Zeaxanthin has enhanced antioxidant capacity with respect to all other xanthophylls in Arabidopsis leaves and functions independent of binding to PSII antennae

Abstract

The ch1 mutant of Arabidopsis (Arabidopsis thaliana) lacks chlorophyll (Chl) b. Leaves of this mutant are devoid of photosystem II (PSII) Chl-protein antenna complexes and have a very low capacity of nonphotochemical quenching (NPQ) of Chl fluorescence. Lhcb5 was the only PSII antenna protein that accumulated to a significant level in ch1 mutant leaves, but the apoprotein did not assemble in vivo with Chls to form a functional antenna. The abundance of Lhca proteins was also reduced to approximately 20% of the wild-type level. ch1 was crossed with various xanthophyll mutants to analyze the antioxidant activity of carotenoids unbound to PSII antenna. Suppression of zeaxanthin by crossing ch1 with npq1 resulted in oxidative stress in high light, while removing other xanthophylls or the PSII protein PsbS had no such effect. The tocopherol-deficient ch1 vte1 double mutant was as sensitive to high light as ch1 npq1, and the triple mutant ch1 npq1 vte1 exhibited an extreme sensitivity to photooxidative stress, indicating that zeaxanthin and tocopherols have cumulative effects. Conversely, constitutive accumulation of zeaxanthin in the ch1 npq2 double mutant led to an increased phototolerance relative to ch1. Comparison of ch1 npq2 with another zeaxanthin-accumulating mutant (ch1 lut2) that lacks lutein suggests that protection of polyunsaturated lipids by zeaxanthin is enhanced when lutein is also present. During photooxidative stress, alpha-tocopherol noticeably decreased in ch1 npq1 and increased in ch1 npq2 relative to ch1, suggesting protection of vitamin E by high zeaxanthin levels. Our results indicate that the antioxidant activity of zeaxanthin, distinct from NPQ, can occur in the absence of PSII light-harvesting complexes. The capacity of zeaxanthin to protect thylakoid membrane lipids is comparable to that of vitamin E but noticeably higher than that of all other xanthophylls of Arabidopsis leaves.

Figure 1.

A, Control wild-type and ch1 Arabidopsis plants showing the pale-green phenotype of the mutant. B, Western blots of Lhcb and Lhca proteins in thylakoids of wild-type and ch1 Arabidopsis. The internal PSII antenna CP47 was also analyzed for normalization purposes. Amounts of Chl loaded on the gel: 0.5, 1, 2, and 4 μg. The blots were repeated three times with similar results. C, Abundance of Lhca and Lhcb proteins normalized to CP47 in wild-type and ch1 thylakoids. Data are mean values of three separate measurements ± sd. [See online article for color version of this figure.]

Figure 2.

A, Nondenaturing Deriphat-PAGE separation of pigmented thylakoid complexes of wild-type and ch1 Arabidopsis. Twenty-five micrograms of Chl/lane were loaded on the gel. B, Tris-sulfate SDS-PAGE of the various green bands eluted from the Deriphat gels.

Figure 3.

A to C, CD spectra of the B2 or B2' band of wild-type and ch1 thylakoids compared with Lhca1 (A), Lhca3 (B), and Lhcb5 (C) recombinant proteins reconstituted with Chl a. Because reconstitution of Lhcb5 antenna is impossible in the complete absence of Chl b, a small amount of Chl b was added to Chl a with a Chl a:b ratio of 8. D, Absorption spectra of the B2 and B2' bands eluted from Deriphat-PAGE gels of wild-type and ch1 thylakoids, respectively (see Fig. 2). The spectra were measured on three different preparations with qualitatively similar results.

Figure 4.

A, Western blot of Lhcb5 in wild-type, lhcb5, ch1, and ch1 lhcb5 leaves. One microgram of Chl was loaded on the gel. Coomassie Blue-stained gel was used as loading control. B, Nondenaturing Deriphat-PAGE separation of pigments photosynthetic complexes of wild-type, lhcb5, ch1, and ch1 lhcb5 thylakoids. Twenty-five micrograms of Chl/lane were loaded. C, Induction curves of Chl fluorescence in leaves infiltrated with 30 μm diuron. Experiments shown in A to C were repeated two times with similar results.

Figure 5.

Pigmented photosynthetic complexes, carotenoids, and NPQ in the ch1 Arabidopsis mutant and in the double mutants ch1 npq1, ch1 npq2, ch1 npq4, and ch1 lut2. A, Nondenaturing gel profiles of thylakoids from the wild type, ch1, and the double mutants. This experiment was repeated three times with similar results. Twenty-five micrograms of Chl/lane were loaded. The different bands were identified by SDS-PAGE, HPLC, and spectrophotometric analyses of excised bands (see text). Lack of lutein in ch1 lut2 appeared to increase the stability of LHCI, which remained attached to the PSI core (no monomeric Lhca band). B, Carotenoid concentrations in the leaves of the different mutants. Plants were exposed to high-light stress (1,300 μmol m⁻² s⁻¹, 8°C, 2 d) before pigment analyses to allow conversion of violaxanthin (Vio) into antheraxanthin (Anthera) and zeaxanthin (Zea). Lut, Lutein; car, β-carotene; Neo, neoxanthin. Data are mean values of four or five measurements + sd. C, NPQ measured in unstressed leaves of the different mutants at different PFDs. For clarity, the NPQ curve of ch1 lut2 (intermediate between that of ch1 and ch1 npq2) is not shown. Data are mean values of three separate experiments ± sd.

Figure 6.

A, Wild-type Arabidopsis and ch1 single and double mutants after high-light stress (1,300 μmol m⁻² s⁻¹, 8°C, 2 d). B, Leaf bleaching was estimated from the decrease in Chl a concentration after high light. C, PSII photoinhibition was measured by the F_v/F_m Chl fluorescence ratio. Data are mean values of three to six measurements (B) or 8 to 12 measurements (C) + sd. Statistical analysis of the differences between ch1 npq2 and ch1 or ch1 lut2: a and c, significantly different with P < 0.05; b, significantly different with P < 0.10 (t test). ch1 npq1 is significantly different from all other mutants for both parameters (P < 0.05, t test), while ch1 npq4 was not significantly different from ch1.

Figure 7.

Lipid peroxidation and photooxidative stress in the ch1 single and double mutants exposed to high-light stress (1,300 μmol m⁻² s⁻¹, 8°C, 2 d). A, Autoluminescence imaging of oxidative stress in the different mutants after high-light stress. The experiments were repeated four times with similar results. Lipid peroxidation was also estimated by thermoluminescence (B) and MDA concentration (C). Data are mean values of three to six measurements + sd. Statistical analysis of the differences between ch1 npq2 and ch1 or ch1 lut2: a and b, significantly different with P < 0.05 and 0.10, respectively (t test). ch1 npq1 is significantly different from all other mutants for both parameters (P < 0.05, t test), while ch1 npq4 was not significantly different from ch1.

Figure 8.

A, Oxidative stress and lipid peroxidation in ch1 lhcb5 and ch1 npq1 lhcb5 compared to ch1 and ch1 npq1 after high-light stress (1,300 μmol m⁻² s⁻¹, 8°C, 2 d). B, Autoluminescence imaging of lipid peroxidation. C, High-temperature thermoluminescence (amplitude of the 135°C TL band). The experiment was done three times with similar results.

Figure 9.

PSII photoinhibition in ch1, ch1 npq1, and ch1 npq4 leaf discs exposed to high-light stress at low temperature (600 or 1,700 μmol photons m⁻² s⁻¹ and 8°C). Photoinhibition was measured by the decrease in the Chl fluorescence parameter F_v/F_m. Black circles, ch1; white circles, ch1 npq1; white triangles, ch1 npq4. Data are mean values of three separate experiments ± sd.

Figure 10.

Responses of ch1 vte1 and ch1 vte1 npq1 in comparison with ch1 and ch1 npq1 to high-light stress (1,300 μmol m⁻² s⁻¹, 8°C, 2 d). A, Picture of the different mutants after high-light stress. B, Autoluminescence imaging. C, Lipid peroxidation measured by the amplitude of the thermoluminescence band at 135°C. D, PSII photoinhibition measured by the decrease in the F_v/F_m Chl fluorescence ratio. Mature, well-developed leaves were selected for the fluorescence measurements. A and B, The experiment was repeated three times with similar results; C and D, data are mean values of three (C) or five to 10 (D) measurements + sd.

Figure 11.

α-Tocopherol level in leaves of ch1 single and double mutants before (in black) and after (in white) high-light stress at low temperature (1,300 μmol photon m⁻² s⁻¹, 8°C, 2 d). Data are mean values of four measurements + sd.