1. Plant Biology
Download icon

Photosynthesis without β-carotene

  1. Pengqi Xu
  2. Volha U Chukhutsina
  3. Wojciech J Nawrocki
  4. Gert Schansker
  5. Ludwik W Bielczynski
  6. Yinghong Lu
  7. Daniel Karcher
  8. Ralph Bock
  9. Roberta Croce  Is a corresponding author
  1. Biophysics of Photosynthesis, Department of Physics and Astronomy, Faculty of Sciences, VU University Amsterdam and LaserLab Amsterdam, Netherlands
  2. Max Planck Institute of Molecular Plant Physiology, Germany
Research Article
  • Cited 2
  • Views 3,215
  • Annotations
Cite this article as: eLife 2020;9:e58984 doi: 10.7554/eLife.58984

Abstract

Carotenoids are essential in oxygenic photosynthesis: they stabilize the pigment–protein complexes, are active in harvesting sunlight and in photoprotection. In plants, they are present as carotenes and their oxygenated derivatives, xanthophylls. While mutant plants lacking xanthophylls are capable of photoautotrophic growth, no plants without carotenes in their photosystems have been reported so far, which has led to the common opinion that carotenes are essential for photosynthesis. Here, we report the first plant that grows photoautotrophically in the absence of carotenes: a tobacco plant containing only the xanthophyll astaxanthin. Surprisingly, both photosystems are fully functional despite their carotenoid-binding sites being occupied by astaxanthin instead of β-carotene or remaining empty (i.e. are not occupied by carotenoids). These plants display non-photochemical quenching, despite the absence of both zeaxanthin and lutein and show that tobacco can regulate the ratio between the two photosystems in a very large dynamic range to optimize electron transport.

eLife digest

Most life on Earth depends on photosynthesis, the process used by plants and many other organisms to store energy from sunlight and produce oxygen. The first steps of photosynthesis, the capture and conversion of sunlight into chemical energy, happen in large assemblies of proteins containing many pigment molecules called photosystems. In plants, the pigments involved in photosynthesis are green chlorophylls and carotenoids. In addition to harvesting light, carotenoids have an important role in preventing damage caused by overexposure to sunlight

There are over one thousand different carotenoids in living beings, but only one, β-carotene, is present in every organism that performs the type of photosynthesis in which oxygen is released, and is thought to be essential for the process. However, this could never be proved because it is impossible to remove β-carotene from cells using typical genetic approaches without affecting all other carotenoids.

Xu et al. used genetic engineering to create tobacco plants that produced a pigment called astaxanthin in place of β-carotene. Astaxanthin is a carotenoid from salmon and shrimp, not normally found in plants. These plants are the first living things known to perform photosynthesis without β-carotene and demonstrate that this pigment is not essential for photosynthesis as long as other carotenoids are present. Xu et al. also show that the photosystems can adapt to using different carotenoids, and can even operate with a reduced number of them.

Xu et al’s findings show the high flexibility of photosynthesis in plants, which are able to incorporate non-native elements to the process. These results are also important in the context of increasing the photosynthetic efficiency, and thus the productivity of crops, since they show that a radical redesign of the photosynthetic machinery is feasible.

Introduction

Carotenoids form a large class of natural pigments responsible for the yellow, orange, and red colors of fruits and leaves (Stange, 2016). In the photosynthetic membranes, they are mainly associated with proteins, forming pigment–protein complexes. Their large absorption cross-section in the blue region of the solar spectrum makes them ideal light-harvesting pigments, especially for aquatic organisms (Croce and van Amerongen, 2014). However, the primary role of carotenoids in photosynthesis is photoprotection. Their capacity to quench chlorophyll (Chl) triplets (thus avoiding their reaction with molecular oxygen and the production of singlet oxygen), and to scavenge singlet oxygen make them essential for the survival of the organism (Frank and Cogdell, 1996; Borth, 1975; Havaux, 1998). In addition, carotenoids are involved in the quenching of singlet excited state Chls in a process known as non-photochemical quenching (NPQ), which controls the level of excited states in the membrane, thus protecting the photosynthetic apparatus from high light damage (Ruban et al., 2012).

Two species of carotenoids are present in the photosynthetic membranes: carotenes and their oxygenated derivatives, xanthophylls. The main carotene, β-carotene (β-car), is associated with the core of photosystems I and II (Umena et al., 2011; Qin et al., 2015), and is present in all organisms performing oxygenic photosynthesis. The xanthophylls (in plants mainly lutein (Lut), neoxanthin (Neo), violaxanthin (Vio) and zeaxanthin (Zea)), instead, are bound to the light-harvesting complexes (LHCs) that act as peripheral antennae increasing the absorption cross-section of both photosystems (Qin et al., 2015; Su et al., 2017). LHCs are able to accommodate different xanthophylls, but they cannot fold in the presence of β-carotene only (Croce et al., 1999; Phillip et al., 2002). Also, PSII assembly has been suggested to require the presence of carotenes (Masamoto et al., 2004), while PSI is stable also in the absence of carotenoids (Masamoto et al., 2004; Santabarbara et al., 2013).

While mutants lacking individual or all xanthophylls but still containing carotenes have been identified for several organisms (e.g. Dall'Osto et al., 2013; Ware et al., 2016; Niyogi et al., 1997; Pogson et al., 1998; Domonkos et al., 2013; Schäfer et al., 2005), mutants lacking carotenes have only been isolated in cyanobacteria and the green alga Chlamydomonas reinhardtii when these organisms were grown heterotrophically (Santabarbara et al., 2013; Sozer et al., 2010; Tóth et al., 2015). In these mutants, no PSII was assembled. This finding, together with the fact that no PSII complexes without carotenes have ever been observed, have suggested that carotenes have a vital role not only in photosynthesis but also for the survival of the plant cell (Dall’Osto et al., 2014). However, this assumption could never be verified, because available mutants without carotenes completely lack carotenoids.

In this work, we have analyzed tobacco (Nicotiana tabacum) plants in which the carotenoid biosynthetic pathway was engineered (by stable transformation of the chloroplast genome) to only produce the ketocarotenoid astaxanthin (Lu et al., 2017Figure 1). The physiological characteristics and the autotrophic growth of these plants demonstrate that photosynthesis without carotenes is possible, at least when plants are grown in laboratory conditions.

Figure 1 with 1 supplement see all
Phenotype of wild-type (WT) and astaxanthin-synthesizing tobacco plants (Asta) and pigment analysis.

WT (A) and Asta (B) plants, 7 and 21 weeks old, respectively. Note that the older leaves in the mutant are less orange than the young leaves. See also Figure 1—figure supplement 1. (C) Chromatographic profiles of the pigments extracted from leaves normalized to the Chl a peak. Neo, neoxanthin; Vio, violaxanthin; Lut, lutein; β-car, β-carotene; Chl, chlorophyll; 4-keto, 4-ketoanteraxanthin; Ast, astaxanthin; Ado, adonixanthin; Can, Canthaxanthin.

Results and discussion

The leaves of the tobacco mutant are orange at an early stage and become greener with age (see Figure 1, Table 1). This is likely due to the high expression of the chloroplast genome in young leaves (Edwards et al., 2010), which in the mutant results in the massive production of astaxanthin. Most of the astaxanthin is present in the form of crystals or aggregates in the chloroplast (Lu et al., 2017). The high-level synthesis of astaxanthin uses a substantial part of the plant’s energy budget and fixed carbon and may contribute to the slow growth of the plants. This negative effect on growth is likely exacerbated by the fact that astaxanthin absorbs most of the incident light, decreasing the number of photons available for photosynthesis. Indeed, the greening of the leaves corresponds to an increase in the growth rate of the plants. Both young and mature leaves of the mutant plants contain only 20% of the Chls per fresh weight as compared to the wild-type (WT), but have a similar (mature leaves) or far higher (young leaves) carotenoid content (Table 1). However, the mutant, at all stages of growth, only contains astaxanthin and traces of by-products of astaxanthin synthesis (Hasunuma et al., 2008) and does not accumulate (<0.005 times the WT) the carotenoids that are usually present in the WT (Figure 1). This result is different from the analysis of previously generated astaxanthin-producing plants that still contained WT carotenoids, although in reduced amounts (Hasunuma et al., 2008; Fujii et al., 2016; Röding et al., 2015). Thus, our engineered tobacco (hereafter referred to as Asta) represents the first organism showing autotrophic growth in the virtual absence of carotenes. In the following, we report the experiments performed on mature leaves, which have a Chl/car similar to the WT.

Table 1
Pigment composition of leaves.
SamplesChl a/bChl/carChl/fresh wt (mg/g)Chl/leaf area # (mg)
WT3.79 ± 0.094.21 ± 0.202.63 ± 0.390.0312 ± 0.0026
Asta mature leaves3.01 ± 0.103.02 ± 0.220.58 ± 0.070.0062 ± 0.0008
Asta young leaves3.16 ± 0.190.76 ± 0.130.55 ± 0.090.0049 ± 0.0011
  1. (#50 mm2; Average values ± SD are shown. n = 10 biological replicas).

Since violaxanthin and lutein are considered to be necessary for the folding of the antenna complexes (Dall'Osto et al., 2006), and β-carotene was thought to be required for PSII assembly and photosynthetic activity (Santabarbara et al., 2013; Sozer et al., 2010; Tóth et al., 2015), we analyzed the effect of their absence on the composition and organization of the photosynthetic apparatus. 2D gel electrophoresis (Figure 2) and immunoblot analyses (Figure 2—figure supplement 1) of thylakoid membranes show that all of the main photosynthetic proteins are present in Asta plants, but the PSII/PSI ratio is far higher than in the WT (Figure 2—figure supplement 1). The LHC/PSII ratio is, however, similar, except for the antenna protein Lhcb5 that is strongly reduced, and for PsbS, the main protein involved in NPQ (Li et al., 2000), which is increased in the mutant (Figure 2—figure supplement 1). PSI-LHCI, ATP synthase and cytochrome b6f have the same mobility in native gels as the WT complexes, indicating that they are stable and have the same supramolecular organization. By contrast, the stability of PSII seems to be affected as the bands corresponding to PSII supercomplexes and LHCII trimers, which are well defined in the WT, are substituted by a smear in the mutant, suggesting that the PSII complexes are more heterogeneous, incompletely assembled or less stable than in the WT (Figure 2).

Figure 2 with 1 supplement see all
Protein composition and supramolecular organization of the photosynthetic complexes in WT and Asta plants.

(A) Thylakoids were solubilized with 1% α-DDM and loaded on a blue-native gel. Second dimension SDS-PAGE of the WT (B) and the Asta mutant (C). The immunoblotting analysis is shown in Figure 2—figure supplement 1.

Next, we investigated the effects of the change in carotenoid composition on the properties of the individual complexes that were isolated from thylakoid membranes and separated by sucrose density gradient ultracentrifugation (Figure 3—figure supplement 1). Pigment analysis (Table 2 and Figure 3—figure supplement 2) confirmed that astaxanthin is the only carotenoid associated with all pigment-binding complexes in Asta plants, while β-carotene is present in PSI in a highly substoichiometric amount (0.15 β-carotene molecules per complex) and it is virtually absent in PSII (0.03 molecules per complex). This means that most of the PSI and PSII complexes in the mutant plants do not contain β-carotene at all. Normalized to Chl, Asta-LHCs contain the same number of carotenoids as the WT monomers, but instead of binding lutein, neoxanthin and violaxanthin, they only bind astaxanthin, indicating that all carotenoid-binding sites are promiscuous and can accommodate different xanthophylls. The pigment analysis also showed that β-carotene can be substituted by astaxanthin in both PSII and PSI cores. However, the higher Chl/car ratio in the isolated Asta complexes compared to the WT complexes indicates that not all sites that are occupied by β-carotene in the WT are occupied by astaxanthin in the mutant complexes, but some are left ‘empty’ in that they are not occupied by carotenoids. Although we cannot exclude the possibility that some of the astaxanthin molecules are more loosely bound and thus are lost during purification, the fact that both PSI and PSII complexes can be purified with a large number of ‘empty’ sites indicates that their occupancy by carotenoids is not crucial for the stability of the complexes.

Table 2
Pigment composition of isolated thylakoidal complexes.
SamplesChl a/bChl/carLute+neo+violaβ-caroAsta and its bypass productsTotal chls*
WT Lhcb
Monomers
2.41 ± 0.014.1 ± 0.032.86 ± 0.010.06 ± 0.01n.d.12
WT LHCII trimer1.40 ± 0.003.60 ± 0.013.90 ± 0.01n.d.n.d.14
Asta-Lhcb mon1.48 ± 0.014.50 ± 0.03n.d.n.d.3.1 ± 0.0214
WT PSII8.88 ± 0.125.00 ± 0.023.47 ± 0.083.93 ± 0.08n.d.37
Asta PSII7.74 ± 0.278.90 ± 0.17n.d.0.03 ± 0.014.12 ± 0.0137
WT PSI-LHCI9.29 ± 0.124.60 ± 0.0214.37 ± 0.1619.56 ± 0.16n.d.156
Asta-PSI-LHCI5.44 ± 0.198.10 ± 0.18n.d.0.15 ± 0.0319.11 ± 0.03156
  1. *Total Chls are based on values reported in the literature for the WT complexes (Qin et al., 2015; Su et al., 2017). The chromatograms are shown in Figure 3—figure supplement 2. (Average values ± SD are shown. n ≥ 3 biological replicas, n.d. = not detected).

Interestingly, absorption (Figure 3—figure supplement 3A) and circular dichroism (Figure 3—figure supplement 3B) spectra of LHCs and PSII core complexes from the WT and the mutant are very similar (see Figure 3—figure supplement 3 for a more detailed explanation) in the Chl absorption regions. This indicates that there are no significant changes in the pigment organization of the complexes and thus in their three-dimensional structure. The only exception is Asta-PSI-LHCI, the fluorescence emission of which showed a 6 nm shift to shorter wavelengths as compared to the WT complex (Figure 3—figure supplement 3C and Figure 3—figure supplement 4). Since the PSI emission at 77 K originates mostly from two specific Chls (called far-red Chls) of Lhca3 and Lhca4 (Morosinotto et al., 2003), we can conclude that the interaction between these Chls is slightly changed in the mutant.

Carotenoids are known to be required for the stability of the pigment-binding holoproteins (Paulsen et al., 1993). Our data measured on the isolated complexes show that the difference in composition between WT and mutant complexes influences the denaturation temperature by only 5–10°C (Figure 3A). This is surprising considering that several of the carotenoid-binding sites in the isolated PSI and PSII are not occupied by carotenoids and indicates that only some of them play a crucial role in protein stability.

Figure 3 with 5 supplements see all
Thermal and photo-stability of photosynthetic complexes.

(A) Thermal denaturation was monitored by following the CD signals in the Qy (Lhcb: 610–700 nm, PSII core: 640–700 nm, PSI-LHCI: 675–735 nm) region at increasing temperature. (B) Photobleaching is measured as the decrease in absorption of the area in the Qy region (600–750 nm) as a function of the length of the high light treatment (Average values ± SD are shown for n = 3 technical replicas.). The purification of the complexes is shown in Figure 3—figure supplement 1. The pigment analysis of the purified complexes is shown in Figure 3—figure supplement 2. The absorption, CD, and fluorescence emission spectra of the complexes are shown in Figure 3—figure supplement 3 for comparison the fluorescence spectra of the thylakoids are shown in Figure 3—figure supplement 4. The raw data used for this figure are provided in Figure 3—figure supplement 5.

Photoprotection via Chl triplet quenching and singlet oxygen scavenging is the primary role of carotenoids in photosynthesis (Siefermann-Harms, 1987). Photobleaching experiments (Figure 3B) show that, while the photostability of LHCs and PSII core is only partially affected by the change in carotenoid composition, Asta-PSI-LHCI is far more sensitive to light than the WT complex. It is likely that this effect on PSI is due to the reduced number of carotenoid molecules associated with the complex, which results in less efficient Chl triplet quenching. However, it is worth noting that, even with a large part of the carotenoid-binding sites not occupied by carotenoids, PSI is more photostable than PSII-WT, in agreement with the fact that, in PSII, carotenoids cannot provide protection by quenching singlet oxygen formed via P680 triplet because of the very high oxidizing potential of PSII (Telfer, 2014).

Next, we investigated the effect of the substitution of carotenes with astaxanthin on the light-harvesting and trapping properties of the photosynthetic complexes in vivo by performing time-resolved fluorescence measurements at 20°C on intact leaves (Figure 4 and Figure 4—figure supplement 1). The PSI kinetics is very similar in the WT (70 ps) and in the mutant (65 ps), and the small difference can be ascribed to the reduced far-red Chl content of Asta-PSI-LHCI, which is known to influence the PSI trapping time (Croce and van Amerongen, 2013). The PSII kinetics in the mutant leaves changes in the presence versus absence of photochemistry (measurements performed with the reaction center (RC) open and closed, respectively) as it does in the WT, indicating that excitation energy transfer occurs in the mutant and the harvested energy is used for photochemistry. However, all the kinetics are faster and the difference between closed and open RC is smaller in mutant than in WT leaves, suggesting that the antenna complexes of the mutant plants are statically quenched in vivo. Measurements on isolated Asta-Lhcb show that this is indeed the case: these complexes are strongly quenched (lifetime of 0.87 ns vs. 3.5 ns in the WT; Figure 4—figure supplement 2) due to the presence of astaxanthin (Liguori et al., 2017). It has also been shown that part of the astaxanthin population can transfer excitation energy to the chlorophylls, thus also acting as light-harvesting pigment (Liguori et al., 2017).

Figure 4 with 2 supplements see all
Normalized fluorescence decay traces measured at room temperature on WT and Asta leaves in the presence (F0) or absence (FM) of photochemistry.

Fluorescence was detected at (A) λ = 685 nm and (B) λ = 720 nm. respectively. Note that, although the decay at 680 nm is dominated by PSII and at 720 nm by PSI, both complexes contribute to the decay at both wavelengths (see Figure 4—figure supplement 1 for the analysis of the complete data set with spectral resolution). The fluorescence decay and analysis of the purified LHCII are shown in Figure 4—figure supplement 2.

The presence of excitation energy transfer from the antenna to the RC in mutant leaves indicates that, although the interactions between the building blocks of the PSII supercomplex are not strong enough to survive purification (see Figure 2), the supercomplexes are functional in vivo meaning that, in the membrane, the subunits are close enough to each other to ensure the delivery of the harvested energy to the reaction center. Indeed, the short excited state lifetime of the antenna (indicative of a constitutively quenched antenna) in the mutant can fully account for the lower maximum quantum efficiency of PSII (FV/FM; Table 3) in mutant plants, which is mainly the result of low fluorescence emission in the absence of photochemistry (FM).

Table 3
In vivo photosystem II parameters.
FV/FM*PSII:PSI ratio (ECS)#PSII:PSI antenna size (ECS)#Relative PSII antenna size (fluorescence)§§
WT0.82 ± 0.011.09 ± 0.121 ± 0.591 ± 0.04
Asta0.43 ± 0.032.6 ± 0.330.33 ± 0.050.33 ± 0.02
  1. (Average values ± SD are shown. *n = 15, #n = 4, §§n = 3 leaves/plants).

Finally, we analyzed the photosynthetic performance of the Asta plants. Electrochromic shift (ECS) of the carotenoid absorption is commonly used to study the function of all major photosynthetic complexes (Bailleul et al., 2015). We verified that the mutant plants exhibit an ECS signal and we determined its light-induced difference spectrum, which agreed with the prediction that astaxanthin is solely responsible for this in vivo Stark effect (Figure 5—figure supplement 1). Using ECS, we observed that the functional PSII/PSI RC ratio was far larger in the mutant than in the WT, in qualitative agreement with our biochemical data (Table 3; Figure 2—figure supplement 1). The difference in the values obtained with the two methods is partially due to the limited quantitative power of immunoblots, but also suggests that some of the PSII cores are not functional. The high PSII/PSI ratio in the mutant seems to be a compensation mechanism for the decrease in the relative functional PSII/PSI antenna size (measured with two independent methods; Table 3 and Figure 5—figure supplement 2) observed in the Asta plants, which is due to the presence of static quenching.

Indeed, comparison of the steady-state photochemical yields of PSII and PSI revealed that, at all light intensities, in both WT and mutant plants, the balance between PSII and PSI photochemistry is maintained (Figure 5A and Figure 5—figure supplements 3 and 4), meaning that the plants are able to compensate for the strong decrease in the PSII functional antenna size by decreasing the PSI/PSII ratio. This means that these plants have the capacity to modulate the PSI/PSII ratio in a large dynamic range. Finally, transient QA reduction and reoxidation kinetics suggest that no significant differences in the PSII electron transfer occur in the mutant plants (Figure 5—figure supplements 2 and 5).

Figure 5 with 7 supplements see all
Photosynthetic performance and photoprotection.

(A) Relation between qL and ΦPSI measured at various light intensities (70–1030 μmol photons m−2 s−1). (B) NPQ kinetics upon transition from the dark-to-light (upward arrow,) and subsequent relaxation in the dark (downward arrow). Average values ± SD are shown for n = 3 leaves/plants. The inset shows amplitude-normalized kinetics. See also Figure 5—figure supplements 17.

The full operational capacity of the electron transport chain permitted us to verify whether the photoprotective regulation is maintained in the mutant plants. As expected, the NPQ amplitude was largely reduced in the mutant (Figure 5B and Figure 5—figure supplement 6), because the ΔpH-induced, PsbS-dependent quenching has to compete with the strong, constitutive astaxanthin quenching in these plants. Note that the difference in NPQ level (1.8 in the WT vs. 0.3 in the mutant) can be fully ascribed to the presence of the static quencher in the mutant, which strongly reduces the maximal fluorescence in both dark (FM) and light (FM) states. This is supported by the NPQ(t) calculation, which permits to correct the apparent NPQ for the presence of a pre-existing quenching, assuming that a decrease of the FV/FM value is solely due to this static quenching. The data show that NPQ(t) is even larger in the mutant than in the WT (Figure 5—figure supplement 7).

Importantly, despite the difference in apparent NPQ amplitude, the kinetics of onset and recovery are identical to those of the WT (inset in Figure 5B) and consistent with qE characteristics. This outcome is particularly striking if one considers that Asta plants lack both lutein and zeaxanthin, which are believed to be essential for NPQ (Niyogi et al., 1998). It is likely that the high amount of PsbS in the mutant (Figure 2—figure supplement 1) can compensate for the lack of the xanthophyll cycle, or that astaxanthin can also be responsible for the dynamic quenching. Whatever the reason for the presence of NPQ in the mutant, our Asta plants clearly show that lutein and zeaxanthin are not absolutely necessary for it.

In summary, we have shown that the carotenoid-binding sites of the core complexes of PSI and PSII are promiscuous. Although they bind carotenes in all known photosynthetic organisms, our data demonstrate that they can also accommodate xanthophylls. This is at variance with the LHCs that can bind various xanthophylls but cannot fold with carotenes. More importantly, we show that both PSI and PSII are stable while most of their carotenoid-binding sites are not occupied by carotenoids and the rest is occupied by an alien xanthophyll. These results indicate that the core complexes are even more robust than the outer antennae and can endure radical changes even in one of their main components. In this respect, it is important to realize that the difference in growth rate between WT and mutant plants is not due to the absence of carotenes, but rather to the presence of astaxanthin that stabilizes the LHCs in a quenched conformation. In conclusion, the substitution of carotenes with the xanthophyll astaxanthin does not impair the functional assembly of the photosynthetic apparatus, nor does it impede efficient electron transfer and NPQ, demonstrating that carotenes are not essential neither for the biosynthesis of the photosynthetic apparatus nor for its function. This finding has important implications not only for our understanding of the structure and function of the photosynthetic apparatus but also for future efforts to design synthetic photosystems with novel and improved properties.

Materials and methods

Tobacco growth and thylakoid isolation

Request a detailed protocol

Seeds from mutant and WT plants were sown on moist filter paper and synchronized at 4°C for 2–3 days before being moved to room temperature until germination. The seedlings were transferred to soil and grown at 22°C under 150–200 μmol photons m−2 s−1 for the WT and 80–120 μmol photons m−2 s−1 for the mutant with 14 hr of light per day. Plants were fed with commercial fertilizer each week. Leaves from WT (5–6 weeks old) and Asta plants (around 20 weeks for younger leaves, 24–30 weeks for older leaves) were used for physiological measurements and thylakoid isolation. WT thylakoid isolation was performed as described in Xu et al., 2015. The centrifuge speed was increased to 4000 g for the first step of the isolation from mutant tobacco.

Pigment analysis

Request a detailed protocol

Pigments from isolated protein–pigment complexes or leaves were extracted with 80% acetone. HPLC was performed as in Xu et al., 2015 with the modification that buffer B was linearly increased from 0 to 100% in 9.2 min. Chlorophyll a/b ratios and chlorophyll/carotenoid ratios were calculated by fitting their individual absorption spectra to measured spectra (Xu et al., 2015). Examples of the fitting of total thylakoids and isolated Lhcbs are shown in Figure 3—figure supplement 2 panels C and D.

Blue-native gel electrophoresis, SDS-PAGE, immunoblotting, and sucrose density gradient centrifugation

Request a detailed protocol

Blue-native gels were performed as described in Järvi et al., 2011 with the modifications described in Bielczynski et al., 2016. The second dimension and the SDS-PAGE were performed as described in Schägger, 2006.

For immunoblot analysis, total protein extracts were separated by SDS-PAGE and transferred to a Protran 0.45 mm nitrocellulose membrane. Specific primary antibodies (Agrisera) were used to detect the target proteins. Chemiluminescence was detected using an ImageQuant LAS 4000 imaging system.

For sucrose density gradient fractionation, thylakoids equivalent to 0.2 mg total chlorophyll were washed with 5 mM EDTA and resuspended in 200 μL 10 mM Hepes (pH 7.5). An equal volume of 1.2% α-DDM was added, mixed gently, and the solubilized thylakoids were centrifuged at 14,000 rpm for 10 min at 4°C. The supernatant was loaded on a 0–1 M sucrose gradient (10 mM Hepes, pH 7.5, 0.03% α-DDM) and centrifuged at 288,000 g for 17 hr. The separated bands were collected with a syringe.

State-steady spectroscopy measurements

Absorption and CD

Request a detailed protocol

Absorption spectra were measured at room temperature with a Varian Cary 4000 UV-Vis-spectrophotometer. CD spectra were recorded using a Chirascan-Plus spectropolarimeter (Applied Photophysics) at 20°C. The OD of the samples was 0.8–1/cm at the maximum of the Qy region.

77 K fluorescence emission

Request a detailed protocol

Low-temperature fluorescence emission spectra were recorded using a Fluorolog 3.22 spectrofluorometer (Jobin Yvon-Spex). For 77 K measurements, a home-built liquid nitrogen-cooled device was used. The samples were excited at 440 nm and the fluorescence emission was detected in the 600–800 nm range. Excitation and emission slit widths were set to 3 nm.

All measurements were performed in the same buffers used for the sucrose gradients.

Photobleaching and protein stability

Photobleaching

Request a detailed protocol

The samples were diluted to an absorbance of around 0.8 at the maximum in the Qy region. The protein–pigment complexes were illuminated with white light (7100 μmol photons m−2 s−1) from a halogen lamp with optic fiber arm. After each interval, the cuvette was removed from the light beam, and the absorption spectra were recorded with a Varian Cary 4000 UV-Vis-spectrophotometer in the range between 600 and 750 nm (Croce et al., 1999).

Protein stability

Request a detailed protocol

The stability of the isolated complexes was tested by measuring the temperature denaturation curve as obtained by monitoring changes in the CD spectra in the Qy region while increasing the temperature from 20 to 90°C (Croce et al., 1999). A 400 μL sample with OD 0.8 at the maximum in the Qy region was used in this measurement.

In vivo time-resolved fluorescence measurements

Request a detailed protocol

Time-resolved fluorescence measurements on leaves were done using a time-correlated single photon-counting (TCSPC) setup as described previously (Chukhutsina et al., 2019). Excitation at 650 nm was used to excite Chl b preferentially. Detached plant leaves were placed between two glass plates and mounted in the rotation cuvette (diameter: 10 cm; thickness: 1 mm). The cuvette was rotated at 1400 rpm while oscillating sideways. Fluorescence was measured in a front-face arrangement from the upper side of the leaves. Time-resolved fluorescence decays were measured at multiple detection wavelengths (between 675 and 690 nm with a wavelength step of 5 nm, and between 700 and 760 nm with a maximal wavelength step of 10 nm). The measurements were done in the presence and in the absence of PSII photochemistry (open (F0) and closed (FM) states, respectively).

  1. F0 was measured in complete darkness after overnight dark adaptation. The repetition rate was then reduced by a Pulse Picker (Spectra Physics) from 40 to 0.8 MHz. The excitation power was 20 μW. Preliminary checks with different powers and repetition rates were done to ensure that the PSII reaction centers (RCs) remained indeed open during the measurement.

  2. To measure leaves with closed PSII RCs (FM), the leaves were incubated for 12 hr in sucrose (0.3 M) with addition of 50 μM 3-(3,4-dichlorophenyl)−1,1-dimethylurea (DCMU). To achieve full closure of the PSII RCs during the measurement, additional blue LED light of low intensity (∼50 μmol photons·m−2·s−1) was used to preilluminate leaves just before detection of the signal. The repetition rate was reduced by a Pulse Picker from 40 MHz to 4 MHz. The excitation power was 100 μW.

The measurement time at a single wavelength was limited to 10 min, to avoid changes in the leaves due to prolonged measurement in the rotating cuvette. To perform an experiment in one state took 2–3 hr. All in vivo measurements were performed at 20°C. The obtained fluorescence decay traces were analyzed globally with the ‘TRFA Data Processing Package’ of the Scientific Software Technologies Center (Belarusian State University, Minsk, Belarus) (Digris et al., 1999). The global analysis methodology is described in van Stokkum et al., 2004. In short, a number of parallel, non-interacting kinetic components was used as a kinetic model, so the total dataset was fitted with function f (t, λ) as follows:

1,2...NDASi(λ)exp(tτi)irf(t,λ),

where the decay-associated spectrum (DASi) is the amplitude factor associated with a decay component i having a decay lifetime τi, and irf(t, λ) was measured using scattering light. Typical full-width at half-maximum (FWHM) values were 28 ± 2 ps.

In vitro time-resolved fluorescence measurement

Request a detailed protocol

Time-resolved fluorescence measurements on isolated LHCII were performed on a FluoTime200 setup (Picoquant). The samples were diluted to an OD of 0.05 cm−1 at the maximum in the Qy region and measured in a 3.5 mL cuvette with a path length of 1 cm at 283 K. Excitation was provided by a 468 nm laser diode (preferential Chl b excitation) operating at 10 MHz repetition rate. The instrument response function (IRF) was obtained by measuring the decay of a pinacyanol iodide dye dissolved in methanol, which has a six ps fluorescence lifetime (van Oort et al., 2008). The resulting IRF FWHM was ∼88 ps. The fluorescence decay kinetics was detected at 680 nm with a channel time spacing of 8 ps. Data analysis was performed by the TRFA DATA software as described above.

ECS-based measurements

Request a detailed protocol

The ECS light-induced difference spectrum was determined according to Bailleul et al., 2015 using a JTS-10 spectrophotometer (BioLogic, Grenoble, France). In brief, the leaf was subjected to a saturating pulse of red light (3000 µmol photons m−2 s−1; 80 ms), and the absorption changes at each wavelength after the pulse were recorded without additional actinic light. The baseline obtained without the saturating pulse was subtracted, and the values between 100 and 200 ms after the pulse (to avoid the contribution of signals due to rapid redox-changes of cytochromes) were averaged. The obtained spectrum closely matches the theoretical ECS spectrum of pure astaxanthin, which is [1-(dfdx)] of astaxanthin-detergent solution spectrum (Bailleul et al., 2010).

The PSII:PSI RC ratio was determined using the JTS-10 spectrophotometer using saturating single-turnover laser flashes (five ns duration) provided by a dye laser pumped with a Nd:YAG laser (Minilite, Continuum) using the protocol described in Nawrocki et al., 2016 but adapted for leaves. For the PSII+PSI signal, the leaf was infiltrated with water, and to obtain a pure PSI signal the leaf was infiltrated with hydroxylamine (HA, 1 mM) and 3-(3,4-dichlorophenyl)−1,1-dimethylurea (DCMU, 10 µM; both from Sigma), after a systematic verification that no variable fluorescence, and thus PSII activity, remained in the leaf. ECS was detected at 546 nm (Asta) and 520–546 nm (WT) using weak white light LED pulses filtered with a 10 nm FWHM interference filter. The peak amplitude at 546 nm allows the detection at the isosbestic point of cyt. b6f haems (Alric et al., 2005). The functional antenna size was measured as described in Nawrocki et al., 2016 but with 300 µmol photos m−2 s−1 red actinic light (630 nm peak) and detecting light as described above. The quantities of active PSII were corrected for the ~20% slowly-opening RCs accumulating after actinic light.

Photosynthetic parameter measurements

PSI and PSII redox state

Request a detailed protocol

Dark-adapted plants were measured with a Dual-PAM-100 (Walz) to record qL and P700 at different actinic light intensities (70–1030 μmol photons·m−2·s−1). Leaves dark-adapted overnight were illuminated for 10 min to attain a steady state. The 820 nm absorbance signal corrected for absorbance changes at 870 nm was used for the analysis of the P700 kinetics. qL was calculated according to Kramer et al., 2004.

NPQ

Request a detailed protocol

Dark-adapted plants were measured by a Dual-PAM-100 (Walz) with a modulated measuring light of 7 μmol photons m−2 s−1 to keep the reaction centers in the open state, and a 4000 μmol photons m−2 s−1 (500 ms) saturating pulse to close the reaction centers. Actinic light of 531 μmol photons m−2 s−1 was used to induce NPQ. NPQ(T) was calculated as described in Tietz et al., 2017.

OJIP

Request a detailed protocol

A HandyPEA (Hansatech Instruments Ltd, UK) was used to measure fluorescence induction upon a dark-to-light transition. Three red LEDs (peak intensity at ~650 nm) were used as a light source, giving approximately 3500 µmol photons m−2 s−1 at the leaf surface. The HandyPEA measured the fluorescence intensity emitted in response to the actinic light. No measuring light was used and between pulses there was no light. To correct for differences in the fluorescence intensity due to differences in the actinic light intensity, the measured fluorescence values were divided by F0 ( = F20µs). Tobacco plants were taken from the growth chamber at the end of the night and were kept in near darkness for at least one hour before leaf clips were attached to a set of 10 leaves. These leaves were measured repetitively during the experiment with 10–15 min of darkness between measurements. Two types of measurements were carried out. To estimate the effective antenna size of PSII, leaves were illuminated with light intensities from 3500 down to 200 µmol photons m−2 s−1, starting at the highest light intensity. To characterize the reoxidation properties of PSII and the rest of the photosynthetic electron transport chain, two strong pulses of light (0.5 s, 3500 µmol photons m−2 s−1) were given to the leaves spaced Δt apart, with Δt between 0.1 and 200 s.

Data availability

All data used for this study are included in the manuscript or in the supporting information.

References

    1. Borth GA
    (1975)
    Treatment of traumatic injuries to the adult and deciduous dentition
    Journal of the Macomb Dental Society 13:8–10.
  1. Book
    1. Dall’Osto L
    2. Bassi R
    3. Ruban A
    (2014) Photoprotective Mechanisms: Carotenoids
    In: Theg SM, Wollman FA, editors. Plastid Biology. New York: Springer. pp. 393–435.
    https://doi.org/10.1007/978-1-4939-1136-3_15

Decision letter

  1. David M Kramer
    Reviewing Editor; Michigan State University, United States
  2. Christian S Hardtke
    Senior Editor; University of Lausanne, Switzerland
  3. David M Kramer
    Reviewer; Michigan State University, United States
  4. David Tiede
    Reviewer; Argonne National Laboratory, United States
  5. Alison Telfer
    Reviewer

In the interests of transparency, eLife publishes the most substantive revision requests and the accompanying author responses.

Acceptance summary:

As indicated by all the reviewers the work is very interesting and has broad implications for the importance of carotene and carotenoids in photosynthesis. The revised version addressed all the issues and nearly all the suggestions proposed by the reviewers. It is anticipated that this publication will inspire additional work that covers the remaining unknowns.

Decision letter after peer review:

Thank you for submitting your article "β-carotene is not essential for photosynthesis in higher plants" for consideration by eLife. Your article has been reviewed by three peer reviewers, including David M Kramer as the Reviewing Editor and Reviewer

#1, and the evaluation has been overseen by Christian Hardtke as the Senior Editor. The following individuals involved in review of your submission have agreed to reveal their identity: David Tiede (Reviewer #2); Alison Telfer (Reviewer #3).

The reviewers have discussed the reviews with one another and the Reviewing Editor has drafted this decision to help you prepare a revised submission.

We would like to draw your attention to changes in our revision policy that we have made in response to COVID-19 (https://elifesciences.org/articles/57162). Specifically, when editors judge that a submitted work as a whole belongs in eLife but that some conclusions require a modest amount of additional new data, as they do with your paper, we are asking that the manuscript be revised to either limit claims to those supported by data in hand, or to explicitly state that the relevant conclusions require additional supporting data.

Our expectation is that the authors will eventually carry out the additional experiments and report on how they affect the relevant conclusions either in a preprint on bioRxiv or medRxiv, or if appropriate, as a Research Advance in eLife, either of which would be linked to the original paper.

The reviews were enthusiastic about the novelty of this work, and its potential implications for understanding the roles of carotenoids in photosynthesis. However, there was a consensus that the claims are too over-reaching. It was not demonstrated that the transgenic plants are not fully devoid of β-carotone and that other carotenoid species were also displaced. Further, while they may be able to grow under lab conditions but would not compete in the field. It is also not clear if there was clear substitution of Asta for other carotenoids, or if certain sites were left empty. Finally, there is an unresolved issue regarding the identification of the NPQ process that was observed in the Asta plants. Is it really equivalent to qE? This past issue could be addressed with some relatively easy experiments. However, considering the difficulty in performing more extensive work, the reviewers feel that a modified version of the paper might be acceptable, if it more realistically reflects the possible interpretations of the results, and eliminates overstatements, including reworking the title and Discussion. More specifically, the focus might be on the surprisingly small effects of substitution of specific carotenoids for others, rather than the essentiality of one specific case.

Reviewer #1:

This manuscript describes the characterization of a very interesting transgenic tobacco plant in which astaxanthin (asta) synthesis has been introduced. Asta accumulates to the extent that it outcompetes most if not all of the native carotenoids in photosynthetic complexes. The surprising finding is that the photosynthetic apparatus appears to function relatively normally, calling into question that textbook view that β-carotene is "essential" for photosynthesis. Overall, this conclusion is quite important for the field, and would encourage the authors to emphasize the danger of using data from knockout mutants to make claims of "essentiality". Their data clearly shows that photosynthesis can operate while β-carotone is largely replaced with (emphasis on these three words) asta. On the other hand, the current text also makes an over-reaching claim by stating (in the title and throughout the text) that β-carotone is not needed for photosynthesis. There are two important points that need to be clearly stated. First, they do not really show that β-carotone is not needed. They could do this by rescuing a mutant that lacks β-carotone by over-expressing asta, but those results are not shown. Second, the title is not accurate because they, rather than showing β-carotone is not needed, a difference compound can substitute for it. That this is possible is, nevertheless, both important and interesting. Although framing the story in this way makes a nice, snappy title, it very much distracts from what is, in fact, interesting and important in this work. The component on NPQ is quite interesting, but could be improved with some additional experiments or analysis.

"These data not only show that PSII is stable in the absence of β-carotene, but also demonstrate that the complexes are functional even when some of the carotenoid-binding sites are empty, as indicated by the higher Chl/car ratio in the Asta complexes compared to the WT complexes."

I am not convinced that this conclusion can be stated with certainty. If the Asta is more weakly bound some of it could have been lost during isolation, resulting in the apparently empty sites.

"Interestingly, the altered carotenoid composition has virtually no effect on the binding and spectroscopic properties of the Chls associated with the LHCs and the PSII core, as revealed by the high similarity of absorption (Figure 3—figure supplement 3A) and circular dichroism (Figure 3—figure supplement 3B) spectra of WT and mutant complexes in the Chl absorption regions…"

This seems to be overstated. There are differences, so it depends on what the definition of "virtually no effect" is. It would be better to tone it down a bit. There are so many cases where knocking out a gene has no apparent effect in the lab, but has strong effects in the field.

"Our data show that the difference in composition between WT and mutant complexes has only a small impact (5-10 °C) on their stability (Figure 3A)."

Similarly, I question whether 5-10 °C is "only a small impact" on stability. The WT would be able to live quite happily after a transient temperature spike, as experienced in the real world, but the asta plant would almost certainly die.

"Next, we investigated the effect of the absence of carotenes on the light-harvesting and trapping properties of the photosynthetic complexes in vivo by performing time-resolved fluorescence measurements on intact leaves (Figure 4 and Figure 4—figure supplement 1)."

Seems to suggest that there are no carotenoids in the leaves, which is obviously not true. This is apparent in many places in the text, and even more so in in the title, which seems to imply that carotenoids are not needed. In fact, they are not gone, only largely (with emphasis) replaced by Asta. What about changing the wording to something like, "…we investigated the effects of outcompeting carotenes with Ast…"?

"We verified that the mutant plants exhibit an ECS signal and we determined its action spectrum, which agreed with the prediction that astaxanthin is solely responsible for this in vivo Stark effect (Figure 5—figure supplement 1)." The data presented in Figure 5—figure supplement 1 is not an action spectrum. It is the light-induced difference spectrum. An action spectrum would be interesting, and would say something about the efficiency of light harvesting by and exciton transfer from Asta to the reaction centers.

"Indeed, comparison of the steady-state photochemical yields of PSII and PSI revealed that, at all light intensities, in both WT and mutant plants, the balance between PSII and PSI photochemistry is maintained (Figure 5A and Figure 5—figure supplement 3), meaning that the plants are able to compensate for the strong decrease in the PSII functional antenna size by decreasing the PSI/PSII ratio."

I disagree with the interpretation of these results at several levels. First, the calculation of NPQ used in the work is assumes that FM (relative fluorescence yield measured in the dark) occurs in the absence of any NPQ. In fact, what is indicated as NPQ is inaccurate because the plants were "pre-quenched" by the presence of Asta, so the NPQ is strongly underestimated. This is, of course, acknowledged in the text, "Note that the difference in NPQ level (1.8 in the WT vs. 0.3 in the mutant) can be fully ascribed to the presence of the static quencher in the mutant, which strongly reduces the maximal fluorescence in both dark (FM) and light (FM' ) states." It would be interesting to try the NPQ(t) parameter, which does not make this assumption.

Second, in a steady-state of linear electron flow, the apparent quantum efficiencies of PSI and PSII must, by definition, be equal, so the figure is not surprising. For example, plots of φII versus PAR are very similar in wild type and mutants deficient in NPQ, e.g. npq4, which lacks the qE response. What is different is that in the wild type, NPQ plays a stronger role in decreasing φII, whereas in npq4, it is accumulation of electrons on QA. Given the data in Figure 5B, I would expect that something similar occurs in the Asta plants, and it would be important to show this.

Third, the qP parameter is not is not a good linear indicator of the redox state of QA, nor is the so-called φI parameter are linear indicators of the quantum efficiency of PSI. The qL parameter might be a more linear indicator of QA redox state and plot of the qL versus the fraction of P700 in its oxidized state might be a better approach.

"It is likely that the high amount of PsbS in the mutant (Figure 2—figure supplement 1) can compensate for the lack of the xanthophyll cycle, or that astaxanthin can also be responsible for the dynamic quenching." It was shown by Li et al. (2009, Plant Cell 21, 1798-1812) that, in mutants lacking the xanthophyll cycle, lutein could restore qE. This reference provides some precedent for the observed persistence of rapid NPQ in line lacking Z.

The results on NPQ are not very thorough and some additional experiments are merited and, with minimal effort, make the paper stronger. In past work, several assays have been used to establish if an observed NPQ is qE, including addition of nigericin to inhibit lumen pH changes, elimination of VDE or ZE activity (e.g. mutation or addition of DTT), mutation of PsbS, or observation of the characteristic shifted spectrum associated with the qE process. It is probably not practical to ask for generation of PsbS/VDE/ZE mutants in the Asta background. However, it is relatively easy to perform the other experiments, and the outcomes would be very helpful in interpreting the results. For instance, Li et al., 2009, see above, found that the substitution of Lut for Z led to a shift in the qE-associated absorbance signal, showing that the onset of qE in the modified line still involved alteration of the spectral properties of a bound carotenoid.

Reviewer #2:

This is an excellent manuscript provides new information on the in vivo function of β-carotene on the assembly and function of PSII, PSI and LHC in higher plants. The manuscript is remarkable because it provides very clear demonstration that, contrary to widely accepted convention, β-carotene is not essential for the assembly and function of PSII and other light-harvesting complexes in tobacco. The results provide an important counterpoint the only other available mutant data from Chlamydomonas. As such, this manuscript will be of significant impact and a reference milestone. The manuscript is nicely placed in scientifically context, expertly documented with a broad suite of diagnostic spectroscopic tools, and generally very well-written.

Reviewer #3:

This is a very interesting paper challenging the belief that carotenes are essential for the stabilisation of PSII and PSI core complexes and hence essential for photosynthetic activity in oxygenic organisms. Mutant tobacco plants, totally lacking β-carotene and the normal xanthophylls, but making the xanthophyll, astaxanthin instead, can grow photosynthetically, all be it at a slower rate than the wild type. The mutant is characterised in this paper and it is shown that both photosystems and their respective light harvesting complexes bind astaxanthin. The paper reads well and is suitable for publication after dealing with the following points.

Note – I have numbered the pages from the title page throughout the review.

1) Abstract – last sentence. The text of the paper does not seem to me to make this point.

2) Figure 1 and Table 1: The figure shows a very orange/red plant but it is said to be 20 weeks old – whereas the table says that mature plants (Materials and methods – 20 weeks) have a very similar Chl/Car ratio to the WT and so I would guess they would be more green. Surely the Asta plant in Figure 1 is younger than 21 weeks? However, astaxanthin maybe be red shifted compared to β-carotene and the normal xanthophylls and so all plants are much redder than the WT? Clarify text.

3) A notable result in the paper is that the Asta mutant in all complexes binds astaxanthin but nearly half the sites are empty in both the PSI and PSII core complexes of Asta whereas the LHCs have only ~20% of the sites unfilled in Asta compared to WT (Table 2). The fact that the plants grow under lab conditions does not mean that the substitution of carotene for a xanthophyll does not have a drastic effect on photosynthetic activity. This should be discussed in more detail than the brief mention in the Results and Discussion.

4) Mention is made that PSI complexes from the mutant are more stable than PSII complexes (Results and Discussion and Figure 3) but there is no mention that PSII is intrinsically less stable than PSI. It is well known that in PSII, carotenoids cannot protect by quenching singlet oxygen formed via P680 triplet because of the very high oxidising potential of PSII.

5) Results and Discussion and Table 3 re ratio PSII/PSI: The text is very confusing and not consistent. The data does not show that this ratio is the same whether measured by ECS or protein levels. ECS is 2.6 times higher while by protein levels the PSI/PSII ratio it is said to be 6 times lower (Results and Discussion). Discuss this difference and use the same ratio PSII/PSI when reporting data and discussing in text.

https://doi.org/10.7554/eLife.58984.sa1

Author response

Reviewer #1:

This manuscript describes the characterization of a very interesting transgenic tobacco plant in which astaxanthin (asta) synthesis has been introduced. Asta accumulates to the extent that it outcompetes most if not all of the native carotenoids in photosynthetic complexes. The surprising finding is that the photosynthetic apparatus appears to function relatively normally, calling into question that textbook view that β-carotene is "essential" for photosynthesis. Overall, this conclusion is quite important for the field, and would encourage the authors to emphasize the danger of using data from knockout mutants to make claims of "essentiality". Their data clearly shows that photosynthesis can operate while β-carotone is largely replaced with (emphasis on these three words) asta. On the other hand, the current text also makes an over-reaching claim by stating (in the title and throughout the text) that β-carotone is not needed for photosynthesis. There are two important points that need to be clearly stated. First, they do not really show that β-carotone is not needed. They could do this by rescuing a mutant that lacks β-carotone by over-expressing asta, but those results are not shown. Second, the title is not accurate because they, rather than showing β-carotone is not needed, a difference compound can substitute for it. That this is possible is, nevertheless, both important and interesting. Although framing the story in this way makes a nice, snappy title, it very much distracts from what is, in fact, interesting and important in this work. The component on NPQ is quite interesting, but could be improved with some additional experiments or analysis.

We are glad the reviewer finds our work important for the field. We have changed the title to “Photosynthesis without β-carotene”. We think this title captures the essence of our work. We agree that, unless we can demonstrate that there is not even only one molecule of β-carotene per plant, we cannot completely exclude its role in photosynthesis. However, the canonical role of β-carotene is linked to its association with PSI and PSII. What we can safely conclude is that, in our plants, the vast majority of the complexes does not contain any β-carotene molecules.

"These data not only show that PSII is stable in the absence of β-carotene, but also demonstrate that the complexes are functional even when some of the carotenoid-binding sites are empty, as indicated by the higher Chl/car ratio in the Asta complexes compared to the WT complexes."

I am not convinced that this conclusion can be stated with certainty. If the Asta is more weakly bound some of it could have been lost during isolation, resulting in the apparently empty sites.

We have now added a sentence to explain that we cannot exclude that astaxanthin is lost upon purification and we have also reworded this sentence to make it clearer that it refers to the isolated complex in which some of the binding sites are “empty” in the sense that they are not occupied by carotenoids. This is now also explicitly mentioned.

"Interestingly, the altered carotenoid composition has virtually no effect on the binding and spectroscopic properties of the Chls associated with the LHCs and the PSII core, as revealed by the high similarity of absorption (Figure 3—figure supplement 3A) and circular dichroism (Figure 3—figure supplement 3B) spectra of WT and mutant complexes in the Chl absorption regions…"

This seems to be overstated. There are differences, so it depends on what the definition of "virtually no effect" is. It would be better to tone it down a bit. There are so many cases where knocking out a gene has no apparent effect in the lab, but has strong effects in the field.

We have reformulated the sentence and we are referring the reader to the Figure 3—figure supplement 3 where we provide a more detailed explanation of the spectra.

"Our data show that the difference in composition between WT and mutant complexes has only a small impact (5-10 °C) on their stability (Figure 3A)."

Similarly, I question whether 5-10 °C is "only a small impact" on stability. The WT would be able to live quite happily after a transient temperature spike, as experienced in the real world, but the asta plant would almost certainly die.

We have rephrased this sentence to make it clearer that we measure isolated complexes and that the results of these measurements are only valid for the isolated complexes.

"Next, we investigated the effect of the absence of carotenes on the light-harvesting and trapping properties of the photosynthetic complexes in vivo by performing time-resolved fluorescence measurements on intact leaves (Figure 4 and Figure 4—figure supplement 1)."

Seems to suggest that there are no carotenoids in the leaves, which is obviously not true. This is apparent in many places in the text, and even more so in in the title, which seems to imply that carotenoids are not needed. In fact, they are not gone, only largely (with emphasis) replaced by Asta. What about changing the wording to something like, "…we investigated the effects of outcompeting carotenes with Ast…"?

We have changed the sentence into “Next, we investigated the effect of the substitution of carotenes with astaxanthin”…. We also changed a similar sentence at the end of the Discussion. We are mentioning throughout the manuscript that our complexes contain astaxanthin and the definition of “carotenes” is already introduced in the Abstract. The title mentioned explicitly β-carotene, and this was done to avoid creating confusion to readers who may not be familiar with the difference between “carotenoids” and “carotenes”.

"We verified that the mutant plants exhibit an ECS signal and we determined its action spectrum, which agreed with the prediction that astaxanthin is solely responsible for this in vivo Stark effect (Figure 5—figure supplement 1)." The data presented in Figure 5—figure supplement 1 is not an action spectrum. It is the light-induced difference spectrum. An action spectrum would be interesting, and would say something about the efficiency of light harvesting by and exciton transfer from Asta to the reaction centers.

We have reformulated the sentence as suggested.

"Indeed, comparison of the steady-state photochemical yields of PSII and PSI revealed that, at all light intensities, in both WT and mutant plants, the balance between PSII and PSI photochemistry is maintained (Figure 5A and Figure 5—figure supplement 3), meaning that the plants are able to compensate for the strong decrease in the PSII functional antenna size by decreasing the PSI/PSII ratio."

I disagree with the interpretation of these results at several levels. First, the calculation of NPQ used in the work is assumes that FM (relative fluorescence yield measured in the dark) occurs in the absence of any NPQ. In fact, what is indicated as NPQ is inaccurate because the plants were "pre-quenched" by the presence of Asta, so the NPQ is strongly underestimated. This is, of course, acknowledged in the text, "Note that the difference in NPQ level (1.8 in the WT vs. 0.3 in the mutant) can be fully ascribed to the presence of the static quencher in the mutant, which strongly reduces the maximal fluorescence in both dark (FM) and light (FM' ) states." It would be interesting to try the NPQ(t) parameter, which does not make this assumption.

We have now used the NPQ(t) parameter, which shows that the level is even higher in the mutant than in the WT, in agreement with the higher amount of PsbS in the former. The data are shown in Figure 5—figure supplement 7.

Second, in a steady-state of linear electron flow, the apparent quantum efficiencies of PSI and PSII must, by definition, be equal, so the figure is not surprising. For example, plots of φII versus PAR are very similar in wild type and mutants deficient in NPQ, e.g. npq4, which lacks the qE response. What is different is that in the wild type, NPQ plays a stronger role in decreasing φII, whereas in npq4, it is accumulation of electrons on QA. Given the data in Figure 5B, I would expect that something similar occurs in the Asta plants, and it would be important to show this.

We agree with the reviewer that the apparent quantum yields in steady-state (and in the linear electron flow regime) will respond in the same way to changes in the light intensity, as will the ETRs of PSII and PSI. This is demonstrated in our work: as the reviewer points out, global parameters relating to PSII (qL) and PSI (phi PSI) both show virtually identical light dependence in asta- and WT plants. Crucially, the similarity of NPQ(t) parameter amplitude and kinetics between asta and WT (Figure 5—figure supplement 7) shows that the presence of static quenching can fully explain the difference in apparent NPQ. In consequence, given that the asta plants exhibit an NPQ(t) of ~2.3 already in darkness, we share reviewer’s opinion that in steady-state, all other parameters unchanged (like in the npq4 example used), the static quenching would be by far the biggest reason for a change/decrease in the PSII ETR light intensity dependence. The reviewer observes that in the npq4 mutant more QA becomes reduced. In Figure 5—figure supplement 3 it is shown that in asta-plants this is not the case, in agreement with the strong static and NPQ quenching in these plants.

Instead, we substantiate by measuring the RC ratio- and antenna sizes per PS that the optimisation of photosynthesis in the mutant is achieved through the regulation of the relative stoichiometries of the photosystems. Unfortunately, this makes the situation significantly more complex than for WT-npq4. We think that quantification of the partitioning of the effect of NPQ and b6f turnover rate for this difference would be difficult due to the superposition of static quenching, changes in PS stoichiometries, altered antenna sizes etc.

Nonetheless, in our opinion such small differences in the qL/phi PSI/light intensity plots rather highlight how remarkably well the mutant adapts to its genetic engineering and accounts for the necessity to maintain PSII and PSI ETRs even, as well as the b6f ETR capacity such that it exhibits pH-dependence (and thus the extent of photosynthetic control) similar to this in the WT.

Third, the qP parameter is not is not a good linear indicator of the redox state of QA, nor is the so-called φI parameter are linear indicators of the quantum efficiency of PSI. The qL parameter might be a more linear indicator of QA redox state and plot of the qL versus the fraction of P700 in its oxidized state might be a better approach.

We have calculated qL and we now show it in Figure 5. We still show qP in the Figure 5—figure supplement 4, to highlight the similar dependence of the data to fPSI or PAR. The fractions of open/closed PSII centres following a pulse were calculated in Figure 5—figure supplement 5.

"It is likely that the high amount of PsbS in the mutant (Figure 2—figure supplement 1) can compensate for the lack of the xanthophyll cycle, or that astaxanthin can also be responsible for the dynamic quenching." It was shown by Li et al. (2009, Plant Cell 21, 1798-1812) that, in mutants lacking the xanthophyll cycle, lutein could restore qE. This reference provides some precedent for the observed persistence of rapid NPQ in line lacking Z.

Both lutein and zeaxanthin are explicitly mentioned in the Abstract and in the Results and Discussion. We have now added the reference to Li et al., 2009.

The results on NPQ are not very thorough and some additional experiments are merited and, with minimal effort, make the paper stronger. In past work, several assays have been used to establish if an observed NPQ is qE, including addition of nigericin to inhibit lumen pH changes, elimination of VDE or ZE activity (e.g. mutation or addition of DTT), mutation of PsbS, or observation of the characteristic shifted spectrum associated with the qE process. It is probably not practical to ask for generation of PsbS/VDE/ZE mutants in the Asta background. However, it is relatively easy to perform the other experiments, and the outcomes would be very helpful in interpreting the results. For instance, Li et al., 2009, see above, found that the substitution of Lut for Z led to a shift in the qE-associated absorbance signal, showing that the onset of qE in the modified line still involved alteration of the spectral properties of a bound carotenoid.

This is unfortunately not possible, because due to Covid-19, we had to empty all our growth chambers and it will take months before we will have plants at the right stage for these measurements. Nonetheless, we are confident that we observe qE, because the relaxation of NPQ is very rapid, with identical kinetics to the WT (see inset in Figure 5B) despite the presence of static quenching. We are also reporting now the NPQ(t) parameter, which shows that the induced quenching is substantial.

Reviewer #3:

This is a very interesting paper challenging the belief that carotenes are essential for the stabilisation of PSII and PSI core complexes and hence essential for photosynthetic activity in oxygenic organisms. Mutant tobacco plants, totally lacking β-carotene and the normal xanthophylls, but making the xanthophyll, astaxanthin instead, can grow photosynthetically, all be it at a slower rate than the wild type. The mutant is characterised in this paper and it is shown that both photosystems and their respective light harvesting complexes bind astaxanthin. The paper reads well and is suitable for publication after dealing with the following points.

Note – I have numbered the pages from the title page throughout the review.

1) Abstract – last sentence. The text of the paper does not seem to me to make this point.

We refer here to the large change in PSI/PSII ratio between WT and the mutant, which is discussed in the last part of the manuscript. We have added a sentence to make the link to the Abstract clearer.

2) Figure 1 and Table 1: The figure shows a very orange/red plant but it is said to be 20 weeks old – whereas the table says that mature plants (Materials and methods – 20 weeks) have a very similar Chl/Car ratio to the WT and so I would guess they would be more green. Surely the Asta plant in Figure 1 is younger than 21 weeks? However, astaxanthin maybe be red shifted compared to β-carotene and the normal xanthophylls and so all plants are much redder than the WT? Clarify text.

We are referring to the age of the leaves. The older leaves are greener than the young ones. This is now mentioned in the legend. We have also added Figure 1—figure supplement 1 with photographs of plants of different ages.

3) A notable result in the paper is that the Asta mutant in all complexes binds astaxanthin but nearly half the sites are empty in both the PSI and PSII core complexes of Asta whereas the LHCs have only ~20% of the sites unfilled in Asta compared to WT (Table 2). The fact that the plants grow under lab conditions does not mean that the substitution of carotene for a xanthophyll does not have a drastic effect on photosynthetic activity. This should be discussed in more detail than the brief mention in the Results and Discussion.

These results are based on purified complexes, we have now better clarified this point and mentioned that we cannot exclude that some of the astaxanthin molecules are lost during purification. We have also added a sentence to the Introduction, mentioning that these plants can grow without β-carotone at least in laboratory conditions.

4) Mention is made that PSI complexes from the mutant are more stable than PSII complexes (Results and Discussion and Figure 3) but there is no mention that PSII is intrinsically less stable than PSI. It is well known that in PSII, carotenoids cannot protect by quenching singlet oxygen formed via P680 triplet because of the very high oxidising potential of PSII.

This was indeed the point that we wanted to make: that PSI with few carotenoids is still better protected than PSII. We have now emphasized this point as suggested.

5) Results and Discussion and Table 3 re ratio PSII/PSI: The text is very confusing and not consistent. The data does not show that this ratio is the same whether measured by ECS or protein levels. ECS is 2.6 times higher while by protein levels the PSI/PSII ratio it is said to be 6 times lower (Results and Discussion). Discuss this difference and use the same ratio PSII/PSI when reporting data and discussing in text.

Thank you. We have now corrected it and discussed the difference.

https://doi.org/10.7554/eLife.58984.sa2

Article and author information

Author details

  1. Pengqi Xu

    Biophysics of Photosynthesis, Department of Physics and Astronomy, Faculty of Sciences, VU University Amsterdam and LaserLab Amsterdam, Amsterdam, Netherlands
    Contribution
    Data curation, Formal analysis, Investigation, Writing - original draft
    Competing interests
    No competing interests declared
  2. Volha U Chukhutsina

    Biophysics of Photosynthesis, Department of Physics and Astronomy, Faculty of Sciences, VU University Amsterdam and LaserLab Amsterdam, Amsterdam, Netherlands
    Contribution
    Formal analysis, Investigation, Methodology, Writing - original draft
    Competing interests
    No competing interests declared
  3. Wojciech J Nawrocki

    Biophysics of Photosynthesis, Department of Physics and Astronomy, Faculty of Sciences, VU University Amsterdam and LaserLab Amsterdam, Amsterdam, Netherlands
    Contribution
    Formal analysis, Investigation, Methodology
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0001-5124-3000
  4. Gert Schansker

    Biophysics of Photosynthesis, Department of Physics and Astronomy, Faculty of Sciences, VU University Amsterdam and LaserLab Amsterdam, Amsterdam, Netherlands
    Contribution
    Formal analysis, Investigation, Methodology, Writing - review and editing
    Competing interests
    No competing interests declared
  5. Ludwik W Bielczynski

    Biophysics of Photosynthesis, Department of Physics and Astronomy, Faculty of Sciences, VU University Amsterdam and LaserLab Amsterdam, Amsterdam, Netherlands
    Contribution
    Investigation
    Competing interests
    No competing interests declared
  6. Yinghong Lu

    Max Planck Institute of Molecular Plant Physiology, Potsdam-Golm, Germany
    Contribution
    Resources
    Competing interests
    No competing interests declared
  7. Daniel Karcher

    Max Planck Institute of Molecular Plant Physiology, Potsdam-Golm, Germany
    Contribution
    Resources
    Competing interests
    No competing interests declared
  8. Ralph Bock

    Max Planck Institute of Molecular Plant Physiology, Potsdam-Golm, Germany
    Contribution
    Conceptualization, Resources, Writing - review and editing
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0001-7502-6940
  9. Roberta Croce

    Biophysics of Photosynthesis, Department of Physics and Astronomy, Faculty of Sciences, VU University Amsterdam and LaserLab Amsterdam, Amsterdam, Netherlands
    Contribution
    Conceptualization, Supervision, Funding acquisition, Writing - original draft, Writing - review and editing
    For correspondence
    r.croce@vu.nl
    Competing interests
    No competing interests declared
    ORCID icon "This ORCID iD identifies the author of this article:" 0000-0003-3469-834X

Funding

Nederlandse Organisatie voor Wetenschappelijk Onderzoek (Vici)

  • Roberta Croce

H2020 European Research Council (ERC CON 281341)

  • Roberta Croce

H2020 European Research Council (ERC ADG 669982)

  • Ralph Bock

The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.

Acknowledgements

The authors thank Benjamin Bailleul for help with ECS spectra determination and Judith Schaefers for help with chromatography. This work was supported by De Nederlandse Organisatie voor Wetenschappelijk Onderzoek (NWO), Earth and Life Sciences (ALW), through a Vici grant and by the European Research Council through the ERC Consolidator grant 281341 (ASAP) (to RC), a grant from the European Research Council (ERC) under the European Union’s Horizon 2020 research, and innovation programme (ERC-ADG-2014; grant agreement 669982) to RB.

Senior Editor

  1. Christian S Hardtke, University of Lausanne, Switzerland

Reviewing Editor

  1. David M Kramer, Michigan State University, United States

Reviewers

  1. David M Kramer, Michigan State University, United States
  2. David Tiede, Argonne National Laboratory, United States
  3. Alison Telfer

Publication history

  1. Received: May 15, 2020
  2. Accepted: September 24, 2020
  3. Accepted Manuscript published: September 25, 2020 (version 1)
  4. Version of Record published: November 3, 2020 (version 2)

Copyright

© 2020, Xu et al.

This article is distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use and redistribution provided that the original author and source are credited.

Metrics

  • 3,215
    Page views
  • 544
    Downloads
  • 2
    Citations

Article citation count generated by polling the highest count across the following sources: Crossref, PubMed Central, Scopus.

Download links

A two-part list of links to download the article, or parts of the article, in various formats.

Downloads (link to download the article as PDF)

Download citations (links to download the citations from this article in formats compatible with various reference manager tools)

Open citations (links to open the citations from this article in various online reference manager services)

Further reading

    1. Plant Biology
    Yan Gong et al.
    Research Article Updated

    Asymmetric and self-renewing divisions build and pattern tissues. In the Arabidopsis stomatal lineage, asymmetric cell divisions, guided by polarly localized cortical proteins, generate most cells on the leaf surface. Systemic and environmental signals modify tissue development, but the mechanisms by which plants incorporate such cues to regulate asymmetric divisions are elusive. In a screen for modulators of cell polarity, we identified CONSTITUTIVE TRIPLE RESPONSE1, a negative regulator of ethylene signaling. We subsequently revealed antagonistic impacts of ethylene and glucose signaling on the self-renewing capacity of stomatal lineage stem cells. Quantitative analysis of cell polarity and fate dynamics showed that developmental information may be encoded in both the spatial and temporal asymmetries of polarity proteins. These results provide a framework for a mechanistic understanding of how nutritional status and environmental factors tune stem-cell behavior in the stomatal lineage, ultimately enabling flexibility in leaf size and cell-type composition.

    1. Plant Biology
    Philipp Schwenk et al.
    Research Article

    Phytochromes are photoreceptors regulating growth and development in plants. Using the model plant Arabidopsis, we identified a novel signalling pathway downstream of the far-red light-sensing phytochrome, phyA, that depends on the highly conserved CCR4-NOT complex. CCR4-NOT is integral to RNA metabolism in yeast and animals, but its function in plants is largely unknown. NOT9B, an Arabidopsis homologue of human CNOT9, is a component of the CCR4-NOT complex, and acts as negative regulator of phyA-specific light signalling when bound to NOT1, the scaffold protein of the complex. Light-activated phyA interacts with and displaces NOT9B from NOT1, suggesting a potential mechanism for light signalling through CCR4-NOT. ARGONAUTE 1 and proteins involved in splicing associate with NOT9B and we show that NOT9B is required for specific phyA-dependent alternative splicing events. Furthermore, association with nuclear localised ARGONAUTE 1 raises the possibility that NOT9B and CCR4-NOT are involved in phyA-modulated gene expression.