Mesophyll-specific phytochromes impact chlorophyll light-harvesting complexes (LHCs) and non-photochemical quenching

ABSTRACT

Phytochromes regulate light-dependent plastid development and plant growth and development. Prior analyses demonstrated that phytochromes regulate expression of Sigma factor 2 (SIG2), which is involved in plastid transcription and coordinates expression of plastid‐ and nuclear‐encoded genes involved in plastid development, as well as plant growth and development. Mutation of SIG2 impacts distinct aspects of photosynthesis, resulting in elevated levels of cyclic electron flow and nonphotochemical quenching (NPQ). As we initially identified SIG2 expression as misregulated in a line lacking phytochromes in mesophyll tissues (i.e., CAB3::pBVR lines), here we report on an investigation of whether photosynthetic parameters such as NPQ are also disrupted in CAB3::pBVR lines. We determined that a specific parameter of NPQ, i.e., energy-dependent quenching (qE) which is a rapidly induced photoprotective mechanism that dissipates stressful absorption of excess light energy during photosynthesis, is disrupted when mesophyll phytochromes are significantly depleted. The observed reduction in NPQ levels in strong CAB3::pBVR lines is associated with a reduction in the accumulation of Lhcb1 proteins and assembly or stability of light-harvesting complexes (LHCs), especially trimeric LHC. These results implicate mesophyll-localized phytochromes in a specific aspect of phytochrome-mediated NPQ, likely through regulation of chlorophyll synthesis and accumulation and the associated impacts on chlorophyll–protein complexes. This role is distinct from the impact of mesophyll phytochrome-dependent control of SIG2 and associated NPQ regulation.

KEYWORDS:

• Chlorophyll–protein complexes
• light-harvesting complexes (LHC)
• light signaling
• non-photochemical
• quenching
• phytochrome
• photomorphogenesis
• photosynthesis
• stress

Introduction

Plants have evolved mechanisms to cope with variations in environmental parameters to support successful growth and development in the environments in which they exist. Although light is vital for plants, excessive light beyond the capacity utilized for growth and development can cause photooxidative damage, resulting in low photosynthetic efficiency.¹ One mechanism for coping with excessive light is non-photochemical quenching (NPQ) of chlorophyll fluorescence in photosystem II (PSII). NPQ is involved in the relaxation of excited chlorophylls by dissipating excess absorbed energy as heat. Since excitation energy of chlorophylls can be re-emitted as fluorescence, NPQ can be quantified by measuring changes in chlorophyll fluorescence due to dissipation of energy through heat release.^1,2 The major component of NPQ is pH- or energy-dependent quenching (qE). qE requires a low pH in the thylakoid lumen, which is generated by electron transport during photosynthesis.³ Another component of NPQ functioning in plants is the photoinhibitory quenching of chlorophyll (qI), caused by photoinhibition of PSII photochemistry.¹ Unlike the fast qE response of NPQ, qI shows slow relaxation of excited chlorophylls.

In Arabidopsis thaliana, several genes related to NPQ have been identified through genetic screening. npq1 mutants lack zeaxanthin and exhibit low levels of NPQ, primarily low qE.⁴ The activation of violaxanthin de-epoxidase (VDE), which is encoded by NPQ1, by low pH status in the thylakoid lumen purportedly contributes to qE in response to excess light.^1,5 NPQ4 encodes the PsbS protein, which is a member of the light-harvesting complex (LHC) protein superfamily and thylakoid membrane integral protein component of PSII. Mutation of NPQ4 causes a lack of qE, suggesting a necessary role of PSII proteins in NPQ.⁶ Protonation of PsbS could initiate qE by a conformational change of the PsbS protein itself or adjacent LHC proteins.⁶ An NPQ state promotes light-dependent interaction of monomeric PsbS with Lhcb1 proteins in LHCII trimers.^7,8 Photosystem I (PSI) proteins also appear to be involved in NPQ. PSI-D, a subunit of PSI, plays an important role in the docking of ferredoxin to PSI. Anti-sense plants of the PSI-D gene exhibit high levels of NPQ along with high levels of accumulation of both VDE and PsbS proteins.⁹

In plants, photoreceptors mediate a number of light responses, including de-etiolation, shade avoidance, stomatal opening, chloroplast development and positioning, and flowering.¹⁰ Several photoreceptors are involved in light stress acclimation to high light. For example, a blue light flavoprotein receptor, phototropin (PHOT2), is important for photoprotective chloroplast positioning.^11,12 In high fluence rates of blue light, phot2 mutants do not show chloroplast avoidance, resulting in low photosynthetic efficiency and photobleaching.¹³ Phytochromes are photoreceptors composed of apoproteins bound to linear tetrapyrrole chromophores, phytochromobilins, and are mainly responsible for red/far-red light responses.¹⁰ Analyses of phytochrome chromophore-deficient mutants have demonstrated that phytochromes are important for maintaining chloroplast number and development as these mutants have reduced numbers of chloroplasts, reduced chlorophyll levels, increased chl a/chl b ratios, and reduced granal stacking in plastids.^14-17 Specific aspects of photosynthesis are also impaired in phytochrome synthesis mutants. High light-grown (400 µmol m⁻² s⁻¹) phytochrome A mutants (phyA-201) exhibit low levels of NPQ¹⁸ and hy1 mutants (hy1-1), which are defective in phytochromobilin chromophore biosynthesis, also show low levels of NPQ,^18,19 suggesting that phytochromes play either direct or indirect roles in regulating NPQ to allow plants to cope with excess absorbed light energy in high light conditions.

Transgenic plants expressing a BILIVERDIN REDUCTASE (BVR) gene, which encodes an enzyme that results in phytochrome chromophore depletion, under the control of tissue-specific promoters have been used for probing tissue- or organ-specific roles of phytochromes.^20-24 For example, mesophyll-specific inactivation of phytochrome is achieved by expression of BVR in plastids of mesophyll tissues of Arabidopsis by using the CAB3 promoter (i.e., CAB3::pBVR). Young seedlings of CAB3::pBVR plants exhibit disruptions in light-dependent growth and developmental responses, including the inhibition of hypocotyl elongation and the stimulation of anthocyanin accumulation.²⁰ In far-red grown CAB3::pBVR seedling lines, both the large and small subunits of Rubisco (RbcL and RbcS) are downregulated at the transcript and protein levels, which indicated specific roles for mesophyll-specific phytochromes in the regulation of photosynthetic factors.²⁴ Mesophyll phytochromes regulate plastid-localized sigma factor 2 (SIG2), which in turn controls the expression of RBCS and LHCB1.2.²⁵ Recent studies of sig2 mutants indicate that these plants have altered photosynthetic parameters, including reduced ΦII, increased NPQ (both qE and qI components), among other altered photosynthetic parameters.²⁶

Given the disruptions in photosynthetic parameters in sig2 and mesophyll phytochrome-dependent regulation of SIG2 expression,^26-28 we measured NPQ and other select photosynthetic parameters in mesophyll-specific phytochrome inactivation lines (i.e., CAB3::pBVR). Since mesophyll tissues are most abundant in leaves, we used fully expanded rosette leaves from white light-grown mature plants for these analyses. CAB3::pBVR lines showed reduced levels of NPQ (namely qE), reduced linear electron flow (LEF), high ratios of chl a/chl b, reduced production of Lhcb1 protein, and defects in the association of chlorophyll with Lhcb1 (especially impacting the accumulation of LHCII trimers). However, the expression of several genes responsible for NPQ (e.g., NPQ1, NPQ4, LHCB1.1, LHCB1.2, and LHCB1.3) was not altered in CAB3::pBVR lines. We proposed that phytochromes in leaves contribute to NPQ by regulating the association of chlorophyll with its partner molecules, which is distinct from the role of phytochrome-dependent regulation of SIG2, which distinctly impacts NPQ and other aspects of photosynthesis through controlling plastid transcription.

Materials and methods

Plant material and growth conditions

No-0 wild-type (WT), 35S::pBVR3, CAB3::pBVR1, CAB3::pBVR2, and CAB3::pBVR3 lines were previously described.^20,29 Plants were grown at 22°C under white light (approximately 100 µmol m⁻² s⁻¹) with a long-day photoperiod (16-h-light and 8-h-dark cycle) for 30 or 45 d, as indicated.

BVR enzyme assays

BVR-specific enzyme activity of soluble protein from rosette leaves of plants was measured spectrophotometrically as described previously.³⁰

Fluorescence imaging and quantification

For qualitative measurement of NPQ, fluorescence images have been obtained using a MAXI version of the IMAGING-PAM M-series chlorophyll fluorescence system (Heinz-Walz Instruments, Germany) coupled with analytical software IMAGING-WIN as described previously,³¹ with modifications described as follows. An AVT Dolphin camera (Allied Vision Technologies) was used in the system. During a 3-min actinic light treatment (531 µmol m⁻² s⁻¹), saturation pulses (2800 µmol m⁻² s⁻¹) at 30 s intervals were applied. In IMAGING-WIN software, NPQ/4 was selected as an image type to generate false-color images of NPQ.

For quantitative measurement of NPQ, light-dependent changes in fluorescence were measured as previously described.^32,33 In detail, an attached, fully expanded rosette leaf (10th or 11th leaf) from 30-d-old plants dark-adapted overnight was used for experiments. NPQ was calculated using the following equation: (Fm-Fm')/Fm', where Fm is maximal fluorescence value from dark-adapted sample and Fm' is the maximal fluorescence value of a light-adapted sample after a saturating pulse (>7,000 µmol m⁻² s⁻¹). For calculation of qE and qI components of NPQ, the maximal fluorescence value after switching off the actinic light (Fm″) was measured. The qE and qI were calculated as (Fm″-Fm′)/Fm′ and (Fm-Fm″)/Fm″, respectively, as described previously.² LEF and maximal quantum yield of PSII (Fv/Fm) were calculated as described.³⁴

Chlorophyll analysis

Extraction of chlorophyll with N,N-dimethylformamide (DMF) was performed using 10 mg of rosette leaves from 30-d-old plants as described.³⁵ The concentration of chlorophyll was calculated according to Inskeep and Bloom.³⁶

SDS-PAGE gel, native green gel, and immunoblot analysis

For SDS-polyacrylamide gel electrophoresis, total soluble proteins were extracted from leaves of 30-d-old plants using 2 × protein sample buffer, which contained 8% (v/v) glycerol, 50 mM Tris-HCl pH 6.8, 2% (v/v) SDS, 0.008% (v/v) bromophenol blue, and 4% (w/v) beta-mercaptoethanol. Proteins were resolved on 15% SDS-PAGE gels. For native green gel electrophoresis, thylakoid membrane-enriched samples were prepared as described in Schwarz,³⁷ with minor modifications. All procedures were performed at 4°C. Leaves were homogenized in ice-cold extraction buffer, containing 50 mM Tris-HCl pH 7.4, 10 mM MgCl₂, and 5 mM KCl. The resulting lysates were filtered through four layers of Kimwipes (Kimberly-Clark) and purified on 80% (v/v) Percoll cushion (Sigma) with centrifugation at 13,000 rpm for 1 h. Thylakoid membrane-enriched samples were dissolved in extraction buffer containing 10% (v/v) glycerol, 2% (v/v) decyl maltoside, and 2% (v/v) octyl glucoside (Sigma). As described byAllen and Staehelin, 8% green gels were prepared.³⁸ Immunoblot analyses for SDS-PAGE and native green gels were performed as described in Lagarias et al.,³⁰ using anti-Lhcb1 antibody (cat. No: AS01 004, Agrisera).

RT-PCR analysis and heatmap

For RT-PCR analysis, total RNA was extracted from leaves using RNeasy® Plant Minikit (Qiagen, CA), and oligo(dT)₁₅ primed-first-strand cDNA was synthesized using a Reverse Transcription System (Promega, WI). RT-PCR was replicated using two biologically independent samples. As internal controls, parts of the UBC21 (At5g25760) or EF1 (At5g60390) genes were amplified in the same PCR reaction tube with NPQ1 (At1g08550), NPQ4 (At1g44575), LHCB1.1 (At1g29920), LHCB1.2 (At1g29910), and LHCB1.3 (At1g29930) primers to demonstrate relative quantity and quality of the cDNA template. Primers were designed using AtRTPrimer³⁹ and the primer sequences are shown in Table 2. For heatmap, mean-normalized values of development (tissues) and light experiments from AtGenExpress expression library (www.weigelworld.org) were subjected to BAR Heatmapper Plus (bar.utoronto.ca). For development experiments, Col-0 WT plants were used. For light experiments, aerial parts (hypocotyl and cotyledons) from 4-d-old Col-0 WT seedlings grown on MS medium treated with different light for either 45 or 240 min were used (www.weigelworld.org).

Table 1. BVR activities in BVR lines.

Plant line^a	BVR activity (µmol h^−1mg^−1)^b	Relative BVR activity^c
No-0WT	ND^d	ND
35S::pBVR3	16.3	1.0
CAB3::pBVR1	34.9	2.1
CAB3::pBVR2	46.2	2.8
CAB3::pBVR3	11.1	0.7

^aPlants were grown under white light (100 µmol m⁻² s⁻¹) with a long-day photoperiod (16-h-light and 8-h-dark cycle) for 30 d. Rosette leaves were harvested and frozen with liquid nitrogen, and soluble protein extracts were used for BVR assay.

^bValues were means from four replicates.

^cRelative to BVR activity in 35S::pBVR3.

^dND: not detected.

Table 2. Primer sequences for RT-PCR.

AGI number	Common name	Primers (forward/reverse)
At1g08550	NPQ1	cttgcgcgttccttattgttcca/ttgcacgaatttttgtacggctga
At1g44575	NPQ4	ggctcaaaccatgctgcttacttca/agggcagatcggacgtttttgc
At1g29910-At1g29930	LHCB1.1-1.3/CAB1-3	gcctccacaatggctctct/cggggtaaagcaagtcctc
At5g25760	UBC21	ccttacgaaggcggtgtttttcag/cggcgaggcgtgtatacatttg
At5g60390	EF1	tagccgcaagactcctttcagattc/gatgtcctggggcatcaatgact

Results

BVR activity in adult plants

We determined the BVR activity in fully expanded rosette leaves from white light-grown adult plants (30-d-old). We measured high BVR activity for CAB3::pBVR1 and CAB3::pBVR2 compared to the BVR activities of 35S::pBVR3 or CAB3::pBVR3 (Table 1). BVR activities in rosette leaves from adult BVR lines were different from those previously tested in entire seedlings grown for only 7 d, which showed the highest BVR activity in CAB3::pBVR2 and 35S::pBVR3 lines.²⁰ These differences may be related to distinct developmental stages, but also are likely impacted by differences in procedure as here we only measure activity in protein samples derived from leaf tissue, whereas prior measurements were with extracts from whole seedlings, the latter of which have tissues that have little to no BVR expression and thus have dilution effects on total soluble protein extracts. Based on our BVR-specific activity assays using leaf tissues from adult plants (Table 1), we considered that CAB3::pBVR1 and CAB3::pBVR2 were strong lines for inactivation of phytochromes in adult leaves compared to CAB3::pBVR3 and 35S::pBVR3.

Phytochrome-dependent regulation of distinct photosynthetic parameters in BVR plants

Constitutive inactivation of phytochromes in 35S::pBVR lines disrupted numerous light responses, including de-etiolation of seedlings and chlorophyll accumulation.^29,30 Mesophyll-specific inactivation of phytochromes in CAB3::pBVR seedlings results in defects of light-mediated responses, including inhibition of hypocotyl elongation, greening, and accumulation of anthocyanin.²⁰ As mature plants (45-d-old), CAB3::pBVR1 and CAB3::pBVR2 were smaller and less green than No-0 WT (Figure 1). CAB3::pBVR1 and CAB3::pBVR2 exhibited low levels of NPQ when assayed using chlorophyll fluorescence imaging, whereas the level of NPQ in CAB3::pBVR3 appeared comparable to levels observed for WT or 35S::pBVR3 (Figure 1).

Figure 1. Growth and non-photochemical quenching (NPQ) phenotypes of No-0 wild-type (WT), constitutive inactivation of phytochrome line (35S::pBVR3), and mesophyll tissue-specific phytochrome inactivation lines (CAB3::pBVR1 [CAB3-1], CAB3::pBVR2 [CAB3-2], and CAB3::pBVR3 [CAB3-3]). Plants were grown under white light (100 µmol m⁻² s⁻¹) with a long-day photoperiod (16-h-light and 8-h-dark cycle) for 45 d. Before photographing, stems were removed from plants. Top panel: Bright light image of the Arabidopsis strains. Bottom panel: NPQ image of the Arabidopsis strains. In false-color NPQ image, white reflects the WT level of NPQ, whereas red reflects a low level of NPQ.

To examine more closely the effect of mesophyll-specific inactivation of phytochromes on NPQ, we assessed distinct components of NPQ in leaves of adult BVR lines under different fluence rates of white light (0–400 µmol m⁻² s⁻¹). NPQ and qE were reduced (>2-fold) in CAB3::pBVR1 and CAB3::pBVR2 compared to WT, 35S::pBVR3, or CAB3::pBVR3 (Figure 2a and b). As plants were grown in 100 µmol m⁻² s⁻¹ of white light for 30 d before examining NPQ levels, the differences were most prominent when plants were exposed to higher fluence rates of light than growth conditions (i.e., >97 µmol m⁻² s⁻¹). Consistent with results for BVR activity assays (Table 1), plants with strong inactivation of phytochromes in mesophyll tissues (i.e., CAB3::pBVR1 or CAB3::pBVR2) showed the greatest reduction in NPQ (qE) levels compared to plants with weak inactivation of phytochromes in mesophyll tissues (i.e., CAB3::pBVR3) or plants with constitutive inactivation of phytochrome in every tissue (i.e., 35S::pBVR3) (Figure 2). Despite low NPQ and qE in CAB3::pBVR1 and CAB3::pBVR2, qI at each fluence rate was only slightly affected in these lines compared with WT (Figure 2c), and significant differences were only observed for the CAB3::pBVR2 line with the highest BVR expression and only a lower light intensities. Thus, the data supported that mesophyll-specific phytochromes have a role in the regulation of NPQ, primarily the qE component.

Figure 2. Measurement of photosynthetic properties of No-0 wild-type (WT), constitutive inactivation of phytochrome line (35S::pBVR3), and mesophyll tissue-specific phytochrome inactivation lines (CAB3::pBVR1 [CAB3-1], CAB3::pBVR2 [CAB3-2], and CAB3::pBVR3 [CAB3-3]). An attached, fully expanded rosette leaf (10th or 11th leaf) from a 30-d-old plant was used for each independent sample. Plants were grown in white light (100 µmol m⁻² s⁻¹) with a long-day photoperiod (16-h-light and 8-h-dark cycle). Before testing, plants were dark-adapted overnight. (a) NPQ as a function of light intensity (μmol m⁻² s⁻¹). (b) Energy-dependent exciton quenching (qE) as a function of light intensity (μmol m⁻² s⁻¹). (c) Photoinhibition (qI) as a function of light intensity (μmol m⁻² s⁻¹). (d) LEF (linear electron flow) as a function of light intensity (μmol m⁻² s⁻¹). (e) Maximal quantum yield of PSII (Fv/Fm). Data were presented as the mean ± SD (n = 5 or 6). (a–e) Unpaired, two-tailed Student’s t test comparing BVR lines to No-0 WT at each light intensity, *p < 0.05; red asterisks, CAB3::pBVR1 compared to WT; blue asterisks, CAB3::pBVR2 compared to WT; and gold asterisks, 35S::pBVR compared to WT. No symbols indicate no significant difference.

We also assessed LEF in BVR lines compared to WT. LEF is linked to the production of NADPH and the synthesis of ATP through the generation of a proton motive force. We observed reduced LEF levels in CAB3::pBVR1 and CAB3::pBVR2 lines (Figure 2d), the severity of which was correlated with BVR activity and associated reductions in phytochrome levels in these lines. These lines also exhibited reduced quantum yield of PSII, i.e., lower Fv/Fm values, which indicates the induction of damage to PSII (Figure 2e).

Chlorophyll content and chl a/chl b ratios in BVR lines

In previous studies, CAB3::pBVR seedlings showed decreased chlorophyll levels and a subtle increase in chl a/chl b ratios.²⁰ In the present studies, adult plants of CAB3::pBVR1 and CAB3::pBVR2 exhibited significant decreases in total chlorophyll content, whereas CAB3::pBVR3 had a level of chlorophyll that was not significantly different from No-0WT (Figure 3a). This age-dependent regulation of chlorophyll levels and plastid development, especially the recovery of defects in mature plants, has been previously reported.^29,40,41 The level of chlorophyll in 35S::pBVR3 was moderately, but significantly reduced compared to No-0 WT (Figure 3a). CAB3::pBVR1 and CAB3::pBVR2 exhibited significant increases in the chl a/chl b ratios (about 40%) compared to other lines (Figure 3b). It has been shown that chl a/chl b ratios are positively correlated with the ratios of PSII cores to light-harvesting chlorophyll–protein complexes (antenna complexes).⁴² Light-harvesting chlorophyll–protein complexes contain the majority of the cellular chl b pool, and the chl a/chl b ratios are related to antenna size.⁴³ Previously, it has been shown that qE occurs in the antenna system of PSII, and the extent of NPQ is positively correlated with the antenna size of PSII.⁴⁴ Our observation of high chl a/chl b ratios in CAB3::pBVR1 and CAB3::pBVR2 suggested a role of mesophyll-specific phytochromes in NPQ, possibly by impacting assembly, composition, or stability of antenna complexes. This impact of mesophyll-specific phytochromes on these processes may be in part through the association of phytochrome with regulating chlorophyll levels⁴⁵ and the quantitative input of phytochromes into the accumulation of Lhcb proteins.¹⁶ Additionally, reduced chlorophyll levels have been previously correlated with disruptions in thylakoid architecture, including grana formation and stacking.¹⁴⁻¹⁷ We also noted reduced grana and disturbed stacking, especially at early developmental stages for leaves lacking phytochromes compared to WT (Supplemental Figure 1). Such disruptions in plastid ultrastructure could also have indirect effects on the photosynthetic parameters measured in CAB3::pBVR lines.

Figure 3. Chlorophyll content and ratio of chlorophyll a (Chl a) and chlorophyll b (Chl b) in No-0 wild-type (WT), 35S::pBVR3 (35S), CAB3::pBVR1 [CAB3-1], CAB3::pBVR2 [CAB3-2], and CAB3::pBVR3 [CAB3-3] lines. Extraction of chlorophyll with DMF (N,N-dimethylformamide) was performed using 10 mg of 30-d-old rosette leaves of plants as described in Figure 2. (a) Total chlorophyll. (b) Ratio of chl a and chl b (chl a/chl b). Data were presented as the means ± SD (n = 5). Unpaired, two-tailed Student’s t test comparing BVR lines with WT, *p < 0.005, **p < 0.0001.

Impact of mesophyll-specific expression of BVR on antenna complexes

To test whether mesophyll-specific phytochromes have roles, which could be either direct or indirect, in the regulation of expression, assembly, or stability of the components of antenna complexes, we examined the accumulation of total leaf proteins from 30-d-old BVR lines using SDS-PAGE gel electrophoresis and Western blot analyses. We found that proteins estimated at ~25 kDa exhibited reduced accumulation in CAB3::pBVR1 and CAB3::pBVR2 lines (Figure 4). The major light-harvesting antenna complex II is composed of up to three different individual subtypes of chlorophyll a/b-binding proteins (Lhcb1–Lhcb3), and Lhcb1 is the most abundant protein in plants.^46,47 Western blot analysis using an anti-Lhcb1 antibody showed that Lhcb1 accumulated to lower levels in CAB3::pBVR1 and CAB3::pBVR2, suggesting that the mesophyll-specific pool of phytochromes is required for regulation of Lhcb1 (Figure 4). In Arabidopsis, Lhcb1 protein is encoded by at least five nuclear genes. The decrease of Lhcb1 protein accumulation in CAB3::pBVR1 and CAB3::pBVR2 suggested roles of mesophyll-specific phytochrome in transcriptional regulation of Lhcb1 genes or in stability of Lhcb1 protein.

Figure 4. Representative SDS-polyacrylamide gel electrophoresis and Western blot analyses of soluble proteins from No-0 wild-type (WT), 35S::pBVR3 (35S), CAB3::pBVR1 [CAB3-1], CAB3::pBVR2 [CAB3-2], and CAB3::pBVR3 [CAB3-3] lines. (Top panel) Total soluble proteins were extracted from 30-d-old plants (rosette leaves) grown under long-day condition. Proteins were resolved on 15% SDS-PAGE gel. Arrow indicated the proteins with reduced accumulation in CAB3-1 and CAB3-2 compared to WT and other lines. (Bottom panel) For Western blot analysis, anti-Lhcb1 antibody (Agrisera, AS01 004) was used.

To investigate the assembly or stability of chlorophyll–Lhcb1 complexes in the BVR lines, thylakoid membrane-enriched fractions solubilized in non-ionic detergents were subjected to native green gel electrophoresis and Western blot analyses (Figure 5). On green gels, several distinctively accumulated chlorophyll-associated protein complexes in CAB3::pBVR1 and CAB3::pBVR2 were detected. Complexes most abundant in WT (denoted as an arrow in Figure 5a) accumulated to lower levels in CAB3::pBVR1 and CAB3::pBVR2. Western blot analyses showed that the chlorophyll–protein complexes which accumulated to lower levels in CAB3::pBVR1 and CAB3::pBVR2 were chlorophyll–Lhcb1-associated complexes (Figure 5b). Considering the size of the complexes on green gels, the strong and weak bands on Western blot should correspond to trimeric LHCII and monomeric LHC associated with chlorophyll, respectively (arrow and arrowhead in Figure 5). Chlorophyll–trimeric LHCII complexes, including Lhcb1, were less abundant in CAB3::pBVR1 and CAB3::pBVR2, whereas chlorophyll–monomeric LHC complexes were not changed to the same degree as trimers in those lines, compared to WT (Figure 5b). Notably, monomers have been identified as intermediates in the formation of trimers;⁴⁸ thus, a lack of mesophyll phytochromes is correlated with a potential disruption in the trimer assembly process. The association of trimeric LHCII with chlorophyll was intact in lines constitutively expressing BVR (35S::pBVR3) or weakly expressing BVR in the mesophyll (CAB3::pBVR3) compared to WT. Whereas phytochromes have been previously associated with accumulation of Lhcb proteins,¹⁶ these data suggested that the pool of mesophyll-specific phytochromes is required for accumulation of Lhcb1 and may impact assembly of LHCII antenna complexes that are necessary for WT levels of NPQ.⁴⁹

Figure 5. Representative native green gel electrophoresis and Western blot analyses of thylakoid membrane-enriched fraction from No-0 wild-type (WT), 35S::pBVR3, CAB3::pBVR1 [CAB3-1], CAB3::pBVR2 [CAB3-2], and CAB3::pBVR3 [CAB3-3] lines. (a) Thylakoid membrane-enriched fractions were prepared using 80% percoll from 30-d-old plants. The solubilized proteins with 2% (v/v) octyl glucoside and 2% (v/v) decyl maltoside were subjected to native green gels. (b) Western blot analysis. Proteins associated with chlorophyll from native green gels were transferred to nitrocellulose membrane, and the membrane blot was subjected to immune reaction with anti-Lhcb1 antibody. Both arrow and arrowhead indicated Lhcb1 proteins associated with chlorophyll with different molecular mass.

Expression of genes required for NPQ in BVR lines

To investigate the role of mesophyll-specific phytochromes in regulating NPQ further, we studied the expression of select genes required for NPQ in BVR lines using RT-PCR analyses. We assessed NPQ1, NPQ4, and LHCB1 genes. NPQ1 (At1g08550) encodes violaxanthin de-epoxidase (VDE), an enzyme required for low-pH-inducible qE.⁴ PsbS protein is encoded by NPQ4 (At1g44575) and is a thylakoid membrane integral protein component of PSII that is necessary for NPQ.⁶ Lhcb1 is encoded by several chlorophyll a/b-binding protein genes, i.e., LHCB1.1 (At1g29920), LHCB1.2 (At1g29910), and LHCB1.3 (At1g29930). Our RT-PCR analyses showed no obvious differences in gene expression in BVR lines compared with No-0 WT (Figure 6a), indicating that mesophyll-specific phytochromes are not required for the transcriptional regulation of NPQ1, NPQ4 or LHCB1.1 to LHCB1.3. Whereas the expression of LHCB1 genes in CAB3::pBVR1 or CAB3::pBVR2 lines were comparable to WT (Figure 6a), Lhcb1 protein accumulation was reduced significantly in those lines (Figure 4), suggesting that mesophyll-specific phytochromes exhibit post-transcriptional or post-translational regulation of products required for NPQ. We examined the expression of NPQ1, NPQ4, LHCB1.1, LHCB1.2, and LHCB1.3 in different tissues at various plant developmental stages and in different light conditions using public Arabidopsis microarray data set and heatmap tools (Figure 6b and c). The expression pattern of these genes was quite similar in various tissues or light conditions (Figure 6b and c). Expression of the genes was upregulated in mesophyll-abundant tissues, such as cotyledons and leaves, and also upregulated in response to light (Figure 6b and c).

Figure 6. Expression of NPQ-related genes in BVR lines. (a) RT-PCR analysis. Rosette leaves from 30-d-old plants of No-0 wild-type (WT), 35S::pBVR3, CAB3::pBVR1 [CAB3-1], CAB3::pBVR2 [CAB3-2], and CAB3::pBVR3 [CAB3-3] were used for RT-PCR analysis. As internal controls, a part of the UBC21 (At5g25760) or EF1 (At5g60390) was amplified in the same PCR reaction with NPQ1, NPQ4, or LHCB1.1-LHCB1.3 (to conserved region of LHCB1.1, LHCB1.2, and LHCB1.3) primers to demonstrate relative quantity and quality of the cDNA template. RT-PCR was repeated using two independent biological samples. (b,c) Heat map showing the expression of NPQ1, NPQ4, or LHCB1.1-LHCB1.3 in different Arabidopsis tissues (b) or different light conditions (c). For heat map, mean-normalized values from AtGenExpress expression library (www.weigelworld.org) and BAR Heatmapper Plus (bar.utoronto.ca) were used.

Discussion

Mesophyll-specific phytochromes regulate distinct photosynthetic parameters

To understand the molecular mechanisms responsible for tissue-specific phytochrome responses, transgenic plants inactivating phytochromes in the mesophyll (i.e., CAB3::pBVR) have been successfully used for photobiological analyses (reviewed in^50,51). Mature CAB3::pBVR1 and CAB3::pBVR2 lines (30-d-old) exhibited stronger BVR activity than 35S::pBVR3 (Table 1). Adult CAB3::pBVR1 and CAB3::pBVR2 plants with strong BVR activity in leaves exhibited low NPQ in qualitative and quantitative assays, whereas 35S::pBVR3 plants exhibited a level of NPQ comparable to WT (Figures 1 and 2a). Among distinct components of NPQ, qE was significantly reduced (about 2-fold) at higher light fluences in CAB3::pBVR1 and CAB3::pBVR2 lines (Figure 2b). By comparison, qI was not changed significantly, except for being significantly reduced in CAB3::pBVR2 at lower light intensities (Figure 2c). The specific effect on qE compared to qI indicates a role for mesophyll-specific phytochromes in impacting NPQ via generation of a pH gradient across the membrane. These results are consistent with a prior observation that phytochromes appear to play a role in protective responses against excessive light absorption through regulation of NPQ ;¹⁸ however, here we link this to a pool of mesophyll-localized phytochromes.

Mesophyll-specific phytochromes also impacted the photosynthetic parameters of LEF and Fv/Fm in the CAB3::pBVR1 and CAB3::pBVR2 lines, specifically. This suggests a disruption in the generation of NADPH and ATP in these lines, which may be associated with PSII damage indicated by the reduction in Fv/Fm. Together, these may contribute to the observed reduction of NPQ responses in these lines.

Post-translational role of mesophyll-specific phytochromes in NPQ

When we evaluated the expression of NPQ1, NPQ4, LHCB1.1, LHCB1.2, and LHCB1.3 genes that are involved in qE, we did not find a change in the expression of those genes in rosette leaves of 30-d-old BVR lines (Figure 6a). However, the accumulation of Lhcb1 proteins was reduced in CAB3::pBVR1 and CAB3::pBVR2 lines (Figure 4). Therefore, the role of mesophyll-specific phytochromes is most likely to regulate NPQ-related genes at the post-transcriptional or post-translational level (e.g., protein stability) in adult leaves. Notably, NPQ1, LHCB1.1, LHCB1.2, and LHCB1.3 genes are significantly downregulated (NPQ1: 2.3-fold; LHCB1.1–LHCB1.3: 3.0-fold), and NPQ4 is moderately downregulated (about 1.7-fold) in far-red-light grown CAB3::pBVR2 seedlings when compared to expression in 35S::pBVR3.⁵² This can be interpreted as a spatial-specific (i.e., developmental stages and tissues-specific) and/or wavelength-dependent role of phytochromes in impacting the transcription of genes involved in NPQ. Indeed, we provide evidence that these genes are differentially regulated in a tissue- and/or light-specific manner (Figure 6b and c).

Role of mesophyll-specific phytochromes in the assembly of antenna complexes and impact on NPQ

We observed a low level of chlorophyll (~7-fold reduction) and high ratio of chl a/chl b (~40% increase) in CAB3::pBVR1 and CAB3::pBVR2 lines compared to WT. Whereas prior analyses of natural phytochrome chromophore-deficient strains associated the reduction in chlorophyll levels with buildup of intermediates such as heme and feedback inhibition of aminolevulinic acid (ALA) synthesis,⁴⁵ BVR expression has distinct impacts on tetrapyrrole homeostasis and associated plastid metabolites/intermediates by nature of the BVR enzymatic activity to reduce biliverdin.⁵³ These studies demonstrated that BVR activity shifts metabolite partitioning between the heme and chlorophyll branches through increasing flow of intermediates through the heme branch in a manner that results in reduced levels of heme due to increased heme oxygenase activity, as well as leading to increased ALA synthesis.⁵³ This BVR-dependent commitment to flux through the heme branch results in inhibition of Mg-porphyrin synthesis and, thus, reduced chlorophyll levels.

The reduced chlorophyll levels of CAB3::pBVR1 and CAB3::pBVR2 likely result in a defect in antenna complex assembly in these lines (Figure 3). In addition, our green gel analyses of chlorophyll–protein complexes showed an impairment in chlorophyll–Lhcb1 stability (most likely LHCII trimers) in CAB3::pBVR1 and CAB3::pBVR2 lines (Figure 5). This differential effect on trimers versus monomers indicates a potential disruption in the trimer synthesis process.⁴⁸ A xanthophyll mutant, lutein-deficient2 (lut2), exhibits reduced levels of qE⁵⁴ and a low stability of trimeric Lhcb1 with chlorophyll.⁵⁵ The reduced stability of Lhcb1 with chlorophyll in lut2 mutants has been translated into a disruption of the organization of the PSII–antenna system, resulting in compromised NPQ.⁵⁵ The reduced level of NPQ with reduced stability of Lhcb1 in lut2 was reminiscent of phenotypes observed in CAB3::pBVR1 and CAB3::pBVR2 (Figures 2 and 5). We, thus, proposed that phytochromes in mesophyll tissues could contribute to NPQ by impacting the organization of the PSII–antenna complexes. In addition to the chlorophyll–Lhcb1 complex, our green gel analyses showed that the largest complexes (presumably super-complexes with PSI and PSII) were also absent in CAB3::pBVR1 and CAB3::pBVR2 (Figure 5a). Thus, it appears that mesophyll-specific phytochromes contribute to the assembly and stability of Lhcb1 with chlorophyll, which contributes to the regulation of NPQ. This role is distinct from the mesophyll-localized phytochrome-dependent regulation of SIG2, which in turn impacts NPQ through transcriptional control of plastid genes.²⁶

High light-grown (400 µmol m⁻² s⁻¹) phytochrome A mutant (phyA-201) exhibits low levels of NPQ compared to WT, whereas phytochrome B mutant phyB-1¹⁸ and sig2 mutants²⁶ show moderately high levels of NPQ. Ambient light-grown double phyAphyB mutants also have moderately higher levels of NPQ compared to WT.⁵⁶ The adult rosette leaves of CAB3::pBVR1 and CAB3::pBVR2 lines have reduced the level of NPQ. Multiple of these lines has chlorophyll deficiencies, yet distinct impacts on NPQ (i.e., either reduced or elevated). Thus, it is not simply a reduction of chlorophyll that results in altered NPQ, but specific impacts on chlorophyll synthesis, regulation of associated proteins, or the regulation of plastid transcription that can yield disparate effects on plant development and NPQ.

Declaration of Interest

Neither author has financial interests or benefits that have arisen from the direct applications of this research.

Acknowledgments

The authors thank Jeffrey Cruz, Deserah Strand, Ben Lucker, and David M. Kramer for assistance with and discussion of chlorophyll fluorescence measurements and green gel analysis. This research was supported by the Office of Science of the US Department of Energy (grant no. DE‐FG02‐91ER2002 to B.L.M.) and general support to B.L.M. from the National Science Foundation (grant no. MCB‐1515002).

Supplementary Materials

Supplemental data for this article can be accessed on the publisher’s website.

Additional information

Funding

This work was supported by the National Science Foundation [MCB‐1515002]; U.S. Department of Energy [DE‐FG02‐91ER20021].