus, they were indicative of likely DNA damage recovery over a long-term exposure period. However, herein, findings for nCeO 2 demonstrated persistent DNA damage under extended exposure conditions. This is critical, as genotoxic effects may be, subsequently, transmitted to future generations with deleterious implications such as a compromised defence towards pests or an inability to adopt adverse environmental conditions.
In turn, this may affect survival and reproduction of algae, thus, compromising ecological balance as algae are food source for higher organisms in the food web. For example, transfer of metal oxide ENPs from algae to daphnia [88] or algae to fish [89] have been reported. In addition, the findings imply that high agglomeration of nCeO 2 in BG-11 medium does not reduce their reactivity and genotoxicity. These findings indicate that DNA damage on algae is ENPs type dependent. Previous findings have demonstrated irreparable DNA damage where affected cells can trigger cell death by activation of apoptosis to eliminate potentially damaged cells [90]. Conversely, herein, findings from apical endpoints after 168 h (Figures 2 and 3) plausibly indicate cell recovery under chronic conditions compared to 72 h. Metagenomic analysis results have demonstrated that microbial communities can protect themselves and recover their functions through keystone taxa, development of resistance, and resilience and functional redundancy [91]. The findings emphasize the importance of including genotoxicity methods in the risk assessment of ENPs on algae. Furthermore, current findings contribute to the limited body of knowledge on the effects of nCeO 2 on algae at low exposure concentrations (µg/L) and long-term exposure conditions, especially using the multi-maker approach that coupled genotoxicity biomarkers with apical endpoints to aid gain complete picture on the effect of these emerging contaminants.

Conclusions
nCeO 2 at 62.5-1000 µg/L exposure concentrations induced significant algal growth inhibition after 72 h, but growth promotion post 96-168 h, irrespective of exposure concentration. After 72 h, nCeO 2 enhanced Chl a content compared to the controls at all tested concentrations. However, after 168 h, no significant changes in Chl a content were observed between non-exposed and nCeO 2 -exposed samples (p > 0.05). The findings demonstrated that the high agglomeration of smaller-sized nCeO 2 do not reduce their reactivity nor hinder their toxicological effects. Furthermore, growth results demonstrated that algal cells could recover under long-term exposure conditions (post 96 h). Assessment of DNA damage using RAPD-PCR showed DNA bands modifications in the form of appearance of new bands and/or disappearance of normal bands compared to the controls. The observed modifications of RAPD-PCR profiles point to likely DNA adducts, DNA breakage, and mutation (point mutations and large rearrangements). In contrast to cell recovery observed after 96 h, DNA damage persisted over 168 h.
Overall, the study provided evidence that exposure duration plays a vital role on the cytotoxic and genotoxic response of P. subcapitata to nCeO 2 . To fully understand the mechanism of ENPs toxicity in algae, we recommend further studies at different endpoints at the molecular (e.g., chromosomal abnormalities, nucleus damage, DNA strand breaks, gene expression) and biochemical (e.g., catalase (CAT), glutathione S-transferase (GST), superoxide dismutase (SOD), etc.) levels. Furthermore, studies should be carried out at low environmentally relevant ENPs concentrations using a more realistic exposure medium (e.g., river water) and under chronic exposure conditions to fully understand the long-term impact of nCeO 2 on non-target aquatic organisms.

Conflicts of Interest:
The authors declare no conflict of interest.