Posidonia oceanica (L.) Delile is an endemic and dominant seagrass species in the Mediterranean Sea, where it forms extensive meadows from the surface down to a maximum depth of about 45 m (Procaccini et al., 2003; Boudouresque et al., 2006). Seagrass meadows play major ecological roles by enhancing biodiversity, supporting high productivity, protecting the geomorphology of the coastline, sequestering global oceanic carbon, and providing a buffering effect against ocean acidification (Duarte et al., 2005; Barbier et al., 2011; Fourqurean et al., 2017; Chou et al., 2018). However, seagrass ecosystems are currently suffering from a worldwide regression in response to several environmental stressors (Orth et al., 2006; Waycott et al., 2009; Marbà and Duarte, 2010). For instance, the reduction in the surface coverage of P. oceanica has been reported to be 34% in the past 50 years (Telesca et al., 2015). Considering that P. oceanica is a climax, slow-growing seagrass species, the natural and anthropogenic perturbations can be particularly critical, as their recovery from perturbations can be very slow or may not even recover at all (Serrano et al., 2011). Among the threats that these ecosystems are facing, eutrophication from the waste waters and aquaculture, shoreline constructions, anchoring and trawling, dredging, introduced species, and climate change (warming and rising of sea-level) are considered the major causes of the decline of P. oceanica meadows over the last decades (Boudouresque et al., 2009; Champenois and Borges, 2018). Most of these impacts potentially or ultimately reduce water transparency and, therefore, the quality and quantity of the irradiance reaching the seagrass canopy (Duarte et al., 2004; Orth et al., 2006). Epiphytic and planktonic algal accumulations from excess anthropogenic nutrients, increased sediment run-off, and resuspension of bottom sediments are primary causes of reduced underwater irradiance in coastal areas (Orth and Moore, 1983; Cambridge et al., 1986; Onuf, 1994). Ocean warming is also regarded as one of the most severe factors of global climate change, expected to cause, under the A1FI (extreme greenhouse gas emissions) scenario, the rise of the ocean surface temperatures between 2.4 and 6.4°C by the year 2100 (IPCC, 2007), along with an increased amplitude and duration of heat waves (abnormally warm seawater episodes reaching up to ~5°C above the mean temperatures recorded) (Marbà and Duarte, 2010; IPCC, 2014; Galli et al., 2017). These changes are predicted to have serious repercussions in the Mediterranean Sea given its confined nature, which makes it more susceptible to temperature increases, with the warming occurring at significantly higher rates compared to open oceans (Diffenbaugh et al., 2007; Vargas-Yáñez et al., 2008; Calvo et al., 2011).

Generally, light and nutrients comprise the energy source and the matter needed for the growth of seagrasses, while the temperature regulates biochemical processes involved in photosynthesis and respiration, thus, predominantly controlling the annual and seasonal production patterns of seagrasses (Lee and Dunton, 1996; Zupo et al., 1997; Lee et al., 2005, 2007). Water temperature and irradiance are usually correlated and display similar seasonal trends, making it difficult to isolate both environmental parameters in relation to seagrass growth and production (Kaldy and Dunton, 2000; Kaldy, 2006). Light requirements for seagrasses are unusually high, being approximately 10–37% of the surface irradiance, compared to the 0.1–1% needed for most of the other marine macrophytes, which is partially attributed to the inefficient carbon concentrating mechanisms for photosynthesis (Invers et al., 2001; Larkum et al., 2006; Zimmerman, 2006). In view of this, seagrasses are highly vulnerable to the deterioration of water clarity, which is evidenced through the reported large-scale losses of meadows worldwide (Dennison et al., 1993; Onuf, 1994; Short and Wyllie-Echeverria, 1996; Erftemeijer et al., 2006). Hence, understanding the light thresholds for seagrass survival is fundamental for the effective management of these valuable habitats (York et al., 2013). Elevated temperatures entail a grand risk of local extinction for cold-adapted plants, such as Mediterranean seagrasses, as they have visibly manifested physiological symptoms of heat stress and of reduced fitness (Beca-Carretero et al., 2018; Marín-Guirao et al., 2018). The consequences of heat waves can be notably damaging on seagrass meadows, promoting shoot mortality and population decline when critical temperature thresholds are surpassed (Díaz-Almela et al., 2009; Marbà and Duarte, 2010; Jordà et al., 2012). In addition, thermal stress can induce the acceleration of the respiration over the photosynthesis rates (Collier and Waycott, 2014; Marín-Guirao et al., 2018), and affect important life-history events, like reproduction, through an increased flowering intensity (Ruiz et al., 2018; Marín-Guirao et al., 2019).

Several studies have been conducted to determine the effects of light reduction on P. oceanica, comparing its response along bathymetric/spatial gradients (Alcoverro et al., 2001; Ruiz and Romero, 2003; Dattolo et al., 2014) and experimentally through the modification of the light environment with shading screens (Ruiz and Romero, 2001; Mazzuca et al., 2009; Serrano et al., 2011; Gacia et al., 2012). Investigations on the impacts of the sea warming, considering the climate change scenarios, have focused on its isolated effects (Marbà and Duarte, 2010; Guerrero-Meseguer et al., 2017; Hernán et al., 2017; Ruiz et al., 2018; Traboni et al., 2018) and the interaction with other stressors (Hendriks et al., 2017; Ontoria et al., 2019; Agawin et al., 2021). However, research on the combined effects of ocean warming and water turbidity in P. oceanica meadows remains relatively scarce, and understanding them is crucial for the future management of these coastal ecosystems. Moreover, no studies have been done, so far, on the effects of these factors on the microorganisms associated with P. oceanica, such as the nitrogen (N₂) fixers.

Biological N₂ fixation, defined as the enzymatic reduction of atmospheric N₂ to ammonium equivalents, takes part in significantly sustaining the high productivity of P. oceanica in the oligotrophic Mediterranean Sea (Garcias-Bonet et al., 2019). Seagrasses harbor diverse communities of epi- and endophytic bacteria associated with their leaves, roots, and rhizomes, that can enhance plant growth through an increased nutrient availability, for instance, via N₂ fixation or by mineralizing organic compounds (Uku et al., 2007; Cole and McGlathery, 2012; Garcias-Bonet et al., 2016). In particular, it has been reported that the N₂ fixation processes associated with the phyllosphere of P. oceanica could potentially supply the total nitrogen demand of the plant (Agawin et al., 2016, 2017). The enzyme complex that catalyzes the biological N₂ fixation process is the nitrogenase, which is composed of two proteins: conventionally, the iron protein or nitrogenase reductase, and the molybdenum iron protein or nitrogenase (Hamisi, 2010). The nifH gene encodes the iron protein, while the nifDK genes encode the molybdenum iron protein (Rubio and Ludden, 2002). Sequencing the nif genes, with nifH being the most sequenced and the marker gene of choice, has allowed the studying of the phylogeny, diversity, and abundance of N₂-fixing microorganisms (Gaby and Buckley, 2012). A significant diversity has been identified among marine diazotrophs, with filamentous organisms primarily including the Trichodesmium sp. and the diatom symbiont Richelia intracellularis, and highly diverse unicellular diazotrophs that comprise Cyanobacteria, Proteobacteria, and Archaea (Moal et al., 2011). At present, three groups of unicellular N₂-fixing cyanobacteria (UCYN) have been described: UCYN-A, B, and C (Zehr et al., 2001; Foster et al., 2007). Groups B and C are nanoplanktonic cells (2–10 μm), which are closely related to the cultivated strains Crocosphaera watsonii and Cyanothece sp., respectively (Church et al., 2005; Foster et al., 2007); whereas, members of Group A are of picoplanktonic size (0.7–1.5 μm) and uncultivated up until now (Biegala and Raimbault, 2008; Goebel et al., 2008).

Nitrogenase activity is influenced by a combination of environmental factors that differ depending on the geographic region and the diazotroph community composition (Mahaffey et al., 2005). Several studies have suggested a strong temperature dependence of the N₂ fixation process at the enzymatic level, and a positive correlation to irradiance (Welsh, 2000; Brauer et al., 2013; Agawin et al., 2017; Garcias-Bonet et al., 2019). Furthermore, as the nitrogenase proteins require iron (Fe), the availability of both phosphorus (P) and Fe are factors that could limit or co-limit the N₂ fixation process in some areas of the oceans (Sañudo-Wilhelmy et al., 2001; Karl et al., 2002; Mills et al., 2004). Primary aquatic producers usually contain external alkaline phosphates; enzymes capable of hydrolyzing organic phosphorus compounds (monoester phosphates), which liberate inorganic phosphorus and increase the availability of this nutrient for growth (Kuenzler and Perras, 1965; Martínez-Crego et al., 2006). Thus, the measurement of alkaline phosphatase activity has been employed as an indicator of the limitation of the phosphorus and the deficiency in algae and seagrasses (Pérez and Romero, 1993; Invers et al., 1995; Steinhart et al., 2002; Fernández-Juárez et al., 2019).

This study assesses the response of P. oceanica and its N₂-fixing community to different combinations of temperature and light levels, in terms of primary production and respiration rates, chlorophyll content, alkaline phosphatase activity, oxidative stress indicators, and N₂ fixation activities of the diazotrophs associated with different plant tissues.

The results obtained for primary productivity and chlorophyll content of Posidonia oceanica suggest an enhancement in these values under saturating light conditions for the plants during winter, which is in line with previous studies that emphasize light availability as the primary factor influencing the photosynthetic performance of this Mediterranean species (Pergent-Martini et al., 1994; Alcoverro et al., 1995). Higher average pH (i.e., more alkaline conditions), occurred in the aquaria under high light treatments (Figure 2B), possibly reflecting the greater buffering effect provided by the plants through increased photosynthesis (Hendriks et al., 2013). Generally, seagrass meadows can induce diurnal variations in the seawater carbon chemistry in relation to their productivity, through the uptake of CO₂ during photosynthesis in the day and the release of CO₂ with respiration at night (Chou et al., 2018; Howard et al., 2018). Hence, the more metabolically intense an ecosystem is, the greater its capacity to affect the seawater pH and its alkalinity (Duarte et al., 2013; Hendriks et al., 2013). In addition, greater chlorophyll production has been denoted as a photo-acclimation response of P. oceanica meadows thriving under high light conditions, which implies an increase in the number of reaction centers, and, consequently, in the capacity of photon absorption and of electron flow rate along the transport chain (Frost-Christensen and Sand-Jensen, 1992; Ruban, 2009; Dattolo et al., 2014). Higher chlorophyll content in P. oceanica leaves is generally related to greater photosynthetic rates (Alcoverro et al., 2001), as it was observed in this study. Although the elevated temperature conditions did not significantly affect the chlorophyll concentrations in this study, a past research done by Pirc (1986) reported that P. oceanica can also positively respond to the increasing temperatures with higher photosynthetic rates, as well as chlorophyll production in the winter and spring months. Similarly, Agawin et al. (2021) reported significantly greater chlorophyll a and b values for P. oceanica when incubated at elevated temperature (17°C) and CO₂ conditions (1200 μmol CO₂ mol⁻¹) in winter.

During the incubations, elevated temperatures did not seem to negatively alter the photosynthetic response, nor promote higher respiration rates in P. oceanica, as opposed to previous findings that demonstrated the disruptive effect of this factor on the productivity of this seagrass species (Collier and Waycott, 2014; Marín-Guirao et al., 2018). This could be partially attributed to the fact that, although the elevated temperature treatment applied in the aquaria was 1°C above the optimum conditions recorded for this seagrass (17–20°C), it stays within its temperature comfort range (13–24°C) (Boudouresque and Meinesz, 1982). However, significantly higher leaf net photosynthetic rates were still exhibited by plants subjected to 15.5°C and high light treatment. On the other hand, in Agawin et al. (2021), positive responses to higher temperature were observed in the photosynthetic activity of P. oceanica, even though leaf respiration rates did not increase, suggesting that ocean warming scenarios may not necessarily have adverse effects on the carbon balance of the plants during winter. This agrees with the higher NPP rates estimated from the phyllosphere of P. oceanica at an elevated temperature and high light conditions during the experiment. Additionally, it should be considered that since the measurements were only made at the end of the experiment, it is possible that the biological parameters might have changed over the course of the incubations, and, thus, the results presented may not entirely address the response of P. oceanica to an abrupt increase of temperature.

The results of this study showed a positive response in the productivity of P. oceanica to light availability in winter. This suggests that anthropogenic activities that cause prolonged periods of a reduced surface irradiance will possibly have more destructive impacts in these ecosystems during winter, rather than an increasing occurrence of heat waves in the Mediterranean (Marbà and Duarte, 2010; Galli et al., 2017). Similarly, Hendriks et al. (2017) reported that low light availability had a negative effect on the photosynthetic performance of P. oceanica under the short-term experimental conditions for summer, while the temperature negatively affected the plants growth. Champenois and Borges (2018) also highlighted the strong association between the GPP of P. oceanica and the interannual variations of the light availability over a decade in the Bay of Revellata, France, as well as the positive correlation to temperature, given that the temperatures recorded had stayed within the comfort range of the seagrass species. Nevertheless, Serrano et al. (2011) demonstrated through in situ shading experiments in Portlligat Bay, Spain, that shallow P. oceanica meadows are more vulnerable to severe light limitation during spring-summer compared with autumn-winter, since it coincides with the plants' favorable growth season when they accumulate reserves for overwintering. The current research was conducted in winter, when the plants could be susceptible to the interaction between an increased seawater temperature and reduced light conditions, as they exhibit lower-upper thermal tolerance limits (Agawin et al., 2021), and the quantity of light they receive for this season is very limited in comparison with spring and summer. Moreover, the majority of experimental evidence on the impacts of elevated temperature on Mediterranean seagrasses have been carried out during the warmer months, hence, this study provides valuable information regarding the potential response of P. oceanica in winter.

The biochemical responses of P. oceanica showed that young leaves subjected to the ambient temperature and high light treatment have exhibited significantly higher ROS production, while top leaves demonstrated greater polyphenols content at an elevated temperature. These observations coincide with studies that reported how factors, such as high irradiance and temperature, may promote the production of ROS in photosynthetic organisms, that in excessive quantities leads to oxidative stress (Choo et al., 2004; Adams et al., 2006; Costa et al., 2015). In this study, the young leaves of P. oceanica that were still in the process of development might have been more vulnerable to the saturating light conditions, which possibly induced the elevated ROS concentrations detected. However, it should also be noted that the high primary production rates unavoidably prompt the production of ROS (Hajiboland, 2014), since these free oxygen radicals are liberated when the photolysis of water molecules by the photosystem II (PSII) occurs during photosynthesis (Lesser, 2006). This corresponds with the significantly higher NPP rates reported for the young leaves at the ambient temperature and high light treatment, hence, the increased amounts of ROS perceived in these tissues could also be related to their high photosynthetic activity. For the top leaves, the higher phenolic compounds measured could be a response of the plants to the elevated temperatures they were exposed to, considering they were not optimum for their functioning and seagrasses, especially for P. oceanica which are sensitive to the quality of environmental conditions (Orth et al., 2006). It may also reflect a thermal stress response of the plants, given the abrupt temperature increase applied in these treatments. Phenolic compounds demonstrate several biological functions that include antioxidant activity, and generally plants activate antioxidant mechanisms to detoxify the generated ROS and to avoid oxidative stress (Cheynier et al., 2013; Costa et al., 2015). The higher mean values of ROS calculated for top leaves under an elevated temperature compared to ambient treatments may support this notion, although the differences were not significant (p < 0.05).

The phosphatase activity in P. oceanica achieved its maximum values at an elevated temperature and low light conditions, which is consistent with the patterns described by Invers et al. (1995), who demonstrated that the increasing temperatures can positively affect APA rates in this seagrass until a certain threshold (24°C). The high activity of this enzyme under an elevated temperature and a low light also matches the significantly enhanced N₂ fixation rates exhibited by this treatment, considering that the energy, in the form of adenosine triphosphate (ATP), to fuel N₂ fixation is dependent on the presence of inorganic phosphorus; therefore, the demand for this nutrient is theoretically induced when the cells are fixing N₂ (Romano et al., 2017; Fernández-Juárez et al., 2019). In contrast, the APA values were on average the lowest under equal temperature and saturating light, regardless of the more favorable pH values for the phosphatases in this treatment, as they were higher compared to the low light treatments. This result could be attributed to the fact that at the final stage of the experiment, the mean phosphate concentration for this treatment was several orders of magnitude higher than the rest (p > 0.05; Table 1), given that APA in seagrasses decreases under elevated phosphate content and, vice versa, increases with phosphorus limitation (Invers et al., 1995; Martínez-Crego et al., 2006; Agawin et al., 2021). As for the significantly lower APA rates demonstrated by the rhizomes, this may suggest that the metabolic activity occurring in this part of the plants is relatively low compared to the others, consequently, its inorganic phosphorus demand is low as well. Generally, the phosphatase activity tends to be greater in the leaves, partially due to the contribution of the epiphytes (Invers et al., 1995).

The estimated N₂ fixation rates in the present study are relatively similar to those previously obtained by other works in P. oceanica meadows during winter (Agawin et al., 2017, 2019). The diazotrophic activity, being significantly higher at an elevated temperature and low light conditions, could be attributed to the strong temperature dependency of the nitrogenase enzyme (Brauer et al., 2013; Agawin et al., 2017; Garcias-Bonet et al., 2019). Further, the low nitrate concentrations exhibited in this treatment at the final stage of the incubations (Table 1) may suggest the existence of a dissolved inorganic nitrogen limitation, which possibly induces N₂ fixing conditions in these aquaria. Nonetheless, the N₂ fixation was considerably lower under an elevated temperature and saturating light conditions, and this could be related to the significantly high GPP values measured for the leaves and phyllosphere of P. oceanica in these tanks. It has been widely documented that the N₂ fixation is an oxygen-sensitive process, since molecular oxygen (O₂) is capable of inactivating the nitrogenase, then causing irreversible damage to the protein structure, as well as inhibiting the synthesis of the enzyme in many diazotrophs (Berman-Frank et al., 2003; Schoffman et al., 2016). Thus, an increased O₂ evolution with an increased photosynthesis may negatively affect the N₂ fixation activities associated with P. oceanica in treatments with increased GPP. The diazotrophic activity showed variability among plant parts, with the roots exhibiting higher average values, particularly the sterilized ones containing the root endophytes, in comparison to the leaves. These findings are consistent with the results of Lehnen et al. (2016) for P. oceanica and Hamisi et al. (2009) in tropical seagrass species. According to the latter authors, higher activities in the rhizosphere could be associated with a high occurrence of heterotrophic diazotrophs aside from the autotrophic bacteria in the phyllosphere. This pattern is also in agreement with what was reported by Agawin et al. (2019), who determined the presence of seasonality in the N₂ fixation process related to P. oceanica meadows along the Mallorcan coast, with generally higher activities associated with the roots during winter.

The RT-qPCR analyses revealed the presence of the three groups of unicellular diazotrophic cyanobacteria in the phyllosphere of P. oceanica, with UCYN-B and -C displaying notably higher transcription levels of the nifH gene in comparison to UCYN-A. Past molecular analyses carried out by Agawin et al. (2017) indicated the presence of members of the UCYN-B: Crocosphaera, and UCYN-C, like Cyanothece, in addition to other genera of the phyla Cyanobacteria, Proteobacteria, Firmicutes, Bacteroidetes, and Archaea in the phyllosphere of P. oceanica. The UCYN-A (Candidatus Atelocyanobacterium thalassa) is regarded as one of the most abundant and widespread N₂-fixing groups in the ocean, proven to live attached or in symbiosis with larger single-celled prymnesiophytes, given that they lack important biosynthetic pathways genes, including oxygenic photosynthesis and carbon fixation (Zehr et al., 2008; Thompson et al., 2012). Considering its association with larger algae and the frequent symbioses between cyanobacteria and multicellular plants as evidenced in terrestrial habitats (Thompson et al., 2012), the presence of this N₂-fixing group in P. oceanica is also plausible. The identified diazotrophic communities showcased varied responses to the treatments, with UCYN-B demonstrating the maximum average of nifH expression at an elevated temperature and a limited light, while UCYN-C had significantly higher values compared to the rest under an ambient temperature and low light conditions, although the highest means for this latter group alone were both under high light treatments. The UCYN-B group, having enhanced activities at elevated temperatures, is consistent based on the literature (Brauer et al., 2013; Agawin et al., 2017), with the generally positive correlation recorded between the temperature and the N₂ fixation. Moreover, past studies have evidenced the thermophilic habit of Crocosphaera (UCYN-B), with warmer sea surface temperatures (26–29 °C) primarily determining its distribution across the oceans (Church et al., 2008; Moisander et al., 2010). The results obtained for UCYN-C might be due to the different light and temperature requirements its representatives possess; consequently, the species have optimum N₂ fixation rates under distinct conditions. On the other hand, although it has been denoted that UCYN-A exhibits a broad temperature range, more associated with cooler waters and a lower temperature optimum than the remaining cyanobacterial groups (Moisander et al., 2010; Cabello et al., 2020), the results obtained from the experiment do not reflect a clear pattern. However, these results should be interpreted with caution, as they may not adequately represent the in situ microbial composition associated with P. oceanica given that the incubation in aquaria might have affected it to a certain extent. Taking into consideration how the different diazotrophic species are adapted to grow and to function under differing conditions, further investigation is required to achieve a better understanding of their potential response to the interaction between the climate change factors and the other stressors.

The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

MG-M and NA designed the experiments. MG-M, VF-J, and JR-C conducted all experiments. All authors led the writing of the manuscript, reviewed, and supervised by the head of the laboratory, NA.

This research was part of the project CTM2016-75457-P, funded by the MINECO/AEI/10.13039/501100011033 and the ERDF.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.