Role of plastid protein phosphatase TAP38 in LHCII dephosphorylation and thylakoid electron flow

Abstract

Short-term changes in illumination elicit alterations in thylakoid protein phosphorylation and reorganization of the photosynthetic machinery. Phosphorylation of LHCII, the light-harvesting complex of photosystem II, facilitates its relocation to photosystem I and permits excitation energy redistribution between the photosystems (state transitions). The protein kinase STN7 is required for LHCII phosphorylation and state transitions in the flowering plant Arabidopsis thaliana. LHCII phosphorylation is reversible, but extensive efforts to identify the protein phosphatase(s) that dephosphorylate LHCII have been unsuccessful. Here, we show that the thylakoid-associated phosphatase TAP38 is required for LHCII dephosphorylation and for the transition from state 2 to state 1 in A. thaliana. In tap38 mutants, thylakoid electron flow is enhanced, resulting in more rapid growth under constant low-light regimes. TAP38 gene overexpression markedly decreases LHCII phosphorylation and inhibits state 1-->2 transition, thus mimicking the stn7 phenotype. Furthermore, the recombinant TAP38 protein is able, in an in vitro assay, to directly dephosphorylate LHCII. The dependence of LHCII dephosphorylation upon TAP38 dosage, together with the in vitro TAP38-mediated dephosphorylation of LHCII, suggests that TAP38 directly acts on LHCII. Although reversible phosphorylation of LHCII and state transitions are crucial for plant fitness under natural light conditions, LHCII hyperphosphorylation associated with an arrest of photosynthesis in state 2 due to inactivation of TAP38 improves photosynthetic performance and plant growth under state 2-favoring light conditions.

Figure 1. Comparison of the TAP38 sequence with those of related proteins from higher plants and moss.

The amino acid sequence of the Arabidopsis TAP38 protein (At4g27800) was compared with related sequences from Populus trichocarpa (POPTRDRAFT_250893), Oryza sativa (Os01g0552300), Picea sitchensis (GenBank: EF676359.1), and Physcomitrella patens (PHYPADRAFT_113608). Black boxes highlight strictly conserved amino acids, and gray boxes closely related ones. Amino acids that constitute the protein phosphatase 2C signature are indicated by asterisks. Putative chloroplast transit peptides (cTPs) are indicated in italics, and the potential transmembrane domain (TM) is highlighted.

Figure 2. Subcellular localization of TAP38.

(A) Full-length TAP38-RFP was transiently expressed in Arabidopsis protoplasts and visualized by fluorescence microscopy. Auto, chlorophyll autofluorescence; DIC, differential interference contrast image; merged, overlay of the two signals; RFP, fusion protein. Scale bar indicates 50 µm. (B) ³⁵S-labeled TAP38 protein, translated in vitro (lane 1, 10% translation product), was incubated with isolated chloroplasts (lane 2), which were subsequently treated with thermolysin to remove adhering precursor proteins (lane 3), prior to SDS-PAGE and autoradiography. m, mature protein; p, precursor. (C) Immunoblot analyses of proteins from WT and tap38-1 leaves. Equal protein amounts were loaded. Filters were immunolabeled with a TAP38-specific antibody. Chl, total chloroplasts; Str, stromal proteins; Thy, thylakoid proteins; Tot, total protein.

Figure 3. Expression of TAP38 in tap38 mutant, TAP38 overexpressor, and WT plants.

(A) Quantification of TAP38 mRNAs by real-time PCR in WT, tap38-1, tap38-2, and oeTAP38 leaves using the primer combination 1 and 2 (as in Figure S1A and S1B). (B) Thylakoid proteins from WT and tap38 mutants were loaded in the corresponding lanes. Reduced amounts of oeTAP38 thylakoids, corresponding to 25% of WT amount were loaded in the lane marked as 0.25× oeTAP38. Additionally, decreasing levels of WT thylakoids were loaded in the lanes indicated as 0.5× WT and 0.25× WT. Filters were immunolabeled with a TAP38-specific antibody raised against the mature TAP38 protein. (C) Thylakoid membranes of WT plants exposed to different light conditions (see Figure 5) were separated by SDS-PAGE. Immunodecoration of the corresponding Western blot was performed using a TAP38-specific antibody raised against the mature protein. A detail of a replicate gel, corresponding to the LHCII migration region, stained with Coomassie Blue is shown as loading control.

Figure 4. TAP38 is required for state transitions.

(A) Red light (R) and red light supplemented with far-red (FR) light were used to induce transitions to state 2 and state 1, respectively. F_M1 and F_M2 represent maximal chlorophyll fluorescence levels in states 1 and 2, respectively. Horizontal bars indicate the length of illumination. Arrows point to the moment when the specific light is switched on/off. Traces are the average of 10 replicates. ML, measuring light. (B) Low-temperature (77 K) fluorescence emission spectra of thylakoids were recorded after exposure of plants to light inducing either state 1 (dashed lines, far-red light of 740 nm) or state 2 (solid lines, low light; 80 µmol m⁻² s⁻¹) (see also Materials and Methods). The excitation wavelength was 475 nm, and spectra were normalized with reference to peak height at 685 nm. Traces are the average of 10 replicates.

Figure 5. Levels of LHCII phosphorylation correlate inversely with TAP38 concentrations.

Left panel, thylakoid proteins extracted from WT (A), tap38-1 (B), and oeTAP38 (C) plants kept in the dark (D; state 1), subsequently exposed to low light (LL; state 2), and then to far-red light for 30, 60, and 120 min (FR₃₀, FR₆₀, FR₁₂₀; state 1) were fractionated by SDS-PAGE. Phosphorylation of LHCII and PSII core proteins was detected by immunoblot analysis with a phosphothreonine-specific antibody. One out of three immunoblots for each genotype is shown. pCAS, phosphorylated CAS ; pCP43, phosphorylated CP43; pD1/D2, phosphorylated PSII-D1/D2; pLHCII, phosphorylated LHCII; Coomassie, portion of Coomassie-stained PA gels, identical to the ones blotted and corresponding to the LHCII migration region, were used as loading control. Right panel, thylakoid proteins of WT (A), tap38-1 (B), and oeTAP38 (C) plants treated as in the left panel were subjected to BN-PAGE analysis. Accumulation of the state 2-associated 670-kDa protein complex correlates with the phosphorylation level of LHCII. Note that tap38-2 behaved very similarly to tap38-1 (data not shown). One out of three BN-PAGEs for each genotype is shown.

Figure 6. Quantification of PSI-LHCI and PSI-LHCI-LHCII complexes under state 2 conditions.

(A) BN-PAGE of identical amounts of thylakoid proteins from WT, tap38-1, and oeTAP38 plants adapted to state 2 (low light; 80 µmol m⁻² s⁻¹). Bands representing the PSI-LHCI-LHCII (1) and PSI-LHCI (2) complexes are indicated. The differences in the separation behavior of the BN-gel in comparison to the ones in Figure 5 are caused by the longer electrophoresis running time. (B) The WT, tap38-1, and oeTAP38 lanes from the BN-PAGE in (A) were fractionated further by denaturing 2D-PAGE. Gels were stained with Coomassie Blue. LHCII, light-harvesting complex of PSII (the bands indicative for the PSI-LHCI-LHCII (1) and PSI-LHCI (2) complexes are encircled); P700, photosystem I reaction center. (C) Densitometric quantification of the spots representing PSI-LHCI-LHCII (spot 1) and PSI-LHCI (spot 2) in (B). Values are averages of three independent 2D gels for each genotype. Bars indicate standard deviations. Note that tap38-2 behaves very similarly to tap38-1 (data not shown).

Figure 7. Recombinant TAP38 directly dephosphorylates pLHCII in vitro.

(A) Equal amounts of pLHCII isolated from tap38-1 mutant plants, treated with or without recombinant TAP38, were separated by SDS-PAGE and immunodecorated with phosphothreonine-specific antibodies. NaF (10 mM) was added to specifically inhibit phosphatase activity. (B) A replicate gel of the samples as in (A) was stained with Coomassie Blue as a loading control. The recombinant TAP38 protein and LHCII bands are shown. (C) Densitometric quantification of the bands in (A), representing the phosphorylation levels of LHCII under the different conditions.

Figure 8. Growth characteristics and photosynthetic performance of tap38 mutant plants.

(A) Phenotypes of 4-wk-old tap38-1, tap38-2, and WT plants grown under low-light conditions (80 µmol m⁻²s⁻¹) on a 12 h/12 h light/dark regime. (B) Growth curve. Leaf areas of 20 plants of each genotype (WT, grey bars; tap83-1, white bars; tap38-2, light grey bars) were measured over a period of 4 wk after germination. Mean values ± standard deviations (SDs; bars) are shown. (C and D), Measurements of light dependence of the photosynthetic parameter 1-qP (C) and effective quantum yield of PSII (ΦII; [D]) of plants grown as in (A). WT, filled grey circles; tap38-1, open circles; tap38-2, filled light-grey circles; oeTAP38, filled black circles; PAR, photosynthetically active radiation in µmol m⁻² s⁻¹. Average values were determined from five independent measurements (SD<5%).