INTRODUCTION
Seagrass beds are among the most productive marine ecosystems, forming habitats for many species and playing significant roles in coastal protection and climate stability (Hughes et al., Reference Hughes, Stachowicz and Williams2009; Fourqurean et al., Reference Fourqurean, Duarte, Kennedy, Marbà, Holmer, Mateo, Apostolaki, Kendrick, Krause-Jensen, McGlathery and Serrano2012; Ondiviela et al., Reference Ondiviela, Losada, Lara, Maza, Galván, Bouma and van Belzen2014). In recent decades there has been a worldwide loss of seagrass (Waycott et al., Reference Waycott, Duarte, Carruthers, Orth, Dennison, Olyarnik, Calladine, Fourqurean, Heck, Randall Hughes, Kendrick, Kenworthy, Short and Williams2009). The overarching reasons for this are habitat destruction and pollution, and the loss of water clarity through sediment loading and eutrophication (Short et al., Reference Short, Polidoro, Livingstone, Carpenter, Bandeira, Bujang, Calumpong, Carruthers, Coles, Dennison, Erftemeijer, Fortes, Freeman, Jagtap, Kamal, Kendrick, Kenworthy, La Nafie, Nasution, Orth, Prathep, Sanciangco, van Tussenbroek, Vergara, Waycott and Zieman2011). About one third of the global area of seagrass meadows has been lost since the 19th century and the current rate of loss is about 7% year−1, which is accelerating (Waycott et al., Reference Waycott, Duarte, Carruthers, Orth, Dennison, Olyarnik, Calladine, Fourqurean, Heck, Randall Hughes, Kendrick, Kenworthy, Short and Williams2009). The importance and vulnerability of seagrass beds underline the need for a thorough understanding of their ecology and adaptability to be able to prevent further losses and help improve restoration activities.
Intertidal seagrasses populate a very variable habitat, characterized by strong fluctuations in temperature, desiccation and irradiance (Björk et al., Reference Björk, Uku, Weil and Beer1999; Silva & Santos, Reference Silva and Santos2003). In particular, seasonal and diurnal fluctuations of the light conditions in the intertidal can be extreme, and at low tide seagrasses can be exposed to irradiances >2000 µmol photons m−2 s−1 of photosynthetically active radiation (PAR). With the incoming tide, in situ irradiance decreases rapidly, particularly in turbid water (Serodio & Catarino, Reference Serodio and Catarino1999). In clear water the decrease in irradiance upon submersion is less pronounced, but short-term light flashes of PAR >5000 µmol photons m−2 s−1 can occur due to the wave-focusing effect (Wing & Patterson, Reference Wing and Patterson1993; Schubert et al., Reference Schubert, Sagert and Forster2001). However, the highest in situ irradiance normally coincides with mid-day low water. In contrast, irradiance at a mid-day high tide is considerably less, but stronger variations in light climate during the day occur (increase of irradiance in the morning, decrease with incoming tide, increase again with outgoing tide, decrease with decreasing light in the evening). Changing light conditions in the intertidal suggests a requirement for rapid adjustment by intertidal photosynthetic organisms, which need to optimize light harvesting when irradiance is low and to protect their photosynthetic apparatus against photodamage when irradiance is high. Mechanisms of photoprotection comprise processes in the antenna complexes of PSII that decrease the transfer of excitation energy to the reaction centre (Franklin et al., Reference Franklin, Osmond, Larkum, Larkum, Douglas and Raven2003). These are dissipative reactions, resulting in a transient reduction in photosynthetic efficiency, which is rapidly reversible. If the decrease in photosynthetic efficiency is due to photoinactivation (also referred to as photodamage) there is a loss of PSII reaction centre function due to a loss of D1 protein and the reversion is much slower or even impossible (Franklin et al., Reference Franklin, Osmond, Larkum, Larkum, Douglas and Raven2003). Both terrestrial and aquatic plants have evolved adaptive physiological traits to cope with high and fluctuating irradiances (Demmig-Adams & Adams, Reference Demmig-Adams and Adams1992; Schwarz et al., Reference Schwarz, Björk, Buluda, Mtolera and Beer2000; Ensminger et al., Reference Ensminger, Xyländer, Hagen and Braune2001; Silva & Santos, Reference Silva and Santos2003; Kono & Terashima, Reference Kono and Terashima2014). While photosynthetic activity usually follows the increase and decrease of diurnal irradiance, midday depressions of photosynthetic activity have been observed (Hanelt et al., Reference Hanelt, Li and Nultsch1994). This is particularly common in macrophytes growing in the intertidal (Huppertz et al., Reference Huppertz, Hanelt and Nultsch1990; Silva & Santos, Reference Silva and Santos2003).
One widely found photoprotective mechanism is the xanthophyll cycle (XC), which includes the rapid and reversible de-epoxidation of violaxanthin (Vx) to antheraxanthin (Ax) and zeaxanthin (Zx) under high light (Demmig-Adams, Reference Demmig-Adams1990; Jahns & Holzwarth, Reference Jahns and Holzwarth2012). Under low light or overnight re-epoxidation from Zx over Ax to Vx takes place (Demmig-Adams, Reference Demmig-Adams1990). The XC is found in higher terrestrial plants (Yamamoto et al., Reference Yamamoto, Nakayama and Chichester1962; Bilger & Björkman, Reference Bilger and Björkman1990; Demmig-Adams, Reference Demmig-Adams1990), diatoms (Lavaud et al., Reference Lavaud, Rousseau, van Gorkom and Etienne2002), dinoflagellates (Ambarsari et al., Reference Ambarsari, Brown, Barlow, Britton and Cummings1997; Brown et al., Reference Brown, Ambarsari, Warner, Fitt, Dunne, Gibb and Cummings1999), brown algae (Uhrmacher et al., Reference Uhrmacher, Hanelt and Nultsch1995) and green algae (Bischof et al., Reference Bischof, Kräbs, Wiencke and Hanelt2002). Although there have been few studies documenting the XC in seagrasses, most have been conducted under laboratory conditions or in outdoor tanks (Dawson & Dennison, Reference Dawson and Dennison1996; Flanigan & Critchley, Reference Flanigan and Critchley1996; Ralph et al., Reference Ralph, Polk, Moore, Orth and Smith2002; Marín-Guirao et al., Reference Marín-Guirao, Ruiz, Sandoval-Gil, Bernardeau-Esteller, Stinco and Meléndez-Martínez2013). To date, only three studies have been carried out in situ (Collier et al., Reference Collier, Lavery, Ralph and Masini2008; García-Sánchez et al., Reference García-Sánchez, Korbee, Pérez-Ruzafa, Marcos, Domínguez, Figueroa and Pérez-Ruzafa2012; Silva et al., Reference Silva, Barrote, Costa, Albano and Santos2013). However, these focused on subtidal rather than intertidal seagrasses where radiation levels are much more variable. In our study areas (New Zealand and Germany), seagrass meadows occur predominantly in the intertidal (Hily et al., Reference Hily, van Katwijk, den Hartog, Green and Short2003; Inglis, Reference Inglis, Green and Short2003). There is clearly a lack of field studies characterizing the essential processes of photoacclimation and photoprotection in seagrass from intertidal habitats, which is the focus of this study. We hypothesize that intertidal seagrasses dynamically adjust the xanthophyll cycle as an efficient photoprotection mechanism in response to highly fluctuating light conditions.
The seagrass genus Zostera comprises nine species and inhabits the intertidal and subtidal along the coast of all continents, except Antarctica (Moore & Short, Reference Moore, Short, Larkum, Orth and Duarte2006), facilitating comparative studies in different geographic regions. Here we wanted to separate general vs specific responses to varying radiation, temperature, and local and regional environmental effects as well as genetic/intrinsic differences between congeneric species. To achieve this we chose two distant study areas – one located in the northern hemisphere in a cold temperate climate (off the island of Sylt in the Northern Frisian Wadden Sea, North Sea) and the other in the southern hemisphere in a warm-temperate climate (Tauranga Harbour, North Island, New Zealand). To take into account different tidal settings we conducted two whole day surveys during summer from pre-dawn until sunset, one when low tide occurred at noon and one when high tide occurred at noon. We used these settings for an in-depth characterization of light-protective mechanisms of seagrass in highly variable light environments.
MATERIALS AND METHODS
Study sites and species
Tauranga Harbour is a meso-tidal estuarine lagoon at the east coast (Bay of Plenty) of the North Island of New Zealand with an average tidal range of ~1.5 m. The study site was located in the northern reaches of the harbour on the outer part of a small sand spit enclosing a sub-estuary (for coordinates see Table 1). The site can be characterized as moderately wave exposed. Median grain sizes of the sediment at this site are around 250 µm and the mud content (<63 µm) is low (<5%) (Kohlmeier et al., Reference Kohlmeier, Pilditch, Bornman and Bischof2014).
Abiotic conditions and coordinates of the study sites in Germany (GER) and New Zealand (NZ).
Noon tide state: HT, high tide at noon; LT, low tide at noon.
The German study area is called Königshafen and is located in the Wadden Sea off the island of Sylt. The area/embayment is well studied and a detailed description can be found in Reise (Reference Reise, Billings, Golley, Lange, Olson and Remmert1985), and a review on the current status in Schumacher et al. (Reference Schumacher, Dolch and Reise2014). The average tidal range is 1.8 m. The study site is called Tonnenlegerbucht and is located in the south-east of the Königshafen (for coordinates see Table 1). The small bay is enclosed by a sand spit in the east and by a dyke to the south and the west. The site can be described as sheltered and is characterized by a mean grain size of 140 µm (Lefebvre and Laurens, unpublished data).
Zostera muelleri Irmisch ex Ascherson (1867) is the genus’ representative in the South-west Pacific and its range covers the coasts of Australia, Papua New Guinea and New Zealand (Moore & Short, Reference Moore, Short, Larkum, Orth and Duarte2006; Matheson et al., Reference Matheson, Lundquist, Gemmill and Pilditch2011). Zostera marina Linnaeus (1753) (eelgrass) is a very widespread species in the northern hemisphere (Moore & Short, Reference Moore, Short, Larkum, Orth and Duarte2006) and in the German Wadden Sea it grows in vast mixed meadows with the dwarf eelgrass Zostera noltei Hornemann (1832) as well as in smaller monospecific meadows. The meadow studied here was dominated by Z. marina, with some small patches of Z. noltei.
Set-up and sampling scheme
Sampling was conducted under summer conditions in January/February 2011 in New Zealand and in July 2011 in Germany. Whole day surveys, one on a day where low tide was at noon (hereafter LT) and one with high tide at noon (hereafter HT) were carried out at both study sites. One survey included continuous recording of incident atmospheric light (as PAR), regular sampling of photosynthetic parameters (via PAM-fluorometry), pigments and in situ light from pre-dawn until sunset. Nine plots, three at the upper, middle and lower levels of the respective seagrass meadow, were marked at each site on the day before the survey. Pilot studies had shown that there were no strong differences in physical, chemical or biological parameters among levels. The positioning of the nine replicate plots at the three levels was done to take natural variability due to vertical position (and thus potential differences in, for example, submersion times and desiccation stress) into account, if necessary. The distance between plots at each level was 10 m and distance between levels was ~50 m. Sampling for PAM-fluorometry, pigment content and in situ light was conducted at ~3 h intervals. For each sampling, one seagrass shoot per plot was chosen. PAM-measurements and sampling for pigment concentrations were conducted on the second or third youngest leaf per shoot. When the seagrass was submerged samples were taken by snorkelling, taking care to prevent sediment resuspension and changes in water transparency.
PAM-Fluorometry
For determination of the effective and maximum quantum yields of PSII, measurements of chlorophyll a fluorescence were conducted with a submersible pulse amplitude modulated fluorometer (Diving-PAM, Walz, Effeltrich, Germany). The effective quantum yield (ΔF/Fm’) was determined using a sample holder (Universal sample holder, Walz, Effeltrich, Germany), which also ensured simultaneous measurements of incident light next to the leaf. After determination of ΔF/Fm’ the leaf was darkened using a leaf clip (Diving LC, Walz, Effeltrich, Germany). After 10 min of darkening the maximum quantum yield (Fv/Fm) was measured. Non-photochemical quenching (NPQ) was calculated: NPQ = (Fm-Fm’)/Fm’. After the PAM-measurements leaf material was sampled for pigment analysis.
Light and temperature
Atmospheric light was measured using a 2π cosine-corrected sensor (LI-COR Li-192, Lincoln, USA) and logged using a data logger (Li-1500, Lincoln, USA). The sensor was located near the study site and measurements were taken every minute, and the mean over 5 min was logged. In situ light was measured next to the seagrass leaves using the mini-quantum sensor of the PAM-fluorometer simultaneously with PAM-measurements. The sensor was calibrated against a 2π cosine-corrected sensor (LI-COR Li-192, Lincoln, USA). Temperature was logged in situ during the whole survey periods at 5 min intervals with TidBit® temperature loggers (Onset Computer Corporation, Bourne, USA).
Pigments
The leaf material previously used for in situ measurements of PAM-fluorescence was cleaned of sediment and visible epiphytes, placed in vials, kept dark, brought back to shore within 15 and 25 min and frozen in liquid nitrogen. Samples were stored until analysis at −80°C. Pigment analysis was conducted following the protocol described in Kohlmeier et al. (Reference Kohlmeier, Pilditch, Bornman and Bischof2014). In short, the lyophilized samples were ground using an homogenizer (Fast Prep, Lysing Matrix D, MP Biomedical, Germany) and extracted in 90% acetone for 24 h at 4°C. Analyses were performed by reversed phase HPLC, applying a gradient according to Wright et al. (Reference Wright, Jeffrey, Mantoura, Llewellyn, Bjørnland, Repeta and Welschmeyer1991). Peaks were detected at 440 nm, and identified and quantified by co-chromatography with pigment standards for chlorophyll a, chlorophyll b, lutein, antheraxanthin (Ax), zeaxanthin (Zx), violaxanthin (Vx), neoxanthin and β-carotene (DHI Lab Products, Hørsholm, Denmark). All extraction procedures were conducted in a low-light environment.
The de-epoxidation state was calculated according to the formula
$${\rm DPS} = \displaystyle{{{\rm Zx} + 0.5{\rm Ax}} \over {{\rm Vx} + {\rm Ax} + {\rm Zx}}}$$
(Thayer & Björkman, Reference Thayer and Björkman1990).
Although we endeavoured to minimize the time between sampling and freezing of the pigment samples, some epoxidation may have taken place before freezing and thus DPS in situ may have been higher.
Statistics
Statistical analyses were performed using R (Version 2.14.0) and R Commander (1.8–4). One-way analysis of variance (ANOVA) followed by Tukey post-hoc tests were performed on pooled pigment data and on selected data from the daily cycles (see below). Where data did not meet the assumptions of normality and variance homogeneity, data transformations were conducted prior to analysis. In cases where data transformations did not lead to normal distribution and/or homogeneity of data, a non-parametric test (Kruskal–Wallis) was performed on un-transformed data, followed by non-parametric post-hoc tests (Wilcoxon with Bonferroni correction).
For NZ_HT there were two sampling times around mid-day and these results were pooled for subsequent analyses. In order to facilitate direct comparisons between the two sites and the two contrasting tidal settings, selected time points within the daily cycles were analysed. To compare the noon LT and HT settings we analysed the New Zealand (NZ) data. To compare the two study areas, the differences between German (GER) HT and NZ_HT were analysed. We could not include the GER_LT data, due to unfavourable weather conditions on the sampling day.
RESULTS
Daily fluence rates during the sampling period were comparable at both sites (~10% lower at the GER site, see Table 1), despite the different geographic latitudes, with longer days at the GER site, but higher maximum irradiance at the NZ location. Maximum incident irradiance at the GER site was 1680 and 1820 µmol photons m−2 s−1 for HT and LT, respectively. At the NZ site the maximum incident irradiance was 2220 and 2250 µmol photons m−2 s−1 for HT and LT, respectively. Maximum in situ irradiance was comparable (between 1300 and 1500 µmol photons m−2 s−1) for both GER surveys and the NZ_HT survey, but was nearly twice as high for NZ_LT (~2900 µmol photons m−2 s−1). On the day of the GER_LT survey the light/weather conditions were unfavourable, with rapidly changing incident light (due to clouds), markedly lower daily fluence (40 mol photons m−2 d−1) and low maximum in situ irradiance of PAR (Table 1, Figure 1).
German site – atmospheric and in situ irradiance (as photosynthetically active radiation – PAR) at noon HT (A) and at noon LT (B), effective quantum yield (ΔF/Fm′) and optimum quantum yield (Fv/Fm) of PSII at noon HT (C) and noon LT (D); non-photochemical quenching (NPQ) and de-epoxidation state (DPS) at noon HT (E) and at noon LT (F); xanthophyll pigment (violaxanthin Vx, antheraxanthin Ax, zeaxanthin Zx) concentrations (as ratio per Chl a) at noon HT (G) and at noon LT (H); N = 3 at pre-dawn and post-sunset measurements of the LT survey, all others N = 9, except atmospheric PAR: n = 1; error bars represent ± 1 SE. The grey area indicates times when the sites were submerged.
Mean in situ temperatures were between 20 and 22°C at both sites and for all four surveys, with comparable minimum temperatures and slightly higher maximum temperatures at the NZ site (Table 1). Submersion times at the study sites ranged between 4 and 6 h, depending on the monthly tidal status and the location within the seagrass meadow (see Table 1).
In all surveys the same general patterns were observed. The effective quantum yield (ΔF/Fm’) decreased with increasing light and it increased again when the light was decreasing (Figures 1C, D and 2C, D). The optimum quantum yield of PSII, Fv/Fm (as a measure of the intrinsic photochemical quantum yield of PS II) decreased slightly in some cases, but it increased again at the last measurement (in the evening) (Figures 1C, D and 2C, D). Both the NPQ and the DPS were increasing and decreasing in parallel with incident light conditions (Figures 1E, F and 2E, F). The pigments of the xanthophyll cycle also showed similar responses in all four surveys: Vx decreased with increasing light and increased again in the evening (Figures 1G, and 2G, H). Ax had intermediate levels and fluctuated only slightly. Zx increased strongly with increasing light conditions, but also decreased again in the evening. A near complete conversion indicated by high levels of Zx, intermediate levels of Ax and low levels of Vx occurred when the irradiance was high (Figures 1A, B, G, H and 2A, B, G, H). Following this, there was a high DPS with levels of up to 0.8 with high irradiance conditions (Figures 1E, F and 2E, F).
New Zealand site – atmospheric and in situ irradiance (as photosynthetically active radiation – PAR) at noon HT (A) and at noon LT (B), effective quantum yield (ΔF/Fm′) and optimum quantum yield (Fv/Fm) of PSII at noon HT (C) and at noon LT (D); non-photochemical quenching (NPQ) and de-epoxidation state (DPS) at noon HT (E) and at noon LT (F); xanthophyll pigment (violaxanthin Vx, antheraxanthin Ax, zeaxanthin Zx) concentrations (as ratio per Chl a) at noon HT (G) and at noon LT (H); N = 9 except atmospheric PAR: N = 1; error bars represent ± 1 SE. The grey area indicates times when the sites were submerged.
The unfavourable weather conditions during the GER_LT meant that a restricted number of tidal conditions could be analysed statistically. For the comparison of NZ_HT and NZ_LT the data showed one clear peak or one minimum in the case of the LT study (a clear peak of irradiance, DPS and NPQ and Zx and a minimum for ΔF/Fm’, Fv/Fm and also for Vx), while the data of the HT study showed a bimodal pattern. There were statistically significant differences for PAR (P<0.001), ΔF/Fm’ (P < 0.001), DPS (P < 0.0001) and for Vx (P < 0.0001), Ax (P = 0.002) and Zx (P < 0.001) (and lutein, P = 0.014; data not shown). There were no statistically significant differences for Fv/Fm or NPQ. In a second analysis (for comparison of study areas), NZ_HT and GER_HT were compared and here we found statistically significant differences for PAR (P < 0.001), Fv/Fm (P = 0.002) and lutein (P = 0.001; data not shown). NPQ, ΔF/Fm’ and pigment concentrations were not statistically different.
Data from each survey were pooled to examine differences in mean pigment content. Absolute pigment concentrations and total pigments (Table 2) were significantly higher at the GER site than at the NZ site (P < 0.01). No statistically significant differences were observed between the two tidal settings within the GER data set. In the comparison of the NZ data set, differences were found for the absolute values (μg pigment per mg dry weight): the XC pool (Vx + Ax + Zx) was significantly higher for LT (P < 0.001). Again there were no differences within the GER data set for the relative values (pigments per Chl a). The NZ data showed that β-carotene and lutein were significantly higher when LT was at noon (P < 0.001). And, as was also seen for the absolute values, higher levels of the XC pool (Vx + Ax + Zx) per Chl a were found (P < 0.001). Also the ratio of chlorophylls to carotenoids was lower when LT was at noon (P < 0.001).
Results of pigment analysis, with the data representing mean values ± standard errors (GER_HT: N = 54; GER_LT: N = 33; NZ_HT: N = 49; NZ_LT: N = 54). For each row of data different letters (a, b, c) indicate significant differences (P < 0.05) among the four surveys.
Chl = Chlorophylls (chlorophyll a and chlorophyll b); Carot = carotenoids (β-carotene, lutein, neoxanthin, violaxanthin (Vx), antheraxanthin (Ax), zeaxanthin (Zx)).
In summary, there was a high activity of the XC with a near complete conversion of Vx to Zx (Figures 1G, H and 2G, H). We have seen similar patterns in both species with rapidly changing pigment compositions following the daily fluctuations in incident PAR (Figures 1A, B and 2A, B). For the NZ seagrass, differences were also observed in pigment concentrations (Table 2). Especially the XC pool sizes changed with tidal settings: when LT occurred at noon, there was a larger XC pool compared with when HT occurred at noon.
DISCUSSION
The results of our study are an illustrative example of the significance of the XC in a highly dynamic light environment. Under most conditions only low or intermediate levels of Ax were observed, indicating a complete de-epoxidation to Zx when levels of PAR were high. This is different to other studies on the XC in seagrasses (Ralph et al., Reference Ralph, Polk, Moore, Orth and Smith2002; Collier et al., Reference Collier, Lavery, Ralph and Masini2008; García-Sánchez et al., Reference García-Sánchez, Korbee, Pérez-Ruzafa, Marcos, Domínguez, Figueroa and Pérez-Ruzafa2012) that found only an incomplete conversion, related to an accumulation of Ax and only very low levels of Zx. This may have been due to the fact that our study was conducted in an intertidal habitat using in situ light conditions (in contrast to subtidal habitat or artificial light in other studies).
The dynamic regulation of the XC indicates its significant role in photoprotection. Values in the literature for the light saturation point (Ek) for Zostera are very variable; they range from 80–900 µmol photons m−2 s−1 (Leuschner & Rees, Reference Leuschner and Rees1993; Vermaat, Reference Vermaat1997; Ralph et al., Reference Ralph, Polk, Moore, Orth and Smith2002; Schwarz, Reference Schwarz2004). Although, we did not conduct photosynthesis vs irradiance curves, our results confirm high light adaptation for both Zostera species, as full operation of the XC was only observed under PAR irradiances above ~900 µmol photons m−2 s−1. Maximum quantum yield (Fv/Fm) dropped only to values of just below 0.5 and at sunset the Fv/Fm had recovered completely, giving no indication of prolonged photoinactivation or photodamage for either species, even at very high PAR. This is further supported by very high NPQ values and an active XC, indicating effective photoprotection mechanisms (Demmig-Adams, Reference Demmig-Adams1998; Goss & Jakob, Reference Goss and Jakob2010).
A dynamic adjustment of the XC pool sizes was also observed. When LT coincided with noon, seagrass leaves at the New Zealand site contained significantly higher concentrations of XC pigments. This is in accordance with studies that have reported higher pool sizes in terrestrial sun plants than in shade plants (Thayer & Björkman, Reference Thayer and Björkman1990; Königer et al., Reference Königer, Harris, Virgo and Winter1995; Demmig-Adams, Reference Demmig-Adams1998) or even sun and shade leaves (Björkman & Demmig-Adams, Reference Björkman, Demmig-Adams, Caldwell and Schulze1994). Other studies have shown that among closely related species there can be strong differences in the type and size of the XC (Matsubara et al., Reference Matsubara, Krause, Aranda, Virgo, Beisel, Jahns and Winter2009; Stamenković et al., Reference Stamenković, Bischof and Hanelt2014). Within one species or even one specimen, the XC pool size can increase significantly within one day (Björkman & Demmig-Adams, Reference Björkman, Demmig-Adams, Caldwell and Schulze1994) to 10 days (Eskling & Åkerlund, Reference Eskling and Åkerlund1998). In the case of Zostera, our results clearly show that XC pool sizes depended on the tidal setting (LT or HT at noon). For subtidal seagrass in tank experiments, Ralph et al. (Reference Ralph, Polk, Moore, Orth and Smith2002) found that the XC pool sizes may increase within one day. In our study we have seen an increase of the XC pool size within one week in the New Zealand data set, correlating with increasing fluence in accordance with the timing of the tides.
Pigments other than those of the XC did not show any clear cycling (data not shown), indicating that alternative energy dissipation pathways, such as the lutein-epoxide cycle that operates in some plants (García-Plazaola et al., Reference García-Plazaola, Hernández, Errasti and Becerril2002; Matsubara et al., Reference Matsubara, Krause, Aranda, Virgo, Beisel, Jahns and Winter2009), was probably not active in the two species examined. However, the levels of lutein were significantly higher in the NZ_LT plants, indicating an important role for lutein in photoprotection (Jahns & Holzwarth, Reference Jahns and Holzwarth2012) as also found for the seagrass Cymodocea nodosa (García-Sánchez et al., Reference García-Sánchez, Korbee, Pérez-Ruzafa, Marcos, Domínguez, Figueroa and Pérez-Ruzafa2012).
For the first time in intertidal seagrasses, our study has demonstrated the high activity of the XC and a near complete conversion from Vx to Zx. We observed similar patterns in both species with rapidly changing pigment compositions following the daily fluctuations in incident PAR. Moreover, at the NZ site, differences were seen in pigment concentrations especially those of the XC between the different tidal settings. Our results therefore corroborate our hypothesis, that intertidal seagrasses dynamically adjust the XC as an efficient photoprotective mechanism in response to highly fluctuating light conditions. Both the high activity of the XC and the dynamic adjustment of XC pool size to tidal conditions appear to be important adaptive features of intertidal seagrass, enabling it to thrive in an extremely variable light environment. Our results underline the known high-light adaptation of seagrasses and help explain the success of the genus Zostera in intertidal habitats. In the context of the global decline of seagrass meadows and increased sedimentation and eutrophication, further work is needed on the impact of light reduction on seagrasses.
ACKNOWLEDGEMENTS
We would like to thank Pauline Robert, Alice Lefebvre and Yuna Laurens for assistance in the field and Petra Kadel and Margit Ludwig for help with fieldwork logistics on Sylt. We also thank Jasmin Heiden, Marie Eschweiler and Britta Meyer-Schlosser for assistance with HPLC analyses.
FINANCIAL SUPPORT
This work was funded by Deutsche Forschungsgemeinschaft (DFG) through the International Research Training Group INTERCOAST (DFG grant number GRK 1598).
