Transcriptome analysis of two inflorescence branching mutants reveals cytokinin is an important regulator in controlling inflorescence architecture in the woody plant Jatropha curcas
In higher plants, inflorescence architecture is an important agronomic trait directly determining seed yield. However, little information is available on the regulatory mechanism of inflorescence development in perennial woody plants. Based on two inflorescence branching mutants, we investigated the transcriptome differences in inflorescence buds between two mutants and wild-type (WT) plants by RNA-Seq to identify the genes and regulatory networks controlling inflorescence architecture in Jatropha curcas L., a perennial woody plant belonging to Euphorbiaceae.
Two inflorescence branching mutants were identified in germplasm collection of Jatropha. The duo xiao hua (dxh) mutant has a seven-order branch inflorescence, and the gynoecy (g) mutant has a three-order branch inflorescence, while WT Jatropha has predominantly four-order branch inflorescence, occasionally the three- or five-order branch inflorescences in fields. Using weighted gene correlation network analysis (WGCNA), we identified several hub genes involved in the cytokinin metabolic pathway from modules highly associated with inflorescence phenotypes. Among them, Jatropha ADENOSINE KINASE 2 (JcADK2), ADENINE PHOSPHORIBOSYL TRANSFERASE 1 (JcAPT1), CYTOKININ OXIDASE 3 (JcCKX3), ISOPENTENYLTRANSFERASE 5 (JcIPT5), LONELY GUY 3 (JcLOG3) and JcLOG5 may participate in cytokinin metabolic pathway in Jatropha. Consistently, exogenous application of cytokinin (6-benzyladenine, 6-BA) on inflorescence buds induced high-branch inflorescence phenotype in both low-branch inflorescence mutant (g) and WT plants. These results suggested that cytokinin is an important regulator in controlling inflorescence branching in Jatropha. In addition, comparative transcriptome analysis showed that Arabidopsis homologous genes Jatropha AGAMOUS-LIKE 6 (JcAGL6), JcAGL24, FRUITFUL (JcFUL), LEAFY (JcLFY), SEPALLATAs (JcSEPs), TERMINAL FLOWER 1 (JcTFL1), and WUSCHEL-RELATED HOMEOBOX 3 (JcWOX3), were differentially expressed in inflorescence buds between dxh and g mutants and WT plants, indicating that they may participate in inflorescence development in Jatropha. The expression of JcTFL1 was downregulated, while the expression of JcLFY and JcAP1 were upregulated in inflorescences in low-branch g mutant.
Cytokinin is an important regulator in controlling inflorescence branching in Jatropha. The regulation of inflorescence architecture by the genes involved in floral development, including TFL1, LFY and AP1, may be conservative in Jatropha and Arabidopsis. Our results provide helpful information for elucidating the regulatory mechanism of inflorescence architecture in Jatropha.
KeywordsBranching Cytokinin Inflorescence Physic nut WGCNA analysis
Activated protein kinase
Adenine phosphoribosyl transferase
Atairp2 target protein
Counts per million mapped reads
Cox-reid profile-adjusted likelihood
Differentially expressed genes
Drought and salt tolerance
Duo xiao hua
False discovery rate
Generalized linear model
Long noncoding RNA
Real-time quantitative PCR
Quantitative trait locus
Rice terminal flower 1/centroradialis homologs
Shoot apical meristem
Small auxin up RNA
S ister of indeterminate spikelet
Suppressor of overexpression of constans 1
Squamosa promoter binding protein-like
Short vegetative phase
Tryptophanaminotransferase of Arabidopsis
Wealthy farmer’s panicle
Weighted gene correlation network analysis
Inflorescence architecture directly influencing plant reproductive success , is a key agronomic factor determining seed yield. In higher plants, inflorescence architectures exhibit remarkable diversity, which is derived from the inflorescence position on the shoot, the flower arrangement within inflorescence and the patterns and timing of flower initiation during reproductive development [2, 3]. Recently, a detailed review has been made on evolutionary developmental biology studies of inflorescence architecture . In inflorescence development, the maintenance of the inflorescence meristem and initiation of the floral meristem are critical to determine inflorescence architecture. In Arabidopsis thaliana, CLAVATA (CLV) pathways regulate meristem maintenance by restricting the expression of WUSCHEL (WUS), which defines the stem cell niche [5, 6]. A mutation in CLAVATA (CLV1, CLV2 or CLV3) results in the accumulation of meristem cells to generate an increased shoot meristem dome . Mutation in WUS causes defective shoot and floral meristems . In maize, thick tassel dwarf1 (td1) and fasciated ear2 (fea2) are orthologs of CLV1 and CLV2, respectively. Loss-of-function mutant td1 exhibits a fascinated ear with extra rows of kernels and a tassel with more dense spikelets . The fea2 mutant exhibits a massive ear inflorescence meristem and increased organ number .
In Arabidopsis, the transition time from the inflorescence to floral meristem is critical in the determination of inflorescence architecture; TERMINAL FLOWER1 (TFL1), LEAFY (LFY) and APETALA1 (AP1) participate in this process and regulate inflorescence branching patterns [10, 11]. TFL mutation promotes the conversion of the inflorescence meristem into the floral meristem, causing an abnormal inflorescence with a compound floral structure . By contrast, the ectopic expression of TFL1 in a transgenic plant induces a highly branched inflorescence phenotype in Arabidopsis . Conversely, LFY and AP1 repress the expression of TFL1 in the floral meristem [11, 14]. Moreover, SUPPRESSOR OF OVEREXPRESSION OF CONSTANS 1 (SOC1), SHORT VEGETATIVE PHASE (SVP), AGAMOUS-LIKE 24 (AGL24), and SEPALLATA 4 (SEP4) redundantly regulate inflorescence architecture by directly suppressing the expression of TFL1; a soc1–2 agl24–3 svp-41 sep4–1 quadruple mutant displays a massive inflorescence branching phenotype .
Overexpression of rice RCN1 and RCN2 and maize ZCN1-ZCN6, which are homologous to Arabidopsis TFL1, causes highly branched inflorescences through maintaining the indeterminacy of the inflorescence meristem [16, 17]. Rice TAWAWA1 (TAW1) regulates inflorescence architecture by suppressing the specification of the spikelet meristem to maintain the indeterminacy of inflorescence architecture . A gain-of-function mutant, tawawa1-D, exhibits a highly branched inflorescence and increased number of spikelets, whereas a TAW1 knockdown mutant displays a small, reduced branching inflorescence resulting from early termination of inflorescence meristems and the formation of spikelet meristems . In rice, the WEALTHY FARMER’S PANICLE (WFP)/IDEAL PLANT ARCHITECTURE (IPA1) locus is linked to a Squamosa Promoter Binding Protein-Like 14 (OsSPL14) gene that can be suppressed by OsmiR156 . A single nucleotide change in OsSPL14 that relieves the repression of OsmiR156 causes increased panicle branching and grain number [19, 20]. Overexpression of a maize UNBRANCHED3 (UB3), an ortholog of rice OsSPL14, dramatically repressed tillering and panicle branching in rice; however, moderate expression of UB3 slightly suppressed tillering, but promoted panicle branching, resulting in an increased grain number per panicle in rice [21, 22].
In maize, two APETALA2-like genes, indeterminate spikelet1 (ids1) and sister of indeterminate spikelet 1 (SID1), are required for inflorescence branching by regulating the initiation of spikelet meristem and floral meristem, and an ids1 sid1 double mutant shows a reduced branching inflorescence phenotype . Three maize genes, RAMOSA1 (RA1), RA2 and RA3, regulate the fate of the inflorescence meristem, and mutations in these genes cause an increased long-branching inflorescence [24, 25, 26].
Plant hormones such as auxin and cytokinin are required for the initiation and outgrowth of axillary meristems that generate inflorescence branches and florets during reproductive development [27, 28]. Several auxin biosynthesis and polar transport components, such as YUCCA (YUC), PIN-FORMED 1 (PIN1) and PINOID (PID), are important regulators in inflorescence development . In Arabidopsis, the YUC gene encodes flavin monooxygenases (FMOs) that catalyze a rate-limiting step in tryptophan-dependent auxin biosynthesis; overexpression of YUC promotes auxin levels during development processes . In maize, sparse inflorescence1 (spi1), a homologous Arabidopsis YUC, encodes a FMO, and spi1 mutation reveals defects in the initiation of axillary meristems and lateral organs, causing reduced branch number and floral organs in inflorescences . Auxin transport proteins PIN1 and PID are required for the distribution of auxin; mutation in pin1 or pid causes a pin-like inflorescence with abnormal flowers because of defects in the initiation of the axillary meristem in Arabidopsis [31, 32]. Maize barreninflorescence2 (bif2), an ortholog of PID, is required for the initiation of the axillary meristem and lateral primordia; the bif2 mutant displays a reduced number of branches, spikelets, florets and kernels in inflorescences . Barren inflorescence1 (bif1) mutation causes a similar phenotype to that of the bif2 mutant ; the barren stalk1 (ba1) mutant displays an unbranched inflorescence without spikelets because of a defect in auxin signaling . In Setaria viridis, SvAUXIN1 is required for inflorescence development and its loss-of-function mutant sparse panicle1 (spp1) displays reduced and uneven inflorescence branching phenotype .
Cytokinin is required for meristem activity and plays a positive role in the shoot meristem . In Arabidopsis, CYTOKININ OXIDASE 3 (CKX3) and CKX5 catalyze the degradation of cytokinin; ckx3 ckx5 double mutant display larger inflorescences and floral meristems because of the accumulation of cytokinin . In rice, a QTL locus, Gn1, encodes an OsCKX2; the reduced expression of OsCKX2 causes higher cytokinin levels in the inflorescence meristem to increase the number of branches and spikelets, leading to enhanced grain yield . A zinc finger transcription factor, DROUGHT AND SALT TOLERANCE (DST), directly regulates the accumulation of cytokinin in the shoot apical meristem (SAM) to promote panicle branching loading to the increase of grain number . The LONELY GUY (LOG) gene encodes an enzyme that participates in the final step of bioactive cytokinin synthesis; mutation in log causes a small inflorescence with reduced branch and spikelet number .
At present, the studies on molecular mechanism of inflorescence architecture are focus on the model plants and few in perennial woody plants because of their long reproductive cycle and the difficulty in the establishment of genetic transformation system. To understand the regulatory mechanism of inflorescence architecture, the investigation of the evolutionary changes in developmental morphology and the identification of key factors controlling inflorescence development are crucial in closely related plant lineages that display different inflorescence structures , especially, in perennial woody plants.
Jatropha curcas L. has high seed oil content and is considered a potential biofuel plant [41, 42]. The Jatropha inflorescence exhibits a dichasial cyme pattern bearing male and female flowers in the same inflorescence. Few female flowers per inflorescence is considered one of the factors leading to poor seed yield in Jatropha . Our previous research showed that co-suppression of JcLFY delayed flower formation, leading to production of more secondary inflorescence branches , but roles of JcTFL1b [44, 45] and JcAP1  in inflorescence branching remain unclear. Paclobutrazol (PAC), an inhibitor of gibberellin biosynthesis, causes compacted inflorescences with short branches, resulting in increased seed yield in Jatropha [47, 48]. Cytokinin treatment on inflorescence buds generated a larger inflorescence with a significantly increased the number of female and total flowers [43, 49, 50], which means a high-branch inflorescence phenotype. These results suggest that cytokinin may play a significant role in the determination of inflorescence architecture in Jatropha. In this study, two Jatropha mutants that exhibit different inflorescence branching traits were used for comparative transcriptome analysis to identify genes and regulation networks that participate in the regulation of inflorescence architecture. Our study will contribute to the understanding of the genetic basis of inflorescence architecture and to the breeding of high-yield Jatropha varieties.
dxh and g mutants have different inflorescence branching phenotypes
Identification of differentially expressed genes
Differentially expressed genes involved in inflorescence development
Differentially expressed genes involved in auxin and cytokinin metabolic and signaling pathways
Confirmation of expression profiles of candidate genes by real-time qPCR
To validate the result of transcriptome analysis, we selected a dozen candidate genes to test their expression patterns across the five group samples using the real-time qPCR method. These genes included JcFUL, JcSEP1, JcSEP2A, JcSEP2B, JcSEP3, JcTFL1, JcCD35911.0, JcCD37832.1, JcCD39325.1, JcCD47611.19, JcCD53029.968 and JcCD53029.1529 (Additional files 9 and 10). Correlation analysis showed that the expression patterns of these genes displayed by RNA-Seq were consistent with that by real-time qPCR (Additional file 11), indicating that the transcriptome results in this study are reliable.
Construction of gene co-expression networks and identification of hub genes that regulate inflorescence branching
Application of 6-benzylaminopurine (6-BA) promoted an increase in inflorescence branches in both g mutant and WT plants
To validate the functions of cytokinin and auxin in inflorescence branching, we applied 6-BA (cytokinin) and 1-naphthaleneacetic acid (NAA, auxin) to the inflorescence buds of low-branch mutant (g) and WT plants, respectively. After 6-BA treatment, more than 70% of inflorescences displayed high-branch phenotype in both g mutant and WT plants, although the growth of some inflorescence buds was arrested (Additional file 15 A, E and C, G). A three-order branch inflorescence became a four-order branch inflorescence in g mutant plants (Additional file 15 B, F and A, E); and a five-order branch inflorescence became a six-order branch inflorescence in WT plants (Additional file 15 D, H and C, G), which is similar to the high-branch inflorescence phenotype of dxh mutant. We postulate that a strong cytokinin activity is present in dxh mutant inflorescence buds than in WT ones, which results in the high-branch inflorescence phenotype while a weak cytokinin activity in g mutant inflorescence buds leads to the low-branch inflorescence phenotype. However, NAA treatment had no effect on inflorescence branching (data not shown). These results indicate that cytokinin is an important regulator in regulating inflorescence branching in Jatropha.
In inflorescence development, cytokinin and auxin are essential for the formation and growth of the inflorescence branch meristem in Arabidopsis and rice [27, 28]. A gene co-expression network analysis showed that six homologous genes involved in the cytokinin metabolic pathway, and four genes in the auxin biosynthetic and signaling pathways are hub genes, which are from modules highly associated with inflorescence architecture phenotypes (Figs. 7 and 8), indicating that cytokinin and auxin likely to play important roles in regulating inflorescence branching in Jatropha. In cytokinin metabolic process, ADK catalyzes the phosphorylation of adenosine to AMP and converts cytokinin nucleosides to nucleotides contributing to intracellular CK homeostasis [54, 55]. ATP1 converts active cytokinin to inactive form and loss of ATP1 activity causes high accumulation of cytokinin bases evoking abnormal cytokinin-regulated responses . Isopentenyltransferases (IPTs) catalyze the formation of isopentenyladenosine 5′-monophosphate (iPMP) from AMP and dimethylallylpyrophosphate (DMAPP), which is the first step of the cytokinin biosynthetic pathways . LOG promotes cytokinin activity, and log mutants display a deficient in the maintenance of the shoot meristem in rice . Cytokinin oxidases/dehydrogenases (CKXs) catalyze the irreversible degradation of cytokinins in cytokinin metabolic pathways . In pairwise low-branch vs. high-branch mutant (gII_dxhII) and low-branch mutant vs. WT (gII_ckII) inflorescences, the expression of JcIPT1 and JcIPT5 was downregulated while the expression of JcLOG1, JcLOG3, JcLOG5, JcCKX1, JcCKX3, and JcCKX7 was upregulated (Fig. 5). These results indicated that cytokinin biosynthesis is decreased, but cytokinin activation and degradation are promoted in low-branch mutant (g) inflorescence compared to those in high-branch mutant (dxh) and WT inflorescences. These effects may cause the low-branch inflorescence phenotype of g mutant.
In Arabidopsis, TRYPTOPHANAMINOTRANSFERASE OF ARABIDOPSIS 1 (TAA1) catalyzes the conversion from L-tryptophan (Trp) to indole-3-pyruvic acid (IPA), and YUC FMOs catalyze the oxidative decarboxylation of IPA to generate indole-3-acetic acid (IAA), both of which are the key enzymes in auxin biosynthesis [59, 60]. The expression of JcTAA1 and JcYUC4 was downregulated, indicating that auxin biosynthesis is decreased in low-branch (g) mutant inflorescence compared to that in high-branch (dxh) mutant inflorescence, along with the downregulated expression of JcPIN1 and JcPID whose homologous genes serve as efflux carriers in auxin polar transport in Arabidopsis [61, 62] (Fig. 5). In addition, the expression of JcSAUR20, an auxin responsive gene, was downregulated, and JcIAMT1 and JcCHS were upregulated, confirming that auxin biosynthesis may be decreased in low-branch mutant (g) inflorescence. In Arabidopsis, IAMT1 encodes an IAA carboxyl methyltransferase that converts IAA to methyl-IAA ester (MeIAA); the overexpression of MeIAA causes dramatic hyponastic leaf phenotypes . CHS encodes the first enzyme in flavonoid biosynthesis that is considered an auxin transport inhibitor . However, there is no reasonable explanation for the upregulated expression of JcAAO1 and JcCYP83B1 genes, which are also involved in the auxin biosynthetic signaling pathways .
The application of 6-BA to inflorescence buds resulted in increased inflorescence branches both in low-branch mutant (g) and WT plants (Additional file 15). In our previous research, 6-BA treatment on WT inflorescence buds significantly promoted total flower number per inflorescence meaning a high-branch inflorescence phenotype, which is positive correlation to 6-BA concentration . Thidiazuron (TDZ), another synthetic compound with cytokinin activity, was also shown to promote initial inflorescence branching in WT plants, although the final branch number will be decreased because of the abortion of flower buds . These results supported that cytokinin is an important regulator in controlling inflorescence branching, which is in agreement with our results of WGCNA analysis. Mutation of several genes involved in the cytokinin metabolic or signaling pathways causes abnormal inflorescence branching phenotypes in Arabidopsis and rice [37, 38, 39], suggesting that cytokinin might has a conserved role in regulating inflorescence branching in different species.
Using WGCNA, we identified several hub genes involved in the cytokinin metabolic pathway from modules highly associated with inflorescence architecture phenotypes. The application of cytokinin (6-BA) to inflorescence buds induced high-branch inflorescences both in low-branch mutant (g) and WT plants. These results supported that cytokinin is an important regulator and may play vital role in controlling inflorescence branching in Jatropha. Several Arabidopsis homologous genes involved in inflorescence development is significantly differentially expressed in inflorescence buds between mutants and WT plants, indicating that they participate in the regulation of inflorescence architecture in Jatropha. Based on the above results, we speculate that the change of inflorescence branching phenotype of two mutants may result from mutations at one or more loci in genome regions that contain genes involved in cytokinin metabolism and/or in inflorescence development. Our results will be helpful for elucidating the regulatory mechanism of inflorescence architecture in Jatropha.
Plant growth conditions and cytokinin (6-benzyladenine, 6-BA) treatment on inflorescence buds
The wild-type (WT), duo xiao hua (dxh) and gynoecy (g) mutant were grown in field in the Xishuangbanna Tropical Botanical Garden (XTBG) of the Chinese Academy of Sciences (21° N, 101° E) located in Mengla County, Yunnan Province, China. WT plant has a four-order branch inflorescence under normal growth conditions, occasionally a three- or five-order branch inflorescence. The dxh mutant has a high-branch inflorescence phenotype, a seven-order branch inflorescence, which was derived from a mutagenized population treated with cobalt-60 gamma rays. Its selfing progeny was selected until the stable high-branch inflorescence phenotype. The g mutant has a low-branch inflorescence phenotype, a three-order branch inflorescence, which was derived from a natural variation . The cutting-propagated plants from single plant with stable phenotype were used for the preparation of experimental materials. Two-year-old cutting-propagated plants were grown in a field at 2 × 2 m per plant at the XTBG.
To confirm the effect of cytokinin on inflorescence branching, middle stage inflorescence buds (about 0.8 cm in diameter) that growth approximately 7–10 days from occurrence of invisible inflorescence bud of low-branch mutant (g) and WT plants were selected for once treatment with 1.0 mM 6-BA solution containing 0.05% Tween-20. The 6-BA and mock solutions were sprayed onto inflorescence buds wetting them to the point of run-off, ten inflorescence buds from three to five plants per treatment. After 2–3 weeks, inflorescence phenotypes were surveyed. Seed oil content was measured by using the minispec mq-one Seed Analyzer (Bruker Optik GmbH, Germany) as described previously , three replicates for each sample.
Statistics of traits of high-branch inflorescence mutant (dxh) and WT
During the suitable period, the total flower number, female flower number, ratio of female-to-male flowers, fruit number, seed number, seed yield, weight of 100 seeds, and seed oil content per inflorescence/infructescence were surveyed in high-branch inflorescence mutant (dxh) and WT, respectively. In total 33 inflorescence/infructescence are surveyed in dxh and 44 ones in WT. Statistical test analysis was performed using the Welch two sample t-test in R software (https://cran.r-project.org).
Collection of samples, RNA isolation and library construction
At initial reproductive period, after removed leaves, shoot tips that can generate inflorescence meristem were harvested from dxh and WT plants. Inflorescence buds that growth approximately 3–4 days from occurrence of invisible inflorescence bud (about 0.4 cm in diameter) were harvested from dxh and g mutants and WT plants. Three shoot tips or inflorescence buds were pooled as one biological replicate for RNA isolation, three replicates per sample. Total RNA extraction, library construction and quality control were performed as previously described . Sequencing was performed on an Illumina Hiseq 2500 platform by Novogene Bioinformatics Technology (Beijing, China).
De novo transcriptome assembly and read mapping
Raw reads were treated with the Fastq_clean  and assessed with FASTQC (http://www.bioinformatics.babraham.ac.uk/projects/fastqc). De novo transcriptome assembly was performed using Trinity (version 2.0.6) with default parameters [68, 69]. In all 122,526 sequences were generated. Bowtie version 1.1.1 (−v 2 -m 10) was used for the mapping of the paired-end reads from each library .
Identification of differentially expressed transcripts
The Corset (version 1.03) was used for abundance estimation of transcripts . Differentially expressed transcripts (DEGs) with a false discovery rate (FDR) of < 0.05 were identified by using the edgeR package . The Venny (version 2.1) was used for the generation of venn diagram of DEGs (http://bioinfogp.cnb.csic.es/tools/venny/index.html). Hierarchical clustering of transcripts was performed using the pheatmap R package (version 1.0.7) (https://github.com/cran/pheatmap).
Annotation of transcripts
A total of 16,206 filtered transcripts used for differentially expressed analysis were annotated with BLASTX search against the Ensembl Plants database (http://plants.ensembl.org) with Evalue < 1.0E-05. Among them, 14,680 transcripts were annotated and 1526 were not found (Additional file 16). Ten transcripts displayed in Fig. 3 were defined as LncRNAs because they have not annotation, not coding protein and length > 200 bp .
Validation of expression profiles of candidate genes by real-time PCR (qPCR)
RNA samples for qPCR are same as RNA-seq ones. The cDNA was synthesized from total RNA (1.0 μg) using a PrimeScript RT Reagent Kit (Takara, Otsu, Japan), for each sample. qPCR was performed on a LightCycler 480 II (Roche, Penzberg, Germany) using the SYBR green I Kit (Roche), with three independent biological replicates for each sample and three technical replicates. JcGAPDH was as the internal reference. Primers for qPCR were listed in Additional file 10. The relative expression levels of genes were calculated by the 2−ΔΔ CT method. Correlation analysis between RNA-Seq and qPCR expression data of the genes is performed with cor.test in R software (https://cran.r-project.org).
Construction and analysis of weighted gene co-expression networks
The raw count data of differentially expressed genes, which were transformed with Log2(x + 1), from edgeR were applied to construct co-expression networks using the R package weighted gene correlation network analysis (WGCNA) . The soft thresholding power of 6 was chosen based on the criterion of approximate scale-free topology. The minimum module size was 30, and modules were merged with the cutoff value of 0.2. The interaction network was visualized using the Cytoscape software .
The authors gratefully acknowledge the Central Laboratory of the Xishuangbanna Tropical Botanical Garden for providing high-performance computing and other research facilities.
Z-FX and M-SC designed the experiments and wrote the manuscript. M-SC analyzed the data and drafted the manuscript. M-SC and G-JW mutated Jatropha to obtain the desired mutant. M-SC, M-LZ, H-YH, XB, B-ZP, QF, Y-BT, MT and JMH carried out experiments. All authors reviewed the final manuscript. All authors agree to be accountable for the content of the work.
This work was supported by funding from the National Natural Science Foundation of China (31670612, 31370595, 31300568, and 31500500), the Program of Chinese Academy of Sciences (kfj-brsn-2018-6-008), and the CAS 135 program (2017XTBG-T02). Funding bodies were not involved in the design of the study, data collection and analysis, or interpretation of the manuscript.
Ethics approval and consent to participate
Consent for publication
The authors declare that they have no competing interests.
- 6.Laux T, Mayer KF, Berger J, Jurgens G. The WUSCHEL gene is required for shoot and floral meristem integrity in Arabidopsis. Development. 1996;122(1):87–96.Google Scholar
- 18.Yoshida A, Sasao M, Yasuno N, Takagi K, Daimon Y, Chen R, Yamazaki R, Tokunaga H, Kitaguchi Y, Sato Y, Nagamura Y, Ushijima T, Kumamaru T, Iida S, Maekawa M, Kyozuka J. TAWAWA1, a regulator of rice inflorescence architecture, functions through the suppression of meristem phase transition. Proc Natl Acad Sci. 2013;110(2):767–72.PubMedCrossRefPubMedCentralGoogle Scholar
- 41.Sato S, Hirakawa H, Isobe S, Fukai E, Watanabe A, Kato M, Kawashima K, Minami C, Muraki A, Nakazaki N, Takahashi C, Nakayama S, Kishida Y, Kohara M, Yamada M, Tsuruoka H, Sasamoto S, Tabata S, Aizu T, Toyoda A, Shin-i T, Minakuchi Y, Kohara Y, Fujiyama A, Tsuchimoto S, Kajiyama S, Makigano E, Ohmido N, Shibagaki N, Cartagena JA, et al. Sequence analysis of the genome of an oil-bearing tree, Jatropha curcas L. DNA Res. 2011;18(1):65–76.PubMedCrossRefPubMedCentralGoogle Scholar
- 48.Song J, Chen M-S, Li J, Niu L, Xu Z. Effects of soil-applied paclobutrazol on the vegetative and reproductive growth of biofuel plant Jatropha curcas. Plant Diver Resour. 2013;35(2):173–9.Google Scholar
- 51.Chen M-S, Pan B-Z, Fu Q, Tao Y-B, Martínez-Herrera J, Niu L, Ni J, Dong Y, Zhao M-L, Xu Z-F. Comparative transcriptome analysis between gynoecious and monoecious plants identifies regulatory networks controlling sex determination in Jatropha curcas. Front Plant Sci. 2017;7:1953.PubMedPubMedCentralGoogle Scholar
- 67.Zhang M, Zhan F, Sun H, Gong X, Fei Z, Gao S: Fastq_clean: an optimized pipeline to clean the Illumina sequencing data with quality control. In: Proceedings of the IEEE International Conference on Bioinformatics and Biomedicine (IEEE BIBM 2014). Belfast, UK: 2014;44–48.Google Scholar
- 68.Grabherr MG, Haas BJ, Yassour M, Levin JZ, Thompson DA, Amit I, Adiconis X, Fan L, Raychowdhury R, Zeng Q, Chen Z, Mauceli E, Hacohen N, Gnirke A, Rhind N, di Palma F, Birren BW, Nusbaum C, Lindblad-Toh K, Friedman N, Regev A. Full-length transcriptome assembly from RNA-Seq data without a reference genome. Nat Biotechnol. 2011;29(7):644–52.PubMedPubMedCentralCrossRefGoogle Scholar
- 69.Haas BJ, Papanicolaou A, Yassour M, Grabherr M, Blood PD, Bowden J, Couger MB, Eccles D, Li B, Lieber M, MacManes MD, Ott M, Orvis J, Pochet N, Strozzi F, Weeks N, Westerman R, William T, Dewey CN, Henschel R, LeDuc RD, Friedman N, Regev A. De novo transcript sequence reconstruction from RNA-seq using the trinity platform for reference generation and analysis. Nat Protoc. 2013;8(8):1494–512.PubMedCrossRefPubMedCentralGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.