Background

Camellia japonica is one of the ten famous traditional flowers in China, which belongs to Camellia, Theaceae. The flower is colorful with ever green leaves, and usually regarded as the ornamental plants. Several studies about flower color have been carried out in Camellia. Through the research of flower gene CnF3H in Camellia nitidissima by Zhou (2015), it was found that the gene was most expressed in stamen, and the expression levels of the other parts from high to low were petals, sepals, pistils and bracts. Through the research on three Camellia F3H genes by Fan et al. (2016), CcF3H was found to be more stable in structure and more conservative in sequence that could be used as the most effective gene of overexpressed exogenous genes for the inheritance and breeding of Camellia floral colors. After a long period of artificial domestication, the ornamental variety of Camellia japonica contains abundant flower pattern changes, but the double flower phenotype of stamen petalody was in the majority. The formation of double flowers is affected by various factors and multiple pathways are found to be involved in double flower formation. Among these, the MADS-box gene family plays a key role in regulating the development of flower organs, which usually concentrates on floral organ development, flowering regulation and diversity of angiosperms (Alvare et al., 2000; Becker and Theissen, 2003; Kaufmann et al., 2005; Hemming and Trevaskis, 2011).

ABC model was initially put forward by Coen and coauthors in last century (Coen and Meyerowitz, 1991). According to the model, the properties of typical flower organs (sepal, petal, stamen, carpel) is mainly regulated by coordinated regulation of function genes of A, B and C (Carpenter and Coen, 1990; Bowman et al., 1991; Coen and Meyerowitz, 1991). Until 1995, the model was developed to ABCD model by Colombo et al. (1995). While at the beginning of this century, a more complex ABCDE model was developed. In general, A, B and C genes separately regulate the development of sepal, petal and stamen, respectively. Moreover, these three classes of genes can co-regulate the development of flower organs. A and B genes can regulate the development of petal together as well as B and C genes regulate the development of stamen. While D gene independently regulates the growth of ovule as well as E gene plays a role in the process of the development of petal, stamen, endoderm and ovule (Ferrario et al., 2004).

All regulatory genes in “ABCDE” model belong to MADS-box gene family except A gene APALATA2. MADS-box gene is a very important transcriptional regulatory factor. In angiosperm, C function gene AGAMOUS (AG) is the largest branch of MADS-box. However, in eudicotyledon Arabidopsis thaliana and Antirrhinum majus, two core branches were proved to exist at the same time: euAG and PLENA (PLE), and functional studies had found that these branches were also involved in the development of internafloral whorls of organs (Fourquin and FerrNdiz, 2012). After a long period of artificial domestication, the ornamental variety of camellia contains abundant changes of flower patterns. Double-flower camellia includes 5 types: semi-double type, peony type, rose type, amenone form and complete double type. It is still not clear about how ABCDE genes regulate the formation of double flower. The study on Camellia japonica gene CjAPL1 and CjAPL2 found that the expression level of A function genes was correlated with the degree of double flower, but overexpressed Arabidopsis plants didn’t have the double flower phenotypes (Sun et al., 2003a). The study on C gene CjAG of Camellia japonica found that two irreversible evolution paths existed in the expression changes of C function genes during the formation of double flower in Camellia japonica: (1) The expression level of C functional gene in complete double Camellia japonica ‘Shibaxueshi’ was entirely lost which was the same as the other double flowers species; (2) In amenone type (‘Jinpanlizhi’), instead of being inhibited, the expression level of C function genes increased, and ectopic expressed in the inner petals. In this study, focusing on the PLE branch of C function genes, gene cloning and evolution analysis were carried out in Camellia japonica. Combined with the gene expression pattern in wild single and double flower varieties, the regulation of CjPLE in flower development and double flower formation was discussed.

1 Results and Analysis

1.1 Homologous clone of CjPLE gene

Through the local BLAST comparison, a full length of 1,019 bp gene numbered c56140.graph_c0, homologous with PLE, was discovered in transcriptome data of Camellia japonica bud (Li et al., 2017), of which the open reading frame (ORF) was 738 bp and encoded a total of 245 amino acids. The 5’ non-coding region was 107 bp and 3’ non-coding region was 174 bp. Based on this sequence information, PCR primer was designed outside the ORF of the gene, and a 901 bp sequence was obtained through specific PCR amplification. Using ORF Finder to analyze CjPLE gene sequence, the result showed that the gene ORF was 780 bp and could encoded a complete protein. The 5’ non-coding region was 60 bp and 3’ non-coding region was 61 bp. Using BioEdit to compare the protein sequences of these two genes, the result showed that c56140.graph_c0 had more 14 amino acids than cloned CjPLE while other amino acids were identical. The above results indicated that the cloned CjPLE gene was a full-length gene and had been submitted to NCBI, the accession numbered was MF278983.

Through the analysis of protein structure encoded by CjPLE gene by CD-Search on NCBI, we found that the encoded protein contained a K-box (E-value: 9.74e-31) domain and a highly conservative region in the specific MADS-box (E-value: 9.42e-46) domain (Figure 1).

Figure 1 Conserved domains of coded protein from CjPLE gene

1.2 Sequence alignment and phylogenetic tree analysis of the homologous gene of CjPLE branch of C functional gene

To investigate the evolutionary relationship between CjPLE and other species, 18 protein sequences, which had high homology with CjPLE (referring to the phylogenetic tree of Alice Tadiello et al.), were selected to construct NJ (Neighbor-joining) phylogenetic tree using MEGA 7.0 (Figure 2). The result showed that CjPLE had the closest evolutionary relationship with TAG1 from tomato, and formed a PLE sub branch with Antirrhinum majus AmPLE (Alice et al., 2009).

Figure 2 Phylogenetic analysis of CjPLE gene and related sequences in other plant species Note: The proteins in this phylogenetic tree and their accession numbers: Apple (Malus domestica), ‘MdAG’AF401637, ‘Md15’CAC80858, ‘Md10’CAA04324; Arabidopsis, ‘SHP/AGL1’NP_191437, ‘SHP2/AGL5’NP_850377, ‘AG’NP_567569, ‘STK/AGL11’NP_001078364; Tomato (Solanum lycopersicum), ‘TAGL1’AAM33099, ‘TAG1’AAM33101; Peach (Amygdalus persica), ‘PpPLE’FJ88413, ‘PpFAR462’FJ184275, ‘PpSTK’ABQ85556; Rose (Rosa rugosa), ‘RrD1’AB025643; Snapdragon (Antirrhinum majus), ‘AmPLE’AAB25101; Strawberry (Fragaria ananassa), ‘FaAG’AAD45814; The outside stray are showed in black part (Alice et al., 2009)

1.3 Gene structure

Through the specific PCR amplification, the cDNA sequence of CjPLE was obtained. To study the complete structure of the gene, we constructed 4 libraries using Universal GenomeWalkerTM 2.0 kit by taking wild Camellia japonica as a template. Besides, two pairs of specific primers were designed to amplify target gene. However, only one pair of primers successfully amplified (Table 1). The primers were directed towards the 5’ terminal and 3’ terminal of the gene, respectively, and contained a repeating area, which ensured the acquisition of whole sequence.

Table 1 Primer sequence for cloning and expression of CjPLE gene in C. japonica

Genome walking result showed that the full length of the gene was 1,195 bp with two small introns of which were 33 bp and 118 bp, respectively (Figure 3).

Figure 3 The structure chart of CjPLE gene

1.4 Study of expression patterns of CjPLE gene from different tissues of Camellia japonica

MADS-box domains were detected in CjPLE through analysis. To further determine the gene function, real-time PCR experiment was carried out using cDNA extracting from wild Camellia japonica and several peony double flower cultivars ‘Hongluzhen’, ‘Rongqiu’, ‘Zhuangyuanhong’. The expression level of the gene in wild Camellia japonica flower bud was used as reference to calculate the relative expression level (Figure 4; Figure 5). CjPLE gene expression was different in various floral organs of wild Camellia japonica, of which the highest expression was carpel. While in peony double flower cultivars ‘Hongluzhen’ and ‘Zhuangyuanhong’, the gene had the highest expression level in stamen and had a higher expression level in inner petals and stamen in ‘Rongqiu’.

Figure 4 Relative expression level of CjPLE gene in different floral organs of wild Camellia japonica Note: YL: Young leaf; LB: Leaf bud; FB: Flower bud; SE: Sepal: PE: Petal; STA: Stamen; SA: Semet; SF: Filament; CA: Carpel

Figure 5 Relative expression level of CjPLE gene in different floral organs of ‘Hongluzhen’, ‘Zhuangyuanhong’ and ‘Rongqiu’ Note: H-SE: Sepals of ‘Hongluzhen’; H-OUTP: Out-petals of ‘Hongluzhen’; H-INP: In-petals of ‘Hongluzhen’; H-STA: Stamens of ‘Hongluzhen’; Z-SE: Sepals of ‘Zhuangyuanhong’; Z-OUTP: Out-petals of ‘Zhuangyuanhong’; Z-INP: In-petals of ‘Zhuangyuanhong’; Z-STA: Stamens of ‘Zhuangyuanhong’; R-SE: Sepals of ‘Rongqiu’; R-OUTP: Out-petals of ‘Rongqiu’; R-INP: In-petals of ‘Rongqiu’; R-STA: Stamens of ‘Rongqiu’

2 Discussion

2.1 Cloning and sequence analysis of homologous gene of PLE branch of C gene family in Camellia japonica

Related studies have been carried out on different genes of ABCDE in Camellia japonica, and the current cloned A function genes included CjAPL1 and CjAPL2 (Sun et al., 2013b); B function genes included CjDEF, CjGLO1, CjGLO2 and CjTM6 (Viaene et al., 2009); C function gene included CjAG. Based on the transcriptional information of Camellia japonica flower organs (Li et al., 2017), a full length 901 bp gene of PLE lineage CjPLE was obtained by homologous cloning. Sequence analysis indicated that the ORF length of the gene was 780 bp. The 5’ non-coding region was 60 bp and 3’ non-coding region was 61 bp. The gene complete coded a protein containing 260 amino acids, which contained highly conservative region of specific MADS-box domain and K-box domain. CjPLE had the closest evolutionary relationship with TAG1, and formed PLE subbranch with AmPLE. Through genomic amplification and sequence analysis, we obtained and analyzed the gene structure of CjPLE, and the results showed that the gene contained two small introns, 33 bp and 118 bp respectively, which was inconsistent with the structure of C function gene in Arabidopsis thaliana and Antirrhinum majus. The analysis of MAD-box family of the whole genome of jujube showed that the genetic structure of different family members was variated from 1 to 12 introns, indicating that the CjPLE gene of Camellia japonica might have the function ambiguity (Zhang et al., 2017).

2.2 CjPLE expression pattern suitable for functional mode of C gene and regulation of development of different double-flower pattern

Real-time fluorescence quantitative PCR was used to analyze the relative expression level of CjPLE gene in different flower organs of wild Camellia japonica, ‘Hongluzhen’, ‘Zhuangyuanhong’ and ‘Rongqiu’. The gene was highly expressed in the carpels of wild Camellia japonica. The gene expression was the highest in the stamens of the peony type double flower varieties ‘Hongluzhen and ‘Zhuangyuanhong’, and was higher in inner petals and stamens of ‘Rongqiu’. The study on C function gene CjAGL6 found that the expression level of CjAGL6 gene was the highest in stigmas, followed was petal stamens, filaments and ovary, and lowest in sepals and petal sepals (Sun et al., 2013a). CjPLE had the same expression pattern with CjAGL6, the general trend was that the expression level was low in outer organs while was high in inner organs. These analyses indicated that the expression pattern of CjPLE conformed to the functional action of C function genes and might play a regulatory role in the development of different double-flower.

3 Materials and Methods

3.1 Materials

The plant materials used in this research were taken from camellia conservation plot at Research Institute of Subtropical Forestry in Chinese Academy of Forestry (Hangzhou, China). The flower buds were stored in -80°C refrigerator after liquid nitrogen refrigeration grinding.

3.2 RNA extraction and cDNA synthesis of Camellia japonica

The total RNA of Camellia japonica bud was extracted by EASYspin Plus Plant RNA Kit (Elade, Beijing), and the first strand of cDNA was then synthesized by reverse transcriptase (Fevmentas, Canada). Through the local BLAST comparison, cDNA sequence that was homologous with PLE was found in transcriptional database of Camellia japonica bud, and primers P-1 and P-2 were designed using Primer 5.0 for specific PCR (Table 1). The target fragment was recovered by gel and connected to T-vector pMD20 (TaKaRa, Dalian), transformed into E. coli DH 5α (TaKaRa, Dalian), and positive monoclonal was selected for sequencing.

3.3 Sequence homologous alignment and phylogenetic tree analysis

Online BLAST (https://blast.ncbi.nlm.nih.gov/Blast.cgi) was applied for homologous sequence alignment of CjPLE gene, and the cDNA opening reading frame of the gene was searched online using ORF Finder (Web). BioEdit was used for sequence assembly of cDNA full-length and alignment of amino acid sequence. Phylogenetic tree was constructed and analyzed by MEGA7.0.

3.4 CjPLE gene structure determined by Genome walking

The full-length CjPLE was amplified using the Universal GenomeWalker^TM 2.0 kit. The young leaf DNA of wild Camellia japonica was extracted with the CTAB DNA rapid extraction kit (Aidlab) of, and the DNA quality was detected by 0.6% agarose gel electrophoresis.

25 µL DNA (0.1 µg/µL) was taken for enzyme digestion by Dra I, EcoR V, Pvu II and Stu I restriction enzymes (as blank control), respectively. The treatment conditions were: 37°C for 2 h, lightly mixed for 5~10 s and kept in 37°C overnight (16~18 h). The enzyme digestion was detected the completeness by 0.6% agarose gel electrophoresis. The DNA was purified and recycled by NucleoSpin Gel and PCR Clean-Up kit. The designed specific primers GSP3 and GSP4 were designed to amplified the sequence by TD-PCR.

3.5 Fluorescent quantitative PCR

Total RNA of various floral organs of Camellia japonica, ‘Hongluzhen’, ‘Rongqiu’ and ‘Zhuangyuanhong’ were extracted by EASYspin Plus Plant RNA Kit. Then, each of them was taken 800 mg to synthesize the first strand of cDNA with reverse transcriptase. According to the known full-length sequence of CjPLE gene, fluorescence quantitative primers PLE-Fa and PLE-Ra were designed (Figure 1). Using GAPDH as internal reference genes, GAPDH-F and GAPDH-R were designed as internal specific primers. The reagent of fluorescence quantitative reaction was SYBR^® Premix Ex Taq^TM II and the instrument was ABI 7300 Real-time PCR (America). The reaction was in 95°C for 30 s, 95°C for 5 s and 60°C for 31 s, repeating 40 cycles. Repeat every test and sample for 3 times, and the data was analyzed by related software of ABI 7300 Real-time PCR, Excel 2003 and 2^-△△CT method (Sun et al., 2013b).

Authors’ contributions

LT was the experiment designer and executor of this research who analyzed the data and wrote the first draft of this thesis. LJY, LXL, FZQ and YW took part in the test result analysis and thesis modification. NS and YHF were the constructors and chargers of this project, who direct experiment design, data analysis, paper writing and revision. All authors read and proved the final paper.

Acknowledgements

This study was jointly funded by the Central Public Welfare Scientific Research Institute (CAFYBB2016SZ001), the National Fund Project (project approval number: 31660049) and the Ningbo Special Agricultural Science and Technology Project (20-14c11002).

References

Alice T., Anna P., Dario Z., Elisabetta C., Lucia C., Giuseppe L., Rotino L.T., and Giorgio C., 2009, A PLENA-like gene of peach is involved in carpel formation and subsequent transformation into a fleshy fruit, 60: 651-661

Becker A., and Theissen G., 2003, The major clades of MADS box genes and their role in the development and evolution of flowering plants, Molecular Phylogenetics and Evolution, 29: 464-489

https://doi.org/10.1016/S1055-7903(03)00207-0

Bowman J.L., Drews G.N., and Meyerowitz E.M., 1991, Expression of the Arabidopsis floral homeotic gene agamous is restricted to specific cell types late in flower development, The Plant Cell, 3(8): 749-758

https://doi.org/10.2307/3869269

PMid:1726485 PMCid: PMC160042

Carpenter R., and Coen E.S., 1990, Floral homeotic mutations produced by transposon-mutagenesis in Antirrhinum majus, Genes Development, 4(9): 1483-1493

https://doi.org/10.1101/gad.4.9.1483

Coen E.S., and Meyerowitz E.M., 1991, The war of the whorls: Genetic interactions controlling flower development, Nature, 353(6339): 31-37

https://doi.org/10.1038/353031a0
PMid:1715520

Colombo L., Franken J., Koetje E., van Went J., Dons H.J., Angenent G.C., and van Tunen A., 1995, The petunia MADS box gene FBP11 determines ovule identity, The Plant Cell, 7(11): 1859-1868

https://doi.org/10.2307/3870193

PMid:8535139 PMCid: PMC161044

Fan J., Huang M.Y., Wu M.M., Ye H., and Zhang X.Y., 2016, Comparison of molecular characteristics, phylogeny and structure of three camellia F3H genes, Jiyinzuxue Yu Yingyong Shengwuxue (Genomics and Applied Biology), 35(5): 1195-1205

Ferrario S., Immink R.G., and Angenent G.C., 2004, Conservation and diversity in flower land, Curr. Opin. Plant Biol., 7(9): 84-91

https://doi.org/10.1016/j.pbi.2003.11.003
PMid:14732446

Fourquin C., and Ferr-Ndiz C., 2012, Functional analyses of agamous family members in Nicotiana benthamiana clarify the evolution of early and late roles of C-function genes in eudicots, The Plant Journal, 71(6): 990-1001

https://doi.org/10.1111/j.1365-313X.2012.05046.x
PMid:22563981

Hemming M.N., and Trevaskis B., 2011, Make hay when the sun shines: The role of MADS-box genes in temperature-dependant seasonal flowering responses, Plant Science, 180(3): 447-453

https://doi.org/10.1016/j.plantsci.2010.12.001
PMid:21421391

Kaufmann K., Melzer R., and Theissen G., 2005, MIKC-type MADS-domain proteins: structural modularity, protein interactions and network evolution in land plants, Genes, 347(12): 183-198

Li X.L., Li J.Y., Fan Z.Q., Liu Z.C., Tanaka T., and Yin H.F., 2017, Global gene expression defines faded whorl specification of double flower domestication in Camellia, Scientific Reports, 7(1): 3197-3201

https://doi.org/10.1038/s41598-017-03575-2
PMid:28600507 PMCid: PMC5466612

Sun Y.K., Li J.Y., Ying H.F., Fan Z.Q., and Zhou X.W., 2013a, Cloning and expression analysis of C function CjAGL6 gene cDNA from Camellia japonica ‘Jinpanlizhi’, Zhiwu Yanjiu (Bulletin of Botanical Research), 33(3): 330-338

Sun Y.K., Li J.Y., Ying H.F., Fan Z.Q., and Zhou X.W., 2013b, cDNA cloning and expression analysis of CjAPL1 gene from Camellia japonica, Linye Kexue Yanjiu (Forest Research), 26(4): 473-479

Viaene T., Vekemans D., Irish V.F., Geeraerts A., Huysmans S., Janssens S., Smets E., and Geuten K., 2009, Pistillata-Duplications as a mode for floral diversification in (Basal) Asterids, Mol. Biol. Evol., 26(11): 2627-2645

https://doi.org/10.1093/molbev/msp181
PMid:19679752

Zhang L.M., Zhao J., Feng C.F., Liu M.J., Wang J.R., and Hu Y.F., 2017, Genome-wide identification, characterization of the MADS-box gene family in Chinese jujube and their involvement in flower development, Scientific Reports, 7(1): 1025-1029

https://doi.org/10.1038/s41598-017-01159-8
PMid:28432294 PMCid: PMC5430891