To investigate the combined effects of OA and NPs, P. donghaiense, pre-acclimated to ambient and OA conditions for 110 generations, was cultured for seven days with simultaneous exposure to 2 factors. One factor was CO₂ (417 and 1045 ppm), while the other was NPs (108.15 ± 8.52 nm) (0, 5, 10, 15 mg L^-1), and the combined treatment groups involved the joint exposure of all different levels of CO₂ and NPs. Low CO₂ concentration (LC, 417 ppm) and high CO₂ concentration (HC, 1045 ppm) were represented the ambient air condition and simulated ocean acidification (OA) condition, respectively. Each treatment comprised of three replicates. The responses of P. donghaiense were recorded as photosynthetic variables (pigments, rETR, and OJIP parameters) and oxidative stress biomarkers (SOD, MDA, and Caro/Chl-a). The details of the experimental design are provided below.

Dinoflagellate P. donghaiense (Strain: PDES) was isolated from coastal waters around Dongtou Island, Zhejiang Province, East China Sea (27°80′N, 121°20′E). Prior to the experiment, P. donghaiense was maintained under simulated OA and ambient CO₂ conditions for approximately 110 generations. The pre-acclimation incubation was conducted in 3-L conical flasks inside growth chambers (RXZ-436C-CO₂, Jiangnan, China) with LED illumination (200 μmol photon m⁻² s⁻¹, 12L:12D). For the 7-d exposure experiments, the pre-acclimated P. donghaiense in the exponential phase was diluted to 1 × 10⁴ cells mL⁻¹ with fresh f/2 medium (Guillard and Ryther, 1962), and incubated at 20°C under LED illumination (200 μmol photon m⁻² s⁻¹, 12L:12D). Three replicates were maintained for each treatment.

Low CO₂ concentrations (417 ppm, LC) and high CO₂ concentrations (1045 ppm, HC) were selected to correspond with the current atmospheric pCO ₂ (Gao et al., 2021) and the predicted value for 2100 (IPCC, 2014), respectively. To maintain consistent CO₂ concentrations during the experiments, cultures were continuously aerated with two different CO₂ concentrations (LC and HC) through a 0.22 μm cellulose acetate filter (Shanghai Xinya, China) at approximately 300 mL min^-1. The levels of CO₂ in the incubators were monitored continuously using portable carbon dioxide meters (77535, AZ, China), and the pH_NBS (National Bureau of Standards) of the culture was measured daily with a portable pH meter (PHS-3E, INESA, Shanghai, China). The carbonate system in the culture was assessed by determining the concentrations of dissolved inorganic carbon (DIC), HCO 3 − , CO 3 2-   and CO₂ from the measured pH and CO₂ concentrations using the CO2SYS software (Lewis and Wallace, 1998) ( Table 1 ).

NPs stock solution (Tianjin Big Goose Technology, China) comprised of 2.5% (w/v) pure polystyrene nanospheres with smooth surfaces. The polystyrene nanospheres are highly hydrophobic and were homogeneously dispersed in 10 mL of deionized water, free of chemical additives and functional groups. The average diameter of the NPs was 108.15 ± 8.52 nm, measured using ImageJ software (National Institutes of Health, USA) ( Supplementary Figure S2 ). Prior to the experiment, the NPs solution was mixed on a vortex mixer (XW-80 A, Haimen Kylin Bell, China) for 2 min. Four nominal NPs concentrations were used: 0, 5, 10, and 15 mg L⁻¹, equivalent to approximately 0, 1 × 10¹³, 2 × 10¹³, and 3 × 10¹³ particles L⁻¹, respectively. The hydrophobicity of the NPs and the aeration of the culture medium in the experiment allowed the cells and NPs to be distributed homogeneously in the culture.

The cell density was determined daily after sampling by counting the cells with a phytoplankton counting chamber (CC-F, Beijing Polytek, China) under a microscope (Eclipse E200MVR, Nikon, Japan). The growth rate (μ) was calculated using the formula: μ = ln (C₇/C₀)/(t₇ - t₀), where C₇ and C₀ represent the cell densities (cells mL^-1) on days 7 and 0, respectively. The growth inhibition was estimated using the following equation: Inhibition (%) = (1 – T/C) × 100, where C and T are the cell densities in the control and treatment samples, respectively.

At Day 7, a sample (20 mL) from each treatment was filtered through a glass fiber filter (Whatman GF/F, 25 mm diameter, nominal pore 0.7 μm) to quantify chlorophyll-a (Chl-a) and carotenoid (Caro) contents. Pigments were extracted after mixing the filtered cells with 4 mL of pure methanol and stored in darkness for 24 hours at 4°C. Then the mixture was centrifuged at 2000 g for 10 min (TG16A - WS centrifuge, Saitexiangyi, China) and the absorbance of the supernatants was determined using a spectrophotometer (U-3900; Hitachi, Japan). The pigment contents were calculated as follows: Chl-a (μg mL^-1) = 16.29 × A_665.2 - 8.54 × A₆₅₂ (Porra, 2002); Caro (μg mL^-1) = 4 × A₄₈₀ (Strickland and Parsons, 1972), where A_665.2, A₆₅₂, and A₄₈₀ are the absorbances at wavelengths of 665.2, 652, and 480 nm, respectively. The pigment content per cell was determined by dividing the pigment concentration by cell density.

Approximately 5 × 10⁶ cells from each treatment group were collected on day 7 after 10 min of centrifugation at 3000 × g (Allegra X-12R Centrifuge, Beckman Coulter, USA). The pellet was resuspended in 5 mL of Milli-Q ultrapure water (SYUP-I-60 L; Shenyuan, China) and lysed at 4°C (JN-Mini Pro; JNBIO, China). The supernatant was obtained by centrifugation at 8000 × g for 10 min at 4°C (Allegra 64R Centrifuge, Beckman Coulter, USA) to assess biochemical indicators. The SOD activity and MDA levels were evaluated according to the manufacturer’s instructions (Nanjing Jiancheng Bioengineering, China).

The rapid light-response curve was obtained using a PSI fluorometer (AquaPenC; Photon System Instruments, Czech Republic). Eight different actinic light intensities (0, 10, 20, 50, 150, 300, 600, and 1000 μmol m^-2 s^-1) were applied for 60 s each. After each light intensity, a 0.8-s saturation light pulse (3000 μmol m^-2 s^-1) captured ΔF/F’_m. The rETR was calculated as follows: rETR = ΔF/F’_m × 0.5 × E, where 0.5 represents the allocation of 50% of actinic light energy to photosystem II (PSII), E is for actinic light, and ΔF/F’_m is the photochemical efficiency of PSII. A hyperbolic tangent function was fitted using this equation (Jassby and Platt, 1976): Y = rETR_max × tanh (α × X/rETR_max), where α denotes light energy utilization efficiency and rETR_max indicates the maximum relative electron transport rate.

On day 7, cells (4 mL) from each replicate were allowed to adapt in the dark for 20 min before measurement. The chlorophyll fluorescence transient (OJIP test) was measured using a PSI fluorometer. The fluorometer measures the transition between O-J-I-P steps: O step is at approximately 20 μs where all PSII reaction centers are open; the J step is at approximately 2 ms; the I step is at approximately 20 ms; and the P step is at about 300 ms where all reaction centers are closed, and the fluorescence intensity reaches its maximum (Strasser and Govindjee, 1992). OJIP curves were drawn to obtain more detailed parameters, including F_v/F_m, V_I, V_J, M₀, Area, S_m, φ_P0, ψ₀, φ_E0, ABS/RC, TR₀/RC, ET₀/RC, DI₀/RC, RC/CS₀, ABS/CS₀, TR₀/CS₀, ET₀/CS₀, DI₀/CS₀, PI_ABS. All data were collected in triplicate, and the values were normalized relative to the LC control treatment (without NPs) to determine the heatmap. The physiological meanings and calculation formulas for these parameters are listed in Supplementary Table S1 (Strasser et al., 2000).

Data are presented as the mean ± standard deviation (SD). Each treatment was performed in triplicate. A one-way ANOVA (LSD) was performed to compare the data among the different treatments on day 7. A two-way ANOVA was employed to evaluate the significant differences and interactions between CO₂ and NP concentrations. All statistical analyses were performed using SPSS (version 26.0, SPSS Inc., Chicago, IL, USA), and figures were plotted using GraphPad Prism (version 9.0, San Diego, CA, USA). Statistical significance was set at p< 0.05.

The carbonate system remained stable in each treatment during the 7-day incubation. The pH values in the LC and HC groups were significantly different (p< 0.05). In HC, HCO 3 − and CO₂ levels increased by 34.3% and 157.6%, respectively, while CO 3 2 - decreased by 31.3% compared to those in LC. The carbonate system in the cultures was not affected by different concentrations of NPs in the treatments within the same CO₂ levels ( Table 1 ).

The growth of P. donghaiense was significantly inhibited by the NPs, except at 5 mg L⁻¹ in HC ( Figure 1 ). NP-induced growth inhibition exhibited a concentration-dependent trend in both LC and HC treatments ( Figure 1B ). The inhibition induced by NPs was 7.74%, 17.56%, and 34.82% in LC at 5 mg L⁻¹, 10 mg L⁻¹, and 15 mg L⁻¹ NPs, respectively. However, HC treatment promoted growth and mitigated the detrimental effects of NPs compared to LC. The lowest NPs concentration (5 mg L^-1) did not inhibit the growth in HC treatment ( Figure 1A ). NP-induced inhibition also decreased to 9.40% (10 mg L⁻¹), 15.94% (15 mg L⁻¹) in HC treatments compared to that in LC ( Figure 1B ).

In LC treatments, NPs increased the Chl-a (19.11%–55.46%) and the Caro (9.56%–40.19%) contents but decreased the Caro/Chl-a ratio ( Figure 2 ). Compared with LC, HC (without NPs) did not change Chl-a or Caro/Chl-a ratio but decreased Caro ( Figure 2 ). However, Chl-a and Caro were decreased in HC+NPs treatments, especially at higher NPs concentrations (10 and 15 mg L⁻¹) compared to LC ( Figures 2A, B ), but Caro/Chl-a ratio did not vary ( Figure 2C ). The two-way ANOVA tests revealed that CO₂ and NPs significantly affected the growth rate (μ), pigments, and CO₂ was the primary influencing factor. The interactive effects of these two factors were evident on μ and pigments ( Table 2 ).

Compared with LC treatment, HC (without NPs) did not significantly alter SOD but enhanced MDA levels ( Figure 3 ). The effects of NPs on the SOD and MDA levels varied in the LC and HC groups. No significant effect of NPs on SOD was detected in the LC treatment, except in the 5 mg L⁻¹ NPs treatment ( Figure 3A ). High NPs concentration (15 mg L⁻¹) significantly increased the MDA content in LC-treated cells ( Figure 3B ). Compared to LC, HC+NPs treatment enhanced SOD activity but reduced MDA content ( Figure 3 ). The two-way ANOVA tests revealed that SOD and MDA levels were significantly affected by CO₂ and NPs, with CO₂ being the major influencing factor. Moreover, the interactive effects of CO₂ and NPs on the SOD and MDA levels were also evident ( Table 2 ).

The rapid light curves and photo-physiological parameters (α, rETR_max, and E_k) revealed the responses of P. donghaiense to different treatments ( Figure 4 ). Compared to LC treatment, HC, LC+NPs, and HC+NPs significantly enhanced the slope of the rETR curve (α), the maximum electron transport rate (rETR_max) and the light saturation point (E_k) ( Table 3 ). However, HC+NPs did not further enhance these three parameters compared with HC (without NPs) ( Table 3 ). The two-way ANOVA tests revealed that CO₂ significantly affected α, while NPs significantly affected rETR_max, α, and E_k. Additionally, an interactive effect between these two factors on rETR_max, α, and E_k was observed ( Table 2 ).

The chlorophyll fluorescence intensity transitions from its minimum level (O-level) to its maximum level (P-level) through two intermediate steps, designated J and I ( Supplementary Figure S3 ). The OJIP curve’s slope increased significantly when exposed to NPs under LC conditions, indicating that the reduction of Q_A was faster ( Supplementary Figure S3A ). The slope change was significantly less in P. donghaiense cultures that were exposed to NPs under HC treatments ( Supplementary Figure S3B ).

The response mechanisms of cells to HC, NPs, and HC+NPs treatments were revealed when the relevant parameters were estimated. Compared to LC (LC0), HC (without NPs) improved F_v/F_m and Area but reduced V_I, V_J, M₀, and S_m ( Figure 5 ). Moreover, energy transfer efficiency parameters showed an increase in response to HC, including the maximum quantum yield of primary photochemistry (φ_P0), the efficiency of a trapped exciton to move downstream (ψ₀), and the probability of an absorbed photon to move downstream (φ_E0) ( Figure 5 ). The parameters related to the energy flux in the reaction center indicated a decrease in response to HC, such as absorption (ABS/RC), trapping (TR₀/RC), and dissipation energy (DI₀/RC) per active reaction center (RC), and an increase in transport energy (ET₀/RC). Furthermore, HC increased the density of the active reaction centers (RC/CS₀), absorption (ABS/CS₀), trapping (TR₀/CS₀), transport energy (ET₀/CS₀) per excited cross section (CS₀) and photosynthetic capacity (PI_ABS), but decreased the dissipation energy (DI₀/CS₀), significantly ( Figures 5 , 6 ).

In LC-treated P. donghaiense, NPs increased the F_v/F_m, φ_P0, ψ₀, φ_E0, ET₀/RC, ET₀/CS₀, and PI_ABS, but decreased the V_I, V_J, M₀, Area, S_m, ABS/RC, TR₀/RC, DI₀/RC, ABS/CS₀, TR₀/CS₀, DI₀/CS₀ ( Figures 5 , 6 ). However, in contrast to LC treatments, NPs enhanced the Area, RC/CS_0, and TR₀/CS₀ in HC treatments ( Figure 5 ).

The presence of NPs had a significant detrimental effect on the growth of P. donghaiense ( Figure 1 ), which was attributed to NP-induced oxidative stress, consistent with previous studies (Natarajan et al., 2020; Sendra et al., 2019; Wang et al., 2021). SOD is an enzyme intricately linked to cellular oxygen metabolism in living organisms and protects organisms against oxidative stress. MDA is a critical byproduct of lipid peroxidation and is widely recognized as an indicator of cellular oxidative damage (Dong et al., 2020). Caro can eliminate excess ROS and inhibit lipid peroxidation (Rezayian et al., 2019), and the Caro/Chl-a ratio is a valuable indicator of non-enzymatic antioxidant capacity. In this study, NP-induced oxidative stress increased MDA levels, but the antioxidant mechanisms (SOD and Caro/Chl-a) were not enhanced ( Figures 2C , 3 ). Therefore, NPs inhibited growth in the LC treatments due to the oxidative stress.

The variations in the physiological state of P. donghaiense cells in the presence of NPs were evident in the levels of Chl-a, rETR, and OJIP. Chl-a is crucial for energy capture and transfer during photosynthesis (Kato et al., 2020), and its increased content can compensate for growth by promoting photosynthesis ( Figure 2A ). This phenomenon is one of the adaptive mechanisms of microalgae in response to external stress (e.g., NPs) (Chen et al., 2022). rETR_max can convert absorbed light energy into chemical energy that flows into the Calvin cycle, and α means photosynthetic efficiency at sub-saturating irradiance (Raniello et al., 2006). The rise in rETR_max and α also implied that the stress response in cells increased photosynthesis under the NPs stress ( Table 3 ). Chlorophyll fluorescence transient analysis (OJIP) provides insights into several photosynthetic parameters. In the LC-treated cells, the slope of the OJIP curve increased significantly with NPs treatment, indicating that the reduction in Q_A was rapid and that P. donghaiense responded vigorously to mitigate the effects of NPs (Chen et al., 2022) ( Supplementary Figure S3A ). ABS/RC, TR₀/RC, and ET₀/RC indicate the efficiency of light absorption, electron trapping, and electron transport per reaction center, respectively. In contrast, DI₀/RC indicates the heat dissipation per reaction center (Rai-Kalal and Jajoo, 2021). Under NPs stress, the decrease of ABS/RC and TR₀/RC but an increase of ET₀/RC in LC treatments represented an improvement in the overall photochemical efficiency of primary photochemistry at the PSII reaction center of P. donghaiense, which is also manifested in the elevated levels of φ_P0 (TR₀/ABS), ψ₀ (ET₀/TR₀) and φ_E0 (ET₀/ABS) ( Figures 5 , 7 ). This might be attributed to the decrease in DI₀/RC, which minimized the energy loss by heat dissipation per reaction center, thereby increasing the photochemical efficiency of primary photochemistry in the presence of NPs ( Figures 5 , 7 ). PI_ABS is considered a sensitive and insightful indicator of stress and is widely used to compare primary photochemical reactions. Under NP-induced stress, PI_ABS increased significantly, suggesting a notable enhancement in the photosynthetic capacity of PSII ( Figure 6 ). Although these changes reflected the stress response of cells to NPs under LC treatment, they did not alleviate NP-induced oxidative stress; therefore, the growth of P. donghaiense was significantly inhibited by NPs. Similarly, the diatom P. tricornutum also exhibited a stress response to improve photosynthesis, as reflected in the OJIP parameters and elevated Chl-a content when exposed to MPs (Chen et al., 2022). However, the response of microalgae to NPs is species-specific. The growth of the cyanobacteria M. aeruginosa was promoted by NPs owing to their strong antioxidant capacity and increased metabolic activity (Wang et al., 2023b).

The accelerated growth of P. donghaiense under OA (indicated by HC) conditions could be attributed to enhanced photosynthesis. The photosynthetic capacity (PI_ABS), photochemical efficiency (φ_P0, ψ_0, and φ_E0), and photosynthetic efficiency (α) were all enhanced by HC compared with their efficiency under LC treatment, respectively ( Figures 5 , 6 , Table 3 ). The increase in the number of active reaction centers (RC/CS₀) enhanced the efficient transfer of the maximum absorbed light energy to these centers ( Figures 5 , 7 ). This increase can be attributed to improved association between light harvesting center II (LHC II) and the PSII complex, resulting in enhanced electron flow from Q_A to Q_B in response to HC (Rai-Kalal and Jajoo, 2021). The significantly increased TR₀/CS₀ and ET₀/CS₀ under HC suggested a heightened energy flux per cross-section available for photosynthesis. This phenomenon may be attributed to a decrease in heat dissipation (DI₀/CS₀), which improved the efficiency of photosynthetic energy utilization ( Figures 5 , 7 ). The enhanced rETR_max more effectively converted the absorbed light energy into chemical energy that flowed into the Calvin cycle ( Table 3 ). Furthermore, HC decreases the allocation of electron transport to photorespiration and increases electron flow to Rubisco carboxylation, improving the efficiency of carbon concentration mechanisms (CCMs) (Robredo et al., 2010). The HC allows for 20% energy savings through downregulation of the CCMs (Hopkinson et al., 2011; Shi et al., 2015), which is expected to further stimulate growth (Fu et al., 2008; Ou et al., 2017) ( Figure 1A ) and photosynthetic rate (Chen et al., 2015; Dason et al., 2004; Wang et al., 2023c) ( Figures 4 , 6 , 7 ).

OA (HC) not only enhanced growth but also mitigated the adverse effects of NPs on P. donghaiense ( Figure 1 ). Higher growth and less NP-induced inhibition were observed in the HC+NPs than in the LC+NPs ( Figure 1 ). Under HC conditions, the heightened photosynthetic activity provided substantial energy to the cells, facilitating the functioning of the antioxidant system and thereby alleviating the oxidative stress induced by NPs (Ren et al., 2023), as indicated by increased SOD activity and decreased MDA content ( Figure 3 ). Therefore, the cells maintained internal homeostasis, which was reflected in the relatively minor variations in OJIP parameters ( Figure 5 ) and PI_ABS ( Figure 6 ) compared with these parameters in LC. The unaltered parameters of α, rETR_max, φ_P0, ψ_0, and φ_E0 ( Table 3 , Figure 5 ) in HC treatment indicated that, in the presence of adequate cellular energy, the energy transfer efficiency did not display compensatory increments in photosynthesis under NPs, which was contradictory to the situation observed under LC. Similarly, OA can enhance the antioxidant capacity of microalgae to alleviate the adverse effects of environmental pollutants, such as cadmium (Dong et al., 2020), copper (Xu et al., 2022), and CuO (Wang et al., 2023a).

Under the impacts of climate change (e.g., OA) and environmental pollution (e.g., NPs), the occurrence of HAB has increased worldwide. Over the last two decades (2003–2020), the global extent of areas affected by blooms expanded by 13.2% (3.97 million km²), and the frequency of global bloom outbreaks increased by 59.2% (Dai et al., 2023). Considering the Pacific coast of East Asia, including Chinese coastal waters, HAB have a significant deleterious influence on aquaculture and the ecological environment (Sakamoto et al., 2021; Yu et al., 2023). Since 1980, approximately 1622 algal blooms have been recorded in Chinese coastal waters (Sakamoto et al., 2021), with a growth rate of 40 ± 4% per decade (Xiao et al., 2019). Furthermore, climate change and environmental pollutants have a combined effect on HAB. OA alleviated the NPs-induced adverse effects and even completely diminished the inhibitory effects of NPs at ≤ 1 × 10¹³ NP particles L^-1. The concentrations of NPs used in the experiments are much higher than those found in the natural environment, such as those in the Southern China Sea (about 2.2 items L⁻¹) and the northern Gulf of Mexico (74 items L⁻¹) (Wang et al., 2022). If these data are extrapolated to natural conditions, the effects of NPs on HAB in future OA scenarios can be disregarded as long as the NPs do not have more than 1 × 10¹³ particles L⁻¹. It can be assumed that HAB species can continuously bloom and threaten coastal communities and public health under future ocean conditions.