The Rice Pentatricopeptide Repeat Gene TCD10 is Needed for Chloroplast Development under Cold Stress

Background Chloroplast plays a vital role in plant development and growth. The pentatricopeptide repeat (PPR) gene family is one of the largest gene families in plants. In addition, cold stress affects a broad spectrum of cellular components, e.g. chloroplast, and metabolism in plants. However, the regulatory mechanism for rice PPR genes on chloroplast development still remains elusive under cold stress. Result In this paper, we characterized a new rice PPR gene mutant tcd10 (thermo-sensitive chlorophyll-deficient mutant 10) that exhibits the albino phenotype, malformed chloroplast and could not survive after the 5-leaf stage when grown at 20 °C, but does the normal phenotype at 32 °C. Map-based cloning, followed by RNA interference and CRISPR/Cas9 genome editing techniques, revealed that TCD10 encoding a novel PPR protein, mainly localized to the chloroplasts, with 27 PPR motifs, is responsible for the mutant phenotype. In addition, TCD10 is specific expression in tissues. The disruption of TCD10 resulted in an evidently reduced expression of chloroplast-associated genes under cold stress (20 °C), whereas they did recovered to normal levels at high temperature (32 °C). These results showed an important role of TCD10 for chloroplast development under cold stress. Conclusions The TCD10 encodes a novel rice PPR protein, mainly located in chloroplasts, which is important for chloroplast development, growth and the maintenance of photosynthetic electron transport and its disorder would lead to an aberrant chloroplast and abnormal expressions in these genes for chloroplast development and photosynthesis in rice under cold stress. Electronic supplementary material The online version of this article (doi:10.1186/s12284-016-0134-1) contains supplementary material, which is available to authorized users.


Background
The formation of normal chloroplasts is crucial for photosynthesis and carbon assimilation in plants. Chloroplast is a semi-autonomous organelle which has its own DNA genome and protein synthesis system, but the only part of the proteins can be synthesized in the chloroplast, the vast majority of proteins are synthesized on the ribosomes of the cytoplasm. Chloroplasts are mostly allocated in the leaves of plants, and its development is closely linked with the leaf development. As known, plastid development from proplastids to mature chloroplasts can be divided into three phases (Chory et al. 1991;Mullet 1993;Kusumi et al. 1997 and. The first phase concerns the activation of plastid replication and plastid DNA synthesis. The second phase is the chloroplast "build-up" stage, which is characterized by the construct of the chloroplast genetic system. At this phase, the nuclear-encoded plastid RNA polymerase (NEP) preferentially transcribes the plastid genes and transcription/translation in the chloroplasts increase strikingly (Hajdukiewicz et al. 1997). In the final phase, the plastid-and nuclear genes for photosynthetic apparatus are massively expressed and principally transcribed by the plastid-encoded RNA polymerase (PEP) (De Santis-Maciossek et al. 1999). Plastid-encoded proteins assemble with imported nuclear-encoded proteins to form photosynthetic and metabolic complexes in chloroplasts. Thus, chloroplast development needs tightly coordinated gene expression between the chloroplast-and nuclear genomes.
Pentatricopeptide repeat (PPR) proteins are one of the largest protein families in plants. The PPR family is characterized by tandem arrays of the PPR motif, a highly degenerate repeat of 35 canonical amino acids (Small and Peeters 2000). PPR proteins harbor between two and 30 PPR motifs, and their tandem alignment allows the modular recognition of specific RNA sequences, generally function in chloroplast or mitochondria (Fujii et al. 2013). Recent studies showed that PPR proteins can promote versatile processes involving RNA, such as editing, splicing, stabilization, and translation (Schmitz-Linneweber and Small 2008;Fujii and Small 2011;Nakamura et al. 2012). In chloroplasts, some PPR proteins have been identified to participate in RNA splicing (Schmitz-Linneweber et al. 2006;De Longevialle et al. 2007;Ichinose et al. 2012), RNA processing (Meierhoff et al. 2003;Hattori et al. 2007), RNA editing (Kotera et al. 2005;Okuda et al. 2007;Chateigner-Boutin et al. 2008;Cai et al. 2009;Yu et al. 2009;Zhou et al. 2009;Tseng et al. 2010;Sosso et al. 2012), translation (Williams and Barkan 2003;Tavares-Carreón et al. 2008), and RNA stability (Yamazaki et al. 2004;Pfalz et al. 2009;Fujii et al. 2013). Despite the fact, a lot of work still to be done is to discriminate the functions of PPR genes in plant development, especially in rice.
Rice (Oryza sativa) is the most important food crop and has also been established as a model species for plant genome research. It is known that a mutation in a PPR gene usually has a strong phenotypic effect (Kocábek et al. 2005). Notably, functional studies of rice PPR proteins remain much sparse. OsPPR1, carrying 11 PPR motifs, is the first identified PPR protein responsible for rice chloroplast development (Gothandam et al. 2005). Antisense transgenic approach was exploited to suppress the expression of OsPPR1 and the obtained transgenic lines displayed chlorophyll-deficiency, albinism and lethality. Another rice PPR protein, YSA, with 16 PPR motifs, also is essential for chloroplast development and its disruption causes a seedling stage-specific albino phenotype (Su et al. 2012). Also the chloroplast-localized PPR protein OsV4 is needed for chloroplast development at early seedling stage under cold stress (Gong et al. 2014). More recently, also we reported that rice ASL3 with 10 PPR motifs is required for chloroplast development and its disruption lead to the death of seedlings (Lin et al. 2015a, b). Herein, we describe a novel thermo-sensitive chlorophyll-deficient mutant, tcd10, which displayed the albino phenotype below 20°C and TCD10 encodes a novel PPR protein, containing 27 PPR motifs, required for chloroplast development and photosynthesis in rice under cold stress.

Phenotypic Analysis of the tcd10 Mutant
The color of the tcd10 mutant seedlings grown at 20°C, 24°C, 28°C and 32°C, respectively, was shown in Fig. 1a. Apparently, the mutant seedlings displayed albino phenotype, thereafter could not survive past 5-leaf stage at 20°C(data not shown); the degree of chlorosis gradually weakened when grown at 24°C and 28°C, but become green as WT plants at 32°C, indicating the low Fig. 1 Characterization of the tcd10 mutants; a 4-leaf stage seedlings of wild type (WT) (right) and tcd10 mutants (left) grown at 20°C, 24°C, 28°C and 32°C, respectively; b, c photosynthetic pigment contents of tcd10 and WT 3-leaf stage seedlings at 20°C and 32°C respectively; Chl, total chlorophyll; Chla, chlorophyll a; Chlb, chlorophyll b; Car, carotenoid thermo-sensitivity of the mutant phenotype. Consist with the observed phenotype, the accumulations of chlorophyll a, b, and carotenoid in tcd10 plants at 20°C (Fig. 1b) were drastically lower than those at 32°C and WT plants (Fig. 1c). It is indicated that the low temperature led to the block of formation of photosynthetic pigments in tcd10 mutants.
The alterations of leaf-color are generally accompanied with the development of chloroplast. To verify this view, we compared the chloroplast ultrastructure of tcd10 and WT leaves under 32°C and 20°C. As expected, WT cells ( Fig. 2) under 32°C and 20°C contained normal chloroplasts with well-organized lamellar structures and were equipped with normally stacked grana and thylakoid membranes. By contrast, most cells under 20°C in tcd10 mutants contained fewer chloroplasts (Fig. 2g), which were theteroplastidic and contained non-pigmented plastids with severely vacuolated and lacked organized lamellar (Fig. 2h). However, the mutant cells at 32°C had not obvious difference with WT plants (Fig. 2c, d). Accordingly, the tcd10 aberrant chloroplast induced by low temperatures caused the decreased accumulation of photosynthetic pigments, so as to produce the mutant phenotype.
We investigated the Fv/Fm values, reflecting the maximum potential capacity of the PSII photochemical reactions in chloroplasts (Krause and Weis 1991). Resultantly, at 20°C, the Fv/Fm value was 0.834 ± 0.04 in WT plants, and undetectable in tcd10 mutants, indicating that the photochemical efficiency of PSII was completely hindered in tcd10 mutants under cold stress. In addition, the max rate (molm −2 s −1 ) of photosynthetic electron transport (rETR max ) was 63.34 ± 0.484 in WT plants and undetectable in tcd10 mutants, showing that the tcd10 mutation completely blocked the electron transport under cold stress. By contrast, at 32°C, all values in the tcd10 mutants (0.559 ± 0.012 for Fv/Fm and 157.200 ± 1.131 for rETR max , respectively) were nearly similar to WT levels (0.642 ± 0.024 for Fv/Fm and 143.800 ± 8.707 for rETR max , respectively), showing that both PSII activity and photosynthetic electron transport in the tcd10 mutants were nearly recovered to WT levels at 32°C.
In addition, the leaf chlorophyll SPAD values (Additional file 1: Figure S1) in tcd10 plants had not significant difference from WT plants after transplanting under nature conditions. However, in contrast to have not great influence on both the final plant height (Additional file 2: Figure S2) and 1000-grain weight (Additional file 3: Figure S3), the panicle number and panicle length was significantly reduced (Additional file 3: Figure S3). These results illustrated the tcd10 mutation affects not only chloroplast development under cold stress and but also the late growth of rice under nature conditions.

Map-based Cloning of TCD10
The genetic analysis (Additional file 4: Table S1) showed that this mutation in tcd10 is a single recessive locus based on the approximate 3(green):1(albino) ratio for phenotypic segregation using F 2 population generated from Guangzhan63S/tcd10. To this end, we randomly selected the 222 F 2 individuals with mutant phenotype and mapped TCD10 between RM475 and RM33 on chromosome 10 (Fig. 3a). To further fine-map TCD10, a total of 5200 F 2 mutant invididuals and developed five InDel molecular markers (Additional file 5: Table S2) were exploited. Ultimately, the TCD10 was localized to 70 kb between ID14874 and ID14944 (Fig. 3b). Using Rice Genome Annotation Project (http://rice.plantbiology.msu.edu/cgi-bin/gbrowse/rice/#search), the target genomic region contained six predicted candidate genes (Fig. 3c). ; e, f, g, h, Chloroplast structure in WT(e, f)and tcd10(g, h) cell at 20°C. c, chloroplast; g, grana lamella stacks By sequencing and analyzing all the candidate genes, compared with wild type, in the tcd10 mutants we only uncovered a 1-bp deletion (A) at position 766 bp from the ATG start codon in the extron of LOC_Os10g28600 (Fig. 4d) which caused pre-termination, suggesting that TCD10 is LOC_Os10g28600.

Knock-down and -out of TCD10 Exhibits the Mutant Phenotype
To further affirm that the malfunction of TCD10 is blame for the mutant phenotype, we used both RNA interference (RNAi) and CRISPR/Cas9 genome editing system to knockdown and knockout LOC_Os10g28600 in WT plants, respectively. In RNAi study, we obtained eleven-one RNAi lines all showing the chlorosis phenotype similar to tcd10 at 20°C (Fig. 3f ). Thus, the downregulated TCD10 expression in RNAi lines (Fig. 3g) could mimic the phenotype of the tcd10 mutant. In CRISPR/Cas9 study, we obtained 38 transgenic T 0 plants carrying two different editing sites (the 1-bp (G) and 17-bp (CCATGGCCGGGTCGGGG) deletion happened at position 316 bp and 306 bp from the ATG start codon in TCD10, respectively) (Additional file 6: Figure S4). Importantly, all transgenic T 1 plants appeared the separation of albino phenotype as tcd10 mutants at 20°C (Fig. 3f). Taken together, these results reconfirmed that LOC_Os10g28600 is TCD10.

Characterization of TCD10 Protein
Determination of the nucleotide sequence of the RT-PCR product revealed that TCD10 encodes a 3790 bp open-reading frame, which is not interrupted by any intron and encodes a 1080-amino acid polypeptide with a calculated molecular mass of 120kD. The TCD10 contains 27 PPR motifs of the protein (Fig. 3e, Additional file 7: Figure S5). In addition, the first 50 amino acids (Additional file 7: Figure S5) were predicted to be a chloroplast transit peptide (CTP) by targetp (http:// www.cbs.dtu.dk/services/TargetP/). In addition, the two editing deletion mutations in the transgenic lines (T1-1and T1-2) and the tcd110 mutation occurred in the 1st and 5th PPR motif, respectively, and all caused the premature termination of translation, finally leading to the complete damage of protein structure (Fig. 3e, Additional file 7: Figure S5). Multiple amino acid sequence alignment showed that TCD10 was similar to many species (Fig. 4a), including Arabidopsis At4g31850/PGR3 with 62.3% of identity (Shikanai et al. 1999;Yamazaki et al. 2004;  Cai et al. 2011;Fujii et al. 2013). The subsequent phylogenetic analysis showed that TCD10 had the closest genetic relationship with one of the predicted proteins in sorghum (Fig. 4b), indicating the high conservation of TCD10 protein in plants.

TCD10 Expression Pattern
To determine expression pattern of TCD10 gene in various tissues, we analyzed the TCD10 expression level of the root, stems, young-leaf, flag-leaf and panicle in wildtype plant tissues. As shown in Fig. 5a, TCD10 in all various tissues were expressed, but in the leaves was relatively higher than that of other tissues, especially in the flag leaf expression was the highest, and expression in roots and stems rarely, which basically is consistent with RiceXPro database prediction results (http://ricexpro.dna.affrc.go.jp/) (Additional file 8: Figure S6). In addition, the expression levels were increased together with leaf age (Fig. 5b), indicating the importance of TCD10 for leaf chloroplast development under cold stress.

The Transcript Expressions of the Associated Genes in the tcd10 Mutants
We investigated the transcript levels of twenty-five genes (Additional file 9: Table S3) associated with Chl biosynthesis, photosynthesis and chloroplast development (CAO1, PORA, YGL1, Cab1R, Cab2R, rbcL, rbcS, psaA, psbA, LhcpII, rps7, OsV4, rpoC, OsRpoTp, V1, V2, RNRL, RNRS, 16SrRNA, rps20, FtsZ, rpl21, and OsDG2) in the tcd10 mutants. As a result, under cold stress (20°C) (Fig. 6a), except for the slight increase in YGL1 encoding Chl synthase, compared with WT plants, the transcripts in those genes for Chl biosynthesis, such as chlorophyllide a oxygenase (CAO1), and NADPH: protochlorophyllide oxidoreductase (PORA), were significantly impeded in tcd10 mutants, consistent with the reduced Chl content (Fig. 1c) and the mutant phenotype (Fig. 1a). Also all the seven photosynthesis-associated genes studied (Cab1R, Cab2R, rbcL, rbcS, psaA, psbA, LhcpII) ( Fig. 6b) were severely hampered in tcd10 mutants, consistent with the undetectable photochemical efficiency of PSII and photosynthetic electron transport in photosystems in tcd10 mutants, thereafter led to the seedling death. Moreover, all transcript accumulations of sixteen genes tested for chloroplast development were abnormally influenced in tcd10 mutants, especially OsV4, OsRpoTp, V1, V2, RNRL, RNRS, 16SrRNA, rpl21 and OsDG2 were considerably down-regulated (Fig. 6c). More notably, at 32°C, the transcript levels of the down-regulated genes at low temperatures in tcd10 mutants were recovered to or even slightly higher than WT plants (Fig. 7a, b, c). Accordingly, under high temperatures (32°C), the tcd10 mutation has very little effect on the associated-genes for Chl biosynthesis, photosynthesis and chloroplast development, consistent with the normal phenotype in tcd10 mutants at 32°C. Taken together, TCD10 is needed for chloroplast development under cold stress.

Subcellular Localization of TCD10
The majority of PPR proteins are predicted to be targeted to either mitochondria or chloroplasts (Small . In this study, TargetP (Emanuelsson et al. 2000, http://www.cbs.dtu.dk/services/TargetP/) and iPSORT (http://ipsort.hgc.jp/) all suggested possible chloroplast localization and secretory channels. To assess possible localization of TCD10, we constructed the recombinant expression vectors pMON530-GFP-TCD10 and transformed into tobacco cells by Agrobacteriummediated infection. Meanwhile, empty GFP vector was used as a control. Resultantly, the vast majority of the green fluorescent signal of the TCD10-GFP fusion protein was coincident with the chlorophyll autofluorescence, and the remaining little parts did not overlap in tobacco cells (Fig. 8b). In contrast, the cells transformed with the empty GFP vector without a specific targeting sequence had green fluorescent signals in both the cytoplasm and the nucleus (Fig. 8a). These findings clearly suggest that TCD10 was mainly localized to the chloroplast in addition to secretory channels.

TCD10 is Needed for Chloroplast Development and Growth under Cold Stress
The chloroplast is a semi-autonomous organelle, which contains about 100 genes, more than 3000 proteins function within it (Leister 2003). The normal chloroplast development needs tightly coordinated expressions between plastid-and nuclear genes. The mutation of those genes could result in chlorophyll deficient/chloroplast defects in plants. In this study, the tcd10 mutants, under low temperatures (20°C), had the severe affected transcript-levels of most genes for Chl biosynthesis, photosynthesis and chloroplast development (Fig. 6), whereas the down-regulated genes at low temperatures could be recovered to normal level or even higher WT levels at high temperature (32°C) (Fig. 7). This discrepancy was attributable to the difference in the chloroplast structure and pigment contents between 20°C and 32°C. In addition, as known, when rice seedlings grow up to the 3/4-leaf stage, a seed's energy is exhausted, stored reserves become insufficient to meet itself the rice growth demand. The continued growing of seedlings requires the external supply of resources from photosynthesis. Hence, we could speculate that the no survival for tcd10 mutants after the 5-leaf stage at 20°C maybe result from the loss of photosynthesis due to the block of electron transport in photosystems under cold stress. Notwithstanding the reason why abnormal chloroplast occurs only under cold stress is not completely claimed yet, it is speculated that the TCD10 function is possibly not prerequisite at higher temperatures, but it is essential/more required for rice chloroplast development under cold stress.

TCD10 Functions at the First Phase of Chloroplast Development Possible via Plastid Division
As described previously, the processes accompanying chloroplast development can be divided into three phases (Mullet 1993;Kusumi et al. 2010 and. It is clearly known that OsPOLP1, encoding one plastidial DNA polymerase (Takeuchi et al. 2007;Kusumi et al. 2010), FtsZ encoding a component of the plastid division machinery (Vitha et al. 2001) are critical for the first phase, determine the number and size of chloroplasts (Takeuchi et al. 2007) and the NEP-transcribed genes which encode PEP cores (e.g. rpoA, rpoB, rpoC) for chloroplast genetic system and the PEP-transcribed genes (e.g. psab, LhcpII) involving PSII function at the second and third phase, respectively. Our data that all transcripts for the PEP-transcribed plastid (e.g. rbcL, psaA and LhcpII) and nuclear genes (rbcS) associated A B Chlorophyl GFP Merged Fig. 8 Subcellular localization of TCD10 protein; a Empty GFP vector without a specific targeting sequence; b TCD10-GFP fusion; The scale bar represents 20 μm with photosynthesis apparatus (Fig. 6b) were obviously down-regulated in the tcd10 mutants showed that TCD10, at least, regulates the third phase of chloroplast development. In addition to V1 for NUS1 protein (Kusumi et al. 1997 and, V2 for guanylate kinase (Sugimoto et al. 2007) and OsV4 for PPR protein (Gong et al. 2014) and ASL2 encoding plastid ribosomal protein L21 (Lin et al. 2015a, b) and TCM5 for Deg protease protein (Zheng et al. 2016), the extremely decreased accumulation of OsRpoTp (Hiratsuka et al. 1989) and the NEP-transcribed gene rpoB and rpoC which encodes PEP core β, β' subunits functioning in the second step suggests that TCD10 also affects the second phase of chloroplast development. Similarly, the significantly reduced transcripts of rbcL, an indicator for the accumulation of polysomes (Barkan 1993), chloroplast 16SrRNA, 23SrRNA and rps7, rps14, rps20 in the tcd10 mutants indicated that the tcd10 mutation affects the assembly and accumulation of plastid ribosomes, resulting in disruption of plastid translation. Similarly, the blockage of the chloroplast 16SrRNA maturation in Arabidopsis rap mutant (Kleinknecht et al. 2014) and rice al1 mutant (Zhang et al. 2016) leads to abnormal chloroplasts. Thus, TCD10 participates in not only the plastid translation, but also the assembly and the accumulation of plastid ribosomes. More notably, the obviously down-regulations in transcripts of the known key genes (OsPoLP1 for one plastidial DNApolymerase (Takeuchi et al. 2007) (>25 fold) and RNRL for (Yoo et al. 2009) (>11 fold), OsDG2 for glycine-rich protein (Jiang et al. 2014a) (>166 fold) at the first step of chloroplast development indicated that TCD10 functions in the first step. As for the slightly increase in FtsZ transcript (<3 fold), it is likely due to the feedback effects. In addition, in view of fewer and smaller of chloroplasts (Fig. 2g) and the abnormal thylakoid structure (Fig. 2h) in tcd10 mutant cells, it suggests that the TCD10 regulates the first phase of chloroplast development probably via plasmid division under cold stress.

Possible Role of TCD10 Protein in Chloroplast RNA Metabolism
PPR protein can specifically identify single stranded RNA sequence, and it is found that the PPR protein is closely related to the transcription process, the shearing process, the editing process and the stability of RNA (Kotera et al. 2005;Okuda et al. 2007;Chateigner-Boutin et al. 2008;Cai et al. 2009;Yu et al. 2009;Zhou et al. 2009;Tseng et al. 2010;Sosso et al. 2012;Williams and Barkan 2003;Tavares-Carreón et al. 2008;Yamazaki et al. 2004;Pfalz et al. 2009;Fujii et al. 2013). With the unveiling of the primary version of the PPR code, manipulation of PPRs to bind to target RNAs is within reach (Barkan et al. 2012), biochemical (Barkan et al. 2012;Kobayashi et al. 2012), statistical (Barkan et al. 2012;Yagi et al. 2013), and evolutionary constraint (Fujii and Small 2011) studies have postulated that the 1st, 3rd, and 6th amino acids within the PPR consensus manipulates RNA base recognition specificity. As stated above, at low temperatures, TCD10 directly or indirectly regulates RNA transcript levels at all three steps of chloroplast development and participates in the plastid translation, the assembly and the accumulation of plastid ribosomes. As for TCD10, how to specifically participate in chloroplast RNA metabolism is unclear yet in detail. However, interestingly, Arabidopsis PGR3 (At4g31850), containing the same 27 PPR motifs, with the highest homolog (62.3% of identity) to TCD10 (Fig. 4a and Additional file 7: Figure S5), was previously studied in very detail (Shikanai et al. 1999;Yamazaki et al. 2004;Cai et al. 2011;Fujii et al. 2013) by mean of three Arabidopsis pgr3-1(the substitution mutation site in the 15th PPR motif), pgr3-2(the substitution mutation in the 12th PPR motif), and pgr3-3 (the substitution mutation in the 27th PPR motif) mutant alleles (Additional file 7: Figure S5). The chloroplast-targeted PGR3 was reported to have three distinct functions: (i) stabilization of photosynthetic electron transport L (petL) operon RNA, (ii) translational activation of petL, and (iii) translational activation of ndhA (Yamazaki et al. 2004;Cai et al. 2011) through binding to the 5′untranslated region (UTR) of petL operon RNA and ndhA (Cai et al. 2011). More interestingly, the 16 N-terminal PPRs in PRG3 was sufficient for function (i) via sequence-specific RNA binding, whereas the 11 Cterminal motifs were essential for functions (ii) and (iii) by activating translation (Fujii et al. 2013). Considering the high homolog between TCD10 and PGR3, it could be speculated that TCD10 may have similar three functions. Since the mutation in tcd10 mutants occurred in the 5th PPR motif site (Fig. 3e), resulting in the changes of structure and loss-of-function of TCD10, thus the tcd10 mutants should loss all three functions aforementioned. Indeed, the severe defects in photosynthetic electron transport pathways in tcd10 mutants as pgr3 mutants (Yamazaki et al. 2004) strongly supported the notion that TCD10 has the stabilization (i) of photosynthetic electron transport under cold stress; Meanwhile, the reduced transcripts of protochlorophyllide oxidoreductase also indirectly indicates the block of (iii) translational activation of ndhA. Regrettably, no evidences in our study can support the (ii) function of translational activation of petL because of our technical limitations. It is notable that, only under cold stress, TCD10 affected chloroplast development unlike Arabidopsis PRG3 without temperatures, showing their obviously different responses to temperatures. These findings could highlight the notion that highly conserved genes within various species might play diverse and complex roles than previously recognized. Hence, further understanding function difference between PGR3 and TCD10 and how TCD10 functions chloroplast development under cold stress is needed.

Conclusions
The rice TCD10 encodes a new PPR protein, mainly located in chloroplast, which is important for chloroplast development, seedling growth and the maintenance of photosynthetic electron transport and its disorder would lead to an aberrant chloroplast and abnormal transcript levels of genes associated with chloroplast development and photosynthesis under cold stress.

Plant Materials and Growth Conditions
The thermo-sensitive chlorophyll-deficient mutant, tcd10, used in this study, was isolated from japonica rice variety, Xiushui11 (WT), treated by 60 Co gamma-radiation in 2006. All rice plants were grown in a rice paddy field in Shanghai (31°11'N, summer season, temperate climate) and Hainan (18°16'N, winter season, subtropical climate), China under local conditions or in the growth chambers.

Phenotype Observation and Photosynthetic Pigment Measurements
The germinated rice-seeds were grown at the chambers with four temperatures (20°C, 24°C, 28°C and 32°C) and 12-h-dark/12-h-light. Each third-leaf 200 mg were sampled at the 3-leaf stage and used to extract the pigments. Photosynthetic pigment (chlorophyll a, b carotenoid) were measured according to previously described methods (Jiang et al. 2014b).
To examine the changes of plant height and leaf chlorophyll content during the whole growth-period, leaf chlorophyll SPAD values, using fast non-destructive chlorophyll meter (SPAD-502, Minolta, Japan) (Peng et al. 1993;Turner and Jund, 1991), and plant height were measured once a week from the 1st weeks after transplanting, respectively. At maturity, panicle-associated traits of rice plants (Additional file 1: Figure S1) were investigated.

Chlorophyll Fluorescence and Transmission Electron Microscopy Analysis
Rice seedlings were cultivated at 20°C and 32°C in growth chambers. Chlorophyll fluorescence analysis for the 3rd leaves at the 3-leaf stage was performed by mean of a PAM-2000 portable chlorophyll fluorometer (MINI-PAM; Walz; http://walz.com). In addition to the max rate of photosynthetic electron transport (rETRmax), the variables Fo, Fm, Fv, and the Fv/Fm ratio were measured and calculated basically according to Meurer et al. (1996) after dark-adapted for 20 min. In addition, the observation of chloroplast ultrastructure in the third leaves was performed according to the previously described methods (Jiang et al. 2014a, b). Finally, samples were examined with a Hitachi-7650 transmission electron microscope.

Map-based Cloning of TCD10
Rice genomic DNA was extracted from young leaves by the CTAB method with minor modifications (Stewart and Via 1993). For genetic analysis and mapping for TCD10 locus, we crossed tcd10 mutant with Guangz-han63S (indica) to obtain F 2 mapping population. Initially, we adopted 81 SSR primers based on data in Gramene (http://www.gramene.org) to determine the chromosome location of TCD10. Then new SSR and InDel markers (Additional file 5: Table S2) were designed using the PREMIRE 5.0 software based on the whole genomic sequences of the japonica Nipponbare variety (Goff et al. 2002) and the indica variety 9311 . In this study, a total of 5200 F 2 mutant individuals were selected for fine-mapping. The functions of candidate genes and respective cDNA sequences were obtaining using TIGR (http://rice.plantbiology.msu.edu/ cgi-bin/gbrowse/rice/) and KOME (http://cdna01.dna.affrc.go.jp/cDNA/index.html), respectively.

RNAi Suppression and Knockout of TCD10
To confirm that TCD10 was blame for the mutant phenotype observed, both RNAi and CRISPR/Cas9 genome editing technology, which is of significance for basic plant research and crop genetic improvement (Belhaj et al. 2015;Ma and Liu 2016;Ma et al. 2016), were preformed. In RNAi experiments, the construct vector pTCK303 with a maize ubiquitin promoter and a rice intron was used as an RNAi vector (Wang et al. 2004). Both anti-sense and sense versions of a specific 780 bp fragment from the coding region of the TCD10 were amplified, and successively inserted into pTCK303, to form the RNAi construct vector pTCK303-dsRNAiTCD10. The primer pairs are 5'CGAGCTCGGTTCCTGATGTTCTTGCC GTG3' (SacI), 5' GACTAGTC GATCAGGTATGCAT CCTTTCG 3' (SpeI) and 5' CGGGATCCCG GTTCCT GATGTTCTTGCCGTG 3 '(BamHI), 5' GGGGTAC CCC GATCAGGTATGCATCCTTTCG 3' (KpnI).
The resultant plasmid (pTCK303-dsRNAiTCD10 and CRISPR/Cas9 expression) and the respective empty vectors were introduced into Agrobacterium tumefaciens EHA105 and used to infect calli of WT plants according to the previously published methods (Hiei et al. 1994). The obtained T 0 transgenic seedlings were grown in a paddy rice field after screening of tolerant-hygromycin and DNA sequencing according to our described methods (Jiang et al. 2014a, b). Then all T 1 seedlings grown at 20°C were used for surveying the segregation of phenotype.

RNA Transcript Analysis
Total RNA was extracted from various fresh tissues by the TRIzol Reagent (Invitrogen; http://www.invitrogen.com) and DNase I treated by an RNeasy kit (Qiagen; http:// www.qiagen.com) and the first-strand cDNA was synthesized with the Revert-Aid first-strand cDNA synthesis kit (Toyobo; http://www.toyobo.co.jp) following the manufacturer's instructions. The specific primers used for RT-PCR assays were shown in Additional file 9: Table S3. Then semiquantitative RT-PCR analysis was performed to assay the tissue expression pattern of TCD10. Real Time qPCR (ABI7500, Applied Biosystems; http://www.appliedbiosystems.com) was used to analyze RNA expression levels for associated-genes of chloroplast development, chlorophyll synthesis and photosynthesis. The relative quantification of gene expression data was analyzed as described by Livak and Schmittgen (2001). Three independent experiments were performed, and each sample was run in triplicate. OsActin was used as an internal control.

TCD10 Subcellular Localization
To investigate the subcellular localization of TCD10, cDNA fragments encoding the N-terminal region of 260 amino acids in TCD10 were amplified by PCR using following primer (5'-GAAGATCTATGTTGGAGGTCTG CTGCTGC-3' and 5'-GGGGTACCCCCTTCATTCTC-CATTTTCGC-3') and introduced into vector pMON530-GFP at the BglII and KpnI sites (the sequence underlined were represented cleavage sites). TCD10 localization was investigated by transient expression of GFP fusion in tobacco (Nicotiana tabacum) cells, and confocal microscopy as described previously (Jiang et al. 2014a, b). Finally, GFP fluorescence in the transformed protoplasts was imaged using a confocal laser-scanning microscopy (LSM 5 PASCAL; ZEISS, http://www.zeiss.com)