1 INTRODUCTION

Harmful algal bloom (HAB) is a result of eutrophication in water, which is considered one of the major marine environmental threats in the world (Anderson et al., 2012). In recent years, eutrophication in the estuary of Changjiang (Yangtze) River and coastal of the East China Sea (ECS) has caused a huge damage to local aquatic systems and other related sectors (He et al., 2013; Shen et al., 2019). Among HAB-forming species, dinoflagellates are accountable for more than 70% of HAB events that have been reported (Smayda, 1997). In addition, dinoflagellate Prorocentrum donghaiense is one of the most common HAB-forming species in the ECS due to its high biomass accumulation, being responsible for the large-scale (ca. 10 000 km²) bloom events and for marine ecosystem damage (Lu et al., 2005; Tang et al., 2006; Zhao et al., 2009). It frequently co-blooms with other toxic algae such as Alexandrium catenella and Karenia mikimotoi, resulting in mass mortalities of farmed fish and other marine fauna (Wang and Wu, 2009; Shen et al., 2019).

On the other hand, global climate changes aggravated HABs in frequency, scale, and severity (Moore et al., 2015; Wells et al., 2015). Recent works reported that phytoplankton, including HAB-forming species, revealed a large latitudinal range shifts on average of over 400 km per decade (Glibert et al., 2014). Ocean acidification is another factor for altering the community structure of phytoplankton and for increased occurrence of dinoflagellate-induced HAB (Howes et al., 2015). Some dinoflagellate species are able to release toxic metabolites lethal to marine organisms and hazardous to public health (Van Dolah, 2000; Zingone and Enevoldsen, 2000; Shang et al., 2020). As a nontoxic species, P. donghaiense often cause a series of HAB over large areas (1 000–10 000 km²) in a long time (> 30 d) (Lu et al., 2005; Glibert et al., 2012). During a HAB occurrence period, P. donghaiense competes with other marine organisms for oxygen and nutrient, risking to those co-existent creatures (Anderson et al., 2002; Lin et al., 2015).

Seaweed aquaculture provides important services for marine ecosystems, e.g., increasing overall biodiversity, sequestrating CO₂, mitigating eutrophication, and thus suppressing HABs (Tang and Gobler, 2011; Yang et al., 2015). Mariculture of seaweeds has been developing rapidly in East Asian countries since the early 20^th century by continuously rising demands for dietary supplements and healthy food (Tseng, 2001; Krumhansl and Scheibling, 2012; Ma et al., 2018). Natural macroalgae populations are vital components of coastal and estuarine ecosystems, have significantly declined in many parts of the world because of the increasing anthropogenic disturbances (Lotze et al., 2006; Sala and Knowlton, 2006; Branch et al., 2013). It have been found that HABs could cause population loss, seedling development delay, thalli bleaching, and photosynthetic rate decrease of natural and maricultured macroalgae (Kim et al., 2015; Ma et al., 2017, 2020). However, the physiological and developmental responses of seaweed to HAB have rarely been documented due to the "unpredictable" nature of algal blooms (Ma et al., 2017; Shang et al., 2020).

Brown seaweed Sargassum fusiformis is an edible marine vegetable consumed widely in the East Asian countries (Ma et al., 2018). In China, coastal waters off Dongtou District, Wenzhou City, Zhejiang Province are the major production region (Fig. 1a). In recent years, farmers began to use zygote-derived seedlings to satisfy the market demands and to promote the commerce (Pang et al., 2006; Lin et al., 2020). To obtain the sexually propagated seedlings, some healthy and strong thalli were left at the aquaculture sites as parental strains during harvest waiting for the thalli to discharge germ cells (Pang et al., 2006; Lin et al., 2020). After fertilization, the eggs developed into oval zygotes within 18-h post-fertilization and into embryos in 24-h post-fertilization (Pang et al., 2006; Lin et al., 2020). A recent study by Lin et al. (2020) found that some juvenile sporophytes of maricultured S. fusiformis could produce matured receptacles (from August to November) that could also discharge germ cells for producing sexually propagated embryos. On the other hand, from May to September, HABs formed by dinoflagellate P. donghaiense occur frequently in the coastal waters of the Dongtou District (Shen et al., 2019). Small- to medium-sized HABs often break out in the aquaculture regions of S. fusiformis (Fig. 1b). In practice, farmers would use cotton strips with which S. fusiformis embryos or seedlings could be attached and collected, and then transferred into a non-HAB region to avoid further damage.

Although HAB inhibits the seaweed farming, which has received much scientific attention (Yang et al., 2015; Inaba et al., 2017), the adverse impact of HAB-forming species on macroalgae growth remain poorly studied (Ma et al., 2017, 2020; Shang et al., 2020). A recent study shows that co-cultivation with toxin-producing dinoflagellate K. mikimotoi inhibited the development, pigment content, and photosynthetic activities of S. fusiformis embryos but not the egg fertilization (Ma et al., 2020). Furthermore, the inhibitory effects are mainly caused by allelochemicals released from K. mikimotoi in a cell density-dependent manner (Shang et al., 2020). As the most important bloom-forming contributor, P. donghaiense could form dense blooms, although it is non-toxin species, it is regarded as a harmful species to maricultured S. fusiformis because the blooms formed by this species that has been found able to affect the plankton community structure (Lin et al., 2014; Chai et al., 2020) and to inhibit the growth of co-cultured microalgae (Shen et al., 2015). Unfortunately, at present, our knowledge of the impact of P. donghaiense blooms on S. fusiformis remains blank.

The sexually propagated embryos of S. fusiformis are more vulnerable than its mature sporophytes (Ma et al., 2020; Shang et al., 2020), and serious consequences such as a deficiency of sexually propagated seedlings and a subsequent reduction of commercial production may happen when the embryos suffered from P. donghaiense-induced HAB. Therefore, we performed a series of laboratory experiments to investigate the effects of P. donghaiense cell density gradients on the development and the photosynthetic ability of S. fusiformis embryos. In addition, the analysis of the O-J-I-P polyphasic rising transient (i.e., JIP-test) was used to quantify the photosystems (PSII and PSI) during the co-cultivation with the HAB-forming species.

2 MATERIAL AND METHOD 2.1 Cultivation condition

The dinoflagellate P. donghaiense was provided by Prof. Douding LU from the Second Institute of Oceanography, Ministry of Natural Resources of the People's Republic of China (MNR) based in Hangzhou, China. The dinoflagellate was grown in sterile f/2 medium (Guillard, 1975) that prepared with seawater (salinity ca. 29) from mariculture region of S. fusiformis. The cultivation was maintained at 22±1 ℃ under cold white light of 100 μmol photons/ (m²·s) in 12-h꞉12-h light-dark photoperiod cycle. All cultivations were manually shaken five times per day until the cells reached the exponential growth phase during which cells were harvested and used for follow-up experiments.

Mature S. fusiformis sporophytes were collected from mariculture field of the Dongtou Fisheries Science and Technology Research Institute in May 2019. Thalli were transported to laboratory in a cooler box within 2 h to avoid thermal and dehydration damage. Epiphytes were removed and the thalli were washed three times to clean other impurities.

To get the fertilized eggs, female and male mature thalli at germ cell release were cultivated together in a weight ratio of 5 to 1. The thalli were incubated in a plastic basin containing filtered natural seawater (salinity ca. 29) and kept in a illumination incubators (GXZ-300D, Ningbo, China) at 22±1 ℃ under cold white light of 100 μmol photons/(m²·s) in 12-h꞉12-h light-dark regime. Seawater was aerated with filtered ambient air (0.22 μm) at 0.5 L/min and the cultures were run for 24 h to ensure complete release and fertilization of eggs. Fertilized eggs were then harvested with a 270-mesh nylon filter (pore diameter=53 μm) and were re-suspended in seawater in beaker.

2.2 Co-cultivation of fertilized eggs (zygotes) of S. fusiformis and P. donghaiense

To investigate the effects of P. donghaiense suspensions on development and photosynthetic activities of the fertilized eggs, re-suspended fertilized eggs were sprinkled evenly on 6-well plates in density of ~30 eggs/cm². After 24-h post-fertilization, zygotes developed into embryos and attached to the bottom of 6-well plates with their filamentous pseudo-roots. To create the different treatments, sterile fresh f/2 medium and P. donghaiense suspensions that concentrated by centrifugation were added to the wells with 6-mL suspension in each well. The final cell densities of P. donghaiense in the wells were 0 (the control), 0.50×1.05, 0.75×10⁵, 1.00×10⁵, and 1.50×10⁵ cells/mL in triplicate on 6-well plates. In addition, to eliminate the effects of nutrition limitation, a semi-continuous cultivation were conducted during the experiment, i.e., the cultivations were renewed to the initial cell density of P. donghaiense every other day by adding fresh f/2 medium to dilute the suspensions according to the actual cell densities. These mono- and co-cultivated embryos were grown at 22 ℃ under cold white light at 100 μmol photons/ (m²·s) in 12-h꞉12-h light-dark scheme for 10 days. The morphological changes of the embryos cultivated under each treatment were examined every other day and their photosynthetic activities detected by rapid light curves, and chlorophyll-a (Chl-a) fluorescence transient analysis (i.e., OJIP) were determined on the 10^th day post-inoculation.

2.3 Observation of morphology and determination of growth of the embryos

The morphological changes of the embryos were examined with an ECLIPSE Ts2 inverted routine microscope (Nikon, Japan) and the digital images were captured with a digital microscope camera (Nikon DS-Ri2). To collect continuous information on the development of the embryos, the same locations of the wells were examined every other day. The lengths and widths of 150 embryos under each treatment, not including the rhizoid lengths, were measured with the Nikon software NIS Elements Ar 3.0 (Nikon, Japan). The embryo volumes were calculated as cylinders because of their cylindrical shape (Shang et al., 2020), and the relative growth rate (RGR, /d) was determined according to the equation of Glenn and Doty (1992): RGR=(lnV₂– lnV₁)/(t₂–t₁), where V₁ and V₂ are the volumes at time t₁ and t₂, respectively.

2.4 Determination of photosynthetic activity of S. fusiformis embryos

To obtain photosynthetic capacity information of the embryos cultivated in the suspensions at different cell densities, the photosynthetic fluorescence parameters of the embryos were detected using a Multi-color-PAM fluorometer (Heinz Walz GmbH, Effeltrich, Germany). Before measurements, embryos were washed more than three times with filtered seawater (0.45 μm) to completely remove P. donghaiense cells attached. Embryos were then scraped from the bottom of the wells and were resuspended in filtered seawater, of which 3-mL sample was used to determine the fluorescence parameters. For measurements of all the photosynthetic parameters, each sample was acclimated to darkness for 15 min to ensure all reaction centers were in open state, and non-photochemical dissipation of excitation energy was in minimum. The minimum fluorescence (F_o) was excited by a very low intensity of measuring light; the maximum fluorescence (F_m) was excited by a saturating flash in a fixed emission wavelength at 480 nm. The variable fluorescence (F_v) was defined as F_v=F_m–F_o and the F_v/F_m was calculated as per Kitajima and Butler (1975): F_v/F_m=(F_m–F_o)/F_m.

Rapid light curves, i.e., the relative electron transport rate (rETR) versus irradiance (I) curves were built under gradient photosynthetically active radiation (PAR) levels (each measurement lasted for 20 s). The parameters of the rETR vs. I curves were analyzed as per Eilers and Peeters (1988): rETR= I/(aI²+bI+c), in which a, b, and c are adjustment parameters. The efficiency of electron transport (α), the saturating light intensity for the photosynthesis (I_k), and the maximum relative electron transport rate (rETR_max), are expressed as functions of the parameters a, b, and c, i.e., α=1/c, I_k=(c/a)^1/2, and rETR_max= 1/[b+2(ac)^1/2].

2.5 Chlorophyll-a fluorescence transient analysis of S. fusiformis embryos

The chlorophyll-a fluorescence (i.e., OJIP) transient, known commonly as the Kautsky curve and shown by photosynthetic organisms under different treatments, can provide detailed information of the structure, conformation, and function of the photosynthetic apparatus, especially the PSII (Strasser et al., 2004). The transients (10 μs) were detected using Multi-color-PAM fluorometer (Heinz Walz GmbH, Effeltrich, Germany) after 15-min darkness acclimation, and analyzed in the JIP-test theory and application to reveal information about the fluxes of photons, excitons, electrons, and further metabolic events (Appenroth et al., 2001; Strasser et al., 2004). The Chl-a fluorescence curves includes the following points: (1) the fluorescence in 50 μs, called F_o, represents the O step of the fluorescence curve, during which all the reaction centers (RCs) were open; (2) after F_o, the photons were transported to primary quinone electron acceptor (Q_A) in 2 ms, when the fluorescence was called J step in fluorescence curve; (3) in the process of photons transported from Q_A to Q_B (the secondary quinone electron acceptor) in 30 ms, when the fluorescence was called Ⅰ step; (4) lastly, a maximum transient in the curve was obtained, it was called P step and at this time the Q_A was fully reduced.

V_J, the relative variable fluorescence intensity at the J step, and M_o, the initial slope of the OJIP curve, were calculated as V_J=(F_{2 ms}–F_o)/(F_m–F_o) and M_o=4×(F_{300 ms}–F_o)/(F_m–F_o). F_{2 ms} and F_{300 ms} represented the fluorescence values at 2 and 300 ms, respectively. The quantum efficiencies of electron transport chain (ETC) included parameters φ_Po, ψ_o, and φ_Eo: φ_Po represent the trapped electrons per absorption, i.e. maximum quantum yield for photochemistry (F_v/F_m); ψ_o means the efficiency that a trapped excition could move an electron into the ETC beyond the primary acceptor plastoquinones (Q_A), i.e. the yield of electron transport of trapped excition; and φ_Eo means the quantum yield of electron transport. These parameters were calculated by the following equations: φ_Po=(1– F_o)/F_m, ψ_o=1–V_J, and φ_Eo=(1–F_o/F_m)×ψ_o.

The specific energy fluxes ratios in electron transport chain included ABS/RC, TR_o/RC, ET_o/RC, and DI_o/RC, representing the energy fluxes ratio, the trapped energy flux ratio, the electron transport flux, and the dissipated energy flux ratio per reaction center, respectively. These ratios were calculated by the following equations: ABS/RC=M_o×(1/V_J)×(1/φ_Po), TR_o/RC=M_o×(1/V_J), ET_o/RC=M_o×(1–V_J)×φ_Po, and DI_o/RC=(ABS/RC)–(TR_o/RC).

Furthermore, the energy flux ratios were represented in terms of "per absorption (/ABS)", including RC/ABS, TR_o/ABS, ET_o/ABS, and DI_o/ ABS, representing the reaction center of PSII, the trapped energy flux ratio, the electron transport flux, and the dissipated energy flux ratio per absorption, respectively. They were derived from the energy fluxes ratio representing in terms of "per active reaction center of PSII (/RC)".

The performance index (PI) is an indicator to vitality, and associated with three independent parts contributing to the photosynthesis: the amount of photosynthetic reaction centers of PSII (RC/ABS); the contribution of the light reactions to primary photochemistry, i.e., the performance due to the trapping probability (P_TR); and the contribution of dark reactions, i.e., the performance due to the conversion of excitation energy to electron transport (P_ET) (Thach et al., 2007). PI can be calculated as per Strasser et al. (2000): PI=(1–F_o/F_m)/(M_o/V_J)×(F_m–F_o)/F_o×(1–V_J)/V_J.

2.6 Determination of the cell density, nutrient, and pH of the P. donghaiense suspensions during the mono- and co-cultivations

The cell number of P. donghaiense suspensions was counted under optical microscope (BX43, Olympus, Japan), and the cell density was determined using the cell numbers in 0.1-mL suspension. In addition, to address the concentration change of nutrients among suspensions during the co-cultivation, the total dissolved nitrogen (TN) and total dissolved phosphorus (TP) were determined every other day. The suspensions were filtered with 0.45-μm cellulose acetate membranes immediately after sampling and the filtrates were used for TN and TP determination. The TN concentration was measured in the high-temperature catalytic oxidation (HTCO) method using automatic TOC-VCPH analyzer (Shimadzu, Japan). The TP concentration was analyzed in phosphomolybdate-blue spectrophotometry. All the TN and TP concentration were measured in triplicate for each treatment. The pH values of the suspensions during the co-cultivation period were measured using a HACH Hq40d multi pH meter.

2.7 Statistical analysis

All data are expressed as the mean and standard deviation. After the testing for normal distribution and homogeneity of variance, one-way analysis of variance (ANOVA) with post-hoc tests (Tukey) were conducted to reveal differences among treatments, and the significance level was set at P=0.05. Statistical analysis was performed using SPSS V.16.0 for Windows (SPSS Inc., Chicago, IL, USA), and all charts were generated using Origin 9.0 (OriginLab, USA).

3 RESULT 3.1 Effect of P. donghaiense cell density on growth of S. fusiformis embryos

The zygotes (fertilized cells) developed into embryos in 24-h post-inoculation by continuous cell division, during which the embryos attached to the well bottom with their filamentous pseudo-roots. In addition, the morphology of the embryos changed from being elliptical to cylindrical in shape by gradually elongating along the long axis, and the broader axis was widened to form a thallus at last (Fig. 2). No significant morphological changes of the embryos could be observed between the mono- and co-cultivations except for the sizes (Fig. 2). When volumes of the embryos were determined as cylinders, the sizes of the embryos co-cultivated with P. donghaiense in the suspensions at 1.00×10⁵ and 1.50×10⁵ cells/mL began to decrease significantly (P < 0.05) below the monocultivated ones on the 10^th day post-inoculation.

On the 10^th day post-inoculation, the volume of the monocultivated embryos was 31.23×10^-3 mm³ while those cultivated in P. donghaiense suspensions at 0.50×10⁵, 0.75×10⁵, 1.00×10⁵, and 1.50×10⁵ cells/mL were 28.54×10^-3, 25.80×10^-3, 22.36×10^-3, and 12.05×10^-3 mm³, respectively (Fig. 3a). In addition, the embryo size decreased with increased cell density of P. donghaiense. The corresponding average RGR of the embryos during the cultivation period was 0.26/d for the mono-cultivation, and 0.25, 0.24, 0.23, and 0.16/d for the co-cultivation, respectively (Fig. 3b). Meanwhile, after 10-day co-cultivation at different cell densities mentioned above, the embryo volume was inhibited from that of the control by 8.61%, 17.39%, 28.40%, and 61.42%, and the RGR by 3.45%, 7.35%, 13.06%, and 37.08%, respectively.

3.2 Effect of P. donghaiense cell density on photosynthetic activity of S. fusiformis embryos

The rapid light curves (RLCs) of the S. fusiformis embryos co-cultivated with P. donghaiense are shown in Fig. 4a, and the critical photosynthetic parameters α and rETR_max derived from the RLCs are shown in Fig. 4b–c. Compared with the control, the co-cultivation resulted in significant (P < 0.05) inhibition on α and rETR_max of the embryos. To be compared, α of the monocultivated embryos was 0.27 electron/ photon, which was greater than those of the co-cultivated in cell densities of 0.50×10⁵, 0.75×10⁵, 1.00×10⁵, and 1.50×10⁵ cells/mL at 0.25, 0.23, 0.20, and 0.20 electron/photon (Fig. 4b), or in percentage of decrease by 7.21%, 14.52%, 26.47%, and 26.65%, respectively. In addition, the rETR_max value of embryos was 162.24 μmol electrons/(m²·s) in the mono-cultivation, and 143.33, 119.08, 108.41, and 95.84 μmol electrons/(m²·s) in the co-cultivation (Fig. 4c), respectively, or in percentage of decrease by 11.66%, 26.60%, 33.18%, and 40.87%, respectively.

3.3 Effect of P. donghaiense cell density on fast chlorophyll-a fluorescence of S. fusiforms embryos

The fast chlorophyll-a fluorescence induction kinetics of the mono- and co-cultivated S. fusiformis embryos are shown in Fig. 5a. To visualize the changes of Chl-a fluorescence, the averages of the raw fluorescence kinetics plotted on a logarithmic time scale from 50 μs to 1 s are shown in Fig. 5b. The Chl-a fluorescence curves started from the initial fluorescence F_o (50 μs) in intensity and finally increased to the peak F_m, and between them, two intermediate steps F_J (about 2 ms) and F_I (about 30 ms) could be observed (Fig. 5a). In the raw fluorescence kinetics, the average F_o and F_m intensities of the co-cultivated embryos decreased with increased cell densities of P. donghaiense (Fig. 5a). However, when the OJIP-curves were plotted with variable fluorescence (V_t), the V_J and V_I intensities of the co-cultivated embryos increased with the increase in cell density of P. donghaiense and were significantly greater than that of the monocultivation (Fig. 5b), although the value of F_m intensity in the raw fluorescence kinetics was lower than that of the monocultivation (Fig. 5a).

3.4 Effect of P. donghaiense cell density on the behavior of PSII in S. fusiforms embryos

The behavior of the photosynthetic parameters that characterize the PSII functioning varied with the P. donghaiense cell densities in the co-cultivation (Fig. 6). All the parameters that derived from the OJIP-curves of the co-cultivated embryos differed significantly (P < 0.05) from those of the monocultivation (Fig. 6; Table 1). For instance, the value of M_o that reflects the accumulation rates of inactive RCs in embryos, increased with the increase in the cell density. In addition, all the values of M_o in co-cultivated embryos were significantly (P < 0.05) higher than those of the monocultivation (Fig. 6; Table 1). Meanwhile, the values of V_J in the co-cultivated embryos also increased with increased cell density of P. donghaiense, which indicated an increase in the proportions of closed RCs at J step during the co-cultivation (Fig. 6; Table 1). In addition, compared with monocultivated ones, all the photosynthetic parameters (φ_Po, φ_o, and φ_Eo) of co-cultivated embryos significantly (P < 0.05) decreased (Fig. 6; Table 1).

Performance index (PI) was decreased in all co-cultivated embryos and the inhibitory effects increased with increased cell density of P. donghaiense. Compared with the control, co-cultivation with P. donghaiense in cell densities of 0.50×10⁵, 0.75×10⁵, 1.00×10⁵, and 1.50×10⁵ cells/mL inhibited the PI by 33.29%, 60.83%, 72.97%, and 70.77% (Fig. 7a & b), respectively. Furthermore, among the parameters, P_TR, RC/ABS, and P_ET contributed to the reduction of PI by 14.26%–40.92%, 7.56%–22.53%, and 21.79%–40.05% (Fig. 7b), respectively. In addition, the density of PSII reaction center (RC/ABS), maximum quantum yield (TR_o/ABS), and the yield of electron transport (ET_o/ABS) of the embryos cultivated in P. donghaiense suspensions at 0.75×10⁵, 1.00×10⁵, and 1.50×10⁵ cells/ mL decreased significantly (P < 0.05) compared with those of the monocultivated embryos. It indicated that co-cultivation with P. donghaiense had a variety of inhibitory effects on PSII of the embryos, including inhibition on the average absorption per reaction centers, the dark reactions in electron transport (P_ET), and the light-dependent reactions (P_TR) (Fig. 8a). The ET_o/RC showed a similar trend to that of the fluorescent transient per absorption (/ABS), whereas the TR_o/RC have no significant difference between control and treated embryos, in general. Meanwhile, the ABS/RC of the co-cultivated embryos, which characterizing the average antenna size, increased significantly (P < 0.05) compared with those the monocultivated (Fig. 8b).

3.5 Changes in nutrients and pH of P. donghaiense suspensions during the mono- and co-cultivations

During the co-cultivation, the TN and TP concentrations of all cultivations are shown in Fig. 9a–e & f–j. At the inoculation time, the TN and TP concentrations of all cultivations were 12.67 and 0.13 mg/L, respectively. Two days after inoculation, TN concentrations in the monocultivation decreased by 28.38%, and so did in the co-cultivation at P. donghaiense cell densities of 0.50×10⁵, 0.75×10⁵, 1.00×10⁵, and 1.50×10⁵ cells/mL by 34.48%, 33.91%, 35.25%, and 33.95%, respectively (Fig. 9a–e). Correspondingly, the TP concentration decreased by 48.90% in the monocultivation, and 54.00%, 58.10%, 69.30%, and 60.70%, respectively in the co-cultivation (Fig. 9f–j). In addition, both TN and TP concentrations in mono- and co-cultivations changed periodically in nearly the same trends during the whole period.

During the inoculation, the pH value of the f/2 medium in mono-cultivation was 8.29, while those of the co-cultivation media, i.e., the P. donghaiense suspensions in cell density of 0.50×10⁵, 0.75×10⁵, 1.00×10⁵, and 1.50×10⁵ cells/mL were 8.12, 8.09, 8.05, and 8.02 (Fig. 9k–o), respectively. Two days after inoculation, the pH value of the f/2 medium decreased to 7.92, and those of the co-cultivation at the cell density gradient increased to 8.17, 8.23, 8.29, and 8.36, respectively (Fig. 9k–o). Furthermore, pH values of all the media varied periodically in nearly the same trends throughout the whole experimental time. For P. donghaiense suspensions in co-cultivation, the cell densities in all set gradients increased about 20% within every 2 days in the semi- continuous culture mode (Fig. 9p–s).

4 DISCUSSION

During a bloom, the recorded cell densities of Prorocentrum species usually varied in a range of 104–106 cells/mL (Lu et al., 2005). In this study, the cell density of P. donghaioense ranged 0.50×10⁵– 1.50×10⁵ cells/mL, close to the normal cell densities during blooms but lower than the highest cell density (3.6×108 cells/mL) reported by Lu et al. (2005), implying that embryos in mariculture sites might have experienced the HABs caused by P. donghaioense at similar cell densities, because they have occurred frequently in the coastal waters of Dongtou District (Shen et al., 2019). The exuded compounds of P. donghaioense could alter the physiology of its competitors by changing their lipid synthesis, cell membrane integrity, and photosynthesis (Poulin et al., 2018). In addition, in a recent research, we found that the inhibitory effects of K. mikimotoi on the S. fusiformis embryos followed a cell density-dependent manner, and they were mainly caused by allelochemicals (Shang et al., 2020). In this study, the inhibitory effects of P. donghaiense on the development (Fig. 2) and photosynthesis (Fig. 4b–c) of S. fusiformis embryos were observed when the cell densities were at 1.00×10⁵ and 1.50×10⁵ cells/mL, and the inhibitory effect increase was observed at higher cell densities, suggesting the inhibitory effects also followed a cell density-dependent manner.

Performance index (PI) was considered a multi-parametric expression of the functional and structural criteria of PSII (Strasser et al., 2000, 2004), which sharply decreased when the cell density of P. donghaiense arrived at 0.50×10⁵ cells/mL (Fig. 7a–b). Moreover, it has been known to be more sensitive to environmental change and correlated well with plant vitality (Hermans et al., 2003). The inhibition on PI parameters indicated that RCs of the co-cultivated embryos were inactivated and the photosynthetic capacities were inhibited. In chlorophyll fluorescence curves, the fluorescence amplitude in O-J phase represents the energy used to reduce Q_A (quinone acceptor) to Q_A^–. The increases in the fluorescence amplitude in V_J step and O-J phase were observed in the co-cultivated embryos (Fig. 5b), which reflected that the electron transport was stocked in Q_A^– because of the photosynthetic elements damage (Kalaji et al., 2012; Zivcak et al., 2014). The following J-I phase (thermal phase) reflected the reduction of the plastoquinone (PQ) pool by reduced Q_A (Schansker et al., 2011; Vredenberg, 2015). Thus the amplitudes in J-I phase reflected the electron transported from Q_A^– to Q_B (the secondary quinone electron acceptor), and the V_I fluorescence reflected the reduction of the PQ pool (Schreiber, 1998; Zivcak et al., 2015). The increases of fluorescence in V_I and J-I phase of the co-cultivated embryos (Fig. 5b) indicated the limitation of electron transport between Q_A and other electron accepters (Q_B, PQ pool), and then the limitation of the electron transport to terminal electron acceptor (PSI reaction center).

The electron transport parameters φ_Po, ψ_o, and φ_Eo of the co-cultivated embryos in P. donghaiense suspensions at all the cell densities decreased significantly (P < 0.05), indicating that the overall quantum efficiency of their photochemical reactors were inhibited, which differed from those of the monocultivated embryos (Fig. 6; Table 1). The energy flux ratios in the process of electron transport provide information of the structure and function of photosynthetic apparatus (Liu et al., 2017; Samborska et al., 2018). In addition, the value of ABS/RC represents the absorption energy per active RC (Appenroth et al., 2001), and the antenna size of the light-harvesting complexes (LHC) per PSII RC complex. In this study, the ABS/RC of the co-cultivated embryos increased significantly (P < 0.05) above those of the monocultivated ones (Fig. 8), indicating that the antenna size of RCs was enlarged and a part of the RCs was inactivated, by which the excitation energy was transferred from the inactive centers to other PSII active units, converted to heat dissipation, and finally increased the DI_o/RC and DI_o/ ABS (Fig. 8a–b).

In this study, the values of ET_o/RC, TR_o/ABS, and ET_o/ABS of the co-cultivated embryos were significantly (P < 0.05) less than the monocultivated ones' (Fig. 8a–b). Among these ratios, the trapped energy (TR_o) was used to restore the Q_A to Q_A^–, and then the electron transport (ET_o), indicating that the photosynthetic electron transport was limited and the energy was used for the reduction of PSII acceptors (Q_A, Q_B, and PQ pool), thus resulting in the decrease of ET_o (Fig. 6; Table 1). The decreases of TR_o/ABS, ET_o/RC, and ET_o/ABS were consistent with the results of JIP-test, showing that the inhibitory effects of P. donghaiense on S. fusiformis embryos were resulted from a comprehensive damage on their photosynthetic systems. Furthermore, the parameters RC/ABS, P_TR, and P_ET were constitutes of the PI, among which the dark reaction (P_ET) contributed largely (21.79%–52.28%) to the PI decline (Fig. 7a–b), thus the dark reaction of photochemistry was inhibited mostly.

The main mechanism of the inhibitory effects between algae was due probably to strong competition for nutrient and the excretion of allelopathic substances (Yang et al., 2015; Ma et al., 2017; Shang et al., 2020). In this study, f/2 media used for the cultivations were rich in N and P nearly 30 times of the natural seawater's, and the media for S. fusiformis monocultivation and K. mikimotoi-S. fusiformis co-cultivation were renewed every day during the experiment. In addition, in both cultivations, the TN changed from 12.67 to 8.46 mg/L and the TP from 0.13 to 0.06 mg/L, showing same trends in every 2 days (Fig. 9a–e, f–j). Therefore, nutrients in all the cultivations were sufficient for the demand of seedlings growth, thus the inhibitory or even lethal effects by nutrient deficiency were unlikely. Moreover, the high pH value has also been taken into account in the allelopathic effects of some autotrophs (Schmidt and Hansen, 2001; Lundholm et al., 2005). Whereas the pH values of both the cultivations ranged from 7.84 to 8.40, and that of the P. donghaioense suspension of the co-cultivation was greater than that of the f/2 media in the monocultivation (Fig. 9k–o), suggesting that the inhibition effects on embryos were irrelevant to the pH. Therefore, it is reasonable to assume that inhibition to the growth and photosynthesis of embryos was caused by allelochemicals that exuded from P. donghaioense.

In the nature, interaction between HAB-forming species by cell contact was thought the key way of promoting the bloom formation (Yamasaki et al., 2007). Uchida et al. (1999) reported that the interaction between K. mikimotoi and Heterocapsa circularisquama depended on the initial cell densities of the two species. In addition, the growth inhibition and the formation of morphologically abnormal cells of Akashiwo sanguinea were induced in a concentration-dependent manner by allelochemicals released from H. akashiwo (Qiu et al., 2012). In a recent study, it was found that the suppression on pigment contents and photosynthetic activities of S. fusiformis embryos by K. mikimotoi were also induced in a cell density-dependent manner (Shang et al., 2020). It has been reported that the secondary substance released from P. donghaiense suppressed the growth of other microalga in bi-algal cultivations (Wang and Tang, 2008; Shen et al., 2015). In China, current mariculture of S. fusiformis depends almost entirely on sexually propagated seedlings (Lin et al., 2020), and thus any adverse effects on the development and growth of the embryos could seriously threaten the seedling stock. In this study, exposure of S. fusiformis embryos to dense P. donghaiense suspensions led to significant inhibition to the growth and photosynthetic activities. In addition, the inhibitory effects of P. donghaiense suspensions on S. fusiformis embryos followed a cell density-dependent manner. Therefore, during the seedling-breeding period, the HABs formed by dinoflagellate P. donghaiense at high cell densities in S. fusiformis farming region could seriously threaten the seedling stock, which would eventually hinder its farming industry.

5 CONCLUSION

The inhibitory effects of P. donghaiense suspensions on growth and photosynthetic activities of S. fusiformis embryos increased with increased cell densities in the suspensions. In addition, the JIP-test analysis showed that the decreases of the maximum relative electron transport rates (rETR_max) and the apparent photosynthetic efficiency (α) in co-cultivated embryos were due to the limitation of electron transport between electron transfer accepters and the inactivation of partial reaction centers, which promoted the conversion of excitation energy to heat dissipation. These results indicate that dense algal bloom formed by P. donghaiense is a threat to S. fusiformis farming industry.

6 DATA AVAILABILITY STATEMENT

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.