Physiological response of a golden tide alga ( Sargassum muticum ) to the interaction of ocean acidification and phosphorus enrichment

The development of golden tides is potentially influenced by global change factors, such as ocean acidification and eutrophication, but related studies are very scarce. In this study, we cultured a golden tide alga, Sargasssum muticum, at two levels of pCO2 (400 and 1000 μatm) and phosphate (0.5 and 40 μM) to investigate the interactive effects of elevated pCO2 and phosphate on the physiological properties of the thalli. Higher pCO2 and phosphate (P) levels alone increased the relative growth rate by 41 and 48 %, the net photosynthetic rate by 46 and 55 %, and the soluble carbohydrates by 33 and 62 %, respectively, while the combination of these two levels did not promote growth or soluble carbohydrates further. The higher levels of pCO2 and P alone also enhanced the nitrate uptake rate by 68 and 36 %, the nitrate reductase activity (NRA) by 89 and 39 %, and the soluble protein by 19 and 15 %, respectively. The nitrate uptake rate and soluble protein was further enhanced, although the nitrate reductase activity was reduced when the higher levels of pCO2 and P worked together. The higher pCO2 and higher P levels alone did not affect the dark respiration rate of the thalli, but together they increased it by 32 % compared to the condition of lower pCO2 and lower P. The neutral effect of the higher levels of pCO2 and higher P on growth and soluble carbohydrates, combined with the promoting effect on soluble protein and dark respiration, suggests that more energy was drawn from carbon assimilation to nitrogen assimilation under conditions of higher pCO2 and higher P; this is most likely to act against the higher pCO2 that caused acid–base perturbation via synthesizing H transport-related protein. Our results indicate that ocean acidification and eutrophication may not boost golden tide events synergistically, although each one has a promoting effect.


Introduction
Sargassum C. Agardh (1820) is the most species-rich genus in the Phaeophyta and has a global distribution (Mattio and Payri, 2011).The species of this genus constitutes an important part of the marine flora and is considered a valuable and unique habitat for a number of highly adapted marine animal species (Laffoley et al., 2011).Some species of Sargassum are economically important and are used in animal fodder, agricultural manure, and alginate production (Ashok-Kumar et al., 2012;Fenoradosoa et al., 2010;González-López et al., 2012).On the other hand, Sargassum is an aggressive genus, and it can rapidly spread and invade new areas (Sfriso and Facca, 2013).The invasion of Sargassum would accordingly compete with indigenous species for nutrients and light, leading to the alteration of the macroalgal community structure (Rueness, 1989;Staehr et al., 2000).For instance, the increased abundance of S. muticum in Limfjorden (Denmark) between 1990 and 1997 led to decreased cover of several indigenous species belonging to the genera Codium, Fucus, and Laminaria, and thus reduced the species richness and diversity of the macroalgal community (Staehr et al., 2000).Recently, species of Sargassum have inundated the coasts along the Gulf of Mexico, West Africa, the Caribbean, and Brazil in unprecedented biomass, which are termed golden tides (Schell et al., 2015;Smetacek and Zingone, 2013).Apart from the negative effect on aesthetics and tourism, the occurrence of golden tides could kill the fish within the algal mass, mainly due to hypoxia or anoxia in the waters caused by decomposition of Sargassum thalli (Cruzrivera et al., 2015).In addition, the dense Sargassum accumulation could clog fishing nets and impede the passage of boats, leading to food shortages for local people who depend on artisanal fisheries (Smetacek and Zingone, 2013).The occurrence of golden tides has been linked to higher nutrient levels in seawater (Lapointe, 1995;Smetacek and Zingone, 2013).The distribution pattern and biomass of Sargassum spp.are environmentdependent (temperature, light, nutrients, etc.) (Ang, 2006;Sfriso and Facca, 2013).
Due to the burning of fossil fuels and changes in land use, the atmospheric concentrations of carbon dioxide increased to the level of 401.72 ppm in July 2016 (http://www.esrl.noaa.gov/gmd/ccgg/trends/global.html), which is an unprecedented high over the last 800 000 years (IPCC, 2013).When CO 2 dissolves in seawater, it forms carbonic acid, and as more CO 2 is taken up by the ocean's surface, the pH decreases, moving towards a less alkaline and therefore more acidic state; this is termed ocean acidification.The mean surface ocean pH has already decreased by 0.1 units since the beginning of the industrial era, corresponding to a 26 % increase in hydrogen ion concentration (IPCC, 2013).By 2100, concentrations of CO 2 (aq) and HCO − 3 are predicted to increase by 192 and 14 %, respectively, and CO 2− 3 is predicted to decrease by 56 % with a concomitant decline in pH to 7.65 (Raven et al., 2005).Increased CO 2 could exert positive, neutral, or negative effects on the physiological properties of macroalgae (Ji et al., 2016;Wu et al., 2008).In terms of Sargassum species, increased CO 2 (800 ppm) enhanced the photosynthetic rate (based on CO 2 uptake) in S. muticum (Longphuirt et al., 2014).On the other hand, the same level of increased CO 2 (750 ppm) did not affect growth, Rubisco's maximal activity, affinity for CO 2 , or quantity in S. vulgare (Alvaro and Mazal, 2002).Furthermore, increased CO 2 (750 ppm) significantly decreased the net photosynthetic rate and light saturation point of S. henslowianum (Chen and Zou, 2014).
Apart from ocean acidification, eutrophication is another environmental challenge.Eutrophication can occur naturally in lakes through the transfer of nutrients from the sediment to the water via living or decomposing macrophytes, resuspension, diffusion, and bioturbation (Carpenter, 1981).However, anthropogenic activities have accelerated the rate and extent of eutrophication (Carpenter et al., 1998).The inevitable urbanization of a growing human population, the increased use of coastal areas, and rising fertilizer use for agricultural intensification has led to accelerated nutrient inputs from land water to coastal waters (Smith et al., 1999).These changes in nutrient availability result in eutrophication, an increasing threat for coastal ecosystems (Bricker et al., 2008).One consequence of eutrophication is that it can lead to algal bloom, such as green tides and golden tides (Smetacek and Zingone, 2013).There are intensive studies regarding the effect of nutrients on the physiological properties of Sargassum species (Hwang et al., 2004;Incera et al., 2009;Lapointe, 1995;Liu and Tan, 2014;Nakahara, 1990).Enrichment of nutrients can usually enhance the growth and photosynthetic parameters of Sargassum.For instance, the growth rate of S. baccularia almost doubled when nutrients increased from 3 µM ammonium plus 0.3 µM phosphate to 5 µM ammonium plus 0.5 µM phosphate (Schaffelke and Klumpp, 1998), and the photosynthetic rates of S. fluitans and S. natans were also 2-fold higher with 0.2 mM PO 3− 4 enrichment compared to the control (Lapointe, 1986).Furthermore, some studies have demonstrated that macroalgae experience more phosphorus limit than nitrogen limit (Lapointe, 1986;Lapointe et al., 1987Lapointe et al., , 1992;;Littler et al., 1991).For instance, nitrogen enrichment did not affect the growth rates of S. fluitans or S. natans, while phosphorus enrichment increased them from 0.03-0.04(control) to 0.05-0.08doublings d −1 (Lapointe, 1986).
Neither ocean acidification nor eutrophication is proceeding in isolation; rather, they occur simultaneously, particularly in coastal areas.The interactive effects of the two factors may be completely different or of greater magnitude compared to the effects of any single stressor.To the best of our knowledge, no studies have been reported regarding the interactive effects of ocean acidification and eutrophication on Sargassum.In this study, we chose the species S. muticum to investigate its responses to the interaction of ocean acidification and eutrophication.S. muticum is an invasive macroalga that commonly inhabits rocky shores (Karlsson and Loo, 1999).It originates from Japan and was imported to the Northern Pacific coast of the United States in the early 20th century (Scagel, 1956).It was also introduced to Europe along with the Japanese oyster in the late 1960s (Jones and Farnham, 1973).Its distribution is now worldwide due to its introduction and subsequent rapid expansion (Cheang et al., 2010).Our study supplies insight into how ocean acidification and eutrophication affect the physiological properties of S. muticum and thus the development of golden tides.
2 Materials and methods 2.1 Sample collection and experimental design S. muticum was collected from lower intertidal rocks on the coast of Lidao, Rongcheng, China (37 • 15 N, 122 • 35 E).The samples were transported to the laboratory in an insulated polystyrene cooler (4-6 • C) within 3 h.Healthy thalli were selected and rinsed with sterile seawater to remove sediments, epiphytes, and small grazers.The thalli were maintained in an intelligent illumination incubator (MGC-250P, Yiheng Technical Co. Ltd., Shanghai, China) for 24 h before the experiment.The temperature in the incubator was set at 20 • C with a 12 h-12 h (light-dark) photoperiod of 150 µmol photons m −2 s −1 photosynthetically active radiation (PAR).After the maintenance, a two-way factorial experiment was set up to investigate the interactive effects of pCO 2 and phosphate on S. muticum.The thalli were placed in 3 L flasks with 2 L of sterile seawater (one thallus per flask) and cultured at fully crossed two pCO 2 (400 µatm, lower pCO 2 , LC; 1000 µatm, higher pCO 2 , HC) and two phosphate (0.5 µM, lower phosphate, LP; 40 µM, higher phosphate, HP) levels with continuous aeration for 13 days.Phosphorus was selected as a nutrient variable, because some findings have displayed that phosphorus, rather than nitrogen, is the primary limiting nutrient for macroalgae (Lapointe, 1986;Lapointe et al., 1987Lapointe et al., , 1992;;Littler et al., 1991).The conditions of natural seawster are 400 µatm pCO 2 and 0.5 µM phosphate.The 400 µatm pCO 2 was achieved by bubbling ambient air and 1000 µatm pCO 2 was obtained through a CO 2 plant chamber (HP1000 G-D, Wuhan Ruihua Instrument & Equipment Ltd, China) with a CO 2 variation of less than 5 %.The higher P level (40 µM) was achieved by adding NaH 2 PO 4 to natural seawater, and the nitrate concentration was set at 200 µM for all treatments to avoid N limit.The media were refreshed every day.

Carbonate chemistry parameters
The seawater pH was recorded with a pH meter (pH 700, Eutech Instruments, Singapore), and total alkalinity (TA) was measured by titrations.The salinity of the seawater was 29.Other carbonate system parameters, which were not directly measured, were calculated via CO2SYS (Pierrot et al., 2006) using the equilibrium constants of K 1 and K 2 for carbonic acid dissociation (Roy et al., 1993).

Measurement of growth
The growth of S. muticum was determined by weighing fresh thalli.The thalli of S. muticum were blotted gently with tissue paper to remove water on the surface from the thalli before weighing them.The relative growth rate (RGR) was estimated as follows: RGR = (ln W t − ln W 0 )/t × 100, where W 0 is the initial fresh weight (FW) and W t is the weight after t days of culture.

Determination of photosynthesis and respiration
The net photosynthetic rate of the thalli was measured by a Clark-type oxygen electrode (Chlorolab-3, Hansatech, Norfolk, UK) at the end of the experiment.Approximately 0.1 g of fresh-weight algae harvested from the culture flask was transferred to the oxygen electrode cuvette with 8 mL of sterilized media, and the media were stirred during measurement.The irradiance and temperature conditions were set the same as in the growth incubators.The increase of oxygen content in seawater within 5 min was defined as the net photosynthetic rate, and the decrease of oxygen content in seawater in darkness within 10 min was defined as the respi-ration rate.The net photosynthetic rate (NPR) and respiration rate were presented as µmol O 2 g −1 FW h −1 .
Photosynthetic rates at different dissolved inorganic carbon (DIC) levels were measured under saturating irradiance of 600 µmol photons m −2 s −1 at the end of the experiment.The various DIC concentrations (0-13.2mM) were obtained by adding different amounts of NaHCO 3 to the Tris-buffered DIC-free seawater.DIC was removed from the natural seawater by reducing pH to approximately 4.0 with the addition of 1.0 M HCl and then sparging for 2 h with pure N 2 gas (99.999 %).Finally, Tris buffer (25 mM) was added and the pH was adjusted to 8.1 with freshly prepared 1 M NaOH and 1 M HCl.The parameters, which are the maximum photosynthetic rate (V max ) and the half saturation constant (K 0.5 , i.e., the DIC concentration required to give half of inorganic carbon (Ci)-saturated maximum rate of photosynthetic O 2 evolution), were calculated from the Michaelis-Menten kinetics equation (Caemmerer and Farquhar, 1981) where [S] is the DIC concentration.

Assessment of photosynthetic pigments
At the end of the experiment, approximately 100 mg of freshweight thalli from each culture condition were ground thoroughly in 2 mL 80 % acetone and placed in darkness for 12 h.Then the homogenate was centrifuged for 10 min at 5000 g and the supernatant was used to determine Chl a content spectrophotometrically according to the equation of Lichtenthaler (1987).

Measurement of nitrate uptake rate
The nitrate uptake rate (NUR) of the thalli was estimated from the decrease of NO − 3 concentration in the culture medium over a given time interval (12 h) during the light period using the following equation: NUR = (N 0 − N t ) × V /W/12, where N 0 is the initial concentration of NO − 3 , N t is the concentration after 12 h, V is the volume of the culture medium, and W is the fresh weight of the thalli in culture.NO − 3 concentration in the seawater was measured according to Strickland and Parsons (1972).

Estimate of nitrate reductase activity
The nitrate reductase activity of the thalli was assayed according to modified in situ method of Corzo and Niell (1991).The measurement was conducted during the local noon period (13:00 UTC + 8 h (Chinese Standard Time)), because the activity of nitrate reductase usually displays circadian periodicity; a maximum during the light period and a minimum in darkness (Deng et al., 1991;Velasco and Whitaker, 1989).Approximately 0.3 g (FW) of thalli from each culture condition was incubated for 1 h at 20 • C in darkness in the reaction www.biogeosciences.net/14/671/2017/Biogeosciences, 14, 671-681, 2017 solution (10 mL), which contained 0.1 M phosphate buffer, 0.1 % propanol (v/v), 50 mM KNO 3 , 0.01 mM glucose, and 0.5 mM EDTA with a pH of 8.0.The mixture was flushed with pure N 2 gas (99.999 %) for 2 min to obtain an anaerobic state before the incubation.The concentration of nitrite produced was determined colorimetrically at 540 nm (Zou, 2005).The NRA was expressed as µmol NO − 2 g −1 FW h −1 .

Analysis of biochemical composition
At the end of the experiment, about 0.2 g of FW thalli from each culture condition were ground in a mortar with distilled water, and soluble carbohydrates were extracted in a water bath of 80 • C for 30 min.After being centrifuged for 10 min at 5000 g, the supernatant was volumed to 25 mL with distilled water, and soluble carbohydrate content was determined by the phenol-sulfuric acid method (Kochert, 1978).Approximately 0.2 g of FW thalli from each culture condition were ground in a mortar with extraction buffer (0.1 mol L −1 phosphate buffer, pH 6.8) and then centrifuged for 10 min at 5000 g.Soluble protein was estimated from the supernatant using the Bradford (1976) assay with bovine serum albumin as a standard.

Data analysis
Results were expressed as means of replicates ± standard deviation.Data were analyzed using the software SPSS v.21.The data under every treatment conformed to a normal distribution (Shapiro-Wilk, P > 0.05), and the variances can be considered equal (Levene's test, P > 0.05).Two-way analysis of variance (ANOVA) was conducted to assess the effects of pCO 2 and P on carbonate parameters, relative growth rate, net photosynthesis rate, V max , K 0.5 , Chl a, nitrate uptake rate, nitrate reductase activity, soluble carbohydrates, soluble protein, and dark respiration rate.Tukey's honest significance difference (HSD) was conducted for a post hoc investigation.A confidence interval of 95 % was set for all tests.

Results
The effects of ocean acidification and P enrichment on seawater carbonate parameters were detected (Table 1).Twoway ANOVA analysis (P = 0.05) showed that pCO 2 had a main effect on all parameters except TA, while P did not affect any parameter.A post hoc Tukey's HSD comparison (P = 0.05) showed that elevated pCO 2 decreased pH by 0.31 at both LP and HP and CO 2− 3 by 45 % (LP) and 45 % (HP), but it increased DIC by 10 % (LP) and 9 % (HP), HCO − 3 by 14 % (LP) and 14 % (HP), and CO 2 by 139 % (LP) and 134 % (HP).
The growth of S. muticum cultured at different pCO 2 and P conditions was recorded (Fig. 1).pCO 2 and P had an interactive effect on the relative growth rate of S. muticum (ANOVA, F = 5.776, df = 1, 8, P = 0.043), and each factor had a main effect (ANOVA, F = 19.145,df = 1, 8, P = 0.002 for pCO 2 ; ANOVA, F = 30.592,df = 1, 8, P = 0.001 for P).A post hoc Tukey's HSD comparison (P = 0.05) showed that the higher levels of pCO 2 and higher P alone increased the relative growth rate by 41 and 48 %, respectively, compared to the relative growth rate (3.1 ± 0.4 %) at lower pCO 2 and lower P.The combination of the higher pCO 2 and higher P levels did not enhance the relative growth rate as much as the sum of the higher pCO 2 alone plus the higher P alone, with an increase of 59.66 %.Although the higher P level increased the relative growth rate at lower pCO 2 , it did not affect the relative growth rate at higher pCO 2 .
In terms of the net photosynthetic rate (Fig. 2), both pCO 2 (ANOVA, F = 26.556,df = 1, 8, P = 0.001) and P had main effects (ANOVA, F = 38.963,df = 1, 8, P < 0.001).A post hoc Tukey's HSD comparison (P = 0.05) showed that the higher pCO 2 level increased the net photosynthetic rates by 46 and 24 % at lower P and higher P, respectively.The higher P level increased the net photosynthetic rates by 55 and 31 % at lower pCO 2 and higher pCO 2 , respectively.The difference in the net photosynthetic rate between LCHP and HCLP was statistically insignificant.
The carbon-saturating maximum photosynthetic rate (V max ) and the half saturation constant (K 0.5 ) obtained from the photosynthesis versus DIC curves (Fig. 3) are shown in Table 2.The pCO 2 and P had an interactive effect on the V max of S. muticum (ANOVA, F = 10.095,df = 1, 8, P = 0.013), and each factor had a main effect (ANOVA, F = 31.402,df = 1, 8, P = 0.001 for pCO 2 ; ANOVA, F = 105.116,df = 1, 8, P < 0.001 for P).A post hoc Tukey's HSD comparison (P = 0.05) showed that the higher pCO 2 level increased the V max by 42 % at lower P, while the increase at higher P was statistically insignificant.The higher Table 1.Parameters of the seawater carbonate system at different CO 2 and phosphate conditions.Measurements and estimation of the parameters are described in the "Materials and methods" section.Data are reported as means ±SD (n = 3).LCLP is the low pCO 2 and low P condition, LCHP is the low pCO 2 and high P condition, HCLP is the high pCO 2 and low P condition, and HCHP is the high pCO 2 and P condition.DIC is dissolved inorganic carbon, and TA is total alkalinity.P level increased the V max at the conditions of both lower pCO 2 (65 %) and higher pCO 2 (24 %) with a larger promoting effect at lower pCO 2 .pCO 2 and P interacted on the K 0.5 of S. muticum (ANOVA, F = 5.928, df = 1, 8, P = 0.041), and each factor had a main effect (ANOVA, F = 14.713, df = 1, 8, P = 0.005 for pCO 2 ; ANOVA, F = 20.857,df = 1, 8, P = 0.002 for P).A post hoc Tukey's HSD comparison (P = 0.05) showed that the higher pCO 2 level increased the K 0.5 by 98 % at lower P but did not affect it at higher P. In con-Figure 3. The photosynthesis versus DIC curves of S. muticum after being cultured under pCO 2 and P conditions for 13 days.Data are reported as means ±SD (n = 3).LCLP is the low pCO 2 and low P condition, LCHP is the low pCO 2 and high P condition, HCLP is the high pCO 2 and low P condition, and HCHP is the high pCO 2 and high P condition.DIC is dissolved inorganic carbon.
trast, the higher P level decreased the K 0.5 by 55 % at higher pCO 2 and the negative effect of the higher P level at lower pCO 2 was insignificant.
To assess the effects of ocean acidification and P enrichment on the nitrogen assimilation in S. muticum, the nitrate uptake rate under various pCO 2 and P treatments was investigated (Fig. 5).Both pCO 2 (ANOVA, F = 139.916,df = 1, 8, P < 0.001) and P (ANOVA, F = 43.923,df = 1, 8, P < 0.001) had main effects on the nitrate uptake rate of S. muticum.The nitrate uptake rates at lower pCO 2 were 0.18 ± 0.01 (LP) and 0.25 ± 0.03 µmol NO − 3 g −1 FW h −1 (HP), respectively.A post hoc Tukey's HSD comparison (P = 0.05) showed that the higher pCO 2 level increased the nitrate uptake rate to 0.31 ± 0.02 µmol NO − 3 g −1 FW h −1 at lower P and to 0.39 ± 0.01 µmol NO − 3 g −1 FW h −1 at higher P, compared to the rates at lower pCO 2 .The higher P level also increased the nitrate uptake rate by 36 % at lower pCO 2 and by 28 % at higher pCO 2 , compared to the rates at lower P.
Figure 5. Nitrate uptake rate of S. muticum after being grown at different pCO 2 and P conditions for 13 days.Data are reported as means ±SD (n = 3).LCLP is the low pCO 2 and low P condition, LCHP is the low pCO 2 and high P condition, HCLP is the high pCO 2 and low P condition, and HCHP is the high pCO 2 and high P condition.Different letters above the error bars indicate significant differences between treatments (P < 0.05).
Figure 6.Nitrate reductase activity (NRA) of S. muticum after being grown at different pCO 2 and P conditions for 13 days.Data are reported as means means ±SD (n = 3).LCLP is the low pCO 2 and low P condition, LCHP is the low pCO 2 and high P condition, HCLP is the high pCO 2 and low P condition, and HCHP is the high pCO 2 and high P condition.Different letters above the error bars indicate significant differences between treatments (P < 0.05).
The soluble carbohydrates (Fig. 7a) and protein (Fig. 7b) were estimated to understand the effects of ocean acidification and P enrichment on the products of carbon and nitrogen assimilation in S. muticum.pCO 2 and P had an interactive effect on the soluble carbohydrates (ANOVA, F = 18.294, df = 1, 8, P = 0.003), and P had a main effect (ANOVA, F = 23.129,df = 1, 8, P = 0.001).The higher P level increased the soluble carbohydrates from 25.40 ± 1.66 to 41.10 ± 1.74 mg g −1 FW at lower pCO 2 but did not alter them at higher pCO 2 .The higher pCO 2 level increased the soluble carbohydrates to 33.72 ± 3.31 mg g −1 FW at lower P, while the decrease of soluble carbohydrates caused by the higher pCO 2 level was not statistically significant at higher P. of S. muticum after being grown at different pCO 2 and P conditions for 13 days.Data are reported as means ±SD (n = 3).LCLP is the low pCO 2 and low P condition, LCHP is the low pCO 2 and high P condition, HCLP is the high pCO 2 and low P condition, and HCHP is the high pCO 2 and high P condition.Different letters above the error bars indicate significant differences between treatments (P < 0.05).
Finally, the effects of ocean acidification and P enrichment on the dark respiration rate of S. muticum were investigated (Fig. 8).pCO 2 and P had an interactive effect on the dark respiration rate (ANOVA, F = 19.584,df = 1, 8, P = 0.002), and each factor had a main effect (ANOVA, F = 6.428, df = 1, 8, P = 0.035 for pCO 2 ; ANOVA, F = 6.754, df = 1, 8, P = 0.032 for P).The higher pCO 2 level increased the dark respiration rate from 14.21 ± 1.94 to 21.24 ± 1.28 µmol O 2 g −1 FW h −1 at higher P but did not affect it at lower P. Likewise, the higher P level increased the respiration rate from 14.15 ± 0.65 to 21.24 ± 1.28 µmol O 2 g −1 FW h −1 at higher pCO 2 but did not change it at lower pCO 2 .

Effects of pCO 2 and P on carbon assimilation
The higher pCO 2 level increased the net photosynthetic rate in S. muticum at lower P in the present study.Although the Figure 8. Dark respiration rate of S. muticum after being grown at different pCO 2 and P conditions for 13 days.Data are reported as means ±SD (n = 3).LCLP is the low pCO 2 and low P condition, LCHP is the low pCO 2 and high P condition, HCLP is the high pCO 2 and low P condition, and HCHP is the high pCO 2 and high P condition.Different letters above the error bars indicate significant differences between treatments (P < 0.05).dissolved inorganic carbon in seawater is around 2 mM, the dominant form is HCO − 3 with CO 2 typically accounting for less than 1 % (Dickson, 2010).In addition, CO 2 in seawater diffuses ∼ 8000 times more slowly than in air (Gao and Campbell, 2014).Furthermore, marine macroalgae have high K 0.5 values (40-70 µM CO 2 ) for Rubisco, the carbon assimilating enzyme (Ji et al., 2016).The evidence above indicates that the CO 2 in seawater should be carbon limited for marine macroalgae.The promoting effect of elevated CO 2 on photosynthesis was also reported in other macroalgae species, such as the green algae Ulva linza (Gao et al., 1999), the red algae Pyropia haitanensis (Zou and Gao, 2002), and the brown algae Petalonia binghamiae (Zou and Gao, 2010).The higher pCO 2 level increased K 0.5 of S. muticum at lower P in the present study, which indicates that a plant grown under conditions of higher pCO 2 reduces its photosynthetic affinity for DIC.This phenomenon is commonly found in both microalgae and macroalgae (Gao and Campbell, 2014;Ji et al., 2016;Wu et al., 2008) and is considered a sign of downregulated CCMs at high CO 2 conditions (Gao and Campbell, 2014).However, this decrease of photosynthetic affinity for DIC did not lead to reduced photosynthesis in S. muticum compared to that at the lower pCO 2 in the present study, mainly because of increased CO 2 availability for Rubisco and depressed photorespiration at the elevated ratio of CO 2 to O 2 , which has been confirmed in the red seaweed Lomentaria articulata (Kübler et al., 1999).
The higher P level also increased the net photosynthetic rate of S. muticum in the present study, which can be partially explained by the decreased K 0.5 at higher P. The decreased K 0.5 is an indication of increased photosynthetic carbon-use capability.Phosphorus is a key macronutrient component for organisms, and high levels of P availability are not only essential for chloroplast DNA and RNA synwww.biogeosciences.net/14/671/2017/Biogeosciences, 14, 671-681, 2017 thesis (Vered and Shlomit, 2008), but are also required for various chloroplast functions referring to the phosphorylation of photosynthetic proteins, the synthesis of phospholipids, and the generation of adenosine triphosphate (ATP; Zer and Ohad, 2003).Therefore, high P levels could speed up the transport of Ci from media to the site of Rubisco by supplying necessary energy.In addition, P enrichment can increase both the activity and the amount of Rubisco (Lauer et al., 1989).Phosphorus, with low concentrations in seawater, is generally considered to be limiting for marine primary producers (Elser et al., 2007;Howarth, 1988;Müller and Mitrovic, 2015).Therefore, adding extra phosphorus to natural seawater can stimulate the photosynthesis of algae.For instance, the midday (12:00) photosynthetic rates increased from 1.3 to 2.3 mg C g −1 DW h −1 for S. natans and from 0.9 to 2.1 mg C g −1 DW h −1 for S. fluitans when 0.2 mM P was added (Lapointe, 1986).In the present study, the addition of 40 µmol P also resulted in a nearly 2-fold increase of the net photosynthetic rate and the V max , which suggests the importance of P in the photosynthesis of this alga.In addition, the higher P level promoted the synthesis of Chl a at the condition of lower pCO 2 , which may also contribute to the increased net photosynthetic rate in S. muticum at higher P.Although P is not a component constituting Chl a, a higher P supply may stimulate the content of Chl a synthesis-related enzymes and thus the production of Chl a.The positive effect of P on Chl a was also reported in S. thunbergii (Nakahara, 1990).On the other hand, the higher P level did not increase the Chl a content at higher pCO 2 in the present study.A possible reason is that there is more ATP available at higher pCO 2 due to the downregulation of CCMs, and thus there is no need to synthesize more Chl a to capture more light for cells, as excessive energy can harm the photosynthesis and growth of algae (Gao et al., 2012;Xu and Gao, 2012).

Effect of pCO 2 and P on nitrogen assimilation
The higher pCO 2 level noticeably enhanced the nitrate uptake rate in S. muticum regardless of P concentration in the present study.This could be attributed to the increased NRA at the condition of higher pCO 2 .The enhanced NRA at the conditions of high CO 2 was also reported in U. rigida (Gordillo et al., 2001), Hizikia fusiforme (Zou, 2005), P. haitanensis (Liu and Zou, 2015), and Corallina officinalis (Hofmann et al., 2013), as well as in the higher plants Plantago major (Fonseca et al., 1997) and tomatoes (Yelle et al., 1987).Taken together, these findings indicate that the response of NRA in plants to elevated CO 2 may be homogeneous.The higher P level also enhanced the nitrate uptake in S. muticum regardless of the pCO 2 level, which could be partially due to the increased NRA at higher P.This is very evident at lower pCO 2 .However, the higher P level decreased the NRA at higher pCO 2 , which did not lead to reduced nitrate uptake.This indicates that there should be other mechanisms to account for the promoting effect of the higher P level on the nitrate uptake.One possible mechanism is the higher P level increasing the availability of ATP required for the active uptake of nitrate across the plasma membrane.The phenomenon of ATP concentration increasing with P level has been found in higher plants (Olivera et al., 2004;Rychter et al., 2006).Apart from S. muticum, the positive effect of a higher P level on nitrate uptake was also reported in the red macroalgae Gracilaria lemaneiformis (Xu et al., 2010) and the higher plant Phaseolus vulgaris (Gniazdowska and Rychter, 2000).The increased nitrate uptake, NRA, and soluble protein at higher P in the present study suggest that high P availability promoted nitrogen assimilation in S. muticum.It is worth noting that the nitrate uptake rates were commonly higher than the corresponding reduction rates of NO − 3 to nitrite NO − 2 by nitrate reductase in the present study, which might be due to the intercellular nitrate storage (Collos, 1982;Lartigue and Sherman, 2005) and the underestimation of RNA measured by the in situ assay (Lartigue and Sherman, 2002).The higher P level increased the nitrate uptake rate and soluble protein at both lower pCO 2 and higher pCO 2 , but it only increased the NRA in S. muticum at lower pCO 2 in the present study.Surprisingly, it decreased the NRA at higher pCO 2 .There may be more than one reason related to interaction of pCO 2 and P. High pCO 2 , on the one hand, could enhance photosynthetic carbon fixation and thus growth by supplying sufficient CO 2 .On the other hand, it also results in the decrease of pH and the increase of seawater acidity, which can disturb the acid-base balance on the cell surface of algae (Flynn et al., 2012).Algae may accordingly allocate additional energy to act against the acid-base perturbation in some way.This hypothesis is supported by increased respiration at higher pCO 2 and higher P in the present study.The increased soluble protein and decreased NRA at higher pCO 2 and higher P suggest that some H + transport-related protein, such as plasma membrane H + -ATPase, might be synthesized to counteract the acid-base perturbation caused by increased pCO 2 and H + .The additional production of an H + transport-related protein, like plasma membrane H + -ATPase, could competitively decrease the synthesis of nitrate reductase.This hypothesis needs further experimental evidence to confirm, even though it could explain the results in the present study.

Connection between carbon and nitrogen assimilation
The increased net photosynthetic rate at higher pCO 2 and higher P did not result in higher soluble carbohydrates compared to higher pCO 2 and lower P.The additional ATP produced by photosynthetic electron transport higher pCO 2 and higher P may be drawn to nitrogen assimilation as more soluble protein was synthesized at higher pCO 2 and higher P. The additional energy allocation to protein synthesis, possibly an H + transport-related protein, to maintain the balance of acidbase hindered the increase of growth, which may be the rea-son that the higher P increased the net photosynthetic rate but not the growth rate at higher pCO 2 .Although synthesized protein can also contribute to the increase of thalli weight, it is not as energy-effective as carbohydrates (Norici et al., 2011;Raven, 1982).It seems that S. muticum tends to maintain a steady state in vivo, even if it can sacrifice growth to some extent, considering that the regulation of the intracellular acid-base balance is crucial for organismal homoeostasis (Flynn et al., 2012;Smith and Raven, 1979).The increased respiration at HC was also demonstrated in G. lemaneiformis (Xu et al., 2010) and U. prolifera (Xu and Gao, 2012).The respiration at higher pCO 2 and lower P did not increase compared to at lower pCO 2 and lower P in the present study, suggesting that action against acid-base perturbation did not commence.The acid-base perturbation at higher pCO 2 and lower P may lead to the decreased photosynthetic rate compared to that at lower pCO 2 and lower P.

Conclusion
Our study, for the first time, demonstrates the combined effects of elevated pCO 2 and P enrichment on the physiological traits of a golden alga, S. muticum.It suggests that the current ocean environment is both CO 2 and P limited for the photosynthesis and growth of S. muticum.Therefore, future ocean acidification and eutrophication may promote the growth of S. muticum and thus the occurrence of golden tide events.S. muticum tends to maintain homoeostasis by taking advantage phosphate at the cost of growth.Accordingly, the combination of ocean acidification and eutrophication may not boost golden tides further compared to ocean acidification or eutrophication alone.

Data availability
The data to this paper can be found in the Supplement.
The Supplement related to this article is available online at doi:10.5194/bg-14-671-2017-supplement.

Figure 1 .
Figure1.Relative growth rate (RGR) of S. muticum grown at different pCO 2 and P conditions for 13 days.Data are reported as means ±SD (n = 3).LCLP is the low pCO 2 and low P condition, LCHP is the low pCO 2 and high P condition, HCLP is the high pCO 2 and low P condition, and HCHP is the high pCO 2 and high P condition.Different letters above the error bars indicate significant differences between treatments (P < 0.05).

Figure 2 .
Figure2.Net photosynthetic rate (NPR) of S. muticum after being grown at different pCO 2 and P conditions for 13 days.Data are reported as means ±SD (n = 3).LCLP is the low pCO 2 and low P condition, LCHP is the low pCO 2 and high P condition, HCLP is the high pCO 2 and low P condition, and HCHP is the high pCO 2 and high P condition.Different letters above the error bars indicate significant differences between treatments (P < 0.05).

Figure 4 .
Figure 4. Chl a content of S. muticum after being grown at different pCO 2 and P conditions for 13 days.Data are reported as means±SD (n = 3).LCLP is the low pCO 2 and low P condition, LCHP is the low pCO 2 and high P condition, HCLP is the high pCO 2 and low P condition, and HCHP is the high pCO 2 and high P condition.Different letters above the error bars indicate significant differences between treatments (P < 0.05).

Figure 7 .
Figure 7.The contents of soluble carbohydrates (a) and protein (b)of S. muticum after being grown at different pCO 2 and P conditions for 13 days.Data are reported as means ±SD (n = 3).LCLP is the low pCO 2 and low P condition, LCHP is the low pCO 2 and high P condition, HCLP is the high pCO 2 and low P condition, and HCHP is the high pCO 2 and high P condition.Different letters above the error bars indicate significant differences between treatments (P < 0.05).

Table 2 .
The carbon-saturating maximum photosynthetic rate (V max , µmol O 2 g −1 FW h −1 ) and half saturation constant (K 0.5 , mM) for S. muticum cultured under different pCO 2 and P conditions for 13 days.