Characterization of two PEBP genes, SrFT and SrMFT, in thermogenic skunk cabbage (Symplocarpus renifolius)

Abstract

Floral thermogenesis has been found in dozens of primitive seed plants and the reproductive organs in these plants produce heat during anthesis. Thus, characterization of the molecular mechanisms underlying flowering is required to fully understand the role of thermogenesis, but this aspect of thermogenic plant development is largely unknown. In this study, extensive database searches and cloning experiments suggest that thermogenic skunk cabbage (Symplocarpus renifolius), which is a member of the family Araceae, possesses two genes encoding phosphatidyl ethanolamine-binding proteins (PEBP), FLOWERING LOCUS T (SrFT) and MOTHER OF FT AND TFL1 (SrMFT). Functional analyses of SrFT and SrMFT in Arabidopsis indicate that SrFT promotes flowering, whereas SrMFT does not. In S. renifolius, the stage- and tissue-specific expression of SrFT was more evident than that of SrMFT. SrFT was highly expressed in flowers and leaves and was mainly localized in fibrovascular tissues. In addition, microarray analysis revealed that, within floral tissues, SrFT was co-regulated with the genes associated with cellular respiration and mitochondrial function, including ALTERNATIVE OXIDASE gene proposed to play a major role in floral thermogenesis. Taken together, these data suggest that, among the PEBP genes, SrFT plays a role in flowering and floral development in the thermogenic skunk cabbage.

Figure 1. The life cycle of skunk cabbage (Symplocarpus renifolius).

S. renifolius is a polycarpic plant that flowers after several years of vegetative growth. In early spring, the flowers begin producing heat when they bloom and terminate heat production when the pollen is released from the anthers. At the end of spring, leaf development is initiated at floral senescence, and environmental conditions determine whether or not flowers will bloom in the following year. In summer, if pollination is successful, the flower becomes a fruit, which contains many seeds. (a) Seeds, (b) seedlings, (c) immature-stage flowers (pre-thermogenic stage), (d) female-stage flowers (thermogenic stage), (e) male-stage flowers (post-thermogenic stage), (f) developed leaves, and (g) fruits. The thermal image of the right panel in (d) was taken using an FLIR SC 620 thermal imager (FLIR SYSTEMS).

Figure 2. Sequence analysis of SrFT, SrMFT, and related proteins.

(a) Alignment of the predicted amino acid sequences of SrFT, SrMFT, and related proteins. The predicted SrFT sequence was aligned with CgFT (Cymbidium goeringii), OnFT (Oncidium Gower Ramsey), OsHd3a (rice), and AtFT (Arabidopsis). The SrMFT sequence was aligned with OsMFT1 and 2 (rice), GmMFT (Glycine max), and AtMFT (Arabidopsis) sequences. Amino acids highlighted in blue, pink, or yellow are conserved across all, more than eighty percent, or more than half of the protein sequences in the alignment, respectively. Arrowheads indicate amino acids (Y84 and Q139) that are critical to FT function. (b) A phylogenetic tree of the predicted amino acid sequences of SrFT and related proteins. (c) A phylogenetic tree of the predicted amino acid sequences of SrMFT and related proteins. The tree was created using MEGA 6.06 with the Clustal W and maximum likelihood method. Bootstrap values above 30 are shown near each branch. The consensus phylogenic trees are shown with bootstrap values from 1000 replications. The accession numbers, locus IDs, and species abbreviations are listed in Supplementary Table 2.

Figure 3. The tissue- or stage-specific expression of two PEBP genes, SrFT and SrMFT.

(a) Tissue-specific expression in plant tissues. (b) Tissue-specific expression in floral tissues. (c) Stage-specific expression in flowers during the transition from floral buds to mature flowers. (d) Stage-specific expression in flowers at the late stage of floral maturation, including immature (pre-thermogenesis), female (strong thermogenesis), bisexual (weak thermogenesis), and male (post-thermogenesis) stages. (e) Stage-specific expression in leaves of juvenile (Juv)- and reproductive (Rep)-phase plants. To prevent saturation, semi-quantitative RT-PCR was analysed with different PCR cycles (Supplementary Fig. 7) and the appropriate numbers of PCR cycles were shown on the right side of each gel. The signals detected here were not from contaminated genomic DNA (Supplementary Fig. 8). In (c,e), potted plants were used to collect samples (Supplementary Fig. 9).

Figure 4. Localization of SrFT mRNA in flowers and leaves.

DIG-labeled antisense RNA probes were hybridized on cross-sections of flowers (a–d) and leaves (e,f). As the control, sense RNA probes were hybridized on cross-sections of flowers (g) and leaves (h). Arrowheads indicate the locations of signals. Pe, petal; Pi, pistil; P, pith; S, stamen; V, vascular bundle; xy, xylem; ph, phloem. Bar =100 μm.

Figure 5. Microarray analysis of four different floral tissues.

(a) SrFT and SrAOX, representative genes for flowering and respiration/thermogenesis, respectively, had similar expression profiles among the different floral tissues. (b) 27,184 genes were classified into four clusters (I-IV) according to their expression profiles in the different floral tissues. Of the resulting 6523, 7259, 7867, and 5535 genes in clusters I, II, III and IV, respectively, 2052, 1981, 2438, and 1808 were annotated with AGI codes, respectively. The annotated genes were classified based on the GO annotation that assigns cellular components and biological processes to each sequence. (c) Categories that had the highest percentage representation in cluster IV of all clusters were extracted. Most notably, many genes involved in mitochondrial function and in electron transport and the energy pathway were found in cluster IV.

Figure 6. Analysis of transgenic Arabidopsis plants that ectopically expressed SrFT or SrFT-GFP.

(a) Phenotypes of 33-day-old wild type (empty vector), 35S::SrFT, and 35S::SrFT-GFP transgenic plants grown in soil under LD conditions. (b,c) Flowering response of wild type, 35S::SrFT, and 35S::SrFT-GFP transgenic plants in soil under LD conditions. Data are means ± standard deviation (SD; N = 10). (d) SrFT mRNA expressed in rosette leaves of 33-day-old wild type, 35S::SrFT, and 35S::SrFT-GFP transgenic plants were analyzed using RT-PCR. Actin is shown as a loading control. (e) Rosette leaves of 33-day-old wild type and 35S::SrFT-GFP transgenic plants were analyzed using SDS-PAGE and then visualized using CBB-staining or immunoblotting with α-GFP antibodies.

Figure 7. Analysis of transgenic Arabidopsis plants that ectopically expressed SrMFT or SrMFT-GFP.

(a) Phenotypes of 46-day-old wild type (empty vector), 35S::SrMFT, and 35S::SrMFT-GFP transgenic plants grown in soil under LD conditions. (b,c) Flowering response of wild type, 35S::SrMFT, and 35S::SrMFT-GFP transgenic plants in soil under LD conditions. Data are means ± standard deviation (SD; N = 10). (d) SrMFT mRNA expressed in rosette leaves of 29-day-old wild type, 35S::SrMFT, and 35S::SrMFT-GFP transgenic plants were analyzed using RT-PCR. Actin is shown as a loading control. (e) Rosette leaves of 29-day-old wild type and 35S::SrMFT-GFP transgenic plants were analyzed using SDS-PAGE and then visualized using CBB-staining or immunoblotting with α-GFP antibodies.