Heat Stress Stimulates Nitric Oxide Production in Symbiodinium microadriaticum: A Possible Linkage between Nitric Oxide and the Coral Bleaching Phenomenon

Abstract

Nitric oxide (NO) is a gas displaying multiple physiological functions in plants, animals and bacteria. The enzymes nitrate reductase and NO synthase have been suggested to be involved in the production of NO in plants and algae, but the implication of those enzymes in NO production under physiological conditions remains obscure. Symbiodinium microadriaticum, commonly referred to as zooxanthellae, is a marine microalga commonly found in symbiotic association with a cnidarian host including reef-building corals. Here we demonstrate NO production in zooxanthellae upon supplementation of either sodium nitrite or l-arginine as a substrate. The nitrite-dependent NO production was detected electrochemically and confirmed by the application of 2-(4-carboxyphenyl)-4,4,5,5-tetramethylimidazoline-1-oxyl-3-oxide (cPTIO), a specific NO scavenger. Cells stained with the diaminofluorescein, DAF-2 DA, an NO fluorescent probe, showed an increase in fluorescence intensity upon supplementation of both sodium nitrite and l-arginine. Microscopic observations of DAF-stained cells verified that NO was produced inside the cells. NO production in S. microadriaticum was found to increase upon exposure of cells to an acute heat stress which also caused a decline in the photosynthetic efficiency of PSII (F_v/F_m). This study provides substantial evidence to confirm that zooxanthellae can synthesize NO even when they are not in a symbiotic association with a coral host. The increase in NO production at high temperatures suggests that heat stress stimulates the microalgal NO production in a temperature-dependent manner. The implications of these findings are discussed in the light of the coral bleaching phenomenon which is associated with elevated sea surface temperature due to global warming.

Introduction

Nitric oxide (NO) is a membrane-permeable gas molecule which is involved in important physiological and developmental functions as well as in defense responses of living organisms. NO is also involved in the chemosensory responses of aquatic invertebrates (Colasanti et al. 1995, Elphick et al. 1995) and in processes such as germination, root elongation, senescence, stomatal closure, disease resistance and responses to both biotic and abiotic stresses in plants (Gould et al. 2003, Zeidler et al. 2004, Modolo et al. 2006, Mur et al. 2006, Valderrama et al. 2007). Recently, some species of microalgae were also shown to produce NO (Mallick et al. 1999, Sakihama et al. 2002, Tischner et al. 2004, Estevez and Puntarulo 2005, Kim et al. 2006, Zheng-Bin et al. 2006). Although the role of NO in microalgae remains unclear, it has been suggested to act as a message factor in the growth status of microalgae because its production is influenced by environmental stimuli such as light, temperature, salinity and trace elements (Zheng-Bin et al. 2006).

In animals, the synthesis of NO is primarily accomplished by a family of enzymes known as NO synthase (NOS) which catalyzes the conversion of l-arginine, NADPH and O₂ to NO, l-citrulline and NADP⁺ (Alderton et al. 2001). Although a NOS-like activity has also been found in plants (Wendehenne et al. 2001) and possibly in microalgae (Trapido-Rosenthal et al. 2001), the identified genes and proteins responsible for this arginine-dependent production of NO in plants present no sequence homologies to the NOS family of enzymes in animals (Crawford 2006, Yamasaki and Cohen 2006). Alternatively, plants and some species of microalgae can produce NO enzymatically through the activity of the nitrate reductase (NR) enzyme (Yamasaki et al. 1999, Rockel et al. 2002, Meyer et al. 2005), through the activity of a variety of mitochondrial enzymes (Tischner et al. 2004, Planchet et al. 2005) or non-enzymatically by the reduction of apoplastic nitrite (Bethke et al. 2004).

Reef-building corals are characterized by an obligate symbiotic relationship between a cnidarian host and a photosynthetic dinoflagellate of the genus Symbiodinium (known as zooxanthellae). Our understanding of the interaction between the host and the symbiont is limited and, although little is known about the physiology of zooxanthellae, recent studies have shown that the photosynthetic apparatus of zooxanthellae is sensitive to a range of environmental conditions: temperature, salinity and light (Ralph et al. 1999, Ralph et al. 2001, Takahashi et al. 2004, Nakamura et al. 2005). Indeed, increases in sea surface temperatures associated with exposure to high light have caused bleaching events over the past two decades (Hoegh-Guldberg 1999). This bleaching phenomenon is characterized by the mass expulsion of symbiotic dinoflagellates from the host tissue or the loss of photosynthetic pigments within individual zooxanthellae that remain (Glynn 1991). Although the exact mechanism involved in the coral bleaching phenomenon remains to be clearly determined, NO has recently emerged as a potential signaling molecule triggering the expulsion of the symbionts from the host cells (Perez and Weis 2006). Under stressful conditions leading to coral bleaching, whether it is the coral host or the algal symbiont which synthesizes NO and actually initiates the algal expulsion from the coral host remains under debate.

In a study performed with zooxanthellae freshly isolated from anemones and with pure cultures of zooxanthellae (Symbiodinium bermudense), a NOS-like activity was reported only for the fraction of zooxanthellae isolated from their cnidarian host and not for the microalgae in culture (Trapido-Rosenthal et al. 2001). These results led the authors to believe that some of the coral host factors were necessary for the algal symbionts to produce NO (Trapido-Rosenthal et al. 2001). These results were, however, contested by others who presumed that the NOS-like activity detected in the microalgae was due to a contamination of the zooxanthellae fraction with some of the host tissues (Perez and Weis 2006). Using an NO-fluorescent probe, Perez and Weis (2006) found the production of NO in symbiotic sea anemone but not in aposymbiotic sea anemones upon exposure to thermal stress. In this same study, NO production could not be detected in zooxanthellae isolated from symbiotic sea anemones nor in cultured zooxanthellae (Perez and Weis 2006). These results led the authors to conclude that the coral host itself and not the symbionts was responsible for the production of NO. Interestingly, recent studies have shown that a variety of microalgal species possess the ability to synthesize NO. It is thus plausible to believe that zooxanthellae, even when not associated with their coral host, also possess such a capacity. The aim of this study was thus to assess the capability of zooxanthellae to synthesize NO when grown under laboratory conditions and to gain some insights concerning the possible pathway of NO production in these microalgae. Additionally, in the context of global warming, the influence of an acute heat stress was assessed on the NO production potential of zooxanthellae.

Results

Electrochemical detection of nitric oxide upon sodium nitrite supplementation

In order to assess the capacity of zooxanthellae to produce NO, a Clark-type electrode was used to detect NO in a suspension of zooxanthellae. Sodium nitrite (NaNO₂) was used as a substrate and supplemented to initiate the reaction (Fig. 1A). Upon sodium nitrite supplementation, the baseline signal detected by the electrode rapidly increased to reach almost 40 pA. The lower trace on the graph shows the response obtained upon supplementation of sodium nitrite to the culture medium containing no zooxanthellae (Fig. 1A). To confirm that the signal detected upon supplementation of sodium nitrite could truly be ascribed to the production of NO, the specific NO scavenger cPTIO [2-(4-carboxyphenyl)-4,4,5,5-tetramethylimidazoline-1-oxyl-3-oxide] was added to the suspension and caused the signal to return quickly to nearly the baseline (Fig. 1B). These results suggest that upon sodium nitrite supplementation, NO was probably produced inside the zooxanthellae and then released inside the culture medium where it was detected by the electrode. The amount of NO released in the culture medium by different species of microalgae has been shown to vary according to the growth status of the cells and to be influenced by multiple environmental factors (Zheng-Bin et al. 2006). In the present study, the amount of NO released into the culture medium upon supplementation of 3 mM of sodium nitrite reached near 0.23 pmol NO per cell (Fig. 1C). Supplementation of different concentrations of sodium nitrite to a suspension of zooxanthellae showed a nitrite concentration-dependent production of NO (Fig. 1C).

Fig. 1

Electrochemical detection of NO production in the culture medium of S. microadriaticum. (A) Clark-type electrode response to sodium nitrite supplementation (3 mM) at 180 s. The upper trace represents the average of six samples (±SE) containing a suspension of S. microadriaticum and the lower trace represents the average of six samples (±SE) containing F/2-Si culture medium with no cells. (B) Typical trace obtained upon supplementation with sodium nitrite (3 mM) and the subsequent addition of the specific NO scavenger cPTIO (1 mM) at 90 s. (C) Nitrite concentration-dependent production of NO. Each bar represents the average of three samples ± SE. Each bar is the average of the highest values (peaks) obtained upon supplementation with 1.5, 3 and 6 mM sodium nitrite. The symbols represent the concentrations of NO standardized by the cell number. (D) Clark-type electrode response to l-arginine supplementation (10 mM) at 180 s. The upper trace represents the average of 16 samples (±SE) containing a suspension of S. microadriaticum, and the lower trace represents the average of eight samples (±SE) containing F/2-Si culture medium with no cells.

In order to determine if the production of NO was solely dependent on the supplementation of sodium nitrite, another trial was performed using l-arginine as a substrate because a NOS-like activity has recently been reported in microalgae (Trapido-Rosenthal et al. 2001). Upon supplementation of l-arginine (10 mM), the signal detected by the electrode slightly increased to reach almost 6 pA. The change in the signal detected was, however, very similar to that of the culture medium without cells (lower trace; Fig. 1D) and too transient to allow the application of a specific NO scavenger (Fig. 1D). Consequently, using the NO electrode, it was impossible to confirm that NO was produced in the presence of l-arginine as a substrate.

Fluorometric detection of nitric oxide upon supplementation of sodium nitrite and l-arginine

As an independent method of NO detection and to ensure that the production of NO detected with the electrode could truly be ascribed to zooxanthellae, an NO-specific fluorescent probe was used to detect the production of NO inside the microalgal cells. For this method, the increase in fluorescence intensity observed in the diaminofluorescein (DAF)-loaded cells is indicative of NO production (Arita et al. 2007). Results obtained during the time-course experiment showed a gradual increase in fluorescence intensity detected at 515 nm for the suspension of zooxanthellae supplemented with sodium nitrite (Fig. 2A). In contrast, DAF-loaded cells not supplemented with sodium nitrite showed no increase in fluorescence intensity (lower trace, Fig. 2A), thereby confirming that zooxanthellae produce NO in a nitrite-dependent manner. In a similar type of experiment, DAF-loaded cells supplemented with l-arginine (10 mM) showed a gradual increase in fluorescence intensity detected at 515 nm (Fig. 2B). When previously treated with l-N^G-nitroarginine (l-NNA), an NOS inhibitor, the fluorescence intensity at 515 nm remained lower compared with that with the l-arginine treatment. Finally, when l-arginine was not supplemented to the DAF-loaded cells, the fluorescence intensity remained low, meaning that NO was not produced if the substrate l-arginine was not supplemented (lower trace, Fig. 2B). These results show that in addition to the nitrite-dependent production of NO found in S. microadriaticum, production of NO dependent on l-arginine supplementation can also occur inside the cells of S. microadriaticum. Interestingly, the fluorescence intensity values reached upon l-arginine supplementation were 10 times lower then those reached upon supplementation with sodium nitrite, thereby suggesting a less important production of NO in the presence of the substrate l-arginine.

Fig. 2

Fluorometric detection of NO production using the NO fluorescence probe DAF-2 DA. (A) NO production upon sodium nitrite supplementation. The upper trace shows the time-course of NO production upon sodium nitrite (3 mM) supplementation at 180 s. The lower trace shows the absence of NO production when sodium nitrite was not provided. Data are the averages of three samples ±SE. (B) NO production upon l-arginine supplementation. The upper trace shows the time-course of NO production upon l-arginine (10 mM) supplementation at 180 s. The middle trace shows the limited production of NO in the presence of the NOS inhibitor l-NNA and l-arginine (10 mM) supplementation at 180 s. The lower trace shows the absence of NO production when the substrate l-arginine was not supplemented to the samples. Except for the ‘No addition’ treatment, data represent the averages of two independent experiments (n = 7; n = 3 for ‘No addition’ treatment) ±SE.

Microscopic visualization of nitric oxide production inside the cells of S. microadriaticum

Epi-fluorescence microscopic observations (Fig 3A–F) showed that when excited by blue light in the 465–495 nm range [fluorescein isothiocyanate (FITC) filter] zooxanthellae loaded with DAF-2 DA fluoresced in green (Fig. 3F). This is because when DAF-2 DA is loaded into the cells it is converted into DAF-2 by cytosolic esterases. N₂O₃, a compound derived from the oxidation of NO, then reacts with DAF-2 to form DAF-2T, a triazole derivative which fluorescesces in green (Kojima et al. 1998). Since this reaction is considered to be specific (Suzuki et al. 2002), the green fluorescence emitted by the DAF-loaded zooxanthellae can be attributed to the production of NO inside the cells. In contrast, cells not loaded with DAF-2 DA showed no fluorescence when excited with blue light in the 465–495 nm range (Fig. 3C) but fluoresced in red when excited with blue light in the 450–490 nm range (Fig. 3B). This red color was attributable to the chlorophyll-induced autofluorescence of the cells. Zooxanthellae loaded with DAF-2 DA, on the other hand, fluoresced in greenish-yellow using the same excitation wavelengths (450–490 nm; Fig. 3E). This color was attributable to the superposition of the red autofluorescence with the green fluorescence from the DAF probe (Fig. 3E). Results from these microscopic observations using the NO fluorescent probe thus confirm that NO is truly produced inside the cells of S. microadriaticum.

Fig. 3

Microscopic observations of NO production in S. microadriaticum. (A–C) Pictures of S. microadriaticum not loaded with DAF; (D–F) pictures of DAF-loaded cells. Light micrographs (A, D) and fluorescence micrographs (B, C, E, F) when cells were excited with blue light either in the 450–490 nm range to visualize the autofluorescence of the cells (chlorophyll) or in the 465–495 nm range to visualize the DAF fluorescence when NO is produced inside the cells.

Possible pathways involved in the production of nitric oxide

In order to assess the different possible pathways involved in the process of NO production, algal cells were incubated in the absence or in the presence of a series of inhibitors and subsequently supplemented either with sodium nitrite or l-arginine. Among the different inhibitors used, tungstate is a common NR inhibitor (Deng et al. 1989), rotenone is a specific inhibitor of mitochondrial complex I (Espoti, 1998), while l-NNA is a NOS inhibitor (Furfine et al. 1993). The fluorescence intensity detected without any addition was higher than that detected (<0.1) for killed zooxanthellae loaded with DAF-2 DA (not shown). As expected, sodium nitrite supplementation to the cells resulted in a significant increase (P = 0.000) in fluorescence intensity compared with that of the cells not supplemented with sodium nitrite (Fig. 4). When the cells were previously treated with the NR inhibitor tungstate and subsequently supplemented with sodium nitrite, the DAF-loaded cells showed a fluorescence intensity significantly lower (P = 0.000) than that of cells supplemented or not with sodium nitrite (Fig. 4), thereby suggesting the implication of NR in this process. Similarly, cells previously treated with rotenone, the mitochondrial inhibitor, and supplemented with sodium nitrite showed a limited production of NO which was also significantly lower than the fluorescence intensity reached in the presence of sodium nitrite only (P = 0.000). Interestingly, when cells were pre-treated with l-NNA, the NOS inhibitor, and supplemented with sodium nitrite, the production of NO was also limited, and significantly lower than the fluorescence intensity reached in the presence of sodium nitrite only (P = 0.000), thereby suggesting the possible implication of a NOS-like activity in the NO production process. When cells were supplemented with the substrate l-arginine instead of sodium nitrite, a significant increase in fluorescence intensity was observed compared with the no addition treatment (P = 0.001). In the presence of either tungstate or rotenone, the production of NO upon supplementation with l-arginine was not significantly different from that of the cells supplemented with l-arginine only (P > 0.05). In contrast, in the presence of the NOS inhibitor l-NNA, the production of NO was significantly lower compared with that of the cells supplemented with l-arginine only (P = 0.001).

Fig. 4

Effects of inhibitors on NO production in S. microadriaticum. NO was detected with DAF-2 DA upon supplementation with either sodium nitrite (3 mM) or l-arginine (10 mM) as a substrate and when using different inhibitors. Sodium tungstate is a common nitrate reductase inhibitor, rotenone is a common mitochondrial inhibitor blocking the flow of electrons at complex I, and L-NNA is a NOS inhibitor. Data are the average of four samples ± SE. Significant differences (at P < 0.05) between the different treatments are indicated by different letters.

Stimulation of nitric oxide production upon heat stress

To determine if NO could be produced as a function of heat stress without the addition of exogenous substrates (e.g. sodium nitrite, l-arginine), zooxanthellae were exposed to 27°C, i.e. their growth temperature, and to 34°C. The results showed a significant increase in fluorescence intensity at 34°C compared with that of the cells incubated at 27°C (Fig. 5A, P = 0.005), thereby suggesting a stimulation of NO production upon heat stress. To make sure that the different fluorescence intensities observed between 27 and 34°C were not due to an unequal loading of the DAF-2 DA into the cells, the NO donor 3-[2-hydroxy-1-(1-methylethyl)-2-nitrosohydrazino]-1-propanamine (NOC5; 0.5 mM) was added exogenously to the DAF-loaded cells at 27°C and left to react for 13 min (Fig. 5B). The fluorescence intensity detected in the presence of the NO donor at 27°C was significantly higher compared with that of the cells not supplemented with the NO donor, thereby proving that the dye had efficiently penetrated inside the cells at 27°C (Fig. 5B). To verify further the relationship between the production of NO and the increase in temperature, zooxanthellae were then exposed to a range of temperatures from 27°C, i.e. their growth temperature, to 41°C. The results showed that the production of NO increased in a linear manner with increasing temperature, with the production of NO being up to 1.5 times more elevated at 41°C than at 27°C (P = 0.000; Fig. 6).

Fig. 5

Effect of heat stress on NO production in S. microadriaticum. (A) Fluorescence intensity is indicative of NO production inside the cells of S. microadriaticum loaded with the NO fluorescent probe after a 2 h exposure to either 27 or 34°C. Data are the averages of four independent experiments (n = 21) ± SE. The asterisk indicates a significant difference at P = 0.005. (B) Fluorescence intensity indicative of NO production inside the cells of S. microadriaticum loaded with the NO fluorescent probe at 27°C in the absence and in the presence of the NO donor NOC5 (final concentration 0.5 mM) for 13 min (n = 3 ± SE). The asterisk indicates a significant difference at P = 0.044.

Fig. 6

Temperature dependence of NO production in S. microadriaticum. The fluorescence intensity which is indicative of NO production inside the cells of S. microadriaticum was measured after exposure of the cells to either 27, 31.5, 34, 37.5 or 41°C for 2 h. Data are the average of five samples ± SE.

In addition to the increased production of NO, the heat stress clearly affected the physiological state of the zooxanthellae, as shown by the dramatic decline in F_v/F_m, a photosynthetic parameter for photochemical efficiency of PSII (van Kooten and Snell, 1990), (P = 0.000; Fig. 7A). The percentage of cell loss also increased, although not significantly (P = 0.107), when the cells were exposed to temperatures >27°C (Fig. 7B). The cell division process was also affected as cells exposed to 27 and 30.5°C maintained a mitotic index (MI) of ∼17% while the MI increased to near 20% when the cells were exposed to 34°C (P = 0.071; Fig. 7B). Temperatures above 34°C caused a decrease in the MI, which reached ∼16 and 13% at 37.5 and 41°C, respectively (Fig. 7B).

Fig. 7

Effects of heat stress on the physiological status of S. microadriaticum. (A) Photosynthetic efficiency (F_v/F_m) measured after cells were exposed to either 27, 31.5, 34, 37.5 or 41°C for 2 h. Data are the average of three samples ± SE. (B) Percentage cell loss and mitotic index during the thermal shock treatment. Each point is the average of five samples ± SE. Significant differences (at P < 0.05) between the different treatments are indicated by different letters.

Discussion

In plants, the production of NO has been shown to be involved in the regulation of cell metabolism, gene expression and plant–pathogen interactions (Zeidler et al. 2004). Compared with the field of plant science, research on NO production by phytoplankton is still in its infancy, and reports of NO production by unicellular microalgae remain scarce. Recently, however, NO was detected in the culture medium of diverse marine microalgae (Kim et al. 2006, Zheng-Bin et al. 2006), released in the gas form from chlorophytes (Mallick et al. 1999, Tischner et al. 2004) and detected inside the cells of bacillariophytes (Vardi et al. 2006), chlorophytes (Sakihama et al. 2002) and raphidophytes (Kim et al. 2006). The production of NO by microalgae was additionally shown to be influenced by an array of environmental factors such as temperature, salinity and light (Zheng-Bin et al. 2006). These interesting findings suggest that, similarly to plants, NO could play the role of a messenger upon microalgae exposure to stress conditions. Whether or not the capacity to synthesize NO is a common feature of all microalgal species, however, remains to be determined. In the present study, S. microadriaticum grown under laboratory conditions was shown to produce NO. This finding is of significant interest because this species of microalgae is commonly found in symbiotic association with a coral host and because NO has recently been suggested to act as a signaling molecule in the coral bleaching phenomenon (Perez and Weis 2006).

In the present study, NO was not only detected in the culture medium of S. microadriaticum using an NO electrode but the combined use of an NO fluorescent probe, fluorescence spectrophotometry and microscopy clearly revealed that NO was produced inside these microalgal cells. Although the l-arginine-dependent production of NO could not be confirmed using the electrode, the production of NO inside the microalgal cells upon supplementation of both sodium nitrite and l-arginine as substrates was confirmed using the NO fluorescent probe. The fact that NO could not be detected using the electrode in the presence of l-arginine may possibly be explained by the low yield of NO produced upon supplementation of this specific substrate. The concentrations of NO found upon supplementation of nitrite were many orders of magnitude greater than what has been reported in the literature for cultured microalgae not supplemented with any substrates (Estevez and Puntarulo 2005, Zheng-Bin et al. 2006). The high concentrations of NO detected in our case can thus be explained by high nitrite concentrations used to stimulate the production of NO.

In addition to the limited number of studies existing concerning the production of NO by microalgae, much more controversy exists concerning the actual pathway of NO production. Some studies performed with microalgae have reported that NO was exclusively produced in a nitrite-dependent manner by NR (Sakihama et al. 2002) or independently of NR (Tischner et al. 2004), while other studies demonstrated rather an l-arginine-dependent production of NO, thereby suggesting the involvement of a NOS-like activity (Kim et al. 2006). Under normal growth conditions, the main role of NR is to catalyze the reduction of nitrate to nitrite. Nitrite is subsequently reduced to ammonium through the activity of the nitrite reductase enzyme (NiR) which is present inside the chloroplasts of microalgae (Fig. 8B). The reduction of nitrite requires reducing equivalents produced by the photosynthetic electron transport chain (Crawford 1995). Under stressful conditions leading to the disruption of the photosynthetic process, nitrite can accumulate inside the cytoplasm (Klepper 1976) and be converted to NO by NR (Yamasaki and Sakihama 2000). Whereas under normal growth conditions the production of NO by NR is assumed to be minimal, under stressful conditions leading to the accumulation of nitrite inside the cytoplasm, NR can display a nitrite reductase activity which leads to the production of NO from nitrite (Yamasaki and Sakihama 2000; Fig. 8B). Since the production of NO by NR has already been detected in plants (Desikan et al. 2002) and in some species of microalgae (Sakihama et al. 2002), it is natural to assume that NO synthesis could be a common feature of microalgae.

Fig. 8

Hypothetical diagram for the NO production pathways in S. microadriaticum. (A) Scheme presenting the action sites of the three inhibitors used in the present study and summary of the nitrite- and l-arginine-dependent pathways of NO production. ET, electron transport system of mitochondria; NR, nitrate reductase; NOS, NO synthase; complex I, NADH-ubiquinone oxidoreductase of mitochondria, l-NNA, a NOS inhibitor. (B) Scheme representing the possible routes of NO production in S. microadriaticum. Reduction of nitrate (NO₃⁻) to nitrite (NO₂⁻) by the NR of the microalgae. Upon its subsequent translocation to the chloroplast, NO₂⁻ is reduced to the ammonium anion (NH₄⁺) by nitrite reductase (NiR). Under stressful conditions (e.g. thermal stress) leading to the disruption of the photosynthetic process, NO₂⁻ can accumulate inside the cytoplasm and be converted to NO by NR. An NO₂⁻-dependent production of NO may also occur inside the mitochondria, although the exact site of NO production remains unclear for S. microadriaticum. The existence of an l-arginine-dependent production of NO is also likely in zooxanthellae. Considering that NO is highly membrane permeable and reactive with the superoxide (O₂⁻) radical, the subsequent reaction of these two molecules is likely to yield peroxynitrite (ONOO⁻) which can in turn have many deleterious effects both on the symbionts and on the coral cells. The green arrows indicate the pathway of NO₃⁻ assimilation and reduction under normal growth conditions. The black arrows indicate the possible routes of NO production under stressful conditions, and the gray arrows indicate the possible reaction occurring between NO and O₂⁻ to yield ONOO⁻.

In the present study, the production of NO was shown to occur at 27°C upon supplementation of sodium nitrite as a substrate. In other studies performed using microalgae, the production of NO was also shown to occur in a nitrite-dependent manner in the green algae Chlamydomonas reinhardtii (Mallick et al. 1999, Sakihama et al. 2002) and Chlorella sorokiniana (Tischner et al. 2004). Using mutants lacking NR, Sakihama et al. (2002) showed direct evidence that NR was involved in the nitrite-dependent NO production observed in C. reinhardtii. In contrast, Tischner et al. (2004) found a nitrite-dependent production of NO in NR-mutant cells and in C. sorokiniana in which NR had been inhibited by tungstate. These authors rather found that the nitrite-dependent production of NO occurred at the mitochondrial level during the flow of electrons from complex III to complex IV via the Cyt bc₁ complex. In the present study, the pathway by which NO is produced has not been investigated directly because mutants of S. microadriaticum are still lacking. Instead, the pathway involved in the production of NO was assessed using a combination of different substrates (sodium nitrite or l-arginine) and inhibitors. Interestingly, our results showed an inhibition of the nitrite-dependent production of NO in the presence of both tungstate, the common NR inhibitor, and rotenone, the mitochondrial inhibitor acting at complex I (Fig. 8A). Our results thus suggest that the nitrite-dependent production of NO could involve both the NR and mitochondrial enzymes. The direct implication of NR in the NO synthesis process cannot, however, be confirmed here because this enzyme may rather constitute an important source of nitrite for subsequent production of NO in the mitochondria (see Crawford, 2006, Modolo et al. 2006). Although our results suggest that the mitochondrial electron transport also reduced nitrite to NO, the exact site of NO production inside the mitochondria might be at complex I or at another location downstream of complex I (Tischner et al. 2004). Alternatively, it is also possible that the use of the complex I inhibitor rotenone caused an increase in mitochondrial superoxide (Li et al. 2003, Amirsadeghi et al. 2007). Considering that NO is a highly membrane-permeable molecule and that it is very reactive with superoxide, NO could easily have diffused inside the mitochondria and reacted with superoxide to yield peroxinitrite (Yamasaki and Sakihama 2000). This would in turn have diminished the concentration of NO, which would explain the low fluorescence detected in the presence of rotenone upon sodium nitrite supplementation.

The inhibition of the NO production observed in the presence of the inhibitor l-NNA and upon subsequent supplementation of sodium nitrite is surprising and may suggest either that NOS inhibitors are not reliable and have side effects or that part of the NO production observed was occurring as a result of a NOS-like activity. Indeed, the sustained production of NO upon supplementation of l-arginine and the inhibition of NO synthesis in the presence of the NOS inhibitor l-NNA suggest that S. microadriaticum can synthesize NO through a NOS-like activity (Fig. 8A, B). These results are in accordance with those found by others who showed an increased NOS-like activity in zooxanthellae exposed to bleaching conditions (Trapido-Rosenthal et al. 2005). Considering the fact that these results have been contested by others (Perez and Weis, 2006), our results confirm that zooxanthellae can produce NO even when they are not associated with a coral host. Although there is still no conclusive evidence for the presence of a NOS enzyme in plants and algae, an l-arginine-dependent NO production has been shown in many studies (Yamasaki and Cohen 2006). As an alternative idea to account for this NOS-like activity, there is a possibility that the l-arginine-dependent NO production may be conducted by multiple enzymes rather than by a single enzyme. The NOS inhibitor l-NNA is an analog of the substrate l-arginine and it should also inhibit other enzyme(s) such as arginase, which is involved in arginine metabolism (Arita et al. 2007). Additionally, one may consider that the arginine-dependent NO production pathway is linked to the nitrite-dependent one by some unknown regulatory mechanism(s); an alternative explanation for the effect of l-NNA on the nitrite-dependent NO production (Figs 4, 8A). It appears that the NO production network remains to be unequivocally determined for S. microadriaticum.

In the present study, heat stress caused the inhibition of photosynthesis (as observed by declines in F_v/F_m) and stimulated the production of NO without the supplementation of exogenous substrates. Although the acute thermal stress used in the present study is not representative of natural conditions where a change in temperature occurs more gradually, it was selected to detect if S. microadriaticum actually possess the capacity to synthesize NO under severe environmental pressure. The increase in NO production with temperature could be explained by increases in enzymatic activities such as that of NR, for example (Yamasaki and Sakihama, 2001, Lartigue and Sherman 2002). In another study, exposure of zooxanthellae to a heat stress (33°C) for 24 h led to an increase in the NOS-like activity (Trapido-Rosenthal et al. 2005). In this study, zooxanthellae from bleaching corals had higher NOS-like activities than zooxanthellae isolated from non-bleaching corals, leading the authors to believe that the elevated NOS-like activity detected in zooxanthellae was related to the coral bleaching phenomenon (Trapido-Rosenthal et al. 2005). On the other hand, the major production of NO reported for a symbiotic sea anemone exposed to elevated temperatures (25 to 33°C) was exclusively attributed to the anemones and not to their symbiotic dinoflagellates (Perez and Weis 2006). This can either be explained by the broad thermal tolerance of the species of Symbiodinium studied (Muller-Parker et al. 2007) or, alternatively, by the use of the NO fluorescent probe DAF-FM which does not permeate well into the cells (Kojima et al. 1999). Nevertheless, Perez and Weis (2006) have proposed an attractive model for the role of NO in the cnidarian bleaching phenomenon in which the exposure of a symbiotic system to elevated temperatures would lead to the inhibition of photosynthesis in the zooxanthellae, to the subsequent production of reactive oxygen species and eventually to the production of NO by the coral host. NO would ultimately react with superoxide (produced by the symbionts) and lead to the production of peroxinitrite which potentially induces many deleterious effects on the symbiotic system (Perez and Weis 2006). Considering that zooxanthellae grown under laboratory conditions also possess the ability to synthesize NO and that this production of NO is increased upon heat stress, one could consider the implication of NO in the coral bleaching mechanism (Fig. 8B).

Materials and Methods

Culture conditions, cell count and determination of the mitotic index

Experiments were performed using axenic cultures of S. microadriaticum (CCMP 829) obtained from the Provasoli-Guillard National Center for Culture of Marine Phytoplankton (West Boothbay Harbour, ME, USA). Algae were grown under batch conditions at 27°C in F/2-Si medium (Guillard 1975). Cultures were grown at 50 μmol photons m⁻² s⁻¹ provided by white fluorescent tubes on a 16 : 8 h light : dark photoperiod. Cell number was determined microscopically using a Fuchs-Rosenthal hemocytometer and a Nikon microscope (Model Eclipse 80i, Japan). The MI was calculated as the percentage of doublet cells out of the total cell count in the sample (Wilkerson et al. 1983). Cells were harvested at mid-exponential growth for each experiment performed (5–8 × 10⁴ cells ml⁻¹).

Electrochemical detection of nitric oxide

Electrochemical detection of NO was performed using a Clark-type NO electrode (ISO-NOP, World Precision Instruments, Inc., Sarasota, FL, USA) connected to an ISO-NO Mark II NO meter (World Precision Instruments, Inc.). The analog signal from the NO meter was digitized using a DUO18 two-channel data acquisition system (World Precision Instruments, Inc.) connected to a computer (Yamasaki et al. 1999). A water-jacket chamber, a circulating bath and a magnetic stirrer were also used to ensure constant temperature and mixing. The NO electrode was inserted into the water-jacketed chamber containing a 4 ml capacity glass vial filled with 3 ml of zooxanthellae culture in the log phase. The vial inside the chamber was sealed with a silicon cap which was punctured in the middle to allow the passage of the electrode. The sample was mixed at constant rate using a magnetic bar controlled by the stirrer. The temperature inside the chamber was maintained constant at 27°C with the circulating bath and the water-jacket. Before use, the NO electrode was carefully calibrated using the chemical generation of NO method recommended by the manufacturer. To initiate the production of NO, two different substrates were supplemented, i.e. sodium nitrite (1.5, 3 and 6 mM) and l-arginine (10 mM). The NO scavenger cPTIO (1 mM) was used to ensure that the signal detected could be ascribed to the production of NO.

Fluorometric detection of nitric oxide

The fluorometric detection of NO was performed using the DAF-2 DA compound which is cell permeable (Daichi Pure Chemicals, Co. Ltd, Tokyo, Japan). Upon reaction with N₂O₃, the immediate unstable product of NO oxidation, DAF-2 DA forms a fluorophore (Kojima et al. 1999). The fluorescence emitted by the fluorophore was detected using a fluorescence spectrophotometer (RF-5300PC, Shimadzu, Kyoto, Japan). For all experiments performed (substrates and inhibitors, and temperature experiments) using the fluorescent NO probe, cells were incubated with DAF-2 DA (7.5 μM final concentration) for 1 h in the dark at 27°C. Cells were then washed twice as follow. Samples were centrifuged (1,000×g for 5 min), the supernatant carefully removed, and the pellet, containing the cells, resuspended with fresh F/2-Si culture medium at 27°C. Samples were then kept in the dark until analyzed with the fluorescence spectrophotometer mentioned above. The excitation and detection wavelengths used were 495 and 515 nm, respectively.

Microscopic detection of nitric oxide

DAF-loaded cells and cells not loaded with DAF were observed using an epi-fluorescence Nikon microscope (Model Eclipse 80i, Japan) equipped with a DXM1200c digital camera. Cells were observed using an FITC filter (blue light excitation at 465–495 nm; emission wavelengths 515–555 nm) to observe the fluorescence emitted due to the production of NO. Cells were also observed with a B-2A filter (blue light excitation at 450–490 nm; emission wavelength >520 nm) to visualize the chlorophyll-induced autofluorescence of the cells (Sakihama et al. 2002).

Experimental procedures using the fluorescent NO probe

In order to assess the potential of S. microadriaticum to synthesize NO inside their cells upon supplementation of different substrates, time-course experiments were performed with DAF-loaded cells (loading conditions: 7.5 μM DAF-2 DA for 1 h at 27°C). For the first experiment, three replicate samples were used for each of the two treatments tested: (i) live cells without sodium nitrite addition and (ii) live cells with sodium nitrite addition. Fluorescence measurements were performed for a total of 10 min but, after 3 min of measurement, 3 mM sodium nitrite was injected into the samples for the second treatment only. Another time-course experiment was performed but this time using DAF-loaded cells (i) not supplemented with l-arginine; (ii) supplemented with l-arginine; and (iii) treated with the NOS inhibitor l-NNA (1 mM) and then supplemented with l-arginine. This experiment with l-arginine was repeated twice (total n = 7). Fluorescence measurements were also performed for a total of 10 min but, after 3 min of measurements, 10 mM l-arginine was injected into the samples for the second and the third treatments.

In order to gain some insight concerning the pathway of NO production, experiments were subsequently performed using different inhibitors and upon subsequent supplementation of either sodium nitrite (3 mM) or l-arginine (10 mM). The DAF-loaded samples (four replicates per treatment) were treated with the different inhibitors for 20 min and then the substrates were added and left to react for 15 min before the fluorescence readings were taken with the fluorescence spectrophotometer. To assess the possible implication of the NR enzyme, sodium tungstate (2 mM) was supplemented to the cells. The inhibitor rotenone (20 μM) was used to block the flow of electrons at complex I inside the mitochondria, and l-NNA (1 mM) was used to inhibit the NOS-like activity in the cells of the microalgae studied.

For the temperature experiment performed at 27 and 34°C, four independent experiments were performed (total n = 21) while for the temperature experiment performed at five different temperatures, one experiment was performed (n = 5). Samples were submitted to a thermal shock and incubated for 2 h in incubators whose temperature was set either to 27, 31.5, 34, 37.5 or 41°C. The irradiance inside the incubators was 50 μmol photons m⁻² s⁻¹, i.e. identical to the growth irradiance. After the temperature treatment, cells were loaded with DAF-2 DA (loading conditions: 7.5 μM DAF-2 DA for 1 h at 27°C) and the fluorescence emitted by the DAF-loaded cells detected using the same fluorescence spectrophotometer. The initial and final cell number in each of these samples was determined microscopically using the hemocytometer and used to calculate the percentage of cell loss (final cell number/initial cell number × 100), while the MI was determined as mentioned above.

Determination of the photosynthetic efficiency (F_v/F_m)

In a separate experiment, but using the exact same settings as for the temperature experiment described above, three replicate samples were submitted to a thermal shock and incubated for 2 h in incubators whose temperature was set either to 27, 31.5, 34, 37.5 or 41°C under an irradiance of 50 μmol photons m⁻² s⁻¹. At the end of the thermal shock treatment, samples were dark-acclimated for 1 h at the in situ temperature in order to allow for the relaxation of the non-photochemical quenching. After this period, the minimal fluorescence (F_o) of the samples was determined using the fluorescence spectrophotometer with an excitation wavelength of 430 nm and a detection wavelength of 680 nm (Vincent 1983). After the reading, DCMU (1 × 10⁻⁵ M) was added the samples which were mixed and kept in the dark for 10 min. This treatment inhibited the photosynthetic electron transfer, and the maximal fluorescence (F_m) arising from the samples was then read using the spectrophotometer. The variable fluorescence (F_v) was calculated as F_m − F_o, and then F_v/F_m was calculated to provide an indication of the physiological state of the cells.

Statistical analyses

Prior to the statistical analysis of the data, the normality of the distribution of the residuals was checked with the Kolmogorov–Smirnov test and the variance homogeneity was checked using the Hartley test (Zar 1984). To identify if significant differences exist between the different combinations of inhibitors and substrates applied to the microalgae (Fig. 4), a two-way analysis of variance (ANOVA; SYSTAT v. 8.0, SPSS Inc., USA) was performed. The dependent variable was the fluorescence intensity and the factors tested were the types of inhibitors and substrates used. Because the distribution of the residuals was not normal, the data were log transformed before performing the ANOVA. When significant differences were revealed, Fisher's LSD (least significant difference) test was used as a post hoc test. To compare the fluorescence intensity reached at 27 and 34°C, and at 27°C in the presence or not of the NO donor (Fig. 5), t-tests were performed. For the temperature experiment performed at five different temperatures, one-way ANOVAs were performed (Figs. 6, 7). The factor tested was the temperature and the dependent variables were either the fluorescence intensity, the F_v/F_m, the percentage cell loss or the MI. For the F_v/F_m, the data were log transformed prior to the ANOVA in order to meet the requirement for the normality of the distribution of the residuals. When a significant difference was found, Fisher's LSD tests were used as post hoc tests.

Funding

Japan Society for the Promotion of Science postdoctoral fellowship (to J.N.B.); Ministry of Education, Science, Sports and Culture, Japan Grant-in-Aid for Scientific Research (B) (to H.Y.)

Abbreviations:

Abbreviations:
 
• ANOVA

  analysis of variance

 
• cPTIO

  2-(4-carboxyphenyl)-4,4,5,5-tetramethylimidazoline-1-oxyl-3-oxide

 
• DAF

  diaminofluorescein

 
• FITC

  fluorescein isothiocyanate

 
• l-NNA

  l-N^G-nitroarginine

 
• LSD

  least significant difference

 
• MI

  mitotic index

 
• NO

  nitric oxide

 
• NOC5

  3-[2-hydroxy-1-(1-methylethyl)-2-nitrosohydrazino]-1-propanamine

 
• NOS

  nitric oxide synthase

 
• NR

  nitrate reductase.