Future CO2-induced seawater acidification mediates the physiological performance of a green alga Ulva linza in different photoperiods

Photoperiods have an important impact on macroalgae living in the intertidal zone. Ocean acidification also influences the physiology of macroalgae. However, little is known about the interaction between ocean acidification and photoperiod on macroalgae. In this study, a green alga Ulva linza was cultured under three different photoperiods (L: D = 8:16, 12:12, 16:8) and two different CO2 levels (LC, 400 ppm; HC, 1,000 ppm) to investigate their responses. The results showed that relative growth rate of U. linza increased with extended light periods under LC but decreased at HC when exposed to the longest light period of 16 h compared to 12 h. Higher CO2 levels enhanced the relative growth rate at a L: D of 8:16, had no effect at 12:12 but reduced RGR at 16:8. At LC, the L: D of 16:8 significantly stimulated maximum quantum yield (Yield). Higher CO2 levels enhanced Yield at L: D of 12:12 and 8:16, had negative effect at 16:8. Non-photochemical quenching (NPQ) increased with increasing light period. High CO2 levels did not affect respiration rate during shorter light periods but enhanced it at a light period of 16 h. Longer light periods had negative effects on Chl a and Chl b content, and high CO2 level also inhibited the synthesis of these pigments. Our data demonstrate the interactive effects of CO2 and photoperiod on the physiological characteristics of the green tide macroalga Ulva linza and indicate that future ocean acidification may hinder the stimulatory effect of long light periods on growth of Ulva species.


INTRODUCTION
Due to the activities of humans, the concentration of atmospheric CO 2 has increased to 400 ppm from 278 ppm during the pre-industrial revolution (Gattuso et al., 2015). It is estimated that the oceans, as a CO 2 sink, have taken up approximately 48% of the fossil-fuel and cement-manufacturing emissions (Sabine et al., 2004). The concentration of atmospheric CO 2 has been predicted to reach almost 1,000 ppm by the end of 21st contributing majority of green tides and U. linza is one of species causing green tides in the Yellow Sea (Fletcher, 1996;Kang et al., 2014;Gao et al., 2017a;Gao et al., 2017b). However, little is known about the interactive effects of ocean acidification and photoperiod on U. linza. Previous study showed that the effect of ocean acidification on diatoms was related to light intensity (Gao et al., 2012). In this study, based on previous studies, we hypothesized the effect of ocean acidification on U. linza may dependent on photoperiod. To test our hypothesis, the physiological responses of the green macroalga Ulva linza under two different CO 2 levels and three different photoperiods were examined.

MATERIALS & METHODS
Thalli collection and culture conditions U. linza was collected from the coastal water of Gaogong peninsula (119.3 • E, 34.5 • N), Lianyungang, Jiangsu Province, China. Gaogong peninsula is a public place and no approval is required for collecting naturally growing Ulva species in China because Ulva species cause green tides in coastal waters of the Yellow Sea. U. linza was identified by morphological characters (Ma et al., 2009). The samples were transferred to the laboratory in a portable cooler (4-6 • C) box within one hour. Healthy thalli (first observed by color and then checked with maximum quantum yield of PSII) were selected and cleaned with filtered natural seawater to remove sediments, visible epiphytes and attached animals. Thalli were pre-cultured in a 500 ml flask in an illuminated incubator (GXZ-500B, Ningbo, China) at 20 • C, with the illumination intensity set at 150 µmol photons m −2 s −1 (12L:12D). Sterilized filtered seawater (salinity 30, supplied with 8 µM NaH 2 PO 4 and 60 µM NaNO 3 ) was used as culture medium and the medium was bubbled with air before being renewed every two days.
To maintain the pH NBS at about 8.12 (LC) and 7.78 (HC) under different photoperiods, the increased algal biomass were removed constantly and the medium was changed every two days and daily variations in pH were maintained at less than 0.05. The experiment lasted 9 days, the physiological factors were taken on the last 3 days.

Estimate of carbonate system parameters
The seawater pH was monitored with a pH meter (pH 700, Eutech Instruments, Singapore) and total alkalinity (TA) was calculated by titrations (Gao et al., 2019). Other parameters of the carbonate system were obtained with CO2SYS software (Pierrot, Lewis & Wallace, 2006), the equilibrium constants K 1 and K 2 for carbonic acid dissociation (Roy et al., 1993).

Measurement of growth
The length of U. linza was recorded every 2 days. The relative growth rate (RGR) of thalli were calculated across the 6-day period before other physiological parameters were measured during the following 3 days. RGR was calculated as follows: RGR = ln (W t /W 0 )/t, where W t is the length after t days culture, W 0 is the initial length.

Chlorophyll fluorescence measurements
Chlorophyll a fluorescence in U. linza was measured with a portable PAM (Pulseamplitude-modulation; AquaPen-P AP-P 100, Chech). Algae were dark adapted for 15 min before the experiment. The parameters were calculated according to the following equations: rETR = PAR × Y(II) × 0.84 × 0.5 (Schreiber, 2004;Zhang, Zhang & Yang, 2017), where rETR is the relative electron transport rate; PAR is the actinic light; Y(II) is the effective quantum yield of PSII. NPQ = Fm − Fm'/1 (Bilger & Schreiber, 1986), where NPQ is the non-photochemical quenching; Fm is the maximum fluorescence value of U. linza when they were adapted in the dark for 15 min; Fm' is the maximum fluorescence value of U. linza under actinic light conditions.

Respiration rate measurements
The respiration rate of U. linza was measured using a Clark-type oxygen electrode (YSI Model 5300, Yellow Springs Instrument Co., USA). The thallus was cut into 1 cm long segments with scissors and the thalli were placed in culture conditions for at least 1 h to decrease the effects of cutting damage. Approximately 0.01 g fresh weight of thalli were placed in the reaction chamber with 8 ml medium. Temperature was controlled at 20 • C with a circulating water bath. The decrease of the oxygen content in the seawater in darkness with seven minutes was defined as the respiration rate.

Measurement of photosynthetic pigments
Chlorophyll a and b were extracted from thalli (about 10 mg FW) with 5 ml methanol at 4 • C for 24 h in the dark. The absorption values were obtained at 652 nm, 663 nm and 665 nm using an ultraviolet spectrophotometer (Ultrospect 3300 pro; Amersham Bioscience, Sweden). The contents of the Chl a and Chl b were estimated using the method of Porra, Thompson & Kriedemann (1989).

Data analysis
All the data are shown as mean ± SD. Origin 9.0 and SPSS 18.0 were used to analyze data. Two-way ANOVA was used to assess the interactive effects of CO 2 levels and photoperiods on relative growth rate, chlorophyll fluorescence parameters, respiration rate and pigment content of U. linza. One-way ANOVA was used to analyze differences under the same conditions. Tukey HSD was conducted for post hoc investigation. Confidence intervals were set at 95%. Data are the mean ± SD (n = 3). DIC, dissolved inorganic carbon; TA, total alkalinity.

RESULTS
Both the elevated CO 2 levels and photoperiod altered carbonate parameters in seawater, and they both had an interactive effect (Tables 1 and 2). The elevated CO 2 decreased pH and CO 2− 3 , increased p CO 2 , DIC, HCO − 3 and CO 2 in the seawater. Increased photoperiod did not affect carbonate parameters at LC but the longest photoperiod increased DIC, HCO − 3 and TA compared to shortest photoperiod. The two-way ANOVA showed that elevated CO 2 and photoperiod had an interactive effect, and both elevated CO 2 levels and the photoperiods had a significant effect on the RGR of U. linza ( Fig. 1 and Table 3). At LC, the RGR of adult U. linza increased with the extended light periods, and the highest RGR occurred at a L: D of 16:8. The effect of CO 2 also varied with photoperiod. Higher CO 2 levels enhanced RGR at L: D of 8:16, but had no effect at 12:12 and reduced it at 16:8.
The Yield and NPQ were measured under different CO 2 levels and photoperiod conditions (Fig. 2). Two-way ANOVA showed that elevated CO 2 and photoperiod had an interactive effect on Yield (Table 4). Higher CO 2 levels increased Yield when thalli were cultured under photoperiods of 8:16 and 12:12 but reduced it under 16:8. Photoperiod had the main effect on NPQ (Table 4). At LC, thalli cultured at L: D of 16:8 had higher NPQ compared to L: D of 8:16 while the difference between 8:16 and 12:12 was insignificant. At HC, NPQ increased with the increase in photoperiod although the increase was not statistically significant. The elevated CO 2 had neutral effect on NPQ of U. linza.
Maximum rETR (rETRmax), efficiency of electron transport (α), and saturating irradiance (I k ) were calculated from the rapid light curves (Fig. 3, Table 5). Photoperiod and elevated CO 2 levels had an interactive effect, and elevated CO 2 levels had a main effect on light-saturated electron transport rate (rETRmax) ( Table 6). Higher CO 2 levels increased rETRmax during the 8:16 and 12:12 photoperiods, but did not affect it at 16:8. A similar pattern was also found for α. In contrast to rETRmax and α, CO 2 did not affect I k while photoperiod had the main effect on it. At LC, I k increased when L: D rose from 8:16 to 12:12 but did not change with the further increase in photoperiod. At HC, I k did not change when L: D rose from 8:16 to 12:12 but was enhanced when L: D increased to 16:8 (Table 5). Notes. CO 2 *photoperiod means the interactive effects of CO 2 and photoperiod, df mean degree of freedom and F means the value of F statistic, and Sig. means p-value.  Notes. CO 2 *photoperiod means the interactive effects of CO 2 and photoperiod, df mean degree of freedom and F means the value of F statistic, and Sig. means p-value.
In addition to photosynthetic parameters, the effects of p CO 2 and photoperiod on the respiration rate of adult U. linza were also investigated (Fig. 4). Photoperiod and elevated CO 2 levels had an interactive effect, and both photoperiod and elevated CO 2 levels had the primary effect on the respiration rate of U. linza (Table 7). Higher CO 2 levels did not affect the respiration rate at photoperiods of 8:16 or 12:12 but increased it by 56.39% at a photoperiod of 16:8.
Changes in photosynthetic pigments of U. linza grown under various conditions are shown in Fig. 5. Two-way ANOVA showed that CO 2 and photoperiod had an interactive effect, and both CO 2 levels and the photoperiods had the main effect on the Chl a content of U. linza (Table 8). Prolonged light periods reduced the synthesis of Chl a in thalli although the difference between the photoperiods of 8:16 and 12:12 at HC was not statistically significant. Higher CO 2 levels reduced Chl a at the photoperiods of 8:16 and 12:12 but did not affect it at a L: D of 12:12. The same trend was found for Chl b. The Chl a/b ratios were all greater than 1.0, suggesting a higher synthesis of Chl a than Chl b under all culture  conditions. Photoperiod and elevated CO 2 levels had an interactive effect on the Chl a/b ratio (Table 8); the higher CO 2 levels increased the ratio at a photoperiod of 12:12 but not at the other photoperiods.     Notes. CO 2 *photoperiod means the interactive effects of CO 2 and photoperiod, df mean degree of freedom and F means the value of F statistic, and Sig. means p-value.

DISCUSSION
In the present study, at LC, extended photoperiods had a positive effect on the relative growth rate of adult U. linza, similar to previous studies on Laminaria sacharina, Porphyra umbilicalis and Ulva prolifera (Fortes & Lüning, 1980;Green & Neefus, 2016;Li et al., 2018). Carbon isotope fractionation experiments suggested that extended photoperiods could enhance growth by influencing inorganic carbon capture and fixation rate in algae (Rost et al., 2003). This hypothesis is supported by the present study where extended photoperiods increased the maximum quantum yield in PS II. On the other hand, the highest growth rates of Compsopogon coeruleus were obtained in shorter light periods (L: D = 8:16) (Zucchi & Necchi, 2001), the highest growth rates of Porphyra umbilicalis was found under neutral Notes. CO 2 *photoperiod means the interactive effects of CO 2 and photoperiod, df mean degree of freedom and F means the value of F statistic, and Sig. means p-value.  Cetin, 2015). Therefore, the effects of photoperiod on algae appear to be species-specific. Although Ulva has efficient mechanisms for CO 2 concentration, the growth of Ulva can be enhanced by elevated CO 2 (Young & Gobler, 2016;Gao et al., 2016a;Gao et al., 2017a;Gao et al., 2017b). However, in this study, we found that the effects of elevated CO 2 levels on the relative growth rate of U. linza depended on photoperiod. CO 2 enhanced the growth of adult U. linza under light/dark conditions of 8:16, had no effect on growth of U. linza under a L: D of 12:12, and reduced the relative growth rate under a L: D of 16:8. High CO 2 levels can down-regulate algal CO 2 concentration mechanisms (CCMs), meaning that energy would be saved, and thus enhancing the relative growth rate of algae (Gao et al., 2012;Gao et al., 2016a;Raven, Beardall & Sánchez-Baracaldo, 2017). This is supported by decreased pigment synthesis in thalli at higher CO 2 levels. However, higher CO 2 levels did not affect the growth rate of U. linza at medium photoperiods. The neutral effects of CO 2 on the growth of U. rigida (Rautenberger et al., 2015) and U. linza (Gao et al., 2018) were also reported. We speculate that this is a compromise between the positive effects of elevated CO 2 and negative effects of decreased pH. The negative effect of decreased pH on growth was documented for the brown alga Sargassum muticum (Xu et al., 2017). Algae might need to consume additional energy to act against acid-based perturbation caused by decreased pH, leading to reduced growth (Xu et al., 2017). This is supported by an enhanced respiration rate at the higher CO 2 levels in this study. The phenomenon of an increased respiration rate of algae under elevated CO 2 concentrations was found in Hizikia fusiformis (Zou, 2005), the microalgae Phaeodactylum tricornutum (Wu, Gao & Riebesell, 2010) and Emiliania huxleyi (Jin et al., 2015).
Furthermore, the higher CO 2 level reduced the growth rate of U. linza under the longest photoperiod in the present study. This may be due to the combination of down-regulated CCMs and excess light energy. The operation of CCMs is an energy-consuming process and the down-regulation of CCMs can result in additional energy (Raven, Beardall & Sánchez-Baracaldo, 2017). Higher light can usually reduce algal photosynthetic activity (Singh & Singh, 2015;Gao et al., 2016a). The energy saved due to down-regulation of CCMs at HC combined with high light intensity could synergistically damage the algal photosystem and photosynthetic rate (Gao et al., 2012;Gao et al., 2016a). Although light intensity did not change among different CO 2 treatments, the lengthened photoperiod may have similar effect to increased light intensity. This argument is supported by increased NPQ at longer photoperiods as NPQ is photo-protective process to dissipate excess light energy. Higher CO 2 levels also reduced maximum quantum yield and stimulated the respiration rate of U. linza during the longest photoperiod in this study, leading to the decrease in growth. Enhanced respiration rate is a signal that organisms are fighting against stress and damage (Xu et al., 2017).
In addition to growth, the interactive effect of CO 2 and photoperiod was also found in photosynthetic parameters. For instance, elevated CO 2 increased Yield, rETRmax and α at shorter photoperiods (L: D of 8:16 and 12:12) but did not affect rETRmax and reduced Yield at longest photoperiod (L: D of 16:8). These findings indicate the close connection between growth and photosynthesis in terms of responding to the combination of CO 2 and photoperiod. It is worth noting that the interaction of CO 2 and photoperiod on growth of U. linza in this study is different from the findings in U. prolifera reported by Li et al. (2018). Elevated CO 2 increased growth of U. prolifera at all three photoperiods (L: D of 12:12, 10:14 and 16:8). The different results may be due to differential physiological property between U. linza and U. prolifera. It has been documented that U. prolifera has a higher tolerance to high light intensity compared to U. linza (Cui et al., 2015). The strong capacity in dealing with high light intensity could contribute to the effect of elevated CO 2 and prolonged photoperiod on growth U. linza was still positive. Integrating our findings with Li et al. (2018) study, the interaction of CO 2 and photoperiod on Ulva species would be species dependent.

CONCLUSIONS
This work is the first attempt to clarify the interaction between light/dark and elevated CO 2 levels on the physiological responses of Ulva linza. We found that the effect of OA on U. linza depended on photoperiod. Outbreaks of green tides during spring and summer in China occur when the photoperiod is reaching its peak. Our findings indicate future OA may hinder the occurrence of green tides dominated by U. linza in combination with extended photoperiods. More environmental factors, such as temperature and nutrient levels, need to be investigated to obtain a more comprehensive understanding on development of green tides in future oceans.

ADDITIONAL INFORMATION AND DECLARATIONS Funding
This study was supported by the Natural Science Foundation of Jiangsu Province (Nos. BK20161295), the Six Talents Peaks in Jiangsu Province (JY-086), and the ''333'' project of Jiangsu Province and Priority Academic Program Development of Jiangsu Higher Education Institutions. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.