Introduction

Ribulose-1,5-bisphosphate carboxylase/oxygenase (Rubisco) is the most plentiful protein in the leaves of light-grown plants and catalyzes the rate-limiting stage of CO2 fixation in photosynthesis. Rubisco is composed of two subunits, large subunit (RbcL) and small subunit (RbcS). The 55-kDa in size RbcL is encoded in chloroplast genome, whereas the 15-kDa in size RbcS is encoded in the nucleus and targeted to chloroplasts by their transit peptide (TP). As shown in Kyozuka et al. (1993), RbcS1 gene originated from Oryza sativa L. cv IR26 was characterized the properties that are light-regulated and mesophyll-specific expressed. Previous report showed high GFP mRNA and protein levels reaching up to 10 % of the total soluble protein. The correct cleavage site of TP:GFP is between the cysteine (residue 47) and methionine (residue 48) during transport to chloroplast of the reporter gene in transgenic plants (Jang et al. 1999). The high level of GFP protein accumulation, primarily reflected in the increase of the mRNA levels, was due to the presence of the functional RbcS1 transit peptide (Jang et al. 2002). Furthermore, the paralogs of RbcS genes have highly conserved coding regions although the levels and patterns of expression can vary.

The RBCS3C gene expression pattern and promoter sequences have been well studied among the five members of a tomato RbcS gene family (RBCS1, 2, 3A, 3B, 3C) (Sugita and Gruissem 1987; Sugita et al. 1987; Manzara et al. 1991; Wanner and Gruissem 1991; Carrasco et al. 1993; Manzara et al. 1993; Meier et al. 1995). The other individual gene of this family has various expression patterns of during temporal treatment and in specific organs. In common, all the five members are expressed high levels in cotyledons and leaves grown on light conditions, but the expression is not detected in roots. The tomato RBCS gene expression is controlled by light similar to other photosynthetic genes. After moving the plant to immediate dark conditions, the transcript levels of RBCS2 and RBCS3A are decreased, while the levels of RBCS1, RBCS3B and RBCS3C come to be undetectable. Soon after placing the plant under light, levels of mRNA rapidly turn back to those of normal conditions. The soybean SRS1 and SRS4 out of six genes of RbcS were expressed topmost and characterized in detail (Berry-Lowe et al. 1982; Shirley et al. 1990). These two genes are highly expressed in leaves grown on light conditions. It means that expression of the two genes is controlled by light. As same as the tomato RbcS genes, after transferring plants to dark conditions, the mRNA levels of both genes are rapidly decreased and turn back to normal levels by restoring to light conditions.

In this study, we identified four different rice small subunit of Rubisco genes by analyzing protein alignment and phylogenic tree. Among them, the RbcS3 promoter and its transit peptide were chosen to drive transgene expression in rice. Several transgenic lines were produced and GFP expression levels were analyzed at the transcript and protein levels. The RbcS3 promoter and its transit peptide were found to be as efficient as RbcS1 for alternative chloroplast-targeted expression.

Materials and methods

Multiple sequence alignments and phylogenetic analysis

Phylogenetic analysis of amino acid sequences from four rice rubisco small subunit (RbcS) genes was carried out using CLUSTALW program which is a multiple sequence alignment tool (Larkin et al. 2007).

Plant materials

Genomic DNA (gDNA) isolation manner from cv. Nipponbare (Oryza sativa L. var. Japonica) was followed by Park et al. (2013b). Through the rice transformation in cv. Nakdong and propagation, ten independent lines from the T3 generations of transgenic plants transformed-Promoter:GFP constructs were selected for subsequent detailed analysis.

Semi-qRT-PCR analysis of endogenous gene expression

To investigate expression of endogenous gene, plants grown in a greenhouse were sampled on the subsequent time and growth stages that are 30 and 60 days after germination (DAG), before heading stage (BH) and dry seed. Total RNAs from the ground powders were then isolated using the GeneAll RiboEx™ LS Kit (GeneAll Biotech, Korea). First-strand cDNA from the isolated total RNAs was synthesized using cDNA synthesis kits (Fermentas, Burlington, Canada). Using the gene-specific primers as shown in Table S2, the RT-PCR was performed with 50 ng of cDNA template. The cycle carried out RT-PCR is as follows as previously reported (Park et al. 2013b); 98 °C for 10 min, followed by 29 cycles at 98 °C for 20 s, 55 °C for 20 s, and 72 °C for 30 s, with one cycle at 10 °C for 10 min. To verify RNA loading, the OsCc1 (AF399666) and OsUbi1 (AK121590) genes were used.

Constructs and transgenic rice production

The promoter RbcS3:TP and RbcS3 fragments were isolated using the promoter-isolated primers as shown in Table S2 and PrimeSTAR DNA polymerase (Takara Bio Inc., Japan). The PCR cycle was set as previously described by Park et al. (2010). The primer pairs are listed in supplementary Table S2. The isolated promoter was recombined into the rice transformation vector as previously described by Park et al. (2010). The rice transformation vector consisted of the CaMV 35S promoter link to the bar gene for use with conventional herbicide-based plant selection. Through tri-parental mating assay, the vectors were integrated into Agrobacterium tumefaciens LBA4404 strain. Calli proliferated from the seeds of rice (cv. Nakdong) were co-cultivated with those Agrobacterium (Hiei et al. 1994), grown the regenerating medium containing with 4 mg/L phosphinothricin until producing transgenics as previously described by Park et al. (2013a).

qRT-PCR analysis

Total RNA was purified from leaves and roots of 30 DAG transgenic and non-transgenic (NT) rice plants using the GeneAll RiboEx™ LS Kit (GeneAll Biotech, Korea) and followed by First-strand cDNA synthesis kit (Fermentas, Burlington, Canada). The real-time qRT-PCR was performed with 50 ng of cDNA template, using the GFP-specific primers listed in Table S2 and a 2× Real-time PCR Pre-mixture with Evagreen (SolGent, Seoul, Korea). The thermocycling was as follows: at 95 °C for 10 min followed by 40 cycles of 95 °C for 20 s, 58 °C for 20 s, 72 °C for 20 s in a 20 μl volume. The fluorescence from the amplicons was detected by a Stratagene Mx3000p Real-Time PCR machine and Mx3005P software v2.02 (Stratagene, La Jolla, CA, USA) as previously described (Jang et al. 2014). Gene-specific primer pairs were designed by testing melting curve analysis as reported in Park et al. (2010). All the qRT-PCR results were calculated from triplicate reactions per sample. The comparative experiments were performed and the results were analyzed using Mx3005P software v2.02 (Stratagene, La Jolla, CA, USA). A reference gene, OsUbi1 (AK121590), was used to normalize the individual cDNA template.

Protoplast isolation and transient transformation

Leaf sheaths from 10-day-old seedlings were cut into 1-mm disks using a sharp razor blade. The pieces were immediately transferred into a 15-ml digestion solution composed of 0.6 M mannitol, 1.5 % cellulase R-10 (Yakult, Japan), 0.75 % macerozyme R-10 (Yakult, Japan), 1 mM CaC12, 0.1 % bovine serum albumin, 5 mM beta-mercaptoethanol and 0.6 mM 4-morpholineethanesulfonic acid (MES) pH 5.7, and incubated for 4–5 h at 28 °C with gentle shaking in the dark conditions. Digested samples were then filtered twice first using a 70-μm pore sixe nylon mesh (BD Bioscience, USA) and second with 40-μm pore to collect the protoplasts. The collected protoplast were then re-suspended in 20 ml of washing buffer (W5) composed of 154 mM NaCl, 125 mM CaC12, 5 mM KC1, 5 mM d-glucose and 2 mM MES, pH 5.7. The protoplast suspensions were centrifuged at 320×g for 8 min to pellet the protoplasts, which were then re-suspended smoothly in 1 ml MMg solution with consisting of 0.6 M mannitol, 15 mM MgCl2and 4 mM MES, pH 5.7. The protoplasts were counted and quantified using a hemocytometer under a microscope (CX21; Olympus, Tokyo, Japan). The constructs of 35S:TP3:GFP, 35S:TP1:GFP and 35S:GFP were then transfected with the isolated protoplasts using polyethylene glycol-mediated transformation. After 16-h incubation at 28 °C, the transiently transformed protoplasts were gathered by carrying out centrifugation at 300×g for 5 min.

Observation of GFP fluorescence

The GFP fluorescence of rice leaf protoplasts isolated from the transgenic plant or transformed temporally was visualized and taken a picture using a confocal laser scanning microscope (Leica TCS SP8 STED, Wetzlar, Germany). GFP fluorescence of dry seeds was detected under a stereomicroscope (SZX9-3122, Olympus, Tokyo, Japan) with GFP filter set as described by Park et al. (2010) and digital camera (C5060-ZOOM, Olympus).

Immunoblotting analysis

Total soluble proteins of 2-week-old leaves were extracted using the extraction buffer containing 50 mM Tris–HCl pH 8.0, 100 mM NaCl, 1 mM dithiothreitol and 1 %Triton X-100 and 100 μM phenylmethylsulphonylfluoride (PMSF). After measuring protein concentrations using the Bradford manner (Bio-Rad, Hercules, CA, USA), protein bands were separated on a 12 % SDS–polyacrylamide gel and transferred onto a polyvinylidene difluoride (PVDF) membrane (Immobilon-P, Millipore Co., Billerica, MA, USA) using a wet transfer tool (Hoefer, San Francisco, CA, USA). The immunoreactive proteins were detected using anti-GFP anti-serum from mouse (Santa cruz biotechnology Inc., CA, USA). The chemi-luminescence signals were detected using ECL kit (GE Healthcare, Buckinghamshire, UK) and visualized a FusionSL (Vilber Lourmat, France) imaging system. Recombinant GFP proteins were used as a positive control (Abcam, Cambridge, UK).

Results

To develop a chloroplast expression system in rice, phylogenic analysis of the four small subunit of Rubisco (RbcS) genes was performed as shown in Fig. 1. Comparison of the nucleotide sequences revealed differences between genes (Fig. S1). All four genes encode similar amino acid sequences with some differences found only in the transit peptide region (Fig. 1b). The predominant expressions of the four RbcS genes in green tissues such as leaf blade, leaf sheath and stem were found by the spatio-temporal transcript profiling (Fig. S2). Expression of the RbcS1 and RbcS2 genes was also detected in flower organs and endosperms whereas that of RbcS3 and RbcS4 genes was not. We have therefore chosen RbcS3 for us to develop a green tissue-specific chloroplast-targeting system in transgenic plants. The endogenous expression of RbcS3 gene was confirmed by semi-qRT-PCR in six different tissues of rice. These were leaf and root tissues at 30 and 60 days after germination (DAG), panicles before pollination and dry seeds (Fig. 2a). The RbcS3 gene was highly expressed only in green tissues, leaves and green panicles at the heading stage which was similar to RbcS1 gene expression.

Fig. 1
figure 1

Homologous genes of small subunits of Rubisco (RbcS) in the Oryza sativa L. a The numbers in front and back of black arrow boxes indicate genomic locations of the four different RbcS genes found in rice chromosome 12. The distance is presented by base-pair numbers between two genes. b Protein alignment of the four rice RbcSs. Amino acid alignment was performed with CLUSTALW (Larkin et al. 2007). The N-terminal 47 amino acids filled with green box are transit peptides. c Phylogenic analysis of the four RbcS proteins. Phylogenetic analysis of the amino acid sequences was performed by rooted phylogenetic tree programs

Full size image
Fig. 2
figure 2

Endogenous expression patterns and structure of rice RbcS3. a Green tissue-specific expression of RbcS3. The endogenous gene expressions were analyzed by semi-qRT-PCR in six different tissues of rice growth. Leaves (L) and roots (R) of 30 and 60 days after germination (DAG) were represented the vegetative stage of growth. Panicles before pollination (less than 10 cm in size) and dry seeds were included. Control genes for equal amount of cDNA loading are OsCc1 (AF399666) and OsUbi1 (AK121590). The gene RbcS1 (Os12g0274700) was used as a positive control for green tissue-specific expression. b Genomic structure of RbcS3 gene. Gray filled boxes represent exons; line between boxes are intron; black bold line is the 5′ upstream region including promoter region; white box is TP3 (transit peptide of RbcS3). The +1 and ATG are referred to as the transcriptional start site and translational start codon, respectively. c Constructs used for rice transformation. RbcS3 promoter (P RbcS3) with and/or without TP3 were linked to the reporter gene green fluorescent protein (GFP) and potato proteinase inhibitor II terminator (T PinII). The phosphinothricin acetyl transferase gene (BAR) as a selection marker gene for rice transformation was connected between the CaMV 35S promoter (P 35S) and the nopaline synthase terminator (T nos)

Full size image

To investigate RbcS3 promoter and transit peptide (TP3) activities, the promoter region with TP3 (1971 base pairs) and without TP3 (1802 base pairs) was isolated from the genomic DNA of rice (Oryza sativa cv. Nipponbare) by PCR. The isolated PCR fragments were linked to the green fluorescent protein gene GFP and inserted into a rice transformation vector (Fig. 2b,c). Thirty independent transgenic lines were produced through Agrobacterium-mediated rice transformation (Hiei et al. 1994). Copy number analysis of transgenes in transgenic plants was determined through TaqMan qPCR using probes specific for the BAR sequence. Ten T3 homozygous single-copy lines were then selected for further analysis.

To measure promoter activity and tissue specificity, ten independent transgenic lines from RbcS3:TP3:GFP and RbcS3:GFP and three from PGD1:GFP were grown in normal greenhouse conditions for 30 days. Total RNAs were isolated from leaf and root tissues at 30 DAG (Fig. 3a). Levels of GFP mRNAs in leaves RbcS3:TP3:GFP and RbcS3:GFP transgenic lines were determined by qRT-PCR and compared with those of PGD1:GFP transgenic lines. The GFP transcript levels of RbcS3:TP3:GFP transgenic lines were significantly higher than RbcS3:GFP transgenic lines. The TP3 sequence increased the transcript levels more than fivefold (Table S1). The GFP transcripts were not detected in roots of RbcS3:TP3:GFP transgenic lines but the levels in leaves were comparable with those of PGD1:GFP transgenic leaves (Fig. 3). These results suggest that the RbcS3 promoter is highly active in leaf tissues with its own functional transit peptide.

Fig. 3
figure 3

Quantitative comparison of promoter activity and tissue specificity in ten independent lines of RbcS3:TP3:GFP and RbcS3:GFP transgenic plants. Total RNAs were isolated from leaf and root tissues of 30-day-old independent T3 lines of transgenic plants RbcS3:TP3:GFP and RbcS3:GFP. Three lines of PGD1:GFP transgenic plants in the same generation were used as a positive control. a Levels of GFP mRNAs were calculated by analyzing qRT-PCR, normalized with an internal control gene OsUbi1. The fold values were drew a picture relative to the level of GFP mRNAs in the lowest expression transgenic lines (marked with an asterisk). b Range of variation in the promoter activity in different transgenic lines. Red bars were indicated median values of each transgenic construct. Data bars represent the mean ± SD of triplicate measurements

Full size image

To determine chloroplast localization of transgene product in RbcS3:TP3:GFP transgenic plants, GFP proteins in leaf protoplasts were examined under confocal laser scanning microscope. Leaf protoplasts isolated from transgenic plants were used for subcellular localization (Fig. 4a). Red autofluorescence from chlorophylls was detected in normal leaf chloroplasts. The GFP fluorescence of the RbcS3:TP3:GFP transgenic protoplasts overlapped with the red fluorescence of chloroplasts while the fluorescence of RbcS3:GFP was detected only in the cytosol. PGD1 and maize Ubi1 promoters were used as controls for cytosolic localization (Fig. 4a). To make a comparison of subcellular targeting efficiency between two transit peptides TP3 and TP1, transient expression assays were performed using wild-type protoplasts. Protoplasts were temporally transformed with three different constructs consisting of TP3:GFP, TP1:GFP or simple GFP driven by the CaMV 35S promoter (Fig. 4b). The GFP fluorescence of 35S:GFP-transformed protoplasts is specifically localized in cytosol, whereas the GFP fluorescence of constructs carrying the TP3 and TP1 is specifically localized in chloroplasts. Moreover, the chloroplasts targeting efficiency of TP3 and TP1 appears to be comparable, as judged by the level of GFP fluorescence in 35S:TP3:GFP- and 35S:TP1:GFP-transformed protoplasts (Fig. 4b).

Fig. 4
figure 4

Subcellular targeting patterns in rice leaf protoplasts. a Transgenic plants harboring RbcS3:TP3:GFP and RbcS3:GFP were analyzed in comparison with the control transgenic plants PGD1:GFP and ZmUbi1:GFP (Park et al. 2010). b Chloroplast localization of GFP protein linked to transit peptide of RbcS3 and RbcS1 were conducted by expressing transiently in wild-type protoplast isolated from 10-day-old rice green seedlings. 35S:GFP was used as control of cytosolic localization. Protoplasts were visualized under confocal laser microscope (Leica). GFP gfp channel, Red red channel, Merged merged channel of Red and GFP image. Scale bar 10 µm. c Constructs for the transient expression assays. CaMV 35S promoter (P 35S) with transit peptide of RbcS3 (TP3) or RbcS1 (TP1) and without transit peptide linked to the GFP reporter gene and the CaMV 35S terminator (T 35S). The three constructs were kindly provided by Professor Sun Hwa Ha of Kyung Hee University

Full size image

To determine the GFP protein levels in the transgenic plants, western blot analysis was performed as shown in Fig. 5. The GFP protein levels in leaves of RbcS3:TP3:GFP plants were 0.12–0.23 % of the total soluble leaf proteins. These levels were higher or comparable to those of PGD1:GFP plants (GFP driven by the constitutive promoter PGD1). Taken together, these results indicate that the fusion of the RbcS3 promoter to its transit peptide, RbcS3:TP3, provides a chloroplast-targeting expression system.

Fig. 5
figure 5

Levels of GFP protein detected through western blot. Total soluble proteins were extracted in the 2-week-old leaf tissues of RbcS3:TP3:GFP transgenic rice plants. Non-transgenic (NT) and PGD1:GFP (GFP driven by the constitutive promoter PGD1) transgenic plants were used as a negative and a positive control, respectively. Numbers with sharp (hash) indicate transgenic lines. Ten microgram of total soluble proteins was loaded on a 12 % SDS–polyacrylamide gel and immunoblotted with anti-GFP antibodies. Levels of GFP proteins were compared with those of the recombinant GFP (rGFP)

Full size image

Discussion

In current study, we analyzed four different rice RbcS genes that encode a small subunit of ribulose bisphosphate carboxylase/oxygenase. They share the same levels of similarity in amino acid sequence with diverged transit peptides. We selected the RbcS3 promoter and its transit peptide for characterization to establish a chloroplast-targeted expression system. The RbcS3 gene is expressed only green tissues such as leaves and leaf sheaths. By promoter analysis and chloroplast localization of GFP proteins in protoplasts of RbcS3:TP3 transgenic plants, RbcS3 promoter and its transit peptide (TP3) sequences were found to direct chloroplast-targeted expression. The RbcS3 promoter was highly active in leaf tissues especially in the presence of its functional transit peptide sequence TP3. These results are consistent with results of RbcS1 and TP1 where TP1 enhanced the levels of transcripts significantly (Jang et al. 2002). Furthermore, transient expressions of constructs, 35S:TP3:GFP and 35S:TP1:GFP in rice leaf protoplasts showed that enhanced expression levels are similar between TP3 and TP1.

Chloroplast-targeted expression system is useful in the field of plant biotechnology, for example, insect-resistant crops. Insecticidal protein originated from Bacillus thuringiensis (Bt) is introduced into various crops for pest management (Barton et al. 1987; Vaeck et al. 1987; Fischhoff et al. 1987). Although constitutive promoters were often used to drive Bt gene expression in insect-resistant transgenic rice (Wu et al. 1997; Cheng et al. 1998; Alam et al. 1999; Tu et al. 2000; Chen et al. 2005; Tang et al. 2006), chloroplast-targeted expression of the gene seems more attractive for two reasons. Obviously, enhanced levels of the expression are the one that plant biotechnologists are seeking for. In fact, chloroplast-targeted expression of synthetic cry1Ac in transgenic rice increased the levels of plant resistance to rice pests tremendously. This was accomplished using RbcS1 and TP1 (Kim et al. 2009). Secondly, green tissue-specific expression system is desired not only because of the potential side effects derived from constant expression of the endotoxin (Kuiper et al. 2001; Shelton et al. 2002; Conner et al. 2003), but also because this approach helps solve the concerns about food safety of GM crops where there should be no or rare detection of the insecticidal protein at the edible parts of crops such as embryo and endosperm of grains (Kuiper et al. 2001; Shelton et al. 2002; Conner et al. 2003). In our study, GFP fluorescence was not detected in the seeds of RbcS3:TP3:GFP and RbcS1:TP1:GFP transgenic plants (Fig. 6). Thus, the RbcS3:TP3 provides an effective alternative for chloroplast-targeted expression system in crop biotechnology.

Fig. 6
figure 6

Observation of GFP fluorescence in transgenic grains. Two transgenic lines harboring RbcS3:TP3:GFP and RbcS3:GFP were analyzed in comparison with the control transgenic lines. The RbcS1:TP1:GFP and RbcS1:GFP (Jang et al. 2002) are positive controls for RbcS3:TP3:GFP and RbcS3:GFP, respectively. PGD1 (Park et al. 2010) and ZmUbi1 (Cornejo et al. 1993) are known constitutive gene promoters. Light and GFP images were captured using stereomicroscope (Oylmpus). Scale bar 2 mm

Full size image