Jatropha curcas ortholog of tomato MADS-box gene 6 (JcTM6) promoter exhibits floral-specific activity in Arabidopsis thaliana

Introduction

Promoter plays a significant role in gene expression regulation. Three types of promoters are currently employed in plant genetic engineering: constitutive, tissue-specific, and inducible promoters (Muthusamy et al., 2017; Potenza, Aleman & Sengupta-Gopalan, 2004). Tissue-specific promoters drive transgene expression in a specific spatiotemporal pattern, which is effective in the modification of agronomic traits of crop plants. For example, the rice (Oryza sativa L.) gene OsGA2ox1 encodes a gibberellin (GA) catabolic enzyme, GA 2-oxidase (Lester et al., 1999; Martin, Proebsting & Hedden, 1999; Thomas, Phillips & Hedden, 1999). When the expression of OsGA2ox1 was driven by the constitutive Actin promoter, transgenic rice plants failed to set grains. To prevent sterility, the promoter of a GA biosynthesis gene, OsGA3ox2, which encodes GA 3-oxidase and is specifically active in shoots, was used to control the expression of OsGA2ox1. As expected, transgenic rice exhibited a semi-dwarf phenotype with normal yield (Sakamoto et al., 2003). GA 20-oxidase is a GA biosynthetic enzyme in plants (Coles et al., 1999). In poplar (Populus spp.), overexpression of the Pinus densiflora GA 20-oxidase gene (PdGA20ox) under the control of the constitutive 35S promoter increased GA levels, thereby accelerating stem growth and plant biomass; however, transgenic poplar plants showed poor leaf development and root growth. When the PdGA20ox gene was driven by a xylem-specific promoter DX15 from poplar, the undesirable phenotypes were reduced (Jeon et al., 2016).

Physic nut (Jatropha curcas L.) is an oilseed plant belonging to the Euphorbiaceae family. The seed oil of Jatropha is a promising feedstock for biodiesel production (Kumar & Sharma, 2008). However, low seed yield, which is mainly caused by low female: male ratio, is a long-standing problem in Jatropha (Raju & Ezradanam, 2002; Rao et al., 2008). Jatropha is a monoecious plant species with male and female flowers on the same inflorescence, and the average ratio of female to male flowers is 1:13−1:29 (Raju & Ezradanam, 2002; Tewari et al., 2007). There are 100−300 flowers in each inflorescence of Jatropha, which only produce approximately 10 fruits (Kumar & Sharma, 2008; Pan & Xu, 2011). Hence, genetic modification of flowering characteristics is critical for Jatropha breeding. Floral-specific promoters play crucial roles in this modification because they can drive efficient expression of functional genes in flowers without affecting the vegetative growth of plants. In pea (Pisum sativum), the PsEND1 promoter exhibits anther-specific activity. Expression of the ribonuclease gene barnase (Gardner, Felsheim & Smith, 2009) in Arabidopsis and Brassica napus under the control of the PsEND1 promoter causes anther ablation at an early developmental stage, leading to male sterility (Roque et al., 2007). Arabidopsis APETALA3 (AP3) promoter was identified as a floral-specific promoter in petunia (Petunia x hybrida). Expression of the Agrobacterium tumefaciens isopentenyltransferase (ipt) gene under the control of the AtAP3 promoter in petunia increased the flower size, without affecting vegetative development (Verdonk et al., 2008). However, analysis of promoters, especially floral-specific promoters, in Jatropha is limited. Although the Jatropha APETALA1 (JcAP1) promoter was recently identified as a reproductive tissue-specific promoter showing high activity in inflorescence buds and seeds (Tao et al., 2016), it is not sufficient to address transgene expression analysis in Jatropha.

In this study, we isolated the promoter of the Jatropha ortholog of TOMATO MADS-BOX GENE 6 (JcTM6), a floral-specific gene. The activity of JcTM6 promoter was evaluated in Arabidopsis using the β-glucuronidase (GUS) reporter gene. The results of GUS staining showed that the JcTM6 promoter was active only in flowers, with the highest activity in stamens. By using this promoter directed a cytokinin biosynthesis gene, Arabidopsis ATP/ADP isopentenyltransferase 4 (AtIPT4) gene (Li et al., 2010), only flower phenotype was changed in transgenic Arabidopsis. Furthermore, deletion analysis showed that an approximately 0.85-kb fragment of the JcTM6 promoter (–1717 to –876 bp) is critical for maintaining its floral-specific expression pattern.

Materials & Methods

Results

JcTM6 expression in Jatropha

We identified the JcTM6 cDNA (GenBank accession no. MN820724) from our Jatropha flower cDNA library constructed previously (Chen et al., 2014). JcTM6 encodes a 230-amino acid protein, which shows high similarity to TM6 homologs from other plant species (Fig. 1A). Phylogenetic analyses showed that JcTM6, which contains the paleoAP3 motif, belongs to the TM6 group, rather than the euAP3 group (Fig. 1B).

To analyze the expression pattern of JcTM6 in Jatropha, qRT-PCR was performed using total RNA extracted from various tissues including roots, stems, leaves, inflorescences, female and male flowers, and pericarps and seeds at 42 DAP. The JcTM6 gene was predominantly expressed in female and male flowers (Fig. 2), indicating that JcTM6 is a flower-specific gene. Furthermore, JcTM6 showed high expression in the stamens of male flowers and petals of male and female flowers but low expression in sepals and pistils (Fig. 2). Thus, the expression pattern of JcTM6 in floral organs is consistent with that of class B genes (Weigel & Meyerowitz, 1994).

Isolation and sequence analysis of JcTM6 promoter

A 1.8-kb fragment of the JcTM6 promoter (Fig. 3A, –1717 to +103 bp; GenBank accession no. MN044579) was isolated from Jatropha genomic DNA by genome walking (Siebert et al., 1995). The transcription start site of JcTM6 was located 103 nt upstream of the translation start codon (Fig. 3A). Analysis of the JcTM6 promoter using the PLACE database (Higo et al., 1999) revealed various putative cis-elements in the 1.8-kb JcTM6 promoter fragment (Fig. 3A) including two CArG boxes, which act as binding sites for MADS-box transcription factors (Irish & Yamamoto, 1995), some pollen-specific elements, including five GTGANTG10 motifs (GTGA) and eight POLLEN1LELAT52 motifs (AGAAA) (Muschietti et al., 1994; Rogers et al., 2001), and a Q element (TGACCT), which shows enhancer-like activity for the pollen-specific expression of maize (Zea mays L.) ZM13 gene (Hamilton, Schwarz & Mascarenhas, 1998).

Activity of the JcTM6 promoter in Arabidopsis

To detect the activity of JcTM6 promoter, a JcTM6 promoter-GUS fusion construct (Fig. 3B) was expressed in Arabidopsis, and GUS staining was monitored in homozygous T3 plants (Fig. 4). No GUS staining was observed in 10-day-old Arabidopsis seedlings (Fig. 4A). Among the five tissues of adult plants examined (including roots, stems, leaves, flowers, and green siliques), GUS staining was detected only in flowers (Figs. 4B–4G). Among all floral organs, GUS staining intensity was the strongest in stamens, followed by sepals and petals, with faint staining in carpels (Fig. S1). Based on the results of GUS staining, we conclude that the JcTM6 promoter functions as a flower-specific promoter in Arabidopsis.

Deletion analysis of the JcTM6 promoter

To analyze the region essential for flower-specific activity of the JcTM6 promoter, we carried out a deletion analysis. A deletion variant of the JcTM6 promoter lacking the region from –1,717 to –876 bp was fused to the GUS gene and transformed into Arabidopsis (Fig. 5A). Compared with the full-length JcTM6 promoter, the deletion was not only active in flowers but also in young leaves (Fig. 5B). Moreover, the deletion showed no promoter activity in stamens but increased activity in sepals and petals (Fig. 5C and 5D). These results indicate that the region from −1,717 to −876 bp is critical for JcTM6 promoter activity in stamens and inhibition of promoter activity in young leaves, sepals, and petals.

JcTM6:AtIPT4 transgenic Arabidopsis produced large flowers

To further verify the floral specificity of JcTM6 promoter, a cytokinin biosynthetic gene (AtIPT4) was expressed under the control of JcTM6 promoter in Arabidopsis. JcTM6:AtIPT4 vector was constructed and was transformed into Arabidopsis plants. A total of 25 independent JcTM6:AtIPT4 lines were obtained. As expected, all transgenic lines showed no vegetative difference from the wild type and most of them produced larger flowers (Fig. 6). Furthermore, the development of siliques was also unaffected. To verify the morphological alteration in flowers that is caused by the transgene, we examined the expression levels of AtIPT4 and the cytokinin signaling genes Arabidopsis histidine kinase 2 (AHK2) (Nishimura et al., 2004) and Arabidopsis response regulator 5 (ARR5) (D’Agostino, Deruère & Kieber, 2000) in wild type and JcTM6:AtIPT4 transgenic plants. The expression level of AtIPT4 in flowers of transgenic lines is significantly higher than that in wild type, whereas the AtIPT4 expression in the leaves of transgenic plants was not different from that in leaves of wile-type plants (Fig. 7A). As expected, higher expression levels of AHK2 and ARR5 were detected in the flowers of transgenic lines (Fig. 7B). These results indicate that the morphological alteration in flowers of JcTM6:AtIPT4 transgenic plants is caused by the flower-specific expression of the transgene driven by the JcTM6 promoter. JcTM6 promoter is indeed a flower-specific promoter.

Discussion

TM6 is a member of the MADS-box gene family, which belongs to the paleoAP3 lineage (Pnueli et al., 1991; Rijpkema et al., 2006; Wu et al., 2011). In tomato (Solanum lycopersicum) and petunia, TM6 functions as a class B gene that plays an essential role in stamen development, although it is mainly expressed in whorls 3 and 4, similar to that of a class C gene (Martino et al., 2006; Rijpkema et al., 2006). In trioecious papaya (Carica papaya) plants, which produce male, female, and hermaphrodite flowers, two TM6 genes were isolated previously (CpTM6-1 and CpTM6-2). Both genes are predominantly expressed in the petals of all sex types and stamens of hermaphrodite and male flowers, although CpTM6-2 is also expressed in leaves (Ackerman et al., 2008). In this study, we identified JcTM6 as a flower-specific gene in Jatropha, with high expression in female and male flowers (Fig. 2). Similar to CpTM6-1, the JcTM6 gene showed high expression in the petals of female and male flowers and stamens of male flowers. Because JcTM6 showed flower-specific expression, we isolated its upstream region from Jatropha genomic DNA and analyzed its activity in Arabidopsis by GUS staining.

In transgenic Arabidopsis, GUS staining showed that the JcTM6 promoter was active only in flowers (Fig. 4), suggesting that the JcTM6 promoter is a flower-specific promoter. AtIPT4 is a cytokinin biosynthesis gene encoding ATP/ADP isopentenyltransferase. The expression of this gene under the control of AP1 promoter results in the alterations in flower number and organs (Li et al., 2010). However, the AtIPT4 driven by JcTM6 promoter only gave rise to the changes in flower organs (Fig. 6), indicating that JcTM6 promoter is active at the late stage of flower development rather than floral meristem. This activity is consistent with the expression pattern of the JcTM6 gene in Jatropha. Recently, Ming et al. (2020) showed that JcTM6 promoter has a high activity in female flowers of Jatropha, suggesting that JcTM6 promoter can drive flower-specific expression of transgenes in different plant species.

When the 842-bp fragment of the JcTM6 promoter (−1,717 to −876 bp) was deleted, the promoter was not only active in flowers but also in young leaves (Fig. 5B). We found that the deleted region contained one of the two CArG box motifs, which are very important for mediating the regulatory effect of MADS-box transcription factors (Dolan & Fields, 1991; Treisman, 1992). In Jatropha, a fragment of the JcAP1 promoter (from −1,313 to −1,057 bp), which contains a CArG box motif, is required for promoter activity in inflorescence buds (Tao et al., 2016). The Arabidopsis AP3 promoter contains three CArG boxes: CArG1 is essential for AP3 promoter activity at all stages of flowering; CArG2 is critical for AP3 expression in petals, and CArG3 represents the binding site of a transcription factor that represses the activity of AP3 promoter during early floral stages (Tilly, Allen & Jack, 1998). Therefore, we propose that the CArG box motif in JcTM6 promoter plays an important role in conferring floral-specific activity in transgenic plants.

Among the floral organs, stamens exhibited the highest activity of JcTM6 promoter (Fig. 4F). This expression pattern could be regulated by pollen-specific elements contained in this promoter, including five GTGA and eight AGAAA motifs. The GTGA motif is critical for the expression of g10 promoter in tobacco pollen because mutation of the GTGA motif reduced g10 promoter activity in pollen (Rogers et al., 2001). The AGAAA motif, which was identified in the tomato late-stage pollen-specific LAT52 promoter, is necessary for promoter activity during pollen maturation (Bate & Twell, 1998). In potato (Solanum tuberosum L.), the GTGA and AGAAA motifs present in the promoter of SBgLR, a pollen-specific gene, are critical for high-level gene expression in pollen (Lang et al., 2008). In the current study, deletion of an 842-bp fragment of the JcTM6 promoter, containing four GTGA and two AGAAA motifs, abolished promoter activity in stamens (Fig. 5D). We assumed that these motifs are essential for the activity of the JcTM6 promoter in stamens. Given the importance of CArG box motifs, it is possible that the GTGA and AGAAA motifs cooperate with the CArG box to regulate JcTM6 promoter activity in stamens. In addition, although the deleted region contained six AGAAA motifs, these motifs do not seem to be required for JcTM6 promoter activity in stamens. Furthermore, the deleted region also contained a 6-bp quantitative element (Q-element), which plays an enhancer-like role (Hamilton, Schwarz & Mascarenhas, 1998). In maize, deletion of the Q-element from the pollen-specific ZM13 promoter reduced the promoter activity by 10-fold (Hamilton et al., 2000). Deletion of the Q-element probably also contributed to the loss of JcTM6 promoter activity in stamens in this study (Fig. 5D). In addition, the deletion variant of the JcTM6 promoter exhibited increased activity in sepals and petals (Fig. 5C and 5D), indicating the presence of potential negative elements in the deleted region, which inhibit promoter activity in sepals and petals. By the deletion analysis of the JcTM6 promoter, we demonstrate the combination of these elements are of great importance to the promoter activity in the flowers, and detailed studies of the functions of these elements will be conducted in the future.

Conclusions

Floral-specific promoters play crucial roles in genetic modification of flowering characteristics. In this study, a 1.8-kb JcTM6 promoter fragment was isolated from Jatropha and characterized as a flower-specific promoter in transgenic Arabidopsis plants. When the region from –1,717 to –876 bp in the JcTM6 promoter was deleted, the promoter lost its flower-specific activity and gained activity in young leaves. Our results suggest that the JcTM6 promoter could be used to drive flower-specific expression of transgenes in plants.

Supplemental Information