Symbiodiniaceae cultures CCMP2465 (Symbiodinium tridacnidorum, formerly clade A3), CCMP2467 (Symbiodinium microadriaticum, formerly clade A1), and CS156 (Fugacium kawagutii formerly clade F1) were grown at 24°C under 50 µmol photons m⁻² s⁻¹ white light for a week in F/2 media until reaching the mid-log growth phase. Cells were centrifuged at 5,000g and re-suspended in fresh F/2 media in such a way that the final Chlorophyll (Chl) concentration was adjusted to 5 µg/ml. Prepared sample was pre-illuminated under 50 µmol photons m⁻² s⁻¹ (growth light condition) for an hour before each measurement.

Flash-induced chlorophyll fluorescence yield (FF) was measured using a double-modulation fluorimeter (FL-3000, Photon Systems Instruments, Brno, Czech Republic) (Trtílek et al., 1997). A 2-ml sample was placed in a cuvette with 1-cm path length and was continuously stirred with a small magnetic stirrer bar in the dark. Four measuring flashes (8 μs, separated with 200-μs intervals, wavelength of 620 nm) were applied to determine minimum fluorescence in the dark (F₀), after which, a single turnover saturating actinic flash (30 μs, wavelength of 639 nm) was given to induce the formation of Q_A ⁻, which resulted in the rise of fluorescence intensity. This is the maximal fluorescence recorded after the single turnover actinic flash, which we specifically denote as F_m(ST). The fluorescence decay resulting from re-oxidation of Q_A ⁻ was measured by applying weak measuring flashes, which have negligible actinic effect and, therefore, do not interfere with the redox state of Q_A, in the time range from 150 μs to 100 s on a logarithmic time scale.

Kautsky induction curve or the polyphasic rise of chlorophyll fluorescence OJIP curve was measured using an FL-3000 fluorimeter (Photon Systems Instruments, Brno, Czech Republic). This method consists of monitoring the kinetics of the chlorophyll fluorescence induction curve in the transition between dark to light adapted state (Kalaji et al., 2017). It measures the transition between the state where all PSII reaction centers are open (referred as O step or F₀ at 20 µs) to the state where all reaction centers are closed and the fluorescence intensity reaches its maximum, nominally called as peak (referred as P step or F_m ~ 300 ms) (Strasser and Govindjee, 1992). Two intermediate steps are also observed typically: the J step (at approximately 2 ms) and the I step (at approximately 20 ms). An automated protocol was applied, which provided red actinic light (wavelength at peak, 650 nm; spectral line half-width, 22 nm) with the intensity of 3,500 μmol m⁻² s⁻¹ for the induction of OJIP fluorescence transients with a length of 3 s.

Post-illumination chlorophyll fluorescence transients (PIFT) were measured using Dual-PAM, Heinz-Walz GmbH, Effeltrich, Germany. Two milliliters of sample was dark adapted for 5 min, then chlorophyll fluorescence was recorded for 6 min. For the first 30 s, the sample was illuminated with a weak measuring light (ML) of approximately 0.1 μmol photons m⁻² s⁻¹, then an actinic light (AL, 50 μmol photons m⁻² s⁻¹) for 3 min was applied, followed by recording the Chl fluorescence rise in the dark. The same ML intensity was used for the entire measurement sequence.

The activity of Photosystem II and Photosystem I (Y(II) and Y(I), respectively) were measured using a Dual-PAM, Heinz-Walz GmbH, Effeltrich, Germany. Eight milliliters of the sample (Chl content of 40 μg/ml) with different heat treatments was filtered onto Whatman GF/C filter paper and dark adapted for 5 min. Because the cultures were filtered and spread evenly on the filter paper in thin layer, high chlorophyll content was required to obtain a good signal. Y(II) and Y(I) were determined simultaneously according to (Hoogenboom et al., 2012). Y(II) is calculated as (F_m ^’-F_t)/F_m ^’, where F_m ^’ is the maximum fluorescence upon the saturation pulse and F_t is steady-state value of fluorescence immediately prior to the flash (Maxwell and Johnson, 2000). In dark-adapted state, Y(II) (or F_v/F_m) is calculated as (F_m − F₀)/F_m, where F_m is the maximal and F₀ is the minimal fluorescence. In analogy to F_m, P_m represents the maximal level of P700 signal obtained by a saturation pulse under far-red illumination, when P700 is fully oxidized (P700⁺). Y(I) = (P_m ^′ − P)/P_m, where P_m′ is the maximal level of P700 signal obtained by a saturation pulse at a defined AL illumination and P is the P700 oxidation level at a defined AL illumination (Klughammer and Schreiber, 1994; Klughammer and Schreiber, 2008). The length of the saturation pulse to determine Y(II) and Y(I) was 300 ms (with an intensity of ~10,000 μmol photons m⁻² s⁻¹). The intensity of ML for F₀ determination was <0.5 μmol photons m⁻² s⁻¹.

Differences in the response to acute heat stress within different species of Symbiodiniaceae were observed by recording the changes in photosynthetic parameters including FF, OJIP, PIFT, and photosystem activity at growth condition (24°C), heated (34°C for 2 h), and recovery (for 2 h). Another set of heat stress experiments was also performed at 24°C, 36°C, 38°C, and 40°C for 1 h to monitor the temperature-dependent changes in flash-induced chlorophyll fluorescence relaxation. Microaerobic condition was created by dark, incubating the sample with 7 U ml⁻¹ of glucose oxidase, 60 U ml⁻¹ of catalase, and 10 mM glucose, for 15 min. This microaerobic condition was complemented with the acute heat stress (for 10 min) and the changes in flash-induced chlorophyll fluorescence relaxation were recorded. The same procedure was applied for all Symbiodiniaceae strains.

Statistical analysis was performed using Origin Pro (Origin-Lab Corporation, Northampton, MA, USA). One-way analysis of variance (ANOVA) was performed on independent samples to detect statistically significant differences between treatments for which Tukey’s post-hoc multiple comparison tests (α = 0.05) were applied. Normality tests were performed using the Kolmogorov-Smirnov method and the homogeneity of variance test was performed using Levene’s method.

In order to reveal the heat-induced changes in the flash-induced chlorophyll fluorescence relaxation, different strains of Symbiodiniaceae were subjected to acute heat stress ( Figure 1 , upper panels).

Acute heat stress (at 34°C) caused a slight increase in the middle phase (0.003–0.1 s) of the fluorescence relaxation, in all investigated strains, but it was the most expressed in S. tridacnidorum 2465, although these changes were minor at this temperature. Increase in the middle phase corresponds to the elevated reduction of the PQ pool. This was confirmed by fluorescence induction measurements; OJIP traces ( Figure 1 , lower panels) showed that the largest increase in J phase (at ~1 ms, which indicates the reduction state of the PQ pool (Hill et al., 2004; Tóth et al., 2007b), occurred in S. tridacnidorum 2465, in agreement with the largest increase in the middle phase of the FF transient ( Figure 1 , upper panels) in S. tridacnidorum 2465. F. kawagutii CS156 displayed the smallest increase in the middle phase of the FF transient, in accordance with the least change in the J phase of the OJIP curve (interestingly, in F. kawagutii CS156, the J phase even decreased at elevated temperatures). The heat-induced changes in the flash-induced fluorescence relaxation traces were nearly completely reversible when the samples were incubated at growth temperature (24°C), in agreement with the decrease of the J phase of the OJIP curves, indicating that the PQ pool becomes more oxidized when the heat-induced changes were relieved.

Elevated heat also caused remarkable changes in the post-illumination Chl fluorescence rise. A transient Chl fluorescence rise occurs after the cessation of light due to the reduction of the PQ pool by stromal reductants; therefore, post-illumination Chl fluorescence rise is indicative of the redox kinetics of the PQ pool (Asada et al., 1993; Reynolds et al., 2008; Gotoh et al., 2010; Deák et al., 2014; Claquin et al., 2021), which showed strain-specific features in Symbiodiniaceae ( Figure 2 ).

The smallest response could be observed in F. kawagutii CS156, where acute heat stress did not remarkably alter the post-illumination Chl fluorescence characteristics ( Figure 2A ). In S. microadriaticum 2467, heat stress accelerated the post-illumination Chl fluorescence rise ( Figure 2C ), whereas S. tridacnidorum 2465 exhibited the largest response ( Figure 2B ). In S. tridacnidorum 2465, it was also observed that the post-illumination fluorescence rise was quite large already at 24°C in darkness, indicating a strong dark-induced reduction of the PQ pool (Hill and Ralph, 2008a). At 34°C, the post-illumination fluorescence rise occurred faster than at 24°C and then slowed down when the cells were re-incubated at 24°C.

In order to reveal the changes in the activity of both photosystems under the acute heat stress, PSII and PSI activities were measured ( Figure 3 ).

Acute heat stress caused no change in the activity of Photosystem I (Y(I)) in F. kawagutii CS156 ( Figure 3A ) and S. microadriaticum 2467 ( Figure 3C ), but it caused a slight elevation in Y(I) in S. tridacnidorum 2465; however, this was not statistically significant ( Figure 3B ). The activity of Photosystem II (Y(II)) decreased significantly in all strains, the largest decrease could be observed in S. tridacnidorum 2465. Y(II) recovered only partially under growth temperature in all the three strains.

To test the inducibility of the wave in fluorescence relaxation in Symbiodiniaceae, cells were exposed to acute heat stress treatment. As shown in Figure 1 , acute heat stress at 34°C caused an increase in the middle phase of flash-induced fluorescence relaxation as a result of reduced PQ pool; however, this condition did not induce the fluorescence wave formation. Therefore, it was investigated whether elevated temperatures (above 34°C) caused any further change in the flash-induced fluorescence relaxation ( Figure 4 ).

Heat treatment with higher temperatures caused marked changes in the flash-induced fluorescence relaxation profiles, and these changes were also strain specific. In F. kawagutii CS156, the flash fluorescence profiles did not change remarkably until 38°C; at 40°C, the middle phase slightly increased indicating the reduced PQ pool ( Figure 4A ). In S. microadriaticum 2467, the PQ pool became strongly reduced at 36°C and the reduction level increased further at 38°C; however, at 40°C, the fluorescence profile became similar to the profile at the growth temperature ( Figure 4C ). The reason of this is probably that the signal became very small at 40°C and, therefore, could not be properly measured. S. tridacnidorum 2465 exhibited reduced PQ pool at 36°C because of the elevated middle phase of fluorescence relaxation; however, increasing the temperature further caused a dip and then a slight increase in fluorescence, reminiscent of a wave phenomenon ( Figure 4B ). These results also indicate the remarkably different heat sensitivity of the different strains, with F. kawagutii CS156 being the least sensitive to elevated temperatures even up to 40°C.

It was investigated whether microaerobic conditions alone, which cause a strongly reduced PQ pool due to the depletion of O₂ which inhibits the terminal oxidases that keep the PQ pool oxidized (Deák et al., 2014; Ermakova et al., 2016), can induce the wave phenomenon in Symbiodiniaceae or not ( Figure 5 ).

Under microaerobic conditions, a pronounced reduction of the PQ pool could be observed, as indicated by the elevated middle phase of the fluorescence relaxation ( Figure 5 ) and by the increased F₀ ( Figure 5 inset). Microaerobic conditions caused similar changes across the different strains; however, the induction of the wave phenomenon did not occur in any of them, in agreement with other microalgae, in which microaerobic conditions were not sufficient either to induce the wave phenomenon (Krishna et al., 2019; Patil et al., 2022).

When microaerobic conditions were applied on heat-treated cells, the transient dip and subsequent rise in fluorescence were more expressed, particularly in S. tridacnidorum 2465 (at 38°C; Figure 6B ). This temperature resulted in the appearance of the wave phenomenon in F. kawagutii CS156 as well ( Figure 6A ); however, the largest amplitude could be observed in S. tridacnidorum 2465 in all cases, in which the fluorescence dip falls below the initial dark-adapted F₀ level ( Supplementary Figures S1–S3 , showing the results at 34°C, 36°C, and 40°C heat treatment). In S. microadriaticum 2467, the wave was not obvious; furthermore, a large decrease was observed in the longer timescales (> 10 s) probably due to cell clumping ( Figure 6C ). This was partially eliminated when 10% ficoll was applied in the cell suspension; however, cell clumping and sedimentation remain an issue in heat-treated samples and microaerobic condition ( Supplementary Figure S4 ). Nevertheless, it is clear that the transient dip and subsequent increase in fluorescence (a wave phenomenon) are the most expressed in S. tridacnidorum 2465.

Flash-induced chlorophyll fluorescence relaxation kinetics is a highly informative tool to monitor several electron transfer processes in cyanobacteria and microalgae: intersystem PSII-PSI electron flow as well as charge recombination and alternative electron transfer processes such as CEF (Vass et al., 1999; Deák et al., 2014; Volgusheva et al., 2016; Krishna et al., 2019; Havurinne and Tyystjärvi, 2020). Under standard conditions, the fluorescence relaxation displays kinetically well-separated and discernible phases. However, under specific or stress conditions, characteristic waves in the fluorescence relaxation could also be identified, which indicate the reduction-reoxidation sequences of the PQ pool. In cyanobacteria, this wave phenomenon was assigned to the operation of the NDH-1 complex, which mediates the electron transfer from ferredoxin at the acceptor side of PSI to the PQ pool (Deák et al., 2014). In microalgae such as Chlamydomonas reinhardtii, the wave phenomenon is related to the activity of NDH-2 (Krishna et al., 2019; Patil et al., 2022). It is important to note that the nature and the manifestation of the wave phenomenon are quite different in the different taxonomical groups. Whereas, in Synechocystis, microaerobic conditions are sufficient to induce the wave, in microalgae, usually, microaerobic treatment is not sufficient to its manifestation. This is due to the different stoichiometry of photosystems. When the PSI : PSII ratio is high, as in cyanobacteria, strong reduction of the PQ pool is sufficient to induce the wave because the electron withdrawal from the PQ pool by PSI exceeds the electron injection to the PQ pool from PSII. When the PSI : PSII ratio is lower, as in eukaryotic microalgae, the two processes (i.e., electron supply to and withdrawal from the PQ pool) are more balanced, which prevents the induction of the wave. As it was shown, if C. reinhardtii, partial inactivation of PSII, which increases the activity ratio of PSI : PSII, in combination with microaerobic treatment, was required to induce the large dip and subsequent rise of fluorescence after the flash (Patil et al., 2022). Therefore, the ratio of photosystems strongly influences the wave phenomenon, which therefore is an informative indicator of the balance of intersystem electron flow.

As heat stress is a severe problem for coral reefs, it is imperative to establish non-intrusive techniques to monitor heat-induced changes in the physiology and health of coral reefs organisms. Symbiodiniaceae species provide the essential photosynthetic products to the coral host to understand the regulation of photosynthetic electron transport processes of the endosymbiont algae. In our work, treating various Symbiodiniaceae strains with acute heat stress that was shown earlier to damage PSII (Hill and Ralph, 2008b), induce CEF (Aihara et al., 2016; Dang et al., 2019), or affect PSI activity and electron donation between PSII and PSI in intact corals (Szabó et al., 2017) caused some alterations in the fluorescence relaxation profile. This manifested itself in elevated middle phase of the fluorescence relaxation, which is related to enhanced reduction of the PQ pool. These changes were confirmed with fast fluorescence induction and post-illumination Chl fluorescence rise measurements, which showed that the strain S. tridacnidorum 2465 exhibiting the largest change in flash-induced fluorescence relaxation profiles displayed the largest (reversible) increase in the PQ pool reduction (largest post-illumination Chl fluorescence rise and largest increase in J phase of OJIP curves) upon acute heat stress. Conversely, the strain F. kawagutii CS156, which did not exhibit remarkable changes in flash-induced fluorescence relaxation, remained relatively unaltered in terms of the redox state of PQ pool. Therefore, the differences in heat sensitivity of the different species of Symbiodiniaceae as revealed by means of flash-induced fluorescence relaxation and accompanying Chl fluorescence measurements corroborated earlier studies, which identified different functional groups of Symbiodiniaceae (Hennige et al., 2009; Fisher et al., 2012; Suggett et al., 2015; Dang et al., 2019). However, it has to be noted that the acute heat treatment in the 34°C range was insufficient for the manifestation of the wave phenomenon in either Symbiodiniaceae strains investigated here.

Because the ratio of the activities of PSI : PSII is one of the crucial factors that determine the inducibility of the wave phenomenon, it was an important question to clarify whether the different strains display temperature dependent loss of PSII activity and if the decrease of PSII activity relative to PSI activity results in the wave formation in fluorescence relaxation. Increasing the temperature up to 40°C revealed that, although the wave formation did not occur in any of the investigated strains by heat treatment alone, the dip and the subsequent increase of fluorescence were the most apparent in S. tridacnidorum 2465 ( Figure 4B ). However, both strong reduction of the PQ pool and partial decrease in PSII activity are typically required in microalgae (C. reinhardtii) for the wave formation (Krishna et al., 2019; Patil et al., 2022). Based on the results of the present work, this is the case for Symbiodiniaceae as well; in heat-treated cells (which caused the loss of PSII activity relative to PSI activity), microaerobic treatment induced the wave formation, which was the most expressed in S. tridacnidorum 2465. The heat-induced changes in flash-induced fluorescence relaxation were also manifested in the relative decrease of variable fluorescence (the difference between maximum fluorescence, F_m(ST), and minimum fluorescence, F₀), increase in F₀, and changes in the fluorescence relaxation profile ( Figure 4 ). Elevated temperatures were shown to primarily impact oxygen evolving capacity and, thereby, PSII activity in corals, although the oxygen evolving complex was not found to be the primary site of damage of the photochemical apparatus during coral bleaching (Hill and Ralph, 2008b). Furthermore, in Symbiodiniaceae, PSII is more sensitive to heat than PSI ( Figure 3 ) (Hoogenboom et al., 2012); therefore, progressively increasing temperature causes progressively increasing imbalance in the electron flow from PSII that reduces the PQ pool and PQ pool oxidation via the forward electron transfer to PSI, which favors the formation of wave phenomenon. However, heat treatment itself up to 40°C was not sufficient to induce the wave; to observe this phenomenon, microaerobic condition was also crucial, which generates strong redox poise for the PQ pool. The tolerance of reef building corals to various stressors determine the ability of coral reef ecosystems to cope with acute and chronic changes in abiotic environmental conditions (Kühl et al., 1995; Ulstrup et al., 2005; Hughes et al., 2020). Although the temperatures applied in our work are typically well above the normal temperature conditions experienced by coral reefs, extreme environments also exist in which the temperature rises from 36°C to 38°C under which heat tolerant corals can survive. Despite the high temperature tolerance threshold of certain corals, they still remain highly susceptible to bleaching when, for example, peak summer temperature exceeds the upper threshold of their thermal tolerance [reviewed, e.g., in (Camp et al., 2018)]. Therefore, we suggest that, although the temperature treatment scenario in our work could not represent widespread or global environmental conditions, extreme climatic conditions similar to the temperatures applied in the current study may occur and, thus, represent naturally relevant scenarios in some cases. Therefore, the wave phenomenon of flash-induced chlorophyll fluorescence relaxation could be a valuable environmental indicator of heat stress in Symbiodiniaceae, which occurs in combination of hypoxic conditions in coral reef ecosystems. The temperature-dependent features of the formation of the wave phenomenon could decipher strain-specific tolerance to heat stress, although it has to be noted that the capacity for different alternative electron flow processes was not found to be unequivocally correlated with long-term heat tolerance in Symbiodiniaceae (Dang et al., 2019).

The formation of the wave was assigned to the operation of the components of CEF, such as the NDH complexes in cyanobacteria (Deák et al., 2014) and some eukaryotic microalgae (Krishna et al., 2019; Patil et al., 2022). Our observations that the wave formation occurred under conditions that was previously found to induce CEF in Symbiodiniaceae (i.e., acute heat stress) (Aihara et al., 2016; Dang et al., 2019) indicate that the wave phenomenon could be potentially assigned to CEF also in this species, although it has to be noted that the wave phenomenon occurred at higher temperatures as compared to the temperature that induced CEF in the earlier studies. The genome of certain strains of Symbiodiniaceae contains copies of both the PGR5/PGRL1 and NDH-2 (Aihara et al., 2016); therefore, it is possible that these pathways play a potential role in the wave formation in this genera; however, at this stage, no experimental evidence can be provided for the involvement of specific components or CEF pathways in the induction of the wave. Other alternative electron transfer pathways in Symbiodiniaceae such as Mehler reaction, PTOX, or FLV proteins (Roberty et al., 2014) could possibly contribute to the regulation of the PQ pool reduction state and the dip-rise kinetics of fluorescence relaxation. However, the wave phenomenon was observed to largest extent in microaerobic condition of heat-treated cells, under which photoreduction of oxygen by these alternative electron transport pathways is not functional. This argues for the involvement of CEF, which also accelerates under microaerobic conditions in Symbiodiniaceae (Aihara et al., 2016). However, similarly to C. reinhardtii, it is quite unlikely that the wave reflects an actual flash-induced cycling of electrons from the acceptor side of PSI to the PQ pool (Patil et al., 2022) because of the rise of the fluorescence after the dip phase occurs in several seconds and the re-reduction of P700⁺ via CEF occurs in ms timescale [e.g., (Alric, 2010)]. Furthermore, it is unlikely that the components like PGR5/PGRL1 mediate fast CEF pathways in Symbiodiniaceae, similar to recent findings in Chlamydomonas as it has been shown that the maximal rate of CEF was similar in the pgrl1 mutant and in wild-type Chlamydomonas (Nawrocki et al., 2019). However, the extent of CEF can be modulated from very low values to a maximal value by modulating the redox state of the donor and acceptor side of PSI (Nawrocki et al., 2019); therefore, it is plausible to investigate the relationship between the different kinetic phases in flash-induced fluorescence relaxation profiles and the kinetics of CEF and its regulatory components, which requires further studies in Symbiodiniaceae.

In conclusion, flash-induced fluorescence relaxation is a sensitive indicator of heat stress in Symbiodiniaceae. These algae displayed the so-called wave phenomenon when the PSII activity is partially decreased and the PQ pool becomes strongly reduced in microaerobic conditions. The wave phenomenon and heat sensitivity of the flash-induced fluorescence relaxation displayed species-specific features; therefore, this method is highly informative about the species-specific characteristics of the redox reactions of the PQ pool. The appearance of the wave phenomenon could possibly be related to CEF processes in Symbiodiniaceae; however, this remains to be investigated further. The flash-induced fluorescence relaxation is therefore a valuable non-intrusive marker of the stress-induced changes in photosynthetic electron transport and therefore could be used as a sensitive detection tool for coral bleaching.