Involvement of Lhcb6 and Lhcb5 in Photosynthesis Regulation in Physcomitrella patens Response to Abiotic Stress

There are a number of highly conserved photosystem II light-harvesting antenna proteins in moss whose functions are unclear. Here, we investigated the involvement of chlorophyll-binding proteins, Lhcb6 and Lhcb5, in light-harvesting and photosynthesis regulation in Physcomitrella patens. Lhcb6 or Lhcb5 knock-out resulted in a disordered thylakoid arrangement, a decrease in the number of grana membranes, and an increase in the number of starch granule. The absence of Lhcb6 or Lhcb5 did not noticeably alter the electron transport rates. However, the non-photochemical quenching activity in the lhcb5 mutant was dramatically reduced when compared to wild-type or lhcb6 plants under abiotic stress. Lhcb5 plants were more sensitive to photo-inhibition, while lhcb6 plants showed little difference compared to the wild-type plants under high-light stress. Moreover, both mutants showed a growth malformation phenotype with reduced chlorophyll content in the gametophyte. These results suggested that Lhcb6 or Lhcb5 played a unique role in plant development, thylakoid organization, and photoprotection of PSII in Physcomitrella, especially when exposed to high light or osmotic environments.


Introduction
To harvest solar energy efficiently, photosynthetic organisms use groups of light-harvesting antenna proteins that bind carotenoids and chlorophylls. In higher plants, these proteins are composed of an inner and outer antenna. The photosystem (PS) II outer antenna consists of different heterotrimers of Lhcb1, Lhcb2, and Lhcb3, which is encoded by Lhcb1, Lhcb2, and Lhcb3 genes, while minor antenna complexes, such as chlorophyll protein 29 (CP29), CP26, and CP24 are encoded by Lhcb4, Lhcb5, and Lhcb6, respectively [1]. A structural analysis of an Lhcb super-complex and PSII organization has revealed that Lhcb5 (also named CP26) is located between the strongly bound LHCII trimer and dimeric PSII core complex, whereas Lhcb4 (also named CP29) is situated on the opposite side of the dimeric PSII core complex which is associated with moderately bound LHCII trimer and Lhcb6 (also named CP24) [2]. Additional LHCII trimers surround the PSII structure by connecting with Lhcb6 to the core of PSII in higher plants [3,4].
There are two mechanisms that regulate the absorption of light energy from Lhc (light-harvesting complex): (1) the light-harvesting antennas quench excess harvested light, which is known as non-photochemical quenching (NPQ) [5,6]; (2) state transition, which is a process associated with PSII

Knock-Out of Lhcb6 or Lhcb5 Altered Chloroplast Organization and Lhc Proteins Accumulation
Although the mutations of Lhcb6 and Lhcb5 in higher plants inhibit the interaction of photosystem II subunits and electron transport rate in grana membranes [22], little is known about the effects of minor antenna proteins in P. patens. To investigate the physiological function of minor antenna proteins, we generated Lhcb6 or Lhcb5 knock-out mutants in Physcomitrella. At least three independent lines were isolated and retained for further characterization. The expression of the gene level and accumulation of protein level were detected in mutants by qRT-PCR and Western blotting. As shown Figure 1B,C, mRNA and protein levels were hardly detected in the corresponding mutants when compared with wild-type. The alterations in Lhc stoichiometry in the thylakoid of mutants were verified by immunoblotting and quantitative PCR analysis ( Figure 2). In lhcb6 1 # , the expression levels of Lhc components Lhca and Lhcb4 were decreased sharply, while Lhcb5, PsbS and violaxanthin de-epoxidase (vde) were increased compared with wild-type plants. Apart from induced Lhcb6, all these genes in lhcb5 1 # displayed the same expression pattern as lhcb6 1 # . In addition, the LhcSR subunit content did not show obvious changes in lhcb6 1 # , but demonstrated an evident decline in lhcb5 1 # . The LhcSR subunit content was reduced in lhcb5 1 # but only changed by a small amount in lhcb6 1 # .
To investigate the effects of Lhcb6 and Lhcb5 on chloroplast organization, ultrathin sections of leaves were analyzed ( Figure 3). Under normal growth conditions, wild-type (WT) plant chloroplasts showed a characteristic organization of stromal membrane with interconnecting grana stacks and large starch granules in most sections. The lhcb6 1 # plants showed a lamella with a disordered arrangement of thylakoids, reduced stacked grana, and increased plastoglobules and starch granules. Chloroplasts from mutant lhcb5 1 # accumulated more starch granules and plastoglobules but displayed a higher ratio of stromal membrane to grana stacks when compared with WT. These results demonstrated that Lhcb6 and Lhcb5 influenced the ultrastructure of chloroplast and the accumulation of thylakoid proteins.

PSII Activity is Markedly Reduced in Mutants under High-Light Treatment
As the chloroplast organization was altered in mutants, we further analyzed PSII function in different mutants using fluorescence measurements. The peaks in the room temperature fluorescence emission spectra of mutants were lower when compared with wild-type at 685 nm (Supplementary Figure S1), suggesting that Lhcb6 and Lhcb5 affected the fluorescence emission spectra of several chlorophylls and the light-harvesting efficiency of PSII. To understand the primary functions of Lhcb6 and Lhcb5 in P. patens, non-invasive chlorophyll fluorometric analyses were performed to investigate the photosynthetic electron transport in wild-type and mutants ( Figure 4, Supplementary Figure S3). The initial Fv/Fm measurements demonstrated that PSII activity was disturbed in mutants and wild-type after high-light treatment ( Figure 4A). To further investigate the PSII activity of mutants under abiotic stress, the maximum (Fm) and minimum (F0) fluorescence, and nonphotochemical quenching (NPQ) of dark-adapted plants were quantitatively determined. Under normal growth conditions, lhcb5 1 # showed significantly decreased NPQ compared to wild-type ( Figure 4C).
The NPQ amplitude has been reported to be dependent on the lumen pH or on the concentration of PsbS [13,23]. Proton pumping into the chloroplast lumen was influenced in lhcb5 1 # rather than lhcb6 1 # and wild-type (Supplementary Figure S2). This result is consistent with the hypothesis that the limitation of NPQ in lhcb5 1 # is partly associated with reduced acidification of the lumen under illumination. The impact of the loss of Lhcb6 or Lhcb5 on photosynthesis was investigated by measuring the electron transport activity (ETA) in CO2-saturating conditions. The electron transport activity (ETA) in CO2 saturating conditions was lower in mutants compared with that in wild-type

PSII Activity is Markedly Reduced in Mutants under High-Light Treatment
As the chloroplast organization was altered in mutants, we further analyzed PSII function in different mutants using fluorescence measurements. The peaks in the room temperature fluorescence emission spectra of mutants were lower when compared with wild-type at 685 nm (Supplementary Figure S1), suggesting that Lhcb6 and Lhcb5 affected the fluorescence emission spectra of several chlorophylls and the light-harvesting efficiency of PSII. To understand the primary functions of Lhcb6 and Lhcb5 in P. patens, non-invasive chlorophyll fluorometric analyses were performed to investigate the photosynthetic electron transport in wild-type and mutants ( Figure 4, Supplementary Figure S3). The initial Fv/Fm measurements demonstrated that PSII activity was disturbed in mutants and wild-type after high-light treatment ( Figure 4A). To further investigate the PSII activity of mutants under abiotic stress, the maximum (Fm) and minimum (F 0 ) fluorescence, and non-photochemical quenching (NPQ) of dark-adapted plants were quantitatively determined. Under normal growth conditions, lhcb5 1 # showed significantly decreased NPQ compared to wild-type ( Figure 4C). Figure S2B). To further characterize the photosynthetic apparatus, the lightresponse curves of PSII quantum yield (Y(II)) and electron transport rate (ETR) were also analyzed. After 3 h of high-light treatment, Y(II) and ETR in lhcb5 1 # were significantly reduced compared with wild-type plants ( Figure 4A). These results indicated that the minor antenna protein Lhcb5 was involved in excitation energy via non-photochemical pathways.

PSII Complexes Accumulation Were Affected after High-Light Treatment
After a short-term high-light treatment, the proteins of the PSII reaction center were damaged. Then, a rapid repair and reassembly process occurred to enable photosynthetic electron transportation [24]. The defected photosynthesis in mutants was possibly caused by a reduced level of protein complexes in the electron transport chain. To further investigate the effects of Lhcb6 and Lhcb5 on the formation of thylakoid membrane protein complexes under abiotic stress, thylakoid membranes (with equal amounts of chlorophyll) in different plants were solubilized in 1% n-dodecylβ-D-maltoside (DM) and the chlorophyll-protein complexes were separated by BN-PAGE. Six protein complexes, PSII-LHCII super-complex, PSII dimer/PSI monomer, PSII core monomer, CP43less PSII core monomer, LHCII trimer, and unassembled protein were resolved ( Figure 5). Under high-light stress, the PSII core monomer of the lhcb5 1 # mutant was reduced, suggesting Lhcb5 might associate with the formation and stability of the PSII complex. Under osmotic stress, the amount of monomer and free pigment of mutants did not show significant differences compared with wild-type plants. These results suggest that light was a critical abiotic factor that limits the photosynthesis efficiency of PSII in Physcomitrella.

Growth Malformation Phenotype of Mutants
Even though there was no significant difference in the protonemal filaments between mutants and wild-type plants, the leafy gametophyte showed large differences. The leafy gametophyte of The NPQ amplitude has been reported to be dependent on the lumen pH or on the concentration of PsbS [13,23]. Proton pumping into the chloroplast lumen was influenced in lhcb5 1 # rather than lhcb6 1 # and wild-type (Supplementary Figure S2). This result is consistent with the hypothesis that the limitation of NPQ in lhcb5 1 # is partly associated with reduced acidification of the lumen under illumination. The impact of the loss of Lhcb6 or Lhcb5 on photosynthesis was investigated by measuring the electron transport activity (ETA) in CO 2 -saturating conditions. The electron transport activity (ETA) in CO 2 saturating conditions was lower in mutants compared with that in wild-type plants (Supplementary Figure S2B). To further characterize the photosynthetic apparatus, the light-response curves of PSII quantum yield (Y(II)) and electron transport rate (ETR) were also analyzed. After 3 h of high-light treatment, Y(II) and ETR in lhcb5 1 # were significantly reduced compared with wild-type plants ( Figure 4A). These results indicated that the minor antenna protein Lhcb5 was involved in excitation energy via non-photochemical pathways.

PSII Complexes Accumulation Were Affected after High-Light Treatment
After a short-term high-light treatment, the proteins of the PSII reaction center were damaged. Then, a rapid repair and reassembly process occurred to enable photosynthetic electron transportation [24]. The defected photosynthesis in mutants was possibly caused by a reduced level of protein complexes in the electron transport chain. To further investigate the effects of Lhcb6 and Lhcb5 on the formation of thylakoid membrane protein complexes under abiotic stress, thylakoid membranes (with equal amounts of chlorophyll) in different plants were solubilized in 1% n-dodecyl-β-D-maltoside (DM) and the chlorophyll-protein complexes were separated by BN-PAGE. Six protein complexes, PSII-LHCII super-complex, PSII dimer/PSI monomer, PSII core monomer, CP43-less PSII core monomer, LHCII trimer, and unassembled protein were resolved ( Figure 5). Under high-light stress, the PSII core monomer of the lhcb5 1 # mutant was reduced, suggesting Lhcb5 might associate with the formation and stability of the PSII complex. Under osmotic stress, the amount of monomer and free pigment of mutants did not show significant differences compared with wild-type plants. These results suggest that light was a critical abiotic factor that limits the photosynthesis efficiency of PSII in Physcomitrella.
lhcb6 plants grew slender and higher, with relatively narrower and longer leaves, while lhcb5 showed a more serious malformation phenotype when compared with lhcb6 or wild-type plants ( Figure 6A, Supplementary Figure S4A). In addition, the absence of Lhcb5 or Lhcb6 resulted in reduced lateral buds and chloroplast cells in protonema (Supplementary Figure S7). In addition, both mutants, especially the lhcb5 plants, displayed a dramatically reduced chlorophyll a+b content and a noticeably lower chlorophyll a/b ratio in comparison with wild-type plants ( Figure 6B,C, Supplementary Figure  S4B,C).
In summary, the Lhcb5 or Lhcb6 deletion mutant showed severe growth inhibition and developmental deficiency. The remarkable growth inhibition in different mutants suggested that Lhcb6 and Lhcb5 played essential roles in leaf development of Physcomitrella. Figure 5. Analysis of pigment-protein complexes in wild-type and mutants by BN-PAGE gel. A freshly isolated thylakoid membrane of wild-type and mutants were solubilized with 1% DM at a chlorophyll concentration of 25 μg. HL means high light stress for 3 h. M1 or M3 means plants under 500 mM mannitol for 1 day or 3 days, respectively. Assignments of the thylakoid membrane macromolecular protein complexes, indicated on the right, were identified according to a previous study [25]. Figure 5. Analysis of pigment-protein complexes in wild-type and mutants by BN-PAGE gel. A freshly isolated thylakoid membrane of wild-type and mutants were solubilized with 1% DM at a chlorophyll concentration of 25 µg. HL means high light stress for 3 h. M1 or M3 means plants under 500 mM mannitol for 1 day or 3 days, respectively. Assignments of the thylakoid membrane macromolecular protein complexes, indicated on the right, were identified according to a previous study [25].

Growth Malformation Phenotype of Mutants
Even though there was no significant difference in the protonemal filaments between mutants and wild-type plants, the leafy gametophyte showed large differences. The leafy gametophyte of lhcb6 plants grew slender and higher, with relatively narrower and longer leaves, while lhcb5 showed a more serious malformation phenotype when compared with lhcb6 or wild-type plants ( Figure 6A, Supplementary Figure S4A). In addition, the absence of Lhcb5 or Lhcb6 resulted in reduced lateral buds and chloroplast cells in protonema (Supplementary Figure S7). In addition, both mutants, especially the lhcb5 plants, displayed a dramatically reduced chlorophyll a+b content and a noticeably lower chlorophyll a/b ratio in comparison with wild-type plants ( Figure 6B,C, Supplementary Figure S4B,C).
In summary, the Lhcb5 or Lhcb6 deletion mutant showed severe growth inhibition and developmental deficiency. The remarkable growth inhibition in different mutants suggested that Lhcb6 and Lhcb5 played essential roles in leaf development of Physcomitrella. The protonema and gametophyte phenotypes of wild-type and mutants. The protonema tissues of wild-type and mutants were grown for one week on ammonium tartrate-free BCD medium. The protonema and gametophyte tissues were grown in this medium for four weeks. The gametophyte tissues were grown in the same medium for six-weeks. The chlorophyll a+b content (B) and chlorophyll a/b ratio (C) of leafy gametophyte tissues. FW, Fresh weight. The data represent the means ± SD of three biological replicates. Statistical significance compared with the wild-type is indicated by asterisks (** p ≤ 0.01, * p ≤ 0.05, Student's t-test).

Discussion
Light-harvesting complexes (Lhc) are members of a large multigene family. They play important roles in regulating photosynthesis in plants' response to environmental stress. The function of minor chlorophyll-binding proteins is in bridging major LHCII antenna to a dimeric PSII core complex [22,23,26]. Both Lhcb6 and Lhcb5 have a single energy transfer process from Chl b to Chl a [27]. Structural analysis of the Lhcb supercomplex and PSII organization has revealed that Lhcb5 is located between the strongly bound LHCII trimer and dimeric PSII core complex, whereas Lhcb6 is always associated with the connection of LHCII trimers and the core of PSII in higher plants [2][3][4]28]. The Physcomitrella has demonstrated a capacity for Lhc-dependent mechanisms in response to environmental conditions in previous research [17,29,30].
To investigate the functions of Lhcb6 and Lhcb5 in Physcomitrella, the PpLhcb6 and PpLhcb5 were inactivated by a targeted gene replacement assay. The mRNA or protein levels of Lhcb6 and Lhcb5 were undetectable in three independent mutation lines ( Figure 1). Moreover, the expression of Lhc components in both mutants were significantly changed ( Figure 2). Interestingly, the lhcb5 1 # showed an obvious increase in Lhcb6, PsbS and vde accumulation, in addition to decreased Lhca and Lhcb4 levels. The lhcb6 1 # mutant showed similar expression levels of pigment-protein components. Under high-light stress, the efficiency of excitation energy trapping and non-photochemical quenching in the absence of the lhcb5 mutant were much lower than WT plant, while the Lhcb6 deletion mutant only displayed a dramatic decrease in non-photochemical quenching (Figure 4). PsbS and LhcSR regulate excess energy dissipation by the xanthophyll cycle, which is an important anticipatory The protonema and gametophyte phenotypes of wild-type and mutants. The protonema tissues of wild-type and mutants were grown for one week on ammonium tartrate-free BCD medium. The protonema and gametophyte tissues were grown in this medium for four weeks. The gametophyte tissues were grown in the same medium for six-weeks. The chlorophyll a+b content (B) and chlorophyll a/b ratio (C) of leafy gametophyte tissues. FW, Fresh weight. The data represent the means ± SD of three biological replicates. Statistical significance compared with the wild-type is indicated by asterisks (** p ≤ 0.01, * p ≤ 0.05, Student's t-test).

Discussion
Light-harvesting complexes (Lhc) are members of a large multigene family. They play important roles in regulating photosynthesis in plants' response to environmental stress. The function of minor chlorophyll-binding proteins is in bridging major LHCII antenna to a dimeric PSII core complex [22,23,26]. Both Lhcb6 and Lhcb5 have a single energy transfer process from Chl b to Chl a [27]. Structural analysis of the Lhcb supercomplex and PSII organization has revealed that Lhcb5 is located between the strongly bound LHCII trimer and dimeric PSII core complex, whereas Lhcb6 is always associated with the connection of LHCII trimers and the core of PSII in higher plants [2][3][4]28]. The Physcomitrella has demonstrated a capacity for Lhc-dependent mechanisms in response to environmental conditions in previous research [17,29,30].
To investigate the functions of Lhcb6 and Lhcb5 in Physcomitrella, the PpLhcb6 and PpLhcb5 were inactivated by a targeted gene replacement assay. The mRNA or protein levels of Lhcb6 and Lhcb5 were undetectable in three independent mutation lines ( Figure 1). Moreover, the expression of Lhc components in both mutants were significantly changed ( Figure 2). Interestingly, the lhcb5 1 # showed an obvious increase in Lhcb6, PsbS and vde accumulation, in addition to decreased Lhca and Lhcb4 levels. The lhcb6 1 # mutant showed similar expression levels of pigment-protein components. Under high-light stress, the efficiency of excitation energy trapping and non-photochemical quenching in the absence of the lhcb5 mutant were much lower than WT plant, while the Lhcb6 deletion mutant only displayed a dramatic decrease in non-photochemical quenching (Figure 4). PsbS and LhcSR regulate excess energy dissipation by the xanthophyll cycle, which is an important anticipatory strategy against photo-inhibition [31]. Lhc proteins are associated with scavenging of reactive oxygen speciecs (ROS) species by switching back and forth between PSI and PSII. The direct quenching of chlorophyll triplet states by xanthophyll is associated with VDE and LhcSR [12,30,32]. The mechanisms which balance light absorption and the amount of electron transport acceptor substrates clarify the evolution of oxygenic photosynthesis [31]. Here, the different expression levels of Lhc genes observed in different mutants might be interpreted as compensatory mechanisms aimed at the dissipation of excess energy.
In higher plants, the minor antenna proteins Lhcb6 and Lhcb5 affect the interactions of PSII subunits and the electron transport rate in grana membranes, especially for limiting plastoquinone diffusion [22,23,26]. Here, the absence of either Lhcb5 or Lhcb6 influenced the ultrastructure of chloroplast and the accumulation of thylakoid proteins, and particularly, reduced the number of stacked grana and increased the amount of starch granules (Figure 3). The disordered arrangement of thylakoid lamella and increased starch granules suggested that Lhcb6 and Lhcb5 might maintain the chloroplast ultrastructure in vivo.
The structural analysis of the Lhc supercomplex and PSII organization revealed that Lhcb5 is located between the strongly bound LHCII trimer and dimeric PSII core complex, while Lhcb6 is associated with the connection of LHCII trimers and the core of PSII in higher plants [2][3][4]28]. The regulation of light-harvesting and energy transfer from LHCII to the PSII core does not require Lhcb5 [3,23]. In the absence of Lhcb5, the NPQ and Y(II) of chlorophyll fluorescence after 3 h of high-light treatment were lower than the plants lacking Lhcb6 (Figure 4, Supplementary Figure S6). After 3 h of high-light treatment or 3 d of osmotic treatment, lhcb5 1 # showed significantly reduced PSII core monomer protein levels in comparison to WT ( Figure 5). Under high-light stress conditions, the Lhcb5 deletion mutant showed reduced PSII core monomer protein levels, and decreased levels of PSII quantum yield, electron transport rate and NPQ, suggesting that the interaction between LHCII and the PSII super-complex was altered. These results implied the function of the minor antenna protein, Lhcb5, might be associated with the formation and stability of PSII super-complex assembly in the thylakoid membrane.
The results described above demonstrated that Lhcb6 and Lhcb5 participated in the accumulation of thylakoid proteins. These two minor antenna proteins probably interacted transiently with PSII complex or some PSII subunits, in a manner similar to several PSII auxiliary factors that interacted with PSII subunits. The cooperative interaction between the PSII core and minor Lhcbs is disrupted, which leads to a decreased affinity when Lhcb6 or Lhcb5 is in low amounts [22]. Furthermore, the inactivation of Lhcb6 or Lhcb5 displayed pleiotropic effects on photosynthetic electron transport. Chlorophyll fluorescence measurements showed that Fv/Fm was slightly reduced in mutants, suggesting no impairment of the PSII complexes (Figure 4), which is similar to previous evidence [33]. A dramatic decrease in NPQ indicated that the photoprotection of PSII was severely inhibited in mutants. These results demonstrated that minor antenna proteins Lhcb6 and Lhcb5 affected the chloroplast ultrastructure, the accumulation of PSII complexes and PSII activity under abiotic stress conditions. Under different abiotic stress conditions, chloroplasts, peroxisomes, and mitochondria accumulated a large amount of reactive oxygen species, which led to cell death and cytotoxicity. Multiple Lhc isoforms assemble with supramolecular photosynthetic complexes to dissipate excess absorbed energy and scavenging ROS. Under high-light stress, Lhc components of the PSII antenna system use non-photochemical quenching to limit the over-reduction of the electron transport chain and the formation of toxic ROS [34][35][36][37]. PsbS and LhcSR regulate the excess energy dissipation by the xanthophyll cycle, which is an important anticipatory strategy in the fight against photo-inhibition [31]. Lhc proteins associate with ROS scavenging by switching back and forth between PSI and PSII, and the direct quenching of chlorophyll triplet states by xanthophyll is associated with VDE and LhcSR [12,30,32]. Compared with the vascular plants, a stronger NPQ and higher capacity of dissipating excitation energy as heat were observed in mosses [17,20,30]. In Physcomitrella, it has been shown that NPQ requires LhcsR and PsbS to regulate excess light energy absorption and dissipate light damage [17,20,30,38]. NPQ is also viewed as a compensatory mechanism in response to overexcited chlorophyll to prevent irreversible damage to PSII [39]. In our study, the mutants showed clear differences in the Lhcb polypeptide composition, especially in PsbS and VDE. The expression of PsbS and vde were increased, while Lhca, Lhcb4, and LhcSR were decreased in the Lhcb5-defective mutant. The alteration of chloroplast organization and photosynthesis efficiency suggested that the absence of Lhcb5 might alter the structure or stability of PSII super-complexes (Figure 3). NPQ activation, in response to illumination with strong actinic light-of dark-adapted plants, was lower in lhcb5 1 # , especially under high-light stress (Figure 4). The distinctly decreased NPQ suggested that the absence of Lhcb5 resulted in the obstruction of excess light energy transfer between LhcII trimer and PSII core complexes. PsbS and xanthophyll control the affinity of different LhcII antenna complexes for protons to participate in the feed-back control of excess light energy that underlies non-photochemical chlorophyll fluorescence quenching [40]. Although the individual antenna components were not quantified in the analyses on the accumulation of PSII complexes, the chlorophyll a+b contents and chlorophyll a/b ratios of mutants suggested there may be some minor changes in the antenna in response to high-light treatment. Similarly, the distinguishable electron transport reflected the lower rate of photosynthetic electron transport in mutants (Supplementary Figure S2). An alternative explanation might reflect the role of Lhcb6 and Lhcb5 in channeling prtons away from PSII due to different pH levels in these plants [3,22]. Partial electron transport reactions localized at the restricted step between the plastoquinone site of PSII and the cytochrome b 6 f complex for electron donors to cytochrome b 6 f are effective in sustaining NADP + reduction [22]. A lack of Lhc proteins might result in the restriction of PQH diffusion from the PSII Q B site to the cytochrome b 6 f complex.
In our findings, the absence of Lhcb5 or Lhcb6 resulted in similar growth-defect phenotypes, such as reduced lateral buds and chloroplast cells in protonema (chloronema and caulonema) and thinner leaves in mature plants ( Figure 6, Supplementary Figures S4 and S7). Interestingly, the mutation of Lhcb5 contributed to a more significant growth-defective phenotype when compared with Lhcb6. The phenotype differences in protonema and gametophyte between WT and mutants implied that Lhcb5 may function slightly differently from Lhcb6. Lacking both genes contributed to a decreased growth ratio in gametophyte, but not in protonema ( Figure 6C, Supplementary Figure S5). Moreover, both mutants also showed disordered lamella arrangements of thylakoid, and abnormal extension of starch granules and plastoglobules in chloroplast, especially in lhcb5 (Figure 3). Excess sunlight damages photosynthetic machinery and limits plant photosynthetic activity, growth, and productivity [39]. Spectroscopic characterization showed that the light-harvesting efficiency decreased more in mutants (Supplementary Figure S1). Under high-light stress, the NPQ of lhcb6 and lhcb5 decreased sharply, implying that the antenna proteins might participate in photo-protection by dissipating excess energy. Under abiotic stress, only lhcb5 mutant showed any inhibition of NPQ, Y(II) and ETR, suggesting that the minor antenna protein, Lhcb5, played an important role in the excitation energy via non-photochemical pathways. The lower photo-inhibition PSII activity in Lhcb-defective mutants might connect with the limited growth rate and malformed phenotype in gametophyte. These results indicated that minor antenna proteins, Lhcb6 and Lhcb5, were involved in the gametophyte development in Physcomitrella. Light may be the critical factor in influencing Physcomitrella to diverge from seed and hydrophyte plants after land colonization.
In high-light conditions, Lhcb5 of Arabidopsis is activated in Zea-mediated photoprotection and catalyzed quenching of the qI type, while it is not phosphorylated in the state transition [14,22,23]. It has been reported that the deletion of Lhcb5 does not cause alterations in several photosynthetic parameters, except in reduced growth [3,22,23]. In the present study, we found the minor antenna proteins Lhcb5 and Lhcb6 of Physcomitrella played an important role in the organization of photosynthetic complexes in grana partitions, participated in electron transport, and excess excitation energy dissipation under high-light stress in higher plants. Moreover, the loss of minor antenna protein Lhcb6 or Lhcb5 resulted in a pale-green leaf and the developmental defect phenotype, which has not been reported in higher plants or other green alga. Function analysis of minor antenna proteins, Lhcb6 and Lhcb5 in P. patens, provides evolutionary insights for the land plants.

Plant Materials and Treatments
The moss P. patens were cultured on modified BCD medium [41] in a growth chamber at 23 ± 1 • C with 16 h of light (55 µmol m −2 s −1 ) and 8 h of dark control light conditions. Uniform leafy shoots and protonema were obtained as described [42]. Two-week-old gametophore colonies were gently ground with a homogenizer, and transferred the homogenate to a cellophane overlay on solid BCD medium, containing 0.75% (w/v) agar, with 5 mM ammonium tartrate, and 0.5% (w/v) glucose. After a week, the cellophane overlay bearing regenerated protonema tissues was transferred onto ammonium tartrate-free BCD medium. The extended leaves and rhizoids were obtained for experimental analysis after three weeks. For light treatments, 5-day-old plants were transferred to 450 µmol m −2 s −1 for 3 h (HL) [20]. For osmotic treatments, the plants were transferred onto fresh agar plates of BCD medium containing 500 mM mannitol [43] for 1 day and 3 days, respectively (M1 and M3).

Protoplast Transformation and Mutant Identification
Genomic P. patens protonema DNA extraction was performed as previously described in [17], and used as a template for PpLhcb5 (locus XM-001752760) and PpLhcb6 (locus XM-001757486) cloning. Two homologous regions (upstream and downstream) of target coding sequences were amplified by PCR from the cDNA library of P. patens tissue and sub-cloned into pTN182 vector ( Figure 1A) [44]. The transformation followed, as described previously in [41], with minor modifications. One-week-old protonemal tissues were collected for protoplast generation and PEG-mediated transformation. The selection and generation of resistant colonies were described previously [17]. The defects of Lhcb5 and Lhcb6 mutants were studied in three independent lines and were parallel. Quantitative PCR and immunoblotting showed that the genes or proteins were absent in both mutants. The primers used in this study were listed in Table S1.

Chlorophyll Content Measurement and EM Scanning
Leaf total chlorophyll was extracted with 80% acetone from two-week-old tissues and measured as previously described [22]. Samples from one-month-old tissues were fixed overnight in 3% glutaraldehyde and 0.1 M sodium cacodylate buffer (pH 6.9) at 4 • C. Ultrathin sections were cut with an ultramicrotome (Reichert-Jung, Ultracut, Wetzlar, Germany) and observed with a transmission electron microscope (FEI Tecnai G2 F20 S-TWIN, Hillsboro, OR, USA) operating at 75 kV [45].

Fluorescence Spectrum Analysis
The fluorescence emission spectrum was performed at room temperature using 436 nm excitation, 5 nm spectral bandwidth and 1 nm spectral resolution with a Hitachi F-4500 spectrofluorometer [46]. Samples were diluted at a chlorophyll concentration of 10 µg/mL in 20 mM Hepes/KOH, pH 7.8, 0.33 M sorbitol, 5 mM MgCl 2 and 70% glycerol (w/v).

Measurement of ∆pH
The kinetics of ∆pH formation across the thylakoid membrane were measured as previously described in [22]. The reaction buffer was as follows: 30 mM Tricine/NaOH, pH 7.8, 0.1 M sorbitol, 5 mM MgCl 2 , 10 mM NaCl, 20 mM KCl, 100 mM methyl viologen, and 2 mM 9-aminoacridine. The chlorophyll concentration in the reaction buffer was 30 µg/mL.

Measurement of Growth Rate
Analysis and identification of gametophyte and protonema tissue of Physcomitrella were photographed directly in their growth media as previously described in [51]. The live-image acquisition of protonema and gametophyte were performed with a light microscope (Nikon DS-Fi1c, Minato, Japan) and a digital camera (Leica DFC495, Wetzlar, Germany), respectively.
The growth-rate analysis of gametophyte in mutants and wild-type plants was carried out as previously described in [52]. The relative growth rate was estimated by measuring the fresh and dry weight of petri dish tissues for the same growth time.

Statistical Analysis
A statistical analysis of the results was carried out from experiments with three or more mean values using one-way analysis of variance (ANOVA). A difference was considered to be statistically significant when p ≤ 0.05 or very significant when p ≤ 0.01.
Supplementary Materials: Supplementary materials can be found at http://www.mdpi.com/1422-0067/20/15/3665/ s1. Figure S1: Spectroscopic characterization of wild-type and mutants; Figure S2: Measurement of trans-thylakoid ∆pH and the activity of the electron transport chain in wild-type and mutants; Figure S3: Time courses for induction and relaxation of NPQ of wild-type and mutants under 500 mM mannitol for 1 and 3 days; Figure S4: The protonema and gametophyte tissues of wild-type and mutants; Figure S5: The relative growth rate (RGR) of wild-type and mutants; Figure S6: Chlorophyll fluorescence parameters of WT and mutants; Figure S7: Micrographs of protonema (chloronema and caulonema) cells and gametophyte of WT and mutants; Table S1: Primers used in this study.
Author Contributions: X.P., D.Z., and H.L. designed the experiments; X.P., X.D. and X.T. performed major experiments of the study; T.T. and B.L. provided assistance in analyzing the data; X.P., X.D., D.Z. and H.L. wrote the manuscript and contributed to discussions.