LHCSR3-Type NPQ Prevents Photoinhibition and Slowed Growth under Fluctuating Light in Chlamydomonas reinhardtii

Natural light intensities can rise several orders of magnitude over subsecond time spans, posing a major challenge for photosynthesis. Fluctuating light tolerance in the green alga Chlamydomonas reinhardtii requires alternative electron pathways, but the role of nonphotochemical quenching (NPQ) is not known. Here, fluctuating light (10 min actinic light followed by 10 min darkness) led to significant increase in NPQ/qE-related proteins, LHCSR1 and LHCSR3, relative to constant light of the same subsaturating or saturating intensity. Elevated levels of LHCSR1/3 increased the ability of cells to safely dissipate excess light energy to heat (i.e., qE-type NPQ) during dark to light transition, as measured with chlorophyll fluorescence. The low qE phenotype of the npq4 mutant, which is unable to produce LHCSR3, was abolished under fluctuating light, showing that LHCSR1 alone enables very high levels of qE. Photosystem (PS) levels were also affected by light treatments; constant light led to lower PsbA levels and Fv/Fm values, while fluctuating light led to lower PsaA and maximum P700+ levels, indicating that constant and fluctuating light induced PSII and PSI photoinhibition, respectively. Under fluctuating light, npq4 suffered more PSI photoinhibition and significantly slower growth rates than parental wild type, whereas npq1 and npq2 mutants affected in xanthophyll carotenoid compositions had identical growth under fluctuating and constant light. Overall, LHCSR3 rather than total qE capacity or zeaxanthin is shown to be important in C. reinhardtii in tolerating fluctuating light, potentially via preventing PSI photoinhibition.


Introduction
Photosynthetic bacteria, algae, and plants are able to cope with rapid fluctuations in light intensity, although this requires significantly more regulation of photosynthesis than light of constant intensity [1,2]. Rates of CO 2 assimilation are limited and cannot always keep pace with rapid changes in light intensity that happen, e.g., as clouds pass across the sun or in sun spots under canopies. Nonphotochemical quenching (NPQ) is a general term for mechanisms that regulate how much light energy is available for photosynthesis [3]. NPQ is considered important in preventing elevated formation of reactive oxygen species (ROS) that otherwise could occur during excess light absorption [4,5]. The so-called qE component of NPQ, which reduces quantum yields of chlorophyll fluorescence in the light-harvesting antennae, is the dominant thermal dissipation pathway driven by pH changes in the thylakoid lumen [6][7][8]. In contrast to the constitutively high qE capacity found in many land plants, the fresh water alga Chlamydomonas reinhardtii has a flexible qE capacity that responds to the environment [9]. Under low light or high light when supplemented with organic carbon (both conditions that do not lead to excess absorption of light energy), the qE capacity of C. reinhardtii is minimal. Only under energetic imbalances, resulting from low CO 2 availability and excess light or exposure or photo-oxidative stress, qE capacity is increased to safely mitigate excess-absorbed light energy [10][11][12].
In C. reinhardtii, thermal dissipation of excess light energy via qE is intricately linked to Light-Harvesting Complex (LHC)-like Stress-Related (LHCSR) thylakoid membrane proteins, LHCSR1 and LHCSR3 [13], which have been lost in plants through evolution. LHCSR1-mediated quenching was shown to occur in LHCII [14,15], or energy transfer via LHCII to PSI [16], while LHCSR3-mediated qE occurs in PSII-LHCII-LHCSR3 supercomplexes [17,18] or in LHCII-LHCSR3 associated to PSI [14]. The pH-regulated xanthophyll cycle (e.g., de-epoxidation of violaxanthin to zeaxanthin) is a ubiquitous response of photosynthetic organisms to high light. Zeaxanthin in proximity to photosynthetic complexes is involved in the qE of plants and some algae, but is not required for high qE in C. reinhardtii [19][20][21]. Further to its role in qE, zeaxanthin has several other stress-associated roles (e.g., an antioxidant [5,21]). Tobacco mutants with an altered xanthophyll cycle and accelerated relaxation of qE had elevated photosynthetic efficiency in the field under naturally fluctuating light conditions, relative to WT, which was attributed to prevention of lost photosynthesis by excess dissipation during a decrease in light intensity [22]. Increasing photosynthetic efficiency via tweaking NPQ, and specifically qE, could be one way to elevate growth rates of lipid-rich algae, which are considered one of the most promising future sources of renewable biofuels [23].
Previously, tolerance to fluctuating light in C. reinhardtii has focused on cyclic electron flow via PGR5 or PGRL1 proteins, and the importance of maintaining the donor side of PSI oxidized via flavodiirons to prevent PSI photoinhibition [24,25]. In the diatom Phaeodactylum tricornutum, qE-type NPQ is part of the fluctuating light acclimation [26], but so far, the role of qE in fluctuating light tolerance in C. reinhardtii has not been elucidated. Here, I investigated how NPQ mechanisms may protect against fluctuating light, by comparing the growth and photosynthetic response of the npq4 mutant deficient in LHCSR3 to wild type (WT) and the npq1 and npq2 mutants with disrupted xanthophyll compositions.

Light Fluctuations, but Not Constant Light, Slowed npq4 Growth
Colony density of npq4 was visually much less than WT-4A when grown under repeated light fluctuations of 10 min exposure and 10 min darkness ( Figure 1). This agreed with increases in fresh weight of WT-4A being 2.5-fold and 6-fold higher than npq4 when fluctuating light intensities were 100 and 500 µmoL quanta m −2 s −1 , respectively, after 6-8 day growth ( Figure S1). Colony growth rates of the two genotypes were very similar under constant saturating (500 µmoL quanta m −2 s −1 ) and subsaturating (100 µmoL quanta m −2 s −1 ) light intensities ( Figure 1 and Figure S1) and also when darkness was replaced with low light intensity (50 µmoL quanta m −2 s −1 ; Figure S1).   Figure 3). This agreed with the higher Y(NPQ) values relative to those found in cells grown in constant light at the same intensity ( Figure 2B). Even at 100 µmoL quanta m −2 s −1 fluctuating light, Y(NPQ) of npq4 increased so much that it equaled WT-4A (Figures 1 and 2B). There was no consistent influence of initial culture dilution on Y(NPQ) values.

Enhanced PSII and PSI Photoinhibition Occurred in npq4 Under Constant and Fluctuating Light, Respectively
Values of F v /F m , here used as a proxy for PSII photoinhibition, were generally higher in WT-4A than npq4 under all light treatments. Notably, at 500 µmoL quanta m −2 s −1 , F v /F m values of npq4 were much lower under constant than fluctuating light (Figures 1 and 2A). This agreed with the lowest PsbA levels (PSII reaction center) found in npq4 under constant light and much higher levels in cells grown under fluctuating light ( Figure 3). However, under fluctuating light WT-4A and npq4 had lower levels of PsaA (PSI reaction center), relative to constant light-treated cells ( Figure 3 and Figure S2). This agrees with significantly altered maximum P700 redox changes (P m ), here used as a proxy for PSI photoinhibition. Lowered P m values were found in both genotypes under fluctuating light, with npq4 significantly lower than WT-4A ( Figure 4).

Xanthophyll Composition Did Not Affect Tolerance to Fluctuating Light
Under fluctuating light, npq4 contained significantly more (p = 0.02) antheraxanthin and slightly more zeaxanthin, and significantly less (p < 0.01) violaxanthin, showing higher activation of the xanthophyll cycle than in WT-4A ( Figure 5A). While it is known that de-epoxidized xanthophylls do not play a significant contribution to qE in C. reinhardtii [19], the xanthophyll cycle is still tightly coupled to high light exposure [27]. To investigate if differences in xanthophyll composition also influenced tolerance to fluctuating light, the NPQ mutants npq1 (no zeaxanthin) and npq2 (only zeaxanthin, and neither violaxanthin nor neoxanthin [27]) were included in the study. No differences in colony growth between npq1 and npq2 were observed under constant or fluctuating light ( Figure 5B), but similar to the other genotypes, fluctuating light increased qE capacity and led to higher F v /F m values relative to cells under constant light of the same intensity ( Figure 5B).

Discussion
Regulation of electron transport is of general importance for efficient photosynthesis and to prevent photoinhibition, especially under fluctuating light [1,28]. NPQ influences electron transport rates [20] and in plants has been shown to have both positive and negative impacts on plant growth, under constant and fluctuating light [22,29]. Growth of npq4 was only mildly affected under constant light (Figure 1), which is in agreement with other studies using LHCSR-deficient mutants at similar light intensities, at least under ambient oxygen tensions [10,30]. C. reinhardtii responded to fluctuating light by increasing levels of LHCSR1 and LHCSR3 (Figure 3), inferring that these qE proteins are important in the dynamic regulation of photosynthesis under fluctuating light. Surprisingly, Y(NPQ) values of npq4 were as high as WT under fluctuating light ( Figure 2B), which must have been strictly mediated by LHCSR1, due to the absence of LHCSR3 in this mutant [13].
LHCSR-mediated qE is dependent on pH [20] and becomes active in response to protonation of luminal-exposed part of the protein [7,8] that naturally occurs under high light. The xanthophyll cycle is another high light-associated process regulated by pH, but has limited contribution to qE of C. reinhardtii and other Chlorophyta alga [19,31]. Absence of an active xanthophyll cycle in the npq1 and npq2 mutants did not affect tolerance to fluctuating light ( Figure 5B). Therefore, higher levels of de-epoxidized xanthophylls in npq4 under fluctuating light compared to WT ( Figure 5A) is unlikely to have contributed to any change in fluctuating light tolerance. Nonetheless, a greater shift in the de-epoxidation ratio of npq4 is indicative of a lower luminal pH than occurred in WT. Since the level of LHCSR-mediated qE is directly associated to pH [20], low pH values may provide an explanation of the particularly high Y(NPQ) values in npq4 in response to fluctuating light ( Figure 2B). We can also be confident from the behavior of npq4 that LHCSR1 is able to induce a large, fast, and reversible pH-dependent quenching of LHCII [15] in response to fluctuating light.
Under fluctuating light, a lower luminal pH and elevated electric field across the thylakoid membrane (∆Ψ) increases incidences of charge recombination, ROS production, and PSII photoinhibition in Arabidopsis thaliana [32]. In contrast, here, F v /F m values of npq4 were higher under fluctuating light, relative to constant light (Figure 2A), indicating that fluctuating light did not induce PSII photoinhibition as much as under constant light. Indeed, higher levels of PsbA, the D1 reaction center of PSII, were likewise found under fluctuating than under constant light ( Figure 3 and Figure S2). PSII photoinhibition directly impacts linear electron flow, potentially influencing PSI photoinhibition [28].
So far, it has been shown that PGR5/PGRL1-mediated cyclic electron flow contributes to fluctuating light tolerance in photosynthetic organisms [25,33], but in C. reinhardtii, at least, the role of flavodiirons in preventing acceptor side limitation of PSI is more critical. In the absence of flavodiirons, fluctuating light leads to PSI photoinhibition [24,25]. This highlights that PSI can be rendered vulnerable under fluctuating light. Under constant light, enhanced PSI photoinhibition has been observed in npq4, but only under elevated oxygen tensions [10], or in cells also deficient in PGRL1-mediated cyclic electron flow [34,35]. Western blotting ( Figure 3) and P700 absorbance measurements (Figure 4) revealed that repeated 10 min light-dark treatments led to decrease in PSI levels, particularly in npq4. Photodamage of PSI can be exacerbated by high rates of electron flow from PSII [36]. Therefore, a lack of LHCSR3 and potentially low qE in npq4 could contribute to enhanced PSI photoinhibition. However, in fluctuating light, under which PSI photoinhibition of npq4 occurred, Y(NPQ) values were the same as in WT ( Figure 2B), indicating that the influence of qE on linear electron flow would have been equal in both genotypes.
Another explanation to enhanced PSI photoinhibition in npq4 under fluctuating light would concern state transitions. This phosphorylation-mediated NPQ mechanism [37] is particularly active in C. reinhardtii during the first few minutes of light-to-dark and dark-to-light acclimation and is affected in npq4 [38]. High light-treated cells are in state 1, but when subjected to darkness they transit to state 2 due to chlororespiration-induced phosphorylation of specific components of the antenna, including LHCII and LHCSR3 [39,40]. During transition to state 2, the majority of LHCII migrates energy transfer to PSI [41], and LHCSR3 also migrates as part of the LHCSR3-LHCII antenna of PSI [40]. Since LHCSR3 can quench LHCII associated with PSI [14], it is possible that LHCSR3 directly protects PSI from photodamage during fluctuating light, when cells are repeatedly exposed to high light in state 2 [38]. Unlike LHCSR3, LHCSR1 is not phosphorylated [34], and therefore unlikely to be in the mobile LHCII fraction during transition to state II. While the large increase in LHCSR1 levels ( Figure 3) suggests that LHCSR1-mediated qE is important under fluctuating light, without LHCSR3, as in npq4, LHCSR1 alone could not prevent PSI photoinhibition (Figures 4 and 5). High LHCSR1 levels may have even promoted PSI photoinhibition, since LHCSR1 has been reported to quench LHCII via PSI [16]. Finally, npq4 was not growth-sensitive to repeated 10-fold increases in light intensity (i.e., when darkness was replaced by low light; Figure S1). Therefore, deactivation of the Calvin-Benson cycle during 10 min of darkness [42] was likely important to reveal the sensitivity of npq4 to fluctuating light.

Conclusions
Overall, LHCSR3-mediated NPQ rather than overall qE capacity is important in tolerating fluctuating light that involves darkness. In agreement with previous studies, PSI showed vulnerability to fluctuating light, and with the use of npq4, it was possible to show that LHCSR3 can protect PSI, a role that LHCSR1 seems not to be able to fulfill. In contrast to the efficient repair cycle of PSII, photoinhibition of PSI is much more costly due to very slow PSI repair rates [43]. Therefore, enhanced photoinhibition of PSI due to absence of LHCSR3 can explain the growth phenotype of npq4 under fluctuating light. Interactions between LHCSR1/3 and LHCII with PSI in energy dissipation in response to dynamic light exposure require further elucidation.

C. reinhardtii Strains and Growth Conditions
The C. reinhardtii LHCSR3-deficient strain npq4 (CC-4614) and its parental WT-4A (CC-4051) were purchased from the Chlamydomonas Centre (www.chlamycollection.org). Liquid cultures were initiated in photoautotrophic media (THP), which was identical to TAP except acetic acid was replaced by HCl. Cultures were adjusted to 5.0, 1.0, and 0.5 µg chlorophyll mL −1 or 12,500, 2500, and 1250 cells (from here-on referred to as 1:1, 1:5, and 1:10 dilutions, respectively), in the 10 µL starting culture pipetted onto THP media containing 1.5% agar. Each Petri dish contained four replicates of each strain placed in an alternating order, with different plates hosting each dilution. The Petri dish lids were placed on, but not sealed with any film or tape to allow gas exchange, before placing in a growth chamber (Percival, PGC-6HO) at 25 • C under subsaturating or saturating light intensity (100 or 500 µmoL photons m −2 s −1 , respectively) with a 24 h time span composed of either constant 12 h illumination or repeated light fluctuations of 10 min illumination and 10 min darkness (see Figure S3 for a profile of the light intensity during one light fluctuation cycle). This light cycle was chosen to enable activation and relaxation of LHCSR3-associated NPQ processes [35]. For comparing growth rates, all colonies of the same genotype were carefully scraped, using a flat-ended spatula, together from the agar and the fresh weight was divided by the time colonies had been growing, which were 6, 7, and 8 days for 1:1, 1:5, and 1:10 dilutions, respectively. Additional cultures initiated from 1:1, 1:5, 1:10, and 1:25 dilutions were treated with fluctuating light as above, except that darkness was replaced by low light (50 µmoL photons m −2 s −1 ) to prevent deactivation of the Calvin-Benson cycle. All dilutions under this treatment were weighed after 7 days growth.

Chlorophyll Fluorescence
At the end of the 12 h constant light treatment on day 6, 7, and 8 for 1:1, 1:5, and 1:10 dilutions, respectively, cultures from both light treatments were moved to darkness for 1 h. After removing the Petri dish lid, chlorophyll fluorescence during a 600 ms saturating pulse was measured with a Maxi Imaging PAM M-series (WALZ). Minimum (F) and maximum fluorescence (F m ) was used to calculate maximum PSII quantum yields (F v /F m ) via (F m −F)/F m . Thereafter, Y(NPQ), as an indicator of the fraction of light energy dissipated to heat via qE, was measured after 30 s at 396 µmoL photons m −2 s −1 with a subsequent saturating pulse and calculated by (F/F m )−(F/F m ), with F m and F m measured before and during light, respectively. Immediately after, cultures were frozen in liquid nitrogen.

Western Blotting of Proteins
For detecting specific proteins via Western blotting, the frozen samples were thawed and extracted in 50-150 µL (depending upon culture amount) of 2% SDS in 50 mM TRIS-HCl, pH 6.8, containing protease inhibitor cocktail (Roche). After centrifugation for 10 min at 4 • C and 16,000 × g, supernatants were measured for protein content using the BCA assay for loading on 12% acrylamide gels at equal protein level (1 µg for PsbA and LHCSR3 and 10 µg for PsaA and LHCSR1; see Figure S2 for showing blots were below saturation point), before running for 1.5 h at 150 V, semi-dry transfer to nitrocellulose membranes, and incubation with antibodies (Agrisera), according to [10].

HPLC of Photosynthetic Pigments
Photosynthetic pigments were measured by HPLC in cultures grown from 1:10 dilution as above, except cells grown across the whole agar plate rather than from individual 10 µL spots (n = 3 plates/genotype). Pigments from ca. 2 mg of lyophilized cultures were extracted in 0.5 mL of ice-cold acetone and measured by absorbance at 440 nm, after separation with an Agilent 1100 HPLC system equipped with a LiChrospher 100 RP-18 column (125 mm × 4 mm, 5 µm), according to [11].

Maximum Photo-Oxidizable P700 Level
Maximum photo-oxidizable P700 levels (P m ) were measured during a 200 ms saturating pulse using a DUAL-PAM (Walz), of cultures grown as for pigment measurements (see Section 2.4). Cells were scraped off the agar and suspended in THP liquid media (see Section 2.1) at equal total chlorophyll concentration of 30 µg mL −1 , and measured according to Roach, Na, Stöggl, and Krieger-Liszkay [10].

Statistics
For chlorophyll fluorescence measurements, four individual colonies were used as replicates for each culture dilution. Data were analyzed by two-way ANOVA in SPSS Statistics 25 (IBM) to reveal p-values between WT and npq4, considering all dilutions collectively (n = 12 colonies) for each light treatment. For comparing colony growth, average fold-difference in fresh weight accumulation for the three culture dilutions (1:1, 1:5, and 1:10) was compared with a Students t-test under each light treatment. Significant differences were considered when p < 0.05.

Supplementary Materials:
The following are available online at http://www.mdpi.com/2223-7747/9/11/1604/s1. Figure S1: Colony growth comparison of LHCSR3 deficient nqq4 mutant relative to wild type (WT 4A) under various light treatments. (Cells were cultured under repeated 10 minute fluctuations (fl.) of light intensity between 0 (moon) and 100 or 500 µmol photons m −2 s −1 (sun), or diurnal (12/12h) constant light c on.) at the same two intensities. Data is shown as fold difference between fresh weight of all WT 4A and npq4 colonies at each initial culture dilution and after 6-8 days culturing (see methods). The average fold difference is depicted by a dashed line. (Growth of nqq4 and WT 4A under light fluctuating between 50 and 500 µmol photons m −2 s −1 (small sun, big sun). Representative images of colonies growing on agar are shown to the left of average colony fresh weight after 7 days of each initial culture dilution. Figure S2: Western blots of proteins from WT and npq4 cells under constant or fluctuating light at 500 µmol photons m −2 s −1 . For PsaA, proteins were loaded at 20 µg (200%) or 10 µg (100%) total protein, and for PsbA and LHCSR1, proteins were loaded at 2 µg (200%) or 1 µg (100%) total protein. All bands are from the same blot and transferred from the same gel. Figure S3: The change in light intensity during 10 min of fluctuating light, as measured with SQ 520 PAR sensor (Apogee Instruments).
Funding: This research received no external funding.
Acknowledgments: I thank Bettina Lehr and Birgit Stenzel (University of Innsbruck) for excellent technical assistance.

Conflicts of Interest:
The author declares no conflict of interest.