Introduction

Species of Sargassum (Fucales, Phaeophyceae) often form dense macroalgal assemblages, which play a central role in the coastal ecosystem by providing habitat and participating in nutrient cycle and primary production (De Wreede 1976, 1978; McCourt 1984; Martin-Smith 1993). Sargassum patens C. Agardh is a temperate brown alga that occurs widely in East Asia, including Japan, Korea, and China, where it forms dense populations on rocky substrata in the upper sublittoral zone (Yoshida 1998; Tseng and Lu 2000; Cho et al. 2012; Titlyanov and Titlyanova 2012). In Japan, this species is distributed along the temperate coasts of Honshu, Shikoku, and Kyushu islands, which are influenced by the warm Kuroshio and Tsushima currents. This species has not been observed in subarctic Hokkaido Island and in the Pacific coast of northern Honshu Island, which is influenced by the cold Oyashio Current, suggesting that relatively low seawater temperatures have negative effects on the survival of this alga and delineates the northern distributional limit (Yoshida 1998). Indeed, the temperature gradient caused by latitudinal variation and water current is a driving factor in biogeographical distributions of algae (Adey and Steneck 2001; Borlongan et al. 2017).

Common as it may seem, the knowledge on how the environment influences the range of distribution and abundance of S. patens in Japan is still insufficient, with few ecological studies (Taniguchi and Yamada 1978; Tsuchiya et al. 2011) and investigations on growth and photosynthesis (Tsukidate 1984; Gao and Umezaki 1988; Haraguchi et al. 2005; Baba 2007; Endo et al. 2013). For instance, Tsuchiya et al. (2012) revealed that the net photosynthetic rates of this species at 12 and 32 °C were significantly lower than those between 16 and 28 °C. However, temperature and PAR conditions for optimum photosynthesis remain to be clarified for this alga. Real-time shifts in the photosynthetic efficiency of other species of marine macroalgae have revealed the ability of seaweeds to adjust to diurnal and vertical variations of underwater PAR, which somehow justify their zonation pattern in the sea (e.g., Sargassum fusiforme (Harvey) Setchell, Kokubu et al. 2015; Sargassum mcclurei Setchell and Sargassum oligocystum Montagne, Terada et al. 2016b; Ecklonia radicosa (Kjellman) Okamura, Terada et al. 2016a). Furthermore, the capacity of Kappaphycus sp. from Okinawa, Japan to recover from photoinhibition was constrained by low temperature (Borlongan et al. 2017), suggesting that both low winter temperatures and PAR influence the northern distributional limit of this species in the region. Decrease in F v /F m associated with temperature, PAR, and PAR + UV combined treatments was also reported in a floating giant kelp, Macrocystis pyrifera (Linnaeus) C. Agardh (Rothäusler et al. 2011).

The impairment of photosynthesis including photosystem II (PSII) deactivation by thermal stress is primarily related to damage and breakdown of photosynthetic pigment-protein complexes and reduced activities of key enzymes in the Calvin cycle (Eggert 2012). Accumulation of H2O2 that inhibit de novo synthesis of D1 protein in PSII likewise accelerates the occurrence of photoinhibition (Takahashi and Murata 2008). Therefore, thermal and irradiance response of the maximum and effective quantum yields in the PSII can be regarded as sensitive and useful parameters to evaluate the optimum and critical environment in the habitat.

In the present study, we measured the dynamics of the effective quantum yield (Φ PSII ) of S. patens in the field under natural sunlight, and evaluated the ability of the species to recover from photoinhibition. The combined effects of temperature and PAR stress on the photosynthetic efficiency of S. patens were also determined and discussed in relation to their regional distribution in Japan.

Materials and methods

In situ measurements of diurnal changes in photosynthetic activity

Underwater measurements of photosynthetic activity of Sargassum patens were carried out by SCUBA (− 0.5 to − 3.4 m depth) at Yojiro (31° 33′ 31″ N 130° 33′ 46″ E; 12 May 2013, 10 January 2014, 23 March 2014) and Sakurajima (31° 35′ 27″ N 130° 35′ 33″ E, 13 June 2014), Kagoshima City, Kyushu Island, Japan, to observe diurnal variation, following the methods of Kokubu et al. (2015). Effective quantum yields (Φ PSII ) and incident PAR on the surface of frond were simultaneously measured every hour from 06:00 to 19:00, using a Diving-PAM (Heinz Walz GmbH, Germany) equipped with a universal sample holder (Diving-USH; fiber-optic angle 60°), applying a saturation pulse of 0.8 Hz (at a PAR setting of 8) as recommended by the manufacturer. The primary laterals on the upper portion of 30 different fronds were selected for measurement. Depth and seawater temperature were also recorded by the Diving-PAM. Given that S. patens occurs in the upper subtidal zone, this alga never gets exposed to air. Sea surface PAR was recorded using a quantum (2π) sensor (LI-190, LI-250A, LI-COR, USA) at least six times an hour, and at a sampling rate of 1 Hz for 30 s. The sky conditions were mostly uniform throughout the survey periods, with very little cloud cover.

Locally weighted smoothing (LOESS) was used to compare trends between PAR and Φ PSII time-series; while an ordinary least squares linear regression was applied to determine the relationship between Φ PSII and underwater PAR.

Sample collection and stock maintenance

Sargassum patens (more than 10 individuals for each experiment) were collected by SCUBA or snorkeling from Sakurajima, Kagoshima City, Kyushu Island, Japan on 12 April 2014, 22 May 2014, 4 June 2015, and 30 July 2015. Samples were directly transported to the laboratory of Kagoshima University in coolers at 20–24 °C, which was the seawater temperature range during the sampling periods. They were held at least overnight but not more than 3 days before examination in an aquarium tank (1500 L) at 33 psu, pH of 8.0–8.1, 20 °C, and 100 μmol photons m−2 s−1 (14:10 h light/dark cycle).

Effect of PAR on the oxygenic photosynthesis at three different temperatures

Net photosynthetic rates were determined at 0, 30, 60, 100, 150, 200, 250, 500, and 1000 μmol photons m−2 s−1 (n = 5 per PAR level), at 8, 20, and 28 °C, respectively. The three temperature treatments were assigned based on the lowest seawater temperature at the northern distributional limit (February; Aomori Prefecture, Honshu Island, 40° 35′ 00″ N 139° 52′ 20″ E), seawater temperature during peak abundance in Kagoshima Prefecture (May), and the highest seawater temperature at the southern distributional limit (August; Kagoshima Prefecture, Kyushu Island). Dissolved oxygen (DO) was measured using DO meters equipped with optical dissolved oxygen sensors (ProODO-BOD, YSI Incorporated, USA). A metal-halide lamp was used as light source (MHN-150D-S, Nichido Ind. Co. Ltd., Osaka, Japan), and a spherical (4π) submersible quantum sensor (LI-192, LI-250A, LI-COR, USA) was used to measure PAR.

Methods for photosynthesis–irradiance (P–E) experiments were described in detail in Kokubu et al. (2015). Explants (i.e., primary laterals on the upper portion of algae) were approximately 267.3 ± 65.3 mg wet weight (mgww; mean ± standard deviation, SD). They were acclimated overnight (12 h) with sterilized natural seawater in an incubator at each temperature treatment. On the day of the experiment, at least five explants were randomly selected and placed in 100 mL BOD bottles containing sterilized natural seawater. The DO sensors were then inserted carefully into the BOD bottles so that no bubbles were trapped. DO concentrations (mg L−1) were measured every 5 min for 30 min after 30-min acclimation to each experimental PAR. Water was continuously stirred throughout the measurement, with the experimental temperatures maintained using a recirculating water bath (Coolnit CL-600R, Taitec, Inc., Japan). The exact volumes of the BOD bottles were determined after the experiment and were used to estimate photosynthesis and dark respiration rates. Seawater medium was renewed after each PAR exposure and measurement to avoid any effects that can be attributed to the depletion of nutrients and dissolved carbon dioxide. P–E experiments started at 0 μmol photons m−2 s−1 and finished at 1000 μmol photons m−2 s−1.

Dark respiration and net photosynthetic rates were estimated by fitting a first-order linear model to the collected data.

Effect of temperature on the oxygenic photosynthesis and dark respiration

Measurements were conducted at nine temperatures (8, 12, 16, 20, 24, 28, 32, 36, and 40 °C; n = 5 per temperature) under an irradiance of 200 μmol photons m−2 s−1 (based on the saturation PAR; Table 1). The different temperatures were obtained with a water bath. Explants (ca. 204.0 ± 7.0 mg wet weight) from primary laterals on the upper portion of S. patens were used. Similar to P–E experiments, DO concentrations were measured every 5 min over a 30-min interval, with 30 min of acclimation to each temperature. Respiration rates were measured after 10 min of dark acclimation by wrapping the BOD bottles with aluminum foil.

Table 1 Mean and 95% Bayesian prediction intervals (95% BPI) of P–E parameters of S. patens at 8, 20, and 28 °C
Full size table

Temperature effect on the maximum quantum yield (F v /F m )

Maxi Imaging-PAM (Heinz Walz GmbH, Germany) measurements followed the procedure from Kokubu et al. (2015) with a few modifications. At least 5-cm-long branches with primary laterals and vesicles from the upper portion of algae were cut and acclimated overnight in the dark at 20 °C. Sections (n = 10 per temperature) were haphazardly selected and placed in a stainless-steel tray (12 × 10 × 3 cm) containing sterilized natural seawater. The tray was then placed on top of a block incubator (BI-535A, Astec, Fukuoka, Japan) for temperature control. A thermocouple (testo 925, testo AG, Lenzkirch, Germany) was used to confirm the water temperature. After a 30-min dark acclimation phase, F v /F m at 0 μmol photons m−2 s−1 were measured as initial values. Sections were then placed in 500-mL flasks wrapped with aluminum foil and were incubated in the dark at nine temperatures (8, 12, 16, 20, 24, 28, 32, 36, and 40 °C) for 72 h (EYELA MTI-201B, Tokyo Rikakikai Co., LTD., Japan). F v /F m at each temperature was measured every 24 h.

Effect of increasing PAR on the effective quantum yields (Φ PSII ) at three different temperatures

Ten branches (5 cm-long) with primary laterals from the upper portion of algae were acclimated overnight in the dark at 8, 20, and 28 °C. After acclimation, F v /F m (n = 10) at each temperature treatment was measured to provide initial values. Specimens were then placed in beakers (1000 mL) containing sterile natural seawater maintained at each experimental temperature in a water bath and were exposed to increasing PAR (200, 400, 600, 800, and 1000 μmol photons m−2 s−1) for 2 h before Φ PSII measurements (n = 10) were made. Temperature and PAR controls used were the same as previously described.

Combined effects of PAR and temperature on F v /F m and Φ PSII , and their potential of recovery

The combined effects of PAR and temperature on quantum yields were studied at two irradiances (200 and 1000 μmol photons m−2 s−1) and at three temperatures (8, 20, and 28 °C). Sargassum patens sections were prepared following the procedure in experiment 4, with an initial overnight acclimation at each experimental temperature in the dark, and subsequent measurement of initial F v /F m (n = 10 per temperature). Samples were then placed in separate beakers (1000 mL) containing sterile natural seawater maintained at a specific temperature in a water bath and were exposed to either 200 or 1000 μmol photons m−2 s−1 (metal-halide lamp) for 12 h. Φ PSII (n = 10 per temperature) were measured every 1 or 2 h of continuous exposure to each PAR treatment. Following the experiment, samples were once more dark-acclimated overnight (i.e., 12 h, corresponding to the night period under natural conditions) at their respective temperatures, and their F v /F m were measured to confirm the possibility of recovery.

Modeling the photosynthetic response to temperature

A Bayesian approach was used to analyze the response of photosynthesis to temperature. To model the response of either gross photosynthesis or maximum quantum yield to temperature, we applied a thermodynamic non-linear model (Eq. 1), which assumes that photosynthesis enters a less active state above some optimal temperature (Thornley and Johnson 2000; Alexandrov and Yamagata 2007). y is the response variable, which is either the gross photosynthetic rate or the maximum quantum yield (F v /F m ). The temperature scale is Kelvin (K). The model has four parameters: y max scales the model to the range of y. K opt is the absolute temperature where y is maximized, H a is the activation energy in kJ mol−1 and H d is the deactivation energy in kJ mol−1. R in this model is the ideal gas constant and has a value of 8.314 J mol−1.

$$ y=\frac{y_{\mathrm{max}}\bullet {H}_{\mathrm{d}}\bullet \exp \left(\frac{H_{\mathrm{a}}\bullet \left(K-{K}_{\mathrm{opt}}\right)}{K\bullet R\bullet {K}_{\mathrm{opt}}}\right)}{\left({H}_{\mathrm{d}}-{H}_{\mathrm{a}}\bullet \left(1-\exp \left({H}_{\mathrm{d}}\bullet \frac{\left(K-{K}_{\mathrm{opt}}\right)}{\left(K\bullet R\bullet {K}_{\mathrm{opt}}\right)}\right)\right)\right)} $$
(1)

F v /F m is mathematically bounded by 0 and 1 and therefore y was assumed to be beta distributed. The gross photosynthesis rate, which is assumed as a hidden state, was estimated by simultaneously fitting the measured respiration rates to the Arrhenius equation (Eq. 2) and the observed net photosynthesis rates to the difference in Eqs. 1 and 2. Under light conditions, both photorespiration and non-photorespiratory (i.e., mitochondrial) reactions result in oxygen consumption (Tcherkez et al. 2008); however, it is not uncommon for the differences between respiration rates under light and dark conditions to be insignificant (Bellasio et al. 2014). Hence, photorespiration was assumed to be adequately described by the dark respiration rate. R 24 is the respiration rate at 24 °C (i.e., mean temperature) and E a is the activation energy. The constant 297.15 is 24 °C scaled in absolute temperature.

$$ {R}_{\mathrm{d}}={R}_{24}\exp \left(-\frac{E_{\mathrm{a}}}{R}\left(\frac{1}{K}-\frac{1}{297.15}\right)\right) $$
(2)

The response of photosynthesis to PAR was examined by modeling the data using an exponential equation (Webb et al. 1974; Jassby and Platt 1976; Platt et al. 1980; Henley 1993) that included a respiration term and a term to model the effects of photoinhibition:

$$ {P}_{\mathrm{net}}={P}_{\mathrm{max}}\left(1-\exp \left(\frac{-\alpha }{P_{\mathrm{max}}}E\right)\right)\exp \left(\frac{-\beta }{P_{\mathrm{max}}}E\right)-{R}_{\mathrm{d}} $$
(3)

where, P net was the net O2 production rate, P max was the maximum O2 production rate, α was the initial slope of the P–E curve, β was a parameter to model the effects of photoinhibition and is set to 0 under conditions of no photoinhibition, E was the incident PAR, and R d was the dark respiration rate. From this model, the saturation PAR (E k) was calculated as P max/α and the compensation PAR (E c) was \( {P}_{\mathrm{max}}\ln \left(\frac{P_{\mathrm{max}}}{\left({P}_{\mathrm{max}}-{R}_{\mathrm{d}}\right)}\right)/\alpha \) when β is 0.

Statistical analysis

Statistical analyses were done using R version 3.3.3 (R Development Core Team 2017), and model fittings were done using RStan version 2.14.2 (Stan Development Team 2017). The parameters were determined by fitting the relevant models (i.e., Eqs. 1, 2, and 3) using Bayesian inference. RStan primarily uses a variant of a Hamiltonian Monte Carlo sampler to construct the posterior distributions of the parameters, and four chains of at least 500,000 samples/chain were generated and assessed for convergence, which provided at least 1000 samples of each of the parameters of interest. Informative normal priors were placed on all of the parameters of the model by using values from previous studies (Kokubu et al. 2015), and a half-cauchy prior was placed on the scale parameter of the models (Gelman 2004, 2006).

A one-way ANOVA was used to examine if continuous PAR exposures affected Φ PSII for each PAR treatment and temperature condition. Time was considered a factor with levels: 0, 12, and 24 h after the start of the experiment (i.e., initial F v /F m , Φ PSII after 12 h, and the final F v /F m after 12 h of darkness). The analysis was conducted on the logit transformed observations and with a generalized least squares approach to model both the heteroscedasticity and any correlation in the residuals caused by the repeated measurements. An initial fit using an AR1 correlation structure indicated that the serial correlations were not important, since the confidence interval of the correlation included the origin. Therefore, the final analysis only includes a model for heteroscedasticity. A likelihood ratio test was used to test the null hypothesis that the PAR exposure treatments had no effect on the observations after the models were fitted using maximum likelihood estimation. When the null hypothesis was rejected, the model was refit using the restricted maximum likelihood, and the effects were examined using the Wald tests (Zuur et al. 2009; Faraway 2016).

Results

Diurnal change of the in situ measurements

Sargassum patens were found on rocky substrata at the upper sublittoral zone in the study site. They were mostly exposed to sunlight throughout the day under fine clear sky. During measurements, the sea was relatively calm without significant wave motion. The actual depth of underwater measurements varied from 0.5–3.4 m during the survey periods. Seawater temperature ranged between 15 °C (10 January 2014) and 23 °C (13 June 2014). Detailed measurements during each survey period are presented in supplementary Table S1.

Both surface and underwater PAR showed typical diurnal changes during our field measurements (Fig. 1a–d). Levels increased from sunrise to noon (maximum surface PAR = 2165 ± 26.88 μmol photons m−2 s−1 at 12:00 on 13 June 2014, Fig. 1d; maximum underwater PAR = 940.7 ± 171.3 μmol photons m−2 s−1 at 11:39–11:52 on 12 May 2013, Fig. 1a) and then decreased until sunset.

Fig. 1
figure 1

Photosynthetically active radiation (PAR) and effective quantum yield (Φ PSII ) measurements conducted at Kagoshima, Japan, on 12 May 2013 (a, e, i), 10 January 2014 (b, f, j), 23 March 2014 (c, g, k), and 13 June 2014 (d, h, l). a–d Diurnal change of the sea surface PAR (circle) and underwater PAR (triangle) on the frond of Sargassum patens. e–h Diurnal change of Φ PSII of S. patens taken in situ. The symbols indicate actual values of the measurement, the line indicates the expected value determined by LOESS smoothing, and the shaded region indicates the 95% confidence interval. i–l Relationship between Φ PSII of S. patens and underwater PAR. Linear regression and the coefficient of determination are: y = −0.0003x + 0.664, R 2 = 0.545 for i; y = −0.0005x + 0.729, R 2 = 0.416 for j; y = −0.0005x + 0.681, R 2 = 0.646 for k; y = −0.0004x + 0.693, R 2 = 0.502 for l, respectively

Full size image

However Φ PSII showed a characteristic decline and recovery during the day (Fig. 1e–h). Values gradually decreased over time by noon or early afternoon (minimum Φ PSII  = 0.30 ± 0.126 at 13:20–13:29 on 12 May 2013; Fig. 1e) and later on returned to initial by sunset. In so doing, Φ PSII was negatively correlated with PAR (y = − 0.0005x + 0.681, R 2 = 0.646, 23 March 2014; Fig. 1k).

Effect of PAR on the oxygenic photosynthesis at three different temperatures

Net photosynthetic (NP) rates determined at 8, 20, and 28 °C, steadily increased with rising PAR and then approached saturation at relatively high PAR (Fig. 2). At 8 °C, NP rates (mean ± SD) were − 0.25 ± 0.16 μg O2 gww −1 min−1 at 0 μmol photons m−2 s−1 and 3.63 ± 0.87 μg O2 gww −1 min−1 at 250 μmol photons m−2 s−1. Rates declined thereafter to 2.51 ± 0.77 μg O2 gww −1 min−1 at 1000 μmol photons m−2 s−1 (Fig. 2a). At 20 °C and 28 °C, NP rates were − 1.45 ± 0.85 and − 2.61 ± 0.16 μg O2 gww −1 min−1 at 0 μmol photons m−2 s−1 and 27.75 ± 2.12 and 19.99 ± 3.04 μg O2 gww −1 min−1 at 1000 μmol photons m−2 s−1, respectively (Fig. 2b, c).

Fig. 2
figure 2

The response of the net photosynthetic rates of Sargassum patens from Kagoshima, Japan, to PAR at a 8 °C, b 20 °C, and c 28 °C. The dots indicate the measured rates (n = 5), the lines indicate the expected value, and the shaded regions indicate the 95% Bayesian prediction interval (BPI) of the model

Full size image

The parameter estimates which describe the significant features of each P–E curve are shown in Table 1. P–E curve at 8 °C depicted photoinhibition (β = − 0.002), which occurred at PAR above the level where maximum net photosynthetic rate occurred (i.e., 250 μmol photons m−2 s−1).

Effect of temperature on the oxygenic photosynthesis and dark respiration

Measured NP rates at 200 μmol photons m−2 s−1 increased from 2.33 ± 1.08 μg O2 gww −1 min−1 at 8 °C to a maximum of 10.14 ± 1.78 μg O2 gww −1 min−1 at 20 °C, then decreased to − 17.53 ± 4.01 μg O2 gww −1 min−1 at 40 °C (Fig. 3a). The gross photosynthesis–temperature (GP–T) model (Fig. 3b) indicated that the optimal temperature (\( {T}_{\mathrm{opt}}^{GP} \)) at maximum gross photosynthetic rate (GP max = 13.32 μg O2 gww −1 min−1) was 26.9 °C. Dark respiration was likewise affected by temperature, with rates increasing from − 0.46 ± 1.09 μg O2 gww −1 min−1 at 8 °C to 11.11 ± 1.28 μg O2 gww −1 min−1 at 40 °C (Fig. 3c). Model parameter estimates are presented in Table 2.

Fig. 3
figure 3

The response of the oxygenic photosynthesis and dark respiration of Sargassum patens from Kagoshima, Japan, to temperature. Net photosynthesis measurements were done at 200 μmol photons m−2 s−1 and the dark respirations at 0 μmol photons m−2 s−1. a The net photosynthesis to temperature determined at 200 μmol photons m−2 s−1. b The modeled gross photosynthetic rates at 200 μmol photons m−2 s−1. Data were derived from the model curve of net photosynthesis (a) and dark respiration (c). c The dark respiration to temperature at 0 μmol photons m−2 s−1. The dots indicate the measured rates (n = 5). See the caption in Fig. 2 for the details

Full size image
Table 2 Mean and 95% Bayesian prediction intervals (95% BPI) of the parameters estimated for the gross photosynthesis–temperature model
Full size table

Maximum quantum yield response to temperature

Throughout the 72-h different temperature exposures, maximum quantum yields (F v /F m ) remained steady at values above 0.7, from 8 to 32 °C (Fig. 4). F v /F m at 36 and 40 °C after 24 h were 0.60 ± 0.06 and 0.26 ± 0.33, respectively. They were almost 0 after 48 and 72 h.

Fig. 4
figure 4

The temperature response of the maximum quantum yield (F v /F m ) in Sargassum patens at nine temperature treatments (8, 12, 16, 20, 24, 28, 32, 36, and 40 °C) after a 24 h, b 48 h, and c 72 h without light. The dots indicate the measured values (n = 10). See the caption in Fig. 2 for the details

Full size image

The sudden decline in F v /F m at 36 and 40 °C, which can be attributed to the deactivation of photosynthetic efficiency, resulted in poor model fits. Therefore, these data were excluded from the analysis. Given the model and data after 72 h, maximum F v /F m of 0.77 occurred at 13.5 °C. Other model parameter estimates are presented in Table 3.

Table 3 Mean and 95% Bayesian prediction intervals (95% BPI) of the parameters estimated for the F v /F m –temperature model in S. patens exposed for 24, 48, and 72 h
Full size table

Effect of increasing PAR on the effective quantum yields at three different temperatures

The decline in quantum yields of S. patens at 8 °C was more abrupt and greater, as compared to those of S. patens at 20 and 28 °C (Fig. 5). From an initial F v /F m of 0.78 ± 0.02, Φ PSII decreased to 0.35 ± 0.05 after 2 h of exposure at 200 μmol photons m−2 s−1 and further dropped to 0.24 ± 0.05 after 2 h at 1000 μmol photons m−2 s−1 (Fig. 5a). At 20 °C, Φ PSII slowly decreased with PAR from initial F v /F m of 0.73 ± 0.02 to Φ PSII of 0.46 ± 0.05 after 2 h of exposure at 1000 μmol photons m−2 s−1 (Fig. 5b). Φ PSII at 28 °C also declined, with greater variation in values as PAR increased to 1000 μmol photons m−2 s−1 (Fig. 5c).

Fig. 5
figure 5

Chronological change of F v /F m and Φ PSII of Sargassum patens from Kagoshima, Japan with increasing PAR at a 8 °C, b 20 °C, and c 28 °C. Samples were acclimated to each PAR for 2 h before measurements. The dots indicate the measured values (n = 10). Initial values were measured as F v /F m

Full size image

Combined effects of PAR and temperature on F v /F m and Φ PSII , and their potential of recovery

Responses of the Φ PSII over 12 h of continuous exposures to 200 (low) and 1000 (high) μmol photons m−2 s−1 at 8, 20, and 28 °C, and recovery of their F v /F m after a 12-h dark acclimation phase were different from each PAR-temperature treatment (Fig. 6).

Fig. 6
figure 6

The hourly response of Φ PSII in Sargassum patens to PAR at 200 μmol photons m−2 s−1 (circle) and 1000 μmol photons m−2 s−1 (triangle), under a, b 8 °C, c, d 20 °C, and e, f 28 °C. The symbols indicate the average of actual values measured (n = 10), and bars indicate standard deviation. Initial values and the values after 12-h dark acclimation were measured as F v /F m

Full size image

At 8 °C, quantum yields of S. patens after 12 h of exposure to low PAR significantly decreased (P < 0.05) from initial F v /F m of 0.73 ± 0.06 (mean ± SD) to Φ PSII of 0.28 ± 0.08. Those exposed to high PAR likewise dropped (P < 0.05) from 0.71 ± 0.04 (mean ± SD) to 0.17 ± 0.06. Their post-dark acclimation F v /F m (0.57 ± 0.06 for low PAR and 0.19 ± 0.07 for high PAR) were significantly different (P < 0.05) from initial values (Fig. 6a, b).

At 20 °C, quantum yields of those exposed to low PAR were somewhat stable, with no significant decline (P = 0.97) in Φ PSII , whereas Φ PSII of samples exposed to high PAR significantly decreased (P < 0.05) from 0.77 ± 0.01 to 0.36 ± 0.03. After overnight dark acclimation, F v /F m of samples under high PAR increased to 0.69 ± 0.02 and yet were significantly different (P < 0.05) from the initial (Fig. 6c, d).

At 28 °C, Φ PSII of S. patens declined (P < 0.05) throughout the 12-h exposure to both low and high PAR, from an initial F v /F m of 0.72 ± 0.03 to Φ PSII of 0.66 ± 0.05 at 200 μmol photons m−2 s−1, and from 0.77 ± 0.03 to 0.32 ± 0.15 at 1000 μmol photons m−2 s−1. The final F v /F m of samples under low PAR returned to initial values, while those under high PAR failed to recover (P < 0.05).

Discussion

In general, S. patens is a perennial species that can live for a few years (Yoshida 1998). In the study site of Kagoshima, this species can be observed throughout the year; however, dominant season with long branches, vesicles, and primary laterals are limited from early spring to early summer (February to June; Tsuchiya et al. 2011).

In the present study, S. patens exhibited a depression in Φ PSII as PAR peaked at noon, followed by an upturn as PAR decreased by sunset. Variations in Φ PSII were somewhat small at sunrise and sunset, but were relatively large at noon (Fig. 1e–h). This may be attributed to the differences in the photosynthetic status of S. patens samples having complicated morphological features (i.e., primary laterals, vesicles, branches, and receptacles; Gao 1991) or to fluctuations in light caused by wave motion (Wing and Patterson 1993; Raven and Hurd 2012). Indeed, the PAR environment in the habitat was unsteady and non-uniform across the upper portions of S. patens (Fig. 1a–d). Nevertheless, observations were uniform throughout the survey, corresponding to the growth (January to March) and maturation (May to June) periods of the seaweed. Such diurnal fluctuations in photosynthetic efficiency was also observed on other species of Sargassum (Kokubu et al. 2015; Terada et al. 2016b) and marine macroalgae (Hanelt et al. 1993; Gévaert et al. 2003; Abdala-Díaz et al. 2006; Tala et al. 2017), revealing the effective protective response mechanism (i.e., dynamic photoinhibition; Gévaert et al. 2003) of the seaweeds to avoid chronic photoinhibition at excessive PAR (Ensminger et al. 2001; Eggert 2012). However, laboratory measurements of 12-h exposure to high PAR (1000 μmol photons m−2 s−1) showed failure of recovery in post-dark acclimation F v /F m on all temperature treatments (Fig. 6b–f), signifying that their repair system might be impaired. Maximum incident PAR on the frond of S. patens ranged between 480 and 940 μmol photons m−2 s−1 (Fig. 1a–d), which is close to critical for the species to suffer from photodamage.

In relation to the photosynthetic response of S. patens to light at various temperatures, no reduction in both oxygenic evolution and quantum yield was observed up to 1000 μmol photons m−2 s−1 at 20 and 28 °C (Fig. 2b, c). Given the high E c and E k at such temperatures (289–401 μmol photons m−2 s−1; Table 1), relatively high PAR is required for this species to reach photosynthetic saturation during summer at the region of southern distributional limit of this species (Kagoshima, Japan; Kokubu et al. 2015). In contrast, NP rates were lowest at 8 °C, with characteristic photoinhibition at PAR greater than 250 μmol photons m−2 s−1 (Fig. 2a). Low temperature limitation of photosynthetic enzymes may have led to such results. Although P–E responses of S. patens varied among temperature treatments, we interpreted these results with caution, as oxygen evolution experiments in the present study were conducted under short-term exposures. Studies of longer time-scale (i.e., seasonal acclimation experiments; Fairhead and Cheshire 2004) are needed to elucidate the capacity of this temperate alga to maintain optimal photosynthetic performance across a range of environmental conditions. Such studies may likewise confirm the worldwide distribution of this temperate seaweeds.

Moreover, depressions in Φ PSII after 2 or 12 h of PAR exposures were more pronounced at 8 °C than at 20 and 28 °C (Figs. 5 and 6). F v /F m of samples exposed to low PAR (200 μmol photons m−2 s−1) at 8 °C (Fig. 6a) failed to recover after the 12-h dark acclimation, while full recovery occurred at higher temperatures (Fig. 6c, e). Low temperatures altered the repair of PSII, given that protein synthesis decreases with declining temperatures (Allakhverdiev and Murata 2004; Takahashi and Murata 2008) and so prevented the seaweed’s full recovery from photoinhibition. Similar result at low temperature and saturating PAR was observed in Kappaphycus sp. from Okinawa, Japan (Solieriaceae; Borlongan et al. 2017), with low winter temperatures as one of the limiting factors for its northern limit of distribution in the western Pacific. Growth of S. patens was also affected by temperature, with decreased rates at 10 °C (Haraguchi et al. 2005; Baba 2007; Endo et al. 2013). The observed inhibitions in oxygen evolution and photochemical efficiency at 8 °C in the present study imply that such low temperature is critical for photosynthesis of S. patens, particularly in the Aomori Prefecture, Honshu Island, as the region of northern distributional limit of this species. Indeed, mean winter seawater temperature in this region is between 6 and 8 °C (2002–2015, Ajigasawa, Aomori Prefecture; JODC 2017). Incident PAR as low as 200 μmol photons m−2 s−1 in winter in this region would cause photodamage to the seaweed. Despite the absence of PAR data in this region, we can assume that it can reach such level in their habitat on clear, sunny days. Nonetheless, in situ measurements and photosynthesis experiments on S. patens from this region are necessary to confirm the possibility of photoinhibition caused by the combined effect of low temperature and PAR. Detailed phenology of this species in Aomori is also necessary. Whereas winter temperatures in Kagoshima seldom reach 8 °C; hence, low temperature-induced photoinhibition will probably not occur for S. patens in this region.

Likewise, temperature influenced the oxygenic photosynthesis, dark respiration and photochemical efficiency of S. patens (Figs. 3 and 4). Temperature response on oxygenic evolution (GP rates) showed a characteristic single peak at 26.9 °C (\( {T}_{\mathrm{opt}}^{GP} \); Fig. 3b), which falls within the range of seawater temperatures recorded at the collecting site (i.e., 15.5–29.3 °C; Kokubu et al. 2015). Optimum temperature range for growth of this species was 20–25 °C (Haraguchi et al. 2005; Baba 2007); however, temperature optima of photosynthesis in the present study are situated well above the temperature optima of growth. This difference is common as growth involves an integration of all metabolic processes, including photosynthesis (Eggert and Wiencke 2000; Eggert 2012). Meanwhile, dark respiration in S. patens was not detectable at 8 °C and still low at 20 °C, but it continuously increased from 24 °C (Fig. 3c), which lead to the decline in NP rates above such temperature (Fig. 3a). As for the F v /F m response of S. patens after 72 h of temperature exposures, values were high and stable from 8 to 32 °C, with the maximum F v /F m occurring at 13.5–16.0 °C (\( {T}_{\mathrm{opt}}^{\raisebox{1ex}{${F}_v$}\!\left/ \!\raisebox{-1ex}{${F}_m$}\right.} \)), but they suddenly dropped to zero at higher temperatures (Fig. 4c). Such F v /F m response, particularly at high temperatures in the present study, somewhat differ from those of previous studies showing a rather gradual decrease with rising temperature after 30 min of temperature exposures (Watanabe et al. 2014a, b, 2016; Terada et al. 2016a, b, c). Besides high temperature, the duration of exposure must have influenced the photochemical efficiency of PSII, which lead to its deactivation after 48 or 72 h. Nevertheless, these results are considered valuable, since temperatures in the natural habitat are rather long term and seldom abruptly change through the day. PSII deactivation due to thermal stress is probably related to one or all of the following: (a) dissociation of specific proteins in the core complex, (b) reduction of electron transport rates due to structural rearrangements in the thylakoid membranes, and (c) to accumulation of H2O2 that inhibit de novo synthesis of D1 protein in PSII (Allakhverdiev et al. 2008; Takahashi and Murata 2008; Roleda 2009; Hanelt and Figueroa 2012; Beer et al. 2014). Given their positive NP rates (Fig. 3a) and high values for F v /F m (Fig. 4c) up to 32 °C, S. patens may not have yet suffered from thermal stress until temperatures exceed such temperature. While summertime seawater temperatures in Kagoshima may only be up to 29.3 °C (Kokubu et al. 2015), they are still considered close to marginal temperatures for S. patens to undergo thermal inhibition. Although this alga is a perennial species, the lower portion of fronds, including the main stipe and holdfast, remains during summer (Tsuchiya et al. 2011). Thus, these parts must have greater high temperature tolerance, in comparison to the upper primary laterals measured in the present study.

In conclusion, based on the field and laboratory measurements, S. patens is well adapted to the environment in the temperate region of Japan. This species may also possibly undergo thermal inhibition at temperatures approaching threshold levels, particularly during summer in Kagoshima. Whereas the combined effects of low temperature and high PAR may cause photodamage to S. patens and so limit the northern distribution of this temperate species in Aomori, Japan.