The Dynamics of Energy Dissipation and Xanthophyll Conversion in Arabidopsis Indicate an Indirect Photoprotective Role of Zeaxanthin in Slowly Inducible and Relaxing Components of Non-photochemical Quenching of Excitation Energy

The dynamics of non-photochemical quenching (NPQ) of chlorophyll fluorescence and the dynamics of xanthophyll conversion under different actinic light conditions were studied in intact leaves of Arabidopsis thaliana. NPQ induction was investigated during up to 180 min illumination at 450, 900, and 1,800 μmol photons m−2 s−1 (μE) and NPQ relaxation after 5, 30, 90, or 180 min of pre-illumination at the same light intensities. The comparison of wild-type plants with mutants affected either in xanthophyll conversion (npq1 and npq2) or PsbS expression (npq4 and L17) or lumen acidification (pgr1) indicated that NPQ states with similar, but not identical characteristics are induced at longer time range (15–60 min) in wild-type and mutant plants. In genotypes with an active xanthophyll conversion, the dynamics of two slowly (10–60 min) inducible and relaxing NPQ components were found to be kinetically correlated with zeaxanthin formation and epoxidation, respectively. However, the extent of NPQ was independent of the amount of zeaxanthin, since higher NPQ values were inducible with increasing actinic light intensities without pronounced changes in the zeaxanthin amount. These data support an indirect role of zeaxanthin in pH-independent NPQ states rather than a specific direct function of zeaxanthin as quencher in long-lasting NPQ processes. Such an indirect function might be related to an allosteric regulation of NPQ processes by zeaxanthin (e.g., through interaction of zeaxanthin at the surface of proteins) or a general photoprotective function of zeaxanthin in the lipid phase of the membrane (e.g., by modulation of the membrane fluidity or by acting as antioxidant). The found concomitant down-regulation of zeaxanthin epoxidation and recovery of photosystem II activity ensures that zeaxanthin is retained in the thylakoid membrane as long as photosystem II activity is inhibited or down-regulated. This regulation supports the view that zeaxanthin can be considered as a kind of light stress memory in chloroplasts, allowing a rapid reactivation of photoprotective NPQ processes in case of recurrent light stress periods.

NPQ dynamics. The induction of NPQ during 180 min of illumination at three different actinic light intensities (450 µE, 900 µE and 1800 µE of white light), and the relaxation of NPQ after pre-illumination for 5 min, 30 min, 90 min and 180 min at each actinic light intensity were determined for (A-C) WT, (D-F) pgr1, (G-I) L17, (J-L) npq4, (M-O), npq2 and (P-R) npq1 plants. During the whole measurements, detached leaves were placed on wet paper in a temperaturecontrolled cuvette (20°C) under permanent supply with ambient air. Mean values of independent 3-6 measurements are shown. The synthesis of Zx during 180 min of illumination at three different actinic  light intensities (450 µE, 900 µE and 1800 µE of white light), and the epoxidation of Zx after different times of pre-illumination at each actinic light intensity were determined for (A-C) WT, (D-F) pgr1, (G-I) L17, (J-L) npq4. For pgr1 plants, Zx epoxidation is shown only for the longer preillumination times of 90 and 180 min, because only very low amounts of Zx were present at the shorter pre-illumination times of 5 and 30 min. During the whole measurements, detached leaves were placed on wet paper in a temperature-controlled cuvette (20°C) under permanent supply with ambient air. Mean values of independent measurements +/-SE are shown. 3-6

Figure S3 Correlation analysis of NPQ and Zx dynamics. (A, B) NPQ induction and Zx synthesis.
(C, D) NPQ relaxation and Zx epoxidation. The data show the results from the experiment with WT plants upon illumination at 900 µE. The data for NPQ induction and Zx synthesis (A, B) were taken from Figure 4B, the NPQ relaxtion and Zx epoxidation (C, D) was analyzed after pre-illumination for 90 min ( Figure 8B). The left side (A, C) shows the data for the total NPQ amplitude, while the data related to the qE component were omitted on the right side (B, D). The determined Pearson's correlation coefficient r is indicated in each panel.

Figure S4 Comparison of NPQ relaxation and Zx epoxidation after 5 min of pre-illumination.
The time course of NPQ relaxation and Zx epoxidation after 5 min of pre-illumination at the three actinic light intensities of 450 µE, 900 µE and 1800 µE is compared for: (A-C) WT, (D-F) npq4, and (G-I) L17. No data for pgr1 plants were included, because only very low amounts of Zx were present after 5 min of pre-illumination. The data were taken from Figure     No data for pgr1 plants were included, because only low amounts of Zx were present after 30 min of pre-illumination. For direct comparison, the data for Zx epoxidation were fitted to match the amplitudes of the slowly relaxing (> 2 min) NPQ components, only. The data were taken from Figure  6 (NPQ) and Figure 7 (Zx). The dashed lines in panels A-C and G-I indicate the NPQ amplitudes after relaxation of qE. The determined Pearson's correlation coefficient r is indicated in each panel.

Figure S9
Comparison of the kinetics of NPQ relaxation and Zx epoxidation after 180 min of pre-illumination. The data for NPQ relaxation (open circles) and Zx epoxidation (filled circles) after 180 min of pre-illumination at the three actinic light intensities of 450 µE, 900 µE and 1800 µE are compared for the four genotypes with an active xanthophyll cycle: (A-C) WT, (D-F) pgr1, (G-I) npq4 and (J-L) L17. For direct comparison, the data for Zx epoxidation were fitted to match the amplitudes of the slowly relaxing (> 2 min) NPQ components, only. The data were taken from Figure  6 (NPQ) and Figure 7 (Zx). The dashed lines in panels A-C and J-L indicate the NPQ amplitudes after relaxation of qE. The determined Pearson's correlation coefficient r is indicated in each panel.

Figure S10 Comparison of NPQ induction upon illumination with white or red actinic light.
The data for NPQ induction with white light (WL, white symbols) were taken from Figure 1. NPQ induction with red light (620 nm, RL, red symbols) was measured with the DUAL PAM 100 (Walz, Effeltrich, Germany). Actinic light intensities are indicated in the Figure. Note, that about 20% lower intensities were used for RL. The data represent mean values (± SD) of 3-5 independent measurements.