Shining Light on Photosynthesis in the Harmful Dinoﬂagellate Karenia mikimotoi –Responses to Short-Term Changes in Temperature, Nitrogen Form, and Availability

: Karenia mikimotoi is a toxic bloom-forming dinoﬂagellate that sometimes co-blooms with Karenia brevis in the Gulf of Mexico, especially on the West Florida Shelf where strong vertical temperature gradients and rapid changes in nitrogen (N) can be found. Here, the short-term interactions of temperature, N form, and availability on photosynthesis–irradiance responses were examined using rapid light curves and PAM ﬂuorometry in order to understand their interactions, and how they may affect photosynthetic yields. Cultures of K. mikimotoi were enriched with either nitrate (NO 3 − ), ammonium (NH 4+ ), or urea with varying amounts (1, 5, 10, 20, 50 µ M-N) and then incubated at temperatures of 15, 20, 25, 30 ◦ C for 1 h. At 15–25 ◦ C, ﬂuorescence parameters (Fv/Fm, rETR) when averaged for all N treatments were comparable. Within a given light intensity, increasing all forms of N concentrations generally led to higher photosynthetic yields. Cells appeared to dynamically balance the “push” due to photon ﬂux pressure and reductant generation, with consumption in overall metabolism (“pull” due to demand). However, at 30 ◦ C, all ﬂuorescence parameters declined precipitously, but differential responses were observed depending on N form. Cells enriched with urea at 30 ◦ C showed a smaller decline in ﬂuorescence parameters than cells treated with NO 3 − or NH 4+ , implying that urea might induce a photoprotective mechanism by increasing metabolic “pull”.


Introduction
Karenia mikimotoi, the close relative of the harmful dinoflagellate Karenia brevis that forms massive annual blooms mostly in the Gulf of Mexico, is a hemolytic and cytotoxic dinoflagellate found in coastal waters around the world, including Asia (e.g., Japan, China, Korea, Hong Kong), Europe (e.g., Norway, United Kingdom, Denmark, Spain), Oceania (e.g., Australia, New Zealand), and America (e.g., United States, Chile) [1,2]. Several Karenia species co-exist and often co-bloom with K. brevis in the Gulf of Mexico, especially on the West Florida Shelf [1,3]. Although highly variable from year-to-year, Karenia spp. blooms on the West Florida Shelf commonly occur from early fall through winter and only rarely persist as substantial blooms throughout the summer [4].
Temperature is an important variable in understanding the physiological success of phytoplankton taxa, but due to its effect on key enzymes and the different temperature optima of enzymes involved in photosynthesis and metabolism of different forms of nitrogen (N), temperature may have different effects on physiological status depending on growth N form. Seasonal temperature changes on the West Florida Shelf range from lows of~15 • C in winter when K. brevis blooms are more common, to highs of >30 • C in midsummer. Sharp vertical gradients in temperature may also exist. For example, Weisberg et al. [5] reported, for the West Florida Shelf for the month of July 2010, that temperatures changed from near 29 • C at the surface to near 17 • C near the bottom, where waters between subspecies, K. mikimotoi is a eurythermal (4-31 • C) and euryhaline (salinity of 9-35) species [1]. While generally thought to be a low-light specialist, K. mikimotoi is also capable of growing well in a wide range of light intensities [1,2] and is positively phototactic. Nutritionally, Karenia spp. are flexible, utilizing various inorganic and organic nutrients and showing phago-mixotrophy [1,[20][21][22][23][24][25]. On the West Florida Shelf, diverse nutrient sources (e.g., land-derived nutrient inputs, microbially regenerated nutrients, regenerated nutrients from decaying fish during toxic blooms, nitrification) may be available to be taken up by Karenia spp. [23]. While Karenia spp. are primarily photo-autotrophic, they may also consume prey, and higher growth rates of K. brevis [23,24] and K. mikimotoi [21,22] have been observed when they have ingested prey organisms compared to growth on inorganic nutrient sources.
In this study, the objective was to analyze photosynthesis in K. mikimotoi, in response to short-term exposures to varying combinations of temperature, N form, and concentration. We focused on short-term (hour time scale) exposure to different temperatures, bracketing the temperatures most commonly experienced on the West Florida Shelf, simulating how cells may respond to environmental changes that may be experienced near coastal fronts, in vertical gradients, or near nutrient sources. Photosynthesis-irradiance responses were examined using rapid light curves. We hypothesized that photosynthetic efficiency in K. mikimotoi will increase with temperature and with increasing N concentrations (all forms), but that with increasing temperatures, photosynthetic efficiency-the relative amount of absorbed light energy that can be used to drive photosynthesis-will increase to a greater extent, with increasing concentrations of chemically reduced forms of N compared to changes with increasing concentrations of NO 3 − .

Algal Culture
A culture of non-axenic K. mikimotoi (ARC165, originally isolated from Venice, Florida in 2006) was obtained from Mote Marine Laboratory, Sarasota, FL, USA. The K. mikimotoi were grown in semi-batch cultures using L1-silicate media with N added as NO 3 − [26]. All culture media were prepared with 0.2 µm filtered nearshore west Florida waters and sterilized by autoclaving. Cultures were grown over 12 h light:12 h dark cycle under 40 µmol photons m −2 s −1 at salinity 32~34 and 20 • C.

Specific Growth Rate, Nutrient Analysis
While maintaining the stock culture, changes in cell number over time were monitored by counting cells with light microscopy using a hemocytometer after fixing with 5% Lugol's solution. The specific growth rate (µ) of K. mikimotoi was calculated according to where X 1 and X 2 are the cell numbers at time t 1 and t 2 , respectively. At the start of the experiment, samples for cell counts and ambient nutrients were collected. Aliquots of culture were filtered through precombusted (450 • C for 2 h) Whatman GF/F filters (Florham Park, NJ, USA) were stored in a −20 • C freezer for further analysis. The filtrates were reserved for ambient nutrient analysis, which was later done via autoanalysis at the Mote Marine Laboratory.

Experimental Setup
To initiate the experiment, the culture was first diluted to a final concentration of 6200 cells ml −1 with autoclaved filtered seawater in order to reduce the concentrations of residual nutrients. Aliquots of the diluted culture were then transferred to 60, 50 mL polypropylene tubes. The tubes were divided into 12 batches of five tubes ( Figure 1). Within each batch, tubes received an enrichment of either NO 3 − , NH 4 + or urea at the following concentrations: 1, 5, 10, 20, 50 µM-N, as 15  To initiate the experiment, the culture was first diluted to a final concentration of 6200 cells ml −1 with autoclaved filtered seawater in order to reduce the concentrations of residual nutrients. Aliquots of the diluted culture were then transferred to 60, 50 mL polypropylene tubes. The tubes were divided into 12 batches of five tubes (Figure 1). Within each batch, tubes received an enrichment of either NO3 − , NH4 + or urea at the following concentrations: 1, 5, 10, 20, 50 µM-N, as 15 N (99% enriched). One batch of each substrate enrichment gradient (NO3 − , NH4 + or urea) was then incubated for 1 h at each of four different temperatures. To do so, tubes were placed in temperature-controlled water baths maintained at 15, 20, 25, 30 °C. Tubes were illuminated continuously by LED lights with 350-400 µmol photons m −2 s −1 as measured with QSL-100 quantum scalar irradiance meter (Biospherical Instruments Inc., San Diego, CA, USA). At the end of the incubation period, tubes were subsampled for photosynthesis measurements and 15 N incorporation as described below. Figure 1. Schematic of the experimental design to measure 15 N incorporation and photosynthesisirradiance responses of K. mikimotoi exposed to different temperatures, N forms, and concentrations. Green tubes represent those enriched with urea; blue tubes-those enriched with NH4 + ; and red tubes-those enriched with NO3 − .

Chl Fluorescence Measurements
After the 1 h incubation at the designated temperature, black multiwell plates (24 wells each) were filled with K. mikimotoi cells from the various tubes (3 mL in each well) and were dark acclimated for 20 min to fully open (i.e., oxidize) the PSII reaction center Chl and to minimize non-photochemical energy dissipation. Fluorescence emissions from PS II were monitored by a MAXI-Imaging PAM M-Series (Heinz Walz GmbH, Effeltrich, Germany), equipped with 44 blue LED lamps. Light intensities were calibrated with a ULM-500 Light Meter & Logger equipped with a micro quantum sensor (Heinz Walz GmbH, Effeltrich, Germany). Rapid light curves were run to monitor PS II fluorescence emissions as a function of rapid changes in irradiance (12 steps for 20 s at each step: 0, 4, 10, 28, 53, 86, 128, 178, 236, 377, 460, 671 µmol photons m −2 s −1 ). The data were exported from the ImagingWin software (Heinz Walz GmbH, Effeltrich, Germany) and fluorescence parameters were calculated (Supplementary Materials, Table S1). For the dark-adapted cells, the maximum quantum yields of PSII photochemistry were calculated as the ratio of variable fluorescence (Fv) to maximum fluorescence (Fm), Fv/Fm [15,27].

Chl Fluorescence Measurements
After the 1 h incubation at the designated temperature, black multiwell plates (24 wells each) were filled with K. mikimotoi cells from the various tubes (3 mL in each well) and were dark acclimated for 20 min to fully open (i.e., oxidize) the PSII reaction center Chl and to minimize non-photochemical energy dissipation. Fluorescence emissions from PS II were monitored by a MAXI-Imaging PAM M-Series (Heinz Walz GmbH, Effeltrich, Germany), equipped with 44 blue LED lamps. Light intensities were calibrated with a ULM-500 Light Meter & Logger equipped with a micro quantum sensor (Heinz Walz GmbH, Effeltrich, Germany). Rapid light curves were run to monitor PS II fluorescence emissions as a function of rapid changes in irradiance (12 steps for 20 s at each step: 0, 4, 10, 28, 53, 86, 128, 178, 236, 377, 460, 671 µmol photons m −2 s −1 ). The data were exported from the ImagingWin software (Heinz Walz GmbH, Effeltrich, Germany) and fluorescence parameters were calculated (Supplementary Materials, Table S1). For the dark-adapted cells, the maximum quantum yields of PSII photochemistry were calculated as the ratio of variable fluorescence (Fv) to maximum fluorescence (Fm), Fv/Fm [15,27]. Subsequently, the effective quantum yield of PSII photochemistry at each light step was calculated as where F'm is light-acclimated maximum fluorescence and Ft is the steady-state value of fluorescence immediately prior to a flash for a light-adapted sample, respectively [14,28].
The quantum yield of non-regulated non-photochemical energy dissipation (YNO) and the quantum yield of regulated non-photochemical energy dissipation (YNPQ) were further calculated to analyze the allocation of PSII excitation energy at each light step [28]. The sum of these three yields equals 1.
The relative electron transfer rate (rETR) of PS II in units of µmol electrons m −2 s −1 was calculated as rETR = 0.5 × YII × PAR × 0.84 (3) which is the product of the effective quantum yield (YII) and photosynthetically active radiation (PAR), adjusting the product by a factor of 0.5, which takes into account that half of the absorbed light energy is directed to PSII, and by the factor 0.84, which is the presumed PAR absorptivity, e.g., [14]. Photosynthesis-irradiance curves were fitted using the formula of Platt et al. [29] and rETR parameters were calculated using the R package 'phytotools' (https://cran.r-project.org/package=phytotools (accessed on 12 April 2021). The photosynthetic quenching parameter, qP, was calculated according to in which minimum fluorescence level of illuminated sample is F'o. In order to convert this to a measure of excitation pressure, the value 1-qP was calculated. This value is also interpreted as the number of photosynthetic reaction centers that are closed [30]. Thus, whereas Fv/Fm is a measure of quantum efficiency of PSII and the efficiency of nonphotochemical quenching (when all reaction centers are considered open and the cells are dark-adapted), any change in qP reflects the degree of reaction center closures under varying light conditions.

15 N Isotope Analysis
After incubation, and immediately after the filling of the microwell plates for photosynthetic measurements, the remaining volumes in each of the experimental tubes were filtered through precombusted (450 • C 2 h) Whatman GF/F filters. The filters were immediately frozen, then subsequently dried and transferred to tin capsules. Isotope analysis was then undertaken by the UC Davis Stable Isotope Facility. Values are reported as 15 N atom percent enrichment of samples at the end of the incubation periods [31].

Statistical Analysis
All statistical analyses were performed with R. The normality of data was checked with Shapiro-Wilks test and the homogeneity of variance was assessed with Levene's test in R package 'car'. Differences between different temperature (15,20,25, 30 • C) and N forms (NO 3 − , NH 4 + , urea) of fluorescence data (e.g., Fv/Fm, rETR parameters) were tested using one-way analysis of variance (ANOVA), followed by Tukey's HSD test for pairwise comparisons.

Initial Cell and Nutrient Conditions
The specific growth rate of the initial undiluted culture at the time of the experiment was 0.23 day −1 . The cells were not nutrient-depleted and the measured concentrations of nutrients after dilution with filtered seawater were 289 µM ± 2.8 and 1.1 ± 0.5 µM for NO 3 − and NH 4 + , respectively, and 7.7 ± 0 µM for PO 4 3− . Urea concentrations were negligible.

PSII Quantum Efficiency and Quantum Yields
Over the range of 15-25 • C, the maximum quantum yields of PSII photochemical efficiency, Fv/Fm, did not differ significantly between the N treatments when the different N concentrations for each temperature were averaged (ANOVA, p > 0.05) (Tables 1 and 2). The Fv/Fm averaged 0.44 ± 0.03 for the 15-25 • C treatments when all N forms and concentrations are taken into account. However, within each N form, Fv/Fm tended to increase with temperature (ANOVA, p < 0.05) ( Table 2) and concentrations ( Figure 2). In contrast, at 30 • C, the Fv/Fm of all the N-enriched samples dropped precipitously. The Fv/Fm of samples enriched with NO 3 − and NH 4 + and incubated at 30 • C dropped to <0.1, while those with urea enrichment fell to a range of 0.1-0.2. The Fv/Fm values of the ureatreated samples (0.17 ± 0.04) were 2-fold higher than those of the NH 4 + -treated samples (0.07 ± 0.03) (Tukey's HSD, p < 0.01) and 9-fold higher than those of the NO 3 − -treated samples (0.02 ± 0.03) (Tukey's HSD, p < 0.001) ( Figure 2 and Table 2). At 30 • C, the Fv/Fm values also showed a general decline when enriched at concentrations >5 µM-N ( Figure 2). This trend was most apparent for samples enriched with urea.

PSII Quantum Efficiency and Quantum Yields
Over the range of 15-25 °C, the maximum quantum yields of PSII photochemical efficiency, Fv/Fm, did not differ significantly between the N treatments when the different N concentrations for each temperature were averaged (ANOVA, p > 0.05) (Tables 1 and  2). The Fv/Fm averaged 0.44 ± 0.03 for the 15-25 °C treatments when all N forms and concentrations are taken into account. However, within each N form, Fv/Fm tended to increase with temperature (ANOVA, p < 0.05) ( Table 2) and concentrations ( Figure 2). In contrast, at 30 °C, the Fv/Fm of all the N-enriched samples dropped precipitously. The Fv/Fm of samples enriched with NO3 − and NH4 + and incubated at 30 °C dropped to <0.1, while those with urea enrichment fell to a range of 0.1-0.2. The Fv/Fm values of the ureatreated samples (0.17 ± 0.04) were 2-fold higher than those of the NH4 + -treated samples (0.07 ± 0.03) (Tukey's HSD, p < 0.01) and 9-fold higher than those of the NO3 − -treated samples (0.02 ± 0.03) (Tukey's HSD, p < 0.001) ( Figure 2 and Table 2). At 30 °C, the Fv/Fm values also showed a general decline when enriched at concentrations >5 µM-N ( Figure  2). This trend was most apparent for samples enriched with urea.    . Effective Photosystem II photochemical efficiency, YII, as a function of irradiance for K. mikimotoi exposed to different temperatures (15, 20, 25, 30 °C) when pulsed with different nitrogen forms (NO3 − , NH4 + , urea) and amounts (1, 5, 10, 20, 50 µM-N). Within each depicted irradiance level, the darker the color of the bars, the higher the enrichment with added N.

Relative Electron Transfer Rate
The rapid light response curves of rETR showed a typical saturating response with little evidence of photoinhibition at 15-25 °C at all N forms and concentrations (Figure 4ai). Only a few treatments showed down-regulation of photochemical reactions, thus exhibited reduction in rETR values at the highest light intensity (e.g., 20 and 50 µM-N of urea at 25 °C). As with values of Fv/Fm and YII, values of rETR were reduced significantly at 30 °C across the light range, but the extent of reduction differed depending on the form of N (ANOVA, p < 0.01) (Figure 4j-l). Both light-limited and light-saturated parts of the

Relative Electron Transfer Rate
The rapid light response curves of rETR showed a typical saturating response with little evidence of photoinhibition at 15-25 • C at all N forms and concentrations (Figure 4a-i). Only a few treatments showed down-regulation of photochemical reactions, thus exhibited reduction in rETR values at the highest light intensity (e.g., 20 and 50 µM-N of urea at 25 • C). As with values of Fv/Fm and YII, values of rETR were reduced significantly at 30 • C across the light range, but the extent of reduction differed depending on the form of N (ANOVA, p < 0.01) (Figure 4j-l). Both light-limited and light-saturated parts of the rETR versus irradiance curves responded to changes in N concentrations with stronger effects on the light-saturated regions at 15-25 • C (Figure 4a Table S2). rETR versus irradiance curves responded to changes in N concentrations with stronger effects on the light-saturated regions at 15-25 °C (Figure 4a-i). For example, the lightlimited rETR of the 1 µM-N NH4 + treatment at 15 °C (0-178 µmol photons m −2 s −1 ) increased 43% compared to the 50-µM-N NH4 + treatment at 15 °C, and the light-saturated rETR of the 1µM-N NH4 + treatment at 15°C (236-460 µmol photons m −2 s −1 ) increased 63% compared to the 50 µM-N NH4 + treatment at 15 °C (rETR parameters including saturation irradiance, Ik, are presented in Supplementary Materials, Table S2). No significant temperature effects on rETRmax were found at 15-25°C at each N source (ANOVA > 0.05) ( Table 2) while N source effects were found at two temperatures (Tukey's HSD, p < 0.05) ( Table 1). The rETRmax values of the NH4 + -enriched cultures (15.6 ± 3.0) were No significant temperature effects on rETR max were found at 15-25 • C at each N source (ANOVA > 0.05) ( Table 2) while N source effects were found at two temperatures (Tukey's HSD, p < 0.05) ( Table 1). The rETR max values of the NH 4 + -enriched cultures (15.6 ± 3.0) were significantly higher than those of samples enriched with NO 3 − (11.2 ± 2.0) at 15 • C (Tukey's HSD, p < 0.05) and rETR max values of the NO 3 − -enriched cultures (0.9 ± 0.8) were markedly lower than those of the NH 4 + -enriched samples (2.4 ± 0.7) (Tukey's HSD, p < 0.05), and the urea-treated samples (3.1 ± 0.6) at 30 • C (Tukey's HSD, p < 0.01) ( Figure 5, Table 2).

Excitation Pressure
The excitation pressure, 1-qP, increased with increasing transient light intensities at 15-25 • C (Figure 6a-i). When cells were fully dark-adapted (I = 0 µmol photons m −2 s −1 ), reaction centers were all open and 1-qP values were low, but by 10 µmol photons m −2 s −1 , an increasing fraction of reactions centers were closed. At each light intensity, excitation pressure decreased with increasing concentrations of N forms (Figure 6a-i). However, cells exposed to 30 • C exhibited closed reaction centers across all nearly all irradiance levels for samples enriched with NO 3 − and NH 4 + , while for samples enriched with urea, full reaction center closure was observed only for samples exposed to >100 µmol photons m −2 s −1 (Figure 6j-l).

Excitation Pressure
The excitation pressure, 1-qP, increased with increasing transient light intensities at 15-25 °C (Figure 6a-i). When cells were fully dark-adapted (I = 0 µmol photons m −2 s −1 ), reaction centers were all open and 1-qP values were low, but by 10 µmol photons m −2 s −1 , an increasing fraction of reactions centers were closed. At each light intensity, excitation pressure decreased with increasing concentrations of N forms (Figure 6a-i). However, cells exposed to 30 °C exhibited closed reaction centers across all nearly all irradiance levels for samples enriched with NO3 − and NH4 + , while for samples enriched with urea, full reaction center closure was observed only for samples exposed to >100 µmol photons m −2 s −1 (Figure 6j-l).
Measures of the quantum yields of regulated and non-regulated non-photochemical energy dissipation (YNPQ and YNO) for all experimental treatments are provided in the in Supplementary Materials (Figures S1 and S2).

15 N Isotope Incorporation
The incorporation of 15 N (as 15 N atom percent) varied with substrate, and temperature. Over the concentration range, the 15 N enrichment of NO3 − was linear for each temperature, and only a slight, but not significant, increase in 15 N incorporation was noted at 30 °C compared to the other temperatures (Figure 7a). Isotopic enrichment levels were highest for NH4 + , and all temperatures indicated a saturating response across the Measures of the quantum yields of regulated and non-regulated non-photochemical energy dissipation (YNPQ and YNO) for all experimental treatments are provided in the in Supplementary Materials (Figures S1 and S2).

15 N Isotope Incorporation
The incorporation of 15 N (as 15 N atom percent) varied with substrate, and temperature. Over the concentration range, the 15 N enrichment of NO 3 − was linear for each temperature, and only a slight, but not significant, increase in 15 N incorporation was noted at 30 • C compared to the other temperatures (Figure 7a). Isotopic enrichment levels were highest for NH 4 + , and all temperatures indicated a saturating response across the concentration gradient. At saturation, the atom % enrichment levels at 25 • C and 30 • C were significantly higher than those measured at 15 • C and 20 • C (Figure 7b). For urea, there was no saturation in the response, but the 15 N atom% values across all concentrations were significantly higher at 30 • C than atom% enrichment values at the other temperatures, which did not differ from each other (Figure 7c).

Discussion
Light, nutrients, and temperatures drive different rates of metabolism within the cell, and the challenge for the cell is to balance energy and reductant generation with that of its consumption. Here, varying each of these factors in short-term exposures, the dynamic balance between the "push" due to photon flux pressure and the metabolic "pull" of diverse metabolic reactions of K. mikimotoi was investigated ( Figure 8). These results are complementary to the dynamic balance theory of photosynthetic regulation [16], which articulated the notion that the concentration of photosynthetic pigments in algae could be explained by the opposing forces of excitation pressure on PSII and degree of reduction of the plastoquinone pool of PSII. Therefore, acclimated pigment concentrations balance the light energy input and biochemical energy use in response to environmental conditions, demonstrating the sensitivity of PSII to metabolic feedback. Once an excitation event occurs in PSII, the downstream redox state determines the electron sink capacity. The redox state around PSII responds rapidly to changes in light availability (and Calvin cycle activity), N reduction and assimilation. Whereas pigment changes with multiple environmental changes were emphasized by Kana et al. [16], in their description of the dynamic balance theory, various fluorescent parameters, including the excitation pressure on PSII, were measured directly herein with changes in light, temperature, and N availability.
The initial hypothesis, that photosynthetic efficiency in K. mikimotoi will increase with temperature and with increasing N concentrations, and that with increasing temperatures, photosynthetic efficiency will increase to a greater extent with increasing concentrations of chemically reduced forms of N, compared to changes with increasing concentrations of NO3 − , was largely borne out, but the highest temperature tested yielded differential stresses depending on N form. Here, the interactions of the different physiological processes measured are described, as well as implications for growth of Karenia on the West Florida Shelf.

Discussion
Light, nutrients, and temperatures drive different rates of metabolism within the cell, and the challenge for the cell is to balance energy and reductant generation with that of its consumption. Here, varying each of these factors in short-term exposures, the dynamic balance between the "push" due to photon flux pressure and the metabolic "pull" of diverse metabolic reactions of K. mikimotoi was investigated ( Figure 8). These results are complementary to the dynamic balance theory of photosynthetic regulation [16], which articulated the notion that the concentration of photosynthetic pigments in algae could be explained by the opposing forces of excitation pressure on PSII and degree of reduction of the plastoquinone pool of PSII. Therefore, acclimated pigment concentrations balance the light energy input and biochemical energy use in response to environmental conditions, demonstrating the sensitivity of PSII to metabolic feedback. Once an excitation event occurs in PSII, the downstream redox state determines the electron sink capacity. The redox state around PSII responds rapidly to changes in light availability (and Calvin cycle activity), N reduction and assimilation. Whereas pigment changes with multiple environmental changes were emphasized by Kana et al. [16], in their description of the dynamic balance theory, various fluorescent parameters, including the excitation pressure on PSII, were measured directly herein with changes in light, temperature, and N availability.

Temperature, Energy (Light) and Material (N) Addition Effects on Photosynthesis
The biochemical reactions of photosynthetic metabolism are generally considered to be more sensitive to temperature than the light reactions, and increase with temperature up to an optimum temperature, above which they decline rapidly [7,32]. Shen et al. [33] reported for K. mikimotoi that the effective quantum yield, YII (pulsed with 70 µmol photons m −2 s −1 ), and rETRmax increased from 16-28 °C although only estimates from 16 °C were statistically lower than those measured at 20, 24, and 28 °C. In this study, generally the maximum quantum yield, Fv/Fm, increased with temperature at each N form (e.g., 0.41 ± 0.01, 0.43 ± 0.04, 0.45 ± 0.02 at 15, 20, 25 °C, respectively, when cells were NO3 −enriched) although differences were statistically not significant in the case of NH4 +enriched cells (Table 2). However, when cells were light saturated, no clear temperature effects were found in rETRmax in the range of 15-25 °C ( Table 2). The cultures from Shen et al. [33] were acclimated to each temperature while our cultures were adapted to 20 °C and exposed to new thermal environments for 1 h. The optimal temperature for maximum photosynthetic rate can increase with increasing growth temperature [34]. At 30 °C, all photosynthetic parameters dropped significantly, suggesting that even a 1 h exposure to this temperature was stressful for K. mikimotoi. Note that herein additional cultures incubated at 30 °C overnight (no extra N addition) did not survive (data not shown).
At all but the highest temperature tested, with increasing light intensity, i.e., increasing "push" due to photon flux pressure (Figure 8), rETR increased progressively until the PSII was saturated with light (Ik) (Figure 4). Although the efficiency of photochemical light capture, YII, decreased with increasing irradiance (Figure 3), enhanced photon flux pressure on PSII increased the quantity of energy channeling to photochemistry, thus, increasing the rETR until Ik. At a given irradiance, rETR increased further with increasing N concentrations. Even though cultures were N-sufficient, photosynthetic activity increased with all forms of N addition when temperature was not at a stressful level (Figure 4). The increased supply of material (N in this case) enhanced metabolic activity, thus raising the demand for energy to run diverse metabolism ("pull" due to demand) (Figure 8). This enhancement was stronger when cells were light saturated. The rETRmax clearly responded to increasing N addition with little variations at 15-25 °C ( Figure 5). The dark-adapted fluorescence parameter, Fv/Fm, was also generally increased with all forms of N addition (Figure 2).

N Form Effects on Photosynthesis at 30 °C
When temperature was not at a stressful level, K. mikimotoi was able to operate its photosynthetic apparatus efficiently with all forms of N tested in this study. However, at The initial hypothesis, that photosynthetic efficiency in K. mikimotoi will increase with temperature and with increasing N concentrations, and that with increasing temperatures, photosynthetic efficiency will increase to a greater extent with increasing concentrations of chemically reduced forms of N, compared to changes with increasing concentrations of NO 3 − , was largely borne out, but the highest temperature tested yielded differential stresses depending on N form. Here, the interactions of the different physiological processes measured are described, as well as implications for growth of Karenia on the West Florida Shelf.

Temperature, Energy (Light) and Material (N) Addition Effects on Photosynthesis
The biochemical reactions of photosynthetic metabolism are generally considered to be more sensitive to temperature than the light reactions, and increase with temperature up to an optimum temperature, above which they decline rapidly [7,32]. Shen et al. [33] reported for K. mikimotoi that the effective quantum yield, YII (pulsed with 70 µmol photons m −2 s −1 ), and rETR max increased from 16-28 • C although only estimates from 16 • C were statistically lower than those measured at 20, 24, and 28 • C. In this study, generally the maximum quantum yield, Fv/Fm, increased with temperature at each N form (e.g., 0.41 ± 0.01, 0.43 ± 0.04, 0.45 ± 0.02 at 15, 20, 25 • C, respectively, when cells were NO 3 − -enriched) although differences were statistically not significant in the case of NH 4 + -enriched cells (Table 2). However, when cells were light saturated, no clear temperature effects were found in rETR max in the range of 15-25 • C ( Table 2). The cultures from Shen et al. [33] were acclimated to each temperature while our cultures were adapted to 20 • C and exposed to new thermal environments for 1 h. The optimal temperature for maximum photosynthetic rate can increase with increasing growth temperature [34]. At 30 • C, all photosynthetic parameters dropped significantly, suggesting that even a 1 h exposure to this temperature was stressful for K. mikimotoi. Note that herein additional cultures incubated at 30 • C overnight (no extra N addition) did not survive (data not shown).
At all but the highest temperature tested, with increasing light intensity, i.e., increasing "push" due to photon flux pressure (Figure 8), rETR increased progressively until the PSII was saturated with light (Ik) (Figure 4). Although the efficiency of photochemical light capture, YII, decreased with increasing irradiance (Figure 3), enhanced photon flux pressure on PSII increased the quantity of energy channeling to photochemistry, thus, increasing the rETR until Ik. At a given irradiance, rETR increased further with increasing N concentrations. Even though cultures were N-sufficient, photosynthetic activity increased with all forms of N addition when temperature was not at a stressful level (Figure 4). The increased supply of material (N in this case) enhanced metabolic activity, thus raising the demand for energy to run diverse metabolism ("pull" due to demand) (Figure 8). This enhancement was stronger when cells were light saturated. The rETR max clearly responded to increasing N addition with little variations at 15-25 • C ( Figure 5). The dark-adapted fluorescence parameter, Fv/Fm, was also generally increased with all forms of N addition ( Figure 2).

N Form Effects on Photosynthesis at 30 • C
When temperature was not at a stressful level, K. mikimotoi was able to operate its photosynthetic apparatus efficiently with all forms of N tested in this study. However, at 30 • C, cells were dissipating most of their absorbed light energy non-photochemically, but at different degrees depending on N form. At the highest temperature, reaction centers were fully closed for the NO 3 − and NH 4 + treatments across most irradiance levels (Figure 6j,k). The NO 3 − -enriched culture showed the strongest stress (Fv/Fm = 0.02, rETR max = 0.9), and then NH 4 + -enriched cells (Fv/Fm = 0.07, rETR max = 2.4; Table 2). Fluorescence parameters were still significantly lower at this high temperature compared to those at 15-25 • C, but when cells were treated with urea, K. mikimotoi experienced less stress than those cells enriched with NO 3 − and NH 4 + (Fv/Fm = 0.17, rETR max = 3.1), implying that urea might induce a photoprotective mechanism.
The 15 N isotope enrichment trends confirmed the differential effects of temperature with the different forms of N ( Figure 7). Little effect of temperature was observed with 15 NO 3 − incorporation, but with 15 NH 4 + , incorporation increased as temperatures rose, and for urea, the highest temperature uniquely yielded the highest isotopic incorporation. Such trends would support the different responses seen in Fv/Fm and rETR with the different N enrichments.
Although K. mikimotoi has been known to use diverse forms of N, the underlying mechanisms for metabolizing each N form are complex with respect to temperature. The inhibition of NR activity at higher temperatures might hinder the first step of NO 3 − metabolism, reduction to NO 2 − [7], thereby decreasing subsequent intracellular N supply. Dai et al. [35] reported that K. mikimotoi grown on NO 3 − could not survive at 30 • C and NR activity was significantly inhibited at this temperature compared to 15-25 • C. On the other hand, the metabolic flux of reduced forms of N such as NH 4 + and urea is expected to be higher at higher temperatures [7]. For example, at higher temperatures, the activity of GS that adds NH 4 + to glutamate by using ATP to produce glutamine increases. Moreover, the activity of GOGAT that combines glutamine to 2-oxoglutarate using electrons can be higher at 30 • C compared to activity rates at lower temperatures [7,36]. All energyconsuming metabolism will increase the energy demand, thereby enhancing the capacity of photochemical use of absorbed light energy. This might partly explain why NH 4 + -treated sample showed four times higher Fv/Fm and three times higher rETR max compared to NO 3 − -treated cells at 30 • C (Figures 2 and 5, Table 2). However, cells enriched with urea showed even higher estimates, implying that energy demand for metabolism could be relatively higher. Dinoflagellates have a urea cycle that helps reallocation of intracellular C and N [36] and several enzymes involved in the urea cycle have been identified in K. mikimotoi, suggesting that this species could also have a complete urea cycle [37]. The increased intracellular recycling of C and N from protein catabolism and photorespiration through the urea cycle could enhance overall metabolic flux, thus increasing the demand for chemical energy. To balance this demand, a higher proportion of absorbed light energy could be funneled into photochemistry, therefore increasing Fv/Fm and rETR. Moreover, generally, urease, which catalyzes the hydrolysis of urea, has a positive relationship with temperature [13,38].

Implications of Temperature Effects for Natural Assemblages
In the field, K. mikimotoi has been observed over a wide range of temperatures, 4-31 • C although some subspecies have narrower temperature ranges (e.g., Norway isolate: 10-25 • C) [1,2,39]. In culture experiments, it has been shown that K. mikimotoi cells can survive and grow well at 16, 20, 24, 28 • C [33,40], but not at 10 and 30 • C [35], suggesting that K. mikimotoi would have a relatively narrow optimum temperature for their growth [1,40]. In general, rising temperatures accelerate enzymatic reactions by increasing the proportions of molecules that have sufficient kinetic energy to react, thus increasing overall metabolic flux (the "pull" due to demand) ( Figure 8) [41,42]. Indeed, the maximum specific growth rates were observed at 24 • C and 25 • C in the previously cited studies [33,35].
Karenia blooms that primarily occurs in the fall months rarely persist throughout hot summer months on the West Florida Shelf. The summers-during which water can >30 • C-may represent a thermal barrier for large blooms to be maintained. However, Karenia cells are occasionally found to sustain substantial blooms in extremely hot surface waters (>30 • C). Generally, West Florida Shelf is characterized by low ambient dissolved inorganic nutrients, but elevated nutrients, especially dissolved organic nutrients (e.g., urea) can occur nearshore [23]. Urea has been widely used as a fertilizer within Florida and some of this nutrient can be delivered to nearshore [43]. This raises the possibility that high temperature stress might be alleviated when urea is available because more photons can be processes photochemically, thus cells can operate photosynthesis more efficiently in very warm conditions with reduced organic nitrogen. Therefore, increased availability of urea might be able to support maintaining Karenia blooms under short-term heat stress.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/phycology2010002/s1 Figure S1: Quantum yield of regulated non-photochemical energy dissipation, YNPQ, as a function of irradiance for K. mikimotoi exposed to different temperatures (15,20,25,30 • C) when pulsed with different nitrogen forms (NO 3 − , NH 4 + , urea) and amounts (1, 5, 10, 20, 50 µM). Within each depicted irradiance level, the darker the color of the bars, the higher the enrichment with added N. YNPQ represents energy loss of excitation energy thorough harmless heat dissipation related to non-photochemical quenching, Figure S2: Quantum yield of non-regulated non-photochemical energy dissipation, YNO as a function of irradiance for K. mikimotoi exposed to different temperatures (15,20,25,30 • C) when pulsed with different nitrogen forms (NO 3 − , NH 4 , urea) and amounts (1, 5, 10, 20, 50 µM). Within each depicted irradiance level, the darker the color of the bars, the higher the enrichment with added N. YNO represents the sum of non-regulated heat dissipation and fluorescence emission ("primarily constitutive losses"), Table S1: Original data of YII, YNPQ, YNO, 1-qP, and rETR, Table S2: rETR parameters, including initial slope of the light curve, alpha; saturation irradiance, Ik; rETRmax and, photoinhibition parameter, beta, as calculated using the R package 'phytotools'.