Physiological Homeostasis Alteration and Cellular Structure Damage of Chlorella Vulgaris Exposed to Silver Nanoparticles with Various Microstructure Morphologies

The toxicity of silver nanoparticles (AgNPs) with single morphology exposed to aquatic organisms had been well revealed in the past decade, but few studies have been carried out to evaluate the toxicity differences between AgNPs with various microstructure morphologies, especially to algae. In this work, Chlorella vulgaris was used as the tested organism to illustrate the differences of toxic effects between silver nanospheres (AgNSs), silver nanocubes (AgNCs) and silver nanoplates (AgPLs) with concentration of 0.5, 1.0, 2.0, 5.0 mg (cid:0) L -1 , based on the algae’s growth (72h), chlorophyll-a content, antioxidant enzyme activity, lipid peroxidation and cell apoptosis (48h). The results showed that the toxicity level exposed to Chlorella vulgaris was in the order of AgPLs> AgNCs> AgNSs. The difference shown indicated that the potential toxicity of AgNPs is primarily depended on their microstructure morphologies. This current study initially revealed the structure-effects of AgNPs on Chlorella vulgaris, provided a scientic basis for aquatic environmental risk assessment.


Introduction
Silver nanoparticles (AgNPs) produced, transported and applicated by humans are ultimately released into the environment and are thus potentially toxic to environmental organisms (Dale et al., 2015).
Besides, in natural environment, Ag + can be reduced into nano silver by dehydrogenase or reducing sugar in microorganisms or plants . AgNPs is highly chemically active and can easily interact with environmental medium (e.g., physical, chemical and biological reaction), which leads to the migration and transformation of AgNPs, eventually affecting their toxic effects ).
Although much work was gone into focusing on the biological effects of AgNPs, which systematically revealed action mechanisms and dose-effect relationship of AgNPs with single morphology (e.g., spherical, rodlike, and hexagonal) to various organisms, the shape-dependent toxicity was less explored nanoplates (AgPLs) and silver nanospheres (AgNSs) and found that AgNSs were highly toxic to Staphylococcus aureus and Escherichia coli (E. coli), yet less toxic than AgPLs (2012). In contrast, some studies reported that the toxicities of AgPLs and AgNSs in P. aeruginosa and E. coli were opposite (Muhammad et al. 2016). Therefore, there is a dispute related to shape-dependent toxicity of AgNPs.
Thus, it is of great importance to explore the toxic differences of AgNPs with various morphologies on organisms in ecosystem, providing direct evidence for the risk assessment of AgNPs with different structure.
As a main primary producer in aquatic systems, algae play a crucial role in the environmental homeostasis of water body. The toxicity of AgNPs to microalgae are known to be related to photosynthetic e ciency inhibition, reactive oxygen species (ROS) generation, metabolism interference, and organelles damage (He et al. 2017;Dorobantu et al. 2015). Whereas, the research concerning the toxic effects of AgNPs with multi-morphologies on algae were still limited. The investigation of structureactivity relationship of between AgNPs and algae is of great scienti c value to the comprehensive and indepth understanding of AgNPs biological toxicity in aquatic environment and the evaluation of safety of water body.
Herein, we compared the toxicities of three AgNPs of various morphologies (AgNSs, AgNCs, and AgPLs) on an alga, Chlorella vulgaris. We determined the growth condition, the chlorophyll-a content, antioxidant enzyme activity, lipid peroxidation degree, and cell apoptosis. The experimental results of study could provide valuable information about the toxicity of AgNPs with various microstructure morphologies to aquatic organisms, which might be useful for assessing ecological risk of AgNPs.

Tested algae
Experiments were carried out with cultures of the unicellular green algae C. vulgaris FACHB-8, purchased from Freshwater Algae Culture Collection at the Institute of Hydrobiology (Wuhan, China).

Algae growth inhibition test
Pre-cultures of C. vulgaris at exponential phase were inoculated into BG11 medium (Table S1) with an initial cell density of 10×10 6 cells/ml for algae growth and chlorophyll-a content assay. Before the inoculation of algae cells, different concentrations of AgNPs with various morphologies were added into the growth media. The nominal concentrations for AgNPs for the test in the medium were 0.5 mg L − 1 , 1.0 mg L − 1 , 2.0 mg L − 1 , 5.0 mg L − 1 (AgNPs concentrations veri ed by inductively coupled plasma mass spectrometry (ICP-MS) showed relatively small deviation). In the exposure test, cultures were grown in 150ml Erlenmeyer asks containing 30ml of BG11 medium containing different concentrations of AgNPs.
Algae cells were grown in medium containing various AgNPs morphologies and concentration for 72h and algae growth was monitored after every 12h. Cell density was measured by cell counting using an optical microscope (Nikon, China). The method proposed by Sartory (1984) was used to determine the content of photosynthetic pigment.

Analysis of antioxidant enzyme activity and lipid peroxidation
Cells at 48h were harvested and analyzed for antioxidant enzyme activity and lipid peroxidation analysis. Superoxide dismutase (SOD) activity was assayed by monitoring the inhibition of reduction of Nitroblue Tetrazolium chloride (NBT) photochemically. The determination of Malondialdehyde (MDA) content was accomplished by the color reaction of Thiobarbituric Acid (TBA) in acid condition (Heath et al. 1968). The ROS level of the algae cells were measured using the ROS assay kit (Beyotime Institute of Biotechnology, Haimen, China). The activity of Peroxidase (POD) and Catalase (CAT) were determined by guaiacol method and UV absorption method respectively.

Analysis of cells apoptosis
After exposure of 48h, centrifuged at 4000rmp for 10min at room temperature and cultured with 195µL Annexin V-FITC binding solution and 5µL Annexin V-FITC for 10min at 24°C. Next, cells were assayed using an Annexin V-FITC apoptosis detection kit (Beyotime Institute of Biotechnology, Haimen, China), the cell apoptosis and cell size were analyzed on a FACS JAZZ ow cytometer (Becton Dickinson, Sanjose, CA, USA) equipped with an argon laser (excitation at 488 nm).

Growth inhibition test
During the experimental process, AgNPs with different concentrations (0.5, 1, 2, 5 mg L − 1 ) had different morphology response to the tested algae. Compared to the control, C. vulgaris was highly sensitive to AgPLs than the others, which showed signi cantly disruption and aggregation (Fig. 1A). After 48h exposure in AgNPs, evident whitening and stunting appeared on algae cells. The result shows that the presence of AgNPs could lead to the aggregation and even rupture of algae cells, thus affecting their normal physiological and biochemical functions. The 48h inhibition rate of C. vulgaris exposed to AgNSs, AgNCs and AgPLs increased with the concentration increasing in a dose related manner (Fig. 1B) (P < 0.5). However, tested concentrations of AgNSs and AgNCs did not induce substantial inhibition on cell growth, with the maximum inhibition effect of 43% and 38% at 5 mg L − 1 . On the contrary, low concentration of AgPLs (0.5 mg L − 1 ) could repress cell growth signi cantly by 60%, indicating a higher toxic effect on cell growth and vitality than AgNSs and AgNCs in this study.
The inhibitory effects of AgNSs, AgNCs and AgPLs on the growth and photosynthetic of C. vulgaris are shown in Fig. 1C-H. The cell density of microalgae is the most intuitive parameter to measure the cell growth, and the chlorophyll-a content can be used as one of the indicators to measure the physiological effects of algae cells (Metzler et al. 2012). It can be seen from the gure that the presence of AgNPs inhibited the cells growth and chlorophyll-a synthesis, the inhibition e ciency had a positive relationship with the concentration of AgNPs. Incubation of algae cells in medium containing AgPLs resulted in signi cant repression on cell growth even at lower concentration of 0.5 mg L − 1 at 72h. Increase of the AgNPs concentration to 5 mg L − 1 impeded C. vulgaris propagation completely. A similar decrease in chlorophyll-a was observed in AgNPs treatment at 0.5, 1, 2 and 5 mg L − 1 , which was in agreement with the cell density histogram. The content of chlorophyll-a was only 26.92%, compared with the control group, as the concentration of AgPLs was 5 mg L − 1 . The chlorophyll-a content and cell density of AgNSs decreased the least, that is, the effect of AgNSs on photosynthetic pigments activity and growth of algae was the weakest. Substantial decreased in chlorophyll-a contents of algae exposed to AgNPs indicated that the photochemical activity was repressed and occurrence of photoinhibition was inevitable. For AgNPs with different morphologies, ake nano silver has a stronger biocidal effect than spherical nanoparticles (Pal et al. 2007), the experimental results con rm this viewpoint to some extent.

Physiological and biochemical test
ROS and MDA are direct indicators of lipid peroxidation and oxidative damage in algae system (Gao et al. 2020). Figure 2 shows the intracellular ROS level ( Fig. 2A) and MDA contents (Fig. 2B) of C. vulgaris as a function of the AgNPs concentration. The intracellular ROS level was signi cantly promoted by increasing AgNPs concentration, whereas relative MDA activity gradually decreased with increasing AgNPs concentrations, resulting a minimum value at 5 of AgNPs. After exposure to 5 mg L − 1 AgPLs for 48h, level of the ROS was the highest, and was 2.7-times those of the control, accordingly. As shown in Fig. 2B, the content of MDA was higher than that of the control group, while MDA content exposed to AgPLs increased with the increase of concentration. MDA is one of the products of lipid peroxidation (Gaschler et al. 2017), the content of MDA will be lower than that of the control group when the cells were damaged and ruptured. The results showed that C. vulgaris was more sensitive to the AgPLs in terms of MDA and ROS.
In order to determine the effects of AgNPs with different morphologies on enantioselective oxidative stress of C. vulgaris, AgNSs, AgNCs and AgPLs exposure group were examined. As seen in Fig. 3, the SOD (Fig. 3A), CAT (Fig. 3B) and POD (Fig. 3C) exhibited similar trend upon exposure to AgNPs after 48h exposure. AgNPs with different morphologies tested differently affected the antioxidant enzyme activities in the algal cells. When C. vulgaris cells exposed to different concentrations of AgNPs (0.5, 1, 2 and 5 mg L − 1 ), SOD activities were enhanced compared with those untreated cells. Besides, the increase of SOD and CAT were most obvious in the AgPLs treated group. SOD enzyme can decompose O 2 − , CAT and POD enzymes are involved in the decomposition of hydrogen peroxide, thus eliminating the in uence of ROS. As expected, under the stimulation of AgNPs, the activity of antioxidant enzymes and the antioxidant capacity increased respectively. In this study, the enhancement on CAT, SOD and POD of cultures exposed to AgNPs signify the involvement of antioxidant enzyme in the antioxidant defense against the ROS. Oxidative stress could enhance the antioxidant capacity of cells, and they couldn't maintain internal stability when the content of reactive oxygen species increased to a certain amount.

Effect of AgNPs on cells membrane damage
The effects of AgNPs on apoptosis of C. vulgaris were evaluated by Annexin V-FITC/PI staining. Cell membrane damage was the primary manifestation of apoptosis (Kundrát et al. 2016), when the cell membrane damaged, Annexin V could bind to phosphatidylserine on the surface of cell membrane, and the uorescence intensity could be detected by FITC and PI dyes. In ow cytometry image, upper left and upper right, lower right quadrants show percentage of early apoptotic cells, advanced stage of apoptotic cells and dead cells, lower left quadrant shows the percentage of live cells respectively. As shown in Fig. 4A, B and C, after exposure to AgNPs with various morphologies for 48 hours, the cell size, complexity and apoptosis of the experimental group were higher than those of the control group.

Conclusion
In this study, the toxicity differences of AgNSs, AgNCs and AgPLs in C. vulgaris were compared based on the growth inhibition, photosynthetic pigment content, antioxidant enzyme activity, lipid peroxidation, morphology and apoptosis of the cells. The growth and photosynthetic pigment content of C. vulgaris were signi cantly inhibited by all three AgNPs. The results of antioxidant enzymes activities and cell apoptosis indicated that AgPLs possessed the highest potent toxicity to C. vulgaris, followed by AgNCs (moderate) and AgNSs (lowest). The experimental ndings indicated that the toxic effects of AgNPs on aquatic organisms may primarily depend on its microstructure morphology rather than on its nano size in some certain cases, especially upon exposure to C. vulgaris. This work pointed out that the microstructure morphology effects and nanoscale effects are of equal importance for understanding the toxicities of AgNPs in aquatic primary producers, whereas the former may results in more potential impact than the latter when it is exposed to some certain algae, e.g., C. vulgaris.
Declarations Figure 1 Determination of AgNPs SEM and cell morphology observation, 48h inhibition rates, cell density and chlorophyll-a contents of C. vulgaris exposed to AgNPs with various morphologies for 3 days. AgNPs SEM images and cell morphology (A). 48h inhibition rates (B). Cell density (C, D, E). Chlorophyll-a content (F, G, H). Lower-case letters demonstrate averages comparisons between treatments by LSD test (p < 0.05).

Figure 2
Changes of intracellular ROS level (A), MDA content (B) in C. vulgaris exposed to AgNPs with different concentrations after 48h. Lower-case letters demonstrate averages comparisons between treatments by LSD test (p < 0.05).

Figure 3
Changes of SOD (A), CAT (B) and POD (C) in C. vulgaris at different concentrations exposed to AgNPs with different concentrations after 48h. Lower-case letters demonstrate averages comparisons between treatments by LSD test (p < 0.05).