Synergistic effects of temperature and light on photoprotection in the model diatom Phaeodactylum tricornutum

Abstract

Diatoms dominate phytoplankton communities in turbulent waters, where light fluctuations can be frequent and intense. Due to this complex environment, these heterokont microalgae display remarkable photoprotection strategies, including a fast Non-Photochemical Quenching (NPQ). However, in nature, several abiotic parameters (such as temperature) can influence the response of photosynthetic organisms to light stress in a synergistic or antagonistic manner. Yet, the combined effects of light and these other drivers on the photosynthetic and photoprotective capacity of diatoms are still poorly understood. In this work, we investigated the impact of short-term temperature and light stress on the model diatom Phaeodactylum tricornutum, combining NPQ induction-recovery assays or light curves with a broad gradient of superimposed temperature treatments (5 to 35°C). We employed mutant lines deficient in NPQ generation (vde KO) or recovery (zep3 KO) and wild type. We found that temperature and light have a synergistic effect: lower temperatures limited both the photosynthetic capacity and NPQ, while the general photophysiological performance was enhanced with warming, up to a heat-stress limit (above 30°C). We discuss the temperature effects on NPQ induction and recovery and propose that these are independent from the energy requirements of the cells and result from altered xanthophyll cycle dynamics. Namely, we found that de-epoxidation activity strongly increases with temperature, outweighing epoxidation and resulting in a positive increase of NPQ with temperature. Finally, we propose that in a short-term time frame, temperature and light have a synergistic and not antagonistic effect, with a positive relationship between increasing temperature and NPQ.

FIGURE 1

YII and NPQ of vde KO , wt and zep3 KO during NPQ induction‐recovery assays performed under different temperature stress (average ± sd, n = 3). Samples were exposed for 30 min to low light (28 μmol photons m⁻² s⁻¹, highlighted in gray) to the corresponding temperature stress, followed by 5 min of NPQ induction (500 μmol photons m⁻² s⁻¹) and 15 min of recovery (28 μmol photons m⁻² s⁻¹, highlighted in gray) at the same temperature. On the x‐axis, time 0 indicates the beginning of the induction phase; an axis break, indicated by a double dash (//), was inserted during the first 30 min of temperature treatment (i.e., between −30 and 0 min). Results from the corresponding statistical analysis are presented in Table S2.

FIGURE 2

YII and NPQ of vde KO , wt and zep3 KO during NPQ induction‐recovery assays performed with a temperature switch between induction (I) and recovery (R) (average ± sd, n = 3). Samples were exposed for 30 min to low light (28 μmol photons m⁻² s⁻¹, highlighted in gray) followed by 5 min of induction (500 μmol photons m⁻² s⁻¹) at the corresponding induction temperature (indicated on top of the panels) and 15 min of recovery (28 μmol photons m⁻² s⁻¹, highlighted in gray) at the corresponding recovery temperature (indicated on top of the panels). Temperature was switched at the end of the induction phase, indicated by a black arrow (↓). On the x‐axis, time 0 indicates the beginning of the induction phase; an axis break, indicated by a double dash (//), was inserted during the first 30 min of temperature treatment (i.e., between −30 and 0 min).

FIGURE 3

Content of diadinoxanthin cycle pigments in wt exposed to different temperatures during NPQ induction‐recovery assays (average ± sd, n = 3). Pigment content is expressed as pigment:chlorophyll a ratio (mol/mol). Samples were exposed for 30 min in low light (28 μmol photons m⁻² s⁻¹, highlighted in gray) to the corresponding temperature, following 5 min of induction (500 μmol photons m⁻² s⁻¹) and 15 min of recovery (28 μmol photons m⁻² s⁻¹, highlighted in gray) under the same temperature. On the x‐axis, time 0 indicates the beginning of the induction phase; an axis break, indicated by a double dash (//), was inserted during the first 30 min of temperature treatment (i.e., between −30 and 0 min). (A) diadinoxanthin (Dd); (B) diatoxanthin (Dt); (C) total diadinoxanthin cycle pool (Dd + Dt). Results from the corresponding statistical analysis are presented in Table S3.

FIGURE 4

YII and NPQ development over time in vde KO , wt and zep3 KO during light curve experiments performed under different temperature stress (average ± sd, n = 3). Samples were exposed for 30 min in low light (28 μmol photons m⁻² s⁻¹, highlighted in gray) to the corresponding temperature stress, followed by 12 light curve steps (12–645 μmol photons m⁻² s⁻¹) and 7.5 min of recovery (28 μmol photons m⁻² s⁻¹, highlighted in gray) under the same temperature. On the x‐axis, time 0 indicates the beginning of the induction phase; an axis break, indicated by a double dash (//), was inserted during the first 30 min of temperature treatment (i.e., between −30 and 0 min). Results from the corresponding statistical analysis are presented in Table S4.

FIGURE 5

Fitted rETR and NPQ versus PAR curves of vde KO , wt and zep3 KO during light curve experiments performed under different temperature stress. After 30 min of temperature treatment (indicated on top of the panels) in low light (28 μmol photons m⁻² s⁻¹), samples were treated with 12 increasing steps of actinic light (12–645 μmol photons m⁻² s⁻¹) under the same temperature. (A) Curves fit. Hollow points represent average values (± sd, n = 3), while solid lines represent the curve fit obtained from all single replicate values (n = 3). Absence of line indicates that fitting was not possible. p‐values (p) indicate the statistical significance of differences between fitted curves at a given temperature and the reference (20°C), according to pairwise t‐test. Differences in rETR were not tested as this parameter carries a.u. (B) rETR curve parameters: Ek and Eopt. (C) NPQ curve parameters: NPQmax and E50; vde KO was excluded from this analysis as these lines always displayed values of NPQ close to 0, determining very poor model fitting. Points represent average values (± sd, n = 3) of curves fitted separately for each replicate. For each time point and within each culture, asterisks indicate the statistical significance of each temperature condition tested against the control (20°C), according to adjusted p‐value of multiple comparison t‐test: *: p < 0.05; **: p < 0.005, *** p < 0.005.

FIGURE 6

Under short time frames, temperature and light have a synergistic effect on NPQ in P. tricornutum . Summary heatmap with the results of our light curve experiment (average NPQ, n = 3). On the × axis, each line represents a temperature treatment. On the y axis, each column represents a single sample (ordered according to sampling time); corresponding light intensity (PAR) is plotted below the × axis.