A Single Nucleotide Variation of CRS2 Affected the Establishment of Photosynthetic System in Rice

Chloroplasts are essential sites for plant photosynthesis, and the biogenesis of the photosynthetic complexes involves the interaction of nuclear genes and chloroplast genes. In this study, we identified a rice pale green leaf mutant, crs2. The crs2 mutant showed different degrees of low chlorophyll phenotypes at different growth stages, especially at the seedling stage. Fine mapping and DNA sequencing of crs2 revealed a single nucleotide substitution (G4120A) in the eighth exons of CRS2, causing a G-to-R mutation of the 229th amino acid of CRS2 (G229R). The results of complementation experiments confirmed that this single-base mutation in crs2 is responsible for the phenotype of the crs2 mutant. CRS2 encodes a chloroplast RNA splicing 2 protein localized in the chloroplast. Western blot results revealed an abnormality in the abundance of the photosynthesis-related protein in crs2. However, the mutation of CRS2 leads to the enhancement of antioxidant enzyme activity, which could reduce ROS levels. Meanwhile, with the release of Rubisco activity, the photosynthetic performance of crs2 was improved. In summary, the G229R mutation in CRS2 causes chloroplast protein abnormalities and affects photosystem performance in rice; the above findings facilitate the elucidation of the physiological mechanism of chloroplast proteins affecting photosynthesis.


Introduction
A chloroplast is a kind of semiautonomous organelle in plant cells and contains approximately 3000 different proteins [1]. The genome of chloroplasts in higher plants is a circular double-stranded structure without histones and the superhelix structure of the nuclear genome. Chloroplast genes can be divided into three categories according to their function: photosystem genes (psaA, psbA, atpA and rbcL, etc.), genesis-related genes (rrn, rpl, rps, trn, rpoA, rpoB, rpoC1 and rpoC2, etc.) and biosynthesis-related genes (ndhA-F, etc.) [2]. Chloroplast RNA requires RNA-binding proteins for RNA post-transcriptional modifications, such as intron splicing and RNA editing, before translation [3]. Many nuclear-encoded proteins involved in chloroplast RNA post-transcriptional modifications have been identified, such as the pentatricopeptide repeat (PPR) protein, chloroplast RNA splicing and ribosome maturation (CMM) domain-containing protein and ribonuclease III domain-containing protein. In rice, AL2 encodes a chloroplast splicing facilitator (CRS1) that, by way of its three CRM domains, is involved in the shearing of group I and II introns [4]. The CRM domain contains a conserved "GxxG" motif, which has the ability to bind RNA [5]. OsPPR6 not only affects ycf3 RNA splicing efficiency but is also involved in ndhB RNA editing. The mutation of OsPPR6 can lead to abnormal chloroplast development and death due to albinism [6]. excess sunlight energy in the form of heat through nonphotochemical quenching (NPQ) to prevent high light stress.
In this study, the chloroplast RNA splicing 2 (crs2) mutant was obtained via ethyl methane sulfonate (EMS)-treated Oryza sativa japonica cv. Shennong 9816 (SN9816). The leaves of the mutants showed a pale green phenotype, and the chlorophyll content was reduced. Map-based cloning and genetic complementation tests revealed that the crs2 mutation was responsible for the abnormal chlorophyll content phenotype. Western blot and qPCR analyses indicated that CRS2 affects the expression of chloroplast rRNA and the relative abundance of major photosynthetic proteins. In addition, the crs2 mutant experienced less thermal dissipation and higher electron transfer efficiency, leading to an increase in the net assimilation rate (An). In conclusion, our study provides a new way in which the regulatory effects of CRS2 on chloroplast development and photosystem performance in rice are revealed.

Phenotype of Mutant crs2
The crs2 mutant was obtained from an EMS mutant pool of the japonica rice cultivar SN9816. Under field conditions, the mutant crs2 leaves showed different degrees of pale green phenotypes. The difference in plant color was most obvious at the seedling stage and then gradually diminished ( Figure 1A-F). The chlorophyll content of crs2 decreased by 79.0% compared with the WT at the seedling stage, while only a 22.2% decrease could be observed at the full-heading stage (Figure 2A). The results of the chlorophyll content were consistent with the phenotypes of the materials in the field, and they indicated that leaf color changes of the crs2 mutants were associated with a low photosynthetic pigment content. dissipate excess sunlight energy in the form of heat through nonphotochemical qu (NPQ) to prevent high light stress.
In this study, the chloroplast RNA splicing 2 (crs2) mutant was obtained via et thane sulfonate (EMS)-treated Oryza sativa japonica cv. Shennong 9816 (SN981 leaves of the mutants showed a pale green phenotype, and the chlorophyll cont reduced. Map-based cloning and genetic complementation tests revealed that the c tation was responsible for the abnormal chlorophyll content phenotype. Western b qPCR analyses indicated that CRS2 affects the expression of chloroplast rRNA relative abundance of major photosynthetic proteins. In addition, the crs2 mutant enced less thermal dissipation and higher electron transfer efficiency, leading t crease in the net assimilation rate (An). In conclusion, our study provides a new which the regulatory effects of CRS2 on chloroplast development and photosyst formance in rice are revealed.

Phenotype of Mutant crs2
The crs2 mutant was obtained from an EMS mutant pool of the japonica rice SN9816. Under field conditions, the mutant crs2 leaves showed different degrees green phenotypes. The difference in plant color was most obvious at the seedlin and then gradually diminished ( Figure 1A-F). The chlorophyll content of crs2 de by 79.0% compared with the WT at the seedling stage, while only a 22.2% decreas be observed at the full-heading stage ( Figure 2A). The results of the chlorophyll were consistent with the phenotypes of the materials in the field, and they indica leaf color changes of the crs2 mutants were associated with a low photosynthetic p content.  CHLD and PORA, which are related to chlorophyll biosynthesis, did not change signifi Moreover, the expression levels of HEMA and CAO increased significantly (Figu Among those of chloroplast development-related genes, the expression levels of V1 an crs2 were significantly lower than those in the WT, while the expression levels of TCD5 and TRXZ were not significantly different between the WT and crs2 ( Figure 2B). However, the results of TEM showed that the mesophyll cells of the WT an plants presented normal chloroplasts with well-developed stacked grana, thylak mellae and thylakoid membranes at the tillering stage ( Figure 2C-H). There were nificant differences in the average thickness of the granum stacks or the num granum stacks per chloroplast between the WT and crs2 ( Figure S1). Although the e sion of the V1 and V2 genes in the crs2 mutant decreased, this probably did not chloroplast development.

Map-Based Cloning of CRS2 and Complementation Analysis
For genetic analysis, the crs2 mutant was crossed with the wild type, and the ob F1 generation hybrids showed normal green leaves. Among all 1137 F2 populatio vidual plants, 861 showed a normal phenotype, and 276 showed a mutant phenotyp ble S1), with a 3:1 ratio of trait isolation (χ 2 = 0.319). These results indicate that the mal pigment phenotype of crs2 is attributable to the inheritance of a single recessive To determine the physical location of the mutant gene, we constructed an F2 popu with crs2 and Kasalath as parents and performed a chromosomal linkage analysis individual plants with a mutant phenotype ( Figure 3A). After expanding the ma population to 212, crs2 was precisely mapped to a 53.9 kb interval between W7 a ( Figure 3B), in which there are seven open-reading frames (ORFs) ( Figure 3C). Then Although the chlorophyll content of crs2 decreased, the relative expression levels of CHLD and PORA, which are related to chlorophyll biosynthesis, did not change significantly. Moreover, the expression levels of HEMA and CAO increased significantly ( Figure 2B). Among those of chloroplast development-related genes, the expression levels of V1 and V2 in crs2 were significantly lower than those in the WT, while the expression levels of TCD5, WSL8 and TRXZ were not significantly different between the WT and crs2 ( Figure 2B).
However, the results of TEM showed that the mesophyll cells of the WT and crs2 plants presented normal chloroplasts with well-developed stacked grana, thylakoid lamellae and thylakoid membranes at the tillering stage ( Figure 2C-H). There were no significant differences in the average thickness of the granum stacks or the number of granum stacks per chloroplast between the WT and crs2 ( Figure S1). Although the expression of the V1 and V2 genes in the crs2 mutant decreased, this probably did not affect chloroplast development.

Map-Based Cloning of CRS2 and Complementation Analysis
For genetic analysis, the crs2 mutant was crossed with the wild type, and the obtained F 1 generation hybrids showed normal green leaves. Among all 1137 F 2 population individual plants, 861 showed a normal phenotype, and 276 showed a mutant phenotype (Table S1), with a 3:1 ratio of trait isolation (χ 2 = 0.319). These results indicate that the abnormal pigment phenotype of crs2 is attributable to the inheritance of a single recessive gene. To determine the physical location of the mutant gene, we constructed an F 2 population with crs2 and Kasalath as parents and performed a chromosomal linkage analysis on 94 individual plants with a mutant phenotype ( Figure 3A). After expanding the mapping population to 212, crs2 was precisely mapped to a 53.9 kb interval between W7 and W8 ( Figure 3B), in which there are seven open-reading frames (ORFs) ( Figure 3C). Then, these seven ORFs were sequenced, and only a single-nucleotide mutation (G4120A) was found at position 4120 bp from the ATG start codon in Os01g0132800, which resulted in the amino acid substitution of Gly229 to Arg229 (G229R) ( Figure 3D-F). To confirm whether the mutation of Os01g0132800 resulted in a mutant phenotype, we transferred the complementary vector containing the CRS2 genome sequence into the crs2 mutant and obtained a total of 19 positive transgenic plants, all of whose mutant phenotype was rescued ( Figure 3G). Pigment contents and gene expression levels were measured, and all of these parameters of the transgenic plants were restored to the WT levels ( Figure 3H,I). Thus, these results confirmed that Os01g0132800 corresponded to the CRS2 gene and that the single-nucleotide substitution of Os01g0132800 is responsible for the abnormal chloroplast translation of the crs2 mutant. The CRS2 gene comprises eight exons and encodes a chloroplast RNA splicing 2 protein composed of 259 amino acids ( Figure 3D). Only one peptidyl-tRNA hydrolase (PTH) domain was found via conserved domain analysis ( Figure 3F). seven ORFs were sequenced, and only a single-nucleotide mutation (G4120A) was found at position 4120 bp from the ATG start codon in Os01g0132800, which resulted in the amino acid substitution of Gly229 to Arg229 (G229R) ( Figure 3D-F). To confirm whether the mutation of Os01g0132800 resulted in a mutant phenotype, we transferred the complementary vector containing the CRS2 genome sequence into the crs2 mutant and obtained a total of 19 positive transgenic plants, all of whose mutant phenotype was rescued ( Figure 3G). Pigment contents and gene expression levels were measured, and all of these parameters of the transgenic plants were restored to the WT levels ( Figure 3H,I). Thus, these results confirmed that Os01g0132800 corresponded to the CRS2 gene and that the singlenucleotide substitution of Os01g0132800 is responsible for the abnormal chloroplast translation of the crs2 mutant. The CRS2 gene comprises eight exons and encodes a chloroplast RNA splicing 2 protein composed of 259 amino acids ( Figure 3D). Only one peptidyl-tRNA hydrolase (PTH) domain was found via conserved domain analysis ( Figure 3F).

Expression Pattern and Subcellular Localization of CRS2
To explore the expression pattern of CRS2 in WT plants, qRT-PCR analyses were performed on the roots, stems, leaves, leaf sheaths and panicles. In general, the CRS2 gene was expressed in all the aboveground parts of the rice plants. The highest expression part was found in the young leaves, followed by the mature leaves ( Figure 4A). Furthermore, a histochemical staining analysis of CRS2-GUS transgenic plants was conducted to confirm the expression pattern of CRS2. The CRS2 gene was expressed in the stems, leaves, leaf sheaths and panicles and was expressed mainly in the leaves, but no CRS2 expression was detected in the roots ( Figure 4B-E). These results were consistent with the qRT-PCR results.
formed on the roots, stems, leaves, leaf sheaths and panicles. In general, the CRS2 gen expressed in all the aboveground parts of the rice plants. The highest expression pa found in the young leaves, followed by the mature leaves ( Figure 4A). Furthermore, tochemical staining analysis of CRS2-GUS transgenic plants was conducted to confir expression pattern of CRS2. The CRS2 gene was expressed in the stems, leaves, leaf sh and panicles and was expressed mainly in the leaves, but no CRS2 expression was de in the roots ( Figure 4B-E). These results were consistent with the qRT-PCR results.
We constructed a subcellular localization vector with pCAMBIA1300-EGFP, uses the CaMV 35S promoter to drive the expression of the CRS2-GFP fusion protei ter transient expression in rice protoplasts, the green fluorescence signal of the CRS2 fusion protein coincided with that of the chloroplast, and empty GFP was present cytoplasm ( Figure 4F). These results confirm that CRS2 was localized in the chloro and was mainly expressed in green tissues.

Abnormal Chloroplast Proteins in crs2
Analysis of chloroplast intron splicing showed that crs2 intron splicing was n fected ( Figure S2), and the expression levels of 23S and 16S rRNA in the chloroplasts significantly lower in crs2 than in the WT ( Figure 5A). The qRT-PCR results showe the expression levels of psaA, psbA, petA and atpA encoded by the chloroplast geno crs2 were significantly higher than those in the WT, but the expression levels of lhcb lhcb2 encoded by nuclear genome were significantly lower ( Figure 5B). Western blot ysis revealed that the abundance of the PsaA, PsbA, PetA, AtpA, Lhcb1 and Lhcb teins in crs2 significantly decreased ( Figure 5C). We constructed a subcellular localization vector with pCAMBIA1300-EGFP, which uses the CaMV 35S promoter to drive the expression of the CRS2-GFP fusion protein. After transient expression in rice protoplasts, the green fluorescence signal of the CRS2-GFP fusion protein coincided with that of the chloroplast, and empty GFP was present in the cytoplasm ( Figure 4F). These results confirm that CRS2 was localized in the chloroplasts and was mainly expressed in green tissues.

Abnormal Chloroplast Proteins in crs2
Analysis of chloroplast intron splicing showed that crs2 intron splicing was not affected ( Figure S2), and the expression levels of 23S and 16S rRNA in the chloroplasts were significantly lower in crs2 than in the WT ( Figure 5A). The qRT-PCR results showed that the expression levels of psaA, psbA, petA and atpA encoded by the chloroplast genome in crs2 were significantly higher than those in the WT, but the expression levels of lhcb1 and lhcb2 encoded by nuclear genome were significantly lower ( Figure 5B). Western blot analysis revealed that the abundance of the PsaA, PsbA, PetA, AtpA, Lhcb1 and Lhcb2 proteins in crs2 significantly decreased ( Figure 5C).

Analysis of Chlorophyll Fluorescence
The reduction in chlorophyll content in the crs2 mutants not only resulted in the abnormal pigment phenotype but also affected the absorption and utilization of light energy in crs2. The chlorophyll fluorescence parameters under steady-state photosynthesis indicated that, compared with that of the WT, the Fv/Fm of crs2 did not change significantly; however, the quantum yield of PSII photochemistry (ΦPSII) and the photochemical quenching coefficient (qP) increased by 33.4 and 26.2%, respectively, and the NPQ decreased by 15.2% ( Figure 6A-D). The chlorophyll fluorescence imaging results were also consistent with the above results ( Figure 6E). The ΦPSII and qP of crs2 are significantly higher than that of the WT, while the NPQ is significantly decreased. The Fv/Fm images showed no significant difference between the WT and crs2. Taken together, these results indicated that the decrease in chlorophyll content in the mutant did not affect the photosynthesis potential at the full head stage. The light energy absorbed by the pigment in the antenna of the PSII reaction center was used more for photosynthetic electron transport, which might contribute to the release of the activity of PSII in the reaction center.

Analysis of Chlorophyll Fluorescence
The reduction in chlorophyll content in the crs2 mutants not only resulted in the abnormal pigment phenotype but also affected the absorption and utilization of light energy in crs2. The chlorophyll fluorescence parameters under steady-state photosynthesis indicated that, compared with that of the WT, the Fv/Fm of crs2 did not change significantly; however, the quantum yield of PSII photochemistry (ΦPSII) and the photochemical quenching coefficient (qP) increased by 33.4 and 26.2%, respectively, and the NPQ decreased by 15.2% ( Figure 6A-D). The chlorophyll fluorescence imaging results were also consistent with the above results ( Figure 6E). The ΦPSII and qP of crs2 are significantly higher than that of the WT, while the NPQ is significantly decreased. The Fv/Fm images showed no significant difference between the WT and crs2. Taken together, these results indicated that the decrease in chlorophyll content in the mutant did not affect the photosynthesis potential at the full head stage. The light energy absorbed by the pigment in the antenna of the PSII reaction center was used more for photosynthetic electron transport, which might contribute to the release of the activity of PSII in the reaction center.

Analysis of Chlorophyll Fluorescence
The reduction in chlorophyll content in the crs2 mutants not only resulted in the abnormal pigment phenotype but also affected the absorption and utilization of light energy in crs2. The chlorophyll fluorescence parameters under steady-state photosynthesis indicated that, compared with that of the WT, the Fv/Fm of crs2 did not change significantly; however, the quantum yield of PSII photochemistry (ΦPSII) and the photochemical quenching coefficient (qP) increased by 33.4 and 26.2%, respectively, and the NPQ decreased by 15.2% ( Figure 6A-D). The chlorophyll fluorescence imaging results were also consistent with the above results ( Figure 6E). The ΦPSII and qP of crs2 are significantly higher than that of the WT, while the NPQ is significantly decreased. The Fv/Fm images showed no significant difference between the WT and crs2. Taken together, these results indicated that the decrease in chlorophyll content in the mutant did not affect the photosynthesis potential at the full head stage. The light energy absorbed by the pigment in the antenna of the PSII reaction center was used more for photosynthetic electron transport, which might contribute to the release of the activity of PSII in the reaction center.

High Light Adaptability Comparison between WT and crs2
Compared with those of the WT, the chlorophyll content and photosynthesis-related protein abundance of crs2 decreased significantly, but the An, intercellular CO 2 concentration (Ci), stomatal conductance (gs) and transpiration rate (E) increased by 12.4, 7.8, 30.9 and 34.2%, respectively. In addition, the water use efficiency (WUE) decreased significantly, and there was no significant difference in carboxylation efficiency (CE) ( Figure 7A). In crs2, the activity of Rubisco and FBA enzymes were significantly increased and were 41.8 and 84.5% higher than those in the WT, respectively ( Figure 7B,C). The light response curve showed that the An of crs2 was significantly higher than that of the WT when the PPFD was ≥800 µmol·m −2 ·s −1 ( Figure 7D). For the antioxidant enzymes, the CAT and POD activities in crs2 significantly increased ( Figure 7I,J), while the SOD and APX activities were not significantly different ( Figure 7H

High Light Adaptability Comparison between WT and crs2
Compared with those of the WT, the chlorophyll content and photosynthesis-related protein abundance of crs2 decreased significantly, but the An, intercellular CO2 concentration (Ci), stomatal conductance (gs) and transpiration rate (E) increased by 12.4, 7.8, 30.9 and 34.2%, respectively. In addition, the water use efficiency (WUE) decreased significantly, and there was no significant difference in carboxylation efficiency (CE) ( Figure  7A). In crs2, the activity of Rubisco and FBA enzymes were significantly increased and were 41.8 and 84.5% higher than those in the WT, respectively ( Figure 7B,C). The light response curve showed that the An of crs2 was significantly higher than that of the WT when the PPFD was ≥800 μmol·m −2 ·s −1 ( Figure 7D). For the antioxidant enzymes, the CAT and POD activities in crs2 significantly increased ( Figure 7I,J), while the SOD and APX activities were not significantly different ( Figure 7H,K). The increase in antioxidant enzyme activity in the leaves of the crs2 mutant indicated that the mutant had a high antioxidant capacity and could reduce the ROS content in mesophyll cells, and compared with the WT, the mutant had a better protective effect on mesophyll cells and chloroplasts. In addition, compared with those in the WT, the content of O2 − , H2O2 and MDA in the crs2 mutant were reduced by 11.8, 14.2 and 26.1%, respectively ( Figure 7E-G). Therefore, crs2 can withstand ROS stress caused by high light intensity.

The G229R Mutation in CRS2 Is Responsible for the Decreased Chlorophyll Content
In this study, the crs2 mutant was obtained by EMS mutagenesis, whose leaf was pale green. Although the chlorophyll content of crs2 was decreased, the expression level of genes involved in chlorophyll synthesis was significantly increased, indicating that the abnormal chlorophyll content of crs2 was not caused by the obstruction of chlorophyll synthesis. Map-based cloning and gene complementation validation indicated that a single nucleotide substitution of crs2 caused the abnormal chlorophyll content of the mutant. CRS2 encodes a chloroplast RNA splicing 2 protein, which is one of the best-characterized chloroplast RNA splicing factors [23]. Jenkins et al., identified a T-DNA insertion mutant zmcrs2 in maize, whose group II intron splicing was abnormal, and the albino leaf phenotype appeared [23]. CRS2 evolved from a peptidyl tRNA hydrolase, but a neofunctionalization resulted in it becoming one of the most important and conserved splicing factors in plant chloroplasts. Due to the lack of RNA binding ability, CRS2 needs to interact with CAF1 and CAF2 and form a splicing complex to participate in RNA splicing [24]. Both CAF1 and CAF2 are CRM domain-containing proteins, in which the CRM domain contains the conserved GxxG sequence and has the ability of RNA binding [5]. Studies on rice crs2 mutants found that CRS2 is also involved in the splicing of a group I intron and chloroplast rRNA maturation [25]. Null mutants of CRS2 are seedling-lethal in maize rice, but the G229R mutation seems to be hypomorphic and has a normal mRNA splicing length and did not show seedling albino lethal phenotypes. In the present study, the G229R mutations led to impaired biogenesis of chloroplast ribosomes and abnormal chloroplast protein abundance. Chlorophyll can be combined with light-harvesting chlorophyll a/b-binding proteins to form light-trapping pigment-protein complexes that capture light energy and rapidly transfer it to the reaction centers to trigger photochemical reactions [26]. The abundance of light-harvesting a/b-binding proteins in crs2 decreases along with that of PsaA and PsbA, thus leading to a decrease in chlorophyll content.

The crs2 Mutant Shows Higher Photosynthetic Efficiency
Chlorophyll fluorescence is an important photosynthesis parameter that reflects the absorption and utilization of light energy by PSII [27]. It is generally believed that the decrease of chlorophyll content may impair the photosynthetic performance, but the results of photosynthetic performance studies in some related mutants are different. For example, the mutation of YGL53 seriously affected chlorophyll synthesis in rice, but the electron transfer ability of the ygl53 mutant was significantly improved, which activated the activity of PSII [28]. Therefore, the chlorophyll fluorescence parameters were further detected in this study. The Fv/Fm value indicates PSII's maximum potential for photosynthetic efficiency, while there was no significant difference observed between the WT and the mutant. ΦPSII represents the actual photochemical efficiency of PSII, and qP reflected the amount of energy consumed and the openness of the PSII reaction center. In this study, qP and ΦPSII were significantly increased in crs2 mutant, while it could be seen that NPQ was significantly decreased. NPQ represents the energy dissipated as heat in a non-photochemical quenching process. Some studies have demonstrated a negative correlation between ΦPSII and NPQ [29]. Similar to ygl53, the lower NPQ of the crs2 mutant indicated that, by reducing heat dissipation, the mutant may increase its electron transfer rate and light energy use efficiency. In addition, Rubisco and FBA are important rate-limiting enzymes in plant photosynthesis, and the improvement of the activities of Rubisco and FBA in crs2 can accelerate the rate of photosynthesis, thus improving the light energy using efficiency [30,31]. The results in the present study indicated that the crs2 mutant achieved a higher ΦPSII by increasing the qP to make better use of light energy, along with the reduction of non-photochemical quenching.

High Protective Enzyme Activity Is the Guarantee of crs2 to Cope with the Intensity of the High Light
It is widely believed that plant leaves tend to synthesize excess amounts of chlorophyll to build larger antennas that capture and absorb more light energy than that demanded by photosynthesis [32]. The excess light energy absorbed needs to be quenched by a variety of light energy dissipation mechanisms to avoid ROS stress and photoinhibition, avoiding the reduction in light energy conversion efficiency [33,34]. Both the chlorophyll content and the abundance of light-trapping antenna proteins in the crs2 mutant decreased, resulting in a smaller size of pigment antennas in the reaction center. Nonetheless, the measurement of photosynthetic indices showed that although the size of the pigment antenna of crs2 decreased, the net photosynthesis rate of crs2 increased significantly. This phenomenon was also observed in indica rice pe-1 mutants, which presented a decreased chlorophyll content but enhanced photosynthesis [35]. Therefore, the smaller pigment antenna of the crs2 mutant could still capture the light energy required for photosynthesis. The intact chloroplast structure indicated that crs2 had normal photosystem function. The increase in Rubisco activity ensured that the crs2 mutant had a high capacity for CO 2 fixation. As the rate-limiting enzyme of photosynthesis, Rubisco is involved in the first step of catalytic photosynthetic carbon assimilation, and Rubisco activity has a positive effect on the CO 2 assimilation rate of plants [36]. There are three pathways of photosynthetic electron transport in higher plants: noncyclic electron transport, cyclic electron transport and pseudocyclic electron transport [37]. In our study, the efficiency of electron transport in the crs2 mutant is improved through optimization of the use of light energy, and the photosystems capture more energy flowing as part of noncyclic electron transport. Due to the increase in ROS scavenging activity, the O 2 − and H 2 O 2 content in crs2 decreased substantially. Normal conditions lead to a balance between ROS generation and scavenging, which is harmless to plants [38]. However, ROS stress can lead to lipid peroxidation and MDA production in plants, and MDA content is an important index for reflecting the degree of oxidative damage to cells [39]. The smaller size of the light-capturing antenna and higher light energy use efficiency of crs2 can reduce the production of ROS, consequently reducing the level of membrane lipid peroxidation and thus reducing the MDA content. Moreover, the increased activities of CAT and POD also reflected the enhanced ROS-scavenging ability of crs2. On the other hand, the reduction in ROS levels and the enhancement of the scavenger system are important for crs2 to prevent ROS stress and photoinhibition under high light intensity. The increase in CAT activity could also enhance the activity of Rubisco [40] and further improve the carbon assimilation ability of crs2.

Plant Materials and Growth Conditions
To map crs2, we constructed an F 2 population derived from a cross of the crs2 mutant and the indica rice variety Kasalath. The seedlings of the materials were cultivated using thermal preservation and dry seedling-nursery technology. The seedlings were sown in mid-April and transplanted 30 days later. All the plants were transplanted under natural conditions into a paddy field of the Rice Research Institute, Shenyang Agricultural University (41 • N, 123 • E).

Measurement of Chlorophyll Content
Fresh leaf tissues (0.2 g) were soaked in 5 mL of 95% ethanol for 48 h in darkness.
The absorbance values of the supernatant at 470, 649 and 665 nm were measured using a Hitachi U5100 spectrophotometer (Hitachi, Tokyo, Japan), and the pigment content was calculated using the method of Arnon [41]. Three biological replicates were performed for each sample.

Transmission Electron Microscopy (TEM)
WT and crs2 fresh leaves were taken and cut into 0.25 cm 2 pieces. The cells were fixed in 3% glutaraldehyde solution and 1% OsO 4 successively and then embedded in resin after being subjected to a dehydration gradient. Then, the leaves were stained with uranyl acetate and alkaline lead citrate and observed under a transmission electron microscope (HITACHI HT7800, Hitachi); the magnifications used for observation were 1.5, 5.0 and 12.0 k×.

Measurement of Photosynthetic Gas Exchange and Chlorophyll Fluorescence Parameters
Photosynthesis and fluorescence parameters were determined on sunny day at full heading stage. Gas exchange parameters of flag leaves were measured using a portable photosynthesis system (CIRAS 3); the photosynthetic photon flux density (PPFD) was set to 1200 µmol·m −2 ·s −1 , and the CO 2 concentration was set to 390 ppm and stabilized using a CO 2 injector system. To determine the response of plants to different light intensities, the net assimilation rates of WT and crs2 were measured under 0, 50, 100, 200, 400, 600, 800, 1000, 1200, 1400, 1600, 1800 and 2000 µmol·m −2 ·s −1 actinic light. After dark treatment with leaf clips for 30 min, an FMS-2 pulse-modulated fluorometer (Hansatech, Norfolk, UK) was used to measure modulated chlorophyll fluorescence, and the chlorophyll fluorescence was observed using a FluorCam 7 chlorophyll fluorescence imaging system (Photon Systems Instruments, Brno, Czech Republic). Five independent individuals were measured for each sample.

Map-Based Cloning and Complementation of CRS2
We obtained 212 plants with mutant epitopes from the constructed F 2 population and extracted DNA for map-based cloning of CRS2. The genomic sequences of Nipponbare (japonica) and 93-11 (indica) were downloaded from the website of the Rice Genome Annotation Project and used to develop simple sequence repeat (SSR) and insertion-deletion (Indel) markers.
To complement the crs2 mutation, the 6734 bp genome sequence was amplified from DNA of wild-type SN9816 leaves and cloned into the binary expression vector pCAM-BIA1300 to construct a pCAMBIA1300-CRS2 vector. The recombinant vector was subsequently transferred into mutant callus by Agrobacterium tumefaciens EHA105. The sequences of all the primers used are listed in Supplementary Table S2.

RNA Extraction and Quantitative Real-Time PCR (qRT-PCR) Analysis
Total RNA was extracted from tissue using a MiniBEST Plant RNA Extraction Kit (9769S, Takara Bio, Inc., Kusatsu, Japan) according to the manufacturer's instructions. The cDNA was reverse-transcribed by using a PrimeScript RT Reagent Kit (RR047Q, Takara Bio, Inc.). qRT-PCR was subsequently performed using SYBR Premix Ex Taq II (RR820A, Takara Bio, Inc.) in an ABI 7500 Real-Time Fluorescent PCR System according to the manufacturer's instructions. The real-time PCR experiments were repeated for three biological replicates.
For chloroplast gene intron splicing analysis, the group I and group II introns of the WT and crs2 genes were amplified and isolated via 1% agarose gel electrophoresis. The primers used for intron splicing and qRT-PCR are listed in Supplementary Table S2. Three independent experiments were repeated for each sample.

Histochemical β-Glucuronidase (GUS) Assays
To conduct active staining of the transgenic plants, a 2123 bp region of the CRS2 promoter was amplified from SN9816 genomic DNA using PCR and ligated into a pCAM-BIA1305 vector such that it drove the expression of a GUS reporter gene. All the vectors were transformed into WT calli by Agrobacterium (EHA105)-mediated transformation. For GUS staining, we used excised tissues of independent T1 transgenic plants. The material was added to X-GluC buffer and incubated at 37 • C for 2 h. Afterward, the material was decolorized by soaking it in ethanol and imaged using a digital camera. The primers used for vector construction are listed in Supplementary Table S2.

Subcellular Localization of the CRS2 Protein
The full-length coding region of CRS2 was amplified using PCR and cloned into a pCAMBIA1300-EGFP vector to generate a p35S-CRS2-GFP fusion construct. The fusion construct was subsequently introduced into and transiently expressed in rice protoplasts. GFP signals were detected using a laser scanning confocal microscope (Zeiss, Oberkochen, Germany). The related primers used are listed in Supplementary Table S2.

Protein Extraction and Western Blotting
Total protein was extracted from rice leaf tissue via a plant protein extraction kit (BC3720, Beijing Solarbio Science & Technology Co., Ltd.), and the protein concentration was determined using the bicinchoninic acid assay (BCA) method. The total protein concentration of WT and crs2 was quantitatively consistent and denatured in boiling water bath for 5 min. The target protein was separated using SDS-PAGE and blotted onto a polyvinylidene difluoride (PVDF) membrane. After incubation with anti-PsaA (AS06 172, Agrisera, Vännäs, Sweden), anti-PsbA (AS05 084), anti-PetA (AS08 306) and anti-AtpA (AS08 304) antibodies, color development was performed using ECL Western Blotting Substrate (PE0010, Beijing Solarbio Science & Technology Co., Ltd.). The anti-β-actin antibody K800001M was used as an internal control.

Statistical Analysis
All of the data were statistically analyzed using Microsoft Excel 2016 (Microsoft Corporation, Washington, DC, USA) and SPSS 22 (SPSS Inc., Chicago, IL, USA). Student's t-test was conducted to compare WT and crs2, and * and ** represent significant differences at the 0.05 and 0.01 levels, respectively. GraphPad Prism 8 (GraphPad Software, Inc., San Diego, CA, USA) and Microsoft Excel were used to construct graphs.

Conclusions
This study indicated that CRS2 in rice encodes a chloroplast RNA splicing 2 protein located in the chloroplast. The missense mutation of crs2 resulted in an abnormality in the abundance of chloroplast photosystem complex proteins, thereby causing a decrease in chlorophyll contents. However, this phenomenon did not affect the chloroplast ultrastructure, and the photosynthesis of crs2 mutants was maintained at normal levels. Taken together, the crs2 mutation causes abnormalities in rice chloroplast proteins and affects photosystem performance. This research helps to elucidate the physiological mechanism of chloroplast proteins affecting photosynthesis.