1 INTRODUCTION

The frequency of green and golden seaweed tides, mainly caused by the seaweed taxa Ulva and Sargassum, has increased in the past decades because of climate change and human activities (Smetacek and Zingone, 2013). Unlike most seaweeds, which rely on hard substrata, floating species increase their biomass by making new thalli and/or offspring during the floating period (Komatsu et al., 2008; Wang et al., 2015). Positive buoyant seaweeds and hitchhiking species extend their distribution range by long-distance rafting with the currents (Macreadie et al., 2011; Bertola et al., 2020). Thus, the movement and longevity of floating seaweed rafts may affect the connectivity of coastal ecology (Rothäusler et al., 2015).

The longevity of floating seaweeds is determined by the balance between their acclimation potential and the biotic/abiotic stress at the sea surface (Tala et al., 2017). The survival of floating seaweeds depends on several biotic and abiotic factors, such as temperature, photosynthetically active radiation (PAR), ultraviolet radiation (UVR), and grazing animals (Rothäusler et al., 2011a). High photo-acclimation potential reflects the ability to minimise the damage caused by ambient light conditions and to retain photosynthetic responses to the changing environment (Wu et al., 2015). This property is important for floating seaweeds, as at the sea surface, seaweeds are exposed to a combination of high light intensity stress, UVR, and changing sea surface temperature, which are distinct from those in benthic habitats (Zhao et al., 2019). UVR is another important factor that affects the photosynthetic performance and survival of intertidal and floating seaweeds (Koch et al., 2016; van Hees et al., 2019). Some seaweed produce UV-absorbing compounds, such as mycosporine-like amino acids (red and green seaweeds) and phlorotannins (brown seaweeds), to avoid damage caused by UVR (Quintano et al., 2019). To cope with high levels of PAR and UVR, floating seaweeds employ some photoprotective strategies; for example, accumulation of UV-absorbing compounds (van Hee et al., 2019), dissipation of excess light energy as heat (Rothäusler et al., 2018), and reducing the content of accumulated pigments (Zhao et al., 2016).

Sargassum horneri (Turner) C. Agardh is a common macroalgal species found along the coast of China that forms extensive subtidal forests, especially in offshore islands in the Northwest Pacific (Komatsu et al., 2008; Li et al., 2020). Large floating patches of S. horneri are often found on the continental shelf west of the Kuroshio Current. Previously, the bloom of S. horneri in 2017 in the Yellow Sea of China was attributed to high water temperature and increased light availability (Qi et al., 2017). In Bohai Bay, vesicles of S. horneri are often found in September (autumn boreal, field observation), and patches of floating thalli appear from September to May the following year. After the reproductive stage, the majority of plants begin to age and lose biomass during the time from May to August. Remote sensing data indicate that floating S. horneri patches in Bohai Bay drift southwards with the currents in winter (Xing et al., 2017). Yatsuya (2008) reported that S. horneri, which detached off the central part of Honshu Island facing the Sea of Japan in January, stayed floating for more than three months. Based on this study, the floating period of S. horneri is much longer than that of the other three Sargassacean species, including Myagropsis myagroides (Mertens ex Turner) Fensholt, Sargassum patens C. Agardh and Sargassum siliquastrum (Turner) C. Agardh (Yatsuya, 2008). The longevity of floating thalli of these species is also greatly affected by the reproductive stage, as thalli begin shedding and withering after sexual reproduction. Thalli of S. horneri detached before maturation had a floating period of 8–14 weeks, while those detached during and after maturation had a floating period of 3–8 weeks and 2 weeks or less, respectively (Yatsuya, 2008).

Our experiments on S. horneri were conducted from October to December, which is the rapid growing stage of the species to avoid the effects of reproductive development (Yu et al., 2019). Because the floating and benthic thalli of S. horneri occupy different ecological niches characterised by contrasting environmental factors, particularly light intensity, we investigated whether these two thalli types exhibit different photo-acclimation potentials to different light intensities. Both thalli types were cultivated at different depths in situ and under different light intensities in the laboratory to elucidate the photo-acclimation mechanisms of this golden tide species.

2 MATERIAL AND METHOD 2.1 Sample collection

All S. horneri samples were collected from Changdao Island, China (37°54′N, 120°43′E) on October 20, 2018. Approximately 20 S. horneri adults, ranging in size from 0.6–1 m, were detached from a nearby Sargassum forest by scuba diving to a depth of 2–3 m. All benthic thalli were cut from the bottom (without holdfast) with knife. Approximately 20 free-floating thalli were collected from the floating mat near the shore (ca. 17 km from the coast).

2.2 Experimental design 2.2.1 In-situ experimental design

The in-situ experiment was conducted in a kelp aquaculture farm in a protected bay. Each thallus was rinsed in seawater to remove epiphytes and sediments. The rinsed thalli were twisted into nylon ropes and suspended in the sea horizontally at 0 m above sea level (masl) or 3 m below sea level (mbsl) using rope and float-balls. Four treatments were established (n=12 for each treatment; Fig. 1a): floating thalli cultured at 0 masl (F0); benthic thalli cultured at 0 masl (B0); floating thalli cultured at 3 mbsl (F3); and benthic thalli cultured at 3 mbsl (B3). The suspended thalli were cultured in the sea from 24 October 2018 to 24 December 2018.

Hydrological parameters, including temperature, salinity, dissolved oxygen, and pH, at each depth were measured every month using an YSI probe (Professional Plus; YSI Inc., USA) (Supplementary Table S1). Light intensity was measured daily (early morning) throughout the culture period at a location near the site of the in-situ experiments using a micro-quantum sensor associated with a diving-pulse amplitude modulated (PAM) fluorometer (WALZ Germany) (Supplementary Fig.S1). The light intensity at 3 mbsl was calculated using the following equation:

where I_z is the light intensity at depth Z; I₀ is the surface light intensity; and K is the light extinction coefficient (K=1.83, calculated using the Secchi disk depth) (Supplementary Fig.S1).

2.2.2 Indoor experimental design

Benthic and floating thalli were rinsed in seawater to remove epiphytes and sediments. The rinsed thalli were stored in a foam plastic box containing crushed ice to control the temperature, and then transferred to the laboratory. Indoor experiments were conducted to compare the responses of benthic and floating thalli to high light intensity. Based on the measurement of light intensity in situ (ranging from 300 to 500 μmol photons/(m²·s) (Supplementary Fig.S1), high light intensity was set at 400 μmol photons/(m²·s). Both types of thalli were cultured in 100-L plastic tanks containing filtered seawater for 2 d (pre-incubation). The seawater was continuously aerated with an air pump, and thalli were cultured in a growth chamber at 15 ℃ under 10-h light/14-h dark photoperiod and 40 μmol photons/(m²·s) light intensity (PAR). Then, fragments (ca. 8 cm each) were obtained from the upper portion of each thallus and cultured in sterile filtered seawater in two growth chambers (GXZ-280C; Ningbo Jiangnan Instrument, China) maintained at 15 ℃, 10-h light/14-h dark photoperiod and either 400 μmol photons/(m²·s) (high-light group; n=20) or 40 μmol photons/(m²·s) (control group; n=20) (Fig. 1b). The filtered seawater was replaced every 2 d. After 21 d, thalli from the high-light group were allowed to recover by placing under 40 μmol photons/(m²·s) light intensity for 6 d.

2.3 Relative growth rates (RGRs)

The length of each thallus (from the base to the top) was measured on 24 October, 24 November, and 24 December during in-situ experiments. The RGR (%/d) in each month was calculated using the following equation (Rueness and Tananger, 1984):

where L₀ is the initial thallus length; L_t is the final length after t days; and t is the number of days.

2.4 Biochemical analyses

Before and after the in-situ experiments, thalli fragments (measuring ~5 cm) were cut from the top of each individual thallus, stored in a box containing crushed ice and then transferred to the laboratory for biochemical analyses. Chlorophyll a (Chl a) and carotenoid (Car) contents were estimated as described previously (Miki et al., 2016). Approximately 500 mg of thallus (wet weight, ww) was ground thoroughly in 5 mL of 90% acetone and then centrifuged. After storage in the dark at 4 ℃ for 24 h, the absorbance of the supernatant was measured at 470, 663, and 646 nm (A₄₇₀, A₆₆₃, A₆₄₆). Pigment concentrations (mg/g ww) were determined according to the following equations (Wellburn, 1994):

The soluble phlorotannin contents of freeze-dried seaweed were extracted and measured using the Folin-Ciocalteu method (van Alstyne, 1995). Tissue (10 mg, dry weight, dw) was ground to fine powder in 3 mL of 80% acetone and extracted overnight. Approximately 300 μL of extract (diluted 1꞉5 with reagent-grade water) was added to 100 μL of the Folin-Ciocalteu reagent (Sigma-Aldrich, Seelze, Germany), and the sample was mixed by shaking for 5 min. The solution was made alkaline by adding 200 μL of 20% sodium carbonate (NaCO₃), and shaken then for an additional 30 s. Subsequently, the sample was incubated in the dark at room temperature (RT) for 45 min and then centrifuged at 2 000×g for 3 min at RT. The absorbance of the supernatant was measured at 730 nm using a microplate spectrophotometer (Eon, BioTek, USA). Purified phloroglucinol (Sigma-Aldrich, Seelze, Germany) was used as a standard. Values were expressed as mg/g dw.

2.5 Chl-a fluorescence measurements

The Chl-a fluorescence of samples cultured in situ and in indoor experiments was measured early in the morning during the first 2 h of the light period. Prior to measuring Chl-a fluorescence, samples were adapted to the dark for 20 min. Chl-a fluorescence (Schreiber et al., 1995) was measured using a Diving-PAM fluorometer (WALZ, Germany). The device employs a red light emitting diode (LED) with emission maximum at 650 nm to excite fluorescence.

The maximum quantum yield (F_v/F_m) of PSII was calculated using the following equation:

where F₀ is the initial fluorescence determined under 3–7 μmol photons/(m²·s) light intensity; F_m is the maximum fluorescence induced after a saturating pulse (5 640 μmol photons/(m²·s)).

The coefficient of photochemical quenching (qP) (Schreiber et al., 1986; Kramer et al., 2004) was measured and calculated as follows:

where F is the steady-state fluorescence at a given actinic light (104 μmol photons/(m²·s)); F_m′ is the maximal fluorescence induced after a saturating pulse (5 640 μmol photons/(m²·s)); F₀′ is the minimal fluorescence induced after a 3-s pulse of far-red light (735 nm; ~1 μmol photons/(m²·s)).

The effective PSII quantum yield [Y(II)] (Schreiber et al., 1995) was calculated as follows:

Once in-situ measurements for each treatment were completed, the samples were immediately returned to the seawater.

After 20-min dark adaptation, rapid light curves (RLCs) were measured with a light intensity gradient, comprising nine different light intensities (0, 104, 186, 322, 463, 621, 893, 1 189, and 1 754 μmol photons/(m²·s)), using a fluorometer (Diving-PAM; Walz, Germany); each light intensity was maintained for 10 s. The relative electron transport rate (rETR; μmol·electrons/(m²·s)) was calculated according to the following equation (Genty et al., 1989):

where PAR is the photosynthetically active radiation (μmol photons/(m²·s)), and F_II is the fraction of chlorophyll associated with PSII being 0.8 in brown seaweeds (Orzymski et al., 1997).

Photosynthetic parameters, namely, maximum relative electron transport rate (rETR_m) and light utilisation efficiency (α) were derived from RLCs using the following equation (Platt and Gallegos, 1980):

where PPFD represents the photosynthetic photon flux density of RLCs.

The light saturation point (E_k) was calculated as follows:

During the measurement of RLC, non-photochemical quenching (NPQ) (Bilger and Björkman, 1990) was calculated as follows:

where F_m and F_m′ values were obtained from the RLCs.

2.6 Statistical analysis

Prior to all statistical analyses, the homogeneity of variances was verified with Levene's test. Statistical significance of the in-situ data (RGRs, photosynthetic parameters and Chl a, carotenoid, and phlorotannin contents) was analysed by two-way analysis of variance (ANOVA), with S. horneri including thallus type (benthic and floating) and sampling depth as fixed factors. For significant differences, a post-hoc Tukey's honestly significant difference (HSD) test was applied. Statistical significance of the variation in photosynthetic parameters of indoor experiments was analysed with repeated measures ANOVA (RM-ANOVA), followed by Tukey's HSD post-hoc test. Statistical significance was considered at P < 0.05. All statistical analyses were conducted using the statistical software package SPSS 24.0

3 RESULT 3.1 In-situ experiment and environmental conditions

Light intensity declined dramatically with an increase in depth (Supplementary Fig.S1). No significant differences were observed in sea temperature, pH and salinity at different depths (Supplementary Table S1). The sea temperature decreased from October (ca. 17.6±1.4 ℃) to December (ca. 4.9±1.3 ℃) (Supplementary Table S1). The dissolved oxygen level increased from October to December at 0 masl (from 6.9 to 12.0 mg/L) and 3 mbsl (from 7.0 to 9.1 mg/L).

In most treatments, the RGR was lower in December than in November (Fig. 2). The results of ANOVA revealed a significant interaction between sampling depth and thallus type in each month (November: F=936.658, P < 0.001; December: F=91.954, P < 0.001) (Supplementary Table S2). Both types of S. horneri thalli at each depth showed positive RGRs, except the floating thalli at 3 mbsl (F3) (Fig. 2). In November, the RGRs of floating thalli cultured at 0 masl (F0) were approximately 2-fold higher than those of benthic thalli cultured at the same depth (B0) (F=9.579, P < 0.001), whereas in December, the RGRs of F0 were lower than those of B0 (F=371.720, P < 0.001) (Fig. 2, Supplementary Table S2). Benthic samples cultured at different depths showed similar RGRs in November; however, B0 samples grew faster than B3 samples in December (Fig. 2).

Before the in-situ experiments, benthic thalli showed significantly higher Chl a and carotenoid contents than floating thalli (Table 1). After a two-month culture period, the Chl a and carotenoid contents of all samples (except F3) decreased (Table 1). The Chl a content of B3 and F3 samples was approximately 5-fold and 6-fold higher than that of B0 and F0 samples, respectively (Table 1). The carotenoid contents of B3 and F3 samples were approximately 2-fold higher than those of B0 and F0 samples (Table 1). The initial phlorotannin concentrations in the benthic and floating thalli were 7.613±0.006 and 8.877±0.012 mg/g dw, respectively (Table 1). Higher phlorotannin concentrations were also detected in the thalli cultured at 0 masl in December (P < 0.05; Supplementary Table S3).

At each depth, benthic thalli displayed significantly higher F_v/F_m values than floating thalli during in-situ experiments, except for B3 vs. F3 in November (P < 0.05; Fig. 3, Supplementary Table S4). Y(II) values in F0 were significantly higher than those in B0 (P < 0.05; Fig. 3). Additionally, Y(II) values of F0 were declined from October to December. During the two-month culture period, both thalli showed a depth adjustment in the qP values (P < 0.05; Supplementary Table S4), with higher values in thalli cultured at 0 masl that those cultured at 3 mbsl (Fig. 3). In addition, qP values of B3 samples were higher than those of F3 samples from November to December (P < 0.05; Fig. 3).

3.2 Indoor experiment

After 21 d of high-light treatment, the photosynthetic performance of both types of thalli decreased, as shown by the decreasing F_v/F_m and Y(II) values (Fig. 4). By contrast, only slight changes in F_v/F_m were found in the control group during the incubation time. The Y(II) values of floating thalli were higher than those of benthic thalli during the first 9 d; however, both thalli types showed similar photosynthetic performance in subsequent days (Fig. 4). In the control group, the Y(II) values of floating thalli were higher than those of benthic thalli (Fig. 4). The Y(II) values of floating and benthic thalli at each treatment are significantly different (F=437.788; P < 0.05; Supplementary Table S5).

The rETR_m of floating thalli was higher than that of benthic thalli in both high-light and control groups (Fig. 4). However, no significant differences were detected between different thalli at each treatment (F=0.971, P > 0.05; Supplementary Table S5). The light utilisation efficiency (α) of floating thalli was higher than that of benthic thalli in both high-light and control groups (Fig. 4). Within the high-light group, the E_k value of floating thalli was higher than that of benthic thalli, except on the day 3 (Fig. 4). After 21 d of indoor culture, the E_k values of both thalli were higher in the high-light group than in the control group (Fig. 4).

The NPQ, which represents heat dissipation ability, of both types of thalli decreased with time in the high-light group (Fig. 4). During the first 9 d, the NPQ of floating thalli was higher in the high-light group than in the control group. By contrast, the NPQ values of benthic thalli in the high-light group were lower than in the control group (Fig. 4).

After 6 d of exposure to low light intensity (40 μmol photons/(m²·s)), both types of high-light-treated thalli recovered gradually, and the effective quantum yield Y(II) exceeded the initial level (Fig. 5).

4 DISCUSSION

New environmental conditions related to the ongoing rapid global climate change are expected to cause shifts of species range and abundance (Duarte et al., 2018; Martínez-Minay et al., 2019), which may benefit seaweeds with high environmental acclimation potential and dispersal ability. Sargassum horneri thalli were assumed to employ efficient mechanisms to respond to changes in solar irradiance at different water depths. Our data suggest that S. horneri possess high photo-acclimation potential that enables adjustments to changes in air-sea interface conditions. The increased photosynthetic rates and various photoprotective strategies may contribute to the long-term persistence and rapid growth of floating S. horneri at the sea surface.

Temperature and solar irradiation are the main factors affecting the reproduction and growth of floating seaweeds at the sea surface (Tala et al., 2016). In this study, high solar irradiance increased the growth rates of S. horneri at the sea surface from October to November when the temperature was approximately 11–18 ℃ (Supplementary Table S1), which is within the optimal growth temperature (T_opt) range (13–18 ℃) of S. horneri (Sun et al., 2008). Lower temperatures may buffer the negative effects of high light intensity (Graiff et al., 2013; Tala et al., 2019) because respiration rates seem to be sensitive to increasing temperature, especially when the temperature exceeds the T_opt for net photosynthesis (Staehr and Wernberg, 2009). By contrast, the reduction in RGRs of S. horneri in December may be attributed to the low ambient temperature (4.94±1.3 ℃), which is much lower than the T_opt of S. horneri. Choi et al. (2008) showed that the RGRs of S. horneri germlings were greater at higher temperatures (15–25 ℃) than at 10 ℃.

Sargassum horneri thalli cultured at different water depths for two months showed different acclimation responses, probably because of changes in variable Chl-a fluorescence of PSII, photosynthetic pigment contents and phlorotannin content. The highest RGR was detected in F0 in November; however, the RGR decreased in December. Accordingly, a reduction in photosynthetic parameters was observed in F0 from November to December. Photo-acclimation potential is usually linked to photosynthate production and/or maximal electron transport (Quintano et al., 2019). Among all treatments, F0 showed the highest Y(II) and qP values, suggesting that high light intensity enhances the plastoquinone acceptor (Q_A) re-oxidation capacity and CO₂ assimilation rates (Hollis and Hüner, 2017). Besides, during indoor experiments, higher rETR_m values were detected in the floating thalli than the benthic thalli, an outcome which could be related to an efficient use of energy (Koch et al., 2016). Similarly, rETR_m is high in apical fronds than in middle and basal fronds of brown seaweed Macrocystis pyrifera at different water depths, reflecting apical fronds are better acclimated to high light (Marambio et al., 2017).

The ability to adjust pigment concentrations reflects the ability to balance the harvesting of light energy with the dissipation of excess excitation energy (Koch et al., 2016), and this ability contributes to the successful persistence of S. horneri at the sea surface. Sargassun horneri cultured at 3 mbsl showed higher Chl a and carotenoid contents than those cultured at the sea surface, implying that thalli at 3 mbsl can make more efficient use of low irradiance than those at 0 masl (Xu and Gao, 2008). Severe high levels of solar irradiance may destroy photosynthetic pigments of thalli at the sea surface. A previous study showed that the contents of Chl a, Chl c, and carotenoids declined in M. pyrifera afloat on the sea surface for 15 d (Rothäusler et al., 2011b). Phlorotannins (polyphenol), the UV-absorbing compounds in brown seaweeds, exhibit high antioxidant and photoprotective capacities. Phlorotannins show dynamic changes with depth, suggesting that the variation is induced by the dose of radiation and UVR, as revealed in other brown seaweeds (Figueroa et al., 2014; Tala et al., 2017). Thus, high photosynthetic rate, increased UV-absorbing compounds and effective energy dissipation contribute to the long-term persistence and accumulation of S. horneri at the sea surface.

To further understand the acclimation strategies of S. horneri, we tested the photoprotective and recovery abilities of S. horneri by placing both types of thalli under high light intensities in the laboratory. After 21 d of indoor culture, both types of thalli showed an increase in E_k and decrease in α under high light intensity, which are the general characteristics of sun-adapted alga (Necchi, 2004). Once excess light has been absorbed, it can be dissipated via several routes, e.g. through the thermal dissipation (represented by NPQ) of excess excitation energy via the xanthophyll cycle (Jahns and Holzwarth, 2012). In the xanthophyll cycle, de-epoxidation of violaxanthin to zeaxanthin efficiently promotes the thermal dissipation of excess excitation energy (Demmig-Adams and Adams III, 1996). Dissipating energy as heat could be an efficient way to prevent the generation of reactive oxygen species (ROS), which increase the extent of photoinhibition by inhibiting the repair of PSII (Takahashi et al., 2009). In the indoor experiments, floating thalli showed stronger thermal dissipation capacity under high light intensity than benthic thalli, indicating that most of the absorbed energy may be passively dissipated in the form of heat and fluorescence (Figueroa et al., 2019). Furthermore, in the present study, rapid recovery rates of Y(II) in benthic and floating thalli indicate that S. horneri thalli adjust their physiological states rapidly to the changing environment.

The values of Y(II), rETR_m, and α were higher in floating thalli than in benthic thalli in indoor experiments, indicating floating thalli utilise the absorbed light through the photochemical reaction more efficiently than benthic thalli (Rothäusler et al., 2011b). In addition, during indoor experiments, the rETR_m of floating thalli was significantly higher than that of benthic thalli under both high and low light intensities, an outcome that could be related to an efficient use of energy (Li et al., 2014). Floating thalli decayed at lower water depth with low light irradiance during in-situ experiments; therefore, we inferred that high light irradiance is needed for the floating thalli to maintain balance between light energy absorbed versus energy utilised. In the field bottom trawl survey, Liu et al. (2018) observed that thousands of tons of S. horneri biomass sank to the bottom of the Yellow Sea of China. Apart from the reproductive stage, low light may also inhibit these thalli to grow and float to the surface again.

5 CONCLUSION

In this study, S. horneri demonstrated a high photo-acclimation potential, which is essential for biomass accumulation and long-term persistence at the sea surface. When afloat, S. horneri thalli were characterised by increased photosynthetic capacity, photoprotective pigments, and UV-absorbing compounds, which prevent the damaging effects of high solar irradiance. Future studies should explore the effects of multiple factors on positive buoyant seaweeds to understand their floating and adaptive strategies to new environmental conditions as well as evaluate and forecast their accumulation time and floating longevity in the near future.

6 DATA AVAILABILITY STATEMENT

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

Electronic supplementary material

Supplementary material (Supplementary Tables S1–S5 & Fig.S1) is available in the online version of this article at https://doi.org/10.1007/s00343-021-0380-y.