A combined reverse transcriptase polymerase chain reaction (RT-PCR) and rapid amplification of cDNA ends (RACE) strategy was used to isolate PI-like gene from pineapple. The full-length gene designated AcPI (Genbank accession number HQ717796) was isolated. AcPI cDNA is 907 bp in length and contains an open reading frame of 594 bp, which encodes a protein of 197 amino acids. It also contains 101 bp 5′ and 183 bp 3′ untranslated regions and a poly (A) tail. Molecular weight was 2.29 kDa and isoelectric point was 9.28. A 650 bp fragment of AcPI with enzyme recognition sites and coding region was obtained and used for transformation. An alignment of the deduced amino acid sequence of AcPI (Ananas comosus), HPI2 (Hyacinthus orientalis, AAD22494), VvPI (Vitis vinifera, ABK59993), BpPI (Betula pendula, CAD32764), MdPI (Malu × domestica, CAC28022), HmPI (Hydrangea macrophylla, BAG68951), BjPI (Brassica juncea, AAY63867), AtPI (Arabidopsis thaliana, BAA06465), MtPI (Medicago truncatula, ACJ36228), OrcPI (Orchis italica, BAC22579), CsPIC2 (Crocus sativus, ABB22781) and AoPI (Alpinia oblongifolia, ABB92623) was performed using the DNAMAN 6.0 program (Figure 1).

Phylogenetic analysis was conducted using BnPI (Brassica napus, ABW74343), PnPI-1 (Papaver nudicaule, AAC42570), CDM86 (Chrysanthemum × morifolium, AAO22986), SoPI (Spinacia oleracea, AAT69985), PePI (Passiflora edulis, AER30449), PPI (Capsicum annuum, ADR83606), NGL9 (Medicago sativa, AAK77938), PsPI (Pisum sativum, CAC37031), TaPI (Trochodendron aralioides, ABQ85946), HoPI (Hyacinthus orientalis, AAD22494), Os05g0423400 (Oryza sativa Japonica Group, BAF17504), FEG1 (Elaeis guineensis, AAQ13915), PIA2 (Crocus sativus, ABB22778), PIA1 (Crocus sativus, ABB22777), MADS2 (Hyacinthus orientalis, AAD22493) and the same complete amino acid sequences as in Figure 1 and generated a rooted tree, by the observed divergency method with DNAMAN 6.0 program (Figure 2). The phylogenetic tree showed that AcPI protein (red-boxed) was more closely related to the other monocot PI proteins than to their dicot counterparts. AcPI had a high identity with CsPIC2 (78.6%), AoPI (77.4%), OrcPI (75.7%) and HPI2 (72.4%).

To determine the detail expression patterns of the AcPI in different tissues and in different stages, the mRNA levels were examined by qRT-PCR. Relative quantification was performed using the comparative cycle threshold (Ct) method, in which simplified arithmetic formulas (2-ΔΔCt) are used to obtain the same result as that yielded by the relative standard curve method, when the target gene and the reference control gene have approximately equal amplification efficiency (94.2% and 94.5% respectively, Figure 3A,B). The expression number in one tissue or stage (AcPI in bract and on 10 days after inducing) was set to a value of 1 and subsequently, expression levels were relative to this number. The qRT-PCR analyses showed that the AcPI was weakly expressed in fruit stems, bracts, leaves and sepals, strongly in petals and carpels, and moderately in apical meristems, flesh, and stamens (Figure 3C).

The expression levels of AcPI in different stages (Figure 3D) decreased a little at 10 days and 20 days after flower inducing with ethephon. It increased at 30 days after flower inducing and reached the highest level at 40 days after flower inducing when the multiple fruit and floral organs were forming.

To investigate the function of AcPI, ectopic expression of AcPI in transgenic Arabidopsis plants was analyzed. Twelve independent PCR-positive 35S::AcPI transgenic T1 plants were obtained (Figure 4). All transgenic plants were phenotypically distinguishable from wild type plants. Firstly, these 35S::AcPI transgenic plants flowered significantly earlier than wild type plants by producing only 8 to 10 small rosette leaves (Figure 5A). When the wild type plants flowered, they had over sixteen big rosette leaves (Figure 5B). Secondly, each 35S::AcPI transgenic plant had 3 to 6 inflorescences or branches (Figure 5C), while the wild type plant had one inflorescence or branch (Figure 5D). No obvious alteration of floral organs was observed in transgenic plants.

Multiple sequence alignment demonstrated that AcPI protein possessed a typical MIKC-type domain structure: the highly conserved MADS (M) domain which is the major determinant of DNA-binding and performs dimerization and accessory factor binding functions [13], the relatively weakly conserved intervening (I) domain which constitutes a molecular determinant for the selective formation of DNA-binding dimmers [28], the less conserved keratin-like (K) domain which is proposed to allow for the formation of an amphipathic helix involved in protein dimerization [13,29], and the divergent C-terminal (C) region which contains a highly conserved PI-motif (FxFRVQPxQPNLQE) [30]. All the conserved regions are related to the function of PI-like proteins [31].

In higher eudicotyledonous angiosperms, the floral organs include sepals, petals, stamens, ovules and carpels. According to the ABCDE model, the identity of these organs is specified by floral homeotic genes of class A, A + B + E, B + C + E, D + E and C + E respectively [27,32]. To explain the Liliaceae flower morphology, van Tunen et al. [23] reported that class B genes were not only expressed in petals and stamens, but also in sepals. Thus the organs of both sepals and petals express class A, plus class B genes and therefore, get the same petaloid identity. Kanno A et al. [33] reported that two DEF-like and one GLO-like genes were expressed in whorls 1, 2 and 3. In addition, TGGLO was also weakly expressed in carpels, leaves, stems and bracts. Expression of B-class genes in Arabidopsis and Antirrhinum was known to be flower-specific [8, 34–36].

In this paper, the expression profile of AcPI is different from B-box genes of eudicot. AcPI was expressed not only in whorls 2 and 3, but also in whorl 1, thus corroborating the modified ABC model [23]. Moreover, AcPI was expressed in whorl 4. AcPI was also expressed in other organs or tissues such as leaves, fruit stems, flesh, apical meristems and bracts. These data indicated a possible function of AcPI during flower development as well as in leaves and fruit development. Busi M et al. [37] reported that B-box genes were involved in other processes than flower development such as the establishment of developing embryos, seed coat and ultimately in root and fruit development. Kim et al. [38] also found AP3 and PI transcripts were detected in petals, stamens and carpels in basal angiosperms such as Amborella and Nuphar. The expression of AcPI reached the highest level at 40 days after flower inducement, when the multiple fruit and floral organs were forming. The results showed the important role of AcPI in fruit and floral organs development and indicated that ethephon may stimulate the expression of AcPI. Ethephon are used widely to induce flowering in pineapples, possibly because when ethephon reaches the shoot apex, it can accelerate the expression of flower-related genes [39]. The effect of ethephon on inducing flowering was affected by temperature, relative humidity, solution pH, and the surface on which solution droplets were placed [39]. The ectopic expression of AcPI proved that AcPI could accelerate flowering and inflorescences or branches forming in Arabidopsis.

Plants of pineapple cultivar Comte de Paris were planted in a vinyl house at natural temperature and light. The apical meristems of young panicle (4–5 cm in height) were cut and collected for cloning cDNA. When the young panicle was 4–5 cm in height, the apical meristems, flesh, fruit stems, bracts, leaves, petals, sepals, stamens and carpels were used for qRT-PCR analyses. The apical meristems at 1 days before flower inducing and 10 days, 20 days, 30 days, 40 days, 50 days after flower inducing with 40% ethephon (200 mg/mL, 30 mL each plant) were also used for qRT-PCR analyses. All the materials were frozen in liquid nitrogen immediately after collection and stored at −80 °C until use.

Total RNA was extracted from apical meristems using Column Plant RNAout 2.0 kit (TIANDZ, Inc, China) following the manufacturer’s instructions. First-strand cDNA was synthesized with M-MLV-Reverse Trancriptase from TAKARA according to the manufacturer’s instructions.

To clone the conserved region of AcPI cDNA, a pair of primers PI-partial-F (5′ CCAGGTC TCCVTCGTCAT 3′) and PI-partial-R (5′ ACACCAGTNAGACCATTC 3′) was designed according to the conserved regions of PI homologous genes from other plants using the Primer Premier 5 software. The PCR amplification was performed with 1 cycle at 94 °C for 3 min; 32 cycles at 94 °C for 0.5 min, 55 °C for 1 min, and 72 °C for 1 min; 1 cycle at 72 °C for 10min. PCR products were isolated and cloned into pMD18-T Vector (TAKARA, Dalian, China) to sequence. The cloned sequence was used to design gene-specific primers to amplify the 5′ and 3′ end of cDNA using the RACE cDNA Amplification Kit (Clontech, Japan). The gene-specific primers were 5′PI-OUT (5′ CCA TTC TGG AGA GCC TCT TCT ATC GG 3′) and 5′PI-IN (5′ GCT CGA TCT GCA TGT TGT CGT TCT CT 3′) for 5′ RACE, 3′PI-OUT (5′ CGG GAA GAT GTC CGA GTA CTG TAG CC 3′) and 3′PI-IN (5′ AAC GAC AAC ATG CAG ATC GAG CTC AG 3′) for 3′ RACE. The first round PCR and the nested amplification were carried out according to the instruction of RACE cDNA Amplification Kit. The PCR products were cloned into pMD18-T vector and sequenced.

The full length cDNA of AcPI was obtained by PCR using the forward primer PIQC-F (5′ AAG CAG TGG TAT CAA CGC AGA GTA 3′) and the reverse primer PIQC-R (5′ ATA GCA GAC AAA GTC GAT GGC AGA 3′). Another pair of primers PImq-F (5′ TGC TCTAGA ATG GGG CGG GGG AAG ATC GAG AT 3′) and PImq-R (5′ CG GGATCC ATA GCA GAC AAA GTC GAT GGC AGA 3′) contained the generated XbaI and BamHI recognition sites respectively to facilitate the transformation of AcPI into Arabidopsis. The cycling condition:1 cycle at 94 °C for 3 min; 35 cycles at 94 °C for 0.5 min, 50 °C for 1.5 min and 72 °C for 1 min; a final extension at 72 °C for 10 min. The PCR products were cloned into pMD18-T vector and sequenced.

Sequence chromatograms were examined and edited using Chromas Version 2.23. Related sequences were found using BLAST [40]. Amino acid alignments with homologous sequences from a range of species were manually edited with the DNAMAN 6.0 program. For determination of amino acid identities, sequences taken from the alignment were pairwise-compared using DNAMAN 6.0. A phylogenetic tree, based on the amino acid sequences, was constructed using DNAMAN 6.0. Gaps appearing in one sequence only, were treated as non-constant characters. The molecular weight and isoelectric point of the gene were analyzed on-line with ExPASy [41].

To study transcription of the AcPI gene by the qRT-PCR, total RNA of the apical meristems in different stages of flesh, fruit stems, bracts, leaves, petals, sepals, stamens and carpels was extracted with Column Plant RNAout 2.0 kit (TIANDZ, Inc, China) following the manufacturer’s instructions. First-strand cDNA was synthesized with PrimeScript^® RT Master Mix (Perfect Real Time, TAKARA, Dalian, China) according to the user manual. These cDNAs were used as templates for qRT-PCR. Each tissue was applied to three replications for gene expression.

To demonstrate that amplification efficiencies for both the AcPI gene and housekeeping gene Ac 18S rRNA were very similar, standard curves were carried out by PCR. The curves were generated for genes AcPI and Ac 18S rRNA using a 5× serial dilution curve, with final quantities of 100, 20, 4, 0.8, 0.16 and 0.032 ng of cDNA.

qRT-PCR was carried out using SYBR^® Premix Ex Taq™ kit (TAKARA, Dalian, China) and the PCR amplification was quantified according to the manufacturer’s protocol. Real time PCR reactions were performed in 25 μL mixtures. The mixture for one reaction contained 12.5 μL 1× SYBR Green PCR Master Mix with 0.5 μL ROXII as a reference dye for real time PCR, 1 μL 10 μM of forward primer, 1 μL 10μM reverse primer and 100 ng of cDNA. No template controls were run to determine contamination and level of primer dimmer formation. To make it possible to compare gene expression levels in the different plant tissues, they were normalized to the expression of Ac 18S rRNA in each tissue. Relative expression of AcPI in different plant tissues was obtained by dividing the average number of copies by the copy number of Ac 18S rRNA for the same tissues.

qRT-PCR reactions were run on a Stratagene Mx3005P detection system (Stratagene 3005P, USA) using the following universal cycling conditions for all amplifications: 1 cycle of 30 s at 95 °C (DNA polymerase activation); 40 cycles of 5 s at 95 °C and 1 min at 55 °C. At the end, a dissociation stage was added: 30 s at 95 °C, 1min at 55 °C and 30 s at 95 °C. Ct values were determined after automatic adjustment of the baseline and manual adjustment of the fluorescence threshold.

The primers used in this qRT-PCR were listed below: PIdl-up (5′ GCA CCA CCA AGA GAT GGC GAT G 3′), PIdl-dn (5′ TAG CAG ACA AAG TCG ATG GCA GAG A 3′). Ac 18S rRNA was used as the housekeeping gene. Ac 18S rRNA-up (5′ ATG GTG GTG ACG GGT GAC 3′), Ac 18S rRNA-dn (5′ AGA CAC TAA AGC GCC CGG TA 3′).

AcPI was excised from the pMD18-T vector using XbaI and BamHI restriction enzymes and inserted into the vector pBI121 under the control of cauliflower mosaic virus 35S promoter. After confirmation of the sequence, the plant expression vector was transformed into Agrobacterium tumefaciens strain GV3101 via the freeze-thaw method [42]. Then the 35S::AcPI was transformed into Arabidopsis thaliana ecotype Columbia plants using a floral dip method [43].

Transformants which had survived in the 1/2 times MS medium containing kanamycin (50 mg/L) were further verified by PCR analysis. For PCR analysis, the Column Plant DNAout kit (TIANDZ, Inc, China) was used to isolate DNA from fresh leaves (100 mg) of T1 transgenic plants and non-transgenic plants. The primers were PImq-F and PImq-R. The cycling condition was the same as that used in amplification of cDNA with enzyme recognition sites.