Ectopic expression of Jatropha curcas APETALA1 (JcAP1) caused early flowering in Arabidopsis, but not in Jatropha

Plant materials and growth conditions

The roots, stems, mature leaves, inflorescence buds, flower buds, male flowers, female flowers and fruits of Jatropha were collected during the summer from Xishuangbanna, Yunnan Province, China. All of the tissues to be prepared for qRT-PCR were immediately frozen in liquid N₂ and stored at −80°C until use. The wild-type (WT) Arabidopsis thaliana Columbia ecotype (Col-0) and the ap1-11 mutant of the same ecotype were purchased from The Arabidopsis Information Resource (TAIR) website (http://www.arabidopsis.org/). The seeds of the Arabidopsis plants were germinated on 1/2 MS medium for one-week. Then, the seedlings were transferred to peat soil in plant growth chambers maintained at 22 ± 2 °C under long-day (16 light/8 h dark) or short-day (8 light/16 h dark) conditions. Phenotype analysis was performed on homozygous (T2) Arabidopsis plants and heterozygous (T0) Jatropha plants. More than 20 plants were used for the characterization of each Arabidopsis genotype. The number of rosette leaves and the number of days between transplantation to soil and appearance of the first visible flower bud were recorded. The aboveground tissues of 15-day-old Arabidopsis seedlings were harvested to analyze mRNA transcription levels.

Cloning of JcAP1 cDNA

Total RNA was extracted from the Jatropha flowers using the protocol described by Ding et al. (2008). First-strand cDNA was synthesized using M-MLV-reverse transcriptase according to the manufacturer’s instructions (TAKARA, Dalian, China). The full-length JcAP1 genomic DNA sequence (Sato et al., 2011) (http://www.kazusa.or.jp/jatropha/) was amplified via PCR using the primers XK928 and XK929 (Table S1), which introduced KpnI and SalI recognition sites, respectively. The JcAP1 cDNA containing full-length coding sequence was amplified from flower cDNA with the same primers. The PCR products were subsequently cloned into the pGEM-T vector (Promega Corporation, Madison, Wisconsin, USA) and sequenced. All primers used in this research were listed in Table S1.

Sequence and phylogenetic analyses

JcAP1 amino acid sequence was deduced according to the coding sequence. Related sequences were identified through a BLAST search (http://www.ncbi.nlm.nih.gov/BLAST/). To determine the amino acid identities, the alignment results were subjected to pairwise comparisons using DNAMAN 6.0. A phylogenetic tree based on the protein sequences was constructed with MEGA 5.0 (http://www.megasoftware.net). The amino acid sequences of AP1s, FULs, and CALs were assembled using ClustalX. A neighbor-joining phylogenetic tree was generated with MEGA 5.0 using the Poisson model, with gamma-distributed rates and 1,000 bootstrap replicates.

Plant expression vector construction and Arabidopsis and Jatropha transformation

To construct the 35S:JcAP1 plant overexpression vector, the JcAP1 sequence was excised from the pGEM-T vector (Promega, Corporation, Madison, Wisconsin, USA) using the restriction enzymes KpnI and SalI. Next, JcAP1 was cloned into the pOCA30 vector containing the CaMV 35S promoter. Transformation of Arabidopsis WT and ap1-11 mutant plants with the Agrobacterium strain EHA105 carrying the 35S:JcAP1 construct was performed using the floral dip method (Clough & Bent, 1998). Transformation of Jatropha with the Agrobacterium strain EHA105 carrying the same construct was performed according to the protocol described by Pan, Fu & Xu (2010) and Fu et al. (2015). All of the transgenic plants were confirmed using genomic PCR and RT-PCR.

Expression analysis via quantitative RT-PCR (qRT-PCR)

The roots, stems, mature leaves, inflorescence buds, flower buds, male flowers, female flowers and fruits of mature Jatropha plants and the aboveground tissues of 15 days Arabidopsis seedlings were collected for qRT-RCR detection. Total RNA was extracted from frozen Jatropha tissues as described by Ding et al. (2008). Total RNA was extracted from frozen Arabidopsis tissues using TRIzol reagent (Transgene, China). First-strand cDNA was synthesized with the PrimeScript^® RT Reagent Kit with gDNA Eraser (TAKARA, Dalian, China). The cDNA templates of first-strand cDNA were diluted 5-fold with sterilized double-distilled water. qRT-PCR was performed using SYBR^® Premix Ex Taq™ II (TAKARA, Dalian, China) on a Roche 480 Real-Time PCR Detection System (Roche, Mannheim, Germany). The primers employed for qRT-PCR are listed in Table S1. qRT-PCR was conducted with three independent biological replicates and three technical replicates for each sample. The data were analyzed using the 2${}^{-\Delta \Delta \mathrm{CT}}$ method described by Livak & Schmittgen (2001). The transcript levels of specific genes were normalized using Jatropha ACTIN1 or Arabidopsis ACTIN2.

Cloning and sequence analysis of JcAP1

A combined reverse transcriptase-polymerase chain reaction (RT-PCR) strategy was used to isolate AP1-like cDNA (JcAP1) from Jatropha. The JcAP1 coding sequence (CDS) (GenBank accession no. KR013222) is comprised of 732 bp and encodes a 243-amino acid protein showing 81%, 79%, 75%, and 71% sequence identity to Vitis vinifera VvAP1 (Calonje et al., 2004), Populus trichocarpa PtAP1 (Tuskan et al., 2006), Coffea arabica CaAP1 (De Oliveira et al., 2014), and AtAP1 (Mandel et al., 1992), respectively.

The genomic sequence of JcAP1 was 4,928 bp and consisted of eight exons and seven introns, which resembles the genomic structure of the Arabidopsis AP1 gene (Mandel et al., 1992). Multiple alignments were performed using the JcAP1 sequence and the sequences of AP1 homologs from other species. The MADS-box domain, K-Box domain, and euAP1 motif were determined (Fig. 1A). Since AP1 is closely related to CAL and FUL, and three genes exhibit high similarity and share redundant functions for floral meristem specification (Bowman et al., 1993), we undertook a phylogenetic analysis of the AP1/FUL MADS-box gene lineage (Fig. 1B), which is also called the SQUA lineage (Krogan & Ashton, 2000). The phylogenetic tree is divided into two clades, AP1 clade and FUL clade. CAL, which appears only in Brassicaceae, is clustered in the AP1 clade (Litt & Irish, 2003). Because CAL and AP1 originated from a recent duplication event <60 million years ago (Alvarez-Buylla, Garcia-Ponce & Garay-Arroyo, 2006). JcAP1 isolated in this study is clustered in the AP1 clade, while the JcFUL is clustered in the FUL clade. Moreover, JcAP1 is more closely related to AP1s of the Euphorbiaceae plants, such as Ricinus communis RcAP1, Manihot esculenta MeAP1 and Plukenetia volubilis PvAP1. In addition, the JcFUL is also closely related to Ricinus communis RcFUL (Fig. 1B). The results indicated that the AP1/FUL MADS-box gene phylogeny follows species phylogeny to interact with other MADS box proteins to confer sepal and petal identity (Mandel et al., 1992).

Expression pattern of JcAP1 in Jatropha

To assess the expression pattern of JcAP1 in Jatropha, we performed a qRT-PCR analysis using RNA extracted from the roots, stems, mature leaves, inflorescence buds, flower buds, male flowers, female flowers and fruits. The morphologies of different developmental stages of flower were shown in Fig. S3. JcAP1 was expressed in the inflorescence buds, flower buds, flowers, and fruits but showed very low expression in the roots, shoots and leaves. The expression profiles revealed that JcAP1 was highly expressed during the later stages of inflorescence buds (IB3) and early stages of flower buds (FB1). During the development of inflorescences, the expression levels of JcAP1 increased, whereas JcAP1 expression decreased during the development of flower organs. In the reproductive organs, the fruits showed the lowest expression level of JcAP1 (Fig. 2A). JcAP1 was expressed in all floral organs, particularly in the sepals and petals (Fig. 2B).

Constitutive overexpression of JcAP1 in Arabidopsis induces early flowering and abnormal flowers

To determine whether JcAP1 is involved in the regulation of flowering time, JcAP1 cDNA driven by the CaMV 35S promoter (Fig. 3A) was transformed into WT Arabidopsis. WT plants under the same growth conditions were used as a control. Transgenic plants were confirmed via qRT-PCR analysis of JcAP1 expression using the aboveground tissues of 15-day-old Arabidopsis seedlings. Thirty-four independent T0 transgenic lines were generated with the 35S:JcAP1 construct. Transgenic plants showed high JcAP1 expression level (Fig. S1A). In the majority of the transgenic lines, bolting occurred notably earlier than in WT plants under both long-day (LD) and short-day (SD) conditions.

We examined the phenotypes of two independent homozygous transgenic lines (L12 and L30) in the T2 generation. Arabidopsis plants ectopically expressing JcAP1 bolted 6–10 days earlier and produced 4–6 fewer rosette leaves than WT plants under LD conditions (Figs. 3B–3D and Table 1). Under SD conditions, Arabidopsis plants ectopically expressing JcAP1 flowered approximately 1–2 months earlier than WT (Figs. 3E–3G and Table 2). Therefore, the overexpression of JcAP1 in Arabidopsis significantly reduced the vegetative growth time.

In contrast to WT plants, the primary shoots of the transgenic plants were converted into compound terminal flowers consisting of two or three pistils surrounded by an abnormal number of sepals, petals, and stamens (Fig. 3I). Furthermore, the secondary shoots produced in cauline and rosette leaf axils were converted into solitary flowers. In extreme transgenic plants, all branches and inflorescences were replaced by solitary flowers (Fig. 3J).

Further analysis indicated that the promotion of flowering and abnormal terminal flowers in 35S:JcAP1 transgenic Arabidopsis was correlated with a significant up-regulation of the floral meristem identity genes AtLFY, AtFUL, AtAP1 and AtCAL and the floral organ identity genes AtAGAMOUS (AtAG), AtAP3 and AtSEPs (AtSEP1, AtSEP2, AtSEP3) (Fig. S1). The expression levels of these genes were highest in transgenic plants L12, whereas the AtTFL1 expression level was slightly down-regulated (Fig. S1). Thus, L12 also showed the most obvious changes in phenotype of extremely early flowering and solitary and terminal flowers (Fig. 3). Thus, the phenotypes of the early-flowering and abnormal terminal flowers produced due to the ectopic expression of JcAP1 in transgenic Arabidopsis were similar to those resulting from AtAP1 overexpression (Mandel & Yanofsky, 1995).

Constitutive overexpression of JcAP1 in ap1-11 mutant Arabidopsis induces early flowering and partially complements the phenotype

To further determine whether JcAP1 can function similarly to AtAP1, the 35S:JcAP1 construct (Fig. 3A) was transformed into Arabidopsis ap1-11 mutant plants. Eight independent T0 transgenic lines were generated and confirmed through qRT-PCR analysis of JcAP1 expression using RNA from aboveground tissues of 15-day-old Arabidopsis seedlings. WT and ap1-11 mutants under the same growth conditions were used as controls. Most of the transgenic lines bolted earlier than the WT and ap1-11 mutant plants under inductive LD conditions. The ap1 mutants didn’t exhibit significantly later flowering than WT (Table 3), which is because three homologous genes AP1, CAL, and FUL play redundant roles in control of flowering time. Each single mutant of ap1, cal or ful exhibited only slightly late flowering, whereas the triple mutant exhibited significantly late flowering (Ferrándiz et al., 2000).

To examine phenotypes, we selected two independent homozygous transgenic lines (C2 and C5) in the T2 generation that showed high JcAP1 expression levels: line C2 and line C5 (Fig. 4I). Complementary transgenic lines C2 and C5 bolted 6–11 days earlier and produced 2–6 fewer rosette leaves than the controls under LD conditions (Figs. 4A–4D and Table 3). In the extreme complementary transgenic line C5, solitary flowers appeared on the axils of rosette and cauline leaves and terminal flowers appeared on the primary shoots (Fig. 4D). The ap1-11 mutant flowers lacked petals, and new secondary floral buds developed on the axils of the bract-like organs present in the first whorl (Fig. 4F). The transgenic mutant C2 and C5 lines restored the development of sepals and petals, and axillary flowers were rarely seen at the bracts (Figs. 4G and 4H). Overexpression of JcAP1 in ap1 mutant leading to early flowering is because we used a strong promoter, the 35S promoter, which drives JcAP1 constitutively expressing. Similar results were found in Arabidopsis ap1 mutant overexpressing the chrysanthemum and lily AP1-like genes (Chen, Lin & Yang, 2008; Shchennikova et al., 2004). Further analysis indicated that the promotion of flowering in the 35S:JcAP1 transgenic Arabidopsis mutant was correlated with a significant up-regulation of the flower meristem identity genes AtLFY, AtSOC1 and floral organ identity genes AtSEPs (Figs. 4J and 4K).

These results demonstrate that the constitutive expression of JcAP1 complements the defect in floral organ development observed in the ap1-11 mutant; thus, JcAP1 functions as an A-class gene in transgenic Arabidopsis.

Overexpression of JcAP1 in Jatropha did not cause early flowering

Transgenic analysis performed in Arabidopsis suggested that JcAP1 might act as a floral identity gene in Jatropha. To test this hypothesis, we generated transgenic Jatropha with the 35S:JcAP1 construct (Fig. 3A) as previously described (Pan, Fu & Xu, 2010). Non-transgenic plants were used as a control. Fifty-five independent transgenic lines were confirmed via PCR using genomic DNA isolated from leaves of 2-month-old plantlets. And the partial results are shown in Fig. 5A. Next, JcAP1 expression levels in fourteen PCR-positive lines were examined through qRT-PCR using RNA extracted from young leaves of 2-month-old plantlets (Fig. 5B). To our surprise, all of these transgenic Jatropha lacked an early-flowering phenotype (Figs. 6A and 6B). When regenerated plantlets were grown in the field for 4 months, flower buds emerged in both transgenic and control plants (Figs. 6C–6H). We chose L2 and L20, which exhibited high and intermediate expression levels, respectively (Fig. 5B), to further analyze the expression levels of several floral identity-related genes in the shoot apices of 6-month-old plantlets. The results showed that the transcript levels of JcLFY, JcSOC1 and JcTFL1s (Fig. S2) were not significantly altered in both transgenic lines. The 35S promoter was highly active in the Jatropha inflorescence buds (Tao et al., 2015), but the inflorescence structure (Figs. 6D, 6F and 6H) and floral organ pattern (Fig. 6I) were not obviously different. These results indicate that JcAP1 may be inadequate to promote flowering and floral organ development by itself in Jatropha.